The Prokaryotes

The Prokaryotes
Eugene Rosenberg (Editor-in-Chief)

Edward F. DeLong, Stephen Lory, Erko Stackebrandt and Fabiano Thompson (Eds.)
The Prokaryotes
Prokaryotic Biology and Symbiotic Associations
Fourth Edition
With 183 Figures and 62 Tables

Editor-in-Chief
Eugene Rosenberg
Department of Molecular Microbiology and Biotechnology
Tel Aviv University
Tel Aviv, Israel
Editors
Edward F. DeLong Fabiano Thompson
Department of Biological Engineering Laboratory of Microbiology, Institute of Biology, Center for
Massachusetts Institute of Technology Health Sciences
Cambridge, MA, USA Federal University of Rio de Janeiro (UFRJ)
Ilha do Fundão, Rio de Janeiro, Brazil
Stephen Lory
Department of Microbiology and Immunology
Harvard Medical School
Boston, MA, USA
Erko Stackebrandt
Leibniz Institute DSMZ-German Collection of Microorganisms
and Cell Cultures
Braunschweig, Germany
ISBN 978-3-642-30193-3 ISBN 978-3-642-30194-0 (eBook)

ISBN 978-3-642-30195-7 (print and electronic bundle)
DOI 10.1007/978-3-642-30194-0
Springer Heidelberg New York Dordrecht London
Library of Congress Control Number: 2012950985
3rd edition: © Springer Science+Business Media, LLC 2006

4th edition: © Springer-Verlag Berlin Heidelberg 2013
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of
translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed.
Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically for the purpose of
being entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is permitted
only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained from
Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution under the respective
Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific
statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher
can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the
material contained herein.
Printed on acid-free paper
Springer is part of Springer ScienceþBusiness Media (www.springer.com)

Foreword
The purpose of this brief foreword is unchanged from the first edition; it is simply to make you, the reader, hungry for the scientific
feast that follows. These 11 volumes (planned) on the prokaryotes offer an expanded scientific menu that displays the biochemical
depth and remarkable physiological and morphological diversity of prokaryote life. The size of the volumes might initially discourage
the unprepared mind from being attracted to the study of prokaryote life, for this landmark assemblage thoroughly documents the
wealth of present knowledge. But in confronting the reader with the state of the art, the Handbook also defines where more work
needs to be done on well-studied bacteria as well as on unusual or poorly studied organisms.
This edition of The Prokaryotes recognizes the almost unbelievable impact that the work of Carl Woese has had in defining
a phylogenetic basis for the microbial world. The concept that the ribosome is a highly conserved structure in all cells and that its
nucleic acid components may serve as a convenient reference point for relating all living things is now generally accepted. At last, the
phylogeny of prokaryotes has a scientific basis, and this is the first serious attempt to present a comprehensive treatise on prokaryotes
along recently defined phylogenetic lines. Although evidence is incomplete for many microbial groups, these volumes make
a statement that clearly illuminates the path to follow.
There are basically two ways of doing research with microbes. A classical approach is first to define the phenomenon to be studied
and then to select the organism accordingly. Another way is to choose a specific organism and go where it leads. The pursuit of an
unusual microbe brings out the latent hunter in all of us. The intellectual challenges of the chase frequently test our ingenuity to the
limit. Sometimes the quarry repeatedly escapes, but the final capture is indeed a wonderful experience. For many of us, these simple
rewards are sufficiently gratifying so that we have chosen to spend our scientific lives studying these unusual creatures. In these
endeavors, many of the strategies and tools as well as much of the philosophy may be traced to the Delft School, passed on to us by our
teachers, Martinus Beijerinck, A. J. Kluyver, and C. B. van Niel, and in turn passed on by us to our students.
In this school, the principles of the selective, enrichment culture technique have been developed and diversified; they have been
a major force in designing and applying new principles for the capture and isolation of microbes from nature. For me, the ‘‘organism
approach’’ has provided rewarding adventures. The organism continually challenges and literally drags the investigator into new areas
where unfamiliar tools may be needed. I believe that organism-oriented research is an important alternative to problem-oriented
research, for new concepts of the future very likely lie in a study of the breadth of microbial life. The physiology, biochemistry, and
ecology of the microbe remain the most powerful attractions. Studies based on classical methods as well as modern genetic
techniques will result in new insights and concepts.
To some readers, this edition of The Prokaryotes may indicate that the field is now mature, that from here on it is a matter of filling
in details. I suspect that this is not the case. Perhaps we have assumed prematurely that we fully understand microbial life. Van Niel
pointed out to his students that—after a lifetime of study—it was a very humbling experience to view in the microscope a sample of
microbes from nature and recognize only a few. Recent evidence suggests that microbes have been evolving for nearly 4 billion years.
Most certainly, those microbes now domesticated and kept in captivity in culture collections represent only a minor portion of the
species that have evolved in this time span. Sometimes we must remind ourselves that evolution is actively taking place at the present
moment. That the eukaryote cell evolved as a chimera of certain prokaryote parts is a generally accepted concept today. Higher as well
as lower eukaryotes evolved in contact with prokaryotes, and evidence surrounds us of the complex interactions between eukaryotes
and prokaryotes as well as among prokaryotes. We have so far only scratched the surface of these biochemical interrelationships.
Perhaps the legume nodule is a pertinent example of nature caught in the act of evolving the ‘‘nitrosome,’’ a unique nitrogen-fixing
organelle. The study of prokaryotes is proceeding at such a fast pace that major advances are occurring yearly. The increase of this
edition to four volumes documents the exciting pace of discoveries.
To prepare a treatise such as The Prokaryotes requires dedicated editors and authors; the task has been enormous. I predict that the
scientific community of microbiologists will again show its appreciation through use of these volumes—such that the pages will
become ‘‘dog-eared’’ and worn as students seek basic information for the hunt. These volumes belong in the laboratory, not in the
library. I believe that a most effective way to introduce students to microbiology is for them to isolate microbes from nature, that is,
from their habitats in soil, water, clinical specimens, or plants. The Prokaryotes enormously simplifies this process and should
encourage the construction of courses that contain a wide spectrum of diverse topics. For the student as well as the advanced
investigator, these volumes should generate excitement.
Happy hunting!
Ralph S. Wolfe
Department of Microbiology
University of Illinois at Urbana-Champaign
Preface
During most of the twentieth century, microbiologists studied pure cultures under defined laboratory conditions in order to uncover
the causative agents of disease and subsequently as ideal model systems to discover the fundamental principles of genetics and
biochemistry. Microbiology as a discipline onto itself, e.g., microbial ecology, diversity, and evolution-based taxonomy, has only
recently been the subject of general interest, partly because of the realization that microorganisms play a key role in the environment.
The development and application of powerful culture-independent molecular techniques and bioinformatics tools has made this
development possible. The fourth edition of the Handbook of the Prokaryotes has been updated and expanded in order to reflect this
new era of microbiology.
The first five volumes of the fourth edition contain 34 updated and 43 entirely new chapters. Most of the new chapters are in the
two new sections: Prokaryotic Communities and Bacteria in Human Health and Disease. A collection of microorganisms occupying
the same physical habitat is called a ‘‘community,’’ and several examples of bacterial communities are presented in the Prokaryotic
Communities section, organized by Edward F. DeLong. Over the last decade, important advances in molecular biology and
bioinformatics have led to the development of innovative culture-independent approaches for describing microbial communities.
These new strategies, based on the analysis of DNA directly extracted from environmental samples, circumvent the steps of isolation
and culturing of microorganisms, which are known for their selectivity leading to a nonrepresentative view of prokaryotic diversity.
Describing bacterial communities is the first step in understanding the complex, interacting microbial systems in the natural world.
The section on Bacteria in Human Health and Disease, organized by Stephen Lory, contains chapters on most of the important
bacterial diseases, each written by an expert in the field. In addition, there are separate general chapters on identification of pathogens
by classical and non-culturing molecular techniques and virulence mechanisms, such as adhesion and bacterial toxins. In recognition
of the recent important research on beneficial bacteria in human health, the section also includes chapters on gut microbiota,
prebiotics, and probiotics. Together with the updated and expanded chapter on Bacterial Pharmaceutical Products, this section is
a valuable resource to graduate students, teachers, and researchers interested in medical microbiology.
Volumes 6–11, organized by Erko Stackebrandt and Fabiano Thompson, contain chapters on each of the ca. 300 known
prokaryotic families. Each chapter presents both the historical and current taxonomy of higher taxa, mostly above the genus level;
molecular analyses (e.g., DDH, MLSA, riboprinting, and MALDI-TOF); genomic and phenetic properties of the taxa covered;
genome analyses including nonchromosomal genetic elements; phenotypic analyses; methods for the enrichment, isolation, and
maintenance of members of the family; ecological studies; clinical relevance; and applications.
As in the third edition, the volumes in the fourth edition are available both as hard copies and e-books, and as eReferences. The
advantages of the online version include no restriction of color illustrations, the possibility of updating chapters continuously and,
most importantly, libraries can place their subscribed copies on their servers, making it available to their community in offices and
laboratories. The editors thank all the chapter authors and the editorial staff of Springer, especially Hanna Hensler-Fritton, Isabel
Ullmann, Daniel Quiñones, Alejandra Kudo, and Audrey Wong, for making this contribution possible.
Eugene Rosenberg
Editor-in-Chief
About the Editors
Eugene Rosenberg (Editor-in-Chief)

Department of Molecular Microbiology and Biotechnology
Tel Aviv University
Tel Aviv
Israel
Eugene Rosenberg holds a Ph.D. in biochemistry from Columbia University (1961) where he described the chemical structures of the
capsules of Hemophilus influenzae, types B, E, and F. His postdoctoral research was performed in organic chemistry under the
guidance of Lord Todd in Cambridge University. He was an assistant and associate professor of microbiology at the University of
California at Los Angeles from 1962 to 1970, where he worked on the biochemistry of Myxococcus xanthus. Since 1970, he has been in
the Department of Molecular Microbiology and Biotechnology, Tel Aviv University, as an associate professor (1970–1974), full
professor (1975–2005), and professor emeritus (2006–present). He has held the Gol Chair in Applied and Environmental Micro-
biology since 1989. He is a member of the American Academy of Microbiology and European Academy of Microbiology. He has been
awarded a Guggenheim Fellowship, a Fogarty International Scholar of the NIH, the Pan Lab Prize of the Society of Industrial
Microbiology, the Proctor & Gamble Prize of the ASM, the Sakov Prize, the Landau Prize, and the Israel Prize for a ‘‘Beautiful Israel.’’
His research has focused on myxobacteriology; hydrocarbon microbiology; surface-active polymers from Acinetobacter; biore-
mediation; coral microbiology; and the role of symbiotic microorganisms in the adaptation, development, behavior, and evolution of
animals and plants. He is the author of about 250 research papers and reviews, 9 books, and 16 patents.
x About the Editors
Edward F. DeLong
Department of Biological Engineering
Massachusetts Institute of Technology
Cambridge, MA
USA
Edward DeLong received his bachelor of science in bacteriology at the University of California, Davis, and his Ph.D. in marine
biology at Scripps Institute of Oceanography at the University of California, San Diego. He was a professor at the University of
California, Santa Barbara, in the Department of Ecology for 7 years, before moving to the Monterey Bay Aquarium Research Institute
where he was a senior scientist and chair of the science department, also for 7 years. He now serves as a professor at the Massachusetts
Institute of Technology in the Department of Biological Engineering, where he holds the Morton and Claire Goulder Family
Professorship in Environmental Systems. DeLong’s scientific interests focus primarily on central questions in marine microbial
genomics, biogeochemistry, ecology, and evolution. A large part of DeLong’s efforts have been devoted to the study of microbes and
microbial processes in the ocean, combining laboratory and field-based approaches. Development and application of genomic,
biochemical, and metabolic approaches to study and exploit microbial communities and processes is his another area of interest.
DeLong is a fellow in the American Academy of Arts and Science, the U.S. National Academy of Science, and the American
Association for the Advancement of Science.
About the Editors xi
Stephen Lory
Harvard Medical School
Boston, MA
USA
Stephen Lory received his Ph.D. degree in Microbiology from the University of California in Los Angeles in 1980. The topic of his
doctoral thesis was the structure-activity relationships of bacterial exotoxins. He carried out his postdoctoral research on the basic
mechanism of protein secretion by Gram-negative bacteria in the Bacterial Physiology Unit at Harvard Medical School. In 1984, he
was appointed assistant professor in the Department of Microbiology at the University of Washington in Seattle, becoming full
professor in 1995. While at the University of Washington, he developed an active research program in host-pathogen interactions
including the role of bacterial adhesion to mammalian cells in virulence and regulation of gene expression by bacterial pathogens. In
2000, he returned to Harvard Medical School where he is currently a professor of microbiology and immunobiology. He is a regular
reviewer of research projects on various scientific panels of governmental and private funding agencies and served for four years on
the Scientific Council of Institute Pasteur in Paris. His current research interests include evolution of bacterial virulence, studies on
post-translational regulation of gene expression in Pseudomonas, and the development of novel antibiotics targeting multi-drug-
resistant opportunistic pathogens.
xii About the Editors
Erko Stackebrandt
Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures
Braunschweig
Germany
Erko Stackebrandt holds a Ph.D. in microbiology from the Ludwig-Maximilians University Munich (1974). During his postdoctoral
research, he worked at the German Culture Collection in Munich (1972–1977), 1978 with Carl Woese at the University of Illinois,
Urbana Champaign, and from 1979 to 1983 he was a member of Karl Schleifer’s research group at the Technical University, Munich.
He habilitated in 1983 and was appointed head of the Departments of Microbiology at the University of Kiel (1984–1990), at the
University of Queensland, Brisbane, Australia (1990–1993), and at the Technical University Braunschweig, where he also was the
director of the DSMZ-German Collection of Microorganisms and Cell Cultures GmbH (1993–2009). He is involved in systematics,
and molecular phylogeny and ecology of Archaea and Bacteria for more than 40 years. He has been involved in many research projects
funded by the German Science Foundation, German Ministry for Science and Technology, and the European Union, working on pure
cultures and microbial communities. His projects include work in soil and peat, Mediterranean coastal waters, North Sea and Baltic
Sea, Antarctic Lakes, Australian soil and artesian wells, formation of stromatolites, as well as on giant ants, holothurians, rumen of
cows, and the digestive tract of koalas. He has been involved in the description and taxonomic revision of more than 650 bacteria taxa
of various ranks. He received a Heisenberg stipend (1982–1983) and his work has been awarded by the Academy of Science at
Göttingen, Bergey’s Trust (Bergey’s Award and Bergey’s Medal), the Technical University Munich, the Australian Society for
Microbiology, and the American Society for Microbiology. He held teaching positions in Kunming, China; Budapest, Hungary;
and Florence, Italy. He has published more than 600 papers in refereed journals and has written more than 80 book chapters. He is the
editor of two Springer journals and served as an associate editor of several international journals and books as well as on national and
international scientific and review panels of the German Research Council, European Science Foundation, European Space Agency,
and the Organisation for Economic Co-Operation and Development.
About the Editors xiii
Fabiano Thompson
Laboratory of Microbiology
Institute of Biology
Center for Health Sciences
Federal University of Rio de Janeiro (UFRJ)
Ilha do Fundão
Rio de Janeiro
Brazil
Fabiano Thompson became the director of research at the Institute of Biology, Federal University of Rio de Janeiro (UFRJ), in 2012.
He was an oceanographer at the Federal University of Rio Grande (Brazil) in 1997. He received his Ph.D. in biochemistry from Ghent
University (Belgium) in 2003, with emphasis on marine microbial taxonomy and biodiversity. Thompson was an associate researcher
in the BCCM/LMG Bacteria Collection (Ghent University) in 2004; professor of genetics in 2006 at the Institute of Biology, UFRJ;
and professor of marine biology in 2011 at the same university. He has been a representative of UFRJ in the National Institute of
Metrology (INMETRO) since 2009. Thompson is the president of the subcommittee on the Systematics of Vibrionaceae–IUMS and
an associate editor of BMC Genomics and Microbial Ecology. The Thompson Lab in Rio currently performs research on marine
microbiology in the Blue Amazon, the realm in the southwestern Atlantic that encompasses a variety of systems, including deep sea,
Cabo Frio upwelling area, Amazonia river-plume continuum, mesophotic reefs, Abrolhos coral reef bank, and Oceanic Islands
(Fernando de Noronha, Saint Peter and Saint Paul, and Trindade).
Table of Contents
Section 1: Biology of Bacteria and Archaea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

1 How We Do, Don’t, and Should Look at Bacteria and Bacteriology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Carl R. Woese
2 What Is a Prokaryote? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
W. Ford Doolittle . Olga Zhaxybayeva
3 Prokaryotes and Their Habitats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Hans G. Schlegel . Holger W. Jannasch
4 Origin of Life: RNA World Versus Autocatalytic Anabolist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Günter Wächtershäuser
5 Morphological and Physiological Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Stephen H. Zinder . Martin Dworkin
6 Prokaryote Characterization and Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

Peter Kämpfer . Stefanie P. Glaeser
7 Principles of Enrichment, Isolation, Cultivation, and Preservation of Prokaryotes . . . . . . . . . . . . . . . . . . . . . . 149

Jörg Overmann
8 Comparative Genomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

Asli Ismihan Ozen . Tammi Vesth . David W. Ussery
9 Defining Taxonomic Ranks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

Konstantinos T. Konstantinidis . Erko Stackebrandt
10 Population Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255

Santiago Castillo-Ramı́rez . Edward J. Feil
11 Public Service Collections and Biological Resource Centers of Microorganisms . . . . . . . . . . . . . . . . . . . . . . . . 267

David Smith . Dagmar Fritze . Erko Stackebrandt
12 Repositories for Patented and Safeguarded Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305

Zea D. V. L. Mayerhoff . Irene von der Weid . Alessandra B. G. Valladão
13 Biotechnology and Applied Microbiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315

Eugene Rosenberg
14 Genomes and Post-genome Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329

Betania Ferraz Quirino . Cristine Chaves Barreto . Georgios J. Pappas . Karsten Zengler . Konstantinos Krampis . Ricardo H. Krüger
Section 2: Symbiotic Associations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
15 Role of Microorganisms in Adaptation, Development, and Evolution of Animals and Plants: The
Hologenome Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
Eugene Rosenberg . Ilana Zilber-Rosenberg
xvi Table of Contents
16 Cyanobacterial-Plant Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359

David G. Adams . Birgitta Bergman . Sandra A. Nierzwicki-Bauer . Paula S. Duggan . Amar N. Rai . Arthur Schüßler
17 Root and Stem Nodule Bacteria of Legumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401

Michael J. Sadowsky . Peter H. Graham . Masayuki Sugawara
18 Symbiotic Associations Between Ciliates and Prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427

Michael Schweikert . Masahiro Fujishima . Hans-Dieter Görtz
19 Bacteriocyte-Associated Endosymbionts of Insects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465

Paul Baumann . Nancy A. Moran . Linda C. Baumann
20 Vibrio fischeri: Squid Symbiosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497

Eric V. Stabb . Karen L. Visick
21 Rumen Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533

Itzhak Mizrahi
22 Symbiotic Associations Between Termites and Prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545

Andreas Brune
23 Marine Chemosynthetic Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579

Colleen M. Cavanaugh . Zoe P. McKiness . Irene L. G. Newton . Frank J. Stewart
List of Contributors
David G. Adams W. Ford Doolittle

Faculty of Biological Sciences Department of Biochemistry and Molecular Biology
University of Leeds Dalhousie University
Leeds Halifax, Nova Scotia
UK Canada
Cristine Chaves Barreto Paula S. Duggan

Genomic Sciences and Biotechnology Program Faculty of Biological Sciences
Universidade Católica de Brası́lia University of Leeds
Brası́lia, DF Leeds
Brazil UK
Linda C. Baumann Martin Dworkin

School of Nursing Department of Microbiology
University of Wisconsin-Madison University of Minnesota Medical School
Madison, WI Minneapolis, MN
USA USA
Edward J. Feil
Paul Baumann
Department of Biology and Biochemistry
Division of Agriculture and Natural Resources
University of Bath
University of California Cooperative Extension
Bath
Davis, CA
UK
USA
Dagmar Fritze
Birgitta Bergman German Collection of Microorganisms and Cell
Department of Botany Cultures GmbH
Stockholm University Braunschweig
Stockholm Germany
Sweden
Masahiro Fujishima
Andreas Brune Department of Environmental Science and Engineering
Max Planck Institute for Terrestrial Microbiology Graduate School of Science and Engineering
Marburg Yamaguchi University
Germany Yamaguchi
Japan
Santiago Castillo-Ramı́rez
Department of Biology and Biochemistry Stefanie P. Glaeser
University of Bath Institut für Angewandte Mikrobiologie
Bath Justus-Liebig-University Giessen
UK Gießen
Germany
Colleen M. Cavanaugh
Biological Laboratories 4081 Peter H. Graham∗
Cambridge, MA
∗
USA Deceased
xviii List of Contributors
Hans-Dieter Görtz Nancy A. Moran

Department of Zoology Biological and Biomedical Sciences Program
Biologisches Institut Yale University
Universität Stuttgart New Haven, CT
Stuttgart USA
Germany
Irene L. G. Newton
Holger W. Jannasch∗ Department of Biology
Indiana University Bloomington
Bloomington, IN
Peter Kämpfer
USA
Institut für Angewandte Mikrobiologie
Justus-Liebig-University Giessen
Gießen Sandra A. Nierzwicki-Bauer
Germany Department of Biology
Rensselaer Polytechnic Institute
Troy, NY
Konstantinos T. Konstantinidis
USA
School of Civil and Environmental Engineering and
School of Biology
Georgia Institute of Technology Jörg Overmann
Atlanta, GA Leibniz-Institut DSMZ-Deutsche Sammlung von
USA Mikroorganismen und Zellkulturen GmbH
Braunschweig
Konstantinos Krampis Germany
J. Craig Venter Institute
Rockville, MD
Asli Ismihan Ozen
USA
Department of Systems Biology
Center for Biological Sequence Analysis, Kemitorvet
Ricardo H. Krüger The Technical University of Denmark
Biology Institute Lyngby
UnB Brası́lia University Denmark
Brası́lia, DF
Brazil
Georgios J. Pappas
Embrapa–Genetic Resources and Biotechnology
Zea D. V. L. Mayerhoff Bioinformatics Laboratory
Brazilian Center of Biological Material Brası́lia, DF
Instituto Nacional da Propriedade Industrial Brazil
Rio de Janeiro and
Brazil Biology Institute
UnB Brası́lia University
Brası́lia, DF
Zoe P. McKiness Brazil
Wildwood, MO
USA
Betania Ferraz Quirino
Genomic Sciences and Biotechnology Program
Itzhak Mizrahi Universidade Católica de Brası́lia
Institute of Animal Science Brası́lia, DF
ARO, Volcani Research Center Brazil
Bet Dagan and
Israel Embrapa–Agrienergy
Brası́lia, DF
∗
Deceased Brazil
List of Contributors xix
Amar N. Rai Frank J. Stewart

Department of Biochemistry School of Biology
North-Eastern Hill University Georgia Institute of Technology
Shillong, Meghalaya Atlanta, GA
India USA
Eugene Rosenberg Masayuki Sugawara

Department of Molecular Microbiology and Biotechnology Department of Soil, Water, and Climate
Tel Aviv University University of Minnesota
Tel Aviv Saint Paul, MN
Israel USA
David W. Ussery
Michael J. Sadowsky
Department of Systems Biology
Department of Soil, Water, and Climate
Center for Biological Sequence Analysis, Kemitorvet
University of Minnesota
The Technical University of Denmark
Saint Paul, MN
Lyngby
USA
Denmark
Hans G. Schlegel Alessandra B. G. Valladão

Göttingen Academy of Sciences and Humanities Brazilian Center of Biological Material
Göttingen Instituto Nacional da Propriedade Industrial
Germany Rio de Janeiro
Brazil
Arthur Schüßler
Genetics Biocenter, University of Munich Tammi Vesth
Munich Department of Systems Biology
Germany Center for Biological Sequence Analysis, Kemitorvet
The Technical University of Denmark
Michael Schweikert Lyngby
Universität Stuttgart Denmark
Stuttgart
Germany
Karen L. Visick
David Smith Loyola University Chicago
CABI Maywood, IL
Egham, Surrey USA
UK
Eric V. Stabb Munich
Department of Microbiology Germany
University of Georgia and
Athens, GA Chapel Hill
USA USA
Erko Stackebrandt Irene von der Weid

Leibniz Institute DSMZ-German Collection of Microorganisms Brazilian Center of Biological Material
and Cell Cultures Instituto Nacional da Propriedade Industrial
Braunschweig Rio de Janeiro
Germany Brazil
xx List of Contributors
Carl R. Woese Ilana Zilber-Rosenberg

The School of Molecular and Cellular Biology Independant Scholar
University of Illinois at Urbana-Champaign Givat Shmuel
Urbana, IL
Israel
USA
Karsten Zengler
Stephen H. Zinder
Department of Bioengineering
Department of Microbiology
University of California, San Diego
La Jolla, CA Cornell University
USA Ithaca, NY
USA
Olga Zhaxybayeva
Department of Biological Sciences
Dartmouth College
Hanover, New Hampshire
USA
Section 1
Biology of Bacteria and Archaea

1 How We Do, Don’t, and Should Look at
Bacteria and Bacteriology
Carl R. Woese
The School of Molecular and Cellular Biology, University of Illinois at Urbana-Champaign,
Urbana, IL, USA
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 The Organismal Structure of the Archaea . . . . . . . . . . . . . . . . . . 13
Microbiology’s Halting Development . . . . . . . . . . . . . . . . . . . . . . . . 4 The Organismal Structure of the Eubacteria . . . . . . . . . . . . . . . 14

The Proteobacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
What Is Bacteriology? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 The Gram-Positive Eubacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
The Cyanobacterial Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Dumbing Down . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 The Chlorobium-Cytophaga Kingdom . . . . . . . . . . . . . . . . . . 16
The Spirochetes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Bacteriology’s Wandering Course . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 The Planctomycetes, Chlamydiae and
Verrucomicrobia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
The Gordian Knot of Bacterial Classification . . . . . . . . . . . . . . . 6 The Deinococcus-Thermus and Chloroflexus Group . . . . 18
The Remaining Eubacterial Kingdoms and Phyla . . . . . . . . 18
The Beginning of the End . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Where to Play? The Sand Castles Have All Background

Been Washed Away . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Microbiology today has a new-found wealth far greater than any
Here Comes the (Molecular) Cavalry . . . . . . . . . . . . . . . . . . . . . . . . 8 it possessed before. The source of that wealth is the universal
Cleaving the Gordian Knot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 phylogenetic tree—the framework essential for understanding
Reaching into the Unknown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 organismal relationships. The power that flows from phyloge-
netic ordering permeates the field. Microbiologists now accom-
Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 plish with ease things that were previously impossible and
approach bacteria in ways that 20 years ago were unthinkable.
Archaea and Eubacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Microbial ecology is no longer the faux ecology it had been—
when defining a niche in organismal terms was not an option.
Metabolism, Membranes, and Walls . . . . . . . . . . . . . . . . . . . . . . . . 10 Today, the field rests on a par with plant and animal ecology and
exceeds them in importance, for it is in the microbial realm that
Translation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 the base and fount of the global ecosystem lie. Studying microbial
diversity used to be the equivalent of hunting through antique
Transcription . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 shops for curios—which resulted in a collection of species no
more connected to one another than the items in a bower bird’s
Genome Replication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 nest. Now all organisms sit on the well-ordered tips of branches
on the universal phylogenetic tree (Woese 1987; Olsen et al.
Horizontal Gene Transfer: How Cells Evolved . . . . . . . . . . . . . . 12 1994), and the study of one, far from being an isolated adventure,
can contribute to the study of all. An interest in bacterial evolution
Microbiology on the Move . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 used to be perceived as metaphysical and worthless. Today, evolu-
tionary relationships are the foundation and motive force behind
Where Is Our Essence? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 a new and resurgent microbiology and hence biology as a whole.
Microbial genomes can be sequenced today in their entirety,
The View from the Top . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 and when measured against the information gained, the cost of
To try to plan for the future without a sense of the past is like trying to plant cut flowers.
(Daniel Boorstein)
E. Rosenberg et al. (eds.), The Prokaryotes – Prokaryotic Biology and Symbiotic Associations, DOI 10.1007/978-3-642-30194-0_2,
# Springer-Verlag Berlin Heidelberg 2013
4 1 How We Do, Don’t, and Should Look at Bacteria and Bacteriology
so doing is actually small—and that cost is continually decreas- One of the rare (if not only) times the problem was brought
ing. An exploration of a given bacterium can now start with to the fore was in 1962 when Stanier and van Niel lamented:
knowledge of all of its genes. Not only are we (potentially) privy ‘‘. . .the abiding intellectual scandal of bacteriology has been the
to the full range of the organism’s biochemistry, but we possess absence of a clear concept of a bacterium.’’ Unfortunately, the
a partial but very useful record of its evolutionary past. authors’ suggested remedy for the malaise was untenable, as we
This historical record, it turns out, extends so far back into shall see, and as a consequence served to encourage the unfor-
biological antiquity that the study of bacterial evolution fuses tunate changeover from the name ‘‘bacteriology’’ to ‘‘microbi-
with the study of the origins of modern cells (Woese 2002). ology,’’ a change that dogs microbiology to this day. A label that
Unfortunately, the universal phylogenetic tree, had denoted the study of what seemed a naturally defined
bacteriology’s essential framework, arrived rather late in the grouping of organisms, ‘‘bacteriology,’’ was replaced by one
course of events. Its belated arrival adversely affected that encompassed an organismal potpourri. One can in principle
not only the development of microbiology but that of all have a biologically valid, unified concept of ‘‘bacteria,’’ but
biology (see below). Consequently, the introductory chapter a comparable unifying concept of the artificial grouping ‘‘micro-
to the third edition of The Prokaryotes is largely devoted to organisms’’ is impossible. The semantic thimblerig involved in
understanding the historical development of bacteriology, this name change was one of the factors that helped to push the
for in its twists and turns lie insights into much of what micro- problem of a ‘‘concept of a bacterium’’ off the table. From a basic
biology and biology are today, as well as a guide to their future scientific perspective, bacteriology in 1992 was worse off than it
development. had been in 1962. But by 1992, bacteriology had a powerful
The editors of the second edition of The Prokaryotes wisely remedy for the situation.
decided to give the volume phylogenetic underpinnings (as best
they could at the time). In this third edition, the trend continues,
and the book’s phylogenetic bent is more pronounced. Bacterial What Is Bacteriology?
taxonomy, which only two decades ago was a dry subject whose
main if not only purpose was determinative classification, iden- Whatever else it is, bacteriology is first and foremost an organ-
tification of species, has now blossomed into an intriguing, ismal science, just as are zoology and botany. As such, its focus is
meaty study in its own right. As you use this third edition of understanding particular naturally defined groupings of organ-
The Prokaryotes, realize that the book is not merely a manual for isms. A critical difference between bacteriology and the other two
identification, cultivation, and determinative classification. Use organismal sciences, however, is that the other two could develop
it for what it really is, a long overdue treatise on comparative as organismal sciences in the absence of molecular characteriza-
bacteriology. If you let your imaginations follow the evolutionary tions, whereas bacteriology could not. Bacterial morphologies are
trails the book lays out, they will lead you to the edge of too simple and variable to have much phylogenetic significance,
microbiology’s future. Most of all, let this work by its structure but phylogeny is embedded in the complex morphologies and
open your eyes to the emerging world of bacterial evolution and variations on morphological themes characteristic of animal and
the challenge that poses to all of biology. This is not the evolution plant life. In the bacterial world, it is physiology (not morphol-
that Darwin and the classical evolutionists had in mind. It is ogy) that predominates and distinguishes. But, physiological
centered on physiologies rather than forms, on molecules rather characteristics have so far proven phylogenetically intractable.
than gross morphologies. And it extends the scientific reach much Whereas among eukaryotes, organization manifests itself
further into the past than was ever before possible. But to begin predominately in structure, organization in the bacterial world
charting the future, we need a map of the past. As in so many occurs mainly in the shifting and subtle biochemical states of the
walks of life, a knowledge of history is the best guide we have. system—something the eye cannot directly apprehend. In the
absence of a knowledge of phylogenetic relationships, a concept
of bacteria was unattainable.
Microbiology’s Halting Development What exactly is an organismal science? Simply put, one that
seeks to understand naturally defined groups of organisms in
Prior to its phylogenetic liberation, microbiology had long been biological terms. Such an understanding has four principal com-
mired down conceptually, sinking ever deeper as twentieth- ponents: (1) structure/function—how the organisms in the
century biology unfolded. This descent into innocence started group are built (organized) and how they work; (2) diversity—
long before that, however, when microbiologists failed to how many and what kinds of organisms the group comprises,
develop a phylogenetic, or ‘‘natural,’’ classification system for and the ways in which the various kinds are similar and different;
bacteria based upon the classical characteristics available to (3) ecology—how the organisms interact with their environments
them. As a consequence, bacteriology’s development became (including other organisms); and (4) evolution—the origins of
increasingly one-sided, so much so that from an organismal the group and how the organisms therein are ancestrally related.
perspective, the discipline developed not at all. The condition This last, evolution, is what underlies and enables the
was chronic, progressing over several generations of microbiol- development of a biological concept of a group. As Dobzhansky
ogists, and so, was scarcely felt, and to the extent that it was felt, famously put it, ‘‘. . .nothing in biology makes sense except in the
the condition was accepted as normal. light of evolution’’ (Dobzhansky 1963). And it was the lack of this
How We Do, Don’t, and Should Look at Bacteria and Bacteriology 1 5
essential evolutionary framework that prevented bacteriology going, the saying goes, any path will take you there. The future
from developing as an organismal science. course of bacteriology (microbiology) is not something that can
With no understanding of phylogenetic relationships, there be left to the vagaries of chance and necessity. How bacteriology
can be no effective study of ecology or diversity. Ecological niches now develops is a matter of utmost concern. The future of all
cannot be defined in organismal terms, and studies in bacterial biology turns upon it.
diversity amount to no more than a catalog of disconnected Although Leeuwenhoek discovered the microbial world dur-
vignettes. It is inconceivable that a zoologist or botanist could go ing the time of Newton, microbiology did not emerge as an
into the field and not be able to distinguish the animals from the effective science until two centuries later. In the last half of the
plants. Yet would-be microbial ecologists were in this exact posi- nineteenth century, the works of scientists like Cohn, Koch, and
tion or worse all the time: they couldn’t distinguish their ‘‘animals’’ Pasteur laid the groundwork. And with the next generation of
from their ‘‘plants’’ or (with a few exceptions) from representatives microbiologists, the Beijerincks, Winogradskys, and Orla-
of any others of the many kingdom-level bacterial taxa. Jensens, the study of microorganisms began in earnest. The
Blocked from developing into a full-fledged organismal sci- grouping of microbiologists that grew around Beijerinck and
ence throughout most of the twentieth century, bacteriology then Kluyver, known as the Delft School, became the dominant
wandered aimlessly, not knowing what it was, where it came influence in microbiology for most of the twentieth century.
from, or what it should become. The field was driven by the Among the school’s members, one finds articulate spokesmen
winds of scientific fad and other outside influences. Any devel- with an overview of their discipline. So, it is to the Delft School
opment our science did undergo was confined to its structure/ that we mainly turn to learn the early history of microbiology.
function aspect and its applied side. With exceptions too rare to Martinus Beijerinck is the actual founder of the Delft School,
impact, twentieth-century microbiologists (and biologists in although Leeuwenhoek, who had lived in Delft, is portrayed as
general) remained unaware of microbiology’s ‘‘identity crisis.’’ its spiritual father (van Niel 1949). Beijerinck is credited, along
with Winogradsky, with the development of enrichment cultur-
Dumbing Down ing, through which methodology microbiologists could begin
a major exploration of the nature and scope of the microbial
What more than anything served to obscure bacteriology’s world. Beijerinck’s contributions were many and varied (van
structural problem was the molecular perspective dominating Iterson et al. 1940): he added greatly to the understanding of
twentieth-century biology. Molecular biology embodies bacterial physiology and diversity. Through his studies on iron
a reductionist fundamentalism that rules out any holistic per- bacteria and the like, he was one of the first to appreciate the
spective. The paradigm assumed that the age-old problem of intimate role bacteria play in geologic processes. He was keenly
biological organization would be automatically solved when aware of developments in genetics (in the higher forms) and was
comprehensive molecular parts-lists for cells were generated. perhaps first to suggest that some of the variation seen in the
Yet even with the parts-lists now in hand, the problem remains microbial world was of mutational origin. His discovery of
with us, awaiting a fresh, constructive, and holistic outlook. The a filterable factor associated with tobacco mosaic disease made
molecular perspective took evolution for granted, found it intel- him a pioneer in virology. And he was keenly aware of Darwin.
lectually wanting, and dismissed its study as trivial. In such a How did this enlightened Dutch scientist view the study of
fundamentalist milieu, the organism per se has only a secondary microorganisms? On the occasion of his being awarded the
existence, shadowy and temporal. For molecular biology, the Leeuwenhoek Medal (microbiology’s highest and his most
organism lies essentially in its collective parts (which molecular prized honor), Beijerinck addressed the question:
biologists felt no compulsion to reassemble into a whole). Little " [M]icrobial ecology. . .is the most necessary and fruitful direction
wonder that under the molecular aegis, bacteriology felt no need
to guide us in organizing our knowledge of that part of nature
to develop into an organismal science, even if it could have done
which deals with the lowest limits of the organic world, and
so. Instead, microbiology followed molecular biology’s lead and
which constantly keeps before our minds the profound problem
slipped into a mechanistic reductionism.
of the origin of life itself (Beijerinck 1905; van Iterson et al. 1940;
Much of the important biochemistry of the last century was
translated by van Niel 1949).
done in bacterial systems. Bacteria—with their enormous pop-
ulation numbers, rapid growth rates, and general ease of This is a sophisticated view of microbiology. The importance
handling—also proved well suited to many of the contemporary of the microbe-environment relationship is foremost, with
studies in genetics and molecular biology. In one sense, great (microbial) evolution constantly in the background, shaping
progress occurred in microbiology over the last five or so and deepening the outlook. To me, Beijerinck’s concept of the
decades. None of it, however, contributed to the development microbial world seems deeper and more holistic than any that
of a concept of bacteria. followed until the present day, when the realities of genomics are
bringing us once more to a similarly broad and inspiring per-
Bacteriology’s Wandering Course spective on bacteria and their world.
In this quote (and in his studies), Beijerinck did not (to my
It is difficult to know where you are going if you don’t know knowledge) speak to the importance of determining the natural
where you came from. And if you don’t know where you are relationships among microorganisms. It is possible he did not
believe (as others of his day did not) that bacteria evolved in Kluyver and van Niel’s classic paper of 1936 on bacterial
a Darwinian fashion, as animals and plants do. I would prefer to classification represents the Delft School outlook. The paper
believe, however—especially given his familiarity with Darwin’s was a reasoned discussion of the problems faced in contempo-
writings (van Iterson et al. 1940)—that Beijerinck saw no way to rary bacterial classification, in the context of which the authors
determine these relationships given the primitive state of bacte- then proposed a system of their own, which they hoped would
riology in his day. Only with the later accumulation of large overcome some of the problems and, so, move bacteriology
numbers of bacterial species (mainly through enrichment cultur- closer to a genuine natural bacterial system. At the time, classi-
ing) would the problem of their classification become a choking fications were largely morphological and their purpose for the
one, and only then, when traditional taxonomic approaches failed most part utilitarian. A salient exception to these approaches had
to yield a satisfactory (natural) classification system, would the been the system of Orla-Jensen (see Kluyver and van Niel 1936),
problem become genuinely acute (Stanier and van Niel 1962). which was based on an evolutionary conjecture, namely, that the
Beijerinck’s successor at Delft was Albert Jan Kluyver. first organisms to evolve were necessarily autotrophs (Oparin and
A brilliant and sophisticated scientist, Kluyver was by training his ocean had yet to enter the picture). Like Beijerinck, Orla-
a biochemist, not a microbiologist. Only upon assuming Jensen appeared to believe that microorganisms held the key to
Beijerinck’s chair did he transform himself into the latter understanding life’s origin, and he constructed a classification that
(Kamp et al. 1959). It is in Kluyver that I see the beginning of hopefully would help to bring out the origin of metabolism. At
bacteriology’s drift away from striving to become an organismal this time, Kluyver and van Niel seemed to share Beijerinck’s belief
discipline. Kluyver is noted, and rightly so, for his fundamental, in the importance of bacteria as evolutionary beacons; at least they
unifying contributions to biochemistry (Kluyver and Donker said: ‘‘A true reconstruction of the course of evolution is the ideal
1926; Kluyver 1931). He pioneered what he called ‘‘comparative of every taxonomist’’ (Kluyver and van Niel 1936).
biochemistry,’’ a field he envisioned as ‘‘. . .benefit[ing] bio- In the Kluyver-van Niel article, one readily senses
chemistry in a manner similar to that in which the concept of bacteriology’s ongoing struggle regarding the bases upon
‘‘comparative anatomy’’ had helped to bring order into [anat- which to develop bacterial classification. Because zoological
omy]’’ (Kluyver 1931; van Niel 1949). Yet in one way, the and botanical classifications are morphologically based, a strong
analogy deceives. Comparative anatomy is basically an organis- precedent existed for putting bacterial classification on a similar
mal, evolutionary pursuit. Comparative biochemistry is not. footing—despite the fact that microbiologists intuitively knew
The latter is simply a way to bring some chemical order to the that bacteria are as fundamentally physiological as animals and
plethora of biochemistries that abound in nature. Kluyver’s plants are morphological. On what levels, to what extent, in what
simile here veils the important distinction between the organism ways, then, are the morphological and various physiological
and its parts. My assertion that Kluyver represented (or came to properties of bacteria taxonomically significant? In the follow-
represent) the biochemical dissection of bacteria rather than an ing, we see Kluyver and van Niel worry the problem (Kluyver
organismal comprehension thereof finds indirect support in van and van Niel 1936, pp. 370, 371):
Niel’s historical account of the Delft School, which details " The question then arises in which [bacterial] characters phylog-
Kluyver’s contributions to biochemistry while failing once to
eny expresses itself. There is no doubt that in this respect mor-
mention Kluyver’s (and his own) concerns with developing
phology remains the first and most reliable guide . . .[although]
a natural bacterial classification (van Niel 1949).
the indispensability of physiological characters for the purpose
of classification has also been generally accepted....
A lack of insight in the fundamentals of metabolism has thus
The Gordian Knot of Bacterial Classification far been the great stumbling block for a rational application of
physiological characteristics in taxonomy, and it also explains the
The Linnaean system had proven extremely useful in structuring
horror with which many systematicians have witnessed their
our knowledge of zoology and botany, making both into respect-
ever increasing use.... The fundamental nature of the energy
able organismal sciences even before Darwin’s time. Darwin’s
providing processes justifies the view that they should be rated
theory did not change the Linnaean classification of animals and
first amongst the physiological characters suitable for
plants all that much; it merely provided it theoretical justifica-
classification....
tion (Darwin 1859).
Early microbiologists were aware of the benefits of ‘‘natural’’ Because the classification sought was a natural one, the kinds
classification. Yet I have never satisfied myself as to the degree of of choices Kluyver and van Niel were entertaining required the
their commitment to or the depth of their appreciation of answers to certain evolutionary questions. But there were no
evolution. Was a natural system merely the most useful classifi- a priori answers—even hints of them. Natural classification then
cation, or were bacteria given the evolutionary significance had to rest on a foundation of conjecture—bolstered by the hope
Beijerinck (quoted above) seemed to accord them? Clearly, that if the scheme were anywhere near phylogenetically valid, it
most microbiologists of the time simply wanted some kind, could bootstrap itself into a true natural classification. (Only
any kind, of useful pigeonholing system. Bergey’s Manual, today is knowledge of evolutionary relationships independent of
a (much criticized though popular) determinative system, pro- [and so can precede the construction of] a taxonomy rather than
vided precisely that (Stanier and van Niel 1941). follow from it.)
There is little point in detailing Kluyver and van Niel’s (Breed 1939; Stanier and van Niel 1941). The overarching per-
attempt at a natural classification. But since the tenor of their spective of the idealists recognized the importance to the disci-
thinking is so instructive as to the microbiological gestalt of their pline’s future of a phylogenetically based taxonomy. The realists’
day, it is worth sampling a bit more. After debating the relative practical perspective was content with the intellectual pauper’s
significance and utility of morphological and physiological char- gruel provided by a taxonomy merely enabling species identifi-
acteristics in bacterial classification, the authors (tentatively) opt cation and a convenient pigeonholing system. Stanier and
for ‘‘. . .the use of morphological criteria as [the] main guiding van Niel deplored this mercantile mentality (Stanier and van
principle. . .above the rank of genera’’ (Kluyver and van Niel 1941):
Niel 1936). " In most biological fields it is considered a truism to state that the
It is basic to biological thought that the complex tends to
only satisfactory basis for the construction of a rational system of
evolve from the simple, and given the presumed primitive nature
classification is the phylogenetic one. Nevertheless, ‘‘realistic’’
of bacteria, Kluyver and van Niel carried this notion to the
bacteriologists show a curious aversion to the attempted use of
extreme:
phylogeny in bacterial systematics.... To what may we ascribe this
" It seems acceptable that the diversity of bacterial forms is the distrust of phylogeny? In part it is undoubtedly due to the
outcome of various independent morphological evolutions unsatisfactory nature of certain systems, purportedly based on
which have had their starting-point in the simplest form both phylogeny, which have been proposed in the past. However, the
existent and conceivable: the sphere (Kluyver and van Niel 1936, mere fact that a particular phylogenetic scheme has been shown
p. 387). to be unsound by later work is not a valid reason for total
rejection of the phylogenetic approach.
From this starting assumption, they picture the aboriginal
spherical bacterium as (somehow) giving rise to four basic The authors then turn to the drawbacks of the alternative:
morphological types: (1) coccoid (the Micrococcaceae), " . . .there is good reason to prefer an admittedly imperfect natural
(2) cells with polar flagella (Pseudomonadaceae), (3) cells with
system to a purely empirical one. A phylogenetic system has at
peritrichous flagella (Bacteriaceae), and (4) a nonmotile line
least a rational basis, and can be altered and improved as new
whose beginnings are streptococci. Each of the four primary
facts come to light; its very weaknesses will suggest the type of
lineages in turn evolves in a quasi-ontogenetic fashion through
experimental work necessary for improvement. On the other
stages of increasing complexity to some ‘‘highest stage of devel-
hand, an empirical system is largely unmodifiable because the
opment,’’ for example, the coccoid forms giving rise to Sarcinae
differential characters employed are arbitrarily chosen and
(packets of cocci), and hence Sarcinae that form spores, or the
usually cannot be altered to any great extent without disrupting
‘‘streptococcal’’ lineage developing through short Gram-positive
the whole system. . . [When the] wide separation of closely
rods to mycobacteria and ultimately the complex actinomycetes
related groups caused by the use of arbitrary differential
(Kluyver and van Niel 1936).
characters. . .makes it impossible to tell with certainty in
Within each main line of morphological descent, morphol-
what order a given organism belongs, an empirical system
ogy increasingly gives way to physiological characters in defining
loses its value....
the lower taxonomic levels (Kluyver and van Niel 1936). In
retrospect, unfortunately, bacterial classification in the early This is incisive commentary! But unfortunately, it is just
twentieth century comes to be little more than a tapestry of about the last one hears the idealist perspective, the last time
Just So Stories. the attitude ‘‘if it can’t be solved today, then we will try anew
tomorrow,’’ is expressed.
That final time seems to be 1946, when van Niel gave a major
The Beginning of the End address concerning bacterial classification at that year’s memo-
rable postwar Cold Spring Harbor Symposium. In the address,
We need to consider one final attempt to devise a natural he analyzed in detail the failures of current and past attempts at
bacterial system, which was proposed by Stanier and van Niel 5 natural bacterial classification. He pointed out the difference
years after the Kluyver and van Niel system (Stanier and van Niel between the successful use of morphological characters in
1941). Again our reason for doing so does not lie in the system plant and animal classification and their lack of utility in the
itself but in the discussion that accompanied it. Theirs was the bacterial case (as mentioned above). He detailed the frustrations
last blush of enthusiasm among microbiologists for developing bacteriologists had experienced in trying to use physiological
a comprehensive natural bacterial classification, and the characteristics for classificatory purposes, concluding that there
last time that having an evolutionary overview of the bacteria is no way one can determine natural relationships in the bacte-
of any kind was vigorously defended (Stanier and van rial world as things then stood (van Niel 1946). But at the end of
Niel 1941). this depressing critique, van Niel held firm to the idealist posi-
Bacterial taxonomists had long been split into two camps: tion: ‘‘[Thus, since] the morphology of a bacterium is of no
the idealists, represented especially by members of the Delft more use in [classification] than is its physiology. . .the search
School, and the realists, represented by the majority of other for a basis upon which a ‘natural system’ can be constructed
bacteriologists, for example, the board of Bergey’s Manual must continue’’ (van Niel 1946, p. 290).
Where to Play? The Sand Castles Have All between these two cell types, upon which rests our only hope of
Been Washed Away more clearly formulating a ‘‘concept of a bacterium’’ (Stanier and
van Niel 1962, pp. 20–21).
The year was 1955 when van Niel returned for the last time (to
In The Microbial World second edition, 1963:
my knowledge) to the subject of natural bacterial classification
(van Niel 1955). By that time, biochemists, following Kluyver’s " All [bacteria] share the distinctive structural properties associ-
pioneering synthesis of microbial biochemistry, had turned their ated with the procaryotic cell, and we can therefore safely infer
attentions in earnest to bacterial systems; sexual recombination a common origin for the whole group in the remote evolutionary
had been discovered in bacteria by Lederberg and Tatum (1946), past; we can also discern four principal sub-groups, blue-green
and geneticists were poised to take advantage, and (as men- algae, myxobacteria, spirochetes, and eubacteria, which seem to
tioned) molecular biologists too were finding bacterial systems be distinct from one another.... Beyond this point, however, any
highly suitable for their studies. The emphasis in bacteriology systematic attempt to construct a detailed scheme of natural
was definitely shifting strongly in the reductionist structure/ relationships becomes the purest speculation, completely
function direction. And van Niel returned to his topic, with unsupported by any sort of evidence. The only possible conclu-
a changed, jaded outlook. sion is, accordingly, that the ultimate scientific goal of biological
Van Niel’s still hopeful (and scientifically proper) perspective classification cannot be achieved in the case of bacteria (Stanier
of 1946—‘‘it hasn’t been done yet, so we must continue’’—had et al. 1963, p. 409).
now become the pessimistic (and scientifically unacceptable)—
A similarly dark and contrived picture was painted in
‘‘it hasn’t been done yet, so it can’t be done,’’ which he expressed
the third, 1970, edition of The Microbial World and carried
this way:
through essentially unchanged into the fourth (Stanier et al.
" What made Winogradsky. . .grant that the systematics of plants 1970, 1976).
and animals on the basis of the Linnean system is defensible,
while contending that a similar classification of bacteria is out of Here Comes the (Molecular) Cavalry
the question? The answer must be obvious to those who recog-
nize in the former an increasingly successful attempt at The sad irony in all this is that while the idealist position was in
reconstructing a phylogenetic history of the higher plants and retreat, with defeat masquerading as victory, and while new
animals. . .and who feel that comparable efforts in the realm of recruits were taught not to look back, as the importance of
the bacteria (and bluegreen algae) are doomed to failure a ‘‘concept of a bacterium’’ (and with it any interest in bacterial
because it does not appear likely that criteria of truly phyloge- evolution) slipped from view, the ground for a genuine phylog-
netic significance can be devised for these organisms (van Niel eny of bacteria, which could transform the discipline, was simul-
1955, pp. 101–102). taneously being prepared. In the early 1950s, Sanger had
The idealists were now in full retreat and in the process took sequenced the first proteins (Sanger and Tuppy 1951; Sanger
an untenable fallback position, namely, that the concept of and Thompson 1953), and the notion that molecular sequence
a bacterium could be developed without resort to phylogenetic comparisons were a rich source of evolutionary information was
relationships. It could be developed simply by knowing the beginning to take hold:
structure/function differences between eukaryotes and ‘‘pro- " Biologists should realize that before long we shall have a subject
karyotes.’’ A little thought shows the folly in this. As emphasized which might be called ‘‘protein taxonomy’’—the study of amino
above, an organismal science must be founded on evolutionary acid sequences of proteins of an organism and the comparison of
relationships. Yet, that is far from what the next generation of them between species. It can be argued that these sequences are
microbiologists were taught. In The Microbial World first edi- the most delicate expression possible of the phenotype of an
tion, 1957: organism and that vast amounts of evolutionary information
may be hidden away within them (Crick 1958, p. 142).
" An eminent contemporary bacteriologist, van Niel, who is noted
for his taxonomic studies on several groups of bacteria, has However, among microbiologists Crick’s prescience fell on
expressed the opinion that it is a waste of time to attempt deafened ears. The molecular and bacteriology paradigms were
a natural system of classification for bacteria, and that bacteriol- somehow on different conceptual planes. However, some
ogists should concentrate instead on the more humble practical ‘‘macrobiologists’’ did see the new molecular royal road to
task of devising determinative keys (Stanier et al. 1957, p. 296). phylogeny (Zuckerkandl and Pauling 1965), and a cottage
industry arose around confirming and (slightly) extending the
In Archives of Microbiology, 1962:
classical phylogenetic tree through sequence comparisons of
" It is now clear that among organisms there are two different cytochrome cs and of a few other molecules (Fitch and
organizational patterns of cells. . .the eucaryotic and procaryotic Margoliash 1967). Yet the record shows that even among these
type. The distinctive property of bacteria and blue-green algae is first-generation molecular evolutionists, there was no recogni-
the procaryotic nature of their cells.... The remaining pages of this tion of the importance of the larger and far more important
essay will be devoted to a discussion of the essential differences problem of using molecular sequence comparisons to infer
bacterial phylogenies, which would have opened the door to the The universal tree shows prokaryotes to comprise not one, but
universal phylogenetic tree. two, unrelated major organismal groups (primary lines of
Microbiologists did get on the molecular evolution band- descent), the Archaea and the Eubacteria. The new perception
wagon but belatedly and half-heartedly. In the main, their efforts of microbial diversity that came out of the universal tree, in
relied upon simple and relatively uninformative methods, such addition to the unexpected discovery of a third highest level
as nucleic acid hybridization (Gillespie and Spiegelman 1965), grouping of organisms, found biology organizationally
and their interests were confined largely to weeding out unprepared. Academic department structures, their organiza-
misclassified species from genera and in properly grouping tion and courses, did not and still do not reflect the new reality.
genera into families. A few protein sequences were attempted Most textbooks do not do so adequately even today. And the
and superficially analyzed, but those responsible saw nothing of funding priorities for governmental supported biology remain
evolutionary value in the comparisons (Ambler et al. 1979a, b; structured in a way that recognizes a bipartite, not tripartite,
Meyer et al. 1986)! The compelling vision of the idealists, the division of life.
overriding concern with building a concept of bacteria on
a comprehensive natural bacterial taxonomy, was gone. (Two
prominent exceptions to the above characterization of microbi- Reaching into the Unknown
ologists’ attitudes toward bacterial taxonomy appear to be
expressed by the Europeans J. De Ley and O. Kandler, in whose While rRNA phylogenies put bacteriology on the road to becom-
works and writings the candle of hope still flickered for a broad ing a full-fledged organismal discipline, they alone did
reaching, if not comprehensive, natural bacterial system not completely resolve bacterial ecology’s core problem: a
[Heberlein et al. 1967; Schleifer and Kandler 1972].) phylogenetic framework is a necessary but not sufficient condi-
tion for developing bacterial ecology. To identify a species, the
bacterial ecologist has first to detect and isolate it. Proper iden-
Cleaving the Gordian Knot tification of a bacterial species was firmly believed to require its
cultivation (in pure culture) in the laboratory. Yet the
It was only a matter of time before the sleeping giant of bacterial vast majority of bacterial species in the typical niche go
phylogeny would be roused by a dose of molecular medicine. By undetected, and among those that can be detected, most defy
the mid-1960s Sanger’s ‘‘oligonucleotide cataloging’’ method laboratory (pure) culture. (Just how severe a problem this was
had come along (Sanger et al. 1965). By 1970, my laboratory did not become apparent until relatively recently, well after
was using it on ribosomal RNAs (Sogin et al. 1971; Woese et al. the fact.)
1974). The universal distribution of rRNA, its large size, its high But, contrary to established belief, isolation of bacteria in
degree of sequence conservation, and a presumed refractoriness laboratory culture is not an absolute requirement for bacterial
of rDNA to horizontal gene displacement argued that the mol- identification and classification. Since bacteria can now be iden-
ecule could be used to derive a universal phylogeny (Fox et al. tified and related merely through the sequence of one molecule,
1977b; Woese 1987). By 1975, we had characterized about 30 rRNA, only that molecule needs to be isolated from the envi-
rRNAs (mainly bacterial) by the method (Woese et al. 1975), ronment to detect, identify, and classify organisms. This realiza-
a number that approached a 100 by the end of the decade. In the tion seems first to have occurred to Norman Pace, in the early
1980s, newer and faster methods permitted effectively full 1980s. Over the next several years, he and his coworkers
sequencing of an rRNA molecule (Lane et al. 1985), which published a series of articles pointing out in principle and
significantly sharpened branching orders in the universal tree. demonstrating in practice the new molecular sequence-based
Today, the sequencing of rRNA (or rDNA) has become trivial, approach to microbial ecology (Stahl et al. 1985; Olsen et al.
and the collection of rRNA sequences now numbers in the tens 1986). The crippling limitations that organismal detection and
of thousands (Maidak et al. 2001). laboratory isolation had previously imposed on microbial ecol-
The most important findings to come out of this new and ogy had vanished!
revolutionary approach to bacterial taxonomy (and taxonomy Pace and coworkers went further, showing that for certain
in general) were that, yes, phylogenetic relationships among purposes it was not even necessary to isolate a molecule (or its
bacteria (distant or not) could be determined, and so, develop- gene) from the environment: ‘‘phylogenetic stains’’ (sets of fluo-
ing a clear concept of a bacterium was now feasible, second in rescent nucleic acid probes for rRNA) could be developed to
importance only to the determining of the universal phyloge- identify individual cells microscopically in situ by species, genus,
netic tree itself (Fox et al. 1980)—one of the great problems that family, etc. (DeLong et al. 1989). Using this powerful new meth-
the nineteenth century in biology bequeathed to the twentieth. odology, a microbiologist could identify any and virtually all
As Darwin had said (Burkhardt and Smith 1990): ‘‘The time will microbial species in a given niche in terms of phylogenetic
come I believe, though I shall not live to see it, when we shall relationships and morphology (and that, obviously, is only the
have very fairly true genealogical trees of each great kingdom of beginning for such a methodology). At last, a genuine bacterial
nature’’ (Darwin Corresp. 6:456). The discovery of the Archaea ecology is emerging! It is impossible to overestimate the impor-
(Fox et al. 1977a; Woese and Fox 1977b) once and for all tance of phylogeny and direct detection of genes to bacterial
exploded the prokaryote/eukaryote phylogenetic myth. ecology’s development.
Perspective level. Yet in looking at the core functionality of the cell, the
enormous gulf between archaea and eubacteria leaps out.
Throughout its history, molecular biology has been guided in
one way or another by ‘‘the problem of the gene,’’ which basically
defined twentieth-century biology. What would molecular biol- Metabolism, Membranes, and Walls
ogy have been like had the gene not been its focus? I ask this
simply to provide perspective in posing a second question: what Let us begin a comparison of the two types with their metabo-
would bacteriology have been like if developing ‘‘a concept of lisms, wherein the differences are relatively few, but still signif-
a bacterium’’ had been its central focus and it had succeeded in icant. The eubacteria are far and away the most metabolically
doing so—allowing bacteriology to become a full-fledged organ- diverse and versatile group of organisms on the planet. The
ismal science? metabolic uniqueness of the archaea is most prominently
Early on, microbiologists were not averse to asking evolu- displayed in methanogenesis, a metabolism confined to the
tionary questions: are all bacteria at base specifically related to Euryarchaeota. The most interesting thing about archaeal
one another? Did bacterial evolution follow a course primarily metabolism may be the unique set of cofactors it employs
laid out by morphological or physiological development? Were (mainly in methanogenesis), for example, the C-1 carriers,
the first bacteria heterotrophs or autotrophs? Were they coenzyme M, methanofuran, and methanopterin; F420, an elec-
mesophiles or thermophiles? Are all photosynthetic bacteria tron carrier analogous to NAD; F430, a nickel-containing por-
related to one another to the exclusion of nonphotosynthetic phyrin akin to heme, vitamin B12, and chlorophyll; and
species? How did the many diverse bacterial metabolisms come methanophenazine, a membrane-bound carrier that functions
into being? How many times did this or that phenotypic feature like a quinone (Wolfe 1992; Beifuss et al. 2000). Viewed meta-
evolve? Does this or that common feature signify relationship or bolically, these cofactors are interesting but not all that unusual
convergent evolution? Unfortunately, all such questions (neces- in the reaction types they catalyze. What is unusual and intrigu-
sarily) were answered only by (untestable) conjectures—not ing, however, are their biosyntheses (the study of which is still in
a very reliable foundation upon which to base a concept of its beginning phases). While the pathways in archaeal cofactor
bacteria. A fabric of conjecture soon comes to be viewed as synthesis themselves tend to be fairly standard in a chemical
metascience (Stanier 1970), and this has the effect (most unfor- sense, the enzymes catalyzing the reactions more often than not
tunate in microbiology’s case) of squelching inquiry. In this way, have no homologs outside of the archaea. As suggested by
bacteriology came to abandon its evolutionary curiosity—much Graham and White (2002), the unity of biochemistry we have
to its and biology’s detriment. known since the time of Kluyver may be as much a matter of
Just as unsubstantiated conjecture leads to stifling curiosity, evolutionary convergence on common biochemical themes as it
phylogenetic knowledge would have done the opposite. Fairly is retention of ancestral metabolic ways.
reliable responses to the early microbiologists’ evolutionary Archaeal membrane structure is unique as well. It is based
musings would have led to the asking of new, more detailed upon ether-linked branched-chain lipids, whereas in the eubac-
questions. Had a reasonable natural bacterial classification come terial and eukaryotic cases, straight-chain ester-linked lipids pre-
on the scene early enough, the spirit of evolutionary inquiry dominate (Kates 1964; Kates et al. 1968; Tornabene and
would not have withered but intensified, and a genuine concept Langworthy 1979). Cell wall structures differ too. The eubacte-
of bacteria would naturally have followed. The development of rial wall characteristically comprises murein (peptidoglycan),
all twentieth-century biology would have been different, while archaeal walls are most often proteinaceous (Kandler and
retaining to some extent a focus on the organism as a whole Hippe 1977; Kandler and König 1985). In Methanobacteriales
and its evolution. Ah well.... species, however, walls comprise pseudomurein, a compound
that, as the name suggests, resembles eubacterial murein (König
and Kandler 1979). Yet the sugars from which murein and
Archaea and Eubacteria pseudomurein are built are largely different, and when the bio-
synthetic mechanisms involved in wall formation are considered,
As discussed above, the character of a prokaryotic type of cell is it is clear that murein and pseudomurein are another example of
expressed far more in physiology (dynamic pattern) than in biochemical convergence (Hartmann and König 1990).
morphology (static pattern), which underlies why bacterial mor-
phologies and the like are so uninformative (even deceiving)
phylogenetically. It also accounts for why the two prokaryotic Translation
cell types appear so similar upon superficial analysis. At the
molecular level, however, this impression of kinship between It is in information processing that the differences between
the two quickly evaporates. In their histories, the two prokary- eubacterial and archaeal cell designs appear in their full glory.
otic types have followed very different paths that from time to Of the three main information processing systems (translation,
time have crossed, leading to the occasional transfer of genes transcription, and genome replication), translation exhibits the
(especially early on), which accounts for much of the similarity most universality in its componentry, the most homology
often taken to signify specific relationship at the organismal between archaea and eubacteria (and eukaryotes). Ribosomal
RNAs, elongation factors, ribosomal proteins, and aminoacyl- componentry (Bell and Jackson 2001). Once again, the eukary-
tRNA synthetases are all basically universal. However, among otes possess an embellished archaeal version.
these universal protein components, a clear distinction exists
between the eubacterial and archaeal versions. The distinction is
characteristically so blatant that it tends to stand out even upon Genome Replication
gross inspection of a sequence alignment (Woese et al. 2000).
(For example, the eubacterial versions of a given sequence often Genome replication is the extreme example of archaeal/eubac-
contain moderate to large blocks of amino acids [located termi- terial difference. The closely related archaeal and eukaryotic
nally or in the interior of the molecule] not seen in the archaeal systems are to a first approximation totally distinct from their
versions and vice versa. In addition, for those columns in an eubacterial counterpart (Olsen and Woese 1996). And, as now
alignment where composition is highly conserved, one often sees might be expected, the mechanism of initiation of chromosome
that the characteristic eubacterial composition differs from that replication in the eubacteria is fundamentally different from that
characteristic of the archaea.) This nearly qualitative distinction basically common to the archaea and eukaryotes (Kelman 2000).
between archaeal and eubacterial versions of a sequence, this In the past, it was common to draw phylogenetic significance
difference in genre between the two, has been called the ‘‘canon- (i.e., specific relationship) from the fact that archaea and
ical phylogenetic pattern’’ (Woese et al. 2000). Differences this eubacteria both have circular chromosomes, in distinction to
extreme are never encountered among the various eubacterial eukaryotic chromosomes, which are linear. In light of the sim-
taxa or among different archaeal taxa (although in each case ilarity between the archaea and eukaryotes in both chromosome
there have been three billion years or so, most of this planet’s replication and the initiation thereof, that conclusion needs
history, in which such differences could have arisen). When revisiting.
eubacterial and archaeal sequences show these differences in It is also of interest that both the eukaryotic and archaeal
genre, their eukaryotic counterparts are almost always of the (euryarchaeal) chromosomes show nucleosome organization—
archaeal genre (Woese et al. 2000). the single archaeal histone responsible being a homolog of the
A minority of ribosomal proteins, however, are not univer- four (related) histones that structure the eukaryotic nucleosome
sally distributed. A relatively small cadre are characteristic of and (Reeve et al. 1997). This is an example of a theme that repeatedly
found only in eubacteria, while a somewhat larger set is common occurs in the chromosome replication mechanism and else-
and confined to the archaea and eukaryotes, with yet a few others where, namely, a job is done by a particular multimeric protein
being found only in eukaryotic ribosomes. Overall, though, complex. The many subunits therein, however, represent only
translation gives the impression of a system that existed in one or a few gene families. In the eukaryotic case, a number of
basically modern form at the root of the phylogenetic tree. different members of a family tend to be represented in the
complex, each present in single copy. In the archaeal case, there
tends to be only one member of the corresponding family and
Transcription that present in multiple copies.
Comparisons between the information processing systems
Transcription presents a similar evolutionary picture but with of Archaea and Eukarya are what give rise to the strong impres-
more, and more pronounced, exceptions to universality (Langer sion of fundamental relationship between the two. However,
and Zillig 1995). The two largest (the catalytic) subunits of the that specific relationship is reflected in relatively few molecules
DNA-dependent RNA polymerase, b and b0 in eubacterial other than those involved in information processing, one exam-
nomenclature, are clearly universal and show the canonical ple being HSP-60. Overall, the relationships among the three
pattern (with again eukaryotic sequences being of the archaeal basic cell types present a complex, fascinating conundrum—a
genre). But that’s about it for homology. For the remaining main delightful challenge.
subunit of the eubacterial RNA polymerase (a), homology The close resemblance between archaeal and eukaryotic ver-
between the eubacterial and archaeal versions is only partial. sions of the three information processing systems has been the
Eubacterial a occurs in two copies in the holoenzyme, whereas source of much speculation as to the origin of the eukaryotic cell,
its archaeal counterpart (found also in eukaryotes) comprises speculation that usually takes the form of invoking a fusion of a
two separate proteins of very different size, each present in single particular bacterial cell (such as an a proteobacterium) with a
copy in the holoenzyme, with parts of each showing homology particular archaeal cell (say, a methanogen; Martin and Mueller
to (somewhat) different parts of eubacterial a (Langer and Zillig, 1998) or sees some hypothetical ancient ‘‘protoeukaryotic’’ cell
1995). The archaeal transcription polymerase has a number of as horizontally acquiring certain metabolic capabilities from the
additional (smaller) subunits, all of which occur in the eukary- eubacteria and information processing capabilities from the
otic enzyme(s), but not in the eubacterial. And as in the case of archaea (Hartman 1984). I do not like either type of explanation.
translation, the eukaryotic version(s) of the transcription poly- The first type gives the eukaryotic cell no character of its own
merase contain(s) a few additional subunits specific to them- before it emerges from the fusion of the two prokaryotic types.
selves (Tjian 1996). When it comes to transcription initiation, In the second, the protoeukaryotic cell suddenly acquires infor-
one sees no homology between the eubacteria and the mation processing systems that it has never before encountered,
archaea: the two mechanisms are different and use different and its basic cellular design is suddenly altered. All such models
miss the essential point in cellular evolution: the core machinery phylogenetic range of many of the alien displacements that have
of the three primary cell types evolved and became established in occurred (Woese 2000). It is common, for example, to see the
each case well before any cells achieved their ‘‘modern’’ type of archaeal genre of a given aminoacyl-tRNA synthetase in one or
organization (Woese 2002). Evolution does not proceed by a few eubacterial lineages (Woese et al. 2000).
suddenly and drastically altering a given cell design (at least A detailed phylogenetic analysis of the 20-odd aminoacyl-
a fairly advanced one). These models ignore the possibility tRNA synthetases (Woese et al. 2000) shows (1) that about two-
(likelihood) that a cell design is stable, homeostatic, in an thirds of the tRNA charging enzymes exhibit the canonical
evolutionary sense. pattern typical of the other components of the cellular informa-
tion processing systems and (2) that in all these cases the canon-
ical pattern has to one extent or another been eroded by HGT,
Horizontal Gene Transfer: How Cells Evolved but (3) that the degree to which and the ways in which that
pattern has been eroded are idiosyncratic, qualitatively different
It might seem that horizontal gene transfer (HGT) would be for each of the enzymes. From these facts, it was concluded that
a complicating factor in understanding how the various major the canonical phylogenetic pattern itself was not produced by
cell types evolved. Yet, far from being a complicating factor, HGT but reflects some earlier evolutionary dynamic (Woese
HGT turns out to be the essence of the process (Woese 2000). et al. 2000). Thus, the canonical pattern represents a (partial)
As is now well known, HGTappears to have had (at certain times genetic record of organismal histories that extends back to the
in evolution) the potential not only to introduce entirely new root of the universal tree. In other words, while HGT has been
functions into cells but also to displace any constituent (gene) in a powerful, pervasive, and shaping force in organismal evolution
the cell with an alien equivalent. What effect has HGT had on the (especially early on), it has not completely obliterated the
genetic record of organismal descent? What is its relationship to ancestral organismal trace that the universal tree represents.
canonical pattern? Understanding here pivots on what the char- As the reader can see, I cannot discuss similarities and
acteristics of HGT are and what factors determine them. differences among the major organismal types without the dis-
The interested reader will find a more extensive and useful cussion shifting into an evolutionary framework. This, however,
discussion of the major problem of evolving cellular organiza- is no peculiarity of mine but rather the fact that such matters
tion and its relevance to the universal phylogenetic tree structure cannot be productively discussed in any other context. So, like it
in Woese (Woese 2000, 2002). or not, microbiology is going to be in the center of evolutionary
The most obvious thing about HGT is its extreme variability, study in the future—and vice versa.
in both frequency and phylogenetic range, from one gene to
another, from one major taxon to another, and likely from one
evolutionary stage to another (Woese et al. 2000). It turns out Microbiology on the Move
that the dominant factor shaping HGT is the organization
(design) of the potential recipient cell (Woese 2002). The various A great deal has happened in bacteriology in the last decade or
components of a cell are in one way or another, to one extent or so. The four pillars upon which a true bacteriology must be built
another, interconnected, variously integrated into the cell’s are now in place, and the long sought ‘‘concept of bacteria’’ is
‘‘design fabric.’’ Some of them are weakly and simply integrated beginning to take shape. Bacteria are now routinely seen in terms
into the fabric, others tightly and complexly so (Woese and Fox of their natural (phylogenetic) relationships; discussions often
1977a; Woese et al. 2000). Components of the first type are in turn on various aspects of their evolution. The sleeping giant of
effect modular: functionally and structurally, they are pretty microbial ecology has finally awakened, and we are beginning to
much self-contained and self-defining; that is, their interactions see the microbial base of the biosphere. Geomicrobiology, which
with the rest of the cell componentry are minimal. It stands to increasingly ties bacteria to geological processes, is a burgeoning
reason that such a modular, loosely coupled, component is far field. Marine microbial ecology is bringing to the fore the
more likely to undergo horizontal gene displacement by some important role that bacteria play in determining the character
alien equivalent than is highly and idiosyncratically integrated of the oceans. Genomic microbiology is beginning to tackle
componentry. some of the knotty problems associated with bacterial evolution,
An ideal system for assessing the character of HGT is the adaptation to environments, and the evolution of the global
translation apparatus, for while most of its componentry, for ecosystem itself. These and similar developments are the poten-
example, the ribosomal proteins and elongation factors, are tial building blocks of a new microbiology. But the concrescence
tightly coupled into the rest of the mechanism (and hence the of twenty-first-century microbiology will not occur unless we,
cell fabric), the tRNA charging enzymes (aminoacyl-tRNA syn- its practitioners, understand what bacteriology is and, thus, what
thetases) are not; the latter are modular and are loosely coupled its future goals are.
into the cellular fabric. And as might then be expected, the To me, microbiology today resembles a peasant who has
ribosomal proteins and elongation factors have been largely suddenly become wealthy. The chains of poverty are gone, but
unaffected by HGT, whereas the aminoacyl-tRNA synthetases the cobwebs that are the habits of poverty remain and hold him
have been quite highly buffeted by it, both in terms of the fast. Obviously, there are no longer scientific barriers to
frequency with which HGT has affected them and the (broad) bacteriology’s becoming the organismal discipline it should
long ago have become. But, who cares anymore? The idealists The View from the Top
among us, whose crucial role it had been to define and develop
the character of bacteriology, to keep the discipline on track, are We have discussed above the major features that distinguish
long gone. The character of bacteriology (like that of the peas- the Archaea and Eubacteria. What remains to be done is
ant) is essentially defined by overlords, external circumstances, provide an organismal overview of the individual domains
be they scientific or practical (societal). Contrast bacteriology to (see > Chaps. 7–9, in Vol. 1). We will go about this in a
disciplines such as genetics and molecular biology, disciplines top-down fashion, starting with the major eubacterial and
that once were strong in being defined from within—disciplines archaeal taxa, leaving the details and definitions of the lower,
whose paradigms embodied major scientific goals, disciplines subordinate taxa to you and the next generation of microbiolo-
that had a grand vision. Microbiology was never that. Today’s gists. I do not claim that these high-level taxa as currently
microbiologists are, unfortunately, the realists of yesteryear. The defined are entirely correct (are genuine naturally defined
field is without overview, without a concept of itself, and its groupings) in all cases, but most of them surely are, and, there-
goals remain shaped by practical considerations and influences fore, our discussing them at this point is useful. Let us start with
from the stronger disciplines. Only an agonizing awareness of the following:
the situation will change that.
The Organismal Structure of the Archaea

Where Is Our Essence?
As we have seen above (and elsewhere), the Archaea represent
It is time to take stock. Microbiology today is not the discipline a unique cell design (Graham et al. 2000). In their cellular
microbiologists knew 30 years ago. Everything is new—the tech- organization, they stand apart from the Eubacteria and, of
nology, the problems that the discipline faces, our training, our course, the Eukarya. The Archaea are also unique in their overall
conceptual framework, and the way microbiology is perceived phylogenetic structure: the domain comprises two characterized
by the other sciences. Only the organisms remain the same. But kingdoms, the Euryarchaeota and the Crenarchaeota—with
is this last even true? The organisms per se may be the same, but the possibility of a third kingdom, the ‘‘Korarchaeota,’’
our perception of them surely isn’t. What is a bacterium to us lurking among the unwashed masses of species that have been
except a synthesis of the various ways in which we experience it? identified only through direct isolation of rRNA genes from
That surely has changed, dramatically so. (Is one’s concept of the environment (Barns et al. 1996). Even a fourth archaeal
Bacteroides and Cytophagas, for example, the same as it previ- kingdom is suggested by the recent discovery of a bizarre
ously was once one knows that these two phenotypically dissim- nanosized exosymbiont associated with particular ones of the
ilar types are actually specifically related to one another? What Crenarchaeota (Huber et al. 2002). What makes archaeal
are Deinococcus and Thermus doing in the same stable? Are phylogeny stand apart from that of the Eubacteria and Eukarya
the myxobacteria that we know today as members of the is the pronounced phenotypic and genotypic distinction between
d-proteobacteria the same as the classical myxobacteria, which the kingdoms Euryarchaeota and Crenarchaeota—the depth of
if they had any affiliation at all, it was to the cyanobacteria?) their phylogenetic split, the genomic differences between them.
The only things that remain at all constant are the prejudices Interkingdom differences of this magnitude are not seen in the
we have inherited from our intellectual forbears. Time to clear other two domains.
the table! The Euryarchaeota appear phenotypically constituted
There are two fundamental challenges that microbiology (designed) around a methanogenic metabolism. They are also
faces today, challenges deep and basic enough that they speak characteristically thermophilic and show relatively high intra-
to the whole of biology. First is the age-old question ‘‘where do cellular concentrations of monovalent ions. To our present
we come from,’’ which can now be addressed in the form of how knowledge, the kingdom has spawned four main methanogenic
cellular organization evolved. The other is the problem of the branches (phyla), the Methanococcales, the Methanobacteriales,
biosphere: what is this incredibly complex yet smooth function- the Methanomicrobiales, and the Methanopyrales (Olsen et al.
ing web of interactions among this incredibly diverse collection 1994).
of organisms—this coherent, self-sustaining, overarching state The euryarchaeal kingdom also harbors other, non-
of the whole? Actually these two problems are at base the same. methanogenic, phenotypes. Interestingly, all but one of these are
And the understanding of both lies in the microbial world, in its offshoots of the lineage that gives rise to the Methanomicrobiales.
diversity, in the (functional) structure of microbial communi- The non-methanogenic phenotypes in question are the extreme
ties, in the organization of microbial cells, in evolutionary and halophiles, the sulfate-reducing archaea (Archaeoglobales), and the
other adaptive responses of microorganisms to environmental Thermoplasmatales. The lone non-methanogen group not stem-
perturbations, and in the incredible interconnectedness of ming from this lineage is the Thermococcales (which comprises the
microbial life in general. In a way, twenty-first-century micro- genera Thermococcus and Pyrococcus).
biology is a return to microbiology’s roots, a rediscovered and The extreme halophiles are facultative aerobes that grow in
development of the grand view of microorganisms that highly saline environments and have been known to microbiolo-
Beijerinck and other early microbiologists shared. gists for ages: they are known for putrefying salted fish among other
things. The extreme halophiles are also notable for the reddish methanogenesis pathway in both directions—which they do in
pigment(s) they produce, which are in part responsible for the metabolizing acetate. Interestingly, their genomes are quite large
striking red color in salt evaporation ponds. Among these pigments by archaeal standards (Deppenmeier et al. 2002; Galagan et al.
is bacterial (now archaeal) rhodopsin, which provides them 2002)—and contain a number of non-archaeal genes!
a photochemistry. Archaeal rhodopsin is central to a light-driven Among the euryarchaeotes, one finds both autotrophic and
transmembrane pump, by means of which the extreme halophiles heterotrophic organisms (which is typical of large organismal
can derive energy. groupings, of course). In the present case, the phylogenetic
Beyond their extremely high (5 M) intracellular potassium arrangement and genomic analysis of species strongly suggest
ion levels and their photochemical ability (unique among cul- that the heterotrophs are derived from the autotrophs, not the
tured bacteria), the extreme halophiles offer little of interest, reverse.
unless it be the fact that as an organismal group, they are The Crenarchaeota, the other of the (classically recog-
phylogenetically very tightly clustered—and so, of relatively nized) archaeal kingdoms, cannot be usefully characterized
recent origin. at this point in time. The reason is twofold. For one, the
One naturally assumes, given the specific (though somewhat group has been relatively poorly studied in the laboratory.
distant) phylogenetic relationship to the Methanomicrobiales, For another, the crenarchaeal species isolated in laboratory
that the metabolism of the halophiles is probably a derived, culture form only a phylogenetically restricted and nonrepre-
degenerate form of the metabolism characteristic of the sentative subset of the kingdom as a whole: direct environ-
Methanomicrobiales. But, the halophile genome shows some- mental gene probing has shown that the crenarchaeal tree
thing different. It is rife with bacterial genes and others that have contains many branchings that are far more deeply placed
no relationships functionally or phylogenetically to than those of the cultured crenarchaeal species (DeLong
methanogens and methanogenesis. 1992; Barns et al. 1994). I will leave you with this impression:
The second grouping in the Methanomicrobiales cluster Crenarchaeota are basically thermophilic organisms whose
comprises the anaerobic sulfate-reducing organisms of the fam- metabolisms somehow feature sulfur prominently. They are
ily Archaeoglobales. Iron reduction is seen in the genus basically anaerobic, though can be facultatively aerobic in
Ferroglobus (Hafenbradl et al. 1996). The members of this some cases. And, as an evolutionist might expect, mesophilic
order, though anaerobic, produce no methane, except for (even psychrophilic) phenotypes also exist. Indeed, one of
a trickle thereof. The biochemical reason is obvious: the gene the major phenotypes in the oceans—the largest biotope on
for the terminal enzyme in the methanogenic pathway, methyl the planet—represents one particular crenarchaeal group, the
CoM reductase, is missing from their genomes. Yet the rest of the Order Cenarchaeales, which abounds throughout the breadth
methanogenic pathway appears intact in these organisms. How- and depth of the oceans, being remarkably in evidence in
ever, the Archaeoglobus sp. runs the pathway backward (oxida- Antarctic waters (DeLong 1997).
tively), producing carbon dioxide as the end product. The I have one closing thought concerning the archaea. The
Archaeoglobales are highly thermophilic, with no known excep- depth of the phylogenetic split between the two archaeal king-
tions, which distinguishes them from their sister lineages, the doms seems the tip of an evolutionary iceberg. We have much to
Methanomicrobiales and extreme halophiles, lineages that do learn evolutionarily from this group of organisms. It saddens me
not appear to harbor hard-core thermophiles. to see how little interest goes into studying the archaea as an
The final aberrant phenotype within the Methanomi- organismal group. And the fact that there exists an entire
crobiales segment of the euryarchaeal tree is the archaeal kingdom, the Korarchaeota, not a single representative
Thermoplasmatales. They are ostensibly wall-less thermophiles of which has been isolated, is doubly distressing.
(unique among the archaea), whose thermoacidophilic pheno-
type superficially resembles that of the Crenarchaeota. And
again, iron metabolism surfaces in the genus Ferroplasma The Organismal Structure of the Eubacteria
(Golyshina et al. 2000). Direct environmental sampling for
rRNA genes reveals that the Thermoplasmatales also encompass The eubacteria are the consummate metabolists: sheer metabolic
a low-temperature marine grouping (DeLong 1992)—whose power and versatility. They are the masters of metabolic theme
phenotype is otherwise unknown. In their genomic makeup, and variation—just as eukaryotes are masters of structural
these organisms are the most phylogenetically cosmopolitan of theme and variation. Someday, we will understand just how
all the archaea. eubacteria could have acquired such a varied metabolic reper-
I find it intriguing that among all Euryarchaeota, it is only toire. Their versatility is the more interesting when one realizes
the Methanomicrobiales (and their ancestral lineage) that have (from genomic analysis) that much of the more limited meta-
proven so evolutionarily inventive. In addition to spawning the bolic repertoire of the Archaea beyond that characteristic of the
above-described aberrant phenotypes, the Methanomicrobiales methanogens and methanogenesis is (as alluded to above) the
proper have been inventive in terms of methanogenesis itself. product of metabolic capabilities (enzymes and pathways)
These are the only methanogens that utilize certain carbon imported from the eubacterial world.
sources other than carbon dioxide (e.g., acetate) in The phylogenetic structure of the Eubacteria is rather differ-
methanogenesis. And they seem to be able to utilize the ent from the archaeal one just discussed. Viewed from afar, the
main eubacterial lineages appear to radiate out from a ‘‘crown exploration of this interesting eubacterial phylum to the tastes
group,’’ with a few atavistic forerunners branching from the of the reader—but not without first mentioning an evolutionary
eubacterial stem slightly earlier (examples being the Aquificales peculiarity of the a-proteobacteria. The a-phylum appears
and the Thermotogales). (It is the crown radiation that spawns (anecdotally) to be (to have been) more prone to the horizontal
the phenotypic richness we characteristically associate with the import of genes than are most or all other proteobacterial phyla,
Eubacteria.) as judged by the number of archaeal genre aminoacyl-tRNA
Among the major (kingdom level) eubacterial taxa—those synthetase genes they contain (Woese et al. 2000). A systematic
dominant by virtue of numbers, phylogenetic diversity, and study of HGT in the a-phylum is definitely called for, especially
ecological impact—there exist several kingdoms and phyla an accounting of when in the phylum’s evolutionary history the
about which we know next to nothing. Only a few, if any, bulk of these transfers occurred.
cultured species represent them, their evolutionary breadth Little more needs be said than already has been about the
and ecological prominence being demonstrated by the large b-phylum. The photosynthetic phenotype therein (represented
numbers of environmentally derived rRNA sequences from the in several lineages) is unlike the photosynthesis found in the
various groupings (Pace 1997). The lesson that has yet to sink in g-phylum (the latter are the purple sulfur bacteria) but akin to the
is that microbiologists know relatively little about the bacterial purple nonsulfur types seen in the a-phylum. The b-phylum is
world. We clearly have no useful idea of what some of the major again a haven for pathogens but also for a varied collection of
environmentally important bacterial groups are and are doing. metabolisms. I am inclined to believe, on the basis of the relative
You would think that a great deal of work would now be going uniqueness of the characteristic b-rRNA sequence, that the
into remedying this situation. This is not the case, however. Ask grouping may have begun as a rapidly evolving offshoot of the
yourselves: do microbiologists have better things to do? g-lineage.
Let us begin with the major (most significant) eubacterial The g-proteobacteria do live up to the implication of the
kingdoms and phyla and then, when appropriate, extend to their designation ‘‘proteo-’’; a great deal of morphological and other
subordinate classes. I more or less lump kingdoms and phyla phenotypic variation is housed therein. Thus, one can make few
here, because—until the proper genomic criteria have been generalizations about the phylum as a whole. The purple pho-
developed to define taxonomic rank on the basis of the proper- tosynthesis characteristic of the g-phylum is, as just mentioned,
ties of naturally defined organismal groups—it is not in some biochemically unique, and the lineages harboring it, the genera
cases possible to distinguish the two. Chromatium, Ectothiorhodospira, and closely related kin branch
early from the g-stem. These photosynthetic g-proteobacteria
deserve far more genomic attention than they have been given
The Proteobacteria to date. I stress this because if photosynthesis is ancestral to the
proteobacteria, at least to the a-b-g cluster therein, then
One of the most, if not the most, ecologically dominant eubac- a genomic understanding of the photosynthetic representatives
terial kingdoms is the Proteobacteria. The kingdom comprises of the kingdom should prove key in the longer term to an under-
five phyla, distinguished by the Greek letters a e. The a, b, and g standing of the group’s evolution.
phyla obviously cluster, which easily emerges by any number of This leaves the d- and -phyla of the proteobacteria. Again,
different phylogenetic analyses, while d and e, which may be a lack of significant genomic coverage of this assemblage—
specifically related, stand apart—but still within the except, of course, for a notable preoccupation with the patho-
proteobacterial kingdom. The b-phylum is actually a major genic helicobacteria—leaves little to be said. Perhaps the most
branching within g. I still consider b to be of phylum rank significant phenotypic characteristic of the group is a lack of
because of the apparent phenotypic diversity within it and its photosynthetic representatives. The reduction of sulfur com-
branching from a section of the g-tree which, except for the pounds is a predominant theme among the d-lineages. One
b-branch, contains no photosynthetic representatives. (A taxo- would like to see a representative genomic sweep of the
nomic precedent for such a ranking is given by the class of d-phylum and some genomic emphasis on the nonpathogenic
mammals arising from within the class of reptiles.) members of the -phylum, such as Wolinella and Thiovulum.
The a-phylum to a first approximation comprises (purple) Mention must be made, of course, of the myxobacteria,
photosynthetic lineages, nonphotosynthetic derivatives thereof, which are one of the four major classically recognized bacterial
and a variety of pathogenic groups, mainly intracellular patho- groups (see quote from The Microbial World second edition,
gens. The most notable of the latter is the rickettsial group above). Under molecular scrutiny, however, the myxobacteria
(which comprises the genera Rickettsia, Wolbachia, Ehrlichia, turn out to be subordinate to, a lineage within, the d-
Anaplasma, and Cowdria). What seem to be free-living relatives proteobacteria. Yet, their complex and beautiful morphologies
of these pathogens have been detected by rRNA analyses from and complicated life cycles give cause to wonder. How could
marine habitats, although none have been cultured. such morphologically rich and metabolically versatile entities
While the closely related agrobacteria and rhizobacteria, phylogenetically reside among the otherwise morphologically
which play such important roles in the plant kingdom, are not drab d-proteobacteria? The genomes of the myxobacteria are
members of the rickettsial group proper, the mitochondria are inordinately large, much larger than other d-proteobacteria.
(Gray et al. 2001). At this point, I’m going to leave further Within their excess DNA must lie the answer.
The Gram-Positive Eubacteria a major evolutionary radiation, open a vast new evolutionary
world for exploration?
This ecologically and phylogenetically major eubacterial It should be noted that chlorophyll g, characteristic of the
kingdom is one of the few classical eubacterial groupings to Gram-positive photosynthetic heliobacteria, is the closest in
survive relatively unscathed the molecular phylogenetic assault structure to the cyanobacterial chlorophylls of all the eubacterial
on bacterial taxonomy. A fair bit of taxonomic reorganization chlorophylls. Tantalizing molecular clues have also suggested
has occurred within the kingdom, and some new members a specific relationship between the Gram-positive eubacteria
that do not possess typical Gram-positive walls have been and the cyanobacteria—none as yet convincing, however.
added. But the two classical traits, endospore formation and
Gram-positive wall, turn out to be phylogenetically valid
characteristics. The only real taxonomic surprise has been the The Chlorobium-Cytophaga Kingdom
addition of the photosynthetic heliobacteria to the group
(Woese et al. 1985a). This grouping and the drawing together of the Cytophaga-
The main phylogenetic split among the Gram-positive Bacteroides phylum (by whatever name) is one of the more
eubacteria separates the ‘‘high G+C’’ group (the actinomycetes, significant triumphs of rRNA-based molecular phylogeny of
mycobacteria, and a number of others) from the ‘‘low G+C’’ the bacteria. As one can infer from the glut of generic names
group. The phylogenetic split between the two is great enough that constitute the phylum (Cytophaga, Flexibacter,
that their specific relationship might be questioned. While the Flavobacterium, Microscilla, Bacteroides, and a host of lesser
coherence of the entire kingdom was apparent in phylogenetic known genera), the grouping lay in separate taxonomic pieces
analyses of the small-subunit rRNAs (Olsen et al. 1994), it before the advent of this molecular approach. The bringing
cannot be reliably demonstrated in my experience using the together of the aerobic cytophagas and their ilk with the anaer-
large subunit rRNAs. What is called for here is an appropriate obic bacteroides was an impressive join. Interestingly, the phy-
test of the coherence of this kingdom based upon genomic lum does not comprise two major classes, split along aerobic/
analysis of the proper Gram-positive genomes—key ones of anaerobic lines. The bacteroides cluster forms a rather superfi-
which have yet to be sequenced, however. cial branching from the main stem of the phylum and has close
While most eubacterial genomes are sequenced for medical specific relatives having a cytophaga phenotype. (None of the
reasons—and this is surely true in the Gram-positive king- latter has piqued the interest of genome sequencers yet.) I would
dom—one would still not single out the kingdom as an evolu- note that the rRNA analysis reveals the coherence of this phylum
tionary training ground for pathogens. Metabolic versatility is very clearly: the members of the group exhibit a clean rRNA
its hallmark. The clostridia (not a genus or family, but signature (Woese et al. 1985b).
a phylogenetically widely dispersed agglomerate of many differ- The specific relationship between cytophagas and the green
ent low G+C Gram-positive lineages) are classically notorious photosynthetic Chlorobium species (the latter an unexpectedly
for the variety of degradative biochemistries of which they are tight phylogenetic cluster for a phylum) was only weakly suggested
collectively capable. Several clostridial genomes have now been by the small-subunit rRNA comparisons. To demonstrate the rela-
sequenced, but many more are needed, and I would hope chosen tionship convincingly, we had to resort to sequencing the large
in a systematic, phylogenetically comprehensive manner. subunit rRNA (Burggraf et al. 1991). And although expected, it is
most reassuring to see the coherence of the kingdom spelled out
in full genomic detail by the comparison of the genome of
The Cyanobacterial Kingdom a Chlorobium species (Eisen et al. 2002) with that of
Porphyromonas, a member of the bacteroides contingent (Paster
This grouping is one of the very few high-level eubacterial taxa et al. 1994). In its entirety, this kingdom, with its spectacular
that are anywhere near phenotypically uniform. Morphological admixture of phenotypes, demands some sort of evolutionary
variety abounds (as expected for a high-level taxon), but chlo- rationalization. So, as you might by now expect, I would encour-
rophyll-based photosynthesis appears a constant metabolic age a concerted, systematic genomic assault on the problem.
theme in all. Classically isolated cyanobacteria contained only
chlorophyll a, whereas the chloroplasts of plants, which are
domesticated cyanobacteria, contain both chlorophylls a and b. The Spirochetes
Cyanobacteria that contain both chlorophylls have more
recently been found (Lewin 1984). Although the chloroplast Since bacteriology’s beginnings, this kingdom has been recog-
lineage(s) traces to the general cyanobacterial cluster, no lineage nized as a major taxon (again see above quote from the second
representing a free-living cyanobacterium shows a specific rela- edition of The Microbial World). The unusual morphology of
tionship to the chloroplast lineage. Indeed the cyanobacterial these organisms together with their unique mode of locomotion
phylogenetic tree has a remarkably bushy appearance. Why this makes them stand out among all bacteria. The kingdom is phe-
is so is again the type of question that an evolutionarily oriented notypically (morphologically) pure, monolithic—with one minor
bacteriology finds interesting. Did the invention of oxygenic exception, an organism originally described as mycoplasma-like
photosynthesis in the ancestor of the whole group kick off but otherwise uncharacterized (K. Stetter, unpublished
BACTERIA
Planctomyces
m
mitochondrio
eriu
act
Rhod erichiafovib
p
vob
Agrobacterium
w tem
Esc
ter
Me
Fla
ocyc
h
ibac
us
ch
th
Th marine lo
Sy
Methanococcus
lob
a
ano
lus
es las s
lor coc
Therm um
Ar lasm
n
Flex
ec
og
ul
op cu
ax
ho
bac
ae
Ch er
op
Gl f
ch
lam oe M lo
ococc
er m
teri
et Ha
t
o
rio
yd ha
Chlo ia bac no
ter
robiu th spirillum
Me Methano
us
m er
Leptonema tha m
no us
py marine Gp. 1 low temp
Clostridium rus Gp. 1
pS p.G low te
m
Bacillus pSL L 12 2 low tem p
m 22
teriu r Gp. p
l l o bac c te 3l
He ba 9 Pyr Sulfoloow tem
r t hro PS1 odi bu p
A pO Th ctiu s
s m
Th pSL
us u ga Roo er
6
m ARCHAEA
uif S6
lex herm oto t
er 50
f op
m
o T
Aq pOP
m ro
lor er
of
EM1 ex
te
pJP 7
pJP 2
Ch us
ilu
Th
7
m
78
Cop
H
rinu
om
o
s
Zea 0.1 changes / site

Cryptomonas
lya
Ach aria
t
Babesiam a
s
eci hyr
Co
rp
u
Po
G
la
am
rd
lium
ia
Par
Dictyoste
ba
Phy
Tri
Entamoe
Naegleria
ch
sar
om
Euglena
um
Trypa
on
Enc
EUCARYA
as
Va
epha
no
irim
soma
orp
lltoz
ha
oon
. Fig. 1.1
Universal phylogenetic tree based on small-subunit (SSU) rRNA sequence. Sixty-four rRNA sequences representative of all known
phylogenetic domains were aligned and a tree was produced using FASTD NAML. That tree was modified, resulting in the composite one
shown, by trimming lineages and adjusting branch points to incorporate results of other analyses. The scale bar corresponds to 0.1
changes per nucleotide
observation) except for an rRNA sequence that places it clearly a lineage of aerobic spirochetes, the leptospiras and their rela-
among the spirochetes (Maidak et al. 2001). tives (some pathogenic).
The spirochete grouping is deeply split phylogenetically into At some early point in the evolutionary history of the spiro-
two major lineages, phyla. The one encompasses the more typical chete kingdom (exactly how early cannot be pinned down yet
spirochetes such as Spirochaeta halophila (and houses the spiro- because genome sequences are not available for the leptospira
chete pathogens to which so much genomic attention has been lineages), the spirochetes experienced a considerable amount of
paid, i.e., Borrelia and Treponema). The other is basically HGT. This assertion, however, is based on a very small sampling
of genes, namely, the aminoacyl-tRNA synthetases. The spiro- The Remaining Eubacterial Kingdoms and Phyla
chetes have replaced more of their ancestral aminoacyl-tRNA
synthetases by synthetases of the archaeal genre than has any With the above, I have come nowhere near to covering the full
other eubacterial group (Woese et al. 2000). range of eubacterial kingdoms and phyla. However, those that
remain fall into one or both of two categories (1) demonstrated
to be ecologically and phylogenetically major groupings by
The Planctomycetes, Chlamydiae and direct rRNA analyses of niches, but not represented by any (or
Verrucomicrobia very few) cultivated examples, and (2) represented by few
directly isolated rRNA clones and no or few cultivated species.
These three phenotypically disparate and phylogenetic distant These bear little comment at this point in time. A phylogenetic
groupings are combined here because fragmentary tree (> Fig. 1.1) generated by N. Pace and colleagues gives some
(unpublished) evidence suggests that they may well turn out to idea of the overall scope of the major eubacterial lineages (Pace
be specifically related to one another (at the kingdom level) when et al. 1997).
the appropriate genomes are available. Only the chlamydiae have I would end this brief, incomplete survey of the major
been given significant attention to date (again for reasons of their organismal groupings by mentioning two eubacterial kingdoms
pathogenicity). The reader is invited to peruse this third edition that have rather limited ecological distribution but are
of The Prokaryotes for evidence of this relationship. represented by genome sequences. Both are among the deepest
As their name suggests, the Planctomyces were originally branching eubacterial lineages (see > Fig. 1.1). The first is
thought to be eukaryotic. Their walls, as are the walls of chla- represented by Aquifex (Huber et al. 1992; Deckert et al. 1998).
mydiae and verrucomicrobia, are not typically eubacterial. It is The second comprises what are now called the Thermotogales
unfortunate that so few examples of the verrucomicrobia exist in (Huber et al. 1986; Nelson et al. 1999). The number of generic
culture, for in nature they are clearly a major and ecologically names in the group, Thermotoga, Geotoga, Petrotoga, and so on,
important group of organisms. The planctomyces and relatives does suggest a somewhat varied ecological distribution for the
also appear to be playing a significant environmental role(s), various members of the kingdom.
although perhaps not at the level of the verrucomicrobia. Now, on to the feast that awaits you in the rest of this third
It has recently been discovered that the Prostheocobacterium edition of The Prokaryotes.
subgroup of the verrucomicrobia carries two tubulin genes, the
only known bacterial sources of these molecules, heretofore
considered unique and essential components of the eukaryote References
cytoskeleton (Jenkins et al. 2002).
Ambler RP, Daniel M, Hermoso J, Meyer TE, Bartsch RG, Kamen MD (1979a)
Cytochrome c2 sequence variation among the recognised species of purple
nonsulphur photosynthetic bacteria. Nature 278:659–660
The Deinococcus-Thermus and Chloroflexus
Ambler RP, Meyer TE, Kamen MD (1979b) Anomalies in amino acid sequence of
Group small cytochromes c and cytochromes c’ from two species of purple photo-
synthetic bacteria. Nature 278:661–662
Here again are two phylum level taxa that, I feel, will someday Barns SM, Fundyga RE, Jeffries MW, Pace NR (1994) Remarkable archaeal
be united into a eubacterial kingdom. The phylogenetic diversity detected in a Yellowstone National Park hot spring environment.
Proc Natl Acad Sci USA 91:1609–1613
grouping of Deinococcus and Thermus, two disparate pheno-
Barns SM, Delwiche CF, Palmer JD, Pace NR (1996) Perspectives on archaeal
types in their own right, is another triumph of molecular diversity, thermophily and monophyly from environmental rRNA
(rRNA-based) phylogenetics (Weisberg et al. 1989). And the sequences. Proc Natl Acad Sci USA 93:9188–9193
genomic sequence data now available more than confirm the Beifuss U, Tietze M, Baumer S, Deppenmeier U (2000) Methanophenazine:
relationship. The Chloroflexus grouping likewise was pulled structure, total synthesis, and function of a new cofactor from methanogenic
Archaea. Angew Chem Int Ed Engl 39:2470–2472
together through rRNA analysis (Olsen et al. 1994). A specific
Beijerinck MW (1905) Versl K Akad Wet Amsterdam 14:168–169
relationship between the two phyla, however, is not convincingly Bell SD, Jackson SP (2001) Mechanism and regulation of transcription in archaea.
demonstrated on this level, merely vaguely suggested. So far, Curr Opin Microbiol 4:208–213
a lack of sufficient genomic sequence data prevents the hypoth- Breed RS (1939) Bergey’s manual of systematic bacteriology, 2nd edn. Williams &
esis being put to critical test—but early results from the genomic Wilkins, Baltimore
Burggraf S, Ching A, Stetter KO, Woese CR (1991) The sequence of Methanos-
front are encouraging, however (R. Overbeek, personal
pirillum hungatei 23S rRNA confirms the specific relationship between the
communication). extreme halophiles and the Methanomicrobiales. Curr Opin Microbiol
The ecological significance of these two phyla is still being 14:358–363
explored (by default as it were, because no one knows a priori Burkhardt F, Smith S (1990) The correspondence of Charles Darwin, vol 6.
what will be uncovered when new niches are probed by Pace’s Cambridge University Press, Cambridge, UK, pp 1856–1857
Crick FHC (1958) The biological replication of macromolecules. Symp Soc Exp
methodology). Environmental sampling of rRNA sequences
Biol 12:138–163
shows there to exist additional major lineages in both the Darwin C (1859) The origin of species by means of natural selection, or
Deinococcus-Thermus and the Chloroflexus phyla, none as yet the preservation of favoured races in the struggle for life. John Murray,
with isolated representatives (N. Pace, personal communication). London, p 69
Deckert G, Warren PV, Gaasterland T, Young WG, Lenox AL, Hartman H (1984) The origin of the eukaryotic cell. Speculations Sci Technol
Graham DE, Overbeek R, Snead MA, Keller M, Aujay M, Huber R, 7:77–81
Feldman RA, Short JM, Olsen GJ, Swanson RV (1998) The complete Hartman H, Fedorov A (2002) The origin of the eukaryotic cell: a genomic
genome of the hyperthermophilic bacterium Aquifex aeolicus. Nature investigation. Proc Natl Acad Sci USA 99:1420–1425
392:353–358 Hartmann E, König H (1990) Comparison of the biosynthesis of the
DeLong EF (1992) Archaea in coastal marine environments. Proc Natl Acad Sci methanobacterial pseudomurein and the eubacterial murein. Naturwis-
USA 89:5685–5689 senschaften 77:472–475
DeLong EF (1997) Marine microbial diversity: the tip of the iceberg. Trends Heberlein GT, De Ley J, Tijtgat R (1967) Deoxyribonucleic acid homology and
Biotechnol 15:203–207 taxonomy of Agrobacterium, Rhizobium, and Chromobacterium. J Bacteriol
DeLong EF, Wickham GS, Pace NR (1989) Phylogenetic stains: ribosomal 94:116–124
RNA-based probes for the identification of single cells. Science 243:1360–1363 Huber R, Langworthy TA, Konig H, Thomm M, Woese CR, Sleytr UB, Stetter KO
Deppenmeier U, Johann A, Hartsch T, Merkl R, Schmitz RA, Martinez-Arias R, (1986) Thermotoga maritima sp. nov. represents a new genus of unique
Henne A, Wiezer A, Baumer S, Jacobi C, Bruggemann H, Lienard T, extremely thermophilic eubacteria growing up to 90 C. Arch Microbiol
Christmann A, Bomeke M, Steckel S, Bhattacharyya A, Lykidis A, Overbeek 144:324–333
R, Klenk HP, Gunsalus RP, Fritz HJ, Gottschalk G (2002) The genome of Huber R, Wilharm T, Huber D, Trincone A, Burggraf S, Konig H, Rachel R,
Methanosarcina mazei: evidence for lateral gene transfer between bacteria Rockinger I, Fricke H, Stetter KO (1992) Aquifex pyrophilus gen. nov., sp.
and archaea. J Mol Microbiol Biotechnol 4:453–461 nov., represents a novel group of marine hyperthermophilic hydrogen-
Dobzhansky T (1963) Evolution, genetics and man. Wiley, New York oxidizing bacteria. Syst Appl Microbiol 15:340–351
Eisen JA, Nelson KE, Paulsen IT, Heidelberg JF, Wu M, Dodson RJ, Deboy R, Huber H, Hohn MJ, Rachel R, Fuchs T, Wimmer VC, Stetter KO (2002) A new
Gwinn ML, Nelson WC, Haft DH, Hickey EK, Peterson JD, Durkin AS, phylum of Archaea represented by a nanosized hyperthermophilic symbiont.
Kolonay JL, Yang F, Holt I, Umayam LA, Mason T, Brenner M, Shea TP, Nature 417:63–67
Parksey D, Nierman WC, Feldblyum TV, Hansen CL, Craven MB, Radune D, Jenkins C, Samudrala R, Anderson I, Hedlund BP, Petroni G, Michailova N,
Vamathevan J, Khouri H, White O, Gruber TM, Ketchum KA, Venter JC, Pinel N, Overbeek R, Rosati G, Staley JT (2002) Genes for the cytoskeletal
Tettelin H, Bryant DA, Fraser CM (2002) The complete genome sequence of protein tubulin in the bacterial genus Prosthecobacter. Proc Natl Acad Sci
Chlorobium tepidum TLS, a photosynthetic, anaerobic, green-sulfur bacte- USA 99:17049–17054
rium. Proc Natl Acad Sci USA 99:9509–9514 Kamp AF, La Riviere JWM, Verhoeven W (eds) (1959) Albert Jan Kluyver: his life
Fitch WM, Margoliash E (1967) Construction of phylogenetic trees. Science and works. North Holland, Amsterdam
155:279–284 Kandler O, Hippe H (1977) Lack of peptidoglycan in the cell walls of
Fox GE, Magrum LJ, Balch WE, Wolfe RS, Woese CR (1977a) Classification of Methanosarcina barkeri. Arch Microbiol 113:57–60
methanogenic bacteria by 16S ribosomal RNA characterization. Proc Natl Kandler O, König H (1985) Cell envelopes of archaebacteria. In: Woese CR,
Acad Sci USA 74:4537–4541 Wolfe RS (eds) The bacteria, vol 8, Archaebacteria. Academic, New York,
Fox GE, Pechman KR, Woese CR (1977b) Comparative cataloging of 16S ribo- pp 413–458
somal ribonucleic acid: molecular approach to procaryotic systematics. Int Kates M (1964) Bacterial lipids. Adv Lipid Res 2:17–90
J Syst Bacteriol 27:44–57 Kates M, Wassef MK, Kushner DJ (1968) Radioisotopic studies on the biosyn-
Fox G, Stackebrandt E, Hespell RB, Gibson J, Maniloff J, Dyer TA, Wolfe RS, Balch thesis of the glyceryl diether lipids of Halobacterium cutirubrum. Can
WE, Tanner RS, Magrum LJ, Zablen LB, Blakemore R, Gupta R, Bonen L, J Biochem 46:971–977
Lewis BJ, Stahl DA, Luehrsen KR, Chen KN, Woese CR (1980) The phylogeny Kelman Z (2000) DNA replication in the third domain (of life). Curr Protein Pept
of prokaryotes. Science 209:457–463 Sci 1:139–154
Galagan JE, Nusbaum C, Roy A, Endrizzi MG, Macdonald P, FitzHugh W, Kluyver AJ (1931) The chemical activities of microorganisms. University Press,
Calvo S, Engels R, Smirnov S, Atnoor D, Brown A, Allen N, Naylor J, London
Stange-Thomann N, DeArellano K, Johnson R, Linton L, McEwan P, Kluyver AJ, Donker HJL (1926) Die Einheit in der Biochemie. Chem Zell Gewiss
McKernan K, Talamas J, Tirrell A, Ye W, Zimmer A, Barber RD, Cann I, 13:134–190
Graham DE, Grahame DA, Guss AM, Hedderich R, Ingram-Smith C, Kluyver AJ, van Niel CB (1936) Prospects for a natural system of classification of
Kuettner HC, Krzycki JA, Leigh JA, Li W, Liu J, Mukhopadhyay B, bacteria. Zentralbl Bakteriol Parasitenkd Infektionskrankh 94:369–403
Reeve JN, Smith K, Springer TA, Umayam LA, White O, White RH, König H, Kandler O (1979) The amino acid sequence of the peptide moiety of the
Conway de Macario E, Ferry JG, Jarrell KF, Jing H, Macario AJ, Paulsen I, pseudomurein from Methanobacterium thermoautotrophicum. Arch
Pritchett M, Sowers KR, Swanson RV, Zinder SH, Lander E, Metcalf WW, Microbiol 121:271–275
Birren B (2002) The genome of M. acetivorans reveals extensive metabolic Lane DJ, Pace B, Olsen GJ, Stahl DA, Sogin ML, Pace NR (1985) Rapid determi-
and physiological diversity. Genome Res 12:532–542 nation of 16S ribosomal RNA sequences for phylogenetic analyses. Proc Natl
Gillespie D, Spiegelman S (1965) A quantitative assay for DNA-RNA hybrids with Acad Sci USA 82:6955–6959
DNA immobilized on a membrane. J Mol Biol 12:829–842 Langer D, Hain J, Thuriaux P, Zillig W (1995) Transcription in archaea: similarity
Golyshina OV, Pivovarova TA, Karavaiko GI, Kondrateva TF, Moore ER, to that in eucarya. Proc Natl Acad Sci USA 92:5768–5772
Abraham WR, Lunsdorf H, Timmis KN, Yakimov MM, Golyshin PN Lederberg J, Tatum EL (1946) Gene recombination in E. coli. Nature 158:558
(2000) Ferroplasma acidiphilum gen. nov., sp. nov., an acidophilic, autotro- Lewin RA (1984) Prochloron—a status report. Phycologia 23:203–208
phic, ferrous-iron-oxidizing, cell-wall-lacking, mesophilic member of the Maidak BL, Cole JR, Lilburn TG, Parker CY Jr, Saxman PR, Farris RJ, Garrity GM,
Ferroplasmaceae fam. nov., comprising a distinct lineage of the Archaea. Olsen GJ, Schmidt TM, Tiedje JM (2001) The RDP-II (ribosomal database
Int J Syst Evol Microbiol 50:997–1006 project). Nucleic Acids Res 29:173–174
Graham DE, White RH (2002) Elucidation of methanogenic coenzyme biosyn- Martin W, Mueller M (1998) The hydrogen hypothesis for the first eukaryote.
theses: from spectroscopy to genomics. Nat Prod Rep 19:133–147 Nature 392:37–41
Graham DE, Overbeek R, Olsen GJ, Woese CR (2000) An archaeal genomic Meyer TE, Cusanovich MA, Kamen MD (1986) Evidence against use of bacterial
signature. Proc Natl Acad Sci USA 97:3304–3308 amino acid sequence data for construction of all-inclusive phylogenetic trees.
Gray MW, Burger G, Lang BF (2001) The origin and early evolution of mito- Proc Natl Acad Sci USA 83:217–220
chondria (Reviews). Genome Biol 2:1018 Nelson KE, Clayton RA, Gill SR, Gwinn ML, Dodson RJ, Haft DH, Hickey EK,
Hafenbradl D, Keller M, Dirmeier R, Rachel R, Rossnagel P, Burggraf S, Huber H, Peterson JD, Nelson WC, Ketchum KA, McDonald L, Utterback TR, Malek
Stetter KO (1996) Ferroglobus placidus gen. nov., sp. nov., a novel hyperther- JA, Linher KD, Garrett MM, Stewart AM, Cotton MD, Pratt MS, Phillips CA,
mophilic archaeum that oxidizes Fe2+ at neutral pH under anoxic condi- Richardson D, Heidelberg J, Sutton GG, Fleischmann RD, Eisen JA, Fraser
tions. Arch Microbiol 166:308–314 CM et al (1999) Evidence for lateral gene transfer between Archaea
and bacteria from genome sequence of Thermotoga maritima. Nature 399: Tjian R (1996) The biochemistry of transcription in eukaryotes: a
323–329 paradigm for multisubunit regulatory complexes. Philos Trans R Soc Lond
Olsen GJ, Woese CR (1996) Lessons from an archaeal genome: what are we B 51:491–499
learning from Methanococcus jannaschii. Trends Genet 12:377–379 Tornabene TG, Langworthy TA (1979) Diphytanyl and dibiphytanyl glycerol ether
Olsen GJ, Lane DJ, Giovannoni SJ, Pace NR, Stahl DA (1986) Microbial ecology lipids of methanogenic archaebacteria. Science 203:51–53
and evolution: a ribosomal RNA approach. Annu Rev Microbiol 40:337–365 van Iterson Jr G, den Dooren de Jong LE, Kluyver AJ (1940) Martinus Willem
Olsen GJ, Woese CR, Overbeek R (1994) The winds of (evolutionary) change: Beijerinck: his life and his work. Delftsch Hoogeschoolfonds, Delft
breathing new life into microbiology. J Bacteriol 176:1–6 van Niel CB (1946) The classification and natural relationships of bacteria. Cold
Pace NR (1997) A molecular view of microbial diversity and the biosphere. Spring Harb Symp Quant Biol 11:285–301
Science 276:734–740 van Niel CB (1949) The ‘‘Delft School’’ and the rise of general microbiology.
Paster BJ, Dewhirst FE, Olsen GJ, Fraser I (1994) Phylogeny of Bacteroides, Bacteriol Rev 13:161–174
Prevotella, and Porphyromonas spp. and related bacteria. J Bacteriol van Niel CB (1955) Classification and taxonomy of the bacteria and blue green
176:725–732 algae. A century of progress in the natural sciences 1853–1953. California
Reeve JN, Sandman K, Daniels CJ (1997) Archaeal histones, nucleosomes, and Academy of Sciences, San Francisco, pp 89–114
transcription initiation. Cell 89:999–1002 Weisberg WG, Giovannoni SJ, Woese CR (1989) The Deinococcus-Thermus
Sanger F, Thompson EOP (1953) The amino-acid sequence in the glycyl chain of phylum and the effect of mRNA composition on phylogenetic tree construc-
insulin. Soc Gen Microbiol Symp 53:353–374 tion. Syst Appl Microbiol 11:128–134
Sanger F, Tuppy H (1951) The amino-acid sequence in the phenylalanyl chain of Woese CR (1987) Bacterial evolution. Microbiol Rev 51:221–271
insulin. Biochem J 49:481–490 Woese CR (2000) Interpreting the universal phylogenetic tree. Proc Natl Acad Sci
Sanger F, Brownlee GG, Barrell BG (1965) A two-dimensional fractionation USA 97:8392–8396
procedure for radioactive nucleotides. J Mol Biol 13:373–398 Woese CR (2002) On the evolution of cells. Proc Natl Acad Sci USA 99:
Schleifer K-H, Kandler O (1972) Peptidoglycan types of bacterial cell walls and 8742–8747
their taxonomic implications. Bacteriol Rev 36:407–477 Woese CR, Fox GE (1977a) The concept of cellular evolution. J Mol
Sogin SJ, Sogin ML, Woese CR (1971) Phylogenetic measurement in procaryotes Evol 10:1–6
by primary structural characterization. J Mol Evol 1:173–184 Woese CR, Fox GE (1977b) The phylogenetic structure of the prokaryotic
Stahl DA, Lane DJ, Olsen GJ, Pace NR (1985) Characterization of a Yellowstone domain: the primary kingdoms. Proc Natl Acad Sci USA 74:5088–5090
hot spring microbial community by 5S rRNA sequences. Appl Environ Woese CR, Sogin ML, Sutton LA (1974) Procaryote phylogeny. I: concerning
Microbiol 49:1379–1384 the relatedness of Aerobacter aerogenes to Escherichia coli. J Mol Evol
Stanier RY (1970) Organization and control in prokaryotic and eukaryotic cells. 3:293–299
In: Charles HP, Knight BCJG (eds) Society for general microbiology sym- Woese CR, Fox GE, Zablen L, Uchida T, Bonen L, Pechman K, Lewis BJ, Stahl
posium 20. Cambridge University Press, Cambridge, UK, pp 1–38 D (1975) Conservation of primary structure in 16S ribosomal RNA. Nature
Stanier RY, van Niel CB (1941) The main outlines of bacterial classification. 254:83–96
J Bacteriol 42:437–466 Woese CR, Debrunner-Vossbrinck BA, Oyaizu H, Stackebrandt E, Ludwig
Stanier RY, van Niel CB (1962) The concept of a bacterium. Arch Mikrobiol W (1985a) Gram-positive bacteria: possible photosynthetic ancestry. Science
42:17–35 229:762–765
Stanier RY, Doudoroff M, Adelberg EA (1957) The microbial world, 1st edn. Woese CR, Stackebrandt E, Macke TJ, Fox GE (1985b) A phylogenetic definition
Prentice-Hall, Engelwood Cliffs of the major eubacterial taxa. Syst Appl Microbiol 6:143–151
Stanier RY, Doudoroff M, Adelberg EA (1963) The microbial world, 2nd edn. Woese CR, Olsen GJ, Ibba M, Söll D (2000) Aminoacyl-tRNA synthetases,
Prentice-Hall, Engelwood Cliffs the genetic code, and the evolutionary process. Microbiol Mol Biol Rev
Stanier RY, Doudoroff M, Adelberg EA (1970) The microbial world, 3rd edn. 64:202–236
Prentice-Hall, Engelwood Cliffs Wolfe RS (1992) Biochemistry of methanogenesis. Biochem Soc Symp 58:41–49
Stanier RY, Adelberg EA, Ingraham JL (1976) The microbial world, 4th edn. Zuckerkandl E, Pauling L (1965) Molecules as documents of evolutionary history.
Prentice-Hall, Engelwood Cliffs J Theor Biol 8:357–366
2 What Is a Prokaryote?
W. Ford Doolittle1 . Olga Zhaxybayeva2
1
Department of Biochemistry and Molecular Biology, Dalhousie University, Halifax, Nova Scotia, Canada
2
Department of Biological Sciences, Dartmouth College, Hanover, New Hampshire, USA
Some Tangled-Together Questions About ‘‘Prokaryote’’ . . . 21 some leading scientists would surely question whether there is
any reason other than misguided adherence to tradition for
History and Fact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 a volume like this, dealing with Bacteria and Archaea as if they
were one kind of thing. Not least, the concerns are pedagogical.
The Concept of a Bacterium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Norman Pace, an unquestioned leader in microbial evolution
and systematics, has written:
Three Domains Versus Two: The Debate with " Because it has long been used by all texts of biology, it is hard to
Ernst Mayr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
stop using the word, prokaryote. But the next time you are
inclined to do so, think what you teach your students: a wrong
Time for a Change? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
idea. (Pace 2006, p. 289)
Paraphyly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 This chapter, then, hopes to explain why such questions

remain alive, and—to some of us at least—still interesting.
‘‘Pro’’ Implies Before . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Our goal is exploration and explication, not resolution, because
we believe that the question ‘‘What is a prokaryote?’’ has no
Negative Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 simple answer. What Evelyn Fox-Keller has recently written
about the nature versus nurture debate applies with equal force
Bacteria and Archaea, and the Fundamental Differences to the two domain versus three domain contretemps (and
Between Them . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 perhaps in most areas of protracted biological contestation).
" There is no single answer to this question, for a number of
Reticulation, the Bête Noire of Cladistics . . . . . . . . . . . . . . . . . . 33
different questions take refuge under its umbrella. Some of the
questions express legitimate and meaningful concerns that can
Coda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
be addressed scientifically, others may be legitimate and mean-
ingful, but perhaps not answerable, and still others simply make
no sense . . . all these different questions are tangled together
It is easy enough to collect scientific data on Escherichia coli
into an insoluble knot, making it all but impossible for us to stay
0157:H7 or Sulfolobus solfataricus strain P2 and to pronounce
clearly focused on a single, well-defined, and meaningful ques-
them ‘‘prokaryotes.’’ But it is not so easy to say what the word
tion. (Fox-Keller 2010, p. 1)
‘‘prokaryote’’ refers to, in any general sense. The issue is ontological.
Does ‘‘prokaryote’’’ refer to a category of things existing indepen-
dently of our categorizing, perhaps a ‘‘natural kind’’ to which any
organism can be surely said to belong or not belong? And if
Some Tangled-Together Questions About
‘‘prokaryote’’ is such a natural kind, is it defined by its essential
‘‘Prokaryote’’
properties or by its genealogical relatedness to other such kinds
(O’Hara 1998)? Or does the word instead denote merely a human " The lessons of the three-domain tree are profound. Instead of
concept, initially constructed on incomplete knowledge and pos-
two kinds of organism, prokaryotes and eukaryotes, there
sibly now hopelessly out-of-date? When we ask whether an
are three: bacteria, eukarya (eukaryotes) and archaea. (Pace
organism belongs to some recognized and named taxon, such
2006, p. 289)
as Archaea, Aves, or E. coli, we at least have in mind both a system
of classification and an understanding of a coherent theory At issue is whether living organisms on this planet are of two
(evolution) that gives us faith in the legitimacy of that system. kinds (prokaryotes and eukaryotes) or instead, three (Bacteria,
But is the same true for ‘‘prokaryotes’’? For ‘‘Prokaryota’’? Archaea, and Eukarya). Begged in this pronouncement by Pace,
These may seem to be abstract philosophical questions, of and in fact the thorniest question taking refuge under the
little relevance to practicing scientists. And yet microbiologists umbrella, is ‘‘What do we mean by kinds of organism’’? Failure
debate fiercely and tediously not only over what ‘‘prokaryote’’ to agree on this keeps the debate alive, and no data will bring it to
means but whether or not the word should be used at all. In just an end because the issue cannot be addressed scientifically, only
the last 2 or 3 years, the dispute has taken on new heat, such that philosophically.
22 2 What Is a Prokaryote?
POLYPHYLETIC GROUP
POLYPHYLETIC GROUP
warm-blooded flying things

swimming things
fishes whales bats therapsids turtles crocodiles pterosaurs birds lizards/snakes
bird
ancestor
mammal
ancestor
PARAPHYLETIC GROUP
reptiles
reptile ancestor
. Fig. 2.1
Polyphyly and paraphyly. See text for explanation
There are two general sorts of things we might mean by taught to believe by Ernst Mayr (Winsor 2006; McOuat 2009),
‘‘kind’’ in this context, and of course an enormous relevant it was unquestionably Darwin’s Origin of Species that redefined
philosophical/biological literature, inadequately cited in the the basis of naturalness in systematics, from phenetics to
summary that follows. First, we might be thinking only in phylogenetics.
terms of current properties, as in a phenetic classification. We " All the foregoing rules and aids and difficulties in classification
might consider prokaryotes and eukaryotes, or alternatively
are explained, if I do not greatly deceive myself, on the view that
Bacteria, Archaea, and Eukarya, to be candidate natural kinds,
the natural system is founded on descent with modification, that
real in a way that a category comprising all organisms taller than
the characters which naturalists consider as showing true affinity
1 m or whose genus name begins with the letter ‘‘E’’ would not be
between any two or more species are those which have been
(Dupré 1981; Hacking 1991; Bird and Tobin 2010). They could
inherited from a common parent, and, in so far, all true classifi-
be thought natural because they are defined by shared essences
cation is genealogical, that community of descent is the hidden
or essential properties, found in all and only representatives of
bond, which naturalists have been unconsciously seeking, and
the kind and somehow determinative of other characteristics.
not some unknown plan of creation, or the enunciation of gen-
Some of the rhetoric around the two domain versus three
eral propositions, and the mere putting together and separating
domain controversy seems to embrace such a view, with ‘‘infor-
objects more or less alike. (Darwin 1859, p. 420)
mational genes,’’ especially those of translation, embodying such
essential properties. This second, phylogenetic rather than phenetic way of seeing
Essence-based natural kind thinking is out of fashion gener- kinds as natural and real is what worked for Darwin and for
ally, however, and a more relaxed but still phenetic, evolution- most biologists today. The primary guarantor of legitimacy for
independent, or ahistorical alternative would be that biological any claim about how many ‘‘kinds of organism’’ there might be
taxa, from species up to domains, are ‘‘cluster’’ kinds, which is community of descent, not shared current properties. We and
share . . . chimpanzees are primates together because we descend from
"
a common ancestral species that would have been considered
families of properties [that] cluster together over time either
a primate. Our many phenetic similarities with chimps justify
because the presence of some properties in the family favors
putting us in the same order because (and only because) they are
the presence of others or because there are underlying internal
taken as evidence of that genealogical relationship. This second,
mechanisms and/or extrinsic contextual mechanisms that tend
historical or phylogenetic, way of being a natural kind might be
to secure the co-occurrence of the properties . . . Cluster kind
called ‘‘tree-essentialism’’ and seen as a consequence of what
realists will readily concede that, depending on the case, envi-
Robert O’Hara (1997) calls ‘‘tree-thinking.’’
ronmental pressures may affect and alter the set of properties
Often, classifications based on phenetic similarities and
associated with a kind over time. Therefore, in such cases, none
common descent or phylogeny converge. Indeed, that they gen-
of the properties . . . themselves need be individually necessary
erally do was at the very heart of Darwin’s case for evolution. But
for kind membership. (Bird and Tobin 2010)
they may sometimes come in conflict, in one of two ways
Although pre-Darwinian systematists were perhaps not so (> Fig. 2.1). First is polyphyly. Insectivorous warm-blooded fly-
wedded to an essentialistic classification as many of us were ing vertebrates include some birds and bats. As a taxon,
What Is a Prokaryote? 2 23
‘‘Insectivorous warm-blooded flying vertebrates’’ would be simultaneously. The grade/clade distinction is a large part of
called polyphyletic, because the most recent common ancestor what is at stake here, but there is more.
of birds and bats was neither warm-blooded nor flying, and
possibly not insectivorous. Such polyphyletic taxa represent
a serious violation of the principle of genealogical classifi-
History and Fact
cation, and are unacceptable to biologists as ‘‘natural’’ taxa,
however similar the organisms so aggregated might be. " Historical narratives in which science appears to advance steadily
Whales as fishes would be an even more egregious example.
in the direction of greater accumulations of factual knowledge
If Bacteria and Archaea do not share a common ancestor that
are now widely scorned as ‘‘whig history.’’ . . . like the stories the
we could consider a prokaryote, Pace’s objection to the use of
Whig political historians used to tell about the steady growth of
‘‘prokaryote’’ (as explained below) would carry considerable
English liberty. Today’s historians are more likely to set them-
weight.
selves the goal of understanding the past ‘‘in its own terms’’
The second sort of phenetic/phylogenetic conflict is
(whatever that might mean) rather than in the light of subse-
paraphyly. Birds and bats are both members of monophyletic
quent developments. This has yielded histories in which knowl-
taxa, in that the classes Aves and Mammalia descend from
edge, rather than continuously increasing, has undergone radical
common ancestors we would consider to be class members
discontinuities and transformations, and in which what subse-
(the last common ancestral bird and the last common ancestral
quently come to be seen as forward movements are deeply
mammal, respectively). Moreover, no non-birds or non-
rooted in contexts that are quite foreign from a modern perspec-
mammals descend from those respective ancestors. But the
tive. (Golinski 1998, p. 4)
class Reptilia, from which both Aves and Mammalia arose then
reveals itself to be a paraphyletic taxon. It includes an ancestor Regrettably, much history of science written by scientists is
(the last common ancestral reptile) and some but not all of its unabashedly and joyfully of the ‘‘whiggish’’ variety scorned by
descendants because others of those descendants are birds and professional historians, portraying an inevitable advance
others still are mammals. Cladists consider such taxa illegiti- (albeit with occasional detours through back alleys) from
mate, instead requiring that we recognize two kinds of what once ignorance and error into truth, seldom admitting that today’s
were called reptiles, synapsids (including mammals) and archo- knowledge may well be seen as relative ignorance tomorrow,
saurians (including dinosaurs and birds). and often overemphasizing the authors’ own roles in that
But unlike polyphyletic groupings, paraphyletic taxa do advance. Perhaps to working scientists accuracy and indepen-
not result in the false unification of species that do not dent confirmation are essential for dealing with experimental
share community of descent, and thus do hardly any violence data, but for historical ‘‘data’’ requiring nuanced interpreta-
to Darwin’s understanding of the evolutionary foundations of tion, not so much.
natural systematics. Indeed, by recognizing that because of Many false stories are propagated by repetitious citation, the
variations in tempo and mode some of an ancestor’s descen- academic equivalent of ‘‘urban myths.’’ Worse, historical narra-
dants resemble it far less than others, the acceptance of tives can be deliberately constructed as foils against which the
paraphyletic taxa, coupled with knowledge of their origins, value of a favored new observation or hypothesis will shine out.
adds explanatory value to classification. Many traditional sys- Ernst Mayr’s magisterial tome The Growth of Biological Thought,
tematists, including most prominently Ernst Mayr, incorpo- as we can see quite clearly now, was no disinterested exercise in
rate such considerations of what Darwin called ‘‘degree of pure historical scholarship (O’Malley 2010)! Moreover, even
difference’’ into their practice. They deploy and defend truly disinterested practitioners of a discipline will inevitably
paraphyletic groupings, given adequate justification (Ashlock have come to whatever is their current understanding of the
1974; Grant 2003). accumulated factual knowledge by different routes, in all hon-
Birds and mammals, with respect to their reptilian ancestors, esty remembered quite differently. Outsiders cannot but be
and eukaryotes with respect to prokaryotes, represent what influenced most by those practitioners who write most persua-
Mayr called evolutionary grades. sively or most often, no matter how idiosyncratic their views.
Not surprisingly, much of the dispute over the use of ‘‘pro-
" . . . the anagenetic component of evolution often leads to the
karyote’’ derives from different readings of that term’s intended
development of definite ‘‘grades’’, or levels of evolutionary
meaning and of the common understandings of microbiologists
change, which must receive recognition in classification. The
who have used it over the last 50 years. A crucial element of the
objection raised by the cladists that this would introduce sub-
recent argument against ‘‘prokaryote’’ has become that it was
jectivity into classification has been rejected by the evolutionary
intended – or has come to be understood by most of its users – to
taxonomist . . . . (Mayr 1982, p. 234)
denote a phylogenetic division rather than (or as well as) a cell
As we hope to show, the argument over ‘‘prokaryote’’ sur- type, a clade rather than a grade, or a level of organization. If the
vives in no small part because its protagonists claim to be former, then we would expect that many of the features in which
disagreeing over scientifically addressable facts when in actuality Bacteria and Archaea are judged to be similar are genuinely
they are arguing about different notions of what are ‘‘natural’’ homologous: if the latter, it is possible that many such features
groups, sometimes indeed from conflicting positions taken are convergent or merely analogous. Considered as a clade
designation, the ‘‘prokaryote/eukaryote dichotomy’’ imposed IA

BACTER EUKARYA ARCHA
EA
upon microbiology a misleadingly reductionist, falsely unitary,
and dangerously non-evolutionary worldview, its detractors
claim (Pace 2006, 2009; Pace et al. 2012).
Perhaps, but a complete reading of the microbiology’s past as
it affects its present is surely more complex. In our view, the
important and still open question about ‘‘prokaryote’’ concerns
its value as a descriptor of a grade or level of cellular organiza-
tion. That it might be also taken or mistaken as the name of
a monophyletic clade (the so-called Prokaryota) is an interesting
side issue. Moreover, whether or not monophyly is properly
attributed to prokaryotes depends rather much on what ‘‘mono-
phyly’’ itself is taken to mean, as does the assumption that any of
the three domains is actually ‘‘monophyletic.’’
The Concept of a Bacterium
Jan Sapp, a professional historian unusually well-read in the ORIGIN

microbial evolutionary/systematic literature and unusually well . Fig. 2.2
acquainted with the discipline’s key figures, has recently The three-domain tree, redrawn from Pace (2006). Structure
highlighted some of the ‘‘urban mythology’’ concerning ‘‘proof the universal Tree of Life on which Pace (2006) bases his case
karyote,’’ while providing what seems a balanced intellectual against the use of ‘‘prokaryote’’
history of the concept in ‘‘its own terms’’ (Sapp 2006, 2009a, b).
The 1999 Berkeley Ph.D. thesis of Susan Spath (1999) earlier
provided important details backgrounding the 1962 publication bacteria as a subject of study from the narrowing focus of
by Roger Stanier and Cornelius van Niel of ‘‘The Concept of infectious disease researchers and the reduction to mere model
a Bacterium’’ in Archiv für Mikroboliogie. organisms threatened by the successes of molecular biologists
This paper is widely considered to be the cornerstone on who saw Bacteria primarily as tools. They hoped to draw a line
which the belief that the most natural way to divide living things above the Bacteria (separating them from eukaryotes) just as
into two kinds is as ‘‘prokaryotes and eukaryotes’’ rested—for Lwoff, in his 1957 ‘‘The Concept of a Virus’’ had drawn a line
almost half a century. In it, Stanier and van Niel wrote: below. Moreover, they intended to quash the false belief,
"
supported by some microscopists, that Bacteria have nuclei.
It is now clear that among organisms there are two different
And lastly they hoped to codify a long-standing but still
organizational patterns of cells, which Chatton (1937) called, with
contested intuition that ‘‘blue-green algae’’—in spite of their
singular prescience, the eucaryotic and procaryotic type. The
photobiochemical affinities to ‘‘higher’’ algae and green plants
distinctive property of bacteria and blue-green algae is the prokary-
(and widespread speculation about their close evolutionary links
otic nature of their cells. It is on this basis that they can be clearly
thereto)— were just another kind of bacterium (indeed,
segregated from all other protists (namely, other algae, protozoa,
‘‘cyanobacteria’’).
and fungi), which have eucaryotic cells. (Stanier and van Niel
Almost as frequently quoted as representative of Stanier
1962, pp. 20–21)
and van Niel’s formulation are these words from the second
Sapp shows Chatton’s ‘‘singular prescience,’’ since lauded (1963) edition of the classic text The Microbial World,
and repetitively cited by many of us who cannot read French, coauthored by Stanier with Michael Doudoroff and
to be mythic. The terms were first coined quite casually by Edward Adelberg, which must have been written at about the
Chatton and came to Stanier and van Niel via his most successful same time.
student, André Lwoff. But once accepted, Spath notes: " In fact, this basic divergence in cellular structure, which
" The terminology introduced by Stanier and van Niel appears to separates the bacteria and blue-green algae from all other cellular
have diffused widely through all branches of biology with little organisms, probably represents the greatest single evolutionary
discussion . . . From the 1960s to the 1980s, the term procaryote discontinuity to be found in the present day world. (Stanier et al.
and eucaryote were usually defined without acknowledgement 1963, p. 85)
of their origin. They became as much a part of standard biolog-
> Figures 2.3a–d illustrate the various ways in which these
ical discourse as ‘molecule’ or ‘DNA.’ (Spath 1999, p. 51)
words might together be interpreted. That great ‘‘evolutionary
Indeed, the prokaryote:eukaryote division seemed to satisfy discontinuity’’ might be seen (and was described in Stanier and
many needs at once. Stanier and his former mentor van Niel van Niel’s ‘‘The Concept of a Bacterium’’) as primarily
were at pains to protect general bacteriology as a science and a structural one, of course the product of evolution (as is all of
a b c d
EUKARYOTES PROKARYOTES
EUKARYOTES
PROKARYOTES
PROKARYOTES
LUCA
e f g h
PROKARYOTES PROKARYOTES
plastids plastids plastids
ndria ndria i
mitocho mitocho
ndria mitocho
LUCA LUCA?
. Fig. 2.3
Eight representations of the possible relationship between prokaryotes and eukaryotes. (a) Two distinct types of cellular organization
(without implications about evolutionary relationship); (b) relationship between types, assuming that all life has a common origin and
that simple precedes complex; (c) prokaryotes and eukaryotes as separately evolving clades, their common ancestor being pre-cellular,
possibly as per Ernst Haeckel; (d) likely the most common view in the mid-to-late twentieth century, the Last Universal Common
Ancestor (LUCA) being a cell or species of the prokaryotic type. (e) Bacteria, Archaea, and eukaryotes as separately evolving clades, their
common ancestor being pre-cellular or ‘‘progenotic’’; (f) as (e), but recognizing a specific Archaeal/eukaryotic ancestor, and taking into
account endosymbioses giving rise to mitochondria and plastids: (g), as (f) but with the Last Universal Common Ancestor (LUCA) being
a cell or species of the prokaryotic type, and Bacteria and Archaea considered a paraphyletic taxon (‘‘Prokaryotes’’); (h), as (f), but
recognizing extensive interdomain lateral gene transfer (LGT). (h) seems most consistent with publications by Woese or Pace cited
herein, although it is (g) that seems to be evoked in > Fig. 2.2 above
biology), but entailing no particular phylogenetic scenario rise to eukaryotes, then that ‘‘evolutionary discontinuity’’ also
(> Fig. 2.3a). However, if one believes that simpler cells inevita- represents a deep branch in the Universal Tree of Life
bly come before more complex ones, then Stanier and van Niel’s (> Fig. 2.3d). The Tree’s root (LUCA, or the Last Universal
line of cell-structure demarcation had to have been crossed at Common Ancestor) must have been a prokaryote.
least once: the prokaryotic cell type having given rise to the Spath and Sapp seem to disagree on how much we are to
eukaryotic cell type (> Fig. 2.3b). Indeed, they thought that understand ‘‘The Concept of a Bacterium’’ to be making
too, and the passage quoted above continues . . . a phylogenetic claim (along the lines of > Fig. 2.3d) on top of
"
the structural differentiation (> Fig. 2.3a) that was so clearly its
It is not too unreasonable to consider that the bacteria and blue-
main intent. The former argues that:
green algae represent vestiges of a stage in the evolution of cells
which, once it achieved a eukaryotic structure in the ancestors of " The procaryote/eucaryote distinction implicitly expressed an
the present-day higher protists, did not undergo any further important taxonomic proposition about the deepest phyloge-
changes through the entire subsequent course of biological netic division among living things. At the time it was formulated,
evolution. (Stanier et al. 1963, p. 85) it seemed obvious to Stanier and van Niel that the procaryotes,
though diverse, belonged to a monophyletic category, as did the
Possibly, one might consider existing prokaryotes and eucaryotes. The distinction provided the basis for ending
eukaryotes as the tips of separate evolutionary lineages, or rather a conviction about the natural world formalized by Aristotle in
clades of lineages, each with its own history and, possibly, the fifth century B.C. and made sacred by Linnaeus in the eigh-
essential characters (> Fig. 2.3c). But if one cannot deny (as teenth century: namely, that all organisms were either plants or
Stanier and van Niel could not) that all living things are related animals. Though challenges to that conviction had been
and believes (as they did) that prokaryotes preceded and gave launched since the middle of the nineteenth century, none had
achieved wide acceptance. The procaryote/eucaryote distinc- cellular organization, each of which embodied, with certain
tion, in contrast, ended the reign of the plant and animal king- limits, the same kinds of evolutionary potentialities. (Stanier
doms as the fundamental bifurcation of living things. When and van Niel 1962, p. 33)
compared at the cellular level, it became evident that plants
The Microbial World does express faith in the ability to
and animals were much more like each other than they were
‘‘safely infer a common origin for the whole group [prokaryotes]
like procaryotes. (Spath 1999)
in the remote evolutionary past’’ (Stanier et al. 1963, p. 409).
Sapp (2006), on the other hand, writes that: Pace et al. 2012 take this as a specific inference about prokaryotic
"
monophyly but more likely Stanier and colleagues had in mind
Stanier and van Niel’s distinction was neither an evolutionary nor
something like > Fig. 2.3d. They were not cladists.
a taxonomic one – at least not as they drew it. In fact, their
One could also complain (as Pace and Woese have) that the
attitude toward an evolutionary-based classification of bacteria
notion that eukaryotes come from prokaryotes (> Fig. 2.3b) is
had taken a sudden change of course prior to 1962. (Sapp 2006,
progressivist and reflects an incomplete commitment to tree-
p. 165)
thinking. Molecular biologists in particular still talk and write
Sapp here refers to the well-documented and often decried about a progression in complexity from prebiotic chemistry-to-
mid-century abandonment by these two most influential figures bacteria-to-yeast-to-Drosophila-to-Homo sapiens in ways that
in microbial systematics of all hope for a satisfactorily complete call to mind The Great Chain of Being (Lovejoy 1936), and ignore
phylogenetic framework for bacteriology, leaving microbial tax- the obvious fact that all extant lineages have equally long evolu-
onomists with at best a useful ‘‘determinative,’’ identification- tionary histories. That prokaryotes and eukaryotes are successive
directed, system of classification (see Sapp 2006; Pace et al. 2012 rungs on the ladder of life is just another example of that pre-
and the next chapter in this volume). Probably a balanced read- Darwinian way of thinking, it might be claimed. But, as argued
ing of ‘‘The Concept of a Bacterium’’ is as a sort of ‘‘at least there below, if one objection to ‘‘prokaryote’’ is that it means nothing
is this’’ proposition: we may never be able to sort out the tree of more than ‘‘not-eukaryote’’ (Pace 2006, 2009), then all that
bacterial evolution, but we can (at least) clearly distinguish > Fig. 2.3b or > Fig. 2.3d imply is that eukaryotes arose from
prokaryotes (of which all were then considered Bacteria) from ancestors we would not now call eukaryotes. Though some
eukaryotes, which (probably) arose from within them. authors (even one of us at one time; Darnell and Doolittle
Whatever the deeper convictions of Stanier and van Niel (1986)) have suggested that the first cells were eukaryote-like,
were in 1962, many microbiologists post-1962 were indeed with prokaryotes being their streamlined descendants (Penny
happy to accept the prokaryote:eukaryote split as and Poole 1999), this seems now an irresponsibly wild
a phylogenetic division as well as cell-structural dichotomy, as conjecture.
sketched in > Fig. 2.3d. It would not however have been seen
(pace Pace and Spath) as the deepest division in the Tree of Life,
as long as one held that eukaryotes likely arose from within the
Three Domains Versus Two: The Debate with
prokaryotes, that is, from one or more already established pro-
Ernst Mayr
karyotic lineages. So (we believe) the default notion throughout
most of the last half of the last century was that it was the
As Woese masterfully summarizes in the next chapter, the pro-
concomitant radical structural reorganization that made the
posal that living things are fundamentally of three kinds, not
prokaryote: eukaryote divergence into the ‘‘greatest single evo-
two, did not have the same easy ride to acceptance as did Stanier
lutionary discontinuity to be found in the present day living
and van Niel’s dichotomy—in no small part because of the
world,’’ not the topology of the implicit Tree of Life (> Fig. 2.3d).
widespread, almost culturally ingrained acceptance of the latter.
For those who cared about such niceties, prokaryotes were
Many of us, biologists, felt or still feel deeply that prokaryotes
already paraphyletic.
versus eukaryotes is ‘‘just the way it is,’’ not from any direct
In any case, ‘‘The Concept of a Bacterium’’ taken ‘‘in its own
observational experience or dispassionate consideration of the
terms’’ is relatively noncommittal about the specific evolution-
‘‘accumulation of factual knowledge,’’ but because it seems
ary connection between prokaryotes and eukaryotes, while
foundational to the biology we were taught and the institutional
admitting to parallel evolutionary processes.
and economic structuring of our disciplines, even our social
" The evolutionary diversification of the prokaryotic protists is lives!
expressed in: (1). gross organization, leading to the existence of Somewhat belatedly, Ernst Mayr (1990, 1991, 1998) vigor-
unicellular, multicellular, and coencytic groups; (2). mode of ously expounded a traditional(ist) objection to Woese’s
cellular locomotion; (3). mode of cell division; and (4). major three domain view, based both on his (Mayr’s) different per-
patterns of energy-yielding metabolism . . . With respect to all spective on how classification should be done and what he
these features, there are parallel modes of evolutionary diversi- thought the Archaea were really like. Mayr contrasts his favored
fication among the eukaryotic protists (i.e. other groups of algae, mode of taxonomic practice, ‘‘Darwinian classification’’ (aka
protozoa and fungi). Consequently, if we look at the microbial ‘‘evolutionary taxonomy’’), to the ‘‘Hennigian cladification’’ to
world in its entirety, we can now see that evolutionary diversifi- which he thinks Woese largely, if incompletely, adheres. With the
cation through time has taken place on two distinct levels of former . . .
" organisms are grouped into taxa on the basis of two criteria, biochemical level; it can be in the basic information processing
similarity and genealogy. A higher taxon recognized by these systems of the cell; and so on. Clearly the vast diversity among
criteria is composed of a group of similar and/or related species birds and among insects is structural diversity, whereas that
descended from their nearest common ancestor. Such a taxon is among the Bacteria or the Archaea is necessarily of the other
called monophyletic. In a cladification, favored by cladists, only types. Dr. Mayr’s is an eye-of-the-beholder type of diversity. It
genealogy is considered. It recognizes branches (clades) of the rests on the incredible capacity of the human eye to distinguish
phylogenetic tree, comprised of the stem species of such minute differences in pattern. But almost all microbial diversity
a branch together with all its descendants. (Mayr 1998, p. 9721, cannot be sensed visually, which means that subtle variations in
Emphasis ours) pattern almost always go undetected. . . When he compares
plant and animal diversity to microbial diversity, Dr. Mayr is
The important difference indeed is in the word ‘‘all.’’ Citing
comparing apples and oranges, and his attempt to apply globally
the already familiar example of reptiles, birds, and mammals
a parochial and subjectively defined concept of diversity serves
Mayr (1998) continues . . .
only to reveal the futility in such an approach. (Woese 1998a,
" In both cases, the cladist removes the branches that gave rise to p. 11045)
the mammals or birds from the reptiles, thereby making the
reptiles, a taxon used in our every-day grouping of animals,
a ‘‘paraphyletic group,’’ not permissible as a formal taxon in
a strictly cladistic arrangement. In both cases, the Darwinian Time for a Change?
taxonomist, who deals with groups rather than with branches,
retains the ancestral groups within the Reptilia and recognizes as In 2006, Norman Pace took up the attack on ‘‘prokaryote’’ with
mammals or birds only those assemblages of species which by renewed vigor. In an pithy one page editorial entitled ‘‘Time for
their diagnosis are characterized as mammals or birds. It was on a Change’’ appearing in Nature (Pace 2006), he drew attention to
this basis that Stanier and van Niel recognized two empires, the what he considered the implications for systematics of the
prokaryotes and the eukaryotes. (Mayr 1998, p. 9721) rooted version of the three-domain rRNA tree (Woese et al.
1990), in which Archaea and Eukarya are sister taxa, splitting
Mayr, like Woese and most biologists today, accepted the
later than Bacteria. His argument as presented then and reiter-
phylogeny displayed in > Fig. 2.3f or > Fig. 2.3g. But he felt that
ated 3 years later (Pace 2009) was fourfold.
the similarities between Bacteria and Archaea were so great and
their collective similarity to eukaryotes so small that the former 1. How, he asks, can we consider Bacteria and Archaea to form
two are legitimately classified into one superkingdom a single group (the Prokaryota) when in fact one of its two
(Prokaryota, as shown in > Fig. 2.3g) and the last into another subgroups, the Archaea, is more closely related to the
(Eukaryota). It was a matter of recognizing that—as a grade— eukaryotes? (See > Fig. 2.2, reproduced from Pace 2006).
the eukaryote lineage had undergone massive, presumably This would be to recognize a paraphyletic taxon (like
selected, changes in the organization of its informational and Reptilia), anathema to cladists.
especially cellular machinery. Moreover, he pointed out that 2. The pro- and eu- prefixes themselves imply that prokaryotes
Woese was not a very thorough-going cladist, because by those gave rise to eukaryotes. But ‘‘the nuclear line of descent is as
principles (and with the branching pattern shown in > Fig. 2.3g) ancient as the archaeal line and not derived from either
Woese should recognize only two primary domains, Bacteria and archaea or bacteria.’’
a second (called Neomura by Cavalier-Smith (2002)) compris- 3. Prokaryotes are defined only negatively, by those eukaryotic
ing two sub-domains, Archaea and Eukarya. Mayr charitably features that they lack. ‘‘No one can define what is
attributed these failings to Woese’s naı̈veté, as he expressed in a prokaryote, only what it is not.’’
a letter to WFD, dated January 31, 2000, that . . . 4. ‘‘Lumping bacteria and archaea conceptually discounts
"
fundamental differences between these two kinds of
Woese came into microbiology from outside of biology and did
organism . . .’’
not (and still does not) understand what classification is all about.
Biologists trained in classification and evolution would like to Thus, he concluded that ‘‘prokaryote’’ embodies
express the fact [of] how different the Prokaryotes are from the a scientifically disprovable false idea . . .
Eukaryotes.
" Prokaryote: gene-sequence comparisons show the tree of life
But, one could easily rebut, Mayr did not understand what consists of Bacteria, Eukarya, and Archaea. The use of the term
Archaea were all about! Woese, in the next chapter, emphasizes ‘‘prokaryote’’ fails to recognize that an idea about life’s origins
the many ways in which Archaea and Bacteria differ in basic cell has been proved wrong. (Pace 2006)
and molecular biology. And in his immediate response to Mayr’s
There is much that can be said in rebuttal and was, for
1998 attack, he cogently noted . . .
instance, in correspondences to Nature by Martin and Koonin
" Diversity can be of many types. It can be at the level of structure (2006), Dolan and Margulis (2007), and Cavalier-Smith (2007)
and organization; it can be anabolic or catabolic enzymatic and later by Whitman (2009) in the Journal of Bacteriology.
diversity; it can be environmental adaptation at the molecular/ An excellent recounting of the published and unpublished
contestation following ‘‘Time for a Change’’ is provided by three-domain tree dictates that features found in eukaryotes,
Sapp (2009a, b). Each of Pace’s four complaints is debatable at but not in Archaea and Bacteria, are more likely to have been
length, and is the subject of one of the next four sections of this invented in the eukaryotic branch than lost in the archaeal
essay. branch. And if, as Pace observes, all it takes to be a prokaryote
is to lack eukaryotic defining traits, then the common
archaeal/eukaryote ancestor surely was a prokaryote, although
Paraphyly (if Archaea and Eukarya are indeed sisters) more Archaea-like
than Bacteria-like.
Pace’s first argument seems to be that prokaryotes comprise Moreover, it seems now equally possible that the eukaryotic
a paraphyletic group. This should thoroughly distress only cla- ‘‘nuclear line of descent’’ is not in fact ‘‘as ancient as the archaeal
dists, which neither Pace nor Woese in other contexts appears to line,’’ but arose from within it! Such an alternative view has been
be. Their above-mentioned refusal to recognize or name around almost as long as Archaea have been recognized, James
Neomura as a clade shows that they are not strict Hennigians, Lake having suggested in the early 1980s, on the basis of ribo-
as does their insistence on the meaningfulness of ‘‘domain’’ some structure, that eukaryotes show a particular affinity to one
status. Degree of difference (especially in the informational of the two then recognized archaeal subdomains, the
transcription, translation, and replication machineries) and the Crenarchaeota, which he called ‘‘eocytes’’ (Lake et al. 1984)
fact that Archaea and Bacteria are separated by deep and long Martin Embley and coworkers (Cox et al. 2008; Foster et al.
branches in the rRNA tree both figure in Pace and Woese’s case 2009) find that sophisticated and conservative phylogenetic
for the domain status of either, and are most consistent with the methods that compensate for compositional and rate heteroge-
practices Mayr styles ‘‘Darwinian classification.’’ neity across sites and across trees are indeed prone to produce
Most microbial systematists, and indeed most biologists the ‘‘eocyte tree’’ in preference to the accepted three-domain
who care at all about microbes would be aware of the benefits tree, when applied to rRNA gene sequences or to a set of several
and dangers of the grade/clade distinction. But many of them dozen highly conserved and ‘‘core’’ genes of translation, tran-
would argue that the elaboration of the cytoskeleton and scription, and replication.
endomembrane systems and the uncoupling of transcription Taking into account also the presence or absence of certain
and translation that seemed to have occurred rapidly and key genes—several ribosomal protein genes, two RNA polymer-
simultaneously in the formation of LECA (the last universal ase subunit genes, genes for components of the cell division
eukaryote ancestor), however that came about, were so rad- machinery and a ubiquitin modifying system, and most excit-
ical and of such consequence as to make the prokaryote- ingly a cell-shape-determining actin ortholog—Guy and Ettema
eukaryote divide still ‘‘the greatest single evolutionary dis- (2011) recently argued for a specific placement of eukaryotes
continuity to be found in the present day world’’ (Embley within what they call the TACK superphylum of Archaea. This
and Martin 2006; Field and Dacks 2009). Thus, as Whitman comprises Crenarchaeota, Thaumarchaeota, Korarchaeota, and
(2009) writes, the recently proposed Aigarchaeota, and specifically excludes
"
Euryarchaeota and Nanoarchaeota. Given the recent and still
Even though the prokaryotes are not monophyletic and the
unstable taxonomic treatment of an expanding archaeal data set,
evolutionary processes they have experienced are extremely
we take this as being a modern version of Lake’s ‘‘eocyte hypoth-
complex, this classification strategy remains useful and knowl-
esis.’’ If this view gains wide acceptance and Archaea themselves
edge of the evolutionary processes which formed modern
become paraphyletic, Pace’s phylogenetic/cladistic argument
organisms provides a great deal of insight into their biological
against prokaryotes would also deconstruct Archaea. At the
properties. (Whitman 2009, p. 2003)
very least we should remain agnostic as to the relationship
between Archaea and the eukaryotic nuclear lineage (Gribaldo
et al. 2010).
‘‘Pro’’ Implies Before
Pace’s second objection, that the ‘‘pro’’ in prokaryotes is mislead-

Negative Definition
ing because it implies an ancestral relationship to eukaryotes,
seems inconsistent with his first, almost disingenuously so. To
Pace’s third argument, that prokaryotes are defined only nega-
call a taxon paraphyletic is to admit that another taxon emerged
tively, has more bite, and requires more discussion. Stanier and
from within it. It is of course commonly understood that con-
van Niel’s definition of ‘‘prokaryote’’ was indeed dismayingly
temporary taxa did not evolve from each other. Humans did not
minimalistic.
evolve from chimpanzees: they instead share with chimps
a common hominid ancestor, which was neither chimp nor " The principal distinguishing features of the procaryotic cell are:
human. But one could be almost certain it was hairier than us, 1. absence of internal membranes which separate the resting
because all the outgroup apes are. Similarly, the common ances- nucleus from the cytoplasm, and isolate the enzymatic machin-
tor of Archaea and eukaryotes was by definition neither one nor ery of photosynthesis and of respiration in specific organelles;
the other. But simple parsimony together with the accepted 2. nuclear division by fission, not by mitosis, a character possibly
related to the presence of a single structure which carries all the a specific archaeal/eukaryote relationship may be demonstrated
genetic information of the cell; and 3. the presence of a cell wall (Cox et al. 2008; Cotton and McInerney 2010), it is still true
which contains a specific mucopeptide as its strengthening ele- (3) that a much larger number, many hundreds, of ‘‘operational’’
ment. (Stanier and van Niel 1962, pp. 32–33) genes make up a shared resource that is common to Bacteria and
Archaea. Such genes are often (on an evolutionary time scale)
Even these ‘‘negative’’ descriptors now seem problematic,
laterally transferred within and between them, and generally
since none is thought true of all prokaryotes. The specific pep-
absent from eukaryotes. Indeed, one might define ‘‘prokaryotes’’
tidoglycan mucopeptide referred to is absent from eukaryotes
‘‘positively’’ on this basis. They are that community of organisms
(except in some photosynthetic organelles of cyanobacterial
that, over an evolutionary timescale, draws on a vast shared pool
origin), but also from many bacterial and all archaeal walls,
of genes encoding diverse metabolic functions seldom if ever
and there is much compositional variation in the walls of both
used by eukaryotes. In a sense, evolution by LGT remains an
prokaryotic domains (Whitman 2009). Fission is not the only
important defining feature of prokaryotes, just as Woese would
way either Bacteria or Archaea divide (Angert 2005; Makarova
have it have been for progenotes, a feature which eukaryotes
et al. 2010). And although no prokaryote has internal mem-
lack, relatively speaking (Woese 1998b, see below).
branes that are likely to be homologous to eukaryotic mem-
In any case, the key systematics question is not so much
branes, it is the case that within the planctomycetes there are
whether there are ‘‘non-negative’’ features uniting Bacteria and
remarkable analogs thereof (Fuerst and Sagulenko 2011), arising
Archaea as whether there is any reason to believe that they derive
by covergence or gene transfer (McInerney et al. 2011).
from a common ancestor that might itself be called a prokaryote,
Still, it is not always so easy to tell negative from positive in
so that they might together be considered a legitimate clade—
such general descriptions of organismal characteristics. The
albeit a paraphyletic one. There are two different questions
absence of functionally differentiated internal membranes has,
embedded here: (1) what do we mean by common ancestor
for instance, been effectively re-cast positively, as the multifunc-
(versus common ancestry), and (2) do we have a sufficiently
tionality of the prokaryotic cytoplasmic membrane, by
refined definition of ‘‘prokaryote’’ to be able to say whether or
Whitman (2009).
not the common ancestor, if it existed, was one? As with nature
" The cytoplasmic membrane is multifunctional in procaryotes and versus nurture, ‘‘these different questions are tangled together
represents the defining structure of the cell. A proton motive into an insoluble knot’’ (Fox-Keller 2010).
force is generated on the cytoplasmic membrane by respiration, In several earlier and recent writings, Woese has argued that
photosynthesis, or ATP hydrolysis to empower key cellular pro- the common ancestor of all three domains was not a prokaryote
cesses such as ATP biosynthesis, NAD+ reduction by reverse or even a population of prokaryotes, but rather ‘‘the progenote,’’
electron transport, nutrient uptake, motility, and secretion. Pro- an inchoate collection of entities still in the throes of evolving the
caryotes utilize membrane transporters on the cell surface to genotype/phenotype coupling. The manifest differences in
assimilate nutrients dissolved in their environment. In many pro- information processing machinery between Bacteria, Archaea,
caryotes, the cytoplasmic membrane possesses a complex topol- and eukaryotes tell us about that early period, he feels, and
ogy composed of lamellae, tubules, or other cytoplasmic suggests a model in which the three domains are best thought
intrusions. In contrast, the cytoplasmic membrane of eucaryotes of as arising independently from an ancestral ‘‘state’’ rather than
is very different in structure and function. (Whitman 2009, an ancestral cell or species, as in > Fig. 2.3e. Of that state, he
p. 2000) wrote (Woese 1998b). . .
" Organismal lineages, and so organisms as we know them, did not
Similarly, Martin and Koonin (2006), in an immediate
exist at these early stages. The universal phylogenetic tree,
response to Pace’s challenge in Nature, point out that the absence
therefore, is not an organismal tree at its base but gradually
of nuclear membranes makes possible the coupling of transcrip-
becomes one as its peripheral branchings emerge. The universal
tion and translation and several prokaryote-specific types of
ancestor is not a discrete entity. It is, rather, a diverse community
regulation, a seemingly ‘‘positive’’ feature. Coupling might
of cells that survives and evolves as a biological unit. This com-
necessitate the absence of spliceosomal introns, which could be
munal ancestor has a physical history but not a genealogical one.
‘‘positivized’’ for prokaryotes as ‘‘having continuous genes.’’
Over time, this ancestor refined into a smaller number of increas-
Coupling might also mandate more direct forms of regulating
ingly complex cell types with the ancestors of the three primary
gene expression (Payankaulam et al. 2010). Indeed, there is
groupings of organisms arising as a result. (Woese 1998b,
considerable literature indicating a common prokaryotic reper-
p. 6854)
toire of transcriptional regulatory factors (Bell and Jackson
2001; Peeters and Charlier 2010). It surely is the case that evolution was different in tempo and
There is also an enormous shared prokaryote-specific pool mode in the ancient pre-cellular past. One of us (WFD; Doolittle
of ‘‘operational’’ genes, as illustrated in > Fig. 2.4 and previously and Brown 1994), called this the period of ‘‘progressive
in Walsh and Doolittle (2005). Thus, though (1) Bacteria and Darwinian evolution’’ because many improvements in the
Archaea are indeed distinguishable from eukaryotes and each speed, accuracy, and efficiency of replication, transcription,
other by the character of the ‘‘informational genes’’ of transcrip- and translation that would have tightened the genotype-
tion, translation, and replication, and (2) for many of these genes phenotype coupling and helped cells adapt to environmental
50 a 87 b >1000
707
40
30
0
72
100
20
10
322
Number of Gene Families

2446
BACTERIA
0 1515
10
50 c 45 d 29
40
115
30 515
244
1
145
20
480
10 103
114 56
0
0 10 20 30 40 50 0 10 20 30 40 50
ARCHAEA
. Fig. 2.4
Distribution of gene families in representative archaeal and bacterial genomes. To reconstruct gene families, we performed all-against-all
BLASTP (E-value cutoff of 10 4) searches (Altschul et al. 1997) of open reading frames (ORFs) in 103 selected genomes. Normalized
bitscores from pairwise BLASTP comparisons were used to cluster ORFs into superfamilies using MCL algorithm (inflation parameter I =
1.1; Enright et al. 2002). Superfamilies with more than four members were further divided into gene families using the BRANCHCLUST
pipeline (Poptsova and Gogarten 2007). To select representative genomes among completed sequenced genomes available in GenBank
we used the following criteria: (a) Only genomes larger than 1 Megabases were considered; (b) The largest representative per Archaeal
genus was taken, resulting in 52 genomes; (c) The largest two representatives per bacterial phylum representing at least different genera
were chosen. For broadly represented phyla (such as proteobacteria and cyanobacteria, among others), additional genomes were added
until the number of selected bacterial genomes was comparable to the number of Archaeal ones (total of 51 genomes). The obtained gene
families’ distribution is visualized in a heat map, where the number of gene families present in x archaeal and y bacterial genomes, shown
on X and Y axes, respectively, is color coded on a logarithmic scale (see color legend on the right of the figure; black values corresponds
to 0 families, and white to >1,000 families). Numbers depicted over a heat map correspond to the total number of gene families in
regions, perimeters of which are delineated by blue lines. Panel A. Comparison of all gene families. Near-universal core consists of 87
gene families. While the bulk of gene families present in at least four genomes can be called domain-specific (2,446 for Bacteria and
1,515 for Archaea), only very small fraction of those is universally present in the majority of the genomes of each domain. A substantial
number of genes (1749) are patchily distributed (shared) across the two domains. Panel B. Randomization of 103 genomes (one of five
replicates is shown in this panel) into two groups resulted in a different distribution of genes, suggesting that the distribution of gene
families among the archaeal and bacterial domains is not random. Panel C. Subset of ‘‘informational’’ gene families (as defined by
‘‘Information Storage and Processing’’ functional category of the Clusters of Orthologous Groups, or COG, database (Tatusov et al.
2003)). Panel D. Subset of ‘‘operational’’ gene families (as defined by ‘‘Metabolism’’ functional category of the COG database)
challenges of all sorts should have been under positive selection ‘‘genealogically traceable,’’ and is indeed the achievement of the
then. True ‘‘progess’’ was being made: organisms would ‘‘community as a whole.’’ The difference of course is that with
have been getting better just at being organisms. After that Homo sapiens it is homologous recombination within the species
(after crossing what Woese calls ‘‘the Darwinian threshold’’), that underwrites the process, whereas for prokaryotes or the
adaptations would have become more ecology-driven, tracking ‘‘cell-like entities’’ comprising the progenote, it is LGT (alterna-
environmental change—even ‘‘reversing’’ themselves, and pro- tively HGT) between ‘‘species.’’
gressive only in a highly contingent and local context. Both within-species recombination and between-species
But whenever and at whatever stage we consider the com- LGT/HGT complicate the concept of ‘‘ancestor’’ (as opposed
mon ancestor of Bacteria and Archaea to have appeared, or to ancestry), in a way not often appreciated by microbial
indeed even if we think there was no ‘‘discrete entity’’ deserving phylogeneticists, who tend to equate gene trees with species
of that name, there is no denying that the two domains share trees. In a recombining sexually reproducing species, alleles at
a common ancestry. The universality of the genetic code of the different loci find their last common ancestors (their coales-
ribosome, and of many of its constituent proteins (Fox 2010), of cents) in different genomes present at different times in the
demonstrably homologous RNA polymerase subunits (Werner population’s past. ‘‘Gene trees’’ are uncoupled from ‘‘species
and Grohmann 2011), and of so many operational enzymes and trees’’ and sexually reproducing organisms have organismal
the biochemistries they promote make it impossible to imagine ancestors from whom they have received no genetic informa-
an independent origin of Bacteria and Archaea. That is, they tion. (We receive half our alleles from each parent but it is
could not have gone from prebiotic chemistry through to the possible if ridiculously improbable that we received none from
first cellular forms that we might consider Bacteria or Archaea one of our grandparents, and in fact certain that, many more
without extensive pooling of information. Even Bill Martin and generations back, we have reproductive ancestors from whom
Mike Russell (2003)—whose origin-of-life scenario entails we received no alleles.)
advancement to the stage of ribosomes and the universal code Similarly, because of LGT, the last common ancestral ver-
among pre-cellular systems maintained in FeS compartments— sions of the many gene families found distributed in Bacteria
must invoke extensive information exchange. and Archaea today will trace to common ancestral versions that
Vetsigian et al. (2006) describe the code as the product of existed in different genomes at different times in the history of
‘‘collective evolution’’ via LGT. . . their ancestral global prokaryotic population. The last common
"
ancestral prokaryotic cell quite possibly housed none of them.
The phylogenetic expression of ambiguity is reticulate evolution.
Some researchers think that a small core (as few a 1%) of all
In reticulate evolution, there is no unique notion of genealogical
genes, mostly involved in translation, have never been subjected
descent: genetic content can be distributed collectively. . . The
to LGT, and thus do find their last common ancestral versions in
players are cell-like entities still in early stages of their evolutions.
the genome of a single cellular Last Universal Common Ances-
The evolutionary dynamic (the ‘‘rules’’) involves communal
tor, or LUCA (Ciccarreli et al. 2007). Others, who may doubt the
descent. The key element in this dynamic is innovation-sharing,
existence of any such never-transferred genes but still hope to
an evolutionary protocol whereby descent with variation from
give a traditional meaning to the Tree of Life, assert that, rather
one ‘‘generation’’ to the next is not genealogically traceable but
than being a single individual or even a single species, ‘‘LUCA
is a descent of a cellular community as a whole. Even if an
was a population’’ (or ancestral ‘‘state’’), indeed a heterogeneous
organismal ancestry were to some extent traceable, it would
population extended through time, possibly over the many tens
have no significance, because it is the community as a unit, not
or hundreds of millions of years consumed by ‘‘progenotic,’’ or
the individual organismal lineages therein, that varies in
‘‘progressive Darwinian’’ evolution (Koonin 2009; Glansdorff
descent. . . The central conjecture in our model is that innova-
et al. 2008, 2009; Di Giulio 2011).
tion-sharing, which involves horizontal transfer of genes and
This seems a radically different use of the word and concept
perhaps other complex elements among the evolving entities
of ‘‘ancestor,’’ which is more commonly understood to single out
[a dynamic far more rampant and pervasive than our current
individuals in earlier populations. And we need not buy into
perception of horizontal gene transfer (HGT)], is required to bring
such a redefinition (or indeed any definition of common ances-
the evolving translation apparatus, its code, and by implication
tor) in order to accept a general model of common ancestry, in
the cell itself to their current condition. (Vetsigian et al. 2006,
which the various features of modern cells evolved ‘‘collectively’’
p. 10696)
(i.e., in no single genomic lineage), in a globally communicating
Importantly, and perhaps ironically, this ‘‘dynamic’’ is also (via LGT) super-population of entities becoming, over the
how one might describe the evolution of a complex trait within course of hundreds of millions of years, ever more like contem-
a modern multicellular sexually reproducing species, ours for porary prokaryotic cells.
instance. Because of recombination, to which mammalian A recasting of the question asked earlier, whether there is any
reproduction is inextricably tied, any multigenic advance is reason to believe that Bacteria and Archaea derive from
similarly going to be the product of ‘‘innovation sharing.’’ a common ancestor that might itself be called a prokaryote, is
Though each contributing mutant allele should be traceable to this. ‘‘Would that ancestral cell or cells in which the last common
a last common ancestor (its ‘‘coalescent’’) arising earlier in the ancestral versions of genes that (it is assumed) have never been
population, the complex trait as a whole will not be laterally transferred – in particular ribosomal RNA and protein
genes – be more ‘‘primitive’’ in terms of its informational become the mitochondrion. And evidence favoring the ‘‘eocyte’’
machineries (their accuracy, efficiency or general character) hypothesis (Cox et al. 2008; Foster et al. 2009) also implies that
than are contemporary prokaryotes?’’ This is a formulation LECA’s ribosomes were already fully functional archaeal types.
that we think most authors who embrace either a single-cell or We cannot from comparative ribosomology alone decide
heterogenous population notion of LUCA might accept. We whether the ancestral cell or cells in which the last common
submit that we do not, in spite of the great appeal of the ancestral versions of ribosomal RNA and ribosomal protein
progenote concept and the various ingenious but tortured argu- genes resided were ‘‘progenotes’’ or ‘‘prokaryotes.’’ If progenotes,
ments put forth in favor a primitive, even RNA-genomed ances- then ‘‘prokaryote’’ becomes polyphyletic, as well as paraphyletic,
tral state (Koonin 2009; Glansdorff et al. 2008, 2009; Di Giulio and the word would have two strikes against it. But this second
2011), know the answer to it. strike depends entirely on how we define both ‘‘prokaryote’’ and
To be sure, Woese’s observation that the many structural ‘‘progenote,’’ not ‘‘meaningful concerns that can be addressed
differences between the informational machineries of Bacteria scientifically’’ and on the existence and nature of LUCA, which is
and Archaea, particularly the ribosomes, reflect their indepen- similarly problematic.
dent evolutionary refinement from a more primitive (less accu-
rate and efficient) ancestral form is the best-articulated
argument along these lines. In concluding a recent survey of Bacteria and Archaea, and the Fundamental
domain-specific features (‘‘signatures’’) of bacterial and archaeal Differences Between Them
ribosomes, he and colleagues (Roberts et al. 2008) wrote . . .
"
Pace’s fourth argument is that disregard of these differences ‘‘rein-
Through our analysis of ribosomal signatures, we have provided
forces an incorrect understanding of biological organization and
a glimpse into the evolutionary past, at the ‘‘base’’ of the [uni-
evolution’’ (Pace 2006). To be sure (rephrasing Stanier et al.
versal phylogenetic tree]. This study has identified the ribosomal
1963) the ‘‘basic divergence in cellular structure which separates
signatures and provided examples of how they are helpful in
the bacteria [and archaea] from each other represents the
understanding the evolutionary dynamic by which the ribosome
greatest single evolutionary discontinuity to be found in the
arose. These signatures give each phylogenetic domain
present day [prokaryotic world].’’ In the next chapter, Woese
a distinctive character and bespeak stages through which the
summarizes those differences—in membrane composition and
evolution of the ribosome must have proceeded, both before the
the machineries of translation, transcription, and replication—
emergence of the individual lineages themselves (in the univer-
that have long been considered diagnostic of the bacterial/
sal ancestral state) and subsequently, separately within each
archaeal discontinuity. It is also the case that ‘‘operational’’
primary lineage. (Roberts et al. 2008, p. 13958)
genes (primarily those of catabolism and anabolism), though
But the differences between archaeal and bacterial ribosomes widely shared between Bacteria and Archaea, are nevertheless
and the very long branches between bacterial and archaeal often preferentially associated with one or the other domain.
domains in rRNA and many other gene trees are not necessarily The gene family presence/absence data presented in > Fig. 2.4
evidence of a different mode and tempo of evolution or of and the comparison to a random partitioning of genes make this
a primitive LUCA at the root of any gene tree. All bifurcating abundantly clear.
trees must have a deepest branching that divides them into two The ‘‘incorrect understanding’’ that concerns Pace is the
subtrees. On average (assuming a random birth-death model) presumption of uniformity among prokaryotes. Stanier and
the two branches will themselves appear unbranched halfway up van Niel were themselves cognizant and celebratory of bacterial
the tree, this being the null model under coalescence theory diversity (Spath 1999; Sapp 2006, 2009a, b), though they were
(Zhaxybayeva and Gogarten 2004). Unusually rapid evolution- not aware of the ‘‘discontinuity’’ between Bacteria and Archaea
ary rates and ancestral primitivity need not be invoked. (nor was anyone at the time). It seems unlikely that they would
Moreover, there are selective (Gupta 2000) and nonselective have been dismissive of the kinds of evidence Woese brings to
(Lukes et al. 2011) forces other than improvement from bear. Indeed, Woese lays much of the blame for what he sees as
a primitive state that might radically transform the translational a woeful neglect of difference on molecular biologists, whose
machinery. Mitochondrial ribosomes, for instance, are vastly reductionist paradigm led them to think that what is true for
different in composition and structure from their alpha- E. coli is true for elephants and (coincidentally) for Bacillus
proteobacterial ancestors (O’Brien 2002), and eukaryote ribo- subtilis and Sulfolobus solfataricus.
somes are larger and more complex than their presumed It is not just the bacterial/archaeal distinction that is painted
archaea-like predecessors. If we accept the scenario in over by the broad brush of molecular biology, however. There
> Fig. 2.3e, these latter differences might bespeak separately are fungi whose molecular genetics are wildly unlike that of
evolved refinements of a primitive progenotic ancestral ribo- Saccharomyces cerevisiae, insects whose population genetics are
some. But most would favor phylogenies more like those in not like that of Drosphila melanogaster, vertebrates whose devel-
> Fig. 2.3f–h. We can infer from the similarity across bacterial opment is strikingly different from that of Gallus domesticus, and
phyla that the bacterial ribosome was already in the final form at primates who do not behave much like us. It is not surprising
the time LECA (the last eukaryote common ancestor) welcomed that biologists who have spent their lives with one model organ-
on board the alpha-proteobacterial symbiont that was to ism want their results to be seen as more generally valid and that
cell and molecular and cell biology textbooks follow their lead, Cyanobacteria, the Proteobacteria, and the Thermoplasmatales,
while teachers of evolution and diversity defend differences but an archaebacterial (euryarchaeotes) group. These signals corre-
reach fewer students. spond to distinct symbiotic partners involved in eukaryote evo-
As Woese himself noted in rebuttal to Mayr, ‘‘Diversity can lution: plastids, mitochondria, and the elusive host lineage.
be of many types,’’ and what we emphasize will depend on our (Pisani et al. 2007, p. 1752)
experience, disciplinary focus, and research goals. That Bacteria
Why, if the Archaeal signal is the weakest, is the Archaeal/
and Archaea appear different in many fundamental biological
eukaryote sister relationship enshrined in textbooks, and the
features is beyond question: whether this degree of difference is
concepts of paraphyly employed so enthusiastically by Pace in
dwarfed by that between prokaryotes and eukaryotes is not
his effort to discredit ‘‘prokaryote’’? Of course, even if the pre-
something that science can decide. This is, as Woese noted in
ponderance of data were all that mattered, prokaryotes would
the passage quoted earlier, to compare apples to oranges.
still be paraphyletic, with eukaryotes seen as sisters to the
Cyanobacteria or Proteobacteria. But the more we come to see
eukaryotes as a chimeric clade, established by symbiosis, cell
Reticulation, the Bête Noire of Cladistics fusion, and LGT, through the active participation of several or
many prokaryotic lines, the more aptly descriptive the ‘‘eu’’ and
Trees based on rRNA are unrootable, because there is no ‘‘pro’’ in ‘‘eukaryotes’’ and ‘‘prokaryotes’’ begin to look. This
outgroup. The accepted rooted version of the three-domain would have been all along the position of the late Lynn Margulis,
rRNA tree (as in > Fig. 2.2) was initially based on a few widely coincidentally (Margulis 1996).
distributed paralogous protein-coding gene families, assumed to In any case, the language of cladistics and the tree-like model
be the products of gene duplications that predated LUCA (Iwabe on which it is based seem almost irrelevant in the face of such
et al. 1989; Gogarten et al. 1989; Brown and Doolittle 1994). The reticulation, or non-tree like signal.
tree based on sequences of one of the duplicates can be used to " Reticulation is thus the bête noire for cladistics, as initially recog-
root that of the other and vice versa, the two trees being in
nized by Hennig. There are a number of sources of homoplasy
principle congruent with each other and with the true Tree of
(incongruency between certain character distributions and the
Life (assuming no LGT). Such duplication-based rootings gen-
cladogram based on maximum parsimony), such as adaptive
erally showed eukaryotes and Archaea to be sisters, an inference
convergence, gene conversion, developmental constraints, mis-
supported by strong and specific similarities between elements
taken coding, and reticulation. The last-named factor is the most
of the archaeal and eukaryotic replication, transcription, and
problematical because it involves the fundamental model of
translation machineries (Pace et al. 2012).
reality underlying cladistic analysis. (Mishler and Theriot 2000,
Of course it was known since the early seventies that there
p. 50)
would be some bacterial contribution to the eukaryotic nuclear
genome, via the mitochondrial symbiosis and organelle-to- Indeed, in 1975, Peter Sneath, a microbiologist and leading
nucleus gene transfer. The initial assumption that this contribu- theoretical systematist, noted that ‘‘reticulate evolution requires
tion would be small—and perhaps a prejudice that it is the consideration of ways to represent the descent of genes, instead
phylogeny of the host that matters most—led to the now largely of entire genomes as in simple branching cladograms’’ (Sneath
unquestioned representation of eukaryotes as sister to Archaea. 1975). In 1992, the systematist David Mindell eloquently
But unexpectedly, estimates of the bacterial contribution defended
have grown enormously as genomic data pour in. In 2004, " . . . the idea that the history of no one symbiont should take
Esser et al., described a surprising result. . .
precedence over another in assessing genealogy (monophyly)
" . . . approximately 75% of yeast genes having homologues among and classification, regardless of size, weight, status as host, or
the present prokaryotic sample share greater amino acid other measure of relative dominance . . . To discount one symbi-
sequence identity to eubacterial than to archaebacterial homo- ont within a holobiont when reconstructing the holobiont’s
logues. At high stringency comparisons, only the eubacterial com- overall phylogenetic history and proposing a classification . . . is
ponent of the yeast genome is detectable. Our findings indicate equivalent to disposal of data. (Mindell 1992, p. 57)
that at the levels of overall amino acid sequence identity and gene
If monophyly is to retain its meaning as containing all
content, yeast shares a sister-group relationship with eubacteria,
descendants of a single ancestor and that ancestor—and no
not with archaebacteria, in contrast to the current phylogenetic
data are to be disposed of—then eukaryotes are not monophy-
paradigm based on ribosomal RNA. (Esser et al. 2004, p. 1643)
letic, nor are Bacteria or Archaea, because of LGT.
More recently, Pisani et al. (2007), using multiple (including A compromise that might allow for a principled if not
photosynthetic) eukaryotic genomes and a larger collection of unarguably ‘‘natural’’ classification in the face of reticulation
bacterial and archaeal genomes reached a similarly startling was suggested in 2007 by David Baum, a noted plant systematist.
conclusion. . . " A primary concordance tree, a tree built from clades that are true
" . . . there are three distinct phylogenetic signals in eukaryotic of a plurality of the genome, provides a useful summary of the
genomes. In order of strength, these link eukaryotes with the dominant phylogenetic history for a group of organisms.
Residual historical signals can also be extracted in the form of Coda

secondary, tertiary, etc. concordance trees, which are built from
clades that are present in the genome but contradict clades on Natural kind thinking has been part of biological classification
the primary concordance tree. (Baum 2007, p. 417) for millennia, of course, as many philosophers of biology and
biologists who practice philosophy have noted. Ernst Mayr
Most of us would react with horror to a proposal to consider called it ‘‘typological thinking.’’ The most extreme form of
reclassifying non-photosynthesizing eukaryotes as sister to such thinking has been described (and then rejected) by the
proteobacteria and photosynthetic eukaryotes as sister to philosopher Ian Hacking as the belief that: . . .
cyanobacteria (though this last view was common until the " There is a unique best taxonomy in terms of natural kinds, that
mid-twentieth century). But this would be arguably more con-
represents nature as it is, and reflects the network of causal laws.
sistent with Darwin’s call for genealogical classification, and
We do not have nor could we have a final taxonomy of anything,
would be based on the same facts and the same understanding
but any objective classification is right or wrong according as it
of what has happened during the evolution of genes, genomes,
captures part of the structure of the one true taxonomy of the
and organisms as is the ‘‘phylogenetic classification’’ enshrined
universe. (Hacking 1991, p. 111)
in the hegemonic three-domain tree.
It is only by ‘‘disposal of data’’—allowing rRNA and a few Such a view, which we too reject, underlies the debate about
associated informational genes to stand in for an evolutionary whether there are three rather than two kinds of living things in
lineage as a whole—that inferences about monophyly, this world. To be sure, there are more or less reasonable ways to
polyphyly, or paraphyly of the three domains can be made to look at diversity. But as to how many kinds there actually are,
appear clean. Two assumptions lie behind this maneuver. First is there should be no more definitely true answer than there would
that informational genes are less likely to be transferred, espe- be to ‘‘How many kinds of people are there in North America?’’
cially across large phylogenetic distances, because of the com- or ‘‘How many kinds of books are there in my town’s public
plexity of the interaction of their products with those of other library?’’
genes (Jain et al. 1999). Thus, they track the history of individual And yet, such natural kind thinking comes naturally to
cell divisions and speciation events better than other genes. The biologists, generating much of the heat in our arguments. Insis-
second, more presumption than assumption, is that it is in this tence that the grade distinction between prokaryotes and
history that a true genealogical classification most naturally eukaryotes must be discarded because it conflicts with the divi-
rests. sion of living things into clades reflects a belief that ‘‘the one true
The first is indeed plausible from several perspectives. Inter- taxonomy of the universe’’ is the Tree of Life—to which concept,
estingly, Cotton and McInerney (2010) recently published as loyal Darwinists, we feel we must cling. And yet, because of
a paper whose title says much about the privileging of the chimerism and extensive LGT, that Tree means much less as
archaeal trace. In ‘‘Eukaryotic genes of archaebacterial origin a description of what organisms are like and how they came to
are more important than the more numerous eubacterial be that way than Darwin himself would have ever imagined
genes, irrespective of function’’ they show that those (quantita- (Doolittle 1999; Martin 2011).
tively) fewer yeast eukaryotic nuclear genes that show archaeal Furthermore, the Tree unambiguously depicts three dis-
sisterhood are disproportionately ‘‘essential to yeast viability, are cretely defined monophyletic clades if and only if we accept as
more highly expressed, and are significantly more highly settled many highly arguable propositions about how evolution
connected and more central in the yeast protein interaction proceeds, about what the data say in that regard, and about how
network,’’ as might be expected if the archaeal heritage were classification is properly to be pursued. Some of these proposi-
more stably associated with the less frangible aspects of cell tions, reviewed here, are that (1) in spite of the testimony of
biology. most of their genes, the ‘‘one true’’ way to look at Eukarya is as
The second, that even when a majority of genes contradict sisters to or emerging from within Archaea; (2) in spite of very
the informational signal, it is the latter we should rely on in sophisticated recent analyses of those genes thought to be most
classification, is not a question of science, answerable by exper- truthful, sisterhood (rather than emergence from within) is the
iment. Classification has less to do with identifying ‘‘natural right way to understand this relationship; (3) Bacteria and
groups’’ and more to do with the philosophy, history, and Archaea are themselves properly described as monophyletic
politics of the idea of ‘‘natural’’ than we generally like to admit. clades, even though it is but a few percent of their genes that
There is much intellectual momentum behind this approach, tell us this, with the preponderance of genes saying that they
but there are serious alternatives (Cavalier-Smith 2002, 2007) comprise a single, albeit highly structured and admittedly very
that cannot be dismissed out of hand, simply because they have slowly mixing population; (4) that it makes sense to speak of
fewer adherents. There is in fact no principled way, given mas- LUCA, either as a single cell or species or (alternatively) as
sive reticulation, of deciding what is the most ‘‘natural’’ way to a heterogeneous population extended over time; (5) that it
classify organisms, or divide all living things up into ‘‘kinds.’’ makes sense to describe LUCA (of whatever type) as something
The various positions taken by microbiology’s leaders, though other than a prokaryote. None of these is provably wrong, but
bolstered by appeals to scientific evidence, are in the end none will ever be proved right, and all are matters of opinion,
rhetorical. not fact. Definition-dependent, supposedly logic-driven
arguments on the use of ‘‘prokaryote’’ seem doomed to failure by to differential relevance of datasets. For the cell biologist of the
the same criteria with which they are undertaken, coupled with school of classification Mayr called ‘‘Darwinian,’’ focusing on
the vast underdetermination of any of our current theories the translational machinery is parochial. And for the microbi-
about early cellular evolution (Vesteg and Krajcovic 2011; ologist working with Bacteria or Archaea, although there are
Forterre 2011). important differences that should inform experiments on gene
Moreover, much of the debate is about the definition of structure and function, in the study of metabolism, regulation,
terms, of which Popper wrote: population genetics, and ecology, this is much less the case.
"
Without knowing in advance it would, in many instances, be
In science, we take care that the statements we make should
hard to say whether a particular published paper on these topics
never depend on the meaning of our terms. Even where the
had a bacterium or an archaeon as its study object.
terms are defined, we never try to derive any information from
It is not necessary to imagine that the world holds some
the definition, or to base any argument upon it. . . . We are always
particular number of basic types in order to understand it, or
conscious that our terms are a little vague (since we have learned
advance biological knowledge. Almost certainly ‘‘prokaryote’’
to use them only in practical applications) and we reach precision
glosses over the many differences between Bacteria and Archaea,
not by reducing their penumbra of vagueness, but rather by
but ‘‘Bacteria’’ glosses over many differences between mycoplas-
keeping well within it, by carefully phrasing our sentences in
mas and planctomycetes, and by what criteria can we judge one
such a way that the possible shades of meaning of our terms
glossing-over more egregious than the other? Prokaryota is
do not matter. This is how we avoid quarrelling about words.
indeed not a legitimate taxonomic group by the principles of
(Popper 2003, pp. 21–22)
cladists, nor are Eukarya, Bacteria, and Archaea, unless we
So, in a more relaxed context, but still one in which we strive ignore or rework those principles to the point that they are no
to keep well within Popper’s ‘‘penumbra of vagueness,’’ is there longer recognizable or useful. To avoid quarreling about words,
anything sensible to say in defense of ‘‘prokaryote’’? we need first to realize that that is what we are doing.
As a phenetically based grade distinction, the prokaryote: The continued use of ‘‘prokaryote’’ with a small ‘‘P’’ seems to
eukaryote dichotomy resonates so strongly largely because us justified by history and utility. If indeed it confuses students
eukaryotes seem so different from either Archaea or Bacteria, to teach them this ‘‘wrong idea,’’ this will be because we have
and all so much like each other, at the level of the cell. Four dumbed our science too far down. Those who come to see that
decades ago, the prevailing view was that eukaryotes arose from ‘‘prokaryote’’ is neither right nor wrong and that its meaning is
within prokaryotes by the gradual evolution of cytoskeleton and contextual will achieve a deeper understanding of microbiology,
endomembrane systems—driven by selection for the acquisition and of science as a human endeavor.
of phagotrophy and followed by the establishment of the endo-
symbionts destined to become mitochondria and plastids (Gray
and Doolittle 1982). Theory predicted that there might be some
surviving primitively amitochondriate phagotrophs still out
Acknowledgments
there. But so far none has been found, and most of the complexi-
This chapter is dedicated to the memory of Lynn Margulis.
fication of internal systems believed to be characteristically
We thank John Archibald and Andrew Roger for comments.
eukaryotic were apparently already in place in LECA. Arguably,
Original work described in this manuscript was supported by
there was very rapid, selection-directed modification of many
the Canadian Institutes of Health Research (WFD) and by the
archaeal-type genes (of the ‘‘important’’ kind identified by
TULA Foundation (OZ).
Cotton and McInerey), such that many now are hard to align
with their still-prokaryotic homologs. To cell biologists working
on the eukaryote side of that ‘‘great discontinuity,’’ it is the ‘‘pro’’
in prokaryotes that gives meaning to the term. Especially for References
symbionticists like Lynn Margulis (1996), but even for more
conservative theorists like John Maynard Smith and Eors Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ
Szathamary (1995), the chimeric fusion event(s) that forged (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database
eu- from prokaryotes was a profoundly important evolutionary search programs. Nucleic Acids Res 25:3389–3402
transition. Angert ER (2005) Alternatives to binary fission in bacteria. Nat Rev Microbiol
3:214–224
Whether archaeal/bacterial differences are comparable in
Ashlock PD (1974) The uses of cladistics. Ann Rev Ecol Syst 5:81–99
importance to the eukaryote:prokaryote distinction is indeed Baum DA (2007) Concordance trees, concordance factors, and the exploration of
a matter of ‘‘comparing apples to oranges.’’ Woese in rebutting reticulate genealogy. Taxon 56:417–426
Mayr asserted that his (Mayr’s) ‘‘attempt to apply globally Bell SD, Jackson SP (2001) Mechanism and regulation of transcription in archaea.
a parochial and subjectively defined concept of diversity serves Curr Opin Microbiol 4:208–213
Bird A, Tobin E (2010) Natural kinds. In: Zalta EN (ed) The Stanford encyclope-
only to reveal the futility in such an approach.’’ But it is hard to
dia of philosophy, summer 2010 edn. http://plato.stanford.edu/archives/
imagine concepts of phenotypic diversity applicable to both sum2010/entries/natural-kinds/
prokaryotes and eukaryotes that are not ‘‘parochial and subjec- Brown JR, Doolittle WF (1994) Root of the universal tree based on aminoacyl-
tively defined,’’ at the very least insofar as they involve choices as tRNA gene duplications. Proc Natl Acad Sci USA 92:2441–2445
Cavalier-Smith T (2002) The neomuran origin of archarebacteria, the Gupta RS (2000) The natural evolutionary relationships among prokaryotes. Crit
negibacterial root of the universal tree and bacterial megaclassification. Int Rev Microbiol 26:111–131
J Syst Evol Microbiol 52:7–76 Guy L, Ettema TJG (2011) The archaeal ‘‘TACK’’ superphylum and the origin of
Cavalier-Smith T (2007) Concept of a bacterium still valid in prokaryote debate. eukaryotes. Trends Microbiol 19:580–587
Nature 446:257 Hacking I (1991) A tradition of natural kinds. Philos Stud 61:109–126
Chatton E (1937) Titres et travaux scientifiques. Sette, Sottano, Italy Iwabe N, Kuma K, Hasegawa M, Osawa S, Miyata T (1989) Evolutionary
Cotton JA, McInerney JO (2010) Eukaryotic genes of archaebacterial origin are relationship of archaebacteria, eubacteria and eukaryotes inferred
more important than the more numerous eubacterial genes, irrespective of from phylogenetic trees of duplicated genes. Proc Natl Acad Sci USA
origin. Proc Natl Acad Sci USA 107:17252–17255 86:9355–9459
Cox CJ, Foster PG, Hirt RP, Harris SR, Embley TM (2008) The archaebacterial Jain R, Rivera MC, Lake JA (1999) Horizontal gene transfer among genomes: the
origin of eukaryotes. Proc Natl Acad Sci USA 105:20356–20361 complexity hypothesis. Proc Natl Acad Sci USA 96:3801–3806
Darnell JE, Doolittle WF (1986) Speculations on the early course of evolution. Koonin EV (2009) On the origin of cells and viruses: primordial virus world
Proc Natl Acad Sci USA 83:1271–1275 scenario. Ann NY Acad Sci 1178:47–64
Darwin CR (1859) On the origin of species by means of natural selection, or Lake JA, Henderson E, Oakes M, Clark MW (1984) Eocytes: a new ribosome
the preservation of favoured races in the struggle for life, 1st edn. John structure indicates a kingdom with a close relationship to eukaryotes. Proc
Murray, London Natl Acad Sci USA 81:3786–3790
Di Giulio M (2011) The Last Universal Common Ancestor (LUCA) and the Lovejoy AO (1936) The great chain of being. Harvard University Press,
ancestors of Archaea and Bacteria were progenotes. J Mol Evol 72:119–126 Cambridge
Dolan MF, Margulis L (2007) Advances in biology reveal truth about prokaryotes. Lukes J, Archibald JM, Keeling PJ, Doolittle WF, Gray MW (2011) How a neutral
Nature 445:21 evolutionary ratchet can build cellular complexity. IUBMB Life 63:528–537
Doolittle WF (1999) Phylogenetic classification and the universal tree. Science Makarova KS, Yutin N, Bell SD, Koonin EV (2010) Evolution of diverse cell
284:2124–2129 division and vesicle formations systems in Archaea. Nat Rev Microbiol
Doolittle WF, Brown JR (1994) Tempo, mode, the progenote, and the universal 8:731–741
root. Proc Natl Acad Sci USA 91:6721–6728 Margulis L (1996) Archaeal-eubacterial mergers in the origin of Eukarya: phylo-
Dupré J (1981) Natural kinds and biological taxa. Philos Rev 90:66–90 genetic classification of life. Proc Natl Acad Sci USA 93:1071–1076
Embley TM, Martin W (2006) Eukaryote evolution, changes and challenges. Martin WF (2011) Early evolution without a tree of life. Biol Direct 6:36
Nature 440:623–630 Martin W, Koonin E (2006) A positive definition of prokaryotes. Nature 442:24
Enright AJ, Van Dongen S, Ouzounis CA (2002) An efficient algorithm for large- Martin WF, Russell MJ (2003) On the origins of cells: a hypothesis for the
scale detection of protein families. Nucl Acids Res 30:1575–1584 evolutionary transitions from abiotic geochemistry to chemoautotrophic
Esser C, Ahmadinejad N, Wiegand C, Rotte C, Sebastiani F, Gelius-Dietrich G, prokaryotes, and from prokaryotes to nucleated cells. Philos Trans R Soc
Henze K, Kretschmann E, Richly E, Leister D, Bryant D, Steel MA, Lond B Biol 358:59–63
Lockhart PJ, Penny D, Martin W (2004) A genome phylogeny for mitochon- Maynard Smith J, Szathmary E (1995) The major transitions in evolution. Oxford
dria among alpha-proteobacteria and a predominantly eubacterial ancestry University Press, Oxford
of yeast nuclear genes. Mol Biol Evol 21:1643–1660 Mayr E (1982) The growth of biological thought. Harvard University Press,
Field MC, Dacks JB (2009) First and last ancestors: reconstructing evolution of Cambridge, MA
the endomembrane system with ESCRTs, vesicle coat proteins, and nuclear Mayr E (1990) A natural system of organisms. Nature 348:491
pore complexes. Curr Opin Cell Biol 21:4–13 Mayr E (1991) More natural classification. Nature 353:122
Forterre P (2011) A new fusion hypothesis for the origin of Eukarya: better than Mayr E (1998) Two empires or three? Proc Natl Acad Sci USA 95:9720–9723
previous ones, but probably also wrong. Res Microbiol 162:77–91 McInerney JO, Martin WF, Koonin EV, Allen JF, Galperin MY, Lane N, Archibald
Foster PG, Cox CJ, Embley TM (2009) The primary divisions of life: JM, Embley TM (2011) Plactomycetes and eukaryotes: a case of analogy not
a phylogenomic approach employing composition-heterogeneous methods. homology. Bioessays 33:810–817
Philos Trans Roy Soc B 364:2197–2207 McOuat G (2009) The origins of ‘‘natural kinds’’: keeping ‘‘essentialism’’ at bay in
Fox GE (2010) Origin and evolution of the ribosome. Cold Spr Harbor Perspect the Age of Reform. Intell Hist Rev 19:211–230
Biol 2010:a003483 Mindell DP (1992) Phylogenetic consequences of symbioses: Eukarya and
Fox-Keller E (2010) The mirage of a space between nature and nurture. Duke eubacteria are not monophyletic taxa. Biosystems 27:53–62
University Press, Durham Mishler BD, Theriot EC (2000) The phylogenetic species concept (sensu Mishler
Fuerst JA, Sagulenko E (2011) Beyond the bacterium: planctomycetes challenge our and Theriot): monophyly, apomorphy and phylogenetic species concepts. In:
concepts of microbial structure and function. Nat Rev Microbiol 9:403–415 Wheeler QD, Meier R (eds) Species concepts and phylogenetic theory.
Glansdorff N, Xu Y, Labedan B (2008) The Last Universal Common AncestoR: Columbia University Press, New York
emergence, constitution and genetic legacy of an elusive forerunner. Biol O’Brien TW (2002) Evolution of a protein-rich mitochondrial ribosome: impli-
Direct 3:29 cations for human genetic disease. Gene 286:73–79
Glansdorff N, Xu Y, Labedan B (2009) The origin of life and the last universal O’Hara, Robert J (1997) Population thinking and tree thinking in systematics.
common ancestor: do we need a change of perspective? Res Microbiol Zool Scr 26(4):323–329
160:522–528 O’Hara RJ (1998) Population thinking and tree thinking in systematics.
Gogarten JP, Kibak H, Dittrich P, Taiz L, Bowman EJ, Bowman BJ, Manolson MF, Zoologica Scripta 26:323–329
Poole RF, Date T, Oshima T, Konishi J, Denda K, Yoshida M (1989) Evolution O’Malley MA (2010) Ersnt Mayr, the tree of life, and philosophy of biology. Biol
of the vacuolar ATPases: implications for the origin of eukaryotes. Proc Natl Philos 25:529–552
Acad Sci USA 86:6661–6665 Pace NR (2006) Time for a change. Nature 441:289
Golinski J (1998) Making natural knowledge: Constructivism and the history of Pace NR (2009) Rebuttal: the modern concept of the prokaryote. J Bacteriol
science. University of Chicago Press, Chicago 191:2006–2007
Grant V (2003) Incongruence between cladistic and taxonomic systems. Am J Bot Pace NR, Sapp J, Goldenfeld N (2012) Phylogeny and beyond: scientific, histor-
90:1263–1270 ical, and conceptual significance of the first tree of life. Proc Natl Acad Sci
Gray MW, Doolittle WF (1982) Has the endosymbiont hypothesis been proven? USA. Early edition on line
Microbiol Rev 46:1–42 Payankaulam S, Li ML, Arnosti DN (2010) Transcriptional repression: conserved
Gribaldo S, Poole AM, Daubin V, Forterre P, Brochier-Armanet C (2010) The and evolved features. Curr Biol 20:R764–R771
origin of eukaryotes and their relationship with the Archaea: are we at Peeters E, Charlier D (2010) The Lrp family of transcription regulators in archaea.
a phylogenomic impasse? Nat Rev Microbiol 8:743–752 Archaea 2010:750457
Penny D, Poole A (1999) The nature of the last universal common ancestor. Curr Tatusov RL, Fedorova ND, Jackson JD, Jacobs AR, Kiryutin B, Koonin EV, Krylov
Opin Genet Dev 9:672–677 DM, Mazumder R, Mekhedov SL, Nikolskaya AN, Rao BS, Smirnov S,
Pisani D, Cotton JA, McInereny JO (2007) Supertrees disentangle the chimerical Sverdlov AV, Vasudevan S, Wolf YI, Yin JJ, Natale DA (2003) The COG
origin of eukaryotic genomes. Mol Biol Evol 24:1752–1760 database: an updated version includes eukaryotes. BMC Bioinformatics 4:41
Popper K (2003) The open society and its enemies, vol 2. Routledge, London (first Vesteg M, Krajcovic J (2011) The falsifiability of the models for the origin of
published 1945) eukaryotes. Curr Genet 57:367–390
Poptsova MS, Gogarten JP (2007) BranchClust: a phylogenetic algorithm for Vetsigian K, Woese D, Goldenfeld N (2006) Collective evolution and the genetic
selecting gene families. BMC Bioinformatics 8:120 code. Proc Natl Acad Sci USA 103:10696–10701
Roberts E, Sethi A, Montoya J, Woese CR, Luthey-Schluten Z (2008) Walsh DA, Doolittle WF (2005) The real ‘domains’ of life. Curr Biol 15:
Molecular signatures of ribosomal evolution. Proc Natl Acad Sci USA R237–R240
105:13953–13958 Werner F, Grohmann D (2011) Evolution of multisubunit RNA polymerases in
Sapp J (2006) Two faces of the prokaryote concept. Int Microbiol 9:163–172 the three domains of life. Nat Rev Microbiol 9:85–98
Sapp J (2009a) The prokaryote-eukaryote dichotomy: meanings and mythology. Whitman WB (2009) The modern concept of the prokaryote. J Bacteriol
Microbiol Mol Biol Rev 69:292–305 191:2000–2005
Sapp J (2009b) The new foundations of evolution. Oxford University Press, Winsor MP (2006) Linnaeus’s biology was not essentialist. Ann Missouri Botan-
Oxford ical Garden 93:2–7
Sneath PHA (1975) Cladistic representation of reticulate evolution. Syst Zool Woese CR (1998a) Default taxonomy: Ernst Mayr’s view of the microbial world.
24:360–368 Proc Natl Acad Sci USA 95:11043–11046
Spath SB (1999) C.B. van Niel and the culture of microbiology, 1920–1965. PhD Woese C (1998b) The universal ancestor. Proc Natl Acad Sci USA 95:6854–6859
thesis, History, University of California, Berkeley Woese CR, Kandler O, Wheelis ML (1990) Towards a natural system of organisms:
Stanier RY, Doudoroff M, Adelberg EA (1963) The microbial world. Prentice- proposal for the domains Archaea, Bacteria and Eucarya. Proc Natl Acad Sci
Hall, Engelwood Cliffs, NJ USA 87:4576–4579
Stanier RY, van Niel CB (1962) The concept of a bacterium. Arch Microbiol Zhaxybayeva O, Gogarten JP (2004) Cladogenesis, coalescence and the evolution
42:17–35 of three domains of life. Trends Genet 20:182–187
3 Prokaryotes and Their Habitats
Hans G. Schlegel . Holger W. Jannasch∗
Göttingen Academy of Sciences and Humanities, Göttingen, Germany
Ecological Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Gradients as Habitats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Ecosystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Habitats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Gradients of a Macroscale: The Stratified Lake . . . . . . . . . . . . 56
Inhabitants of an Ecosystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Gradients of a Microscale: The Sediments . . . . . . . . . . . . . . . . . . 58
Ecological Niches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Horizontal Gradients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Bacterial Metabolism as an Ecological Determinant . . . . . . . 43 Possible Mechanisms for Finding or Remaining in the
Beneficial Layer of the Gradient . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Modes of Energy Conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Chemotaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Aerotaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Anaerobic Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Phototaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Magnetotaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Growth with Inorganic Electron Donors . . . . . . . . . . . . . . . . . . . 45 Buoyancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Fixation of Molecular Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Microbial Associations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Range of Organic Substrates for Growth . . . . . . . . . . . . . . . . . . . 45 Interspecies Hydrogen Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Extremes of Environmental Conditions Allowing Bacterial Bidirectional Transfer of Small Molecules . . . . . . . . . . . . . . . . . . 62

Growth and Survival . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Low Temperatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Specific Aquatic Ecosystems: Marine Environments . . . . . . . . 62
High Temperatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Water Stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 The Deep Sea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Salinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Hydrothermal Vents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Environments of Extreme pH Values . . . . . . . . . . . . . . . . . . . . . . . 49
Acid Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Marine Anoxic Ecosystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Alkaline Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Eukaryotes as Habitats for Bacteria . . . . . . . . . . . . . . . . . . . . . . . . 67
Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Animal Habitats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Low-Nutrient Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Intestinal Tract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Human Intestinal Tract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Resting Stages and Survival . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 Rumen and Reticulum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Hindgut Fermentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Light as an Extreme Environment . . . . . . . . . . . . . . . . . . . . . . . . . . 52 Bacteria of the Human Skin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Surfaces as Habitats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Molecular Microbial Ecology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Liquid–Solid Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Liquid–Gas Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
∗
Deceased
40 3 Prokaryotes and Their Habitats
Prokaryotes are well recognized as essential members of the reflects the development of metabolism during the evolution of
biosphere. They inhabit all possible locations in which life exists life, enables them to live in many parts of the biosphere, includ-
from those offering ideal conditions for growth and reproducing several where eukaryotes are not able to exist.
tion to those representing extreme environments at the border- The vegetative microbial cell, with its relatively large reactive
line of abiotic conditions. surface, responds quickly to changing physicochemical conditions
The ubiquity of microorganisms is based on three major of its immediate surrounding. As a consequence, the effective
properties: their small size for easy dispersal by air and water, habitat of a microorganism is its microhabitat, the immediate
their metabolic versatility and flexibility, and their ability to surrounding of the cell in a compatible scale of space and time as
tolerate unfavorable conditions. A predominant population is determined by the radius of its metabolic action and interaction.
commonly composed of species able to grow under the Naming microorganisms for their occurrence in certain char-
particular conditions of a habitat. Many other species may acteristic macrohabitats, for example, soil and water bacteria, is
also be present but in low numbers of individuals. As a rule, of limited use. The two apparently very different habitats, soil
ecosystems of indistinct physicochemical and nutritional and water, can be characterized as representing different pro-
characteristics, such as many soils or seawater, which neither portions of the two phases, solid surface and water. The contin-
suppress nor specifically support microbial growth, usually carry uum of habitats ranges from highly arid desert soil with no or
low numbers of microorganisms but a high diversity of species. firmly bound pore water to offshore pelagic seawater containing
In contrast, ecosystems of strong environmental characteristics, a minimum of suspended particulate matter. Within the range of
such as acid mine waters, salt brines, and hot springs, commonly suitable physicochemical conditions, the abundance of microor-
contain high cell numbers of very few species. ganisms in an ecosystem is determined by the availability of the
Experimental enrichment procedures bring about the required energy and carbon sources and essential nutrients. All
predominance of certain species by controlling the supply of the more or less specific environments—for example, the surface
specific nutrients or the use of certain physicochemical of leaves or skin, intestinal tracts, and symbiotically or parasiti-
conditions. If the growth conditions of a particular microorgan- cally invaded tissues—conform to this general description.
ism are known and reproducible, enrichment and isolation The concept of microenvironments eliminates the sharp
usually pose no problem. But if the particular requirements for dividing lines between aquatic, terrestrial, and even medical
growth of an organism are unknown, isolation procedures may be microbiology. Indeed, in ecological research, the distinction
difficult to discover (Pfennig 1961; Schlegel and Pfennig 1961). between these academic disciplines is now more and
For that reason, a number of organisms long known from more deemphasized by encompassing them under the label of
microscopical observations, such a Thiovulum or Achromatium, environmental or biogeochemical microbiology. This chapter
have not yet been isolated in pure culture. Furthermore, organ- does not try to cover all the habitats of all organisms treated in
isms that have hitherto unknown growth characteristics and that this handbook; the individual habitats and their characteristics are
are too small and inconspicuous for easy microscopical recog- considered chapter by chapter for single species or physiological
nition have often escaped detection. An excellent example is groups of prokaryotes. This chapter reviews the versatility
Desulfuromonas acetoxidans (Pfennig and Biebl 1976). of prokaryotic metabolism in relation to a few principles
In characterizing an ecosystem microbiologically, it is that determine the distribution of prokaryotes in nature.
important to distinguish between (1) organisms introduced The principal methods for the enrichment and isolation
incidentally by air, soil runoff, etc., physiologically just making of the major metabolic types of microorganisms were discov-
the best of it, and (2) organisms typically adapted to the partic- ered within a relatively short time. Details of the techniques
ular habitat and not occurring in any other except in the form of developed by Winogradsky and Beijerinck are dispersed
survival stages. An example of the former is Escherichia coli, as through the journals. Their collected papers (Beijerinck
frequently found in polluted waters. An example of the latter is 1921–1940; Winogradsky 1949) are treasure troves for microbi-
the above-mentioned Thiovulum sp., whose need for dissolved ologists; only one contemporary compilation exists (Stockhau-
oxygen and hydrogen sulfide at the same time requires a high sen 1907). Since the enrichment principles and methods were
motility combined with chemotactic orientation in an aquatic the subject of a symposium (Schlegel 1965), several reviews
oxic/anoxic interface. have appeared (Aaronson 1970; Schlegel and Jannasch 1967;
Although their morphological differentiation is limited, Veldkamp 1970).
prokaryotes have evolved a number of structural and chemical The aim of this handbook is to encourage biologists to
mechanisms that enable them to inhabit various extreme continue and intensify the search for bacteria in their natural
environments. The presence of a specified pigment, for instance, environments, define habitats and ecological niches, and
protects a cell against detrimental radiation or may provide understand the flux of matter and energy through the biosphere.
for the absorption of light energy at specific wavelengths One may remember that in many soil and water samples, there
encountered in deeper water. Some filamentous cyanobacteria are more kinds of bacteria present than we can cultivate. Fur-
show a certain degree of cell differentiation, a feature that permits thermore, much data on the flux of carbon and of trace gases
the fixation of elemental nitrogen concomitantly with oxygenic through ecosystems cannot yet be accounted for by the bacteria
photosynthesis in oligotrophic environments. More importantly, cultivated so far. The gaps need to be filled by laboratory and
however, the metabolic versatility of the prokaryotes, which field studies.
Prokaryotes and Their Habitats 3 41
Ecological Terminology organism. For each microorganism, at least one habitat can be
defined where it lives and grows and from which it may be
Ecosystems recovered and isolated. Within a distinct ecosystem, a microor-
ganism usually has only one kind of habitat. However, a micro-
‘‘The ultimate aim of ecology is to understand the relationships organism may have two or more habitats, each in a different
of all organisms to their environment’’ (Hungate 1962). ecosystem. For example, many rhizobia are able to grow in the
The principles of ecology were developed independently for soil as well as intracellularly in root nodules; opportunist path-
the sciences dealing with plants, animals, and microorganisms. ogenic bacteria such as Pseudomonas aeruginosa, P. pseudomallei,
Identical terms are used to designate ecological units and and Erwinia live in aquatic habitats as well as in mammalian
parameters when dealing with macro- and microorganisms, hosts; and luminescent bacteria are met as free-living forms as
but the implications of these terms are slightly different. well as living symbiotically in light organs of fish and inverte-
The differences sometimes lead to considerable confusion brates. Some bacteria are associated with both plant and animal
(Whittaker et al. 1973). Furthermore, microorganisms, hosts. A particular species of cellulolytic bacteria or of
especially bacteria, exhibit properties not encountered in higher methanogenic bacteria may have its habitat in a lake bottom
forms of life. The ecology of microorganisms has, therefore, sediment, in the rumen, and in the sewage-sludge fermenter. In
been studied separately, and emphasis has been given to those other words, the habitat designates the street and house number of
microorganisms and their habitats that live under conditions a particular organism; some organisms may have several addresses.
not tolerated by the majority of the higher forms of life When a species or a particular group of microorganisms is
(Alexander 1971, 1977; Hungate 1962; Kushner 1978). The discussed, the word habitat is often used to denote an ecosystem,
recognition of the principles of microbial ecology is not yet that is, the entirety of the biotic and abiotic factors to which
complete and requires studies in many areas of experimental the particular microorganism is exposed in its dwelling place.
science (Atlas and Bartha 1987; Belkin et al. 1986a, b, 1987; This meaning of the term is in general usage in bacterial
Stetter 1986, 1989; Megusar and Gantar 1986). Therefore, the ecology. Since the operational area of many bacteria is small,
usage of terms in microbial ecology is not uniform and partially the term ‘‘microhabitat’’ is frequently used synonymously with
deviates from that used for the ecology of macroorganisms. ‘‘microecosystem’’ or with ‘‘microenvironment.’’
A short survey of the present usage of some terms found in
this chapter is presented below.
The ecology of microorganisms is concerned with the Inhabitants of an Ecosystem
relationships between different species of microorganisms and
between microorganisms and the environment. The basic unit of Each ecosystem harbors a variety of diverse microorganisms.
ecology is the community or biocoenosis. The community has Two principal categories of microorganisms are distinguished:
to be considered in relation to the physical and chemical (1) autochthonous or indigenous (resident) microorganisms
characteristics of the site. Both the biotic components—the and (2) zymogenous, allochthonous, or nonindigenous
community—and the abiotic, physicochemical components (transient) microorganisms. This concept of two large groups
make up the ecosystem. The abiotic component of the ecosystem of microorganisms, which are different with respect to
is frequently referred to as the environment, although when their function, was first suggested by Winogradsky (1925). This
speaking about a particular member of the community, for concept, originally developed in soil microbiology (Winogradsky
example, a species or its population, the term environment is 1925, 1926, 1947, 1949), has been adopted in general ecology
often used to designate both the physicochemical and the (Alexander 1971; Savage 1977a). Autochthonous microorgan-
biotic components of the ecosystem. In this case, the term isms in the strict sense are those that are native to a distinct soil
environment is used synonymously with ‘‘habitat.’’ (e.g., humic soil) and are always present. The population of
Ecosystems differ from each other with respect to their autochthonous bacteria does not depend on exterior organic
extension. Ecosystems may be as large as a pond, a lake, a river, matter or on periodic increases of nutrients. Instead it reflects
or a fir forest, or they may be as small as the rumen, the intestinal a more or less constant flux of nutrients. Indigenous microor-
tract of an earthworm, or the rhizosphere of a plant. Some ganisms are known for many ecosystems, such as humus soil, the
ecosystems, such as a cultivated field or the human skin, are intestinal tract of animals, the rumen, the skin, and plant leaves.
characterized by weak (normal) environmental factors. The terms zymogenous, allochthonous, or nonindigenous
Other ecosystems, such as hot springs or solar salt ponds, are refer to microorganisms that are dependent on occasional
characterized by strong environmental factors which cause a increases of nutrient concentrations or the occurrence of
strict selection among potential inhabitants. unusual substrates. As aliens and transients, they may
persist in the absence of their required substrate or substrate
concentration in dormant or starvation-resistant cell stages.
Habitats The ecosystem may be occupied by metabolically
highly specialized organisms that are restricted to a distinct
Within an ecosystem, a habitat for each species can be defined. environment. These microorganisms are easily recognized as
The habitat is the location or dwelling place of a particular belonging to the indigenous flora of the ecosystem. Various
obligately parasitic bacteria are specialized to certain animals or Methanobacterium ruminantium (Latham and Wolin 1977), and
plants or even to certain tissues; some bacteria have only been Clostridium thermocellum and Methanobacterium thermoauto-
isolated from hot springs, from sewage, or from other extreme trophicum (Winfrey and Zeikus 1977). Therefore, the organism
environments. At the opposite extreme are the ubiquitous bac- fulfilling a certain function in a particular ecosystem is deter-
teria that are able to grow in distinctly different environments. mined by many physical and chemical factors and by all of the
Examples are soil and water bacteria that occasionally become organisms constituting the community.
opportunistic pathogens in animals (Pseudomonas aeruginosa, P. Studies on bacteria in pure culture attempt to define their
pseudomallei) or that may switch from plants to animals entire ecological potential. The conclusions derived from pure-
(Erwinia). culture studies can be used to describe the potential niche of the
particular bacterium. In nature, the predicted niche may be
narrowed by competition. For example, it is possible to grow a
Ecological Niches pure culture of Chromatium vinosum in the laboratory in a
medium containing malate as the sole electron donor (Thiele
The term ecological niche was originally used in the sense of an 1968; van Niel 1932, 1936). In its natural habitat, Chromatium
area or location, the properties of which enabled an organism to vinosum would not dominate on this medium; members of the
live there. The functional concept of the niche led to some Rhodospirillaceae would outgrow the Chromatiaceae.
confusion. The different meanings were discussed by Whittaker, Finding the explanation for the dominance of a particular
Levin, and Root (1973). Now ‘‘ecologists use the term ecological species in the habitat requires nutritional and kinetic pure-
niche to mean the role that the organisms play in the ecosystem’’ culture studies. For example, in aquatic environments where
(Odum 1977). The ecological niche means the function of hydrogen sulfide is generated in the bottom sediment, bright
a species in a community of species in the ecosystem. Each purple or salmon-colored suspensions of the phototrophic bac-
species or strain of a bacterium has particular nutritional terium Chromatium okenii or Thiospirillum jenense are often
requirements, kinetic properties, biochemical abilities, and seen. Microscopic inspection reveals the presence of several
structural particularities; furthermore, it has distinct degrees of small species of the Chromatiaceae in addition to the large
tolerance to environmental conditions. The ability or inability to species. Both types evidently coexist in their habitat. However,
fulfill a particular function in a distinct ecosystem is dependent when samples of these natural enrichments are used for inocu-
on those properties. lating enrichment cultures in the laboratory, the dominant large
For example, many cellulose-hydrolyzing bacteria are not bacterial species are overgrown by small species of the same
highly specialized. They occupy a broad niche. However, in physiological group, such as C. vinosum (Pfennig 1965;
some ecosystems, they may be restricted to cellulose utilization Schlegel and Pfennig 1961). The dominance of the large species
only, because they cannot compete with other organotrophic in the natural habitat was not understood until Van Gemerden
bacteria in performing other functions such as glucose (1974) studied the competition of a representative of the large
fermentation. In these ecosystems, they occupy a narrow niche. species, C. weissei, and a representative of the small species,
This description agrees with the general observation that C. vinosum, for the same substrate, hydrogen sulfide, in
the actual distribution of a species or population is usually chemostat culture.
more limited than its predicted one. In other words, real The dominance of C. weissei in the natural habitat turned
niches are always smaller than potential niches. Secondary out to depend upon the diurnal rhythm of illumination. The two
factors determine the predominance of one species among species differ from each other with respect to the affinity (the Ks
those which could potentially occupy the niche. In the rumen, of intact cells) for hydrogen sulfide and to the rate of its uptake
for example, among the cellulolytic bacteria, only those that and oxidation to sulfur globules stored intracellularly. The max-
are able to hydrolyze cellulose under anaerobic conditions and imal rate of hydrogen sulfide uptake (on the basis of total cell
to generate energy for growth by fermentation will be suited number) by C. weissei is about two and a half times higher than
to fulfill the function of cellulose hydrolysis. Furthermore, that by C. vinosum. In contrast, C. vinosum has a lower Ks for
cellulose degradation in the rumen requires tolerance to hydrogen sulfide and has a higher maximum growth rate. On the
the particular environmental conditions in the rumen, such as basis of experimental evidence, the dominance or the coexis-
temperatures up to 39–40 C and presence of various fatty tence of C. weissei in the natural habitat can be explained: when
acids, enzymes, ammonia, and gases. Finally, the function of both strains are growing in continuous light, most of the hydro-
the cellulolytic population of the rumen is influenced by gen sulfide will be consumed by C. vinosum due to its high
the activities of other members of the community. In some growth rate and high affinity for hydrogen sulfide. In the dark,
rumen bacteria, the metabolic processes are either impeded hydrogen sulfide accumulates; on illumination, the major
or modified if a product of their fermentation, hydrogen, is amount of the accumulated hydrogen sulfide is oxidized by
not removed by methanogenic bacteria. In this case, interspecies C. weissei, and sulfur is stored intracellularly, allowing at
hydrogen transfer may be involved, as demonstrated for least moderate growth for a few hours. Under intermittent
the S-organism and strain MoH of Methanobacterium illumination, both species can coexist. The example
omelianskii (Reddy et al. 1972a), Ruminococcus albus and Vibrio demonstrates that competition in the natural habitat can be
succinogenes (Iannotti et al. 1973), Ruminococcus flavefaciens and based on kinetic properties. It further emphasizes the necessity
of pure-culture studies for understanding the ecological niches Modes of Energy Conversion
of some microorganisms.
The importance of the substrate affinity, expressed in the Ks All forms of life use energy for maintenance as well as for
constant, can be easily demonstrated by growing mixed bacterial biosynthesis of cell material. Biochemical energy is generated
populations in the chemostat at varying dilution rates. At high by metabolic reactions. Energy sources are organic or inorganic
dilution rates with substrate excess, the fast-growing bacteria are substrates that are taken up from the environment. In the cell,
favored. At low dilution rates, when the substrate limits growth, they are converted via a series of metabolic pathways. These
the organism endowed with high substrate affinity (low Ks) pathways fulfill two essential functions: they provide precursors
will successfully compete. Continuous culture enrichment for the macromolecular cell constituents and they make energy
procedures with natural populations are based on these spe- available for biosynthetic and other energy-requiring processes.
cies-specific substrate affinities (Jannasch 1967). Competition The principle of ‘‘unity in biochemistry’’ (Kluyver and
studies were conducted with psychro- and mesophilic bacteria Donker 1925, 1926) is one of the few lasting dogmas of this
(Harder and Veldkamp 1971), a spirillum, and a rod-shaped century. Although the original concept referred to the basic
bacterium in phosphate-limited medium (Kuenen et al. 1977). energy relationships of organisms, it is now clear that ‘‘unity’’
Such studies have been used to describe the characteristics of expresses the assumption that the biochemistry of all organisms
prokaryotes living in habitats where growth is largely limited by on this planet follows a few basic principles, for example, the
carbon compounds utilized by a large number of competing uniformity of cell constituents, the universality of ATP as
species. The use of the chemostat and the kinetics of microbial the principal carrier of biological energy, the universality of the
growth at characteristic ecological niches have been repeatedly genetic code, and the distribution of the degradative pathways,
reviewed (Jannasch and Mateles 1974; Veldkamp 1976; the respiratory chain, and the basic mechanisms of cellular
Veldkamp and Jannasch 1972). energy conversion. Even the main metabolic pathways are
Understanding the ecological niche of a bacterial species is almost identical in all organisms. Only a few groups of bacteria
necessary to design enrichment culture conditions. Elective have modified patterns of central metabolic routes. The
enrichment requires knowing more about a bacterium than its metabolic pathways arose during evolution. The biochemical
basic features. The knowledge of general growth conditions is apparatus that is typical for aerobic organisms developed when
not sufficient. For elective enrichment, only those properties can oxygen became available.
be considered which allow successful competition within The prokaryotes developed billions of years before the atmo-
a mixed population. The determining feature for occupying sphere became aerobic and before carbohydrates became abun-
the actual narrow niche may be as obvious as in the case of dant products of primary biomass production. They may be
the dinitrogen-fixing Azotobacter, abundantly growing and regarded as relics of the evolution of life. During their early
dominating in a sample of soil to which sucrose has been evolution, the prokaryotes learned to exploit a multitude of
added without providing a source of combined nitrogen. The sources of energy and cell carbon different from the predomi-
determining feature, however, may still be hidden, as in the case nant present-day nutrient sources. The morphological unifor-
of enriching for green sulfur bacteria (Chlorobiaceae), which mity of prokaryotes contrasts astonishingly with their
compete successfully at low light intensities but succumb to versatility in substrate utilization, peripheral metabolic path-
purple sulfur bacteria at high light intensities (Biebl and Pfennig ways, and modes of energy conversion. The most outstanding
1978; Pfennig and Cohen-Bazire 1967). metabolic capacities, which are either restricted to only a few
eukaryotes or completely lacking among eukaryotes, are anaer-
obic growth, use of inorganic electron donors for growth, fixa-
Bacterial Metabolism as an Ecological tion of molecular nitrogen, and the utilization of methane and
Determinant a few polymers.
Many ecosystems, both natural and artificial, are almost exclu-

sively occupied by prokaryotes. Environments characterized by Anaerobic Growth
the absence of eukaryotes are the strictly anaerobic regions of
freshwater lakes and of marine estuary sediments and any The ability to grow indefinitely under anaerobic conditions
organic material fermented or decomposed under the exclusion is almost exclusively confined to prokaryotes (see Chap. 13,
of air (sauerkraut, silage, biodigestion with methane produc- ‘‘The Anaerobic Way of Life’’ in Vol. 2). In the majority of
tion). Prokaryotic cells have also adapted almost exclusively to eukaryotes, anaerobic energy generation is only a transient pro-
certain extreme conditions of temperature, acidity, and salinity. cess occurring during intense activity and exposure to hypoxic
The exclusiveness of bacteria is based on unique metabolic environments (Bennett 1978). There are a few exceptions;
capabilities either absent from or only rudimentarily present among the protozoa, two ciliate groups, holotrichs and
in eukaryotes. The uniqueness of bacteria pertains to their entodiniomorphs, live in the rumen (Hungate 1975). Ent-
modes of energy conversion, to their wide range of growth sub- amoeba, Diplomonas, and Trichomonas lack mitochondria and
strates, and to their tolerance toward extreme environmental grow under strictly anaerobic conditions (Bauchop 1971; Müller
conditions. 1975; Steinbüchel 1986; Williams 1986; Stumm and Zwart 1986;
. Table 3.1
Groups of anaerobic phototrophic bacteria
Growth
Anaerobically in Aerobically in Photosynthetic electron Sulfur
Bacterial group Typical species light dark donors deposited
Nonsulfur purple Rhodospirillum + (+) H2, organic, (S2) (Extracellularly)a
bacteria rubrum
Purple sulfur bacteria Chromatium okenii + S2, S0, S2O32, H2 Intracellularly
Green sulfur bacteria Chlorobium limicola + S2, S0, S2O32, H2 Extracellularly
2
Chloroflexus group Chloroflexus + + Organic (S ) (Extracellularly)
aurantiacus
Heliobacterium Heliobacterium + Organic –
chlorum
a
Parentheses indicate substrates used by only a few strains or species
Hobson 1988). Among the helminths, there are various faculta- hydrogen sulfide, sulfur, or hydrogen (Gromet-Elhanan 1977;
tive anaerobes, such as Ascaris lumbricoides, Trichuris vulpis, Jones 1977). Two major groups are different by their pigmenta-
Trichinella spiralis larvae, various tapeworms, and Schistosoma tion: the purple and the green bacteria. On the basis of their
mansoni (Fairbairn 1970; Hochachka and Mustafa 1972; source of reducing power, the purple and green bacteria are
Hochachka and Somero 1973). The question of whether fungi subdivided into sulfur and nonsulfur bacteria. The purple and
are able to grow anaerobically has now been definitely settled. An green sulfur bacteria can use reduced sulfur compounds and
early statement [‘‘One of the major metabolic differences oxidize them via elemental sulfur to sulfate. The nonsulfur
between fungi and bacteria is that there are no anaerobic moulds purple bacteria require reduced organic compounds or, when
either obligate or facultative’’ (Foster 1949)] has been shown to using hydrogen sulfide, oxidize it to sulfur or sulfate. In contrast
be wrong by experimental evidence for anaerobic growth of to the purple sulfur bacteria, they lack the ability to oxidize
species of Mucor (Bartnicki-Garcia and Nickerson 1962) and elemental sulfur. The green bacteria comprise two groups,
Fusarium (Gunner and Alexander 1964), Aqualinderella the Chlorobium group and the Chloroflexus group. The chlorobia
fermentans (Held 1970; Held et al. 1969), and Neocallimastix use hydrogen sulfide as reductant and comprise strictly anaero-
frontalis (Mountfort and Asher 1983; Orpin and Joblin 1988). bic bacteria; the latter are facultative and versatile (Pfennig 1979;
The latter belongs to the Chytridiomycetes and ferments poly- Trüper 1976). The majority of phototrophic bacteria require
saccharides, cellulose included, with the formation of lactic and light for growth; a few species can equally well obtain energy
acetic acids and H2 and CO2. via aerobic respiration.
The contribution of eukaryotes to anaerobic decomposition Anaerobic respiratory energy conversion is similar to aerobic
appears to be negligibly small. The presence of various protozoa, respiration with respect to the electron donors, which may be
fungi, and lower metazoa in the anaerobic layers of coastal either organic or inorganic compounds. Oxygen serves as
marine sediments, especially within the sulfide system, and the ultimate electron acceptor in aerobic respiratory energy
their disappearance when the sediment becomes oxidized generation, but anaerobic respiration depends on the presence
suggest that the number of eukaryotes able to live under anaer- of inorganic compounds, which, under anaerobic conditions,
obic conditions is large (Fenchel 1969; Fenchel and Riedl 1970; are reduced. The physiological groups listed in > Table 3.2 are
Schroff and Schöttler 1977; Zebe 1977). differentiated with respect to the electron acceptors used and to
The forms of anaerobic metabolism are briefly discussed respective end products of the respiratory process. Each of these
here because of the uniqueness of anaerobic metabolism in physiological groups comprises strains and species belonging
prokaryotes and its function in whole ecosystems. The energy to various taxonomic groups.
for bacterial growth under anaerobic conditions can be Fermentative energy generation depends on organic com-
provided by three fundamental types of anaerobic energy pounds serving as electron donors and as electron acceptors.
generation: (1) anoxygenic photosynthesis, (2) anaerobic These compounds are usually two different metabolites derived
respiratory energy generation, and (3) fermentative energy from sugar by cleavage or, in a few cases, derived from two
generation. different compounds. Fermentation is accompanied by the pro-
Anoxygenic photosynthesis is the light-driven process that duction of more or less reduced compounds, such as alcohols,
anaerobic phototrophic bacteria use to generate energy organic acids, ammonia and hydrogen and of carbon dioxide as
(> Table 3.1). Unlike green plants, these bacteria are not able the oxidized product.
to use water as the ultimate reductant; consequently, they do not The fermentative bacteria are usually separated into groups
evolve oxygen, and they require the presence of other reduced on the basis of one or several fermentation products which
compounds such as organic acids, alcohols, carbohydrates, reflect their metabolic pathways (> Table 3.3).
. Table 3.2
Physiological groups of bacteria able to grow under anaerobic conditions using external electron acceptors for electron transport
(‘‘aerobic respiration’’)
Bacterial group Typical species Metabolic process Electron acceptor Reduction products(s)

Denitrifiers Pseudomonas denitrificans Nitrate respiration NO3 N2, N2O, NO2
Sulfate reducers Desulfovibrio vulgaris Sulfate respiration SO4 2
S2
Sulfur reducers Desulfuromonas acetoxidans Sulfur respiration S0 S2
Methanogenic bacteria Methanobacterium thermoautotrophicum Carbonate respiration CO2 CH4
Acetogenic bacteria Acetobacterium woodii Carbonate respiration CO2 CH3–COOH
Succinogenic bacteria Wolinella succinogenes Fumarate respiration Fumarate Succinate
Iron reducers Pseudomonas GS-15 Iron respiration Fe3+ Fe2+
Growth with Inorganic Electron Donors There are a few compounds that can be utilized exclusively by
prokaryotes. Among the low-molecular-weight compounds,
The ability to use inorganic compounds as electron donors methane is oxidized only by highly specialized bacteria, the
for growth, called lithotrophy, is exclusively restricted to methylotrophs. Some of these bacteria are obligate
prokaryotes (see Chap. 14, ‘‘The Chemolithotrophic Prokary- methylotrophs, being able to use methane, methanol, and
otes’’ in Vol. 2; and Chap. 4, ‘‘The H2-Metabolizing Prokary- dimethyl ether as substrates.
otes’’; Chap. 15, ‘‘The Colorless Sulfur Bacteria’’ and Chap. 3, No macromolecular compounds are used only by prokary-
‘‘Oxidation of Inorganic Nitrogen Compounds as an Energy otes. Cellulose, hemicelluloses, pectins, xylanes, galactan, man-
Source’’ in Vol. 3). The electrons are used for electron transport nans, chitin, and others are subject to degradation by bacteria
phosphorylation, either aerobically with oxygen or anaerobically and fungi; some polymers are hydrolyzed by protozoa and a few
with inorganic compounds (nitrate, sulfate, carbonate) as the metazoa. Under anaerobic conditions, however, the degradation
ultimate electron acceptors. The electrons also serve to reduce of polysaccharides is almost exclusively confined to prokaryotes.
carbon dioxide, which is the common carbon source for
a physiological group of bacteria called chemolithoautotrophs.
These bacteria are usually separated into groups on the basis Extremes of Environmental Conditions
of their electron donors (> Table 3.4). In memory of S.N. Allowing Bacterial Growth and Survival
Winogradsky, who developed the concept of chemolithotrophy
and autotrophy in 1887–1891, a symposium on lithoautotrophy Room or body temperature, the oxygen partial pressure of the
was held in 1987 (Schlegel and Bowien 1989). atmosphere, neutrality of pH, and abundant nutrients
supporting luxurious growth of the kinds of organisms studied
in the laboratory are generally considered to be ‘‘normal’’
Fixation of Molecular Nitrogen conditions for growth of microorganisms. Any conditions
substantially deviating therefrom are regarded as being extreme.
Only prokaryotes are able to fix molecular nitrogen The ecologically minded biologist will define those conditions
(dinitrogen = N2) and to grow in the absence of a source of under which the greatest species diversity develops as normal.
combined nitrogen. Dinitrogen fixation requires the presence of Environmental conditions different from the norm have an
a special enzyme system, nitrogenase, and particular environmental elective effect on organisms; as a rule, the species diversity
conditions. The ability to fix dinitrogen is distributed among many decreases with the increase of environmental adversity. The
species of oxygenic and anoxygenic phototrophic bacteria, chemo- greater the severity of an adverse environmental factor,
autotrophic and chemoheterotrophic bacteria, and both aerobic the smaller is the number of species of actively growing
and anaerobic bacteria—in short, in all major physiological groups. microorganisms. In addition, among the organisms adapted to
There is almost no correlation with the taxonomic unit. A recent extreme conditions, the abundance of physiological types
symposium considered all aspects of research in this area (Bothe is restricted. However, this trend of limitation of species and
et al. 1988). number of individuals may not be as pronounced as is presently
assumed. It may just reflect reluctance to examine extreme
environments and to isolate new organisms, many of which
Range of Organic Substrates for Growth cannot easily be grown in laboratory cultures.
In general, the microorganisms that tolerate and grow under
The bacteria as a whole are considered to be omnipotent with the most extreme conditions are obligately adapted to their
respect to their substrate spectrum. All natural (biosynthetic) particular environment and fail to grow at lower intensities of
compounds are subject to degradation by microorganisms. the same environmental factor. Such an organism has acquired
. Table 3.3
Groups of fermentative bacteria able to grow under anaerobic conditions and their fermentation products
Fermentation characterizing bacterial Fermentation product

groups Typical species Substrate Major Minor
Ethanol fermentation Zymomonas mobilis Glucose Ethanol CO2
Lactate fermentation:
Homofermentative Lactobacillus casei Glucose Lactate
Heterofermentative Leuconostoc mesenteroides Glucose Lactate Ethanol, CO2
Heterofermentative Bifidobacterium bifidum Glucose Acetate Lactate
Butyrate fermentation Clostridium butyricum Glucose Butyrate Acetate + H2 + CO2
Clostridium acetobutylicum Glucose Butyrate, butanol Acetone, 2-propanol
Clostridium kluyveri Ethanol + Butyrate Caproate, H2
acetate
Homoacetate fermentation Clostridium aceticum Fructose Acetate
Propionate and succinate fermentation Propionibacterium Sugars, lactate Propionate Succinate
pentosaceum
Veillonella alcalescens Lactate Propionate Acetate, H2, CO2
Bacteroides ruminicola Sugars Propionate
Mixed acid and butanediol fermentation Escherichia coli Glucose Lactate, ethanol, Formate, H2 + CO2,
acetate succinate
Enterobacter aerogenes Glucose 2,3-Butanediol, Formate, H2 + CO2
ethanol
Nitrogenous compounds fermentation Clostridium Glutamate Butyrate Acetate, CO2, NH3
tetanomorphum
Clostridium sticklandii Lysine Butyrate Acetate, NH3
Clostridium oroticum Orotate Acetate CO2, NH3
. Table 3.4
Groups of bacteria able to use inorganic electron donors for growth (‘‘chemolithoautotrophs’’)
Electron Electron Carbon

Bacterial group Typical species Metabolic process donor acceptor source Product
Hydrogen-oxidizing bacteria Alcaligenes eutrophus H2 oxidation H2 O2 CO2 H2 O
Carbon monoxide-oxidizing bacteria Pseudomonas CO oxidation CO O2 CO2 CO2
carboxydovorans
Ammonium-oxidizing bacteria Nitrosomonas europaea Ammonium oxidation NH4+ O2 CO2 NO2
Nitrite-oxidizing bacteria Nitrobacter winogradskyi Nitrite oxidation NO2 O2 CO2 NO3
Sulfur-oxidizing bacteria Thiobacillus thiooxidans Sulfur oxidation S, S2O32 O2 CO2 SO42
Iron-oxidizing bacteria Thiobacillus ferrooxidans Iron oxidation Fe 2+
O2 CO2 Fe3+
Methanogenic bacteria Methanobacterium Methanogenesis H2 CO2 CO2 CH4
thermoautotrophicum
Acetogenic bacteria Acetobacterium woodii Acetogenesis H2 CO2 CO2 CH3‐COOH
the ability to grow in one extreme environment at the expense of numbers of species and of high specialization in habitats of
its ability to grow in others. Less rigid or extreme environmental extreme conditions, such as low and high temperature, high
conditions are tolerated by a greater number of organisms; some salt concentrations, low moisture, low pH, and low-nutrient
of these are obligately bound to these conditions, and others concentrations. Much attention has been given during recent
grow there facultatively. There are several examples of low years both to microorganisms that either tolerate or are
dependent on extreme conditions and to their habitats. Life in High Temperatures

extreme environments was the subject of several symposia, and
emphasis was paid to the microorganisms, to the regulatory or Growth at temperatures higher than about 60 C is restricted to
molecular mechanisms that make life under adverse conditions the prokaryotes. Habitats with such temperatures include piles
possible, and to the possible existence of life on other planets of self-heating hay, compost, or other organic materials
(Alexander 1971, 1976; Brock 1969; Ellwood et al. 1980; (Hussain 1973), circulating hot or cooling water in industrial
Heinen 1974; Hochachka and Somero 1973; Kushner 1971, plants, and hot springs and other geothermal sources.
1978; Shilo 1979; Brock, 1986a, b, Stetter 1986, 1989; Larsen The ecology of hot springs has been especially well studied
1986; Tindall and Trüper 1986). and reviewed (Brock 1967, 1970, 1978, 1986; Castenholz 1969,
Thanks to the amount of attention that the ‘‘extremophiles’’ 1979). The runoff of a hot spring is an ideal location to study the
and their habitats have received during recent years, there is no upper temperatures for continual growth of organisms. As in
need to add another exhaustive review. However, it is likely that most extreme habitats, the species diversity decreases
more as yet unidentified microorganisms representing various with increasing severity of the environmental factor. When the
metabolic groups are present in extreme environments. They temperature gradient rises from 50 C to 70 C, thermophilic
can only be discovered by experimental approaches based on bacteria such as the yellow-pigmented Thermus aquaticus
new ideas on possibly existing organisms. The following short (Brock and Freeze 1969), which has an optimum temperature
survey is meant to encourage relevant research. for growth at 70 C and a maximum at 79 C, can be found. In
many runoff channels, thick mats consisting of the cyanobacte-
Low Temperatures rium Synechococcus lividus and the phototrophic, gliding,
filamentous bacterium Chloroflexus aurantiacus are present
The temperature of the natural environments of water and soil (Pierson and Castenholz 1974). In these mats, several layers of
bacteria is distinctly lower than their optimum growth temper- Chloroflexus are covered by a thin surface layer containing
ature. The average soil temperature in the temperature climate a mixture of Synechococcus and Chloroflexus (Bauld and Brock
zone is 12 C, and 90 % of the ocean water is 5 C or colder. 1973; Madigan and Brock 1977).
Many mesophilic bacteria do not find their optimum growth Several strains of the facultative autotroph Sulfolobus
temperature in their natural habitats; they are, however, able to acidocaldarius, with a growth pH range of 0.9–5.8 and temper-
tolerate the seasonal fluctuations of temperature, which in the ature optimum of 70–75 C, and of an extreme thermophile
summer can easily span 30 C or more. This tolerance is lacking growing at 85 C have been isolated from acid, hot, aqueous, and
in psychrophilic bacteria. As defined by Morita (1975), soil habitats (Brock et al. 1971, 1972; Fliermans and Brock 1972).
‘‘the psychrophiles are those organisms having an optimum Thermophilic Thiobacillus-type bacteria growing at 60 C or
growth temperature of 15 C or lower, a maximal temperature 75 C were found in Icelandic thermal areas (Le Roux et al.
for growth at about 20 C and a minimal temperature for 1977). The habitats par excellence of these bacteria are volcanic
growth at 0 C or lower.’’ Many bacteria that have been isolated hot springs where magmatic hydrogen sulfide is oxidized to
from Arctic and Antarctic Ocean waters and sediments have elemental sulfur and sulfuric acid. Thermophilic iron- and sul-
optimal growth temperatures of about 10 C but do not survive fur-oxidizing bacteria may be involved in metal leaching from
exposure to 20 C. True psychrophiles will, therefore, be low-grade ore (Brierley 1977; Brierley and Lockwood 1977;
found only in habitats that never become warmer than 20 C. Brierley 1978; Golovacheva and Karavaiko 1978; Levi and
At temperatures below 10 C, psychrophiles do have selective Linkletter 1989; Hughes and Poole 1989).
advantage as demonstrated in continuous culture experiments Yellow-pigmented Thermus strains have also been isolated
(Harder and Veldkamp 1968, 1971). One of the ecological from Icelandic hot springs (Pask-Hughes and Williams 1977).
niches of psychrophiles in the ocean is chitin degradation. Nonpigmented thermophilic bacteria, otherwise similar to
The vast majority of psychrophilic bacteria are members of Thermus aquaticus, were found in laundry heaters (Brock and
Pseudomonas, Flavobacterium, Achromobacter, and Alcaligenes Boylen 1973), hot tap water (Pask-Hughes and Williams 1975),
(Rose 1968). and a stream receiving hot-water effluents (Ramaley and Hixson
Food preservation added another habitat for psychrophilic 1970). The range between 50 C and 70 C is occupied by
bacteria (Schmidt-Lorenz 1967). However, the majority of bac- a multitude of bacteria, among them members of the genera
teria that grow on food near the freezing point are apparently Bacillus, Thermoactinomyces (Cross 1968), Methanobacterium
facultative psychrophiles or psychrotrophs that have tempera- (Zeikus and Wolfe 1972), Methylococcus, and others. Several
ture maxima for growth above 20 C but are able to grow at low thermophilic bacteria find excellent growth conditions in
temperatures also. With respect to growth at temperatures down canned foods (Gillespy and Thorpe 1968) and sugar (Scarr
to 18 C, the capacities of prokaryotes do not seem to exceed 1968).
those of the eukaryotes. Our knowledge of microbial life at high temperatures has
Microbial life at low temperatures has been comprehensively been greatly extended by studies of continental and marine hot
reviewed and discussed on the basis of ecological aspects (Baross springs of volcanic origin. A considerable number of organisms
and Morita 1978) and molecular aspects (Inniss 1975; Inniss and have been isolated that cover a range for optimal growth at
Ingraham 1978). 85–105 C and are generally termed extreme thermophiles
or hyperthermophiles (Stetter 1986; Stetter et al. 1990). These Although many thermophilic water and soil bacteria were
organisms are mainly archaebacteria and appear to be so well reported to be present only in samples of thermal habitats,
adapted to these high temperatures that they do not grow below experience indicates that they are much more widespread. The
60 C, and some of them do not grow below 80 C. Acidic isolation of thermophilic, hydrogen-oxidizing bacteria from hot
terrestrial solfataric environments have primarily yielded acido- springs and the failure to isolate them from other places indi-
philic, hyperthermophilic autotrophs of the order Sulfolobales cated a very narrow distribution of these highly specialized
that aerobically oxidize H2S and S , with the exception of bacteria (Goto et al. 1977). Surprisingly, attempts to isolate
Acidianus, which is able to grow anaerobically on H2 and S thermophilic, hydrogen-oxidizing bacteria from cold lake sedi-
forming H2S (Huber et al. 1987a). The same is true for some ments (Aragno 1978; Schenk and Aragno 1979) and from the
neutrophilic isolates that exist in neutral and anaerobic zones of oxidation ponds of a sugar factory (Schlegel unpublished) were
terrestrial solfataras and are members of the genera Thermoproteus successful. The search for thermophiles among other highly
and Pyrobaculum (Zillig et al. 1981, 1982; Stetter and Zillig specialized bacteria resulted in equally encouraging findings,
1985). A facultatively autotrophic isolate (Pyrobaculum for example, thermophilic, nitrifying bacteria (Golovacheva
islandicum) grows by sulfur respiration of organic matter 1976), iron-oxidizing bacteria (Brierley 1978), sulfur-oxidizing
(Huber et al. 1987a). bacilli (Golovacheva and Karavaiko 1978), and carbon monox-
Another habitat that has recently yielded a number of ide-oxidizing bacteria (Meyer unpublished). These recent find-
extremely thermophilic bacterial isolates is the deep-sea hydro- ings indicate that the ability to grow at temperatures higher than
thermal hot vents found at depths from 1,800 to 3,700 m. The 50 C is not confined to only a few metabolic types.
high pressures make possible the presence of liquid water at Research in the ecology of thermophilic microorganisms has
temperatures far beyond the 1-atm boiling point of water, that recently yielded spectacular results; they are summarized by
is, at a depth of 2,500 m, extruding vent water will not boil below Brock (1978), Tansey and Brock (1978), Brock (1986), and
approximately 460 C. This habitat provides an ideal situation to Castenholz (1979).
look for the upper temperature limit of prokaryotic life. Early
indications that growth might occur at 250 C (Baross and
Deming 1983) have not yet been confirmed, although active Water Stress
growth at 110 C has been found in a culture of a new group
of methanogenic archaebacteria (Methanopyrus, Huber et al. Without water, life is not possible. The normal environment of
1989). While this organism grows with its optimum doubling microorganisms is an aqueous solution. The water body may
time of about 1 h just below 100 C, isolates of the genus differ in size, and the water may be free or adsorbed to external
Pyrodictium (Stetter et al. 1983) grow optimally at 105 C. or internal surfaces of materials. The availability of water to
Pyrodictium, as well as a large number of physiologically diverse microorganisms is a function not only of the water content of
hyperthermophilic archaebacteria, was isolated from a shallow a material but also of solution and adsorptive factors. In order to
marine hot spring (Stetter 1986) and an erupting volcano compare solutions and solid materials with respect to available
(Huber et al. 1990). water, the parameters ‘‘water activity’’ (aw, ranging from 0 to 1.0)
Using 35SO42, hyperthermophilic sulfate reduction has and ‘‘relative humidity’’ (expressed as a percentage) are used.
been demonstrated in sediment cores freshly collected These parameters express the ratio of water in the vapor phase to
from a hydrothermal vent field (Jørgensen et al. 1990), and the the amount of water the air would contain when vapor-saturated
organisms responsible (Archaeoglobus profundus, Burggraf et al. at a given temperature. Another parameter, the ‘‘water poten-
1990) could be isolated at the same time. This particular tial,’’ is based on the free energy of molecules in water and makes
vent field, located in the Gulf of California at a depth of differentiation of the matric and solute components of the
2,000 m, is overlayed by 3–400 m of organic-rich sediment system easy (Brown 1978; Griffin and Luard 1979; Smith 1978;
(Jannasch 1989). In the upper 60 cm of these sediments, down- Griffin 1981).
ward temperature gradients of 3–180 C have been measured, Microbial growth is possible in the range of water activity
overlayed by ambient bottom water with a temperature of between 0.998 and 0.6 (Duckworth 1975; Mossel 1975; Mossel
2.1 C. In these sediments, methanogenic hyperthermophiles and Ingram 1955). Bacteria are not very successful at extracting
of the genus Methanococcus (Jones et al. 1983, 1989; Zhao et al. water from environments of lowest water activities. Fungi lead
1988) appear to be common in addition to those of the above- the list of organisms arranged with respect to the lowest water
mentioned Methanopyrus. The majority of the newly described activity at which they can grow. The osmotolerant yeast, Sac-
hyperthermophiles were found in both the deep and the shallow charomyces rouxii, and the osmophilic mold, Xeromyces bisporus,
marine hot vents. This includes the genera Pyrococcus, are able to grow down to aw = 0.60; even Aspergillus glaucus can
Pyrodictium, and Staphylothermus and the only eubacterial survive at aw = 0.80. In contrast, the majority of bacteria need
genus of extreme thermophiles, Thermotoga (Fiala et al. 1986; water activities higher than 0.98 (which is the aw of seawater at
Belkin et al. 1986; Huber et al. 1986; Jannasch et al. 1988; 25 C); only a few bacteria grow at 0.95–0.91. The halophilic
Windberger et al. 1989). Some of them depend on the reduction bacteria are exceptions, growing at aw = 0.75. The data are in
of elemental sulfur, while others conduct an unknown type of accordance with the results of measurements on the microbial
fermentation. degradation of plant residues and straw mixed with field soil.
If the soil was kept in equilibrium with pure water, the rate of following differences between freshwater and saline ecosystems
carbon dioxide evolution was maximal. With decreasing humid- cannot be overemphasized. As long as a saline water body or
ity, the activity of bacteria decreased and reached zero at 92 % a water-soaked sediment is aerobic, the water potential may be
relative humidity. the dominating factor among the selective environmental con-
One has to distinguish between two effects of humidity, that ditions for the growth of bacteria. The sequence of dominance of
on the activity and that on the viability of microorganisms. Some selective factors changes as soon as the location becomes
bacterial species are more resistant to drying than others. After anaerobic. Many saline waters, either marine or salt-polluted
some weeks of air-drying in the laboratory, the remaining viable inland waters, contain sulfate; in the absence of oxygen, sulfate is
population consists of bacteria that are able to form endospores, the preferred terminal electron acceptor in anoxic environments
cysts, and other resting stages resistant to desiccation (Boylen and gives rise to the generation of hydrogen sulfide. The con-
1973; Robinson et al. 1965). However, under certain conditions, centration of hydrogen sulfide may then exert a selective pres-
even vegetative cells survive (Clark 1967). Survival of vegetative sure on the bacterial flora in the location and govern the
cells depends on the type of soil, the velocity of the drying composition of the anaerobic food chain. In many freshwater
process, and the water activity. Several reports indicate that lakes, the sulfate concentration is just high enough to provide
processes like nitrification, sulfur oxidation, and nonsymbiotic sufficient hydrogen sulfide for securing the redox potential nec-
nitrogen fixation proceeded after rewetting without any soil essary for methanogens to grow, but in marine ecosystems, such
reinoculation. Apparently, the endospores of some bacilli can as sublittoral and estuarine sediments, the concentration of
survive for hundreds of years. On the basis of viable counts made sulfate and, in consequence, that of hydrogen sulfide exceed
on soil granules from the roots of plants gathered and dried in the threshold where methanogenesis is possible (Cappenberg
1640 (Herbarium, Royal Botanical Gardens, Kew), ‘‘one can 1974a, b; Winfrey and Zeikus 1977).
estimate that a ton of dry soil would still contain a few viable
spores even after 1,000 years’’ (Sneath 1962).
Environments of Extreme pH Values
Salinity Hydrogen and hydroxyl ions are the most mobile of all ions. The
concentration of these ions affects the growth of microorgan-
High salt concentrations represent a special case of low water isms either directly or indirectly via its influence on the ionic
activity. Seawater (aw = 0.98) is not tolerated by the majority of state and the availability of many inorganic ions and metabolites
bacteria living in soil and in freshwater. Ecosystems containing to the cells. The majority of bacteria prefer neutral or slightly
salt (sodium chloride) at saturating concentrations are inhabited alkaline conditions.
by only a few organisms. Many bacteria (Briston et al. 1974) as
well as the flagellated alga Dunaliella viridis and the brine shrimp
Artemia salina can tolerate such high salt concentrations, but Acid Environments
only members of the prokaryotes find their optimum growth
conditions there. The extreme halophiles have their best growth The hydrogen ion concentrations of the ocean and the major
at 20–30 % salt, moderate halophiles at 5–20 %, and slight part of the land vary only within a narrow range, allowing
halophiles at 2–5 % (Dundas 1977; Larsen 1967, 1971, 1973, growth of both fungi and bacteria. Only volcanic lakes and
1986; Tindall and Trüper 1986). soils of recently claimed land have drastically reduced pH values.
Extremely halophilic bacteria, such as Halobacterium An extreme aquatic habitat of low pH is represented by the
cutirubrum, H. salinarum, and Halococcus morrhuae, are distrib- drainage of coal mines and coal mine refuse piles; the acidity is
uted in evaporation ponds for the production of solar salt and due to the oxidation of pyritic minerals, reduced iron, and sulfur
occasionally on salted fish and hides. They are easily recognized compounds associated with coal (Colmer et al. 1950; Leathen
by their red color caused by carotenoids. Moderately halophilic et al. 1953). Sulfuric acid produced by thiobacilli causes a
bacteria live in similar habitats, namely, salt brines and mud, and decrease of the pH down to 2.0 or even 0.7. Iron- and sulfur-
are often found in curing brines for meat, fish, and vegetables. oxidizing bacteria are present wherever mine water enters a
About a dozen different bacteria are reported to find their stream. Thiobacillus ferrooxidans, involved in metal-leaching
optimum growth conditions in the range of 5–20 % sodium processes (Tuovinen and Kelly 1972), can tolerate even pH 1.0,
chloride. The majority of the bacteria inhabiting the sea and as well as ions of copper, cobalt, zinc, nickel, and iron
marine mud require 2–5 % sodium chloride; some of these slight up to extremely high concentrations (Torma 1977). Organic
halophiles fail to grow at lower salt concentrations. acids are excreted in laboratory cultures (Schnaitman and
The general aspects of halophilism and life in high salt Lundgren 1965); their ecological significance is unknown.
concentrations have been recently discussed by Kushner In acidic water, most Gram-positive, aerobic and anaerobic, het-
(1978); Lanyi (1979); Brock (1979); Bayley and Morton (1978, erotrophic bacteria die quickly, while yeasts and molds
1979); and Csonka (1989). predominate (Marchlewitz and Schwartz 1961). Among the het-
From the ecological point of view, salinity plays erotrophic bacteria, only Bacillus, Pseudomonas, Achromobacter
a more important role than the water potential suggests. The (Tuttle et al. 1968), Flavobacterium acidurans (Millar 1973),
and other slime-forming bacteria (Dugan et al. 1970) were Schweinfurth and Lewin in 1898, these shallow salt pans harbor
encountered. Waters of acid hot springs are further habitats a rich population of halophilic and phototrophic microorgan-
of acidophilic bacteria, such as thermophilic bacteria related isms. Sodium is the major cation of the saturated brine (98 %);
to Sulfolobus acidocaldarius (Brierley and Brierley 1973; the major anions are chloride, sulfate, and carbonate/bicarbon-
Brock et al. 1971; De Rosaet et al. 1975) and, at lower ate (56, 26, and 17 %, respectively; Jannasch 1957). Species of the
temperatures, of mesophilic thiobacilli. genus Ectothiorhodospira and other phototrophic bacteria have
Many lakes, bogs, pine forests, and tea soils are slightly been isolated from these and other alkaline waters (Grant et al.
acidic, with pH values between 3 and 5.5. Bacterial life in these 1979; Imhoff and Trüper 1977). Alkaline lakes such as those in
locations is scarce, and, especially under anoxic conditions, Ethiopia and Anatolia, with ambient values of pH 10–11, are
degradation of plant polymers is slow. Nitrification proceeds also worth study.
slowly in acidic soils, and the question of whether nitrification Life at extreme pH values has been discussed by Langworthy
in these areas is partly due to autotrophic nitrifying bacteria (1978), Horikoshi and Akiba (1982), and Krulwich and Guffanti
(Bhuiya and Walker 1977) and to heterotrophs is still open (1983).
(Focht and Verstraete 1977). Acid bogs apparently offer
the marginal conditions under which some rarely encountered
bacteria, such as Planctomyces, Bactoderma, Caulobacter, and Oxygen
Microcyclus (Hirsch and Pankratz 1970), as well as other stalked
and prosthecate bacteria, can thrive (Henrici and Johnson 1935; When considering oxygen as an environmental factor, at first
Schmidt 1971). glance, the anoxic environment appears to be the extreme one.
However, if the peculiarities of generating biochemical energy
under anaerobic conditions are not considered, the anoxic envi-
Alkaline Environments ronment is not difficult for life. Oxygen causes troubles for living
cells. It plays a dual role: it acts as an effective electron acceptor,
The environment where human or animal urine undergoes making energy conversion with high efficiency possible;
a urea fermentation cannot be ignored because of the penetrat- however, oxygen can be considered toxic to life processes that
ing odor of ammonia. The causative agents of this noticeable depend on slow and thoroughly controlled oxidations. The
process were first studied by Pasteur in 1862 and designated toxic effect is intensified in light. Oxygen is reduced by
Torule ammoniacale. The ureolytic bacterium, Bacillus pasteurii, univalent or single electron steps, and by the uptake of
is an alkalophile that requires pH values of 8.5 or higher and one electron, the superoxide anion radical (or simply the
high ammonia concentrations (Wiley and Stokes 1963). superoxide) is formed, which by further reactions gives rise to
Sporosarcina ureae grows at lower alkalinities (pH 8.5) (Mazanec hydrogen peroxide and to the hydroxyl radical; this radical is the
et al. 1965). most reactive and, therefore, the most detrimental product.
The isolation of alkaliphilic bacteria, able to grow in culture From the onset of photosynthetic oxygen production in the
media of pH 10–11, is apparently rather easy. Such variants early atmosphere of this planet, it was the major task of bio-
growing up to pH 11 were isolated after adapting Bacillus chemical evolution to develop an enzyme system for tetravalent
circulans (Chislett and Kushner 1961). Bacillus alcalophilus reduction of oxygen, cytochrome oxidase, which does not
growing at pH 8–10 was isolated from human feces (Vedder release any toxic intermediates, and to develop elaborate detox-
1934) and from dried sewage sludge (Boyer et al. 1973). Several ification systems (Fridovich 1974, 1975, 1976). Although forms
Bacillus strains originated from projects on the production of of life are still susceptible to oxygen toxicity (Gottlieb 1971),
alkali-tolerant enzymes and were isolated from indigo balls and there now exist a variety of defense mechanisms in different
plain soil (Ohta et al. 1975). organisms (Hassan and Fridovich 1979; Morris 1975, 1976,
The szik (salt and alkali) soils with pH values of 8–9 in the 1978, 1979).
Hungarian lowlands have been extensively studied with regard The rise of the oxygen level in the early atmosphere may
to soil ecology and agriculture; however, their bacterial flora was have started not only the evolution of aerobic organisms, but
not characterized in detail (Bokor 1933). Alkali soils in India also that of modestly oxygen-tolerant, anaerobic, fermentative
were reported to contain only few microorganisms, the total bacteria as well. There is a wide spectrum of degrees of oxygen
count decreasing drastically with increasing pH (Rupela and tolerance, starting with the strict anaerobes such as Bacteroides,
Tauro 1973). With these exceptions, alkaline environments Butyrivibrio, Fusobacterium, Megasphaera, Peptococcus,
have not attracted much attention. Investigations of strictly Ruminococcus, Selenomonas, Succinivibrio, Methanobacterium,
alkaline springs in northern California led to the isolation Methanococcus, Methanospirillum, and Methanosarcina and con-
of an anaerobic spore-former and an aerobic, pigmented bacte- tinuing to the moderately oxygen-tolerant clostridia (C. tetani,
rium. The latter grew between pH 8 and 11.4 with the optimum C sporogenes) and the highly oxygen-tolerant clostridia
within the range of pH 9–9.5 (Souza and Deal 1977; Souza et al. (C. perfringens, C. acetobutylicum) and to the majority of the
1974). The lakes of the Wadi el Natrun in Egypt are lactic acid bacteria, which can grow in the presence of air
most interesting aquatic habitats, with pH values ranging from (Loesche 1969; Morris and O’Brien 1971; O’Brien and O’ Morris
9 to 11. Topographically and geochemically described by 1971; Uesugi and Yajima 1978).
Although the obligately anaerobic bacteria have in com- example, at 2%, 20%, 40%, and 100% oxygen and whether
mon a hypersensitivity to oxygen, they are apparently not correlations between oxygen-tolerance thresholds and habitats
restricted to places like mud and intestinal tracts but can are encountered.
also be found in seemingly aerobic locations. The wide distri-
bution of Clostridium can be explained by their possession of
oxygen-insensitive spores. However, non-spore-forming, strictly Low-Nutrient Environments
anaerobic bacteria are similarly widely dispersed, perhaps
because under field conditions, the anaerobes are associated The most commonly cited low-nutrient environments are desert
with oxygen-consuming bacteria that keep the local oxygen soils, oligotrophic lakes, and, most prominently, the oceans.
concentration low. These apparent discrepancies caution one Considering suspended detrital matter, decaying plankton
not to overemphasize the results of pure-culture studies when organisms, and fecal pellets, isolated high-nutrient habitats
extrapolating to the behavior of microorganisms in their natural abound in various parts of the sea. However, if this amount is
environment. divided by the vast volume of the ocean, which amounts to
In some cases, the oxygen concentration is the deciding about 90% of the biosphere, the dissolved organic carbon in
factor in the ability of the organism to develop potential seawater is rarely more than 1 mg/l. Particulate organic carbon
metabolic activities. The oxygen concentration may deter- commonly amounts to 10–20% of the total organic carbon.
mine whether various nitrogen-fixing bacteria can occupy A large fraction of the latter is ‘‘refractory,’’ that is, unavailable
a niche in an environment low in combined nitrogen. In for microbial attack (Barber 1968; Menzel and Ryther 1970).
contrast to members of the Azotobacter group, which fix The description of low-nutrient habitats is, therefore, more
nitrogen in the presence of air (Mulder and Brotonegoro realistically based on the flux of nutrients across the ecosystem
1974), the nitrogen-fixing, hydrogen-oxidizing bacterium rather than on the standing nutrient concentrations
Xanthobacter autotrophicus (Berndt et al. 1976; Wiegel and (Hirsch 1979).
Schlegel 1976; Wiegel et al. 1978) is able to grow with molecular Many prokaryotes will survive in such conditions, as amply
nitrogen as the sole nitrogen source only at oxygen concentra- documented by the fact that poor soils and natural waters have
tions below 2% oxygen (by volume), while in the presence of often been used as a source for the isolation of microorganisms.
combined nitrogen, the bacterium grows under air. Other Low-nutrient habitats may be extreme with respect to stress
nitrogen-fixing bacteria, such as Azospirillum lipoferum (Okon exerted toward some vegetative cells, but these conditions are
et al. 1976), Aquaspirillum fasciculus (Strength et al. 1976), and hardly selective.
Methylosinus species (De De Bont and Mulder 1974), respond to Vegetatively growing cells may be carried into the ocean
oxygen in a similar manner. A number of aerobic marine bacte- from land runoff or be exposed to low-nutrient levels after
ria have been demonstrated to be microaerophilic; they survive separation from relatively nutrient-rich particles. These cells
but do not grow in air-saturated seawater medium (Jannasch may adopt one of a variety of strategies in order to cope with
1977). starvation conditions. Two general strategies have been
Whether there are differences among microorganisms with pinpointed whereby: (1) the survival of the individual cell is
respect to their tolerance to oxygen at higher concentrations achieved by the most efficient use of the limited amount of food
than that of air-saturated water is not known. Many strictly available or (2) the survival of the species is achieved by pro-
aerobic bacteria that form colonies from single cells on petri ducing the maximum amount of progeny in the form of dor-
dishes exposed to air can tolerate gas mixtures up to 40% oxygen mant stages capable of immediate growth and multiplication at
by volume but fail to grow at 50%. Also, 100% oxygen is usually the renewed availability of nutrients.
considered to suppress growth; however, many cyanobacteria In terms of population dynamics studied in continuous
that form blooms are certainly tolerant to high oxygen culture, the ‘‘K-strategists’’ have been described as organisms
concentrations. During photosynthesis in the sun, blooms of adapted to highly efficient growth or uptake at low substrate
cyanobacteria and algal mats near the water surface, as well as concentrations, as indicated by a low substrate saturation con-
the lawn of submerged water plants like Elodea canadensis, stant (K[s]). In contrast, mmax-strategists (mmax = maximum
Chara fragilis, or Ranunculus aquaticus in ditches, shallow growth rate, in nonprokaryotes also called r-strategists) will
ponds, and estuaries, are covered with bubbles of oxygen. outgrow the former in the presence of relatively high-nutrient
This growth indicates that the water body must contain higher concentrations (Koch 1979).
concentrations of oxygen than in equilibrium with air and Early continuous culture studies on the synthesis and
that the epiphytic and other bacteria are exposed to oxygen activity of prokaryotic enzyme systems led to the discovery of
concentrations supersaturated with respect to oxygen. It is not catabolic derepression by Gorini (1960) in the presence of
known whether bacteria exist able to grow at these oxygen extremely low concentrations of the growth-limiting substrate.
concentrations of almost 100 %. Furthermore, it would be Subsequently, the mechanisms of adaptation to low-nutrient
interesting to know whether among the variety of environments have been intensively studied (Tempest and
bacteria a continuum of thresholds of oxygen tolerance exists Neijssel 1976, 1979). These studies resulted in the discovery of
or whether there are certain discontinuities of threshold the high-affinity mechanism for assimilating ammonia into
concentrations for certain species or metabolic types, for glutamate via glutamine synthetase and glutamate synthase
(Tempest et al. 1970, 1973) and of dual mechanisms for the The rod-shaped cells of the genus Azotobacter frequently
assimilation of glycerol: the glycerol kinase route at carbon- turn into spherically shaped cells called cysts. Unlike endospores,
limiting conditions and the glycerol dehydrogenase route in the vegetative rod as a whole is transformed into a cyst. These
glycerol-sufficient environments (Neijssel et al. 1975). Matin cysts share with endospores structural rigidity and resistance to
(1979) reported the onset of a multiple substrate utilization desiccation and to ultraviolet radiation; however, they lack heat
technique as a response to decreasing dilution rates and pools resistance (Sadoff 1975). Similarly, myxobacteria (Myxococcus)
of intracellular metabolites. Adaptation of distinctive species to turn into spherical cells called myxospores (Sadoff 1973; Voelz
low-nutrient environments is well illustrated by a comparison and Dworkin 1962). In the genus Methylocystis, the rodlike cells
with respect to starvation and survival of a freshwater Spirillum round up and turn into desiccation-resistant cysts similar to
sp., apparently belonging to the oligotrophic flora, with those of Azotobacter species. Methylomonas and Methylococcus
a Pseudomonas sp. adapted to environments richer in nutrients form spherical cells that are neither thick walled nor resistant to
(Matin and Veldkamp 1978). The spirillum accumulated desiccation (Whittenbury et al. 1970). A modest resistance to
poly-b-hydroxybutyric acid (PHB) during carbon-limited desiccation is shown by Arthrobacter coccoid cells (Boylen 1973;
growth in the chemostat; the stored amount was Ensign and Wolfe 1964; Veldkamp et al. 1963).
highest at the lowest dilution rate examined. After growth at Bacteria that are not able to transform into resting stages are
D = 0.03–0.05 h1, the spirillum was much more resistant to less apt to survive under adverse environmental conditions.
starvation than the pseudomonad that did not accumulate PHB Their resistance, however, varies from species to species; it may
(Matin et al. 1979). The survival value of storage materials and be high as in mycobacteria or arthrobacters, which survive
the regulation of their synthesis in various types of bacteria may desiccation for several weeks, or low as in Neisseria gonorrhoea
assist the understanding of adaptation to low-nutrient or Treponema pallida, which respond to desiccation by immedi-
environments. ate death. Furthermore, maintenance of the viability of vegeta-
One of the most important characteristics of low-nutrient tive cells under a mild environmental stress is a function of
habitats is the strong influence of solid surfaces in the colonizing a multitude of factors, such as growth conditions, content of
of microbial films and layers (Marshall 1976). In a summary storage materials, speed of transition from growth to resting
of work done on the effect of inert particulate material conditions, and many others (Strange 1976). Many observations
suspended in low-nutrient aquatic habitats, Jannasch and on the effect of drying and rehydration of vegetative bacteria
Pritchard (1972) included studies using the chemostat for have been made during long-term preservation in culture
the enrichment of bacterial strains that exhibit a tendency for collections (Bousfield and MacKenzie 1976; Lapage et al. 1970;
specific attachment mechanisms. In Caulobacteriales, Martin 1964; Miller and Simons 1962). These observations
typical low-nutrient organisms, Poindexter (1979) postulates indicate that the survival of bacteria depends on many factors
a morphological–physiological mechanism that regulates an and may vary among species and environments. Many aspects
efficient metabolic response to low-nutrient conditions, were reviewed by Kjelleberg et al. (1987) and Matin et al. (1989).
including the role of holdfast organelles.
Life under conditions of low-nutrient concentrations was
a topic of a Dahlem Konferenz (Jannasch 1979; Koch 1979; Light as an Extreme Environment
Poindexter 1979; Rittenberg 1979) and an international
symposium on microbial ecology (Hattori et al. 1989). The well-known predominance of pigmented colonies on nutri-
ent-agar plates exposed for some time to air, compared to plates
inoculated with soil, indicates the selective action of sunlight
and air and the survival value of pigments. Yellow, orange, and
Resting Stages and Survival red pigments point to members of the genera Micrococcus,
Corynebacterium, Mycobacterium, and Nocardia, which contain
In order to survive under adverse environmental conditions, carotenoids. In nonphototrophic bacteria, there is a good
a few bacterial groups are able to form resting cells, which are negative correlation between the presence of carotenoids in
more resistant to deleterious environmental conditions than the cell and the anaerobic way of life; carotenoids are rarely
their vegetative cells. The transformation occurs when the found in strict anaerobes except the phototrophic bacteria.
metabolic activities decline because of nutrient depletion or Selective effects similar to those observed in air are exerted in
transfer to a growth-limiting environment. The resting cells are several sunlight-exposed environments, such as leaf surfaces,
characterized by thick, frequently multilayered walls. Typical is freshwaters, and salt brine. The radiation resistance of
endospore formation by the genera Bacillus and Clostridium. pigmented bacteria has been studied with Halobacterium
Many endospores are extremely resistant to heat, desiccation, salinarium (Dundas and Larsen 1962), Rhodospirillum rubrum,
radiation, and chemical agents. The formation of exospores is, as Corynebacterium poinsettiae (Cohen-Bazire and Stainer 1958),
far as is known, restricted to the methane-utilizing bacterium, and other microorganisms.
Methylosinus trichosporium; the thermoresistant exospores are Resistance to radiation also seems to be positively correlated
formed by a process similar to budding of vegetative cells with a high GC content of the cellular DNA; this correlation
(Whittenbury et al. 1970). agrees with the conclusion that inactivation of cells by ultraviolet
light is mainly due to thymine dimerization (Singer and Ames low (Acher and Juven 1977), the ecological significance of pho-
1970). The hypothesis was in accordance with the distribution of tosensitization by natural pigments liberated by autolysis of part
GC contents of the purple and green bacteria, the majority of of the population is far from clear. Carotenoid pigments also
which possess DNAs with 60–70 mol% GC (Mandel et al. 1971). exert a protective effect against photosensitized killing, as has
However, in contrast to the anoxygenic phototrophic bacteria, been shown for Micrococcus luteus (Anwar et al. 1977; Matthews
the cyanobacteria show a markedly different mean DNA base and Sistrom 1959; Prebble and Huda 1977).
composition, the majority of species having only 40–50 mol%
GC (Herdman et al. 1979). Cyanobacteria obviously need solar
radiation and are more exposed to ultraviolet light than most Surfaces as Habitats
bacteria. One would have expected them to possess a high GC
content, if the hypothesis of Singer and Ames holds true. Solar The microbiologist dealing with homogeneous suspensions of
radiation should, therefore, not be considered as the selective bacteria for physiological studies usually pays little attention to
factor that led to the great divergence of DNA base composition the growth of bacteria on surfaces. In nature, however, the
among prokaryotes. The effect of high irradiation, mainly of growth of bacteria as a surface film may be more frequent and
ultraviolet-light damage to DNA and its repair, has been recently important than growth in homogeneous suspensions (Marshall
reviewed by Nasim and James (1978). 1976, 1984; Fletcher and Marshall 1982) (see also Chap. 11,
The bactericidal action of diffuse sunlight has not gained ‘‘Planktonic Versus Sessile Life of Prokaryotes’’ in Vol. 2). Three
much attention. The growth-retarding and lethal action of kinds of interfaces have to be considered: (1) liquid–solid inter-
wavelengths of light not absorbed by DNA has been shown for faces, such as stationary objects in a river, the tidal zone, and
Serratia marcescens (Swart-Füchtbauer and Rippel-Baldes 1951), ponds; the inner surfaces of the mouth and the intestinal tract;
Nitrobacter winogradskyi (Müller-Neuglück and Engel 1961), the outer surfaces of rocks piled up for bacterial leaching of ore
and Nitrosomonas europaea (Schön and Engel 1962). The or of rocks stacked in columns for wastewater purification in
inhibitory effect is apparently most pronounced and therefore trickling filters; food particles in the digestive system; and silt
easily recognizable with slowly growing bacteria. Further studies particles in rivers or marine environments; (2) liquid–gas inter-
were reviewed by Jagger (1983) and Cornax et al. (1990). faces, such as the water surface of a pond and gas bubbles in the
Light exerts its killing effect on microorganisms primarily ocean; and (3) liquid–liquid interfaces, such as oil droplets in an
in the presence of oxygen; in its absence, the effect is smaller by aqueous solution.
several orders of magnitude. The killing effect is apparently
due to photooxidation reactions that are mediated by singlet
oxygen. This highly toxic oxygen type is produced by Liquid–Solid Interfaces
photosensitized activation of triplet oxygen. Protection against
single oxygen is readily provided by carotenoids that possess Adhesion to surfaces is an important ecological determinant in
more than seven conjugated double bonds. These carotenoids many ecosystems. The colonization of a suitable liquid–solid
are able to quench singlet oxygen as well as excited photosen- interface may be the prerequisite for exploitation of a habitat.
sitizers very effectively. Experimental results and a literature Many organisms that inhabit rushing rivers stick to solid sur-
review ‘‘on the mechanisms for coping with the stress of faces. In open ecosystems, sessile animals as well as bacteria take
photosensitized oxidations’’ have been presented by Krinsky advantage of being fixed to a privileged location. In a running
(1979). river that receives organic waste from upstream, slow-growing
The information on the tolerance to intense solar radiation bacteria could not compete with fast-growing bacteria if they
is still scarce. The generalizations on the protective effect were not attached to stationary objects at a location where food
of carotenoids just outlined are mainly based on studies of is still available in concentrations sufficient for growth. Growth
anaerobic, phototrophic bacteria. Among the aerobic bacteria, in locations of this kind is restricted to those bacteria that can
only those containing carotenoids were considered. There attach themselves to a solid surface. Thick plaits of Sphaerotilus
are many more bacteria characterized by remarkable natans were found in rivers where wastewater of sugar factories
pigments different from carotenoids, such as pyocyanine was discharged (Cohn 1881; Demoll and Liebmann 1952;
(Pseudomonas aeruginosa), prodigiosin (Serratia marcescens) Dondero 1961, 1975; Kolkwitz 1904–1906). A similar habitat
and related pigments (Gerber 1975), indochromes, violacein, for mass development is held by Leptothrix ochracea in field
xanthomonadines, and others. It would be interesting to know drains (Dondero 1975), by Crenothrix polyspora in water wells
the ecological significance of these pigments in the natural (Völker et al. 1977; Wolfe 1960), and by Toxothrix trichogenes in
habitats of the bacteria containing them. iron springs (Krul et al. 1970; van Veen et al. 1978). The trickling
The sensitivity of bacteria and protozoa to visible light and filters of waste-disposal plants resemble the river habitat. In
oxygen can be increased by the addition to the medium of some of these habitats, Sphaerotilus, Caulobacter, Asticcacaulis,
sensitizing pigments such as methylene blue, toluidine blue, Zoogloea ramigera, and other stalked, prosthecate, or slime-
eosin, or acridine orange. Although this photodynamic effect forming bacteria have been found in addition. In many of
has long been known and is advantageously applied on animal these habitats, the bacteria form thick, slimy layers on rocks
farms to keep the microbial number in drinking water and pebbles.
Bacteria are able to stick to solid surfaces in several ways. The

outer layers of the bacterial cell envelope (Costerton et al. 1974)
evidently play a special role. The cells are either attached by
a sheath as in the case of Sphaerotilus, Leptothrix, and Leucothrix;
by holdfast substances, which are apparently a gumlike outer
layer of the cell envelope and consist of polymers (Umbreit and
Pate 1978); or by nonflagellar, filamentous appendages such as
fimbriae or pili. The attachment of bacteria to solid surfaces by
polymers is apparently widely distributed.
As mentioned above, at high flow rates of a nutrient
medium, sessile bacteria are able to exploit the nutritional
opportunities of the habitat better than nonattached forms.
Cells that do not stick to the surfaces of stationary objects
are washed out into another ecosystem; the nutrient exchange
is not facilitated by the movement of the medium past the
cell. The ability to adhere to surfaces confers a selective
. Fig. 3.1 advantage on certain metabolic types. Studies on the fine
Microflora on the surface of the brown alga Ascophyllum nodosum. structure of extracellular polysaccharide fibers (Cagle 1975)
Scanning electron micrograph (Courtesy of Ralph Mitchell) gave rise to a generalized view of the ability of bacterial cells
to adhere to surfaces. The fiber network that extends from
the bacterial surface, called bacterial ‘‘glycocalyx,’’ mediates
In the marine environment, Leucothrix mucor holds a similar adhesion not only to abiotic surfaces but also to other cells,
position to Sphaerotilus in freshwater. L. mucor forms sessile either host or prey.
filaments and grows as a bacterial epiphyte on seaweed, such as Adhesion is a way for a bacterium to enter a habitat. The
filamentous red and green algae living in habitats where water adhesion of the causative agent of caries, Streptococcus mutans,
flows and aeration is good (Kelly and Brock 1969; Raj 1977). The to the enamel surface of the tooth, the adhesion of Vibrio
filament is attached to the seaweed by a holdfast (Harold and cholerae to intestinal cells, and other bacterium–host relation-
Stanier 1955) that is part of the sheath (Pringsheim 1957). By ships are examples for the potential role of glycocalyx formation
scanning electron microscopy (> Fig. 3.1), bacterial lawns were in the establishment of infectious diseases (Costerton et al. 1978,
revealed that consisted of long and short end-attached bacteria, 1987). Several unidentified, segmented filamentous bacteria
including Leucothrix mucor and flexibacteria (Cundell et al. attached to the epithelium of the mucosa of the small bowels
1977). Thiothrix nivea was observed to grow in tufts attached of mice have been revealed by scanning electron microscopy
to pebbles in the runoff of a sulfur spring in Seattle, Washington (Blumershine and Savage 1978).
(Bland and Staley 1978). For utilization of some solid substrates, a close contact is
In rivers and seawater, small, floating particles such as silt, necessary between the bacterium and its substrate. Cytophaga
clay, or detritus particles have a remarkable growth-promoting cells adhere closely to cellulose fibers, the rods being aligned with
effect on microorganisms. At low-nutrient concentrations, the orientation of the microfibrils. Cellulose degradation by
which are suboptimal for efficient growth in homogeneous Cytophaga and Sporocytophaga requires direct contact between
suspension, the addition of suspended particles favors the the bacteria and the cellulose fibers (Berg et al. 1972; Stanier
growth of bacteria (Heukelekian and Heller 1940) due to 1942). The attachment of rumen bacteria to plant particles
adsorption of nutrients to the particle surface. The attachment (Akin 1976; Akin and Amos 1975) is well documented.
of bacteria to dense suspensions of silt particles, as is character- Ruminococcus species were shown to adhere strongly to cotton
istic for certain river waters, has a marked effect on growth and cellulose fibers and to the cell walls of leaf sections of rye grass,
oxygen uptake (Jannasch 1955). The degree of attachment was evidently by means of their prominent glycoprotein coats
affected by the concentration of the dissolved organic substrate (Latham et al. 1978; Minato and Suto 1978; Patterson et al.
(Jannasch 1958; Jannasch and Pritchard 1972). Dense growth of 1975). Further examples concern the adherence of starch-
bacteria on suspended particles may lead to anoxic microenvi- digesting bacteria to starch grains, of chitin-hydrolyzing bacteria
ronments, as demonstrated by model experiments on denitrifi- to chitin, of Sulfolobus to sulfur globules (Weiss 1973), and of
cation (Jannasch 1960, 1978). a thermophilic sulfur oxidizer to sulfide minerals (Golovacheva
The ability to attach to surfaces may be considered as 1979). Knowledge about the specific mechanisms of attachment
a means to escape low-nutrient environments. Interfaces play of bacteria to solid substrates suggests methods for their
an eminent role in the transport and accumulation of nutrients, enrichment.
such as polysaccharides or proteins. Not only do these nutrients Specific interactions between the bacterial cell coat and the
accumulate at solid-liquid interfaces, but they even accumulate plant cell wall apparently precede the invasion of rhizobia into
at liquid–gas interfaces, and gas bubbles serve as vectors for the root hairs of their host plants. Phytohemagglutinins or
nutrients (Marshall 1979). lectins (Liener 1976) of the plant may be involved in the
recognition process by binding only the corresponding bacteria a diverse bacterial flora, sometimes called the bacterial neuston.
and not bacteria that infect other legumes (Dazzo et al. 1978; Common to the liquid–air interfaces are (1) direct exposure to
Kato et al. 1979; Marx 1977). the oxygen of the air, (2) accumulation of hydrophobic sub-
Fimbriation is apparently another means of bacterial attach- stances either originating from the air or from the water body,
ment to submerged objects (Hirsch and Pankratz 1970). The and (3) high light intensities.
possession of filamentous, nonflagellar appendages seems to be The surface film of lakes and ponds formed in summer
distributed mainly among members of the Enterobacteriaceae during calm weather harbors bacteria such as Pseudomonas,
and Pseudomonadaceae but is not restricted to members of Caulobacter, Hyphomicrobium, Nevskia, Flavobacterium,
these families (Ottow 1975). The ecological importance of fim- Alcaligenes, and Micrococcus (Babenzien 1965, 1967; Hirsch
briae or pili may consist in their initiating both attachment to 1974; Hirsch and Pankratz, 1970). Bacteria that contain gas
solid surfaces as well as contact with other members of the vacuoles may dominate under certain conditions (Walsby 1978).
community. The solid surface may be a substrate of low solubil- The surface layer of the ocean is a separate environment
ity, such as sulfur oxidized by Sulfolobus (Brierley 1978; Weiss (Zaitsev 1971). It has gained special attention because oil pollu-
1973) or Thiobacillus A2 (Korhonen et al. 1978). Attachment to tion has increased and wide areas of the ocean are occasionally
surfaces has not been studied as well as the phenomenon covered with petroleum (Bartha and Atlas 1977). A considerable
called star formation. The formation of stars, rosettes, or proportion of organic substances accumulating in the marine
cellular aggregates was originally observed when strains of neuston is released from planktonic organisms; fatty acids,
Agrobacterium tumefaciens and Rhizobium were studied lipids, polysaccharides, hydrocarbons, and proteins have been
(Stapp and Bortels 1931; Stapp and Knösel 1954). Since both identified in the surface film (Wangersky 1976). The marine
Agrobacterium tumefaciens and rhizobia normally invade plant neuston is the habitat of various marine bacteria (Bezdek and
tissues and must attach to the surface of the plant beforehand, Carlucci 1972).
the connection between attachment, important in the natural The surface film of water baths in laboratories offers a special
habitat, and the formation of aggregates as observed in slide habitat for bacteria that either utilize traces of conventional
cultures under the microscope is obvious. The formation of the substrates inadvertently added to the water or that utilize vola-
frequently very regular rosette-like aggregates in pure cultures of tile or gaseous substrates present in laboratory air (Leifson
bacteria, such as Pseudomonas ‘‘rhodos’’ (Marx and Heumann 1962). The gum formed within several weeks contains
1962), Pseudomonas ‘‘echinoides’’ (Heumann and Marx 1964; Bacteriogloea, Hyphomicrobium, Caulobacter, Mycobacterium,
Mayer 1971; Mayer and Schmitt 1971), and Agrobacterium and Nocardia.
‘‘luteum’’ (Ahrens et al. 1968) may be regarded as
a consequence of the presence of polar fimbriae and slime in
these bacteria. Similar aggregates are formed on the surface of Gradients as Habitats
liquid media (Ahrens et al. 1968).
Aggregation may be either specific, as in the cases of bacterial In nature, the distribution of nutrients or environmental factors
conjugation, host cell infection, and growth at special surfaces, is generally patchy. In many ecosystems, vertical gradients of
or may be due to a rather nonspecific process (Fletcher and Loeb concentrations of various nutrients exist. The nutrients pro-
1979), when extracellular polymers such as complex polysaccha- duced in sediments diffuse into the aqueous layer and, in the
rides and polyamino acids are excreted or are exposed at the absence of mixing by convection or currents, form
cellular surface under varying physiological conditions, espe- a concentration gradient. Such gradients are to be expected
cially during the stationary growth and the death phases and at either in a microscale in sediments, soil, or shallow water bodies
low-nutrient concentrations. The formation of microbial aggre- or in a macroscale in lakes or oceans. The gradients may either be
gates is of considerable importance in the soil, in floc formation, of substances used as nutrients or of physical factors such as
in activated sludge treatment, and in industrial biomass produc- redox potential, temperature, and radiation. Of primary impor-
tion (Harris and Mitchell 1973). tance for growth and distribution of microorganisms are the
Microbial adhesion has recently grown into a self-contained concentration gradients of organic acids, hydrogen sulfide, car-
field of research (see Chap. 11, ‘‘Planktonic Versus Sessile Life bon dioxide, and oxygen.
of Prokaryotes’’ in Vol. 2). The phenomenon is studied on the The major amount of organic matter that is decomposed by
physiological (Costerton et al. 1981, 1985; Berkeley et al. 1980), microorganisms in nature consists of fresh plant litter. Another
biochemical (Kefford et al. 1982), molecular (Switalski et al. fraction enters the soil or sediments of aquatic ecosystems
1989), and biotechnological level (Savage and Fletcher 1985; after partial degradation by herbivorous animals. Only
Characklis and Marshall 1990). a comparatively small fraction of the plant tissue is degraded
by animals; the major part, cellulose and lignin, is released and
becomes the major substrate of the microbial food chain.
Liquid–Gas Interfaces Terrestrial and aquatic ecosystems differ greatly with respect
to further degradation of the primary biomass and the cellulose-
The liquid–gas interface is a unique habitat in many respects. rich detritus derived therefrom. On the land, the solid organic
Depending on the ecosystem, this environment harbors matter stays on or near the surface of the soil and is decomposed
LIGHT
Aerobic primary production

ALGAE, CYANOBACTERIA by oxygenic photosynthesis
minerals
and CO2
Decomposition, Consumption
BACTERIA, ZOOPLANKTON, ANIMALS

EPILIMNION
Secondary production
cellular
material
THERMOKLINE, CHEMOKLINE
Anaerobic primary production

PURPLE AND GREEN SULFUR BACTERIA by anoxygenic photosynthesis
H2S sulfate
HYPOLIMNION
reduction
CO2
CH4 Anaerobic decomposition
(fermentations, sulfate reduction,
minerals
methanogenesis)
SEDIMENT, MUD
. Fig. 3.2
Diagram of production, consumption, and decomposition in an aquatic ecosystem with a chemocline
mainly aerobically. Only part of the organic matter is a variety of diverse living conditions for microorganisms. The
transported into the soil by animals as vectors and is degraded study of concentration profiles and of the zonation of aquatic
by microorganisms either aerobically or anaerobically, ecosystems and the assignment of the microorganisms to the
depending on aeration and moisture content. Undoubtedly, particular conditions of their habitat is a major concern of
diffusion processes and concentration gradients play a role in microbial ecology. Several lakes have been studied as model
the soil. However, due to the heterogeneous distribution of systems (Clark and Walsby 1978b; Gorlenko et al. 1977;
organic matter, the basic features of anaerobic and aerobic Kuznezow 1959, 1977). The stratified lake will be presented
food chains cannot easily be recognized in terrestrial systems. here as an example. The principles derived from these studies
In contrast, in a stagnant body of water, the solid organic can be easily applied to zonations and processes occurring in
matter produced in the surface layers sinks, either immediately water-logged soils, tundras, and rice fields. They apply with
or after passage through a short, animal food chain, to the slight variation also in shallow ponds, sediments, and shallow
bottom of the water body. There microbial degradation occurs. mats in the runoffs of hot sulfur springs.
Due to its low solubility in water, the oxygen may soon be
exhausted in the course of the initial aerobic degradation of
organic substances. In highly productive waters and in the Gradients of a Macroscale: The Stratified Lake
absence of convection, the deeper layers become anoxic, which
results in anaerobic microbial degradation processes. These pro- Many lakes are temporarily or permanently stratified (> Figs. 3.2
cesses give rise to the production of various soluble or gaseous and > 3.3). Stratification occurs when a dense, either cold or
substances such as organic acids, hydrogen sulfide, hydrogen, saline (salt-rich) water body is overlayered by a less dense, either
methane, carbon dioxide, and ammonia. These products freely warm or salt-free water body. The stratification patterns vary
diffuse upward, each forming its own concentration gradient. with different climate zones. In temperate climates, normal
The chemical conditions may be further diversified by the stratification occurs in freshwater lakes more than 10 m in
liberation of ions from the sediment. These upward gradients are depth.
overlayered by the downward gradients of temperature, light, The holomictic stratified lake may serve as an example of the
and oxygen. Thus, due to the gradients of various nutrients and stratification and gradient-forming process (Overbeck 1972;
other environmental factors, a stagnant water body offers Pfennig 1979; Sorokin 1970). After total circulation in winter
0 As soon as anaerobic conditions have been established in

1
the bottom layer or in the total hypolimnion, the anaerobic
6
conversion processes, which started with the degradation of
5 O2 solid material, mainly cellulose, in the sediment, continue
in the water body. Primary soluble fermentation products,
T 9 10 hydrogen gas included, are used to reduce sulfate to hydrogen
2
Depth (m)
10 sulfide. The major amount of hydrogen sulfide originates from

3 sulfate reduction in the free water. Simultaneously, nitrate is
CH4
8 reduced to nitrite, which transiently accumulates and forms
15 4 a distinct layer, and then to nitrogen.
H2S The upper part of the epilimnion is the zone of primary
production by the oxygenic phototrophic organisms, such as
20 7 photosynthetic higher plants, algae, diatoms, flagellates,
green algae, and cyanobacteria. These primary producers are
5
accompanied by bacteria, protozoa, and metazoa that consume
Concentrations Rates Biomass
and temperature photosynthesized biomass, which results in secondary
production and release of cellulosic detritus (Fenchel and
. Fig. 3.3 Jørgensen 1977). The thermally stratified lake is a very common
Idealized vertical profile of a freshwater lake in the temperate habitat for cyanobacteria that contain gas vacuoles. Oscillatoria
climate zone in summer, showing concentrations, conversion rubescens, O. agardhii, Aphanizomenon flos-aquae, and
rates, and biomass. The figure is based on drawings and data of Microcystis aeruginosa form stable populations near the bottom
Sorokin (1970), Gorlenko, Dubinina, and Kuznezow (1977), and of the epilimnion (Clark and Walsby 1978a, b).
Overbeck (1972). Symbols: T temperature, (1–5, conversion rates) The chemocline and the hypolimnion are the favored
1 CO2 fixation in the light (oxygenic photosynthesis), 2 CO2 habitats of the anaerobic prokaryotes. Provided hydrogen sulfide
fixation in the dark, 3 CO2 fixation in the light (anoxygenic is present and the light intensity allows, anoxygenic
photosynthesis), 4 and 5 sulfate reduction, 6 biomass of algae and phototrophic bacteria develop underneath the chemocline and
cyanobacteria, 7 total bacterial biomass, 8 biomass of form a second layer of primary biomass production (Sorokin
phototrophic bacteria, 9 biomass of protozoa, 10 biomass of 1970). The upper hypolimnion just below the chemocline is the
Copepoda and Cladocera habitat of the purple and green sulfur bacteria, among which the
brown forms dominate (> Fig. 3.2). They produce biomass by
time, the warming of the surface layers during the summer yields anoxygenic photosynthesis from carbon dioxide and hydrogen
a less dense upper layer (epilimnion) of circulating, aerated sulfide. Bacteria that are either buoyant by gas vacuoles or motile
water and a dense, cold, stagnant bottom layer (hypolimnion). by flagellation dominate. The buoyancy of Lamprocystis,
The intermediate layer (metalimnion) is characterized by Amoebobacter, Thiodictyon, Thiopedia, Pelodictyon, and
a temperature gradient (the thermocline) and, after oxygen Ancalochloris is apparently just sufficient to allow them to float
deprivation and anaerobic decomposition of organic matter in in the heavy cold water beneath but insufficient to float in the
the hypolimnion, by a chemical gradient (chemocline) also. light warm water above the thermocline. Among the motile
In the epilimnion, which is exposed to sunlight, biomass is purple sulfur bacteria (Chromatiaceae), the large Chromatium
produced by phototrophic cyanobacteria, diatoms, and green species (C. okenii, C. weissei, C. warmingii, C. buderi) and
algae; allochthonous matter from the surroundings usually adds Thiospirillum live in this layer.
to the total organic material. Part of the biomass usually sinks to Further details on the ecology of photosynthetic
the hypolimnion and to the bottom of the lake where it is bacteria with emphasis on the distribution in stratified lakes
degraded. Degradation is accompanied by oxygen consumption, were presented in a comprehensive review (Pfennig 1979).
resulting in a decrease in oxygen concentration and, finally, in The efficiency of biomass production by anoxygenic photosyn-
anoxic conditions, first in the sediment, then in the bottom layer. thesis is remarkable. Part of the biomass that by slight
Continued anaerobic degradation results in the production of turbulences reaches the upper chemocline is grazed by
organic fermentation products as well as hydrogen, methane, copepods, cladocera, and protozoa and transported into upper
hydrogen sulfide, and carbon dioxide, which diffuse upward aerobic regions. Photosynthesis by the purple and green sulfur
forming concentration gradients. Methane, which is the major bacteria is accompanied by sulfate production. The sulfate
carbonaceous substance released from the sediment, is produced is immediately reduced to hydrogen sulfide.
discharged in the form of gas bubbles. Part of the methane is As documented by measurements of the sulfate-reducing
dissolved in the water as it moves upward and is oxidized by activity and cell counts of sulfate-reducing bacteria in various
methane-utilizing aerobic bacteria. The quick removal of oxygen depths, the activity of sulfate reduction has two maxima: one
from the hypolimnion is primarily due to the quick dispersal of in the hypolimnion bottom layers and one below the
methane and the growth of methane-oxidizing bacteria. Finally, chemocline (Sorokin 1970). While the reduction power for
the hypolimnion becomes totally anoxic. the first process originates from the sediment, sulfate
reduction below the chemocline seems to be supplied by indig- sulfide concentration indicate the significant role of the purple
enous solutes released from cells within the region, either by sulfur bacteria in the photic zones of these coastal areas. The
excretion or by decay. bacteria respond to the exhaustion of sulfide and the alternation
The metalimnion is a layer of high biological activity. of the oxygen tension by chemotactic (aerotactic) movements
Because of its richness in inorganic nutrients as compared to and disappear in the course of illumination into the sediments.
the epilimnion, it is inhabited by a few aerobic photosynthetic The movements parallel the diurnal vertical migrations observed
prokaryotes that can tolerate hydrogen sulfide and anaerobic in stratified lakes. Depending on the particular environmental
conditions. One of them is Oscillatoria limnetica, which has been condition, the ecological niche of sulfide oxidation may be
isolated from the sulfide-rich layers of Solar Lake (Eilat, Israel) occupied by colorless sulfur bacteria such as Thiovulum,
and studied in detail (Cohen et al. 1975; Padan 1979a, b). It is Macromonas, Achromatium, and Beggiatoa or by phototrophic
able to use hydrogen sulfide as an electron donor in purple sulfur bacteria.
a photosystem I-driven reaction and to photoreduce carbon A characteristic ecosystem, with a very steep gradient from
dioxide; hydrogen sulfide is oxidized to elemental sulfur. Once the anoxic hydrogen sulfide-rich zone to the oxic layer, has been
the system has been induced, O. limnetica can grow under described as the ‘‘Farbstreifen-Sandwatt’’ (Hoffmann 1942); the
anaerobic conditions (Oren and Padan 1978; Oren and Shilo name was coined due to the succession of green, red, and black
1979). After they return to the oxic zones, the cells switch layers of the sand of marine coastal areas. The phenomenon is
from anoxygenic to oxygenic photosynthesis. This study of restricted to localities with a high level of ground water, sand of
O. limnetica explained the predominance of cyanobacteria in grain diameter to provide capillary ascension almost up to the
sulfide-containing microaerophilic habitats or in habitats of sand surface, sufficient organic material and sulfate in the lower
frequently alternating conditions (Padan 1979). sediment layers, and high light intensities to penetrate the upper-
Because of the oxidation of hydrogen sulfide by the purple most layers of sand. The green layer contains cyanobacteria
sulfur bacteria in the light and diffusion of oxygen from the exclusively, demonstrating their tolerance to hydrogen sulfide
epilimnion, the chemocline moves downward several meters compared to green algae as emphasized by Padan (1979). The red
during the daytime. Diurnal vertical fluctuations of the redox layer underneath is an almost monospecific culture of purple
discontinuity layer obviously represent a major selective factor sulfur bacteria, which is in direct contact with the black zone.
for the organisms occupying this habitat. These observations confirm the ecological significance of the
complementary spectral absorption of green plants and purple
bacteria (Buder 1919) and are in accordance with measurements
Gradients of a Microscale: The Sediments showing that long wavelengths penetrate in sand further than
short wavelengths (Fenchel and Staarup 1971), in contrast to
Among the various types of sediments, only the coastal marine water (Pfennig 1967).
sediment and the lacustrine sediment will be discussed here. In Multilayered microbial communities in aquatic ecosystems
principle, the degradative and biosynthetic processes that occur resembling the ‘‘Farbstreifen-Sandwatt’’ have been observed in
in sediments are similar to those in stratified lakes. The areas of hot springs and in solar ponds. Various kinds of
organic material that supplies the energy for life processes in mats have been described and studied in detail (Cohen and
the sediment is mainly allochthonous. Either leaves or large Rosenberg 1989).
algae are buried in the sand of coastal marine areas or detritus A particular environment is apparently required by the
is incorporated into the mud. sulfide-oxidizing, nonphotosynthetic bacteria such as Beggiatoa,
If sulfate is present in excess, as in marine ecosystems, the Thiothrix, Achromatium, Macromonas, Thiovulum, and
anaerobic food chain yields mainly hydrogen sulfide and gives Thiospira. Their habitats appear to be areas where the gradients
rise to the activities of bacteria involved in the sulfur cycle. of both hydrogen sulfide and oxygen are overlapping or, as in
Habitats of this sort gained early attention by the smell of running waters, where the hydrogen sulfide-providing sediment
hydrogen sulfide and the bright red layers and bloom of is covered by running water containing oxygen.
purple sulfur bacteria and have repeatedly been described (see Due to the instability of hydrogen sulfide in the presence of
Bavendamm 1924). There are numerous kinds of sulfureta, most oxygen, a habitat requiring the simultaneous presence of both
of them estuaries, limans, salt marshes, tidal flats, and swamps. compounds will be small and transient. The coexistence of
In typical sulfureta of littoral marine areas, buried organic hydrogen sulfide and oxygen has already been convincingly
material is covered by sand, and the water layer is 10–20 cm deep demonstrated by Beijerinck (1895). He introduced a solution
(Fenchel 1969). Hydrogen sulfide production may be high. The of hydrogen sulfide containing a small amount of oxygen into an
sulfide concentration in a few centimeters’ depth of these sulfide anaerobic suspension of luminous bacteria—precautions to
systems may easily reach 10 mM H2S (Fenchel and Riedl 1970). exclude the access of air were taken. Luminescence occurred,
The diffusing hydrogen sulfide may give rise to emission to the indicating that the hydrogen sulfide solution contained free
atmosphere during the night. During the day, the emission of oxygen and that the affinity of the bacteria for oxygen is high.
hydrogen sulfide amounts to only 4% of that during the night Hydrogen sulfide and oxygen will be simultaneously available
(Hansen et al. 1978). The diurnal fluctuations of the hydrogen for a long period of time only in areas to which both compounds
are continuously supplied. Observations in nature and when Horizontal Gradients

attempting to grow isolates of colorless sulfur-oxidizing bacte-
ria are in agreement with this idea (Bland and Staley 1978; Horizontal gradients of environmental factors exist where water
la Riviére 1963, 1965; Strohl and Larkin 1979; Chap. 15, ‘‘The of unusual properties and contents is discharged into running
Colorless Sulfur Bacteria’’ in Vol. 3). In a study on the water. Gradients are formed in the vicinity of hot springs; sulfur
chemolithotrophic nature of Thiovulum species (Wirsen and springs; effluents of coal, iron, and salt mines; rivers; ditches;
Jannasch 1978), the difficulty of obtaining pure cultures is and drainage tubes.
ascribed to the instability of the oxygen/hydrogen sulfide envi- The water discharged from sulfur springs gives rise to
ronment. The individual cells of Thiovulum aggregate in ‘‘veils’’ a gradient in sulfide concentrations, providing sulfide-rich
by extruding slime threads, a possible means of stabilizing an water and photoanaerobic conditions at the origin and
oxygen/hydrogen sulfide interface for a period of time (la Riviére photoaerobic conditions downstream. Hot sulfur springs as
1963). they occur in New Zealand, North Island (Castenholz 1976),
Careful studies using microelectrodes for the measurement and Yellowstone National Park (Brock 1978; Castenholz 1977)
of oxygen and hydrogen sulfide have been carried out on provide gradients of temperature, sulfide and oxygen
Beggiatoa populations at natural oxic/anoxic interfaces of estu- concentration, pH value, and secondarily, in the concentration
arine sediments (Jørgensen 1982; Jørgensen and Revsbech of minerals and organic nutrients; near the spring, the effluent
1983). In Beggiatoa ‘‘plates’’ prepared in artificial agar-gelled sulfide-rich water of 60–70 C enables the gliding green
interfaces or O2/H2S gradient cultures, chemoautotrophy was bacterium, Chloroflexus, to grow and form reddish orange
demonstrated (Nelson and Jannasch 1983), and the growth or orange-green mats. Downstream, this green bacterium is
pattern and yield were studied in detail (Nelson et al. 1986). accompanied by the cyanobacteria Oscillatoria, Synechococcus,
The thickness of these filament plates never reached more than and, eventually, Mastigocladus. The patterns do not vary greatly
a few mm. at different localities (Bauld and Brock 1973; Castenholz 1976,
The unusual occurrence of thick mats of Beggiatoa-like 1977, 1979).
filaments was observed on the surface of hydrothermal
sediments covering hot vents at a depth of 2,000 m in the
Guaymas Basin of the Gulf of California (Nelson et al. 1989). Possible Mechanisms for Finding or
These mats were 3-cm thick on the sediments and up to 30-cm Remaining in the Beneficial Layer of the
thick between stands of vestimentiferan tube worms, the Gradient
characteristic hosts of symbiotic, sulfide-oxidizing prokaryotes
(see below). The unusual mass development of Beggiatoa Motility and buoyancy may be involved in keeping the
at the hydrothermal vents is speculated to be due to bacteria in horizons of favorable growth conditions. Many
(1) a more efficient flux of H2S and O2 than by mere diffusion, motile bacteria manage, by means of their chemotactic response
(2) a continuous supply of an organic growth factor, (3) facul- mechanisms, to reach and stay in an area where the concentra-
tative chemoautotrophy providing an efficient metabolic flexi- tion of a nutrient or of oxygen is optimal. Bacteria that are
bility in case of intermittent or turbulent vent emissions, and motile by flagellation, as well as gliding bacteria, are able to
(4) an optimal temperature gradient (Jannasch et al. 1989). move to and stay in their beneficial environment. Gas vacuoles,
Furthermore, the unusually large diameter of the filaments (up occurring exclusively in aquatic microorganisms (Cohen-Bazire
to 122 mm) and an ‘‘empty’’ inner space (a liquid vacuole filling et al. 1969), provide buoyancy and may be a means by
more than 80% of the cell, the cytoplasm being distributed only which bacteria maintain their vertical position (Clark and
along the outer cell wall) may contribute to their growth and Walsby 1978a, b).
survival in this environment. These allow the cells to grow to an
enormous size without diffusional problems, thus providing
them with a structural rigidity necessary for exploiting a larger Chemotaxis
ambient space than smaller cells can without support of a
substratum (e.g., sediment). Moreover, an inner low-redox Chemotaxis is widespread among motile microorganisms
reservoir may enable these organisms to survive in the absence (Weibull 1960) and is of ecological importance in many systems
of reduced chemical substrates during flushing by oxygenated (Chet and Mitchell 1976). Usually, the chemotactic response
ambient seawater for a certain period of time. mechanisms result in the accumulation of microorganisms in
The classical filamentous and stalk-forming iron bacteria areas of favorable metabolic conditions and depletion in areas of
have to be considered as gradient organisms also. Bacteria such adverse conditions. However, there is almost no correlation
as Leptothrix ochracea and related bacteria, as well as Gallionella between the properties of a substance to act as an attractant
ferruginea, grow in drainage tubes in moist fields, preferably in and to serve as an energy source (Adler 1974; Weibull 1960). The
the narrow zone between the anaerobic bottom layer, which effect of an attractant has been shown to increase swimming
is the source of ferrous iron, and the flowing water, which periods toward the attractant after tumbling, while the
contains oxygen. swimming periods in the opposite direction are shortened
(Koshland 1974, 1976, 1980, 1981; Repaske and Adler 1981; (Mann et al. 1990). These crystals are often aligned in chains
Adler 1988; Hazelbauer 1988). The role of chemotaxis in the and associated with single crystals of nonmagnetic pyrite (FeS2).
vertical distribution is still a matter of speculation.
Buoyancy
Aerotaxis
Assuming that gas vacuoles provide buoyancy to the cells
The response of bacteria to oxygen may be a dominant mecha- (Walsby 1975, 1977), there may be two ways by which distinct
nism enabling bacteria to find and remain in an environment layers are formed. The bacteria may be able to migrate to the
favorable with respect to oxygen concentration. Vertical move- favorable depth, or the gas vacuoles may provide neutral buoy-
ments of clouds of phototrophic bacteria in Winogradsky col- ancy to guarantee a long residence time to the bacteria at depths
umns observed during diurnal cycles of light and dark may serve supporting their growth. The second alternative has been
as a model for investigating the response of bacterial populations supported by experimental evidence (Clark and Walsby 1978b).
toward oxygen and substrates. Gas vacuoles may also serve the distribution of cells and
resting stages; this function is exemplified by the gas vacuoles
of some clostridia, which are only formed during the transition
Phototaxis phase from the vegetative cell to the spore (Duda and
Makaer’eva 1977).
Phototaxis is apparently shared by all anoxygenic phototrophic An interesting alteration of buoyancy and its significance in
bacteria that are motile by flagellation. The phototactic response the natural habitat has been described in the case of
mechanism results in keeping the bacteria in an area of favorable Metallogenium. Metallogenium has been found in all oligotro-
light intensity after they accidentally enter. The response, phic lakes that contain manganese and also in mesotrophic and
reversal of flagellar movement, usually occurs when the cells eutrophic lakes. Its development occurs during the periods of
enter an area of lower light intensity. Although phototaxis has circulation. The vertical distribution of Metallogenium in
been well studied in laboratory cultures (Hustede et al. 1989), these lakes is dependent on the concentration of manganese.
investigations relevant to the significance of phototactic behav- Manganese is soluble in the reduced state. If oxidized by
ior in the vertical distribution of phototrophic bacteria in lakes Metallogenium, manganese is precipitated at the cell surface,
are lacking. The function of photokinesis, the initiation or and the heavy manganese oxides make the cells sink down to
acceleration of linear velocity by light, has not yet been studied the sediment. Under anaerobic conditions, manganese is
by ecologists (Nultsch 1975). reduced, the buoyancy is increased, and the cells return to the
aerobic layers. Thus, the movement of Metallogenium is strictly
correlated to the availability of manganese and oxygen.
Magnetotaxis
Bacteria containing magnetite (Fe3O4) crystals have been dis- Microbial Associations
covered (Blakemore 1975) and isolated (Blakemore et al. 1979;
Frankel et al. 1979). It is hypothesized that the downward Under natural conditions, the various types of prokaryotes live
orientation of the cell within the Earth’s inclined magnetic in more or less close associations. Interactions between bacteria
field lines on the northern hemisphere provides favorable in a common habitat may be either weak or absent; only species
growth conditions for these motile and strongly microaerophilic with quite dissimilar nutrient requirements can be expected to
organisms. However, when the studies were repeated in New show neutralism. The interactions may be strong and may be
Zealand and Australia, very similar organisms were found that characterized as mutualism, commensalism, or parasitism. The
exhibited a reversed polarity (Blakemore et al. 1980). relationships may be loose or tight, facultative or obligatory.
The anaerobic formation of magnetite by a marine Several examples of microbial associations studied in open
magnetotactic bacterium (designated strain MV-1) was demon- systems have been reviewed by Meers (1973).
strated using nitrous oxide as an electron acceptor (Bazylinski The most spectacular examples of syntrophic or of mutual-
et al. 1988). This finding obliterated the notion that the istic associations have been discovered in the course of metabolic
biological production of magnetite can only occur in aerobic studies on bacterial cultures presumed to be pure cultures.
top sediments. Some nonmagnetotactic, dissimilatory One example concerns the partners of the culture of
iron-reducing bacteria were described by Lovley and Phillips Methanobacterium omelianskii and illustrates a symbiosis with
(1987) and found also to synthesize extracellular magnetite unidirectional substrate supply of mutual benefit. The other
from hydrous ferric oxide under anaerobic conditions. concerns symbiotic associations with bidirectional transfer of
The formation or ‘‘biomineralization’’ of a ferrimagnetic iron small molecules involved in energetic coupling.
sulfide called greigite (Fe3S4) has been reported in a multicellular Another striking discovery is the symbiotic association
magnetotactic bacterium that has not been isolated but is between chemolithotrophic sulfide- and thiosulfate-oxidizing
common in brackish sulfide-rich water and sediments prokaryotes and newly described marine invertebrates found
. Table 3.5
Examples of interspecies hydrogen transfer by coculturing hydrogen-utilizing (methanogenic) bacteria with various H2-producing
bacterial species
Substrate H2-producing species; products of pure culture H2-utilizing species; products of coculture Reference
Glucose Ruminococcus albus; ethanol, acetate, H2, CO2 Vibrio succinogenes; acetate, succinate Iannotti et al. (1973)
Glucose Selenomonas ruminantium; lactate Methanobacterium ruminantium; acetate, methane, Chen and Wolin
CO2 (1977)
Cellulose Clostridium thermocellum; ethanol, acetate, H2, Methanobacterium thermoautotrophicum; acetate, Weimer and Zeikus
CO2 methane, CO2 (1977)
Cellulose Ruminococcus flavefaciens; succinate, acetate, Methanobacterium ruminantium; acetate, Latham and Wolin
formate, H2, CO2 methanol, CO2 (1977)
to live tightly clustered around deep-sea hydrothermal vents hydrogen transfer is the most significant example of
(Cavanaugh et al. 1981; Felbeck et al. 1981; Belkin et al. 1986; a symbiosis with unidirectional substrate supply from which
Distel et al. 1988; Fisher et al. 1989). Furthermore, a new genus mutual benefit is drawn.
of marine blue mussels was found to contain methylotrophic The principle of interspecies hydrogen transfer became clear
endosymbionts in their gill cells enabling the animal to thrive when the culture called Methanobacterium omelianskii was sep-
near methane emissions in the vicinity of ‘‘cold seeps’’ on the arated into a strain that produced hydrogen (S-organism) and
seafloor (Childress et al. 1986). A similar association was later a strain that oxidized hydrogen (Moh-strain) (Bryant et al. 1967;
described in certain polychaetes collected from marine sedi- see also Chap. 2, ‘‘Virulence Strategies of Plant Pathogenic
ments at the Skagerak in the Baltic Sea (Schmaljohann and Bacteria’’ in Vol. 3). These strains carry out the following
Flügel 1987). reactions:
S-organism : 2C2 H5 OH þ 2H2 O

Interspecies Hydrogen Transfer ! 2CH3 COOH þ 4H2
Moh-strain : 4H2 þ CO2 ! CH4 þ 2H2 O
The concentration of many metabolites in the cell is below
measurable levels. The intermediates of reactions, especially The S-organism grows poorly as a pure culture in media with
those which are catalyzed by multienzyme complexes, scarcely ethanol or other utilizable alcohols because the accumulated
reach measurable concentrations. The intermediates are used hydrogen inhibits growth. The S-organism lacks the ability to
as substrates of subsequent enzyme reactions at the same site dispose of electrons resulting from the oxidation of ethanol via
or in the same compartment where they are produced. The electron sinks other than hydrogen. For efficient growth of the
absence of measurable concentrations of a compound does S-organism, a hydrogen-utilizing bacterium has to be included
not mean that the compound is not involved in a metabolic to remove the hydrogen produced (Reddy et al. 1972a, 1972b).
reaction. The methanogenic bacteria keep the partial pressure of
Similar situations are encountered in many ecosystems hydrogen low; for the bovine rumen, a value of about three
where very important intermediates of anaerobic food chains times 104 atm (two times 107 M) was reported (Hungate
are present at scarcely detectable concentrations. In the rumen, 1967). Therefore, both organisms benefit from this symbiotic
lactate and hydrogen are present only at low concentrations, association. The methanogenic bacterium is continuously
although they are major products of the predominating anaer- supplied with hydrogen, and the hydrogen-producing organism
obic fermentations; in contrast, acetate, butyrate, propionate, can even degrade and generate energy from substrates such
valerate, and formate, which are not subject to further fermen- as lactate or ethanol, which, for thermodynamic reasons
tative conversion, may accumulate. In anaerobic ecosystems, (Wolin 1976), could not be degraded when hydrogen accumu-
only the end produces of anaerobic food chains accumulate or lated in the medium (Wolin and Miller 1982).
are released into the environment. The concentration of some As a result of interspecies hydrogen transfer between part-
degradation products, such as hydrogen, acetate, lactate, and ners of close symbiotic associations, therefore, special ecological
ethanol, is low. These compounds are consumed as soon as niches can be occupied. The principle has been exemplified by
they are produced. coculturing methanogenic bacteria with various hydrogen-
Metabolic products may be inhibitory to the cells that pro- producing bacterial species (> Table 3.5).
duce them; subsequent utilization of inhibitory products by For the fermentative degradation of substrates such as lactate
commensals, therefore, is useful to the ecosystem. In several or ethanol, the removal of hydrogen by a hydrogen-utilizing
cases, the cooperation of two or more organisms is essential, partner is obligatory; without continuous removal of hydrogen,
since a substrate would not be degradable at all if the concen- anaerobic growth on these substrates is slight. For the degrada-
tration of the product were not kept very low. Interspecies tion of other substrates, glucose included, the removal of
hydrogen is not obligatory; however, it enables the cell to obtain found capable of oxidizing acetate to carbon dioxide linked to
more energy than would be otherwise possible. the reduction of sulfur to hydrogen sulfide (Pfennig and Biebl
In media that contain little sulfate, Desulfovibrio vulgaris 1976). This bacterium, Desulfuromonas acetoxidans, was isolated
grows only modestly on ethanol or lactate producing acetate from a culture called Chloropseudomonas ethylica, which had
and hydrogen. Like the S-organism, it lacks the ability to pro- been held as a stable mixed culture for many years. The
duce a sink for electrons other than protons. In coculture with syntrophic association of the green phototrophic bacterium,
Methanobacterium, ethanol and lactate were actively fermented Chlorobium, and of D. acetoxidans occurs because hydrogen
with the production of acetate and methane (Bryant et al. 1977). sulfide and sulfur permit energetic coupling of the life processes
Stable methanogenic mixed cultures, which converted glucose to of both partners, as shown below:
methane, were also obtained by enrichment in the chemostat
(Siñeriz and Pirt 1977). Benzoate, which under anaerobic
CO2 S0 CO2
conditions can be utilized by phototrophic bacteria in the light light Cell material
or by chemoorganotrophs through anaerobic respiration, has
Chlorobium Desulfuromonas
been shown to be biodegradable even under fermentation
conditions (Nottingham and Hungate 1969; Boyd et al. 1983; Cell material S2 Ethanol
Sleat and Robinson 1984; Tschech 1989); methane and carbon
dioxide are the main fermentation products. Apparently, In mixed cultures, low concentrations of hydrogen sulfide
a microbial consortium of more than two bacteria is involved (7–8 mg/l) are sufficient to allow maximum growth of
in methane production from benzoate (Ferry and Wolfe 1976). Desulfuromonas strains and of Chlorobium or Prosthecochloris
The examples discussed above of interspecies substrate strains (Biebl and Pfennig 1978). The cell yield of the green
transfer concerned the unidirectional transfer of substrates bacterium is not limited by the hydrogen sulfide added to the
within food chains. In the next section, bidirectional transfer culture; it depends rather on the amount of ethanol, the hydrogen
will be discussed. donor.
To understand the function of a species in a habitat doubtless
requires studies in mixed culture. Only in mixed culture under
conditions of competition is the true actual niche revealed.
Bidirectional Transfer of Small Molecules According to van Niel (1955), Winogradsky ‘‘argued convinc-
ingly that pure culture studies may reveal characteristics that can
A bidirectional transfer of small molecules is the basis of some express themselves only in the absence of potential competitors.’’
symbiotic associations in which sulfate (or sulfur) and hydrogen The examples presented demonstrate the necessity for pure-
sulfide are involved in the energetic coupling of two different culture studies. Ironically enough, superb examples for the
strains of bacteria kept in mixed culture. mixed culture concept were noticed when it was discovered
Sulfate-reducing bacteria such as Desulfovibrio vulgaris that the ‘‘pure’’ cultures were not really pure.
growing with organic acids as hydrogen donors provide the
hydrogen sulfide used for anoxygenic photosynthesis by
Chromatium, which in turn reoxidizes hydrogen sulfide with Specific Aquatic Ecosystems: Marine
the production of sulfate. The occurrence of this light-driven Environments
sulfur cycle has been repeatedly described. When sulfate is
continuously added to the system, part of the sulfur intracellu- Although the borderline between the aquatic and terrestrial
larly accumulated in the Chromatium cells may leave the cycle habitat of microorganisms is diffuse, certain physical and chem-
and be deposited. The deposition of sulfur in certain Cyrenaican ical properties of water characterize the predominantly aqueous
(North African) lakes has been attributed mainly to this cycle environment distinctly with respect to the indigenous microbial
(Butlin and Postgate 1954), which is as follows: population. Inorganic and organic nutrients are available in
dissolved and ionic form and are highly mobile and constantly
dispersed by diffusion, convection, and currents. These charac-
Acetate, CO2, light SO42− Acetate, CO2 terizations of nutrient regime, together with the high heat capac-
Cell material ity of water, provide the characteristic constancy of the
Chromatium Desulfovibrio environmental conditions upon which many of the typical
aquatic microorganisms depend.
Cell material S2− Ethanol Another typical characteristic of aquatic environments is the
variety of interfaces that are of special microbiological impor-
Although the direct reduction of sulfur was already observed tance: the surface film (neuston); the interface between oxygen-
80 years ago (Beijerinck 1895), a chemoorganotrophic bacte- ated and anoxic water or sediment; horizontal layering of water
rium able to use elemental sulfur as hydrogen acceptor for with respect to gradients of light, temperature, and salinity in
growth was not known until the 1970s when a bacterium was stagnant waters; and finally the aqueous–solid interface of
submerged surfaces. These interfaces provide specific and Relating specific groups of prokaryotes with specific aquatic
often highly selective environmental conditions for the accumu- macrohabitats such as ponds, lagoons, rivers, or parts of the
lation and growth of microorganisms on the micro- as well as ocean is of limited value. A number of characteristic traits of
on the macroscale (see above section > ‘‘Surfaces as Habitats’’ these ecosystems, however, are of microbiological interest and
and Chap. 11, ‘‘Planktonic Versus Sessile Life of Prokaryotes’’ are dealt with in the following sections.
in Vol. 2). Seventy-one percent of the globe is covered by seawater,
The literature in ‘‘aquatic microbiology’’ largely deals with three-quarters of which lies below a depth of 1,000 m. Compar-
environments on the macroscale: the qualitative and quantitative ing the mean depth of the ocean to an arbitrarily assumed
assessment of populations in springs, ponds, rivers, lakes, and thickness of the terrestrial biosphere (depth of live soil and the
parts of the ocean. In addition, applied problems led to the zone of plant and animal life on and above the surface) at 38 m
investigation of groundwaters, acid mine waters, sewage lagoons, results in an aquatic/terrestrial volume ratio of 99/1. Productiv-
cooling water in pipe systems, etc. Only some of these studies ity, on the other hand, is related to surface area and the avail-
focused specifically on the particular aquatic characteristics. ability of light and nutrients. It is limited in the sea to the
It was stated earlier that the lack of water-preserving mecha- phototrophic surface layer and is estimated, in spite of the
nisms makes prokaryotes as a group appear to be typically aquatic above-mentioned volume ratio, in the same order of magnitude
organisms. Over and beyond this fact, certain microorganisms as the terrestrial photosynthetic production.
are specifically adapted to life at the typical conditions of The ambient concentration of dissolved organic carbon in
aquatic environments. An example is Thiovulum, a highly motile seawater lies in the range of 0.3–1.5 mg/l (Menzel and Ryther
and large (spherical cells of up to 25-mm diameter) 1970). Particulate organic carbon reaches not more than one-
chemolithotrophic organism that oxidizes hydrogen sulfide tenth to one-fifth of that concentration. Much of this material,
aerobically (see Chap. 15, ‘‘The Colorless Sulfur Bacteria’’ in especially in deeper waters, is ‘‘refractory’’ to microbial attack
Vol. 3). Growth of Thiovulum depends on its ability to locate (> ‘‘Low-Nutrient Environments’’). Except for a few nutrient-
in areas of the short-lived coexistence of hydrogen sulfide rich areas of high oxygen consumption by microorganisms,
and oxygen as spontaneously reacting compounds (Wirsen and offshore waters are aerobic down to the greatest depths. The
Jannasch 1978). The biology of the typically aquatic Caulobacter only attempt to cover the area of marine microbiology was made
group has been reviewed by Poindexter (1964) and Schmidt by ZoBell in 1946. More recent literature reviews deal with
(1971). These prokaryotes are uniquely adjusted to metabolize particular problems (see below). Initial efforts in reviewing the
at low-nutrient levels (Poindexter 1979) while largely attached to systematics of the characteristic indigenous bacterial flora, obli-
submerged solid surfaces with the aid of characteristic appendages, gately and facultatively aerobic rods, were carried out by
the prosthecae. Baumann et al. (1972). Luminescent bacteria, largely restricted
A large number of typically aquatic microorganisms are to the marine environment, are described as belonging to a new
microaerophilic, a point of evolutionary interest, and take genus of the Enterobacteriaceae, the Beneckea, and to the genus
advantage of the fact that water provides a barrier against Photobacterium (Baumann and Baumann 1977; Reichelt and
a quick replenishment of oxygen, as determined by the rates of Baumann 1973). The former occur as free-living forms and the
dissolution and diffusion. Except for areas of high mixing, as in latter as free-living and quite distinct symbiotic forms living in
mountain streams, or of high photosynthesis, as in the surface specific organs of marine fishes and invertebrates (Greenberg
layer of eutrophic lakes, concentrations of dissolved oxygen in et al. 1979; Hastings and Nealson 1977).
most natural waters are well below saturation values.
The occurrence of a large number of other microorganisms,
which can be easily and repeatedly isolated from natural waters,
is not necessarily diagnostic for their aquatic specialization but The Deep Sea
merely for (1) their ability to survive well under these conditions
or (2) a high rate of introduction from other habitats. Obvious The deep sea comprises the largest volume of seawater charac-
examples are the seasonal occurrence of typical soil bacteria in terized by the absence of light, by constant temperature around
streams and rivers or the common existence of enteric bacteria 2–3 C, by limited input of organic energy sources from the
in waters polluted by domestic sewage. remote surface water, and by considerable hydrostatic pressure.
The limited ability of enteric bacteria, as an extreme example, Because of the continuous increase of the last factor with depth,
to compete for survival in natural waters indicates another gen- about 1 atm every 10 m, there is no particular depth at which the
eral characteristic of aquatic habitats. The constant dispersion of start of the deep sea can be defined. Oceanographers distinguish
dissolved nutrients by diffusion and currents has a dilution effect between shelf waters of up to 200 m deep, the continental slope
and leads to generally low concentrations of most of the sub- area ranging to a depth of about 3,000 m, and the abyssal plains
strates essential for microbial growth. A central theme of aquatic reaching 6,000 m. The deep trenches may reach depths of up to
microbiology is the study of the physiology of growth and metab- 11,000 m, but the total area of the world’s ocean below 6,000 m
olism in the presence of low-nutrient concentrations (see section covers not more than 1.2 %, and the total volume amounts to
> ‘‘Low-Nutrient Environments’’ in this chapter). only 0.01 % (Sverdrup et al. 1942).
although their seawater volume is only 0.1% of the oceans’ total.

A Thus, the results are more of physiological than of ecological
A⬘ significance.
Since this pioneering work appeared, large numbers of
B barophilic bacteria have been isolated from various depths
RATE OF GROWTH
(Deming and Colwell 1982; Deming et al. 1981; Yayanos et al.

1982; Jannasch et al. 1982; Jannasch and Wirsen 1982). The data
C
generally indicate a close association between barophilism and
psychrophilism, confirming earlier observations (Wirsen and
D Jannasch 1975) on high barotolerance found primarily, but not
exclusively, in psychrophilic isolates. The large variability of
barophilism seems to merge gradually with barotolerant growth
HYDROSTATIC PRESSURE characteristics.
A direct comparison of all the published data is difficult, since
. Fig. 3.4
they have been obtained in different growth media. However, the
Summarizing scheme of growth responses by marine bacterial
general situation can be schematically depicted as in > Fig. 3.4,
isolates to hydrostatic pressure. A low barotolerance, A0 high
where A and A0 indicate a range of pressure between low and
barotolerance, B barophilic, C obligate barophilic, D slope of line
high barotolerance. Barophilic bacteria (B) also exhibit
indicates decrease of activity with increasing pressure (Modified
a considerable range of optimal growth pressures and are termed
from Jannasch and Taylor 1984)
obligate barophiles (C) if they cannot grow at normal atmo-
spheric pressure. The slope indicated by line D is a simplified
expression of the overall decrease of microbial activity as
The remoteness of the deep sea from productive surface a function of increasing pressure with depth. When considering
waters or from coastal input of organic energy sources causes microbial growth in the deep ocean, it should not be forgotten
the sparsity of life in deep waters. Yet a large number of bacteria that both the nutrient concentration (mainly the available
can be isolated from sediments of almost any depth. The early organic carbon) and the temperature are of predominant
work in deep-sea microbiology was reviewed by ZoBell (1970) importance. It is now technically possible to sample enrich and
and Morita (1976). Interest in this area was rekindled by exper- isolate bacteria from the deep sea in pure culture while
imental work on deep-sea in situ incubation (Jannasch and maintaining full pressure continuously (Jannasch et al. 1982).
Wirsen 1973).
The hydrostatic pressure and its possible effect on prokary-
ote metabolism (Marquis 1976; Marquis and Matsumara 1978) Hydrothermal Vents
are of special interest. Based on physicochemical considerations,
hydrostatic pressure may be assumed to affect metabolic The decreasing cell density observed in the oceans with increasing
processes only through differential volume changes. Unless depth has been explained as the result of the increasing remote-
dissolution and liberation of gases are involved, in principle, ness from the photosynthetically productive surface waters.
little or no effect can be expected at pressures below 1,000 atm. A striking exception to this general rule was discovered when
A high variability of barotolerance in marine bacteria, mostly copious populations of large, sessile invertebrates were found at a
documented in terms of growth in various media, has long been depth of 2,700 m clustered around warm-water vents of volcanic
known but is not fully understood yet. This is principally also origin in the Galapagos Rift area (Ballard 1977). The hydrogen
true for ‘‘barophilism,’’ a term introduced by ZoBell and Johnson sulfide content of the extruding waters (> Fig. 3.5) was shown
(1949) and today defined as optimal growth response at pres- to provide the energy (the electron donor) for the
sures higher than normal atmospheric pressure. Pure cultures of chemosynthetically rather than photosynthetically sustained
true barophilic bacteria only first became available at the time ecosystem either by free-living or by symbiotic sulfide- and
Dietz and Yayanos (1978) introduced a technique for their thiosulfate-oxidizing prokaryotes (Jannasch and Wirsen 1979;
isolation by using pressurized enrichments from decaying Grassle 1986; Jannasch 1989). Thus, these deep-sea animal com-
deep-sea amphipods in an organic-rich, silica-gel medium. Sub- munities are supported by terrestrial rather than solar energy
sequent isolations from materials collected at depths of 5,800 (> Fig. 3.6), the caveat to this statement being the use of
and 10,500 m yielded organisms that grew optimally at pressures photosynthetically produced oxygen in the pathway for aerobic
of about 500 and 690 atm (Yayanos et al. 1979, 1981). While sulfur oxidation. Although quantitatively of less importance for
some isolates grew at normal atmospheric pressure but at a rate the production of organic carbon, anaerobic chemosynthesis has
30-fold lower than at optimum pressure, others did not grow at also been demonstrated by the isolation of methanogens from
all after decompression and were called ‘‘obligate barophiles.’’ various sites at hydrothermal vents (Jones et al. 1989;
Because of the more striking psychrophilic/barophilic behavior Huber et al. 1989). Many of the aerobic chemoautotrophic and
of isolates from greater depths, much of the earlier work con- heterotrophic isolates from the warm and hot hydrothermal
centrated on deep-sea trenches, that is, at depths below 6,000 m, vent sites were extremely thermophilic (see above).
On mixing:
Ca2+ + SO42− CaSO4 FeS, Mn2+ + O2 FeO(OH), MnO2

2+ +
Fe + H2S FeS + 2H
Plume
Pillow lava
6°–23°C 350°C
Sheet lava
Seawater Sedimentation
Permeation
20°–100°C
Shallow
(10–200 m)
En route
precipitation:
Deep FeS, FeS2
(1–3 km) CuFeS2
350°C
Hydrothermal contour
fluid
SO42− S2− HCO3− CO2, CH4 Mg2+ > 2+

< Mn Ca
2+
Fe2+ Cu2+ H+
Basalt
. Fig. 3.5
Major geochemical processes occurring within the oceanic crust and on the deep-sea floor. As seawater penetrates several km into the
crust, it is heated to 350–400 C, reacts with basaltic rocks, and leaches various chemical species into solution. The highly reduced
‘‘hydrothermal fluid’’ rises and reaches the seafloor either directly (hot vents) or after mixing with cold, oxygenated seawater before
emission (warm vents). On mixing, polymetal sulfides and calcium sulfate (anhydrite) precipitate, either within subsurface lava conduits
or as ‘‘chimneys’’ and suspended particulate matter in the ‘‘black smokers’’ (Modified from Jannasch and Mottl 1985)
Marine Anoxic Ecosystems masses on the scale of the Black Sea (see below). The high
sodium concentration of seawater has been the key determinant
Shallow estuaries are similar to rich freshwater environments, for typically marine bacteria by constituting strict requirements
except for frequently and sometimes drastically changing salin- for the cation (McLeod 1968).
ities and for the presence of certain characteristic anions and The largest anoxic ecosystems in the biosphere are marine.
cations. The former is dealt with in the chapters on the habitats The Black Sea contains no free oxygen from a depth of about
of halotolerant and halophilic prokaryotes. Next to chloride, the 150 m down to the bottom at a 2,000-m depth. In an extensive
anion most characteristic for the marine environment is sulfate. team effort, the geological, chemical, and biological characteris-
In organically rich marine environments, the microbial reduc- tics have been studied (Degens and Ross 1974). Skopintsev,
tion of sulfate to sulfide initiates the cycling of sulfur com- Karpov, and Vershinina (1959) and Sorokin (1964) have done
pounds with a number of concomitant environmental earlier microbiological work, finding that about 95 % of the
phenomena: the production of hydrogen sulfide, the precipita- sulfide present in the deeper Black Sea water stems from sulfate
tion of ferrous (and other heavy metal) sulfide often followed by reduction and that the oxidation of sulfide at the oxic/anoxic
the formation of a disulfide (e.g., pyrite), the deposition of interface accounts for the large amounts of dark carbon dioxide
elementary sulfur, and the rich populations of sulfur-oxidizing fixation (4–6 mg C/m3·day). The auxiliary role of the cycling of
prokaryotes at the oxic/anoxic interface. Acidification rarely sulfur compounds in the turnover of organic and inorganic
occurs, due to the high buffer capacity of seawater. Anaerobic carbon has been described (Jannasch et al. 1974), and the
marine basins, chemically stabilized by high concentrations of biomass and total number of microorganisms in the depth of
sulfide, range from small estuarine pockets to anoxic water the Black Sea have been measured (Mitskevich 1979).
Solar Energy
hv
CO2 + H2O (CH2O) + O2 Photosynthesis
CO2 + H2O + H2S + O2 (CH2O) + H2SO4 Chemosynthesis, aerobic
2CO2 + 6H2 (CH2O) + CH4 + 3H2O Chemosynthesis, anaerobic
Terrestrial Energy
. Fig. 3.6
Scheme of the energy supply pathway for photosynthesis and chemosynthesis, showing the role of free oxygen (O2) at deep-sea
hydrothermal vents (From Jannasch 1989)
Oxygen (in mM)

A series of research cruises on the Black Sea (April–Septem- 5 10 15 20
ber 1988) has resulted in some unusual oceanographic data. The 40
oxic/anoxic interface, historically at depths between 125 and 50
175 m in the 2,000-m water column, had risen within the last O2
decade to 80–90 m (> Fig. 3.7). At the same time, a maximum
for bacteriochlorophyll e was detected at the same depth where B
hydrogen sulfide first appears on a downward profile in the 75
water column (Repeta et al. 1989). Such a high concentration H2S
of a prokaryotic pigment has never been observed before in the
Black Sea, and it exceeded the concentration of the phytoplank-
tonic chlorophyll a in the oxic waters above (> Fig. 3.8). This 100
Depth (m)
first observation of mass bacterial photosynthesis in offshore

waters indicates that the rising sulfide level has reached
the lower range of the photic zone. Slow-growing isolates
125
from this zone appear to be adapted to low light intensities
and are undistinguishable from Chlorobium phaeobacteroides O2
(N. Pfennig and H. Cypionka, personal communication).
Other organisms isolated from the interface are aerobic, neutro- 150
philic, obligately chemolithotrophic sulfide oxidizers, tentatively
A
placed into the genus Thiomicrospira. A number of new sulfide-
reducing and methanogenic bacteria were also isolated from the
sediment. 175
The Cariaco Trench off the Venezuelan coast is another
permanently anoxic basin of a depth of 1,400 m (Richards H2S
1975; Richards and Vaccaro 1958). Another site is the Orca
Basin in the northern Gulf of Mexico, which features an anoxic 200
and saline layer of bottom water at a depth of 2,300 m (Shokes 2 5 10 15 20
Hydrogen sulfide (in mM)
et al. 1977). Trüper (1969) and Watson and Waterbury (1969)
investigated anoxic hot brines (56 C) found at the bottom of the . Fig. 3.7
Red Sea and ascribed the absence of readily growing bacteria to Profiles of dissolved oxygen and hydrogen sulfide in the upper
toxic concentrations of heavy metals. A unique environment water column of the Black Sea, central western gyre. (A, data from
for the discovery of a number of new metabolic types of Sorokin 1972; B, data from Repeta et al. 1989)
Concentration (ng/I) Two kinds of special abilities of the prokaryotes are exploited
0 200 400 600 800 1000 by the eukaryote and have opened highly specialized habitats:
0 the ability to fix nitrogen and to hydrolyze cellulose.
Nitrogen fixation is exploited by plants. The bacterial
symbionts are either held as ectosymbionts in intercellular
20 spaces of leaves, stems, or at the root surface or as endosymbi-
Chlorophyll-a onts in root nodules and rhizothamnia. Animals do not take
much advantage of nitrogen-fixing bacteria; nitrogen fixation
has only been observed in the hindgut of termites (Benemann
40 1973; Breznak et al. 1973) and in the gut of humans who eat
a carbohydrate-rich diet (Bergensen and Hipsley 1970).
Depth (meters)
The ability to hydrolyze cellulose is lacking in animals that

are higher on the evolutionary scale than the mollusks, with
60
exception of the silverfish, Lepisma lineata. Symbiotic relation-
ships between animals and cellulolytic protozoa and bacteria
were established because of the abundance of cellulose as food
80 H2S interface and the general inability of animals to produce cellulolytic
enzymes.
Associations of eukaryotic hosts with bacteria have evolu-
tionary aspects. When the plants started to colonize the land and
100
Bacteriochlorophyll-e to shape the prerequisites for the evolution of the higher forms
of life, the bacteria had already acquired a high degree of bio-
chemical and physiological fitness—presumably the present-day
120 status. Instead of being added to a preexisting system of higher
organisms as, for example, plants and animals colonizing the
. Fig. 3.8
recently emerged island of Surtsey, the bacteria colonized the
Distribution of phytoplanktonic and bacterial chlorophylls in the
eukaryotic host during all stages of the host’s evolution.
upper water column of the Black Sea, central western gyre (Data
The present-day prokaryote–eukaryote relationships must be
from Repeta et al. 1989)
considered the result of a long selection process. Excluding
parasitism, one may hypothesize that in all cases of stable asso-
ciations, the relationship is of mutual advantage to the partners.
The benefit may be as obvious as in the cases of nitrogen fixation
prokaryotes, especially the anaerobically photosynthesizing in higher plants and of cellulose digestion in animals, or the
cyanobacteria, is the Solar Lake near Eilat in the Gulf of Aqaba, benefit may just be a protective mechanism of seemingly very
which has been described in a series of papers (see Cohen et al. little importance, such as preventing harmful bacteria from
1977). The relatively low concentration of methane in entering the potential habitat.
most of the anoxic marine waters as compared to anoxic
freshwaters is in agreement with the competitive behavior of
sulfate versus carbon dioxide reduction originally suggested by Animal Habitats
Cappenberg (1974a, b) and discussed in detail by Rudd and
Taylor (1979). Eliminating parasitism from this discussion, only some benign
relationships will be considered. In this context, the endosym-
biosis of protozoa and bacteria deserves mention. Progress in
Eukaryotes as Habitats for Bacteria this area is enormous and has been reviewed in the handbooks
edited by Schwemmler and Schenk (1980) and Schenk and
Eukaryotes present a multitude of habitats for bacteria. The Schwemmler (1983). The inner and outer surfaces of animals,
surfaces, cavities, crevices, and intercellular spaces open to the the intestinal tract, the skin, and several organs have to be
air, as well as the intestinal tracts, exudates, and excretory regarded as microbial ecosystems with specific populations of
substances, offer opportunities for the growth of many bacte- indigenous and nonindigenous bacteria.
ria. During evolution, more or less close associations or sym-
bioses between the eukaryotic organisms and bacteria have
developed. The symbiotic relationship between the host, the Intestinal Tract
larger partner, and the bacterium is either neutralistic, for
example, when bacteria feed on the waste products of the Among the associations of bacteria and animals, the intestinal
host, or mutualistic. Both of these relationships will be tract is the ecosystem with the highest population density. The
discussed here. number of bacterial cells within the intestinal tract may even
exceed the number of host cells. With respect to the human composition of the saliva, and the succession within the bacterial
gut, this ratio was dramatically described by Savage (1977a): flora (Russell and Melville 1978). The interaction of the oral
‘‘the normal human organism can be said to be composed of streptococci and the mucosal and enamel surfaces deserves spe-
over 1014 cells, of which only about 10 % are animal cells.’’ The cial emphasis; special methods have been developed to investi-
ratio differs from species to species through the animal gate the specificity of interactions (Rutter and Abbott 1978),
kingdom. which are partially due to the kind and properties of extracellu-
The intestinal tract is an open ecosystem that resembles lar polysaccharides (Ebisu et al. 1975; Germaine et al. 1974).
a tube receiving food at one end and releasing waste at the The intestinal tract contains climax populations of indige-
other. In omnivores and carnivores, food digestion is accom- nous as well as allochthonous bacteria, yeasts, and protozoa.
plished by the animal’s own intestinal digestive enzymes with- Hundreds of different bacteria have been isolated from the
out involvement of microbes. There are two modifications of intestinal tract of many animals. In many cases, it is difficult to
this ‘‘straight tube’’ model. Both types are found especially in decide whether a certain species is autochthonous or not and
herbivores, among the invertebrates as well as the vertebrates. which physicochemical niche it fills. Applying modern ecological
Due to the particular nature of the diet of herbivorous animals, theory (Alexander 1971) to the ecosystem of the gastrointestinal
which is rich in celluloses, hemicelluloses, and pectins but is tract of mammals (Savage 1977a, b), criteria for determining
low in protein, the tract has been adapted to accommodate autochthony of microorganisms isolated from the gastrointesti-
microorganisms. Extensions of the tract function as fermenta- nal tract have been developed. ‘‘Autochthonous gastrointestinal
tion vessels and harbor protozoa and bacteria able to convert microorganisms (1) can grow anaerobically, (2) are always found
the plant polymers into microbial cells and degradation prod- in normal adults, (3) colonize particular areas of the tract,
ucts. These fermentation vessels may be located either anteri- (4) colonize their habitats during succession in infant animals,
orly or posteriorly to the areas of the intestinal tract where (5) maintain stable population levels in climax communities in
the gut contents are digested by the enzymes of the animal. In normal adults, and (6) may associate intimately with the mucosal
the ruminants, a special compartment of the stomach, the epithelium in the area colonized’’ (Savage 1977a). These criteria
rumen–reticulum, is used as the fermentation vat. Its huge are useful for distinguishing indigenous microorganisms from
size guarantees a long residence time for the plant material to nonindigenous ones. For detailed information, the excellent
be degraded by cellulolytic and other microorganisms. Similar reviews on relevant problems and on the present status of
vessels destined for foregut fermentation are present in camels, knowledge by Savage (1977a, b), Clarke (1977), Bauchop
kangaroos, hippopotamuses, and leaf-eating apes. The second (1977), Costerton et al. (1981a, b, 1987), Costerton and Cheng
modification of the straight tube is represented by extended (1981), Bitton and Marshall (1980), Marshall (1984), and
compartments or blind sacs (ceca) near the end of the intesti- Tannock (1990) should be consulted.
nal tract. Their function is also that of a fermentation vessel; Because of its high acidity, the human stomach is not pop-
examples for hindgut fermentation are found among verte- ulated by bacteria. Behind the pylorus, fewer than 10 bacterial
brates (horses, pigs, guinea pigs, rats, rabbits) and insects cells per milliliter have been counted. From the pylorus via the
(termites, wood roaches). duodenum and jejunum to the ileum, the number of bacteria
increases; up to 1011 bacteria per gram of feces have been
counted; Escherichia coli represents less than 1 % of the popula-
Human Intestinal Tract tion. The predominant genus is Bacteroides, comprising species
such as B. fragilis. The Bacteroides are followed by
The tract comprises six major areas: mouth, esophagus, stom- Fusobacterium, Eubacterium, and Peptostreptococcus (Moore
ach, small intestine, cecum, and large intestine. Each may be the and Holdeman 1974), all strictly anaerobic, Gram-negative
habitat of a particular bacterial flora. And these habitats may rods. Other inhabitants belong to the streptococci, lactobacilli,
be subdivided into further categories, such as epithelial and bifidobacteria (Drasar and Hill 1974; Drasar and Barrow
(the bacteria grow in association with epithelial surfaces), 1985). The dependence of the composition of the flora on the
lumenal (bacteria live free in the lumen), and crystal (bacteria kind of food eaten has not yet been exhaustively studied. The
live in crypts). Different areas harbor different bacteria, and each changes caused by the oral application of antibacterial agents
may be the primary habitat for a certain bacterium or a group of like antibiotics and sulfonamides or by special foods such as
bacteria. The habitats are more or less delineated and offer rather garlic or cabbage have not been examined with respect to qual-
constant conditions. itative and quantitative changes of the bacterial flora.
As was recognized very early (see Miller 1890), the mouth, The specific niches of the bacteria in the gastrointestinal tract
nasopharynx, and throat are inhabited by many aerobic and are not easy to comprehend. The nutritional conditions are
anaerobic bacteria. Streptococcus salivarius, S. mutans, Veillonella certainly optimal for many other bacteria which do not occur
alcalescens, Treponema dentium, Fusobacterium, Actinomyces, there. It is not known whether one or the other special property
lactobacilli, corynebacteria, and cocci belong to the normal of intestinal bacteria as discussed by Prins (1977) is responsible
flora of the human mouth (Hardie and Bowden 1974). for a true autochthony or whether their common occurrence in
The mouth may be considered as an ecosystem of its own char- this habitat is correlated with specific chemical or structural
acterized by its anatomical and physiological development, the properties.
One could imagine that for lumenal bacteria, high growth The rumen may be considered to be a semicontinuous culture
rates are required and that bacterial attachment sites and mech- of microorganisms (Hungate 1975).
anisms favor the colonization of epithelia. The selection of only Between 1010 and 1011 bacteria and 105 and 106 protozoa
a few species of bacteria to grow in the gastrointestinal habitats inhabit each milliliter of rumen contents. Most of the bacteria
may, among other factors, be due to the tolerance to pH, to are strict anaerobes and are especially adapted to that habitat.
lipolytic, peptolytic, and saccharolytic enzymes, to detergent- The oxidation–reduction potential amounts to 0.35 V. Because
like bile acids, and to degradation products. Whether certain many rumen bacteria are instantly killed by oxygen,
structural details of the bacterial cell envelope are common to all special techniques for manipulating nutrient media, inocula,
indigenous inhabitants of the lower intestine is an open question and cultures in the absence of oxygen had to be
(Costerton et al. 1974, 1981a; Martin 1969). developed (Holdemann et al. 1977; Hungate 1950). The nitrogen
Basically, in many intestinal tracts, the presence of source of most rumen bacteria is ammonia. Many
microorganisms is not necessary. Many animals can grow as rumen bacteria require a carbon dioxide-rich (10 % CO2) atmo-
gnotobiotes without microorganisms (Coates and Fuller 1977). sphere (Dehority 1971). Ruminococcus albus, R. flavefaciens,
Studies with such animals have shown that the intestinal micro- Bacteroides succinogenes, Butyrivibrio fibrisolvens, Eubacterium
flora confers a kind of resistance to intruding pathogens. cellulosolvens, and Clostridium lochheadii are the primary species
While Vibrio cholerae and Shigella dysenteriae colonize the of cellulose-digesting bacteria of the rumen. The cellobiose and
intestinal tract of germ-free rats easily, they do not colonize glucose produced from cellulose are fermented by a variety of
the normal gut easily. Stability of the microbial population is bacteria which utilize the pectins, starch, fructosans, proteins,
a major factor in the gastrointestinal ecosystem. and lipids as well. Fermentative degradation of these
The use of gnotobiotic animals will help to explore neutral- compounds leads to the accumulation of fatty acids, carbon
istic, mutualistic, and parasitic relationships of bacteria and dioxide, and hydrogen. The latter two gases are combined
their host. However, even simple quantitative and qualitative by Methanobacterium ruminantium to give methane.
studies on the populations in intestinal habitats are urgently Besides the bacteria mentioned above, there are many
needed. It was not long ago that Escherichia coli was considered others able to ferment noncellulose carbohydrates or the
the dominant bacterium in the human intestinal tract. With the products of carbohydrate fermentation. These bacteria can
application to the human gut of methods originally developed only be mentioned here: Bacteroides amylophilus, B. ruminicola,
for studying the ruminant bacteria, it was found that the Succinimonas amylolytica, Selenomonas ruminantium, Strepto-
number of strict anaerobes exceeded the oxygen-tolerant coccus bovis, Veillonella alcalescens, Lachnospira multiparus,
anaerobes such as E. coli by a thousandfold. The routine Peptostreptococcus elsdenii, and Desulfotomaculum ruminis.
application of anaerobic techniques to other vertebrate or In addition to these bacteria, which have been grown in pure
invertebrate intestinal tracts will certainly result in similar culture, there are some that have been identified microscopically
unexpected relationships. only. Quin’s oval and Eadie’s oval have been grown in mixed
culture (Orpin 1972, 1973). The filamentous bacterium
Oscillospira has been found mainly in the rumen of sheep.
Rumen and Reticulum Lampropedia is an obligatory aerobe; however, it is found in
high numbers in the sheep rumen.
The best-studied ecosystem of foregut fermentation and most The rumen microflora is in a delicate equilibrium. If,
probably of all anaerobic ecosystems is the rumen. Rumen for example, rumen methanogenesis is eliminated by the
symbiosis is an excellent example of a highly developed ecosys- addition of chloroform to the rumen contents, a series of
tem (Hobson 1988). The principles derived from these studies changes occur: gaseous hydrogen accumulates immediately,
have had profound influence on the study of microbial ecology and the increased partial pressure of hydrogen results in
(Hungate 1966, 1975; Savage 1977a, b). an increase of the ratio of propionic to acetic acid. These
The main sources of carbohydrates for ruminants are hay, observations indicate that interspecies hydrogen transfer plays
straw, and grass. About 50 % of dried grass consists of fructosans an important role in the rumen (Iannotti et al. 1973). Hydrogen
and xylanes, and the rest of the carbohydrate fraction is cellulose. is an intermediate in substrate conversions in the rumen
The alimentary tract of ruminants is adapted to this special diet. (Hungate 1967, 1975).
The first two parts of the bovine stomach, the rumen and the
reticulum, serve as a microbial digestive vessel; the total volume
is 100–250 l. This ruminoreticulum provides ideal conditions Hindgut Fermentation
for the growth of many microorganisms. The temperature is
constant at 37–39 C; 100–200 l of saliva are secreted per day; it Although it is not as efficient for the animal as the rumen-type
contains phosphate (10–50 mM), bicarbonate (100–140 mM), fermentation, hindgut fermentation is widely distributed among
and urea (10 mM nitrogen) and is a well-buffered (pH 5.8–7.3) herbivorous animals, vertebrates as well as insects. The fermen-
solution. Nutrients are periodically added in the form of tative organisms and their function in the blind sacs (ceca) and
well-macerated cellulose-containing fodder; the mixture is the large intestine of animals have been less well studied than
mechanically agitated by the contractions of the rumen. those involved in rumen symbiosis (McBee 1977). The majority
of studies on hindgut fermentations were limited to anatomical– procedures for isolation, cultivation, and purification become
morphological descriptions, and preserved specimens of insects less critical. The disadvantage is that specifically expressed met-
have been used to recognize or describe microorganisms. Recent abolic activities of the phylogenetically identified organisms or
investigations show that the isolation and cultivation of bacteria their ecological functions in situ or in vitro are not readily
so far only recognized in cross sections of cecal mucosa epithelia apparent from the data. After Torsvick and Goksoyr (1978)
is possible, although with difficulties (Lee and Phillips 1978). pioneered a method for DNA extraction from whole natural
The comprehensive investigations of Buchner (1953) provide an populations, hybridization probes were applied to the study of
excellent basis and incentive for studies on the axenic culture of the microbial population in the rumen (Stahl et al. 1988), in soil
protozoa and bacteria present in the insect gut and its appendi- (Holben et al. 1988), and of communities of marine planktonic
ces. Investigations on the protozoal and bacterial symbionts microorganisms (Giovannoni et al. 1990), of microbial hot
probably involved in the degradation of plant polymers or springs (Stahl et al. 1985), and of symbiotic prokaryotes (Distel
having other important functions are urgently needed et al. 1988; Unterman et al. 1989). Much of the present work
(Eutick et al. 1978; see Foglesong et al. 1975; McBee 1977). continues to emphasize methodology (Somerville et al. 1989;
Investigations on the isolation and characterization of Weller and Ward 1989), especially the development of gene
heterotrophic bacteria from hindguts of the wood-eating probes for individual species or genera and other methods to
termite, Reticulitermes flavipes, revealed the predominance of assess detection and abundance of microbes at specific habitats
streptococci and Bacteroides species and indicate the existence or within natural populations.
of a unidirectional food chain from glucose via lactate to propi-
onate and acetate (Schultz and Breznak 1978, 1979).
Outlook
Bacteria of the Human Skin Although the habitat is not specified in the description of
a species, it is a major characteristic of a species. Microorgan-
The human skin is a rather homogeneous habitat with respect to isms have their habitats in various ecosystems and microenvi-
temperature. However, moisture conditions vary and create ronments. Unfortunately, a discussion of the environments of
distinctive habitats for a characteristic flora. The available microorganisms can never be complete. The purpose of the
water is the most important environmental factor influencing present chapter was to demonstrate the multiplicity of micro-
the size of cutaneous populations. The occlusion of the relatively bial environments with emphasis on the close relationship to
dry skin of the forearm results in a rapid increase of the the diversity of prokaryotes, their metabolic peculiarities,
bacterial population by a factor of 104 (from an initial colony potential capabilities, and constraints. Obviously, more exam-
count of 3 103 to 3.8 107 cells/cm2) by the fourth day of ples could have been selected, and more will be found in the
occlusion. The high relative humidity of the axilla results in the literature every month. The discoveries of new morphological
survival of dense populations in this region (Marples 1965, 1974, and metabolic types of prokaryotes, as well as of unsuspected
1976; Noble and Somerville 1974; Woodroffe and Shaw 1974). relationships and interactions between microorganisms and
Many common bacteria, such as the corynebacteria (‘‘diphthe- their environments, over the last decades can be related, in
roids’’), mycobacteria, micrococci, and streptococci, in addition most cases, to the study of microbial ecosystems which had
to yeasts and other fungi, are inhabitants of the human skin. received no particular attention before. The rumen and its
Nutrients are provided by sweat. Fatty acids may act as inhabitants have been discussed as one of the most striking
antibacterial agents and effect a counterselection. The bacteria examples of this; the illuminated hydrogen sulfide–oxygen
inhabiting the sweat glands (sebaceous glands) and hair follicles interfaces of lakes and lagoons are others. Highly developed
cannot be reached by normal cleaning of the skin and guarantee symbiotic relationships between bacteria and metazoa in the
fast repopulation of the skin after cleaning (Noble and Pitcher luminous organs and in the hindguts of mammals and insects
1979; Rosebury 1972). as well as between bacteria and protozoa have only been
marginally mentioned. The ecosystems consisting of solid
matter, such as the various kinds of fertile soils, virgin volcanic
Molecular Microbial Ecology soils, and desert soils as well as the rhizospheres of plants, have
not been dealt with at all in this chapter.
Although the sequencing and hybridization of ribosomal RNAs New artificial microbial ecosystems are arising in increas-
primarily concern the phylogenetic characteristics of organisms ing numbers in, for instance, the cooling systems of nuclear
(see > Chap. 1, ‘‘How We Do, Don’t and Should Look at Bacte- power plants and the aquatic and terrestrial dumping sites of
ria and Bacteriology’’ in Vol. 1 and Woese 1987), first results new, chemically synthesized organic compounds, including
show that a search for uses of this technique in microbial ecology herbicides, pesticides, and widely used organic solvents.
is well justified (Pace et al. 1986; Giovannoni et al. 1988; DeLong Special evolutionary pressure is being exerted on prokaryotes
et al. 1988). The advantage of identifying microorganisms by the ever-changing fungicidal and bactericidal environments
within complex natural populations by the use of their individ- of hospitals. Microbial populations will also be the first to
ual ribosomal RNA signatures is that the often difficult respond to global environmental changes such as those
following the thinning of the ozone layer and the greenhouse Baumann P, Baumann L (1977) Biology of the marine enterobacteria: genera
Beneckea and Photobacterium. Ann Rev Microbiol 31:39–61
effect. One may also be realistic enough to include here the
Baumann L, Baumann P, Mandel M, Allen RD (1972) Taxonomy of aerobic
exposure of this planet’s microorganisms to extraterrestrial marine bacteria. J Bacteriol 110:402–429
conditions some of which, for example, certain layers in the Bavendamm W (1924) Die farblosen und roten Schwefelbakterien des Süss-und
atmosphere of Venus, may indeed sustain life of some known Salzwassers Pflanzenforschung 2:1–156
chemolithotrophic bacteria. In any case, it can be expected Bayley ST, Morton RA (1978) Recent developments in the molecular biology of
extremely halophilic bacteria. CRC Crit Rev Microbiol 6:151–205
that studies of prokaryotes in relation to their natural envi-
Bayley ST, Morton RA (1979) Biochemical evolution of halobacteria. In: Shilo M
ronments will remain a most exciting and productive area of (ed) Strategies of microbial life in extreme environments. Verlag Chemie,
microbiology. Weinheim, pp 109–124
Bazylinski DA, Frankel RB, Jannasch HW (1988) Anaerobic magnetite
production by a marine, magnetotactic bacterium. Nature 334:518–519
Beijerinck MW (1895) Über Spirillum desulfuricans als Ursache von
References Sulfatreduktion Zentralblatt für Bakteriologie, Parasitenkunde Infektion-
skrankheiten und Hygiene, Abt. 2, 1:1–9
Aaronson S (1970) Experimental microbial ecology. Academic, New York Beijerinck MW (1921–1940) Verzammelde Geschriften Nijhoff, Den Haag, pp 1–6
Acher AJ, Juven BJ (1977) Destruction of coliforms in water and Belkin S, Wirsen CO, Jannasch HW (1986a) A new sulfur-reducing, extremely
sewage by dye-sensitized photooxidation. Appl Environ Microbiol thermophilic eubacterium from a submarine thermal vent. Appl Environ
33:1019–1022 Microbiol 51:1180–1185
Adler J (1974) Chemoreception and chemotaxis in bacteria. In: Jaenicke L (ed) Belkin S, Nelson DC, Jannasch HW (1986b) Symbiotic assimilation of CO2 in
Biochemistry of sensory functions. Springer, Berlin two hydrothermal vent animals, the mussel Bathymodiolus thermophilus
Adler J (1988) Chemotaxis: old and new. Botanica Acta 101:93–100 and the tube worm Riftia pachyptila. Biol Bull 170:110–121
Ahrens R, Moll G, Rheinheimer G (1968) Die Rolle der Fimbrien bei der Benemann JR (1973) Nitrogen fixation in termites. Science 181:164–165
eigenartigen Sternbildung von Agrobacterium luteum. Archiv für Bennett AF (1978) Activity metabolism of the lower vertebrates. Ann Rev Physiol
Mikrobiologie 63:321–330 40:447–469
Akin DE (1976) Ultrastructure of rumen bacterial attachment to forage cell walls. Berg B, van Hofsten B, Pettersson G (1972) Electron microscopic observations on
Appl Environ Microbiol 31:562–568 the degradation of cellulose fibers by Cellvibrio fulvus and Sporocytophaga
Akin DE, Amos WE (1975) Rumen bacterial degradation of forage cell walls myxococcoides. J Appl Bacteriol 35:215–219
investigated by electron microscopy. Appl Microbiol 29:692–701 Bergensen FJ, Hipsley EH (1970) The presence of N2-fixing bacteria in the
Alexander M (1971) Microbial ecology. Wiley, New YorK intestine of man and animals. J Gen Microbiol 60:61–65
Alexander M (1976) Natural selection and the ecology of microbial adaption Berkeley RCW, Lynch JM, Melling J, Rutter PR, Vincent B (1980) Microbial
in a biosphere. In: Heinrich MR (ed) Extreme environments. Mechanisms adhesion to surfaces. Ellis Horwood, Chichester
of microbial adaption. Academic, New York, pp 3–25 Berndt H, Ostwal K-P, Lalucat J, Schumann CH, Mayer F, Schlegel HG (1976)
Alexander M (1977) Introduction to soil microbiology, 2nd edn. Wiley, Identification and physiological characterization of the nitrogen fixing
New York bacterium Corynebacterium autotrophicum GZ 29. Arch Microbiol
Anwar M, Khan TH, Prebble J, Zagalski PF (1977) Membrane-bound carotenoid 108:17–26
in Micrococcus luteus protects naphthoquinone from photodynamic action. Bezdek HG, Carlucci AF (1972) Surface concentration of marine bacteria. Limnol
Nature 270:538–540 Oceanogr 17:566–569
Aragno M (1978) Enrichment, isolation and preliminary characterization of Bhuiya ZH, Walker N (1977) Autotrophic nitrifying bacteria in acid tea soils from
a thermophilic, endospore-forming hydrogen bacterium. FEMS Microbiol Bangladesh and Sri Lanka. J Appl Bacteriol 42:253–257
Lett 3:13–15 Biebl H, Pfennig N (1978) Growth yield of green sulfur bacteria in mixed cultures
Atlas RM, Bartha R (1987) Microbial ecology: fundamentals and applications, with sulfur and sulfate reducing bacteria. Arch Microbiol 117:9–16
2nd edn. CA The Benjamin/Cummings, Menlo-Park Bitton G, Marshall KC (1980) Adsorption of microorganisms to surfaces. Wiley,
Babenzien H-D (1965) Über Vorkommen und Kultur von Nevskia ramosa New York
Zentralblatt für Bakteriologie, Parasitenkunde, Infektionskrankheiten und Blakemore RP (1975) Magnetotactic bacteria. Science 190:377–379
Hygiene, Abt. 1, Suppl 1:111–116 Blakemore RP, Maratea D, Wolfe RS (1979) Isolation and pure culture of
Babenzien H-D (1967) Zur Biologie von Nevskia ramosa. Zeitschrift für a freshwater magnetic spirillum in chemically defined medium. J Bacteriol
Allgemeine Mikrobiologie 7:89–96 140:720–729
Ballard RD (1977) Notes on a major oceanographic find. Oceanus 20:35–44 Blakemore RP, Frankel RB, Kalmijn AJ (1980) South-seeking magnetotactic
Barber RT (1968) Dissolved organic carbon from deep waters resists microbial bacteria in the Southern Hemisphere. Nature 286:384–385
oxidation. Nature 220:274–275 Bland JA, Staley JT (1978) Observations on the biology of Thiothrix. Arch
Baross JA, Deming JW (1983) Growth of black smoker bacteria at temperatures of Microbiol 117:79–87
at least 250 C. Nature 303:423–426 Blumershine RV, Savage DC (1978) Filamentous microbes indigenous to the
Baross JA, Morita RY (1978) Microbial life at low temperatures: ecological aspects. murine small bowel: a scanning electron microscopic study of their mor-
In: Kushner DJ (ed) Microbial life in extreme environments. Academic, phology and attachment to the epithelium. Microb Ecol 4:95–103
London, pp 9–71 Bokor R (1933) Die Mikrobiologie der Szik-(Salz-oder Alkali-) Böden mit
Bartha R, Atlas RM (1977) The microbiology of aquatic oil spills. Adv Appl besonderer Berücksichtigung ihrer Fruchtbarmachung. In: Fehäer D (ed)
Microbiol 22:225–266 Untersuchungen über die Mikrobiologie des Waldbodens. Springer, Berlin,
Bartnicki-Garcia S, Nickerson WJ (1962) Nutrition, growth and morphogenesis pp 221–258
of Mucor rouxii. J Bacteriol 84:841–858 Bothe H, De Bruijn FJ, Newton WE (eds) (1988) Nitrogen fixation: hundred years
Bauchop T (1971) Mechanism of hydrogen formation in Tritrichomonas foetus. after. Gustav-Fischer-Verlag, Stuttgart
J Gen Microbiol 68:27–33 Bousfield IJ, MacKenzie AR (1976) Inactivation of bacteria by freeze-drying. Soc
Bauchop T (1977) Foregut fermentation. In: Clarke RTJ, Bauchop T (eds) Appl Bacteriol Symp Ser 5:329–344
Microbial ecology of the gut. Academic, London, pp 223–310 Boyd SA, Shelton DR, Berry D, Tiedje JM (1983) Anaerobic biodegradation
Bauld J, Brock TD (1973) Ecological studies of Chloroflexus, a gliding photosyn- of phenolic compounds in digested sludge. Appl Environ Microbiol
thetic bacterium. Archiv für Mikrobiologie 92:267–284 46:50–54
Boyer EV, Ingle MB, Merver GD (1973) Bacillus alcalophilus subsp. halodurans Inhibition experiments. Antonie van Leeuwenhoek J Microbiol Serol
subsp. nov.: an alkaline-amylase-producing, alkalophilic organism. Int J Syst 40:297–306
Bacteriol 23:238–242 Castenholz RW (1969) Thermophilic blue-green algae and the thermal environ-
Boylen CW (1973) Survival of Arthrobacter crystallopoietes during prolonged ment. Bacteriol Rev 33:476–504
periods of extreme desiccation. J Bacteriol 113:33–37 Castenholz RW (1976) The effect of sulfide on the blue-green algae of hot springs.
Breznak JA, Brill WJ, Mertins JW, Coppel HC (1973) Nitrogen fixation in I. New Zealand and Iceland. J Phycol 12:54–68
termites. Nature 244:577–579 Castenholz RW (1977) The effect of sulfide on the blue-green algae of hot springs.
Brierley CL (1977) Thermophilic microorganisms in extraction of metals from II. Yellowstone National Park. Microb Ecol 3:79–105
ores. In: Underkofler LA (ed) Developments in industrial microbiology. Castenholz RW (1979) Evolution and ecology of thermophilic microorganisms.
Proceedings of the 33rd general meeting of the society for industrial micro- In: Shilo M (ed) Strategies of microbial life in extreme environments. Verlag
biology, vol 18. Washington American Institute of Biological Sciences, Chemie, Weinheim, pp 373–392
Washington, DC, pp 273–284 Cavanaugh CM, Gardiner SL, Jones MLS, Jannasch HW, Waterbury JB
Brierley CL (1978a) Bacterial leaching. CRC Crit Rev Microbiol 6:207–262 (1981) Prokaryotic cells in the hydrothermal vent tube worm. Riftia
Brierley JA (1978b) Thermophilic iron-oxidizing bacteria found in copper pachyptila: possible chemoautotrophic symbionts. Science 213:340–342
leaching dumps. Appl Environ Microbiol 36:523–525 Characklis WG, Marshall KC (1990) Biofilms. Wiley Interscience, New York
Brierley CL, Brierley JA (1973) A chemoautotrophic and thermophilic microor- Chen M, Wolin MJ (1977) Influence of CH4 production by Methanobacterium
ganism isolated from an acid hot spring. Can J Microbiol 19:183–188 ruminantium on the fermentation of glucose and lactate by Selenomonas
Brierley JA, Lockwood SJ (1977) The occurrence of thermophilic iron-oxidizing ruminantium. Appl Environ Microbiol 34:756–759
bacteria in a copper leaching system. FEMS Microbiol Lett 2:163–165 Chet I, Mitchell R (1976) Ecological aspects of microbial chemotactic behavior.
Briston J, Courtois D, Denis F (1974) Microbiological study of a hypersaline lake Ann Rev Microbiol 30:221–239
in French Somaliland. Appl Microbiol 27:819–822 Childress JJ, Fisher CR, Brook JM, Kennicutt MC II, Bridigare R, Anderson AE
Brock TD (1967) Life at high temperatures. Science 158:1012–1019 (1986) A methanotrophic marine molluscan (Bivalvia, Mytilidae) symbiosis:
Brock TD (1969) Microbial growth under extreme conditions. Soc Gen Microbiol mussels fueled by gas. Science 233:1306–1308
Symp 19:15–41 Chislett ME, Kushner DJ (1961) A strain of Bacillus circulans capable of growing
Brock TD (1970) High temperature systems. Ann Rev Ecol Syst 1:191–220 under highly alkaline conditions. J Gen Microbiol 24:187–190
Brock TD (1978) Thermophilic microorganisms and life at high temperatures. Clark FE (1967) Bacteria in soil. In: Burges A, Raw F (eds) Soil biology. Academic,
Springer, New York London, pp 15–49
Brock TD (1979) Ecology of saline lakes. In: Shilo M (ed) Strategies of microbial Clark AE, Walsby AE (1978a) The occurrence of gas-vacuolate bacteria in lakes.
life in extreme environments. Verlag Chemie, Weinheim, pp 39–47 Arch Microbiol 118:223–228
Brock TD (ed) (1986) Thermophiles: general, molecular and applied microbiology. Clark AE, Walsby AE (1978b) The development and vertical distribution of
Wiley, New York populations of gas-vacuolate bacteria in a eutrophic, monomictic lake.
Brock TD (1987) The study of microorganisms in situ: progress and problems. In: Arch Microbiol 118:229–233
Fletscher M, Gray TRG, Jones JG (eds) Ecology of microbial communities. Clarke RTJ (1977) The gut and its micro-organisms. In: Clarke RTJ, Bauchop T
Cambridge University Press, Cambridge, pp 1–17 (eds) Microbial ecology of the gut. Academic, London, pp 35–71
Brock TD, Boylen KL (1973) Presence of thermophilic bacteria in laundry and Coates ME, Fuller R (1977) The gnotobiotic animal in the study of gut microbi-
domestic hot-water heaters. Appl Microbiol 25:72–76 ology. In: Clark RTJ, Bauchop T (eds) Microbial ecology of the gut.
Brock TD, Freeze H (1969) Thermus aquaticus gen. n. and sp. n., a non- Academic, London, pp 311–346
sporulating extreme thermophile. J Bacteriol 98:289–297 Cohen Y, Rosenberg E (1989) Microbial mats: physiological ecology of benthic
Brock TD, Brock ML, Bott TL, Edwards MR (1971) Microbial life at 90 C: The microbial communities. American Society for Microbiology, Washington, DC
sulfur bacteria of Boulder Spring. J Bacteriol 107:303–314 Cohen Y, Padan E, Shilo M (1975) Facultative anoxygenic photosynthesis in the
Brock TD, Brock KM, Belly RT, Weiss RL (1972) Sulfolobus: a new genus of cyanobacterium Oscillatoria limnetica. J Bacteriol 123:855–861
sulfur-oxidizing bacteria living at low pH and high temperature. Archiv für Cohen Y, Krumbein WE, Goldberg M, Shilo M (1977) Solar lake (Sinai). I.
Mikrobiologie 84:54–68 Physical and chemical limnology. Limnol Oceanogr 22:597–608
Brown AD (1978) Microbial water stress. Bacteriol Rev 40:803–846 Cohen-Bazire G, Stainer RY (1958) Inhibition of carotenoid synthesis in photo-
Bryant MP, Wolin EA, Wolin MJ, Wolfe RS (1967) Methanobacillus omelianskii, synthetic bacteria. Nature 181:250–252
a symbiotic association of two species of bacteria. Archiv für Mikrobiologie Cohen-Bazire G, Kunisawa R, Pfennig N (1969) Comparative study of the struc-
59:20–31 ture of gas vacuoles. J Bacteriol 100:1049–1061
Bryant MP, Campbell LL, Reddy CA, Crabill MR (1977) Growth of Desulfovibrio Cohn F (1881) Gutachten über die Abwässer verschiedener Zuckerfabriken im
in lactate or ethanol media low in sulfate in association with H2-utilizing Winter 1881 Quoted from Kolkwitz 1906
methanogenic bacteria. Appl Environ Microbiol 33:1162–1169 Colmer AR, Temple KL, Hinkle ME (1950) An iron-oxidizing bacterium from the
Buchner P (1953) Endosymbiose der Tiere mit pflanzlichen Mikroorganismen. drainage of some bituminous coal mines. J Bacteriol 59:317–328
Birkhäuser Verlag, Basel Cornax R, Morinigo MA, Romero P, Borrego JJ (1990) Survival of pathogenic
Buder J (1919) Zur Biologie des Bacteriopurpurins und der Purpurbakterien. microorganisms in seawater. Curr Microbiology 220:293–298
Jahrbücher der Wissenschaftlichen Botanik 58:525–628 Costerton JW, Cheng KJ (1981) The Bacterial glycocalyx in nature and disease.
Burggraf S, Jannasch HW, Nicolaus B, Stetter KO (1990) Archaeoglobus profundus Ann Rev Microbiol 35:299–324
sp. nov., represents a new species of within the sulfate-reducing Costerton JW, Ingram JM, Cheng K-J (1974) Structure and function of the cell
archaebacteria. Syst Appl Microbiol 10:24–28 envelope of Gram-negative bacteria. Bacteriol Rev 38:87–110
Butlin KR, Postgate JR (1954) The microbiological formation of sulphur in Costerton JW, Geesey GG, Cheng K-J (1978) How bacteria stick. Sci Am 238:86–95
Cyrenaican lakes. In: Cloudsley-Thompson JL (ed) Biology of deserts. Insti- Costerton JW, Irvin RT, Cheng KJ (1981) The role of bacterial surface structures
tute of Biology, London, pp 112–122 in pathogenesis. Crit Rev Microbiol 8:303–338
Cagle GD (1975) Fine structure and distribution of extracellular polymer sur- Costerton JW, Marrie TJ, Cheng KJ (1985) Phenomena of bacterial adhesion. In:
rounding selected aerobic bacteria. Can J Microbiol 21:395–408 Savage DC, Fletcher M (eds) Bacterium adhesion. Plenum, New York, pp 3–43
Cappenberg TH-E (1974a) Interrelations between sulfate-reducing and methane- Costerton JW, Cheng K-J, Geesey GG, Ladd TI, Nickel JC, Dasgupta M, Marrie TJ
producing bacteria in bottom deposits of a fresh-water lake I. Field of (1987) Bacterial biofilms in nature and disease. Ann Rev Microbiol 41:435–464
observations. Antonie van Leeuwenhoek J Microbiol Serol 40:285–295 Cross T (1968) Thermophilic actinomycetes. J Appl Bacteriol 31:36–53
Cappenberg TH-E (1974b) Interrelations between sulfate-reducing and Csonka LN (1989) Physiological and genetic responses of bacteria to osmotic
methane-producing bacteria in bottom deposits of a fresh-water lake. II. stress. Microbiol Rev 53:121–147
Cundell AM, Sleeter TD, Mitchell R (1977) Microbial populations associated with thermophilic submarine heterotrophic archaebacteria growing up to 98 C.
the surface of the brown alga Ascophyllum nodosum. Microb Ecol 4:81–91 Syst Appl Microbiol 8:106–113
Dazzo FB, Yanke WE, Brill WJ (1978) Trifoliin: a RHIZOBIUM recognition Fisher CR, Childress JJ, Minnich E (1989) Autotrophic carbon fixation
protein from white clover. Biochim Biophys Acta 539:276–286 by the chemoautotrophic symbionts of Riftia pachyptila. Biol Bull
De Bont JAM, Mulder EG (1974) Nitrogen fixation and co-oxidization of ethyl- 177:372–385
ene by a methane-utilizing bacterium. J Gen Microbiol 83:113–121 Fletcher M, Loeb GI (1979) Influence of substratum characteristics on the
Degens ET, Ross DA (1974) The Black Sea: geology, chemistry and biology attachment of a marine pseudomonad to solid surfaces. Appl Environ
Memoir 20. American Association of Petroleum, Tulsa Microbiol 37:67–72
Dehority BA (1971) Carbon dioxide requirement of various species of rumen Fletcher M, Marshall KC (1982) Are solid surfaces of ecological significance to
bacteria. J Bacteriol 105:70–76 aquatic bacteria? In: Marshall KC (ed) Advances in microbial ecology. Ple-
DeLong EF, Wickham GS, Pace NR (1988) Phylogenetic stains: ribosomal num, New York
RNA-based probes for the identification of single cells. Science Fliermans CB, Brock TD (1972) Ecology of sulfur-oxidizing bacteria in hot acid
243:1360–1363 soils. J Bacteriol 111:343–350
Deming JW, Colwell RR (1982) Barophilic bacteria associated with digestive Focht DD, Verstraete W (1977) Biochemical ecology of nitrification and denitri-
tracts of abyssal holothurians. Appl Environ Microbiol 44:1222–1230 fication. Adv Microb Ecol 1:135–214
Deming JW, Tabor PS, Colwell RR (1981) Barophilic growth from intestinal tracts Foglesong MA, Walker DH Jr, Puffer JS, Markovetz AJ (1975) Ultrastructural
of deep-sea invertebrates. Microb Ecol 7:85–94 morphology of some prokaryotic microorganisms associated with the hind-
Demoll R, Liebmann H (1952) Über die Verteilung von Sphaerotilus natans im gut of cockroaches. J Bacteriol 123:336–345
Fluss. Schweizerische Zeitschrift für Hydrologie 14:289–297 Foster JW (1949) Chemical activities of fungi. Academic, New York
Dietz AS, Yayanos AA (1978) Silica gel for isolating and studying bacteria under Frankel RB, Blakemore RP, Wolfe RS (1979) Magnetite in freshwater
hydrostatic pressure. Appl Environ Microbiol 36:966–968 magnetotactic bacteria. Science 203:1355–1356
Distel DL, Lane DJ, Olsen GJ, Giovannoni SJ, Pace B, Pace NR, Stahl DA, Fridovich I (1974) Superoxide dismutases. Adv Enzymol 41:35–97
Felbeck H (1988) Sulfur-oxidizing bacterial endosymbionts: analysis of Fridovich I (1975) Oxygen: boon and bane. Am Sci 63:54–59
phylogeny and specificity by 16S rRNA sequences. J Bacteriol 170:2506–2510 Fridovich I (1976) Oxygen radicals, hydrogen peroxide, and oxygen toxicity.
Dondero NC (1961) Sphaerotilus, its nature and economic significance. Adv Appl In: Pryor WA (ed) Free radicals in biology, vol 1. Academic, New York,
Microbiol 3:77–107 pp 239–277
Dondero NC (1975) The Sphaerotilus-Leptothrix group. Ann Rev Microbiol Gerber NN (1975) Prodigiosin-like pigments. CRC Crit Rev Microbiol 3:469–485
29:407–465 Germaine GR, Chludzinski AM, Schachtele CF (1974) Streptococcus mutans
Drasar BS, Barrow PA (1985) Intestinal microbiology (Aspects of Microbiology 10). dextransucrase: requirement for primer dextran. J Bacteriol 120:287–294
van Nostrand Reinholf, England Gillespy TG, Thorpe RH (1968) Occurrence and significance of thermophiles in
Drasar BS, Hill MJ (1974) Human intestinal flora. Academic, London canned foods. J Appl Bacteriol 31:59–65
Duckworth RB (1975) Water relations in foods. Proceedings of an international Giovannoni SJ, DeLong EF, Olsen GJ, Pace NR (1988) Phylogenetic group-
symposium in Glasgow, Sept 1974. Academic, London specific oligodeoxynucleotide probes for identification of single microbial
Duda VI, Makaer’eva DE (1977) Morphogenesis and function of gas caps on cells. J Bacteriol 170:720–726
spores of anaerobic bacteria of the genus Clostridium. Mikrobiologiya Giovannoni SJ, Britschgi TB, Moyer CL, Field KG (1990) Genetic diversity in
46:689–694, In Russian, with English summary Sargasso Sea bacterioplankton. Nature 345(6270):60–63
Dugan PR, MacMillan CB, Pfister RM (1970) Aerobic heterotrophic bacteria Golovacheva RS (1976) Thermophilic nitrifying bacteria from hot springs.
indigenous to pH 2.8 acid mine water: microscopic examination of acid Mikrobiologiya 45:377–379, In Russian, with English summary
streamers. J Bacteriol 101:973–981 Golovacheva RS (1979) Attachment of Sulfobacillus thermosulfidooxidans cells to
Dundas IED (1977) Physiology of Halobacteriaceae. Adv Microb Physiol the surface of sulfide minerals. Mikrobiologiya 48:528–533, In Russian, with
15:85–120 English summary
Dundas ID, Larsen H (1962) The physiological role of the carotenoid pigments of Golovacheva RS, Karavaiko GI (1978) Sulfobacillus, a new genus of thermophilic
Halobacterium salinarium. Archiv für Mikrobiologie 44:233–239 sporeforming bacteria. Mikrobiologiya 47:815–822, In Russian, with English
Ebisu S, Kato K, Kotani S, Misaki A (1975) Structural differences in fructans summary
elaborated by Streptococcus mutans and S. salivarius. J Biochem 78:879–887 Gorini L (1960) Antagonism between substrate and repressor in controlling the
Ellwood DC, Hedger JN, Latham MH, Lynch JM, Slater JH (1980) Contemporary formation of a biosynthetic enzyme. Proc Nat Acad Sci USA 46:682–690
microbial ecology. Academic, London Gorlenko WM, Dubinina GA, Kuznezow SJ (1977) Ecology of aquatic microor-
Ensign JC, Wolfe RS (1964) Nutritional control of morphogenesis in Arthrobacter ganisms. Nauka, Moscow
crystallopoietes. J Bacteriol 87:924–932 Goto E, Kodama T, Minoda Y (1977) Isolation and culture conditions of ther-
Eutick ML, O’Brien RW, Slaytor M (1978) Bacteria from the gut of Australian mophilic hydrogen bacteria. Agric Biol Chem 41:685–690
termites. Appl Environ Microbiol 35:823–828 Gottlieb SF (1971) Effect of hyperbaric oxygen on microorganisms. Ann Rev
Fairbairn D (1970) Biochemical adaptation and loss of genetic capacity in Microbiol 25:111–152
helminth parasites. Biol Rev 45:29–72 Grant WD, Mills AA, Schofield AK (1979) An alkalophilic species of Ectothior-
Felbeck H, Childress JJ, Somero GN (1981) Calvin-Benson cycle and sulphide hodospira from a Kenyan soda lake. J Gen Microbiol 110:137–142
oxidation enzymes in -animals from sulphiderich habitats. Nature 293:291–293 Grassle JF (1986) The ecology of deep-sea hydrothermal vent communities. Adv
Fenchel TM (1969) The ecology of marine microbenthos. IV. Structure and Mar Biol Ecol 23:301–362
function of the benthic ecosystem. Ophelia 6:1–182 Greenberg EP, Hastings JW, Ulitzur S (1979) Induction of luciferase synthesis in
Fenchel TM, Jørgensen BB (1977) Detritus food chains in aquatic ecosystems: the Beneckea harveyi by other marine bacteria. Arch Microbiol 120:87–91
role of bacteria. Adv Microb Ecol 1:1–58 Griffin DM (1981) Water and microbial stress. Adv Microb Ecol 5:91–136
Fenchel TM, Riedl RJ (1970) The sulfide system: a new biotic community Griffin DM, Luard EJ (1979) Water stress and microbial ecology. In: Shilo M (ed)
underneath the oxidized layer of marine sand bottoms. Mar Biol 7:255–268 Strategies of microbial life in extreme environments. Verlag Chemie,
Fenchel TM, Staarup BJ (1971) Vertical distribution of photosynthetic pigments Weinheim, pp 49–63
and the penetration of light in marine sediments. Oikos 22:172–182 Gromet-Elhanan Z (1977) Electron transport and photophosphorylation in
Ferry JG, Wolfe RS (1976) Anaerobic degradation of benzoate to methane by photosynthetic bacteria. In: Trebst A, Avron M (eds) Encyclopaedia of
a microbial consortium. Arch Microbiol 107:33–40 plant physiology, vol 5. Springer, Berlin, pp 637–662
Fiala G, Stetter KO, Jannasch HW, Langworthy TA, Madon J (1986) Gunner HB, Alexander M (1964) Anaerobic growth of Fusarium oxysporum.
Staphylothermus marinus sp. nov. represents a novel genus of extremely J Bacteriol 87:1309–1316
Hansen MH, Ingvorsen K, Jørgensen BB (1978) Mechanisms of hydrogen sulfide Huber R, Huber G, Segerer A, Seger J, Stetter KO (1987a) Aerobic and anaerobic
release from coastal marine sediments to the atmosphere. Limnol Oceanogr extremely thermophilic autotrophs. In: van Verseveld HW, Duine JA (eds)
23:68–76 Proceedings of the 5th international symposium. Martinus Nijhoff, Dor-
Harder W, Veldkamp H (1968) Physiology of an obligate psychrophilic marine drecht, pp 44–51
Pseudomonas species. J Appl Bacteriol 31:12–33 Huber R, Kristjansson J, Stetter KO (1987b) Pyrobaculum gen. nov., a new
Harder W, Veldkamp H (1971) Competition of marine psychrophilic bacteria at low genus of neutrophilic, rod-shaped archaebacteria from continental solfataras
temperatures. Antonie van Leeuwenhoek J Microbiol Serol 37:51–63 growing optimally at 100 C. Arch Microbiol 149:95–101
Hardie JM, Bowden GH (1974) The normal microbial flora of the mouth. In: Huber R, Kurr M, Jannasch HW, Stetter KO (1989) A novel group of abyssal
Skinner FA, Carr JG (eds) The normal microbial flora of man. Academic, methanogenic archaebacteria (Methanopyrus) growing at 110 C. Nature
London, pp 47–83 342:833–834
Harold R, Stanier RY (1955) The genera Leucothrix and Thiothrix. Bacteriol Rev Huber R, Stoffers P, Cheminee JL, Richnow HH, Stetter KO (1990) Hyperther-
19:49–58 mophilic archaebacteria within the crater and open-sea plume of erupting
Harris RH, Mitchell R (1973) The role of polymers in microbial aggregation. Ann Macdonald Seamount. Nature 345:179–181
Rev Microbiol 27:27–50 Hughes MN, Poole RK (1989) Metals and microorganisms. Chapman and Hall,
Hassan HM, Fridovich I (1979) Superoxide dismutase and its role for survival in London
the presence of oxygen. In: Shilo M (ed) Strategies of microbial life in Hungate RE (1950) The anaerobic mesophilic cellulolytic bacteria. Bacteriol Rev
extreme environments. Verlag Chemie, Weinheim, pp 179–193 14:1–49
Hastings JW, Nealson KH (1977) Bacterial bioluminescence. Ann Rev Microbiol Hungate RE (1962) Ecology of bacteria. In: Gunsalus JC, Stanier RY (eds)
31:549–595 The bacteria, vol IV, The physiology of growth. Academic, New York,
Hattori T, Ishida Y, Maruyama Y, Morita RY, Uchida A (1989) Recent advances in pp 95–119
microbial ecology. In: Proceedings of the 5th international symposium on Hungate RE (1966) The rumen and its microbes. Academic, New York
microbial ecology, Japan Science Society Press, Tokyo Hungate RE (1967) Hydrogen as an intermediate in the rumen fermentation.
Hazelbauer GL (1988) The bacterial chemosensory system. Can J Microbiol Archiv für Mikrobiologie 59:158–164
34:466–474 Hungate RE (1975) The rumen microbial ecosystem. Ann Rev Ecol Syst 6:39–66
Heinen W (1974) Proceedings of the first European workshop on microbial Hussain HM (1973) Ökologische Untersuchungen über die Bedeutung
adaptation to extreme environments. Biosystems 6:57–80 thermophiler Mikroorganismen für die Selbsterhitzung von Heu. Zeitschrift
Held AA (1970) Nutrition and fermentative energy metabolism of the water mold für Allgemeine Mikrobiologie 13:323–334
Aqualinderella fermentans. Mycologia 62:339–358 Hustede E, Liebergesell M, Schlegel HG (1989) The photophobic response
Held AA, Emerson R, Fuller MS, Gleason FH (1969) Blastocladia and of various sulfur and nonsulfur purple bacteria. Photochem Photobiol
Aqualinderella: fermentative water molds with high carbon dioxide optima. 50:809–815
Science 165:706–708 Iannotti EL, Kafkewit D, Wolin MJ, Bryant MP (1973) Glucose fermentation
Henrici AT, Johnson DE (1935) Studies of freshwater bacteria. II. Stalked bacteria, products of Ruminococcus albus grown in continuous culture with Vibrio
a new order of Schizomycetes. J Bacteriol 30:61–86 succinogenes: changes caused by interspecies transfer of H2. J Bacteriol
Herdman M, Janvier M, Waterbury JB, Rippka R, Stanier RY 114:1231–1240
(1979) Deoxyribonucleic acid base composition of cyanobacteria. J Gen Imhoff JF, Trüper HG (1977) Ectothiorhodospira halochloris sp. nov., a new
Microbiol 111:63–71 extremely halophilic bacterium containing bacteriochlorophyll b. Arch
Heukelekian H, Heller A (1940) Relation between food concentration and surface Microbiol 114:115–121
for bacterial growth. J Bacteriol 40:547–558 Inniss WE (1975) Interaction of temperature and psychrophilic microorganisms.
Heumann W, Marx R (1964) Feinstruktur und Funktion der Fimbrien bei dem Ann Rev Microbiol 29:445–465
sternbildenden Bakterium Pseudomonas echinoides. Archiv für Mikrobiologie Inniss WE, Ingraham JL (1978) Microbial life at low temperatures: mechanisms
47:325–337 and molecular aspects. In: Kushner DJ (ed) Microbial life in extreme envi-
Hirsch P (1974) Budding bacteria. Ann Rev Microbiol 28:392–444 ronments. Academic, London, pp 73–104
Hirsch P (1979) Life under conditions of low nutrient concentrations. In: Shilo Jaggar J (1983) Physiological effects of near-ultraviolet radiation on bacteria.
M (ed) Strategies of microbial life in extreme environments. Verlag Chemie, Photochem Photobiol Rev 7:1–75
Weinheim, pp 357–372 Jannasch HW (1955) Zur Ökologie der zymogenen planktischen Bakterienflora
Hirsch P, Pankratz St H (1970) Study of bacterial populations in natural envi- natürlicher Gewässer. Archiv für Mikrobiologie 23:146–180
ronments by use of submerged electron microscope grids. Zeitschrift für Jannasch HW (1957) Die bakterielle Rotfärbung der Salzseen des Wadi Natrun
Allgemeine Mikrobiologie 10:589–605 (Ägypten). Archiv für Hydrobiologie 53:425–433
Hobson PN (1988) The rumen microbial ecosystem. Elsevier Science, London Jannasch HW (1958) Studies of planktonic bacteria by means of a direct
Hochachka PW, Mustafa T (1972) Invertebrate facultative anaerobiosis. Science membrane filter method. J Gen Microbiol 18:609–620
178:1056–1060 Jannasch HW (1960) Versuche über Denitrifikation und die Verfügbarketi
Hochachka PW, Somero GN (1973) Strategies of biochemical adaptation. WB des Sauerstoffes in Wasser und Schlamm. Archiv für Hydrobiologie
Saunders, London 56:335–369
Hoffmann C (1942) Beiträge zur Vegetation des FarbstreifenSandwattes. Kieler Jannasch HW (1967) Enrichment of aquatic bacteria in continuous culture.
Meeresforschungen 4:85–108 Archiv für Mikrobiologie 59:165–173
Holben WE, Jansson J, Chelm B, Tiedje T (1988) DNA probe method for Jannasch HW (1977) Growth kinetics of aquatic bacteria. Soc Appl Bacteriol
detection of specific microorganisms in the soil bacterial community. Appl Symp Ser 6:55–68
Environ Microbiol 54:703–711 Jannasch HW (1978) Microorganisms and their aquatic habitat. In: Krumbein
Holdemann LV, Cato EP, Moore WEC (1977) Anaerobe laboratory WE (ed) Environmental biogeochemistry and geomicrobiology, vol 1. Ann
manual, 4th edn. Virginia Polytechnic Institute and State University, Arbor Scientific, Ann Arbor, pp 17–24
Blacksburg Jannasch HW (1979) Microbial ecology of aquatic low-nutrient habitats. In: Shilo
Horikoshi K, Akiba T (1982) Alkalophilic microorganisms. A new microbial M (ed) Strategies of microbial life in extreme environments. Verlag Chemie,
world. Japan Scientific Societies Press/Springer, Berlin/New York Weinheim, pp 243–260
Huber R, Langworthy TA, König H, Thomm M, Woese CR, Sleytr UB, Stetter KO Jannasch HW (1984) Microbial processes at deep sea hydrothermal
(1986) Thermotoga maritima sp. nov. represents a new genus of unique vents. In: Rona PA, Bostrom K, Laubier L, Smith KL (eds)
extremely thermophilic eubacteria growing up to 90 C. Arch Microbiol Hydrothermal processes at seafloor spreading centers. Plenum, New York,
144:324–333 pp 677–709
Jannasch HW (1989) Chemosynthetically sustained ecosystems in the deep sea. Koch AL (1979) Microbial growth in low concentrations of nutrients. In: Shilo
In: Schlegel HG, Bowien B (eds) Autotrophic bacteria. Springer, Berlin, M (ed) Strategies of microbial life in extreme environments. Verlag Chemie,
pp 147–166 Weinheim, pp 261–279
Jannasch HW, Mateles RI (1974) Experimental bacterial ecology studies in con- Kolkwitz R (1904–1906) Mykologie und Reinigung der städtischen und der
tinuous culture. Adv Microb Physiol 11:165–212 Zuckerfabriksabwässer 391. In: Lafar F (ed) Handbuch der technischen
Jannasch HW, Pritchard PH (1972) The role of inert particulate matter in the Mykologie, vol 3. Jena Gustav, Fischer Verlag
activity of aquatic microorganisms. In: Melchiorri-Santolinie U, Hopton JW Korhonen TK, Nurmiaho E-L, Tuovinen OH (1978) Fimbriation in Thiobacillus
(ed) Detritus and its role in aquatic ecosystems. Memorie dell’Istituto Italiano A2. FEMS Microbiol Lett 3:195–198
di Idrobiologia Dott Marco de Marchi Pallanza Italy Suppl 29:289–308 Koshland DE Jr (1974) The chemotactic response in bacteria. In: Jaenicke L (ed)
Jannasch HW, Taylor CD (1984) Deep-sea microbiology. Ann Rev Microbiol Biochemistry of sensory functions. Springer, Berlin, pp 133–160
37:487–514 Koshland DE Jr (1976) Bacterial chemotaxis as a simple model for a sensory
Jannasch HW, Wirsen CO (1973) Deep-sea microorganisms: in situ response to system. Trends Biochem Sci 1:1–3
nutrient enrichment. Science 180:641–643 Koshland DE (1980) Bacterial chemotaxis as a model behavioral system. Raven,
Jannasch HW, Wirsen CO (1977) Retrieval of concentrated and undecompressed New York
microbial populations from the deep sea. Appl Environ Microbiol 33:642–646 Koshland DE (1981) Biochemistry of sensing and adaptation in a simple bacterial
Jannasch HW, Wirsen CO (1979) Chemosynthetic primary production at East system. Ann Rev Biochem 50:765–782
Pacific sea floor spreading centers. Bioscience 29:592–598 Krinsky NI (1979) Carotenoid pigments: multiple mechanisms for coping with
Jannasch HW, Wirsen CO (1981) Morphological survey of microbial mats near the stress of photosensitized oxidations. In: Shilo M (ed) Strategies of
deep sea thermal vents. Appl Environ Microbiol 41:528–538 microbial life in extreme environments. Verlag Chemie, Weinheim,
Jannasch HW, Wirsen CO (1982) Microbial activities in undecompressed and pp 163–177
decompressed deep-seawater samples. Appl Environ Microbiol 43:1116–1124 Krul JM, Hirsch P, Staley JT (1970) Toxothrix trichogenes (Chol.) Beger et
Jannasch HW, Trüper HG, Tuttle JH (1974) The microbial sulfur cycle in the Bringmann: the organism and its biology. Antonie van Leeuwenhoek
Black Sea. Dergens ET, Ross DA (ed) The Black Sea: its geology, chemistry J Microbiol Serol 36:409–420
and biology, Memoir 20. American Association of Petroleum, Tulsa Krulwich TA, Guffanti AA (1983) Physiology of acidophilic and alkalophilic
Jannasch HW, Wirsen CO, Taylor CD (1976) Undecompressed microbial bacteria. Adv Microb Physiol 24:173–214
populations from the deep sea. Appl Environ Microbiol 32:360–367 Kuenen JG, Boonstra HG, Schröder J, Veldkamp H (1977) Competition for
Jannasch HW, Wirsen CO, Taylor CD (1982) Deep-sea bacteria: isolation in the inorganic substrates among chemoorganotrophic and chemolithotrophic
absence of decompression. Science 216:1315–1317 bacteria. Microb Ecol 3:119–130
Jannasch HW, Huber R, Belkin S, Stetter KO (1988) Thermotoga neapolitana sp. Kushner DJ (1971) Life in extreme environments. In: Buvet R, Ponnamperuma C
nov. of the extremely thermophilic, eubacterial genus Thermotoga. Arch (eds) Chemical evolution and origin of life. North-Holland, Amsterdam,
Microbiol 150:103–104 pp 485–491
Jannasch HW, Nelson DC, Wirsen CO (1989) Massive natural occurrence of Kushner DJ (1978) Life in high salt and solute concentrations: halophilic bacteria.
unusually large bacteria (Beggiatoa sp.) at a hydrothermal deep-sea vent In: Kushner DJ (ed) Microbial life in extreme environments. Academic,
site. Nature 342:834–836 London, pp 317–368
Jones OTG (1977) Electron transport and ATP synthesis in the photosynthetic Kuznezow SI (1959) Die Rolle der Mikroorganismen im Stoffkreislauf der Seen.
bacteria. In: Haddock BA, Hamilton WA (eds) Microbial energetics. Berlin VEB Deutscher Verlag der Wissenschaften
Cambridge University Press, Cambridge, pp 151–183 Kuznezow SI (1977) Trends in the development of ecological microbiology. In:
Jones WJ, Leigh JA, Mayer F, Woese CR, Wolfe RS (1983) Methanococcus Droop MR, Jannasch HW (eds) Advances in aquatic microbiology. Aca-
jannaschii, sp. nov., an extremely thermophilic methanogen from demic, London, pp 1–48
a submarine hydrothermal vent. Arch Microbiol 163:154–261 la Riviére JWM (1963) Cultivation and properties of Thiovulum majus Hinze. In:
Jones WJ, Stugard CE, Jannasch HW (1989) Comparison of Oppenheimer CH (ed) Marine microbiology. Charles C Thomas, Spring-
thermophilic methanogens from submarine hydrothermal vents. Arch field, pp 61–72
Microbiol 151:314–318 la Riviére JWM (1965) Enrichment of colorless sulfur bacteria Zentralblatt für
Jørgensen BB (1982) Ecology of the bacteria of the sulfur cycle with special Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene, Abt. 1,
reference to anoxic-oxic interface environments. Phil Tran R Soc Lond Biol Suppl 1:17–27
Sci 298:543–561 Langworthy TA (1978) Microbial life in extreme pH values. In: Kushner
Jørgensen BB, Revsbech NP (1983) Colorless sulfur bacteria, Beggiatoa spp. and DJ (ed) Microbial life in extreme environments. Academic, London,
Thiovulum spp., in O2 and H2S microgradients. Appl Environ Microbiol pp 279–315
45:1261–1270 Lanyi JK (1979) Physical-chemical aspects of salt-dependence in Halobacteria. In:
Jørgensen BB, Zawacki LX, Jannasch HW (1990) Thermophilic bacterial sulfate Shilo M (ed) Strategies of microbial life in extreme environments. Verlag
reduction in deep-sea sediments at the Guaymas Basin hydrothermal vent Chemie, Weinheim, pp 93–107
site (Gulf of California). Deep-Sea Res 37:695–710 Lapage SP, Shelton JE, Mitchell TG, MacKenzie AR (1970) Culture collections and
Kato G, Maruyama Y, Nakamura M (1979) Role of lectins and lipopolysaccha- the preservation of bacteria. In: Norris JR, Ribbons DW (eds) Methods in
rides in the recognition process of specific legume-Rhizobium symbiosis. microbiology, vol 3A. Academic, London
Agric Biol Chem 43:1085–1092 Larsen H (1967) Biochemical aspects of extreme halophilism. Adv Microb Physiol
Kefford B, Kjelleberg S, Marshall KC (1982) Bacterial scavenging: utilization of 1:97–132
fatty acids localized at a solid-liquid interface. Arch Microbiol 133:257–260 Larsen H (1971) Halophilism, microbial. In: McGraw-Hill encyclopedia of sci-
Kelly MT, Brock TD (1969) Physiological ecology of Leucothrix mucor. J Gen ence and technology, 3rd edn. McGraw-Hill, New York
Microbiol 59:153–162 Larsen H (1973) The halobacteria’s confusion to biology. The fourth A.J. Kluyver
Kjelleberg S, Hermansson M, Mardén P, Jones GW (1987) The transient phase memorial lecture delivered before the Netherlands Society for Microbiology,
between growth and non growth of heterotrophic bacteria, with emphasis on April 1972 at the Delft University of Technology. Antonie van Leeuwenhoek
the marine environment. Annu Rev Microbiol 41:25–49 J Microbiol Serol 39:383–396
Kluyver AJ, Donker HJL (1925) The unity of the chemistry of the fermentative Larsen H (1986) Halophilic and halotolerant microorganisms—an overview and
sugar dissimilation processes of microbes. Proc Roy Acad Amsterdam historical perspective. FEMS Microbiol Rev 39:3–7
28:297–313 Latham MJ, Wolin MJ (1977) Fermentation of cellulose by Ruminococcus
Kluyver AJ, Donker HJL (1926) Die Einheit in der Biochemie Chemie der Zelle flavefaciens in the presence and absence of Methanobacterium ruminantium.
und Gewebe 13:134–190 Appl Environ Microbiol 34:297–301
Latham MJ, Brooker BE, Pettipher GL, Harris PJ (1978) Ruminococcus flavefaciens Matin A, Auger EA, Blum PH, Schultz JE (1989) Genetic basis of
cell coat and adhesion to cotton cellulose and to cell walls in leaves of starvation survival in nondifferentiating bacteria. Ann Rev Microbiol
perennial ryegrass (Lolium perenne). Appl Environ Microbiol 35:156–165 43:293–316
Le Roux NW, Wakerley DS, Hunt SD (1977) Thermophilic Thiobacillus-type Matthews MM, Sistrom WR (1959) Function of carotenoid pigments in non-
bacteria from Icelandic thermal areas. J Gen Microbiol 100:197–201 photosynthetic bacteria. Nature 184:1892–1893
Leathen WW, Braley SA Sr, McIntyre LD (1953) The role of bacteria in Mayer F (1971) Elektronenmikroskopische Untersuchung der Fimbrienkon-
the formation of acid from certain sulfuritic constituents associated traktion bei dem sternbildenden Bodenbakterium Pseudomonas echinoides.
with bituminous coal. II. Ferrous iron oxidizing bacteria. Appl Microbiol Archiv für Mikrobiologie 76:166–173
1:65–68 Mayer F, Schmitt R (1971) Elektronenmikroskopische, diffraktometrische und
Lee A, Phillips M (1978) Isolation and cultivation of spirochetes and other spiral- disc-elektrophoretische Untersuchungen an Fimbrien des sternbildenden
shaped bacteria associated with the cecal mucosa of rats and mice. Appl Bodenbakteriums Pseudomonas echinoides und einer nicht-sternbildenden.
Environ Microbiol 35:610–613 Mutante Archiv für Mikrobiologie 79:311–326
Leifson E (1962) The bacterial flora of distilled and stored water. I. General observa- Mazanec K, Kocur M, Martinec T (1965) Electron microscopy of ultrathin
tions, techniques and ecology. Int Bull Bacteriol Nomen Taxon 12:133–153 sections of Sporosarcina ureae. J Bacteriol 90:808–816
Levi P, Linkletter A (1989) Metals, microorganisms and biotechnology. In: McBee RH (1977) Fermentation in the hindgut. In: Clarke RTJ, Bauchop T (eds)
Hughes MN, Poole RK (eds) Metals and microorganisms. Chapman and Microbial ecology of the gut. Academic, London, pp 185–222
Hall, London McLeod RA (1968) On the role of inorganic ions in the physiology of marine
Liener IE (1976) Phytohemagglutinins (phytolectins). Ann Rev Plant Physiol bacteria. Adv Microbiol Sea 1:95
27:291–319 Meers JL (1973) Growth of bacteria in mixed cultures. CRC Crit Rev Microbiol
Loesche WJ (1969) Oxygen sensitivity of various anaerobic bacteria. Appl 2:139–184
Microbiol 18:723–727 Megusar F, Gantar M (eds) (1986) Perspectives in microbial ecology. Slovene
Lovley DR, Phillips EJP (1987) Novel mode of microbial energy metabolism: Society for Microbiology Ljubljana, Yugoslavia
organic carbon oxidation coupled to dissimilatory reduction of iron or Menzel DW, Ryther JH (1970) Distribution and cycling of organic matter in the
manganese. Appl Environ Microbiol 54:1472–1480 oceans. In: Hood DW (ed) Organic matter in natural waters. Alaska Institute
Madigan MT, Brock TD (1977) Adaptation by hot springs phototrophs to of Marine-Sciences, Alaska
reduced light intensities. Arch Microbiol 113:111–120 Millar WN (1973) Heterotrophic bacterial population in acid coal mine water:
Mandel M, Leadbetter ER, Pfennig N, Trüper HG (1971) Deoxyribonucleic acid Flavobacterium acidurans, sp. n. Int J Syst Bacteriol 23:142–150
base compositions of phototrophic bacteria. Int J Syst Bacteriol 21:222–230 Miller WD (1890) The micro-organisms of the human mouth. Basel Karger,
Mann S, Sparks NH, Frankel RB, Bazylinski DA, Jannasch HW (1990) Biominer- Philadelphia, unaltered reprint from original work
alization of ferrimagnetic greigite (FeS) and iron pyrite (FeS) in Miller RE, Simons LA (1962) Survival of bacteria after twenty-one years in the
a magnetotactic bacterium. Nature 343:258–261 dried state. J Bacteriol 84:1111–1114
Marchlewitz B, Schwartz W (1961) Untersuchungen über die Mikroben- Minato H, Suto T (1978) Technique for fractionation of bacteria in rumen
Assoziation saurer Grubenwässer. Zeitschrift für Allgemeine Mikrobiologie microbial ecosystem. II. Attachment of bacteria isolated from bovine
1:100–114 rumen to cellulose powder in vitro and elution of bacteria attached there-
Marples MJ (1965) The ecology of the human skin. Thomas Charles C, from. J Gen Appl Microbiol 24:1–16
Springfield Mitskevich IN (1979) The total number of biomass of microorganisms in deep
Marples MJ (1974) The normal microbial flora of the skin. In: Skinner FA, Carr waters of the Black Sea (In Russian, with English summary). Mikrobiologiya
JG (eds) The normal microbial flora of man. Academic, London, pp 7–12 48:552–557
Marples MJ (1976) Life on the human skin. Sci Am 220:108–115 Moore WEC, Holdeman LV (1974) Human fecal flora: the normal flora of 20
Marquis RE (1976) High-pressure microbial physiology. In: Rose AH, Tempest Japanese-Hawaiians. Appl Microbiol 27:961–979
DW (eds) Advances in microbial physiology, vol 14. Academic, London, Morita RY (1975) Psychrophilic bacteria. Bacteriol Rev 39:144–167
pp 159–241 Morita RY (1976) Survival of bacteria in cold and moderate hydrostatic pressure
Marquis RE, Matsumara P (1978) Microbial life under pressure. In: Kushner DJ environments with special reference to psychrophilic and barophilic bacte-
(ed) Microbial life in extreme environments. Academic, London, pp 105–158 ria. In: Gray TGR, Postgate JR (eds) The survival of vegetative microbes.
Marshall KC (1976) Interfaces in microbial ecology. Harvard University Press, Cambridge University Press, Cambridge, pp 279–298
Cambridge/London Morris JG (1975) The physiology of obligate anaerobiosis. Adv Microb Physiol
Marshall KC (1979) Growth at interfaces. In: Shilo M (ed) Strategies of microbial 12:169–246
life in extreme environments. Verlag Chemie, Weinheim Morris JG (1976) Fifth Stenhouse-Williams Memorial Lecture—oxygen and the
Marshall KC (ed) (1984) Microbial adhesion and aggregation. Dahlem workshop obligate anaerobe. J Appl Bacteriol 40:229–244
reports, Life Science, vol 31. Springer, New York Morris JG (1978) The biochemistry of anaerobiosis. Biochem Soc Trans 6:353–356
Martin SM (1964) Conservation of microorganisms. Ann Rev Microbiol 18:1–16 Morris JG (1979) Nature of oxygen toxicity in anaerobic microorganisms. In:
Martin HH (1969) Die Struktur der Zellwand bei Gram-negativen. Bakterien Shilo M (ed) Strategies of microbial life in extreme environments. Verlag
Arzneimittel-Forschung 19:266–272 Chemie, Weinheim, pp 149–162
Marx JL (1977) Looking at lectins: do they function in recognition processes? Morris JG, O’Brien RW (1971) Oxygen and clostridia: a review. In: Barker
Science 196:1429–1430 AN, Gould GW, Wolf J (eds) Spore research 1971. Academic, London,
Marx R, Heumann W (1962) Uber Geisselfeinstrukturen und Fimbrien bei zwei pp 1–37
Psuedomonas-Stämmen. Archiv für Mikrobiologie 43:245–254 Mossel DAA (1975) Water and micro-organisms in foods—a synthesis.
Matin A (1979) Microbial regulatory mechanisms at low nutrient concentrations In: Duckworth RB (ed) Water relations of foods. Academic, London,
as studies in chemostat. In: Shilo M (ed) Strategies of microbial life in pp 347–361
extreme environments. Verlag Chemie, Weinheim, pp 323–339 Mossel DAA, Ingram M (1955) The physiology of the microbial spoilage of foods.
Matin A, Veldkamp H (1978) Physiological basis of the selective advantage J Appl Bacteriol 18:232–268
of a Spirillum sp. in a carbon-limited environment. J Gen Microbiol Mountfort DO, Asher RA (1983) Role of catabolic regulatory mechanism in
105:187–197 control of carbohydrate utilization by the rumen anaerobic fungus
Matin A, Veldhuis C, Stegemann V, Veenhuis M (1979) Selective advantage Neocallimastix frontalis. Appl Environ Microbiol 46:1331–1338
of a Spirillum sp. in a carbon-limited environment. Accumulation of Mulder EG, Brotonegoro S (1974) Free-living heterotrophic nitrogen-fixing
poly-b-hydroxybutyric acid and its role in starvation. J Gen Microbiol bacteria. In: Quispel A (ed) The biology of nitrogen fixation. North-Holland,
112:349–355 Amsterdam, pp 37–85
Müller M (1975) Biochemistry of protozoan microbodies: peroxisomes, Pfennig N (1967) Photosynthetic bacteria. Ann Rev Microbiol 21:286–324
glycerophosphate oxidase bodies, hydrogenosomes. Annu Rev Microbiol Pfennig N (1979) General physiology and ecology of photosynthetic bacteria. In:
29:467–483 Sistrom WR, Clayton R (eds) Photosynthetic bacteria. Plenum, New York,
Müller-Neuglück M, Engel H (1961) Photoinaktivierung von Nitrobacter pp 3–18
winogradskyi Buch. Archiv für Mikrobiologie 39:130–138 Pfennig N, Biebl H (1976) Desulfuromonas acetoxidans gen. nov. and sp. nov.,
Nasim A, James AP (1978) Life under conditions of high irradiation. In: Kushner DJ a new anaerobic, sulfur-reducing, acetate-oxidizing bacterium. Arch
(ed) Microbial life in extreme environments. Academic, London, pp 409–439 Microbiol 110:3–12
Neijssel OM, Hueting S, Crabbendam KJ, Tempest DW (1975) Dual pathways of Pfennig N, Cohen-Bazire G (1967) Some properties of the green bacterium
glycerol assimilation in Klebsiella aerogenes NCIB 418. Their regulation and Pelodictyon clathratiforme. Arch Mikrobiol 59:226–236
possible functional significance. Arch Microbiol 104:83–87 Pierson BK, Castenholz RW (1974) A phototrophic gliding filamentous bacte-
Nelson DC, Jannasch HW (1983) Chemoautotrophic growth of a marine rium of hot springs. Chloroflexus aurantiacus, gen. and sp. nov. Arch
Beggiatoa in sulfide-gradient cultures. Arch Microbiol 136:262–269 Microbiol 100:5–24
Nelson DC, Jørgensen BB, Revsbech NP (1986) Growth pattern and yield of Poindexter JS (1964) Biological properties and classification of the Caulobacter
a chemoautotrophic Beggiatoa sp. in oxygen-sulfide microgradients. Appl group. Bacteriol Rev 28:231–295
Environ Microbiol 53:225–233 Poindexter JS (1979) Morphological adaptation to low nutrient concentrations.
Nelson DC, Wirsen CO, Jannasch HW (1989) Characterization of large, autotro- In: Shilo M (ed) Strategies of microbial life in extreme environments. Verlag
phic Beggiatoa spp. abundant at hydrothermal vents of the Guaymas Basin. Chemie, Weinheim, pp 341–356
Appl Environ Microbiol 55:2909–2917 Prebble J, Huda S (1977) The photosensitivity of the malate oxidase system of
Noble WC, Pitcher DG (1979) Microbial ecology of the human skin. Adv Microb a pigmented strain and a carotenoidless mutant of Sarcina lutea (Micrococcus
Ecol 2:245–289 luteus). Arch Microbiol 113:39–42
Noble WC, Somerville DA (1974) Microbiology of human skin. Saunders, Pringsheim EG (1957) Observations on Leucothrix mucor and Leucothrix
London cohaercus nov. sp. with a survey of colorless filamentous organisms. Bacteriol
Nottingham PM, Hungate RE (1969) Methanogenic fermentation of benzoate. Rev 21:69–81
J Bacteriol 98:1170–1172 Prins RA (1977) Biochemical activities of gut microorganisms. In: Clarke
Nultsch W (1975) Phototaxis and photokinesis. In: Carlile MJ (ed) Primitive RTJ, Bauchop T (eds) Microbial ecology of the gut. Academic, London,
sensory and communication systems: the taxes and tropisms of microorgan- pp 73–183
isms and cells. Academic, London, pp 29–90 Raj HD (1977) Leucothrix. CRC Crit Rev Microbiol 5:271–301
O’Brien RW, Morris JG (1971) Oxygen and the growth and metabolism of Ramaley RF, Hixson J (1970) Isolation of a nonpigmented, thermophilic bacte-
Clostridium acetobutylicum. J Gen Microbiol 68:307–318 rium similar to Thermus aquaticus. J Bacteriol 103:527–528
Odum EP (1977) Ecology: the link between the natural and the social sciences, Reddy CA, Bryant MP, Wolin MJ (1972a) Characteristics of S organism isolated
2nd edn. Holt, Rinehart & Winston, London from Methanobacillus omelianskii. J Bacteriol 109:539–545
Ohta K, Kiyomiya A, Koyama N, Nosoh Y (1975) The basis of the alkalophilic Reddy CA, Bryant MP, Wolin MJ (1972b) Ferredoxin-independent conversion
property of a species of bacillus. J Gen Microbiol 86:259–266 of acetaldehyde to acetate and H2 in extracts of S organism. J Bacteriol
Okon Y, Albrecht SL, Burris RH (1976) Factors affecting growth and nitrogen 110:133–138
fixation of Spirillum lipoferum. J Bacteriol 127:1248–1254 Reichelt JL, Baumann P (1973) Taxonomy of the marine, luminous bacteria. Arch
Oren A, Padan E (1978) Induction of anaerobic, photoautotrophic growth in the für Mikrobiol 94:283–330
cyanobacterium Oscillatoria limnetica. J Bacteriol 133:558–563 Repaske DR, Adler J (1981) Change in intracellular pH of escherichia coli
Oren A, Shilo M (1979) Anaerobic heterotrophic dark metabolism in the cyano- mediates the chemotactic response to certain attractants and repellents.
bacterium Oscillatoria limnetica: sulfur respiration and lactate fermentation. J Bacteriol 145:1196–1208
Arch Microbiol 122:77–84 Repeta DJ, Simpson DJ, Jørgensen BB, Jannasch HW (1989) Evidence for
Orpin CG (1972) The culture in vitro of the rumen bacterium Quin’s oval. anoxygenic photosynthesis from the distribution of bacteriochlorophylls in
J GenMicrobiol 73:523–530 the Black Sea. Nature 342:69–72
Orpin CG (1973) The intracellular polysaccharide of the rumen bacterium Eadie’s Richards FR (1975) The Cariaco basin (Trench). Oceanogr Marine Biol Ann Rev
oval. Archiv für Mikrobiologie 90:247–254 13:11–67
Orpin CG, Joblin KN (1988) The rumen anaerobic fungi. In: Hobson PN (ed) Richards FA, Vaccaro RF (1958) The Cariaco Trench, an anaerobic basin in the
The rumen microbial ecosystem. Elsevier Science, London Caribbean Sea. Deep-Sea Res 3:214–228
Ottow JCG (1975) Ecology, physiology, and genetics of fimbriae and pili. Annu Rittenberg SC (1979) Bdellovibrio: a model of biological interactions in nutrient
Rev Microbiol 29:79–108 impoverished environments? In: Shilo M (ed) Strategies of microbial life in
Overbeck J (1972) Zur Struktur und Funktion des aquatischen Ökosystems extreme environments. Verlag Chemie, Weinheim, pp 305–322
Berichte der Deutschen Botanischen Gesellschaft 85:553–579 Robinson JB, Salonius PO, Chase FE (1965) A note on the differential response of
Pace NR, Stahl D, Lane DJ, Olsen GJ (1986) The analysis of natural microbial Arthrobacter spp. and Pseudomonas spp. to drying in soil. Can J Microbiol
populations by ribosomal RNA sequences. Adv Microbiol Ecol 9:1–55 11:746–748
Padan E (1979a) Facultative anoxygenic photosynthesis in cyanobacteria. Ann Rose AH (1968) Physiology of microorganisms at low temperatures. J Appl
Rev Plant Physiol 30:27–40 Bacteriol 31:1–11
Padan E (1979b) Impact of facultatively anaerobic photoautotrophic metabolism Rosebury T (1972) Der Reinlichkeitstick. Hoffmann & Campe Verlag, Hamburg
on ecology of cyanobacteria (blue-green algae). Adv Microb Ecol 3:1–48 Rudd JWM, Taylor CD (1980) Methane cycling in aquatic environments. Adv
Pask-Hughes RA, Williams RAD (1975) Extremely thermophilic Gram-negative Aquat Microbiol 2:77–150
bacteria from hot tap water. J Gen Microbiol 88:321–328 Rupela OP, Tauro P (1973) Isolation and characterization of Thiobacillus from
Pask-Hughes RA, Williams RAD (1977) Yellow-pigmented strains of Thermus alkali soils. Soil Biol Biochem 5:891–897
spp. from Icelandic hot springs. J Gen Microbiol 102:375–383 Russell C, Melville TH (1978) A review: bacteria in the human mouth. J Appl
Patterson H, Irvin R, Costerton JW, Cheng K-J (1975) Ultrastructure and adhe- Bacteriol 44:163–181
sion properties of Ruminococcus albus. J Bacteriol 122:278–287 Rutter PR, Abbott A (1978) A study of the interaction between oral streptococci
Pfennig N (1961) Eine vollsynthetische Nährlösung zur selektiven. Anreicherung and hard surfaces. J Gen Microbiol 105:219–226
einiger Schwefelpurpurbakterien Naturwissenschaften 48:136 Sadoff HL (1973) Comparative aspects of morphogenesis in three prokaryotic
Pfennig N (1965) Anreicherungskulturen für rote und grüne Schwefelbakterien genera. Ann Rev Microbiol 27:133–153
Zentralblatt für Bakteriologie, Parasitenkunde, Infektionskrankheiten und Sadoff HL (1975) Encystment and germination in Azotobacter vinelandii.
Hygiene, Abt. 1, Suppl 1:179–189, 503–504 Bacteriol Rev 39:516–539
Savage DC (1977a) Microbial ecology of the gastrointestinal tract. Ann Rev Souza KA, Deal PH (1977) Characterization of a novel extremely alkaline bacte-
Microbiol 31:107–133 rium. J Gen Microbiol 101:103–109
Savage DC (1977b) Interactions between the host and its microbes. In: Clarke Souza KA, Deal PH, Mack HM, Turnbill CE (1974) Growth and reproduction of
RTJ, Bauchop T (eds) Microbial ecology of the gut. Academic, London microorganisms under extremely alkaline conditions. Appl Microbiol
Savage DC, Fletcher M (1985) Bacterial adhesion. Plenum, New York 28:1066–1068
Scarr MP (1968) Thermophiles in sugar. J Appl Bacteriol 31:66–74 Stahl DA, Lane DJ, Olsen GJ, Pace NR (1985) Characterization of a Yellowstone
Schenk A, Aragno M (1979) Bacillus schlegelii, a new species of thermophilic, hot spring microbial community by 5S rRNA sequences. Appl Environ
facultatively chemolithoautotrophic bacterium oxidizing molecular hydro- Microbiol 49:1379–1384
gen. J Gen Microbiol 115:333–341 Stahl DA, Flesher B, Mansfield HR, Montgomery L (1988) Use of phylogenetically
Schenk EA, Schwemmler W (1983) Endocytobiology II. Walter de Gruyter, Berlin based hybridization probes for studies of ruminal microbial ecology. Appl
Schlegel HG (ed) (1965) Anreicherungskultur und Mutantenauslese Zentralblatt Environ Microbiol 54:1079–1084
für Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene, Abt. Stanier RY (1942) The cytophaga group: a contribution to the biology of
1 Orig., Suppl. 1:179–189 Mycobacteria. Bacteriol Rev 6:143–196
Schlegel HG, Bowien B (1989) Autotrophic bacteria. Science Tech/Springer, Stapp C, Bortels H (1931) Der Pflanzenkrebs und sein Erreger Pseudomonas
Madison/New York tumefaciens. II. Mitteilung: Über den Lebenskreislauf von Pseudomonas
Schlegel HG, Jannasch HW (1967) Enrichment cultures. Ann Rev Microbiol tumefaciens Zeitschrift für Parasitenkunde 4:101–125
21:49–70 Stapp C, Knösel D (1954) Zur Genetik sternbildender Bakterien Zentralblatt für
Schlegel HG, Pfennig N (1961) Die Anreicherungskultur einiger Schwefelpurpur- Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene, Abt. 2.
bakterien. Archiv für Mikrobiologie 38:1–39 108:244–259
Schmaljohann R, Flügel HJ (1987) Methane-oxidizing bacteria in Pogonophora. Steinbüchel A (1986) Anaerobic pyruvate metabolism of Trichomonas foetus and
Sarsia 72:91–98 Trichomonas vaginalis hydrogenosomes. Mole Biochem Parasitol 20:57–65
Schmidt JM (1971) Prosthecate bacteria. Ann Rev Microbiol 25:93–110 Stetter KO (1986) Diversity of extremely thermophilic archaebacteria. In: Brock
Schmidt-Lorenz W (1967) Behavior of microorganisms at low temperatures. TD (ed) Thermophiles, general, molecular and applied microbiology. Wiley,
Bulletin de l’Institut International du Froid 1–59 New York, pp 39–74
Schnaitman C, Lundgren DG (1965) Organic compounds in the spent medium of Stetter KO (1989) Extremely thermophilic chemolithoautotrophic
Ferrobacillus ferrooxidans. Can J Microbiol 11:23–27 archaebacteria. In: Schlegel HG, Bowien B (eds) Autotrophic bacteria. Sci-
Schön GH, Engel H (1962) Der Einfluss des Lichtes auf Nitrosomonas europaea. ence Tech, Madison, pp 167–176
Win Archiv für Mikrobiologie 42:415–428 Stetter KO, Zillig W (1985) Thermoplasma and the thermophilic sulfur-
Schroff G, Schöttler U (1977) Anaerobic reduction of fumarate in the body wall dependent archaebacteria. In: Wolfe RS, Woese CR (eds) The bacteria,
musculature of Arenicola marina (Polychaeta). J Comp Physiol 116:325–336 vol 8. Academic, New York, pp 85–170
Schultz JE, Breznak JA (1978) Heterotrophic bacteria present in hindguts of Stetter KO, König H, Stackebrandt E (1983) Pyrodictium gen. nov., a new genus of
wood-eating termites [Reticulitermes flavipes (Kollar)]. Appl Environ submarine disc-shaped sulphur reducing archaebacteria growing optimally
Microbiol 35:930–936 at 105 C. Syst Appl Microbiol 4:535–551
Schultz JE, Breznak JA (1979) Cross-feeding of lactate between Streptococcus lactis Stetter KO, Fiala G, Huber G, Huber R, Segerer A (1990) Hyperthermophilic
and Bacteroides sp. isolated from termite hindguts. Appl Environ Microbiol microorganisms. FEMS Microbiol Rev 75:117–124
37:1206–1210 Stockhausen F (1907) Okologie, ‘‘Anhäufungen’’ nach Beijerinck. Institut für
Schweinfurth G, Lewin L (1898) Beiträge zur Topographie und Geochemie des Gärungsgewerbe, Berlin
ägyptischen Natron-thals. Zeitschrift für die Gesamte Erdkunde 33:1–25 Strange RE (1976) Microbiol response to mild stress. Meadowfield Press, Durham
Schwemmler W, Schenk EA (1980) Endocytobiology. Walter de Gruyter, Berlin Strength WJ, Isani B, Linn DM, Williams FD, Vandermolen GE, Laughon BE,
Shilo M (ed) (1979) Strategies of microbial life in extreme environments. Verlag Krieg NR (1976) Isolation and characterization of Aquaspirillum fascilus sp.
Chemie, Weinheim nov., a rod-shaped, nitrogen-fixing bacterium having unusual flagella. Int
Shokes RF, Trabant PK, Presley BJ, Reid DF (1977) Anoxic, hypersaline basin in J Syst Bacteriol 26:253–268
the northern. Gulf of Mexico Sci 196:1443–1446 Strohl WR, Larkin JM (1979) Enumeration, isolation, and characterization
Siñeriz F, Pirt SJ (1977) Methane production from glucose by a mixed culture of Beggiatoa from freshwater sediments. Appl Environ Microbiol 36:755–770
of bacteria in the chemostat: the role of Citrobacter. J Gen Microbiol Stumm CK, Zwart KB (1986) Symbiosis of protozoa with hydrogen-utilizing
101:57–64 methanogens. Microbiol Sci 3:100–105
Singer CE, Ames BN (1970) Sunlight ultraviolet and bacterial DNA base ratios. Sverdrup HW, Johnson MW, Fleming RH (1942) The oceans. Prentice-Hall, London
Science 170:822–826 Swart-Füchtbauer H, Rippel-Baldes A (1951) Die baktericide Wirkung des
Skopintsev BA, Karpov AV, Vershinina OA (1959) Study of the dynamics of some Sonnenlichtes. Archiv für Mikrobiologie 16:358–362
sulfur compounds in the Black Sea under experimental conditions. Soviet Switalski L, Hook M, Beachey EH (1989) Molecular mechanisms of microbial
Oceanogr Ser 4:55–72, English translation adhesion. Springer, New York
Sleat R, Robinson JP (1984) The bacteriology of anaerobic degradation of aro- Tannock GW (1990) The Microecology of lactobacilli inhabiting the gastrointes-
matic compounds. J Appl Bacteriol 57:381–394 tinal tract. Adv Microb Ecol 11:147–171
Smith DW (1978) Water relations of microorganisms in nature. In: Tansey MR, Brock TD (1978) Microbial life at high temperatures: ecological
Kushner DJ (ed) Microbial life in extreme environments. Academic, London, aspects. In: Kushner DJ (ed) Microbial life in extreme environments.
pp 369–380 Academic, London, pp 159–216
Sneath PHA (1962) Longevity of micro-organisms. Nature 195:643–646 Tempest DW, Neijssel OM (1976) Microbial adaptation of low-nutrient environ-
Somerville CC, Knight IT, Straube WL, Colwell RR (1989) Simple, rapid method ments. In: Dean ACR, Ellwood DC, Evans CGT, Melling J (eds) Continuous
for direct isolation of nucleic acids from aquatic environments. Appl Environ culture 6: applications and new fields. Ellis Horwood, Chichester,
Microbiol 55:548–554 pp 283–296
Sorokin YI (1964) On the primary production and bacterial activities in the Black Tempest DW, Neijssel OM (1979) Eco-physiological aspects of microbial
Sea. Journal du Conseil Conseil International pour l’Exploration de la Mer growth in aerobic nutrient-limited environments. Adv Microb Ecol
29:41–60 2:105–153
Sorokin YI (1970) Interrelations between sulphur and carbon turnover in Tempest DW, Meers JL, Brown CM (1970) Synthesis of glutamate in Aerobacter
meromictic lakes. Archiv für Hydrobiologie 66:391–446 aerogenes by a hitherto unknown route. Biochem J 117:405–407
Sorokin YI (1972) The bacterial population and process of sulfide oxidation in the Tempest DW, Meers JL, Brown CM (1973) Glutamate synthetase (Gogat): a key
Black Sea. J Cons Int Explor Mer 34:423–454 enzyme in the assimilation of ammonia by prokaryotic organisms. In:
Prusiner S, Stadtman ER (eds) The enzymes of glutamine metabolism. Wangersky PJ (1976) The surface film as a physical environment. Ann Rev Ecol
Academic, New York, pp 167–182 Syst 7:161–176
Thiele HH (1968) Die Verwertung einfacher organischer Substrate durch Watson SW, Waterbury JB (1969) The sterile hot brines of the Red Sea. In: Degens
Thiorhodaceae. Archiv für Mikrobiologie 60:124–138 ET, Ross DA (eds) Hot brines and recent heavy metal deposit in the Red Sea.
Tindall BJ, Trüper HG (1986) Ecophysiology of the aerobic halophilic Springer, New York, pp 272–281
archaebacteria. Syst Appl Microbiol 7:202–212 Weibull C (1960) Movement. In: Gunsalus IC, Stanier RY (eds) The bacteria,
Torma AE (1977) The role of Thiobacillus ferrooxidans in hydrometallurgical vol 1, Structure. Academic, New York, pp 153–205
processes. Adv Biochem Engin 6:1–37 Weimer PJ, Zeikus JG (1977) Fermentation of cellulose and cellobiose by Clos-
Torsvik VL, Goksoyr J (1978) Determination of bacterial DNA in soil. Soil Biol tridium thermocellum in the absence and presence of Methanobacterium
Biochem 10:7–12 thermoautotrophicum. Appl Environ Microbiol 33:289–297
Trüper HG (1969) Bacterial sulfate reduction in the Red Sea hot brines. In: Weiner RM, Devine RA, Powell DM, Dagasan L, Moore RL (1985) Hyphomonas
Degens ET, Ross DA (eds) Hot brines and recent heavy metal deposits in oceanitis spec. nov., H. hirschiana spec. nov. and H. jannaschiana spec. nov.
the Red Sea. Springer, New York, pp 262–271 Int J System Bact 35:237–243
Trüper HG (1976) Higher taxa of the phototrophic bacteria: Chloroflexaceae fam. Weiss RL (1973) Attachment of bacteria to sulphur in extreme environments.
nov., a family for the gliding filamentous, phototrophic ‘‘green’’ bacteria. Int J Gen Microbiol 77:501–507
J Syst Bacteriol 26:74–75 Weller R, Ward DM (1989) Selective recovery of 16 S rRNA sequences from
Tschech A (1989) Der anaerobe Abbau von aromatischen Verbindungen Forum. natural microbial communities in the form of cDNA. Appl Environ
Mikrobiologie 12:251–264 Microbiol 55:1818–1822
Tuovinen OH, Kelly DP (1972) Biology of Thiobacillus ferrooxidans in relation to Whittaker RH, Levin SA, Root RB (1973) Niche, habitat and ecotope. Am Nat
the microbiological leaching of sulphide ores. Zeitschrift für Allgemeine 107:321–338
Mikrobiologie 12:311–346 Whittenbury R, Davies SL, Davey JF (1970) Exospores and cysts formed by
Tuttle JH, Randles CI, Dugan PR (1968) Activity of microorganisms in acid mine methane-utilizing bacteria. J Gen Microbiol 61:219–226
water. I. Influence of acid water on aerobic heterotrophs of a normal stream. Wiegel J, Schlegel HG (1976) Enrichment and isolation of nitrogen fixing hydro-
J Bacteriol 95:1495–1503 gen bacteria. Arch Microbiol 107:139–142
Uesugi I, Yajima M (1978) Oxygen and ‘‘strictly anaerobic’’ intestinal bacteria. I. Wiegel J, Wilke D, Baumgarten J, Opitz R, Schlegel HG (1978) Transfer of
Effects of dissolved oxygen on growth. Zeitschrift für Allgemeine the nitrogen fixing hydrogen bacterium Corynebacterium
Mikrobiologie 18:287–295 autotrophicum (Baumgarten et al.) to Xanthobacter gen. nov. Int J Syst
Umbreit TH, Pate JL (1978) Characterization of the holdfast region of wild-type Bacteriol 28:573–581
cells of holdfast mutants of Asticcacaulis biprosthecum. Arch Microbiol Wiley WR, Stokes JL (1963) Effect of pH and ammonium ions on the permeabil-
118:157–168 ity of Bacillus pasteurii. J Bacteriol 86:1152–1156
Unterman PM, Baumann P, McLean DL (1989) Pea aphid symbiont relationships Williams AG (1986) Rumen holotrich ciliate protozoa. Microbiol Rev 50:25–49
established by analysis of 16S rRNAs. J Bacteriol 171:2970–2974 Windberger E, Huber R, Trincone A, Fricke H, Stetter KO (1989) Thermotoga
van Gemerden H (1974) Coexistence of organisms competing for the same thermarum sp. nov. and Thermotoga neapolitana occurring in African con-
substrate: an example among the purple sulfur bacteria. Microb Ecol tinental sulfataric springs. Arch Microbiol 151:506–512
1:104–119 Winfrey MR, Zeikus JG (1977) Effect of sulfate on carbon and electron flow
van Niel CB (1932) On the morphology and physiology of the purple and green during microbial methanogenesis in fresh-water sediments. Appl Environ
sulphur bacteria. Archiv für Mikrobiologie 3:1–112 Microbiol 33:275–281
van Niel CB (1936) On the metabolism of the Thiorhodaceae. Archiv für Winogradksy SN (1926) Études sur la microbiologie du sol. Sur les microbes
Mikrobiologie 7:323–358 fixateurs d’azote. Annales de l’Institut Pasteur 40:455–520
van Niel CB (1955) The microbe as a whole. In: Waksman SA (ed) Perspectives Winogradsky SN (1925) Etudes sur la microbiologie du sol. I. Sur la m;aaethode.
and horizons in microbiology. Rutgers University Press, New Brunswick, Annales de l’Institut Pasteur 39:299–354
pp 3–12 Winogradsky SN (1947) Principles de la Microbiologie Ecologique. Antonie van
van Veen WL, Mulder EG, Deinema MH (1978) The Sphaerotilus-Leptothrix Leeuwenhoek J Microbiol Serol 12:5–15
group of bacteria. Microbiol Rev 42:329–356 Winogradsky SN (1949) Microbiologie du sol: problèmes et méthodes. Masson et
Vedder A (1934) Bacillus alcalophilus sp. nov., benevens enkle ervaringen met Cie, Paris
sterk alcalische voedingsbodems. Antonie van Leeuwenhoek J Microbiol Wirsen CO, Jannasch HW (1975) Activity of marine psychrophilic bacteria at
Serol 1:141–147 elevated hydrostatic pressures and low temperatures. Mar Biol 31:201–209
Veldkamp H (1970) Enrichment cultures of prokaryotic organisms. In: Norris JR, Wirsen CO, Jannasch HW (1978) Physiological and morphological observations
Ribbons DW (eds) Methods in microbiology, vol 3A. Academic, London, on Thiovulum sp. J Bacteriol 136:765–774
pp 305–361 Woese CR (1987) Bacterial evolution. Microbiol Rev 51:221–271
Veldkamp H (1976) Continuous culture in microbial physiology and ecology Wolfe RS (1960) Observations and studies of Crenothrix polyspora. J Am Water
patterns of progress. Meadowfield Press, Durham Works Assoc 52:915–918
Veldkamp H, Jannasch HW (1972) Mixed culture studies with the chemostat. Wolin MJ (1976) Interactions between H2-producing and methane-producing
J Appl Chem Biotechnol 22:105–123 species. In: Schlegel HG, Gottschalk G, Pfennig N (eds) Microbial
Veldkamp H, van den Berg G, Zevenhuizen LPTM (1963) Glutamic acid produc- production and utilization of gases. Goltze, Göttingen, pp 141–150
tion by Arthrobacter globiformis. Antonie van Leeuwenhoek J Microbiol Serol Wolin MJ, Miller TL (1982) Interspecies hydrogen transfer: 15 years later.
29:35–51 ASM-News 48:561–565
Voelz H, Dworkin M (1962) Fine structure of Myxococcus xanthus during mor- Woodroffe RCS, Shaw DA (1974) Natural control and ecology of microbial
phogenesis. J Bacteriol 84:943–952 populations on skin and hair. In: Skinner FA, Carr JG (eds) The normal
Völker H, Schweisfurth R, Hirsch P (1977) Morphology and ultrastructure of microbial flora of man. Academic, London, pp 13–34
Crenothrix polyspora Cohn. J Bacteriol 131:306–313 Yayanos AA, Dietz AS (1983) Death of a hadal deep-sea bacterium after decom-
Walsby AE (1970) The gas vesicles of aquatic prokaryotes Regulations between pression. Science 220:497–498, Washington
structure and function in the prokaryotic cell. In: Society for General Micro- Yayanos AA, Dietz AS, Van Boxtel R (1979) Isolation of a deep-sea barophilic
biology Symposium, vol 28. Cambridge University Press, London, New York, bacterium and some of its growth characteristics. Science 205:808–810,
Melbourne, pp 327–357 Washington
Walsby AE (1975) Gas vesicles. Ann Rev Plant Physiol 26:427–439 Yayanos AA, Dietz AS, Van Boxtel R (1981) Obligately barophilic bacterium from
Walsby AE (1977) The gas vacuoles of blue-green algae. Sci Am 237:90–97 the Marianas Trench. Proc Nat Acad Sci USA 78:5212–5215
Yayanos AA, Dietz AS, Van Boxtel R (1982) Dependence of reproduction rate on archaebacteria isolated from Icelandic solfataras Zlb. Bakt. Hyg., I. Abt.
pressure as a hallmark of deep-sea bacteria. Appl Environ Microbiol Orig. C 2:205–227
44:1356–1361 Zillig W, Schnabel R, Tu J, Stetter KO (1982) The phylogeny of archaebacteria,
Zaitsev Yu-P (1971) Vinogradov KA (ed) Marine neustonology. Keter Press, including novel anaerobic thermoacidophiles in the light of RNA polymerase
Jerusalem structure. Naturwiss 69:197–204
Zebe E (1977) Anaerober Stoffwechsel bei wirbellosen Tieren Vorträge der ZoBell CE (1946) Marine microbiology, a monograph on hydrobacteriology.
Rheinisch-Westfälischen Akademie der Wiseenschaften Chronica Botanica, Waltham
Zeikus JG, Wolfe RS (1972) Methanobacterium thermoautotrophicus sp. n., an ZoBell CE (1970) Pressure effects of morphology and life processes. In: Zimmer-
anaerobic, autotrophic, extreme thermophile. J Bacteriol 109:707–713 mann A (ed) High pressure effects on cellular processes. Academic, London,
Zhao H, Wood AG, Widdel F, Bryant MP (1988) An extremely thermophilic pp 85–130
Methanococcus from a deep-sea hydrothermal vent and its plasmids. Arch ZoBell CE, Johnson FH (1949) The influence of hydrostatic pressure on the
Microbiol 150:178–183 growth and viability of terrestrial and marine bacteria. J Bacteriol 57:179–189
Zillig W, Stetter KO, Schäfer W, Janekovic D, Wunderl S, Holz I, Palm P (1981) ZoBell CE, Morita RY (1957) Barophilic bacteria in some deep sea sediments.
Thermoproteales: a novel type of extremely thermoacidophilic anaerobic J Bacteriol 73:563–568
4 Origin of Life: RNA World Versus
Autocatalytic Anabolist
Munich, Germany
Chapel Hill, USA
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Sources for Nutrients and Energy

Sources for Nutrients and Energy . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Nutrients required by the anabolist theory must have always been
available as volcanic-hydrothermal fluids (CO2, CO, COS,
Induction, Reproduction, Evolution . . . . . . . . . . . . . . . . . . . . . . . . 81 CH3SH, HCN, P4O10, H2S, H2, N2, NH3) with chemical poten-
tials due to quenching to low temperatures of, say 100–160 C.
Evolutionary Connex with Extant Genetic Machinery . . . . . 82 They show multiple interrelated functions by contact to crustal
transition metal minerals, notably by the reaction FeS + H2S !
Evolutionary Connex with Extant Cellular Structures . . . . . 83 FeS2 + 2H+ + 2e (E0 = 620 mV) as electron source for
metabolic reactions and N2-fixation (Dörr et al. 2003).
Evolutionary Connex with Extant Metabolism . . . . . . . . . . . . . 84 The RNA world theory postulates a ‘‘prebiotic broth’’ with
activated nucleotides formed from preaccumulated organic
Temperature, Phylogeny, and Time Frame of Life . . . . . . . . . . 85 compounds. Experiments typically follow a modular scheme:
nucleosides made from bases and ribose with subsequent acti-
Explanatory Power and Explanatory Fallacies . . . . . . . . . . . . . 86 vation. Besides requiring mutually incompatible reaction con-
ditions (Shapiro 1986), they typically result in intractable
Conclusion and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 mixtures (Schwartz 2008), if they are chemically feasible at all.
For solving some of these problems an alternative synthesis of
pyrimidine nucleotides departs from an aqueous, phosphate-
Introduction buffered solution of cyanamide with sequential additions of
pure glycolaldehyde, pure glyceraldehyde, and pure
The deep past of the Earth is unobservable. Therefore, the cyanoacetylene (Powner et al. 2009)—reactants that pose more
problem of the origin of life can only be approached by problems than they solve. In view of these problems it has been
a theory. Such theories are scientific to the extent that they suggested that the first reproducing ‘‘informational’’ molecules
explain known facts of biology, predict unknown facts of chem- were proteins, with thioesters as energy source for polyconden-
istry, agree with unproblematic geohistoric theories, and are sation (De Duve 1991). This idea of a ‘‘protein world’’ has never
compatible with universal laws of physics and chemistry. These matured to a theory.
requirements follow from Popper’s methodology of science
(Popper 1959). Only two theories on the origin of life appear
to be detailed enough for an evaluation along these lines: Induction, Reproduction, Evolution
1. The RNA world theory (Gesteland et al. 1999; Cech 2011;
Both approaches assume a molecular autocatalytic system that
Szostak 2011) postulates an origin with a population of
forms de novo in a certain chemical context (induction), and
RNA-like molecules that replicated in a cold prebiotic
subsequently promotes its own synthesis (reproduction) with
broth of activated nucleotides, and served at the same time
modifications (evolution). For the original RNA world theory
as catalysts (ribozymes) for replication and for metabolic
induction means de novo formation of a population of oligo- or
transformations.
polynucleotides in a prebiotic broth. There is experimental
2. The autocatalytic anabolist theory (Wächtershäuser 1992,
support for nucleotides that are activated by nonprebiotic imid-
2006, 2010) posits a hot origin by transition metal-catalyzed
azole-phosphate groups in combination with clay catalysis
carbon fixation that evolved by organic products bonding as
(Prakash et al. 2009). A spontaneous onset of replication and
catalyst-promoting ligands to metal centers (also termed FeS
evolution is too improbable to be expected under such experi-
world in view of the dominance of Fe and S).
mental conditions, the required strand separation by temporary
These two theories are here compared with regard to their elevation of temperature (Kuhn 1972) adding to the problem.
relative empirical merit. Selected references are cited for reasons For such polycondensations liquid water is a problem, notably at
of being very relevant, very recent, or very forgotten. elevated temperatures. Experiments with preestablished RNAs
82 4 Origin of Life: RNA World Versus Autocatalytic Anabolist
feedback effect
= stabilizing by
metabolic reproduction
feed-forward effect
= innovation by
metabolic evolution
. Fig. 4.1
Product ligand effects (A: starting material; B, C: organic products; Km, K0 m: transition metal catalysts; dashed arrow: ligand effect; double
arrow: evolutionary change)
and nucleotides in the absence of a sophisticated catalyst Co2(CO)8/CH3SH/CO (Huber et al. 2012), or from Ni(OH)2/
(mainly due to Orgel and his school) have provided something (CO,H2)/CN (Huber et al. 2012); peptides from a-amino acids
like a conclusive falsification. As a consequence the experimental by activation with CO/H2S via COS (Huber et al. 2003); and a-
program shifted to in vitro evolution experiments with the hydroxy/amino acids (or derivatives) from HCN/Ni(OH)2/(CO,
aim to create catalytic RNA (a ribozyme) that could serve as H2) (Huber et al. 2012). Phenyl-acetate and phenyl-lactate (via
catalyst for its own replication from nucleoside triphosphates. phenyl-pyruvate) form by reaction of benzyl mercaptan with
Experiments in this direction were successful, but the demon- HCN/CO/Ni(OH)2. Yield of the latter reaction increases by
strated requirement of long ribozymes of at least 165 nucleotides addition of Gly or Ala, which demonstrates the chemical possi-
renders their emergence in a prebiotic broth extremely improb- bility of a feed-forward effect (Huber et al. 2012).
able (Johnston et al. 2001). From there the research program
shifted to RNA replication inside lipid vesicles (Szostak et al.
2001). For an origin in a prebiotic broth this move compounded Evolutionary Connex with Extant Genetic
the problems further by requiring a high content of Machinery
lipids, while in another sense it leads us to the realization that
the genetic machinery must have originated inside a cellular The RNA world theory assumes a phase of life, whereby the
entity. catalytic biopolymers consisted exclusively of ribozymes.
According to the anabolist theory reproduction and evolu- Hence the name ‘‘RNA world.’’ With the appearance of the ur-
tion are induced by the reductive formation of low-molecular- ribosome began the translation of RNA sequences into protein
weight organic compounds from volcanic C1-compounds by sequences. This gave rise to an RNA-protein world with a step-
transition metal catalysis. For these redox reactions liquid by-step replacement of ribozymes by enzymes (enzymatic take-
water is a benefit rather than a detriment. Key to understanding over). Much later, RNA as replicating informational molecule
the possibility of reproduction and evolution at this simple was replaced by DNA, with ribonucleotide reductases appearing
molecular level is the recognition that some of the produced first as ribozymes to be replaced later by enzymes.
organic compounds have functional groups, by which they may By the anabolist theory life began with a direct mechanism of
become ligands of transition metal centers, increasing their evolution, organic products, e.g., peptides, feeding directly as
catalytic activity. There are two effects of an organic product ligands into metallocatalysts. Originally, the components of
B (> Fig. 4.1): Feedback effect on catalyst for producing the nucleic acids (ribose, bases, nucleosides, nucleotides) earned
same organic product B (metabolic reproduction); or feed- their keep also as ligands; and so did oligonucleotides giving
forward effect on catalyst for producing another organic prod- rise to catalytic metalloribonucleo particles. The earliest oligo-
uct C (metabolic evolution). New ligands improve catalysts, nucleotide sequences were not hereditary. Nevertheless base
which then elicit new metabolic reactions with new products pairing would have started soon, first as structural determinant
that become further new ligands and so forth. In this manner the and later as a new kind of catalysis by sequence complementa-
metabolism evolves by terminal extension or lateral branching tion. Now information transfer and its fidelity began to become
of its pathways, by closing reaction cycles, by recruiting new important, ushering in a massively parallel evolution of the
nutrients or new catalytic metals, or by conquering new envi- incipient genetic machinery: from sequence copying to replica-
ronments. By these effects the metabolism self-expands at an tion and from peptide formation to translation, drawing in
accelerating pace, resulting in an avalanche breakthrough. more and more amino acids, and generating ever longer coded
Demonstrated induction reactions comprise formation of peptide/protein ligands. DNA as stable information storage
RSH from CO2/FeS/H2S (Heinen and Lauwers 1996); RSH and molecule may well have followed transient catalytic RNA on its
RCOOH from CO/NiS/H2S (Loison et al. 2010); COS, CH3SH, heels. In this way an indirect mechanism of evolution appeared,
and CH3COSCH3 from CO/NiS/H2S via COS (Huber and with accidental sequence mutations being translated into mod-
Wächtershäuser 1997); CH3-CO-COOH from CO/FeS/ ified proteins for metalloenzymes, followed by selection: life’s
C9H19SH at 200 MPa and 250 C (Cody et al. 2000), from great randomizer invented by life itself.
Origin of Life: RNA World Versus Autocatalytic Anabolist 4 83
Much of the early history of life is recorded in the structures

of the ribosome (Harish and Caetano-Anollés 2012). By gener-
ating mutually compatible rooted phylogenetic trees directly
from rRNA substructures and from ribosomal protein domain
structures in accordance with cladistic principles it was possible
to establish a timeline of ribosomal evolution, correlating sub-
structures with ages. This revealed a gradual coevolution of
ribosomal RNAs and ribosomal proteins with the ribosomal
center for peptide bond formation having less antiquity than
the ribonucleoprotein ratchet of the small subunit. The authors
suggest that these oldest regions of the ribosome are ‘‘relics of an
ancient ribonucleoprotein world,’’ which corresponds nicely
with the structural complementarity of mutual scaffolding of
RNAs and peptides before the dawn of the genetic machinery
(Carter and Kraut 1974). These findings contradict the central
thesis of the RNA world theory that ribozymes ‘‘invented’’ pep-
tide bond formation, while in accordance with the anabolist
theory RNA is predated by metallopeptides.
Evolutionary Connex with Extant Cellular

Structures
A multicomponent genetic machinery could not become

established prior to cellularization. For the RNA world theory
this means a nutritional paradox: RNA replication requires ionic . Fig. 4.2
or hydrophilic nutrients, for which a lipid membrane is imper- Formation of deprotonated anionic acetyl-thioester (AT)
meable. A lipid vesicle that is competent for containment would
cause starvation and a nonstarving, leaky vesicle would be inef-
fective as means for containment. intermediate (AT) may form by a redox reaction with
According the anabolist theory the metabolism evolves to the FeS/H2S involving a nucleophilic attack of a sulfhydryl anion
point where it generates lipids intrinsically. These are inert and on the a-mercapto group of thioglycolyl thioester (TGT)
accumulate on the mineral surfaces with the effect of surface (reductive elimination of HSSH en route to pyrite). Desulfuri-
lipophilization and reduced activity of H2O and H3O+. Thereby zation of thioglycolic acid by FeS/H2S (Blöchl et al. 1992) and
all kinds of condensation reactions are protected from the conversion of thioglycolic acid to acetyl-amide by FeS/H2S (Kel-
hydrolytic effect of liquid water (collective feed-forward effect) ler et al. 1994) support this proposal. TGT might form reduc-
including the formation of lipids by condensation reactions tively from CO and NiS/H2S. The anionic intermediate (AT)
(collective feedback effect). Further accumulation will eventually may in turn become reversibly carboxylated to form malonyl-
lead first to mineral-supported bilayer membranes and later to thioester (MT).
semi-cellular structures bounded partly by the mineral base and For lipid synthesis (> Fig. 4.3) the anionic intermediate

partly by a self-supporting lipid membrane. At this stage the ( AT) would engage in a subsequent nucleophilic attack on
surface metabolism coexists with a membrane metabolism and the thioester group of acetyl-thioester (AT) to generate 3-keto-
a cytosol metabolism and the genetic machinery can begin to butyryl-thioester (3KBT) en route to fatty acid lipids by the chain
come into the world of life. Ultimately autonomous lipid vesicles length arithmetic 2 + 2 + 2 + 2 + 2 + 2 = 12 (pathway A); or on the
are formed with mineral inclusions. A nutritional paradox does thioester group of malonyl-thioester (MT) to generate 3-keto-
not arise, since bilayer lipid membranes are permeable for the glutaryl-thioester (3KGT), 3-hydroxy-glutaryl-thioester (3HGT),
small, neutral inorganic nutrients. glutaconyl-thioester (GT) and its decarboxylation product, an
In view of their low water solubility and strong bonding to anionic form of butenyl-thioester (-BT) en route to fatty acid
transition metal centers relatively short alkyl mercaptans would lipids by the chain length arithmetic 4 + 4 + 4 = 12 (pathway B);
have been effective initial lipids, soon to be replaced by long- or on the keto group of 3-keto-butyryl-thioester (3KBT) to
chain fatty acids. All extant forms of lipid biosynthesis may go generate 3-hydroxy-3-methyl-glutaryl-thioester (3H3MGT)
back to a common precursor (> Figs. 4.2, > 4.3). However, and 3-methyl-glutaconyl-thioester (3MGT) and its decarboxyl-
primordial condensation of acetyl-thioester (AT) requires con- ation product, an anionic form of 3-methyl-butenyl-thioester
version to an anionic intermediate (AT) by deprotonation in a- (3MBT) en route to isoprenoid acid lipids by the chain length
position. This in turn requires a strong base, which cannot exist arithmetic 4 + 4 + 4 = 12 (pathway C). Pathways B and C are
under aqueous conditions. Alternatively, the anionic favored, because they require fewer steps than pathway A.
CH3 CH2
CO-SCH3 a CO-SCH3 HOOC CO-SCH3

– a
AT AT MT
c O
CO-SCH3 HOOC CO-SCH3
O
b 3KGT
3KBT
OH
HOOC CO-SCH3 OH
CO-SCH3 HOOC CO-SCH3
–H2O 3H3MGT
3HGT
HOOC CO-SCH3
HOOC CO-SCH3
fatty acid lipids by
multiple repetitions of GT
–CO2 3MGT –CO2
steps a and b
CO-SCH3 CO-SCH3
a –3MBT a –BT
Pathway A Pathway C Pathway B
. Fig. 4.3
Family of anionic species for lipid synthesis (a = condensation with thioester group; b = elimination of oxo group; c = condensation with
keto group)
They are vinylogs of pathway A. Pathway B generates unsatu- and a compensatory ‘‘altruistic feedback’’ effect promoting
rated straight-chain fatty acids and pathway C generates unsat- the preexisting metabolism (vitalizer effect). It may persist by
urated branched isoprenoid acids. All three carbanionic species virtue of its egotistic feedback component even though the
may co-react. chemical conditions for its de novo synthesis may have vanished.
A crucial consequence of the formation of closed cellular Therefore, with every new dual feedback, the metabolism
structures is the automatic formation of an ion (H+) gradient switches into a new, relatively stable expanded state. This con-
across the membrane in conjunction with the appearance of stitutes a stabilizing memory effect. Extant cellular organisms
electron transport chains and a new kind of energy source: are replete with vitalizers, e.g., DNA, ribosomes, translocases,
chemiosmosis. It would at first have operated by coupling exer- coenzymes.
gonic and endergonic redox reactions while coupling with mem- Autotrophic metabolisms have a particular need for integra-
brane-bound ATPases would have appeared later. tion, since carbon flows from volcanic C1-compounds into
organic carbon skeletons of increasing size (C2 ! C3 ! C4).
In an expanding metabolism the traffic through these early steps
Evolutionary Connex with Extant Metabolism increases drastically and quickly causes a bottleneck, notably in
case of limitations of nutrients, transition metals, reductants,
Extant metabolisms are highly integrated with multifunctional and ligand feedback. This metabolic burden is today relieved by
catalysts and tightly interconnected pathways and cycles. By the a special kind of stabilizing feedback: The reductive citric acid
anabolist theory integration must have been at work from the cycle (RCC), which is really a thioester cycle. It is autocatalytic,
start. If a ligand effect promotes one pathway at the expense of all doubling with every turn, and potentially goes back to a very
others, the overall metabolism is weakened. This generates early primitive version, perhaps with thio acids instead of
a selective force for multi-catalytic ligand effects. Also, the less thioesters and FeS/H2S as reductant/catalyst (Wächtershäuser
integrated periphery of the metabolism would evolve faster than 1990). In view of its complex arithmetic (2 + 1 + 1 + 1 + 1 !
its center. 6 ! 4 + 2), a simpler version (2 + 1 + 1 ! 4 ! 2 + 2) has been
If a new pathway branches from a preestablished metabo- suggested, with cleavage at the malyl-thioester(acid) level
lism and a product of said new pathway feeds back into its (Wächtershäuser 1998a). Other extant autocatalytic CO2-
own synthesis, we speak of an egotistic or parasitic catalyst. It fixation pathways (Fuchs 2011) share with the RCC acetyl-
may be seen as the earliest form of a virus (virulyst). thioester and succinyl-thioester as common pivot points and
A dramatically different situation arises, if a product promise deep insights into early metabolic evolution. In extant
of a branch pathway exhibits a ‘‘dual feedback’’ effect: an metabolisms a protein cycle is operative with needed proteins
‘‘egotistic feedback’’ effect directly promoting its own synthesis being synthesized by the ribosome while unneeded proteins are
a O
–2[H] RHC
CO + H2N-CHR-COOH
aa NiS/H2S HN O c
b c CO2 c CO2 c CO2
aa aa-aa aa-aa2 aa-aan–1 aa-aan

CO CO CO CO CO
NiS/H2S NiS/H2S NiS/H2S
c h-aa h-aa2 h-aan–1 h-aan
H2O H2O H2O H 2O
u-aa u-aa2 u-aan–1 u-aan
aa H O aa H2O aa H2O aa H2O
2
c O
–2[H] RHC RHC COOH
+H2O
CO + H2N-CHR-CONHR′
NiS/H2S HN N R′ NiS/H2S HN NH R′
aa-aan
O h-aan O u-aan
. Fig. 4.4
Peptide cycle: (a) activating reaction; (b) concatenated cycles; (c) degradative reactions (aa = amino acid unit; h = hydantoin ring;
u = urea group; NHR0 = -(NH-CHR-CO)n-OH)
degraded by the proteasome. Early peptides are in an analogous origin of life may have been hyperthermophiles that evolved
situation. They are synthesized and degraded by the same cata- down the temperature scale with polyphyletic appearance of
lytic system, NiS/CO/H2S, thus forming a concatenation of mesophiles (Woese 1987).
peptide cycles (> Fig. 4.4) and a dynamic peptide library, The RNA world theory is committed to a mesophilic origin
which is self-selecting since peptides are not only stabilized by of life and makes the implicit assumption of symmetry between
binding to pyrite surfaces (Schreiner et al. 2011) or other cata- thermally upward and downward evolution. Temperature is all-
lytic transition metal species but also promote their own pervasive, profoundly influencing all structures and processes.
synthesis. A thermally upward adaptation could only occur for one (the
Homochirality unites biochemistry and sets it apart from the weakest) trait at a time, and by tiny increments. This has the
abiotic world. How did this occur? Extant RNA is homochiral, consequence that thermally upward evolution of whole organ-
yet the RNA world theory assumes a prebiotic broth of racemic isms is extremely slow (probabilistically forbidden). Thermally
or prochiral nucleotides, which means that any broth-born RNA downward evolution, by contrast, proceeds by opportunistic
would have been heterochiral and fallen stillborn from the losses of thermal traits, e.g., losses of CysS-metal bonds as
copying process. By the anabolist theory the racemic state of cross-linkers: evolution by degeneration rather than adaptation.
organic products is a benefit, because it broadens the space of Let us look at three main consequences: (1) Species that appear
feedback possibilities. Homochirality is a later ‘‘invention’’ of at successively higher branch points in the tree of life will have
life, perhaps forced in complementary fashion for L-amino successively fewer thermal stabilizers. (2) Organisms with lower
acids and D-ribose by RNA-peptide complexation (Carter and thermal status will harbor some traits of a higher thermal status
Kraut 1974). By the influence of chiral pyrite crystals enantios- (thermal relics). (3) A recently acquired low thermal status may
electivity may be predetermined and not a frozen accident be readily reversed (thermal atavism). This explains a striking
(Wächtershäuser 1992). fact concerning ancient proteins with metal-binding pairs of
CXXC signatures. Some of these have the same CXXC signatures
in all homologues, while others (e.g., aminoacyl synthetases and
Temperature, Phylogeny, and Time Frame of ribosomal proteins) show on average a decrease (but not an
Life increase!) of frequency of these signatures in the direction of
higher branch points. In the course of thermally downward evo-
The discovery of hyperthermophilic Bacteria and Archaea has lution such thermal traits may readily deteriorate one C at a time,
made room for the previously unthinkable possibility that the while in the course of thermally upward evolution the functionally
last universal common ancestor (LUCA) and by extension the required introduction of whole signatures (e.g., CXXC—CXXC)
into the right places by covariation would be highly improbable. gene clusters is statistically forbidden. Now, by overlapping
The earliest peptides would have bonded to transition metals alignment of gene clusters of different organisms, it was possible
(Woese 1972), and longer ones to extended mineral surfaces, to reconstruct a hypothetical segment of the genome of LUCA
and self-supporting folding structures would first have been (Wächtershäuser 1998b). It comprises genes for transcription,
acquired by local cross-linking to metals and much later by translation, and protein translocation. This means that LUCA
a multitude of distributed weak group interactions. had already a cellular lipid membrane, an organized DNA
Informational RNAs cannot fold well at high temperatures, genome, RNA, genetic code, modern sets of bases and amino
which means that the RNA theory requires a mesophilic or acids, and an extended metabolism for supporting all that.
psychrophilic origin (Moulton et al. 2000). The anabolist theory The above results support the proposal that organisms at the
allows for an origin of RNA at high temperature. It merely dawn of LUCA were ‘‘pre-cells’’ in the sense of being modern in
assumes that the earliest RNAs were bonded to surfaces and almost every respect, except for an intense lateral transfer of
only those with a proper sequence and high G + C content lifted metabolic genes, but not of genes for the genetic machinery
off as well-folded, self-supporting structures stabilized by (Kandler 1998). Owing to the primitivity of its enzymes LUCA
peptides. Later, with the formation of longer ribosomal proteins must have generated its chiral phosphoglycerol lipids as race-
by the incipient genetic machinery and of enzymes for stabiliz- mates. Racemic lipids are able to form stable membranes with
ing RNAs by secondary modifications, the restriction of the a patchwork of chirally segregated lipid domains and with cel-
RNA sequence space became more and more relaxed. By com- lular structures undergoing frequent fusions and fissions as in
parative analysis of RNA sequences (from which highly variable rapidly growing Thermococcus coalescens (Kuwabara et al. 2005),
positions were eliminated) it was possible to reconstruct ances- thereby generating a first subset of pre-cells with a predomi-
tral RNAs, with very high G + C contents for LUCA and also for nance of one lipid enantiomer and a second subset of pre-cells
the common ancestors of all bacteria, and all archaea, thus with a predominance of the other lipid enantiomer. Between
strongly supporting a hyperthermophilic origin of life these two subsets fusions would have been less probable than
(Di Giulio 2003). within each subset. They would have served as placeholders for
The oldest microfossils with remains of cell walls of uniform the later independent emergence of the phylogenetic domains of
thickness, carbon content with 13C-depletion and pyrite the Bacteria and Archaea by the appearance of enantioselective
deposits have been dated to 3.4 billion years ago (Wacey et al. enzymes for lipid synthesis—origin of speciation preordained in
2011), more than a billion years after the origin of the Solar the universal laws of physical chemistry. From this perspective
System (4.567 billion years ago). It appears from careful analysis the Eukarya are seen as the result of an endosymbiosis between
of Hadean crystals of extremely stable ZrSiO4 (zircon) found a population of bacterial host cells and a population of pre-cells
preserved as detrital particles in much younger sedimentary host as guests, that later turned into the cell nucleus (Wächtershäuser
rocks, that as early as 4.4 billion years ago the Earth had cooled 1992, 2003, 2006, 2010).
down sufficiently to bear an ocean of liquid water (Wilde et al.
2001; Mojzsis et al. 2001), which must have been hot. For the
anabolist theory this finding makes room for an early Hadean Explanatory Power and Explanatory Fallacies
origin of life, at a time when volcanic-hydrothermal sites must
have been rampant, thus providing enough time for biochemi- Explanatory power: The anabolist theory is based on the heuristic
cal/cellular evolution. The theory is compatible with and even of biochemical retrodiction (common precursor functions for
requires a hot origin for kinetic reasons (Stockbridge et al. 2010). disparate successor functions), which generates automatically
Hadean life could have survived intense bombardment with a backward convergence toward the pioneer organism of life
very large impactors simply by protection inside rocks and conversely explanatory power for numerous traits: concen-
thrown into orbit around the Earth (Cockel 2006). An RNA tric pattern of metabolism; FeS-proteins; metalloenzymes;
world origin would have had to wait until the Earth cooled organic ligands; inorganic ligands; organic sulfhydryl groups
down and heavy bombardment had subsided, i.e., until after with catalytic or redox functions; CO-fixation; multifunc-
3.8 billion years ago. Thus, by the empirical facts of geology an tionality of catalysts. Extant amino acid activation by ATP and
anabolist origin had plenty of time while the RNA world theory is synthetases may be seen as reversal of presumptive primordial
squeezed for time. formation of ATP by energy transfer from amino acids activated
Attempts to employ comparative analysis of polymer by COS or CO/H2S. Catalytic enzyme surfaces are seen as func-
sequences for reconstructing LUCA (as clue for the origin of tional successors of catalytic mineral surfaces. Bioorganic car-
life) are hampered by lateral gene transfer, notably of genes boxylate, thiolate, or phosphate groups are explained as derived
concerning the metabolism (Woese 1998). The situation may from mineral bonding mediators. Finally, extant digital
be remedied by comparing bacterial and archaeal genomes for sequence information is explained by a takeover from primor-
the presence of more or less extended conserved clusters of genes dial analog information with ribosomes operating as digital-to-
for multicomponent information processing systems. Lateral analogue converters.
transfer of a cluster of several genes encoding components of Explanatory fallacies: The theory of primordial RNA has no
such systems is expected to lead to functionally impaired or explanatory power, because the explanans is more complex than
corrupted hybrid systems. Therefore, lateral transfer of such the explanandum. The proposal of ribozymes as sole biocatalysts
nick A
rolling circle
replication
annealing
nick B sticky
end
sticky
end circularization
two identical plasmids
with displaced nicks
. Fig. 4.5
Hypothetical rolling circle replication of multiple LUCA DNA plasmids with nicks in both strands (takeover from rolling circle
transcription by expression of ribonucleotide reductase)
lost its explanatory power, with the discovery that the peptidyl Conclusion and Outlook
transfer center is predated by other nucleoprotein regions of the
ribosome. The proposal of RNA as hereditary polymer is explan- The RNA world theory posits the origin of life as highly improb-
atory for conversion of ribonucleotides to deoxyribonucleotides able one-time event that is knowable only in principle. By
and for RNA in telomerases, but not for RNA primers since contrast, the anabolist theory assumes an origin of life as rapid,
DNA primers work well in Archaea (Chemnitz Galal et al. 2012), chemically predetermined, singular, and directional process:
and not for viral RNA genomes since viruses lack cytosol. a universal chemical law of life’s origin. Fundamentally different
Nucleotidyl rests of coenzymes are better explained as activation experimental programs transpire from the two theories. Within
for covalent bonding to enzymes (Wächtershäuser 1992), rather the RNA world theory experiments have the role of rationalizing
than as relics of ‘‘handles’’ for RNA bonding. Finally, the partial aspects of the theory. The anabolist theory by contrast has
nonrelation between archaeal and bacterial DNA replicases is the ultimate experimental goal to recreate the pioneer organism
better explained by rolling circle replication of LUCA DNA in vitro and to watch it evolve.
(> Fig. 4.5) than by a LUCA RNA genome.
An imaginative proposal postulates a primordial seafloor
mound that grew like a chemical garden by repeated precipita- Acknowledgments
tion and bursting of an FeS membrane at the boundary between
hot, alkaline, H2S-laden vent water, and cold, acidic, Fe2+-laden I express my gratitude to Claudia Huber for dedicated and
ocean water, thereby acquiring an interior open-cell structure. ingenious experimental testing of the anabolist theory, to
Early evolution in its entirety supposedly happened inside this Helmut Simon, Adelbert Bacher and Mathias Groll for provid-
mound within the time frame of its growth: initial carbon ing laboratory facilities and support, to Deutsche Forschungs-
fixation (oceanic CO2 + vent H2) at the bottom, later RNA gemeinschaft for financial support, to Otto Kandler, the great
catalysis and replication in the middle, LUCA at the top, with catalyst without whom none of this would have come about, and
Bacteria and Archaea breaking loose and entering the ocean— to Dorothy Wächtershäuser for years of encouragement and
formation of ATP by chemiosmosis across the top membrane help in formulating this and other papers.
serving as continuous source of fuel (Martin and Russell 2003).
The proposal fails because (a) RNA cannot exist under the hot,
alkaline interior conditions, (b) an experimental FeS layer References
‘‘behaved more like a permeable reactive barrier than
Blöchl E, Keller M, Wächtershäuser G, Stetter KO (1992) Reactions depending on
a membrane’’ (Rickard and Luther 2007) and thus cannot sup-
iron sulfide and linking geochemistry with biochemistry. Proc Natl Acad Sci
port chemiosmosis, and (c) experimental FeS cells, the main USA 89:8117–8120
support of this proposal, turned out to be a freeze-drying artifact Carter CW, Kraut J (1974) A proposed model for interaction of polypeptides with
(Wächtershäuser 2006). RNA. Proc Natl Acad Sci USA 72:283–287
Cech TR (ed) (2011) The RNA worlds in context. Cold Spring Harb Perspect Biol. Mojzsis SJ, Harrison TM, Pidgeon RT (2001) Oxygen-isotope evidence from
doi:10.1101/cshperspect.a006742 ancient zircons for liquid water at the Earth’s surface 4.300 Myr ago. Nature
Chemnitz Galal W, Pan M, Kelman Z, Hurwitz J (2012) Characterization of the 409:178–181
DNA Primase complex isolated from the archaeon, Thermococcus Moulton V, Gardner PP, Pointon RF, Creamer LK, Jameson GB, Penny
kodakarensis. J Biol Chem 287(20):16209–16219 D (2000) RNA folding argues against a hot-start origin of life. J Molec
Cockel CS (2006) The origin and emergence of life under impact bombardment. Evol 51:416–421
Phil Trans R Soc B 361:1845–1875 Powner MW, Gerland B, Sutherland JD (2009) Synthesis of activated pyrimidine
Cody GD, Boctor NZ, Filley TR, Hazen RM, Scott JH, Sharma A, Yoder HS Jr ribonucleotides in prebiotically plausible conditions. Nature 459:239–242
(2000) Primordial carbonylated iron-sulfur compounds and the synthesis of Popper KR (1959) The logic of scientific discovery. Hutchinson, London
pyruvate. Science 289:1337–1340 Prakash CJ, Aldersley MF, Delano JW, Ferris JP (2009) Mechanism of montmo-
De Duve C (1991) Blueprint for a cell: the nature and origin of life. Neil Patterson, rillonite catalysis in the formation of RNA oligomers. J Am Chem Soc
Burlington 131:13369–13374
Di Giulio M (2003) The universal ancestor and the ancestor of bacteria were Rickard D, Luther GW III (2007) Chemistry of iron sulfides. Chem Rev 107:
hyperthermophiles. J Mol Evol 57:721–730 514–562
Dörr M, Käßbohrer J, Grunert R, Kreisel G, Brand WA, Werner RA, Geilmann H, Schreiner E, Nair NN, Wittekindt C, Marx D (2011) Peptide synthesis in aqueous
Apfel C, Robl C, Weigand W (2003) A possible prebiotic formation of environments: the role of extreme conditions and pyrite mineral surfaces on
ammonia from dinitrogen on iron-sulfide surfaces. Angew Chem Int Ed formation and hydrolysis of peptides. J Am Chem Soc 133:8216–8226
42:1540–1543 Schwartz AW (2008) Intractable mixtures and the origin of life. In: Herdewijn P,
Gesteland RF, Cech TR, Atkins JF (eds) (1999) The RNA world: the nature of Kisakürek MV (eds) Origin of life: chemical approach. Verlag Helvetica
modern RNA suggests a prebiotic RNA. Cold Spring Harbor Laboratory Chimica Acta, Zürich, pp 175–183
Press, Cold Spring Harbor Shapiro R (1986) Origins: a sceptic’s guide to the creation of life on Earth.
Fuchs G (2011) Alternative pathways of carbon dioxide fixation: insights into the Summit Books, New York
early evolution of life? Annu Rev Microbiol 65:631–658 Stockbridge RB, Lewis CA Jr, Wolfenden R (2010) Impact of temperature on the
Harish A, Caetano-Anollés G (2012) Ribosomal history reveals origin of modern time required for the establishment of primordial biochemistry, and for the
protein synthesis. PLoS One 7:e32776 evolution of enzymes. Proc Natl Acad Sci 107:22102–22105
Heinen W, Lauwers AM (1996) Organic sulfur compounds resulting from the Szostak JW (2011) An optimal degree of physical and chemical heterogeneity for
interaction of iron sulfide, hydrogen sulfide and carbon dioxide in an the origin of life? Phil Trans R Soc B 366:2894–2901
anaerobic aqueous environment. Orig Life Evol Biosph 26:131–150 Szostak JW, Bartel DP, Luisi PL (2001) Synthesizing life. Nature 409:387–390
Huber C, Wächtershäuser G (1997) Activated acetic acid by carbon fixation on Wacey D, Kilburn MR, Saunders M, Cliff J, Brasier MD (2011) Microfossils of
(Fe, Ni)S under primordial conditions. Science 276:245–247 sulphur-metabolizing cells in 3.4-billion-year-old rocks of Western Australia.
Huber C, Eisenreich W, Hecht S, Wächtershäuser G (2003) A possible primordial Nature Geosci 4:698–702
peptide cycle. Science 301:938–940 Wächtershäuser G (1990) Evolution of the first metabolic cycles. Proc Natl Acad
Huber C, Kraus F, Hanzlik M, Eisenreich W, Wächtershäuser G (2012) Elements Sci USA 87:200–204
of metabolic evolution. Chem Eur J 18:2063–2080 Wächtershäuser G (1992) Groundworks for an evolutionary biochemistry: the
Johnston WK, Unrau PJ, Lawrence MS, Glasner ME, Bartel DP (2001) RNA- iron-sulfur world. Prog Biophys Mol Biol 58:85–201
catalyzed RNA-polymerization: accurate and general RNA-templated primer Wächtershäuser G (1998a) The case for a hyperthermophilic, chemolithoau-
extension. Science 292:1319–1325 totrophic origin of life in an iron-sulfur world. In: Wiegel J, Adams MWW
Kandler O (1998) The early diversification of life and the origin of the three (eds) Thermophiles: the keyes to molecular evolution and the origin of life?
domains: a proposal. In: Wiegel J, Adams MWW (eds) Thermophiles: the Taylor & Francis, London, pp 47–57
keys to molecular evolution and the origin of life. Taylor & Francis, London, Wächtershäuser G (1998b) Towards a reconstruction of ancestral genomes by
pp 19–31 gene cluster alignment. Syst Appl Microbiol 21:473–477
Keller M, Blöchl E, Wächtershäuser G, Stetter KO (1994) Formation of amide Wächtershäuser G (2003) From pre-cells to eukarya — a tale of two lipids. Mol
bonds without condensation agent and implications for the origin of life. Microbiol 47:13–22
Nature 368:836–838 Wächtershäuser G (2006) From volcanic origins of chemoautotrophic origin
Kuhn H (1972) Selbstorganisation molekularer systeme und die evolution des of life to bacteria, archaea and eukarya. Phil Trans R Soc B Lond 361:
genetischen apparats. Angew Chem 84:838–862 1787–1808
Kuwabara T, Minaba M, Ogi N, Kammekura M (2005) Thermococcus coalescens Wächtershäuser G (2010) Chemoautotrophic origin of life: the iron-sulfur world
sp. nov., a cell-fusing hyperthermophilic archaeon from Suiyo Seamount. Int hypothesis. In: Barton LL, Mandl M, Loy A (eds) Geomicrobiology: molec-
J Syst Microbiol 55:2507–2514 ular and environmental perspective. Springer, Dordrecht, pp 1–35
Loison A, Dubant S, Adam P, Albrecht P (2010) Elucidation of an iterative process Wilde SA, Valley JW, Peck WH, Graham CM (2001) Evidence from detrital
of carbon-carbon bond formation of prebiotic significance. Astrobiology zircons for the existence of continental crust and oceans on the Earth 4.4
10:973–988 Gyr ago. Nature 409:175–178
Martin W, Russell MJ (2003) On the origin of cells: an hypothesis for the Woese CR (1972) The emergence of genetic organization. In: Ponnamperuma
evolutionary transitions from abiotic geochemistry to chemoautotrophic C (ed) Exobiology. North-Holland, Amsterdam, pp 301–341
prokaryotes, and from prokaryotes to nucleated cells. Phil Trans R Soc B Woese CR (1987) Bacterial evolution. Microbiol Rev 51:221–271
358:27–85 Woese CR (1998) The universal ancestor. Proc Natl Acad Sci USA 95:6854–6859
5 Morphological and Physiological Diversity
Stephen H. Zinder1 . Martin Dworkin2
1
Department of Microbiology, Cornell University, Ithaca, NY, USA
2
Department of Microbiology, University of Minnesota Medical School, Minneapolis, MN, USA
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Introduction
Diversity in Cell Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Over 5,000 species and nearly 1,000 genera of isolated prokary-
otes were tabulated as of 1999 (Garrity and Holt 2000). More-
Diversity in Cell Shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 over, studies examining 16S ribosomal DNA in natural
populations have provided convincing evidence that these cul-
Diversity in Cell Grouping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 tured organisms are just the ‘‘tip of the iceberg’’ with several
entire phyla/divisions having no or few cultured representatives
The Cell Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 (Hugenholtz et al. 1998). Prokaryotic diversity is an immensely
valuable resource—not only as a source of an almost infinite
Diversity in Capsules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 variety of metabolic capabilities, enzymes, and genes, but also as
a veritable cornucopia of strategies for dealing with the world. If
Cell Wall Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 what we wish to understand is not only how an organism
operates but also how what it does enables it to deal with an
Motility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 extremely variable and occasionally hostile environment, then
Swimming Motility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 the study of microbial diversity truly holds the answers.
Swarming Motility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Much attention has focused on understanding the properties
Twitching Motility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 of a relatively few organisms (such as Escherichia coli and some
Gliding Motility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 of its close relatives) in ever greater depth and detail that is
Motility in Archaea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 unmatched in any other cellular organism. This understanding
has tended to serve as a paradigm for understanding bacteria in
Cell Division Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 general, and in a larger sense, all organisms. Its ultimate expression
is attributed to Jacques Monod:SprS ‘‘What was true for
Developmental Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Escherichia coli would be true for the elephant’’ (Judson 1979),
although he was actually referring to an ‘‘old axiom’’ of A. J.
Metabolic Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Kluyver on the unity in biochemistry: ‘‘From elephant to butyric
acid bacterium—it is all the same’’ (Singleton 2000). In any
Anaerobic Fermentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 event, we have come to understand that though some funda-
mental strategies are indeed universal, each organism has
Anaerobic Respiration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 evolved its own combination of tactics to arrive at the solution
to its peculiar problems. It is thus appropriate, indeed often even
Aerobic Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 useful, to continue the pursuit of the unusual microbe, so
eloquently referred to by Ralph Wolfe in the foreword to this
Lithotrophy and Methanotrophy . . . . . . . . . . . . . . . . . . . . . . . . . . 110 treatise.
Among the many rewards of this pursuit has been the recent
Phototrophy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 recognition that, in addition to the diversity among the familiar
and commonly recognized bacteria, there is yet a higher level of
Autotrophic CO2 Fixation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 diversity. The existence of the Archaea has led to the recognition
of a group of prokaryotes that not only does not share many of
Nitrogen Fixation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 the properties considered fundamental to bacteria (Stanier and
van Niel 1962) but also has some properties, most notably
Adaptation to Environmental Extremes . . . . . . . . . . . . . . . . . . . 118 transcription, more closely resembling those of eukaryotes
(Reeve et al. 1997).
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
90 5 Morphological and Physiological Diversity
Prokaryotes (Monera)
Domain: Bacteria Archaea Eukarya
Methanogens
Bacteroides-
Flavobacterium
group
Thermotoga
group Protista
. Fig. 5.1
Phylogenetic tree based on 16S rRNA. The three domains are shown, as well as the five classically defined kingdoms (From Woese et al.
1990)
To get a sense of the evolutionary diversity of prokaryotes, Bacteria can show enormous diversity in cell size. For example,
one need only examine the universal 16S rDNA tree (> Fig. 5.1) one of us works with a disk-shaped organism, Dehalococcoides
to conclude that about two thirds of evolutionary diversity is ethenogenes, with a diameter of 0.4–0.5 mm and a height of
prokaryotic, and indeed most of eukaryotic evolutionary 0.1–0.2 mm so that its volume is near 0.024 mm3, about one
diversity is also microbial, mainly consisting of protists (Sogin twentieth that of E. coli. On the other hand, Epulopiscium
and Silberman 1998). Some ‘‘macrobiologists,’’ notably the emi- fishelsoni, found in the gastrointestinal tracts of certain fish, are
nent evolutionary biologist Ernst Mayr (1998), insist that the cigar-shaped cells with a diameter as great as 80 mm and lengths
phenotypic diversity of prokaryotes pales before that of eukary- up to 600 mm (Angert et al. 1993). The volume of these cells is
otes, which includes ‘‘jellyfish, butterflies, dinosaurs, humming- near 2 million mm3, about eight orders of magnitude larger than
birds, yeasts, giant kelp, and giant sequoias.’’ Woese in his reply of D. ethenogenes. Another example of prokaryotic gigantism,
(Woese 1998) justly states that Mayr is comparing apples and Thiomargarita, is a sphere (nearly a millimeter in diameter)
oranges, and that the human visual system has evolved to discern composed of a thin layer of cytoplasm around a large internal
differences in plants and animals as a matter of survival. Whereas, vacuole in which nitrate, concentrated from the surrounding
as we will demonstrate presently, there is ample morphological water, can reach concentrations as high as 0.8 M (Schulz et al.
variation in prokaryotes; their specialty is metabolic variation and 1999). This nitrate-filled vacuole allows the organisms to oxidize
adaptation to habitats and niches. Can eukaryotes use sulfur or sulfide under anaerobic conditions using the nitrate as a respi-
a pesticide as their lunch, fix nitrogen, or grow in boiling water? ratory electron acceptor. Curiously, no examples of archaeal cells
Clearly not. Indeed, some of Mayr’s naturalist colleagues at Har- more than a few mm in diameter are known (> Fig. 5.2).
vard now appreciate the importance of prokaryotes. Stephen Jay An advantage of existing as micrometer-sized cells is a greater
Gould (1996) states that ‘‘on any reasonable or fair criterion, surface-to-volume ratio. For example, if one compares spherical
bacteria are—and always have been—the dominant form of life cells with 1-mm and 10-mm diameters, it requires 1,000 of the
on earth,’’ and E. O. Wilson speculates at the end of his autobi- smaller cells (ca. 0.5 mm3) to equal the larger cell (ca. 500 mm3) in
ography (Wilson 1995) that were he to start over again in the volume. Moreover, the smaller cells have 10 times greater total
twenty-first century, he would become a microbial ecologist. surface area (ca. 3,140 mm2) than the larger one (ca. 314 mm2),
It is not our intention in this chapter to construct a bestiary. allowing them to present more surface to the environment for
Nor is it our goal to provide an exhaustive description of pro- uptake of scarce nutrients. The smaller size apparently obviates
karyotic diversity; in effect, that is the entire content of this the need for the organellar compartmentation and internal
book. Our choice of examples of diversity has been selective, membrane systems generally found in eukaryotes because mol-
and this chapter shall be limited to using a variety of prokaryotes ecules can traverse the distances in the prokaryotic cell more
to illustrate the diversity of structural, physiological, and meta- easily by diffusion. Indeed, the cells of Epulopiscium contain an
bolic strategies used by bacteria to adapt to the world. extensive internal membrane system (Clements and Bullivant
1991), indicating that their large size necessitates such a system.
Diversity in Cell Size

Diversity in Cell Shape
The conventional wisdom that prokaryotic cell diameters are
typically near 1 mm, whereas eukaryotic cells are 10 mm or larger The controversy between the pleomorphists and the
is generally true, but significant exceptions exist. Members of the monomorphists led to a view of bacterial morphology first
Morphological and Physiological Diversity 5 91
consequent increase in surface area available for nutrient uptake.

It has also been suggested (Pate and Ordal 1965) that the char-
acteristic stalks of Caulobacter (and perhaps also Asticcacaulis)
function to increase the surface area of the cell yet further.
Conversely, one may suggest that organisms that characteristi-
cally find themselves under conditions of ample nutrient levels
would not be disadvantaged by low S/V ratios and may even find
Epulopiscium
spherical shapes advantageous for other reasons. For example,
one could imagine that an organism whose cell surface was
subject to assault by antibodies or other host-defense factors
E. coli might evolve a coccal shape that would minimize the amount of
accessible surface.
It is conceivable that the shape of a cell may also be related to
optimal hydrodynamic properties. It may not be a coincidence
that motile organisms are rarely spherical; the number of flagel-
lated cocci is quite small. Gliding bacteria are generally rod
Paramecium shaped (> Fig. 5.6) although an interesting exception to this is
found in Isosphaera pallida, a budding, chain-forming, gliding
coccus (> Fig. 5.4). In addition, the spiral shape of organisms
. Fig. 5.2
such as spirochetes may be causally related to the ability of the
Comparison of Epulopiscium with cells of the protist Paramecium
organism to screw its way through a relatively highly viscous
and E. coli (Photo courtesy of E. Angert)
medium. In fact, it has been shown that the motility of spiro-
chetes is enhanced at higher than normal viscosities (Canale-
propounded by Ferdinand Cohn (1875) and hammered into Parola 1978).
a dogma by the subsequent work of Robert Koch. The classical It was appropriate at one time to view any departure from
textbook view of bacteria described them as spherical or cylin- the rod-coccus-spiral morphology as a reflection of abnormal
drical, the latter category divided between straight and curved conditions. It is now quite clear that there are a variety of cell
rods. This view, based on the observations made on laboratory- morphologies, each of which is uniquely appropriate for the
grown cultures cultivated under conditions designed to result in ecological niche of that cell.
a maximum growth rate and morphological uniformity, came to
represent a picture of the ‘‘true’’ morphology of bacteria. It
obscured the spectrum of enormous morphological diversity Diversity in Cell Grouping
manifested by the bacteria. It also obscured the fact that that
diversity was often the result of life cycles that represented, for It is also interesting to think about the functions of cell group-
bacteria, survival strategies in addition to those of maximizing ings. Is it possible that the characteristic clustering of the staph-
growth rates. ylococci further reduces the amount of accessible cell surface?
The shape of a cell is not a trivial or casual aspect of its And what is the conceivable function of the striking tendency of
adaptation to its environment but is instead a strategic conse- Thiopedia and Lampropedia (the latter a chemotrophic aerobe
quence of this adaptation. Nor is the shape of a cell limited in any and the former a phototrophic anaerobe) to grow as sheets of
fundamental way to a rod, coccus, or spiral. Ferdinand Cohn in cells (> Fig. 5.3)?
1875 divided bacteria into these three forms in one of the first One instance where a characteristic grouping has been ratio-
systematic attempts to classify bacteria. It is now quite clear that nalized is in the myxobacteria. These organisms undergo
the variety of cell shapes and groupings far exceeds those offered a complex life cycle and feed on insoluble macromolecular
in early views. There are bacteria that are amorphous, ovoid, debris by excreting a complex of hydrolytic enzymes. They are
square, stellate, filamentous, or stalked (> Fig. 5.3). They may be thus at the mercy of diffusion both of their enzymes away from
grouped as pairs, clumps, chains, rosettes, cuboid packets, the cell and of the hydrolyzed nutrients back to the cell. How-
flat squares, networks, mycelia, or fruiting bodies (> Figs. 5.4, ever, the complex life cycle of these organisms which causes
5.5, 5.6). them to periodically collect into densely packed fruiting bodies,
What are the possible functions of cell shape and grouping? as well as to form swarming, gliding masses of cells (> Fig. 5.6b),
Certainly one of the most obvious relates to the surface-volume guarantees a high local concentration of periodically excreted
(S/V) ratio of a cell referred to above in the context of cell size. enzymes. This has been shown to optimize their feeding
The sphere has the lowest possible S/V ratio, and as a cell (Rosenberg et al. 1977) and has been referred to as a ‘‘microbial
becomes longer and thinner, its S/V ratio increases. One can wolf-pack effect’’ (Dworkin 1973).
imagine that a cell such as the oligotrophic Caulobacter, whose Though the collection of resting cells compressed in the
lifestyle involves the scavenging of extremely low concentrations fruiting body represents the potential for releasing a high-
of nutrients, will benefit from a rod shape, and from the density population once germination occurs, one can only
a b
. Fig. 5.3
A variety of cell shapes found among prokaryotes. (a) Prosthecomicrobium. (b) Simonsiella. (c) Cylindrospermum. (d) Square
bacterium. (e) Rhodomicrobium. (f) Asticcacaulis. (g) Thiopedia rosea (a from Staley 1968; b from Pangborn et al. 1977; c from Stanier et al.
1981; d from Walsby 1980, with permission; f from Pate and Ordal 1965, with permission; e courtesy of P. Hirsch; and g courtesy of
N. Pfennig)
the nature of attachment to a substrate or to another cell. The

cell surface also determines what gets into and out of the cell and
GV the nature of that transport process.
bud Diversity in Capsules
Microbiologists have always had a fondness for working with

organisms that grow in liquid in a nicely dispersed state. This has
tended to obscure the fact that, in nature, most bacteria are not
growing in a dispersed state but are clinging tenaciously to
a substratum as a clump of cells or in a biofilm. Many organisms
are surrounded by a more or less amorphous layer of slime,
generally designated as the glycocalyx.
Furthermore, as was pointed out originally by Whittenbury
and Dow (1977) and later emphasized by Costerton et al. (1986),
the physiological properties of planktonic bacteria growing in
a dispersed state are substantially different from those of the
same cells when they are in their sessile state, embedded in
a matrix of glycocalyces as pure- or mixed-culture biofilms
(Costerton et al. 1999).
The principal component of the glycocalyx may be either
polypeptide, as is the case with a number of species of Bacillus, or
polysaccharide, characteristic of both Gram-positive and Gram-
. Fig. 5.4 negative organisms. These layers have a variety of functions
Isosphaera pallida (From Giovannoni et al. 1987) Arrows denote depending on the organism and the nature of the layer. They
new cells growing as buds forming between cells, and gv denotes serve in some cases to facilitate attachment to surfaces, either
gas vesicles cellular or otherwise, and the nature of the capsule must, in
some way, determine the kind of substratum to which the cell
can attach.
speculate as to the function of fruiting body complexity. It is The capsule has been shown to interfere with the function of
feasible, however, that each germinated cluster of cells has an the phagocytic cells of the immune system and may form
optimal size, and that the number of cells in each of the a barrier to phage adsorption. Cook and Colvin (1980) have
sporangiole packages arrayed on the fruiting body determines shown that the cellulosic fibers produced by Acetobacter serve to
this size. maintain these highly aerobic organisms at the liquid-air inter-
There are other groupings that are familiar, for example, face. Finally, it has been suggested that the hygroscopic quality of
chain formation by cocci or by rods (> Fig. 5.7a, b), clusters the glycocalyx may help a cell to resist desiccation. Based on such
(e.g., Stomatococcus, > Fig. 5.7c), sheathed cells (e.g., a variety of functions that a glycocalyx may play, it is no surprise
Sphaerotilus, > Fig. 5.7e), mycelium formation (e.g., Streptomy- that the composition of this structure varies so widely.
ces, > Fig. 5.7f), rosette formation (e.g., Caulobacter, Another aspect of capsular diversity relates to the role of
> Fig. 5.7d), cuboidal packets (e.g., Sarcina, > Fig. 5.7g), and a capsule in interfering with the ability of phagocytic cells to
networks (e.g., Rhodomicrobium, > Fig. 5.3). These groupings engulf their bacterial prey. Certainly the ability of the pneumo-
are probably not casual or accidental properties of the cells but coccus to manifest a variety of capsular polysaccharide types
are likely functional and should be viewed as other manifesta- plays an important role in the evasion of host defenses by the
tions of the diversity of prokaryotic adaptation to the environ- bacterium; the generation of diversity of the cell surface is
ment. Understanding how these various groupings contribute to a general strategy for compromising host-defense mechanisms.
a cell’s fitness lingers as one of the challenges of structure-
function studies.
Cell Wall Diversity
The Cell Surface In Stanier and van Niel’s classic description of the prokaryotic
cell (Stanier and van Niel 1962), one of the few positive charac-
The interface between a bacterial cell and its surroundings is teristics of prokaryotes (as opposed to lack of nucleus, mito-
where the cell will first have to deal with a variable environment. chondria, etc.) was the presence of a muramic acid-containing
It is thus the part of the bacterial cell where one might expect to peptidoglycan cell wall. Indeed, because this structure is not
find the greatest diversity. It is the cell surface that determines present in eukaryotes, its synthesis is a primary target of several
. Fig. 5.5
Fruiting bodies of myxobacteria. (a) Myxococcus fulvus. (b) Myxococcus stipitatus. (c) Stigmatella aurantiaca (a and b from Reichenbach
and Dworkin 1981b; c courtesy of H. Reichenbach)
. Fig. 5.6
Gliding bacteria. (a) Nannocystis exedens. (b) Myxococcus fulvus. (c) Cytophaga sp. (d) Flexibacter elegans. (e) Vitreoscilla stercoraria.
(f) Lysobacter sp. (g) Herpetosiphon giganteus (From Reichenbach and Dworkin 1981a)
important antibiotics, such as the penicillin/cephalosporin nearly all members of the Bacteria do indeed have a muramic
group. At that time, cell walls were divided into Gram-positive, acid-containing cell wall, many have protein cell walls, such as
with its thick peptidoglycan layer, and Gram-negative, with planktomycetes and chlamydia (Weisburg et al. 1986). The
a thin peptidoglycan layer and outer membrane (Beveridge Gram-negative cell wall seems to be the default phenotype in
1995). Stanier and van Niel were aware of mycoplasmas and the Bacteria, being found in most branches including the deeply
similar organisms lacking a cell wall. It is now clear that though branching Aquificae and Thermotogae. Moreover, even some
. Fig. 5.7
A variety of cell arrangements among the prokaryotes. (a) Phase-contrast micrograph of chains of Streptococcus lactis. (b) Chain of
rod-shaped cells of Lactobacillus gasseri. (c) Irregular cluster of Stomatococcus mucilaginosus. (d) Rosette of Caulobacter. (e) Motile rods
within a sheath, Sphaerotilus natans. (f) Branched filament, Streptomyces sp. (g) Three-dimensional packet of cocci, Sarcina maxima.
(h) Two-dimensional packet of cocci, Amoebobacter pedioformis (a courtesy of T. D. Brock; b from Kandler and Weiss 1986; c from
Schleifer 1986; d from Poindexter 1964; e from Stokes 1954; f from Kutzner 1981; g from Holt and Canale-Parola 1967; and h from
Eichler and Pfennig 1986)
deep branches of the Firmacutes contain organisms (such as condition that the substance in question is soluble rather than
Sporomusa and Desulfotomaculum) with Gram-negative cell particulate, and that the milieu is thus an aqueous one. This
wall structures. On the other hand, there are those who propose serves to rationalize the observation that all the bacteria for
that the Gram-positive cell wall is the default structure (Gupta which chemotaxis has been demonstrated are motile by means
1998). of flagella, with which they swim through an aqueous milieu. In
Arguably, the default cell wall architecture in the Archaea is fact, it is reasonable to suggest that the only conceivable function
the protein S-layer, a paracrystalline two-dimensional array of of swimming motility in bacteria is to allow the cell to travel
protein subunits (Sleytr et al. 1993). S-layers also are found along a perceived concentration gradient. The importance for
outside the cell wall in many bacteria. In the genera chemotaxis and motility to the hunt for food is underlined by
Methanosarcina and Halococcus, there is a thick fibrous polysac- the finding that many naphthalene-degrading Pseudomonas
charide layer outside the S-layer, apparently serving as a corset to strains possess plasmids which carry both the necessary biodeg-
prevent lysis at relatively low osmolarities (Kandler 1994; Sowers radation genes and genes encoding chemotaxis toward naphtha-
et al. 1993). In two methanogenic orders, the Methanobac- lene (Grimm and Harwood 1997), thereby forming
teriales and Methanothermales, the cell walls consist of a thick a biodegradative package. Similarly, prokaryotes can position
peptidoglycan layer called ‘‘pseudomurein’’ which causes them themselves in a gradient of light, a phenomenon called
to stain Gram-positive. Pseudomurein is similar to eubacterial ‘‘phototaxis.’’
murein but differs in a number of interesting aspects. It contains
N-acetyl talosamineuronic acid instead of N-acetyl muramic
acid, the carbohydrate that, along with N-acetyl glucosamine, Swimming Motility
represents the carbohydrate backbone of murein. The amino
acids in the peptide chain of the archaebacteria are all L-isomers In all cases where they have been examined carefully, flagella in
and the glycosidic linkage is a-1,3 rather than 1,4. the Bacteria are fundamentally similar. They are usually com-
posed of one, and in some cases two or three, species of a self-
assembling protein, generically designated as ‘‘flagellin.’’ This is
Motility attached, via a hook-shaped adaptor, to a motor-like basal body
whereby the flagellar filament is inserted into the cell membrane.
Motility by multicellular eukaryotes serves a number of different The flagellin subunits are translocated down the center of the
functions. It enables an organism to escape from a predator or flagellar shaft to the distal growth point where flagella are syn-
some other threat, it facilitates movement toward a potential thesized. An interesting adaptive use of this system is type III
mate, it allows developmental movement, and perhaps most secretion found in pathogens, in which an assembly with protein
importantly, it allows an organism to seek out and move toward subunits homologous to those in the flagellar filament is used to
a supply of food. inject toxic proteins directly into eukaryotic host cells (Macnab
In the case of prokaryotes, it seems unlikely that cells can 1999). Diversity is manifested by the number and arrangement
perceive the presence of predators such as protozoa, leukocytes, of the flagella around the cell. This varies from the polarly
slime molds, phage, or Bdellovibrio sufficiently in advance to flagellated cell with a single flagellum (> Fig. 5.8) to the pro-
allow them to exercise an avoidance strategy. Bacteria, however, fusely and peritrichously flagellated cells (> Fig. 5.11). In
can detect gradients of repellent chemicals and respond by between, there are cells with bipolar flagellation (> Fig. 5.9) or
moving away (see below). with polar tufts of flagella (> Fig. 5.10).
With regard to mating, it does not seem that motility to A most unusual style of flagellation is found among the
facilitate mating interactions occurs among the bacteria. In spirochetes. Rather than being freely turning propellers, the
those cases where a mating pheromone has, in fact, been dem- flagella are enshrouded by an outer cell envelope. It has been
onstrated (Dunny et al. 1978), the participating organisms are suggested (Berg 1976) that their motion within this outer sheath
nonmotile and depend on random motion to effect contact. generates a torque that, in combination with the helical structure
The function of the pheromone is to prepare the organisms of the cell, propels the cell in a screw-like fashion. It has been
for the events involved in adhesion and gene transfer. Prokary- shown that, for certain of the spirochetes (Canale-Parola 1978),
otic movement to accomplish a developmental event seems flagellar motility is enhanced in a high-viscosity medium. This
to be limited to the myxobacteria, where it allows the cells peculiar variation on flagellar organization may have evolved in
to maintain a high population density and to move into aggre- response to high-viscosity environments that would hamper or
gation centers as a prelude to fruiting body formation prevent ordinary flagellar motility, allowing the spirochetes to
(Spormann 1999). move through viscous media like a corkscrew through a cork.
The most common function of motility among the prokary- Interestingly, spirochete motility seems to have at least three
otes seems to be to allow the organisms to position themselves modes: swimming forward, swimming backward, and flexing
optimally in their microenvironment, for example, to move (Fosnaugh and Greenberg 1988).
along a concentration gradient toward a food source or away Only flagellated prokaryotes move, in effect, via a rotating
from a repellant (see Chap. 15, ‘‘Bacterial Behavior’’ in Vol. 2). propeller principle. In this sense it is completely unlike the
The notion of a concentration gradient has implicit in it the movement of conventional eukaryotic flagella, whose action is
. Fig. 5.8 . Fig. 5.10

Polarly flagellated cell of Pseudomonas andropogonis. Flagellar fascicle of Spirillum volutans (From Kreig 1984)
Bar = 0.2 mm (From Palleroni 1984)
. Fig. 5.9 . Fig. 5.11

Bipolarly flagellated cell of a magnetotactic spirillum (From Peritrichously flagellated swarmer cell of Proteus mirabilis (From
Balkwill et al. 1980) Hoeniger 1965)
based instead on the sinusoidal beating of the flagellum within for gliding motility have been proposed (Burchard 1984). They
a single plane. One might ask either how the bacterium manages range from moving tracks (Lapidus and Berg 1982), rotating
to operate an intracellular, rotating structure or, on the other organelles (Pate and Chang 1979), polarized excretion of surfac-
hand, why there are no fish with propellers. The answer may lie tants (Keller et al. 1983), and contractile fibers (Burchard 1984).
in two features of the biophysics of movement. First, the effi- No one of these proposals is supported by other than those who
ciency of converting input energy to thrust is about 96–98 % for have suggested them, and it now appears that there may be more
an oscillating, flexible foil (such as the eukaryotic flagellum; than one mechanism of gliding (Spormann 1999).
Katz and Weihs 1979); this is substantially higher than what The gliding bacteria, though customarily grouped together,
can be obtained with a propeller-type mechanism. Second, the have in common only the facts that they move by gliding and
bacterium does not have this option; the tremendous viscous that they all have Gram-negative cell wall architecture. The
drag exerted on a small cell would damp out the oscillations of motility mechanisms seem to be different, and the organisms
a sinusoidally beating flagellum. A screw-like motion is its solu- cross the entire spectrum of physiological types.
tion to the need to move through an aqueous but (from the While there is no hard evidence, it seems appropriate that
viewpoint of the bacterium) tar-like milieu. those organisms whose nutrition depends on the extracellular
solubilization of particulate material would glide rather than
swim to gain access to their substrates. Though this seems to
Swarming Motility be the case for the myxobacteria, it is not clearly the case for the
cyanobacteria or other gliding phototrophs.
Certain species of Proteus and of Vibrio carry out a kind of
motility called ‘‘swarming’’ (Williams 1978) that is an alternative
to their normal, swimming motility. Under certain conditions, Motility in Archaea
usually the perception of a solid surface or of an increased
viscosity (in the case of Vibrio), the cell morphology and the Some archaea show a flagellar-based motility, though gliding
manner of flagellation change. The cells become longer and, in motility has never been detected in the Archaea. Whereas the
the case of Proteus, filamentous, and the sparsely or polarly archaeal flagellar assembly superficially resembles those of the
flagellated cells become profusely peritrichous (> Fig. 5.11). In Bacteria, it lacks a hook region, and, interestingly, the amino
Proteus, when the cells are placed on an agar surface under the acid sequences of the subunits of the flagellar shaft resemble
appropriate conditions, the swarming phase is periodic and those of type IV pili in bacteria, rather than those of flagellins
alternates with a growth phase, during which shorter, less exten- (Jarrell et al. 1996). Moreover, whereas bacterial flagellins are
sively flagellated swimming cells are formed (Matsuyama et al. translocated outside the cell via a sec-independent transport
2000). system involving transport through the interior of the flagellum
shaft, the flagellar subunits in the Archaea are transported using
a standard sec-dependent system (Jarrell et al. 1996). Homo-
Twitching Motility logues of methyl-accepting chemotaxis proteins and the CheA-
CheY two-component signal transduction system (Rudolph and
Certain of the Gram-negative cocci and coccobacilli, such as Oesterhelt 1996) have been found in the haloarchaea. Indeed,
Neisseria, Moraxella and Acinetobacter, manifest a peculiar type the haloarchaea have special sensory rhodopsins that, instead of
of movement known as ‘‘twitching’’ (Henrichsen 1983). The pumping ions as do bacteriorhodopsin and halorhodopsin,
cells exhibit jerking or jumping movements of a few microme- couple with the methyl-accepting chemotaxis proteins to form
ters in any direction. It now appears that retraction of type IV a phototaxis system (Perazzona and Spudich 1999).
pili is responsible for this motility (Merz et al. 2000).
Cell Division Strategies

Gliding Motility
Bacterial multiplication is usually thought of as being synony-
There are, however, alternative mechanisms for motility in pro- mous with binary transverse fission, the division strategy used
karyotes. An entire group of bacteria, colloquially termed ‘‘the by most of the commonly studied bacteria. This process has been
gliding bacteria,’’ move by an undefined mechanism. These intensively investigated, and, though many of the regulatory
organisms are able to glide across solid surfaces such as agar processes of division remain unknown, the descriptive details
and occasionally glass or plastic. The mechanism or mechanisms of the process are clearly defined (Ingraham et al. 1983). There is,
of such motility are not understood. The cells do not inch, however, a variety of other division strategies found among the
wiggle, or sidle. There seem to be no organelles of locomotion, bacteria, once one strays from the old laboratory standbys. Thus,
although various authors have, on occasion, seen structures that division by budding among such organisms as Hyphomicrobium
they have interpreted as locomotory organelles (Lünsdorf and and Rhodomicrobium (> Fig. 5.3e), by mycelial extension and
Reichenbach 1989; Pate and Chang 1979). Various mechanisms subsequent fragmentation among the filamentous
. Fig. 5.12
Multiple fission and release of baeocytes from Dermocarpa. The numbers represent the hours that have elapsed during the growth of the
initial, small baeocyte (From Stanier et al. 1981)
actinomycetes (> Fig. 5.7f), and by multiple fission among the

pleurocapsalean cyanobacteria (> Fig. 5.12) has been described
among the bacteria.
While it does seem a useful and interesting exercise to won-
der what particular advantages are associated with the different
strategies for cell division, this is not a subject that has received
a great deal of attention from microbiologists. Nevertheless, it
does seem obvious that these different mechanisms of cell divi-
sion must be functionally related to the larger aspects of the
biology of the cell. Thus, for example, mycelial extension
followed by fragmentation seems a reasonable way to alternate
filamentous and single-celled modes. And the unequal division
of the stalked cell of Caulobacter (> Fig. 5.13) leading to the
generation of a flagellated swarmer cell must certainly reflect the
division of the organism’s activity between a sessile, stalked stem
cell and a motile cell that may expand the territory occupied by
that clone. Certainly, other speculations are possible to rational-
ize other modes of prokaryotic cell division.
Developmental Diversity
Chapter 16, ‘‘Prokaryotic Life Cycles’’ in Vol. 2 presents a more

extensive discussion of the traditional view (and a more detailed
description) of development among bacteria. The purpose of
this section is to point out that development represents an
additional set of strategies for diversity.
Development implies a set of alternative states. These alter-
natives may be expressed as a function of time, in which case we
think of the alternative states as parts of a life cycle; on the other . Fig. 5.13
hand, the heterogeneity may be spatial, with different parts of Generation of swarmer cell from stalked cell of Caulobacter
the cell differentiated so as to fulfill specific functions. An crescentus (From Poindexter 1964)
example of the latter would be the stalked cell of Caulobacter

(> Fig. 5.13). In the case of either temporal or spatial alterna-
tives, prokaryotic development generates considerably greater
diversity than was afforded by the traditional nineteenth-
century view of bacteria as rods, cocci, or spirilla.
The generation of diversity among bacteria by developmen-
tal morphogenesis or by differentiation expands the kinds of
niches that the bacteria can occupy. As an example, the ability of
Caulobacter to form both a swarmer cell and a stalked cell allows
it to separate the processes of growth and reproduction from
dispersal. As an oligotroph, that is, an organism designed to feed
on relatively low concentrations of nutrients, its ability to seek
out, detect, and move toward such low concentrations must be
optimized. That is the presumptive function of the swarmer cell.
Once such a site has been discovered, it is the function of the
stalked cell to attach to a surface and begin the process of
feeding, growth, and reproduction. This sort of alternation
between a sessile, feeding and reproductive stage and
a nongrowing, swarming stage is also characteristic of organisms
such as the actinoplanes, in which a mycelial phase alternates
with a motile zoospore-like cellular stage. This type of develop-
mental strategy facilitates the colonization of sites that might
otherwise be inaccessible or inappropriate.
The characteristic swarm of the myxobacteria is another
developmental strategy used by a bacterium to optimize its . Fig. 5.14
feeding. The myxobacteria feed by excreting a battery of potent Swarm edge of Stigmatella erecta at low magnification.
hydrolytic enzymes that degrade proteins, peptidoglycan, poly- Bar = 110 mm (From Reichenbach 1984)
saccharides, lipids, and nucleic acids. The cells are thus at the
mercy of the process of diffusion of enzymes away from the cell
and of low-molecular-weight products of hydrolysis toward the
cell. The myxobacteria optimize this process by traveling as Sporolactobacillus, Sporosarcina, and Thermoactinomyces
a swarm—a high population density of cells (> Fig. 5.14). The (> Fig. 5.15). In all of these organisms, the essential features of
rate of cell growth on such substrates increases as a function of the spore and of the process of sporulation are quite similar.
the cell density—a reflection of the fact that the cells are feeding Another type of sporulation among the Gram-positive bacteria
cooperatively (Rosenberg et al. 1977). This sort of microbial is found among the actinomycetes. Actinospores include
‘‘wolf-pack effect’’ has been suggested to rationalize the overall the nonmotile conidial spores of Streptomyces (> Fig. 5.16),
life cycle of the myxobacteria (Dworkin 1973). the sporangiospores of Actinoplanes (> Fig. 5.17), and the endo-
The ability of organisms to convert from a vegetative, grow- spores of Thermoactinomyces (> Fig. 5.18), which, despite their
ing stage to a resting stage allows them to persist in environ- traditional taxonomic distance from the Bacillaceae, share many
ments that might periodically become inhospitable to normal, of the properties of Bacillus endospores. In this context, it is
prokaryotic life. As was pointed out earlier, the ability to grow in interesting that molecular techniques have revealed a moderate
an extreme environment is invariably at the expense of the but distinct phylogenetic relationship between Bacillus
ability to grow in the so-called normal environment. However, stearothermophilus and Thermoactinomyces vulgaris
morphogenesis to a resistant, metabolically quiescent resting cell (Stackebrandt et al. 1987).
allows the organism to survive in the absence of growth until The exospores of Gram-negative bacteria are, in general, less
conditions are once again suitable for growth. A remarkable resistant to environmental extremes than are endospores. They
feature of this adaptation is that the state of almost total quies- are, however, metabolically quiescent and substantially more
cence is juxtaposed with the ability to respond to the resistant to desiccation, physical breakage, and environmental
reappearance of conditions suitable for growth by germinating extremes than the corresponding vegetative cells. Among the
in an almost hair-trigger fashion. Gram-negative bacteria, resistant resting cells that have been
One particular type of spore is referred to as an ‘‘endospore’’ fairly well characterized are found among the myxobacteria, as
and is found among the Gram-positive bacteria. This resting cell myxospores; among the cyanobacteria, as akinetes; and among
is formed within the vegetative cell and is usually released as Azotobacter and related genera, as azotocysts. Less well charac-
a free spore as the process of sporulation is completed. terized resting cells have been described for the photosynthetic
Endosporulation occurs widely across traditional taxonomic bacterium Rhodomicrobium (Dow and Whittenbury 1979) and
lines and is found among the genera Bacillus, Clostridium, some of the methylotrophic bacteria (Whittenbury et al. 1970).
. Fig. 5.15
Endospores of (a) Bacillus fastidiosus, (b) Clostridium butyricum, (c) Sarcina ureae, and (d) Thermoactinomyces dichotomicus (a courtesy of
S. C. Holt; b from Gottschalk et al. 1981; c from Sneath 1986; and d from Cross 1981)
Myxobacterial resting cells are contained in the characteristic

fruiting bodies formed by the myxobacteria (> Fig. 5.5), and
their shapes vary from the round, resistant, optically refractile
cells formed by the genus Myxococcus (> Fig. 5.19a) to the slightly
shortened, oval rods formed by Stigmatella (> Fig. 5.19b).
The myxospore of Myxococcus xanthus is formed by the short-
ening and rounding up of the entire rod-shaped vegetative cell
and is finally enclosed by a multilayered spore coat/capsule
(White 1984). The Myxococcus xanthus myxospore is the only
myxobacterial resting cell that has been well characterized both
structurally and biochemically.
The cyanobacteria form a variety of cell types that have been
considered to be spores or resting cells. These include the
hormocysts of Westiella, the exospores of Chamaesiphon, the
endospores or baeocytes of the Pleurocapsales (> Fig. 5.12),
and the akinetes of Anabaena and other genera. The
cyanobacteria, in general, have not been characterized from
a physiological or biochemical point of view with the same
intensity that many of the eubacteria have. Thus, information
about the properties of cyanobacterial resting cells is quite
sparse. Most of the available information centers on the akinetes
of Anabaena, and even in that case, the difficulty of obtaining
large populations of relatively pure akinetes has limited the
available information (Nichols and Adams 1982). While the
. Fig. 5.16 akinetes of Anabaena are more resistant to extreme low temper-
Phase-contrast micrographs of vegetative hyphae and ature and desiccation than the corresponding vegetative cells,
conidiospores of Streptomyces coelicolor. CW cross walls, IS their respiratory rate is twice that of the vegetative cells.
immature spores, and MS mature spores. Bar = 10 mm (From > Figure 9.20 illustrates the akinetes of Anabaena.
Chater and Hopwood 1973)

. Fig. 5.17
Electron micrographs of thin sections of sporangiospores and sporangia of Actinoplanes. (a) Mature sporangium, 4 days old. (b, c, and d)
Immature sporangium, 2 days old. (e, f), and (g) Mature sporangium, 4 days old (From Lechevalier and Holbert 1965)
Cells of the Gram-negative, nitrogen-fixing genus Azotobac- more resistant to desiccation and seem to be truly metabolically
ter are able to convert to cysts when the cells have exhausted the quiescent. (See Sudo and Dworkin (1973) for an extensive,
nutrients of the growth medium (Sadoff 1976). The resting cells comparative survey of prokaryotic resting cell properties.)
also can be induced by placing the cells in a medium containing Another interesting type of life cycle is shown by intracellular
hydroxybutyrate as the carbon source. Azotobacter cysts pathogens of the chlamydia group. Cells of the chlamydia must
(> Fig. 5.21) are superficially similar to myxospores, in that be able to carry out three very different kinds of processes.
they are only slightly more resistant to temperature extremes Because they are not transmitted by any sort of a vector, but
than the corresponding vegetative cells but are considerably may exist free in the environment, they must be able to persist in
on bacterial metabolism (also in the third edition), written well

before the modern era of molecular biochemistry or bacterial
genetics. In it she states:
‘‘Bacteria may be tentatively regarded as biochemical exper-
imenters; owing to their small size and rapid growth, variations
must arise very much more frequently than in more differenti-
ated forms of life, and they can in addition afford to occupy
more precarious positions in the natural economy than larger
organisms with more exacting requirements.’’
In terms of anabolism, all organisms share essentially the
same pathways for biosynthesis of protein, nucleic acids, carbo-
hydrates, and lipids. Prokaryotes can biosynthesize certain com-
pounds like vitamin B12 or certain antibiotics not found in the
eukaryotes, but one must also give due credit to the biosynthetic
capabilities of the fungi, and especially those of the plants that
biosynthesize an incredible diversity of chemical compounds
including hydrocarbons, aromatic compounds, heterocyclics,
and alkaloids. Indeed, the formidable biodegradative abilities
of soil microorganisms can be partially attributed to the selective
pressures of diverse plant compounds in their environment
serving as potential growth substrates.
Two crucial aspects of anabolism are the fixation of carbon
and nitrogen, processes that are essential for primary production
. Fig. 5.18
of biomass on earth. It has long been known that fixation of
Electron micrograph of a thin section of a spore of
nitrogen is a solely prokaryotic process. Moreover, if one con-
Thermoactinomyces sacchari (From Lacy 1971)
siders chloroplasts to be descended from cyanobacteria, which is
clearly the case (Moreira et al. 2000), then essentially all fixation
of carbon on earth is also prokaryotic.
a nutrient-free, desiccated environment, subject to the normal In the following sections, we will give an overview of various
variations of environmental conditions. Second, they must be metabolic modes found in prokaryotes, discussing catabolic
able, from this state, to infect a specific host and to enter the host diversity and the fixation of carbon and nitrogen.
cell. Third, they must then be able to grow and reproduce
intracellularly. Chlamydia accomplishes these processes by alter-
nating between two states—the elementary body (a small, dense, Anaerobic Fermentation
resistant, and nongrowing cell) and the reticulate body (the
vegetative form of the cell that can grow and reproduce). Fermentation can be defined as the utilization of an organic
Readers interested in a more detailed description of the compound in the absence of external electron acceptors, includ-
various groups of prokaryotes that undergo development are ing oxygen. Microbes are unique in their ability to exploit
referred to the excellent recent monograph on the subject (Brun anaerobic environments, and Louis Pasteur’s insight into ‘‘La
and Shimkets 2000). vie sans l’air’’ opened up a whole new area of metabolism.
Among the microbes, the prokaryotes have most extensively
exploited this modus vivendi. As is the case with almost all
Metabolic Diversity aspects of biological diversity, there are few clear-cut categories
but rather a spectrum of differences that diminish as one learns
Catabolism is the part of metabolism involved in conservation of more about the property. Thus, the spectrum of relations to
energy that can be used for biosynthesis and other cellular oxygen includes those organisms that cannot use oxygen as
functions. Energy can be conserved from chemical reactions a terminal electron acceptor and are in fact damaged by exposure
(chemotrophy) or from light (phototrophy). Catabolic diversity to oxygen, those that are likewise obligately anaerobic but are
in prokaryotes greatly exceeds that in eukaryotes. There are indifferent to the presence of oxygen, those facultative organ-
many modes of metabolism, including anaerobic respiration, isms that have the option of metabolizing either aerobically or
or lithotrophy, which eukaryotes cannot perform; even in anaerobically, and finally those organisms that are obliged to use
those that the two cell types share, such as fermentation or oxygen as a terminal electron acceptor.
photosynthesis, the eukaryotes are greatly outstripped by the Eukaryotes can ferment a few common carbohydrates such
prokaryotes in terms of substrates utilized and metabolic as starch, cellulose, glucose, or sucrose to a limited number of
modes. Majorie Stephenson (1949) expressed this concept pre- products: lactate or ethanol and CO2, or, in the case of
sciently in the introduction to the first edition of her textbook hydrogenosome-containing anaerobic protists, to ethanol,
. Fig. 5.19
Myxospores of myxobacteria. (a) Myxococcus xanthus. (b) Stigmatella aurantiaca (Courtesy of H. Reichenbach)
. Fig. 5.20
Akinetes of Anabaena, positioned on both sides of a heterocyst
. Fig. 5.21
(From Carr 1979)
Vegetative cells (top) and azotocysts (bottom) of Azotobacter
vinelandii (Courtesy of H. Sadoff)
acetate, and CO2 and H2. The fermentative abilities of prokary- carbohydrates is the Embden-Meyerhof-Parnas (EMP) pathway
otes are more extensive. found in eukaryotes and many prokaryotes. The Entner-
Carbohydrates often are the main organic substrate available Doudoroff (ED) pathway involves oxidation of glucose-6-
for fermentation. The canonical pathway for utilization of phosphate to glucuronic acid-6-phosphate and leads to the
Embden-Meyerhof glucose Entner-Doudoroff

pathway pathway
ATP
ADP
glucose-6-P
ATP NADP+
ADP NADPH + H+
fructose-1,6-
diphosphate 2-keto-3-deoxy-6-phosphogluconic acid
2 triose pyruvic triose

phosphate acid phosphate
2NAD+ 2 P P NADP+
2H+ + 2NADH NADPH + H+
4ADP 2ADP
4ATP 2ATP
2 pyruvic acid pyruvic

acid
Net Reactions:
glucose + 2ADP + 2NAD+ + 2 P glucose + ADP + NAD+ + NADP+ + P

2 pyruvic acid + 2ATP + 2NADH + 2H+ 2 pyruvic acid + ATP + NADH
+ NADPH + 2H+
. Fig. 5.22
The Embden-Meyerhof and Entner-Doudoroff pathways (Adapted from Stanier et al. 1979)
conservation of only a single ATP per hexose via substrate-level Moreover, instead of the standard glyceraldehyde 3-phosphate
phosphorylation (Conway 1992, > Fig. 5.22). The ED pathway is dehydrogenase, which uses NAD+ as an electron acceptor
common in the Proteobacteria, especially aerobes, but can be and produces 1,3-diphosphoglycerate, Pyrococcus and certain
found in some clostridia and bacilli and even some eukaryotic other Archaea possess a novel tungsten-containing glyceralde-
microbes such as Entamoeba histolytica and Penicillium notatum hyde 3-phosphate/ferredoxin oxidoreductase (Adams 1993)
(Conway 1992). Most of the organisms that use the ED pathway that produces 3-phosphoglycerate. Thus, the net high-energy
are aerobes, but the Proteobacterium Zymomonas mobilis uses it phosphodiester bonds conserved from hexose to pyruvate by this
as a primary fermentative pathway for hexoses. Escherichia coli pathway are zero. The ATP from acetyl-CoA produced from pyru-
possesses both the EMP and the ED pathways and apparently vate, however, is conserved by a novel ADP-dependent acetyl-CoA
uses the latter when dining on uronic acids in the gut (Peekhaus synthetase (previously described acetyl-CoA synthetases cleave ATP
and Conway 1998). Other glycolytic pathways in the Bacteria to AMP and pyrophosphate in the acetyl-CoA synthesis direction).
include that used by heterolactic Gram-positive bacteria (also In contrast, Thermotoga maritima, a phylogenetically deep-
leading to the conservation of a single ATP per hexose), the branching hyperthermophilic fermentative bacterium, converts
pathway found in bifidobacteria, and the pentose phosphate hexoses to pyruvate using the conventional EMP pathway and
pathway, which is generally used for biosynthesis of five carbon conserves ATP from acetyl-CoA using phosphotransacetylase
sugars for nucleic acids (Gottschalk 1986). and acetate kinase, as is found in mesophilic bacteria. The
When sugar-fermenting Archaea were examined for glyco- advantage of the glyceraldehyde 3-phosphate/ferredoxin oxido-
lytic pathways, they were shown to have interesting modifica- reductase over the NAD+-utilizing dehydrogenase is unclear, but
tions of both the EMP and the ED pathways (Kengen et al. 1996; one possibility is that ferredoxin is a very strong electron donor
Selig et al. 1997). For example, the hyperthermophile capable of reducing protons to H2. In contrast, H2 production
Pyrococcus furiosus, which ferments hexoses to acetate, hydro- from NADH is thermodynamically unfavorable at H2 partial
gen, CO2, and alanine, was found to utilize a modified EMP pressures above 103 atm (Wolin and Miller 1982). Therefore,
pathway in which ADP rather than ATP serves as the phosphoryl the modified EMP pathway is less likely to be inhibited under
donor for hexokinase and phosphofructokinase (> Fig. 5.23). conditions of high H2 partial pressures. Indeed, growth of
Thermotoga maritima Pyrococcus furiosus

Glucose Glucose
ATP ADP
Glucokinase Glucokinase
ADP AMP
Glucose-6-P Glucose-6-P
Fructose-6-P Fructose-6-P
ATP ADP
Phospofructo-
Phospfructokinase
kinase
ADP AMP
Fructose-1,6 di-P Fructose-1,6 di-P
DHA-P GA-3-P DHA-P GA-3-P

2NAD+ + P 2 Fdox
GA-3-P Dehydrogenase GA-3-P/Fd
2NADH oxidoreductase 2 Fdred H2
2 1,3-Di-P-Glycerate 2 3-P-Glycerate
2 ADP
2 ATP
2 3-P-Glycerate 2P-EP
2 ADP
2 ATP
2 Pyruvate
2P-EP
2 Fdox + CoA
2 ADP Pyruvate/Fd
oxidoreductase 2 Fdred
2 ATP H2
2 Acetyl-CoA
2 Pyruvate ADP-forming 2 ADP + 2 P
2 Fdox + 2 CoA acetyl-CoA
synthetase 2 ATP + 2 CoA
2 Fdred
2 Acetate
2 Acetyl-CoA
Phosphotrans- 2P
acetylase 2 CoA
2 Acetyl-P
2 ADP
Acetate kinase
2 ATP
2 Acetate
. Fig. 5.23
Comparison of glycolytic pathways in Thermotoga maritima and Pyrococcus furiosus. P phosphate, DHA dihydroxyacetone, GA
glyceraldehyde, P-EP phosphoenolpyruvate, Fd ferredoxin, and CoA coenzyme A
T. maritima on sugars becomes inhibited unless an electron phosphorylation occurs at a later step in the ED pathway (Con-
acceptor such as elemental sulfur or a hydrogen-utilizing way 1992).
methanogen is added to remove H2. Besides carbohydrates, amino acids, derived from the hydro-
Archaea possess other variants on the EMP pathway lysis of proteins by proteases, are fermented by a variety of
(Menendez et al. 1999). In Desulfurococcus, ATP is used in prokaryotes, particularly the clostridia, either singly or in
both hexokinase and phosphofructokinase as in the classic pairs, as in the Stickland reaction (Gottschalk 1986). The prod-
EMP pathway, whereas in Thermoproteus, ATP is used by hexo- ucts of amino acid fermentation, amines (including putrescine
kinase and pyrophosphate is used by a phosphofructokinase, as and cadaverine), branched-chain fatty acids, and sulfide and
has been found in some eubacteria and eukaryotes. All EMP mercaptans from sulfur-containing amino acids, essentially
pathways studied in the Archaea thus far utilize glyceraldehyde define the word putrid. Our great olfactory sensitivity to these
3-phosphate/ferredoxin oxidoreductase instead of the NAD+- compounds no doubt has played a selective role in protecting us
utilizing dehydrogenase. A modification of the ED pathway in from eating spoiled food potentially containing botulin and
certain Archaea (including Sulfolobus, Thermoplasma, and other toxins. The purines and pyrimidines derived from nucleic
Halobacterium) has been described in which glucose is not acids are also readily fermented by prokaryotes.
phosphorylated before oxidation to gluconic acid, and
Originally it was thought that only a small number of sugars assimilate nitrate and nitrite but can be used as a respiratory
and amino acids were fermented by prokaryotes. As anaerobic process by E. coli, for example (Stewart 1994). The energetics of
culture techniques have improved, an astounding diversity of using nitrate and nitrate compounds as electron acceptors is
compounds has been found to serve as substrates for fermenta- comparable to that of using O2, and most of the organisms
tion, including aromatic compounds and even saturated alkane reducing nitrogen oxides are facultative aerobes.
hydrocarbons (Zengler et al. 1999). Many of these fermentations In the final decade of the twentieth century, recognition of
take place in the presence of a hydrogen-consuming anaerobe the importance of Fe(III) as an electron acceptor in anaerobic
such as a methanogen. The removal of hydrogen (or acetate in habitats increased. Whereas at low pH, Fe(III) mainly exists as
some cases) allows the fermentation to be energetically favor- the free ion and the Fe(III)/Fe(II) oxidation-reduction potential
able. This syntrophic interaction is called ‘‘interspecies hydrogen is near +0.77v, at circumneutral pH values, Fe(III) exists pri-
transfer’’ (Schink 1997). marily as hydroxide precipitates, which causes the oxidation/
Indeed, our concepts of fermentation have had to change reduction potential to be between +0.2 and 0.2v (Widdel et al.
over the past few decades. It was generally considered that 1993). Amorphous ferric hydroxide precipitates are much more
fermentations involved substrate-level phosphorylation. How- bioavailable than are more crystalline ones. Most known Fe(III)
ever, consider the example of Propionigenium modestum, which reducers are members of the Proteobacteria, but even certain
grows by decarboxylating succinate to propionate by the follow- deep-branching thermophilic Bacteria and Archaea can reduce
ing reaction: Fe(III) (Vargas et al. 1998), suggesting a role for this process
early in life’s history on earth. Also, certain purple photosyn-
O H H O H H O
thetic bacteria oxidize Fe(II) to Fe(III) (Widdel et al. 1993),
H+ + C C C C H C C C + CO2
making for a photosynthetic producer-consumer cycle analo-
O– H H O– H H O–
gous to those for oxygen or sulfur.
Succinate Propionate Sulfur compounds can serve as electron acceptors, and it is
apparent from > Table 5.1 that the amount of energy available
There is no net redox reaction, only a decarboxylation. from their reduction is considerably less than that of the pre-
Readers might rightly guess that the pathway is not a simple ceding compounds. Sulfate is an abundant form of sulfur, espe-
decarboxylation, but rather involves coenzyme A derivatives and cially in seawater, where its concentration is near 28 mM. Two
includes a rearrangement to methylmalonyl-CoA. However, the genera, Desulfovibrio and Desulfotomaculum, were essentially the
substrate-level phosphorylations in this pathway provide a net only ones known until better anaerobic techniques were applied,
ATP yield of zero. Instead, it was shown that membrane-bound revealing an enormous diversity of organisms. Most sulfate
methylmalonyl-CoA decarboxylase pumps sodium ions with reducers belong to the d- and e-subphyla of the Proteobacteria
each decarboxylation it carries out. A sodium-dependent ver- or to the Firmicutes (Gram-positive bacteria) in the Bacteria and
sion of an F1F0-ATPase can then conserve ATP from the sodium to the genus Archaeoglobus in the Archaea. Sulfate reducers
gradient across the cell membrane. Thus, in the absence of and other organisms can generally utilize sulfite or thiosulfate,
electron transport, a chemiosmotic potential is generated in and these two compounds can be used even in the absence of
this organism, as well as in others carrying out similar decar- organic compounds in an inorganic ‘‘fermentation’’ to sulfide
boxylations (Dimroth and Schink 1998). and sulfate (Bak and Pfennig 1987).
Elemental sulfur also is used as an electron acceptor by
diverse anaerobes, either as a respiratory electron acceptor lead-
Anaerobic Respiration ing to energy conservation by an electron transport chain or
simply as a way to recycle NADH to NAD+ for certain fermen-
For aerobic respiration, eukaryotes are dependent upon mito- tative organisms such as Thermotoga. Elemental sulfur is
chondria, clearly derived from endosymbiotic prokaryotes in the reduced to sulfide by many hyperthermophiles growing under
a-subphylum of the Proteobacteria (Gray et al. 1999). If the conditions of ‘‘fire and brimstone,’’ and sulfur reduction is
mitochondria were once capable of using electron acceptors a good candidate for the first respiratory process on earth.
other than oxygen, they no longer are, nor have the eukaryotes When other electron acceptors in anaerobic habitats are
evolved other mechanisms to use alternate electron acceptors. absent or depleted, the acceptor remaining is carbon dioxide,
Thus, anaerobic respiration is solely the domain of prokaryotes. which can be used by either methanogens or acetogens.
Moreover, prokaryotes can utilize a large range of electron Methanogens represent the predominant phenotype in the
acceptors other than oxygen (> Table 5.1). Euryarchaeota phylum in the Archaea, demonstrating an enor-
Nitrate and nitrite are produced aerobically from ammonia mous phylogenetic and morphological diversity. Methanogens
by nitrifiers (see below) and can be used as electron acceptors by use only a small number of simple substrates (the most complex
diverse members of both the Bacteria and the Archaea (Zumft is acetic acid) and an intricate pathway for reduction of one-
1997). Nitrate is initially reduced to nitrite, and nitrite can be carbon units to methane (DiMarco et al. 1990). This pathway
reduced either to N2 and N2O gas via the denitrification pathway was thought to contain many unique enzymes and cofactors, but
or to ammonia. The former process is more common as a large portion of the pathway is now known to be used to oxidize
a respiratory process, whereas the latter is often used to formaldehyde to CO2 in aerobic methylotrophic bacteria
. Table 5.1
Electron acceptor utilization for respiration
Reactants Products DG0 /H2 (kJ) Representative organisms

O2 + 2 H2 2 H2O 237 Homo sapiens
Escherichia coli
Sulfolobus acidocaldarius

NO3 + H2 NO2 + H2O 163 Escherichia coli
Pyrobaculum aerophilum

NO2 + 2 H + 3 H2+
NH4+ + 2 H2O 145 Escherichia coli
2 NO2 + 2 H+ + 3 H2 N2 + 4 H2O 265 Pseudomans stuzeri
Pyrobaculum aerophilum
2 Fe(OH)3 + 2 HCO3 +
+ 2 H + H2 2 FeCO3 + 6 H2 118 Geobacter metallireducens
Shewanella putrefaciens
SO42 + H+ + 4 H2 HS + 4 H2O 38 Desulfovibrio desulfuricans
Archaeoglobus fulgidus
S + H2 HS + H+ 28 Desulfuromonas acetoxidans
Pyrodictium brockii

HCO3 + 4 H2 + H +
CH4 + 3 H2O 34 Methanococcus jannaschii
Methanospirillum hungatei
2 HCO3 + 4 H2 + H +
CH3COO + 4 H2O 26 Acetobacterium woodii
Fumarate2 + H2 Succinate2 86 Escherichia coli
(CH3)2SO + H2 (CH3)2 S + H2O 124 Escherichia coli
Rhodobacter capsulatus
R–Cl R–H 170 Desulfomonile tiedjei
Dehalococcoides ethenogenes
ClO4 + 4 H2
Cl + 4 H2O 268 Ideonella dechloratans
a
H2 is used as a model electron donor for the sake of comparisons of the reactions, and its use does not imply that the representative organisms use H2. Most DG0
values are taken from Thauer et al. 1977
(Chistoserdova et al. 1998). Moreover, coenzyme M, the imme- hierarchy of competitive exclusion is best explained by
diate precursor of methane in methanogens, has been found in an a threshold model based on thermodynamic principles, so that
aerobic alkane-oxidizing Xanthobacter species (Allen et al. 1999). if the substrate is H2, sulfate reducers are capable of utilizing that
Thus, there is indeed unity in biochemistry (Singleton 2000). H2 at concentrations below that at which methanogens can
Acetogens are anaerobic eubacteria generally in the Firmicutes conserve energy from it (Cord-Ruwisch et al. 1988).
capable of reducing two CO2 moieties to the methyl and acetyl Besides these biogeochemically important electron accep-
groups of acetyl-CoA (see section > ‘‘Autotrophic CO2 Fixation’’ tors, a wide diversity of others can be utilized, and we will
in this chapter) by a pathway utilizing an enzyme complex touch on only a few. It is generally considered that electron
sometimes called the ‘‘carbon monoxide dehydrogenase/acetyl- acceptors in anaerobic respiration are inorganic, but, for exam-
coenzyme A synthetase complex’’ (Ragsdale 1997). A variant on ple, E. coli and many other organisms capable of anaerobic
this enzyme complex is used by methanogenic Archaea to fix growth can use fumarate or dimethyl sulfoxide (Weiner et al.
carbon dioxide for autotrophic growth and by acetate-utilizing 1988) as respiratory electron acceptors. Indeed, during its ‘‘fer-
methanogens to split acetyl-coenzyme A. Unlike the mentation’’ of glucose, E. coli produces succinate, which is the
methanogens, the acetogens are metabolically versatile, using product of the respiratory reduction of fumarate.
a wide variety of substrates for acetogenesis. Besides its importance to the biogeochemical cycles of car-
In most anaerobic habitats, the amount of electron donor is bon, nitrogen, and sulfur, anaerobic respiration plays an impor-
limiting, and prokaryotes capable of anaerobic respiration are in tant role in biodegradation of pollutants. One important set of
fierce competition for those electrons. It has long been known reactions is that involving reductive dechlorination of chlori-
that the outcome of this competition correlates with the free nated organics, which are among the most important of pollut-
energy available from the reactions, so that denitrifiers could ants. The most highly chlorinated organics are often resistant to
outcompete sulfate reducers, which, in turn, outcompete aerobic attack, but there is considerable energy available for
methanogens, which generally outcompete acetogens. This conservation if they can use the chlorinated organic compound
a Paracoccus denitrificans
4H+ 2H+ 2H+
+
2H
NADH Rieske FeS O + 2H+
Cu-cyt aa3
NADH dehydrogenase UQ cyt b/c1 cyt c cyt c oxidase
complex complex complex
H2O
b Escherichia coli
4H+ (Na+) 2H+

2H+
O + 2H+
NADH Cu-cyt bo3
NADH dehydrogenase UQ quinol oxidase
complex complex
H2O
+
2H
c Thiobacillus ferrooxidans
O + 2H+
Cu-cyt aa3
RC cyt c cyt c oxidase
2Fe2+ complex
H2O
d Sulfolobus acidocaldarius 2H+ 2H+

2H+
O + 2H+
Rieske FeS Cu-cyt aa3
NADH cyt b cyt c oxidase
NADH CQ
dehydrogenase complex complex
H2O
. Fig. 5.24
Electron transport in various prokaryotic aerobes: (a) Paracoccus denitrificans, (b) Escherichia coli, (c) Thiobacillus ferrooxidans, and
(d) Sulfolobus acidocaldarius. UQ ubiquinone, cyt cytochrome, RC rusticyanin, and CQ caldariellaquinone
as a respiratory electron acceptor (Mohn and Tiedje 1992), contaminated several groundwater aquifer sites, to chloride
a process sometimes called ‘‘dehalorespiration.’’ It was first dem- (Coates et al. 1999; Malmquist et al. 1994).
onstrated that 3-chlorobenzoate could serve as an electron
acceptor for energy conservation via reductive dechlorination
for an organism called Desulfomonile tiedjei (DeWeerd et al. Aerobic Metabolism
1990; Suflita et al. 1982), a member of the d-Proteobacteria.
Since then, organisms have been shown to conserve energy by In aerobic eukaryotes, respiration is carried out by mitochon-
reductive dechlorination of chlorophenols and chlorinated eth- dria, which are now known to be descended from the
enes (Holliger et al. 1998; Maymó-Gatell et al. 1997; Mohn and a-Proteobacteria (Gray et al. 1999). Thus, all respiration on
Tiedje 1992), the latter including the solvents tetrachloroethene earth is prokaryotic. In essentially all aerobic organisms, elec-
and trichloroethene, which are particularly pervasive ground- trons travel down an electron transport chain from the organic
water pollutants. Curiously, although evidence for anaerobic compound to oxygen in a manner such that protons, or some-
breakdown of less chlorinated methanes is ample (Mägli et al. times sodium ions, are pumped out of the cell or mitochon-
1996; Messmer et al. 1993), respiratory utilization of chloroform drion, leading to development of an electric potential that can be
and carbon tetrachloride has not been described. converted to ATP by membrane-associated ATPases (Saraste
Other environmental applications utilizing anaerobic respi- 1999).
ration involve reduction of various metals besides Fe(III) Diversity of prokaryotic electron transport chains can be
(Lovley and Coates 2000). For example, Desulfovibrio considerable. For example, the a-Proteobacterium Paracoccus
desulfuricans can reduce U(VI) to U(IV), which is denitrificans utilizes a complex transport chain very closely
a precipitate, and allows immobilization of radioactive wastes resembling that in mitochondria consisting of three large mem-
(Lovley and Phillips 1992). Another novel reaction with envi- brane-bound enzyme complexes (> Fig. 5.24a). The first is an
ronmental potential is the respiratory reduction of perchlorate, NADH/ubiquinone oxidoreductase complex (NADH dehydro-
which is part of rocket propellant mixtures and has genase or complex I) containing a bound flavin and several iron/
sulfur centers. The second is a ubiquinol/cytochrome
c-oxidoreductase complex (cytochrome bc1 complex or complex nature, as well as many nonnatural synthetic compounds (xeno-
III), which contains heme groups, and the high potential Rieske biotics). There are exceptions to this doctrine of microbial
iron/sulfur center carries out a Q cycle (Saraste 1999), thereby infallibility (Alexander 1981) including some polymers and
conserving energy from ubiquinone transport. Cytochrome c is some smaller molecules that do not resemble natural substrates,
reduced in the periplasm and transfers its electrons to a copper- leading to persistence of some toxic chemicals in the environ-
and heme-containing cytochrome c oxidase (complex IV), ment. Nevertheless, a novel compound existing in an environ-
which passes the electrons from a periplasmic cytochrome c to ment at a reasonably high concentration may eventually select
oxygen. Protons are pumped out of the cell by each of the for organisms capable of using it. For example, Burkholderia
complexes, and by the asymmetric reduction of ubiquinone on cepacia strain AC1100 has, apparently by mutation and genetic
the inside of the cell membrane and its oxidation of the outside exchange, developed a pathway to utilize the herbicide 2,4,5-T,
(the Q loop) and by the consumption of protons inside the cell originally considered nondegradable, as a growth substrate
by the reduction of oxygen to water. It should be mentioned that (Haugland et al. 1990; Huebner et al. 1998).
P. denitrificans actually has several different terminal oxidases
that it regulates in response to growth conditions (De Gier et al.
1994). Lithotrophy and Methanotrophy
Escherichia coli has a simpler electron transport chain that
lacks complex III, so that electrons are passed directly from the Another metabolic capability found uniquely in prokaryotes is
quinol to an oxidase complex (> Fig. 5.24b). Thus, it does not lithotrophy, the ability to use inorganic electron donors for
pump as many protons per oxygen as does P. denitrificans. energy conservation. Their ability to oxidize inorganic nitrogen
Moreover, under low oxygen conditions, E. coli induces more and sulfur compounds makes prokaryotic lithotrophs impor-
of a quinol oxidase complex containing hemes b and d. This tant components of the biogeochemical cycles of these elements.
complex does not pump protons at all, but has a lower Km for Both aerobes and anaerobes can oxidize inorganic substrates,
oxygen, which is a useful trade-off. such as hydrogen oxidation by methanogens, but we will confine
Thiobacillus ferrooxidans is an acidophile which transports our discussion mainly to aerobes. Because they use inorganic
electrons from Fe(II) to oxygen, and inasmuch as the Fe(II)/ substrates for energy conservation, lithotrophs are often auto-
Fe(III) couple is near +0.77 v, Fe(II) cannot reduce NADH trophs, that is, they fix carbon dioxide, but some can incorporate
(0.32 v) or quinones (ca. 0.0 v). Therefore, T. ferrooxidans organic carbon. Methanotrophy, the ability to utilize methane as
has a truncated electron transport chain in which electrons an electron donor, has only been found in certain aerobic mem-
flow from Fe(II) to a small periplasmic copper-containing pro- bers of the Proteobacteria.
tein called ‘‘rusticyanin,’’ then to cytochrome c, and finally to The nitrifiers, bacteria that oxidize ammonia and nitrite to
a typical oxidase (> Fig. 5.24c). Thus, T. ferrooxidans conserves nitrate, were originally characterized in the classic studies of
less energy per electron than do organisms using reactions in Winogradsky. One set of organisms, sometimes called
which NADH is the electron donor. ‘‘nitrosofiers,’’ oxidizes ammonia to nitrite, followed by oxida-
As an example of an archaeal electron transport chain, the tion of nitrite to nitrate. The reason that oxidation of ammonia
thermoacidophile Sulfolobus acidocaldarius can use elemental to nitrate requires two separate microbial groups is not under-
sulfur or organic compounds as electron donors for its aerobic stood. Most extant nitrifiers are members of the Proteobacteria,
growth. When growing on organic compounds, it transports although the nitrite oxidizer Nitrospira is in a distinct phylum,
electrons from NADH to oxygen using an electron transport and all can fix CO2 by the Calvin cycle. Because organic matter
chain that is similar to those in eubacteria, but with some generally inhibits nitrofiers, Winogradsky used silica gel plates
interesting differences (> Fig. 5.24d). Its NADH dehydrogenase rather than agar to isolate them.
complex is relatively small and does not appear to pump pro- The aerobic oxidation of ammonia begins with oxidation to
tons; it uses a sulfur-containing quinone called ‘‘calderiel- hydroxylamine by ammonia monooxygenase, making that pro-
laquinone,’’ and the electrons then go through a supercomplex cess obligately aerobic. However, an anaerobic oxidation pro-
equivalent to the two found in Paracoccus, but lacking cyto- cess, called the ‘‘anammox reaction,’’ has been found in which
chrome c (Schäfer et al. 1996). The main subunit of the terminal ammonia is the electron donor and nitrite is the electron accep-
oxidase complex shows genetic relatedness to other copper- tor (> Table 5.2). The organism responsible for this reaction has
heme oxidases, which form a gene family (Garcia-Horsman not been isolated, but strong molecular biological evidence
et al. 1994) that also includes nitric oxide reductase from deni- shows that the predominant organism in the enrichment culture
trifiers, suggesting that the different reductases had a common is a member of the phylum Planctomycetes (Strous et al. 1999).
origin. The ‘‘colorless’’ sulfur bacteria are so named to differentiate
The energetics of aerobic respiration is so energetically them from the purple sulfur and green sulfur photosynthetic
favorable that the oxidation of essentially any organic com- bacteria. They oxidize reduced sulfur compounds, often the
pound is thermodynamically feasible, and the only limitation product of sulfate reducers in anaerobic zones, so they are
for an organism to utilize a particular organic compound is often found at anaerobic/aerobic interfaces. Typically they can
devising an energy-conserving breakdown pathway. Typically, oxidize sulfide, elemental sulfur (to which they can sometimes
microorganisms can break down any compound made by be attached or which they store in cellular vacuoles), or
. Table 5.2
Reactions carried out by aerobic lithotrophs and methanotrophs
Reaction DG0 /2e (kJ) Organism type Example species

NH4+ + 1.5 O2 ! NO2 + 2 H+ + 2 H2O 137 Nitrosofying bacteria Nitrosomonas europaea
NO2 + 0.5 O2 ! NO3 76 Nitrifying bacteria Nitrobacter winogradskyi
NH4+ + NO2 ! N2 + 2 H2O 238 Anammox organisms Brocadia anammoxidans
S + 1.5 O2 ! SO42 + 2 H+ 196 ‘‘Colorless’’ sulfur bacteria Thiobacillus thiooxidans
2 Fe 2+
+ 2 H + 0.5 O2 ! 2 Fe
+ 3+
+ H2O 66 a
Iron bacteria Thiobacillus ferrooxidans
2 FeS2 + 7.5 O2 + 7 H2O ! 2 Fe(OH)3 + 8 H+ + 4 SO42 164a Iron bacteria Thiobacillus ferrooxidans
Metallosphaera sedula
H2 + 0.5 O2 ! H2O 237 Hydrogen bacteria Ralstonia eutropha
CH4 + 2 O2 ! HCO3 + H+ + H2O 203 Methanotrophs Methylococcus capsulatus
Abbreviation: Anammox Anaerobic ammonium oxidation
a
Values for pH = 2
thiosulfate. Because their main metabolic product is sulfuric plug at the bottom while oxygen diffused from the top. Micro-
acid, some of the sulfur oxidizers are acidophiles. Many of the organisms grew as a band at the iron/oxygen interface in these
Gram-negative rods, both acidophilic and neutrophilic, which cultures, from which neutrophilic iron oxidizers were isolated.
carry out this reaction, were named ‘‘Thiobacillus.’’ However, At low pH, ferrous iron is stable in the presence of oxygen,
these organisms are scattered throughout the Proteobacteria and and Thiobacillus ferrooxidans and other iron-oxidizing organ-
many will require new generic assignments. There are also some isms such as Leptospirillum ferrooxidans can be readily isolated
filamentous sulfur-oxidizing bacteria such as Beggiatoa and from acidophilic environments. Acidophilic Archaea such as
Thiothrix, which are also in the proteobacteria (although Sulfolobus acidocaldarius and the newly described mesophile
Beggiatoa, because of its close morphological resemblance to Ferroplasma acidarmanus (Edwards et al. 2000) can also oxidize
the filamentous cyanobacterium Oscillatoria, was once consid- ferrous iron. Most organisms that can oxidize ferrous iron and
ered by some to be a colorless cyanobacterium). There are also sulfur compounds play a role in the leaching of metal sulfide
archaeal sulfur oxidizers, such as Sulfolobus. minerals such as pyrite (FeS2). The exposure of pyrite minerals
Some sulfur oxidizers can use nitrate as an electron acceptor. to oxygen during coal mining leads to their oxidation to sulfuric
One interesting example is the large (ca. 50 mm in diameter) acid, resulting in acid mine drainage. On the other hand, this
gliding filaments called Thioploca, which form large mats in ability has been taken advantage of to recover metals such as
upwelling zones. When these filaments are in the aerobic zone, copper from low-grade sulfide ores. Thus, the iron-oxidizing
they concentrate nitrate from seawater, where its concentration prokaryotes can have positive or negative economic effects.
is near 25 mM, into intracellular vacuoles, where the nitrate The ability to oxidize hydrogen (which only requires the
concentrations can reach the remarkably high concentration of ability to link a hydrogenase with an electron transport chain)
5 M. They can then glide down into the anaerobic parts of the is widespread in the Bacteria and Archaea. Because of labile iron/
mat, where they can then use the nitrate for oxidation of sulfide sulfur centers in hydrogenases, often hydrogen oxidation occurs
(Fossing et al. 1995). An even larger organism, up to 1 mm in under microaerophilic conditions in aerobes. Some hydrogen
diameter, is the spherical organism, called ‘‘Thiomargarita’’ bacteria can grow as autotrophs, but many will incorporate
(sulfur-pearl), which consists of a thin film of cytoplasm around organic carbon when available.
a large nitrate-filled vesicle. These organisms are uniquely poised Methanotrophy, the ability to oxidize methane, has only
to exploit the anaerobic/aerobic interface by using an electron been found in two clusters in the Proteobacteria. Methylotrophy,
acceptor that allows them to use sulfide in anaerobic layers the ability to use single-carbon compounds such as methanol, is
where aerobic metabolism is excluded. more widespread, occurring in many bacteria and archaea, as
Ferrous iron spontaneously oxidizes at pH 7, which has ham- well as in certain yeasts. The first step in aerobic methanotrophs
pered studies on neutrophilic iron oxidizers. Several organisms, is the hydroxylation of methane to methanol by methane
such as the sheathed bacterium Leptothrix and the stalked bacte- monooxygenase, making the process obligatorily
rium Gallionella, are known to precipitate iron and manganese aerobic. However, there is considerable evidence that methane
oxides, but it is still unclear what role they play in their metab- may be oxidized under anaerobic conditions (Boetius et al.
olism. More recently, Emerson and Moyer (1997) used agar 2000), but the organisms involved have thus far proved elusive.
gradient cultures in which reduced iron diffused from an agar It was long believed that the intermediates after methanol in the
. Table 5.3
Properties of photosynthetic bacteria and chloroplasts
Phylogenetic Reaction center Accessory Primary electron CO2 fixation

group chlorophyll pigments Electron donors acceptor pathway
Purple nonsulfur a and b bcl a or b chl a or b Organic compounds, H2, Q Calvin cycle
bacteria Proteobacteria Fe2+, H2S (at low
concentration)
Purple sulfur g Proteobacteria bcl a chl a or b H2S, H2, some organic Q Calvin cycle
bacteria compounds, etc.
Heliobacteria Firmacutes bcl g — Organic compounds Fd Calvin cycle
Green sulfur Chlorobi bcl a bcl c,d,e H2S, S2O32, H2 Fd Reverse TCA cycle
bacteria (chlorosomes)
Green nonsulfur Chloroflexi bcl a bcl c,d,e Organic compounds, H2, Q Hydroxypropionate
bacteria (chlorosomes) H2S cycle
Cyanobacteria Cyanobacteria chl a Phycobilins or H2O Q (PSI) Calvin cycle
chl b Fd (PSII)
Chloroplasts Cyanobacteria chl a Phycobilins or H2O Q (PSI) Calvin cycle
chl b Fd (PSII)
Abbreviations: Q Quinone, Fd Ferredoxin, TCA Tricarboxylic acid, PS Photosystem
pathway of methane oxidation were free one-carbon com- endosymbiotic event (Moreira et al. 2000). There are two general
pounds such as formaldehyde and formate (Dworkin and Foster classes of phototrophy in prokaryotes. The first is one based on
1956), but it has recently been shown that the pathway in many the now inappropriately named ‘‘bacteriorhodopsin’’ (it should
methanotrophs and methylotrophs closely resembles that in be ‘‘archaerhodopsin’’) found in halophilic Archaea. Bacterio-
methanogenic Archaea in which intermediates are bound to rhodopsin is a polypeptide with a retinal prosthetic group,
tetrahydromethanopterin and methanofuran and involves which allows it to pump protons, thereby generating a proton-
which enzymes homologous to those in methanogens motive force. This system seems to be supplemental to normal
(Chistoserdova et al. 1998). heterotrophic growth in these organisms, although its actual
Eukaryotes are unable to utilize inorganic energy sources or ecophysiological role is unclear. Besides bacteriorhodopsin, the
methane, and therefore, some living in habitats in which these halobacteria can possess halorhodopsin, a light-driven chloride
energy sources are abundant, such as anaerobic/aerobic sedi- pump, and two types of sensory rhodopsins used in phototaxis
ment interfaces or near undersea spreading centers, have entered (Spudich 1993). Recently, a DNA fragment cloned from ocean
symbioses with lithotrophs or methanotrophs. One of the best water was found to have, besides a 16S rDNA gene, which
known examples is the giant tube worm Riftia living near under- showed it to be from a proteobacterium, a gene encoding rho-
sea spreading centers and using intracellular bacteria that oxi- dopsin-like protein (Beja et al. 2000). When expressed in E. coli
dize sulfide and fix CO2 via the Calvin cycle (Robinson et al. and provided with retinal, this protein pumped protons, and it
1998). A wide variety of marine invertebrates, especially clams has been given the name ‘‘proteorhodopsin.’’
and mussels, have either sulfide-oxidizing or methane-oxidizing The major type of phototrophy is that based on chlorophylls.
symbionts, sometimes both (Distel and Cavanaugh 1994). Thus, Among the prokaryotes, there is considerable diversity of pig-
the eukaryotic hosts have used the prokaryotes to augment their ments, photosystems, and electron donors, all adapted to
metabolic capabilities, much as eukaryotic hosts did when enter- a variety of ecological niches. The photosynthetic bacteria can
ing symbioses with the ancestors of the mitochondria and chlo- be placed into five groups, the purple bacteria (traditionally
roplasts or with symbiotic bacteria providing needed nutrients divided into the purple sulfur and nonsulfur bacteria, based on
(Moran and Baumann 2000). their use of reduced sulfur or organic compounds as the pre-
ferred electron donor), the heliobacteria, the green sulfur bacte-
ria, the green nonsulfur bacteria, and the cyanobacteria
Phototrophy (> Table 5.3). Phylogenetic analysis of the 16S rDNA of the
photosynthetic bacteria (Stackebrandt et al. 1996) shows that
Phototrophy, the ability to use light as an energy source, is yet these groups occupy different phyla.
another invention of prokaryotes. Photosynthesis in eukaryotes The purple bacteria are found in three subphyla of the
is carried out in chloroplasts, conclusively shown to be derived diverse phylum Proteobacteria, which includes many typical
from cyanobacteria and probably the result of a single ‘‘Gram-negatives’’ such as E. coli, Pseudomonas, and Rhizobium.
The purple nonsulfur bacteria are spread within the a- and synthesis in cyanobacteria and chloroplasts appear to share
b-subphyla, and the purple sulfur bacteria are associated with a common evolutionary origin (Tomitani et al. 1999), but
the g-subphylum. This distribution strongly suggests that the whether that is due to the ancestor of the cyanobacteria and
ancestor of these groups was itself a ‘‘purple bacterium’’ and that chloroplasts containing chl b or horizontal gene transfer has yet
the current nonphotosynthetic organisms in the Proteobacteria, to be determined.
like E. coli, evolved from photosynthetic ancestors. Van Niel’s insight that the photosynthetic reduction of CO2
Similarly, it was a surprise when the heliobacteria, a group in green plants was a consequence of [H] released by the pho-
physiologically similar to the purple nonsulfur bacteria, were tolysis of water (van Niel 1949) was a stroke of genius and led to
found to be members of the Gram-positive phylum Firmacutes, a unifying equation of photosynthesis: CO2 + 2 H2A ! CH2O
although in this case, only a single small branch within the H2O + A, where A could be O, S, an organic constituent, or even
phylum contains photosynthetic organisms. The green sulfur nothing in the case of H2. It also rationalized the diversity among
bacteria form essentially their own separate phylum, the the photosynthetic bacteria, as it became clear that water in the
Chlorobia. According to 16S rDNA phylogeny, the green cyanobacteria reduced sulfur in the green and purple sulfur
nonsulfur bacteria, despite similarities of their photopigments, bacteria, and organic molecules in the purple nonsulfur bacteria
are not closely related to the green sulfur bacteria and are the all served the same purpose—that of providing reducing power
founding members of the phylum Chloroflexus. Interestingly, for biosynthesis. And each of these sources of reducing power
the phylogeny of genes for chlorophyll synthesis in the green made it possible for that particular group of photosynthetic
sulfur and nonsulfur bacteria indicates a relatively close affinity prokaryotes to exploit a particular ecological niche: water in
of the two groups, suggesting horizontal gene transfer of the the case of the cyanobacteria, areas rich in reduced sulfur for
photosynthesis genes (Xiong et al. 2000). Finally, the the green and purple sulfur bacteria, and areas with organic
cyanobacteria also form a separate phylum, which also includes substrates for the purple and green nonsulfur bacteria.
chloroplasts. Photosynthetic pigments have adapted further. All of the
The cyanobacteria and chloroplasts contain chlorophyll a chlorophyll molecules have the same basic ground plan—a
(chl a) as their primary photosynthetic pigment in their reaction substituted Mg-tetraporphyrin. However, the nature of the sub-
centers, which absorbs light maximally at 680–700 nm, stitutions and of the sequence of conjugated double bonds
depending on whether it is in photosystem I or II. The other generates a series of absorption spectra that span from 680 nm
photosynthetic bacteria possess only a single photosystem, most for the major absorption peak of chlorophyll a in the
containing mainly bacteriochlorophyll a (bcl a), which absorbs cyanobacteria to 1,035 nm for the major absorption maximum
maximally in the infrared at 800–870 nm. Exceptions include for bacteriochlorophyll b in certain purple bacteria. Thus, the
certain purple nonsulfur bacteria which utilize bcl b, which photosynthetic prokaryotes are able to span a broad range of
absorbs maximally in the far infrared at 1,050 nm, and the photic zones. The green and purple bacteria usually occupy
heliobacteria, which utilize the interesting pigment bcl g, shallow, aquatic areas, underlying the oxygenic cyanobacteria
which absorbs maximally at 790 nm and breaks down to chl a that filter out the relatively short, visible wavelengths of light and
upon exposure to oxygen. transmit the longer, near-infrared wavelengths. This orientation
All phototrophs contain carotenoids (which make the pur- is consistent with the fact that photosynthesis in the green and
ple bacteria ‘‘purple’’ and green bacteria green) as accessory purple bacteria is an anoxygenic process and their photosynthe-
pigments to widen the spectrum of light used. In the purple sis is obligately anaerobic.
bacteria, the structure of the antenna complex in the membrane,
which feeds excitation energy into the reaction center, forms
a set of concentric circles around the reaction center (Cogdell Autotrophic CO2 Fixation
et al. 1999). In the green sulfur and nonsulfur bacteria, light
energy is transmitted to the reaction center in the cell membrane Carbon dioxide is an abundant and available source of carbon,
by the chlorosome (a macromolecular complex containing pro- but it must be reduced to cellular organic carbon at approxi-
tein and bacteriochlorophylls c, d, and e), which is associated mately the level of CH2O, a process that requires reducing power
with the inside face of the cell membrane. In most cyanobacteria, and usually requires energy. Organisms that use CO2 as their
the main accessory pigments that transmit energy to photosys- primary carbon source are autotrophs and serve as primary
tem II are the phycobilins (tetrapyrrole-containing proteins, producers in ecosystems. Autotrophic carbon dioxide fixation
phycocyanin being the one that confers the blue part of the in eukaryotes is represented solely by the Calvin cycle, though
blue-green color of most members of this group). Chloroplasts among the prokaryotes, there are several distinct pathways of
can contain either phycobilins, as in the case of the red algae, or autotrophic CO2 fixation: (1) the Calvin cycle, (2) the reverse
chl b, as found in plants and green algae, as the main accessory TCA cycle, (3) acetogenesis, and (4) the hydroxypropionate
pigment. Finding cyanobacteria with chl b suggested that they pathway.
were the ancestors of the green plant chloroplasts; however, The Calvin cycle, which was discovered initially in the green
phylogenetic analyses have demonstrated that these algae, is also found among many of the photolithotrophic and
‘‘prochlorophytes’’ are simply distinct strains of cyanobacteria chemolithotrophic Bacteria, where it serves as the major mech-
with no direct relationship with chloroplasts. The genes for chl b anism for carbon assimilation (> Fig. 5.25). The first step of CO2
CO2
CO2
CO2
Ru-P2
PGA
Ru-5-P PGA
Ru-P2 PGA
PGA GAP
Ru-5-P
Ru-P2 PGA
PGA GAP
Ru-5-P
GAP
Ri-5-P GAP
GAP
Xu-5-P
Su-7-P GAP
Xu-5-P DAP
Su-P2 DAP
F-P2
E-4-P F-6-P
. Fig. 5.25
The Calvin cycle. Abbreviations: Ru-P2 ribulose-1,5-bisphosphatem PGA 3-phosphoglycerate, GAP glyceraldehyde 3-phosphate, DAP
dihydroxyacetone phosphate, F-P2 fructose-1,6-bisphosphate, F-6-P fructose-6-phosphate, E-4-P erythrose-4-phosphate, Su-P2
sedoheptulose-1,7-bisphosphate, Su-7-P sedoheptulose-7-phosphate, Xu-5-P xylulose-5-phosphate, Ri-5-P ribose-5-phosphate, Ru-5-P
ribulose-5-phosphate (Adapted from Gottschalk 1986)
fixation is the ribulose-1,5-bisphosphate carboxylase/oxygenase into acetyl-CoA and pyruvate, rather than in the conventional
(Rubisco) reaction, resulting in the conversion of one molecule oxidative direction (> Fig. 5.26). Most, but not all, of the
of ribulose-1,5-bisphosphate and CO2 to two molecules of enzymes will catalyze the reactions in the reverse direction.
3-phosphoglyceric acid. These are then reduced to glyceralde- Thus, the Chlorobiaceae have replaced succinic dehydrogenase
hyde 3-phosphate, followed by several rearrangements to regen- with fumarate reductase, substituted an a-ketoglutarate dehy-
erate ribulose-1,5-bisphosphate. Rubisco from the Bacteria has drogenase-ferredoxin oxidoreductase for the conventional
been classified into two types. Type I is found in plants, a-ketoglutarate dehydrogenase complex, and replaced the irre-
cyanobacteria, and several other prokaryotes and has a subunit versible citrate synthase with an ATP-citrate lyase. Since its
structure consisting of eight large and eight small subunits. Type discovery in the Chlorobiaceae, the reductive TCA cycle has
II Rubisco is found in certain purple nonsulfur bacteria and also been found in Desulfobacter hydrogenophilus in the
a few other organisms and has been found to be a dimer of d-subphylum of the Proteobacteria; in members of the
large subunits. The phylogenetic tree for Rubisco shows many Aquificae, a deeply branching hyperthermophilic phylum in
branching orders considerably different from that of the 16S the Bacteria; and in anaerobic members of the Crenarchaeota.
rRNA tree, and this has been considered strong evidence for It was thought to be present in some aerobic Crenarchaeota, but
rampant genetic transfer of this gene (Delwiche and Palmer recent results indicate that a different pathway functions in those
1996). Moreover, genes resembling those encoding Rubisco organisms (see below).
have been found in archaeal genomes, and it has been shown As mentioned previously in the discussion of anaerobic
that the corresponding proteins have Rubisco activity (Maeda respiration, acetogenic bacteria can convert H2 and CO2 into
et al. 1999; Watson et al. 1999). These Rubisco homologues form acetate, the equivalent of fixing CO2 into organic matter. Indeed,
yet another phylogenetic cluster, and the metabolic role of these many acetogens can grow in mineral medium using CO2 as
enzymes in their host organisms has yet to be determined a carbon source and are therefore autotrophs. The pathway,
because most evidence indicates that CO2 in methanogenic shown in > Fig. 5.27, leads to the fixation of two moles of
archaea is fixed via the acetogenic pathway (see below). carbon into acetate. The methyl group of acetate is synthesized
The Chlorobiaceae, known colloquially as the ‘‘green sulfur by reducing one-carbon units to methyl-tetrahydrofolate. This
bacteria,’’ were shown not to use the Calvin cycle for CO2 methyl group is transferred to a corrinoid-containing iron-
fixation. Instead, they run the tricarboxylic acid (TCA) cycle in nickel-sulfur enzyme complex called ‘‘carbon monoxide dehy-
the reverse, reductive direction, using it to fix CO2 eventually drogenase’’ or ‘‘acetate synthetase/decarbonylase.’’ The methyl
CO2 PEP
AMP + Pi ATP
oxaloacetate pyruvate
Pi
oxaloacetate CoA
2H 2H
ATP-citrate acetyl-CoA CO2

L-malate lyase
AMP + Pi
H2O ATP + CoA
fumarate citrate
2H
fumarate
reductase H 2O
succinate cis-aconitate
ATP H2O
CoA ADP + Pi
isocitrate
succinyl-CoA
α-ketoglutarate
synthase 2H
CO2 CoA
2H CO2
α-ketoglutarate
. Fig. 5.26
The reductive or reversed tricarboxylic acid cycle (Adapted from Gottschalk 1986)
group is transferred to the cobalt of a corrinoid, and CO2 is Nitrogen Fixation

reduced to the equivalent of carbon monoxide. These are then
assembled into an enzyme-bound acetyl group, which is released Fixation of N2 is carried out solely by prokaryotes. Until the
as acetyl-coenzyme A (CoA), which can be conserved as ATP in twentieth century, when a large amount of nitrogen was fixed
catabolism or can be used for biosynthesis. This pathway costs anthropogenically by the Born-Haber process, essentially all
a cell only one ATP per two carbons fixed to acetyl-CoA. This nitrogen found in eukaryotes was originally fixed by prokaryotes
pathway is also found in certain sulfate reducers in the and transferred to them either indirectly through the food chain
d-subphylum of the Proteobacteria (Menendez et al. 1999). or directly in symbiosis, as in the case of the symbioses between
A variation of this pathway using methanopterin derivatives legumes and rhizobia found in their root nodules. Nitrogen
rather than folates is found in autotrophic methanogenic fixation is found in a variety of physiological types of prokaryotes,
Archaea as well as in the sulfate-reducing Archaeoglobus occurring in aerobes, microaerophiles, anaerobes, phototrophs,
(Menendez et al. 1999). Thus, the acetogenic pathway is respon- and free-living as well as symbiotic bacteria. Phylogenetically it is
sible for much of the CO2 fixation occurring in anoxic habitats. widespread in the Bacteria, including members of the phyla
Finally, the pathway for CO2 fixation in the green nonsulfur Proteobacteria, Firmacutes, Actinobacteria, Chlorobia, and
photosynthetic bacterium Chloroflexus aurantiacus was not clear Cyanobacteria. In the Archaea, nitrogen fixation has only been
until it was demonstrated (Strauss and Fuchs 1993) that a novel found thus far in the methanogens (Lobo and Zinder 1992).
pathway based on carboxylation of acetyl-CoA and propionyl- The enzyme that is responsible for the process, nitrogenase,
CoA (> Fig. 5.28), sometimes called the ‘‘3-hydroxypropionate is a complex of two metalloprotein components. One compo-
pathway,’’ is responsible for CO2 fixation to glyoxalate in these nent usually contains an iron/sulfur/molybdenum cluster that is
organisms. The key enzymes of this pathway are acetyl-CoA considered the active site for dinitrogen reduction, as well as an
carboxylase and propionyl-CoA carboxylase. This pathway has unusual iron/sulfur cluster called the ‘‘P cluster’’ (Dean et al.
also been found in aerobic lithotrophic Crenarchaeota such as 1993). The second component is sometimes called the ‘‘Fe pro-
Sulfolobus (Menendez et al. 1999). The distribution of these four tein’’ and contains a single 4Fe-4S cluster per homodimer and
autotrophic pathways is summarized in > Table 5.4. transfers electrons to the first component, accompanied by
hydrolysis of approximately two ATPs to ADPs per electron
transferred. Because this process is so energetically expensive, it
is not surprising that most free-living organisms regulate expres-
sion of nitrogenase genes (Berger et al. 1994) and activity
CO2 CO2
2H
Formate
dehydrogenase
HCOOH
N-formyl- ATP HN-THF
tetrahydrofolate
synthetase ADP + Pi H2O
O
H C N THF
2H
2H
H2O
2H
H3C-N-THF
Co-Enz
H 2O
HN-THF
H3C-Co-Enz (CO)– ACDS
HN-THF = Tetrahydrofolate
O
Co-Enz = Corrinoid enzyme H3C C ACDS
ACDS = Acetate decarbonylase/synthetase
HS-CoA
ACDS
O
H3C C S-CoA
Pi
Acetate
kinase HS-CoA
O
H3C C Pi
Phosphotrans- ADP
acetylase
ATP
O
H3C C O–
. Fig. 5.27
The acetogenic pathway
(Ludden and Roberts 1988). Alternative nitrogenases exist in oxygen poisoning. Among the members of the genus Azotobac-
which molybdenum (Mo) is replaced by vanadium (V) or in ter, there are three mechanisms for nitrogenase protection.
which no metal other than iron has been found (Bishop and These are respiratory protection, conformational protection,
Premakumar 1992). The nitrogenase protein components con- and oxygen regulation of nitrogenase synthesis (Kennedy and
tain many highly conserved amino acid sequence motifs Toukdarian 1987). Respiratory protection occurs because Azo-
involved in cofactor binding and subunit interactions. Interest- tobacter can consume oxygen much faster than its rate of entry
ingly, these proteins are homologous to proteins involved in into the cell. These unusually high rates of respiration thus result
reductive steps of chlorophyll a biosynthesis (Fujita and Bauer in maintaining the nitrogenase in an essentially anoxic environ-
2000), providing a link between the processes of nitrogen fixa- ment. Indeed, limiting Azotobacter respiration increases their
tion and photosynthesis. sensitivity to oxygen during nitrogen fixation. Conformational
Both protein components of nitrogenase are extremely sen- protection is a result of the ability of Azotobacter to synthesize
sitive to oxygen, and the bacteria fixing nitrogen aerobically have another FeS protein that enters into an association with the
evolved a variety of strategies to protect the nitrogenase from nitrogenase complex and protects it from O2 inactivation
4H
O O
Cell carbon C C
HO H
CO2 Glyoxalate
2H 2H
OH O O O
O O O
C C CH2 C C C C C C H3C C
HO H2 HO H2 H HO H2 S-CoA S-CoA
3-hydroxypropionate CoA-SH Malonyl-CoA ATP Acetyl-CoA
Malonate
CoA-SH semialdehyde OH
O O
+ ATP C C C C
H2O H H2
HO S-CoA
O OH Matyl-CoA
C C CH2
CoA-S H2
3-hydroxypropionyl-CoA OH
O O
C C C C
HO H H2 OH
H2O
Malate
O CoA-SH
C C CH2
CoA-S H H2O
O O
Acetyl-CoA C C C C
HO H H OH
2H
Fumarate
CO2 O OH
O C O O O O
O 2H
C C CH3 C C CH3 C C C C C C C C
CoA-S H2 H
CoA-S CoA-S H2 H2 OH HO H2 H2 OH
ATP
Propionyl-CoA Methylmalonyl-CoA Succinyl-CoA Succinate
. Fig. 5.28
The 3-hydroxypropionate cycle (From Menendez et al. 1999)
(Moshiri et al. 1994). During this association, the complex is not generate any O2; photosystem I is still operative and con-
unable to manifest any nitrogenase activity. tinues to generate ATP by photophosphorylation. Second, it is
Many organisms are not as adept as Azotobacter at surrounded by a laminated structure consisting of a series of
protecting their nitrogenase from O2 and fix nitrogen only unique glycolipids that seem to act as a physical barrier to
under microaerophilic conditions, even though they otherwise prevent O2 from penetrating into the cell. Thus, the cell sepa-
are not microaerophiles. One example of this is the rhizobia rates its nitrogenase both from endogenous as well as exogenous
(Spaink 2000). They were not shown to be able to fix nitrogen O2. The heterocyst can feed the fixed, reduced nitrogen products
until 1975, nearly a century after their isolation by Beijerinck. to the adjoining vegetative cells, from which it receives the
When in the plant root nodule, the rhizobia are protected by the reducing power necessary to convert dinitrogen to amino
heme protein leghemoglobin, which has an extremely high acids. To add to the elegance of the solution, the heterocysts
affinity for O2 and transports the O2 necessary for Rhizobium are interspersed along the filament, spaced so as to provide an
growth, simultaneously preventing the access of the O2 to the optimum supply of fixed nitrogen to the growing and dividing
nitrogenase (Dilworth and Appleby 1979). vegetative cells. A peptide signal, similar to those used for quo-
The nitrogen-fixing cyanobacteria are presented with an rum sensing by Gram-positive bacteria, is used to regulate this
even greater challenge than other aerobes because O2 is one of spacing (Yoon and Golden 1998). The actinomycete Frankia,
the main products of their photosynthetic metabolism. which fixes nitrogen in nodules in alder trees and several shrubs,
Anabaena and other related filamentous, nitrogen-fixing utilizes a similar solution, forming specialized cell aggregates
cyanobacteria solve the problem of O2 poisoning of nitrogenase called ‘‘vesicles’’ (Benson and Silvester 1993).
by segregating the nitrogen-fixing enzymes in a specialized cell Some of the nonheterocystous cyanobacteria have solved the
called ‘‘the heterocyst.’’ The heterocyst insulates the nitrogenase problem of nitrogen fixation and photosynthetic O2 evolution
from O2 in two ways. First, it lacks photosystem II and thus does by separating the two processes in time rather than in space.
. Table 5.4 can grow up to 113 C and survive 1 h of autoclaving at 121 C

CO2 fixation pathways found in prokaryotes (Blochl et al. 1997), Halobacterium salinarum grows optimally at
5.2 M salt (Kushner 1985), Ferroplasma grows at pH 0 (Edwards
Pathway Representative organisms Phylum
et al. 2000), and a wide variety of organisms can grow at tem-
Calvin cycle Plant chloroplasts Cyanobacteria peratures below 0 C (Morita 1975). It should be pointed out
Anabaena cylindrical Cyanobacteria that the organisms specifically referred to above are Archaea.
Rhodobacter sphaeroides Proteobacteria While the ability to thrive under extreme conditions is not an
Thiobacillus ferrooxidans Proteobacteria exclusive property of the Archaea, they seem to be able to
withstand high temperatures, at least partially attributable to
Methanococcus jannaschiia Euryarchaeota
their unusual ether-linked lipids.
Reductive TCA cycle Chlorobium limicola Chlorobi
In most cases, the mechanisms whereby the organisms are
Desulfobacter Proteobacteria able to thrive in an extreme environment have reduced or
hydrogenophilus
eliminated the ability of that organism to tolerate the normal
Aquifex pyrophilus Aquificae or common environment. In other words, the adaptation to the
Thermoproteus Crenarchaeota extreme environment has not extended a particular organism’s
neutrophilus milieu but rather replaced one optimum with another. Thus, the
Reductive acetyl- Clostridium Firmacutes ability to tolerate, indeed the requirement for, a high salt con-
CoA pathway thermoaceticum centration renders Halobacterium salinarum fragile in a normal
Desulfobacterium Proteobacteria osmotic milieu, Pyrolobus cannot grow at below 80 C, and
autotrophicum Ferroplasma cannot grow at a neutral pH.
Methanococcus jannaschiib Euryarchaeota The ability to exploit a wide variety of ecological niches and
Ferroglobus placidus Euryarchaeota to grow optimally at environmental extremes that are intolerable
Hydroxypropionate Chloroflexus aurantiacus Chloroflexi for higher, more complex organisms is a result of the intense
cycle Sulfolobus metallicus Crenarchaeota physiological and metabolic specialization among the prokary-
otes. In a sense, then, the ability of a complex, multicellular
Data from Menendez et al. (1999)
organism to deal with its environment is a compromise among
Abbreviations: TCA Tricarboxylic acid, and CoA Coenzyme A
a
The function of Rubisco found in methanogens and other Euryarchaeota are
the various specialized and differentiated cells; the extremes that
not presently known such an organism can tolerate are the lowest common denom-
b
Pathway uses methanofuran and tetrahydromethanopterin instead of inator of the properties of its individual cells. That is the price
tetrahydrofolate paid for the operational complexity of the multicellular organ-
ism. On the other hand, the small size, structural simplicity, and
Thus, nitrogenase is synthesized, and nitrogen fixation takes unicellular nature of the prokaryote are admirably suited to
place in the dark. During the photoperiod, the nitrogenase permit adaptation to a wide variety of environmental extremes,
formed during the previous dark period is presumably destroyed with each adaptive response narrowly focused on dealing opti-
(Stal and Krumbein 1985). Indeed, this was the first evidence for mally with that particular extreme.
a biological clock in prokaryotes (Johnson and Golden 1999). It
is still not clear, however, how the filamentous nonheterocystous
colonial cyanobacterium Trichodesmium fixes nitrogen during
Conclusion
photosynthesis (Capone et al. 1997).
Finally, it should be mentioned that a thermophilic actino-
It is appropriate that we end this chapter with a quotation from
mycete utilizes a nitrogenase enzyme complex in which
the late mentor of one of us (M.D.), Professor Jackson W. Foster
a modified Mo-Fe protein is coupled to a carbon monoxide
(Foster 1964):
dehydrogenase (Ribbe et al. 1997). This system is much more
O2 resistant than the standard nitrogenases, and it is not clear " The source of the microbiologist’s strength, and at the same
why similar systems have not been found in other aerobes. In time, his refuge, is the infinitely large numbers and varieties of
summary, the variety of mechanisms devised by the prokaryotes microbes known or presumed to be extant. Compounded with
for protecting nitrogenase from O2 poisoning is an impressive those that can be modified artificially, this pool represents an
example of the strategic versatility of the prokaryotes. infinitely versatile catalyst. Because, figuratively speaking, it cat-
alyzes innumerable reactions and transformations with virtually
any material occurring in nature, the microbe has to be regarded
Adaptation to Environmental Extremes as a prime, natural resource of all men regardless of national
boundaries.... As a means of emphasizing. . . the diverse, myste-
The prokaryotes not only can tolerate the broadest spectrum of rious, and gratifying potential of the microbe, I paraphrase
environmental extremes of any group of organisms, but some of a trenchant saying made famous many years ago by the influen-
their optimum conditions for growth would roast, freeze, acid- tial American magazine, the Ladies Home Journal. Whereas they
ify, or shrivel up most other organisms. Thus, Pyrolobus fumarii were concerned with women, and I with microbes, the small
perversion I make does no injustice to either – ‘‘Never underes- Coates JD, Michaelidou U, Bruce RA, O’Connor SM, Crespi JN, Achenbach LA
(1999) Ubiquity and diversity of dissimilatory (per)chlorate-reducing bac-
timate the power of the Microbe.’’
teria. Appl Environ Microbiol 65:5234–5241
Cogdell RJ, Isaacs NW, Howard TD, McLuskey K, Fraser NJ, Prince SM
(1999) How photosynthetic bacteria harvest solar energy. J Bacteriol
181:3869–3879
References Cohn F (1875) Untersuchungen über Bacterien. Beiträge zur Biologie der
Pflanzen 1:127–222
Adams MWW (1993) Enzymes and proteins from organisms that grow near and Conway T (1992) The Entner-Doudoroff pathway: history, physiology and
above 100 degrees C. Annu Rev Microbiol 47:627–648 molecular biology. FEMS Microbiol Rev 9:1–27
Alexander M (1981) Biodegradation of chemicals of environmental concern. Cook KE, Colvin JR (1980) Evidence for a beneficial influence of cellulose
Science 211:132–138 production on growth of Acetobacter xylinum in liquid medium. Curr
Allen JR, Clark DD, Krum JG, Ensign SA (1999) A role for coenzyme M (2- Microbiol 3:203–205
mercaptoethanesulfonic acid) in a bacterial pathway of aliphatic epoxide Cord-Ruwisch R, Steitz H-J, Conrad R (1988) The capacity of hydrogenotrophic
carboxylation. Proc Natl Acad Sci USA 96:8432–8437 anaerobic bacteria to compete for traces of hydrogen depends on the redox
Angert ER, Clements KD, Pace NR (1993) The largest bacterium. Nature potential of the terminal electron acceptor. Arch Microbiol 149:350–357
362:239–241 Costerton JW, Nickel JC, Ladd TI (1986) Suitable methods for the comparative
Bak F, Pfennig N (1987) Chemolithotrophic growth of Desulfovibrio study of free-living and surface-associated bacterial populations. In:
sulfodismutans sp. nov. by disproportionation of inorganic sulfur com- Poindexter JS, Leadbetter EL (eds) Bacteria in nature, vol 2. Plenum Press,
pounds. Arch Microbiol 147:184–189 New York, pp 49–84
Balkwill DL, Maratea D, Blakemore RP (1980) Ultrastructure of a magnetotactic Costerton JW, Stewart PS, Greenberg EP (1999) Bacterial biofilms: a common
spirillum. J Bacteriol 141:1399–1408 cause of persistent infections. Science 284:1318–1322
Beja O, Aravind L, Koonin EV, Suzuki MT, Hadd A, Nguyen LP, Jovanovich SB, Cross T (1981) The monosporic actinomycetes. In: Starr MP, Stolp H, Trüper HG,
Gates CM, Feldman RA, Spudich JL, Spudich EN, DeLong EF (2000) Bacte- Balows A, Schlegel HG (eds) The prokaryotes. Springer, Berlin, pp 2102–
rial rhodopsin: evidence for a new type of phototrophy in the sea. Science 2091, www.prokaryotes.com
289:1902–1906 De Gier JW, Lubben M, Reijnders WN, Tipker CA, Slotboom DJ, van Spanning
Benson DR, Silvester WB (1993) Biology of Frankia strains, actinomycete sym- RJ, Stouthamer AH, van der Oost J (1994) The terminal oxidases of
bionts of actinorhizal plants. Microbiol Rev 57:293–319 Paracoccus denitrificans. Mol Microbiol 13:183–196
Berg HC (1976) How spirochaetes may move. J Theor Biol 56:269–273 Dean DR, Bolin JT, Zheng LM (1993) Nitrogenase metalloclusters—structures,
Berger DK, Narberhaus F, Kustu S (1994) The isolated catalytic domain of NIFA, organization, and synthesis. J Bacteriol 175:6737–6744
bacterial enhancer-binding protein, activates transcription in vitro: activa- Delwiche CF, Palmer JD (1996) Rampant horizontal transfer and duplication of
tion is inhibited by NIFL. Proc Natl Acad Sci USA 91:103–107 rubisco genes in eubacteria and plastids. Mol Biol Evol 13:873–882
Beveridge TJ (1995) The periplasmic space and periplasm in gram-positive and DeWeerd KA, Mandelco L, Tanner RS, Woese CR, Suflita JM (1990) Desulfomonile
Gram-negative bacteria. ASM News 61:125–130 tiedjei gen. nov. and sp. nov., a novel anaerobic dehalogenating, sulfate-
Bishop PE, Premakumar R (1992) Alternative nitrogen fixation systems. In: reducing bacterium. Arch Microbiol 154:23–30
Stacey G, Burris RH, Evans HJ (eds) Biological nitrogen fixation. Chapman Dilworth MJ, Appleby CA (1979) Leghemogiobin and Rhizohium hemoproteins.
and Hall, New York, pp 736–762 In: Hardy RWF, Bottomly F, Burns RC (eds) A treatise on dinitrogen fixation.
Blochl E, Rachel R, Burggraf S, Hafenbradl D, Jannasch HW, Stetter KO Sec I, II. Wiley, New York, pp 691–764
(1997) Pyrolobus fumarii, gen. and sp. nov., represents a novel group of DiMarco AA, Bobik TA, Wolfe RS (1990) Unusual coenzymes of methanogenesis.
archaea, extending the upper temperature limit for life to 113 degrees C. Annu Rev Biochem 59:355–394
Extremophiles 1:14–21 Dimroth P, Schink B (1998) Energy conservation in the decarboxylation of
Boetius A, Ravenschlag K, Schubert CJ, Rickert D, Widdel F, Gieseke A, Amann R, dicarboxylic acids by fermenting bacteria. Arch Microbiol 170:69–77
Jorgensen BB, Witte U, Pfannkuche O (2000) A marine microbial consor- Distel DL, Cavanaugh CM (1994) Independent phylogenetic origins of
tium apparently mediating anaerobic oxidation of methane. Nature methanotrophic and chemoautotrophic bacterial endosymbioses in marine
407:623–626 bivalves. J Bacteriol 176:1932–1938
Brun YV, Shimkets LJ (2000) In: Brun YV, Shimkets LJ (eds) Prokaryotic devel- Dow CS, Whittenbury R (1979) Prosthecate bacteria. In: Parish JH (ed) Devel-
opment. ASM Press, Washington, DC, pp 1–500 opmental biology of prokaryotes. University of California Press, Berkeley
Burchard RP (1984) Gliding motility and taxes. In: Rosenberg E (ed) Dunny CM, Brown BL, Clewell DB (1978) Induced cell aggregation and mating in
Myxobacteria: development and cell interactions. Springer, New York, Streptococcus faecalis: evidence for a bacterial sex hormone. Proc Nat Acad
pp 139–161 Sci USA 75:3479–3483
Canale-Parola E (1978) Motility and chemotaxis in spirochaetes. Annu Rev Dworkin M (1973) Cell-cell interactions in the myxobacteria. In: Ashworth JM,
Microbiol 32:69–99 Smith JE (eds) Microbial differentiation. Cambridge University Press,
Capone DG, Zehr JP, Paerl HW, Bergman B, Carpenter EJ (1997) Trichodesmium, London, pp 125–142
a globally significant marine cyanobacterium. Science 276:1221–1229 Dworkin M, Foster JW (1956) Studies on Pseudomonas methanica (Sohngen) nov.
Carr NG (1979) Differentiation in filamentous cyanobacteria. In: Parish JH (ed) comb. J Bacteriol 72:646–659
Developmental biology of prokaryotes. University of California Press, Berke- Edwards KJ, Bond PL, Gihring TM, Banfield JF (2000) An archaeal iron-oxidizing
ley, pp 167–201 extreme acidophile important in acid mine drainage. Science 287:1796–1799
Chater KE, Hopwood DA (1973) Differentiation in actinomycetes. In: Ashworth Eichler B, Pfennig N (1986) Characterization of a new platelet-forming purple
JM, Smith JE (eds) Microbial differentiation. Cambridge University Press, sulfur bacterium. Amoebobacter pedioformis sp. nov. Arch Microbiol
London, pp 143–160 146:295–300
Chistoserdova L, Vorholt JA, Thauer RK, Lidstrom ME (1998) C1 transfer Emerson D, Moyer C (1997) Isolation and characterization of novel iron-
enzymes and coenzymes linking methylotrophic bacteria and methanogenic oxidizing bacteria that grow at circumneutral pH. Appl Environ Microbiol
Archaea. Science 281:99–102 63:4784–4792
Clements KD, Bullivant S (1991) An unusual symbiont from the gut of Fosnaugh K, Greenberg EP (1988) Motility and chemotaxis of Spirochaeta
surgeonfishes may be the largest known prokaryote. J Bacteriol aurantia: computer-assisted motion analysis. J Bacteriol 170:1768–1774
173:5359–5362 Fossing H, Gallardo VA, Jorgensen BB, Huttel M, Nielsen LP, Schulz H, Canfield
DE, Forster S, Glud RN, Gundersen JK, Kuver J, Ramsing NB, Teske A,
Thamdrup B, Ulloa O (1995) Concentration and transport of nitrate by the Kengen SWM, Stams AJM, de Vos WM (1996) Sugar metabolism in hyperther-
mat-forming sulphur bacterium Thioploca. Nature 374:713–715 mophiles. FEMS Microbiol Rev 18:119–137
Foster JW (1964) Speculative discourse on where and how microbiological Kennedy C, Toukdarian A (1987) Genetics of azotobacters: applications to nitro-
science as such may advance with application. In: Starr MP (ed) Global gen fixation and related aspects of metabolism. Annu Rev Microbiol
impacts of applied microbiology. Wiley, New York, pp 61–73 41:227–258
Fujita Y, Bauer CE (2000) Reconstitution of light-independent protochloro- Kreig N (1984) Aerobic, microaerophilic, motile, helical/vibroid Gram-negative
phyllide reductase from purified bchl and BchN-BchB subunits: in vitro bacteria. In: Krieg NR, Holt JG (eds) Bergey’s manual of determinative
confirmation of nitrogenase-like features of a bacteriochlorophyll biosyn- bacteriology, vol 1. Williams and Wilkins, Baltimore, pp 71–124, http://
thesis enzyme. J Biol Chem 275:23583–23588 www.cme.msu.edu//bergeys
Garrity GM, Holt JG (2001) A road map to the manual. In: Garrity GA, Boone Kutzner HJ (1981) The family Streptomycetaceae. In: Starr MP, Stolp H, Trüper
DR, Castenholz RW (eds) Bergey’s manual of systematic bacteriology, vol 1, HG, Balows A, Schlegel HG (eds) The prokaryotes. Springer, Berlin,
2nd edn. Springer, New York, pp 119–166, http://www.cme.msu.edu// pp 2028–2090
bergeys Lacy J (1971) Thermoactinomyces sacchari sp. nov., a thermophilic actinomycete
Giovannoni SJ, Schabatach E, Castenholz R (1987) Isosphaera pallida, gen. and causing bagassosis. J Gen Microbiol 66:327–338
comb. nov., a gliding, budding eubacterium from hot springs. Arch Lapidus IR, Berg HC (1982) Gliding motility of Cytophaga strain U67. J Bacteriol
Microbiol 147:276–284 151:384–398
Gottschalk G (1986) Bacterial metabolism. Springer, New York, pp 1–380 Lechevalier H, Holbert PE (1965) Electron microscopic observations of the
Gottschalk G, Andreeson JR, Hippe H (1981) The genus Clostridium (non- sporangial structure of a strain of Actinoplanes. J Bacteriol 89:217–222
medical aspects). In: Starr MP, Stolp H, Trüper HG, Balows A, Schlegel HG Lobo AL, Zinder SH (1992) Nitrogen fixation by methanogenic bacteria. In:
(eds) The prokaryotes. Springer, Berlin, pp 1176–1803, http://www.prokary- Evans H, Stacey G, Burris R (eds) Biological nitrogen fixation. Routledge,
otes.com Chapman and Hall, New York, pp 191–211
Gould SJ (1996) The power of the modal bacter, or why the tail can’t wag the dog. Lovley DR, Coates JD (2000) Novel forms of anaerobic respiration of environ-
Full House Three Rivers Press, New York, pp 167–216 mental relevance. Curr Opin Microbiol 3:252–256
Gray MW, Burger G, Lang BF (1999) Mitochondrial evolution. Science Lovley DR, Phillips EJ (1992) Reduction of uranium by Desulfovibrio
283:1476–1481 desulfuricans. Appl Environ Microbiol 58:850–856
Grimm AC, Harwood CS (1997) Chemotaxis of Pseudomonas spp. to the Ludden PW, Roberts EA (1988) Regulation of nitrogenase activity by reversible
polyaromatic hydrocarbon naphthalene. Appl Environ Microbiol ADP-ribosylation of dinitrogenase reductase. In: Boethe H, de Bruijn FJ,
63:4111–4115 Newton WE (eds) Nitrogen fixation: hundred years after. Gustav Fischer
Gupta RS (1998) Protein phylogenies and signature sequences: a reappraisal of Verlag, Stuttgart, pp 157–162
evolutionary relationships among archaebacteria, eubacteria, and eukary- Lünsdorf H, Reichenbach H (1989) Ultrastructural details of the apparatus of
otes. Microbiol Mol Biol Rev 62:1435–1491 gliding motility of Myxococcus fulvus (Myxobacterales). J Gen Microbiol
Haugland RA, Schlemm DJ, Lyons RP, Sferra PR, Chakrabarty AM (1990) 135:1633–1641
Degradation of the chlorinated phenoxyacetate herbicides 2,4- Macnab RM (1999) The bacterial flagellum: reversible rotary propellor and
dichlorophenoxyacetic acid and 2,4,5, thrichlorophenoxyacetic acid in pure type III export apparatus. J Bacteriol 181:7149–7153
and mixed bacterial cultures. Appl Environ Microbiol 56:1357–1362 Maeda N, Kitano K, Fukui T, Ezaki S, Atomi H, Miki K, Imanaka T
Henrichsen J (1983) Twitching motility. Annu Rev Microbiol 37:81–93 (1999) Ribulose bisphosphate carboxylase/oxygenase from the hyperther-
Hoeniger JFM (1965) Development of flagella by Proteus mirabilis. J Gen mophilic archaeon Pyrococcus kodakaraensis KOD1 is composed
Microbiol 40:29–42 solely of large subunits and forms a pentagonal structure. J Mol Biol
Holliger C, Wohlfarth G, Diekert G (1998) Reductive dechlorination in the energy 293:57–66
metabolism of anaerobic bacteria. FEMS Microbiol Rev 22:383–398 Mägli A, Wendt M, Leisinger T (1996) Isolation and characterization of
Holt SC, Canale-Parola E (1967) Fine structure of Sarcina maxima and Sarcina Dehlobacterium formicoaceticum gen. nov. sp. nov., a strictly anaerobic bac-
ventriculi. J Bacteriol 93:399–410 terium utilizing dichloromethane as a source of carbon and energy. Arch
Huebner A, Danganan CE, Xun L, Chakrabarty AM (1998) Genes for 2,4,5- Microbiol 166:101–108
trichlorophenoxyacetic acid metabolism in Burkholderia cepacia AC1100: Malmquist A, Welander T, Moore E, Ternstrom A, Molin G, Stentrom IM
characterization of the tftC and tftD genes and locations of the tft operons (1994) Edeonella dechloratans, gen. nov., sp. nov., a new bacterium capable
on multiple replicons. Appl Environ Microbiol 64:2086–2093 of growing anaerobically with chlorate as an electron acceptor. Syst Appl
Hugenholtz P, Goebel BM, Pace NR (1998) Impact of culture-independent Microbiol 17:58–64
studies on the emerging phylogenetic view of bacterial diversity. J Bacteriol Matsuyama T, Takagi Y, Nakagawa Y, Itoh H, Wakita J, Matsushita M (2000)
180:4765–4774 Dynamic aspects of the structured cell population in a swarming colony of
Ingraham JL, Maaløe O, Neidhardt FC (1983) Growth of the bacterial cell. Proteus mirabilis. J Bacteriol 182:385–393
Sinauer Associates, Sunderland Maymó-Gatell X, Chien YT, Gossett JM, Zinder SH (1997) Isolation of
Jarrell KF, Bayley DP, Kostyukova AS (1996) The archaeal flagellum: a unique a bacterium that reductively dechlorinates tetrachloroethene to ethene.
motility structure. J Bacteriol 178:5057–5064 Science 276:1568–1571
Johnson CH, Golden SS (1999) Circadian programs in cyanobacteria: adaptive- Mayr E (1998) Two empires or three? Proc Natl Acad Sci USA 95:9720–9723
ness and mechanism. Annu Rev Microbiol 53:389–409 Menendez C, Bauer Z, Huber H, Gad’on N, Stetter KO, Fuchs G (1999) Presence
Judson HF (1979) Afterward: always the same impasse. In: The eighth day of of acetyl coenzyme A (CoA) carboxylase and propionyl-CoA carboxylase
creation. Simon and Schuster, New York, pp 605–616 in autotrophic Crenarchaeota and indication for operation of
Kandler O (1994) Cell wall biochemistry and three-domain concept of life. Syst a 3-hydroxypropionate cycle in autotrophic carbon fixation. J Bacteriol
Appl Microbiol 16:501–509 181:1088–1098
Kandler O, Weiss N (1986) Regular, non-sporing Gram-positive rods. In: Sneath Merz AJ, So M, Sheetz MP (2000) Pilus retraction powers bacterial twitching
PHA, Mair NS, Sharpe ME, Holt JG (eds) Bergey’s manual of systematic motility. Nature 407:98–102
bacteriology, vol 2. Williams and Wilkins, Baltimore, pp 1208–1260, http:// Messmer M, Wohlfarth G, Diekert G (1993) Methyl chloride metabolism of the
www.cme.msu.edu//bergeys strictly anaerobic, methyl chloride-utilizing homoacetogen strain MC. Arch
Katz J, Weihs D (1979) Large amplitude on steady motion of a flexible slender Microbiol 160:383–387
propulsor. J Fluid Mech 90:713–723 Mohn WW, Tiedje JM (1992) Microbial reductive dehalogenation. Microbiol Rev
Keller KH, Grady M, Dworkin M (1983) Surface tension gradients: feasible model 56:482–507
for gliding motility of Myxococcus xanthus. J Bacteriol 155:1358–1366
Moran NA, Baumann P (2000) Bacterial endosymbionts in animals. Curr Opin Schulz HN, Brinkhoff T, Ferdelman TG, Marine MH, Teske A, Jorgensen BB
Microbiol 3:270–275 (1999) Dense populations of a giant sulfur bacterium in Namibian shelf
Moreira D, Le Guyader H, Phillippe H (2000) The origin of red algae and the sediments. Science 284:493–495
evolution of chloroplasts. Nature 405:69–72 Selig M, Xavier KB, Santos H, Schonheit P (1997) Comparative analysis of
Morita RY (1975) Psychrophilic bacteria. Bacteriol Rev 39:144–167 Embden-Meyerhof and Entner-Doudoroff glycolytic pathways in hyperther-
Moshiri F, Kim JW, Fu C, Maier RJ (1994) The FeSII protein of Azotobacter mophilic archaea and the bacterium Thermotoga. Arch Microbiol
vinelandii is not essential for aerobic nitrogen fixation, but confers signifi- 167:217–232
cant protection to oxygen-mediated inactivation of nitrogenase in vitro and Singleton R (2000) Historical roots. ASM News 66:510–511
in vivo. Mol Microbiol 14:101–114 Sleytr UB, Messner P, Pum D, Sara M (1993) Crystalline bacterial cell surface
Nichols JM, Adams DG (1982) Akinetes. In: Carr NG, Whitton BA (eds) The layers. Mol Microbiol 10:911–916
biology of cyanobacteria. University of California Press, Berkeley, Sneath PHA (1986) Endospore forming Gram-positive rods and cocci. In: Krieg
pp 387–412 NR, Holt JG, Krieg NR, Holt JG (eds) Bergey’s manual of determinative
Palleroni NJ (1984) Gram-negative aerobic rods and cocci. In: Krieg NR, Holt JG bacteriology, vol 2. Williams and Wilkins, Baltimore, pp 1104–1207, http://
(eds) Bergey’s manual of determinative bacteriology, vol 1. Williams and www.cme.msu.edu//bergeys
Wilkins, Baltimore, pp 140–406, http://www.cme.msu.edu//bergeys Sogin ML, Silberman JD (1998) Evolution of the protists and protistan parasites
Pangborn J, Kuhn DA, Woods JR (1977) Dorsal-ventral differentiation in from the perspective of molecular systematics. Int J Parasitol 28:11–20
Simonsiella and other aspects of its morphology and ultrastructure. Arch Sowers KR, Boone JE, Gunsalus RP (1993) Disaggregation of Methanosarcina
Microbiol 113:197–204 spp. and growth as single cells at elevated osmolarity. Appl Environ
Pate JL, Chang LYE (1979) Evidence that gliding motility in prokaryotic cells is Microbiol 59:3832–3839
driven by rotary assemblies in the cell envelope. Curr Microbiol 2:257–262 Spaink HP (2000) Root nodulation and infection factors produced by rhizobial
Pate JL, Ordal EJ (1965) The fine structure of two unusual stalked bacteria. J Cell bacteria. Annu Rev Microbiol 54:257–288
Biol 27:133–150 Spormann AM (1999) Gliding motility in bacteria: insights from studies of
Peekhaus N, Conway T (1998) What’s for dinner?: Entner-Doudoroff metabolism Myxococcus xanthus. Microbiol Mol Biol Rev 63:621–641
in escherichia coli. J Bacteriol 180:3495–3502 Spudich JL (1993) Color sensing in the archaea—a eukaryotic-like receptor
Perazzona B, Spudich JL (1999) Identification of methylation sites and effects of coupled to a prokaryotic transducer. J Bacteriol 175:7755–7761
phototaxis stimuli on transducer methylation in Halobacterium salinarum. Stackebrandt E, Ludwig W, Weizenegger M, Dorn S, McGill TJ, Fox GE,
J Bacteriol 181:5676–5683 Woese CW, Schubert W, Schleifer KH (1987) Comparative 16S rRNA oligo-
Poindexter JS (1964) Biological properties and classification of the Caulobacter nucleotide analyses and murein types of round-spore-forming bacilli and
group. Bacteriol Rev 28:231–295 non-spore-forming relatives. J Gen Microbiol 133:2523–2529
Ragsdale SW (1997) The eastern and western branches of the Wood/Ljungdahl Stackebrandt E, Rainey FA, Ward-Rainey N (1996) Anoxygenic phototrophy
pathway: how the east and west were won. Biofactors 6:3–11 across the phylogenetic spectrum: current understanding and future per-
Reeve JN, Sandman K, Daniels CJ (1997) Archaeal histones, nucleosomes, and spectives. Arch Microbiol 166:211–223
transcription initiation. Cell 89:999–1002 Stal LJ, Krumbein WE (1985) Nitrogenase activity in the non-heterocystous
Reichenbach H (1984) Myxobacteria: a most peculiar group of social prokaryotes. cyanobacterium Oscillatoria sp. grown under alternating light–dark cycles.
In: Rosenberg E (ed) Myxobacteria: development and cell interactions. Arch Microbiol 143:67–71
Springer, Berlin, pp 1–50 Staley JT (1968) Prosthecomicrobium and Ancalomicrobium: new prosthecate
Reichenbach H, Dworkin M (1981a) Introduction to the gliding bacteria. In: freshwater bacteria. J Bacteriol 95:1921–1942
Starr MP, Stolp H, Trüper HG, Balows A, Schlegel HG (eds) The prokaryotes. Stanier RY, van Niel CB (1962) The concept of a bacterium. Arch Mikrobiol
Springer, Berlin, pp 316–327, http://www.prokaryotes.com 42:17–35
Reichenbach H, Dworkin M (1981b) The order myxobacterales. In: Starr MP, Stanier RY, Adelberg EA, Ingraham JL, Wheelis ML (1979) Microbial metabolism:
Stolp H, Trüper HG, Balows A, Schlegel HG (eds) The prokaryotes. Springer, the generation of ATP Introduction to the microbial World. Prentice-Hall,
Berlin, pp 328–355, http://www.prokaryotes.com New York, pp 154–186
Ribbe M, Gadkari D, Meyer O (1997) N2 fixation by Streptomyces thermoauto- Stanier RY, Pfennig N, Trüper HG (1981) Introduction to the phototrophic
trophicus involves a molybdenum-dinitrogenase and a manganese- prokaryotes. In: Starr MP, Stolp H, Trüper HG, Balows A, Schlegel HG
superoxide oxidoreductase that couple N2 reduction to the oxidation of (eds) The prokaryotes. Springer, Berlin, pp 197–211, http://www.prokary-
superoxide produced from O2 by a molybdenum-CO dehydrogenase. otes.com
J Biol Chem 272:26627–26633 Stephenson M (1949) Bacterial metabolism, 3rd edn. Longmans, Green, London
Robinson JJ, Stein JL, Cavanaugh CM (1998) Cloning and sequencing of a form II Stewart V (1994) Regulation of nitrate and nitrite reductase synthesis in
ribulose-1,5-biphosphate carboxylase/oxygenase from the bacterial symbi- enterobacteria. Antonie van Leeuwenhoek 66:37–45
ont of the hydrothermal vent tubeworm Riftia pachyptila. J Bacteriol Stokes J (1954) Studies on the filamentous sheathed iron bacterium Sphaerotilus
180:1596–1599 natans. J Bacteriol 67:278–291
Rosenberg E, Keller KH, Dworkin M (1977) Cell density-dependent growth of Strauss G, Fuchs G (1993) Enzymes of a novel autotrophic CO2 fixation pathway
Myxococcus xanthus on casein. J Bacteriol 129:770–779 in the phototrophic bacterium Chloroflexus aurantiacus, the 3-
Rudolph J, Oesterhelt D (1996) Deletion analysis of the che operon in the hydroxypropionate cycle. Eur J Biochem 215:633–643
archaeon Halobacterium salinarium. J Mol Biol 258:548–554 Strous M, Fuerst JA, Kramer EH, Logemann S, Muyzer G, van de Pas-
Sadoff HL (1976) Encystment and germination in Azotobacter vinelandii. Schoonen KT, Webb R, Kuenen JG, Jetten MS (1999) Missing lithotroph
Bacteriol Rev 39:516–539 identified as new planctomycete. Nature 400:446–449
Saraste M (1999) Oxidative phosphorylation at the fin de siecle. Science Sudo SZ, Dworkin M (1973) Comparative biology of prokaryotic resting cells.
283:1488–1493 Adv Microbiol Physiol 9:153–224
Schäfer G, Purschke W, Schmidt CL (1996) On the origin of respiration: electron Suflita JM, Horowitz A, Shelton DR, Tiedje JM (1982) Dehalogenation: a novel
transport proteins from archaea to man. FEMS Microbiol Rev 18:173–188 pathway for the anaerobic biodegradation of haloaromatic compounds.
Schink B (1997) Energetics of syntrophic cooperation in methanogenic degrada- Science 218:1115–1117
tion. Microbiol Rev 61:262–280 Thauer RK, Jungermann K, Decker K (1977) Energy conservation in
Schleifer KH (1986) Gram-positive cocci. In: Sneath PHA, Mair NS, Sharpe ME, chemotrophic anaerobic bacteria. Bacteriol Rev 41:100–180
Holt JG (eds) Bergey’s manual of systematic bacteriology, vol 2. Williams and Tomitani A, Okada K, Miyashita H, Matthijs HC, Ohno T, Tanaka A (1999)
Wilkins, Baltimore, pp 999–1002, http://www.cme.msu.edu//bergeys Chlorophyll b and phycobilins in the common ancestor of cyanobacteria and
chloroplasts. Nature 400:159–162
van Niel CB (1949) The comparative biochemistry of photosynthesis. In: Widdel F, Schnell S, Heising S, Ehrenreich A, Assmus B, Schink B (1993)
Franck J, Loomis WE (eds) Photosynthesis in plants, vol 400. Iowa State Ferrous iron oxidation by anoxygenic phtotrophic bacteria. Nature
College Press, Ames, pp 437–495 362:834–836
Vargas M, Kashefi K, Blunt-Harris EL, Lovley DR (1998) Microbiological evi- Williams FD (1978) Nature of the swarming phenomenon in Proteus. Annu Rev
dence for Fe(III) reduction on early Earth. Nature 395:65–67 Microbiol 32:101–122
Walsby AE (1980) A square bacterium. Nature 283:69–71 Wilson EO (1995) Naturalist. Warner Books, New York, pp 101–122
Watson GM, Yu JP, Tabita FR (1999) Unusual ribulose 1,5-bisphosphate carbox- Woese CR (1998) Default taxonomy: Ernst Mayr’s view of the microbial world.
ylase/oxygenase of anoxic Archaea. J Bacteriol 181:1569–1575 Proc Natl Acad Sci USA 95:11043–11046
Weiner JH, MacIsaac DP, Bishop RE, Bilous PT (1988) Purification and proper- Woese CR, Kandler O, Wheelis ML (1990) Towards a natural system of organisms:
ties of Escherichia coli dimethyl sulfoxide reductase, an iron-sulfur proposal for the domains Archaea, Bacteria, and Eucarya. Proc Natl Acad Sci
molybdoenzyme with broad substrate specificity. J Bacteriol 170:1505–1510 USA 87:4576–4579
Weisburg WG, Hatch TP, Woese CR (1986) Eubacterial origin of chlamydiae. Wolin MJ, Miller TL (1982) Interspecies hydrogen transfer: 15 years later. ASM
J Bacteriol 167:570–574 News 48:561–565
White D (1984) Structure and function of myxobacteria cells and fruiting bodies. Xiong J, Fischer WM, Inoue K, Nakahara M, Bauer CE (2000) Molecular evidence
In: Rosenberg E (ed) Myxobacteria: development and cell interactions. for the early evolution of photosynthesis. Science 289:1724–1730
Springer, Berlin, pp 51–67 Yoon HS, Golden JW (1998) Heterocyst pattern formation controlled by
Whittenbury R, Dow CS (1977) Morphogenesis and differentiation in a diffusible peptide. Science 282:935–938
Rhodomicrobium vannielii and other budding and prosthecatebacteria. Zengler K, Richnow HH, Rossello-Mora R, Michaelis W, Widdel F (1999) Meth-
Bacteriol Rev 41:754–808 ane formation from long-chain alkanes by anaerobic microorganisms.
Whittenbury R, Davies SL, Davy JF (1970) Exospores and cysts formed by Nature 401:266–269
methane-utilizing bacteria. J Gen Microbiol 61:219–226 Zumft WG (1997) Cell biology and molecular basis of denitrification. Microbiol
Mol Biol Rev 61:533–616
6 Prokaryote Characterization and
Identification
Peter Kämpfer . Stefanie P. Glaeser
Institut für Angewandte Mikrobiologie, Justus-Liebig-University Giessen, Gießen, Germany
Systematics of Prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Cytoplasmatic Compounds . . . . . . . . . . . . . . . . . . . . . . . . . 142

Some Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Other Features of Taxonomic Value . . . . . . . . . . . . . . . . . . . . . 142
Nomenclature of Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

Importance of Type Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
The Provisional Status Candidatus . . . . . . . . . . . . . . . . . . . . . . 128
Minimal Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 Systematics of Prokaryotes∗
From the Beginning of Prokaryotic Classification and Some Definitions

Towards a Modern Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Systematics can be defined as the scientific study of organisms
Current Procedures for the Characterization and with the ultimate goal to characterize and arrange organisms in
Identification of Bacterial and Archaeal Strains . . . . . . . . . . 130 an orderly manner. Systematics also might be defined as ‘‘the
Genotypic Traits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 study of organismal diversity and interrelationships.’’ As pointed
Phylogenetic Assignment Based on 16S rRNA Gene out already by Cowan (1968), systematics includes taxonomy
Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 and aspects of ecology, biochemistry, microscopy, pathology,
DNA–DNA Hybridization . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 genetics, and molecular biology.
DNA Base Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Taxonomy and systematics are often used synonymously,
Phylogenetic Assignment Based on Alternative but it is more appropriate to regard taxonomy as a part of
Phylogenetic Markers Including Protein Coding systematics (Tindall et al. 2007). Taxonomy was defined as
Genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 ‘‘the art of biological classification’’ (Stanier et al. 1986) but
MultiLocus Sequence Analysis (MLSA) . . . . . . . . . . . . . 133 also as the theoretical study of classification, including its
Genome-Based Analysis: GEBA Project: Toward bases, principles, and rules (Simpson 1961). As stated by
a New Area in Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Cowan (1968), taxonomy is traditionally divided into
The Problem of Horizontal Gene
1. Classification, the orderly arrangement of units into taxo-
Transfer (HGT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
nomic groups
Nucleic Acid Fingerprinting . . . . . . . . . . . . . . . . . . . . . . . . . 135
2. Nomenclature, the labeling of units defined by
Phenotypic Traits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
classification
Colony Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
3. Identification of unknowns to units defined by classification
Cell Shape and Cell Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
and labeled by nomenclature
Staining Behavior of the Cell . . . . . . . . . . . . . . . . . . . . . . . . 136
Cellular Motility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 Identification is the practical application on the foundation
Flagellation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 of classification and nomenclature. Classification and identifi-
Sporulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 cation are sometimes confused but classification is rather
Cellular Inclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 a prerequisite for identification. Identification may be under-
Pigmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 stood either transitively (identification of unknowns with units
Bacteriochlorophylls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 defined by classification) or intransitively (to describe the iden-
Cytochromes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 tity of a species as such and use it as a basis for classification)
Ultrastructural Characters . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 (Trüper and Schleifer 2006). Two functions of taxonomy have
Physiological Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 been considered by Stanier et al. (1986), the identification and
Numerical Taxonomy and Numerical description of basic taxonomic units or species and the devising
Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 of methods to arrange and catalogue these units.
New Ways for Phenotypic Characterization . . . . . . . . 139
Chemotaxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Cell Wall and Membrane Compounds . . . . . . . . . . . . . . 139 ∗
The basis of this section has been the excellent short review of Tindall et al.
(2007) supplemented with further details given by Tindall et al. (2010)
124 6 Prokaryote Characterization and Identification
New prokaryotic isolates

(pure culture)
Partial or nearly full length Fatty acid profiling,

16S rRNA gene sequence MALDI-TOF,
API/Biolog-Systems
First assignment to already

described taxa (genus level)
Full-length 16S rRNA gene (First) characterization/

sequence analysis Identification
Phylogenetic affiliation to most

closely related taxa (genus level)
Identification of a new taxon
Selection of type Selection of criteria

strains needed for needed for detailed
direct comparison characterization
Genotypic Phenotypic characterization:

characterization Phenotype Chemotype
(Detailed)
16S rRNA gene sequence Cell and colony Fatty acid profiles
(if not done already) Morphology Polar lipid profiles
Characterization
DNA Mol% G+C content Physiological/ Quinone pattern
DNA-DNA hybridization biochemical properties Polyamine pattern
Further gene sequences/ Pigmentation Cell wall structure
MLSA Substrate utilization Peptidoglycans
(Partial) genome sequences Enzyme activities Teichoic acids
MALDI-TOF
Comparative analysis
Assignment to Differentiation of
Identification existing taxa related taxa Classification
(genus or species) (genus or species)
Definition of a new taxon Nomenclature

(genus of species)
Deposit of type strains Submission of gene sequences

In two culture collections to GenBank/EMBL/DDBJ
Proposal of a new taxon

(genus or species)
. Fig. 6.1
Flow diagram represents the steps currently performed for taxonomic characterization and description of new taxa. Furthermore the
flow diagram outlines the interrelationship of characterization, identification, classification and nomenclature in taxonomy. Parts of the
diagram were adapted from Rainey and Oren (2012) and Trüper and Schleifer (2006)
The flow diagram in > Fig. 6.1 visualizes the interrelation- a population of ideally genotypically identical cells, with the
ship of the classification, nomenclature, and identification. cell as basic unit.
It should be the primary aim of taxonomy to provide In contrast, taxonomic categories (including the ‘‘basic
a classification that can be used for a wide variety of purposes unit,’’ the species) are mental representations sometimes defined
(including identification) and serving all microbiological dis- as ‘‘units of knowledge’’ rather than physical objects.
ciplines. The basic physical object in microbiology to be There are many ways to classify organisms, and there is
classified is the organism, which is (in most cases) represented an ongoing debate about this issue, but in any case any
as pure bacterial or archaeal culture, which represents resulting classification system remains an abstract (mental)
Prokaryote Characterization and Identification 6 125
representation. Wayne et al. (1987) pointed out already that an a genotypic, most often only single-step phylogenetic taxonomy
ideal taxonomy would involve one system, which should be based on the 16S rRNA gene and a few more gene sequences.
a hierarchical system, and in microbiology, the ultimate goal Genes and genomes, however, do not function on their own
should be to establish a system that mirrors the taxonomic and can display their potential only within the cell as the basic
relationships as an ‘‘order in nature,’’ which is now most unit of evolution (Tindall et al. 2007). It is the phenotype and
often associated with ‘‘evolutionary order’’ back to the origin the natural selection that ‘‘drive’’ evolution in a given environ-
of life (Tindall et al. 2007). ment. In this context, the ‘‘polyphasic taxonomic approach’’ is
A major step forward toward this goal was the analysis of the still important, and novel insights into genomes and other
small subunit ribosomal RNA (16S rRNA) gene, which has ‘‘omic’’ sciences should be brought in a more strict and detailed
clearly revolutionized prokaryotic taxonomic studies. For the context with the phenotype (Kämpfer 2012).
first time, the establishment of a hierarchical taxonomic system
on the basis of a molecular marker was possible.
A reliable classification system is a prerequisite for the Nomenclature of Bacteria
scientific work with microorganisms to keep track of the
large variety of those organisms. The currently applied classifi- The nomenclature of prokaryotes in a (hierarchical) taxonomic
cation is an operational-based approach, which depends more or system is part of the general scheme for a system of all organisms
less on data-driven analyses and is not based on a unifying developed already by Carl von Linné and published in 1735 in
theory. his Systema Naturae. To this system, the highest taxonomic ranks
A set of cutoff values and specific criteria are used for the called ‘‘domains’’ were added after the discovery of the three
delineation of group or taxa that share specific values of simi- domains of life by the pioneering work on the small subunit
larities. The most well-known ‘‘cutoff value’’ used for the differ- rRNA (16S rRNA) sequences of Woese and colleagues (e.g.,
entiation of species is still the 70% DNA–DNA similarity value Woese and Fox 1977; Stackebrandt and Woese 1981; Woese
and the 5% difference in melting temperature of the respective 1987; Woese et al. 1990).
DNA–DNA hybrids (Wayne et al. 1987). Based on 16S rRNA gene data, all prokaryotes are classified
But the currently applied system includes in addition to this into the domains ‘‘Archaea’’ or ‘‘Bacteria,’’ which are subdivided
cutoff value still a polyphasic approach (Colwell 1970; in a hierarchical manner into the lower nonoverlapping ranks:
Vandamme et al. 1996), which comprises phenotypic, and fur- ‘‘phylum,’’ ‘‘class,’’ ‘‘order,’’ ‘‘family,’’ ‘‘genus,’’ and ‘‘species’’
ther genotypic information. The phenotypic analysis includes (Brenner et al. 2001), and all these ranks are sometimes (not
morphological, physiological, and by definition also chemotax- consistently) subdivided into lower ranks using the suffix ‘‘sub-’’
onomic properties. As outlined below, phenotypic data, espe- like, for example, ‘‘suborder’’ or ‘‘subspecies.’’
cially chemotaxonomic data, may also often reflect phylogenetic The basic unit of the taxonomic hierarchy is essentially the
relationships of prokaryotes and are often suitable to assign and/ species. The definition of a prokaryotic species was and is still
or differentiate new taxa to related ones. The genotypic under discussion among microbiologist of different disciplines.
approach in contrast is based on genomic relationships and Different ‘‘species definitions’’ and ‘‘species concepts’’ have been
includes DNA–DNA hybridization (DDH) studies and more proposed (for a comprehensive review, see Rosselló-Mora and
and more also comparative sequence analysis of homologous Amann 2001).
genes, which are used as phylogenetic markers (primarily the Cowan (1978) already addressed three meanings for the term
small ribosomal subunit, 16S rRNA gene). Different theory- species: a category (a mental representation), a taxonomic
based concepts have recently been brought into discussion to be group, and a concept. The category indicates that the species is
used for classification as the ecotype-based (Koeppel et al. 2008) a taxonomic rank below the genus rank in a hierarchical system.
or the metapopulation-based concept (Achtman and Wagner Cowan (1968) even pointed out that a species is ‘‘a group of
2008). Those concepts are important for the elucidation of organisms defined more or less subjectively by the criteria cho-
specific microbes in an ecological and also evolutionary context, sen by the taxonomist to show the best advantage as far as
but it is problematic to apply those systems as tools for routine possible and putting into practice his individual concept of
analysis for identification of the broad range of diversity of what a species is.’’ The usefulness of working with the concept
prokaryotes (Schleifer 2009; Kämpfer and Glaeser 2012). As of a species was not denied by Cowan, but he reminded the user
previously outlined by Schleifer (2009), ‘‘it should be kept in that the species does not exist and does not represent a natural
mind that the end-users need a pragmatic classification system entity. It should be mentioned here that all three meanings given
that can serve as a tool in routine identification. A classification above are intrinsically connected. A description of a taxonomic
that is of little use to microbiologists, no matter how sophisti- group (or unit) and also of a concept is only possible if one uses
cated a scheme is, will soon be ignored or significantly modified a word for the category ‘‘species,’’ which directly leads to the
(Staley and Krieg 1984).’’ name of this category and, hence, nomenclature.
The practicability of the polyphasic approach has been ‘‘Species’’ are regarded as the fundamental units in taxon-
documented in bacterial taxonomy (Vandamme et al. 1996; omy, and microbial strains are most often assigned to the taxo-
Schleifer 2009; Tindall et al. 2010), but there is a tendency nomic rank ‘‘species’’ or ‘‘subspecies’’ and at least to the genus
to replace the sometimes time-consuming procedures by level. With this nomenclatural assignment (and simultaneously
the assignment of the meaning behind this name), the link of infrasubspecific taxon, one strain or a set of strains shows similar
a mental representation (the taxonomic rank) to a ‘‘physical or identical properties and therefore is treated as a taxonomic
object’’ is most obvious. Type strains serve as reference points group. The term ‘‘infrasubspecific’’ is used to refer to the kind of
in the cases of the basic taxonomic unit species and subspecies. taxa below subspecies.
The nomenclature of the ‘‘lower’’ranks, that is, ‘‘family,’’ ‘‘genus,’’ Subdivisions usually contain the suffixes ‘‘-type,’’ ‘‘-var,’’ or
‘‘species,’’ and ‘‘subspecies,’’is regulated by rules of thenomenclatural ‘‘-form.’’ However, in Appendix 10, it is recommended, to use
system, ‘‘The Bacteriological Code of Nomenclature’’ (Lapage et al. the suffix ‘‘-var’’ instead of ‘‘-type’’ to avoid confusion with the
1992), that has been developed several decades ago and serves use of the term ‘‘type’’ in nomenclatural context (Rule 15). The
microbiologists as a solid and indispensable foundation (Lapage following suffixes for subdivisions are recommended, whereas
et al. 1992; Skerman et al. 1989; Sneath 2005). suffixes given in parenthesis should be avoided: biovar (biotype,
The 1990 revision of the International Code of Nomenclature physiological type), chemovar, chemoform (chemotype), culti-
of Bacteria supersedes all previous editions and has been applied var, forma specialis (special form), morphovar (morphotype),
from the date of publication in 1992 and consists of 7 general pathovar (pathotype), phagovar (phagotype, lysotype), phase,
considerations, 9 principles, 65 rules (some supplemented by serovar (serotype), and state (Trüper and Schleifer 2006).
recommendations which do not have the force of rules), advi- A ‘‘biovar’’ (instead of biotype or physiological type) is based
sory notes (subdivided into suggestions for authors and pub- on biochemical or physiological properties; a ‘‘chemovar’’ is based
lishers, advices regarding the quotations of authors, and names on the production of a specific chemical, and a ‘‘chemoform’’
and the maintenance of type strains), 10 appendices, and the (instead of chemotype) is based on the chemical constitution.
statutes of the International Committee on Systematic Bacteri- A ‘‘cultivar’’ refers to special cultivation properties, and a ‘‘forma
ology and the statutes of the Bacteriology and Applied Micro- specialis’’ (instead of special from) refers to a parasitic, symbiotic,
biology Division of the International Union of Microbiological or commensal bacterium which can be distinguished by adapta-
Societies (IUMS). tion to a particular host or habitat (here the scientific name of the
The rules of the code give advice for the naming of taxa, host in the genitive should be used for nomenclature).
nomenclatural types including their designation, the priority A ‘‘morphovar’’ (instead of morphotype) reflects specific mor-
and the publication of the names, and the citation of authors phological characteristics, a ‘‘pathovar’’ (instead of pathotype)
and names. There are rules about changes in the names of taxa as refers to specific hosts, a ‘‘phagovar’’ (instead of phagotype) is
a result of transference, union, or change in rank; about illegit- linked to a specific bacteriophage, and a ‘‘serovar’’ (instead of
imate and legitimate state of names and epithets; and the serotype), to antigenic characteristics.
replacement, rejection, and conservation of names and epithets. As pointed out by Tindall (1999), ‘‘the Bacteriological Code
A number of rules and Appendix 9 (Orthography) refer to the does not attempt to regulate either which methods are to be
correct usage of Latin orthography. Since the revision of the used, or which taxonomic interpretation of a given problem is
Bacteriological Code, in 1990, Appendix 9 is stronger, incorpo- correct (i.e., there is no regulated ‘‘official’’ taxonomy). The
rated in the code as a recommendation. In 2005, it has been Bacteriological Code deals solely with the way in which names
accepted at the Plenary Meetings of the Judicial Commission are assigned to organisms, and which of these are to be used.’’
and the ICSP during the IUMS Congress in San Francisco, CA, This system has been in operation since 1 January 1980 and
USA (July 23–29, 2005) that a new edition of the Appendix 9 was developed in order to clarify uncertainties of about 40,000
should be available online: http://ijs.sgmjournals.org/cgi/ names found in the scientific literature over the last hundreds of
content/full/59/8/2107 (Trüper and Euzéby 2009). years (Tindall et al. 2007). With the publication of the Approved
According to the general consideration 7, the word taxon List of Bacterial Names (Skerman et al. 1980), only around 2,000
(plural: taxa) is used in the code for any taxonomic group of names were incorporated into the system.
organisms, and as laid down in Principle 6, the correct name of From 1 January 1980 onward, the proposal of a name for a
a taxon is based upon valid publication, legitimacy (named prokaryotic organism representing a novel taxon must be
according to the rules), and priority of publication (for more accompanied by specific criteria laid down in the revised word-
details, see below). ing of Rule 27 of the Bacteriological Code (Lapage et al. 1992;
Each taxon consists of one or more elements. For each Sneath 2001; Tindall et al. 2006). These include the elements,
named taxon of the various taxonomic categories that the new name or new combination is clearly stated and
(> Table 6.1), a nomenclatural ‘‘type’’ shall be designated, that indicated as such (i.e., fam. nov., gen. nov., sp. nov., comb. nov.,
is, the element of the taxon to which the name is permanently etc.), in order to show clearly the intention of the author(s) to
associated. Appendix 4 contains all conserved and rejected create a new name, the presentation of the derivation (etymol-
names of bacterial taxa, and Appendix 5, the Opinions of ogy) of a new name (and if necessary of a new combination), the
the Judicial Commission of the International Committee description of the properties of the taxon, and the accessibility of
on Systematic Bacteriology, which have the same force as all these information. The major element of the presented sys-
the rules. Appendix 10 of the International Code of Nomencla- tem is the process of valid publication of a name of a taxonomic
ture of Bacteria deals with the infrasubspecific subdivision, rank (subspecies, species, genus, family), which constitutes
which is only based on selected ‘‘utilities’’ but not based on a form of official registration/indexing of that name through
DNA-DNA Hybridization (DDH)-based differentiation. In an a centralized system.
. Table 6.1
Overview of taxonomic ranks, respective suffixes, and appropriate types as listed in the bacteriological code. Overview of taxonomic
ranks, their suffixes, and appropriate types as listed in the Bacteriological Code revisions 1975 and 1990 (Lapage et al. 1992). The
depicted example was presented previously by Tindall et al. (2006). The ranks of subfamily, tribe, subtribe, and subgenus are not widely
used at present
Taxonomic Latin
category suffix Examplesa Type (example)
Order -ales Pseudomonadales Pseudomonas (genus)
Suborder -ineae Pseudomonadineae Pseudomonas (genus)
Family -aceae Pseudomonadaceae Pseudomonas (genus)
Subfamily -oideae Pseudomonadoideaeb Pseudomonas (genus)
Tribe -eae Pseudomonadeae Pseudomonas (genus)
Subtribe -ineae Pseudomonadinaeb Pseudomonas (genus)
Genus Pseudomonas Pseudomonas aeruginosa (species)
Species Pseudomonas aeruginosa RH 815 = ATCC 10145 = CCEB 481 = CCUG 551 = CCUG 28447 = CCUG 29297 = CFBP
2466 = CIP 100720 = DSM 50071 = IBCS 277 = IFO (now NBRC) 12689 = JCM
5962 = LMG 1242 = NCCB 76039 = NCIB (now NCIMB) 8295 = NCTC 10332 = NRRL
B-771 = VKM B-588
Pseudomonas aeruginosa RH 815 = ATCC 10145 = CCEB 481 = CCUG 551 = CCUG 28447 = CCUG 29297 = CFBP
Subspecies subsp. aeruginosac 2466 = CIP 100720 = DSM 50071 = IBCS 277 = IFO (now NBRC) 12689 = JCM
5962 = LMG 1242 = NCCB 76039 = NCIB (now NCIMB) 8295 = NCTC 10332 = NRRL
B-771 = VKM B-588
a
In the presented examples (according to Tindall et al. 2006), the stem above the genus level is Pseudomonad
b
However, subfamily and subtribe are included in the Bacteriological Code, revisions 1975 and 1990; the respective names are not validly published
c
The subspecies name Pseudomonas aeruginosa subsp. aeruginosa is not validly published, but if Pseudomonas aeruginosa would be divided into subspecies, the
name would be created automatically (Rule 46)
Furthermore, the nomenclatural type of the taxon must be in at least two recognized culture collections in two different
designated. In addition, the name must appear in the Interna- countries, and since August 2002, the IJSEM, as the official
tional Journal of Systematic and Evolutionary Microbiology organ of the International Committee on Systematics of Pro-
(IJSEM) and make reference to these requirements listed. This karyotes (ICSP), has asked that authors provide documented
may be via original publication in the IJSB/IJSEM or via publi- evidence from the collections confirming deposition and avail-
cation in another journal and inclusion of the relevant informa- ability of type strains. The accession numbers assigned to the
tion in the validation lists (which must include reference to the strain by the culture collections must be quoted in the published
type, taxonomic rank, and where the remaining information is description.
to be found). Rule 30(3a) states: ‘‘As of 1 January 2001 the description of
a new species, or new combinations previously represented by
viable cultures must include the designation of a type strain, and
Importance of Type Material a viable culture of that strain must be deposited in at least two
publicly accessible service collections in different countries from
One of the most important criteria in classification is the desig- which subcultures must be available. The designations allotted to
nation of the nomenclatural type. As pointed out by Tindall the strain by the culture collections should be quoted in the
(2008) and Tindall and Garrity (2008), the type serves as published description. Evidence must be presented that the cul-
a reference point for a taxon, and in the case of the proposal of tures are present, viable, and available at the time of publication.’’
a novel species or subspecies, these types are represented by There have been several discussions about this requirement
‘‘physical objects,’’ the type strains (Lapage et al. 1992). Those followed by the proposal of standards for strain deposits
‘‘representative’’ strains must be made available without restric- (Tindall 2008; Tindall and Garrity 2008; Kämpfer 2010), and
tions to the scientific community. there is clearly a necessity to ensure that higher standards are
Since January 2001 (De Vos and Trüper 2000; Labeda 2000), being applied to taxonomic work, enabling the key elements of
it is a requirement laid down in the Bacteriological Code that verifying existing experimental data and expanding on the data
authors proposing novel species, novel subspecies, and new set associated with authentic biological material (type strains) to
combinations have to provide evidence that types are deposited be carried out.
The Provisional Status Candidatus order Halobacteriales (Oren et al. 1997), family Halomonadaceae
(Arahal et al. 2007), genus Helicobacter (Dewhirst et al. 2000),
The provisional status Candidatus was introduced in 1994 by suborder Micrococcineae (Schumann et al. 2009), class Mollicutes
Murray and Schleifer (1994) followed by Murray and (division Tenericutes, order Mycoplasmatales) (International
Stackebrandt (1995) for certain putative taxa that could not be Committee on Systematic Bacteriology Subcommittee on the
described in sufficient detail as a novel taxon. The designation Taxonomy of Mycoplasmatales 1972; Brown et al. 2007), genus
Candidatus is not a rank but a status that is currently not Mycobacterium (Lévy-Frébault and Portaels 1992), family
formally recognized in the International Code of Nomenclature Pasteurellaceae (Christensen et al. 2007), and Staphylococci
of Bacteria. According to the ‘‘Ad Hoc Committee for the re- (Freney et al. 1999).
evaluation of the species definition in bacteriology’’
(Stackebrandt et al. 2002), microbiologists are encouraged to
use the status Candidatus for well-characterized but as-yet- From the Beginning of Prokaryotic
uncultured prokaryotes. Classification and Towards a Modern
The phylogenetic relatedness of the as-yet-uncultured pro- Taxonomy
karyote has to be determined and their authenticity should have
been revealed by, for example, in situ probing using fluorescence The consideration of the classification of prokaryotes from the
in situ hybridization or other similar techniques. Beside geno- beginning in the late nineteenth century until today illustrated
typic information, phenotypic properties should also be the tremendous changes in the classification of prokaryotes,
included in the description of a Candidatus species including which was primarily affected by the development of new tech-
structural, metabolic, physiological, and reproductive features. niques. This topic will only be briefly covered in this chapter. It is
The names of the category Candidatus should be written as addressed in further detail in another chapter in this volume.
follows: Candidatus in italics and the subsequent name(s) in At the end of the nineteenth century, Ferdinand Cohn used
roman type with an initial capital letter for the genus name. morphological features as different cell shapes (cocci, small or
The entire name should be in quotation marks (e.g., elongated rods, or spirals) to classify bacteria into six genera at
‘‘Candidatus Magnospira bakii’’). It was decided by the Judicial that time as members of the plants. Based on the morphological
Commission of the International Committee on Systematics of similarities, Cohn (1875) grouped the ‘‘common bacteria’’ and
Prokaryotes that the concept Candidatus should be mentioned the cyanobacteria (later on referred as blue-green algae) together
in the main body of the Bacteriological Code; however, the to the Schizophyta. Pathogenic bacteria were in the focus of
names have still no standing in nomenclature (De Vos et al. interest at that time which is reflected by the fact that many
2005). Many Bacteria and Archaea, which have currently well-known pathogenic bacteria were described at the end of
a Candidatus status (more than 200 in 2009) are endosymbionts the nineteenth century. The pathogenic potential and general
or parasites of eukaryotes, which belong to the phyla and orders growth requirements were the most important taxonomic
Mollicutes, Chlamydiales, and Rickettsiales, respectively. Other markers, which were used beside morphology to classify bacteria.
as-yet-cultured prokaryotes with a Candidatus status are only In the first half of the twentieth century, beside morphology,
available as special enrichments or cocultures (e.g., syntrophic numerous biochemical and physiological features were used to
organisms, freshwater Actinobacteria) or occurring in unusual classify and identify bacteria. Later on, enzymes and metabolic
habitats, revealing unusual metabolic properties (Anammox, pathways were additionally elucidated. In the first edition of
Fe-oxidizing bacteria). Bergey’s Manual of Determinative Bacteriology (Bergey et al.
1923), phenotypic properties were used for the classification of
Minimal Standards bacteria as ‘‘typically unicellular plants,’’ the so-called
Schizomycetes. Even in the 7th edition of Bergey’s Manual
Minimal standards are documents which are compiled by (Breed et al. 1957), bacteria were still classified as members of
experts within the framework of subcommittees set up within plants (Protophyta, primitive plants). As first propagated by
the International Committee on Systematic Bacteriology/Inter- A. Lwoff, R. Stanier and C. B. van Niel described in 1962 the
national Committee on Systematics of Prokaryotes (ICSB/ICSP; division into prokaryotic (bacteria) and eukaryotic (animal,
Lapage et al. 1992) to provide detailed information how specific plants) organisms (Stanier and Van Niel 1962), and then in the
groups of organisms should be characterized. Their role is cov- eighth edition of the Bergey’s Manual (Buchanan and Gibbons
ered by Recommendation 30 (formerly Recommendation 30b) 1974), bacteria were no longer recognized as plants but assigned
of Rule 30 of the Bacteriological Code (Lapage et al. 1992), as as members of the kingdom Prokaryotae. At that time, pheno-
modified at the 1999 meetings of the ICSB and its Judicial typically, properties including gram-staining, morphology, and
Commission (De Vos and Trüper 2000; Labeda 2000). Really oxygen requirements were used to classify bacteria. However, the
updated minimal standards are only available for a very phenotype-based classification alone led to a grouping of phe-
restricted group of organisms covered by the subcommittees. notypically coherent but as nowadays well-known genetically
Some updated minimal standards, for example, have been different bacteria.
published for aerobic, endospore-forming bacteria (Logan In the 1970s and 1980s, comprehensive sets of phenotypic
et al. 2009), family Flavobacteriaceae (Bernardet et al. 2002), data were used to classify bacteria by using numerical studies
(numerical taxonomy; Goodfellow 1977; Sneath 1971; Sneath the integrated use of phenotypic and genotypic characteristics,
and Sokal 1973, and several others). Numerical taxonomy that is, the polyphasic taxonomy (Colwell 1970) is necessary for
improved the accuracy of phenotypic identification by increas- the delineation of taxa at all levels from kingdom to genus
ing the number of tests used and the calculation of coefficients of (Murray et al. 1990). In the majority of cases, there is a good
similarities between strains and species (Sneath and Sokal 1973). evidence in bacterial systematics of a congruence between the
The results of the numerical studies were tabulated in tables of distribution of specific chemical markers and the relative posi-
t organisms versus n characters and specific operation taxo- tion of species in phylogenetic trees. Chemotaxonomic markers
nomic units (OTUs). The characters, which were compared, can therefore help to assign new taxa to related taxa but can also
should come from various different categories of properties as help to delimit groups of related species (Murray et al. 1990;
morphology, physiology, or biochemistry and were equally Tindall et al. 2010).
weighted in the analyses. The number of common characteris- With the increasing importance of the 16S rRNA gene for
tics used was considered as a quantitative measurement of tax- delineation of bacterial taxa and the increase of available
onomic relatedness, although this did not mean that the sequences for almost all taxa, Stackebrandt and Goebel
compared organisms were also phylogenetically related. (1994) tried to correlate numerous published DDH data with
In the last 50 years, tremendous achievements have been 16S rRNA gene sequence similarity data and came to the conclu-
made in the development of DNA-based analyses for the char- sion that the correlation plot of the two parameters was not
acterization of microorganisms. The classification based on linear. They pointed out that each method is strong in those
genotypic methods enabled more and more the investigation areas of relationships in which the other method sometimes
of the phylogenetic relationships among prokaryotes. DNA- fails to reliably show relationships. Sequence analysis of 16S
based analysis started with the estimation of the guanine and rRNA genes is regarded as more suitable from the level of
cytosine nucleosides ratio within the total genomic DNA domains (starting at about 55% similarity) to genera or in
followed by the introduction of DNA–DNA hybridization some cases moderately related species, that is, below 97.5%
(DDH) methods of genomic DNA (Johnson and Ordal 1968), rRNA sequence similarity. Above this value, DNA-DNA
which is still recognized as the ‘‘genotypic standard’’ for the reassociation values were found to be very low (<20%) or as
delineation of prokaryotic species (Wayne et al. 1987; high as 100%. They concluded in addition that DDH remains
Stackebrandt et al. 2002; Tindall et al. 2010). An important the method for measuring the degree of relatedness between
step forward was the development of the PCR-based sequencing highly related organisms and provided evidence that at 16S
technology (Saiki et al. 1988) and the introduction of the 16S rRNA gene sequence similarity values below about 97.5%, it
rRNA gene as a molecular marker in taxonomy. Several other was unlikely that two organisms shared more than 60–70%
genes (e.g., rpoB, gyrA, recA, hsp60, atpD, just to name a few) DNA similarity. In general, strong evidence was provided that
were introduced as alternative or supporting molecular markers sequence analyses of 16S rRNA genes are not the appropriate
in prokaryotic taxonomy, and the rapid development of high- method to replace DNA–DNA reassociation for the delineation
throughput sequencing technologies will lead to more and more of species and measurement of infraspecies relationships.
genome-based genotypic analysis. Hence, a major critical point is the resolving power of the
The work of C. R. Woese was a ‘‘milestone’’ for the start of 16S rRNA gene sequence below the genus level because of its
genotype-based classification of prokaryotes. Carl Woese and conserved structure. This means that branching pattern at the
coworkers were the first who demonstrated the usefulness of periphery of phylogenetic trees (sometimes at the genus level,
small subunit (SSU) rRNA as a universal phylogenetic marker to but most often at the species level) cannot reliably reflect phy-
reflect phylogenetic relationships (Fox et al. 1977). logeny in the sense of common ancestry (independent from the
Data obtained from comparative sequence analyses of the ‘‘treeing algorithm’’). Organisms sharing very similar or even
small ribosomal subunit (the 16S rRNA gene sequence) identical 16S rRNA gene sequences may be more diverse at the
provided for the first time a basis to study phylogenetic whole genome level than those having more variable positions
relationships among all bacteria. With the investigation of (Stackebrandt and Goebel 1994).
partial sequences of 16S ribosomal RNA (rRNA) genes, the The development of new sequencing technologies makes it
Archaea (which were originally assigned as Archaebacteria) now feasible to look in detail at the genome sequence because
were found to constitute a unique group leading to the descrip- gene and genome sequences can be generated in a relatively short
tion of a separate kingdom (Woese and Fox 1977). Other period of time, and it can be foreseen that the wealth of the new
phylogenetic markers including other ribosomal RNA genes, data can and will be used for a critical evaluation of the taxo-
the 5S and 23S rRNA, or other macromolecules besides nomic system. Several new genomic approaches, based on total or
rRNAs, for example, the elongation factor Tu or the b-subunit partial genome comparisons, including concatenated sequences
of ATP-synthase (Schleifer and Ludwig 1989), confirmed the of all conserved or selected genes (summarized by Cole et al.
relationships obtained by the 16S rRNA gene sequence–based 2010), confirmed the 16S-rRNA-gene-based hierarchical system
approach. at least at the genus level and above. As a consequence, the 16S-
Differences in evolutionary rates in various groups of organ- rRNA-gene-based hierarchical structure still remains for the
isms and other considerations can be obstacles preventing the time the backbone of prokaryotic systematics (Cole et al. 2010;
single use of gene sequences in delineating taxa. Therefore, still Ludwig 2010).
Genome sequence–based studies start to overcome the prob- step to characterize a new bacterial isolate is the PCR amplifica-
lems associated with DDH experiments. The average nucleotide tion and sequencing of the 16S rRNA gene to determine its
identity (ANI) index of shared genes between two genomes phylogenetic affiliation (> Fig. 6.1). Expert-maintained web-
(Konstantinidis and Tiedje 2005) showed among several tested based databases such as the search tool EzTaxon that contains
genomes-derived parameters the best correlation with DDH the 16S rRNA gene sequences of all type strains enables a fast first
values. The first evaluation of ANI values was based on assignment of a new strain to related taxa (Chun et al. 2007;
a pairwise genome comparison including all shared orthologous updated version: Kim et al. 2012).
protein-coding genes (Konstantinidis and Tiedje 2005). Based In some areas, phenotypic identification systems including
on those studies, an ANI value of 94% was recommended that fatty acid profiles or MALDI-TOF analyses and the investigation
could reflect the bacterial species boundary of 70% DDH simi- of physiological properties by commercial test systems are also
larity. Similar results were observed by the comparison of often used for the identification of an unknown, especially in
genomes that were artificially sectioned into 1020 nucleotide routine identification procedures. Here, also complete and updated
fragments (Goris et al. 2007). Richter and Rosselló-Móra respective databases are the prerequisites for this identification.
(2009) recommended the use of an even narrower boundary of For the allocation of an unknown to a genus, detailed phy-
95–96% ANI to circumscribe prokaryotic species and showed logenetic calculations based on nearly full-length 16S rRNA gene
that the comparison of at least 20% of genomes generated by sequences are now in most cases ‘‘state of the art.’’ This is
simple randomly sequencing will be enough to determine reli- followed by calculations of pairwise similarities and the con-
able ANI values. Rosselló-Móra (2012) had a closer look to struction of phylogenetic trees to determine the exact phyloge-
methods and their potential for a genome-based taxonomy. netic relationship of the new strain to closest related taxa. The
In a recently published review, Klenk and Göker (2010) gave results of 16S rRNA gene sequence similarity calculations often
an outlook to a genome-based classification of Archaea and Bacte- indicates whether or not the new strain may represent a new
ria. They concluded that microbial systematics and genomics species or even a new genus and thereby provides the direction
have not been completely reconciled due to the intrinsic diffi- for the following detailed taxonomic investigations. At 16S
culties of inferring reasonable phylogenies from genome rRNA gene sequence similarity levels below 97%, a new species
sequences, in part due to the significant role of lateral gene or even genus assignment may be quite likely. Above 97% gene
transfer (LGT) in microbial evolution. In addition, they pointed sequence similarities, further genotypic methods have to be
out that the consequences of using different methods for applied to elucidate if the strain/strains represent a new species
orthology detection, sequence alignment, and alignment filter- or has to be assigned to an existing one. In the current species
ing, and use of different reference sequences, may all have concept, two organisms sharing a DNA–DNA hybridization
a significant effect of phylogenetic reconstructions and are not value of higher than 70% represent the same species (Wayne
yet fully understood (Klenk and Göker 2010; Ludwig 2010). et al. 1987; Stackebrandt et al. 2002). If 16S rRNA gene sequence
However, it can be foreseen, that the comparison of whole similarities are below 95%, it is quite likely that a new genus has to
genomes will play a more and more important role in the be described (for more details, see below). If a new taxon will be
classification of prokaryotes. described, the strains or at least the provisional type strain must
be studied in great detail in order to provide evidence that the
novel taxon can be clearly differentiated from closest related taxa
Current Procedures for the Characterization also on the basis of phenotypic (including chemotaxonomic)
and Identification of Bacterial and Archaeal characteristics. Distinguishing phenotypic differences is required
Strains for the description of a new species. If such differences are not
found, groups of similar bacteria that appear to be genetically
The currently operational ‘‘polyphasic approach’’ includes phe- distinct should be described by other terms (e.g., genospecies/
notypic (including chemotaxonomic) and genotypic traits to genomovars). An increasing number of taxonomic studies include
characterize a prokaryotic taxa. DNA sequence–based methods the investigation of alternative genetic markers in addition to the
are nowadays more and more the preferred method for the phylogenetic analysis based on the 16S rRNA gene sequence to
initial identification and characterization of a prokaryotic strain obtain a higher resolution at lower taxonomic ranks. Multilocus
(see > Fig. 6.1). sequence analysis (MLSA) is most often used for this purpose
For classification, it is recommended to include more than including the investigation of a set of partial sequences of house-
just the type strain for the description of a new species (e.g., keeping genes. It has been suggested recently that MLSA may
Christensen et al. 2001; Felis and Dellaglio 2007) because those replace DDH analysis (see below). Also, other genotypic as geno-
characteristics can be variable within species, and this can only mic fingerprint methods are often applied to differentiate taxa at
be determined by analyzing more than one strain. However this the lower taxonomic ranks, for example, strains within species.
is only a recommendation and cannot be mandatory. Current recommendations for the characterization of new
All properties, which are common, perhaps even unique for prokaryotic taxa have been summarized recently by Tindall et al.
a strain, should be included in the description. Variable proper- (2010). These recommendations may serve as general guidelines.
ties should be included as well and can be an indicator for If minimal standards are available for the respective taxa a new
subgroupings within a taxonomic group. Most often, the initial strain belongs to, those should also be considered in addition.
If a new species is assigned to an existing genus, differences Recently, some suggestions have been made for lower
of the new species and existing species within the genus have to boundaries (Stackebrandt and Ebers 2006; Yarza et al. 2008;
be provided-if the genus is too heterogeneous (e.g., the genera Tindall et al. 2010); however, it has to be mentioned again, the
Bacillus and Clostridium) or contains a large number of species delineation of species cannot be only based on the 16S rRNA
(e.g., Streptomyces), then the authors must give scientific-base gene sequence similarities, and results obtained by several other
arguments for the selection of only a set of species of the genus methods have to be considered as well.
for comparative analysis. The type species of the respective genus The importance of the phylogenetic analysis based on the
must be included. In the case where a new strain seems to 16S rRNA gene sequence as molecular marker for the identifi-
represent a new genus, this genus has to be clearly differentiated cation and classification of prokaryotes makes it exceptionally
from all ‘‘closely related’’ genera. important to perform this analysis as standardized as possible.
Currently, no elaborative predictions are available, but several
recommendations have been given (Tindall et al. 2010; Peplies
Genotypic Traits et al. 2008).
Several critical steps have to be considered for the analysis of
Genotypic and phenotypic methods that are typically applied in the 16S rRNA gene sequences, including the sequence qualities,
polyphasic taxonomic approaches for characterizing new taxa sequence alignments, and the calculation of pairwise sequences
are summarized in > Fig. 6.1. similarities and of phylogenetic trees. Only high-quality nearly
Genotypic traits of an organism are those within its genetic full-length sequences (at least 1,300 nucleotides) should be used
material, the genome. The term genomic is convenient to refer for analysis. Not only the new sequence but also the reference
to large amounts of information in the genome or derived from sequences have to fulfill those requirements. Therefore, sequence
it very directly with few extraneous influences. Messenger RNA qualities should be checked with caution before any analysis and
and ribosomal RNA can be regarded as formally intermediates especially before sequences are deposited in databases, used for
between genotype and phenotype, but they reflect the genome so a publication, or sent to culture collections for the deposited of
faithfully that they can in practice be treated as genotypic the respective strains. Sequences should be checked for ambigu-
(Sneath 1989). ities, the consensus of primary and secondary structures, and for
potential gene heterogeneities, which can occur by the direct
sequencing of PCR fragments. It is recommended to use a set of
Phylogenetic Assignment Based on 16S rRNA well-aligned reference sequences for a final check of the sequence
Gene Sequences quality. The use of sequences available at the primary databases
(GenBank/EMBL/DDBJ) has to be performed with caution
A 16S rRNA gene sequence–based approach alone cannot because the databases contain several incorrect labeled and
describe a new species but can provide the first evidence if low-quality sequences, which have not undergone an official
a new isolate represents a new species or even a new genus. quality control before deposit.
A 16S rRNA gene sequence similarity of 97% can be used as The alignment itself is a critical step for all phylogenetic
an indication for a new species. There are, however, many analysis because it is the basic for pairwise sequence similarity
examples in the literature that two strains with less than 97% calculations and the construction of phylogenetic trees. It is
16S rRNA gene sequence similarity are not members of the same highly recommended to use expert-maintained seed alignments
species (Amann et al. 1992; Collins et al. 1991; Fox et al. 1992; that contain quality-checked sequences, as provided by ARB
Martinez-Murcia and Collins 1990; Martinez-Murcia et al. [www.arb-home.de], RDP [http://rdp.cme.msu.edu/], SILVA
1992). The prerequisite for such a conclusion is that the [www.arb-silva.de], or LTP [www.arb-silva.de/projects/living-
sequence comparison based on nearly full-length high-quality tree/]. Reference sequences (of type strains) can also be
sequences and a well-performed sequence alignment (see downloaded from the databases and aligned in multiple align-
below). If 16S rRNA gene sequence similarities are above 97% ment programs as ClustalW followed by a manual editing.
(over full pairwise comparisons), DNA–DNA hybridization or However, those programs did not enable the inclusion of sec-
gene sequences showing a higher resolution must be considered. ondary structure information, as it is, for example, possible in
16S rRNA gene similarity values (over full pairwise compar- ARB. Dependent on the investigated taxa, secondary structure
isons) of 95% gives often evidence that a new strain may information can be very helpful for correct alignments especially
represent a new genus. of highly variable regions of the 16S rRNA gene. The procedure
While high attention was focused on the resolution power of the sequence alignment should be stated clearly in
of the 16S rRNA gene at the species level, the delineation at a publication, and new sequences and the alignments of all
the genus level based on16S rRNA gene sequence similarities is sequences included in the analysis for a publication must be
less clear. Within some genera, high 16S rRNA gene sequence made available for editors and reviewers when a manuscript is
similarities occur, for example, within the Enterobacteriaceae submitted.
or Actinomycetales. In contrast, other genera contain species Pairwise nucleotide sequence similarity values of high-
with lower sequence similarities, for example, the genus quality sequences should then also be performed with caution.
Flavobacterium. It is recommended to perform the pairwise sequence similarity
calculations in ARB, PHYDIT, and jPHYDIT; local alignment The resolution power of the 16S rRNA gene is limited most
programs as BLAST should not be used. Pairwise sequence often to distinction at the genus level. At the species level, other
similarities should not based on corrected evolutionary dis- methods have to be considered if better resolution/distinction is
tances as, for example, calculated with the Jukes and Cantor, or required, and DDH and phylogenetic analysis of other molecu-
several other evolutionary models. Here again, the procedure of lar markers as single-gene analysis or in a set of genes have to be
the similarity calculations should be clearly stated. additional performed.
Phylogenetic trees are the graphical representation of For the characterization of specific taxa and the differentia-
phylogenetic relationships of the new taxa to related taxa. All tion of those taxa from closest related taxa, sometime signature
sequence used for the tree construction have to be full-length nucleotides within the 16S rRNA gene sequence are determined
sequences. and listed in a species description. The presence and absence of
The preferable methods for tree calculations are the maxi- those signature nucleotides should be compared by the proposal
mum parsimony and maximum likelihood methods because of new species.
both methods are based on evolutionary models. It is
recommended to use distance-matrix-based treeing methods as
the neighbor-joining methods only for first assignment of new DNA–DNA Hybridization
taxa (Ludwig and Klenk 2001; Peplies et al. 2008; Tindall et al.
2010). In most studies, however, neighbor-joining methods are DNA–DNA and DNA–RNA hybridizations were introduced
depicted in manuscripts because only a limited number of into prokaryotes systematics in the 1960s (Brenner et al. 1967;
sequences could be investigated with the maximum likelihood De Ley et al. 1966; Johnson and Ordal 1968; McCarthy and
method in a short time and bootstrap analyses are not possible Bolton 1963; Pace and Campbell 1971; Palleroni et al. 1973).
by most maximum likelihood applications. However, there are Later, DNA–RNA hybridizations were replaced by 16S rRNA
tools for maximum likelihood analyses (PHYML, RAXML) gene sequencing because a good correlation was obtained for
available, which enable the inclusion of appropriate data sets results of DNA–RNA experiments and 16S rRNA gene sequenc-
and the application of bootstrap analysis. In general, different ing. DNA–DNA hybridization (or reassociation) (DDH) is still
treeing methods should be applied in parallel to improve the the recommended standard to confirm if strains that share more
calculated phylogenetic relationships. than 97% 16S rRNA gene sequence identity represent different
It is recommended to use sequence conservation profiles or the same species. If the new taxon shows 97% 16S rRNA gene
determined for the group of interest and higher ranks as filters sequence similarity to more than one related species, DDH
to recognize branch attraction effects, which may result from should be performed with all regarding type strains to ensure
pleomorphic sites, and to test the stability of a tree topology that there is enough dissimilarity to support classification of the
(Ludwig and Klenk 2001; Peplies et al. 2008). The application of strain(s) as a new taxon. DDH must also be performed in cases
a filter results in the successive removing of alignment columns where the new taxon contains more than a single strain in order
according to the variability of the positions. The most often to show that all members of the taxon have a high degree of
applied filter is a 50% base frequency filter, which will exclude hybridization among each other and thereby indeed represent
a complete alignment column if the frequency of the most the same species.
abundant nucleotide is below 50% (Peplies et al. 2008). To A number of techniques can be used for DDH experiments.
proof the tree topology, the use of 30% and 40% filters are also Most of them have been validated and show comparable results
recommended. If filters are used, tree topologies of trees gener- (Grimont et al. 1980; Rosselló-Móra 2006). It is recommended
ated with and without a filter have to be compared, and an that the method (with all modifications) must be clearly cited
alignment with the assignment of the filtered positions must and by the use of a novel method, the authors must provide
be made available. The use of filters and the interpretations evidence that the method produces comparable results to the
relying on the use of a filter must be clearly stated. established methods. As recommended by Tindall et al. (2010),
The choice of a reference out-group sequence can also have the DDH data of the type strain of a proposed new species with
an important effect on the tree topology. Sequences used as out- the type strain(s) of the closest related species with validly
groups should not be too distantly related to the investigated published names (if >97% 16S rRNA gene sequence similarity
groups. Single distantly related organisms as out-group may lead occurs) and with all other strains of the proposed new species
to branch extraction. must be provided. Depending on the used method, at least one
An accurate performed analysis of 16S rRNA gene based reciprocal value of the closest related species to the new type
data is the prerequisite for the interpretation of the phylo- strain must be provided. If reciprocal values cannot be supplied,
genetic relationships. However, the use of different treeing for example, by the use of spectrophotometric methods without
methods for the evaluation of the same data set does not DNA labeling, standard deviations of at least three analyses
identify effects such as gene transfer, convergent, or parallel should be given.
evolution. Therefore, it is required to include other studies The most often used DDH method in prokaryote systemat-
that can also reflect evolution including phylogenetic analy- ics is the relative binding ratio (RBR) methods, where results are
sis based on alternative genes but also nonsequence-based given in percentage binding or percentage hybridization. For the
methods. application of the RBR method in prokaryotic systematics, it was
recommended that only DNA from strains of at least 5 C like the 23S rRNA gene, genes encoding translation elongation
differences in melting temperature (deltaTm) should be com- and initiation factors, RNA polymerase subunits, ATPase sub-
pared. The temperature, which is used for DDH must be clearly units, DNA gyrases, the RecA, and heat shock proteins give very
stated. Commonly, hybridizations are performed at Tm-30 and/ similar tree topologies when compared with the topology of the
or Tm-15 (stringent conditions). The GC contents of the strains 16S rRNA genes (Ludwig 2010). Rather conserved and ubiqui-
under study are normally used to calculate the melting temper- tously distributed genes of the core genome are potential phy-
ature for each strain. logenetic markers for the genotypic classification of less related
DDH results have to be critically considered; however, prokaryotes. Thus, comparative sequence analysis of certain core
a DDH value of 70% has been recommended as species bound- genes, including rRNA genes, may be useful to obtain the phy-
ary (Brenner 1973; Johnson 1973, 1984; Wayne et al. 1987). This logenetic relationships of higher taxa. For the classification of
value should not be used strictly without considering other lower taxa, as the delineation of species, character genes may be
results (Ursing et al. 1995). All strains that cannot be clearly suitable phylogenetic markers. Genes coding for key phenotypic
discriminated by a stable phenotypic property should be differences may also play an important role to proof classifica-
grouped into a single species. Hence, a single species may con- tion of bacteria and to enable a higher resolution of taxa espe-
tain groups of strains with even less than 50% binding, which cially at the species level.
cannot be distinguishable by means of other properties tested at
the time. In contrast, a single species may contain several geno-
mic groups or genomovars (Ursing et al. 1995) that may be MultiLocus Sequence Analysis (MLSA)
reclassified as new species if they can be clear and stable discrim-
inated by phenotypic properties. Multilocus sequence analysis (MLSA) was outlined as an ‘‘inter-
DNA–DNA hybridization is at the time of this writing still an mediate resolution’’ between the 16S rRNA gene and genome-
important standard for the description of novel species (Tindall based approaches. MLSA is a method for the genotypic charac-
et al. 2010), but it is often criticized to be time-consuming, terization of a selected group of prokaryotes, most often strains
difficult to perform, and prone to high experimental errors of one species; species of a genus, and also of closely related
(Rosselló-Móra 2006). With the increasing applicability of geno- genera are investigated. Typically partial sequences of a set of
mic sequencing technologies, it can be foreseen that this method 5–12 housekeeping genes are investigated and either compared
will be replaced by methods relying on genomic information as individual data sets or combined to concatenated sequences,
derived on sequence information. Currently applied techniques, which were then used for phylogenetic analysis.
which will have the potential to replace DDH, are the calculation MLSA based on multilocus sequence typing (MLST), which
of the average nucleotide identity (ANI) of two genomes, which was introduced by Maiden et al. (1998) for epidemiological
are compared (see below). Recently, multilocus sequence analysis studies of pathogenic bacteria where it is currently one of the
(MLSA) of approximately 5–12 housekeeping genes has been most successfully applied methods. In contrast to MLST, which
suggested as an alternative for the application of DDH (Gevers is focused on the designation of sequence types based on base
et al. 2005). variation within sequences independent of the sequence simi-
larities, MLSA is focused on the investigation of the phylogenetic
relationship among closely related taxa. Sequence similarities are
DNA Base Composition calculated and phylogenetic relationships are investigated by the
construction of phylogenetic trees. Ideally, MLSA analysis is
The guanine–cytosine (GC) content of DNA (DNA base com- based on both DNA and amino acid sequences. Criteria for the
position) is an important character which is still required for the selection of genes are that they have to be present in all investi-
description of a new genus. Two strains differing by more than gated strains and they need to have conserved regions to gener-
10 mol% should not be considered as members of the same ate primers for gene amplification but however have to provide
genus; however otherwise, a higher similarity of the DNA base enough sequence differences to distinguish between the investi-
composition does not necessarily imply that the two strains are gated taxa. The selection of genes also depends on the taxa,
closely related. It has been shown that the DNA base composi- which are studied. Different genes may be suitable if different
tion is a valuable character to distinguish between nonrelated species are investigated or if an intraspecies analysis is performed
bacteria, such as staphylococci (30–35 mol% GC) and micro- (Cole et al. 2010).
cocci (70–75 mol% GC). The advantage of MLSA to the 16S-rRNA-gene-sequence-
based phylogeny is the use of protein coding housekeeping
genes, which also have a conserved function but a higher degree
Phylogenetic Assignment Based on Alternative of resolution than the 16S rRNA gene. The use of multiple genes
Phylogenetic Markers Including Protein Coding provides more informative nucleotide sites and buffers against
Genes the distorting effects of recombination of one of the loci.
MLSA enables the better phylogenetic resolution of deeply
There are not many alternative genetic markers available that branching clusters and helps to delineate genotypic clustering
occur in all genomes. The comparative analyses of some of them, within a genus or species (Soria-Carrasco et al. 2007). It was
recommended by Tindall et al. (2010) that if species delineation approach are gene sequence data obtained by whole, draft, or
is based on MLSA schemes, then it has to be validated by partial genome sequences. In the future, the rMLST or rMLSA
DNA–DNA hybridization data. approach is promising to be ‘‘a natural extension to the 16S
The application of MLSA for taxonomic classification may rRNA gene based analysis in microbial taxonomy’’ (Jolley et al.
be problematic for several reasons. A selection of a proper set of 2012).
genes is not systematically evaluated. As it has been shown so far,
different sets of genes are required for the classification of
different taxa and lineages. An overall applicable set of genes Genome-Based Analysis: GEBA Project: Toward
does not exist. In contrast to the 16S rRNA genes, the design of a New Area in Taxonomy
primers for the amplification of housekeeping genes may be
difficult or sometimes even impossible also for very closely The expected decrease in the costs for sequencing whole
related species or even strains within a species. A set of 5–12 genomes, together with the technical advances that have been
genes has been recommended for MLSA studies (Stackebrandt made recently, will make it more feasible that in the near future
et al. 2002; Tindall et al. 2010), but often, smaller sets including routine sequencing of more prokaryote genomes will be
3–5 genes are used, and even 12 genes correspond only to realistic. The detailed investigation of genome sequences may
a minor fraction of the genome. Most often only small gene then overcome the problems associated with DDH experiments.
fragments are investigated, which leads to a restricted phyloge- However, these analyses and the interpretation of the data
netic content. As already pointed out for 16S rRNA gene should be done very carefully. The key issue for taxonomic
sequence analysis, the alignment is a critical step also in protein analysis is a reliable annotation of genes in a genome since
coding gene sequence analysis. Here, the open reading frames identifying gene homologies (preferably orthologues) is of cen-
should be considered, and nucleotide sequences should be tral importance.
aligned according to the translated amino acid sequences. Espe- In the last years, several genome indexes were established for
cially if single amino acid deletions occur, amino-acid-based pairwise genome comparison for taxonomy purposes. The aver-
prealignments are often a prerequisite for a correct alignment age nucleotide identity (ANI, Konstantinidis et al. 2006) and
and for reliable phylogenetic analysis. Often, only nucleotide maximal unique matches (MUM, Deloger et al. 2009) are the
sequences are considered, but the third codon positions however most common indexes, and it has been suggested that they will
are more variable, and amino-acid-based sequence may be be able to substitute DDH. This may especially be true for ANI
performed in parallel to prove the stability of the obtained already in the near future. It was demonstrated that results of
phylogenetic relationships. MLSA analysis within a genus some- ANI and DDH correlate well and a range of about 95–96% ANI
times investigated with different MLSA schemes and most often reflects the current boundary of 70% DDH similarity (Goris
not all species of a genus are included in the analysis. Often, only et al. 2007; Richter and Rosselló-Móra 2009). Goris et al.
a selection of type strains of species of a genus or all genera under (2007) also performed the comparison of genomes that were
consideration are not available and thereby not considered by artificially sectioned into 1020 nucleotide fragments and showed
the delineation of new species. For example, the clinically rele- that the comparison of at least 20% of genomes generated by
vant Acinetobacter calcoaceticus—A. baumannii complex (ABC- simple randomly sequencing will be enough to determine reli-
complex) has been intensively studied by even two different able ANI values.
MLST schemes (Bartual et al. 2005; Wisplinghoff et al. 2008; So far, the number of genome sequenced type strains is still
Diancourt et al. 2010). Based on the MLST scheme published by too small to apply genome sequence–based taxonomy. The
Diancourt et al. (2010), a detailed phylogenetic analysis (MLSA) Genomic Encyclopedia of Bacteria and Archaea (GEBA) project
was performed leading to the proposal of two new species, was the first project which was initiated for the genome sequenc-
A. pittii and A. nosocomiales, which represented the genomic ing of type strains of almost all bacteria and archaea (http://
species 3 and 13TU (within the ABC-complex). Despite those www.jgi.doe.gov/programs/GEBA/index.html). Most of
detailed analysis, MLSA has so far not been applied and evalu- genome sequencing projects so far were focused on the inves-
ated for the whole genus Acinetobacter (Kämpfer and Glaeser tigation of pathogens or strains of relevance in other context.
2012). The primary criterion for the selection of type strains in the
Recently, Jolley and Maiden introduced a combined taxo- initial phase of the GEBA project is their taxonomic position in
nomic and typing approach, the ribosomal multilocus sequence the tree of life. Genome information obtained in the GEBA
typing (rMLST), which is based on the analysis of 53 bacterial project will be made available by publishing information in
ribosomal protein subunit (rps) gene loci. In contrast to other a short genome report in the open access journal Standards in
MLST/MLSA systems, the rMLST approach is promising to be Genomics Sciences (http://standardsingenomics.org/index.
suitable for the whole domain bacteria with a resolution down to php/sigen/index). In the currently ongoing pilot GEBA project,
the species or subspecies level (Jolley and Maiden 2010; Jolley the JGI performs in collaboration with the DSMZ the sequenc-
et al. 2012). At least all of the genes are present in prokaryotes, ing of 100 bacterial and archaeal genomes. The selection of
widely distributed over the genome and represent proteins, and organisms from sequencing is based on their phylogenetic posi-
are thereby under stabilizing selection because of the functional tions in the tree of life determined by 16S rRNA gene sequence
conservation. However, so far the prerequisite for an rMLST analysis.
The Problem of Horizontal Gene Transfer (HGT) Phenotypic Traits
The increasing investigation of gene sequence and genome- Phenotypic characterizations, including morphology and phys-
based studies shows a remarkable occurrence of horizontal iology, belong to the oldest methods used in prokaryote system-
gene transfer (HGT) affecting genotypic analysis and atics. But these methods are still very important for the initial
impacts a clear resolution of bacteria of clusters of organ- characterization of a strain or a set of strains. Phenotypic traits
isms which derive from a common ancestor. The dimension are the observable characteristics that result from the expression
of the effect of gene flow, especially lateral gene transfer, on of genes of an organism (Moore et al. 2010) but can largely be
microbial evolution and thereby on microbial taxonomy is modulated by environmental or other conditions (e.g., growth
still unclear. conditions, like temperature, pH value, etc.). For phenotypic
Investigations of potential HGT events between bacteria analysis, it is important that strains which under comparison
will get more and more important in microbial evolution. It should grow under the same conditions and cells/cultures in the
has also been shown that changes in gene flow caused by same growth stage should be compared. A detailed and highly
ecological factors may even lead to an incipient merging recommended book chapter on phenotypic characterization of
of two bacterial species, as it has been demonstrated for prokaryotes with detailed references to methods has been pro-
Campylobacter jejuni and C. coli which reverses the process of vided by Tindall et al. (2007).
speciation. It has been published recently that 80% of all genes
were subjected to HGT at some point in their history (Dagan
et al. 2008), even though most estimates are lower (Lienau and Colony Morphology
DeSalle 2009).
The gene content and gene order phylogenies of prokaryotes Strains grown on media show different colony morphologies,
may be strongly affected by gene loss (‘‘big/small genome attrac- including shape, size, color, and structure of the colony. Some
tion’’ and HGT; Wolf et al. 2001). Selection pressure may operate strains form rapidly growing colonies, which spread out, while
on whole sets of genes that are functionally linked to each other. other strains form compact, slow growing colonies. Because
In conclusion, Klenk and Göker (2010) summarized that each strain can show great morphological variation depending
genome-scale data will improve microbial taxonomy consider- on temperature, pH, atmosphere, age, and growth medium,
ably once sufficient coverage of major lineages based on type observations should be made on standardized media and growth
strains has been obtained, and a more detailed insight into the conditions. Colony shape and size, surface, outer edge, produc-
processes given above is provided. They concluded further that if tion of slime, elevation, odor, opacity, swarming behavior, con-
phenotypic information is integrated in a standardized manner sistency, changing color, and any visible additional structures or
in the course of such ventures, genomics as the basis of all -omics features should be noted accurately as they are important
methods might open the door to a truly ‘‘holistic’’ approach to parameters for the characterization of strains. Often, colony
classification. morphology on specific media helps to identify clinical isolates
quickly. The color of colonies on agar surfaces or in deep agar-
shake cultures may help in identification. Examples are Serratia
Nucleic Acid Fingerprinting marcescens within the Enterobacteriaceae, Halobacterium as an
extremely halophilic bacterium, and phototrophic bacteria.
Rapid and high-resolution differentiation of closely related
taxa especially for the differentiation at the subspecies and/or
strain level can be performed by genomic fingerprint analysis Cell Shape and Cell Size
using different methods as amplified fragment-length poly-
morphism PCR (AFLP), random amplified polymorphic Cell shape and cell size vary greatly between species of Bac-
DNA (RAPD) analysis, repetitive element primed PCR (rep- teria and Archaea. This becomes more and more apparent
PCR) using different primers that bind to highly conserved, with the development of better methods to visualize prokary-
repetitive DNA sequences distributed all over the genome, otic cells and to analyze the structure of cells at the subcel-
including repetitive extragenic palindromic (REP)-PCR, lular and molecular level. The most traditional tool for
enterobacterial repetitive intergenic consensus sequences visualizing prokaryotic cells is the light microscope. But
(ERIC)-PCR, and BOX-PCR (derived from the boxA ele- often, a higher resolution is needed to differentiate cell mor-
ment) and (GTG)5-PCR, and ribotyping. A standardized phologies. Electron microscopy serves this purpose. The mor-
application to enable a comparison of results is only possible phology of whole cells can best be differentiated with
for AFLP and ribotyping. For other fingerprint methods, it is scanning electron micrographs, while inner structures, for
important that all strains under consideration are investi- example, the presence of cytoplasmic inclusions or internal
gated in parallel. For the differentiation of a new species to membrane structures can be determined with transmission
related taxa, most of the fingerprint techniques are not suit- electron microscopy. Regardless of the method, cell shape and
able, but they can be applied to show if strains of new taxa form should be described in detail and supported by appro-
are a clone or different representatives of a new species. priate photographs. Samples should be prepared by covering
the slides with thin layers of agar to get the best light micro- ‘‘nonmotile.’’ Nonmotile strains do not show motility at all and
scope image results (Pfennig and Wagener 1986). have no means of locomotion by flagella or gliding. If a cell is
The two most common cell morphologies are rods and cocci. temporarily not moving, it should be described as ‘‘immotile.’’
Rods can have a uniform size and rounded ends or the ends can
have characteristic forms. Cocci often come with a regular
spherical shape, but they can also show different shapes, for Flagellation
example, cocci, which have just divided can seem relatively
flat. Other morphological forms are spirilla and spirochetes, Flagellation of cells can differ regarding the position and the
which can differ in amplitude and wavelength, spirals, or vibrios. number of the flagella. Flagella can be polar, subpolar, or
Next to this, cells can be stalked, sheathed, or divided by budding laterally inserted. They can occur in one location, as tufts of
with its characteristic form. Cell shapes and forms are not only flagella, or distributed around the cell at different locations.
important taxonomic features; they can also play a role in Flagella cannot always be observed by light microscopy. In
medical microbiology to identify pathogenic microorganisms. these cases, cells should be examined using electron micros-
Next to the shape of single cells, it should be noted if cells form copy. Sometimes, flagella may be lost during fixation of cells.
aggregates, for example, myxobacteria, branching hyphae, Nevertheless, flagellation is still recommended as a criterion
mycelia, or sporangia, which are important features to differen- for the classification of bacteria. For further details, see the
tiate actinomycetes. Other features that should be observed are, review by Rhodes (1965).
for example, apolar growth, the formation of prosthecae,
budding, or branching as well as the possible undergoing of
a life cycle. Sporulation
Spores are divided into endospores and exospores. Organisms,

Staining Behavior of the Cell which form endospores, should be examined regarding the
position of the spores and their size in relation to the size of
One of the oldest staining procedures, which is still widely in use the vegetative cell. Endospores can be easily detected under the
and which is an easy and fast method, is the Gram stain (Gram light microscope with phase contrast by examining living cul-
1884). With Gram staining, cell wall structures of prokaryotes tures or Gram staining. If this does not work well, a specific
can be distinguished between cell walls where the dye complex is spore stain on the basis of malachite green can be used. Endo-
washed out (Gram negative) and cell walls where the dye com- spores often are formed under limited nutrient conditions and
plex is retained (Gram positive). Problems can arise if cells are are an important criterion for identification and classification.
too old because the Gram stain reaction alters with the age of For the stimulation of endospore growth, often selective media
cells. Organisms which have a defective cell wall also stain as have to be used; details are summarized by Logan et al. (2009).
Gram negative, although they might be Gram positive. Some Exospores should be described regarding their form and
species react Gram positive in the exponential growth phase and other sporulation features, for example, the formation of
become Gram negative in the stationary phase. In certain sporangia. Shape and color are especially important for the
groups, for example, the Halobacteria, cells have to be fixated identification of actinomycetes.
before staining.
A staining method for organisms, which contain mycolic
acids, is the acid fast staining. This staining procedure is used Cellular Inclusions
mainly for Mycobacteria. Cell components can be stained as well.
Lipophilic cellular inclusions (e.g., polyhydroxybutyric acid) can Cellular inclusions are intracellular structures except spores, for
be stained with Sudan black; extracellular polymers can be example, polyhydroxybutyrate granules, gas vacuoles, phospho-
stained with India ink. rus, and sulfur granules. They are visible under the light micro-
scope and should be studied using appropriate methods like
specific staining procedures. Cellular inclusions of different
Cellular Motility kinds are widespread among bacteria. Polyphosphate granules
can be found in a variety of bacteria, but they play also an
The form and speed of mobility can differ between microorgan- important role for the presumptive identification of clinically
isms; therefore, it should be described precisely. Motility can be relevant bacteria like Corynebacterium diphtheriae.
differentiated, for example, between slow or rapid movements Polyphosphate are included in cells as volutine granules, which
and motility caused by flagella or gliding. Some organisms devel- appear reddish when stained with toluidine blue due to their
oped their own characteristic form of motility, for example, basophilic character. Another name for these volutine granules
myxobacteria, flexibacteria, Cyanobacteria, Chloroflexaceae, and is metachromatic granules. Staining methods include the
Beggiatoaceae. Best results can be observed if fresh cultures are Albert’s and Neisser’s stain, where the polymetaphosphate bod-
examined. Although one has to be aware that motility is difficult ies are blue-black after staining, while the cytoplasm stains light
to observe and cells, which are not moving, are not necessarily brown. Another cellular structures used for differentiation are
globules of elemental sulfur, which can often be found in cytochrome pattern bcaa3 and o occurs predominantly in
phototrophic sulfur bacteria. Protein crystal on the other hand Gram-positive bacteria and the pattern bcdo and a1 is often
can be used to differentiate Bacillus thuringiensis. found in Gram-negative bacteria (Jones and Poole 1985). How-
ever, other studies have revealed that cytochrome d is also
present in some Gram-positive bacteria, in particular in micro-
Pigmentation cocci and coryneform bacteria (Faller et al. 1980; Faller and
Schleifer 1981). Cytochrome c is often absent in both Gram-
Pigmentation of cells can be studied under the microscope or by negative and Gram-positive bacteria, and enterobacteria can be
analyzing the color of cell suspensions or colonies. The study of easily separated from pseudomonads since the former do not
single cells is of less importance than the other two but can help contain cytochrome c and therefore are oxidase negative.
to differentiate pigmented bacteria, for example, purple sulfur When cytochrome patterns are used for taxonomic studies,
bacteria of the family Chromatiaceae, which have a conspicuous it has to be taken into account that growth conditions (nutri-
purple-red carotenoid okenone. tion, growth phase) can influence quantitatively and, to a lesser
Of greater importance is the color of cell suspensions. With extent, qualitatively the cytochrome content of a bacterial cul-
absorption spectra, the identification of pigment-containing ture (Faller et al. 1980; Faller and Schleifer 1981). To assess the
bacteria, for example, phototrophic bacteria, is possible. cytochrome-synthesizing capacity of a strain, cells from both the
Cellular pigments can be soluble in water or organic solvents. exponential and the stationary growth phase (as a minimum)
Pigments, which can be found typically in cells, include caroten- have to be examined.
oids, flexirubins, (bacterio)chlorophyll, melanin, and pyocyanin.
Several methods exist to identify these pigments. Carotenoids
turn blue if concentrated sulfuric acid is added; flexirubins change Ultrastructural Characters
their color reversible when observed under acid or alkaline con-
ditions; (bacterio)chlorophyll can be solved in organic solvents The ultrastructure of cells can be studied by electron microscopy,
and, after saponification, also in water; if treated with acids, which has turned out to be useful in identification. With elec-
(bacteria) phaeophytin is formed; and pyocyanin changes its tron microscopy, cell wall layers, typical intracytoplasmic
color depending on the pH and fluoresces at 360 nm. membrane systems, or characteristics, which differ between
Gram-positive and Gram-negative cells, become visible.
Intracytoplasmic membrane systems are especially important
Bacteriochlorophylls for the identification of phototrophic (cyanobacteria and purple
bacteria), methylotrophic, and chemolithoautotrophic
The determination and differentiation of bacteriochlorophylls can (thiobacilli and nitrifiers) bacteria. Chlorosomes are character-
be helpful in the classification of especially phototrophic bacteria istic for Chlorobiaceae (Staehelin et al. 1978) and phycobilisomes
(Oelze 1985). The majority of the species of the Rhodospir- for cyanobacteria. Carboxysomes are found in autotrophic bac-
illaceae, Ectothiorhodospiraceae, and Chromatiaceae contain teria as polyhedral bodies (Shively 1974) and are characterized
bchl a only. However, some of them possess bchl b instead. by the polymerized form of the enzyme ribulose bisphosphate
In the Chlorobiaceae and Chloroflexaceae, bchl c, d, and e are carboxylase, which appear as conspicuous polyhedral bodies in
the main pigments. They are always accompanied by small autotrophic bacteria only.
amounts of bchl a. Bchl a is also found in Erythrobacter, whereas
Heliobacterium chlorum contains bchl g.
Physiological Properties
Cytochromes In the past, identification of microorganisms was performed

mainly by methods that we now refer to as ‘‘conventional pro-
Bacterial cytochromes are involved in a wide variety of redox cedures.’’ Those included physiological reactions in tubed media
processes such as aerobic and anaerobic respiration and photo- and observation of physical characteristics, coupled with the
synthetic electron transfer. Most of the cytochromes are associ- already mentioned results of Gram staining, other morpholog-
ated with the cytoplasmic membrane, but some exceptions have ical properties, and (sometimes) antimicrobial susceptibility
been determined. Some c-type cytochromes and various profiles.
oxygenases are located within the cytoplasm or the periplasmic The physiological characterization of a new strain in com-
space. Cytochrome patterns of bacteria can be used as a valuable parison to related taxa includes the determination of enzyme
chemotaxonomic character since, unlike the mitochondrial activities, growth on specific substrates, and the determination
respiratory chains of eukaryotes, bacterial respiratory chains of metabolic activity without associated growth (Tindall et al.
contain a greater variation of different cytochromes (Jones and 2010). Tindall et al. (2007) as well as Krieg and Padgett (2011)
Poole 1985). Conventional difference spectrophotometry, pref- give excellent overviews for a broad range of routine tests,
erably at low temperatures, is sufficient to determine the cyto- which can be applied for the physiological identification of
chrome pattern (Faller et al. 1980). It has been reported that the prokaryotic taxa.
It should be determined if a new strain is phototrophic, minimal standards (see above) can be used to determine which
chemoautotrophic, or chemoheterotrophic. The new taxa physiological tests are required for respective taxa.
should furthermore be characterized with respect to the require- The activity of a broad range of metabolically active enzymes
ments of molecular oxygen and/or the growth in presence of are determined to characterize the metabolic proportion of
molecular oxygen: aerobic, microaerophilic, anaerobic, and fac- a new strain, for example, activity of glycosidases (measure-
ultative anaerobic growth. ments by the used of chromogenic or fluorogenic substances
Important applied tests for the investigation of the presence linked to respective substrates), lipases, peptidases, phenylalanin
and activity of respiratory enzymes are the oxidase and deaminase, or phosphatases, to name a few, are determined.
catalase tests. The investigation of the oxidation and fermentation of dif-
Oxidases transfer electrons from a donor molecule to molec- ferent compounds (mostly sugars) as well as the determination
ular oxygen and thereby occur most likely in aerobic and/or of carbon substrate utilization pattern is mandatory for the
microaerophilic microorganisms, which use molecular oxygen characterization of new taxa.
as terminal electron acceptor. An artificial electron acceptor can Within the last decades, several methods simply miniatur-
be used as test reagent (e.g., N, N, N0 ,N0 -tetramethyl-p- ized commonly used biochemical reactions into a more conve-
phenylenediamine (TMPD)) which turns to a colored nient format, and this system-dependent approach has become
compound after reaction. Bacteria harbor cytochromes, mostly more and more standard, especially for the identification of
associated with cytoplasmic membranes, which are involved in clinically relevant bacteria. These methods are based on sets of
a wide variety of redox processes, as the aerobic or anaerobic substrates that are carefully selected for their positive and neg-
respiration or photosynthetic electron transfers. Cytochrome c is ative reactions. These patterns create metabolic profiles that are
often absent in both Gram-negative and Gram-positive bacteria. compared with established databases, in most cases by numeri-
For example, enterobacteria can be distinguished from pseudo- cal procedures.
monads regarding the absence of cytochrome c (determined as The numerical identification procedures developed by
oxidase negative bacteria) in the former ones. The comparison Sneath and coworkers in the 1960s and 1970s (Sneath and
of cytochrome patterns can also be used for the differentiation of Sokal 1973) are either manual or automated. For all systems
micrococci and staphylococci because the later one usually lack combined from carefully selected tests, the backbone of accuracy
cytochrome c and d. is the strength and utility of the database. Databases are
Catalases are enzymes that prevent oxidative damage caused constructed by using known and well-characterized strains and
by hydrogen peroxide via the disproportionation of hydrogen include the type strains of most taxa. The major disadvantage of
peroxide to water and oxygen. The presence and activity of this approach is that atypical microorganisms often cannot be
catalase can be easily tested by ‘‘bubble formation’’ on cultures reliably identified by these commercial systems. Furthermore,
(grown on agar plates) after the addition of some drops of a 3% the number of species included in the databases may vary from
hydrogen peroxide solution. just a few for some manual assays to thousands for automated
Temperature-, pH-, and salinity (%w/v NaCl)-dependent instruments.
growth range and growth optima are routinely performed for Database maintenance must be a continuous process and
the characterization of new taxa. Optimal growth conditions software upgrades incorporating major taxonomic changes
often depend on the environmental condition of the original must be provided by the manufacturer at regular intervals.
habitat of the strains under study and can thereby already differ This poses more and more a problem because the majority of
between closely related taxa which were originally isolated from novel species in the last 10 years have been proposed on a single
different environments, as, for example, freshwater (low salt strain (the type strain) only. Hence, the physiological diversity of
content) or marine systems (high salt content). For the test of these species are unknown and cannot be considered in the
pH-dependent growth, a puffer should be used to stabilize the databases.
pH of the used medium. Buffers have to be selected critically System identifications are nowadays computer-based and
regarding the puffer capacity, the use of a puffer as substrate, and done by algorithm-based decision making. In some cases, the
the possible negative effect of a puffer to the tested organisms. list of identifications is compiled into a preprinted index, which
The most often used buffer systems are phosphates buffers. may then be consulted to manually convert the organism’s
Carbon utilization or assimilation pattern is often a routine biochemical profile number into an identification result. In
in test for the characterization of new taxa including the strains many cases, Bayes’ theorem, or modifications of it, is often the
of new taxa and of related taxa which have to be compared in basis of algorithm construction from data matrices.
a respective study. It is important to note here that microbiologists should not
Nitrate reduction or denitrification, as well as the require- always become dependent upon these likelihoods and percent-
ments of CO2 or hydrogen for growth, or the use of H2 as an ages when interpretive judgment would suggest an alternative
energy and electron source for some autotrophic bacteria are taxonomic conclusion. Bacteria may not react in specific tests
metabolic properties which have to be tested for specific taxa. as expected in a commercial system, and even though
Several other physiological properties can and have to be tested a legitimate result is produced (e.g., lactose-positive Salmonella
dependent on the taxa of interest. Comparison to the character- spp. or H2S-positive Escherichia coli), these results may be
ization of related taxa or requirements listed in respective misleading.
Physiological and/or biochemical tests are in the majority of linkage clusters. Another way to express relations between
cases determined by the reactions of individual organisms with OTUs after an analysis is by the use of dendrograms. Kaneko
different substrates in a system that gives often a color change or (1979) introduced the correlative similarity coefficient as
a change in turbidity in case of a positive reaction. The accuracy a new criterion for forming dendrograms. Other coefficients
of these reactions is heavily dependent upon the users following for numerical taxonomy have been evaluated by Austin and
exactly the directions of the manufacturer regarding the prepa- Colwell (1977).
ration of the inoculum, adjustment of the inoculum density, the Detailed information about the use of computer analysis in
incubation conditions, and test interpretation. numerical taxonomy can be obtained from Sneath (1972, 1977,
Many tests rely upon one or a combination of several indi- 1979a, b, c, 1980a, b, c). Computer-based identification systems
cators. These may include (1) pH changes resulting from acid have also been described, for example, by Edwards (1978);
formation and/or alkalinization as a result of the utilization of Kellogg (1979); Schindler et al. (1979); Beers and Lockhard
the substrate, (2) turbidity changes resulting from the growth (1962); Gyllenberg (1965); Holmes and Hill (1985); and Lapage
and utilization of the substrate, (3) enzymatic reactions on et al. (1970, 1973).
a chromogenic and/or fluorogenic substrate that allow the
release of a colored or fluorescent compound, (4) tetrazolium-
based indicators of metabolic activity (here often dehydrogenase New Ways for Phenotypic Characterization
activities) in the presence of a variety of carbon sources,
(5) detection of volatile or nonvolatile acids. Additional tests Recently, more sophisticated phenotyping systems analyzing
for microbial identification that use other means of detecting different cellular fractions have been introduced into systemat-
a positive response for a given substrate may also be included. ics, for example, the matrix-assisted-laser desorption/ionization
time of flight mass spectrometry (MALDI-TOF MS) (Welker
and Moore 2011; Moore and Rosselló-Móra 2011) or the high-
Numerical Taxonomy and Numerical field ion cyclotron Fourier transform mass spectroscopy (ICP-
Identification FT MS) technique, which are referred to metabolomics
(Rosselló-Móra 2012). However, despite the advantages of
Numerical taxonomy is the well-established means for the these systems to generate high numbers of data, which can be
assessment and evaluation of phenotypic data (Goodfellow stored in databases, the general restrictions of cultivation-based
1977; Sneath 1971; Sneath and Sokal 1973; Sokal and Sneath dependencies of the cultivation conditions also apply to these
1963; Hill 1974). In numerical studies, results are tabulated in techniques.
a table of t organisms versus n characters, and operational
taxonomic units (OTUs) are assigned for each individual strain.
Characters under study should be independent and should come Chemotaxonomy
about equally from the various different categories of properties
(morphology, physiology, biochemistry, serology, etc.). For sta- Chemotaxonomic characterization is based on the investiga-
tistical reasons, the total number of characters should be above tion of various structural cellular compounds of prokaryotes.
60. Whereas in conventional taxonomy, so-called characteristic The most important investigated structural compound are
tests and differentiating media are in use (with special impor- fatty acids, polar lipids, respiratory chinones, and for some
tance for identification), numerical taxonomy principally allots extent pigments of the outer membrane(s), cytoplasmatic
the same weight to each character because there is no logical polyamines or the peptidoglycan, teichoic acids, mycolic
alternative that would allow independence from the personal acids, or lipopolysaccharides (LPS), which are compounds of
opinion of the scientist. the outer cell layer.
The number of common characters is considered as The order of the importance of specific chemical com-
a quantitative measure for the taxonomic relationship (not pounds for the characterization of new taxa primarily
phylogenetic ‘‘relatedness’’). As an important consequence, not depended on the chemotaxonomic trades used for the charac-
all members (species) of a group must have one special property terization of closest related taxa. As outlined above, the iden-
in common. Similarities are quantitatively expressed by the tification of closest related taxa is at most often done by the
matching coefficient of Sokal and Michener (Skerman 1967), phylogenetic affiliation based on nearly full-length 16S rRNA
usually expressed in percent: gene sequences.
Sum of positive and negative matches
SSM ¼
Total number of tests
Cell Wall and Membrane Compounds
The similarities between each OTU and every other OTU
under study are set out in tables in the form of a triangular Peptidoglycan
similarity matrix (also called a Sneath diagram). By forming Peptidoglycan is the most predominant cell wall compound of
linkage groups and performing cluster analyses, the numerical both, Gram-negative and Gram-positive bacteria. Peptidoglycan
taxonomist finally arrives at a rearranged matrix that depicts is a heteropolymer of glycan strand with alternating ß-1,4-linked
residues of N-acetylglucosamine and N-acetylmuramic acid, Lipopolysaccharides (LPS)

which are cross-linked by short peptides. N-acetylmuramic Few routine studies are now carried out, but it is evident from
acid is a derivative of glucosamine and occurs in cells specifically work over the past 6 decades that both the nature of the sugar
as a component of the peptidoglycan. Gram-negative bacteria present in lipid A as well as the nature of the fatty acids and the
harbor a monolayer of peptidoglycan, which is remarkable way they are linked (ester and/or amide linked) to the sugar is of
uniform in its composition. The multilayer peptidoglycan of significance (Hase and Rietschel 1976; Weckesser and Mayer
Gram-positive bacteria is characterized by a great variety in the 1988). The chain length of the fatty acids in the LPS may also
chemical composition. Therefore, the discriminatory power of differ significantly from those found in the polar lipids. When
peptidoglycan structures is apparently restricted to Gram- carrying out whole cell fatty acid analysis, the fatty acids from
positive bacteria. Especially for Proteobacteria and Bacteroidetes, the LPS will also be extracted, and this should be borne in mind
no characteristic variations could be determined (Schleifer and when interpreting the data.
Kandler 1972).
The advantage of peptidoglycan is that they are thought to Mycolic Acids
be less affected by mutations, which could alter the peptidogly- Mycolic acids are characteristic lipid compound of the cell
can structure, and is less affected by phenotypic variations envelope of mycobacteria and related high-GC Gram-positive
(Schleifer et al. 1976). bacteria (Brennan 1988). The main fraction of mycolic acids are
The analysis of the peptidoglycan structure can be ester-linked fraction of arabinogalactan which is attached to
performed at different levels. At first, characteristic diamino peptidoglycan.
acid in the cross-linking peptide can be determined. Analysis For bacteria that produce mycolic acids, they can be used as
of the peptidoglycan type (A type: cross-linkage of the two addition taxonomic markers because it was shown that the
peptide side chains via amino acid 3 of one peptide subunit to phylogenetic relationships based on the 16S rRNA gene
amino acid 4 of the other peptide subunit; B type: cross-linkage sequence analysis and the grouping based on the length of
of the two peptide side chains via amino acid 2 of the one mycolate side chain correlated well with each other. It has also
peptide subunit to amino acid 4 of the other peptide subunit), been of taxonomic relevance if taxa of groups, which are
mode of cross-linkage (direct or interpeptide bridge and amino characterized by the content of mycolic acids, are lacking.
acids in the bridge), and complete amino acid composition As described above, mycolic acids can be detected by the
provides more detailed information. B-type peptidoglycan is staining with carbol fuchsin (acid fast staining) but also detailed
characteristic for all genera of the family Microbacteriaceae and investigates after extraction and derivatization by thin-layer
members of the Erysipelothrix/Holdemania group. All other chromatography (Dobson et al. 1985), HPLC (Willumsen et al.
murein-containing bacteria so far analyzed exhibit the A type. 2001), gas chromatography (Rainey et al. 1995; Müller et al.
The amino acid composition of the peptide side chain, including 1998; Torkko et al. 2003), or mass spectrometry (Fujita et al.
the characteristic diamino acid is usually common to all species 2005a, b).
of a genus. However, a higher degree of variability has been
detected in the mode of cross-linkage between the peptide side Fatty Acids and Isoprenoid Side Chains
chains, which often differ among species of certain genera (e.g., Almost all Bacteria contain fatty acids ester linked to the glycerol
members of the genus Microbacterium) but may differ also as typical membrane constituents. The comparison of fatty acid
between strains of a single species as reported for M. luteus. pattern can be used to assign new strains to related taxa because
Analysis of the peptidoglycan structure is a requirement for all the fatty acid composition generally does not fluctuate signifi-
members of novel Gram-positive genera when they are cantly within a taxonomic group. Otherwise, differences
described, and at least the amino acid composition should be obtained between closely related strains can also be used to
provided for every novel Gram-positive species. In the majority differentiate related taxa from each other. But, it has to be
of cases, the amino acid composition of the peptide side chain of critically question if, for example, branched chain fatty acids
the type species of a novel genus may be shared by future species are reported in a group that otherwise synthesizes straight chain
assigned to the genus, and hence, it should be listed in the genus and unsaturated fatty acids. In a similar manner, the presence or
description as a characteristic trait. The complete composition absence of hydroxylated fatty acids is generally characteristic,
of the peptidoglycan of a novel species of a genus that has and their unexpected absence or presence should also be treated
previously been described should be in agreement with the with caution.
characteristics of the genus and may provide differences in the The MIDI system generally used for fatty acid analysis pro-
interpeptide bridge useful for differentiation from other species. vides a comprehensive database set, which however is certainly
A list of peptidoglycan variations can be found at http://www. not complete and thereby may give discrepancies by the identi-
dsmz.de/species/murein.htm. This system is slightly different to fication that need to be clarified, for example, compounds that
that developed by Schleifer and Kandler (1972). are currently not included in the database has to be assessed. In
Pseudomurein, a characteristic peptidoglycan, has been general, all components that constitute 1% or more of the fatty
detected in some Gram-positive members of the Archaea, in acids must be reported; if major peaks were not identified, they
which N-acetylmuramic acid is replaced by N-acetyltalosuronic will not be included in the peak-naming table, but their presence
acid (König et al. 1982, 1983). must be reported and the equivalent chain length (ECL) should
be given. This will allow any future work on the elucidation of Polar Lipids
the structure to link to the ECL given in publications. In addition to the analysis of the fatty acids composition, polar
It is important to determine fatty acid patterns of strains, lipids are independent chemotaxonomic markers for prokary-
which should be compared, from cells that were cultured under otic classification. Prokaryotes harbor a vast diversity of polar
identical condition (medium, temperature) and were in the lipids associated with cellular membranes. Among those phos-
same physiological growth stage prior to fatty acid extraction. phate lipids are the most commonly known polar lipids; how-
Exceptions may be made if strains could not be grown under the ever, several others are also present, which are so far not all
same conditions. But this must then be carefully documented. identified (Ratledge and Wilkinson 1988). For taxonomic pur-
The MIDI system extracts fatty acids from intact cells (Miller poses, polar lipid of strains are extracted and resolved by two-
1982), and comparisons with work that has been carried out on dimensional thin-layer chromatography. Standardized condi-
the lipid fraction(s) extracted from cells or on fatty acid methyl tions are an indispensable prerequisite for those analyses. Several
ester mixtures that have previously been separated into different thin layer chromatograms have to be generated to obtain
classes by thin layer chromatography are not sensible and falsify a detailed analysis for the polar lipid profile of an organism. At
the interpretation fatty acid compositions. first, polar lipids are stained with a reagent that visualizes all
Several Bacteria also synthesize ether-linked lipids in addition polar lipids. It is recommended to scan those thin layer plates
to fatty acids, which have either straight chains or simple with high resolution (300 dpi) and to provide the polar lipid
branched (not isoprenoid) side chains or monounsaturated profiles as 8-bit gray-scale images with a size of 7 7 cm
derivatives, with the point of unsaturation adjacent to the ether (Tindall et al. 2010).
bond (i.e., plasmalogens or vinyl ethers). Fatty acid methyl ester Different spraying reagents have to be used to stain specific
for gas chromatographic fatty acid analysis is prepared under characteristic functional groups of polar lipids including phos-
acidic conditions. Under those conditions, it comes to the hydro- phates, alpha-glycols, sugars, free amino groups, quaternary
lysis of plasmalogens, and/or vinyl ethers and dimethyl acetals are nitrogen, and primary and secondary amines. For most of the
formed, which elute with the fatty acid methyl esters. If the stains, separate thin layer plates have to be prepared. Based on
presence of those compounds are known or suspected, their the staining reactions and Rf values, various polar lipids can be
presence/absence should be investigated and reported. However, identified. As it is known so far, membranes of Bacteria contain
gas chromatographic results have to be interpreted with care, as it phospholipids, glycolipids, phosphoglycolipids, aminopho-
has been reported by Tindall et al. (2010) that the presence of spholipids, amino-acid-derived lipids, capnines, sphingolipids
dimethly acetals can be misinterpreted in the presence of (glyco- or phosphosphingolipids), and also hopanoids. For
hydroxylated fatty acids, which elute with the same retention archaea, only phospholipids, glycolipids, and phosphogly-
time (e.g., Moore et al. 1994; Helander and Haikara 1995). colipids are known as membrane substitutions.
Nonplasmalogenic mono- and diethers also occur in the All identified polar lipids should be assigned on the polar
members of the Bacteria (Langworthy et al. 1983; Rütters et al. lipid profile stained with the unspecific staining reagent. If
2001) but do not cochromatograph with mono- or diethers unknown lipids were determined, which were stained by specific
from members of the Archaea. But, however, similar methods reagents, they should be termed accordingly, for example, as
can be applicable for analysis. ‘‘unidentified phospholipids’’ or ‘‘glycolipid.’’
Specific modified fatty acids occur in a variety of polar lipids A specific characteristic for some taxa of Bacteria
of some Bacteria, which, for example, are formed by condensa- are polar lipids with hydroxylated fatty acids which can
tion reactions with small molecular weight molecule as amino cause changes of Rf values compared to the respective
acids. Sphingolipids (Naka et al. 2003), capnines (Godchaux and compounds without hydroxylated fatty acids (Cox and
Leadbetter 1984), and alkylamines (Anderson 1983) are well- Wilkinson 1989; Kroppenstedt et al. 1990; Kroppenstedt and
known examples. Goodfellow 1991).
Spots of polar lipid profiles are often referred to be single
Hopanoids compounds if they are, for example, labeled as phosphatidyl-
Sterols are normally not synthesized by prokaryotes. There are glycerol. But however, it has to be considered that dependent on
only two genuine exceptions, namely, Methylococcus capsulatus the investigated organism, a variety of fatty acids can be bound
and Nannocystis exedens (Tornabene 1985). However, in recent to glycerin, and each of the fatty acid combinations results in
years, hopanoids and other polyterpenoids have been discovered a new compound. Therefore, spots should be more appropri-
in prokaryotes (Ourisson et al. 1987), which can be considered ately treated as being a class of compounds. In any one organism,
as sterol surrogates. Hopanoids have been found in various the spot labeled phosphatidylglycerol may contain several dif-
eubacteria. For instance, they are present in most cyanobacteria, ferent compounds with different fatty acid compositions.
methylotrophs, Rhodospirillaceae, acetic acid bacteria, and vari-
ous other Gram-negative and Gram-positive eubacteria. They Respiratory Lipoquinone Systems
have not been found in archaebacteria, purple sulfur bacteria, or Respiratory lipoquinones (isoprenoid quinones) are widely dis-
Enterobacteriaceae. Although archaebacteria do not contain tributed in both anaerobic and aerobic organisms, within the
hopanoids, they have phytanyl and bisphytanyl ethers that are Bacteria and Archaea, and play an important role in electron
also polyterpenoids. transport. Based on the structural composition, two major
classes can be divided, the naphthoquinones (menaquinones) Cytoplasmatic Compounds

and benzoquinones (including ubiquinones, rhodoquinones,
and plastoquinones). A third class includes derivate of Polyamines
benzothiophene derivatives, which however seemed to be Polyamines have been detected in a wide range of concentrations
restricted to members of the order Sulfolobales (Tindall 2005). in most prokaryotes.
Ubiquinones are wildly distributed in bacteria and eukaryotes They are small molecular weight compounds located in
whereas plastoquinones occur in plants, algae, and the cytoplasm with a broad range of functions including the
cyanobacteria. stabilization of DNA, involvement in gene expression, and
Isoprenoid quinones can be easily degraded and therefore intracellular compensation of extracellular changes in osmotic
samples before and during extraction should be handled very conditions (Feuerstein et al. 1991; Eraso and Kaplan 2009).
carefully. Biomass and extracts should be flushed with N2 before Certain polyamines are also known to be covalently linked to
they are stored at 20 C to prevent oxidation, and dimmed the peptidoglycan of members of the Sporomusa-Pectinatus-
light condition is recommended for extraction to avoid photo- Selenomonas evolutionary group (Hirao et al. 2000; Kamio
oxidation. The analysis of quinones is performed by HPLC. et al. 1981a, b; Kamio and Nakamura 1987). Polyamine pattern
Before analysis, ubiquinones and menaquinones are often sepa- can be used to support to differentiate and define taxonomic
rated by thin layer chromatography (TCL) and purified from groupings (Busse and Auling 1988; Hamana and Matsuzaki
respective TLC plates. 1992; Altenburger et al. 1997; Busse and Schumann 1999;
Ubiquinones and menaquinones vary in the length of the Busse 2011). The most commonly detected polyamines are 1,3-
isoprenyl units in polyprenyl side changes, which are denoted diaminopropane, putrescine, 2-hydroxyputrescine, cadervarine,
with Q-n or MK-n, with n representing the number of isoprenyl sym-norspermidine, spermidine, sym-homospermidine, and
units. The side-chain lengths recorded to date is in the range of spermine.
5–15 isoprenoids units. The isoprenoid side chains of The analysis of polyamines should be performed from cells
menaquinones can be stereospecific hydrogenated (saturated), of a standardized physiological age; it is recommended to har-
which is marked by the appreciation Hn with n representing the vest the cells at the late exponential growth phase because a high
number of hydrogen in the isoprenoid side chain. polyamine content is obtained in cells of this growth phase
Menaquinones then are assigned as MK-n(Hn). (For more (Busse and Auling 1988). The analysis of polyamines is carried
details, see da Costa et al. 2011.) out by gradient HPLC with fluorescence detection after poly-
The occurrence of ubiquinones seems to be restricted to the amine extraction by acidic hydrolysis and derivatization with
Alphaproteobacteria, Gammaproteobacteria, and Betaproteo- dansylchloride (Scherer and Kneifel 1983). The content of poly-
bacteria (Tindall 2005). If those are reported for other taxa, the amines is commonly given in mmol per g dry or wet weight of
results should be treated with caution. The most abundant cell biomass. For classification, the identification of the predom-
ubiquinone in Alphaproteobacteria is Q-10, but taxa with Q-8 inant polyamine is the most important criteria (Busse 2011).
were also determined. The variety of ubiquinone in Gammapro- The presence or absence of some minor compounds however
teobacteria is higher, and characteristic ubiquinones are Q-7 to can also be very important, for example, for the differentiation
Q-14, whereas Q-10 was not determined as a single predomi- of species of closely related genera.
nant ubiquinone in this class. For example Legionella
spp. typically synthesize more than one predominant ubiqui-
none, with chain lengths between Q-10 and Q-14 (Collins and Other Features of Taxonomic Value
Gilbart 1983; Karr et al. 1982). Menaquinones and
rhodoquinone are also produced by some taxa of the Alphapro- It should be remembered that prokaryotes are chemically
teobacteria and Gammaproteobacteria. The relative amount of diverse, and the presence of compounds such as teichoic and
those lipoquinones found in the cells depends on the used teichuronic acids (Fischer 1988; Neuhaus and Baddiley 2003;
growth substrates and the use of oxygen as a terminal electron Hancock 1994), arabinogalactans (Brennan 1988), and other
acceptor. heteropolysaccharides (Hancock 1994; König 1994; Kandler
For some methanotrophic bacteria, ubiquinones with mod- and Hippe 1977; Schleifer et al. 1982) or the substitution of
ified side chains were determined (Collins 1994); in the majority sphingoglycolipids for lipopolysaccharides in members of the
of Bacteria and Archaea lineages, derivates of menaquinone were family Sphingomonadaceae all contribute to differentiating var-
found, including demethyl-menaquinones, monomethylmena- ious taxonomic groups (see, e.g., Lechevalier and Lechevalier
quinones, dimethylmenaquinones, and menathioquinones 1970) within the prokaryotes (members of the Bacteria and
(Collins 1985; Collins 1994; Tindall 2005). Archaea) and that this information is encoded somewhere on
Isoprenoid side chains in high-GC Gram-positive bacteria the genome. Rahman et al. (2009a, b) have provided such
show different patterns of hydrogenation of menaquinones, an overview of the distribution of teichoic and lipoteichoic
whereas it is characteristic for members of the Deltaproteo- acids in actinomycetes, linking that information to the underly-
bacteria or Halobacteria that only one point of the side changes ing genetic information encoding critical steps in their
are saturated (Collins 1994). biosynthesis.
Conclusion Bergey DH, Harrison FC, Breed RS, Hammer BW, Hantoon FM (eds)
(1923) Bergey’s manual of determinative bacteriology, 1st edn. Williams
and Wilkins, Baltimore
Moore et al. (2010) has pointed out recently that taxonomy Bernardet JF, Nakagawa Y, Holmes B (2002) Proposed minimal standards for
serves many purposes. It should be stable and predictable. describing new taxa of the family Flavobacteriaceae and emended description
A most comprehensive phenotypic and genotypic characteriza- of the family. Int J Syst Bacteriol 52:1049–1070
tion (in the framework of a polyphasic taxonomic approach) is Breed RS, Murray EGD, Smith NR (eds) (1957) Bergey’s manual of determinative
bacteriology, 7th edn. Williams & Wilkins, Baltimore
still necessary in characterization and classification (which is
Brennan PJ (1988) Mycobacterium and other actinomycetes. In: Ratledge C,
a prerequisite of identification). As already pointed out by Wilkinson SG (eds) Microbial lipids, vol 1. Academic, London, pp 203–298
Wayne et al. (1987), an ideal taxonomy would involve one Brenner DJ (1973) Deoxyribonucleic acid reassociation in the taxonomy of
system (a hierarchical system), and in microbiology, the ulti- enteric bacteria. Int J Syst Bacteriol 23:298–307
mate ambition would be to establish a system that mirrors the Brenner DJ, Martin MA, Hoyer BH (1967) Deoxyribonucleic acid homologies
among some bacteria. J Bacteriol 94:486–487
taxonomic relationships as an ‘‘order in nature,’’ which is now
Brenner DJ, Staley JT, Krieg NR (2001) Classification of prokaryotic organisms
most often associated with ‘‘evolutionary order’’ back to the and the concept of bacterial speciation. In: Garrity GM (ed) Bergey’s manual
origin of life. The circumscription of the taxonomic rank ‘‘spe- of systematic bacteriology, vol 1, 2nd edn. Springer, New York, pp 27–31
cies’’ provided by Stackebrandt et al. (2002) was given as an open Brown DR, Whitcomb RF, Bradbury JM (2007) Revised minimal standards for
and workable ‘‘definition’’ allowing the integration of new infor- description of new species of the class Mollicutes (division Tenericutes). Int
J Syst Evol Microbiol 57:2703–2719
mation and new developments (which can be expected in the era
Buchanan RE, Gibbons NE (eds) (1974) Bergey’s manual of determinative bac-
of ‘‘omics’’): ‘‘A species is a category that circumscribes teriology, 8th edn. Williams & Wilkins, Baltimore
a (preferably) genomically coherent group of individual iso- Busse H-J (2011) Polyamines. In: Rainey F, Oren A (eds) Taxonomy of pro-
lates/strains sharing a high degree of similarity in (many) inde- karyotes, vol 38, Methods in microbiology. Elsevier, Amsterdam, pp 15–60
pendent features, comparatively tested under highly Busse H-J, Auling G (1988) Polyamine pattern as a chemotaxonomic marker
within the Proteobacteria. Syst Appl Microbiol 11:1–8
standardized conditions.’’ The introduction of quality standards
Busse H-J, Schumann P (1999) Polyamine profiles within genera of the class
is highly desirable. At present, both traditional and also novel Actinobacteria with LL-diaminopimelic acid in the peptidoglycan. Int J Syst
phenotypic approaches are important for the characterization of Bacteriol 49:179–184
novel species and genera. The development of new methods and Christensen H, Bisgaard M, Frederiksen W, Mutters R, Kuhnert P, Olsen JE
the improvement of existing methods for phenotypic character- (2001) Is characterization of a single isolate sufficient for valid publication
of a new genus or species? Proposal to modify recommendation 30b of the
ization should be encouraged. Genomic (and other omic)
bacteriological code (1990 revision). Int J Syst Evol Microbiol 51:2221–2225
approaches will provide a rich source of data that should be Christensen H, Kuhnert P, Busse H-J, Frederiksen WC, Bisgaard M (2007) Pro-
carefully investigated in regard to the information behind. It is posed minimal standards for the description of genera, species and subspe-
hoped that we may be able to provide a better picture of the vast cies of the Pasteurellaceae. Int J Syst Evol Microbiol 57:166–178
biodiversity of the prokaryotic world based on a taxonomic Chun J, Lee J-H, Jung Y, Kim M, Kim S, Kim BK, Lim Y-W (2007) EzTaxon:
a web-based tool for the identification of prokaryotes based on 16S ribo-
scheme that reflects nature (Kämpfer 2012).
somal RNA gene sequences. Int J Syst Evol Microbiol 5:2259–2261
Cohn F (1875) Untersuchungen über Bacterien II. Beitr Biol Pflanz 1(3):141–208
Cole JR, Konstandinidis K, Farris RJ, Tiedje JM (2010) Microbial diversity and
References phylogeny: extending from rRNAs to genomes. In: Liu W-T, Jackson JK (eds)
Environmental molecular microbiology. Caister Academic Press, Norfolk,
Achtman M, Wagner M (2008) Microbial diversity and the genetic nature of pp 1–19
microbial species. Nat Rev Microbiol 6:431–440 Collins MD (1985) Isoprenoid quinone analyses in bacterial classification and
Altenburger P, Kämpfer P, Akimov VN, Lubitz W, Busse H-J (1997) Polyamine identification. In: Goodfellow M, Minnikin DE (eds) Chemical methods in
distribution in actinomycetes with Group B peptidoglycan and species of the bacterial systematics, vol 20, SAB technical series. Academic, London,
genera Brevibacterium, Corynebacterium, and Tsukamurella. Int J Syst pp 267–287
Bacteriol 47:270–277 Collins MD (1994) Isoprenoid quinones. In: Goodfellow M, O’Donnell AG (eds)
Amann RI, Lin C, Key R, Montgomery L, Stahl DA (1992) Diversity among Chemical methods in prokaryotic systematics. Wiley, Chichester,
Fibrobacter isolates: towards a phylogenetic and habitat-based classification. pp 345–401
Syst Appl Microbiol 15:23–31 Collins MD, Gilbart J (1983) New members of the coenzyme Q series from the
Anderson R (1983) Alkylamines: novel lipid constituents in Deinococcus Legionellaceae. FEMS Microbiol Lett 16:251–255
radiodurans. Biochim Biophys Acta 753:266–268 Collins MD, Rodrigues U, Ash C, Aguirre M, Farrow JAE, Martinez-Murcia A,
Arahal DR, Vreeland RH, Litchfield CD, Mormile MR, Tindall BJ, Oren A, Phillips BA, Williams AM, Wallbanks S (1991) Phylogenetic analysis of the
Bejar V, Quesada V, Ventosa A (2007) Recommended minimal standards genus Lactobacillus and related lactic acid bacteria as determined by reverse
for describing new taxa of the family Halomonadaceae. Int J Syst transcriptase sequencing of 16S rRNA. FEMS Microbiol Lett 77:5–12
Evol Microbiol 57:2436–2446 Colwell RR (1970) Polyphasic taxonomy of bacteria. In: Iizuka H, Hasegawa T
Austin B, Colwell RR (1977) Evaluation of some coefficients for use in numerical (eds) Culture collections of microorganisms. University of Tokyo Press,
taxonomy of microorganisms. Int J Syst Bacteriol 27:204–210 Tokyo, pp 421–436
Bartual SG, Seifert H, Hippler C, Luzon MA, Wisplinghoff H, Rodriguez-Valera F Cowan ST (1968) A dictionary of microbial taxonomic usage. Oliver & Boyd,
(2005) Development of a multilocus sequence typing scheme for character- Edinburgh
ization of clinical isolates of Acinetobacter baumannii. J Clin Microbiol Cowan ST (1978) Hill LR (ed) Dictionary of microbial taxonomy. Cambridge
43:4382–4390 University Press, Cambridge
Beers RJ, Lockhard WR (1962) Experimental methods in computer taxonomy. Cox AD, Wilkinson SG (1989) Polar lipids and fatty acids of Pseudomonas cepacia.
J Gen Microbiol 28:633–640 Biochim Biophys Acta 1001:60–67
Da Costa MS, Albuquerque L, Nobre MF, Wait R (2011) The extraction and Gevers D, Cohan FM, Lawrence JG, Spratt BG, Coenye T, Feil EJ et al (2005)
identification of respiratory lipoquinones of prokaryotes and their sue in Re-evaluating prokaryotic species. Nat Rev Microbiol 3:733–739
taxonomy. In: Rainey F, Oren A (eds) Taxonomy of prokaryotes, vol 38, Godchaux W, Leadbetter ER (1984) Sulfonolipids of gliding bacteria. Structure of
Methods in microbiology. Elsevier, Amsterdam, pp 197–206 the N-acylaminosulfonates. J Biol Chem 259:2982–2990
Dagan T, Artzy-Randrup Y, Martin W (2008) Modular networks and cumulative Goodfellow M (1977) Numerical taxonomy. In: Laskin AI, Lechevalier HA (eds)
impact of lateral transfer in prokaryote genome evolution. Proc Natl Acad CRC handbook of microbiology, vol 1, 2nd edn. CRC Press, Cleveland,
Sci USA 105:10039–10044 pp 579–596
De Ley J, Park IW, Tijtgat R, van Ermengem J (1966) DNA homology and Goris J, Konstantinidis KT, Klappenbach JA, Coenye T, Vandamme P,
taxonomy of Pseudomonas and Xanthomonas. Microbiology 42:43–56 Tiedje JM (2007) DNA-DNA hybridization values and their
De Vos P, Trüper HG (2000) Judicial Commission on the International relationship to whole-genome sequence similarities. Int J Syst
Committee of Systematic Bacteriology. IXth International (IUMS) congress Evol Microbiol 57:81–91
of bacteriology and applied microbiology. Minutes of the meetings, 14, 15, Gram C (1884) Ueber die isolierte Färbung der Schizomyceten in Schnitt- und
and 18 August 1999, Sydney, Australia. Int J Syst Evol Microbiol Trockenpärparaten. Forschitte Med 2:185–189
50:2239–2244 Grimont PAD, Popoff MY, Grimont F, Coynault C, Lemelin M (1980) Reproduc-
De Vos P, Trüper HG, Tindall BJ (2005) Judicial Commission of the International ibility and correlation study of three deoxyribonucleic acid hybridization
Committee on Systematics of Prokaryotes. Xth international (IUMS) con- procedures. Curr Microbiol 4:325–330
gress of bacteriology and applied microbiology. Minutes of the Meetings. Int Gyllenberg HG (1965) A model for computer identification of microorganisms.
J Syst Evol Microbiol 55:525–532 J Gen Microbiol 39:401–405
Deloger M, El Karoui M, Petit M-A (2009) A genomic distance based on MUM Hamana K, Matsuzaki S (1992) Polyamines as a chemotaxonomic marker in
indicates discontinuity between most bacterial species and genera. J Bacteriol bacterial systematics. Crit Rev Microbiol 18:261–283
191:91–99 Hancock IC (1994) Analysis of cell wall constituents of Gram-positive bacteria.
Dewhirst FE, Fox JG, On SLW (2000) Recommended minimal standards for In: Goodfellow M, O’Donnell AG (eds) Chemical methods in prokaryotic
describing new species of the genus Helicobacter. Int J Syst Evol Microbiol systematics. Wiley, Chichester, pp 63–84
50:2231–2237 Hase S, Rietschel ET (1976) Isolation and analysis of the lipid A backbone. Lipid
Diancourt L, Passet V, Nemec A, Dijkshoorn L, Brisse S (2010) The population A structure of lipopolysaccharides from various bacterial groups. Eur
structure of Acinetobacter baumannii: expanding multiresistant clones from J Biochem 63:101–107
an ancestral susceptible genetic pool. PLoS One 5:e10034 Helander IM, Haikara A (1995) Cellular fatty acyl and alkenyl residues in
Dobson G, Minnikin DE, Minnikin SM, Parlett JH, Goodfellow M (1985) Sys- Megasphaera and Pectinatus species: contrasting profiles and detection of
tematic analysis of complex mycobacterial lipids. In: Goodfellow M, beer spoilage. Microbiol 141:1131–1137
Minnikin DE (eds) Chemical methods in bacterial systematics, vol 20, SAB Hill LR (1974) Theoretical aspects of numerical identification. Int J Syst Bacteriol
technical series. Academic, London, pp 237–266 24:494–499
Edwards JG (1978) Computer-assisted identification of unknown. In: Sharpe AN, Hirao T, Sato M, Shirahata AY, Kamio Y (2000) Covalent linkage of polyamines to
Clark DS (eds) Mechanizing microbiology. Charles C. Thomas, Springfield, peptidoglycan in Anaerovibrio lipolytica. J Bacteriol 182:1154–1157
pp 280–292 Holmes B, Hill LR (1985) Computers in diagnostic bacteriology, including
Eraso JM, Kaplan S (2009) Regulation of gene expression by PrrA in Rhodobacter identification. In: Goodfellow M, Jones D, Priest FG (eds) Computer assisted
sphaeroides 2.4.1: role of polyamines and DNA topology. J Bacteriol bacterial systematics. Academic, London, pp 265–287
191(13):4341–4352 International Committee on Systematic Bacteriology Subcommittee on the Tax-
Faller AM, Schleifer KH (1981) Effect of growth phase and oxygen supply on onomy of Mycoplasmatales (1972) Proposal for minimal standards for
cytochrome composition and morphology of Arthrobacter crystallopoietes. descriptions of new species of the order Mycoplasmatales. Int J Syst Bacteriol
Curr Microbiol 6:253–258 22:184–188
Faller AM, Götz F, Schleifer KH (1980) Cytochrome patterns of staphylococci and Johnson JL (1973) Use of nucleic-acid homologies in the taxonomy of anaerobic
micrococci and their taxonomic implications Zentralbl. Bakteriol Mikrobiol bacteria. Int J Syst Bacteriol 23:308–315
Hyg Abt I Orig C1:26–39 Johnson JL (1984) Nucleic acids in bacterial classification. In: Krieg NR, Holt JG
Felis GE, Dellaglio F (2007) On species descriptions based on a single strain: (eds) Bergey’s manual of systematic bacteriology, vol 1. Williams & Wilkins,
proposal to introduce the status species proponenda (sp. pr.). Int J Syst Baltimore, pp 8–11
Evol Microbiol 57:2185–2187 Johnson JL, Ordal EJ (1968) Deoxyribonucleic acid homology in bacterial tax-
Feuerstein BG, Williams LD, Basu HS, Marton LJ (1991) Implications onomy. The effect of incubation temperature and reaction specificity.
and concepts of polyamine-nucleic acid interactions. J Cell Biochem J Bacteriol 95:893–900
46(1):37–47 Jolley KA, Maiden MC (2010) BIGSdb: scalable analysis of bacterial genome
Fischer W (1988) Physiology of lipoteichoic acids in bacteria. Adv Microb Physiol variation at the population level. BMC Bioinformatics 11:595
29:233–302 Jolley KA, Bliss CM, Bennett JS, Bratcher HB, Brehony C, Colles FM et al
Fox GE, Pechman KR, Woese CR (1977) Comparative cataloging of 16S ribo- (2012) Ribosomal multilocus sequence typing: universal characterization
somal ribonucleic acid: molecular approach to prokaryotic systematics. Int of bacteria from domain to strain. Microbiology 158:1005–1015
J Syst Bacteriol 27:44–57 Jones CW, Poole RK (1985) The analysis of cytochromes. In: Gottschalk G (ed)
Fox GE, Wisotzkey JD, Jurtshuk P Jr (1992) How close is close: 16S rRNA Methods in microbiology, vol 18. Academic, London, pp 285–328
sequence identity may not be sufficient to guarantee species identity. Int Kamio Y, Nakamura K (1987) Putrescine and cadaverine are constituents of
J Syst Bacteriol 42:166–170 peptidoglycan in Veillonella alcalescens and Veillonella parvula. J Bacteriol
Freney J, Kloos WE, Hajek V, Webster JA, Bes M, Brun Y, Vernozy-Rozand C 169:2881–2884
(1999) Recommended minimal standards for description of new staphylo- Kamio Y, Itoh Y, Terawaki Y (1981a) Chemical structure of peptidoglycan in
coccal species. Int J Syst Bacteriol 49:489–502 Selenomonas ruminantium: cadaverine links covalently to the D-glutamic
Fujita Y, Naka N, Doi T, Yano I (2005a) Direct molecular mass determination of acid residue of peptidoglycan. J Bacteriol 146:49–53
trehalose monomycolate from 11 species of mycobacteria by MALDI-TOF Kamio Y, Itoh Y, Terawaki Y, Kusano T (1981b) Cadaverine is covalently linked to
mass spectrometry. Microbiology 151:1443–1452 peptidoglycan in Selenomonas ruminantium. J Bacteriol 145:122–128
Fujita Y, Naka N, McNeil MR, Yano I (2005b) Intact molecular characterization Kämpfer P (2010) The importance of phenotype for bacterial systematics. Bull
of cord factor (trehalose 6,60 -dimycolate) from nine species of BISMiS 1(1):7–15
mycobacteria by MALDI-TOF mass spectrometry. Microbiology 151: Kämpfer P (2012) Systematics of prokaryotes: the state of the art. Antonie Van
3403–3416 Leeuwenhoek 101(1):3–11
Kämpfer P, Glaeser SP (2012) Prokaryotic taxonomy in the sequencing era—the Ludwig W (2010) Molecular phylogeny of microorganisms: is rRNA still a useful
polyphasic approach revisited. Environ Microbiol 14:291–317 marker? In: Oren A, Papke RT (eds) Molecular phylogeny of microorgan-
Kandler O, Hippe H (1977) Lack of peptidoglycan in the cell walls of isms. Caister Academic Press, Norfolk
Methanosarcina barkeri. Arch Microbiol 113:57–60 Ludwig W, Klenk H-P (2001) Overview: a phylogenetic backbone and taxonomic
Kaneko T (1979) Correlative similarity coefficient: new criterion for forming framework of prokaryotes. In: Garrity GM (ed) Bergey’s manual of system-
dendrograms. Int J Syst Bacteriol 29:188–193 atic bacteriology, 2nd edn. Springer, New York, pp 49–65
Karr DE, Bibb WF, Moss CW (1982) Isoprenoid quinones of the genus Legionella. Maiden MC, Bygraves JA, Feil E, Morelli G, Russel JE, Urwin R et al
J Clin Microbiol 15:1044–1048 (1998) Multilocus sequence typing: a portable approach to the identification
Kellogg ST (1979) MICRID: a computer-assisted microbial identification system. of clones within populations of pathogenic microorganisms. Proc Natl Acad
Appl Environ Microbiol 38:559–563 Sci USA 95:3140–3145
Kim O-S, Cho Y-J, Lee K, Yoon S-H, Kim M, Na H et al (2012) Introducing Martinez-Murcia AJ, Collins MD (1990) A phylogenetic analysis of the genus
EzTaxon-e: a prokaryotic 16S rRNA gene sequence database with Leuconostoc based on reverse transcriptase sequencing of 16S rRNA. FEMS
phylotypes that represent uncultured species. Int J Syst Evol Microbiol Microbiol Lett 70:73–84
62:716–721 Martinez-Murcia AJ, Benlloch S, Collins MD (1992) Phylogenetic interrelation-
Klenk H-P, Göker M (2010) En route to a genome-based classification of archaea ships of members of the genera Aeromonas and Plesiomonas as determined by
and bacteria? Syst Appl Microbiol 33(4):175–182 16S ribosomal DNA sequencing: lack of congruence with results of
Koeppel A, Perry EB, Sikorski J, Krizanc D, Warner A, Ward DM et al (2008) Iden- DNA-DNA hybridisations. Int J Syst Bacteriol 42:412–421
tifying the fundamental units of bacterial diversity: a paradigm shift to McCarthy BJ, Bolton ET (1963) An approach to the measurement of genetic
incorporate ecology into bacterial systematics. Proc Natl Acad Sci USA relatedness among organisms. Proc Natl Acad Sci USA 50:156–164
105:2504–2509 Miller LT (1982) A single derivatization method for bacterial fatty acid methyl
König H (1994) Analysis of archaeal cell envelopes. In: Goodfellow M, O’Donnell esters including hydroxy acids. J Clin Microbiol 16:584–586
AG (eds) Chemical methods in prokaryotic systematics. Wiley, Chichester, Moore ERB, Rosselló-Móra R (2011) MALDI-TOF MS: a return to phenotyping
pp 85–119 in microbial identification? Syst Appl Microbiol 34:1
König H, Kralik R, Kandler O (1982) Structure and modifications of Moore LVH, Bourne DM, Moore WEC (1994) Comparative distribution and
pseudomurein in Methanobacteriales. Zbl Bakt Hyg I Abt Orig C 3:179–191 taxonomic value of cellular fatty acids in thirty-three genera of anaerobic
König H, Kandler O, Jensen M, Rietschel ET (1983) The primary structure of the Gram-negative bacilli. Int J Syst Bacteriol 44:338–347
glycan moiety of the pseudomurein from Methanobacterium thermoauto- Moore ERB, Mihaylova SA, Vandamme P, Krichevsky MI, Dijkshoorn L (2010)
trophicum. Hoppe-Seyler’s Z Physiol Chem 364:627–636 Microbial systematics and taxonomy: relevance for a microbial commons.
Konstantinidis KT, Tiedje JM (2005) Genomic insights into the species definition Res Microbiol 161:430–438
for prokaryotes. Proc Natl Acad Sci USA 102:2567–2572 Müller K-D, Schmid EN, Kroppenstedt RM (1998) Improved identification of
Konstantinidis KT, Ramette A, Tiedje JM (2006) The bacterial species definition mycobacteria by using the microbial identification system in combination
in the genomic era. Philos Trans R Soc Lond B Biol Sci 361:1929–1940 with additional trimethylsulfonium hydroxide pyrolysis. J Clin Microbiol
Krieg NR, Padgett PJ (2011) Phenotypic and physiological characterization 36:2477–2480
methods. In: Rainey F, Oren A (eds) Taxonomy of prokaryotes, vol 38, Murray RGE, Schleifer KH (1994) Taxonomic notes: a proposal for recording the
Methods in microbiology. Elsevier, Amsterdam, pp 15–60 properties of putative taxa of prokaryotes. Int J Syst Bacteriol 44:174–176
Kroppenstedt RM, Goodfellow M (1991) The family Thermomonosporaceae. In: Murray RGE, Stackebrandt E (1995) Taxonomic note: implementation of the
Balows A, Trüper HG, Dworkin M, Harder W, Schleifer KH (eds) The provisional status Candidatus for incompletely described prokaryotes. Int
prokaryotes, 2nd edn. Springer, New York, pp 1085–1114 J Syst Bacteriol 45:186–187
Kroppenstedt RM, Stackebrandt E, Goodfellow M (1990) Taxonomic revision of Murray RGE, Brenner DJ, Colwell RR, De Vos P, Goodfellow M, Grimont PAD,
the actinomycete genera Actinomadura and Microtetraspora. Syst Appl Pfennig N, Stackebrandt E, Zavarzin GA (1990) Report of the ad hoc
Microbiol 13:148–160 committee on approaches to taxonomy within the Proteobacteria. Int
Labeda DP (2000) International Committee on Systematic Bacteriology. IXth J Syst Bacteriol 40:213–215
international (IUMS) congress of bacteriology and applied microbiology. Naka T, Fujiwaraa N, Yanoc I, Maedaa S, Doed M, Minaminob M, Ikedab N,
Minutes of the meetings, 14 an 17 August 1999, Sydney, Australia. Int J Syst Katob Y, Watabee K, Kumazawaf Y, Tomiyasug I, Kobayashia K (2003)
Evol Microbiol 50:2245–2247 Structural analysis of sphingophospholipids derived from Sphingobacterium
Langworthy TA, Holzer G, Zeikus JG, Tornabene TG (1983) Iso- and anteiso- spiritivorum, the type species of genus Sphingobacterium. Biochim Biophys
branched glycerol diethers of the thermophilic anaerobes Thermodesulfo- Acta 1635:83–92
bacterium commune. Syst Appl Microbiol 4:1–17 Neuhaus FC, Baddiley J (2003) A continuum of anionic charge: structures and
Lapage SP, Bascomb B, Willcox WR, Curtis MA (1970) Computer identification functions of D-alanyl-teichoic acids in Gram-positive bacteria. Microbiol
of bacteria. In: Baillie A, Gilbert RJ (eds) Automation, mechanization, and Mol Biol Rev 67:686–723
data handling in microbiology. Academic, London, pp 1–22 Oelze J (1985) Analysis of bacteriochlorophylls. In: Gottschalk G (ed) Methods in
Lapage SP, Bascomb B, Willcox WR, Curtis MA (1973) Identification of bacteria microbiology, vol 18. Academic, London, pp 257–284
by computer: general aspects and perspectives. J Gen Microbiol 77:273–299 Oren A, Ventosa A, Grant WD (1997) Proposed minimal standards for descrip-
Lapage SP, Sneath PHA, Lessel EF, Skerman VBD, Seeliger HPR, Clark WA tion of new taxa in the order Halobacteriales. Int J Syst Evol Microbiol
(1992) International code of nomenclature of bacteria (1990 revision). 47:233–238
American Society for Microbiology, Washington, DC Ourisson G, Rohmer M, Poralla K (1987) Prokaryotic hopanoides and other
Lechevalier MP, Lechevalier H (1970) Chemical composition as a criterion in the polyterpenoidsterol surrogates. Annu Rev Microbiol 41:301–333
classification of aerobic actinomycetes. Int J Syst Bacteriol 20:435–443 Pace B, Campbell LL (1971) Homology of ribosomal ribonucleic acid of diverse
Lévy-Frébault VV, Portaels F (1992) Proposed minimal standards for the genus bacterial species with Escherichia coli and Bacillus stearothermophilus.
Mycobacterium and for description of new slowly growing Mycobacterium J Bacteriol 107:543–547
species. Int J Syst Bacteriol 42:315–323 Palleroni NJ, Kunisawa R, Contopoulou R, Doudoroff M (1973) Nucleic acid
Lienau EK, DeSalle R (2009) Evidence, content and corroboration and the tree of homologies in the genus Pseudomonas. Int J Syst Bacteriol 23:333–339
life. Acta Biotheor 57:187–199 Peplies J, Kottmann R, Ludwig W, Glöckner F-O (2008) A standard operating
Logan NA, Berge O, Bishop AH, Busse H-J, De Vos P, Fritze D, Heyndrickx M, procedure for phylogenetic inference (SOPPI) using (rRNA) marker genes.
Kämpfer P, Rabinovitch L, Salkinoja-Salonen MS, Seldin L, Ventosa A (2009) Syst Appl Microbiol 31:251–257
Proposed minimal standards for describing new taxa of aerobic, endospore- Pfennig N, Wagener S (1986) An improved method of preparing wet mounts for
forming bacteria. Int J Syst Evol Microbiol 59:2114–2121 photomicrographs of microorganisms. J Microbiol Methods 4:303–306
Rahman O, Dover LG, Sutcliffe IC (2009a) Lipoteichoic acid biosynthesis: two Sneath PHA (1972) Computer taxonomy. In: Norris JR, Ribbons DW (eds)
steps forwards, one step sideways? Trends Microbiol 17:219–225 Methods in microbiology, vol 7A. Academic, London, pp 29–98
Rahman O, Pfitzenmaier M, Pester O, Morath S, Cummings SP, Hartung T, Sneath PHA (1977) The maintenance of large numbers of strains of microorgan-
Sutcliffe IC (2009b) Macroamphiphilic components of thermophilic actino- isms, and the implications for culture collections. FEMS Microbiol Lett
mycetes: identification of lipoteichoic acid in Thermobifida fusca. J Bacteriol 1:333–334
191:152–160 Sneath PHA (1979a) BASIC program for a significance test for two clusters in
Rainey FA, Klatte S, Kroppenstedt RM, Stackebrandt E (1995) Dietzia, new genus Euclidian space as measured by their overlap. Comput Geosci 5:143–155
including Dietzia maris comb. nov., formerly Rhodococcus maris. Int J Syst Sneath PHA (1979b) BASIC program for identification of an unknown with
Bacteriol 45:32–36 presence-absence data against an identification matrix of percent positive
Rainey FA, Oren A (2012) Taxonomy of prokaryotes – introduction. Methods characters. Comput Geosci 5:195–213
Microbiol 38:1–5 Sneath PHA (1979c) BASIC program for character separation indices from an
Ratledge C, Wilkinson SG (1988) Microbial lipids, vol 1. Academic, London identification matrix of percent positive characters. Comput Geosci
Rhodes ME (1965) Flagellation as a criterion for the classification of bacteria. 5:349–357
Bacteriol Rev 29:442–465 Sneath PHA (1980a) BASIC program for the most diagnostic properties of groups
Richter M, Rosselló-Móra R (2009) Shifting the genomic gold standard from an identification matrix of percent positive characters. Comput Geosci
for the prokaryotic species definition. Proc Natl Acad Sci USA 6:21–26
106:19126–19131 Sneath PHA (1980b) BASIC program for determining the best identification
Rosselló-Móra R (2006) DNA-DNA reassociation methods applied to microbial scores possible for the most typical example when compared with an iden-
taxonomy and their critical evaluation. In: Stackebrandt E (ed) Molecular tification matrix of percent positive characters. Comput Geosci 6:27–34
identification, systematics, and population structure of prokaryotes. Sneath PHA (1980c) BASIC program for determining overlap between groups in
Springer, Heidelberg/Berlin, pp 23–50 an identification matrix of percent positive characters. Comput Geosci
Rosselló-Móra R (2012) Towards a taxonomy of bacteria and archaea based on 6:267–278
interactive and cumulative data repositories. Environ Microbiol 14:318–334 Sneath PHA (1989) Analysis and interpretation of sequence data for bacterial
Rosselló-Mora R, Amann R (2001) The species concept for prokaryotes. FEMS systematics—the view of a numerical taxonomist. Syst Appl Microbiol
Microbiol Rev 25:39–67 12:15–31
Rütters H, Sass H, Cypionka H, Rullkötter J (2001) Monalkylether phospholipids Sneath PHA (2001) Bacterial nomenclature. In: Garrity GM (ed) Bergey’s manual
in the sulfate-reducing bacteria Desulfosarcina vaiabilis and Desulforhabdus of systematic bacteriology, vol 1, 2nd edn. Springer, New York, pp 83–88
aminigenus. Arch Microbiol 176:435–442 Sneath PHA (2005) The preparation of the approved lists of bacterial names. Int
Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, J Syst Evol Microbiol 55:2247–2249
Erlich HA (1988) Primer-directed enzymatic amplification of DNA with Sneath PHA, Sokal RR (1973) Numerical taxonomy, the principles and practice of
a thermostable DNA polymerase. Science 239:487–491 numerical classification. Freeman, San Francisco
Scherer P, Kneifel H (1983) Distribution of polyamines in methanogenic bacteria. Sokal RR, Sneath PHA (1963) Numerical taxonomy. Freeman, San Francisco
J Bacteriol 154:1315–1322 Soria-Carrasco V, Valens-Vadell M, Peña A, Antón J, Amann R, Castresana J,
Schindler J, Duben J, Lysenko O (1979) Computer-aided numerical identification Rosselló-Móra R (2007) Phylogenetic position of Salinibacter ruber based on
of Gram-negative fermentative rods on a desk-top computer. J Appl concatenated protein alignments. Syst Appl Microbiol 30:171–179
Bacteriol 47:45–51 Stackebrandt E, Ebers J (2006) Taxonomic parameters revisited: tarnished gold
Schleifer K-H (2009) Classification of bacteria and archaea: past, present and standards. Microbiol Today 8:6–9
future. Syst Appl Microbiol 32:533–542 Stackebrandt E, Goebel BM (1994) Taxonomic note: a place for DNA-DNA
Schleifer KH, Kandler O (1972) Peptidoglycan types of bacterial cell walls and reassociation and 16S rRNA sequence analysis in the present species defini-
their taxonomic implications. Bacteriol Rev 36:407–477 tion in bacteriology. Int J Syst Bacteriol 44:846–849
Schleifer KH, Ludwig W (1989) Phylogenetic relationships among bacteria. In: Stackebrandt E, Woese CR (1981) Towards a phylogeny of the Actinomycetes and
Fernholm B, Bremer K, Jörnvall H (eds) The hierarchy of life. Elsevier, related organism. Curr Microbiol 5:197–202
Amsterdam, pp 103–117 Stackebrandt E, Frederiksen W, Garrity GM, Grimont PAD, Kämpfer P, Maiden
Schleifer KH, Hammes WP, Kandler O (1976) Effect of endogenous and exoge- MCJ et al (2002) Report of the ad hoc committee for the re-evaluation
nous factors on the primary structures of bacterial peptidoglycan. Adv of the species definition in bacteriology. Int J Syst Evol Microbiol 52:
Microb Physiol 13:245–292 1043–1047
Schleifer K-H, Steber J, Mayer H (1982) Chemical composition and struc- Staehelin LA, Golecki JR, Fuller RC, Drews G (1978) Visualization of the supra-
ture of the cell wall of Halococcus morrhuae. Zbl Bakt Hyg I Abt Orig molecular architecture of chlorosomes (chlorobium type vesicles) in freeze-
C 3:171–178 fractured cells of Chloroflexus aurantiacus. Arch Microbiol 119:269–277
Schumann P, Kämpfer P, Busse H-J, Evtushenko LI, For the Subcommittee on the Staley JT, Krieg NR (1984) Classification of prokaryotic organisms: an overview.
Taxonomy of the suborder Micrococcineae of the International Committee In: Murray RG, Brenner DJ, Bryant MP, Holt JG, Krieg NR, Moulder JW,
on Systematics of Prokaryotes (2009) Proposed minimal standards for Pfennig N, Sneath PHA, Staley JT (eds) Bergey’s manual of systematic
describing new genera and species of the suborder Micrococcineae. Int bacteriology, vol 1. Williams and Wilkins, Baltimore, pp 1–4
J Syst Bacteriol 59:1823–1849 Stanier RY, Van Niel CB (1962) The concept of a bacterium. Arch Microbiol
Shively JM (1974) Inclusion bodies of prokaryotes. Annu Rev Microbiol 42:7–35
28:167–187 Stanier RY, Ingraham JL, Wheelis ML, Paintes PR (1986) The microbial world,
Simpson GG (1961) Principles of animal taxonomy. Columbia University Press, 5th edn. Prentice-Hall, Englewood Cliffs
New York Tindall BJ (1999) Misunderstanding the bacteriological code. Int J Syst Bacteriol
Skerman VBD (1967) A guide to the identification of the genera of bacteria, 49:1313–1316
2nd edn. Williams & Wilkins, Baltimore Tindall BJ (2005) Respiratory lipoquinones as biomarkers. In: Akkermans A, de
Skerman VBD, McGowan V, Sneath PHA (1980) Approved lists of bacterial Bruijn F, van Elsas D (eds) Molecular microbial ecology manual, 2nd edn.
names. Int J Syst Bacteriol 30:225–420 Kluwer, Dordrecht, Section 4.1.5, Supplement 1
Skerman VDB, McGowan V, Sneath PHA (1989) The approved lists of bacterial Tindall BJ (2008) Confirmation of deposit, but confirmation of what? Int J Syst
names. Int J Syst Bacteriol 30:225–420 Evol Microbiol 58:1785–1787
Sneath PHA (1971) Theoretical aspects of microbiological taxonomy. In: Pérez- Tindall BJ, Garrity GM (2008) Proposals to clarify how type strains are deposited
Miravete A, Peláez D (eds) Recent advances in microbiology. Asociacióon and made available to the scientific community for the purpose of systematic
Mexicana de Microbiologı́a, Mexico City, pp 581–586 research. Int J Syst Evol Microbiol 58:1987–1990
Tindall BJ, Kämpfer P, Euzéby JP, Oren A (2006) Valid publication of names of Wayne LG, Brenner DJ, Colwell RR, Grimont PAD, Kandler O, Krichevsky MI,
prokaryotes according to the rules of nomenclature: past history and current Moore LH, Moore WEC, Murray RGE, Stackebrandt E, Starr MP, Trüper HG
practice. Int J Syst Evol Microbiol 56:2715–2720 (1987) Report of the ad hoc committee on reconciliation of approaches to
Tindall BJ, Sikorski J, Smibert RM, Krieg NR (2007) Phenotypic characteriza- bacterial systematics. Int J Syst Bacteriol 37:463–464
tion and the principles of comparative systematics. In: Reddy CA, Weckesser J, Mayer H (1988) Different lipid A types in lipopolysaccharides of
Beveridge TJ, Breznak JA, Marzluf G, Schmidt TM, Snyder LR (eds) phototrophic and related non-phototrophic bacteria. FEMS Microbiol Rev
Methods for general and molecular microbiology. ASM Press, Washington, 54:1431–1454
DC, pp 330–393 Welker M, Moore ER (2011) Applications of whole-cell matrix-assisted laser-
Tindall BJ, Rosselló-Mora R, Busse H-J, Ludwig W, Kämpfer P (2010) Notes on desorption/ionization time-of-flight mass spectrometry in systematic
the characterization of prokaryote strains for taxonomic purposes. Int J Syst microbiology. Syst Appl Microbiol 34:2–11
Evol Microbiol 60:249–266 Willumsen P, Karlson U, Stackebrandt E, Kroppenstedt RM (2001) Mycobacte-
Torkko P, Katila M-L, Kontro M (2003) Gas-chromatographic lipid profiles in rium frederiksbergense sp. nov., a novel polycyclic aromatic hydrocarbon-
identification of currently known slowly growing environmental degrading Mycobacterium species. Int J Syst Evol Microbiol 51:1715–1722
mycobacteria. J Med Microbiol 52:315–323 Wisplinghoff H, Hippler C, Bartual SG, Haefs C, Stefanik D, Higgins PG,
Tornabene TG (1985) Lipid analysis and the relationship to chemotaxonomy. In: Seifert H (2008) Molecular epidemiology of clinical Acinetobacter baumannii
Gottschalk G (ed) Methods in microbiology, vol 18. Academic, London, and Acinetobacter genomic species 13TU isolates using a multilocus sequenc-
pp 209–234 ing typing scheme. Clin Microbiol Infect 14:708–715
Trüper HG, Euzéby JP (2009) International code of nomenclature of pro- Woese CR (1987) Bacterial evolution. Microbiol Rev 51:221–271
karyotes. Appendix 9: orthography. Int J Syst Evol Microbiol 59(Pt 8): Woese CR, Fox G (1977) Phylogenetic structure of the prokaryotic domain: the
2107–2113 primary kingdoms. Proc Natl Acad Sci USA 74:5088–5090
Trüper HG, Schleifer KH (2006) Prokaryote characterization and identification. Woese CR, Kandler O, Wheelis ML (1990) Towards a natural system of organisms:
In: Dworkin M, Falkow S, Rosenberg E, Schleifer KH, Stackebrandt E (eds) proposla for the domains: archaea. Bacteria and Eucarya. Proc Natl Acad Sci
The prokaryotes: a handbook on the biology of bacteria, vol 1, 3rd edn. USA 87:4576–4579
Springer, New York, pp 58–79 Wolf Y, Rogozin I, Grishin N, Tatusov R, Koonin E (2001) Genome trees
Ursing J, Rosselló-Mora RA, Garcia-Valdez E, Lalucat J (1995) Taxonomic note: constructed using five different approaches suggest new major bacterial
a pragmatic approach to the nomenclature of phenotypically similar geno- clades. BMC Evol Biol 1:8
mic groups. Int J Syst Bacteriol 45:604 Yarza P, Richter M, Peplies J, Euzeby J, Amann R, Schleifer KH, Ludwig W,
Vandamme P, Pot B, Gillis M, De Vos P, Kersters K, Swings J (1996) Polyphasic Glöckner FO, Rossello-Mora R (2008) The all-species living tree project:
taxonomy, a consensus approach to bacterial systematics. Microbiol Rev A 16S rRNA-based phylogenetic tree of all sequenced type strains. Syst Appl
60:407–438 Microbiol 31:241–250
7 Principles of Enrichment, Isolation,
Cultivation, and Preservation of
Prokaryotes
Jörg Overmann
Leibniz-Institut DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH,
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 The Viable but Nonculturable State . . . . . . . . . . . . . . . . . 177

Lysogenic Phages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Historical Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 Low Nutrient Concentrations, ‘‘Oligotrophic Bacteria,’’
and ‘‘Ultramicrobacteria’’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Prokaryotic Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 Growth on Multiple Substrates . . . . . . . . . . . . . . . . . . . . . . . . . . 178
Parameters of Prokaryotic Growth . . . . . . . . . . . . . . . . . . . . . . 154 Effect of Inhomogeneities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Modes of Prokaryotic Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 Interactions with Other Microorganisms . . . . . . . . . . . . . . . 180
Requirements for Prokaryotic Growth . . . . . . . . . . . . . . . . . . 156 Neutralism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
General Composition of the Prokaryotic Cell . . . . . . . 156 Competition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
Growth Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 Amensalism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Predation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Osmolarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Commensalism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 Synergism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Hydrostatic Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 Syntrophic Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Symbiosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Treatment of Growth Media and Equipment . . . . . . . . . . . . . . 168
Types of Culture Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 Enrichment Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Media Preparation and Sterilization . . . . . . . . . . . . . . . . . . . . 169
Handling of Glassware and Equipment . . . . . . . . . . . . . . . . . 171 Isolation of Prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Removal and Exclusion of O2, Cultivation of
Anaerobes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Alternative and Novel Concepts for Cultivation . . . . . . . . . . 187
Ionic Composition of the Growth Media . . . . . . . . . . . . . . . 187
Culture Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Exploring Novel Types of Metabolism . . . . . . . . . . . . . . . . . . 187
Batch Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Cultivation-Independent Physiological Testing
Fed-Batch Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 of Target Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
Recycling Fermenter, Dialysis Culture, Retentostat, Types and Concentrations of Growth Substrates . . . . . . . 188
and Recyclostat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 Cocultivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
Chemostat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 Chemical Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
Redox-Controlled Sulfidostat . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Solid Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
Auxostat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Conservation of Prokaryotic Cultures . . . . . . . . . . . . . . . . . . . . . 190
Possible Reasons for ‘‘Nonculturability’’ . . . . . . . . . . . . . . . . . . 175 Maintenance of Working Cultures . . . . . . . . . . . . . . . . . . . . . . 190
Physiological State of Prokaryotic Cells . . . . . . . . . . . . . . . . . 176 Preservation of Stock Cultures . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Starvation Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 Spore-Forming Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Presence of Dead Cells, Prevention, and Reversal Lyophilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
of Cellular Damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 Freezing and Ultrafreezing . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Dormancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
Substrate-Accelerated Death . . . . . . . . . . . . . . . . . . . . . . . . . 176 Culture Collections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
150 7 Principles of Enrichment, Isolation, Cultivation, and Preservation of Prokaryotes
Introduction strain with a 16S rRNA sequence corresponding to the dominant

strains in the natural community (Felske et al. 1999). Entire
Currently, a total of 9,409 prokaryotic species are recognized (as bacterial phyla such as the Acidobacteria that are highly abun-
of January 2012, validly published names not including dant in the soil environment or the OP10 phylum (now
homotypic and heterotypic synonyms, comb. nov. and nomina Armatimonadetes) have resisted cultivation for a long time. At
nova; DSMZ 2012; Euzéby 2012). By comparison, the number of the same time, culture-independent methods have shown that
small subunit ribosomal RNA (SSU rRNA) gene sequences uncultured phylotypes are physiologically active in situ (Bernard
deposited in public databases keeps increasing exponentially et al. 2000) and that active cells can constitute a high percentage
(Pruesse et al. 2007; Yarza et al. 2008) and surmounted the of total cell counts (Ouverney and Fuhrman 1999; Cottrell and
species numbers already some 15 years ago (> Fig. 7.1). Mean- Kirchman 2000; Janssen et al. 2002).
while, a total of 2,492,653 sequences are available of which The not-yet-cultured prokaryotes may exhibit a different
2,282,670 are prokaryotic whereas only 33,842 originate from physiology that does not match established cultivation
cultured strains (SILVA 2012). In line with these cumulative methods (e.g., ultraoligotrophic growth characteristics, see sec-
data, culture-independent analyses of DNA reassociation kinet- tion > ‘‘Low Nutrient Concentrations, ‘‘Oligotrophic Bacteria,’’
ics and of 16S rRNA gene sequences in individual environmental and ‘‘Ultramicrobacteria’’’’). Indeed, key reactions in biogeo-
samples also indicate that prokaryotic diversity is poorly chemical cycles often appear to be mediated not by the fre-
represented by the species cultivated so far. Thus, estimates of quently isolated prokaryotes but rather by other prokaryotes
bacterial species numbers in just one type of soil reached values that are phylogenetically unrelated and that have escaped culti-
of up to 53,000 (Sandaa et al. 1999; Roesch et al. 2007). Further- vation for a long time or until now remained uncultured. Exam-
more, molecular investigations of 16S rRNA gene sequences in ples include members of the Nitrosospira group which (instead
natural bacterial assemblages typically yielded many more of Nitrosomonas spp.) catalyze nitrification in at least some
sequence types than those recovered by cultivation-based environments, most notably in wastewater treatment plants as
approaches (Fuhrman et al. 1992; Ward et al. 1992; Barns et al. the world’s largest biotechnological process (Hastings et al. 1998;
1994; DeLong et al. 1994; Hiorns et al. 1997; Kuske et al. 1997; Schramm et al. 1999). Other examples are Achromatium
Ludwig et al. 1997; Suzuki et al. 1997b; Gich et al. 2001; Béjà et al. spp. and coccoid magnetotactic bacteria participating in the
2002; Roesch et al. 2007). In light of these findings, the earlier sulfur cycle (Spring et al. 1993), archaea which mediate anaero-
estimates of the fraction of already cultured bacterial species bic methane consumption (Hinrichs et al. 1999), autotrophic
of 12–20% (Wayne et al. 1987; Bull et al. 1992) or even the planktonic marine Crenarchaeota that oxidize ammonia and
commonly cited estimate of 1% appears to be far too exhibit an extraordinary high affinity to ammonia (Hallam
optimistic. Based on recent estimates of total bacterial species et al. 2006; Martens-Habbena et al. 2009) and novel type II
numbers (107–109; Dykhuizen 1998; Curtis et al. 2002), the methanotrophic bacteria oxidizing methane at atmospheric
value more likely ranges between 0.1% and 0.001% and may concentrations (Holmes et al. 1999; Roslev and Iversen 1999).
be even lower (compare the higher estimates of bacterial species Some environments harbor whole groups of entirely unknown
numbers in Sogin et al. 2006; Harwood and Buckley 2008). bacteria and archaea, which most likely are important for the
Concomitantly to the explosive increase of 16S rRNA gene biogeochemical transformations (Hugenholtz et al. 1998). These
sequence numbers, the number of full genome sequences in include for instance members of the green filamentous bacteria
Genbank has approached 2,000 (NCBI 2012). Together, com- (Coolen et al. 2002) or freshwater actinobacteria (Glöckner et al.
plete and incomplete genome sequencing projects listed in the 2000) that appear to utilize N-acetyl-D-glucosamine as com-
Genome Online Database (GOLD 2012) amount to 11,221, of mensals of aquatic chitin degraders (Beier and Bertilsson
which 80% are prokaryotic. The increase in new genome 2011). In addition, culture-independent analyses of large
sequencing projects now has reached a pace (>2,000 year1) genome fragments retrieved from natural samples have uncov-
that clearly surpasses the current rate of new species descriptions ered bacteria present in the natural environment that have
(622 species year1, from 2005 on) (> Fig. 7.1). This implies that a previously unrecognized physiology (Béjà et al. 2000). Only
the rate of cultivation of novel bacterial species has become in a few instances could numerically important bacterial species
insufficient to tap the currently available potential of genome also be isolated (Ferris et al. 1996; Kalmbach et al. 1997).
sequencing facilities. Obviously, the future discovery of novel With the aid of modern molecular methods, some informa-
metabolic pathways, regulation mechanisms, novel metabolic tion on the physiological capabilities of ‘‘not-yet-cultured pro-
capacities, but also novel antibiotics (Jensen et al. 2007) will karyotes’’ can be obtained even down to the single-cell level
largely depend on an improved access to novel types of (Ouverney and Fuhrman 2000; Wagner et al. 2006; Beier and
prokaryotes. Bertilsson 2011). Though culture-independent techniques have
In most cases, the 16S rRNA gene inventories of natural improved, they still provide only limited insights into the phys-
bacterial communities differ from the sequences of strains iso- iology of prokaryotes. Therefore, the role of not-yet-cultured
lated from the same or similar samples (e.g., Hiorns et al. 1997; prokaryotes in the environment cannot be fully appreciated
Suzuki et al. 1997; Hugenholtz et al. 1998; Bernard et al. 2000). until these microorganisms are available for detailed physiolog-
Even a large cultivation campaign during which 659 bacterial ical and molecular studies. Until now, additional and at the same
isolates were obtained from grassland soil did not yield any time innovative novel cultivation efforts often still are
Principles of Enrichment, Isolation, Cultivation, and Preservation of Prokaryotes 7 151
. Fig. 7.1
Increase in validly described species, numbers of small subunit ribosomal RNA sequences (SILVA 2012), and numbers of genome
sequences (GOLD 2012) since the appearance of the approved list of bacterial names in 1980. This graph is an extended version of that in
Overmann (2002b) and Yarza et al. (2008). Species data were obtained from Euzéby (2012) and the Deutsche Sammlung für
Mikroorganismen und Zellkulturen (DSMZ 2012), and include homotypic and heterotypic synonyms, comb. nov. and nomina nova.
Ninety-two percent of the small subunit rRNA sequences are prokaryotic. Genome numbers include complete and incomplete
sequencing projects; about 80% of the genomes are prokaryotic
a precondition for gaining access to a large portion of microbial diversity because of systematic reasons. More importantly,
diversity. In a limited number of cases, abundant phylotypes a closer look at the cultivated fraction of prokaryotic diversity
(Rappe et al. 2002; Gich et al. 2005) and representatives of reveals that, to date, it is heavily biased toward Actinobacteria,
previously largely uncultured (sub)phyla such as Acidobacteria Firmicutes, Alpha- and Gammaproteobacteria, and
(Janssen et al. 2002; Koch et al. 2008; Eichorst et al. 2007), Bacteroidetes (Hugenholtz 2002; Handelsmann 2004)
mesophilic Crenarchaea (Könneke et al. 2005), and members (> Fig. 7.2). About 50% of the currently recognized bacterial
of the OP10 candidate phylum (Tamaki et al. 2011) could be phyla are still poorly covered by cultivated representatives
isolated. Also, some symbiotic forms have become available, (> Fig. 7.2) (Schloss and Handelsman 2004). Furthermore, the
such as the symbionts of phototrophic consortia (Vogl et al. majority of novel isolates is phylogenetically rather closely
2006). related to previously described species (i.e., >95% 16S rRNA
Some of the isolates showed exceptionally high affinity gene sequence similarity; > Fig. 7.2). Of the 472 novel species
toward growth substrates matching substrate concentrations in that were described in the International Journal of Systematic and
the natural environment (Button et al. 1998; Martens-Habbena Evolutionary Microbiology, only 19 exhibited a similarity of
et al. 2009). In most other cases, however, it has remained 90.4% in 16S rRNA gene sequence to the closest cultivated
unclear which physiological traits prevented earlier isolation of relative (> Fig. 7.2) and hence represented novel higher taxa.
these bacteria. In addition, the discrepancy between 16S rRNA Thus, rather than continuing to isolate the easily to obtain and
gene inventories and cultivated phylotypes may simply be the phylogenetically and physiologically similar strains, a major
result of the low number of cultivation attempts, since even effort is required to elucidate the genomic and physiological
conventional cultivation trials continue to yield novel novelties of enigmatic groups of prokaryotes. Given the most
phylotypes of bacteria (Pinhassi et al. 1997; Suzuki et al. 1997). likely vast number of total bacterial species (see above) and the
The advent of 16S rRNA methodology in the mid-1990s per- highly time-consuming nature of cultivation trials, future culti-
mitted a significantly increased rate of discovery of novel species vation-based approaches to microbial diversity need to be
from a mere (144.0 5.4) species year1 until the year 1994 to conducted in a much more targeted manner. Blind sampling of
a meanwhile stable value of (621.9 8.2) species year1 from the environment does not necessarily yield a representative view
the year 2005 on (> Fig. 7.1), although the procedure for species of bacterial diversity and needs to be complemented by focusing
descriptions has been streamlined and standardized consider- on undersampled and geographically disjunct bacterial groups
ably (Kämpfer et al. 2003). (Schloss and Handelsman 2004). Of the 19 novel species men-
Even at this increased rate, however, current cultivation tioned above, 10 were isolated from undersampled environ-
efforts appear far too limited to adequately cover prokaryotic ments like geothermal soils, thermal springs, mud volcanos, oil
. Fig. 7.2
Similarity of 472 novel species described in the International Journal of Systematic and Evolutionary Microbiology, volume 61 (2011) with
the phylogenetically closest cultured relative. Similarity is given as percentage sequence identity of the 16S rRNA gene. Insert depicts
percentages of different (sub)phyla among the novel species
production wells, an artesian well, sulfur mats, or cryptoen- successful enrichment, isolation, and cultivation of prokaryotes
dolithic microbial communities. Nine isolates were recovered critically depend on the choice of appropriate growth media and
in oligotrophic media and three from cocultures with incubation conditions. Our present methods for the enrich-
Escherichia coli or amoeba. ment, isolation, and cultivation still are largely based on con-
In conclusion, the following strategies for an improved cov- cepts developed first and foremost for the isolation of medically
erage of prokaryotic diversity by cultivation appear most important bacteria (see section > ‘‘Historical Perspective’’).
promising: In the subsequent chapters, the requirements of prokaryotic
cells for growth in the laboratory and the different existing
– Development and further improvement of synthetic cultiva-
current methods for their cultivation are discussed.
tion media that mimic the composition and ionic strength of
the natural environment
– Increasing the number of systematic cultivation trials and
Historical Perspective
improving their throughput by development of rapid
screening tools to identify enrichments of phylogenetically
The work of Louis Pasteur (1822–1895) marks the beginning of
deeply branching prokaryotes
scientific microbiology and focused on the issue of spontaneous
– Determination of physiological traits by culture-
generation. The theory of spontaneous generation assumed that
independent methods such as MAR-FISH, SIP, nanoSIMS
mice, frogs, and ‘‘lower’’ forms of life could develop spontane-
(Wagner et al. 2006), metatranscriptomics and
ously in decaying organic matter and mud. Pasteur’s experi-
metaproteomics, and stimulation experiments (Beier and
ments with swan-necked flasks ruled out spontaneous
Bertilsson 2011) or of the physiological potential by
generation and at the same time laid the foundations for aseptic
metagenomic (single-cell) analyses and development of
manipulation and sterilization. Questioning the belief of emi-
targeted cultivation trials based on this information
nent chemists of his time, Pasteur provided the first notion of
– Improved long-term conservation protocols for reliable
the microbial nature of fermentation in milk (Pasteur 1862) and
storage of unaltered bioresources
of anaerobic life. Other studies revealed that fermentation of
– Focusing -omics and physiological studies on selected novel
sugar to alcohol, wine to vinegar, and the putrefaction of meat
isolates that are known to exhibit highly unusual metabolic
all were caused by microorganisms. Each particular kind of
traits
fermentation was accompanied by the development of
If cultivation approaches are not improved further, bacterial a specific type of microorganism.
diversity and evolution might continue to remain largely Until 1884, the bacterial pleomorphism versus monomor-
uncharted, and prokaryotic diversity cannot be exploited in phism controversy was a central issue of microbiology. The
a systematic fashion for biotechnological purposes. The description of rust and smut fungi and their morphological
changes during development had caused confusion among the advanced the concept of chemosynthesis on the basis of his
early bacteriologists, and it was assumed that, depending solely studies of the colorless sulfur bacteria, Thiothrix and Beggiatoa
on growth conditions, one and the same bacterium is able to (Winogradsky 1949), concluding that these aerobic bacteria
appear in different forms (pleomorphic), can cause completely obtain their energy for autotrophic growth by oxidizing reduced
different diseases, and form totally different metabolic products. sulfur compounds (such as sulfides and sulfur) to sulfate.
Of outstanding importance for our current thinking in micro- The work of Beijerinck, with the contributions from
biology is the contribution of Robert Koch (1843–1910), the Winogradsky, led to the development of the enrichment culture
founder of medical bacteriology. In his experimental work, Koch technique. Instead of directly isolating microorganisms from
established a series of criteria, the so-called Koch’s postulates, for nature by exposing nonselective growth media to some environ-
the identification of causative agents of infectious diseases: ment and allowing chance to dictate what grew, Beijerinck
proposed a different approach. Tailoring culture conditions to
1. The microorganism should be constantly present in animals
favor microbes with a particular metabolic activity usually leads
suffering from the disease but should not be present in
to a rapid enrichment of the desired organism, even if its original
healthy individuals.
numbers are very low in the sample (Van Iterson et al. 1940).
2. The microorganism must be cultivated in pure culture out-
One of the early examples of the application of this principle was
side the diseased animal.
the discovery and isolation of Spirillum (Desulfovibrio)
3. Healthy animals infected with these pure cultures must
desulfuricans, described in a preliminary paper in 1894. Besides
display the characteristic disease symptoms.
root nodule bacteria, Lactobacillus species, and others,
4. The microorganism should be reisolated from the experi-
Beijerinck also isolated the first pure cultures of aerobic nitro-
mental animals and shown to be the same as the original.
gen-fixing bacteria from a culture obtained by Winogradsky.
In 1876, Koch isolated the anthrax bacillus from diseased Winogradsky applied enrichment of selective cultures for his
cattle and conclusively demonstrated that the large nonmotile research on sulfur bacteria and nitrifying bacteria and also
bacilli caused the disease. In 1873, Joseph Lister (1827–1912) used this technique to identify nitrogen-fixing bacteria in
had introduced the serial dilution technique to achieve pure soil. In the years to come after Beijerinck, the success of the
cultures. However, this procedure did not yield pure cultures batch enrichment culture technique was demonstrated over and
in the hands of every researcher. Pure culture techniques over again in the work of Kluyver, Van Niel, Stanier, and
employing solid nutrient media containing gelatin were devel- their students and associates (Veldkamp 1965; Van Niel 1967;
oped by Anton de Bary (1831–1888) and Oscar Brefeld (1839– Pfennig 1993). Very early, however, soil microbiologists noted
1926) for the study of fungi (Bull and Slater 1982). Although the a pronounced discrepancy between numbers of colony-forming
latter scientists succeeded in cultivating fungi on solid media, bacteria and the numbers of cells that could be discerned by
their techniques were not suited for the isolation and growth of directly when observing the original samples. This led to the
bacteria for several reasons: various bacteria were found to suggestion that the variety of bacteria in soil was so enormous
liquefy gelatin, on hot days it melted spontaneously, and gela- that separating the different forms was impractical (Russell
tin-based media could not be incubated at the temperature that 1923).
most human pathogens needed to grow in a convenient time At a time when the discovery of a bewildering variety of
span. Starting out with cut surfaces of boiled potatoes and after microbial forms and activities reached its height, Albert J.
adopting hydrated gelatin films as cultivation media, agar- Kluyver (1884–1957) and H.J.L. Donker published a seminal
solidified media (around 1881), covered culture dishes, and paper on the unity in biochemistry (Kluyver and Donker
staining techniques were developed in the laboratory of Robert 1926). This synthetic paper was based on an integration of the
Koch and, in nearly unaltered form, have remained major tools ideas of Carl Neuberg, Heinrich Wieland, Otto Warburg, and
in bacteriology and medical microbiology until today. After A. Harden. According to this concept, cells contained, in addi-
employing agar media, Koch announced his discovery of the tion to a number of hydrolases, various oxidoreductases catalyz-
tubercle bacillus as the causative agent of tuberculosis in 1882, at ing the transfer of hydrogen from one molecule, the hydrogen
the time when this disease had many victims in Europe (H) donor, to another, the H acceptor, along a gradient of
(Groeschel 1982). Two years later, he published the discovery energy. Hydrogen acceptors other than oxygen might also be
of the cholera bacillus. used. The first report on the view that photosynthesis can be
Subsequently, microorganisms were found to be also of considered as a light-dependent reaction in which different sub-
crucial importance for the geochemical cycles. Ferdinand Cohn strates, specific for the different kinds of photosynthetic organ-
was the first to realize the role of microorganisms in the trans- isms, serve as H donors for the reduction of carbon dioxide
formation of organic matter and inorganic substances on earth came from Van Niel (1930). Van Niel (1967) described a general
(Cohn 1872). The most significant contributions to the knowl- formula of photosynthesis. His work was extended greatly by
edge of the various types of microorganisms responsible for N. Pfennig and his students, who grew many of the anaerobic
specific chemical transformations in nature (especially for the photosynthetic bacteria in pure culture (Pfennig 1993).
nitrogen and sulfur cycles) came largely from the laboratories of Early attempts to devise truly anoxic culturing techniques
Sergius Winogradsky (1856–1953) and Martinus Willem included the use of deep-agar shake tubes gassed with hydrogen,
Beijerinck (1851–1931). During the 1880s, Winogradsky the use of pyrogallol to remove oxygen, sealed tubes, and the
isolation of colonies by picking with capillary pipettes. Still, the osmolarity and can be assessed from the increase in bacterial
only anaerobes isolated until 1940 were sporeformers and sev- cell number or in bacterial biomass. This distinction has to be
eral non-spore-forming bacteria of clinical importance, while made since under certain conditions, growth is not completely
identities of the majority of the anaerobic prokaryotes in sedi- balanced, and the increase in biomass is not paralleled by an
ments, soils, and the gastrointestinal tract remained unknown. increase in cell numbers or vice versa. As an example, cells in
This was in part due to the fact that techniques were insufficient the early stationary phase can undergo reductive division
to reach the necessary anaerobic conditions but also to the fact whereby cell numbers continue to increase while an almost
that the media used were not simulating the microbial habitats. constant overall biomass of the culture is maintained (Kolter
Hungate (1950), the first student of Van Niel from the United et al. 1993).
States, developed an anoxic roll tube technique based on various Upon inoculation of bacterial cells into fresh growth
approaches from the older literature and used habitat- medium, a period without growth is frequently observed (the
simulating media in the isolation of cellulolytic rumen bacteria so-called lag phase) during which adaptation of the cells occurs
(Hungate 1966, 1969, 1985). The Hungate technique was by, for example, synthesis of cellular enzymes. Afterward, a phase
very successful and has been widely adapted by microbiologists of exponential growth occurs until cell growth is limited by
from various other fields (dentistry, sewage, and sediment substrate availability and by accumulation of toxic metabolic
microbiology). products or unfavorable pH (stationary phase). The stationary
Many soil bacteria were studied in pure culture after their phase is then typically followed by a phase of prokaryotic bio-
discovery to examine their basic metabolism and metabolic mass reduction due to lysis of cells. If changes in bacterial cell
capacities. Winogradsky emphasized that the conclusions numbers N are followed during the exponential phase, the
drawn from laboratory studies on the behavior of bacteria in specific division rate n and the generation time g are used for
the natural habitat may be misleading because laboratory strains the mathematical description of growth (N0 denoting the initial
might only be artifacts, that is, selected by the chosen nutrient cell number, and t the time of incubation):
medium, temperature, aeration, and agitation. Consequently, he N ¼ N0 2vt ð7:1Þ
already warned against the use of bacterial strains from culture
collections as typical wild-type cells and, together with some of 1
g¼ ð7:2Þ
his students, designed methods to study bacteria in situ, within n
their natural habitats. Winogradsky’s concepts on soil microbi-
ology and on the differentiation between autochthonous and In a growing culture, the concentration of prokaryotic bio-
allochthonous populations in the soil are the basis of relevant, mass X increases autocatalytically (a first-order reaction), that is,
current research. the instantaneous biomass changes dX/dt are proportional to
The majority of physiological investigations have been lim- the biomass present (Lengeler et al. 1999). Here the specific
ited to pure laboratory cultures. According to the statement of growth rate m and the doubling time td are used to describe
Oscar Brefeld, work with impure cultures yields nothing but prokaryotic growth:
nonsense and Penicillium glaucum (Brefeld 1881). However, dX
¼mX ð7:3Þ
Koch’s methods may also have delayed the study of microbial dt
interactions as they occur in complex natural microbial com- and hence
munities, for example, in soil (Winogradsky 1949). In fact,
interactions between microorganisms under natural conditions X ¼ X0 e mt ð7:4Þ
may lead to transformations that are unknown from pure cul- The biomass doubling time is then calculated according to
tures. The shortcomings of pure culture techniques are illus-
ln 2
trated best by the interspecies hydrogen transfer described first td ¼ ð7:5Þ
by M.J. Wolin in 1975, as well as by the inhibition of growth of m
a Micrococcus culture by a chance Penicillium contaminant Whereas g = td, the two parameters m and n are related
(Fleming 1929), which led to the discovery of antibiotics. For according to
an in-depth understanding of microbial physiology, microbial
m ¼ n ln 2 ð7:6Þ
transformations therefore have to be studied also in natural
mixed microbial communities. The specific growth rate is determined by the substrate
concentration which, based on the theoretical consideration of
Monod (Monod 1942, 1950), can be described by
Prokaryotic Growth
S
m ¼ mmax ð7:7Þ
Ks þ S
Parameters of Prokaryotic Growth
where mmax is the maximum value of the specific growth rate
Bacteria multiply in media of appropriate composition and in attained at unlimiting substrate concentrations S and Ks (the
the presence of suitable substrates. The rate of growth is depen- saturation constant numerically equal to the substrate concen-
dent on substrate concentration, temperature, pH, and tration at which m = 0.5 mmax).
A
thus higher than that of ammonium assimilation of some
diatoms, but the specific affinity of ‘‘Candidatus Nitrosopumilus
B maritimus’’ surpasses that of the oligotrophic diatom
Thalassiosira pseudonana by a factor of 30.
mmax (A) = mmax (B)
Growth rate µ
Modes of Prokaryotic Growth
One of the most pronounced differences between prokaryotes

and eukaryotes is the high metabolic versatility of the former.
Among the outstanding metabolic properties of prokaryotes is
the use of inorganic electron donors for energy generation,
Ks(A) < Ks(B)
anaerobic growth, and the fixation of molecular nitrogen. In
Substrate concentration S eukaryotes, these properties are either restricted to a few species
or completely lacking. Novel ways of energy generation continue
to be uncovered (Zengler et al. 1999; Schink and Friedrich 2000;
Ettwig et al. 2010).
A According to the concept of unity in biochemistry, redox
reactions are central to life. Only a few exceptions exist to the
mmax (A) > mmax (C) rule that biological energy is produced in redox processes
Growth rate µ
(Thauer et al. 1977). Such catabolic nonredox processes comprise

the arginine fermentation in Enterococcus faecalis, xanthine
fermentation in Clostridium cylindrosporum, and the pyruvate
fermentation to acetate and formate in Proteus rettgeri. In these
C cases, substrates are metabolized by lytic rather than redox
reactions, and energy is conserved by substrate level phosphor-
ylation. Also, photophosphorylation in Halobacterium halobium
proceeds without the participation of an electron chain.
Ks(A) = Ks(C)
In the case of chemotrophic prokaryotes, the energy-driving
Substrate concentration S cellular processes is derived from an exergonic chemical reaction
. Fig. 7.3 (> Table 7.1), during which an exogenously supplied substrate
Significance of the substrate affinity of growth (mmax/Ks, i.e., the reduces an external electron acceptor (e.g., during respiration
initial slope of the m vs. S curve) for three species A, B, and C. with oxygen, nitrate, or sulfate). Alternatively, a metabolic prod-
Comparison of substrate-dependent specific growth rates for uct of the substrate reduces an electron acceptor formed intra-
two strains A and B with the same maximum growth rate but cellularly from the substrate (fermentation).
different Ks values (upper panel) and for two strains A and C with The standard free energy change of a specific type of micro-
the same Ks value but different maximum growth rates mmax bial metabolism can be assessed from the difference between the
(bottom) standard redox potential of the electron donors (E00 [don]) and
that of the electron acceptors (E00 [acc]; tabulated in Thauer et al.
1977), according to
For comparison of growth characteristics between different
DG00 ¼ n F DE00 ¼ n F ½E0 0 ðaccÞ E0 0 ðdonÞ ð7:8Þ
cultures, the Ks value has been frequently invoked as a measure
of affinity for a given substrate. At similar mmax, microorganisms where n is the number of electrons transferred and F is the
with a lower KS indeed have a selective advantage (> Fig. 7.3, Faraday constant (96.48 kJ·V1).
upper panel). However, the competitive advantage of microor- The standard free energy change of a reaction can also be
ganisms is also determined by the value of its mmax relative to that calculated from the free energy of formation (Gf0) of the prod-
of its competitors. Thus, prokaryotes with similar Ks values may ucts and reactants (Thauer et al. 1977) and, in case of energy-
differ substantially in their growth rate due to their different yielding reactions involving H+, the number of protons formed
affinity (> Fig. 7.3, bottom panel), which is calculated as the (m) and the free energy of formation of a proton at pH 7 and
ratio mmax/Ks (Schut et al. 1997). As an example, the recently 25 C (DG f0 [H+]):
isolated ‘‘Candidatus Nitrosopumilus maritimus’’ exhibits an X X
extraordinarily high affinity of ammonium uptake and capable DG 0 0 ¼ Gf0 ðProductsÞ Gf0 ðreactantsÞ
of depleting ammonium below the detection limit of 10 nM þ m DGf ðHþ Þ
while reaching maximum growth rates similar to the low affinity X X ð7:9Þ
¼ Gf0 ðproductsÞ Gf0 ðreactantsÞ
ammonia-oxidizing bacteria (Martens-Habbena et al. 2009).
The KS value for ammonia for this archaeon is 0.132 mM and þ m ð39:83 kJÞ
. Table 7.1 the principal source of cellular carbon. It has to be noted,

Classification of microbial metabolism on the basis of energy however, that some inorganic carbon-fixing enzymes, such as
source, the type of electron donor, and the type of carbon source acetyl-CoA carboxylase or propionyl-CoA carboxylase (present
in certain autotrophs), as well as anaplerotic enzymes such
Energy source
as phosphoenolpyruvate (PEP) carboxylase or biosynthetic
Chemical Chemo- enzymes such as carbamoyl phosphate synthetase fix bicarbon-
Light Photo- ate instead of CO2 (Neuhard and Kelln 1996; van der Meer et al.
Electron donor 2001). As a rule, these microbes use light or the oxidation of
Inorganic -litho- inorganic compounds for the generation of metabolic energy;
hence, they grow as photolithoautotrophs or chemolithoau-
Organic -organo-
totrophs (> Table 7.1), respectively. Facultative autotrophs
Carbon source
can also grow at the expense of organic carbon compounds,
CO2 -auto- whereas mixotrophs can utilize CO2 and organic carbon com-
Organic -hetero- pounds simultaneously. One and the same strain can grow
Both -mixo- chemolithoautotrophically under one set of environmental con-
-troph ditions but may switch to chemoorganoheterotrophic growth
under different conditions. Consequently, the metabolic types
listed in > Table 7.1 are not mutually exclusive but may occur in
the same prokaryote. In the case of heterotrophic bacteria,
organic carbon compounds are used not only for biomass syn-
(Madigan et al. 2000a). It has been suggested that microbes can thesis but also for energy generation. Approximately half of the
exploit a reaction for growth, if the Gibbs free energy is organic carbon compounds are assimilated by aerobic microor-
20 kJ·(mol substrate)1 and hence sufficient for the transloca- ganisms. Fermenting bacteria assimilate significantly less, typi-
tion of one proton across the cytoplasmic membrane (Schink cally 10–20%, while the remaining carbon substrate is required
1991, 1997). However, experiments with syntrophic cultures as an energy source.
indicate that even smaller amounts of energy may be utilized The redox state of carbon in biomass (in <C4H8O2N>, on
and that syntrophic associations with methanogens metabolize average 0.25) is more reduced than many carbon sources
near thermodynamic equilibrium at values as little as utilized by prokaryotes. This situation is most pronounced in
d G0 = 4.5 kJ·(mol substrate)1 (Scholten and Conrad 2000; the case of autotrophic growth, but also during growth on
Jackson and McInerney 2002). In ecological niches where organic carbon sources such as formate, glycolate, malate, fuma-
available energy is at a minimum, prokaryotes are hence capable rate, oxaloacetate, or citrate. The substrates used in microbial
of maximizing energy conservation when thermodynamic con- metabolism therefore not only serve to generate energy in the
straints begin to limit substrate degradation. It has been case of chemotrophic species, but also yield reducing power for
suggested that the underlying mechanisms control changes in the cells to carry out a variety of reductive processes during
cellular phosphorylation potential or the electron-motive synthesis of new cell biomass. In the classification of microbial
membrane potential of the syntrophically metabolizing metabolism, the (inorganic or organic) nature of the
bacteria and that these mechanisms may be triggered by signals electron donor for these reductive processes is considered
from their hydrogen-scavenging partner (Jackson and (> Table 7.1).
McInerney 2002).
In contrast to chemotrophs, phototrophic prokaryotes
exploit electromagnetic energy for growth. During photosyn- Requirements for Prokaryotic Growth
thesis, light absorption converts a cellular pigment (the reaction
center chlorophyll or bacteriochlorophyll) from a weak to General Composition of the Prokaryotic Cell
a strong reductant. Electrons are transferred to associated elec-
tron carriers and—during cyclic electron transport (cyclic pho- For the understanding of the thermodynamics of bacterial
tophosphorylation)—finally returned to the oxidized reaction growth and as a basis for the design of appropriate growth
center pigment. As a principal difference from chemotrophic media, knowledge of the composition of prokaryotic biomass
prokaryotes, phototrophic prokaryotes can convert light energy is essential. By comparing the ratio of macroelements in
into chemical energy (a proton gradient) without the need of an a medium recipe with that in bacterial biomass, it can be deter-
external electron donor. mined which of the elements will limit growth of the cells in
Organic or inorganic carbon compounds represent the most a culture. Of the more than 100 elements that appear in the
important nutrients for prokaryotic growth. Traditionally, pro- periodic table, some 35–40 are considered essential. Six non-
karyotes have been divided into two major groups with respect metals (C, O, H, N, S, and P), together with four metals (K, Mg,
to their carbon requirement, namely, autotrophs and hetero- Fe, and Ca) comprise an average of 98% of the dry weight
trophs. Autotrophic prokaryotes assimilate carbon dioxide as of prokaryotes. A good first approximation of the molar
. Table 7.2
Macroelements and their physiological functions
Element % dry weight mMa,b Physiological functions

c
C 48–59 Main constituent of organic cellular material
O 13.1–23.9c Organic material and cytoplasmic water
H 6.4–8c Organic material and cytoplasmic water
c
N 13.6–14.7 Proteins, nucleic acids, and coenzymes
S 0.9–1.4c Cysteine, methionine in proteins, coenzymes thiamine pyrophosphate, coenzyme A,
biotin, and a-lipoic acid
P 2–5.4d Nucleic acids, nucleotides, phospholipids, teichoic acids, and coenzymes
K 0.03–1.4 18–800 4500 Predominant monovalent cation in cytoplasm, maintenance of cell osmolarity,
cofactor of some enzymes (e.g., pyruvate kinase, peptidyl transferase, L-malate
dehydrogenase; Walderhaug et al. 1987)
Mg 0.007–0.05 6–46e,f Predominant intracellular divalent cation, cofactor of many enzymes (e.g., kinases), in
phosphate esters (coordinating phosphoryl oxygen atoms), ribosomes, membranes,
and cell wall
Ca 0.00006–0.06 0.01–9.5g Present in exoenzymes (a-amylases and proteases) and in cell walls, role in
transformation, Ca++-dipicolinate is an important component of endospores,
possibly in many putative EF-hand Ca++-binding proteins such as calsymin
Na 1.45 1400 Bacterial oxaloacetate, glutaconyl-CoA, and methylmalonyl-CoA decarboxylases;
NADH-quinone reductase of halophilic Vibrio spp. (Skulachev 1987), transport,
H4MPT: CoM methyltransferase of methanogenic archaea
Cl 0.05 31–1600f,h Essential for active uptake of compatible solutes and flagella formation in halophiles,
for example, H. halophilus
Fe 0.003-0.02i 1.2–7.7 Present in cytochromes, ferredoxins, and other Fe–S proteins, cofactor in enzymes
1.8k (some dehydratases)
Abbreviations: CoA, coenzyme A; NADH, reduced beta-nicotinamide adenine dinucleotide; and H4MPT: CoM, reduced molybdopterin coenzyme M
a
Interconversion between % dry weight and mM was done based on the cytosolic volume of E. coli of 0.9 mL · (mg protein)1, equivalent to 0.45 mL · (mg dry
mass)1 (Gangola and Rosen 1987)
b
Intracellular concentrations in halophilic archaea or bacteria are given in italics
c
From Norland et al. (1995) and Battley (1995)
d
From Damoglou and Dawes (1968) and Norland et al. (1995)
e
From Schmidt et al. (1971); up to 60 mM in moderately halophilic eubacteria (Shindler et al. 1977)
f
From Battley (1995)
g
In E. coli depending on extracellular concentrations (Gangola and Rosen 1987). The intracellular concentration of free Ca2+ is 90–700 nM
h
In H. halophilus (Roeßler and Müller 1998). Chloride has been shown to be obligatory also for growth of Paracoccus denitrificans, Aeromonas hydrophila,
Escherichia coli, Proteus mirabilis, P. vulgaris, Vibrio fischeri, Bacillus megaterium, B. subtilis, Staphylococcus aureus, Corynebacterium glutamicum, and Thermus
thermophilus (Roeßler and Müller 2002)
i
Niehaus et al. (1991), calculated from their value of 1.6–6.9 mmol Fe · (g protein)1, and Braun (1997), based on the cellular amount of iron ions of 105 cell1 and
a cytosolic volume of 0.14 mm3. Intracellular volume calculated from a protein content of E. coli of 156 fg · cell1 (Neidhardt and Umbarger 1996) and the cytosolic
volume given in footnote (a). Similar values are also found in Abdul-Tehrani et al. (1999)
k
Rouf (1964)
composition of prokaryotic biomass is given by the formula It should be noted, however, that certain gliding bacteria, for
<C4H8O2N>, although a more elaborate formula, namely, example, the marine filamentous sulfate-reducing bacterium
C4H6.4O1.5NP0.09 S0.024, has been published (Battley 1995). The Desulfonema magnum, strictly require calcium concentrations
total ionic constituents of the cytoplasm account for 1% of its of 4 mM (Widdel et al. 1983). This appears to be related to
dry weight. Consequently, the so-called macroelements a higher Ca2+ requirement of gliding motility (Castenholz 1973;
(> Table 7.2) are needed in relatively high concentrations in Burchard 1980). While total iron content of bacterial cells is
the growth medium. Most bacteria do not require Na, although in the millimolar range, concentrations of free iron ions are in
many marine bacteria, certain phototrophs, and some strict the tens of micromolar range (Yamamoto et al. 2004). Yet,
anaerobes require this element. Chloride has been found to be certain bacteria grow independently of iron. The Lactobacilli
obligatory for the growth of a variety of bacteria (> Table 7.2). were the first of these iron-independent organisms identified.
Their metabolic independence of iron explains their ability A large fraction of phosphorus is bound in RNA. As a result,
to grow in milk which represents an iron-restricted medium phosphorus demand increases with the specific growth rate of
due to the high concentrations of the iron-binding protein the cells. No reduced phosphorus compounds are stable, and
lactoferrin. The genome sequence of the spirochete Borrelia cellular phosphorus is in the oxidation state of phosphate.
burgdorferi does not seem to encode iron proteins, and the Hence, inorganic phosphate is the usual source of this element
spirochete Treponema pallidum also lacks requirement for iron for microbial nutrition. Since phosphate is frequently also
(Andrews et al. 2003). used as a pH buffer, it is added in excess of the growth require-
The range of organic carbon compounds utilizable by het- ment. However, high concentrations of inorganic phosphates
erotrophs is vast; virtually any compound synthesized by bio- may be growth inhibitory to at least some aquatic bacteria
logical processes, as well as many xenobiotica (compounds (Bartscht et al. 1999). Organic phosphorus sources such as
synthesized in the laboratory which do not originate in glycerophosphate are an alternative supply.
nature), can be degraded by microbes. Different species of Cellular sulfur is mostly in the reduced state and present in
heterotrophic prokaryotes utilize considerably different num- the sulfur-containing amino acids cysteine and methionine.
bers and kinds of carbon substrates. Some, such as the pseu- Many bacteria are capable of assimilatory sulfate reduction and
domonads, are versatile and are known to utilize over 100 thus can utilize sulfate to satisfy the sulfur requirements for
different carbon compounds as the sole source of carbon and growth. Other prokaryotes, such as about half of the strains of
energy. Their substrates include carbohydrates, sugar acids, anoxygenic phototrophic bacteria or methanogenic archaea,
polyols, fatty acids, primary alcohols, amino acids, and aro- require sulfur in a reduced form as either sulfide or an organic
matic substances. In contrast to these versatile bacteria, several compound with a thiol group, such as cysteine.
groups exist that are limited in their ability to decompose Potassium is the principal inorganic cation in the cell and
organic compounds. In this category are the obligate reaches a concentration of about 300 mM in the cytoplasm.
methylotrophs that only use methane, methanol, dimethyl Although much of it is bound in the ribosomes, it is also
ethers, and a few other compounds. Highly specialized species a cofactor of some enzymes, is required for carbohydrate metab-
are restricted to the use of only one type of organic carbon olism, and is involved in many transport processes and osmoreg-
substrate, for example, Bacillus fastidiosus to uric acid. Many ulation. Usually, an inorganic potassium salt (K2SO4 or KH2PO4)
chemoorganotrophs still require carbon dioxide in small is added to the growth medium to satisfy this requirement.
amounts for anaplerotic reactions, for example, the synthesis The magnesium requirement of bacteria is principally that of
of oxaloacetate by PEP carboxylase, PEP carboxykinase, PEP bacterial ribosomes. Magnesium also functions as an enzyme
carboxytransphosphorylase, and by pyruvate carboxylase cofactor and is present in cell walls and membranes. It is usually
(Wood-Werkman reactions). Since carbon dioxide is produced supplied as magnesium sulfate. Iron is required (though not by
during catabolism of organic compounds, it does not normally Borrelia burgdorferi) at micromolar concentrations, whereas
become a limiting nutritional factor. However, some bacteria, the trace elements listed in > Table 7.3 occur in only minute
such as Neisseria and Brucella, require higher concentrations of fractions of between 106% and 0.008% dry weight in the
carbon dioxide (up to 10%) in the atmosphere for good growth bacterial cell (Rouf 1964). Trace elements are toxic at higher
in organic media, a need that must be considered in isolating concentrations and thus should be present in growth media at
and cultivating such organisms. concentrations between 0.01 and 1 mM. The presence of some
Oxygen and hydrogen are derived from water and, in the trace elements can actually prevent the growth of certain pro-
case of heterotrophic bacteria, from the organic carbon source. karyotes. Thus, tungsten (the heaviest metal with well-
Molecular O2 is required in only few exceptions, where an OH documented functions in bacteria) is misincorporated into the
group is introduced by mono- or dioxygenases during the bio- nitrogenase of spirochetes from the termite hindgut resulting in
synthesis of cell constituents. Remarkably, exogenous oxygen an inactive enzyme and in this manner prevents diazotrophic
does not seem to be mandatory in all cases. The recently discov- growth in laboratory culture (Lilburn et al. 2001). In this case,
ered ‘‘Candidatus Methylomirabilis oxyfera’’ couples anaerobic tungsten has to be removed and increased amounts of molyb-
methane oxidation with the reduction of nitrite to dinitrogen, denum (final concentrations of 5 mM) have to be added to the
apparently by conversion of two intermediate nitric oxide mol- medium. Many iron-oxidizing bacteria are strongly inhibited by
ecules to dinitrogen and oxygen. The latter is then used to tungstate (Sugio et al. 2001).
oxidize methane (Ettwig et al. 2010). In cellular material, nitro- Since divalent and trivalent metal cations tend to form
gen is incorporated in the reduced form as amino groups. Many insoluble hydroxides or phosphates at neutral to alkaline pH,
prokaryotes are capable of assimilatory nitrate reduction and they may become unavailable to bacteria in growth media.
therefore grow with nitrate as nitrogen source. A number of Concentrated stock solutions of inorganic salts therefore are
aerobic and anaerobic bacteria are capable of nitrogen fixation often kept anoxically at acidic pH, and minimal amounts of
and thus grow in media devoid of a combined nitrogen source. complexing agents (ethylenediamine-N,N,N0 ,N0 -tetraacetic
Some prokaryotes, such as anoxygenic phototrophs, depend on acid [EDTA] and nitrilotriacetic acid [NTA]) are
reduced nitrogen in the form of ammonium salts. Others require incorporated.
amino acids or oligo- or polypeptides as a source of cellular In addition, a variety of microelements (Mn, Co, Cu,
nitrogen. Mo, Zn, Ni, V, and B) is required (> Table 7.3).
. Table 7.3 therefore decreases the need for de novo biosynthesis in the cell,
Microelements: their source and physiological functions if the compounds can be transported into the cell. Generally, the
growth factors recognized are (1) amino acids, (2) purines and
Element Source Physiological functions
pyrimidines, and (3) vitamins. The latter are required only in
small amounts (on the order of mg·L1, or between 0.1 and
2+
Mn Mn Superoxide dismutase, photosystem II, some
enzymes (PEP carboxykinase and citrate 1 mM). The biological function of a number of vitamins is listed
synthase) in > Table 7.4. Synthetic stock solutions of vitamins
Co Co2+ Coenzyme B12-containing enzymes should be kept dark, cold, anoxic, and sterile (by filtration)
(glutamate mutase and methylmalonyl-CoA and at a slightly acidic pH. They should be added to the medium
mutase) after autoclaving the latter.
Cu Cu2+ Cytochrome oxidase, plastocyanin, nitrite However, many prokaryotes obligately depend on the pres-
reductase, oxygenases, and superoxide ence of growth factors in the medium because the cells are
dismutasesa unable to synthesize all cell constituents from a single carbon
Mo MoO42 Nitrate reductase, nitrogenase, xanthine source. In more rare cases, unusual compounds such as porphy-
dehydrogenase, and formate dehydrogenase rins (hemin in the case of Haemophilus spp.), short-branched
Zn Zn2+ Carbonic anhydrase, alcohol dehydrogenase, fatty acids (e.g., 2-methyl-n-butyric acid in the case of
alkaline phosphatase, aldolase, RNA and DNA Methanobacterium ruminantium) or straight-chain fatty
polymerase, and proteinases acids, mevalonic acid, cholesterol (Mycoplasma), choline, beta-
Ni Ni2+ Urease, hydrogenase, and cofactor F430 ine, and polyamines are required. Some vitamins are required
V VOSO4 V-dependent nitrogenase, and only during growth on specific substrates (such as B12 by
bromoperoxidase Escherichia coli). Essential growth factors may include a few
Se SeO32 Formate dehydrogenase, hydrogenase, and vitamins, such as biotin and p-aminobenzoic acid in the case of
glycine reductase in clostridia Clostridium kluyveri, or a large variety of compounds (as in the
W WO42 Some formate dehydrogenases case of Lactobacilli, which grow in media supplemented with
(commensurate with its higher bioavailability peptone, Tween 80, acetate, and Mn2+ salts [MRS-medium]).
replaces Mo in some anaerobic bacteria) Many lactic acid bacteria including Leuconostoc have
B H3BO3 Formation of heterocysts, akinetes in a requirement for Mn2+ (Boyaval 1989). Yeast extract is often
cyanobacteria; antibiotics (boromycins, employed as a convenient source of most vitamins and also
tartrolons) in Streptomyces, Sorangium; amino acids. Freshly prepared yeast autolysate has been found
possibly cross-linking of pyranoses in cellular to be superior to commercially available dried preparations
envelope,b and AI-2 (cyclic borate diester)c,d
(Leadbetter et al. 1999; Vogl et al. 2006). However, the addition
PEP phosphoenolpyruvate, V-dependent vanadium-dependent, AI-2 of growth factors such as yeast extract or casamino acids also
autoinducer-2 clearly inhibits growth of some photolithoautotrophic or
a
Certain cyanobacteria such as Synechococcus are very sensitive to Cu, which
chemolithoautotrophic bacteria even at very low concentrations
therefore has to be omitted from the respective growth media
b
Loomis and Durst (1992) of 0.01% (w/v) (Overmann and Pfennig 1989).
c
However, boric acid acts as a bacteriostatic agent under certain circum- In many instances, the nature of additional growth factors
stances (Lum and Meers 1989) required for the growth of fastidious prokaryotes is not
d
Chen et al. (2002) known. Besides yeast extract, also fresh rumen extract,
fermented rumen extract or swine manure, or filtered hindgut
homogenates of beetles larvae (Geissinger et al. 2009) have
been successfully employed. Sterilized rumen extract contains
a number of volatile, particularly short-branched, fatty acids,
Certain prokaryotes, in particular strict anaerobes, often also but also vitamins and hemin. Sludge supernatant is another
require selenium and tungsten. Recently, the requirement for complex source of supplines but has been applied less frequently.
boron has been recognized as being rather widespread, since The polyol Tween 80 (polyoxyethylene sorbitan monooleate)
many bacteria employ the AI-2 autoinducer, a cyclic borate can be used as a water-soluble source of the long-chain fatty
diester. acid oleate (cis-9-octadecenoic acid) and in addition consists of
linoleate, palmitate, and stearate as minor compounds. Cold soil
extract has been observed to stimulate growth of certain soil
Growth Factors bacteria.
Growth factors may also stimulate microbial growth
Many bacteria grow with a single carbon compound added to in an indirect fashion. Recently, evidence has accumulated
the medium. Additional organic compounds often stimulate the for a stimulatory role of humic substances and extracellular
growth of microorganisms when present in small concentra- quinones (e.g., anthraquinone-2,6-disulfonate) in dissimila-
tions. These so-called growth factors represent building blocks tory iron reduction (Lovley and Blunt-Harris 1999). These
of major cell constituents, and their addition to the medium redox-active compounds can serve as electron shuttles between
. Table 7.4 grow even at pH values of 1 (e.g., Acidithiobacillus and

Functions of vitamins and related compounds in prokaryotes Sulfolobus). Prokaryotes with optima above pH 8.5 are called
alkaliphiles and grow up to values of pH 11.5 (e.g., Sporosarcina
Vitamin Physiological functions
pasteurii). Many enzymes and structural components in these
p-Aminobenzoic Precursor of tetrahydrofolic acid involved in extremophiles cannot function properly at the very high or low
acid one-carbon transfer pH values found in their external environment, and the intra-
Biotin Carbon dioxide fixation and release cellular pH is kept at relatively constant values (Krulwich and
Coenzyme M Methane formation Guffanti 1983, 1989; Padan and Schuldinger 1986; Matin 1990).
Folic acid Tetrahydrofolic acid involved in one-carbon In general, the internal pH of acidophiles is regulated to a value
metabolism of 6.0–7.0, and alkaliphiles maintain an internal pH 1–2 units
Lipoic acida Transfer of acyl groups, for example, prosthetic lower than the external value.
group of pyruvate dehydrogenase complex The majority of natural environments possess pH values
Thiamine (B1) Thiamine pyrophosphate is prosthetic group of between 5 and 9, and the pH optima for the growth of most
decarboxylases and transketolases prokaryotes (neutrophiles) fall well within this range.
Riboflavin (B2) Precursor of FMN and FAD, prosthetic group of Neutrophiles keep the internal pH slightly more alkaline than
flavoproteins, and redox reactions this value outside. For example, Escherichia coli regulates its
Pyridoxine (B6) Pyridoxal phosphate coenzyme of
internal pH at 7.4–7.8 during growth over the external pH range
transaminases and amino acid decarboxylases of 5.0–9.0 (Slonczewski and Foster 1996). A much increased
cultivation success for methanotrophs, Acidobacteria, and
Cyanocobalamin Molecular rearrangement reactions (e.g.,
(B12) glutamate mutase) methanogenic Archaea from northern acidic wetlands has been
attained by employing low-ionic-strength, low-nutrient acidic
Pantothenic acid Precursor of CoA, prosthetic group of acyl
carrier proteins, and participant in metabolism
(pH 4.0–5.5) media (Dedysh 2011).
of fatty acids During the growth of microorganisms in cultures, pro-
nounced pH changes can occur because organic acids (especially
Nicotinic acid Precursor of NAD and NADP+ and found in
many dehydrogenases during fermentation) and ammonia from nitrogen-containing
compounds are produced. When exposed to pH values at the
Vitamin K Precursor of menaquinone and electron carrier
in respiratory chains
upper or lower end of the permissive range of growth, prokary-
otes may exhibit metabolic properties not displayed at normal
Hemin Precursor of cytochromes, and involved in
pH values. A well-known example is the shift toward production
redox reactions
of nonacidic end products by fermentative bacteria (Graham
FMN flavin mononucleotide, FAD flavin adenine dinucleotide, CoA coenzyme and Lund 1983; Gottwald and Gottschalk 1985; Ferchichi et al.
A, NAD nicotinamide adenine dinucleotide, NADP nicotinamide adenine dinu-
1986; Huang et al. 1986; Forsberg 1987; Hommes et al. 1989).
cleotide phosphate
a
Dithiooctanoic acid Weak organic acids also influence the cytoplasmic pH in a direct
From Gottschalk (1985) manner since in their undissociated form they are lipophilic and
therefore rapidly diffuse through cell membranes, ultimately
conducting hydrogen ions along the transmembrane gradient.
As a result, the intracellular pH is lowered. At low pH values, the
the cells of dissimilatory iron-reducing bacteria and their effect of extracellular pH in microbial growth is therefore mag-
insoluble electron acceptors like Fe(III) oxides. In addition, nified in the presence of organic acids. Hence, permeant acids
organic ligands such as EDTA, NTA, n-methyliminodiacetic such as benzoic, propionic, and sorbic acids are inhibitors of
acid, ethanol diglycine, deferoxamine, or phosphates, if microbial growth (and in fact are used as food preservatives),
added to growth media, make Fe(III) more available and and their potency increases with decreasing pH (Ingraham and
increase rates of microbial reduction significantly (Lovley et al. Marr 1996). However, the sensitivity of bacterial cells to low pH
1996b). values is less pronounced in the stationary phase (Ingraham and
Marr 1996).
In high-nutrient-strength complex media, the various acidic
pH and basic functional groups of organic molecules often provide
sufficient buffering capacity. In mineral media and low-nutri-
In natural environments, the pH value varies from below 1 in ent-strength complex media, continuing acidification or alka-
acidic springs to over 11 in soda lakes (Brock 1978; Grant and linization of the growth medium will rapidly lead to arrest of
Tindall 1986; Madigan et al. 2000a). Prokaryotes can be isolated growth of prokaryotes. Therefore, the pH has to be maintained
from these environments and are capable of growing at such pH within the permissive range for growth using appropriate
values (Brock 1978; Langworthy 1978; Horikoshi and Akiba organic or inorganic buffer systems (> Table 7.5). Below pH 3,
1982; Krulwich and Guffanti 1989). Species with optima for the actual concentration of hydronium ions becomes high
growth below pH 5.5 are usually called acidophiles and can enough to obviate the need for buffering.
. Table 7.5
pH buffers used for culturing prokaryotes
Buffer pKa Buffering range Concentration References

HOMOPIPES 4.55 (37 C) 4.0–5.0 Slonczewski and Foster (1996)
MES 5.96 (37 C) 5.5–6.5 Slonczewski and Foster (1996)
CO2(H2CO3)/NaHCO3a 6.35 (25 C) 5.4–7.8 30 mM Widdel and Bak (1992)

PIPES 6.66 (37 C) 6.0–7.0 Slonczewski and Foster (1996)
KH2PO4/K2HPO4 7.2 (25 C) 5.8–7.8 10 mM Bast (2001)
MOPS 7.2 (25 C) 6.5–7.7 10 mM Bartscht et al. (1999)
7.01 (37 C) Slonczewski and Foster (1996)
HEPES 7.5 (25 C) 6.8–8.2 10 mM Bartscht et al. (1999)
Tris · HCl 8.08 (25 C) 7.2–9.0 10 mM Bast (2001)
TAPS 8.11 (37 C) 7.5–8.5 Slonczewski and Foster (1996)
CAPSO 9.43 (37 C) 9.0–10.0 Slonczewski and Foster (1996)

CAPS 10.08 (37 C) 9.5–10.5 Slonczewski and Foster (1996)
NaHCO3/Na2CO3 10.4 (25 C) 8.5–11.0 50 mM Horikoshi and Akiba (1982)
CAPS 3-(cyclohexylamino)-1-propanesulfonic acid, CAPSO 3-(cyclohexylamino)-2-hydroxy-1-propanesulfonic acid, HEPES N-[2-hydroxyethyl] piperazine-N0 -[2-
ethanesulfonic acid], HOMOPIPES homopiperazine-N,N0 -bis-2(ethanesulfonic acid), MES 2-(N-morpholino)ethanesulfonic acid, MOPS 3-[N-morpholino]
propanesulfonic acid, PIPES piperazine-N,N0 -bis-(2-ethanesulfonic acid), TAPS 3-[N-Tris(hydroxymethyl)methylamino]propanesulfonic acid, Tris · HCl tris
[hydroxymethyl]aminomethane
a
Applicable in closed gas-tight vessels only
Not all buffers listed can be applied for the isolation and of 20–50 mg L1 to maintain pH of growth media at 4.0–5.0
cultivation of all prokaryotes. Phosphate buffers, although (Pankratov and Dedysh 2011).
widely employed, inhibit the growth of bacteria from low- For the preparation of low pH solid media, silica gel has to be
nutrient environments such as freshwater lakes even at compar- employed as solidifying agent (see section > ‘‘Treatment of
atively low concentrations of 10 mM (Bartscht et al. 1999) and Growth Media’’).
were also found to act inhibitory for Acidobacteria from acidic
wetlands (Dedysh 2011). At higher concentrations (30 mM),
phosphate buffers are inhibitory also to many other microor- Osmolarity
ganisms (Bast 2001). A second disadvantage of phosphate
buffers is their tendency to form precipitates with Ca2+, Mg2+, Microorganisms differ considerably in the amount of water they
and Fe3+ ions already at rather low concentrations (e.g., 7 mM require in their immediate surroundings (Brown 1976; Kushner
PO43 and 0.7 mM Ca2+). In these cases, the Ca2+ concentration 1978; Yancey et al. 1982). At higher concentrations of aqueous
needs to be decreased to 200 mM, which (as calculated from the solutions, forces between solutes and solvent lead to a lower
cellular Ca2+ content; > Table 7.2) still permits a sufficient cell ‘‘effective concentration’’—the so-called activity—of water.
yield of 7 g dry weight·L1 (for exceptions such as gliding Availability of water is measured as the water activity aw, which
bacteria, see section > ‘‘General Composition of the Prokaryotic is defined by the mole fraction of water Nw and the activity
Cell’’). The sulfonate buffers listed in > Table 7.5 have been coefficient gw according to Schopfer and Brennicke (1999)
found to be useful in several instances (Slonczewski and Foster
aw ¼ g w N w ð7:10Þ
1996; Bartscht et al. 1999). Substituted amines such as TRIS (tris
[hydroxymethyl]aminomethane) and triethanolamine should According to Raoult’s law, the water activity in a dilute
be avoided as they cross the cytoplasmic membrane in their solution is equal to the ratio of the vapor pressure P of the
deprotonated form. In general, the ionic form of a buffering solution relative to the vapor pressure P0 of pure water at the
substance is less toxic than its undissociated form, since the same temperature:
charged molecules penetrate the cellular membrane much P n1
more slowly. Consequently, cationic buffers (e.g., TRIS) are aw ¼ ¼ ð7:11Þ
P0 n1 þ n2
used at pH values below their pK, while anionic buffers are
employed at a pH above their pK value. In addition to the buffers Here, n1 is the number of moles of the solvent and n2 the number
listed above, alginic acid has been employed at a concentration of moles of ideal solute. Values of water activity range between
0 and 1. The majority of microorganisms known grow well only (Reed and Walsby 1985; Overmann et al. 1991; Ingraham and
at values between 0.9 and 1.0 (Nichols et al. 1999), as they are Marr 1996; Lengeler et al. 1999).
commonly found in aquatic and soil environments. Freshwater Two major strategies have been found by which prokaryotes
media used for routine cultivation usually have aw values 0.99, adjust intracellular aw during changes of extracellular salt con-
whereas seawater (3.5% NaCl) has a value of approximately 0.98. centrations. Aerobic, extremely halophilic Archaea
Salt lakes in which Halobacterium and Halococcus thrive are (Halobacteriales) and anaerobic, halophilic bacteria
saturated with NaCl, and aw values are around 0.75. The lowest (Haloanaerobiales) accumulate inorganic ions, especially K+
value reported at which growth of microorganisms still occurs is and Cl (Oren 1986, 1999; Galinski and Trüper 1994). As
0.61 for the xerophytic fungus Xeromyces bisporus, which can a special adaptation to the resulting high intracellular ion con-
grow on dry foods. centrations, these prokaryotes contain salt-adapted or even salt-
Since the chemical potential of water mw is determined by requiring enzymes that contain a large fraction of polar and
the chemical potential of the pure solvent mw and the water acidic amino acids (especially aspartate and glutamate) but
activity by only a low fraction of hydrophobic amino acids. Aspartate and
mw ¼ mw þ RT ln aw ð7:12Þ glutamate residues bind intracellular water especially well. Con-
sequently, correct folding of these enzymes occurs despite low
(where R is the gas constant and T the absolute temperature), intracellular water concentrations (Oren 1999).
higher concentrations of solutes decrease the chemical poten- At lower external osmolalities (1 Osm), turgor pressure in
tial of water. Most prokaryotes maintain an internal activity E. coli is also mainly regulated by increasing intracellular K+
of water lower than that of the surrounding medium. At the concentrations (and its counterion, glutamate) parallel to an
same time, the cytoplasmic membrane is highly permeable to increase in external osmolality. The intracellular concentration
water, and bacteria have a high surface-to-volume ratio. of Na+ does not vary directly with external osmolality, and
Essentially, the chemical potential of intracellular water intracellular concentrations of the divalent polyamine putres-
must be equal to that outside. As a result of the lower intra- cine are reduced by excretion of this compound into the
cellular water activity, a tendency exists for water to enter the medium, thereby decreasing the intracellular concentration
cell. The direct consequence is osmosis, defined as diffusion of from 50 to 5 mM. It has been suggested that putrescine bound
solvent (water) through a semipermeable membrane in the to nucleic acids is displaced by high K+ concentrations (Csonka
direction of a decreasing chemical potential of the solvent. and Epstein 1996).
During this process, the free enthalpy of the entire system As a second type of osmoregulation, small soluble organic
decreases. molecules are accumulated upon an increase in extracellular salt
Osmotic pressure p is the hydrostatic pressure (in MPa) that concentrations. This type is found widely distributed among
must be applied to a solution of lower chemical potential of the bacteria. The solutes accumulated do not strongly interfere
solvent to stop net diffusion of solvent through the semiperme- with cellular metabolism and have thus been termed ‘‘compat-
able membrane ible solutes’’ (Brown and Simpson 1972; Brown 1976). In addi-
p Vwo ¼ RT ln aw ð7:13Þ tion, these molecules stabilize cellular proteins and increase their
solubility (Cayley et al. 1992; Bolen 2001). Thus, E. coli cells
with V w being the partial molal volume of water. In diluted accumulate trehalose (O-a-D-glucosyl[1 ! 1]-a-D-glucoside),
solutions, > Eq. 7.13 can be simplified (with N2 = n2/(n1 + n2); accompanied by a reduction in the concentration of K+ and
ln aw = ln(1 N2) N2; and n2 n1): glutamate. It has been suggested that this decrease in ionic
strength relieves the inhibition of transcription and translation
RT R T n2
p¼ N 2
o ¼ R T c2 ð7:14Þ and that the cytoplasmic membrane is stabilized by interaction
Vwo Vw n1
of compatible solutes with the polar head groups of the phos-
where c2 denotes the concentration of the solute. Hence, pholipids (Ingraham and Marr 1996). Various compatible
a concentration of 0.1 M of a solute at 25 C results in an osmotic solutes, zwitterionic derivatives of amino acids, sugars, or
pressure of 2.48 bar or 0.248 MPa. If n different solutes are polyols, have been identified and include proline, glutamate,
present, the osmotic pressure amounts to glutamine, alanine, a-aminobutyrate, glycine betaine (N,N,N-
X trimethylglycine), ectoine, NE-acetyl-b-lysine, Nd-
p¼RT n ci ð7:15Þ acetylornithine, b-dimethylsulfoniopropionate, sucrose, and
i
choline (Csonka and Epstein 1996). Some of these, especially
The presence of a cell wall permits bacteria to maintain the proline, glycine betaine, and choline, are also termed
intracellular osmotic pressure, termed ‘‘turgor pressure.’’ Main- ‘‘osmoprotectants,’’ since they can be added to the growth
tenance of this turgor pressure is essential for the growth and medium where they are taken up by the bacterial cells and alleviate
division of the prokaryotic cell. It is therefore understandable osmotic stress effects through direct interactions with intracellu-
that bacteria regulate their turgor pressure over a wide range of lar macromolecules. It has also been proposed that glycine betaine
extracellular values of aw. Gram-negative bacteria have turgor increases the fraction of free water in the cell, and it is believed that
pressures of 0.3–0.6 MPa, while Gram-positive bacteria such as osmoprotectants stabilize and protect enzymes by being excluded
Bacillus sp. maintain a turgor pressure of up to 2 MPa from the protein surface (Cayley et al. 1992). Thus, tolerance of
E. coli toward high external osmolalities can be increased further 1986; Brock 1987; Wiegel 1990). On the basis of their minimum,
by the addition of osmotic protectants such as proline, glycine optimum, and maximum temperatures for growth (the so-
betaine, choline (which is metabolized to glycine betaine), called cardinal temperatures Tmin, Topt, Tmax), organisms are
trimethyl-g-aminobutyrate, b-alanine betaine, taurine betaine, currently divided into four major groups: psychrophiles (also
carnitine, or even MOPS (3-[N-morpholino] propanesulfonic called stenopsychrophiles; Cavicchioli 2006) (Topt 20 C),
acid) to the growth medium (Csonka and Hanson 1991). The mesophiles (Topt between 20 C and 42 C), thermophiles (Topt,
latter is a buffer commonly used in growth media, which has 42–70 C), and hyperthermophiles (Topt, >70 C). In addition,
been shown to accumulate in E. coli at high osmolality (Cayley the term ‘‘psychrotolerant’’ (also called eurypsychrophile; pre-
et al. 1989). Glycine betaine is the most effective osmotic pro- viously ‘‘psychrotrophic’’) has been coined to describe those
tectant of E. coli and Salmonella typhimurium. Accumulation prokaryotes capable of growing at temperatures between 0 C
of glycine betaine reduces the concentration of endogenous and 5 C but reach maximum temperatures of growth exceeding
trehalose as well as ionic solutes. 25 C (Morita 1975). Different temperature ranges for psychro-
The osmoregulatory potential varies between microorgan- philic and psychrotolerant prokaryotes have been used, however
isms, which can be divided according to their osmotic tolerance. (Isaksen and Jørgensen 1996).
This classification has been used most often for NaCl tolerance. Generally, the optimum temperature for growth is only a few
Nonhalophiles thus are organisms capable of growth at NaCl degrees Celsius lower than Tmax. Consequently, it is advisable to
concentrations of >0.2 M, whereas moderate halophiles and perform routine cultivation at incubation temperatures below
marine species usually grow from 0.2 to 3.5 M NaCl, and Topt. It is very difficult to determine the minimum growth
extreme halophiles grow from 1 to 5.5 M NaCl (Kushner 1978; temperature, as doubling times can become extremely long at
Yancey et al. 1982; Epstein 1986; Imhoff 1986; Larsen 1986). the low end of the temperature range.
As a rule of thumb, the rate constants of chemical reactions
increase by a factor of about 2 (the so-called Q10 value; range,
Temperature 1.5–4) when the temperature is increased by 10 C. This
exponential dependence on temperature exists for biochemical
Prokaryotes can grow at temperatures between 17 C or even reactions (e.g., firefly bioluminescence reaction) and likewise
lower (Morita 1975; Baross and Morita 1978; Mazur 1980; for microbial metabolism (e.g., sulfate reduction; Bak and
Carpenter et al. 2000) and +113 C (in the case of the archaeon Pfennig 1991; Isaksen and Jørgensen 1996) and microbial
Pyrolobus fumarii; Blochl et al. 1997; Stetter 2001). The upper growth. Hence, the specific growth rate doubles with a 10 C
limit for microbial growth is suggested to be between 110 C and increase in temperature for many bacteria. Like rate constants
150 C owing to constraints on the thermostability of several of chemical reactions, the temperature dependence of the
essential cell components. The half-life for ATP under these microbial growth rate between the minimum and optimum
conditions is <1 s, and it is 1 ms for polynucleotides temperature can be described by the Arrhenius equation
(Bernhardt et al. 1984; White 1984; Jaenicke 1988). At the (Arrhenius 1889):
other extreme, over 80% of the volume of the biosphere Eo
m ¼ A e RT ð7:16Þ
permanently resides at a temperature below 5 C which is mainly
due to a rather constant temperature of about 2 C in two-thirds in which m represents the specific growth rate, E a the activation
of deep ocean water (Graumann and Marahiel 1996). The energy, R the gas constant (8.31 J K1 mol1), T the temperature
cryosphere (i.e., the part of the planet that is permanently (in K), and the constant A the collision or frequency factor
frozen) covers 15% of the Earth’s surface, and 33% of the surface (in h1), which in chemical reactions describes the collision
is seasonally covered by snow. Other, artificially, low- frequency and orientation of reacting molecules. If applied to
temperature environments encompass refrigerated appliances bacterial growth or physiological activity, Ea does not represent
and products. Viable bacteria have been discovered in snow, activation energy in the chemical sense but rather a measure of
glacier ice, and permafrost soils and even in 1-million-year-old the temperature response of the bacteria. Consequently, Ea has
and 3,600-m deep Antarctic ice layers. Physiological activity has sometimes also been termed ‘‘the temperature characteristic’’
been detected down to temperatures as low as 33 C which has (King and Nedwell 1984). This equation permits an assessment
been attributed to the presence of 50-nm-thin films of liquid at of the temperature dependence of growth or physiological activ-
the surface of ice crystals and to microdiameter liquid ity within certain limits. If the logarithm of m is plotted versus
water veins caused by ionic impurities in the crystal structure the reciprocal of T, a straight line is obtained with a negative
(Carpenter et al. 2000; Karl et al. 1999; Bidle et al. 2007; slope of Ea/2.303·R (Ingraham 1962; Harder and Veldkamp
Christner 2010; Achberger et al. 2011). For bacteria belonging 1971; Ratkowsky et al. 1983):
to the Deinococcus-Thermus group, evidence has been obtained Ea 1
for low rates of bacterial DNA and protein synthesis at ambient ln m ¼ þ ln A ð7:17Þ
R T
subzero temperatures of 12 C to 17 C (Carpenter et al.
2000). Typical Ea values for the specific growth rate or the physio-
Individual strains can grow over a temperature range of 10 C logical activity of bacteria are in the range of 23–132 kJ mol1
to maximally 60 C (usually 30–35 C; Wiegel and Ljungdahl (Bak and Pfennig 1991; Ingraham and Marr 1996;
Knoblauch et al. 1999). On the basis of equation (> 7.3), the chaperones or proteases, exhibit a large (10–20-fold) but tran-
Q10-value can be calculated according to: sient increase in synthetic rate upon temperature upshift (Gross
1996).
Ea 10
Q10 ¼ exp ð7:18Þ In mesophilic bacteria, a specific group of 14 cold shock
R T ðT þ 10Þ
proteins (none of which is a heat shock protein) is produced
At the low, and especially at the high, temperature end, the during the period of growth cessation following a shift from
data of the logarithmic plot of log m versus T 1 deviate from 37 C to 10 C in E. coli, and at least 75 proteins participate in the
linearity. In the case of psychrophiles, the slope of the Arrhenius cold shock response of Bacillus subtilis (Graumann and
plot is linear down to 0 C, whereas for psychrotolerants, it Marahiel 1996). Considerable evidence suggests that the inabil-
deviates from linearity at about 5 C, and for mesophilic bacte- ity to synthesize protein determines the minimum temperature
ria, it tends to deviate from a straight line at even higher tem- of growth in E. coli and that the sensitive step is the initiation of
peratures (Harder and Veldkamp 1971). At high temperatures, translation (Ingraham and Marr 1996; Graumann and Marahiel
the growth rate decreases sharply due to the thermal inactivation 1996). Additional limiting factors of growth at low temperature
of enzymes and disruption of the membrane structure. It should are the fluidity of the cytoplasmic membrane and rate of local
be mentioned that alternative models for the relationship melting of DNA by RNA polymerase (Graumann and Marahiel
between temperature and growth rate constant have been devel- 1996). In psychrophiles, cell membranes tend to contain more
oped which fit the experimental data of many bacterial strains unsaturated fatty acids and short-chain fatty acids than mem-
more precisely (Ratkowsky et al. 1983). branes of mesophiles (Bhakoo and Herbert 1979; Chan et al.
One important factor that determines minimum and max- 1971). Finally, the affinity for substrate uptake is decreased at
imum temperatures for growth is the fluidity of the membrane low temperatures (Nedwell and Rutter 1994). Psychrophiles
lipids, affected particularly by the ratio of monounsaturated and synthesize enzymes with high catalytic activities at low temper-
saturated fatty acids, their length, and by proportion of cyclic atures (Feller et al. 1994b; Trimbur et al. 1994) and produce
fatty acids (Russel 1984; Herbert 1986; Russel and Fukunaga more enzymes upon a decrease in temperature (Feller et al.
1990; Jones et al. 2002). Cold adaptation is related to incorpo- 1994a). The lowest optimum temperature for an extracellular
ration of larger proportions of unsaturated fatty acids. Fatty enzyme recovered from a bacterial culture was 20 C for
acids containing one or more double bonds take a more a protease. The corresponding bacterial isolate produced the
expanded conformation than fatty acids with saturated bonds maximum amount of protease at 1 C, thereby counteracting
and cause less dense packing of the lipids and a higher fluidity of the effects of very low temperatures on the catalytic efficiency of
the membrane (Feller 2007). If the viscosity of the membrane the enzyme (Huston et al. 2000). Chitobiase and leucine-
cannot be maintained within certain limits, it may ultimately aminopeptidase in situ even have lower Topt values of 15 C
become leaky to ions at high temperatures. A second factor is the (Huston et al. 2000). Enhancement of the catalytic activity of
solute transport capability of the membrane (Baxter and psychrophilic enzymes is attributed to an increased flexibility of
Gibbons 1962; Rose and Evison 1965), which is dependent on some of their structural components. Thus, a higher content of
its fluidity. Thirdly, the forces governing the formation of proper non-charged polar amino acids, particularly glutamine and
tertiary and quartenary structure of proteins (Jaenicke 1988) threonine, and hydrophobic amino acids, and fewer charged
limit bacterial growth especially in the high temperature range. residues in the solvent-accessible area were detected in psychro-
The first enzyme inactivated (e.g., homoserine transsuccinylase philic Archaea, which probably destabilize the surface of these
in Escherichia coli; Ingraham and Marr 1996) determines Tmax. proteins and reduce the activation energy of the protein-
Hence, the latter can be increased when the product of this substrate transition state and increase catalytic efficiency at low
critical enzyme is supplied exogenously (methionine in temperatures (Cavicchioli 2006). At the same time, these fea-
E. coli). Several lines of evidence suggest that the inability to tures lead to a reduction of their thermostability. Cold-active
initiate translation and hence to synthesize protein determines enzymes thus exhibit a pronounced heat lability and sensitivity to
the minimum temperature for growth (Ingraham and Marr protein denaturants (Lonhienne et al. 2000). As another aspect
1996). Finally, an important factor determining maximum and of psychrophily, posttranscriptional modification of tRNA is
minimum growth temperatures is the temperature sensitivity of much less pronounced in psychrophiles than in mesophiles.
regulatory mechanisms, which in case of malfunctioning may However, significantly higher levels of dihydrouridine occur in
cause fatal imbalances in cellular metabolism. psychrophiles as compared to psychrotolerants or mesophiles
At very high temperatures >80 C, proton circuits can no (Dalluge et al. 1997). This latter finding has been explained by
longer be maintained owing to a high proton permeability of the the high conformational flexibility, which is maintained by
cytoplasmic membrane. Hyperthermophiles switch to energy a high dihydrouridine content of tRNA molecules.
coupling via Na+ since the permeability of the membrane to Not only the overall specific growth rate but also the phys-
this ion is less affected by high temperature (Lengeler et al. iology of a given strain changes upon changes in incubation
1999). Also, the pattern of proteins changes significantly with temperature. Examples are the formation of different fermenta-
growth temperatures outside the normal range. Changes that tion products (Jung et al. 1974), change in yield and mainte-
occur at higher temperatures are under the control of the heat nance coefficients (Brooke et al. 1989), altered specific
shock response. About 20 proteins, many of them molecular extracellular xylanase activity (Suh et al. 1988), changes in the
affinity for H2 and acetate consumption by Methanosarcina 1995; Knoblauch and Jørgensen 1999; Knoblauch et al. 1999;
barkeri (Westermann et al. 1989), changes in the content of Madigan et al. 2000a). An exemplary list covering the diversity of
unsaturated or branched fatty acids in membrane lipids (Russel recently described psychrophilic bacteria is presented in
1984; Kaneda 1991), and formation of secondary metabolites > Table 7.6.
like pigments, for example, prodigiosin formed only below 30 C Most temperate environments, like deeper parts of the water
by Serratia marcescens (Burkhardt 1992). Thus, different column of lakes and deeper soil strata, are permanently below
enzymes in one and the same prokaryotic strain appear to 20 C. It therefore appears possible that bacteria in these envi-
exhibit considerable differences in thermostability. ronments are adapted to these lower temperatures. Indeed, first
Besides the increased content of unsaturated fatty acids of experiments indicate that the fraction of lake water bacteria
the cytoplasmic membrane, an increased flexibility of tRNAs, growing in artificial liquid media reaches maximum values at
and the lower temperature optima of enzymes, bacterial adap- 16 C (Bussmann et al. 2001). Hence, the preferred temperature
tations to low-temperature environment also include the for- used for enrichments of freshwater or marine planktonic bacte-
mation of cyroprotectants such as soluble carbohydrates and ria is 15 C. Some psychrotolerant bacteria have been isolated
polyalcohols and the synthesis of antifreeze proteins or ice- from packed food after storage at 2–4 C for several weeks (Broda
binding proteins. These latter adaptations occur in bacteria et al. 2000; Kato et al. 2000). Like psychrophilic bacteria,
that are frequently exposed to subzero temperatures. The sea- psychrotolerants also fall in various phylogenetic groups such
ice bacterium Colwellia psychrerythraea produces extracellular as low G + C Gram-positive bacteria (including acetogens;
polymeric substances that were proposed to act as cryoprotec- Nozhevnikova et al. 2001), high G + C Gram-positive bacteria,
tant enabling survival in brine-filled ice veins (Cavicchioli 2006). Proteobacteria, Bacteroidetes, but also methanogenic archaea
The antifreeze protein (AFP) of Marinomonas primoryensis (Nozhevnikova et al. 2001; > Table 7.6). Sulfate-reducing bacte-
lowers the freezing point of water further than any of the other ria isolated from cold sediments are psychrotolerants with
known AFP of fish or insects (Garnham et al. 2008). Due to their respect to growth rate and show a Topt of 18–19 C, whereas the
specific tertiary structure, AFPs (in a broader sense also called maximum growth yield was attained at much lower tempera-
‘‘ice-structuring proteins’’ or ‘‘ice-binding proteins’’) adsorb to tures of 0 C and 12 C (Isaksen and Jørgensen 1996). However,
the surface of the ice crystals, resulting in a thermodynamically the sulfate-reducing activity of bacteria in Antarctic sediments
less favorable mode of ice crystal growth. AFPs often lower the had a temperature optimum well above the in situ temperature,
freezing point, whereas the melting point remains largely exhibiting a mesophilic response (Isaksen and Jørgensen 1996).
unaltered (thermal hysteresis). They are highly effective and Consequently, the temperature response of a given bacterial
can protect cells at concentrations 200–500-fold lower than strain cannot be judged solely on the basis of respiratory activity
NaCl but without increasing significantly the osmotic pressure or growth rate alone.
of the medium (Crevel et al. 2002). The irreversible binding of
the 34-kDa bacterial AFP of Marinomonas primoryensis to ice
crystals is enabled by the highly ordered structure of hydrogen Hydrostatic Pressure
bonds that result in an array of ice-like surface waters and by
hydrophobic interactions (Garnham et al. 2011). Even if freezing Besides temperature, osmolarity, and pH, hydrostatic pressure
is not prevented, ice-structuring proteins protect cells during has been shown to directly influence the growth of prokaryotes.
freezing and thawing. They inhibit recrystallization and the Hydrostatic pressure is a decisive environmental variable in the
formation of large ice crystals in an highly effective manner, deep sea. At nearly 11,000 m, the Challenger Deep is the deepest
stabilizing the initially formed small ice crystals and thereby known oceanic site where pressure values greater than 100 MPa
preventing mechanical damage of the cellular envelope (Chao (1,000 atm) exist. Since the oceans cover about 71% of the
et al. 1996; Tomczak et al. 2003). Such ice-binding proteins earth’s surface at a mean depth of 3,700 m, the high-pressure,
have been detected in Actinobacteria, Firmicutes, Bacteroidetes cold habitat (37 MPa, 5 C) represents the largest portion of
(e.g., Flavobacterium), and Proteobacteria (Pseudomonas, the biosphere by volume.
Marinomonas, and Colwelia spp.) (Christner 2010). As early as 1872, the Challenger expedition revealed the
If some simple procedural precautions are observed, psy- occurrence of living material from depths of at least 8,000 m.
chrophilic microorganisms can be readily isolated from natural Twelve years later, the bacteria found in samples from the deep
environments. Besides using precooled pipettes, media, dilu- sea were shown to be more pressure tolerant than terrestrial
ents, etc., the inoculum should not be exposed to lethal temper- species (Marquis and Matsumura 1978). These findings have
atures above 20 C (room temperature). The number of known now been confirmed. Barophilic microorganisms, recently also
psychrophiles and psychrotolerant bacteria increases steadily termed ‘‘piezophilic,’’ are defined as those well adapted to
(i.e., by 24 novel species in the year 2011). The isolates obtained growth at high pressure (ZoBell and Johnson 1949; Jannasch
belong to the Archaea, low G + C Gram-positive bacteria, high and Taylor 1984), and hence they exhibit optimum growth rates
G + C Gram-positive bacteria, alpha-, beta-, gamma-, and delta- at elevated pressures. Barophilic prokaryotes are usually found
subclasses of the Proteobacteria (and include purple nonsulfur below a depth of 2,000 m in the ocean. At high hydrostatic
bacteria, methanotrophs, and sulfate-reducing bacteria), and pressures and 2 C, doubling times as short as 7 h have been
the Bacteroidetes (Bowman et al. 1997c; Gosink and Staley observed (Yayanos 1986). Obligate barophiles grow only at
. Table 7.6
Some psychrophilic and psychrotolerant bacteria described to date
Group Species Strain References

Psychrophiles
Archaea Methanogenium frigidum SMCC 459WT Franzmann et al. (1997)
Low G + C Gram+ Bacillus marinus DSM 1297T Ruger et al. (2000)
T
Sporosarcina psychrophila DSM 3 Euzéby 2001
High G + C Gram+ Alpinimonas psychrophila DSM 23737T Schumann et al. (2012)
T
Arthrobacter psychrolactophilus ATCC 700733 Loveland-Curtze et al. (1999)
Clostridium vincentii DSM10228T Mountfort et al. (1997)
Cryobacterium psychrophilum NCIMB 2068T Suzuki et al. (1997)
T
Ditzia psychralcaliphila NCIMB13777 Yumoto et al. (2002)
Frigoribacterium faeni DSM 10309T Kämpfer et al. (2000)
Subtercula boreus DSM 13056T Mannisto et al. (2000)
T
Subtercula frigoramans DSM 13057 Mannisto et al. (2000)
Bacteroidetes Flavobacterium frigidarium ATCC 700810T Humphry et al. (2001)
Flavobacterium xueshanense NBRC 106479T Dong et al. (2011)
T
Flavobacterium urumqiense NBRC 106480 Dong et al. (2011)
Gelidibacter algens ACAM 536T Bowman et al. (1997b)
T
Hymenobacter psychrophilus DSM 22290 Zhang et al. (2011b)
Polaribacter franzmannii ATCC 700399T Gosink et al. (1998)
Polaribacter filamentus ATCC 700397T Gosink et al. (1998)
T
Polaribacter irgensii ATCC 700398 Gosink et al. (1998)
Psychroflexus torquis ACAM623T Bowman et al. (1998a)
Psychroserpens burtonensis ACAM 188T Bowman et al. (1997b)
T
Alphaproteobacteria Devosia psychrophila DSM 22950 Zhang et al. (2012)
Devosia glacialis LMG 26051T Zhang et al. (2012)
Sphingomonas alpina DSM 22537T Margesin et al. (2012)
T
Betaproteobacteria Glaciimonas immobilis DSM 23240 Zhang et al. (2011a)
Polaromonas vacuolata 34-PT Irgens et al. (1996)
T
Polaromonas glacialis DSM24062 Margesin et al. (2011)
Polaromonas cryoconiti DSM 24248T Margesin et al. (2011)
Rhodoferax antarcticus ATCC700587T Madigan et al. (2000b)
Gammaproteobacteria Acinetobacter calcoaceticus LP009 Pratuangdejkul and Dharmsthiti (2000)
Glaciecola pallidula ATCC 700757T Bowman et al. (1998b)
Glaciecola punicea ATCC 700756T Bowman et al. (1998b)
T
Methylosphaera hansonii ACAM 549 Bowman et al. (1997a)
Moritella marina ATCC 15381T Urakawa et al. (1998)
Psychrobacter pacificensis IFO 16270T Maruyama et al. (2000)
T
Psychromonas antarcticus DSM 10704 Mountfort et al. (1998)
Shewanella frigidimarina ATCC 700753T Bozal et al. (2002)
T
Shewanella gelidimarina ATCC 700752 Bowman et al. (1997c)
Shewanella livingstonensis LMG 19866T Bozal et al. (2002)
Thiocapsa sp. Ant.Rd Madigan (1998)
T
Deltaproteobacteria Desulfofaba gelida DSM 12344 Knoblauch and Jørgensen (1999)
Desulfofrigus fragile DSM 12345T Knoblauch and Jørgensen (1999)
Desulfofrigus oceanense DSM 12341T Knoblauch and Jørgensen (1999)
T
Desulfotalea arctica DSM 12342 Knoblauch and Jørgensen (1999)
Desulfotalea psychrophila DSM 12343T Knoblauch and Jørgensen (1999)
. Table 7.6 (continued)
Group Species Strain References

Psychrotolerants
Archaea Methanococcoides burtonii DSM 6242T Franzmann et al. (1992)
Bacteroidetes Flavobacterium sinopsychrotolerans JCM 16398T Xu et al. (2011)
Gelidibacter sp. IC158 Nichols et al. (1999)
Pedobacter arcticus A12T Zhou et al. (2012)
Psychroflexus gondwanense ACAM 48T Bowman et al. (1998b)
T
Low G + C Gram+ Acetobacterium tundrae DSM 9173 Simankova et al. (2000)
Carnobacterium funditum DSM 5970T Franzmann et al. (1991)
Clostridium gasigenes DSM 12272T Broda et al. (2000)
T
Lactobacillus bavaricus DSM 20269 Euzéby (2001)
Lactobacillus algidus JCM 10491T Kato et al. (2000)
High G + C Gram+ Arthrobacter flavus MTCC 3476T Reddy et al. (2000)
T
Arthrobacter livingstonensis DSM 22825 Ganzert et al. (2011a)
Arthrobacter cryotolerans DSM 22826T Ganzert et al. (2011a)
Arthrobacter globiformis DSM 20124T Euzéby (2001)
Brevibacterium NCIMB 13216 Nedwell and Rutter (1994)
Leifsonia psychrotolerans DSM 22824T Ganzert et al. (2011b)
Micrococcus agilis Siebert and Hirsch (1988)
Micrococcus roseus Siebert and Hirsch (1988)
Modestobacter marinus DSM 45201T Xiao et al. (2011)
Betaproteobacteria Hydrogenophaga pseudoflava NCIMB 13215 Nedwell and Rutter (1994)
Gammaproteobacteria Colwellia chukchiensis DSM 22576T Yu et al. (2011)
Pseudomonas alcaliphila IAM 14884T Yumoto et al. (2001)
DSM Deutsche Sammlung von Mikroorganismen; ATCC American Type Culture Collection; NCIMB National Collections of Industrial Food and Marine Bacteria;
ACAM Australian Collection of Antarctic Microorganisms; IFO Institute for Fermentation Culture Collection; LMG Universiteit Gent, Laboratorium voor
Mikrobiologie, Gent, Belgium; JCM Japanese Collection of Microorganisms; MTCC Microbial Type Culture Collection at Institute of Microbial Technology,
Chandigarh, India; IAM Institute of Applied Microbiology, University of Tokyo
pressures exceeding 0.1 MPa (1 atm). In studies of more than Pressure-retaining devices have been designed which allow
100 bacterial strains isolated from depths between 2,000 and the sampling of barophilic prokaryotes without decompression
7,000 m, obligate barophile has been detected only in isolates (Jannasch et al. 1976; Yayanos 1978). Phylogenetically, many
from 6,350 m (Yayanos 1986). The maximum pressure per- barophilic bacteria fall in the gamma-subclass of Proteobacteria
mitting growth of an obligate barophile was determined as (DeLong et al. 1997). The major genera of cultivated
115 MPa (Deming et al. 1988). Some barophilic bacteria grow barophiles include the g-proteobacterial genera Shewanella,
at pressures >100 MPa (in one case even 130 MPa, a value Photobacterium, Colwellia, and Moritella. In addition, mem-
reached nowhere in the ocean; Yayanos 1986). Many abyssal bers of other physiological and phylogenetic groups have been
and hadal prokaryotes were shown to be barophilic (Yayanos described, for example, the sulfate-reducing Desulfovibrio
1986). On the contrary, barotolerant microorganisms, which are profundus (Bale et al. 1997). Most characterized barophiles are
also abundant at great depth, grow fastest at 0.1 MPa and more thus closely related to shallow-water marine bacteria. Most of
slowly as hydrostatic pressure is increased. Barophilic bacteria the prokaryotes isolated are also psychrophiles, and some are
isolated from the deep sea are also psychrophiles (Topt 8–10 C). capable of growing at very low nutrient concentrations (Deming
However, the maximum temperature of growth (Tmax) is higher and Colwell 1985; Deming 1986). The latter observation indi-
at high pressure (Yayanos 1986). On the other hand, barophilic cates that barophiles are not confined to nutrient-rich niches,
bacteria show barotolerant properties if cultured not at such as fecal pellets and inside higher organisms, but are also
10–15 C, but at 4 C, which corresponds to the actual found free in the water column and in the sediment. Since at
temperature of their environment, and decreasing substrate least some barophilic prokaryotes are oligotrophs, high-pressure
concentrations induce a more efficient barophilic response in continuous culture techniques had to be developed which now
certain deep-sea psychrophiles (Wirsen and Molyneaux 1999). permit a study of barophilic prokaryotes at pressures of up to
71 MPa and at low and precisely controlled nutrient concentra- (Bartlett et al. 1993; Bartlett and Welch 1995). In addition to
tions (Jannasch et al. 1996; Wirsen and Molyneaux 1999). hydrostatic pressure, OmpH is induced by carbon starvation and
High pressures affect different aspects of cell structure and subject to catabolite control (Bartlett and Welch 1995). Addi-
metabolism, such as membrane structure, transcription, trans- tional pressure-regulated operons have been identified, and for
lation, and the quarternary structure of enzymes, and hence instance comprise a gene involved in the assembly of the cyto-
their activity (Marquis 1976; Marquis and Matsumura 1978; chrome bd complex (Li et al. 1998). Alternative RNA polymerase
MacDonald 1984; DeLong and Yayamos 1986; Morita 1986; s factors (e.g., the rpoE gene product) and modifiers are
Wirsen et al. 1987; Jaenicke 1988; Somero 1992; Welch et al. involved in genetic regulation by hydrostatic pressure (Chi and
1993). On the basis of direction of the change in molecular Bartlett 1995; Nakasone et al. 1998).
volume, biochemical reactions can be slowed down or acceler- High hydrostatic pressures also affect the bacterial cyto-
ated. In fact, increased hydrostatic pressure can accelerate the skeleton. Cells of E. coli form filaments and fail to produce
fructose bisphosphate reaction in a barophilic organism, while septa for cell division at 40 MPa, indicating that cell division is
decreasing it in a non-barophilic one (Hochachka et al. 1972). directly affected under these conditions. Based on evidence
However, most of the biological reactions are slowed down at from in vitro experiments, it has been suggested that high
pressures of 30 MPa or more (Ingraham and Marr 1996). hydrostatic pressure results in dissociation of the bacterial
Escherichia coli is moderately barotolerant and withstands cytoskeleton and growth of the cells (Ishii et al. 2004). Finally,
a maximum pressure of 56 MPa in complex medium, whereas hydrostatic pressure appears to activate cell wall hydrolase
strains of Lactococcus lactis subsp. cremoris are unaffected by activity or to increase cell wall accessibility to the enzyme
200 MPa and inactive at pressures as high as 400–800 MPa (Malone et al. 2002).
(Malone et al. 2002). Five MPa cause a detectable decrease in
growth rate. The pressure sensitive steps are polysome formation
(Schulz et al. 1976) and translocation. Genetic changes in ribo- Treatment of Growth Media and Equipment
some structure can increase the barotolerance of E. coli
(Ingraham and Marr 1996). Types of Culture Media
Several molecular adaptations are thought to be required for
barophily. High pressure and low temperature in the deep-sea Conventional bacteriological culture media are provided as
habitat affect the physicochemical properties of the cytoplasmic either liquid broths or solid media. Liquid media are used in
membrane by tighter packing and by restricting the rotational studies of growth and metabolism in which homogenous media
motion of acyl chains. Changes in membrane fluidity (which is conditions are mandatory. Usually, optical density can be
negatively influenced by high pressures) are known to occur followed easily in liquid media and subsamples for the analysis
through changes in the ratio of unsaturated over saturated of substrates and metabolic products can be withdrawn. In
fatty acids in membrane phospholipids (MacDonald 1984; addition, many bacteria, especially from planktonic samples,
DeLong and Yayamos 1986; Wirsen et al. 1987), similarly to do not appear to grow on solid media and have to be isolated
responses to temperature changes. Many piezophiles in the in liquid dilution series. To maintain an adequate supply of
deep sea contain high proportions of unsaturated fatty acids in oxygen, cultures of aerobic prokaryotes need to be shaken vig-
the cytoplasmic membrane. In some bacteria, these proportions orously, leaving a large head space (gas phase volume).
increase concomitantly with increasing growth pressure. Solid media were originally designed for the enumeration
Omega-3 polyunsaturated fatty acids such as eicosapentaenoic and isolation of bacteria but are now also used routinely for
acid (C20:5) and docosahexaenoic acid (C22:6) are characteris- general culture work. On solid media, colony morphology and
tics of piezophilic and psychrophilic bacteria. In the psychro- other properties (such as swarming over the agar surface) can be
philic piezophile Shewanella violacea DSS12 that was isolated easily observed. Extracellular enzymes originating from the cells
from a water depth of 5,110 m, eicosapentaenoic acid prevents but diffusing into the surrounding noncolonized media can be
the cytoplasmic membrane from becoming hyperfluid and detected as a result of their action on insoluble substrates
largely maintains membrane stability against changes in hydro- (cellulose, starch, and lipid emulsions), which can be
static pressure (Usui et al. 2012). This permits S. violacea to grow maintained evenly distributed in solid as opposed to liquid
optimally at 30 MPa. media. Hydrolysis of starch can be visualized after flooding
The barophilic bacterium Photobacterium profundum plates with iodine, since only the intact starch molecules pro-
strain SS9 preferentially synthesizes a 37-kDa protein, desig- duce the characteristic deep-blue to purple color. Lipids can be
nated OmpH in response to an elevated pressure of 290 atm. stained with Sudan black. Furthermore, the effects of antibiotic
(29 MPa). On the basis of amino acid sequence comparison, substances on colony growth can be tested on solid media.
OmpH is an outer membrane porin and possibly especially Solidifying agents include agar, Gelrite, and silica gel. Agar is
adapted to a high pressure (Bartlett et al. 1989), and it may used most frequently and is a sulfated polygalactan (D-galactose
represent a member of a high-pressure regulon (Bartlett and and 3,6-anhydrogalactose linked by 1 ! 3 and 1 ! 4 bonds)
Welch 1995). Control of the abundance of OmpH is probably produced by marine red algae of the genera Gelidium,
regulated at the transcriptional level. Unusual putative regula- Pterocladia, and Gracilaria. This polymer is degraded by only
tory sequences have been found upstream of the ompH gene very few bacteria such as some Cytophaga, Pseudomonas, and
Vibrio spp. Unusual is the large difference between the temper- fosfomycin (100 [maximum concentrations in mg·L1]) acting
atures for melting (100 C) and solidification (40 C). Con- against bacteria, and cycloheximide (100), tunicamycin (0.25),
sequently, many temperature-sensitive constituents of the colchicine (20), or cordycepin (25) inhibiting eukarya; (2) dyes
media, for example, vitamins, may be added at temperatures (such as crystal violet, methylene blue, or brilliant green) inhibit
slightly above 40 C. Agar-containing media should not be many Gram-positive bacteria; (3) high concentrations of glycine
adjusted to pH values <6.0 before sterilization, because the and LiCl, which permit the growth of many staphylococci but
agar may be hydrolyzed. If lower pH values are required, the not that of physiologically similar other bacteria; (4) bile salts,
adjustment should be done by aseptic addition of acid after heat which permit the growth of enteric bacteria but not that of many
treatment or, alternatively, solid silica gel media should be used. other bacteria; and (5) bromoethane sulfonate (BES) for inhi-
Agar may contain variable amounts of impurities such as Ca, bition of methanogenic archaea, or sodium molybdate for the
Mg, and other minerals (Bromke and Hammel 1987). One inhibition of sulfate-reducing bacteria. However, the inhibitory
means of reducing soluble nonpolymeric contaminants is the effects vary among different phylogenetic groups of bacteria.
repeated washing in distilled water (e.g., five times in double- Despite their alleged broad range action, certain antibiotics can
distilled water, using 300% of the final volume; Widdel and Bak thus be used to selectively enrich particular phylogenetic groups
1992). Alternatively, agar has been purified by subsequent of bacteria. Quite early, it was noted that kanamycin favors the
extraction with acetone and ethanol. For direct plating of enrichment of Flavobacteria (Flint 1985) and rifampicin is rou-
oligotrophs from the marine environment and if even the tinely used to enrich for spirochetes (Stanton and Canale-Parola
washed agar inhibits growth, a glass filter may be used as 1979). Recent examples for selective cultivation of novel types of
a substitute. bacteria include the isolation of a fermentative spirochete from
Various oligotrophic bacteria have been found to be unable sulfur mats in the presence of rifampicin (Dubinina et al. 2011),
to grow on agar-containing media (Giovannoni and Stingl 2007; the isolation of the phylogenetically deeply branching
Dedysh 2011). Gellan gum (or Gelrite, Phytagel, Sigma-Aldrich) betaproteobacterium Parasutterella secunda in oxacillin-
is another alternative for solidification of microbial media. containing media (Morotomi et al. 2011), and the isolation of
Harris (1985) proposed gellan gum as an agar substitute since Fibrisoma limi, a novel type of Bacteroidetes, as well as two novel
agar at high concentrations may be toxic to methanogens. Gellan types of thermophilic aerobic heterotrophs of the phylum
gum (which is produced by strains of Sphingomonas spp.) is an Chloroflexi in the presence of kanamycin (Filippini et al. 2011;
anionic acidic heteropolysaccharide consisting of glucose, Yabe et al. 2011). Similarly, lithium toxicity varies among micro-
glucuronic acid, and rhamnose that overcomes some of the organisms and has been added for the selective growth of
toxic effects that agar has on some groups of microorganisms Bifidobacterium spp. (Lapierre et al. 1992).
(Ferris et al. 1996; Liesack et al. 1997) and starts to solidify upon Selective media are especially used for the isolation, detec-
addition of Ca++; the concentration of divalent cations influ- tion, and recognition of pathogenic bacteria from mixed cul-
ences gel strength and solidification temperature (Bast 2001). tures but are equally important for the isolation of selected
However, the polymerization of gellan gum requires the addi- groups of slow-growing bacteria from environmental samples.
tion of substantial concentrations of Ca++ that potentially In a strict sense, however, all media are at least slightly selective.
inhibit growth of bacteria adapted to oligotrophic conditions. Differentiating media are designed to distinguish one type
The effect of gellan gum does not seem to be restricted to of microorganism from another in a mixed culture.
oligotrophic bacteria, however. Thus, substituting agar with A differentiating medium contains a special ingredient that
gellan gum resulted in a significantly improved cultivation changes during growth of a certain type of bacterium. They are
success of bacteria from a eutrophic freshwater sediment designed to differentiate between morphologically or physiolog-
(Tamaki et al. 2005). Most notably, gellan gum has recently ically similar microorganisms (e.g., hemolytic reaction on blood
been employed simultaneously as the sole growth substrate agar, urease, production of acetoin, and reduction of tellurite).
for the isolation of Chthonomonas calidirosea, a novel Differentiating media can be selective or nonselective.
member of the Armatimonadetes (the former OP10 phylum) Assay media are used for the quantification of organic sub-
(Lee et al. 2011). stances such as vitamins, amino acids, and growth factors in
Silica gel media have been developed for use when solid bioassays, in which the growth response of a certain organism
media free of any organic contamination are needed, low pH is requiring the factor is directly proportional to the concentration
desired, or agar-degrading microorganisms are to be cultivated of the factor under investigation.
(Funk and Krulwich 1964; Bast 2001).
Media may affect the growth of microorganisms
nonselectively or selectively. Selective media favor the growth Media Preparation and Sterilization
of only some bacteria by the inclusion of a particular substrate as
a carbon/energy source, by the presence or absence of specific Heat-labile supplements such as serum, vitamins, or growth
nutrients (e.g., nitrogen or vitamins), or by the presence of factors (e.g., freshly prepared yeast extract or fermented rumen
compounds with differential toxicity. Inhibitory compounds fluid) are added to the basal medium after sterilization to avoid
include (1) antibiotics with rifampicin (2.4), kanamycin (5.0), deterioration. This is also a recommended practice for com-
erythromycin (50), penicillin (50), tetracycline (100), and pounds that might react with other medium ingredients during
autoclaving. Glucose and other sugars, when autoclaved with radiation, and other treatments. Hence, sterilization proce-
salts such as phosphate, may form inhibitory sugar phosphates. dures need to be designed to kill the most resistant forms of
The carbonyl groups of reducing sugars react with free amino life, namely, the endospores of Gram-positive bacteria. Of the
groups of primary amines and may result in the formation of latter, Moorella thermoacetica forms unusually heat-resistant
toxic Maillard reaction (or ‘‘browning’’ reaction) products such endospores which reach decimal reduction times of up to
as furfurals or furaldehyde. Oxygen or oxidation products accel- 111 min at 121 C (Byrer et al. 2000). Consequently, purified
erate Maillard reactions. Reducing agents such as cysteine and suspensions of bacterial endospores are used as indicators for
sulfide will be oxidized by other medium ingredients during the effectiveness of the sterilization process. Routine sterilization
autoclaving and form toxic radicals (Carlsson et al. 1979; procedures are designed to provide a wide margin so that the
Cypionka et al. 1985) and thus have to be added separately chance of having even a single survivor is less than one in
afterward. The formation of mineral precipitates (e.g., struvite a million.
or MgNH4PO4; Schink et al. 2002) can often be avoided by Direct heat, dry heat, and moist heat are the three most
separate sterilization of solutions of the calcium, magnesium, common methods of sterilization. For sterilization by direct
and/or iron salts, which are added to the cooled medium. Addi- heat, objects are exposed to an open flame, and the adhering
tion of chelating agents such as EDTA helps to prevent the microorganisms are quickly burned. Small equipment such as
precipitation in some cases. In many instances, the microele- inoculating needles and loops, forceps, open ends of culture
ments are present in adequate amounts as contaminants of the tubes, and Pasteur pipettes are routinely flame-sterilized. In
mineral salts used in media or as contaminants of glassware and addition, combustion is the method of choice for destruction
water. In several cases, however, microelements need to be added of disposables and contaminated wound dressings in hospitals.
separately to the growth media. Dispensing of media is usually When employing a Bunsen burner for sterilization, it has to be
carried out after cooling to below 50 C to avoid condensation. kept in mind that air supply must be regulated such that the
Treatments can be distinguished on the basis of their effect flame generated is completely blue (not yellow-orange) and
on total and viable cell numbers with exposure time. Bacterio- that the point where the highest temperature is reached is
static substances inhibit growth of bacterial cells, while total cell found atop of the inner (dark-blue) cone within the flame.
numbers and the viable cell count (determined, e.g., by plating A yellow color of the flame indicates the presence of sooty
of cells after diluting out the inhibitory substance) remain con- particles, hence incomplete combustion and lower tempera-
stant. Bactericidal compounds lead to a decrease, hence an tures. Dry heat is used to sterilize empty glassware and other
irreversible damage, of viable cells, while total cell numbers heat-resistant objects such as laboratory instruments, surgical
remain constant. Finally, bacteriolytic compounds lead to tools, glass syringes, needles, mineral oils, and dry powders of
a decrease in viable as well as total cell numbers owing to heat-stable substances and involves baking at 170 C for 2 h in
prokaryotic cell lysis. a hot-air oven. The objects should be protected from subse-
Sterilization—the complete inactivation or removal of quent contamination by wrapping them in aluminum foil
microorganisms—is achieved by applying heat or irradiation prior to sterilization.
(physical methods) or by treatment with toxic compounds and Obviously, dry sterilization cannot be used to sterilize liq-
gases (chemical methods). Gases and liquids can also be steril- uids, which would boil at temperatures above 100 C at atmo-
ized by filtration through filters with extremely small pores spheric pressure. Also, the method is not suited for the
(preferably 0.1 mm). Disinfection is a procedure that results sterilization of heat-sensitive materials such as cotton wool,
in the inactivation of only a fraction of the microbial population plastics, and rubber. In addition, heat conduction is less rapid
from an object or from a culture and is most often used to in moist air, and dried bacterial cells and spores intrinsically have
inactivate pathogens. When a pure culture is exposed to a higher heat resistance than wet cells and wet spores, which
a lethal agent, the kinetics of death are exponential, that is, makes sterilization by dry heat a more time-consuming process.
when the logarithm of the number of survivors (usually deter- Therefore, moist heat is the most effective and most commonly
mined by counting colony-forming units on a suitable medium) used method for sterilization. Autoclaving denotes a heat treat-
is plotted against time, a straight line is obtained whose down- ment with a water-saturated atmosphere under pressure. In
ward slope is called the ‘‘death rate.’’ The time course of killing is principle, the autoclave represents a type of pressure cooker in
described by the D-value, which gives the decimal reduction which a pressure of 1 atm. and a temperature of 121 C are
time, that is, the time it takes for a tenfold reduction in the reached simply by heating or by inflow of preformed pressurized
microbial population at a particular temperature. The Z-value is steam. Only if the autoclave chamber is completely filled with
the number of degrees that the temperature must be raised to steam can a temperature of 121 C be reached. This temperature,
reduce the D-value tenfold, hence provided a sufficient exposure time is chosen, is sufficient to kill
T1 T2 even bacterial endospores. Therefore, the temperature has to be
Z¼ ð7:19Þ monitored when checking the reliability of the sterilization
log D1 log D2
procedure. As a rule of thumb, 50 min of autoclaving are suffi-
The actual number of survivors is then determined by the cient to sterilize 1 L of liquid even when contaminated with
initial size of the population and the death rate. Differences endospore-forming bacteria (but compare Moorella
exist among microbial species in their resistance to heat, thermoacetica). After sterilization, the steam is allowed to
escape slowly to prevent boiling of liquids, which would other- surgical supplies (sponges), optical equipment, and samples of
wise occur upon a sudden drop in pressure. plant material (if sterile seeds are required; Shockey and
Pasteurization is a mild treatment with moist heat employed Dehority 1989). While polypropylene plasticware can be
to control spoilage of food products and to extend their shelf life autoclaved without problems, polyethylene polymers,
without significant decay of heat-labile constituents such as polystyrole, and some polysulfonates or polyfluoroethylene
vitamins. One treatment is the low-temperature-long-time materials are more temperature sensitive. Microorganisms are
(LTLT) procedure in which, for example, milk is heated for killed by exposing them to toxic chemicals, mostly propylene
30 min at 63 C. In the more frequently used high-tempera- oxide, b-propiolactone, or the most widely used ethylene oxide
ture-short-time (HTST) or flash procedure, heating lasts for (EtO). The latter is an alkylating agent that reacts with
20 s at a temperature of 72 C followed by rapid cooling to hydroxyl-, sulfhydryl, and amino groups in proteins and nucleic
minimize undesirable changes in taste and nutrient content. acids. Ethylene oxide volatilizes above 10.8 C, and sterilization
Pasteurization reduces the number of viable cells by between involves an exposure of materials for at least 4 h to EtO gas in
97% and 99% and is intended to eliminate pathogens. It was a closed chamber, after which they must be thoroughly flushed
introduced by L. Pasteur to control the spoilage of wine. How- for 8–12 h with sterile inert gas or air. EtO is explosive, flamma-
ever, even vegetative cells of certain non-spore-forming bacterial ble, causes skin burns, and is highly toxic. Therefore, nonflam-
species, such as Microbacterium lacticum, Enterococcus spp., mable mixtures of EtO and freon or carbon dioxide with the
and Coxiella burnetii, are capable of surviving pasteurization same microbicidal activity as EtO alone are in use.
and lead to subsequent food spoilage. For decontamination of laboratory surfaces, clear phenolics
Filtration is the method of choice for the sterilization of and hypochlorites (3%) are most commonly used, but alcohols
heat-sensitive liquids and gases, which are passed through filter (ethanol at a final concentration of 70%) or mixtures of alcohols
material with pores small enough to retain the microorganisms. and formaldehyde and iodophores are most effective against
Membrane filters consisting of mixed cellulose esters, polycar- spores.
bonate, polytetrafluoroethylene bonded to polyethylene, or Ionizing radiation is employed in large laboratories for the
polypropylene are most frequently used for sterilization of liq- sterilization of heat-sensitive solid objects such as powdered
uids. Pore size of membrane filters is precisely determined dur- pharmaceuticals, disposable plastics, or clothing. Because it has
ing manufacture, and different pore sizes between 0.05 and the power of penetrating solids, radiation is also used to retard
12 mm are currently available. Passage time is inversely related or eliminate spoilage of foods (Murray 1989). Gamma radiation
to pore-size diameter, and permeability can be affected by the emitted by a radioactive cobalt source is most commonly
chemical or electrostatic properties of the filtrate. In some applied. A high dose of 2.5 Mrad is sufficient to kill microor-
instances (such as natural water samples containing larger con- ganisms, spores, and viruses, but chemical changes in media are
centrations of dissolved organic matter present in colloidal possible. Gamma rays interact with water molecules to produce
form), filters will be rapidly clogged. The application of prefilters ions (OH) and free radicals (OH ) that can significantly alter
with larger pore sizes will reduce clogging of the membrane. and destroy many different biomolecules in the cell.
Certain prokaryotes, such as Flexibacter (Little et al. 1987),
mycoplasmas, or spirochetes, may pass membrane filters. For
safe sterilization, 0.1-mm-pore-size membrane filters should be Handling of Glassware and Equipment
employed since the cells of some prokaryotes are known to have
diameters of around 0.2 mm. In most cases, viruses cannot be An important element of the handling of glassware is its proper
removed by filtration, however. Depth filters clog less rapidly cleaning. New glassware needs special treatment for the removal
and are therefore frequently used for the clarification of liquids. of free alkali. Detergents alone often are not sufficient to remove
Depth filters consist of a matrix of randomly oriented fibers adsorbed compounds that may later inhibit growth especially of
bound together in a tortuous maze of flow channels. However, fastidious bacteria from natural bacterioplankton. Soaking in
depth filters differ considerably from membrane filters since 0.2 N HCl often alleviates these problems. Detergents such as
organisms are trapped within the matrix and gradually go Mucasol® have been found to aid in proper cleaning of glass-
through and hence eventually contaminate the filtrate. This is ware. Glassware should be rinsed thoroughly with bidistilled
not the case for dry (sterilized) cotton plugs which therefore can water after treatment with detergents, since even traces of the
be employed for the sterilization of gases. Also, high-efficiency latter are inhibitory to certain bacteria such as certain
particulate air filters (HEPA) have become available commer- cyanobacteria or marine oligotrophs (Gottschal et al. 1991).
cially for the filtration of large volumes of air, for example, to In extreme cases, in which prokaryotes from oligotrophic
supply clean rooms or laminar flow cabinets. Also, glass pipettes environments are to be isolated, it is necessary to rigorously
plugged with cotton wool or plastic pipette tips equipped with clean water samplers, storage containers, and culture vessels
sterile filters should be used in the laboratory, especially where with acid followed by repeated washings with ultrapure
dealing with medically important bacteria. deionized or double quartz-distilled water (Waterbury 1991).
Chemical sterilization is the method of choice for sterilizing Soaking in appropriate detergent for 1 week and subsequent
solid objects that cannot be treated without damage by physical rinsing with especially clean water followed by 1 week of soaking
methods, such as certain disposable plasticware, plastic tubing, in 0.5 N HCl and repeated washing with clean water has been
recommended for the culturing of cyanobacteria (Waterbury Reductants frequently employed are sulfide or cysteine.
1991). At least in some cases, such as obligate oligotrophic A sulfur-free reducing agent is titanium (III) citrate. Owing to
marine heterotrophs or cyanobacteria of the genera complex formation between titanium and citrate, Ti(OH)3-
Prochlorococcus or Trichodesmium, special precautions against precipitate formation is minimized (Zehnder and Wuhrman
traces of organic or inorganic contaminants are mandatory 1976). The solution can be prepared by adding 5 mL of a 15%
(Chisholm et al. 1992; Orcutt et al. 2002; Giovannoni and Stingl (w/v) titanium (III) chloride solution to 50 mL of a 0.2 M
2007). Non-glass systems consisting of Teflon® vessels are used sodium citrate solution, followed by neutralization with satu-
in which natural seawater is sterilized by gentle tyndallization in rated sodium carbonate. Moench and Zeikus (1983) reported an
a microwave. Furthermore, Millipore Q water is employed for easy method for preparing this reductant with the use of
dilution and cultures grown in Nalgene polycarbonate flasks. nitriloacetic acid (NTA) instead of citrate as the complexing
Because of the potential danger involved in working with agent. However, NTA inhibits growth of certain prokaryotes.
infectious microorganisms, irrespective of their known patho- Furthermore, titanium (III) citrate may be inhibitory to anaer-
genic nature, the requirements for sterile working conditions obic prokaryotes at low growth rates (Wachenheim and Hespell
and safe handling of contaminated glassware and equipment 1984). In cases where organic reducing agents are to be avoided
should be rigid. All used glassware and other materials should and free sulfide is toxic to the prokaryotes, amorphous ferrous
be autoclaved first before unloading in the wash-up room. sulfide has been employed as a reducing agent. This reductant
Sterile rooms and sterile cabinets supplied with gas, electricity, can be easily prepared in the laboratory and reacts much more
sterile air, and ultraviolet (UV) irradiation are useful, because they rapidly with O2 than does either soluble sulfide or cysteine
considerably reduce the possibility of air contamination. For the (Brock and O’Dea 1977). After preparation, media are dispensed
sterilization of inoculation rooms and other work areas, UV anoxically as described by Widdel and Bak (1992).
irradiation (wavelength, 260 nm) is often applied. Laminar flow Three-electrode poised-potential amperometric culture sys-
(clean air) cabinets are designed to provide a work area that is tems have been developed which consist of a platinum counter
protected from the environment and are useful for preventing electrode, a platinum working electrode, and an AgCl-Ag reference
airborne contamination when handling sterile media, such as electrode connected to a potentiostat (Emde et al. 1989; Emde and
during aseptically dispensing sterile fluids and culture media. Schink 1990; Ohmura et al. 2002). These systems can be used to
Filtered air is passed in a vertical or horizontal unidirectional grow anaerobic prokaryotes at a carefully controlled constant
(laminar) flow through the cabinet. This air is made sterile by redox potential despite continuing redox reactions and flow of
filtering through high-efficiency air filters that can remove parti- electrons to the working electrode. During growth experiments,
cles down to 0.3 mm. In one type of cabinet, the flow of filtered air the redox potential of the growth medium and the electron flow
is directed toward the front. In a second type of cabinet, the flow between working and counter electrodes can be recorded.
occurs vertically downward, forming a curtain of sterile air and is Even fastidious anaerobic prokaryotes such as methanogenic
subsequently in part recirculated. For decontamination of lami- archaea can now be cultivated on solid agar media, if anaerobic
nar flow hoods, 3% hypochloric acid can be used. Laminar flow chambers are employed (Leedle and Hespell 1980). This adap-
hoods are certified for handling microorganisms of different tation of the technique for anaerobes permits the application of
hazard classes. For handling bacteria from natural samples and replica-plating techniques to ecological and genetic studies of
with unknown properties, safety class II cabinets should be used. bacterial populations from anaerobic habitats. Sufficient CO2
(10–20%) should be present in the gas phase to maintain the pH
of the medium if a bicarbonate buffer system is used.
Removal and Exclusion of O2, Cultivation of
Anaerobes
Culture Systems
The fundamental methodology for the cultivation of anaerobic
prokaryotes is the Hungate technique (Stewart and Bryant 1988; Batch Culture
Widdel and Bak 1992). Modifications of the original Hungate
technique include the use of butyl rubber stoppers (Hungate The typical sequence of lag phase, exponential phase, and sta-
1966) and of serum bottles closed with crimp-closure aluminum tionary phase (see section > ‘‘Prokaryotic Growth’’) is observed
seals holding butyl rubber stoppers (Miller and Wolin 1974), the when prokaryotes are grown in batch culture. In this most
syringe technique (Macy et al. 1972), and the use of pressurized frequently employed culture system, cells grow suspended in
tubes and vessels for the culture of methanogens (Balch and a medium containing sufficient carbon and energy sources and
Wolfe 1976). Usually, anoxic media contain a bicarbonate-CO2 other required nutrients to allow growth at maximum rate for
buffer system (> Table 7.5), mineral salts, and a reducing agent. a limited period of time. Since fresh supply of essential compo-
Depending on the requirements of the prokaryotes to be culti- nents and removal of metabolic waste products do not occur
vated, clarified rumen fluid or low concentrations of yeast after inoculation, batch cultures are closed systems, and cells
extract are added. Resazurin can be used as a redox indicator, grow in a continuously changing environment, which shifts
provided it does not inhibit growth of the prokaryotes as in the from a nonlimiting supply of nutrients to conditions of starva-
case of many phototrophic sulfur bacteria. tion. Because of its relative simplicity and ease of operation, the
batch culture is nevertheless most widely used as a routine Gond et al. 1986; Dawson et al. 1988; Kole et al. 1988). Especially
procedure for propagation of bacteria both in research and for industrial fermentations, the possibility of turning a fed-
industry. Various degrees of control can be obtained by regulat- batch culture into a ‘‘repeated fed-batch culture’’ by withdraw-
ing important parameters such as pH, temperature, and oxygen ing part of the culture volume at regular time intervals allows the
tension. productive growth phase in principle to be extended indefinitely.
A high-throughput variant of the batch culture technique in The major difference with truly continuous culture systems
combination with high-throughput screening has recently remains that the volume is not kept constant, thus introducing
become popular to recover novel types of bacteria from the permanent transience in growth rate, which is possibly essential
natural environment, particularly aquatic oligotrophs (see for certain metabolic activities.
section > Low Nutrient Concentrations, ‘‘Oligotrophic Bacte-
ria,’’ and ‘‘Ultramicrobacteria’’). In the ‘‘dilution-to-extinction’’
approach, a large number of small volumes of appropriate media Recycling Fermenter, Dialysis Culture,
are prepared in microtiter plates. As another measure to gener- Retentostat, and Recyclostat
ate a large number of positive cultures, each well is inoculated
with 1–5 cells instead of preparing dilution across the microtiter To overcome the problem of permanently changing volumes and
plates (Connon and Giovannoni 2002). Inoculation of the wells concentrations of metabolic end products, the recycling fermen-
can be automated by use of, for example, the Microdrop® ter was introduced. In this type of fermenter, either complete or
pipetting robot (Bruns et al. 2003a). Instead of monitoring partial recycling of the biomass is accomplished, but the culture
growth by classical approaches such as turbidity measurements, liquid is continuously replenished with fresh medium and
membrane arrays are prepared from cultures and bacterial cells removed at the same rate. This is achieved by a cross-flow type
are detected by fluorescence microscopy (Connon and of external membrane that retains prokaryotic cells while
Giovannoni 2002). This provides a much higher sensitivity allowing spent medium to leave the fermenter (Müller and
(103 cell mL1) for the detection of low-density oligotrophic Babel 1996; Ahn et al. 2001). Typically, cross-flow dialysis mem-
cultures than standard optical density measurements (typically branes with large total membrane areas together with peristaltic
105–106 cells mL1). If the cultivation success in the particular pumps are employed for efficient and rapid filtration without
growth medium is lower (i.e., 1–10%), the wells of microtiter clogging of the membrane (Pörtner and Märkl 1998). This type
plates can be inoculated with about 50–200 cells to yield of fermenter has proven an especially valuable tool to obtain
a sufficiently large number of positive cultures (Gich et al. cultures of high cell density since inhibitory metabolic products
2005). Cultivation success can be calculated from the number such as short-chain fatty acids or alcohols are removed. Another
of dividing cells and the total number of cells inoculated (Bruns important use is to study growth rate dependence of microbial
et al. 2003a) (see section > ‘‘Isolation of Prokaryotes’’). Inocu- metabolism at very long generation times (i.e., very low specific
lating a larger number of cells also enables microbial interactions growth rates; Chesbro et al. 1979; van Verseveld et al. 1984;
to be established which in some cases leads to the enrichment of Chesbro 1988; Bulthuis et al. 1989; Müller and Babel 1996).
additional and previously uncultured bacterial types that Under natural conditions, the transition from exponential to
depend on accompanying microorganisms. stationary phase is expected to occur much more gradually than
in conventional batch cultures (Mason and Egli 1993). Hence,
physiological properties, such as maintenance energy require-
Fed-Batch Culture ments and resuscitation, of cells at extreme substrate limitation
can be studied best in retentostats (Mason and Egli 1993; Tappe
A further elaboration of the batch cultivation technique was et al. 1996, 1999). A very simple means of prolonging the
developed especially for certain industrial fermentations transition time between the exponential and stationary phase
(Yoshida et al. 1973; Pirt 1974, 1975). In fed-batch cultures, is through biphasic culture. In this type of culture, a bottom gel
a continuous supply of fresh medium is fed to a batch culture layer containing 4% agar is overlain by liquid medium. This
as soon as the substrate concentrations drop to low (sometimes arrangement results in a rapid exponential growth based on the
growth-limiting) levels. As a result, prokaryotic cells continue to nutrients in the liquid, and a subsequent prolonged deceleration
grow. Since no medium is allowed to flow out, the culture phase of growth when cells utilize substrates slowly diffusing out
volume and cell biomass increase. At the same time, the ratio of the agar layer (Chesbro 1988).
of biomass per amount of nutrients entering the vessel increases
such that nutrient limitation becomes more severe over time and
the microbial growth rate decreases continuously. Ultimately, Chemostat
fed-batch culture systems provide the means to largely extend
the transition between the exponential and stationary growth The main difference between continuous culture and other
phases. For some industrial fermentations, these very conditions batch-type cultures is that it is a typical open system in which
have proven vital for optimal production of metabolites, fresh nutrient medium is added at a constant rate to a well-
for example, citric acids, penicillin, some enzymes, and alcohols mixed culture while the volume is kept constant through an
(Yoshida et al. 1973; Esener et al. 1981; Cleland and Enfors 1983; overflow device (> Fig. 7.4). In general, this culture system is
recorder T air
pH– pO2– pO2–
meter meter regulator T
T N2
T CO2
CO2
8 kPa
pump
sampler
medium
thermostat
. Fig. 7.4
Example for a chemostat arrangement designed for the continuous cultivation of anaerobic or microaerophilic bacteria. The partial
pressure of oxygen can be set to any value between 0% and 21%. A simple regulation of pH is possible by manually controlling the CO2
flow into the chemostat culture with a high precision needle valve and using a CO2/bicarbonate buffering system (From Overmann and
Pfennig 1992)
designed to provide a culture growing permanently in an expo- reservoir medium SR (Monod 1950; Novick and Szilard 1950;
nential fashion at a constant submaximum growth rate. The Herbert et al. 1956; Tempest 1970; Pirt 1975; Calcott 1981;
growth rate is dictated by the rate at which the limiting nutrient Gottschal 1990). The combined effect of growth and dilution
is fed to the culture. In this type of arrangement, the continuous by the inflowing medium will eventually result in a steady state
culture is termed ‘‘a chemostat.’’ In continuous culture in which no further change in biomass concentration X occurs:
approaches with a chemostat, the continuously changing condi- dX
tions characteristic of a batch culture are eliminated (Veldkamp ¼ ðm DÞ X ¼ 0 ð7:20Þ
dt
and Kuenen 1973), and relatively large populations of prokary-
At this point, the specific growth rate exactly balances the
otic cells with constant physiological state can be maintained in
dilution rate, and therefore a chosen rate of culture dilution fixes
the presence of low concentrations of a limiting nutrient, which
the specific growth rate of the culture at a value below mmax.
resemble those under natural conditions. It has to be acknowl-
When the specific growth rate is described by the Monod equa-
edged, however, that in certain natural systems, nutrient supply
tion (Monod 1942)
and cell removal are not closely linked; these environments
therefore do not exhibit steady-state conditions (Wirsen and S
m ¼ mmax ð7:21Þ
Molyneaux 1999). Chemostats offer a tremendously powerful Ks þ S
tool for studying the physiology and ecology of prokaryotes. Its in which Ks is the half-saturation constant for growth, the actual
characteristics render the chemostat one of the most widely used substrate concentration S in the culture is fixed at a low, rate-
culturing devices for studying microbial metabolism under care- limiting value. In the steady state, the substrate concentration
fully controlled environmental conditions in both pure and and biomass density are
mixed cultures (Tempest 1970; Veldkamp 1977; Tempest and
D
Neyssel 1978; Matin 1981; Kuenen and Harder 1982; Gottschal S ¼ Ks ð7:22Þ
1986, 1990; Gottschal and Dijkhuizen 1988; Overmann and mmax D
Pfennig 1992; Van den Ende et al. 1996, 1997).
X ¼ Y ðSR SÞ ð7:23Þ
In a chemostat, growth of prokaryotes is determined by the
dilution rate D, defined as the rate of nutrient supply F (dimen- if the biomass yield Y (the amount of biomass formed per
sion: volume per time) divided by the volume of the culture substrate consumed) remains constant over the range of dilution
vessel V, and by the concentration of the limiting nutrient in the rates employed.
For the optimal production of some microbial metabolites, Auxostat

a continuous culture system is required in which the cells pass
through various stages of growth (Ricica and Dobersky 1981; An essential property of chemostat cultures is that their rate of
Thompson et al. 1983; Parkes and Senior 1988). In these growth is fixed by the rate at which fresh medium is fed to the
coupled systems, two or more chemostats are arranged serially, culture. Although this rate can be varied over a considerable
by connecting the outlet of the first chemostat to the inlet of the range of values, steady state cannot be obtained near mmax, and
second. Conditions vary between the different chemostat ves- washout of the cells occurs at the critical dilution rate (Pirt
sels, for example, with respect to dilution rate, O2 supply, 1975). Furthermore, unbalanced growth (and in some cases
temperature, or substrate supply. Culturing different prokary- washout) may occur when inhibitory metabolites accumulate
otes in coupled chemostats for instance permits the study of or potentially toxic substrates are used. An alternative for
sequential mineralization of recalcitrant organic compounds. obtaining controlled growth at an appreciable and constant cell
Two-stage cascade chemostats are also employed to investigate density in such cases is by switching over to some type of
predator-prey relationships, in which bacteria grown in the first internal control of the rate of medium supply. Such a control
stage are transferred to a second one to serve as food for must be based on a growth-dependent parameter. The first
bacterivorous protozoa (Jost et al. 1973; Swift et al. 1982; continuous culture with internal control was named
Sambanis and Fredrickson 1987; Simek et al. 1997; Pernthaler ‘‘turbidostat’’ because the feedback control was based on mea-
et al. 2001). surements of culture turbidity (Myers and Clark 1944; Bryson
Even more complex, but rarely used, systems are the bidi- and Szybalski 1952). However, a continuous and accurate mea-
rectionally linked multistage chemostats, which permit the surement of turbidity represents a major problem in these sys-
mutual exchange of bacteria and/or nutrients. In a ‘‘gradostat’’ tems because of wall growth, inhomogeneity of the culture,
(Lovitt and Wimpenny 1981), a series of chemostats (linked by etc. Therefore, other parameters directly dependent on the cul-
tubing) is fed from both sides with media of different, some- ture density are measured with electrodes in more recent sys-
times complementary, compositions. The inoculated bacteria tems. Parameters include CO2, O2, pH, redox potential,
will be exposed to different physicochemical conditions in the fermentation products, and sulfide (Watson 1969; Martin and
different chemostat vessels. In ‘‘bidirectional compound diffu- Hempfling 1976; Oltmann et al. 1978; Kjaergaard and Jørgensen
sion-linked chemostats,’’ the various chemostat vessels are not 1979; Schauer et al. 1982; Cypionka 1986; de la Broise and
linked by tubing but through membranes, permitting diffusion Durand 1989). These more recent designs have proven reliable,
of solutes but not bacteria (Keith and Herbert 1985). Multistage and their use is most rewarding in studying growth of microbes
chemostats are especially suited to study prokaryotes under in the presence of inhibitory concentrations of substrates or
conditions prevailing in the natural habitat, since heterogeneous products, and possibly also in selecting mutant strains exhibiting
conditions of environmental parameters and interactions the highest growth rates under such conditions.
between different species can be reproduced with these systems.
For growth studies at extremely low nutrient concentrations,
the chemostat is not a suitable tool since extremely low dilution Possible Reasons for ‘‘Nonculturability’’
rates ( 0.5 h1) give rise to inhomogeneities due to mixing
problems and low steady-state biomass concentrations. As The titer of colony-forming units obtained from a given sample
a more realistic approach, a retentostat can be employed which in almost all cases is significantly lower than the actual titer of
offers the advantage of studying physiological properties of prokaryotic cells in the sample as determined by culture-
cells at extreme substrate limitation (Mason and Egli 1993; independent microscopic techniques. Only few exceptions
Tappe et al. 1999). have been published so far (e.g., Button et al. 1993). This obser-
vation has been termed the ‘‘great plate count anomaly’’ and is
attributed to several factors. One simple explanation would be
that only a fraction of the prokaryotic cells are culturable under
Redox-Controlled Sulfidostat a certain set of conditions such that no single medium will allow
growth of all types of prokaryotes. Indeed, the efficiency of plate
As a special type of continuous culture, the redox-controlled counting could be significantly increased when 25 different
sulfidostat permits the cultivation of phototrophic sulfur- media were employed instead of a single medium (Balestra and
oxidizing bacteria under constant concentrations of hydrogen Misaghi 1997). However, in most cases, the numerically domi-
sulfide (Sánchez et al. 1996). In this type of culture, a constant nant species of prokaryotes from natural samples are not recov-
concentration of sulfide is maintained despite light intensity ered. Certain bacteria and archaea, among them the mesophilic
variations that affect photosynthetic rate and hence sulfide Crenarchaeota which constitute a fraction of up to 34% of the
oxidation. A redox controller modulates the rate at which prokaryotic plankton in subpolar or polar latitudes (DeLong
the medium is pumped into the culture and therefore governs et al. 1994), the clone T78 group of the green gliding bacteria
the dilution rate. In a similar manner, the system can adjust to (Coolen et al. 2002), or most of the Acidobacteria detected by
new rates of sulfide oxidation caused by changes in light molecular methods in soils (Barns et al. 1999), for a long time
intensity. escaped cultivation in numerous different types of media.
It therefore appears reasonable to suggest that (1) cells of Barer and Harwood 1999), caused for instance by oxidative
not-yet-cultured prokaryotes in natural samples are in a specific stress. Hence, dormant cells of Micrococcus luteus are permeant
physiological state which prevents them from growing in con- to certain fluorescent stains but restore the cytoplasmic mem-
ventional cultivation media and/or (2) the physiology of not- brane barrier upon resuscitation (Kaprelyants et al. 1996). Other
yet-cultured species of prokaryotes is fundamentally different factors that may limit multiplication are limiting nutrients or an
from that of known prokaryotes such that cultivation methods (as yet unexplained) upper limit of cell density as described for
applied do not meet the requirements for growth. Sphingomonas sp. strain RB2256 (Schut et al. 1997). Injury of
DNA may activate the SOS response which includes expression
of SulA, a protein that interacts with the tubulin-like protein
Physiological State of Prokaryotic Cells FtsZ, thereby preventing septation and thus resulting in
a filamentous growth of cells (Bi and Lutkenhaus 1993; Walker
Starvation Response 1996). Detoxifying enzymes like catalase and superoxide
dismutase are involved in the prevention of oxidative stress.
The gene product of rpoS, namely, the transcription factor sS, is Heat shock proteins, peptide methionine sulfoxide reductase,
involved in cellular responses to a diverse number of stresses and glutathione reductase are involved in the reversal of damage
(Loewen et al. 1998). Induction of rpoS results in an increased within the cell (Dukan and Nyström 1998; Barer and Harwood
survival of cells under unfavorable conditions (Munro et al. 1999).
1995). Accordingly, and dependent on the growth state of the
cells, RpoS (sS) can positively influence the culturability of
E. coli and Salmonella typhimurium in oligotrophic seawater Dormancy
(Munro et al. 1995). However, sS is also involved in the transi-
tion to stationary phase, and cAMP (in a complex with the Dormancy is defined as a reversible state of low metabolic
cAMP receptor protein [CRP-cAMP]) acts as a negative regula- activity in which viability is maintained. This physiological
tor of the transcription of rpoS (Loewen et al. 1998). Addition of state has been studied extensively for Micrococcus luteus. Incu-
extracellular cAMP has the same effect (Lange and Hengge- bation of stationary-phase cultures at room temperatures for
Aronis 1991). Scavenging transporters (such as LamB), which several months results in large numbers of dormant cells. In the
are regulated by cAMP or endoinduction, are turned on at case of Micrococcus luteus, dormant cells show reduced activity,
higher substrate concentrations than RpoS-dependent functions as exemplified by uptake of the membrane energization-
(Notley and Ferenci 1996). Therefore it appears feasible that, by sensitive dye rhodamine-123 (Kaprelyants and Kell 1993). How-
addition of extracellular cAMP, cells could be maintained more ever, such temporarily nonculturable cells can be resuscitated in
easily in a nutrient scavenging state and hindered from entering the presence of supernatants from growing M. luteus cultures
the protective stationary-phase response, which potentially (Kaprelyants et al. 1994, 1996). The agent responsible for resus-
could facilitate cultivation in the absence of other stress factors. citation of dormant cells (Rpf, the resuscitation promoting
In fact, addition of cAMP to low-nutrient liquid artificial fresh- factor) was identified as a 17-kDa protein exported by M. luteus
water or marine media resulted in significant increase of culti- (Mukamolova et al. 1998). In E. coli, the protein SdiA regulates
vation success compared to controls containing AMP (Bruns the ftsQAZ cluster of essential cell division genes and the P2
et al. 2002, 2003b). promoter (Wang et al. 1991) and is a member of the LuxR
subfamily of transcriptional activators. The expression of SdiA
is regulated in turn by a factor released by growing cells into the
Presence of Dead Cells, Prevention, and Reversal medium (Garcia-Lara et al. 1996). The factor was identified as
of Cellular Damage an n-acyl homoserine lactone (Sitnikov et al. 1996). Cell-cell
signaling between cells of the same clone therefore may be
The presence of nongrowing cells has been shown in laboratory important in regulating cell division.
cultures of Enterobacter aerogenes, which at maximum dou-
bling times of 100 h contained up to 50% of cells not capable of
growing on agar plates (Tempest et al. 1967). Since culture- Substrate-Accelerated Death
independent methods such as microautoradiography or the
direct viable count technique have revealed that up to 50%, If a growth-limiting substrate (e.g., glycerol, glucose, ribose,
and in some cases even 90%, of the prokaryotic cells may be phosphate, or ammonia) is added in concentrations of
metabolically active (Kogure et al. 1979; Fry 1990; Karner and 1–10 mM to cells previously starved for the same substrate
Fuhrman 1997), dead cells may actually not be present at high (Postgate and Hunter 1963, 1964; Calcott and Postgate 1972),
numbers in the natural environment. However, elongating cells growth of the cells is inhibited. This phenomenon has been
may be capable of completing only a limited number of division called ‘‘substrate-accelerated death.’’ Cells challenged with this
cycles, which prevents detection of growth by standard methods substrate exhibit low levels of cAMP, and in the presence of
(Binnerup et al. 1993). This limitation of division cycles can be extracellular cAMP, substrate-accelerated death disappears
caused by injury (e.g., to the cell envelope, proteins, or DNA; (Calcott et al. 1972). Also, the addition of Mg2+ protects cells
against substrate-accelerated death. Substrate-accelerated death well (Giovannoni and Stingl 2007), while organic carbon con-
has been reported to occur in Klebsiella, Escherichia, Strepto- centrations exceed those in seawater by 170 times, not consid-
coccus, Azotobacter, Arthrobacter, and Mycobacteria (Calcott ering the large fraction of recalcitrant organic matter in the
and Calvert 1981). The significance of substrate-accelerated natural environment. It should be noted that even terrestrial
death on recovery of bacteria from environmental samples bacteria are limited by the availability of organic carbon sub-
remains unknown. However, it has recently been demonstrated strates in some soils despite the much higher total organic
that addition of cAMP at low concentrations (10 mM) in certain carbon concentrations in these environments (Alden et al.
cases can significantly increase the culturability for natural sam- 2001). Commensurate with this situation, strongly diluted
ples (Bruns et al. 2002). nutrient broth yielded a significantly improved culturability of
soil bacteria (Janssen et al. 2002; Koch et al. 2008; Dedysh 2011).
Media supplemented with comparatively low concentrations of
The Viable but Nonculturable State complex carbon sources like yeast extract and peptone (i.e.,
0.25 g L1 each or less) have been established for the isolation
In addition, it has been proposed that certain bacteria may of fastidious bacteria such as Planctomycetes (Schlesner 1986;
acquire a state termed ‘‘viable but nonculturable’’ (VBNC; Xu Staley et al. 1992), for Armatimonadetes (the former OP10
et al. 1982; Roszak and Colwell 1987). In the proposed VBNC phylum; Tamaki et al. 2011), and for novel types of freshwater
state, bacterial cells, especially of pathogenic bacteria like Vibrio sphingomonads (Chen et al. 2012). Using concentrations of
cholerae or Campylobacter jejuni (Bovill and Mackey 1997), are 5 mg of organic carbon·L1, which are supplied as casamino
believed to become temporarily nonculturable until they are acids, or employing straight filtered autoclaved seawater can
exposed to an environment that stimulates their resuscitation. increase the apparent cultivation success (determined as the
In Vibrio vulnificus, exposure of cells to temperatures around ratio of most probable numbers in relation to total cell counts)
4 C results in a decline of culturability without cell lysis (Oliver up to 20–60% (Button et al. 1993; Eguchi et al. 2001) and has
1995). Temperature upshift results in a resuscitation of these also allowed the isolation of first strains of the abundant marine
nonculturable cells (Whitesides and Oliver 1997). Those cells photoautotroph Prochlorococcus marinus (Chisholm et al. 1992).
that retain the capacity for resuscitation appear to maintain Subsequently, a combination of miniaturization of culture vol-
physiological activity. Over 35 species belonging to 17 umes, molecular screening techniques, and clean techniques for
proteobacterial genera have been reported to show this phenom- handling of seawater (Connon and Giovannoni 2002) allowed
enon (McDougald and Kjelleberg 1999). Replacement of the the isolation of a considerable number of heterotrophic marine
term VBNC by ‘‘temporarily nonculturable’’ or ‘‘not immedi- oligotrophs (Rappe et al. 2002; Cho and Giovannoni 2004).
ately culturable’’ has been proposed since it is unclear whether Alternatively, microencapsulation (Zengler et al. 2002) or sepa-
VBNC is the result of a specific programmed differentiation or ration of cells from the surrounding environment in diffusion
an adaptive process (Barer and Harwood 1999). The view that chambers (Kaeberlein et al. 2002) or microbial culture chips
the VBNC state is a single phenomenon and a valid operational (Ingham et al. 2007) allow an exchange of natural substrates at
concept has been challenged (Barer and Harwood 1999). in situ concentrations and may provide initial enrichments of
interesting novel bacteria.
In addition, the qualitative composition of carbon sources
Lysogenic Phages determines the cultivation success of naturally occurring pro-
karyotes. Complex media have been shown to yield higher
Lysogenic phages may be induced upon plating onto agar media numbers and a higher diversity of isolates than similar media
by SOS response-related induction (Barer and Harwood 1999), containing only one defined carbon source at the same concen-
leading to a rapid decrease of culturability. tration: single substrates yielded almost exclusively strains of the
gamma-subclass of Proteobacteria (Uphoff et al. 2001). For soil
pseudomonads, nutrient-poor media containing only
Low Nutrient Concentrations, ‘‘Oligotrophic 15 mg C·L1 have been shown to yield higher numbers of
Bacteria,’’ and ‘‘Ultramicrobacteria’’ culturable cells than conventional organic-rich media (Aagot
et al. 2001). Strongly diluted nutrient broth resulted in signifi-
The concentration of nutrients critically affects the outcome of cantly increased numbers of Acidobacteria, Actinobacteria,
cultivation experiments (Pinhassi and Berman 2003). Whereas a-Proteobacteria, low G + C Gram-positive bacteria, and
early in the history of microbiology the ionic of media were Verrucomicrobia from soil samples (Janssen et al. 2002). This
adjusted to concentrations commensurate with the natural envi- combined evidence suggests that the majority of heterotrophic
ronment, carbon-poor (‘‘oligotrophic’’) media have only prokaryotic cells in many natural samples require lower concen-
recently been used to tap the large diversity of bacterioplankton trations of organic carbon for growth. The cultivation success of
by systematic cultivation approaches. As an example, the con- freshwater planktonic bacteria could also be increased in media
centrations of all major elements except carbon, nitrogen, and with reduced nitrogen and phosphorus concentrations. In the
phosphorous in marine broth 2,216 that was established in 1941 latter case, nutrient levels were decreased to match the maximal
(ZoBell 1941) mimic the average concentrations in seawater very ambient levels in the planktonic environment (50 mM nitrogen
and 1.5 mM phosphorus; Eilers et al. 2001). In general, viable cell high-nutrient microbiological media. Oligotrophic bacteria are
counts of freshwater or marine planktonic bacteria reached on defined as those that on first cultivation develop on media
agar-solidified media are orders of magnitude smaller than those containing 1–15 mg C·L1 (Kuznetsov et al. 1979). Conse-
obtained as most probable numbers in liquid extinction dilution quently, such bacteria cannot be cultivated directly on nutri-
series (Eguchi et al. 2001; Jaspers et al. 2001). ent-rich agar (Vancanneyt et al. 2001). Actually, the range of
Ultramicrobacteria are defined by a cell volume of <0.1 mm3 carbon concentrations reported for the pelagic and deep ocean is
(Eguchi et al. 2001) and prevail in oligotrophic marine waters 30–200 mM, corresponding to 0.36–2.4 mg C·L1 (Jannasch
(Schut et al. 1997). In this respect, an important question is et al. 1996). Therefore, bacto yeast extract (Difco) is added
whether ultramicrocells represent a distinct class of prokaryotes from autoclaved stock solutions to a final concentration of
that maintain their small biovolume independently of their 10–1 mg·L1 (corresponding to 3.3–0.33 mg C·L1 or 270–
growth status or whether they are capable of becoming larger. 27 mM carbon; Jannasch et al. 1996) in media designed to
In many bacterial species, nutrient deprivation results in pro- recover marine pelagic prokaryotes. It has been demonstrated,
nounced changes of cell morphology and size (Torella and however, that prolonged incubation over 1 year at 5–8 C or
Morita 1981; Morita 1982; Amy and Morita 1983; Kjelleberg storage and monthly reculturing at 15 C can yield cultures
et al. 1993). Upon starvation, many nongrowing marine isolates which eventually can multiply on high-nutrient laboratory
produce ultramicrocells by reductive division, and these can have media (Eguchi et al. 2001; Vancanneyt et al. 2001). Obligate
volumes as little as 1% that of rapidly growing cells. The resulting oligotrophs are prokaryotes that typically fail to grow at high
daughter cells retain measurable metabolic activity, at least in an substrate concentrations even after multiple passages in the labo-
initial period following their formation (Amy and Morita 1983). ratory. Examples are members of the alphaproteobacterial SAR11
It appears likely that such ultramicrocells contribute signifi- clade or of the gammaproteobacterial OMG clade (Rappe et al.
cantly to observed populations of prokaryotes in ocean waters. 2002; Cho and Giovannoni 2004). In contrast, facultative
However, in cells of other species, like for instance oligotrophs such as Sphingomonas alaskensis RB2256 can
Sphingomonas alaskensis within the a-Proteobacteria (Eguchi (eventually) be cultivated on standard microbiological media.
et al. 2001; Vancanneyt et al. 2001) or the Verrucomicrobiales, Oligotrophic bacteria may represent novel types of bacteria
a small cell volume (0.03–0.06 mm3) is a stable characteristic and with unknown physiological properties. Indeed, the recently
independent of substrate concentrations (Schut et al. 1993; described ammonia-oxidizing archaeon ‘‘Candidatus
Janssen et al. 1997). Freshwater Actinobacteria are ultramicro- Nitrosopumilus maritimus’’ exhibits the highest affinity for
bacteria that have recently been successfully enriched based on ammonia reported to date (Martens-Habbena et al. 2009).
their cell size, employing 0.2-mm-pore-size filters in a Notably, this archaeon is inhibited in growth at ammonia con-
prefiltration step to eliminate accompanying, faster-growing centrations of 2 mM. Another example is the presence of
copiotrophic bacteria (Hahn et al. 2003). For successful labora- a novel pathway for assimilation of dimethylsulfoniopropionate
tory cultivation of these bacteria, an acclimatization procedure in ‘‘Candidatus Pelagibacter ubique’’ and other marine plank-
was employed that encompassed an increase of the temperature in tonic bacteria (Reisch et al. 2011). In contrast to all other
2 C-steps to the final incubation temperature of 15 C and step- (mostly copiotrophic) bacteria investigated to date, the marine
wise increases of the concentrations of organic carbon substrates ultramicrobacterium Sphingomonas alaskensis RB2256
in the medium were from 5 mg to 1 g L1. The freshwater exhibits a high level of inherent stress resistance toward oxidative
Actinobacteria so far could be grown only slowly and as cocul- stress (hydrogen peroxide [H2O2]), but no starvation-induced
tures with other heterotrophic bacteria on agar plates, forming stress against H2O2 (Ostrowski et al. 2001). Stress resistance is
very small colonies (Hahn 2009). Ultramicrobacteria form negatively correlated to growth rate, but only marginal changes
a dominant fraction of freshwater and marine bacterioplankton in catalase activity were observed; hence, other factors must be
where they represent typical oligotrophs. Another abundant critical to the stress resistance of this strain.
group of obligate freshwater ultramicrobacteria are members of
the betaproteobacterial Polynucleobacter necessarius cluster that
are facultative oligotrophs (Hahn 2003). Small cell size does not Growth on Multiple Substrates
always correlate with ultraoligotrophic characteristics, however.
The recent isolation of the first representative of the termite group In spite of the fact that bacteria in natural environments grow in
1 phylum (now Elusimicrobia) was enabled by selecting for the presence of a large diversity of compounds, most laboratory
ultramicrobacteria through filtration of beetles gut homogenate studies have focused on growth with single substrates. When
through 0.2-mm-pore-size membrane filters but employing bacteria are grown in batch culture with more than one growth
anoxic cultivation media with standard concentrations (i.e., substrate, sequential utilization of these substrates is often
2 mM) of organic carbon substrates (Geissinger et al. 2009). observed. In some cases, this results in a typical diauxic growth
Although debated for a considerable time (Morita 1982), it pattern (Monod 1942; Lengeler et al. 1999) in which the sub-
was eventually recognized (Kuznetsov et al. 1979) that certain strate that is used first represses the synthesis of enzymes
bacteria are capable of replicating at nutrient concentrations required for the utilization of the other substrates. Only after
present in seawater but cannot grow in the standard (almost) complete utilization of the first substrate is growth on
the second one induced. It is noteworthy that sugars may cause bacteria will in most cases be missed when batch-type enrich-
catabolite repression of inducible enzyme systems even if the ment techniques are used, even if mixtures of several substrates
sugars are not utilized themselves (Pastan and Perlman 1969). are present. Instead, chemostat cultures need to be employed for
In enteric bacteria, cAMP is part of the pleiotropic crp activation enrichment.
system which regulates most peripheral catabolic operons and
carbohydrate transport systems, and it typically mediates carbon
catabolite repression (Lengeler et al. 1999). Sugars transported Effect of Inhomogeneities
by the phosphotransferase system (PTS) decrease intracellular
cAMP levels which in turn prevents the expression of genes Inhomogeneities can have a profound effect on the physiology of
necessary for the uptake of non-PTS sugars. After depletion of prokaryotes. Many environments such as soils, sediments, and
glucose, intracellular cAMP concentrations rise from 0.3 up to marine snow are highly heterogeneous, and therefore microbial
3 mM, and other sugars can be taken up and metabolized after substrates and bacterial productivity are distributed in micro-
a lag period required for the induction of the necessary genes. scale patches of variable concentration and size (Azam 1998;
This regulatory pattern leads to a biphasic growth curve. The Ploug et al. 1999). Under such conditions, steep gradients, for
intermediate lag phase can be abolished by addition of extracel- example, of molecular oxygen, may result in a close proximity of
lular cAMP when added at millimolar concentrations (Epstein aerobic and anaerobic species and transformations (Revsbech
et al. 1975), whereas the growth rate itself does not change and Jørgensen 1986).
(Okada et al. 1981). The involvement of cAMP in regulation of Attachment of prokaryotic cells has been shown to be of
catabolic enzymes has also been demonstrated for a wide range significance under some conditions. Some compounds are
of nonenteric bacteria, including other g-Proteobacteria, the optimally metabolized only if oxygen-dependent and strictly
a-Proteobacteria, b-Proteobacteria, and the cyanobacteria anoxic steps are coupled by diffusion of metabolic products
(Botsford and Harman 1992). and thus proceed in close proximity. This principle has been
However, this type of clear-cut diauxic pattern with a lag demonstrated in an elegant study of the degradation of DDT
period between consumption of the first and the second sub- (1,1,1-trichloro-2,2-bis[4-chlorophenyl]ethane), using calcium
strate is by no means very common. More often, a gradual alginate beads for immobilization of bacteria (Beunink and
transition between the use of two (or more) substrates is Rehm 1988). In this system, the attached cells of Alcaligenes
observed, and in some cases, no enzyme repression is evident species and Enterobacter cloacae performed reductive dechlo-
at all (Harder and Dijkhuizen 1982; Gottschal 1986). Moreover, rination of DDT inside the beads, and the partly dechlorinated
it is questionable whether distinct preferences for one substrate products were metabolized oxidatively. Since highly chlorinated
would be functionally valuable in nutrient-poor natural envi- compounds are dehalogenated more readily under anoxic con-
ronments with many different substrates available at the same ditions, this same principle may hold for many other haloge-
time. Under nutrient limitation, one would rather expect organ- nated xenobiotics as well.
isms to develop physiological strategies enabling them to make Besides affinity for a given substrate and cell yield, attachment
use of several nutrients simultaneously. Growth in chemostats to solid surface is an additional determinant of the outcome of
under limitation of mixtures of different substrates has provided competition between prokaryotes with similar physiology.
ample evidence that a multitude of substrates, serving similar Attachment to solid surfaces can thus change the metabolic inter-
physiological functions, can be growth limiting at the same time relations between competing bacteria. When growing in suspen-
(Harder and Dijkhuizen 1982; Egli et al. 1983; Gottschal 1986; sion, Pseudomonas putida R1 and Acinetobacter C6 compete
Gottschal and Dijkhuizen 1988). Low Ks values have been found for the substrate benzyl alcohol, with Acinetobacter largely
in marine isolates (Schut et al. 1995). However, the reported Vmax outcompeting P. putida. However, both strains formed stable
values are generally so low that specific affinities and oligotro- and structured biofilms when glass as solid substrate was avail-
phic capacities are insufficient to allow growth on single sub- able (Christensen et al. 2002). Under the latter conditions,
strates. In a marine coryneform bacterium, the presence of P. putida is capable of utilizing benzoate which Acinetobacter
amino acids can enhance the uptake of glucose and lowers the uses only inefficiently and therefore excretes to a large extend.
threshold concentrations for growth (Law and Button 1977). On In the natural environment, nonpolar organic carbon sub-
the other hand, growth on alanine of Sphingomonas alaskensis strates occur in the adsorbed state. In the adsorbed state, these
strain RB2256 occurred with a lower affinity in the presence of substrates are often not directly available, and rates of desorp-
glucose (Schut et al. 1995). tion control the rate of degradation. However, some prokaryotes
Metabolically versatile bacteria exist that are specialized in can degrade adsorbed substrates more rapidly than can be
using several different substrates at the same time if the latter are accounted for by the rates of desorption into the aqueous
present at growth-limiting concentrations (Laanbroek et al. phase (Harms and Zehnder 1995). Prokaryotes thus differ in
1979; Dykhuizen and Davies 1980; Gottschal and Kuenen their capability of using adsorbed substrates (Guerin and Boyd
1980; Beudeker et al. 1982; Legan and Owens 1988). Since 1997; Stringfellow and Aitken 1994; Crocker et al. 1995). Con-
these prokaryotes usually display lower maximum growth rates sequently, enrichment strategies, in which especially nonpolar
relative to more specialized species, metabolically versatile substrates are offered in the adsorbed state, may provide a more
relevant low-bioavailability environment, and hence lead to the various physicochemical factors but also by interactions with
enrichment and isolation of novel types of prokaryotes (Tang other microorganisms. Cells may need to communicate with
et al. 1998; Grosser et al. 2000). each other for growth. The metabolic interaction between the
carbohydrate-fermenting Streptococcus gordonii and the lactic
acid-fermenting Veillonella atypica not only involves the
Interactions with Other Microorganisms exchange of lactic acid but also involves signaling between
both partners that leads to an increased expression of a-amylase
Most likely, one considerable problem in current enrichment by S. gordonii. In an open flow-through system (similar to
and cultivation techniques is that microbial interactions cannot human oral biofilms which contain both bacterial species), the
be reproduced adequately (see > ‘‘Alternative and Novel Con- diffusible signal functions only over short distances on the order
cepts for Cultivation’’). In nature, prokaryotes reach cell densi- of 1 mm (Egland et al. 2004). These experimental results empha-
ties of about 106 mL1 in most aquatic environments, 109 cm3 size the relevance of spatial organization for the metabolic activ-
in sediments, and 1011 cm3 in soils. If a homogenous distribu- ity and mutual control of bacteria.
tion of the cells is assumed, the average cell-to-cell distance at So far, the compounds known to be exchanged between
these increasing densities would amount to 112, 10 and 1 mm, prokaryotes that can promote growth of otherwise unculturable
respectively. Over such small distances, transport of small mol- bacteria include signal compounds (as in the case of quorum
ecules by molecular diffusion proceeds at a rapid rate and takes sensing) or cAMP (Bruns et al. 2002) or possibly oligopeptides
between microseconds and a few seconds (Overmann 2002a). (Nichols et al. 2008), growth factors such as vitamins (Graber
Owing to the laws of three-dimensional diffusion, the flux of and Breznak 2005), siderophores (Guan et al. 2000), essential
metabolites experienced in the vicinity of a prokaryotic cell by nutrients (Tripp et al. 2008), carbon sources (Brown and
another one decreases to as little as 0.01% when the cell-to-cell Whiteley 2007), and other compounds directly involved in
distance increases to 10 mm (Overmann and Schubert 2002). energy metabolism (mostly electron donors/electron acceptors
This simple calculation indicates that (1) interactions between such as hydrogen or inorganic sulfur compounds in syntrophic
prokaryotes may influence their growth under natural conditions interactions). Signaling compounds also comprise small pep-
and most likely need to be considered in cultivation attempts of tides such as the 5-amino-acid peptide LQPEV that at 3.5 nM
some not-yet-cultured bacteria and (2) a strong selective pressure induces growth of a Psychrobacter strain on standard media
must exist for interacting prokaryotic cells to maintain close (Nichols et al. 2008). Besides enabling prokaryotes to perform
spatial proximity. novel syntrophic reactions, interactions can also lead to altered
At high cell densities, even monospecific associations of kinetics of microbial transformations. For example, the high-
some bacteria exhibit physiological traits that differ from those affinity oxidation of methane observed in soil could be
of dilute cultures. In the case of quorum sensing, the excretion of reproduced in a coculture of a nonnovel methanotroph with
autoinducer molecules signals high cell density and triggers light a Variovorax strain (Dunfield et al. 1999). This finding indicates
production, expression of virulence factors, and swarming that the physiology and ecology of prokaryotes can only be
(Fuqua and Greenberg 1998; Basler and Losick 2006) or prevents completely appreciated if interactions with other prokaryotes
cell aggregation (Puskas et al. 1997). Similarly, myxobacteria are considered as well. The experimental setup to exploit inter-
exhibit complex social interactions: when deprived of nutrients, actions with living accompanying bacteria for the cultivation of
they enter into a complex developmental cycle that results in the novel types of bacteria is typically based on diffusion chambers
formation of a multicellular fruiting body that contains (Kaeberlein et al. 2002; Nichols et al. 2008; > Fig. 7.5, see section
myxospores (Reichenbach 1984). Other cases of monospecific > ‘‘Cocultivation’’). An alternative approach is the microencap-
associations have been described occasionally (Lins and Farina sulation technique (Zengler et al. 2002).
1999). However, it has also been found that different types of Overall, eight types of interspecies interrelationships can be
bacteria can form associations. These comprise highly struc- distinguished based on the effect of each of the two populations
tured associations of defined composition, so-called consortia (Atlas and Bartha 1993; > Table 7.7), ranging from mutualistic
(Schink 1991; Overmann 2002a). In addition, less-structured to antagonistic interactions. It has to be kept in mind, however,
assemblages, like microcolonies, netlike structures, biofilms, or that the different categories listed in > Table 7.7 represent only
aggregates of up to 18 different prokaryotic genera, have com- concepts, which in many cases fail to account for all facets of
monly been observed in natural samples (Weise and microbial interactions.
Rheinheimer 1978; Paerl 1982; Alldredge and Youngbluth
1985; van Gemerden et al. 1989; Dubinina et al. 1993;
Kolenbrander and London 1993; Seitz et al. 1993; Overmann Neutralism
et al. 1996; Jacobi et al. 1997; Moissl et al. 2002). Least structured
and hence difficult to detect are 10–100-mm-large patches of This describes the situation of a complete lack of interaction
free-living bacterial cells, but evidence has accumulated for the between two populations. This situation is more likely to occur
existence of such inhomogeneities in the pelagic habitat between populations with very different metabolic capabilities.
(Krembs et al. 1998). Under certain natural conditions, the However, neutralism is defined in a negative way and therefore is
growth of prokaryotes is therefore not only influenced by the most difficult to verify experimentally. Most likely interactions
dialysis membrane Competition

on teflon support
Since energy and nutrient sources are often present in limiting

concentrations, competition for growth-limiting resources is
screw cap one of the major types of interactions between cells of one
population or between different prokaryotes, and it results in
a reduction of growth rate. Eventually, it may lead to the exclu-
sion of the slower-growing species, a process also termed ‘‘com-
petitive exclusion’’ (Gause 1934; Hardin 1960). Since the
introduction of continuous culture techniques which allowed
cultivation under conditions of permanent nutrient limitation,
the competition in mixed cultures has been studied under a great
variety of environmental conditions (Powell 1958; Veldkamp
and Jannasch 1972; Fredrickson 1977; Veldkamp 1977;
glass Fredrickson and Stephanopoulos 1981; Kuenen and Gottschal
culture 1982; Kuenen and Harder 1982; Gottschal and Dijkhuizen 1988;
vessel Visscher et al. 1992b). These studies were mostly concerned with
. Fig. 7.5 simple and pure competition for single growth-limiting nutrient
A simple dialysis coculture setup that permits the growth of two in the absence of other interactions, and the results supported
separate prokaryotic cultures while contact via a dialysis the competitive exclusion principle. The outcome of the com-
membrane is maintained. For growth of aerobic cultures, the petition between two or more species is merely dependent on the
screw caps are replaced by cotton plugs and the glass vessels are shape of the m versus S relationship of the competitors
only half full. Homogeneous cell suspensions can be maintained (> Fig. 7.2). Those prokaryotes which reach a higher specific
by adding a small stirring bar to each of the two chambers and growth rate at a given substrate concentration always
placing the set up on a magnetic stirrer outcompete the slower-growing species. In some instances, the
m versus S curves cross, and as a result, the outcome of the
competition depends on the dilution rate chosen. Apparently,
. Table 7.7
certain species are much better adapted to growth at very low
Types of interspecies interactions
substrate concentrations (and exhibit a relatively low mmax
value), whereas others are more specialized in growth at high
Name of rates in the presence of high substrate concentrations. Examples
interaction Effect of interaction have been reported for aerobic and anaerobic heterotrophs,
Neutralism Neither population affects the other chemolithotrophic species, and phototrophic organisms
Competition Populations inhibit each other when (Harder and Veldkamp 1971; Jannasch and Mateles 1974;
resources are in limiting supply Fredrickson 1977; Harder et al. 1977; Mur et al. 1977; Veldkamp
Amensalism Population 1 is negatively affected by 2, but
1977; Matin and Veldkamp 1978; Kristjansson et al. 1982; Lovley
2 is not affected by 1 et al. 1982; Laanbroek et al. 1983, 1984; King 1984; Kuenen and
Robertson 1984; Robinson and Tiedje 1984; Veldkamp et al.
Parasitism Population 1 consumes population 2 but
usually not in a destructive manner 1984; Legan et al. 1987; Legan and Owens 1988). Competition
between microorganisms exists not only for organic carbon but
Predation Population 1 consumes population 2 in
a destructive manner
also for nutrients like phosphate (Rhee 1972; Currie and Kalff
1984).
Commensalism Population 1 benefits from population 2
without affecting it in a negative sense
Synergism Both populations benefit from the
Amensalism
(protocooperation) interaction, which is not obligatory
Mutualism Both populations benefit from the
The production of lactic acid by lactic acid bacteria and sulfuric
(symbiosis) interaction, which is obligatory
acid by Thiobacillus thiooxidans, which inhibit other non-
acidophilic microorganisms, are examples of amensalistic rela-
tionships. Similarly, the inhibition of Salmonella enterica by
are absent at very low population densities and if prokaryotes acetate and propionate produced by Clostridium
form physiologically largely inactive resting stages (Atlas and lactatifermentans at low pH (5.8) has been demonstrated (van
Bartha 1993). However, it has been demonstrated that even der Wielen et al. 2002). At these low pH values, considerable
endospores can affect the surrounding environment by the pres- acetate and propionate present as undissociated acids penetrate
ence of extracellular enzymes (i.e., Mn2+ oxidase; Francis and the cytoplasmic membrane and hence decrease the membrane
Tebo 2002). potential (see section > ‘‘pH’’).
Production of inhibitory compounds has been observed Commensalism

more frequently among particle-attached marine bacteria
than their free-living counterparts and may be a mechanism In the nonobligatory commensal relationship, one population
to deter other potential colonizers from the nutrient-rich par- benefits, for example, from growth factors excreted by a second
ticle environment (Long and Azam 2001). Members of the population, while the latter remains unaffected. This type of
marine Roseobacter clade are dominant primary colonizers of relationship thus is unidirectional in character. Commensalism
surfaces in the coastal environment (Dang and Lovell 2000) and within one and the same population is of significance if, for
show the strongest antagonistic activity (Long and Azam 2001). instance, the growth substrates are insoluble (such as lignin or
The marine antagonistic Phaeobacter strain 27-4, a member of cellulose). These substrates are made available by extracellular
the Roseobacter clade, has been found to synthesize enzymes, and while the exoenzymes themselves are kept at the
tropodithietic acid and thiotropocin that inhibit Vibrio cell surface (i.e., in the periplasmic space in the case of Gram-
anguillarum and Vibrio splendidus (Bruhn et al. 2005) and negative bacteria or by attachment to the cytoplasmic mem-
hence may act as a probiotic against infection of fish larvae in brane in the case of Gram-positives), the resulting substrates
commercial fish farms. often are rapidly lost by diffusion from the vicinity of single cells.
In contrast, soluble products can be utilized at high efficiency if
cell densities are high. As an example, Myxococcus xanthus does
Predation not grow on insoluble casein at cell densities lower than
103 mL1, whereas growth rates increased with cell densities
The best-documented example for predatory prokaryotes is above this value (Rosenberg et al. 1977). This effect of cell
bacteria of the genus Bdellovibrio which attack Gram-negative density is not observed on prehydrolyzed casein.
bacteria by attaching to their prey, penetrating the cell wall, Not in all cases does the limitation of two populations of
and subsequently multiplying within the periplasmic space. prokaryotes by one substrate lead to competitive exclusion, but
Multiplication of Bdellovibrio sp. occurs at the expense of it can result in stable mixed cultures. The underlying reasons
cellular components of the host cell and leads to the formation include the occurrence of additional interactions, especially
of 4–20 daughter cells by segmentation within 2–3 h after commensalism and mutualism, the formation of self-inhibitory
infection (Stolp and Starr 1963, 1965; Varon and Shilo 1980; products, the presence of predators, selective adhesion, fluctua-
Shilo 1984). The cell content of the host is partially degraded and tions in physical parameters (pH, temperature, light, and oxic/
utilized. Although growth and survival of wild-type anoxic conditions), or an alternating supply of differing growth-
Bdellovibrio spp. is strictly dependent on the availability of limiting substrates (Bungay and Bungay 1968; Megee et al. 1972;
appropriate prey cells, this parasite differs fundamentally from Jost et al. 1973; Meers 1973; van Gemerden 1974; Meyer et al.
viruses in that it does not depend directly on the metabolic 1975; Lee et al. 1976; Fredrickson 1977; de Freitas and
machinery of the host cell. Bdellovibrio species and similar Fredrickson 1978; Slater and Bull 1978; Gottschal et al. 1979;
bacteria appear to be widespread, having been isolated from Miura et al. 1980; Bull and Slater 1982; Kuenen and Robertson
many different aquatic and terrestrial ecosystems (Varon and 1984; Kuenen et al. 1985). For example, purple sulfur bacteria
Shilo 1980; Burnham and Conti 1984). Additional predatory and colorless sulfur bacteria, which are expected to compete for
bacteria that require cell contact have been described. H2S as electron-donating substrate, are found to thrive in high
Vampirococcus sp. is a nonmotile, Gram-negative anaerobic population densities in the same layer of microbial mats
bacterium which occurs as an epibiont of phototrophic (Visscher et al. 1992a). This has been explained by the removal
Chromatium spp. to which it adheres by means of specific of oxygen by colorless sulfur oxidizers and the formation of
attachment structures without penetrating the outer cell layers. incompletely oxidized inorganic sulfur intermediates, which
Concomitant to growth and division of the epibiont, the host then serve as alternative electron donor for the anoxygenic
cell cytoplasm is degraded, leaving behind an almost empty cell phototroph (Visscher et al. 1992b; van den Ende et al. 1996).
envelope (Guerrero et al. 1986). A second type of predatory Also, competition can be alleviated and stable cocultures
bacterium, Daptobacter sp., penetrates the cell envelope of obtained if conditions for competition do not exist over an
cells of various genera of the Chromatiaceae and degrades entire 24-h period but only for a shorter time interval (van
the cytoplasm of its prey. The Gram-negative Daptobacter Gemerden 1974). This fact for instance can be exploited for
is a facultative predator, facultatively anaerobic and motile the enrichment of more fastidious purple sulfur bacteria (like
(Guerrero et al. 1986). Other bacteria can lyse prokaryotes the large-celled Chromatium spp.) by using light/dark cycles
without direct contact. An isolate of Stenotrophomonas instead of continuous light.
maltophilia was found to lyse cells of Chlorobium limicola Commensalism between nonrelated bacteria has been
and several heterotrophic bacteria (Nogales et al. 1997). detected in a wide variety of cases. In freshwater bacterioplankton,
Finally, myxobacteria can cause lysis of susceptible strains at chitinolytic Flavobacteria colonize chitin particles which they
some distance, apparently with the aid of exoenzymes. The hydrolyze and solubilize, thereby supporting uptake and utiliza-
myxobacteria derive their nutrition from material released by tion of N-acetyl-D-glucosamine by Actinobacteria of the ubiqui-
the lysed cells. tous freshwater AcI cluster (Beier and Bertilsson 2011).
As another example, Propionibacterium shermanii can photolithoautotrophic partner is often less evident and may
grow in mixed continuous culture at the expense of lactate comprise the supply of vitamins (Jones 1982), the reduction of
produced by Lactobacillus plantarum from glucose, the oxygen concentration around heterocysts (Bunt 1961; Paerl
growth-limiting nutrient in the chemostat (Lee et al. 1976). 1978, 1982)—which has been disputed on physical grounds
Lactate was also the mediator in a commensal relationship (Overmann 2002a)—and the formation of CO2 (Lange 1971).
between Streptococcus mutans and Veillonella alcalescens. A reciprocal relationship was also suggested in a mesophilic
Both strains are commonly found in dental plaque and were estuarine microbial mat system in which the cyanobacterium
shown to coexist in mixed continuous cultures supplemented Microcoleus chthonoplastes excreted organic matter that stim-
with glucose as the limiting nutrient (Mikx and van der Hoeven ulated growth of Thiocapsa roseopersicina, which in turn
1975). In this mixed culture, the lactate produced by S. mutans prevented accumulation of excess amounts of hydrogen sulfide
was metabolized to acetate, propionate, and ethanol, which, (de Wit and van Gemerden 1988).
owing to lower dissociation constants, might reduce the demin- Many different types of interspecies interactions occur in
eralization of tooth enamel. Also, consumption of lactate by laminated microbial ecosystems, or so-called microbial mats,
a second organism (Pseudomonas stutzeri) under anaerobic which are found on seashores, estuarine areas, salt marshes,
conditions and in the presence of nitrate in a lactose-limited and along the effluents from geothermal springs. Metabolic
mixed chemostat culture results in a marked stimulation of the interactions include (1) the excretion of photosynthates and
growth yield of Lactococcus cremoris (Otto et al. 1980). The cell lysis of cyanobacteria as the primary colonizers, (2) respira-
latter is capable of an electrogenic lactate export during which tion of the organic substances by heterotrophic bacteria,
two H+ are translocated across the cytoplasmic membrane thereby generating anoxic conditions, (3) fermentation and
together with one lactate molecule, thereby generating reduction of sulfate by anaerobic bacteria, and (4) oxidation of
a membrane potential. Since the driving force of this transport the reduced sulfur compounds by anoxygenic phototrophs and
is the transmembrane lactate gradient, the stimulatory interac- chemolithotrophs (Stal et al. 1985). In addition, a multitude of
tion is based on the very low external lactate concentration interactions, most of them beneficial, occur between the differ-
maintained by Pseudomonas. This enables S. cremoris to gain ent groups of prokaryotes (Overmann and van Gemerden 2000).
more energy from the efflux of lactate than at higher external In microbial mats, sulfate-reducing bacteria can reach their
concentrations (Michels et al. 1979). Commensalistic interac- highest abundance at the surface, rather than at the deeper,
tions are also established in mixed cultures where either stimu- permanently anoxic sediment layers. Accordingly, the sulfate-
latory compounds are formed or inhibitory compounds are reducing bacterium Desulfovibrio desulfuricans has been found
being removed. As an example of the removal of inhibitory to be capable of syntrophic growth with the colorless sulfur-
compounds, methane-consuming pseudomonads grew in oxidizing bacterium Thiobacillus thioparus under oxygen lim-
a mixed culture with a Hyphomicrobium species, the latter itation and with lactate as the electron-donating substrate (van
removing small inhibitory amounts of methanol formed during den Ende et al. 1997). In this association, Thiobacillus removed
methane oxidation (Wilkinson et al. 1974). molecular oxygen inhibiting Desulfovibrio and provided the
latter with soluble polysulfide compounds as electron acceptor.
Desulfovibrio in turn oxidized lactate to acetate, forming sulfide
Synergism concomitantly, which served again as electron-donating sub-
strate of Thiobacillus.
Synergism (protocooperation) occurs between Lactobacillus
arabinosus and Enterococcus faecalis in a minimal medium
(Nurmikko 1956) in which each organism is unable to grow Syntrophic Interactions
on its own. The synergism is based on the fact that S. faecalis
requires folic acid, which is produced by Lactobacillus, whereas Under certain conditions, synergism between different groups of
the latter requires phenylalanine, which is produced by Strepto- anaerobic prokaryotes is essential for the degradation of organic
coccus. Reciprocal stimulation also can occur in yogurt, where matter. In contrast to aerobic decomposition, anaerobic miner-
Lactobacillus bulgaricus produces amino acids that stimulate alization involves the participation of different groups of
the growth of Streptococcus thermophilus. The latter produces prokaryotes with only a limited physiological flexibility. Conse-
small quantities of formic acid, which in turn stimulates quently, a tight cooperation of mixed populations exists. In this
L. bulgaricus (Driessen 1981). Chemolithotrophic and multistep process, polymeric organic matter (such as cellulose,
phototrophic organisms growing with CO2 as carbon source proteins, and lipids) is first hydrolyzed to oligomers and mono-
are known to excrete substantial quantities (>50%) of the mers that are subsequently fermented (Hungate 1960; Bryant
carbon fixed as carbohydrates, peptides, amino acids, lipids, 1976; Laanbroek and Veldkamp 1982; Wolin 1982; Nedwell
or vitamins (Clark and Schmidt 1966; Czeczuga 1968; Fogg 1984). The first step in anaerobic degradation of sugars, most
1971; Cohen et al. 1979; Jones 1982; Coveney 1982; Wolter amino acids, and other readily fermentable substrates is the
1982; Soendergaard et al. 1985; Bateson and Ward 1988). formation of H2, CO2, formate, alcohols, acetate and other
In these interactions, the benefit for the chemo- or short-chain fatty acids, sulfide, and ammonia. In the presence
of oxidized sulfur compounds, further mineralization may pro- and CO2 (instead of to ethanol) if cocultured with Wolinella
ceed directly through activity of sulfate-reducing bacteria, succinogenes, which uses hydrogen as electron donor in the
which, as a group, can directly oxidize the (long-chain) fatty reduction of fumarate to succinate (Ianotti et al. 1973). Hydro-
acids, amino acids, alcohols, aromatic compounds, and hydrogen transfer results in an additional energy gain for R. albus
gen (Widdel 1988). In the absence of sulfate, CO2 will serve as as more acetyl-CoA is converted to acetate in an ATP-
the major electron acceptor resulting in the formation of meth- yielding route. Shifts in fermentation products were also
ane. Methanogenic bacteria, however, have a very narrow range demonstrated for cocultures of Clostridium thermocellum
of substrates (mainly H2, acetate, formate, methanol, and several and Methanobacterium thermautotrophicum (Weimer and
methylamines; Oremland 1988). Hence, an additional group of Zeikus 1977).
prokaryotes, the ‘‘acetogenic’’ bacteria, are involved in the for- A second type of syntrophic interaction has been found in
mation of suitable methane precursors. In the absence of sulfate, cocultures of green sulfur bacteria with sulfur- or sulfate-
the subsequent fermentative degradation of some of the prod- reducing bacteria (Wolfe and Pfennig 1977; Biebl and Pfennig
ucts like propionate, butyrate, or benzoate to hydrogen, CO2, 1978; Warthmann et al. 1992). Under conditions of limitation by
and acetate is an endergonic process under standard conditions, inorganic sulfur compounds as they prevail in freshwater eco-
however. A syntrophic relation, the so-called interspecies systems, the sulfide-producing bacterium relies on the activity of
hydrogen transfer, is established between hydrogen-producing the sulfide-oxidizing phototroph for the availability of the elec-
bacteria (also called ‘‘obligate proton-reducing bacteria’’) on tron acceptor, while the activity of the green sulfur bacteria is
one hand and hydrogen-consuming prokaryotes such as controlled by the activity of the sulfide-producing organism. In
methanogens on the other. this way, a closed sulfur cycle is established through which each
Anaerobic chemotrophic syntrophic cocultures are the only sulfur atom cycles many times (Pfennig 1980).
associations of prokaryotes that so far have been investigated in
sufficient detail to understand the physiological basis of the
interaction (Schink 1991). Interspecies hydrogen transfer in Symbiosis
these cocultures is decreased to values below 10 Pa (Zehnder
and Stumm 1988), which renders the oxidation of short-chain Mutual relationships between different prokaryotes render them
fatty acids exergonic. Thus, Syntrophomonas wolfei in coculture capable of occupying niches that could not be occupied by either
can metabolize even-numbered fatty acids such as butyrate, organism alone. ‘‘Symbiosis’’ in its original sense is a close
caproate, and caprylate to H2 and acetate; odd-numbered fatty association between different species of organisms (de Bary
acids such as valerate and heptanoate are metabolized to propi- 1879). Currently, symbiosis is often defined as an obligatory
onate, acetate, and hydrogen (Dwyer et al. 1988; McInerney et al. and highly specific mutual relationship that benefits both part-
1979, 1981). Syntrophobacter wolinii was shown to degrade ners (Atlas and Bartha 1993; > Table 7.7). Most symbioses of
propionate to acetate, CO2, and H2 in coculture with an H2- prokaryotes known to date are with eukaryotes. Besides
consuming Desulfovibrio species (Boone and Bryant 1980). the endosymbiosis of chloroplasts (cyanobacteria) or
Although acetate can be cleaved to CH4 and CO2 by several a-mitochondria (Proteobacteria; Margulis 1981), the root
methanogenic species, a thermophilic acetate-oxidizing organ- nodule symbiosis of rhizobia with legumes is another well-
ism was shown to convert acetate to CO2 and H2 only in studied example. In the latter association, the bacteria provide
coculture with Methanobacterium thermautotrophicum combined nitrogen and the plant host dicarboxylic acids. In
(Zinder and Koch 1984). Benzoate and several other aromatic their symbiosis with anaerobic protozoa, prokaryotes appear
compounds were also shown to be degraded by cocultures of to act as hydrogen sinks in interspecies hydrogen transfer
a proton-reducing species and methanogens, sulfate-reducing (Hackstein et al. 1999). In ectosymbiotic association with
bacteria, or both (Mountfort and Bryant 1985; Dolfing and marine nematodes, especially adapted g-Proteobacteria oxidize
Tiedje 1986; Mountfort and Asher 1986; Szewzyk and Schink sulfide with concomitant CO2 fixation (Ott et al. 1991), as do the
1989). Although hydrogen represents the best-documented endosymbionts of the same group in gutless marine oligochaetes
agent for transfer of reducing equivalents, formate serves the (Dubilier et al. 1995). Cellulolytic nitrogen-fixing bacteria in
same purpose in species lacking hydrogenases (Thauer et al. shipworms (Bivalvia: Terebrenidae) provide exoenzymes neces-
1975; Thiele and Zeikus 1988; Thiele et al. 1988). Syntrophic sary for the digestion of cellulose or keratene (Waterbury et al.
associations can involve even three different physiological types 1983). Bacterial symbionts of insects synthesize nutrients such as
of prokaryotes, for example, two different sulfate-reducing bac- vitamins (Aksoy 1995) or recycle amino acids (Douglas 1998).
teria and one methanogenic archaeon (Thiele and Zeikus 1988). Intracellular bacterial symbionts of protozoa can provide toxins,
In other fermentations, proton reduction and subsequent which in turn are of selective value for the host in competition
hydrogen transfer is not a strict requirement since the reaction is with protozoal competitors (Görtz and Brigge 1998). In other
exergonic under standard conditions, but interspecies hydrogen cases, the cytotoxicity of the protozoal host is increased by
transfer shifts the fermentation pattern toward the production endosymbiotic bacteria (Fritsche et al. 1998). Most of the
of more oxidized products plus hydrogen in contrast to the prokaryotic symbionts of higher organisms could not be cul-
reduced fermentation products formed in pure culture. For tured to date and thus have only been identified by 16S
example, Ruminococcus albus ferments glucose to acetate, H2, rDNA analyses.
In contrast to the hundreds of cases of symbiotic relation- By far the most common method of enrichments is the batch
ships between prokaryotes and eukaryotes, only very few exam- culture technique, resulting in the selection of those prokaryotes
ples of symbioses between different prokaryotes have been that attain the maximum specific growth rate under the condi-
described to date (Overmann and Schubert 2002). However, tions chosen. Since at the same time, bacteria are growing for
evidence from very different environments, which range from a substantial period of time at high substrate concentrations,
the termite hindgut and dental plaque to the chemocline of batch culture enrichments frequently select for prokaryotes with
stratified lakes and deep-sea sediments, indicates that many high maximum growth rates and low substrate specificity (the so-
more symbiotic associations exist among prokaryotes called zymogenous organisms [Winogradsky 1949], also referred
(Overmann 2002a; Overmann and Schubert 2002). As indicated to as ‘‘R-strategists’’ [Andrews 1984; Andrews and Harris 1986]).
by microscopic studies, morphologically conspicuous associa- It has been argued for many years that prokaryotes obtained in
tions of prokaryotes exist in the form of ‘‘consortia,’’ in which this way are not representative of those which dominate micro-
two or more prokaryotes maintain a permanent cell-to-cell bial transformations in the natural environment (Jannasch
contact (Hirsch 1984; Schink 1991; Trüper and Pfennig 1971), 1967a, b; Veldkamp and Jannasch 1972; Veldkamp 1977; Kuenen
and most likely represent the extreme case of a mutual and Harder 1982; Gottschal 1986; Poindexter and Leadbetter
interdependence of different prokaryotes. To date, some 16 1986; Gottschal and Dijkhuizen 1988). Nevertheless, nonlimiting
different types of such consortia have been described (Huber growth conditions possibly also occur under natural conditions at
et al. 2002; Overmann 2002a; Overmann and Schubert 2002) least as transient phenomena, for instance after the burial of
and are comprised of very different phylogenetic groups, for organic matter (leaves and fecal pellets), during algal blooms,
example, filamentous cyanobacteria associated with heterotro- tidal flooding, or upon sudden changes in physicochemical con-
phic bacteria, associations between giant sulfur-oxidizing ditions. Similarly, in gastrointestinal tracts, feeding may provide
g-Proteobacteria (Thioploca) and sulfate-reducing Desulfonema, for conditions most suitable for a rapid response of opportunistic
or the aggregates of sulfate-reducing bacteria with methanogenic microbial species. The batch culture technique can be employed
archaea, which were detected in methane-hydrate-rich marine for the enrichment and isolation of K-strategists, if the latter
sediments. An additional novel type of association has been outnumber the accompanying R-strategists in the original sam-
described (Nanoarchaeum equitans; Huber et al. 2002). ple and are capable of (albeit slow) growth in the medium used.
Recently, advancements have been made toward a more detailed In these cases, serial dilution series are established, such that
understanding of the structure and function of phototrophic R-strategists are absent at higher dilutions and cannot overgrow
consortia consisting of green sulfur bacterial epibionts the target prokaryotes (Ferris et al. 1996; Sekiguchi et al. 2001).
surrounding a colorless motile central Betaproteobacterium The second, less frequently employed technique is the use
(Fröstl and Overmann 1998, 2000; Overmann et al. 1998; of a chemostat with one or more growth-limiting nutrients in
Tuschak et al. 1999; Overmann and Schubert 2002). the inflowing medium. Using this approach, organisms
specialized in growing at very low substrate concentrations and
usually exhibiting relatively low mmax values and fairly high
Enrichment Techniques substrate specificities (so-called autochthonous prokaryotes or
K-strategists) can be obtained and studied (Andrews 1984;
When the desired prokaryote is present in very small numbers, Jannasch 1967a; Jannasch 1967b; Schlegel and Jannasch 1967;
the enrichment culture technique is used to enable this partic- Harder et al. 1977; Kuenen and Harder 1982; Poindexter and
ular type of microorganism to grow faster than all others in the Leadbetter 1986; Gottschal and Dijkhuizen 1988). Another
sample. During selective enrichment, a natural sample harbor- advantage of continuous cultures systems over batch cultures
ing many different microorganisms is kept under conditions is that physical and chemical parameters can be maintained
which favor the growth of a particular physiological type of constant over prolonged periods. Alternatively, these conditions
prokaryote or a group of prokaryotes, thus allowing them to may be changed in a controlled fashion, such as during light/
grow faster than accompanying physiologically different types of dark cycles or oxic/anoxic transitions.
microorganisms (Schlegel and Jannasch 1967; Aaronson 1970;
Norris and Ribbons 1970; Veldkamp 1977; Gerhardt et al. 1981;
Poindexter and Leadbetter 1986; Austin 1988). Usually, no Isolation of Prokaryotes
inhibitory ingredient is added to the medium, but the enrich-
ment medium is designed to favor growth of the desired organ- The purpose of isolation is the generation of a single clone, that
ism. The technique was essentially developed by microbiologists is, a population consisting of bacteria all derived from a single
(e.g., Beijerinck and Winogradsky) early in the twentieth cen- cell. The strains obtained in this manner are an essential, but not
tury. The enrichment technique is applied if (1) a prokaryote sufficient (see section > ‘‘Interactions with Other Microorgan-
with particular metabolic properties is known to exist or (2) it is isms’’), prerequisite for thorough, in-depth, and unambiguous
to be determined whether prokaryotes exist which grow under studies of the ultrastructure, physiology, molecular biology,
specific conditions, for example, high/low temperature, high genetics, and autecology of prokaryotes. When free from other
osmolarity, extreme pH values, or with xenobiotic carbon (contaminating) microorganisms, a culture is called ‘‘axenic.’’ Not
substrates. all individuals in this population need to be genetically identical as
mutation and selection of mutants can take place, but this het- Anaerobic bacteria most frequently are isolated in deep-agar
erogeneity is accommodated in the pure culture concept. dilution series in prereduced media (see section > ‘‘Removal
Since pure cultures almost never occur in nature, their prep- and Exclusion of O2, Cultivation of Anaerobes’’ in this chapter).
aration involves the physical separation of single cells from The first tube is inoculated and mixed, and an aliquot (usually
others, inoculation into sterile medium, and incubation under a tenth of the volume) is transferred to the following tube. In this
conditions allowing axenic growth. In general, samples are manner, a serial dilution series is obtained, which is subse-
diluted by means of different methods and concomitantly placed quently solidified by cooling in a 20 C water bath, rapidly gassed
on a nonselective medium. Either direct samples or aliquots with N2 or a mixture of N2/CO2, and sealed with gas-tight butyl
from appropriate enrichments may be used for isolation of rubber stoppers.
strains. Separation of single prokaryotic cells can be accom- Liquid dilution series are preferentially employed for more
plished by streaking on solid media such as agar plates, separa- sensitive bacteria that do not readily grow on agar-solidified
tion in liquid media as occurs during preparation of pour plates, media or for inocula with cell titers too high for direct plating
in liquid dilution series, or with the aid of micromanipulators. on solid media. An additional advantage of this technique is that
Each colony on solid media or culture developing in the highest physical separation of bacterial cells in serial dilutions may
positive dilutions potentially has grown from a single cell. promote growth of those bacteria which are otherwise inhibited
For inoculation with diluted sample material, a small amount by antibacterial substances (e.g., bacteriolytic enzymes) pro-
of liquid (100 mL for agar plates) containing not more than duced by the accompanying bacteria (Talamoto et al. 1994).
about 200 cells is brought on the surface of an agar plate and Heterogeneous liquid samples are first homogenized by shaking
spread evenly by the use of a Drigalski spatula. Samples with or agitation in some sort of mechanical device, for example,
higher cell densities, for instance material from cell colonies, are a tube containing sterile glass beads, or in a mixer. A nontoxic
streaked on the surface with a sterile loop after dipping the latter in detergent and complexing agents, for example, Tween 80 (final
the sample. One of the most efficient ways of streaking is the concentration, 0.05% v/v) and sodium pyrophosphate (10 mM
‘‘13-streak method’’ (Cypionka 1999), which involves intermit- in 10 mM n-[2-hydroxyethyl] piperazine-N0 -[2-ethanesulfonic
tent sterilization of the loop and provides the highest chance to acid] [HEPES] buffer), respectively, may be added for detach-
obtain single, well-isolated colonies from samples of widely dif- ment of cells adhering to solids. Subsequently, serial dilutions
ferent cell densities. Physically separated cells grow out into dis- (mostly in 1:10 dilution steps) are prepared in sterile medium.
tinct populations visible on the solidified medium as small heaps Directly after inoculation, each tube has to be mixed carefully,
or spots, called ‘‘colonies,’’ varying in shape, color, and size. When and a fresh glass pipette or pipette tip must be employed for the
streaking an alleged pure culture, all colonies should be identical following dilution step. By inoculating tubes with cell suspen-
in appearance, color, and texture, but in crowded parts of the sions so dilute that some tubes receive just one single cell or no
plate, the size of the colonies will be smaller owing to nutrient cell at all, pure cultures may ultimately be obtained, and most
depletion. Consequently, size difference as a criterion can only be probable numbers (MPN) of culturable bacteria can be calcu-
applied to plates with well-separated (>5 mm) colonies. Motile lated. This method is based on the assumption that bacteria are
and swarming prokaryotes cannot easily be isolated on agar plates normally distributed in the liquid media. Numbers of viable cells
unless the plates are thoroughly dried and chemicals (mostly from natural samples obtained in this way frequently surpass
detergents like Pril® at a concentration of 0.2% w/v, as used in those obtained on agar media (Bartscht et al. 1999; Jaspers et al.
Pril-Mannitol-Agar [Merck]; Pietzsch 1967; Reusse and Meyer 2001). However, the opposite was observed for soil samples,
1972) added to disintegrate bacterial flagella and thus stop which yielded highest numbers of cultivated bacteria on
motility. Some bacteria cannot be cultivated or isolated on improved solid media (Janssen et al. 2002). More recently, the
solid agar media and require sloppy agar or a liquid medium. MPN technique has been adapted and increasingly used to
An alternate method is the inoculation of agar plates in the recover ultraoligotrophic and highly fastidious marine plank-
liquid stage, shortly before solidification (the so-called pour- tonic bacteria by dilution-to-extinction culturing. In these
plate technique). A known volume of a bacterial suspension is approaches, sterilized seawater was used as growth medium
mixed with molten agar (kept at 42 C to just prevent solidifica- and permitted the isolation of Sphingopyxis alaskensis from
tion) and then poured into a sterile Petri dish where the mixture cold, coastal oceanic samples (Button et al. 1993) and the
is allowed to solidify. With some experience, mixing can also be enrichment of ‘‘Candidatus Pelagibacter ubique,’’ the first cul-
performed in the Petri dish itself. Upon incubation, cells will tured representative of the most abundant heterotrophic
grow into colonies that mostly develop below the surface. The bacterioplankton group in the sea (Rappe et al. 2002).
latter colonies often show a lenticular shape. Gas-forming colo- One drawback of MPN dilution series when employed for
nies may produce cracks in the surrounding agar. The pour- the isolation of bacterial strains is that only few cultures can be
plate technique is suited for the isolation of organisms requiring obtained from the highest positive dilutions (i.e., based on the
oxygen at pressures lower than atmospheric (microaerobic bac- number of parallel dilution series, usually 3–8). However, pro-
teria). It is not easy, however, to isolate or count subsurface vided the average viability for cells in a given sample is known,
colonies or to assess colony morphology as a criterion for culture the number of pure cultures produced can be increased by orders
purity. Also, some prokaryotes may be killed by exposure to the of magnitude without increasing the total number of cultures
higher temperatures of the molten agar. inoculated. To this end, one appropriate size of inoculum is
chosen which statistically contains 0.5 viable cells for parallel ‘‘not-yet-cultured prokaryotes’’ can now be obtained even
inoculation of media. A very efficient way of automated inocu- down to the single-cell level (Ouverney and Fuhrman 2000;
lation is the MICRODROP® technique, in which microdroplets Wagner et al. 2006; Beier and Bertilsson 2011). The information
(volume of 100 pL) of a cell suspension are generated by gained by molecular approaches allows to develop even more
a piezoelectric actuator and positioned directly on agar plates targeted cultivation strategies. Chemotaxis assays have been
(Schober et al. 1993) or into each well of microtiter plates filled developed that permit in situ testing of the response of particular
with liquid media. If the procedure is automated, large numbers bacteria to selected substrates and at the same time enrich for the
of inoculation can be performed very rapidly (e.g., 100 inocula- desired bacterial type (Overmann 2005). It should be stressed,
tions in less than a minute). If each droplet statistically contains however, that recent progress in the successful cultivation (often
0.5 viable bacterial cells, the number of positive positions on the also isolation) of previously unculturable prokaryotes cannot be
agar plate or of wells in a microtiter plate follows a Poisson attributed to a single technological breakthrough but rather an
distribution and the most probable number of culturable bacte- educated combination of different, often innovative, methods.
ria can be calculated from fraction p of positive tubes (i.e., the The importance of applying a combination of different growth
number of positive tubes divided by the total number of tubes n) media rather than a single ‘‘universal’’ medium has already been
according to Button et al. (1993) demonstrated 15 years ago (Balestra and Misaghi 1997). A thor-
MNP ¼ lnð1 pÞ ð7:24aÞ ough analysis of a bacterial genome sequence may reveal media
supplements required for axenic growth as in the case of the
The standard deviation is human pathogenic actinobacterium Tropheryma whipplei, the
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi agent of Whipple’s disease that is unable to synthesize several
p
S:D: ¼ ð7:24bÞ amino acids de novo. Subsequently, a cell-free culture medium
½n ð1 pÞ
for this bacterium could be established (Renesto et al. 2003).
A much more tedious isolation procedure involves the use of However, it will remain a challenging task to identify more
micromanipulator devices (Skerman 1968), which nowadays complex requirements such as those based on oligotrophic
include computer-assisted electronic controls, and is used to properties or the dependence on biotic interactions with other
physically separate single cells or morphologically defined sym- microorganisms based on a genome sequence alone. Therefore,
biotic consortia under an inverted microscope (Fröhlich and it is to be expected that careful and high-throughput growth
König 1999; Fröstl and Overmann 2000). The isolated cells can experiments will remain a crucial element of cultivation trials
then be transferred into sterile media. This latter method is also in the foreseeable future.
especially valuable if morphologically conspicuous prokaryotes,
which are easily overgrown by other microorganisms, are to be
isolated. This procedure has been used to isolate cyanobacteria, Ionic Composition of the Growth Media
myxobacteria, budding bacteria, and manganese-oxidizing bac-
teria and phototrophic consortia (Fröhlich and König 1999; Fully synthetic media have been developed which correspond to
Fröstl and Overmann 2000; Sly and Arunpairojana 1987). the ionic strength of freshwater (Bartscht et al. 1999) or seawater
(Schut et al. 1993; Sass et al. 2001). For the cultivation of soil
prokaryotes, ionic concentrations have been adapted to those
Alternative and Novel Concepts for typically found in soil solution (Angle et al. 1991; Grosser et al.
Cultivation 2000; for Winogradsky’s salt solution, compare Aagot et al.
2001). Synthetic seawater medium has been used in several
The usually low cultivation success and the molecular detection investigations of the culturability of marine bacteria (Schut
of entire not-yet-cultured groups of prokaryotes (see section et al. 1993; Eguchi et al. 2001; Sass et al. 2001). The various
> ‘‘Introduction’’ in this chapter) strongly indicate that specially recipes differ slightly with respect to trace element content, salt
adapted or novel techniques have to be developed for the isola- composition, and the type of buffer used, and they represent an
tion of not-yet-cultured prokaryotes from natural samples. Most improvement over other types of media employed previously.
importantly, novel or alternative cultivation techniques should Reduction or elimination of phosphate and copper concentra-
account for those factors most likely responsible for the low tions, respectively, has proven very successful for the isolation of
culturability. ‘‘unculturable’’ cyanobacteria (Waterbury 1991). Recently, arti-
Meanwhile, state-of-the-art molecular techniques are fre- ficial freshwater media have been developed which facilitate the
quently employed to identify phylotypes associated with bio- cultivation of planktonic bacteria from limnic sources (Bartscht
geochemical transformations. Molecular approaches are not et al. 1999; Jaspers et al. 2001).
limited to monitor abundances of certain phylotypes but
also include metagenomic techniques that helped to identify
entirely unexpected physiological properties of selected Exploring Novel Types of Metabolism
phylotypes such as the widespread occurrence and utilization
of proteorhodopsin by marine Proteobacteria (Béjà et al. 2000). The potential of any reaction to drive prokaryotic growth can be
Some information on the physiological capabilities of predicted based on a calculation of the free energy change.
A negative value of the free energy change indicates that cells, subsequent immunocapture of BrdU-labeled DNA, and
a postulated reaction can actually promote microbial growth. sequencing of the corresponding 16S rDNA sequences (Urbach
This information can then be used to tailor enrichments specif- et al. 1999). An important limitation of the latter approach is
ically for microorganisms that perform the transformation of that not all prokaryotes assimilate BrdU, even when metaboli-
interest. As one prominent example, anaerobic ammonium oxi- cally active.
dation (the so-called anammox process) was predicted theoret- As a fourth approach to identify potential growth substrates
ically by Broda (1977) and later discovered to occur in certain and optimize growth conditions for yet-to-be-cultured prokary-
members of the phylum Planctomycetes (Jetten et al. 1998). otes, specific oligonucleotide probes can be developed which are
Similarly, the proper combination of untraditional electron directed against the internal transcribed spacer region of the
donors and acceptors based on thermodynamical calculations rRNA (rrn) operon and used for FISH to detect selectively,
has led to the successful enrichment and isolation of an anaer- physiologically active cells (Cangelosi and Brabant 1997; Licht
obic arsenite-oxidizing autotrophic gammaproteobacterium et al. 1999; Oerther et al. 2000; Schmid et al. 2001). Lately, the
(Oremland et al. 2002) and a phosphite-oxidizing autotrophic quantification of ITS transcripts via qPCR was also used to
deltaproteobacterium (Schink et al. 2002). Novel types of micro- monitor physiological activity in situ (Marschall et al. 2010).
bial reactions may also be postulated on the basis of measure- All of the above techniques in principle can be employed to
ments of concentration changes in potential microbial perform substrate tests for target prokaryotes under close to in
substrates and products in the environment (Zengler et al. situ conditions without prior cultivation of the microorganism
1999; Schink and Friedrich 2000). of interest.
Cultivation-Independent Physiological Testing Types and Concentrations of Growth Substrates

of Target Bacteria
Sometimes, specific substrates can selectively promote the
The enrichment and isolation of numerically abundant prokary- growth of numerically abundant bacteria (González et al.
otes or specific groups of interest can be followed by a phyloge- 1997). However, target bacteria are frequently slow-growing
netic screening of enrichments, for example, by fluorescent in K-strategists and hence easily overgrown by the more rare but
situ hybridization (FISH; Kane et al. 1993; Spring et al. 2000). rapidly dividing large-celled R-strategist bacteria during enrich-
FISH has been used in a combined approach to identify poten- ment in the presence of higher substrate concentrations. Conse-
tial growth substrates of target prokaryotes for subsequent quently, one novel approach for the isolation of bacteria from
enrichment and isolation experiments. In this approach, the natural samples is the utilization of low substrate concentrations
uptake of radiolabeled substrates by target bacteria can be of carbon substrates added to synthetic media of appropriate
followed if microautoradiography is combined with FISH ionic strength (see section > ‘‘Ionic Composition of Growth
using oligonucleotide probes designed for selected prokaryotes Media’’). In addition, reduction of the concentrations of N and
(Lee et al. 1999; Cottrell and Kirchman 2000; Gray et al. 2000; P compounds (which at certain times may represent the growth-
Nielsen et al. 2000; Ouverney and Fuhrman 2000). Pitfalls of this limiting natural substrates) has been found to significantly
technique are (1) that only one growth substrate can be tested at increase colony counts (Eilers et al. 2001). An alternative
a time and (2) that the added substrate may be rapidly degraded approach is to employ sterile filtered seawater directly (Button
by other (nontarget) microorganisms and that target prokary- et al. 1993; Connon and Giovannoni 2002; Rappe et al. 2002),
otes are actually the only organisms present capable of assimi- which already contains naturally occurring organic carbon sub-
lating the (unidentified) metabolites. strates as potential substrates for the growth of prokaryotes.
Stable isotope probing is a parallel approach based on label- Media for the cultivation of soil bacteria can be supplemented
ing with 13C-labeled substrates of genomic DNA of selected with cold soil extract (Olsen and Bakken 1987) as a carbon and
prokaryotes. Active prokaryotes can subsequently be identified energy source or as a source of supplies. In some instances,
by 16S rDNA sequencing of the heavy DNA (Radajewski et al. application of low-nutrient media has yielded comparably high
2000). A limitation of the latter technique is that (1) only those cultivation success (up to 60%; Button et al. 1993). Application
substrates can be tested which are utilized by very few types of of complex mixtures of organic carbon substrates from natural
prokaryotes (e.g., C1-compounds), (2) a large excess of labeled sources can also increase growth of selected groups of anaerobic
substrate and long incubation times are necessary to maximize bacteria. For example, anaerobically prepared sterilized sludge
13
C-uptake and obtain sufficiently heavier genomic DNA— or sediment slurries resulted in higher MPN of sulfate-reducing
possibly resulting in pronounced shifts in microbial community bacteria as compared to conventional media (Vester and
structure, and (3) unknown intermediates may be rapidly gen- Ingvorsen 1998).
erated by other bacteria, thus preventing a correct identification As an alternative, typical K-strategists can be
of potential growth substrates. preconcentrated and potential R-strategists selectively excluded
Thirdly, prokaryotes stimulated by defined substrates can be by a prefiltration step during which the inoculum is filtered
identified on the basis of the incorporation of bromodeox- through sterile 0.8-mm- or 0.2-mm-pore-size membrane filters.
yuridine (BrdU) into genomic DNA by metabolically active This treatment can significantly increase the enrichment
efficiency of certain target bacteria. As an example, members of organic carbon compounds (e.g., acetate) are assimilated
the Holophaga/Acidobacterium lineage or the freshwater heterotrophically (Lovley et al. 1999; Coates et al. 2002). This
Actinobacterium cluster could be successfully enriched from latter type of metabolism has been shown for S. alga,
freshwater lake sediments or lake water, respectively, using G. metallireducens, and G. fermentans but also for
prefiltration (Spring et al. 2000; Hahn et al. 2003). the denitrifying bacterium Paracoccus denitrificans,
Especially adapted cultivation techniques have been devel- the a-Proteobacterium Agrobacterium strain PB, the
oped to meet the exact physiological requirements of fastidious b-Proteobacteria Dechloromonas agitata, Dechloromonas
bacteria. Gradient plates or gradient tubes represent a simple strain JJ, Azoarcus strain HA, the g-Proteobacteria Pseudomo-
means to cultivate prokaryotes dependent on two substrates, nas strain BU, Pseudomonas strain NMX, and Marinobacter
each present at a defined and very narrow concentration range. strain SBS, and a g-Proteobacterium (strain KC; Lovley et al.
Examples include chemolithotrophic sulfur-oxidizing bacteria, 1999; Coates et al. 2002). These microorganisms are capable of
which are naturally occurring in highly stratified environments exploiting a part of the large pool of humic substances in soils
such as the chemocline of stratified lakes or sediments. In such and sediments and gain a competitive advantage over prokary-
cases, artificial agar-stabilized oxygen-sulfide-countergradients otes that depend on limited organic compounds (e.g., acetate) as
have been used very successfully to isolate for the first time energy source.
Beggiatoa alba or magnetotactic cocci (Nelson et al. 1986;
Meldrum et al. 1993). In a second type of system, a quartz
sand core is employed for gradient stabilization and is Cocultivation
sandwiched between a lower anoxic sulfide-containing compart-
ment and an upper oxic compartment (benthic gradient cham- Often, positive interactions between cells of one clone are
ber; Pringault et al. 1996). As a great advantage of such gradient required for growth of a single cell. This is exemplified by the
systems, the bacteria of interest (by means of their own physio- fact that in the case of fastidious prokaryotes, small inocula can
logical activity) precisely create the necessary environmental result in an extended lag period or a complete failure of growth.
conditions for growth (Jørgensen 1982). This effect has been attributed to the presence of low-molecular-
The testing of novel classes of substrates has led to the weight metabolic intermediates excreted by living cells whose
discovery of novel physiological capabilities of known bacterial introduction into fresh medium depends on the size of the
species or even to the enrichment and isolation of novel bacteria inoculum. This type of isolation difficulties can be remedied
of a broad phylogenetic diversity (Coates et al. 2002). Many by preparing a sterile filtrate of spent enrichment culture
investigations have focused on the role of humic substances on medium and including this filtrate as a major constituent of
bacterial growth. Humic substances consist of a skeleton of alkyl the new medium on which single cells of fastidious microorgan-
or aromatic units which are cross-linked mainly by oxygen and isms are to be isolated (Atlas and Bartha 1993).
nitrogen groups and which carry carboxylic acid, phenolic and As a third point of concern, evidence has accumulated that
alcoholic hydroxyls, ketone, and quinone groups (Schulten et al. different prokaryotes may actually depend on phylogenetically
1991). Contrary to the conservative view of humic substances as nonrelated partner bacteria for a successful enrichment (see
being refractory, they in fact can be utilized in different ways by section > ‘‘Interactions with Other Microorganisms’’). For
a large number of prokaryotes. Besides the utilization as growth example, bacteria of the Holophaga/Acidobacterium lineage in
substrates by Rhodococcus spp. (Goodfellow 1992a), and prob- enrichment cultures as well as natural microbial communities
ably also by Actinoplanes and related genera (Vobis 1992), are frequently accompanied by Alphaproteobacteria of the
Microbispora, Streptosporangium (Goodfellow 1992b), and Beijerinckia group (Felske et al. 1998; Nogales et al. 1999;
Pedomicrobium spp. (Poindexter 1992), humic substances rep- Radajewski et al. 2000; Spring et al. 2000). Another example is
resent redox-active compounds that act as electron carriers in the not-yet-cultured Betaproteobacterium associated with green
abiotic as well as biotic redox reactions (Coates et al. 2002). The sulfur bacteria in phototrophic consortia (Fröstl and Overmann
redox-active moieties in these reactions are the quinone groups 2000) and to date not enriched or grown in pure culture. Double
(Lovley et al. 1996a; Scott et al. 1998). Humic compounds can culture vessels for cocultivation of different bacteria have been
act as an electron acceptor for respiratory and fermentative described earlier (Jannasch and Mateles 1974; Wirsen and
bacteria belonging to diverse bacterial lineages, such as Jannasch 1978). Meanwhile, simple incubation devices have
the high G + C Gram-positive bacteria (Propionibacterium been developed which permit the growth of two different pro-
freudenreichii), g-Proteobacteria (Shewanella algae), karyotes in coculture but separate them by a membrane perme-
d-Proteobacteria (Geobacter metallireducens), or the able to metabolites. These devices range from simple O-rings
Holophaga/Acidbacterium division (e.g., Geothrix fermentans, covered with membrane filters (Kaeberlein et al. 2002) over
Benz et al. 1998; Coates et al. 1998; Lovley et al. 1999). Humic commercially available tissue culture inserts (Nichols et al.
compounds as water-soluble carriers can shuttle electrons 2008) to specific cultivation vessels that facilitate cocultivation
between microorganisms and insoluble terminal electron accep- for aerobic but also anaerobic bacteria (> Fig. 7.5).
tors (e.g., Fe(III) oxides; Lovley et al. 1996). In addition, humic The addition of cAMP, which represents one type of bacterial
substances have been shown to be utilized as electron-donating signal compound at least in some cases, resulted in a significant
substrates during respiration with nitrate or fumarate, while increase of culturability in natural samples (Bruns et al. 2002).
An isolated Roseovarius strain exhibited increased cell yield in isolation of aquatic Planctomycetes (Schlesner 1994). This
the presence of cAMP. Similarly, the addition of pyrophosphate method uses sterilized glass cover slips that are inserted vertically
to cultures of E. coli has been proposed to confer upon the cells into sterilized agar (often water or soft agar) at the bottom of
a better capacity to use carbon sources, induce biosynthetic a Petri dish. For the inoculation and growth of oligotrophic
processes, and enhance stationary-phase survival of E. coli cells aquatic prokaryotes, the agar is overlaid with 10 mL of sample
(Biville et al. 1996). water and incubated for several weeks. For monitoring of
growth, individual cover slips can be recovered and inspected
by phase contrast light microscopy. Also, the presence of specific
Chemical Protection bacterial target groups can be investigated by fluorescence in situ
hybridization (Dedysh 2011). Recently, polyurethane foam
Upon exposure to heat, ethanol, or osmotic stress, bacterial cells soaked with various agar media was employed during in situ
may stop growing while their metabolism continues. As a result of incubation experiments and shown to trap a large variety of
this uncoupling, a burst of free radical production occurs which different marine bacteria (Yasumoto-Hirose et al. 2006).
may be lethal to the cells. Exponentially growing bacterial cells are Gliding bacteria, like the green sulfur bacterium
significantly more sensitive to this so-called suicide response than Chloroherpeton thalassium, require a solid matrix for growth.
cells in the stationary phase (Aldsworth et al. 1999). Entry of For initial isolation, washed soft agar at a final concentration of
E. coli into stationary-phase growth leads to expression of 0.8% has been found to be the most suitable matrix. On this
genes for resistance mechanisms under the control of the alter- matrix, spreading of the filaments during growth leads to a fluffy
native sigma factor ss, and they include those for resistance appearance of the colonies. A gelling agent needs to be added
against oxidative stress such as katE (Loewen and Hengge- even to pure cultures for growth, however. Good results have
Aronis 1994). In addition, the presence of oxidizing compounds, been obtained by adding Gelrite (4% w/v stock solution) to
in particular molecular oxygen, itself can cause an inhibition of a final concentration of 0.025% which produces a visible
sensitive bacteria. Facultative anaerobes and microaerophiles can increase in viscosity of the liquid medium and significantly
be grown only under reduced oxygen tension. Under full oxygen stimulates cell division of the gliding bacteria. In a similar fash-
tension, these bacteria form cocultures with aerobes that lower ion, the filamentous gliding multicellular sulfate-reducing
concentrations in liquid enrichments or colonies on agar media. Desulfonema spp. can be maintained on insoluble aluminum
Isolation of target bacteria then requires reduced oxygen tension. phosphate precipitate which can readily be generated in the
The recovery after starvation stress can be significantly enhanced anoxic growth medium by addition of AlCl3 and Na2CO3
by inclusion of catalase in the agar medium even for bacteria like (Widdel et al. 1983).
Escherichia coli that should itself be capable of H2O2 detoxifica-
tion (Mizunoe et al. 1999). At least in some bacteria, cellular
damage by oxidative stress can be prevented by the addition of Conservation of Prokaryotic Cultures
H2O2-degrading compounds such as catalase, sodium pyruvate,
or a-oxoglutaric acid (Martin et al. 1976; Brewer et al. 1977; For future studies, as reference for standardized assays and tests,
Mossel et al. 1980; Mizunoe et al. 1999). The addition of ascorbic for taxonomic purposes, and as valuable stock for biotechno-
acid to the growth medium or lowering of the oxygen partial logical applications, strains of prokaryotes need to be maintained
pressure has similar effects (Jannasch and Mateles 1974). and preserved over extended periods of time. During repeated
The addition of chemical protectants or antioxidants may cultivation cycles, mutations can arise that change major proper-
thus improve the recovery of abundant bacteria from natural ties of strains. As a prominent example, Bdellovibrio strains that
samples. Accordingly, addition of activated charcoal as another are typically host dependent for growth can acquire the ability for
scavenger of toxic oxygen forms increased the number of colony- independent growth. For long-term maintenance, cultures are
forming units in samples from the deep terrestrial subsurface therefore routinely stored in a lyophilized or deep-frozen form
(Stevens 1995). to prolong their viability and to reduce changes due to the
occurrence of mutations. Genetic changes may even during con-
servation, however, since conservation also selects for adapted
Solid Surfaces variants among the same cell population and genetic drift can
lead to rapid genetic changes upon transfer of culture small
Solid surfaces may lead to stimulation of cell division and volumes after conservation.
growth of starved bacteria (Kjelleberg et al. 1982, see section
> ‘‘Effect of Inhomogeneities’’). So far, however, the capability
of attachment has rarely been exploited as a selective factor for Maintenance of Working Cultures
the enrichment and isolation of novel types of prokaryotes. The
cover slip cultivation technique was originally developed for the Usually, maintenance of working cultures requires the periodic
growth of streptomycetes for microscopic examination (Kawato transfer of strains to fresh minimal media. Aerobic strains are
and Shinobu 1959) and subsequently established for the streaked on cotton-plugged agar slants or agar plates, whereas
stab cultures in agar deeps are produced for facultative anaer- procedure. However, not all bacteria can be preserved by this
obes or microaerophiles and closed by screw caps. Obligate method (e.g., purple sulfur bacteria [Chromatiaceae]) (Pfennig
anaerobes are transferred to fresh liquid media. After cultiva- and Trüper 1989). For example, about 65% of the culture stocks
tion, cultures are stored in the refrigerator in the dark at 5–8 C held in the DSMZ culture collection are provided as lyophilized
to reduce metabolic activity while maintaining viability. The cultures.
frequency of transfer should be kept to a minimum, but it has Freshly grown cells are harvested by centrifugation and
to be determined separately for each organism: some require placed in an ampoule containing suspending medium and fro-
transfer after just a few days, while some sporeformers can zen at 60 C to 80 C. Rapid freezing is often accomplished in
survive in liquid cultures for years. During storage, agar slants a metal chamber with dry ice (solid CO2) and ethanol. The water
and stab cultures must be prevented from drying. To avoid is then removed from the frozen state directly by sublimation in
selection of mutants, it is advisable to subculture from the a vacuum (<1 Pa). This prevents formation of a liquid water
whole plate or slant and not just from one colony. phase. Freeze-dryers consist of a manifold connected to
For different physiological groups of bacteria, specific main- a vacuum pump. The ampoule is sealed and stored in a cool
tenance requirements exist. Working cultures of rumen bacteria place. The suspending medium is critical for the rate of survival
and other fermentative anaerobes can be maintained in slants or during the freeze-drying process and for the rate of death of
sloppy agar (0.5–0.7% [w/v]) prepared from nonselective media. dried bacteria during storage. Among the most commonly used
The cultures are incubated at 39 C until growth is just apparent suspending media are serum plus 30% glucose nutrient broths
and then stored at 4 C. Such cultures remain viable for a few or sterile skimmed milk fortified with 5% sucrose, sterilized in
weeks at least, although the viable count is decreased. Purple 3-mL amounts. Survival rates of Gram-positive and Gram-
sulfur bacteria of the family Chromatiaceae are precultured in negative bacteria and yeasts during the freeze-drying procedure
sulfide-containing media until the sulfide is depleted and intra- range between 8% and 85% and for subsequent long-term
cellular sulfur has been formed as the intermediate oxidation archiving range between 85% and 98% (Miyamoto-Shinohara
product in the cells; cultures in this phase of growth have et al. 2006). More robust bacterial strains could thus be con-
a chalky appearance macroscopically in reflected light. Subse- served over 50 years in the freeze-dried state (Gherna and Reddy
quently, cultures are transferred to the dark at 4 C and can be 2007).
stored for at least 3 months. Green sulfur bacteria (Chlorobiaceae) A percentage of survivors of freeze-thaw stress may lose the
are maintained best after the accumulated sulfur has also been ability to form colonies on a minimal salts-glucose medium. In
depleted. However, cultures of many microorganisms kept at 4 C E. coli, this effect was not due to a particular nutrient that was
still show slow growth and turnover of cellular components. absent but rather to a perturbation of the cell control system
Special preservation of cultures is required for extended storage since cyclic GMP or ppGpp singly or in combination partly
and to prevent genetic and metabolic alterations. restored the efficiency of plating on a minimal medium of
frozen-thawed cells. Mutagenic effects of freeze-thaw stress
have been shown to be related to single-stranded breaks in
Preservation of Stock Cultures DNA, an effect that might be similar to that of ionizing radia-
tion. The damage can be repaired in nutrient media. In E. coli,
Spore-Forming Bacteria both the uvrA-, polA-dependent (excision repair) and the error-
prone rec-, polA-dependent (recombinational repair) DNA
One method of long-term preservation for spore-forming bac- repair pathways are required for repair of freeze-thaw-induced
teria is based on the use of sterile soil or sand. A sample of soil or DNA damage.
sand is sterilized in a screw-capped bottle by autoclaving for
several hours on at least two successive days. Then, 1 mL of the
suspension of the organism is added, and the contents dried in Freezing and Ultrafreezing
a vacuum desiccator with the cap of the tube loosely closed.
When the contents are dry, the cap is closed tightly and the bottle Cultures of rumen bacteria can be kept viable for up to one year
stored in the refrigerator. Even easier is the use of sterile filter by adding 20% (v/v) glycerol as a cryoprotectant and storage
paper discs or strips soaked with the bacterial suspension. The at 20 C (Teather 1982). Freezing and archiving of bacterial
discs are stored as described for sand above. One may keep strains at 70 C has been successful for less fastidious bacteria
several discs in one screw-cap container and remove one disc and for time periods of 1–3 years (Gherna and Reddy 2007).
at a time, taking care not to contaminate the remaining discs. Novel approaches comprise loading of porous glass or ceramic
beads that are frozen directly without culture supernatant. Indi-
vidual beads can then be retrieved while effectively avoiding
Lyophilization thawing of the entire sample.
Extra- and intracellular ice crystal formation represent the
Lyophilization (Tindall 2007) is often preferred over major problems during freezing and ultrafreezing of living cells.
cryoconservation (see below) due to the simpler handling During the freezing process, ice crystal formation in the
extracellular matrix results in an osmotic gradient across the References

cytoplasmic membrane which during slow freezing rates causes
an osmotic loss of water and osmotic stress of the cells. Addition Aagot N, Nybroe O, Nielsen P, Johnsen K (2001) An altered Pseudomonas
of extracellular cryoprotectants interferes with ice crystal growth diversity is recovered from soil by using nutrient-poor Pseudomonas-
selective soil extract media. Appl Environ Microbiol 67:5233–5239
and decreases the temperature of ice crystal formation. Extra-
Aaronson S (1970) Experimental microbial ecology. Academic, New York
cellular cryoprotectants that have been tested are skimmed Abdul-Tehrani H, Hudson AJ, Chang YS, Timms AR, Hawkins C, Williams JM,
milk powder (20% w/v), sucrose (12% w/v) or other sugars Harrison PM, Guest JR, Andrews SC (1999) Ferritin mutants of Escherichia
(raffinose, lactose, trehalose), sugar alcohols (inositol, sorbitol; coli are iron deficient and growth impaired, and fur mutants are iron
10% w/v), glycine betaine, methanol, ethylen glycol, propandiole, deficient. J Bacteriol 181:1415–1428
Achberger AM, Brox TI, Skidmore ML, Christner BC (2011) Expression and partial
2-methyl-2,4-pentandiole, and polymers such as dextran,
characterization of an ice-binding protein from a bacterium isolated at a depth
polyglycol, hydroxyethyl starch, methyl cellulose, polyvinylpyr- of 3,519 m in the Vostok ice core, Antarctica. Front Microbiol 2:255
rolidone, Ficoll, hyaluronan, bovine serum albumin, or activated Ahn WS, Park SJ, Lee SY (2001) Production of poly (3-hydroxybutyrate) from
charcoal (Hubalek 2003; Gherna and Reddy 2007). whey by cell recycle fed-batch culture of recombinant Escherichia coli.
Rapid freezing prevents water to leave the bacterial cell and Biotechnol Lett 23:235–240
Aksoy S (1995) Wigglesworthia gen. nov. and Wigglesworthia glossinidia sp. nov.,
but results in the formation of intracellular ice crystals that also
taxa consisting of the mycetocyte-associated, primary endosymbionts of
may cause damage and permeabilization of the cytoplasmic Tsetse flies. Int J Syst Bacteriol 45:848–851
membrane similar to extracellular crystals. Membrane- Alden L, Demoling F, Baath E (2001) Rapid method of determining factors
permeable cryoprotectants such as glycerol or dimethyl sulfox- limiting bacterial growth in soil. Appl Environ Microbiol 67:1830–1838
ide bind intracellular water, raise the osmotic pressure and hence Aldsworth TG, Sharman RL, Dodd CER (1999) Bacterial suicide through stress.
Cell Mol Life Sci 56:378–383
interfere with intracellular ice crystal formation. Typical con-
Alldredge AL, Youngbluth MJ (1985) The significance of macroscopic aggregates
centrations applied for glycerol are 10–25% (v/v) and for (marine snow) as sites for heterotrophic bacterial production in the meso-
dimethyl sulfoxide are 5–10% (v/v). Due to the slow diffusion pelagic zone of the subtropical Atlantic. Deep-Sea Res 32:1445–1456
of glycerol through the cytoplasmic membrane, the cryoprotec- Amy PS, Morita RY (1983) Starvation-survival patterns of sixteen freshly isolated
tive effect of glycerol is only weak. DMSO is highly effective in open-ocean bacteria. Appl Environ Microbiol 45:1109–1115
Andrews JH (1984) Relevance of r-and K-theory to the ecology of plant patho-
promoting vitrification (i.e., formation of a viscous glassy aggre-
gens. In: Klug MJ, Reddy CA (eds) Current perspectives in microbial ecology.
gate state of water without ice crystals) of the cells but at the ASM Press, Washington, DC, pp 1–7
same time is cytotoxic. Therefore, DMSO must be removed Andrews JH, Harris RF (1986) r-and K-selection and microbial ecology. In:
rapidly upon thawing of the cells and before cultivation. Marshall KC (ed) Advanced microbiology and ecology, vol 9. Plenum,
Another complication of cryoconservation is oxidative New York, pp 1–7
Andrews SC, Robinson AK, Rodriguez-Quinones F (2003) Bacterial iron homeo-
stress. Antioxidative protecting enzymes are damaged by freez-
stasis. FEMS Microbiol Rev 27:215–237
ing which may result in peroxidation of membrane phospho- Angle JS, McGrath SP, Chaney RL (1991) New culture medium containing ionic
lipids. In eukaryotic cells, formation of toxic oxygen species has concentrations of nutrients similar to concentrations found in the soil
been prevented by addition of reducing agents such as ascorbate solution. Appl Environ Microbiol 57:3674–3676
(Tarin and Trounson 1993) or detoxifying enzymes such as Arrhenius S (1889) Über die Reaktionsgeschwindigkeit bei der Inversion von
Rohrzucker durch Säuren. Z Phys Chem 4:226–248
superoxide dismutase (Dinara et al. 2001).
Atlas RM, Bartha R (1993) Microbial ecology, 3rd edn. Benjamin/Cummings,
In order to promote vitrification of bacterial cells, rapid Redwood City
cooling is typically performed in the presence of cryoprotec- Austin B (1988) Methods in aquatic bacteriology. Wiley, Chichester
tants and at the temperature of liquid nitrogen (196 C) or in Azam F (1998) Microbial control of oceanic carbon flux: The plot thickens.
the gaseous phase of liquid nitrogen (150 C) which simulta- Science 280:694–696
Bak F, Pfennig N (1991) Microbial sulfate reduction in littoral sediment of Lake
neously excludes molecular oxygen (Hespell and Bryant 1981;
Constance. FEMS Microbiol Ecol 85:31–42
Pfennig and Trüper 1989). Successful cryoconservation Balch WE, Wolfe RS (1976) New approach to the cultivation of methanogenic
protocols comprise the use of narrow glass capillaries that are bacteria: 2-mercaptoethane sulfonic acid (HS-CoM)-dependent growth of
sealed directly after filling with culture suspension and subse- Methanobacterium ruminantium in a pressurized atmosphere. Appl Environ
quently dropped into liquid nitrogen (Hippe 1991; Tindall Microbiol 32:781–791
Bale SJ, Goodman K, Rochelle PA, Marchesi JR, Fry JC, Weightman AJ, Parkes RJ
2007). Yet, different strains of the same bacterial species can
(1997) Desulfovibrio profundus sp. nov., a novel barophilic sulfate-reducing
perform very differently during conservation, and an increasing bacterium from deep sediment layers in the Japan Sea. Int J Syst Bacteriol
number of novel isolates cannot be conserved using established 47:515–521
protocols. Balestra GM, Misaghi IL (1997) Increasing the efficiency of the plate counting
method for estimating bacterial diversity. J Microbiol Methods 30:111–117
Barer MR, Harwood CR (1999) Bacterial viability and culturability. Adv Microb
Culture Collections Physiol 41:93–137
Barns SM, Fundyga RE, Jeffries MW, Pace NR (1994) Remarkable archaeal
diversity detected in a Yellowstone National Park hot spring environment.
A complete list of culture collections worldwide may be obtained
from the World Data Center for Microorganisms maintained by Barns SM, Takala SL, Kuske CR (1999) Wide distribution and diversity of
the World Federation for Culture Collections (WFCC) at members of the bacterial kingdom Acidobacterium in the environment.
(http://www.wfcc.info/ccinfo/collection/). Appl Environ Microbiol 65:1731–1737
Baross JA, Morita RY (1978) Microbial life at low temperatures: ecological Boone DR, Bryant MP (1980) Propionate degrading bacterium Syntrophobacter
aspects. In: Kushner DJ (ed) Microbial life in extreme environments. Aca- wolinii sp. nov., gen. nov. from methanogenic ecosystems. Appl Environ
demic, London, pp 9–17 Microbiol 33:1162–1169
Bartlett DH, Welch TJ (1995) ompH gene expression is regulated by multiple Botsford JL, Harman JG (1992) Cyclic AMP in prokaryotes. Microbiol Rev
environmental cues in addition to high pressure in the deep-sea bacterium 56:100–122
Photobacterium species strain SS9. J Bacteriol 177:1008–1016 Bovill RA, Mackey BM (1997) Resuscitation of ‘‘non-culturable’’ cells from aged
Bartlett D, Wright M, Yayanos AA, Silverman M (1989) Isolation of a gene regulated cultures of Campylobacter jejuni. Microbiology 143:1575–1581
by hydrostatic pressure in a deep-sea bacterium. Nature 342:572–574 Bowman JP, McCammon SA, Brown JL, Nichols PD, McMeekin TA (1997a)
Bartlett DH, Chi E, Wright ME (1993) Sequence of the ompH gene from the deep- Psychroserpens burtonensis gen. nov., sp. nov., and Gelidibacter algens, gen.
sea bacterium Photobacterium SS9. Gene 131:125–128 nov., sp. nov., psychrophilic bacteria isolated from Antarctic lacustrine and
Bartscht K, Cypionka H, Overmann J (1999) Evaluation of cell activity and of sea ice habitats. Int J Syst Evol Microbiol 47:670–677
methods for the cultivation of bacteria from a natural lake community. Bowman JP, McCammon SA, Nichols DS, Skerratt JH, Rea SM, Nichols PE,
FEMS Microbiol Ecol 28:249–259 McMeekin TA (1997b) Shewanella gelidimarina sp. nov. and Shewanella
Basler BL, Losick R (2006) Bacterially speaking. Cell 125:237–246 frigidimarina sp. nov., novel Antarctic species with the ability to produce
Bast E (2001) Mikrobiologische methoden, 2nd edn. Spektrum Akad, Berlin eicosapentaenoic acid (20:5w3) and grow anaerobically by dissimilatory Fe
Bateson MM, Ward DM (1988) Photoexcretion and fate of glycolate in a hot (III) reduction. Int J Syst Bacteriol 47:1040–1047
spring cyanobacterial mat. Appl Environ Microbiol 54:1738–1743 Bowman JP, McCammon SA, Skerratt JH (1997c) Methylosphaera hansonii gen.
Battley EH (1995) An apparent anomaly in the calculation of ash-free dry weights nov., sp. nov., a psychrophilic, group I methanotroph from Antarctic
for the determination of cellular yields. Appl Environ Microbiol 61: marine-salinity, meromictic lakes. Microbiology 143:1451–1459
1655–1657 Bowman JP, McCammon SA, Brown JL, McMeekin TA (1998a) Glaciecola
Baxter RM, Gibbons NE (1962) Observations on the physiology of punicea gen. nov., sp. nov., and Glaciecola pallidula gen. nov., sp. nov.:
psychrophilism in a yeast. Can J Microbiol 8:511–517 psychrophilic bacteria from Antarctic sea-ice habitats. Int J Syst
Beier S, Bertilsson S (2011) Uncoupling of chitinase activity and uptake of Evol Microbiol 48:1213–1222
hydrolysis products in freshwater bacterioplankton. Limnol Oceanogr Bowman JP, McCammon SA, Lewis T, Skerratt JH, Brown JL, Nichols DS,
56:1179–1188 McMeekin TA (1998b) Psychroflexus torquis gen. nov., sp. nov.,
Béjà O, Aravind L, Koonin EV, Suzuki MT, Hadd A, Nguyen LP, Jovanovich SB, a psychrophilic species from Antarctic sea ice, and reclassification of
Gates CM, Feldman RA, Spudich JL, Spudich EN, DeLong EF (2000) Bacte- Flavobacterium gondwanense (Dobson et al. 1993) as Psychroflexus
rial rhodopsin: evidence for a new type of phototrophy in the sea. Science gondwandense gen. nov., comb. nov. Microbiology 144:1601–1609
289:1902–1906 Boyaval P (1989) Lactic acid bacteria and metal ions. Lait 69:87–113
Béjà O, Suzuki MT, Heidelberg JF, Nelson WC, Preston CM, Hamadas T, Eisen JA, Bozal N, Montes MJ, Tudela E, Jimenez F, Guinea J (2002) Shewanella
Fraser CM, DeLong EF (2002) Unsuspected diversity among marine aerobic frigidimarina and Shewanella livingstonesis sp. nov. isolated from Antarctic
anoxygenic phototrophs. Nature 415:630–633 coastal areas. Int J Syst Evol Microbiol 52:195–205
Benz M, Schink B, Brune A (1998) Humic acid reduction by Propionibacterium Braun V (1997) Avoidance of iron toxicity through regulation of bacterial iron
freudenreichii and other fermenting bacteria. Appl Environ Microbiol transport. Biol Chem 378:779–786
64:4507–4512 Brefeld O (1881) Botanische Untersuchungen Über Schimmelpilze:
Bernard L, Schäfer H, Joux F, Courties C, Muyzer G, Lebaron P (2000) Genetic Culturmethoden. Felix, Leipzig
diversity of total, active and culturable marine bacteria in coastal seawater. Brewer DG, Martin SE, Ordal ZJ (1977) Beneficial effects of catalase or pyruvate
Aquat Microb Ecol 23:1–11 in a most-probable-number technique for the detection of Staphylococcus
Bernhardt G, Lüdemann H-D, Jaenicke R, König H, Stetter KO (1984) Biomole- aureus. Appl Environ Microbiol 34:797–800
cules are unstable under ‘‘black smoker’’ conditions. Naturwissenschaften Brock TD (1978) Thermophilic microorganisms and life at high temperatures.
71:583–586 Springer, New York
Beudeker RF, Gottschal JC, Kuenen JG (1982) Reactivity versus flexibility in Brock TD (1987) Introduction: an overview of the thermophiles. In: Brock TD
thiobacilli. Antonie Van Leeuwenhoek 48:39–51 (ed) Thermophiles: general molecular and applied microbiology. Wiley, New
Beunink J, Rehm HJ (1988) Synchronous anaerobic and aerobic degradation of York, pp 1–16
DDT by an immobilized mixed culture system. Appl Microbiol Biotechnol Brock TD, O’Dea K (1977) Amorphous ferrous sulfide as a reducing agent for
29:72–80 culture of anaerobes. Appl Environ Microbiol 33:254–256
Bhakoo M, Herbert RA (1979) The effects of temperature on the fatty acid and Broda E (1977) Two kinds of lithotrophs missing in nature. Z Allg Mikrobiol
phospholipid composition of four obligately psychrophilic Vibrio spp. Arch 17:491–493
Microbiol 121:121–127 Broda DM, Saul DJ, Lawson PA, Bell RG, Musgrave DR (2000) Clostridium
Bi E, Lutkenhaus J (1993) Cell division inhibitors SulA and MinCD prevent gasigenes sp. nov., a psychrophile causing spoilage of vacuum-packed meat.
formation of the FtsZ ring. J Bacteriol 175:1118–1125 Int J Syst Evol Microbiol 50:107–118
Bidle KD, Lee SH, Marchant DR, Falskowski PG (2007) Fossil genes and microbes Bromke B, Hammel JM (1987) Gelatin as a complete endogenous source of
in the oldest ice on Earth. Proc Natl Acad Sci USA 104:13455–13460 calcium for Serratia marcescens protease activity. J Microbiol Methods
Biebl H, Pfennig N (1978) Growth yields of green sulfur bacteria in mixed 6:253–256
cultures with sulfur and sulfate reducing bacteria. Arch Microbiol 117:9–16 Brooke AG, Watling EM, Attwood MM, Tempest DW (1989) Environmental
Binnerup SJ, Jensen DF, Thordal-Christensen H, Sorgensen J (1993) Detection of control of metabolic fluxes in thermotolerant methylotrophic Bacillus
viable, but non-culturable Pseudomonas fluorescens DF57 in soil using strains. Arch Microbiol 151:268–273
a microcolony epifluorescence technique. FEMS Microbiol Ecol 12:97–195 Brown AD (1976) Microbial water stress. Bacteriol Rev 40:803–846
Biville F, Laurent-Winter C, Danchin A (1996) In vivo positive effects of exoge- Brown AD, Simpson JR (1972) Water relations of sugar-tolerant yeasts: the role of
nous pyrophosphate on Escherichia coli cell growth and stationary phase intracellular polyols. J Gen Microbiol 72:589–591
survival. Res Microbiol 147:597–608 Brown SA, Whiteley M (2007) A novel exclusion mechanism for carbon resource
Blochl E, Rachel R, Burggraf S, Hafenbradl D, Jannasch HW, Stetter KO partitioning in Aggregatibacter actinomycetemcomitans. J Bacteriol 189:
(1997) Pyrolobus fumarii, gen. nov. and sp. nov., represents a novel group 6407–6414
of archaea, extending the upper temperature limit for life to 113 C. Bruhn JB, Nielsen KF, Hjelm M, Hansen M, Brescani J, Schulz S, Gram L (2005)
Extremophiles 1:14–21 Ecology, inhibitory activity, and morphogenesis of a marine antagonistic
Bolen DW (2001) Protein stabilization by naturally occurring osmolytes. bacterium belonging to the Roseobacter clade. Appl Environ Microbiol
Methods Mol Biol 168:17–36 71:7263–7270
Bruns A, Cypionka H, Overmann J (2002) Cyclic AMP and acyl homoserine Cayley S, Lewis BA, Record MT (1992) Origins of osmoprotective properties of
lactones increase the cultivation efficiency of heterotrophic bacteria from the betaine and proline in Escherichia coli K-12. J Bacteriol 174:1586–1595
central Baltic Sea. Appl Environ Microbiol 68:3978–3987 Chan M, Himes RH, Akagi JM (1971) Fatty acid composition of thermophilic,
Bruns A, Hoffelner H, Overmann J (2003a) A novel approach for high through- mesophilic, and psychrophilic clostridia. J Bacteriol 106:876–881
put assays and the isolation of planktonic bacteria. FEMS Microbiol Ecol Chao H, Davies PL, Carpenter JF (1996) Effects of antifreeze proteins in red blood
45:161–171 cell survival during cryopreservation. J Exp Biol 199:2071–2076
Bruns A, Nübel U, Cypionka H, Overmann J (2003b) Effect of signal compounds Chen X, Schauder S, Potier N, Van Dorsselaer A, Pelczer I, Bassler BL,
and incubation conditions on the culturability of freshwater bacterio- Hughson FM (2002) Structural identification of a bacterial quorum-sensing
plankton. Appl Environ Microbiol 69:1980–1989 signal containing boron. Nature 415:545–549
Bryant MP (1976) The microbiology of anaerobic degradation and Chen H, Jogler M, Rohde M, Klenk H-P, Busse H-J, Tindall B, Spröer C,
methanogenesis with special reference to sewage. In: Schlegel HW, Barnea Overmann J (2012) Reclassification and amended description of Caulobacter
J (eds) Microbial energy conversion. Erick Goltze KG, Göttingen, leidyi as Sphingomonas leidyi comb. nov., and emendation of the genus
pp 107–1167 Sphingomonas. Int J Syst Evol Microbiol (in press)
Bryson V, Szybalski W (1952) Microbial selection. Science 116:45–51 Chesbro W (1988) The domains of slow bacterial growth. Can J Microbiol
Bull AT, Slater JH (1982) Microbial interactions and community structure. In: 34:427–435
Bull AT, Slater JH (eds) Microbial interactions and communities, vol 1. Chesbro WR, Evans T, Eifert R (1979) Very slow growth of Escherichia coli.
Academic, London, pp 13–44 J Bacteriol 139:625–638
Bull AT, Goodfellow M, Slater JH (1992) Biodiversity as a source of innovation in Chi E, Bartlett DH (1995) An rpoE-like locus controls outer membrane protein
biotechnology. Annu Rev Microbiol 46:219–252 synthesis and growth at cold temperatures and high pressures in the deep-sea
Bulthuis BA, Koningstein GM, Stouthamer AH, van Verseveld HW bacterium Photobacterium sp. strain SS9. Mol Microbiol 17:713–726
(1989) A comparison between aerobic growth of Bacillus licheniformis in Chisholm SW, Frankel SL, Goericke R, Olson RJ, Palenik B, Waterbury JB, West-
continuous culture and partial-recycling fermenter, with contributions to Johnsrud L, Zettler ER (1992) Prochiorococcus marinus nov. gen. nov. sp.: an
the discussion on maintenance energy demand. Arch Microbiol 152:499–507 oxyphototrophic marine prokaryote containing divinyl chlorophyll a and b.
Bungay HR, Bungay ML (1968) Microbial interactions in continuous culture. Arch Microbiol 157:297–300
Adv Appl Microbiol 10:269–290 Cho JC, Giovannoni SJ (2004) Cultivation and growth characteristics of a diverse
Bunt JG (1961) Nitrogen-fixing blue-green algae in Australian rice soils. Nature group of oligotrophic marine Gammaproteobacteria. Appl Environ
192:479–480 Microbiol 70:432–440
Burchard RP (1980) Gliding motility of bacteria. Bioscience 30:157–162 Christensen BB, Haagensen JAJ, Heydorn A, Molin S (2002) Metabolic commen-
Burkhardt F (1992) Mikrobiologische diagnostik georg. Thieme, New York salism and competition in a two-species microbial consortium. Appl Envi-
Burnham JC, Conti SF (1984) Genus Bdellovibrio. In: Krieg NR (ed) Bergey’s ron Microbiol 68:2495–2502
manual of systematic bacteriology, vol 1. Williams and Wilkins, Baltimore, Christner BC (2010) Bioprospecting for microbial products that affect ice crystal
pp 118–124 formation and growth. Appl Microbiol Biotechnol 85:481–489
Bussmann I, Philipp B, Schink B (2001) Factors influencing the cultivability of Clark C, Schmidt EL (1966) Effect of mixed culture on Nitrosomonas europaea
lake water bacteria. J Microbiol Methods 47:41–50 simulated by uptake and utilization of pyruvate. J Bacteriol 91:367–373
Button DK, Schut F, Quang P, Martin R, Robertson BR (1993) Viability and Cleland N, Enfors SO (1983) Control of glucose fed batch cultivations of E. coli by
isolation of marine bacteria by dilution culture: theory, procedures, and means of an oxygen stabilized enzyme electrode. Eur J Appl Microbiol
initial results. Appl Environ Microbiol 59:881–891 Biotechnol 18:141–147
Button DK, Robertson BR, Lepp PW, Schmidt TM (1998) A small, dilute- Coates JD, Ellis DJ, Blunt-Harris EL, Gaw CV, Roden E, Lovley DR (1998) Recov-
cytoplasm, high-affinity, novel bacterium isolated by extinction culture ery of humic-reducing bacteria from a diversity of environments. Appl
and having kinetic constants compatible with growth at ambient concentra- Environ Microbiol 64:1504–1509
tions of dissolved nutrients in seawater. Appl Environ Microbiol 64: Coates JD, Cole KA, Chakraborty R, O’Connor SM, Achenbach LA (2002) Diver-
4467–4476 sity and ubiquity of bacteria capable of utilizing humic substances as electron
Byrer DE, Rainey FA, Wiegel J (2000) Novel strains of Moorella thermoacetica donors for anaerobic respiration. Appl Environ Microbiol 68:2445–2453
form unusually heat-resistant spores. Arch Microbiol 174:334–339 Cohen Y, de Jonge I, Kuenen JG (1979) Excretion of glycolate by Thiobacillus
Calcott PH (1981) The construction and operation of continuous cultures. In: neapolitanus grown in continuous culture. Arch Microbiol 122:189–194
Calcott PH (ed) Continuous cultures of cells, vol 1. CRC Press, Boca Raton, Cohn F (1872) Über bacterien, die kleinsten lebenden. Wesen Carl Habel, Berlin
pp 13–26 Connon SA, Giovannoni SJ (2002) High-throughput methods for culturing
Calcott PH, Calvert TJ (1981) Characterization of 30 :50 cyclic AMP phosphodies- microorganisms in very-low-nutrient media yield diverse new marine iso-
terase in Klebsiella aerogenes and its role in substrate accelerated death. J Gen lates. Appl Environ Microbiol 68:3878–3885
Microbiol 122:313–321 Coolen MJL, Cypionka H, Smock A, Sass H, Overmann J (2002) Ongoing
Calcott PH, Postgate JR (1972) On substrate-accelerated death in Klebsiella modification of Mediterranean Pleistocene sapropels mediated by prokary-
aerogenes. J Gen Microbiol 70:115–122 otes. Science 296:2407–2410
Calcott PH, Montague W, Postgate JR (1972) The levels of cyclic AMP during Cottrell MT, Kirchman DL (2000) Natural assemblages of marine proteobacteria
substrate accelerated death. J Gen Microbiol 73:197–200 and members of the Cytophaga-Flavobacteria cluster consuming low-and
Cangelosi GA, Brabant WH (1997) Depletion of pre-16S rRNA in starved high-molecular-weight dissolved organic matter. Appl Environ Microbiol
Escherichia coli cells. J Bacteriol 179:4457–4463 66:1692–1697
Carlsson J, Granberg GPD, Nyberg GK, Edlund M-BK (1979) Bactericidal effect Coveney MF (1982) Bacterial uptake of photosynthetic carbon from freshwater
of cysteine exposed to atmospheric oxygen. Appl Environ Microbiol 37: phytoplankton. Oikos 38:8–20
383–390 Crevel RWR, Fedyk JK, Spurgeon MJ (2002) Antifreeze proteins: characteristics,
Carpenter EJ, Lin S, Capone DG (2000) Bacterial activity in South Pole snow. occurrence and human exposure. Food Chem Toxicol 40:899–903
Appl Environ Microbiol 66:4514–4517 Crocker FH, Guerin WF, Boyd SA (1995) Bioavailability of naphthalene sorbed to
Castenholz RW (1973) Movements. In: Carr NG, Whitton BA (eds) The biology cationic surfactant-modified smectite clay. Environ Sci Technol 29:
of blue-green algae. Blackwell, London, pp 320–339 2953–2958
Cavicchioli R (2006) Cold-adapted archaea. Nat Rev Microbiol 4:331–343 Csonka LN (1989) Physiological and genetic responses of bacteria to osmotic
Cayley S, Record MT, Lewis BA (1989) Accumulation of 3-[N-morpholino] stress. Microbiol Rev 53:121–147
propanesulfonate by osmotically stressed Escherichia coli K-12. J Bacteriol Csonka LN, Epstein W (1996) Osmoregulation. In: Neidhardt FC (ed) Escherichia
171:3597–3602 coli and Salmonella, 2nd edn. ASM Press, Washington, DC, pp 1210–1223
Csonka LN, Hanson A (1991) Prokaryotic osmoregulation: genetics and physi- Douglas AE (1998) Nutritional interactions in insect-microbial symbioses: aphids
ology. Annu Rev Microbiol 45:569–606 and their symbiotic bacteria Buchnera. Annu Rev Entomol 43:17–37
Currie DJ, Kalff J (1984) Can bacteria outcompete phytoplankton for phospho- Driessen FM (1981) Protocooperation of yoghurt bacteria in continuous cultures.
rus? A chemostat test. Microb Ecol 10:205–216 In: Buschell ME, Slater JH (eds) Mixed culture fermentations. Academic,
Curtis TP, Sloan WT, Scannell JW (2002) Estimating prokaryotic diversity and its London, pp 99–120
limits. Proc Natl Acad Sci USA 99:10494–10499 Dubilier N, Giere O, Distel DL, Cavanaugh CM (1995) Characterization of
Cypionka H (1986) Sulfide-controlled continuous culture of sulfate-reducing chemoautotrophic bacterial symbionts in a gutless marine worm
bacteria. J Microbiol Methods 5:1–9 (Oligochaeta, Annelida) by phylogenetic 16S rRNA sequence analysis and
Cypionka H (1999) Grundlagen der mikrobiologie. Springer, New York in situ hybridization. Appl Environ Microbiol 61:2346–2350
Cypionka H, Widdel F, Pfennig N (1985) Survival of sulfate-reducing bacteria Dubinina GA, Leshcheva NV, Grabovich MY (1993) The colorless sulfur bacte-
after oxygen stress, and growth in sulfate-free oxygen-sulfide gradients. rium Thiodendron is actually a symbiotic association of spirochetes and
FEMS Microbiol Ecol 31:39–45 sulfidogens. Microbiology 62:432–444
Czeczuga B (1968) An attempt to determine the primary production of the green Dubinina GA, Grabovich MY, Leshcheva NV, Rainey FA, Gavrish E (2011)
sulphur bacteria, Chlorobium limicola Nads, (Chlorobacteriaceae). Spirochaeta perfilievii sp. nov., an oxygen-tolerant, sulfide-oxidizing, sulfur-
Hydrobiologia 31:317–333 and thiosulfate-reducing spirochaete isolated from a saline spring. Int J Syst
Dalluge JJ, Hamamoto T, Horikoshi K, Morita RY, Stetter KO, McCloskey JA Evol Microbiol 61:110–117
(1997) Posttranscriptional modification of tRNA in psychrophilic bacteria. Dukan S, Nyström T (1998) Bacterial senescence: stasis results in increased and
J Bacteriol 179:1918–1923 differential oxidation of cytoplasmic proteins leading to developmental
Damoglou AP, Dawes EA (1968) Studies on the lipid content and phosphate induction of the heat shock regulon. Genes Dev 12:3431–3441
requirement for glucose-and acetate-grown Escherichia coli. Biochem Dunfield PF, Liesack W, Henckel T, Knowles R, Conrad R (1999) High-affinity
J 110:775–781 methane oxidation by a soil enrichment culture containing a type II
Dang HY, Lovell CR (2000) Bacterial primary colonization and early succession methanotroph. Appl Environ Microbiol 65:1009–1014
on surfaces in marine waters as determined by amplified rRNA gene restric- Dwyer DF, Weeg-Aerssens E, Shelton DR, Tiedje JM (1988) Bioenergetic condi-
tion analysis and sequence analysis of 16S rRNA genes. Appl Environ tions of butyrate metabolism by a syntrophic, anaerobic bacterium in
Microbiol 66:467–475 coculture with hydrogen oxidizing methanogenic and sulfidogenic bacteria.
Dawson MW, Maddox IS, Boag IF, Brooks JD (1988) Application of fed-batch Appl Environ Microbiol 54:1354–1359
culture to citric acid production by Aspergillus niger: the effects of dilution Dykhuizen DE (1998) Santa Rosalia revisited: why are there so many species of
rate and dissolved oxygen tension. Biotechnol Bioeng 32:220–226 bacteria? Antonie Van Leeuwenhoek 73:25–33
De Bary A (1879) Die Erscheinung der Symbiose Naturforschung Versammlung Dykhuizen D, Davies M (1980) An experimental model: bacterial specialists and
Cassel generalists competing in chemostats. Ecology 61:1213–1227
de Freitas MJ, Fredrickson AG (1978) Inhibition as a factor in the maintenance of Egland PG, Palmer RJ, Kolenbrander PE (2004) Interspecies communication
the diversity of microbial ecosystems. J Gen Microbiol 106:307–320 in Streptococcus gordonii-Veillonella atypica biofilms: signaling in flow
de la Broise D, Durand A (1989) Osmotic, biomass, and oxygen effects on the conditions requires juxtaposition. Proc Natl Acad Sci USA 101:16917–16922
growth rate of Fusarium oxysporum using a dissolved oxygen controlled Egli T, Lindley ND, Quayle JR (1983) Regulation of enzyme synthesis and
turbidstat. Biotechnol Bioeng 33:699–705 variation of residual methanol concentration during carbon limited growth
De Wit R, van Gemerden H (1988) Interactions between phototrophic bacteria in of Kloeckera ap. 2201. on mixtures of methanol and glucose. J Gen Microbiol
sediment ecosystems. Hydrobiol Bull 22:135–145 129:1269–1281
Dedysh SN (2011) Cultivating uncultured bacteria from northern wetlands: Eguchi M, Ostrowski M, Fegatella F, Bowman J, Nichols D, Nishino T, Cavicchioli
knowledge gained and remaining gaps. Front Microbiol 2:184 R (2001) Sphingomonas alaskensis AFO1, an abundant oligotrophic
DeLong EF, Yayamos AA (1986) Biochemical function and ecological significance ultramicrobacterium from the North Pacific. Appl Environ Microbiol
of novel bacterial lipids in deep-sea prokaryotes. Appl Environ Microbiol 67:4945–4954
51:730–737 Eichorst SA, Breznak JA, Schmidt TM (2007) Isolation and characterization of
DeLong EF, Ying Wu K, Prézelin BB, Jovine RVM (1994) High abundance of soil bacteria that define Terriglobus gen. nov., in the phylum Acidobacteria.
Archaea in Antarctic marine picoplankton. Nature 371:695–697 Appl Environ Microbiol 73:2708–2717
DeLong EF, Franks DG, Yayanos AA (1997) Evolutionary relationships of culti- Eilers H, Pernthaler J, Peplies J, Glöckner FO, Gerdts G, Amann R (2001)
vated psychrophilic and barophilic deep-sea bacteria. Appl Environ Isolation of novel pelagic bacteria from the German Bight and their seasonal
Microbiol 63:2105–2108 contributions to surface picoplankton. Appl Environ Microbiol 67:
Deming JW (1986) Ecological strategies of barophilic bacteria in the deep ocean. 5134–5142
Microbiol Sci 3:205–207 Emde R, Schink B (1990) Oxidation of glycerol, lactate, and propionate by
Deming JW, Colwell RR (1985) Observations of barophilic microbial activity in Propionibacterium freudenreichii in a poised-potential amperometric culture
samples of sediments and intercepted particulates from the Demerara abyssal system. Arch Microbiol 153:506–512
plain. Appl Environ Microbiol 50:1002–1006 Emde R, Swain A, Schink B (1989) Anaerobic oxidation of glycerol by Escherichia
Deming JW, Somers LW, Straube WL, Swartz DG, MacDonell MT (1988) Isola- coli in an amperometric poised-potential culture system. Appl Microbiol
tion of an obligately barophilic bacterium and description of a new genus, Biotechnol 32:170–175
Colwellia gen. nov. Syst Appl Microbiol 10:152–160 Epstein W (1986) Osmoregulation by potassium transport in Escherichia coli.
Dinara S, Sengoku K, Tamate K et al (2001) Effects of supplementation with free FEMS Microbiol Rev 39:73–78
radical scavengers on the survival and fertilization rates of mouse Epstein W, Rothman-Denes LB, Hesse J (1975) Adenosine 30 :50 -cyclic
cryopreserved oocytes. Human Reprod 16:1976–1981 monophosphate as mediator of catabolite repression in Escherichia coli.
Dobell C (1932) Anthony van Leeuwenhoek and his ‘‘little animals’’. Harcourt Proc Natl Acad Sci USA 72:2300–2304
Brace, New York Esener AA, Roels JA, Kossen NW (1981) Fed batch culture: modelling and
Dolfing J, Tiedje JM (1986) Hydrogen cycling in a three-tiered food web growing application in the study of microbial energetics. Biotechnol Bioeng
on the methanogenic conversion of 3-chlorobenzoate. FEMS Microbiol Ecol 22:1851–1871
38:293–298 Ettwig KF, Butler MK, LePaslier D, Pelletier E, Mangenot S, Kuypers MMM et al
Dong K, Liu H, Zhang J, Zhou Y, Xin Y (2011) Flavobacterium xueshanense (2010) Nitrite-driven anaerobic methane oxidation by oxygenic bacteria.
sp. nov. and Flavobacterium urumqiense sp. nov., two psychrophilic Nature 464:543–548
bacteria isolated from the China No. 1 glacier. Int J Syst Evol Microbiol Euzéby JP, Tindall BJ (2001) Nomenclatural type of orders: corrections necessary
62:1151–1157 according to Rules 15 and 21a of the Bacteriological Code (1990 Revision),
and designation of appropriate nomenclatural types of classes and sub- Fröstl JM, Overmann J (2000) Phylogenetic affiliation of the bacteria that con-
classes. Request for an opinion. Int J Syst Evol Microbiol 51(Pt 2):725–727 stitute phototrophic consortia. Arch Microbiol 174:50–58
Euzéby JP (2012) List of bacterial names with standing in nomenclature. http:// Fry JC (1990) Direct methods and biomass estimation. In: Grigorova R, Norris JR
www.bacterio.cict.fr (eds) Methods in microbiology, vol 22. Academic, London, pp 41–85
Evans CG, Herbert D, Tempest DW (1970) The continuous cultivation of Fuhrman JA, McCallum K, Davis AA (1992) Novel major archaebacterial group
microorganisms. 2. Construction of a chemostat. In: Norris JR, from marine plankton. Nature 356:148–149
Ribbons DW (eds) Methods in microbiology, vol 2. Academic, London, Fuhrman JA, McCallum K, Davis AA (1993) Phylogenetic diversity of subsurface
pp 277–327 marine microbial communities from the Atlantic and Pacific Oceans. Appl
Feller G (2007) Life at low temperatures: is disorder the driving force? Environ Microbiol 59:1294–1302
Extremophiles 11:211–216 Funk HB, Krulwich TA (1964) Preparation of clear silica gels that can be streaked.
Feller G, Narinx E, Arpigny JL, Zekhnini Z, Swings J, Gerday C (1994a) Temper- J Bacteriol 88:1200–1201
ature dependence of growth, enzyme secretion and activity of psychrophilic Fuqua C, Greenberg EP (1998) Self perception in bacteria: quorum sensing with
Antarctic bacteria. Appl Microbiol Biotechnol 41:477–479 acylated homoserine lactones. Curr Opin Microbiol 1:50–58
Feller G, Payan F, Theys F, Quian M, Haser R, Gerday C (1994b) Stability and Galinski EA, Trüper HG (1994) Microbial behaviour in salt-stressed ecosystems.
structural analysis of a-amylase from the Antarctic psychrophile FEMS Microbiol Rev 15:95–108
Alteromonas haloplanctis A23. Eur J Biochem 222:441–447 Gangola P, Rosen BP (1987) Maintenance of intracellular calcium in Escherichia
Felske A, Wolterink A, van Lis R, Akkermans ADL (1998) Phylogeny of the main coli. J Biol Chem 262:12570–12574
bacterial 16S rRNA sequences in Drentse A grassland soils (The Nether- Ganzert L, Bajerski F, Mangelsdorf K, Lipski A, Wagner D (2011a) Arthrobacter
lands). Appl Environ Microbiol 64:871–879 livingstonensis sp. nov. and Arthrobacter cryotolerans sp. nov., salt-tolerant
Felske A, Wolterink A, van Lis R, de Vos WM, Akkermans ADL (1999) Searching and psychrotolerant species from Antarctic soil. Int J Syst Evol Microbiol
for predominant soil bacteria: 16S rDNA cloning versus strain cultivation. 61:979–984
FEMS Microbiol Ecol 30:137–145 Ganzert L, Bajerski F, Mangelsdorf K, Lipski A, Wagner D (2011b) Leifsonia
Ferchichi M, Hemme D, Bouillaune C (1986) Influence of oxygen and pH on psychrotolerans sp. nov., a psychrotolerant species of the family Microbac-
methanethiol production from L-methionine by Brevibacterium linens teriaceae from Livingston Island, Antarctica. Int J Syst Evol Microbiol
CNRZ 918. Appl Environ Microbiol 51:725–729 61:1938–1943
Ferris MJ, Ruff-Roberts AL, Kopczynski ED, Bateson MM, Ward DM Garcia-Lara J, Shang LH, Rothfield LI (1996) An extracellular factor regulates
(1996) Enrichment culture and microscopy conceal diverse thermophilic expression of sdiA, a transcriptional activator of cell division genes in
Synechococcus populations in a single hot spring microbial mat habitat. Escherichia coli. J Bacteriol 178:2742–2748
Appl Environ Microbiol 62:1045–1050 Garnham CP, Gilbert JA, Hartman CP, Campbell RL, Laybourn-Parry J, Davies PL
Filippini M, Kaech A, Ziegler U, Bagheri HC (2011) Fibrisoma limi gen. nov., sp. (2008) A Ca2+-dependent bacterial antifreeze protein domain has a novel
nov., a filamentous bacterium isolated from tidal flats. Int J Syst b-helical ice-binding fold. Biochem J 411:171–180
Evol Microbiol 61:1418–1424 Garnham CP, Campbell RL, Dabies PL (2011) Anchored clathrate waters bind
Fleming A (1929) On the antibacterial action of cultures of a Penicillium, with antifreeze proteins to ice. Proc Natl Acad Sci USA 108:7363–7367
special reference to their use in the isolation of B. influenzae. Br J Exp Pathol Gause GF (1934) The struggle for existence. Williams and Wilkins, Baltimore
10:226–236 Geissinger O, Herlemann DPR, Mörschel E, Maier UG, Brune A (2009) The
Flint KP (1985) A note on a selective agar medium for the enumeration of ultramicrobacterium ‘‘Elusimicrobium minutum’’ gen. nov., sp. nov., the
Flavobacterium species in water. J Appl Bacteriol 59:561–566 first cultivated representative of the termite group 1 phylum. Appl Environ
Fogg GE (1971) Extracellular products of algae in freshwater. Arch Hydrobiol Microbiol 75:2831–2840
Beih Ergebn Limnol 5:1–25 Genomes Online Database (GOLD) (2012) http://www.genomesonline.org/cgi-
Forsberg CW (1987) Production of 1,3-propanediol from glycerol by Clostridium bin/GOLD/index.cgi?page_requested=Statistics
acetobutylicum and other Clostridium species. Appl Environ Microbiol Gerhardt P, Murray RGE, Costilow RN, Nester EW, Wood WA, Krieg NR,
53:639–643 Phillips GB (1981) Manual of methods for general bacteriology. American
Francis CA, Tebo BM (2002) Enzymatic manganese(II) oxidation by metaboli- Society for Microbiology, Washington, DC
cally dormant spores of diverse Bacillus species. Appl Environ Microbiol Gherna RL, Reddy CA (2007) Culture preservation. In: Reddy CA (ed) Methods
68:874–880 for general and molecular microbiology. ASM Press, Washington, DC,
Franzmann PD, Höpfl P, Weiss N, Tindall BJ (1991) Psychrotrophic, lactic acid- pp 1019–1033
producing bacteria from anoxic waters in Ace Lake, Antarctica: Gich F, Garcia-Gil J, Overmann J (2001) Previously unknown and phylogeneti-
Carnobacterium funditum sp. nov. and Carnobacterium alterfunditum sp. cally diverse members of the green nonsulfur bacteria are indigenous to
nov. Arch Microbiol 156:255–262 freshwater lakes. Arch Microbiol 177:1–10
Franzmann PD, Springer N, Ludwig W, Conway de Macario E, Rohde M (1992) Gich F, Schubert K, Bruns A, Hoffelner H, Overmann J (2005) Specific detection,
A methanogenic archaeon from Ace Lake, Antarctica: Methanococcoides isolation and characterization of selected, previously uncultured members of
burtonii sp. nov. Syst Appl Microbiol 15:573–581 freshwater bacterioplankton. Appl Environ Microbiol 71:5908–5919
Franzmann PD, Liu Y, Balkwill DL, Aldrich HC, Conway de Macario E, Boone DR Giovannoni S, Stingl U (2007) The importance of culturing bacterioplankton in
(1997) Methanogenium frigidum sp. nov., a psychrophilic H2-using the ‘omics’ age. Nature 5:820–826
methanogen from Ace Lake Antarctica. Int J Syst Bacteriol 47:1068–1072 Glöckner FO, Zaichikov E, Belkova N, Denissova L, Pernthaler J, Perthaler A,
Fredrickson AG (1977) Behaviour of mixed cultures of microorganisms. Annu Amann R (2000) Comparative 16S rRNA analysis of lake bacterioplankton
Rev Microbiol 31:63–87 reveals globally distributed phylogenetic clusters including an abundant
Fredrickson AG, Stephanopoulos G (1981) Microbial competition. Science group of actinobacteria. Appl Environ Microbiol 66:5053–5065
213:972–979 Gond O, Engasser JM, Matta-El-Amouri C, Petitdemange H (1986) The acetone
Fritsche TR, Sobek D, Gautom RK (1998) Enhancement of in vitro cytopathoge- butanol fermentation on glucose and xylose. II: Regulation and kinetics in
nicity by Acanthamoeba spp. following acquisition of bacterial endosymbi- fed batch cultures. Biotechnol Bioeng 28:167–175
onts. FEMS Microbiol Lett 166:231–236 González JM, Mayer F, Moran MA, Hodson RE, Withman WB (1997) Sagittula
Fröhlich J, König H (1999) Rapid isolation of single microbial cells from mixed stellata gen. nov., sp. nov., a lignin-transforming bacterium from a coastal
natural and laboratory populations with the aid of a micromanipulator. Syst environment. Int J Syst Bacteriol 47:773–780
Appl Microbiol 22:249–257 Goodfellow M (1992a) The family Nocardiaceae. In: Balows A, Trüper HG,
Fröstl JM, Overmann J (1998) Physiology and tactic response of the phototrophic Dworkin M, Schleifer K-H (eds) The prokaryotes, 2nd edn. Springer, New
consortium ‘‘Chlorochromatium aggregatum’’. Arch Microbiol 169:129–135 York, pp 1188–1213
Goodfellow M (1992b) The family Streptosporangiaceae. In: Balows A, Trüper HG, Guerin WF, Boyd SA (1997) Bioavailability of naphthalene associated with
Dworkin M, Schleifer K-H (eds) The prokaryotes, 2nd edn. Springer, New natural and synthetic sorbents. Water Res 51:1504–1512
York, pp 1115–1138 Guerrero R, Pedros-Alió C, Esteve I, Mas J, Chase D, Margulis L (1986) Predatory
Görtz H-D, Brigge T (1998) Intracellular bacteria in protozoa. Naturwis- prokaryotes: predation and primary consumption evolved in bacteria. Proc
senschaften 85:359–368 Natl Acad Sci USA 83:2138–2142
Gosink JJ, Staley JT (1995) Biodiversity of gas vacuolate bacteria from Antarctic Hackstein JH, Akhmanova A, Boxma B, Harhangi HR, Voncken FG
sea ice and water. Appl Environ Microbiol 61:3486–3489 (1999) Hydrogenosomes: eukaryotic adaptations to anaerobic environ-
Gosink JJ, Woese CR, Staley JT (1998) Polaribacter gen. nov., with three new ments. Trends Microbiol 7:441–447
species, P. irgensii sp. nov., P. franzmannii sp. nov. and P. filamentus so. nov., Hahn MW (2003) Isolation of strains belonging to the cosmopolitan
gas vacuolate polar marine bacteria of the Cytophaga-Flavobacterium- Polynucleobacter necessarius cluster from freshwater habitats located in
Bacteroides group and reclassification of ‘‘Flectobacillus glomeratus’’ as three climatic zones. Appl Environ Microbiol 69:5248–5254
Polaribacter glomeratus comb. nov. Int J Syst Evol Microbiol 48:223–235 Hahn MW (2009) Description of seven candidate species affiliated with the
Gottschal JC (1986) Mixed substrate utilization by mixed cultures. In: Leadbetter phylum Actinobacteria, representing planktonic freshwater bacteria. Int
ER, Poindexter JS (eds) Bacteria in nature, vol 2. Plenum, New York, pp 261– J Syst Evol Microbiol 59:112–117
292 Hahn MW, Lünsdorf H, Wu Q, Schauer M, Höfle MG, Boenigk J, Stadler P (2003)
Gottschal JC (1990) Different types of continuous culture in ecological studies. Isolation of novel ultramicrobacteria classified as Actinobacteria from five
In: Norris JR, Grigorova R (eds) Methods in microbiology, vol 22. Academic, freshwater habitats in Europe and Asia. Appl Environ Microbiol 69:1442–
London, pp 87–124 1451
Gottschal JC, Dijkhuizen L (1988) The place of the continuous culture in eco- Hallam SJ, Mincer TJ, Schleper C, Preston CM, Roberts K, Richardson PM,
logical research. In: Wimpenny JWT (ed) Handbook of laboratory model DeLong EF (2006) Pathways of carbon assimilation and ammonia oxidation
systems for microbial ecosystems, vol 1. CRC Press, Boca Raton, pp 19–79 suggested by environmental genomic analyses of marine Crenarchaeota.
Gottschal JC, Kuenen JG (1980) Selective enrichment of facultatively PLoS Biol 4:e95
chemolithotrophic thiobacilli and related organisms in continuous culture. Handelsmann J (2004) Metagenomics: application of genomics to uncultured
FEMS Microbiol Lett 7:241–247 microorganisms. Microbiol Mol Biol Rev 68:669–685
Gottschal JC, Morris JG (1981) The induction of acetone and butanol production Harder W, Dijkhuizen L (1982) Strategies of mixed substrate utilization in
in cultures of Clostridium acetobutylicum by elevated concentrations of microorganisms. Philos Trans R Soc Lond B Biol Sci 297:459–480
acetate and butyrate. FEMS Microbiol Lett 12:385–389 Harder W, Veldkamp H (1971) Competition of marine psychrophilic bacteria at
Gottschal JC, de Vries S, Kuenen JG (1979) Competition between the facultatively low temperatures. Antonie Van Leeuwenhoek 37:51–63
chemolithotrophic Thiobacillus A2, an obligately chemolithotrophic Harder W, Kuenen JG, Matin A (1977) Microbial selection in continuous culture.
Thiobacillus and a heterotrophic Spirillum for inorganic and organic sub- J Appl Bacteriol 43:1–24
strates. Arch Microbiol 121:241–249 Hardin G (1960) The competitive exclusion principle. Science 131:1292–1297
Gottschal JC, Harder W, Prins RA (1991) Principles of enrichment, isolation, Harms H, Zehnder AJB (1995) Bioavailability of sorbed 3-chlorodibenzofuran.
cultivation, and preservation. In: Balows A, Trüper HG, Dworkin M, Appl Environ Microbiol 61:27–33
Schleifer K-H (eds) The prokaryotes, 2nd edn. Springer, New York, Harris JE (1985) Gelrite as an agar substitute for the cultivation of mesophilic
pp 149–196 Methanobacterium and Methanobrevibacter species. Appl Environ Microbiol
Gottschalk G (1985) Bacterial metabolism. Springer, New York 50:1107–1109
Gottwald M, Gottschalk G (1985) The internal pH of Clostridium acetobutylicum Harwood C, Buckley M (2008) The uncharted microbial world: microbes and
and its effect on the shift from acid to solvent formation. Arch Microbiol their activities in the environment. A report from the American Academy of
143:42–46 Microbiology. American Academy of Microbiology, Washington, DC, 37 p
Graber JR, Breznak JA (2005) Folate cross-feeding supports symbiotic Hastings RC, Saunders JR, Hall GH, Pickup RW, McCarthy AJ (1998) Application
homoacetogenic spirochetes. Appl Environ Microbiol 71:1883–1889 of molecular biological techniques to a seasonal study of ammonia oxidation
Graham AF, Lund BM (1983) The effect of alkaline pH on growth and metabolic in a eutrophic freshwater lake. Appl Environ Microbiol 64:3674–3682
products of a motile, yellow-pigmented Streptococcus sp. J Gen Microbiol Herbert RA (1986) The ecology and physiology of psychrophilic microorganisms.
129:2429–2435 In: Herbert RA, Codd GA (eds) Microbes in extreme environments. Aca-
Grant WD, Tindall BJ (1986) The alkaline saline environment. In: Herbert RA, demic, London, pp 1–23
Codd GA (eds) Microbes in extreme environments. Academic, London, Herbert D, Elsworth R, Telling RC (1956) The continuous culture of bacteria:
pp 25–54 a theoretical and experimental study. J Gen Microbiol 14:601–622
Graumann P, Marahiel MA (1996) Some like it cold: response of microorganisms Hespell RB, Bryant MP (1981) The genera Butyrivibrio, Succinivibrio,
to cold shock. Arch Microbiol 166:293–300 Lachnospira and Selenomonas. In: Starr MP, Stolp H, Trüper HG, Balows
Gray ND, Howarth R, Pickup RW, Jones JG, Head IM (2000) Use of combined A, Schlegel HG (eds) The prokaryotes. Springer, New York, pp 1479–1494
microautoradiography and fluorescence in situ hybridization to determine Hinrichs K-U, Hayes JM, Sylva SP, Brewer PG, DeLong EF (1999) Methane-
carbon metabolism in mixed natural communities of uncultured bacteria consuming archaebacteria in marine sediments. Nature 398:802–805
from the genus Achromatium. Appl Environ Microbiol 66:4518–4522 Hiorns WD, Methé BA, Nierzwicki-Bauer SA, Zehr JP (1997) Bacterial diversity
Groeschel DHM (1982) The etiology of tuberculosis: a tribute to Robert Koch on in Adirondack Mountain lakes as revealed by 16S rRNA gene sequence
the occasion of the centenary of his discovery of the tubercle bacillus. ASM analysis. Appl Environ Microbiol 63:2957–2960
News 48:248–250 Hippe H (1991) Maintenance of methanogenic bacteria. In: Kirsop BE, Doyle
Gross CA (1996) Function and regulation of the heat shock proteins. In: A (eds) Maintenance of microorganisms and cultured cells, 2nd edn. Aca-
Neidhardt FC (ed) Escherichia coli and Salmonella, 2nd edn. ASM Press, demic, London, pp 101–114
Washington, DC, pp 1382–1399 Hirsch P (1984) Microcolony formation and consortia. In: Marshall KC (ed)
Grosser RJ, Friedrich M, Ward DM, Inskeep WP (2000) Effect of model sorptive Microbial adhesion and aggregation: Dahlem Konferenzen. Springer, New
phases on phenanthrene biodegradation: different enrichment conditions York, pp 373–393
influence bioavailability and selection of phenanthrene-degrading isolates. Hochachka PW, Moon TW, Mustafa T (1972) The adaptation of enzymes
Appl Environ Microbiol 66:2695–2702 to pressure in abyssal and midwater fishes. In: Sleigh MA, Macdonald AG
Guan LL, Onuki H, Kamino K (2000) Bacteria growth stimulation with exoge- (eds) The effects of pressure on living organisms. Academic, London,
nous siderophore and synthetic N-acyl homoserine lactone autoinducers pp 175–195
under iron-limited and low-nutrient conditions. Appl Environ Microbiol Holmes AJ, Roslev P, McDonald IR, Iversen N, Henriksen K, Murrell JC
66:2797–2803 (1999) Characterization of methanotrophic bacterial populations in soil
showing atmospheric methane uptake. Appl Environ Microbiol 65: comitans’’ and the myxobacterium Chondromyces crocatus. Appl Environ
3312–3318 Microbiol 63:719–723
Hommes RWJ, Postma PW, Tempest DW, Neijssel OM (1989) The influence of Jaenicke R (1988) Molecular mechanisms of adaptation of bacteria to extreme
the culture pH value on the direct glucose oxidation pathway in Klebsiella environments. Forum Microbiol 11:435–440
pneumoniae NCTC 418. Arch Microbiol 151:261–267 Jannasch HW (1967a) Enrichment of aquatic bacteria in continuous culture.
Hooke R (1665) Micrographia: or some physiological descriptions of minute Arch Mikrobiol 59:165–173
bodies made by magnifying glasses with observations and inquiries there- Jannasch HW (1967b) Growth of marine bacteria at limiting concentrations of
upon. John Martyn and James Allestry, London organic carbon in seawater. Limnol Oceanogr 12:264–271
Horikoshi K, Akiba T (1982) Alkalophilic microorganisms. Springer, Jannasch HW, Mateles RI (1974) Experimental bacterial ecology studied in
New York continuous culture. Adv Microb Physiol 11:165–212
Huang L, Forsberg CW, Gibbins LN (1986) Influence of external pH and fermen- Jannasch HW, Taylor CD (1984) Deep-sea microbiology. Annu Rev Microbiol
tation products on Clostridium acetobutylicum intracellular pH and cellular 38:487–514
distribution of fermentation products. Appl Environ Microbiol 51:1230– Jannasch HW, Wirsen CO, Taylor CD (1976) Undecompressed microbial
1234 populations from the deep sea. Appl Environ Microbiol 32:360–367
Hubalek Z (2003) Protectants used in the cryopreservation of microorganisms. Jannasch HW, Wirsen CO, Doherty KW (1996) A pressurized chemostat for the
Cryobiology 46:205–229 study of marine barophilic and oligotrophic bacteria. Appl Environ
Huber H, Hohn MJ, Rachel R, Fuchs T, Wimmer VC, Stetter KO (2002) A new Microbiol 62:1593–1596
phylum of Archaea represented by a nanosized hyperthermophilic symbiont. Janssen PH, Schuhmann A, Mörschel E, Rainey FA (1997) Novel anaerobic
Nature 417:63–67 ultramicrobacteria belonging to the Verrucomicrobiales lineage of bacterial
Hugenholtz P (2002) Exploring prokaryotic diversity in the genomic era. Genome decent isolated by dilution culture from anoxic rice paddy soil. Appl Environ
Biol Rev 3:0003.1–0003.8 Microbiol 63:1382–1388
Hugenholtz P, Pitulle C, Hershberger KL, Pace NR (1998) Novel division level Janssen PH, Yates PS, Grinton BE, Taylor PM, Sait M (2002) Improved
bacterial diversity in a yellowstone hot spring. J Bacteriol 180:366–376 culturability of soil bacteria and isolation in pure culture of novel members
Humphry DR, George A, Black GW, Cummings SP (2001) Flavobacterium of the divisions Acidobacteria, Actinobacteria, Proteobacteria, and
frigidarium sp. nov., an aerobic psychrophilic, xylanolytic and laminariolytic Verrucomicrobia. Appl Environ Microbiol 68:2391–2396
bacterium from Antarctica. Int J Syst Evol Microbiol 51:1235–1243 Jaspers E, Nauhaus K, Cypionka H, Overmann J (2001) Multitude and temporal
Hungate RE (1950) The anaerobic mesophilic cellulolytic bacteria. Bacteriol Rev variability of ecological niches as indicated by the diversity of cultivated
14:1–49 bacterioplankton. FEMS Microbiol Ecol 36:153–164
Hungate RE (1960) Microbial ecology of the rumen. Symposium: selected topics Jensen PR, Williams PG, Oh DC, Zeigler L, Fennical W (2007) Species-specific
in microbial ecology. Bacteriol Rev 24:353–364 secondary metabolite production in marine actinomycetes of the genus
Hungate RE (1966) The rumen and its microbes. Academic, New York Salinispora. Appl Environ Microbiol 73:1146–1152
Hungate RE (1969) A roll tube method for cultivation of strict anaerobes. In: Jetten MSM, Strous M, van de Pas-Schoonen KT, Schalk J, van Dongen UGJM,
Norris JR, Ribbons WD (eds) Methods in microbiology, vol 3B. Academic, van de Graaf AA, Logemann S, Muyzer G, van Loosdrecht MCM, Kuenen JG
London, pp 117–132 (1998) The anaerobic oxidation of ammonium. FEMS Microbiol Rev
Hungate RE (1985) Anaerobic biotransformations of organic matter. In: 22:421–437
Leadbetter ER, Poindexter JS (eds) Bacteria in nature, vol 1. Plenum, New Jones AK (1982) The interaction of algae and bacteria. In: Bull AT, Slater JH (eds)
York, pp 39–96 Microbial interactions and communities. Academic, London, pp 189–247
Huston AL, Krieger-Brockett BB, Deming JW (2000) Remarkably low tempera- Jones SL, Drouin P, Wilkinson BJ, Morse PD (2002) Correlation of long-range
ture optima for extracellular enzyme activity form Arctic bacteria and sea ice. membrane order with temperature-dependent growth characteristics of
Environ Microbiol 2:383–388 parent and a cold-sensitive, branched-chain-fatty-acid-deficient mutant of
Ianotti EL, Kafkewitz D, Wolin MJ, Bryant MP (1973) Glucose fermentation Listeria monocytogenes. Arch Microbiol 177:217–222
products of Ruminococcus albus grown in continuous culture with Vibrio Jørgensen BB (1982) Ecology of the bacteria of the sulphur cycle with special
succinogenes. J Bacteriol 114:1231–1240 reference to anoxic-oxic interface environments. Philos Trans R Soc Lond
Imhoff JF (1986) Osmoregulation and compatible solutes in eubacteria. FEMS B Biol Sci 298:543–561
Microbiol Rev 39:57–66 Jost JL, Drake JF, Fredrickson AG, Tsuchiya HM (1973) Interactions of Tetrahy-
Ingham CJ, Sprenkels A, Bomer J, Molenaar D, van der Berg A, van Hylckama mena pyriformis, Escherichia coli, Azotobacter vinelandii and glucose in
Vlieg JET, de Vos WM (2007) The micro-Petri dish, a million-well growth a minimal medium. J Bacteriol 113:834–840
chip for the culture and high-throughput screening of microorganisms. Proc Jung L, Jost R, Stoll E, Zuber H (1974) Metabolic differences in Bacillus stearother-
Natl Acad Sci USA 104:18217–18222 mophilus grown at 55 C and 37 C. Arch Microbiol 95:125–138
Ingraham JL (1962) Temperature relationships. In: Gunsalus IC, Stanier RY (eds) Junge K, Eicken H, Deming JW (2004) Bacterial activity at 2 to 20 C in Arctic
The bacteria, vol 4. Academic, New York, pp 265–296 wintertime sea ice. Appl Environ Microbiol 70:550–557
Ingraham JL, Marr AG (1996) Effect of temperature, pressure, pH, and osmotic Kaeberlein T, Lewis K, Epstein SS (2002) Isolating the ‘‘uncultivable’’ microor-
stress on growth. In: Neidhardt FC (ed) Escherichia coli and Salmonella ganisms in pure culture in a simulated natural environment. Science
typhimurium. ASM Press, Washington, DC, pp 1570–1578 296:1127–1129
Irgens RL, Gosink JJ, Staley JT (1996) Polaromonas vacuolata gen. nov., sp. nov., Kalmbach S, Manz W, Szewyk U (1997) Isolation of new bacterial species from
a psychrophilic, marine gas vacuolate bacterium from Antarctica. Int J Syst drinking water biofilms and proof of their in situ dominance with highly
Evol Microbiol 46:822–826 specific 16S rRNA probes. Appl Environ Microbiol 63:4164–4170
Isaksen MF, Jørgensen BB (1996) Adaptation of psychrophilic and psychrotrophic Kämpfer P, Rainey FA, Andersson MA, Nurmiaho Lassila EL, Ulrych U, Busse HJ,
sulfate-reducing bacteria to permanently cold marine environments. Appl Weiss N, Mikkola R, Salkinoja-Salonen M (2000) Frigoribacterium faeni gen.
Environ Microbiol 62:408–414 nov., sp. nov., a novel psychrophilic genus of the family Microbacteriaceae.
Ishii A, Sato T, Wachi M, Nagai K, Kato C (2004) Effects of high hydrostatic Int J Syst Evol Microbiol 50:355–363
pressure on bacterial cytoskeleton FtsZ polymers in vivo and in vitro. Kämpfer P, Buczolits S, Albrecht A, Busse H-J, Stackebrandt E (2003) Towards
Microbiology 150:1965–1972 a standardized format for the description of a novel species (of an established
Jackson BE, McInerney MJ (2002) Anaerobic microbial metabolism can proceed genus): Ochrobactrum gallinifaecis sp. nov. Int J Syst Evol Microbiol 53:
close to thermodynamic limits. Nature 415:454–456 893–896
Jacobi CA, Aßmus B, Reichenbach H, Stackebrandt E (1997) Molecular evidence Kane MD, Poulsen LK, Stahl DA (1993) Monitoring the enrichment and isolation
for association between the Sphingobacterium-like organism ‘‘Candidatus of sulfate-reducing bacteria by using oligonucleotide probes designed from
environmentally derived 16S rRNA sequences. Appl Environ Microbiol Kristjansson JR, Schönheit P, Thauer RK (1982) Different Km values for hydrogen
59:682–686 of methanogenic bacteria and sulfate reducing bacteria: an explanation for
Kaneda T (1991) Iso-fatty and anteiso-fatty acids in bacteria-biosynthesis, func- the apparent inhibition of methanogenesis by sulfate. Arch Microbiol
tion, and taxonomic significance. Microbiol Rev 55:288–302 131:278–282
Kaprelyants AS, Kell DB (1993) Dormancy in stationary-phase cultures of Micro- Krulwich TA, Guffanti AA (1983) Physiology of acidophilic and alkalophilic
coccus luteus—flow cytometric analysis of starvation and resuscitation. Appl bacteria. Adv Microb Physiol 24:173–214
Environ Microbiol 59:3187–3196 Krulwich TA, Guffanti AA (1989) Alkalophilic bacteria. Annu Rev Microbiol
Kaprelyants AS, Mukamolova GV, Kell DB (1994) Estimation of dormant Micro- 43:435–463
coccus luteus cells by penicillin lysis and by resuscitation in cell free spent Kuenen JG, Gottschal JC (1982) Competition among chemolithotrophs and
culture medium at high dilution. FEMS Microbiol Lett 115:347–352 methylotrophs and their interactions with heterotrophic bacteria. In: Bull
Kaprelyants AS, Mukamolova GV, Davey HM, Kell DB (1996) Quantitative AT, Slater JH (eds) Microbial interactions and communities, vol 1. Aca-
analysis of the physiological heterogeneity within starved cultures of Micro- demic, London, pp 153–187
coccus luteus by flow cytometry and cell sorting. Appl Environ Microbiol Kuenen JG, Harder W (1982) Microbial competition in continuous culture. In:
62:1311–1316 Burns RG, Slater JH (eds) Experimental microbial ecology. Blackwell,
Karl DM, Bird DF, Björkman K, Houlihan T, Shackelford R, Tupas L (1999) Oxford, pp 342–367
Microorganisms in the accreted ice of Lake Vostok, Antarctica. Science Kuenen JG, Robertson LA (1984) Competition among chemolithotrophic bacte-
286:2144–2147 ria under aerobic and anaerobic conditions. In: Klug MJ, Reddy CA (eds)
Karner M, Fuhrman JA (1997) Determination of active bacterioplankton: Current perspectives in microbial ecology. ASM Press, Washington, DC,
a comparison of universal 16S rRNA probes, autoradiography, and nucleoid pp 306–313
staining. Appl Environ Microbiol 63:1208–1213 Kuenen JG, Robertson LA, van Gemerden H (1985) Microbial interactions
Kato Y, Sakala RM, Hayashidani H, Kiuchi A, Kaneuchi C, Ogawa M (2000) Lacto- among aerobic and anaerobic sulfur-oxidizing bacteria. Adv Microbiol
bacillus algidus sp. nov., a psychrophilic lactic acid bacterium isolated from Ecol 8:1–59
vacuum-packaged refrigerated beef. Int J Syst Evol Microbiol 50:1143–1149 Kushner DJ (1978) Life in high salt and solute concentrations: halophilic bacteria.
Kawato M, Shinobu R (1959) On Streptomyces herbaricolor nov. sp.; supplement: In: Kushner DJ (ed) Microbial life in extreme environments. Academic,
a simple technique for the microscopical observation. Mem Osaka Univ Lib London, pp 318–368
Arts Educ 8:114 Kuske CR, Barns SM, Busch JD (1997) Diverse uncultivated bacterial groups from
Keith SM, Herbert RA (1985) The application of compound bi-directional flow soils of the arid southwestern United States that are present in many geo-
diffusion chemostats to the study of microbial interactions. FEMS Microbiol graphical regions. Appl Environ Microbiol 63:3614–3621
Ecol 31:239–248 Kuznetsov SI, Dubinina GA, Laptev NA (1979) Biology of oligotrophic bacteria.
King GM (1984) Utilization of hydrogen, acetate, and ‘‘noncompetitive’’ substrates Annu Rev Microbiol 33:377–387
by methanogenic bacteria in marine sediments. Geomicrobiol J 3:275–306 Laanbroek HJ, Veldkamp H (1982) Microbial interactions in sediment commu-
King D, Nedwell DB (1984) Changes in the nitrate-reducing community of an nities. Philos Trans R Soc Lond B Biol Sci 297:533–550
anaerobic saltmarsh sediment in response to seasonal selection by tempera- Laanbroek HJ, Smit AJ, Klein Nulend G, Veldkamp H (1979) Competition for L-
ture. J Gen Microbiol 130:2935–2941 glutamate between specialized and versatile Clostridium species. Arch
Kjaergaard L, Jørgensen BB (1979) Redox potential as a state variable in fermen- Microbiol 120:61–66
tation systems. Biotechnol Bioeng Symp 9:85–94 Laanbroek HJ, Geerlings HJ, Peynenburg AACM, Siesling J (1983) Competition
Kjelleberg S, Humphrey BA, Marshall KC (1982) Effect of interfaces on small, for L-lactate between Desulfovibrio, Veillonella, and Acetobacterium species
starved bacteria. Appl Environ Microbiol 43:1166–1172 isolated from anaerobic intertidal sediments. Microb Ecol 9:341–354
Kjelleberg S, Flardh KBG, Nystrom T, Moriarty DJW (1993) Growth limitation Laanbroek HJ, Geerlings HJ, Sijtsma L, Veldkamp H (1984) Competition for
and starvation in bacteria. In: Ford TE (ed) Aquatic microbiology: an sulfate and ethanol among Desulfobacter, Desulfobulbus, and Desulfovibrio
ecological approach. Blackwell, Oxford, pp 298–320 species isolated from intertidal sediments. Appl Environ Microbiol 47:
Kluyver AJ, Donker HJL (1926) Unity in biochemistry. Chem Zelle Gewebe 329–334
13:134–190 Lange W (1971) Enhancement of algal growth in cyanophyta-bacteria systems by
Knoblauch C, Jørgensen BB (1999) Effect of temperature on sulphate reduction, carbonaceous compounds. Can J Microbiol 17:303–314
growth rate and growth yield in five psychrophilic sulphate-reducing bacte- Lange R, Hengge-Aronis R (1991) Identification of a central regulator of station-
ria from Arctic sediments. Environ Microbiol 1:457–467 ary-phase gene expression in Escherichia coli. Mol Microbiol 5:49–59
Knoblauch C, Jørgensen BB, Harder J (1999) Community size and metabolic Langworthy TA (1978) Microbial life in extreme pH values. In: Kushner DJ (ed)
rates of psychrophilic sulfate-reducing bacteria in arctic marine sediments. Microbial life in extreme environments. Academic, New York, pp 279–317
Appl Environ Microbiol 65:4230–4233 Lapierre L, Undeland P, Cox LJ (1992) Lithium chloride-sodium propionate agar
Koch IH, Gich F, Dunfield PF, Overmann J (2008) Edaphobacter modestus gen. for the enumeration of bifidobacteria in fermented dairy products. J Dairy
nov., sp. nov., and Edaphobacter aggregans sp. nov., two novel acidobacteria Sci 75:1192–1196
isolated from alpine and forest soils. Int J Syst Evol Bacteriol 58:1114–1122 Larsen H (1986) Halophilic and halotolerant microorganisms—an overview and
Kogure K, Simidu U, Taga N (1979) A tentative direct microscopic method for historical perspective. FEMS Microbiol Rev 39:3–7
counting living marine bacteria. Can J Microbiol 25:415–420 Law AT, Button DK (1977) Multiple-carbon-source-limited growth kinetics of
Kole MM, Draper I, Gerson DF (1988) Protease production by Bacillus subtilis in a marine coryneform bacterium. J Bacteriol 129:115–123
oxygen controlled, glucose fed-batch fermentations. Appl Microbiol Leadbetter JR, Schmidt TM, Graber JR, Breznak JA (1999) Acetogenesis from H2
Biotechnol 28:404–408 Plus CO2 by Spirochetes from Termite Guts. Science 283:686–689
Kolenbrander PE, London J (1993) Adhere today, here tomorrow: oral bacterial Lee IH, Fredrickson AG, Tsuchiya HM (1976) Dynamics of mixed cultures of
adherence. J Bacteriol 175:3247–3252 Lactobacillus plantarum and Propionibacterium shermanii. Biotechnol
Kolter R, Siegele DA, Tormo A (1993) The stationary phase of the bacterial life Bioeng 18:513–526
cycle. Annu Rev Microbiol 47:855–874 Lee N, Nielsen PH, Andrasen KH, Juretschko S, Nielsen JL, Schleifer K-H, Wagner
Könneke M, Bernhard AE, de la Torre JR, Walker CB, Waterbury JB, Stahl DA M (1999) Combination of fluorescent in situ hybridization and microautor-
(2005) Isolation of an autotrophic ammonia-oxidizing marine archaeon. adiography—a new tool for structure: function analyses in microbial ecol-
Nature 437:543–645 ogy. Appl Environ Microbiol 65:1289–1297
Krembs C, Juhl AR, Long RA, Azam F (1998) Nanoscale patchiness of bacteria in Lee KC-Y, Dunfield PF, Morgan XC, Crowe MA, Houghton KM, Vyssotski M,
lake water studied with the spatial information preservation method. Limnol Ryan JLJ, Lagutin K, McDonald IR, Stott MB (2011) Chthonomonas
Oceanogr 43:307–314 calidirosea gen. nov., sp. nov., an aerobic, pigmented, thermophilic
micro-organism of a novel bacterial class, Chthonomonadetes classis nov., of Lum KT, Meers PD (1989) Boric acid converts urine into an effective bacterio-
the newly described phylum Armatimonadetes originally designated candi- static transport medium. J Infect 18:51–58
date division OP10. Int J Syst Evol Microbiol 61:2482–2490 MacDonald AG (1984) The effects of pressure on the molecular structure and
Leedle JAZ, Hespell RB (1980) Differential carbohydrate media and anaerobic physiological functions of cell membranes. Philos Trans R Soc Lond B Biol
replica plating technique in delineating carbohydrate utilizing subgroups in Sci 304:47–68
rumen bacterial populations. Appl Environ Microbiol 39:709–719 Macy JM, Snellen JE, Hungate RE (1972) Use of syringe methods for anaerobiosis.
Legan JD, Owens JD (1988) Bacterial competition for methylamine: computer Am J Clin Nutr 25:1318–1323
simulation of a three-strain continuous culture supplied continuously or Madigan MT (1998) Isolation and characterization of psychrophilic purple
alternatively with two nutrients. FEMS Microbiol Ecol 53:307–314 bacteria from Antarctica. In: Peschek GA, Löffelhardt W, Schmetterer G
Legan JD, Owens JD, Chilvers GA (1987) Competition between specialist and (eds) The phototrophic prokaryotes. Kluwer/Plenum, New York,
generalist methylotrophic bacteria for an intermittent supply of methyl- pp 699–706
amine. J Gen Microbiol 133:1061–1073 Madigan MT, Martinko JM, Parker J (2000a) Biology of microorganisms. Pren-
Lengeler JW, Drews G, Schlegel HG (1999) Biology of the prokaryotes. Thieme, tice-Hall International, Upper Saddle River
New York Madigan MT, Jung DO, Woese CR, Achenbach LA (2000b) Rhodoferax antarcticus
Li L, Kato C, Nogi Y, Horikoshi K (1998) Distribution of the pressure-regulated sp. nov., a moderately psychrophilic purple nonsulfur bacterium isolated
operons in deep-sea bacteria. FEMS Microbiol Lett 159:159–166 from an Antarctic microbial mat. Arch Microbiol 173:269–277
Licht TR, Tolker-Nielsen T, Holmstrøm K, Krogfelt KA, Molin S (1999) Inhibi- Malone AS, Shellhammer TH, Courtney PD (2002) Effects of high pressure on the
tion of Escherichia coli precursor-16S rRNA processing by mouse intestinal viability, morphology, lysis and cell wall hydrolase activity of Lactococcus
contents. Environ Microbiol 1:23–32 lactis subsp. cremoris. Appl Environ Microbiol 68:4357–4363
Liesack W, Janssen PH, Rainey FA, Ward-Rainey NL, Stackebrandt E (1997) Mannisto MK, Schumann P, Rainey FA, Kampfer P, Tsitko I, Tiirola MA,
Microbial diversity in soil: the need for a combined approach using molec- Salkinoja-Salonen MS (2000) Subtercola boreus gen. nov., sp. nov., and
ular and cultivation techniques. In: van Elsas JD, Trevors JT, Wellington Subtercola frigoramans sp. nov., two new psychrophilic actinobacteria
EMH (eds) Modern soil microbiology. Marcel Dekker, New York, isolated from boreal groundwater. Int J Syst Evol Microbiol 50:1731–1739
pp 375–439 Margesin R, Spröer C, Zhang DC, Busse HJ (2011) Polaromonas glacialis sp. nov.
Lilburn TG, Kim KS, Ostrom NE, Byzek KR, Leadbetter JR, Breznak JA and Polaromonas cryoconiti sp. nov., two novel bacteria from alpine glacier
(2001) Nitrogen fixation by symbiotic and free-living spirochetes. Science cryoconite. Int J Syst Evol Microbiol. doi:10.1099/ijs.0.037556-0
292:2495–2498 Margesin R, Zhang DC, Busse HJ (2012) Sphingomonas alpina sp. nov.,
Lins U, Farina M (1999) Organization of cells in magnetotactic multicellular a psychrophilic bacterium isolated from alpine soil. Int J Syst
aggregates. Microbiol Res 154:9–13 Evol Microbiol 62:1558–1563
Little B, Gerchakov S, Udey L (1987) A method for sterilization of natural Margulis L (1981) Symbiosis in cell evolution. Freeman, San Francisco
seawater. J Microbiol Methods 7:193–200 Marquis RE (1976) High pressure microbial physiology. Adv Microb Physiol
Loewen PC, Hengge-Aronis R (1994) The role of the sigma factor s6 (Kat F) in 14:159–239
bacterial global regulation. Annu Rev Microbiol 48:53–80 Marquis RE, Matsumura P (1978) Microbial life under pressure. In: Kushner DJ
Loewen PC, Hu B, Stutinsky J, Sparling R (1998) Regulation in the rpoS regulon (ed) Microbial life in extreme environments. Academic, London, pp 105–158
of Escherichia coli. Can J Microbiol 44:707–717 Marschall E, Jogler M, Henssge U, Overmann J (2010) Large scale distribution
Long RA, Azam F (2001) Antagonistic interactions among marine pelagic bacte- and activity patterns of an extremely low-light adapted population of green
ria. Appl Environ Microbiol 67:4975–4983 sulfur bacteria in the Black Sea. Environ Microbiol 12:1348–1362
Lonhienne T, Gerday C, Feller G (2000) Psychrophilic enzymes: revisiting the Martens-Habbena W, Berube PM, Urakawa H, de la Torre JR, Stahl DA
thermodynamic parameters of activation may explain local flexibility. (2009) Ammonia oxidation kinetics determine niche separation of nitrifying
Biochim Biophys Acta 1543:1–10 Archaea and Bacteria. Nature 461:976–979
Loomis WD, Durst RW (1992) Chemistry and biology of boron. Biofactors Martin GA, Hempfling WP (1976) A method for the regulation of microbial
3:229–239 population density during continuous culture at high growth rates. Arch
Loveland-Curtze J, Sheridan PP, Gutshall KR, Brenchley JE (1999) Biochemical Microbiol 107:41–47
and phylogenetic analyses of psychrophilic isolates belonging to the Martin SE, Flowers RS, Ordal ZJ (1976) Catalase: its effect on microbial enumer-
Arthrobacter subgroup and description of Arthrobacter psychrolactophilus, ation. Appl Environ Microbiol 32:731–734
sp. nov. Arch Microbiol 171:355–363 Maruyama A, Honda D, Yamamoto H, Kitamura K, Higashihara T (2000) Phy-
Lovitt RW, Wimpenny JWT (1981) The gradostat: a bidirectional compound logenetic analysis of psychrophilic bacteria isolated from the Japan Trench,
chemostat and its application in microbiological research. J Gen Microbiol including a description of the deep-sea species Psychrobacter pacificensis sp.
127:261–268 nov. Int J Syst Evol Microbiol 50:835–846
Lovley DR, Blunt-Harris EL (1999) Role of humic-bound iron as an electron Mason CA, Egli T (1993) Dynamics of microbial growth in the decelerating and
transfer agent in dissimilatory Fe(III) reduction. Appl Environ Microbiol stationary phase of batch culture. In: Kjelleberg S (ed) Starvation in bacteria.
65:4252–4254 Plenum, New York, pp 81–98
Lovley DR, Dwyer DF, Klug MJ (1982) Kinetic analysis of competition between Matin A (1981) Regulation of enzyme synthesis as studied in continuous culture.
sulfate reducers and methanogens for hydrogen in sediments. Appl Environ In: Calcott PH (ed) Continuous culture of cells, vol 2. CRC Press, Boca
Microbiol 43:1373–1379 Raton, pp 69–97
Lovley DR, Coates JD, Blunt-Harris EL, Phillips EJP, Woodward JC (1996a) Matin A (1990) Keeping a neutral cytoplasm: the bioenergetics of obligate
Humic substances as electron acceptors for microbial respiration. Nature acidophiles. FEMS Microbiol Rev 75:307–318
382:445–448 Matin A, Veldkamp H (1978) Physiological basis of the selective advantage of
Lovley DR, Woodward JC, Chapelle FH (1996b) Rapid anaerobic benzene oxi- a Spirillum sp. in a carbon-limited environment. J Gen Microbiol 105:
dation with a variety of chelated Fe(III) forms. Appl Environ Microbiol 187–197
62:288–291 Mazur P (1980) Limits to life at low temperatures and at reduced water contents
Lovley DR, Fraga JL, Coates JD, Blunt-Harris EL (1999) Humics as an electron and water activities. Orig Life 10:137–159
donor for anaerobic respiration. Environ Microbiol 1:89–98 McDougald D, Kjelleberg S (1999) New perspectives on the viable but
Ludwig W, Bauer SH, Bauer H, Held I, Kirchhof G, Schulze R, Huber I, Spring S, nonculturable response. Biologia 54:617–623
Hartmann A, Schleifer K-H (1997) Detection and in situ identification of McInerney MJ, Bryant MD, Pfennig N (1979) Anaerobic bacterium that degrades
representatives of a widely distributed bacterial phylum. FEMS Microbiol fatty acids in syntrophic association with methanogens. Arch Microbiol
Lett 153:181–190 122:129–135
McInerney MJ, Bryant MP, Hespell RB, Costerton JW (1981) Syntrophomonas Müller RH, Babel W (1996) Measurement of growth at very low rates (m ≧ 0), an
wolfei gen. nov. sp. nov., an anaerobic, syntrophic, fatty acid-oxidizing approach to study the energy requirement for the survival of Alcaligenes
bacterium. Appl Environ Microbiol 41:1029–1039 eutrophus JMP 134. Appl Environ Microbiol 62:147–151
Meers JL (1973) Growth of bacteria in mixed cultures CRC. Crit Rev Microbiol Munro PM, Flatau GN, Clément RL, Gauthier MJ (1995) Influence of the RpoS
2:139–184 (KatF) sigma factor on maintenance of viability and culturability of
Megee RD 3rd, Drake JF, Fredrickson AG, Tsuchiya HM (1972) Studies in Escherichia coli and Salmonella typhimurium in seawater. Appl Environ
intermicrobial symbiosis. Saccharomyces cerevisiae and Lactobacillus casei. Microbiol 61:1853–1858
Can J Microbiol 18:1733–1742 Mur LR, Gons HJ, van Liere L (1977) Some experiments on the competition
Meldrum FC, Heywood BR, Mann S, Frankel RB, Bazylinski DA (1993) Electron between green algae and blue-green bacteria in light-limited environments.
microscopy study of magnetosomes in a cultured coccoid magnetotactic FEMS Microbiol Lett 1:335–338
bacterium. Proc R Soc Lond B Biol Sci 251:231–236 Murray DR (1989) Biology of food radiation. Wiley, Chichester
Meyer JS, Tsuchiya HM, Fredrickson AG (1975) Dynamics of mixed populations Myers J, Clark LB (1944) Culture conditions and the development of the photo-
having complementary metabolism. Biotechnol Bioeng 17:1065–1081 synthetic mechanism. II: an apparatus for the continuous culture of Chlo-
Michels PAM, Michels JPJ, Boonstra J, Konings WN (1979) Generation of an rella. J Gen Physiol 28:103–112
electrochemical proton gradient in bacteria by the excretion of metabolic Nakamura I, Ogimoto K, Izumu H (1995) ATP-dependent calcium release from
end-products. FEMS Microbiol Lett 5:357–364 binding site in Streptococcus bovis: bound versus free pools. J Gen Appl
Mikx FJM, van der Hoeven JS (1975) Symbiosis of Streptococcus mutans and Microbiol 41:389–398
Veillonella alcalescens in mixed continuous cultures. Arch Oral Biol 20: Nakasone K, Ikegami A, Kato C, Usami R, Horikoshi K (1998) Mechanisms of
407–410 gene expression controlled by pressure in deep-sea microorganisms.
Miller TL, Wolin MJ (1974) A serum bottle modification of the Hungate tech- Extremophiles 2:149–154
nique for cultivating obligate anaerobes. Appl Microbiol 27:985–987 National Center for Biotechnology Information (2012) http://www.ncbi.nlm.nih.
Miura Y, Tanaka H, Okazaki M (1980) Stability analysis of commensal and gov/genomes/MICROBES/microbial_growth.html
mutual relations with competitive assimilation in continuous mixed culture. Nedwell DB (1984) The input and mineralization of organic carbon in anaerobic
Biotechnol Bioeng 22:929–948 aquatic sediments. Adv Microbiol Ecol 7:93–131
Miyamoto-Shinohara Y, Sukenobe J, Imaizumi T, Nakahara T (2006) Survival Nedwell DB, Rutter M (1994) Influence of temperature on growth rate and
curves for microbial species stored by freeze-drying. Cryobiology 52:27–32 competition between two psychrotolerant Antarctic bacteria: low tempera-
Mizunoe Y, Wai SN, Takade A, Yoshida S (1999) Restoration of culturability of ture diminishes affinity for substrate uptake. Appl Environ Microbiol
starvation-stressed and low-temperature-stressed Escherichia coli O157 cells 60:1984–1992
using H2O2-degrading compounds. Arch Microbiol 172:63–67 Neidhardt FC, Umbarger HE (1996) Chemical composition of Escherichia coli. In:
Moench TT, Zeikus JG (1983) An improved preparation method for a titanium Neidhardt FC (ed) Escherichia coli and Salmonella, 2nd edn. ASM Press,
(III) media reductant. J Microbiol Methods 1:199–202 Washington, DC, pp 13–16
Moissl C, Rudolph C, Huber R (2002) Natural communities of novel archaea and Nelson DC, Jørgensen BB, Revsbech NP (1986) Growth pattern and yield of
bacteria with a string-of-pearls-like morphology: molecular analysis of the a chemoautotrophic Beggiatoa sp. in oxygen-sulfide microgradients. Appl
bacterial partners. Appl Environ Microbiol 68:933–937 Environ Microbiol 52:225–233
Monod J (1942) Recherches sur la croissance des cultures bacteriennes. Hermann, Neuhard J, Kelln RA (1996) Biosynthesis and conversions of pyrimidines. In:
Paris Neidhardt FC (ed) Escherichia coli and Salmonella, 2nd edn. ASM Press,
Monod J (1950) La technique de culture continue: théorie et applications. Ann Washington, DC, pp 580–599
Inst Pasteur 79:390–410 Nichols DS, Greenhill AR, Shadbolt CT, Ross T, McMeekin TA (1999) Physico-
Morita RY (1975) Psychrophilic bacteria. Bacteriol Rev 39:144–167 chemical parameters for growth of the sea ice bacteria Glaciecola punicea
Morita RY (1982) Starvation—survival of heterotrophs in the marine environ- ACAM 611T and Gelidibacter sp. strain IC158. Appl Environ Microbiol
ment. Adv Microbiol Ecol 6:171–198 65:3757–3760
Morita RY (1986) Pressure as an extreme environment. In: Herbert RA, Nichols D, Lewis K, Orjala J, Mo S, Ortenberg R, O’Connor P, Zhao C, Vouros P,
Codd GA (eds) Microbes in extreme environments. Academic, London, Kaeberlein T, Epstein SS (2008) Short peptide induces an ‘‘uncultivable’’
pp 171–185 microorganism to grow in vitro. Appl Environ Microbiol 74:4889–4897
Morotomi M, Nagai F, Watanabe Y (2011) Parasutterella secunda sp. nov., isolated Niehaus F, Hantke K, Unden G (1991) Iron content and FNR-dependent gene
from human faeces and proposal of Sutterellaceae fam. nov. in the order regulation in Escherichia coli. FEMS Microbiol Lett 84:319–324
Burkholderiales. Int J Syst Evol Microbiol 61:637–643 Nielsen PH, de Muro MA, Nielsen JL (2000) Studies on the in situ physiology of
Mossel DAA, Veldman A, Eelderink I (1980) Comparison of the effects of liquid Thiothrix spp. present in activated sludge. Environ Microbiol 2:389–398
medium repair and the incorporation of catalase in Macconkey type media Nogales B, Guerrero R, Esteve I (1997) A heterotrophic bacterium inhibits
on the recovery of Enterobacteriaceae sublethally stressed by freezing. J Appl growth of several species of the genus Chlorobium. Arch Microbiol 167:
Bacteriol 49:405–419 396–399
Mountfort DO, Asher RA (1986) Isolation from a methanogenic ferulate Nogales B, Moore ERB, Abraham WR, Timmis KN (1999) Identification of the
degrading consortium of an anaerobe that converts methoxyl groups of metabolically active members of a bacterial community in a polychlorinated
aromatic acids to volatile fatty acids. Arch Microbiol 144:55–61 biphenyl-polluted moorland soil. Environ Microbiol 1:199–212
Mountfort DO, Bryant MP (1985) Isolation and characterization of an anaerobic Norland S, Fagerbakke KM, Heldal M (1995) Light element analysis of individual
syntrophic benzoate degrading bacterium from sewage sludge. Arch bacteria by X-ray microanalysis. Appl Environ Microbiol 61:1357–1362
Microbiol 133:249–256 Norris JR, Ribbons DW (1970) Methods in microbiology, vol 3. Academic,
Mountfort DO, Rainey FA, Burghardt J, Kaspar HF, Stackebrandt E (1997) New York, pp 1–506
Clostridium vincentii sp. nov., a new anaerobic, saccharolytic, psychrophilic Notley L, Ferenci T (1996) Induction of RpoS-dependent functions in glucose-
bacterium isolated from low-salinity pond sediment of the McMurdo ice limited continuous culture: what levels of nutrient limitation induces the
shelf, Antarctica. Arch Microbiol 167:54–60 stationary phase of Escherichia coli. J Bacteriol 178:1465–1468
Mountfort DO, Rainey FA, Burghardt J, Kaspar HF, Stackebrandt E (1998) Novick A, Szilard L (1950) Description of the chemostat. Science 112:715–716
Psychromonas antarcticus gen. nov., sp. nov., a new aerotolerant anaerobic, Nozhevnikova AN, Simankova MV, Parshina SN, Kotsyurbenko OR (2001)
halophilic psychrophile isolated from pond sediment of the McMurdo ice Temperature characteristics of methanogenic archaea and acetogenic
shelf, Antarctica. Arch Microbiol 169:231–238 bacteria isolated from cold environments. Water Sci Technol 44:41–48
Mukamolova GV, Kaprelyants AS, Young DI, Young M, Kell DB (1998) A bacterial Nurmikko V (1956) Biochemical factors affecting symbiosis among bacteria.
cytokine. Proc Natl Acad Sci USA 95:8916–8921 Experientia 12:245–249
Oerther DB, Pernthaler J, Schramm A, Amann R, Raskin L (2000) Monitoring Overmann J, van Gemerden H (2000) Microbial interactions involving sulfur
precursor 16S rRNAs of Acinetobacter spp. in activated sludge wastewater bacteria: implications for the ecology and evolution of bacterial communi-
treatment systems. Appl Environ Microbiol 66:2154–2165 ties. FEMS Microbiol Rev 24:591–599
Ohmura N, Matsumoto N, Sasaki K, Saiki H (2002) Electrochemical regeneration Overmann J, Lehmann S, Pfennig N (1991) Gas vesicle formation and buoyancy
of Fe(III) to support growth on anaerobic iron respiration. Appl Environ regulation in Pelodictyon phaeoclathratiforme (green sulfur bacteria). Arch
Microbiol 68:405–407 Microbiol 157:29–37
Okada T, Ueyama K, Niya S, Kanazawa H, Futai M, Tsuchiya T (1981) Role of Overmann J, Beatty JT, Hall KJ (1996) Purple sulfur bacteria control the growth of
inducer exclusion in preferential utilization of glucose over melibiose in aerobic heterotrophic bacterioplankton in a meromictic salt lake. Appl
diauxic growth of Escherichia coli. J Bacteriol 146:1030–1037 Environ Microbiol 62:3251–3258
Oliver JD (1995) The viable but non-culturable state in the human pathogen Overmann J, Tuschak C, Fröstl J, Sass H (1998) The ecological niche of the
Vibrio vulnificus. FEMS Microbiol Lett 133:203–208 consortium ‘‘Pelochromatium roseum’’. Arch Microbiol 169:120–128
Olsen RA, Bakken LR (1987) Viability of soil bacteria: optimization of plate- Padan E, Schuldinger S (1986) Intracellular pH regulation in bacterial cells.
counting technique and comparison between total counts and plate counts Methods Enzymol 125:337–352
within different size groups. Microb Ecol 13:59–74 Paerl H (1978) Role of heterotrophic bacteria in promoting N2-fixation by
Oltmann LF, Schoenmaker GS, Reijnders WNM, Stouthamer AH (1978) Modifi- Anabaena in aquatic habitats. Microb Ecol 4:215–231
cation of the pH auxostat culture method for the mass cultivation of Paerl HW (1982) Interactions with bacteria. In: Carr NG, Whitton NG (eds) The
bacteria. Biotechnol Bioeng 20:921–925 biology of cyanobacteria. University of California Press, Los Angeles,
Orcutt KM, Rasmussen U, Webb EA, Waterbury JB, Gundersen K, Bergman pp 441–461
B (2002) Characterization of Trichodesmium spp. by genetic techniques. Pankratov TA, Dedysh SN (2011) Granulicella paludicola gen. nov., sp. nov.,
Appl Environ Microbiol 68:2236–2245 Granulicella pectinivorans sp. nov., Granulicella aggregans sp. nov. and
Oremland RS (1988) Biogeochemistry of methanogenic bacteria. In: Zehnder AJB Granulicella rosea sp. nov., acidophilic, polymer-degrading acidobacteria
(ed) Biology of anaerobic microorganisms. Wiley, New York, pp 641–706 from Sphagnum peat bogs. Int J Syst Evol Microbiol 60:2951–2959
Oremland RS, Hoeft SE, Santini JM, Bano N, Hollibough RA (2002) Anaerobic Parkes RJ, Senior E (1988) Multistage chemostats and other models for
oxidation of arsenite in Mono Lake water and by a facultative, arsenite- studying anoxic ecosystems. In: Wimpenny JWT (ed) Handbook of labora-
oxidizing chemoautotroph, strain MLHE-1. Appl Environ Microbiol tory model systems for microbial ecosystems, vol 1. CRC Press, Boca Raton,
68:4795–4802 pp 51–71
Oren A (1986) Intracellular salt concentrations of the halophilic eubacteria Pastan I, Perlman R (1969) Repression of b-galactosidase synthesis by glucose in
Haloaerobium praevalens and Halobacteroides halobius. Can J Microbiol phosphotransferase mutants of Escherichia coli. Repression in the absence of
32:4–9 glucose phosphorylation. J Biol Chem 244:5836–5842
Oren A (1999) Bioenergetic aspects of halophilism. Microbiol Mol Biol Rev Pasteur L (1862) Mémoire sur les corpuscules qui existent dans l’atmosphère:
63:334–348 examen de la doctrine des génerations spontanées. Ann Chém Phys 64:5–110
Ostrowski M, Cavicchioloi R, Blaauw M, Gottschal JC (2001) Specific growth rate Pernthaler J, Posch T, Simek K, Vrba J, Pernthaler A, Glöckner FO, Nübel U,
plays a critical role in hydrogen peroxide resistance of the marine oligotro- Psenner R, Amann R (2001) Predator-specific enrichment of Actinobacteria
phic ultramicrobacterium Sphingomonas alaskensis strain RB2256. Appl from a cosmopolitan freshwater clade in mixed continuous culture. Appl
Environ Microbiol 67:1292–1299 Environ Microbiol 67:2145–2155
Ott JA, Novak R, Schiemer F, Hentschel U, Nebelsiek M, Polz M (1991) Tackling Pfennig N (1980) Syntrophic mixed cultures and symbiotic consortia with
the sulfide gradient: a novel strategy involving marine nematodes and che- phototrophic bacteria: a review. In: Gottschalk G, Pfennig N, Werner HH
moautotrophic ectosymbionts. Mar Ecol 12:261–279 (eds) Anaerobes and anaerobic infections. Fischer, Stuttgart, pp 127–131
Otto R, Hugenholtz J, Konings WN, Veldkamp H (1980) Increase of molar Pfennig N (1993) Reflections of a microbiologist, or how to learn from the
growth yield of Streptococcus cremoris for lactose as a consequence of lactate microbes. Annu Rev Microbiol 47:1–29
consumption by Pseudomonas stutzeri in mixed culture. FEMS Microbiol Pfennig N, Trüper HG (1989) Anoxygenic phototrophic bacteria. In: Staley JT,
Lett 9:85–88 Bryant MP, Pfennig N, Holt JG (eds) Bergey’s manual of systematic bacteri-
Ouverney CC, Fuhrman JA (1999) Combined microautoradiography—16S ology, vol 3. Williams and Wilkins, Baltimore, pp 1635–1709
rRNA probe technique for determination of radioisotope uptake by specific Pietzsch O (1967) Ein Nährboden zur Schwärmhemmung und gleichzeitigen
microbial cell types in situ. Appl Environ Microbiol 65:1746–1752 Unterscheidung der Proteuskeime von Salmonellen. Fleischwirtschaft
Ouverney CC, Fuhrman JA (2000) Marine planktonic Archaea take up amino 1:31–32
acids. Appl Environ Microbiol 66:4829–4833 Pinhassi J, Berman T (2003) Differential growth response of colony-forming
Overmann J (2002a) Phototrophic consortia: a tight cooperation between non- a- and g-Proteobacteria in dilution culture and nutrient addition experi-
related eubacteria. In: Seckbach J (ed) Symbiosis. Mechanisms and model ments from Lake Kinneret (Israel), the Eastern Mediterranean Sea, and the
systems. Kluwer, Dordrecht, pp 239–255 Gulf of Eilat. Appl Environ Microbiol 69:199–211
Overmann J (2002b) Principles of enrichment, isolation, cultivation, and preser- Pinhassi J, Zweifel UL, Hagström Å (1997) Dominant marine bacterioplankton
vation of bacteria. In: Dworkin M et al (eds) The prokaryotes: an evolving species found among colony-forming bacteria. Appl Environ Microbiol
electronic resource for the microbiological community, 3rd edn (latest 63:3359–3366
update release 3.11, September 2002). Springer, New York. http://ep. Pirt SJ (1974) The theory of fed batch culture with reference to the penicillin
springer-ny.com:6336/contents/ or http://141.150.157.117:8080/prokPROD/ fermentation. J Appl Chem Biotechnol 24:415–4224
index.htm Pirt SJ (1975) Principles of microbe and cell cultivation. Blackwell, Oxford
Overmann J (2005) Chemotaxis and behavioural physiology of not-yet-cultivated Ploug H, Grossart HP, Azam F, Jørgensen BB (1999) Photosynthesis, respiration,
microbes, vol 397, Methods in enzymology. Elsevier, San Diego, pp 133–147, and carbon turnover in sinking marine snow from surface waters of South-
Chap II.8 ern California Bight: implications for the carbon cycle in the ocean. Mar Ecol
Overmann J, Pfennig N (1989) Pelodictyon phaeoclathratiforme sp. nov., a new Prog Ser 179:1–11
brown-colored member of the Chlorobiaceae forming net-like colonies. Poindexter JS (1992) Dimorphic protshecate bacteria: the genera Caulobacter,
Arch Microbiol 152:401–406 Asticcacaulis, Hyphomicrobium, Pedomicrobium, Hyphomonas, and
Overmann J, Pfennig N (1992) Continuous chemotrophic growth and respiration Thiodendron. In: Balows A, Trüper HG, Dworkin M, Schleifer K-H (eds)
of Chromatiaceae species at low oxygen concentrations. Arch Microbiol The prokaryotes, 2nd edn. Springer, New York, pp 2176–2196
158:59–67 Poindexter JS, Leadbetter ER (1986) Enrichment cultures in bacterial ecology. In:
Overmann J, Schubert K (2002) Phototrophic consortia: model systems for Poindexter JS, Leadbetter ER (eds) Bacteria in nature, vol 2. Plenum, New
symbiotic interrelations between prokaryotes. Arch Microbiol 177:201–208 York, pp 229–260
Porter JR (1976) Antony van Leeuwenhoek: tercentenary of is discovery of Rose AH, Evison LM (1965) Studies on the biochemical basis of the minimum
bacteria. Microbiol Rev 40:260–269 temperature for growth of certain psychrophilic and mesophilic microor-
Pörtner R, Märkl H (1998) Dialysis cultures. Appl Microbiol Biotechnol 50: ganisms. J Gen Microbiol 38:131–141
403–414 Rosenberg E, Keller KH, Dworkin M (1977) Cell-density dependent growth of
Postgate JR, Hunter JR (1963) Acceleration of bacterial death by growth sub- Myxococcus xanthus on casein. J Bacteriol 129:770–777
strates. Nature 198:273–280 Roslev P, Iversen N (1999) Radioactive fingerprinting of microorganisms that
Postgate JR, Hunter JR (1964) Accelerated death of Aerobacter aerogenes starved in oxidize atmospheric methane in different soils. Appl Environ Microbiol
the presence of growth-limiting substrates. J Gen Microbiol 34:459–473 65:4064–4070
Powell EO (1958) Criteria for the growth of contaminants and mutants in Roszak DB, Colwell RR (1987) Survival strategies of bacteria in the natural
continuous culture. J Gen Microbiol 18:259–268 environment. Microbiol Rev 51:365–379
Pratuangdejkul J, Dharmsthiti S (2000) Purification and characterization of lipase Rouf MA (1964) Spectrochemical analysis of inorganic elements in bacteria.
from psychrophilic Acinetobacter calcoaceticus LP009. Microbiol Res 155: J Bacteriol 88:1545–1549
95–100 Ruger HJ, Fritze D, Sproer C (2000) New psychrophilic and psychrotolerant
Price B (2000) A habitat for psychrophiles in deep Antarctic ice. Proc Natl Acad Bacillus marinus strains from tropical and polar deep-sea sediments and
Sci USA 97:1247–1251 emended description of the species. Int J Syst Evol Microbiol 50:1305–1313
Pringault O, de Wit R, Caumette P (1996) A benthic gradient chamber for Russel NJ (1984) Mechanisms of thermal adaptation in bacteria: blueprints for
culturing phototrophic sulfur bacteria on reconstituted sediments. FEMS survival TIBS. Trends Biol Sci 9:108–112
Microbiol Ecol 20:237–250 Russel NJ (2000) Toward a molecular understanding of cold activity of enzymes
Pruesse E, Quast C, Knittel K, Fuchs B, Ludwig W, Peplies J, Glöckner FO from psychrophiles. Extremophiles 4:83–90
(2007) SILVA: a comprehensive online resource for quality checked and Russel NJ, Fukunaga N (1990) A comparison of thermal adaptation of membrane
aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids lipids in psychrophilic and thermophilic bacteria. FEMS Microbiol Rev
Res 35:7188–7196 74:171–182
Puskas A, Greenberg EP, Kaplan S, Schaefer AL (1997) A quorum-sensing system Russell EJ (1923) The micro-organisms of the soil. Longmans, Green, London
in the free-living photosynthetic bacterium Rhodobacter sphaeroides. Sambanis A, Fredrickson AG (1987) Long-term studies of ciliate-bacterial inter-
J Bacteriol 179:7530–7537 actions: use of a chemostat fed with bacteria grown in a separate chemostat.
Radajewski S, Ineson P, Parekh NR, Murrell JC (2000) Stable-isotope probing as J Gen Microbiol 133:1619–1630
a tool in microbial ecology. Nature 403:646–649 Sánchez O, van Gemerden H, Mas J (1996) Description of a redox-controlled
Rappe MS, Connon SA, Vergin KL, Giovannoni SJ (2002) Cultivation of the sulfidostat fro the growth of sulfide-oxidizing phototrophs. Appl Environ
ubiquitous SAR11 marine bacterioplankton clade. Nature 418:630–633 Microbiol 62:3640–3645
Ratkowsky PA, Lowry RR, McMeekin TA, Stokes AN, Chandler RE (1983) Model Sandaa R-A, Torsvik V, Enger Ø, Daae FL, Castberg T, Hahn D (1999) Analysis of
for bacterial culture growth rate throughout the entire biokinetic tempera- bacterial communities in heavy metal-contaminated soils at different levels
ture range. J Bacteriol 154:1222–1226 of resolution. FEMS Microbiol Ecol 30:237–251
Reddy GS, Aggarwal RK, Matsumoto GI, Shivaji S (2000) Arthrobacter flavus sp. Sass A, Sass H, Coolen MJL, Cypionka H, Overmann J (2001) Microbial com-
nov., a psychrophilic bacterium isolated from a pond in McMurdo Dry munities in the chemocline of a hypersaline deep-sea basin (Urania Basin,
Valley, Antarctica. Int J Syst Evol Microbiol 50:1553–1561 Mediterranean Sea). Appl Environ Microbiol 67:5392–5402
Reed RH, Walsby AE (1985) Changes in turgor pressure in response to increases Schauer NL, Brown DP, Ferry JG (1982) Kinetics of formate metabolism in
in external NaCl concentration in the gas-vacuolate cyanobacterium Methanobacterium formicicum and Methanospirillum hungatei. Appl Environ
Microcystis sp. Arch Microbiol 143:290–296 Microbiol 44:549–554
Reichenbach H (1984) Myxobacteria: a most peculiar group of social prokaryotes. Schink B (1991) Syntrophism among prokaryotes. In: Balows A, Trüper HG,
In: Rosenberg E (ed) Myxobacteria: development and cell interactions. Dworkin M, Schleifer K-H (eds) The prokaryotes, 2nd edn. Springer,
Springer, New York, pp 1–50 New York, pp 276–299
Reisch CR, Stoudemayer MJ, Varaljay VA, Amster IJ, Moran MA, Whitman WB Schink B (1997) Energetics of syntrophic cooperation in methanogenic degrada-
(2011) Novel pathway for assimilation of dimethylsulphoniopropionate tion. Microbiol Mol Biol Rev 61:262–280
widespread in marine bacteria. Nature 12:208–211 Schink B, Friedrich M (2000) Phosphite oxidation by sulphate reduction. Nature
Renesto P, Crapoulet N, Ogata H, La Scola B, Vestris G, Claveri J-M, Raoult D 406:37
(2003) Genome-based design of a cell-free culture medium for Tropheryma Schink B, Thiemann V, Laue H, Friedrich MW (2002) Desulfotignum
whipplei. Lancet 362:447–449 phosphitoxidans sp. nov., a new marine sulfate reducer that oxidizes phos-
Reusse U, Meyer A (1972) Der ‘‘Pril-Mannit-Agar’’ in der Salmonellen- phite to phosphate. Arch Microbiol 177:381–391
Diagnostik. Zbl Bakteriol I Abt Orig 219:555–557 Schlegel HG, Jannasch HW (1967) Enrichment cultures. Annu Rev Microbiol
Revsbech NP, Jørgensen BB (1986) Microelectrodes: their use in microbial ecol- 21:49–70
ogy. Adv Microbiol Ecol 9:293–352 Schlesner H (1986) Pirellula marina sp. nov., a budding peptidoglycanless bacte-
Rhee GY (1972) Competition between an algae and an aquatic bacterium for rium from brackish water. Syst Appl Microbiol 8:177–180
phosphate. Limnol Oceanogr 17:505–514 Schlesner H (1994) The development of media suitable for the microorganisms
Ricica J, Dobersky P (1981) Complex systems. In: Calcott PH (ed) Continuous morphologically resembling Planctomyces spp., Pirellula spp., and other
cultures of cells, vol 1. CRC Press, Boca Raton, pp 63–96 Planctomycetales from various aquatic habitats using dilute media. Syst
Robinson JA, Tiedje JM (1984) Competition between sulfate-reducing and Appl Microbiol 17:135–145
methanogenic bacteria for H2 under resting and growing conditions. Arch Schloss PD, Handelsman J (2004) Status of the microbial census. Microbiol Mol
Microbiol 137:26–32 Biol Rev 68:686–691
Roesch LFW, Fulthorpe RR, Riva A, Casella G, Hadwin AKM, Kent AD, Schmid M, Schmitz-Esser S, Jetten M, Wagner M (2001) 16S–23S rDNA
Daroub SH, Camargo FAO, Farmerie WG, Triplett EW (2007) Pyrosequencing intergenic spacer and 23S rDNA of anaerobic ammonium-oxidizing bacte-
enumerates and contrasts soil microbial diversity. ISME J 1:283–290 ria: implications for phylogeny and in situ detection. Environ Microbiol
Roeßler M, Müller V (1998) Quantitative and physiological analyses of chloride 3:450–459
dependence of growth of Halobacillus halophilus. Appl Environ Microbiol Schmidt GB, Rosano CL, Hurwitz C (1971) Evidence for a magnesium pump in
64:3813–3817 Bacillus cereus T. J Bacteriol 105:150–155
Roeßler M, Müller V (2002) Chloride, a new environmental signal molecule Schober A, Günther R, Schwienhorst A, Döring M, Lindemann BF (1993) Accu-
involved in gene regulation in a moderately halophilic bacterium, rate high-speed liquid handling of very small biological samples.
Halobacillus halophilus. J Bacteriol 184:6207–6215 Biotechniques 15:324–329
Scholten JCM, Conrad R (2000) Energetics of syntrophic propionate oxidation in Slonczewski J, Foster JW (1996) pH-regulated genes and survival at extreme pH.
defined batch and chemostat cultures. Appl Environ Microbiol 66: In: Neidhardt FC (ed) Escherichia coli and Salmonella, 2nd edn. ASM Press,
2934–2942 Washington, DC, pp 1539–1549
Schopfer P, Brennicke A (1999) Pflanzenphysiologie, 5th edn. Springer, New York Sly LI, Arunpairojana V (1987) Isolation of manganese-oxidizing Pedomicrobium
Schramm A, de Beer D, van den Heuvel JC, Ottengraf S, Amann R (1999) cultures from water by micromanipulation. J Microbiol Methods 6:177–182
Microscale distribution of populations and activities of Nitrosospira and Soendergaard M, Riemann B, Jörgensen NOG (1985) Extracellular organic
Nitrospira spp. along a macroscale gradient in a nitrifying bioreactor: quan- carbon (EOC) released by phytoplankton and bacterial production. Oikos
tification by in situ hybridization and the use of microsensors. Appl Environ 45:323–332
Microbiol 65:3690–3696 Sogin ML, Morrison HG, Huber JA, Welch DM, Huse SM, Neal PR, Arrieta JM,
Schulten HR, Plage B, Schnitzer M (1991) A chemical structure for humic sub- Herndl GJ (2006) Microbial diversity in the deep sea and the underexplored
stances. Naturwissenschaften 78:31–312 ‘‘rare biosphere’’. Proc Natl Acad Sci USA 103:12115–12120
Schulz E, Lüdemann H-D, Jaenicke R (1976) High-pressure equilibrium studies Somero GN (1992) Adaptations to high hydrostatic pressure. Annu Rev Physiol
on the dissociation-association of E. coli ribosomes. FEBS Lett 64:40–43 54:557–577
Schumann P, Zhang DC, Redzic M, Margesin R (2012) Alpinimonas psychrophila Spring S, Amann R, Ludwig W, Schleifer K-H, van Gemerden H, Petersen N
gen. nov., sp. nov., a novel actinobacterium of the family Microbacteriaceae (1993) Dominating role of an unusual magnetotactic bacterium in the
isolated from alpine lacier cryoconite. Int J Syst Evol Microbiol. doi:10.1099/ micraerobic zone of a freshwater sediment. Appl Environ Microbiol
ijs.0.036160-0 59:2397–2403
Schut F, de Vries EJ, Gottschal JC, Robertson BR, Harder W, Prins RA, Button DK Spring S, Schulze R, Overmann J, Schleifer K-H (2000) Identification and char-
(1993) Isolation of typical marine bacteria by dilution culture: growth, acterization of ecologically significant prokaryotes in the sediment of fresh-
maintenance, and characteristics of isolates under laboratory conditions. water lakes: molecular and cultivation studies. FEMS Microbiol Rev
Appl Environ Microbiol 59:2150–2160 24:573–590
Schut F, Jansen M, Pedro Gomes TM, Gottschal JC, Harder W, Prins RA Stal LJ, van Gemerden H, Krumbein WE (1985) Structure and development of
(1995) Substrate uptake and utilization by a marine ultramicrobacterium. a benthic marine microbial mat. FEMS Microbiol Ecol 31:111–125
Microbiology 141:351–361 Staley JT, Fuerst JA, Giovannoni SJ, Schlesner H (1992) The order
Schut F, Prins RA, Gottschal JC (1997) Oligotrophy and pelagic marine bacteria: Planctomycetales and the genera Planctomyces, Pirellula, gemmata and
facts and fiction. Aquat Microb Ecol 12:177–202 Isosphaera. In: Balows A, Dworkin M, Harder W, Schleifer K-H,
Scott DT, McKnight DM, Blunt-Harris EL, Kolesar SE, Lovley DR (1998) Quinone Trüper HG (eds) The prokaryotes. Springer, New York, pp 3710–3731
moieties act as electron acceptors in the reduction of humic substances by Stanton TB, Canale-Parola E (1979) Enumeration and selective isolation of
humic-reducing microorganisms. Environ Sci Technol 32:2984–2989 rumen spirochetes. Appl Environ Microbiol 38:965–973
Seitz AP, Nielsen TH, Overmann J (1993) Physiology of purple sulfur bacteria Stetter KO (2001) Genus III: Pyrolobus. In: Boone DR, Castenholz RW (eds)
forming macroscopic aggregates in Great Sippewissett Salt Marsh, Massa- Bergey’s manual of systematic bacteriology, vol 1. Springer, New York,
chusetts. FEMS Microbiol Ecol 12:225–236 pp 186–197
Sekiguchi Y, Takahashi H, Kamagata Y, Ohashi A, Harada H (2001) In situ Stevens TO (1995) Optimization of media for enumeration and isolation of
detection, isolation, and physiological properties of a thin filamentous aerobic heterotrophic bacteria from the deep terrestrial subsurface.
microorganism abundant in methanogenic granular sludges: a novel isolate J Microbiol Methods 21:293–303
affiliated with a clone cluster, the green non-sulfur bacteria, subdivision I. Stewart CS, Bryant MP (1988) The rumen bacteria. In: Hobson PN (ed) The
Appl Environ Microbiol 67:5740–5749 rumen microbial ecosystem. Elsevier Applied Science, London, pp 21–75
Shilo M (1984) Bdellovibrio as a predator. In: Klug MJ, Reddy CA (eds) Current Stolp H, Starr MP (1963) Bdellovibrio bacteriovorus gen. et sp. n., a predatory
perspectives in microbial ecology. ASM Press, Washington, DC, pp 334–339 ectoparasitic and bacteriolytic microorganism. Antonie Van Leeuwenhoek
Shindler DB, Wydro RM, Kushner DJ (1977) Cell-bound cations in the moder- 29:217–248
ately halophilic bacterium Vibrio costicola. J Bacteriol 130:698–703 Stolp H, Starr MP (1965) Bacteriolysis. Annu Rev Microbiol 19:79–104
Shockey WL, Dehority BA (1989) Comparison of two methods for enumeration Stringfellow WT, Aitken MD (1994) Comparative physiology of phenanthrene
of anaerobe numbers on forages and evaluation of ethylene oxide treatment degradation by two dissimilar pseudomonads isolated from a creosote-
for forage sterilization. Appl Environ Microbiol 55:1766–1768 contaminated soil. Can J Microbiol 40:432–438
Siebert J, Hirsch P (1988) Characterization of 15 selected coccal bacteria isolated Sugio T, Kuwano H, Negishi A, Maeda T, Takeuchi F, Kamimura K (2001)
from Antarctic rock and soil samples from the McMurdo-Dry Valleys Mechanism of growth inhibition by tungsten in Acidithiobacillus
(South-Victoria Land). Polar Biol 9:37–44 ferrooxidans. Biosci Biotechnol Biochem 65:555–562
SILVA (2012) http://www.arb-silva.de/documentation/background/release-108/ Suh DH, Becker TC, Sands JA, Montenecourt BS (1988) Effects of temperature on
Simankova MV, Kotsyurbenko OR, Stackebrandt E, Kostrikina NA, Lysenko AM, xylanase secretion by Trichoderma reesei. Biotechnol Bioeng 32:821–825
Osipov GA, Nozhevnikova AN (2000) Acetobacterium tundrae sp. nov., a new Suzuki K, Sasaki J, Uramoto M, Nakase T, Komagata K (1997a) Cryobacterium
psychrophilic acetogenic bacterium from tundra soil. Arch Microbiol psychrophilum gen. nov., sp. nov., nom. rev., comb. nov., an obligately
174:440–447 psychophilic actinomycete to accommodate ‘‘Curtobacterium
Simek K, Vrba J, Pernthaler J, Posch T, Hartman P, Nedoma J, Psenner R (1997) psychrophilum’’, Inoue and Komagata 1976. Int J Syst Evol Microbiol
Morphological and compositional shifts in an experimental bacterial com- 47:474–478
munity influenced by protists with contrasting feeding modes. Appl Environ Suzuki MT, Rappé MS, Haimberger ZW, Winfield H, Adair N, Ströbel J,
Microbiol 63:587–595 Giovannoni SJ (1997b) Bacterial diversity among small-subunit rRNA gene
Sitnikov DM, Schineller JB, Baldwin TO (1996) Control of cell division in clones and cellular isolates from the same seawater sample. Appl Environ
Escherichia coli: regulation of transcription of ftsQA involves both rpoS Microbiol 63:983–989
and SdiA-mediated autoinduction. Proc Natl Acad Sci USA 93:336–341 Sweerts JPRA, de Beer D (1989) Microelectrode measurements of nitrate gradi-
Skerman VDB (1968) A new type of micromanipulator and microforge. J Gen ents in the littoral and profundal sediments of a meso-eutrophic lake (Lake
Microbiol 54:287–297 Vechten, The Netherlands). Appl Environ Microbiol 55:754–757
Skulachev VP (1987) Bacterial sodium transport: bioenergetic functions of Swift ST, Najita IY, Ohtaguchi K, Fredrickson AG (1982) Continuous culture of
sodium ions. In: Rosen BP, Silver S (eds) Ion transport in prokaryotes. the ciliate Tetrahymena pyriformis on Escherichia coli. Biotechnol Bioeng
Academic, New York, pp 131–164 24:1953–1964
Slater JH, Bull AT (1978) Interactions between microbial populations. In: Bull AT, Szewzyk U, Schink B (1989) Degradation of hydroquinone, gentisate, and ben-
Meadow PM (eds) Companion to microbiology. Longman, London, zoate by a fermenting bacterium in pure or defined mixed culture. Arch
pp 181–206 Microbiol 15:541–545
Talamoto S, Yamada K, Ezura Y (1994) Producing of bacteriolytic enzymes during Tuschak C, Glaeser J, Overmann J (1999) Specific detection of green sulfur
the growth of a marine bacterium Alteromonas sp. no. 8-R. J Gen Appl bacteria by in situ hybridization with a fluorescent labeled oligonucleotide
Microbiol 40:499–508 probe. Arch Microbiol 171:265–272
Tamaki H, Sekiguchi Y, Hanada S, Nakamura K, Nomura N, Matsumara M, Uphoff HU, Felske A, Fehr W, Wagner-Döbler I (2001) The microbial diversity in
Kamagata Y (2005) Comparative analysis of bacterial diversity in freshwater picoplankton enrichment cultures: a molecular screening of marine isolates.
sediment of a shallow eutrophic lake by molecular and improved cultivation- FEMS Microbiol Ecol 35:249–258
based techniques. Appl Environ Microbiol 71:2162–2169 Urakawa H, Kita-Tsukamoto K, Steven SE, Ohwada K, Colwell RR
Tamaki H, Tanaka Y, Matsuzawa H, Muramatsu M, Meng X, Hanada S, Mori K, (1998) A proposal to transfer Vibrio marinus (Russell 1891) to a new
Kamagata Y (2011) Armatimonas rosea gen. nov., sp. nov., of a novel bacterial genus Moritella gen. nov. as Moritella marina comb. nov. FEMS Microbiol
phylum, Armatimonadetes phyl. nov., formally called the candidate phylum Lett 165:373–378
OP10. Int J Syst Evol Microbiol 61:1442–1447 Urbach E, Vergin KL, Giovannoni SJ (1999) Immunochemical detection and
Tang W-C, White JC, Alexander M (1998) Utilization of sorbed compounds by isolation of DNA from metabolically active bacteria. Appl Environ Microbiol
microorganisms specifically isolated for that purpose. Appl Microbiol 65:1207–1213
Biotechnol 49:117–121 Usui K, Hiraishi T, Kawamoto J, Kurihara T, Nogi Y, Kato C, Abe F (2012)
Tappe W, Tomaschewski C, Rittershaus S, Groeneweg J (1996) Cultivation of Eicosapentaenoic acid plays a role in stabilizing dynamic membrane struc-
nitrifying bacteria in the retentostat, a simple fermenter with internal bio- ture in the deep-sea piezophile Shewanella violacea: a study employing high-
mass retention. FEMS Microbiol Ecol 19:47–52 pressure time-resolved fluorescence anisotropy measurement. Biochim
Tappe W, Laverman A, Bohland M, Braster M, Rittershaus S, Groeneweg J, van Biophys Acta 1818:574–583
Versefeld HW (1999) Maintenance energy demand and starvation recovery Van den Ende FPL, Laverman AM, van Gemerden H (1996) Coexistence of
dynamics of Nitrosomonas europaea and Nitrobacter winogradskyi cultivated aerobic chemotrophic and anaerobic phototrophic sulfur bacteria under
in a retentostat with complete biomass retention. Appl Environ Microbiol oxygen limitations. FEMS Microbiol Ecol 19:141–151
65:2471–2477 Van den Ende FP, Meier J, van Gemerden H (1997) Syntrophic growth of sulfate-
Tarin JJ, Trounson AO (1993) Effect of stimulation or inhibition of lipid perox- reducing bacteria and colorless sulfur bacteria during oxygen limitation.
idation on freezing-thawing of mouse embryos. Biol Reprod 49:1362–1368 FEMS Microbiol Ecol 23:65–80
Teather RM (1982) Maintenance of laboratory strains of obligately anaerobic van der Meer MTJ, Schouten S, van Dongen BE, Rijpstra WIC, Fuchs G,
rumen bacteria. Appl Environ Microbiol 44:499–501 Sinninghe Damsté JS, de Leeuw JW, Ward DM (2001) Biosynthetic controls
Tempest DW (1970) The continuous culture in microbial research. Adv Microb on the 13C contents of organic components in the photoautotrophic bacte-
Physiol 4:223–250 rium Chloroflexus aurantiacus. J Biol Chem 276:10971–10976
Tempest DW, Neyssel OM (1978) Eco-physiological aspects of microbial growth Van der Wielen PWJJ, Lipman LJA, van Knapen F, Biesterveld S (2002) Compet-
in aerobic nutrient limited environments. Adv Microbiol Ecol 2:105–153 itive exclusion of Salmonella enterica serovar enteritidis by Lactobacillus
Tempest DW, Herbert F, Phipps PJ (1967) Studies on the growth of Aerobacter crispatus and Clostridium lactatifermentans in a sequencing fed-batch cul-
aerogenes at low dilution rates in a chemostat. In: Powell EO, Evans CGT, ture. Appl Environ Microbiol 68:555–559
Strange RE, Tempest DW (eds) Microbial physiology and continuous cul- van Gemerden H (1974) Coexistence of organisms competing for the same
ture. HMSO, London, pp 240–254 substrate: an example among purple sulfur bacteria. Microb Ecol 1:104–110
Thauer RK, Käufer B, Fuchs G (1975) The active species of ‘‘CO2’’ utilized by van Gemerden H, Mas J (1995) Ecology of phototrophic sulfur bacteria. In:
reduced ferredoxin: CO2 oxidoreductase from Clostridium pasteurianum. Blankenship RE, Madigan MT, Bauer CE (eds) Anoxygenic phototrophic
Eur J Biochem 27:282–290 bacteria. Kluwer, Dordrecht, pp 49–85
Thauer RK, Jungermann K, Decker K (1977) Energy conservation in van Gemerden H, Tughan CS, De Witz R, Herbert RA (1989) Laminated micro-
chemotrophic anaerobic bacteria. Bacteriol Rev 41:100–180 bial ecosystems on sheltered beaches in Scapa Flow, Orkney Islands. FEMS
Thiele JH, Zeikus JG (1988) Control of interspecies electron flow during anaerobic Microbiol Ecol 62:87–102
digestion: significance of formate transfer versus hydrogen transfer during Van Iterson G Jr, Den Dooren de Jong LE, Kluyver AJ (1940) Martinus Willem
syntrophic methanogenesis in flocs. Appl Environ Microbiol 54:20–29 Beijerinck, his life and work. Martinus Nijhoff, The Hague
Thiele JH, Chartrain M, Zeikus JG (1988) Control of interspecies electron flow Van Niel CB (1930) Contributions to marine biology. Stanford University Press,
during anaerobic digestion: role of floc formation in syntrophic Stanford, pp 161–169
methanogenesis. Appl Environ Microbiol 54:10–19 Van Niel CB (1967) Prefatory chapter. The education of a microbiologist: some
Thompson LA, Nedwell DB, Balba MT, Banat IM, Senior E (1983) The use of reflections. Annu Rev Microbiol 21:1–30
multiple vessel, open flow systems to investigate carbon flow in anaerobic Van Verseveld HW, Chesbro WR, Braster M, Stouthamer AH (1984) Eubacteria
microbial communities. Microb Ecol 9:189–199 have 3 growth modes keyed to nutrient-flow—consequences for the concept
Tindall BJ (2007) Vacuum-drying and cryopreservation of prokaryotes. In: Day JG, of maintenance and maximal growth yield. Arch Microbiol 137:176–184
Stacey GN (eds) Cryopreservation and freeze-drying protocols, vol 368, 2nd Vancanneyt M, Schut F, Snauwaert C, Goris J, Swings J, Gottschal JC
edn, Methods in molecular biology. Humana Press, Totowa, pp 73–97 (2001) Sphingomonas alaskensis sp. nov., a dominant bacterium from
Tomczak MM, Hincha DK, Crowe JH, Harding MM, Haymet ADJ (2003) The a marine oligotrophic environment. Int J Syst Evol Microbiol 51:73–79
effect of hydrophobic analogues of the type I winter flounder antifreeze Varon M, Shilo M (1980) Ecology of aquatic Bdellovibrios. Adv Aquat Microbiol
protein on lipid bilayers. FEBS Lett 551:13–19 2:1–48
Torella F, Morita RY (1981) Microcultural study of bacterial size changes and Veldkamp H (1965) Enrichment cultures—history and prospects. In: Schlegel
microcolony and ultramicrocolony formation by heterotrophic bacteria in HG, Kröger E (eds) Anreicherungskultur und Mutantenauslese: zentralbl.
seawater. Appl Environ Microbiol 41:518–527 Bakteriol. Parasitenkd. Infektionskrankh. Hygiene I. Abteilung Gustav.
Trimbur DE, Gutshall KR, Prema P, Brenchley JE (1994) Characterization of Fischer, Stuttgart, Supplement 1, pp 1–13
a psychrotrophic Arthrobacter gene and its cold-active b-galactosidase. Veldkamp H (1977) Ecological studies with the chemostat. Adv Microbiol Ecol
Appl Environ Microbiol 60:4544–4552 1:59–94
Tripp HJ, Kitner JB, Schwalbach MS, Dacey JW, Wilhelm LJ, Giovannoni SJ Veldkamp H, Jannasch HW (1972) Mixed culture studies with the chemostat.
(2008) SAR11 marine bacteria require exogenous reduced sulphur for J Appl Chem Biotechnol 22:105–123
growth. Nature 452:741–744 Veldkamp H, Kuenen JG (1973) The chemostat as a model system for ecological
Trüper HG, Pfennig N (1971) Family of phototrophic green sulfur bacteria: studies. Bull Ecol Res Commun 17:347–355
Chlorobiaceae Copeland, the correct family name; rejection of Veldkamp H, van Gemerden H, Harder W, Laanbroek HJ (1984) Microbial
Chlorobacterium Lauterborn; and the taxonomic situation of the consor- competition. In: Klug MJ, Reddy CA (eds) Current perspectives in microbial
tium-forming species. Int J Syst Bacteriol 21:8–10 ecology. ASM Press, Washington, DC, pp 279–290
Vester F, Ingvorsen K (1998) Improved most-probable-number method to detect Widdel F, Bak F (1992) Gram-negative mesophilic sulfate-reducing bacteria. In:
sulfate-reducing bacteria with natural media and a radiotracer. Appl Environ Balows A, Trüper HG, Dworkin M, Schleifer K-H (eds) The prokaryotes,
Microbiol 64:1700–1707 2nd edn. Springer, New York, pp 3352–3378
Visscher PT, Prins RA, van Gemerden H (1992a) Rates of sulfate reduction and Widdel F, Kohring G-W, Mayer F (1983) Studies on dissimilatory sulfate-
thiosulfate consumption in a marine microbial mat. FEMS Microbiol Ecol reducing bacteria that decompose fatty acids. III: characterization of the
86:283–294 filamentous gliding Desulfonema limicola gen. nov. sp. nov., and
Visscher PT, van den Ende FP, Schaub BEM, van Gemerden H (1992b) Compe- Desulfonema magnum sp. nov. Arch Microbiol 134:286–294
tition between anoxygenic phototrophic bacteria and colorless sulfur bacte- Wiegel J (1990) Temperature spans for growth: hypothesis and discussion. FEMS
ria in a microbial mat. FEMS Microbiol Ecol 101:51–58 Microbiol Rev 75:155–170
Vobis G (1992) The genus, Actinoplanes and related genera. In: Balows A, Wiegel J, Ljungdahl LG (1986) The importance of thermophilic bacteria in
Trüper HG, Dworkin M, Schleifer K-H (eds) The prokaryotes, 2nd edn. biotechnology CRC. Crit Rev Biotechnol 3:39–107
Springer, New York, pp 1029–1060 Wilkinson TG, Topiwala HH, Hamer G (1974) Interactions in a mixed bacterial
Vogl K, Glaeser J, Pfannes KR, Wanner G, Overmann J (2006) Chlorobium population growing on methane in continuous culture. Biotechnol Bioeng
chlorochromatii sp. nov., a symbiotic green sulfur bacterium isolated from 16:41–59
the phototrophic consortium ‘‘Chlorochromatium aggregatum’’. Arch Winogradsky S (1949) Microbiologie du sol: oeuvres complèetes. Marson, Paris
Microbiol 185:363–372 Wirsen CO, Jannasch HW (1978) Physiological and morphological observations
Wachenheim DE, Hespell RB (1984) Inhibitory effects of titanium (III) citrate on on Thiovulum sp. J Bacteriol 136:765–774
enumeration of bacteria from rumen contents. Appl Environ Microbiol Wirsen CO, Molyneaux SJ (1999) A study of deep-sea natural microbial
48:444–445 populations and barophilic pure cultures using a high-pressure chemostat.
Wagner M, Nielsen PH, Loy A, Nielsen JL, Daims H (2006) Linking microbial Appl Environ Microbiol 65:5314–5321
community structure with function: fluorescence in situ hybridization- Wirsen CO, Jannasch HW, Wakeham SG, Cannel EA (1987) Membrane lipids of
microautoradiography and isotope arrays. Curr Opin Biotechnol 17:1–9 a psychrophilic and barophilic deep-sea bacterium. Curr Microbiol
Walderhaug MO, Dosch DC, Epstein W (1987) Potassium transport in bacteria. 14:319–322
In: Rosen BP, Silver S (eds) Ion transport in prokaryotes. Academic, New Wolfe RS, Pfennig N (1977) Reduction of sulfur by Spirillum 5175 and
York, pp 85–130 syntrophism with Chlorobium. Appl Environ Microbiol 33:427–433
Walker GC (1996) The SOS response of Escherichia coli. In: Neidhardt FC (ed) Wolin MJ (1982) Hydrogen transfer in microbial communities. In: Bull AT, Slater
Escherichia coli and Salmonella, vol 1. ASM Press, Washington, DC, JH (eds) Microbial interactions and communities, vol 1. Academic, London,
pp 1400–1416 pp 323–356
Wang X, de Boer PAJ, Rothfield LI (1991) A factor that positively regulates cell Wolter K (1982) Bacterial incorporation of organic substances released by natural
division by activating transcription of the major cluster of essential cell phytoplankton populations. Mar Ecol Prog Ser 7:287–295
division genes of Escherichia coli. EMBO J 10:3363–3372 Xiao J, Luo Y, Xu J, Xie S, Xu J (2011) Modestobacter marinus sp. nov.,
Ward DM, Bateson MM, Weller R, Ruff-Roberts AL (1992) Ribosomal RNA analysis a psychrotolerant actinobacterium from deep-sea sediment, and emended
of microorganisms as they occur in nature. Adv Microbiol Ecol 12:219–286 description of the genus Modestobacter. Int J Syst Evol Microbiol 61:
Warthmann R, Cypionka H, Pfennig N (1992) Photoproduction of H2 from 1710–1714
acetate by syntrophic cocultures of green sulfur bacteria and sulfur-reducing Xu HS, Roberts N, Singleton FL, Attwell RW, Grimes DJ, Colwell RR (1982)
bacteria. Arch Microbiol 157:343–348 Survival and viability of nonculturable Escherichia coli and Vibrio cholerae in
Waterbury JB (1991) The cyanobacteria—isolation, purification, and identifica- the estuarine and marine environment. Microbiol Ecol 8:313–323
tion. In: Balows A, Trüper HG, Dworkin M, Schleifer K-H (eds) The pro- Xu M, Xin Y, Tian J, Dong K, Yu Y, Zhang J, Liu H, Zhou Y (2011) Flavobacterium
karyotes, 2nd edn. Springer, New York, pp 149–196 sinopsychrotolerans sp. nov., isolated from a glacier. Int J Syst Evol Microbiol
Waterbury JB, Calloway CB, Turner RD (1983) A cellulolytic nitrogen-fixing 61:20–24
bacterium cultured from the gland of Deshayes in shipworms (Bivalvia: Yabe S, Aiba Y, Sakai Y, Hazaka M, Yokota A (2011) Thermogemmatispora
Terebrenidae). Science 221:1401–1403 onikobensis gen. nov., sp. nov. and Thermogemmatispora foliorum sp. nov.,
Watson TG (1969) Steady state operation of a continuous culture at maximum isolated from fallen leaves on geothermal soils, and description of
growth rate by control of carbon-dioxide production. J Gen Microbiol 59:83–89 Thermogemmatisporaceae fam. nov. and Thermogemmatisporales ord.
Wayne LG, Brenner DJ, Colwell RR, Grimont PAD, Kandler O, Krichevsky MI, nov. within the class Ktedonobacteria. Int J Syst Evol Microbiol 61:903–910
Moore LH, Moore WEC, Murray RGE, Stackebrandt E, Starr MP, Trüper HG Yamamoto Y, Fukui K, Koujin N, Ohya H, Kimura K, Kamio Y (2004) Regulation
(1987) Report of the ad hoc committee on reconciliation of approaches to of the intracellular free iron pool by Dpr provides oxygen tolerance to
bacterial systematics. Int J Syst Bacteriol 37:463–464 Streptococcus mutans. J Bacteriol 186:5997–6002
Weimer PJ, Zeikus JG (1977) Fermentation of cellulose and cellobiose by Clos- Yancey PH, Clark ME, Hand SC, Bowlus RD, Somero GN (1982) Living with
tridium thermocellum in the absence and presence of Methanobacterium water stress: evolution of osmolyte systems. Science 217:1214–1227
thermoautotrophicum. Appl Environ Microbiol 33:289–297 Yarza P, Richter M, Peplies J, Euzeby J, Amann R, Schleifer K-H, Ludwig W,
Weise W, Rheinheimer G (1978) Scanning electron microscopy and Glöckner FO, Rosselo-Mora R (2008) The all-species living tree project: a 16S
epifluorescence investigation of bacterial colonization of marine sand sedi- rRNA-based phylogenetic tree of all sequenced type strains. Syst Appl
ments. Microb Ecol 4:175–188 Microbiol 31:241–250
Welch TJ, Farewell A, Neidhardt FC, Bartlett D (1993) Stress response of Yasumoto-Hirose M, Nishijima M, Ngirchechol MK, Kanoh K, Shizuri Y, Miki W
Escherichia coli to elevated hydrostatic pressure. J Bacteriol 175:7170–7177 (2006) Isolation of marine bacteria by in situ culture on media-
Westermann P, Ahring BK, Mah RA (1989) Temperature compensation in supplemented polyurethane foam. Mar Biotechnol 8:227–237
Methanosarcina barkeri by modulation of hydrogen and acetate affinity. Yayanos AA (1978) Recovery and maintenance of live amphipods at a pressure of
Appl Environ Microbiol 55:1262–1266 580 bars from an ocean depth of 5700 meters. Science 200:1056–1059
White RH (1984) Hydrolytic stability of biomolecules at high temperatures and Yayanos AA (1986) Evolutional and ecological implications of the properties of
its implications for life at 250 C. Nature 310:430–432 deep-sea barophilic bacteria. Proc Natl Acad Sci USA 83:9542–9546
Whitesides MD, Oliver JD (1997) Resuscitation of Vibrio vulnificus from the Yoshida F, Yamane T, Nakamoto K (1973) Fed-batch hydrocarbon fermentation
viable but non-culturable state. Appl Environ Microbiol 63:1002–1005 with colloidal emulsion feed. Biotechnol Bioeng 15:257–270
Widdel F (1988) Microbiology and ecology of sulfate-and sulfur-reducing bacte- Yu Y, Li H-R, Zeng Y-Y (2011) Colwellia chukchiensis sp. nov., a psychrotolerant
ria. In: Zehnder AJB (ed) Biology of anaerobic microorganisms. Wiley, New bacterium isolated from the Arctic Ocean. Int J Syst Evol Microbiol
York, pp 469–585 61:850–853
Yumoto I, Yamazaki K, Hishinuma M, Nodasaka Y, Nakajima K, Inoue N, Kawa- Zhang DC, Rezic M, Schinner F, Margesin R (2011a) Glaciimonas immobilis gen.
saki K (2001) Pseudomonas alcaliphila sp. nov., a novel facultatively psychro- nov., sp. nov., a member of the family Oxalobacteraceae isolated from alpine
philic alkaliphile isolated from seawater. Int J Syst Evol Microbiol 51:349–355 glacier cryoconite. Int J Syst Evol Microbiol 61:2186–2190
Yumoto I, Nakamura A, Iwata H, Kusumoto K, Nodasaka Y, Matsuyama H (2002) Zhang DC, Busse H-J, Liu H-C, Zhou Y-G, Schinner F, Margesin R (2011b)
Dietzia psychralcaliphila sp. nov., a novel, facultatively psychrophilic Hymenobacter psychrophilus sp. nov., a psychrophilic bacterium isolated
alkaliphile that grows on hydrocarbons. Int J Syst Evol Microbiol 52:85–90 from soil. Int J Syst Evol Microbiol 61:859–863
Zehnder AJB, Stumm W (1988) Geochemistry and biogeochemistry of anaerobic Zhang DC, Redzic M, Liu HC, Zhou YG, Schinner F, Margesin R (2012) Devosia
habitats. In: Zehnder AJB (ed) Biology of anaerobic microorganisms. Wiley, psychrophila sp. nov. and Devosia glacialis sp. nov., two novel bacteria from
New York, pp 1–38 alpine glacier cryoconite. Int J Syst Evol Microbiol 62:710–715
Zehnder AJB, Wuhrman K (1976) Titanium (III) citrate as a non-toxic oxidation- Zhou Z, Jiang F, Wang S, Peng F, Dai J, Li W, Fang C (2012) Pedobacter arcticus sp.
reduction buffering system for the culture of obligate anaerobes. Science nov., a facultative psychrophile isolated from Arctic soil and emended
194:1165–1166 descriptions of the genus Pedobacter, P. heparinus, P. daechungensis,
Deutsche Sammlung für Mikroorganismen und Zellkulturen (2012) Bacterial P. terricola, P. glucosidilyticus and P. lentus. Int J Syst Evol Microbiol
nomenclature up-to-date. http://www.dsmz.de/bacterial-diversity/bacte- 62:1963–1969
rial-nomenclature-up-to-date.html Zinder SH, Koch M (1984) Non-aceticlastic methanogenesis from acetate: acetate
Zengler K, Richnow HH, Rosselló-Mora R, Michaelis W, Widdel F (1999) Meth- oxidation by a thermophilic syntrophic coculture. Arch Microbiol 138:263–272
ane formation from long-chain alkanes by anaerobic microorganisms. ZoBell CE (1941) Studies on marine bacteria. J Mar Res 4:42–75
Nature 401:266–269 Zobell CE, Johnson FH (1949) The influence of hydrostatic pressure on the
Zengler K, Toledo G, Rappe M, Elkins J, Mathur EJ, Short JM, Keller M (2002) growth and viability of terrestrial and marine bacteria. J Bacteriol
Cultivating the uncultured. Proc Natl Acad Sci USA 99:15681–15686 57:179–189
8 Comparative Genomics
Asli Ismihan Ozen . Tammi Vesth . David W. Ussery
Department of Systems Biology, Center for Biological Sequence Analysis, Kemitorvet, The Technical
University of Denmark, Lyngby, Denmark
Prokaryotic Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 dependent on the development of microbial techniques such as

isolating and growing microorganisms in pure cultures, staining
Current Taxonomy of Prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . . 210 and microscope observations. In their early observations, lack of
guidelines on naming inevitably led to a vast number of invalid
The Explosion of Sequenced Bacterial Genomes . . . . . . . . . . 210 names and synonyms. Ferdinand Cohn made the first classifica-
tion system of bacteria in 1872; six genera of bacteria were
Statistics on Prokaryotic Genomes . . . . . . . . . . . . . . . . . . . . . . . . 211 classified based on their shape, cellular structures, pigmentation,
Data Growth over the Years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 and metabolic activities (Cohn 1872).
Taxonomy Analysis, Most Sequenced Phyla At the beginning of the twentieth century, besides morphol-
and Genera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212 ogy, the use of physiological and biochemical information could
Basic Genome Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 be incorporated. European scientists also proposed physiology,
Thousands of Genome Sequences . . . . . . . . . . . . . . . . . . . . . . . 216 metabolism, pigments, and pathogenicity as new systems for
Whole-Genome-Based Tools for Taxonomy . . . . . . . . . . . . 216 classification. However, some of these methods were then criti-
rRNA Phylogenetic Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 cized for being not important for assessing taxonomic ranks.
Average Nucleotide Identities (ANI) and Tetra Later, advances in biochemistry and molecular biology from the
Nucleotide Frequency Calculations . . . . . . . . . . . . . . . . . . . . . 218 isolation of nucleic acids to elucidation of macromolecular
BLASTMatrix Using Reciprocal Best Hits . . . . . . . . . . . . . . . 220 structure of proteins and nucleic acids led to the foundation of
Composition Vector Trees (CVTree) . . . . . . . . . . . . . . . . . . . . 222 genomic sciences. Development of computers in the 1950s was
Pan-genome Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 another important step in bacterial taxonomy, where they were
first used for analysis of phenotypic and molecular data.
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 Between the years 1960–1980, numerical taxonomy and chemo-
taxonomy were on the rise (Stackebrandt 2006; Schleifer 2009).
In late 1950s, scientists were able to identify the molecules
Prokaryotic Classification conserved throughout history of life, such as proteins, DNA, or
RNA molecules. The idea of using these molecules as blueprints
Classification covers the theory and practice of how to order of the evolutionary history of organisms emerged in the 1960s
characterized organisms into different groups based on their (Zuckerkandl et al. 1962). Tertiary structure and sequence anal-
degree of relatedness. Together with identification and nomen- ysis of molecules, such as cytochrome C, ferrodoxins, and fibri-
clature, classification is a part of taxonomy, a science that deals nopeptides, and also immunological approaches were being used
with the relatedness of organisms. The goal of many taxonomists afterward. However, the interest in these methods decreased as
is to have a classification system that reflects the natural relation- rapid sequencing techniques for DNA became more significant.
ships among organisms. This natural system has been depicted The first genotypic approach that allowed bacteriologists to
mostly as phylogeny (Doolittle 1999)—or an evolutionary classify prokaryotes on the basis of their phylogenetic related-
tree—which is a diagram that shows ancestor-descendant rela- ness was DNA-DNA hybridization (DDH) (Wayne et al. 1987).
tionships of organisms based on their evolutionary history. In the following years, more genotypic studies, including com-
However, inferring a true phylogeny for prokaryotic organisms parative analysis of Ribosomal RNA (ribonucleic acid) genes
is very challenging due to the diversity of these organisms, as well and protein-coding gene sequence, allowed more insight to the
as frequent horizontal transfer of genes. relationships of prokaryotes (Schleifer 2009). The small subunit
Prokaryotes, known as unicellular organisms with no rRNA (16S rRNA in prokaryotes) was shown to be one of the
nuclear membrane structure, have a history of more than 3.5 universally conserved molecules became the primary molecule
billion years on earth, yet humans have been aware of them for of interest. Being ubiquitous, having functional consistency,
only the past few centuries, after first being described by Robert genetic stability, appropriate size, and independently evolving
Hooke in the seventeenth century. Louis Pasteur and other domains caused this molecule to be chosen for phylogenetic
scientists of the nineteenth century described microorganisms analysis and this approach became a classical tool for taxonomy
in detail, and began to categorize them. Their classification was (Harayama and Kasai 2006). An important study by Carl Woese
210 8 Comparative Genomics
revolutionized bacterial taxonomy, proposing the new kingdom classification system based on evolutionary history, clear clus-
of Archaebacteria (Woese and Fox 1977). His later studies con- ters of taxonomic units in a phylogenetic tree are seen such that
cluded in a phylogenetic scheme of three main branches of life species that share a common ancestor would form the genus
(Bacteria, Archaea, and Eukarya) that he called Domains (Woese and genera that share a common ancestor form a family and so
et al. 1990). forth. Major sources for bacterial names and taxonomical order
In other genotypic classifications, many protein-coding are Bergey’s Manual of Systematic Bacteriology (Brenner et al.
genes were used for phylogenetic relatedness, some of which 2005a), Bergey’s Taxonomic Outlines (http://www.bergeys.org/
are recA, gyrB, genes of some chaperonins, RNA polymerase outlines.html), and the comprehensive list available at The Tax-
subunits (i.e., rpoB) or sigma factors (rpoD), elongation factor onomic Outline of Bacteria and Archaea (TOBA) journal
G (fus). The most accepted criteria for selection of these proteins (Garrity et al. 2007).
is such that, they should not be subjected to horizontal gene Taxonomy tools historically have been mainly based on
transfer (HGT), should be present in all bacteria, preferably in laborious laboratory experiments trying to characterize bacteria
single copies and at least two highly conserved regions for the based on their phenotypic and biochemical properties until
design of PCR primers (Yamamoto and Harayama 1996). These molecular approaches and sequencing technologies were devel-
properties give them an advantage of being more appropriate for oped. Today, such research can be handled using robotic and
phylogenetic analysis of closely related bacteria than 16S rRNA computational techniques, where most of the knowledge gained
analysis. from results rely on the data that is being handled.
In addition to the single gene based methods, Multi Locus
Sequence Typing (MLST) has been widely used for genotypic
characterization and classification of prokaryotes by comparing The Explosion of Sequenced Bacterial
multiple housekeeping gene sequences (Maiden et al. 1998). Genomes
However, usually a different set of genes is useful for different
set of organisms, and some difficulties occur in primer design for Biological data generated by researchers worldwide has been
amplification of genes in all strains if the analysis is not growing with a tremendous rate, especially with the advances
conducted all in silico. A widely used website and database in molecular biology techniques in the past 50 years. Much of
currently is mlst.net (Aanensen and Spratt 2005). this vast information can now be accessed through biological
databases that hold records for experimental data, sequence
data, classification schemes, literature, and some also provide
Current Taxonomy of Prokaryotes computational analysis tools.
A part of this huge biological information is the genomic
Classification is done by comparing a newly identified organism sequences. In modern molecular biology and genetics,
with the collection of previously classified organisms and then a ‘‘genome’’ is the entirety of an organism’s hereditary informa-
assigning it with a previously described or new species. If tion. Therefore, genomics can be referred to as the science of
a bacterial species is considered novel, the proper naming for genome analysis. As such the field of comparative microbial
the new or existing taxa are made by nomenclature that is based genomics (CMG) work with comparing the entire DNA material
on the International Code of Nomenclature of Bacteria (Lapage of a microbial organism to other organisms. The first two com-
et al. 1992), also named as the Bacteriological Code. Nomencla- plete bacterial genome sequences were published in 1995. As the
ture is, however, subject to changes because classification is technologies advanced and the sequencing cost went down,
a dynamic process. The publication of names for novel prokary- many more sequences were being published and more databases
otic taxa is made in the International Journal of Systematic and were established to handle this information. One of the most
Evolutionary Microbiology (IJSEM), which is the official journal used databases is based on GenBank, now located as part of the
for this purpose. IJSEM also publishes ‘‘Validation Lists’’ which National Center for Biotechnology Information, NCBI (http://
contain new names published in other journals (Tindall et al. www.ncbi.nlm.nih.gov/genome/browse/). The NCBI GenBank
2006). An updated list of approved names for microorganisms holds the nucleotide sequence data from expression sequence
based on the international rules can also be found at The tag (EST), genome survey sequences, other high-throughput
DSMZ—Deutsche Sammlung von Mikroorganismen und sequences such as whole-genome sequences and genome anno-
Zellkulturen GmbH (German Collection of Microorganisms tations of thousands of organisms. Both prokaryotic and
and Cell Cultures) depository (http://www.dsmz.de). eukaryotic data is available (Benson et al. 2008). GenBank is
In taxonomy, groups of organisms that are brought a part of an international collaboration called International
together based on shared properties are called ‘‘taxa’’ or Sequence Database Collaboration, which also consists of DNA
‘‘ranks,’’ and prokaryotic taxonomy makes use of several Data Base of Japan (DDBJ) and the European Molecular Biology
ranks or levels. The current classification scheme has laboratory (EMBL). Another part of NCBI that is highly related
a hierarchical structure, where the higher taxonomic ranks to this chapter is NCBI Taxonomy. Although claiming not to be
consist of the lower ranked groups. In other words, higher a primary source, NCBI provides taxonomical information that
taxa (e.g., genus) contain lower taxa (e.g., species). In an ideal is gathered from various sources.
Comparative Genomics 8 211
400
300
Number of genomes
200
100
10
1
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Year
. Fig. 8.1
Genomes published and deposited to public NCBI GenBank since 1995 (Data gathered from NCBI (http://www.ncbi.nlm.nih.gov/
genomes/lproks.cgi, Jan. 2012))
In May 2011, NCBI GenBank contained around 1,500 than others, a clear standard for genome publication has yet to
genome sequences labeled as ‘‘finished.’’ Six months later, at be established (Médigue and Moszer 2007).
the time of writing (November 2011), this number has gone
up to 1,790. Currently (Feb. 2012), the NCBI ‘‘Genome Projects’’
is changing to ‘‘BioProjects,’’ in order to relate genomic infor- Statistics on Prokaryotic Genomes
mation to other data types, such as the transcriptome, prote-
ome, and metagenome. With such a large amount of data, it is interesting to
In addition to NCBI (GenBank) and EMBL (Nucleotide see the trends in the basic statistics of the genomic data and
Sequence Database), a source for genomic information is the comparisons on different taxonomic levels and years of
Genomes Online database (GOLD). GOLD aims to provide an sequence publications. The data presented in this section is
accurate and complete set of finished and ongoing genome taken from the NCBI complete genomes list in Jan. 2012 and
projects with a broad range of information on each project. GenBank files for 1,500 sequenced genomes were downloaded
The sequence data itself is not stored in the database, however, (November 2011).
external links to where the data can be found is given, most of
which are to the NCBI Genome Project pages. GOLD also pro-
vides taxonomical information, though not the primary source Data Growth over the Years
(Bernal et al. 2001; Liolios et al. 2010).
One part of comparative microbial genomics is to monitor > Figure 8.1 illustrates how many genomes were published each
the available microbial genomic data. Even though the year, since 1995. The two first complete genomes to be
sequences available may only be a small fraction of the real sequenced and deposited was Mycoplasma genitalium G37
world, the information gathered is growing every day. It took (Fraser et al. 1995) and Haemophilus influenzae Rd KW20
14 years to sequence the first thousand bacterial genomes (Fleischmann et al. 1995). From 1995 until 1999, only 25
(1995–2009), and already in 2012, less than 3 years later, the genomes were published as complete and they covered 14
two thousandth genome sequence has been deposited to phyla. Of these the Archaeal genomes constitute a large portion
GenBank. Not only has the cost of genome sequencing decrease compared to the fraction today (around 31% of the 25 compared
dramatically but also the time and effort put into the task has to 99 out of 1,500 (6.6%)). The genomes from this first period of
also gone down. Also the computational power and software to genome sequencing cover a large span of the microbial land-
handle sequencing data is being revolutionized and fast assem- scape with no obvious medical bias. From 2000 to 2010 the
bly and interpretation is increasing the number of published number steadily increased from 26 to 1,423 with the major
genomes (Ansorge 2009). The increase in genome data has given phyla covered being Firmicutes, Gamma, and Alpha subdivisions
rise to a whole new area of problems when it comes to publica- of Proteobacteria. It is possible that producing a complete genome
tion and sharing of data. Databases usually have their own sequence is becoming less popular due to the improvement
formatting of the raw data and though some are more used in software that can work on draft genomes (Chain et al. 2009).
a
400
Count of occurance, genomes in progress
300
200
100
Chlamydiae/Verrucomicrobia (CHL−VER)
Bacteroidetes/Chlorobi (BAC−CHLO)
Gammaproteobacteria (GAM−PRO)
Epsilonproteobacteria (EP−PRO)
Alphaproteobacteria (ALP−PRO)
Deltaproteobacteria (DEL−PRO)
Deinococcus−Thermus (DEITH)
Betaproteobacteria (BET−PRO)
Nanoarchaeota (NAN)
Planctomycetes (PLA)
Crenarchaeota (CRE)
Euryarchaeota (EUR)
Cyanobacteria (CYA)
Actinobacteria (ACT)
Chloroflexi (CHLFX)
Thermotogae (THE)
Fusobacteria (FUS)
Acidobacteria (ACI)
Spirochaetes (SPI)
Aquificae (AQU)
Firmicutes (FIR)
Other
b
80
Count of occurrence, genomes in progress
60
40
20
0
(GAM−PRO) Pseudomonas
(CHL−VER) Chlamydophila
(GAM−PRO) Xanthomonas
(GAM−PRO) Acinetobacter
(GAM−PRO) Haemophilus
(EP−PRO) Campylobacter
(DEL−PRO) Desulfovibrio
(BET−PRO) Burkholderia
(GAM−PRO) Shewanella
(GAM−PRO) Escherichia
(GAM−PRO) Salmonella
(GAM−PRO) Francisella
(EP−PRO) Helicobacter
(ACT) Corynebacterium
(CYA) Prochlorococcus
(CHL−VER) Chlamydia
(GAM−PRO) Buchnera
(CYA) Synechococcus
(ALP−PRO) Rickettsia
(ACT) Bifidobacterium
(BET−PRO) Neisseria
(ACT) Mycobacterium
(GAM−PRO) Yersinia
(FIR) Staphylococcus
(ALP−PRO) Brucella
(ACT) Streptomyces
(FIR) Streptococcus
(GAM−PRO) Vibrio
(FIR) Lactobacillus
(FIR) Mycoplasma
(CRE) Sulfolobus
(FIR) Clostridium
(SPI) Borrelia
(FIR) Bacillus
(FIR) Listeria
. Fig. 8.2
Number of genomes sequences from each phyla and genera. Only genera with more than 10 representative genomes are shown
The cost of sequencing and the development of cheaper the GOLD database (http://genomesonline.org, March 2011)
sequencing methods have most definitely had an impact on the over 73% of the listed Streptococcus and more than 64% of
rate of sequencing (Sboner et al. 2011). Escherichia are labeled as pathogens.
It is likely that some organisms are sequenced because of
their medical relevance. Organisms belonging to the Escherichia
Taxonomy Analysis, Most Sequenced Phyla and genus have a considerable role within the medical world, with
Genera Escherichia coli being the cause of serious food poisoning. The
tendency of pathogens to be more often sequenced is, however,
> Figure 8.2a shows the number of genomes within each phyla; not as strong overall. The fraction of pathogens within each
Firmicutes and Gamma Proteobacteria are the most highly genus varies from 7% (Lactobacillus) to 98% (Listeria), for the
represented. The plot in > Fig. 8.2b shows genera with more six different genera belonging to the Firmicutes. For all the
than 10 sequenced genomes. The genus of Streptococcus is highly genera listed in > Fig. 8.2b the range covers everything from
overrepresented (63 genomes) while the closest other group is 3% (Synechococcus) and 100% (Borrelia and Rickettsia,
Escherichia (45 genomes). According to supporting data from supporting data from GOLD). It should be noted that the
annotation of ‘‘pathogen’’ is not always accurate; for example, Chlamydiae/Verrucomicrobia, and Nanoarchaeota. Of the largest
the first sequenced genome, Haemophilus influenzae is listed as genomes (more than 8,000 kb, 32 genomes), members of
a ‘‘pathogen,’’ although the strain sequenced (Rd KW20) is Actinobacteria are prevalent (14 genomes) ranging from 925 kb
a nonpathogenic strain. to 11,937 kb. The largest genome, as of May 2011, at
In any event, it is clear that there is a strong sequencing bias, 13,033.779 kb, was Sorangium cellulosum So ce 56 (soil-dwelling
making the available data for certain phyla and genera consid- bacteria) (Schneiker et al. 2007).
erably more than others. It could be expected that more patho- An interesting perspective on genome size is the focus on the
gens than nonpathogens would be sequenced, as the immediate minimal genome for a free-living organism. Defining the minimal
interest in these organisms is larger. However, the bias is not genome is a science in itself and has been heavily discussed in the
directly linked to pathogenicity, as some genera are sequenced scientific community (Galperin 2006). In 1995, the genome of
more often though not being serious pathogens. For example, Mycoplasma genitalium (a parasite) was published, and at that
species like Escherichia coli (urinary tract infections, simple time this was thought to be the smallest genome of any free-
diarrhea, dysentery-like conditions) include pathogens but are living organism (Fraser et al. 1995). Of the 1,500 genomes in this
not as severe as other species like Borrelia (Lyme disease) or study, M. genitalium is the 18th smallest genome. Upon closer
Listeria (Listeriosis in newborn infants, elderly patients, and inspection, the eight smallest ‘‘genomes’’ are described as phage,
immunocompromised). Escherichia coli is, however, Integrating and conjugative elements (ICEs), pathogenicity
a significant player in the financial aspect of medical relevance. island, or genomic island, so not free-living organisms. These
These organisms, though rarely lethal, can occur frequently in nongenomic sequences have been reported to GenBank, and
the population, and still require treatment; this is a burden on since have been deleted from the list of genomes during this
any healthcare system. Another factor could be the economic work. Other genomes smaller than M. genitalium consist of
cost, as some organisms grow less easily or replicate very slowly Buchnera (an endosymbiant (Pérez-Brocal et al. 2006) and
making experiments long and expensive. Some pathogens Nanoarchaeum) another symbiont (Waters et al. 2003).
require extreme safety procedures when cultured and this con- The remaining seven genomes are Candidatus species, from
sumes time, space, and money. The historic factor could also be proposed genera, and all of these are described as symbionts
partly responsible for sequencing bias. Some organisms became (McCutcheon et al. 2009). It is worth mentioning that the
model organisms from the early stage of microbiology and as smallest genome of a ‘‘true’’ free-living organism (as opposed
such, many procedures are optimized for these organisms. to parasites like M. genitalium) is considerably larger, containing
Unfortunately, due to the large variety within the microbial more than a thousand protein-encoding genes. Two proposed
world, many organisms will not respond well to procedures ‘‘minimal free-living organisms’’ are Pelagibacter ubique (het-
developed with Escherichia coli as the template. Taxonomical erotroph, 133th smallest in this study; DeWall and Cheng 2011)
bias in sequencing data is, as stated, a complex and multifaceted and Prochlorococcus marinus (autotroph, 209th smallest in this
discussion that will probably never end. However, these tenden- study; Moya et al. 2009). Note that these genomes are still
cies should be kept in mind when accessing the data available for smaller than the largest viral genomes (Arslan et al. 2011).
analysis. Another genome statistics commonly used is the percentage
of AT (> Fig. 8.3b), which is calculated as the average AT content
of all the DNA sequence. Genomes with high AT content include
Basic Genome Statistics Candidatus Zinderia insecticola CARI (86%, Beta
Proteobacteria), Candidatus Carsonella ruddii PV DNA (83%,
The sequences of 1,500 genomes have been obtained from NCBI Gamma Proteobacteria), and Buchnera aphidicola str. Cc (Cinara
GenBank and analyzed according to basic statistical parameters. cedri; 80%, Gamma Proteobacteria). These are all extremely
Here, basic genome statistics refers to certain DNA properties of small genomes. They are also all symbiotic organisms living
the genome, such as genome size, frequencies of A and T bases in inside insects, the spittlebug Clastoptera arizonana, jumping
the DNA, and bias on the third codon positions for the open plant lice and plant lice, respectively. Genomes with low AT
reading frames. (high GC) content include Anaeromyxobacter dehalogenans
> Figure 8.3a is a box and whisker plot showing the variation (Delta Proteobacteria; Sanford et al. 2002) and Cellulomonas
of genome sizes within each phylum. As seen, several phyla have flavigena (Actinobacteria; Abt et al. 2010), both with around
a wide distribution of sizes. Phyla containing only a few genomes 25% AT. These genomes consist of an anaerobic and aerobic
(less than five sequences) show very little size variation that soil-bacteria, respectively. The AT content within each phyla
could be the result of sequencing several closely related strains. shows some specificity with phyla like Acidobacteria,
For most phyla size is not a key feature, although Chlamydia and Actinobacteria, and Deinococcus/Thermus having a significant
Nanoarchaeota are expected to be small genomes. The genomes skew toward low AT and Fusobacteria, Epsilon Proteobacteria,
within the Firmicutes are distributed over a broad spectrum and Aquificae having a skew toward high AT (> Fig. 8.3b).
(580–8,300 kb), while Gamma Proteobacteria size varies between Can the AT content of an organism be related to its size? The
32 7,215 kb. Large genomes are often seen within answer can be both ‘‘yes’’ and ‘‘no.’’ The numbers from 1,500
Planctomycetes, Beta Proteobacteria, and Actinobacteria while genomes show that these two properties are not always propor-
small genomes are found within Epsilon Proteobacteria, tional to each other. However, for very large and small genomes,
a G
12000 G
G
10000 G
G
Genome size (kb)
G G
G
8000
G
G
6000 G
G
G
G
G
4000
G
G
G G
G G
G
2000
Chlamydiae/Verrucomicrobia
Archaea Nanoarchaeota
Archaea Crenarchaeota
Archaea Euryarchaeota
Proteobacteria Gamma
Bacteroidetes/Chlorobi
Proteobacteria Epsilon
Deinococcus-Thermus
Proteobacteria Alpha
Proteobacteria Delta
Proteobacteria Beta
Planctomycetes
Archaea Other
Other Bacteria
Cyanobacteria
Actinobacteria
Acidobacteria
Thermotogae
Spirochaetes
Fusobacteria
Chloroflexi
Firmicutes
Aquificae
b
G
G
80 G
G
G
G
G
70
G
G
G
Percent AT
G
60 G
G
G
50
G
G
G
G
G
G
40 G
G G
G
G G
30 G
G
Archaea Crenarchaeota
Proteobacteria Gamma
Bacteroidetes/Chlorobi
Proteobacteria Epsilon
Deinococcus-Thermus
Proteobacteria Alpha
Proteobacteria Delta
Proteobacteria Beta
Planctomycetes
Archaea Other
Other Bacteria
Cyanobacteria
Actinobacteria
Acidobacteria
Thermotogae
Spirochaetes
Fusobacteria
Chloroflexi
Firmicutes
Aquificae
. Fig. 8.3
Boxplots showing the distribution of genome size (in kilo base-pairs, panel a) and AT content (in percentage, panel b) for each phyla
(as described by NCBI Taxonomy). The middle bar is the 50 % percentile, the bottom and top of the box are the 25 % and 75 % percentiles
(Q1 and Q3, respectively). Whisker bars extend to the most extreme data point which is no more than +/ 1.58IQR/sqrt (n), where IQR is
the interquartile range (IQR = Q3 Q1). Any data point that exceeds this limit is plotted as an individual data point (outlier). The genome
size was calculated as the sum of lengths of all contigs
the answer can be ‘‘yes.’’ > Figure 8.4 shows a scatterplot of a strong correlation between these two properties of a genome
genome size and AT content (Pearson correlation coefficient of (Pearson correlation coefficient of 0.94). The third codon
0.48), showing that small genomes have high AT content and position is the most variable position for the codon and this is
large genomes have low AT content. The analysis also shows where the largest variation in base use would be expected. The
a cloud around the middle values, indicating that average size correlation was therefore expected and shows that high AT
corresponds to an AT content with high fluctuations. content in a genome correlates with a bias close to 1 (which
On the other hand, an interesting relation is seen between AT is 100% AT in the third codon) and low AT content correlates
content and bias in third codon position. > Figure 8.5 illustrates with a bias close to +1 (which is 100% GC in the third codon).
80 factor(Group)
Acidbacteria
Actinobacteria
Aquificae
70 Archaea Crenarchaeota
Archaea Other
60
Bacteroidetes.Chlorobi
Chlamydiae.Verrucomicrobia
Percent AT
Chloroflexi
50 Cyanobacteria
Deinococcus-Thermus
Firmicutes
Fusobacteria
40 Other Bacteria
Planctomycetes
Proteobacteria alpha
Proteobacteria beta
30 Proteobacteria delta
Proteobacteria epsilon
Proteobacteria gamma
20 Spirochaetes
Thermotogae
2000 4000 6000 8000 10000 12000

Genome size (kb)
. Fig. 8.4
Scatter plot showing percentage AT compared to total genome size (kb) for 1,500 genome sequences. The Pearson Correlation
Coefficient (PCC) for this data is 0.48, which shows a medium correlation. PCC is often used to measure the linear dependence between
two variables, and takes a value between +1 and 1, where 0 reflects no linear correlation
Phyla
80
Acidobacteria
Actinobacteria
Alphaproteobacteria
Aquificae
70 Bacteroidetes/Chlorobi
Betaproteobacteria
Chloroflexi
AT percentage
Crenarchaeota
60
Cyanobacteria
Deinococcus-Thermus
Deltaprotebacteria
Epsilonproteobacteria
50 Euryarchaeota
Firmicutes
Fusobacteria
Gammaprotebacteria
Nanoarchaeota
40
Other Archaea
Other Bacteria
Planctomycetes
Spirochaetes
30 Thermotogae
–0.5 0.0 0.5

Bias in third position
. Fig. 8.5
Scatter plot showing percentage AT compared to base bias in third codon position for 1,500 genome sequences. Bias is calculated so that
100 % A or T in third codon position gives a score of 1, 100 % G or C in third position gives a score of +1. The Pearson Correlation
Coefficient for this data is 0.94, which shows a strong correlation. PCC is often used to measure the linear dependence between two
variables, and takes a value between +1 and 1, where 0 reflects no linear correlation
Thousands of Genome Sequences the Average Nucleotide Identity (ANI) between pairs of genomes
or the Average Amino acid Identity (AAI) of the shared genes
Availability of thousands of genomes makes it possible to inves- between two genomes. The study of Goris et al. on pairwise
tigate phylogenies based on genomic information and see how comparison of complete sequenced genomes showed the ANI of
current taxonomy is affected. The development of many com- the core genes show results similar to analysis of 16S rRNA
putational tools and increasing computational power makes it sequence identity and DDH similarity values, concluding that
possible to compare whole genomes in a reasonable time, yet a 70% DDH value corresponds to 95% ANI. Hence ANI has
comparison of thousands of whole genomes is still a tedious been shown to be an alternative to the tedious DDH method
process. Therefore, three data sets were selected that represent (Goris et al. 2007). Another genome-based method, AAI, has
different taxonomic levels of prokaryotes. The first data set is been shown to result in strong correlation between 16S rRNA
chosen to cover a wide coverage of all the prokaryotic organisms gene identity and AAI-based phylogenetic trees congruent with
(126 genomes and 23 phyla). The second data set is core genome-based trees (Konstantinidis and Tiedje 2005).
a representative of a well-defined prokaryotic family There are many methods and types of data used to build
(Enterobactericiae family, 50 genomes). The third one is chosen evolutionary trees. The results of the whole-genome-based tools
as an example of a prokaryotic species and close relatives are usually values representing similarity between organisms,
(Escherichia coli, Escherichia fergusonii, Shigella). Different com- which can then be converted to distance-based phylogenies.
putational methods that we have encountered to be fitting in the Distance methods constitute a major part of phylogenetic anal-
current taxonomy of prokaryotes were shown for each data set in ysis. ‘‘Least squares’’ is one of these methods where the sum of
the following sections. squares of difference between the observed and the predicted
distances of a tree should be minimized. Unweighted (Cavalli-
Sforza and Edwards 1967) and weighted (Fitch and Margoliash
Whole-Genome-Based Tools for Taxonomy 1967; Beyer et al. 1974) algorithms are suggested for least
squares. Minimum Evolution, Neighbor joining, and UPGMA
The previous section showed the growth in available sequence are all distance-based methods. There are also methods that rely
data as well as the bias and diversity in this data. This large on probabilities of evolutionary change. Maximum likelihood is
diversity and coverage opens the doors to large-scale phyloge- one of them, where different evolutionary rates can be taken into
netic analysis of genome sequences. As a result, great insight into account and several models can be implemented (Felsenstein
bacterial evolution and diversity has come from comparison of 2004). The evolutionary models and the distance methods
many microbial genome sequences in the last decade. The dif- should be chosen carefully when phylogenies are generated, as
ferences, even between strains of a distinct taxonomic cluster, they might result in different results even for a small set.
show that bacteria represent a great diversity, which led to the
formation of the hypothetical concepts of ‘‘pan-genomes’’ and
‘‘core-genomes.’’ The pan-genome contains the total number of rRNA Phylogenetic Trees
genes found in the gene pool of a set of genomes (Ussery et al.
2009) and can be viewed in three separate parts. The part that In this section the 16S rRNA and 23S rRNA comparison of 126
consists of conserved essential genes common to all genomes various organisms from all bacterial and archaeal phyla is
compared (core-genome). It has been seen that core-genomes of presented (> Fig. 8.6). This data set represents a collection of
phylogenetically coherent groups contain genes that are less distantly related prokaryotic organisms. The first criteria for the
prone to horizontal gene transfer and are more stable such as selection of organisms for this dataset, was to get the largest and
housekeeping genes. The genes essential for colonization, sur- smallest genomes from each phyla (taxonomy reference is
vival, or adaptation to a specific environment are thought to Genome metadata from NCBI and GOLD). More organisms
form the lifestyle genes, which can also be named as the ‘‘shell’’ from each phyla were selected from different environments or
for frequently occurring genes. The third part is called ‘‘acces- host associations, in order to get a less biased data in total.
sory’’ or ‘‘cloud’’ genes, as these are rarely found, often strain Ribosomal RNA sequences of all 126 genomes were
specific and nonessential (Lapierre and Gogarten 2009). Though predicted using RNAmmer program (Lagesen et al. 2007).
hypothetical, these terms can serve use for defining and classi- Foreach genome one 16S and 23S rRNA sequence was selected
fying bacteria. These different ‘‘genomes’’ can be used to explain based on the highest RNAmmer score and appropriate length
the differences and similarities between species or genera, and (Lagesen et al. 2010). The length requirements were between
visualized by pan-genome trees (Snipen and Ussery 2010). An 1,400 and 1,700 bp for 16s rRNA sequences, 2,500 and
elaborate work on the comparisons of genomic DNA using 3,800 bp for 23S rRNA sequences. Once the RNA sequences
oligonucleotide-based methods and proteomes with a pan- were gathered they were aligned using CLUSTALW with default
genome approach was presented in a study by Bohlin et al., parameters (10 for gap opening penalty, 0.20 gap extension
where Brucella species were classified using 32 genomes (Bohlin penalty, 30% Delay divergent sequences, 0.5 for DNA transitions
et al. 2010). weight, IUB for DNA weight matrix) (Larkin et al. 2007).
Other genome-based methods include measures for replace- After obtaining the alignments, the phylogenetic trees were
ment of the DDH (DNA-DNA-Hybridization) analysis, such as constructed using MEGA5 (Tamura et al. 2011) and
. Fig. 8.6
(a)16S rRNA and (b) 23S rRNA tree with NJ method and 1,000 bootstrap resamplings from ClustalW alignment. The trees are viewed and
colored with MEGA5. Branch lengths are measured in the number of substitutions per site. Each phylum is collapsed when possible,
except classes of Proteobacteria were collapsed instead of phyla
Neighbor-Joining (NJ) with 1,000 bootstrap resamplings. The method, to make it more similar to DDH, was made by ran-
bootstrap values in these phylogenies were transformed to per- domly chopping up the genome sequences in 1,020 nucleotide
centages. They give a statistical measure for how reliable fragments regardless of whether or not they correspond to any
a branch separation is. Therefore, higher percentages support ORFs. The fragments from two genomes are aligned using
a stronger evidence of grouping, meaning a more prominent a BLAST (Altschul et al. 1990) algorithm or a fast alignment
common ancestor, whereas lower percentages mean the separa- tool such as MUMmer without fragmentation.
tion on that branch is statistically insignificant. In this section, 50 genomes from different genera of the
Ribosomal RNA phylogenies are usually able to distinguish Enterobacteriaceae family (data gathered from NCBI GenBank)
the domains, phyla, and genera in a given set. The distances on are compared based on their ANI values. ANI calculations were
this type of phylogeny show the divergence in the rRNA performed as explained in the paper by Richter and Rosello-
sequences. According to the bootstrap values in the 16S rRNA Mora’s using Jspecies (Richter and Rosselló-Móra 2009). The
phylogenetic tree (> Fig. 8.6a), the phyla level clusters are sig- genome sequence comparison was based on nucleotide MUM-
nificant with higher than 80% bootstrap value on their roots. To mer (NUCmer) which is a fast DNA alignment tool for large-
better illustrate this, the two different clades were left scale comparisons (Delcher et al. 1999). MUMmer aligns two
uncollapsed while the remaining phyla clades were collapsed. given genome sequences based on maximal unique matches
The correspondence of the significant clades to phyla in pro- (MUMs) between the sequences. A ‘‘MUM’’ is an exact string
karyotes is, however, an expected result. Phylum, as a taxonomic match that occurs once in each genome. Once the MUMs are
unit is not defined by the official nomenclature (International identified, they are sorted in ascending order according to their
Code of Nomenclature of Bacteria (Lapage et al. 1992)). The positions in the genomes. After the global MUM-alignment, the
highest rank according to the official nomenclature is a class; gaps between them are closed based on the properties of the
however, the rank phylum is also being used in prokaryotic gaps. A gap can be a single nucleotide polymorphism, an inser-
taxonomy quite often and seems to serve practical use for the tion or deletion where a large sequence is found in one but not
taxonomists. Historically, phyla were referred as divisions the other genome, tandem repeats, or polymorphic regions. If
and Prokaryotes, as one of the superkingdoms proposed by Whit- gaps are found, they are aligned using the Smith-Waterman
taker and Margulis, were divided into three divisions based on cell algorithm (Smith and Waterman 1981).
wall structure or absence (Whittaker and Margulis 1978; Gibbons Comparisons based on the tetranucleotide frequencies were
and Murray 1978). Although the classification largely changed calculated using Jspecies and the algorithms from Teeling et al.
since the division of archaeal phyla (Murray 1989) were discov- (2004). In this method, all possible combinations of
ered, most of these names are still in use today. In the 2nd tetranucleotide frequencies (256 frequencies) for each sequence
edition of Bergey’s Manual, phylum was defined as the major is calculated and their z-scores are computed based on the
prokaryotic lineages, based on the 16S rDNA sequence data and difference between the observed and the expected frequencies
used as main organizational unit (Brenner et al. 2005b). for a genomic fragment. The similarity between the two
Ribosomal RNA based phylogenies usually involves the 16S sequences (or genomic fragments) in terms of having similar
rRNA subunit comparisons. Here we show that 23S rRNA phy- patterns of tetranucleotides is addressed by calculating the Pear-
logeny can also be useful. When the same dataset is analyzed son correlation coefficient for their z-scores. Similar patterns are
using 23S rRNA genes, the bootstrap values are generally higher expected to correlate and therefore have higher correlation coef-
than 16S rRNA phylogenies (> Fig. 8.6b). There are exceptions ficients, whereas the distant patterns would have lower correla-
to this, for example, the bootstrap value on the roots of the tion coefficients. Oligonucleotide frequencies are thought to
Gamma Proteobacteria clade that is higher on the 16S rRNA tree. carry species-specific signals, where longer signatures carry
The generally higher values for the 23S rRNA analysis might be more signals. Thus, closely related organisms are expected to
due to the size or information content of the sequences and show similar distribution of the usage of these signatures.
maybe due to the different mutation rates of the genes. The > Figure 8.7 shows a pairwise genome comparison of ANI
separation of phyla on > Fig. 8.6b is significant, but the order value (heatmap). The genomes are manually ordered based on
is different, which might lead to the idea of having different 16S rRNA similarities. It is seen that DNA similarity within
relationships among different phyla. However, since the boot- a genus is higher compared to the similarity between genera. It
strap values are very low at that level, it is still not relevant to is therefore possible to distinguish groups of genus and species
conclude how close Firmicutes is to Proteobacteria based on based on their DNA similarity. For comparison, Tetra Nucleo-
rRNA phylogeny. tide frequencies were calculated for the same data and ordered
based on 16S rRNA similarity. The two heatmaps are expected to
show similar results, with ANI values above 96% identity would
Average Nucleotide Identities (ANI) and Tetra correspond to very high Tetra Nucleotide frequencies correlation
Nucleotide Frequency Calculations coefficients of 0.99 (Richter and Rosselló-Móra 2009). It is
seen that within genera, sequences are highly correlated based on
Average nucleotide identity was developed as an alternative to tetranucleotide signature usage. Changing the order of the
the DDH values, and was initially based on comparison of all matrix based on hierarchical clustering might give a better res-
shared genes among two genomes. Later on an advance in the olution using Tetra Nucleotide frequencies.
a
Shigella_dysenteriae_Sd197
Shigella_boydii_Sb227
. Fig. 8.7 (continued)

Escherichia_coli_O157.H7_str_Sakai
Shigella_sonnei_Ss046
Shigella_flexneri_5_str_8401
Escherichia_coli_str_K.12_substr_DH10B
Escherichia_fergusonii_ATCC_35469
Salmonella_enterica_subsp_enterica_serovar_Paratyphi_A_str
Salmonella_enterica_subsp_enterica_serovar_Typhi_str_CT18
Salmonella_enterica_subsp_enterica_serovar_Typhimurium_str_LT2
Citrobacter_koseri_ATCC_BAA.895
Citrobacter_rodentium_ICC168
Enterobacter_cloacae_subsp_cloacae_ATCC_13047
Cronobacter_sakazakii_ATCC_BAA.894
Cronobacter_turicensis_z3032
Erwinia_pyrifoliae_Ep1_96
Erwinia_amylovora_ATCC_49946
Erwinia_tasmaniensis_Et1_99
Enterobacter_sp_638
Klebsiella_variicola_At.22
Klebsiella_pneumoniae_342
Klebsiella_pneumoniae_subsp_pneumoniae_MGH_78578
Pantoea_ananatis_LMG_20103
Dickeya_dadantii_Ech703
Dickeya_zeae_Ech1591
Pectobacterium_carotovorum_subsp_carotovorum_PC1
Pectobacterium_atrosepticum_SCRI1043
Pectobacterium_wasabiae_WPP163
Serratia_proteamaculans_568
70
Yersinia_enterocolitica_subsp_enterocolitica_8081
Yersinia_pseudotuberculosis_IP_32953
Yersinia_pestis_Angola
80
Yersinia_pestis_CO92
Yersinia_pestis_Antiqua
Yersinia_pestis_Nepal516
Sodalis_glossinidius_str_morsitans
ANI value
Edwardsiella_ictaluri_93.146
90
Edwardsiella_tarda_EIB202
Proteus_mirabilis_HI4320
Xenorhabdus_bovienii_SS.2004
Candidatus_Hamiltonella_defensa_5AT
Photorhabdus_asymbiotica
Photorhabdus_luminescens_subsp_laumondii_TTO1
Buchnera_aphidicola_str_APS
Buchnera_aphidicola_str_Bp
Candidatus_Blochmannia_floridanus
Candidatus_Blochmannia_pennsylvanicus_str_BPEN
Wigglesworthia_glossinidia_endosymbiont_of_Glossina_brevipalpis
Candidatus_Riesia_pediculicola_USDA
Comparative Genomics
8
219
b
Escherichia_coli_O157:H7_str_Sakai
Escherichia_coli_str_K−12_substr_DH10B
Citrobacter_koseri_ATCC_BAA−895
Cronobacter_sakazakii_ATCC_BAA−894
Enterobacter_sp_638
Klebsiella_variicola_At−22
Edwardsiella_ictaluri_93−146
Xenorhabdus_bovienii_SS−2004
Escherichia_coli_O157.H7_str_Sakai
Escherichia_coli_str_K.12_substr_DH10B
Citrobacter_koseri_ATCC_BAA.895
Cronobacter_sakazakii_ATCC_BAA.894
Enterobacter_sp_638
Klebsiella_variicola_At.22
Edwardsiella_ictaluri_93.146
Xenorhabdus_bovienii_SS.2004
0.4 0.6 0.8

PCC
. Fig. 8.7
(a) Shows heatmap for ANI values between each genome retrieved by pairwise alignments with MUMmer. Darker colors indicate
higher percentages. The columns and rows are ordered based on the clusters in the 16S rRNA NJ tree. (b) Shows Pearson Correlation
coefficients yielded by tetranucleotide frequency calculations between genomes of Genera set. The comparison based on Tetra
calculations shows higher similarity between organisms as the colors get darker
BLASTMatrix Using Reciprocal Best Hits is based on an all-against-all BLASTP analysis where all proteins
are compared to all other proteins in the dataset. Then
The dataset of 50 genomes was used to illustrate the a reciprocal best-hit criterion is implemented. According to
BLASTMatrix method. Each proteome was pairwise compared this, a BLAST hit will be considered significant only if the length
using BLASTP using a reciprocal best-hit criterion. The method of the alignment is at least 95% of the longest protein and has
Escherichia colistr K−12 substr DH10B

Shigella boydii Sb227
Escherichia fergusonii ATCC35469
Shigella dysenteriae Sd197
Shigella flexneri 5 str8401
Escherichia coli O157:H7 strSakai
Shigella sonnei Ss046
Citrobacter koseri ATCCBAA−895
Citrobacterrodentium ICC168
Salmonella enterica subsp enterica serovar Paratyphi A str
Salmonella enterica subsp enterica serovar Typhi str CT18
Salmonella enterica subsp enterica serovar Typhimurium strt LT2
Cronobacter sakazakii ATCCBAA−894
Cronobacter turicensis z3032
Erwinia amylovora ATCC49946
Erwinia pyrifoliae Ep1 96
Erwinia tasmaniensis Et1 99
Enterobacter cloacae subsp cloacae ATCC 13047
Enterobactersp 638
Klebsiella pneumoniae 342
Klebsiella pneumoniae subsp pneumoniae MGH78578
Klebsiella variicola At−22
Pantoea ananatis LMG20103
Dickeya dadantiiEch703
Dickeya zeae Ech1591
Pectobacterium atrosepticum SCRI1043
Pectobacterium carotovorum subsp carotovorum PC1
Pectobacterium wasabiae WPP163
Serratia proteamaculans 568
Yersinia enterocolitica subsp enterocolitica 8081
Yersinia pestis Angol
Yersinia pestis Antiqu
Yersinia pestis CO92
Yersinia pestis Nepal516
Yersinia pseudotuberculosis IP31758
Yersinia pseudotuberculosis IP32953
Sodalis glossinidius strmorsit
Edwardsiella ictaluri 93−146
Edwardsiella tarda EIB202
Proteus mirabilis HI4320
Xenorhabdus bovienii SS−2004
Photorhabdus asymbiotic
Photorhabdus luminescens subsp laumondii TTO1
Candidatus Hamiltonella defensa 5AT
Buchnera aphidicola strAPS
Buchnera aphidicola strBp
Candidatus Blochmannia floridanus
Candidatus Blochmannia pennsylvanicus strBPEN
Wigglesworthia glossinidia endosymbiont of Glossina brevipalpi
Candidatus Riesia pediculicola USDA
Klebsiella.pneumoniae.subsp.pneumoniae.MGH.78578
Shigella.boydii.Sb227
Shigella.sonnei.Ss046
Klebsiella.pneumoniae.342
Yersinia.pestis.Nepal516
Edwardsiella.ictaluri.93.146
Candidatus.Blochmannia.floridanus
Candidatus.Riesia.pediculicola.USDA
Escherichia.coli.str.K.12.substr.DH10B
Escherichia.fergusonii.ATCC.35469
Shigella.dysenteriae.Sd197
Shigella.flexneri.5.str.8401
Escherichia.coli.O157.H7.str.Sakai
Citrobacter.koseri.ATCC.BAA.895
Citrobacter.rodentium.ICC168
Salmonella.enterica.subsp.enterica.serovar.Paratyphi.A.str
Salmonella.enterica.subsp.enterica.serovar.Typhi.str.CT18
Salmonella.enterica.subsp.enterica.serovar.Typhimurium.str.LT2
Cronobacter.sakazakii.ATCC.BAA.894
Cronobacter.turicensis.z3032
Erwinia.amylovora.ATCC.49946
Erwinia.pyrifoliae.Ep1.96
Erwinia.tasmaniensis.Et1.99
Enterobacter.cloacae.subsp.cloacae.ATCC.13047
Enterobacter.sp.638
Klebsiella.variicola.At.22
Pantoea.ananatis.LMG.20103
Dickeya.dadantii.Ech703
Dickeya.zeae.Ech1591
Pectobacterium.atrosepticum.SCRI1043
Pectobacterium.carotovorum.subsp.carotovorum.PC1
Pectobacterium.wasabiae.WPP163
Serratia.proteamaculans.568
Yersinia.enterocolitica.subsp.enterocolitica.8081
Yersinia.pestis.Angol
Yersinia.pestis.Antiqu
Yersinia.pestis.CO92
Yersinia.pseudotuberculosis.IP.31758
Yersinia.pseudotuberculosis.IP.32953
Sodalis.glossinidius.str.morsit
Edwardsiella.tarda.EIB202
Proteus.mirabilis.HI4320
Xenorhabdus.bovienii.SS.2004
Photorhabdus.asymbiotic
Photorhabdus.luminescens.subsp.laumondii.TTO1
Candidatus.Hamiltonella.defensa.5AT
Buchnera.aphidicola.str.APS
Buchnera.aphidicola.str.Bp
Candidatus.Blochmannia.pennsylvanicus.str.BPEN
Wigglesworthia.glossinidia.endosymbiont.of.Glossina.brevipalpi
20 80
Percent
. Fig. 8.8
Proteome comparison between the genomes of Enterobacteriaceae based on BLASTP searches
95% sequence identity. There are three possible outcomes for best hits for protein X and vice-a-versa. When a proteome is
each protein: (1) The protein does not have any significant hits BLAST searched against itself, a protein will have a hit to itself
to any other protein, (2) the protein has one or more hits where and it might have another hit to another protein in the same
the hit with the highest bitscore is chosen as best hit, and (3) the genome. The results are then put through homology reduction
protein has more than one hit where there can be several best by the HOBOHM algorithm, in order to reduce the self-hits
hits. Once identified based on this three possibilities, best hits for (Hobohm et al. 1992).
each protein are stored. For each proteome, a hit will be counted The output of the BLASTMatrix analysis is a matrix of
if the best hit of protein X in proteome A is also the best hit of the numbers. > Figure 8.8 shows a heatmap of these values though
corresponding protein Y on proteome B. In the case where there not the actual value. In the discussion below the actual values are
are several best hits, the hit is counted if protein Y is one of the presented. In the matrix for the Enterobacteriaceae family, the
42 Citrobacter_rodentium_ICC168_uid43089
13 Citrobacter_koseri_ATCC_BAA_895_uid58143
45 Enterobacter_cloacae_ATCC_13047_uid48363
18 Enterobacter_638_uid58727
100 Cronobacter_turicensis_z3032_uid40821
18 Cronobacter_sakazakii_ATCC_BAA_894_uid58145
94Klebsiella_variicola_At_22_uid42113
25 94 Klebsiella_pneumoniae_342_uid59145
Klebsiella_pneumoniae_MGH_78578_uid57619
39 Salmonella_enterica_serovar_Typhi_CT18_uid57793
39 Salmonella_enterica_serovar_Paratyphi_A_AKU_12601_uid59269
46 Salmonella_enterica_serovar_Typhimurium_LT2_uid57799
21 Shigella_dysenteriae_Sd197_uid58213
Escherichia_fergusonii_ATCC_35469_uid59375
12 59 Escherichia_coli_O157_H7_Sakai_uid57781
Escherichia_coli_K_12_substr__DH10B_uid58979
36
52 Shigella_sonnei_Ss046_uid58217
37 Shigella_boydii_Sb227_uid58215
Shigella_flexneri_5_8401_uid58583
11 Sodalis_glossinidius__morsitans__uid58553
1 Pantoea_ananatis_LMG_20103_uid46807
Xenorhabdus_nematophila_ATCC_19061_uid49133
0 23 Proteus_mirabilis_HI4320_uid61599
Dickeya_zeae_Ech1591_uid59297
66 0 43 Dickeya_dadantii_Ech703_uid59363
32 Yersinia_enterocolitica_8081_uid57741
1 Serratia_proteamaculans_568_uid58725
6 50 Photorhabdus_luminescens_laumondii_TTO1_uid61593
Photorhabdus_asymbiotica_ATCC_43949_uid59243
100 Edwardsiella_tarda_EIB202_uid41819
12 Edwardsiella_ictaluri_93_146_uid59403
61 Erwinia_pyrifoliae_Ep1_96_uid40659
24 50 Erwinia_amylovora_ATCC_49946_uid46943
Erwinia_tasmaniensis_Et1_99_uid59029
42 Pectobacterium_carotovorum_PC1_uid59295
42 Pectobacterium_atrosepticum_SCRI1043_uid57957
73 Pectobacterium_wasabiae_WPP163_uid41297
55 Wigglesworthia_glossinidia_endosymbiont_of_Glossina_brevipalpis_uid57853
13
0 Candidatus_Riesia_pediculicola_USDA_uid46841
Buchnera_aphidicola_Bp__Baizongia_pistaciae__uid57827
215 Buchnera_aphidicola_APS__Acyrthosiphon_pisum__uid57805
5 Candidatus_Hamiltonella_defensa_5AT__Acyrthosiphon_pisum__uid59289
24 Candidatus_Blochmannia_pennsylvanicus_BPEN_uid58329
Candidatus_Blochmannia_floridanus_uid57999
46Yersinia_pestis_Nepal516_uid58609
46Yersinia_pestis_Antiqua_uid58607
77 Yersinia_pestis_CO92_uid57621
76 Yersinia_pestis_Angola_uid58485
84Yersinia_pseudotuberculosis_IP_32953_uid58157
Yersinia_pseudotuberculosis_IP_31758_uid58487
7 6 5 4 3 2 1 0
Relative distance
. Fig. 8.9
CVTree for the Genera set, generated by using ‘‘euclidian’’ distances with ‘‘ward’’ linkage hierarchical clustering. The values on branches
indicate bootstrap values
order is based on the rRNA phylogenies, because a clear separa- a Klebsiella pneumoniae because it has 94% similarity to the K.
tion of each genus was possible. According to this matrix, prote- pneumoniae 342. Another issue is the presence of the reduced
ome similarity within some genera might seem to be higher than genomes in this family. In this plot, the reduced genomes are
others; however, the darkest colors around the diagonal (90%) seen in the upper right corner (> Fig. 8.8). The percentage of
are due to the comparison within a species level, rather than proteins that these organisms share with the others is very high
different species in the same genus. Average proteome similarity because of the small size of their proteomes, which creates a dark
between Escherichia and Shigella clusters is 73%. Homology band on the top of the matrix and a lighter one on the right hand
between Salmonella strains are 80–96% while Citrobacter species side. This might be due to these genomes containing generally
show an average of 72% similarity. The similarity values for conserved core genes of the whole family, and they have actually
Cronobacter are 88% on average, 80% for Erwinia, around 70% very few accessory genes compared to the other genomes. They
for Enterobacter. Klebsiella species have up to 94% similarity, also have very low internal homology. A Shigella genome, on the
Dickeya species around 74% and Pectobacterium species have other hand, has up to 30% internal homology. The largest
75–83% similarity (> Fig. 8.8). In the Yersinia cluster, similarity proteome, Sodalis, shares 32–40% with the genomes that are
values between the strains of same species are 92–98% for not among the reduced genomes group. Except the reduced and
Y. pestis and 88% for Y. pseudotuberculosis. Homology between expanded genomes, the similarity levels among different genera
Yersinia species is around 88%. Edwardsiella species have 81%, range between 40% and 72%.
Buchnera strains have around 89%, Candidatus Blochmannia
strains have around 96% similarity.
It is not clear if there should be a proteome similarity cut-off Composition Vector Trees (CVTree)
to specify genera, however, the values within a genus are gener-
ally between 70% and 80% and within the species it is higher, In this section, a CVTree for the 50 Enterobacteriaceae genomes
80–98%. For example, Klebsiella variicola can actually be is presented. In this method, frequencies of overlapping

80
Shigella flexneri 5 str 8401
Escherichia coli UMN026
100
68 Escherichia coli 042
pathogenic E. coli 27 Escherichia coli SMS−3−5
Escherichia coli IAI39
20 Escherichia coli O157:H7 str EDL933
100
Escherichia coli O157:H7 str Sakai
100
Escherichia coli O157:H7 str TW14359
100 100
Escherichia coli O157:H7 str EC4115
Escherichia coli O55:H7 str CB9615
Escherichia coli str K−12 substr W3110
98
98 Escherichia coli str K−12 substr MG1655
73 Escherichia coli DH1
20 73 Escherichia coli str K−12 substr DH10B
98 Escherichia coli BW2952
79
Escherichia coli ATCC 8739
97
Escherichia coli BL21−Gold−DE3pLysS AG
67
78 67 Escherichia coli BL21 DE3
Escherichia coli B str REL606
Escherichia coli HS
37 Escherichia coli O26:H11 str 11368
75
75 Escherichia coli O103:H2 str 12009
Escherichia coli O111:H− str 11128
83 98 Escherichia coli IAI1
81
77 Escherichia coli SE11
94 Escherichia coli E24377A
100 Escherichia coli 55989
Escherichia coli APECO1
78
Escherichia coli S88
77
Escherichia coli UTI89
81 81 95
Escherichia coli IHE3034
66 Escherichia coli CFT073
100 80 Escherichia coli 536
79 Escherichia coli O127:H6 str E234869
Escherichia coli ED1
Escherichia fergusonii ATCC 35469
0.15 0.10 0.05 0.00

Relative manhattan distance
. Fig. 8.10
Pan-genome tree for the Escherichia and Shigella set based on the ‘‘shell’’ genes of the data set. The gray boxes indicate pathogenic
strains of E. coli
oligopeptides of length K are calculated. The random back- ‘‘Euclidian’’ distance function and hierarchical clustering using
ground is subtracted from these frequencies with a (K-2) order the ‘‘ward’’ method (Ward 1963). For statistical significance
Markov model to avoid the bias from neutral mutations, 100 bootstrap samplings were made on these trees, giving
highlighting the selective evolution. After this procedure, the values between 0 and 100, showing the reliability for
pairwise distance is calculated using the correlation between two each branches.
organisms. The method describes the resulting distances as The CVTree for the Enterobactericiae family can be seen in
D = (1 C)/2, where C is the correlation between genomes > Fig. 8.9. In this tree, Yersinia pestis and pseudotuberculosis
and D is the distance (Qi et al. 2004). The updated CVTree species are clearly separated from the rest of the taxa with 73%
method with a faster and more stable web server performance bootstrap value, whereas Yersinia enterocolitica remains clus-
is described in Xu and Hao (2009). The computation was tered with Serratia. Two main clusters are seen after this, where
performed on the web server using proteomes for all genomes the first consists of E. coli, Shigella, Salmonella, Klebsiella,
and a K-parameter set to 6. Cronobacter, Citrobacter, and Enterobacter. The branching
The outcome from the analysis is a distance matrix based order is very similar to rRNA phylogeny as well.
on amino acid sequence comparison, which is then used to To summarize, the CVTree is an alignment-free method,
generate phylogenetic trees using neighbor joining. The where the sequence similarities are investigated by the use of
web server provides an NJtree, made using the PHYLIP pack- hexamer frequencies (K = 6); therefore, it is also quite fast. It is
age, with a Newick format output. However, the NJtree seen that the distance matrix generated by the algorithm does
did not provide bootstrap values so the data was instead not give a clear picture for classification when very diverse
analyzed using R, version 2.11. With this approach, the dis- organisms are analyzed. On a genera level, all the genera are
tance matrix from the CVTrees output was used to generate intact and the clustering is very similar to the rRNA trees. On the
phylogenetic trees with bootstrap value. In > Fig. 8.9, new species level, Shigella and E. fergusonii ATCC 35469 are separated
distances were calculated from the distance matrix using the from E. coli.
Escherichia coli BW2952

68
Escherichia coli str K-12 substr MG1655
91
Escherichia coli str K-12 substr DH10B
74
Escherichia coli 55989
49 57 Escherichia coli 536
Escherichia fergusonii ATCC 35469
Escherichia coli O127 H6 str E2348 69
39 44 Shigella flexneri 5 str 8401
44 Escherichia coli ED1
40 Escherichia coli IHE3
Escherichia coli CFT073
Escherichia coli BL21-Gold-DE3pLysS AG
31 Escherichia coli APEC O1
Escherichia coli S88 pathogenic E. coli
64
73 Escherichia coli UTI89
Escherichia coli HS
70 Escherichia coli O111 H- str 11128
45 Escherichia coli O26 H11 str 11368
88 Escherichia coli O157 H7 str EDL933
45 Escherichia coli DH1
40 Escherichia coli str K-12 substr W3110
Escherichia coli O103 H2 str 12009
88
Escherichia coli SE11
63 Escherichia coli IAI39
Escherichia coli 042
85 55 65 Escherichia coli SMS-3-5
Escherichia coli ATCC 8739
Escherichia coli BL21 DE3
83
Escherichia coli B str REL606
85
58 Escherichia coli E24377A
35 Escherichia coli UMN026
Escherichia coli O157 H7 str Sakai
87 Escherichia coli IAI1
51 Escherichia coli O157 H7 str EC4115
43 Escherichia coli O157 H7 str TW14359
Escherichia coli O55 H7 str CB9615
0.001
. Fig. 8.11
16S rRNA phylogenetic tree for Escherichia and Shigella genomes. Sequences are obtained as explained in rRNA phylogenies previously.
The tree was generated using ClustalW alignments with neighbor-joining method and statistically tested with 1,000 bootstrap
resamplings
Pan-genome Trees Once the gene families are assigned, a matrix is constructed
containing the gene families in columns and the genomes in
Pan-genome family trees were generated using BLASTP rows, having 1 for the presence of that gene family in the
(Altschul et al. 1990) similarity between each proteome. corresponding genome, 0 otherwise. The tree is constructed by
According to this, genes that have a significant hit to each calculating Manhattan distances from this matrix and making
other are considered to be in one gene family, where the signif- hierarchical clustering using the Unweighted Pair-Group
icance cut-off is chosen for each BLAST hit (50% identity over Method with Arithmetic mean (UPGMA) algorithm. For the
an alignment with a length of at least 50% of the longest gene). stabilome view, the gene families that are represented in only one
genome (ORFans) are weighted down and the tree is based on . Table 8.1
‘‘shell’’ genes between the genomes (Snipen and Ussery 2010). Methods that can be used for investigating inter- and intra-taxa
In this section, a pan-genome tree for genomes of 36 relationships. The highest level is the Three Domains of life, and
Escherichia coli and 4 Shigella species was constructed (data the lowest level is within the Strains
gathered from NCBI GenBank) (> Fig. 8.10). This tree, com-
Taxonomic levels Inter-taxa Intra-taxa
pared to a 16S rRNA neighbor-joining phylogeny (> Fig. 8.11),
shows more clear separation on the pathogenicity of the E. coli Superkingdom 16S and 23S rRNA
strains. The pan-genome tree shows relationships among phylogeny
different strains of a family with a higher resolution. Shigella Phyla None 16S and 23S rRNA
species and E. fergusonii ATCC 35469 are also clearly separated phylogeny
from the E. coli, showing that they show clear differences in Genus 16S and 23S rRNA 16S and 23S rRNA
their proteome composition, therefore, they can be separated phylogeny, phylogeny ANIm and
from E. coli. BLASTMatrix Tetra, CVTree,
BLASTMatrix,
Pan-genome tree
Summary Species 16S and 23S rRNA 16S and 23S rRNA
phylogeny, phylogeny,
BLASTMatrix, BLASTMatrix, Pan-
This chapter presents analysis on genomic data that is in public
Pan-genome tree, genome tree, CVTtee,
databases and comparative genomics approaches to taxonomy
CVTtee, ANI and ANI and Tetra
based on rRNA, DNA, and protein molecular sequences. Tetra
It is clear that the available genome data is biased which
Strain BLASTMatrix, Pan-genome tree
cannot be attributed to any one reason. However, the monocul-
Pan-genome tree,
ture approach to genome sequencing is causing a significant CVTree
skew in sequencing data. From experiments, it is known that
the diversity in the microbial world is tremendous, but in the
statistical results, this diversity is not covered in the sequenced
data. The advances in metagenomic sequencing and the
sequencing of noncultivable cells will in time result in a much differences between all the methods in the sense of using the
more realistic view of the bacterial world than is seen now. In the sequence information; some use the sequence directly and make
mean time, scientists should be aware of the bias in sequence use of alignments and some reduces this information content
data and not believe that what is sequenced so far is representa- into vectors of numbers. The latter can be thought more as
tive, even with roughly 2,000 genomes finished, and another numerical taxonomy, where several properties of organisms are
3,000 ‘‘draft’’ genomes available; there are currently about measured and statistical significance tests and clustering
30,000 bacterial genomes available in the ‘‘short-read archives,’’ methods are used to analyze relationships. As a result of all this
which in principle could be assembled, sometimes into less than analysis, it is suggested that there might not actually be one
100 pieces—this means that there are ~35,000 genomes available unified theory on taxonomy of living things, but several which
now, and within a year or two, the number is likely to be in the classify well in different taxonomic levels. These methods are
hundreds of thousands. It seems likely that in the near future, shown in > Table 8.1, where a method explaining the relation-
draft genomes will become more common; if done properly and ships among different taxa are referred as inter-taxa, and
assembled well into only a few contigs, these draft genomes can methods that can delineate specific taxa from others regardless
provide useful information for core- and pan-genomes of of their relations with others are referred as intra-taxa.
a given taxa. However, one can hope that in the not too distant As seen from the results, the largest phylum in terms of
future, emerging third-generation sequencing technology will having the highest bacterial genome sequence projects,
allow for the economical production of high quality full-length Proteobacteria, seems to be a well-defined taxa, where most of
genomes for more reliable and robust information. the methods catches the clustering, and the classes of Alpha,
The amount of available data makes sequence-based taxon- Beta, Gamma, Delta, and Epsilon Proteobacteria are usually
omy inevitable. In this chapter, organisms with different levels of coherent within themselves based on rRNA base taxonomy.
taxonomic relations were selected. Since the reference taxonomy Second largest phyla, Firmicutes are usually clearly separated
used is the current taxonomy, the results shown have been into two groups of different classes. Another group,
selected to be as close as possible to current taxonomy. rRNA Cyanobacteria is actually a subdivision, because they do not
phylogenies were used since it is a classical approach. Although have any classes or orders defined. Its members usually cluster
the method is based on a single gene, the more conservative together in many methods. All the other phyla generally have
nature of the rRNA genes gives them a unique advantage of their members clustered together. This makes sense in classifi-
identifying more distantly related organisms. Whole-genome cation, if looking for clearly separated groups. However, from an
approaches, on the other hand, are more precise for understand- evolutionary point of view, this result is not enough to under-
ing relations among closely related organisms. There are also stand the ancestral relations of different phyla.
In the levels of genera, relationships for Enterobacteriaceae Cavalli-Sforza LL, Edwards AWF (1967) Phylogenetic analysis: Models and esti-
mation procedures. Evolution 32:550–570
family can be seen in many methods. These families include
Chain P, Grafham D, Fulton R, Fitzgerald M (2009) Genome project standards in
some clinically and industrially important bacteria. Genera a new era of sequencing. Science 326:4–5
inside Enterobacteriaceae are clearly separated into different Cohn F (1872) Untersuchungen tiber Bakterien II. Beitr Biol Pflanz 1:127–224
clusters, although sometimes the Enterobacter genus clusters Delcher A, Kasif S, Fleischmann R, Peterson J, White O, Salzberg S (1999)
separately. The relations of reduced genomes vary in different Alignment of whole genomes. Nucleic Acids Res 27:2369–2376
DeWall MT, Cheng DW (2011) The minimal genome–a metabolic and environ-
methods but generally they are separated from the rest of the
mental comparison. Brief Funct Genomic Proteomic 105:312–315
genera. They share the core genes with the rest of the family and Doolittle W (1999) Phylogenetic classification and the universal tree. Science
have very few accessory genes. 284:2124–2128
On a species level it is seen that even though in the same Felsenstein J (2004) Inferring phylogenies. Sinauer Associates, Sunderland
species, bacteria can be very diverse in terms of proteome con- Fitch WM, Margoliash E (1967) Construction of phylogenetic trees. Science
155:279–284
tent. Distinguishing between different types of E. coli strains are
Fleischmann RD, Adams MD, White O, Clayton RA, Kirkness EF, Kerlavage AR,
not possible with classical methods, but with proteome compar- Bult CJ, Tomb JF, Dougherty BA, Merrick JM (1995) Whole-genome ran-
isons. It is also clear that although they are historically known to dom sequencing and assembly of Haemophilus influenzae Rd. Science
be similar, Shigella and E. coli can be distinguished from each 269:496–512
other with proteome comparison. Fraser CM, Gocayne JD, Others (1995) The minimal gene complement of Myco-
plasma genitalium. Science 270:397–403
In theory, classification simply depends on how one defines
Galperin MY (2006) The minimal genome keeps growing. Environ Microbiol
the relatedness among the entities of the system. However, in 84:569–573
reality the choice of the characters when building a system is not Garrity GM, Lilburn GT, Cole JR, Harrison SH, Euzeby J, Tindall BJ (2007) Intro-
always simple. Whole-genome-based tools do not consistently duction to the taxonomic outline of bacteria and archaea (TOBA) Release
agree with current taxonomy. In order to make them match 7.7. The Taxonomic Outline of Bacteria and Archaea. http://www.taxonomi-
coutline.org/index.php/toba/article/download/190/223. Accessed 23 Feb
the current taxonomical system, different methods should
2012
be used for the different levels of taxonomy. The emergence of Gibbons N, Murray R (1978) Proposals concerning the higher taxa of bacteria. Int
large amount of molecular and genomic data made it evident J Syst Evol Microbiol 28:1–6
that there is not one universal method to naturally classify Goris J, Konstantinidis K, Klappenbach J, Coenye T, Vandamme P, Tiedje J (2007)
prokaryotes. Taxonomists should therefore keep the skepticism DNA-DNA hybridization values and their relationship to whole-genome
sequence similarities. Int J Syst Evol Microbiol 57:81–91
when using genomic data and using the common
Harayama S, Kasai H (2006) Bacterial phylogeny reconstruction from
methodologies. molecular sequences. In: Stackebrandt E (ed) Molecular identification, sys-
tematics, and population structure of prokaryotes. Springer, Berlin/
New York, pp 105–140
Hobohm U, Scharf M, Schneider R, Sander C (1992) Selection of representative
References protein data sets. Protein Sci 1:409–417
Konstantinidis K, Tiedje J (2005) Genomic insights that advance the species
Aanensen D, Spratt B (2005) The multilocus sequence typing network: mlst. net. definition for prokaryotes. Proc Natl Acad Sci USA 102:2567–2572
Nucleic Acids Res 33:W728–W733 Lagesen K, Hallin P, Rødland EA, Staerfeldt HH, Rognes T, Ussery DW (2007)
Abt B, Foster B, Others (2010) Complete genome sequence of cellulomonas RNAmmer: consistent and rapid annotation of ribosomal RNA genes.
flavigena type strain (134). Stand Genomic Sci 3:15–25 Nucleic Acids Res 35:3100–3108
Altschul S, Gish W, Miller W, Myers E, Lipman D (1990) Basic local alignment Lagesen K, Ussery DW, Wassenaar TM (2010) Genome update: the 1000th
search tool. J Mol Biol 215:403–410 genome–a cautionary tale. Microbiology 156:603–608
Ansorge WJ (2009) Next-generation DNA sequencing techniques. New Lapage S, Sneath P, Lessel E, Skerman V, Seeliger H, Clark W (1992) International
Biotechnol 25:195–203 code of nomenclature of bacteria: bacteriological code, 1990 revision. Amer-
Arslan D, Legendre M, Seltzer V, Abergel C, Claverie J (2011) Distant Mimivirus ican Society of Microbiology, Washington, DC
relative with a larger genome highlights the fundamental features of Lapierre P, Gogarten J (2009) Estimating the size of the bacterial pan-genome.
Megaviridae. Proc Natl Acad Sci 108:17486–17491 Trends Genet 25:107–110
Benson DA, Karsch-Mizrachi I, Lipman DJ, Ostell J, Wheeler DL (2008) GenBank. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H,
Nucleic Acids Res 36:D25–D30 Valentin F, Wallace IM, Wilm A, Lopez R, Thompson JD, Gibson TJ,
Bernal A, Ear U, Kyrpides N (2001) Genomes OnLine Database GOLD: a monitor Higgins DG (2007) Clustal W and Clustal X version 2.0. Bioinformatics
of genome projects world-wide. Nucleic Acids Res 29:126–127 23:2947–2948
Beyer WA, Stein M, Smith TF, Ulam S (1974) A molecular sequence metric and Liolios K, Chen I-MA, Mavromatis K, Tavernarakis N, Hugenholtz P, Markowitz
evolutionary trees. Math Biosci 19:9–25 VM, Kyrpides NC (2010) The genomes on line database GOLD) in 2009:
Bohlin J, Snipen L, Cloeckaert A, Lagesen K, Ussery D, Kristoffersen AB, status of genomic and metagenomic projects and their associated metadata.
Godfroid J (2010) Genomic comparisons of Brucella spp and closely related Nucleic Acids Res 38:D346–D354
bacteria using base compositional and proteome based methods. BMC Maiden M, Bygraves J, Feil E, Morelli G, Russell J, Urwin R, Zhang Q, Zhou J,
Evol Biol 10:249 Zurth K, Caugant D, Others (1998) Multilocus sequence typing: a portable
Brenner DJ, Krieg NR, Staley JT, Garrity GM, Boone DR, Vos P, Goodfellow M, approach to the identification of clones within populations of pathogenic
Rainey FA, Schleifer KH (2005a) Bergeys manual® of systematic bacteriol- microorganisms. Proc Natl Acad Sci USA 95:3140–3145
ogy, vol 2. Springer, Boston McCutcheon JP, McDonald BR, Moran NA (2009) Origin of an alternative genetic
Brenner DJ, Staley JT, Krieg NR (2005b) Classification of procaryotic organisms code in the extremely small and GC-rich genome of a bacterial symbiont.
and the concept of bacterial speciation. In: Brenner DJ, Krieg NR, Staley JT, PLoS Genet 57:e1000565
Garrity GM (eds) Bergey’s manual® of systematic bacteriology. Springer, Médigue C, Moszer I (2007) Annotation, comparison and databases for hundreds
Boston, pp 27–32 of bacterial genomes. Res Microbiol 158:24–36
Moya A, Gil R, Latorre A, Peretó J, Pilar Garcillán-Barcia M, de la Cruz F (2009) Teeling H, Meyerdierks A, Bauer M, Amann R, Glöckner FO (2004) Application
Toward minimal bacterial cells: evolution vs. design. FEMS Microbiol Rev of tetranucleotide frequencies for the assignment of genomic fragments.
331:225–235 Environ Microbiol 6:938–947
Murray RGE (1989) The higher taxa, or, a place for everything. . .? In: Williams Tindall BJ, Kämpfer P, Euzéby JP, Oren A (2001) Valid publication of names of
ST, Sharpe ME, Holt JG (eds) Bergey’s manual of systematic bacteriology, prokaryotes according to the rules of nomenclature: past history and current
vol 4, 1st edn. Williams & Wilkins, Baltimore, pp 2329–2332 practice. Int J Syst Evol Microbiol 56:2715–2720
Pérez-Brocal V, Gil R, Ramos S, Lamelas A, Postigo M, Michelena JM, Silva FJ, Ussery D, Wassenaar T, Borini S (2009) Computing for comparative microbial
Moya A, Latorre A (2006) A small microbial genome: the end of a long genomics: bioinformatics for microbiologists. Springer, London
symbiotic relationship? Science 314:312–313 Ward JH, Jr. (1963) Hierarchical grouping to optimize an objective function.
Qi J, Luo H, Hao B (2004) CVTree: a phylogenetic tree reconstruction tool based J Am Stat Assoc 48:236–244
on whole genomes. Nucleic Acids Res 32:W45–W47 Waters E, Hohn MJ, Others (2003) The genome of Nanoarchaeum equitans:
Richter M, Rosselló-Móra R (2009) Shifting the genomic gold standard for the insights into early archaeal evolution and derived parasitism. Proc Natl
prokaryotic species definition. Proc Natl Acad Sci 106:19126–19131 Acad Sci USA 100:12984–12988
Sanford RA, Cole JR, Tiedje JM, Al SET, Icrobiol APPLENM (2002) Characteri- Wayne L, Brenner D, Colwell R, Grimont P, Kandler O, Krichevsky M, Moore L,
zation and description of Anaeromyxobacter dehalogenans gen. nov., sp. nov., Moore W, Murray R, Stackebrandt E, Others (1987) International
an aryl-halorespiring facultative anaerobic myxobacterium. Appl Environ committee on systematic bacteriology. Report of the ad hoc committee on
Microbiol 68:893–900 reconciliation of approaches to bacterial systematics. Int J Syst Evol
Sboner A, Mu XJ, Greenbaum D, Auerbach RK, Gerstein MB (2011) The real cost Microbiol 37:463–464
of sequencing: higher than you think! Genome Biol 12:125 Whittaker R, Margulis L (1978) Protist classification and the kingdoms of organ-
Schleifer KH (2009) Classification of Bacteria and Archaea: past, present and isms. Biosystems 10:3–18
future. Syst Appl Microbiol 32:533–542 Woese C, Fox G (1977) Phylogenetic structure of the prokaryotic domain: the
Schneiker S, Perlova O, Others (2007) Complete genome sequence of the primary kingdoms. Proc Natl Acad Sci USA 74:5088–5090
myxobacterium Sorangium cellulosum. Nat Biotechnol 25:1281–1289 Woese C, Kandler O, Wheelis M (1990) Towards a natural system of organisms:
Smith T, Waterman M (1981) Identification of common molecular subsequences. proposal for the domains Archaea, Bacteria, and Eucarya. Proc Natl Acad Sci
J Mol Biol 147:195–197 USA 87:4576–4579
Snipen L, Ussery DW (2010) Standard operating procedure for computing Xu Z, Hao B (2009) CVTree update: a newly designed phylogenetic study plat-
pangenome trees. Stand Genomic Sci 2:135–141 form using composition vectors and whole genomes. Nucleic Acids Res
Stackebrandt E (2006) Exciting times: the challenge to be a bacterial systematist. 37:174–178
Molecular Identification, Systematics, and Population Structure of Yamamoto S, Harayama S (1996) Phylogenetic analysis of acinetobacter
Prokaryotes 1–21 strains based on the nucleotide sequences of gyrB genes and on the
Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011) MEGA5: amino acid sequences of their products. Int J Syst Evol Microbiol 46:
molecular evolutionary genetics analysis using maximum likelihood, evolu- 506–511
tionary distance, and maximum parsimony methods. Mol Biol Zuckerkandl E, Pauling LB, Kasha M, Pullman B (1962) Molecular disease,
Evol 28(10):2731–2739 evolution, and genetic heterogeneity. Academic, New York, pp 189–225
9 Defining Taxonomic Ranks
Konstantinos T. Konstantinidis1 . Erko Stackebrandt2
1
School of Civil and Environmental Engineering and School of Biology, Georgia Institute of Technology,
Atlanta, GA, USA
2
Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures, Braunschweig,
Germany
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 routine. Indeed, the problem of changing names of taxa is an

Past Classification Attempts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 inherent character of classification, which can be explained by
the desire of taxonomists to provide the user of taxonomy with
Phylogeny Is Based on Homology . . . . . . . . . . . . . . . . . . . . . . . . . . 232 a system that, to their opinion, optimally reflects the natural
relatedness between the taxa. Looking back in the history of
A Word About Horizontal Gene Transfer (HGT) . . . . . . . . . . 233 microbiology, the restricted interests in classification can be
explained by the enormous problems of past generations of
Main Phylogenetic Parameters for Classification of systematists to order the increasing wealth of phenotypic and
Ranks Above the Genus Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 genotypic properties of the rapidly growing numbers of bacterial
isolates. The user of taxonomy was confronted with constantly
Main Phylogenetic Parameters for Classification of changing classification concepts and classification systems, tax-
Genera and Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236 onomic rearrangements, and synonymy of names. Problems also
arose from the terminology: While some regard systematics and
Classification Is a Dynamic Process (Revised from taxonomy synonymous, others define taxonomy as the theory
Stackebrandt 1992) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 and practice of classifying organisms, while systematics cover
DNA-DNA Hybridization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 a broader scope including the evolutionary and phylogenetic
Translating Traditional Standards to Portable Sequence components. For many researchers the only aspect of taxonomy
Information: The ANI Approach . . . . . . . . . . . . . . . . . . . . . . . . 239 they come into contact with is nomenclature and may only
The Use of Typing Methods and MALDI TOF . . . . . . . . . . 240 notice this component when they are confronted with name
Correlation of Individual Phylogenetic Parameters . . . . . 241 changes. However, systematics does not include only the naming
Comparison of Phylogenetic Patterns Derived from 16S of organisms but also includes the following features
rRNA Gene Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 (Stackebrandt et al. 1999): Classification is the generation of as
much data on the properties of a novel isolate as possible in
Delineation of Genera: Pragmatic Approach . . . . . . . . . . . . . . 242 order to decide whether, by the process of identification, e.g.,
comparison of these data with the database of previously classi-
The Prokaryotic Species: A Historical View . . . . . . . . . . . . . . . 243 fied organisms, these characters justify the affiliation of the
isolate to a described species, or whether a new species need to
The Pragmatic Species Definition . . . . . . . . . . . . . . . . . . . . . . . . . 245 be described for the isolate. Classification includes the theory
Delineating a Species in the Frame of the Pragmatic and process of ordering the characterized organisms into a single
Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 or multiple classification systems. Nomenclature is the naming
of the appropriate taxa within the realm of a classification sys-
Definition of a Subspecies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 tem, and it includes subjective changes in names whenever novel
insights into the taxonomic weight of characters stimulate tax-
Species Concept in the Era of Metagenomics . . . . . . . . . . . . . . 247 onomists to change the rank of taxa. This chapter will not cover
the nomenclatural aspects of taxonomy as this is governed and
updated in detail by the articles of the Bacteriological Code
Introduction (Lapage et al. 1975).
As outlined previously (Stackebrandt 1992), several classifi-
Those who have chosen systematics, classification, and taxon- cation systems exist in parallel, and no classification system can
omy as a research topic have learned to consider this complex claim to be the sole classification. Any two systems do not need
topic an exciting and important biological discipline. For others, to match in the clustering of organisms they represent and as
these topics are mainly dull subjects, which, through changes in long as a system does not fail to do what it sets out to do, it
names of microbial taxa, may cause confusion in their daily cannot be described as wrong, or in error.
230 9 Defining Taxonomic Ranks
. Table 9.1
Examples for the classification of prokaryotic species according to risk groups (European Community classification)
Genus Risk group 1 Risk group 2 Risk group 3

Chlamydia Not known C. trachomatis C. psittaci
Bacillus B. circulans None B. anthracis
Burkholderia Bu. cocovenenans Bu. cepacia Bu. mallei
Bu. antropogonis Bu. vietnamensis Bu. pseudomallei
Francisella Not known F. tularensis subsp. holarctica F. tularensis subsp. tularensis subsp. tularensis
Mycobacterium M. asiaticum M. avium subsp. avium M. leprae
M. fallax M. chelonae M. tuberculosis
. Table 9.2
Examples for a classification system that is based on phenotypic properties. Members of the family Bacteroidaceae are described to be
Gram-negative, fermentative anaerobic organisms (Holt et al. 1994)
Family Genera Main diagnostic differences

Bacteroidaceae Bacteroides Peritrichous straight rod; produces a mixture of fermentation products from carbohydrate and
peptone; butyrate not a major product
Fusobacterium Nonmotile straight rod; butyrate is a major product
Leptotrichia Nonmotile straight rod; lactate is the sole major fermentation product
Butyrivibrio Motile, not peritrichous curved rod; butyric acid is the major fermentation product
Succinimonas Short motile rod or coccobacilli; single polar flagellum; succinate and acetate are major fermentation
products
Succinivibrio Motile helical or spiral-shaped cell; single polar flagellum; succinate and acetate are major fermentation
products
Anaerobiospirillum Motile helical or spiral-shaped cell; bipolar tufts of flagella; succinate and acetate are major
fermentation products
Wolinella Motile helical curved, or straight rod, single polar flagellum; either hydrogen or formate as electron
donors for reduction of fumarate to succinate; carbohydrates not fermented
Selenomonas Motile crescent-shaped cell, tufts of flagella on concave side; fermentation products propionate and
acetate
Anaerovibrio Motile curved cells, single polar flagellum, lipolytic; fermentation products propionate and acetate
Pectinatus Motile curved cells, lateral flagella aligned on concave side; fermentation products propionate and
acetate
There are systems that group microorganisms on the basis of Yet another trend and the one favored today, takes into
their increasing degrees of risk to humans, animals, and plants. consideration the similarities in homologous molecules which
Here, organisms are artificially, pragmatically classified into risk leads to the grouping of organisms according to their evolution-
groups according to their degree of pathogenicity or risk poten- ary relatedness. The classification process then circumscribes the
tial, and this system does not serve any other purpose emerging clusters by a wide range of characteristics originating
(> Table 9.1). from the genomic and epigenetic level. This genealogy-based
Another system focuses on the rapid and reliable identifica- classification system is the most comprehensive one in terms of
tion of bacteria for which knowledge about phylogenetic relat- overall understanding of the biology of the organisms, including
edness is not mandatory (> Table 9.2). In such diagnostic the evolution of core processes of genetics, biochemistry, and
system, used in the past, affiliation of an isolate to a genus and physiology. This approach, which was outlined two decades ago
species includes Gram-stain reaction, oxygen requirement and (Wayne et al. 1987), is now applied by the vast majority of
morphology, chemotaxonomy, numerical phenetic analyses, microbiologists. In the following text, the term ‘‘phylogeny’’ is
usage of rapid diagnostic kits (e.g., API and Vitek systems used when evolutionary relatedness among members of taxa or
[bioMerieux] and BIOLOG Microbial Identification Systems natural populations has been explored, via means of molecular
[Biolog Inc/]), and combinations of selected physiological tests. sequence data or DNA-DNA hybridization approaches.
Defining Taxonomic Ranks 9 231
. Table 9.3
Examples of changes in higher classification following phylogenetic-polyphasic taxonomic analyses as exemplified by the fate of some
species of Bacteroides (> Table 9.2)
Reclassification following
Traditional classification phylogenetic assessment Affiliation to higher taxon 2000 Affiliation to higher taxon 2010
Bacteroides amylophilus Ruminobacter amylophilus Phylum Proteobacteria Family
Class Gammaproteobacteria Succinivibrionaceae
Bacteroides bivius Prevotella bivia Phylum Bacteroidetes Family Prevotellaceae
Family Bacteroidaceae
Bacteroides cappilosus Pseudoflavonifractor cappilosus Not described Phylum Firmicutes
Family Clostridiaceae
Bacteroides endodontalis Porphyromonas endodontalis Phylum Bacteroidetes Family Porphyromonadaceae
Bacteroides distasonis Parabacteroides distasonis Not described Family Porphyromonadaceae
Bacteroides forsythus Tannerella forsythus Not described Phylum Bacteroidetes
Family Porphyromonadaceae
Bacteroides furcosus Anaerorhabdus furcosa Phylum Bacteroidetes Unchanged
Bacteroides gracilis Campylobacter gracilis Phylum Proteobacteria Family Unchanged
Campylobacteraceae
Bacteroides hypermegas Megamonas hypermegale Phylum Firmicutes Phylum Firmicutes
Family Veillonellaceae Family ‘‘Acidamniococcaceae’’
Bacteroides microfuscus Rikenella microfuscus Phylum Bacteroidetes Family Rikenellaceae
Bacteroides multiacidus Mitsuokella multiacida Phylum Bacteroidetes Phylum Firmicutes
Family Veillonellaceae
Bacteroides nodosus Dichelobacter nodosus Phylum Proteobacteria Unchanged
Class Gammaproteobacteria
Family Cardiobacteriaceae
Bacteroides ochraceus Capnocytophaga ochraceae Phylum Bacteroidetes Phylum Bacteroidetes
Family Flavobacteriaceae
Bacteroides splanchnicus Odoribacter splanchnicus Phylum Bacteroidetes Phylum Bacteroidetes
Family Porphyromonadaceae
Bacteroides praeacutus Tisseriella praeacuta Phylum Firmicutes Family Veillonellaceae
Bacteroides pseudosintes Dialister pneumosintes Phylum Firmicutes Phylum Firmicutes
Family ‘‘Acidamniococcaceae’’
Bacteroides putredinis Alistipes putredinis Phylum Bacteroidetes Phylum Bacteroidetes
Family Rikenellaceae
Bacteroides succinogenes Fibrobacter succinogenes Unassigned Phylum Fibrobacteres
Family Fibrobacteraceae
Bacteroides termitidis Sebaldella termitidis Phylum Fusobacterium Unchanged
Family Leptotrichiaceae
The fact that the genealogy is derived from gene sequence relatedness as the most reliable means for classification
similarities leads to the advantage of working with a reliable, (> Table 9.3). The reclassification of a species makes it necessary
objective, and stable basis for identification and classification. to redefine its properties. Above all, analysis of the genomic
The dramatic changes that occurred in the classification of relatedness of a species will provide information on its phyloge-
species of Bacteroides, lumped together in the past on the basis netic position, i.e., who its nearest neighbor(s) is. However, in
of a few superficial properties, is an excellent example of the shift many cases, the position will not provide information on other
toward a system that is primarily based on phylogenetic properties needed to make decisions whether this species can be
considered a species of a known genus or whether this species most of the characters used for systematics in the past were
forms the nucleus of a novel genus. These conclusions depend inappropriate for unraveling a phylogenetic framework.
upon the results of a targeted search for a wide array of pheno- During the beginning of the twentieth century, the number
typic and genomic properties. of determinable properties expanded dramatically, and conse-
This chapter introduces the importance of working with quently, the number of species increased. Early taxonomists
sequences of homologous genes and gene products as the basis usually worked in a clinical environment, leading to the descrip-
for providing an objective framework of the order at which tion and identification of mainly pathogenic species. At the
lineages of prokaryotic organisms evolved. It then describes the beginning of the twentieth century, new systems were proposed
(subjective) decision-making strategy of how bacteriologists on in which the accent shifted from morphology to physiology,
the basis of this phylogenetic framework define the ranks of metabolism, pigments, and pathogenicity (Migula 1900;
species and genera. Special emphasis is placed here on the Orla-Jensen 1909; Pringsheim 1923; Prévot 1938; Kluyver and
pragmatic definition of the species. It also deals with problems van Niel 1936; Stanier and van Niel 1941). In order to get a grasp
of delineating ranks above the genus level for which only a few on the wealth of information and to harmonize the different
common characters are available, and it discusses recent findings systems, a single unifying formal system of bacterial classifica-
that higher taxa of the same rank are often nonequivalent taxa. tion was established by Buchanan (1916, 1918) in accordance
with the systems of higher animals and plants. This system
provided the basis for Bergey’s Manual of Determinative Bacte-
Past Classification Attempts riology, which, in the many editions that followed over the years,
presented better than any other source the most useful references
Ranks or taxa have been introduced in the classification of for identification but, following the tradition, retained
biological specimens to facilitate communication among scien- a nomenclature that was believed to reflect phylogenetic rela-
tists and to order the living matter in a hierarchic way according tionships. Attempts to either construct a single formal classifi-
to morphological, physiological, ecological, and genomic fea- cation system or to work with several systems in parallel were
tures present in one rank but absent in the others. The basis of both criticized by Kluyver and van Niel (1936). These scientists
any system is the species and the genus, and according to the suggested that rather than searching for a natural system taxon-
binomial system (Linnaeus 1753), the description of a type omists should develop determinative keys to provide the easiest
species is not possible without describing a genus and vice possible identification of species and genera. However, as the
versa a genus cannot be described without a species. This is characters used for the establishment of the system were chosen
a clear and simple recommendation, and the simplicity explains according to the subjectivity of the taxonomist, it was admitted
why the binomial system is still in use for the naming of organ- that an empirical system was largely unmodifiable. Conse-
isms within the three primary domains of life (Woese 1987), the quently, the whole system was disrupted when novel characters
Archaea, the Bacteria, and the Eukarya (Woese et al. 1990). were taken as the basis for the establishment of a new classifica-
Considering the way description of biological material was tion system. The main advantage of the empirical system was
done prior to Linnaeus, Moreno (1997) states, ‘‘The wisdom of seen in its immediate practical utility, but it became obvious that
Linnaeus was not only to create a comprehensive classification even this advantage disappeared when the differential characters
system, but more importantly, a useful one.’’ The definition of were not mutually exclusive. Scientists of this era, lasting to the
a species has been debated extensively since the publication of end of the third quarter of last century, recognized the impor-
the key work On the Origin of Species (Darwin 1859). The debate tance of developing a natural classification system (Stanier and
concentrated on animals and plants but excluded the prokary- van Niel 1941) but attempts for doing so were not considered
otes mainly because of the lack of an evolutionary record. achievable. The question then remains why it was impossible for
Bacterial classification as a science began with the contribu- past generations of microbiologists to develop a phylogenetic
tion of the botanist Cohn (Cohan 2002, 2006) in the second half framework of prokaryotes? In hindsight, the answer is quite
of the nineteenth century. He was the first to raise the fundamen- easy: Without information about fundamental genetic informa-
tal question whether bacteria, like plants and animals, can be tion and without understanding the mechanisms of heredity and
arranged in species and genera, and he presented a classification the technical ability to unravel the structure of these units, early
scheme, composed of six genera, which was based on morpho- attempts were prone to failure.
logical criteria. However, he clearly pointed out that the restric-
tion to morphological properties is insufficient as he noticed that
similarity in shape does not exclude the bacteria to differ from Phylogeny Is Based on Homology
each other in physiological characters. Cohn regarded genera as
natural entities, but he described species as largely provisional. As Phylogenetic systematics is seeking congruency between the lines
judged from today’s view, early microbial systematists, having to of descents evolved through time and the supraspecific taxa
rely on superficial characters which were easy to observe and to described by taxonomists. Prerequisite for the description of
describe, were not in a position to fully acknowledge the a taxon of any rank in a phylogenetic system is the recognition
complexity of the bacterial cell. It took almost 100 years for that all members to be included originate from one ancestral form
bacteriologists to gather enough information to recognize that and that homologous traits evolved in the ancestral form are
found in their descendants. The question then remained which of greater the complexity of the elements at that level and the
the several thousand semantides present in a prokaryotic cell to smaller the parts of elements that have to be affected to bring
use for phylogenetic studies. The establishment of a system which about a significant change. Under favorable conditions of this
is set up to include all species requires the presence of phyloge- kind, recognition of many differences between two elements
netic markers that should be ubiquitously distributed, function- does not preclude the recognition of their similarity.’’ The cor-
ally equivalent, and homologous. These markers should be rectness of this hypothesis was justified by the impressive phy-
homologous apomorphic characters that evolved only once (syn- logenetic trees of gene and proteins sequences.
apomorphy) but not by convergence. Homology is the sharing by
two taxa of a property that is derived from the same or equivalent
property of the nearest common ancestor. Furthermore, the fossil A Word About Horizontal Gene Transfer
record, morphological complexity, and comparative anatomy, (HGT)
extremely useful properties of eukaryotes for determining homol-
ogies, are lacking in the morphological and developmental simple Genomic studies during the past decade have revealed that
prokaryotes, which lead to the consequence that a phylogenetic bacterial (and archaeal) genomes are much more dynamic and
classification system became only available after the theoretical diverse than previously anticipated. For instance, it is not
and methodological basis had been laid some 30 years ago. uncommon that strains of the same species would differ in up
One of the main intellectual breakthroughs concerning the to 30 % of the genes in their genomes (Konstantinidis and Tiedje
potential of unraveling the phylogeny of life was provided by 2005b). Horizontal (as opposed to vertical descent) gene transfer
Zuckerkandl and Pauling (1965). They recognized that contem- (HGT) and gene loss account for the majority of this immense
porary organisms are the products of historical events and that all genome plasticity and diversity (Lawrence 2002). It has been
cellular structures reflect their evolutionary history. These scien- argued that HGT might be so pervasive in the prokaryotic world
tists also commented that in the case of microorganisms, histor- that the prokaryotic tree of life is actually a web of life and that
ical documents of early evolutionary events can only be found at the bacterial phylogeny cannot be reconstructed faithfully based
the primary structure level of homologous informative molecules. on the sequence analysis of genes (Doolittle and Bapteste 2007).
Number and composition of sequence differences that exist HGT (mediated by homologous recombination) might also be
among proteins and genes coding for rRNA and proteins reflect so frequent and unbiased (neutral) that it can serve as the force
their phylogenies and consequently allow recognizing pairs or of population cohesion (Gevers et al. 2005), similar to the role of
groups of organisms which originated from a common ancestor. sex in sexual organisms (discuss later in this chapter). Nonethe-
The basis for unraveling relatedness is provided by sequence less, several other studies have shown that genes involved in
analysis of certain biological molecules, the semantides. Based central cellular processes such as DNA replication and protein
on their information content, three categories can be identified: translation machinery are subjected to HGT much less frequent
than metabolic genes because the transfer of the former genes
1. Semantides, i.e., genes or their transcripts (DNA [primary
typically confers no selective advantage to the recipient cell
semantides], RNA [secondary semantides], and proteins
(Gogarten and Townsend 2005; Beiko et al. 2005). Indeed,
[tertiary semantides]). Sequences of these molecules are
HGT of 16S rRNA genes has been documented for only a few
molecular chronometers, records of evolution, as they indi-
cases (e.g., Miller et al. 2005). As pointed out by van Berkum
rectly measure the time elapsed since their origin, and the
et al. (2003), deciding taxonomic relationship based solely on
comparative analysis of the primary structure provide
16S rRNA gene sequence divergence to reflect phylogenetic
a powerful approach to measure evolutionary relationships.
relatedness may be misleading; tree topologies based upon 16S
2. Episemantic molecules to be used in comparative studies
rRNA gene sequences of members of rhizobia differ from those
(e.g., peptidoglycan, isoprenoid quinones, polar lipids,
based upon 23S rRNA genes and internally transcribed space
membrane constituents) are synthesized under the control
region sequences, and the authors suggested gene conversion of
of tertiary semantides, and above all, it is the chemical
certain stretches within the 16S rRNA gene to be responsible for
composition of cell constituents that have received consid-
the discordant phylogenies. Accordingly, although the real phy-
erable attention (Schleifer and Kandler 1972). Episemantic
logeny of organisms is probably impossible to be accurately
molecules were not considered useful for deriving evolution-
reconstructed, even with complete genome sequences available,
ary conclusions because enzymes with different primary
an average ‘‘consensus’’ phylogeny may have been attained.
structures can lead to the synthesis of identical episemantic
or similar molecules in different organisms as long as the
active enzymatic sites are similar.
Main Phylogenetic Parameters for
3. Asemantic molecules are not produced by the organisms
Classification of Ranks Above the Genus
themselves and do not express any of the historic informa-
tion that organisms contain (e.g., exogeneously supplied
Level
vitamins, phosphate ions, oxygen, viruses).
The two main components in determining phylogenetic
Zuckerkandl and Pauling state that ‘‘at any level of integra- relationships among the prokaryotes – sequence analysis of
tion, the amount of history preserved will be the greater, the the semantides DNA, RNA, and proteins (Zuckerkandl and
Pauling 1965) and DNA-DNA hybridization techniques—were a major portion of the genome. In principle, the primary struc-
developed approximately at the same time, around the mid- ture of the most widely analyzed 16S rRNA gene can be regarded
1960s. Historically, the molecular approaches to evolution as representing a miniaturized approximation of the core
involved sequence analyses of proteins, such as cytochrome C, genome’s history, though, due to its size of only 1,540 bases,
fibrinopeptides, and ferrodoxins, as well as immunological with much less resolution power. It should be noted that rrn
approaches, such as immunodiffusion and microcomplement operons, consisting of genes coding for rRNAs and the
fixation. However, the latter methods as well as protein sequenc- intrageneric spacers, are in most organisms present in multiple
ing have lost significance when rapid sequencing techniques for copies per genome. While a single copy is often found in slow
DNA became available. Above all, the 16S rRNA gene is the most growing organisms, copy numbers of up to 14 may be found in
widely analyzed molecule as it has certain advantages, not pro- some organisms, e.g., Firmicutes. Nucleotide variations
vided by other rRNA molecules (5S rRNA is too short and 23S (microheterogeneity) often occur among the multiple gene cop-
rRNA is too large to be sequenced routinely) or genes coding for ies, in most cases in variable helices, which may cause problems
proteins (e.g., degeneration of the code makes it difficult to in deciphering sequences generated by the PCR technique. If
design universal PCR primers). The primary structure of the only a low percentage of copies are subjected to microheter-
16S rRNA gene is apt to cover an enormous geological time span, ogeneity, a PCR-based analysis run will detect such events as
most likely covering almost the whole time span since the origin noise. Thus, the selection of cloned genes may have an influence
of prokaryotes (3.8 Gy ago). The molecule has been selected as on the positioning of an organisms, especially in those taxa in
suitable phylogenetic marker mainly because it is ubiquitously which species share high sequence similarities. With the advent
distributed and orthologous among all forms of life, functionally of fully sequenced genomes, the extent of microheterogeneity
constant, genetically stable, of an appropriate size, and possesses will be unraveled, allowing even the elucidation of possible HGT
independently evolving domains within the molecule. events of rRNA genes.
Sequence analysis of rRNA genes and certain other genes has The branching patterns based upon 16S rRNA and 16S
become a rapid standard technique, and the sequences generated rDNA sequences have sometimes surprised biologists in the
have a very low error rate. The restrictions in the use as past mainly because it suggested that the characters traditionally
a phylogenetic marker are due to certain intrinsic properties of used to cluster organisms show restricted phylogenetic signifi-
the molecule: In comparison to the billions of years that have cance. Prominent early examples of phenotypic characteristics
passed since its origin, the number of informative positions, i.e., shown to have failed to circumscribe higher taxa in the past
positions that changed over time in proportion to the time are the chemical composition of peptidoglycan (Kandler and
elapsed, within the molecular sequences is limited. As H. König 1985; Stackebrandt 1985; Schleifer et al. 1990), aerobic
a consequence, the majority of evolutionary change that did metabolism (Fox et al. 1980; Seewaldt et al. 1982), spore forma-
occur at the organism level will remain undetected at the tion (Ash et al. 1991; Collins et al. 1994; Stackebrandt and
sequence level of rRNA. Another restriction is that in most Rainey 1997), biosynthetic pathways (Balch et al. 1979;
prokaryotic organisms, the multigenic rRNA operon is present Fowler et al. 1986; Stackebrandt et al. 1988), and photosynthesis
in several copies in the genome ranging from 2 to 14 (Farelly (Gibson et al. 1985; Woese et al. 1985; Stackebrandt et al. 1988;
et al. 1995). PCR amplification will mask possibly occurring Imhoff et al. 1998a, b). Today, the use of 16S rRNA gene
intracistronic microheterogeneities which consequently may sequences in bacteriology is so widely accepted that students
obscure elucidation of the fine differences of relatedness of consider sequencing of this molecule a classical approach. The
closely related organisms. Sequence analysis of individually sequencing and analysis strategies and the main results for the
cloned operons can unravel these heterogeneities, but these few evolution, phylogeny, classification, and identification are now
changes are typically regarded ‘‘noise,’’ hence without phyloge- textbook knowledge. The need to include into phylogenetic
netic implications in most instances. Higher order structures of studies genes with a higher resolution power than that of the
rRNA molecules facilitate sequence alignment, which today can 16S rRNA gene arose once the conserved primary structure of
be analyzed by a wide range of publicly available databases (e.g., rrn genes excluded the fine resolution of closely related species
RDP II, ARB, Greengenes) and algorithms (e.g., maximum and genera (e.g., see certain members of the genus Aeromonas
likelihood, parsimony, neighbor-joining, Bayesian methods), [Alperi et al. 2010]). On the other hand, protein sequences were
facilitating checks the robustness of resulting branching pat- evaluate to provide better resolution of deep branches, were 16S
terns. Numerous factors have been shown to influence the rRNA genes sequence analysis resulted in a forklike separation of
branching pattern (topology), which is a dynamic construct taxa, implying a sudden evolutionary burst of lineages (e.g., Fox
and will change with any new sequence included or region et al. 1980). Protein genes analyzed for these purposes were,
selected for analysis. Nevertheless, the topology of trees which among others, the b-subunit of ATP synthases (Amann et al.
are generated on the basis of genes subjected to the same fate in 1988), HSP70 (dnak) genes (Gupta and Golding 1993), DNA-
evolution is rather stable and robust constructs as demonstrated dependent RNA polymerase (Klenk and Zillig 1994), or groEL
by results of comparative analyses of other conservative mole- (chaparonin) genes (Douglas 1998). These sequences, however,
cules responsible for central cellular functions. Thus, trees based did not always agree in their phylogenetic assessment: While one
on rRNAs (early studies) and rRNA genes not only reflect the set of genes strongly supported the results of 16S rRNA phylog-
evolution of these molecules but most likely the evolution of enies, other genes represented a different picture of their own
evolution. The increasing availability of draft and complete . Table 9.4

genome sequences will facilitate the search for appropriate Number of taxonomic ranks of prokaryotes each compiled at the
genes, and especially in the absence of a wide range of universal end of four decadesa
markers, for sets of taxon-specific genes. Lately, with the avail-
Rank 1980 1990 2000 2010
ability of thousands of genomes of Archaea, Bacteria, and
Eukarya, relatedness among organisms is elevated from the Domain – 2 2 2
level of individual genes to sets of core genes. By and large, Phylum/division – – 14 34
independent approaches for tree construction from completely Class 7 12 14 70
sequenced genomes (Wolf et al. 2001; Cicarelli et al. 2006) Order 21 30 39 106
support the main 16S rDNA gene-based phylogenetic lineages
Family 68 96 151 220
of prokaryotes, though the order of branches may differ in some
Genus 300 545 988 1,676
cases.
The criteria for describing higher ranks do not follow Species 1,841 3,233 5,315 8,659
a coherent strategy as neither rules nor minimal requirements Subspecies 132 138 116 115
for their description are laid down. Ranks governed by the Code a
According to Euzeby (http://www.bacterio.cict.fr/)
of Nomenclature are regularly validated by listing them in noti-
fication tables published in the International Journal of System-
atic and Evolutionary Microbiology (http://ijs.sgmjournals.org/).
It should also be noted that taxonomic categories above the rank 10000
of class (classis) are not covered by the rules of the Bacteriolog-
ical Code though the names of several phyla/divisions have been
included in validations lists. While the phylum (division)
Firmicutes (Gibbons and Murray 1978) was created in the pre-
nucleic acid sequencing era, almost all other phyla were 1000
described later, once the outline of the phylogenetic tree became
species
visible. Higher ranks are mainly created on the basis of phylo-
Logarithmic scale
genetic coherency and distinctness from neighboring taxa based, genus

primarily, on the 16S rRNA gene phylogeny. Recent overall views family
of major higher ranks were provided by Gibbons and Murray 100
(1978) and by Cavalier-Smith (2002), but, though validly order
published, the names introduced did not always find their way class
into the vocabulary of taxonomists. While authors usually con-
centrate on the description of species, genera, and families, phylum
higher ranks have been considered less important. This changed 10
by the release of the so far four published volumes of the 2nd
edition of Bergey’s Manual of Systematic Bacteriology (from 2001
to 2012). The increase of the number of phyla, class, and orders
over the past decade (> Table 9.4, > Fig. 9.1) is mainly due to the
desire of the editors to provide a most complete hierarchic 1
1980 1990 2000 2010
system. As a consequence of the availability of such framework,
authors are encouraged to include higher ranks for novel taxa . Fig. 9.1
whenever appropriate. Logarithmic display of the evolution of descriptions of higher
Many ranks above the order were mainly defined for organ- ranks over the past 30 years
isms having 16S rRNA gene sequences related to those of mem-
bers of the type order. This strategy has been outlined in depth
by the editors of Bergey’s Manual (http://www.taxonomi- Proteobacteria, the authors refrain from indicating any pheno-
coutline.org/). The attempt to support these ranks by phenotypic traits because of the enormous phenotypic heterogeneity
typic properties, often result in very broad descriptions, which of its members.
would not allow proper classification without the aid of molec- To summarize, the ranks above the species rank are defined
ular sequences. As phenotypic properties are deduced from mainly on the basis of the 16S rRNA gene phylogeny, but no
shared properties of most or all taxa defined by molecular definite standards exist for the genetic relatedness among the
analyses to belong to the same lineage, the circumscription is organisms grouped at the same rank. Konstantinidis and Tiedje
a reflection of the phylogenetic depth and phenotypic heteroge- (2005a) published an independent assessment of the higher taxa
neity of its members. > Table 9.5 gives some examples for phyla by comparing the shared gene content and genetic relatedness
of the domain Bacteria. In most of these cases, the phylum among 175 fully sequenced bacterial genomes. Genetic related-
description is based on that of the order. In the case of ness was assessed using the average amino acid identity (AAI) of
. Table 9.5 inconsistencies of the taxonomy. For instance, the genetic relat-
Definition of some higher ranks, according to volumes 1–3 of edness among the Prochlorococcus marinus or the Buchnera
Bergey’s Manual of Systematic Bacteriology, 2nd ed (2001–2009) aphidicola genomes is far too low, compared with the remaining
data set, to justify their inclusion in the same species. Similarly,
Key properties as indicated in the phylum
the Treponema and Leptospira genomes are assigned to the same
Phylum description
order due to their common spirochete-like morphology, even
Aquificae/ Rod-shaped, moderate thermophilic to though they are as divergent from each other as some organisms
Aquificales hyperthermophilic. Includes microaerophilic of different classes, even phyla, are. Such cases are apparently due
Reysenbach chemolithotrophic hydrogen oxidizers; most to historic reasons and old taxonomic assignments, which need
(2001) members are motile
to be adjusted in order for a system that is (more) predictive of
Deinococci A phylogenetic lineage that contains species the genetic relatedness of the grouped taxa can emerge
Garrity and Holt of the genus Deinococcus as a coherent group.
(Konstantinidis and Tiedje 2007). Taxonomic changes will also
(2001a) A set of 16SRNA-rDNA signature nucleotides
be necessary in the future in order to adjust the single-gene 16S
defines the order
rRNA gene-based hierarchy of ranks to the emerging genome-
Nitrospirae Defined mainly on phylogenetic grounds.
based taxonomy.
Garrity and Holt Gram-negative, curved, vibroid, or spiral
(2001b) shaped. Metabolically diverse, most members
are aerobic chemolithotrophs including
nitrifiers, dissimilatory sulfate reducers, and Main Phylogenetic Parameters for
magnetotactic forms Classification of Genera and Species
Chloroflexi Gram-negative, filamentous, exhibiting
Garrity and Holt gliding motility. Peptidoglycan contains Until the late 1960s of the last century, methods for describing
(2001c) L-ornithine as the diamino acid. species and genera were based upon phenotypic, epigenetic
Lipopolysacchride-containing outer properties (> Table 9.6). This is exemplified for the characteri-
membrane not present zation of Vibrio cholerae over the past 120 years. Looking at the
Chlorobi Defined on the basis of phylogenetic analyis description given by Lehmann and Neumann (1896), morpho-
Iino et al. (2010) of 16S rRNA gene sequences. Gram-negative logical properties of cells and colonies, growth conditions, sim-
bacteria that grow under strictly anaerobic ple physiological and chemical reactions, as well as range of
conditions susceptible animals dominated. This description remained
Proteobacteria Defined on the basis of phylogenetic analyis more or less unchanged during seven editions of Bergey’s Man-
Garrity et al. (2005) of 16S rRNA gene sequences. Gram-negative ual of Determinative Bacteriology but was supplemented with the
bacteria classified in six classes characterization of antigens and typing of proteins and cell wall
Firmicutes/ Rigid, muramic acid-containing cell wall. Most components, which were introduced in the 1950s. The major
Bacillales members Gram-positive, some Gram- increase in knowledge occurred between 1960 and 2000, by the
Gibbons and negative staining. Phenotypically diverse: introduction of numerical taxonomic methods among other
Murray (1978) spherical, straight curved, helical rods or methods, providing huge databases of substrate utilization and
filaments, flagella may be present. With or
reactions toward changing physicochemical conditions (manual
without heat-resistant endospores. Aerobes,
version of today’s BIOLOG Phenotype MicroArrays™), and by
facultative or strict anaerobe. Thermophiles
and holophiles present. Most of them are chemotaxonomic analyses of cellular constituents such as pep-
chemoorganotrophs, a few are anoxygenic tidoglycan, cellular fatty acids, phospholipids, polyamins, and
phothohetrotrophs. The mol% G + C content isoprenoid quinones. Ecological parameters such as dormant
of DNA is generally >50 stages and epidemiological traits were likewise considered as
were results of morphological and genomic makeup and lytic
specificity of bacteriophages. Although MLST and MLSA studies
on vibrios were already available at the release of Bergey’s Man-
the shared genes between two strains (orthologs and paralogs), ual in 2005, these data were not included in the description of
a genome-derived robust measure of genetic relatedness. Plot- V. cholerae. The main progress concerning the elucidation of
ting 16S rRNA gene identity against AAI showed that, in general, relationships between strains and species and the delineation of
there is a gradient of genetic cohesiveness, i.e., higher ranks tend species and genera started with the introduction of DNA-DNA
to group more diverse genomes than lower ranks. However, the (DDH) and DNA-rRNA hybridization techniques (see below).
analysis also revealed that adjacent ranks (e.g., phylum vs. class) While the latter method was soon replaced by sequence analyses
show, on average, 30.7 % overlap, meaning that 30.7 % of the of the 16S rRNA gene, the DDH approach is still considered
pairs of organisms showing the exact same genetic and gene- a fundamental element of any species description. Today the
content relatedness to each other belong to different ranks and advanced MLST and MLSA approaches allow a much deeper
hence, the current system is of limited predictive power in this insight into the population structure of species with far-reaching
respect. The overlap between nonadjacent ranks (e.g., phylum conclusions on the evolution, epidemiology, systematics, and
vs. family) is generally limited and attributable to clear ecology.
. Table 9.6
Transfer of increased taxonomic knowledge into the description of the species Vibrio cholerae (V. comma)
Bergey’s Manual of Determinative Bergey’s Manual of Systematic Bacteriology,

Bakteriologische Diagnostik, Bacteriology, 7. Aufl. (1957), 2. Aufl. (2005),
Lehmann and Neumann (1896), V. comma (Schroeter 1886) (Winslow V. cholerae Pacini 1854, 411AL
Vibrio cholerae V. cholerae (Koch) Buchner et al. 1920) (Farmer et al. 2005)
Culture Morphology, luminescence, size, As 1899, plus tolerance against acidic As 1899
description colony form and color, motility; and alkaline conditions
life span in patient material;
resistance
Growth Gelatin liquification, growth in As 1866, plus starch hydrolysis, Litmus As 1899, plus Na+-requirement
requirements agar, bouillon serum, blood, milk, milk reaction
potato; relation toward O2
Chemical Pigments, gas and acid As 1899, plus formation of NO2 from As 1899, plus broad substrate utilization
reactions production, enzymes, formation NO3 spectrum; fermentation end products,
of H2S, indol; toxins transport systems; degradation pathways
Pathogenicity, Mouse, rat, guinea pig, rabbit Not covered Antibiotic resistance; population analyses;
immunity pandemic analysis
Habitat Patient, healthy person, As 1899 Life cycle, resting stages (‘‘viable but
environment nonculturable state’’); ecology
Chemical – Typing of proteins, phospholipids- and As 1957, plus cytochrome and siderophore
structures polysaccharides, O- and H-antigens analyses; antigen structures; mol% DNA G + C
Genomic – – DNA restriction, genome size chromosome
properties number, plasmids, genome analyses, gene
transfer, bacteriophages, phage typing;
bacteriocins
Relationships – – DNA-DNA- and rRNA-DNA-hybridization, 16S
rRNA-sequences, immunological
relationships of enzymes and other proteins
Classification Is a Dynamic Process (Revised new insights. One problem still remains: The original advantage
from Stackebrandt 1992) of the tree—its objectivity (insofar as this is possible)—is weak-
ened by subjective clustering of organisms due to the differences
Microbiologists are aware that the available phylogenetic- in the emphasis taxonomists place on phenotypic characters. As
branching patterns reflect the actual situation in nature only in previous decades, the one system (or parts thereof) with
very incompletely. Phylogenetic reconstructions are based on highest practicability will succeed against competing systems
inferred homologies but, unless witnessed by the evolutionary with less persuasive arguments (see, e.g., the system devised by
history of taxa, i.e., by fossil record data, the latter cannot be Cavalier Smith (2002) vs. the one used by the editors of Bergey’s
considered definitive (Rothschild et al. 1986). The isolated posi- Manual). The ultimate goal is to establish a hierarchic system
tion of certain taxa, well defined by genotype and phenotype where all taxa show phylogenetic coherency and, at least for
today, may disappear tomorrow with more organisms investi- ranks below the family level, a great deal of phenotypic coher-
gated. Thus, whenever new information requires corrections, ency as well. On the other hand, sufficient differences need to be
either within established taxa or in neighboring groups, flexibil- known to distinguish taxa from each other by stable and easily
ity is called for and changes have to be made for the benefit of determinable characters. While phylogenetic coherency is easy
a better agreement between phylogeny and taxonomy. The main to define, the term ‘‘phenotypic coherency’’ varies according to
advantages of the phylogenetic system lie in its stability: only the the taxonomist. Again, practical considerations must come
rank (either vertical or horizontal) within the hierarchic struc- before petty splitting or lumping.
ture, but not its place, will be changed—as happened in past Prerequisite is a profound knowledge about the phylogenetic
systems (a comparison of Bergeys Manual of Determinative clustering of members of the higher taxon in question. An
Bacteriology from the first through eighth edition is instructive). optimal survey would work with coded, unnamed organisms
Even the most convincing tree is always in a dynamic state; to judge the resulting branching pattern without prejudice. The
this forces taxonomists to stay flexible in order to adjust not only study should include the type strain of all type species. The
established ranks but also to modify nomenclature according to resulting pattern depicts the relative branching order that,
depending on the size of the database and the selection of (De Ley 1970; Grimont et al. 1980; Huss et al. 1983; Baumann
reference organisms, will immediately yield information about et al. 1983; De Ley 1970; Schleifer and Stackebrandt 1983).
the phylogenetic homogeneity of a group of isolates. In the Comparative DNA-DNA hybridization studies on the same strains
second step, the branching pattern is superimposed on pheno- using different techniques were in good congruence (see Schleifer
typic data with the goal to delineate clusters of organisms that and Stackebrandt 1983 for examples). Some novel techniques
are phylogenetic coherent and easy to recognize by phenotypic and variations of established methods have been introduced
characters. This is not only prerequisite for identification, but such as hybridization in microdilution wells (Ezaki et al. 1989;
also to decide which of the possibly several available branching Hara et al. 1991; Kaznowski 1995), the random-primed labeling
patterns is the most likely one to reflect phylogeny most closely. and signal amplification system (Amersham Life Science), or
It should be mentioned in this context that in the presence of detection of double-stranded, DIG-labeled DNA with anti-DIG
varying evolutionary rates, species with the highest nucleic acid antibodies conjugated with alkaline phosphatase (Lind and
sequence similarity are not necessarily the most closely related Ursing 1986; Ziemke et al. 1998). Evaluation of these novel
ones; while programs that optimize branch length take care of methods with the established ones (S1-, renaturation- and filter
this problem, numerical phenetic analyses would in fact cluster methods) has not been performed except for the microplate
the respective species as neighbors. The combination of taxon- technique (Ezaki et al. 1989) and the renaturation technique,
describing characters will, in most cases, not be predictable, and where good correlations were observed (Goris et al. 1999).
the search will have to be extended to features not considered of Relationships are usually expressed in terms of DNA simi-
taxonomic significance previously. Still many phylogenetic larities. It should be noted that due to the failure to unravel the
coherent taxa exist for which appropriate characters have not underlying processes of renaturation, the expression ‘‘DNA sim-
been found yet. ilarity’’ but not ‘‘DNA homology’’ should be used in connection
with DNA reassociation techniques. Wayne and colleagues
(Wayne et al. 1987) recommended the use of a second parame-
DNA-DNA Hybridization ter, the Tm(e) (e for eluted) value, especially in those cases where,
under optimal hybridization conditions, fine resolution in DNA
The DNA-DNA hybridization (DDH) method has received similarities is needed. The lower the degree of reassociation
much attention in the literature because of its outstanding role specificity (relaxed conditions), the higher the degree of mis-
in the classification of lower ranks. While several publications matches. When the reassociated DNA strands are remelted, the
highlight the need, but also the restrictions, of this method as an melting temperature Tm indicates the degree of thermal insta-
integral part of the polyphasic taxonomy (e.g., Stackebrandt and bility caused by incomplete reassociation between heterologous
Goebel 1994; Rosselló-Mora and Amann 2001; Rosselló-Móra DNA fragments. As compared to values of homologous reac-
2006; Stackebrandt and Ebers 2006), others challenge the poly- tions, the Tm(e) value of the heterologous reaction is thus an
phasic approach including DDH as it is not encompassed by indication of the degree of mispaired bases in the hybrid formed
a theoretical ‘‘species concept’’ (Cohan 2002; Gevers et al. 2005; after reassociation. However, as an inverse linear correlation
Achtman and Wagner 2008; Ward 1998). Despite the shortcom- exists between Tm(e) and DNA similarity, which makes determi-
ings, one should, however, not forget the enormous influence nation of both parameters somewhat redundant (Grimont et al.
DDH had on stabilizing bacterial classification of species and 1980; Baumann et al. 1983), determination of Tm(e) values is
genera. This method was the first molecular approach used usually not included in DNA-DNA reassociation studies.
routinely for measuring degrees of relatedness and the first Reproducibility, together with small sampling error (Sneath
phylogenetic one to be generally accepted for improving bacte- 1989), is an obvious advantage of DDH. Its limitation is the
rial classification. Despite the rapid deposition of draft or fully accessibility of equipment, i.e., a thermo-controlled spectropho-
sequenced genomes, DDH is still the most rapid and inexpensive tometer, the accessibility of DNA of sufficient quantity and purity,
of all phylogenetic methods for measuring the degree of nucle- e.g., for many archaea and lithoautotrophic bacteria and
otide similarity of the entire genome. The main drawback of this uncultivated organisms. The limited resolution power has been
approach, though, is the lack of information on which genes of recognized from the very first experiments. It has been calculated
the two strains compared contribute to the similarity and which that for reassociation under optimal hybridization conditions
genes do not, because of not being shared between the strains or (25 C below the Tm of the DNA), the two DNA strands must
showing lower sequence similarities. DNA concentration, frag- exhibit at least 80 % sequence complementarity. Depending on
ment size, temperature, DNA mol% G + C content, concentra- the sequence similarity of the reassociating single strands,
tion of salt, and denaturing agent influence the hybridization a difference of about 20 % is then spread between 0 % (no
process. Though unsatisfactory in the lack of transparency, the hybridization) and 100 % (as defined by maximal reassociation
simple similarity values obtained offer an advantage over those obtained with homologous DNA strands). It is therefore obvious
techniques that involve the comparison of individual genes or that a given DNA similarity value does not reflect the actual degree
gene products only. Several hybridization techniques exist that of sequence similarity at the level of the primary structure of DNA.
have been thoroughly tested to determine the influence of var- As measured with experimentally introduced mispairings,
ious experimental parameters, and all have been compared to thermal stabilities have been estimated to decrease from 1 % to
each other with respect to reproducibility and limitations 2.2 % for each percent mispairing (Bautz and Bautz 1964;
25
(52)
E. coli K12 vs. HS (ANI = ~99%)
E. coli K12 vs. E. albertii (ANI = ~90%)
E. coli K12 vs. Salmonella enterica LT2 (ANI = ~80%)
20
Percent of total genes in the genome

E. coli K12 vs. Yersinia pestis KIM (ANI = ~71%)
15 Genome average
(ANI = ~80%)
Highly conserved
Hypervariable genes
rRNA operon
Membrane proteins
HGT genes
Hypotheticals
10 Transposases
0
60 65 70 75 80 85 90 95 100
Nucleotide Identity to E. coli K12 ortholog (%)
. Fig. 9.2
ANI values among selected Enterobacteriaceae genomes. The figure shows the distribution of the nucleotide identities (x-axis) among
orthologous genes shared between E. coli K12 and four other Enterobacteriaceae genomes (y-axis), which show increasing genetic
relatedness to strain K12 (see figure key). Note that the gene sequence identities represent a, more or less, normal distribution around
the genome average value (ANI) for each pair of genomes, the distribution has a smaller standard deviation for more closely related
genomes, and the distribution starts deviating from normal for divergent genomes (Y. pestis), which is consistent with the fact that the
ANI approach is not robust for genomes showing less than about 80 % ANI. The genes that typically deviate the most from the genome
average in terms of their degree of sequence conservation (outliers) are denoted in the E. coli K12 versus Salmonella enterica LT2
comparison (solid squares). The patterns shown apply to other prokaryotic families and genera
Britten and Kohne 1968; Ullman and McCarthy 1973). Although independently of the genome size or morphological properties
these experiments have been performed on short stretches and (e.g., Gram-positive vs. Gram-negative) of the organisms com-
not on complete genomes, one can nevertheless argue that pared. These results are also comparable to those obtained
organisms that share 70 % DNA similarities share at least 96 % earlier based on short stretches of DNA (Johnson 1973). Thus,
DNA sequence identity (Johnson 1973). organisms that share higher than 95 % ANI should be expected
to belong to the same species accordingly to current recommen-
dations for species delineation. Similarly, it was found that the
Translating Traditional Standards to Portable 95 % ANI value corresponds to 98.5 % 16S rRNA gene sequence
Sequence Information: The ANI Approach identity; hence, organisms that show less than 98.5 % 16S rRNA
gene sequence identity to all characterized type strains should be
Several attempts to replace traditional cumbersome classifica- expected to represent a new species (Konstantinidis and Tiedje
tion techniques such as the DNA-DNA hybridization (DDH) 2007). The latter results are also consistent with those reported
with more user-friendly approaches have been made recently. In by Stackebrandt and Ebers (2006) based on comparisons
perhaps the most promising of these studies, Goris and col- between DDH and 16S rRNA gene sequence identity.
leagues proposed the genome-aggregate average nucleotide Similar to DDH, the resolution level of ANI is restricted to
identity (ANI) as a means of replacing DDH and translating closely related organisms, showing between 80 % and 100 %
DDH values to sequence identity information. ANI is the aver- ANI among each other. > Figure 9.2 represents an example
age nucleotide identity of all homologous genes shared between for some members of the family Enterobacteriaceae showing
two genomes, and the work of Goris and colleagues showed that different degrees of relationships. This is primarily due to the
it is the genome-derived parameter that best correlates with fact that multiple substitutions have frequently occurred in
DDH values, among several parameters evaluated (e.g., G + C homologous nucleotide positions that differ between genomes
% content, number of shared genes). In particular, it was found related at about the 80 % ANI level or lower. Such substitutions
that the 70 % DDH standard corresponds tightly to 95 % ANI, are not taken into account in the ANI calculation; thus, lower
than 80 % ANI values correspond to increasingly longer diver- food industry as the availability of an automated system
gent times between the genomes compared (Konstantinidis et al. (RiboPrinter® System, Dupont) and an extensive database of
2006). Further, no DNA fragments that showed lower than 80 % relevant strains facilitates handling and analyses. Other
nucleotide identity were found to cross-hybridize during DDH methods, such as analyses of multiple locus VNTR (variable
experiments (Goris et al. 2007); thus, this level of sequence number of tandem repeat) (MLVA), random amplified poly-
identity appears to represent the low boundary of robust reso- morphic DNA (RAPD), repetitive extragenic palindromic-PCR
lution for both the ANI and the DDH methods. Below the 80 % (REP-PCR), enterobacterial repetitive intergenic consensus
nucleotide level, the average amino acid identity (AAI) should sequences (ERIC-PCR), amplified ribosomal DNA-restriction
be used instead. To date, no cases have been reported where ANI analysis (ARDRA), and rRNA intergenic spacer analysis
fails to precisely represent the genetic relatedness among strains (RISA), are hardly portable and are mainly used as screening
of the same or closely related species. Hence, this approach holds tools to narrow down a large strain collection to subgroups with
great potential for prokaryotic taxonomy [see for instance different genotypes. As described by Christensen et al. (2007),
(Rossello-Mora 2005; Lilburn et al. 2006; Staley 2009)] especially ‘‘. . .however, all methods [the ones just mentioned] might
since obtaining the genomic sequence of a microorganism potentially be of use in testing the diversity of isolates. REP-
becomes an increasingly easier task nowadays. PCR, ERIC-PCR and BOX-PCR are only suitable for assignment
A simple way to calculate ANI between any two complete to a certain species based on obvious similarities in the banding
genome sequences is to identify, using reciprocal best match patterns. On the other hand, banding patterns without any
blast searches (nucleotide level; blastn), the conserved gene similarity do not necessarily demonstrate that two strains are
sequences between the genomes and calculate their average members of different species.’’ This notion correlates with the
nucleotide identity based on the identities reported by the possibility that discrete centers of genetic diversity exist within
blast algorithm. It is recommended that users run blastn with strains defining a species. Comparison of methods have been
default settings (tested in blast versions 2.2.17 to 2.2.25) as long published by, e.g., Scheldeman et al. (2004), Rademaker et al.
as the genomes compared show between 80 % and 100 % ANI. If (2000), and Rodas et al. (2005). As compared to the more than
instead of gene sequences, one uses 1-Kb long nonoverlapping 600 annually named species, only a few authors included in their
consecutive fragments of the genome sequence, which simulate description studies more than a single typing method, e.g.,
well the sheered DNA fragments produced during the most subspecies of Aeromonas hydrophila (Huys et al. 2002). This is
popular DDH protocols, the same ANI values are essentially perhaps not surprising, as more than 70 % of novel species
obtained. The ANI values can also be robustly estimated based descriptions embrace a single strain, the type strain, only
on draft genomes, i.e., genomes sequenced at 10X coverage or (Stackebrandt 2010).
better, as this has been implemented, for instance, in the freely The introduction of matrix-assisted laser desorption/ioniza-
available software JSpecies (Richter and Rossello-Mora 2009). tion time-of-flight mass spectrometry (MALDI-TOF MS) into
the routine microbiology laboratory at the end of the 1990s has
been a breakthrough in the rapid characterization of bacteria at
The Use of Typing Methods and MALDI TOF the strain, species, and genus levels (van Baar 2000; Lay 2001;
Shah et al. 2002). As little as about 5 103 cells are necessary for
Typing methods are routinely used for the characterization of reliable MALDI-TOF analysis (Hsieh et al. 2008). The remark-
strains at the species and subspecies level. The literature on these able reproducibility of the MALDI-TOF approach is due to the
approaches is vast, and key publications were compiled by fact that many of the individual single charged proteins of size
Gürtler and Mayall (2001). The use of comparative one- 2,000–20,000 m/z (mass/charge; daltons) present in high abun-
dimensional polyacrylamide gel electrophoresis of whole- dance in the cell include many ribosomal proteins. Being part of
organism protein pattern has been used widely in species the cellular translational machinery, MALDI protein finger-
descriptions in the 1990s, but nowadays, after the emergence prints are therefore not significantly influenced by variability
of a wide array of molecular DNA- and rRNA-based techniques, in environmental or growth conditions (there are some notable
protein patterns are rarely included in descriptions. The exceptions such as polymers formed during prolonged cold
methods for generating one-dimensional patterns of proteins storage or presence of spores).
and nucleic acids need to be highly standardized to allow intra- As cell extraction procedures and machine-specific settings
and interlaboratory comparisons. Among the methods, it is will have an influence on the recorded mass spectral profiles,
especially the macrorestriction of DNA after pulsed field gel their direct comparison with existing databases (e.g.,
electrophoresis (PFGE) that gives highly standardized patterns. BioTyperTM [Bruker Daltonics] or Saramis [bioMérieux]) is
Though not commonly used in bacterial classification, PFGE not immediately straightforward. However, by using the same
was commonly considered a gold standard in epidemiological type of mass spectrometer and keeping growth and extraction
studies of pathogenic organisms before the introduction of procedures constant, this method is unchallenged for the pur-
MLST and MLSA techniques. Two other methods with similar pose of identification and authentication in terms of preparation
highly standardized protocols are AFLP (amplified fragment- effort required, speed (30 min), and identification accuracy. The
length polymorphism PCR) and ribotyping. The latter method identification of strain-specific protein sequences by subsequent
is predominantly used in the medical environment and by the mass spectroscopic methods is only a question of time.
Firstly, the spectrum and number of fully sequences genomes are appears to be genetically much more homogeneous than other
expanding enormously, and secondly, the necessary analytical species of the genus (Maiden and Dingle 2008).
equipment will be easier to access through increased -omics Comparison of different DDH methods used in the deter-
studies. mination of relatedness has shown excellent agreement when
The question remains, however, how the spectra can be used closely related organisms are compared. However, when param-
in the classification process. The presence/absence of peaks of eters from more distantly related organisms are used, it has been
a defined mass between any pair of organisms is conventionally shown difficult to merge data from different techniques (Huss
transformed into a similarity matrix and then into a graphic et al. 1983; Grimont et al. 1980). The borderline of 70 % simi-
display of relationship, such as a dendrogram (Stackebrandt larity obtained under optimal hybridization conditions is
et al. 2005; Cousin et al. 2008). The identification scores given recommended for species differentiation (Wayne et al. 1987)
by the Biotyper software defines 3 categories such as ‘‘not valid’’ because a close agreement between phenotypic and genetic
for a score below 1.7, ‘‘valid at the genus level’’ for scores similarities was demonstrated once the DNA homology reached
between 1.7 and 2, and ‘‘valid for the species level’’ for values the 70 % level in numerous studies. Values from 30 % to 70 %
above 2.0. Several studies have used these threshold values for reflect a moderate degree of relatedness while values become
the identification of medical strains. With a minor failure rate, increasingly unreliable once they fall below the 30 % level, under
all studies, including 158 staphylococcal strains of six species which conditions taxonomic conclusions should be avoided.
from prosthetic joint infection (Harris et al. 2010), 602 Staphy- One has, however, to consider that the original recommendation
lococcus aureus strains, and 412 strains of 20 different non- for species delineation was derived mainly from the experience
S. aureus species of a medical culture collection (Szabados made with numerous strains of enterobacterial species
et al. 2010), as well as 304 g-negative stool isolates (He et al. (Steigerwalt et al. 1976; Brenner 1991). Thus, transferring the
2010), were correctly classified to the species level. Strains of results found for a phylogenetically very shallow group of mainly
Shigella species and EHEC isolates could not be distinguished eukaryote-associated organisms to the ancient and enormously
from typical E. coli strains, which is not surprising considering diverse structure of the two prokaryotic domains is grossly
the genomic closeness of these organisms. In the study of underestimating the different mechanisms as well as the mode
Prod’hom et al. (2011), including 122 strains from blood culture and tempo at which organisms develop. One has to remember
pellets, 79 strains were correctly identified, while 21 % of the that the delineation value is an artificial value used to structure
strains gave scores below 1.7; half of the latter were phenotypi- the bacterial world at the level of species in a coherent way.
cally identified as Staphylococcus epidermidis. The inability of the Correlation blots determined in the early 1990s for the
BioTyper to correctly identify strains points toward one of the parameters of the two most widely used approaches in prokary-
main problems of any identification databases, which can iden- otic phylogeny, the DDH and 16S rRNA gene sequence identity,
tify only as correctly as the deposited patterns allow. justified the continuous use of the DNA-DNA reassociation
technique. In case these plots would have shown a linear rela-
tionship between intraspecies DNA similarities above 70 % and
Correlation of Individual Phylogenetic 16S rDNA sequence similarities above 97.0 %, the DNA hybrid-
Parameters ization method would have disappeared over night. However,
the situation was different: As the 16S rRNA gene is
It is generally accepted that if two organisms have highly similar a miniaturized mirror image of the genome though too
DNA, they are closely related genetically. With increasing simi- constrained in its function to follow immediately the changes
larities between strain X and a type strain of a defined species that occur in less conserved molecules, the relationship between
A in either one or a combination of taxonomic parameters, e.g., these two parameters was curvilinear, with the rrn sequences
DDH, gene sequences, DNA and MALDI-TOF protein patterns, showing almost no divergence at DDH values above 70 %
ANI parameters, or even full genome sequences, corroborated (Amann et al. 1992; Fox et al. 1992; Stackebrandt and Goebel
by phenotypic similarities, the rational is also increased to clas- 1994). Each approach is strong in those areas of relationships in
sify strain X as a member of species A. This strategy has been which the other method fails to reliably depict relationships.
successfully applied during the history of microbiology. How- Sequence analysis has proven to be a reliable marker for the
ever, advances in genomic studies clearly indicated that vast phylogeny of organisms between the levels of domains (around
genomic differences may occur even among strains assigned to 55–60 % 16S rDNA gene sequence similarities) to moderately
the same species, such as those expressed by differences in related species (around 97 % similarities). Above this value,
genome size, e.g., Escherichia coli O157:H7 with a genome size DNA values can be as low as 55 % or as high as 100 %. Several
of 5.44 Mb possesses 1,346 genes not found in E. coli K-12 with organisms are known to share 99.8 % or even 100 % rRNA gene
a genome size of 4,64 Mb. Also, the MLST/MLSA approach sees sequence similarity that belong to different species because the
discrete centers of variance in most of the species investigated DNA reassociation values were below the 70 % threshold value.
(Lester et al. (2008) and Maiden and Dingle (2008) to name only Even if one considers the noise of DNA reassociation values
two). Interestingly, even species of the same genus may show originating from different laboratories using different
a different tempo and/or mode of evolution as exemplified in reassociation methods, the evidence is strong enough to state
species of the genus Campylobacter, where Campylobacter fetis that the resolution power of DNA-DNA reassociation is
significantly higher than that of 16S rDNA sequences. The find- including that of the variable regions, can be compared. The
ing that below a 16S rDNA gene sequence similarity value of branching pattern obtained will change when this small dataset
97 % the corresponding DNA-DNA reassociation value will not is embedded in a larger one composed of sequences of members
be higher than 60 % has led to the recommendation of families, orders, classes, and so on. At each level of ranks,
(Stackebrandt and Goebel 1994) that, at this and lower level of information will be lost, either by removal of variable regions or
sequence similarity, no DNA pairing studies need to be trimming of stem and loop structures to the minimum length
performed as the strain concerned will be highly unlikely to be common to all members of the data set and by omission of those
a member of a described species. Above 97 % rRNA similarities, regions for which ambiguous sequence information is provided.
however, taxonomists were encouraged to perform DDH to Each of these steps will most likely lead to changes in the
verify the species status of isolates. Using a significantly broader branching pattern of any lineage. Thus, the picture unraveled
data set, the 97 % threshold value defined in 1994 was corrected from the inclusion of thousands of sequences in a single data is
to 98.5 % in 2007 (Stackebrandt and Ebers 2006). not more than an approximation of the phylogeny. The litera-
Konstantinidis and colleagues (2006) also showed that even ture is full of examples that demonstrate the changes of the
a few genes in the genome, such as the 6–10 genes sequenced in phylogenetic relatedness within genera and families through
typical MLSA applications, could provide reliable estimates of the influence of new entries in the database. Branching patterns
the genome-aggregate ANI values. The correlation between the supported by high bootstrap values indicate that for a given
average nucleotide identity of 10 randomly selected genes in the branching point the statistical analysis supports the order of
genome and the genomic ANI was always significant and ranged lineages—but this statistical analysis is per se no indication
from 0.4 to 0.9 (Kendall t), depending on the genes used. In that the pattern reflects the natural relationship with a similar
particular, if selected genes do not represent fast evolving (e.g., degree of confidence.
surface proteins, transposases) or slowly evolving (e.g., rRNA Assuming that no algorithm exists that provides a tree
operon genes) genes (see > Fig. 9.2) then the ANI value of reflecting precisely the evolution of prokaryotes we must then
a small subset of genes approximates well the genome-aggregate accept the pattern(s) that appears to be the most plausible one. It
ANI. This study also revealed that ANI values correlate strongly may be the one showing the highest degree of topographic
(r2 > 0.9, p-value <0.001) with the genetic relatedness values similarity to patterns derived from different informative mole-
estimated using state of the art phylogenetic approaches, such as cules. With the availability of a wide range of MLSA-based and
maximum likelihood analysis of concatenated gene sequences. full genome-based trees, the branching order and intrataxon
Yet, ANI is a much simpler parameter to estimate, and the ANI coherency of 16S rRNA gene trees could be scrutinized. By and
calculation is less computationally intensive compared to the large the delineation of the major clusters, especially at the
alternative phylogenetic approaches because it employs the fast domain, phylum, and class level could be confirmed, though
blast algorithm and is performed in a pair-wise fashion similar the branching order often showed significant deviations.
to DDH.
Delineation of Genera: Pragmatic Approach

Comparison of Phylogenetic Patterns Derived
from 16S rRNA Gene Sequences The definition of a genus, given by Cowan in 1968 (Cowan
1968), has not been changed with the input of molecular data.
A previous comment (Stackebrandt 1992) that many trees are Cowan states that the genus is ‘‘one of the basic ranks in the
not comparable as they were generated on the basis of partial hierarchical systems used in biology, and probably the highest
sequences and different treeing methods is of historic interest rank with any significance in microbiology. In position between
only, as today mostly nearly complete sequences are compared ‘FAMILY’ and ‘SPECIES’, it is best considered as a collection of
using two or three different treeing algorithms of proven reso- species with many characters in common; unfortunately no one
lution power and statistical significance. The former inability of has indicated the extent of this sharing of characters, and it is
computers programs to handle the enormous amount of data in purely a matter of personal judgement (....) as to what consti-
a reasonable time without omitting either the number of refer- tutes a genus. Like the SPECIES, the genus is a subjective concept
ence organisms or sequence information became obsolete with without any foundation in fact.’’
increasing computing power and more sophisticated algo- A significant finding of the analysis of rRNA, rDNA, and
rithms. Some problems still remain which are firstly the subjec- DNA-DNA reassociation studies was to point out the genetic
tive omission of parts of the sequence judged to be of less heterogeneity of many phenotypically defined genera. The new
phylogenetic importance than other parts and secondly the results were often in discord with the classification of species
sometimes arbitrary omission of sequences available for mem- (Stackebrandt and Woese 1984): The working basis, i.e., the
bers of the taxon under investigation. However, one must dif- genus, had to be redefined or dissected or the type species was
ferentiate between different goals that determine the selection of found to actually be a member of a different genus. Examples
sequences to use. For the elucidation of intrageneric relation- have been described for Methanobacterium, Azospirillum, Pseu-
ships, the number of sequences is mostly restricted to those of domonas, Bacillus, Clostridium, Streptococcus, Flavobacterium,
type strains. In this case the complete sequence information, Bacteroides, Arthrobacter, Micrococcus, Brevibacterium, Nocardia,
and several genera of phototrophic organisms, but it should be genera. The correlation with phylogenetic analysis is so high that
stressed that almost each genus was involved in the reclassi- the finding of a new combination of such patterns is indicative of
fication process to varying degree. Some prominent examples the presence of a new genus (Embley and Stackebrandt 1994).
are listed in Stackebrandt (2000). Now, after 30 years of reclassi- Morphological, chemotaxonomic, and growth properties:
fication the situation is much more stable and reclassifications The main basis for dissection of the former genus Bacillus
are rare and mostly happen when intrafamily relationships (today the family Bacillaceae embraces more than 40 genera
change with the description of novel genera belonging to the [http://www.ncbi.nlm.nih.gov/Taxonomy/Browser/]) has been
family. the extensive phylogenetic analysis of its species (originally by
Scientists became aware of the high degree of horizontal gene Ash et al. 1991 and Rainey and Stackebrandt 1993). The separa-
transfer among prokaryotes (see above) during genome evolution into genera is mainly based upon the chemical structure of
tion, which challenges the use of universal genes in determining the peptidoglycan, cellular fatty acids, polyamins, polar lipids,
the whole range of phylogenetic relationships. A second wave of cell shape, spore shape, anaerobic growth, optimum pH, and
analyses of protein-coding genes was initiated by multilocus growth in 10 % NaCl. As compared to the actinomycete genera,
sequence analysis (MLSA) of housekeeping genes, a method the phylogenetic significance of chemotaxonomic properties is
derived from multilocus sequence typing (Maiden et al. 1996) much lower. The most species-rich genus, Bacillus (>165 spe-
to follow the epidemiology of pathogens and to unravel relation- cies), is heterogeneous with respect to amino acid composition
ships among closely related species (as opposed to the whole of peptidoglycan, spore shape, anaerobic growth, presence of
prokaryotic tree covered by universal genes). Usually a set of swollen sporangium, and other features, and it can be expected
three to ten sequences of orthologous protein genes (e.g., lepA, that this genus will be subject of even further dissection.
gyrB, fusA, pheS, rpoA/B, or atpA) are analyzed, using highly
conserved primers for PCR amplification and sequencing. Either
concatenated or individually, phylogenetic analyses contribute The Prokaryotic Species: A Historical View
significantly to the clarification of strains and species to species
and genus, respectively. Results often display chromosomal The rank ‘‘species’’ as defined today is an artificial, a pragmatic
crossover (HTG) events among species, warning against the construct that serves the main goal of identifying new isolates
practice of single gene-based identification. An ad hoc commit- and, if appropriate, providing sufficient properties to allow their
tee for the re-evaluation of the species definition in bacteriology classification. The previous paragraphs have already indicated
(Stackebrandt et al. 2002) strongly encouraged taxonomists to the scientific and conceptional limitations of this construct,
include this method in future species descriptions and taxo- which become increasingly obvious in the light of the insights
nomic studies, a suggestion which has been widely accepted from recent comparative MLST and MLSA as well as whole-
(e.g., Naser et al. 2005; Thompson et al. 2008; Rong and genome studies. One hundred and thirty years of history of
Huang 2010). Criteria for the use of gene sequences for the bacterial classification is now open to scrutiny, and, as in the
characterization of prokaryotic strains, including criteria for past, the majority of taxonomists will welcome the input of
quality check of sequences and alignment, use of alternative population geneticists, molecular biologists, and ecologists to
treeing methods and helpful hints for the interpretation of 16S either improve the definition or to suggest a concept that better
rRNA gene data have been compiled by Tindall et al. (2010). fits the evolutionary processes. However, a transition will not be
A new genus has to be described when a strain or a strain easy to achieve as the pragmatic definition of a prokaryotic
cluster is shown to branch outside the radiation of a validly species is well accepted among bacteriologists. During
named genus and the isolated phylogenetic position is accom- a process of revision and constant adaptation, new insights
panied by the presence of distinct phenotypic properties not into cell structure and cell function are incorporated, taking
found among the neighboring genera. On the other hand, the into account the microbiologists’ perception. As Staley and
placement of a new taxon with a novel pattern of phenetic Krieg (1984) indicated, ‘‘a classification that is of little use to
properties within the radiation of a genus may point toward its the microbiologist, no matter how fine a scheme or who devised,
taxonomic heterogeneity, which consequently may lead to the it will be ignored or significantly modified.’’ This is true for each
dissection of the genus. The decision about which phenotypic level within a hierarchic system and the history of microbiology
properties to use for the circumscription of a novel genus is up has witnessed ample examples for the description and rejection
to the taxonomist but depends to some extend upon the descrip- of such systems. Even more prone to the microbiologists’ idea
tion of the neighboring genera. The genus-specific characteris- about the concept of a species are the higher ranks which are
tics must be present in each species of the genus or the genus almost completely defined by subjective arguments—to a level
description needs amendment. The following gives two exam- where the importance of working with taxa above the rank of
ples for the usefulness of phenotypic properties to corroborate genera is considered trivial and unimportant (O’Hara 1994) (see
phylogeny-based delineations: also above). The importance of any hierarchic system goes
Chemotaxonomic properties: In the order Actinomycetales, beyond the main function of classification and identification.
the high degree of chemical diversity in the peptidoglycan, fatty Higher ranks are developed to explain and to understand the
acids, polar lipids, menaquinones, whole cell sugars, or teichoic evolution of organisms and parts thereof, based upon the knowl-
acid offers superb diversity at the epigenetic level to delineate edge that has been available at the time of their establishment.
In bacteriology, time has seen various hierarchic systems and species and never to none, (2) all descents of a species are
proposed phylogenetic paths which in hindsight failed because member of that species, (3) members of a species should be
they were not based upon the natural relationship of the organ- phylogenetically related, and (4) the species so defined should
isms concerned but on properties that were believed to express apply to taxa that coincide more or less with the intuitively
natural relationships: morphology (Cohn 1872; Stanier and van recognized species. Most obviously, points (1) to (3) are already
Niel 1941), pigmentation, physiology (Orla-Jensen 1909; matched in bacteriology while point (4) has failed significantly
Margulis 1981), and cell constituents (Schleifer and Kandler in the past because of inability to classify a prokaryotic species by
1972). Some of these attempts represented important contribu- intuition.
tions in their time because the classification system based upon The species definition applied today does not incorporate
them and some of their constituent does actually reflect phylo- the concept of how strains may develop into entities named
genetic divergence (e.g., peptidoglycan structure, lipids, fatty ‘‘natural’’ species. Several factors have been identified through
acids). The order of phylogenetic lineages guides the bacteriol- intensive multilocus sequence typing of housekeeping genes
ogists to the two basic units, the genus and species, without the (Maiden et al. 1996), RAPDs and multilocus enzyme electro-
need for a superimposed system. Actually, there is no immediate phoreses (Selander et al. 1994; Istock et al. 1996), to contribute
need to work with a hierarchic system but is tempting to do so in to the evolution of the genome. Some organisms, e.g., Neisseria
order to comprehensively classify according to similarities and and Rhizobium species, as well as enterobacterial species
differences and evolutionary traits. Today, we see the emergence (Guttman and Dykhuizen 1994), are subjected to reticulate
of higher taxa along the phylogenetic structure, and, as in sys- events or panmixis (Maynart-Smith et al. 1993; Istock et al.
tems of plants and animals, taxa of the same rank are not 1996) in which clonal relationships, due to mutational events
necessarily comparable units and described in a coherent way. and vertically transmitted accessory genetic elements, are
Also, we must be aware that only a small fraction of prokaryotic perturbated by horizontal genetic transfer, e.g., conjugation,
species are described, and new entries will not only change the and phage transduction DNA transformation (Achtman 1998).
description of the higher ranks but may change the composition Other strains, which are mostly endosymbionts and obligate
of taxa as well. However, the advantage of basing the hierarchic pathogenic organisms, such as members of the genera
structure on a rational basis, i.e., on semantides reflecting the Bartonella, Brucella, and Rickettsia, are mainly clonal because
organisms’ evolutionary history, makes it highly likely that they are only rarely subjected to horizontal gene transfer. In
changes within the system will occur mainly within ranks of some species, the recombination among strains of the same
a common genealogical lineage and not, as in the past, affect and species is more frequent among strains of different species
possibly change the units of remotely related taxa. (e.g., the enterobacteria), which may lead to the homogenization
Surprising perhaps to microbiologists, there are some zool- of the gene pool of the interacting organisms (Guttman and
ogists (Hull 1976) and botanists (Bachmann 1998) who discuss Dykhuizen 1994). In an attempt to come to a biological species
the possibility that the species as an objective basic unit of definition for bacteria, it has been proposed (Dykhuizen and
taxonomy as defined in the past does not exist. The nonexis- Green 1991) to base the species definition on the following
tence, to be more precise the inability of microbiologists to observations: (1) Phylogenetic trees from different genes from
define a species as an objective category, a product of natural members of a single species should be the same. (2) Phylogenetic
selection that after sufficient studies identifies itself to the tax- trees from different genes from members of different species
onomist, has been recognized more than 30 years ago. Bacteri- should be the different (as shown for two genes from seven
ologists in particular follow guidelines and recommendations in species of Neisseria). Without questioning the validity of this
order to provide stability, reproducibility, and coherency in approach, it is obvious that this strategy is far beyond a routine
taxonomy—although the final decision about species descrip- method at the present status of sequence analysis, especially as
tion is still based upon his or her subjective judgment. This several strains of a single species must be investigated, and
concept does not include the role of reproductive isolation, although worth discussing, this approach cannot replace the
i.e., the barriers to horizontal gene transfer over large phyloge- present pragmatic species definition.
netic distances; it does not even try to explain the mode of The current pragmatic approach to the definition of species
speciation. One may be amused upon such a naive approach is also not taking into account the ecological niche, although the
to handle the ‘‘species’’ matter—but once you decide that source of isolation should be part of a species description.
a species cannot be recognized as a natural entity, it is only However, the isolation site can be considered to reflect the actual
consequent to work with the compromise of a working defini- place or the role in the ecosystem that the organism occupies
tion. This strategy has facilitated the practice of taxonomy—a only in those instances in which the organism has a close depen-
way that may also be gone by protozoologists, mycologists, and dence to its environment, e.g., the rhizoplane, rhizosphere, in
algologists. As Bachmann (1998) points out, the most useful endosymbiotic, and pathogenic relationships. The description
general species definition would be the one that allows ‘‘the marine water, fresh water, mud, sediment, soil, rumen, skin, and
largest number of individual organisms be unequivocally be so on, is too superficial to describe the exact niche in complex
assigned to species so that some basic conditions are satisfied.’’ environmental samples. Knowledge about the site of speciation
These conditions are (1) membership of a strain to only one and the environmental selection of members of a clonal
population may be helpful in explaining the path of evolution hierarchic structure represents a solid foundation, emerging
and the mode of speciation—but it provides no clue in the from 100 years of changes in nomenclature and relatedness of
definition at what level subpopulations may be regarded as taxa at all levels of ranks. In order to maintain and to safeguard
individual species. prokaryotic systematics, taxonomists developed a conservative
The evolutionary record of molecular sequences provides attitude, and they will await a concept of proof that the emerging
a basis for phylogenetic studies, which has allowed incorporating broad range of molecular methods and theories derived there
the bacterial world within the constraints of genealogical studies. from will be of benefit and manageable as a framework to either
But this information alone does not per se conclude on the substitute or to add to the present approach.
decision about what a species is and how a species has to be
defined—it just puts the prokaryotes on the same level with
animals and the higher evolved plants in the discussion about The Pragmatic Species Definition
the concept of the species category as a general unit for biodi-
versity, evolution, and taxonomy. The mere availability of The definition of a prokaryotic species embraces the phyloge-
molecular data does not allow biologists to define the category netic component given by Cracraft (1983) for a phylogenetic
species as a comparable biological entity throughout the diver- species as ‘‘the smallest diagnosable cluster of individual organ-
sity of organisms. The dilemma must be seen in the fact that isms within which there is a parental pattern of ancestry and
biologists themselves are not clear about the definition ‘‘species,’’ descendent’’ and the taxonomic component given by Colwell
for which a theoretical basis is lacking (Bachmann 1998). The (1970) as ‘‘a group of related organisms that is distinguished
phylogenetic species concept (Cracraft 1983), the taxonomic from similar groups by a constellation of significant genotypic,
species concept (Staley and Krieg 1984), the biological species phenotypic, and ecological characteristics.’’ The definition used
concept (Dobzhansky 1937; Istock et al. 1996) (disregarding today includes components of a geno-, or genomo-species,
asexual reproduction entirely), and the ecological species con- a taxo-species, and a pheno-species, which reflect the different
cept (Istock et al. 1996), and more concepts, all have their concepts of species descriptions in the past decades. From
strengths and weaknesses, and each of them stresses different a pragmatic point of view, all these facets are incorporated into
aspects of biology and evolution. This topic has been discussed a single definition, though the terms are still in use. An ‘‘opti-
by Gevers et al. (2005) in the light of multilocus nucleotide- mal’’ species would be the one that, at the same time, is not only
sequence-based approaches and has recently been in the focus of representing a phylogenetically and phenotypically coherent
intense discussions. Modes of horizontal gene transfer and unit but also one that naturally exists. However, except for
homologous recombination (Lawrence and Retchless 2009), as some strains of pathogenic species, the ecological niche is either
well as population genetic and macroevolutionary theories, are not known or the number of isolated strains is too small to
considered necessary for the evaluation of interspecific and conclude on their original habitat (e.g., airborne isolates; aquatic
intraspecific variation. Barraclough et al. (2009), Cohan and isolates that entered rivers, lakes, and the ocean by terrestrial
Koeppel (2008), and Fraser et al. (2009) added the ecological runoffs; or sediment isolates). It should also be remembered that
theory to the genetic theory of speciation. Doolittle and for the description of the type strain, a single strain is selected
Zhaxybayeva (2009), reviewing existing theories, argue that out of a broad diversity of naturally occurring prokaryotic
‘‘there is no principled way in which questions about prokary- organisms, which may reflect a continuum of genetic and epi-
otic species, such as how many there are, how large their genetic diversity. Strains that are sufficiently similar in those
populations are, or how globally they are distributed, can be aspects used today in prokaryotic taxonomy, i.e., mainly DNA-
answered.’’ Metagenomic data in combination with alternative DNA reassociation, are considered members of this species. This
theories, however, will allow meaningful questions about the concept of selecting species has been described as the arbitrary
biological processes of speciation to be addressed and one day species concept (Staley 1997). The combination of arbitrary
one may develop alternatives to the present pragmatic approach. selection and artificial species delineation is admittedly arguable
Thane (2009) come to a similar conclusion arguing that ‘‘species and open to discussion, which has recently been intensified with
as being discrete clusters or monophyletic lineages are at odds the advent of genomic information. The entity species
with most of the data, suggesting that taxon circumscription can circumscribed by bacteriologists cannot be compared to that of
only proceed by informed compromise, pragmatism, and subjec- zoology and botany. For example, Homo sapiens and its closest
tivity.’’ In this respect, the conclusion is similar to the perspective relatives, the higher evolved apes, comprising about 200 species,
of 40 years ago, when Cowan (1968) already bluntly stated in the are related by higher than 75 % DNA reassociation (Sibley et al.
Dictionary of Microbial Taxonomic Usage that the category 1990); transferring this concept to bacteriology, all these species
species does not exist and does not represent a natural entity. would fall into a single bacterial species delineated by the 70 %
The aspect, theories, and evolving methods reflected upon in threshold value. In the absence of knowledge of (an) underlying
the past 10 years on the ‘‘species’’ concept will sooner or later speciation process(es), the pragmatic approach to the species
influence the today’s pragmatic species definition and the guard- definition in bacteriology has nevertheless been extremely useful
ians of taxonomy will have to adapt to the changing wind. This and the success of the definition is measured by its widespread
will not be an easy goal to achieve as the current scaffold of the acceptance.
Delineating a Species in the Frame of the interested to determine the more precise affiliation of the isolate
Pragmatic Definition and will continue with the classification process. The way to
proceed depends upon the number of species in the phylogenetic
In the daily routine, a new isolate is subjected to an identifica- vicinity of the isolate.
tion process, which may be different from laboratory to labora-
tory and from taxon to taxon. Many scientists, however, are not 1. In case the isolate falls within the boundaries of a genus, the
interested in a fine resolution of relatedness or they are not in description of this genus will guide the few key properties to
a position to go through the laborious identification process. analyze, which will demonstrate that the isolate actually
Some initial superficial tests are performed, like determination should be considered as a member of this genus. If the
of colony morphology and pigmentation, shape, spore forma- species within this genus are separated by distinct pheno-
tion, Gram-stain, and relationship to oxygen. The aim of the typic properties (which one should assume but which is not
study defines the identification procedures to follow, but even always the case), these should be searched for in the isolate,
with the availability of a broad range of molecular tools, the and, if present, the isolate has been identified. If not, it is
phylogenetic diversity is rarely fully explored for superficially recommended to perform DDH or ANI studies to determine
similar strains. whether the isolate is the nucleus of a new species. DDH
Often, the classification process starts when the isolate does similarities of lower than about 70 % are indicative of a new
not match the description of 1 of the about 8,750 species validly species and properties that distinguish the new species from
described today (http://www.bacterio.cict.fr/number.html). In the established ones should be sought to identify. In those
most cases today, this is performed by examining the phyloge- cases where the DDH similarity between an isolate and
netic position of a novel isolate determined by sequence analysis a described species is 70 %, this information will usually
of its 16S rRNA genes, as the database of prokaryotic strains is not be recorded unless the new strain leads to the description
enormous, covering more than 95 % of the described species. of a subspecies or to an emendation of the species
This approach may change with the advent of inexpensive and description.
fast generation of draft genome sequences, foreseen as early as The recommendation (Wayne et al. 1987) to delineate
1999 (Relman and Strauss 1999) and further elaborated by species in genomic terms at a threshold value of around 70 %
Buckley and Roberts (2007) and Riley and Buckley (2008). In DDH similarity are guidelines but should not be applied as
order to search for the closest relative for which 16S rDNA gene fixed rules. Though the majority of species are actually
sequence data are publicly available, taxonomists make use of described as suggested, there are a few exceptions, which
electronic tools such as, to name the most widely used ones, the are indicated by Stackebrandt (2000):
BLAST system ([email protected]; http://www.ncbi. Despite the recommended value of 70 % DNA similarity,
nih.gov/cgi-bin/BLAST/nph-newblast), the Ribosomal Data- taxonomists working with some defined prokaryotic groups
base Project ([email protected]; http://www.cme.msu. have altered this value to come to a better correlation
edu/RDP), the ARB program ([email protected] between phenotypic and genotypic similarities. Within the
muenchen.de; http://www.mikro.biologie.tu-muenchen.de), family Pasteurellaceae, a DNA reassociation value of at least
the GreenGenes database (http://greengenes.lbl.gov/cgi-bin/ 85 % describes a species. Similar values have been found for
nph-index.cgi), or the living tree project embracing over 8,000 the interspecies relatedness of Bordetella pertussis,
curated sequences, each of which represents a single type of B. parapertussis, and B. bronchiseptica (Kloos et al. 1981)
strain of a classified species (http://www.arb-silva.de/projects/ and between members of the spotted fever group of
living-tree/). The search in the public databases will show the Rickettsia (Walker 1989).
phylogenetic distance to the isolates’ nearest neighbor(s), but 2. In case the 16S rDNA gene sequence similarity values indicate
the quality of the search depends upon the completeness of the an approximately equidistant relationship to members of
16S rDNA gene database, both, in terms of available sequences different genera, the diagnostic properties given for these
and species analyzed as well as the completeness or not of the genera must be tested for the isolate. Such highly related
sequences obtained. In this respect, we recommend the use of genera have been described, e.g., in the Actinomycetales,
the living tree project database. Once the approximate nearest Flavobacteriaceae, Alphaproteobacteria, Gammaproteo-
phylogenetic neighbors have been identified, it is recommended bacteria, and Firmicutes. If the properties match those of
to search the public databases in order to not miss any recent one of the genera, the identification process will be restricted
entries. The recent publication of Tindall et al. (2010) includes to members of this genus, and one has to proceed as indicated
guidelines for proper analyses of 16S rRNA gene sequences, above. If the analysis of the genus-specific properties will
including the description of novel species. reveal no unambiguous match with any of the neighboring
The result of the phylogenetic analysis will have conse- genera, it is likely that the isolate represents yet another
quences on the strategy to continue. Let us assume that the closely related genus of this genus cluster, and the description
16S rDNA gene similarity values will be higher than 98.5 % to of the new species and the new genus will go hand in hand.
its nearest neighbor. Many scientists will be satisfied knowing 3. In case the new isolate shares less than 98.5 % sequence
the approximate phylogenetic position and will not continue similarity to the nearest phylogenetic neighbor, DHH stud-
with the identification process. Others, however, will be ies are unnecessary to perform as similarity values will range
clearly below the 70 % reassociation borderline value The previous difficulties to determine intraspecies variations
recommended for the present species definition. consistent with a specific function explain the rather low and
In contrast to the rather stringent genomic definition of over the years rather constant number of subspecies descriptions
a ‘‘species,’’ the phenotypic characterization of a new species (see > Table 9.4), which are mainly found in species of
is very variable. Information about the phylogenetic position medical species (e.g., M. avium, M. tuberculosis, Campylobacter
of a putative new type strain facilitates the selection of such lari, Francisella tularensis) and technological importance (e.g.,
features and guides the search for taxonomically relevant Bacillus subtilis, Acetobacter xylinus, Photorhadus luminescens,
properties. The properties to be investigated depend upon Xylella fastidiosa, to name a few recent descriptions). The criteria
firstly on those indicated as being specific for the genus and to delineate subspecies are at the discretion of the taxonomist
secondly on the set of characters already indicated to be of and generally no rationale for the subspecies description is
discriminatory value for species described for the genus. given, other than the indication of discrete genomic centers of
Extensive morphological and ultrastructural characteri- variance: For example, the description of the three subspecies of
zation must be presented especially for species of novel Pseudomonas chlororaphis, P. chlororaphis subsp. chlororaphis,
genera. Records on enrichment and isolation, motility, col- P. chlororaphis subsp. aureofaciens, and P. chlororaphis subsp.
ony characterization, optimal growth conditions, growth aurantiaca, are based on strains of previously separate species
requirement and substrates, and base composition of DNA (Peix et al. 2007) which showed higher than 75 % DDH simi-
are part of a set of characterizing features (Tindall et al. larities, quantitative differences in fatty acids, phylogenetic
2010). Analysis of special features, such as antigenic charac- placement of strains based upon 16S rRNA, atpD, recA, and
terization, is required for certain taxa. Many of the proper- carA gene sequences and differences in five phenotypic proper-
ties to be provided for the description of a species are listed ties. The presence of different DDH groups, electrophoretic
either in the descriptions of minimal standards, which are motility of certain enzymes, and 16S rRNA gene typing led to
available for species of some genera, or they are compiled in the description of three subspecies of Fusobacterium nucleatum
Bergey’s Manual of Systematic Bacteriology. (Dzink et al. 1990; Gharbia and Shah 1992). In yet another case
(Yanagida et al. 1997), the two subspecies of Sporolactobacillus
nakayama share lower than 65 % DDH similarities, which, in the
Definition of a Subspecies presence of distinct phenotypic differences, would have allowed
the description of two separate species.
The rank subspecies is the lowest rank covered by the Code of It is desirable to base descriptions of taxonomic ranks on
Nomenclature. Its definition is somewhat vague as, unlike in the easily identifiable properties—a criterion that is not always met
definition of the species, no molecular or phenetic threshold at the subspecies rank. In the absence of chemical or metabolic
values are indicated at which intrageneric clusters are separated differences, recognized molecular intraspecific clusters can be
from each other. According to Staley and Krieg (1984), described on the basis of reproducible typing patterns, including
a subspecies is ‘‘based on minor but consistent phenotypic MALDI TOF profiles, signature nucleotides of genes coding for
variations within the species or on genetically determined clus- proteins and rRNA, presence/absence of genes, chromosome
ters of strains within the species,’’ while Wayne et al. (1987) topology and the like. These centers of intraspecific variations
argue that ‘‘Subspecies designations can be used for genetically may be the origin of future species and thus, are of scientific
close organisms that diverge in phenotype. . . .. . . There is a need value to be named. The question whether a formal naming is
for further guidelines for designation of subspecies.’’ The defi- indicated for subspecies with no recognized metabolic function,
nition used in botany and zoology, i.e., ‘‘a geographically defined host affiliation, or pathogenicity remains open, requiring
grouping of local populations which differs taxonomically from a formal decision by an internally recognized body.
similar subdivisions of species’’ (Gorilla gorilla subsp. gorilla is
the western lowland gorilla, and Gorilla gorilla subsp. graueri is
the eastern lowland gorilla), has not influenced the definition Species Concept in the Era of Metagenomics
used for prokaryotes yet. This may change with the advent of
high-resolution molecular analysis, apt to reveal restricted geo- The species concept issue has been discussed extensively earlier
graphical distribution among strains. Multispacer typing in this chapter and elsewhere (e.g., Rosselló-Mora and Amann
[MST], MLST, and MLSA analyses did already influence recent 2001). Here, we focus on the recent developments and the
descriptions of subspecies, e.g., Rickettsia conorii (Zhu et al. challenges remaining, giving first a brief introduction into
2005), while comparative genome sequence analyses supported metagenomics, which is necessary for the remaining of this
a polyphasic approach to describe two subspecies of Bacillus section.
amyloliquefaciens (Borris et al. 2010). MLSA have been used to Metagenomics represents a new field, where the power of
investigate the molecular coherence and relatedness of subspe- genomic tools such as cloning and sequencing of small pieces
cies, e.g., in Mycobacterium avium (Turenne et al. 2008), and from a DNA sample is applied to entire communities of microbes,
even used to combine several species and subspecies of the Strep- bypassing the need to isolate and culture community members
tomyces griseus clade (Rong and Huang 2010) and Streptomyces (Handelsman et al. 2007). Thus, metagenomics can reveal the
albidoflavus clade (Rong et al. 2009). total naturally occurring diversity and provide novel insights into
a 100 b
90
Genetic discontinuity
Nucleotide identity
80
70
60
50
0 0.5 1.0 1.5 Scale
Position in Prochlorococcus marinus genome (in Mb) 0.01
. Fig. 9.3
Sequence-discrete populations of photosynthetic Prochlorococcus marinus recovered in metagenomic datasets from the Sargasso Sea. The
whole metagenomic dataset from the Sargasso Sea reported by Venter and colleagues (2004) was searched against the genome
sequence of P. marinus strain AS9601, which represents the high-light surface Prochlorococcus, as described in Konstantinidis and
DeLong (2008). The graph represents the nucleotide identity of each metagenomic sequence that shows at least 60 % nucleotide
identity to the reference genome sequence (y-axis) plotted against the position of the genome that the metagenomic sequence maps on
(x-axis) (panel A). The analysis reveals that Prochlorococcus forms a sequence-discrete population, which is composed of genotypes that
show at least 90 % nucleotide identity among themselves (blue dots; each dot represents a different metagenomic sequence; each
sequence is about 1 kb long) and are clearly divergent from their co-occurring genotypes in the sample (red dots). The few reads of
intermediate genetic relatedness (gray dots) represent primarily sequence errors or artifacts and, secondarily, ecologically divergent
genotypes, since under the same environmental conditions (i.e., they are co-occurring in the sample) they show substantially lower
abundance based on the number of reads recovered and as explained in Konstantinidis and DeLong (2008). The sequence-discrete
population is also obvious based on a phylogentic approach, using a tree built from randomly drawn fully overlapping metagenomic
reads (panel B). Similar patterns of diversity were observed for numerous other marine microbial groups
the gene content and function of microbial populations, includ- community, if any, based on their relative abundance in situ and
ing those that are resistant to laboratory cultivation. Numerous lower interpopulation ANI values (typically <85 % ANI) rela-
examples of successful application of metagenomics to the study tive to the intrapopulation ANI. These genotypes also showed
of both low-diversity (e.g., biofilms) and high-diversity (e.g., significantly smaller gene-content differences compared to the
complex) natural microbial communities have been conducted differences seen within many commonly defined bacterial spe-
recently, e.g., (Tyson et al. 2004; Rusch et al. 2007; Konstantinidis cies and exhibited substantial levels of genetic exchange among
and DeLong 2008). Therefore, metagenomics hold great potential themselves, mediated by homologous recombination
to provide new insights into the most fundamental question in (Konstantinidis and DeLong 2008). Although more habitats
taxonomy, namely, whether or not the prokaryotic diversity is (e.g., soil, sediment) and samples over time need to be analyzed
organized in discrete units, species. before more robust interpretations can emerge, it is intriguing to
In all metagenomic studies performed today, the microbial hypothesize that the sequence-discrete populations identified
populations comprising the communities sampled were found may correspond to genuine species and represent the important
to represent discrete, sequence-defined populations, with intra- units of microbial diversity. The genotypes of these populations
population genomic relatedness ranging from 94 % to 100 % meet several of the desirable properties expected for strains of
ANI depending on the population considered, at least for the the same species such as the genetic distinctiveness and similar
abundant populations that were adequately sampled by the ecophysiological properties. It is also interesting to point out
sequencing coverage achieved in the studies (Tyson et al. that the areas of genetic discontinuities usually observed
2004; Rusch et al. 2007; Konstantinidis and DeLong 2008). between sequence-discrete populations in the previous
The abundant populations sampled included both autotrophic metagenomics studies (85–95 % ANI) correspond roughly to
and heterotrophic populations of several prokaryotic phyla, the 70 % DDH threshold, indicating that there might be an
including archaeal ones, indicating that the patterns observed important underlying mechanism(s) that accounts for the pat-
may be universal. > Figure 9.3 gives such an example for mem- terns of species diversity described by taxonomists over the past
bers of the Sargasso Sea population of P. marinus. Additional few decades (see also discussion about recombination below).
analyses showed that the genotypes of a population were clearly In contrast, when genotypes from different populations and
distinguishable from their closest co-occurring relatives in the habitats characterized by dissimilar physicochemical properties
were compared, a genetic continuum, as opposed to sequence- mechanism. The former can potentially serve as a population
discrete clusters, was frequently observed. For instance, when cohesion mechanism if it is rampant and unbiased, similar to the
Konstantinidis and DeLong compared populations of ammonia role of sex in higher eukaryotes. If, on the other hand, homol-
oxidizing Crenarchaea group I from different depths in the ogous recombination is driven by selection for a few gene func-
Pacific Ocean, they recovered organisms that showed the com- tions important in the environment then it is unlikely to lead to
plete range of genetic relatedness among themselves from 70 % genome homogenization (except for the genes under selection)
to 100 % ANI. However, the Crenarchaea populations were and population cohesion. Several cases of selection-driven
found to be sequence-discrete within a specific depth (intrapop- recombination have been documented in recent high-resolution
ulation ANI values >95 %), which is consistent with indepen- genomic studies such as for pathogenic Campylobacter and
dent studies suggesting that microbial life is stratified with depth Streptococcus populations, which preferentially exchange only
in the ocean’s interior (e.g., DeLong et al. 2006), resulting from genes related to antibiotic resistance and evasion of host
the unique physicochemical properties characterizing different immune system (Caro-Quintero et al. 2009; Beres et al. 2009).
depths (e.g., light intensity, nutrient availability, temperature, Nonetheless, substantial levels of unbiased intrapopulation
salinity, hydrostatic pressure). A subsequent study identified homologous recombination were detected for several
several genomic adaptations that presumably account for the sequence-discrete marine populations and populations of
spatial (depth) isolation of the sequence-discrete populations biofilms related to acid mine drainage, using various methods
such as the preferential use of specific amino acids in the protein (Tyson et al. 2004; Konstantinidis and DeLong 2008). Homolo-
sequences of organisms from deeper waters that maintain pro- gous recombination frequency was also shown to drop logarith-
tein tertiary structure under high hydrostatic pressure mically with increasing evolutionary divergence of the
(Konstantinidis et al. 2009). These findings confirm that in recombining sequences, particularly in the range of 80–90 %
order to better understand the drivers of genetic variation in sequence nucleotide identity (Zawadzki et al. 1995; Eppley et al.
bacterial populations, within-community genomic variation 2007) that roughly corresponded to the areas of genetic discon-
should be analyzed, focusing ideally on abundant populations tinuities identified between sequence-discrete populations.
that are less likely to represent transient and/or allochthonous Although these studies were based on short, error-prone
members of the community, as opposed to genotypes recovered metagenomic sequences, recent whole-genome-based analyses
from different populations and habitats. The findings also imply of Shewanella baltica strains, which co-occur in the Baltic Sea,
that several previous studies that have recovered continuous revealed that homologous recombination was affecting (purg-
genetic gradient as opposed to discrete populations might have ing) more nucleotide substitution positions than those created
been biased by the samples analyzed and the limitation of the by point mutations (sexual evolution) (Caro-Quintero et al.
culture-based approaches to not distinguish between transient 2011). These findings make homologous recombination an
and abundant community members. Interestingly, intriguing candidate mechanism responsible for population
Konstantinidis and DeLong (2008) also found that populations cohesion and for maintaining the sequence-discrete populations
of Crenarchaea and photosynthetic Prochlorococcus from similar identified in metagenomic studies. However, more quantitative
depths in the Pacific and Atlantic Oceans are indistinguishable data on the role of homologous recombination is needed to
from each other at the ANI level, indicating panmictic support this hypothesis and exclude alternatives. For instance,
populations across the World’s two largest oceans (see also to what the extent the patterns seen in the S. baltica case repre-
Konstantinidis et al. 2009). sent the norm as opposed to a rare event remains currently
Several species concepts based on recombination frequency unclear. Related to this, contemporary methods (McVean et al.
(Fraser et al. 2007) or population sweeps caused by periodic 2002; Kosakovsky Pond et al. 2006) frequently vary by an order
natural selection (Acinas et al. 2004; Cohan 2006) have been of magnitude in their estimates of recombination rates (Eppley
advanced to explain the maintenance of sequence clusters such et al. 2007; Konstantinidis and DeLong 2008). This is primarily
as the sequence-discrete populations discussed above. Alterna- due to the low accuracy in detecting recombination among very
tive explanations such as population bottlenecks and random closely related organisms and to the fact that several important
birth/extinction are less favorable and probably applicable to parameters such as the effective population size remain specu-
more restricted microbial populations and habitats, such as the lative for natural populations (Fraser et al. 2007).
vertically transmitted microbial pathogens (Moran 2007), com- On the other hand, the genotypes of a population could
pared to the populations in open environments and complex cohere together via means of a shared ecological niche, as this
communities discussed above. Recent studies have provided has been shown to be the case for several microbial populations
evidence in support of both of the previous two major species and extensively discussed theoretically in various versions of the
concepts; however, the number of populations and samples per ecological species concept (e.g., Acinas et al. 2004; Cohan 2006;
population analyzed to date are simply too small to allow for Ward 2006). It is also important to realize that values of a few
universal conclusions to emerge. We discuss the emerging new percent nucleotide sequence divergence (e.g., 97–98 % ANI)
insights and highlight the remaining open questions below. correspond to long periods of time elapsed since the last com-
HGT can introduce novel DNA into a genome through mon ancestor of the genomes related at this level, in the order of
a homologous (the sequences are sufficiently similar to allow thousands of years (Lawrence and Ochman 1998). During such
one displacing the other) or nonhomologous recombination long evolutionary time, it is likely that genotypes that are more
divergent compared to the genotypes comprising the population References

have become extinct (e.g., by accumulating deleterious muta-
tions), differentiated ecologically (speciated), or simply been Achtman M (1998) Microevolution during epidemic spread of Neisseria
transported to other habitats. Thus, the ecological species con- meningitidis. Electrophoresis 19:593–596
Achtman M, Wagner M (2008) Microbial diversity and the genetic nature of
cept can explain the sequence-discrete populations observed in
microbial species. Nat Rev Microbiol 6:431–440
metagenomic studies; albeit direct experimental evidence in Acinas SG, Klepac-Ceraj V, Hunt DE, Pharino C, Ceraj I, Distel DL, Polz MF
support of the previous interpretations is difficult to obtain. (2004) Fine-scale phylogenetic architecture of a complex bacterial commu-
Ecological niche differentiation could also lead to subpopu- nity. Nature 430:551–554
lations, also called ecotypes (Cohan 2006), and eventually to Alperi A, Martı́nez-Murcia AJ, Monera A, Saavedra MJ, Figueras MJ (2010)
Aeromonas fluvialis sp. nov., isolated from a Spanish river. Int J Syst
speciation. Although the ecotype is not recognized as an official
Evol Microbiol 60:72–77
rank of the bacterial taxonomy, ecotypes are commonly defined Amann R, Ludwig W, Schleifer KH (1988) Beta-subunit of ATP-synthase: a useful
as collections of strains that preserve the full phenotypic and marker for studying the phylogenetic relationship of eubacteria. J Gen
ecological potential that characterizes the species and are able to Microbiol 134:2815–2821
exploit only a slightly different ecological niche compared to Amann RI, Lin C, Key R, Montgomery L, Stahl DA (1992) Diversity among
Fibrobacter strains: towards a phylogenetic classification. Syst Appl
closely related species, such as growth on a new carbon substrate.
Microbiol 15:23–31
Therefore, an ecotype should be expected to have only a small Ash C, Farrow JAE, Wallbanks S, Collins MD (1991) Phylogenetic heterogeneity
gene-content (or expression) difference compared with other of the genus Bacillus revealed by comparative analysis of small-subunit
ecotypes of the same species, or if larger differences exist, they ribosomal RNA sequences. Lett Appl Microbiol 13:202–206
would be encoded by unstable parts of the genome such as Bachmann K (1998) Species as units of diversity: an outdated concept. Theor
Biosci 117:213–230
plasmids. The major challenge in identifying and characterizing
Balch WE, Fox GE, Magrum LJ, Woese CR, Wolfe RS (1979) Methanogens:
ecotypes, and, in general, species based on the ecological species reevaluation of a unique biological group. Micribiol Rev 43:260–296
concept, is to define the ecological niche. Niche is Barraclough TG, Hughes M, Ashford-Hodges N, Fujisawa T (2009) Inferring
a multidimensional space that cannot be measured precisely evolutionarily significant units of bacterial diversity from broad environ-
with current technologies; for some researchers, defining the mental surveys of single-locus data. Biol Lett 23:425–428
Baumann P, Baumann L, Woolkalis MJ, Bang SS (1983) Evolutionary relation-
ecological niche of microorganisms within most natural habitats
ships in Vibrio and Photobacterium a basis for a natural classification. Annu
on Earth represents a futile exercise. Accordingly, ecotypes are Rev Microbiol 37:369–398
typically detected indirectly, through the phylogenetic analysis Bautz EKF, Bautz FA (1964) The influence of non-complementary bases on the
of a single or a few genes from the population(s) under scrutiny, stability of ordered polynucleotides. Proc Natl Acad Sci USA 52:1476–1481
assuming that the appearance of a distinct clade in the phylogeny Beiko RG, Harlow TJ, Ragan MA (2005) Highways of gene sharing in prokaryotes.
(equated to ecotype) is accompanied by ecological (ecotype)
Beres SB, Carroll RK, Shea PR, Sitkiewicz I, Martinez-Gutierrez JC et al
differentiation. However, the genes frequently employed for (2009) Molecular complexity of successive bacterial epidemics deconvoluted
such purposes are not responsible for the distinctive ecophysio- by comparative pathogenomics. Proc Natl Acad Sci USA 107:4371–4376
logical trait of the ecotype (e.g., they are housekeeping genes); Borriss R, Chen X, Rueckert C, Blom J, Becker A, Baumgarth B, Fan B, Pukall R,
thus, when the distinct clade becomes detectable in the phylog- Schumann P, Spröer C, Junge H, Vater J, Pühler A, Klenk H-P (2010)
Relationship of Bacillus amyloliquefaciens clades associated with strains
eny, it has probably diverged enough to be recognized as
DSM7T and FZB42: a proposal for Bacillus amyloliquefaciens subsp.
a distinct species rather than an ecotype. Due to the inability amyloliquefaciens subsp. nov. and Bacillus amyloliquefaciens subsp.
to define the ecological niche and the ambiguity associated with plantarum subsp. nov. based on their discriminating complete genome
deciding when a clade should be defined as a species versus sequences. Int J Syst Evol Microbiol. First published on Sep 3, 2010 as doi:
ecotype, the ecotype concept remains of limited practical use 10.1099/ijs.0.023267-0
Brenner DJ (1991) Additional genera of the Enterobacteriaceae. In: Balows A,
in taxonomy. With the recent developments in genome sequenc-
Trüper HG, Dworkin M, Harder W, Schleifer K-H (eds) The prokaryotes.
ing technologies, however, which allow the inexpensive identi- A handbook on the biology of bacteria: ecophysiology, isolation, applica-
fication of all genes encoded by the genome of ecotypes, and the tions. Springer, New York, pp 2922–2937
availability of techniques such as nanoSIMS (Wagner 2009), Britten RJ, Kohne DE (1968) Repeated sequences in DNA. Science 161:529–540
which allow detailed measurements of many cell activities at Buchanan RE (1916) Studies in the nomenclature and classification
of the bacteria I. The problem of bacterial nomenclature. J Bacteriol
the individual cell level in situ, it might be possible to robustly
1:591–596
detect and characterize ecotypes in the not-too-distant future. Buchanan RE (1918) Studies in the nomenclature and classification of the
The recent technological developments in genomics and bacteria V. Subgroups and genera of the Bacteriaceae. J Bacteriol 3:27–61
metagenomics have propelled the microbial systematic and phy- Buckley M, Roberts RJ (2007) Reconciling microbial systematics and
logeny studies in a new era. We believe that it is within reach that genomics. Colloquium Report. The American Society for Microbiology,
Washington, DC
new robust conclusions about what is the relative importance of
Caro-Quintero A, Rodriguez-Castano GP, Konstantinidis KT (2009) Genomic
the genetic mechanisms generating and sustaining population insights into the convergence and pathogenicity factors of Campylobacter
diversity and whether or not there is a continuum of diversity as jejuni and Campylobacter coli species. J Bacteriol 191:5824–5831
opposed to discrete taxa will emerge soon. Until then, the Caro-Quintero A, Deng J, Auchtung J, Brettar I, Höfle MG, Klappenbach J,
current pragmatic species definition and the 16S rRNA gene- Konstantinidis KT (2011) Unprecedented levels of horizontal gene transfer
among spatially co-occurring Shewanella bacteria from the Baltic Sea. ISME J
based phylogeny remain reliable and useful means to guide
5:131–140
prokaryotic systematics.
Cavalier-Smith T (2002) The neomuran origin of archaebacteria, the Staley JR (eds) Bergey’s manual systematic bacteriology: the proteobacteria,
negibacterial root of the universal tree, and bacterial mega classification. 2nd edn. Springer, New York, pp 494–546
Int J Syst Evol Microbiol 52:7–76 Fowler VJ, Widdel F, Pfennig N, Stackebrandt E (1986) Phylogenetic relationships
Christensen H, Kuhnert P, Busse H-J, Frederiksen WC, Bisgaard M (2007) Pro- of sulfate- and sulfur-reducing eubacteria. Syst Appl Microbiol 8:32–41
posed minimal standards for the description of genera, species and subspe- Fox GE, Stackebrandt E, Hespell RB, Gibson J, Maniloff J, Dyer TA, Wolfe RS,
cies of the Pasteurellaceae. Int J Syst Evol Microbiol 57:166–178 Balch WE, Tanner R, Magrum L, Zablen L-B, Blakemore R, Gupta R, Lewis
Cicarelli FD, Doerks T, von Mering C, Creevey CJ, Snel B, Bork P (2006) Toward BJ, Stahl DA, Luehrsen R, Chen KN, Woese CR (1980) The phylogeny of
automatic reconstruction of a highly resolved tree of life. Science 311: prokaryotes. Science 209:457–463
1283–1287 Fox GE, Wisotzkey JD, Jurtshuk P Jr (1992) How close is close: 16S RNA sequence
Cohan FM (2002) What are bacterial species? Annu Rev Microbiol 56:457–487 identity may not be sufficient to guarantee species identity. Int J Syst
Cohan FM (2006) Towards a conceptual and operational union of bacterial Bacteriol 42:166–170
systematics, ecology, and evolution. Philos Trans R Soc Lond B Biol Sci Fraser C, Hanage WP, Spratt BG (2007) Recombination and the nature of
361:1985–1996 bacterial speciation. Science 315:476–480
Cohan FM, Koeppel AF (2008) The origins of ecological diversity in prokaryotes. Fraser C, Alm EJ, Polz MF, Spratt BG, Hanage WP (2009) The bacterial species
Curr Biol 18R:1024–1034 challenge: making sense of genetic and ecological diversity. Science 323:741–746
Cohn F (1872) Untersuchungen über Bacterien. Beitr Biol Pfl 2:127–224 Garrity GM, Holt JG (2001a) Class I. Deinococci class. nov. In: Boone DR,
Collins MD, Lawson PA, Willems A, Cordoba JJ, Fernandez-Garayzabal J, Castenholz RW, Garrity GM (eds) Bergey’s manual of systematic bacteriol-
Garcia P, Cai J, Hippe H, Farrow JAE (1994) The phylogeny of the genus ogy, vol 1, 2nd edn, The Archaea and the deeply branching and phototrophic
Clostridium: proposal of five new genera and eleven new species combina- bacteria. Springer, New York, p 395
tions. Int J Syst Bacteriol 44:812–826 Garrity GM, Holt JG (2001b) Phylum BVIII. Nitrospirae phy. nov. In: Boone DR,
Colwell RR (1970) Polyphasic taxonomy of the genus Vibrio: numerica taxonomy Castenholz RW, Garrity GM (eds) Bergey’s manual of systematic bacteriol-
of Vibrio cholerae, Vibrio parahaemolyticus, and related Vibrio species. ogy. Springer, New York, pp 451–464
J Bacteriol 104:410–433 Garrity GM, Holt JG (2001c) Phylum BVI. Chloroflexi phy. nov. In: Boone DR,
Cousin S, Brambilla E, Yang J, Stackebrandt E (2008) Culturable aerobic bacteria Castenholz RW, Garrity GM (eds) Bergey’s manual of systematic bacteriol-
from the upstream region of a karst water rivulet. Int Microbiol 11:91–100 ogy, vol 1, 2nd edn, The Archaea and the deeply branching and phototrophic
Cowan ST (1968) A dictionary of microbial taxonomic usage. Oliver & Boyd, bacteria. Springer, New York, pp 427–446
Edinburgh Garrity GM, Garrity GM, Bell JA, Lilburn T (2005) Phylum XIV. Proteobacteria
Cracraft J (1983) Species concept and speciation analysis. Current ornithology. phyl. nov. In: Brenner DJ, Krieg NR, Staley JT, Garrity GM (eds) Bergey’s
Plenum Press, New York, pp 159–187 manual of systematic bacteriology, vol 2, 2nd edn, The proteobacteria,
Darwin C (1859) On the origin of species. Murray, London part B: the gammaproteobacteria. Springer, New York, p 1
De Ley J (1970) Reexamination of the association between melting point, buoyant Gevers D, Cohan FM, Lawrence JG, Spratt BG, Coenye T, Feil EJ, Stackebrandt E,
density, and chemical base composition of deoxyribonucleic acid. J Bacteriol Van de Peer Y, Vandamme P, Thompson FL, Swings J (2005) Re-evaluating
101:738–754 prokaryotic species. Nat Rev Microbiol 3:733–739
DeLong EF, Preston CM, Mincer T, Rich V, Hallam SJ, Frigaard NU, Martinez A, Gharbia SE, Shah HN (1992) Fusobacterium nucleatum subsp. fusiforme
Sullivan MB, Edwards R, Brito BR, Chisholm SW, Karl DM (2006) Commu- subsp. nov. and Fusobacterium nucleatum subsp. animalis subsp. nov. as
nity genomics among stratified microbial assemblages in the ocean’s interior. additional subspecies within Fusobacterium nucleatum. Int J Syst Bacteriol
Science 311:496–503 42:296–298
Dobzhansky T (1937) Genetics and the origin of species. Columbia University Gibbons NE, Murray RGE (1978) Proposals concerning the higher taxa of
Press, New York bacteria. Int J Syst Bacteriol 28:1–6
Doolittle WF, Bapteste E (2007) Pattern pluralism and the tree of life hypothesis. Gibson J, Ludwig W, Stackebrandt E, Woese CR (1985) The phylogeny of the
Proc Natl Acad Sci USA 104:2043–2049 green photosynthetic bacteria: absence of a close relationship between
Doolittle WF, Zhaxybayeva O (2009) On the origin of prokaryotic species. Chlorobium and Chloroflexus. Syst Appl Microbiol 6:152–156
Genome Res 19:744–756 Gogarten JP, Townsend JP (2005) Horizontal gene transfer, genome innovation
Douglas AE (1998) Nutritional interactions in insect-microbial symbioses: aphids and evolution. Nat Rev Microbiol 3:679–687
and their symbiotic bacteria Buchnera. Annu Rev Entomol 43:17–37 Goris J, Suzuki K, De Vos P, Nakase T, Kersters K (1999) Evaluation of
Dykhuizen DE, Green L (1991) Recombination in Escherichia coli and the defi- a microplate DNA-DNA hybridization method compared to the initial
nition of biological species. J Bacteriol 173:7257–7268 renaturation method. Can J Microbiol 44:1148–1153
Dzink JL, Sheehan MT, Socransky SS (1990) Proposal of three subspecies of Goris J, Konstantinidis KT, Coenye T, Vandamme P, Tiedje JM (2007) DNA-DNA
Fusobacterium nucleatum Knorr 1922: Fusobacterium nucleatum subsp. hybridization values and their relation to whole genome sequence. Int J Syst
nucleatum subsp. nov., comb. nov.; Fusobacterium nucleatum subsp. Evol Microbiol 57:81–91
polymorphum subsp. nov., nom. rev., comb. nov.; and Fusobacterium Grimont PAD, Popoff MY, Grimont F, Coynault C, Lemelin M (1980) Reproduc-
nucleatum subsp. vincentii subsp. nov., nom. rev., comb. nov. Int J Syst ibility and correlation study of three deoxyribonucleic acid hybridization
Bacteriol 40:74–78 procedures. Curr Microbiol 4:325–330
Embley MT, Stackebrandt E (1994) The molecular phylogeny and systematics of Gupta RS, Golding GB (1993) Evolution of HSP70 gene and its implications
the Actinomycetes. Annu Rev Microbiol 48:257–289 regarding relationships between Archaebacteria, Eubacteria and Eukaryotes.
Eppley JM, Tyson GW, Getz WM, Banfield JF (2007) Genetic exchange across J Mol Evol 37:573–582
a species boundary in the archaeal genus Ferroplasma. Genetics 177:407–416 Gürtler V, Mayall BC (2001) Genomic approaches to typing, taxonomy and
Ezaki T, Hashimoto Y, Yabuuchi E (1989) Fluorimetric deoxyribonucleic acid- evolution of bacterial isolates. Int J Syst Evol Microbiol 51:3–16
deoxyribonucleic acid hybridization in microdilution wells as an alternative Guttman DS, Dykhuizen DE (1994) Clonal divergence in Escherichia coli as
to membrane filter hybridization in which radioisotopes are used to deter- a result of recombination, not mutation. Science 266:1380–1383
mine genetic relatedness among bacterial strains. Int J Syst Bacteriol 39: Handelsman J, Tiedje J, Alvarez-Cohen L, Ashburner M, Cann I, Delong E,
224–229 Doolittle W, Fraser-Ligget C, Godzik A, Gordon J, Riley M, Schmidt T
Farelly V, Rainey FA, Stackebrandt E (1995) Effect of genomic size and rrn gene (2007) The news of metagenomics: revealing the secrets of our microbial
copy number on PCR amplification of 16S rRNA genes from a mixture of planet. The National Academies Press, Washington, DC
bacterial species. Appl Environ Microbiol 61:2798–2801 Hara T, Shimoda T, Nonaka K, Ogata S (1991) Colorimetric detection of DNA-
Farmer JJ III, Janda JM, Brenner FW, Cameron DN, Birkhead KM DNA hybridization in microdilution wells for taxonomic application on
(2005) Genus 1. Vibrio Pacini 1854, 411AL. In: Brenner DJ, Krieg NR, bacteria strains. J Ferment Bioeng 72:122–124
Harris LG, El-Bouri K, Johnston S, Rees E, Frommelt L, Siemssen N, Christner M, Konstantinidis KT, Tiedje JM (2005b) Genomic insights that advance the species
Davies AP, Rohde H, Mack D (2010) Rapid identification of staphylococci concept for prokaryotes. Proc Natl Acad Sci USA 102:2567–2572
from prosthetic joint infections using MALDI-TOF mass-spectrometry. Int Konstantinidis KT, Tiedje JM (2007) Prokaryotic taxonomy and phylogeny in the
J Artif Organs 33:568–574 genomic era: advancements and challenges ahead. Curr Opin Microbiol
He Y, Li H, Lu X, Stratton CW, Tang YW (2010) Mass spectrometry biotyper 105:504–509
system identifies enteric bacterial pathogens directly from colonies grown on Konstantinidis KT, Ramette A, Tiedje JM (2006) Toward a more robust assess-
selective stool culture media. J Clin Microbiol 48:3888–3892 ment of intra-species diversity using fewer genetic markers. Appl Environ
Holt JG, Krieg NR, Sneath PHA, Staley JT, Williams ST (1994) Bergey’s manual of Microbiol 72:7286–7293
determinative bacteriology, 9th edn. Williams & Wilkins, Baltimore Konstantinidis KT, Braff J, Karl DM, DeLong EF (2009) Comparative
Hsieh S-Y, Tseng C-L, Lee Y-S, Kuo A-J, Sun C-F, Lin Y-H, Chen J-K (2008) Highly metagenomic analysis of a microbial community residing at a depth of
efficient classification and identification of human pathogenic bacteria by 4,000 meters at station ALOHA in the North Pacific subtropical gyre. Appl
MALDI-TOF MS. Mol Cell Proteomics 7:448–456 Environ Microbiol 75:5345–5355
Hull D (1976) Are species really individuals? Syst Zool 25:174–191 Kosakovsky Pond SL, Posada D, Gravenor MB, Woelk CH, Frost SD (2006) Auto-
Huss VAR, Festl H, Schleifer KH (1983) Studies on the spectrophotometric mated phylogenetic detection of recombination using a genetic algorithm.
determination of DNA hybridization from renaturation rates. Syst Appl Mol Biol Evol 23:1891–1901
Microbiol 4:184–192 Lapage SP, Sneath PHA, Lessel EF, Skerman VBD, Seeliger HPR, Clark WA (eds)
Huys G, Kämpfer P, Albert MJ, Kuhn I, Denys R, Swings J (2002) Aeromonas (1975) International code of nomenclature of bacteria (1975 revision).
hydrophila subsp. dhakensis subsp. nov., isolated from children with diar- American Society for Microbiology, Washington, DC
rhoea in Bangladesh, and extended description of Aeromonas hydrophila Lawrence JG (2002) Gene transfer in bacteria: speciation without species? Theor
subsp. hydrophila Chester 1901 Stanier 1943 Approved Lists 1980. Int J Syst Popul Biol 61:449–460
Evol Microbiol 52:705–712 Lawrence JG, Ochman H (1998) Molecular archaeology of the Escherichia coli
Iino T, Mori K, Uchino Y, Nakagawa T, Harayama S, Suzuki KI (2010) genome. Proc Natl Acad Sci USA 95:9413–9417
Ignavibacterium album gen. nov., sp. nov., a moderately thermophilic Lawrence JG, Retchless AC (2009) The interplay of homologous recombination
anaerobic bacterium isolated from microbial mats at a terrestrial hot and horizontal gene transfer in bacterial speciation. Methods Mol Biol
spring and proposal of Ignavibacteria classis nov., for a novel lineage at 532:29–53
the periphery of green sulfur bacteria. Int J Syst Evol Microbiol 60: Lay JO Jr (2001) MALDI-TOF mass spectrometry of bacteria. Mass Spectrom Rev
1376–1382 20(4):172–194
Imhoff JF, Süling J, Petri R (1998a) Phylogenetic relationships among the Lehmann KB, Neumann R (1896) Atlas und Grundriss der Bakteriologie und
Chromatiaceae, their taxomonic reclassification and description of the new Lehrbuch der speziellen Bakteriologischen Diagnostik, 1st edn. Munich J.F,
genera Allochromatium, Halochromatium, Isochromatium, Marichromatium, Lehmann
Thiococcus, Thiohalocapsa and Thermochromatium. Int J Syst Bacteriol Lester CH, Sandvang D, Olsen SS, Schønheyder HC, Jarløv JO, Bangsborg J,
48:1129–1143 Hansen DS, Jensen TG, Frimodt-Møller N, Hammerum AM, Lester CH,
Imhoff JF, Petri R, Süling J (1998b) Reclassification of species of the spiral-shaped Sandvang D, Olsen SS, Schønheyder HC, Jarløv JO, Bangsborg J, Hansen DS,
phototrophic purple non-sulfur bacteria of the g-Proteobacteria: description Jensen TG, Frimodt-Møller N, Hammerum AM, DANRES Study Group
of the genera Phaeospirillum gen. nov., Rhodovibrio gen. nov., (2008) Emergence of ampicillin-resistant Enterococcus faecium in Danish
Rhodothalassium gen. nov. and Roseospiragen nov., as well as transfer of hospitals. J Antimicrob Chemother 62:1203–1206
Rhodospirillum fulvum to Phaeospirillum fulvum comb. nov., of Lilburn TG, Harrison SH, Cole JR, Garrity GM (2006) Computational aspects of
Rhodospirillum molischianum to Phaeospirillum molischianum comb. nov., systematic biology. Brief Bioinform 7:186–195
of Rhodospirillum salinarum to Rhodovibrion salinarum comb. nov., of Lind E, Ursing J (1986) Clinical strains of Enterobacter aggomeruns synonyms:
Rhodospirillum sodomense to Rhodovibrio sodomensis comb. nov., of Erwinia herbicola,Envinia milletiae identified by DNA-DNA-hybridization.
Rhodospirillum salexigens to Rhodothalassium salexigens comb. nov. and of Acta Pathol Microbiol Scand B Microbiol Immunol 94:205–213
Rhodospirillum medosalinum to Roseospira mediosalina. Int J Syst Bacteriol Linnaeus C (1753) Species plantarum. Holmiae
48:793–798 Maiden CJM, Dingle KC (2008) Campylobacter. In: Nachamkin I, Szymanski
Istock CA, Bell JA, Ferguson N, Istock NL (1996) Bacterial species and evolution: CM, Blaser MJ (eds) Population biology of Campylobacter jejuni and related
theoretical and practical perspectives. J Ind Microbiol 17:137–150 organisms, 3rd edn. American Society for Microbiology Press, Washington,
Johnson JL (1973) The use of nucleic acid homologies in the taxonomy of DC, pp 27–40
anaerobic bacteria. Int J Syst Bacteriol 23:308–315 Maiden MC, Bygraves JA, Feil E, Morelli G, Russell JE, Urwin R, Zhang Q, Zhou J,
Kandler O, König H (1985) Cell envelopes of archaebacteria. In: Woese CR, Zurth K, Caugant DA, Feavers IM, Achtman M, Spratt BG (1996) Multilocus
Wolfe RS (eds) Archaebacteria (The bacteria. A treatise on structure and sequence typing: a portable approach to the identification of clones within
function), vol VIII. Academic, New York, pp 413–457 populations of pathogenic microorganisms. Proc Natl Acad Sci USA
Kaznowski A (1995) A method of colorimetric DNA-DNA hybridization in 95:3140–3145
microplates with covalently immobilized DNA for identification of Margulis L (1981) Symbiosis in cell evolution. W.H. Freeman, San Francisco
Aeromonas sp. Med Microbiol Lett 4:362–369 Maynart-Smith J, Smith NH, O’Rourke M, Spratt BG (1993) How clonal are
Klenk HP, Zillig W (1994) DNA-dependent RNA polymerase subunit B as a tool bacteria? Proc Natl Acad Sci USA 90:4384–4388
for phylogenetic reconstructions: branching topology of the archaeal McVean G, Awadalla P, Fearnhead P (2002) A coalescent-based method for
domain. J Mol Evol 38:4420–4432 detecting and estimating recombination from gene sequences. Genetics
Kloos WE, Mohapatra N, Dobrogosz J, Ezell JW, Manclark CR (1981) Deoxyri- 160:1231–1241
bonucleotide sequence relatioships among Bordetella species. Int J Syst Migula W (1900) System der Bakterien. Gustav Fischer, Jena
Bacteriol 31:173–176 Miller SR, Augustine S, Olson TL, Blankenship RE, Selker J, Wood AM (2005) Dis-
Kluyver AJ, van Niel CB (1936) Prospects for a natural system of classification of covery of a free-living chlorophyll d-producing cyanobacterium with
bacteria. Zbl Bakt Abt 2(94):369–403 a hybrid proteobacterial/cyanobacterial small-subunit rRNA gene. Proc
Konstantinidis KT, DeLong EF (2008) Genomic patterns of recombination, clonal Natl Acad Sci USA 102:850–855
divergence and environment in marine microbial populations. ISME J Moran NA (2007) Symbiosis as an adaptive process and source of phenotypic
2:1052–1065 complexity. Proc Natl Acad Sci USA 104:8627–8633
Konstantinidis KT, Tiedje JM (2005a) Towards a genome-based taxonomy for Moreno E (1997) In search of a bacterial species definition. Rev Biol Trop
prokaryotes. J Bacteriol 187:6258–6264 45:753–771
Naser SM, Thompson FL, Hoste B, Gevers D, Dawyndt P, Vancanneyt M, Swings J Scheldeman P, Goossens K, Rodriguez-Diaz M, Pil A, Goris J, Herman L, De Vos P,
(2005) Application of multilocus sequence analysis MLSA for rapid identi- Logan NA, Heyndrickx M (2004) Paenibacillus lactis sp. nov. isolated from
fication of Enterococcus species based on rpoA and pheS genes. Microbiol raw and heat-treated milk. Int J Syst Evol Microbiol 54:885–891
(Reading) 151:2141–2150 Schleifer K-H, Kandler O (1972) The peptidoglycan types of bacterial cell walls
O‘Hara RJ (1994) Evolutionary history and the species problem. Amer Zool and their taxonomic implications. Bacteriol Rev 36:407–477
34:12–22 Schleifer K-H, Stackebrandt E (1983) Molecular systematics of prokaryotes. Annu
Orla-Jensen S (1909) Die Hauptlinien des natürlichen Bakteriensystems. Zbl Bakt Rev Microbiol 37:143–187
Abt 2(22):305–346 Schleifer K-H, Leuteritz M, Weiss N, Ludwig W, Kirchhof G, Seidel-Rüfer H
Peix A, Valverde A, Rivas R, Igual JM, Ramrez-Bahena M-H, Mateos PF, (1990) Taxonomic study of anaerobic, Gram-negative, rod-shaped bacteria
Santa-Regina I, Rodrguez-Barrueco C, Martnez-Molina E, Velazquez from breweries: emended description of Pectinatus cerevisiiphilus and
C (2007) Reclassification of Pseudomonas aurantiaca as a synonym of Pseu- description of Pectinatus frisingensis sp. nov., Selenomonas lacticifex sp.
domonas chlororaphis and proposal of three subspecies, P. chlororaphis subsp. nov., Zymophilus raffinosivorans gen. nov., sp. nov., and Zymophilus
chlororaphis subsp. nov., P. chlororaphis subsp. aureofaciens subsp. nov., paucivorans sp. nov. Int J Syst Bacteriol 40:19–27
comb. nov. and P. chlororaphis subsp. aurantiaca subsp. nov., comb. nov. Schroeter J (1886) Microspira comma. In: Cohn F (ed) Kryprogamen-Flora von
Int J Syst Evol Microbiol 57:1286–1290 Schlesien 3:168
Prévot AR (1938) Études de systématique bactérienne. Ann Inst Pasteur Seewaldt E, Schleifer K-H, Bock E, Stackebrandt E (1982) The close phylogenetic
60:285–307 relationship of Nitrobacter and Rhodopseudomonas palustris. Arch Microbiol
Pringsheim EG (1923) Zur Kritik der Bakteriensystematik. Lotus 71:357–377 131:287–290
Prod’hom G, Bizzini A, Durussel C, Bille J, Gilbert Greub G (2011) MALDI-TOF Selander RK, Li J, Boyd F, Wang F-S, Nelson K (1994) DNA sequence analysis of
mass spectrometry for direct bacterial identification from positive ammo- the genetic structure of populations of Salmonella enterica and Escherichia
nium chloride lysed blood culture pellets. J Clin Microbiol. doi:10.1128/ coli. In: Priest FG, Ramos-Cormenzana A, Tindall R (eds) Bacterial system-
JCM.01780-09 atics and diversity. Plenum Press, New York, pp 17–50
Rademaker JLW, Hoste B, Louws FJ, Kersters K, Swings J, Vauterin L, Vauterin P, Shah HN, Keys CJ, Schmid O, Gharbia SE (2002) Matrix-assisted laser desorp-
de Bruijn FJ (2000) Comparison of AFLP and rep-PCR genomic fingerprint- tion/ionisation time of flight mass spectrometry and proteomics; a new era
ing with DNA–DNA homology studies: Xanthomonas as a model system. Int in anaerobic microbiology. Clin Infect Dis 35:58–64
J Syst Evol Microbiol 50:665–677 Sibley CG, Comstock JA, Ahlquist JE (1990) DNA hybridization evidence of
Rainey FA, Stackebrandt E (1993) Phylogenetic evidence for the relationship of homoid phylogeny: a reanalysis of the data. J Mol Evol 30:202–236
Saccharococcus thermophilus to Bacillus thermoglucosidasius, Bacillus Sneath PAH (1989) Analysis and interpretation of sequence data for bacterial
kaustophilus and Bacillus stearothermophilus. Syst Appl Microbiol systematics: the view of a numerical taxonomist. Syst Appl Microbiol
16:224–226 12:15–31
Relman DA, Strauss E (1999) Microbial genomes: blueprint for life. American Stackebrandt E (1992) The prokaryotes. a handbook on the biology of bacteria:
Society for Microbiology Press, Washington, DC ecophysiology, isolation, identification, applications. In: Balows A, Trüper
Reysenbach A-L (2001) Phylum BI. Aquificae phy. nov. In: Boone DR, Castenholz HG, Dworkin M, Harder W, Schleifer K-H (eds) Unifying phylogeny and
RW, Garrity GM (eds) Bergey’s manual of systematic bacteriology, vol 1, 2nd phenotypic properties. Springer, New York, pp 19–47
edn, The Archaea and the deeply branching and phototrophic bacteria. Stackebrandt E (2000) The prokaryotes: an evolving electronic resource for the
Springer, New York, pp 359–367 microbiological community. In: Dworkin M, Falkow S, Rosenberg E,
Richter M, Rossello-Mora R (2009) Shifting the genomic gold standard for Schleifer KH, Stackebrandt E (eds) Defining taxonomic ranks. Springer,
the prokaryotic species definition. Proc Natl Acad Sci USA 106: New York, pp 29–57
19126–19131 Stackebrandt E (2010) Diversification and focusing: strategies of microbial cul-
Riley M, Buckley M (2008) Large-scale sequencing: the future of genomic sci- ture collections. Trends Microbiol 18:283–287
ences? American Society for Microbiology Press, Washington, DC Stackebrandt E, Ebers J (2006) Taxonomic parameters revisited: tarnished gold
Rodas AM, Ferrer S, Pardo I (2005) Polyphasic study of wine Lactobacillus strains: standards. Microbiol Today 33:152–155
taxonomic implications. Int J Syst Evol Microbiol 55:197–207 Stackebrandt E, Goebel BM (1994) A place for DNA-DNA reassociation and 16S
Rong X, Huang Y (2010) Taxonomic evaluation of the Streptomyces griseus clade rRNA sequence analysis in the present species definition in bacteriology. Int
using multilocus sequence analysis and DNA-DNA hybridization, with pro- J Syst Bacteriol 44:846–849
posal to combine 29 species and three subspecies as 11 genomic species. Int Stackebrandt E, Rainey FA (1997) Phylogenetic relationships. In: Rood IJ,
J Syst Evol Microbiol 60:696–703 McClane BA, Songer JB, Titball RW (eds) The clostridia: molecular biology
Rong X, Guop Y, Huag Y (2009) Proposal to reclassify the Streptomyces and pathogenesis. Academic, New York, pp 3–19
albidoflavus clade on the basis of multilocus sequence analysis and DNA- Stackebrandt E, Woese CR (1984) The phylogeny of prokaryotes. Microbiol Sci
DNA hybridization, and taxonomic elucidation of Streptomyces griseus 1:117–122
subsp. solvifaciens. Syst Appl Microbiol 32:314–322 Stackebrandt E (1985) The significance of "wall types" in phylogenetically based
Rossello-Mora R (2005) Updating prokaryotic taxonomy. J Bacteriol 187: taxonomic studies on actinomycetes. In: Szabó G, Biró S. Goodfellow M
6255–6257 (eds) Sixth Int Symp on Actinomycete Biology. Akademiai Kiado, Budapest,
Rosselló-Móra R (2006) DNA-DNA reassociation methods applied to microbial pp 497–606
taxonomy and their critical evaluation. In: Stackebrandt E (ed) Molecular Stackebrandt E, Embley M, Weckesser J (1988) Phylogenetic, evolutionary, and
identification, systematics, and population structure of prokaryotes. taxonomic aspects of phototrophic eubacteria. In: Olson JM, Ormerod JG,
Springer, Berlin, pp 23–50 Amesz J, Stackebrandt E, Trüper HG (eds) Green photosynthetic bacteria.
Rosselló-Mora R, Amann R (2001) The species concept for prokaryotes. FEMS Plenum Press, New York, pp 201–215
Microbiol Rev 25:39–67 Stackebrandt E, Tindall B, Ludwig W, Goodfellow M (1999) Prokaryotic diversity
Rothschild LJ, Ragan A, Coleman AW, Heywood P, Gerbi SA (1986) and systematics. In: Lengeler J, Drews G, Schlegel HG (eds) Biology of the
Are rRNA sequence comparisons the Rosetta stone of phylogeny? Cell prokaryotes. Thieme, Stuttgart, pp 675–720
47(5):640 Stackebrandt E, Frederiksen W, Garrity GM, Grimont PAD, Kämpfer P,
Rusch DB, Halpern AL, Sutton G, Heidelberg KB, Williamson S, Yooseph S, Maiden MC, Nesme X, Rossello-Mora R, Swings J, Trüper HG, Vauterin L,
Wu D, Eisen JA, Hoffman JM, Remington K et al (2007) The sorcerer II Ward AC, Whitman WB (2002) Report of the ad hoc committee for the re-
global ocean sampling expedition: Northwest Atlantic through Eastern evaluation of the species definition in bacteriology. Int J Syst Evol Microbiol
Tropical Pacific. PLoS Biol 5:e77 52:1043–1052
Stackebrandt E, Päuker O, Erhard M (2005) Grouping myxococci (Corallococcus) Wagner M (2009) Single-cell ecophysiology of microbes as revealed by Raman
strains by matrix-assisted laser desorption ionization time-of-flight MALDI microspectroscopy or secondary ion mass spectrometry imaging. Annu Rev
TOF mass spectrometry: comparison with gene sequence phylogenies. Curr Microbiol 63:411–429
Microbiol 50:71–77 Ward DM (1998) A natural species concept for prokaryotes. Curr Opin Microbiol
Staley JT (1997) Biodiversity: are microbial species threatened? Curr Opin 1:271–277
Biotechnol 8:340–345 Ward DM (2006) A macrobiological perspective on microbial species. Microbe
Staley J (2009) The phylogenomics species concept. Microbiol Today 36:82–83 1:269–278
Staley JT, Krieg NR (1984) Bacterial classification I. Classification of prokaryotic Walker DH (1989) Rocky Mountain spotted fever: a new disease in need of
organisms: an overview. In: Williams ST, Sharpe ME, Holt JG (eds) Bergey’s microbiological concern. Clin Microbiol Rev 2:227–240
manual of systematic bacteriology. Williams & Wilkins, Baltimore, pp 1–4 Wayne L, Brenner DJ, Colwell RR, Grimont PAD, Kandler O, Krichevsky MI,
Stanier RY, van Niel CB (1941) The main outlines of bacterial classification. Moore LH, Moore WEC, Murray RGE, Stackebrandt E, Starr MP, Trüper HG
J Bacteriol 42:437–466 (1987) International committee on systematic bacteriology: report of the ad
Steigerwalt AG, Fanning GR, Fife-Ashbury MA, Brenner DJ (1976) DNA related- hoc committee on reconciliation of approaches to bacterial systematics. Int
ness among species of Enterobacter and Serratia. Can J Microbiol 22:121–137 J Syst Bacteriol 37:463–464
Szabados F, Woloszyn J, Richter C, Kaase M, Gatermann SG (2010) A Biotyper 2.0 Winslow C-EA, Broadhurst J, Buchanan RE (1920) The families and genera of the
Bruker Daltonics-based score cut-off of actually 2 is superior to molecular bacteria. J Bacteriol 3:191–229
Staphylococcus aureus identification. J Med Microbiol 59:787–790 Woese CR (1987) Bacterial evolution. Microbiol Rev 51:221–271
Thane PR (2009) A critique of prokaryotic species concepts. Methods Mol Biol Woese CR, Debrunner-Vossbrinck B, Oyaizu H, Stackebrandt E, Ludwig W
532:379–395 (1985) Gram-positive bacteria: possible photosynthetic ancestry. Science
Thompson CC, Thompson FL, Vicente ACP (2008) Identification of Vibrio 229:762–765
cholerae and Vibrio mimicus by multilocus sequence analysis MLSA. Int Woese CR, Kandler O, Wheelis ML (1990) Towards a natural system of organisms:
J Syst Evol Microbiol 58:617–621 proposal for the domains archaea, bacteria and eucaryas. Proc Natl Acad Sci
Tindall BJ, Rosselló-Móra R, Busse H-J, Ludwig W, Kämpfer P (2010) Notes on USA 87:4576–4579
the characterization of prokaryote strains for taxonomic purposes. Int J Syst Wolf YI, Rogozin IB, Grishin NV, Tatusov RL, Koonin EV (2001) Genome trees
Evol Microbiol 60:249–266 constructed using five different approaches suggest new major bacterial
Turenne CY, Collins DM, Alexander DC, Behr MA (2008) Mycobacterium avium clades. BMC Evol Biol 1:8–28
subsp. paratuberculosis and M. avium subsp. avium are independently Yanagida F, Suzuki K-I, Kozaki M, Komagata K (1997) Proposal of Sporolacto-
evolved pathogenic clones of a much broader group of M. avium organisms. bacillus nakayamae subsp. nakayamae sp. nov., subsp. nov., Sporolacto-
J Bacteriol 190:2479–2487 bacillus nakayamae subsp. racemicus subsp. nov., Sporolactobacillus terrae
Tyson GW, Chapman J, Hugenholtz P, Allen EE, Ram RJ, Richardson PM, sp. nov., Sporolactobacillus kofuensis sp. nov., and Sporolactobacillus lactosus
Solovyev VV, Rubin EM, Rokhsar DS, Banfield JF (2004) Community struc- sp. nov. Int J Syst Bacteriol 47:499–504
ture and metabolism through reconstruction of microbial genomes from the Zawadzki P, Roberts MS, Cohan FM (1995) The log-linear relationship between
environment. Nature 428:37–43 sexual isolation and sequence divergence in Bacillus transformation is
Ullman JS, McCarthy BJ (1973) The relationship between mismatched base pairs robust. Genetics 140:917–932
and the thermal stability of DNA duplexes. Biochim Biophys Acta 294:416–424 Zhu Y, Fournier PE, Eremeeva M, Raoult D (2005) Proposal to create subspecies
van Baar BL (2000) Characterisation of bacteria by matrix-assisted laser desorp- of Rickettsia conorii based on multi-locus sequence typing and an emended
tion/ionisation and electrospray mass spectrometry. FEMS Microbiol Rev description of Rickettsia conorii. BMC Microbiol 5:11. doi:10.1186/1471-
24:193–219 2180-5-11
van Berkum P, Terefework Z, Paulin L, Suomalainen S, Lindström K, Eardly BD Ziemke F, Höfle MG, Lalucat J, Rosello-Mora R (1998) Reclassification of Owen’s
(2003) Discordant phylogenies within the rrnl of rhizobia. J Bacteriol genomic group II as Shewanella baltica sp. nov. Int J Syst Bacteriol
185:2988–2998 48:179–186
Venter JC, Remington K, Heidelberg JF, Halpern AL, Rusch D et al (2004) Environ- Zuckerkandl E, Pauling L (1965) Molecules as documents of evolutionary history.
mental genome shotgun sequencing of the Sargasso sea. Science 304:66–74 J Theor Biol 8:357–366
10 Population Genetics
Santiago Castillo-Ramı́rez . Edward J. Feil
Department of Biology and Biochemistry, University of Bath, Bath, UK
Introduction and Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 was of immediate practical benefit for population biology and
epidemiological studies, in that it enabled primer design for any
The Population Structure of Staphylococcus aureus . . . . . . 256 gene present on the sequenced strain. The contemporaneous
advances of capillary-based automated sequencers and the abil-
Zooming in on Single S. aureus Clones: ST5 . . . . . . . . . . . . . . 258 ity to maintain and readily access databases on the Internet
provided the means not only to sequence several gene loci
Zooming in on Single S. aureus Clones: ST239 . . . . . . . . . . . 258 for large population samples (100s of isolates), but also to
instantly interrogate these data from anywhere in the world.
A New Dawn for Next-Generation Sequencing . . . . . . . . . . . . 259 The development of Multilocus Sequence Typing (MLST)
in 1998 aimed to exploit these advances for pathogenic
A Paucity of Homoplasies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 bacteria (Maiden et al. 1998; Maiden 2006). Although clear
sampling biases have meant that the dual questions of epidemi-
Dating with Confidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 ological surveillance and population biology have, at times, been
uncomfortable bedfellows (Maiden 2006), these datasets have
The Purging of Deleterious Mutations Over Time . . . . . . . . 261 significantly impacted on both fronts.
Microarray technology represented another important
Synthesizing Selection and Ecology Using Next-Generation advance. Combined with the rapidly increasing genomics
Sequence Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 datasets, this approach confirmed a fundamental duality within
bacterial chromosomes, the stable ‘‘core genome’’ versus the
Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 dynamic ‘‘accessory’’ or ‘‘non-core’’ genome (Feil 2004). For
any given taxon (e.g., named species), the core genome is defined
simply as all those genes universally present in all strains. In
Introduction and Background contrast, the non-core genome is composed of genes variably
present or absent. Microarray data revealed that the non-core
Molecular data come in waves. The first nucleotide-sequence- genome is not randomly interspersed throughout the chromo-
based population studies on bacteria were published in the some, but is largely confined to regions of high gene content
late 1980s and early 1990s and typically focused on one or two variability, and that the locations of these variable regions are the
gene sequences from around a dozen bacterial isolates (DuBose same in all strains (Alm and Trust 1999; Dorrell et al. 2001;
et al. 1988; Milkman and Bridges 1990; Nelson et al. 1991; Fitzgerald et al. 2001). Complete genome sequencing revealed
Nelson and Selander 1992, 1994; Boyd et al. 1994; ). Despite a second key feature of the non-core genome, that of atypical
the modest size of these datasets by contemporary standards, base composition (GC content), GC skew, and codon bias,
these early works laid the foundation for a number of interre- which implicates extrachromosomal origins and the role of
lated debates: (1) the extent to which bacterial evolution is tree- mobile genetic elements (Ochman et al. 2000). A useful analogy
like or network-like, (2) the search for a meaningful ‘‘species’’ is to consider the core genome as the cell’s ‘‘operating system’’
concept, (3) biogeography and the extent to which ‘‘everything is (encoding housekeeping metabolism and DNA processing), and
everywhere,’’ and (4) the roles of selection (ecological adapta- the non-core genome as software modules providing specialized
tion) and neutrality (drift) in shaping population structure. functionality (> Fig. 10.1). Thus, the non-core genome can play
More specifically, data generated during the late 1980s and a major phenotypic role (e.g., by conferring antibiotic resistance
early 1990s provided unequivocal evidence that at least some or virulence traits) (Holden et al. 2004a), but the rapidity by
species experience high rates of genome-wide homologous which such genes can be gained or lost may make these accessory
recombination. These early data also brought into sharp relief genes poor markers for reconstructing evolutionary history
the importance (and difficulties) in assembling representative (Turner and Feil 2007).
population samples in bacteria (Smith et al. 1993; Fleischmann Molecular typing protocols such as MLST represent prag-
et al. 1995). matic trade-offs between inter-strain discrimination, the ame-
The generation of complete genome sequences for major nability of the data to evolutionary analyses, and the time and
human pathogens from 1995 onward (Fleischmann et al. 1995) expense required to characterize large population samples. The
256 10 Population Genetics
“Core” genome: “Non-core” (Accessory) genome:

e.g.: transcription, translation e.g. pathogenicity (toxins, adhesins),
(“Informational” genes), antibiotic resistance,
cell envelope, secretion systems and effectors,
key metabolic pathways restriction / modification,
(“housekeeping genes”), secondary metabolism
DNA replication
. Fig. 10.1
The duality of the bacterial genome into core and accessory genes. The ‘‘accessory’’ genes are acquired by horizontal transfer, a process
which usually involves a vector such as a plasmid or phage. Such genes can help the recipient bacterium to exploit new niches; thus
this model supports a saltational view of bacterial evolution
optimal solution depends both on the taxon being considered, techniques based on hypervariable repeat regions have been
and on the aims of the study; there is no universally ‘‘correct’’ established for some monomorphic species such as Mycobacte-
typing method. For MLSTa small sample of core genes (typically rium sp.; it is not always clear to what extent the patterns of
seven) are sequenced for each isolate, and each strain is defined divergence in the repeat loci are homoplasious, or conflict in
on the basis of the combined alleles observed over all loci. Freely some other way with the rare single nucleotide polymorphisms
available databases house MLST data for thousands of isolates (SNPs) present elsewhere in the genome (Smith et al. 2003,
for many major pathogens (Chan et al. 2001). While this 2006).
approach has been very successful for large-scale epidemiologi- Full genome sequencing will, of course, provide ultimate
cal surveillance and for understanding migration rates and discriminatory power, but until recently it has not been possible
evolutionary dynamics, it has two serious drawbacks. First, as to generate genome-wide datasets for large population samples.
MLST is based exclusively on a small sample of core genes, it is Latest advances in next-generation sequencing technology are
blind to large-scale genome rearrangements and changes in the set to close this technology gap. Platforms, such as the Illumina
non-core genome. A striking example of this is provided Genome Analyzer (IGA), Roche 454, and SOLiD, provide the
by a comparison between two related species: Burkholderia means to identify genome-wide SNPs for large population sam-
pseudomallei (an opportunistic pathogen of man) and B. ples. Below we discuss how this technology has been recently
mallei (an equine pathogen). While phylogenetic analysis applied to the important pathogen Staphylococcus aureus, and
based on the MLST data suggested that B. mallei represents how the high discriminatory power of the data can be used in an
a specialized ‘‘clone’’ nested within the B. pseudomallei popula- epidemiological setting. We will go on to emphasize how these
tion (Gevers et al. 2005), comparative genomics revealed that the data might also provide information on more fundamental
former had undergone extensive genome degradation (the dele- questions like how ecological (or epidemiological) specialisms
tion of around 1 Mb of DNA) and revealed a complete loss of of specific clones might impact on short-term molecular
synteny with B. pseudomallei (Holden et al. 2004b). These evolution.
dramatic changes reveal much about the ecological and molec-
ular dynamics underpinning the emergence of B. mallei, but
would have been completely missed on the basis of MLST data The Population Structure of Staphylococcus
alone. aureus
The second drawback of MLST is that a small sample of
seven (or so) core genes provides insufficient discrimination for S. aureus is a low-GC Gram-positive bacterium which asymp-
highly monomorphic species (recently reviewed in (Achtman tomatically colonizes the skin and anterior nares (nostrils) of
2008)) or, similarly, for single clonal complexes within more approximately one third of the human population at any given
diverse species. Cases where a single ‘‘type’’ predominates point in time (Peacock et al. 2001; Nulens et al. 2005; van
on a local scale present a pertinent problem for epidemiological Belkum 2006). The species is also commonly recovered from
surveillance, as the majority of isolates recovered are domesticated animals, particularly cows, pigs, and chickens
indistinguishable by standard procedures. Alternative (Vanderhaeghen et al. 2010), though the true ecological range
Population Genetics 10 257
. Fig. 10.2
Simplified clonal population structure of S. aureus visualized using MLST/eBURST (www.mlst.net/eburst.mlst.net). Each circle represents
a unique MLST haplotype. The size of the circle represents the frequency of the haplotype. Haplotypes differing by only one of the
seven loci are linked. The vast majority of isolates correspond to a small number of haplotypes and clusters of haplotypes. Distances,
angles, and positions have no meaning
in wild animals and the environment is not known. Infection by consistently delimit the S. aureus population in to the same
S. aureus can cause a number of conditions ranging in severity discrete clusters, or clonal complexes. These can be visualized
from boils to life-threatening endocarditis. Serious infections are using UPGMA dendrograms or a simple clustering algorithm
much more common within health-care settings, such as hospi- called BURST, implemented as the freely available eBURST (Feil
tals and nursing homes, where disease management is substan- et al. 2004) (> Fig. 10.2) or goeBURST (Francisco et al. 2009). As
tially hampered by the spread of resistance to b-lactam it is based on nucleotide sequence data, MLST data is also
antibiotics (initially penicillin, then methicillin-resistant amenable to phylogenetic analysis. These analyses have also
S. aureus; MRSA). Resistance to methicillin is conferred via the revealed that hospital-acquired (HA-) MRSA isolates are partic-
horizontal acquisition of a large (20–60 Kb) chromosomal cas- ularly clonal, with a very small number of clusters accounting for
sette (SCCmec), which is thought to have been introduced from almost all the cases of infection worldwide(Crisostomo et al.
naturally resistance commensal staphylococcal species on mul- 2001; Aires de Sousa et al. 2005; Gomes et al. 2006; Conceicao
tiple occasions (Enright et al. 2002). In recent years, sporadic et al. 2007). Methicillin-sensitive S. aureus (MSSA) are more
infection by MRSA has become more common in the commu- diverse, as are community-acquired (CA-) MRSA isolates. This
nity, outside of health-care settings (Deleo et al. 2010). has led to the view that HA-MRSA strains, which are more likely
Several typing methods have been used for epidemiological to be multiply resistant than CA-MRSA strains and are relatively
surveillance of this species, including pulse-field gel electropho- rare outside of health-care settings, are specifically adapted to
resis (PFGE), MLST (Cookson et al. 2007), and MLVA (multiple the hospital environment.
loci VNTR analysis, where VNTR stands for variable number of S. aureus experiences homologous recombination at rela-
tandem repeats) (Melles et al. 2009; Schouls et al. 2009). VNTR tively low frequency. This is thought to explain in part why
loci are hypervariable microsatellites, the most notable example discrete sub-clusters have emerged and been maintained in
in S. aureus being the spa gene which has been used extensively the population (Feil et al. 2003), and why it is possible to
for typing studies in this species (Mellmann et al. 2008; Basset reconstruct relatively robust phylogenies (Cooper and Feil
et al. 2009). While these methods present a range of utility for 2006), at least in comparison to the naturally transformable
more detailed evolutionary analyses, they (more or less) and frequently recombining species, such as Neisseria
meningitidis (Maiden 2008). Although there is some evidence This study, therefore, points to the potential of using high-
for phage specificity in different clusters (Waldron and Lindsay resolution sequence data to reconstruct local transmission
2006), the evidence that these clusters present barriers to gene dynamics, and to reveal microevolutionary changes as clones
flow, and hence equate to biological species, remains equivocal. expand locally. An important caveat is that the rate of immigra-
Discounting the recently emerged (and massively over-sampled) tion of variants from other locales should be sufficiently low so
HA-MRSA subgroups, the adaptive relevance of clusters within as not to obscure local patterns of migration and diversification.
the broader population remains unclear. Nevertheless, the con- As ‘‘local’’ might mean anything from a single hospital ward
sistent delineation of the same clonal complexes, even from gene to a whole country, and rates of diversification and short-
content data generated using microarrays, underline that they and long-range migration patterns may vary widely between
are real biological entities, and as such are both of evolutionary taxa, so it is unlikely that there will be any general rules
interest and of epidemiological utility (Turner and Feil 2007). concerning the utility of high-resolution data for understanding
Taking the species as a whole, MLST, MLVA, and PFGE all do transmission routes, or how this is likely to change with epide-
an excellent job in assigning isolates to one or other of these miological scale.
groups, and in revealing the changing frequencies of the clusters The data of Nubel et al. also pointed to multiple independent
over time and space (over years and decades, and on both acquisitions of the SCCmec element responsible for methicillin
national and international scales). Considering HA-MRSA, typ- resistance, even within ST5. This argued against the prevailing
ically a very small number of clones tend to predominate at model of the rapid global dissemination of a few MRSA strains
a given location at any given point in time, a result of sequential in favor of a model of frequent emergence of ‘‘home-grown’’
waves of infection (de Lencastre et al. 2007). This epidemiolog- locally restricted MRSA clones, many of which belong to the
ical pattern lessens the utility of MLST and other typing schemes same MLST-defined haplotype owing to local clonal expansion.
to track strain transmission patterns on local scales (within and This has important implications concerning the high mobility of
between hospitals) over short time scales (weeks and months) the SCCmec element, and its wide distribution within staphylo-
because isolates belonging to a single clonal group are often very coccal reservoirs. This observation also has a practical relevance,
difficult to distinguish. However, the typing data clearly reveals in that genome-wide SNP data should resolve cohesive subtypes
minor variation within clonal complexes, usually manifest as within ST5, each corresponding to independent acquisitions of
single point mutation on the basis of MLST, occasionally differ- SCCmec, which could potentially be distinguished by tailor-
ent spa types, and minor banding pattern changes by PFGE. made typing protocols based on PCR. A more recent paper
Thus, more powerful sequencing studies should be able to dis- using similar methodology focused on one such sub-cluster
sect out sufficient variation within clonal complexes in order to within the broader ST5 group, called ST225 (Nubel et al.
provide clues as to the emergence, spread, and demise of these 2010). The universal presence of a defining deletion within
clusters, and to understand evolutionary processes in bacteria SCCmec showed that this cluster corresponds to a single acqui-
over very short time periods. Although next-generation sition of this element, and the authors were able to posit that this
sequencing platforms might not currently be available for rou- sub-group was introduced into central Europe from the
tine epidemiological surveillance, such data will greatly facilitate USA approximately a decade ago, and has subsequently spread
the development of rapid and cheap typing protocols based on rapidly between hospitals.
PCR assays tailored for specific lineages which may provide
a useful stop-gap until such a time (say, between 5 and
10 years) when next-generation sequencing is commonplace. Zooming in on Single S. aureus Clones: ST239
One of the earliest HA-MRSA clones, and currently the most

Zooming in on Single S. aureus Clones: ST5 dominant globally is called ST239. This haplotype evolved via
a homologous recombination event between unrelated S. aureus
Initial studies aimed at increasing the resolution of MLST by strains (i.e., belonging to different clonal complexes) which
supplementing the data with hypervariable loci (Robinson et al. resulted in the replacement of 20% of the recipient genome
2005; Kuhn et al. 2006) provided further evidence in support of (>600 Kb) (Robinson and Enright 2004). This event remains the
low rates of recombination, and facilitated the reconstruction of largest single homologous replacement in bacteria described in
more robust phylogenetic trees, but shed little extra light on the the literature and arose only once via a completely unknown
patterns of diversification within clonal complexes over time mechanism (> Fig. 10.3). This makes this hybrid clade unusual
and space. A leap forward was taken by Nubel et al. (2008), in that it can be defined by a single and unambiguous
who used a mutation discovery procedure based on denaturing marker. Any strain exhibiting this replacement can be confi-
high-performance liquid chromatography (dHPLC) to identify dently assumed to have descended from the original recipient.
SNPs in 45.5 Kb (1.6% of the genome) in 135 S. aureus isolates Further, as ‘‘reversals’’ are implausible, any isolate that does not
from 22 countries. All the isolates were identical by MLST exhibit this replacement can be excluded (Feil et al. 2008).
(corresponding to the multilocus sequence haplotype ST5). Although the occasional de novo point mutation means that
The authors detected sufficient variation to reveal striking geo- not all members of this clade are identical by MLST, for ease of
graphical structuring, indicating the presence of local variants. terminology we will use the term ST239 to include all
ST30
100
100
89
100
XXX 100
100
100 100
~600-Kb 100 94
100
ST239 100
100
ST 239 ST8 100

100
85
85
xxxxx ST30-like mosaic 100
85
ST8-like backbone 84
98 87
87
50 changes
87
100
100
. Fig. 10.3
A representative phylogeny of S. aureus generated using concatenated data from 40 gene loci (17.8 Kb), reconstructed using MrBayes.
Values at the nodes are posterior probabilities (see Cooper and Feil 2006). The hybrid ST239 genome is the result of a large-scale
recombination event between unrelated lineages ST8 and ST30 (marked on the tree). 600 Kb of the ST8 genome has been replaced by
the homologous region from an ST30-like donor (see Robinson and Enright 2004). Only one example of each of the major clonal
complexes depicted in > Fig. 10.2 is included
members of this clade as defined by the presence of the large and 1990s but has subsequently been replaced by other strains
replacement. (although it is still common in Eastern Europe) (Conceicao et al.
All ST239 isolates are resistant to methicillin (MRSA) and all 2007). Following an outbreak in a London hospital (Edgeworth
isolates possess the same ‘‘type’’ (variant) of SCCmec cassette et al. 2007), a variant of ST239 called TW20 was sequenced
(although minor SCCmec variation is detected). This again (Holden et al. 2010). The clinical significance, widespread dis-
points to common ancestry, and is not consistent with multiple semination, availability of a reference sequence, and unusual
independent acquisitions of SCCmec within this clade. The evolution and epidemiology of ST239 makes it an ideal candi-
specific type common to all ST239 isolates is known as type III date for high-resolution analysis. Smyth et al. identified all the
which, at over 60-Kb, is the largest SCCmec cassette yet mutations within 15 Kb of sequence of 111 ST239 isolates
described in S. aureus. Type III SCCmec elements confer multi- representing 34 years and 29 countries (Smyth et al. 2010).
ple resistance, and have only been observed in ST239 isolates. Again, these authors noted geographical structuring on
ST239 is responsible for 90% of hospital-acquired MRSA a continental scale, and identified European, South American,
infection throughout most of mainland Asia (from the Middle and Asian sub-clades. These data also pointed to repeated and
East to China), and much of South America (including Brazil) independent deletions within the SCCmec element during
(Diekema et al. 2001; Chongtrakool et al. 2006; Xu et al. 2009). diversification of the ST239 clade, but (unlike ST5) there was
However, similar to ST225, ST239 is almost exclusively observed no evidence for multiple acquisitions of completely different
within hospital environments. This is thought to be because the SCCmec types.
large SCCmec cassette confers a fitness cost which renders it
uncompetitive in the community. This implicates direct hospi-
tal-to-hospital transmission as playing a key role in its global A New Dawn for Next-Generation Sequencing
dissemination.
ST239 is also unusual in that there is both epidemiological The ST239 clade was also chosen as the focus for the first
and experimental evidence pointing to increased virulence population study which utilized the Illumina Genome Analyzer
(Amaral et al. 2005; Edgeworth et al. 2007). ST239 was also the (IGA) platform to identify genome-wide SNPs and INDELs
predominant MRSA clone in Western Europe during the 1980s compared to TW20 as a reference sequence. Harris and
colleagues used index adapters to create individually tagged and from adjacent wards in the same block in the hospital. This
genomic libraries in order to rapidly generate whole-genome observation supports the possibility that this approach could
DNA sequence data for a large number of isolates (Harris et al. provide information on transmission chains even at the scale of
2010). The authors characterized 62 ST239 isolates, 42 of which a single hospital, which has clear implications for infection
were globally representative, while the remaining 20 were control.
isolated from a single hospital in northeast Thailand over a More large-scale transmission events were clearly evident in
7-month period. The reference TW20 strain was also the data of Harris et al. Portugal experienced two waves of
re-sequenced as a control. The study was thus designed to infection due to ST239 during the 1990s. Isolates recovered
simultaneously address both extremes of epidemiological scale: from the second wave showed subtle band differences by PFGE
the global diversity of this clone and the utility of next- from the first wave and were more similar to isolates from Brazil
generation sequencing for highly localized epidemiology within than those previously isolated from Portugal. The latter clone
a single health-care setting. from Portugal had been previously dubbed the ‘‘Brazillian
The study identified 6,714 high-quality SNPs, but these were clone’’ on the basis of the PFGE profiles (Sanches et al. 1998),
not equally distributed throughout the genome. Regions of high and the three Portuguese isolates from this second wave
density SNPs were clearly identified, and these also tended to included in the study of Harris et al. clearly cluster with the
correspond to regions with relatively low coverage. These diverse South American isolates. This then confirmed that this second
regions represented the non-core genome, principally consisting wave of infection in Portugal was seeded from South America.
of mobile genetic elements (MGEs), such as SCCmec, genomic More surprisingly, the TW20 reference genome, which origi-
islands, conjugative elements, and prophage. Harris et al. nated from a recent outbreak in an Intensive Care Unit in
defined core regions simply and conservatively as all regions London (Edgeworth et al. 2007), clustered with the Thai isolates,
>1 Kb which were mapped to a high quality in all isolates. thus implicating a hitherto unexpected Southeast Asian origin of
This definition provides a subtly different set of genes assigned this outbreak strain.
as ‘‘accessory’’ than previous microarray and comparative geno-
mic studies on the broader S. aureus population. For example,
any accessory elements present in the founding ST239 genome A Paucity of Homoplasies
which had been stably inherited were assigned as core, whereas
in contrast any small INDELs which have arisen since the emer- A common feature of all the datasets discussed thus far, and also
gence of ST239 will mean that the corresponding region will be for those of some monomorphic species, such as Salmonella
assigned as non-core, even though they may be ‘‘native’’ to the Typhi (Roumagnac et al. 2006; Holt et al. 2008), is the lack of
genome. phylogenetic conflict in the data due to homoplasy (at least, once
The vast majority of those SNPs thus assigned as the non-core regions have been excluded). While this is of
corresponding to non-core regions are likely to have been obvious benefit in enabling easy and robust tree construction,
acquired by horizontal transfer, as has been confirmed subse- does it betray a deeper significance? Homoplasies are character
quently (Castillo-Ramirez et al. 2011). For this reason, they were states (in this case SNPs) that are observed in unrelated lineages,
excluded from the phylogenetic analysis, leaving 4,310 variable and there are three possible means by which they might be
sites in the core genome. generated. The first is mutational reversal in other sequences,
Similar to the studies mentioned above, Harris et al. noted such that homoplasies are in fact identical by descent but have
striking geographical structuring in the data from their tree, been lost in intervening sequences. In the context of the current
which was rooted using ST8, the original recipient of the large discussion, this scenario is so unlikely it can be effectively
replacement (Harris et al. 2010). The increased resolution pro- discounted. The second possibility is that identical mutational
vided by IGA also allowed a number of specific inferences events may arise de novo independently in different lineages.
concerning the emergence and spread of ST239. First, it is The third possibility is recombination, in this case SNPs emerge
apparent that European isolates are more diverse than Asian or de novo by mutation in one lineage and are then horizontally
South American isolates. This observation points to a European transferred to other lineages.
origin of ST239, which is consistent with the fact that it was first Harris et al. noted only 38 homoplasies among the 4310
recorded in this continent, but has subsequently been largely SNPs in the core genome (<0.1%) (Harris et al. 2010). Although
replaced, at least in Western Europe. Second, isolates from S. recombination is known to be rare in S. aureus, it does occa-
America are extremely homogenous, despite the fact that they sionally happen, the large replacement in ST239 being a notable
represent Brazil, Chile, Argentina, and Uruguay and were recov- example. Given the unlikely alternatives, it is, therefore, reason-
ered from 1993 to 1998. This strongly points to a single intro- able to assume that at least some of the observed homoplasies
duction of ST239 into South America followed by dramatic have arisen through recombination, and this is supported sim-
spread. Third, no two isolates were identical even when exclud- ply by the physical clustering of these SNPs. Harris et al. also
ing the non-core genome, and this includes isolates recovered considered the selective consequences of the observed homo-
days apart from the same ward in Sappasithiprasong hospital in plasies, and noted that almost a third of them correspond to
northeast Thailand. Furthermore, five isolates were differenti- mutations known to confer antibiotic resistance. This means
ated by only 14 SNPs, and these were isolated within a few weeks they will confer a strong selective advantage and be much
more likely to be observed than neutral or slightly deleterious mutation rate as one moves toward the very tip of the trees
changes, which are more likely to be quickly lost through drift. (Balbi and Feil 2007). This effect has recently been discussed in
Indeed, an important message from the study of Harris et al. is terms of two categories: a mutation rate and a substitution rate
that examining homoplasies is likely to be a powerful means to (Achtman 2008; Nubel et al. 2010). The mutation rate is simply
identify changes which confer an adaptive advantage, and in the rate at which mutations are generated de novo. The high rate
particular those conferring drug resistance. Homoplasies can, of change within the groups and subgroups described is assumed
therefore, be explained either by occasional recombination or by to approximate this rate. The substitution rate refers to all the
positive selection. Occasionally, these forces may work in con- mutations which have subsequently become fixed within the
cert through hitch-hiking. For example, in the data of Harries population; this is necessarily much slower than the mutation
et al. four synonymous homoplasies clustering within approxi- rate, as most mutations will be lost by selection or drift.
mately 1 Kb are noted. These are likely to have hitch-hiked with Although classically correct, this interpretation remains
two mutations conferring trimethoprim resistance which were something of an oversimplification. Prokaryotes do not con-
located 200–300 bp away. form to idealized sexual (eukaryotic) populations, and what is
true for elephants is not always true for E. coli.
Difficulties arise because it is not always obvious where the
Dating with Confidence boundary of a ‘‘population,’’ as defined by the limits of drift,
should lie. One is inclined to use the (often pragmatically)
The discussion above hints at a discrepancy in the rate of molec- named species to define such a population, but the assumption
ular evolution over extremely short time scales even when com- that these two boundaries actually coincide is rarely, if ever,
pared to more modestly related lineages within the named species. explicitly examined. Indeed, despite the recent soul searching
The relative paucity of homoplasies within the ST239 clone sug- on the nature of bacterial ‘‘species’’ (Hanage et al. 2006), the less
gests that recombination rates in S. aureus might be lower within loaded term ‘‘population’’ continues to be deployed rather casu-
ST239 than in the broader population. However, an equivalent ally. To put the problem succinctly, it is currently not clear
interpretation is that recombination rates are similar, or even whether the term ‘‘fixation’’ (or ‘‘substitution’’) should apply
slightly higher (as might be expected given the high sequence only to those mutations which are present in all strains in the
identity), but that mutation rates are much higher. The estima- named species, or whether the concept may also apply on the
tion of mutation rates (hence dating the emergence of clades) level of clonal complex. In the highly monomorphic species,
represents a central theme in the high-resolution studies of such as Bacillus anthracis or Mycobacterium tuberculosis, the
bacterial populations currently being published, and (at least question is simple as named species equates to a single clonal
for S. aureus) a clear consensus is emerging. The three lineage, and likely a single ‘‘ecotype.’’ In such cases, there is little
intraclonal S. aureus studies discussed above propose strikingly justification for further ‘‘splitting.’’ However, in more diverse
similar figures: 3.3–4.6 106 per site per year (Smyth et al. species, such as S. aureus, E. coli, or N. meningitidis, there is
2010), 2.5–4.0 106 (Harris et al. 2010), and 1.2 106 to 2.9 substantial substructuring, that is to say, a number of different
106 (Nubel et al. 2010). Harris et al. dated the emergence of clonal lineages which may represent different adaptive peaks,
ST239 to the mid-to-late 1960s, whereas the date estimate of and hence different populations. One might then consider a type
Smyth et al. was approximately a decade earlier. Harris et al. also of ‘‘hierarchical fixation,’’ encompassing different phylogenetic
noted that this rate equates to approximately 1 SNP every levels from clonal complex, major clades of related complexes,
6 weeks (in the core genome), and Smyth et al. similarly opined and whole species. It follows that every level should also corre-
that this clone is ‘‘measurably evolving.’’ spond to a different rate of change, getting progressively slower
The date estimates given above are approximately 100-fold as one moves back in the tree, and representing intermediate
higher than the standard estimate of 3 108 for E. coli rates between ‘‘mutation’’ and ‘‘substitution.’’ This is an impor-
(Achtman et al. 1999), but approximately 10-fold slower than tant question, as dating estimates form a central plank of the
estimates for Campylobacter jejuni (Wilson et al. 2009), analyses in re-sequencing studies. Although the relationship
Helicobacter pylori (Falush et al. 2001), and Neisseria gonorrhoeae between ‘‘mutation’’ rate and divergence has not yet been exam-
(Perez-Losada et al. 2007). If we cancel out generation times by ined at the intraspecies level for bacteria, one might predict
assuming they are similar between species (although they are in (based on the studies discussed below) a log-linear relationship,
fact notoriously difficult to measure for natural populations) such that there is a rapid drop of rate moving back from the
then the faster estimates for these latter species are still easily most fine-scale clusters.
explained by the fact that they all recombine at a much higher
frequency than S. aureus. This means that the majority of the
mutations observed in these studies will not have arisen de novo The Purging of Deleterious Mutations
but will have been horizontally acquired. However, this does not Over Time
explain why all these estimates are so fast relative to the canon-
ical estimate for E. coli reflects. The key to understanding this is The decrease in the rate of change moving back from the tips of
the fact that these estimates were based on extremely closely trees is due to the accumulative purging of slightly deleterious
related isolates, and there is an apparent acceleration of mutations. Purifying selection acts as a sieve in removing these
Time gene transfer (a phenomenon yet to be described for bacteria)

Early divergence:
(Rocha and Feil 2010).
nt divergence = 0.1%
high dN/dS (~0.8) Whatever force is responsible for the maintenance of GC
content in bacterial genomes, in certain taxa this force appears to
be operating very weakly, if at all. Insect endosymbionts, such as
Buchnera aphidicola, are characterized by very high AT contents
Late divergence: nt divergence = 1%
low dN/dS (~0.1)
and very high dN/dS ratios, both of which may be consistent
with weak purging of slightly deleterious mutations (Moran
. Fig. 10.4 et al. 2009). Such species are transmitted from mother to off-
The selective purging of slightly deleterious non-synonymous spring transovarially down the generations (vertically), a lifestyle
mutations over time results in a decrease in the dN/dS ratio (see that imposes repeated bottlenecking, thus lowering the Ne.
Rocha et al. 2006). The figures of divergence and dN/dS given are However, as such a specialization also results in ecological iso-
approximate for S. aureus. Early divergence corresponds to lation and hence an absence of recombination, the high AT
comparisons within a clonal complex, and late divergence content would also be predicted by a lack of biased gene con-
corresponds to comparisons between unrelated clonal complexes version. For this reason, high AT endosymbiont genomes do not
currently constitute equivocal evidence for a selective mainte-
nance of high GC content in other, more typical, genomes.
The fact that the GC content in most bacterial genomes
mutations from the population, but this purging does not occur resides at a point higher than predicted by mutation pressure
instantaneously; the excess mutations observed at the very tips alone suggests that GC ! AT mutations must be lost over time
of the tree represent those slightly deleterious changes destined more frequently than AT ! GC mutations, in much the same
for subsequent loss. According to the nearly neutral model, the way that non-synonymous mutations are lost preferentially
efficiency of purifying selection is determined by the selection to synonymous mutations. Balbi et al. examined the selective
coefficient(s) and the effective population size (Ne). If Ne s < 1, purging over time of both non-synonymous mutations and
then the chances of the mutation becoming fixed are predicted GC ! AT mutations (relative to their counterparts) through
to be as if the mutation were neutral, otherwise they are a comparative genomic analysis of Escherichia coli and Shigella
predicted to be slightly deleterious and stand a lower chance of (Balbi et al. 2009). The Ne of E. coli can be assumed to be very
becoming fixed. The progressive purging of slightly deleterious large, as this is a very generalist species, which occupies a wide
mutations can be easily plotted over time. As de novo non- range of animal guts (usually without harm to the host), and is
synonymous mutations are on average more likely to be slightly able to survive in soil and water (Hartl et al. 1994). The four
deleterious than synonymous mutations, they should be Shigella named ‘‘species’’ are the causative agents of bacillary
enriched between very closely related genomes but preferentially dysentery and are essentially specialized clones of E. coli that
removed over time (> Fig. 10.4). Rocha et al. confirmed this have independently acquired a large ‘‘invasion’’ plasmid which
effect by demonstrating that dN/dS ratio between pairs of confers on them the ability to invade host cells and replicate
genomes decreases with increasing divergence (Rocha et al. intracellularly (Lan and Reeves 2002). This adaptation to
2006), and this effect has subsequently been confirmed by a specialized lifestyle might be accompanied by a reduction in
other studies (Hughes et al. 2008; Kryazhimskiy and Plotkin Ne, in a way somewhat analogous to ‘‘island’’ populations of
2008; Garcia Pelayo et al. 2009; Novichkov et al. 2009). Further- eukaryotes (Johnson and Seger 2001; Balbi and Feil 2007).
more, Rocha et al. showed by simulation that the trajectory of Balbi et al. examined the molecular evolutionary conse-
this decline is dependent upon both Ne and s, as predicted under quences of this niche restriction by comparing both the pro-
the nearly neutral theory. portions of GC ! AT polymorphisms over the reverse and dN/
It is important to note that there is no reason to suppose that dS in the E. coli and Shigella genomes. They confirmed two key
this effect is reserved for non-synonymous changes. Although predictions: (1) that slightly deleterious mutations (synony-
the dN/dS ratio is a very convenient metric, the same principle mous or GC ! AT) are selectively purged over time, and
should apply to any kind of slightly deleterious change relative to (2) slightly deleterious mutations are enriched in Shigella
a more neutral counterpart. A curious, yet striking, example may relative to E. coli (when divergence time is considered).
be the maintenance of GC content in bacterial genomes. Com- > Figure 10.5 reveals a log-linear purging of two types of slightly
parative genomics analyses of a large range of bacterial taxa have deleterious mutations over time A: GC- > AT mutations; B:
revealed that GC ! AT mutations are far more common than Non-synonymous mutations. In both cases, this purging is less
the reverse (Hershberg and Petrov 2010) (Hildebrand et al. marked in Shigella than it is in E. coli. Note that only fourfold
2010). This bias is so strong that at mutational equilibrium the degenerate sites are used for the GC ! AT analysis, hence these
base composition of most bacterial genomes would rest at 20% two effects are independent. This also means that if a selective
GC. As base composition is typically much higher than this, force is responsible for the maintenance of GC content, it operates
there must be some force which acts to favor GC over AT. As on synonymous (and intergenic) sites, so is not directly related to
recently discussed, a number of selective explanations have been amino-acid sequence. These observations, therefore, challenge the
proposed, as well as the purely mechanistic explanation of biased widely held view that synonymous changes can be assumed to be
0.3
E. coli
2.5 0.25
Shigellae
+AT/+GC 4-fold deg. sites only
Internal branches
2 0.2
dN/dS
1.5 0.15
1 0.1
0.5 0.05
0 0
3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4 4.1 4.2 4.3 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4 4.1 4.2 4.3
Log (Divergence Time) Log (Divergence Time)
. Fig. 10.5
The selective purging of GC ! AT mutations and non-synonymous mutations (relative to their counterparts) over time in Shigella and
E. coli. The effect is weaker in Shigella than in E. coli because the former has a smaller effective population size. ‘‘Divergence time’’ is
measured as the total number of SNPs, and the relationships are log-linear (see Balbi et al. 2009)
neutral. Transversions behave in the same way (relative to transi- (corresponding to various mobile non-core elements) show
tions), but this is because transversions are more likely to be non- a relative enrichment of synonymous changes relative to the
synonymous (Balbi et al. 2009). core genome. Essentially, those SNPs acquired by horizontal
transfer are more likely to be synonymous because they have
already passed through a selective filter in the wider population.
Synthesizing Selection and Ecology Using An analysis of the large replacement present in ST239 supports
Next-Generation Sequence Data this argument, and a similar enrichment of synonymous changes
within recombined blocks is noted in Clostridium difficile as
The relative preponderance of non-synonymous changes noted using the data of He et al. (Castillo-Ramirez et al. 2011)
between very closely related genomes, with subsequent purifica- (He et al. 2010).
tion of these SNPs over time, has been repeatedly confirmed for In the study by Balbi et al. on Shigella genomes described
S. aureus and a number of other species (Larsson et al. 2009), above, it was noted that the decline in dN/dS over time was
(Holt et al. 2008), (He et al. 2010). However, the increased moderated in Shigella relative to E. coli because ecological spe-
resolution afforded by next-generation sequencing will reveal cialization has resulted in a decrease in the effective population
the dynamics of this process in greater detail. Revisiting the data size and hence weaker purifying selection (Balbi et al. 2009).
of Harris et al., Castillo-Ramirez et al. plotted the dN/dS of Such a model provides a direct bridge between ecology and
ST239 isolates against divergence for the core and non-core molecular evolution, which may also have a bearing on the
genomes separately (Castillo-Ramirez et al. 2011). They noted analysis of the ST239 S. aureus clone. As discussed, this clone
that non-synonymous SNPs were proportionately far more is highly specialized to the hospital environment, and competes
common in the core genome than the non-core genome. Fur- very poorly in the community, probably as a consequence of the
thermore, although the dN/dS ratio declined against divergence high fitness cost associated with multiple drug resistance and
when non-core SNPs were considered, no obvious purging of a very large SCCmec element. This then raises the possibility that
non-synonymous changes over time was noted in the core SNPs, this ecological specialism will similarly reduce the power of
apart from a steep initial decrease owing to the rapid purging of purifying selection in this clone, leading to an accumulation of
a highly deleterious class. The authors suggest that the relative slightly deleterious mutations. In other words, it is possible that
enrichment of non-synonymous changes in the core, relative to hospitals may be acting as ecological islands. Such a process
the non-core, reflects the fact that many of the SNPs in the non- could be of key importance in understanding the change in
core genome have been acquired via horizontal transfer, while clonal composition of a bacterial population over time. Most
those in the core genome have arisen by de novo mutation. This attempts to understand clonal replacement, where one clone
means that the core SNPs are, on average, younger and (by the within a given locale rapidly disappears and another takes its
argument outlined above) should, therefore, contain a higher place, begin from the perspective of trying to understand what it
proportion of non-synonymous mutations. This effect is appar- is that makes a new clone successful. In this context, it may also
ent from > Fig. 10.6 where regions of high SNP density be fruitful to consider what makes old clones less successful; that
900
ϕSa1(TW20)
synonymous
800 2 non-synonymous
total
700
600
Number of SNPs
500
400
300
200 SCCmec III vSaβ

SaPI1
4 Tn552
1 3
100 5
0
180
330
480
630
780
930
1080
1230
1380
1530
1680
1830
1980
2130
2280
2430
2580
2730
2880
3030
Clockwise distance from origin (Kb)
. Fig. 10.6
The distribution of SNPs across the TW20 (reference) genome. The five peaks of high SNP density correspond to well-characterized
components of the accessory genome; SCCmec III (a resistance casette), wSa1(TW20) (a prophage), SaPI1 (a pathogenicity island), nSab
(a genomic island), and Tn552 (a transposon). The figure illustrates that for peaks 2–5 synonymous SNPs (dashed line) are more common
than non-synonymous SNPs, whereas the reverse is true for regions between the peaks. Other well-characterized elements known to be
present in the TW20 genome do not correspond to a clear peak of SNP density because they are less variable between ST239 isolates
is, to what extent clones can consolidate their advantage in ‘‘species’’) boundaries. As the trajectories depend both on Ne
a population over time in the face of deleterious mutations, and s, so comparisons may also be drawn between different
and whether the rapid spread of new ‘‘successful’’ clones is mutation types, or between different classes of genes. Many
inevitably self-limiting. This may have some bearing on the other analytical approaches can be taken with these data, for
ongoing decline and replacement of S. aureus ST239 in many example to understand the dynamics of gene loss and gain, the
parts of Europe and Asia, and such a model is consistent with the size of the accessory gene ‘‘library’’ from which any strain of
apparently weak purifying selection acting on this clone evi- a given population may access, and how evolutionary processes
dence by the poor purging of non-synonymous mutations. vary according to genomic position. A comprehensive recent
study on multiple E. coli genomes provides an excellent template
for such studies (Touchon et al. 2009).
Concluding Remarks
Next-generation sequencing clearly heralds a new dawn in bac- Acknowledgments

terial microevolution. Similar to previous developments, such as
MLST, initial studies using the new technology have (for obvi- SCR is funded by the TROCAR consortium (FP7-HEALTH
ous reasons) focused on pathogenic bacteria, with a view to #223031). We are grateful to colleagues present at the first
increased power of epidemiological surveillance and under- international PERMAFROST meeting, Bormio, Italy, on March
standing patterns of transmission. However, here we have con- 5–8 2010, and for their fruitful discussions.
sidered how these data might also provide information on basic
evolutionary and population-level processes and the molecular
consequences of lifestyle changes. These future directions are References
certain to impact on our understanding of the ecology of envi-
ronmental species, as well as the epidemiology of important Achtman M (2008) Evolution, population structure, and phylogeography of
genetically monomorphic bacterial pathogens. Annu Rev Microbiol
pathogens. A consideration of the strength of purifying selec-
62:53–70
tion, and how this relates to the ecology of the clone//species, Achtman M, Zurth K, Morelli G, Torrea G, Guiyoule A, Carniel E (1999) Yersinia
may even help to define the limits of drift, thus finally providing pestis, the cause of plague, is a recently emerged clone of Yersinia pseudotu-
a conceptual basis for identifying population (and hence berculosis. Proc Natl Acad Sci USA 96:14043–14048
Aires de Sousa M, Conceicao T, Simas C, de Lencastre H (2005) Comparison of Edgeworth JD, Yadegarfar G, Pathak S, Batra R, Cockfield JD, Wyncoll D, Beale R,
genetic backgrounds of methicillin-resistant and -susceptible Staphylococcus Lindsay JA (2007) An outbreak in an intensive care unit of a strain of
aureus isolates from Portuguese hospitals and the community. J Clin methicillin-resistant Staphylococcus aureus sequence type 239 associated
Microbiol 43:5150–5157 with an increased rate of vascular access device-related bacteremia. Clin
Alm RA, Trust TJ (1999) Analysis of the genetic diversity of Helicobacter pylori:the Infect Dis 44:493–501
tale of two genomes. J Mol Med 77:834–846 Enright MC, Robinson DA, Randle G, Feil EJ, Grundmann H, Spratt BG
Amaral MM, Coelho LR, Flores RP, Souza RR, Silva-Carvalho MC, Teixeira LA, (2002) The evolutionary history of methicillin-resistant Staphylococcus
Ferreira-Carvalho BT, Figueiredo AM (2005) The predominant variant of aureus (MRSA). Proc Natl Acad Sci USA 99:7687–7692
the Brazilian epidemic clonal complex of methicillin-resistant Staphylococ- Falush D, Kraft C, Taylor NS, Correa P, Fox JG, Achtman M, Suerbaum S (2001)
cus aureus has an enhanced ability to produce biofilm and to adhere to and Recombination and mutation during long-term gastric colonization by
invade airway epithelial cells. J Infect Dis 192(5):801–810 Helicobacter pylori: estimates of clock rates, recombination size, and minimal
Balbi KJ, Feil EJ (2007) The rise and fall of deleterious mutation. Res Microbiol age. Proc Natl Acad Sci USA 98:15056–15061
158:779–786 Feil EJ (2004) Small change: keeping pace with microevolution. Nat Rev
Balbi KJ, Rocha EP, Feil EJ (2009) The temporal dynamics of slightly Microbiol 2:483–495
deleterious mutations in Escherichia coli and Shigella spp. Mol Biol Feil EJ, Cooper JE, Grundmann H et al (2003) How clonal is Staphylococcus
Evol 26:345–355 aureus? J Bacteriol 185:3307–3316
Basset P, Hammer NB, Kuhn G, Vogel V, Sakwinska O, Blanc DS (2009) Staphy- Feil EJ, Li BC, Aanensen DM, Hanage WP, Spratt BG (2004) eBURST:
lococcus aureus clfB and spa alleles of the repeat regions are segregated into inferring patterns of evolutionary descent among clusters of related
major phylogenetic lineages. Infect Genet Evol 9:941–947 bacterial genotypes from multilocus sequence typing data. J Bacteriol
Boyd EF, Nelson K, Wang FS, Whittam TS, Selander RK (1994) Molecular genetic 186:1518–1530
basis of allelic polymorphism in malate dehydrogenase (mdh) in natural Feil EJ, Nickerson EK, Chantratita N et al (2008) Rapid detection of the pandemic
populations of Escherichia coli and Salmonella enterica. Proc Natl Acad Sci methicillin-resistant Staphylococcus aureus clone ST 239, a dominant strain
USA 91:1280–1284 in Asian hospitals. J Clin Microbiol 46:1520–1522
Castillo-Ramirez S, Harris SR, Holden MT, He M, Parkhill J, Bentley SD, Feil EJ Fitzgerald JR, Monday SR, Foster TJ, Bohach GA, Hartigan PJ, Meaney WJ, Smyth
(2011) The impact of recombination on dN/dS within recently emerged CJ (2001) Characterization of a putative pathogenicity island from bovine
bacterial clones. PLoS Pathog 7:e1002129 Staphylococcus aureus encoding multiple superantigens. J Bacteriol 183(1):
Chan MS, Maiden MC, Spratt BG (2001) Database-driven multi locus sequence 63–70
typing (MLST) of bacterial pathogens. Bioinformatics 17:1077–1083 Fleischmann RD, Adams MD, White O et al (1995) Whole-genome random
Chongtrakool P, Ito T, Ma XX, Kondo Y, Trakulsomboon S, Tiensasitorn C, sequencing and assembly of Haemophilus influenzae Rd. Science
Jamklang M, Chavalit T, Song JH, Hiramatsu K (2006) Staphylococcal 269:496–512
cassette chromosome mec (SCCmec) typing of methicillin-resistant Staph- Francisco AP, Bugalho M, Ramirez M, Carrico JA (2009) Global optimal eBURST
ylococcus aureus strains isolated in 11 Asian countries: a proposal for a new analysis of multilocus typing data using a graphic matroid approach. BMC
nomenclature for SCCmec elements. Antimicrob Agents Chemother 50 Bioinformatics 10:152
(3):1001–1012 Garcia Pelayo MC, Uplekar S, Keniry A et al (2009) A comprehensive survey of
Conceicao T, Aires-de-Sousa M, Fuzi M, Toth A, Paszti J, Ungvari E, van Leeuwen single nucleotide polymorphisms (SNPs) across Mycobacterium bovis strains
WB, van Belkum A, Grundmann H, de Lencastre H (2007) Replacement of and M. bovis BCG vaccine strains refines the genealogy and defines a minimal
methicillin-resistant Staphylococcus aureus clones in Hungary over time: set of SNPs that separate virulent M. bovis strains and M. bovis BCG strains.
a 10-year surveillance study. Clin Microbiol Infect 13:971–979 Infect Immun 77:2230–2238
Cooper JE, Feil EJ (2006) The phylogeny of Staphylococcus aureus—Which genes Gevers D, Cohan FM, Lawrence JG et al (2005) Opinion: re-evaluating prokary-
make the best intra-species markers? Microbiology 152:1297–1305 otic species. Nat Rev Microbiol 3:733–739
Cookson BD, Robinson DA, Monk AB, Murchan S, Deplano A, de Ryck R, Gomes AR, Westh H, de Lencastre H (2006) Origins and evolution of methicillin-
Struelens MJ, Scheel C, Fussing V, Salmenlinna S et al (2007) Evaluation of resistant Staphylococcus aureus clonal lineages. Antimicrob Agents
molecular typing methods in characterizing a European collection of epi- Chemother 50:3237–3244
demic methicillin-resistant Staphylococcus aureus strains: the HARMONY Hanage WP, Fraser C, Spratt BG (2006) Sequences, sequence clusters and bacte-
collection. J Clin Microbiol 45(6):1830–1837 rial species. Philos Trans R Soc Lond B Biol Sci 361:1917–1927
Crisostomo MI, Westh H, Tomasz A, Chung M, Oliveira DC, de Lencastre H Harris SR, Feil EJ, Holden MT et al (2010) Evolution of MRSA during hospital
(2001) The evolution of methicillin resistance in Staphylococcus aureus: transmission and intercontinental spread. Science 327:469–474
similarity of genetic backgrounds in historically early methicillin-susceptible Hartl DL, Moriyama EN, Sawyer SA (1994) Selection intensity for codon bias.
and -resistant isolates and contemporary epidemic clones. Proc Natl Acad Sci Genetics 138:227–234
USA 98:9865–9870 He M, Sebaihia M, Lawley TD et al (2010) Evolutionary dynamics of
de Lencastre H, Oliveira D, Tomasz A (2007) Antibiotic resistant Staphylococcus Clostridium difficile over short and long time scales. Proc Natl Acad Sci
aureus: a paradigm of adaptive power. Curr Opin Microbiol 10:428–435 USA 107:7527–7532
Deleo FR, Otto M, Kreiswirth BN, Chambers HF (2010) Community-associated Hershberg R, Petrov DA (2010) Evidence that mutation is universally biased
methicillin-resistant Staphylococcus aureus. Lancet 375:1557–1568 towards AT in bacteria. PLoS Genet 6:e1001115
Diekema DJ, Pfaller MA, Schmitz FJ, Smayevsky J, Bell J, Jones RN, Beach M Hildebrand F, Meyer A, Eyre-Walker A (2010) Evidence of selection upon geno-
(2001) Survey of infections due to Staphylococcus species: frequency of mic GC-content in bacteria. PLoS Genet 6:e1001107
occurrence and antimicrobial susceptibility of isolates collected in the United Holden MT, Feil EJ, Lindsay JA et al (2004a) Complete genomes of two clinical
States, Canada, Latin America, Europe, and the Western Pacific region for Staphylococcus aureus strains: evidence for the rapid evolution of virulence
the SENTRYAntimicrobial Surveillance Program, 1997–1999. Clin Infect Dis and drug resistance. Proc Natl Acad Sci USA 101:9786–9791
32(Suppl 2):S114–132 Holden MT, Titball RW, Peacock SJ et al (2004b) Genomic plasticity of the
Dorrell N, Mangan JA, Laing KG, Hinds J, Linton D, Al-Ghusein H, Barrell BG, causative agent of melioidosis, Burkholderia pseudomallei. Proc Natl Acad
Parkhill J, Stoker NG, Karlyshev AV et al (2001) Whole genome comparison Sci USA 101:14240–14245
of Campylobacter jejuni human isolates using a low-cost microarray reveals Holden MT, Lindsay JA, Corton C, Quail MA, Cockfield JD, Pathak S, Batra R,
extensive genetic diversity. Genome Res 11(10):1706–1715 Parkhill J, Bentley SD, Edgeworth JD (2010) Genome sequence of a recently
DuBose RF, Dykhuizen DE, Hartl DL (1988) Genetic exchange among natural emerged, highly transmissible, multi-antibiotic- and antiseptic-resistant var-
isolates of bacteria: recombination within the phoA gene of Escherichia coli. iant of methicillin-resistant Staphylococcus aureus, sequence type 239 (TW).
Proc Natl Acad Sci USA 85:7036–7040 J Bacteriol 192:888–892
Holt KE, Parkhill J, Mazzoni CJ et al (2008) High-throughput sequencing Ochman H, Lawrence JG, Groisman EA (2000) Lateral gene transfer and the
provides insights into genome variation and evolution in Salmonella nature of bacterial innovation. Nature 405:299–304
Typhi. Nat Genet 40:987–993 Peacock SJ, de Silva I, Lowy FD (2001) What determines nasal carriage of
Hughes AL, Friedman R, Rivailler P, French JO (2008) Synonymous and Staphylococcus aureus? Trends Microbiol 9(12):605–610
nonsynonymous polymorphisms versus divergences in bacterial genomes. Perez-Losada M, Crandall KA, Zenilman J, Viscidi RP (2007) Temporal trends in
Mol Biol Evol 25:2199–2209 gonococcal population genetics in a high prevalence urban community.
Johnson KP, Seger J (2001) Elevated rates of nonsynonymous substitution in Infect Genet Evol 7:271–278
island birds. Mol Biol Evol 18:874–881 Robinson DA, Enright MC (2004) Evolution of Staphylococcus aureus by large
Kryazhimskiy S, Plotkin JB (2008) The population genetics of dN/dS. PLoS Genet chromosomal replacements. J Bacteriol 186:1060–1064
4:e1000304 Robinson DA, Monk AB, Cooper JE, Feil EJ, Enright MC (2005) Evolutionary
Kuhn G, Francioli P, Blanc DS (2006) Evidence for clonal evolution among highly genetics of the accessory gene regulator (agr) locus in Staphylococcus aureus.
polymorphic genes in methicillin-resistant Staphylococcus aureus. J Bacteriol J Bacteriol 187:8312–8321
188:169–178 Rocha EP, Feil EJ (2010) Mutational patterns cannot explain genome composi-
Lan R, Reeves PR (2002) Escherichia coli in disguise: molecular origins of Shigella. tion: are there any neutral sites in the genomes of bacteria? PLoS Genet 6:
Microbes Infect 4:1125–1132 e1001104
Larsson P, Elfsmark D, Svensson K, Wikstrom P, Forsman M, Brettin T, Keim P, Rocha EP, Smith JM, Hurst LD, Holden MT, Cooper JE, Smith NH, Feil EJ
Johansson A (2009) Molecular evolutionary consequences of niche restric- (2006) Comparisons of dN/dS are time dependent for closely related bacte-
tion in Francisella tularensis, a facultative intracellular pathogen. PLoS rial genomes. J Theor Biol 239:226–235
Pathog 5:e1000472 Roumagnac P, Weill FX, Dolecek C et al (2006) Evolutionary history of Salmonella
Maiden MC (2006) Multilocus sequence typing of bacteria. Annu Rev Microbiol typhi. Science 314:1301–1304
60:561–588 Sanches IS, Saraiva ZC, Tendeiro TC, Serra JM, Dias DC, de Lencastre H (1998)
Maiden MC (2008) Population genomics: diversity and virulence in the neisseria. Extensive intra-hospital spread of a methicillin-resistant staphylococcal
Curr Opin Microbiol 11:467–471 clone. Int J Infect Dis 3:26–31
Maiden MC, Bygraves JA, Feil E, Morelli G, Russell JE, Urwin R, Zhang Q, Zhou J, Schouls LM, Spalburg EC, van Luit M, Huijsdens XW, Pluister GN, van Santen-
Zurth K, Caugant DA et al (1998) Multilocus sequence typing: a portable Verheuvel MG, van der Heide HG, Grundmann H, Heck ME, de Neeling AJ
approach to the identification of clones within populations of pathogenic (2009) Multiple-locus variable number tandem repeat analysis of Staphylo-
microorganisms. Proc Natl Acad Sci USA 95(6):3140–3145 coccus aureus: comparison with pulsed-field gel electrophoresis and spa-
Melles DC, Schouls L, Francois P, Herzig S, Verbrugh HA, van Belkum A, typing. PLoS One 4:e5082
Schrenzel J (2009) High-throughput typing of Staphylococcus aureus by Smith JM, Smith NH, O’Rourke M, Spratt BG (1993) How clonal are bacteria?
amplified fragment length polymorphism (AFLP) or multi-locus variable Proc Natl Acad Sci USA 90:4384–4388
number of tandem repeat analysis (MLVA) reveals consistent strain related- Smith NH, Dale J, Inwald J, Palmer S, Gordon SV, Hewinson RG, Smith JM
ness. Eur J Clin Microbiol Infect Dis 28:39–45 (2003) The population structure of Mycobacterium bovis in Great Britain:
Mellmann A, Weniger T, Berssenbrugge C, Keckevoet U, Friedrich AW, clonal expansion. Proc Natl Acad Sci USA 100:15271–15275
Harmsen D, Grundmann H (2008) Characterization of clonal relatedness Smith NH, Gordon SV, de la Rua-Domenech R, Clifton-Hadley RS, Hewinson
among the natural population of Staphylococcus aureus strains by using spa RG (2006) Bottlenecks and broomsticks: the molecular evolution of Myco-
sequence typing and the BURP (based upon repeat patterns) algorithm. bacterium bovis. Nat Rev Microbiol 4:670–681
J Clin Microbiol 46:2805–2808 Smyth DS, McDougal LK, Gran FW, Manoharan A, Enright MC, Song JH, de
Milkman R, Bridges MM (1990) Molecular evolution of the Escherichia coli Lencastre H, Robinson DA (2010) Population structure of a hybrid clonal
chromosome. III. Clonal frames. Genetics 126:505–517 group of methicillin-resistant Staphylococcus aureus, ST239-MRSA-III. PLoS
Moran NA, McLaughlin HJ, Sorek R (2009) The dynamics and time scale of One 5:e8582
ongoing genomic erosion in symbiotic bacteria. Science 323:379–382 Touchon M, Hoede C, Tenaillon O et al (2009) Organised genome dynamics in
Nelson K, Selander RK (1992) Evolutionary genetics of the proline permease gene the Escherichia coli species results in highly diverse adaptive paths. PLoS
(putP) and the control region of the proline utilization operon in Genet 5:e1000344
populations of Salmonella and Escherichia coli. J Bacteriol 174:6886–6895 Turner KM, Feil EJ (2007) The secret life of the multilocus sequence type. Int
Nelson K, Selander RK (1994) Intergeneric transfer and recombination of the J Antimicrob Agents 29:129–135
6-phosphogluconate dehydrogenase gene (gnd) in enteric bacteria. Proc Natl van Belkum A (2006) Staphylococcal colonization and infection: homeostasis
Acad Sci USA 91:10227–10231 versus disbalance of human (innate) immunity and bacterial virulence. Curr
Nelson K, Whittam TS, Selander RK (1991) Nucleotide polymorphism and Opin Infect Dis 19(4):339–344
evolution in the glyceraldehyde-3-phosphate dehydrogenase gene (gapA) Vanderhaeghen W, Hermans K, Haesebrouck F, Butaye P (2010) Methicillin-
in natural populations of Salmonella and Escherichia coli. Proc Natl Acad resistant Staphylococcus aureus (MRSA) in food production animals.
Sci USA 88:6667–6671 Epidemiol Infect 138:606–625
Novichkov PS, Wolf YI, Dubchak I, Koonin EV (2009) Trends in prokaryotic Waldron DE, Lindsay JA (2006) Sau1: a novel lineage-specific type I restriction-
evolution revealed by comparison of closely related bacterial and archaeal modification system that blocks horizontal gene transfer into Staphylococcus
genomes. J Bacteriol 191:65–73 aureus and between S. aureus isolates of different lineages. J Bacteriol
Nubel U, Roumagnac P, Feldkamp M et al (2008) Frequent emergence and limited 188:5578–5585
geographic dispersal of methicillin-resistant Staphylococcus aureus. Proc Natl Wilson DJ, Gabriel E, Leatherbarrow AJ, Cheesbrough J, Gee S, Bolton E, Fox A,
Acad Sci USA 105:14130–14135 Hart CA, Diggle PJ, Fearnhead P (2009) Rapid evolution and the importance
Nubel U, Dordel J, Kurt K et al (2010) A timescale for evolution, population of recombination to the gastroenteric pathogen Campylobacter jejuni. Mol
expansion, and spatial spread of an emerging clone of methicillin-resistant Biol Evol 26:385–397
Staphylococcus aureus. PLoS Pathog 6:e1000855 Xu BL, Zhang G, Ye HF, Feil EJ, Chen GR, Zhou XM, Zhan XM, Chen SM,
Nulens E, Gould I, MacKenzie F, Deplano A, Cookson B, Alp E, Bouza E, Voss A Pan WB (2009) Predominance of the Hungarian clone (ST 239-III)
(2005) Staphylococcus aureus carriage among participants at the 13th Euro- among hospital-acquired meticillin-resistant Staphylococcus aureus
pean Congress of Clinical Microbiology and Infectious Diseases. Eur J Clin isolates recovered throughout mainland China. J Hosp Infect 71(3):
Microbiol Infect Dis 24(2):145–148 245–255
11 Public Service Collections and Biological
Resource Centers of Microorganisms
David Smith1 . Dagmar Fritze2 . Erko Stackebrandt3
1
CABI, Egham, Surrey, UK
2
German Collection of Microorganisms and Cell Cultures GmbH, Braunschweig, Germany
3
Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures, Braunschweig,
Germany
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 Recent EU Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290

MIRRI: A New Initiative Within the EU Strategy
Prokaryotic Holdings in Public Service Collections . . . . . . . 270 Level of ESFRI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
Gaps and Strengths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
Non-type Strains as an Important Source of Toward a Global Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
Biodiversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
Capacity Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
The Paradigm Shift: From Collections to Biological
Resource Centers (BRCs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 Authenticity Check of Microorganisms . . . . . . . . . . . . . . . . . . . . 293
Meeting the Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Introduction of Quality Management Systems, Access to Microbial Biodiversity Data . . . . . . . . . . . . . . . . . . . . . 295
Accreditation, OECD Best Practice Guidelines . . . . . . . . . 277 CODATA, MINE, CABRI, and GBIF: Examples of Past
Quality and Certification: Costs and Benefits . . . . . . . . . . . 279 and Ongoing Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
User Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 The OECD Best Practice Guidelines . . . . . . . . . . . . . . . . . . . . 296
BRC Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 Minimum and Recommended Data Sets . . . . . . . . . . . . . . . . 297
Financial Sustainability of Public Service Collections
and BRCs and Networks in Both Developed and Long-Term Storage of Microorganisms: At the Very Heart
Developing Nations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 of Collection Activities and Responsibilities . . . . . . . . . . . . . . . 297
Financial Models for Biological Resource Centers . . . . . . 280 Why Do We Need Long-Term Preservation? . . . . . . . . . . . . 297
Culture Collection Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Developing Income Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Choice of Preservation Technique . . . . . . . . . . . . . . . . . . . . . . . . . 299
Commercial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
Research Program Funding . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Factors Influencing the Survival Rate of Microorganisms
Government Department Support . . . . . . . . . . . . . . . . . . 282 During Freezing and Drying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Sponsorship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Protective Suspension Media for Freezing or (Freeze-)
Other Financial Aspects of Operating a BRC Drying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Network (as Identified by the OECD Workshop Simple Methods for In-House Purposes . . . . . . . . . . . . . . . . 300
on Funding Models) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
Strategic Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282 Preferred Methodologies for Long-Term Storage: Freeze-
Operational Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 Drying and Cryo-storage in Liquid Nitrogen . . . . . . . . . . . . . 301
General Aspects of the Freeze-Drying Process . . . . . . . . . . 301
Collections and Their Users: The Need to Know Each Practical Aspects of the Freeze-Drying Process . . . . . . . . . 301
Other Better . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 Media for Cultivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Age of Cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Scientific-Technical Cooperation Among Microbial Ampoules for Freeze-Drying . . . . . . . . . . . . . . . . . . . . . . . . 301
Culture Collections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284 Protective Medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Sterility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Examples for Joint Activities and Global Cooperation . . . . 289 End Vacuum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
EU Projects Tackling Quality Issues of Data and the Sealing Off of Ampoules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Biological Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 True Freeze-Drying Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
268 11 Public Service Collections and Biological Resource Centers of Microorganisms
Centrifugal Freezing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302 a EU-funded network of 15 European culture collections of

The Double-Vial, Liquid-Drying Method, as Applied microorganisms and cell cultures (2001–2004).; ECCO, Euro-
in the DSMZ for Bacteria and Fungi . . . . . . . . . . . . . . . . . . . . 302 pean Culture Collection’s Organisation (http://www.eccosite.
Opening of Ampoules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302 org), a network of European Culture Collections and BRCs
Cryopreservation In or Above Liquid Nitrogen . . . . . . . . . 302 (see > Box 11.4).; EMbaRC, European Consortium of Microbial
General Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302 Resource Centers (http://www.embarc.eu/) networking the
Miniaturized Method for the Cryopreservation of BCCM™/LMG, Belgium; CECT, Spain; CIP, France; DSMZ,
Bacterial Cultures: The Glass Capillary Method . . . . . . . . 303 Germany and two French research collections INRA-CIRM-BP
in Tours and CIRM-BIA, Rennes), aiming to improve, coordi-
nate and validate microbial resource center (MRC) delivery to
Abbreviations ACM, The Asian Consortium for the Conserva- European and International researchers from both public and
tion and Sustainable Use of Microbial Resources (http://www. private sectors (> Box 11.4).; ENBI, European Network of Bio-
nbrc.nite.go.jp/e/project01-e.html), a network of Asian culture diversity Information (www.enbi.org), an EU funded project
collections and BRCs (see > Box 11.5).; ATCC™, American Type established to include all European national nodes of the Global
Culture Collection; Manassas; VA, USA.; BCC, Biotech culture Biodiversity Information Facility (GBIF). (2003–2006).; ESFRI,
collection, Bangkok, Thailand.; BCCM™/LMG, Belgian Co- European Strategy Forum for Research Infrastructures
Ordinated Collections of Micro-Organisms, Laboratorium a strategic instrument to develop the scientific integration of
voor Microbiologie, University Gent, Belgium.; BCCUSP, Europe and to strengthen its international outreach (http://ec.
Brazilian Cyanobacteria Collection, University Sao Paulo, Bra- europa.eu/research/infrastructures/index_en.cfm?pg=esfri);
zil.; BRC, In the context of this chapter defined as a microbial FEMS, Federation of European Microbiological Societies
Biological Resource Center (sensu OECD), a CC running under (http://www.fems-microbiology.org).; GBIF, Global Biodiver-
a defined quality management system which yet needs to be sity Information Facility (http://www.gbif.org/), an interna-
agreed upon by the stakeholders.; CABI, CAB International, tional government-initiated and funded initiative focused on
Egham, UK; CABRI, Common Access to Biological Resources making biodiversity data free and openly available online.;
and Information (www.cabri.org), a EU-funded network of GBRCN, Global Biological Resource Center Network (http://
eight European Collections (1996–1999), (http://www.cabri. www.gbrcn.org), a project following work in the OECD to
org).; CBD, Convention on Biological Diversity (http://www. improve access to high quality biological resources and infor-
cbd.int/), a global agreement addressing all aspects of biological mation to support research and biotechnology as a platform for
diversity: genetic resources, species, and ecosystems. Their pro- a knowledge-based bio-economy.; INRA CIRM-BIA, Center
tection, sustainable use and access to including benefit sharing of International de Ressources Microbiennes - Bacteries d’Interet
the advantages arising from their use.; CBS, Centraalbureau voor Alimentaire, Institut National de la Recherche Agronomique,
Schimmelcultures, Utrecht, The Netherlands.; CC, In the con- Rennes, France.; INRA CIRM-BP, Center International de
text of this chapter defined as a microbial Culture Collection, Ressources Microbiennes – Bacteries Pathogenes, Institut
a general term of a facility accessioning and maintaining micro- National de la Recherche Agronomique, Nouzilly, France.;
bial resources (prokaryotes, fungi, yeast), DNA, plasmids, KACC, Korean Agricultural Culture Collection, National Insti-
phages, and material derived therefrom. Public Culture Collec- tute of Agricultural Science and Technology, Suwon, Republic of
tions provide this material to users. For a comprehensive list of Korea.; KTCT, Korean Collection for Type Cultures, Korea
abbreviations see WDCM and http://www.bacterio.cict.fr/col- Research Institute of Bioscience and Biotechnology, Taejon,
lections.html.; CCAP, Culture Collection of Algae and Protozoa, Republic of Korea.; LMG, The Belgian Consortium of Collec-
Scottish Marine Institute, Oban, Argyll, UK.; CCMM, Moroccan tions of Microorganisms (BCCM™), represented by the
Coordinated Collections of Micro-organisms, Morocco.; CCMP, Universiteit Gent, Belgium.; MINE, Microbial Information Net-
Culture Collection of Marine Phytoplankton, Bigelow Labora- work Europe, an EU-funded network of European culture col-
tory for Ocean Sciences, West Boothbay Harbor, Maine USA.; lections, running between 1986–1989 and 1990–1993.; MIRRI,
CCTCC, Chinese Center for Type Cultures Collections, Wuhan Microbial Resource Research Infrastructure (http://www.mirri.
University, Wuhan, Hubei, China.; CCUG, Culture Collection org/), a pan- European distributed research infrastructure
of the University of Göteborg, Institute of Clinical Bacteriology, established on the European Strategy Forum for Research Infra-
Immunology, and Virology, Göteborg, Sweden.; CECT, structures (ESFRI) road map with the goal to improve access to
Colección Española de Cultivos Tipo, Valencia, Spain.; the microbial resources and services that are needed to accelerate
CGMCC, China General Microbiological Culture Collection research and discovery processes.; MOSAICC, Micro-Organisms
Center, Institute of Microbiology, Chinese Academy of Sciences, Sustainable use and Access regulation International Code of
Beijing, PR China.; CIP, Collection of the Institut Pasteur, Paris, Conduct, an EU-funded project (1997–1999), a tool to support
France; CPCC, Canadian Phycological Culture Center (formerly the implementation of CBD at the microbial level, in accordance
known as UTCC), University of Waterloo, Waterloo, ON, Canada.; with other relevant rules of international and national laws.;
DSMZ, Deutsche Sammlung von Mikroorganismen und MUM, Microtheca do Universidade do Minho, Braga, Portugal.;
Zellkulturen GmbH, Braunschweig, Germany.; EBRCN, European NBRC, Biological Resource Center, National Institute of Tech-
Biological Resource Centers Network (http://www.ebrcn.net), nology and Evaluation, Chiba Pref., Japan.; NCAIM, National
Public Service Collections and Biological Resource Centers of Microorganisms 11 269
Collection of Agricultural and Industrial Microorganisms, Before the globalization of information by the internet,
Department of Microbiology and Biotechnology, University of printed catalogues were the only means by which the users
Horticulture and Food Industry, Budapest, Hungary.; NCCB, could have an insight into collection holdings. Thus, collections
Netherlands Culture Collection of Bacteria, Utrecht, The Neth- usually served the national market, and only a few highly visible
erlands.; NCIMB, National Collection of Industrial and Marine collections operated on an international level. The common
Bacteria, National Collections of Industrial, Food and Marine goals of collections and their same basic operations for collec-
Bacteria, Aberdeen, UK.; NCTC, National Collection of Type tion functioning triggered the need for better cooperation which
Cultures, Central Public Health Laboratory, London, UK.; started after the mid-1980s. National networks were created,
NRRL, Northern Regional Research Center, Agricultural some of which, such as the Belgian Network and more recently
Research Service Culture Collection, National Center for Agri- the Chinese and Brazilian networks, proved successful, while
cultural Utilization Research, US Department of Agriculture, others, such the British and the US networks of culture collec-
Peoria, Illinois, USA.; OECD, Organisation for Economic Co- tions, passed through some stormy times. Regional networks
operation and Development (http://www.oecd.org/).; SAG, Cul- have operated since the early 1980s, first in Europe but now also
ture Collection of Algae Sammlung von Algenkulturen, in East Asia. The need for better harmonization among those
University Göttingen, Göttingen, Germany.; UCL, The Belgian collections which provide users in academia and bio-industry
Consortium of Collections of Microorganisms (BCCM™), with biological material was recognized by Organization for
Université Catholique de Louvain, Belgium.; WDCM, World Economic Cooperation and Development (OECD). From
Data Center of Microorganisms, an activity of the WFCC, pro- 1998, this organization, together with collection managers and
viding an electronic gateway to databases on microbes and cell representatives of governments and industry from member and
lines and resources on biodiversity, molecular biology and associated states, developed a strategic plan to improve the
genomes (see > Box 11.3).; WFCC, World Federation for quality of management and operation of collections for their
Culture Collections (http://www.wfcc.info/), a Federation own benefit and for the benefit of the user, specifically to play
within the International Union of Microbiological Societies a key role in underpinning the knowledge-based bio-economy.
(IUMS, http://www.iums.org) (see > Box 11.2). The growing market in Biotech R&D in the USA, China, India,
Brazil, and Europe required the improvement of the operation
of providers of biological material as their authenticity, purity,
Introduction and the availability of associated database resources were con-
sidered indispensible for the success of downstream processes.
Culture collections have been preserving organisms and supplying As a result, the role of those collections that agreed to follow
them for research and development for over a century. The term the route of higher quality has changed dramatically, culminat-
‘‘culture collection’’ actually does not reflect a common standard, ing in the introduction of the term Biological Resource Center
however, since tasks, holdings, size, funding system, affiliation, (BRC) to reflect the delivery of services and products compliant
mandate, and other parameters differ widely. Though the basic with a standard agreed by national authorities. BRCs focus on
principles of operation are the same, namely, accessioning, main- the following quality criteria:
tenance, and provision of microorganisms, collections may sig-
● Achieving the primary objective to maintain strains in
nificantly differ from each other, and hardly any two collections
a viable state without morphological, physiological, or
operate under the same system. To name a few extremes, printed
genetic change
or electronic catalogues may be missing completely, while other
● Implementing best practice in the provision of services by
collections display their holdings electronically in a most profes-
ensuring:
sional way, while collections may maintain a very regional selec-
– Authentication of biological materials
tion of microbial strains of a narrow taxonomic or physiological
– Validity of data
range; others try to accession the complete range of validly named
– Continued availability and reproducibility of materials
type strains, and yet others access their strains from geographically
– Safe and legitimate shipping
diverse regions; some collections are affiliated and funded as part
– Legitimate acquisition of biological material
of an academic institute; others receive strong governmental
– Compliance with biosafety and biosecurity guidance
support, while others are in charge of nonpublic strains used by
– Protection of intellectual property rights, particularly for
the biotech industry. Though the range of collection types is vast
patents
and many are not even visible to the public, information is
● Applying long-term methods of preservation essential to ensure
available for those 592 collections (status May 2011, including
availability of biological materials for the long-term
those with non-microorganism holdings) which are registered at
– Selection of most suitable method
the MIRCEN-World Data Center for Microorganisms (WDCM)
– Optimization
(http://www.wfcc.info/index.php/wdcmdb/) which is overseen
– Viability, purity, and stability
by the World Federation for Culture Collections (WFCC).
Statistics on these collections and summaries of kind and Meeting the requirements of BRC status requires investment
number of holdings, number of staff, funding system, and and change. While it will be feasible for some of the well-funded
services offered are compiled in the WDCM. public service collections to implement international accepted
guidelines, others will need more support, both strategically as 160000
well as financially. The mandate to develop a framework for the

evolution of collections to BRC status is being undertaken by the 140000
Global Biological Resource Center Network, which is supported all strains

120000 type strains
by regional activities (e.g., MIRRI in Europe). Additional exper-
non-type strains
tise in various areas of collection-related issues, such as the
validly named type
CBD, intellectual property issues, material transfer agreement, 100000
strains
biosecurity, and the like, is being harnessed to deliver common
policy and strategy for implementation. An overview of the legal 80000
and regulatory environment in which microbiologists and in

particular microbial service collections dwell and their reactions 60000
is given in Fritze (2010).
This chapter will highlight some of the recent developments, 40000
focusing on the core activities of any type of collection of
microorganisms, and will describe in detail the way forward to 20000
achieve the goal of their global networking.
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Prokaryotic Holdings in Public Service . Fig. 11.1

Collections Cumulative display of accessions of type and non-type strains by
selected public collections over a period of 11 years. Collections
On 30 May 2011, a total of 1,751,439 microbial resources were included are ATCCTM, USA; BCC, Thailand; BCCM™/LMG, Belgium;
listed in the 592 culture collections in 68 countries registered in CCM, Czech Republic; CCMM, Morocco; CCUG, Sweden; CCTCC and
the WDCM: about 761,000 of them were bacteria and 506,500 of CGMCC, China; CECT, Spain; CIP, France; DSMZ, Germany; KACC
them were fungi. Other holdings embrace bacteriophages, plant and KCTC, Korea; NBRC, Japan (starting 2002); NCAIM, Hungary;
and animal viruses, microscopic algae, protozoa, dediffer- NCCB, the Netherlands; and NCIMB, the UK
entiated plant cells, and immortalized human and animal cell
lines. Collections of microorganisms maintain two categories of access and overcomes costly shipping and lengthy documentation
strains with relevance to taxonomy, type strains, and non-type procedures. The graph in > Fig. 11.1, showing the cumulative
strains. For prokaryotes, the deposition of type strains, the accessions, clearly displays the significant uptake of non-type
nomenclatural type to which the binominal species designation strains from 2005 on which may be a reflection of increased
is linked, is mandatory (Tindall et al. 2006). This means that no interest in biodiversity studies and the emergence of the bio-
name will be valid without the written confirmation of at least economy. The same trend is visible in the duplication of num-
two public collections in two different countries that the respec- bers of new type strains around 2005, a direct reflection of the
tive strain has been deposited without restrictions and checked isolation of novel pheno- and genotypes. It must be noted that
for authenticity by the original depositor (for descriptions not all public collections follow the same accessioning strategy:
before 2006, the type may be available in a single collection while some (e.g., BCC, CCUG, CGMCC, CCMM, CCTCC, or
only). This procedure was internationally accepted in order to NCCB; see Abbreviations) concentrate on the deposition of
make the type available as reference for scientific studies. Non- non-type strains (75 %), others (e.g., ATCC™, BCCM™/
type strains can be either authenticated strains of a described LMG, CCM, CECT, DSMZ, CIP, KACC, NCAIM, or NBRC)
species or be any taxonomically less well-identified strain found accession about as many type strains as non-type strains (35–60
worthwhile maintaining by scientists and collection managers %), while among the collections surveyed, only NCIMB and
for their specific properties. A 2011 survey on holdings of pro- KCTC concentrate on the deposition of type strains (70–75 %).
karyotes in several of the major and some of the smaller public While deposition of type strains is free of charge, the increas-
collections (see legend to > Fig. 11.1) clearly indicated the higher ing number of newly described species (> Fig. 11.1) challenges
number of deposits of non-type strains over type strains. Over the collections by increasing the necessary manpower and mainte-
past 11 years, about 136,000 strains were accessioned by these nance costs. These costs include not only the administrative
collections, about 80 % of which were non-type strains. In the responsibilities according to the CBD and other rules and regu-
same period, 5,412 type strains were validly named and described, lations (Smith et al. 2008) but also identification, authentica-
meaning that type strains were, on average, distributed to five tion, maintenance, and long-term storage, as well as the
public collections. Duplication in collections at a reasonable level generation of accompanying bio-information and regular tech-
is considered good practice as backup. The decision to maintain nological updates and training. Some revenues are generated by
copies of strains depends upon a wide range of scientific and user- providing strains to users, but this income does not cover the
related interests: it guarantees the long-term maintenance of the costs involved in state-of-the-art maintenance of resources. This
prime reference strain in bacteriology, while the rapid provision indicates that even with the descriptions reaching a plateau in
of these strains to users at the regional/national level facilitates the next years (due to the lack of systematists, not of novel
800
700
600
500
400
300
Europa
200 Japan
Korea
100
USA
0
2004 China
2005
2006 others
2007
2008
. Fig. 11.2
Deposition of new strains in culture collections per country/region between 2004 and 2008 (Courtesy of Ken Suzuki, NBRC, Japan)
organisms), the workload of maintaining and providing novel of under-sampled or low diversity) phyla are represented by a few
biodiversity under high quality standards remains a huge task type strains only. The policy of mandatory deposition, however,
for accessioning collections. guarantees that the type strain is available from at least one public
As shown in > Fig. 11.2, European collections were used most collection (e.g., the respective type strains of species of Fibrobacter
frequently as repositories for new strains over the last years (data and Lentisphaera are held in the ATCCTM and ATCCTM and
available from 2004 to 2008). The decline of the USA collections as KCTC, respectively). Obvious gaps detected are within the
primary source for deposition is worth noting as is the increased Tenericutes (formerly Mollicutes) and within Cyanobacteria.
deposition in Korea, Japan, and China, countries that today Mollicutes embrace primarily parasites of various animals and
contribute most to the description of novel species. It is to the plants, living within the host’s cells. Their maintenance often
advantage of new public collections, mainly in East Asia and requires host tissues which are out of range for most resource
Brazil, that their planning fell in the times of increasing awareness centers. Collections of Cyanobacteria exist in several countries
of microbial diversity and the genomic revolution, resulting in the (e.g., ATCC™ and CCMP, USA; PCC, France; UTCC, HAMBI,
proper provision of infrastructure for their future tasks. Finland; Canada; CCAP, UK; SAG, Germany; BCCUSP, Brazil)
and often in conjunction with collections of eukaryotic algae.
At the species level gaps are obvious in the so-called rare
Gaps and Strengths species, here, the more specialized collections show their
strength, especially in holdings of the extremophiles. These
Public collections do not only differ in numbers and types of species are usually less frequently requested than species of
material deposited, they also differ significantly in the range of medical and biotechnological interest, and their maintenance
taxa maintained. A worldwide comparative analysis of holdings is more demanding than those of the majority of aerobic
of prokaryotes in WFCC/WDCM member collections, as and heterotrophic species in the phyla Actinobacteria,
displayed individually in their respective strain catalogues, is Proteobacteria, Firmicutes, and Bacteroidetes. Other gaps are
not available; a 2009 survey of some West European collections, also seen among some of the genera of obligate endosymbionts
members of the EMbaRC project (see Abbreviations), most and pathogens, as well as among the obligate chemolithotrophs.
likely mirrors the situation at a more regional level, such as Only 12 % of all genera described are not covered by any of the
those existing in North America and East Asia. The range of six EMbaRC collections surveyed. These are either those recently
gaps at the generic level is rather small as the majority of phyla described, embracing obligate endosymbionts or fastidious path-
are covered at least by some strains. At the genus level, most ogens. More telling than genus numbers, however, is the availabil-
phyla are covered above 80 % (> Table 11.1); only the mono- ity of type strains and range of diversity covered at the strain level.
generic phyla Fibrobacteres and Lentisphaera are not covered in In this respect, too, collections differ widely from each other (for
any of these collections, and some of the ‘‘rare’’ (rare in the sense EMbaRC collections, see > Table 11.2). Certainly, the history of
. Table 11.1 Based upon the range of genera and species covered, collec-
In percentage coverage of bacterial diversity at the genus level by tions fall into one of several types. One type, represented by,
six EMbaRC collections (Survey from 2009) for example, DSMZ, ATCC™, or NBRC, covers in-depth genera
and type strains. Members of a second type do not, or do not
Number of Number of
only, concentrate on phylogenetic diversity, habitat, metabo-
described genera Percent
Taxon genera covered coverage lism, or ecology but are also specifically strong in holdings of
intra-generic and often intraspecies diversity, representing either
Archaea 91 86 95 a pathogenic potential (to humans, animals or plants), or organ-
Bacteria isms relevant to biotechnology (food, agriculture, pharmacy).
Aquificae 12 12 100 The CIP, ATCC™, NCTC, or CCUG are well known for their
Thermotogae 6 5 83 holdings of pathogens, while the BCCM™/LMG collection,
Thermodesulfobacteria 4 4 100 NRRL, and especially collections in subtropical and tropical
Deinococcus/Thermus 6 6 100
regions maintain in-depth diversity of nitrogen-fixing bacteria
and plant pathogens. The collections maintain a higher number
Chrysiogenetes 1 1 100
of strains per taxon than those of the first category. The research
Chlorobi 3 3 100
collections usually cover a very narrow spectrum of genus diver-
Chloroflexi 13 10 77 sity but in-depth coverage of intraspecies diversity. This is clearly
Thermomicrobia 1 1 100 shown by the example of the two INRA collections in which
Nitrospirae 3 2 67 holdings are strains involved in either milk or cheese processing
Deferribacteres 6 5 83 (CIRM-BIA) or in pathogenicity (CIRM-BP).
Synergistetes 5 3 60
Planctomycetes 9 7 78
Non-type Strains as an Important Source of
Fusobacteria 9 7 78
Biodiversity
Chlamydiae 5 1 20
Spirochaetes 14 7 50 The increasing deposition of non-type strains is due to several
Fibrobacteres 2 0 0 factors. Firstly, the strains can originate from a collection’s own
Acidobacteria 6 5 83 research, enlarging the holdings of those taxa which are in the
Verrucomicrobia 13 8 62 prime focus of individual curators. Though often the number of
Dictyoglomi 1 1 100 isolates originating from environmental studies is too high to
Gemmatimonadetes 1 1 100
maintain the complete set of the strains isolated, the short intra-
collection distances and the in-house availability of maintenance
Lentisphaerae 1 0 0
procedures favor rapid and competent deposition.
Bacteroidetes 181 148 82
A second source of strains originates from research labora-
Firmicutes 312 289 93 tories. These facilities may work independently or in a network
Actinobacteria 238 228 96 of collaboration, including public collections. With the advent
Proteobacteria of molecular ecology and the growing awareness of the huge, yet
Alphaproteobacteria 235 188 80 unexplored biodiversity of microorganisms from the early 1990s
Betaproteobacteria 143 131 92 on, increased funding has strongly supported research in geno-
Gammaproteobacteria 269 232 86
mics, metagenomic, and functional diversity. Microbiological
research also benefitted from increased funding for sampling
Deltaproteobacteria 84 76 90
due to the awareness that linking sequences to function requires
Epsilonproteobacteria 15 12 80
research on the organism itself and due to the inquisitiveness of
scientists to isolate the organism for which the molecular data
a collection significantly determines the size and phylogenetic signaled phylogenetic uniqueness. The heydays of the recovery
affiliations of individual holdings, and the expertise of curators is of extremophiles and the discovery and description of novel
determined to a great extent by the history and tradition of phyla and classes fall into the last two decades. While many of
collections (to name a few, the actinobacterial collection in the the more recently described novel taxa rarely embrace more than
DSMZ and NBRC, Japan; the Bacillus and mycelium-forming the type strain of the type species only, it can be assumed that
actinomycetes holdings in the NRRL, USA; the Lactobacillus more strains of these taxa are hidden in research collections.
holdings in the CIP, or the Vibrio, Pseudomonas, or Enterococcus Normally, the result of the lack of proper identification and
collections in the BCCM™/LMG). The history, including the demanding circumscription protocols, not to mention the lack
research emphasis of the collection founders, also explains the of funding for taxonomic research, which is not matched at all to
strength in methods used in-house for authentication and char- the support for expensive sampling expeditions and isolation
acterization and the range of skills offered to the public, for regimes. It is mostly up to the individual scientist to identify
example, providing identification service and training courses. non-type strains that are worthwhile depositing, and this is done
. Table 11.2
Comparison of genera and type strains described per phylum and holdings in EMbaRC collections, representing a regional network
(Survey from 2009)
DSMZ CIP BCCMTM/LMG CECT INRA CIRM-BP INRA CIRM-BIA

Phyla Genera/species describeda Germany France Belgium Spain France France
Archaea 91/379
Crenarchaeota 26/56 24/52 – – – – –
Euryarchaeota 65/322 58/276 – – 12/20 – –
Bacteria 1,595/9,056
Aquificae 12/27 12/25 1/1 – – – –
Thermotogae 6/31 5/28 2/3 – – – –
Thermodesulfobacteria 4/8 4/8 1/1 1/1 – – –
Deinococcus/Thermus 6/61 6/52 2/15 3/26 1/2 – –
Chrysiogenetes 1/1 1/1 – – – – –
Chlorobi 3/15 3/9 3/8 – – – –
Chloroflexi 13/20 9/12 – – – – –
Thermomicrobia 1/2 1/1 – – – – –
Nitrospirae 3/10 2/6 1/1 – – – –
Deferribacteres 6/9 4/6 2/2 – – – –
Synergistetes 5/5 2/2 1/1 – – – –
Planctomycetes 9/14 6/9 – – – – –
Fusobacteria 9/45 5/33 4/15 1/1 – – –
Chlamydiae 5/10 1/1 – – – – –
Spirochaetes 14/112 7/42 4/26 – – – –
Fibrobacteres 1/4 – – – – – –
Acidobacteria 6/7 5/6 – – – – –
Verrucomicrobia 13/31 5/7 3/3 – – – –
Dictyoglomi 1/3 1/3 – – – – –
Gemmatimonadetes 1/1 1/1 – – – – –
Lentisphaera 1/1 – – – – – –
Bacteroidetes 181/711 105/453 107/340 76/249 13//18 5/2 –
Firmicutes, Tenericutes 312/1,962 269/1.752 146/936 76/730 42/247 1/2 7/73
Actinobacteria 238/2,381 226/2.293 147/1.069 79/569 46/191 2/3 3/40
Proteobacteria 746/3,582 572/2.618 367/1.645 277/1.334 126/351 50/157 –
Total number of strains 19.735 22.452 16.505 2.658 2.239 3.082
a
It should be noted that the total number of genera and species includes synonyms (see http://www.bacterio.cict.fr/number.html). This is due to the fact that the
name of some species appears in two or more genera (synonyms), depending on the number of reclassifications (names once validly published or notified will
remain valid irrespective of its present classification). The deviations from the actual numbers will remain uncorrected as the visualization of holdings in the
individual collections will only be slightly affected
in an environment of underfunding of public collections. Gen- has been the post-emeritus transfer of the collection containing
erally, the number and kind of non-type strains to be deposited myxobacteria and cytophagas from Hans Reichenbach to the
is a matter of negotiation between scientist and curator. Rarely DSMZ or the Seeliger collection of Listeria strains transferred to
are entire research collections transferred to public collections; Mark Achtman, Cork, Ireland.
in most cases, such collections consist of unique holdings which Surprisingly, it is only recently that funding bodies have
are at high risk of being discarded or transferred to a new facility developed an interest in microbial collections established in
with unknown long-term perspective. The WFCC established the course of projects funded with taxpayer’s money. Starting
a specific task group in order to provide a focal point of call for with the long-term and secure availability of taxon-affiliated
any collection (industrial/private/academic) which itself con- data of mainly eukaryotes, such as the barcode of life sequences,
siders to be endangered or in need of help or advice with respect environmental observatory data, remote sensing, or geographic
to its future sustainability. One example for a successful rescue atlas of plants and animals (to name only a few), microbiological
information was restricted to gene and genome sequences. (mainly staphylococci, mycobacteria, Clostridia, enterobacteria,
But it is the living culture that is needed in order to verify data Acinetobacter, Burkholderia, Chlamydia) which accumulate rap-
and to explore new scientific horizons on the basis of the depos- idly in daily hospital routine and represent in almost every case,
ited information. This situation is slowly changing as funding isolates of described species and few exhibiting new properties.
agencies place more emphasis on measures for appropriate In the second survey, only 0.94 % found their way into public
maintenance of research collections and their evaluation; they collections. Release of material and/or deposition in public
foster collaboration with the expertise of curators working in collections is left to the authors’ discretion; although some
acknowledged collections; they even financially support basic journals may have a stricter implementation policy than others,
taxonomic groundwork in order to allow collection curators to enforcement mechanisms do not exist for those frequent cases
objectively judge the novelty of strains which could complement where authors deny sharing the requested material.
their holdings with the goal to broaden the biodiversity of taxa During a recent EMbaRC meeting of editors, collection
for research in general. Only a small fraction of strains managers, and authors, it was confirmed that, though access to
maintained in research collections will be deposited in public published material may work smoothly among scientists in
collections, unless a completely novel long-term storage strategy certain disciplines and tightly knit scientific communities, access
is developed. Here, the same criteria could be applied to those overall is dismal. Discussion on a strategy to enhance and facil-
listed in > Box 11.1 for post-publication deposits. It should also itate access to microbial resources was done with awareness that
be stressed that researchers should communicate specific experi- deposition of all microbial strains is neither necessary nor achiev-
ence on growth and maintenance for fastidious strains to collec- able under the present funding system of public repositories. The
tion curators prior to deposition in public collections. Usually, rationale for recommending deposition in public collections was
curators lack experience needed for members of mainly higher not based on the concern that authors are incapable of short-term
and novel taxa for which no strain had been deposited before, and handling of research strains; it was based on the fact that microbial
they need to be informed before the arrival of such strains in order resource centers have decades of experience in handling,
to optimally preserve them long-term. The number of isolates in safeguarding, and shipping a wide range of diverse materials that
research collections worldwide cannot be estimated, and, likewise, is otherwise prone to involuntary extinction by negligence or
the percentage of novel strains worthwhile depositing cannot be deliberate clearing of laboratory holdings. Against this background,
assessed. The transfer of research isolates to the safe environment a set of selection criteria were recommended that would allow all
of a public collection will only be decided through an intense stakeholders to prioritize material for deposition (> Box 11.1).
dialog between curators and researchers in academia.
The third source of strains is currently almost unavailable to
the scientific public: namely, those strains which are included in
Box 11.1 Post-publication Deposit of Microbial Strains to
the scientific literature. The instructions to authors of almost all
Underpin Good Practice in Science
peer-reviewed journals state that the editors expect that new and
variant organisms, viruses, and vectors described in journals will Despite recommendations to release to the community microbial
be made available to all qualified members of the scientific resources post-publication, the reality is far from satisfying.
community. Some journals even explicitly encourage authors A recent workshop discussed the need for a coordinated and
to deposit important strains in publicly accessible culture col- effective deposition policy and proposed a set of criteria to
lections and to refer to the collections and strain numbers in the facilitate deposition into public service collections (Biological
text (e.g., FEMS journals). The Guide to Authors in Nature Resource Centers) of ‘‘key’’ prokaryotic strains.
publications states that resources should be made available in The workshop participants decided against a mandatory
order to allow others to ‘‘replicate and build upon the authors’ post-publication deposition of microbial strains but agreed on
published claims’’ (http://www.nature.com/nature/authors/gta/ a set of criteria based on the phylogenetic, metabolic, and geno-
#a1.3), to check when aberrant results are discovered or to mic uniqueness of ‘‘key’’ strains worthy of deposit. The checklist
reevaluate the strains when new technologies are available. would also contain the contact addresses of a range of public
Though the number of deposits of non-type strains indi- service collections together with their taxonomic priorities to
cated in > Fig. 11.1 over a period of 11 years sounds impressive, it facilitate contact between authors and curators. Completion of
is only a minute fraction of strains annually included in scientific this checklist would be mandatory prior to manuscript submis-
studies. To give only two examples, in the first two issues of sion. The definition of ‘‘key’’ strains should be seen as a first but
Volume 46 (2008) of the Journal of Clinical Microbiology, around not exclusive step to initiate the strain sharing strategy; environ-
32,000 strains of mostly clinical origin were included, while about mental samples, including as-yet-uncultured microorganisms,
20,000 strains were included in the publications of the 2008 metagenome libraries, and other material should also be consid-
volumes of ten European microbiology journals covering mostly ered medium-term. The following criteria were agreed upon:
applied and ecological topics (Stackebrandt 2010). However,
● Uniqueness, based on a cutoff point of 98% of 16S rRNA gene
hardly any of them were deposited in public collections
sequence similarity to the most closely related species with
for long-term availability. In the first example, only 0.03 %
a validated name. This sequence is currently the gold standard
of strains investigated were deposited which is perhaps not
for phylogenetic affiliation of an isolate at the genus level.
surprising considering the taxonomic affiliation of these strains
● Metabolic uniqueness, based on the presence of a new path- The workshop did not address the financial consequences of
way, modification of an existing pathway, metabolic differ- enhanced deposition but, considering the urgency to act now,
ences compared to the type strain or the production of novel funding agencies need to reevaluate their responsibilities by
products. providing long-term and increasing support for public reposito-
● Genomic uniqueness, such as significant differences (20%) ries to allow these tasks to be performed (Emerson and Wilson
in genome size, genome architecture, or new regulatory 2009; Stackebrandt 2010, 2011)
mechanisms.
● Resources and parts thereof with fully sequenced genomes E. Garay-Aubán4 . P. Kämpfer5 . Y. Lecuona6 . J. Prosser7 . R.
(prokaryotes, phages, plasmids). Roséllo-Mora8 . K.-H. Schleifer9 . K. Smalla10 .
● A second strain of those species or subspecies for which only D. Smith11 . E. Stackebrandt12
4
the type strain has been deposited. For 79% of new species CECT-Spanish Type Culture Collection, Parc Cientific University
described in 2009, only the type strain is available. of Valencia, Paterna, Valencia, Spain
5
Department for Applied Microbiology, Justus-Liebig University,
A survey among scientists was carried out to determine the
Giessen, Germany
reasons for the lack of materials in public service collections and 6
EMbaRC, INRA, Rennes, France
whether they felt improved access was needed. Of the 3,950 7
Institute of Biological and Environmental Sciences, University of
scientists in 49 countries who were asked to participate, 517
Aberdeen, Aberdeen, UK
responded (13.1%). When asked if they had encountered prob- 8
Department of Microbiology, University of the Balearic Islands,
lems in accessing strains, 76.8% indicated that they had encoun-
Palma de Mallorca, Spain
tered problems, frequently to always, when asking for strains. 9
Department of Microbiology, Technical University Munich,
Around 50% indicated that they received no response at all,
Emil-Ramann-Str. 4, Freising, Germany
others were requested to pay for the strain and some were 10
Federal Research Center for Cultivated Crops, Institute for
denied access because of patent issues. Almost 87% agreed
Epidemiology and Pathogen Diagnostics, Julius-Kühn Institut,
there was a need to improve access to microbial resources and
79% agreed that journal publication guidelines should request 11
CABI Egham, UK
that strains with particular properties, such as those listed above, 12
DSMZ- Deutsche Sammlung von Mikroorganismen und
must be deposited in public culture collections to maintain them
Zellkulturen GmbH, Braunschweig, Germany
for further research.
This response suggests that the responders believe that
a behavioral change is necessary and that journals should request
that strains associated with publications be deposited. The rea-
sons given for lack of response or failure to receive strains were
The Paradigm Shift: From Collections to
specifically indicated by about 100 scientists but are subject to
Biological Resource Centers (BRCs)
conjecture, being a mixture of guesses and author citations. In
Culture collections need to adopt new technologies and work
approximately 39% of cases it is feared that researchers simply
toward providing what today’s research community needs. BRCs
want to protect their research from exploitation by others. This
need to work together to address these needs through coordi-
appears to be the very opposite of the philosophy behind pub-
nated and harmonized approaches and sharing tasks in a cost-
lication and dissemination of results and conflicts with accepted
effective and appropriate manner.
scientific principles and morals. About 31% referred to the
It is absolutely essential that any service industry moves with
authors response that strains were lost or were unavailable for
the times and the needs of its users. Culture collections are no
nonspecified reasons; 25% referred to quarantine, customs, and
different and have been adapting to change and increasing chal-
biosafety regulations as severe obstacles for releasing strains,
lenges. They have been providing a public service for over
problems that would be better solved by international, experi-
a century essentially collecting and distributing organisms. The
enced BRCs than by individual scientists. Additionally, to protect
core function of providing an authentically named strain has
the investment made using public funds, the research funders
remained but broadened to characterization of their holdings to
must also consider whether they make similar deposit and avail-
a greater extent. Such change is most often an independent deci-
ability requirements on material subject to their funded research.
sion dependent on the sector in which the collection or its host
The workshop participants stressed that authors should make
institution focuses with the consequence that public service col-
every reasonable effort to make material available, if they do
lections have deviated in collection focus and service provided.
not deposit material in public collections, it should be because
Naturally, collections have learnt from one another in introducing
their strains do not meet the above criteria; it was also considered
new approaches and products that have worked for others as well
important that journals and funding agencies police their policies
as introducing their own innovations. More recently, certainly over
and have a mechanism for accepting complaints where access to
the last four decades, change has been driven by consensus through
material is denied. Journals were recommended to introduce
communities such as the WFCC. The WFCC introduced guide-
a mechanism for active agreement by authors to make material
lines for the establishment and operation of culture collections
available when they submit an article.
(http://www.wfcc.info/guideline.htm) to help collections to
provide the best service to the scientific community. A coordinated ● Develop new systems and technologies for the long-term
approach to microbial and cell culture resource has been fostered. maintenance and distribution of large numbers of diverse
New collections are created while others disappear. Over 980 biological resources
WDCM registration numbers have been issued, but only 592 ● Coordinate curation, as well as development and networking
collections remain (> Box 11.3). This has been attributed to of informatics tools for data analysis, comparison, and
several causes, the retirement of individuals who maintained visualization
them, a change of focus of the scientist or institution or the loss ● Ensure that the scientific community has access to affordable
of funding. In the late 1980s, the Japanese government listened to products and services
their scientific community and challenged the OECD to address
The development of BRCs to make available high-quality
sustainability and development of culture collections. The OECD
biological materials for research and development was consid-
responded with the Biological Resource Center Initiative which
ered necessary to underpin biotechnology. BRCs must focus on
now describes the modern-day culture collection as a Biological
adding value to their biological materials and link more inti-
Resource Center and defines them as follows:
mately to the life sciences and bio-industry to help deliver the
" Biological Resource Centers are an essential part of the infrastruc- developing bio-economy. The OECD report ‘‘The bio-economy
ture underpinning biotechnology. They consist of service providers to 2030: designing a policy agenda’’ (2011) emphasizes that the
and repositories of the living cells, genomes of organisms, and biological sciences are adding value to a host of products and
information relating to heredity and the functions of biological services, supporting what some have labeled the ‘‘bio-economy.’’
systems. BRCs contain collections of culturable organisms The report explains that from a broad economic perspective, the
(e.g., microorganisms, plant, animal and human cells), replicable bio-economy refers to the set of economic activities relating to
parts of these (e.g., genomes, plasmids, viruses, cDNAs), viable but the invention, development, production, and use of biological
not yet culturable organisms cells and tissues, as well as data bases products and processes. If it continues on course, the bio-
containing molecular, physiological and structural information rel- economy could make major socioeconomic contributions in
evant to these collections and related bioinformatics’’ (Definition OECD and non-OECD countries. These benefits are expected
based on the one adopted at the 1999 Tokyo Workshop on to improve health outcomes, boost the productivity of agricul-
Biological Resource Centers, where the concept of BRCs as an ture and industrial processes, and enhance environmental
outgrowth of conventional pre-genomics ex situ collections of sustainability. The bio-economy’s success is not, however,
biological materials was developed – and incorporating scientific guaranteed: harnessing its potential will require coordinated
developments since 1999.) BRCs must meet the high standards of policy action by governments to reap the benefits of the biotech-
quality and expertise demanded by the international community nology revolution. The plethora of uses of microorganisms, not
of scientists and industry for the delivery of biological information least their use as reference strains in taxonomy or use in stan-
and materials. They should provide access to biological resources dards, demands access to expertise, technologies, and data
on which R&D in the life sciences and the advancement of analysis.
biotechnology depends. (http://oecdpublications.gfi-nb.com/ The change in how science research is conducted today,
cgi-bin/oecdbookshop.storefront) utilizing new technologies and information, requires culture
collections to adapt in order to provide the resources in a way
that will facilitate their use and enable an accelerated discovery
chain. The transition to the OECD envisaged BRC is the first
Meeting the Challenges
step. The modern-day BRC can support countries by
establishing a means to release the potential of their microbial
The OECD report (2001) on BRCs stresses that to cope with the
resources to provide solutions to national economic, environ-
massive expansion of biological resources, including living
mental, food, and healthcare problems and consequently
biological materials and data on genomics, BRCs need to
contribute to achieving the United Nations Millennium Devel-
opment Goals. This ambitious agenda for reducing poverty and
● Contribute to the coordination of efforts to conserve biodi-
improving lives can be partially delivered by better management
versity and to provide access to natural and engineered
and utilization of biological resources:
biological resources
● Assist in the development of a coordinated international ● Improve livelihoods (Millennium Goal – MG 1).
system for decision-making to guide appropriate acquisi- ● Provide new sources of food and reduce agricultural losses
tion, maintenance, and distribution of biological resources (MG1).
so as to avoid unnecessary duplication of effort while ● Lead to discovery of new drugs and treatments of disease to
preserving critical levels of biodiversity reduce child mortality and improve maternal health (MG 4,
● Modernize to incorporate the latest developments in web- 5, and 6).
based electronic communication, bio-informational science, ● Understand and contribute to environmental stability
and informatics technologies (MG7).
● Coordinate and unify catalogues and databases to meet the ● Develop a global partnership in the conservation and utili-
requirements of science in the developing post-genomics era zation of microbial resources for development (MG8).
● Resources created from the above can be mobilized to pro- development of BRCs. There are several examples of existing
mote gender equality and empower women (MG3) and guidelines for microbial and cell culture collections available:
achieve universal primary education.
● The WFCC Guidelines for the establishment and operation of
The OECD BRC Initiative took into consideration that the collections of microorganisms (http://www.cabri.org)
growing worldwide demand for biological resources provides ● The Microbial Information Network for Europe (MINE)
good reasons for greatly increasing the number and quality of project standards for the member collections (Hawksworth
culture collections. Only a very few large national centers are and Schipper 1989; Stalpers et al. 1990)
able to perform a comprehensive role. A higher number and ● Common Access to Biological Resources and Information
a broader geographic coverage of high-quality public service (CABRI) guidelines (http://www.cabri.org)
collections are needed to reach this goal. The development,
There are also a number of general nonspecific standards
expansion, and survival of these face many challenges. These
and norms that can be applied to microbial laboratories,
include those posed by the molecular revolution (genomics and
such as
the information revealed by DNA sequencing), accelerating
efforts to conserve biodiversity, funding uncertainties that ● Good Laboratory Practice (GLP)
threaten stability, the need for adequate quality assurance and ● ISO 9001 quality management systems
constraints on access to biological resources within countries ● ISO 17025 general requirements for the competence of test-
and across international borders resulting from private ing and calibration laboratories
industry’s protection of investments and industrial secrecy, ● ISO Guide 34 general requirements for the competence of
import/export regulations, intellectual property rights, safety reference material producers
issues, and ethical concerns about the uses of genes and other
Industry is expressing the need for quality control and stan-
biological resources (OECD 2001).
dards within collections. Although publications on collection
management and methodology give information on protocols
Introduction of Quality Management Systems, and procedures, a quality management system must go further
Accreditation, OECD Best Practice Guidelines and set minimum standards (Smith et al. 2001). The CABRI
electronic catalogue project made available a set of guidelines to
The modern-day collection or BRC is an entity compliant with aid collections to put in place best practice (CABRI 2002). These
appropriate national law, regulations and policies, and operates cover critical elements in the handling, storage, characterization,
to internationally validated criteria. The impact of legislation on and distribution of microorganisms and cell cultures and the
the collection, handling and distribution is enormous. Keeping handling of associated information. The EU project (QLRT-
pace with new and changing legislation is absolutely essential 2000-00221) European Biological Resource Centers Network
and places an additional burden on the culture collection. To (EBRCN) ran in parallel to the OECD Task Force. The EBRCN
demonstrate that a BRC is implementing best practice, a third- consortium supported the work of the OECD Task Force by
party independent assessment process is needed. The OECD drawing together the key elements of the above-mentioned
agreed best practice guidance to enable the delivery of high- guidelines and standards to form the basis of the OECD best
quality materials ensuring they are authentic, preserved by practice guidelines for BRCs (OECD 2007). This was formulated
‘‘state-of-the-art’’ technology and that all associated informa- at two levels, the general criteria that can be applied to all BRCs
tion is validated. The OECD best practice guidelines for BRCs and secondly, organism domain specific criteria that are applied
(OECD 2007) outline a process for certification or accreditation to BRCs based on the biological materials they hold. Currently,
of BRCs. The BRC must apply for accreditation through two sets of general guidelines (‘‘general best practice guidelines
a process approved by national governments but either through for all BRCs’’ and ‘‘Best Practice Guidelines on Biosecurity for
an accreditation body recognized by government or through BRCs’’) exist in parallel to two sets of domain-specific guidelines
a transparent accreditation procedure recognized by govern- (‘‘best practice guidelines for the microorganism domain’’ and
ment or directly by government. There are a number of ways ‘‘best practice guidelines for human-derived material’’).
this might be achieved, but it is considered that the process
should be based upon existing systems. Many collections ● The general best practice guidelines for all BRCs cover
although implementing best practices may not wish to go this – Organizational requirements
far. The most appropriate model is one that sets the baseline for – Equipment use, calibration, testing, and maintenance
authentic, well-preserved, and validated strains and requires records
development in excellence (see below). The BRC seeks to add – Documentation management
value to its holdings by further research on the characterization – Data management, processing, and publication
of the strains held to enable improvement in their public service – Preparation of media and reagents
role. It is envisaged that not all culture collections will become – Accession of deposits to the BRC
BRCs, but best practice should be implemented nevertheless. – Preservation and maintenance
The OECD BRC Task Force considered that the establish- – Supply
ment of a common quality standard was a key issue in the – Quality audit and quality review
● The best practice guidelines for the microorganism domain and the core processes in the laboratories. But in addition to
cover these traditional key issues of quality management, mainly cov-
– Staff qualifications and training ered and endorsed by the global standards ISO 9001 or ISO
– Hygiene and biosafety 17025, modern culture collections are faced with far-reaching
– Equipment use, calibration, testing, and maintenance issues in their shift toward the modern BRC as defined by the
records OECD. These issues include social and socioeconomic tasks,
– Preparation of samples sustainable financial management, balancing of commercial
– Information provided with the biological material and scientific interests as equitable stakeholders, linkage to
supplied innovative information technologies, and realization of govern-
Additionally, specific guidance was prepared to cover mental and cross-national legislation in the fields of biotechnol-
potential dual-use organisms and to ensure BRCs ogy and security interests. The transformational change from
implemented practice to ensure biosecurity. Dual-use is a national though networked repository for biological material
a term to refer to any technology or material which can be toward a multitask facility being part of a global infrastructure
used for peaceful and military aims. Thus, the exploitation of for the emerging knowledge-based bio-economy requires not
biological material and biotechnology is an essential part of only an enlargement of managerial requirements but also a new
the international dual-use regulations with regard to bioter- mutual standard in quality management. Taking up the neces-
rorism and bioweapons. sity to standardize and systemize the core activities of a BRC
● The best practice guidelines on biosecurity for BRCs cover within a special tailored guideline covering most of the key
– Assessing biosecurity risks of biological material issues, the OECD best practice guidelines have not only a high
– New acquisitions/reassessment of inventory coverage of all requirements in one single standard, but in
– Biosecurity risk management practices addition, they cover a broad spectrum of the requirements
– Physical security of BRCs delivered by other standards. In fact, the new guidelines for
– Security management of personnel and visitors BRCs offer an integrated approach to support a culture collec-
– Incident response plan tion in their own development as well as in reaching a high
– Material control and accountability compliance level in many normative aspects.
– Supply and transport security But, having the new set of guidelines for BRCs demands
a new approach for third-party assessment. Especially in
Over 20 WDCM registered collections have some form of consideration that the transition of the guidelines into one’s
certification or accreditation to demonstrate the provision of own organization, the handling of increasing requirements
quality services and material. The OECD best practice guidelines coeval to diminishing financial support and the exposure to
extend such certification criteria to address BRC operations regulatory compliance is an unsolved problem left to the indi-
more specifically setting a benchmark for culture collections vidual BRC and their quality managers. Thus, the internation-
worldwide. Mechanisms to ensure collections adopt these ally organized, German Federal Ministry of Education and
standards to deliver high quality should be put in place Research funded project, ‘‘GBRCN – Global Biological Resource
(OECD 2001). Although the OECD BRC Task Force wished to Center Network’’ – is working on an assessment model based on
see high quality proven through an independent third-party the excellence principle. Implementing the OECD guidelines
auditing process, for example, certification, it wishes to base it brings a multidimensional capability into an organization; the
on existing systems and internationally accepted scientifically principle of the GBRCN of sharing and continuously improving
based quality criteria. these capabilities among all BRCs and culture collections is
The actual system of international standards is focusing the opening the way toward an excellence model in performance
complementary aspects of management, technical skills, prod- and in the delivery of the OECD requirements. The excellence
uct conformity, and process stability. The OECD best practice approach would resolve the restrictive regime of a standard and
guidelines added the aspect of regulatory affairs to this comple- its full compliance assessment by offering a stepwise develop-
mentary system. Each standard is specialized to enhance the ment in accordance to the available resources and demands.
compliance of an organization to a single aspect (e.g., ISO Self-evaluation and third-party audits will still be the important
9001 –> management). Are the OECD guidelines really a new instruments to gain confidence in the system and recognition
approach going further than the already existing standards, of the delivered results. However, the new assessment model
establishing a complementary system with an integrative char- is not propagating the golden way; it shows that many
acter covering all aspects? Can the OECD guidelines in addition approaches will lead to excellent quality in services, material,
answer both the Task Force and the OECD demands? These and science.
main questions can be answered by identifying the issues, Currently, the discussion is ongoing, whether the OECD best
which impact significantly on the implementation and mainte- practice guidelines will remain a part of the so-called GxP world,
nance of quality management in culture collections and BRCs. for example, Good Laboratory Practice or if they should become
These issues reflect a variety of managerial operations and per- an ISO standard, thus broadening the existing set of standards
spectives including continuous improvement, organizational with the special requirements for culture collections and their
behavior, human resources management, customer relations, living biological material.
Quality and Certification: Costs and Benefits ● Confidence that the materials are fit for purpose
● Assurance that national law, policies, and procedures have
Whatever system is selected, there will be costs associated with been followed
the achievement and implementation of the standards, and it is
therefore important that the benefits to users and the BRCs
themselves are clear. Some of these needs and benefits are BRC Benefits
outlined in the OECD report (OECD 2001). If the user benefits
from the certification or accreditation of BRCs through better ● Recognition that they operate to international scientifically
access to authentic and reproducible materials in a transparent based quality criteria
and traceable way, how does the BRC benefit? There is an ever- ● An international mark of quality
increasing demand for authentic materials as more and more ● Raised profile
industries are adopting certification or accreditation as a means ● Sharing of tasks
to demonstrate quality and competence. This may be the driving ● Common policies and procedures
force for the business elements of a BRCs strategy for long-term ● Competitive edge
sustainability, but it is also an increasing requirement to satisfy ● Level playing field
the funders of research who seek high-quality science and ● Common access to data enabling links to be made to other
solutions. international initiatives without duplication of effort
It is imperative that organisms utilized in biotechnology are ● Common approach to data access, sharing, and
maintained in a way that will ensure that they retain their full interoperability
capacity. BRCs must ensure a high-quality product that will give ● Improved data usage
reproducible results. To achieve this, BRCs must apply quality ● Collaborative research and development
control and assurance measures to maintain these standards,
taking into account the needs of users and of the facilities and Inevitably, introducing the requirements of the standard and
resources available. The need for common standards is evident accreditation or certification procedures to the collections to
as the task of maintaining representative samples of microbial achieve the status of a BRC will be costly. However, used cor-
diversity cannot be achieved by one collection alone. Therefore, rectly, it can attract investment in the development of BRCs, and
it is essential that a worldwide network of collections interacts to the outcome will be beneficial to all concerned.
provide the coverage required by the user. In order that
a customer of such a network would get a consistent level of
service and quality, it is necessary to set standards for all collec- Financial Sustainability of Public Service
tions to attain. Collections and BRCs and Networks in Both
Standards also provide a useful target for new collections to Developed and Developing Nations
achieve, but it must be remembered that standards should
become part of the operations and not be a set of rules Implementing common standards and improving operations
implemented separately. Their aim is to ensure good quality have additional costs. Despite there not being one model for the
and traceability and encourage improvement and further devel- operational and financial sustainability of a collection, we can
opment. Standards must fit the operation and not add excessive learn from the experience of existing culture collections. Studies
unnecessary burden. Implementing standards for operation by the OECD BRC Task Force and the EMbaRC consortium
allows collections to convince investors to establish the facilities, provide working models for BRCs. Although culture collections
skills, and mechanisms needed to participate in international or BRCs have similar activities and objectives, they can be quite
activities. However, it is not sufficient to set the standard and different in size, scope, and function. They may be described as
then forget about it. A process for review and update must be put either specialist or generalist collections, be small, based around
in place to ensure that new technologies can be brought in to an individual researcher or research team, or be large public
improve the standard. service collections and a multitude of structures in between with
The advantages of certified or accredited BRCs forming differing financial models supporting each. Culture collection
a network can be split into two groups, those that give benefits revenues traditionally come from supply, preservation, and vari-
to the users and those that benefit the BRC itself although several ous services associated with these, such as identification, charac-
could fall in both categories. terization, or specialist consultancies. Culture collections also
participate in research or service contracts, but most rely on
some form of governmental or host institutional funding.
User Benefits A variety of activities relate directly to quality control, collection
development, and operation that may include opportunities for
● A one-stop shop where both high-quality biological mate- some additional cost-recovery activities. Among several potential
rials and the information associated with them can be found new sources of revenue is the generation of genomics and prote-
● Conformity of both quality and authenticity of biological omics data that complement and add value to the biological
materials but also of processes and procedures to access them materials themselves. The degree to which such activities may
actually provide support, sufficient to ensure financial sustain- being planned and at a standard that users would expect. If secure
ability of a BRC, is unproven. There are a few centers only that resources are limited, in general, it is preferable to restrict the
purport to declare themselves self financially sustainable. primary objectives of the collection to those which it has a strong
The OECD model of BRCs includes considerable diversity of probability of maintaining in the long-term. The financial models
funding mechanisms for individual centers. However, it is to be provided by existing culture collections of various types are well
expected that most BRCs will require some degree of commit- recognized and include
ment to core funding by their respective national governments.
● The ‘‘General Collection’’ – often a national/regional facility.
Other kinds of funding sources include support from industry,
– ‘‘Popular’’ items for distribution can guarantee income.
grants from agencies that support research, cost recovery
– Archive function requires subsidy.
through fees-for-service, development of databases, and other
● The ‘‘Specialist Collection’’ – usually more localized.
tools that complement the core role of BRCs, for example, even
● The ‘‘Institutional Collection’’ can provide internal institutional
funding from charitable sources, especially those associated with
service or wider external community/network service.
public health or sustainable development. Furthermore, BRCs
● The ‘‘Research Collection’’ provides a service relevant to one
should be encouraged to coordinate their pricing policies and
or more research interest.
other activities to best serve their essential functions in response
to the needs of sectors that depend on their biological resources. These models vary considerably in the proportion of income
The different approaches do not only rely on the expertise derived from the various sources defined below. It must be
and function of the different collection hosts; but additionally, emphasized, however, that the larger the archiving function
the needs and capacities of individual countries vary. Specifi- carried out for strategic reasons rather than supply, the greater
cally, the needs of developing countries must be understood and is the need for public and private subsidy.
accommodated. The OECD BRC Task Force advocated that
national governments should identify collections and centers
already capable of being designated as BRCs or forming Financial Models for Biological Resource Centers
a network and build upon and improve these rather than
starting up new BRCs, especially in developing countries where The diversification of activities in the transition from the ‘‘Cul-
resources are limited. Similarly, partnerships must be developed ture Collection’’ to a BRC anticipates additional sources of
among BRCs and appropriate existing agencies, identifying their revenue, both from existing activities and projects related to
capacities and interests in terms of support for BRCs. A survey new technology-based partnerships. Two types of income
carried out by Stromberg et al. (2012) of 103 WFCC affiliated streams are recognized. ‘‘Existing’’ income streams are those
member collections, comprising mainly public service collec- that support existing models of culture collections. ‘‘Antici-
tions, demonstrated quite different balances between revenues pated’’ income streams which represent activities in which
and public support. Fifteen percent receive no public funding; BRCs will or may participate in and that may generate recover-
11 % received 1–40 % funding; 8 % receive 41–60 %; 13 % are able income from stakeholders.
61–80 % funded; 54 % receive more than 80 % public funding. Existing Income Streams for BRCs
These differ widely from the total statistics for the WDCM
● Government support
which shows that the majority of the remaining 450 plus collec-
● Private industrial support for participation in the function-
tions are not publicly funded and overall only about half the
ing of BRCs
registered collections receive governmental support.
● Private industrial support for internal restricted BRC
It is evident that at the outset when a collection is being
activities
considered, and before it is established, the financial plan and its
● Public and private foundation support
sustainability must be designed. The WFCC guidelines (Anon
● Public fundraising
2010) state that the long-term support needed to enable collections
● Fees for supply of biological resources and technical
to provide professional services must be considered, including
materials
appropriate operational facilities, the staffing levels to allow oper-
● Provision of specialist services and technical consulting
ation at a high standard and the training level of staff with research
expertise
expertise related to the aims of the collection. The WFCC guidance
● Research income (grants and contracts)
presents funding as a key consideration. Administration and
● Fees for repository service (safe deposits and patent strain
funding arrangements for collections require a long-term commit-
maintenance)
ment from the parent organization. Support solely in the form of
● Provision of technical courses
short-term contracts or without any allocation of core funding is
● Exploitation of and adding value to genetic resources
inappropriate for service collections, aiming to provide long-term
storage and supply services. Even the establishment of small in- The understanding of current income lines for culture collec-
house collections requires an ongoing source of direct, or indirect, tions was assisted by the survey carried out by Stromberg et al.
financial support from a parent body. It is important to consider (2012) where they report that 67 % of affiliated member collec-
the level of funding, both now, and what it is likely to be in the tions of the WFCC charge supply fees for strains provided. They go
future. This must be adequate to provide the range of services on to report that on average, 23 % of the recipients of cultures are
External sources of funding of

EMbaRC BRCs in 2009
100%
other
75%
research / private companies
research / public contracts
50% consulting
technical training
patent / exploitation
25%
services
resource deposit
resource sale
0%
M
M
IP
BC T
BS
MT
AB
SM
EC
IR
U
C
C
M
C
C
C
C
D
A-
R
IN
. Fig. 11.3
Funding sources of EMbaRC BRCs compared for 2009 (see Abbreviations)
in industry, whereas 60 % are in academia or hospitals. They also Culture Collection Funding
report that there is a relatively small overlap in strains held, with
between 1 % and 17 % of their holdings sourced from other There should be a balance between governmental support,
collections. They report that 38 % are received with some sort of commercial, and other income lines to provide support for
MTA, providing terms and conditions of supply. There is, how- collections. There are several collections that are supported
ever, an imbalance on where these collections are, with 82 % of by governments but rarely are they fully supported. The
them being in OECD countries and 10 % in the USA. Individually, government supports 235 of the 592 culture collections
although having overlap in the most popular reference strains, registered with the WDCM, a further 56 are semi-govern-
they have limited coverage of biodiversity, but together in mental, 218 are supported by university, 17 are supported by
a network, they could offer good coverage. Inevitably, there is an industry, and 25 are private. It can be argued that govern-
element of competition for the market, hence the determination of mental funding is essential and appropriate but even long-
collections to store the ‘‘best sellers.’’ term stability of such funds may eventually be under threat.
Potential or ‘‘Anticipated’’ Income Streams for BRCs from, for Culture collections perform many functions for governments
example, the supply of services for not least helping them meet their obligations to the Con-
vention on Biological Diversity and making available biolog-
● cDNA libraries, genomic libraries, filter sets, clones, plates, ical resources to underpin science, education, and the
and PCR products economy. Such government funding is usually balanced
● Microarrays and reagents against the income received for the various services and
● RNAi resources products offered by the collection. This leaves very little for
● Accreditation/standardization-added value products and investment and to enable the collections to improve their
services coverage and incorporate new and advancing technologies.
● Data storage and retrieval Collections need sound and innovative business plans to
● Software development/collaborations – data mining tools allow them to keep pace with the ever-increasing demands
● Technology development/collaborations – LIMS/robotics of their users.
● Sequence database annotation/phenotypic analysis The EMbaRC project is examining sustainability of BRCs
● Linking genomics databases to proteomics and has compared the revenue lines of the partner collec-
● MLST (multilocus sequence typing) and population studies tions. These collections are in the main well-established
● Product discovery, manufacture, and supply (potentially public service collections with a long history of providing
spun out to independent companies) products and services. They show that they have common
products and services, but the balance on how important
However, it is debatable whether all these activities have each individual line is to each collection is significantly
a market and offer worthwhile returns. Often such services are different.
offered by specialist organizations, and the competition can be The funding sources of some major West European collec-
quite tough. tions (EMbaRC) for the year 2009 are shown in > Fig. 11.3.
The level of public funding was mainly in the range of 65–92 % Government Department Support
but with the exception of CABI which has no specific national
funding but which has overall member country contribution of ● Provision of services to governments to help them achieve their
around 3 % with more invested in the collection maintenance conservation and utilization of biodiversity commitments,
activity. Other funding sources for the EMbaRC collections in their environmental policies, and their commitment to poverty
2009 were alleviation
● Maintenance and bench fees – 0–16 %
● Resource supply – 2–75 %
Sponsorship
● Resource deposit – 0–10 %
● Services – 0–39 %
● Attracting donations to cover costs of biological resource
● Technical training – 0–1 %
provision, establishing a consortium of research program
● Consulting – 0–7 %
funders and sponsors
● Research/public contracts – 0–94 %
● Research/private companies – 0–14 %
Other Financial Aspects of Operating a BRC
Network (as Identified by the OECD Workshop
Developing Income Lines on Funding Models)
Not only do collections need to find novel ways of funding While a uniform structure of funding is not necessarily critical,
but also need to keep abreast and harness new technologies many BRCs will require a significant component of government
to produce information on the strains they hold, adding funding. Some guarantee of ongoing funding is necessary to
value with the aim to provide today’s users with the infor- ensure that their essential functions remain reliable for R&D
mation they need. It is not always possible to establish these and support of biotechnology. Collections will be put at risk if
technologies in-house, but it is possible to establish partner- a BRC network operates at the expense of individual BRC
ships with manufacturers, other collections, or institutions funding resulting in the individual BRC’s folding for lack of
with the expertise and facilities. Bioinformatics is of increas- support. The following points were derived from the SWOT
ing importance to the operation of collections, and new (Strengths, Weaknesses, Opportunities, and Threats) analysis
ways of collecting, storing, analyzing, presenting, and inter- of financial models of BRCs and a proposed model for BRC
rogating information are required to make best use of development.
biodiversity information. Molecular techniques are increas-
ing in use to differentiate between strains and identification.
Collections should be adopting such techniques to offer as Strategic Implications
services to users to counter the costs of utilizing these
techniques for checking stability and authenticity of the ● A prerequisite for any network is that it is built upon stan-
strains they supply. dards and accreditation. This defines the network. This may
There are a number of ways BRCs can develop their individ- require further investment.
ual business plans. However, it is crucial that BRCs do not ● A national strategy that provides core financial support for
become commercial entities; they must not compromise their a national BRC (or BRCs) should be viewed as a prerequisite
public service role. Having said that they must do all they can do for participation in the international BRC network, to
to reduce their public cost, a delicate balance is required. Some ensure that the network is sustainable.
avenues that can be explored are outlined below. ● The international BRC network will be built upon national
initiatives that in turn will evolve from existing activities
(including culture collections). These activities are already
Commercial based upon a range of income streams with varying levels of
government support.
● Development and ownership of spin-off biotechnology ● Governments will be fundamental partners in the creation of
companies, generally through partnerships, sale of products, national BRCs contributing to the international network,
and services as well as consultancy regardless of the level of financial support.
● Many existing culture collections will not wish to participate
in the BRC network if this is inappropriate to their aims or
Research Program Funding goals, or if this is not justifiable given the level of investment
required to raise/alter standards. Links to enable BRCs to
● A series of projects to meet donor requirements, engaging draw resources from such centers will need to be created.
research program funders to protect their investments by ● Governments need to recognize that BRCs will take
paying for deposits in collections a regulated role in the supply and maintenance of
dangerous/pathogenic organisms. This important core Collections and Their Users: The Need to
aspect of BRCs provides a controlled framework for Know Each Other Better
the availability of these sensitive resources. In turn,
fulfilling this role requires a level of financial Though not commonly encountered, collections are encouraged
commitment. to conduct market studies and carry out regular surveys on
● BRCs must use the opportunity of establishing an interna- customer satisfaction and buying behavior. Collections requir-
tional network to seek sponsorship from a variety of new ing a substantial proportion of their budget by generating rev-
sources of support (national, international, public, private, enues usually have in place a dedicated user-oriented
and industry). management plan which, according to a functioning quality
management, must be improved constantly. Knowledge about
user demands and requests is indispensible for strengthening
Operational Implications their market position. Therefore, collections should have access
to certain basic information, such as
● BRCs have to take a prominent role in capacity building and
● Who is the user?
ensure a link between research-based collections and the
For example, where are they working: in academia, bio-
BRC and the ultimate user.
industry, food, or clinical sector or in public health or
● BRCs need to function as a strategic, national repository for
schools?
key academic and industrial research resources, which will
● What are the needs of the user? Are curators aware of them
in turn provide an income stream. This is unlikely to
and is the management in a position to react quickly to
operate on the basis of full cost recovery from supply
satisfy user demands? Examples are as follows:
income.
Post-order communication
● Governments and their funding agencies must ensure
Quality and type of packaging
that products derived from publicly funded research pro-
Modalities of shipment
grams are deposited in BRCs as part of the conditions
Correctness of delivered goods, such as authenticity and
attached to any award. (This could result in a small
product information on safety aspects and handling
element of the grant allocated to this task as appropriate –
Option to establish contacts
see below.)
Amicability and qualification of collections’ contact person
● BRCs need to provide greater support to research-based
Goodwill policy in case of replacement shipment
collections in terms of training and advice on standards,
Handling of complaints
quality control and integrate more with the national activi-
Lead and delivery times
ties in key-related priority research areas (e.g., model organ-
Internet accessibility on information to
ism research consortia).
Cultivation conditions
● Governments must ensure that infrastructure aspects of the
Spectrum of services
support for research are funded through relevant research
Databases
programs.
New resources and products and new developments
● BRCs must create partnerships with centers of excellence and
developing new technologies and databases to ensure that The necessity to develop a more intense communication
linkage is possible between these leading edge aspects of between collections and users is driven by the need to establish
research and the physical resources held in BRCs. long-term and repeated use of the collection and its services.
Once satisfied with some basic principles, such as high quality,
It is anticipated that all of these strategic and operational short delivery time, and correct and timely information, the user
changes relevant to the national role of BRCs will enhance their will more likely as not become a loyal customer, independent of
position in providing services of benefit to the scientific com- the fees for resources and services. The collection should develop
munity and thus in turn benefit them by maximizing the poten- a specific affiliation with the user (worldwide highly recognized
tial for financial support. brands have achieved this goal), and the collection should facil-
A key element for discussion, however, remains the degree itate this relationship by documenting the advantages to be
to which BRCs may benefit from the direct commercial exploi- linked to justify this very resource center. Some are as follows:
tation of the resources that they hold. ‘‘Ownership’’ as contact to a nationally/internationally leading collection; pro-
a concept has, to a large degree, been avoided in the past with cess reliability, such as provision of non-contaminated resources
the BRC acting as a ‘‘custodian’’ of the resource. Widespread allowing reproducible results; range of a defined (either broad or
introduction of Material Transfer Agreements and implications specific) selection of products and services; close scientific sup-
that IPR and reach-through are requirements for access to port and consultancy, guaranteeing quality in products and
resources would fundamentally alter the relationship between processes; and long experience in taxonomy and identification,
depositor, user, and the BRC. National mechanisms for handling of recalcitrant, pathogenic and other delicate material,
implementing the Nagoya ABS protocol (CBD 2011) could as well as expert knowledge in shipping packaging and import
impact heavily here. and export rules and regulations.
As compared to the collection-user relationship of 10 years collections and have coordinated some efforts. They have
ago, significant changes have already been introduced, for exam- brought together metadata on their members to central
ple, by establishment of a quality management (QM) system; points and have helped keep members up to date with the
one of the central features of a functioning QM system is the progress of science, changing legislation, and collaborative
continuous process of improvement. This not only includes the opportunities through newsletters, conferences, and work-
strengthening of the collection-user relationship as mentioned shops. However, coordinated strategies for ensuring compre-
above but also offers some of the advantages the customer is hensive coverage of species and the diversity within them are
used to receive from well-managed online shops in other market yet to be put in place. Projects and individual initiatives have
segments. Top priority of online shopping is the option to pay by made some progress, but consolidating the many initiatives
credit card. The latter issue is still a moot point in Europe as, in that are working toward this goal is crucial to establish
contrast to the situation in North America and a few other a systematic and networked approach. This would bring
countries, most organizations do not allow payment by this advantages to both the users and the collections themselves
means or do not provide credit cards to their employers. but importantly provide an infrastructure to underpin
Another sector with room for improvement is the need for the research and development, enabling the harnessing of micro-
collection to accompany the customer through the ordering and bial and cell diversity to contribute toward providing solu-
delivery process. Once an order is launched, basic information tions to the world’s big challenges.
on entry date, confirmation of an order, order status, and date of The WFCC has been promoting the activities of culture
dispatch should be provided. It must be the goal to have the collections for over four decades and has done a tremendous
material shipped within a few days – if not, the user should at job to help establish a sound operational basis (> Box 11.2). It
least be informed about possible delays. Such requirements are was first to try and establish minimum standards through their
clearly described in the OECD best practice guidelines (OECD guidelines (Anon 2010), common standards form the platform
2007). on which networking is based. The WFCC, as are most culture
We are aware that online shopping for consumer goods collection organizations, is a community that exchanges views
cannot be fully compared with the provision of living organ- and ideas. Often, this results in the uptake of common
isms, which require a lengthy process of customer authoriza- approaches, but the organization has no mandate to affect
tion, and administrative effort on export, import, and shipping institutional changes in policies and practices. This impedes
regulations, not to mention delays of shipment of active cul- the introduction of coordinated approaches. At the regional
tures. It is, however, not overstated to indicate that most public level organizations such as the European (ECCO, > Box 11.4)
service collections have not attempted to assess the satisfaction and Asian (ACM, > Box 11.5) networks work on behalf of
level of their users. Here, one can learn from commercial collections. They have been very successful in bringing project
resource centers with which noncommercial collections com- consortia together to seek project funding to solve common
pete in the same market. Collections should not hesitate to operational problems or address common research issues.
learn from the best cases of other organizations, and they There are over 20 national federations that do similar things at
should learn to react quickly to customer needs and demands. the country level. However, a lot of work still needs to be done
This requires the establishment of a client-led marketing pol- both by collections and governments if the goal to harness the
icy, most efficiently executed by a professional marketing unit. power of microbial diversity is to be realized. We need to harness
Public service collections must recognize the need to make the properties and products of microorganisms more efficiently
themselves more attractive through regular press releases of if we are to tackle the big global challenges of today in poverty
collection-related scientific headlines, by increased publica- alleviation, food security, healthcare, climate change, and the
tions in international peer-reviewed journals, attractive train- environment.
ing courses, and involvement in teaching and public lectures. The OECD emphasizes that biological resources, such as
These activities are already followed by larger public collections microorganisms and their derivatives, are the essential raw
but continuously necessary to accompany measures of the core material for the advancement of biotechnology (OECD 2001).
mandate and motive, that is, the provision of high-quality However, they go on; scientific progress and the resulting growth
and non-contaminated biological references to support scien- of the knowledge-based bio-economy will depend on the facil-
tists in their goal to obtain reproducible data at the highest itated and safe access to ex-situ held living biological material
scientific level. and its availability in an adequate and comparable quality
worldwide. It is understood that this, in turn, requires putting
in place coordinated policy actions by all stakeholders involved.
Scientific-Technical Cooperation Among To meet the increasing demands of the scientific community for
Microbial Culture Collections comprehensive, up-to-date, and easy to access living biological
material available from microbiological culture collections and
Although culture collection organizations have existed for related information, a series of coordinated activities were initi-
many decades, they or their modern-day versions, the ated in different regions worldwide, leading to network activities
BRCs, have never been fully networked. National, regional to foster communication and research among collections for the
and global organizations have endeavored to help promote benefits of users and science.
Box 11.2 World Federation for Culture Collections Box 11.3 The World Data Center for Microorganisms
The World Federation for Culture Collections (WFCC) is a key global In 1982, the World Data Center on Microorganisms hosted by the
organization originated from an IAMS ‘‘section on Culture Collections’’ University of Queensland in Australia issued the World Directory
formed in 1963, which was reorganized as the World Federation for of Collections of Cultures of Microorganisms (> Fig. 11.1). The
Culture Collections in 1970. From 1973, it was recognized as editors of the directory, Vicki F. McGowan and V. B. D. Skerman,
a multidisciplinary commission of the International Union of Biological articulated the roles of culture collections and the data center,
Sciences (IUBS) and since the separation of the International Union of ‘‘Culture collections occupy a central position in microbiology
Microbiological Societies (IUMS) from IUBS in 1979, it has operated as because effective research demands adequate and reliable
an inter-union commission. It seeks to promote activities that support sources of properly preserve cultures. As a result of their function
the interests of culture collections and their users. Member collections as repositories of living organisms, culture collections promote
of the WFCC register with the World Data Center for Micro-organisms microbiological research. Increased demands for historical infor-
(WDCM, > Box 11.3). A congress is held every 3 years to discuss mation and strain data have created a need for easily accessible
advances in technology and common policies with regard to biodi- and up-to-date files of important information on the location and
versity and the role of culture collections. The WFCC keeps its characteristics of cultures. Such needs can be met by the develop-
members informed on matters relevant to collections in its Newslet- ment of an adaptable system for storing, retrieving and exchang-
ter and has working programmes addressing patent depositions, ing information which can be used by all microbiologists.’’
biosafety and biosecurity, safeguard of endangered collections, The WDCM relocated in 1986 to RIKEN, Saitama, Japan, and
capacity building, and quality standards. Since 1986, the WFCC has then again in 1999 to the National Institute of Genetics, Japan,
overseen the activities of the WDCM which is now the data center for and introduced an online database and website to capture and
the WFCC and Microbial Resource Centers (MIRCENs) Network. diffuse information on culture collections and their holding. In
The WFCC is the largest independent global organisation that the meantime, the number of culture collections registered in
represents professional individuals and culture collections, which WDCM has increased year by year. However, it is to be noted that
preserve biodiversity and enable their proper use. They target living WDCM has issued 983 IDs to culture collections, that is, the
microorganisms, cell lines, viruses and parts and derivatives of community has lost about 300 culture collections since the
them. Key values are authenticity and genetic integrity of the first was registered. The WDCM collections hold in excess of
material and validity of the information provided. The WFCC sup- 1.7 million strains: 44% are fungi, 43% bacteria, 2% viruses,
ports the professionals, organizations and individuals with interests 1% live cells, and 10% others (including plasmids, plant, animal
in culture collection activities through networking, providing infor- cells, and algae).
mation and expertise, and facilitating communication; facilitating In 1999, WDCM organized a symposium with the title of
access to the collection resources; providing training and promot- ‘‘Microbial Resources Centers in 21st Century – New Para-
ing partnerships; encouraging the development and implementa- digm’’ back to back with the 1st OECD Meeting on Culture
tion of quality and security procedures and the use of common Collections. This was the moment when the concept of Bio-
standards and regulations; representing member interests in inter- logical Resource Centers was born. The participants of the
national organizations and fora; and promoting the establishment two meetings recognized the impacts of biodiversity, geno-
of culture collections and their perpetuation. mics, and informatics on culture collections and agreed that
There are over 120 culture collections affiliated to the WFCC culture collections had to evolve to become BRCs to meet
who have agreed to implement the WFCC guidelines (Guidelines their needs and those of users.
for the Establishment and Operation of Culture Collections- Anon The online database of the world directory named CCINFO
2010) and who contribute to the delivery of its objectives. In the includes information on 592 culture collections in 68 countries as
growing bio-economy, WFCC’s members face increasing global of March 2011; 235 of them are supported by the government, 56
demands for worldwide and controlled access to biological of them are semi-governmental, 218 of them are supported by
resources, public security, industrial quality of their holdings university, 17 of them are supported by industry, and 25 of them
and associated data and long-term genetic stability of the mate- are private; 226 collections produce catalogues of holdings and
rial. Key to the use of microorganisms from culture collections is there are 3,051 people working in them. These culture collections
the retention of their properties as research and development preserve 1,751,439 microbes. WDCM functions as an information
must be based on authentic and well-preserved biological mate- hub of culture collections and their customers. The WDCM is now
rial. The WFCC have been helping collections in this respect for hosted by the Chinese Academy of Science Institute of Microbi-
over 4 decades. It is a goal that strains of organisms be supplied ology since April 2011 and it is expected that the functions of
from member collections with traceability, conforming to WDCM will be expanded to cover aspects of biodiversity, geno-
national and international regulatory requirements, and that are mics, and advanced information and communication technolo-
preserved in such a way as to retain their full potential. gies (ICT). The URL addresses of the websites of WFCC and WDCM
stay as they are, namely, http://www.wfcc.info/ and http://www.
David Smith wdcm.org/, after the relocation of WDCM.
CABI, Egham, UK
Hideaki Sugawara Congress of Culture Collection (ICCC-10) held at Tsukuba in 2004.

DNA Data Bank of Japan, National Institute of Genetics, Mishima, Heads of culture collections and government officers were
Japan involved in the meeting as well as research microbiologists. The
objective of the consortium is to promote collaboration among
Government or public organizations in Asian countries for the
Such activities are not new for regional networks (> Boxes 11.4 purposes of enhancing conservation and suitable use of micro-
and > 11.5) or national networks, for example, the Brazilian bial resources in Asia. Currently the members are from Cambodia,
initiative (> Box 11.6). China, Indonesia, Japan, Korea, Laos, Malaysia, Mongolia,
Myanmar, Philippines, Thailand, and Vietnam.
The ongoing activities of the consortium are:
Box 11.4 European Culture Collections’ Organization and its
1. Exchange of views and information of the current policy of
Networking Activities
the Asian countries for science, technology, and the related
The European culture collections have collaborated since 1982 matters
when the European Culture Collection Curators’ Organisation was 2. Establishment and management of the network of culture
established to bring together the managers of the major public collections including a common database
service collections in Europe to discuss common policy, exchange 3. Enhancement of public awareness on the consortium’s activ-
technologies, and seek collaborative projects. The organization ities for the conservation and sustainable use of microbial
opened itself to staff and users of microorganisms and is now resources in consideration of the Convention of Biological
named the European Culture Collections’ Organisation (ECCO). Diversity
There are currently >65 members, including 57 collections hold- 4. Development of human resources for handling of microbial
ing approximately 350,000 strains. The members have been resources technically and legally
involved in producing practical approaches to international 5. Promotion of research and development on microbial
rules and regulations. An initiative led by the Belgian Co- resources and their application in industrial and other uses
ordinated Collections of Microorganisms (BCCMTM) produced 6. Establishment of a common scheme for international transfer
a code of practice for collections to operate within the Budapest of microbial resources
Treaty and the EU project Microorganisms, sustainable access and 7. Scientific meetings (seminars, workshops, training courses,
use, International Code of Conduct (MOSAICC) provided model etc.) and other related activities
guidelines for the operation within the spirit of the Convention
The General Assembly Meeting of the ACM has been held
on Biological Diversity. Several collaborative projects originated
annually since 2004 in different countries. Task Forces for
through discussions between ECCO members that have placed
Bioresource Information Management, for Human Resource
the European Collections at the cutting edge of culture collection
Development and for Management of Material Transfer are
activities and research. The most recent initiative is the EMbaRC
set up. These activities will be of value for the standardization
project. They have resulted in technical guidelines and focused
and authorization of international transfer of microbial
information documents covering requirements with which mod-
resources. Many Asian microbiologists are eager to study the
ern-day microbial collections are challenged. Substantial input
microbiological diversity in the nature of various natural envi-
was given by ECCO to the BRC initiative and the recent demon-
ronments. ACM is also expected to achieve the rule of Interna-
stration project for a Global Biological Resource Center Network
tional Code of Prokaryote Nomenclature in compliance
(GBRCN). On a global level, the latter project aims to build
with the laws and regulations relevant to the Convention on
a structured long-lasting network which will pave the way for
Biological Diversity.
collections to meet user needs. It addresses technical, legal, and
The 7th ACM meeting was held in Japan again in 2010,
administrative challenges presented in this globalized, fast-
the International Year of Biodiversity, and adopted the Kazusa
developing world.
Statement on 15th October as follows:
Dagmar Fritze ● Kazusa Statement

DSMZ- Deutsche Sammlung von Mikroorganismen und In the 7th ACM Meeting at Department of Biotechnology,
Zellkulturen GmbH, Braunschweig, Germany National Institute of Technology and Evaluation (NITE) in
Kazusa, Chiba Prefecture in Japan, members of the Asian
Consortium for the Conservation and Sustainable Use of
Microbial Resources (ACM) recognize that:
Microorganisms such as filamentous fungi, yeasts, mush-
Box 11.5 The Asian Consortium for Conservation and Sus-
rooms, bacteria, archaea, and microalgae play important
tainable Utilization of Microbial Resources (ACM)
roles in the global ecosystem either directly or indirectly
The Asian Consortium for the Conservation and Sustainable Use The diversity of isolated microbes only account for less than
of Microbial Resources was established by the consensus of par- 10% of the total species, which means that many novel
ticipants of 12 Asian countries during the 10th International and yet-to be-discovered microbes inhabit the earth
The diversity of microbes is endangered by global climate integrated infrastructure of distributed biological resource cen-
changes, habitat changes, over exploitation, and ecosys- ters. The goal is not only to underpin the actual needs for biolog-
tem destruction ical material from industry and academia, but to foster innovation
Long-term laboratory preservation of microbes is technically in biotechnology, addressing issues related to emerging legal,
well attainable technical, and sanitary barriers associated with the access of
Microorganisms are crucial biological resources to academia, biological material and genetic resources to the global market.
biotechnology, and bio-industries contributing to tech-
nology, economy and social developments
History of the Brazilian Resource Centers
They have also reached the following agreements:
Network
1. Prompt action of each country toward ex situ conserva-
tion of microbes is imperative.
The need for consolidating a network of microbial collections in
2. For effective ex situ conservation of microbes, interna-
Brazil was discussed for the first time at the Second International
tional research cooperation is essential.
Congress of Culture Collections held in São Paulo in 1973. The
3. Active international research cooperation needs to be
recommendations from this congress organized by the Brazilian
promoted by establishing a scheme to facilitate interna-
Society of Microbiology (SBM) in collaboration with the WFCC
tional transfer of microbial resources, further provision of
influenced the 1980s and 1990s strategies for development of
technical cooperation, and capacity building in full com-
biotechnology in Brazil. A key enabling instrument for improving
pliance with the principles of the Convention on Biolog-
microbial resource centers in Brazil was the implementation of
ical Diversity.
the Program for Human Resources in Strategic Areas (Programa
4. For clarification of endangered microbes and conserva-
de Recursos Humanos em Áreas Estratégicas -RHAE) funded by the
tion areas, a list of domestic microbes should be created.
Ministry of Science and Technology (MCT). The RHAE program
5. Microbial taxonomists should take an initiative on the
developed in collaboration with WFCC promoted the training of
creation of such a list with the support of international
a number of experts in international collections associated with
research cooperation.
the organization of nearly 50 training courses and seminars
6. Demand for and importance of microbial taxonomists
focused on issues related to collections management, preserva-
should therefore be well recognized in each country, so
tion techniques, and microbial taxonomy.
that having training programs in place for microbial tax-
In the late 1990s the effort carried out by the Organization for
onomists who can keep inter-generational continuity
Economic Cooperation and Development (OECD) to discuss the
seems imperative.
critical role of microbial collections as infrastructure to underpin
To achieve the intention of this Kazusa Statement, the biotechnological innovation was crucial to renew the discussion
establishment of a Microbial Resource Center (MRC) in each on the need to improve the bio-collections infrastructure in Brazil.
country is necessary. By establishing the MRC, the training The OECD report on ‘‘Biological Resource Centers: underpinning
program for microbial taxonomists, legal management of the future of life sciences and biotechnology’’ (http://www.oecd.
microbial resources, and the creation of the list of domestic org/dataoecd/55/48/2487422.pdf) published in 2001 was the cat-
microbes through exploration, characterization, conservation, alyst factor for the establishment of a Task Force to discuss the
and sustainable utilization of these microbial resources can be conformity assessment of biological material. The result of the
carried out. Furthermore, the MRC can make a significant work carried out by the Task Force is summarized in the docu-
contribution to the development of the bio-industry by ment Sistema de Avaliação da Conformidade de Material Biológico
providing scientific and technical services to various users. The (MCT 2002, System for the Conformity Assessment of Biological
MRCs in countries must endeavor to maintain close coordination Material) (http://www.ctnbio.gov.br/upd_blob/0000/10.pdf). This
with each other and dedicate to exploration and promotion of timely report summarizes the state of the art of Brazilian collec-
utilization of microbes. tions within the framework of the international scenario and
discusses the challenges and opportunities associated with the
Ken-ichiro Suzuki implementation of a conformity assessment system for biological
Biological Resource Center, National Institute of Technology and material in Brazil. The assessment of the local legal framework
Evaluation, Chiba Pref, Japan compared to international norms and guidelines, associated with
a proposal for capacity building was key to guide the participa-
tion of Brazilian experts at the OECD Biological Resource Center
Task Force that resulted in the publication of the ‘‘OECD Best
Practice Guidelines for Biological Resource Centers’’ (http://www.
Box 11.6 The Brazilian Network
gbrcn.org/fileadmin/gbrcn/media/OECD_guidelines_for_brc.pdf)
The increasing demand for high-quality biological material and and the ‘‘OECD Best Practice Guidelines for on Biosecurity
information as a consequence of the growth of the Brazilian bio- for BRCs’’ (http://www.oecd.org/dataoecd/6/27/38778261.pdf)
based economy is requesting the implementation of strategies in 2007. The Brazilian document on conformity assessment and
and funding mechanisms to enhance and consolidate an the OECD guidelines were incorporated as appropriate in the
MCT capacity building strategy and discussed at biannual events storage, distribution, taxonomic characterization, and identifica-
organized by the Brazilian Society of Microbiology (SBM). The tion of Leishmania and associated information. CLIOC services
importance of SBM support to Brazilian microbial collections in meet the needs of public research and educational institutions,
this decade was reviewed by Canhos et al. (2007). industry in general, offering assistance and technical and
Based on the recommendations of the ‘‘System for the Con- scientific consultancy, training and development of specific
formity Assessment of Biological Material,’’ the MCT launched research projects.
a capacity building program to improve quality management in The MCT’s capacity building program focused on quality
selected service collections. The institutional arrangements and management in selected microbial collections as candidates to
the effort to reorganize the institutional systems of collections in acquire the status of Biological Resource Center (BRC). It allowed
Brazil were reviewed by Canhos et al. (2009). the participation of CBMAI and CLIOC in the Demonstration
The need to develop strategies focused on the reorganization project for a Global Biological Resource Center Network
and consolidation of the infrastructure to support biotechnolog- (GBRCN) (http://www.gbrcn.org/). This project is supported by
ical innovation in Brazil was addressed by the Presidential Decree the German Federal Ministry of Research and Education (BMBF)
604 (http://www.planalto.gov.br/ccivil_03/_ato2007-2010/2007/ following work in the OECD to improve access to high-quality
decreto/d6041.htm) signed on 8 February 2007. The decree biological resources and information to support research and
establishes the National Policy for the Development of Biotech- biotechnology as a platform for a knowledge-based bio-
nology and makes specific recommendations for the moderniza- economy.
tion of microbial collections as a key step in the implementation MCT’s program aiming at the establishment of the BR-BRCN is
of the Brazilian Network of Biological Resource Centers (Br-BRCN). being implemented in close coordination with the activities
sponsored by Brazilian Ministry of Development; Industry and
Foreign Trade (MDIC) focused on the establishment of
Networking Model and Institutional
a Depository Authority for patent purposes at the National Insti-
Arrangements
tute of Metrology, Standardization, and Industrial Quality
(INMETRO) in association with the National Institute for Industrial
As opposed to establishing a large national center to provide
Property (INPI); and the implementation of the INMETRO program
a wide range of biological materials and specialized services, the
for certification and/or accreditation of Biological Resource Cen-
Brazilian strategy is focused on the consolidation of a distributed
ters in Brazil.
network of specialized resource centers to meet the growing
demands of the user community. The reorganization and quality
management enhancement of networked collections in organi- Information System Architecture
zations such as Fiocruz (Fundação Oswaldo Cruz) (http://www.
fiocruz.br//cgi/cgilua.exe/sys/start.htm?infoid=5574&sid=17) To support the consolidation of the Brazilian network of resource
and Embrapa (Empresa Brasileira de Agropecuária) (http://www. centers the MCT is funding the development of the mSICol soft-
embrapa.br), and specialized collections like CBMAI, the Brazilian ware and implementation of SIColNet.
Collection of Environmental and Industrial Microorganisms The mSICol is a collection management software to support
(http://webdrm.cpqba.unicamp.br/cbmai/english/index.php) digital documentation and traceability of all processes associated
represent important starters in the strategy for the implementa- with day-to-day management of microbial collections, including
tion of the Br-BRCN. methods and procedures for strain authentication, preservation
Fiocruz coordinates one of the best-structured networks of techniques, stock control, quality management procedures, and
epidemiological control and public health in the world and hosts distribution of strains and biological reagents. The software is
several microbial collections with holdings ranging from archaea, a multiplatform system, designed to be compatible with different
bacteria, and fungi to protozoa. For the last 5 years, Fiocruz has data management systems. It has multi-user and multi-language
been working on the harmonization of procedures and protocols, capability and supports the installation of multiple collections
focusing on quality management based on ISO/IEC 17025/05 and and sub-collections. It is designed to document specific fields of
OECD Best Practice Guidelines for Biological Resource Centers. importance to microbial collections based on the WFCC Guide-
This program is supported by the installation of the information lines for Operation and Management of Collections of Cultures of
management software, which is at the moment integrating Microorganisms (Second Edition, 1999) (http://www.wfcc.nig.ac.
data from 11 collections at Fiocruz with the System for Collections jp/GuideFinal.html), the Common Access on Biological Resource
of Biotechnological Interest (SIColNet) (http://sicol.splink.org.br/). and Information (CABRI) Guidelines (http://www.cabri.org/guide-
The Fiocruz Leishmania Collection (CLIOC) (http://clioc.fiocruz.br/ lines.html), and the OECD Guidelines for Quality Management
index?) a Reference Collection of the World Health Organization and is fully compatible with the DarwinCore extension for micro-
(WHO) which is being prepared to be the core collection of bial strains (http://rs.tdwg.org/dwc/). The database provides
the Fiocruz BRC. Its experience will be replicated to the other a specific view that allows the exchange of data using TDWG
culture collections at the institution. CLIOC has a specialized Access Protocol for information Retrieval (Tapir) (http://www.tdwg.
holding with more than 1,000 Leishmania strains, mainly from org/dav/subgroups/tapir/1.0/docs/tdwg_tapir_specification_2010-05-
the New World. CLIOC’s mission is dedicated to preservation, 05.htm) in a simple and immediate way. The system is being
continuously developed to accommodate new features and Examples for Joint Activities and Global
requirements, including reports of daily activities and indicators Cooperation
of the collection holdings including taxonomic profile, geo-
graphic distribution of deposits, clients, and services provided. EU Projects Tackling Quality Issues of Data and
SIColNet allows the dynamic integration of strain data avail- the Biological Material
able in Brazilian collections with relevant information sources
ranging from molecular to ecosystems databases. Alignment The Microbial Information Europe (MINE) project was an early
with emerging technologies and adoption of internationally ambitious, taxonomy-based start with a clear user-driven data
agreed standards and protocols to secure systems interoperabil- selection. The main goals were the harmonization and digitiza-
ity are key features of SIColNet architecture. The Virtual Catalogue tion of data on over 150,000 catalogue strains, standardization of
of Strains based on the architecture developed for speciesLink formats, and contents of fields of databases, as well as common
(http://splink.cria.org.br) allows the dynamic integration of strain thesauri (Gams et al. 1988; Stalpers et al. 1990).
data with information on host organisms (botanical and zoological On this important base, the Common Access to Biological
information). Using a simple system of mapping and filtering of Material and Information (CABRI) project was built to offer
sensitive data, the system allows data providers to have full control online access to data across various collection databases in
over the data served to the network, with an appropriate crediting Europe with search options through individual and combined
system. Each collection determines what data is restricted and catalogues of collections. The great merit of this project was to
what is public. Through a web interface, users may search and develop guidelines for two of the service aspects of culture
retrieve nonsensitive data in different formats, may rapidly and collections: (1) for the handling of the biological material and
efficiently visualize species occurrence data on maps, and also (2) for the handling of the related data. The laboratory side of
have access to a number of indicators. The system also provides the guidelines covered aspects of accession, authentication,
reports on each collection’s profile, based on metadata and on the maintenance, storage, and supply for such kinds of biological
analysis of online data and reports on data quality. material as bacteria, archaea, fungi, yeasts, animal, human, and
plant cells, and genomic material. A compilation of model in-
house procedures was added. On the data side, minimal data sets
Future Developments (MDS) and recommended data sets (RDS) were agreed
outlining the minimum amount of data that should accompany
It is expected that, in 2012, the Depository Authority for patent a particular strain or culture when it is put into a publicly
purposes at the National Institute of Metrology, Standardization, accessible catalogue. Both types of guidelines aimed at raising
and Industrial Quality (INMETRO) will be in operation and that the scientific and technical quality of holdings and data to better
regulatory framework for the accreditation of resource centers support modern research and application. CABRI was later
candidates to acquire the BRC status will be in place in Brazil. incorporated as a core activity into the EBRCN project, and
Fiocruz is working on the implementation of a large-scale/long- CABRI guidelines are still available today.
term effort to establish a Biological Resource Center for Health A project with minor microbial participation was the ENBI
(BRC - Health) – unique in the world –focused on the study, a regional complementation of GBIF (see below). It formed an
preservation and distribution of microorganisms and biological intermediate level between national GBIF activities and the global
materials relevant to neglected diseases; innovation in epidemi- GBIF level. One interesting outcome of this work was a monograph
ology surveillance; as well as the development and production of on digital imaging of biological specimens from the zoological,
bio-compounds directed to diagnosis, vaccines, and drugs. botanical, and microbiological areas. This work aimed at setting
and publishing standards for improved quality photography. For
V. Canhos17 . E. Cupolillo18 . M. da Silva19 . C. Pirmez19 the microbial side, among other items, the microscopical mounting
17
Reference Center on Environmental Information (CRIA), method of ‘‘agar slides’’ was presented (Fritze 2005).
Brazil, UK Providers and users of biological material worked together in
18
Fiocruz Leishmania Collection (CLIOC), Rio de Janeiro, Brazil the MOSAICC project on the development of a system for
19
Fiocruz Vice-Presidency for Research and Reference appropriate management of access to and transfer of microbio-
Laboratories Rio de Janeiro-RJ, logical resources (see report on http://bccm.belspo.be/projects/).
Brazil The goal was in particular to help implement the provisions of
the CBD concerning Prior Informed Consent (PIC) and mutu-
ally agreed terms (MAT) in the context of monitoring transbor-
Similarly, individual collections cooperated in other global ini- der movements of biological material.
tiatives or in those of other communities. All of these initiatives Within the European Biological Resource Centers’ Network
have common goals: that the enormous amount and range of project (EBRCN, 2001–2004), emphasis was laid on the devel-
both the living biological material itself and the data pertaining opment of information documents concerning the various reg-
to this material are placed at the rapid disposal of researchers in ulatory issues around collection work, such as classification of
academia and industry under full legal compliance and curated microorganisms on the basis of risk, Convention on Biological
in a pure and authentic form. Diversity (CBD), intellectual property rights, regulations
governing packaging and shipping (2010 update available) and Toward a Global Network
control of distribution of possibly dangerous microorganisms
(http://www.ebrcn.eu/). To deliver their services, BRCs preserve their holdings using
long-term storage techniques such as cryopreservation and
lyophilization depending upon organism type. Quite often,
Recent EU Activities these techniques require optimization to enable not only
survival but also retention of properties. Seldom can a single
The ongoing European Consortium of Microbial Resources collection invest in preservation research, and often the
Centers (EMbaRC; 2009–2012) project includes work on improvement and testing of new techniques is done through
improving protocols for the authentication and preservation of projects. Networks can support each other to carry out research.
cultures and combines the additional aspects of training and Not every collection has the ability to handle every strain they
research activities. Within the area of compliance with regula- are offered, and networks can share the burden with organisms
tory requirements, emphasis is laid on the development of being deposited in BRCs which have the expertise and facilities
a Biosecurity Code of conduct. It aims to improve, coordinate, to handle them. There is extensive legislation that impacts upon
and validate microbial resource center delivery to European and access to, the safe handling, distribution, and use of biological
international researchers from both public and private sectors. resources (Fritze and Weihs 2000; Smith and Rohde 2007, 2008).
The EMbaRC project is a mixture of networking, access, train- A number of culture collection organizations exist to help
ing, and research (http://www.embarc.eu). collections keep up to date in a constantly changing legal
framework notably biosecurity, shipping regulations and
ethical access and use, common information resources can be
MIRRI: A New Initiative Within the EU Strategy established and common procedures implemented across the
Level of ESFRI network to ensure compliance. Therefore, networks can increase
single BRC capacity.
A new initiative to strengthen the European innovation capac- The work of the culture collection organizations has been
ities was developed by the European Strategy Forum for invaluable and has only been limited by their voluntary
Research Infrastructures. The microbiology collection commu- nature, relying on input of dedicated people as and when
nity led by the GBRCN secretariat (see below) and supported by they can contribute. Collections need to increase the avail-
ECCO and EMbaRC worked with ESFRI state representatives to ability of biological material for the verification of experi-
place the Microbial Resources Research Infrastructure (MIRRI) mental data and the authenticity of reference material used
on the ESFRI road map and as such is a priority for research in research. Deplorably, the scientific literature is full of data
funding. The resultant high-quality global platform will be which cannot be verified because the material is either no
designed to accommodate the future needs of biotechnology longer available and/or the material once used to generate
and biomedicine. the data has changed or deteriorated. This challenge needs to
MIRRI will bring together European microbial resource be met with a coordinated approach requiring an infrastruc-
collections with stakeholders (their users, policy makers, poten- ture to support it. Such strategies cannot be achieved
tial funders and the plethora of microbial research efforts) by projects with a defined lifespan. At a global level, the
aiming at improving access to enhanced quality microbial GBRCN aims at bringing together regional efforts such as
resources in an appropriate legal framework, thus underpinning those of ECCO help disseminate the outputs of projects such
and driving life sciences research. Emphasis will be laid on the as EMbaRC as well as the Asian initiatives (> Box 11.5) and
conservation of biodiversity, on services for research and appli- national activities such as those in Brazil (> Box 11.6) will play
cation, as well as CBD-ABS, biosafety, biosecurity, and bio-risk a key role.
matters. The overall aim is to support research, development, The GBRCN demonstration project emanates from
and bio-economy by improving access to and use of the micro- an OECD Working Party on biotechnology initiative. For
bial material. On this platform, strong interaction of all kinds of boosting the activities, a small central secretariat is presently
stakeholder working with microbiological material will be supported by the German Ministry of Research and Educa-
enabled, ranging from scientific and industrial providers and tion (BMBF) to coordinate activities to deliver improved
users to collections and policy makers, as well as to regulatory support to the life sciences. High-quality research in the
bodies and others. Non-European participation is strongly life sciences and innovative solutions to global problems
encouraged. MIRRI will integrate services and resources, bridg- requires access to high-quality biological materials and asso-
ing the gap between the organism and provision of innovative ciated information. No one single entity can provide the
solutions. MIRRI will as well provide coherence in the applica- necessary coverage of organisms and data; therefore, the
tion of quality standards, homogeneity in data storage and enormous task of maintaining biodiversity must be shared.
management, and workload sharing to help release the hidden Although the goal of the GBRCN would be to bring together
potential of microorganisms. All 57 microbial resource center BRCs from all four domains, animal, plants, human-derived
members of ECCO in the 26 European countries are invited to material and microorganisms, the project focus was on
join the initial consortium of collaborators in this initiative. microorganisms.
The GBRCN demonstration project secretariat coordinates To this end, the GBRCN and EMbaRC projects have designed
some activities of candidate microbial domain BRCs in 15 coun- a Biosecurity Code of conduct for BRCs which, when finalized,
tries in order to deliver: will be (morally) binding for GBRCN members. The Biosecurity
Code of conduct for BRCs sets out an undertaking by microbial
● The establishment of a network differentiated from existing
BRCs to tackle their responsibilities and provides a baseline for
organizations
their operation.
● The implementation of OECD best practice in BRCs (OECD
2007) assessed by independent third parties
● A strategy for the full GBRCN defining its infrastructure and
Box 11.7 Biosecurity
governance mechanisms, its secretariat’s structure, and func-
tion with a program of activities Research on biological material and the resulting knowledge
have benefitted mankind in many respects, ranging from basic
A GBRCN will also help the science community address the
science to applied agriculture to medicine and biotechnology.
maintenance of biodiversity range and magnitude, meeting
However, as so often, scientific results can also be used for
biosecurity requirements, bridging gaps in our knowledge and
malicious purposes – the dual use potential. This possibility
protecting investments in research. A GBRCN will support the
includes not only information, but also access to the biological
BRCs keeping abreast of modern scientific developments, meet-
material itself. At first the political accent was on biological
ing quality needs for research, supplying authentic cultures,
warfare. Bioweapons are attractive because they are relatively
supplying standardized biological material for testing and qual-
cheap, leave the infrastructure intact, are self-perpetuating but
ity control, developing comparable methodology and harmo-
may allow immunization, and have a delayed onset. Hence
nizing procedures and reconciling research and development
political activities concentrated on arms control, resulting in
demands with compliance with regulations.
the Biological and Toxic Weapons Convention (1972, BTWC)
Exploitation of biological materials must be in compliance
with the aim to prohibit the development, possession, and
with conventions, treaties, and law, for example, the CBD. The
use of biological weapons, and in the Australia group,
CBD requires that Prior Informed Consent (PIC) be obtained in
which intends to prevent the supply of harmful organisms to
the country where organisms are to be collected. Terms, on
malafide third parties.
which any benefits will be shared, must be agreed. The benefits
When Ivins in 2001 sent a series of letters with contents
may be monetary but could be nonmonetary such as informa-
contaminated with Bacillus anthrax spores, the risk of misuse of
tion, technology transfer or training. If the organism is passed
microorganisms suddenly became apparent; he changed the
on to a third party, it must be under terms agreed by the
microbial world. Although the number of victims was limited
country of origin. This will entail the use of material transfer
(22 infected, 7 deceased), the consequences were severe as the
agreements between supplier and recipient to ensure benefit
public was shocked. The horror scenario of a mad scientist threat-
sharing with, at least, the country of origin. Access and benefit
ening society had suddenly become reality. The trust in a world
sharing rules must be followed by those countries having
containing only scientific institutions with sufficient instruments
signed the Nagoya ABS Protocol (CBD 2011). The national
of self-control had been shattered. Biosecurity issues became
implementation of the protocol may well impede access and
a major concern for politicians, who immediately reacted by
exchange of materials and information. In this context, the
increasing the budget for biological warfare research and region-
collection community will have to work toward a mutually
ally by radiating imported parcels, thus jeopardizing the sharing
beneficial multilateral operational framework to facilitate sci-
of all biological material. The latter was remedied quickly, but the
ence and the discovery process.
need for well-executed and transparent biosecurity regulations
Biosecurity (> Box 11.7) impacts heavily on the operations
and the raising of public awareness remained. The task to restore
of public service microbial domain Biological Resource Centers,
the trust was taken up by both international organizations and
hence the activities of the WFCC and GBRCN. The GBRCN and
the scientific community, and two major contributions were
the European EMbaRC project promote the implementation of
made to provide clear and reliable guidance: the OECD Best
OECD BRC best practice which includes the biosecurity guid-
Practice guidelines for BRCs (2007), including the OECD Best
ance as well as aspects of biosafety, particularly in regard to
Practice guidelines on Biosecurity for BRCs, and the Laboratory
implementation of national legislation. Concerns exist on finan-
Bio-risk Management Standard, CWA 15793 (2008).
cial constraints of BRCs/culture collections to implement best
In analogy with biosafety, biosecurity also recognizes four risk
practice regarding biosecurity, particularly with the requirement
categories. They are labeled negligible, low, moderate, and high
of risk assessment. Another key concern is the lack of easy access
risk, and by necessity the definitions for these categories do not
to regulations and other information regarding national rules
allow unambiguous classification. Moreover, they are based on
and regulations governing the movement of materials. It is
the biosafety classification, and hence focused on threats against
evident that culture collections adopt compliant procedures
humans, not crops.
firstly governed by national laws but specifically compliant
The international standardization of lists dealing with the
with the Biological and Toxin Weapons Convention (BTWC).
organism content of biosecurity risk groups, or at least with
They must endeavor to reduce the potential for misuse of bio-
those directly affecting humans, would be advantageous, but
logical agents, toxins, or associated information or technologies.
the political impediments are considerable and have not been management throughout the organization, provide adequate
solved yet. With regards to plant pathogens, the situation is even resources and identify opportunities for improvement and pre-
worse, because national legislation only concentrates on national vention. They are responsible for the appointment of qualified
interests, e.g., in the absence of a host in the respective country, staff and for subsequent training to maintain the desired quality.
a pathogen for that host is not considered to represent a risk. They have to convince the funding bodies of the necessity for
However, potential abusers may obtain such material from good biosafety and biosecurity management and to provide the
research groups or BRCs in those countries where such an organ- personnel with a supportive environment, involving working
ism is not on the quarantine list. This must be prevented by space and equipment. They also are responsible for compliance
all means. with legal requirements, communication to staff and relevant
In order to decide on the necessary biosecurity measure- third parties, and for a reliable and appropriate risk assessment.
ments for a specific organism, a risk assessment has to be Finally they are responsible for screening the outgoing informa-
performed. As the potential targets for dual use are not only tion on potential dual use. In practice they should appoint
humans but also crops, life stock, or the human environment in a Biosecurity Officer to ensure internal compliance with the
general; these elements have also to be considered, and the adopted regulations.
biosecurity classification cannot be a simple translation or adap- In order to obtain maximal collaboration of staff, it is essential
tation of an existing biosafety classification. Although for the that the awareness level be high. It is necessary to devote specific
human aspects there may be a considerable similarity, there will and sufficient attention to the education and additional training
also be significant differences. For example, the highest biosafety of all staff to the risks of misuse of biological material, information
category contains only viruses, while the highest biosecurity and life sciences research and the requirements of regulations in
category contains also Bacillus anthracis, Francisella tularensis, this context. This requires not only training in, but also auditing
and Yersinia pestis, next to the toxin producer Clostridium botuli- of, knowledge and practices with regards to biosecurity. More-
num. Especially where risks for crops are concerned, the role of over it is also the responsibility of a biological resource collection
fungi (which with a single possible exception does not qualify for (BRC) or research group to inform involved third parties on their
the highest biosecurity level with regard to humans) becomes responsibilities, for example, when high-risk biological material is
important. shipped to authorized users.
The requirements for a risk assessment in compliance with Within an organization, accountability is an essential element.
the OECD Best Practice guidelines are high. When sources of Both management and staff should be aware of the presence,
potential harm have been identified, the following elements location, and risk of the organism they are using. BRCs should
have to be considered to assess the potential for misuse: maintain and update inventories of the biological material in their
custody. Any finding or suspicion of misuse of biological material,
– Availability of the organism in nature
information, or technology has to be reported immediately and
– Requirements (necessary skills and knowledge) for isolation
directly to competent persons or commissions within the organi-
and reproduction
zation. To maintain trust in these institutions, persons reporting
– Environmental viability (survival chances)
on misuse have to be protected. They must not suffer any adverse
– Conditions for dispersal (air or contact)
effects from their actions.
In case of virulence, knowledge on the infective dose, path- Restriction to the accessibility of potential dual use biological
ogenicity, lethality, incubation time, and transmissibility is material is a vital element in biosecurity management.
required, as is information on the presence of effective Depending on the appropriate biosecurity risk level resulting
countermeasures. from the biosecurity risk assessment, physical security has to be
It will be clear that due to the high demand, many organiza- selected. For a low level, a generally secure area is sufficient, but
tions and institutions cannot fully comply with these require- for a moderate or high-risk level, respectively, a restricted area or
ments, and ways have to be found to remedy this. In practice, a high-security area, with different degrees of containment, are
bio-risk assessment is performed by comparison, which includes necessary. Staff and visitors have to be screened before access is
the biological material, its interaction with the substrate, dispersal allowed to areas in which potential dual-use biological materials
system, knowledge of properties of taxonomic relatives, even are stored or used.
tests on the organism itself. Data on virulence are usually absent, Scientists and BRCs should undertake an information risk
or scattered in the literature, and a tedious search for such assessment to determine what information presents a potential
properties need to be performed. In practice it has worked well biosecurity risk and steps need to be taken to protect such
as far as the author is aware, but in order to comply with the information that could be used to locate the material and facili-
guidelines, collaboration with other facilities and access to spe- tate theft. During assessment or application procedures and
cialist knowledge has to be established. In this scenario, the during the execution of research projects on potential dual-use
outreach of international societies could play an essential role. aspects, emerging threads have to be considered. Any risk that
Within an organization having to deal with biosecurity issues, publication of results on potential dual-use organisms will
any respective measures need to be implemented as part of the contribute to misuse of that knowledge has to be minimized.
quality management system and regulated and supported by the Access of unauthorized persons to internal and external
senior management. It is their task to integrate bio-risk e-mails, post, telephone calls, and data storage concerning
information about potential dual-use research or potential dual- ● To assist approved BRCs to gain higher levels of compliance
use materials has to be prevented and communication of sensi- to the OECD best practice guidelines (BPG)
tive information has to be regulated. ● To support political and socioeconomic decision-making
Transport of biological material classified as moderate or high processes in the fields of BRC related bio-economy
requires special conditions, both inside and outside the organi-
zation. Inside the organization, transport containers are required
and high-risk material may never be unattended outside the Authenticity Check of Microorganisms
high-security area. In case of transport to third parties, both
these parties and the transporters have to be screened for both One of the most crucial tasks of public collections is the delivery
their capacity to deal with the material and their intentions to of authentic material that matches most closely material origi-
prevent dual use, in consultation with the relevant authorities nally accepted by the collection. Changes can occur through
and parties. Export control has to be performed in accordance human error during routine maintenance steps of sub-
with applicable regulations. cultivation or intrinsically due to one of the laws of biology,
For packaging, the WHO guidelines on International Regula- namely, populations of organisms change genetically and irre-
tions for the Packaging and Transport of Infectious Substances versibly through time (http://hunstem.uhd.edu/ABOUT/).
and the International Air Transport Association (IATA) Dangerous Especially in microorganisms, detection of such changes is chal-
Goods Regulations (DGR), or other applicable regulations, for lenging as short generation times and high mutation rates
example, for road transport regulations in various countries, unavoidably lead to the formation of heterogeneous cultures.
should be utilized. Diversification occurs at the genetic and epigenetic level during
International projects and organizations are now working on growth and repeated cultivation as has been observed for
a Code of Conduct for BRCs, combined with an inventory of decades and experimentally proven more recently (Lenski and
problems occurring in practice when implementing the guidelines. Travisano 1994; Rainey and Travisano 1998). As pointed out by
They are also involved in the setting up of a database, which should Arber (2003) using Escherichia coli as an example, spontaneous
allow fast and reliable information on legislation and regulations genetic mutations occur in single cells at a rate of 0.1–1 % of cells
per country, for example, import and export regulations for micro- per generation during exponential growth and also in the sta-
organisms, transport regulations, quarantine organisms, biosafety tionary phase. If cells were singled out, the genetic variation
and biosecurity regulations, classification lists for human patho- would be subjected to natural selection and consequently to
gens, animal pathogens and plant pathogens. It should also con- reproductive and geographic isolation. The observation of eco-
tain a list of experts that could advise on biosecurity items. types reported for strain-rich species, differing in physiology,
genome content, and ecology (Cohan and Perry 2007), already
Joost Stalpers starts at the strain level.
CBS, Centraalbureau voor Schimmelcultures, Utrecht, Collection curators are aware of the risk of increased micro-
The Netherlands evolution by repeated sub-cultivation, most obvious by the loss
of plasmids and changes in colony morphology, and have
reacted by keeping the numbers of culture transfers to
a minimum. As minute changes may occur during growth and
Capacity Building sudden environmental changes such as refrigeration and freez-
ing, most public service collections in their terms of delivery
The paucity of national BRCs and the challenges outlined above include a clause in the paragraph concerning the quality of
necessitate an efficient coordinated program for capacity building. products phrases such as the one used by the DSMZ: ‘‘Customer
Again, a task that is better organized and delivered by a network. is aware of the products and services being biological material
The major regions or countries of the world are serviced by collec- and therefore subject to changes of quality beyond the control of
tions providing whole organisms, but there are gaps particularly in DSMZ. DSMZ’s high quality standard supports availability of
countries rich in diversity but poor in resources. There are about 60 pure, viable and authentic biological material. DSMZ shall not
major collections supplying cultures in Europe while there are very be deemed to have guaranteed certain properties of the products
few in Africa. It is necessary to establish expertise and facilities to and services except if it has expressly confirmed such guarantee
access conserve and utilize microbial diversity. Capacity building is in writing.’’
one of GBRCN’s pillars for the sustainable development of Biolog- In order to ensure as far as possible the originality of the
ical Resource Centers. Through the transfer of knowledge as well as cultures, strains deposited are preserved with a minimum num-
skills and abilities, GBRCN supports BRC candidates as well as ber of transfers. The type of preservation method depends on the
acknowledged BRCs and accompanies them through the develop- organism, but freezing a few straws or alternatively glass capil-
ment process. The focus of the capacity building is laries with material from the original culture received in liquid
nitrogen is good practice (see http://www.cabri.org/guidelines/
● To help countries to establish BRCs micro-organisms/M204.html). Depending on the estimated
● To enable BRC candidates to become BRCs and GBRCN number of requests by users, this master stock should contain
members between 6 and 50 straws/capillaries, one of which is used for
the production of lyophilized cultures. The number of replicates the applicability of these techniques to the level of individual
of this stock, too, depends upon the estimated user demand. laboratories. Molecular methods can differ widely in their ability
Authentication differs from the characterization of a strain to differentiate strains, and the user should be aware of the
in goals and effort. Characterization attempts to circumscribe strength and restrictions of the individual methods. An excellent
a novel organism by as many properties of taxonomic value as summary of early techniques available in the 1990s, their intra-
possible to allow subsequent unambiguous identification. The and interlaboratory reproducibility, equipment needed, costs,
range of properties to be determined is usually set by the prop- and duration, has been given by Olive and Bean (1999).
erties defined to be useful for members of taxa (usually species Among the typing methods, the ribotyping (Grimont and
and genera) described previously. Authentication may only use Grimont 1986) approach has been successfully established in
a subset of such properties or, especially today, molecular traits several public and industrial collections, executed either manu-
that had not been included in the original description. In both ally or by the automated Riboprinter microbial characterization
cases, characterization and authentication, purity checks of the system (E. I. du Pont de Nemours & Co., Inc.). It has especially
culture under investigation, is mandatory. In the pre-molecular facilitated strain identification over the past 10 years through the
era, authenticity checks relied upon microscopy, culture charac- ease and reproducibility of the robot functionality. The advan-
teristics (e.g., pigmentation or colony characteristics) and basic tage of ribotyping over the amplified rDNA restriction analysis
physiological tests. As the latter requires time for growth, con- (ARDRA) (Pukall et al. 1998) approach lies in the use of the
firmation of identity took at least a few days. The introduction of entire rrn operons rather than of individual 16S rRNA genes. In
semiautomated commercial substrate utilization and enzymatic that restriction sites also occur within structural genes, the
reaction panels shortened the time for authentication. These spacer regions including tRNAs and the flanking DNA regions,
tests work well for those taxa which were among the hundreds the diversity of fragments hybridizing to a fluorescently labeled
of strains used for the construction of associated databases, such rrn probe is much higher than those of ARDRA.
as those used in the food and pharmaceutical industry and on The recent introduction of the MALDI-TOF approach is
species found routinely in the clinical environment. As databases a major step forward in a fast and cheap, though highly reliable,
are not available for strains of new species, the use of such authentication at the strain level. While riboprinting works on
identification systems often delivers disappointing identification the gene level, the highly amplified ribosomal proteins are the
results and misidentifications. Also, as nowadays descriptions of main targets for the MALDI-TOF spectrometric method. The
newly named species embrace a single strain only, the type resolution power of this method is significantly higher than that
strain, the spectrum of reactions is usually too narrow to unam- of comparative 16S rRNA gene sequence, often being suitable for
biguously identify additional strains of such species which may affiliation of strains to the subspecies level. However, as the
deviate in their physiological reaction from the type strain. species as defined today does not necessarily reflect genomic
Nevertheless, species descriptions published over the past 20 coherency, the range of intraspecies dissimilarities determined
years rely on commercial kits for circumscribing substrate utili- for strains may vary between different taxa. A linear TOF mass
zation patterns, acid production and the activity of enzymes and spectrometer operates on the principle that when a group of ions
only rarely manually prepared media for checking such proper- of differing mass/charge (m/z) ratios move through a region of
ties (e.g., Smibert and Krieg 1994). Consequently, in order to constant electric field, they will traverse this region in a time
obtain more reliable test results an ad hoc committee for the which depends upon their m/z ratios (for more details and the
reevaluation of the species definition in bacteriology principle of a reflection TOF-MS, see http://www.abrf.org/
(Stackebrandt et al. 2002), and more recently Tindall et al. abrfnews/1997/june1997/jun97lennon.html). As not all proteins
(2010) strongly encourage authors of new descriptions to refrain occurring in a cell will be charged, only a small fraction of
from commercial kits for studying metabolic reactions. protein masses are detected, usually in the mass range between
For culture collections, the molecular era started with the 2,000 and 12,000 m/z. The advantages of working with this
introduction of one-dimensional protein patterns (Kersters and approach, especially for the identification of those species for
De Ley 1975), restriction enzyme and PCR-based techniques and which a high number of strains (mainly of clinical origin), have
16S rRNA gene sequence analyses (Fischer et al. 1983). Although been analyzed, and spectra deposited in databases have been
the latter method has the advantage of being highly reproducible outlined in Chapter X of Konstantinidis and Stackebrandt
and allows the generation of a cumulative database, the highly (2011). Public service collections usually cover a broad range
conservative primary structure does not allow discrimination at of diverse microorganisms for which commercial MALDI-TOF
the strain level. Two other molecular techniques, DNA-DNA databases are less well suited. In this case, the authentication
hybridization (DDH) (Johnson and Ordal 1968) and DNA- laboratory has to generate tailor-made in-house databases for
rRNA hybridization (DeLey and De Smedt 1975), have been specific taxa. The method has been used to verify the authentic-
used for classification at the family and genus level, respectively. ity of many type strains of different taxa maintained in different
However, they have failed to authenticate at the strain level. Even EMbaRC collections (Schumann, Bizet, Arahal, and de Vos,
the DDH method lacks reproducibility to affiliate isolates to unpublished). The majority of these strains showed highly sim-
strains. Several molecular typing methods (PFGE, AFLP, ilar MALDI-TOF protein profiles, indicating that microevolu-
RAPD, RFLP, and more) are more reliable for strain authentication does not affect the masses of housekeeping proteins and
tion though the lack of a portable database somewhat restricts ribosomal proteins; the few strains identified as being
misclassified or mislabeled have been discarded and exchanged important aspects of microbial taxonomy, ecology, physiology,
with authentic ones. MALDI-TOF appears to be the method of and biochemistry, also including data pertaining to the practi-
choice for routine identity checks of strains in long-term storage cal applications of microorganisms. The disadvantage is that
and for authenticity checks before shipment. entries need to be added manually, thus prone to errors and
omissions.
The Common Access to Biological Resources and Informa-
Access to Microbial Biodiversity Data tion (CABRI, see further details above) has its roots directly in
the previous MINE project from which it inherited the dedicated
A wealth of invaluable information for academic and industrial taxonomic data structure. Its main goals are to increase aware-
users as well as regulatory bodies is available in resource collec- ness of the scientific user community of the quality and variety
tions, and research data are scattered in hundreds of thousands of European culture collections and to ease access to informa-
of publications since the beginning of microbiology. To put this tion and material. In order to reach this objective, the project has
treasure at hand, individual approaches will not suffice any more implemented a unified access to culture collection catalogues of
to meet the global needs. Modern information technology devel- participating collections, by also guaranteeing a common level
opments will offer increasingly easier access to data, while work- of quality of material and related information. The final achieve-
ing toward interoperability of databases will make searching ment of the project has been the development of an online ‘‘one-
through countless databases all over the world practicable. In stop-shop’’ for biological resources (www.cabri.org) which
order to archive a comprehensive overview on published data allows the user to check on the availability of a particular item,
and to offer these data to the user, standardization of data interrogating one or more catalogues at the same time, and to
production, data recording, and data presentation should be pre-order the required biological resources, once located.
harmonized, at least to allow text mining software to efficiently CABRI membership is open to any recognized European BRC,
extract data. The unstructured recording of phenotypic data in willing and able to work at the CABRI quality levels.
current species descriptions of prokaryotes is a simple example Participating resource centers agree to find consensus for
of the present-day failure to cope with strain-associated data. It data fields and content type, harmonize procedures and agreeing
is, however, promising to see recent attempts to standardize the on equivalent methods and procedures and producing guide-
electronic exchange of meta-information about microorganisms lines for each type of biological material. CABRI can be seen as
which led to the definition of the Microbiological Common a pioneer model for an integrated, while distributed database
Language (MCL) (Verslyppe et al. 2010) within the framework which is searchable through a common gate. CABRI is currently
of the Bacterial Commons (Dawyndt et al. 2006). available through the main website (see http://www.cabri.org/)
and four web mirrors (see http://www.it.cabri.org/, http://www.
be.cabri.org/, http://www.fr.cabri.org/, http://www.cn.cabri.org/).
CODATA, MINE, CABRI, and GBIF: Examples of Three distinct data sets were defined for each biological
Past and Ongoing Activities material. The minimum data set (MDS) consists of mandatory
information needed to identify a unique strain or cell line:
A number of initiatives aiming at the coordination and stan- strains for which this information is not available cannot be
dardization of activities to give access to information available inserted into the catalogue since they lack essential data. The
at culture collections have been set up. Among them, WDCM recommended data set (RDS) includes useful information for an
(see > Box 11.3), CODATA, MINE, and CABRI have set base- improved description of the material. These data should always
lines and provided model roles. be included in the catalogue, when available. The Full Data Set
CODATA, the Committee on Data for Science and Technol- (FDS) provides all remaining information that is available at the
ogy, is an interdisciplinary Scientific Committee of the Interna- collection for a strain or cell line. Since the individual CABRI
tional Council for Science (ICSU) founded over 30 years ago. Its catalogues are independently built, each collection can have its
main objective is to foster and advance science and technology own FDS, although information which is available in the FDS
through developing and sharing knowledge about data and the undergoes a harmonization effort (http://www.cabri.org/guide-
activities that work with data (www.codata.org). lines/catalogue/CPexport.html).
MINE (see for details further above) was designed to har- At an international level, major attention has recently been
monize and computerize data on >150,000 strains of microor- devoted to bioinformatics for biodiversity. Discussions initiated
ganisms in European culture collections and to develop through the OECD Working Group on Biological Informatics
a common database on microorganisms held in culture collec- 1996–1999 have led to the Global Biodiversity Information
tions. Requirements for efficient data recording and computer- Facility (GBIF). GBIF aims to provide a worldwide network of
ization were established by the 12 participating European interlinked biodiversity databases so that the worlds’ scientific
countries together with details of the data structure used, the biodiversity data can be made freely available. It is based on an
hard- and software configurations, data entry procedures and implementing agreement signed by governments and interested
online access. The main merit of MINE has been the develop- organizations. The initial focus of GBIF is on species- and
ment of an internationally agreed format: 135 fields for fungi specimen-level data. It works in close contact with existing and
and yeasts and 145 fields for bacteria which cover the most ongoing activities with similar or complementary goals on
a national and international level. Implementation of GBIF is ● Spell checking for every field should be a basic
nationally and/or regionally driven (see http://www.gbif.org/). requirement.
However, microbial data are still today represented only on ● International English should be chosen as a preferred
a smaller scale. The extended Darwin Core (as developed by language of data (in addition to local language if
the Brazilian group) reflects better the needs of microbiology different).
than the ABCD Schema and will be the basis for the envisioned ● A standardised approach should be adopted to certain
future information resource within GBRCN. scientific symbols (to avoid any errors due to incorrect
reading of a character set, standard ASCII alternatives to
symbols should be used (examples follow):
The OECD Best Practice Guidelines
43. BRCs should adopt procedures to detect errors in data to
improve their quality and consistency. This is an essential part of
In the OECD BRC best practice guidelines, detailed require-
information management and should be both applied to the
ments are formulated for the processing of biodiversity data.
input of new data as well as to preexisting information in current
Clear reference is taken to previous works such as CABRI, GBIF
databases:
and WDCM. The following is an excerpt from OECD best
practice guidelines – general (DSTI/STP/BIO(2007)9/FINAL), ● For existing data, a series of checks should be carried out to
heading Data and Informatics which defines the responsibilities ascertain their validity and completeness. As more BRCs
of resource centers and depositors concerning quality assurance: become associated, more searches should be made for com-
mon classes of error to allow more efficient error correction.
● For new data, wherever possible, inputting should be
Box 11.8 Data and informatics checked against authorized lists of not only scientific names
but also thesaurus/ontology to prevent errors such as
39. The BRC should manage and store data and produce elec-
mistyping.
tronic catalogues based on authenticated and validated
● BRCs should present evidence that they have applied a
information.
recognised protocol appropriate for each data element (A com-
prehensive treatment of Data Cleaning can be found in Chap-
man, A.D., Principles and Methods of Data Cleaning – Primary
Data Management
Species and Species-Occurrence Data, Version 1.0, Publisher -
Global Biodiversity Information Facility (GBIF), 2005.
40. Depositors are responsible for assuring the quality of data
associated with the biological material. The BRC may require
evidence to assure the validity of the data.
Data Processing
41. The authentication of data may differ from center to
center, but a BRC should:
44. The informatics system employed by BRCs should provide
● Provide traceability of data through a history of modifications appropriate facilities for information management, linkage and
(dates and signatures of inputs, validations, modifications exchange.
and deletions). 45. The databases should contain either information relating
● Give signature for data entry, validation, modification or to strains held by a BRC (which at least, should be retained as long
deletion. as a strain remains viable), or other relevant data items or com-
posite data needed by the BRC (e.g. users records). On the loss of
42. The BRC should use a standard terminology and formats
a strain the database record should be either printed and stored
for data management and exchange and standard protocols for
on file or copied to a digital archive before the entry is removed
data transmission to networks (domain, regional or global
from the working database, placed in reserve or annotated to
networks):
indicate that it is no longer available as living material.
1. Select data format, data representation and data transporta- 46. The BRC should preferably choose standard data schema
tion taking into consideration existing standards for data and protocols to make the databases distributed and interoper-
processing, e.g. DarwinCore/DiGIR and ABCD schema/ able. Confidential data should be clearly identified in relation with
BioCASE for strain data, CCINFO for the organizational infor- user authentication capability, encryption techniques and other
mation of BRCs. related information security tools.
2. Check vocabulary against standard reference lists or thesauri. 47. The informatics system should ensure regular data back-
3. Keep consistency among BRCs for searching and retrieving of up. Off-site storage of data is desirable. Data archives should be
information from catalogues and databases: maintained in accordance with the maintenance of the biological
● Each biological material record should contain resource storage policy. The support of these archives should be
a Minimum Data Set, a Recommended Data Set, and/or regularly updated according to its physical characteristics (obso-
a Full Data Set in accordance with domain specific criteria. lescence) and to software compatibility.
48. BRCs should introduce appropriate measures (protocols, . Table 11.3

tools and standards) in their own informatics systems to assure Minimum (MDS) and recommended data set (RDS) bacteria
reasonable security of information. There are existing systems,
Recommended data set (in addition to
e.g. authentication by user ID and password, encryption, encryp-
Minimum data set MDS)
tion of messages and restriction of IP addresses that may provide
the basis for such measures. Backup-files should be stored in Accession number Serovar
secure cabinets. Other collection Other names
From: OECD Best Practice Guidelines –General (DSTI/STP/BIO numbers
(2007)9/FINAL) Name Isolated from
Infrasubspecific names Mutant
Organism type Genotype
Restrictions on Literature
Minimum and Recommended Data Sets distribution
Status
Based on the previous works in MINE and CABRI, the concept
History of deposit
of minimum and recommended data sets was adopted by the
OECD guidelines and slightly updated, in particular with a view Geographic origin
to the necessary information of ‘‘country of origin’’ for Conditions for growth
implementing the requirements of the Convention on Biological Form of supply
Diversity. It should be noted, however, that these attempts
defined data fields but have failed to set up a standardized,
open infrastructure that allows electronic processing.
> Table 11.3 shows as example the data sets for bacteria. Similar laboratories. It is obvious that in terms of strain mainte-
data sets are also available for filamentous fungi, yeasts, pro- nance, these facilities have different requirements than those
tozoa, cyanobacteria, archaea, virus, plasmids, phages, and required in a public culture collection. As usually neither
cDNA/cDNA libraries. equipment nor staff experience is available in the research
laboratories, some basic procedures for proper curation of
strains and information should be implemented. Still today,
Long-Term Storage of Microorganisms: At the bacterial cultures are often maintained over years through
Very Heart of Collection Activities and periodical transfer onto fresh media. However, this practice
Responsibilities is not only considerably time and material consuming but
presents risks of contamination, selection of mutants, loss,
Why Do We Need Long-Term Preservation? and mislabeling, and therefore the method should be
avoided. Scientists should identify as early as possible those
Proper preservation methods suspend metabolic activities strains worth maintaining from their research. As has been
instantly while retaining viability and genetic and physiological outlined above (> Box 11.1), scientists have the obligation to
stability of the specimen. This is the basis for the safe and long- share with peers those strains included in scientific publica-
term maintenance of strains of microorganisms of scientific or tions, meaning that subcultures of such strains need to be
industrial interest, which is an inevitable prerequisite for con- available and prepared in a manner that optimally preserves
tinuous and efficient research and production. Availability of the properties of the original culture. It is our recommen-
strains maintained in a genetic and physiological unchanged dation to contact public service collections to guarantee
state must be guaranteed over years. For example, important availability and dispatch according to national rules and
production strains, reference strains for research and testing and regulations.
other strains with valuable properties should be available for A number of reliable methods for short, medium, and
comparative examinations even after decades. Or, to give long-term maintenance of microorganisms have been devel-
another example, sometimes years after a certain bacterial spe- oped (Kirsop and Doyle 1991; Day and Stacey 2007). Some of
cies has been described, it becomes clear that strains of this the short-term methods are simple and may be performed in
species have important, useful properties. It is then most helpful any laboratory but may not be successful for a broad range
to have the reference strain available in the most original state of organisms. For long-term methods, more sophisticated
and ready to be shipped. If such a strain is lost, time, informa- equipment is required; these, in turn, seem to be effective
tion, and research funds are lost as well, and the re-isolation of for most microorganisms. However, it must be stated that
strains with exactly the same properties as that of the type strain there is no universal preservation method for all microor-
is highly unlikely. ganisms. Different taxonomical groups, even strains within
Despite the fundamental significance of reliable availabil- a species, may react differently to the various preservation
ity of pure and stable cultures, quite often, only little atten- conditions. > Table 11.4 gives an overview of different long-
tion is paid to the maintenance of these cultures in research term preservation methods.
298
11
. Table 11.4
Comparison of some methods used for long-term preservation
Costs for Working time Working time Working time Dependence

material and consumed consumed over consumed on Survival of Genetic Space Danger of from power
Method equipment initially storage time supply organisms stability needed Suitability for shipping contamination supply
Periodic Low Low High High 2–6 Low High Active cultures High Cool room
transfer monthsa acceptable required
Drying Low Medium Medium High 1/2–7 yearsa High Low Difficult for individual High Can be
over silica replica; needs independent
gel reactivation
Deep Medium Medium Medium High 1/2–7 yearsa Medium Medium Needs reactivation or Depending on Dependent
freezing shipping on dry ice method; high
80 C
Freeze High High Low Low > 40 yearsb Highd Medium Very good Low Can be
drying independent
Public Service Collections and Biological Resource Centers of Microorganisms
Cryo- High Low Low High > 40 years High Low Needs reactivation Low Can be
storage in ‘‘indefinite’’b independent
LN2
a
Depending upon taxon and strain
b
In service collections, experience indicates that organisms that survive the first ~5 years will survive ‘‘indefinitely’’ in all probability
c
Experience shows that if organisms survive the initial freezing and, upon recovery, the thawing process, they will survive indefinitely
d
If drying is performed too extensively, bound H2O may be removed; this may result in DNA damage
Choice of Preservation Technique As a conclusion, the safest storage for microorganisms to be

preserved long-term is provided by liquid nitrogen. Nitrogen is
Each of the different methods has its advantages and disadvan- also much safer than other liquefied gases as it does not burn, is
tages which we need to know to ensure the right one is applied to not toxic, and is cheaper than other gases. Nevertheless, care
meet the special conditions or needs in each circumstance. The must be taken to avoid suffocation due to displacement of
selection should be decided upon after comparison of the con- oxygen. However, when considering continuous supply of cul-
ditioning factors of a given method, the available equipment and tures, drying, and storage of microorganisms under vacuum is
the needs of the user. the method of choice for preservation.
Some general aspects should be considered with any choice
of methodology:
Factors Influencing the Survival Rate of
● Viability should be maintained as high as possible.
● Genetic changes should be avoided as far as possible (this is
Microorganisms During Freezing and Drying
also most likely supported by a high survival rate).
Successful preservation of microorganisms not only depends on
● The risk of contaminating the preserved culture should be as
the application of an appropriate cooling, drying, thawing or
low as possible.
rehydration regime. Other factors (> Table 11.5) determined by
Some technical aspects refer to effort/efficiency balance the organism itself have been shown to be important, such as
which is especially important when larger numbers of organisms type and strain of the organism, growth conditions, nutritional
need to be processed: status, and growth phase. Additionally, the diluting medium and
growth medium used for reactivation and determination of
● The number of ampoules or replicas to be prepared should
viability will influence their recovery.
be considered according to the procedure chosen (expense of
Nevertheless, for most procedures, basic factors determining
work, material or space for storage, authentication proce-
survival of the preserved organisms are similar.
dure, demand).
● If cultures are to be supplied to third parties, they must be
present in an appropriate form; public collections almost
Protective Suspension Media for Freezing or
always assume that strains will be requested.
(Freeze-) Drying
● The present or envisaged market need for a given culture.
● The availability of space in the facility, its financial situation,
For the protection of cells against damage during drying, freez-
and personnel expertise and availability.
ing, and freeze-drying as well as during storage, microorganisms
1. Drying methods have the advantage that, once the culture have to be suspended in a protective suspension medium. The
has been successfully preserved, material can be stored composition of these media may depend upon the type of
independently from power supply. This can be done in organism to be preserved. Some examples of protective suspen-
the dark at ambient temperature, though storage at lower sion media are given below. However, as far as possible, widely
temperatures between +10 C and 20 C extends applicable routine methods should be established. Especially
shelf life considerably. Dried specimen (‘‘ampoules’’) can with larger collections, the laborious and time-consuming indi-
be used perfectly for shipping of cultures, and the method vidual treatment of each culture cannot be afforded.
bears relative low costs for material. Protecting effects of compounds have been assigned to
2. Storage at ultralow temperatures (below 139 C), for maintaining macromolecular structures (replacing H2O mole-
example, the procedure of preservation in liquid nitrogen cules) (Suggett 1975), protecting against O2 or oxygen radicals
(LN2) at 196 C (‘‘cryo-storage’’) is much faster and (Lion and Bergmann 1961), avoiding damage to membranes
more reliable than any other methods. As a drawback, (Morichi 1970) or maintaining a certain level of residual mois-
costs for material are higher, and preserved cultures need ture (Nei 1974; Danilova et al. 1980).
to be revitalized and incubated before shipping. For drying processes, complex organic substances, for exam-
3. Temperatures of 70 C to 90 C as generated by ple, skimmed milk, serum, and peptones and also pure sub-
electrical deep freezers or by solid CO2 (below 78 C) stances, for example, sugars, amino acids, and mixtures thereof
have been shown to give useful results for the preserva- have been proven supportive for keeping high viability.
tion of some types of microorganisms. Freezing at Substances protecting living cells against freeze-thaw injury
80 C is often applied especially in research groups. (cryoprotectants) can, on the one hand, be compounds with
This can be an acceptable method for medium-term defined low molecular weight, such as glycerol, dimethyl-
storage of cultures maintained for own, in-house use sulfoxide (DMSO), methanol, or sugars. On the other hand,
within a research group. However, these temperatures these can be compounds with defined high molecular weight,
range within the margin where water migration into cells such as starch, hydroxyethyl starch (HES), or polyvinylpyr-
is possible, and therefore, cells have to be carefully rolidone (PVP) and undefined substances, such as proteins,
protected by appropriate additives to obtain reasonable malt extracts, or blood (Farrant 1969; Fry 1966; Fuller 2004;
periods of storage. Heckly 1978).
. Table 11.5 Non-penetrating cryoprotectants work differently as they

Factors influencing preservation
● Cause an osmotic dehydration of cells
Factors Comments ● Reduce extracellular salt concentration
● Influence extracellular ice formation
General Kind of organism Size, taxon, strain, suspendability
● May stabilize membrane structures
Culture Nutritional state
conditions An extensive compilation of protective suspension media
Age Mid to late log phase suitable for the freezing and freeze-drying of microbial strains
Concentration of The higher the bettera can be found on www.cabri.org (> guidelines > microorgan-
cells isms > Part 3: Guidelines for maintaining deposits – Appendixes
Media for Protective media, stabilizing – M/1998/3.00 Appendix 3).
preparing the structure, replacing H2O,
suspension protection against O2, retention
of residual moisture Simple Methods for In-House Purposes
Freezing Freezing and Storage at as low as possible
storage Cultures may need to be maintained by simple methods, for
temperature example, as cultures on slants which are over-layered with sterile
Freezing velocity paraffin oil, in distilled water or by simple drying methods. In
Cryoprotectant Depending on the permeability any case, periodical transfer onto fresh media should be avoided
of the membranes due to the fairly high danger of contamination and physiological
Drying Drying and genetic changes (see above).
temperature Cultures may be dried in earth, sand, pumice-stone, above
Drying velocity silica gel, or on porcelain or glass beads; however, some facts
should be kept in mind:
Residual moisture Influenced by length of the
drying process, temperature of ● A protective medium like skimmed milk or skimmed milk
cold trap and final vacuum with myo-inositol, serum, or nutrient broth should be used.
(ideally 10 1 to 10 2 mbar) ● The amount to be dried should be as small as possible.
Storage Best under vacuum (without O2) ● The drying process should not take too long.
conditions like: 4–10 C recommended ● The dried cultures should be stored in the cold, if possible
Gas
under vacuum and dark.
atmosphere
Temperature It is recommended to use such preparations rather for
Method of Medium in-house purposes than to use them for supply to third parties.
reactivation Temperature The problem usually accompanied with these methods is that the
Rehydration time With freeze-drying: allow the storage receptacle needs to be opened many times to remove
material to rehydrate for 10 min the required sample of the dried organism. This immediately
Thawing velocity With LN2: plunge into 37 C
presents the danger of contamination and negative impact on
warm water survival rates.
a
When drying over silica gel on glass or porcelain beads is
Concerning the cell concentration mentioned above, it has been demon-
chosen, the bacterial suspension is surface-dried but without
strated that the survival rate is positively influenced by an initially higher
concentration of the cell suspension. While for various organisms with direct contact with the drying agent. Silica gel develops consid-
an initial cell density between 107 and 108, a drop by two log levels was erable heat when taking up water, which may be harmful for the
observed; no drop was observed when the initial cell density ranged between organisms attached. Silica gel with blue indicator (toxic CoCl2)
1010 and 1011 may be used, though today it is recommended to avoid CoCl2
and silica gel with other indicators are available. The amount of
drying agent should be sufficiently large so that only the
smaller part of the silica gel will change in color during the
The effects of penetrating cryoprotectants are manifold,
drying process.
as they
The method of drying over silica gel in gelatine disks, orig-
● Partially replace intracellular water inally described by Stamp (1947), uses a protective medium
● Thus prevent a too high increase of salt concentration containing peptone, meat extract, gelatin, and sodium ascor-
● May also replace water molecules for the stabilization of bate. The complex organic compounds are used for stabilizing
proteins and membranes macromolecular structures and/or to serve as physical barriers
● Influence ice crystal formation to maintain a certain residual water content. Na ascorbate is
● Like DMSO (NOTE: toxic), increases the permeability of added as an oxygen radical trap, as it is suspected that cell
membranes damage occurs through oxygen radicals. During hardening and
drying of the gelatine drops, small disks are formed which Protective Medium
may be stored in presterilized screw cap tubes containing dry
silica gel. For many microorganisms, skimmed milk has been proven an
effective protective agent. To avoid caramelization, skimmed milk
should be autoclaved in small amounts at 115 C for only 13 min.
Preferred Methodologies for Long-Term Thorough sterility testing is therefore necessary, particularly for
Storage: Freeze-Drying and Cryo-storage in the presence of heat-resistant spores of thermophilic organisms.
Liquid Nitrogen Thus, testing should be performed at 30 C as well as 55 C.
General Aspects of the Freeze-Drying Process

Sterility
Preservation of microorganisms by drying under vacuum from
the frozen state (through sublimation of ice) has been used for During preparation of freeze-dried cultures, both, the cultures as
more than 60 years. Methods and equipment have been developed well as the personnel, must be protected from contamination or
over the years and nowadays present a reliable and effective hazard. In parallel to the common safety precautions for micro-
preservation method for most bacteria, fungi, and yeasts. During biological work, it must be observed that during the drying
the freeze-drying process, wet material is frozen and the ice process cell material may escape the ampoule in the form of
directly transferred into the gas phase. The ice sublimes without fine particles and contaminate the vacuum chamber or the
melting. The porous cake resulting from this has, in principle, the whole freeze-drying apparatus. To avoid this, the ampoules
same size and shape as the original frozen mass. Through adding must be provided with a filtering closure, which, simultaneously,
of water or culture medium, the original state is reconstituted. In will avoid contamination of the culture when air is allowed to
general, freeze-dried material is highly soluble. However, it should enter the vacuum chamber after the first or primary drying
be noted that freeze-dried organisms are extremely susceptible to (when true freeze-drying is applied).
oxygen. To exclude this negative effect, the cultures should be
stored in glass ampoules sealed under vacuum.
End Vacuum
The evacuation process must be monitored. Optimally, a final

Practical Aspects of the Freeze-Drying Process
vacuum between 10 1 and 10 2 mbar should be reached to
guarantee good survival rates over longer periods.
Media for Cultivation
Note: The use of silica gel as moisture indicator as used with
the double-vial ampoule is meant as an ‘‘optical help’’ only to
Microorganisms should be cultivated on media which allow
indicate loss of vacuum during storage. The change in color of
good growth and from which they can be harvested easily.
the indicator early on in the drying process does not mean that
Incubation on agar slopes is preferred. In the case of liquid
a sufficiently deep vacuum has already been reached.
cultures, these must be centrifuged before suspending in the
protective medium.
Sealing Off of Ampoules
Age of Cultures Sealing off ampoules is done to maintain a vacuum. Care
should be taken that the tips are perfectly sealed and rounded
Fast-growing organisms are harvested generally after about 24 so that cracks or breakage during storage can be avoided. In
h of incubation. This is around the mid to late logarithmic practice, freeze-drying is performed in various ways adapted to
phase. Slow growing organisms must be incubated adequately. specific needs.
Spore-forming bacteria and fungi are incubated until optimal
spore formation.
True Freeze-Drying Process
Ampoules for Freeze-Drying With this method, the bacterial suspension is mixed with pro-
tective medium, then frozen and transferred to the vacuum
Within a small margin, the dimensions of the ampoules are of chamber in the frozen state. Vacuum is applied before the
minor importance. However, with the ‘‘single-vial-method,’’ suspension starts melting, and water is removed by sublimation.
vials may be constricted by hand if the inner diameter is around A full description of the procedure can be found in www.
6 mm. When preparing the ‘‘double-vial ampoule’’ (see CABRI cabri.org (Guidelines > ‘‘Click here to read the guidelines’’ >
guideline for more details), outer tubes with an inner diameter Microorganisms > Part 3: Maintaining deposits > Appendixes;
of about 14 mm are recommended. With these, using an edited and amended > M/1998/3.00 Appendix 5.08 ’Preservation
ampoule constrictor machine is recommended. of Bacteria by Freeze – Drying (True Freeze-drying)). In M/1998/
3.00 Appendix 5.08.1, a flow chart of the freeze-drying procedure ● Contamination of the culture through air entering the
is shown. For recording each step of the preservation procedure ampoule when opening
and results of viability checks, protocol form M/1998/3.00 Appen- ● Release of fine particles of the dried bacterial mass into the
dix 5.08.2 is suggested. air (the sudden inrush of air when cracking an ampoule may
result in a back surge of particle-loaded air), thus contami-
nating the air of the laboratory
Centrifugal Freezing
Note: If cultures belong to hazard group 1, (freeze-) dried
To shorten the exposure time to air/oxygen, the decreasing tem- cultures in ampoules sealed under vacuum can be opened in an
perature in the freeze-drying chamber due to evaporation can be ordinary transfer cabinet. In other cases, ampoules should be
an alternative method for freezing, as the removal of water under opened in a biohazard safety cabinet of the appropriate level.
vacuum results in a quick loss of about 10 % water in a relative A full description of the methods can be found under www.
short time. As this is an energy-consuming process, the residual cabri.org (> guidelines > microorganisms. Part 3: Guidelines
suspension will freeze. To avoid the strong frothing, which would for maintaining deposits – Appendixes M/1998/3.00 M/1998/
normally occur and which would expel some of the contents from 3.00 Appendix 5.14).
the ampoules due to the release of gas, the samples are centrifuged
during the evaporation process until the material is frozen.
A full description of the procedure can be found in www. Cryopreservation In or Above Liquid Nitrogen
cabri.org (Guidelines > ‘‘Click here to read the guidelines’’ >
Microorganisms > Part 3: Maintaining deposits > Appendixes; General Aspects
edited and amended. Flow Diagram M/1998/3.00 Appendix
5.10.1 ‘‘Centrifugal Freeze-Drying’’). Freezing of living cells or parts of them to very low temperatures
and storage at these temperatures stops metabolic activities and
retains viability and genetic stability of the specimens. Even
The Double-Vial, Liquid-Drying Method, as molecular motions are significantly reduced at sufficiently low
Applied in the DSMZ for Bacteria and Fungi temperatures and cease below 139 C. These characteristics
make cryogenic storage very attractive for the long-term preser-
This modified method, applied for example by the DSMZ for vation of living cells.
a wide spectrum of prokaryotes and fungi, includes a drying step Studies, as early as around 1900, have shown that microor-
from the liquid state. The advantage is less stressful for the cells ganisms can withstand freezing down to ultralow temperatures
and less water vapor developed. Due to the much smaller (liquid air, liquid hydrogen). The discoveries of Polge et al. (1949)
amount of water vapor, the drying process may be even run and of Lovelock and Bishop (1959), that glycerol and dimethyl-
without a freezing chamber. sulfoxide (DMSO) protect living cells against freezing damage,
The principle includes the transfer of a small drop of a heavy greatly influenced the further development of the technology in
suspension of organisms in fresh medium onto the porous cake of this field. Considerable progress has been made over subsequent
a pre-dried skim milk pellet. The proportion of the drop of sus- decades with regard to the control of the freezing and thawing
pension to dried skim milk is such that the amount of liquid is process to obtain optimal results. A broad range of living cells –
absorbed at once and totally. This method can then be combined from the small-sized prokaryotic to the larger sized eukaryotic
with the ‘‘double-vial method,’’ where the small, cotton plug stop- cells (such as fungi; protozoa; algae; and plant, animal, and human
pered vial, containing the dried skim milk and the drop of suspen- cells and even tissues) – can be retained viable for long periods by
sion, is inserted into a bigger tube. This tube is then constricted and low temperature storage (Reed 2008; Day and Stacey 2007).
connected to the manifold of the freeze-drying machine. The safest cryo-storage for both, the organisms to be pre-
Note: As cells are under extreme stress, the time lapse served and the personnel, is that provided through liquid nitrogen:
between transfer of drops onto the milk cake and connection 196 C. Compared with other liquefied gases, nitrogen is safer – it
of the constricted vials to the manifold of the freeze-dryer should does not burn, is not toxic – and is cheaper than other more rare
be as short as possible. gases. Excellent storage containers and additional equipment is
A full description of the method can be found under www. supplied by several manufacturers in many countries.
cabri.org (> guidelines > microorganisms. Part 3: Guidelines for Living cells consist mainly of water, and in the protective or
maintaining deposits – Appendixes M/1998/3.00 Appendix 5.11). growth media, they are surrounded by water containing differ-
ent amounts of electrolytes. Ice crystal formation occurring
during freezing inside or outside the cells removes liquid water.
Opening of Ampoules This may impact negatively on cells which normally depend on
a balanced ionic environment and hydration state of their
When opening ampoules that had been sealed under vacuum, macromolecules. Shrinkage of cells and ice crystals may be
especially the one vial preparations, care should be taken to responsible for damage to the cytoplasmic membrane (Morris
avoid the following hazards: 1981). However, it should be borne in mind that ice crystal
formation not only occurs during freezing but also when cells are This chapter has been prepared under the EMbaRC project
thawed slowly to subzero temperatures; therefore, rapid thawing (EU Seventh Framework Programme Research Infrastructures
is recommended. (INFRA-2008-1.1.2.9: Biological Resources Centers (BRCs) for
To safeguard cells from freezing injuries, cryoprotectants are microorganisms (Grant agreement number: FP7- 228310) and
added to the freezing suspension. For this purpose, certain for the GBRCN Demonstration Project, financed by the
defined low molecular weight or high molecular weight (see Bundesministerium für Bildung und Forschung (BMBF), the
further above) compounds or undefined complex substances German Federal Ministry of Research and Education.
are applied. A common characteristic of such compounds is
that they are nonionic polar molecules with a pronounced abil-
ity to H-bonding. Those compounds penetrating the cell mem- References
brane should be applied in molar concentrations (0.5–1.5 M),
while non-penetrating agents are used at much lower (around Anon (2010) The WFCC Guidelines for the establishment and operation of culture
collections. http://www.wfcc.info/guideline.html. Accessed 03 July 2011
0.01 M) concentrations.
Arber W (2003) Elements for a theory of molecular evolution. Gene 317:3–11
A full description of the method can be found under www. CABRI (2002) Common Access to Biological Resources and Information
cabri.org (> guidelines > microorganisms. Part 3: Guidelines http//:www.cabri.org
for maintaining deposits – Appendixes M/1998/3.00 M/1998/ Canhos VP, Sette LD, Cupolillo E, Tigano MS, Vazoller RF (2007) O papel da Sociedade
3.00 Appendix 5.04). Brasileira de Microbiologia no suporte à consolidação da Rede Brasileira de
Coleções de Culturas de Microrganismos. Microbiologia in foco 2:40–48
Canhos VP, Ferraz de Souza RD, Ávila JPC (2009) Avanços na implementação
da Rede Brasileira de Centros de Recursos Biológicos. Microbiologia in foco
Miniaturized Method for the Cryopreservation of 9:46–47
Bacterial Cultures: The Glass Capillary Method CBD Nagoya Protocol (2011) https://www.cbd.int/abs/
Cohan FM, Perry EB (2007) A systematics for discovering the fundamental. Units
of bacterial diversity. Curr Biol 17:R373–R386
The limited capacity of storage containers, together with
Danilova MV, Nadirova JM, Emtseva TV (1980) The viability’s dependence of
increasing costs for equipment and nitrogen, creates problems lyophilized and longly stored bacteria on the quantity of residual humidity.
when larger numbers of different organisms, each in multiple Izv Akad Nauk SSSR, Ser Biol 3:449–451
replicas, have to be preserved. Miniaturized methods which have Dawyndt P, Dedeurwaerdere T, Swings J (2006) Explorating and exploiting
been developed by several investigators may help to overcome microbiological commons: contributions of bioinformatics and intellectual
property rights in sharing biological information. Int Soc Sci J 188:249–258
this problem (Hippe 1991). The basic idea of these methods is to
Day JD, Stacey G (eds) (2007) Cryopreservation and freeze-drying protocols.
reduce the volume of cell suspension to be preserved and of the In: Series: methods in molecular biology, 368, 2nd edn. Humana Press, ISBN
unit holding it. Use of small plastic straws or glass capillary tubes 1-58829-377-7
replacing the common vials or ampoules is now standard. DeLey J, De Smedt TJ (1975) Improvements of the membrane filter method for
A full description of the glass capillary method is given in DNA: rRNA hybridization. Ant van Leeuwenh 41:287–307
Emerson D, Wilson W (2009) Giving microbial diversity a home. Nat Rev
www.cabri.org (> guidelines > ‘‘click here to read the guide-
Microbiol 7:758
lines’’ > microorganisms > part 3: maintaining deposits: appen- Farrant J (1969) Is there a common mechanism of protection of living cells by
dixes; edited and amended > m/1998/3.00 appendix 5.01 polyvinylpyrrolidone and glycerol during freezing? Nature 222:1175, London
Preservation of bacteria by freezing and low temperature storage Fischer A, Kroppenstedt RM, Stackebrandt E (1983) Molecular-genetic and
in glass capillary tubes). chemotaxonomic studies on Actinomadura and Nocardiopsis. J Gen
Microbiol 129:3433–3446
Often, filamentous bacteria and fungi cannot be suspended fine
Fritze D (2005) Digital imaging of prokaryotes for taxonomic purposes. In:
enough to apply the above capillary method. Especially the fastid- Häuser CL, Steiner A, Holstein J, Scoble MJ (eds) Digital imaging of biolog-
ious, non-sporulating strains need special attention. For such ical type specimens. A manual of best practice. Results from a study of the
organisms, a smart procedure has been developed in which young european network for biodiversity information. Staatliches Museum für
mycelium together with the agar on which the organisms is growing Naturkunde, Stuttgart, viii + 304 pp, ISBN 3-00-017240-8, pp 153–171
Fritze D (2010) A common basis for facilitated legitimate exchange of biological
is punched out using short pieces of PVC straws. These can be the
materials proposed by the European Culture Collections’ Organisation. Int
larger common drinking straws or the thinner straws used for J Commons 4:507–527
artificial insemination in cattle breeding. Before punching out Fritze D, Weihs V (2000) Systematics and legislation. In: Priest F,
growth and agar, the growth is covered with a 10 % glycerol solution Goodfellow M (eds) Applied microbial systematics. Chapman and Hall,
(take care for thorough wetting of the growth), which is poured off London, pp 447–469
Fry RM (1966) Freezing and drying of bacteria. In: Meryman HT (ed) Cryobiol-
after about 2 h. Several of such straws can then be assembled in
ogy. Academic, London/New York, pp 665–696
a cryo-vial which is stored in the vapor phase of liquid nitrogen. Fuller BJ (2004) Cryoprotectants: the essential antifreezes to protect life in the
frozen state. CryoLetters 25:375–388
Gams W, Hennebert GL, Stalpers JA, Jansens D, Schipper MAA, Smith J,
Acknowledgments Yarrow D, Hawksworth DL (1988) Structuring strain data for the storage
and retrieval of information on fungi and yeasts in MINE, Microbial Infor-
mation Network Europe. J Gen Microbiol 134:1667–1689
We wish to thank all contributing collections for sharing their Grimont F, Grimont PAD (1986) Ribosomal ribonucleic acid gene restriction
data with us. Special thanks go to Dunja Martin, Hanover, for patterns as potential taxonomic tools. Ann Inst Pasteur/Microbiol
critically reading and for improving this chapter. 137B:165–175
Hawksworth DL, Schipper MAA (1989) Criteria for consideration in the accred- Reed BM (2008) Plant cryopreservation – a practical guide. Springer, Heidelberg.
itation of culture collections participating in MINE, the Microbial Informa- ISBN 978-0-387-72275-7
tion Network Europe. MIRCEN J 5:277–281 Smibert RM, Krieg NR (1994) Phenotypic characterization. In: Gerhardt P,
Heckly RJ (1978) Preservation of microorganisms. Adv Appl Microbiol Murray RGE, Wood WA, Krieg NR (eds) Methods for general and molecular
24:1–53 bacteriology. American Society for Microbiology, Washington, DC,
Hippe H (1991) Maintenance of methanogenic bacteria. In: Kirsop BE, pp 607–655
Doyle A (eds) Maintenance of microorganisms and cultures cells, 2nd edn. Smith D, Rohde C (2007) Biological resource centres compliance with law.
Academic, London, pp 101–113 Society for Microbiology, UK. http://www.sgm.ac.uk/pubs/micro_today/
Johnson JL, Ordal EJ (1968) Deoxyribonucleic acid homology in bacterial tax- pdf/0299brc.pdf
onomy: effect of incubation temperature on reaction specificity. J Bacteriol Smith D, Rohde C (2008) Safety in microbiology. Laboratory Manager Issue,
95:893–900 vol 125. Croner, UK, pp 4–6
Kersters K, De Ley J (1975) Identification and grouping of bacteria by numerical Smith D, Ryan MJ, Day JG (2001) The UKNCC biological resource: properties,
analysis of their electrophoretic patterns. J Gen Microbiol 87:333–342 maintenance and management. UKNCC Secretariat, Egham. ISBN
Kirsop BE, Doyle A (eds) (1991) Maintenance of microorganisms and cultured 0954028503
cells: a manual of laboratory methods. Academic, London, 308 pp Smith D, Ryan MJ, Stackebrandt E (2008) The ex situ conservation of microor-
Konstantinidis K, Stackebrandt E (2011) Defining taxonomic ranks. In: Rosen- ganisms: aiming at a certified quality management. In: Doelle HW,
berg E, DeLong EF, Thompson F, Lory S, Stackebrandt E (eds) The pro- DaSilva EJ (eds) Biotechnology. Encyclopedia of Life Support Systems
karyotes: a handbook on the biology of bacteria, 4th edn. Springer, New York (EOLSS). Developed under the Auspices of the UNESCO, Eolss, Oxford,
Lenski RE, Travisano M (1994) Dynamics of adaptation and diversification: UK. http://wwweolss.net
a 10,000-generation experiment with bacterial populations. Proc Natl Acad Stackebrandt E (2010) Diversification and focusing: strategies of microbial cul-
Sci USA 91:6808–6814 ture collections. Trends Microbiol 18:283–287
Lion MB, Bergmann ED (1961) The effect of oxygen on freeze-dried Escherichia Stackebrandt E (2011) Letter to the editor. Int J Syst Evol Microbiol 61:479–481
coli. J Gen Microbiol 24:191–200 Stackebrandt E, Frederiksen W, Garrity GM, Grimont PAD, Kämpfer P,
Lovelock JE, Bishop MWH (1959) Prevention of freezing damage to living cells by Maiden MCJ, Nesme X, Rossello-Mora R, Swings J, Trüper HG,
dimethyl sulfoxide. Nature 183:1394–1395 Vauterin L, Ward AC, Whitman WB (2002) Report of the ad hoc committee
Morichi T (1970) Nature and action of protective solutes in freeze-drying of for the re-evaluation of the species definition in bacteriology. Int J Syst
bacteria. In: Iizuka H, Hasegawa T (eds) Culture collections of microorgan- Evol Microbiol 52:1043–1047
isms. University Park Press, Baltimore, pp 351–361 Stalpers JA, Kracht M, Jansens D, De Ley J, Van der Toorn J, Smith J, Claus D,
Morris GJ (1981) Cryopreservation: an introduction to cryopreservation in Hippe H (1990) Structuring strain data for storage and retrieval of informa-
culture collections. Culture Centre of Algae and Protozoa, Cumbria tion on bacteria in MINE, Microbial Information Network Europe. Syst Appl
Nei T (1974) Some aspects of freezing and drying of microorganisms on the basis Microbiol 13:92–103
of cellular water. Cryobiology 10:403–408 Stamp TC (1947) The preservation of bacteria by drying. J Gen Microbiol
OECD (2001) Biological resource centres – underpinning the future of life 1:251–265
sciences and biotechnology. (http://oecdpublications.gfi-nb.com/cgi-bin/ Stromberg PM, Dedeurwaerdere T, Pascual U (2012) The contribution of public
oecdbookshop.storefront) networks to knowledge accumulation: ex situ collections in microbial
OECD (2007) Best practice guidelines for biological resource centres (Online), research (in press)
http://www.oecd.org/dataoecd/7/13/GGTSPU-styx2.bba.de-7664-3281383- Suggett A (1975) Water-carbohydrate interactions. In: Duckworth RB (ed) Water
DAT/38777417.pdf Accessed 20 Jan 2011 relations of food. Food Science and Technology, a series of monographs.
OECD (2011) The bio-economy to 2030: designing a policy agenda. The Academic Press, London, pp 573–586
Organization for Economic Cooperation and Development (OECD), Paris Tindall P, Kämpfer P, Euzéby J, Oren A (2006) Valid publication of names of
Olive DM, Bean P (1999) Principles and applications of methods for DNA-based prokaryotes according to the rules of nomenclature: past history and current
typing of microbial organisms. J Clin Microbiol 37:1661–1669 practice. Int J Syst Evol Microbiol 56:2715–2720
Polge C, Smith AU, Parkes S (1949) Revival of spermatozoa after dehydration at Tindall BJ, Rosselló-Móra R, Busse HJ, Ludwig W, Kämpfer P (2010) Notes on the
low temperatures. Nature 164:666 characterization of prokaryote strains for taxonomic purposes. Int J Syst
Pukall R, Brambilla E, Stackebrandt E (1998) Automated fragment length analysis Evol Microbiol 60:249–266
of fluorescently-labeled 16S rDNA after digestion with 4-base cutting restric- Verslyppe B, Kottmann R, De Smet W, De Baets B, De Vos P, Dawyndt P
tion enzymes. J Microbiol Meth 32:55–64 (2010) Microbiological common language (MCL): a standard for electronic
Rainey PB, Travisano M (1998) Adaptive radiation in a heterogeneous environ- information exchange in the microbial commons. Res Microbiol
ment. Nature 394:69–72 161:439–445
12 Repositories for Patented and
Safeguarded Material
Zea D. V. L. Mayerhoff . Irene von der Weid . Alessandra B. G. Valladão
Brazilian Center of Biological Material, Instituto Nacional da Propriedade Industrial, Rio de Janeiro,
Brazil
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 viable sample in an official culture collection, so that the technol-
The Role of the Intellectual Property System in ogy can be made available as a living organism to other parties.
Technological Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 This chapter discusses international arrangements for ensur-
Culture Collections and the Preservation of ing that there are culture collections able to support interna-
Microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306 tional patent obligations by accepting deposits of biological
material. In addition, it discusses the main features of patent
The Budapest Treaty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 and safe deposit of biological material, as well as its importance
for biotechnology development.
International Depositary Authorities . . . . . . . . . . . . . . . . . . . . . 308
Geographical Distribution of IDAs . . . . . . . . . . . . . . . . . . . . . . 308
Definition of Depositable Materials . . . . . . . . . . . . . . . . . . . . . 308 The Role of the Intellectual Property System in
Technological Development
Patent Deposit and Safe Deposit of Biological
Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 Technological innovation has been long considered
The Deposit of Biological Material for a determining factor in economical development and social
Patent Purposes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 changes. At the beginning of the twentieth century, Schumpeter,
Depositing a Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 in his Theory of Economic Development, defined development as
Main Tests Performed by the IDA . . . . . . . . . . . . . . . . . . . 310 the process of discontinuous change and disequilibrium brought
Methods of Preservation for Biological Materials . . . 311 about by innovation. Schumpeter described innovation as the
When Can an IDA Refuse a Sample? . . . . . . . . . . . . . . . . 311 introduction of a new good or a new method of production, the
Dispatch of the Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 opening of a new market, the conquest of a new source of supply
Length of the Deposit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 of raw materials or half-manufactured goods, and the carrying
Safe Deposit of Biological Material . . . . . . . . . . . . . . . . . . . . . . 313 out of the new organization of any industry (O’Hara 1994).
Intellectual property refers to a collection of specific rights,
Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 which attach to the results of intellectual activity such as in the
industrial, commercial, scientific, literary, and artistic fields. An
intellectual property right is a legal right based on national law
Introduction encompassing that particular type of intellectual property.
Industrial property is a branch of intellectual property system
Microorganisms have been used in a broad range of biotechno- that covers inventions, industrial designs, trademarks, service
logical processes for a very long time, especially for fermentation marks, commercial names and designations, including indica-
of food and drink, but their industrial use has intensified in tions of source and appellations of origin, and protection against
recent decades as a consequence of advances in handling and unfair competition (Kalanje 2005; WIPO 2004). Intellectual
characterization techniques for microorganisms and their property is considered to play an important role in creating an
products. innovative environment since the effective use of intellectual
The intellectual property rights of many microbial-based property tools reduces risks for innovators and enhances the
technologies are protected by patents. The use of the patent competitiveness of technology-based enterprises, regardless of
system requires disclosure of novel aspects of the technology to whether they are commercializing new or improved products,
be protected in order to guarantee a flux of information between or providing service on the basis of a new or improved technology
inventors, and thus stimulate the development of new technol- (Kalanje 2005).
ogies or the improvement of existing ones. The territoriality principle of intellectual property deter-
In the case of patent applications that involve the use of mines that protection for a result of intellectual activity, such
biological material, a patent office can require the deposit of a as a patent or an industrial design, is restricted to the territory
306 12 Repositories for Patented and Safeguarded Material
where it was granted and that this granting has to be made in Article 1 of the Convention on Biological Diversity (CBD 1992),
accordance within the legal rules and scope defined by that the international legal instrument signed by 150 government
territory. Common ground rules have been adopted interna- leaders at the 1992 Rio Earth Summit with the aim of promoting
tionally over the past 150 years through multilateral treaties, global sustainable development. Sustainable economic develop-
which resulted in a basic harmonization of the intellectual prop- ment includes both the discovery of new products to promote
erty system at international level. human well-being and the preservation of endangered speci-
Patents play a major role in the protection of technological mens or ecosystems. The preservation of biological material
creations. International rules for granting patents have been the used in biotechnological inventions by depositing a sample
object of important treaties, such as the Paris Convention for the with a trusted repository institution can contribute to these
Protection of Industrial Property, dating from 1883 (Paris Con- major objectives of the convention.
vention 1883), and the Agreement on Trade-Related Aspects of Biological materials such as microorganisms have been used
Intellectual Property Rights (TRIPS 1994). Under the Paris in a broad range of biotechnology-based processes for thousands
Convention, a patent is a temporary and exclusive right granted, of years, but their use has intensified in recent decades due to
upon request, by a state to an inventor or a legally entitled advances in our understanding of the genetic code, of its orga-
person, for the commercial use of a new technology. Under the nization, and of gene expression. Microbial genetics has played
TRIPS Agreement, the member states should grant patents to a pivotal role in our understanding of metabolic pathways and
eligible inventions, in any field of technology, and the term of enabled the production of both natural and genetically
protection available should be for a minimum of 20 years. engineered products, such as antibiotics.
The basic requirements for the granting of a patent are Microorganisms have been widely used for fermented food
novelty, the involvement of an inventive step, and the industrial and beverage production for several thousand years by societies
application of the subject to be protected. Full disclosure of the all over the world. These uses were expanded to the manufacture
invention is also a common requirement for granting a patent. of different products at an industrial level in the first half of the
All patent applications are compulsorily published and there- twentieth century. This new era began during World War I, when
fore, the information disclosed can be easily accessed, aiming to processes to produce chemicals for munition manufacturing
guarantee the flux of information among developers, thus stim- were developed in England and Germany. Many fermentation,
ulating the development of new technologies, or the improve- bioconversion, and enzymatic processes were soon developed,
ment of existing ones, by competitors. In most cases, the yielding useful products with large markets, including amino
disclosure, which has to be sufficiently detailed as to enable acids, nucleotides, vitamins, organic acids, solvents, vaccines,
others ‘‘skilled in the art’’ to reproduce the process or product polysaccharides, and antibiotics (Demain 2000).
themselves, can be presented as a written description of the A new change came in the mid-1970s with the discovery and
invention. The patenting of inventions in which living material harnessing of basic subcellular processes, such as replication,
plays an essential role can be difficult to describe due to the recombination, DNA repair, and gene organization and expres-
inherent unpredictability and complexity of biological systems sion (gene transcription, translation, and regulation), that have
(Fritze and Weihs 2001). The problem of disclosure in the revolutionized approaches to the study of life sciences. It is now
patenting of biotechnologies involving live biological material possible to undertake detailed studies of many difficult and
was solved by enabling patent offices to enforce the requirement more complex organisms that were previously refractory to
that viable (living) samples be deposited in recognized culture genetic analysis. Traditional industrial microbiology merged
collections; the technology can then be made available as a living with recombinant DNA technology, launching the modern bio-
organism to other inventors. technology era in which we have harnessed many cellular func-
Although the patentability of microorganisms is permitted tions for the production of primary and secondary metabolites,
under Article 27 of the TRIPS Agreement, some signatory states bioconversions, and enzymes (Chandler 2008).
do not grant patents to natural microorganisms, arguing that The advent of the modern industrial biotechnology made a
they by themselves are not considered inventions. However, even major impact in the business world through a novel, high-value
in these countries, besides the deposit of genetically modified products such as biopharmaceuticals (recombinant protein
microorganisms, the deposit of natural biological material can drugs, vaccines, and monoclonal antibodies), produced a revolu-
also be required when its description is essential for the repro- tion in agriculture, and markedly increased the market for micro-
ducibility of a patentable process in which they are involved. bial enzymes (Demain 2000). According to O’Hara (1994),
developments in biotechnology since the early 1970s have intro-
duced a new regime of accumulation in the global economy. In
Culture Collections and the Preservation of this context, the patented microorganisms that underpin indus-
Microorganisms trial biotechnology and the information about them have both
become valuable economic resources in their own right.
Over the last decades, the international community has paid A vast number of different biological materials are used in
increasing attention to the adoption of measures to promote industrial biotechnology. The process of identifying an indus-
human development in a sustainable basis. The sustainable use trially useful microbial strain can require great scientific effort
of biological diversity is one of the three objectives set out in for its isolation, selection, investigation, and genetic
Repositories for Patented and Safeguarded Material 12 307
modification into different strains with diverse purposes (Donev a patent, and ensuring that the invention’s benefits are made
2001). Patent protection is often sought for useful microorgan- available to the public. Disclosure encourages innovation and
isms, especially where investments in the genetic engineering of dissemination of knowledge and is intended to balance the
special characteristics have been made. interests of the inventor with the needs of society.
The preservation and maintenance of these useful biological The disclosure of an invention is normally achieved by
materials in culture collections is important for guaranteeing the means of a written description, supplemented when necessary
continuity of the innovation process. Viable and accurately char- by drawings. In the case of inventions involving the use of
acterized biological organisms must be available to allow the biological material, these patent disclosure requirements may
development of reproducible bioprocesses and products. Con- be difficult to fulfill. Industrial property offices in many coun-
trols for the purity and identity of strains are necessary to achieve tries overcome this difficulty by recommending that the written
optimal yields and well-defined products. These are not trivial description of an invention involving the use of a new organism
undertakings, and in many cases, scientists do not possess the should be supplemented by its deposition in a recognized cul-
resources or interest needed for optimal long-term preservation ture collection, in a viable form.
of their material and management of associated data. This is one The territoriality principle of intellectual property rights
reason for the advent of specialized culture collections that act as means that the granting of a patent is ruled by national law in
long-term repositories for large numbers of organisms, stored each territory. Inventors seeking a patent in various countries
under state-of-the-art conditions that ensure strain stability and would face a complex and costly task if they had to deposit
continuity of access (Daniel and Prasad 2010). samples in each country since the recognition of a depository
The preservation of microbial viability, specificity, activity, institution is also independent for each patent office. This has
and immunogenicity is necessary for standardization and indus- been the driver for the rationalization of deposition procedures
trial preparation of high-quality products and also provides at an international level.
a basis for future development (Donev 2001). The wide tempo- The Budapest Treaty was created to implement such recom-
ral gaps that may exist between the isolation and the character- mendations. It was first proposed by the United Kingdom to the
ization of a strain’s useful properties, and its eventual industrial Executive Committee of the Paris Union that WIPO would
application, render long-term depositories essential (Daniel and study the possibilities of creating an international Treaty for
Prasad 2010). The discovery of a heat-stable DNA polymerase the deposit of microorganisms and related material.
enzyme in the thermophilic bacterium Thermus aquaticus illus- A committee of experts was established and after three sessions
trates the importance of the temporal gap. This novel enzyme is (1974, 1975, and 1976) prepared a draft of the Treaty and its
now used to catalyze the polymerase chain reaction (PCR) regulations, which was submitted to a diplomatic conference,
process widely used in molecular biology for amplifying DNA held in Budapest in 1977.
sequences. The PCR process revolutionized biotechnology The Budapest Treaty on the International Recognition of the
research and won the 1993 Nobel Prize for its discoverer, Dr. Deposit of Microorganisms for the Purposes of Patent Proce-
Kary Mullis. While the bacterial species was identified and dure (commonly known as Budapest Treaty) was signed by 18
reported in 1969, the use of its polymerase enzyme in PCR was states on the 28 April 1977 and came into force in August 1980.
patented only in 1990 (Cetus Corporation 1990). Long-term The Treaty permits the deposit of microorganisms at Interna-
storage of T. aquaticus strains was essential for the successful tional Depositary Authorities (IDAs), recognized for the pur-
outcome of this technology during its long development period. poses of patent procedure (Budapest Treaty 1977). As of May
There are many culture collections all over the world for the 2011, 75 countries are party to the Budapest Treaty. The Treaty is
long-term conservation and preservation of microbial diversity. open to States which are party to the Paris Convention and is
These collections are the custodians of genetic resources of vital administered by the World Intellectual Property Organization
importance to science and society (Cameotra 2007). Culture (WIPO 2011).
collections are also needed to store material that can no longer Briefly, the main characteristics of the Treaty are that (1) all
be maintained by research institutions or to duplicate important Contracting States recognize the deposit of a microorganism
reference strains so as to provide simplified user access and for with any IDA and (2) any deposit of a microorganism with an
safety reasons. Culture collections also establish exchange net- IDA shall be accepted for the purposes of patent procedure by
works among themselves and with researchers (Daniel and the patent offices of the Contracting States and by any regional
Prasad 2010). office who filed a declaration of acceptance.
The main feature of the Budapest Treaty is that a Contracting
State must recognize the deposit of a sample with any IDA,
The Budapest Treaty irrespective of whether the IDA is on or outside the territory of
the said State. This allows patent applicants to make one deposit
An important characteristic of the intellectual property system is with an IDA anywhere in the world, rather than having to
that patent laws worldwide require details of an invention to be arrange deposits in each country where patent protection is
fully disclosed in order for others skilled in the relevant field to sought. This procedure benefits patent applicants by reducing
be able to replicate it. The detailed and accurate description of costs and setting up a mechanism to facilitate appropriate third
the invention is a condition for receiving and maintaining party access to deposited biological materials. In addition, the
80 IDA status is acquired after acceptance of a communication

Number of IDAs from the Contracting State to the Director General of WIPO
70 Number of Contracting (Budapest Treaty, 1977—art. 7), a United Nations agency based
States in Switzerland that is responsible for promoting international
60 intellectual property protection.
For a depository institution to qualify for IDA status,
50
WIPO has set some standard guidelines, including that the
IDA should be located within the territory of a Contracting
40
State and it should have a continuous existence with the neces-
sary staff and facilities to perform its scientific and administra-
30
tive tasks under the Treaty. The IDA must be impartial
and objective, which means that it should make its services
20
available on the same terms to any depositor. The IDA can accept
any or specific types of microorganism; > Fig. 12.2 describes the
10
kinds of microorganisms received by different IDA. The Treaty
0
does not define the term ‘‘microorganism’’ thus allowing a broad
interpretation of the term, which includes unicellular and
1981 1986 1991 1996 2001 2006 2011
multicellular organisms, DNA sequences, enzymes, seeds, etc
. Fig. 12.1 (see > Fig. 12.2 and following section for more detail).
Numbers of the recognized International Depositary Authority It should test the viability of biological material promptly
(IDAs) and numbers of Contracting States of the Budapest Treaty after receipt and issue viability statements to the depositor.
since it was in force (WIPO 2011) In addition, the IDA could also perform training, identification,
consultation, and other services, and publish catalogues on
its holdings. The IDA should comply with the requirements of
Treaty increases the security of the depositor because it estab- secrecy and, at the same time, supply samples of deposited
lishes a uniform system of deposit, recognition, and furnishing biological material only to persons entitled to receive them
of samples of microbial samples. according to the regulations of the Budapest Treaty (Budapest
The number of recognized IDAs and Contracting States of Treaty 1977; Fritze and Weihs 2001; Sekar and Kandavel 2004).
the Budapest Treaty are demonstrated in > Fig. 12.1. Currently,
75 countries have signed the Budapest Treaty and 40 culture
collections are recognized as IDAs. The latest institutions to have Geographical Distribution of IDAs
acquired the IDA status were CBA (Australia), recognized in
February 2010; VTTCC (Finland) in August 2010; and MCC Under the Budapest Treaty, Contracting States are not obliged to
(India) in April 2011. Two countries have recently become establish an IDA. Contracting States that decide to establish an
parties to the Budapest Treaty: Morocco and Chile, since July IDA on their territory must provide assurances that the IDA
and August 2011, respectively (WIPO 2011). fulfills Treaty requirements, and they must guarantee its perma-
In addition to the Contracting States, other intergovernmen- nent existence, a supply of necessary equipment, staff, and
tal industrial property organizations, such as the European Pat- knowledge. Of the 75 Contracting States, 22 have at least one
ent Office (EPO), the Eurasian Patent Organization (EAPO), IDA on their territory. Some countries have more than one
and the African Regional Intellectual Property Organization institution with IDA status: the United States of America,
(ARIPO), have made declarations of acceptance of the effects India, Poland, Australia, Japan, Italy; China and Spain, have
of the Treaty. two IDAs each; the Republic of Korea and Russian Federation
As shown in > Fig. 12.1, the numbers of Contracting States have three IDAs each; and the United Kingdom has seven IDAs
and the number of institutions that have obtained IDA status on its territory (for updates, see www.wipo.int/budapest). The
have increased in recent years. It is important to notice that majority of these institutions are concentrated in Europe, with
among Budapest Treaty signatories, only two nations are located 26 recognized IDAs; there are 9 IDAs in Asia, only 3 in North
in South America: Peru and Chile (parties to the Treaty in 2009 America, and 2 in Oceania. However, there are no institutions
and 2011 respectively), and three in Africa: South Africa, with IDA status in South and Central America or on the African
Tunisia, and Morocco (in 1997, 2004 and 2011 respectively). continent.
International Depositary Authorities Definition of Depositable Materials
An IDA can be defined as a scientific institution located on the As noted above, the term ‘‘microorganism’’ is not defined in the
territory of a Contracting State, which accepts deposits of micro- Treaty, which has created considerable confusion. In general,
organisms, and is able to store and furnish all samples in storage. whether a particular deposit is a microorganism or not matters
Hybridomas
4 9 15
4 Embryos
10
Human cells
6 13
Animal cells
6
Seeds
17 Bacteria
Archeae
24
Filamentous Fungi
5
Nematodes
Yeasts
7 Consortium
Anaerobes
30 Phages
13 Algae/microalgae
Genetic material
8 biosafety level > 2
4
Plant virus
8 Animal virus
27 Protozoa
31 1 Plant cells
. Fig. 12.2
Kinds of biological material accepted for patent deposits by the 40 recognized IDAs. The numbers on graph indicate the amount of IDAs
that accept the kind of microorganism listed
less than whether deposition is necessary for the purposes of nematodes, are accepted only by some IDAs (personal commu-
patent disclosure. Any biological material that plays an essential nication from IDAs). > Figure 12.2 represents data collected by
role in a given invention, and without which the invention accessing all 40 IDAs in order to identify the kinds of biological
would not be reproducible, needs to be deposited. For the material they will accept for patent purposes. The numbers of
purposes of patent protection, the term ‘‘microorganism’’ IDAs that are able to accept each kind of biological material is
often applies to diverse biological material including, for exam- also shown in > Fig. 12.2.
ple, viruses, bacteria, actinomycetes, yeasts, filamentous fungi, The technical challenge faced by IDAs and culture collections
mushrooms, protozoa, unicellular algae, cell lines of plants or is to preserve viable life forms for a long period of time (at least
animals, fused cells, transformants and vectors used in genetic 30 years), without causing genetic changes. An important aspect
engineering, variants, plant cells, DNA, and RNA (> Fig. 12.2). of an IDA is to ensure that material in storage is viable and can be
The European Union has decided to discontinue the use of cultured and propagated identically. Not every biological mate-
the term microorganism; instead, it has decided to use the term rial meets the requirements for deposition, as some microorgan-
‘‘biological material,’’ encompassing any material that contains isms are not amenable to preservation methods for long-term
genetic information and is capable of replicating itself or of storage (Fritze and Weihs 2001).
being reproduced in a biological system (Sekar and Kandavel Sekar and Kandavel (2004) have identified some future
2004). necessities for IDAs. In the near future, IDAs should prepare
IDAs vary in the nature of the biological material accepted to deal with the emerging need to store different kinds of
for deposition (> Fig. 12.2). An analysis of the type of microor- biological material, such as microbial consortia (natural or
ganism accepted by the recognized IDAs highlights wide varia- artificial assemblages of interacting microorganisms), or via-
tions in their ability to handle diverse life forms. In general, ble but not yet culturable microorganisms. At present, only
nonpathogenic bacteria and fungi and yeasts and genetic mate- eight IDAs are able to receive and store microbial consortia.
rial (usually plasmids in hosts) are accepted by most IDAs However, mixtures of microbial cultures of more than two
(around 60–75% of IDAs depending on the material), whereas components are usually not accepted, while mixtures of two
cell cultures and hybridomas are accepted by 32–37% of IDAs as components are only accepted if these can easily be distin-
demonstrated on > Fig. 12.2. guished, if the composition of the mixture is defined, and its
About 17–25% of IDAs receive biological material such as components are identifiable and can be preserved in the
algae, anaerobes (strict anaerobes are accepted by even fewer desired ratio (personal communication from IDAs). At pre-
IDAs), plant cells, and animal or plant viruses. Some kinds of sent, no IDA has mentioned the possibility of preserving
biological material, such as embryos, archeae, protozoa, plant viable, but not yet culturable microorganisms for patent
seeds, pathogenic organisms (of biosafety level 2 or higher), or purposes.
Depositor submits sample (e.g. frozen or feeze-dried samples)
If bad
Viability and other examination of the deposited material by the IDA
If good
Preservation of the material through two different methods (if possible);

Sample is returned to the depositor for verification of properties
A certificate of deposit and viability is provided;

A patent deposit number is assigned;
Deposit date is set as the date of receiving the viable material.
Deposit of the patent application at the patent office
. Fig. 12.3
Outline of technical and administrative procedure for the deposition of biological material at an IDA (Adapted from Sekar and Kandavel
(2004))
Patent Deposit and Safe Deposit of Biological The cost for making a deposit under the Budapest Treaty is
Material levied as a single fee at the beginning of the deposit process and
varies depending on the depositary institution, the kind of
The Deposit of Biological Material for Patent biological material deposited, and the details of services ren-
Purposes dered. The fee is valid for the full 30 years of storage.
Depositing a Sample
Main Tests Performed by the IDA
Before sending the biological samples to the chosen institution,
the depositor must come into contact with it to be sure about the After receipt of the biological material, the IDA submits the
form and amount of material to be sent. Each IDA can deter- sample to its main tests for viability and purity. The depositor
mine how it should receive the material to be tested and then may be requested to indicate the most suitable methods of
preserved. cultivation for conducting these tests. If the material received
The technical and administrative procedure for sample depo- is not approved, a new deposit shall be made and new tests of
sition is as follows: (1) A microbial culture arrives at the deposi- viability and purity are conducted (> Fig. 12.3).
tary authority. (2) Its documentation is checked. (3) The viability After the approval of the tests, a receipt confirming reception
and purity of the organism is checked. (4) An accession number is and acceptance of material can be drafted by the institution and
assigned to the culture after viability and purity have been proven. sent to the depositor, who should then include these data on the
(5) The depositor receives official statements of receipt and via- body of the patent document. This receipt must contain informa-
bility which should be used for filing the patent application. (6) In tion about the IDA, the applicant, the biological material identi-
the case of bacteria, fungi, yeasts, and plant cell cultures, the fication references supplied by the applicant, and the filing date
biological material is usually subcultured and a stock of samples and serial number assigned to the deposit by the IDA (Budapest
is—as applicable—preserved by storage in liquid nitrogen and/or Treaty 1977—Rule 7). If the scientific description and/or taxo-
freeze-drying. All other biological material will usually not be nomic designation of the material have not been listed on the
subcultured, but stored in liquid nitrogen or in freezers at around deposit, the depositor may indicate them later by means of
80 C in the same form as it was received from the depositor. written communication (Budapest Treaty 1977—Rule 8).
(7) Viability of the stored cultures is inspected regularly during the IDAs are obliged to promptly test the viability of each bio-
minimum storage period of at least 30 years (Fritze and Weihs logical material deposited with it and also at reasonable inter-
2001; Sekar and Kandavel 2004, > Fig. 12.3). vals, depending on the kind of material and its storage
. Table 12.1
Examples of biological material preservation methods (Weihs 2010)
Genetic
Method Successfully preserved organisms Shelf life stability
Storage under paraffin oil Yeasts, fungi, some bacteria Fungi: 5–20 years Low
Bacteria: 2–5 years
Storage in distilled water Yeasts, filamentous fungi, actinomycetes, not 1–5 years low
enterobacteria
Drying in gelatin discs Enterobacteria, staphylococci, pseudomonas, 1/2–7 years Medium
spore-forming fungi
Storage in sterile soil, sand etc. Spore-forming bacteria and fungi 10–15 years Good
Non-spore formers 1–5 years
L-drying Bacteria, fungi, yeasts, animal viruses, protozoa 2–5 years Good
Drying on glass beads or porcelain rings Fungi, bacteria, mycoplasms 5–10 years Good
Sporeformers: 10–15
years
Freeze-drying Bacteria, some yeasts >40 years Good
Storage at domestic refrigerator Bacteria Several weeks/ Low
temperature months
Storage at 20 C in glycerol Bacteria Several months–2 Medium
years
Storage at 60 C to 80 C in glycerol Bacteria 5 year Good
Storage in liquid nitrogen at 196 C Bacteria, fungi, yeasts, plant cell cultures, animal cell >30 years Good
cultures
Maintenance on glass beads at 60 C to Bacteria >10 years Good
80 C
conditions, or at any time, if necessary for technical reasons and institutions conserve materials with at least two different pres-
at any time, upon the request of the depositor (Budapest Treaty ervation methods. > Table 12.1 shows some of the methods used
1977—Rule 10). to preserve a range of biological materials and the resulting shelf
life and genetic stability.
Methods of Preservation for Biological Materials

When Can an IDA Refuse a Sample?
Preservation methods can vary between different institutions,
since each of these has the autonomy to decide which procedure In some cases an IDA can refuse to receive a depositor’s biological
it will follow in its daily work routine. Moreover, the intrinsic material. This can take place (1) when the biological material is
characteristics of each type of biological material to be preserved not part of the IDA’s scope of work, and it cannot guarantee
will also influence the choice of the most appropriate conserva- correct storage and handling; (2) when the properties of the
tion method. biological material are so exceptional that the IDA is technically
Regardless of the preservation method adopted, institutions not in a position to perform tasks in relation to it; and (3) when
should maintain the biological material genetically stable and the deposit is received in a condition which clearly indicates that
without contamination. For the reduction of cellular metabo- the material is missing, or which precludes its acceptance for
lism and concomitant maintenance of material viability, the scientific reasons (Budapest Treaty 1977—Rule 6). The deposit
preservation methods most commonly employed are cryogenic is not considered valid in such cases. In the first two situations,
storage, for example, through storage in liquid nitrogen at a new deposit shall be held at another institution that can
196 C, and deprivation of water in the cell through the provide guarantees on the material deposited. In the latter
process of lyophilization. These methods combine long storage situation, and at the discretion of the depositor, a new shipment
times with good genetic stability. of the material can be sent to the same institution (receiving
To ensure the safety of biological material in case of a new filing date, see > Fig. 12.3) or to another institution that
unforeseen problems with a preservation method, most can also provide guarantees on the material deposited.
Filing of the Patent Publication of the Patent Grant

Application Patent Application
Patent Procedure Timeline

Available for:
Patent Office Patent Office
Depositor Depositor
Authorized Third Party Authorized Third Party
+
Certified Third Parties
. Fig. 12.4
Provision of biological material samples under Budapest Treaty rules (Adapted from Weihs (2010))
Dispatch of the Sample

Rule 11.1
600 Rule 11.2 3000
The rules for sending samples of biological material should
Rule 11.3
comply with the rules established by each country. Rule 11 of
the Budapest Treaty restricts the transmission of biological 500 Total Deposits 2500
material samples to signatories to the Treaty or to countries
that adopt Treaty rules despite not being signatories. 400 2000
According to the Budapest Treaty, samples of material can be
sent to: 300 1500
– Interested Industrial Property Offices at any time on request
to the IDA (Rule 11.1) 200 1000
– The Depositor, or with the authorization of the depositor, to
authorized third parties at any time on request to the IDA
100 500
(Rule 11.2)
– Parties legally entitled by confirmation of the request by the
responsible patent office (certified third parties) after the 0 0
2001 2002 2003 2004 2005 2006 2007 2008 2009
publication of the patent deposit (Rule 11.3)
. Fig. 12.5
In respect of patents granted and published by an industrial Samples dispatched by IDAs according to Rule 11 of the Budapest
property office, that office may from time to time communicate Treaty (left axis—bar graph) and the total number of deposits
to any IDA a list of the deposit accession numbers it has assigned received by IDAs under the Budapest Treaty (right axis—line
to biological material referred to in specific patents it has issued. graph) (WIPO 2011)
The IDA shall, under request, furnish a sample of any biological
material with a listed accession number. For deposited material
whose accession numbers have been communicated, the indus-
trial property office shall not be required to provide certifica- started to increase again. In 2009 about 2,500 deposits of bio-
tions (Rule 11.3b). > Figure 12.4 shows a scheme to facilitate an logical material were performed under the Budapest Treaty.
understanding of how biological samples are provided under the > Figure 12.5 also shows the numbers of cultures furnished
Budapest Treaty. under the Rule 11 of the Budapest Treaty (bar graphs). Remark-
Under the Treaty rules, sending samples to the patent offices ably, independently of the year, very few samples of biological
is free of charge, while sending samples to the depositor, an material were dispatched to patent offices (Rule 11.1) while
authorized third party, or certified third parties is accomplished about 300–500 cultures per year were furnished to depositors,
through the payment of fees prescribed by each IDA. authorized third parties (Rule 11.2) or legally entitled parties
Once an IDA has furnished a sample to any interested party, (Rule 11.3).
other than the depositor, it shall promptly notify the depositor > Table 12.2 represents the ten major IDAs, the number of
in writing of that fact, as well as of the date on which the said deposits accepted, and the number of samples furnished under
sample was furnished, and of the name and address of the Rule 11 of the Treaty Regulations from 2001 to 2009.
industrial property office, of the authorized party, of the certi-
fied party, or of the requesting party, to whom or to which the
sample was furnished. Length of the Deposit
The line graph in > Fig. 12.5 represents the total number of
deposits of biological material for patent purposes on all IDAs According to Rule 9 of the Treaty, any biological material depos-
from 2001 to 2009. The deposits dropped around 2003 and then ited with an IDA shall be stored by that authority, with all the
. Table 12.2 Concluding Remarks

Numbers of deposits of biological material for patent purposes
and samples furnished under Rule 11 of the regulations of the Biotechnology processes are increasingly used for industrial
Budapest Treaty by the major International Depositary Authori- production despite great advances in chemical routes to synthe-
ties (2001–2009) (WIPO 2011) size many products of interest. This is mainly due to the ability
of microorganisms to naturally metabolize a wide range of
International depositary Deposited Furnished
compounds which are still too complex for synthetic chemical
authority (country) samples samples
routes. Genetic engineering has given even greater control on
ATCC (United States of America) 7,302 6,721 our ability to harness the use of microbes and cell lines for
CGMCC (China) 3,009 96 industrial production. Biotechnology therefore remains a very
IPOD (Japan) 2,439 664 promising field for technological and economic development,
DSMZ (Deutschland) 2,211 720 and this can be seen from the increasing number of patent
CCTCC (China) 1,793 52 applications that refer to biochemistry and genetic engineering
discoveries. The preservation of biological material in interna-
KCTC (Republic of Korea) 1,727 152
tionally recognized culture collections plays a key role
CNCM (France) 1,701 462
supporting biotechnology-related patents. The main advantages
NRRL (United States of America) 932 3,075 of the Budapest Treaty include the simplification and cost reduc-
KCCM (Republic of Korea) 822 71 tion of patent procedures, the prevention of certain risks in the
NCIMB (United Kingdom) 602 115 field of biotechnology, enhanced research, and development
through access to deposited biological material and promotion
of cooperation and exchange between IDAs.
Even so, the necessity for physical deposits of biological
care necessary to keep it viable and uncontaminated, for a period materials for patent purposes is being increasingly questioned.
of at least 5 years after the most recent request for the furnishing This is due to the idea that DNA sequences, submitted in the
of a sample of the deposited material and, in any case, for form of computer data only, might already be comprehensive
a period of at least 30 years after the date of the initial deposit. enough to fulfill the needs for full disclosure of an invention and
Regarding secrecy aspects, no IDA shall give information to be able to substitute for the deposition of complete and viable
anyone about whether a biological material has been deposited biological material. At present most applicants still prefer to
with it under the Budapest Treaty. Furthermore, it shall not deposit as comprehensively as possible to be sure that the inven-
disclose to anyone information concerning material deposited tion is completely disclosed and rendered reworkable, so that it
with it under the Treaty except to an authority or a party legally will not be challengeable by competitors (Fritze and Weihs
entitled to obtain a sample of a material according to Rule 11 of 2001).
the Treaty. The deposit of biological material for patent purpose
requires optimal conditions to guarantee long-term preserva-
tion. Due to its needs of specific expertise and resources,
Safe Deposit of Biological Material a culture collection is not always able to maintain a wide range
of samples, and for this reason, sometimes it can be beneficial for
Cell cultures, microorganisms, and other biological materials are them to operate in a coordinated network to achieve a broader
valuable assets for organizations, and the loss of these materials preservation activity.
can be costly. Several culture collections around the world pro- Several initiatives have been proposed by developing coun-
vide safe deposit services for corporations, government labora- tries to achieve the necessary infrastructure to set up an IDA,
tories, academic institutions, and others. which could intensify international coordination and enhance
A culture collection that offers safe deposit services must the preservation of a country or region’s natural gene pool.
ensure security and confidentiality of the material deposited. Although a depositor can deposit a culture in any IDA belonging
Preservation methods are defined for each collection and can to any country, the presence of an IDA in the country of patent
vary depending on the type of material to be preserved. In safe origin may result in a more affordable deposition process, par-
deposits, the depositor can determine how long the material will ticularly in developing countries, and build up important local
remain preserved in the collection, unlike for the deposit of expertise (Sekar and Kandavel 2004).
biological material for patent purposes. Culture collections, including suitably resourced IDAs, play
In safe deposits all rights to cultures remain with the depos- a role in promoting and disseminating the use of biological
itor and all information concerning deposited material is materials with important characteristics, by providing con-
retained in confidence. Also, culture material is available only trolled storage conditions and the supervision of specialized
to the depositor or an individual designated in writing by the professionals. With the rapid loss of habitat, it may become
depositor. At some culture collections, vials of each culture can important for IDAs established in developing or emerging econ-
be returned to the depositor for analysis after a storage period to omies to specialize not just in the storage of patented organisms
ensure the viability of material received (ATCC 2011). but in the preservation of their regional biodiversity.
The establishment of well-structured culture collections to pre- Demain AL (2000) Small bugs, big business: the economic power of the microbe.
Biotechnol Adv 18:499–514
serve the biodiversity of megadiverse countries would appear to
Donev T (2001) Methods for conservation of industrial microorganisms. Part 1:
be a crucial question; especially where a vast amount of material, Methods for conservation of industrial microorganisms. National Bank
with great potential for biologically based products and pro- for Industrial Microorganisms and Cell Cultures, Bulgaria. Available at
cesses to benefit human welfare, has not yet been isolated and http://www.nbimcc.org/bg/Conservation%20methods.pdf. Accessed 29
identified. May 2011
Fritze D, Weihs V (2001) Deposition of biological material for patent protection
in biotechnology. Appl Microbiol Biotechnol 57:443–450
Kalanje CM (2005) Role of intellectual property in innovation and new
References product development. Available at http://wipo.int/sme/en/documents/pdf/
ip_innovation_development.pdf. Accessed 29 May 2011
ATCC (American Type Culture Collection) (2011) Available at http://www.atcc. O’Hara PA (1994) An institutionalist review of long wave theories:
org/DepositServices/SafeDepository/tabid/238/Default.aspx. Accessed 10 Schumpeterian innovation, modes of regulation, and social structures of
May 2011 accumulation. J Econ Issues 2:489–500
Budapest treaty on the international recognition of the deposit of microorgan- Paris convention for the protection of industrial property. Done at 20 March 1883
isms for the purposes of patent procedure: done at 28 Apr 1977 and amended and amended on 28 Sept 1979. (Paris). WIPO. Available at http://www.wipo.
on 26 Sept 1980, and regulations (as in force since 31 Jan 1981). (Budapest): int/treaties/en/ip/paris/trtdocs_wo020.html. Accessed 29 May 2011
WIPO. Available at http://www.wipo.int/export/sites/www/treaties/en/regis- Sekar S, Kandavel D (2004) The future of patent deposition of microorganisms?
tration/budapest/pdf/trtdocs_wo002.pdf. Accessed 10 May 2011 Trends Biotechnol 22:213–218
Cameotra SS (2007) Preservation of microorganisms as deposits for patent TRIPS – Agreement on trade-related aspects of intellectual property rights. Done
applications. Biochem Biophys Res Commun 353:849–850 at 15 Apr 1994 (Morocco). WTO. Available at http://www.wto.org/english/
Cetus Corporation, Mullis KB, Erlich HA, Gelfand DH, Horn G, Saiki RK tratop_e/trips_e/t_agm0_e.htm. Accessed 29 May 2011
(1990) Process for amplifying, detecting, and/or cloning nucleic acid Weihs V (2010) IDAs -30 years of experiences world-wide. 12th international
sequences using a thermostable enzyme. US4965188. 29 Oct 1990 conference on culture collection, Florianópolis, SC, Brazil. Available at
Chandler M (2008) Microbiology – what now? Res Microbiol 159:51–58 http://www.iccc12.info/presentations/vweihs.pdf. Accessed 10 May 2011
Convention on biological diversity. Done at 1992 (Rio de Janeiro). United WIPO (2004) Intellectual property handbook: policy, law and use. WIPO publi-
Nations. Available at http://www.cbd.int/convention/text/. Accessed 20 cation no. 489. Available at http://www.wipo.int/about-ip/en/iprm/.
May 2011 Accessed 10 May 2011
Daniel H-M, Prasad GS (2010) The role of culture collections as an interface between WIPO (World Intellectual Property Organization) (2011) Available at http://
providers and users: the example of yeasts. Res Microbiol 161:488–496 www.wipo.int/treaties/en/registration/budapest/. Accessed 10 May 2011
13 Biotechnology and Applied Microbiology
Eugene Rosenberg
Department of Molecular Microbiology and Biotechnology, Tel Aviv University, Tel Aviv, Israel
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 Introduction
The Fermentation Controversy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 There is no agreed upon definition of biotechnology. Definitions
vary from the most noble—the uses of biology for the benefit of
Classical Applied Microbiology (1900–1940) . . . . . . . . . . . . . . 317 man—to the most prosaic—the use of biology to make money.
A balanced definition might be the use of living organisms (or
Isolation of Cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 materials from living organisms) to perform defined processes for
industrial, agricultural, and health applications. Applied micro-
Screening Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 biology is a large part of biotechnology because microorganisms
have a broadly diverse synthetic and degradative potential, which
Strain Improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 is easily manipulated genetically, and they can readily grow under
controlled conditions in large fermenters. The so-called ‘‘new
Biochemical Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 biotechnology’’ generally refers to the use of recombinant
DNA technology to construct useful new strains. However,
The Antibiotic Era . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 a considerable amount of ‘‘new biotechnology’’ does not use
The Discovery of Penicillin and Aftermath . . . . . . . . . . . . . 319 recombinant DNA technology but rather structural biology as a
Limitations of Penicillin and the Development basis for rational drug design, monoclonal antibodies as
of Other Antibiotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 biosensors, and enzymes from extremophiles to facilitate
The Role of Antibiotic Research in the Development industrial production. In addition, conventional industrial
of Microbial Biochemistry, Molecular Genetics microbiological processes apply many of the new biotechnologies.
and Biotechnology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 This edition of the Prokaryotes contains an entire section
dealing with the following specific biotechnology and applied
The New Biotechnology: Applications of microbiology topics: bioremediation, microbial biofilms,
Genetic Engineering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322 biodeterioration, bacterial enzymes, bacteria in food and
beverage production, organic acid and solvent production,
Applied Environmental Microbiology: Bioremediation . . . 323 amino acid and vitamin production, recombinant DNA protein
Petroleum Pollution and Bioremediation . . . . . . . . . . . . . . . 323 production in bacteria, bacterial pharmaceutical products,
The Use of GEMs in Bioremediation . . . . . . . . . . . . . . . . . . . . 324 biofuels, biosurfactants, and repositories for patented materials.
In addition, each chapter dealing with a particular genus
Health-Related Applied Microbiology . . . . . . . . . . . . . . . . . . . . . 325 (or a group of genera) has a section on applications. In this
Drug Discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 introductory chapter, I attempt to put applied microbiology and
Diagnostic Microbiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 biotechnology into a historical perspective and give an overview
of some of the general principles related to the biotechnology of
Food Microbiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 prokaryotes. It should be emphasized at the outset that both
applied and basic microbiology have been and always will be
Biosafety and Legal Protection in Biotechnology . . . . . . . . . . 326 intertwined. Applied microbiology without basic microbiology
is lame and limited, and a major driving force of basic microbi-
Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 ology is the solution of applied problems.
Environmental Microbiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
Microbial Associations with Animals, Plants,
and Humans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 The Fermentation Controversy
Microbial Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
For thousands of years, microbiological processes (such as those
for producing wine, beer, vinegar, bread, pickles, sauerkraut,
and various milk products such as yogurts, cheeses, and butter)
were developed by trial and error, without any understanding
of the underlying principles. During the last half of the
nineteenth century, the controversy regarding the causative
316 13 Biotechnology and Applied Microbiology
agent of fermentation stimulated experimentation that serves Liebig’s views were widely accepted, partly because of his
as the scientific basis of today’s applied microbiology and powerful influence in the scientific world and partly because of
biochemistry. a desire among chemists to avoid conceding an important
For centuries fermentation had a significance that was chemical change to the domain of biology.
almost equivalent to what we would now call a chemical reac- In 1857, Pasteur published his first paper in the field of
tion, an error that probably arose from the vigorous bubbling fermentation. The publication dealt with lactic acid fermenta-
seen during the process. The conviction that fermentation was tion, not alcoholic fermentation. Utilizing the finest micro-
strictly a chemical event gained further support during the early scopes of the time, Pasteur discovered that souring of milk was
part of the nineteenth century. French chemists led by Lavoisier correlated with the growth of a microorganism, but one consid-
and Gay-Lussac determined that alcoholic fermentation could erably smaller than the beer yeast. During the next few years
be described as the chemical conversion of one molecule of Pasteur extended these studies to other fermentative processes,
glucose (C6H12O6) into two molecules of ethanol (C2H5OH) such as the formation of butyric acid and lactic acid. In each case
and two molecules of CO2. It was of course known that yeast he was able to demonstrate the involvement of a specific and
must be added to the glucose solution to ensure a reproducible characteristic microorganism; alcoholic fermentation was
and rapid fermentation. The function of the yeast, according to always accompanied by yeasts, lactic acid fermentation by
the chemists, was merely to act as a chemical catalyst. Then, in nonmotile bacteria, and the butyric acid fermentation by motile,
1837, Theodor Schwann and Charles Cagniard-Latour indepen- rod-shaped bacteria. Thus, Pasteur not only disposed of one of
dently published studies that indicated yeast was a living micro- the opposition’s strongest arguments, but also provided strong
organism. Prior to their publication, yeast was considered circumstantial evidence for the biological theory of fermenta-
merely as a proteinaceous chemical substance. One reason it tion. Pasteur then attacked the crucial problem, alcoholic
was difficult to ascertain whether yeasts were alive was that, fermentation. Liebig had argued that the proteinaceous
like most other fungi, they are not motile. The cellular nature material, released from yeast during its decomposition,
of yeast was discovered only when improved microscopes catalyzed the splitting of sugar. Pasteur countered this argument
became available, and it is most likely that improved micro- by developing a protein-free medium for the growth of yeast. He
scopes enabled the two workers to come up with the same found that yeast could grow in a medium composed of glucose,
observation at approximately the same time. Schwann and ammonium salts, and some incinerated yeast. If this medium
Cagniard-Latour also observed that alcoholic fermentation was kept sterile, neither growth nor fermentation took place. As
always began with the first appearance of yeast, progressed soon as the medium was inoculated with even a trace of yeast,
only with their multiplication, and ceased as soon as their growth commenced, fermentation ensued, and the quantity of
growth stopped, so both scientists concluded that alcohol is alcohol produced paralleled the multiplication of the yeast.
a by-product of yeast growth. In this protein-free medium, Pasteur was able to show that
The biological theory of fermentation advanced by fermentation took place without the decomposition of yeast.
Cagniard-Latour and Schwann was immediately attacked by In fact, the yeast synthesized protein at the expense of the sugar
the leading chemists of the time. The eminent Swedish physical and ammonium salts. Thus, Pasteur concluded in 1860 that
chemist, Jons Jakob Berzelius, reviewed the papers of ‘‘Fermentation is a biological process, it is the subvisible
Cagniard-Latour and Schwann in his Jahresbericht for 1839 organisms which cause the changes in the fermentation process.
and dismissed microscopic evidence as of no value in what What’s more, there are different kinds of microbes for each kind
was obviously a purely chemical problem. According to of fermentation. I am of the opinion that alcoholic fermentation
Berzelius, nothing was living in yeast. ‘‘It was only a chemical never occurs without simultaneous organization, development
substance which precipitated during the fermentation of and multiplication of cells, or continued life of the cells already
beer and which had the usual shape of a non-crystalline formed. The results expressed in this memoir seem to me to be
precipitate.’’ In addition, Liebig published a paper containing completely opposed to the opinion of Liebig and Berzeliu.’’
several important arguments against the biological theory of (Dubos 1950).
fermentation. Liebig’s two major points can be summarized Pasteur argued effectively and, more importantly, all the data
as follows: were on his side. Thus, the so-called vitalistic theory of fermen-
tation predominated until 1897, when an accidental discovery
1. Certain types of fermentation such as the lactic acid (souring by Eduard Buchner finally resolved the controversy and threw
of milk) and acetic acid (formation of vinegar) can occur in open the door to modern catalytic biochemistry based on
the complete absence of yeast. enzymes.
2. It is not necessary to conclude that, even if yeast were Buchner was attempting to obtain from yeast an extract that
a living organism, fermentation is a biological process. might have medicinal value. After several unsuccessful trials, he
Yeast is a remarkably unstable substance, which, as discovered that mixing yeast cells with fine sand and grinding
a consequence of its own death and decomposition, catalyzes the mixture in a mortar and pestle disrupted the cells. After
the splitting of sugar. Thus, fermentation is essentially filtering the mash to remove the sand and any unbroken cells,
a chemical change catalyzed by breakdown products of a clear yeast juice was obtained. The juice, however, soon became
the yeast. contaminated with bacterial growth. Since the extract was for
Biotechnology and Applied Microbiology 13 317
human consumption, Buchner could not utilize ordinary at 70 C. Also, the preferred inoculum would be a soil sample
antiseptics to prevent the spoilage. Therefore, he attempted to taken from a high temperature and pH site that contains
preserve the yeast extract by adding large quantities of sugar. The decaying plant material. After several transfers in the same
yeast extract began to bubble soon after the sugar was added. medium, the mixed culture is streaked onto an agar medium
Careful analysis revealed that the sugar was decomposing to containing the same nutrients. The resulting colonies of micro-
carbon dioxide and ethyl alcohol. Fermentation had proceeded organisms must contain an extracellular xylanase that is active at
in the absence of living cells. Buchner’s achievement inaugurated 70 C and pH 9.0. Enrichment cultures of this type have been
a new era in the study of alcoholic fermentation and other successfully employed to obtain a wide variety of useful strains.
metabolic processes. Reactions that normally took place only The power of the enrichment culture technique depends on the
in vivo could now be studied in vitro. The agents which are enormous metabolic diversity of microorganisms and on
present in cell extracts and which catalyze these reactions were the ingenuity of the investigator to establish proper restrictive
called enzymes, from the Greek ‘‘en zyme,’’ meaning ‘‘in yeast.’’ growth conditions.
In 1907, Eduard Buchner received the Nobel Prize in Chemistry
‘‘for his biochemical researches and his discovery of cell-less
fermentation.’’ Screening Procedures
The enrichment culture technique is valuable only when growth
Classical Applied Microbiology (1900–1940) conditions select for the desired strain. The selection is usually
based on the degradation of a particular substrate or the
After it was demonstrated that living yeast was the causative resistance to a particular environmental parameter, for example,
agent for the alcohol fermentation, rapid progress was made in high and low temperature, salt and pH values, antibiotics and
isolating and identifying the microorganisms responsible for toxic metals. However, it is generally not possible to select
other fermentations. As a result, considerable practical knowl- directly for microorganisms that produce or overproduce useful
edge was gained, commercial media developed, producer strains chemicals, such as fermentation products, antibiotics, and other
improved, and the fermentation industry grew and diversified drugs. In such cases, screening procedures have to be employed.
(> Table 13.1). As an example, commercial bioproduction of The success of the Weizmann process for producing acetone and
citric acid was begun in New York in 1923, and by 1940 over 10 butanol depended on screening a large number of microorgan-
million kilograms were produced per annum. Much of this early isms for a strain (C. acetobutylicum, Weizmann) that produced
work on industrial fermentations has been reviewed by Prescott high levels of these solvents. In recent years, many new screening
and Dunn (1940). procedures have been developed and automated for the discov-
With the advent of World War I, it became necessary to ery of new products (White et al. 1986). Screening procedures
develop a fermentation method for making acetone, which is are based on either chemical or functional assays, using intact
used for the manufacture of explosives. The US government cells or subcellular preparations. An example of a functional
purchased two large distilleries in Indiana and established the assay that uses intact cells is antibacterial screening by the agar
Weizmann process, which made acetone from maize by plate diffusion assay (Gerhardt 1981). The lambda prophage
Clostridium acetobutylicum fermentation. After the war, the induction assay for screening anticancer drugs is an example of
Weizmann process was used mainly to produce butyl alcohol an assay that uses subcellular preparations (Price et al. 1964). In
needed in the manufacture of automobile lacquers. During the recent years, high-throughput screening (HTS) using robotics,
development of the Weizmann and other industrial fermenta- data processing and control software, liquid handling devices,
tion processes, several important and general aspects of applied and sensitive detectors has been applied for drug discovery
microbiology became established. (Zhang 2011).
Isolation of Cultures Strain Improvement
Isolation of a pure culture of the desired microorganism is often Microorganisms, freshly isolated from nature, have highly reg-
the first step in an applied microbiology project. Selective ulated metabolic systems designed to prevent the
media and growth conditions developed for the isolation and overproduction of biochemicals. Genetically altering the strain
cultivation of specific groups of bacteria and archaea are so that it overproduces a desired product is an important step in
described throughout this handbook. In addition, microorgan- industrial microbiology. Three such general procedures have
isms that carry out a specific selectable function can be isolated been used successfully: mutation, genetic recombination, and
by enrichment culture procedures (Kreig 1981; Veldamp 1970; gene cloning. Each technique has its advantages and disadvan-
Jones and Krieg 1984). For example, if the applied goal is to tages. Often two or three of these techniques can be used in
obtain a xylanase that is active at pH 9.0 and 70 C, then an tandem to obtain a stable improved strain. Success in bringing
enrichment culture that contains xylan as the sole carbon and a fermentation product to market and consequently competing
energy source would be adjusted to pH 9.0 and incubated in that market depends on continuous strain improvement
. Table 13.1
Production of industrial chemicals by fermentationa
Fermentation Microorganism Commercial use

Ethyl alcohol Saccharomyces cerevisiae Industrial alcohol
Acetone-butanol Clostridium acetobutylicum Synthetic rubber, explosives, lacquers, and solvents
Acetone-ethanol Bacillus acetolyticus Solvents
Acetic acid Acetobacter Vinegar
Lactic acid Lactobacilli, Streptococci Textile and leather industries
Propionic acid Propionic acid bacteria Solvent
Citric acid Aspergillus niger Medicines, food and beverages
Gluconic acid Penicillium chrysogenum Pharmaceutical industry, cleaning
Gallic acid Penicillium glacum Dye industry
Fumaric acid Rhizopus Food and beverages
Mannitol White Asperg Resin production, food and beverages
Dihydroxy acetone Acetobacter Artificial tanning and chemical industry
a
See Chap. 1, ’’Organic Acid and Solvent Production‘‘ in Vol. 4
. Table 13.2
Strain improvement in penicillin production
Year Penicillin strain Origin Yield (g/l)

1929 Penicillium (Fleming) Chance contamination 0.01
1941 NRRL-832 Isolated in Belgium 0.04
1943 NRRL-1951 Isolated from a melon 0.15
1944 X-1612 X-ray mutant of NRRL 1951 0.30
1945 Q-176 UV mutant of X-1612 0.55
1949 49–133 Spontaneous mutants of Q-176a 1.2
1990 Commercial strains Nitrogen mustard mutants of 49–133a >7.0
a
Several steps of mutation and selection were used to obtain this overproducing strain
programs. Screening and selection methods for strain improve- improvement program is that today it is possible to produce
ment have been reviewed by Elander (1966), Aharonowitz and enough penicillin at a low cost to treat anyone who needs the
Cohen (1981), Queener and Lively (1986), Silverman et al. antibiotic, whereas before only a few serious cases could be
(1998), and Demain and Adrio (2008). treated at a high cost.
Historically, mutant screening was the first systematic Genetic recombination of advantageous mutations from
method to improve industrial strains. The lineage of several mutant strains is a useful procedure for strain improve-
strain improvement for penicillin production is shown in ment. It allows one to combine advantageous mutations from
> Table 13.2. After an exhaustive screening of natural different sources. Also, genetic recombination makes possible
Penicillium strains, one isolate (NRRL-1951) was obtained that the removal of deleterious secondary mutations. For example,
produced ten times more penicillin than the initial isolate of when a culture is mutagenized and then a bacterium is selected,
Fleming. Starting in 1944, strain NRRL-1951 was used as the which overproduces the desired product, the bacterium
parent strain for a program of mutation and selection by several may also contain mutations that interfere with growth. By
groups of investigators. The results were remarkable: strain backcrossing the mutant strain into the wild-type, it is possible
X-1612, an x-ray-induced mutant of NRRL-1951, produced to screen for strains that contain the useful mutation without
0.3 g of penicillin per liter; strain Q-176, an ultraviolet the deleterious one. This procedure of ‘‘cleaning up’’ the strains
light-induced mutant of X-1612, yielded 0.55 g per liter; and is particularly important when multiple mutation steps are
descendants of strain Q-176 (e.g., strain 49–133) in turn employed. The concept of genetic recombination for strain
produced 1.2 g; and currently used commercial strains improvement has also been applied to natural strains (Zhang
yield more than 7 g per liter. The significance of this strain et al. 2002).
Gene manipulation is the third and most recent technique of product formation, growth rates, growth yields, changes in pH
strain improvement. This method requires a good understand- values during growth, and nutrient requirements. Since the cost
ing of the molecular genetics and biochemical pathway that is of producing a fermentation product is often influenced largely
involved in the biosynthesis of the desired product. Gene manip- by the costs of the nutrient feed stocks, it is important to
ulation can be used to overcome rate-limiting reactions by examine inexpensive commercial sources of nutrients, such as
increasing the production of specific enzymes. This can be corn syrup and casein hydrolysates, even in the flask experi-
obtained by cloning the gene and increasing its copy number, ments. Small fermenters (1–50 l) are used to determine the
by altering promoter strength and ribosome-binding sites, and optimum biochemical engineering parameters that are required
eliminating undesirable properties such as product inhibition. in large-scale industrial fermentations, for example, oxygen
In addition, gene manipulation can be used to generate new demand. Many economic evaluations cannot be performed
products by combining genes from different microorganisms. accurately using laboratory equipment, and therefore pilot
This latter technique—combinatorial genetics—has been used plants must sometimes be built. Scale-up of fermentations has
successfully to produce new macrolide antibiotics (Hutchinson been reviewed by Trilli (1981) and Reisman (1993).
and Fujii 1995). The series of steps used to concentrate and purify the desired
product is referred to as the downstream process. Each down-
stream process is tailored to fit the specific properties of the
Biochemical Engineering desired product. The first step is usually the separation of the
cells from the broth by continuous centrifugation or hollow fiber
To commercialize the production of fermentation products by filtration. If the product is in the cells, then the cells are
microorganisms, it was necessary to scale up productions from disrupted and the product purified by specific biochemical
flasks to large fermentation tanks. This requires a combination procedures. If the product is in the cell-free broth, then the
of microbiological and engineering skills. The first problem that next step is concentration of the product by removal of
had to be overcome was the need for sterilization and prevention the water. Depending on the molecular size and charge of the
of contamination by undesirable microorganisms. The funda- product, the water can be removed by ultrafiltration, precipita-
mental principles of sterilization, developed by Pasteur and tion, or distillation of the product. Again, the final purification
others, had to be modified considerably when going from flasks of the fermentation product depends on its specific biochemical
and Petri dishes to very large stainless steel tanks. Not only did properties. Typical procedures include liquid-liquid extractions,
engineering techniques have to be applied to sterilize the adsorption chromatography, ion-exchange chromatography,
medium, but techniques had to be developed for the sterile and gel filtration.
introduction of the inoculum, maintenance of pH and temper-
ature, and the introduction of large quantities of oxygen (air)
under sterile conditions. The Antibiotic Era
Understanding and controlling a fermentation process
depends on the data obtained from biosensors and instrumen- See Chap. 9, ‘‘Bacterial Pharmaceutical Products’’ in Vol. 4.
tation. The most important parameters to measure and control
are pH, temperature, and concentrations of dissolved oxygen
and substrate. If these parameters are not maintained within The Discovery of Penicillin and Aftermath
a narrow range, the synthesis of the desired product will decline
and in the worst case, the culture will die. Each of these param- The discovery and development of antibiotics is one of the
eters must be measured online. In modern fermenters, the greatest scientific achievements of the twentieth century. Like
information obtained by the sensors is fed into computers that many events of major historical significance, the contributing
then automatically adjust the conditions of the fermentation factors were varied and complex. Microbiologists, chemists,
broth to a predetermined value. Dissolved oxygen can be regu- fermentation engineers, medical doctors, businessmen, indus-
lated by increasing air flow into the broth or agitation speed of trialists, lawyers, and government officials played major roles.
the rotors, pH can be adjusted by pumping in acid or base, and The motivating forces ranged from pure scientific curiosity and
temperature can be controlled by heating or cooling with circu- the desire to alleviate suffering to economic gain and the
lating cold water. Additional parameters of biological signifi- pressures of the Second World War.
cance that can be measured include culture turbidity, exit O2 The story, which has been told many times, begins in 1928
and CO2, and product concentration. Instrumentation for mea- with a fortuitous observation by Alexander Fleming, a
suring and regulating fermentation parameters and for bacteriologist working in St. Mary’s Hospital in London.
computer control of microbial processes has been reviewed by Fleming was growing a disease-causing staphylococcus in Petri
Wang (1981) and Guerreiro et al. (1997). dishes containing nutrient agar. An airborne fungal spore fell
Scale-up of fermentation processes generally proceeds from inadvertently onto the agar and began to multiply. As the con-
flasks to small fermenters to large fermenters (Gaden 1981). taminating mold grew, a halo, or clear area, developed around
Certain physiological parameters can be determined in flask the mold colony. Such accidents must surely have happened to
experiments, such as optimum temperature for growth and hundreds of bacteriologists before Fleming, but the ruined agar
plates were simply discarded in disgust. Fleming, however, more importantly, the research staff was experienced in mold
realized the significance of the clear area around the mold fermentations. After a few days of intense discussions, Florey
colony. The fungus must have secreted something into the returned to England, leaving his trusted assistant, Norman
medium that inhibited the growth of the staphylococci. He Heatley, to teach the Americans what was then known about
therefore isolated the fungus and repeated the experiment, this growing Penicillium and measuring the quantity of penicillin
time intentionally adding the fungus to the bacterial culture. produced. Progress was rapid. As described above, strain
Again the fungus secreted a product that killed bacteria in improvement by mutation and selection yielded strains that
the neighborhood of the mold colony. Fleming went one step produced higher and higher levels of antibiotic (> Table 13.2).
further. He grew large batches of the mold and then separated By 1943, the major bioengineering problems had been resolved
the mycelial mass from the culture medium. The mold-free juice by a cooperative effort of the Peoria laboratory, universities, and
he prepared was still lethal to the bacteria. Since the mold was industry, and the first aerated stainless steel tank for penicillin
a species of Penicillium, Fleming called the antibacterial material production was put into operation.
in the juice penicillin. Shortly thereafter he attempted to One of the ironies of the penicillin saga is that England,
concentrate and purify penicillin, failed, and abandoned the where the drug was first discovered and shown to be effective,
project. There followed a period of 10 years during which there had to pay US companies royalties after the Second World War
were no significant developments in penicillin research. Several for the technical know-how to produce penicillin. Fleming, of
suggestions have been put forth to explain this 10-year lag course, never attempted to patent penicillin. It is unfortunate
between the discovery of penicillin and its development as but true that even if he had, the patents would have expired
a potent chemotherapeutic agent. Although Fleming wrote before the drug was produced commercially. The Oxford group,
in 1929 that penicillin ‘‘may be an efficient antiseptic for on the other hand, could have obtained considerable financial
applications to, or injection into, areas infested with penicillin- reward for developing the penicillin extraction procedure,
sensitive microbes,’’ he failed to convince the scientific but instead they gave their knowledge freely to the world.
community. Fleming gave up because he was not a good enough The pioneering research on penicillin, however, did not go
chemist to purify the unstable molecule and lacked the money unrecognized: The 1944 Nobel Prize in medicine was shared by
necessary to hire chemists to help him. With the millions of Fleming, Chain and Florey. For the immigrant biochemist
dollars being spent today on medical research, it is difficult to Chain, there was the added pleasure of knowing that he was
comprehend how an established scientist like Fleming could not a major contributor to the development of a drug that saved the
then obtain the $5,000 per year that he needed. Finally, it should lives of thousands of young men who fought to liberate his
be mentioned that scientists are not always the best promoters former homeland—sweet justice.
of their own ideas. Fleming represents the classic example
of a scientist whose original findings were subsequently
rediscovered, developed, and exploited by others. Limitations of Penicillin and the Development
As a consequence of the rise of fascism in the 1930s, many of Other Antibiotics
outstanding scientists were forced to flee from Hitler’s Germany.
One of these refugees was the biochemist Ernst Chain. Shortly There are at least three serious limitations to the chemothera-
after arriving in England, Chain joined with Howard Florey, peutic use of penicillin. First, it cannot be taken orally.
a pathologist at Oxford, in a systematic search for antibacterial The molecule is rapidly decomposed in the acid of the human
substances. A careful library study led Chain to select penicillin stomach. Second, it is not effective against several bacteria,
as the target of research. A relatively short time after receiving including a number of Gram-negative pathogens. Third,
the penicillin-producing mold from Fleming, the team of Chain a significant number of people are allergic to it. At least three
and Florey succeeded in developing techniques for the purifica- approaches have been used by applied microbiologists and
tion of penicillin. The first animal experiment with the partially chemists to overcome these limitations: (1) synthesis of
purified penicillin was extremely encouraging. Subsequent stud- chemical derivatives of penicillin (semisynthetic penicillins)
ies demonstrated that penicillin was not toxic to mammals, that are more resistant to acid and penicillinase and inhibitory
including humans. Preliminary tests with humans suffering to a wider range of bacteria than the natural penicillin G, (2) use
from incurable bacterial infections showed that penicillin was of natural inhibitors of penicillinase, such as clavulanic acid, that
a miraculous drug. By 1941, the pressing problem was how to can work synergistically with b-lactam antibiotics to extend
produce enough penicillin for general clinical use. With the their range, and (3) development of many other antibiotics
outbreak of World War II, the need for drugs effective against for clinical use.
battle wounds became even more urgent. Florey and Chain went The great success of penicillin encouraged microbiologists to
to the Ministry of Health for help. A decision was made at the search for other antibiotics. A key figure in this search was the
highest level of government to assist the Oxford group in making soil microbiologist Selman Waksman of Rutgers University in
more penicillin. Florey was sent to Peoria, Illinois, the headquar- New Jersey. Between 1939 and 1945, Waksman and his associates
ters of the newly established Fermentation Division of the US examined thousands of microbes for their ability to produce
Department of Agriculture. The laboratory in Peoria was ideally antibacterial substances. A major conclusion that emerged from
suited for the project. The proper equipment was available and, the Rutgers study was that spore-forming bacteria in the soil,
. Table 13.3
Antibiotics used in chemotherapy
Antibiotic Source Used medically in treating Mode of action

Penicillin Fungus (Penicillium) Gram-positive infections Blocks bacterial cell-wall synthesis
Ampicillin (Penbritin) Semisynthetic Gram-positive and Gram-negative infections Blocks bacterial cell-wall synthesis
pencicillin
Cephalosporins Fungus and Gram-positive and Gram-negative infections Blocks bacterial cell-wall synthesis
Streptomycesa
Streptomycin Streptomyces Tuberculosis; Gram-negative intestinal tract Inhibits protein synthesis in bacteria
infections
Oxytetracycline (Terramycin) Streptomyces Most bacterial infections Inhibits protein synthesis in bacteria
Chlortetracycline Streptomyces Most bacterial infections Inhibits protein synthesis in bacteria
Chloramphenicol Streptomyces Typhoid fever; Rocky Mountain spotted fever Inhibits protein synthesis in bacteria
(Chloromycetin)b
Polymyxin (Aerosporin) Bacillus Gram-negative wound infections Destroys bacterial cytoplasmic
membrane
Bacitracin Bacillus Topical infections of eye and skin; burn Blocks bacterial cell-wall synthesis
infections
Novobiocin Streptomyces As penicillin; also penicillin-resistant Blocks bacterial nucleic acid
staphylococci synthesis
Erythromycin Streptomyces As penicillin; also penicillin-resistant Inhibits protein synthesis in bacteria
staphylococci
Griseofulvin Fungus and Fungal infections (taken orally) Destroys fungal cytoplasmic
Streptomycesa membrane
Polyenes (Nystatin) Streptomyces Systemic fungal infections Destroys fungal cytoplasmic
membrane
a
Initially discovered as a fungus, now known also to be produced by Streptomyces species
b
Initially discovered in a Streptomyces, now made chemically
especially the Streptomyces, were a rich source of antibiotics. the time the cells are sporulating. Because of these correlations,
From the thousands of antibacterial substances tested, Waksman some microbiologists have hypothesized that the natural role of
isolated, characterized and patented several important anti- antibiotics is to regulate the formation of spores in microbes.
biotics, including streptomycin, neomycin, and actinomycin. However, many mutants do not produce antibiotics yet sporu-
Streptomycin, first reported in 1944, quickly emerged as late normally. It is, therefore, reasonable to conclude that anti-
a valuable treatment for tuberculosis and other infections caused biotics have no single function, playing different natural roles in
by penicillin-resistant Gram-negative bacteria. With Waksman different microorganisms. (3) Medically, the most important
pointing his finger to the soil, the ‘‘gold rush’’ was on for new trait of an antibiotic is high therapeutic index, an indicator of
and better antibiotics. Thousands of active compounds have selective killing and low toxicity. Each antibiotic interferes with
been found, identified chemically, and screened for potential an essential and specific microbial function.
clinical use. Although most of the active compounds were
shown to be too toxic for general use, about 50 of them have
proven therapeutic value and are currently produced commer- The Role of Antibiotic Research in the
cially for medical and veterinary use. A partial list of important Development of Microbial Biochemistry,
antibiotics is shown in > Table 13.3. Molecular Genetics and Biotechnology
What are the common characteristics of antibiotics?
(1) From the chemical point of view, they are an extremely Much of what we now know about the bacterial cell wall and
diverse group of organic molecules. The only chemical feature biopolymer synthesis has come from studies using specific anti-
antibiotics seem to share is their relatively small size. Molecules biotics to inhibit synthesis. Furthermore, antibiotics were key
have fewer than 40 carbon atoms. (2) Ecologically, antibiotics tools in the development of microbial genetics. For example,
are produced almost exclusively by spore-forming soil microor- most auxotrophic mutants were obtained using the penicillin
ganisms. Furthermore, antibiotics are synthesized at precisely selection method (Davis 1948). Antibiotic resistance continues
to be the most important selection marker for genetic recombi- . Table 13.4
nation. The discovery of plasmids was initially connected to Some of the commercial applications of recombinant DNA
research on multidrug resistance or R factors. Even today, most technology
of DNA recombinant technology relies heavily on the use of
Applications Examples
antibiotics and antibiotic-resistance genes to construct useful
strains. Antibiotic research is an excellent example of how Production of mammalian Insulin, blood proteins, interferons,
applied and basic research overlap, interact, and mutually proteins in bacteria growth hormones, and tumor necrosis
benefit each other. factor
Vaccines Hepatitis B, rabies, measles, polyvalent
vaccines, and DNA vaccines
The New Biotechnology: Applications of Fermentation products Increased yields of antibiotics,
Genetic Engineering. restriction enzymes, proteases,
xylanases, amino acids, and vitamins
Cloning genes in bacteria is already a valuable technology in Diagnostics Diagnosis of pathogenic microbes and
both basic biological research and practical applications (Glazer genetic diseases
and Nikaido 1995). In basic research, genetic engineering in New antibiotics Hybrid polyketides
bacteria has been used to examine the underlying mechanisms Transgenic plants and Herbicide, insect and microbial
of DNA replication, gene recombination, and gene expression of animals disease resistant plants; cows that
viral, bacterial, and eukaryotic genes. The number of commer- produce pharmaceutical products in
cial applications of gene engineering continues to grow their milk
(> Table 13.4). Molecular biology of bacteria led the way for Gene therapy Introduction of cloned adenosine
the development of the biotechnology industry and today deaminase (ADA) into immune
molecular biology is the major driving force in pharmaceutical deficient patients lacking ADA
research (Ferrer-Miralles et al. 2009). Recombinant DNA Environmental Cloning and overexpression of genes
technology can be used to more efficiently carry out an existing biotechnology involved in the degradation of
process or develop entirely new products. One of the first pollutants
examples was the production of human insulin in Escherichia
coli. After the insulin gene was chemically synthesized and
inserted downstream from a suitable E. coli promoter, the cell for use in humans was against hepatitis B virus. Recombinant
was able to synthesize a fusion protein, which was then split into vaccines are also available for animal diseases, such as
the separate insulin peptides. These peptides, once connected by Newcastle’s and fowl pox diseases in poultry. Several trials are
disulfide bonds, were the final product that is identical to human currently being carried out with polyvalent (subunits from more
pancreatic insulin and now commercially available. Microbially than one protein) vaccines against acquired immunodeficiency
produced human insulin is not immunogenic and is purer and syndrome (AIDS). A new approach is to use specific, cloned
less expensive to produce than porcine or bovine insulin, which DNA itself as the vaccine. The DNA is taken up by the animal cell
was previously used. Several other human protein pharmaceu- and then transcribed and translated into an active immunogenic
ticals are now produced in microorganisms including anticancer protein. If successful, these DNA vaccines would be safe
agents, such as a- and b- interferon; human growth hormone for and inexpensive.
the treatment of dwarfism; nerve growth factor; lysozyme for the The fermentation industry has benefited greatly from
treatment of inflammations; and several blood proteins, such as recombinant DNA technology. The technology is used in strain
hemoglobins used as blood substitutes (Kumar 1995), improvement programs to obtain not only more efficient
erythroprotein to treat anemia, urokinase to assist in blood processes but also new products. Most of the DNA restriction
clotting, and tissue plasminogen activator to dissolve clots. enzymes used in biotechnology are produced in E. coli from
The number of new pharmaceutical products by recombinant cloned DNA that has been put into high expression vectors.
technology in 2005 comprised 134 approved products, with an Similarly, many of the commercial enzymes (such as proteinases
approximate sales value of 35 billion dollars (Lawrence 2006). used in the detergent and food industries, xylanases and amy-
Safe and highly effective vaccines can be produced by recom- lases used in the textile and paper industries, and glucose
binant DNA technology. Instead of using killed or attenuated isomerase used to make high fructose corn syrup) are produced
pathogenic bacteria or viruses, subunit vaccines are produced by in bacteria that are easy to grow in fermenters, for example,
cloning part of the protein that is on the outer surface of E. coli and Bacillus species (Godfrey and West 1996).
a bacterium or virus. The subunit protein vaccine is then An interesting approach to the production of new antibiotics is
produced in a microorganism, usually E. coli. The subunit to place selected genes from two different polyketide antibiotic-
vaccines produce a rapid and high level of protective immunity producing bacteria into a recipient bacterium that can
with no possibility of transmitting the infection. Furthermore, then synthesize a new polyketide antibiotic (McDaniel et al.
vaccines can be produced against microbes that are difficult to 1993). Another important application of recombinant DNA
grow in vitro. The first recombinant subunit vaccine approved technology is the construction of transgenic plants and animals.
Such organisms have great potential for enhancing agricultural processes. It is primarily by microorganisms that waste products,
productivity and improving the nutritional and storage proper- including most pollutants, are broken down. Microbes have
ties of meats and vegetables. Also in development are crop enormous biodegradation potential. They occupy all possible
plants, used for the production of recombinant vaccines, ecological niches, from deep in the earth to high in the atmo-
including edible vaccines against cholera and enteric diarrhea. sphere, and collectively are able to grow over a wide range of
In many of the plant engineering constructs, Agrobacterium temperatures, pH values, salt concentrations, and oxygen
tumefaciens is used to introduce desired genes into the plant. tensions. There is not a single known natural organic compound
Monsanto Company (St. Louis, MO) is presently marketing that cannot be degraded by one or more microbial species.
a transgenic soybean plant that has an Agrobacterium-derived If microbes have such enormous degradative power, why do
gene that confers resistance to Roundup, a glyphosate herbicide. we have pollution? There are two fundamental reasons: (1) man-
Genetic engineering also has been used to protect plants from made conditions can become unfavorable for rapid degradation
viral diseases. Calgene Corporation (Davis, CA) produces and (2) certain man-made synthetic compounds are resistant to
a transgenic tomato that spoils more slowly than normal biodegradation. The former is dictated by the ‘‘law of the
tomatoes because it contains an inhibitor of pectinase transcrip- minimum,’’ that is, growth is limited by one essential compo-
tion. Transgenic animals are used extensively in research as well nent. For example, if a large amount of petroleum is dumped
as for the commercial production of proteins of pharmaceutical into the sea, biodegradation may not be at an appreciable rate
value, such as a-1-antitrysin in sheep and tissue plasminogen because utilizable forms of nitrogen and phosphorus are not
activator in goats. Biotechnology is also involved with human available—even if many hydrocarbon-degrading bacteria are at
genes and, in the long run, recombinant DNA technology will the site. Nitrogen and phosphorus are, of course, essential
probably have its most dramatic effect on the human genome. components for all living cells. Another condition, which is
As Robert Sinsheimer stated 30 years ago: unfavorable for biodegradation, is high concentration of organic
"
material. Two good examples of this are whiskey and jam in
For eon after eon, creature has given rise to creature upon this
which ethyl alcohol and sugar are, respectively, at unnatural
earth—blindly, each generation usually like the former, occasion-
levels that resist microbial breakdown. One of the reasons that
ally—by accident—a little different. Of all the creatures that have
organic matter is degraded so slowly in garbage dumps is
lived upon this earth we are the first to understand this process.
that the consolidation and concentration of materials make
The ultimate significance of this understanding of the very basis
biodegradation difficult. In the above examples, natural bacteria
of heredity is incalculable. It will change all the eons to come.
can completely degrade (mineralize) these materials, whether
(Rosenberg 1991)
they are hydrocarbons in crude oil or cellulose in newspaper, if
Already, PCR technology is widely used in forensic medicine the conditions are favorable. However, several man-made
to determine parentage, to analyze evolutionary relationships unnatural compounds, such as certain halogenated aromatics
between modern man, early man, and pre-man, and so forth. In and plastics, are not biodegradable under any circumstances.
one of the largest biological research programs ever conducted, They simply have not been on the planet long enough for
the human genome project, the entire nucleotide sequence of bacteria to have evolved degradative mechanisms.
the human genome will soon be determined. By comparing the Bioremediation is the use of living organisms, usually
DNA sequence of healthy and diseased individuals, the precise bacteria, to prevent or clean up pollution. Bioremediation is
position and aberration in the ‘‘bad’’ gene may be found. a large and growing area of applied microbiology. Hundreds of
Gene therapy, through genetic engineering, offers the hope companies around the world are involved in treating water and
of replacing disease-causing genes with healthy ones, thereby soil that is polluted with industrial wastes, such as petroleum
providing preventive medicine in the most profound sense. products, herbicides, pesticides, toxic metals, dioxenes, and
polychlorinated biphenyls (PCBs), and explosives like TNT.
There are two general approaches to bioremediation. The first
Applied Environmental Microbiology: and most successful method is to optimize conditions so
Bioremediation bacteria native to the polluted site can more rapidly degrade
the pollutants. The second method, which has had limited
See Chaps. 10, ‘‘Biosurfactants’’ and Chap. 11, ‘‘Bioremediation’’ success to date, is to introduce more efficient degraders to the
in Vol. 4. polluted site, or in the case of nonbiodegradable synthetic
Albert Einstein defined the environment as ‘‘anything which is pollutants, to construct genetically engineered microorganisms
not me.’’ This definition is interesting for two reasons: it has (GEMs) that can degrade the pollutant. The following examples
enormous scope, and it points the finger away from each of us. will illustrate these two approaches.
Individuals, industries, and societies have assumed incorrectly
that if the environment is not ‘‘them,’’ they do not have to be
responsible for it. The result is worldwide pollution, which Petroleum Pollution and Bioremediation
degrades the quality of our lives and threatens our very existence.
When considering pollution, one must focus on the During the twentieth century, the demand for petroleum as
turnover of matter, which is fundamentally a series of microbial a source of energy and as a primary raw material for the chemical
. Table 13.5 . Table 13.6

Requirements for biodegradation of petroleum Bioremediation of hydrocarbon-contaminated sand at an Israeli
beach
A. Microorganisms with:
1. Hydrocarbon-oxidizing enzymes Hydrocarbon degradation (%)
2. Ability to adhere to hydrocarbons Day Control (untreated) Experimental (N3P added)b
3. Mechanism for desorption from hydrocarbons 10 10 20
4. Emulsifier-producing potential 20 20 41
B. Water 30 22 50
C. Oxygen 40 18 60
D. Relatively large amounts of utilizable nitrogen and phosphorus 50 20 65
E. Iron, magnesium, and other essential elements 70 10 70
90 0 75
120 0 88
industry has increased world production to about 3,500 million a
Data from Rosenberg et al. (1996)
metric tons per year. Approximately 1 % or 35 million b
Approximately 30,000 m2 of heavy oil contaminated beach were treated with
tons finds its way to soils, lakes, and seas. The toxicity of 15 g urea-formaldehyde polymers (containing utilizable N and P) per kilogram
petroleum to microorganisms, plants, animals, and man is sand. The area was plowed twice weekly
well established. The polycyclic aromatic hydrocarbon (PAH)
fraction of petroleum is particularly toxic and is on the United
States Environmental Protection Agency priority pollutant list frequent plowing and to underground reservoirs by pumping
(Cernilia 1992). aerated water through them. Soils can be amended with utilizable
It has been known for over 80 years that certain bacteria nitrogen and phosphorus salts. However, no satisfactory method
are able to degrade petroleum hydrocarbons and use them as exists that provides the nitrogen and phosphorus necessary for
a sole source of carbon and energy. Effective bioremediation bioremediation of oil spills in open systems, such as the sea. The
of petroleum pollution relies heavily on a fundamental ideal nitrogen and phosphorus source should adhere to the oil,
understanding of the requirements for petroleum biodegrada- rather than be diluted with water. Recently, it has been
tion (> Table 13.5). Generally, the presence of hydrocarbon- shown that water-insoluble fertilizers, such as uric acid and
degrading microorganisms is not the limiting factor in urea-formaldehyde polymers, can be effective in stimulating
petroleum bioremediation. Hydrocarbon-oxidizing bacteria biodegradation of hydrocarbons in the marine environment
have been isolated from a wide variety of natural aquatic and (Koren et al. 2003; Knezevich et al. 2006). An example of
terrestrial environments (Rosenberg 1991), including bioremediation of beach sand by addition of urea-formaldehyde
hydrothermal vent sites (Baylinski et al. 1989). Furthermore, polymers is shown in > Table 13.6. Overall, the bioremediation
the addition of oil causes a rapid shift to increased numbers of treatment resulted in the degradation of 88 % of the oil after 4
hydrocarbon degraders. For example, Song and Bartha (1990) months. During the same period, there was virtually no change
demonstrated that the number of hydrocarbon-degrading in the oil concentration of the untreated control plot.
microorganisms increased from 4104 to 31010 per g of soil
4 weeks after addition of jet fuel and inorganic nutrients.
The practical conclusion from this and other studies is that The Use of GEMs in Bioremediation
seeding an oil-polluted area with microorganisms is generally
not necessary, and GEMs are not required for petroleum The first GEM to be patented in the USA was a Pseudomonas that
bioremediation. was designed for degrading different hydrocarbons and
In addition to having the biochemical potential for constructed from genes derived from different bacteria
degrading one or more classes of hydrocarbons, an efficient (Chakrabarty and Gunsalus 1971). The great potential of GEMs
hydrocarbon degrader must be able to adhere to and desorb for bioremediation was publicized by the news media, which
from water-insoluble hydrocarbons. The role of cell surface referred to them as ‘‘superbugs.’’ Unfortunately, the potential of
hydrophobicity, adhesins, and bioemulsifiers in the growth GEMs in bioremediation has not been realized as yet for several
process has been reviewed (Rosenberg and Ron 1996; Ron and reasons. (1) Safety regulations have restricted the release of
Rosenberg 2009). GEMs into the environment; in fact, most of the research carried
The most important factor determining the success of out on GEMs and bioremediation has been diverted to studying
a petroleum pollution bioremediation project is the ability to the fate of GEMs in the environment. (2) Genetically engineered
provide the optimum conditions for the hydrocarbon-degrading microorganisms do not appear to have any significant advantage
bacteria. The major requirements are adequate sources of oxygen, over a mixture of existing bacteria for the in situ degradation of
nitrogen, and phosphorus. Oxygen can be added to soils by natural pollutants, such as crude oil. (3) Also, to construct
GEMs that effectively degrade recalcitrant synthetic pollutants patient may be at risk, or in the case of a water supply or
has proved difficult. In spite of these problems, progress is being consumer product, many people could be infected during the
made in constructing GEMs for degrading haloaromatic diagnosis period. Therefore, it would be advantageous to have
compounds (Reineke 1998; Timmis et al. 1994; Cases and more rapid and sensitive techniques.
Lorenzo 2005) and other pollutants (Chen et al. 1999). When During the last 10 years, many new biotechnology companies
enhanced biodegradative pathways are combined with the have emerged that have as their major focus the development of
genetic potential to catabolize these pollutants in the natural specific diagnostic tests that do not depend on culturing micro-
environment, then mineralization rates will likely improve. One organisms. Rather, these new biotechnology methods rely on
possible approach is to clone genes for biosurfactant production amplification of a species-specific DNA target, of the probe, or
into GEMs that have good catabolic potential. of the signal that is generated (Hammes and Higgins 1995;
Whelen and Persing 1996). In addition to great sensitivity and
rapidity, these methods can detect pathogens that are difficult to
Health-Related Applied Microbiology culture, such as Tropheryma whippelii (Relman et al. 1992),
Chlamydia trachomatis (Chapin-Robertson 1993), and Neisseria
See Chap. 9, ‘‘Bacterial Pharmaceutical Products’’ in Vol. 4. gonorrhoeae (Hale et al. 1993). Nucleic acid probes have already
replaced classical morphology and biochemical tests in clinical
laboratories for the definitive identification of mycobacteria
Drug Discovery (Keller et al. 2002). Considerable effort is now being employed
to develop biotechnology-based detection methods for drug
One of the most exciting areas of applied microbiology involves resistance in clinically important organisms (Eron et al. 1992).
the discovery of new drugs for human and animal health. Sec- Advances in approaches to DNA-based diagnostics have been
ondary metabolites produced by microorganisms provide a vast reviewed (Whitcombe et al. 1998; Tang and Stratton 2011).
array of potentially useful drugs. While the search for new drugs
from Streptomyces and other spore-forming bacteria continues
to be productive, new sources of bacteria are now being Food Microbiology
exploited, such as marine bacteria (Fenical 1993, 1997; Hill
and Fenical 2010). In addition to expanding the pool of micro- See Chap. 8, ‘‘Bacteria in Food and Beverage Production’’
organisms in the search for novel pharmaceuticals, researchers in Vol. 4.
are now employing new technologies for drug screening (Steele Food microbiology, one of the traditional areas of applied
and Stowers 1991; Lloyd 1998; Schmid et al. 1999). These microbiology, is changing dramatically. Microorganisms play
include robotic techniques for plating test bacteria and deliver- a direct role in producing and processing the healthy foods
ing dilutions of bacterial extracts, use of monoclonal antibodies we eat and drink. Meat and milk are available to us only
and fluorescent techniques and combinatorial biosynthesis because microbes, in symbiotic relationships with certain
(Hutchinson 1998), and screens based on specific enzymatic animals, are able to digest cellulose. Different microorganisms
reactions. These new techniques allow for a more rapid and convert milk to a variety of yogurts, cheeses, and butter. Still
comprehensive screening for new drugs. other microbes preserve plant materials in the form of tasty
products such as pickles and sauerkraut. Alcoholic fermenta-
tion by yeast is needed for bread as well as all alcoholic
Diagnostic Microbiology beverages. Hagedom and Kaphammer (1994) have reviewed
the role of microbial biocatalysis in the generation of flavor
Detecting, classifying, and enumerating microorganisms in and fragrance chemicals.
drinking water, foods and beverages, and human tissues and An interesting recent development in food microbiology is
fluids is a large area of applied microbiology that is constantly the use of ‘‘probiotic bacteria’’ to improve animal and human
undergoing improvements. During most of the twentieth cen- food (Bengmark 1998; Holzapfel 1998; Guarino 1997). Probiotic
tury, microbiology laboratories developed many dependable bacteria are generally isolated from the mouth, stomach, and
approaches for detecting microorganisms, especially pathogens, upper and lower bowel of healthy humans and animals. These
using microscopy coupled to staining techniques, culturing in normal resident bacteria are beneficial, providing vitamin K,
selective media, and immunoserology (Koneman et al. 1992). helping to process nutrients, preventing colonization of tran-
This field of diagnostic microbiology has proved invaluable to sient bacteria (nonspecific defense mechanism), developing the
physicians treating patients, food producers controlling the immune system, and stimulating phagocytic activity. The future
quality of their products, and health officials monitoring water of probiotics will depend upon the development of safe strains
supplies and consumer products. However, many of these diag- that are well adapted to occupy specific niches in the body and
nostic tests are slow, expensive, and insensitive. For example, promote health. When added to food or used as starters in food
culturing bacteria from urine or blood and then testing them for fermentations, these strains can provide the beneficial effects of
antibiotic sensitivity can take several days. During this time, the probiotics to man and animals.
Biosafety and Legal Protection in Environmental Microbiology

Biotechnology
The rapid increase in human population and technological
The safety of research microbiologists and legal protection for developments is degrading our environment at an alarming
their discoveries are two subjects, which have become more rate, causing a lowering of the quality of life and serious health
important in recent years in applied microbiology and biotech- problems. As a recent example, the Deepwater Horizon oil
nology. Unfortunately, most microbiologists do not get an spill released 140 million gallons of crude oil into the Gulf of
adequate education in these topics. Many laboratory techniques Mexico, causing extensive damage to marine and wildlife
create aerosols, which can lead to the inhalation of undetected habitats and to the Gulf ’s fishing and tourism industries
infectious agents and the occupational illnesses among (Biello 2010). Since microorganisms are largely responsible for
laboratory workers who handle infectious materials. Based the turnover of matter on this planet, applied environmental
upon available data, preventative measures have been developed microbiology will become more important. Not only will
which provide safeguards for scientific personnel and the bioremediation develop further scientifically, technologically,
environment. These safeguards are collectively called contain- and commercially, but also selected microbes will be used
ment practices. Leiberman et al. (1986) have written a concise under controlled conditions to recycle waste and to prevent
review on biosafety in biotechnology, which should be studied pollution by replacing pollution-producing chemical processes.
by all practicing microbiologists. The so-called ‘‘green revolution’’ will be carried out by
To derive financial benefit from an original and potentially microorganisms supplied by a new generation of applied
useful discovery in biotechnology, it should be patented microbiologists.
(see > Chap. 12, ‘‘Repositories for Patented and Safeguarded An outcome of research in environmental microbiology will
Material’’ in Vol. 1). This is particularly important for research be the emergence of new principles of ecology. For the same
scientists at universities because once a patent application has reasons that microbes were ideal organisms to study the
been filed, it is possible to disclose the invention (scientific principles of biochemistry and genetics, they will be used to
publications, grant applications, seminars, etc.) and still have develop fundamental principles of ecology. The use of higher
legal protection. In the USA, it is possible to file a patent within 1 animals and plants for deriving general ecological principles is
year of the publication date. The culture of science demands free limited by our inability to carry out controlled experiments.
exchange of ideas and data. Patenting makes this possible, while Animal and plant ecologists can make accurate observations,
still protecting the rights of the scientist. Another reason for ask interesting questions and present hypotheses, but the
patenting a discovery is that it allows the owner of the patent to microbial ecologist can obtain quantitative data in controlled
more safely market the invention. An applied microbiologist at experiments, using molecular probes and other tools of modern
a university or research institution cannot generally convert his microbiology. The combined effort of animal, plant, and micro-
discovery into a commercial product. It is therefore necessary bial ecologists should lead to a better understanding of how
for the institution to find a partner or sell the patent and know- organisms interact with their environment.
how. The procedures for obtaining a patent have been outlined
by Saliwanchik (1986).
Microbial Associations with Animals, Plants,
and Humans
Outlook
Bacterial symbionts play a key role in maintaining the health of
The border between basic and applied microbiology, which has all animals, plants, and humans. Cooperation between the
always been tenuous, is disappearing. The period in which normal microbiota and the host generally leads to improved
a discovery in basic microbiology becomes an applied project fitness of the holobiont, by the host outsourcing (Zilber-
is often very short. Some of the most basic discoveries in micro- Rosenberg and Rosenberg 2008) different kinds of functions to
biology are now made by scientists working on applied problems its microbiota and vice versa. Protection against infectious dis-
in universities, research institutions, and biotechnology compa- ease is one of the important attributes of the resident
nies. The rapid progress taking place in applied microbiology microbiota. The exact mechanism is unknown, but it has been
and biotechnology, made possible by almost a century of fun- suggested that resident bacteria occupy binding sites needed by
damental work on microbial biochemistry, physiology, and pathogens for adhesion in addition to releasing antibacterials
genetics, makes it difficult to predict which areas will develop active against pathogens. Another important known beneficial
most rapidly in the next few decades. Clearly, the use of bacteria functions of microbiota is participation in the development and
in DNA-based technologies will continue to grow and find new normal function of the innate and adaptive immune systems in
applications. Already, a large percentage of the corn grown in the the gut (O’Hara and Shanahan 2006; Ivanov and Littman 2011)
USA contains cloned Bacillus thuringiensis protein as an endog- while creating a permissive, noninflammatory environment for
enous insecticide. In addition, I predict the following three their own presence (Hapfelmeier et al. 2010). The microbiota
general subjects will soon emerge as highly significant in applied also play a key role in angiogenesis, (Stappenbeck et al. 2002),
microbiology and biotechnology. they synthesize and excrete vitamins in excess of their own
needs, which can be absorbed as nutrients by their host. For Chakrabarty AM, Gunsalus IC (1971) Degradative pathways specified by
extrachromosomal gene clusters. Pseudomonas Genet 68:510
example, in humans, enteric bacteria secrete Vitamin K and
Chapin-Robertson K (1993) Use of molecular diagnostics in sexually transmitted
Vitamin B12, and lactic acid bacteria produce certain B-vitamins diseases: critical assessment. Diagn Microbiol Infect Dis 16:173–184
(Mai et al. 2010). The human gut microbiota is a complex Chen W, Bruhlmann F, Richins RD, Mulchandani A (1999) Engineering of
ecosystem that plays an essential role in the catabolism of improved microbes and enzymes for bioremediation. Curr Opin Biotechnol
dietary fibers, and fat metabolism (obesity) and related disor- 10:137–141
Davis BD (1948) Isolation of biochemically deficient mutants of bacteria by
ders (Ley et al. 2006; Turnbaugh et al. 2006; Cani and
penicillin. J Am Chem Soc 70:4267
Delzenne 2009). Demain AL, Adrio JL (2008) Strain improvement for production of pharmaceu-
As we learn more about these interactions, use of selected ticals and other microbial metabolites by fermentation. Prog Drug Res
and improved bacterial strains will grow: in agriculture, for 6:253–289
improving yields and preventing spoilage; in animal husbandry, Dubos RJ (1950) Louis Pasteur: free lance of science. Little, Brown and Company,
Boston
for improving yields and nutritional value of meat, milk, and
Elander RP (1966) Two decades of strain development in antibiotic-producing
eggs; in humans, for maintaining health and deterring diseases microorganisms. Dev Ind Microbiol 7:61–73
(Zilber-Rosenberg and Rosenberg 2011). The development of Eron JJ, Gorczyc P, Kaplan JC, D’Aquila RT (1992) Susceptibility testing by
defined prebiotics and probiotics will become an important polymerase chain reaction DNA quantitation: a method to measure drug
applied microbial technology. resistance of human immunodeficiency virus type 1 isolated. Proc Natl Acad
Sci 89:3241–3245
Fenical W (1993) Chemical studies of marine bacteria: developing a new resource.
Chem Rev 93:1673–1683
Microbial Enzymes Fenical W (1997) New pharmaceuticals from marine organisms. Trends
Biotechnol 15:339–341
Microbial enzymes are currently being used commercially, Ferrer-Miralles N, Domingo-Espin J, Corcero JL, Vasquez E, Villaverde A (2009)
Microbial factories for recombinant pharmaceuticals. Microb Cell Fact
primarily for degradative reactions, for example, proteinases,
8:17–25
cellulases, amylases, and lipases. This area will be further devel- Gaden EL (1981) Production methods in industrial microbiology. Sci Am
oped with the introduction of more efficient enzymes and 245:134–145
enzymes from extremophiles with increased heat tolerance and Gerhardt P (ed) (1981) Manual of methods for general bacteriology. ASM,
extended pH range. However, the commercial potential of Washington, DC
Glazer AN, Nikaido H (1995) Microbial biotechnology: fundamentals of applied
microbial enzymes used to synthesize drugs and other fine and
microbiology. W. H. Freeman, New York
bulk chemicals is even greater. With the accumulating data on Godfrey T, West S (1996) Industrial enzymology. In: Godfrey T, West S (eds) The
biosynthetic pathways and the ability to overproduce specific application of enzymes in industry, 2nd edn. Stockton, New York
enzymes, it should be possible to mix and match enzymes both Guarino A (1997) Oral bacterial therapy reduces the duration of symptoms and
in vivo and in vitro to efficiently produce existing valuable of viral excretion in children with milk diarrhea. J Pediatr Gastroenterol Nutr
25:516–519
products as well as new materials. In this regard, bacterially
Guerreiro MA, Andrietta SR, Maugeri F (1997) Expert system for the design of an
produced polyhydroxybutyrate and its copolymers have industrial fermentation plant for the production of alcohol. J Chem Technol
attracted considerable attention as environmentally degradable Biot 68:163–170
plastics for a wide range of agricultural, industrial, and medical Hagedom S, Kaphammer B (1994) Microbial biocatalysis in the generation of
applications (Kumaravel et al. 2010). flavor and fragrance chemicals. Annu Rev Microbiol 48:773–800
Hale YM, Melton ME, Lewis JS, Willis DE (1993) Evaluation of PACE 2 Neiseria
This short list is certainly not comprehensive; many other
gonorrhoeae assay by three public health laboratories. J Clin Microbiol
subjects of applied microbiology will expand and new areas will 31:451–453
be created. In short, the future of applied microbiology and Hammes BD, Higgins SJ (1995) Gene probes: a practical approach. Oxford
biotechnology is bright. University Press, New York, pp 1–288
Hapfelmeier S, Lawson MAE, Slack E et al (2010) Reversible microbial coloniza-
tion of germ-free mice reveals the dynamics of IgA immune responses.
Science 328:1705–1709
References Hill RT, Fenical W (2010) Pharmaceuticals from marine natural products: surge
or ebb? Curr Opin Biotechnol 6:777–779
Aharonowitz Y, Cohen G (1981) The microbiological production of pharmaceu- Holzapfel HH (1998) Overview of gut flora and probiotics. Int Food Microbiol
ticals. Sci Am 245:106–119 41:85–101
Baylinski DA, Wirsen CE, Jannasch HW (1989) Microbial utilization of naturally Hutchinson CR (1998) Combinational biosynthesis for new drug discovery. Curr
occurring hydrocarbons at the Guaymas Basin hydrothermal vent site. Appl Opin Microbiol 1:319–329
Environ Microbiol 55:2832–2836 Hutchinson CR, Fujii I (1995) Polyketide synthase gene manipulation:
Bengmark S (1998) Ecological control of the gastrointestinal tract. Gut 42:2–7 a structure-function approach in engineering novel antibiotics. Annu Rev
Biello D (2010) Meet the microbes eating the gulf oil spill. Sci Am 303:14–17 Microbiol 49:201–238
Cani PD, Delzenne NM (2009) The role of the gut microbiota in energy metab- Ivanov II, Littman DR (2011) Modulation of immune homeostasis by commensal
olism and metabolic disease. Curr Pharm Des 15:1546–1558 bacteria. Curr Opin Microbiol 14:106–114
Cases I, Lorenzo V (2005) Genetically modified organisms for the environment: Jones D, Krieg NR (1984) Seriology and chemotaxonomy. In: Krieg NR, Holt JC
stories of success and failure and what we have learned from them. Int (eds) Bergey’s manual of systematic bacteriology. Williams and Wilkins,
Microbiol 8:213–222 Baltimore, pp 15–18
Cernilia CE (1992) Biodegradation of polycyclic aromatic hydrocarbons. Keller AP, Beggs ML, Amthor B, Bruns F, Meissner P, Haas WH (2002) Evidence of
Biodegradation 3:351–368 the presence of IS1245 and IS1311 or closely related insertion elements in
nontuberculous mycobacteria outside of the Mycobacterium avium complex. Rosenberg E, Legmann R, Kushmaro A, Adler E, Abir H, Ron EZ (1996) Oil
J Clin Microbiol 40:1869–1872 bioremediation using insoluble nitrogen source. J Biotechnol 51:273–278
Knezevich V, Koren O, Ron EZ, Rosenberg E (2006) Petroleum bioremediation in Saliwanchik R (1986) Legal protection for biotechnology. In: Demain AL,
seawater using guano as the fertilizer. Bioremed J 10:83–91 Solomon NA (eds) Industrial microbiology and biotechnology. ASM,
Koneman EW, Allen SD, Janda WM, Schreckenberger PC, Winn WC Jr (1992) Washington, DC, pp 389–401
Diagnostic microbiology. 4th ed. Philadelphia: JB Lippincott Schmid I, Sattler I, Grabley S et al (1999) Natural products in high throughput
Koren O, Knezvic V, Ron EZ, Rosenberg E (2003) Petroleum bioremediation screening: automatic high-quality sample preparation. J Biomol Screen
using water insoluble uric acid as the nitrogen source. Appl Environ 4:15–25
Microbiol 69:6337–6339 Silverman L, Campbell R, Broach JR (1998) New assay technologies for
Kreig NR (1981) Enrichment and isolation. In: Gerhardt P (ed) Manual of high-throughput screening. Curr Opin Chem Biol 2:397–403
methods for general bacteriology. ASM, Washington, DC, pp 112–142 Song HG, Bartha R (1990) Effects of jet fuel spills on the microbial community in
Kumar R (1995) Recombinant hemoglobins as blood substitutes: a biotechnology soil. Appl Environ Microbiol 56:646–651
perspective. Proc Soc Exp Biol 208:150–158 Stappenbeck TS, Hooper LV, Gordon JI (2002) Developmental regulation
Kumaravel S, Hema R, Lakshmi R (2010) Production of polyhydroxybutyrate of intestinal angiogenesis by Peneth cells. Proc Natl Acad Sci USA
(bioplastic) and its biodegradation by Pseudomonas lemoignei and Aspergillus 99:15451–15455
niger. E-J Chem 7(S1):S536–S542 Steele DB, Stowers MD (1991) Techniques for selection of industrially important
Lawrence S (2006) Biotech blockbusters consolidate markets. Nat Biotechnol microorganisms. Annu Rev Microbiol 45:89–106
24:1466 Tang YW, Stratton CW (2011) Advanced techniques in diagnostic microbiology
Leiberman DG, Fink R, Schaffer F (1986) Biosafety in biotechnology. In: Demain overview. Springer, New York
AL, Solomon NA (eds) Industrial microbiology and biotechnology. ASM, Timmis KW, Steffan RJ, Unterman R (1994) Designing microorganisms for the
Washington, DC, pp 402–409 treatment of toxic wastes. Annu Rev Microbiol 48:525–557
Ley RE, Turnbaugh PJ, Klein S, Gordon JI (2006) Human gut microbes associated Trilli A (1981) Scale-up of fermentation. In: Demain AL, Solomon NA (eds)
with obesity. Nature 444:1022–1023 Industrial microbiology and biotechnology. ASM, Washington, DC,
Lloyd A (1998) High-throughput screening. Drug Discov Today 3:566–569 pp 277–307
Mai V, Ukhanova M, Baer DJ (2010) Understanding the extent and sources of Turnbaugh PJ, Ley RE, Mahowald MA et al (2006) An obesity-associated
variation in gut microbiota studies; a prerequisite for establishing associa- gut microbiome with increased capacity for energy harvest. Nature
tions with disease. Diversity 2:1085–1096 444:1027–1031
McDaniel R, Ebert-Khosla S, Hopwood DA, Khosla C (1993) Engineered biosyn- Veldamp H (1970) Enrichment cultures of prokaryotic organisms. In: Noris JR,
thesis of novel polyketides. Science 262:1546–1550 Ribbons DW (eds) Methods in microbiology, vol 6. Academic, New York,
O’Hara AM, Shanahan F (2006) The gut flora as a forgotten organ. EMBO Rep p 305
7:688–693 Wang HY (1981) Bioinstrumentation and computer control of fermentation
Prescott SC, Dunn CG (1940) Industrial microbiology. McGraw-Hill, New York processes. In: Demain AL, Solomon NA (eds) Industrial microbiology and
Price KE, Buck RE, Lein J (1964) System for detecting inducers of lysogenic biotechnology. ASM, Washington, DC, pp 308–320
Escherichia coli W1709 (gamma) and its applicability as a screen for anti- Whelen AC, Persing DH (1996) The role of nucleic acid amplification and
neoplastic antibiotics. Appl Microbiol 12:428–435 detection in the clinical microbiology laboratory. Annu Rev Microbiol
Queener SW, Lively DH (1986) Screening and selection for strain improvement. 50:349–373
In: Demain AL, Solomon NA (eds) Industrial microbiology and biotechnol- Whitcombe D, Newton CR, Little S (1998) DNA-based diagnostics. Curr Opin
ogy. ASM, Washington, DC, pp 155–169 Biotechnol 9:602–608
Reineke W (1998) Development of hybrid strains for the mineralization of White RJ, Maiese WM, Greenstein M (1986) In: Demain AL, Solomon NA (eds)
chloroaromatics by patchwork assembly. Annu Rev Microbiol 52:287–332 Industrial microbiology and biotechnology. ASM Press, Washington, DC,
Reisman HB (1993) Problems in scale-up of biotechnology production processes. pp 24–31
Crit Rev Biotechnol 13:195–253 Zhang XHD (2011) Optimal high-throughput screening: practical experimental
Relman DA, Schmidt TM, MacDermott RP, Falkow S (1992) Identification of the design and data analysis for genome-scale RNAi research. Cambridge
uncultured bacillus of Whipple’s disease. N Eng J Med 327:293–301 University Press, Cambridge
Ron EZ, Rosenberg E (2009) Role of biosurfactants. In: Timmis KN (ed) Hand- Zhang YX, Perry K, Vinci VA et al (2002) Genome shuffling leads to rapid
book of hydrocarbon and lipid microbiology. Springer, Heidelberg phenotype improvement in bacteria. Nature 415:644–646
Rosenberg E (1991) Hydrocarbon-oxidizing bacteria. In: Balows A, Truper HG, Zilber-Rosenberg I, Rosenberg E (2008) Role of microorganisms in the evolution
Dworkin M, Harder W, Schleifer K-H (eds) The prokaryotes, 2nd edn. of animals and plants: the hologenome theory of evolution. FEMS Microbiol
Springer, New York, pp 446–459 Rev 32:723–735
Rosenberg E, Ron EZ (1996) Bioremediation of petroleum contamination. Zilber-Rosenberg I, Rosenberg E (2011) Prebiotics and probiotics within the
In: Crawford RL, Crawford DL (eds) Bioremediation: principles and framework of the hologenome concept. J Microbial Biochem, doi:10.4172/
applications. Cambridge University Press, New York, pp 100–124 1948-5948.S1-001
14 Genomes and Post-genome Technology
Betania Ferraz Quirino1,2 . Cristine Chaves Barreto1 . Georgios J. Pappas3,6 . Karsten Zengler4 .
Konstantinos Krampis5 . Ricardo H. Krüger6
1
Genomic Sciences and Biotechnology Program, Universidade Católica de Brası́lia, Brası́lia, DF, Brazil
2
Embrapa-Agrienergy, Brası́lia, DF, Brazil
3
Embrapa-Genetic Resources and Biotechnology, Bioinformatics Laboratory, Brası́lia, DF, Brazil
4
Department of Bioengineering, University of California, San Diego, La Jolla, CA, USA
5
J. Craig Venter Institute, Rockville, MD, USA
6
Biology Institute, UnB Brası́lia University, Brası́lia, DF, Brazil
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 its classification. DNA-DNA hybridization, 16S rRNA gene

sequencing, multilocus house-keeping genes sequencing, and
Application of Prokaryotic Intraspecies super-trees are the standards for bacterial classification. The
Genomic Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330 first complete bacterial genome sequenced was that of
Haemophilus influenza. This genome was published in the early
Culture-Independent Exploitation of Microbial Genomes: 1990s (Fleischmann et al. 1995) and was undoubtedly a seminal
Metagenomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 achievement for the microbiology field. Now, with our access to
whole genome information, even from ‘‘uncultivable’’ microbe
Single Cell Genomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336 representatives, we have a better understanding of what
a microbe is capable of doing, its role in the microbial commu-
Using Genome and Post-genome Information to Make nity, and its role in the environment. Access to individual
a Product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337 genomes allowed comparison among genomes from different
species, among genomes from different isolates from the same
DNA Sequencing and Bioinformatics . . . . . . . . . . . . . . . . . . . . . 338 species (pan-genome), and finally metagenomes (the collective
Working with Large Amounts of Data . . . . . . . . . . . . . . . . . . 338 community genome) (Medini et al. 2005, 2008; Ansorge 2009;
Analytical Pipelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340 MacLean et al. 2009).
Metagenomics Pipelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340 Some of the immediate benefits we have already reaped from
genomic studies are the understanding of the general organiza-
Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340 tion of genomes, the constituents and regulators of metabolic
pathways, virulence factors, and other genes involved in host-
pathogen interactions. However, as with any major scientific
Introduction advancement, this knowledge paved the road for uncountable
other inquiries that the structural sequence of a given genome is
Until a few years ago, microbial biotechnology was limited by the not capable of providing clear answers to (e.g., a universal bac-
use of cultivated microorganisms. However, most of the microor- teria species concept). It is clear, as billions of bases pour into
ganisms that can be seen through a microscope or be detected databases every day, that as we move from genomes, single cell
using molecular biology tools are labeled as non-cultivable (or genome, pan-genomes to metagenomes and other ‘‘omes,’’ we
not-yet-cultivated). Nowadays, this previously unreachable geno- need to think about how to make the most of this information,
mic information can be accessed without the need of having the not only from a basic biology standpoint but from a biotech-
microorganism in pure culture. Microbiology is at one of those nology perspective.
inflection points in history, and it is expected that this newly The availability of genome sequences has allowed a better
accessible wealth of microorganisms will bring significant contri- understanding of genome evolution, lateral gene transfer, and
butions to mankind over the next decades. This will be achieved by genomic plasticity. Furthermore, we should be able to explore
the development of new biotechnological products for plant and this genomic plasticity and ability to adapt to ever-changing
animal disease, diagnostics, management, and control. Further- environmental conditions to bio-prospect in a systematic fash-
more, there will be significant contributions to improve the quality ion the universe of biochemical reactions and compounds pro-
of crops and food production yields for an ever-growing human duced and develop products – if we are clever enough to design
population, development of new biocatalysts, new processes, and bioassays. In addition, the tools to manipulate the genome are
applications for bioremediation and energy production. being developed and it is now possible to reduce genome size to
Although not absolutely necessary for biotechnological a minimum to sustain life in free form. It is expected that this
application, one of the first steps to study a microorganism is development will allow the custom design of a microorganism to
330 14 Genomes and Post-genome Technology
perform particular reactions with high efficiency and yield such cell, and we end describing the advancements in the field of
as hydrogen production (Gibson et al. 2008b; Glass et al. 2009; bioinformatics. Whenever possible, the implications and
Lartigue et al. 2009). We are now able to not only change small advancements of these post-genome technologies to biotechnol-
parts of a genome but also to create an entire new entity with the ogy will be discussed. A schematic form of how these themes are
sole purpose of providing us with a very specific function – the connected can be found in > Fig. 14.1.
basis of biotechnology (Lartigue et al. 2007; Gibson et al. 2008a,
2010; Benders et al. 2010).
One important achievement obtained over the last two Application of Prokaryotic Intraspecies
decades is the development of ‘‘culture-independent’’ techniques. Genomic Diversity
Initially fingerprinting techniques such as DGGE, RISA, T-RFLP,
DNA-microarrays, among others, targeting gene markers as the Since the landmark of 1,000 bacterial genomes in 2010, and 100
16S rRNA gene (one of the most frequently used genes) allowed archaeal genomes in 2011, several strains belonging to the same
the detection of shifts in the microbial communities according to species were sequenced revealing that the intraspecies genetic
changes in the environment, time, and perturbations such as diversity of prokaryotes was higher than previously thought
pollution. Soon after, metagenomics (the study of a pool of (Lagesen et al. 2010; Brochier-Armanet et al. 2011). The geno-
environmental genomes) became more popular, allowing the mic analysis of multiple strains revealed that the number of
metabolic potential of the microbial community to be probed genes belonging to a single bacterial species was larger than the
and increased the understanding of the functional aspects of these number of genes present in one strain, leading to the concept of
communities (Rondon et al. 2000; Lopez-Garcia and Moreira a pan-genome for bacteria (Medini et al. 2005; Tettelin et al.
2008; Chistoserdova 2010; Chistoserdovai 2010; Fernandez- 2008; Mira et al. 2010).
Arrojo et al. 2010; Simon and Daniel 2011). The pan-genome is the sum of all genes present in members
Growing a microorganism in the laboratory has its advan- of one taxonomic group; initially, it was used to describe the
tages. For example, if an organism is already naturally capable of repertoire of genes shared by one species but later was applied to
producing a certain compound, it is expected that this organism other taxonomic groups, such as genus, and even for all bacteria
can cope with its production. For instance, a microorganism (Lefébure and Stanhope 2007; Tettelin et al. 2008; Lapierre and
producing an antibiotic compound, invariably, will not succumb Gogarten 2009; Ussery et al. 2009; Lukjancenko et al. 2010; Mira
to its expression, whereas heterologous expression will certainly et al. 2010). This concept was first used in 2005 in a study on
need several adaptations to avoid killing the host. However, it is Streptococcus agalactiae strains, 10 years after the first bacterial
well known that several microorganisms that produce com- genome was published. Eight genomes of S. agalactiae were
pounds of biotechnological interest are not yet cultivated. Now, compared in order to determine the set of genes shared by all
with the development of single cell genomics, we have the ability strains for further development of a universal vaccine against
to locate a specific cell in a complex microbial background, to Group B Streptococcus (GBS) (Tettelin et al. 2005). The average
isolate it, and sequence its genome (Walker and Parkhill 2008; number of identified genes was 2,200 genes per genome; 1,800 of
Siegl et al. 2010; de Jager and Siezen 2011). With its genome these genes were present in all strains studied. In other words,
information in hand, we can now ask why this microorganism is approximately 80 % of the genes were part of the S. agalactiae
not able to grow in standard laboratory culture medium and common core genome. However, the estimations on the number
adapt the growth conditions accordingly to culture it. of genomes required to fully describe S. agalactiae pan-genome
Regardless of one´s scientific interest, one major concern would be over hundreds, since it was estimated that for every
today is the incredible amount of information pouring from extra genome added to the analysis, an average of 33 new genes
hundreds and hundreds of next generation sequencing (NGS) would be revealed (Maione et al. 2005; Tettelin et al. 2005). The
machines every single day worldwide (Gomez-Alvarez et al. analysis of this pan-genome indicated a series of putative anti-
2009; Kunin et al. 2009; Morgan et al. 2010; Schwartz et al. gens that could be used to generate a vaccine against GBS.
2011; Suzuki et al. 2011). This amount of information over- Among them only four were considered good antigens, and
whelms the most prepared bioinformaticians, not to mention although they were effective against a wide range of strains,
the poor microbiologist trying to understand his single favorite none alone could be considered a universal vaccine (Maione
bug in the laboratory. Trained personnel, infrastructure for et al. 2005; Muzzi et al. 2007).
processing, storing and analyzing data have become some of the Similar to the S. agalactiae pan-genome study, clinical appli-
major bottlenecks in the field of genomics and post-genomics cations were among the objectives of a study comparing the
(Several Authors 2008; Horner et al. 2009; Schadt et al. 2010). genomes of 13 pathogenic and nonpathogenic strains of
This chapter will cover all the aspects mentioned above of Haemophilus influenzae. In this case, the goal was the develop-
how the information acquired by genome sequencing has ment of a hybridization chip to identify clinical pathogenic iso-
impacted microbiology and what are the emerging post- lates at low cost. The H. influenzae pan-genome was estimated to
genomic technologies. Initially, we will discuss prokaryotic be between 4,425 and 6,052 genes, which is at least three times
intraspecies genome diversity. Next, we will cover the diversity larger than the number of genes that a single strain may contain.
at the community level and how this can be exploited, then we The core genome was estimated to contain only 51 % of the
will discuss the sequencing of the genome from an individual genes (Hogg et al. 2007).
Genomes and Post-genome Technology 14 331
. Fig. 14.1
Schematic of the main topics covered in this chapter. The genomes of microorganisms present in the environment can be easily
sequenced, regardless if we are able to cultivate them in the laboratory, if they are present as part of a community, or if we can only
separate them as single cells. Bioinformatics is crucial for the organization and analyzes of genome sequencing data. The metabolic
wealth present in these microbial genomes is being used for the development of products such as enzymes with applications in various
industries
It is only natural to think about the Escherichia coli pan- approximately three times the average number of gene families
genome; after all, this is the bacterial species with the largest described for Escherichia coli 0157:H7 str. EC4196i (Rasko et al.
number of strains with genome sequenced. A study on the pan- 2008; Lukjancenko et al. 2010).
genome of E. coli using 51 strains showed that, considering Species such as S. agalactiae, H. influenza, and E. coli are
an average size of 5,000 genes for a genome, approximately considered to have ‘‘open pan-genomes’’ which means that every
20 % of the genes are shared among all strains. A previous time a genome is added to the analysis, new genes, or genes not
work using 21 strains had suggested that the core genome, previously found in any strain, will be identified. Some species
defined by the presence of a gene in all strains, contained roughly present a ‘‘closed pan-genome’’ indicating that the number of
50 % of these genes. The E. coli pan-genome is currently esti- new genes to be identified by the addition of a genome is zero.
mated to contain 13,926 gene families, which correspond to That is the case of Bacillus anthracis, Staphylococcus aureus, and,
Ureaplasma urealyticum (Tettelin et al. 2008). B. anthracis was would belong to the same species (Konstantinidis et al. 2006;
the first example of a closed pan-genome since the analysis of Goris et al. 2007). It should be mentioned that the relationship
eight genomes showed that the number of new genes added to between ANI and 16S rRNA gene phylogeny seems to be strong
the pan-genome is zero as the fourth genome was added. In fact, as well; in general, if two species show higher than 97 % simi-
B. anthracis pan-genome was considered an extreme case of larity in the 16S rRNA gene nucleotide sequence, they will
a closed pan-genome and it was previously described that these present higher than 95 % ANI (Konstantinidis and Tiedje
strains are a clonal population derived from Bacillus cereus that 2005). Although it may seem that there are many available
carries the anthrax toxin plasmid (Tettelin et al. 2005). sequences of bacterial genomes, they represent a small fraction
It is important to understand that the models used for each of the bacterial diversity and to use genome data for defining
pan-genome analysis were distinct depending on the study, and bacterial species seems premature.
in some cases, the interpretation whether a pan-genome should Another important observation of intraspecies variation is
be considered open or closed may differ among authors; simi- that, in general, bacteria with larger genomes tend to show
larly, the choice of methods for analysis on comparative geno- higher intraspecific variations, and it has been proposed
mics is far from being standardized (Hogg et al. 2007; Tettelin that free-living species will present a larger pan-genome
et al. 2008; Klenk and Göker 2010). Whether or not these terms (Konstantinidis et al. 2006). The pan-genome of Burkholderia
will be largely used to define the intraspecies diversity in bacte- pseudomallei and other Burkholderia was estimated and, as
rial genomes is yet to be determined. observed for E. coli, the B. pseudomallei pan-genome was esti-
The core genome usually comprises house-keeping genes mated to be three times larger than a typical genome, based on
involved in cell envelope synthesis, regulation of gene expres- a strain average of 5,000 genes. However, in this case, the core
sion, and protein transport; however, several genes are ‘‘hypo- genome corresponds to 80–85 % of one strain genome. For
thetical proteins’’ or ‘‘proteins of unknown function’’ (Tettelin B. mallei, the core genome was estimated to comprise 58 % of
et al. 2005; Mira et al. 2010). Genes that are present in some the genes present in all strains, although the genome is smaller
strains, but not all, are known by different designations: probably because it is a pathogen whereas some B. pseudomallei
‘‘expandable genome,’’ ‘‘distributed genome,’’ or ‘‘accessory strains are free-living bacteria (Konstantinidis et al. 2006; Ussery
genome’’; and usually contain a large number of transposable et al. 2009).
elements. Some authors prefer a more relaxed concept for a core The majority of multi-strain genomic analysis focused on
genome, for example, a gene should be present in at least 99 % or pathogenic bacteria, not only for the clinical applications but
95 % of the strains, which maybe more useful for taxonomic because these are the ones with the highest number of cultivated
purposes (Konstantinidis et al. 2006; Lapierre and Gogarten representatives. Some studies on intraspecies variation included
2009; Kislyuk et al. 2011). pathogenic and nonpathogenic bacteria, but there are few
On the other hand, the study of the accessory genome including only free-living microorganisms. Archaeal pan-
recognized the extent of the importance of horizontal gene genomes are rare and often these studies overlap with
transfer (HGT) to bacterial evolution. It was first believed that metagenomic analysis. For example, comparative genomic anal-
the major source of evolutionary variation in bacteria would ysis of Haloquadratum walsbyi, a haloarchaeon common in
come from spontaneous mutations, the present number of saltern crystallizer ponds, showed that this species presents low
genomes available demonstrated that HGT is one of the major genetic variability among strains isolated from geographically
contributors to the genetic variability found in prokaryotic distant ponds as far as Spain and Australia. In fact, H. walsbyi
genomes. The majority of sequences present in the accessory seems to be the only species in this genus and strains vary in their
genomes are involved in GHT processes, they are genomic 16S rRNA gene sequences by less than 2 %. In one study com-
islands, phage-related sequences, and transposon-related prising sequences from three distant ponds located in Australia,
sequences (Tettelin et al. 2005; Doolittle and Papke 2006; Ehrlich almost all sequences present in the genome of H. walsbyi type
et al. 2010). species were retrieved from the collective genome of the
Initially, it was proposed that a core set of genes could be microbes inhabiting a saltern pond. It was estimated that the
used to define a bacterial species (Lan and Reeves 2001); how- core genome would contain 84 % of the genes found in the type
ever, for the reasons explained above, defining a core genome is species. This low level of variation is not a common feature
neither a simple task nor a standardized procedure. Studies have identified in halophilic archaea; other genera inhabiting saltern
compared genomic analysis of multiple strains with other ponds show a higher number of species, but their pan-genomes
methods for typing such as ‘‘multilocus sequence typing’’ or to are yet to be studied (Bolhuis et al. 2006; Legault et al. 2006;
the ‘‘gold standard’’ method to define bacterial species: DNA- Cuadros-Orellana et al. 2007; Oh et al. 2010; Dyall-Smith et al.
DNA hybridization (DDH) or the ‘‘average nucleotide identity’’ 2011).
(ANI) (Konstantinidis and Tiedje 2005; Konstantinidis et al. The pan-genome concept was created to improve the iden-
2006; Deloger et al. 2009; Hall et al. 2010). It was previously tification of clinically relevant bacteria and to develop new
demonstrated that ANI can be used to determine the level of strategies for pathogen control, but the striking discovery of
relatedness between two bacteria and that this index correlates the intraspecies genetic variation has contributed to bacterial
nicely with DDH, in general two strains that present higher than ecology and evolution in at least in two ways: (1) it has led to an
95 % ANI will present higher than 70 % DDH; therefore, they in-depth debate on a genome-based definition for prokaryotic
. Fig. 14.2
Number of publications in PubMed that presented the word ‘‘pan-genome’’ in the title or abstract from 2005 to the present (July 2011)
species, and (2) it has shown the extent and importance of DNA libraries that are then screened by various means to reveal
horizontal gene transfer to prokaryotic evolution. Other appli- clones with a biological activity of interest.
cations of this knowledge may come in the future. In fact, The first step to construct a metagenomic DNA library is to
the number of publications that include the word pan-genome obtain total microbial community DNA. Direct DNA extraction
has increased every year (> Fig. 14.2), demonstrating that allows both the cultivated and uncultivated components of
microbiologists find this concept useful enough to continue a microbial community to be represented in DNA libraries,
applying it. and these can be exploited for biotechnological purposes.
While some studies describe the selection of their samples for
DNA extraction and library construction based on the overall
Culture-Independent Exploitation of microbial diversity, others have attempted to exploit the natural
Microbial Genomes: Metagenomics enrichment of a specific microbial community – sometimes over
long evolutionary time – to a particular environment. For exam-
A major challenge in post-genome biology is to uncover gene ple, one can look for enzymes that can withstand high temper-
function and assemble together gene sets into useful metabolic atures in hot spring microbial communities (Rhee et al. 2005),
pathways for biotechnological purposes. Bioinformatics has for esterases that work in cold conditions in the arctic soil
been instrumental for the organization of the avalanche of (Yu et al. 2011), for cellulases and other enzymes involved in
genome sequence information so that gene families and specific deconstructing plant cell walls in the microbial communities of
domains are identified. However, there are still single genes, gene the gut, rumen, and feces of various vertebrate and invertebrate
families, and entire gene clusters for which there are no leads animals (e.g., cow (Zhao et al. 2010), wallaby (Pope et al. 2010),
regarding their biological function, despite bioinformatics ana- termites (Liu et al. 2011), earthworm (Beloqui et al. 2010)) and
lyses. Therefore, there is no substitute for lab bench work in this for bioremediation enzymes in the microbial communities of
post-genome era. There are many studies published in which the herbicide contaminated soil (Lu et al. 2011). However, knowl-
function of a specific gene has been uncovered by mutational edge about bacterial physiology is warranted when choosing
analysis (e.g., targeted mutagenesis, transposon mutagenesis) a specific microbial community for a particular project. For
and comparison with the wild type strain under a number of instance, many acidophiles have a circumneutral internal pH
growth conditions. due to a variety of homeostasis mechanisms (reviewed in
A different approach focuses on the biotechnological appli- (Baker-Austin and Dopson 2007)), and therefore a microbial
cation rather than on any specific gene and is common in the community from a low pH environment would not necessarily
metagenomics field. As previously mentioned, the term be an appropriate source of enzymes tolerant to low pH.
‘‘metagenome’’ here is used in its original context referring to Although there are many interesting environments to be
the collection of the bacterial genomes from a microbial com- explored, there are technical difficulties to obtain microbial
munity (Rondon et al. 2000). In the metagenomics approach, DNA for library construction. Each environmental sample will
total DNA of the microbial community of choice is directly bring different challenges. Some microorganisms may be hard to
extracted from the environment and can be used to construct lyse. When extracting DNA to construct an expression library
from the cow’s rumen, one needs to ensure that DNA from the kept by leaf-cutter ants from Panama (Suen et al. 2010). These
plant fraction is not overrepresented in the library. When ants harvest fresh leaves to cultivate a fungus – which can be
the sample of choice is water, the scarcity of microbial cells in considered a form of farming – that is its primary food source.
the sample may require concentration by filtration prior to Pyrosequencing of the community metagenome followed by
processing. When dealing with low biomass samples, it may a carbohydrate-active enzyme characterization was performed
also be necessary to amplify the small amount of extracted and 28 families of glycosyl hydrolases, carbohydrate esterases,
DNA before library construction by multiple displacement and polysaccharide lyases were identified. The presence of these
amplification (MDA). Sampling and obtaining replicates may enzymes able to degrade cellulose and hemicellulose was most
also require special attention in projects dealing with aquatic similar to the enzyme profile of the bovine rumen, which can be
microbial communities as the species composition of these used as evidence that the fungus garden can be considered an
environments may vary significantly according to the depth, external digestive system (Suen et al. 2010). Similar work has
season, currents, and temperature in which the water samples been performed to study the cow rumen and the human feces
were collected. For other environments, contaminants that may microbiota (Brulc et al. 2009; Arumugam et al. 2011).
degrade DNA and that are incompatible with downstream Another approach to perform a functional study of
molecular biology applications are the main challenge. Humic a microbial community is to sequence the metatranscriptome
acids that are co-extracted with DNA inhibit enzymatic reaction (i.e., sequencing the environmental cDNA derived from
and are a common problem when dealing with soil samples. Clay mRNA). Although the mRNA complement of a microbial com-
particles can absorb organic matter including DNA and interfere munity reflects more accurately its metabolic functions, there
with DNA extraction (Purdy 2005). High iron content in soils are fewer studies using this approach than the metagenomic
can lead to DNA breakage by Fenton reaction in which hydroxyl approach due to the technical difficulties associated with
radicals are produced from the reaction of iron with hydrogen mRNA extraction from the environment. Interestingly, compar-
peroxides (Harrison and Arosio 1996). Therefore, there is no ison between the total microbial community (metagenome) and
universal protocol to obtain microbial DNA from the environ- the most active community (metatranscriptome) may reveal
ment; new protocols need to be developed or old ones from differences such as those found in the coniferous forest soil
similar environments optimized for each new environment where low abundance fungal species seem to be highly active in
explored (de Castro et al. 2011). plant decomposition (Baldrian et al. 2011). Furthermore,
Once metagenomic DNA is obtained in high enough quan- metatranscriptome data has been used to design a growth
tity and quality (e.g., not degraded and free of inhibiting sub- medium for the cultivation of a Rikenella-like bacterium that
stances), three main approaches can be taken to study gene inhabits the medicinal leech gut, a previously uncultivated and
function. Firstly, traditional technologies or NGS can be used the most abundant endosymbiont of this microbiome (Bomar
to sequence metagenomic DNA that can be assembled into near- et al. 2011).
complete or partial individual genomes when the complexity of Although functional analysis of genes identified in the
the community is low. This has been accomplished for different microbial community of interest through sequencing and com-
microbial communities including that of an acidophilic biofilm, parison with databases can be quite useful, as previously
the nutrient-limited open ocean in the Sargasso sea, and anaer- discussed, these sequence-based approaches will not identify
obic ammonium oxidation bioreactor (Tyson et al. 2004; Venter genes that are not similar to previously known genes. A third
et al. 2004; Strous et al. 2006). Once assembled, the genome is sequence-independent approach to study gene function within
analyzed in various ways and annotated. The genes identified are a metagenome starts with the construction of metagenomic
categorized according to their putative functions such as signal DNA libraries. These can be used to screen for the activity of
transduction, transcription, energy metabolism, and unknown interest. Indeed, this approach has allowed the identification of
function. New genes such as the 782 rhodopsin-like photore- enzymes with low similarity to known enzymes and even
ceptors found in the Sargasso Sea work can be identified (Venter a new family of enzymes (Beloqui et al. 2006; Kim et al. 2009;
et al. 2004). Information about which biochemical pathways are Pang et al. 2009; Park et al. 2011).
present is obtained and sometimes new ones can be deduced. An For metagenomic DNA library construction, different vec-
example of the latter is a new metabolic pathway for anaerobic tors including plasmids, phages, fosmids, cosmids have been
ammonium oxidation, which is a sink to fixed nitrogen in the successfully used. Most works have used E. coli as an expression
oceans (Strous et al. 2006). These metabolic genomic analyses host, although there are reports of libraries constructed in
can also be useful to understand how microorganisms adapt to a-proteobacteria (e.g., Agrobacterium tumefaciens, Caulobacter
specific environments and to understand the role a particular vibrioides, Rhizobium leguminosarum), b-proteobacteria
microorganism plays in a microbial community. Microbial (e.g., Burkholderia graminis, Ralstonia metallidurans), and
diversity as well as the species richness present in the original g-proteobacteria (e.g., Pseudomonas fluorescens, P. putida,
sample can also be estimated. Xanthomonas campestris) (reviewed in Taupp et al. 2011). The
More commonly, sequencing of metagenomic DNA will choice of insert size for construction of expression metagenomic
allow the analysis of gene function without the individual libraries will affect how intense the library will need to be
genomes present being assembled. An interesting example of screened, as the same number of clones from a large insert
this approach was the study of the microbiota of fungus gardens library will represent more metagenomic DNA if a large insert
library is being screened than a small insert library. However, and thus requiring less harsh chemicals to be used. In this case,
once a clone showing the activity of interest is identified, it is the original xylanase obtained from soil environmental samples
harder to identify the gene responsible for the activity in large was mutagenized to improve its thermostability (Dumon et al.
insert clones. 2008). One mutant with changes at multiple sites had a Tm 25 C
A metagenomic DNA library can be submitted to essentially higher than the parent enzyme, while retaining similar catalytic
three different types of functional screens to identify clones with properties (Dumon et al. 2008).
the biological activity of interest: (1) direct activity screens, As with any method, function-based screening of
(2) promoter activity screens, and (3) selection of clones by metagenomic DNA libraries has limitations. The first hurdle that
complementation. Direct activity screens are still the most com- needs to be overcome is to get the proteins expressed. As previously
mon. They are well suited for the identification of activities with mentioned, different hosts with different codon bias and cellular
biotechnological application such as antibiotic, amylase, ami- machinery for protein expression may be useful. There are patents
dase, nitrilase, oxidoreductase, esterase/lipase, cellulase, for broad-host range vectors that will function in different hosts.
xylanase, pectinase, b-glucosidase, protease, agarase, chitinase, Furthermore, it is important that the host strain is devoid of
and phytase (reviewed in Simon et al. 2009). Some of these proteases that may hydrolyze the protein one is interested in
screens rely on the hydrolysis of the substrate molecule present expressing. Another problem is the low level of expression of clones
in the solid growth medium (e.g., starch for amylase, milk which may lead to false negatives. This problem can be partly
protein for protease) and thus formation of a clear halo around addressed by optimizing assay protocols such as by extending
positive clones. Other screens are based on the hydrolysis of incubation times to increase sensitivity. The opposite problem
a synthetic substrate added to the growth medium that mimics where a clone will initially give a positive result that cannot latter
the natural substrate releasing a molecule that can be detected by be reproduced (i.e., false positives) is also common. Furthermore,
a change in fluorescence when excited by the appropriate wave- when dealing with the functional diversity and metabolic com-
length (e.g., 4-methylumbelliferyl-b-D-glucoside, methylum- plexity of metagenomes in a screen, one may discover that the
belliferyl-b-D-cellobioside). Some screens such as those for method used to target a particular activity is not as specific as
cellulase activity where halos around positive colonies are previously thought. For example, in our lab (Quirino et al.,
sought for in soluble cellulose (CMC) containing-plates unpublished data), we have found that halos around clones plated
followed by Congo red staining are popular because they are in starch-containing medium and staining with I2 vapor may lead
inexpensive. Although they are labor intensive as they are not only to clones with amylase activity but may also reveal
destructive, requiring that the clones are stored prior to the clones with amydase activity, an entirely different enzyme class.
assay or that a replica of each plate is made, the local accumu- Promoter activity–based screens have been designed to iden-
lation of an enzyme at a high concentration may facilitate tify novel catabolic genes based on the induction of operons by
detection of certain enzyme activities. Liquid-based screens are a substrate of choice. This technique is called substrate-induced
also used and have the advantage of being quantitative. gene expression screening or SIGEX (Uchiyama et al. 2005) and
Screens using solid as well as liquid growth medium can be is based on the fact that in many cases, catabolic gene expression
automated for high-throughput screens of metagenomic librar- is generally trigged by a substrate and that catabolic genes are
ies. Colony picker robots, liquid handlers, and integrated spec- often part of operons. Therefore, a system was developed where
trophotometers are commercially available. These can be used operons from metagenomic origin are inserted upstream of the
on 96- or 384-well format microplates. Robotization of screen- reporter gene GFP. Clones in liquid culture from such libraries
ing procedures allows high-throughput experiments. There are are screened in the presence of the desired inducer molecule
a few examples of the success of this strategy from the company using fluorescence-activated cell sorting (FACS). This procedure
Verenium (U.S.A.). One such example is the alpha-amylase allows the selection of positive clones in high throughput.
Fuelzyme®, used commercially for the production of ethanol Although promising, there have been few examples of the
by corn wet milling process. Fuelzyme® is the chimeric product biotechnological exploitation of metagenomes for the discovery
obtained by direct evolution of three archaeal enzymes, each of new regulatory elements per se rather than enzymes. For
with different properties, obtained by screening metagenomic example, constitutive promoters can be identified by cloning
DNA from the depths of the ocean floor (Richardson et al. metagenomic DNA fragments upstream of promoter-less
2002). Another example of a metagenomic DNA library reporter genes such as GFP or DsRed and screening for the
derived-product is a phospholipase C commercialized under reporter gene activity (Han et al. 2008). One of the possible
the name Purifine® PLC. This enzyme is used in the edible oil problems in metagenomic screens may be the insufficient rec-
industry for the removal of the phosphorus from phospholipids. ognition of promoters, and one application of this technology is
High phosphorus oils are present in different species such as to identify new promoters from metagenomic DNA to construct
soybean and canola. The removal of phosphorus is part of the oil new expression vectors.
refining process (www.verenium.com). Yet another example of A variation of function-based screens is targeted screens for
a product from metagenomic origin is a xylanase used in a particular activity by complementation of a cell mutant. In this
the paper industry commercialized under the name Luminase® approach, borrowed from traditional microbial genetics, the
PB-100. This xylanase hydrolyses hemicellulose making the pulp metagenomic library clones are used to complement the pheno-
bleaching process (i.e., process used to make paper whiter) easier type of a mutant of interest. When the phenotype is no growth
under specific conditions and this condition can be restored by Single Cell Genomics
metagenomic clones, a selection rather than a screen can be
performed. Phenotypic selection is a powerful method for find- The metagenomic approach has revolutionized microbiology as
ing an activity of interest as it allows much greater numbers of it has widened its boundaries revealing a whole new world of
clones to be analyzed than traditional screens. This approach has microorganisms. However, access to the rare components of the
been now used to find a number of gene activities from community is still a challenge. It is not known what the impor-
metagenomic libraries (Majernik et al. 2001; Riesenfeld et al. tance of these low abundance species in the community is but it
2004), many of which with biotechnological applications. Such is believed that if environmental conditions change, these species
approach was used to obtain a xylose isomerase from a garden may become dominant in the community (Sogin et al. 2006).
compost metagenomic library (Parachin and Gorwa-Grauslund Therefore, these species can be thought of as genetic reservoirs of
2011). The plasmid library was used to complement an E. coli biological functions. While next generation sequencing of 16S
strain that had a nonfunctional xylA gene to restore growth in rRNA gene will reveal the presence of these rare species, single
xylose-containing medium. Xylose is present in hemicellulose, cell genomics – the ability to obtain a full (or partial) genome
and for complete fermentation of plant biomass to make cellu- derived from a single cell – provides a strategy to access the gene
losic ethanol by yeast, it is necessary that the yeast strain be able content of a cell in a targeted fashion.
to ferment xylose. From a practical standpoint, the first step in obtaining whole
DNA polymerases have a number of applications in molec- genome information from a single cell is isolating it. There are
ular biology such as DNA sequencing, probe labeling, and various methods published in the literature that describe isola-
mutagenic PCR. Nine new genes encoding DNA polymerase tion of single microbial cells. Traditionally, single cells have been
I proteins or domains were identified by complementation of isolated by dilution to extinction (Button et al. 1993). Although
the E. coli polA mutant with metagenomic library clones derived this method has a great advantage for certain cultivation pro-
from glacial ice (Simon et al. 2009). cedures (Connon and Giovannoni 2002), it is seldom applied to
Novel polyhydroxyalkanoate (PHA) synthase genes that may amplify the genome of a single cell since it routinely operates
be used industrially to produce biodegradable plastics were with too large of a volume which is incompatible with down-
identified from a soil metagenomic library by complementation stream amplification procedures. More suitable methods to
of Sinorhizobium meliloti PHA synthesis mutants (Schallmey obtain a single cell for subsequent genome amplification include
et al. 2011). The metagenomic library was transferred to optical tweezers and microfluidics, laser-capture microscopy, as
S. meliloti by triparental conjugation and the screens for PHA well as fluorescence-activated cell sorting (FACS) (Porter et al.
synthase were based on the restoration of mucoid phenotype or 1993; Frohlich and Konig 1999; Ferrari et al. 2004; Ishoy et al.
Nile red staining on YM agar. 2006; Marcy et al. 2007; Taylor et al. 2009). Methods that employ
Although function-based screens are the method of choice a microscope, such as microfluidics and laser-capture micros-
when one is interested in finding new classes of proteins, one can copy, have the advantage that the cell can be visualized before
also use the library to screen for specific activities in sequence- amplification where obvious contamination will easily be spot-
based screens. Depending on how the sequence-based screen is ted. Furthermore, effects of environmental perturbations can be
designed, entirely novel proteins can also be identified using this observed in real time. Based on its high-throughput nature,
approach. For instance, a new blue-light sensitive protein was FACS has an advantage when underrepresented microbes within
uncovered in a microarray-based screen of 2,500 clones of a soil a community are being targeted. A combination of FACS with
metagenome cosmid library in which the microarray contained fluorescence in situ hybridization (FISH) allows for the targeting
oligonucleotides derived from LOV (light, oxygen, and voltage) of organisms that represent only a small fraction of the total
domain consensus sequences of blue-light photoreceptor genes community (Podar et al. 2007).
(Pathak et al. 2009). Following successful single cell isolation, nucleic acids can be
Once positive clones are obtained in functional screens, the amplified. Amplification can target a set of genes or the whole
gene responsible for the phenotype needs to be identified. This genome. A set of genes can be amplified (e.g., by digital PCR) to
can be accomplished by sequencing the insert by traditional screen for rare variables within a population, to link a particular
(Sanger) or high-throughput sequencing methods and organism to a particular function, or for example to study viral
subcloning of open reading frames for expression followed by infection patterns within a population (Ottesen et al. 2006; Zeng
testing for the activity of interest. This can be quite laborious if et al. 2010; Tadmor et al. 2011). Whole genomes from single cells
one is dealing with large fragments. However, transposon muta- can be amplified using multiple displacement amplification
genesis can be used to narrow in to the gene (Wang et al. 2006). (MDA) (Dean et al. 2001). Due to the nature of MDA, the
Screens are constantly evolving and allowing new proteins to coverage for microorganisms is typically less than 100 %
be identified and studied. As more high-throughput and mini- (Woyke et al. 2010) and can vary substantially depending on
aturized screens are developed, more products of metagenome the organism (Raghunathan et al. 2005; Zhang et al. 2006; Marcy
origin will certainly emerge. However, for more businesses based et al. 2007; Mussmann et al. 2007; Podar et al. 2007;
on the genetic wealth of metagenomes to become lucrative, not Stepanauskas and Sieracki 2007; Rodrigue et al. 2009; Woyke
only technical problems need to be solved, but it will necessary et al. 2009; Blainey et al. 2011). Similar to accomplishments for
to be in tune with the market. Archaea and Bacteria, whole genome amplification has
successfully been applied to gain insights into the genomes and symbiotic or other host-associated microorganisms, where
lifestyles of eukaryotic organisms, such as marine protists (Hey- transcription and translation of a biomolecule can be
wood et al. 2011; Yoon et al. 2011). tightly correlated with the host physiology, making it even
Subsequent sequencing of amplified genomic DNA can be harder to decipher by traditional methods. For example, the
performed with various sequencing and next generation isolation and structure determination of bryostatin (i.e.,
sequencing platforms. Depending on sequence quality and a compound which demonstrated potential in treating certain
length, the genome assembly relies on computational tools to types of cancer and Alzheimer’s disease), the link of its biosyn-
stitch the genome together (Nagarajan et al. 2010; Chitsaz et al. thesis to a bacterial endosymbiont, and the identification of
2011). This genetic information is providing insight into the genes involved in the synthesis took almost two decades (Sharp
diversity that exists on the genomic level within a natural pop- et al. 2007).
ulation. We learned from pure culture studies that cell individ- After a biologically active compound is identified, it is nec-
uality can be much larger than originally expected. Studying the essary to decide between two major options in order to produce
evolution of bacterial genomes in the laboratory confirmed that it. The first includes a robust microorganism (mainly
microbial genomes exhibit an enormous plasticity (Portnoy Escherichia, Bacillus, Pseudomonas, or Saccharomyces) that is
et al. 2011). Single cell genomics now provides a suitable tool genetically engineered to produce a desired product. The second
to evaluate this phenomenon in the natural environment. How- option takes advantage of an organism that possesses a desired
ever, cell individuality is often controlled on the transcriptional biosynthetic pathway which can subsequently be isolated
and translational level (Zengler 2009). Deciphering cell individ- and propagated in the laboratory. Although less common for
uality on the transcriptional level for single cells thus far requires most industrial applications, the isolation of new organisms can
conversion of mRNA to cDNA and subsequent amplification. become necessary, especially when synthesis of the biomolecule
Although progress has been made studying the transcriptome of is encoded in complex pathways and spread throughout
a single cell for eukaryotes (Tang et al. 2009; Taniguchi et al. the organisms’ genome, as is often the case in natural products.
2009), mRNA-seq for single bacterial cells is still in its infancy Having established production strains with a long history
and has proven more difficult because of the reduced amount of in industrial settings, however, is greatly advantageous over
mRNA that can be obtained from bacterial cells (Taniguchi et al. newly isolated microorganisms when it comes to producing
2010). However, recent developments have made it possible to biomolecules for applied purposes, since critical factors, such
amplify mRNA from a single bacterial cell, thus enabling single as scale-up processes and phage resistance, have already been
cell transcriptomic studies for bacteria (Kang et al. 2011). evaluated.
Where metagenomic and other top-down approaches cast As stated before, a metagenomic approach can be used to
a wide net to obtain as much information from the environment obtain sequence information or, in case of expression libraries,
as possible, single cell genome sequencing can be considered to screen for a desired trait. Using recombinantly generated
a bottom-up approach in which detailed information for single expression libraries has been shown to be successful for individ-
organisms is stitched together to generate the bigger picture ual enzymes (Robertson et al. 2004), but it is more challenging
(Zengler 2009). for whole biosynthetic pathways because various enzymes
involved in a pathway can be located on different parts of
a genome and therefore will not be represented in a single
large insert clone. In principle, it is possible to express any
Using Genome and Post-genome Information particular enzyme or whole pathways recombinantly in
to Make a Product a production host. Regarding the latter, the field of metabolic
engineering has made great advances in understanding meta-
Genomes and post-genomic tools are at the heart of any genetic bolic processes and using these for obtaining this product. This
or metabolic engineering effort. The genes involved in the syn- can be achieved by modifying a microorganism by recombinant
thesis of molecules of biotechnological interest can be identified DNA methods in various ways such as by removing competing
in the genomes of cultivated or uncultivated microorganisms or networks, adding new ones to feed or to speed up the major one,
from metagenomes. After the desired trait is linked to a gene or and adding new enzymes to create new products (> Fig. 14.3). In
a pathway, the next step is to find a strategy to obtain the product this approach, it is important to monitor the yields of the desired
in commercially compatible quantities. product and the other components of the metabolic network
Several examples from the natural products field show that and make the necessary fine adjustments to improve it even
a desired chemical entity found in nature often cannot be further. Usually, it is not possible to predict the overall behavior
retrieved in large enough quantities to make it a viable com- of the complete network. Therefore, it is important to obtain
mercial product (Newman and Cragg 2005). Natural products information about the final product yield, as well as about other
are often constituted of complex chemical structures which are systems’ components to identify bottlenecks. Modifications in
difficult to synthesize organically in a cost-effective manner the engineered metabolic pathway can then be proposed and
(Wender and Miller 2009). Identifying the organism that tested, starting another cycle of improvements. As expected,
produces it, its chemical structure, and the genes encoding for this procedure can be very time and resource consuming. It
its synthesis can be a long-term project. This is true especially in may be necessary, for example, to reverse enzymatic reactions,
Inhibitors Activators
in/out
E E transporters
Enzyme concentration
Proteins or
metabolites
1 2 1 2
Feedback
Transfer pathway
2 1
3
Redirecting
metabolites flow
3 3
Pathway complementation
4
3
2
Subunits expression or cofactors P
4 4 New products
1 and activities
Channeling pathway
P Product stability
Energy
Product activity
Grow inhibition or product lethality
. Fig. 14.3
Schematic of some network modifications accomplished by metabolic engineering. Red rectangles and the purple circle (‘‘P’’) represent
the synthetic pathway and the product of interest. The blue circle (‘‘E’’) represents the first enzyme of this pathway. The magenta
rectangles represent regulators/activators necessary for the production of new enzyme subunits or cofactors responsible for changing or
enhancing the product P activity. The green rectangles represent an entirely new pathway transplanted to the host to cross-feed directly
or indirectly into the main synthetic pathway. Pink squares represent a new added pathway to modify the final product further as
a consequence producing a new one (black circle ‘‘P’’). Red and yellow circles represent pathways added or enhanced to increase the
energy (yellow stars) availability
synchronize transcription and translation levels, change DNA Sequencing and Bioinformatics
a particular enzyme for another with more suitable kinetic
properties, or avoid the undesirable formation of inclusion Working with Large Amounts of Data
bodies. An example of a successful metabolically engineered
product is the production of amorpha-4,11-diene, the precursor The advent of next generation sequencing technologies (NGS)
of artemisinin which is used in the treatment of malaria by resulted in an unprecedented gain of resolution in genomic and
metabolic engineering (Tsuruta et al. 2009). metagenomic experiments as a result of orders of magnitude
. Table 14.1
Storage requirements at two stages of data analysis, for the different sequencing technologies
Roche/454 Illumina/Solexa ABI/SOLiD

Data units: Gigabyte (GB)a GS20 FLX Ti I II IIx 1 2 3
Stage 1 images 10 10 30 500 1,100 2,800 1,800 2,500 1,900
b
Stage 2 sequence reads 0.5 1 4 30 50 100 140 600
a
The first row of numbers at each stage refers to fragment data, while the second row is for paired-end library data
b
Data are not available (Source: Genome Center at Washington University, Dooling 2010, http://www.politigenomics.com/next-generation-sequencing-
informatics)
sampling capacity. One generally overlooked aspect is that the acquiring the sequence is only the first step, and must be
amount of data generated by NGS surpasses processing capabil- followed by large-scale computational analysis to process the
ities of conventional desktop computers and that a sizeable data, test hypotheses, and draw scientific insights. Therefore,
computational infrastructure is necessary to manipulate giga- investment in a sequencing instrument must be accompanied
bytes of raw data. Three data generation and transformation by substantial investment in computer hardware, skilled infor-
stages are present during genome sequencing: The first involves matics support, and bioinformaticians competent in configur-
collection of the light intensity values captured by the image ing and using specific software to analyze the data.
sensor of the instrument and emitted by the fluorescent nucle- For laboratories acquiring a sequencing instrument, web-
otides added during each sequencing cycle; the second stage data based bioinformatic tools are not an option for data analysis
results from quantification and quality check of the intensities since they do not offer enough computational and storage
and their conversion to corresponding nucleotides and sequence capacity. For example, bioinformatics tools available through
reads using specialized algorithms; and the third where scientific the website of the National Center for Biotechnology Informa-
value is produced from the sequence data, by applying assembly tion (NCBI, http://ncbi.nlm.nih.gov) allow input of only few
algorithms to the sequence reads in order to reconstruct the MegaBytes in size, while most sequence datasets are in the order
complete genome. Having acquired the complete genome of gigabytes or even terabytes (> Table 14.1). Few sequencing
sequence, it is then processed further using bioinformatic pipe- runs can overwhelm disk systems available to most laboratories,
lines for gene calling, functional annotation, and discovery of and costs for data disks must be considered in addition to
other interesting biological features on the genome. the expense for acquiring the sequencing instrument.
As sequencing data go from first to third stage, their scien- Hard disk prices per GB have been following a downward
tific value is increased, while their size is reduced. For example, trend similar to the price per sequenced DNA base, but this
the light intensity values are saved by the sequence instrument in has been happening at a reduced rate, halving roughly every
large image files which have no value for scientific analysis after 14 months. Therefore, the required investment for data storage
they are quantified and converted to sequence reads, other than systems and overall costs for building an informatics infrastruc-
used as quality reference for the reads. On the other hand, ture can surpass the cost of the sequencer by many orders of
sequence reads are small, plain text files which are used to magnitude.
reconstruct the complete genome sequence with assembly algo- Given the significant costs that can result from data storage
rithms. Assembled genomes have the most value, as they can be and computation systems, data management plans should be
used in discovery of new genes, comparative genomics, and drawn on the early stages of designing sequencing experiments
evolutionary studies. Examples of data size generated from spe- and submitting proposals involving generation of large amounts
cific sequencing instruments are available on the market of genomic data. An alternative option to investing in informat-
presented in > Table 14.1, and it is apparent that data trans- ics infrastructure is obtaining storage capacity from a cloud
formations from stage one to stage three result in data reduction computing service. Cloud services provide researchers with the
from 22-fold (Illumina II) to 3-fold (SOLiD 3). ability to store data and perform bioinformatic analysis on
Since the introduction of the ABI SOLiD instrument in a practically unlimited pool of virtual machine (VM) servers,
2007, sequencing technologies continue to move in a direction without owning or maintaining any computer hardware. The
where throughput is increasing while the cost per sequenced charge model used by cloud service providers is similar to
DNA base is decreasing, dropping in half about every 5 months. utilities such as electricity, and customers are billed based on
Furthermore, the introduction of benchtop, low-cost genome the amount of computational resources consumed. This can
sequencers such as MiSeq by Illumina (http://www.illumina. work better for smaller research laboratories acquiring a low-
com/systems/miseq.ilmn) and GS Junior by 454 (http://www. cost, benchtop sequencing instrument, instead of investing in
gsjunior.com) has made complete sequencing of bacterial computer hardware and data center infrastructure for which the
genomes affordable to small laboratories. Nonetheless, cost cannot be justified for only a few experiments.
An example of a bioinformatics platform using cloud com- workflow management system that offers highly customizable
puting technologies is J. Craig Venter Institute’s (JCVI) Cloud pipelines for metagenomic data processing. The CAMERA por-
Biolinux (http://www.cloudbiolinux.org), which consists of tal (Sun et al. 2011) not only provides metagenomic dataset
a VM designed to run on cloud computing platforms. The querying and download capabilities, but it also makes available
system offers a suite of bioinformatics software including more a series of pre-built workflows acting on common metagenomic
than 100 preinstalled and configured sequence analysis tools, data analyses, such as raw data preprocessing, BLAST-based
coupled with an intuitive graphical interface. The software similarity searches, sequence clustering and assembly, functional
included with JCVI Cloud Biolinux runs on a practically unlim- annotation, and community diversity measures. More impor-
ited pool of high capacity VM servers that can be rented on tantly, it offers a collaborative environment for data sharing and
demand from cloud computing vendors such as Amazon EC2. integration.
This provides researchers with a large-scale, virtualized infor- MG-RAST is a comprehensive analysis server providing
matics infrastructure, without the financial or time burden of annotation and comparative analysis of metagenomic data, as
owning and maintaining hardware. Research on cloud and well as quality control and multivariate sample comparison tools
virtualized computing solutions for bioinformatics can democ- (Meyer et al. 2008). The latest version (3.0) is cloud computing
ratize access to computational resources for smaller laboratories, ready, capable of handling short reads (at least 75 bp), and
which use next generation sequencing or other high-throughput supports metadata to describe samples. WebMGA (Wu et al.
genomics technologies for biological experiments. 2011) is a web server providing a wide range of tools for
metagenomic analysis, such as gene prediction, taxonomic anal-
ysis, and functional annotation.
Analytical Pipelines Nevertheless, web-based workflows face some limitations,
such as the general lack of flexibility to adapt the workflows
Once the raw sequencing data is generated, it should be stressed and data security. Also some constraints can be set by the remote
that a series of steps are necessary before any biological insights servers to reduce storage and processing loads. Stand-alone
could be drawn, since all sequencing technologies have inherent software packages like SmashCommunity (Arumugam et al.
technical biases. As such, the exploration of metagenomic 2010) provide an all-encompassing list of features to perform
datasets, both at quantitative and qualitative levels, requires functional annotation of metagenomes. Another option is to
elaborate analytical protocols to face this complex setting. completely avoid pre-built pipelines and manually invoke
General guidelines for both functional and community a series of task-specific programs to perform the analysis.
metagenomics processing have been comprehensively described There are advantages in this scenario, such as greater control of
previously (Wooley et al. 2010). A detailed protocol for func- program-specific parameters, maintenance of intermediate files,
tional (shotgun) metagenomics data processing was described and freedom to create streamlined processing protocols. This
recently (Tanenbaum et al. 2010) and is illustrative of how comes at the expense of requiring bioinformatics experts to cope
several programs should be chained together to accomplish with installation, maintenance, and programming in the Linux
a reliable functional annotation. This process involves the pre- operating system, the ‘‘de facto’’ bioinformatics computational
diction of coding and noncoding genes followed by application environment. In this respect, CloVR (Angiuoli et al. 2011) was
of several similarity search algorithms to assign biologically recently released, providing a Linux virtual machine with
relevant attributes like gene ontology, Enzyme Commission preinstalled software to perform automated pipelines for sin-
(EC) numbers, and protein domains. gle-genome, shotgun, or 16S rRNA gene metagenomic data.
Metagenomics Pipelines Concluding Remarks
As seen previously, the analytical process of metagenomic A number of post-genomic tools have been developed to study
datasets involves the coordinated execution of several specialized genomes, regardless if they come from a cultivated or
programs, each receiving and emitting data files in different uncultivated microorganism, if they come from a microbial
formats, in a concerted flow of events. There is an intrinsic community or single cell. The challenge now is to make this
modularity and, for each task, a marked heterogeneity in terms information useful through the development of products.
of software options, parameter settings, and execution times. As Whenever possible, scientists should be even more proactive in
more data and analytical tools are continuously generated, it is seeking new application opportunities for their research keeping
important to stress the complexity of the whole process, which is a strong link with the market. If there is no understanding of the
increasingly turning out to be more demanding than data gen- market needs from the early stages of the process, the result may
eration itself. be an interesting academic work with no market value. With the
Automated web-based workflows provide a user-friendly scarcity of public funding and an ever greater interest of tax-
alternative to execute initial metagenomic analyses, effectively payers in how their money is spent, scientists should push
abstracting the end user from software installation and mainte- creative solutions to convert the knowledge about genomes
nance. Galaxy (Kosakovsky Pond et al. 2009) is a generic web into everyday technologies that can benefit the average citizen.
This is a great opportunity to show the value of science and how Blainey PC, Mosier AC, Potanina A, Francis CA, Quake SR (2011) Genome
of a low-salinity ammonia-oxidizing archaeon determined by single-cell
basic and applied science are interconnected, as the knowledge
and metagenomic analysis. PLoS One 6:e16626
about genomes is transformed into post-genome technologies Bolhuis H, Palm P, Wende A, Falb M, Rampp M, Rodriguez-Valera F, Pfeiffer F,
and commercial products. Oesterhelt D (2006) The genome of the square archaeon Haloquadratum
walsbyi: life at the limits of water activity. BMC Genomics 7:169
Bomar L, Maltz M, Colston S, Graf J (2011) Directed culturing of microorgan-
isms using metatranscriptomics. MBio 2(2):e00012-11
Acknowledgments Brochier-Armanet C, Forterre P, Gribaldo S (2011) Phylogeny and evolution of the
Archaea: one hundred genomes later. Curr Opin Microbiol 14:274
We thank Samuel Dias Araújo Júnior for making figures and Brulc JM, Antonopoulos DA, Miller ME, Wilson MK, Yannarell AC, Dinsdale EA,
Dr. David Weiner for providing information about commercial Edwards RE, Frank ED, Emerson JB, Wacklin P, Coutinho PM, Henrissat B,
enzyme products, and also CAPES, FAP-DF, and CNPq for Nelson KE, White BA (2009) Gene-centric metagenomics of the fiber-
adherent bovine rumen microbiome reveals forage specific glycoside hydro-
funding.
lases. Proc Natl Acad Sci USA 106:1948–1953
Button DK, Schut F, Quang P, Martin R, Robertson BR (1993) Viability and
isolation of marine bacteria by dilution culture: theory, procedures, and
initial results. Appl Environ Microbiol 59:881–891
References Chistoserdova L (2010) Recent progress and new challenges in metagenomics for
biotechnology. Biotechnol Lett 32:1351–1359
Angiuoli SV, Matalka M, Gussman A, Galens K, Vangala M, Riley DR, Arze C, Chistoserdovai L (2010) Functional metagenomics: recent advances and future
White JR, White O, Fricke WF (2011) CloVR: a virtual machine for auto- challenges. Biotechnol Genet Eng Rev 26:335–352
mated and portable sequence analysis from the desktop using cloud com- Chitsaz H, Yee-Greenbaum JL, Tesler G, Lombardo M-J, Dupont CL, Badger JH,
puting. BMC Bioinformatics 12:356 Novotny M, Rusch DB, Fraser LJ, Gormley NA, Schulz-Trieglaff O, Smith GP,
Ansorge WJ (2009) Next-generation DNA sequencing techniques. N Biotechnol Evers DJ, Pevzner PA, Lasken RS (2011) De Novo assembly of bacterial
25:195–203 genomes from single cells. Nat Biotechnol 29:915
Arumugam M, Harrington ED, Foerstner KU, Raes J, Bork P (2010) Connon SA, Giovannoni SJ (2002) High-throughput methods for culturing
SmashCommunity: a metagenomic annotation and analysis tool. Bioinfor- microorganisms in very-low-nutrient media yield diverse new marine
matics 26:2977–2978 isolates. Appl Environ Microbiol 68:3878–3885
Arumugam M, Raes J, Pelletier E, Le Paslier D, Yamada T, Mende DR, Fernandes Cuadros-Orellana S, Martin-Cuadrado AB, Legault B, D’Auria G, Zhaxybayeva O,
GR, Tap J, Bruls T, Batto JM, Bertalan M, Borruel N, Casellas F, Fernandez L, Papke RT, Rodriguez-Valera F (2007) Genomic plasticity in prokaryotes: the
Gautier L, Hansen T, Hattori M, Hayashi T, Kleerebezem M, Kurokawa K, case of the square haloarchaeon. ISME J 1:235–245
Leclerc M, Levenez F, Manichanh C, Nielsen HB, Nielsen T, Pons N, Poulain J, de Castro AP, Quirino BF, Allen H, Williamson LL, Handelsman J, Kruger RH
Qin J, Sicheritz-Ponten T, Tims S, Torrents D, Ugarte E, Zoetendal EG, Wang J, (2011) Construction and validation of two metagenomic DNA libraries from
Guarner F, Pedersen O, de Vos WM, Brunak S, Dore J, Antolin M, Cerrado soil with high clay content. Biotechnol Lett 33:2169
Artiguenave F, Blottiere HM, Almeida M, Brechot C, Cara C, Chervaux C, de Jager V, Siezen RJ (2011) Single-cell genomics: unravelling the genomes of
Cultrone A, Delorme C, Denariaz G, Dervyn R, Foerstner KU, Friss C, unculturable microorganisms. Microb Biotechnol 4:431–437
van de Guchte M, Guedon E, Haimet F, Huber W, van Hylckama-Vlieg J, Dean FB, Nelson JR, Giesler TL, Lasken RS (2001) Rapid amplification of plasmid
Jamet A, Juste C, Kaci G, Knol J, Lakhdari O, Layec S, Le Roux K, Maguin E, and phage DNA using Phi 29 DNA polymerase and multiply-primed rolling
Merieux A, Melo Minardi R, M’Rini C, Muller J, Oozeer R, Parkhill J, circle amplification. Genome Res 11:1095–1099
Renault P, Rescigno M, Sanchez N, Sunagawa S, Torrejon A, Turner K, Deloger M, El Karoui M, Petit MA (2009) A genomic distance based on MUM
Vandemeulebrouck G, Varela E, Winogradsky Y, Zeller G, Weissenbach J, indicates discontinuity between most bacterial species and genera. J Bacteriol
Ehrlich SD, Bork P (2011) Enterotypes of the human gut microbiome. Nature 191:91–99
473:174–180 Doolittle WF, Papke RT (2006) Genomics and the bacterial species problem.
Baker-Austin C, Dopson M (2007) Life in acid: pH homeostasis in acidophiles. Genome Biol 7:116
Trends Microbiol 15:165–171 Dumon C, Varvak A, Wall MA, Flint JE, Lewis RJ, Lakey JH, Morland C,
Baldrian P, Kolarik M, Stursova M, Kopecky J, Valaskova V, Vetrovsky T, Zifcakova Luginbuhl P, Healey S, Todaro T, DeSantis G, Sun M, Parra-Gessert L, Tan
L, Snajdr J, Ridl J, Vlcek C, Voriskova J (2011) Active and total microbial X, Weiner DP, Gilbert HJ (2008) Engineering hyperthermostability into
communities in forest soil are largely different and highly stratified during a GH11 xylanase is mediated by subtle changes to protein structure. J Biol
decomposition. ISME J 6:248–258 Chem 283:22557–22564
Beloqui A, Pita M, Polaina J, Martinez-Arias A, Golyshina OV, Zumarraga M, Dyall-Smith ML, Pfeiffer F, Klee K, Palm P, Gross K, Schuster SC, Rampp M,
Yakimov MM, Garcia-Arellano H, Alcalde M, Fernandez VM, Elborough K, Oesterhelt D (2011) Haloquadratum walsbyi: limited diversity in a global
Andreu JM, Ballesteros A, Plou FJ, Timmis KN, Ferrer M, Golyshin PN pond. PLoS One 6:e20968
(2006) Novel polyphenol oxidase mined from a metagenome expression Ehrlich GD, Ahmed A, Earl J, Hiller NL, Costerton JW, Stoodley P, Post JC,
library of bovine rumen: biochemical properties, structural analysis, and DeMeo P, Hu FZ (2010) The distributed genome hypothesis as a rubric for
phylogenetic relationships. J Biol Chem 281:22933–22942 understanding evolution in situ during chronic bacterial biofilm infectious
Beloqui A, Nechitaylo TY, Lopez-Cortes N, Ghazi A, Guazzaroni ME, Polaina J, processes. FEMS Immunol Med Microbiol 59:269–279
Strittmatter AW, Reva O, Waliczek A, Yakimov MM, Golyshina OV, Ferrer M, Fernandez-Arrojo L, Guazzaroni ME, Lopez-Cortes N, Beloqui A, Ferrer M
Golyshin PN (2010) Diversity of glycosyl hydrolases from cellulose-depleting (2010) Metagenomic era for biocatalyst identification. Curr Opin Biotechnol
communities enriched from casts of two earthworm species. Appl Environ 21:725–733
Microbiol 76:5934–5946 Ferrari BC, Oregaard G, Sorensen SJ (2004) Recovery of GFP-labeled bacteria for
Benders GA, Noskov VN, Denisova EA, Lartigue C, Gibson DG, Assad-Garcia N, culturing and molecular analysis after cell sorting using a benchtop flow
Chuang RY, Carrera W, Moodie M, Algire MA, Phan Q, Alperovich N, cytometer. Microb Ecol 48:239–245
Vashee S, Merryman C, Venter JC, Smith HO, Glass JI, Hutchison CA Fleischmann RD, Adams MD, White O, Clayton RA, Kirkness EF, Kerlavage AR,
3rd (2010) Cloning whole bacterial genomes in yeast. Nucleic Acids Res Bult CJ, Tomb JF, Dougherty BA, Merrick JM et al (1995) Whole-genome
38:2558–2569 random sequencing and assembly of Haemophilus influenzae Rd. Science
Several Authors (2008) Big data. Nature 455:1. Nature special issue 269:496–512
Frohlich J, Konig H (1999) Rapid isolation of single microbial cells from mixed Kunin V, Engelbrektson A, Ochman H, Hugenholtz P (2009) Wrinkles in the rare
natural and laboratory populations with the aid of a micromanipulator. Syst biosphere: pyrosequencing errors can lead to artificial inflation of diversity
Appl Microbiol 22:249–257 estimates. Environ Microbiol 12:118–123
Gibson DG, Benders GA, Andrews-Pfannkoch C, Denisova EA, Baden-Tillson H, Lagesen K, Ussery DW, Wassenaar TM (2010) Genome update: the 1000th
Zaveri J, Stockwell TB, Brownley A, Thomas DW, Algire MA, Merryman C, genome–a cautionary tale. Microbiology 156:603–608
Young L, Noskov VN, Glass JI, Venter JC, Hutchison CA 3rd, Smith HO Lan R, Reeves PR (2001) When does a clone deserve a name? A perspective on
(2008a) Complete chemical synthesis, assembly, and cloning of a bacterial species based on population genetics. Trends Microbiol 9:419–424
Mycoplasma genitalium genome. Science 319:1215–1220 Lapierre P, Gogarten JP (2009) Estimating the size of the bacterial pan-genome.
Gibson DG, Benders GA, Axelrod KC, Zaveri J, Algire MA, Moodie M, Monta- Trends Genet 25:107–110
gue MG, Venter JC, Smith HO, Hutchison CA 3rd (2008b) One-step Lartigue C, Glass JI, Alperovich N, Pieper R, Parmar PP, Hutchison CA 3rd, Smith
assembly in yeast of 25 overlapping DNA fragments to form a complete HO, Venter JC (2007) Genome transplantation in bacteria: changing one
synthetic Mycoplasma genitalium genome. Proc Natl Acad Sci USA species to another. Science 317:632–638
105:20404–20409 Lartigue C, Vashee S, Algire MA, Chuang RY, Benders GA, Ma L, Noskov VN,
Gibson DG, Glass JI, Lartigue C, Noskov VN, Chuang RY, Algire MA, Benders Denisova EA, Gibson DG, Assad-Garcia N, Alperovich N, Thomas DW,
GA, Montague MG, Ma L, Moodie MM, Merryman C, Vashee S, Merryman C, Hutchison CA 3rd, Smith HO, Venter JC, Glass JI (2009) Cre-
Krishnakumar R, Assad-Garcia N, Andrews-Pfannkoch C, Denisova EA, ating bacterial strains from genomes that have been cloned and engineered in
Young L, Qi ZQ, Segall-Shapiro TH, Calvey CH, Parmar PP, Hutchison CA yeast. Science 325:1693–1696
3rd, Smith HO, Venter JC (2010) Creation of a bacterial cell controlled by Lefébure T, Stanhope MJ (2007) Evolution of the core and pan-genome of
a chemically synthesized genome. Science 329:52–56 Streptococcus: positive selection, recombination, and genome composition.
Glass JI, Hutchison CA 3rd, Smith HO, Venter JC (2009) A systems biology tour Genome Biol 8:R71
de force for a near-minimal bacterium. Mol Syst Biol 5:330 Legault BA, Lopez-Lopez A, Alba-Casado JC, Doolittle WF, Bolhuis H,
Gomez-Alvarez V, Teal TK, Schmidt TM (2009) Systematic artifacts in Rodriguez-Valera F, Papke RT (2006) Environmental genomics of
metagenomes from complex microbial communities. ISME J 3:1314–1317 ‘‘Haloquadratum walsbyi’’ in a saltern crystallizer indicates a large pool of
Goris J, Konstantinidis KT, Klappenbach JA, Coenye T, Vandamme P, Tiedje JM accessory genes in an otherwise coherent species. BMC Genomics 7:171
(2007) DNA-DNA hybridization values and their relationship to whole- Liu N, Yan X, Zhang M, Xie L, Wang Q, Huang Y, Zhou X, Wang S, Zhou Z (2011)
genome sequence similarities. Int J Syst Evol Microbiol 57:81–91 Microbiome of fungus-growing termites: a new reservoir for lignocellulase
Hall BG, Ehrlich GD, Hu FZ (2010) Pan-genome analysis provides much higher genes. Appl Environ Microbiol 77:48–56
strain typing resolution than multi-locus sequence typing. Microbiology Lopez-Garcia P, Moreira D (2008) Tracking microbial biodiversity through
156:1060–1068 molecular and genomic ecology. Res Microbiol 159:67–73
Han S, Lee J, Kim W, Shin HJ, Kim G (2008) Screening of promoters from Lu Y, Yu Y, Zhou R, Sun W, Dai C, Wan P, Zhang L, Hao D, Ren H (2011) Cloning
metagenomic DNA and their use for the construction of expression vectors. and characterisation of a novel 2,4-dichlorophenol hydroxylase from
J Microbiol Biotechnol 18:1634–1640 a metagenomic library derived from polychlorinated biphenyl-contaminated
Harrison PM, Arosio P (1996) The ferritins: molecular properties, iron storage soil. Biotechnol Lett 33:1159–1167
function and cellular regulation. Biochim Biophys Acta 1275:161–203 Lukjancenko O, Wassenaar TM, Ussery DW (2010) Comparison of 61 Sequenced
Heywood JL, Sieracki ME, Bellows W, Poulton NJ, Stepanauskas R (2011) Escherichia coli Genomes. Microb Ecol 60:708–720
Capturing diversity of marine heterotrophic protists: one cell at a time. MacLean D, Jones JD, Studholme DJ (2009) Application of ‘next-generation’
ISME J 5:674–684 sequencing technologies to microbial genetics. Nat Rev Microbiol 7:287–296
Hogg JS, Hu FZ, Janto B, Boissy R, Hayes J, Keefe R, Post JC, Ehrlich GD Maione D, Margarit I, Rinaudo CD, Masignani V, Mora M, Scarselli M, Tettelin H,
(2007) Characterization and modeling of the Haemophilus influenzae core Brettoni C, Iacobini ET, Rosini R, D’Agostino N, Miorin L, Buccato S,
and supragenomes based on the complete genomic sequences of Rd and 12 Mariani M, Galli G, Nogarotto R, Nardi Dei V, Vegni F, Fraser C, Mancuso
clinical nontypeable strains. Genome Biol 8:R103 G, Teti G, Madoff LC, Paoletti LC, Rappuoli R, Kasper DL, Telford JL, Grandi
Horner DS, Pavesi G, Castrignano T, De Meo PD, Liuni S, Sammeth M, Picardi G (2005) Identification of a universal Group B Streptococcus vaccine by
E, Pesole G (2009) Bioinformatics approaches for genomics and post multiple genome screen. Science 309:148–150
genomics applications of next-generation sequencing. Brief Bioinform Majernik A, Gottschalk G, Daniel R (2001) Screening of environmental DNA
11:181–197 libraries for the presence of genes conferring Na(+)(Li(+))/H(+) antiporter
Ishoy T, Kvist T, Westermann P, Ahring BK (2006) An improved method for single activity on Escherichia coli: characterization of the recovered genes and the
cell isolation of prokaryotes from meso-, thermo- and hyperthermophilic corresponding gene products. J Bacteriol 183:6645–6653
environments using micromanipulation. Appl Microbiol Biotechnol Marcy Y, Ouverney C, Bik EM, Losekann T, Ivanova N, Martin HG, Szeto E, Platt
69:510–514 D, Hugenholtz P, Relman DA, Quake SR (2007) Dissecting biological ‘‘dark
Kang Y, Norris MH, Zarzycki-Siek J, Nierman WC, Donachie SP, Hoang TT matter’’ with single-cell genetic analysis of rare and uncultivated TM7
(2011) Transcript amplification from single bacterium for transcriptome microbes from the human mouth. Proc Natl Acad Sci USA 104:11889–11894
analysis. Genome Res 21:925–935 Medini D, Donati C, Tettelin H, Masignani V, Rappuoli R (2005) The microbial
Kim EY, Oh KH, Lee MH, Kang CH, Oh TK, Yoon JH (2009) Novel cold-adapted pan-genome. Curr Opin Genet Dev 15:589–594
alkaline lipase from an intertidal flat metagenome and proposal for a new Medini D, Serruto D, Parkhill J, Relman DA, Donati C, Moxon R, Falkow S,
family of bacterial lipases. Appl Environ Microbiol 75:257–260 Rappuoli R (2008) Microbiology in the post-genomic era. Nat Rev Microbiol
Kislyuk AO, Haegeman B, Bergman NH, Weitz JS (2011) Genomic fluidity: an 6:419–430
integrative view of gene diversity within microbial populations. BMC Geno- Meyer F, Paarmann D, D’Souza M, Olson R, Glass EM, Kubal M, Paczian T,
mics 12:32 Rodriguez A, Stevens R, Wilke A, Wilkening J, Edwards RA (2008) The
Klenk HP, Göker M (2010) En route to a genome-based classification of Archaea metagenomics RAST server – a public resource for the automatic phyloge-
and Bacteria? Syst Appl Microbiol 33:175–182 netic and functional analysis of metagenomes. BMC Bioinformatics 9:386
Konstantinidis KT, Tiedje JM (2005) Genomic insights that advance the species Mira A, Martin-Cuadrado AB, D’Auria G, Rodriguez-Valera F (2010) The
definition for prokaryotes. Proc Natl Acad Sci USA 102:2567–2572 bacterial pan-genome:a new paradigm in microbiology. Int Microbiol
Konstantinidis KT, Ramette A, Tiedje JM (2006) The bacterial species definition 13:45–57
in the genomic era. Philos Trans R Soc Lond B Biol Sci 361:1929–1940 Morgan JL, Darling AE, Eisen JA (2010) Metagenomic sequencing of an in vitro-
Kosakovsky Pond S, Wadhawan S, Chiaromonte F, Ananda G, Chung WY, Taylor simulated microbial community. PLoS One 5:e10209
J, Nekrutenko A (2009) Windshield splatter analysis with the Galaxy Mussmann M, Hu FZ, Richter M, de Beer D, Preisler A, Jorgensen BB,
metagenomic pipeline. Genome Res 19:2144–2153 Huntemann M, Glockner FO, Amann R, Koopman WJ, Lasken RS, Janto
B, Hogg J, Stoodley P, Boissy R, Ehrlich GD (2007) Insights into the genome (2004) Exploring nitrilase sequence space for enantioselective catalysis.
of large sulfur bacteria revealed by analysis of single filaments. PLoS Appl Environ Microbiol 70:2429–2436
Biol 5:e230 Rodrigue S, Malmstrom RR, Berlin AM, Birren BW, Henn MR, Chisholm SW
Muzzi A, Masignani V, Rappuoli R (2007) The pan-genome: towards (2009) Whole genome amplification and de novo assembly of single bacterial
a knowledge-based discovery of novel targets for vaccines and antibacterials. cells. PLoS One 4:e6864
Drug Discov Today 12:429–439 Rondon MR, August PR, Bettermann AD, Brady SF, Grossman TH, Liles MR,
Nagarajan H, Butler JE, Klimes A, Qiu Y, Zengler K, Ward J, Young ND, Methe Loiacono KA, Lynch BA, MacNeil IA, Minor C, Tiong CL, Gilman M,
BA, Palsson BO, Lovley DR, Barrett CL (2010) De Novo assembly of the Osburne MS, Clardy J, Handelsman J, Goodman RM (2000) Cloning the
complete genome of an enhanced electricity-producing variant of Geobacter soil metagenome: a strategy for accessing the genetic and functional diversity
sulfurreducens using only short reads. PLoS One 5:e10922 of uncultured microorganisms. Appl Environ Microbiol 66:2541–2547
Newman DJ, Cragg GM (2005) The discovery of anticancer drugs from natural Schadt EE, Linderman MD, Sorenson J, Lee L, Nolan GP (2010) Computational
sources. In: Zhang L, Demain AL (eds) Natural products: drug discovery and solutions to large-scale data management and analysis. Nat Rev Genet
therapeutic medicine. Humana Press, Totowan, pp 128–168 11:647–657
Oh D, Porter K, Russ B, Burns D, Dyall-Smith M (2010) Diversity of Schallmey M, Ly A, Wang C, Meglei G, Voget S, Streit WR, Driscoll BT, Charles
Haloquadratum and other haloarchaea in three, geographically distant, Aus- TC (2011) Harvesting of novel polyhydroxyalkanaote (PHA) synthase
tralian saltern crystallizer ponds. Extremophiles 14:161–169 encoding genes from a soil metagenome library using phenotypic screening.
Ottesen EA, Hong JW, Quake SR, Leadbetter JR (2006) Microfluidic digital PCR FEMS Microbiol Lett 321:150–156
enables multigene analysis of individual environmental bacteria. Science Schwartz S, Oren R, Ast G (2011) Detection and removal of biases in the analysis
314:1464–1467 of next-generation sequencing reads. PLoS One 6:e16685
Pang H, Zhang P, Duan CJ, Mo XC, Tang JL, Feng JX (2009) Identification of Sharp KH, Davidson SK, Haygood MG (2007) Localization of ‘Candidatus
cellulase genes from the metagenomes of compost soils and functional Endobugula sertula’ and the bryostatins throughout the life cycle of the
characterization of one novel endoglucanase. Curr Microbiol 58:404–408 bryozoan Bugula neritina. ISME J 1:693–702
Parachin NS, Gorwa-Grauslund MF (2011) Isolation of xylose isomerases by Siegl A, Kamke J, Hochmuth T, Piel J, Richter M, Liang C, Dandekar T, Hentschel
sequence- and function-based screening from a soil metagenome library. U (2010) Single-cell genomics reveals the lifestyle of Poribacteria, a candidate
Biotechnol Biofuels 4:9 phylum symbiotically associated with marine sponges. ISME J 5:61–70
Park SY, Shin HJ, Kim GJ (2011) Screening and identification of a novel esterase Simon C, Daniel R (2011) Metagenomic analyses: past and future trends. Appl
EstPE from a metagenomic DNA library. J Microbiol 49:7–14 Environ Microbiol 77:1153–1161
Pathak GP, Ehrenreich A, Losi A, Streit WR, Gartner W (2009) Novel blue light- Simon C, Herath J, Rockstroh S, Daniel R (2009) Rapid identification of genes
sensitive proteins from a metagenomic approach. Environ Microbiol encoding DNA polymerases by function-based screening of metagenomic
11:2388–2399 libraries derived from glacial ice. Appl Environ Microbiol 75:2964–2968
Podar M, Abulencia CB, Walcher M, Hutchison D, Zengler K, Garcia JA, Holland Sogin ML, Morrison HG, Huber JA, Mark Welch D, Huse SM, Neal PR, Arrieta
T, Cotton D, Hauser L, Keller M (2007) Targeted access to the genomes of JM, Herndl GJ (2006) Microbial diversity in the deep sea and the
low-abundance organisms in complex microbial communities. Appl underexplored ‘‘rare biosphere’’. Proc Natl Acad Sci USA 103:12115–12120
Environ Microbiol 73:3205–3214 Stepanauskas R, Sieracki ME (2007) Matching phylogeny and metabolism in the
Pope PB, Denman SE, Jones M, Tringe SG, Barry K, Malfatti SA, McHardy AC, uncultured marine bacteria, one cell at a time. Proc Natl Acad Sci USA
Cheng JF, Hugenholtz P, McSweeney CS, Morrison M (2010) Adaptation to 104:9052–9057
herbivory by the Tammar wallaby includes bacterial and glycoside hydrolase Strous M, Pelletier E, Mangenot S, Rattei T, Lehner A, Taylor MW, Horn M,
profiles different from other herbivores. Proc Natl Acad Sci USA Daims H, Bartol-Mavel D, Wincker P, Barbe V, Fonknechten N, Vallenet D,
107:14793–14798 Segurens B, Schenowitz-Truong C, Medigue C, Collingro A, Snel B, Dutilh
Porter J, Edwards C, Morgan JA, Pickup RW (1993) Rapid, automated separation BE, Op den Camp HJ, van der Drift C, Cirpus I, van de Pas-Schoonen KT,
of specific bacteria from lake water and sewage by flow cytometry and cell Harhangi HR, van Niftrik L, Schmid M, Keltjens J, van de Vossenberg J,
sorting. Appl Environ Microbiol 59:3327–3333 Kartal B, Meier H, Frishman D, Huynen MA, Mewes HW, Weissenbach J,
Portnoy VA, Bezdan D, Zengler K (2011) Adaptive laboratory evolution- Jetten MS, Wagner M, Le Paslier D (2006) Deciphering the evolution and
harnessing the power of biology for metabolic engineering. Curr Opin metabolism of an anammox bacterium from a community genome. Nature
Biotechnol 22:590–594 440:790–794
Purdy KJ (2005) Nucleic acid recovery from complex environmental samples. Suen G, Scott JJ, Aylward FO, Adams SM, Tringe SG, Pinto-Tomas AA, Foster CE,
Methods Enzymol 397:271–292 Pauly M, Weimer PJ, Barry KW, Goodwin LA, Bouffard P, Li L, Osterberger J,
Raghunathan A, Ferguson HR Jr, Bornarth CJ, Song W, Driscoll M, Lasken RS Harkins TT, Slater SC, Donohue TJ, Currie CR (2010) An insect herbivore
(2005) Genomic DNA amplification from a single bacterium. Appl Environ microbiome with high plant biomass-degrading capacity. PLoS Genet
Microbiol 71:3342–3347 6:e1001129
Rasko DA, Rosovitz MJ, Myers GSA, Mongodin EF, Fricke WF, Gajer P, Crabtree J, Sun S, Chen J, Li W, Altintas I, Lin A, Peltier S, Stocks K, Allen EE, Ellisman M,
Sebaihia M, Thomson NR, Chaudhuri R, Henderson IR, Sperandio V, Ravel J Grethe J, Wooley J (2011) Community cyberinfrastructure for advanced
(2008) The pangenome structure of Escherichia coli: comparative microbial ecology research and analysis: the CAMERA resource. Nucleic
genomic analysis of E. coli commensal and pathogenic isolates. J Bacteriol Acids Res 39:D546–D551
190:6881–6893 Suzuki S, Ono N, Furusawa C, Ying BW, Yomo T (2011) Comparison of sequence
Rhee JK, Ahn DG, Kim YG, Oh JW (2005) New thermophilic and thermostable reads obtained from three next-generation sequencing platforms. PLoS One
esterase with sequence similarity to the hormone-sensitive lipase family, 6:e19534
cloned from a metagenomic library. Appl Environ Microbiol 71:817–825 Tadmor AD, Ottesen EA, Leadbetter JR, Phillips R (2011) Probing individual
Richardson TH, Tan X, Frey G, Callen W, Cabell M, Lam D, Macomber J, Short environmental bacteria for viruses by using microfluidic digital PCR. Science
JM, Robertson DE, Miller C (2002) A novel, high performance enzyme for 333:58–62
starch liquefaction. Discovery and optimization of a low pH, thermostable Tanenbaum DM, Goll J, Murphy S, Kumar P, Zafar N, Thiagarajan M, Madupu R,
alpha-amylase. J Biol Chem 277:26501–26507 Davidsen T, Kagan L, Kravitz S, Rusch DB, Yooseph S (2010) The JCVI
Riesenfeld CS, Goodman RM, Handelsman J (2004) Uncultured soil bacteria standard operating procedure for annotating prokaryotic metagenomic
are a reservoir of new antibiotic resistance genes. Environ Microbiol 6:981–989 shotgun sequencing data. Stand Genomic Sci 2:229–237
Robertson DE, Chaplin JA, DeSantis G, Podar M, Madden M, Chi E, Richardson Tang F, Barbacioru C, Wang Y, Nordman E, Lee C, Xu N, Wang X, Bodeau J,
T, Milan A, Miller M, Weiner DP, Wong K, McQuaid J, Farwell B, Preston LA, Tuch BB, Siddiqui A, Lao K, Surani MA (2009) mRNA-Seq whole-
Tan X, Snead MA, Keller M, Mathur E, Kretz PL, Burk MJ, Short JM transcriptome analysis of a single cell. Nat Methods 6:377–382
Taniguchi K, Kajiyama T, Kambara H (2009) Quantitative analysis of gene Nealson K, White O, Peterson J, Hoffman J, Parsons R, Baden-Tillson H,
expression in a single cell by qPCR. Nat Methods 6:503–506 Pfannkoch C, Rogers YH, Smith HO (2004) Environmental genome shotgun
Taniguchi Y, Choi PJ, Li GW, Chen H, Babu M, Hearn J, Emili A, Xie XS sequencing of the Sargasso Sea. Science 304:66–74
(2010) Quantifying E. coli proteome and transcriptome with single- Walker A, Parkhill J (2008) Single-cell genomics. Nat Rev Microbiol 6:176–177
molecule sensitivity in single cells. Science 329:533–538 Wang C, Meek DJ, Panchal P, Boruvka N, Archibald FS, Driscoll BT, Charles TC
Taupp M, Mewis K, Hallam SJ (2011) The art and design of functional (2006) Isolation of poly-3-hydroxybutyrate metabolism genes from complex
metagenomic screens. Curr Opin Biotechnol 22:465–472 microbial communities by phenotypic complementation of bacterial
Taylor RJ, Falconnet D, Niemisto A, Ramsey SA, Prinz S, Shmulevich I, Galitski T, mutants. Appl Environ Microbiol 72:384–391
Hansen CL (2009) Dynamic analysis of MAPK signaling using a high- Wender PA, Miller BL (2009) Synthesis at the molecular frontier. Nature 460:
throughput microfluidic single-cell imaging platform. Proc Natl Acad Sci 197–201
USA 106:3758–3763 Wooley JC, Godzik A, Friedberg I (2010) A primer on metagenomics. PLoS
Tettelin H, Masignani V, Cieslewicz MJ, Donati C, Medini D, Ward NL, Angiuoli Comput Biol 6:e1000667
SV, Crabtree J, Jones AL, Durkin AS, Deboy RT, Davidsen TM, Mora M, Woyke T, Xie G, Copeland A, Gonzalez JM, Han C, Kiss H, Saw JH, Senin P,
Scarselli M, Margarity Ros I, Peterson JD, Hauser CR, Sundaram JP, Nelson Yang C, Chatterji S, Cheng JF, Eisen JA, Sieracki ME, Stepanauskas
WC, Madupu R, Brinkac LM, Dodson RJ, Rosovitz MJ, Sullivan SA, R (2009) Assembling the marine metagenome, one cell at a time. PLoS
Daugherty SC, Haft DH, Selengut J, Gwinn ML, Zhou L, Zafar N, Khouri One 4:e5299
H, Radune D, Dimitrov G, Watkins K, O’Connor KJ, Smith S, Utterback TR, Woyke T, Tighe D, Mavromatis K, Clum A, Copeland A, Schackwitz W, Lapidus
White O, Rubens CE, Grandi G, Madoff LC, Kasper DL, Telford JL, Wessels A, Wu D, McCutcheon JP, McDonald BR, Moran NA, Bristow J, Cheng JF
MR, Rappuoli R, Fraser CM (2005) Genome analysis of multiple pathogenic (2010) One bacterial cell, one complete genome. PLoS One 5:e10314
isolates of Streptococcus agalactiae: implications for the microbial ‘‘pan- Wu S, Zhu Z, Fu L, Niu B, Li W (2011) WebMGA: a customizable web server for
genome’’. Proc Natl Acad Sci USA 102:13950–13955 fast metagenomic sequence analysis. BMC Genomics 12:444
Tettelin H, Riley D, Cattuto C, Medini D (2008) Comparative genomics: the Yoon HS, Price DC, Stepanauskas R, Rajah VD, Sieracki ME, Wilson WH, Yang
bacterial pan-genome. Curr Opin Microbiol 11:472–477 EC, Duffy S, Bhattacharya D (2011) Single-cell genomics reveals organismal
Tsuruta H, Paddon CJ, Eng D, Lenihan JR, Horning T, Anthony LC, Regentin R, interactions in uncultivated marine protists. Science 332:714–717
Keasling JD, Renninger NS, Newman JD (2009) High-level production of Yu EY, Kwon MA, Lee M, Oh JY, Choi JE, Lee JY, Song BK, Hahm DH,
amorpha-4,11-diene, a precursor of the antimalarial agent artemisinin, in Song JK (2011) Isolation and characterization of cold-active family VIII
Escherichia coli. PLoS One 4:e4489 esterases from an arctic soil metagenome. Appl Microbiol Biotechnol
Tyson GW, Chapman J, Hugenholtz P, Allen EE, Ram RJ, Richardson PM, 90:573–581
Solovyev VV, Rubin EM, Rokhsar DS, Banfield JF (2004) Community struc- Zeng Y, Novak R, Shuga J, Smith MT, Mathies RA (2010) High-performance
ture and metabolism through reconstruction of microbial genomes from the single cell genetic analysis using microfluidic emulsion generator arrays.
environment. Nature 428:37–43 Anal Chem 82:3183–3190
Uchiyama T, Abe T, Ikemura T, Watanabe K (2005) Substrate-induced gene- Zengler K (2009) Central role of the cell in microbial ecology. Microbiol Mol Biol
expression screening of environmental metagenome libraries for isolation Rev 73:712–729
of catabolic genes. Nat Biotechnol 23:88–93 Zhang K, Martiny AC, Reppas NB, Barry KW, Malek J, Chisholm SW, Church GM
Ussery DW, Kiil K, Lagesen K, Sicheritz-Ponten T, Bohlin J, Wassenaar TM (2006) Sequencing genomes from single cells by polymerase cloning. Nat
(2009) The genus Burkholderia: analysis of 56 genomic sequences. Genome Biotechnol 24:680–686
Dyn 6:140–157 Zhao S, Wang J, Bu D, Liu K, Zhu Y, Dong Z, Yu Z (2010) Novel glycoside
Venter JC, Remington K, Heidelberg JF, Halpern AL, Rusch D, Eisen JA, Wu D, hydrolases identified by screening a Chinese Holstein dairy cow rumen-
Paulsen I, Nelson KE, Nelson W, Fouts DE, Levy S, Knap AH, Lomas MW, derived metagenome library. Appl Environ Microbiol 76:6701–6705
Section 2
Symbiotic Associations
15 Role of Microorganisms in
Adaptation, Development, and Evolution
of Animals and Plants: The Hologenome
Concept
Eugene Rosenberg1 . Ilana Zilber-Rosenberg2
1
Department of Molecular Microbiology and Biotechnology, Tel Aviv University, Tel Aviv, Israel
2
Independant Scholar, Givat Shmuel, Israel
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347 systems because size can be measured by cell number or by

genome size, and in the case of many systems, the microbiota
Microbes and the Origin of Eukaryotic Cells and outnumbers its host. In spite of these limitations, the definition
Multicellular Organisms: The First Hologenomes . . . . . . . . . 348 of host and symbiont based on size continues to be widely used.
Endosymbionts and exosymbionts refer to microbiota living
The Hologenome Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348 inside or outside host cells, respectively. Symbioses take many
forms, mostly different levels of mutualism, where both the host
All Animals and Plants Establish Symbiotic Relationships and the symbiont benefit from the interaction, and to a much
with Microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348 smaller degree—parasitism, where the symbiont benefits and the
host suffers damage. Commensalism is defined as the close
Cooperation Between the Host and the Microbiota association of two or more dissimilar organisms where the
Contributes to the Fitness of the Holobiont . . . . . . . . . . . . . . . 350 association is advantageous to one and does not affect the
other(s). These types of symbioses may change under different
Invertebrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350 circumstances. For example, Wolbachia is generally detrimental
Insect Holobionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350 (parasitic) to its insect host; however, it can also be beneficial
Squid Light Organ–Vibrio Symbiosis . . . . . . . . . . . . . . . . . . . 352 (mutualistic): Wolbachia is essential for the production of
Coral: Bacteria Symbiosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352 mature oocytes in a parasitic wasp (Dedeine et al. 2001).
Another example is the bacterium Vibrio shiloi, which is only
Vertebrates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352 parasitic when the temperature is raised and virulence genes are
expressed (Rosenberg and Falkovitz 2004). The holobiont refers
Plant: Microbe Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353 to the host and its symbiotic microbiota (Margulis 1993; Rohwer
et al. 2002), and the hologenome is defined as the sum of the
Transmission of Symbionts Between Holobiont genetic information of the host and its microbiota (Rosenberg
Generations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354 et al. 2007; Zilber-Rosenberg and Rosenberg 2008). The term
microbiome was defined by J. Lederberg as the sum of the genetic
Genetic Variation in Holobionts . . . . . . . . . . . . . . . . . . . . . . . . . . . 354 information of the microbiota (Hooper and Gordon 2001).
Until recently, studies on symbioses have concentrated on
Prebiotics and Probiotics as Applications of the a single primary symbiont and its host. However, with the advent
Hologenome Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 of molecular (culture-independent) techniques in microbiology
during the last 20 years, it is now clear that all animals and plants
live in close association with hundreds or thousands of different
Introduction microbial species. In many cases, the number of symbiotic micro-
organisms and their combined genetic information far exceed
It is useful to begin by providing some common definitions. The that of their host. In the last few years, it has been demonstrated
term symbiosis was first coined by Anton de Bary in the mid that these diverse microbiota with their large microbiomes play
nineteenth century as ‘‘the living together of different species.’’ a remarkable role in the lives of animals and plants.
This broad definition is still generally accepted. The symbiotic Evolutionary developmental biology is based on the princi-
system is usually constructed from a large partner termed the ple that evolution arises from hereditable changes in develop-
host and smaller partners called symbionts. This arbitrary divi- ment (Gilbert et al. 2010). In the past, the focus of these changes
sion by dimension between host and symbiont may not fit all has been on the host genome (genetic and epigenetic) and
348 15 Role of Microorganisms in Adaptation, Development, and Evolution of Animals and Plants: The Hologenome Concept
occasionally on the genome of a specific primary symbiont (co- eukaryotic organism is purely speculative. It is reasonable to
evolution). In this chapter, we shall summarize the hologenome assume that at least some of the early eukaryotic cells, formed
concept and then discuss the contribution of diverse microbial by the fusion of two or more prokaryotes, had the genetic
symbionts to fitness (adaptation, survival, development, growth, information that would allow for cell-to-cell interactions and
and reproduction) and evolution of representative animals and the formation of multicellular eukaryotic structures as described
plants, thereby demonstrating the idea that holobionts have above. The earliest animal that still exists is the sponge. What can
developed, lived, survived, and evolved together. The other chap- it tell us about the early evolution of animals? Costerton and
ters in this section deal with specific host: bacteria symbioses. colleagues (1995) have compared modern sponges to biofilms
because both lack tissues and organs, but are composed of a three-
dimensional matrix that allows for flow of water, nutrients,
Microbes and the Origin of Eukaryotic Cells metabolites, and oxygen. Modern sponges are well known for
and Multicellular Organisms: The First containing large complex microbial symbiotic communities—
Hologenomes more than half of the biomass of a sponge is bacteria (Reitner
and Schumann-Kindel 1997). The fossil record of sponges dem-
Prokaryotic life first appeared on Earth about 3.5 billion years onstrates their ancient association with bacteria, further indi-
ago (Schopf 1993), whereas animals first appear in the fossil cating that prokaryotic symbionts were essential components of
record approximately 500 million years ago (Hickman 2005). animals from their very beginning (Hickman 2005).
Thus, microbes were the only form of life on Earth for most of its In light of the available information, it is likely that animal
history. During the initial 3 billion years of prokaryotic life, there and plant cells arose from prokaryotic organisms by fusion,
was time for metabolic diversification and specialization. aggregated into multicellular complexes, initially using prokary-
According to the endosymbiotic theory, eukaryotic cells arose otic genetic information, and differentiated into animals and
when free-living bacteria were taken inside another bacterial cell plants, always in close association with microorganisms. During
as an endosymbiont. Mitochondria developed from evolution, animals and plants acquired additional structures and
proteobacteria, in particular, Rickettsia prowazekii or close rela- functions either by changing their DNA or by acquiring addi-
tives (Andersson et al. 1998) and chloroplasts from tional symbionts. Good examples of the latter are ruminants
cyanobacteria (Falcon et al. 2010). The origin of the nucleus is (Dehority 2003) and termites (Minkley et al. 2006), which
not clear, but one hypothesis is that it arose when ancient archaea, evolved the ability to utilize cellulose as a nutrient by incorpo-
similar to modern methanogenic archaea, invaded and lived rating cellulose-decomposing microorganisms, thereby avoiding
within bacteria similar to modern myxobacteria, eventually the very slow process of evolving novel efficient enzymes by
forming the early nucleus (Pennisi 2004). This theory is analo- themselves. This is an example of outsourcing functions from
gous to the accepted theory for the origin of eukaryotic mito- eukaryotes to prokaryotes that continues to the present.
chondria and chloroplasts. According to Margulis and Sagan
(2001), ‘‘Life did not take over the globe by combat, but by
networking’’ (i.e., by cooperation or formation of hologenomes). The Hologenome Concept
The concept that mitochondria are bacteria goes back more
than 100 years. Portier (1918) wrote ‘‘Each living cell contains in The hologenome theory of evolution considers the holobiont
its protoplasm formations which histologists designate by the with its hologenome, acting in consortium, as a unit of selection
name mitochondria. These organelles are, for me, nothing other in evolution (Rosenberg et al. 2007; Zilber-Rosenberg and
than symbiotic bacteria, which I call symbiotes.’’ However, it was Rosenberg 2008; Sharon et al. 2010). The hologenome theory
not until the 1970s that biochemical evidence demonstrated the posits that (1) all animals and plants harbor abundant and
bacterial origin of mitochondria, chloroplasts, and probably diverse microorganisms acquiring from their host a sheltered
centrioles. Recently, Davidov and Jurkevitch (2009) suggested and nutrient-rich environment, (2) these microbial symbionts
that eukaryotes arose from predatory bacteria, like Bdellovibrio, affect the fitness of the holobiont and in turn are affected by it,
that penetrated into larger bacteria and became endosymbionts. (3) variation in the hologenome can be brought about by
With regard to multicellularity, it is reasonable to assume changes in either the host genome or the microbial population
that early prokaryotes developed means of interacting among genomes (microbiome), and (4) these variations, including
themselves and with other groups in a manner that was advan- those of the microbiome, can be transmitted from one genera-
tageous, i.e., allowed them to survive and multiply more effi- tion to the next with fidelity and thus may also influence evolu-
ciently. Cell-to-cell adhesion and signaling are two mechanisms tion of the holobiont.
that are widespread in the bacterial world. Myxobacteria, for
example, when starved of nutrients, produce signals and aggre-
gate by gliding chemotaxis in order to construct species-specific All Animals and Plants Establish Symbiotic
fruiting bodies consisting of thousands of cells (Reichenbach Relationships with Microorganisms
1984). The interactions of different groups of bacteria in
biofilms and microbial mats with the exchange of metabolites As discussed above, eukaryotes presumably arose from prokary-
can be considered symbioses. The origin of the first multicellular otes and have remained in close relationship with them ever
Role of Microorganisms in Adaptation, Development, and Evolution of Animals and Plants: The Hologenome Concept 15 349
since. It is therefore not surprising that the surfaces of animals . Table 15.1
and plants contain a great abundance and variety of microor- Number of microbial species associated with representative ani-
ganisms. In addition, some microorganisms are able to grow mals and plantsa
inside animal or plant cells, i.e., endosymbionts. The number of
Minimum number
microbial cells and their combined genetic information often far
of microbial
exceed that of their hosts. For example, using the powerful Host species References
technique of metagenomics, it has recently been reported that
the number of bacterial genes present in the human gut exceeds Invertebrates
3.3 million (Qin et al. 2010). By comparison, the human host Drosophila 74 Mateos et al. (2006)
genome contains about 20,500 genes (She et al. 2004). This large melanogaster
number of microbial genes considerably increases the potential Marine sponge 1,694 Webster et al.
for adaptation by the holobiont. In addition, wherever (2001)
microbes exist, bacteriophages also occur, and they probably Coral Oculina 400 Koren and
play an important role in microbiota dynamics, which will be patagonica Rosenberg (2006)
discussed later. Tunicate 40 Martı́nez-Garcı́a
Because the vast majority of microorganisms which have et al. (2007)
been observed on or in animal and plant tissues cannot be Termite gut 367 Hongoh et al.
cultured, current research on the diversity of microorganisms (2005)
associated with a particular species relies primarily on culture- Vertebrates
free DNA-based technology. Although censuses of microorgan- Human skin 113 Grice et al. (2009)
isms associated with different animal and plant species are only Human gut 1,050 Qin et al. (2010)
in an early stage, certain interesting generalizations have
Great ape gut 8,914 Ochman et al. (2010)
emerged: (1) The diversity of microbial species associated with
Human subgingival 756 Dewhirst et al.
a particular animal or plant species is high (> Table 15.1).
plaque (2010)
(2) The host associated microbial community is very different
Bovine rumen 341 Edwards et al.
from the community in the surrounding environment (Chelius
(2004)
and Triplett 2001; Rohwer et al. 2002; Sharp et al. 2007). (3) In
some cases, it has been shown that similar, but not identical, Reindeer rumen 700 Sundset et al.
(2007)
microbial populations are found on the same eukaryote species
that are geographically separated, while different populations Pig gut 375 Leser et al. (2002)
are found on different species at the same location (Rohwer et al. Plants
2002; Lambais et al. 2006; Fraune and Bosch 2007). (4) Different Leaf of the plant 617 Lambais et al. (2006)
microbial communities often dominate different tissues of the Trichilia catigua
same organism (Tannock 1995; Koren and Rosenberg 2006; Plant seed 46 Cankar et al. (2005)
Dethlefsen et al. 2006; Reid et al. 2011). (5) In several cases Roots of Zea mays 74 Chelius and Triplett
where a large diversity of associated bacterial species exists, (2001)
certain bacterial groups dominate (Koren and Rosenberg 2006; a
These are minimum numbers because no one has sequenced enough to
Ley et al. 2008; Redford et al. 2010). detect rare species
The association of microorganisms with hosts can take many
different forms. Some may be transitory and have little effect on
adaptation or evolution of the holobiont. At the other extreme, idea that microbial diversity can play a critical role under con-
there are several examples of well-studied long-lasting interac- ditions of fluctuating environments has been referred to as the
tions (e.g., the rumen system) between host and microorgan- insurance policy hypothesis (Yachi and Loreau 1999). Another
isms, which can lead to total dependence of one on the other. factor that contributes to bacterial diversity is bacteriophages. It
Between these two extremes lies a gradient of interactions of has been established that high concentrations of bacteriophages
varying strengths, including pathogenesis. Let us consider some are present in animal and plant tissues (Breibart et al. 2003). If
factors that determine the diversity of microorganisms associ- any microorganism becomes too abundant, it may be lysed by
ated with the holobiont. We shall first consider those character- bacteriophages. This concept, referred to as the ‘‘kill the win-
istics that would result in high diversity. Many microorganisms ners’’ hypothesis (Thingstad and Lignell 1997), is supported by
are specialists. Given that hosts provide a variety of different mathematical models of the bacteria: bacteriophage dynamics
niches that can change with the developmental stage of the host, (Weitz et al. 2005).
the diet, and other environmental factors, a diverse microbial On the other hand, there exist opposing forces that limit the
community is established, with different microbial strains filling number of strains that can survive and become established in the
the different niches. This microbial diversity, and therefore its holobiont, notably the innate and adaptive immune systems.
versatility, may allow the holobiont as a whole to function more The innate or nonspecific immune system is the first line of
optimally and adapt more rapidly to changing conditions. The defense and includes physical barriers, antimicrobial molecules,
enzymes, specific binding proteins for microbial attachment Invertebrates

(e.g., the peptidoglycan-binding protein and lectin complement
system), production of reactive oxygen species and phagocytes Insect Holobionts
(Iwanaga and Lee 2005). Interestingly, resident symbiotic bac-
teria are also part of the innate immune system—by occupying Insects are the most diverse animal group on earth, embracing
potent adhesion sites and by producing antibiotics (Ritchie several million species (Wilson 1992). Insect–microbe symbioses
2006). The adaptive or the specific immune system in vertebrates take many forms: Some are endocellular and many more are
includes specific recognition of ‘‘foreign’’ microorganisms, gener- extracellular. In some insects, obligate mutualistic bacteria are
ation of responses to eliminate these microorganisms, and devel- harbored in specialized host cells. For example, Buchnera
opment of immunological memory to hasten the response to aphidicola is harbored intracellularly within bacteriocytes in
subsequent infections with the same microbe. In essence, the the abdominal body cavity of almost all aphids and provides
immune system of the host is responsible for both limiting the essential amino acids that are lacking in the phloem sap diet of
types of microorganisms that can survive within the host and the insects (Douglas 1998), and Wigglesworthia glossinidia is
recognizing and accommodating the normal microbiota, thereby localized in a midgut-associated bacteriome of tsetse flies and
regulating the kinds of microorganisms that can reside in the plays pivotal roles in biosynthesis of B vitamins that are deficient
holobiont. Also, it is important to note that plants have evolved in the blood diet of the insects (Akman et al. 2002). These obligate
myriad phytochemicals, whose purpose is to prevent infection by P-symbionts usually share long evolutionary histories with their
harmful microorganisms (Wallace 2004) and enable coexistence hosts and, in most cases, the host cannot survive without the
with beneficial ones (Smith et al. 1999; Stougaard 2000; endosymbiont, or the elimination of the endosymbiont has
Wilkinson 2001). The human gut is an example in which, a deleterious effect on the fitness of the host (Baumann et al. 2006).
although one finds a substantial diversity of bacterial strains, Wolbachia, a Gram-negative bacterium of the alpha-
they belong by and large to only 30–40 species, which themselves proteobacteria group, is a common obligate intracellular para-
belong to two main bacterial divisions (out of 70 divisions site of insects and other invertebrates. It is probably the most
identified), the Firmicutes and the Bacteroidetes, while the ubiquitous endosymbiont on the planet (Dedeine et al. 2001)
Archaea are represented mainly by only one strain, Methanobre- and is maternally transmitted through the cytoplasm of eggs.
vibacter smithii (Ley et al. 2006a). First recognized as the cause of some incompatible crosses in
insects (Yen and Barr 1971), Wolbachia has since been identified
as a cause of parthenogenesis, feminization of male hosts, and
Cooperation Between the Host and the male killing in different arthropod taxa (Veneti et al. 2005).
Microbiota Contributes to the Fitness of the Recent studies using molecular techniques have brought new
Holobiont insights into the mechanisms by which microbial symbionts
digest cellulose in the small intestine of insects (Watanabe and
Considering the hologenome concept, we argue that the coop- Tokuda 2010). If the available diet of an insect changes from
eration between the normal microbiota and the host generally simple sugars to complex polysaccharides, those symbionts,
leads to improved fitness of the holobiont, by the host which contain the appropriate polysaccharidases will amplify
outsourcing (Gilbert et al. 2010) different kinds of functions to in number, depolymerize the polysaccharides, and allow the
its microbiota and vice versa. > Table 15.2 displays various insect to grow efficiently. Termites, for example, have
representative symbiotic systems, indicating some of the ways a multitude of different microorganisms in their hindgut
in which the microorganisms contribute to the fitness of the (Warnecke et al. 2007) that are largely responsible for the break-
holobiont. The first cases listed in the table are the mitochondria down of lignocelluloses (Breznak and Brune 1994) and nitrogen
and chloroplasts, which can be considered (extreme) symbionts fixation (Golichenkov et al. 2002). It has been shown that dif-
which are responsible for respiration and photosynthesis, ferent bacterial phylogenetic groups are present in the different
respectively. Animal–microbe symbioses take many forms: Few gut compartments (Schmitt-Wagner et al. 2003).
are endocellular and termed ‘‘primary’’ symbionts (P-symbi- It has also been suggested that microbes have been powerful
onts) and the vast majority are extracellular symbionts. While selective agents in the development of social behavior in insects,
some of the symbioses show absolute dependency, most of the such as ants, bees, wasps, and termites (Stow and Beattie 2008):
symbiosis systems, as indicated in > Table 15.2, are not based On one hand, close contact between community members
on life or death interactions, but rather the microbial partners ensures that beneficial microorganisms are transmitted from
contribute in different degrees to the holobiont’s well-being. The one generation to the next; on the other hand, it provides ideal
hologenome concept emphasizes the importance not only of the conditions for transfer of contagious diseases. To help solve this
intracellular symbionts but also of the diverse and dynamic problem, many social insects contain symbiotic bacteria, which
extracellular microbial symbionts that are present in all animals. produce antibiotics active against pathogens (Currie et al. 2006).
The enormous genetic richness of these symbionts can play Diet-induced mating preference in Drosophila was reported
a major role in adaptation and evolution of holobionts during many years ago (Dodd 1989); however, the mechanism was
times of environmental change. Some examples of well- unknown until a recent demonstration that changing the diet
described symbioses are summarized below. caused an amplification of a particular bacterial symbiont,
. Table 15.2
Modes of transmission of symbionts and their contribution to the fitness of the holobiont
Mode of transmission of
Holobiont: microbiota microorganisms Microbial contribution References
General
All eukaryotes: Mitochondria Cytoplasmic inheritance Respiration Andersson et al. (1998)
Plants: Chloroplasts Cytoplasmic inheritance Photosynthesis Falcon et al. (2010)
Invertebrates
Aphids: Buchnera sp. Via intracellular bacteria in Provision of specific required amino Baumann et al. (2006), Douglas (1998)
(primary-endosymbiont) bacteriocytes; present in ova acids lacking in the plant sap diet
Aphids: Secondary Via intracellular bacteria in Growth at high temperature; Sandström et al. (2001), Russell et al.
endosymbionts addition to environment resistance to parasites (2003)
Termite: Microbiota in hind Feces of adult termites fed to Utilizable energy and carbon; Abe et al. (2000), Minkley et al. (2006),
gut newly hatched juveniles nitrogen metabolism, recognition Watanabe and Tokuda (2010)
signal from odor of bacterial
metabolites
Anthropods/nematodes: Intracellular transmission via Fertility and sex determination Veneti et al. (2005)
Wolbachia spp. egg cytoplasm
Drosophila: Vibrio plantarum Feces of mother placed on Mating preference Sharon et al. (2010)
egg mass
Stinkbug midgut: Specific transmission via More efficient food utilization Kikuchi et al. (2007)
Burkholderia environment
Squid nidamental gland: Via cover of eggs originating Protection of eggs and embryos Kaufman et al. (1998), Barbieri et al.
Microbiota from the gland against pathogens (2001)
Squid light organ: Vibrio Environmental from Camouflage against predators McFall-Ngai (1999)
fischeri surrounding water
Corals: Microbiota From the environment and Photosynthesis (intracellular algae); Rohwer et al. (2002), Buddemeier et al.
by vegetative reproduction nitrogen fixation; protection against (2004), Rosenberg et al. (2007)
pathogens
Sponges: Microbiota Environmental in addition to Breakdown of complex polymers; Webster et al. (2001), Hickman (2005),
possible transmission from nitrogen cycling; protection against Taylor et al. (2007)
parent pathogens
Vertebrates
Cow rumen: Microbiota Physical contact with Provision of all nutritional needs from Dehority (2003), Russell and Rychlik
parents and via grass cellulose (2001)
contaminated with feces
and sputum
Whale forestomach: Physical contact with mother Provision of nutrition from chitin and Herwig et al. (1984), Olsen et al. (2000)
Microbiota other complex organics
Human gut and mouse Via passage through the Protection against pathogens; Silva et al. (2004), Stecher and Hardt
model: Microbiota birth canal, physical contact stimulation of immune system; (2008), Stappenbeck et al. (2002),
and from environment angiogenesis; vitamin synthesis; fiber O’Hara and Shanahan (2006), Mai et al.
breakdown; fat metabolism, obesity (2010), Dethlefsen et al. (2006), Ley
and related disorders et al. (2006a), Cani and Delzenne
(2009)
Plants
Legume plants: Rhizobium Environmental from Nitrogen fixation Stougaard (2000), Jones et al. (2007),
surrounding Ott et al. (2009)
Plant: Growth promoting Environmental from Protection against pathogens; Singh et al. (2004), Lugtenberg and
rhizobacteria surrounding soil nitrogen metabolism; acceleration of Kamilova (2009), Lindow and Brandle
mineralization; carbon cycling; salt (2003), Whipps et al. (2008)
tolerance
Rice plants: Azoarcus sp. From surrounding soil Associative nitrogen fixation Hurek and Reinhold-Hurek (2003)
Lactobacillus plantarum, and that this bacterium was responsible Vertebrates

for the mating preference (Sharon et al. 2010). The combination
of partial geographic separation and bacterial-induced mating Symbioses between diverse microbiota and vertebrates have
preference could reduce interbreeding of the populations. been studied in a variety of animals, including ruminants
Slower changes in the host genome would further enhance the (Dehority 2003), chickens (Abbas Hilmi et al. 2007), whales
mating preference. The stronger the mating preference, the (Olsen et al. 1994), gorillas (Frey et al. 2006), and rats (Brooks
greater the chance that two populations will become sexually et al. 2003). We would like to present here briefly what is
isolated, and biologists (Coyne 1992; Schluter 2009) have argued probably the best studied metabolic system in vertebrates,
that the emergence of sexual isolation is the central event in the namely, the human/mouse gut symbiosis. This system has pro-
evolution of species. These data strongly support the vided a wealth of detailed information on how diverse extracel-
hologenome theory of evolution which posits that microorgan- lular symbionts contribute to the health of animal holobionts.
isms can play a key role in the evolution of animals and plants The number of bacteria in and on a typical healthy human is
(Zilber-Rosenberg and Rosenberg 2008). 1014 to 1015. Most of these microbes are not merely in transit but
rather inhabit defined niches. Although about 90% of the
human microbiota are found in the gastrointestinal tract, there
Squid Light Organ–Vibrio Symbiosis is a high abundance and diversity of microbes on all surfaces of
the human body, including skin, oral cavity, nasal cavity, phar-
The symbiosis between the Hawaiian bobtail squid Euprymna ynx, esophagus, and urogenital tract.
scolopes and the luminous bacterium Vibrio fischeri is one of best Protection against infectious disease is one of the important
studied systems that demonstrate how a bacterial symbiont can attributes of the resident microbiota. Most bacterial pathogens
play a role in the development of an animal organ (Ruby 1996; infect their human hosts predominantly via mucosal surfaces of
Nyholm and McFall-Ngai 2004). Following fertilization of the the respiratory, urogenital, or gastrointestinal tracts. In addition
eggs within the female, the embryos develop an immature light to mechanical and immunological barriers, mucosal surfaces are
organ that is free of bacteria but has three pores leading to protected against pathogen infection by the high concentration
separate epithelial-lined crypts. The female host lays clutches of microbiota colonizing the mucosa. The exact mechanism is
of hundreds of eggs, which hatch almost synchronously at dusk. unknown, but it has been suggested that resident bacteria
Within hours after hatching, the juvenile squid becomes colo- occupy binding sites needed by pathogens for adhesion in addi-
nized by V. fischeri, which triggers morphogenesis of the light tion to releasing antibacterials active against pathogens (Belden
organ (Montgomery and McFall-Ngai 1994). and Harris 2007). As an experimental example of bacterial
protection against infection, mice were treated with
Bifidobacterium longum, part of the normal microbiota, and
Coral: Bacteria Symbiosis then infected with the pathogen Salmonella typhimurium. The
mice that received B. longum survived, whereas the control
The hundreds of different bacterial species that are associated group (Salmonella alone) all died within a few days (Silva et al.
with the coral mucus, tissues, and skeleton are essential for coral 2004). Interestingly, it has been demonstrated that immune
health. A substantial part of the coral nitrogen requirement is response to integral microbiota via IgM differs from its reaction
provided by nitrogen-fixing coral bacteria (Shashar et al. 1994). to pathogenic microorganisms (Hapfelmeier et al. 2010).
Some of the bacterial symbionts degrade complex polysaccha- Another important known beneficial function of microbiota
rides, such as chitin (Ducklow and Mitchel 1979), thereby pro- is participation in the development and normal function of the
viding nutrients to the coral holobiont; others are able to protect innate and adaptive immune systems in the gut (O’Hara and
the coral against pathogens by producing antibiotics (Ritchie Shanahan 2006; Ivanov and Littman 2011) while creating
2006; Nissimov et al. 2009; Shnit-Orland and Kushmaro 2009). a permissive, noninflammatory environment for their own pres-
Microbial symbionts of corals also play a major role in ence (Hapfelmeier et al. 2010). The gut, in spite of providing
adaptation to changing environmental conditions. When sea- a diversity of niches for microbial colonization, imposes strict
water temperature exceeds the normal maximum by a few requirements for their survival, the important ones being adap-
degrees, corals lose their symbiotic zooxanthellae, a process tation to digestive enzymes, evading the potent innate and
referred to as bleaching. If the process is not reversed in adaptive immune systems, escaping washout from the gut, and
a reasonable time, the coral will die. The adaptive hypothesis the ability to live anaerobically. These strict requirements will
of coral bleaching (Buddemeier et al. 2004) puts forth the force a narrowing of the variability of microorganisms, leaving
concept that expulsion of the algae allows more temperature- only those that are able to create a viable and well-adapted
resistant zooxanthellae to infect the coral and establish a more holobiont with their host. Bacteria are critical in promoting
favorable symbiosis. Another adaptive process of bleached corals the early development of the gut’s mucosal immune system
is the amplification of cyanobacteria in the coral skeleton. The both in terms of its physical components and its function and
photosynthetic products of these bacteria are transferred to the continue to play a role later in life in its operation. The
host and have been suggested to help them survive the bleaching microbiota also plays a key role in angiogenesis, the structural
episodes (Fine and Loya 2002). buildup of blood vessels (Stappenbeck et al. 2002). Bacteria
found in the gut synthesize and excrete vitamins in excess of cooperation for development, growth, and survival (van der
their own needs, which can be absorbed as nutrients by their Heijden et al. 2008).
host. For example, in humans, enteric bacteria secrete Vitamin Studies on the microbiology of plants have been performed
K and Vitamin B12, and lactic acid bacteria produce certain with microorganisms found in three main locations: around the
B vitamins (Mai et al. 2010). roots (rhizosphere), on the leaves, stems, flowers, and fruit
Moreover, germ-free animals are deficient in Vitamin K to (phyllosphere), and inside plant cells (endophytes). The great
the extent that it is necessary to supplement their diets (Komai majority of microorganisms have different degrees of beneficial
et al. 1988). The human gut microbiota is a complex ecosystem relationships with plants, whereas only a small minority is
that plays an essential role in the catabolism of dietary fibers, the parasitic. The close cooperation between plants and microorgan-
part of plant material in our diet that is not metabolized in the isms necessitates overcoming the plant’s immune response and
upper digestive tract, because the human genome does not often using some of its components together with other elements of
encode adequate enzymes (Dethlefsen et al. 2006). the plant and some functions of the microbiota for enabling this
Germ-free animals, born and grown under sterile condi- interaction to occur (Bucher et al. 2009; Bednarek et al. 2010).
tions, are a useful tool for studying the relationship between In addition to fungi, many bacterial species interact with
host and its microbiota. Studies on germ-free mice exhibit plant roots. The multitude of bacterial species contribute to
significant differences in gut development, function, and regu- carbon transfer to soil, nitrogen fixation, nitrate reduction,
lation when compared with mice grown conventionally, i.e., mineralization of organic materials, maintenance of soil struc-
possessing normal gut microbiota. The germ-free mice demon- ture and water cycling, protection against pathogens and other
strate enlarged ceca (Wostmann 1981), a slow digested food stress conditions, all of which promote plant growth directly or
transit time (Abrams and Bishop 1967), altered kinetics of indirectly (Singh et al. 2004; Lugtenberg and Kamilova 2009).
epithelia turnover in the small intestine (Savage et al. 1981), an Rhizosphere microbial communities differ between plant spe-
increased caloric intake (Wostmann et al. 1983), and a greater cies (Innes et al. 2004; Berg and Smalla 2009), between ecotypes
susceptibility to infection (Silva et al. 2004). The influence of within species (Micallef et al. 2009), between different develop-
microbiota on energy metabolism in germ-free mental stages of a given plant (Weisskopf et al. 2006), and from
conventionalized mice was observed within 2 weeks of the intro- those present in bulk soil (Broz et al. 2007). Microorganisms in
duction of microbiota (Bäckhed et al. 2007). It included micro- the rhizosphere are selected for their functional abilities no less
bial fermentation of polysaccharides not digested by the host, than for their taxonomy (Singh et al. 2004). Moreover, it has
absorption of the microbially produced short-chain fatty acids, been shown that plant’s specific exudates are major contributors
more efficient absorption of the monosaccharides from the to the plant specificity of rhizosphere microbiota (Somers et al.
intestine, conversion of breakdown products in the liver to 2004; Singh et al. 2004; Berg and Smalla 2009).
more complex lipids, and microbial regulation of host genes Large populations of microorganisms live also in the
that promote fat deposition in adipocytes. These events were phyllosphere. Archaea, filamentous fungi, and yeast are present
accompanied by lower food intake and a higher metabolic rate. in the phyllosphere, but bacteria are considered to be the dom-
It has been shown in mice (Ley et al. 2005; Turnbaugh et al. inant microbial inhabitants present on the plant surface and
2006) and humans (Ley et al. 2006b) that obesity is correlated within the plant tissue (Whipps et al. 2008). Stressful conditions
with different bacterial communities and that a gradual transi- on the leaves, such as extreme temperatures and dryness, irradi-
tion occurs in humans from the obese microbiota to the lean ation, and oxidative stress in addition to poor nutrient availabil-
microbiota during a course of a restrictive energy intake ity, determine the kinds of bacteria, their mode of growth, and
(Ley et al. 2006b). Moreover, obese microbiota have been their activities (Lindow and Brandl 2003). Most information on
implicated in obesity-related metabolic disorders such as the microbial communities in the phyllosphere has been
type 2 diabetes, inflammation, disordered lipid metabolism, established using culture-dependant methods, and much of it
and atherosclerosis, in addition to fatty liver, primarily via is on pathogenic bacteria and fungi. The global surface area of
bacterial Gram-negative LPS (lipopolysaccharide) metabolic the phyllosphere, estimated to be 4 108 km2, harbors a bacte-
effects (Cani and Delzenne 2009; Caesar et al. 2010; Abu-Shanab rial population in the region of 1026 cells including 2–3 106
et al. 2010). species (Whipps et al. 2008). Interestingly, culture-independent
techniques have revealed that similarly to the human gut, these
species fall within a relatively small number of dominant phyla,
Plant: Microbe Symbioses the proteobacteria being the most abundant on leaves (Delmotte
et al. 2009; Redford et al. 2010). This phenomenon is in accor-
Plant abundance, diversity, and activities are essential for life on dance with the special conditions known to occur in the
our planet, and microorganisms play a central role in all three phyllosphere, demanding specific adaptations and activities
phenomena. Microorganisms supply plants with nutrients, play (Delmotte et al. 2009). Given the high mass of phyllosphere
a role in establishment of plants and the development of root microbiota, it is likely that they play an important role in global
systems and in protection against pathogens and other environ- transformation of matter, including recycling of carbon and
mental stress conditions. Moreover, it is estimated that about nitrogen. In addition, they contribute to the plant’s fitness
20,000 species of plants are obligatorily dependant on microbial mainly through spatial protection against pathogens, promotion
of growth, and deterrence of herbivores (Lindow and Brandl (McFall-Ngai 1999). Furthermore, the squid provides a habitat
2003; Whipps et al. 2008). Microorganisms are unevenly distrib- in which only V. fischeri that emits light is able to maintain
uted, mainly on the lower part of leaves, as single or aggregated a stable association (McFall-Ngai 1999; Visick et al. 2000).
microorganisms (Whipps et al. 2008). Culture-independent Thus, even in transfer via the environment (often referred to as
methods have shown that by and large community pattern of horizontal transfer), the holobiont is reconstituted faithfully.
the phyllosphere bacteria correlates with the tree phylogeny even Direct contact is another slightly less direct mode of trans-
across continents (Yang et al. 2001; Redford et al. 2010), though mission demonstrated in mammals in which many of the sym-
not all studies are in agreement (Whipps et al. 2008). In addi- bionts are derived during passage through the birth canal or
tion, though bacterial leaf communities differ between seasons, subsequently by close physical contact with parent or family and
similar ones are found on leaves sampled during the same community members. In humans, for example, a greater simi-
season, and this pattern is predictable from year to year (Ercolani larity was observed within-family members as compared to
1991; Redford and Fierer 2009). between families (Zoetendal et al. 2001) and within the same
The best studied plant: bacteria symbiosis involves nitrogen- European population as compared with between different Euro-
fixing legume holobionts. Several specialized kinds of bacteria, pean populations (Mueller et al. 2006).
including the most studied, Rhizobium, engage in symbiotic Some animals and most plants can develop from cells other
relationships with peas, soybeans, and other legumes to convert than gametes, namely, from somatic cells. The most striking
nitrogen gas into ammonia and further into organic nitrogen- example is vegetative reproduction in plants. When a fragment
containing compounds. Rhizobia are highly specific for their of a plant falls to the earth, it may root and grow into a fully
plant host. Their specificity arises, in part, from chemical ‘‘cross- developed plant. In such cases, it will clearly contain some of the
talk’’ between the bacteria and plant (Hardison 1996). Thus, symbionts of the original plant (direct transfer). In addition, it
a two-way conversation between the bacterium and its plant will most likely incorporate microorganisms from the soil adja-
host is responsible for the development of the nodule and its cent to the parent.
nitrogen-fixing capability. The Rhizobium–legume symbiosis is Most studies demonstrating the transfer of symbionts from
discussed in detail in another chapter of this section of the book. parent to offspring are short term. However, there is also evi-
dence that symbionts are maintained for many generations.
Fraune and Bosch (2007) demonstrated the specificity and
Transmission of Symbionts Between long-term accuracy of transmission in the metazoan Hydra. Sim-
Holobiont Generations ilarly to many plants, Hydra reproduce vegetatively (by budding)
and sexually. The researchers showed, first, that two different
The hologenome theory of evolution relies on ensuring the species of Hydra were colonized by different communities of
continuity of partnerships between holobiont generations. microorganisms and, second, in both cases, the two species of
> Table 15.2 presents some of the diverse modes of transmission Hydra were populated with similar microorganisms both in the
of symbionts in animals and plants. Mitochondria and chloro- laboratory and in nature, even after more than 30 years of
plasts are transmitted by the most direct mode, namely, cyto- maintaining the animals in the laboratory. In addition, it has
plasmic inheritance. Direct transmission from parent to been reported recently that the gut microbiota among great ape
offspring also occurs with other symbionts where the microor- species is phylogenetically conserved over evolutionary timescales
ganisms are in or on the reproductive cells. For example, in the and has diverged in a manner consistent with vertical inheritance
aphid–Buchnera symbiosis, bacteria are intracellularly situated (Ochman et al. 2010). In humans, the conservation of specific
in bacteriocytes and are transferred to and transmitted via the strains of Helicobacter pylori between generations has been used
eggs (Baumann et al. 2006). Another slightly less direct mode of as a window into human migration (Devi et al. 2006).
transmission is used in the termite hindgut: microbiota symbi-
osis where feces of adult termites (containing abundant micro-
organisms) are fed to newly hatched juveniles by workers in the Genetic Variation in Holobionts
colony (Abe et al. 2000). Similarly, in the bovine rumen–
microbiota symbiosis, the offspring acquire the microbiota by Genetic variation is the raw material for evolution. Genetic
feeding on grass that is contaminated with feces and sputum variation in holobionts can arise from changes in either the
from their parents, as well as by passage through the birth canal host or the symbiotic microbiota genomes. Variation in host
(Dehority 2003). In general, inoculating newborns with the feces genomes occurs during sexual reproduction, chromosome
of their parents is widespread in the animal world. rearrangements, and ultimately by mutation. These same pro-
A less direct, but precise, mode of transmission is exempli- cesses occur in microorganisms with the noteworthy difference
fied in the squid light organ: Vibrio fischeri symbiosis where the that in haploid bacteria, recombination occurs, within the same
high specificity of the light organ for V. fischeri has evolved species, by conjugation, transduction, and DNA transformation,
together with the need to acquire the motile bacteria from the and, between species, by horizontal gene transfer. In addition,
surrounding seawater. The adult squid releases large amounts of changes in the hologenome of the holobiont can occur by two
V. fischeri into the water at dawn every day, assuring that other processes that are specific to holobionts: microbial ampli-
sufficient symbionts are available to colonize the hatchlings fication and acquisition of novel strains from the environment.
Microbial amplification is the most rapid and easy to under- probiotics appears to capture our current knowledge of the
stand mode of variation in holobionts. It involves changes in the subject: ‘‘Live microorganisms which, when administered in
relative numbers of the diverse types of associated microorgan- adequate amounts, confer a health benefit on the host’’
isms that can occur as a result of changing temperatures (for (Hoffman et al. 2008). Unlike variation of the hologenome by
plants and invertebrates), nutrient availability or quality, expo- acquisition of microbes from the environment, probiotic tech-
sure to antibiotics, or other environmental factors. The nology involves the nonrandom introduction of specific bacteria
holobiont is a dynamic entity with certain microorganisms to improve the health of the host. The effects can even be
multiplying and others decreasing in number as a function of transferred to the next generation as was shown in a study
local conditions. Accordingly, an increase in the number of (Luoto et al. 2010a, b) where pregnant women treated with
a particular microbe is equivalent to gene amplification. Con- Lactobacillus rhamnosus GG and Bifidobacterium lactis had
sidering the large amount of genetic information encoded in the reduced frequency of gestational diabetes mellitus and dimin-
diverse microbial population of holobionts, as mentioned ished risk of larger birth size in affected cases. In view of the fact
above, microbial amplification is a powerful mechanism for that large birth size is a risk factor for later obesity, the present
adapting to changing conditions. In fact, changes of symbiont results are of significance for public health in demonstrating that
populations as a function of external factors are well this risk is not only modifiable but also transferred from one
documented (Weimer et al. 1999; Russell and Rychlik 2001; de generation to the next. Going one step further, we speculate that
la Cruz and Davies 2005; Koren and Rosenberg 2006; Blaser and since probiotics can be transmitted to future generations, they
Falkow 2009). Acquiring new symbionts from the environment could, in principle, be used to treat not only metabolic and
is another mechanism for introducing variation into holobionts. alimentary tract diseases such as diabetes, coronary heart dis-
Animals and plants come in contact with billions of microor- ease, and alimentary tract diseases, but also genetic and behav-
ganisms during their lifetime. It is reasonable to assume that ioral human diseases. In addition, probiotics could also be used
occasionally, as a random event, some of these microorganisms more intensely in agriculture to combat diseases and improve
will find a niche and become established in the host. Unlike yields and achieve specific characteristics in plants and animals.
microbial amplification, acquiring novel symbionts can intro-
duce entirely new genes into the holobiont. Under the appro-
priate conditions, the new symbionts may become more
abundant and affect the phenotype of the holobiont. It is often
References
technically difficult to distinguish between amplification of
Abbas Hilmi HT, Surakka A, Apajalahti J, Saris PE (2007) Identification of the
a low abundant indigenous microbe and acquisition from the most abundant Lactobacillus species in the crop of 1- and 5-week-old broiler
environment. chickens. Appl Environ Microbiol 73:7867–7873
Abe T, Bignell DE, Higashi M (eds) (2000) Termites: evolution, sociality, symbi-
oses, ecology. Kluwer, Dordrecht
Abrams GD, Bishop JE (1967) Effect of normal microbial flora on gastrointestinal
Prebiotics and Probiotics as Applications of
motility. Proc Soc Exp Biol Med 126:301–304
the Hologenome Concept Abu-Shanab A, Eamonn MM, Quigley EMM (2010) The role of the gut
microbiota in nonalcoholic fatty liver disease. Nat Rev Gastro Hepatol
Prebiotics and probiotics are two technological applications of 7:691–701
the hologenome concept (Zilber-Rosenberg and Rosenberg Akman LA, Yamashita H, Watanabe A et al (2002) Genome sequence of the
endocellular obligate symbiont of tsetse flies, Wigglesworthia glosssinidia.
2011). Prebiotics has been defined as follows: ‘‘a selectively
Nat Genet 32:402–407
fermented ingredient that allows specific changes, both in the Andersson SGE, Zomorodipour A, Andersson JO (1998) The genome sequence of
composition and/or activity in the gastrointestinal microflora Rickettsia prowazekii and the origin of mitochondria. Nature 396:133–143
that confers benefits upon host well-being and health’’ (Gibson Barbieri E, Paster BJ, Hughes D, Zurek L, Moser DP, Teske A, Sogin ML
et al. 2004). The implication of this definition is that prebiotics in (2001) Phylogenetic characterization of epibiotic bacteria in the accessory
nidamental gland and egg capsules of the squid Loligo pealei (Cephalopoda:
the form of functional carbohydrates change the gut microbiota
Loliginidae). Environ Microbiol 3:151–167
and its metabolic activity similarly to natural fiber present in the Bäckhed F, Manchester JK, Semenkovich CF, Gordon JI (2007) Mechanisms
normal human diet, except that prebiotics are targeted at specific underlying the resistance to diet-induced obesity in germ-free mice. Proc
bacteria and are degraded in a specific form and therefore are Natl Acad Sci USA 104:979–984
more controlled. In essence, prebiotics cause variation in the Baumann P, Moran NA, Baumann L (2006) Bacteriocyte-associated endosymbi-
onts of insects. In: Dworkin M, Rosenberg E, Schleifer KH, Stackebrandt E
holobiont by amplification of beneficial bacteria. Both fiber and
(eds) The prokaryotes, vol 1. Springer, New York, pp 403–438
prebiotics have been shown to exert health benefits, the main ones Bednarek P, Kwon C, Schulze-Lefert P (2010) Not a peripheral issue: secretion in
being: improving or stabilizing microbiota, improvement of plant–microbe interactions. Curr Opin Plant Biol 13:378–387
intestinal function, increase of mineral absorption, modulation Belden LK, Harris RN (2007) Infectious diseases in wildlife: the community
of gastrointestinal excretions, improving energy metabolism and ecology context. Front Ecol Environ 5(10):533–539
Berg G, Smalla K (2009) Plant species and soil type cooperatively shape the
satiety, modulation of intestinal immune functions, and reduc-
structure and function of microbial communities in the rhizosphere. FEMS
tion in the risk of intestinal infection. Microbiol Ecol 68:1–13
Probiotics have been defined in different ways since 1965 Blaser MJ, Falkow S (2009) What are the consequences of the disappearing
when initially proposed. The following accepted definition of human microbiota? Nature Rev Microbiol 7:887–894
Breibart M, Hewson I, Felts B et al (2003) Metagenomic analysis of an uncultured Fraune S, Bosch TCH (2007) Long-term maintenance of species-specific bacterial
viral community from human feces. J Bacteriol 185:6220–6223 microbiota in the basal metazoan Hydra. Proc Natl Acad Sci USA 104:13146–
Breznak JA, Brune A (1994) Role of microorganisms in the digestion of lignocel- 13151
lulose by termites. Ann Rev Entomol 39:453–487 Frey JC, Rothman JM, Pell AN et al (2006) Fecal bacterial diversity in a wild
Brooks SPJ, McAllister M, Sandoz M, Kalmokoff ML (2003) Culture-independent gorilla. Appl Environ Microbiol 72:3788–3792
phylogenetic analysis of the faecal flora of the rat. Can J Microbiol Gibson GR, Probert HM, Van Loo J, Rastall RA, Roberfroid MB (2004) Dietary
49:589–601 modulation of the human colonic microbiota: updating the concept of
Broz AK, Manter DK, Vivanco JM (2007) Soil fungal abundance and prebiotics. Nutr Res Rev 17(2):259–275
diversity: another victim of the invasive plant Centaurea maculosa. ISMEJ Gilbert SF, McDonald E, Boyle N et al (2010) Symbiosis as a source of selectable
1:763–765 epigenetic variation: taking the heat for the big guy. Philos Trans R Soc Lond B
Bucher M, Wegmüller S, Drissner D (2009) Chasing the structures of small 365:671–678
molecules in arbuscular mycorrhizal signaling. Curr Opin Plant Biol Golichenkov MV, Kostina NV, Ul’yanova TA et al (2002) Specific features of
12:500–507 nitrogen fixation and denitrification in termites Neotermes castaneus,
Buddemeier RW, Baker AC, Fautin DG, Jacobs JR (2004) The adaptive hypothesis Zootermopsis angusticollis, and Reticulitermes lucifugus. Biol Bull 29:172–175
of bleaching. In: Rosenberg E, Loya Y (eds) Coral health and disease. Grice EA, Kong HH, Contan S et al (2009) Topographical and temporal diversity
Springer, New York, pp 427–444 of the human skin microbiome. Science 324:1190–1192
Caesar R, Fåk F, Bäckhed F (2010) Effects of gut microbiota on obesity and Hapfelmeier S, Lawson MAE, Slack E et al (2010) Reversible microbial coloniza-
atherosclerosis via modulation of inflammation and lipid metabolism. tion of germ-free mice reveals the dynamics of IgA immune responses.
J Intern Med 268:320–328 Science 328:1705–1709
Cani PD, Delzenne NM (2009) The role of the gut microbiota in energy metab- Hardison RC (1996) A brief history of hemoglobins: plant, animal, protist, and
olism and metabolic disease. Curr Pharm Des 15:1546–1558 bacteria. Proc Natl Acad Sci USA 93:5675–5679
Cankar K, Kraigher H, Ravnikar M, Rupnik M (2005) Bacterial endophytes from Herwig RP, Staley JT, Nerini MK, Braham HW (1984) Baleen whales: preliminary
seeds of Norway spruce (Picea abies L. Karst). FEMS Microbiol Lett 244:341– evidence for forestomach microbial fermentation. Appl Environ Microbiol
345 47:421–423
Chelius MK, Triplett EW (2001) The diversity of Archaea and bacteria in associ- Hickman CS (2005) How have bacteria contributed to the evolution of
ation with the roots of Zea mays. Microb Ecol 41:252–263 multicellular animals? In: McFall-Ngai MJ, Henderson B, Ruby EG (eds)
Costerton JW, Lewandowski Z, Caldwell DE, Korber DR, Lappin-Scott HM The influence of cooperative bacteria on animal host biology. Cambridge
(1995) Microbial biofilms. Annu Rev Microbiol 49:711–745 University Press, New York, pp 3–33
Coyne JA (1992) Genetics and speciation. Nature 355:511–515 Hoffman FA, Heimbach JT, Sanders ME, Hibberd PL (2008) Executive Summary:
Currie CR, Poulsen M, Mendenhall J et al (2006) Coevolved crypts and exocrine Scientific and regulatory challenges of development of probiotics as food and
glands support mutualistic bacteria in fungus-growing ants. Science drugs. Clinical Infect Dis 46:S53–S57
311:81–83 Hongoh Y, Deevong P, Inoue T et al (2005) Intra- and interspecific comparisons
Davidov Y, Jurkevitch E (2009) Predation between prokaryotes and the origin of of bacterial diversity and community structure support coevolution of gut
eukaryotes. Bioessays 31:748–757 microbiota and termite host. Appl Environ Microbiol 71:6590–6599
Dedeine F, Vavre F, Fleury F et al (2001) Removing symbiotic Wolbachia bacteria Hooper LV, Gordon JI (2001) Commensal host-bacterial relationships in the gut.
specifically inhibits oogenesis in a parasitic wasp. Proc Natl Acad Sci USA Science 292:1115–1118
98:6247–6252 Hurek T, Reinhold-Hurek B (2003) Azoarcus sp. strain BH72 as a model for
Dehority BA (2003) Rumen microbiology. Nottingham University Press, nitrogen-fixing grass endophytes. J Biotechnol 106:169–178
Nottingham Innes L, Hobbs PJ, Bardgett RD (2004) The impacts of individual plant species on
de la Cruz F, Davies J (2005) Industrial revolution and microbial evolution. In: rhizosphere microbial communities in soils of different fertility. Biol Fertil
McFall-Ngai MJ, Henderson B, Ruby EG (eds) (2005) The influence of Soils 40:7–13
cooperative bacteria on animal host biology. Cambridge University Press, Ivanov II, Littman DR (2011) Modulation of immune homeostasis by commensal
New York, pp 73–82 bacteria. Curr Opin Microbiol 14:106–114
Delmotte N, Knief C, Chaffron S et al (2009) Community proteogenomics reveals Iwanaga S, Lee BL (2005) Recent advances in the innate immunity of invertebrate
insights into the physiology of phyllosphere bacteria. Proc Natl Acad Sci USA animals. J Biochem Mol Biol 38:128–150
106:16428–16433 Jones KM, Kobayashi H, Davies BW, Taga ME, Walker GC (2007) How rhizobial
Dethlefsen L, Eckburg PB, Bik EM, Relman DA (2006) Assembly of the human symbionts invade plants: the Sinorhizobium medicago model. Nat Rev
intestinal microbiota. Trends Ecol Evol 21:517–523 Microbiol 5:619–633
Devi SM, Ahmed I, Khan AA et al (2006) Genomes of Helicobacter pylori from Kaufman MR, Ikeda Y, Patton C, Van Dykhuizen G, Epel D (1998) Bacterial
native Peruvians suggest a mixture of ancestral and modern lineages and symbionts colonize the accessory nidamental gland of the squid Loligo
reveal a western type cag-pathogenicity island. BMC Genomics 7:191 opalescens via horizontal transmission. Bio Bull 194:36–43
Dewhirst FE, Chen T, Izard J et al (2010) The human oral microbiome. J Bacteriol Kikuchi Y, Hosokawa T, Fukatsu T (2007) Insect–microbe mutualism
192:341–345 without vertical transmission: a stinkbug acquires a beneficial gut symbiont
Dodd DMB (1989) Reproductive isolation as a consequence of adaptive diver- from the environment every generation. Appl Environ Microbiol
gence in Drosophila pseudoobscura. Evolution 43:1308–1311 73:4308–4316
Douglas AE (1998) Nutritional interactions in insect-microbial symbioses: aphids Komai M, Shirakawa H, Kimura S (1988) Newly developed model for vitamin
and their symbiotic bacteria Buchnera. Annu Rev Entomol 43:17–37 K deficiency in germfree mice. Int J Vitam Nutr Res 58(1):55–59
Ducklow HW, Mitchel R (1979) Bacterial populations and adaptations in the Koren O, Rosenberg E (2006) Bacteria associated with mucus and tissues of the
mucus layers on living corals. Limnol Oceanogr 24:715–725 coral Oculina patagonica in summer and winter. Appl Environ Microbiol
Edwards JE, McEwan NR, Travis AJ, Wallace RJ (2004) 16S rDNA library-based 72:5254–5259
analysis of ruminal bacterial diversity. Anton Van Leeuw 86:263–281 Lambais MR, Crowley DE, Cury JC, Büll RC, Rodrigues RR (2006) Bacterial
Ercolani GL (1991) Distribution of epiphytic bacteria on olive leaves and the diversity in tree canopies of the Atlantic Forest. Science 312:1917
influence of leaf age and sampling time. Microb Ecol 21:35–48 Leser TD, Amenuvor JZ, Jensen TK et al (2002) Culture-independent analysis of
Falcon L, Magallon S, Castillo A (2010) Dating the cyanobacterial ancestor of the gut bacteria: the pig gastrointestinal tract microbiota revisited. Appl Environ
chloroplast. ISME J 4:777–783 Microbiol 68:673–690
Fine M, Loya Y (2002) Endolithic algae: an alternative source of photoassimilates Ley RE, Bäckhed F, Turnbaugh P et al (2005) Obesity alters gut microbial ecology.
during coral bleaching. Proc R Soc Lond B 269:1205–1210 Proc Natl Acad Sci USA 102:11070–11075
Ley RE, Peterson DA, Gordon JI (2006a) Ecological and evolutionary forces Reichenbach H (1984) Myxobacteria: a most peculiar group of social prokaryotes.
shaping microbial diversity in the human intestine. Cell 124:837–848 In: Rosenberg E (ed) Myxobacteria: development and cell interactions.
Ley RE, Turnbaugh PJ, Klein S, Gordon JI (2006b) Human gut microbes associ- Springer, New York, pp 1–50
ated with obesity. Nature 444:1022–1023 Reid G, Younes JA, Van der Mei HC et al (2011) Microbiota restoration: natural
Ley RE, Lozupone CA, Hamady M, Knight R, Gordon JI (2008) Worlds within and supplemented recovery of human microbial communities. Nat Rev
worlds: evolution of the vertebrate gut microbiota. Nature Rev Microbiol Microbiol 9:27–38
6:776–788 Reitner J, Schumann-Kindel G (1997) Pyrite in mineralized sponge tissue-
Lindow SE, Brandl MT (2003) Microbiology of the phyllosphere. Appl Environ product of sulfate reducing sponge related bacteria. Facies 36:272–276
Microbiol 69:1875–1883 Ritchie KB (2006) Regulation of microbial populations by coral surface mucus
Lugtenberg B, Kamilova F (2009) Plant-growth-promoting Rhizobacteria. Annu and mucus-associated bacteria. Mar Ecol Prog Ser 322:1–14
Rev Microbiol 63:541–556 Rohwer F, Seguritan V, Azam F, Knowlton N (2002) Diversity and distribution of
Luoto R, Laitinen K, Nermes M, Isolauri E (2010a) Impact of maternal probiotic- coral-associated bacteria. Mar Ecol Prog Ser 243:1–10
supplemented dietary counseling on pregnancy outcome and prenatal and Rosenberg E, Falkovitz L (2004) The Vibrio shiloi/Oculina patagonica model
postnatal growth: a double-blind, placebo-controlled study. Br J Nutr system of coral bleaching. Ann Rev Microbiol 58:143–159
103:1792–1799 Rosenberg E, Koren O, Reshef L, Efrony R, Zilber-Rosenberg I (2007) The role of
Luoto R, Kalliomaki M, Laitinen K, Isolalauri E (2010b) The impact of perinatal microorganisms in coral health, disease and evolution. Nat Rev Microbiol
intervention on the development of overweight and obesity: follow-up study 5:355–362
from birth to 10 years. Int J Obes 34:1531–1537 Ruby EG (1996) Lessons from a cooperative bacterial-animal association: the
Mai V, Ukhanova M, Baer DJ (2010) Understanding the extent and sources of Vibrio fischeri-Euprymna scolopes light organ symbiosis. Annu Rev Microbiol
variation in gut microbiota studies; a prerequisite for establishing associa- 50:591–624
tions with disease. Diversity 2:1085–1096 Russell JB, Rychlik JL (2001) Factors that alter rumen ecology. Science
Margulis L (1993) Symbiosis in cell evolution: microbial communities in the 292:1119–1122
Archean and Proterozoic eons, 2nd edn. Freeman, New York Russell JA, Latorre A, Sabater-Muñoz B, Moya A, Moran NA (2003) Side-stepping
Margulis L, Sagan D (2001) Marvellous microbes. Resurgence 206:10–12 secondary symbionts: widespread horizontal transfer across and beyond the
Mart´ınez-Garc´ıa M, D´ıaz-Valdéz M,Wanner G et al (2007) Microbial commu- Aphidoidea. Mol Ecol 12:1061–1075
nity associated with the colonial ascidian Cyctodytes dellechiajei. Environ Sandström JP, Russel JA, White JP, Moran NA (2001) Independent origins
Microbiol 9:521–534 and horizontal transfer of bacterial symbionts of aphids. Mol Ecol
Mateos M, Castrezana SJ, Nankivell BJ, Estes AM, Markow TA, Moran NA 10:217–228
(2006) Heritable endosymbionts of Drosophila. Genetics 174:363–376 Savage DC, Siegel JD, Snellen JE, Whitt DD (1981) Transit time of epithelial cells
McFall-Ngai MJ (1999) Consequences of evolving with bacterial symbionts: in the small intestines of germfree mice and ex-germfree mice associated with
insights from the squid–Vibrio association. Ann Rev Ecol Syst 30:235–256 indigenous microorganisms. Appl Environ Microbiol 42:996–1001
Micallef SA, Shiaris MP, Colon-Carmona A (2009) Influence of Arabidopsis Schluter D (2009) Evidence for ecological speciation and its alternative. Science
thaliana accessions on rhizobacterial communities and natural variation. 323:737–741
J Exp Bot 60:1729–1742 Schmitt-Wagner D, Friedrich MW, Wagner B, Brune A (2003) Phylogenetic
Minkley N, Fujita A, Brune A, Kirchner WH (2006) Nest specificity of the diversity, abundance, and axial distribution of bacteria in the intestinal
bacterial community in termite guts (Hodotermes mossambicus). Insect Soc tract of two soil-feeding termites. Appl Environ Microbiol 69:6007–6017
53:339–344 Schopf JW (1993) Microfossils of the early Archean Apex chert: new evidence of
Montgomery MK, McFall-Ngai M (1994) Bacterial symbionts induce host organ the antiquity of life. Science 260:640–646
morphogenesis during early postembryonic development of the squid Sharon G, Segal D, Ringo JM, Zilber-Rosenberg I, Rosenberg E (2010) Commen-
Euprymna scolopes. Development 120:1719–1729 sal bacteria play a role in mating preference of Drosophila melanogaster. Proc
Mueller S, Saunier K, Hanisch C et al (2006) Differences in fecal microbiota in Natl Acad Sci USA 107:20051–20056
different European study populations in relation to age, gender and country: Sharp KH, Eam B, Faulkner DJ, Haygood MG (2007) Vertical transmission of
a cross-sectional study. Appl Environ Microbiol 72:1027–1033 diverse microbes in the tropical sponge Corticium sp. Appl Environ
Nissimov J, Rosenberg E, Munn C (2009) Antimicrobial properties of resident Microbiol 73:622–629
coral mucus bacteria of Oculina patagonica. FEMS Microb Lett 292:210–215 Shashar N, Cohen Y, Loya Y, Sar N (1994) Nitrogen fixation (acetylene reduction)
Nyholm SV, McFall-Ngai M (2004) The winnowing: establishing the squid vibrio in stony corals: evidence for coral–bacteria interactions. Mar Ecol Prog Ser
symbiosis. Nat Rev Microbiol 2:632–642 111:259–264
Ochman HM, Worobey CH, Kuo JB et al (2010) Evolutionary relationships of She X, Jiang Z, Clark RA, Liu G et al (2004) Shotgun sequence assembly and
wild hominids recapitulated by gut microbial communities. PLoS Biol 8: recent segmental duplications within the human genome. Nature
e1000546. online 431:927–930
O’Hara AM, Shanahan F (2006) The gut flora as a forgotten organ. EMBO Rep Shnit-Orland M, Kushmaro A (2009) Coral mucus-associated bacteria: a possible
7:688–693 first line of defense. FEMS Microbiol Ecol 67:371–380
Olsen MA, Aagnes TH, Mathiesen SD (1994) Digestion of herring by indigenous Silva AM, Barbosa FH, Duarte R et al (2004) Effect of Bifidobacterium longum
bacteria in the minke whale forestomach. Appl Environ Microbiol 60:4445–4455 ingestion on experimental salmonellosis in mice. J Appl Microbiol 97:29–37
Ott T, Sullivan J, James EK et al (2009) Absence of symbiotic leghemoglobins Singh BK, Millard P, Whiteley AS, Murrell JC (2004) Unravelling rhizosphere–
alters bacteroid and plant cell differentiation during development of Lotus microbial interactions: opportunities and limitations. Trends Microbiol
japonicus root nodules. Mol Plant Microbe Interact 22:800–808 12:386–393
Pennisi E (2004) Evolutionary biology. The birth of the nucleus. Science Smith KP, Handelsman J, Goodman RM (1999) Genetic basis in plants for
305:766–768 interactions with disease-suppressive bacteria. Proc Natl Acad Sci USA
Portier P (1918) Les symbiotes. Masson, Paris 96:4786–4790
Qin J, Li R, Raes J et al (2010) A human gut microbial gene catalogue established Somers E, Vanderleyden J, Srinivasan M (2004) Rhizosphere bacterial signalling:
by metagenomic sequencing. Nature 464:59–65 a love parade beneath our feet. Crit Rev Microbiol 30:205–240
Redford AJ, Fierer N (2009) Bacterial succession on the leaf surface: a novel Stappenbeck TS, Hooper LV, Gordon JI (2002) Developmental regulation of
system for studying successional dynamics. Microb Ecol 58:189–198 intestinal angiogenesis by Peneth cells. Proc Natl Acad Sci USA 99:15451–
Redford AJ, Bowers RM, Knight R et al (2010) The ecology of the phyllosphere: 15455
geographic and phylogenetic variability in the distribution of bacteria on tree Stecher B, Hardt WD (2008) The role of microbiota in infectious disease. Trends
leaves. Environ Microbiol 12:2885–2893 Microbiol 16:107–114
Stougaard J (2000) Regulators and regulation of legume root nodule develop- Weimer PJ, Waghorn GC, Odt CL et al (1999) Effect of diet on populations of
ment. Plant Physiol 124:531–540 three species of ruminal cellulolytic bacteria in lactating dairy cows. J Dairy
Stow A, Beattie A (2008) Chemical and genetic defenses against disease in insect Sci 82:122–134
societies. Brain Behav Immun 22:1009–1013 Weisskopf L, Abou-Mansour E, Fromin N et al (2006) White lupin has developed
Sundset MA, Praesteng KE, Cann IK, Mathiesen SD, Mackie RI (2007) Novel a complex strategy to limit microbial degradation of secreted citrate required
rumen bacterial diversity in two geographically separated sub-species of for phosphate acquisition. Plant Cell Environ 29:919–927
reindeer. Microb Ecol 54:424–438 Weitz JS, Hartman H, Levin SA (2005) Coevolutionary arms races between
Tannock G (1995) Normal microflora. Chapman & Hall, London bacteria and bacteriophage. Proc Natl Acad Sci USA 102:9535–9540
Taylor MW, Radax R, Steger D, Wagner M (2007) Sponge associated microor- Whipps JM, Hand P, Pink D, Bending GD (2008) Phyllosphere microbiology with
ganisms: evolution, ecology and biotechnological potentials. Microbiol Mol special reference to diversity and plant genotype. J Appl Microbiol
Biol Rev 71:295–347 105:1744–1755
Thingstad TF, Lignell R (1997) Theoretical models for the control of bacterial Wilkinson DM (2001) Mycorrhizal evolution. Trends Ecol Evol 16:64–65
growth rate, abundance, diversity and carbon demand. Aquat Microb Ecol Wilson EO (1992) The diversity of life. W. W. Norton, New York
13:19–27 Wostmann BS (1981) The germ-free animal in nutritional studies. Annu Rev Nutr
Turnbaugh PJ, Ley RE, Mahowald MA et al (2006) An obesity-associated gut 1:257–297
microbiome with increased capacity for energy harvest. Nature 444:1027–1031 Wostmann BS, Larkin C, Moriarty A, Bruckner-Kardoss E (1983) Dietary intake,
van der Heijden MGA, Bardgett RD, van Straalen NM (2008) The unseen energy metabolism and excretory losses of adult male germfree Wistar rats.
majority: soil microbes as drivers of plant diversity and productivity in Lab Anim Sci 33:46–50
terrestrial ecosystems. Ecol Lett 11:296–310 Yachi S, Loreau M (1999) Biodiversity and ecosystem productivity in a fluctuating
Veneti ZL, Reuter M, Montenegro H et al (2005) Interactions between inherited environment: the insurance hypothesis. Proc Natl Acad Sci USA
bacteria and their hosts: the Wolbachia paradigm. In: McFall Ngai MJ, 96:1463–1468
Henderson B, Ruby EG (eds) The influence of cooperative bacteria on Yang CH, Crowley DE, Borneman J, Keen NT (2001) Microbial phyllosphere
animal host biology. Cambridge University Press, New York, pp 119–141 populations are more complex than previously realized. Proc Natl Acad Sci
Visick KL, Foster J, Doino J, McFall-Ngai MJ, Ruby EG (2000) Vibrio fischeri lux USA 98:3889–3894
genes play an important role in colonization and development of the host Yen JH, Barr AR (1971) New hypothesis of the cause of cytoplasmic incompat-
light organ. J Bact 182:4578–4586 ibility in Culex pipens. Nature 232:657–658
Wallace RJ (2004) Antimicrobial properties of plant secondary metabolites. Proc Zilber-Rosenberg I, Rosenberg E (2008) Role of microorganisms in the evolution
Nutr Soc 63:621–629 of animals and plants: the hologenome theory of evolution. FEMS Microbiol
Warnecke F, Luginbu P, Ivanova N et al (2007) Metagenomic and functional analysis Rev 32:723–735
of hindgut microbiota of a wood-feeding higher termite. Nature 450:560–565 Zilber-Rosenberg I, Rosenberg E (2011) Prebiotics and probiotics within the
Watanabe H, Tokuda G (2010) Cellulolytic systems in insects. Annu Rev Entomol framework of the hologenome concept. J Microb Biochem Technol http://
55:609–632 dx.doi.org/10.4172/1948-5948.S1-001
Webster NS, Wilson KJ, Blackall LL, Hill RT (2001) Phylogenetic diversity of Zoetendal EG, Akkermans ADL, Akkermans-van Vliet WM et al (2001) A host
bacteria associated with the marine sponge Rhopaloeides odorabile. Appl genotype affects the bacterial community in the human gastrointestinal
Environ Microbiol 67:434–444 tract. Microb Ecol Health Dis 13:129–134
16 Cyanobacterial-Plant Symbioses
David G. Adams1 . Birgitta Bergman2 . Sandra A. Nierzwicki-Bauer3 . Paula S. Duggan1 .
Amar N. Rai4 . Arthur Schüßler5
1
Faculty of Biological Sciences, University of Leeds, Leeds, UK
2
Department of Botany, Stockholm University, Stockholm, Sweden
3
Department of Biology, Rensselaer Polytechnic Institute, Troy, NY, USA
4
Department of Biochemistry, North-Eastern Hill University, Shillong, Meghalaya, India
5
Genetics Biocenter, University of Munich, Munich, Germany
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360 The Azolla Symbiosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
Cyanobacterial Symbioses with Hornworts Taxonomy and Distribution . . . . . . . . . . . . . . . . . . . . . . . . . 372
and Liverworts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360 Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
Bryophyte Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360 General Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
The Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361 The Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
Bryophyte Structures and Their Infection . . . . . . . . . . . . . . 362 Cyanobacterial Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
Bryophyte-Cyanobacterium Signal Exchange . . . . . . . . . . . 362 Bacterial Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
Host-Cyanobiont Interactions Post Infection . . . . . . . . . . . 364 Host Structures and the Infection Process . . . . . . . . . . . . . . 374
Cell Division Control and Hormogonia The Leaf Cavity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
Formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364 The Infection Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
Morphological Modifications to Bryophyte and Host-Symbiont Signal Exchange . . . . . . . . . . . . . . . . . . . . . . . . 375
Symbiont . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365 Host-Cyanobiont Interactions Post Infection . . . . . . . . . . . 375
Nitrogen Fixation and Transfer of Fixed Nitrogen . . . . . . 365 Morphological Modifications to Host and
Carbon Dioxide Assimilation and Transfer of Cyanobacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
Carbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366 Nitrogen Fixation and Transfer of Fixed
Genetic Analysis of the Nostoc-Anthoceros Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
Association . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366 Carbon Assimilation and Transfer of
The hrm Operon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366 Fixed Carbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
sigH and ctpH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 Ecological Importance: Friend or Foe? . . . . . . . . . . . . . . . . . . 376
tprN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 Cyanobionts Becoming a Nitrogen-Fixing
ntcA, hetR, and hetF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 ‘‘Organelle’’? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
Interactions in the Nostoc-Gunnera Symbiosis . . . . . . . . . . . 367 The Cycad Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377

The Nostoc-Gunnera Symbiosis . . . . . . . . . . . . . . . . . . . . . . . . . 367 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
A Unique Endosymbiosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368 Cycads and Toxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
The Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368 The Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
Specificity and Recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368 Establishment of Symbiosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
The Site of Gunnera Infection: The Gland . . . . . . . . . . . . . . 368 Symbiotic Competence: Fit to Infect? . . . . . . . . . . . . . . . . . . . 378
The Infection Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369 Other Symbiotic Competence Factors . . . . . . . . . . . . . . 379
Hormogonium Differentiation . . . . . . . . . . . . . . . . . . . . . . 369 Carbon Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
Entrance and Penetration . . . . . . . . . . . . . . . . . . . . . . . . . . . 370 Nitrogen Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
Signal Exchange Between the Cyanobacterium and Other Adaptations to Life in Symbiosis . . . . . . . . . . . . . . . . . 380
the Host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370
Host-Cyanobacterial Interactions Post Infection . . . . . . . 370 Cyanolichens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Nitrogen Fixation and the Transfer of Nitrogen . . . . . . . . 371 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Carbon Assimilation and the Transfer of The Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Fixed Carbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371 Mycobionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Ecological Importance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372 Cyanobionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372 The Lichen Thallus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
360 16 Cyanobacterial-Plant Symbioses
Recognition and Signal Exchange Between Partners . . . . 381 agents in most plant symbioses; some plants enhance their
Structural-Functional Changes . . . . . . . . . . . . . . . . . . . . . . . . . . 382 chances of infection by producing chemical signals that stimu-
Nutrient Exchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383 late hormogonia formation and also chemoattractants that
Ecological Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383 direct hormogonia into the plant tissue. Cyanobacteria are not
Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384 restricted to the roots of plants but can infect thalli, stems, and
leaves. The major hosts are bryophytes (see the section
The Geosiphon pyriformis - Nostoc Endocyanosis and Its > ‘‘Cyanobacterial Symbioses with Hornworts and Liverworts’’
Relationship to the Arbuscular Mycorrhiza (AM) . . . . . . . . . 384 in this chapter), the angiosperm Gunnera (see the section
The Geosiphon pyriformis Symbiosis . . . . . . . . . . . . . . . . . . . . 384 > ‘‘Interactions in the Nostoc-Gunnera Symbiosis’’ in this
The Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384 chapter), the aquatic fern Azolla (see the section > ‘‘The Azolla
Geosiphon pyriformis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384 Symbiosis’’ in this chapter), fungi (forming lichens; see the
Nostoc punctiforme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385 section > ‘‘Cyanolichens’’ in this chapter), the fungus Geosiphon
Infection Process, Development, and Structure of the (see the section > ‘‘The Geosiphon pyriformis: Nostoc
Symbiosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386 Endocyanosis and its Relationship to the Arbuscular Mycorrhiza
Infection Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386 (AM)’’ in this chapter), and cycads (see the section on > ‘‘The
Development of the Symbiosis . . . . . . . . . . . . . . . . . . . . . . 387 Cycad Symbioses’’ in this chapter).
Structure and Compartmentation of the
Geosiphon Bladder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
Element Composition and Distribution . . . . . . . . . . . . 389 Cyanobacterial Symbioses with Hornworts
Signal Exchange Between Host and and Liverworts
Cyanobacterium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
Host-Cyanobiont Interactions Post Infection . . . . . . . . . . . 390 Bryophyte Symbioses
Morphological Modifications of the Symbiosis
Partners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390 The division Bryophyta consists of the Hepaticae (liverworts),
N2 and CO2 Fixation and Transfer . . . . . . . . . . . . . . . . . . 391 the Anthocerotae (hornworts), and the Musci (mosses), all of
Ecological Importance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391 which are small, nonvascular terrestrial plants, some of which
Why the Symbiosis Is Mutualistic . . . . . . . . . . . . . . . . . . . 391 form epiphytic or endophytic associations with cyanobacteria,
Evolutionary Implications with Ecological primarily of the genus Nostoc (Adams 2002a, b; Meeks 2003;
Meaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 Solheim et al. 2004; Adams et al. 2006, 2012; Adams and Duggan
A Network Between Fungi, Cyanobacteria, and 2008; Bergman et al. 2007, 2008). Moss-associated cyanobacteria
Plants? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 are mostly epiphytic (Solheim and Zielke 2002; Solheim et al.
2004; Gentili et al. 2005), apart from those found in two Sphag-
num species in which the cyanobacteria are found in water-filled,
Introduction hyaline (dead) cells, where they may be protected from the
acidic bog environment (Solheim and Zielke 2002). Even these
Cyanobacteria are unique in the wide range of symbiotic asso- associations can be considered epiphytic as the hyaline cells are
ciations they form with eukaryotic hosts including plants, fungi, connected via pores to the outside environment. A wide range of
sponges, and protists (for reviews, see Adams 2000; Adams and cyanobacteria, including members of the non-heterocystous,
Duggan 2012; Adams et al. 2012; Rai et al. 2000; Rai et al. 2002; filamentous genera Phormidium and Oscillatoria and even
Bergman et al. 2003, 2008). Cyanobacteria are photoautotrophs, the unicellular Microcystis, have been found as moss
and in many cases facultative heterotrophs and nitrogen fixers, and epiphytes (Solheim et al. 2004), although members of the
can provide nonphotosynthetic hosts with both nitrogen and car- filamentous, heterocyst-producing genera Nostoc, Stigonema,
bon. Even if the benefit to the host is clear, that to the cyanobacteria and Calothrix are the most common (DeLuca et al. 2002, 2007;
is less obvious. They often receive carbon from photosynthetic Gentili et al. 2005; Houle et al. 2006). These epiphytic
hosts, but they are capable of carbon fixation themselves. Perhaps, associations will not be discussed further here, but they are of
in the enclosed environment provided by the host, a more likely ecological importance as they are commonly the major source
advantage is protection from predation and from environmental of combined nitrogen in ecosystems where mosses are
extremes, such as high light intensity and desiccation. abundant, such as northern hemisphere forests (Zielke et al.
The cyanobacterial symbionts of plants all possess at least 2002, 2005; Solheim and Zielke 2002; Nilsson and Wardle
two essential characteristics—the ability to differentiate hetero- 2005; DeLuca et al. 2008; see also Adams et al. 2012).
cysts, which are specialized nitrogen-fixing cells (for reviews, see In their natural habitat, the liverworts and hornworts grow
Adams and Duggan 1999; Zhang et al. 2006), and hormogonia, as a prostrate gametophyte thallus a few centimeters in length,
which are short, gliding filaments that lack heterocysts and attached to the substratum by primitive roots known as rhizoids.
provide a means of dispersal (Adams 2000; Meeks 2003, 2009; Mature symbiotic colonies can be seen as dark spots 0.5–1.0 mm
Gusev et al. 2002; Meeks et al. 2002; Meeks and Elhai 2002; in diameter within the plant tissue (> Fig. 16.1). Of the more
Bergman et al. 2007). The hormogonia serve as the infective than 340 liverwort genera, only four are known to develop
Cyanobacterial-Plant Symbioses 16 361
that lack heterocysts and provide a means of dispersal (Campbell

and Meeks 1989; Johansson and Bergman 1994; Bergman et al.
1996). Heterocysts fix dinitrogen for both partners, and the
motile hormogonia, which are a transient phase of the life
cycle, enable the otherwise immotile cyanobacterial filaments
to gain entry to the plant host (see the section > ‘‘Bryophyte
Structures and Their Infection’’ in this chapter). The symbioti-
cally competent cyanobacteria are hormogonia-forming strains
of mostly the genus Nostoc, although Calothrix and
Chlorogloeopsis strains have been shown to reconstitute the
symbiosis with Blasia and Phaeoceros (West and Adams 1997).
In the field, a single liverwort or hornwort thallus can become
infected by many different Nostoc strains (West and Adams 1997;
West et al. 1999; Costa et al. 2001; Rasmussen and Nilsson 2002;
Adams and Duggan 2008).
Hormogonia differentiation is triggered by environmental
stimuli, including the dilution of liquid cultures, or their transfer
to solid medium or exposure to red light (Herdman and Rippka
1988; Tandeau de Marsac 1994). Their formation can also be
triggered by exudates from plants such as Anthoceros (Campbell
and Meeks 1989), Blasia (Knight and Adams 1996), Gunnera
(Rasmussen et al. 1994), and wheat roots (Gantar et al. 1993;
Knight and Adams 1996). The first 24 h of Nostoc punctiforme
. Fig. 16.1 hormogonia development, induced by hormogonia-inducing
The liverwort Blasia pusilla, collected from the wild, showing the factor (HIF, see section > ‘‘Bryophyte Structures and Their
thick midribs of the thallus surrounded by the dark spots of Nostoc Infection’’ in this chapter) or combined nitrogen starvation
colonies (From Adams (2000), with permission) (Campbell et al. 2007, 2008), is characterized by many changes
in gene expression, with the transcription of 944 genes
upregulated and 856 downregulated (Campbell et al. 2007).
associations with cyanobacteria: two (Marchantia and Porella) The upregulated genes reflect the importance of signal sensing
forming epiphytic associations and two (Blasia and Cavicularia) and chemotaxis because a majority of the encoded proteins are
forming endophytic associations (Meeks 1990). Four of the six involved in signal transduction and transcriptional regulation,
hornwort genera (Anthoceros, Phaeoceros, Notothylas, and and others have putative roles in chemotaxis and pilus biogen-
Dendroceros) form endophytic associations (Meeks 1990). The esis (Meeks et al. 2001; Klint et al. 2006; Campbell et al. 2007).
epiphytic associations are more common than once thought but Abundant type IV pili (Tfp) cover the surface of Nostoc
are poorly understood (Dalton and Chatfield 1985; Brasell hormogonia but are absent from vegetative cells (> Fig. 16.2).
et al. 1986), whereas the endophytic associations have been In a wide range of bacteria, Tfp have roles in adhesion, motility,
well studied because of the ease with which they can be grown pathogenesis, and DNA uptake (Mattick 2002; Nudleman and
in the laboratory. The hornworts Anthoceros and Phaeoceros and Kaiser 2004; Burrows 2005). Both adhesion (to the plant surface)
the liverwort Blasia can all be grown conveniently in shaken and motility (together with chemotaxis, to locate the host plant
liquid culture (> Fig. 16.3b), with or without their symbiotic symbiotic structures) are likely to be essential factors in the
partners, and can be readily reinfected with their original successful infection of plants. Tfp are involved in motility in
partner or with cyanobionts from Gunnera, cycads, lichens, some unicellular cyanobacteria (Bhaya 2004) and may also have
and even some free-living strains (Enderlin and Meeks 1983; a role in the gliding of hormogonia, although this is so far
Meeks 1988, 1990, 2003; Kimura and Nakano 1990; Babic 1996; unproven (Duggan et al. 2007). In Nostoc punctiforme, the
West and Adams 1997; Adams 2002a, b; Duckett et al. 2004; mutation of genes such as pilT and pilD, thought to be involved
Adams and Duggan 2008; > Fig. 16.3c). in Tfp function, greatly reduces the infectivity of the mutant
hormogonia in the liverwort Blasia (Duggan et al. 2007). How-
ever, it is not clear if this is due to loss of motility (and with it,
The Symbionts chemotaxis) or interference with another potential function of
the pili, such as recognition of, or adhesion to, the plant surface.
For a cyanobacterium to establish a successful plant symbiosis, it The ability of hormogonia to infect a particular host can be
must possess at least two essential characteristics—the ability to affected by subtle aspects of their behavior. For example, the
differentiate both heterocysts, which are specialized nitrogen- infection frequency of Nostoc punctiforme hormogonia in
fixing cells (for reviews, see Adams and Duggan 1999; Zhang the liverwort Blasia is influenced by mutations in cyaC, which
et al. 2006), and hormogonia, which are short, gliding filaments encodes adenylate cyclase, the enzyme responsible for the
. Fig. 16.2
Pili on the surface of Nostoc punctiforme hormogonia. Pili are absent from the cell surface of vegetative filaments (a) but are abundant on
the surface of hormogonia (b). Scale bars represent 1 mm. For electron microscopy, platinum was evaporated onto the surface of
each sample which was then viewed using a JEOL1200EX transmission electron microscope at 80 kV (From Duggan et al. (2007) with
permission)
biosynthesis of the intracellular messenger cAMP, adenosine which resemble stomata but are not thought to be related
30 , 50 -cyclic monophosphate (Adams and Duggan 2008; Chapman (Villarreal and Renzaglia 2006), are formed by the separation
et al. 2008). However, mutation in two different domains of this of adjacent epidermal cells, and their formation is followed by
multi-domain enzyme results in hormogonia with very different the development of a slime cavity directly beneath the cleft
infection frequencies in Blasia, one having a three- to fourfold (Renzaglia et al. 2000). Blasia auricles have two slime papillae,
greater infection frequency than the wild type and the other one of which (the inner slime papilla) partly fills the auricle
showing a 75 % reduction in frequency compared with the cavity, whereas the other (the outer slime papilla) arises from the
wild type (Chapman et al. 2008). The explanation of these thallus adjacent to the auricle (> Fig. 16.3d). In the hornwort
different infection phenotypes is not readily apparent, as both Leiosporoceros dussii (> Fig. 16.4a), the slime cavities take the
mutants have cellular cAMP levels 25 % of the wild type, and the form of elongated mucilage-filled ‘‘canals’’ (> Fig. 16.4b) that
mutant hormogonia, induced in the presence of Blasia, show no result from the separation of plant cell walls along their middle
differences in their frequency, motility, or piliation. lamellae and are connected to the outside by mucilage clefts
(> Figs. 16.4c, d) through which Nostoc can gain entry.
Branching of the canals results in an integrated network,
Bryophyte Structures and Their Infection enabling the symbiont to invade the whole thallus (Villarreal
and Renzaglia 2006). The cyanobacteria enter Blasia auricles,
In the bryophyte-cyanobacteria symbioses, the symbionts infect and presumably hornwort slime cavities, as hormogonia (see the
existing plant structures. In the liverwort Blasia, the preceding section > ‘‘The Symbionts’’), whereupon they lose
cyanobacteria occupy roughly spherical structures, known as motility and differentiate heterocysts (Kimura and Nakano
auricles, on the underside of the thallus (> Fig. 16.3c, d). These 1990; Babic 1996).
develop from a three-celled mucilage hair that undergoes
extensive elaboration (Renzaglia et al. 2000). The thallus of the
hornworts Anthoceros and Phaeoceros is much thicker than that Bryophyte-Cyanobacterium Signal Exchange
of Blasia, and the cyanobacteria are found in slime cavities,
within the thallus, that open to the ventral surface via slit-like Anthoceros punctatus releases an unidentified, low-
pores or mucilage clefts (> Fig. 16.3a). The mucilage clefts, molecular-mass, heat-labile product that stimulates
. Fig. 16.3
The hornwort and liverwort symbioses. (a) Fluorescence micrograph of the hornwort Phaeoceros sp. stained with calcofluor. Hormogonia
gain entry to the slime cavities within the thallus via slit-like entrances (one of which is arrowed). (b) View of the liverwort Blasia
pusilla grown free of cyanobacteria in shaken liquid medium in an Erlenmeyer flask (viewed from below). (c) Liquid-grown Blasia
pusilla infected in the laboratory with two different Nostoc strains, one brown pigmented (the two auricles to the left) and the other
blue-green. (d) Fluorescence micrograph of an uninfected Blasia auricle stained with calcofluor. The auricle has one inner (lower arrow)
and one outer (upper arrow) slime papilla. Bars 50 mm (Photographs (a) and (d) courtesy of S. Babic. (a, d) From Adams (2000) with
permission; (b) from Adams (2002a) with permission; (d) from Adams and Duggan (1999) with permission)
hormogonia formation in Nostoc strains (Campbell and Meeks As a symbiotic colony develops, filamentous protrusions
1989). This hormogonia-inducing factor (HIF) seems to be grow from the host plant into the colony, possibly to enhance
produced as a result of nitrogen starvation, as it is not present nutrient exchange between host and symbiont (see the section
when the hornwort is cultured in medium containing excess > ‘‘Morphological Modifications to Bryophyte and Symbiont’’
NH4+. Compounds with similar activity to HIF are found in in this chapter). What signal induces these changes in the host is
Gunnera stem gland mucilage (Rasmussen et al. 1994), wheat not known; however, arabinogalactan proteins (AGPs) are released
root exudates (Gantar et al. 1993), and Blasia exudates (Babic by many cyanobacteria (Bergman et al. 1996; Jackson et al. 2012),
1996; Watts et al. 1999; Watts 2000). To attract hormogonia, and such AGPs are thought to have important roles in plant
a potential host must release a chemoattractant, such as that growth and development (Pennell 1992). Liverworts also pro-
produced by the liverwort Blasia when nitrogen starved (Knight duce AGPs (Basile 1990), the inner and outer slime papillae of
and Adams 1996; Watts 2000; Adams and Duggan 2008). Blasia and the slime cavity of Phaeoceros staining with both Yariv
However, hormogonia chemoattractants can also be produced reagent, which is specific for AGPs, and with anti-AGP mono-
by nonhost plants such as Trifolium repens (Nilsson et al. 2006) clonal antibodies (Watts 2000; Jackson et al. 2012).
and germinating wheat seeds (Knight and Adams 1996; Watts Another group of potential signaling molecules in
2000; Adams and Duggan 2008). Although the chemical identity cyanobacteria-plant symbioses is the flavonoids; these are
of these chemoattractants is not known, they are thought to be secreted by legumes and are involved in the initial signaling in
sugar-based molecules (Watts 2000), and in keeping with this, the symbiosis with Rhizobium, by binding to the transcriptional
simple sugars such as arabinose, glucose, and galactose are activator NodD (Fisher and Long 1992). Seed rinse from
known to attract hormogonia (Nilsson et al. 2006). Gunnera, an angiosperm that forms symbiosis with Nostoc, can
. Fig. 16.4
The hornwort Leiosporoceros dussii with symbiotic Nostoc. (a) The young rosette to the left lacks the upright sporophytes that are
abundant on the surface of the older thallus to the right. (b) The Nostoc colonies can be seen as long ‘‘strands’’ (some of which
are arrowed) within the thallus, parallel to the main axis. S = sporophyte. (c) Light micrograph of a nearly transverse section of the
mucilage clefts (arrows) that serve as the point of entry for cyanobacterial infection; the filaments of Nostoc subsequently
spread through channels that result from the separation of hornwort cells along their middle lamellae. (d) Scanning electron
micrograph of a mucilage cleft. Bars 10 mm in (a), 2 mm in (b), 15 mm in (c) and 20 mm in (d) (From Villarreal and Renzaglia (2006)
with permission)
induce expression of nod genes in Rhizobium (Bergman et al. Host-Cyanobiont Interactions Post Infection
1996; Rasmussen et al. 1996; Rai et al. 2000), and the flavonoid
naringin induces expression of hrmA (see the sections Cell Division Control and Hormogonia
> ‘‘Cell Division Control and Hormogonia Formation’’ and Formation
> ‘‘The Hrm Operon’’ in this chapter) in Nostoc punctiforme
(Cohen and Yamasaki 2000). Expression of the N. punctiforme In symbiosis with Anthoceros, the doubling time of Nostoc
hrmA gene is also induced by a combination of components, can be 240 h, compared with 45 h in the free-living state
including deoxyanthocyanins, found in extracts of the (Meeks 1990). This slowed growth of the cyanobiont ensures
water fern Azolla which forms symbioses with Anabaena that its growth rate matches that of the host plant. The mecha-
(Cohen et al. 2002). nism of this growth control is unknown, but it seems not to be
The lectins are another group of signaling compounds nitrogen limitation, even though the host takes most of
of importance in bacterial symbioses. Although little is the nitrogen fixed by its partner (see the section > ‘‘Nitrogen
known about their potential involvement in cyanobacteria- Fixation and Transfer of Fixed Nitrogen’’ in this chapter).
plant symbioses, they are produced by the plant host in As well as controlling the growth rate of the cyanobiont, the
bryophyte and Azolla symbioses and bind to sugars on host must control hormogonia formation. Prior to infection,
the surface of symbiotic Nostoc strains (Lehr et al. 2000; the host plant stimulates the development of hormogonia
see also: Rai et al. 2000; Adams 2000; Rikkinen 2002; Adams in potential partners by releasing HIF (see the section
et al. 2006, 2012). They have also been suggested to be involved > ‘‘Bryophyte-Cyanobacterium Signal Exchange’’ in this
in fungus-partner recognition in lichens (Lehr et al. 2000; chapter). However, once infection has occurred, the plant must
Elifio et al. 2000; Rikkinen 2002; Legaz et al. 2004; Sacristan prevent hormogonia differentiation because hormogonia lack
et al. 2006). heterocysts and so cannot form a viable, nitrogen-fixing colony.
A hormogonia repressing factor (HRF), found in aqueous . Table 16.1

extracts of Anthoceros tissue (Cohen and Meeks 1997; Meeks Summary of morphological and physiological changes in
1998), inhibits HIF-induced hormogonia formation in wild- cyanobacteria symbiotically associated with hornworts and
type N. punctiforme. The expression of two genes, hrmA and liverworts
hrmU, is induced by HRF but not by HIF. These observations
Characteristic Hornworts Liverworts
imply that the gene products of the hrmUA operon block hor-
mogonium formation, perhaps by the production of an inhib- Plant structure infected Slime cavities Auricles
itor or by the catabolism of an activator (Cohen and Meeks 1997; Cyanobiont Nostoc Nostoc a
see the section > ‘‘Genetic Analysis of the Nostoc-Anthoceros Location of cyanobiont Intercellular Intercellular
Association’’ in this chapter). Heterocyst frequency (%) b
30–50 30–50
Nitrogenase specific activityd 443 n.d.
Glutamine synthetase:
Morphological Modifications to Bryophyte and
Symbiont Amount of proteinc 86 n.d.
Specific activityc 38 n.d.
The cells of hornwort-associated Nostoc are often enlarged and
+
Form of combined N released NH4 NH4 +
d
show irregularities of shape compared with the same strains Light-dependent CO2 fixation (%) 12 n.d.
grown free living (Meeks and Elhai 2002). In free-living RuBisCo:
cyanobacteria, heterocyst frequency is typically 4–10 % of cells, Amount of proteind 100 n.d.
whereas in symbiosis with hornworts and liverworts, frequencies
Specific activityd 12 n.d.
are usually considerably higher (Adams 2000; Adams et al. 2012;
> Table 16.1). Although, in at least Anthoceros, some heterocysts Abbreviations: RuBisCo ribulose bisphosphate carboxylase/oxygenase, n.d., not
determined, though likely to be similar to hornwort data
seem to be senescent or dead (Meeks 1990), the increase in a
The symbionts are Nostoc spp. in almost all cases; there have been rare
heterocyst frequency is still correlated with elevated rates of reports of Calothrix spp. as symbionts
nitrogen fixation. Because heterocysts are unable to fix CO2, b
Heterocyst frequencies are expressed as a percentage of total cells. Typical
this elevated heterocyst frequency results in a loss of CO2-fixing values for free-living cyanobacteria are 4–10 %
c
capacity, which can be compensated by the supply of carbon Values are for the symbiont as a percentage of the same cyanobacterium in
the free-living state
skeletons by the host. In Anthoceros, and presumably all endo- d
Values are expressed as a percentage of those for the free-living
phytic bryophyte associations, nitrogenase gene expression and cyanobacteria
heterocyst development in the symbiotically associated Nostoc From Steinberg and Meeks (1989, 1991), Meeks (1990), Rai (1990), and Berg-
appear to be controlled by plant signals and are independent of man et al. (1992a)
the nitrogen status of the cyanobiont (Campbell and Meeks
1992; Meeks 2003, 2009).
Morphological changes are also observed in the bryophyte the reduced photosynthetic capacity of the cyanobiont and must
following infection. In both Blasia and Anthoceros, branched, rely on reduced carbon derived from the plant.
multicellular filaments grow from the wall of the symbiotic Nitrogen fixed by the cyanobiont is released to the plant as
cavity and invade the colony, increasing the surface area of ammonia (> Table 16.1) in both Anthoceros (Rodgers and
contact between the cyanobacteria and the bryophyte (Rodgers Stewart 1974; Stewart and Rogers 1977; Meeks et al. 1985a;
and Stewart 1974; Rodgers and Stewart 1977; Duckett et al. 1977; Meeks et al. 1985b) and Blasia (Rodgers and Stewart 1974;
Renzaglia 1982; Kimura and Nakano 1990; Gorelova et al. 1996). Stewart and Rogers 1977), and initial uptake of the ammonia
In Blasia, these filaments are derived from the inner slime papilla occurs via the glutamine synthetase-glutamate synthase
and possess transfer cell morphology, implying an involvement (GS-GOGAT) pathway of the host (Meeks et al. 1983, 1985b;
in nutrient exchange. However, such wall ingrowths are not Meeks 1990; Rai 1990). In Anthoceros, the cyanobiont retains as
found in other hornworts, including Leiosporoceros (Villarreal little as 20 % of the nitrogen it fixes (Meeks et al. 1985a) yet
and Renzaglia 2006). shows no signs of nitrogen deprivation. Ammonia is released by
the cyanobiont as a consequence of decreased activity of gluta-
mine synthetase, the first enzyme in the GS-GOGAT pathway,
Nitrogen Fixation and Transfer of Fixed Nitrogen which is the primary route of ammonia assimilation in
cyanobacteria (Muro-Pastor et al. 2005; Flores and Herrero
The elevated rate of nitrogen (N2) fixation in bryophyte- 2005). In Anthoceros, the decreased activity of GS appears to be
associated cyanobacteria broadly correlates with the increased the result of an undetermined posttranslational modification of
heterocyst frequency in symbiosis (> Table 16.1). The N2 fixa- the enzyme because the amount of GS protein differs little in
tion rate of the Anthoceros-Nostoc association is 4- to 35-fold filaments of free-living and symbiotically associated Nostoc
higher than that of free-living Nostoc (Steinberg and Meeks (Joseph and Meeks 1987; Lee et al. 1988; Meeks 1990, 2003,
1991). Such a high rate of N2 fixation cannot be supported by 2009; Meeks and Elhai 2002; > Table 16.1).
Carbon Dioxide Assimilation and Transfer of these techniques to identify a number of genes involved in the
Carbon initial infection of Anthoceros. This has been aided by the avail-
ability of the complete genome sequence of Nostoc punctiforme
The Calvin cycle is the primary route of CO2 fixation in free-living {DOE Joint Genome Institute website} (see [{http://www.jgi.
and symbiotically associated cyanobacteria, with ribulose-1, doe.gov}]).
5-bisphosphate carboxylase/oxygenase (RuBisCo) as the primary
carboxylating enzyme (Tabita 1994). The rate of light-dependent
The hrm Operon
CO2 fixation in the Nostoc symbiont of Anthoceros immediately
after its separation from symbiosis is eightfold lower than that of
In a transposon mutant of Nostoc 29133, characterized by an
the same cyanobacterium in the free-living state (Steinberg and
increased rate of initial infection of Anthoceros (Cohen and
Meeks 1989; Meeks 1990; > Table 16.1). However, the level of
Meeks 1997; > Table 16.2), Meeks et al. (1999) identified
RuBisCo protein is similar in the two cases (Rai et al. 1989;
two open reading frames (ORFs), hrmU and hrmA, flanking
Steinberg and Meeks 1989; Meeks 1990, 2003; Meeks and Elhai
the site of transposition (> Fig. 16.5). hrmA has no significant
2002), implying that activity is regulated by an unidentified
similarity to sequences in major databases, whereas hrmU has
posttranslational modification of the enzyme (Steinberg and
similarity to the sequences of mannonate oxidoreductase
Meeks 1989; Meeks 1990, 2003; Meeks and Elhai 2002). The
genes and 2-keto-3-deoxygluconate dehydrogenase genes.
cyanobiont therefore grows photoheterotrophically, receiving
Expression of hrmUA is induced by an aqueous extract of
fixed carbon from its photosynthetic host, probably in the
A. punctatus but not by the hormogonium-inducing factor,
form of sucrose (Stewart and Rogers 1977; Steinberg and
HIF. The aqueous extract appears to contain a hormogonium-
Meeks 1991). In at least Anthoceros, the presence of glycogen
repressing factor (HRF) because it suppresses HIF-induced
granules in the cells of symbiotically associated Nostoc implies
hormogonia formation in the wild type but not in the mutant.
that the symbiont is not starved of carbon (Meeks 1990).
Whereas HIF is released into the growth medium, HRF
is probably released into the symbiotic cavity, suppressing
Genetic Analysis of the Nostoc-Anthoceros further hormogonium formation and permitting heterocyst
Association differentiation.
At the 50 end of hrmUA, three other ORFs (hrmI, hrmR, and
Meeks and coworkers have developed genetic techniques, hrmK) are followed by two ORFs coding for unknown
including transposon mutagenesis, for the analysis of the proteins, followed by hrmE, which has similarity to an aldehyde
symbiotically competent cyanobacterium Nostoc punctiforme reductase (> Fig. 16.5). HrmI shows similarity to uronate isom-
strain ATCC 29133 (Cohen et al. 1994, 1998) and have used erase, HrmR to the LacI/GalR family of transcriptional
. Table 16.2
Effect of insertion mutations on the symbiotic infectiveness (expressed in column two as the number of symbiotic colonies per unit of
host tissue) and effectiveness (expressed in column three as acetylene reduction activity per g fresh weight of host tissue and in column
four as acetylene reduction activity per symbiotic colony) of Nostoc 29133 strains in association with Anthoceros punctatus
nmol C2H2 reduced per min per:

Strain (gene) Colonies per mg dry wt per mg Chl a g of fresh weight Colony (103) Gene induction factor(s)
ATCC 29133 (WT) 0.21 6.3 12.4 n.d.
UCD 328 (hrmA) 1.6 6.1 8.6 HRF
UCD 398 (sigH) 1.2 8.0 10.1 HIF
UCD 400 (tprN) 0.49 10.4 6.7 HIF and HRF
Abbreviations: Chl a, chlorophyll a; HRF, aqueous extract of A. punctatus containing hormogonium-repressing factor identified as inducing the hrm operon; WT,
wild type; HIF, exudate of A. punctatus containing hormogonium-inducing factor; and n.d., not determined (From Meeks et al. (1999). The standard deviations and
number of replicates have been omitted for simplicity)
500 bp
hrmE unk unk hrmK hrmR hrmI hrmU hrmA
. Fig. 16.5
Map of the open reading frames in the hrm locus of Nostoc punctiforme. The direction of transcription is indicated by the arrows. Unk are
unknown proteins. Sizes are approximate (Adapted from Campbell et al. (2003))
repressors, and HrmK to gluconate kinases. HrmR is a DNA- ntcA, hetR, and hetF
binding protein that binds sugar ligands and represses transcrip-
tion of hrmR and hrmE (Campbell et al. 2003). Galacturonate Nostoc punctiforme (Nostoc 29133) strains unable to develop
abolishes in vitro binding of HrmR to DNA, implying that the in heterocysts because of mutations in either hetR or hetF can
vivo inducer may be a sugar molecule similar to or containing still infect Anthoceros at a frequency similar to that of the wild
galacturonate. These observations led Meeks and coworkers to type, despite being incapable of forming a functional nitrogen-
propose the following model for the way in which the HRF fixing symbiosis (Wong and Meeks 2002). hetR is thought to
external signal is transduced into Nostoc. HRF enters the Nostoc be the primary activator of heterocyst development
cell and it, or a derivative similar to galacturonate, binds to (Wolk 2000; Golden and Yoon 2003; Zhang et al. 2006), and
HrmR, rendering it incapable of binding to the hrmR the HetF protein seems to be a positive activator of heterocyst
and hrmE operator regions; this derepresses transcription of differentiation, enhancing transcription of hetR and ensuring
these genes, leading to inhibition of hormogonia formation that HetR is localized to developing heterocysts (Wong and
(Campbell et al. 2003). Meeks 2001).
In cyanobacteria, NtcA functions as a nitrogen-
dependent global regulator (Herrero et al. 2004) and controls
sigH and ctpH the transcription of a number of genes, including hetR
(Fiedler et al. 2001; Herrero et al. 2001). The Nostoc
Mutation of the Nostoc 29133 sigH gene, which encodes an punctiforme ntcA mutant, UCD 444, forms motile hormogonia
alternative RNA polymerase sigma subunit, produces no obvi- with wild-type morphology but at only 5–15 % of the wild-
ous phenotype in filaments grown in medium with or without type frequency (Wong and Meeks 2002). However, rather
combined nitrogen but results in an increased infection pheno- than infecting Anthoceros at a reduced frequency, as might
type when they are cocultured with A. punctatus (Campbell et al. be expected, the ntcA mutant fails to infect at all. This
1998; Meeks et al. 1999; Meeks and Elhai 2002; Meeks 2003; noninfective phenotype can be complemented with copies
> Table 16.2). Transcription of sigH is induced by Anthoceros of ntcA.
HIF, but not by HRF, and hrmA transcription is not altered in
a sigH mutant. Thus, although the hrmA and sigH mutants both
have an increased infection phenotype, it seems likely that Interactions in the Nostoc-Gunnera Symbiosis
increased infection has a different basis in the two strains
(Meeks et al. 1999). The Nostoc-Gunnera Symbiosis
The gene ctpH lies immediately 50 of sigH and encodes a
protein with significant similarity to carboxy-terminal proteases Although cyanobacterial-plant symbioses are the most wide-
of the cyanobacterium Synechocystis PCC 6803 (Meeks et al. spread of the nitrogen-fixing symbioses, with hosts through-
1999). In Synechocystis 6803, this gene is required for processing out the plant kingdom, those symbioses with angiosperms
the carboxy-terminal portion of the photosystem II D1 protein (flowering plants) are presently restricted to one
in the thylakoid lumen (Anbudurai et al. 1994). However, in monogeneric family, the Gunneraceae. This contrasts with
Nostoc 29133, ctpH seems to have a different physiological role the more recently evolved rhizobia- or Frankia-angiosperm
because it is not transcribed under vegetative growth conditions, symbioses, which involve a considerably wider angiosperm host
but transcription is induced by Anthoceros HIF. The significance range. The scarcity is also unexpected as angiosperms form the
of this is not understood. ecologically most successful plant division on earth, an area
discussed in recent reviews by Osborne and Bergman (2009)
and by Usher et al. (2007). In addition, cyanobacteria are
tprN globally widespread with a morphological variation surpass-
ing most other prokaryotes. In spite of this, the
Lying 30 of the gene devR, expression of which is essential for cyanobacterial range is narrow, with only one cyanobacterial
heterocyst maturation is the gene tprN, which encodes a protein genus, Nostoc, functioning as microsymbiont in Gunnera. How-
with similarity to tetratricopeptide repeat proteins (Campbell ever, as the Gunneraceae is one of the oldest angiosperm families
et al. 1996). These proteins have been studied primarily in and with Gunnera and cyanobacterial fossils dating to some
eukaryotes in which they are required for a variety of functions 90 million years ago (Ma) and three billion years ago (Ba),
from cell cycle control to transcription repression and protein respectively, this symbiosis is likely to have persisted for a long
transport (Lamb et al. 1995). Inactivation of tprN in Nostoc time. Prior to the establishment of the Nostoc-Gunnera symbio-
29133 has no apparent phenotypic effect in the free-living sis, however, the same or a similar cyanobacterial genus may also
growth state, but the mutant infects Anthoceros at about twice have given rise to chloroplasts by entering some ancestral
the level of the wild type (> Table 16.2). Transcription of tprN eukaryotic cell/organism. Indeed, the chloroplast genome of
occurs during vegetative growth but increases in the presence of Arabidopsis is more similar to that of Nostoc than to the unicel-
both HIF and HRF (Meeks et al. 1999). The significance of this lular cyanobacteria tested (Martin et al. 2002; Deusch et al.
in the infection process is not known. 2008). This ancient endosymbiotic event (or series of events)
was the origin of all plants and algae and therefore totally found in cyanobionts of Gunnera spp. sampled from natural
revolutionized our biosphere and atmosphere (via oxygenic stands in Chile (Guevara et al. 2002) using the same fingerprint-
photosynthesis). ing technique. 16S rRNA analyses also demonstrate that all
Gunnera isolates examined belong to the genus Nostoc
(Rasmussen and Svenning 2001). Svenning et al. (2005) dem-
A Unique Endosymbiosis onstrated that some cyanobacteria isolated from various
Gunnera spp. may form a distinct clade (based on the complete
Although the Nostoc-Gunnera symbiosis was first described by 16S rDNA gene sequence) suggesting host specificity, although
Reinke in 1873 (Reinke 1873), understanding of the infection a few Gunnera isolates did not conform to this clade. Later it was
mechanism in this unique angiosperm symbiosis is incomplete. suggested that most cyanobionts are affiliated to two clusters in
In contrast to the other cyanobacterial-plant symbioses, the which they are intermixed with free-living cyanobacteria
Gunnera symbiosis is exclusively intracellular. Still, being (Papaefthimiou et al. 2008).
a facultative symbiosis, the cyanobiont is easily separated from
the plant and may be grown independently, and the symbiosis
can be reconstituted under laboratory conditions. This makes Specificity and Recognition
the Nostoc-Gunnera symbiosis an excellent model for identifying
mechanisms involved in plant endosymbioses and indirectly in Although all Nostoc strains form hormogonia (the plant coloni-
plastid evolution. Also, since it is the only plant symbiosis in zation units) per definition, still only certain strains of Nostoc are
which the cyanobacterium penetrates into the plant cells, the accepted as symbionts, which suggests the existence of other
symbiotic development in Gunnera may have evolved further selective recognition mechanisms (see Rasmussen and Nilsson
than that in all other plant symbioses in which the cyanobacte- 2002). The intracellular position of the cyanobiont in Gunnera
rium remains extracellular. may also impose more severe restrictions on symbiotic partner
recognition than in other intercellular, and possibly less
intimate, plant symbioses. On the other hand, cyanobacterial
The Symbionts isolates from cycads and bryophytes readily invade Gunnera cells
and vice versa.
The genus Gunnera was named by C. von Linné in honor of the
Norwegian bishop Gunnérus, a person Linné admired. The
approximately 30–50 Gunnera species are mostly subtropical to The Site of Gunnera Infection: The Gland
tropical perennial herbs, the exception being the smallest,
G. herteri, which is annual (Wanntorp et al. 2001; Osborne and Infection occurs via a peculiar bright red gland (> Fig. 16.6),
Sprent 2002). The Gunnera plants are composed of large com- already clearly visible at the developing cotyledon (see Bergman
pound spikes and are rhizomatous, or more seldom stolonifer- 2002; Bergman and Osborne 2002; Bergman et al. 2003; Chiu
ous, and have rhubarb-like leaves. Plant sizes vary considerably; et al. 2005; Khamar et al. 2010). The development of the glands is
some are gigantic and may be the largest herbs on earth, such a response to the nitrogen status of the Gunnera plant, and they
as species in South America, Hawaii, and Asia, whereas others only fully develop under nitrogen-deplete conditions. The plant
are small and creeping, such as the stoloniferous species in
New Zealand. In nature, Gunnera spp. seem to be invariably
infected by cyanobacteria (Wanntorp et al. 2001; Osborne and
Sprent 2002).
Ever since the discovery of this peculiar symbiosis (Reinke
1873), cyanobacteria of the genus Nostoc, which are filamentous
and differentiate heterocysts, have been identified as the sole
cyanobionts (see Meeks et al. 2001; Meeks and Elhai 2002;
Bergman et al. 2003). The phenotypic range of the cyanobiont
of Gunnera is wide in terms of morphology, pigmentation, and
colony shape and size, which is obvious when isolates are culti-
vated (Bergman et al. 1992b; Rasmussen and Nilsson 2002;
Svenning et al. 2005; Papaefthimiou et al. 2008). A genotypic
variation has also been verified using genetic fingerprinting of 45
cultured isolates originating from 11 Gunnera species (Nilsson
et al. 2000; Rasmussen and Svenning 2001) and natural
cyanobacteria freshly collected from different Gunnera growing
in Chile (Guevara et al. 2002). One specific Gunnera plant may . Fig. 16.6
also occasionally be infected with more than one Nostoc strain Gunnera seedling with red stem glands out of which viscous
(Nilsson et al. 2000), while no variation within one plant was mucilage is released
The involvement of other microorganisms in the establish-

Mucilage Channel Infected Gunnera cell
ment of the Gunnera symbiosis, as proposed by Towata (1985), is
not likely. This can be demonstrated by, for instance, reconsti-
tution experiments under sterile laboratory conditions (Silvester
and McNamara 1976; Johansson and Bergman 1992). In addi-
tion, some cells of the gland have heavy tannin depositions,
which have been suggested to prevent the invasion of non-
compatible or unwanted microorganisms (fungi and bacteria),
which often reside together with cyanobacteria in the channel
Free-living Motile hormogonia Symbiotic filament mucilage (Towata 1985).
filament with with multiple
few heterocysts heterocysts
. Fig. 16.7 The Infection Process

Schematic illustration of the Nostoc infection process in Gunnera.
Vegetative cells of Nostoc with heterocysts (5–10 %) are attracted The focus has so far primarily been on morphological and adaptive
to the mucilage pouring out of the Gunnera stem and stolon changes in the cyanobiont. The plasticity of Nostoc, in this respect,
glands. The motile hormogonial stage is induced by the mucilage, is utilized by the plant throughout the colonization process and
and the cyanobacterium proceeds toward the interior of the is likely a key factor contributing to its success as a Gunnera
gland. At the bottom of the channel, the cyanobacterial filaments symbiont. A typical feature of the Nostoc-Gunnera symbioses is
penetrate the plant cell walls, and intact Nostoc filaments enter the tight regulation by the plant of cyanobacterial behavior such
the Gunnera cells. After internalization, a cyanobacterial as cell division (considerably slowed down in planta), cell dif-
phenotype with larger cells and supernumerous heterocysts (up ferentiation (the development of supernumerous heterocysts),
to 80 %) develops. The arrow indicates the direction of infection and physiological performance (high nitrogen fixation rates).
hormone auxin is positively involved, potentially communicat- Hormogonium Differentiation

ing the C:N status of the plant (Chiu et al. 2005; Khamar et al.
2010). Gland development is further accelerated under A terrestrial cyanobacterium like Nostoc would (under normal
N-limited conditions if modest levels of sugars (0.5–1.0 % free-living conditions) primarily occur as nonmotile, vegetative
sucrose) are added (Chiu et al. 2005). This strengthens the filaments with heterocysts at regular intervals (about 5–10 % of
significant role of the glands in plant nitrogen acquisition. the total cell number; > Fig. 16.7). On contact with Gunnera,
Furthermore, high levels of carbohydrates (glucose and fruc- the gland and the plant apex are, however, soon covered by
tose), known to support symbiotic nitrogen fixation (Wouters a cyanobacterial ‘‘biofilm’’ composed of tightly packed hormo-
et al. 2000), accumulate in the mature glands prior to the gonia (Osborne et al. 1991; Johansson and Bergman 1992;
colonization by the cyanobacterium, while in Nostoc-colonized Johansson and Bergman 1994; Chiu et al. 2005). Differentiation
glands (in which nitrogen is replenished via nitrogen fixation) of these small-celled motile hormogonia is essential for the
soluble sugar quantities are highly reduced (Khamar et al. 2010). whole Gunnera colonization and cell penetration process; they
The glands secrete a carbohydrate-rich mucilage act as a means for the cyanobacterium both to reach and to
(> Fig. 16.7) when non-infected, and new glands continuously invade the Gunnera organ (the gland; > Fig. 16.7). The mucilage
develop at the base of each new leaf petiole, i.e., near the growing has a pivotal role during this process (Rasmussen et al. 1994). It
stem apices, which also become covered by the mucilage. is composed of highly glycosylated arabinogalactan proteins
Cyanobacteria-colonized glands are closed and do not release (AGPs; Rasmussen et al. 1996) and stimulates not only growth
mucilage. Although root primordia were earlier suggested to be but also hormogonium differentiation. A low-molecular-weight
the point of entry (Schaede 1951), the present consensus is that (<12 kDa), heat-labile protein, not yet characterized, which acts as
glands are the sole cyanobacterial entry point (Silvester and hormogonium-inducing factor (HIF), has been identified in the
McNamara 1976; Bonnett and Silvester 1981; Towata 1985; mucilage (Rasmussen et al. 1994). In contrast, the soluble sugars
Johansson and Bergman 1992; Khamar et al. 2010). It has been of Nostoc-colonized glands inhibit hormogonium differentia-
proposed that these modified glands should be termed ‘‘nod- tion (Khamar et al. 2010). This is needed to stimulate heterocyst
ules’’ (Silvester and McNamara 1976) and indeed a distinct and differentiation and nitrogen fixation, the ‘‘essence’’ of the sym-
well-functioning symbiotic ‘‘organ,’’ restricted in time and space, biosis. Molecular mechanisms behind the induction of hormo-
develops below the gland surface on colonization. Each gland, and gonia and their differentiation are still largely unexplored.
possibly also each of the channels that penetrate into the gland Preliminary studies, using subtractive hybridization and
(see the section > ‘‘The Infection Process’’ in this chapter), proteomics (two-dimensional [2-D] gel electrophoresis coupled
functions as an independent colonization conduit, which would to mass spectrometry) of soluble Nostoc proteins treated with
explain why several cyanobionts may be found inside one indi- Gunnera mucilage show that the induction of hormogonium
vidual gland (Johansson and Bergman 1992; Nilsson et al. 2000). differentiation is also reflected in a differential expression of
genes and proteins, whose expression is either up- or all stages of the infection process and that this includes several
downshifted or both. For instance, three mucilage-induced hie fundamental cyanobacterial processes such as growth, cell divi-
(host-induced expression) genes have been identified, including sion, cell differentiation, ammonia assimilation, and phototactic
a putative precursor of a pheromone-like signaling peptide behavior. The question is whether this is triggered by plant
(HieA), an outer membrane or secreted glycoprotein (HieB), compounds or by the environment within the plant. For
and a protein probably involved in adaptation to acidity (HieC; instance, the symbiotic tissue is low in oxygen and light, which
Liaimer et al. 2001). The latter may be important as the Gunnera may have consequences for gene expression.
mucilage has a pH of 4–5 (Rasmussen et al. 1994), a pH at the Another open question is to what extent the release of the auxin
lower limit of the cyanobacterial tolerance range. Another set of IAA (indole-3-acetic acid) by Nostoc (Sergeeva et al. 2002) acts as
proteins was also identified as being differentially expressed in a signal or influences the development of the symbiotic Gunnera
hormogonia (Klint et al. 2006). These proteins, which were tissues. The influence of auxin has recently been stressed (Chiu
predominantly surface associated, may have roles in motility, et al. 2005). Indeed, cyanobacteria seem to have the potential to
recognition, adhesion, as well as in communication with produce major phytohormones (Liaimer and Bergman 2003)
host plants. The mucilage therefore appears to have important and also to release ‘‘AGP-like’’ proteoglucans, which may also
functions at earlier stages of the Gunnera infection process. influence plant development (Bergman et al. 1996).
Entrance and Penetration Host-Cyanobacterial Interactions Post Infection
The Gunnera glands are composed of a set of up to nine papillae Internalization of the cyanobiont elicits novel, dramatic modi-
surrounding a central papilla (Johansson and Bergman 1992; fications of cyanobacterial morphology and function
Uheda and Silvester 2001; Chiu et al. 2005). Between the papil- (> Fig. 16.7). Because hormogonia lack heterocysts, they are
lae, and leading into the stem tissue, are deep invaginations unable to fix nitrogen. Thus, the hormogonium stage is lethal
through which the mucilage is released. The hormogonia use under free-living conditions (unless combined nitrogen is avail-
these narrow channels to enter the dark interior of the Gunnera able), and hence is, of necessity, transient. Redifferentiation into
stems (> Fig. 16.7). As this is against the normal positive pho- vegetative filaments with heterocysts occurs after 1–2 days. The
totactic behavior of Nostoc, a potent attractant must be released maintenance of a continuous vegetative stage with heterocysts is
by the plant, possibly carried by the mucilage. Motility is crucial a prerequisite for the symbiosis to persist as an efficient provider
at this stage, as the direction of infection is opposite to that of the of combined nitrogen. Repression of hormogonium differenti-
flow of mucilage. Upon reaching the bottom of the gland chan- ation in Gunnera may be achieved by homologues to the hor-
nels, the cyanobacterium penetrates the thin walls of smaller mogonium-repressing factor(s) (HRFs) identified in the
meristematic and dividing cells lining the channel (Silvester and bryophyte symbiosis (see Meeks and Elhai 2002). One compo-
McNamara 1976; Johansson and Bergman 1992; Johansson and nent of this repression machinery may be the inhibitory effects
Bergman 1994; Uheda and Silvester 2001). A delimited tissue on hormogonium development by the soluble sugars present in
of Nostoc-infected Gunnera cells is formed within a few days of Nostoc-colonized glands (Khamar et al. 2010).
inoculation. The mechanism(s) involved in the actual host cell When inside the Gunnera cells, the cyanobacterial cells
penetration is still unknown, although Towata (1985) suggested enlarge, and cell division is considerably restricted (Söderbäck
the occurrence of pectolytic or cellulolytic activities in the muci- and Bergman 1992). In addition, the filaments remain
lage of G. kaalensis. Also lining the channel are the thick-walled surrounded by the host cell plasmalemma through the pinocy-
secretory cells releasing the mucilage (Towata 1985). tosis process. This membrane, like the peribacteroid membrane
in Rhizobium-legume symbioses, acts as the interface between
the symbionts through which the exchange of metabolites takes
Signal Exchange Between the Cyanobacterium place. The Gunnera cells eventually become filled with
and the Host cyanobacterial filaments, which soon start to differentiate an
abnormally high frequency of heterocysts (> Fig. 16.7). Once
Besides HIFs, the plant signals involved in hormogonium dif- infection is complete, the host must tightly control cyanobiont
ferentiation still await genetic identification and chemical char- growth to avoid being outgrown, and this may explain the
acterization, as do the cellular response signaling cascades in enlargement of cyanobacterial cells typical for the endosymbi-
Nostoc. In this context, a highly interesting question is whether otic stage (see Rai et al. 2000; Bergman 2002).
the differentiation of hormogonia resulting from a biotic stim- The dramatic morphological transitions seen in Nostoc on
ulus (such as Gunnera mucilage) triggers specific genes (such as entering Gunnera cells are also reflected in the transcription of
those involved in ‘‘symbiotic competence’’) but not those trig- genes (and the corresponding proteins) related to heterocyst
gered by any abiotic stimulus (such as red light). Also interesting differentiation and nitrogen fixation (see Table 1 in Bergman
are mechanisms involved in the initial rapid cell division and the 2002). For instance, the expression of the hetR gene (the master
machinery behind motility. All studies do, however, verify that gene for heterocyst differentiation) correlates positively with the
the plant influences cyanobacterial morphology and behavior at increase in heterocyst frequency, as does the expression of the
Myrothamnus Africa, Madagascar
G. herteri Uruguay, Brazil
G. perpensa Africa, Madagascar

G. dentata New Zealand
8 G. cordifolia Tasmania
99 6
G. hamiltonii New Zealand
99
G. monoica New Zealand
1 G. macrophylla Malaya
52
G. magellanica Andes
G. petaloidea Hawaii
2
86 G. chilensis Andes
1
70 G. pilosa Andes
. Fig. 16.8
Phylogram of 11 Gunnera species representing varying sizes and geographical origin. Myrothamnus is an African shrub-like plant
growing in dry areas, as opposed to Gunnera species that prefer wet environments characterized by high humidity and high rainfall
(From Wanntorp et al. (2001))
nitrogen-responsive transcription factor encoded by ntcA, whereas (Silvester and Smith 1969; Silvester 1976; Bonnett and Silvester
nifH expression is (as expected) already high, close to the grow- 1981; Osborne et al. 1992; Khamar et al. 2010). The heterocysts
ing apex. By contrast, the expression of the glnB gene, encoding act as the nitrogen-producing entities, holding all the nitroge-
the signal transduction protein PII, decreases along the same nase (Söderbäck et al. 1990; Söderbäck 1992), and are capable of
symbiotic profile (Wang et al. 2004). The overexpression of supporting the entire symbiosis with combined nitrogen. In addi-
both hetR and ntcA and the contrasting downregulation of tion, the cyanobacterium attains enhanced nitrogen fixation capac-
glnB are features indicating important regulatory differences ities compared to its free-living relatives (Silvester 1976; Bonnett
between the symbiotic and free-living life stages. Later, Ekman and Silvester 1981). This may be related to the high heterocyst
et al. (2006) identified a differential protein expression pattern in frequency or to an enhanced nitrogen starvation signal caused
a cyanobacterium isolated from Gunnera manicata when using by the continuous N-drainage from the cyanobiont.
proteomic analysis. Changes were primarily related to cell enve- Up to 90 % of the nitrogen fixed is exported from the
lope and membrane-associated proteins and to changes in cel- cyanobacterium to the host (Silvester et al. 1996). This is likely
lular activities of C and N metabolism, including upregulation due to downregulation of glutamine synthetase protein levels,
of nitrogenase and proteins of the oxidative pentose phosphate specifically in heterocysts, as well as other activities in symbiosis
pathway and a downregulation of Calvin-Benson cycle enzymes. (Söderbäck 1992). As in most nitrogen-fixing plant symbioses,
The significance of these findings in relation to cyanobacterial the nitrogen fixed is released primarily as NH4+ (Silvester et al.
cell differentiation and the establishment and maintenance of an 1996). The Nostoc-infected Gunnera tissues are always well
efficient nitrogen-fixing cyanobacterial-plant symbiosis now invested with vascular strands that facilitate exchange of metab-
needs to be further explored. olites such as nitrogen and carbohydrates (see > Fig. 16.8 in
Cross-sectioning of rhizomes of mature plants reveals the Bergman et al. 1992b). Multiple vascular strands (polystele)
final outcome of the symbiosis: distinct and bright blue-green persist in Gunnera, which may be reminiscent of an aquatic
pigmented but restricted and delimited cyanobacterial colonies ancestry (Osborne et al. 1991). Stock and Silvester (1994)
seen scattered in the rhizome or along the stolons of the smaller showed, using pulse-chase labeling with 15N, that the nitrogen
Gunnera plants (Osborne et al. 1991). However, the sites of fixed was efficiently transported from mature to young parts
infection comprise only a small proportion of the total plant (with lower heterocyst frequencies) in G. monoica stolons and
biomass, particularly in the large Gunnera species. that N-translocation occurs via the phloem.
Nitrogen Fixation and the Transfer of Nitrogen Carbon Assimilation and the Transfer of
Fixed Carbon
As with most other plant symbioses, the main function of
the cyanobacterium in Gunnera is to cover the total combined As Nostoc inside the Gunnera cells is excluded from light, the
nitrogen requirement of the host via nitrogen fixation host must supply the cyanobiont with fixed carbon via its
photosynthesis. Hence, the cyanobiont must adapt to Conclusions

a heterotrophic, or at least a mixotrophic, mode of life
to generate enough reductant and ATP to support the From a cyanobacterial perspective, the Nostoc-Gunnera symbi-
demanding nitrogen fixation process (Söderbäck and Bergman osis may on the one hand seem wasteful; the cyanobiont merely
1992, 1993; Black et al. 2002; Khamar et al. 2010). Nevertheless, functions as an N-producing entity with highly suppressed
total pigment and ribulose-1,5-bisphosphate carboxylase levels growth and is possibly deprived of producing a new generation
remain constant along the developmental sequence, from young of cyanobionts, being enclosed in tissues in a long-lived plant.
to old parts, although values decrease if related to cell volume as On the other hand, it may be beneficial; the cyanobiont no
this increases in older cells (Söderbäck and Bergman 1992). doubt extends its ecological niche to also include symbiotically
The high frequency of heterocysts also drastically competent cells of an angiosperm. In this way, the cyanobacte-
diminishes the number of vegetative cells, but the use of gas rium not only gains access to plant leaves and roots and their
chromatography with mass spectrometry (GC-MS) has nutrient acquisition capacities but it also finds shelter from all
shown that still only the vegetative cells are actively taking possible predators, being the sole organism in this ‘‘golden cage.’’
up C (Black et al. 2002). Following 14C translocation in The data obtained so far clearly show that cyanobacterial
Nostoc-infected Gunnera stolons reveals that the Nostoc- morphology and protein and gene expression patterns are dras-
infected tissues at the apex of G. magellanica stolons are tically affected prior to, during, and after the establishment of
particularly efficient sinks for newly fixed plant carbon the Nostoc-Gunnera symbiosis, although no symbiosis-specific
(Söderbäck and Bergman 1993). The phloem of Gunnera genes and proteins, equivalent to the nod genes and Nod-factors
has the unusual capacity to contemporaneously transport in the Rhizobium-legume symbioses, have yet been discovered.
N outward and C inward toward the symbiotic tissue However, it seems logical to assume that equally advanced
(Stock and Silvester 1994). A tight interaction of nitrogen molecular mechanisms must persist in a cyanobacterial-
and carbon metabolism in the Gunnera symbioses is also angiosperm endosymbiosis to generate this potentially very
suggested (see, e.g., Chiu et al. 2005; Khamar et al. 2010). long-lived, well-coordinated, and successful interaction.
hetR expression in symbiotically competent Nostoc (PCC 9229)
is negligible in the absence of a carbon source in
darkness but pronounced in the presence of fructose (Wouters The Azolla Symbiosis
et al. 2000).
Introduction
Ecological Importance Taxonomy and Distribution
Our understanding of the ecology and significance (e.g., as The Azolla symbiosis is a mutualistic association between the
a nitrogen fixer) of this ancient plant and its cyanobiont is aquatic fern Azolla, the filamentous, heterocystous, nitrogen-
still rudimentary. For a detailed review of the ecology of fixing cyanobacterium Nostoc (formerly classified as Anabaena),
Gunnera, the reader is referred to Osborne and Sprent (2002). and endosymbiotic bacteria. The genus Azolla has been reported
The geographic range of Gunnera was considerably wider in to contain seven extant species that are divided into two sections
the past when the climate was more favorable (Osborne et al. on the basis of spore morphology. Section Azolla (New World
1991; Osborne and Sprent 2002). Today, Gunnera species) has included A. caroliniana, A. mexicana, A. filiculoides,
typically grows in super-humid habitats and often at high ele- A. microphylla, and A. rubra. However, the taxonomy of the New
vations or on steep cliffs. The genus is found in all continents, World species of Azolla has been the subject of much debate. In
except in Europe and polar regions (see Wanntorp et al. 2001; 2004, a comprehensive review of the literature was carried out
Osborne and Sprent 2002). Some large Gunnera species along with original observations of type specimens using optical
were introduced into Europe as ornamental plants at the end and scanning electron microscopy (Evrard and Van Hove 2004).
of the nineteenth century, and eventually some plants escaped This study confirmed the opinion of some that A. caroliniana
and became invasive in, e.g., western Ireland, the Channel and A. microphylla are synonyms of the previously described
Islands, and the Azores (Osborne et al. 1991; Osborne and A. filiculoides. To clarify the taxonomic classification, the authors
Sprent 2002). suggested the need to rehabilitate the Mettenius concept,
The genetics (rbcL and rps16 introns) of Gunnera plant and then according to the priority rule, the section Azolla
species have recently been analyzed (> Fig. 16.8). The large species must be named A. cristata and A. filiculoides.
species in South America and Hawaii distinctly group together Section Rhizosperma (Old World species) includes A. pinnata
in one clade, the often smaller species of New Zealand and and A. nilotica. Geographically, A. pinnata is found in Australia,
Southeast Asia group in another, while G. perpensa (the first New Zealand, Japan, Asia, and Africa, and A. nilotica is primarily
Gunnera to be described by Linné) and G. herteri (with the found in Africa (Saunders and Fowler 1993). Species from the
smallest size) are sister groups, representing Africa and Brazil, section Azolla are more widely distributed around the world and
respectively (Wanntorp et al. 2001). are found in Europe, Asia, Africa, Australia, and America.
However, the distribution of A. rubra is restricted to demonstrated, major as well as some minor cyanobacteria may
New Zealand and Australia (Large and Braggins 1993). The be present in the association, with the more readily cultured
distribution of some species has been impacted by human effects cyanobacteria representing minor symbionts (Gebhardt and
(Janes 1998a). Nierzwicki-Bauer 1991). The other possible explanation is that
the isolates presumably obtained from the association are actu-
ally epiphytes. However, recent studies of the genetic diversity of
Morphology cultured cyanobionts of diverse species of Azolla revealed
a genetic distinctness of the cultured Azolla cyanobionts as
The plant’s shape, color, and size change significantly under compared to free-living cyanobacterial strains of the genera
different growth conditions (Janes 1998b). The rhizome is Anabaena and Nostoc and symbiotic Nostoc strains from
branched, bearing alternate leaves that are bilobed. The ventral Anthoceros, Cycas, and Gunnera (Sood et al. 2008). These find-
lobe is transparent and serves to float the plant on the surface of ings support the coexistence of minor species rather than epi-
the water, whereas the dorsal photosynthetic lobe contains phytes. Regardless, based on molecular studies, it has not been
a leaf cavity in which the symbionts are found. The roots are demonstrated that the major cyanobacterial symbiont from
adventitious. The shedding of roots and branches is related to the association can be cultured in a free-living state. In fact,
environmental and physiological factors and enables the plant with the genome sequencing of the major cyanobiont of an
to reproduce via vegetative fragmentation. Factors affecting the A. filiculoides strain, there is now strong evidence (described
growth of Azolla include genotype, temperature, light (intensity, below) that there has been ongoing selective streamlining of
quality, and photoperiod), water chemistry (including pH, the cyanobiont genome which has resulted in an organism
salinity, and nutrients), and influence of pests and diseases devoted to nitrogen fixation and devoid of autonomous growth
(see Singh and Singh 1997). (Ran et al. 2010).
Given the challenge of studying the cyanobacterial symbi-
onts in a free-living state, direct molecular studies have been
General Characteristics used for accurate identification. Restriction fragment length
polymorphism (RFLP) analyses (Gebhardt and Nierzwicki-
The association has been most frequently used as an alternative Bauer 1991), polymerase chain reaction (PCR) fingerprinting
nitrogen fertilizer in rice fields, as well as a supplemental animal (Zheng et al. 1999), random amplified polymorphic DNAs
fodder. Azolla provides the cyanobiont with nutrients, including (RAPDs; Van Coppenolle et al. 1995), as well as fluorescence in
fixed carbon, and the cyanobiont provides the host with com- situ hybridizations (FISH) (Bushnell 1998) have been used to
bined nitrogen (via nitrogen fixation). The exact role of the examine the identity of the symbiotic cyanobacteria. Regardless
endosymbiotic bacteria in the association remains unclear; of the approach used, the cyanobiont referred to as ‘‘Anabaena
however, some possible functions have been suggested. azollae’’ has in most instances been described as being somewhat
related to Anabaena or Nostoc (Plazinski et al. 1990b; Gebhardt
and Nierzwicki-Bauer 1991). A study (Baker et al. 2003) using
The Symbionts comparisons of sequences of the phycocyanin intergenic spacer
and a fragment of the 16S rRNA gene places the Azolla
Cyanobacterial Symbionts cyanobiont in the order Nostocales but in a separate group
from Anabaena or Nostoc. Additionally, near full-length
Identification (1,500 bp) 16S rRNA sequencing and phylogenetic analysis of
The filamentous, heterocystous, nitrogen-fixing cyanobacterial major cyanobionts from a variety of Azolla species yielded sim-
symbionts in the Azolla association have been extensively stud- ilar results (Milano 2003). In 1989, Komarek and Anagnostidis
ied using both traditional and modern molecular techniques. In placed the Azolla cyanobiont in a revised genus named
addition to characterization of the cyanobionts after they have ‘‘Trichormus’’ on the basis of morphology (Komarek and
been directly extracted from the association, there have been Anagnostidis 1989). This is not inconsistent with the most
a number of studies in which cyanobacteria were isolated and recent molecular-based findings.
cultured, in attempts to study the cyanobacterial symbionts in Recently, the genome sequencing of a cyanobacterium
a free-living state. While numerous researchers have reported from Azolla filiculoides leaf cavities has provided the most
success in isolation and cultivation of the symbiotically associ- comprehensive information on its identity (Ran et al. 2010).
ated cyanobacteria (Newton and Herman 1979; Tel-Or et al. Surprisingly, the phylogenetic analysis places the cyanobiont
1983; Gebhardt and Nierzwicki-Bauer 1991; see Braun-Howland (Nostoc azollae 0708) most closely with Raphidiopsis brookii D9
and Nierzwicki-Bauer 1990), molecular studies (primarily based and Cylindrospermopsis raciborskii CS 505, the two multicellular
on restriction fragment length polymorphism [RFLP] analyses) cyanobacteria with the smallest known genomes (Stucken et al.
have indicated that none of the isolates represent the 2010). However, it shares the highest number of protein groups
major cyanobacterial symbionts in the association (Gebhardt with Nostoc sp. PCC 7120, Anabaena variabilis ATCC 29413, and
and Nierzwicki-Bauer 1991). Though not conclusively N. punctiforme PCC 73102 (Ran et al. 2010).
The taxonomy of the cyanobionts is generally in agreement PCR-DDGE and electron microscopy revealed a complex and
with the taxonomy of the host plant (Plazinski 1990; Van divergent bacterial community with Bacillus cereus as the dom-
Coppenolle et al. 1993; Zheng et al. 1999). These findings, inant species (Zheng et al. 2008).
taken in conjunction with the continuous maintenance of the
symbiosis throughout the life cycle of the plant (see the section
> ‘‘The Infection Process’’ below), suggest coevolution of the Host Structures and the Infection Process
cyanobionts and the host plant.
The Leaf Cavity
Developmental Profile Along the Main Stem Axis
The growth of the endophyte is coordinated with the growth of In the association, the symbionts reside in a leaf cavity, an
the plant. In the apical meristem and younger leaves, the extracellular compartment in the dorsal lobe of the leaf. In
cyanobacterial vegetative cells are smaller than in older leaves mature leaves, the symbionts (cyanobacteria and bacteria) are
and undergo frequent cell divisions. Increases in leaf age are located in the periphery of the leaf cavity in mucilaginous
accompanied by a decrease in cell division and increased size material between internal (Nierzwicki-Bauer et al. 1989) and
of the cyanobacterial vegetative cells, as well as increased hetero- external envelopes (Uheda and Kitoh 1991). Electron micro-
cyst frequencies (Hill 1975). The number of heterocysts and the scopic analysis combined with specific staining showed that
nitrogen fixation rates vary in leaves of different ages, as well as the inner envelope does not have a tripartite structure typical
in different Azolla species (Hill 1977). Heterocyst frequencies of a membrane and is rich in lipids (Nierzwicki-Bauer et al.
can reach up to 20–30 % of the cells within a filament in the 1989). The external three-layered envelope is believed to contain
symbiotically associated cyanobiont. These are much higher cutinic and suberic substances, as revealed by response to chem-
than the typical 10 % heterocyst frequency in free-living ical treatments of degradation using hot alkali methanol
Anabaena/Nostoc species. (de Roissart et al. 1994).
The adaxial epidermis of the leaf cavity contains a pore that
is surrounded by two cell layers (Veys et al. 1999, 2000).
Bacterial Symbionts One layer inside the pore is composed of teat-shaped cells
that are extended from the adaxial epidermis. The other layer
The presence of bacteria residing within the leaf cavity of Azolla corresponds to the inner epidermis, which lines the inside of the
has been recognized for many years (Carrapico 1991), yet still cavity. Three to four tiers of teat cells form a cone-like pore with
unclear is the specific function(s) of most of the bacteria in this an average diameter at the base of 80 mm. The pore opening is
symbiosis. Initial attempts to study the symbiotic bacteria larger in younger leaves, and the morphology of the teat cells
employed traditional microbiological, biochemical, and physio- suggests that their function is as a physical barrier to prevent
logical techniques for the identification of bacteria isolated particles and organisms from entering the cavity and the
from a variety of Azolla species. Utilizing these approaches, symbionts from exiting (Veys et al. 2002).
there were many reports of Arthrobacter spp. (most frequently
A. globiformis) occurring in symbiotic association with Azolla
(Gates et al. 1980; Wallace and Gates 1986; Forni et al. 1989, The Infection Process
1990; Nierzwicki-Bauer and Aulfinger 1991; Shannon et al.
1993). Agrobacterium has also been reported to be isolated Azolla is a heterosporous water fern that is capable of both sexual
from different Azolla species (Plazinski et al. 1990a; Shannon and asexual reproduction. Unlike any of the other cyanobacterial
et al. 1993; Serrano et al. 1999). Other bacteria, such as symbioses, the host is in continual association with the symbi-
Staphylococcus sp., Rhodococcus spp., Corynebacterium jeikeium, onts, making this the only known permanent symbiosis. Thus,
and Weeksella zoohelcum, were identified by BIOLOG and API rather than reinfect Azolla, the symbionts retain coordinated
tests as being in association with Azolla (Serrano et al. 1999). growth in association with the host throughout its life cycle.
A detailed review of the identification of bacteria isolated from Descriptions of the processes involved in maintaining the con-
Azolla species is provided in Lechno-Yossef and Nierzwicki- tinual association during sexual and asexual reproduction are
Bauer (2002). Molecular techniques, in particular 16S rDNA described briefly below.
gene amplification, cloning, screening, sequencing, and
phylogenetic analysis, have provided more detailed information Sexual Reproduction
on the identity of the symbiotic bacteria (Lechno-Yossef 2002; Sporulation is the sexual reproduction process in Azolla. During
Milano 2003). In the accessions studied, sequence similarity sexual reproduction, the host produces both mega- and
found that the most abundant bacterial symbionts in microsporocarps. The partitioning of the cyanobacterial fila-
A. caroliniana and A. filiculoides were Frateuria aurantia and ments into the developing sporocarps and the reestablishment
Agrobacterium albertimagni and in A. mexicana, Agrobacterium of the symbiosis following embryogenesis were originally
tumefaciens (Lechno-Yossef 2002). More recent research described for A. mexicana (Perkins and Peters 1993; Peters and
studying the endophytic bacteria within A. microphylla using Perkins 1993). The symbionts that are used as inoculum to the
developing sporocarps come from the dorsal lobe of the same Azolla and Nostoc, has been somewhat successful (Watanabe
leaf in which the sporocarps are developing. Recently, 1994; Watanabe and Van Hove 1996). For example, Nostoc
a comprehensive study of cellular responses in the from A. microphylla (MI4031) was successfully introduced into
cyanobacterial symbionts during its vertical transfer via A. filiculoides (FI1034) by exchange of the indusium cap of the
megasporocarps between plant generations in the A. microphylla megaspore (Lin et al. 1989). Successful sexual hybridizations
symbiosis was reported (Zheng et al. 2009). During colonization between A. microphylla (megasporocarp) and A. filiculoides
of the megasporocarp, the cyanobacterium entered through (microsporocarp; Wei et al. 1988; Do et al. 1989), between A.
pores at the top of the indusium as motile hormogonium fila- filiculoides (megasporocarp) and A. microphylla
ments. Subsequently, the cells differentiated into akinetes in (microsporocarp; Watanabe et al. 1993), and between A.
a synchronized manner. Also discovered was that this process mexicana and A. microphylla (Zimmerman et al. 1991) have
was accompanied by cytoplasmic reorganizations within the also been reported. The key to these successes has been having
cyanobionts and the release of numerous membrane vesicles, the cyanobacteria at the appropriate stage of development (dur-
most of which contained DNA, and the formation of a highly ing akinete germination and vegetative cell growth) that mimics
structured biofilm (Zheng et al. 2009). These data revealed what naturally occurs in situ.
complex adaptations in the cyanobacterium during transition
between plant generations that merit further investigation.
The cyanobiont akinetes (which function as spores) and the Host-Symbiont Signal Exchange
bacterial symbionts (which do not always show ultrastructural
characteristics of spore envelopes; Aulfinger et al. 1991) found in The recognition between Azolla and Nostoc azollae is facilitated
the megasporocarps are transferred to the developing spores and by lectins in both the plant (Mellor et al. 1981) and the
sporelings. After separation of the megasporocarp from the cyanobionts (Kobiler et al. 1981, 1982). Additionally,
plant, part of the indusium is shed, and the proximal half bacteria isolated from A. pinnata and A. filiculoides have been
becomes the indusium cap. The symbionts reside in a space shown to contain lectins (Serrano et al. 1999). The presence of
called ‘‘the inoculation chamber’’ (Peters and Perkins 1993), Rhizobiaceae symbionts in association with different Azolla
located between the indusium cap and the apical membrane of species and cultures examined would suggest that this group of
the megasporocarp. Following fertilization and the beginning of bacteria has a role in the symbiosis. Plazinski et al. (1991)
embryogenesis, the symbionts resume metabolic activity. With showed that the nodL and nodABC genes gave hybridization
the assistance of cotyledonary hairs, the symbionts are intro- signals to a plasmid and the chromosome of the isolate AFSR-
duced into the embryonic leaf before it displaces the indusium 1 from A. filiculoides. These authors suggest that the nod genes, if
cap (Peters and Perkins 1993). Leaves, which grow from the active in the bacterial symbionts of Azolla, play a regulatory role
meristem, are initially unlobed but contain a structure similar in the development of the symbiosis or in the maintenance of
to the leaf cavity that contains the symbionts. As the frondling bacterial association with the plant.
continues to grow, the symbionts are distributed into the
developing leaf cavities by a mechanism similar to the transfer
mechanism used during asexual reproduction via vegetative Host-Cyanobiont Interactions Post Infection
fragmentation (see next section).
Morphological Modifications to Host and
Asexual Reproduction Cyanobacteria
The main form of reproduction in Azolla is vegetative fragmen-
tation. The apical meristem of each branch contains a colony of The leaf cavity and inner and external envelopes of Azolla do not
undifferentiated cyanobacterial cells. Cyanobacterial filaments appear to be present only when it is symbiotically associated
from the apical colony are introduced into the leaf primordium with the cyanobionts. These structures are present in both
before the development of the leaf and leaf cavity are complete. Nostoc-free and Nostoc-containing Azolla (Nierzwicki-Bauer
The partitioning of the endophytes into the developing leaves is et al. 1989). This evidence excludes the involvement of the
facilitated by entanglement around primary branched hair cyanobionts in the formation of these structures. However,
(PBH) cells of Azolla (Calvert and Peters 1981). The leaf cavity given that the cyanobiont-free plants examined still contained
starts to develop and engulf the cyanobacterial colony in the symbiotic bacteria, a possible role of bacteria in the synthesis of
fourth or fifth leaf along the stem axis. In this way, symbionts are the leaf cavity envelopes cannot yet be excluded.
inoculated into every leaf cavity that is formed. The develop-
ment of the leaf cavity is also accompanied by the formation of
simple hair cells by Azolla (Peters and Calvert 1983). Nitrogen Fixation and Transfer of Fixed Nitrogen
‘‘Artificial’’ In Vitro Infection of Cyanobacteria Nitrogen fixation is carried out by the heterocysts of the
In sporulating Azolla, sexual hybridization between different cyanobiont. In leaf cavities of different ages along the stem
Azolla species, as well as the formation of new combinations of axis of Azolla, the heterocyst frequencies and nitrogen fixation
rates vary. Nitrogen fixation, as determined by the acetylene growth of Azolla into thick mats also makes it effective in
reduction assay, occurs in the apical (younger) leaves but not suppressing weed growth. Owing to its high protein content,
in the stem apex, increases and reaches a peak in leaves of middle Azolla is used as a fodder for sheep, pigs, ducks, etc. The ability
age, and then decreases in the older leaves (Canini et al. 1990). of Azolla to remove nitrates and phosphorous from water has
Ammonium, the product of nitrogen fixation, is released from resulted in improvement of water quality. Additionally, Azolla
heterocysts and assimilated by Azolla into glutamate using the has been used to remove heavy metals from water. Ten useful
glutamine synthetase (GS)-glutamate synthase (GOGAT) sys- characteristics attributed to this association have been described
tem (Peters and Calvert 1983). Nitrogenous compounds in the (Van Hove and Lejeune 1996; Lejeune et al. 1999), with the
form of glutamate, glutamine, ammonia, and other glutamate capacity to fix atmospheric nitrogen, high productivity, high
derivatives are transferred from the mature leaf cavities to the protein content, and a depressive influence on both aquatic
stem apex (Peters et al. 1985). In Nostoc azollae, the activity and weeds and NH3 volatilization being considered unquestionably
protein content of GS are only 5–10 % of that of free-living useful.
Anabaena (Orr and Haselkorn 1982). However, the nitrogen The same characteristic feature that makes Azolla useful for
fixation activity is much higher because of the increased number weed suppression and biofertilization of fields (namely, the
of heterocysts. Some of the bacteria found in this association can ability to grow in thick mats) also results in a number of negative
fix nitrogen. Immunoelectron microscopy studies using anti- ecological impacts. For example, growth of Azolla mats in
bodies against the Fe and FeMo protein subunits of nitrogenase streams in Zimbabwe has been shown to have a negative impact
revealed that a subset of the bacteria in the A. caroliniana and on animal biodiversity (Gratwicke and Marshall 2001). In many
A. filiculoides associations contained these nitrogenase subunits regions where Azolla is an invasive species, it has overgrown
(Lindblad et al. 1991). The potential nitrogen-fixing contribu- many native species. In efforts to control Azolla growth, biolog-
tion of the bacteria in the association separate from that of ical controls such as the introduction of a frond-feeding weevil
cyanobacterial symbionts could not be measured because they (McConnachie et al. 2004) or the flea beetle (Hill and
coexist in the leaf cavities and, once removed, are likely to have Oberholzer 2002) are being explored. Thus, the overall ecolog-
altered capabilities. ical impact of the Azolla association continues to expand and
may reach even to Mars, since Azolla is currently being used in
studies examining possible bioregenerative life support on Mars
Carbon Assimilation and Transfer of Fixed (http://www.highmars.org/niac/niac04.html, The Caves of
Carbon Mars Project website).
The cyanobiont, Nostoc azollae, has photosynthetic capabilities;

however, in the symbiotic state, it is believed to contribute less Cyanobionts Becoming a Nitrogen-Fixing
than 5 % of the total CO2 fixed in the association (Kaplan and ‘‘Organelle’’?
Peters 1988). Pulse-chase studies have shown that sucrose from
the plant is supplied to and accumulated by the cyanobiont The most important recent advancements related to the Azolla
(Peters et al. 1985). Simple hair cells of Azolla are involved in symbiosis have been proteomic (Ekman et al. 2008) and geno-
the transport of sugars from the photosynthetic mesophyll cells mic analyses (Ran et al. 2010) of the cyanobacterium of the
to the leaf cavity. Simple hair cells have ATPase activity in their A. filiculoides symbiosis. In brief, proteomic analyses revealed
plasmalemma and some accumulation of starch in their chloro- that processes related to energy production, nitrogen, and car-
plasts (which do not possess ribulose 1, 5 bisphosphate carbox- bon metabolism and stress-related functions (e.g., superoxide
ylase, RuBisCO), suggesting active transfer of sugars from the dismutase and peroxiredoxins) were upregulated in the
simple hairs to the leaf cavity (Carrapico and Tavares 1989). cyanobiont compared with a free-living strain, whereas photo-
Additionally, primary branched hair cells, having the morphol- synthesis and metabolic turnover rates were downregulated
ogy of a transfer cell, are believed to be involved in nutrient (Ekman et al. 2008). Genome sequencing of the cyanobiont of
transfer from the plant to the cyanobiont(s) (Peters et al. 1985). A. filiculoides strongly suggests that these cyanobionts are at the
initial phase of a transition from a free-living organism to
a nitrogen-fixing plant entity. There has been coevolution
Ecological Importance: Friend or Foe? between Azolla and the cyanobiont with genome degradation
and signs of reductive genome evolution resulting in an organ-
The Azolla symbiosis is of tremendous ecological importance, ism devoted to nitrogen fixation and devoid of autonomous
having both positive and negative impacts. On the positive side, growth (Ran et al. 2010). Noteworthy is the loss of function
the association has been extensively used as a biofertilizer, pro- within gene categories for basic metabolic processes such as
viding a source of combined nitrogen in the form of ammo- glycolysis, replication, and nutrient uptake. This genomic anal-
nium, thereby reducing or eliminating the need for the addition ysis now opens the door for obtaining a much better under-
of chemical fertilizers. This role has been most extensively used standing of this ecologically and evolutionarily important
in conjunction with rice paddies or fertilization of fields. The symbiosis.
The Cycad Symbioses
Introduction
First appearing in the Pennsylvanian era some 300 million

years ago, cycads are the most primitive and longest-lived of
present day seed plants (gymnosperms). They are also the
only gymnosperms that enter into symbiotic associations
with cyanobacteria. Once plentiful during the Jurassic, occupy-
ing far reaching habitats stretching from Alaska and Siberia
to the Antarctic, they are now found in diminishing numbers
in certain subtropical and tropical regions of mostly the
southern hemisphere, including Australia, parts of Southern
Asia, and South Africa (Brenner et al. 2003; Vessey et al. 2005).
There are two or three extant families with some 300 species in
10 genera (Chaw et al. 2005; Vessey et al. 2005; Bergman et al.
2007; see also Lindblad 2009). Cycads are long-lived, cone-
bearing evergreen palm-like plants which can reproduce either
asexually (producing stem offshoots or suckers) or sexually.
They grow terrestrially, with the exception of Zamia
pseudoparasitica which is the only true epiphytic cycad and
hangs from branches by the tap and lateral roots
(Stevenson 1993).
In general, cycads have a stout trunk with a large crown
of tough spiny leaves and can vary in height from a few tens of
centimeters to almost 20 m at maturity. Most cycads
produce different root types—a thick taproot that extends
some 9–12 m beneath the soil surface, lateral roots, and coralloid . Fig. 16.9
roots which are highly specialized lateral roots named for The cycad-Nostoc symbiosis. (a) A cycad coralloid root, the site of
their resemblance to coral. Coralloid roots exhibit negative cyanobacterial infection. (b) Transverse section of the root
geotropism, growing sideways and upward toward the soil sur- showing the dark cyanobacterial band between the inner and
face, and are the sites in which symbiotic cyanobacteria can outer cortical layers [(a) From Lindblad et al. (1985a) with
be found (Costa and Lindblad 2002; Lindblad and Costa permission. (b) From Rai et al. (2000) with permission]
2002; Brenner et al. 2003; Vessey et al. 2005; Bergman et al.
2007, 2008).
Cyanobacteria living in association with cycads were first
reported in the nineteenth century (Reinke 1872), and this Heterotrophic bacteria have been found associated with
partnership is still the only known example of a naturally occur- cyanobacteria recovered from coralloid roots (Chang et al.
ring plant root-cyanobacterial symbiosis. All cycad species 1988), although these bacteria have not been found in the
examined to date are able to form symbiosis with nitrogen- cyanobacterial zone within the coralloid cortex (Grilli Caiola
fixing cyanobacteria, visible as a dark blue-green band 1980). However, bacteria have been found inside the periderm of
(the cyanobacterial zone) between the inner and outer coralloid the coralloid roots of several different cycads (Joubert et al.
root cortex (> Fig. 16.9). The association between the cycad 1989). It has been suggested that phenolic substances,
Zamia furfuracea and its native cyanobiont Nostoc sp. strain detected in the mucilaginous material of the cyanobacterial
FUR 94201 has been separated and reconstituted successfully zone and in the cortical cells surrounding the cyanobacterial
under axenic laboratory conditions (Ow et al. 1999). The same zone, have antimicrobial properties that exclude organisms
authors also showed for the first time that a cycad symbiosis other than cyanobacteria (Grilli Caiola 1980; Obukowicz
could be established with the soil cyanobacterium Nostoc 2S9B, et al. 1981). Phenolics are among the most widespread
a strain previously shown to form loose associations with wheat plant secondary metabolites and have been shown to func-
roots. The ability of cycads to thrive in nutrient-poor soils is tion as signaling molecules in the establishment of legume-
often attributed to the associations they form with rhizobial symbiosis and vesicular-arbuscular mycorrhiza,
cyanobacteria. Nitrogen fixation in cycads not only contributes and there is some evidence that phenolic compounds may
to the nitrogen metabolism of the plant but also up to also influence the formation of symbiosis between
18.8 kg N ha1 year1 to the local nitrogen economy (see: Rai cyanobacteria and cycads, as well as their metabolism (see
et al. 2000; Vessey et al. 2005). Lobakova et al. 2004).
Cycads and Toxicity studies have revealed little if any specificity between cycads and
their cyanobionts (Lindblad et al. 1989; Lotti et al. 1996; Costa
The toxic properties of cycads have been noted for centuries, et al. 1999; Zheng et al. 2002; Costa et al. 2004; Gehringer et al.
with the azoxyglycosides, cycasin and macrozamine, and the 2010); the cyanobiont species found within a host plant is
neurotoxic nonprotein amino acid b-methylamino-L-alanine probably determined by the predominant symbiotically compe-
(BMAA), receiving the most documentation. BMAA synthesis tent species available within the immediate rhizosphere
has been associated with the cyanobionts of cycads (Cox et al. (Gehringer et al. 2010).
2003) as well as the host plant (e.g., Vega and Bell 1967;
Polsky et al. 1972; Marler et al. 2010) and has since been found
in all known groups of free-living cyanobacteria (Cox et al. 2005; Establishment of Symbiosis
Banack et al. 2007). BMAA has been linked to the high incidence
of the progressive neurodegenerative disease amyotrophic lateral Coralloid root development begins with the initiation of
sclerosis/parkinsonism-dementia complex (ALS-PDC) in the precoralloid root formation in the seedling (Rai et al. 2000;
Chamorro people on the island of Guam. They were thought Lindblad 2009). Cyanobacteria are not found in precoralloids,
to have acquired damaging levels of the toxin through the and their presence is not necessary for the initiation of
ingestion of cycad seed-eating flying foxes in which the toxin is precoralloid development (Staff and Ahern 1993), although
believed to have been ‘‘biomagnified’’ (Banack and Cox 2003; exposure to light is considered significant in many cycad genera
Cox et al. 2003; Banack et al. 2006). However, this hypothesis is (Webb 1983a, b). Infection by cyanobacteria can occur at any
controversial (see Marler et al. 2010; Snyder and Marler 2011) stage of precoralloid root maturation, and their presence triggers
not least because of difficulties in reliable separation and detec- further growth and the developmental changes required to
tion of BMAA. Some groups have confirmed the presence of transform precoralloids into coralloids. The mode of entry of
BMAA in cyanobacteria and cycad seeds (Esterhuizen and the cyanobacteria is still unclear, although suggested access
Downing 2008; Spáčil et al. 2010), whereas contradictory results points have included lenticels, breaks in the dermal tissue, or
have been obtained following analysis of underivatized samples via the papillose sheath (see Bergman et al. 2007; Lindblad
(e.g., Rosén and Hellenäs 2008; Krüger et al. 2010). These latter 2009). In addition, bacteria and fungi in the cycad rhizosphere
studies were, however, able to detect 2, 4-diaminobutyric acid may cause local degradation of the cell wall, enabling the
(DAB), a neurotoxic isomer of BMAA (Rosén and Hellenäs cyanobacteria to penetrate the root (Lobakova et al. 2003).
2008; Krüger et al. 2010). The recent work by Banack et al. Following entry into the root, the cyanobiont migrates toward
(2010) addressed the issues concerning effective BMAA analysis, the cyanobacterial zone, between the inner and outer cortex,
and they were able to reliably and consistently separate BMAA through a channel created through the outer cortex, believed to
from 2,4 DAB as well as distinguish it from other compounds be caused by the separation, distortion, and destruction of
previously mistaken for BMAA during chloroformate cortical cells (Nathanielsz and Staff 1975).
derivitization for GC analysis (Banack et al. 2010). They con- The process of coralloid formation is irreversible, with one of
cluded that cyanobacteria do indeed produce BMAA and its the most significant changes being a conversion from negative to
neurotoxic structural isomer 2, 4-DAB. positive geotropism, resulting in growth sideways and upward
The biological significance of the toxins remains unclear, toward the soil surface. Other changes include the loss of the
although possible roles include protection from herbivory, com- papillose sheath, proliferation of apical lenticels, and early dif-
petition with other plants, or antibacterial and antifungal ferentiation of the conspicuous cyanobacterial zone (Ahern and
defense mechanisms (Castillo-Guevara and Rico-Gray 2003). Staff 1994; see also Rai et al. 2000). Some cycad cells within the
BMAA is, however, heavily concentrated in the coralloid roots, cyanobacterial zone undergo a distinct differentiation process to
raising the suggestion that the primary function of this toxin is interconnect the two adjacent cortical layers, and this may facil-
not anti-herbivory (Marler et al. 2010). An alternative role might itate the transfer of nutrients between the partners (Lindblad
be in communication involved in the initiation and mainte- et al. 1985a; Lindblad et al. 1991; Pate et al. 1988; Costa and
nance of the symbiotic association (Marler et al. 2010; Snyder Lindblad 2002; Vessey et al. 2005; see also Lindblad 2009).
and Marler 2011). Although the cyanobiont location is extracellular, there have
been reports of intracellular cyanobionts in Cycas revoluta
Thunb. and Macrozamia communis L. (see Lindblad 2009 and
The Symbionts references therein).
The cyanobionts of cycads are filamentous, heterocystous spe-

cies, largely restricted to the genus Nostoc, although Calothrix Symbiotic Competence: Fit to Infect?
spp. have occasionally been found (Grobbelaar et al. 1987; Costa
and Lindblad 2002; Rasmussen and Nilsson 2002; Gehringer So, how do cyanobacteria living in the soil reach the sites
et al. 2010; Thajuddin et al. 2010). Cycads can host multiple of infection within the coralloid roots of cycads? As is the case
cyanobacterial strains in single plants as well as in single roots with other plant hosts, cyanobacteria entering into functional
(Zheng et al. 2002; Thajuddin et al. 2010). However, taxonomic symbiosis with cycads produce transient motile filaments
known as hormogonia that are highly adapted to sense and of Cycas circinalis which are known to develop at or close to the
respond to environmental stimuli, including chemicals released soil surface where the cyanobacteria may be exposed to light
by potential host plants. Water extracts from macerated seeds of (Perraju et al. 1986). Nevertheless, it is unclear why the
the cycad Zamia furfuracea induce the development of motile dark-associated cycad cyanobionts retain a full photosynthetic
hormogonia (Ow et al. 1999) and exhibit some chemoattractive apparatus, associated pigments, and carbon-fixing potential
properties. Chemotaxis of hormogonia in response to plant- (cellular levels of RuBisCo are comparable with those in free-
derived attractants is likely to be of particular importance living counterparts).
in the infection of cycad tissue, where the sites of infection
receive little or no light, because the plant-derived signals must
be able to override the natural phototactic response of the Nitrogen Metabolism
hormogonia.
Nitrogenase protein is restricted to the heterocysts, including
contiguous heterocysts, in both free-living heterocystous
Other Symbiotic Competence Factors cyanobacteria and those in symbiosis with cycads (Bergman
et al. 1986). Nitrogenase activity is some three- to fivefold higher
Laboratory attempts to reconstitute a functional symbiosis in cycad-associated Nostoc than in the free-living cultures
between Nostoc PCC 73102 (originally isolated from the cycad (Lindblad et al. 1985b). Although nitrogenase activity parallels
Macrozamia sp.) and Nostoc ATCC 29133 (believed to be the increasing heterocyst frequency, it reaches a maximum at het-
same isolate as PCC 73102 but with a different laboratory erocyst frequencies of around 25–35 % (Lindblad et al. 1985b) at
history and morphology; see Ow et al. 1999 and references a location where single heterocysts predominate, and declines
there in) have been unsuccessful (Ow et al. 1999). This is sur- thereafter, although heterocyst frequency continues to increase
prising because, as described elsewhere in this chapter, Nostoc and contiguous heterocysts become common (Lindblad et al.
ATCC 29133 readily enters into functional symbiosis with a wide 1985b). The possibility that some of the heterocysts within these
range of plants such as liverworts, hornworts, and the angio- clusters are metabolically inactive cannot be ruled out (Bergman
sperm Gunnera. This implies the existence of a recognition and et al. 1986). Indeed, the decrease in nitrogenase activity observed
compatibility selection process that is able to select certain in older parts of the coralloid roots of Cycas and Zamia is
cyanobacteria and exclude others. Alternatively, successful believed to be in part due to the aging of the cyanobacteria
cyanobionts may have evolved mechanisms to protect against located there (Lindblad 1990).
or disguise themselves from the host’s natural defense system. Cyanobacteria freshly isolated from older coralloid root
sections, as well as those located close to the apex of Macrozamia
riedlei, an Australian cycad in which the coralloid roots can
Carbon Metabolism develop up to 0.5 m below the soil, show marked light-
stimulated nitrogenase activity (measured by both acetylene
As the coralloid roots of most cycad species are located beneath reduction and 15N2 fixation), providing they are maintained
the soil surface, their cyanobionts receive little or no light and under low (<1 %) O2 levels (Lindblad et al. 1991). The low
are assumed to have a heterotrophic metabolism, possibly using level of nitrogenase activity recorded in freshly isolated
fixed carbon supplied by the host and/or by their own dark CO2 cyanobionts incubated in darkness probably results from loss
fixation (Lindblad 2009). Freshly isolated cyanobionts from or damage of heterotrophic mechanisms that previously func-
Cycas revoluta coralloid roots fail to fix CO2 in vivo under tioned to provide the necessary ATP to support nitrogenase
light or dark conditions (Lindblad et al. 1987), but their crude activity within the intact coralloid roots (Lindblad et al. 1991;
extracts have similar activities of RuBisCo and phosphoribu- see also Lindblad 2009). The authors suggested that damage
lokinase to those in free-living cultures (Lindblad et al. 1987). caused by separation of the cyanobacteria from the host plant
The lack of CO2 fixation by the cyanobiont might be due to disrupts the intercellular microenvironment and any biochem-
a reversible inhibitor of RuBisCo, which is lost by dilution in the ical interactions provided by the intact cyanobiont-coralloid
in vitro assay (see Meeks and Elhai 2002). Alternatively, the root association (Lindblad et al. 1991).
photosystems of the cyanobiont may be nonfunctional, or In free-living cyanobacteria, the ammonia derived
other enzymes of the Calvin cycle may be lacking (Lindblad from nitrogen fixation is primarily assimilated via the
et al. 1987). glutamine synthetase-glutamate synthase (GS-GOGAT) path-
In a study of the Nostoc-like cyanobiont of Zamia skinneri, way (Muro-Pastor et al. 2005; Flores and Herrero 2005). In
a fully developed photosynthetic apparatus containing thyla- many cyanobacteria-plant symbioses, the release, to the host,
koids with distinct phycobilisomes, harboring phycobili- of ammonia that would otherwise be assimilated by the
proteins, was revealed (Lindblad et al. 1985a). Carboxysomes cyanobiont is achieved partly by a host-mediated decrease in
were also noted, although photosynthesis is unlikely to occur as GS activity in the cyanobiont. However, xylem sap analysis of
the coralloid roots were collected from below the soil surface. freshly detached Nostoc-colonized coralloid roots has
However, high rates of photosynthetic oxygen evolution revealed that the nitrogen fixed by the cyanobiont is instead
have been found in the cyanobacteria isolated from the roots translocated to the host as either a combination of the amino
acids citrulline and glutamine (in Zamiaceae) or glutamine cyanobacteria (tripartite lichens). A lichen thallus is quite dis-
alone (although in the Boweniaceae and Cycadaceae glutamic tinct in appearance from either of its symbionts, and its name
acid is also present; Pate et al. 1988; Costa and Lindblad 2002; refers to the dominating fungal partner (the mycobiont). Lichen
Bergman et al. 2007). The possibility that ammonium is not the thalli represent an integration of the mycobiont’s heterotrophic
transferred N-solute in cycad associations is supported by metabolism and the autotrophic metabolism of the photosyn-
the findings that cyanobionts of Cycas revoluta, Ceratozamia thetic partners (the photobionts: green algae and
mexicana, and Zamia skinneri all have high in vitro GS cyanobacteria). In tripartite lichens, the cyanobacterial partner
activity and GS protein levels similar to those found in (the cyanobiont) is also referred to as the ‘‘secondary
free-living cyanobacteria, including strains originally isolated photobiont,’’ whereas the green algal partner is referred to as
from cycads (Lindblad and Bergman 1986). the ‘‘primary photobiont.’’ All lichens having a cyanobiont,
Despite supplying the host with most of the nitrogen either as the sole photobiont or as a secondary photobiont, are
they fix, the cyanobionts do not show signs of nitrogen starva- called ‘‘cyanolichens.’’ For lack of space, this review on
tion, as they retain abundant sources of combined nitrogen. cyanolichens is brief. The reader can find further details in
These include cyanophycin, which is a specialized nitrogen books and reviews elsewhere (Galun 1988; Ahmadjian 1993;
reserve (consisting of a copolymer of arginine and aspartic Nash 1996; Rai et al. 2002; Rikkinen 2002). The journal Bryol-
acid), carboxysomes, and phycobiliproteins (accessory ogist regularly lists recent literature on lichens, and a literature
photopigments), all of which can be degraded under conditions search is also possible at Mattick’s Literature Index website
of nitrogen starvation (Meeks and Elhai 2002). ([{http://www.toyen.uio.no/botanisk/bot-mus/lav/sok_rll.htm}]).
There are approximately 1,550 known species of
cyanolichens, representing roughly 12–13 % of all known
Other Adaptations to Life in Symbiosis lichens; among these, two-thirds are bipartite and the rest tri-
partite species. Lichen symbioses are thought to have arisen
Cycad cyanobionts generally show little morphological change independently on several occasions. An estimated 100
compared with their free-living counterparts, apart from lichenization events have occurred during diversification of
an increased heterocyst frequency. Baulina and Lobakova extant fungi (Aptroot 1998; see also Rikkinen 2002).
(2003a, b) observed cyanobionts with vegetative cells and het-
erocysts showing considerable degradation of the peptidoglycan
layer. However, it is not clear if such cells are functional or are in The Symbionts
various states of senescence. In free-living cyanobacteria, het-
erocysts largely occur singly at almost regularly spaced intervals Mycobionts
within a filament of photosynthetic vegetative cells (reviewed by
Zhang et al. 2006). In return for the fixed carbon (possibly The current classification of fungi is in transition, and molecular
sucrose; see Meeks and Elhai 2002 for further discussion) they approaches are being used to fine-tune it (Tehler et al. 2000; see
receive from neighboring vegetative cells, the heterocyst pro- also Rikkinen 2002). Approximately 13,500 species of lichen-
vides fixed nitrogen. In symbiosis, there is a developmental forming fungi presently belong mostly to the Ascomycetes
gradient of heterocysts from low (a heterocyst frequency of (98 %) and very few to the Basidiomycetes (1.6 %) and fungi
16.7 % of total cells) in the growing tips of the coralloid roots imperfecti (0.4 %). About 15–18 orders of Ascomycetes (nearly
to high (46 %) in the base (older parts) of the roots. Indeed, at 130 genera from 50 families) include lichen-forming taxa (see
the very growing tips of the coralloid roots of some cycad species Rikkinen 2002). Most are from two orders, the Lecanorales and
are short ‘‘free-living’’ type filaments that resemble hormogonia. Lichinales. Nearly 1,700 species of fungi associate with different
There are few heterocysts, those present being restricted mostly types of cyanobacteria. A fairly comprehensive list of these has
to the filament poles (Grilli Caiola 1980). In various cycad been provided earlier (Rikkinen 2002).
species, multiple contiguous heterocysts (double to quadruple)
are found frequently in older root tissue but rarely at the tip (see
Lindblad et al. 1985a, 1985b; Lindblad et al. 1991: also Cyanobionts
reviewed by Lindblad 2009). However, these contiguous
heterocysts in older tissue may not be metabolically active A variety of heterocyst-producing and unicellular cyanobacteria
(Bergman et al. 1986). occur as cyanobionts in cyanolichens where the mycobiont is an
ascomycete. Among heterocystous forms, Nostoc is the most
common. Others are Scytonema, Calothrix, Dichothrix, and
Cyanolichens Fischerella (including Hypomorpha, Stigonema, and
Mastigocladus). Unicellular forms that occur as cyanobionts in
Introduction cyanolichens include Gloeocapsa (also Chroococcus), Gloeothece,
Synechocystis (also Aphanocapsa), Chroococcidiopsis, Hyella, and
Lichens are associations of symbiotic fungi and green algae Myxosarcina (see Rai et al. 2000; Rikkinen 2002). The range of
(bipartite lichens) or symbiotic fungi, green algae, and cyanobionts in cyanolichens where the mycobiont is
a basidiomycete is rather limited. Only two cyanobacteria a preexisting thallus (direct transmission) or from fresh synthesis
(Chroococcus and Scytonema) are reported as cyanobionts in (fresh acquisition of cyanobiont from the environment). The
basidiolichens (see Schenk 1992). former mode of transmission allows prolonged continuity of the
Analyses of tRNALeu (UAA) introns and 16S rDNA partners. Similar modes of cyanobiont acquisition also apply to
sequences have been used as genetic markers to study the diver- the development of cephalodia (see Rai et al. 2000). Cyanobionts
sity of Nostoc cyanobionts (Paulsrud and Lindblad 1998; are essential for the formation of thalli or cephalodia in
Paulsrud et al. 1998, 2000, 2001; Lohtander et al. 2002; Rikkinen cyanolichens. They may stimulate thallus morphogenesis but
et al. 2002). These studies have shown that genetic do not determine the kind of thallus formed; the mycobiont
variation among lichen-forming Nostoc strains is considerable. determines the structure and chemistry of a cyanolichen.
Within symbiotic Nostoc strains, there seem to be several Different lichen fungi form different lichen thalli even if
subgroups. For example, one subgroup of Nostoc strains seems associating with the same cyanobiont (see Rai 1990; Rai et al.
to occur only in epiphytic cyanolichens, whereas another 2000, 2002).
includes strains that occur as cyanobionts in terricolous Because they are slow growing, the initiation and develop-
cyanolichens and other symbiotic systems (Rikkinen 2002; ment of lichens is difficult to study in nature. Development of
Rikkinen et al. 2002). Miura and Yokoto (2006) have reported a lichen thallus afresh involves germination of the mycobiont
the occurrence of two cyanobionts in the same lichen. Based on spore, development of the hyphal mat, contact, recognition, and
morphological observations and 16 s rDNA sequences of acquisition of the cyanobiont, and structural-functional integra-
cyanobacterial isolates from lichens, they reported the tion of the symbionts. While a thallus may result within months
occurrence of Nostoc, Calothrix, Cylindrospermum, Phormidium, when starting from propagules, it takes years when starting from
Leptolyngbya, Microcystis, and Chroococcidiopsis. isolated partners. During laboratory synthesis of lichens, the
partners initially form undifferentiated aggregates that later dif-
ferentiate into thalli (see Rai et al. 2000). Fresh synthesis in nature
The Lichen Thallus may also start from mycobiont hyphae that become detached
and acquire a fresh cyanobiont (Smith and Douglas 1987).
Lichen thalli have a stable and organized structure quite distinct Development of each cephalodium is a new event. External
from any of their symbionts. The thalli appear to be crustose cephalodia develop on the main thallus by entrapment of
(small lobes and scales; e.g., Collema), foliose (flat and dorsi- a cyanobiont by hairs on the thallus surface, followed by involve-
ventral lobes; e.g., Peltigera), or fruiticose (round or flat thalli, ment of medullary hyphae immediately below. Internal
upright, or hanging down from the substratum; e.g., cephalodia may develop in a similar fashion starting with
Stereocaulon). In foliose or fruticose thalli, the fungal hyphae cyanobiont entrapment by cortical hyphae or rhizines. The
form an outer pseudoparenchymatous zone (the cortex) that cyanobiont, enmeshed by a thick layer of mycobiont, is pressed
covers or encloses a more loosely interwoven medulla. Within into the thallus where the cephalodium eventually develops.
the thallus, the partners remain extracellular to each other and New cephalodia may develop from hormogonia released by
can be isolated and grown in culture, but the symbiosis is fairly earlier cephalodia (Stocker-Wörgötter 1995), ensuring
stable in nature because of the balanced and synchronized cyanobiont homogeneity among cephalodia of a thallus. In
growth and development of the symbionts. Thinner cell walls laboratory synthesis, however, cephalodia developed by attach-
(less sheath material) and specialized hyphae and haustoria, ment of hyphae from primordia (containing cyanobiont and
showing transfer cell ultrastructure, enable close contact mycobiont) to the green thallus (Stocker-Wörgötter and Turk
between the mycobiont and the cyanobiont. Since the bulk of 1994). The latter mode of cephalodia development, if prevalent
the thallus consists of the heterotrophic mycobiont, the thallus in nature, should cause considerable heterogeneity among sym-
interior is microaerobic (see Rai et al. 2000). biont populations within a single thallus, but this is not the case.
In bipartite lichens, cyanobionts either are dispersed Occasional reports of different cyanobionts (see Rai 1990) or
throughout the thallus (e.g., Collema) or occupy a distinct different strains of a cyanobiont (Paulsrud et al. 2000) among
layer below the upper cortex (e.g., P. canina). In tripartite cephalodia of a single thallus may, however, indicate instances of
lichens, the cyanobiont is located in cephalodia, which occur at cephalodia development by capture of a fresh cyanobiont in
the upper surface of the thallus (external cephalodia; e.g., in P. some lichens. Entry of the cyanobiont for development of inter-
aphthosa) or inside the medulla (internal cephalodia; e.g., in nal cephalodia is from the lower surface of the thallus, but
Nephroma arcticum). In some cases, internal cephalodia are occasionally, when the cyanobiont enters from above, the
found close to the lower surface of the thallus (e.g., in P. venosa). phycobiont layer is pressed deep into the medulla.
In tripartite lichens, direct contact between the cyanobiont and
the phycobiont (green algal partner) is never direct.
Lichen symbioses perpetuate by direct transmission of the Recognition and Signal Exchange Between
cyanobiont from one generation to the next and, as a result of Partners
the acquisition, by the mycobiont of fresh cyanobiont from the
environment. For example, a lichen thallus can develop from For the right symbionts to enter into a lichen symbiosis, signal
propagules (phyllidia, isidia, soredia, and hormocystangia) of exchange must occur between the partners. Transformation of
Nostoc colonies into the symbiotic state occurred without the lower surface (e.g., Peltigera venosa), may be minimal due to
necessity for direct contact with the mycobiont during low light and RuBisCo. In the tripartite Nephroma arcticum, the
resynthesis of Peltigera praetextata (Yoshimura and Yamamoto Nostoc cyanobiont has 70 % fewer carboxysomes compared with
1991). This suggests that the substance responsible for Nostoc that in the bipartite P. canina (Bergman and Rai 1989).
transformation may be a diffusible soluble substance from the In free-living cyanobacteria, heterocysts are regularly
mycobiont. The exact identity of such a substance is not known, spaced and represent about 5–10 % of the cell population.
but lichen-forming fungi do produce a large number of unique There is a change in the spacing pattern of heterocysts and an
secondary metabolites and compounds, and their possible roles increase in their frequency in the cyanobionts in tripartite
in signal exchange need to be investigated. Lectins (glycopro- lichens (heterocyst frequency 15–35 %) but not in bipartite
teins) of mycobiont origin have been implicated in the recogni- lichens. Heterocyst frequency correlates with the status of
tion of the cyanobiont by a mycobiont (Rai 1990; Kardish et al. fixed carbon in the cyanobiont; in bipartite lichens, the
1991; Lehr et al. 1995, 2000). Cyanobiont cell surfaces possess cyanobiont bears the burden of providing both fixed nitrogen
specific sugars, fimbriae (pili), and in some cases, lectins, which and fixed carbon to the mycobiont, whereas in tripartite
may have a role in recognition and adherence (Stewart et al. lichens, it provides fixed nitrogen only. Indeed heterocyst
1983; Kardish et al. 1991; see Rai et al. 2000). frequency increases when Nostoc isolates are grown in the
Direct observations, lectin-binding experiments, and dark with sugars. In many cyanobacterial-plant symbioses,
tRNALeu intron analysis all indicate a broader cyanobiont- where the cyanobiont receives fixed carbon from the plant
mycobiont specificity in lichens than that in other host, heterocyst frequencies of up to 80 % can occur (see Rai
cyanobacterial symbioses. Different lichen species can have the et al. 2002).
same cyanobiont, and different cyanobionts have been reported In free-living cyanobacteria, glutamine synthetase (GS) is
among cephalodia of a single lichen thallus. Different Nostoc the primary ammonia assimilating enzyme, and GS levels in
strains have been found in different lichen species from the same heterocysts are twofold higher than those in vegetative cells
site, while different lichen species from distant places had the (Bergman et al. 1985). In cyanobionts, the GS activity and
same Nostoc strain. In chimeroid thalli, both bipartite and tri- protein levels decrease by over 90 %, and the remaining GS is
partite morphotypes are reported to have the same cyanobiont uniformly distributed among heterocysts and vegetative cells
strain (Paulsrud et al. 1998, 2000, 2001). Overall, there is a great (Bergman and Rai 1989; Rai 2002). GS activity is undetectable
deal of cyanobiont diversity among the lichens, and much of it in the mycobiont, but mycobiont hyphae in contact with
might be contributed by the mode of cyanobiont acquisition cyanobiont cells show high levels of nicotinamide adenine dinu-
during the development of the lichen thallus and cephalodia cleotide phosphate (NADP+)-dependent glutamate dehydroge-
(Rai et al. 2000). nase (GDH) activity.
Many cyanolichens share similar environmental require- Nitrogen fixation occurs in all lichens containing hetero-
ments and may depend on a common pool of cyanobionts. cystous cyanobionts. The rates are higher in tripartite lichens
Many cyanolichen species having identical cyanobiont owing to the higher heterocyst frequency of the cyanobiont
strains co-occur in a particular habitat, forming character- (see Rai et al. 2000; Rai 2002). In contrast to free-living forms,
istic communities or ‘‘guilds’’ (Rikkinen et al. 2002). Within cyanobionts in bipartite lichens and in excised cephalodia con-
a guild, the cyanobionts of all lichens are closely related, but the tinue to fix N2 even in the presence of nitrate or ammonia
mycobionts are not. While some guilds include different (Stewart and Rowell 1977; Rai et al. 1980). However, nitrogen
mycobiont genera or even families, some closely related fixation by the cyanobiont in cephalodia attached to the main
mycobionts belong to different guilds (associate with different thallus of the tripartite lichen P. aphthosa was repressed by
types of cyanobionts). nitrate and ammonia. The effect was obviously mediated via
the phycobiont. Significant levels of N2 fixation have also been
reported in darkness, and under these conditions ammonia
Structural-Functional Changes has an inhibitory effect (Rai et al. 1981a, 1983b). As in the
free-living forms, nitrogenase is located only in the heterocysts,
Cyanobionts undergo structural-functional changes in the despite the microaerobic conditions in lichen thalli (Bergman
symbiosis that permit a close interaction and development et al. 1986).
of nutrient exchange between the partners. These changes The extent of the changes described above varies from young
include increased cell size, altered cell shape, lack of to older, more mature parts of the thallus. While growth rate
polyphosphate reserves, fewer carboxysomes, less sheath gradually declines, cell division and GS levels, the levels of N2
material, and slower growth and cell division (Rai et al. fixation, CO2 fixation, and heterocyst frequency of the
2000). cyanobiont increase. There is a parallel increase in the GDH
The cyanobionts are photosynthetically active and fix CO2 activity in the mycobiont (Rowell et al. 1985; Hill 1989;
via the C3 pathway. In addition, there is a significant level of dark Rai 2002). Still undetermined is whether these changes are
CO2 fixation (15–20 % of that in the light) via the C4 pathway caused by the mycobiont or by endogenous regulation due to
(Rai et al. 2000; Palmqvist 2002). However, CO2 fixation by special environmental conditions offered by the host in the
cyanobionts in internal cephalodia, particularly those on the symbiosis.
N2
NITROGENASE
Other amino
NH3 NH4+
acids
GS-GOGAT
a?
CYANOBIONT d
NH3 NH4+
d a?
MYCOBIONT
NH3 NH4+
GDH
Other amino
Glutamate
acids
GOT
GPT
Aspartate
APT Alanine MAIN
THALLUS
. Fig. 16.10
Pathways of N-metabolism in P. aphthosa. a, active transport; d, diffusion. GS glutamine synthetase, GOGAT glutamate synthase,
GOT glutamate-oxaloacetate transaminase, GPT glutamate-pyruvate transaminase, GDH glutamate dehydrogenase, APT aspartate-
pyruvate transaminase
Nutrient Exchange the isolation of the cyanobiont, indicating the influence of

mycobiont and symbiotic conditions in the thallus on this pro-
Most studies on nutrient exchange relate to carbon and nitrogen cess (Smith and Douglas 1987; Meindl and Loos 1990; Rai 1990;
transfer from the cyanobiont to the mycobiont in foliose lichens, Palmqvist 2002).
particularly Peltigera species. Such nutrient transfer is biotrophic 15
N tracer studies in P. aphthosa (tripartite) and P. canina
in nature and varies along the lichen thallus. From young to (bipartite) have concluded that fixed N is transferred from
mature parts of the lichen thallus, the cyanobiont increases cyanobiont to the mycobiont as ammonia (Rai et al. 1981b,
fixation and release of nitrogen and carbon. Specialized 1983a). Over 90 % of the N2 fixed in P. aphthosa (and
mycobiont hyphae and haustoria showing transfer cell ultra- about 50 % in P. canina) is released by the cyanobiont because
structure (TCU) may play an important role in the nutrient GS in heterocysts is repressed. The partitioning of fixed N among
exchange (see Rai et al. 2000). the partners is proportionate to their contribution to the thallus
In bipartite cyanolichens, 70–80 % of the CO2 fixed is composition. In P. aphthosa, the ammonium released by
released by the cyanobiont to the mycobiont. The transfer of the cyanobiont is primarily assimilated by the mycobiont
fixed carbon occurs mostly in the light and in the form of in cephalodia, and the phycobiont receives fixed N via
glucose. Cyanobionts in tripartite lichens transfer little (<5 % the mycobiont. The mechanism of ammonia release by the
of CO2 fixed) or no fixed C to the mycobiont. Their primary role cyanobiont and its uptake by the mycobiont at the
seems to be the provision of fixed nitrogen. It would be inter- cyanobiont-mycobiont interface have not been investigated.
esting to know whether the cyanobionts in internal cephalodia However, diffusion of NH3 from heterocysts can occur in the
occurring deep in the medulla or on the undersurface of a lichen absence of ammonia assimilation by GS. Ammonia assimilation
thallus actually receive any fixed carbon from the phycobiont in the mycobiont occurs via GDH followed by aminotransfer-
(either directly or via the mycobiont). The glucose transferred to ases. In pulse-chase experiments, much of the 15N label accu-
the mycobiont is converted to mannitol, which serves as both a C mulated as alanine in the mycobiont of P. aphthosa cephalodia.
source and a physiological buffer. Mannitol production by Alanine could be the principal compound transferred to the rest
lichenized fungi could be an effective way of sequestering the of thallus (> Fig. 16.10).
fixed carbon since the other partners cannot use it. The mech-
anism underlying glucose transfer is not fully understood, but
the glucose is thought to originate from a glucan pool rather Ecological Significance
than directly from CO2 fixation. Altered cell wall synthesis may
lead to a diversion of sugars from cell wall synthesis to simple Lichens are ubiquitous, occurring in terrestrial as well as aquatic
release. Release of glucose declines sharply and stops soon after habitats from the equator to the highest latitudes, at sea level to
9,000-m altitude, and in the wettest to driest habitats. They are

excellent colonizers of nutrient-poor habitats (sand dunes,
rocks, forest floors, and the surfaces of other vegetation), form
dominant vegetation in tundra and arctic-alpine regions, and
contribute significantly to the N economy of these ecosystems.
Lichens are good bioindicators of air pollution.
Cyanobionts endow mycobionts with N and C autotrophy
and thereby widen their potential habitats. In a lichen thallus,
cyanobionts gain a safe habitat and protection from uncertainty
of fluctuating nutrient availability and climatic conditions in
nature.
Outlook
Many interesting aspects of the lichen symbioses remain to be . Fig. 16.11

elucidated. These include release and uptake of nutrients at the The Geosiphon-Nostoc symbiosis, isolated from a laboratory
cyanobiont-mycobiont interface, cyanobiont acquisition, and culture on natural substrate, incubated in liquid medium. The
regulatory mechanisms enforcing synchronized growth and dark bladders are about 1.5 mm in length. The insert shows
development of the partners. Furthermore, whether the struc- Geosiphon pyriformis spores, which have a diameter of about
tural-functional changes in symbionts are a result of endogenous 250 mm
regulation due to the symbiotic environment (e.g.,
microaerobiosis, restricted growth, and cell division) or whether fungus is used in its orthographically correct form, Geosiphon
they are directly caused by the mycobiont will need to be resolved. pyriformis (Schüßler 2002).
The Geosiphon pyriformis - Nostoc The Symbionts

Endocyanosis and Its Relationship to the
Arbuscular Mycorrhiza (AM) Geosiphon pyriformis
The Geosiphon pyriformis Symbiosis After its original description as Botrydium pyriforme, a siphonal
alga (Kützing 1849), Geosiphon pyriformis (as G. pyriforme) was
The fungus Geosiphon pyriformis (Kütz.) v. Wettstein (von recognized as a phycomycete (fungus with aseptate hyphae;
Wettstein 1915) forms the only known fungal endocyanosis Knapp 1933). Sixty years later, based on suggestions by Walter
(endocytobiotic association with cyanobacteria). The coeno- Gams, it was suggested that Geosiphon could be related to
cytic fungus forms unicellular, multinucleated cells (‘‘bladders’’) Glomus-like fungi (Mollenhauer 1992). Such fungi form the
of up to 2 mm in size (> Fig. 16.11), harboring endosymbiotic, arbuscular mycorrhiza (AM) symbiosis with land plants which
filamentous cyanobacteria of the genus Nostoc. There have been is extremely important ecologically and economically, therefore
only six reports describing this symbiosis in nature at locations verification of the phylogenetic relationship of Geosiphon would
ranging from eastern Germany to Austria. Probably, the symbi- make it conceivable that Geosiphon may also be capable of such
osis is geographically widespread in Central Europe but, due to association.
its small size, rarely reported. Presently, field sites around the Because the systematics of AM fungi in the last century was
small village of Bieber in the Spessart Mountains (Germany) are based mainly on the characteristics of their spore structure,
the only known stable natural habitats worldwide (Mollenhauer morphological and ultrastructural criteria of Geosiphon spores
1992; Schüßler and Wolf 2005). were compared with those of some AM fungi (Schüßler et al.
The species name ‘‘Geosiphon pyriforme’’ was sometimes 1994). This indeed revealed similarities between G. pyriformis
used for the fungus as well as for the symbiosis because the latter and AM fungi like Diversispora epigaea BEG47 (at that time
was often regarded as a ‘‘phycomycetous lichen.’’ Nowadays named Glomus versiforme; see Schüßler et al. 2011). Final evi-
endosymbiotic associations are usually excluded from lichen dence showing that Geosiphon is closely related to AM fungi was
definitions (Hawksworth and Honegger 1994). Thus, the species based on small subunit ribosomal RNA (SSU rRNA) gene
name should be used for the fungus only, also because phyloge- sequences (Gehrig et al. 1996). The AM fungi, together with
netically the Geosiphon fungus belongs to the arbuscular mycor- Geosiphon, formed a distinct clade not closely related to any
rhiza (AM)-forming and related fungi, the Glomeromycota other group of the zygomycetes. Further sequence analyses
(> Fig. 16.12). Here, the association between the fungus and (Schüßler 1999; Redecker et al. 2000b) showed that Geosiphon
cyanobacteria is referred to as the Geosiphon-Nostoc symbiosis, is closely related to an AM fungus forming two different spore
or simply the Geosiphon symbiosis, and the species name of the morphs, at that time named Acaulospora gerdemannii.
. Fig. 16.12
Phylogenetic tree of AM fungi (Glomeromycota), including Geosiphon (Modified and updated from Schüßler et al. 2001; Schüßler and
Walker 2010, http://www.amf-phylogeny.com). Families (represented by red branches) and genera are given, except for Entrophospora
and Otospora, for which placement is yet unclear
Nowadays, the clade containing these lineages is defined as 80 % of all vascular plants studied (Brundrett 2009) and
the order Archaeosporales, which represents one of the basal moreover also by lower plants (Read et al. 2000; Schüßler 2000),
main phylogenetic lineages in the phylum containing the AM despite their lack of roots. Considering this huge number of
fungi and Geosiphon, the Glomeromycota (Schüßler et al. 2001). plants that form AM, it is obvious that the AM must be one of
In this order, the Geosiphonaceae clusters as sister to the the most important factors in land ecosystems (Smith and Read
Ambisporaceae, thus appearing to be more derived than the 2008).
Archaeosporaceae, which branch earlier. This means that
Geosiphon does not represent a sister lineage to the AM fungi,
as was sometimes wrongly suggested. It was the analysis of the Nostoc punctiforme
phylogeny of Geosiphon that eventually led to the erection of the
Glomeromycota, a widely accepted fungal phylum and, eventu- The endosymbiont in the Geosiphon symbiosis (> Fig. 16.13) is
ally to the phylogenetically based, revised classification of the N. punctiforme, which belongs to a clade of cyanobacteria
Glomeromycota (Schüßler and Walker 2010). containing many symbiosis-forming members. In laboratory
The Geosiphon-Nostoc symbiosis attracted interest from cultures (Schüßler and Wolf 2005), a strain that originally was
the field of AM research. The AM symbiosis is formed by isolated from the Geosiphon symbiosis was used (Mollenhauer
. Fig. 16.13
Endosymbiotic Nostoc cells (about 7 6 mm in size), within
a Geosiphon bladder. One heterocyst is in focus (arrowhead)
1992). However, various other strains of N. punctiforme from

other symbiotic systems (e.g., Anthoceros, Blasia, Gunnera) are . Fig. 16.14
also capable of forming symbiosis with G. pyriformis. In the field, Electron micrograph of a ‘‘bacteria like organism’’ (BLO) in
G. pyriformis was usually found together with Anthoceros, and Geosiphon pyriformis. BLOs have a diameter of about 0.5 mm and
the cyanobionts of G. pyriformis associate in symbioses with are not enclosed by a host membrane (arrow). The insert shows
Anthoceros and Blasia (Mollenhauer 1992). the plasma membrane of the BLO (arrowhead), as well as the thick
It has to be noted that Geosiphon harbors another prokary- murein sacculus. Recent studies show them to be Mycoplasma-
otic endosymbiont, the so-called BLOs (bacteria-like organisms; related, despite the Gram-positive appearance
> Fig. 16.14), which are not enclosed by a fungal host membrane
but live freely in the cytoplasm (Schüßler et al. 1996; Schüßler

2012). These endosymbiotic bacteria, and those living in most of characterization of symbiosis-related genes is facilitated by use
the AM fungi that were studied for their occurrence, have the of this symbiosis (e.g., Schüßler et al. 2006).
same typical ultrastructure. Because they are found in very
diverse branches of the Glomeromycota, they were considered
to be widespread, Gram-positive glomeromycotan symbionts Infection Process, Development, and Structure
(Schüßler et al. 1994). of the Symbiosis
The BLOs are indeed ancestral and typical endobacteria in
AM fungi. New findings regarding their phylogeny and occur- Infection Process
rence in very diverse AM fungal lineages (Naumann et al. 2010)
showed that the BLOs are related to the cell-wall-lacking Both symbiosis partners live in the upper layer and on the
Mollicutes. We now know that they are monophyletic and later- surface of humid soil, where they make contact. The interaction
ally transferred within the AM fungi for more than 450 million is considered to be specific for two reasons: (1) Only certain
years. Their phylogeny and biotrophic lifestyle are shared with Nostoc punctiforme strains can form this symbiosis. (2) For
the related mycoplasmas, despite the obvious difference of a successful interaction with the fungus, Nostoc has to be differ-
possessing a murein sacculus. entiated into a specific stage represented by an early immobile
The Geosiphon symbiosis is facultative for one of the partners stage of the cyanobacterial developmental cycle, the so-called
(Nostoc can be cultivated without the fungus) and obligate for primordium (Mollenhauer et al. 1996). The motile filaments
the other one (Geosiphon is obligatory symbiotic). It is conceiv- (hormogonia) and late primordial, as well as vegetative stages
able that the fungus is not restricted to the cyanobacteria as of Nostoc, are not recognized by the fungus. When contacting
symbiotic partner but also forms symbioses with land plants Nostoc, the tip of the fungal hypha bulges out and surrounds part
(see below). However, this assumption is still speculative. of a cyanobacterial filament, thus incorporating the Nostoc cells
Regardless, Geosiphon belongs to the Glomeromycota, and the (> Fig. 16.15). Usually, 5–15 Nostoc cells are taken up during this
Nostoc symbiosis bears functional and structural similarities to process, whereas the heterocysts are never incorporated but ‘‘cut
the AM. Thus, the Geosiphon-Nostoc symbiosis can play a role as off ’’ by the fungus (see below). These events are documented in
a model symbiosis (Schüßler 2012) for the AM, which is difficult a scientific film available in German and English (Mollenhauer
to investigate but extremely important. For example, the and Mollenhauer 1997).
. Fig. 16.16
. Fig. 16.15 Young Geosiphon bladders, 100–150 mm in size, formed on the
Confocal laser scanning microscopy (CLSM) projection of a short fungal mycelium 7–10 days after initial uptake of the
hypha branching from a main hypha (horizontally oriented, cyanobacteria (left). The irregular structures in the background
4–6 mm in diameter) and ‘‘bulging out’’ to enclose a part of are vegetatively growing Nostoc colonies. At the right, two mature
a Nostoc filament. The extracellular polysaccharides of Nostoc and bladders of about 1 mm length, together with a young bladder,
the outer layer of the fungal cell wall are labeled by the are shown
fluorescence-coupled lectin ConA (green). The Nostoc cells (red
autofluorescence, ~4 3 mm in size) that are taken up by the
fungal structure show strong deformations and irregular and Each bladder represents a polyenergid cell, coenocytic with the
reduced pigment fluorescence fungal mycelium, in which the symbiotic Nostoc cells divide
and become physiologically active. Laboratory culturing exper-
iments have shown that, as for AM, phosphate limitation
(1–2 mM) of the nutrient solution triggers the stable establish-
Development of the Symbiosis ment of the symbiosis. N limitation seems not to be a crucial
factor. The same situation is found in the natural habitat, so
Studies on the development of the Geosiphon-Nostoc symbiosis P limitation seems to be a driving factor for the establishment of
showed that a successful interaction depends on the appropriate this symbiosis.
developmental stage of the cyanobacterium (Mollenhauer et al. Within the first hours after incorporation into the fungal
1996; Wolf and Schüßler 2005). The life cycle of Nostoc starts cytoplasm, the Nostoc filaments become heavily deformed, and
from akinetes (spore-like resting stages) leading to vegetative some cells may die during this process. The photosynthetic
colonies. These colonies release motile trichomes (hormogonia) pigments degrade considerably (> Fig. 16.15; Mollenhauer
which are positively phototactic in dim light and negatively in et al. 1996; Schüßler and Wolf 2005). These alterations and
strong light. As a consequence, the hormogonia often congre- significant changes in ultrastructure suggest that during the
gate just below the soil surface where they spread and meet their initial state of endocytotic life, the incorporated cyanobacteria
symbiotic partners. They eventually undergo a transformation suffer severe stress. Within 2–3 days, the enclosed Nostoc cells
into an aseriate stage called a primordium. This then differenti- recover and begin to multiply and grow to reach as much as six
ates into so-called vegetative cells, which divide and form gelat- times the volume of free-living cells (Schüßler et al. 1996;
inous colonies (‘‘thalli’’). Only the very early primordial stage of Mollenhauer and Mollenhauer 1997). Under phosphate limita-
Nostoc can interact with the fungal partner to give rise to the tion, the endosymbiotic cyanobacteria divide much faster and
symbiotic consortium. form a much higher biomass compared with the free-living ones
The life cycle of the fungal partner starts from resting spores (unpublished). In the symbiosis, the Nostoc cells arrange in
formed in the upper soil layer. The spores (Schüßler et al. 1994) filaments in which heterocysts are formed with the same fre-
germinate by the outgrowth of a hypha (sometimes more than quency as in the filaments outside the bladders (if cultured
one), which branches to form a small mycelium of up to 2–3 cm under nitrogen limitation). Mature Geosiphon bladders can
in the soil. When a hyphal tip contacts a compatible early Nostoc then reach more than 2 mm in length and up to 6 months in
primordium, the fungal hypha bulges out just below the apex. lab cultures. They possess a turgor pressure of about 0.6 MPa
This bulging process is repeated several times so that eventually (=6 bar) (Schüßler et al. 1995).
the hyphal tip forms an irregularly shaped structure surrounding
a part of a Nostoc primordium. After this incorporation into the
fungal hypha, large amounts of cytoplasm stream into this Structure and Compartmentation of the
Nostoc-containing structure, which then starts swelling and Geosiphon Bladder
develops the fungal bladder (> Fig. 16.16).
Each individual incorporation event results in the formation The Geosiphon bladder is effectively a multikaryotic cell, coeno-
of a single pear-shaped aboveground bladder (Knapp 1933). cytic with the fungal mycelium in the soil. It shows a strong
M
BLO
vacuole
Nostoc heterocyst SM
symbiosome-
membrane
fungal plasma-
membrane
cell wall
hypha lipid
V
soil
CW
PM
. Fig. 16.17
Schematic representation of the compartmentation of the Geosiphon-Nostoc symbiosis (left). At the right, a magnification of the
peripheral part of a bladder is shown. The Nostoc cells are about 6 mm in diameter. Drawings are based on electron microscopical
observations. BLO bacteria-like organism, CW cell wall, M mitochondrion, N nucleus, NC, Nostoc cell, PM plasma membrane,
SM symbiosome membrane, V vacuole
polarity and has a photosynthetically active region in the apical wall permeability, but can be improved by using microwave
part of the cell exposed to light and air and a whitish-appearing acceleration.
storage region, containing many lipid droplets, in the basal part Preparation of the G. pyriformis spores for electron micros-
embedded in the soil surface. The center of the bladder is highly copy (Schüßler et al. 1994) is even more difficult. This problem,
vacuolated. Schematic drawings of the compartmentation of caused by the thick spore wall being only slowly permeable to
Geosiphon are shown in > Fig. 16.17. Ultrastructural observa- fixatives, also exists with other glomeromycotan species (Maia
tions show the G. pyriformis symbiosis as a system with very et al. 1993). Two main storage compounds occur inside the
close contact between the partners. In fact, it is a symbiotic spores: lipid droplets of different sizes and ‘‘structured granules’’
consortium of three organisms: (1) the fungus, supplying the that occupy about 25 % of the volume. The latter are discussed
consortium with inorganic nutrients like phosphate, trace ele- below with respect to element analysis. They show
ments, and water; (2) the cyanobacteria, supplying the consor- paracrystalline inclusions, as are also found in spores of some
tium with carbohydrates by photosynthesis and, at least under other glomeromycotan fungi. Small vacuoles are found in ger-
some conditions, nitrogen compounds by N2 fixation; (3) the minating spores and hyphae, often containing dark deposits.
‘‘bacteria-like organisms’’ (BLOs), which are Mollicutes-related These are similar to the deposits in AM fungi and probably
endobacteria, with yet unknown function. polyphosphate precipitates. The ultrastructure of the Geosiphon
Within the bladders, the cyanobacteria are located periph- symbiosis was first studied by Schnepf (1964), and this was the
erally in a single, cup-shaped (often invaginated) compartment, crucial investigation leading to the theory of the compartmen-
the symbiosome. The Nostoc cells divide and are physiologically tation of the eukaryotic cell. The space between the symbiosome
active as endosymbionts in this compartment. Within the cyto- membrane and the wall of the enclosed Nostoc cells is only
plasm of the fungus, glycogen granules exist as storage com- 30–40 nm wide and contains a layer of electron microscopically
pounds. No dictyosomes are found; microtubules can rarely be opaque and amorphous-appearing material which was origi-
observed. Fixation of the bladders during preparation for elec- nally assumed to be slime produced by the endosymbiont
tron microscopy is often inadequate, probably due to the low cell (Schnepf 1964). Later ultrastructural and confocal laser
. Fig. 16.18
The symbiotic interface and bidirectional nutrient flows in the Geosiphon symbiosis, in comparison with those in the arbuscular
mycorrhiza (AM) (From Schüßler et al. 2006)
scanning microscopical (CLSM) studies by means of affinity element requirements and general element composition. Con-
techniques revealed that this amorphous layer inside the sidering the fact that these fungi supply the majority of land
symbiosome contains chitin (Schüßler et al. 1996), confirmed plants with inorganic elements, studies on the element compo-
by labeling with wheat germ agglutinin (WGA)-gold conjugates. sition and transport processes are interesting topics. We have
Thus, the electron opaque layer within the symbiosome repre- used PIXE (proton induced X-ray emission) measurements to
sents a ‘‘rudimentary’’ fungal cell wall, showing that the obtain the first indications of the macro- and microelement
symbiosome membrane surrounding the Nostoc cells is homol- composition of the spores and symbiotic bladders. The element
ogous with a fungal plasma membrane. content of some subcellular compartments could be quantita-
Clear similarities exist between the fungal cell wall material tively measured and, by a differential approach, calculated.
present in the symbiosome space of the Geosiphon symbiosis and PIXE, combined with STIM (scanning transmission ion micros-
the thin arbuscular cell wall bordering the symbiotic AM fungus copy), allowed elemental concentrations to be absolutely quan-
from the colonized plant cell: Both are electron dense after OsO4 tified with a lateral resolution in the 1 mm range and with high
fixation, about 30–40 nm thick, and show the same amorphous accuracy and precision (Maetz et al. 1999a).
structure and appearance. In general, the ultrastructural appear- Studies on the G. pyriformis symbiotic bladders (Maetz et al.
ance of G. pyriformis is similar to that of AM fungi. Considering 1999b) showed that the fungal partner of the symbiosis while
also the phylogenetic position of G. pyriformis and the known or grown on a nutrient-poor solution (e.g., containing 1 mM
proposed nutrient flows between the symbiotic partners, it has phosphate) accumulated P in high concentrations (about
been suggested that the symbiotic interface in the AM and the 2 %), but not in the symbiosome. The P is probably stored as
Geosiphon symbiosis are homologous (Schüßler et al. 1996). polyphosphate in the vacuoles, as for AM (and many other)
The main difference between the symbioses is the relation of fungi. High amounts of Cl (about 2.5 %) and K (about 8 %),
macro- and microbiont. In the Geosiphon symbiosis, the photo- which appear to play major roles in osmoregulation of the
autotrophic partner (cyanobacterium) is the microsymbiont, fungus, are found (all values given here are related to dry weight,
whereas in the AM, it is the macrosymbiont (plant) ppm = mg/g DW). The symbiosome (including the
(> Figs. 16.17, > 16.18). cyanobacteria) contains only small amounts of these elements.
This is in line with a presumed high concentration of monova-
lent ions in the fungal vacuoles. The macroelements Mg, S, and
Element Composition and Distribution Ca and the microelements Fe, Mn, Cu, and Zn occur in concen-
trations comparable with those found in plants. The Se concen-
It is not yet known why AM fungi cannot be cultured axenically. tration is below 1 ppm. Mo is present within the symbiosome in
Also, there is little information available about their trace very low amounts, compared with the rest of the bladder,
symbiotic bladders is blocked at an early stage of development.

Yet other N. punctiforme strains are not incorporated at all.
Further evidence for a specific recognition process is the fact
that, among the various developmental stages of Nostoc, only the
early primordia, existing for 3–12 h during the life cycle, are
incorporated by the fungus. Not only is the physiological activity
of the primordia different from the other stages of the Nostoc life
cycle (Bilger et al. 1994) but so is the composition of the gelat-
inous envelope. When differentiating into ‘‘symbiosis-
compatible’’ primordia, a mannose-containing slime is pro-
. Fig. 16.19 duced by the cells, whereas other sugars within the extracellular
A ‘‘symbiosis network’’ between cyanobacteria, fungi and plants. glycoconjugates can be detected only in earlier or later stages of
Associations or interactions which are highlighted with white the life cycle (Schüßler et al. 1997). The heterocysts (specialized
arrows are hypothetical. The BLOs in glomeromycotan fungi play N2-fixing cells), differentiating at regular spacing along the
an unknown but probably important role in this network of filaments of the Nostoc primordia when grown under nitrogen
intimate associations limitation, always remain outside the fungal hypha during the
incorporation process (Mollenhauer et al. 1996). They are not
surrounded by a newly appearing mannose-containing
although Mo is a constituent of nitrogenase, required for N2 glycoconjugate (Schüßler et al. 1997), also indicating a specific
fixation of the cyanobacteria. Reasons for this might be that recognition of the early primordial cell surface by the fungus.
other Mo enzymes (e.g., nitrate reductase, sulfite oxidase) Thus, alterations of extracellular glycoconjugates could be
occur in sufficient amounts in the fungal cytoplasm or that involved in partner recognition. Some unpublished data further
Mo is located in the fungal vacuole. Mn and Ni, by contrast, indicate a lectin-mediated process.
are present in the symbiosome in much higher concentrations
than in the rest of the bladder. Much of the Mn (approximately
50 ppm, which is comparable to values found in plant leaves) is Host-Cyanobiont Interactions Post Infection
probably contained in the water-cleaving Mn protein of photo-
system II. Some may be from other enzymes, e.g., Mn- Morphological Modifications of the Symbiosis
superoxide dismutase (SOD). A likely candidate for enzymes Partners
containing 50 ppm Ni is cyanobacterial (or secreted fungal)
urease; other Ni-containing bacterial enzymes are Ni-SOD and The most obvious morphological change taking place after
NiFe-hydrogenases. partner recognition is the formation of the Geosiphon bladder.
Unpublished results on the element composition of the Mature bladders represent large cells, which are coenocytic with
Geosiphon spores show that the structured granules (SGs), the mycelium. They show a clear polarity, with the photosyn-
which are 4–6 mm in diameter, located each within a vesicle, thetically active symbiotic compartment (symbiosome) located
together occupy about 25 % of the spore volume and contain in the apical part of the bladder (> Figs. 16.16, > 16.17).
most of the total P, K, and S. The S concentration of the spore cell The symbiosome is derived from the invaginated plasma
wall is 0.25 %, probably because of high protein content, as membrane of the fungus and contains the cyanobacteria,
shown for an AM fungus (Bonfante and Grippiolo 1984). Com- which are morphologically modified by increasing in volume
pared with the bladders, Cl and K are concentrated within the about six- to eightfold compared with free-living vegetative cells.
spores in much lower amounts. This is probably caused by the high osmotic pressure inside the
bladders. In many plant symbioses, cyanobacteria are known to
increase in size (Bergman et al. 1992a; Grilli Caiola 1992),
Signal Exchange Between Host and probably as a reaction to the higher osmotic pressure of the
Cyanobacterium surrounding medium. High NaCl concentrations are also
known to cause an increase in volume of cyanobacteria
It is not known what triggers the recognition process and the (Erdmann and Schiewer 1984). For Geosiphon bladders, the
morphological changes during the establishment of the symbi- iso-osmolar concentration of sorbitol was measured with oil-
osis. Microscopical studies give no hints for any chemotactic or filled microcapillaries and determined to be 220–230 mM,
otherwise directed growth toward the respective symbiosis part- corresponding to a turgor pressure (P) of about 0.6 MPa
ner, but the symbiosis-compatible Nostoc stage can be synchro- (Schüßler et al. 1995).
nized (Schüßler and Wolf 2005) to study this in more detail. However, despite the increase in size, the Nostoc cells inside
Cells of particular strains of N. punctiforme can be incorporated the Geosiphon bladder have an almost normal ultrastructure.
by Geosiphon, resulting in the formation of functional symbio- They contain a high number of thylakoids and carboxysomes;
ses. For other strains, although incorporated, the formation of one alteration is that the outer membrane is hardly recognizable
electron microscopically. Heterocysts are formed with the same other cell wall types, is very small (Schüßler et al. 1995). Such
frequency as in free-living colonies, but their cell wall is thinner a pore size is too small for an efficient permeation by, e.g., hexose
in the symbiosis, possibly indicating a lower surrounding O2 molecules from the outside, but it allows permeation of inor-
concentration. ganic hydrated ions like phosphate. If the hyphal cell wall also
has such a small pore size, the fungus would not be capable of
saprobic acquisition of organic molecules such as glucose,
N2 and CO2 Fixation and Transfer sucrose, and larger amino acids. However, cell wall permeability
is a complex topic, and the thin hyphae formed by AM fungi
14
C-tracer studies have shown that the Geosiphon bladders fix known as ‘‘branching absorbing structures’’ might possess dif-
CO2 both in light and in darkness, whereas the rate of CO2 ferent cell wall permeability.
fixation in light is much higher (Kluge et al. 1991). In light, Because AM fungi obtain up to 20 % of the plant-fixed CO2,
largely phosphate esters, poly-glucans, free sugars (including putatively as monosaccharides, the study of glomeromycotan
trehalose and raffinose), some amino acids, and organic acids sugar transporters that could play a role in C transfer from
trap 14C. In darkness only malic and fumaric acids together with plants to AM fungi is important. Only one such
some amino acids appear as labeled products. High photosyn- glomeromycotan monosaccharide transporter had been charac-
thetic activity of the endosymbiotic Nostoc cells is also shown by terized, and this (GpMST1) was from the Geosiphon symbiosis
photosystem II chlorophyll-fluorescence kinetics (Bilger et al. (Schüßler et al. 2006). This putatively symbiosome membrane-
1994). The symbiotic Nostoc cells achieve much higher steady- located transporter was demonstrated also to transport sugars
state quantum yields and electron transport rates when com- potentially deriving from plant-cell-wall material (> Fig. 16.18).
pared with free-living Nostoc. The GpMST1 sequence moreover provided valuable data for the
The capability of N2 fixation is indicated by the occurrence isolation of homologues from other AM fungi and could even-
of heterocysts, and considerable nitrogenase activity is shown tually lead to the better understanding of C flows in the AM.
for the bladders (Kluge et al. 1992). In contrast to symbioses of
Nostoc with plants, where usually a great increase of the hetero-
cyst frequency indicates N2 fixation as the major role of the Ecological Importance
cyanobacteria, in Geosiphon the relative heterocyst number
does not change. Here, the major role of the endosymbiotic Why the Symbiosis Is Mutualistic
Nostoc is photosynthesis. However, matter exchange between
the partners is still poorly investigated, and it is even possible The fungus in the Geosiphon symbiosis belongs to the
that the second bacterial endosymbiont (BLO; > Fig. 16.14), Glomeromycota (> Fig. 16.12) and is, like these, obligatorily
which is typical for many glomeromycotan fungi, may contrib- symbiotic. It is not yet known why glomeromycotan fungi are not
ute to N2 fixation. capable of nonsymbiotic life. Possibly it will be feasible in the future
For the endosymbiotic Nostoc cyanobacteria, all inorganic to develop special culture methods for in vitro growth of AM fungi,
nutrients except N have to be provided by the fungus, as the including Geosiphon. Generally, in nature, glomeromycotan
cyanobacteria live intracellularly. As shown by electrophysiolog- fungi seem incapable of saprotrophic existence but are depen-
ical experiments (unpublished), inorganic ions (nitrate, chlo- dent on their symbiosis partners for C delivery. For Geosiphon
ride) and small organic molecules (e.g., glycine, cysteine) lead to bladders, it was shown that only small molecules can pass the cell
rapid, transient depolarization of the plasma membrane poten- wall (Schüßler et al. 1995). The narrow pores do not allow the
tial of the G. pyriformis bladders, indicating that these substances uptake of hexoses or disaccharides from the outside, but perme-
are actively taken up from the outside. By contrast, there are no ation of inorganic ions like phosphate can occur. This might
changes in membrane potential if hexoses and larger amino reflect the situation in nature. However, it is also very possible
acids are applied. In addition, metabolism of radioactively that the fine hyphae growing into the substrate show higher
labeled hexoses by the bladders cannot be detected after the permeability. In any case, by incorporating Nostoc, the fungus
usual incubation times. Low cell wall permeability was consid- obtains the required organic compounds.
ered the likely reason for the lack of uptake of monosaccharides. Nostoc also benefits from the cooperation with the fungal
This theory is supported by observations showing that the host, which probably facilitates the supply of the endosymbiont
presence of solutes with large molecule radii leads to irreversible with water, phosphate, and also CO2. It is interesting that all
cytorrhysis, i.e., collapse of the whole bladder including the cell inorganic nutrients, except N, have to be delivered by the fungus,
wall, whereas in the presence of small solutes, plasmolysis since the cyanobacteria live intracellularly. It should also be kept
occurred (or cytorrhysis was quickly reversed). This different in mind that the establishment of the Geosiphon symbiosis, as is
transport behavior is presumably due to the selective permeabil- usually true for AM, is strongly promoted by P limitation, which
ity of the bladder wall. is a severe stress for the photobiont. The endosymbiotic Nostoc
By systematically using solutes with known molecular radii, cells thus divide and grow faster and bigger than their free-living
it was shown that the limiting pore radius of the G. pyriformis sisters. Preliminary studies moreover show that the intracellular
bladder wall is approximately 0.5 nm, which, compared with cyanobacteria are protected against heavy metals, which
accumulate in the fungus (Scheloske et al. 2001). Therefore, as in representing a symbiotic association between
the AM, the photobiont seems to be protected against abiotic a glomeromycotan fungus and a photoautotrophic prokaryote,
stress factors in the Geosiphon symbiosis. may reflect such an ancestral partnership, and thus, indirectly
but substantially, supports the hypothesis regarding pre-
Embryophyta associations of AM fungal predecessors
Evolutionary Implications with Ecological (Pirozynski and Malloch 1975). It is very plausible to assume
Meaning that at the beginning of terrestrial plant life, also other associa-
tions between glomeromycotan fungi and photoautotrophic
Most vascular plant species form AM (Smith and Read 2008), organisms (like the ubiquitous cyanobacteria) existed. The pre-
including gametophytes and sporophytes of many ferns (Peter- sent knowledge regarding AM fungi and AM symbiosis evolu-
son et al. 1981) and Lycopodiaceae (Schmid and Oberwinkler tion was recently discussed and reviewed (Schüßler and Walker
1993). Also, except for mosses, all groups of bryophytes contain 2011).
species with AM associations (Ligrone 1988; Ligrone and Lopes In summary, glomeromycotan fungi may have adapted to
1989; Stahl 1949; Fonseca et al. 2009), indicating an early origin symbiotic life more than 500 MY ago. Without fossil support,
of the AM symbiosis. this is speculative, but G. pyriformis clearly confirms the ability
In fact, the AM fungi have an ancient fossil record. Many of of glomeromycotan fungi to form symbioses with even prokary-
the oldest and best preserved 400 MY old AM fungal fossils in otic photoautotrophic organisms. Therefore, cyanobacterial
association with plants are known from the Rhynie chert, radio- symbioses formed by glomeromycotan fungi could have been
metrically dated to the early Devonian (e.g., Remy et al. 1994; an ecologically important step for the colonization of the land
Dotzler et al. 2009). The oldest known fossils of what appear to habitat.
be AM fungal spores and hyphae are from 460 MY old Ordo- Arbuscular mycorrhizal fungi form symbioses with most
vician dolomite rock of Wisconsin (Redecker et al. 2000a), and it land plants, and individual AM fungi can be symbiotic with
was concluded that terrestrial AM fungi already existed at a time widely divergent photoautotrophs such as hornworts and vas-
when the land flora most likely consisted only of bryophyte-like cular plants (Schüßler 2000). The genetic base for these interac-
‘‘lower’’ plants. tions is highly conserved (Wang et al. 2010). One can speculate
From fossil cryptospore assemblages sharing characters with that there might even be very ancestral symbiotic mechanisms in
those of extant liverworts (found in what was eastern Gond- the AM that can be found in the Geosiphon-Nostoc symbiosis.
wana; Rubinstein et al. 2010), it is estimated that land plants are There are some very fundamental and conserved mechanisms of
more than 470 MY old (Early Middle Ordovician). The diversity plant-microorganism interactions present among the different
of these assemblages implies an earlier, perhaps even Cambrian, AM(like) associations. When conducting ecophysiological
origin of embryophytes. Early vascular plants already existed studies involving plants, it is important to consider that in
420 MY ago (Middle Silurian; Cai et al. 1996). A recent molec- nature the mycorrhizal fungal partners are the main facilitators
ular clock study (Smith et al. 2010) suggested an origin of land of nutrient uptake, rather than the plant roots alone. If, as is
plants around 477 MY, but this dating in fact refers to the split thought, the mechanisms of nutrient acquisition by land plants
between bryophytes and the remaining lineages, not the (pre- coevolved since their origin with the AM fungi, ecologically and
sumably earlier) origin of the land plant lineage itself. Therefore, economically important questions might be answered by using
a minimum age of 420 MY for the liverwort-vascular plant the Geosiphon symbiosis as a model.
divergence must be assumed, and bryophyte-like land plants
were already present 510–470 MY ago.
Altogether, these data provide support for the hypothesis A Network Between Fungi, Cyanobacteria, and
(Pirozynski and Malloch 1975) that AM fungi symbioses played Plants?
a crucial role in the colonization of the land by plants, evolving
from a partnership between two aquatic types of organisms, Against the above described evolutionary background, the inter-
algae, and ‘‘oomycetous’’ fungi (the authors recognized the esting question arises as to whether G. pyriformis itself can act as
difference between AM fungi and other ‘‘phycomycetes,’’ and a fungal partner to form AM. Unpublished results showed that
thus interpreted them as ‘‘oomycetes,’’ which are nowadays Geosiphon rDNA can be PCR amplified from plant roots and
known not to be fungi), as the initial step of land plant evolu- bryophytes growing in the natural habitat of Geosiphon. How-
tion. A mycotrophic lifestyle could have been essential for an ever, it cannot be completely ruled out that the sensitive nested
efficient supply of plants with water and nutrients from the soil PCR approach is detecting tiny amounts of DNA from externally
(Malloch et al. 1980; Marschner and Dell 1994). However, attached hyphae. Future studies at sites where the cyanobacteria
molecular clock estimates always date the origin of the AM symbiosis of Geosiphon never occurs will have to show whether
fungal lineage to be at least 50, possibly more than 200 MY Geosiphon is indeed forming an AM. If this were to be the case,
earlier than that of land plants. If this holds true, it implies a complex network of ecological importance may be imagined
that there were other types of associations formed by AM (> Fig. 16.19).
fungi before land plants existed, whether saprobically, parasiti- The molecular probes to screen for the occurrence of
cally, or already mutualistically. Geosiphon pyriformis, Geosiphon in the soil and plant roots have been developed.
If Geosiphon indeed forms AM with plants, a complex network Basile DV (1990) Morphological role of hydroxyproline containing proteins in
liverworts. In: Chopra RN, Bhatia SC (eds) Bryophyte development: physi-
of biotic interactions would exist in the natural habitat. Within
ology and biochemistry. CRC Press, Boca Raton, pp 225–243
such a network, symbiotic Nostoc could be exchanged between Baulina OI, Lobakova ES (2003a) Atypical cell forms overproducing extracellular
Geosiphon and bryophytes, and Geosiphon could simultaneously substances in populations of cycad cyanobionts. Microbiology 72:701–712
form endosymbiosis with Nostoc and AM with plants, thus, e.g., Baulina OI, Lobakova ES (2003b) Heterocysts with reduced cell walls in
transporting and delivering N2 fixed by the cyanobacteria to the populations of cycad cyanobionts. Microbiology 72:713–722
Bergman B (2002) The Nostoc-Gunnera symbiosis. In: Rai AN, Bergman B,
plants. Unfortunately, funding applications have been rejected,
Rasmussen U (eds) Cyanobacteria in symbiosis. Kluwer, Dordrecht,
with the consequence that laboratory cultures are no longer pp 207–232
available. Bergman B, Osborne B (2002) The Gunnera-Nostoc symbiosis. In: Osborne B (ed)
Commentaries on cyanobacterial symbioses. Royal Irish Academy, Dublin,
pp 35–39
References Bergman B, Rai AN (1989) The Nostoc-Nephroma symbiosis: localization, distri-
bution pattern and levels of key proteins involved in nitrogen and carbon
Adams DG (2000) Symbiotic interactions. In: Whitton B, Potts M (eds) Ecology metabolism of the cyanobiont. Physiol Plant 77:216–224
of cyanobacteria: their diversity in time and space. Kluwer, Dordrecht, Bergman B, Lindblad P, Pettersson A, Renström E, Tiberg E (1985) Immuno-gold
pp 523–561 localization of glutamine synthetase in a nitrogen-fixing cyanobacterium
Adams DG (2002a) Cyanobacteria in symbiosis with hornworts and liverworts. (Anabaena cylindrica). Planta 166:329–334
In: Rai AN, Bergman B, Rasmussen U (eds) Cyanobacteria in symbiosis. Bergman B, Lindblad P, Rai AN (1986) Nitrogenase in free-living and symbiotic
Kluwer, Dordrecht, pp 117–135 cyanobacteria: immunoelectron microscopic localization. FEMS Microbiol
Adams DG (2002b) The liverwort-cyanobacterial symbiosis. Biol Environ Proc Lett 35:75–78
R Ir Acad 102B:27–30 Bergman B, Rai AN, Johansson C, Söderbäck E (1992a) Cyanobacterial-plant
Adams DG, Duggan P (1999) Heterocyst and akinete differentiation in symbioses. Symbiosis 14:61–81
cyanobacteria: Tansley Review No. 107. New Phytol 144:3–33 Bergman B, Johansson C, Söderbäck E (1992b) The Nostoc-Gunnera symbiosis.
Adams DG, Duggan PS (2008) Cyanobacteria-bryophyte symbioses. J Exp Bot New Phytol 122:379–400
59:1047–1058 Bergman B, Matveyev A, Rasmussen U (1996) Chemical signalling in
Adams DG, Duggan PS (2012) Signalling in cyanobacteria-plant symbioses. cyanobacterial-plant symbioses. Trends Plant Sci 1:191–197
In: S. Perotto S, Baluska F (eds) Signaling and communication in plant Bergman B, Rasmussen U, Rai AN (2003) Cyanobacterial associations. In:
symbiosis. Springer, Heidelberg, pp 93–121 Elmerich C, Newton WE (eds) Associative nitrogen-fixing bacteria and
Adams DG, Duggan PS, Jackson O (2012) Cyanobacterial symbioses. In: Whitton cyanobacterial associations. Kluwer, Dordrecht
B (ed) Ecology of cyanobacteria II: their diversity in time and space. Springer, Bergman B, Rasmussen U, Rai AN (2007) Cyanobacterial associations. In:
Heidelberg, pp 593–647 Elmerich C, Newton WE (eds) Associative and endophytic nitrogen-
Adams DG, Bergman B, Nierzwicki-Bauer SA, Rai AN, Schussler A (2006) fixing bacteria and cyanobacterial associations. Kluwer, Dordrecht,
Cyanobacterial-plant symbioses. In: Dworkin M, Falkow S, Rosenberg E, pp 257–301
Schleifer K-H, Stackebrandt E (eds) The prokaryotes. A handbook on the Bergman B, Ran L, Adams DG (2008) Cyanobacterial-plant symbioses: signaling
biology of bacteria, vol 1, 3rd edn, Symbiotic associations, biotechnology, and development. In: Herrero A, Flores E (eds) The cyanobacteria: molecular
applied microbiology. Springer, New York, pp 331–363 biology, Genomics and evolution. Caister Academic Press, Norfolk,
Ahern CP, Staff IA (1994) Symbiosis in cycads: the origin and development pp 447–473
of coralloid roots in Macrozamia communis (Cycadaceae). Am J Bot Bhaya D (2004) Light matters: phototaxis and signal transduction in unicellular
81:1559–1570 cyanobacteria. Mol Microbiol 53:745–754
Ahmadjian V (1993) The lichen symbiosis. Wiley, New York Bilger W, Büdel B, Mollenhauer R, Mollenhauer D (1994) Photosynthetic activity
Anbudurai PR, Mor TS, Ohad I, Shestakov SV, Pakrasi HB (1994) The ctpA of two developmental stages of a Nostoc strain (Cyanobacteria) isolated from
gene encodes the C-terminal processing protease for the D1 protein of Geosiphon pyriforme (Mycota). J Phycol 30:225–230
the photosystem II reaction center complex. Proc Natl Acad Sci USA Black KG, Parsons R, Osborne BA (2002) Uptake and metabolism of glucose in
91:8082–8086 the Nostoc-Gunnera symbiosis. New Phytol 153:297–305
Aptroot A (1998) Aspects of the integration of the taxonomy of lichenized and Bonfante P, Grippiolo R (1984) Cytochemical and biochemical observations on
non-lichenized pyrenocarpoous ascomycetes. Lichenologist 30:501–514 the cell wall of the spore of Glomus epigaeum. Protoplasma 123:140–151
Aulfinger H, Braun-Howland EB, Kannaiyan S, Nierzwicki-Bauer SA (1991) Bonnett HT, Silvester WB (1981) Specificity in the Nostoc-Gunnera endosymbi-
Ultrastructural changes of the endosymbionts of Azolla microphylla osis. New Phytol 89:121–128
during megaspore germination and early plantlet development. Can J Bot Brasell HM, Davies SK, Mattay JP (1986) Nitrogen fixation associated with
69:2489–2496 bryophytes colonizing burnt sites in Southern Tasmania, Australia. J Bryol
Babic S (1996) Hormogonia formation and the establishment of symbiotic 14:139–149
associations between cyanobacteria and the bryophytes Blasia and Braun-Howland EB, Nierzwicki-Bauer SA (1990) Azolla-Anabaena symbiosis:
Phaeoceros. PhD thesis, University of Leeds, Leeds biochemistry, physiology, ultrastructure and molecular biology. In: Rai AN
Baker J, Entsch B, McKay DB (2003) The cyanobiont in an Azolla fern is neither (ed) CRC handbook of symbiotic cyanobacteria. CRC Press, Boca Raton,
Anabaena nor Nostoc. FEMS Microbiol Lett 229:43–47 pp 65–117
Banack SA, Cox PA (2003) Biomagnification of cycad neurotoxins in flying foxes: Brenner ED, Stevenson DW, Twigg RW (2003) Cycads: evolutionary innovations
implications for ALS-PDC in Guam. Neurology 61:387–389 and the role of plant-derived neurotoxins. Trends Plant Sci 8:446–452
Banack SA, Murch SJ, Cox PA (2006) Neurotoxic flying foxes as dietary items for Brundrett MC (2009) Mycorrhizal associations and other means of nutrition of
the Chamorro people, Marianas Islands. J Ethnopharmacol 106:97–104 vascular plants: understanding the global diversity of host plants by resolving
Banack SA, Johnson HE, Cheng R, Cox PA (2007) Production of the neurotoxin conflicting information and developing reliable means of diagnosis. Plant
BMAA by a marine cyanobacterium. Mar Drugs 5:180–196 Soil 320:37–77
Banack SA, Downing TG, Spácil Z, Purdie EL, Metcalf JS, Downing S, Esterhuizen Burrows LL (2005) Weapons of mass retraction. Mol Microbiol 57:878–888
M, Codd GA, Cox PA (2010) Distinguishing the cyanobacterial neurotoxin Bushnell T (1998) Molecular identification of endosymbionts present in the leaf
b-N-methylamino-L-alanine (BMAA) from its structural isomer 2, cavities of Azolla caroliniana and Azolla mexicana. PhD thesis, Rensselaer
4-diaminobutyric acid (2, 4-DAB). Toxicon 56:868–879 Polytechnic Institute, Troy, pp 1–222
Cai CY, Ouyang S, Wang Y, Fang ZJ, Rong JY, Geng LY, Li XX (1996) An Early Cohen MF, Sakihama Y, Takagi YC, Ichiba T, Yamasaki H (2002) Synergistic effect
Silurian vascular plant. Nature 379:592 of doexyanthocyanins from symbiotic fern Azolla spp. on hrmA gene induc-
Calvert HE, Peters GA (1981) The Azolla-Anabaena relationship. IX: Morpho- tion in the cyanobacterium Nostoc punctiforme. Mol Plant Microbe Interact
logical analysis of leaf cavity hair populations. New Phytol 89:327–335 15:875–882
Campbell EL, Meeks JC (1989) Characteristics of hormogonia formation by Costa J-L, Lindblad P (2002) Cyanobacteria in symbiosis with cycads. In: Rai AN,
symbiotic Nostoc spp. in response to the presence of Anthoceros punctatus Bergman B, Rasmussen U (eds) Cyanobacteria in symbiosis. Kluwer,
or its extracellular products. Appl Environ Microbiol 55:125–131 Dordrecht, pp 195–205
Campbell EL, Meeks JC (1992) Evidence for plant-mediated regulation of nitro- Costa JL, Paulsrud P, Lindblad P (1999) Cyanobiont diversity within coralloid
genase expression in the Anthoceros-Nostoc symbiotic association. J Gen roots of selected cycad species. FEMS Microbiol Ecol 28:85–91
Microbiol 138:473–480 Costa J-L, Paulsrud P, Rikkinen J, Lindblad P (2001) Genetic diversity of Nostoc
Campbell EL, Hagen KD, Cohen MF, Summers ML, Meeks JC (1996) The devR symbionts endophytically associated with two bryophyte species. Appl
gene product is characteristic of receivers of two-component regulatory Environ Microbiol 67:4393–4396
systems and is essential for heterocyst development in the filamentous Costa JL, Romero EM, Lindblad P (2004) Sequence based data supports a single
cyanobacterium Nostoc sp. strain ATCC 29133. J Bacteriol 178:2037–2043 Nostoc strain in individual coralloid roots of cycads. FEMS Microbiol Ecol
Campbell EL, Brahamsha B, Meeks JC (1998) Mutation of an alternative sigma 49:481–487
factor in the cyanobacterium Nostoc punctiforme results in increased infec- Cox PA, Banack SA, Murch SJ (2003) Biomagnification of cyanobacterial neuro-
tion of its symbiotic plant partner, Anthoceros punctatus. J Bacteriol toxins and neurodegenerative disease among the Chamorro people of Guam.
180:4938–4941 Proc Natl Acad Sci USA 100:13380–13383
Campbell EL, Wong FCY, Meeks JC (2003) DNA binding properties of the HrmR Cox PA, Banack SA, Murch SJ, Rasmussen U, Tien G, Bidigare RR, Metcalf JS,
protein of Nostoc punctiforme responsible for transcriptional regulation Morrison LF, Codd GA, Bergman B (2005) Diverse taxa of cyanobacteria
of genes involved in the differentiation of hormogonia. Mol Microbiol produce beta-N-methylamino-L-alanine, a neurotoxic amino acid. Proc Natl
47:573–582 Acad Sci USA 102:5074–5078
Campbell EL, Summers ML, Christman H, Martin ME, Meeks JC (2007) Dalton DA, Chatfield JM (1985) A new nitrogen-fixing cyanophyte-hepatic
Global gene expression patterns of Nostoc punctiforme in steady state association: Nostoc and Porella. Am J Bot 72:781–784
dinitrogen-grown heterocyst-containing cultures, and at single time de Roissart P, Jucqued C, Watetkeyn L, Berghmans P, Van Hove C (1994) First
points during the differentiation of akinetes and hormogonia. J Bacteriol evidence for the cutinic nature of the envelope at the interface of Azolla and
189:5247–5256 its endophytes. In: Hegazi NA (ed) Nitrogen fixation with non-legumes: the
Campbell EL, Christman H, Meeks JC (2008) DNA microarray comparisons of sixth international symposium. The American University in Cairo Press,
plant factor- and nitrogen deprivation-induced hormogonia reveal decision- Cairo, pp 133–138
making transcriptional regulation patterns in Nostoc punctiforme. J Bacteriol DeLuca TH, Zackrisson O, Nilsson MC, Sellstedt A (2002) Quantifying nitrogen-
190:7382–7391 fixation in feather moss carpets of boreal forests. Nature 419:917–920
Canini A, Caiola MG, Mascini M (1990) Ammonium content, nitrogenase DeLuca TH, Zackrisson O, Gentili F, Sellstedt A, Nilsson MC (2007) Ecosystem
activity and heterocyst frequency within the leaf cavities of Azolla filiculoides controls on nitrogen fixation in boreal feather moss communities. Oecologia
Lam. FEMS Microbiol Lett 71:205–210 152:121–130
Carrapico F (1991) Are bacteria the 3rd partner of the Azolla-Anabaena symbi- DeLuca TH, Zackrisson O, Gundale MJ, Nilsson M-C (2008) Ecosystem feed-
osis? Plant Soil 137:157–160 backs and nitrogen fixation in boreal forests. Science 320:1181
Carrapico F, Tavares R (1989) New data on the Azolla-Anabaena symbiosis II: Deusch O, Landan G, Roettger M, Gruenheit N, Kowallik KVet al (2008) Genes of
cytochemical and immunocytochemical aspects. In: Skinner FA, Boddey cyanobacterial origin in plant nuclear genomes point to a heterocyst forming
RM, Fendrik I (eds) Nitrogen fixation with non-legumes. Kluwer, plastid ancestor. Mol Biol Evol 25:748–761
Dordrecht, pp 95–100 Do VC, Watanabe I, Zimmerman WJ, Lumpkin TA, de Waha Gaillonville T
Castillo-Guevara C, Rico-Gray V (2003) The role of macrozamin and cycasin (1989) Sexual hybridization among Azolla species. Can J Bot 67:3482–3485
in cycads (Cycadales) as antiherbivore defenses. J Torrey Bot Soc 130: Dotzler N, Walker C, Krings M, Hass H, Kerp H, Taylor TN, Agerer R (2009)
206–217 Acaulosporoid glomeromycotan spores with a germination shield from the
Chang DCN, Grobbelaar N, Coetzee J (1988) SEM observations on 400-million-year-old Rhynie chert. Mycol Prog 8:9–18
cyanobacteria-infected cycad coralloid roots. S Afr J Bot 54:491–495 Duckett JG, Prasad AKSK, Davies DA, Walker S (1977) A cytological analysis of
Chapman KE, Duggan PS, Billington NA, Adams DG (2008) Mutation at differ- the Nostoc-bryophyte relationship. New Phytol 79:349–362
ent sites in the Nostoc punctiforme cyaC gene, encoding the multiple-domain Duckett JG, Burch J, Fletcher PW, Matcham HW, Read DJ, Russell AJ,
enzyme adenylate cyclase, results in different levels of infection of the host Pressel S (2004) In vitro cultivation of bryophytes: a review of practicalities,
plant Blasia pusilla. J Bacteriol 190:1843–1847 problems, progress and promise. J Bryol 26:3–20
Chaw S-M, Walters TW, Chang C-C, Hu S-H, Chen S-H (2005) A phylogeny of Duggan PS, Gottardello P, Adams DG (2007) Molecular analysis of genes in
cycads (Cycadales) inferred from chloroplast matK gene, trnK intron, and Nostoc punctiforme involved in pilus biogenesis and plant infection.
nuclear rDNA ITS region. Mol Phylogenet Evol 37:214–234 J Bacteriol 189:4547–4551
Chiu WL, Peters GA, Levielle G, Still PC, Cousins S, Osborne B, Elhai J (2005) Ekman M, Tollbäck P, Klint J, Bergman B (2006) Protein expression profiles in an
Nitrogen deprivation stimulates symbiotic gland development in Gunnera endosymbiotic cyanobacterium revealed by a proteomic approach. Mol
manicata. Plant Physiol 139:224–230 Plant Microbe Interact 19:1251–1261
Cohen MF, Meeks JC (1997) A hormogonium regulating locus, hrmUA, of the Ekman M, Tollback P, Bergman B (2008) Proteomic analysis of the cyanobacte-
cyanobacterium Nostoc punctiforme strain ATCC 29133 and its response to rium of the Azolla symbiosis: identity, adaptation, and NifH modification.
an extract of a symbiotic plant partner Anthoceros punctatus. Mol Plant J Exp Bot 59:1023–1034
Microbe Interact 10:280–289 Elifio SL, Da Silva M, Iacomini M, Gorin PAJ (2000) A lectin from the lichenized
Cohen MF, Yamasaki H (2000) Flavonoid-induced expression of a symbiosis- basidiomycete Dictyonema glabratum. New Phytol 148:327–334
related gene in the cyanobacterium Nostoc punctiforme. J Bacteriol Enderlin CS, Meeks JC (1983) Pure culture and reconstitution of the Anthoceros-
182:4644–4646 Nostoc symbiotic association. Planta 158:157–165
Cohen MF, Wallis JG, Campbell EL, Meeks JC (1994) Transposon mutagenesis of Erdmann N, Schiewer U (1984) Cell size changes as indicator of salt resistance of
Nostoc sp. strain ATCC 29133, a filamentous cyanobacterium with multiple blue green algae. Arch Hydrobiol Suppl (Algol Stud) 67:431–439
cellular differentiation alternatives. Microbiology 140:3233–3240 Esterhuizen M, Downing T (2008) b-N-methylamino-L-alanine (BMAA)
Cohen MF, Meeks JC, Cai Y, Wolk CP (1998) Transposon mutagenesis of hetero- in novel South African cyanobacterial isolates. Ecotoxicol Environ Saf
cyst-forming filamentous cyanobacteria. Methods Enzymol 297:3–17 71:309–313
Evrard C, Van Hove C (2004) Taxonomy of the American Azolla species Hill DJ (1975) The pattern of development of Anabaena in the Azolla-Anabaena
(Azollaceae): a critical review. Syst Geogr Plant 74:301–318 symbiosis. Planta 122:178–184
Fiedler G, Muro-Pastor A, Flores E, Maldener I (2001) NtcA-dependent expres- Hill DJ (1977) The role of Anabaena in the Azolla-Anabaena symbiosis.
sion of the devBCA operon, encoding a heterocyst-specific ATP-binding New Phytol 78:611–616
cassette transporter in Anabaena spp. J Bacteriol 183:3795–3799 Hill DJ (1989) The control of cell cycle in microbial symbionts. New Phytol
Fisher RF, Long SR (1992) Rhizobium-plant signal exchange. Nature 357:655–660 112:175–184
Flores E, Herrero A (2005) Nitrogen assimilation and nitrogen control in Hill MP, Oberholzer IG (2002) Laboratory host range testing of the flea beetle,
cyanobacteria. Biochem Soc Trans 33:164–167 Pseudolampsis guttata (Leconte) (Coleoptera: Chrysomelidae), a potential
Fonseca HMAC, Ferreira JIL, Berbara RLL, Zatorre NP (2009) Dominance of natural enemy for red water fern, Azolla filiculoides Lamarck (Pteridophyta:
Paris-type morphology on mycothallus of Lunularia cruciata colonised by Azollaceae) in South Africa. Coleopt Bull 56:79–83
Glomus proliferum. Braz J Microbiol 40:96–101 Houle D, Gauthier SB, Paquet S, Planas D, Warren A (2006) Identification of two
Forni C, Grilli Caiola M, Gentili S (1989) Bacteria in the Azolla-Anabaena genera of N2-fixing cyanobacteria growing on three feather moss species in
symbiosis. In: Skinner FA, Boddey RM, Fendrik I (eds) Nitrogen fixation boreal forests of Quebec, Canada. Can J Bot 84:1025–1029
with non-legumes. Kluwer, Dordrecht, pp 83–88 Jackson O, Taylor O, Adams DG, Knox JP (2012) Arabinogalactan proteins occur
Forni C, Gentili S, Van Hove C, Grilli Caiola M (1990) Isolation and character- in the free-living cyanobacterium genus Nostoc and in plant–Nostoc symbi-
ization of the bacteria living in the sporocarps of Azolla filiculoides Lam. Ann oses. Mol Plant-Microbe Interactions 25:1338–1349
Microbiol 40:235–243 Janes R (1998a) Growth and survival of Azolla filiculoides in Britain. I. Vegetative
Galun M (1988) Handbook of lichenology. CRC Press, Boca Raton reproduction. New Phytol 138:367–375
Gantar M, Kerby NW, Rowell P (1993) Colonization of wheat (Triticum vulgare L.) Janes R (1998b) Growth and survival of Azolla filiculoides in Britain. II. Sexual
by N2-fixing cyanobacteria. III: the role of a hormogonia-promoting factor. reproduction. New Phytol 138:377–384
New Phytol 124:505–513 Johansson C, Bergman B (1992) Early events during the establishment of
Gates JE, Fisher RW, Candler RA (1980) The occurrence of corynoform bacteria Gunnera/Nostoc symbiosis. Planta 188:403–413
in the leaf cavity of Azolla. Arch Microbiol 127:163–165 Johansson C, Bergman B (1994) Reconstruction of the symbiosis of Gunnera
Gebhardt JS, Nierzwicki-Bauer SA (1991) Identification of a common manicata Linden: cyanobacterial specificity. New Phytol 126:643–652
cyanobacterial symbiont associated with Azolla spp. through molecular Joseph CM, Meeks JC (1987) Regulation of expression of glutamine synthetase in
and morphological characterization of free-living and symbiotic a symbiotic Nostoc strain associated with Anthoceros punctatus. J Bacteriol
cyanobacteria. Appl Environ Microbiol 57:2141–2146 169:2471–2475
Gehrig H, Schüßler A, Kluge M (1996) Geosiphon pyriforme, a fungus forming Joubert L, Grobbelaar N, Coetzee J (1989) In situ studies of the ultrastructure
endocytobiosis with Nostoc (Cyanobacteria), is an ancestral member of the of the cyanobacteria in the coralloid roots of Encephalartos arenarius,
Glomales: evidence by SSU rRNA analysis. J Mol Evol 43:71–81 E. transvenosus and E. woodii (Cycadales). Phycologia 28:197–205
Gehringer MM, Pengelly JJL, Cuddy WS, Fieker C, Forster PI, Neilan BA (2010) Kaplan D, Peters GA (1988) Interaction of carbon metabolism in the Azolla-
Host selection of symbiotic cyanobacteria in 31 species of the Australian cycad Anabaena symbiosis. Symbiosis 6:53–68
genus: Macrozamia (Zamiaceae). Mol Plant Microbe Interact 23:811–822 Kardish N, Silberstein L, Fleminger G, Galun M (1991) Lectins from the
Gentili F, Nilsson MC, Zackrisson O, DeLuca TH, Sellstedt A (2005) Physiological lichen Nephroma laevigatum Ach: localization and function. Symbiosis
and molecular diversity of feather moss associative N2-fixing cyanobacteria. 11:47–62
J Exp Bot 56:3121–3127 Khamar HJ, Breathwaite EK, Prasse CE, Fraley ER, Secor CR, Chibane FL, Elhai J,
Golden JW, Yoon HS (2003) Heterocyst development in Anabaena. Curr Opin Chiu WL (2010) Multiple roles of soluble sugars in the establishment of
Microbiol 6:557–563 Gunnera-Nostoc endosymbiosis. Plant Physiol 154:1381–1389
Gorelova OA, Baulina OI, Shchelmanova AG, Korzhenevskaya TG, Gusev MV Kimura J, Nakano T (1990) Reconstitution of a Blasia-Nostoc symbiotic associ-
(1996) Heteromorphism of the cyanobacterium Nostoc sp., a microsymbiont ation under axenic conditions. Nova Hedwig 50:191–200
of the Blasia pusilla moss. Microbiology 65:719–726 Klint J, Ran L, Rasmussen U, Bergman B (2006) Identification of developmentally
Gratwicke B, Marshall BE (2001) The impact of Azolla filiculoides Lam. on animal regulated proteins in cyanobacterial hormogonia using a proteomic
biodiversity in streams in Zimbabwe. Afr J Ecol 39:216–218 approach. Symbiosis 41:87–95
Grilli Caiola M (1980) On the phycobionts of the cycad coralloid roots. Kluge M, Mollenhauer D, Mollenhauer R (1991) Photosynthetic carbon assimi-
New Phytol 85:537–544 lation in Geosiphon pyriforme (Kützing) F.v. Wettstein, an endosymbiotic
Grilli Caiola M (1992) Cyanobacteria in symbiosis with bryophytes and tracheo- association of fungus and cyanobacterium. Planta 185:311–315
phytes. Biopress, Bristol Kluge M, Mollenhauer D, Mollenhauer R, Kape R (1992) Geosiphon pyriforme, an
Grobbelaar N, Scott WE, Hattingh W, Marshall J (1987) The identification of the endosymbiotic consortium of a fungus and a cyanobacterium (Nostoc), fixes
coralloid endophytes of the southern African cycads and the ability of nitrogen. Bot Acta 105:343–344
the isolates to fix dinitrogen. S Afr J Bot 53:111–118 Knapp E (1933) Über Geosiphon pyriformis Fr.v. Wettst., eine intrazelluläre
Guevara R, Armesto JJ, Caru M (2002) Genetic diversity of Nostoc Pilz-Algen-Symbiose. Ber Dtsch Bot Ges 51:210–217
microsymbionts from Gunnera tinctoria revealed by PCR-STRR fingerprint- Knight CD, Adams DG (1996) A method for studying chemotaxis in nitrogen-
ing. Microb Ecol 44:127–136 fixing cyanobacterium-plant symbioses. Physiol Mol Plant Pathol 49:73–77
Gusev MV, Baulina OI, Gorelova OA, Lobakova ES, Korzhenevshaya TG (2002) Kobiler D, Cohen-Sharon A, Tel-Or E (1981) Recognition between the N2-fixing
Artificial cyanobacterium-plant symbioses. In: Rai AN, Bergman B, Rasmussen Anabaena and the water fern Azolla. FEBS Lett 133:157–160
U (eds) Cyanobacteria in symbiosis. Kluwer, Dordrecht, pp 253–312 Kobiler D, Cohen-Sharon A, Tel-Or E (1982) Lectins are involved in the recog-
Hawksworth D, Honegger R (1994) The lichen thallus: a symbiotic phenotype of nition between Anabaena and Azolla Israel. J Bot 31:324–328
nutritionally specialized fungi and its response to gall producers. In: Williams Komarek J, Anagnostidis K (1989) Trichormus azollae (Strasb.): modern
MAJ (ed) Plant galls: organisms, interactions, populations. Clarendon, approaches to the classification system of cyanophytes 4-nostocales. Arch
Oxford, The Systematics Association, Special Volume No. 4977–98 Hydrobiol Algol Stud 56:303–345
Herdman M, Rippka R (1988) Cellular differentiation: hormogonia and Krüger T, Mönch B, Oppenhäuser S, Luckas B (2010) LC-MS/MS determination
baeocytes. Methods Enzymol 167:232–242 of the isomeric neurotoxins BMAA (b-N-methylamino-L-alanine) and DAB
Herrero A, Muro-Pastor AM, Flores E (2001) Nitrogen control in cyanobacteria. (2,4-diaminobutyric acid) in cyanobacteria and seeds of Cycas revolute and
J Bacteriol 183:411–425 Lathyrus latifolius. Toxicon 55:547–557
Herrero A, Muro-Pastor AM, Valladares A, Flores E (2004) Cellular differentia- Kützing FT (1849) Species algarum. F.A. Brockhaus, Leipzig
tion and the NtcA transcription factor in filamentous cyanobacteria. FEMS Lamb JR, Tugendreich S, Hieter P (1995) Tetratrico peptide repeat interactions: to
Microbiol Rev 28:469–487 TPR or not to TPR? Trends Biochem Sci 20:257–259
Large MF, Braggins JE (1993) Spore morphology of New Zealand Azolla Maetz M, Przybylowicz WJ, Mesjasz-Przybylowicz J, Schüßler A, Traxel K (1999a)
filiculoides Lam. (Salviniaceae) New Zealand. J Bot 31:419–423 Low-dose nuclear microscopy as a necessity for accurate quantitative
Lechno-Yossef S (2002) Identification and characterization of bacteria associated microanalysis of biological samples. Nucl Instrum Methods Phys Res
with the water fern Azolla sp. PhD thesis, Rensselaer Polytechnic Institute, B 158:292–298
Troy, pp 1–132 Maetz M, Schüßler A, Wallianos A, Traxel K (1999b) Subcellular trace element
Lechno-Yossef S, Nierzwicki-Bauer SA (2002) Azolla-Anabaena symbiosis. In: distribution in Geosiphon pyriforme. Nucl Instrum Methods Phys Res B
Rai AN, Bergman B, Rasmussen U (eds) Cyanobacteria in symbiosis. 150:200–207
Kluwer, Dordrecht, pp 153–178 Maia LC, Kimbrough JW, Erdos G (1993) Problems with fixation and embedding
Lee KY, Joseph CM, Meeks JC (1988) Glutamine synthetase specific activity and of arbuscular mycorrhizal fungi (Glomales). Mycologia 85:323–330
protein concentration in symbiotic Anabaena associated with Azolla Malloch DW, Pirozynski KA, Raven PH (1980) Ecological and evolutionary
caroliniana. Antonie Van Leeuwenhoek 54:345–355 significance of mycorrhizal symbiosis in vascular plants (a review). Proc
Legaz M-E, Fontaniella B, Millanes A-M, Vicente C (2004) Secreted arginases Natl Acad Sci USA 77:2113–2118
from phylogenetically far-related species act as cross-recognition factors for Marler TE, Snyder LR, Shaw CA (2010) Cycas micronesica (Cycadales) plants
two different algal cells. Eur J Cell Biol 83:435–446 devoid of endophytic cyanobacteria increase in beta-methylamino-L-alanine.
Lehr H, Fleminger G, Galun M (1995) Lectin from the lichen Peltigera Toxicon 56:563–568
membranacea (Ach.) Nyl.: characterization and function. Symbiosis 18:1–13 Marschner H, Dell B (1994) Nutrient uptake in mycorrhizal symbiosis. Plant Soil
Lehr H, Galun M, Ott S, Jahns HM, Fleminger G (2000) Cephalodia of the lichen 159:89–102
Peltigera aphthosa (L.) Willd.: specific recognition of the compatible Martin W, Rujan T, Richly E, Hansen A, Cornelsen S, Lins T, Leister D, Stoebe B,
photobiont. Symbiosis 29:357–365 Hasegawa M, Penny D (2002) Evolutionary analysis of Arabidopsis,
Lejeune A, Cagauan A, Van Hove C (1999) Azolla research and development: cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thou-
recent trends and priorities. Symbiosis 27:333–351 sands of cyanobacterial genes in the nucleus. Proc Natl Acad Sci USA
Liaimer A, Bergman B (2003) Phytohormones in cyanobacteria: occurrence and 99:12246–12251
perspectives. In: Proceedings from plant-microbe interaction conference, St. Mattick JS (2002) Type IV pili and twitching motility. Annu Rev Microbiol
Petersburg 56:289–314
Liaimer A, Matveyev A, Bergman B (2001) Isolation of host plant induced cDNAs McConnachie AJ, Hill MP, Byrne MJ (2004) Field assessment of a frond-feeding
from Nostoc sp. PCC 9229 forming symbiosis with the angiosperm Gunnera weevil, a successful biological control agent of red waterfern, Azolla
spp. Symbiosis 31:293–307 filiculoides, in southern Africa. Biol Control 29:326–331
Ligrone R (1988) Ultrastructure of a fungal endophyte in Phaeoceros laevis (L.) Meeks JC (1988) Symbiotic associations. Methods Enzymol 167:113–121
Prosk. (Anthocerophyta). Bot Gaz 149:92–100 Meeks JC (1990) Cyanobacterial-bryophyte associations. In: Rai AN (ed)
Ligrone R, Lopes C (1989) Cytology and development of mycorrhiza-like infec- CRC handbook of symbiotic cyanobacteria. CRC Press, Boca Raton,
tion in the gametophyte of Conocephalum conicum (L.) Dum. pp 43–63
(Marchantiales, Hepatophyta). New Phytol 111:423–433 Meeks JC (1998) Symbiosis between nitrogen-fixing cyanobacteria and plants.
Lin C, Liu ZZ, Zheng DY, Tang LF, Watanabe I (1989) Re-establishment of Bioscience 48:266–276
symbiosis to Anabaena free Azolla. Sci China Ser B Chem 32:551–559 Meeks JC (2003) Symbiotic interactions between Nostoc punctiforme,
Lindblad P (1990) Nitrogen and carbon metabolism in coralloid roots of cycads. a multicellular cyanobacterium, and the hornwort Anthoceros punctatus.
Advances in cycad research I. Mem N Y Bot Gard 57:104–113 Symbiosis 35:55–71
Lindblad P (2009) Cyanobacteria in symbiosis with cycads. In: Pawlovski K (ed) Meeks JC (2009) Physiological adaptations in nitrogen-fixing Nostoc-plant sym-
Prokaryotic endosymbionts in plants, vol 8, Microbiology monographs. biotic associations. In: Pawlowski K (ed) Prokaryotic symbionts in plants,
Springer, Berlin/Heidelberg, pp 225–233 vol 8, Microbiol monograph. Springer, Berlin, pp 181–205
Lindblad P, Bergman B (1986) Glutamine synthetases: activity and localization in Meeks JC, Elhai J (2002) Regulation of cellular differentiation in filamentous
cyanobacteria of the cycads Cycas revolute and Zamia skinneri. Planta 169:1–7 cyanobacteria in free-living and plant-associated symbiotic growth states.
Lindblad P, Costa J-L (2002) The cyanobacterial-cycad symbiosis. Biol Environ Microbiol Mol Biol Rev 66:94–121
Proc R Ir Acad 102B:31–33 Meeks JC, Enderlin CS, Wycoff KL, Chapman JS, Joseph CM (1983) Assimilation
Lindblad P, Hällblom L, Bergman B (1985a) The cyanobacterium-Zamia symbi- of 13NH4+ by Anthoceros grown with and without symbiotic Nostoc. Planta
osis: C2H2 reduction and heterocyst frequency. Symbiosis 1:19–28 158:384–391
Lindblad P, Bergman B, Hofsten AV, Hallbom L, Nylund JE (1985b) The cyano- Meeks JC, Enderlin CS, Joesph CM, Chapman JS, Lollar MWL (1985a) Fixation
bacterium-Zamia symbiosis: an ultrastructural study. New Phytol of [13N]; N2 and transfer of fixed nitrogen in the Anthoceros-Nostoc symbiotic
101:707–716 association. Planta 164:406–414
Lindblad P, Rai AN, Bergman B (1987) The Cycas revoluta-Nostoc symbiosis: Meeks JC, Steinberg N, Joseph CM, Enderlin CS, Jorgensen PA, Peters GA
enzyme activities of nitrogen and carbon metabolism in the cyanobiont. (1985b) Assimilation of exogenous and dinitrogen-derived 13NH4+ by
J Gen Microbiol 133:1695–1699 Anabaena azollae separated from Azolla caroliniana Willd. Arch Microbiol
Lindblad P, Haselkorn R, Bergman B, Nierzwicki-Bauer SA (1989) Comparison of 142:229–233
DNA restriction fragment length polymorphisms of Nostoc strains in and Meeks JC, Campbell E, Hagen K, Hanson T, Hitzeman N, Wong F (1999)
from cycads. Arch Microbiol 152:20–24 Developmental alternatives of symbiotic Nostoc punctiforme in response to
Lindblad P, Atkins CA, Pate JS (1991) N2-fixation by freshly isolated Nostoc from its symbiotic partner Anthoceros punctatus. In: Peschek GA, Löffelhardt W,
coralloid roots of the cycad Macrozamia riedlei (Firsch. ex Gand.) Garnd. Schmetterer G (eds) The phototrophic prokaryotes. Kluwer/Plenum,
Plant Physiol 96:753–759 New York, pp 665–678
Lobakova ES, Orazova MK, Dobrovol’skaya TG (2003) Microbial complexes Meeks JC, Elhai J, Thiel T, Potts M, Larimer F, Lamerdin J, Predki P, Atlas R (2001)
occurring on the apogeotropic roots and in the rhizosphere of cycad plants. An overview of the genome of Nostoc punctiforme, a multicellular, symbiotic
Microbiology 72:628–633 cyanobacterium. Photosynth Res 70:85–106
Lobakova ES, Dubravina GA, Zagoskina NV (2004) Formation of phenolic com- Meeks JC, Campbell EL, Summers ML, Wong FC (2002) Cellular differentiation
pounds in apogeotrophic roots of cycad plants. Russ J Plant Physiol 51:486–493 in the cyanobacterium Nostoc punctiforme. Arch Microbiol 178:395–403
Lohtander K, Oksanen I, Rikkinen J (2002) A phylogenetic study of Nephroma Meindl D, Loos E (1990) Release of glucose by Nostoc species isolated from the
(lichen-forming Ascomycota). Mycol Res 106:777–787 lichen Peltigera horizontalis. Plant Sci 72:283–288
Lotti F, Giovanetti L, Margheri MC, Ventura S, Materassi R (1996) Diversity of Mellor RB, Gadd GM, Rowell P, Stewart WDP (1981) A phytohaemagglutinin
DNA methylation pattern and total DNA restriction pattern in symbiotic from the Azolla-Anabaena symbiosis. Biochem Biophys Res Commun
Nostoc. World Microbiol Biotechnol 12:38–42 99:1348–1353
Milano V (2003) Identification of microbionts associated with the water fern Pate JS, Lindblad P, Atkins CA (1988) Pathways of assimilation and transfer of
Azolla. Masters thesis, Rensselaer Polytechnic Institute, Troy, pp 1–49 the fixed nitrogen in coralloid roots of cycad-Nostoc symbioses. Planta
Miura S, Yokoto A (2006) Isolation and characterization of cyanobacteria from 176:461–471
lichen. J Gen Appl Microbiol 52:365–375 Paulsrud P, Lindblad P (1998) Sequence variation of the tRNALeu intron as
Mollenhauer D (1992) Geosiphon pyriformis. In: Reisser W (ed) Algae and a marker for genetic diversity and specificity of symbiotic cyanobacteria in
symbiosis: plants, animals, fungi, viruses, interactions explored. Biopress, some lichens. Appl Environ Microbiol 64:310–315
Bristol, pp 339–351 Paulsrud P, Rikkinen J, Lindblad P (1998) Cyanobiont specificity in some
Mollenhauer D, Mollenhauer R (1997) Endosymbiosis between Nostoc and Nostoc-containing lichens and in a Peltigera aphthosa photosymbiodeme.
Geosiphon pyriforme. Institut für den Wissenschaftlichen Film (IWF), New Phytol 139:517–528
Göttingen, Film-no. C1955 Paulsrud P, Rikkinen J, Lindblad P (2000) Spatial patterns of photobiont diversity
Mollenhauer D, Mollenhauer R, Kluge M (1996) Studies on initiation and in some Nostoc-containing lichens. New Phytol 146:291–299
development of the partner association in Geosiphon pyriforme (Kütz.) Paulsrud P, Rikkinen J, Lindblad P (2001) Field experiments on cyanobacterial
v. Wettstein, a unique endocytobiotic system of a fungus (Glomales) specificity in Peltigera aphthosa. New Phytol 152:117–123
and the cyanobacterium Nostoc punctiforme (Kütz.) Hariot. Protoplasma Pennell RI (1992) Cell surface arabinogalactan proteins, arabinogalactans
193:3–9 and plant development. In: Collow JA, Green JR (eds) Perspectives in
Muro-Pastor AM, Reyes JC, Florencio FJ (2005) Ammonium assimilation in plant cell recognition. Cambridge University Press, Cambridge, UK,
cyanobacteria. Photosynth Res 83:135–150 pp 105–121
Nash TH (1996) Lichen biology. Cambridge University Press, Cambridge, UK Perkins SK, Peters GA (1993) The Azolla-Anabaena symbiosis: endophyte conti-
Nathanielsz CP, Staff IA (1975) A mode of entry of blue-green algae into the nuity in the Azolla life-cycle is facilitated by epidermal trichomes I:
apogeotrophic roots of Macrozamia communis. Am J Bot 62:232–235 partitioning of the endophytic Anabaena into developing sporocarps.
Naumann M, Schüßler A, Bonfante P (2010) The obligate endobacteria of New Phytol 123:53–64
arbuscular mycorrhizal fungi are ancient heritable components related to Perraju BTVV, Rai AN, Kumar AR, Singh HN (1986) Cycas circinalis-Anabaena
the Mollicutes. ISME J 4:862–871 cycadeae symbiosis: photosynthesis and the enzymes of nitrogen and hydro-
Newton JW, Herman AI (1979) Isolation of cyanobacteria from the aquatic fern, gen metabolism in symbiotic and cultured Anabaena cycadeae. Symbiosis
Azolla. Arch Microbiol 120:161–165 1:239–251
Nierzwicki-Bauer SA, Aulfinger H (1991) Occurrence and ultrastructural char- Peters GA, Calvert HE (1983) The Azolla-Anabaena azollae symbiosis. A contin-
acterization of bacteria in association with and isolated from Azolla uum of interaction strategies. In: Goff LJ (ed) Algal symbiosis. Cambridge
caroliniana. Appl Environ Microbiol 57:3629–3636 University Press, Cambridge, UK, pp 109–145
Nierzwicki-Bauer SA, Aulfinger H, Braun-Howland EB (1989) Ultrastructural Peters GA, Perkins SK (1993) The Azolla-Anabaena symbiosis: endophyte conti-
characterization of an inner envelope that confines Azolla endosymbionts to nuity in the Azolla life-cycle is facilitated by epidermal trichomes II:
the leaf cavity periphery. Can J Bot 67:2711–2719 reestablishment of the symbiosis following gametogenesis and embryogen-
Nilsson MC, Wardle DA (2005) Understory vegetation as a forest ecosystem esis. New Phytol 123:65–75
driver: evidence from the northern Swedish boreal forest. Front Ecol Environ Peters GA, Kaplan D, Meeks JC, Buzby KM, Marsh BH, Corbin JL (1985) Aspects
3:421–428 of nitrogen and carbon interchange in the Azolla-Anabaena symbiosis. In:
Nilsson M, Bergman B, Rasmussen U (2000) Cyanobacterial diversity in geo- Ludden PW, Burris JE (eds) Nitrogen fixation and CO2 metabolism. Elsevier,
graphically related and distant host plants of the genus Gunnera. Arch New York, pp 213–222
Microbiol 173:97–102 Peterson RL, Howarth MJ, Whittier DP (1981) Interactions between a fungal
Nilsson M, Rasmussen U, Bergman B (2006) Cyanobacterial chemotaxis to endophyte and gametophyte cells in Psilotum nudum. Can J Bot 59:711–720
extracts of host and nonhost plants. FEMS Microbiol Ecol 55:382–390 Pirozynski KA, Malloch DW (1975) The origin of land plants: a matter of
Nudleman E, Kaiser D (2004) Pulling together with type IV pili. J Mol Microbiol mycotrophism. Biosystems 6:153–164
Biotechnol 7:52–62 Plazinski J (1990) The Azolla-Anabaena symbiosis. In: Gresshoff PM (ed)
Obukowicz M, Schaller M, Kennedy GS (1981) Ultrastructure and phenolic Molecular biology of symbiotic nitrogen fixation. CRC Press, Boca Raton,
histochemistry of the Cycas revoluta-Anabaena symbiosis. New Phytol pp 51–75
87:751–759 Plazinski J, Taylor R, Shaw W, Croft L, Rolfe BG, Gunning BES (1990a) Isolation
Orr J, Haselkorn R (1982) Regulation of glutamine synthetase activity of Agrobacterium sp. strain from the Azolla leaf cavity. FEMS Microbiol Lett
and synthesis in free-living and symbiotic Anabaena spp. J Bacteriol 70:55–59
152:626–635 Plazinski J, Zheng Q, Taylor R, Croft L, Rolfe BG, Gunning BES (1990b) DNA
Osborne B, Bergman B (2009) Why does Gunnera do it and other angiosperms probes show genetic variation in cyanobacterial symbionts of the Azolla fern
don’t? In: Pawlowski K (ed) Prokaryotic symbionts in plants, vol 8, Micro- and a closer relationship to free-living Nostoc strains than to free-living
biology monographs. Springer, Heidelberg, pp 207–224 Anabaena strains. Appl Environ Microbiol 56:1263–1270
Osborne BA, Sprent JI (2002) Ecology of Nostoc-Gunnera symbiosis. In: Rai AN, Plazinski J, Croft L, Taylor R, Zheng Q, Rolfe BG, Gunning BES (1991) Indigenous
Bergman B, Rasmussen U (eds) Cyanobacteria in symbiosis. Kluwer, plasmids in Anabaena azollae: their taxonomic distribution and existence of
Dordrecht, pp 207–232 regions of homology with symbiotic genes of rhizobium. Can J Microbiol
Osborne B, Doris F, Cullen A, McDonald R, Campbell G, Steer M (1991) Gunnera 37:171–181
tinctoria: an unusual invader. Bioscience 41:224–234 Polsky FI, Nunn PB, Bell EA (1972) Distribution and toxicity of a-amino-b-
Osborne BA, Cullen A, Jones PW, Campbell GJ (1992) Use of nitrogen by methylaminopropionic acid. Fed Proc 31:1473–1475
the Nostoc-Gunnera tinctoria (Molina) Mirbel symbiosis. New Phytol Rai AN (1990) Cyanobacterial-fungal symbioses: the cyanolichens. In: Rai AN
120:481–487 (ed) Handbook of symbiotic cyanobacteria. CRC Press, Boca Raton, pp 9–41
Ow MC, Gantar M, Elhai J (1999) Reconstitution of a cycad-cyanobacterial Rai AN (2002) Cyanolichens: nitrogen metabolism. In: Rai AN, Bergman B,
association. Symbiosis 27:125–134 Rasmussen U (eds) Cyanobacteria in symbiosis. Kluwer, Dordrecht,
Palmqvist K (2002) Cyanolichens: carbon metabolism. In: Rai AN, Bergman B, pp 97–115
Rasmussen U (eds) Cyanobacteria in symbiosis. Kluwer, Dordrecht, Rai AN, Rowell P, Stewart WDP (1980) NH4+ assimilation and nitrogenase
pp 73–96 regulation in the lichen Peltigera aphthosa Willd. New Phytol 85:545–555
Papaefthimiou D, Hrouzek P, Mugnai MA, Lukesova A, Turicchia S, Rasmussen Rai AN, Rowell P, Stewart WDP (1981a) Nitrogenase activity and dark CO2
U, Ventura S (2008) Differential patterns of evolution and distribution of the fixation in the lichen Peltigera aphthosa Willd. Planta 151:256–264
symbiotic behaviour in nostocacean cyanobacteria. Int J Syst Evol Microbiol Rai AN, Rowell P, Stewart WDP (1981b) 15N2 incorporation and metabolism in
58:553–564 the lichen Peltigera aphthosa Willd. Planta 152:544–552
Rai AN, Rowell P, Stewart WDP (1983a) Interactions between cyanobacterium Rubinstein CV, Gerrienne P, De La Puente GS, Astini RA, Steemans P (2010) Early
and fungus during 15N2-incorporation and metabolism in the lichen Middle Ordovician evidence for land plants in Argentina (eastern Gond-
Peltigera aphthosa Willd. Arch Microbiol 134:136–142 wana). New Phytol 188:365–369
Rai AN, Rowell P, Stewart WDP (1983b) Mycobiont-cyanobiont interactions Sacristan M, Millanes A-M, Legaz M-E, Vicente C (2006) A lichen lectin specif-
during dark nitrogen fixation by the lichen Peltigera aphthosa. Physiol ically binds to the alpha-1,4-polygalactoside moiety of urease located in the
Plant 57:285–290 cell wall of homologous algae. Plant Signal Behav 1:23–27
Rai AN, Borthakur M, Singh S, Bergman B (1989) Anthoceros-Nostoc symbiosis: Saunders RMK, Fowler K (1993) The supraspecific taxonomy and evolution of
immunoelectronmicroscopic localization of nitrogenase, glutamine synthe- the fern genus Azolla (Azollaceae). Plant Syst Evol 184:175–193
tase, phycoerythrin and ribulose-1,5-bisphosphate carboxylase/oxygenase in Schaede R (1951) Über die Blaualgensymbiose von Gunnera. Planta 39:154–170
the cyanobiont and the cultured (free-living) isolate Nostoc 7801. J Gen Scheloske S, Maetz M, Schüßler A (2001) Heavy metal uptake of Geosiphon
Microbiol 135:385–395 pyriforme. Nucl Instrum Methods Phys Res Sect B 181:659–663
Rai AN, Söderbäck E, Bergman B (2000) Tansley Review No. 116, cyanobacte- Schenk HEA (1992) Cyanobacterial symbioses. In: Balows A, Trüper HG,
rium-plant symbioses. New Phytol 147:449–481 Dworkin M, Harder W, Schleifer K-H (eds) The prokaryotes, vol 4. Springer,
Rai AN, Bergman B, Rasmussen U (eds) (2002) Cyanobacteria in symbiosis. Berlin, pp 3819–3854
Kluwer, Dordrecht Schmid E, Oberwinkler F (1993) Mycorrhiza-like interactions between the
Ran L, Larsson J, Vigil-Stenman T, Nylander J, Ininbergs K, Zheng W, Lapidus A, achlorophyllous gametophyte of Lycopodium clavatum L. and its fungal
Lowry S, Haselkorn R, Bergman B (2010) Genome erosion in a nitrogen- endophyte studied by light and electron microscopy. New Phytol 124:69–81
fixing vertically transmitted endosymbiotic multicellular cyanobacterium. Schnepf E (1964) Zur Feinstruktur von Geosiphon pyriforme. Arch Mikrobiol
PLoS One 5:e11486 49:112–131
Rasmussen U, Nilsson M (2002) Cyanobacterial diversity and specificity in plant Schüßler A (1999) Glomales SSU rRNA gene diversity. New Phytol 144:205–207
symbioses. In: Rai AN, Bergman B, Rasmussen U (eds) Cyanobacteria in Schüßler A (2000) Glomus claroideum forms an arbuscular mycorrhiza-like
symbiosis. Kluwer, Dordrecht, pp 313–328 symbiosis with the hornwort Anthoceros punctatus. Mycorrhiza 10:15–21
Rasmussen U, Svenning M (2001) Characterization by genotypic methods of Schüßler A (2002) Molecular phylogeny, taxonomy, and evolution of Geosiphon
symbiotic Nostoc strains isolated from five species of Gunnera. Arch pyriformis and arbuscular mycorrhizal fungi. Plant Soil 244:75–83
Microbiol 176:204–210 Schüßler A (2012) The Geosiphon–Nostoc endosymbiosis and its role as a model
Rasmussen U, Johansson C, Bergman B (1994) Early communication in the for arbuscular mycorrhiza research. In: Hock B (ed) The mycota IX—fungal
Gunnera-Nostoc symbiosis: plant-induced cell differentiation and protein associations, 2nd edn. Springer, Berlin/Heidelberg (in press)
synthesis in the cyanobacterium. Mol Plant Microbe Interact 7:696–702 Schüßler A, Walker C (2010) The Glomeromycota: a species list with new families
Rasmussen U, Johansson C, Renglin A, Petersson C, Bergman B (1996) A molec- and new genera published in libraries at The Royal Botanic Garden
ular characterization of the Gunnera-Nostoc symbiosis: comparison with Edinburgh, The Royal Botanic Garden Kew, Botanische Staatssammlung
Rhizobium-and Agrobacterium-plant interactions. New Phytol 133:391–398 Munich, and Oregon State University. Electronic version freely available
Read DJ, Duckett JG, Francis R, Ligrone R, Russel A (2000) Symbiotic fungal online at www.amf-phylogeny.com. Gloucester
associations in ‘‘lower’’ land plants. Philos Trans R Soc Lond B 355:815–831 Schüßler A, Walker C (2011) Evolution of the ‘plant-symbiotic’ fungal phylum,
Redecker D, Kodner R, Graham LE (2000a) Glomalean fungi from the Ordovi- Glomeromycota. In: Pöggeler S, Wöstemeyer J (eds) Evolution of fungi and
cian. Science 289:1920–1921 fungal-like organisms. Springer, Berlin Heidelberg, pp 163–185
Redecker D, Morton JB, Bruns TD (2000b) Ancestral lineages of arbuscular Schüßler A, Wolf E (2005) Geosiphon pyriformis—a glomeromycotan soil fungus
mycorrhizal fungi. Mol Phylogenet Evol 14:276–284 forming endosymbiosis with cyanobacteria. In: Declerck S, Strullu D-G,
Reinke J (1872) Ueber die anatomische Verhaltnisse einiger Arten von Gunnera L. Fortin A (eds) In vitro culture of mycorrhizas. Springer, Berlin/Heidelberg,
Nachrichten der. Konigliche Gesellschaft der Wissenschaften Universitat, pp 271–289
Gottingen, pp 100–108 Schüßler A, Mollenhauer D, Schnepf E, Kluge M (1994) Geosiphon pyriforme, an
Reinke J (1873) Untersuchungen über die Morphologie der Vegetationsorgane endosymbiotic association of fungus and cyanobacteria: the spore structure
von Gunnera. In: Reinke J (ed) Morphologische Abhandlungen. Verlag W. resembles that of arbuscular mycorrhizal (AM) fungi. Bot Acta 107:36–45
Engelman, Leipzig, pp 45–123 Schüßler A, Schnepf E, Mollenhauer D, Kluge M (1995) The fungal bladders of
Remy W, Taylor TN, Hass H, Kerp H (1994) Four hundred-million-year-old the endocyanosis Geosiphon pyriforme, a Glomus-related fungus: cell wall
vesicular arbuscular mycorrhizae. Proc Natl Acad Sci USA 91:11841–11843 permeability indicates a limiting pore radius of only 0.5 nm. Protoplasma
Renzaglia KS (1982) A comparative developmental investigation of the gameto- 185:131–139
phyte generation in the Metzgeriales (Hepatophyta). Bryophyt Bibl 24:1–238 Schüßler A, Bonfante P, Schnepf E, Mollenhauer D, Kluge M (1996) Character-
Renzaglia KS, Duff RJ, Nickrent DL, Garbary D (2000) Vegetative and reproduc- ization of the Geosiphon pyriforme symbiosome by affinity techniques: con-
tive innovations of early land plants: implications for a unified phylogeny. focal laser scanning microscopy (CLSM) and electron microscopy.
Trans R Soc Lond 355:769–793 Protoplasma 190:53–67
Rikkinen J (2002) Cyanolichens: an evolutionary overview. In: Rai AN, Bergman Schüßler A, Meyer T, Gehrig H, Kluge M (1997) Variations of lectin binding sites
B, Rasmussen U (eds) Cyanobacteria in symbiosis. Kluwer, Dordrecht, in extracellular glycoconjugates during the life cycle of Nostoc punctiforme,
pp 31–72 a potentially endosymbiotic cyanobacterium. Eur J Phycol 32:233–239
Rikkinen J, Oksanen I, Lohtander K (2002) Lichen guilds share related Schüßler A, Schwarzott D, Walker C (2001) A new fungal phylum, the
cyanobacterial symbionts. Science 297:357 Glomeromycota: phylogeny and evolution. Mycol Res 105:1413–1421
Rodgers GA, Stewart WDP (1974) Physiological interrelations of the blue-green Schüßler A, Martin H, Cohen D, Fitz M, Wipf D (2006) Characterization of
alga Nostoc with the liverworts Anthoceros and Blasia. Br Phycol J 9:223 a carbohydrate transporter from symbiotic glomeromycotan fungi. Nature
Rodgers GA, Stewart WDP (1977) The cyanophyte-hepatic symbiosis. I: 444:933–936
morphology and physiology. New Phytol 78:441–458 Schüßler A, Krüger M, Walker C (2011) Revealing natural relationships among
Rosén J, Hellenäs K (2008) Determination of the neurotoxin BMAA (b-N- arbuscular mycorrhizal fungi: culture line BEG47 represents Diversispora
methylamino-L-alanine) to cycad seed and cyanobacteria by LC-MS/MS epigaea, not Glomus versiforme. PLoS One 6:e23333
(liquid chromatography tandem mass spectrometry). Analyst Sergeeva E, Liaimer A, Bergman B (2002) Evidence for biosynthesis and release
133:1785–1789 of the phytohormone indole-3-acetic acid by cyanobacteria. Planta
Rowell P, Rai AN, Stewart WDP (1985) Studies on the nitrogen metabolism of the 215:229–238
lichen Peltigera aphthosa and Peltigera canina. In: Brown DH (ed) Lichen Serrano R, Carrapico F, Vidal R (1999) The presence of lectins in bacteria
physiology and cell biology. Plenum, New York, pp 145–160 associated with the Azolla-Anabaena symbiosis. Symbiosis 27:169–178
Shannon BT, Gates JE, McCowen SM (1993) DNA base composition of eubacteria Stocker-Wörgötter E (1995) Experimental cultivation of lichens and lichen sym-
isolated from 4 species of Azolla. Symbiosis 15:165–175 bionts. Can J Bot 73S:579–589
Silvester WB (1976) Endophyte adaptation in Gunnera-Nostoc symbiosis. In: Stocker-Wörgötter E, Turk R (1994) Artificial resynthesis of the
Nutman PS (ed) Symbiotic nitrogen fixation in plants. Cambridge photosymbiodeme Peltigera leucophlebia under laboratory conditions. Cryp-
University Press, Cambridge, UK, pp 521–541 togam Bot 4:300–308
Silvester WB, McNamara PJ (1976) The infection process and ultrastructure of Stucken K, John U, Cembella A, Murillo A, Soto-Liebe K et al (2010) The smallest
the Gunnera-Nostoc symbiosis. New Phytol 77:135–141 known genomes of multicellular and toxic cyanobacteria: comparison, min-
Silvester WB, Smith DR (1969) Nitrogen fixation by Gunnera-Nostoc symbiosis. imal gene sets for linked traits and the evolutionary implications. PLoS One
Nature 224:1231 5:e9235
Silvester WB, Parsons R, Watt PW (1996) Direct measurement of release and Svenning MM, Eriksson T, Rasmussen U (2005) Phylogeny of symbiotic
assimilation of ammonia in the Gunnera-Nostoc symbiosis. New Phytol cyanobacteria within the genus Nostoc based on 16S rDNA sequence ana-
132:617–625 lyses. Arch Microbiol 183:19–26
Singh P, Singh D (1997) Azolla-Anabaena symbiosis. In: Ddarwal K (ed) Tabita FR (1994) The biochemistry and molecular regulation of carbon dioxide
Biotechnological approaches in soil microorganisms for sustainable crop metabolism in cyanobacteria. In: Bryant DA (ed) The molecular biology of
production. Scientific Press, Jodhpur, pp 93–107 cyanobacteria. Kluwer, Dordrecht, pp 437–467
Smith DC, Douglas AE (1987) The biology of symbiosis. Edward Arnold, London Tandeau de Marsac N (1994) Differentiation of hormogonia and relationships
Smith SE, Read DJ (2008) Mycorrhizal symbiosis, 3rd edn. Academic, London with other biological processes. In: Bryant DA (ed) The molecular biology of
Smith SA, Beaulieu JM, Donoghue MJ (2010) An uncorrelated relaxed-clock cyanobacteria. Kluwer, Amsterdam, pp 825–842
analysis suggests an earlier origin for flowering plants. Proc Natl Acad Sci Tehler A, Farris JS, Lipscomb DL, Kallersjo M (2000) Phylogenetic analysis of the
107:5897–5902 fungi based on large rDNA data sets. Mycologia 92:459–474
Snyder LR, Marler TE (2011) Rethinking cycad metabolite research. Commun Tel-Or E, Sandovsky T, Kobiler D, Arad C, Weinberg R (1983) The unique
Integr Biol 4:86–88 symbiotic properties of Anabaena in the water fern Azolla. In: Papageorgiou
Söderbäck E (1992) Developmental patterns in the Nostoc-Gunnera symbiosis. GC, Packer L (eds) Prokaryotes: cell differentiation and function. Elsevier,
Doctoral thesis, Stockholm University, Stockholm New York, pp 303–314
Söderbäck E, Bergman B (1992) The Nostoc-Gunnera magellanica symbiosis: Thajuddin N, Muralitharan G, Sundaramoorthy M, Ramamoorthy R,
phycobiliproteins, carboxysomes and Rubisco in the microsymbiont. Physiol Ramachandran S, Akbarsha MA, Gunasekaran M (2010) Morphological
Plant 8:425–432 and genetic diversity of symbiotic cyanobacteria from cycads. J Basic
Söderbäck E, Bergman B (1993) The Nostoc-Gunnera symbiosis: carbon fixation Microbiol 50:254–265
and translocation. Physiol Plant 89:125–132 Towata EM (1985) Morphometric and cytochemical ultrastructural analyses of
Söderbäck E, Lindblad P, Bergman B (1990) Developmental patterns related to the Gunnera kaalensis/Nostoc symbiosis. Bot Gaz 146:293–301
nitrogen fixation in the Nostoc-Gunnera magellanica Lam symbiosis. Planta Uheda E, Kitoh S (1991) Electron microscopic observations of the envelopes of
182:355–362 isolated algal packets of Azolla. Can J Bot 69:1418–1419
Solheim B, Zielke M (2002) Associations between cyanobacteria and mosses. In: Uheda E, Silvester WB (2001) The role of papillae during the infection process in
Rai AN, Bergman B, Rasmussen U (eds) Cyanobacteria in symbiosis. Kluwer, the Gunnera-Nostoc symbiosis. Plant Cell Physiol 42:780–783
Dordrecht, pp 137–152 Usher K, Bergman B, Raven J (2007) Exploring cyanobacterial mutualisms. Annu
Solheim B, Wiggen H, Roberg S, Spaink HP (2004) Associations between arctic Rev Ecol Evol Syst 38:255–273
cyanobacteria and mosses. Symbiosis 37:169–187 Van Coppenolle B, Watanabe I, Van-Hove C, Second G, Huang N, McCouch SR
Sood A, Prasanna R, Prasanna BM, Singh PK (2008) Genetic diversity among and (1993) Genetic diversity and phylogeny analysis of Azolla based on DNA
within cultured cyanobionts of diverse species of Azolla. Folia Microbiol amplification by arbitrary primers. Genome 36:686–693
53:35–43 Van Coppenolle B, McCouch SR, Watanabe I, Huang N, Van Hove C (1995)
Spáčil Z, Eriksson J, Jonasson S, Rasmussen U, Ilag LL, Bergman B (2010) Genetic diversity and phylogeny analysis of Anabaena azollae based on RFLP
Analytical protocol for identification of BMAA and DAB in biological detected in Azolla-Anabaena azollae DNA complexes using nif gene probes.
samples. Analyst 135:127–132 Theor Appl Genet 91:589–597
Staff IA, Ahern CP (1993) Symbiosis in cycads with special reference to Van Hove C, Lejeune A (1996) Does Azolla have any future in agriculture? In:
Macrozamia communis. In: Stevenson DW, Norstog KJ (eds) The biology, Rahman M (ed) Biological nitrogen fixation associated with rice production.
structure and systematics of the Cycadales. Palm and Cycad Societies of Kluwer, Dordrecht, pp 83–94
Australia, Queensland, pp 200–210 Vega A, Bell EA (1967) a-Amino-b-methylaminopropionic acid, a new amino
Stahl M (1949) Die Mycorrhiza der Lebermoose mit besonderer Berücksichtigung acid from seeds of Cycas circinalis. Phytochemistry 6:759–762
der thallösen Formen. Planta 37:103–148 Vessey JK, Pawlowski K, Bergman B (2005) Root-based N2-fixing symbioses:
Steinberg NA, Meeks JC (1989) Photosynthetic CO2 fixation and ribulose legumes, actinorhizal plants, Parasponia sp and cycads. Plant Soil 274:51–78
bisphosphate carboxylase/oxygenase activity of Nostoc sp. strain UCD 7801 Veys P, Waterkeyn L, Lejeune A, Van Hove C (1999) The pore of the leaf cavity of
in symbiotic association with Anthoceros punctatus. J Bacteriol 171:6227–6233 Azolla: morphology, cytochemistry and possible functions. Symbiosis
Steinberg NA, Meeks JC (1991) Physiological sources of reductant for nitrogen 27:33–57
fixation activity in Nostoc sp. strain UCD 7801 in symbiotic association with Veys P, Lejeune A, Van Hove C (2000) The pore of the leaf cavity of Azolla:
Anthoceros punctatus. J Bacteriol 173:7324–7329 interspecific morphological differences and continuity between the cavity
Stevenson DW (1993) The Zamiaceae in Panama with comments on phytogeog- envelopes. Symbiosis 29:33–47
raphy and species relationships. Brittonia 45:1–16 Veys P, Lejeune A, Van Hove C (2002) The pore of the leaf cavity of Azolla: teat-
Stewart WDP, Rogers GA (1977) The cyanophyte-hepatic symbiosis. II: cell differentiation and cell wall projections. Protoplasma 219:31–42
nitrogen fixation and the interchange of nitrogen and carbon. New Phytol Villarreal JC, Renzaglia KS (2006) Structure and development of Nostoc strands in
78:459–471 Leiosporoceros dussii (Anthocerotophyta): a novel symbiosis in land plants.
Stewart WDP, Rowell P (1977) Modifications of nitrogen-fixing algae in lichen Am J Bot 93:693–705
symbiosis. Nature (London) 265:371–372 Von Wettstein F (1915) Geosiphon Fr. v. Wettst., eine neue, interessante Siphonee
Stewart WDP, Rowell P, Rai AN (1983) Cyanobacteria-eukaryotic plant symbio- sterr. Bot Z 65:145–156
ses. Ann Inst Pasteur Microbiol 134B:205–228 Wallace WH, Gates JE (1986) Identification of eubacteria isolated from leaf
Stock PA, Silvester WB (1994) Phloem transport of recently fixed nitrogen in the cavities of 4 species of the N-fixing Azolla fern as Arthrobacter Conn and
Gunnera-Nostoc symbiosis. New Phytol 126:259–266 Dimmick. Appl Environ Microbiol 52:425–429
Wang C-M, Ekman M, Bergman B (2004) Expression of cyanobacterial genes spectral confocal laser scanning microscopy and spectral unmixing. Plant
involved in heterocyst differentiation and dinitrogen fixation along Cell Environ 28:480–491
a plant symbiosis development profile. Mol Plant Microbe Interact 17(4): Wolk CP (2000) Heterocyst formation in Anabaena. In: Brun YV, Shimkets LJ
436–443 (eds) Prokaryotic development. ASM Press, Washington, DC, pp 83–104
Wang B, Yeun LH, Xue J-Y, Liu Y, Ane J-M, Qiu Y-L (2010) Presence of three Wong FCY, Meeks JC (2001) The hetF gene product is essential to heterocyst
mycorrhizal genes in the common ancestor of land plants suggests a key role of differentiation and affects HetR function in the cyanobacterium Nostoc
mycorrhizas in the colonization of land by plants. New Phytol 186:514–525 punctiforme. J Bacteriol 183:2654–2661
Wanntorp L, Wanntorp H-E, Oxelman B, Källersjö M (2001) Phylogeny of Wong FCY, Meeks JC (2002) Establishment of a functional symbiosis between the
Gunnera. Plant Syst Evol 226:85–107 cyanobacterium Nostoc punctiforme and the bryophyte Anthoceros punctatus
Watanabe I (1994) Genetic enhancement and Azolla collection-problems in requires genes involved in nitrogen control and initiation of heterocyst
applying Azolla-Anabaena symbiosis. In: Hegazi NA, Fayez M, Monib M differentiation. Microbiology 148:315–323
(eds) Nitrogen fixation with non-legumes. The American University in Cairo Wouters J, Janson S, Bergman B (2000) The effect of exogenous carbohydrates on
Press, Cairo, pp 437–450 nitrogen fixation and hetR expression in Nostoc PCC 9229 forming symbiosis
Watanabe I, Van Hove C (1996) Phylogenetic, molecular, and breeding aspects of with Gunnera. Symbiosis 28:63–76
Azolla-Anabaena symbiosis. In: Camus JM, Gibby M, Jones RJ (eds) Pteri- Yoshimura I, Yamamoto Y (1991) Development of Peltigera praetextata lichen
dology in perspective. Royal Botanic Gardens, Kew, pp 611–619 thalli in culture. Symbiosis 11:109–117
Watanabe I, Lapis-Tenorio MT, Ventura TS, Padre BC (1993) Sexual hybrids of Zhang C-C, Laurent S, Sakr S, Peng L, Bedu S (2006) Heterocyst differentiation
Azolla filiculoides with A. microphylla. Soil Sci Plant Nutr 39:669–676 and pattern formation in cyanobacteria: a chorus of signals. Mol Microbiol
Watts S (2000) Signalling in the Nostoc-plant symbiosis. PhD thesis, University of 59:367–375
Leeds, Leeds Zheng WW, Nilsson M, Bergman B, Rasmussen U (1999) Genetic diversity and
Watts SD, Knight CD, Adams DG (1999) Characterisation of plant exudates classification of cyanobacteria in different Azolla species by the use of PCR
inducing chemotaxis in nitrogen-fixing cyanobacteria. In: Peschek GA, fingerprinting. Theor Appl Genet 99:1187–1193
Löffelhardt W, Schmetterer G (eds) The phototrophic prokaryotes. Zheng WW, Song TY, Bao XD, Bergman B, Rasmussen U (2002) High
Kluwer/Plenum, New York, pp 679–684 cyanobacterial diversity in coralloid roots of cycads revealed by PCR finger-
Webb DT (1983a) Nodulation in light- and dark-grown Macrozamia communis L. printing. FEMS Microbiol Ecol 40:215–222
Johnson seedlings in sterile culture. Ann Bot 52:545–547 Zheng S-P, Chen B, Guan X, Zheng W (2008) Diversity analysis of endophytic
Webb DT (1983b) Developmental anatomy of light-induced root nodulation by bacteria within Azolla microphylla using PCR-DGGE and electron micros-
Zamia pumila L. seedlings in sterile culture. Am J Bot 70:1109–1117 copy. Chin J Agric Biotechnol 16:508–514
Wei WX, Jin GY, Zhang N, Chen J (1988) Studies of hybridization in Azolla. In: Zheng W, Bergman B, Chen B, Zheng S, Xiang G, Rasmussen U (2009) Cellular
Singh KH, Kramer KU (eds) Proceedings of the international symposium on responses in the cyanobacterial symbiont during its vertical transfer between
systematic pteriodology. China Sciientific and Technical Press, Beijing, plant generations in the Azolla microphylla symbiosis. New Phytol 181:53–61
pp 135–139 Zielke M, Ekker AS, Olsen RA, Spjelkavik S, Solheim B (2002) The influence of
West N, Adams DG (1997) Phenotypic and genotypic comparison of symbiotic abiotic factors on biological nitrogen fixation in different types of vegetation
and free-living cyanobacteria from a single field site. Appl Environ Microbiol in the High Arctic, Svalbard Arctic. Arctic Antarct Alpine Res 34:293–299
63:4479–4484 Zielke M, Solheim B, Spjelkavik S, Olsen RA (2005) Nitrogen fixation in the high
West NJ, Adams DG, Sisson PR, Freeman R, Hawkey PM (1999) Pyrolysis mass arctic: role of vegetation and environmental conditions. Arctic Antarct
spectrometry analysis of free-living and symbiotic cyanobacteria. Antonie Alpine Res 37:372–378
Van Leeuwenhoek 75:201–206 Zimmerman WJ, Watanabe I, Lumpkin TA (1991) The Anabaena-Azolla symbi-
Wolf E, Schüßler A (2005) Phycobiliprotein fluorescence of Nostoc punctiforme osis: diversity and relatedness of neotropical host taxa. Plant Soil
changes during the life cycle and chromatic adaptation: characterization by 137:167–170
17 Root and Stem Nodule Bacteria of
Legumes
Michael J. Sadowsky . Peter H. Graham∗ . Masayuki Sugawara
Department of Soil, Water, and Climate, University of Minnesota, Saint Paul, MN, USA
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401 leguminous plants (peas, beans, clover, etc.). Not all legumes
form nodules with rhizobia.
Phylogeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401 Within the nodules, rhizobia convert atmospheric
dinitrogen (N2) gas into ammonia. This fixed nitrogen (N) is
Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404 subsequently assimilated by the host and improves plant growth
and productivity. Approximately 300 million hectares (Mha) of
Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 legumes are grown worldwide, and they collectively fix about
60 Tg (6 107 metric tons) of N each year (Kinzig and Socolow
Strain Isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406 1994). Overall, N2 fixation supplies about 50 % of the N used in
Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406 agriculture, and because the fixed N is used directly by the host
Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406 plant without exogenous addition to the soil, the process is
Isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407 generally considered environmentally friendly (Vance 1998). Fix-
Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407 ation rates vary with plant species, length of the growing season,
presence of a suitable microsymbiont, and environmental con-
Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407 ditions, but rates commonly are in the range of 100–200 kg of N2
fixed ha 1 year 1 (Sparrow et al. 1995; Unkovich et al. 1997).
Cultivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408 Because of the practical benefits of nodulation and N2 fixation,
the rhizobia have been extensively studied, particularly with
Preservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408 respect to the genetic basis for their symbiotic interactions,
The Nodulation Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409 their host specificity, and their nitrogen-fixation potential. How-
ever, the rhizobia are also good saprophytes, with soil populations
Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409 of 103–104 rhizobia g 1 being common in soils previously used
Nodulation Genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409 for legume growth. Thus, the ecological attributes of these
Genotype-Specific Nodulation Genes . . . . . . . . . . . . . . . . . . . 410 organisms also have been studied extensively.
Signal Exchange and Induction of Nod Genes . . . . . . . . . . 411
Extracellular Nodulation Factors . . . . . . . . . . . . . . . . . . . . . . . . 411
Nitrogen Fixation Genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411 Phylogeny
Other Genes Involved in Symbiotic Nitrogen
Fixation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412 The taxonomy of the organisms producing root and stem nod-
ules on legumes is in a state of flux. Currently 92 species of
Genomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412 nodule bacteria in 12 genera have been defined (> Table 17.1).
Although this ever-changing taxonomy affects what the organ-
Ecology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413 isms are called and how they are distinguished, it has little
impact on their phylogenetic relationships. Small subunit
Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414 rRNA sequence analysis (SSU rRNA) supports division of
these organisms into three major groups (Rhizobium [including
Agrobacterium, Allorhizobium, Ensifer, and Mesorhizobium],
Introduction Bradyrhizobium, and Azorhizobium) and Devosia and
Methylobacterium within the a subclass of the Proteobacteria
The root and stem nodule bacteria of legumes (collectively called (Martinez-Romero and Caballero-Mellado 1996; Young and
the rhizobia) comprise a genetically diverse group of microor- Haukka 1996). More recently, however, several b-proteobacteria,
ganisms characterized by the ability to produce hypertrophies including Burkholderia, Ralstonia, and Cupriavidus, have been
(swellings) or nodules on the stems or roots of most, but not all, shown to nodulate some legumes (Chen et al. 2001, 2003, 2005;
∗
This chapter is dedicated to the memory of Peter H. Graham (1937–2009).
402 17 Root and Stem Nodule Bacteria of Legumes
. Table 17.1 . Table 17.1 (continued)

List of current, validly published species of root and stem nodule Species References
bacteria
M. opportunistum Nandasena et al. (2009)
Species References M. plurifarium de Lajudie et al. (1998b)
Alpha-proteobacteria M. robiniae Zhou et al. (2010)
Rhizobium Frank (1889) M. shangrilense Lu et al. (2009)
R. alamii Berge et al. (2009) M. septentrionale Gao et al. (2004)
R. alkalisoli Li et al. (2009) M. tarimense Han et al. (2008a)
R. cellulosilyticum Garcia-Fraile et al. (2007) M. temperatum Gao et al. (2004)
R. daejeonense Quan et al. (2005) M. tianshanense Chen et al. (1995), Jarvis et al. (1997)
R. etli Segovia et al. (1993) Ensifer (formerly Casida (1982), de Lajudie et al. (1994),
Sinorhizobium) Young (2003)
R. galegae Lindstrom (1989)
E. abri Ogasawara et al. (2003)
R. gallicum Amarger et al. (1997)
E. adhaerens Casida (1982)
R. giardinii Amarger et al. (1997)
E. arboris Nick et al. (1999)
R. hainanense Chen et al. (1997)
E. fredii Scholla and Elkan (1984), de Lajudie
R. huautlense Wang et al. (1998)
et al. (1994)
R. indigoferae Wei et al. (2002)
E. garamanticus Merabet et al. (2010)
R. leguminosarum Frank (1889)
E. indiaense Ogasawara et al. (2003)
R. loessense Wei et al. (2003)
E. kostiensis Nick et al. (1999)
R. lusitanum Valverde et al. (2006)
E. kummerowiae Wei et al. (2002)
R. mesosinicum Lin et al. (2009)
E. medicae Rome et al. (1996)
R. miluonense Gu et al. (2008)
E. meliloti Dangeard (1926), de Lajudie et al.
R. mongolense van Berkum et al. (1998) (1994)
R. multihospitium Han et al. (2008b) E. mexicanus Lloret et al. (2007)
R. oryzae Peng et al. (2008) E. numidicus Merabet et al. (2010)
R. phaseoli Ramirez-Bahena et al. (2008) E. saheli de Lajudie et al. (1994)
R. pisi Ramirez-Bahena et al. (2008) E. sojae Li et al. (2010)
R. tibeticum Hou et al. (2009) E. teranga de Lajudie et al. (1994)
R. sullae Squartini et al. (2002) Sinorhizobium Toledo et al. (2003)
R. tropici Martinez-Romero et al. (1991) americanuma
R. tubonense Zhang et al. (2011) Sinorhizobium morelensea Wang et al. (2002)
R. undicola de Lajudie et al. (1998a), Young et al. Bradyrhizobium Jordan (1982)
(2001) B. canariense Vinuesa et al. (2005)
R. vignae Ren et al. (2011) B. elkanii Kuykendall et al. (1992)
R. yanglingense Tan et al. (2001) B. iriomotense Islam et al. (2008)
Mesorhizobium Jarvis et al. (1997) B. japonicum Kirchner (1896), Jordan (1982)
M. albiziae Wang et al. (2007) B. jicama Ramirez-Bahena et al. (2009)
M. alhagi Chen et al. (2010) B. liaoningense Xu et al. (1995)
M. amorphae Wang et al. (1999) B. pachyrhizi Ramirez-Bahena et al. (2009)
M. australicum Nandasena et al. (2009) B. yuanmingense Yao et al. (2002)
M. camelthorni Chen et al. (2011) Azorhizobium Dreyfus et al. (1988)
M. caraganae Guan et al. (2008) A. caulinodans Dreyfus et al. (1988)
M. chacoense Velazquez et al. (2001) A. doebereinerae Moreira et al. (2006)
M. ciceri Nour et al. (1994), Jarvis et al. (1997) Methylobacterium Patt et al. (1976)
M. gobiense Han et al. (2008a) Met. nodulans Jourand et al. (2004)
M. huakuii Chen et al. (1991), Jarvis et al. (1997) Devosia Nakagawa et al. (1996)
M. loti Jarvis et al. (1982, 1997) D. neptuniae Rivas et al. (2003)
M. mediterraneum Nour et al. (1994), Jarvis et al. (1997) Ochrobactrum Holmes et al. (1988)
M. metallidurans Vidal et al. (2009) O. cytisi Zurdo-Piñeiro et al. (2007)
Root and Stem Nodule Bacteria of Legumes 17 403
Species References
O. lupini Trujillo et al. (2005)
Phyllobacterium Knosel (1962)
P. trifolii Valverde et al. (2005)
P. ifriqiyense Mantelin et al. (2006)
P. leguminum Mantelin et al. (2006)
Shinella An et al. (2006)
S. kummerowiae Lin et al. (2008)
Beta-proteobacteria
Burkholderia Yabuuchi et al. (1992)
Bu. caribensis Vandamme et al. (2002)
Bu. cepacia Rasolomampianina et al. (2005)
Bu. mimosarum Chen et al. (2006)
Bu. nodosa Chen et al. (2007)
Bu. phymatum Vandamme et al. (2002)
Bu. sabiae Chen et al. (2008)
Bu. tuberum Vandamme et al. (2002)
Cupriavidus Makkar and Casida (1987)
C. taiwanensis Chen et al. (2001)
a
Not formally renamed as Ensifer
. Fig. 17.1
Initiation of nodule formation on the roots of Phaseolus vulgaris
(L.) by Rhizobium etli 8 days after inoculation (Photo by permission
Moulin et al. 2001). Regardless of which taxonomic class these
of M. H. Chaverra)
microorganisms belong to, the root/stem nodule bacteria are
collectively referred to as the rhizobia.
Wang et al. (1998) showed that Bradyrhizobium and Caesalpinioideae to 97 % in the Papilionoideae. Because these
Azorhizobium are only distantly related to fast-growing Rhizo- groups of legumes differ in the frequency of nodulation (and
bium and their relatives. Also amalgamation of Rhizobium and because Rhizobium, Bradyrhizobium, and Azorhizobium are so
Agrobacterium has been proposed on a number of occasions different), it has been suggested that the ability to nodulate and
(Graham 1964; Heberlein et al. 1967; De Ley 1968; Sawada fix N2 could have arisen on more than one occasion (Parker
et al. 1993; Parker 1957), suggesting that the rhizobia may have 1968; Doyle 1994; Doyle and Doyle 1997). Doyle (1994)
evolved from plant pathogenic bacteria. Nonpathogenic suggested that nodulation arose on at least three previous occa-
Agrobacterium are well-known as nodule contaminants (Hofer sions, including in the genus Chamaecrista. Species within
1941; Graham 1976; de Lajudie et al. 1999) and often are con- Chamaecrista can be distinguished from the non-nodulated,
fused with the nodule-forming rhizobia. Relative to the large but closely related, Cassia on the basis of randomly amplified
number of species of Rhizobium that have been described, only polymorphic DNA (RAPD) analysis (Whitty et al. 1994). Within
a limited number of Bradyrhizobium and Azorhizobium species Chamaecrista, species differ in the retention and release of
have been distinguished, although nine Bradyrhizobium species rhizobia from infection threads during differentiation (Naisbitt
are now recognized (Risal et al. 2010). This is likely to change as et al. 1992). It seems unlikely that legumes as different as
additional tropical legume species are studied. Additional Phaseolus and Acacia could nodulate with both Rhizobium and
groups of bradyrhizobia have already been identified, but Bradyrhizobium (Lange 1961; Michiels et al. 1998).
not detailed phylogenetically (So et al. 1994; Graham et al. The rhizobia associated with a particular legume host can
1995). Moreover, links between Rhizobium and related root- show significant diversity (Pinero et al. 1988; Souza et al. 1994).
nodule bacteria (Phyllobacterium, Devosia, Methylobacterium, However, some caution in interpreting results from biodiversity
Brucella, and Bartonella) and between Bradyrhizobium and studies is advisable because a number of studies predate recent
Rhodopseudomonas, Blastobacter, and Afipia have been described phylogenetic advances and changes in rhizobial taxonomy and
but need additional study (> Fig. 17.1). could have included more than one species of rhizobia. As new
The ability to form nodules is restricted to a clade of plants legumes are commercialized and exploited, more studies exam-
including both legume and actinorhizal species. However, not all ining the extent of legume/microsymbiont biodiversity near the
legumes bear nodules. The percentage of plant species with legume’s center of origin and exploration of the consequences of
nodules increased from only 23 % in the more primitive founder effects are warranted.
Taxonomy
Rhizobia have traditionally been a difficult group to classify.

Early researchers considered all rhizobia part of a single species
that could nodulate any legume. Subsequently, each rhizobial
strain was shown to only nodulate certain specific legumes. This
led to the concept of cross-inoculation groups, with rhizobia
being distinguished according to the legumes each could
nodulate. Thus, rhizobia from alfalfa would generally nodulate
medic species and vice versa, but neither would nodulate clover.
Using this approach, more than 20 different cross-inoculation
groups were identified, and a number of these were raised to
species status within the genus Rhizobium (Fred et al. 1932).
Fred et al. (1932) stated, ‘‘It seems true that the ability of an
organism to infect certain plants and not others is as fixed and
definite as any phase of the physiology of the organism. . .we feel
justified in regarding it as the prime character in species differ-
entiation.’’ While host specificity is still important in the iden-
tification of rhizobia, it is often at odds with results from
numerical and phylogenetic studies (Graham 1964; De Ley and
Rassell 1965; Heberlein et al. 1967; Moffett and Colwell 1968).
The demonstration that the nodulation genes in Rhizobium are
plasmid borne (Nuti et al. 1979; Brewin et al. 1980), or located
on chromosomal symbiotic islands, and move between soil
microorganisms has further weakened infection-based taxo-
. Fig. 17.2
nomic analyses.
A Bradyrhizobium japonicum cell showing significant
The 1984 edition of (http://www.cme.msu.edu/bergeys/)
polyhydroxybutyrate accumulation (Photo by T. McDermott, used
Bergey’s Manual of Systematic Bacteriology divides the
with permission)
Rhizobiaceae into four groups, including three genera of nodule-
or gall-forming bacteria, Rhizobium, Bradyrhizobium, and
Agrobacterium (Jordan 1984). The reduced emphasis on host Ochrobactrum lupini (Trujillo et al. 2005), Ochrobactrum cytisi
range and the availability of several new phenotypic and phylo- (Zurdo-Piñeiro et al. 2007), and Shinella kummerowiae (Lin
genetic techniques have resulted in the proliferation of new et al. 2008), are members of the a-proteobacteria, whereas
species and genera of nodule bacteria. Excellent review article several beta-proteobacterial species, including Cupriavidus
on the taxonomy of rhizobia are provided by Sawada et al. (formerly Ralstonia; Chen et al. 2001) and Burkholderia (Moulin
(2003), Willems (2006), and Weir (2011) provides a great up- et al. 2001; Vandamme et al. 2002), also have the ability to form
to-date website (http://www.rhizobia.co.nz/taxonomy/rhizobia. nitrogen-fixing root nodules on some legumes (> Fig. 17.2).
html) on the current taxonomy of rhizobia. With new species of root-nodule bacteria now justified using
As of March 2011, 12 genera and 92 species of rhizobia were a polyphasic approach that includes both phenotypic, and phy-
recognized (Weir 2011). Most of these rhizobial species belong logenetic traits (Graham et al. 1991), and genomic analyses
to the a-proteobacteria and are members of the genera (MacLean et al. 2007), the further description of new species of
Allorhizobium, Azorhizobium, Rhizobium, Mesorhizobium, Ensifer rhizobia based solely on simple characteristics has become
(formerly Sinorhizobium), or Bradyrhizobium (Martinez- increasingly problematic. In the second edition of (http://www.
Romero and Caballero-Mellado 1996; Young and Haukka prokaryotes.com) The Prokaryotes, Elkan and Bunn (1994) listed
1996). Although the genus Sinorhizobium was described by phenotypic traits useful in the distinction of Rhizobium,
Chen et al. in 1988, some recent studies show that Sinorhizobium Azorhizobium, and Bradyrhizobium. To do so again is
and the genus Ensifer (Casida 1982) belong to a single taxon. a daunting and perhaps not a useful task because older species
Ensifer is the earlier heterotypic synonym (it was named first) descriptions may include more than one organism and because
and thus takes priority (Young 2003). This means that all differences in the tests applied and methods used can impact
Sinorhizobium spp. must be renamed as Ensifer spp. according results and their interpretation.
to the Bacteriological Code. This has not been universally agreed > Table 17.2 lists carbon source utilization differences for
upon there has been some discussion in the literature if Ensifer or many species of root and stem nodule bacteria in the
Sinorhizobium is correct (Young 2010). a-proteobacteria. It was compiled from a number of different
In addition to these divisions, recent research has shown that studies (Jarvis et al. 1997; de Lajudie et al. 1998b; Wang et al.
some root-nodule bacteria, including Devosia neptuniae (Rivas 1999) and will likely change as new species are identified. Addi-
et al. 2002), Methylobacterium nodulans (Sy et al. 2001), tional distinctive phenotypic differences are urgently needed.
. Table 17.2
Differences in the carbon compounds used for growth among genera of root-nodule bacteria in the a-proteobacteriaa
Genus of nodule bacteriaa

Carbon source Rhizobium Ensifer (Sinorhizobium) Mesorhizobium Allorhizobium Bradyrhizobium Azorhizobium
b
Adonitol + + + +/
D-Arabinose + + + +
L-Arabinose + + (+) + +
D-Cellobiose + + +
L-Fucose + +/ +/
Inositol + + +/ +
Gluconate + (+) + +
Lactose + + (+) +
L-ysine +/ ( )
DL-Malate (+) (+) +/ + (+) +
D-Maltose + + + +
D-Mannose + + (+) + +
Mannitol + (+) + + (+)
D-Mellibiose + + ( )
D-Raffinose + + +/
Ribose + + + +
L-Rhamnose + + + + (+)
Sucrose + + + ( ) +
Trehalose + + + ( ) (+)
D-Xylose + + + +
a
Includes data from Elkan and Bunn (1994), de Lajudie et al. (1994, 1998a), Rome et al. (1996), Jarvis et al. (1997), and Wang et al. (1999).
b
Symbols: + positive reaction, negative reaction, +/ discriminatory within the genus, (+) mainly positive reaction, ( ) mainly negative reaction
Analysis of rhizobial fatty acid methyl esters (FAME), using prairie is but one ecosystem in which plant diversity and pro-
gas chromatography (Jarvis and Tighe 1994; Jarvis et al. 1996), ductivity is controlled, in large part, by N availability (Collins
has been recommended as a relatively simple and inexpensive et al. 1998). Although they are thought to be solely soil sapro-
method for the identification of fast-growing rhizobia. Rhizobial phytes, rhizobia can also be found in aquatic systems associated
FAME profiles correctly identified nearly 95 % of almost 200 with water-growing leguminous plants. Owing to cultural and
strains evaluated by Jarvis and Tighe (1994) and Jarvis et al. agricultural practices, the migration of birds and animals, and
(1996). These studies only erred in identifying some Ensifer atmospheric deposition of soil particles, there are relatively few
fredii as E. meliloti and Rhizobium etli as R. leguminosarum, soils in the world that do not contain some rhizobia. Moreover,
and vice versa (Graham et al. 1999). Ballen and Graham rhizobia have been shown to exist in soils for a relatively long
(unpublished observations) have also shown that R. etli, time in the absence of a host plant (Bottomley 1992; Brunel et al.
R. gallicum, and strains from Dalea and Onobrychis overlap by 1988; Kucey and Hynes 1989; Sanginga et al. 1994; Slattery and
FAME analysis. Similarly, FAME profiles have been used to Coventry 1993; Weaver et al. 1972), indicating that they are also
distinguish slow-growing Bradyrhizobium japonicum and efficient autochthonous soil bacteria.
B. elkanii (Kuykendall et al. 1992; So et al. 1994; Graham et al. Rhizobia have been recognized as being important for the
1995); although in each case additional isolates were identified functioning of soil ecosystems for centuries (Fred et al. 1932).
that did not group with these species. Shortly, after legume root nodules were shown conclusively to
assimilate atmospheric N2 (Hellriegel and Wilfarth 1888),
Nodbe and Hiltner were granted a patent for the use of these
Habitat microorganisms as legume inoculants (Elkan and Bunn 1994).
This and subsequent farming and cultural practices have led to
Rhizobia, through their ability to fix N2 in symbiosis with the dissemination of rhizobia on a global basis.
legumes, play a central role in the N supply of most natural Rhizobia in soils may be introduced by application of com-
terrestrial and aquatic ecosystems. The American tall grass mercial inoculants or, as in many cases, be the normal flora
present as microsymbionts of an indigenous legume. Inoculants (Date 1982; Date and Halliday 1987; Somasegaran and Hoben
applied to seed, as recommended by their manufacturer, achieve 1994). Extensive collection and conservation is necessary
inoculation rates of 103–106 rhizobia seed 1 (Somasegaran because many isolates will prove to be ineffective in symbiosis
and Hoben 1994). This corresponds to application rates of up or host/strain interactions will be significant. For example, in the
to 8 1010 rhizobia ha 1 (Brockwell and Bottomley 1995). At case of Stylosanthes scabra, more than 1,000 isolates were evalu-
these rates, inoculant strains often dominate in nodulation in ated before a strain suitable for use in commercial legume
the first year of a newly introduced crop (Brockwell et al. 1982; inoculation was identified (Date 1997).
Gibson et al. 1976; Singleton and Tavares 1986). Moreover,
inoculant strains contribute to the rapid buildup of rhizobia in
the soil once nodules senesce and release large numbers of viable Collection
rhizobia into the soil system (McDermott et al. 1987; Sutton
1983). Several studies have documented that inoculant strains The collection of rhizobia is most commonly undertaken as part
dominate in nodules 5–15 years after initial inoculation (Brunel of a plant introduction program to supply suitable host germ-
et al. 1988; Diatloff 1977; Lindstrom et al. 1990). It should be plasm with the rhizobia they need for symbiosis (see The
noted, however, that not all introduced legumes receive inocu- National Plant Germplasm Collection System [http://www.ars-
lation, and in such situations, seed, soil, or aerial contamination grin.gov]). Ideally, the collection of nodules should coincide
will usually lead to some initial nodule formation. A buildup of with early season growth and well-watered conditions. However,
soil rhizobial populations typically occurs over a period of 4–5 where collection involves remote geographic regions, sample
years (Sadowsky and Graham 1998). Moreover, diverse rhizobial acquisition may be delayed until plant maturity when most
populations can develop in association with species that are not nodules may have senesced. Nodule collection may also be
initially indigenous to a particular region (Leung et al. 1994). limited where the plant species in question is endangered, and
Although it is thought that rhizobia in soil have a clonal popu- therefore, no plant harvest is permitted. In both of these cases,
lation structure, genetic recombination between groups of soil soil may be used as a source of rhizobia, using surface-sterilized
rhizobia may be contributing to diversity in soils (Demezas et al. seed of an appropriate host to ‘‘trap’’ nodule bacteria. As
1995; Sullivan et al. 1995). It has been demonstrated that soil discussed above, however, this may bias the type and efficacy
rhizobia can transfer plasmids (Jarvis et al. 1989; Kinkle and of rhizobia identified.
Schmidt 1991; Thomas et al. 1994; Young and Wexler 1988) and Collection of rhizobia also may be undertaken to study
chromosomal symbiotic genes (Sullivan et al. 1995). the biodiversity of indigenous microorganisms and
The rhizobia obtained from any given soil habitat are dras- macroorganisms, the success in nodulation, or the soil estab-
tically influenced by the common method of isolation. This lishment of bioengineered organisms. In some cases, the culture
usually involves the use of serial dilutions of soil and inoculation of rhizobia from nodules may be unnecessary because enough
on a trap host, followed by recovery from nodules (Somasegaran cell material may be present in soils or plant tissue to directly
and Hoben 1994). This procedure often underestimates the characterize nodule occupants using serological or genetic
numbers of rhizobia in the soil and biases diversity determina- methods (Sadowsky and Graham 1998). Somasegaran and
tions (Dye et al. 1995). Numerous studies have documented the Hoben (1994) list several culture collections of rhizobia
influence of a trap host on the recovery of particular groups of throughout the world. In addition, the USDA-ARS National
rhizobia from soils (Bottomley et al. 1994; Bromfield et al. 1995; Rhizobium Resource Collection (http://bldg6.arsusda.gov/
Brunel et al. 1996; Keatinge et al. 1995; Kumar Rao et al. 1982; pberkum/Public/collection/index.htm) provides a searchable
van Berkum et al. 1995). Selective culture media, when available, database of rhizobia grouped by legume host, and several culture
will most likely prove useful in determining the identity of collections that specialize in legume inoculants can be found at
natural populations of rhizobia in soil (Gault and the WFCC-MIRCEN World Data Centre for Microorganisms
Schwinghamer 1993; Tong and Sadowsky 1993). (http://www.wfcc.nig.ac.jp/datacenter.html).
Lastly, the legume host itself has been shown to strongly
influence the prevalence and type of rhizobia in soils (Bottomley
1992). How this occurs is not known, but it is thought to be due Sampling
to nonspecific, root-exudate-enhanced growth of rhizobia in the
rhizosphere, multiplication and release of rhizobia from the Sampling of plants and nodules should be done from
nodule, and selection by the trap host of particular groups of undisturbed locations and, where possible, from healthy plants.
rhizobia from mixed soil populations (Sadowsky and Graham Accurate site description and record keeping are essential. The
1998). number of nodules needed to be sampled can vary with the
reason for collection. Where the aim is to identify inoculant-
quality rhizobia, 15–20 nodules per plant, taken from the crown
Strain Isolation region of the host root system, are usually sufficient. In contrast,
if the goal is to evaluate strain biodiversity in soil, a large number
Several studies have detailed methods for the collection, sam- of nodules should be collected from as much of the root system
pling, isolation, authentication, and maintenance of rhizobia as possible. Ease of collection may vary; stoloniferous species
may have nodules on adventitious roots within 1–2 cm of the Rhizobium, Mesorhizobium, Sinorhizobium, and
surface (Date 1982), while nodules on tree species may be at Allorhizobium strains will generally produce moist, gummy
a great depth in the soil at some distance from the trunk of the colonies on YEM medium that are 4–6 mm in diameter after
tree. Collected nodules should be protected in Vacutainers or in 7 days incubation. On medium containing BTB, the colonies
vials containing a desiccant (e.g., silica gel) overlain by cotton and surrounding medium are yellow due to acid production by
wool (Somasegaran and Hoben 1994). the microorganisms. The slower-growing bradyrhizobia
produce smaller colonies, usually only 1–2 mm diameter after
7–10 days incubation, which are raised and mucoid. The colo-
Isolation nies and surrounding medium are blue in color on YEM
containing BTB. Most nodule isolates will produce white or
Successful isolation of rhizobia from nodules depends on the
cream-colored colonies, though some isolates produce melanin
quality of nodules recovered. When nodules have been stored
(Cubo et al. 1988), or in the case of bradyrhizobia, a rust red
dry over silica gel or CaCl2, they must first be allowed to imbibe
pigmentation in older colonies. Since some rhizobia have
(sterile) water fully before being surface sterilized. Rhizobia also
recently been shown to be members of b-proteobacteria, care
can be frequently recovered from nodules obtained from intact
must be taken in making any assessment of taxonomic identity
root system frozen at 20 C. Sodium hypochlorite (3 %),
based on morphological criteria.
hydrogen peroxide (3 %), and acidified mercuric chloride
(0.1 %) are all effective surface sterilants. The former is usually
preferred due to its low-cost, ready availability, and ease of
disposal. Surface sterilization procedures are described in detail Authentication
by Vincent (1970) and Somasegaran and Hoben (1994).
Yeast extract mannitol (YEM) medium is commonly used in Authentication of rhizobia usually involves completion of
the routine isolation and subculture of rhizobia. Many different Koch’s postulates with the host from which strains were origi-
formulations for this medium exist (Vincent 1970; Somasegaran nally isolated. Somasegaran and Hoben (1994) provide details of
and Hoben 1994). That used in our laboratory contains: this methodology. Inoculated seedlings produced from surface-
sterilized seed of a suitable legume host are typically grown in
Mannitol 10.0 g sterile low N plant growth medium or on seedling agar in large
MgSO4·7H2O 0.2 g test tubes, growth pouches, or in Leonard jars (Vincent 1970).
NaCl 0.1 g Plants are examined for nodulation after 25–30 days of incuba-
tion under lights. The presence of nodules on uninoculated
K2HPO4 0.5 g
control plants invalidates the experiment.
CaCl2·2H2O 0.2 g
FeCl3·6H2O 0.01 g
Yeast extract 1.0 g
Identification
Agar 20.0 g
Distilled water 1l
The identity of rhizobia or bradyrhizobia often requires
pH 6.7–7.0 a multiphasic approach using many of the techniques employed
in naming new genera, species, and strains of rhizobia (Graham
Sterilize by autoclaving 20 min at 103 H 103 Pa (15 lb/in2) et al. 1991). Members of the International Subcommittee on
pressure. Rhizobium and Agrobacterium, a subcommittee of the Interna-
Since many rhizobia produce copious quantities of extracel- tional Committee on Systematic Bacteriology of the Interna-
lular polysaccharides on this medium, which makes colony tional Union of Microbiological Societies, have recommended
purification difficult, many researchers prefer to use AG a minimal set of criteria for naming new species and genera of
(Sadowsky et al. 1987b) and TY media (Beringer et al. 1978) nodule bacteria (Graham et al. 1995). These criteria are also
for isolating slow- and fast-growing rhizobia from nodules. useful for identifying the genus and species status of unknown
The YEM medium may be amended with cycloheximide rhizobia isolated from nodules. In addition to biochemical,
(20 mg/ml) to reduce fungal contamination and bromothymol cultural, and symbiotic data, 16S rRNA (rDNA) sequencing
blue (BTB; 25 mg/ml) or Congo red (25 mg/ml) to facilitate (Young 1996; Young and Haukka 1996), DNA-DNA hybridiza-
identification of rhizobia. These should be filter-sterilized sepa- tion (Scholla and Elkan 1984), FAME (Graham et al. 1995), and
rately and added to autoclaved, molten YEM medium before multilocus enzyme electrophoresis (MLEE) (Strain et al. 1995)
plates are poured (Vincent 1970). Tong and Sadowsky (1993) data are of primary importance for identifying rhizobia isolated
described a selective medium specific for Bradyrhizobium, based from newly surveyed legumes.
primarily on the heavy metal tolerance of these organisms. Following strain authentication, it is often useful to geneti-
Media selective for fast-growing rhizobia have been described cally mark these isolates to facilitate identification in subsequent
(Barber 1979; Louvrier et al. 1995), but they have not proven ecological, genetic, or plant studies. This can be done using
generally effective. a variety of techniques. These include intrinsic resistance to
a series of different antibiotics (Josey et al. 1979) or the selection As stated earlier, most rhizobia grow well in YEM medium
of mutants resistant to high levels of antibiotics. In the latter (Vincent 1970), though most produce copious quantities of
case, selected mutants must be evaluated to show that the capsular- and exopolysaccharides in this medium, limiting its
acquisition of antibiotic resistance has not influenced nodula- use in biochemical and genetic studies. The bradyrhizobia grow
tion, N2 fixation or competitive abilities. fairly slowly in this medium, with generation times greater than
Strains can also be identified using strain or group-specific 6 h. Rhizobia and bradyrhizobia shift the pH of this medium;
antibodies (Sadowsky et al. 1987a; Schmidt et al. 1968). Anti- the rhizobia produce acid and the bradyrhizobia, alkaline
bodies, which are typically produced in rabbits to somatic by-products, from growth. Polysaccharide production can be
whole-cell antigens, are useful in strain identification because drastically reduced in fast-growing rhizobia by cultivation in
they do not require genetic modification of strains. Agglutina- TY medium (Beringer et al. 1978), containing (g/l) tryptone
tion, immunodiffusion, immunofluorescence, and ELISA tech- (5.0), yeast extract (3.0), and CaCl2·2H2O (0.87), pH 6.9. In
niques all have found wide acceptance in serological this medium, turbid cultures, up to 109 cells ml 1, can be
identification of rhizobia (Dudman 1977; Humphrey and obtained after overnight incubation at 28 C. The slow-growing
Vincent 1965; Kishinevsky and Jones 1987; Schmidt et al. bradyrhizobia do not grow in TY medium. A growth medium
1968). The fluorescent antibody technique is especially useful useful for near polysaccharide-free growth by bradyrhizobia is
because it allows for the direct in situ examination of rhizobia in AG medium (Sadowsky et al. 1987b). This medium, which pro-
soil and nodules (Bohlool and Schmidt 1973), using direct or motes rapid growth of B. japonicum and B. elkanii strains,
sandwich labeling procedures. contains (g/l) HEPES (0.13), MES (0.11), FeCl3·6H2O
Strains also can be genetically modified with (0.0067), MgSO4·7H2O (0.18), CaCl2·2H2O (0.013), Na2SO4
b-glucuronidase (GUS) (Wilson et al. 1995, 1999), green fluo- (0.25), NH4Cl (0.32), Na2HPO4 (0.125), arabinose (1.0), Na-
rescent protein (GFP) (Gage et al. 1996), and lux (Chabot et al. gluconate (1.0), and yeast extract (1.0), pH 6.9. Several defined,
1996) genes, and these strains have proven especially useful in minimal, media are also used for the growth of rhizobia
ecological studies. Again, however, it is essential that such genet- (Vincent 1970; Somasegaran and Hoben 1994), especially for
ically marked strains be plant tested before use in ecological, biochemical and molecular biological studies. We also use AG
symbiotic, or field studies. medium without arabinose, gluconate, and yeast extract as
DNA fingerprinting techniques have been used to identify a minimal medium for the cultivation of prototrophic rhizobia
and study biodiversity of rhizobial strains (Sadowsky 1994; and in genetic mating studies.
Versalovic et al. 1998; Sadowsky and Hur 1998; Demezas Rhizobia for legume inoculants can be grown in shake flasks
1998). Initially, DNA fingerprints of strains were generated or in fermentors (Somasegaran and Hoben 1994). Lorda and
following restriction enzyme digestion of total genomic DNA Balatii (1996) described a glycerol-based culture medium capa-
(Glynn et al. 1985; Demezas 1998). More recently, however, ble of producing approximately 1010 cells/ml, even in shake flask
restriction fragment polymorphism (RFLP) analysis techniques, culture. In contrast, Stephens and Rask (2000) suggest that
DNA hybridization probes, and DNA primers corresponding to carbon-limited media be used to produce legume inoculants,
repetitive elements, coupled to the polymerase chain reaction to condition rhizobia to the less favorable conditions found
(PCR) technique, have been used in strain identification, com- in soil.
petition, and biodiversity studies (de Bruijn 1992; Judd et al.
1993; Sadowsky 1994; Sadowsky et al. 1990; Wheatcroft and
Watson 1988). Preservation
Rhizobium strains in frequent use are usually maintained on

Cultivation YMA, TY, or AG agar slants in screw-capped test tubes stored
at 6–10 C. Longer-term storage is achieved by lyophilization
Rhizobia are relatively robust, ubiquitous, aerobic bacteria with with 10 % glycerol or 10 % sucrose and 5 % peptone as cryo-
the ability to utilize many different substrates (carbon [C] and protectants or by storage at 70 C in 15 % glycerol. Gherna
nitrogen [N] sources) for growth (Parke and Ornston 1984). (1994) details methodologies for lyophilization and storage at
Consequently, rhizobia can be cultivated on a large variety of 70 C. Changes in Rhizobium characteristics, especially sym-
complex and defined culture media. Only a limited number of biosis, with repeated growth on laboratory media have been
rhizobia grow on highly enriched media, such as nutrient broth reported (Herridge and Roughley 1975) and must be of concern.
or LB medium. Medium used in the cultivation of rhizobia For this reason, it is strongly suggested that repetitive
depends on the species of nodule bacteria, growth characteristics subculturing or colony restreaking not be done for the fast-
desired, and the method of cultivation. Most rhizobia are growing rhizobia. Since symbiosis-related genes are chromo-
mesophiles and can grow in shake cultures at 25–30 C. How- some-borne in the slow-growing bradyrhizobia, subculturing
ever, rhizobia isolated from legumes grown in the Canadian does not appear to be a problem with this group. Some inoculant
High Arctic grow well at 5 C (Prévost et al. 1987), and high companies maintain large numbers of ampoules of each Rhizo-
temperature tolerant strains have been isolated in Africa and bium strain in the freeze-dried state Rhizobium and routinely
Brazil. replace all working cultures at 3-month to 1-year intervals.
The Nodulation Process lateral roots as the first-formed nodules senesce. The number
of nodules produced on each legume host is tightly controlled
The nodulation process requires molecular communication by the host and rhizobial genotype, the efficiency of the symbi-
between both symbiotic partners and involves the induction otic interaction, and by environmental factors such as soil
and repression of a large number of bacterial and plant genes. N level and the presence of existing nodules (Caetano-Annoles
Free-living rhizobia infect and form N2-fixing symbioses with 1997; Sagan and Gresshoff 1996; Singleton and Stockinger
legumes in a series of discrete stages or steps. Stages in the 1983).
process include proliferation of rhizobia in the rhizosphere, Nodule shape in legumes is determined by the host plant and
recognition of host by rhizobia, attachment of rhizobia to sus- is regulated by the pattern of cortical cell divisions. There are two
ceptible root-hair cells, root-hair curling and infection-thread basic types of nodules that are formed on legumes: determinant
formation, initiation of nodule primordium, and transforma- and indeterminant (Franssen et al. 1992). Indeterminant nod-
tion of free-living rhizobia into N2-fixing bacteroids. ules are most commonly formed in symbioses between the fast-
Rhizobia infect their respective host plants and induce root growing root-nodule bacteria, such as members of the genera
or stem nodules using several different mechanisms. Infection Ensifer or Rhizobium, and temperate legumes (pea, clover, and
through root hairs is commonly seen with most legumes (Hadri alfalfa). In contrast, determinate nodules are typically induced
et al. 1998). Rhizobia can also invade the host plant by entry by the bradyrhizobia and some rhizobial species strains and are
through wounds, cracks, or lesions caused by emergence of more common on tropical legumes, such as soybean and bean.
secondary roots (Boogerd and van Rossum 1997), as occurs in Morphologically, indeterminate nodules have defined, persis-
peanut and Stylosanthes. In these cases, rhizobia spread tent apical meristems and are elongated and sometimes lobed,
intercellularly. There are instances where the same rhizobia whereas determinant nodules do not have persistent meristems
infect one legume through root hairs and another via cracks or and are usually round (Hadri et al. 1998).
wounds (Sen and Weaver 1988). Lastly, rhizobia may initiate
infection of the host via cavities surrounding adventitious root
primordia on the stems of Sesbania, Aeschynomene, Neptunia,
and Discolobium (Boivin et al. 1997; Giraud et al. 2007). As
Genetics
above, one bacterium may produce both stem and root nodules
Understanding the genetics of the rhizobia has, in most cases,
on different legume plants.
centered on the genetics of nodulation and symbiotic
In root-hair infection, rhizobia attach to susceptible root
N2-fixation, key characters that set the rhizobia apart from
hairs within minutes of inoculation or contact with the host
other soil bacteria. Recent advances in molecular biology and
plant. Rhizobial cells often attach perpendicular to the root-hair
genetics have elucidated a large number of genes with symbiotic
cell. It has been suggested that adhesion is initially mediated by
functions. Though many of these genes are clustered together
the calcium (Ca)-binding protein rhicadhesin, or by plant
(on the chromosome in some organisms and on symbiotic
lectins, and subsequent bonding via production of cellulose
plasmids in others), additional genes may be dispersed or
fibrils (Kijne 1992). It is hypothesized that rhizobia produce
located on different replicons. Consequently, all symbiotically
localized hydrolysis of the root-hair cell wall. Subsequent pene-
related genes will most likely not be found until total sequencing
tration of rhizobia through the cell wall leads to root-hair curl-
and functional genomic efforts are completed. Because the scope
ing, which may be visible 6–18 h after inoculation. The
of this chapter is broad, more detailed information on the
proportion of root hairs infected is low, the percentage of these
genetics of nodulation and N2 fixation can be found in several
giving rise to nodules is low and highly variable, and aborted
reviews (Boivin et al. 1997; Denarie et al. 1996; Jones et al. 2007;
root hairs can frequently be found.
Niner and Hirsch 1998; Pueppke 1996; Spaink 1995; Schultze
Within the root hair, rhizobia are enclosed within a plant-
and Kondorosi 1998; van der Drift et al. 1998).
derived infection thread and move down the root hair in the
direction of the root cortex (Jones et al. 2007). Cell division in
the root cortex, in advance of the approaching infection thread,
leads to the production of nodule primordia (Kijne 1992). Nodulation Genes
Spread of the infection thread among cells of the nodule pri-
mordium follows, with the release of rhizobia into host cortex by In the last several years, a large number of bacterial genes
an endocytotic process. Rhizobia are never free in the cytoplasm have been identified which are involved in the formation of
but rather are surrounded by a host-derived peribacteroid mem- nodules on leguminous plants. Collectively, more than 65 nod-
brane, which serves to compartmentalize the rhizobia into ulation genes have been identified in rhizobia, although each
a symbiosome. One to several rhizobia can be confined to strain may only have a subset of these. Niner and Hirsch (1998),
a single symbiosome. Nodulation is usually visible 6–18 days Pueppke (1996), and Bladergroen and Spaink (1998) provide
after inoculation, but this varies considerably with the selection a more complete description of the function of a majority of
of bacterial strain and host cultivar, the inoculant density and these genes.
placement, and the temperature. Initially, nodulation is heaviest Several studies have shown that relatively few genes are
in the crown of the root, with secondary nodules appearing on required for nodulation of legumes (Göttfert 1993; Long et al.
1985; Long 1989; van Rhijn and Vanderleyden 1995). In the case genes and Nod factor are not required for symbiosis for these
of the fast-growing rhizobia, a majority of nodulation genes are strains and implies that an alternative nodulation strategy exists
located on large, indigenous, symbiotic (Sym), and often self- in the Bradyrhizobium-Aeschynomene symbiotic interaction
transmissible, plasmids (Broughton et al. 1984; Hombrecher (Giraud et al. 2007; Bonaldi et al. 2010).
et al. 1981; Kondorosi et al. 1989). The complete genomic
sequence of the symbiotic plasmid from Rhizobium sp. strain
NGR234, a Rhizobium strain with broad nodulation ability Genotype-Specific Nodulation Genes
(Pueppke and Broughton 1999), is currently available. In
E. meliloti, the symbiont of alfalfa, nodulation genes (located Although many hsn genes have been identified in Rhizobium and
on an 8.5 kb fragment of the Sym plasmid) contain sequences Bradyrhizobium, there are only limited reports on the identifi-
necessary for the nodulation of a wide variety of legume hosts cation of genotype-specific nodulation (GSN) genes in the
(Kondorosi et al. 1989; Truchet et al. 1991). These genes, referred rhizobia (Sadowsky et al. 1991). The GSN genes specifically
to as ‘‘common nodulation’’ genes and designated nodA, nodB, refer to those bacterial genes that allow nodulation of specific
and nodC, have homologues in other fast- and slow-growing plant genotypes within a given legume species. For example,
species. The common nodulation genes are involved in biosyn- strain TOM nodulates the pea genotype Pisum sativum cv.
thesis of the chitin backbone of Nod factor (see below) and Afghanistan (Lie 1978; Lie et al. 1978), but European R.
organized in a similar cluster in most rhizobia (Long et al. leguminosarum bv. viceae strains fail to nodulate this host.
1985; Long 1989; van Rhijn and Vanderleyden 1995). A fourth Some GSN-like genes have been found on plasmid pRL5JI of
gene, nodD, is regulatory and together with plant flavonoid strain TOM (Gotz et al. 1985; Hombrecher et al. 1984). Davis
signals (see below) activates transcription of other inducible et al. (1988) have identified a single gene on this plasmid, nodX,
nod genes (Long 1989; Martinez et al. 1990; van Brussel et al. mediating the O-acetylation of Nod factors (Firmin et al. 1993),
1990). R. leguminosarum bv. viceae and trifolii have single copies which is necessary for the nodulation of ‘‘Afghanistan’’ peas. In
of nodD. The symbionts E. meliloti, B. japonicum, and M. loti E. fredii strain USDA 257, two other GSN-like loci, nolC
have multiple copies of nodD (Göttfert et al. 1986, 1990; Honma (Krishnan and Pueppke 1991) and nolBTUVW (Meinhardt
and Ausubel 1987; Kaneko et al. 2000). In some instances, nodD et al. 1993), allow this strain to nodulate primitive lines of
also appears to impart host-specificity functions (Spaink et al. soybean, but not improved soybean varieties, such as ‘‘McCall’’
1987). Another nodulation gene cluster, originally designated (Heron et al. 1989). In each case, Tn5 insertions in the gene
hsn (for host-specific nodulation), is closely linked to the com- regions allow E. fredii to nodulate commercial soybean cultivars.
mon nodulation region in E. meliloti and controls nodulation of Phenotypically, these regions are similar to that reported by
specific legume genera (Bachem et al. 1986; Horvath et al. 1986). Djordjevic et al. (1985) and Innes et al. (1985) for clover
Mutations in the hsn genes (designated nodFEGH) cannot be rhizobia. More recently, however, Lewis-Henderson and
complemented with Sym plasmids from other species of Rhizo- Djordjevic (1991a) reported that nodM in R. leguminosarum
bium. Analogous hsn genes also have been isolated from R. bv. trifolii is a GSN which prevents effective nodulation of
leguminosarum bv. trifolii (Djordjevic et al. 1985; Rolfe et al. subterraneum cv. Woogenellup (Lewis-Henderson and
1985), leguminosarum bv. viceae (Wijffelman et al. 1985), and Djordjevic 1991b). Analysis of R. leguminosarum bv. trifolii
from Rhizobium strain MPIK3030 (Bachem et al. 1986; Bassam strain TA1 demonstrated that this strain also lacks nodT and
et al. 1986; Broughton et al. 1984; Lewin et al. 1987). An hsn gene that introduction of nodT from R. leguminosarum bv. viceae
linked to the common nodulation region in B. japonicum strain strain ANU843 into TA1 allows effective nodulation of
USDA 110 was also reported (Nieuwkoop et al. 1987). This ‘‘Woogenellup’’ (Lewis-Henderson and Djordjevic 1991a). The
sequence, subsequently called nodZ (Dockendorff et al. 1994), GSN genes can act in either a positive or negative manner
was shown to be involved in the host-specific nodulation of (Djordjevic et al. 1987a; Sadowsky et al. 1990), insertions in
siratro, but not soybean. Hahn and Hennecke (1988) and a negatively acting nodulation gene extending host range and
Göttfert et al. (1990) have identified another hsn locus in B. insertions in a positively acting GSN gene limiting host range. B.
japonicum strain USDA 110, nodVW, which is essential for the japonicum serogroup 123 strains are restricted for nodulation by
nodulation of siratro, mungbean, and cowpea, but not soybean. PI 377578 (Cregan and Keyser 1986). The B. japonicum nolA
In B. japonicum strain USDA 110, the essential nodulation genes gene, identified in strain USDA 110, is a positively acting gene
are located on the chromosome in several transcriptional units that allows serogroup 123 strains to nodulate PI 377578
in the order: nolZ, nolA, nodD2, nodD1, nodYABCSUIJmolMNO (Sadowsky et al. 1991). We recently identified a mutant of
(Dockendorff et al. 1994). Unlike other rhizobial nodD genes, B. japonicum strain USDA 110 that has the ability to overcome
the B. japonicum nodD1 is induced by the flavonoids genistein nodulation restriction conditioned by soybean PI 417566
and daidzein (Banfalvi et al. 1988; Kosslak et al. 1987) and by (Lohrke et al. 1995). More recently, the soybean genes Rj2 and
xanthones (Zaat et al. 1987). Rfg1, responsible for restricting nodulation by specific strains of
Intriguingly, Bradyrhizobium strains BTAi1 and ORS278, B. japonicum and E. fredii, respectively, were positionally cloned
which induce nodules on both the root and stem of the aquatic and found to be members of the Toll-interleukin receptor/nucle-
legume Aeschynomene, do not possess functional nodABC genes otide-binding site/leucine-rich repeat class of plant resistance
(Giraud et al. 2007). This indicates that canonical nodABC proteins (Yang et al. 2010).
Signal Exchange and Induction of Nod Genes various substituents. For example, the E. meliloti genes nodP,
nodQ, and nodH are involved in the sulfation of the Nod factor
Although the regulation of nodulation genes in rhizobia is still reducing sugar (Faucher et al. 1989; Roche et al. 1991). Disrup-
not fully understood, we know a lot about communication tion of any of these genes affects host specificity. Rhizobium
between rhizobia and susceptible legume hosts. Flavonoid signal spp. NGR234, which can nodulate over 125 different legume
molecules present in root and seed exudates are necessary for species (Pueppke and Broughton 1999), produce diverse (more
nod gene expression (Banfalvi et al. 1988; Boundy-Mills et al. than 18) Nod factors, which vary in the substituents attached to
1994; Djordjevic et al. 1987b; Fellay et al. 1995; Göttfert et al. a similar backbone structure (Price et al. 1992).
1988; Innes et al. 1985; Kosslak et al. 1987; Long 1989; Mulligan Purified Nod factors, which are structurally similar to those
and Long 1985; Olson et al. 1985; Peters et al. 1986; Price et al. produced by the appropriate rhizobial symbiont, can induce
1992; Sadowsky et al. 1988; van Brussel et al. 1990; Zaat et al. nodules on the specific host plant in the absence of
1987). Other Sym plasmid-borne genes are also induced by root a bacterium (Downie 1998; Mergaert et al. 1993; Relic et al.
exudates in E. fredii and Rhizobium sp. strain NGR234 (Boundy- 1993; Schultze et al. 1992; Truchet et al. 1991). Nod factors
Mills et al. 1994; Fellay et al. 1995; Olson et al. 1985; Sadowsky from several strains of B. japonicum have been characterized
et al. 1988). Flavones, isoflavones, flavanols, flavanones, and (Carlson et al. 1993; Sanjuan et al. 1992). The functions of nod
closely related compounds have been identified as nod gene genes and the basic structure of Nod factors for B. japonicum and
inducers, and each is specific for a particular legume-Rhizobium several species of the genus Rhizobium can be found in Downie
interaction (Schlaman et al. 1998). Flavonoid compounds are (1998), and recent review can be found in Murray (2011).
only one of several determinants of host specificity. Spaink et al.
(1991) reported differential induction of nodD in various fast-
growing rhizobia by a range of flavonoids and exudates. Induc- Nitrogen Fixation Genes
tion of nodulation genes requires the regulatory nodD gene
product (Long 1989; Mulligan and Long 1985; Shearman et al. Nitrogen fixation is the natural process, either biological or
1986). The inducer apparently binds NodD, causing a change in abiotic, by which nitrogen gas (N2) in the atmosphere is
conformation (Kondorosi et al. 1988; Fisher and Long 1989). converted into ammonium (NH4). Only bacteria containing
Activated NodD then binds to a regulatory, promoter-like the enzyme nitrogenase can reduce N2 to ammonium. This is
sequence, found upstream of rhizobial nod genes, the Nod box the only known enzyme that can carry out this energetically
(Hong et al. 1987; Horvath et al. 1987; Kondorosi et al. 1988; unfavorable reaction. As described above, legume plants enter
Rostas et al. 1986; Shearman et al. 1986). Repressor proteins have into a symbiotic interaction with nitrogen-fixing rhizobia
also been suggested to play a role in nod gene regulation resulting in the formation of root or stem nodules. Nitrogenase
(Kondorosi et al. 1988), a repressor encoded by the nolR gene is very sensitive to inactivation by oxygen, and within the nod-
has been identified in E. meliloti strain 41 (Kondorosi et al. 1989, ules, leghemoglobin maintains a low-oxygen concentration to
1991), and a repressor encoded by the nolA and nodD2 has been protect enzyme function (nanomolar range) (Downie 2005).
identified in B. japonicum strain USDA 110 (Sadowsky et al. Two major types of N2 fixation genes have been described,
1991; Göttfert et al. 1992; Loh and Stacey 2003). This section has nif genes and fix genes. The nif refer to genes involved in the
recently been reviewed by Downie (2010) and the reader is N2-fixation process and have structurally and functionally
directed here. related genes in the free-living diazotrophic microorganism,
Klebsiella pneumoniae. K. pneumoniae was the first N2-fixing
microorganism studied in detail (Kennedy 1989). As with the
Extracellular Nodulation Factors nodulation genes, a majority of the nif genes are plasmid borne
and contiguous in the rhizobia. In contrast, the nif genes are
One of the primary functions of nod genes is the production of chromosomally located in the bradyrhizobia. The N2-fixation
extracellular lipochitooligosaccharide (LCO) molecules, also process is catalyzed by the enzyme complex nitrogenase,
known as Nod factors (Carlson et al. 1993, 1994). These mole- encoded by the nifDK and nifH genes. The fix genes are also
cules, acting at 10 8 to 10 9 M, can (1) stimulate the plant to involved in the N2-fixation process but have no similar struc-
produce more nod gene inducers (van Brussel et al. 1990), tural or functional homologues in K. pneumoniae. The organi-
(2) deform root hairs on homologous hosts (Banfalvi and zation of nif genes varies in the rhizobia (Kaminski et al. 1998).
Kondorosi 1989; Faucher et al. 1989), and (3) initiate cell divi- Nitrogenase consists of two protein subunits,
sion in the root cortex (Lerouge et al. 1990; Price et al. 1992; a molybdenum-iron (MoFe) protein and an iron-containing
Sanjuan et al. 1992; Schultze et al. 1992; Relic et al. 1993; Spaink (Fe) protein. These structural components of the nitrogenase
et al. 1991). In E. meliloti, these signal molecules are acetylated enzyme complex are often referred to as subunits I and II,
and sulfated glucosamine oligosaccharides (Lerouge et al. 1990). respectively. The nifK and nifD genes encode the MoFe protein
Similar molecules have been identified in other legume symbi- subunits. A FeMo cofactor (FeMo-Co) is required for activation
otic systems (Pueppke (1996) and Downie (1998, 2010) for of the MoFe protein and is assembled from the nifB, nifV, nifN,
a review). Numerous observations support the theory that hsn and nifE genes. The nifH gene encodes the Fe subunit protein. In
genes control host specificity by decorating Nod factors with K. pneumoniae, there are at least 20 nif-specific genes that are
. Table 17.3
Other bacterial genes involved in symbiotic nitrogen fixation
Gene designation Phenotype or function References

exo Exopolysaccharide Becker and Puhler (1998), Glazebrook and Walker (1989)
hup Hydrogen uptake Maier (1986)
gln Glutamine synthase Carlson et al. (1987)
dct Dicarboxylate transport Finan et al. (1983), Jiang et al. (1989)
nfe Nodulation efficiency Sanjuan and Olivares (1989)
ndv b-1,2 Glucans Breedveld and Miller (1998)
lps Lipopolysaccharide Carlson et al. (1987)
bacA Bacteroid development Glazebrook et al. (1993)
tts Type III secretion system Krause et al. (2002)
virB Type IV secretion system Hubber et al. (2004)
acdS, rtx Inhibition of plant ethylene biosynthesis Ma et al. (2003), Duodu et al. (1999)
pur Purine biosynthesis Noel et al. (1988), Giraud et al. (2007)
localized in about 8 operons (Dean and Jacobson 1992). Though been completed and are currently available online
the organization of nif genes in other organisms varies tremen- (> Table 17.4), and approximately 150 more are currently
dously (Downie 1998), nifHD and nifK are conserved in dispa- being sequenced. In most of the genome-sequenced rhizobia,
rate N2-fixing organisms and rhizobia (Ruvkin and Ausubel the genes involved in nodulation and nitrogen-fixation genes are
1980). The gene products NifA and NifL control the regulation clustered on large and often self-transmissible, megaplasmids
of all other nif genes. Whereas NifA is positive activator of (pSyms) or are located within large genomic islands, referred to
transcription of nif operons, NifL is involved in negative control. as symbiotic islands (SIs). These features in many ways empha-
In K. pneumoniae and several other free-living diazotrophic size the accessory nature of the symbiosis-related genes and their
microbes, nif gene expression is regulated by oxygen and nitro- ability to be acquired by microorganisms via horizontal gene
gen levels (Merrick 1992). Ammonia (NH3) causes NifL to act as transfer (MacLean et al. 2007).
a negative control and prevents the activator function of NifA. A SI present in Mesorhizobium loti strain ICMP3153 was
This has been referred to as the N control system and has been found capable of transforming nonsymbiotic strains of M. loti
shown to regulate several enzymes that are capable of producing into symbiotic counterparts (Sullivan and Ronson 1998). Large
NH3. Merrick (1992) and Dean and Jacobson (1992) give excel- symbiotic islands of 611 and 681 kb have been found in M. loti
lent in-depth reviews of the structure and regulation of N2 MAFF303099 and B. japonicum strain USDA 110, respectively
fixation in free-living and symbiotic bacteria. (Kaneko et al. 2000, 2002). Although the widespread transmis-
sibility of these islands has yet to be confirmed, the integration of
symbiotic islands into the M. loti and B. japonicum genomes
Other Genes Involved in Symbiotic Nitrogen occurs within phe-tRNA and val-tRNA genes, respectively
Fixation (Kaneko et al. 2000, 2002). The association of symbiotic islands,
in both rhizobial species, with a phage-related integrase implies
Other plasmid- and chromosomally borne bacterial genes also the symbiotic islands may have originated from the ancient
have been found to function indirectly in nodulation and sym- integration of a bacteriophage. Beside the SI, 14 smaller genomic
biotic N2-fixation in rhizobia (> Table 17.3). Recent review islands, with lower GC content, were also found in the genome
articles on the structure and function of these and other symbi- of B. japonicum USDA 110 (Kaneko et al. 2002). Since some of
osis-related genes are provided by Pueppke (1996), Spaink these genomic islands were found to be missing in several strains
(1995), Long et al. (1999), Sugawara et al. (2006), and Kobayashi of B. japonicum, the 14 genomic islands were thought to be likely
and Broughton (2008). inserted into the ancestor genome of USDA 110 via horizontal
gene transfer events (Itakura et al. 2009). In contrast to what has
been found in the slower-growing bradyrhizobia, most of the
Genomics genes involved in symbioses in genome-sequenced Ensifer and
Rhizobium strains are located on symbiotic plasmids, pSyms. In
Several approaches have been used to define and study the many cases, the pSyms have been shown to be transferred among
involvement of whole bacterial genomes in the symbiotic pro- bacteria via conjugation (Rao et al. 1994; Freiberg et al. 1997;
cess. As of this date, about 15 rhizobial genome sequences have Brom et al. 2004; Pérez-Mendoza et al. 2004).
. Table 17.4
Architecture of sequenced rhizobial genomes
Rhizobial strain No. replicons Total size (bp) Genbank accession no. References
Alpha-proteobacteria
A. caulinodans ORS571 1 5,369,772 AP009384 Lee et al. (2008)
B. japonicum USDA 110 1 9,105,828 BA000040 Kaneko et al. (2002)
Bradyrhizobium sp. BTAi1 2 8,493,513 CP000494, CP000495 Giraud et al. (2007)
Bradyrhizobium sp. ORS278 1 7,456,587 CU234118 Giraud et al. (2007)
Ensifer (Sinorhizobium) medicae WSM419 4 6,817,576 CP000738–CP000741 Reeve et al. (2010)
Ensifer (Sinorhizobium) meliloti 1021 3 6,691,694 AL591688, AL591985, AE006469 Galibert et al. (2001)
Mesorhizobium loti MAFF303099 3 7,596,297 AP002994–AP003017 Kaneko et al. (2000)
Methylobacterium nodulans ORS2060 8 8,839,022 CP001349–CP001356 Unpublished
Rhizobium etli CFN42 7 6,530,228 CP000133–CP000138 González et al. (2006)
Rhizobium etli CIAT652 4 6,448,048 CP001074–CP001077 Unpublished
Rhizobium leguminosarum bv. trifolii WSM1325 6 7,418,122 CP001622–CP001627 Unpublished
Rhizobium leguminosarum bv. trifolii WSM2304 5 6,872,702 CP001191–CP001195 Unpublished
Rhizobium leguminosarum bv. viciae 3841 7 7,751,309 AM236080–AM236086 Young et al. (2006)
Rhizobium sp. NGR234 3 6,891,900 CP000874, CP001389, U00090 Schmeisser et al. (2009)
Beta-proteobacteria
Burkholderia phymatum STM815 4 8,676,562 CP001043–CP001046 Unpublished
Cupriavidus (Ralstonia) taiwanensis LMG19424 3 6,476,522 CU633749–CU633751 Amadou et al. (2008)
Ecology nodule tissue (McDermott et al. 1987). When these nodules

senesce at the end of the growing season, large numbers of
Rhizobia are relatively unique among the majority of soil micro- rhizobia are released into the soil. Nodule bacteroids are subject
organisms in that they have an extensive soil phase as free-living, to changes in surface chemistry (Roest et al. 1995) and are
saprophytic, heterotrophic microorganisms, yet they have the susceptible to osmotic and other soil stresses (Sutton 1983).
ability to form species-specific, N2-fixing symbiotic associations However, many of the released organisms manage to persist as
with legumes. The ability to form N2-fixing nodules affords free-living, heterotrophic, saprophytes in the soil until
unique opportunities for the rhizobia. When a legume crop is a susceptible legume is again planted. As a consequence of this,
grown in soil for the first time, few rhizobia are likely to be most soils contain at least some rhizobia, and a dramatic
present, and in most instances, inoculation will most likely be buildup in their numbers occurs when a leguminous host is
needed for adequate nodulation and subsequent N2 fixation included in a crop rotation, pasture, or natural setting.
(Date 1991; Diatloff 1977). In contrast, soils surrounding Ellis et al. (1984) reported soil populations of bradyrhizobia
legumes that have been planted for several years usually contain approaching 106 cells g 1 in soils of the American Midwest
relatively large numbers of rhizobia and do not require added following cultivation of soybean, and rhizosphere populations
strains. Numerous studies have documented that legume inoc- can reach 108 cells g 1 (Bottomley 1992). The distribution of
ulants added to soils containing relatively small populations of rhizobia in soil is not uniform. Postma et al. (1990) reported that
rhizobia usually give rise to only a small percentage of the the greatest numbers of rhizobia are associated with soil aggre-
nodules formed (Thies et al. 1991; Date 1991; Ellis et al. 1984; gates of larger than 50 mm, and Mendes and Bottomley (1998)
Ham 1978). Despite intensive investigations over the last 30+ noted that the percentage of Rhizobium recovered from aggre-
years, however, some of the factors that influence the survival gates of different sizes varied over the course of a growing
and the persistence of rhizobia in the soil, and their ecology and season.
competitiveness for nodulation sites on the host, are only now Rhizobia are excellent soil saprophytes and can persist for
beginning to be understood. It is beyond the scope of this many years in the absence of their host (Brunel et al. 1988; Kucey
chapter to present all that is known about the ecology and soil and Hynes 1989; Bottomley 1992). Chatel et al. (1968) used the
biology of the rhizobia. The reader is directed to more extensive term saprophytic competence to describe this ability, but the
reviews by Bottomley (1992), McInnes et al. (2004) and factors involved have yet to be determined. Even though Bushby
Sadowsky and Graham (1998) on this material. (1990) noted surface electrophoretic charge in bradyrhizobia
The establishment of the symbiotic state results in the pro- correlated to the pH of soils from which they came, Rynne
duction of nodule populations of more than 1010 rhizobia g 1 et al. (1994) found no correlation between catabolic ability and
strain persistence. More recently, Ratcliff et al. (2008) suggested problem in sandy soils amended with montmorillonite and
that there is a relationship between strain survival and PHB illite. Along with temperature, soil water stress can have
accumulation in rhizobia. Inoculant strains used at the time a profound influence on the survival of rhizobia in soil, affecting
a particular host was introduced may still occupy a large per- the induction and repression of a large number of bacterial genes
centage of the nodules formed on that host 10–15 years after (Cytryn et al. 2007).
their introduction (Diatloff 1977; Brunel et al. 1988; Lindstrom Although rhizobia comprise only 0.1–8.0 % of the total bac-
et al. 1990). However, many studies have shown that inoculant terial population in soil and 0.01–0.14 % of its biomass (Bottomley
strains may also decline in nodule representation over time or 1992; Schortemeyer et al. 1997), their biodiversity in soil and the
quite quickly disappear from soil. factors which can affect it have been extensively studied. The
The growth of rhizobia in the rhizosphere may also be development of improved techniques for strain fingerprinting,
stimulated by specific root exudates (Van Egaraat 1975). particularly restriction fragment length polymorphism (RFLP)
Rhizobia, in turn, also stimulate growth and respiration of and PCR analyses (de Bruijn 1992; Judd et al. 1993; Dye et al.
leguminous plants (Phillips et al. 1999). Several research studies 1995; Madrzak et al. 1995; Richardson et al. 1995; Brunel et al.
have sought to create biased rhizospheres, in which plants 1996; Labes et al. 1996; Paffetti et al. 1996; Rome et al. 1996;
transformed to synthesize opines and favor the growth of Sadowsky and Hur 1998), multilocus enzyme electrophoresis
rhizosphere bacteria utilizing this substrate (Rossbach et al. (MEE) (Pinero et al. 1988; Eardly et al. 1990; Demezas et al.
1994; Oger et al. 1997; Savka and Farrand 1997). The inability 1991, 1995; Bottomley et al. 1994; Dupuy et al. 1994; Souza et al.
of strains to compete for nodulation sites on the host legume 1994; Strain et al. 1994, 1995), and SDS-PAGE analysis of total
does not necessarily mean their displacement from the soil cell proteins (Roberts et al. 1980; Dupuy et al. 1994), has been
population. Bromfield et al. (1995) compared populations of used extensively to study strain biodiversity.
E. meliloti recovered from soil and nodules and found significant Population biodiversity among the rhizobia for a particular
differences in the frequency with which particular genotypes legume species tends to be greatest near the center of origin/
were recovered. Similarly, Segovia et al. (1991) found the pop- domestication of that legume (Lie et al. 1987). Pinero et al.
ulation of noninfective bean rhizobia in soil numerically supe- (1988) recorded a mean genetic distance per enzyme locus of
rior to those capable of inducing nodule formation. 0.691 for 51 isolates of etli from the Mesoamerican center of
Environmental factors, particularly soil pH, temperature, origin for Phaseolus vulgaris (L.), while Souza et al. (1994)
and water availability, often affect rhizobial survival in soil and grouped 372 bean rhizobia into seven clusters comprising 95
the balance between particular genotypes. In soils of pH >7.0, electrophoretic types. In the later study, rhizobia isolated from
Brockwell et al. (1991) found an average of 89,000 E. meliloti g 1 wild bean populations grouped by location and Phaseolus spe-
soil, whereas in soils of pH <6.0, the number was only 37 g 1. cies, whereas those from cultivated beans were very heteroge-
Even more striking is the replacement of the normal bean neous. An emerging consideration in data such as this is the
microsymbiont etli by the acid-tolerant tropici to which beans promiscuity of Phaseolus vulgaris. This host is nodulated by at
were introduced (Anyango et al. 1995; Hungria et al. 1997). This least five different species of rhizobia (Michiels et al. 1998),
change occurred in the relatively short time since Spanish and necessitating care in distinguishing between intra- and
Portuguese colonization of Latin America and despite the fact extraspecific diversity. At the other extreme, the movement of
that tropici is less competitive than etli in nodule formation with rhizobia as seed-borne contaminants (Perez-Ramirez et al. 1998)
beans (Martinez-Romero and Rosenblueth 1990; Chaverra and can give the impression of limited biodiversity, more properly
Graham 1992). In contrast, Richardson and Simpson (1989) identified as founder effects (Hagen and Hamrick 1996). The
found that many rhizobia from acid soils are sensitive to acidity, environmental stresses noted above also can have profound
suggesting that microniches in soils protect these microorgan- effects on the biodiversity of rhizobia in soil. In an extreme
isms from extremes of soil pH. case (Hirsch 1996), application of manure containing heavy
Temperature also has a marked influence on survival and metals reduced the biodiversity of R. leguminosarum bv. trifolii
persistence of rhizobial strains. Eaglesham et al. (1981) found in soil to a single biotype. Caballero-Mellado and Martinez-
cowpea rhizobia from the hot, dry Sahelian savannah of West Romero (1999) also reported fertilizer effects on strain biodi-
Africa to be temperature tolerant, with good growth at 37 C. versity in soil.
More than 90 % of the strains isolated from this region grew well
to 40 C, whereas rhizobia from the more humid Onne region of
West Africa generally failed to grow at this temperature. Soil Applications
temperature might contribute to the number of noninfective
rhizobia found in some soils (Segovia et al. 1991). Temperature It is unlikely that soils contain appropriate rhizobia when
effects appear to be both strain and soil dependent. Marshall a legume species is planted in a new area for the first time. In
(1964) studied a clover nodulation problem in which autumn- these cases, inoculation is usually required for adequate nodu-
sown plants nodulated well but frequently failed to do so in lation and N2 fixation. Yield increases following inoculation
subsequent regrowth. He found that Bradyrhizobium sp. with appropriate inoculant-quality rhizobia can exceed 50 %,
(Lupinus) was less susceptible than R. leguminosarum bv. trifolii with clear differences between inoculated and uninoculated
to high soil temperatures but also noted amelioration of this plants as shown in > Fig. 17.3.
. Fig. 17.4
. Fig. 17.3
PulseR™: A presterilized-peat-based inoculant for soybean
Response of soybean to inoculation in newly cultivated areas of
(Photograph: H. Mc Ives, Agribiotics)
Puerto Rico. The yellow-green plants in the center of the picture
were not inoculated with Rhizobium (Courtesy of R. Stewart Smith)
use of the correct rhizobia. Specificity in nodule formation

Where plants are not inoculated with rhizobia in the first between host and rhizobia has been referred to already; speci-
year of introduction into a new area, nodulation will most likely ficities between host and Rhizobium in terms of N2 fixation also
be limited to that coming from seed-borne or aerial contami- exist (Burton 1967) and impact inoculant strain selection. EMD
nants. It is common that these rhizobia are less than fully Crop Bioscience (Milwaukee, WI), formerly the Nitragin Com-
effective in symbiosis with their host (Guar and Lowther pany, provides more than 100 different strain preparations for
1980). Over several years of cultivation, these rhizobia will legume inoculation (Smith 1988). This company now produces
increase in numbers in the soil, limiting subsequent inoculation legume inoculants supplemented with LCOs (Optimize®) to
response. It is important, therefore, that inoculation with an enhance nodulation. Other legume inoculant companies also
inoculant-quality organism lead to early establishment and per- produce products with chemical additives to enhance
sistence of effective rhizobia in the soil, ensuring long-term nodulation.
benefits. Where this is done, the original inoculant strain(s) Consistent quality in the inoculant material is also essential
may still dominate in nodulation 10–15 years later (Brunel but can be surprisingly variable. A number of countries regulate
et al. 1988; Kucey and Hynes 1989; Bottomley 1992). In Thai- inoculant quality. Thus, for example, inoculants sold in Canada,
land, 80 % of farmers in new areas of soybean production are Australia, Uruguay, and France must contain in excess of 108 or
willing to inoculate, but only 30 % of farmers in older 109 rhizobia g 1 peat carrier and be essentially contaminant free
established areas do, and in many cases, the latter apply fertilizer (Lupwayi et al. 2000). Inoculant quality in these countries is
N (Hall and Clark 1995). often controlled by a government testing agency and subject to
Inoculation of legumes with suitable rhizobia has been prac- law. Best results have traditionally been with inoculant formu-
ticed for more than 100 years. Initially, this involved the collec- lations that are sterilized before introduction of the rhizobia
tion of soil from areas where particular legumes had established (D. J. Hume and J. A. Omelian, personal communication).
successfully and the mixing of soil and seed prior to planting. These maintain higher numbers of rhizobia and have a longer
Inoculation with pure cultures of rhizobia followed, and today shelf life than unsterile peat or granular inoculants. Whereas the
the provision of rhizobial inoculants to farmers, home gar- large-scale production of inoculants is a relatively simple pro-
deners, and others is a multimillion dollar industry with approx- cess, not all inoculants meet the quality standards mentioned
imately 4 t peat culture sold annually in the United States alone above. For example, the 18 Argentine inoculants examined by
(Burton 1980). More than 80 % of this is for two crops, soybean Gomez et al. (1997) ranged in cell count from 0 to 109 rhizobia
and alfalfa. In Brazil where there was an early commitment to N2 g 1, and 14 contained more contaminants than rhizobia. More
fixation as the principal source of plant N, Dobereiner et al. recent liquid formulations have shown good survival of rhizobia
(1995) estimate the benefits of symbiotic and associative sym- in storage and on seeds and show great promise in delivering
bioses at more than $1.8 billion each year. Most inoculant large numbers of effective strains under field conditions
preparations are peat based (Smith 1992; Brockwell and (Albareda et al. 2008; Singleton et al. 2002; Walley et al. 2004).
Bottomley 1995), but frozen, granular, liquid, and other prepa- Environmental conditions also can affect inoculation suc-
rations also are used. Applied as recommended, these prepara- cess. The effect of soil acidity on rhizobia has already been
tions will supply between 109 and 1013 rhizobia ha 1, equivalent mentioned and has necessitated both a search for more acid-
to 103–106 rhizobia seed 1 (Lupwayi et al. 2000) (> Fig. 17.4). tolerant rhizobia (Graham et al. 1982, 1994; Howieson et al.
Many factors have to be considered if inoculation is to 1988; Howieson and Ewing 1989) or the use of pelleted, inocu-
function properly. Most important in legume inoculation is lated seed that provides a neutral environment prior to infection
(Somasegaran and Hoben 1994). Higher inoculation rates may Bonaldi K, Gourion B, Fardoux J, Hannibal L, Cartieaux F, Boursot M, Vallenet
D, Chaintreuil C, Prin Y, Nouwen N, Giraud E (2010) Large-scale transposon
be needed where high temperatures at seeding limit rhizobial
mutagenesis of photosynthetic Bradyrhizobium sp. strain ORS278 reveals
survival (Smith et al. 1981; Smith and del Rio Escurra 1982). new genetic loci putatively important for Nod-independent symbiosis with
A recent trend is for mixed inoculants, for example, containing Aeschynomene indica. Mol Plant Microbe Interact 23:760–770
Rhizobium plus biocontrol, phosphate-solubilizing, or growth Boogerd FC, van Rossum D (1997) Nodulation of groundnut by Bradyrhizobium
hormone–producing organisms (Rice et al. 1995; Burdmann – a simple infection process by crack entry. FEMS Microbiol Rev 21:5–27
Bottomley PJ (1992) Ecology of Bradyrhizobium and Rhizobium. In: Stacey G,
et al. 1996; Xi et al. 1996; Zhang et al. 1997). Results of
Burris R, Evans HJ (eds) Biological nitrogen fixation. Chapman and Hall,
coinoculation experiments with Azospirillum are promising; New York, p 943
more detailed field experimentation is needed to establish the Bottomley PJ, Cheng HH, Strain SR (1994) Genetic structure and symbiotic
value of the other combined formulations. A concern in all such characteristics of a Bradyrhizobium population recovered from a pasture
preparations must be the compatibility and survival of the soil. Appl Environ Microbiol 60:1754–1761
Boundy-Mills KL, Kosslak RM, Tully RE, Pueppke SG, Lohrke S, Sadowsky MJ
various inoculant organisms used.
(1994) Induction of the Rhizobium fredii nod box-independent nodulation
gene nolJ requires a functional nodD1 gene. Mol Plant Microbe Interact
7:305–308
References Breedveld MW, Miller KJ (1998) Cell-surface g-glycans. In: Spaink HP,
Kondorosi A, Hooykaas PJJ (eds) The Rhizobiaceae. Kluwer, Dordrecht,
Albareda M, Rodrı́guez-Navarro DN, Camacho M, Temprano FJ (2008) Alterna- pp 81–96
tives to peat as a carrier for rhizobia inoculants: solid and liquid formula- Brewin NJ, Beringer JE, Johnston AWB (1980) Plasmid mediated transfer of host
tions. Soil Biol Biochem 40:2771–2779 range specificity between two strains of Rhizobium leguminosarum. J Gen
Amadou C, Pascal G, Mangenot S, Glew M, Bontemps C, Capela D, Carrère S, Microbiol 120:413–420
Cruveiller S, Dossat C, Lajus A, Marchetti M, Poinsot V, Rouy Z, Servin B, Brockwell J, Bottomley PJ (1995) Recent advances in inoculant technology and
Saad M, Schenowitz C, Barbe V, Batut J, Médigue C, Masson-Boivin prospects for the future. Soil Biol Biochem 27:683–697
C (2008) Genome sequence of the beta-rhizobium Cupriavidus taiwanensis Brockwell J, Gault RR, Zorin M, Roberts MJ (1982) Effects of environmental
and comparative genomics of rhizobia. Genome Res 18:1472–1483 variables on the competition between inoculum strains and naturalized
Amarger N, Macheret V, Laguerre G (1997) Rhizobium gallicum sp. nov. and populations of Rhizobium trifolii for nodulation of Trifolium subterraneum
Rhizobium giardinii sp. nov., from Phaseolus vulgaris nodules. Int J Syst L. and on rhizobia persistence in the soil. Aust J Agric Res 33:803–815
Bacteriol 47:996–1006 Brockwell J, Pilka A, Holliday RA (1991) Soil pH is a major determinant of the
An DS, Im WT, Yang HC, Lee ST (2006) Shinella granuli gen. nov., sp. nov., and numbers of naturally-occurring Rhizobium meliloti in non-cultivated soils in
proposal of the reclassification of Zoogloea ramigera ATCC 19623 as Shinella central New South Wales, Australia. Aust J Exp Agric 31:211–220
zoogloeoides sp. nov. Int J Syst Evol Microbiol 56:443–444 Brom S, Girard L, Tun-Garrido C, Garcia SA de los, Bustos P, González V, Romero
Anyango B, Wilson KJ, Beynon JL, Giller KE (1995) Diversity of rhizobia D (2004) Transfer of the symbiotic plasmid of Rhizobium etli CFN42 requires
nodulating Phaseolus vulgaris L in two Kenyan soils with contrasting pHs. cointegration with p42a, which may be mediated by site-specific recombi-
Appl Environ Microbiol 61:4016–4021 nation. J Bacteriol 186:7538–7548
Bachem CW, Banfalvi Z, Kondorosi E, Schell J, Kondorosi A (1986) Identification Bromfield ESP, Barran LR, Wheatcroft R (1995) Relative genetic structure of
of host range determinants in the Rhizobium species MPIK3030. Mol Gen a population of Rhizobium meliloti isolated directly from soil and from
Genet 203:42–48 nodules of alfalfa (Medicago sativa) and sweet clover (Melilotus alba). Mol
Banfalvi Z, Kondorosi A (1989) Production of root hair deformation factors by Ecol 4:183–188
Rhizobium meliloti nodulation genes in Escherichia coli: HsnD (nodH) is Broughton WJ, Heycke N, Meyer ZA, Pankhurst CE (1984) Plasmid linked nif
involved in plant host-specific modification of the nodABC factor. J Mol Biol and nod genes in fast-growing rhizobia that nodulate Glycine max,
13:1–12 Psophocarpus tetragonolobus and Vigna unguiculata. Proc Natl Acad Sci
Banfalvi Z, Nieuwkoop A, Schell M, Best L, Stacey G (1988) Regulation of nod USA 81:3093–3097
expression in Bradyrhizobium japonicum. Mol Gen Genet 214:420–424 Brunel B, Cleyet-Marel JC, Normand P, Bardin R (1988) Stability of
Barber LE (1979) Use of selective agents for recovery of Rhizobium meliloti. Soil Bradyrhizobium japonicum inoculants after introduction into soil. Appl
Sci Soc Am J 43:1145–1148 Environ Microbiol 54:2636–2642
Bassam BJ, Rolfe BG, Djordjevic MA (1986) Macroptilium atropurpureum Brunel B, Rome S, Ziani R, Cleyet-Marel JC (1996) Comparison of/nucleotide
(siratro) host specificity genes are linked to a nodD-like gene in the broad diversity and symbiotic diversity of Rhizobium meliloti populations from
host range Rhizobium strain NGR234. Mol Gen Genet 203:49–57 annual Medicago species. FEMS Microbiol Ecol 19:71–82
Becker A, Puhler A (1998) Production of exopolysaccharides. In: Spaink HP, Burdmann S, Sarig S, Kigel J, Okon Y (1996) Field inoculation of common bean
Kondorosi A, Hooykaas PJJ (eds) The Rhizobiaceae. Kluwer, Dordrecht, (Phaseolus vulgaris L.) and chickpea (Cicer arietinum L.) with Azospirillum
pp 97–118 brasilense strain CD. Symbiosis 212:41–48
Berge O, Lodhi A, Brandelet G, Santaella C, Roncato M, Christen R, Heulin T, Burton JC (1967) Rhizobium culture and use. In: Peppler HJ (ed) Microbial
Achouak W (2009) Rhizobium alamii sp. nov., an exopolysaccharide- technology. Van Nostrand-Reinhold, New York, pp 1–33
producing species isolated from legume and non-legume rhizospheres. Int Burton JC (1980) Modern concepts in legume inoculation. In: Graham PH,
J Syst Evol Microbiol 59:367–372 Harris SC (eds) BNF technology for tropical agriculture. CIAT, Colombia,
Beringer JE, Beynon JL, Buchanan-Wollaston AV, Johnston AWB (1978) Transfer pp 105–114
of the drug-resistance transposon Tn5 to Rhizobium. Nature 276:633–634 Bushby HVA (1990) The role of bacterial surface charge in the ecology of root
Bladergroen MR, Spaink HP (1998) Genes and signal molecules involved in the nodule bacteria: An hypothesis. Soil Biol Biochem 22:1–9
Rhizobia-leguminosae symbiosis. Curr Opin Plant Biol 1:353–359 Caballero-Mellado J, Martinez-Romero E (1999) Soil fertilization limits the
Bohlool BB, Schmidt EL (1973) Persistence and competition aspects of Rhizobium genetic diversity of Rhizobium in bean nodules. Symbiosis 26:111–121
japonicum observed in soil by immunofluorescence microscopy. Soil Sci Soc Caetano-Annoles G (1997) Molecular dissection and improvement of the nodule
Am Proc 37:561–564 symbiosis in legumes. Field Crops Res 53:47–68
Boivin C, Ndoye I, Molouba F, Delajudie P, Dupuy N, Dreyfus B (1997) Stem Carlson RW, Kalembasa S, Turowski D, Pachori P, Noel KD (1987) Characteriza-
nodulation in legumes–diversity, mechanisms, and unusual characteristics. tion of the lipopolysaccharide from a Rhizobium phaseoli mutant that is
Crit Rev Plant Sci 16:1–30 defective in infection thread development. J Bacteriol 169:4923–4928
Carlson RW, Sanjuan J, Bhat UR, Glushka J, Spaink HP, Wijfjes AHM, van Brussel Cubo MT, Buendia-Claveria AA, Beringer JE, Ruiz-Sainz JE (1988) Melanin
AAN, Stokkermans TJW, Peters NK, Stacey G (1993) The structures and production by Rhizobium strains. Appl Environ Microbiol 54:1812–1817
biological activities of the lipo-oligosaccharide nodulation signals produced Cytryn EJ, Sangurdekar DP, Streeter JG, Franck WL, Chang W-S, Stacey G,
by type I and II strains of Bradyrhizobium japonicum. J Biol Chem Emerich DW, Joshi T, Xu D, Sadowsk MJ (2007) Transcriptional and phys-
268:18372–18381 iological responses of Bradyrhizobium japonicum to desiccation-induced
Carlson RW, Price NPJ, Stacey G (1994) The biosynthesis of rhizobial lipo- stress. J Bacteriol 189:6751–6762
oligosaccharide nodulation signal molecules. Mol Plant-Microbe Interact Dangeard PA (1926) Recherches sur les tubercles radicaux des Le´gumineuses.
7:684–695 Botaniste (Paris) 16:1–275
Casida LE Jr (1982) Ensifer adhaerens gen. nov., sp. nov.: A bacterial predator of Date RA (1982) Collection, isolation, characterization and conservation of
bacteria in soil. Int J Syst Bacteriol 32:339–345 Rhizobium. In: Vincent JM (ed) Nitrogen fixation in legumes. Academic,
Chabot R, Antoun H, Kloepper JW, Beauchamp CJ (1996) Root colonization of Sydney, pp 95–109
maize and lettuce by bioluminescent Rhizobium leguminosarum biovar Date RA (1991) Nodulation success and persistence of recommended inoculum
phaseoli. Appl Environ Microbiol 62:2767–2772 strains for subtropical and tropical forage legumes in Northern Australia.
Chatel DL, Greenwood RM, Parker CA (1968) Saprophytic competence as an Soil Biol Biochem 23:533–541
important characteristic in the selection of Rhizobium for inoculation. In: Date RA (1997) The contribution of R and D on root nodule bacteria to future
Proceedings of the Ninth International Congress of Soil Science Transactions, cultivars of tropical forage legumes. Trop Grassl 31:350–354
Vol. 2 (pp. 65–73). Adelaide. Date RA, Halliday J (1987) Collection, isolation, cultivation and maintenance of
Chaverra MH, Graham PH (1992) Cultivar variation in traits affecting early rhizobia. In: Elkan GH (ed) Symbiotic nitrogen fixation technology. Marcel
nodulation of common bean. Crop Sci 32:1432–1436 Dekker, New York, pp 1–27
Chen WX, Yan GH, Li JL (1988) Numeric taxonomic study of the fast-growing Davis EO, Evans IJ, Johnston AWB (1988) Identification of nodX, a gene that
soybean rhizobia and a proposal that Rhizobium fredii be assigned to allows Rhizobium leguminosarum biovar viceae strain TOM to nodulate
Sinorhizobium gen. nov. Int J Syst Bacteriol 38:392–397 Afghanistan peas. Mol Gen Genet 212:531–535
Chen WX, Li GS, Qi YL, Wang ET, Yuan HL, Li JL (1991) Rhizobium huakuii sp. De Bruijn FJ (1992) Use of repetitive (repetitive extragenic element and entero-
nov. isolated from the root nodules of Astragalus sinicus. Int J Syst Bacteriol bacterial repetitive intergenic consensus) sequences and the polymerase
41:275–280 chain reaction to fingerprint the genomes of Rhizobium meliloti isolates
Chen W, Wang E, Wang S, Li Y, Chen X, Li Y (1995) Characteristics of Rhizobium and other soil bacteria. Appl Environ Microbiol 58:2180–2187
tianshanense sp. nov., a moderately and slowly growing root nodule bacte- de Lajudie P, Willems A, Pot B, Dewettinck D, Maestrojuan G, Neyra M, Collins
rium isolated from an arid saline environment in Xinjiang, People’s Republic MD, Dreyfus B, Kersters K, Gillis M (1994) Polyphasic taxonomy of rhizobia:
of China. Int J Syst Bacteriol 45:153–159 emendation of the genus Sinorhizobium and description of Sinorhizobium
Chen WX, Tan ZY, Gao JL, Li Y, Wang ET (1997) Rhizobium hainanense sp. nov., meliloti comb. nov., Sinorhizobium saheli sp. nov., and Sinorhizobium teranga
isolated from tropical legumes. Int J Syst Bacteriol 47:870–873 sp. nov. Int J Syst Bacteriol 44:715–733
Chen WM, Laevens S, Lee TM, Coenye T, De Vos P, Mergeay M, Vandamme de Lajudie P, Laurent-Fulele E, Willems A, Torck U, Coopman R, Collins MD,
P (2001) Ralstonia taiwanensis sp. nov., isolated from root nodules of Kersters K, Dreyfus B, Gillis M (1998a) Allorhizobium undicola gen. nov., sp.
Mimosa species and sputum of a cystic fibrosis patient. Int J Syst nov., nitrogen-fixing bacteria that efficiently nodulate Neptunia natans in
Evol Microbiol 51:1729–1735 Senegal. Int J Syst Bacteriol 48:1277–1290
Chen WM, Moulin L, Bontemps C, Vandamme P, Bena G, Boivin-Masson de Lajudie P, Willems A, Nick G, Moreira F, Molouba F, Hoste B, Torck U, Neyra
C (2003) Legume symbiotic nitrogen fixation by beta-proteobacteria is M, Collins MD, Lindstrom K, Dreyfus B, Gillis M (1998b) Characterization
widespread in nature. J Bacteriol 185:7266–7272 of tropical tree rhizobia and description of Mesorhizobium plurifarium
Chen WM, de Faria SM, Straliotto R, Pitard RM, Simoes-Araujo JL, Chou JH, sp. nov. Int J Syst Bacteriol 48:369–382
Chou YJ, Barrios E, Prescott AR, Elliot GN, Sprent JI, Young JPW, James EK de Lajudie P, Willems A, Nick G, Mohamed SH, Torck U, Coopman R, Filali-
(2005) Proof that Burkholderia forms effective symbioses with legumes: Maltouf A, Kersters K, Dreyfus B, Lindstrom K, Gillis M (1999)
A study of novel Mimosa-nodulating strains from South America. Appl Agrobacterium bv. 1 strains isolated from nodules of tropical legumes. Syst
Environ Microbiol 71:7461–7471 Appl Microbiol 22:119–132
Chen WM, James EK, Coenye T, Chou JH, Barrios E, de Faria SM, Elliott GN, De Ley J (1968) DNA base composition and hybridization in the taxonomy of
Sheu SY, Sprent JI, Vandamme P (2006) Burkholderia mimosarum sp. nov., phytopathogenic bacteria. Ann Rev Phytopathol 6:63–90
isolated from root nodules of Mimosa spp. from Taiwan and South America. De Ley J, Rassell A (1965) DNA base composition, flagellation and taxonomy of
Int J Syst Evol Microbiol 56:1847–1851 the genus Rhizobium. J Gen Microbiol 41:85–91
Chen WM, de Faria SM, James EK, Elliott GN, Lin KY, Chou JH, Sheu SY, Dean DR, Jacobson MR (1992) Biochemical genetics of nitrogenase. In: Stacey G,
Cnockaert M, Sprent JI, Vandamme P (2007) Burkholderia nodosa Burris B, Evans HJ (eds) Biological nitrogen fixation. Chapman and Hall,
sp. nov., isolated from root nodules of the woody Brazilian legumes New York, pp 763–834
Mimosa bimucronata and Mimosa scabrella. Int J Syst Evol Microbiol Demezas DH (1998) Fingerprinting bacterial genomes using restriction fragment
57:1055–1059 length polymorphisms. In: de Bruijn FJ, Lupski JR, Weinstock GM (eds)
Chen WM, de Faria SM, Chou JH, James EK, Elliott GN, Sprent JI, Bontemps C, Bacterial genomes: physical structure and analysis. Chapman and Hall, New
Young JPW, Vandamme P (2008) Burkholderia sabiae sp. nov., isolated York, pp 383–398
from root nodules of Mimosa caesalpiniifolia. Int J Syst Evol Microbiol Demezas DH, Reardon TB, Watson JM, Gibson AH (1991) Genetic diversity
58:2174–2179 among Rhizobium leguminosarum bv trifolii strains revealed by allozyme
Chen WM, Zhu WF, Bontemps C, Young JP, Wei GH (2010) Mesorhizobium and restriction length polymorphisms analyses. Appl Environ Microbiol
alhagi sp. nov., isolated from root nodules of the wild legume Alhagi 59:1702–1708
sparsifolia. Int J Syst Evol Microbiol 60:958–962 Demezas DH, Reardon TB, Strain SR, Watson JM, Gibson AH (1995) Diversity
Chen WM, Zhu WF, Bontemps C, Young JPW, Wei GH (2011) Mesorhizobium and genetic structure of a natural population of Rhizobium leguminosarum
camelthorni sp. nov., isolated from Alhagi sparsifolia. Int J Syst bv trifolii isolated from Trifolium subterraneum. Mol Ecol 4:209–220
Evol Microbiol 61:574–579 Denarie J, Debelle F, Prome JC (1996) Rhizobium lipo-chitooligosaccharide
Collins SL, Knapp AK, Briggs JM, Blair JM, Steinauer EM (1998) Modulation nodulation factors signaling molecules mediating recognition and morpho-
of diversity by grazing and mowing in native tallgrass prairie. Science genesis. Annu Rev Biochem 65:503–535
280:745–747 Diatloff A (1977) Ecological studies of root nodule bacteria introduced into field
Cregan PB, Keyser HH (1986) Host restriction of nodulation by Bradyrhizobium environments. 6: antigenic and symbiotic stability in Lotononis rhizobia
japonicum strain USDA 123 in soybean. Crop Sci 26:911–916 over a 12 year period. Soil Biol Biochem 9:85–88
Djordjevic MA, Sargent CL, Innes RW, Kuempel PL, Rolfe BG (1985) Host-range O-acetylation of the Rhizobium leguminosarum lipo-oligosaccharide nodu-
genes also affect competitiveness in Rhizobium trifolii. In: Evans HJ, lation factor. Mol Microbiol 10:351–360
Bottomley PJ, Newton WE (eds) Nitrogen fixation research progress. Fisher RF, Long SR (1989) DNA footprint analysis of the transcriptional activator
Martinus Nijhoff, Dordrecht, p 117 proteins Nod-D1 and Nod-D3 on inducible nod gene promoters. J Bacteriol
Djordjevic MA, Gabriel DW, Rolfe BG (1987a) Rhizobium – the refined parasite of 171:5492–5502
legumes. Ann Rev Phytopathol 25:145–168 Frank B (1889) Über die Pariseten in den Wurzelanschwellungen der
Djordjevic MA, Redmond JW, Batley M, Rolfe BG (1987b) Clovers secrete specific Papilionaceen. Botanische Zeitung 37:377–388, 393–400
phenolic compounds which either stimulate or repress nod gene expression Franssen HJ, Nap J-P, Bisseling T (1992) Nodulins in root nodule development.
in Rhizobium trifolii. EMBO J 6:173–1179 In: Stacey G, Burris B, Evans HJ (eds) Biological nitrogen fixation. Chapman
Dobereiner J, Urquiaga S, Boddey RM (1995) Alternatives for nitrogen nutrition and Hall, New York, pp 598–624
of crops in tropical agriculture. Fertilizer Res 42:339–346 Fred EB, Baldwin IL, McCoy E (1932) Root nodule bacteria and leguminous
Dockendorff TC, Sanjuan J, Grob P, Stacey G (1994) NolA represses nod plants. University of Wisconsin, Madison, WI, p 343
gene expression in Bradyrhizobium japonicum. Mol Plant-Microbe Interact Freiberg C, Fellay R, Bairoch A, Broughton WJ, Rosenthal A, Perret X (1997)
7:596–602 Molecular basis of symbiosis between Rhizobium and legumes. Nature
Downie JA (1998) Functions of rhizobial nodulation genes. In: Spaink HP, 387:394–401
Kondorosi A, Hooykaas PJJ (eds) The Rhizobiaceae. Kluwer, Dordrecht, Gage DJ, Bobo T, Long SR (1996) Use of green fluorescent protein to visualize the
pp 387–402 early events of symbiosis between Rhizobium meliloti and alfalfa (Medicago
Downie JA (2005) Legume haemoglobins: Symbiotic nitrogen fixation needs sativa). J Bacteriol 178:7159–7166
bloody nodules. Curr Biol 15:196–198 Galibert F, Finan TM, Long SR, Puhler A, Abola P, Ampe F, Barloy-Hubler F,
Downie JA (2010) The roles of extracellular proteins, polysaccharides and signals Barnett MJ, Becker A, Boistard P, Bothe G, Boutry M, Bowser L, Buhrmester
in the interactions of rhizobia with legume roots. FEMS Microbiol Rev J, Cadieu E, Capela D, Chain P, Cowie A, Davis RW, Dreano S, Federspiel NA,
34:150–170 Fisher RF, Gloux S, Godrie T, Goffeau A, Golding B, Gouzy J, Gurjal M,
Doyle JJ (1994) Phylogeny of the legume family: An approach to understanding Hernandez-Lucas I, Hong A, Huizar L, Hyman RW, Jones T, Kahn D, Kahn
the origins of nodulation. Annu Rev Ecol Syst 25:325–349 ML, Kalman S, Keating DH, Kiss E, Komp C, Lelaure V, Masuy D, Palm C,
Doyle JJ, Doyle JL (1997) In: Legocki A et al (eds) Biological fixation of nitrogen Peck MC, Pohl TM, Portetelle D, Purnelle B, Ramsperger U, Surzycki R,
for ecology and sustainable agriculture, vol G39, NATO ASI series. Springer, Thebault P, Vandenbol M, Vorholter FJ, Weidner S, Wells DH, Wong K, Yeh
New York, pp 307–312 KC, Batut J (2001) The composite genome of the legume symbiont
Dreyfus B, Garcia JL, Gillis M (1988) Characterization of Azorhizobium Sinorhizobium meliloti. Science 293:668–672
caulinodans gen. nov., sp. nov., a stem-nodulating nitrogen-fixing bacterium Gao JL, Turner SL, Kan FL, Wang ET, Tan ZY, Qiu YH, Terefework Z, Young JPW,
isolated from Sesbania rostrata. Int J Syst Bacteriol 38:89–98 Lindstrom K, Chen WX (2004) Mesorhizobium septentrionale sp. nov. and
Dudman WF (1977) Serological methods and their application to dinitrogen- Mesorhizobium temperatum sp. nov., isolated from Astragalus adsurgens
fixing organisms. In: Treatise on dinitrogen-fixation, vol 4. Wiley, New York, growing in the northern regions of China. Int J Syst Evol Microbiol
pp 487–508 54:2003–2012
Duodu S, Bhuvaneswari TV, Stokkermans TJW, Peters NK (1999) A positive role Garcia-Fraile P, Rivas R, Willems A, Peix A, Martens M, Martinez-Molina E,
for Rhizobitoxine in Rhizobium-legume symbiosis. Mol Plant Microbe Inter- Mateos PF, Velazquez E (2007) Rhizobium cellulosilyticum sp. nov. isolated
act 12:1082–1089 from sawdust of Populus alba. Int J Syst Evol Microbiol 57:844–848
Dupuy N, Willems A, Pot B, Dewettinck D, Vandenbruaene I, Maestrojuan G, Gault RR, Schwinghamer EA (1993) Direct isolation of Bradyrhizobium
Dreyfus B, Kersters K, Collins MD, Gillis M (1994) Phenotypic and geno- japonicum from soil. Soil Biol Biochem 25:1161–1166
typic characterization of bradyrhizobia nodulating the leguminous tree Gherna RL (1994) Culture preservation. In: Gerhardt P et al (eds) Methods for
Acacia albida. Int J Syst Bacteriol 44:461–473 general and molecular bacteriology. ASM Press, Washington, DC
Dye M, Skot L, Mytton LR, Harrison SP, Dooley JJ, Cresswell A (1995) A study of Gibson AH, Date RH, Ireland JA, Brockwell J (1976) A comparison of compet-
Rhizobium leguminosarum biovar trifolii populations from soil extracts using itiveness and persistence among five strains of Rhizobium trifolii. Soil Biol
randomly amplified polymorphic DNA profiles. Can J Microbiol 41:336–344 Biochem 8:395–401
Eaglesham A, Seaman B, Ahmad H, Hassouna S, Ayanaba A, Mulongoy K (1981) Giraud E, Moulin L, Vallenet D, Barbe V, Cytryn E, Avarre JC, Jaubert M,
High temperature tolerant ‘‘cowpea’’ rhizobia. In: Gibson AH, Newton WE Simon D, Cartieaux F, Prin Y, Bena G, Hannibal L, Fardoux J,
(eds) Current perspectives in nitrogen fixation. Australian Academy of Kojadinovic M, Vuillet L, Lajus A, Cruveiller S, Rouy Z, Mangenot S,
Science, Canberra, p 346 Segurens B, Dossat C, Franck WL, Chang WS, Saunders E, Bruce D,
Eardly BD, Materon LA, Smith NH, Johnson DA, Rumbaugh MD, Selander RK Richardson P, Normand P, Dreyfus B, Pignol D, Stacey G, Emerich D,
(1990) Genetic structure of natural populations of the nitrogen-fixing bac- Verméglio A, Médigue C, Sadowsky M (2007) Legumes symbioses: Absence
terium Rhizobium meliloti. Appl Environ Microbiol 56:187–194 of Nod genes in photosynthetic bradyrhizobia. Science 316:1307–1312
Elkan GH, Bunn CR (1994) The rhizobia. In: Balows A et al (eds) The pro- Glazebrook J, Walker GC (1989) A novel exopolysaccharide can function in place
karyotes. Springer, Berlin, pp 2197–2213, http://www.prokaryotes. of the calcofluor-binding exopolysaccharide in nodulation of alfalfa by
com&rcub Rhizobium meliloti. Cell 56:661–672
Ellis WR, Ham GE, Schmidt EL (1984) Persistence and recovery of Rhizobium Glazebrook J, Ichige A, Walker GC (1993) A Rhizobium meliloti homolog of the
japonicum inoculum in a field soil. Agron J 76:573–576 Escherichia coli peptide-antibiotic transport protein SbmA is essential for
Faucher C, Camut S, Denarie J, Truchet G (1989) The nodH and nodQ host range bacteroid development. Genes Dev 7:1485–1497
genes of Rhizobium meliloti behave as avirulence genes in R. leguminosarum Glynn P, Higgins P, Squartini A, O’Gara F (1985) Strain identification in Rhizo-
bv viceae and determine changes in the production of plant-specific extra- bium trifolii using DNA restriction analysis, plasmid DNA profiles, and
cellular signals. Mol Plant-Microbe Interact 2:291–300 intrinsic antibiotic resistances. FEMS Microbiol Lett 30:177–182
Fellay R, Perret X, Viprey V, Broughton WJ (1995) Organization of host-inducible Gomez M, Silva N, Hartman A, Sagardoy M, Catroux G (1997) Evaluation of
transcripts on the symbiotic plasmid of Rhizobium sp. NGR234. Mol commercial soybean inoculants from Argentina World. J Microbiol
Microbiol 16:657–667 Biotechnol 13:167–173
Finan TM, Wood JM, Jordon DC (1983) Symbiotic properties of C4-dicarboxylic González V, Santamarı́a RI, Bustos P, Hernández-González I, Medrano-Soto A,
acid transport mutants of Rhizobium leguminosarum. J Bacteriol Moreno-Hagelsieb G, Janga SC, Ramı́rez MA, Jiménez-Jacinto V, Collado-
154:1403–1413 Vides J, Dávila G (2006) The partitioned Rhizobium etli genome: Genetic and
Firmin JL, Wilson KE, Carlson RW, Davies AE, Downie JA (1993) Resistance to metabolic redundancy in seven interacting replicons. Proc Natl Acad Sci
nodulation of cv. Afghanistan peas is overcome by nodX, which mediates an USA 103:3834–3839
Göttfert M (1993) Regulation and function of rhizobial nodulation genes. FEMS Heberlein GT, De Ley J, Tyjtgat R (1967) Deoxyribonucleic acid homology and
Microbiol Lett 104:39–64 taxonomy of Agrobacterium, Rhizobium and Chromobacterium. J Bacteriol
Göttfert M, Horvath B, Kondorosi E, Putnoky P, Rodriguez-Quniones F, 94:116–124
Kondorosi A (1986) At least two nodD genes are necessary for efficient Hellriegel H, Wilfarth H (1888) Untersuchungen ueber die stickstoffnahrung der
nodulation of alfalfa by Rhizobium meliloti. J Mol Biol 191:411–420 Gramineen und Leguminsen. Beilageheft su der Zeitschrift des Verein
Göttfert M, Webber J, Hennecke H (1988) Induction of a nodA-lacZ fusion in Rubenzucker-Industrie Detchen Reichs Unknown, p 234
Bradyrhizobium japonicum by an isoflavone. J Plant Physiol 132:394–397 Heron DS, Ersek T, Krishan HB, Pueppke SG (1989) Nodulation mutants of
Göttfert M, Grob P, Hennecke H (1990) Proposed regulatory pathway encoded by Rhizobium fredii USDA 257. Mol Plant Microbe Interact 2:4–10
the nodV and the nodW genes, determinants of host specificity in Herridge DF, Roughley RJ (1975) Variation in colony characteristics and symbi-
Bradyrhizobium japonicum. Proc Natl Acad Sci USA 87:2680–2684 otic effectiveness of Rhizobium. J Appl Bacteriol 38:19–27
Göttfert M, Holzhäuser D, Bäni D, Hennecke H (1992) Structural and functional Hirsch PR (1996) Population dynamics of indigenous and genetically modified
analysis of two different nodD genes in Bradyrhizobium japonicum USDA110. rhizobia in the field. New Phytol 133:159–171
Mol Plant Microbe Interact 5:257–265 Hofer AW (1941) A characterization of Bacterium radiobacter (Beijerinck and
Gotz R, Evans IJ, Downie JA, Johnston AWB (1985) Identification of host-range Van Delden) Lohnis. J Bacteriol 41:193–224
DNA which allows Rhizobium leguminosarum strain TOM to nodulate cv. Holmes B, Popoff M, Kiredjian M, Kersters K (1988) Ochrobactrum anthropi gen.
Afghanistan peas. Mol Gen Genet 201:296–300 nov., sp. nov. from human clinical specimens and previously known as group
Graham PH (1964) The application of computer techniques to the taxonomy of Vd. Int J Syst Bacteriol 38:406–416
the root-nodule bacteria of legumes. J Gen Microbiol 35:511–517 Hombrecher G, Brewin NJ, Johnston AWB (1981) Linkage of genes for nitroge-
Graham PH (1976) Identification and classification of root nodule bacteria. In: nase and nodulation ability on plasmids in Rhizobium leguminosarum and R.
Nutman PS (ed) Symbiotic nitrogen fixation in plants. Cambridge Univer- phaseoli. Mol Gen Genet 182:133–136
sity Press, Cambridge, UK, pp 99–112 Hombrecher G, Gotz R, Dibb NJ, Downie JA, Johnston AWB, Brewin NJ
Graham PH, Viteri SE, Mackie F, Vargas AAT, Palacios A (1982) Variation in acid (1984) Cloning and mutagenesis of nodulation genes from Rhizobium
soil tolerance among strains of Rhizobium phaseoli. Field Crops Res 5:121–128 leguminosarum TOM, a strain with extended host range. Mol Gen Genet
Graham PH, Sadowsky MJ, Keyser HH, Barnet YM, Bradley RS, Cooper JE, De 194:293–298
Ley J, Jarvis BWD, Roslycky EB, Strijdom BW, Young JPW (1991) Proposed Hong G-F, Burn JE, Johnston AWB (1987) Evidence that DNA involved in the
minimal standards for the description of new genera and species of root-and expression of nodulation (nod) genes in Rhizobium binds to the regulatory
stem-nodulating bacteria. Int J Syst Bacteriol 41:582–587 gene nodD. Nucleic Acids Res 15:9677–9690
Graham PH, Draeger KJ, Ferrey ML, Conroy MJ, Hammer BE, Martinez E, Honma MA, Ausubel FM (1987) Rhizobium meliloti has three functional
Aarons SR, Quinto C (1994) Acid pH tolerance in strains of Rhizobium copies of the nodD symbiotic regulatory element. Proc Natl Acad Sci
and Bradyrhizobium, and initial studies on the basis for pH tolerance of USA 84:8558
Rhizobium tropici UMR1899. Can J Microbiol 40:198–207 Horvath B, Kondorosi E, John M, Schmidt J, Torok I, Gyorgypal Z, Barabas I,
Graham PH, Sadowsky MJ, Tighe SW, Thompson JA, Date RA, Howieson JG, Wieneke U, Schell J, Kondorosi A (1986) Organization, structure, and
Thomas R (1995) Differences among strains of Bradyrhizobium in fatty acid– symbiotic function of Rhizobium meliloti nodulation genes determining
methyl ester analysis. Can J Microbiol 41:1038–1042 host specificity for alfalfa. Cell 46:335–343
Graham PH, Ballen KL, Montealegre C, Jones R, Fischer B, Luque E (1999) Horvath B, Bachem CW, Schell J, Kondorosi A (1987) Host-specific regulation of
Characterization of rhizobia associated with Dalea spp. in natural prairies nodulation genes in Rhizobium is mediated by a plant-signal, interacting
and revegetation areas in Minnesota. In: Martinez E, Hernandez G (eds) with the nodD gene products. EMBO J 6:841–848
Highlights in nitrogen fixation research. Plenum Press, New York Hou BC, Wang ET, Li Y, Jia RZ, Chen WF, Gao Y, Dong RJ, Chen WX (2009) Rhi-
Gu CT, Wang ET, Tian CF, Han TX, Chen WF, Sui XH, Chen WX (2008) Rhizo- zobium tibeticum sp. nov., a symbiotic bacterium isolated from Trigonella
bium miluonense sp. nov., a symbiotic bacterium isolated from Lespedeza archiducis-nicolai (Sirj.) Vassilcz. Int J Syst Evol Microbiol 59:3051–3057
root nodules. Int J Syst Evol Microbiol 58:1364–1368 Howieson JG, Ewing MA (1989) Annual species of Medicago differ greatly in their
Guan SH, Chen WF, Wang ET, Lu YL, Yan XR, Zhang XX, Chen WX (2008) ability to nodulate on acid soils. Aust J Agric Res 40:843–850
Mesorhizobium caraganae sp. nov., a novel rhizobial species nodulated with Howieson JG, Ewing MA, Dantuono MF (1988) Selection for acid tolerance in
Caragana spp. in China. Int J Syst Evol Microbiol 58:2646–2653 Rhizobium meliloti. Plant Soil 105:179–188
Guar YD, Lowther WL (1980) Distribution, symbiotic effectiveness and fluores- Hubber A, Vergunst AC, Sullivan JT, Hooykaas PJ, Ronson CW (2004) Symbiotic
cent antibody reaction of naturalized populations of Rhizobium trifolii in phenotypes and translocated effector proteins of the Mesorhizobium loti
Otago, New Zealand soils NZ. J Agric Res 23:529–532 strain R7A VirB/D4 type IV secretion system. Mol Microbiol 54:561–574
Hadri A-E, Spaink HP, Bisseling T, Brewin NJ (1998) Diversity of root nodulation Humphrey BA, Vincent JM (1965) The effect of calcium nutrition on the
and rhizobial infection processes. In: Spaink HP, Kondorosi A, Hooykaas PJJ production of diffusible antigens by Rhizobium trifolii. J Gen Microbiol
(eds) The Rhizobiaceae. Kluwer, Dordrecht, pp 348–360 41:109–118
Hagen MJ, Hamrick JL (1996) Population level processes in Rhizobium Hungria M, Andrade DD, Colozzi A, Balota EL (1997) Interactions among soil
leguminosarum bv trifolii: The role of founder effects. Mol Ecol 5:707–714 microorganisms and bean and maize grown in monoculture or
Hahn M, Hennecke H (1988) Cloning and mapping of a novel nodulation region intercropped. Pesq Agropec Brasil 32:807–818
from Bradyrhizobium japonicum by genetic complementation of a deletion Innes RW, Kuempel PL, Plazinski J, Canter-Cremers H, Rolfe BG, Djordjevic MA
mutant. Appl Environ Microbiol 54:55–61 (1985) Plant factors induce expression of nodulation and host-range genes in
Hall A, Clark N (1995) Coping with change, complexity and diversity in agricul- R. trifolii. Mol Gen Genet 201:426–432
ture-the case of rhizobial inoculants in Thailand. World Dev 23:1601–1614 Islam MS, Kawasaki H, Muramatsu Y, Nakagawa Y, Seki T (2008) Bradyrhizobium
Ham GE (1978) Interactions of Glycine max and Rhizobium japonicum. In: iriomotense sp. nov., isolated from a tumor-like root of the legume Entada
Summerfield RJ, Bunting AH (eds) Advances in legume science. Royal koshunensis from Iriomote Island in Japan. Biosci Biotechnol Biochem
Botanic Gardens Kew, London, pp 289–296 72:1416–1429
Han TX, Han LL, Wu LJ, Chen WF, Sui XH, Gu JG, Wang ET, Chen WX (2008a) Itakura M, Saeki K, Omori H, Yokoyama T, Kaneko T, Tabata S, Ohwada T, Tajima
Mesorhizobium gobiense sp. nov. and Mesorhizobium tarimense sp. nov., S, Uchiumi T, Honma K, Fujita K, Iwata H, Saeki Y, Hara Y, Ikeda S, Eda S,
isolated from wild legumes growing in desert soils of Xinjiang, China. Int Mitsui H, Minamisawa K (2009) Genomic comparison of Bradyrhizobium
J Syst Evol Microbiol 58:2610–2618 japonicum strains with different symbiotic nitrogen-fixing capabilities and
Han TX, Wang ET, Wu JL, Chen WF, Gu JG, Gu CT, Tian CF, Chen WX (2008b) other Bradyrhizobiaceae members. ISME J 3:326–339
Rhizobium multihospitium sp. nov., isolated from multiple legume species Jarvis BDW, Tighe SW (1994) Rapid identification of Rhizobium species based on
native of Xinjiang, China. Int J Syst Evol Microbiol 58:1693–1699 cellular fatty acid analysis. Plant Soil 161:31–41
Jarvis BDW, Pankhurst CE, Patel JJ (1982) Rhizobium loti, a new species of legume Knosel D (1962) Prufung von Bakterien auf Fahigkeit zur Sternbildung. Zentralbl
root nodule bacteria. Int J Syst Bacteriol 32:378–380 Bakteriol Parasitenkd Infedtionskr Hyg II Abt 116:79–100
Jarvis BWD, Ward LJH, Slade EA (1989) Expression by soil bacteria of nodulation Kobayashi H, Broughton WJ (2008) Fine-tuning of symbiotic genes in rhizobia:
genes from Rhizobium leguminosarum bv. trifolii. Appl Environ Microbiol flavonoid signal transduction cascade. In: Dilworth MJ, James EK, Sprent JI,
55:1426–1434 Newton WE (eds) Nitrogen-fixing leguminous symbioses. Springer,
Jarvis BDW, Sivakumaran S, Tighe SW, Gillis M (1996) Identification of Dordrecht, pp 117–152
Agrobacterium and Rhizobium species based on cellular fatty acid composi- Kondorosi E, Gyuris J, Schmidt J, John M, Duda E, Schell J, Kondorosi A (1988)
tion. Plant Soil 184:143–158 Positive and negative control of nodulation genes in Rhizobium meliloti
Jarvis BDW, van Berkum P, Chen WX, Nour SM, Fernandez M, Cleyet-Marel JC, strain 41. In: Verma DPS, Palacios R (eds) Molecular microbe-plant inter-
Gillis M (1997) Transfer of Rhizobium loti, Rhizobium huakuii, Rhizobium actions. APS Press, St. Paul, p 73
cicer,i Rhizobium mediterraneum and Rhizobium tianshansense to Kondorosi E, Gyuris J, Schmidt J, John M, Duda E, Hoffman B, Schell J,
Mesorhizobium gen. nov. Int J Syst Bacteriol 47:895–898 Kondorosi A (1989) Positive and negative control of nod gene expression
Jiang J, Gu BH, Albright LM, Nixon BT (1989) Conservation between coding and in Rhizobium meliloti is required for optimal nodulation. EMBO
regulatory elements of Rhizobium meliloti and Rhizobium leguminosarum dct J 8:1331–1340
genes. J Bacteriol 171:44–53 Kondorosi E, Pierre M, Cren M, Haumann U, Buire M, Hoffman B, Schell J,
Jones KM, Kobayashi H, Davies BW, Taga ME, Walker GC (2007) Invasion of the Kondorosi A (1991) Identification of nolR, a negatively transacting factor
plant host by symbiotic rhizobial bacteria: molecular clues from controlling the nod regulon in Rhizobium meliloti. J Mol Biol 222:885–896
Sinorhizobium meliloti and the plant Medicago truncatula. Nat Rev Microbiol Kosslak RM, Bookland R, Barkei J, Paaren HE, Applebaum ER (1987) Induction
5:619–633 of Bradyrhizobium japonicum common nod genes by isoflavones isolated
Jordan DC (1982) Transfer of Rhizobium japonicum Buchanan 1980 to from Glycine max. Proc Natl Acad Sci USA 84:7428–7432
Bradyrhizobium gen. nov., a genus of slow-growing, root nodule bacteria Krause A, Doerfel A, Göttfert M (2002) Mutational and transcriptional analysis of
from leguminous plants. Int J Syst Bacteriol 32:136–139 the type III secretion system of Bradyrhizobium japonicum. Mol Plant
Jordan DC (1984) Family III, Rhizobiaceae Conn 1938. In: Krieg NR, Holt JG Microbe Interact 15:1228–1235
(eds) Bergey’s manual of systematic bacteriology, 1st edn. Williams and Krishnan HB, Pueppke SG (1991) nolC, a Rhizobium fredii gene involved in
Wilkins, Baltimore, pp 234–242, http://www.cme.msu.edu/bergeys/&rcub cultivar-specific nodulation of soybean, shares homology with a heat-shock
Josey DP, Beynon JL, Johnston AWB, Beringer JE (1979) Strain identification in gene. Mol Microbiol 5:737–745
Rhizobium using intrinsic antibiotic resistance. J Appl Bacteriol 46:343–350 Kucey RMN, Hynes MF (1989) Populations of Rhizobium leguminosarum biovars
Jourand P, Giraud E, Bena G, Sy A, Willems A, Gillis M, Dreyfus B, de Lajudie phaseoli and viciae in fields after bean or pea in rotation with nonlegumes.
P (2004) Methylobacterium nodulans sp. nov., for a group of aerobic, facul- Can J Microbiol 35:661–667
tatively methylotrophic, legume root-nodule-forming and nitrogen-fixing Kumar Rao JVDK, Dart PJ, Usha Khan M (1982) Cowpea group Rhizobium in
bacteria. Int J Syst Evol Microbiol 54:2269–2273 soils of the semiarid tropics. In: Graham PH, Harris SC (eds) BNF technol-
Judd AK, Schneider M, Sadowsky MJ, de Bruijn FJ (1993) Use of repetitive ogy for tropical agriculture. CIAT, Cali. Colombia, pp 291–295
sequences and the polymerase chain reaction technique to classify genetically Kuykendall LD, Saxena B, Devine TE, Udell SE (1992) Genetic diversity in
related Bradyrhizobium japonicum serocluster 123 strains. Appl Environ Bradyrhizobium japonicum Jordan 1982 and a proposal for Bradyrhizobium
Microbiol 59:1702–1708 elkanii sp. nov. Can J Microbiol 38:501–505
Kaminski PA, Batutu J, Boistard P (1998) A survey of symbiotic nitrogen fixation Labes G, Ulrich A, Lentzch P (1996) Influence of bovine slurry deposition on the
by Rhizobia. In: Spaink HP, Kondorosi A, Hooykaas PJJ (eds) The structure of nodulating Rhizobium leguminosarum bv viciae soil populations
Rhizobiaceae. Kluwer, Dordrecht, pp 432–460 in a natural habitat. Appl Environ Microbiol 62:1717–1722
Kaneko T, Nakamura Y, Sato S, Asamizu E, Kato T, Sasamoto S, Watanabe A, Lange RT (1961) Nodule bacteria associated with the indigenous Leguminosae of
Idesawa K, Ishikawa A, Kawashima K, Kimura T, Kishida Y, Kiyokawa C, South Western Australia. J Gen Microbiol 26:351–359
Kohara M, Matsumoto M, Matsuno A, Mochizuki Y, Nakayama S, Nakazaki Lee KB, De Backer P, Aono T, Liu CT, Suzuki S, Suzuki T, Kaneko T, Yamada M,
N, Shimpo S, Sugimoto M, Takeuchi C, Yamada M, Tabata S (2000) Com- Tabata S, Kupfer DM, Najar FZ, Wiley GB, Roe B, Binnewies TT, Ussery DW,
plete genome structure of the nitrogen-fixing symbiotic bacterium D’Haeze W, Herder JD, Gevers D, Vereecke D, Holsters M, Oyaizu H (2008)
Mesorhizobium loti. DNA Res 7:331–338 The genome of the versatile nitrogen fixer Azorhizobium caulinodans
Kaneko T, Nakamura Y, Sato S, Minamisawa K, Uchiumi T, Sasamoto S, ORS571. BMC Genomics 9:271
Watanabe A, Idesawa K, Iriguchi M, Kawashima K, Kohara M, Matsumoto Lerouge P, Roche P, Faucher C, Maillet F, Truchet G, Prome JC, Denarie J (1990)
M, Shimpo S, Tsuruoka H, Wada T, Yamada M, Tabata S (2002) Complete Symbiotic host specificity of Rhizobium meliloti is determined by
genomic sequence of nitrogen-fixing symbiotic bacterium Bradyrhizobium a sulphated and acylated glucosamine oligosaccharide signal. Nature
japonicum USDA110. DNA Res 9:189–197 344:781–784
Keatinge JDH, Beck DP, Materon LA, Yurtsever N, Karuc K, Altuntas S (1995) The Leung K, Strain SR, de Bruijn FJ, Bottomley PJ (1994) Genotypic and
role of rhizobial diversity in legume crop productivity in the west Asian phenotypic comparisons of chromosomal types within an indigenous soil
highlands. 4: Rhizobium ciceri. Expl Agric 31:501–507 population of Rhizobium leguminosarum bv. trifolii. Appl Environ Microbiol
Kennedy C (1989) The genetics of nitrogen fixation. In: Hopwood DA, Chater KE 60:416–426
(eds) Genetics of bacterial diversity. Academic, New York, pp 107–127 Lewin A, Rossenberg C, Meyer H, Wong ACH, Nelson L, Manen JF, Stanley J,
Kijne JW (1992) The Rhizobium infection process. In: Stacey G, Burris B, Dowling DN, Denarie J, Broughton WJ (1987) Multiple host-specificity loci
Evans HJ (eds) Biological nitrogen fixation. Chapman and Hall, New York, of the broad host-range Rhizobium sp. NGR234 selected using the widely
pp 349–398 compatible legume Vigna unguiculata. Plant Mol Biol 8:447–459
Kinkle BK, Schmidt EL (1991) Transfer of the pea symbiotic plasmid pJB5JI in Lewis-Henderson WR, Djordjevic MA (1991a) A cultivar specific interaction
non-sterile soil. Appl Environ Microbiol 57:3264–3269 between Rhizobium leguminosarum bv. trifolii and subterranean clover is
Kinzig AP, Socolow RH (1994) Is nitrogen fertilizer use nearing a balance? Reply conditioned by nodM, other bacterial cultivar specificity genes and a single
Physics Today 47:24–35 recessive host gene. J Bacteriol 173:2791–2799
Kirchner O (1896) Die wurzelnollchen der sojabohne. Beitrage Biologie Pflanzen Lewis-Henderson WR, Djordjevic MA (1991b) nodT, a positively-acting cultivar
Cohn’s 7:213–223 specificity determinant controlling nodulation of Trifolium subterraneum by
Kishinevsky BD, Jones DG (1987) Enzyme-linked immunosorbent assay Rhizobium leguminosarum bv. trifolii. Plant Mol Biol 16:515–526
(ELISA) for the detection and identification of Rhizobium strains. In: Elkan Li LY, Chen WF, Li HL, Wang ET, Chen WX (2009) Rhizobium alkalisoli sp. nov.,
GH (ed) Symbiotic nitrogen fixation technology. Marcel Dekker, New York, isolated from Caragana intermedia growing in saline-alkaline soils in the
pp 157–184 north of China. Int J Syst Evol Microbiol 59:3006–3011
Li QQ, Wang ET, Chang YL, Zhang YZ, Zhang YM, Sui XH, Chen WF, Chen WX Mantelin S, Fischer-Le Saux M, Zakhia F, Bena G, Bonneau S, Jeder H, de Lajudie
(2010) Ensifer (Sinorhizobium) sojae sp. nov., isolated from root nodules of P, Cleyet-Marel JC (2006) Emended description of the genus Phyllobacterium
Glycine max grown in saline-alkaline soils in Hebei province of China. Int and description of four novel species associated with plant roots:
J Syst Evol Microbiol 61(Pt 8):1981–1988 Phyllobacterium bourgognense sp. nov., Phyllobacterium ifriqiyense sp. nov.,
Lie TA (1978) Symbiotic specialization in pea plants: The requirement of specific Phyllobacterium leguminum sp. nov. and Phyllobacterium brassicacearum sp.
Rhizobium strains for peas from Afghanistan. Ann Appl Biol 88:462–465 nov. Int J Syst Evol Microbiol 56:827–839
Lie TA, Winarno R, Timmermans PCJM (1978) Rhizobium strains isolated from Marshall KC (1964) Survival of root nodule bacteria in dry soils exposed to high
wild and cultivated legumes: suppression of nodulation by a non-nodulating temperatures. Aust J Agric Res 15:273–281
Rhizobium strain. In: Loutit MW, Miles JAR (eds) Microbial ecology. Martinez E, Romero D, Palacios R (1990) The Rhizobium genome. Crit Rev Plant
Springer, Berlin Sci 9:59–93
Lie TA, Goktan D, Pijnenborg J, Anlarsal E (1987) Co-evolution of the legume- Martinez-Romero E, Caballero-Mellado J (1996) Rhizobium phylogenies and
Rhizobium association. Plant Soil 100:171–181 bacterial genetic diversity. Crit Rev Plant Sci 15:113–140
Lin DX, Chen WF, Wang FQ, Hu D, Wang ET, Sui XH, Chen WX (2009) Rhizo- Martinez-Romero E, Rosenblueth M (1990) Increased bean Phaseolus vulgaris L.
bium mesosinicum sp. nov., isolated from root nodules of three different nodulation competitiveness of genetically modified strains. Appl Environ
legumes. Int J Syst Evol Microbiol 59:1919–1923 Microbiol 56:2384–2388
Lin DX, Wang ET, Tang H, Han TX, He YR, Guan SH, Chen WX (2008) Shinella Martinez-Romero E, Segovia L, Mercante FM, Franco AA, Graham P, Pardo MA
kummerowiae sp. nov., a symbiotic bacterium isolated from root nodules of (1991) Rhizobium tropici, a novel species nodulating Phaseolus vulgaris L.
the herbal legume Kummerowia stipulacea. Int J Syst Evol Microbiol beans and Leucaena sp. trees. Int J Syst Bacteriol 41:417–426
58:1409–1413 McDermott TR, Graham PH, Brandwein DM (1987) Viability of Bradyrhizobium
Lindstrom K (1989) Rhizobium galegae, a new species of legume root nodule japonicum bacteroids. Arch Microbiol 148:100–106
bacteria. Int J Syst Bacteriol 39:365–367 McInnes A, Thies JE, Abbott LK, Howieson JG (2004) Structure and diversity
Lindstrom K, Lipsanen P, Kaijalainen S (1990) Stability of markers used for among rhizobial strains, populations and communities–a review. Soil Biol
identification of two Rhizobium galegae inoculant strains after five years in Biochem 36:1295–1308
the field. Appl Environ Microbiol 56:444–450 Meinhardt LW, Krishnan HB, Balatti PA, Pueppke SG (1993) Molecular cloning and
Lloret L, Ormeno-Orrillo E, Rincon R, Martinez-Romero J, Rogel-Hernandez characterization of a Sym-plasmid locus that regulates cultivar-specific nodu-
MA, Martinez-Romero E (2007) Ensifer mexicanus sp. nov. a new species lation of soybean by Rhizobium fredii USDA 257. Mol Microbiol 9:17–27
nodulating Acacia angustissima (Mill.) Kuntze in Mexico. Syst Appl Mendes IC, Bottomley PJ (1998) Distribution of a population of Rhizobium
Microbiol 30:280–290 leguminosarum bv trifolii among different size classes of soil aggregates.
Loh J, Stacey G (2003) Nodulation gene regulation in Bradyrhizobium japonicum: Appl Environ Microbiol 64:970–975
A unique integration of global regulatory circuits. Appl Environ Microbiol Merabet C, Martens M, Mahdhi M, Zakhia F, Sy A, Le Roux C, Domergue O,
69:10–17 Coopman R, Bekki A, Mars M, Willems A, de Lajudie P (2010) Multilocus
Lohrke SM, Orf JH, Martı́nez-Romero E, Sadowsky MJ (1995) Host-controlled sequence analysis of root nodule isolates from Lotus arabicus (Senegal), Lotus
restriction of nodulation by Bradyrhizobium japonicum strains in serogroup creticus, Argyrolobium uniflorum and Medicago sativa (Tunisia) and descrip-
110. Appl Environ Microbiol 61:2378–2383 tion of Ensifer numidicus sp. nov. and Ensifer garamanticus sp. nov. Int J Syst
Long SR (1989) Rhizobium-legume nodulation: Life together in the underground. Evol Microbiol 60:664–674
Cell 56:203–214 Mergaert P, van Montagu M, Prome J-C, Holsters M (1993) Three unusual
Long SR, Egelhoff T, Fisher RF, Jacobs TW, Mulligan JT (1985) Fine structure studies modifications, a D-arabinosyl, an N-methyl, and a carbamoyl group, are
of R. meliloti nodDABC genes. In: Evans HJ, Bottomley PJ, Newton WE (eds) present on the Nod factors of Azorhizobium caulinodans strain ORS571. Proc
Nitrogen fixation research progress. Martinus Nijhoff, Boston, pp 87–94 Natl Acad Sci USA 90:1551–1555
Long S, Reed JW, Himawan J, Walker GC (1999) Genetic analsysis of a clsuter of Merrick MJ (1992) Regulation of nitrogen fixation genes in free-living and
genes required for synthesis of the calcofluor-binding exopolysaccharide of symbiotic bacteria. In: Stacey G, Burris RH, Evans HJ (eds) Biological
Rhizobium meliloti. J Bacteriol 170:4239–4248 nitrogen fixation. Chapman and Hall, New York, pp 835–876
Lorda GS, Balatii A (1996) Production of high cell concentrations of Rhizobium Michiels J, Dombrecht B, Vermeiren N, Xi CW, Luyten E, Vanderleyden J (1998)
and Bradyrhizobium. In: Balatti AP, Freire JR (eds) Legume inoculants: Phaseolus vulgaris L. is a non-selective host for nodulation. FEMS Microbiol
Selection and characterization of strains – production, use, and manage- Ecol 26:193–205
ment. Editorial Kingraf Buenos Aires, Argentina Moffett ML, Colwell RR (1968) Adansonian analysis of the Rhizobiaceae. J Gen
Louvrier P, Laguerre G, Amarger N (1995) Semi selective medium for isolation of Microbiol 51:245–266
Rhizobium leguminosarum from soils. Soil Biol Biochem 27:919–924 Moreira FMS, Cruz L, de Faria SM, Marsh T, Martinez-Romero E, de Oliveira
Lu YL, Chen WF, Han LL, Wang ET, Zhang XX, Chen WX, Han SZ Pedrosa F, Pitard RM, Young JPW (2006) Azorhizobium doebereinerae sp.
(2009) Mesorhizobium shangrilense sp. nov., isolated from root nodules of nov. microsymbiont of Sesbania virgata (Caz.) Pers. Syst Appl Microbiol
Caragana spp. Int J Syst Evol Microbiol 59:3012–3018 29:197–206
Lupwayi NZ, Olsen PE, Sande ES, Keyser HH, Collins MM, Singleton PW, Moulin L, Munive A, Dreyfus B, Boivin-Masson C (2001) Nodulation of legumes
Rice WA (2000) Inoculant quality and its evaluation. Field Crops Res by members of the b-subclass of Proteobacteria. Nature 411:948–950
65:259–270 Mulligan JT, Long SR (1985) Induction of Rhizobium meliloti nodC expression by
Ma W, Guinel FC, Glick BR (2003) Rhizobium leguminosarum biovar viciae plant exudate requires nodD. Proc Natl Acad Sci USA 82:6609–6613
1-aminocyclopropane-1-carboxylate deaminase promotes nodulation of Murray JD (2011) Invasion by invitation: Rhizobial infection in legumes. Mol
pea plants. Appl Environ Microbiol 69:4396–4402 Plant Microbe Interact 24:631–639
MacLean AM, Finan TM, Sadowsky MJ (2007) Genomes of the symbiotic nitro- Naisbitt T, James E, Sprent JI (1992) The evolutionary significance of the legume genus
gen-fixing bacteria of legumes. Plant Physiol 144:615–622 Chamaecrista, as determined by nodule structure. New Phytol 122:487–492
Madrzak CJ, Golinska B, Kroliczak J, Pudelko K, Lazewska D, Lampka B, Nakagawa Y, Sakane T, Yokota A (1996) Transfer of ‘Pseudomonas riboflavina’
Sadowsky MJ (1995) Diversity among field populations of Bradyrhizobium (Foster 1944), a gram-negative, motile rod with long-chain 3-hydroxy fatty
japonicum in Poland. Appl Environ Microbiol 61:1194–1200 acids, to Devosia riboflavina gen. nov., sp. nov., nom. rev. Int J Syst Bacteriol
Maier RJ (1986) Biochemistry, regulation, and genetics of hydrogen oxidation in 46:16–22
Rhizobium japonicum CRC. Crit Rev Biotechnol 3:17–38 Nandasena KG, O’Hara GW, Tiwari RP, Willems A, Howieson JG (2009)
Makkar NS, Casida LE (1987) Cupriavidus necator gen. nov., sp. nov. – Mesorhizobium australicum sp. nov. and Mesorhizobium opportunistum sp.
A nonobligate bacterial predator of bacteria in soil. Int J Syst Bacteriol nov. isolated from Biserrula pelecinus L. growing in Australia. Int J Syst
37:323–326 Evol Microbiol 59:2140–2157
Nick G, de Lajudie P, Eardly BD, Suomalainen S, Paulin L, Zhang X, Gillis M, Price NJ, Relic B, Talmont F, Lewin A, Prome D, Pueppke SG, Maillet F, Denarie J,
Lindstrom K (1999) Sinorhizobium arboris sp. nov. and Sinorhizobium Broughton WJ (1992) Broad host range Rhizobium species strain NGR234
kostiense sp. nov., isolated from leguminous trees in Sudan and Kenya. Int secretes a family of carbamoylated and fucosylated nodulation signals that
J Syst Bacteriol 49:1359–1368 are O-acetylated or sulphated. Mol Microbiol 6:3575–3584
Nieuwkoop AJ, Banfalvi Z, Deshmane N, Gerhold D, Schell M, Sirotkin K, Stacey Pueppke SG (1996) The genetic and biochemical basis for nodulation of legumes
G (1987) A locus encoding host range is linked to the common nodulation by rhizobia. Crit Rev Biotechnol 16:1–51
genes of Bradyrhizobium japonicum. J Bacteriol 169:2631–2638 Pueppke SG, Broughton WJ (1999) Rhizobium strain NGR234 and R. fredii
Niner BM, Hirsch AM (1998) How many Rhizobium genes, in addition to nod, USDA257 share exceptionally broad nested host ranges. Mol Plant Microbe
nif/fix, and exo, are needed for nodule development and function. Symbiosis Interact 12:293–318
24:51–102 Quan ZX, Bae HS, Baek JH, Chen WF, Im WT, Lee ST (2005) Rhizobium
Noel KD, Diebold RJ, Cava JR, Brink BA (1988) Rhizobial purine and pyrimidine daejeonense sp. nov., isolated from a cyanide treatment bioreactor. Int
auxotrophs: nutrient supplementation, genetic analysis, and the J Syst Evol Microbiol 55:2543–2549
symbiotic requirement for the novo purine biosynthesis. Arch Micobiol Ramirez-Bahena MH, Garcia-Fraile P, Peix A, Valverde A, Rivas R, Igual JM,
149:499–506 Mateos PF, Martinez-Molina E, Velazquez E (2008) Revision of the
Nour SM, Fernandez MP, Normand P, Cleyet-Marel JC (1994) Rhizobium ciceri taxonomic status of the species Rhizobium leguminosarum (Frank 1879)
sp. nov., consisting of strains that nodulate chickpeas (Cicer arietinum L.). Frank 1889AL, Rhizobium phaseoli Dangeard 1926AL and Rhizobium
Int J Syst Bacteriol 44:511–522 trifolii Dangeard 1926AL. R. trifolii is a later synonym of R. leguminosarum.
Nuti MP, Lepidi AA, Prakash RK, Shilperoot RA, Cannon FC (1979) Evidence for Reclassification of the strain R. leguminosarum DSM 30132
nitrogen fixation (nif) genes on indigenous Rhizobium plasmids. Nature (=NCIMB 11478) as Rhizobium pisi sp. nov. Int J Syst Evol Microbiol
(London) 282:533–535 58:2484–2490
Ogasawara M, Suzuki T, Mutoh I, Annapurna K, Arora NK, Nishimura Y, Ramirez-Bahena MH, Peix A, Rivas P, Camacho M, Rodriguez-Navarro DN,
Maheshwari DK (2003) Sinorhizobium indiaense sp. nov. and Sinorhizobium Camacho P, Mateos PF, Martinez-Molina E, Velazquez E (2009)
abri sp. nov. isolated from tropical legumes, Sesbania rostrata and Abrus Bradyrhizobium pachyrhizi sp. nov. and Bradyrhizobium jicamae sp. nov.
precatorius, respectively. Symbiosis 34:53–68 isolated from effective nodules of Pachyrhizus erosus. Int J Syst
Oger P, Petit A, Dessaux Y (1997) Genetically engineered plants producing opines Evol Microbiol 59:1929–1934
alter their biological environment. Nat Biotechnol 15:369–372 Rao JR, Fenton M, Jarvis BDW (1994) Symbiotic plasmid transfer in Rhizobium
Olson ER, Sadowsky MJ, Verma DPS (1985) Identification of genes involved in leguminosarum biovar trifolii and competition between the inoculant
the Rhizobium-legume symbiosis by Mu-dI(kan, lac)-generated transcrip- strain ICMP2163 and transconjugant soil bacteria. Soil Biol Biochem
tion fusions. Nat Biotechnol 3:143–149 26:339–351
Paffetti G, Scotti C, Gnocchi S, Francelli S, Bazzicalupo M (1996) Genetic Rasolomampianina R, Bailly X, Fetiarison R, Rabevohitra R, Béna G, Ramaroson
diversity of an Italian Rhizobium meliloti population from different Medicago L, Raherimandimby M, Moulin L, De Lajudie P, Dreyfus B, Avarre JC
sativa varieties. Appl Environ Microbiol 62:2279–2285 (2005) Nitrogen-fixing nodules from rose wood legume trees (Dalbergia
Parke D, Ornston LN (1984) Nutritional diversity of Rhizobiaceae revealed by spp.) endemic to Madagascar host seven different genera belonging to
auxanography. J Gen Microbiol 130:1743–1750 alpha- and beta-Proteobacteria. Mol Ecol 14:4135–4146
Parker CA (1957) Evolution of nitrogen-fixing symbiosis in higher plants. Nature Ratcliff WC, Kadam SV, Denison RF (2008) Poly-3-hydroxybutyrate supports
179:593–594 survival and reproduction in starving rhizobia. FEMS Microbiol Ecol
Parker CA (1968) On the evolution of symbiosis in legumes. In: Jensen HL (ed) 65:391–399
Festskrift til Hans Laurits Jensen. Gadgaard Nielsens Bogtrykkeri, Lemvig, Reeve W, Chain P, O’Hara G, Ardley J, Nandesena K, Bräu L, Tiwari R, Malfatti S,
pp 107–115 Kiss H, Lapidus A, Copeland A, Nolan M, Land M, Hauser L, Chang YJ,
Patt TE, Cole GC, Hanson RS (1976) Methylobacterium, a new genus of faculta- Ivanova N, Mavromatis K, Markowitz V, Kyrpides N, Gollagher M, Yates R,
tively methylotrophic bacteria. Int J Syst Bacteriol 26:226–229 Dilworth M, Howieson J (2010) Complete genome sequence of the Medicago
Peng G, Yuan Q, Li H, Zhang W, Tan Z (2008) Rhizobium oryzae sp. nov., isolated microsymbiont Ensifer (Sinorhizobium) medicae strain WSM419. Stand
from the wild rice Oryza alta. Int J Syst Evol Microbiol 58:2158–2163 Genomic Sci 28:77–86
Pérez-Mendoza D, Domı́nguez-Ferreras A, Muñoz S, Soto MJ, Olivares J, Brom S, Relic B, Talmont F, Kopcinska J, Golinowski W, Prome J-C, Broughton WJ
Girard L, Herrera-Cervera JA, Sanjuán J (2004) Identification of functional (1993) Biological activity of Rhizobium sp. NGR234 Nod factors on
mob regions in Rhizobium etli: evidence for self-transmissibility of the sym- Macroptilium atropurpureum. Mol Plant Microbe Interact 6:764–774
biotic plasmid pRetCFN42d. J Bacteriol 186:5753–5761 Ren DW, Chen WF, Sui XH, Wang ET, Chen WX (2011) Rhizobium vignae sp.
Perez-Ramirez NO, Rogel MA, Wang E, Castellanos JZ, Martinez-Romero nov., a symbiotic bacterium isolated from multiple legume species nov. Int
E (1998) Seeds of Phaseolus vulgaris bean carry Rhizobium etli. FEMS J Syst Evol Microbiol 61:580–586
Microbiol Ecol 26:289–296 Rice WA, Olsen PE, Leggett ME (1995) Co-culture of Rhizobium meliloti and
Peters NK, Frost JW, Long SR (1986) A plant flavone, luteolin, induces expression a phosphorus solubilizing fungus (Penicillium bilaii) in sterile peat. Soil Biol
of Rhizobium meliloti nodulation genes. Science 233:977–979 Biochem 27:703–705
Phillips DA, Joseph CM, Yang GP, Martinez-Romero E, Sanborn JR, Richardson AE, Simpson RJ (1989) Acid-tolerance and symbiotic effectiveness of
Volpin H (1999) Identification of lumichrome as a Sinorhizobium enhancer Rhizobium trifolii associated with a Trifolium subterraneum L. based pasture
of alfalfa root respiration and shoot growth. Proc Natl Acad Sci USA growing in an acid soil. Soil Biol Biochem 21:87–95
96:12275–12280 Richardson AE, Viccars LA, Watson JM, Gibson AH (1995) Differentiation of
Pinero D, Martinez E, Selander RK (1988) Genetic diversity and relationships Rhizobium strains using the polymerase chain reaction with random and
among isolates of Rhizobium leguminosarum biovar phaseoli. Appl Environ directed primers. Soil Biol Biochem 27:515–524
Microbiol 54:2825–2832 Risal CP, Yokoyama T, Ohkama-Ohtsu N, Djedidi S, Sekimoto H (2010) Genetic
Postma J, Hok-a-Hin CH, van Veen JA (1990) Role of microniches in protecting diversity of native soybean bradyrhizobia from different topographical
introduced Rhizobium leguminosarum biovar trifolii against competition and regions along the southern slopes of the Himalayan Mountains in Nepal.
predation in soil. Appl Environ Microbiol 56:495–502 Syst Appl Mirobiol 33:416–425
Prévost D, Bordeleau LM, Caudry-Reznick S, Schulman HM, Antoun H (1987) Rivas R, Velázquez E, Willems A, Vizcaı́no N, Subba-Rao NS, Mateos PF, Gillis M,
Characteristics of rhizobia isolated from three legumes indigenous to the Dazzo FB, Martı́nez-Molina E (2002) A new species of Devosia that forms
high arctic: Astragalus alpinus, Oxytropis maydelliana, and Oxytropis a unique nitrogen-fixing root nodule symbiosis with the aquatic legume
arctobia. Plant Soil 98:313–324 Neptunia natans. Appl Environ Microbiol 68:5217–5222
Rivas R, Willems A, Subba-Rao NS, Mateos PF, Dazzo FB, Kroppenstedt RM, Sanginga N, Danso SKA, Mulongoy K, Ojeifo AA (1994) Persistence and
Martinez-Molina E, Gillis M, Vizcaino N (2003) Description of Devosia recovery of introduced Rhizobium 10 years after inoculation on Leucaena
neptuniae sp. nov. that nodulates and fixes nitrogen in symbiosis with leucocephala grown on an alfisol in southwestern Nigeria. Plant Soil
Neptunia natans, an aquatic legume from India. Syst Appl Microbiol 159:199–204
26:47–53 Sanjuan J, Olivares J (1989) Implication of nifA in regulation of genes located
Roberts GP, Leps WT, Silver LE, Brill WJ (1980) Use of two dimensional poly- a Rhizobium meliloti cryptic plasmid that effect nodulation efficiency.
acrylamide gel electrophoresis to identify and classify Rhizobium strains. J Bacteriol 171:4154–4161
Appl Environ Microbiol 39:414–422 Sanjuan J, Carlson RW, Spaink HP, Bhat UP, Barbour WM, Glushka J, Stacey
Roche P, Debelle F, Maillet F, Lerouge P, Faucher C, Truchet G, Dénarié J, Prome G (1992) A 2-O-methylfucose moiety is present in the lipo-oligosaccharide
JC (1991) Molecular basis of symbiotic host specificity in Rhizobium meliloti: nodulation signal of Bradyrhizobium japonicum. Proc Natl Acad Sci
nodH and nodPQ genes encode the sulfation of lipo-oligosaccharides. Cell USA 89:8789–8793
67:1131–1143 Savka MA, Farrand SK (1997) Modification of rhizobacterial populations by
Roest HP, Goosenderoo L, Wijffelman CA, de Maagd RA, Lugtenberg BJJ engineering bacterium utilization of a novel plant source. Nat Biotechnol
(1995) Outer membrane protein changes during bacteroid development 15:363–368
are independent of nitrogen fixation and differ between indeterminate and Sawada H, Ieki H, Oyaizu H, Matsumoto S (1993) Proposal for rejection of
determinate nodulating host plants of Rhizobium leguminosarum. Mol Plant Agrobacterium tumefaciens and revised descriptions for the genus
Microbe Interact 8:14–22 Agrobacterium and for Agrobacterium radiobacter and Agrobacterium
Rolfe BG, Innes RW, Schofield PR, Watson JW, Sargent CL, Kuempel PL, Plazinski rhizogenes. Int J Syst Bacteriol 43:694–702
J, Canter-Cremers H, Djordjevic MA (1985) Plant-secreted factors influence Sawada H, Kuykendall LD, Young JM (2003) Changing concepts in the system-
the expression of R. trifolii nodulation and host-range genes. In: Evans HJ, atics of bacterial nitrogen-fixing legume symbionts. J Gen Appl Microbiol
Bottomly PJ, Newton WE (eds) Nitrogen fixation research progress. 49:155–179
Martinus Nijhoff Publishers, Boston, pp 79–85 Schlaman HRM, Phillips DA, Kondorosi E (1998) Genetic organization and
Rome S, Fernandez MP, Brunel B, Normand P, Cleyet-Marel JC (1996) transcriptional regulation of rhizobial nodulation genes. In: Spaink HP,
Sinorhizobium medicae sp. nov. isolated from annual Medicago spp. Int Kondorosi A, Hooykaas PJJ (eds) The Rhizobiacea. Kluwer, Dordrecht,
J Syst Bacteriol 46:972–980 pp 351–386
Rossbach S, Kulpa DA, Rossbach U, De Bruijn FJ (1994) Molecular and genetic Schmeisser C, Liesegang H, Krysciak D, Bakkou N, Le Quéré A, Wollherr A,
characterization of the rhizopine catabolism (mocabrc) genes of Rhizobium Heinemeyer I, Morgenstern B, Pommerening-Röser A, Flores M, Palacios R,
meliloti l5–30. Mol Gen Genet 245:11–24 Brenner S, Gottschalk G, Schmitz RA, Broughton WJ, Perret X,
Rostas K, Kondorosi E, Horvath B, Simoncsits A, Kondorosi A (1986) Conserva- Strittmatter AW, Streit WR (2009) Rhizobium sp. strain NGR234 possesses
tion of extended promoter regions of nodulation genes in Rhizobium. Proc a remarkable number of secretion systems. Appl Environ Microbiol
Natl Acad Sci USA 83:1757–1761 75:4035–4045
Ruvkin GB, Ausubel FM (1980) Interspecies homology of nitrogenase genes. Proc Schmidt EL, Bankole RO, Bohlool BB (1968) Fluorescent antibody approach to
Natl Acad Sci USA 77:191–195 study rhizobia in soil. J Bacteriol 95:1987–1992
Rynne FG, Glenn AR, Dilworth MJ (1994) Effect of mutations in aromatic Scholla MH, Elkan GH (1984) Rhizobium fredii sp. nov., a fast-growing species
catabolism on the persistence and competitiveness of Rhizobium that effectively nodulates soybeans. Int J Syst Bacteriol 34:484–486
leguminosarum bv. trifolii. Soil Biol Biochem 26:703–710 Schortemeyer M, Santruckova H, Sadowsky MJ (1997) Relationship between root
Sadowsky MJ (1994) Microbial DNA fingerprinting and restriction fragment length density and soil microorganisms in the rhizospheres of white clover
length polymorphism analysis. In: Weaver RW, Angle JS, Bottomley P (eds) and perennial ryegrass. Comm Soil Sci Plant Anal 28:1675–1682
Methods of soil analysis, chemical and microbiological properties of soils. Schultze M, Kondorosi A (1998) Regulation of symbiotic root nodule develop-
ASA-SSSA, Madison, pp 647–664 ment. Ann Rev Genet 32:33–57
Sadowsky MJ, Graham PH (1998) Soil biology of the Rhizobiaceae. In: Spaink HP, Schultze M, Quiclet-Sire B, Kondorosi E, Virelizier H, Glushka N, Endre G, Gero
Kondorosi A, Hooykaas PJJ (eds) The Rhizobiacea. Kluwer, Dordrecht, D, Kondorosi A (1992) Rhizobium meliloti produces a family of sulphated
pp 155–172 lipooligosaccharides exhibiting different degrees of plant host specificity.
Sadowsky MJ, Hur HG (1998) Use of endogenous repeated sequences to finger- Proc Natl Acad Sci USA 89:192–196
print bacterial genomic DNA. In: Lupski JR, Weinstock G, de Bruijn FJ (eds) Segovia L, Pinero D, Palacios R, Martinez-Romero E (1991) Genetic structure of
Bacterial genomes: structure and analysis. Chapman and Hall, New York a soil population of nonsymbiotic Rhizobium leguminosarum. Appl Environ
Sadowsky MJ, Bohlool BB, Keyser HH (1987a) Serological relatedness of Rhizo- Microbiol 57:426–433
bium fredii to other rhizobia and to the bradyrhizobia. Appl Environ Segovia L, Young JP, Martinez-Romero E (1993) Reclassification of American
Microbiol 53:1785–1789 Rhizobium leguminosarum biovar phaseoli type I strains as Rhizobium etli sp.
Sadowsky MJ, Tully RE, Cregan PB, Keyser HH (1987b) Genetic diversity in nov. Int J Syst Bacteriol 43:374–377
Bradyrhizobium japonicum serogroup 123 and its relation to genotype spe- Sen D, Weaver RW (1988) Nitrogenase acetylene activities of isolated peanut and
cific nodulation of soybeans. Appl Environ Microbiol 53:2624–2630 cowpea bacteroids at optimal oxygen availability and comparison with whole
Sadowsky MJ, Olson ER, Foster VE, Kosslak RM, Verma DPS (1988) Two host- nodule activities. J Exp Bot 35:785–789
inducible genes of Rhizobium fredii and the characterization of the inducing Shearman CA, Rossen L, Johnston AWB, Downie JA (1986) The Rhizobium
compound. J Bacteriol 170:171–178 leguminosarum nodulation gene nodF encodes a polypeptide similar to acyl
Sadowsky MJ, Cregan PB, Rodriguez-Quinones F, Keyser HH (1990) Microbial carrier protein and is regulated by nodD plus a factor in pea root exudate.
influence on gene-for-gene interactions in legume-Rhizobium symbioses. EMBO J 5:647
Plant and Soil 129:53–60 Singleton PW, Stockinger KR (1983) Compensation against ineffective nodula-
Sadowsky MJ, Cregan PB, Gottfert M, Sharma A, Gerhold D, Rodriguez- tion in soybean (Glycine max). Crop Sci 23:69–72
Quniones F, Keyser HH, Henneke H, Stacey G (1991) The Bradyrhizobium Singleton PW, Tavares JW (1986) Inoculation response of legumes in relation to
japonicum nolA gene and its involvement in the genotype-specific nodula- the number and effectiveness of indigenous Rhizobium populations. Appl
tion of soybeans. Proc Natl Acad Sci USA 88:637–641 Environ Microbiol 51:1013–1018
Sagan M, Gresshoff PM (1996) Developmental mapping of nodulation events in Singleton P, Keyser HH, Sande E (2002) Development and evaluation of
pea (Pisum sativum L.) using supernodulating plant genotypes and bacterial liquid inoculants. In: Herridge D (ed) Inoculants and nitrogen fixation
variability reveals both plant and Rhizobium control of nodulation regula- of legumes in vietnam, ACIAR Proceedings 109e. ACIAR, Canberra,
tion. Plant Sci 117:167–179 pp 52–66
Slattery JF, Coventry DR (1993) Variation of soil populations of Rhizobium Thomas PM, Golly KF, Zyskind JW, Virginia RA (1994) Variation of clonal,
leguminosarum bv. trifolii and the occurrence of inoculant rhizobia in nod- mesquite-associated rhizobial and bradyrhizobial populations from surface
ules of subterranean clover after pasture renovation in north eastern Victo- and deep soils by symbiotic gene region restriction fragment length poly-
ria. Soil Biol Biochem 25:1725–1730 morphism and plasmid profile analysis. Appl Environ Microbiol 60:1146–
Smith RS (1988) Nitragin bran inoculants technical bulletin. Lipha Tech, Mil- 1153
waukee WI, p 9 Toledo I, Lloret L, Martinez-Romero E (2003) Sinorhizobium americanus sp. nov.,
Smith RS (1992) Legume inoculant formulation and application. Can J Microbiol a new Sinorhizobium species nodulating native Acacia spp. in Mexico. Syst
38:485–492 Appl Microbiol 26:54–64
Smith RS, del Rio Escurra GA (1982) Soybean inoculant types and rates evaluated Tong Z, Sadowsky MJ (1993) A selective medium for the isolation and quantifi-
under dry and irrigated field conditions. J Agr Univ Puert Rico 66:241–249 cation of Bradyrhizobium japonicum and Bradyrhizobium elkanii strains
Smith RS, Ellis MA, Smith RE (1981) Effect of Rhizobium japonicum inoculation from soils and inoculants. Appl Environ Microbiol 60:581–586
rates on soybean nodulation in a tropical soil. Agron J 73:505–508 Truchet G, Roche P, Lerouge P, Vasse J, Camut S (1991) Sulphated lipo-
So RB, Ladha JK, Young JPW (1994) Photosynthetic symbionts of Aeschynomene oligosaccharide signals of the symbiotic prokaryote Rhizobium meliloti elicit
form a cluster with bradyrhizobia on the basis of fatty acid and r RNA root nodule organogenesis on the host plant Medicago sativa. Nature
analysis. Int J Syst Bacteriol 44:392–403 351:670–673
Somasegaran P, Hoben HJ (1994) Handbook for rhizobia. Springer, New York, Trujillo ME, Willems A, Abril A, Planchuelo AM, Rivas R, Ludeña D, Mateos PF,
p 450 Martı́nez-Molina E, Velázquez E (2005) Nodulation of Lupinus albus by strains
Souza V, Eguiarte L, Avila G, Cappello G, Gallardo C, Montoya J, Pinero D (1994) of Ochrobactrum lupini sp. nov. Appl Environ Microbiol 71:1318–1327
Genetic structure of Rhizobium etli biovar phaseoli associated with wild and Unkovich MJ, Pate JS, Sandford P (1997) Nitrogen fixation by annual legumes in
cultivated bean plants (Phaseolus vulgaris and Phaseolus coccineus in Morelos Australian Mediterranean agriculture. Aust J Agric Res 48:267–293
Mexico). Appl Environ Microbiol 60:1260–1268 Valverde A, Velazquez E, Fernandez-Santos F, Vizcaino N, Rivas R, Mateos PF,
Spaink HP (1995) The molecular basis of infection and nodulation by rhizobia – Martinez-Molina E, Igual JM, Willems A (2005) Phyllobacterium trifolii sp.
the ins and outs of sympathogenesis. Annu Rev Phytopathol 33:345–368 nov., nodulating Trifolium and Lupinus in Spanish soils. Int J Syst
Spaink HP, Wijffelman CA, Pees E, Okker RJH, Lugtenberg BJJ (1987) Rhizobium Evol Microbiol 55:1985–1989
nodulation gene nodD as a determinant of host specificity. Nature Valverde A, Igual JM, Peix A, Cervantes E, Velazquez E (2006) Rhizobium
328:337–340 lusitanum sp. nov. a bacterium that nodulates Phaseolus vulgaris. Int J Syst
Spaink HP, Sheeley DM, van Brussel AAN, Glushka J, York WS, Tak T, Geiger O, Evol Microbiol 56:2631–2637
Kennedy EP, Reinholdt VN, Lugtenberg BJJ (1991) A novel highly unsatu- van Berkum P, Beyene D, Vera FT, Keyser HH (1995) Variability among Rhizo-
rated fatty acid moiety of lipooligosaccharide signals determines host spec- bium strains originating from nodules of Vicia faba. Appl Environ Microbiol
ificity of Rhizobium. Nature 354:124–130 61:2649–2653
Sparrow SD, Cochran VL, Sparrow EB (1995) Dinitrogen fixation by seven van Berkum P, Beyene D, Bao G, Campbell TA, Eardly BD (1998) Rhizobium
legume crops in Alaska. Agron J 87:34–41 mongolense sp. nov. is one of three rhizobial genotypes identified which
Squartini A, Struffi P, Doring H, Selenska-Pobell S, Tola E, Giacomini A, nodulate and form nitrogen-fixing symbioses with Medicago ruthenica
Vendramin E, Velazquez E, Mateos PF, Martı́nez-Molina E, Dazzo FB, Casella [(L.) Ledebour]. Int J Syst Bacteriol 48:13–22
S, Nuti MP (2002) Rhizobium sullae sp. nov. (formerly ‘Rhizobium hedysari’), van Brussel A, Recourt K, Pees E, Spaink HP, Tak T, Wijffelman C, Kijne J,
the root-nodule microsymbiont of Hedysarum coronarium L. Int J Syst Lugtenberg BB (1990) A biovar specific signal of Rhizobium leguminosarum
Evol Microbiol 52:1267–1276 bv. viceae induces increased nodulation gene-inducing activity in root exu-
Stephens JHG, Rask H (2000) Inoculant production and formulation. Field date of Vicia sativa sub-sp. nigra. J Bacteriol 172:5394–5401
Crops Res 65:249–258 van der Drift KMGM, Olsthoorn MMA, Brull LP, Blok-Tip L, Thomas-Oates JE
Strain SR, Leung K, Whittam TS, de Bruijn FJ, Bottomley PJ (1994) Genetic (1998) Mass spectrometric analysis of lipo-chitin oligosaccharides signal
structure of Rhizobium leguminosarum biovar trifolii and viciae populations molecules mediating the host-specific legume-Rhizobium symbiosis. Mass
found in two Oregon soils under different plant communities. Appl Environ Spectrom Rev 17:75–95
Microbiol 60:2772–2778 Van Egaraat AWSM (1975) The possible role of homoserine in the development of
Strain SR, Whittam TS, Bottomley PJ (1995) Analysis of genetic structure in soil Rhizobium leguminosarum in the rhizosphere of pea seedlings. Plant Soil
populations of Rhizobium leguminosarum recovered from the USA and UK. 42:381–386
Mol Ecol 4:105–114 van Rhijn P, Vanderleyden J (1995) The Rhizobium-plant symbiosis. Microbiol
Sugawara M, Okazaki S, Nukui N, Ezura H, Mitsui H, Minamisawa K (2006) Rev 59:124–142
Rhizobitoxine modulates plant-microbe interactions by ethylene inhibition. Vance CP (1998) Legume symbiotic nitrogen fixation: agronomic aspects. In:
Biotechnol Adv 24:382–388 Spaink HP et al (eds) The Rhizobiaceae. Kluwer, Dordrecht, pp 509–530
Sullivan JT, Ronson CW (1998) Evolution of rhizobia by acquisition of a 500-kb Vandamme P, Goris J, Chen WM, de Vos P, Willems A (2002) Burkholderia
symbiosis island that integrates into a phe-tRNA gene. Proc Natl Acad Sci tuberum sp. nov. and Burkholderia phymatum sp. nov., nodulate the roots
USA 95:5145–5149 of tropical legumes. Syst Appl Microbiol 25:507–512
Sullivan JT, Patrick HN, Lowther WL, Scott DB, Ronson CW (1995) Nodulating Velazquez E, Igual JM, Willems A, Fernandez MP, Munoz E, Mateos PF, Abril A,
strains of Rhizobium loti arise through chromosomal symbiotic gene transfer Toro N, Normand P, Cervantes E, Gillis M, Martinez-Molina E (2001)
in the environment. Proc Natl Acad Sci USA 92:8985–8989 Mesorhizobium chacoense sp. nov. a novel species that nodulates Prosopis
Sutton WD (1983) Nodule development and senescence. In: Broughton WJ (ed) alba in the Chaco Arido region (Argentina). Int J Syst Evol Microbiol
Nitrogen fixation, vol 3, Legumes. Oxford University Press, Oxford, pp 144–212 51:1011–1021
Sy A, Giraud E, Jourand P, Garcia N, Willems A, de Lajudie P, Prin Y, Neyra M, Versalovic J, de Bruijn FJ, Lupksi JR (1998) Repetive sequence-based PCR
Gillis M, Boivin-Masson C, Dreyfus B (2001) Methylotrophic (rep-PCR) DNA fingerprinting of bacterial genomes. In: de Bruijn FJ,
Methylobacterium bacteria nodulate and fix nitrogen in symbiosis with Lupski JR, Weinstock GM (eds) Bacterial genomes: physical structure and
legumes. J Bacteriol 183:214–220 analysis. Chapman and Hall, New York, pp 437–456
Tan ZY, Kan FL, Peng GX, Wang ET, Reinhold-Hurek B, Chen WX (2001) Rhizo- Vidal C, Chaintreuil C, Berge O, Maure L, Escarre J, Bena G, Brunel B, Cleyet-
bium yanglingense sp. nov., isolated from arid and semi-arid regions in Marel JC (2009) Mesorhizobium metallidurans sp. nov., a metal-resistant
China. Int J Syst Evol Microbiol 51:909–914 symbiont of Anthyllis vulneria growing on metallicolous soil in Languedoc
Thies JE, Singleton PW, Bohlool BB (1991) Influence of the size of indigenous France. Int J Syst Evol Microbiol 59:850–855
rhizobial populations on establishment and symbiotic performance of intro- Vincent JM (1970) A manual for the practical study of root-nodule bacteria,
duced rhizobia on field-grown legumes. Appl Environ Microbiol 57:19–28 vol 15, IBP handbook. Blackwell, Oxford, p 164
Vinuesa P, Leon-Barrios M, Silva C, Willems A, Jarabo-Lorenzo A, Perez-Galdona Xu LM, Ge C, Cui Z, Li J, Fan H (1995) Bradyrhizobium liaoningense sp. nov.,
R, Werner D, Martinez-Romero E (2005) Bradyrhizobium canariense sp. isolated from the root nodules of soybeans. Int J Syst Bacteriol 45:706–711
nov., an acid-tolerant endosymbiont that nodulates endemic genistoid Yabuuchi E, Kosako Y, Oyaizu H, Yano I, Hotta H, Hashimoto Y, Ezaki T, Arakawa
legumes (Papilionoideae: Genisteae) from the Canary Islands, along with M (1992) Proposal of Burkholderia gen. nov. and transfer of seven species of
Bradyrhizobium japonicum bv. genistearum, Bradyrhizobium genospecies the genus Pseudomonas homology group II to the new genus, with the type
alpha and Bradyrhizobium genospecies beta. Int J Syst Evol Microbiol species Burkholderia cepacia (Palleroni and Holmes 1981) comb. nov.
55:569–575 Microbiol Immunol 36:1251–1275
Walley F, Clayton G, Gan Y, Lafond G (2004) Performance of rhizobial inoculant Yang S, Tang F, Gao M, Krishnan HB, Zhu H (2010) R-gene-controlled host
formulations in the field. Crop Manag. doi:10.1094/CM-2004-0301-03-RV specificity in the legume-rhizobia symbiosis. Proc Natl Acad Sci
Wang ET, van Berkum P, Beyene D, Sui XH, Dorado O, Chen WX, Martinez- USA 107:18735–18740
Romero E (1998) Rhizobium huautlense sp. nov., a symbiont of Sesbania Yao ZY, Kan FL, Wang ET, Wei GH, Chen WX (2002) Characterization of rhizobia
herbacea that has a close phylogenetic relationship with Rhizobium galegae. that nodulate legume species of the genus Lespedeza and description
Int J Syst Bacteriol 48:687–699 of Bradyrhizobium yuanmingense sp. nov. Int J Syst Evol Microbiol
Wang ET, van Berkum P, Sui XH, Beyene D, Chen WX, Martinez-Romero 52:2219–2230
E (1999) Diversity of rhizobia associated with Amorpha fruticosa isolated Young JPW (1996) Phylogeny and taxonomy of rhizobia. Plant Soil 186:45–52
from Chinese soils and description of Mesorhizobium amorphae sp. nov. Int Young JM (2003) The genus name Ensifer Casida 1982 takes priority over
J Syst Bacteriol 49:51–65 Sinorhizobium Chen 1988, and Sinorhizobium morelense Wang et al. 2002 is
Wang ET, Tan ZY, Willems A, Fernandez-Lopez M, Reinhold-Hurek B, Martinez- a later synonym of Ensifer adhaerens Casida 1982. Is the combination
Romero E (2002) Sinorhizobium morelense sp. nov., a Leucaena leucocephala- ‘Sinorhizobium adhaerens’ (Casida 1982) Willems et al. legitimate? Request
associated bacterium that is highly resistant to multiple antibiotics. Int J Syst for an opinion. Int J Syst Evol Microbiol 53:2107–2110
Evol Microbiol 52:1687–1693 Young JM (2010) Sinorhizobium versus Ensifer: may a taxonomy subcommittee of
Wang FQ, Wang ET, Liu J, Chen Q, Sui XH, Chen WF, Chen WX the ICSP contradict the Judicial Commission? Int J Syst Evol Microbiol
(2007) Mesorhizobium albiziae sp. nov., a novel bacterium that nodulates 60:1711–1713
Albizia kalkora in a subtropical region of China. Int J Syst Evol Microbiol Young JPW, Haukka KE (1996) Diversity and phylogeny of rhizobia. New Phytol
57:1192–1199 133:87–94
Weaver RW, Frederick LR, Dumenil LC (1972) Effect of soybean cropping and soil Young JPW, Wexler M (1988) Sym plasmid and chromosomal genotypes are
properties on numbers of Rhizobium japonicum in Iowa soils. Soil Sci correlated in field populations of Rhizobium leguminosarum. J Gen
114:137–141 Microbiol 134:2731–2739
Wei GH, Wang ET, Tan ZY, Zhu ME, Chen WX (2002) Rhizobium indigoferae sp. Young JM, Kuykendall LD, Martinez-Romero E, Kerr A, Sawada H (2001)
nov. and Sinorhizobium kummerowiae sp. nov., respectively isolated from A revision of Rhizobium Frank 1889, with an emended description of the
Indigofera spp. and Kummerowia stipulacea. Int J Syst Evol Microbiol genus, and the inclusion of all species of Agrobacterium Conn 1942 and
52:2231–2239 Allorhizobium undicola de Lajudieet al. 1998 as new combinations: Rhizo-
Wei GH, Tan ZY, Zhu ME, Wang ET, Han SZ, Chen WX (2003) Characterization bium radiobacter, R. rhizogenes, R. rubi, R. undicola and R. vitis. Int J Syst
of rhizobia isolated from legume species within the genera Astragalus and Evol Microbiol 51:89–103
Lespedeza grown in the Loess Plateau region of China and description of Young JP, Crossman LC, Johnston AW, Thomson NR, Ghazoui ZF, Hull KH,
Rhizobium loessense sp. nov. Int J Syst Evol Microbiol 53:1575–1583 Wexler M, Curson AR, Todd JD, Poole PS, Mauchline TH, East AK, Quail
Weir BS (2011) The current taxonomy of rhizobia. New Zealand rhizobia website. MA, Churcher C, Arrowsmith C, Cherevach I, Chillingworth T, Clarke K,
http://www.rhizobia.co.nz/taxonomy/rhizobia.html. Last updated 13 March Cronin A, Davis P, Fraser A, Hance Z, Hauser H, Jagels K, Moule S, Mungall
2011 K, Norbertczak H, Rabbinowitsch E, Sanders M, Simmonds M, Whitehead S,
Wheatcroft R, Watson RJ (1988) A positive strain identification method for Parkhill J (2006) The genome of Rhizobium leguminosarum has recognizable
Rhizobium meliloti. Appl Environ Microbiol 54:574–576 core and accessory components. Genome Biol 7:R34
Whitty PW, Powell W, Sprent JI (1994) Molecular separation of genera in Zaat SAJ, Wijffelman CA, Spaink HP, van Brussel AAN, Okker RJH, Lugtenberg
Cassiinae (Leguminosae), and analysis of variation in the nodulating species BJJ (1987) Induction of the nodA promoter of Rhizobium leguminosarum
of Chamaecrista. Mol Ecol 3:507–515 Sym plasmid pRL1JI by plant flavanones and flavones. J Bacteriol
Wijffelman CA, Pees E, van Brussel AA, Priem M, Okker R, Lugtenberg BJ 169:198–204
(1985) Analysis of the nodulation region of the Rhizobium leguminosarum Zhang F, Dashti N, Hynes RK, Smith NL (1997) Plant growth promoting
Sym plasmid pRL1JI. In: Evans HJ, Bottomly PJ, Newton WE (eds) Nitrogen rhizobacteria and soybean (Glycine max (L.) Merr) growth and physiology
fixation research progress. Martinus Nijhoff, Boston, p 127 at suboptimal root temperatures. Ann Bot 79:243–249
Willems A (2006) The taxonomy of rhizobia: an overview. Plant Soil 287:3–14 Zhang RJ, Hou BC, Wang ET, Li Y, Zhang XX, Chen WX (2011) Rhizobium
Wilson KJ, Sessitsch A, Corbo JC, Giller KE, Akkermans ADL, Jefferson RA tubonense sp. nov., isolated from root nodules of Oxytropis glabra. Int J Syst
(1995) B-glucuronidase (GUS) transposons for ecological and genetic Evol Microbiol 61:512–517
studies of rhizobia and other Gram-negative bacteria. Microbiology Zhou PF, Chen WM, Wei GH (2010) Mesorhizobium robiniae sp. nov., isolated
141:1691–1705 from root nodules of Robinia pseudoacacia. Int J Syst Evol Microbiol
Wilson KJ, Parra A, Botero L (1999) Application of the GUS marker gene 60:2552–2556
technique to high-throughput screening of rhizobial competition. Can Zurdo-Piñeiro JL, Rivas R, Trujillo ME, Vizcaı́no N, Carrasco JA, Chamber M,
J Microbiol 45:678–685 Palomares A, Mateos PF, Martı́nez-Molina E, Velázquez E (2007)
Xi K, Stephens JHG, Verma PR (1996) Application of formulated rhizobacteria Ochrobactrum cytisi sp. nov., isolated from nodules of Cytisus scoparius in
against root rot of field pea. Plant Path 45:1150–1158 Spain. Int J Syst Evol Microbiol 57:784–788
18 Symbiotic Associations Between Ciliates
and Prokaryotes
Michael Schweikert1 . Masahiro Fujishima2 . Hans-Dieter Görtz 3
1
Universität Stuttgart, Stuttgart, Germany
2
Department of Environmental Science and Engineering, Graduate School of Science and Engineering,
Yamaguchi University, Yamaguchi, Japan
3
Department of Zoology, Biologisches Institut, Universität Stuttgart, Stuttgart, Germany
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427 The Taxa of Prokaryotic Symbionts of Euplotes . . . . . . . . . 449

Genus Polynucleobacter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449
The Ciliate Cell as a Microcosm . . . . . . . . . . . . . . . . . . . . . . . . . . . 428
The Omikron-Like Endosymbionts of Euplotes . . . . . . . . . . . 450
The Nuclei of Ciliates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 428 Symbionts of Euplotes without Binomial Names . . . . . . . . 451
The Epsilon Symbionts of Euplotes . . . . . . . . . . . . . . . . . . . . . . 451
The History of Symbiont Research in Ciliates . . . . . . . . . . . . . 429 The Eta Symbiont of Euplotes . . . . . . . . . . . . . . . . . . . . . . . . . . . 451
The B1 Symbiont of Euplotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451
Terms in Symbiosis Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430 Inconspicuous Symbionts of Euplotes . . . . . . . . . . . . . . . . . . . 451
Prokaryotic Symbionts of Paramecium . . . . . . . . . . . . . . . . . . . . 430 Ectosymbionts of Euplotidium . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452

Habitat and Biology of Epixenosomes of
Habitat and Biology of the Bacterial Symbionts of Euplotidium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452
Paramecium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430 Identification of Epixenosomes of Euplotidium . . . . . . . . . 453
The Killer Trait in Paramecium . . . . . . . . . . . . . . . . . . . . . . . . . . . 432 The Xenosomes of Parauronema acutum . . . . . . . . . . . . . . . . . . 453
Habitat of the Xenosomes of Parauronema acutum . . . . . 453
Biology of Holosporaceae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434 Isolation of Xenosomes from Parauronema acutum . . . . 453
Isolation of Symbiont-Bearing Paramecium . . . . . . . . . . . . 436 Identification of Xenosomes from
Culturing Paramecia with Symbionts . . . . . . . . . . . . . . . . . . . 438 Parauronema acutum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454
Other Culture Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439 Prokaryotic Symbionts of Ciliates from Anaerobic

Purification of Symbionts from Paramecium . . . . . . . . . . . 440 Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454
Identification of the Paramecium Symbionts . . . . . . . . . . . 440
Habitat of Prokaryotic Symbionts of Ciliates from
Orcein Staining of Intracellular Bacteria . . . . . . . . . . . . . . . . . 441 Anaerobic Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 454
The Taxa of Prokaryotic Symbionts of
Paramecium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441 Isolation of Prokaryotic Symbionts of Ciliates from
Genus Caedibacter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441 Anaerobic Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
Genus Pseudocaedibacter . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
Genus Lyticum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443 Identification of Prokaryotic Symbionts of Ciliates from
Genus Pseudolyticum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443 Anaerobic Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
Genus Tectibacter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
Genus Nonospora . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444 Electronic Supplementary Material . . . . . . . . . . . . . . . . . . . . . . . 458
Genus Holospora . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444
Genus Paraholospora . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446
Introduction
Symbionts of Paramecium Without Binomial Names . . . . . 446
Ciliates are frequently colonized by bacteria. This is partly due to
Prokaryotic Symbionts of Euplotes . . . . . . . . . . . . . . . . . . . . . . . . 446 the cytology and biology of these highly evolved heterotrophic
Habitat of the Euplotes Symbionts . . . . . . . . . . . . . . . . . . . . . . 448 protozoa (Hausmann and Bradbury 1996). As phagotrophic
Isolation of the Euplotes Symbionts . . . . . . . . . . . . . . . . . . . . . 448 predators on microorganisms, they bear a high risk of microbial,
Identification of the Euplotes Symbionts . . . . . . . . . . . . . . . . 448 namely, bacterial infections: Ingested bacteria may resist
428 18 Symbiotic Associations Between Ciliates and Prokaryotes
digestion, escape from the phagosomes, and persist in the cells as again. An even smaller number have been described with
endocytobionts. Traditionally, intracellular bacteria in ciliates validated binomial names in keeping with the international
have been termed ‘‘symbionts’’ (Preer et al. 1974a). Phagocytosis rules of nomenclature (for prokaryotic endosymbionts, see
appears to be the usual way infectious bacteria enter their host Stackebrandt et al. 2002). Especially these validly described
cells. In addition to intracellular bacteria, ectosymbionts may be endosymbionts will be discussed in detail in this chapter. A lag
intimately associated with ciliates. The most intriguing example phase in the description of new symbionts in ciliates for several
is certainly the epixenosomes of Euplotidium, which even defend years may be partly due to the fact that intracellular bacteria
their host against predators (Petroni et al. 2000; see also could not be investigated with classical microbiological
> ‘‘Terms in Symbiosis Research’’). methods, since they cannot be grown outside their host cells.
The vast majority of bacterial symbionts of ciliates comprise New powerful techniques for detection and phylogenetic classi-
a variety of bacteria in many unrelated genera of different sub- fication facilitate localization and identification of endosymbi-
groups of proteobacteria. They are alike in that their habitat otic bacteria. The increasing number of data in public databases
is the cytoplasm or the nucleus of a ciliate cell. Within the allows to detect their distinct phycogenetic relationship.
cytoplasm, some bacteria associate with organelles, the
endomembrane system, or cytoskeletal elements specifically.
Many of these symbionts appear to be well adapted to their The Ciliate Cell as a Microcosm
environment; they are no longer free living and have genomes
that are reduced in size, indicating a lengthy period of symbiont- Ciliates—phylum Ciliophora—are large protozoa that offer a
host association. Most of the symbionts are not infectious, but variety of suitable habitats within their cells with plenty of space
a few are Holospora: infectious bacteria of the nuclei. These for bacteria (> Fig. 18.1). The ciliate cell is characterized by
infectious symbionts have developed specific features that guar- a highly organized cortex consisting of the plasma membrane,
antee uptake and transport to the intracellular sites where they flat vesicles termed ‘‘alveoles’’ underneath the plasma mem-
can multiply. In most cases, it is not clear whether the symbionts brane, and a network of microtubules and other cytoskeletal
provide their hosts with a selective advantage; under laboratory elements. Cilia, the motile organelles of ciliates, are anchored
conditions, most of the symbionts have proved to be dispens- to the cytoskeleton.
able. On the other hand, it is significant that the majority of cells Ciliates are heterotrophic. Because of their solid cortex of
of Paramecium biaurelia freshly collected from nature contain three membranes and multiple cytoskeletal systems, phagocyto-
symbionts (Beale et al. 1969), and work by Landis (Landis 1981, sis occurs only at a preformed structure termed the ‘‘oral
1987) and Kusch et al. (2001) gives evidence that paramecia that apparatus.’’ This is a complex metaorganelle consisting of
bear bacteria of the genus Caedibacter (formerly called ‘‘kappa a funnel-like pouch with batteries of cilia and a highly ordered
particles’’) have a selective advantage over those that are symbi- cytoskeletal basket at the bottom of which new phagosomes are
ont-free, see > ‘‘The Killer Trait in Paramecium.’’ Moreover, formed. Newly formed phagosomes are acidified and fuse with
Polynucleobacter necessarius (formerly called ‘‘omikron parti- primary lysosomes (Fok and Allen 1988). As known from other
cles’’) and the closely related omikron-like symbionts occurring professional phagocytes such as macrophages, ingested bacteria
in several freshwater Euplotes species have been shown to be are attacked by acidification, oxidative burst, and lysosomal
necessary for survival of their hosts (Heckmann 1975; enzymes; see > ‘‘The Nuclei of Ciliates.’’ In spite of these attacks,
Heckmann et al. 1983; See also > ‘‘Prokaryotic Symbionts of certain bacteria with appropriate features may survive and even
Euplotes’’). Since in these cases the hosts depend on their sym- escape from the phagosomes. In the cytoplasm of ciliates, bacteria
bionts and those in turn depend upon their hosts, they are no prevail in symbiontophorous vesicles or naked (i.e., not encircled
longer free living—the distinction between ‘‘symbiont’’ on the by host membranes). Symbionts may also live in the large,
one hand and ‘‘organelle’’ on the other hand becomes blurred. somatic macronuclei as well as in the much smaller, generative
Furthermore, Polynucleobacter necessarius is the first micronuclei. Of those bacteria that are, however, nuclear specific,
documented example of a prokaryote, separated in an obligate certain species are only found in micronuclei, and others only in
endocytobiotic form and a free-living form within the same macronuclei, of their specific hosts. Bacteria have even been found
bacterial species (Hahn et al. 2009). in the perinuclear space and in the endoplasmic reticulum (Fokin
Ciliates can usually be handled easily, and some of them— and Karpov 1995). The ciliate cell thus may be regarded as a
particularly species of the Paramecium aurelia complex—have microcosm that might contain even different symbionts at any
been investigated so thoroughly that they can be manipulated in given time. A striking example may be found in the giant ciliates
a variety of ways. This has led to a wealth of information about of the genus Spirostomum (e.g., Fokin et al. 2005a).
their symbionts that has been reviewed extensively by Preer et al.
(1974a), Soldo (1974), Quackenbush (1988), and Pond et al.
(1989). It is not possible to list all the types of symbionts of The Nuclei of Ciliates
ciliates encountered. Their number is very large, and certainly,
many more will be added to this list. Few of them, however, have Characteristically, ciliates have two types of nuclei: small gener-
been deposited in stock cultures and have been investigated ative micronuclei and large somatic macronuclei (Paulin 1996).
in such a way that they can easily be identified when found Ciliates multiply by binary division and sexually propagate by
Symbiotic Associations Between Ciliates and Prokaryotes 18 429
not adapted to this process; bacteria are then digested in the

process of resorption. New macronuclei developing after conju-
gation are free of bacteria unless they are infected anew.
In spite of the problems arising for bacteria colonizing nuclei,
endonuclear symbionts are frequently found in ciliates because in
the nuclei, symbionts should (1) have access to the most complete
supply of metabolites, (2) be more assured of distribution/
segregation to the daughter cells at cell division, and (3) be
protected from cellular defense mechanisms. The last reason
is more obvious when the fact that symbionts in nuclei are
naked—not encircled by host membranes—and less vulnerable
to attacks by lytic enzymes, which would be deleterious to nuclear
structures such as chromatin, is considered.
The History of Symbiont Research in Ciliates

Prokaryotes living in ciliates were first noticed over a century
ago by Müller (1856). Rod-shaped structures were observed in
the macronuclei and micronuclei of a number of ciliates and,
less commonly, in their cytoplasm. In the beginning, it was not
clear whether they were parasites or spermatozoa, because the
micronucleus was considered to be a testis and the macronu-
cleus an ovary, while chromosome filaments and endonuclear
symbionts were mistaken for spermatozoa. This view was
corrected by Bütschli (1876), who also wrote the first review
. Fig. 18.1 on parasites in ciliates (Bütschli 1889). Accounts of early obser-
The ciliated cell. Both macronucleus and micronucleus (8 and 10) vations of bacteria in protozoa that followed this initial period were
are encircled by the nuclear envelope with nuclear pores (not reviewed by Kirby (1941), Wichterman (1953), and Ball (1969).
shown). Note that the nuclear envelopes have a perinuclear space Interest in prokaryotic symbionts of ciliates arose again in
that may be colonized by certain bacteria. The oral apparatus/ the 1950s when it was discovered that a killer phenotype is
cytostome is the only site where phagocytosis takes place. frequently associated with them. The possibility that certain
Intracellular compartments in a ciliate from where intracellular ciliate strains kill other strains by liberating a toxic agent into
bacteria were reported: 1: inside a food vacuole, 2: embraced by the medium was first expressed by Sonneborn (1938), who
a peribacterial membrane, 3: inside the lumen of rER, 4: inside noted this phenomenon in paramecia during experiments on
mitochondria, 5: associated with mitochondria, 6: in the mating types that involved mixing different strains (see > ‘‘The
cytoplasm, 7: in the lumen of the nuclear envelope, 8: inside the Killer Trait in Paramecium’’). He found that under certain
macronucleus, 9: associated with cell cortex structures, and 10: conditions, conjugation could be brought about between killers
inside the micronucleus, outer line = cytoskeleton, inner line = and sensitive paramecia so that genetic analysis of these traits
fibrillar skeletal elements (e.g., myonemes) became technically feasible. Sonneborn (1943) demonstrated
that the killer phenotype was an inherited trait transmitted via
cytoplasmic particles, which he named ‘‘kappa.’’ His findings
conjugation. The latter includes meiosis of the micronucleus, aroused great interest among geneticists and other biologists
resorption of the old macronucleus, and formation of a new because it furnished one of the first clear examples of
macronucleus from the synkaryon that will also give rise to the a cytoplasmic genetic factor. That kappa was an endosymbiont
new micronucleus. Whereas cytoplasmic symbionts may be not known at that time. From data obtained in studies using
simply distributed to daughter cells during binary division, X-rays, Preer (1948b) determined that kappa was similar in size
endonuclear symbionts may even make use of the nuclear divi- to bacteria. He subsequently demonstrated the presence of
sion machinery. Both micronuclear mitosis and macronuclear kappa in the cytoplasm of killer paramecia as Feulgen-staining
division are closed: The nuclear envelopes are maintained, keep- bodies (Preer 1950). In the following years, cytological, bio-
ing nucleoplasm and cytoplasm separated throughout the cell chemical, and physiological studies by a number of workers
cycle. Endonuclear symbionts (see > ‘‘Terms in Symbiosis established that kappa was actually a Gram-negative bacterium.
Research’’) are therefore caught in the nuclei unless they have In 1974, it was given a binomial designation—Caedobacter
developed means of passing through the nuclear envelopes. taeniospiralis, which has since been changed to Caedibacter
During conjugation of the host cell, resorption of the old mac- taeniospiralis (for a detailed review, see Preer et al. 1974, and
ronucleus may be deleterious to endonuclear symbionts that are Preer and Preer 1984).
After the initial discovery of the first killer paramecia, other Endosymbiosis—the symbiont lives within its host, intracellular
types were found. Siegel (1953) described ‘‘mate killers,’’ whose symbiosis.
toxins act only after cell-to-cell contact is made during conjuga- Episymbiosis—the symbiont lives attached to the surface of its
tion, and Schneller (1958) described ‘‘rapid-lysis’’ killers, which host cell.
may injure sensitives in 10 min and kill them in 30 min, a process Episymbiont—microorganism living attached to the surface of
that is much more rapid than when kappa mediates killing. Over a host cell.
decades, bacterial symbionts had been studied most extensively Epibiont—synonymous with episymbiont.
in the Paramecium aurelia species complex (for reviews, see Ectosymbiont—synonymous with episymbiont.
Beale et al. 1969; Preer et al. 1972, 1974; Preer and Preer 1984; Xenosome—intracellular symbiont in a broad sense.
Gibson 1974; and Soldo 1974). They have also been found and Endocytobiont—intracellular symbiont in a broad sense of
studied in other ciliates, particularly in Paramecium caudatum symbiosis.
(for reviews; Görtz 1983, 2001 ; Görtz and Fokin 2009; Fokin Epixenosome—episymbiont in a broad sense.
and Görtz 2009; Fujishima 2009) and in Euplotes species (see
For definitions of the terms, see Henry (1966), Read (1970),
> ‘‘Prokaryotic Symbionts of Euplotes’’). Many cases of simul-
Corliss (1985), Margulis and Fester (1991), Sitte (1993), and
taneous infections of ciliate cells by two or more different bac-
Rosati (1999).
teria have been observed (e.g., Görtz 1992; Fokin et al. 2005). In
each of the killer paramecia, particles were found that later
proved to be prokaryotic symbionts. One only needed to wash
the paramecia free of extracellular bacteria, crush the ciliates, Prokaryotic Symbionts of Paramecium
and observe the resulting preparations in a phase contrast
microscope to ascertain the presence of the symbionts (see With respect to endosymbionts, Paramecium is by far the
> ‘‘Identification of Symbionts in Paramecium’’). This proce- best-studied ciliate genus. In the Paramecium aurelia species
dure also revealed, however, that some strains carry endosymbi- complex, consisting of 14 sibling species described by and
onts without showing any kind of killing ability. The symbionts of named P. primaurelia to P. quadecaurelia, many different types
nonkiller paramecia were named ‘‘nu’’ (Sonneborn et al. 1959). of endosymbionts have been discovered. They have been
The presence of endo- and episymbionts of marine and thoroughly reviewed by and descriptions appeared in earlier
freshwater ciliates living in anaerobic habitats has been generat- editions of this book (Preer 1981; Heckmann and Görtz 1991).
ing increasing interest among ecologists and microbiologists. In Valuable information on isolation and identification of
these habitats, many of the symbionts were identified as symbionts is taken from these earlier editions. In the present
methanogenic bacteria (Stumm and Vogels 1989; Fenchel and edition, descriptions of further endosymbionts were added, and
Finlay 1991a; Finlay and Fenchel 1992; see > ‘‘Prokaryotic Sym- the information about many symbionts has increased owing to
bionts of Ciliates from Anaerobic Environments’’). the availability of new techniques. Among the symbionts where
In recent years, it became obvious to find many more pro- a wealth of data has been presented are the Holospora bacteria.
karyotic symbionts of ciliates once further host species are They differ from most of the Paramecium symbionts in being
studied. The description of episymbionts that have a defensive infectious and in their ability to invade nuclei for reproduction.
function for their host Euplotidium by Rosati, Verni, and col- The holosporas are being investigated to elucidate the
leagues (Rosati et al. 1999) is a fantastic example; see ‘‘Prokary- mechanisms that allow a prokaryote to invade a eukaryote.
otic Ectosymbionts of Euplotidium.’’ (Only recently, these Being relatively large, the holospora can be seen even at
episymbionts have been identified as belonging to the relatively low magnification and are therefore used to monitor
Verrucomicrobia.) their route of infection and the changes they undergo after
entering a cell.
Terms in Symbiosis Research Habitat and Biology of the Bacterial

The following terms are used in this chapter or in the references
Symbionts of Paramecium
cited:
Paramecia are aerobic ciliates from freshwater and brackish
Symbiosis—the living together of dissimilarly named organ- water habitats. Many of the paramecia brought in from nature
isms. Sometimes, the term is used in a restricted sense in are found to contain symbionts. Usually, there are hundreds and
the sense of mutualism. sometimes even thousands of symbionts per paramecium (see
Mutualism—associations which involve mutual benefit (symbi- TEM Bild Cc23 and 3). Endosymbionts may be found in the
osis in its restricted sense). micronucleus, macronucleus, perinuclear space, and cytoplasm
Parasitism—associations in which there is overt exploitation of of paramecia. Different types of symbiont invade different parts
one associate. of the ciliate cell, and they are, moreover, often adapted to one
. Fig. 18.2
Vegetative macronucleus of Paramecium biaurelia stock 562. The
spiral endosymbionts filling the macronucleus are cells of
Holospora caryophila. A few symbionts are also visible in the
cytoplasm. Osmium-lacto-orcein preparation. Whole mount,
bright phase contrast. Bar = 10 mm (From Preer 1969)
. Fig. 18.3
Paramecium species only. The cell compartment in which Paramecium tetraurelia stock 239 bearing endosymbiont Lyticum
a symbiont multiplies and the Paramecium species in which flagellatum, seen as dark-staining rods in the cytoplasm. Osmium-
multiplication occurs are therefore important taxonomic char- lacto-orcein preparation, whole mount, dark phase contrast.
acters. Though it had been observed that bacterial endosymbi- Bar = 10 mm (From Preer, 1969)
onts may prevent secondary infections by another bacterium in
laboratory cultures (Görtz 1983), double and even triple infec-
tions are found in cells of natural populations (Görtz 1992; it should be mentioned that the P. aurelia species 3, 7, 9, 10, 11,
Kusch et al. 2000). 12, 13, and 14 have never been found to contain symbionts
Several symbionts have been shown to require the presence (Preer and Preer 1984); however, not all species of the P. aurelia
of specific Paramecium genes for their maintenance complex have been studied with the same thoroughness. It has
(Sonneborn 1943; Siegel 1953; Schneller et al. 1959; Gibson been argued that symbionts profit from living inside a parame-
and Beale 1961; Fujishima and Fujita 1985). It is not known cium by being better protected from predation, as compared
whether the genes that assure maintenance of the symbionts are with free-living species of bacteria, and that symbionts are
active—e.g., providing the symbionts with some essential provided with a convenient and abundant supply of nutrients
metabolite—or whether they are merely inactive alleles, the (Beale et al. 1969). Which metabolites of the host are used,
active ones preventing growth of an ‘‘invader.’’ In this context, however, is known rarely.
Although many of the symbionts have a smaller genome during the period of cell-to-cell contact at conjugation, the
size than free-living bacteria (Soldo and Godoy 1973a; Hahn toxin producers are called ‘‘mate killers.’’ Different killer stocks
et al. 2009) and some of the associations were suggested to be of Paramecium induce different prelethal symptoms in sensitives
very ancient (Preer 1977), no indications for a transfer of mixed with killers. These symptoms include spinning,
genes from symbiont to host nucleus have been discovered as vacuolization, paralysis, formation of aboral humps, and rapid
has been found in the case of mitochondria (Gellissen and lysis. The symptoms were taken as crucial characters for identi-
Michaelis 1987). Schmidt (1984), studying the association of fication of a symbiont. The value of such symptoms for identi-
Caedibacter varicaedens with P. biaurelia, was unable to fication and determination of the symbionts must, however, be
obtain evidence for a sharing of the translational systems of doubted; see > ‘‘Identification of Symbionts in Paramecium.’’ In
host and symbiont. His observations indicate that all addition to making their hosts capable of producing toxins, the
major proteins found in Caedibacter are synthesized in the symbionts also confer upon the host-specific resistance to
symbiont itself. the toxins produced. When a symbiont is lost from a killer strain,
In only a few cases, it has been demonstrated that host cells the paramecia lose both toxin production and toxin resistance
may profit from bearing symbionts; Soldo (1963) and Soldo and (Sonneborn 1959).
Godoy (1973b) found that it was not necessary to provide Killer strains have been reported not only for species of the
a Paramecium stock bearing Lyticum flagellatum (formerly P. aurelia complex but also for P. caudatum (Schmidt et al.
called ‘‘lambda particles’’) with folic acid, while the same stock 1987b, 1988), P. bursaria (Chen 1955; Dorner 1957), and
freed of this symbiont required the vitamin. Holtzman (1959) P. polycaryum (Takayanagi and Hayashi 1964). In the latter two
observed that P. pentaurelia bearing Pseudocaedibacter falsus was species, however, symbionts were not observed, although it is
more resistant to killer paramecia bearing Lyticum flagellatum likely that they were present and were responsible for the killing
than P. pentaurelia strains that were free of symbionts, and properties of the paramecia.
Landis (1981, 1987) showed that under natural conditions, The most information on killer bacteria was collected from
paramecia with killer properties have a selective advantage symbiotic bacterial genus Caedibacter (for review see
over nonkillers; see > ‘‘The Killer Trait in Paramecium.’’ On Schrallhammer and Schweikert 2009). A feature unique to the
the other hand, in the laboratory, the symbionts are all dispens- killer symbionts of the genus Caedibacter is the ability to
able, and many of them are lost when paramecia are cultured for produce R bodies. These are proteinaceous ribbons, 20–30 mm
some time. The reason for such loss is in most cases a rapid long, coiled inside the bacterial cell to form a hollow
multiplication of the paramecia, resulting in a dilution of the cylindrical structure (> Figs. 18.4, > 18.5, and > 18.6). This
symbionts and then in their loss. Although there is little structure has a diameter of about 0.4 mm in all species
information about what occurs in nature, it appears unlikely
that reinfection of paramecia that have lost their intracellular
bacteria plays a major role for most of the symbionts
(Kusch et al. 2001).
Most symbionts of ciliates are not infectious and are prop-
agated in the host population with the divisions of host cells
only. An exception to this rule is provided by symbionts of
the Holosporaceae. In addition to being propagated with
host cell division, they develop forms specialized for infection
(infectious forms) that are released and infect new cells
upon being taken up with food. Unlike most other endosym-
bionts, they tend to harm their hosts and can therefore be
regarded as parasites, although it cannot be excluded that
under certain conditions, the host cells may have advantages
from the presence of these symbionts (Görtz 1983). Interac-
tions between Holospora and its host Paramecium are described
in more detail in the section entitled ‘‘> Biology of the
Holosporaceae.’’
The Killer Trait in Paramecium

. Fig. 18.4
Many of the symbionts of Paramecium confer on their hosts the Electron micrograph of longitudinal and cross sections of
ability to produce toxins capable of killing sensitive Paramecium Caedibacter varicaedens, endosymbiont of Paramecium biaurelia
strains of the same species and even of other species, if the toxins stock. Longitudinal sections display an R-body each. Note
are liberated into the medium. The toxin producers are called the numerous dark-stained phages inside the coiled R-body.
‘‘killers’’ and their victims ‘‘sensitives.’’ If the toxins act only Bar = 1mm. (in courtesy of Anna Steyer)
stretched R body
native R body
. Fig. 18.6
R bodies of Caedibacter. R bodies (refractile bodies) are coiled
proteinaceous ribbons. By appropriated triggers, such as acid pH,
they are induced to unroll
. Fig. 18.5 1974). Dilts (1976) isolated plasmid DNA from C. taeniospiralis
Electron micrograph of an R body isolated from Caedibacter 51 and suggested that the extrachromosomal DNA might be the
taeniospiralis of Paramecium tetraurelia stock 51. The R body determinant of the R bodies. Further investigation revealed that
begins to unroll from the inside. Negative staining with plasmids are present in all strains of C. taeniospiralis and that
phosphotungstic acid. Bar = 1 mm (From Preer et al. 1972) they are highly homologous as determined by restriction map-
ping (Quackenbush 1983). Further evidence that the genetic
determinant for R-body synthesis resides on the plasmid was
and is about 0.4 mm long, except in C. caryophilus where presented by Quackenbush and Burbach (1983), who cloned
the R bodies are approximately 0.8 mm in width and portions of a plasmid and obtained expression of the R-body-
length. The R bodies unroll when ingested into a phagosome encoding sequences in Escherichia coli. Analysis of various
and also under certain in vitro conditions, e.g., when placed at subclones allowed them to determine the approximate location
low pH (Preer et al. 1966; Schrallhammer and Schweikert 2009). of the R-body-encoding sequence. The DNA required for type 51
It has been suggested that the R bodies play an important R-body synthesis is about 1.8 kbp in size and has been
role in the killing mediated by Caedibacter. Mueller (1963) and completely sequenced (see Pond et al. 1989; Heruth et al.
Smith-Sonneborn and Van Wagtendonk (1964) demonstrated (1994)). However, none of the R-body-producing E. coli clones
that only Caedibacter particles that contained R bodies were was found to be toxic to sensitive paramecia, although function
toxic to sensitive paramecia. In addition, R bodies purified of the R body was successfully tested. The DNA sequence
from certain strains of C. varicaedens have been shown to be required for toxin production has not yet been located. It is
toxic to sensitive paramecia (Preer and Preer 1964; Preer et al. assumed to reside on the plasmid, too (Pond et al. 1989). The
1972). However, neither the toxin itself nor its mode of action R bodies may unroll and destroy phagosomal membranes when
have been identified nor is it known how paramecia are ingested, and apparently they play a role in killing mediated by
protected from the toxic action of their own symbionts. Caedibacter (Preer et al. 1974), but their presence is not required
Early observations suggested that the genetic determinants for the host cell to resist being killed.
of R bodies are plasmids or bacteriophages that have lost the R bodies of some Caedibacter species are associated with
ability to lyse their host cells—the symbiotic bacteria—upon phage capsids (> Fig. 18.4). The proteins of these R bodies and
maturation of the virions (Preer and Preer 1967; Preer et al. perhaps also the toxin are encoded on bacteriophage genomes
(Quackenbush 1988; Schmidt et al. 1988; Pond et al. 1989; Fokin photosynthetic bacterium (Favinger et al. 1989). Since
and Görtz 1993). Only Caedibacter bearing phages or plasmids R bodies are often present at low frequencies, it appears likely
may confer the killer trait to their hosts. Heruth et al. (1994) that R-body-producing bacteria are more common than previ-
have characterized and sequenced three genes (rebA, rebB, and ously suspected. With respect to the shape of the ends and the
rebC) for synthesis and assembly of R-body synthesis in mode of unrolling, the different types of R body vary (for details,
C. taeniospiralis. These genes are independent transcriptional see Preer et al. 1974; Quackenbush 1988, and Pond et al. 1989).
units on a plasmid. Two polymerization events were found to
be involved in R-body assembly: One event requires RebB and
RebC; the other requires all three proteins. The RebC protein is Biology of Holosporaceae
apparently involved in posttranslational modifications of RebA
and RebB, both of them showing peptide species with different All Holospora species known up to the present are residents of
molecular weights. The role of a fourth protein whose gene has the micro- or macronuclei of Paramecium; see > ‘‘The Ciliate
not been sequenced remains unknown (Kanabrocki et al. 1986; Cell as a Microcosm.’’ The bacteria grow and multiply exclu-
Heruth et al. 1994). sively in the host nuclei (> Figs. 18.2 and > 18.7) and have
Dilts and Quackenbush (1986) have provided evidence that evolved mechanisms to invade nuclei as well as to escape from
R bodies are required for the killing trait to be expressed but intact nuclei (Hafkine 1890; Ossipov et al. 1975; Ossipov 1981;
not for resistance of the ciliate host to killing mediated by Görtz 1983; Fokin and Sabaneyeva 1997; Abamo et al. 2008;
C. taeniospiralis. They described a mutant strain of Fujishima 2009; Iwatani et al. 2005; Sabaneyeva et al. 2009;
C. taeniospiralis 169 that simultaneously lost the ability to pro- Eschbach et al. 2009). To date, two kinds of models for
duce R bodies and to kill sensitive paramecia but still rendered Holospora’s escape from the host digestive vacuole had been
its host resistant to killing. Investigations of the R-body- proposed: a transport vesicle model (> Fig. 18.8) and a disrup-
encoding plasmid isolated from the mutant revealed that tion of the digestive vacuole model (Fujishima 2009). Holospora
a transposon-like element had been inserted within the species are host specific, infecting only certain species of Para-
R-body-encoding region, thereby eliminating R-body produc- mecium, and nucleus specific, infecting only the micronucleus or
tion. Two separate mutational events occurred in the same cell, the macronucleus. Host cells remain vital with Holospora infec-
one inactivating the R-body-encoding sequence and the other tions, and it has been shown that Holospora-bearing paramecia
inactivating the toxin-encoding sequence. This shows that are able to grow, divide, and even mate. However, conjugation is
R bodies are crucial to expression of the killer trait. Their exact unsuccessful, as new macronuclei are apparently not functional
role remains, however, unknown. Evidence indicates that their (Görtz and Fujishima 1983), and growth and physiological
action probably involves delivery of the toxin to the sensitive conditions of infected cells may be affected, depending on
paramecium and to its target site by unrolling and penetrating culture conditions. In starving Paramecium or after inhibition
the food vacuole membrane (Dilts 1986). Evidence that R bodies of host protein synthesis, most bacteria differentiate into
are required for killing but not for resistance of the host has also the infectious form (Fujishima 1993; Dohra and Fujishima
been obtained for C. caryophilus. Paramecia bearing mutant 1999b). Paramecia seem to have mechanisms to cure themselves
C. caryophilus with blocked R-body synthesis no longer showed from endonuclear symbionts. Namely, in ‘‘wrong’’ hosts,
a killer trait but were resistant to the toxin (Schmidt et al. 1988). infectious forms are digested having been phagocytosed or, if
The R bodies of other species of the genus were shown to be
associated with icosahedral viral capsids (Preer and Jurand 1968;
Grimes and Preer 1971). The capsids were in most cases found to
contain DNA (Preer et al. 1971). The relationships between
genomes of different phages of kappa and between R-body-
encoding plasmids and kappaphage genomes have been studied
by restriction endonuclease analysis and by DNA-DNA hybrid-
ization (Quackenbush 1978; Quackenbush et al. 1986). These
studies demonstrate that the R-body-encoding plasmids show
little or no homology with kappaphages and that there is also
considerable diversity among the kappaphages.
R bodies have also been reported in the free-living bacteria
Pseudomonas taeniospiralis (Lalucat et al. 1979), P. avenae (Wells
and Horne 1983), and Marinomonas mediterranea (Hernández-
Romero et al. 2003). However, with respect to antigenicity,
genetic determinants, and other features, these R bodies appear
to be unrelated to the R bodies of Caedibacter species (Bedingfield . Fig. 18.7
et al. 1984; Gibson 1984; Meenaghan et al. 1984; Kanabrocki Holospora obtusa in the macronucleus of Paramecium caudatum.
et al. 1986; Lalucat et al. 1986). Additionally, R bodies IF infectious form and RF reproductive form. Bar 1 mm (From Görtz
have been reported for Rhodospirillum centenum, a nonsulfur et al. 1988)
nucleus
4 6
5
3 7
2 8
phagosome
1
9
. Fig. 18.8
Infection cycle of Holospora obtusa. 1 The infectious form of the
bacterium (long rod) is ingested by a Paramecium caudatum cell
into a phagosome (food vacuole). 2 Microfilaments are nucleated . Fig. 18.9
at the phagosomal membrane covering the bacterium protruding Phase contrast micrograph of Holospora obtusa isolated from
from the phagosome. 3 The bacterium is sluiced out of the Paramecium caudatum. Note the peculiar appearance of the
phagosome. It remains encircled by a host membrane that forms infectious forms reflecting their fine structure (> Fig. 18.8). Inset:
a transport vesicle. Later on, the transport vesicle is surrounded developmental stages of Holospora obtusa. On the left,
by vesicles of endoplasmic reticulum. These vesicles finally form a bacterium with three constrictions (arrowheads) undergoing
a secondary transport vesicle. 4 While the inner (phagosome- multiple divisions is seen
derived) membrane disintegrates, the membranes originating
from endoplasmic reticulum fuse with the membranes of the
nucleus, and the bacterium is incorporated into the Fujishima 1999a; Fujishima et al. 1990a; Wiemann and Görtz
macronucleus. 5 The nucleus of the bacterium constricts at several 1991; Abamo et al. 2008). The cytoplasm is condensed and
sites. 6 The bacterium divides into 4–10 short rods. By this located toward one end of the symbiont, while a voluminous
multiple division, the reproductive form is established. 7 The periplasmic area is located mainly at the other end. The peri-
reproductive form multiplies by binary fission. Some of the short plasm consists of fine granular, strongly osmiophilic material.
rods grow longer and differentiate into infectious forms, see Some less osmiophilic material is located at the end, distal from
‘‘Development of Holospora.’’ 8 At the division of the host cell, the cytoplasm. Obviously, the infectious form is physiologically
mature infectious forms are collected in the connecting piece of inactive. It is a resting stage that may remain vital and infectious
the dividing nucleus, and 9 are later released from the host cell outside host cells for days to weeks. Infectious forms were
(Ossipov 1981; Görtz and Wiemann 1989; Wiemann and Görtz therefore regarded as spores (‘‘holospores’’; Hafkine 1890;
1989) Ossipov 1981). The mode of development and reactivation is,
however, different from formation and hatching of endospores
(Görtz and Wiemann 1989 and Wiemann 1989; > Fig. 18.11).
When an infective form has been taken up by a paramecium
they enter the nucleus, may later be lysed within the nucleus and is on its way into the nucleus, the periplasmic material
(Skoblo et al. 1990; Ossipov et al. 1993; Fokin and Skovorodkin disappears while the cytoplasm expands (Dohra et al. 1994.;
1991, 1997; Fokin et al. 2005). This lysis may be due to an Fujishima et al. 1997) (> Fig. 18.11). It is assumed that the
unknown mechanism of cellular defense. After lysis of bacteria periplasm of the infectious form contains substances that inter-
in the nucleus, host cells remain vital nevertheless. act with host membranes during the infection process (Görtz
Not only the infection cycle but also the development of the et al. 1988; Görtz and Wiemann 1989). The further development
Holosporaceae is unique. The bacteria show a developmental of the infectious form into the short, reproductive form is
cycle with a long infectious form and a short reproductive form completed in the nucleus. Induction of the development of
(> Fig. 18.9). It is, however, not just the length that characterizes infectious forms appears to be activated by specific triggers of
infectious forms but also their specific organization. Even short host cells. Acidification of the phagosome and possibly also the
forms are infectious (Kawai and Fujishima 2000). The infectious actions of oxidative radicals and lysosomal enzymes may be such
form is unique among bacteria. It is polarly organized and has triggers for the development of infectious forms of Holospora
a voluminous periplasm (> Figs. 18.7, > 18.10, > 18.11, and and may initiate a recognition mechanism depending upon
> 18.12) that contains a number of stage-specific proteins, protein-protein interactions between bacterium and host cell
some of which appear to be released during the infection process (Wiemann and Görtz 1991; Fujishima et al. 1997; Iwatani et al.
(Görtz and Wiemann 1987; Dohra et al. 1994, 1997; Dohra and 2005). Apparently, there is a stepwise interaction between
OM IM
Periplasm Cytoplasm
. Fig. 18.10
Diagram of infectious form of Holospora. The cytoplasm is condensed and occupies only half the volume of the bacterial cell.
Accordingly, the periplasm is extremely voluminous. IM the inner membrane (cell membrane), OM the outer membrane, and T, the less
electron dense, apically oriented material in the periplasm (special tip)
invading bacterium and host cell. This is indicated by the to this, the outer membrane of the infectious form was found to
observation that the equipment of the invading infectious contain only very few IMPs (Görtz et al. 1989), which corre-
form with surface proteins changes dramatically (Görtz et al. sponds to the observation that the infectious form shows only
1992; Iwatani et al. 2005). a limited number of surface proteins (Görtz et al. 1992). These
Certain proteins were immunolocalized in the periplasm of differences between the two forms correlate with a difference in
the infectious form (> Fig. 18.13). During the invasion process, the behavior of the two forms during the division of the host
some of these proteins were found on the surface of the bacteria nuclei. The infectious forms of H. obtusa are concentrated in the
after their ingestion into the phagosome, and others were asso- connecting piece around the separation spindle of the dividing
ciated with the phagosomal membrane (Görtz et al. 1990; nucleus and are later released from the host cell. Outer mem-
Fujishima et al. 1990b; Dohra et al. 1994, 1997; Wiemann and brane of the reproductive forms has high affinity to the host
Görtz 1991). This is what would be expected if such proteins chromatin so that the bacteria are transported to the poles and
were used for communication with host membranes. Released in this way, get into the daughter nuclei (Ossipov et al. 1975;
periplasmic proteins could also protect the bacteria against Ossipov 1981; Görtz and Dieckmann 1980; Wiemann 1989;
lysosomal enzymes of the host or inactivate such enzymes. Wiemann and Görtz 1989). ). Apparently, H. obtusa makes use
Dohra et al. (1997) have sequenced the gene of a small periplas- of the division apparatus of the host nucleus for their distribu-
mic protein of 5.4 kDa. The gene is highly expressed in the tion to the daughter cells.
intermediate form, a transitional stage in the development The physiology of Holospora and the triggers needed at
from the reproductive into the infectious form of the bacterium. different stages of its development are still enigmatic. It appears
It has been suggested that the protein may function in the that during development of the infectious into the reproductive
recognition process in the early phase of infection. Amino form, protein synthesis of the host cell is needed (Dohra and
acid sequence similarities with other polypeptides have not Fujishima 1999b). It has also been shown that induction of the
been found. Binding experiments of biotinylated macronuclear development of the reproductive form into the infectious form
proteins on blotted Holospora proteins had revealed binding to depends upon protein synthesis of the host cell (Fujishima
a 50-kDa polypeptide (Ehrsam and Görtz 1999). The 50-kDa 1993). Freiburg (1985) presented evidence that RNA polymerase
protein was immunolocalized in the periplasm and on the of infected macronuclei of Paramecium had a 5-fold higher
surface of H. obtusa on sections for light and electron activity than that of uninfected nuclei (Freiburg 1985). Unlike
microscopy and may function in the recognition process endosymbionts of Euplotes species (Fujishima and Heckmann
during infection. Dohra et al. (1998) have identified a 1983; Heckmann 1983), Amoeba proteus (Jeon and Jeon 1976),
GroEL-like protein in H. obtusa that is selectively expressed in Holospora species is not a necessary endosymbiont for the host’s
the reproductive form. survival. However, Holospora-bearing Paramecium cells can
The infectious form differs from the reproductive form not acquire heat-shock resistance (Hori and Fujishima 2003;
only in morphology but also in the protein patterns produced Fujishima et al. 2005; Hori et al. 2008) and osmotic-shock
on sodium dodecyl sulfate-polyacrylamide gel electrophoresis resistance (Smurov and Fokin 1999); therefore, Paramecium
(SDS-PAGE; > Figs. 18.14 and > 18.15). Some proteins specific cells become adapted to unsuitable environments from their
for the infectious form of H. obtusa are located in the periplasm, habitats through Holospora species. Furthermore, differential
as shown by means of immune electron microscopy display reverse transcribed PCR shows that H. obtusa alters mul-
(> Fig. 18.15) using poly- and monoclonal antibodies as probes tiple gene expression of the host after establishing endosymbio-
against these proteins (Wiemann and Görtz 1991; Dohra et al. sis (Nakamura et al. 2004).
1994; Fujishima et al. 1997).
The fine structure of the reproductive forms of species of
Holospora is that of Gram-negative bacteria. Freeze-fracture Isolation of Symbiont-Bearing Paramecium
studies of H. obtusa revealed that the outer membrane of the
reproductive form has a density of intramembrane particles Prokaryotic symbionts of Paramecium are not free living. They
(IMPs) that is similar to that of the inner membrane. In contrast can only be found in their host cells. To investigate the
RF
1
IF
cytoplasm
periplasm
. Fig. 18.11 . Fig. 18.12

Developmental cycle of Holospora. 1 Reproductive forms (RF) Scanning electron micrograph (SEM) of the isolated infectious
multiply by binary fission. 2 Some reproductive forms grow form of Holospora obtusa. (a) One pole (arrows) of the infectious
longer. 3 After having reached their final length, the cytoplasm forms appears rough compared to the other pole. Bar = 1 mm.
condenses and proteins are deposited in the periplasm until the (b) Ends of the bacterium at higher magnification. Bar = 0.1 mm
periplasm makes up more than half of the total cell volume. At the (From Fujishima et al. 1990b)
end of this differentiation, the infectious form (IF) is established.
4 When an infectious form is ingested into a phagosome of
a suitable host, it starts to develop into the reproductive form. The rivers with the help of a dissection microscope. For most species
cytoplasm expands, while periplasmic proteins are assumed to be of Paramecium, the highest abundance is found in the eulittoral
secreted. 5 Once the periplasm has almost completely or in the microaerobic areas just above the anaerobic layer of
disappeared, the bacterium constricts at several points and detritus on the water level, where the abundance of bacteria and
undergoes a multiple division. By this division, the reproductive small eukaryotes is high. In the pelagial—the free water body of
form is established anew a lake—the abundance of paramecia may be low. Few species of
Paramecium are found in brackish water, and no marine species
are known (Wichterman 1953; Fokin and Chivilev 1999). Para-
symbionts, paramecia have to be isolated from natural habitats. mecia do not form cysts and, unless they have been introduced
The techniques used in collecting and cultivating paramecia with the water added, cannot be obtained from infusions of
have been described by Sonneborn (1950, 1970). Paramecia are straw or hay. When brought in from nature, the paramecia
easily detected in samples of pond or lake water and even in must be grown slowly at low temperature (about 16 C) without
. Fig. 18.13
Immunoelectron micrograph of Holospora obtusa in the
macronucleus of Paramecium caudatum, treated with
a monoclonal antibody specific for a 39-kDa periplasmic protein
of the infectious form. Secondary label with gold anti-mouse IgG.
Bar = 0.2 mm (From Dohra et al. 1994)
. Fig. 18.15
Two-dimensional sodium dodecyl sulfate-polyacrylamide gel
electrophoresis (SDS-PAGE) gels of the reproductive (a) and the
infectious (b) forms of Holospora obtusa. Spots being unique for
either of the forms were labeled with single arrowheads, and those
common to both of the forms were indicated with white-rimmed
arrowheads (From Fujishima et al. 1990)
A culture containing only one type of symbiont may be

obtained, therefore, by isolating a single paramecium and grow-
ing it into a clone. If a strain of Paramecium does carry more
than one type of symbiont, it is often possible to obtain pure
. Fig. 18.14 cultures by growing the strain at a high fission rate until the
SDS-polyacrylamide gel electrophoresis of total protein of symbionts are diluted down to no more than one symbiont per
Holospora obtusa and Escherichia coli. Lane 1, H. obtusa, infectious paramecium. Cells isolated at that time and cultured at a low
form (from 2 107 cells); lane 2, H. obtusa, reproductive form fission rate will grow into populations containing one type of
(from 2 108 cells); and lane 3, E. coli (from 5 107 cells). The symbiont only. This technique has also been used to separate
position and size (kDa) of marker proteins are indicated on the different types of symbionts (Preer Jr. 1948a).
right (From Görtz et al. 1988)
Culturing Paramecia with Symbionts

adding any antibiotics, if the purpose is to look for symbionts.
Their presence can be monitored as described in the section A method of culturing paramecia involves the use of bacterized
‘‘> Identification of Symbionts in Paramecium.’’ cereal leaf medium (Sonneborn 1970). For a stock solution,
If the paramecia are grown at high fission rates, their sym- 75 g of dried cereal leaves (a mixture of wheat and barley leaves
bionts often become diluted and may even be lost. Natural that were harvested before growth of ears) are boiled for 15 min
populations of Paramecium are infected with symbionts to vary- in 1 l of distilled water. The solution is then filtered, diluted with
ing degrees. Most strains carry only one kind of symbiont. water (1: 30), buffered with Na2HPO4, and autoclaved. A day
before use, the autoclaved medium is inoculated with diameter 25 mm,

Enterobacter aerogenes, Klebsiella pneumoniae, Micrococcus about 7 mm deep
luteus, or another bacterium suitable as food for Paramecium.
The medium should have a pH of about 7. The growth rate of
Paramecium can be regulated by varying the temperature and
the amount of bacterized medium that is added to a culture.
Other Culture Methods three spot depression slide

. Fig. 18.16
An even simpler method is to culture paramecia in autoclaved Three-spot depression slide. Each depression has the volume of
straw or hay infusions. However, this method is less reproducible about 1.5 ml. The depression slides may be labeled by pencil on
than the one using cereal leaf medium. For preparing the straw the etched sides. Kept in a moist chamber, the slides may be
or hay medium, 3–5 g of straw or hay per liter of distilled water is covered by glass plates supported by the elevated edges of the
boiled, filtered, and buffered with Na2HPO4 (pH 6.8–7.0) and slides
then autoclaved. After cooling and a day before use, the medium
is inoculated with bacteria.
Unfiltered straw medium, with the straw remaining in the Before it is used, the medium is inoculated with a small number
medium, may be used for maintaining stock cultures. In these of Chlamydomonas and incubated under light for 2 days.
cultures, paramecia with or without symbionts can be kept for A single bacterium-free paramecium, put into a tube half-filled
months at low temperatures (6–16 C) without special care. with this Chlamydomonas medium, usually will multiply and
Another, even safer, method for maintaining stocks is to freeze ingest most of the algae within 3–4 days.
the paramecia and to keep them in liquid nitrogen, from which Most of the axenic media used for growing paramecia are
they may be recovered with their symbionts on thawing (Simon based on a recipe initially designed by Soldo et al. (1966). Recent
and Schneller 1973). A great number of Paramecium stocks modifications by Thiele et al. (1980) and Schönefeld et al.
bearing symbionts are maintained in the frozen state at the (1986) have proved particularly valuable for large-scale
American Type Culture Collection, Rockville, MD, USA. It cultures of P. tetraurelia. A modification (Soldo 1987) that
appears, however, that some symbionts may be lost after should be capable of supporting the axenic growth of paramecia
prolonged maintenance in liquid nitrogen. and many other ciliates of freshwater and marine origin is
Some strains of Paramecium can also be grown axenically. shown below.
For sterile growth, the paramecia must first be washed free of Medium for Axenic Growth of Marine and Freshwater
bacteria. This is achieved by allowing the paramecia to swim Ciliates (Soldo 1987)
through sterile medium for a time sufficient to permit the
bacteria in the food vacuoles to be eliminated (Sonneborn Proteose peptone 10.00 g
1950; Van Wagtendonk and Soldo 1970). A simple technique Trypticase 5.0 g
modified from Heatherington (1934) is to place several parame- Yeast nucleic acid 1.0 g
cia at one edge of a hole of a depression slide filled with wash Biopterin 0.5 mg
fluid, allow them to swim to the other side, and then to transfer
Folic acid 0.5 mg
them to a new depression. The procedure should be repeated
four times before the cells are left in the fifth wash for an hour. Nicotinamide 2.5 mg
They should then be transferred to a new depression, and this D-Pantothenate, Ca 7.5 mg
should be repeated hourly for another 4 h. A helpful tool for this Pyridoxal hydrochloride 2.5 mg
method and for the growth of small numbers of paramecia is the Riboflavin 2.5 mg
so-called three-spot depression slide, also called the ‘‘Sonneborn Thiamine hydrochloride 0.01 mg
slide.’’ This is a thick slide with three depressions (> Fig. 18.16).
DL-Thioctic acid 0.01 mg
For sterile washing, the paramecia must be taken up each time in
Phospholipid (oleate-containing) 250 mg
a new sterile micropipette with as little fluid as possible; slides
and wash fluid have to be sterile. The use of antibiotics to obtain Stigmasterol 2 mg
bacteria-free cultures is to be avoided because such substances Distilled water (or seawater) 1l
may harm the symbionts.
A method for culturing paramecia without bacteria involves The medium is prepared in distilled water for freshwater
use of the photoautotrophic alga Chlamydomonas reinhardtii as ciliates and in seawater (density 1.015–1.026 g/ml) for marine
a food organism (Preer et al. 1974). The medium contains 1 g of forms. The final pH is 7.2. Stigmasterol is added from a stock
yeast autolysate, 0.25 g sodium acetate, 0.625 g of cereal leaves, solution (0.5 g of stigmasterol dissolved in 100 ml of absolute
and 0. 125 g of Na2HPO4 in 1 l of double-distilled water. The ethanol, stored at 4 C in a tightly capped plastic bottle) by
medium is dispensed into test tubes, autoclaved, and stored. injection into the culture medium from a syringe.
It is important to transfer paramecia gradually from holosporas sediments at 350g within 10 min, while the repro-
a bacterized to an axenic medium; the ciliates need to be allowed ductive form remains in the supernatant (Görtz et al. 1988).
to adapt slowly. Several procedures have been proposed for this Protocols for isolating holosporas directly from cell homog-
transfer (see, e.g., Van Wagtendonk and Soldo 1970, and Fok and enates have been published by Fujishima and Nagahara (1984b),
Allen 1979). An adapting medium (called ‘‘VS medium’’) based Schmidt et al. (1987a), Görtz et al. (1988), Fujishima et al.
on that of Allen and Nerad (1978) is one containing all the (1990a), and Kawai and Fujishima (2000). Infected cells are
vitamins of the axenic medium given above plus stigmasterol, homogenized in sodium phosphate buffer with a hand homog-
but not the other components, the pH being adjusted to 7.0. In enizer. The Teflon pestle should fit tightly. The homogenate is
VS medium, bacteria (e.g., Klebsiella pneumoniae previously then centrifuged at about 3,000g for 10 min. The pellet is
grown in a tryptone medium) are suspended and adjusted to resuspended in buffer and centrifuged in a preformed continu-
OD590 = 3.0 by dilution. The bacteria are then dispensed in 1 ml ous gradient of 70 % Percoll for 12 min at 40,000g. To maintain
portions in screw-cap tubes and placed in a deep freeze. The final the gradient, it is advisable to use a centrifuge with an acceler-
medium is prepared, about a week’s supply at a time, by adding ation rate control. For H. obtusa, H. elegans, H. undulata, and
1 ml of the frozen bacterized VS medium to 9 ml of unbacterized H. recra, the infectious forms then concentrate in a sharp band
VS medium. This medium is autoclaved and inoculated with that is usually uncontaminated with food bacteria and cell debris
paramecia. The protozoa grow in this medium at a rate of 1/2–2 (microscopical control). The reproductive forms do not form
fissions per day. Not all stocks of paramecium can be adapted to a band in a continuous Percoll gradient. These forms can be
an axenic medium. Furthermore, some media were found to obtained with the help of a discontinuous gradient (at 10,000g)
support growth of paramecia, but the latter were not able to where—depending on the species—they form a layer above ca.
maintain their symbionts under these conditions. 60 % Percoll. Most of the cell organelles of P. caudatum do not
enter at the 50 % Percoll step.
The isolation procedures described here have been helpful for
Purification of Symbionts from Paramecium studying ultrastructural details and protein composition of symbi-
onts, their R bodies, and their DNA. Attempts to culture Holospora
The purification of symbionts from paramecia (i.e., separation endosymbionts outside their hosts have been unsuccessful, as
from host cell material) has been achieved in a number of ways: has been the case for other symbionts of aerobic ciliates.
passage of cell homogenates through ion-exchange cellulose
columns (Smith 1961; Mueller 1963), through filter paper col-
umns (Preer et al. 1966), and centrifugation (Soldo et al. 1970) Identification of the Paramecium Symbionts
have all been used. For many symbionts, Percoll gradient cen-
trifugation proved to be the method of choice (Görtz et al. 1988; Most of the endosymbionts are easy to observe. Paramecia are
Fujishima et al. 1990a). Depending on the type of symbionts, collected from ponds, lakes, or rivers and brought into the
different methods of purification may be necessary. For example, laboratory. With the help of a micropipette, a few cells are placed
the R-body-containing Caedibacter caryophila can best be iso- on a microscope slide, covered with a cover glass, and then
lated with the help of a discontinuous 70 % Percoll gradient, observed with a light microscope using phase contrast optics.
whereas isolation of Caedibacter caryophila containing Crushing of the paramecia is sometimes necessary (Preer and
no R bodies is better achieved by means of an ECTEOLA Stark 1953). The classification of the endosymbionts of Parame-
(anion exchanger) column (Schmidt et al. 1987c, 1988). For cium is traditionally based on the following characteristics: mor-
the isolation of holosporas, two different approaches have been phology and killing activity, the species in which they live, DNA
followed. One method starts with homogenization of the host base ratios calculated from the buoyant density (Bd) or from
cells, and the other involves separation of the macronuclei from thermal denaturation profiles (Tm), and interstrain DNA/DNA
the cytoplasm and then isolation of the symbionts. The latter hybridizations. Type strains, which have been deposited at the
method involving prior isolation of nuclei (Freiburg 1985) American Type Culture Collection (ATCC), are listed below by
proved especially useful for preparing clean reproductive forms their ATCC numbers. Some type strains have also been depos-
of the symbionts because it avoids contamination of the bacteria ited in the culture collection of the Laboratory of Invertebrate
with food vacuoles. For isolation of nuclei, the paramecia are Zoology, Biological Research Institute, St. Petersburg State
lysed in a buffer containing 10-mM Tris, pH 7.9, 0.25-M sucrose, University, Russia, and can be obtained from that source.
3 mM-CaCl2, 8 mM-MgCl2 plus 0.1-mM phenylmethylsulfonyl Since endosymbionts do not multiply outside their hosts,
fluoride, 0.1 % (w/v) spermidine, and 0.2 % Nonidet P-40 by characterization based on metabolism and growth, as is custom-
gently stirring the cells in an ice bath and subsequently passing ary for bacteria, is not possible. Instead, morphological and
the suspension 5–10 times through a 20-ml pipette. The nuclei biological features and nowadays, sequencing of ribosomal
are concentrated on a cushion of 1.6 M sucrose by spinning for genes after PCR amplification and classification by fluorescence
10 min at 700g. The purified nuclei are then homogenized in in situ hybridization (FISH) using specific oligonucleotide
a Mg-free buffer, and the bacteria are pelleted. Two different probes complementary to ribosomal sequences (> Fig. 18.17))
forms of the bacteria, the infectious and reproductive forms, can unique for different taxonomic levels (from species to universal
be separated by sedimentation. The infectious form of for all bacteria) have become powerful tools for classification of
specific as well as nucleus specific, i.e., they infect only one

type of nucleus, either the micronuclei or the macronuclei
(Ossipov 1973; Ossipov et al. 1975; > Fig. 18.1; see > The Ciliate
Cell as a Microcosm). Holospora bacteria establish short, repro-
ductive forms that undergo binary fission as well as long, infec-
tive forms that leave the paramecium and infect others very
efficiently (Ossipov and Podlipaev 1977; Wiemann and Görtz
1989; Wiemann 1989; Fujishima et al. 1990b; > Figs. 18.9,
> 18.10, and > 18.11). Namely, the infectious forms have
a structure unique among bacteria, and their structure may

well be used for identification.
Orcein Staining of Intracellular Bacteria
A very useful staining technique that allows observations of

intracellular bacteria in regular bright field and/or phase con-
trast is given by Beale and Jurand (1966): Paramecia are placed in
a small drop on a slide, and as much of the fluid as possible is
withdrawn by a micropipette or filter paper. The paramecia are
lightly fixed by exposure to OsO4 vapor for 6–10 s and immedi-
ately stained with a small drop of lacto-aceto-orcein (1 g of
. Fig. 18.17
orcein dissolved in 25 ml of hot 45 % acetic acid, mixed with
Fluorescence in situ hybridization (FISH) using an oligonucleotide
25 ml of lactic acid, diluted with water 1:1, and then filtered).
probe specific for Holospora obtusa. Short and long rods of
A coverslip is placed over the drop of stained paramecia and is
holosporas in the host macronucleus are labeled. Faintly,
lightly pressed down, flattening but not disrupting the cells. The
phagosomes with prey bacteria are visible (From Fokin et al. 1996)
preparations can then be observed with a 100X oil immersion
objective. To remove lipids that sometimes obscure observation
uncultured intracellular bacteria (Amann et al. 1991, 1995). of endosymbionts, the paramecia can be treated with a drop of
Nonetheless, detection and sometimes a preliminary identifica- acetone or a 3:1 mixture of ethanol and acetic acid before
tion are still possible with the help of a light microscope staining. Orcein-stained preparations can be made permanent
equipped with phase contrast, namely, after staining with orcein by either applying Vaseline around the edges of the coverslip or
(Beale and Jurand 1966; see > ‘‘Orcein Staining of Intracellular by the following procedure. The slide with the orcein prepara-
Bacteria’’). tion and coverslip is rapidly cooled in liquid nitrogen. The slide
In addition to microscopic observation, killer tests should be is immediately lifted off the coverslip with a razor blade. Cells
performed when a stock is brought in from nature. Standard will remain on the slide. The preparation is dehydrated by
sensitives (e.g., cells of stock 152 of P. triaurelia) may be dipping it briefly into 50 %, 70 %, and absolute ethanol and
employed in such tests. Equal volumes of the culture to be tested adding a small drop of Euparal or other ethanol-soluble resins.
and a culture of sensitive cells are mixed in a depression slide and The resin layer between slide and coverslip should be extremely
observed with appropriate controls for prelethal effects. Mate thin for optimal optical conditions.
killing can only be detected by mating sensitive cells with sym-
biont bearers. Any symbiont-free strain that will mate with an
unknown strain is usually adequate. The mode of killing and the The Taxa of Prokaryotic Symbionts of
features of R bodies (if present), i.e., shape of the ends and the Paramecium
mode of unrolling of the different types of R body (> Fig. 18.6),
have been used for classification and determination of killer Genus Caedibacter
symbionts (for details, see Preer et al. 1974; Quackenbush 1988,
and Pond et al. 1989). It has now become evident that R bodies Cells are straight rods or coccobacilli, 0.4–1.0 mm in diameter
and their genetic basis, plasmids or phages, are not adequate for and 1.0–4.0 mm in length (Preer and Preer 1982). Usually less
classification of their bacterial hosts (Beier et al. 2002). than 10 % but sometimes up to 50 % of the symbionts contain
In contrast to the killer trait, other biological features may refractile (R) bodies (> Figs. 18.4, > 18.5, and > 18.6). Cells
still be of great value for the classification and determination of containing R bodies are usually wider and longer than those
certain symbiotic bacteria. Such features may be the high infec- that do not. In addition to R bodies, many spherical phage-like
tivity, unique life cycle, and also unique morphology of the structures or covalently closed circular DNA plasmids are
infectious forms of the species of the genus Holospora; see present in the central lumen of the cylinder formed by the
> ‘‘Biology of the Holosporaceae.’’ These bacteria are host structure. The symbionts are Gram negative and nonmotile.
The G + C content of the DNA is 40–44 mol%. Earlier work of Caedibacter paraconjugatus
this extensively studied genus has been reviewed by Preer et al. Cells are small rods and contain phage-like structures
1974; Preer and Preer 1984; and Pond et al. 1989 and recently (Quackenbush 1982). Less than 1 % of the symbionts contain
Schrallhammer and Schweikert 2009. R bodies of the C. varicaedens type. They are found in the
An analysis of the 16S rDNA of Caedibacter caryophilus, a cytoplasm of P. biaurelia. Ingestion of symbionts by sensitive
killer symbiont dwelling in the macronucleus of Paramecium strains does not produce any observable toxic effect. Cell-to-cell
caudatum, revealed an unusual insertion of 194 bp that was not contact between host and sensitive paramecia is required for
present in mature 16S rRNA (Springer et al. 1993). It was shown toxic effects to be observed in the sensitive paramecia (mate
that C. caryophila contained fragmented 16S rRNA. Insertion of killers). The type strain is found in ATCC strain 30638 (stock
more than 100 bp has been reported for bacterial 23S rDNAs 570 of P. biaurelia; see Preer et al. 1974).
(Roller et al. 1992). Comparable intervening sequences in 16S
rRNAs have not been described in free-living or intracellular Caedibacter caryophilus
bacteria. Results obtained by Beier et al. (2002) give evidence Cells are rods, 0.4–0.7 mm wide and up to 2.5 mm long (Schmidt
that the genus Caedibacter is polyphyletic. Apparently, R bodies et al. 1987b; Euzeby 1997). Those with R bodies are larger than
and their genetic basis, plasmids or phages, are not adequate for those without R bodies. The G + C content is 34.6 mol%.
classification of their bacterial hosts. The following species have They are found in the macronucleus of P. caudatum.
been described: The R bodies unroll from the inside (with blunt outer terminus
and sharp inner terminus) and are associated with phages.
Caedibacter taeniospiralis Width of R bodies after isolation is 0.8 mm. Sensitive strains
Cells are rods, 0.4–0.7 mm in diameter and 1.0–2.5 mm long are killed by paralysis. Coinfections with Holospora obtusa,
(Preer and Preer 1982). The G + C content is 41 mol%. They H. undulata, and H. elegans may occur in natural populations
are found in the cytoplasm of P. tetraurelia only. Their R bodies (H.-D. Görtz, unpublished observations). According to its
(> Fig. 18.5) unroll from the inside and contain plasmids. ribosomal gene sequences, C. caryophilus has been affiliated to
Ingestion of R-body-containing symbionts by sensitive parame- the a-subgroup of Proteobacteria, the closest relative being
cia results in the development of clear small blisters on their Holospora obtusa (Springer et al. 1993). The type strain is
surface in 2–3 h. Between 4 and 5 h, a bulge develops first in the found in ATCC strain 50168 (clone C221 of P. caudatum; see
posterior part of the oral side and later moves to the posterior Schmidt et al. 1987b).
aboral side. The position of the bulge gave rise to the designation
‘‘hump killer.’’ Death takes place in 1–2 days, the corpses
remaining intact for some time. Caedibacter taeniospiralis of Genus Pseudocaedibacter
P. tetraurelia stock 51K has been affiliated to g-subgroup of
Proteobacteria according to its ribosomal gene sequences Cells are rods, 0.25–0.7 mm in diameter and 0.5–4.0 mm long
(Beier et al. 2002). The type strain is found in ATCC strain (Quackenbush 1982). They do not produce R-body-containing
30632 (stock 51 of P. tetraurelia; see Preer et al. 1974). cells. The symbionts are Gram negative and nonmotile. The
G + C content of the DNA is 35–39 mol%. The genus includes
Caedibacter varicaedens some species that confer a killer trait on their hosts, some that
The cells are rods, 0.4–1.9 mm in diameter and 2.0–4.0 mm long render them mate killers, and others that do not produce any
(Quackenbush 1982). The G + C content of the DNA is 40–41 killing ability. Four species have been described:
mol%. They are found in the cytoplasm of P. biaurelia. Different
strains cause vacuolization, paralysis, or rapid reverse rotation Pseudocaedibacter conjugatus
while swimming (spin-killing) of sensitive cells. After 4–6 h of Formerly called mu particles, these cells are rods, 0.3–0.5 mm in
swimming, vigorous rotation to the right becomes nearly diameter and 1.0–4.0 mm long (Quackenbush 1982). The G + C
uninterrupted and then becomes slower. R bodies (> Fig. 18.4) content of the DNA is 35–37 mol%. They are found in the cytoplasm
unroll from the outside. The outer terminus of the unrolled of P. primaurelia and P. octaurelia where a mate-killer phenotype
R body is blunt. The R body is usually associated with bacteri- of the hosts is produced. The type strain is found in ATCC strain
ophage capsids. The type strain is found in ATCC 30637 (stock 7 30796 (stock 540 of P. primaurelia; see Preer et al. 1974).
of P. biaurelia; see Preer et al. 1974).
Pseudocaedibacter minutus
Caedibacter pseudomutans Formerly known as ‘‘gamma particles,’’ these cells are rods, often
These are cigar-shaped rods, approximately 0.5 mm in diameter double but also single rods 0.25–0.35 mm in diameter and 0.5–1
and 1.5 mm long (Quackenbush 1982). The G + C content is 44 mm long (Quackenbush 1982). The G + C content of the DNA is
mol%. They are found in the cytoplasm of P. tetraurelia and 38 mol%. They are found in the cytoplasm of P. octaurelia. In the
cause rapid reverse rotations of sensitive paramecia while swim- host cell, the symbiont is surrounded by an extra set of mem-
ming (spin-killers). The R bodies are of the C. varicaedens type. branes, apparently continuous with the endoplasmic reticulum
The type strain is found in ATCC strain 30633 (strain 51 ml of of the host. The paramecia which bear these symbionts are very
P. tetraurelia; see Preer et al. 1974). strong killers that cause sensitives to develop vacuoles, swell,
finally become spherical, and lyse. Death occurs after about 8 h.

The type strain is found in ATCC strain 30699 (stock 214 of
P. octaurelia; see Preer et al. 1974).
Pseudocaedibacter falsus
Formerly called ‘‘nu’’ and ‘‘pi particles,’’ these cells are rods,
0.4–0.7 mm in diameter and 1.0–1.5 mm long. The
G + C content of the DNA is 36 mol% (Quackenbush 1982).
They have no known toxic actions and are found in the cyto-
plasm of P. biaurelia, P. tetraurelia, and P. pentaurelia. The type
strain is found in ATCC strain 30640 (stock 1010 of P. biaurelia;
see Preer et al. 1974).
Pseudocaedibacter glomeratus . Fig. 18.18

These are small rods, about 0.3 mm wide and up to 1.2 mm long Electron micrograph of Lyticum flagellatum of Paramecium
(Fokin and Ossipov 1986). They are found in the cytoplasm of octaurelia stock 327. Negative staining with phosphotungstic
P. pentaurelia and have no known toxic actions. Symbionts are acid. Bar = 1 mm (From Preer et al. 1974a)
individually enclosed in vacuoles which are tightly associated
with the endoplasmic reticulum. The type strain is strain Bp 171
of P. pentaurelia (deposited in the culture collection of the
Laboratory of Invertebrate Zoology, Biological Research Insti- DNA is 27 mol% in one strain and 49 mol% in another. They are
tute, St. Petersburg State University, Russia). found in the cytoplasm of P. tetraurelia and P. octaurelia. The
type strain is found in ATTC strain 30700 (stock 299 of
Caedobacter chlorellopellens P. octaurelia; see Preer et al. 1974).
This organism, a symbiont living in the cytoplasm of P. bursaria
and found to be antagonistic to symbiotic chlorellae, was named Lyticum sinuosum
‘‘Caedobacter chlorellopellens’’ by Skoblo et al. (1985), who did Formerly, these were called ‘‘sigma particles’’ (Preer and Preer
not know that the genus name Caedobacter had been changed to 1982). They are curved or spiral rods, 0.7–0.9 mm in diameter
Caedibacter. Caedibacter was then restricted to symbionts that and 2.0–10 mm long, sometimes forming chains of 2–3 cells. The
produce R bodies and confer killer traits upon their host para- G + C content is 45 mol%. They are found in the cytoplasm of
mecia (Preer and Preer 1982). The symbiont in P. bursaria is egg P. biaurelia. The type strain is found in ATCC strain 30696 (stock
shaped, rarely rod shaped, 0.35 mm wide and up to 1.4 mm long. 114 of P. biaurelia; see Preer et al. 1974).
Since no R bodies were observed and the cells have no known
toxic actions on sensitives, placement of the species in the genus
Pseudocaedibacter seems more plausible. A renomination of the Genus Pseudolyticum
species seems necessary but should be done only after ribosomal
sequences have been obtained. These are large symbionts with numerous flagella (Boss et al.
1987). No motility has been observed. They may contain refrac-
tile bodies consisting of polyhydroxybutyric acid. No killer
Genus Lyticum activity and no infectivity could be detected. Only one species
is known:
Cells are large rods, 0.6–0.8 mm in diameter and 3.0–5.0 mm long
(Preer and Preer 1982). They resemble bacilli in general appear-
Pseudolyticum multiflagellatum
ance and bear numerous peritrichous flagella (> Figs. 18.3 and
Cells are straight rods, 1.0–2.0 mm in diameter and 3.5–14 mm
> 18.18) but are not obviously motile. They are enclosed in
long (Boss et al. 1987). They are found in the cytoplasm of
vacuoles in the cytoplasm of their hosts. They do not contain
P. caudatum. The symbionts are individually enclosed in vacu-
R bodies. The G + C content is 27 mol% in one type and 45–49
oles. The membrane of these vacuoles forms numerous pro-
mol% in another type. The symbionts are Gram negative. They
jections that are continuous with the endoplasmic reticulum.
produce toxins that kill sensitive paramecia by lysing them
No toxic actions are known.
within 10–20 min (rapid-lysis killing). Two species have been
described:
Lyticum flagellatum Genus Tectibacter

Formerly known as ‘‘lambda particles,’’ these cells are straight
rods, 0.6–0.8 mm in diameter and 2.0–4.0 mm long (> Figs. 18.3 The species of this genus are distinguished from other symbionts
and > 18.18) (Preer and Preer 1982). The G + C content of the by a layer of electron-dense material surrounding the outer of
the two membranes (Preer and Preer 1982). The symbionts have originally assumed (Schmidt et al. 1987a). The differences at
sparse peritrichous flagella and show slight motility on occasion. the molecular level may suggest that the genus Holospora has
They appear not to be toxic. For a more detailed description, see coevolved with ciliates for a long time.
Preer et al. (1974). Only one species has been described: Holospora were the first intracellular bacteria in Paramecium
for which the phylogenetic position was determined (Amann
Tectibacter vulgaris et al. 1991). Holospora obtusa, H. elegans, and H. undulata
Formerly known as delta particles, these cells are straight rods, belong to the a-group of Proteobacteria. The closest relative
0.4–0.7 mm in diameter and 1.0–2.0 mm long, and are Gram among other symbionts in ciliates was found to be Caedibacter
negative (Preer and Preer 1982). Hosts include the following: caryophilus, and the closest relatives among other bacteria found
P. primaurelia, P. biaurelia, P. tetraurelia, P. sexaurelia, and up to now are Rickettsia and Ehrlichia (Amann et al. 1991;
P. octaurelia. The type strain is found in ATCC strain 30697 Springer et al. 1993). It is tempting to regard the striking biology
(stock 225 of P. sexaurelia; see Preer 1974a). (developmental cycle, host specificity for Paramecium, etc.) and
the unique morphology of the infectious form as homologous
features proving the close relationship and monophyletic origin
Genus Nonospora of these bacteria. New observations, however, cast doubt on this
possibility (Fokin et al. 1996).
These rodlike symbionts live in the macronucleus (Fokin et al. The behavior of the infectious forms of certain Holospora
1987a). Flagella have not been observed. No toxic actions are species to assemble in the connecting piece of the dividing host
known. The symbionts are retained in macronuclear fragments nucleus is certainly highly advanced and must be regarded as an
of exconjugants and enter macronuclear anlagen by fusion of old apomorphic feature. This behavior ensures that the infectious
fragments with the anlagen (Fokin et al. 1987a). Only one forms are specifically collected and released by the host cell.
species has been described: Holospora caryophila, H. bacillata, and H. curvata do not share
this feature with the other holosporas (Fokin et al. 1996; Fokin
Nonospora macronucleata and Sabaneyeva 1997), and it is not clear how the bacteria of
These symbionts are rodlike, 0.2–0.3 mm in diameter and mostly these species leave their host nuclei. They either are more prim-
about 1.0 mm long, sometimes forming chains up to 10 mm long itive than the other holosporas, as a quantitative separation of
(Fokin et al. 1987a; Demar-Gervais 1976). The surface of the infectious forms and reproductive forms is not observed, or are
symbionts appears irregularly wavy in the electron microscope. phylogenetically not closely related. Two species groups may
They are found in the macronucleus of P. caudatum, often presently be distinguished in the genus Holospora
clustered in the center of the nucleus. (> Table 18.1). It has been hypothesized that the unique behavior
of the infectious form to deposit enormous amounts of periplas-
mic materials polarly could be encoded on a plasmid or phage
Genus Holospora genome. However, no plasmid or phage genome has been found
(M. S. Rautian et al., unpublished results). It is consistent with
Symbionts of this genus live in the micronucleus or the macro- this observation that the ‘‘more advanced’’ species tested (of which
nucleus (see > The Ciliate Cell as a Microcosm) of Paramecium the infectious forms are collected in the separation spindle) were
and exist in two forms: a short rod 1.0–3.0 mm long and 0.5–1.0 labeled by in situ hybridization using an oligonucleotide probe
mm wide that can replicate; and a long form 5.0–20 mm long and designed for H. obtusa Fokin et al. 1996).
0.8–1 mm wide that cannot replicate (> Figs. 18.9–18.11) An unnamed symbiont of this Holospora-type has repeatedly
(Gromov and Ossipov 1981; Hafkine 1890). The long form is been observed in Stentor multiformis and in S. polymorphus
released from the host and can infect other paramecia (Görtz and Wiemann 1987); another one has been observed in
(> Fig. 18.8). The infective form is differentiated into a refractile the peritrich ciliate Zoothamnium pelagicum by Laval (1970).
portion with a less electron-dense, pale tip, and a posterior part Similar morphology and life cycle were also observed in bacterial
that contains typical bacterial cytoplasm, stains with DNA- symbionts in the ciliates Trithigmostoma cucullulus (Görtz and
specific dyes, and appears dark in phase contrast (> Fig. 18.9). Maier 1991) and Spirostomum sp. (Fokin et al. 2005). These
Nine species have been described (see > Table 18.1): Holospora observations indicate that symbionts of the genus Holospora
undulata, H. obtusa, H. elegans, H. acuminata, H. caryophila, are not restricted to Paramecium. Some strains of Paramecium
H. recta, H. curviuscula, H. bacillata, and H. curvata. were found to be harmed after infection with holosporas,
The striking similarity of the biology and cytology of the whereas other strains apparently remained unaffected.
different Holospora species led to their classification in one
genus, although a comparison of the protein patterns of Holospora undulata
H. obtusa and H. elegans revealed great differences. Moreover, This organism lives in the micronucleus of P. caudatum (Hafkine
the hybridization of the total DNAs and a comparison of DNA 1890; Gromov and Ossipov 1981). Two forms are seen: a short,
banding patterns after digestion with restriction enzymes indi- spindle-shaped, reproductive form, about 0.8 mm in diameter
cate that the two species may not be as closely related as and 1.5–2.0 mm long, and a long, spiral-shaped infectious form
. Table 18.1
Descriptions of Holospora spp.
Holosporaa Morphology and size of infectious form Host Paramecium Nucleus Species groupb
‘‘H. acuminata’’ Straight, tapered ends, length 4–6 mm P. bursaria Micronucleus I
‘‘H. bacillata’’ Straight, ends rounded, length 5–17 mm P. woodruffi Macronucleus II
P. calkinsi
H. caryophila Spiral, tapered ends, length 5–6 mm P. biaurelia Macronucleus II
P. novaurelia
P. caudatum
‘‘H. curvata’’ Curved, length 12–20 mm P. calkinsi Macronucleus II
‘‘H. curviuscula’’ Curved, tapered ends, length 6–10 mm P. bursaria Macronucleus I
H. elegans Straight, tapered ends, length 7–18 mm P. caudatum Micronucleus I
H. obtusa Straight, tapered ends, length 7–20 mm P. caudatum Macronucleus I
‘‘H. recta’’ Straight, tapered ends, length 10–15 mm P. caudatum Micronucleus I
H. undulata Spiral, tapered ends, length 7–15 mm P. caudatum Micronucleus I
a
Names in quotation marks are not validated
b
Species of group I are characterized by an escape mechanism of the infectious form from the host nucleus that is regarded as higher evolved. The infectious
forms are collected in the connecting piece of the dividing nucleus and later released by the host cell (> Fig. 18.8). Bacteria of species of group I react with
a ‘‘Holospora-specific’’ oligonucleotide probe in fluorescence in situ hybridization (Fokin et al. 1996). For species of group II, the exact mode by which infectious
forms leave their host nuclei is not known. In this group, infectious forms are not collected in the connecting piece of a dividing nucleus (Fokin et al. 1996; Fokin
and Sabaneyeva 1997). Holospora species of group II do not react with a ‘‘Holospora-specific’’ oligonucleotide probe in fluorescence in situ hybridization
that has tapered ends and is 7.0–15 mm long. The type strain is fusiform rods with tapered ends, 0.8–1.0 mm in diameter and
found in clone MI-48 of P. caudatum (deposited in the culture 4.0–6.0 mm long. The type strain is found in stock AC61-10
collection of the Laboratory of Invertebrate Zoology, Biological of P. bursaria (deposited in the culture collection of the Labora-
Research Institute, St. Petersburg State University, Russia). tory of Invertebrate Zoology, Biological Research Institute,
St. Petersburg State University, Russia).
Holospora obtusa
This organism lives in the macronucleus (> Fig. 18.7; Hafkine Holospora caryophila
1890; Gromov and Ossipov 1981). Reproductive forms are short These organisms were formerly known as ‘‘alpha particles’’
rods, 0.8 mm in diameter and 1.5–2.5 mm long. Infectious forms (Preer 1969), live in the macronucleus of P. biaurelia
are long rods with rounded ends, 0.8–1.0 mm in diameter and (> Fig. 18.2), and may also be found in P. caudatum (Görtz
7.0–20 mm long. Reflecting the fine-structural organization, the 1987; Preer and Preer 1982). Reproductive forms are thin, fusi-
infectious form shows a refractile part and a dark part in phase form rods, 0.4 mm in diameter and 1.0–3.0 mm long. Infectious
contrast light microscopy (> Fig. 18.9). The type strain is found forms are spiral shaped with tapered ends, 0.5 mm in diameter
in clone M-115 of P. caudatum (deposited in the culture collec- and 5.0–6.0 mm long. The type strain is found in ATCC strain
tion of the Laboratory of Invertebrate Zoology, Biological 30694 (stock 562 of P. tetraurelia; see Preer et al. 1974).
Research Institute, St. Petersburg State University, Russia).
Holospora recta
Holospora elegans This organism lives in the micronucleus of P. caudatum (Fokin
This organism lives in the micronucleus (Hafkine 1890; Preer 1991). Two forms are seen: a short, spindle-shaped, reproductive
and Preer 1982). Reproductive forms are short rods, 0.8 mm in form, about 0.8 mm in diameter and 1.5–2.0 mm long and a long,
diameter and 1.5–2.0 mm long. Infectious forms are long rods straight infectious form (one end rounded, one end tapered)
with tapered ends, 0.6–0.8 mm in diameter and 7.0–18 mm long. 10–15 mm long.
The type strain is found in ATCC strain 50008 (stock C1O1 of Whereas all other Holospora spp. listed here definitely appear
P. caudatum; Görtz and Dieckmann 1980). to be good species, this status has been questioned for H. recta by
Rautian and Ossipov (M. S. Rautian and D. V. Ossipov (1992),
Holospora acuminata personal communication), who found that a strain of H. elegans
This organism lives in the micronucleus of P. bursaria (Ossipov had a number of individuals with H. recta-like features. The
et al. 1980). Reproductive forms are fusiform rods, 0.6 mm in question may finally be resolved after molecular data have been
diameter and 2.0–2.5 mm long. Infectious forms are straight obtained from analyses of the different strains.
Holospora curviuscula P. duboscqui. The symbiont is 0.3 mm in diameter and 0.7–1.4

Cells live in the macronucleus of P. bursaria (Borchsenius et al. mm long, looks spindle shaped, and is Gram negative. No killing
1983). They infect only certain strains of three syngens, and in activity was observed when symbiont bearers were tested against
other strains, the development into the infectious form is not nonsymbiont bearers.
completed. Occasionally, infection of both macro- and A symbiont studied by Estève (1957) occurs in the macro-
micronuclei was observed. Reproductive forms are fusiform nucleus of P. caudatum and confers a killer trait on its host.
rods, 0.8 mm in diameter and 1.5–2.0 mm long. Infectious When it was investigated cytologically, a greatly enlarged mac-
forms are slightly curved rods with tapered ends, approximately ronucleus was observed to contain numerous kappa-like bacte-
0.8 mm in diameter and 6.0–10 mm long. ria, some of which contained R bodies. Electron micrographs of
this bacterium showed spherical phages inside the R bodies.
Holospora bacillata Schmidt and coauthors assumed that the symbiont is identical
This organism lives in the macronucleus of Paramecium calkinsi with Caedibacter caryophilus described by them (Schmidt et al.
(Fokin and Sabaneyeva 1993) and Paramecium woodruffi 1987b).
(Fokin et al. 1996; Fokin and Sabaneyeva 1997). Infectious A so far unnamed symbiont found by Görtz and Freiburg
forms are 5–17.0 mm long, 0.7–0.8 mm wide, straight, rounded (1984) living in the micronucleus of P. bursaria is a small rod,
at both ends. 0.5 mm in diameter and up to 2 mm long. Its ultrastructure
suggests that it is a Gram-negative bacterium. No flagella were
Holospora curvata found, and no killing capacity of its host was observed. Another
Cells live in the macronucleus of Paramecium calkinsi symbiont of the nonkiller type was discovered in the cytoplasm
(Fokin and Sabaneyeva 1993). Infectious forms are 12–20 mm of P. woodruffi, a ciliate living in brackish water. The symbiont is
long, 0.7–0.9 mm wide, curved, and rounded at both ends. a Gram-negative rod (0.2–0.8 mm in diameter and 0.6–2.5 mm
long) that lacks flagella and contains hexagonal viroid particles
(Fokin et al. 1987b).
Genus Paraholospora In P. sexaurelia isolated from an aquarium with tropical fish,
a bacterial endosymbiont was observed by Görtz (1981) to
Recently, a new bacterial genus Paraholospora with a single invade the macronucleus but not the micronucleus. Once
species described as ‘‘Candidatus Paraholospora nucleivisitans’’ the symbiont has entered the macronucleus, it tends to
was found (Eschbach et al. 2009). In phylogenetic trees of disappear from the cytoplasm. The bacterium is slightly
the SSU rDNA sequence, it branches at the basis of the curved with a diameter of 0.5–0.8 mm and up to 25 mm
Holosporaceae as a sister group. It differs from members of long. When present in the cytoplasm, the symbiont tends to
the genus Holospora in the intracellular cytoplasmic location be closely associated with food vacuoles. Once the symbiont
and the missing life cycle. has entered the macronucleus, it multiplies there without
causing nuclear hypertrophy. After autogamy, the symbiont
Candidatus Paraholospora nucleivisitans is found only in the cytoplasm and not in new macronu-
Cells live in the cytoplasm and occasionally in the macronucleus clear anlagen. It remains there until a new infection of the
of Paramecium sexaurelia (Eschbach et al. 2009). Bacteria are macronucleus occurs. The bacterium was observed again in
bent or spiral shaped with a diameter of 0.5–0.8 mm and up to a P. sexaurelia isolated by Fokin (Przybosz and Fokin 1997) from
25 mm in length. Cells are absent in the micronucleus. No spore a pond in the Wilhelma (Zoological and Botanical Garden,
formation or life cycle could be observed. Bacteria could not be Stuttgart).
cultivated outside their host. > Table 18.2 summarizes the taxa of prokaryotic symbionts
of Paramecium and provides a determinative key to their

identification.
Symbionts of Paramecium Without Binomial
Names
Prokaryotic Symbionts of Euplotes
Jenkins (1970) described a Gram-negative bacterium living
within bulbous distensions of the outer membrane of the Endosymbiotic bacteria are also very common in Euplotes,
nuclear envelopes of both the micro- and macronucleus of a ciliate genus that comprises both freshwater and marine spe-
a strain of P. multimicronucleatum. This symbiont is a very cies (Heckmann 1983; Petroni et al. 2001). So far, only one of the
short rod, sometimes appearing nearly coccoid, and approxi- symbionts has been given a binomial name. The others are still
mately 0.35 mm in diameter with longer forms reaching 0.7 mm referred to by Greek letters, as was formerly customary for
in length. It was named ‘‘epsilon.’’ cytoplasmic elements. Several Euplotes symbionts remained
Another symbiont, reported to occur in the perinuclear unnamed or were provisionally given Latin letters as names
space, has been described by Fokin (1988). It was found when they were encountered. However, most of these symbionts
inside the nuclear cisternae of the macronuclear envelope of are not well characterized.
. Table 18.2 . Table 18.2 (continued)

Key to the prokaryotic symbionts of Paramecium Pseudolyticum multiflagellatum
I. Host paramecia are killers or mate killers b. Host: P. primaurelia; rods, up to 2 mm long, Gram negative with
a thick surface layer visible in EM; with sparse flagella, occasionally
A. Between 2 % and 50 % of the symbiont population contains
slight motility; coexists with other symbionts such as Caedibacter
R bodies
and Pseudocaedibacter
1. Host paramecia are killers
Tectibacter vulgaris
a. Kill by producing aboral humps on sensitive paramecia; R bodies
B. Symbionts are present almost exclusively in the nuclei
unroll from inside at low pH, reroll at high pH; found in P. tetraurelia
1. Symbionts exclusively in the micronucleus, show two
Caedibacter taeniospiralis
morphologically distinct forms, and are highly infectious
b. Kill in ways other than producing aboral humps
a. Long form of symbiont (infectious form) spiral shaped, both
(1) Found in P. tetraurelia; R bodies unroll from the outside ends tapered, 7–15 mm long; host: P. caudatum
irreversibly when exposed to high temperature or certain
Holospora undulata
detergents
b. Long form of symbiont (infectious form) straight rod with
Caedibacter pseudomutans
tapered ends
(2) Found in P. biaurelia; R bodies unroll from outside irreversibly
(1) Host: P. caudatum; long form 7–18 mm long
when exposed to high temperature or certain detergents
Holospora elegans
Caedibacter varicaedens
(2) Host: P. bursaria; long form 4–6 mm long
(3) Found in P. caudatum, found in the macronucleus only; R bodies
unroll from inside Holospora acuminata
Caedibacter caryophilus c. Long form of symbiont (infectious form) straight rod, one end
tapered, one end rounded, host: P. caudatum
2. Host paramecia are mate killers
Holospora recta
Caedibacter paraconjugatus
2. Symbionts exclusively in the macronucleus
B. Symbiont population does not contain R bodies
a. Symbionts are highly infectious, two morphologically distinct
1. Host paramecia are killers
forms often observable
a. Rapid-lysis killer; symbionts are large flagellated cells
(1) Long form slightly spiral shaped, with tapered ends, 5–6 mm
(1) Straight rods found in P. tetraurelia, P. octaurelia long; host: P. biaurelia (sometimes also observed in P. caudatum)
Lyticum flagellatum Holospora caryophila (> Fig. 18.2)
(2) Sinuous rods found in P. biaurelia (2) Long form straight rod, with rounded ends, 7–20 mm long; host:
Lyticum sinuosum (> Fig. 18.3) P. caudatum
b. Kill by vacuolization, symbionts are very small cells, often (3) Long form straight rod, with rounded ends, 5–17 mm long; host:
doublets P. calkinsi or P. woodruffi
Pseudocaedibacter minutus Holospora bacillata
2. Host paramecia are mate killers (4) Long form slightly curved, with tapered ends, 6–10 mm long;
Pseudocaedibacter conjugatus host: P. bursaria
II. Host paramecia are nonkillers (5) Long form curved, both ends rounded, 12–20 mm long, host:
P. calkinsi
A. Symbionts are present only in the cytoplasm
Holospora curvata
1. Symbionts lack flagella
b. Symbionts are weakly infectious, only one morphological form
a. Host: P. bursaria; egg shaped, rarely rod shaped, up to 1.4 mm
observable. Host: P. caudatum; rods small, about 1 mm long,
long. Antagonistic to symbiotic algae (Chlorella) of the host
sometimes forming ‘‘chains’’ that are up to 10 mm long; mostly
Caedobacter chlorellopellens aggregated in the center of the macronucleus; may coexist with
b. Hosts: P. biaurelia, P. tetraurelia, P. pentaurelia; rods, up to 1.2 mm H. obtusa or H. undulata
long Nonospora macronucleata
Pseudocaedibacter falsus C. Symbionts are present in the cytoplasm but occasionally invade
c. Host: P. pentaurelia; rods, up to 1.2 mm long; the symbiont- the macronucleus
containing vacuole is surrounded by endoplasmic reticulum 1. Bacteria are bent or spiral shaped with a diameter of 0.5–0.8 mm
Pseudocaedibacter glomeratus and up to 25 mm in length. Host: Paramecium sexaurelia
2. Symbionts with flagella ‘‘Candidatus Paraholospora nucleivisitans’’ Paraholospora
a. Host: P. caudatum; rods; 3.5–14 mm long; refractile inclusions nucleivisitans
(polyhydroxy butyric acid); with numerous flagella but no motility Based on the key proposed by Preer (1981) and on information from articles
observed cited in this chapter
Habitat of the Euplotes Symbionts
All except one of the bacterial symbionts of Euplotes are confined

to the cytoplasm; see > ‘‘The Ciliate Cell as a Microcosm.’’ They
can easily be observed with a phase contrast microscope. Fixa-
tion and staining of Euplotes with aceto-carmine or lacto-aceto-
orcein can be helpful for the observation and identification of
these endosymbionts; see > ‘‘Orcein Staining of Intracellular
Bacteria.’’ When stained, Polynucleobacter necessarius (formerly
omikron) and the closely related ‘‘omikron-like’’ bacteria reveal
many nucleoids (Heckmann 1975; Heckmann et al. 1983;
> Figs. 18.19 and > 18.20). It appears that most of the Euplotes
symbionts cannot grow outside their hosts, although this has

been investigated thoroughly only for P. necessarius. Like
Caedibacter and other symbionts of the P. aurelia species com-
plex, the Euplotes symbionts appear not to be infectious, at least . Fig. 18.19
under laboratory conditions, and some are killer symbionts Fluorescence micrograph of Polynucleobacter necessarius in the
similar to the killer symbionts of Paramecium; see > ‘‘The Killer cytoplasm of a slightly crushed Euplotes aediculatus cell after
Trait in Paramecium.’’ The symbionts found to date have not staining with N,N0 -diethylpseudoisocyanin chloride. The
been observed to be harmful to their hosts, which, however, in symbionts are revealed by the DNA-specific yellow fluorescence
some cases are converted by infection into killers or mate killers. of their nucleoids. Ma macronucleus and Mi micronucleus.
Polynucleobacter necessarius and the closely related omikron-like Bar = 1 mm (From Heckmann 1975)
bacteria are essential for their hosts (Heckmann 1983; Fujishima
and Heckmann 1983).
Isolation of the Euplotes Symbionts
Euplotes minuta, E. crassus, and E. vannus are marine species.

They can be isolated from water samples containing algae or
detritus collected at the seashore. All these species can be easily
grown in the laboratory in seawater with Dunaliella salina as the
food organism (for details, see Heckmann 1963). The other
Euplotes species listed in > Table 18.3 are freshwater organisms
that may be collected from ponds. They can be grown in
a diluted soil medium (Ruthmann and Heckmann 1961) or in
culture medium for Euplotes (CME; Kuhlmann and Heckmann
1989) and fed with Chlorogonium elongatum or Chilomonas
paramecium. Most Euplotes species can also utilize bacteria as
food.
. Fig. 18.20
Since the symbionts may differ in different strains of a host
Electron micrograph of Polynucleobacter necessarius. Longitudinal
species, it is advisable to start Euplotes cultures from single cells
section. The ultrastructure of this symbiont resembles that of
so that the cultures maintained in the laboratory form clones.
Gram-negative bacteria. Bar = 0.5 mm (From Heckmann 1975)
In most cases, this will ensure that one is dealing with homoge-
neous populations of endosymbionts. However, double and
triple infections have also been observed.
Polynucleobacter necessarius is the only endosymbiont Identification of the Euplotes Symbionts
of Euplotes known so far that has been isolated. In this
case, cells of strain 15 of E. aediculatus were homogenized The bacterial symbionts of Euplotes can be detected by direct
mechanically, and the symbionts were then purified by observation with a phase contrast microscope, by observation of
applying the homogenate to an ECTEOLA column, followed physiological effects produced on other Euplotes cells that lack
by elution with phosphate buffer (Heckmann 1975; symbionts, or by a combination of these methods. In the past,
Schmidt 1982). The procedure basically followed the one identification of symbionts of Euplotes has also been done using
developed for the isolation of kappa symbionts from Parame- these methods. Now, at least for Polynucleobacter necessarius,
cium by Smith-Sonneborn and Van Wagtendonk (1964) and formerly omikron particles, identification by FISH using appro-
resulted in very clean preparations of symbionts. priate oligonucleotide probes has become possible.
. Table 18.3
Characteristics of the prokaryotic endosymbionts of the genus Euplotes
Symbiont designation Euplotes speciesa Site Killing type References

Polynucleobacter necessarius E. aediculatus (15) Cy NK Heckmann and Schmidt (1987)
Polynucleobacter sp. E. harpa Cy ND Vannini et al. (2005)
Candidatus Francisella noatunensis subsp. endociliophorad E. raikovi Cy NK Schrallhammer et al. (2010)
d
Candidatus Devosia euplotes E. magnicirratus Cy ND Vannini et al. (2004)
Candidatus Anadelfobacter velesd E. harpa Cy ND Vannini et al. (2010)
d
Candidatus Cyrtobacter comes E. harpa Cy ND Vannini et al. (2010)
Omikron-like symbiont E. eurystomus (25) Cy NK Heckmann et al. (1983)
Omikron-like symbiont E. plumipes (24) Cy NK Heckmann et al. (1983)
Omikron-like symbiont E. daidaleos (13) Cy NK Heckmann et al. (1983)
Omikron-like symbiont E. octocarinatus (11) Cy NK Heckmann et al. (1983)
Omikron-like symbiont E. patella (5) Cy NK Heckmann et al. (1983)
Omikron-like symbiont E. woodruffi (22) Cy NK Heckmann et al. (1983)
Omikron-like symbiont E. moebiusi Cy ND Foissner (1978)
Omikron-like symbiont E. harpa Cy ND Petroni et al. (2001)
Epsilon E. minuta (K1,K2, K3) Cy K Heckmann et al. (1967)
Epsilon-like E. minuta (VF17) Cy K + MK Heckmann et al. (1967)
Eta E. crassus Cy K Nobili et al. (1976)
A E. crassus Cy NK Rosati et al. (1976)
B1 E. crassus (C8) Cy MK Dini and Luporini (1976)
B3 E. crassus Cy NK Rosati et al. (1976)
C E. crassus Cy NK Rosati et al. (1976)
D E. crassus ND ND Rosati et al. (1976)
Unnamed E. crassus Ma NK Rosati and Verni (1975)
Unnamed E. crassus Cy K + MK Demar-Gervais and Genermont (1975)
Unnamedb E. aediculatus Cy NK Heckmann et al. (1983)
c
Unnamed E. octocarinatus Cy NK Heckmann et al. (1983))
Unnamed E. harpa Cy ND Petroni et al. (2001)
Abbreviations: Cy cytoplasm, Ma macronucleus, NK nonkiller, K killer, MK mate killer, and ND not determined
a
The typical strain designation is given in parentheses
b
Unnamed symbiont that is not omikron-like in E. aediculatus stocks 7 and 10
c
Unnamed symbiont observed in addition to omikron-like symbionts in stock 11 of E. octocarinatus
d
Scientific names given for uncultivated prokaryotes according to Stackebrandt et al. (2002)
The Taxa of Prokaryotic Symbionts of Euplotes Polynucleobacter necessarius

This species was formerly called ‘‘omikron’’ (Heckmann and
Genus Polynucleobacter Schmidt 1987). Cells are slightly curved rods, about 0.3 mm in
diameter and 2.5–7.5 mm long. If stained with DNA-specific
These are obligate endosymbiotic bacteria living in the cyto- dyes, usually 3–9, but in some cases, up to 12 intensely stained
plasm of freshwater ciliates of the genus Euplotes (Heckmann and regularly spaced dots become visible (> Fig. 18.19). They are
and Schmidt 1987). Characterized by multiple nucleoids, these considered to be nucleoids. When examined with the electron
symbionts are essential for their host species and are nonmotile microscope, these nucleoids differ from those of most free-living
and Gram negative. The type species is Polynucleobacter bacteria by exhibiting an electron-dense central core that resem-
necessarius. bles the chromatin of eukaryotes (> Fig. 18.20). Whether this
Polynucleobacter was found to belong to the b-subclass core is formed by proteins associated with DNA or some other
of Proteobacteria and shows the closest relationship to material is not clear (Heckmann 1975). Recently, it was shown
members of the Burkholderiaceae (Hahn et al. 2009; Springer that free-living Polynucleobacter necessarius is distributed world-
et al. 1996). wide (Hahn 2003). Polynucleobacter necessarius strains revealed
the closest phylogenetic relationship between free-living and (Vannini et al. 2005). The bacteria were located in the cytoplasm
obligate endosymbiotic bacteria ever found (Vannini et al. of the ciliate cells embraced by a peribacterial membrane. Cells
2007). As a result, the revision of P. necessarius led to the of E. harpa were found to be associated with a second endosym-
differentiation of free-living and endosymbiotic species at the biotic prokaryote affiliated with a-proteobacteria.
subspecies level. The free-living subspecies is proposed to be
named Polynucleobacter necessarius subsp. asymbioticus, while Candidatus Devosia euplotes Cells are about 0.5 mm wide and
the endosymbiotic subspecies is named P. necessarius up to 2.5 mm long and located embraced by a peribacterial
subsp. necessarius (Hahn et al. 2009). Both subspecies display membrane in the cytoplasm of their host Euplotes magnicirratus
a 99.1–99.4 % sequence similarity for the 16S rDNA gene. They (Vannini et al. 2004). Bacteria belong to a-proteobacteria.
differ significantly in genome size, visibility of nucleoids, cell
length, and culturability (Hahn et al. 2009). Candidatus Francisella noatunensis subsp. endociliophora Bac-
The symbionts are individually contained in vesicles, to terial Gram-negative endosymbionts are between 0.2–0.4 mm
which ribosomes are often attached. Polynucleobacter necessarius wide and 0.9–1.9 mm long and located dispersed in the cyto-
subsp. necessarius reproduces by transverse binary fission in a plasm of Euplotes raikovi without surrounding membranes
typical bacterial manner. However, the fission products have often (Schrallhammer et al. 2010).
been found to differ in size. Frequently, a 7 mm long rod containing
eight to nine nucleoids was observed to bud off a 2.5 mm piece that Candidatus Anadelfobacter veles These are Gram-negative bac-
contained three nucleoids (Heckmann 1975). teria of the order Rickettsiales, inhabiting the cytoplasm of Euplotes
The G + C content of the DNA of endosymbiotic harpa (Vannini et al. 2010). Cells are rod shaped and up to 2 mm
P. necessarius is 48 mol% by thermal denaturation. A value that wide and 6 mm long and surrounded by a host membrane within
was 2.8 mol% lower was found when the G + C content was the cytoplasm of the ciliate. Found in an coinfection with bacteria
calculated from the buoyant density, which was 1.7036 g/cm3 of the genus Polynucleobacter (b-Proteobacteria).
(Schmidt 1982). The differences could be caused by rare bases;
however, no DNA chemistry has been done for intracellular Candidatus Cyrtobacter comes These are rod-shaped
P. necessarius. The average DNA content of the symbiont was a-proteobacteria of the order Rickettsiales with an irregular sur-
determined to be 5.8 10–3 pg. Taking into account the average face, visible in ultrathin section for transmission electron micros-
number of nucleoids, a DNA content of 0.5 109 Da per copy (Vannini et al. 2010). Cells are up to 3 mm wide, up to 5 mm
nucleoid was calculated. This value is in close agreement with long and living without a surrounding peribacterial membrane
the value of the kinetic complexity of the DNA, which was deter- in the cytoplasm of Euplotes harpa. Found in an coinfection with
mined to be 0.57 109 Da when corrected for the G + C content bacteria of the genus Polynucleobacter (b-Proteobacteria).
(Schmidt 1982). Hahn et al. (2009) determined a genome size of
1.5–2.5Mbp for both subspecies of P. necessarius and estimated
a G + C content of about 44–46 mol%. Equally, small genomes The Omikron-Like Endosymbionts of
have been reported for Lyticum flagellatum, Pseudocaedibacter Euplotes
conjugatus, and P. falsus, endosymbionts of the P. aurelia species
complex (Soldo and Godoy 1973a); for xenosomes of These bacteria are very similar in appearance to P. necessarius.
Parauronema acutum (Figueroa-de Soto and Soldo 1977); and They were found in six freshwater Euplotes species
for chlamydias, rickettsias, and mycoplasmas (Bak et al. 1969). (E. eurystomus, E. plumipes, E. octocarinatus, E. patella, and
The genomes of these organisms are the smallest known for cells. E. woodruffi) all having 9 type 1 fronto-ventral cirrus (FVC)
They code for only about 700–900 proteins. pattern (9 FVC double dargyrome; Gates and Curds 1979) like
Heckmann (1975) has shown that it is possible to remove E. aediculatus, the bearer of omikron, and are considered to be
P. necessarius from its host E. aediculatus by treating a rapidly closely related. The symbionts may differ from P. necessarius in
growing culture with penicillin (100–500 units/ml) for 5–6 days. size, shape, and other features, but they always share with
Aposymbiotic hosts may undergo one or two fissions but then P. necessarius the characteristic of multiple nucleoids
stop multiplying and die about 15–20 days after the last fission. (Heckmann et al. 1983). Symbionts of this type, but about
The same results were obtained with several other antibiotics or twice as large as P. necessarius, were first noticed by Fauré-
with sufficiently high doses of X-rays. Fremiet (1952) in stocks of E. patella and E. eurystomus. He
Reinfection and rescue of E. aediculatus have been achieved also observed at that time that small doses of penicillin led to
either by adding a cell homogenate from symbiont bearers or by a loss of symbionts and to death of the host ciliates. He therefore
injecting symbiont-containing cytoplasm (Heckmann 1975; suggested that the symbionts might be essential for survival of
Fujishima and Heckmann 1984a). The type strain is found in the ciliate. This was later confirmed when Heckmann et al.
ATCC strain 30859 (clone 15 of E. aediculatus; see Heckmann (1983) subjected a large number of stocks of Euplotes species
and Schmidt 1987). with a 9 type 1 cirrus pattern to a penicillin treatment of the kind
A phylogenetically and morphologically different endosymbi- that had earlier been found to remove P. necessarius. They
otic Polynucleobacter was described in Euplotes harpa from found that all the ciliates stopped dividing and eventually died
a brackish water of river estuaries of different geographical regions when their symbionts were removed. Heckmann et al. (1986)
suggest that all Euplotes species with a 9 type 1 cirrus pattern into a healthy-looking exconjugant, whereas the other one failed
suffer from a common deficiency that arose in a common ances- to develop a large macronuclear anlage, became quiescent, and
tor of this group of organisms. This ancestor must have lived in finally died. The surviving exconjugant clones had the same
symbiosis with a prokaryote that compensated for the acquired cortical pattern as stock VF17 and were all found to act as killers.
deficiency. Polynucleobacter necessarius and the omikron-like They were therefore cytoplasmic descendants of the mate-killer
symbionts are considered to be progeny of this prokaryote. parent. That they were true hybrids, i.e., that they had received
a gametic nucleus from the sensitive partner, was shown from
presence of certain genetic markers followed in these crosses.
Symbionts of Euplotes without Binomial Names
Whether VF17 was host to a mixed population of symbionts—
with one type responsible for the killer trait and the other one for
As for the symbionts of Paramecium, many bacterial symbionts
the mate-killer trait, both types being so similar that they could
of Euplotes have not been given binomial names. For the time
not be distinguished morphologically from each other—or
being, there is no information about their phylogenetic position.
whether a single class of particles present in VF17 determined
Nevertheless, they are included because they may be identified in
both traits cannot be decided. Both the killer trait and the
further isolations by their host specificities and other distinct
mate-killer trait were lost concurrently with aging of the stock
biological traits.
(Frankel 1973).
The Epsilon Symbionts of Euplotes

The Eta Symbiont of Euplotes
The epsilon symbionts from killer strains K1, K3, and K7 of
E. minuta, collected from the Mediterranean Sea at Villefranche- The eta symbiont, which was observed only by electron micros-
sur-Mer (France), were described by Heckmann et al. (1967). They copy, is round or oval shaped with an average diameter of 0.9
are small rods, about 0.4 mm in diameter and 0.8–2.5 mm long, mm. It is bound by three juxtaposed membranes. The outermost
when observed in freshly crushed cells with bright phase optics, and membrane has ribosomes attached to it. The interior of the
appear somewhat darker than free-living bacteria. When the killer symbiont is granular with a few scattered fibrous strands (Rosati
strain K3 was fixed in Schaudinn’s fluid, hydrolyzed in 1 N HCl, and Verni 1977). The symbiont was found in strains of E. crassus
and stained with basic fuchsin, as described by Dippell and Chao collected by Luporini (1974) from a site on the coast of Somalia
in Sonneborn (1950), several dozen darkly stained bodies were on the Indian Ocean. It was named ‘‘eta’’ by Nobili et al. (1976),
found scattered in the cytoplasm. Their staining properties, size, who investigated the killer trait conferred on its host by this
shape, and cellular distribution resembled that of Caedibacter symbiont. Sensitive testers, when exposed to the culture fluid of
species and other endosymbionts of Paramecium. Since epsilon killer strains or to a homogenate prepared from cells of these
particles were found only in killer strains, it is assumed that they lines, developed a large vacuole in the posterior part of the body
are responsible for the killer phenotype of E. minuta. within 6–24 h. Affected cells are gradually transformed into
Siegel and Heckmann (1966) noticed that certain strains of transparent spheres that eventually ‘‘explode.’’
E. minuta killed certain other strains of this species when con-
jugation occurred. It was found that the sensitive cells formed
vacuoles, ceased their normal swimming movements, settled to The B1 Symbiont of Euplotes
the bottom of the culture vessel, and finally lysed. These changes
occurred within 1–4 days after the strains had been mixed. No A symbiont very similar in morphology to eta appears to be
evidence for killing was observed when cells of the three killer responsible for the mate-killer properties of strain C8 of
strains were mixed together. E. crassus; it was named ‘‘B8.’’ When cells of this strain were
A good example of the difficulties encountered in the iden- treated with penicillin, both the mate-killer trait and the sym-
tification of endosymbionts is provided by the killer particles of biont disappeared. Dini and Luporini (1982) provided evidence
stock VF17 of E. minuta. Heckmann et al. (1967) had noticed for the existence of specific host genes required for maintenance
that this strain caused symptoms in sensitives slightly different of the B8 particles.
from those observed in mixtures with the other E. minuta killers.
Here, vacuole coalescence generally proceeded to the point
where only one large vacuole was present in the posterior region Inconspicuous Symbionts of Euplotes
of the sensitive, giving the whole affected organism a pear-
shaped form. Nevertheless, no morphological differences The other symbionts of Euplotes listed in > Table 18.3 (A, B3, C,
between these symbionts and epsilon particles could be detected. D, and the unnamed ones) have been discovered by chance during
Later, when stock VF17 was employed for cross-breeding exper- electron microscopic investigations of certain strains. No special
iments in a study of the genetic control of cortical pattern in functions of these symbionts are known nor have any effects of the
E. minuta, it was found that this stock not only killed sensitive symbionts on their hosts been reported. Interestingly, in two
strains via particles liberated into the medium but also acted as cases, they have been observed in addition to omikron-like sym-
a mate killer. From a pair of conjugants, one generally developed bionts (Heckmann et al. 1983; Petroni et al. 2001).
Ectosymbionts of Euplotidium (Rosati et al. 1993a). On top of the extrusive structure in the
distal part of the cell body, an electron-dense, dome-shaped
Peculiar ectosymbionts have been observed on the surface of the structure lies underneath the cell membrane (> Fig. 18.23).
marine ciliate Euplotidium itoi (> Fig. 18.21) and other species The dome-shaped structure contains DNA and basic proteins,
of the genus (Verni and Rosati 1990). Though they have been and its ultrastructure resembles eukaryotic chromatin. The
studied in detail, their phylogenetic position remained obscure extrusive apparatus may be ejected upon appropriate stimuli
over years until their relationship to Verrucomicrobia was eluci- (Rosati et al. 1993b).
dated (Petroni et al. 2000). It is the only ciliate symbiont belong-
ing to Verrucomicrobia, while most symbionts of ciliates belong
to the Proteobacteria. The authors found two morphologically Habitat and Biology of Epixenosomes of
different forms of the particles. At stage I of the life cycle, the Euplotidium
organisms are spherical (1 mm in diameter; > Fig. 18.22)
and have a simple bacteria-like organization of their cell. The ciliate hosts of epixenosomes, Euplotidium itoi and
These cells apparently divide by binary fission. Stage E. arenarium, were collected from tide pools along the rocky
I epixenosomes may transform into stage II by gradually shore of the Ligurian Sea. As has been shown by Rosati, Verni,
acquiring a more complex structure. Fully developed, stage II and collaborators, epixenosomes have defensive function for the
epixenosomes are larger (2.5 mm long) and egg shaped host cell (Rosati et al. 1999). Predators such as the ciliate
(> Fig. 18.23). They contain a sophisticated extrusive structure Litonotus easily ingested Euplotidium without epixenosomes,
of a ribbon coiled around a central core surrounded by a basket whereas they were not able to ingest Euplotidium with
built of bundles of regularly arranged tubules. The tubules have epixenosomes unless the ejecting capacity of the bacteria was
a number of features in common with eukaryotic microtubules inhibited.
. Fig. 18.21
Scanning electron micrograph of Euplotidium itoi, dorsal view. . Fig. 18.22
Epixenosomes (arrows) are arranged in a band on the cortex of the Epixenosomes (arrow) of stage I. Transmission electron
ciliate. Bar = 100 mm (From Petroni et al. 2000. © 2000 National micrograph of thin section. Bar = 1 mm (From Petroni et al. 2000.
Academy of Sciences, USA) © 2000 National Academy of Sciences, USA)
The Xenosomes of Parauronema acutum
The term ‘‘xenosome’’ was coined by Soldo to denote the

infectious particles that he found in the cytoplasm of the marine
ciliate Parauronema acutum (Soldo et al. 1974). They were later
shown to be small bacteria that possess multicopy genomes
and resemble Caedibacter taeniospiralis (kappa) and other
cytoplasmic symbionts of Paramecium in many respects; see
> ‘‘The Prokaryotic Symbionts of Paramecium.’’ Two types of
xenosomes were distinguished: killer xenosomes, which inhibit

growth when taken up by susceptible ciliates, especially those of
the genus Uronema (Soldo and Brickson 1978), and nonkiller
xenosomes (Soldo et al. 1987). The xenosomes have not yet
received binomial names.
Habitat of the Xenosomes of Parauronema

acutum
Parauronema acutum is a small hymenostome marine ciliate that

can be maintained axenically. The culture medium is basically
that developed for species of the Paramecium aurelia complex by
Soldo and Van Wagtendonk (1969), distilled water being
replaced by seawater (density 1.015–1.026 g/ml) and modified
to contain asolectin (500 mg/ml) as the sole source of lipid
(Soldo et al. 1974). The association with xenosomes is stable.
Symbiont-bearing strains have now been maintained for more
than 15 years.
. Fig. 18.23 Xenosomes released from host cells can infect symbiont-free
Epixenosomes of stage II. Sections of two bacteria at different strains of P. acutum. A single xenosome is capable of infecting
levels. DZ apical (dome-shaped) zone, EA extrusive apparatus, and a susceptible cell. However, a threshold of 100–200 xenosomes
BT microtubule-like elements forming a basket around the appears to be required before a single xenosome can infect
extrusive apparatus. Bar = 1 mm (From Petroni et al. 2000. © 2000 a potential host (Soldo 1983). While phagocytosis is the usual
National Academy of Sciences, USA) way infectious bacteria enter host ciliates, Soldo and Brickson
(1978) found that the symbionts in Parauronema acutum,
termed ‘‘xenosomes’’ by those authors, penetrate the cell mem-
Epixenosomes are arranged in a broad cortical band on the brane at sites of the cortex other than the oral apparatus.
surface of host cells (> Fig. 18.21). Individual bacteria appear Soldo and coworkers measured the oxygen consumption of
fixed in cup-shaped indentations of the ciliate cortex. At these purified xenosome preparations and reported a rate of oxygen
sites, the cortex of host cells is different in symbiont-bearing uptake of 1.3 nmoles O2/min/mg of protein, which is higher
ciliates compared to symbiont-free cells (Verni and Rosati 1990; than the oxygen uptake of Bdellovibrio and of the same order of
Rosati 1999). magnitude as that of rickettsiae and Caedibacter taeniospiralis.
Oxygen uptake is about 20 times lower than that of E. coli. The
rate of oxygen consumption was found to be stimulated by
Identification of Epixenosomes of Euplotidium various fatty acids, by intermediates of the glycolytic pathway,
and by intermediates of the citrate cycle with the exception of
While it appears easy to identify epixenosomes of Euplotidium citrate itself, which had no effect. Cyanide was found to be
as such by their unique site specificity, arrangement on the a potent inhibitor of oxygen consumption of xenosomes
host cell, and ultrastructure, their phylogenetic position long (Soldo 1983).
remained obscure. Earlier attempts of classification of these
organisms did not give definite results. Only by using
a Eukaryotes forward primer and a backward primer designed Isolation of Xenosomes from Parauronema
for Archaea did amplification of the small ribosomal subunit acutum
(SSU) rRNA gene become possible (Petroni et al. 2000). A probe
designed for part of the sequence labeled epixenosomes on the Xenosomes can be isolated from Parauronema acutum by the
host E. arenarium. same procedures as those used for the isolation of Paramecium
symbionts (Soldo and Godoy 1974). It has been shown that Fenchel 1989; Stumm and Vogels 1989; Fenchel and Finlay
killer xenosomes isolated in this way and purified with the 1991a; Narayanan et al. 2009). It has been shown that anaerobic
help of Percoll gradients retain their ability to infect P. acutum ciliates generate their energy by converting carbohydrates to
and remain capable of killing sensitive Uronema strains lactate, acetate, and butyrate and that they can remove reducing
(Soldo et al. 1986a). Results obtained from experiments in equivalents in the form of H2 (Müller 1988). Production of H2
which isolated killer xenosomes lost their capacity to kill after by proton reduction involves the enzyme hydrogenase. However,
they had been treated with various enzymes or had been coated this enzyme functions well only if the concentration of H2 is kept
with antibodies directed against the xenosomes indicate that the low (<10–5 atm). This requirement is apparently achieved by
toxic principle of the xenosomes is a protein present at or near methanogenic bacteria, which consume H2 and tend to be
the surface (Soldo 1987). abundant in these habitats.
Identification of Xenosomes from Habitat of Prokaryotic Symbionts of Ciliates

Parauronema acutum from Anaerobic Environments
These symbionts are rod-shaped, Gram-negative bacteria, 0.3 An episymbiotic association with methanogenic bacteria was
mm in diameter and 0.8 mm long. The G + C content of the DNA described for 11 species of rumen ciliates of the family
is 33.9 mol%. The symbionts occur in the cytoplasm of the Ophryoscolecidae (Vogels et al. 1980). The attached bacteria
marine ciliate Parauronema acutum in numbers ranging from were rods 0.9–3.8 mm long and 0.6–0.7 mm wide that occurred
50 to 300 per cell. Negative staining reveals the presence of as clusters or long chains. The symbionts were identified as
flagella, which provide the symbionts with a spinning and methanogens (probably Methanobrevibacter ruminantium) on
darting motility when released from the host (Soldo 1987). the basis of specific fluorescent coenzymes (F350 and F420;
The xenosomes were found to contain inclusions in the form > Figs. 18.24 and > 18.25). Experiments with a fistulated sheep
of helical arrays. These structures (called ‘‘H bodies’’) are about revealed a decrease of the association frequency of methanogenic
0.6 mm long and 0.026 mm wide. They occur singly and in bacteria with rumen ciliates when the hydrogen concentration in
multiples and extend almost the entire length of the symbiont. the rumen fluid increased and the reverse when hydrogen
The H bodies were found in over 50 % of killer and nonkiller became scarce (Stumm et al. 1982). This is interpreted in favor
xenosomes (Soldo et al. 1987). of an interspecies transfer of hydrogen. The finding that rumen
The genome size of xenosomes was found to be only 515 kbp, ciliates have hydrogenosomes, showing strong hydrogenase
which is much smaller than that of free-living bacteria. Analyt- activity, supports this view (Yarlett et al. 1981). Stumm and
ical measurements and data from sedimentation rate analyses Vogels (1989) propose that the attachment of hydrogen-
led to the conclusion that the chromosomal DNA exists in the consuming methanogens to rumen ciliates facilitates hydrogen
form of nine circularly permuted double-stranded DNA mole- removal, which is advantageous to both the ciliates and to their
cules, each about 515 kbp in length (Soldo et al. 1983). Both the episymbionts. Fenchel and Finlay (1990; 1991a), however, have
small size and the multicopy nature of the genome are typical of argued that the symbiosis is not of great advantage for the
bacterial symbionts rather than for free-living bacteria (Soldo ciliates, since the external hydrogen tension is low and diffusion
and Godoy 1973a; Soldo 1987). pathways are very short, the problem being greater for the
Killer and nonkiller xenosomes both contain plasmids bacteria to get hold of the H2 that is produced than it is for
(Soldo et al. 1986b; Soldo 1987). The nonkiller plasmids consist the ciliate to get rid of it. Fenchel and Finlay (1991b) state
of two circular DNA duplexes, each 63 kbp in length. The killer that the presence of methanogens does not seem to be vital
plasmid consists of four circular 63-kbp DNA duplexes. When to the ciliates in any case that has been studied.
a ciliate, which previously harbored killer xenosomes but had Episymbiotic bacteria have been known from sand-dwelling
been freed of them, is infected with nonkiller xenosomes, it ciliates of anaerobic habitats for many years (Sauerbrey 1928;
becomes a killer. Soldo et al. (1986b) report that together with Kahl 1933, 1935; Fauré-Fremiet 1950b, 1951). They are firmly
this transformation, the restriction pattern of the extrachromo- attached to the cell surface and appear to differ in size, shape,
somal DNA is altered. The mechanism by which this transfor- pigmentation, and wall structure from species to species
mation takes place is unknown. (Fenchel et al. 1977; Fenchel and Finlay 1991a). Fauré-Fremiet
(1950a) carried out a study of the symbionts of two species of the
genus Kentrophoros (formerly Centrophorella). He investigated
Prokaryotic Symbionts of Ciliates from their symbionts and found them to be attached to the
Anaerobic Environments nonciliated dorsal side of these ribbon-shaped ciliates. The
oblong symbionts were densely packed and protruded perpen-
Ciliates living in an anaerobic habitat, such as the rumen or dicularly so that they appeared like bristles of a brush. They were
sewage sludge rich in hydrogen sulfide, show special adapta- Gram-negative, nonmotile rods and were observed to divide by
tions. They usually lack mitochondria, bear hydrogenosomes, longitudinal fission. Besides containing rather large quantities of
and are often associated with methanogenic bacteria (Finlay and polysaccharides (indicated by iodine treatment), they contained
. Fig. 18.24
Metopus contortus containing methanogenic endosymbionts: The methanogenic bacteria are located parallel to the kineties (ciliary
rows) or the inner side of the cell. Cells are fixed with 1.2 % formaldehyde and 0.3 % glutaraldehyde. (a) Bright field micrograph. (b)
Epifluorescence micrograph. Bars = 10 mm (From Van Bruggen et al. 1986)
dark refractive sulfur globules so that the host appeared black.

The dark pigmentation is a strong indication that the symbionts
are purple sulfur bacteria (Chromatiaceae). Fauré-Fremiet, who
found the symbionts exclusively on the ciliates and neither on
accompanying sand grains nor on glass slides that had been in
the sand in which these ciliates lived for several days, considered
the symbionts as obligately episymbiotic. Electron microscope
studies by Raikov (1971, 1974) confirmed Fauré-Fremiet’s
observations and led to the discovery that the episymbiotic
bacteria are phagocytized. He studied Kentrophoros fistulosum
and K. latum and found that these ciliates take up the
episymbiotic bacteria by pseudopodia-like cytoplasmic protru-
sions and enclose them in food vacuoles. This may occur at any
place on the nonciliated body side of these ciliates, which have
no special mouth structures. The unique form of phagocytosis
discovered in these species has been named ‘‘random phagocy-
tosis.’’ Based on the content of the food vacuole of these ciliates,
their epizoic bacteria appear to be their main nutrition source.
Fourteen Kentrophoros species are now known. They all live in
the interstices of sandy marine sediments.
Fenchel et al. (1977) examined marine sediment-dwelling
ciliates for cytochrome oxidase activity and for fine-structural
details. They found that many of the ciliates of this habitat lacked
cytochrome oxidase activity and mitochondria but contained
microbodies that were identified as hydrogenosomes (Van
Bruggen et al. 1986). In addition, these ciliates harbored
. Fig. 18.25 a species-specific flora of epi- and endosymbiotic bacteria.
Autofluorescence of endosymbiotic methanogens in Metopus Large numbers of symbionts per cell were observed, ranging
palaeformis (From Finlay and Fenchel 1992) from about 1,000 bacteria per ciliate, corresponding to less
than 1 % of the host’s biomass, to about 100,000 bacteria per

ciliate, corresponding to about 20 % of the ciliate’s biomass.
Some of these anaerobic ciliates were found to contain both epi-
and endosymbionts, and often more than one type of
episymbiont was recognized. Fenchel et al. (1977) hypothesized
that the bacteria and the ciliates interact metabolically.
Anaerobic conditions also exist in sediments that are rich in
decaying plant material, such as those of freshwater ponds, lakes,
ditches, swamps, and wastewater bioreactors. This habitat, also
called ‘‘Faulschlamm’’ in German, was named ‘‘Sapropel’’ by
Lauterborn (1901). Endosymbiotic bacteria in sapropelic ciliates
were first described by Fauré- Fremiet (1909). Later, Liebmann
(1937, 1938) noted that all the sapropelic ciliates that he inves-
tigated contained rodlike bacteria in their cytoplasm. In recent
years, this habitat was thoroughly investigated (Stumm and
Vogels 1989; Fenchel and Finlay 1991a; Hackstein and Stumm
1994), and it is now, next to the rumen, the best-studied anaer-
obic habitat. It is inhabited by large numbers of anaerobic pro-
tozoa and methanogenic bacteria. A survey of sapropelic ciliates
by means of fluorescent microscopy revealed the presence of
methanogenic bacteria inside the cells; they were spread
throughout the cytoplasm in considerable quantities
(Van Bruggen et al. 1983).
Electron microscopic investigations of sapropelic ciliates
revealed the absence of mitochondria and the presence of
microbodies. Because the bacteria were found to consume
hydrogen, it has been hypothesized that the symbiont-associated
microbodies are hydrogenosomes (Van Bruggen et al. 1986). In
Metopus striatus, a Gram-positive rod-shaped bacterium was
regularly found to be in close association with
a hydrogenosome consisting of a granular matrix surrounded . Fig. 18.26
by a membrane (Van Bruggen et al. 1984). The bacterium was Symbiotic methanogen with hydrogenosomes in a transmission
isolated and identified as Methanobacterium formicicum. electron micrograph of the ciliate Metopus contortus (From Finlay
A similar or even closer association of hydrogenosomes and and Fenchel 1989. Reprinted from Finlay and Fenchel 1989,
methanogenic bacteria is reported for Metopus contortus (Van © 1989, with permission from Elsevier Science)
Bruggen et al. 1986; Finlay and Fenchel 1989; > Fig. 18.26),
Plagiopyla nasuta (Goosen et al. 1988), and Plagiopyla frontata
(Embley and Finlay 1994). A direct proof was provided by Zwart endosymbionts which are more closely related to free-living
et al. (1988), who demonstrated hydrogenase activity in methanogens than to each other (Embley and Finlay 1994).
microbodies of Plagiopyla nasuta and Trimyema compressum. While many marine anaerobic ciliates have endosymbiotic
Whereas methanogens are tightly packed with hydrogenosomes methanogens, ectosymbiotic bacteria that are often found in
in ciliates such as M. contortus and Plagiopyla frontata addition to the endosymbionts are never methanogens. Fenchel
(> Fig. 18.27) and Trimyema species (e.g., Finlay and Fenchel and Finlay (1991a) assumed that the ectosymbionts are sulfate-
1989, 1991a), the long methanogens found in Metopus reducing bacteria. Epibiotic bacteria have been reported also on
palaeformis are not closely associated (Finlay and Fenchel 1991; the peritrichous ciliate Zoothamnium niveum collected in about
> Fig. 18.28). Symbionts in this ciliate appear to divide simul- 0.5–2.0 m depth on the vertical walls of a tidal channel cut into
taneously, and growth rates of ciliates and methanogens are mangrove peat off the island of Twin Cays (Belize Barrier Reef,
approximately equivalent (Fenchel and Finlay 1991b; Finlay Carribean Sea; Bauer-Nebelsick et al. 1996a, b). The association
and Fenchel 1992). A comparison of the phylogenetic diversity appeared to be obligatory (the ciliate being unable to survive
of endosymbiotic methanogens and anaerobic ciliates revealed without its symbiont), and it was also shown that oxygen and
that symbioses have formed repeatedly and independently sulfide were needed simultaneously. The white color of the bacte-
(Finlay et al. 1993; Embley and Finlay 1994) with endosymbionts ria is assumed to represent inclusions of elemental sulfur used as
from the genus Methanobacterium and from the Methanomi- storage within the sulfide-oxidizing process. The bacteria were
crobiales. The authors found no congruence between the host and proven to be autotrophic. Epibiotic bacteria have been reported
endosymbiont trees and thus no evidence of parallel speciation. for a number of different species of Zoothamnium (cited in Bauer-
Congeneric hosts in the genus Plagiopyla and Metopus contain Nebelsick et al. 1996a) but have not been further investigated.
. Fig. 18.27
Stacks of alternating hydrogenosomes (dark bodies) and
methanogens in a transmission electron micrograph of Plagiopyla
frontata (From Embley and Finlay 1994)
. Fig. 18.28
A transmission electron micrograph of Metopus palaeformis with
long rod-shaped methanogens. h hydrogenosomes and dv
Isolation of Prokaryotic Symbionts of Ciliates digestive vacuole. Bar = 1 mm (From Finlay and Fenchel 1991)
from Anaerobic Environments
For enrichment and culture of sapropelic ciliates, samples from

anoxic natural sediments are introduced into mineral media Identification of Prokaryotic Symbionts of
under reducing conditions. Two media which have successfully Ciliates from Anaerobic Environments
been used for this purpose are given by Wagener and Pfennig
(1987). They also reported the first monoxenic large-scale The symbionts of several sapropelic ciliates have been identified
culture of an anaerobic ciliate. They enriched the ciliate as methanogenic bacteria. The endosymbionts of species of the
Trimyema compressum from anoxic mud samples and then ciliate Metopus (> Fig. 18.24) were first recognized to be
established a pure culture. The ciliate was fed with a bacterial methanogens by their autofluorescence when irradiated with
strain isolated from the enrichment culture that proved capable short-wavelength blue light (Van Bruggen et al. 1983, 1986).
of serving as food. During continued cultivation, T. compressum According to Doddema and Vogels (1978), epifluorescence
gradually lost its endosymbionts. microscopy can detect the presence of the deazaflavin coen-
An isolation method for methanogenic bacteria from zyme F420 and the pterin compound F342, both of which are
sapropelic ciliates has been described by Van Bruggen et al. specific for methanogens. The methanogenic character of
(1986): A small number of ciliates is collected with a pipette the endosymbionts has subsequently been proven by studies
and is washed free of potential contaminants in an isolation of the isolated symbionts and measurements of their meth-
medium under a dissection microscope. The microscope is ane production in situ. The symbionts were finally identi-
covered with a plastic hood flushed with nitrogen. The ciliates fied as Methanobacterium formicicum (Van Bruggen et al.
are then homogenized in an anaerobic glove box. The homog- 1984). The methanogenic symbionts of Plagiopyla nasuta were
enate with the symbionts is plated on solid isolation medium also identified as M. formicicum (Goosen et al. 1988). This is
containing penicillin G or lysozyme, which does not affect a bacterium that often occurs in the sapropel; it is a slender,
methanogenic bacteria. The homogenate is incubated at nonmotile rod, with a diameter of 0.4 mm and a length of 2–7
22–37 C under an atmosphere of H2/CO2 (80/20). In liquid mm. It was identified by colony form, cell morphology, temper-
media, growth can be followed by measuring the methane ature and pH optimum, substrate specificity, DNA base compo-
production. sition, and type of coenzyme F420. It appears likely that
M. formicicum is preadapted for endosymbiosis because Amann R, Springer N, Ludwig W, Görtz H-D, Schleifer K-H (1991) Identification
in situ and phylogeny of uncultured bacterial endosymbionts. Nature
symbiont-free lines of the ciliate Trimyema compressum could
351:161–164
also successfully be infected with this bacterium (Wagener Amann RI, Ludwig W, Schleifer K-H (1995) Phylogenetic identification and in
et al. 1990). situ detection of individual microbial cells without cultivation. Microbiol
Methanoplanus endosymbiosus was isolated from the marine Rev 59:143–169
ciliate Metopus contortus (Van Bruggen et al. 1986). In liquid Bak AL, Black FT, Christiansen C, Freundt EA (1969) Genome size of mycoplas-
mal DNA. Nature 224:1209–1210
media, the isolated bacteria were irregular, nonmotile,
Ball GH (1969) Organisms living on and in protozoa. In: Chen TT (ed) Research
nonsporeforming discs with a diameter of 1.6–3.4 mm. In in protozoology. Pergamon, Oxford, pp 565–718
a side view, they appeared as rods that were sometimes Bauer-Nebelsick M, Bardele CF, Ott J (1996a) A symbiosis between Zoothamnium
branched. The symbionts were osmotically fragile and lysed niveum (Hemprich and Ehrenberg 1831) Ehrenberg 1838 (Oligohyme-
immediately when suspended in water. Growth and nophorea, Peritrichida) and thiotrophic bacteria. I: redescription of the
species based on light and scanning electron microscopic technique. Eur
methanogenesis were observed with H2/CO2 or formate as sub-
J Protistol 32:18–30
strates, with generation times of 7 or 12 h, respectively. The Bauer-Nebelsick M, Bardele CF, Ott J (1996b) Electron microscopic studies of
temperature range from growth was between 16 and 36 C, Zoothamnium niveum (Hemprich & Ehrenberg 1831) Ehrenberg 1838
with an optimum at 32 C. The optimal pH range for growth (Oligohymenophorea, Peritrichida), a ciliate with ectosymbiotic,
was 6.8–7.3. Tungsten and NaCl at a concentration of chemmoautotrophic bacteria. Eur J Protistol 32:202–215
Beale GH, Jurand A (1966) Three different types of mate-killer (mu) particle in
0.25 M were required for optimal growth. The G + C content
Paramecium aurelia (syngen 1). J Cell Sci 1:31–34
of the DNA is 38.7 mol%. Beale GH, Jurand A, Preer JR Jr (1969) The classes of endosymbionts of Parame-
Methanosaeta sp. found as endosymbionts in the ciliate cium aurelia. J Cell Sci 5:65–91
Metopus es (Narayanan et al. 2009). Bacteria are belonging to Bedingfield G, Gibson I, Horne RW (1984) A comparative study of the structure
archaeal methanogens. Archaea are rod shaped with an average of isolated refractile bodies (R-bodies) from Paramecium 1: the effect of
treatment with EDTA and EGTA. Micron Microsc Acta 15:235–246
length of about 3.4 mm which can be cultivated in a
Beier CL, Horn M, Michel R, Schweikert M, Görtz H-D, Wagner M (2002) The
Methanosaeta-specific medium. genus Caedibacter comprises endosymbionts of Paramecium spp. related to
For detecting methanogenic endosymbionts, epifluorescence the Rickettsiales (Alphaproteobacteria) and to Francisella tularensis
microscopy proved to be a powerful method. Even when (Gammaproteobacteria). Appl Environ Microbiol 68(12):6043–6050
symbionts are inside cells, their blue autofluorescence may Borchsenius ON, Skoblo II, Ossipov DV (1983) Holospora curviuscula – a new
species of macronuclear symbiotic bacteria of Paramecium bursaria.
be visible. Endosymbionts that fluoresce have been observed
Cytologia 25:91–97
in the sapropelic ciliates Brachonella spiralis, Caenomorpha Boss AO-L, Borchsenius ON, Ossipov DV (1987) Pseudolyticum multiflagellatum
medusula, C. universalis, Lacrymaria cucumis, Metopus es, n. g, n. sp. – a new symbiotic bacterium in the cytoplasm of Paramecium
M. laminarius (Van Bruggen et al. 1983), Parablepharisma caudatum (Ciliata, Protozoa). Cytologia 29:94–97
pellitum, Metopus palaeformis, and others (Fenchel and Finlay Bütschli O (1876) Studien über die ersten entwicklungsvorgänge der eizelle, die
zelltheilung und die conjugation der infusorien. Abh Senckenb Naturforsch
1991b). They have not been found in ciliates from aerobic
Ges 10:213–464
habitats. Bütschli O (1889) Protozoa. Abt. 111: infusoria und system der radiolaria. In:
Bronn HG (ed) Klassen und ordnungen des thierreichs, vol 1. C.F. Winter,
Leipzig, pp 1098–2035
Chen TT (1955) Paramecin 34, a killer substance produced by Paramecium
Electronic Supplementary Material bursaria. Proc Soc Exp Biol Med 88:541–543
Corliss JO (1985) Concept, definition, prevalence, and host-interactions of
Video sequence showing killer trait. For the video, see the online xenosomes (cytoplasmic and nuclear endosymbionts). J Protozool
32:373–376
version of The Prokaryotes.
Demar-Gervais C, Genermont J (1976) Recherches génétiques sur Euplotes
Video sequence showing infection of Paramecium by vannus: les charactères ‘‘monster’’ et ‘‘killer’’. J Protozool 23 (Suppl) 8A
Holospora. For the video, see the online version of The Dilts JA (1976) Covalently closed, circular DNA in kappa endosymbionts of
Prokaryotes. Paramecium. Genet Res 627:161–170
Video sequence showing release of Holospora. For the video, Dilts JA (1986) The importance of the refractile body in expression of the killer
trait in Paramecium. In: Wickner RB, Hinnebusch A, Lambowitz AM,
see the online version of The Prokaryotes.
Gunsalus IC, Hollaender A (eds) Extrachromosomal elements in lower
Video sequence showing the development of Holospora. For eukaryotes. Plenum, New York
the video, see the online version of The Prokaryotes Dilts JA, Quackenbush RL (1986) A mutation in the R body-coding
sequence destroys expression of the killer trait in P tetraurelia. Science
232:641–643
Dini F, Luporini P (1976) The mate-killer trait in a stock of Euplotes crassus
References (Dujardin) (Ciliata, Hypotrichida). Monitore Zool Ital (NS) 10:15–24
Dini F, Luporini P (1982) The inheritance of the mate-killer trait in Euplotes
Abamo F, Dohra H, Fujishima M (2008) Fate of the 83-kDa periplamic crassus (Hypotrichida, Ciliophora). Protistologica 18:179–184
protein of the infectious form of the endonuclear symbiotic bacterium Doddema HJ, Vogels GD (1978) Improved identification of methanogenic
Holospora obtusa during the infection process. FEMS Microbiol Lett bacteria by fluorescence microscopy. Appl Environ Microbiol 36:752–754
280:21–27 Dohra H, Fujishima M (1999a) Cell structure of the infectious form of Holospora,
Allen SL, Nerad TA (1978) Method for the simultaneous establishment of many an endonuclear symbiotic bacterium of the ciliate Paramecium. Zool Sci
axenic cultures of Paramecium. J Protozool 25:134–139 16:93–98
Dohra H, Fujishima M (1999b) Effects of antibiotics on the early infection Fok A, Allen RD (1979) Axenic Paramecium caudatum. 1: mass culture and
process of the macronuclear endosymbiotic bacterium Holospora obtusa of structure. J Protozool 6:463–470
Paramecium caudatum. FEMS Microbiol Lett 179:473–477 Fok A, Allen RD (1988) The lysosome system. In: Görtz HD (ed) Paramecium.
Dohra H, Fujishima M, Hoshide K (1994) Monoclonal antibodies specific for Springer, Berlin, pp 301–324
periplasmic materials of the macronuclear specific bacterium Holospora Fokin SI (1988) A bacterial symbiont of the macronucleus perinuclear space in
obtusa of the ciliate Paramecium caudatum. Eur J Protistol 30:288–294 the ciliate Paramecium duboscqui. Cytologia 30:632–635
Dohra H, Yamamoto K, Fujishima M, Ishikawa H (1997) Cloning and sequencing Fokin SI (1991) Holospora recta sp. nov., a micronucleus-specific endobiont of the
of gene coding for a periplasmic 5.4 kDa peptide of the macronucleus- ciliate Paramecium caudatum (Russian with English summary). Cytologia
specific symbiont Holospora obtusa of the ciliate Paramecium caudatum. 33:135–141
Zool Sci 14:69–75 Fokin SI, Chivilev SM (1999) Brackish water Paramecium species and Parame-
Dohra H, Fujishima M, Ishikawa H (1998) Structure and expression of a GroE- cium polycaryum: morphometric analysis and some biological peculiarities.
homologous operon of a macronucleus-specific symbiont Holospora obtusa Acta Protozool 38:105–117
of the ciliate Paramecium caudatum. J Eukaryot Microbiol 4:71–79 Fokin SI, Görtz HD (1993) Caedibacter macronucleorum sp. nov., a bacterium
Dorner RW (1957) Stability of paramecin 34 at different temperatures and pH inhabiting the macronucleus of Paramecium duboscqui. Arch Protistenkd
values. Science 126:1243–1244 143:319–324
Ehrsam E, Görtz H-D (1999) Surface proteins of the gram-negative bacterium Fokin SI, Görtz HD (2009) Holospora bacteria in Paramecium and their
Holospora obtusa bind to macronuclear proteins of its host Paramecium characterization. In: Fujishima M (ed) Endosymbionts in Paramecium.
caudatum. Eur J Protistol 35:304–308 Microbiology Monpographs, vol 12. Springer, Berlin, Heidelberg,
Embley TM, Finlay BJ (1994) The use of small subunit rRNA sequences to unravel pp161–199
the relationships between anaerobic ciliates and their methanogen endosym- Fokin S, Karpov S (1995) Bacterial endocytobionts inhabiting the perinuclear
bionts. Microbiology 140:225–235 space of Protista. Endocyt Cell Res 11:81–94
Eschbach E, Pfannkuchen M, Schweikert M, Drutschmann D, Brümmer F, Fokin Fokin SI, Ossipov DV (1986) Pseudocaedibacter glomeratus n. sp. – A cytoplasmic
S, Ludwig W, Görtz HD (2009) ‘‘Candidatus Paraholospora nucleivisitans’’, an symbiont of the ciliate Paramecium pentaurelia. Cytologia 28:1000–1004
intracellular bacterium in Paramecium sexaurelia shuttles between the cyto- Fokin SI, Sabaneyeva EV (1993) Bacterial endocytobionts of the ciliate Parame-
plasm and the nucleus of its host. Sys Appl Microbiol 32:490–500 cium calkinsi. Eur J Protistol 29:390–395
Estève J-C (1957) Une population de type ‘‘killer’’ chez Paramecium caudatum Fokin SI, Sabaneyeva E (1997) Release of endonucleobiotic bacteria Holospora
(Ehrenberg). Protistologica 14:201–207 bacillata and Holospora curvata from the macronucleus of their host
Euzeby JP (1997) Revised nomenclature of specific or subspecific epithets that do cells Paramecium woodruffi and Paramecium calkinsi. Endocyt Cell Res
not agree in gender with generic names that end in-bacter. Int J Syst Bacteriol 12:49–55
47(2):585–592 Fokin SI, Skovorodkin IN (1991) Holospora undulata – endonucleobiont of the
Fauré-Fremiet E (1909) Sur un cas de symbiose par un infusoir cilié. CR Soc Biol ciliate Paramecium caudatum in search for micronucleus. Cytology 33:64–75
67:113–114 Fokin SI, Skovorodkin IN (1997) Experimental analysis of the resistance of
Fauré-Fremiet E (1950a) Caulobactéries épizoiques associées aux Centrophorella Paramecium caudatum (Ciliophora) against infection by bacterium
(Cilies holotriches). Bull Soc Zool France 75:134–137 Holospora undulata. Eur J Protisatol 33:214–218
Fauré-Fremiet E (1950b) Ecologie des ciliés psammophiles litteraux. Bull Biol Fokin SI, Ossipov DV, Skoblo II, Rautian MS, Sabaneyeva EV (1987a) Nonospora
France-Belgique 84:35–75 macronucleata g. n., sp. n–a vegetative nucleus symbiont of the ciliate
Fauré-Fremiet E (1951) The marine sand dwelling ciliates of cape cod. Biol Bull Paramecium caudatum. Cytologia 29:963–970
100:59–70 Fokin SI, Boss AO-L, Ossipov DV (1987b) Virus-containing cytoplasmic symbi-
Fauré-Fremiet E (1952) Symbionts bacteriens des ciliés du genre Euplotes. CR ont of the ciliate Paramecium woodruffi. Cytologia 29:1303–1306
Acad Sci 235:402–403 Fokin SI, Brigge T, Brenner J, Görtz H-D (1996) Holospora species infecting the
Favinger J, Stadtwald R, Gest H (1989) Rhodospirillum centenum, sp. nov., nuclei of Paramecium appear to belong into two groups of bacteria. Eur
a thermotolerant cyst-forming anoxygenic photosynthetic bacterium. Ant J Protistol 32(suppl 1):19–24
V Leeuwenhoek 55:291–296 Fokin SI, Schweikert M, Brümmer F, Gőrtz H-D (2005a) Spirostomum
Fenchel T, Finlay BJ (1990) Anaerobic free-living protozoa: growth efficiencies spp. (Ciliophora, Protista) – a new suitable system for endocytobiosis
and the structure of communities. FEMS Microbiol Ecol 74:269–276 research. Protoplasma 255:93–102
Fenchel T, Finlay BJ (1991a) The biology of free-living anaerobic ciliates. Eur Fokin SI, Schweikert M, Fujishima M (2005b) Recovery of the ciliate Paramecium
J Protistol 26:200–216 multimicronucleatum following bacterial infection with Holospora obtusa.
Fenchel T, Finlay BJ (1991b) Synchronous division of an endosymbiotic Eur J Protistol 41:129–138
methanogenic bacterium in the anaerobic ciliate Plagiopyla frontata Kahl. Frankel J (1973) Dimensions of control of cortical patterns in Euplotes: the role of
J Protozool 38:22–28 preexisting structure, the clonal life cycle, and the genotype. J Exp Zool
Fenchel T, Perry T, Thane A (1977) Anaerobiosis and symbiosis with bacteria in 183:71–94
free-living ciliates. J Protozool 24:154–163 Freiburg M (1985) Isolation and characterization of macronuclei of Paramecium
Figueroa-de Soto VE, Soldo AT (1977) Characterization and base sequence caudatum infected with the macronucleus-specific bacterium Holospora
homologies of the DNA of xenosomes and related particles. Int Congr obtusa. J Cell Sci 73:389–398
Protozool 5:339 Fujishima M (1993) Control of morphological changes of the endonuclear
Finlay BL, Fenchel T (1989) Hydrogenosomes in some anaerobic protozoa symbiont Holospora of the ciliate Paramecium. In: Sato S, Ishida M, Ishikawa
resembling mitochondria. FEMS Microbiol Lett 65:187–190 H (eds) Endocytobiology, vol V. Tübingen University Press, Tübingen,
Finlay BL, Fenchel T (1991) An anaerobic protozoon, with symbiotic methanogens, pp 505–508
living in municipal landfill material. FEMS Microbiol Lett 65:169–180 Fujishima M (2009) Infection and maintenance of Holospora species in Parame-
Finlay BL, Fenchel T (1992) An anaerobic ciliate as a natural chemostat for the cium caudatum. In: Endosymbionts in Paramecium. Microbiology
growth of endosymbiotic methanogens. Eur J Protistol 28:127–137 Monpographs Vol 12, Fujishima, M. (Ed.). Springer Berlin, Heidelberg,
Finlay BJ, Embley TM, Fenchel T (1993) A new polymorphic methanogen, closely pp. 201–225
related to Methanocorpusculum parvum, living in stable symbiosis within the Fujishima J, Fujita M (1985) Infection and maintenance of Holospora obtusa,
anaerobic ciliate Trimyema sp. J Gen Microbiol 139:371–378 a macronuclear-specific bacterium of the ciliate Paramecium caudatum.
Foissner W (1978) Euplotes moebiusi f. quadricirratus (Ciliata, Hypotrichida) II. J Cell Sci 76:179–187
The fine structure of some cytoplasmic organelles. Naturk Jahrb Stadt Linz Fujishima M, Heckmann K (1984a) Intra-and interspecific transfer of endosym-
23:17–24 bionts in Euplotes. J Exp Zool 230:339–345
Fujishima M, Nagahara K (1984b) Isolation of endonuclear symbiont Holospora Görtz H-D, Lellig S, Miosga O, Wiemann M (1990) Changes in fine structure
obtusa from mass cultures of Paramecium caudatum. In: Proceedings of the and polypeptide pattern during development of Holospora obtusa, a bacte-
annual meeting of the zoological society of Japan, vol 55. Toryokosku Press, rium infecting the macronucleus of Paramecium caudatum. J Bacteriol
Morioka, Japan, pp 36–40 172:5664–5669
Fujishima J, Nagahara K, Kojima K (1990a) Changes in morphology, buoyant Görtz H-D, Benting J, Ansorge I, Freiburg M (1992) Cell surface proteins of the
density and protein composition in differentiation from the reproductive infectious forms of the symbiotic bacterium Holospora obtusa. Symbiosis
short form to the infectious long form of Holospora obtusa, a macronucleus- 14:391–397
specific symbiont of the ciliate Paramecium caudatum. Zool Sci 7:849–860 Grimes GW, Preer JR Jr (1971) Further observations on the correlation between
Fujishima M, Sawabe H, Iwatsuki K (1990b) Scanning electron microscopic kappa and phage-like particles in Paramecium. Genet Res 18:115–116
observation of differentiation from the reproductive short form to the Gromov BV, Ossipov DV (1981) Holospora (ex Hafkine 1890) nom. rev., a genus of
infectious long form of Holospora obtusa. J Protozool 37:123–128 bacteria inhabiting the nuclei of paramecia. Int J Syst Bacteriol 31:348–352
Fujishima M, Dohra H, Kawai M (1997) Quantitative changes in periplasmic Hackstein JHP, Stumm CK (1994) Methane production in terrestrial arthropods.
proteins of the macronucleus-specific bacterium Holospora obtusa in the Proc Natl Acad Sci 91:5441–5445
infection process of the ciliate Paramecium caudatum. J Euk Microbiol Hafkine M-W (1890) Maladies infectieuses des paramécies. Ann Inst Pasteur
44:636–642 (Paris) 4:148–162
Fujishima M, Kawai M, Yamamoto M (2005) Paramecium caudatum acquires Hahn MW (2003) Isolation of strains belonging to the cosmopolitan
heat-shock resistance in ciliary movement by infection with the endonuclear Polynucleobacter necessarius cluster from freshwater habitats located in
symbiotic bacterium Holospora. Zool Sci 23:1161 three climatic zones. Appl Environ Microbiol 69:5248–5254
Gates MA, Curds CR (1979) The dargyroome of the genus Euplotes. Bull Br Mus Hahn MW, Lang E, Brandt U, Wu QL, Scheuerl T (2009) Emended description
Nat Hist (Zool) 35:127–200 of the genus Polynucleobacter and the species P necessarius and proposal
Gellissen G, Michaelis G (1987) Gene transfer: mitochondria to nucleus. In: Lee of two subspecies P. necessarius subspecies necessarius subsp. nov. and
JJ, Fredrick R (eds) Endocytobiology, vol III. Academy of Science, New York, P. necessarius subsp. A symbioticus subsp nov. Int J Syst Evol Microbiol
pp 391–401 59(8):2002–2009
Gibson I (1974) The endosymbionts of Paramecium. Crit Rev Microbiol 3:243–273 Hausmann K, Bradbury PC (eds) (1996) Ciliates—cells as organisms. Gustav
Gibson I (1984) A comparison of the refractile bodies (R-bodies) of certain Fischer-Verlag, Stuttgart
bacteria. III. Nucleotide sequence homologies and R-body function. Micron Heatherington S (1934) The sterilization of protozoa. Biol Bull 67:315–321
Microsc Acta 15:253–273 Heckmann K (1963) Paarungstypsystem und genabhängige Paarungstypdiffer-
Gibson I, Beale GH (1961) Genic basis of the mate-killer trait in Paramecium enzierung bei dem hypotrichen Ciliaten Euplotes vannus. O F Müller Arch
aurelia stock 540. Genet Res 2:82–91 Protistenk 106:393–421
Goosen NK, Hormenans AMC, Hillebrand SJW, Stumm CK, Vogels GD (1988) Heckmann K (1975) Omikron, ein essentieller Endosymbiont von Euplotes
Cultivation of the sapropelic ciliate Plagiopyla nasuta Stein and isolation aediculatus. J Protozool 22:97–104
of the endosymbiont Methanobacterium formicicum. Arch Microbiol Heckmann K (1983) Endosymbionts of Euplotes. In: Jeon KW (ed) Intracellular
150:165–170 symbiosis International review of cytology. Academic, New York,
Görtz H-D (1981) Ein neues symbiontisches Bakterium in Paramecium pp 111–144
sexaurelia. Verh Dtsch Zool Ges 74:227 Heckmann K, Frankel J (1968) Genic control of cortical pattern in Euplotes. J Exp
Görtz H-D (1983) Endonuclear symbionts in ciliates. In: Jeon KW (ed) Intracel- Zool 168:11–38
lular symbiosis. International review of cytology, supplement 14. Academic, Heckmann K, Görtz HD (1991) Prokaryotic symbionts of ciliates. In: Balows A,
New York, pp 145–176 Trüper HG, Dworkin M, Harder W, Schleifer KD (eds) The prokaryotes,
Görtz H-D (1987) Different endocytobionts simultaneously colonizing ciliate 2nd edn. Springer, New York, pp 3865–3890
cells. Ann NY Acad Sci 503:261–268 Heckmann K, Schmidt HJ (1987) Polynucleobacter necessarius gen. nov., spe nov.,
Görtz H-D (1992) Bactéries endosymbiotiques de Paramecium. Ann Biolog an obligately endosymbiotic bacterium living in the cytoplasm of Euplotes
73:41–49 aediculatus. Int J Syst Bacteriol 37:456–457
Görtz H-D (2001) Intracellular bacteria in ciliates. Int Microbiol 4(3):143–150 Heckmann K, Preer JR Jr, Straetling WH (1967) Cytoplasmic particles in the
Görtz H-D, Dieckmann J (1980) Life cycle and infectivity of Holospora elegans killers of Euplotes minuta and their relationship to the killer substance.
Hafkine, a micronucleus-specific symbiont of Paramecium caudatum J Protozool 1:360–363
(Ehrenberg). Protistologica 16:591–603 Heckmann K, ten Hagen R, Görtz H-D (1983) Fresh-water Euplotes species with a
Görtz H-D, Fokin SI (2009) Diversity of endosymbiotic bacteria in Paramecium. 9 type 1 cirrus pattern depend upon endosymbionts. J Protozool 30:284–289
In: Fujishima M (ed) Endosymbionts in Paramecium. Microbiology Heckmann K, Schmidt HJ, Fujishima M (1986) Omikron and omikron-like
Monpographs, vol 12. Springer, Berlin/Heidelberg, pp 131–160 endosymbionts of Euplotes. In: Wickner RB, Hinnebusch A, Lambowitz
Görtz H-D, Freiburg M (1984) Bacterial symbionts in the micronucleus of AM, Gunsalus IC, Hollaender A (eds) Extrachromosomal elements of
Paramecium bursaria. Endocyt Cell Res 1:37–46 lower eukaryotes. Plenum, New York, pp 327–335
Görtz H-D, Fujishima M (1983) Conjugation and meiosis of Paramecium Henry SM (1966) Symbiosis: vol I. Associations of microorganisms, plants, and
caudatum infected with the micronucleus-specific bacterium Holospora marine organisms. Academic, New York/London
elegans. Eur J Cell Biol 32:86–91 Hernández-Romero D, Lucas-Elı́o P, López-Serrano D, Solano F, Sanchez-Amat A
Görtz H-D, Maier G (1991) A bacterial infection in a ciliate from sewage sludge. (2003) Marinomonas mediterranea is a lysogenic bacterium that synthesizes
Endocyt Cell Res 8:45–52 R-bodies. Microbiology 149:2679–2686
Görtz H-D, Wiemann M (1987) Colonization of the ciliate Stentor multiformis by Heruth DP, Pond FR, Dilts JA, Quackenbush RL (1994) Characterization of
three different endocytobionts. Endocyt Cell Res 4:177–184 genetic determinants for R body synthesis and assembly in Caedibacter
Görtz H-D, Wiemann M (1989) Route of infection of the bacteria Holospora taeniospiralis 47 and 116. J Bacteriol 176:3559–3567
elegans and Holospora obtusa into the nuclei of Paramecium caudatum. Eur Holtzman HE (1959) A kappa-like particle in a non-killer stock of Paramecium
J Protistol 24:101–1090 aurelia, syngen 5. J Protozool 6(Suppl):26
Görtz H-D, Freiburg M, Wiemann M (1988) Polypeptide differences between Hori M, Fujishima M (2003) The endosymbionts bacterium Holospora obtusa
infectious and reproductive forms of Holospora obtusa, an endonucleobiotic enhances heat-shock gene expression of the host Paramecium caudatum.
bacterium in Paramecium caudatum. Endocyt Cell Res 5:233–244 J Eukaryot Microbiol 50:293–298
Görtz H-D, Ahlers N, Robenek H (1989) Ultrastructure of the infectious and Hori M, Fujii K, Fujishima M (2008) Micronucleus-specific bacterium Holospora
reproductive forms of Holospora obtusa, a bacterium infecting the macronu- elegans irreversibly enhances stress gene expression of the host Paramecium
cleus of Paramecium caudatum. J Gen Microbiol 135:3079–3085 caudatum. J Eukaryot Microbiol 55:515–521
Iwatani K, Dohra H, Lang FB, Burger G, Hori M, Fujishima M (2005) Translo- Narayanan N, Krishnakumar B, Anupama VN, Manilal VB (2009) Methanosaeta
cation of an 89-kDa periplasmic protein is associated with Holospora infec- sp., the major endosymbionts of Metopus es. Res Microbiol 160:600–607
tion. Biochem Biophys Res Comm 337:1198–2015 Nobili R, Rosati G, Verni F (1976) The killer-trait in Euplotes crassus. Boll Zool
Jenkins RA (1970) The fine structure of a nuclear envelope associated endosym- 43:251–258
biont of Paramecium. J Gen Microbiol 61:355–359 Ossipov DV (1973) Specific infectious specificity of the omega-particles, micronuclear
Jeon KW, Jeon MS (1976) Scanning electron microscope observations of Amoeba symbiotic bacteria of Paramecium caudatum. Cytologia 15:211–217
proteus during phagocytosis. J Protozool 23:83–86 Ossipov DV (1981) Problems of nuclear heteromorphism in the unicellular
Kahl A (1933) Ciliata libera et ectocommensalia. In: Grimpe G, Wagler E (eds) organisms. Nauka, Leningrad
Die Tierwelt der Nord-und Ostsee, Teil II Akademische Verlagsgesellschaft Ossipov DV, Podlipaev SA (1977) Electron microscope examinations of early
Becker & Erler, Leipzig, Germany, pp 29–146 stages of infection of Paramecium caudatum by bacterial symbionts of the
Kahl A (1935) Urtiere oder protozoa I: Wimpertiere oder ciliata (infusoria). 4: macronucleus (Iota-bacteria). Acta Protozool 16:289–308
peritricha und chonotricha nachtrag. In: Dahl F (ed) Die tierwelt Ossipov DV, Skoblo II, Rautian MS (1975) Iota-particles, macronuclear symbiotic
Deutschlands. Gustav Fischer, Jena, pp 806–842 bacteria of the ciliate Paramecium caudatum clone M 115. Acta Protozool
Kanabrocki JA, Lalucat J, Cox BJ, Quackenbush RL (1986) Comparative study of 14:263–280
refractile (R) bodies and their genetic determinants: relationship of type 51 Ossipov DV, Skoblo II, Borchsenius ON, Rautian MS, Podlipaev SA (1980)
R bodies to R bodies produced by Pseudomonas taeniospiralis. J Bacteriol Holospora acuminata – a new species of symbiotic bacterium from
168:1019–1022 the micronucleus of the cilı́ate Paramecium bursaria. Focke Cytolog
Kawai M, Fujishima M (2000) Invasion of the macronucleus of Paramecium 22:922–929
caudatum by the bacterium Holospora obtusa: fates of the bacteria and timing Ossipov DV, Skoblo II, Borchsenius ON, Lebedeva NA (1993) Interactions
of invasion steps. Eur J Protistol 36:46–52 between Paramecium bursaria (Protozoa, Ciliophora, Hymenostomatida)
Kirby H Jr (1941) Organisms living on and in protozoa. In: Calkins GN, Summers and their nuclear symbionts. I: phenomenon of symbiogenic lysis of the
FM (eds) Protozoa in biological research. Columbia University Press, New bacterium Holospora acuminata. Eur J Protistol 29:61–71
York, pp 1009–1113 Paulin JJ (1996) Morphology and cytology of ciliate. In: Hausmann K, Bradbury
Kuhlmann H-W, Heckmann K (1989) Adolescence in Euplotes octocarinatus. PC (eds) Ciliates: cells as organisms. Gustav Fischer-Verlag, Stuttgart, pp 1–40
J Exp Zool 251:316–328 Petroni G, Spring S, Schleifer KH, Verni F, Rosati G (2000) Defensive extrusive
Kusch J, Stremmel M, Schweikert M, Adams V, Schmidt HJ (2000) The toxic ectosymbionts of Euplotidium (Ciliophora) that contain microtubule-like
symbiont Caedibacter caryophila in the cytoplasm of Paramecium novaurelia. structures are bacteria related to Verrucomicrobia. Proc Natl Acad Sci USA
Microb Ecol 40:330–335 97:1813–1817
Kusch J, Czubatinski L, Wegmann S, Hübner M, Alter M, Albrecht P (2001) Petroni G, Dini F, Verni F, Rosati G (2001) A molecular approach to the tangled
Competitive advantages of Caedibacter-infected paramecia. Protist intragenic relationships underlying phylogeny in Euplotes (Ciliophora,
153:47–58 Spirotricha). Molec Phylogenet Evol 22(1):118–130
Lalucat J, Meyer O, Mayer F, Pares R, Schlegel HG (1979) R bodies in Pond FR, Gibson I, Lalucat J, Quackenbush RL (1989) R-body producing bacte-
newly isolated free-living hydrogen-oxidizing bacteria. Arch Microbiol ria. Microbiol Rev 53:25–67
121:9–15 Preer JR Jr (1948a) A study of some properties of the cytoplasmic factor ‘‘kappa’’
Lalucat J, Wells B, Gibson I (1986) Relationships between R bodies of certain in Paramecium aurelia, variety 2. Genetics 33:349–404
bacteria. Micron Microsc Acta 17:243–245 Preer JR Jr (1948b) The killer cytoplasmic factor kappa: Its rate of reproduction, the
Landis WG (1981) The ecology, role of the killer trait, and interactions of five number of particles per cell, and its size. Am Natural 82:35–42
species of the Paramecium aurelia complex inhabiting the littoral zone. Can Preer JR Jr (1950) Microscopically visible bodies in the cytoplasm of the ‘‘killer’’
J Zool 59:1734–1743 strains of Paramecium aurelia. Genetics 35:344–362
Landis WG (1987) Factors determining the frequency of the killer trait within Preer LB (1969) Alpha, an infectious macronuclear symbiont of Paramecium
populations of the Paramecium aurelia complex. Genetics 115:197–206 aurelia. J Protozool 16:570–578
Lauterborn R (1901) Die ‘‘sapropelische’’ Lebewelt. Zool Anz 24:50–55 Preer JR Jr (1977) The killer system in Paramecium-kappa and its viruses.
Laval M (1970) Présence de bactéries intranucléaires chez Zoothamnium Microbiology 1977:576–578
pelagicum. In: 7th Internernational Congress on Electron Microscoscopy, Preer LB (1981) Prokaryotic symbionts of Paramecium. In: Starr MP, Stolp H,
Grenoble, pp 403–404 Trüper HG, Balows A, Schlegel HG (eds) The prokaryotes. Springer, New
Liebmann H (1937) Bakteriensymbiose bei Faulschlammziliaten. Biol Zbl York, pp 2127–2136
57:442–445 Preer JR Jr, Jurand A (1968) The relationship between virus-like particles and
Liebmann H (1938) Biologie und Chemismus der Bleilochsperre. Arch Hydrobiol R bodies of Paramecium aurelia. Genet Res 12:331–340
33:1–81 Preer LB, Preer JR Jr (1964) Killing activity from lysed particles of Paramecium.
Luporini P (1974) Killer and autogamous strains of Euplotes crassus Dujardin Genet Res 5:230–239
from the Somalien coast. Monit Zool Ital (NS) 5(Suppl):129–132 Preer JR Jr, Preer LB (1967) Virus-like bodies in killer-paramecia. Proc Natl Acad
Margulis L, Fester R (1991) Symbiosis as the source of evolutionary innovation. Sci USA 58:1774–1781
MIT Press, Cambridge, MA Preer JR Jr, Preer LB (1982) Revival of names of protozoan endosymbionts
Meenaghan H, Reilly K, Wells B, Gibson I (1984) A comparison of the refractile and proposal of Holospora caryophila nom. nov. Int J Syst Bacteriol
bodies (R-bodies) of certain bacteria. II: effects of pH on the structure. 32:140–141
Micron Microsci Acta 15:247–252 Preer JR Jr, Preer LB (1984) Endosymbionts of protozoa. In: Krieg NR (ed)
Mueller JA (1963) Separation of kappa particles with infective activity from those Bergey’s manual of systematic bacteriology, vol 1. Williams and Wilkins,
with killing activity and identification of the infective particles in Parame- Baltimore, pp 795–811
cium aurelia. Exp Cell Res 30:492–508 Preer JR Jr, Stark P (1953) Cytological observations on the cytoplasmic factor
Müller J (1856) Beobachtungen an Infusorien. Monatsber Königl Preub Akad ‘‘kappa’’ in Paramecium aurelia. Exp Cell Res 5:478–491
Wissensch, Berlin, Juli, pp 389–393 Preer JR Jr, Hufnagel LA, Preer LB (1966) Structure and behavior of ‘‘R’’ bodies
Müller M (1988) Energy metabolism of protozoa without mitochondria. Ann Rev from killer paramecia. J Ultrastr Res 15:131–143
Microbiol 42:465–488 Preer JR Jr, Preer LB, Rudman B, Jurand A (1971) Isolation and composition of
Nakamura Y, Aki M, Aikawa T, Hori M, Fujishima M (2004) Differences in gene bacteriophage-like particles from kappa of killer paramecia. Mole Gen Genet
expression of the ciliate Paramecium caudatum caused by endonuclear sym- 111:202–208
biosis with Holospora obtusa, revealed using differential display reverse Preer LB, Jurand A, Preer JR Jr, Rudman B (1972) The classes of kappa in
transcribed PCR. FEMS Microbiol Lett 240:209–213 Paramecium aurelia. J Cell Sci 11:581–600
Preer JR Jr, Preer LB, Jurand A (1974a) Kappa and other endosymbionts of Schmidt HJ, Freiburg M, Görtz HD (1987a) Comparison of the infectious forms
Paramecium aurelia. Bacteriol Rev 38:113–163 of two bacterial endonucleobionts, Holospora elegans and H. obtusa, from the
Preer LB, Rudman B, Preer JR Jr (1974b) Induction of R bodies by ultraviolet light ciliate Paramecium caudatum. Microbios 49:189–197
in killer paramecia. J Gen Microbiol 80:209–215 Schmidt HJ, Görtz HD, Quackenbush RL (1987b) Caedibacter caryophila sp. nov.,
Przybosz E, Fokin SI (1997) Species of the Paramecium aurelia complex a killer symbiont inhabiting the macronucleus of Paramecium caudatum. Int
Sonneborn in Germany. Arch Protistenkd 148:167–172 J Syst Bacteriol 37:459–462
Quackenbush RL (1978) Genetic relationships among bacterial endosymbionts of Schmidt HJ, Pond FR, Görtz HD (1987c) Refractile bodies (R bodies)
Paramecium aurelia: polynucleotide sequence relationship among members from the macronuclear killer particle Caedibacter caryophila. J Cell Sci
of Caedobacter. J Gen Microbiol 108:181–187 88:177–184
Quackenbush RL (1982) Validation of the publication of new names and new Schmidt HJ, Görtz HD, Pond FR, Quackenbush RL (1988) Characterization of
combinations previously published outside the IJSB. List No. 8. Int J Syst Caedibacter endonucleobionts from the macronucleus of Paramecium
Bacteriol 32:266–268 caudatum and the identification of a mutant with blocked R-body synthesis.
Quackenbush RL (1983) Plasmids from bacterial endosymbionts of hump-killer Exp Cell Res 174:49–57
paramecia. Plasmid 9:298–306 Schneller MV (1958) A new type of killing action in a stock of Paramecium aurelia
Quackenbush RL (1988) Endosymbionts of killer paramecia. In: Görtz HD (ed) from Panama. Proc Ind Acad Sci 67:302
Paramecium. Springer, Berlin, pp 406–418 Schneller MV, Sonneborn TM, Mueller JA (1959) The genetic control of kappa-
Quackenbush RL, Burbach JA (1983) Cloning and expression of DNA sequences like particles in Paramecium aurelia. Genetics 44:533–534
associated with the killer trait of Paramecium tetraurelia stock 47. Proc Natl Schönefeld U, Alfermann AW, Schultz JE (1986) Economic mass cultivation of
Acad Sci USA 80:250–254 Paramecium tetraurelia on a 200-liter scale. J Protozool 33:222–225
Quackenbush RL, Cox BJ, Kanabrocki JA (1986) Extrachromosomal elements of Schrallhammer M, Schweikert M (2009) The killer effect of Paramecium and its
extrachromosomal elements of Paramecium and their extrachromosomal causative agents. In: Fujishima M (ed) Endosymbionts in Paramecium.
elements. In: Wickner RB, Hinnebusch A, Lambowitz AM, Gunsalus IC, Microbiology Monpographs, vol 12. Springer, Dortrecht/Heidelberg/
Hollaender A (eds) Extrachromosomal elements in lower eukaryotes. London/New York, pp 228–246
Plenum, New York, pp 265–278 Schrallhammer M, Schweikert M, Vallesi A, Verni F, Petroni G (2010) Detection
Raikov IB (1971) Bactéries épizoiques et mode de nutrition du cilié psammophile of a novel subspecies of Francisella noatunensis as endosymbiont of Euplotes
Kentrophorus fistulosum Fauré-Fremiet (étude au microscope electronique). raikovi. Microb Ecol 61:455–464
Protistologica 7:365–378 Siegel RW (1953) A genetic analysis of the mate-killer trait in Paramecium aurelia,
Raikov IB (1974) Etude ultrastructurale des bactéries épizoiques et éndozoiques variety 8. Genetics 38:550–560
de Kentrophorus latum Raikov, cilié holotriche mésopsammique. Cahiers de Siegel RW, Heckmann K (1966) Inheritance of autogamy and the killer trait in
Biologie Marine 15:379–393 Euplotes minuta. J Protozool 13:34–48
Read CP (1970) Parasitism and symbiology. Ronald Press, New York Simon EM, Schneller MV (1973) The preservation of ciliated protozoa at low
Roller C, Ludwig W, Schleifer K-H (1992) Gram-positive bacteria with a high temperature. Cryobiology 10:421–426
DNA G-C content are characterized by a common insertion with their 23S Sitte P (1993) Symbiogenetic evolution of complex cells and complex plastids.
rRNA genes. J Gen Microbiol 138:1167–1175 Eur J Protistol 29:131–143
Rosati G (1999) Epixenosomes: symbionts of the hypotrich ciliate Euplotidium Skoblo II, Lebedeva NA (1986) Infection of the nuclear apparatus of Paramecium
itoi. Symbiosis 26:1–24 bursaria (Ciliata) by the symbiotic bacterium Holospora curviuscula.
Rosati G, Verni F (1975) Macronuclear symbionts in Euplotes crassus (Ciliata, Cytologia 28:367–372
Hypotrichida). Bull Zool 42:231–232 Skoblo II, Borchsenius ON, Lebedeva NA, Ossipov DV (1985) A new species of
Rosati G, Verni F (1977) Bacteria-like endosymbionts in Euplotes crassus. In: symbiotic bacteria of Paramecium bursaria, Ciliophora, Protozoa. Cytologia
Hutner SH (ed) 5th Iternational Congress of Protozoology, Abstract Society 27:1292–1297
of Protozoology, New York, 443 pp Skoblo II, Borchsenius ON, Lebedeva NA, Ossipov DV (1990) Symbiogenic lysis
Rosati G, Verni F, Luporini P (1976) Cytoplasmic bacteria-like endosymbionts in of bacteria Holospora acuminata in the generative nucleus of the ciliate
Euplotes crassus (Dujardin) (Ciliata, Hypotrichida). Monitore Zool Ital (NS) Paramecium bursaria. Cytologia 32:515–519
10:449–460 Smith IE (1961) Purification of kappa particles of Paramecium aurelia, stock 51.
Rosati G, Lenzi P, Verni F (1993a) ‘‘Epixenosomes’’: peculiar epibionts of the Am Zool 1:390
protozoon ciliate Euplotidium itoi. Do their cytoplasmic tubules consist of Smith-Sonneborn JE, Van Wagtendonk WJ (1964) Purification and chemical
tubulin? Micron 24:465–471 characterization of kappa of stock 51, Paramecium aurelia. Exp Cell Res
Rosati G, Verni F, Lenzi P (1993b) ‘‘Epixenosomes’’: peculiar epibionts of the 33:50–59
ciliate Euplotidium itoi. The formation of the extrusive apparatus and the Smurov AO, Fokin SI (1999) Resistence of Paramecium species (Ciliophor,
ejecting mechanism. Eur J Protistol 29:238–245 Peniculia) to salinity of environment. Protistology 1:43–53
Rosati G, Petroni G, Quochi S, Modeo L, Verni F (1999) Epixenosomes: peculiar Soldo AT (1963) Axenic culture of Paramecium: some observations on the growth
epibionts of the hyprotrich ciliate Euplotidium itoi defend their host against behavior and nutritional requirements of a particle-bearing strain of Para-
predators. J Eukaryot Microbiol 46:278–282 mecium aurelia 299 lambda. Ann NY Acad Sci 108:380–388
Ruthmann A, Heckmann K (1961) Formwechsel und struktur des makronukleus Soldo AT (1974) Intracellular particles in Paramecium. In: Van Wagtendonk WJ
von Bursaria truncatella. Arch Protistenkd 105:313–340 (ed) Paramecium: a current survey. Elsevier, Amsterdam, pp 377–442
Sabaneyeva EV, Derkacheva ME, Benken KA, Fokin SI, Vainio S, Skovorodkin IN Soldo AT (1983) The biology of the xenosome, an intracellular symbiont. In: Jeon
(2009) Actin-based mechanism of holospora obtusa trafficking in KW (ed) Intracellular symbiosis. Int Rev Cytol, Suppl 14. Academic Press,
Paramecium caudatum. Protist 160:205–219 New York, pp 79–109
Sauerbrey E (1928) Beobachtungen über einige neue oder wenig bekannte marine Soldo AT (1987) Parauronema acutum and its xenosomes, a model system.
Ciliaten. Arch Protistenkd 61:355–407 J Protozool 34:447–451
Schmidt HJ (1982) Isolation of omikron-endosymbionts from mass cultures of Soldo AT, Brickson SA (1978) Observation on the ultrastructure, mode of infec-
Euplotes aediculatus and characterization of their DNA. Exp Cell Res tivity and host range of xenosomes. Tissue Cell 10:609–618
140:417–425 Soldo AT, Godoy GA (1973a) Molecular complexity of Paramecium symbiont
Schmidt HJ (1984) Studies on protein synthesis in kappa particles. J Gen lambda deoxyribonucleic acid: evidence for the presence of a multicopy
Microbiol 130:1517–1523 genome. J Mol Biol 73:93–108
Soldo AT, Godoy GA (1973b) Observations on the production of folic acid by Takayanagi T, Hayashi S (1964) Cytological and cytogenetical studies on Para-
symbiont lambda particles of Paramecium aurelia stock 299. J Protozool mecium polycarium V Lethal interactions in certain stocks. J Protozool
20:502 11:128–132
Soldo AT, Godoy GA (1974) The molecular complexity of mu and pi symbiont Thiele JO, Honer-Schmid J, Wahl GK, Schultz JE (1980) A new method for axenic
DNA of Paramecium aurelia. Nucl Acid Res 1:387–396 mass cultivation of Paramecium tetraurelia. J Protozool 27:118–121
Soldo AT, Van Wagtendonk WJ (1969) The nutrition of Paramecium aurelia, Van Bruggen JJA, Stumm CK, Vogels GD (1983) Symbiosis of methanogenic
stock 299. J Protozool 16:500–506 bacteria and sapropelic protozoa. Arch Microbiol 136:89–95
Soldo AT, Godoy GA, Van Wagtendonk WJ (1966) Growth of particle-bearing and Van Bruggen JJA, Zwart KB, Van Assema RM, Stumm CK, Vogels GD
particle-free Paramecium aurelia in axenic culture. J Protozool 13:492–497 (1984) Methanobacterium formicicum, an endosymbiont of the anaerobic
Soldo AT, Van Wagtendonk WJ, Godoy GA (1970) Nucleic acid and protein ciliate Methopus striatus. McMurrich Arch Microbiol 139:1–7
content of purified endosymbiote particles of Paramecium aurelia. Biochim Van Bruggen JJA, Zwart KB, Hermans JGF, Van Hove EM, Stumm CK, Vogels GD
Biophys Acta 204:325–333 (1986) Isolation and characterization of Methanoplanus endosymbiosus sp
Soldo AT, Godoy GA, Brickson SA (1974) Infectious particles in a marine ciliate. nov., an endosymbiont of the marine sapropelic ciliate Methopus contortus.
Nature 249:284–286 Quennerstedt Arch Microbiol 144:367–374
Soldo AT, Brickson SA, Larin F (1983) The size and structure of the DNA genome Van Wagtendonk WJ, Soldo AT (1970) Methods used in the axenic cultivation of
of symbiont xenosome particles in the ciliate Parauronema acutum. J Gen Paramecium aureli. In: Prescott M (ed) Methods in cell physiology, vol 4.
Microbiol 129:1317–1325 Academic, New York, pp 117–130
Soldo AT, Brickson SA, Castiglione GA, Freytag AF (1986a) Association of Vannini C, Rosati G, Verni F, Petroni G (2004) Identification of the bacterial
transformation of xenosomes from nonkiller to killer with extrachromo- endosymbionts of the marine ciliate Euplotes magnicirratus (Ciliophora,
somal DNA. J Bacteriol 169:96–102 Hypotrichia) and proposal of ‘‘Candidatus Devosia euplotis’’. Int J Sys
Soldo AT, Brickson SA, Freytag A (1986b) Structure and characterization of Evol Microbiol 54:1151–1156
extrachromosomal DNA of killer xenosomes, intracellular symbionts of Vannini C, Petroni G, Verni F, Rosati G (2005) Polynucleobacter bacteria in the
a marine protozoan. In: Wickner RB, Hinnebusch A, Lam AM, Gunsalus brackish-water species Euplotes harpa (Ciliata, Hypotrichia). J Euk Microbol
IC, Hollaender A (eds) Extrachromosomal elements in lower eukaryotes. 52:116–122
Plenum, New York, pp 291–302 Vannini C, Ferrantini F, Schleifer KH, Ludwig W, Verni F, Petroni G (2010)
Soldo AT, Brickson SA, Castiglione GA (1987) Inclusion bodies in xenosomes purified ‘‘Candidatus Anadelfobacter veles’’ and ‘‘Candidatus Cyrtobacter comes’’ two
on percoll gradients as revealed by electron microscopy. J Protozool 34:6–9 new Rickettsiales species hosted by the protist ciliate Euplotes harpa
Sonneborn TM (1938) Mating types, toxic interactions and heredity in Parame- (Ciliophora, Spirotrichea). Appl Env Microbiol 76:4047–4054
cium aurelia. Science 88:503 Vannini C, Pöckl M, Petroni G, Wu QL, Lang E, Stackebrandt E, Schrallhammer M,
Sonneborn TM (1943) Gene and cytoplasm. I: the determination and inheritance Richardson PM, Hahn MW (2007) Endosymbiosis in statu nascendi: close
of the killer character in variety 4 of P aurelia. II: the bearing of determina- phylogenetic relationship between obligatory endosymbiotic and obligately
tion and inheritance of characters in P aurelia on problems of cytoplasmic free-living Polynucleobacter strains (Betaproteobacteria). Environ Microbiol
inheritance, pneumococcus transformation, mutations and development. 9(2):347–359.
Proc Natl Acad Sci USA 29:329–343 Verni F, Rosati G (1990) Peculiar epibionts in Euplotidium itoi (Ciliata,
Sonneborn TM (1950) Methods in the general biology and genetics of Parame- Hypotrichida). J Protozool 37:337–343
cium aurelia. J Exp Zool 113:87–148 Vogels GD, Hoppe WF, Stumm CK (1980) Association of methanogenic bacteria
Sonneborn TM (1959) Kappa and related particles in Paramecium. Adv Virus Res with rumen ciliates. Appl Environ Microbiol 40:608–612
6:229–356 Wagener S, Pfennig N (1987) Monoxenic culture of the anaerobic ciliate
Sonneborn TM (1970) Methods in Paramecium research. In: Prescott M (ed) Trimyema compressum Lackey. Arch Microbiol 149:4–11
Methods in cell physiology, vol 4. Academic, New York, pp 241–339 Wagener S, Bardele CF, Pfennig N (1990) Functional integration of
Sonneborn TM (1975) The Paramecium aurelia complex of fourteen sibling Methanobacterium formicicum into the anaerobic ciliate Trimyema
species. Trans Am Microsci Soc 94:155–178 compressum. Arch Microbiol 153:496–501
Sonneborn TM, Mueller JA, Schneller MV (1959) The classes of kappa-like Wells B, Horne RW (1983) The ultrastructure of Pseudomonas avena. II: intracel-
particles in Paramecium aurelia. Anat Rec 134:642 lular refractile (R body) structure. Micron Microsc Acta 14:329–344
Springer N, Ludwig W, Amann R, Schmidt HJ, Görtz HD, Schleifer KH (1993) Wichterman R (1953) The biology of Paramecium. Plenum Press, New York
Occurrence of fragmented 16S rRNA in an obligate bacterial endosymbiont Wiemann M (1989) The release of Holospora obtusa from Paramecium caudatum
of Paramecium caudatum. Proc Natl Acad Sci USA 90:9892–9895 observed with a new device for extended in vivo microscopy. J Protozool
Springer N, Amann R, Ludwig W, Schleifer K-H, Schmidt HJ (1996) 36:176–179
Polynucleobacter necessarius, an obligate bacterial endosymbiont of the Wiemann J, Görtz H-D (1989) Release of the endonucleobiotic bacterium
hypotrichous ciliate Euplotes aediculatus, is a member of the beta-subclass Holospora elegans from its host cell Paramecium caudatum. Eur J Protistol
of Proteobacteria. FEMS Microbiol Lett 135:333–336 25:100–108
Stackebrandt E, Frederiksen W, Garrity GM, Grimont PA, Kampfe P, Maiden MC, Wiemann M, Görtz H-D (1991) Identification and localization of major stage-
Nesme X, Rossello-Mora R, Swings J, Truper HG, Vauterin L, Ward AC, specific polypeptides of infectious Holospora obtusa with monoclonal anti-
Whitma WB (2002) Report of the ad hoc committee for the re-evaluation of bodies. J Bacteriol 173:4842–4850
the species definition in bacteriology. IJSEM 52:1043–1047 Yarlett N, Hann AC, Lloyd D, Williams A (1981) Hydrogenosomes in the rumen
Stumm CK, Vogels GD (1989) Relations of autotrophic bacteria to protozoa. protozoon Dasytricha ruminantium Schuberg. Biochem J 200:365–372
In: Schlegel HG, Bwien B (eds) Autotrophic bacteria. Springer, Berlin, Zwart KB, Goosen NK, Van Schijndel MW, Broers CAM, Stumme CK, Vogels GD
pp 177–191 (1988) Cytochemical localization of hydrogenase activity in the anaerobic
Stumm CK, Gijzen HJ, Vogels GD (1982) Association of methanogenic bacteria protozoa Trichomonas vaginalis, Plagiopyla nasuta and Trimyema
with bovine rumen ciliates. Br J Nutr 47:95–99 compressum. J Gen Microbiol 134:2165–2170
19 Bacteriocyte-Associated Endosymbionts
of Insects
Paul Baumann1 . Nancy A. Moran2 . Linda C. Baumann3
1
Division of Agriculture and Natural Resources, University of California Cooperative Extension,
Davis, CA, USA
2
Biological and Biomedical Sciences Program, Yale University, New Haven, CT, USA
3
School of Nursing, University of Wisconsin-Madison, Madison, WI, USA
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465 Introduction

Symbionts of Insects Which Utilize Plant Sap as Food . . . . 467 Intracellular associations between bacteria and insects are wide-
spread in nature (Baumann and Moran 1997; Buchner 1965;
Aphid Endosymbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468 Dasch et al. 1984; Douglas 1989; Houk and Griffiths 1980).
Buchnera: The Primary Endosymbiont of Aphids . . . . . . 468 Extensive studies of the natural history of such associations
Phylogeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468 have led to the conclusion that they are commonly found in
Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469 insects that utilize diets containing an excess of one class of
Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470 compounds but a deficiency of some essential nutrients
Physiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 472 (Buchner 1965; Dadd 1985). It was thought that the function
Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 476 of the endosymbionts was to rectify this imbalance by the syn-
Secondary Endosymbionts of Aphids . . . . . . . . . . . . . . . . . . . 484 thesis of these essential nutrients for the host. Extensive compi-
Absence of a Stable Bacterial Flora in Aphid Guts . . . . . . 485 lations of the occurrence of endosymbionts in different groups
of insects are found in Buchner (1965) and Dasch et al. (1984).
Endosymbionts of Other Plant Sap-Utilizing Insects . . . . . . 485 Because most of the prokaryotes involved in such associations
Psyllid Endosymbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 485 are not cultivable on common laboratory media, their character-
Whitefly Endosymbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 485 ization had to await the development of recombinant DNA meth-
Mealybug Endosymbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 486 odology. The past 10 years have witnessed the initiation of studies
Tsetse Fly Endosymbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 486 on the intracellular association of prokaryotes with a variety of
Wigglesworthia: The Primary Endosymbiont of insect hosts. In this chapter, we will provide an overview of the
Tsetse Flies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 486 evolution and, where possible, genetics and physiology of such
Sodalis: The Secondary Endosymbiont recently studied associations. A summary of some of their features
of Tsetse Flies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 487 is presented in > Table 19.1, and the phylogeny of the endosym-
Sitophilus (Weevils) Endosymbionts . . . . . . . . . . . . . . . . . . . . 487 bionts based on 16S rDNA is presented in > Fig. 19.1.
Phylogeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 The diversity of symbiotic associations and problems of
Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 definitions have been previously discussed and will not be con-
Physiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 sidered here (Smith and Douglas 1987; Werren and O’Neill
Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 1997). Some of the phylogenetic studies have included few
host taxa and are thus not entirely conclusive; nevertheless,
Comparisons with Other Associations . . . . . . . . . . . . . . . . . . . . . 488 current results suggest that most of the associations considered
Carpenter Ant Endosymbionts . . . . . . . . . . . . . . . . . . . . . . . . . . 488 in this chapter have common features and represent a relatively
Phylogeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488 well-defined type. To aid presentation, we will describe these
Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489 common features, which are established from recent, largely
Blattabacterium-Endosymbionts of Cockroaches molecular, studies as well as from older investigations based on
and Termites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489 morphological analyses. References to the earlier studies are
Phylogeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489 found in Buchner (1965), who arrived at similar conclusions.
Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489 References to recent studies are given as each association is
Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490 considered.
Physiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490 The associations listed in > Table 19.1 and > Fig. 19.1 are
Isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490 the results of infections of various insect lineages with different
Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491 prokaryotes. These associations became stable, resulting in the
Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 492 emergence of a new composite (of host and endosymbiont)
466 19 Bacteriocyte-Associated Endosymbionts of Insects
. Table 19.1
General properties of the considered endosymbiotic associationsa
Host 16S rRNA group or other taxonomic

categoryb Principal host food source Symbiont designation designation
Order: Homoptera
Suborder: Sternorrhyncha
Superfamily: Aphidoidea
Aphids Phloem sap Buchnera aphidicola g-Proteobacteria
S-endosymbiont Enterobacteriaceaec
Superfamily: Psylloidea
Psyllids Phloem sap P-endosymbiont g-Proteobacteria
S-endosymbiont g-Proteobacteria
Superfamily: Aleyrodoidea
Whiteflies Phloem sap P-endosymbiont g-Proteobacteria
S-endosymbiont Enterobacteriaceae
Superfamily: Coccoidea
Family: Pseudococcidae
Mealybugs Phloem sap P-endosymbiont b-Proteobacteria
Order: Diptera
Family: Muscidae
Genus: Glossina
Tsetse flies Vertebrate blood Wigglesworthia glossinidia g-Proteobacteria
(P-endosymbiont)
Sodalis glossinidius Enterobacteriaceae
(S-endosymbiont)
Order: Coleoptera
Family: Curculionidae
Genus: Sitophilus
Weevils Stored grain P-endosymbiont Enterobacteriaceae
Order: Hymenoptera
Family: Formicidae
Genus: Camponotus
Carpenter Plant nectar, honeydew, detritus, and other P-endosymbiont g-Proteobacteria
ants sources
Order: Orthoptera
Superfamily: Blattoidea
Cockroaches Universalists Blattabacterium cuenoti Flavobacterium-Bacteroides group
Order: Isoptera
Family: Mastotermitidae
Genus: Mastotermes
Termites Dead wood Blattabacterium cuenoti Flavobacterium-Bacteroides group
a
See text for references
b
Taxonomy of the host according to Borror et al. (1989)
c
As defined by Brenner (1984)
organism. The endosymbiont became heritable through the endosymbionts is congruent with the phylogeny of the hosts.
acquisition of mechanisms ensuring vertical, maternal transmis- With some exceptions, heritable associations tend to become
sion to progeny. The association also became obligate, or mutualistic (Lipsitch et al. 1995; Werren and O’Neill 1997). In
beneficial, for host growth. Because the host depended on most cases, the host cannot survive without the endosymbiont,
the association, and because horizontal or infectious transmis- or the elimination of the endosymbiont has a deleterious effect.
sion between hosts did not occur, the phylogeny of the Although the advantage for the host is in most cases apparent,
Bacteriocyte-Associated Endosymbionts of Insects 19 467
Buchnera P-endosymbiont
of aphids
Endosymbionts of
carpenter ants
P-endosymbiont
Wigglesworthia of tsetse flies
Enterobacteriaceae
P-endosymbiont of
Sitophilus (weevil) g
S-endosymbiont of
aphids, tsetse flies, whiteflies
Escherichia coli, Proteus vulgaris
P-endosymbionts of psyllids
P-endosymbionts of whiteflies
Endosymbionts of
mealybugs b
endosymbionts
Blattabacterium Flavobacterium-
of cockroaches
Bacteroides
and termites
. Fig. 19.1
Phylogenetic tree resulting from parsimony analysis of insect endosymbionts based on 16S rDNA sequence analysis. P-, primary
endosymbiont; S-, secondary endosymbiont; Greek letters, subdivisions of the Proteobacteria. References given in text
the advantage for the endosymbiont is not always clear. Perhaps host lineages and has properties characteristic of a parasite
it is more correct to think that the host domesticates the endo- (O’Neill et al. 1997). Recent work may necessitate a modification
symbiont for its own welfare, utilizing functions that are present of this conclusion because newly discovered Wolbachia in filarial
in the prokaryote but lacking in the host (Douglas and Smith worms appear to be essential for host survival and may show
1989; Maynard Smith and Szathmáry 1995). In this chapter, the phylogenetic congruence with their hosts, indicating vertical
organism which is present in all the species of an insect group evolution (Bandi et al. 1998, 1999).
and which appears to be of essential value to the host is desig-
nated by either its scientific name or, if one is lacking, by the
term primary (P-) endosymbiont. Symbionts of Insects Which Utilize Plant Sap
Superimposed on this fundamental association may be asso- as Food
ciations with additional endosymbionts. Although these are
heritable, they appear to be the result of multiple independent Aphids, psyllids, whiteflies, and mealybugs share a number of
infections, horizontal transmission, or both. Since these endo- common structural and nutritional properties (Borror et al.
symbionts may be absent in some hosts, their contribution to 1989) and constitute four separate lineages within the suborder
the welfare of the organism may not be major or essential. These Sternorrhyncha (order Homoptera; Campbell et al. 1994; von
organisms are designated as secondary (S-) endosymbionts. In Dohlen and Moran 1995). All of these groups feed predomi-
this connection, it is relevant that some bacterial strains may nantly or exclusively on plant phloem sap. This mode of life
exist within the body cavity of insects for long periods without necessitates the penetration of plant tissue by flexible tubular
obvious deleterious effects, thus serving as potential endosym- mouthparts (stylets) and the ingestion of plant phloem sap. This
biont precursors of endosymbiotic associations (Boman and diet is unbalanced, as it is rich in carbohydrates but deficient in
Hultmark 1987; Faye 1978). amino acids and other nitrogenous compounds (Dadd 1985;
These conclusions are tentative and, with the possible excep- Sandström and Pettersson 1994). Because of the low concentra-
tion of the Sitophilus-endosymbiont association, are probably tions of nitrogenous compounds, phloem-feeding insects ingest
applicable to most of the endosymbiotic associations considered a large amount of plant sap and then excrete the excess sugar as
in this chapter. One well-studied, contrasting association is honeydew. This mode of feeding is conducive to the transmis-
between many arthropods and the intracellular prokaryote, sion of plant viruses, and members of the Homoptera are
Wolbachia (O’Neill et al. 1997). Although this organism is typ- important vectors of viral plant pathogens (Blackman and
ically heritable, being transmitted maternally, results of phylo- Eastop 1984; Gray and Banerjee 1999; Sylvester 1985). In addi-
genetic studies imply some incidence of horizontal exchange tion, these insect populations may reach enormous numbers on
between very different lineages. Wolbachia causes a number of plants, causing nutrient deprivation, leaf curling, and gall for-
different reproductive alterations favoring the spread of infected mation (Borror et al. 1989).
In spite of these common properties, aphids, whiteflies, (Berlyn 1998). The organization of the rRNA genes into two
psyllids, and mealybugs have different prokaryotic transcription units suggests a possible relationship between
P-endosymbionts (> Table 19.1, > Fig. 19.1). These insects, Buchnera and the endosymbionts of carpenter ants.
like other animals, require ten essential amino acids, and endo- The results of phylogenetic analyses involving all of the
symbionts are thought to upgrade the diet by synthesizing these currently available Buchnera sequence information are presented
essential amino acids and providing them to the host (Baumann in > Fig. 19.2. Most of the characterized endosymbionts are
et al. 1995, 1997a, b; Douglas 1989; Moran and Telang 1998). Of from the family Aphididae. Based on 16S rDNA, Buchnera
these four symbiotic associations, the most extensively studied is forms one clade within which two well-supported subclades
that between Buchnera (the P-endosymbiont) and aphids. This are apparent. These are the aphids of the Aphididae and the Sc
association will be considered in some detail and followed by and Mr from the tribe Fordini in the family Pemphigidae.
a brief discussion of three other associations. Additional studies using a portion of trpB (> Fig. 19.2b) con-
firmed some of these relationships and provided further resolu-
tion within the genus Uroleucon. These relationships are in
Aphid Endosymbionts broad agreement with the results of evolutionary studies of
plasmid-associated trpE, leuBCD, and repA1 (> Fig. 19.2c, d, e;
Buchnera: The Primary Endosymbiont of Aphids Baumann et al. 1997b, 1999b; Bracho et al. 1995; Rouhbakhsh
et al. 1996, 1997; Silva et al. 1998; van Ham et al. 1997, 1999).
Phylogeny Within the genus Uroleucon, the phylogeny based on trpB is in
good agreement with the more extensive analysis of host phy-
The initial characterization of the 16S rDNA gene of Buchnera logeny based on mitochondrial and nuclear genes (Clark et al.
involved the use of an Escherichia coli 16S rDNA hybridization 2000; Moran et al. 1999).
probe to perform a restriction enzyme and Southern blot anal- The congruence of phylogenies derived from Buchnera chro-
ysis on total Acyrthosiphon pisum DNA, which established that mosomal and plasmid genes, as well as host mitochondrial and
this gene was present as a single copy (Unterman et al. 1989). nuclear genes, is strong evidence for a vertical mode of evolution
Subsequently, three overlapping DNA fragments were cloned with no exchange of either bacteria or plasmids among host
and the 16S rDNA gene was sequenced. In addition, bacteriomes lineages (Moran and Baumann 1994). An implication of the
were dissected from the aphid and the DNA purified. Restriction congruence between the phylogenies of Buchnera and
enzyme and Southern blot analysis gave the same results with corresponding aphid hosts is that dates for branch points
whole aphid DNA and DNA obtained from dissected inferred from fossil aphids can be extended to ancestral
bacteriomes, indicating that the bacteriomes were the source of Buchnera (Moran et al. 1993). A further implication is that
endosymbiont DNA. In all subsequent studies, the 16S rDNA modern Buchnera descend from an infection of a common
was obtained by PCR amplification using whole aphid DNA ancestor of all modern aphids. From the aphid fossil record,
preparations cloned into plasmid vectors and then sequenced we can infer that this infection by a free-living bacterium must
(Munson et al. 1991b). have occurred at least 150–250 million years ago. The divergence
Based on 16S rDNA analysis, Buchnera is a distinct lineage in 16S rDNA of modern Buchnera is consistent with this hypoth-
within the g-3 subgroup of the Proteobacteria (> Fig. 19.1; esis of an ancient infection followed by cospeciation of Buchnera
Moran et al. 1993; Munson et al. 1991b; Unterman et al. 1989; and hosts.
van Ham et al. 1997). The closest known organisms are the Buchnera shows a rate of base substitution in its 16S rDNA
endosymbionts of carpenter ants, endosymbionts of tsetse flies that is about twice as great as that in related free-living bacteria
(Wigglesworthia), and members of the Enterobacteriaceae as based both on relative rate comparisons with free-living taxa and
defined by Brenner (Brenner 1984; Aksoy 1995a, b; Schröder on comparisons of rates calibrated with respect to absolute time
et al. 1996). Phylogenetic analyses based on 16S rDNA indicate (Clark et al. 1999b; Moran 1996). The elevated substitution rate
that these organisms are four separate lineages but do not permit of Buchnera relative to that of related free-living bacteria extends
firm conclusions regarding their relationships to one another. to genes encoding proteins (Brynnel et al. 1998; Clark et al.
Buchnera contains a single copy of rRNA genes which are 1999b; Moran 1996; Wernegreen and Moran 1999). Based on
arranged as two transcription units, 16S rRNA and tRNAGlu- calibrated rates for protein-coding genes, synonymous sites
23S rRNA-5S rRNA (Munson et al. 1993; Rouhbakhsh and evolve about twice as fast and nonsynonymous sites about six
Baumann 1995). This organization of the rRNA genes into two times as fast in Buchnera as in E. coli/Salmonella typhimurium,
transcription units is somewhat rare but also has been found in based on an absolute time scale (> Table 19.3). The rate differ-
Wolbachia (Bensaadi-Merchermek et al. 1995) and Rickettsia ences are considerably greater on a scale based on generations,
(Andersson et al. 1998), organisms which are in the since Buchnera appears to have fewer generations per year than
a-subdivision of the Proteobacteria and also associated with do natural populations of enteric bacteria (Clark et al. 1999b).
insects. In the endosymbionts of carpenter ants, the rRNA The most plausible explanation for the faster rate of substi-
genes are also split into two transcription units (C. Sauer and tution in Buchnera is that the population structure of Buchnera,
R. Gross, personal communication), whereas in Wigglesworthia involving strictly vertical transmission of a small inoculum
(Aksoy 1995b) and the Enterobacteriaceae, the order is 16S–23S between hosts, results in greater levels of genetic drift, which
a b c d e
Chromosome Chromosome Plasmid Plasmid Plasmid
16S rDNA Sg trpB 64 Sg trpE 80 Sg 100 Sg Sg
78 Rp 100 Rp 100 Rp Rp 99 96 Rp
66 Rm Rm Rm leuB 70 Ri
Urd leuC Rc
81 leuD
94 Uas repA2
Uam
79 Uae
78
63 Uja
Usl
73 Usn Usn
Usn Usn
92 A Uo 100 A
100 82
85 Urp 64
62 Uc Uc
97 Uh 99
Urr Urr Urr
68 Ue Ue
Ujl 78 Mdr
Ml 60 98 Mrs
Mp 72 As
100
Ap Ap Ap Ap
Dn Dn Dn Dn Dn
Cv D
Ppo
Mk M Mk
Ts T Ts
Sc
Pb
Ps Ec Ppo-repAl
70
95 Mr P Mr*
Sc 100 Sc*
Tc Tc
92 Ec
Enterobacterlaceae Ec
Pv
Ra
. Fig. 19.2
Phylogenetic trees resulting from parsimony analyses using Buchnera (a) 16S rDNA, (b) trpB, (c) trpE, (d) leuB, leuC, and leuD, and (e)
repA1. Numbers at nodes are bootstrap percentages from parsimony searches (1,000 replicates). Abbreviations designating the insect
hosts are given in > Table 19.2. In (a), Enterobacteriaceae: Ec(E. coli), Pv(P. vulgaris); Ra(Ruminobacter amylophilus). Dashed lines in (a)
designate aphid species within one family: A Aphididae, D Drepanosiphidae, M Mindaridae, T Thelaxidae, P Pemphigidae. Dashed line in
(e) designates aphids within one family. Underlined abbreviations in (c) and (e) refer to aphid species not included in the other analyses. *
in (c) designates chromosomal genes. For references see text
can increase the fixation rate of mildly deleterious mutations. hosts, based on comparisons of pairwise divergences of
Several observations support this explanation. First, the rate corresponding aphid and Buchnera taxa (Moran et al. 1995).
increase is found at all genes and is concentrated at sites, such Thus, the hypothesis of a universal rate of substitution in rDNA
as nonsynonymous sites, that are expected to be under selective is not even approximately true.
constraint (Moran 1996; Wernegreen and Moran 1999). Second,
polypeptide compositions are consistently biased toward amino
acids that allow more adenine and thymine in the DNA sequence, Taxonomy
indicating that mutational bias has affected protein evolution.
Third, faster substitution rates in 16S rDNA are observed in The genus Buchnera contains one species, Buchnera aphidicola,
other insect endosymbionts that share a similar transmission and the type strain is the endosymbiont of the aphid Schizaphis
mode, suggesting that the endosymbiotic lifestyle has repeatedly graminum (Munson et al. 1991a). Currently this species name
produced the same changes in patterns of molecular evolution. designates the lineage consisting of the P-endosymbionts of
Finally, the 16S rRNA secondary structure of Buchnera and other aphids. There are over 4,000 species of aphids (Blackman and
endosymbionts has lower thermal stability than that of related Eastop 1984; Remaudière and Remaudière 1997) of which only
free-living bacteria, as expected if the DNA base substitutions are 35 have been characterized by molecular methods. Consequently
mildly deleterious (Lambert and Moran 1998). our conclusions are based on a very small sample of aphid species.
The 16S rDNA substitution rate of Buchnera is about 35 Although 16S rDNA has been useful for showing the monophy-
times greater than that of homologous regions of 18S rDNA of letic origin of aphid endosymbiosis and the establishment of
. Table 19.2
Abbreviations of aphid species used in this chapter
Abbreviation Aphid species Family Tribe

Rc Rhopalosiphum cerasifoliae Aphididae Aphidini
Ri Rhopalosiphum insertum Aphididae Aphidini
Rm Rhopalosiphum maidis Aphididae Aphidini
Rp Rhopalosiphum padi Aphididae Aphidini
Sg Schizaphis graminum Aphididae Aphidini
Ap Acyrthosiphon pisum Aphididae Macrosiphini
As Aulacorthum solani Aphididae Macrosiphini
Dn Diuraphis noxia Aphididae Macrosiphini
Ml Macrosiphoniella ludovicianae Aphididae Macrosiphini
Mp Myzus persicae Aphididae Macrosiphini
Mdr Metopolophium dirhodum Aphididae Macrosiphini
Mrs Macrosiphum rosae Aphididae Macrosiphini
Uae Uroleucon aeneum Aphididae Macrosiphini
Uam Uroleucon ambrosiae Aphididae Macrosiphini
Uas Uroleucon astronomus Aphididae Macrosiphini
Uc Uroleucon caligatum Aphididae Macrosiphini
Ue Uroleucon erigeronense Aphididae Macrosiphini
Uh Uroleucon helianthicola Aphididae Macrosiphini
Uja Uroleucon jaceae Aphididae Macrosiphini
Ujl Uroleucon jaceicola Aphididae Macrosiphini
Uo Uroleucon obscurum Aphididae Macrosiphini
Urd Uroleucon rudbeckiae Aphididae Macrosiphini
Urp Uroleucon rapunculoidis Aphididae Macrosiphini
Urr Uroleucon rurale Aphididae Macrosiphini
Usl Uroleucon solidaginis Aphididae Macrosiphini
Usn Uroleucon sonchi Aphididae Macrosiphini
Ppo Pterocomma populeum Aphididae Pterocommatinae
Cv Chaitophorus viminalis Drepanosiphidae Chaitophorini
Mk Mindarus kinseyi Mindaridae Mindarini
Mr Melaphis rhois Pemphigidae Fordini
Sc Schlechtendalia chinensis Pemphigidae Fordini
Pb Pemphigus betae Pemphigidae Pemphigini
Ps Pemphigus spyrothecae Pemphigidae Pemphigini
Tc Tetraneura caerulescens Pemphigidae Eriosomatini
Ts Thelaxes suberi Thelaxidae
major aphid subgroups, it is far too conserved to be useful for Habitat

defining relationships among endosymbionts of closely related
aphids. Some success has been obtained by the use of other, less Location and Ultrastructure
conserved, molecules (> Fig. 19.2). The 16S rDNA sequence During their reproductive phase, aphids contain within their
difference of Buchnera in Sg and Sc (the most distantly related body cavity a bilobed structure called a bacteriome consisting of
aphids) is about the same as that between E. coli and Proteus 60–90 uninucleate, polyploid cells called bacteriocytes (Douglas
vulgaris. Thus, subsequent studies using less conserved mole- and Dixon 1987). These cells are filled with host-derived vesicles
cules will probably indicate that Buchnera should be subdivided containing Buchnera (> Fig. 19.3a). This organism is spherical
into new species. So far, no studies have addressed the range of or oval in shape, 2–4 m in diameter, with a cell wall consisting of
variation within endosymbionts of a single aphid species. two-unit membranes, as is characteristic of Gram-negative
. Table 19.3
Substitution rates in Buchnera and enteric bacteriaa
Synonymous rate Nonsynonymous rate

Species pair Estimated time of divergence Absoluteb Generationc Absoluteb Generationc
Buchnera (Sg/Dn) 50–70 MY 6.8–9.5 0.14–0.19 1.0–1.4 0.02–0.03
Buchnera (Sc/Mr) 50–70 MY 5.1–7.2 0.17–0.24 1.1–1.6 0.04–0.05
E. coli/S. typhimurium 100–150 MY 2.9–4.4 0.03–0.04 0.1–0.2 0.001–0.002
a
Based on comparisons of over 5,100 codons (Clark et al. 1999b)
b
Substitutions/site/10 years
c
Substitutions/site/10 generations
Number of Endosymbionts
Buchnera contains one copy of the 16S rRNA gene (Munson
et al. 1991b, 1993; Unterman et al. 1989) and one copy of groEL
(Ohtaka et al. 1992; Hassan et al. 1996). The number of copies of
Buchnera 16S rRNA genes in the aphid Sg was studied by quan-
titative PCR (Baumann and Baumann 1994) and was found to
be 0.5–1.2 107 mg1 aphid wet weight. Using quantitative
hybridization of a Buchnera groEL probe, the number of genome
copies in Ap was estimated at 1–2 107 mg1 aphid wet weight
(Humphreys and Douglas 1997). In both of these studies, the
number of Buchnera cells was assumed to be identical to the
number of genome copies. However, a recent study has demon-
strated that Buchnera (Ap) is polyploid, containing an average of
about 120 genomes per cell (Komaki and Ishikawa 1999). If the
average number of genomes per endosymbiont is relatively
constant, then the number of endosymbionts is about 100-fold
less than estimated previously, or about 105 mg1 aphid wet
weight. This value is considerably lower than the estimates for
Ap of 1.6–1.8 106 endosymbionts mg1 aphid wet weight,
based on microscopic enumeration of the endosymbionts
(Harrison et al. 1989).
Growth and Reproduction

In their most active stage, aphids are wingless females, which
reproduce by parthenogenesis, giving birth to live young. There
is telescoping of generations in that the mother aphid contains
embryos that, in turn, may contain other embryos (Dixon 1973,
1992). Studies on the growth of Sg (Baumann and Baumann
1994) have indicated that newly born aphids weigh 24 mg and
contain 2 105 copies of the Buchnera genome. The increase in
. Fig. 19.3 the number of endosymbiont genomes parallels the increase in
Electron micrographs of Buchnera, the P-endosymbiont of aphids. the weight of the aphid. The maximum weight is reached
(a) Endosymbionts within a bacteriocyte, bar = 1 m. (b) Higher in 9–10 days at which time the aphid weighs 540 mg and contains
magnification showing the Gram-negative cell wall (large arrow) 5.6 106 Buchnera genomes. The endosymbionts are
and the vesicle membrane (small arrow), bar = 0.5 m (Photos partitioned between maternal and embryonic bacteriocytes. In
courtesy of Marv Kinsey and Don McLean) a mature aphid, most of the Buchnera genomes are found in the
embryos (Humphreys and Douglas 1997). The first young are
bacteria (> Fig. 19.3b; Akhtar and van Emden 1994; Griffiths born in 8–9 days; each aphid can produce 60–80 live young
and Beck 1973; Hinde 1971b; McLean and Houk 1973). A thin during its lifetime. Douglas and Dixon (1987) showed that,
layer corresponding to peptidoglycan has been detected (Houk during the period of growth, there is a concomitant increase in
et al. 1977). The presence of peptidoglycan also is indicated by the maternal bacteriocyte volume as well as a small drop in
penicillin-induced alterations of the cell wall as well as by chem- bacteriocyte number. In the adult aphid, the number of maternal
ical analysis (Griffiths and Beck 1974; Houk et al. 1977). bacteriocytes in the bacteriome undergoes a sharp decrease
probably due, in part, to their dispersion within the abdomen, her collaborators (Douglas and Prosser 1992; Wilkinson 1998).
their degradation, as well as the degradation of Buchnera The S-endosymbiont probably does not perform any essential
(Brough and Dixon 1990; Douglas and Dixon 1987; Griffiths functions for the host (see section on > ‘‘Secondary Endosym-
and Beck 1973; Hinde 1971a). bionts of Aphids’’ in this chapter); nevertheless, the use of
Aphids may also produce sexual forms with the females antibiotics eliminates both endosymbionts, and consequently
depositing eggs that overwinter and hatch in the spring. the observed effects of this loss may not be attributable solely
Buchnera is maternally transmitted (transovarial transmission) to the loss of Buchnera.
to both developing embryos and eggs. Maternal bacteriocytes Currently one of the more complete studies involves the
adjacent to an embryo near the blastoderm stage form a small essential amino acid tryptophan. Using a strain of Ap containing
opening through which the endosymbionts pass. Buchnera then an S-endosymbiont, Douglas and Prosser (1992) have shown
moves through the intervening hemolymph and enters a nearby a sparing effect of tryptophan in chlortetracycline-containing
opening on the oocyte surface. During early embryonic devel- synthetic diets on aphid growth. In addition, they detected low
opment, the presumptive bacteriocytes form, and the endosym- levels of tryptophan synthase in Buchnera and found that activ-
bionts migrate into these cells (Buchner 1965; Blackman 1987; ity was absent in chlortetracycline-treated and thus endosymbi-
Hinde 1971a). Symbiont invasion of eggs also occurs from the ont-free aphids. The assays used (Smith and Yanofsky 1962)
dispersed bacteriocytes, and they can be observed as an aggregate crude extracts of whole aphids as well as preparations enriched
at the posterior pole of the mature egg (Buchner 1965; Brough in endosymbionts and measured the disappearance of the sub-
and Dixon 1990). Buchnera and bacteriomes appear to be nearly strate indole and not the appearance of the product tryptophan.
universal in aphids (Buchner 1965). However, some species of Indole or indole derivatives may be substrates for a variety of
the tribe Cerataphidini lack both and instead contain yeastlike reactions catalyzed by enzymes found in crude extracts of
extracellular symbionts within their body cavity (Buchner 1965; insects. No information was provided about the linearity of
Fukatsu and Ishikawa 1992a, 1996). Some species of aphids may increasing enzyme activity with increasing crude extract con-
produce dwarf males and/or sterile female soldiers that may also centration. In spite of possible difficulties with this assay, the
lack endosymbionts (Buchner 1965; Fukatsu and Ishikawa dependence of the reaction on ‘‘the substrate [sic] pyridoxal
1992b; Fukatsu et al. 1994). phosphate’’ (Douglas and Prosser 1992) is consistent with it
being a measure of tryptophan synthase activity.
Using synthetic diets containing radiolabeled sulfate, it was
Physiology shown that Buchnera can reduce this compound to the level of
hydrogen sulfide and incorporate it into methionine and cyste-
Nutrition and Metabolism ine and that these amino acids are found in aphid tissue
Plant sap, the diet of aphids, has an excess of carbohydrate (Douglas 1988). Using 14C-radiolabeled amino acids, it was
relative to amino acids and other nitrogenous compounds found that the synthesis of the essential amino acids arginine,
(Dadd 1985; Douglas 1998; Sandström and Pettersson 1994; threonine, isoleucine, and lysine was reduced or eliminated by
Sandström and Moran 1999). Aphids, like other insects, are the inclusion of rifampicin in the diet (Liadouze et al. 1996).
thought to require 10 preformed amino acids, and these essen- Sasaki and Ishikawa (1995) also showed that treatment of aphids
tial amino acids are present in low amounts in plant sap. Some with rifampicin eliminated the incorporation of dietary
15
species of aphids can grow on synthetic diets even in the absence N-glutamine into the essential amino acids arginine, histidine,
of essential amino acids. Adding antibiotics to such diets results isoleucine and/or leucine, phenylalanine, threonine, and valine.
in the elimination of endosymbionts and the failure of the Glutamine is the predominant amino acid in phloem and
aphids to reproduce. There is some sparing effect when the also in aphid hemolymph (Sandström and Pettersson 1994;
essential amino acids are included in the antibiotic-containing Sasaki et al. 1990). Isolated bacteriocytes were found to take up
diet. These experiments have generally been interpreted as indi- glutamine and convert it to glutamate, which subsequently was
cating that one of the functions of Buchnera is the synthesis of taken up by Buchnera (Sasaki and Ishikawa 1995). Isolated
essential amino acids for the aphid host (reviewed by Baumann endosymbionts incorporated the nitrogen of glutamine into the
et al. 1995; Douglas 1998). A major problem is that compared essential amino acids isoleucine, leucine, valine, and phenylala-
with growth on plants, growth on artificial diets is poor and nine as well as a number of other amino acids, and these amino
generally limited to a few generations. In addition, aphid growth acids were excreted into the suspending medium. Whitehead and
on complete synthetic diets in the presence of antibiotics is even Douglas, however, could not find any evidence for excretion of
worse, indicating that Buchnera provides nutrients or functions amino acids by Buchnera (cited in Douglas 1997).
that cannot be provided by the artificial diets. The effects of Using synthetic diets, Nakabachi and Ishikawa (1999) dem-
antibiotics on a number of aphid properties have been recently onstrated a requirement for riboflavin by rifampicin-treated
reviewed (Wilkinson 1998). There is a further complication with aphids. These results indicate that Buchnera is required for the
some of the nutritional studies, in which physiological effects synthesis of at least one vitamin for the aphid host.
have been attributed to the removal of Buchnera. The aphid Whitehead and Douglas (1993) isolated vesicles containing
strain used may also contain S-endosymbionts, as is true of the Buchnera and showed that they readily took up acetic, glutamic,
strain of the aphid Ap used in the studies of A. E. Douglas and and aspartic acid as well as tricarboxylic acid cycle intermediates
and oxidized them to CO2. Oxygen consumption was also . Table 19.4
detected and was greatly reduced by KCN. These results suggest Genes of Buchnera from the aphid S. graminuma
the presence of a tricarboxylic acid cycle in the endosymbionts
Gene Linkage
and indicate a respiratory metabolism. The latter conclusion is
symbol Gene product description groupb
consistent with the presence of a gene for a subunit of NADH
dehydrogenase I, an enzyme involved in the generation of I. Small-Molecule Metabolism
a proton motive force during respiration, and of all the genes B. Energy metabolism
for ATP synthase, a membrane-associated enzyme which utilizes 1. Glycolysis
the proton motive force for the synthesis of ATP (> Table 19.4). gap A Glyceraldehyde-3-phosphate 13
dehydrogenase
Gene Expression tpiA Triosephosphate isomerase 3
Buchnera messenger RNA (mRNA) has been detected for 5. Pentose phosphate pathway
a variety of genes encoding proteins involved in amino acid
(a) Oxidative branch
biosynthesis (> Table 19.4). This includes genes for amino
gnd Gluconate-6-phosphate dehydrogenase 2
acids of the glutamate (argA) and aspartate (thrB) families
(Nakabachi and Ishikawa 1997), the shikimate pathway (aroH), 7. Respiration
as well as the biosynthetic pathway for tryptophan (trpE, trpD, (a) Aerobic
trpA), branched-chain amino acids (ilvI, ilvD, leuA), and histi- nuoC(D)c NADH dehydrogenase I, subunits cd
dine (hisG; Baumann et al. 1999a). Buchnera mRNA has been (c) Electron transport
detected also for a gene involved in the biosynthesis of riboflavin fdx Ferredoxin 1
(ribE; Nakabachi and Ishikawa 1999) as well as for genes fpr Ferredoxin-NADP reductase 12
involved in the heat shock response (groEL, groES, dnaK, dnaJ;
9. ATP proton motive force
> Table 19.4; Sato and Ishikawa 1997a, b). Numerous Buchnera
atp A ATP synthase, a-subunit 1
proteins have also been detected by immunological methods.
These include GroEL, GroES, and DnaK (Kakeda and Ishikawa atp B ATP synthase, subunit a 1
1991; Sato and Ishikawa 1997b) as well as ribosomal protein S1 atp C ATP synthase, E-subunit 1
(the product of rpsA; Clark et al. 1996) and the protein involved atp D ATP synthase, b-subunit 1
in septum formation during cell division (the product of ftsZ; atp E ATP synthase, subunit c 1
> Table 19.4; Baumann and Baumann 1998).
atp F ATP synthase, subunit b 1
In bacteria, rRNA genes are transcribed from strong pro- atp G ATP synthase, g-subunit 1
moters. Comparisons of the regions upstream of rRNA genes
atp H ATP synthase, -subunit 1
from six species of Buchnera indicated conservation of sequences
D. Amino acid biosynthesis
resembling the 35 and 10 regions of s70 promoters as well as
boxA and boxC (Munson et al. 1993; Rouhbakhsh 1995). Similar 1. Glutamate family
putative 35 and 10 regions were found in Buchnera plasmids argAd N-acetylglutamate synthase
containing genes for tryptophan and leucine biosynthesis argH Argininosuccinate lyase 7
(Baumann et al. 1999b; Rouhbakhsh et al. 1996; Silva et al. 2. Aspartate family
1998). dapD Succinyldiaminopimelate aminotransferase 4
thrA Aspartokinase I
GroEL Overproduction and Its Significance
thrBd Homoserine kinase
In Buchnera, the chaperonin, GroEL, constitutes a major frac-
3. Serine family
tion of the total protein (Sato and Ishikawa 1997a). In addition,
GroEL is present in aphid hemolymph (van den Heuvel et al. cysE Serine acetyltransferase 7
1994). Overproduction of GroEL is a characteristic of some serC Phosphoserine aminotransferase 3
endosymbionts and pathogens in the intracellular environment 4. Aromatic amino acid family
(Hogenhout et al. 1998). This protein mediates the folding of aroA 5-Enolpyruvylshikimate-3-phosphate 3
peptides into their functional forms as well as the repair of synthase
damaged proteins (Gross 1996). Buchnera GroEL is able to aroC Chorismate synthase 2
complement E. coli mutants (Ohtaka et al. 1992). GroEL has aroE Dehydroshikimate reductase 8
been localized in maternal and embryonic Buchnera by immu-
aroH 3-Deoxy-D-arabino-heptolosonate-7- 10
nohistochemistry (Fukatsu and Ishikawa 1992c). Electron phosphate synthetase (DAHP synthetase)
micrographs indicate that the purified Buchnera GroEL has the
trpA Tryptophan synthase, A protein 5
characteristic double-ring appearance observed with the E. coli
trpB Tryptophan synthase, B protein 5
protein (Filichkin et al. 1997; Hara and Ishikawa 1990). The
endosymbiont protein has ATPase activity, and in the presence
of E. coli, GroES could reconstitute denatured Rhodospirillum
. Table 19.4 (continued) . Table 19.4 (continued)
Gene Linkage Gene Linkage

symbol Gene product description groupb symbol Gene product description groupb
trpC(F) Indole-3-glycerol-phosphate synthetase 5 rrl 23S rRNA 8
and N-(5-phosphoribosyl)anthranilate rrs 16S rRNA 11
isomerase
2. Ribosomal protein synthesis and modification
trpD Phosphoribosylanthranilate transferase 5
rplL 50S ribosomal protein L7/L 12 9
trpE (p)e Anthranilate synthase, A subunit 16
rplT 50S ribosomal protein L20 10
trpG (p)e Anthranilate synthase, B subunit 16
rpmH 50S ribosomal protein L34 1
(glutamine amidotransferase)
rpmI 50S ribosomal protein L35 10
5. Histidine
rpsA 30S ribosomal protein S1 3
hisA N-(50 -phospho-L-ribosyl-formimino)-5- 2
amino-1-(50 -phosphoribosyl)-4- rpsD 30S ribosomal protein S4 14
imidazolecarboxamide isomerase rpsK 30S ribosomal protein S11 14
hisB Imidazoleglycerol-phosphate dehydratase 2 4. tRNAs
and histidinol-phosphate phosphatase tRNAGlu Glutamate-tRNA 8
hisC Histidinol-phosphate aminotransferase 2 tRNAPhe Phenylalanine-tRNA 1
hisD Histidinol dehydrogenase 2 tRNATrp Tryptophan-tRNA 1
hisF Imidazoleglycerol-phosphate synthase 2 5. Aminoacyl tRNA synthetases and their modification
subunit (with HisH) argS Arginine tRNA synthetase 11
hisG ATP phosphoribosyltransferase 2 aspS Aspartic tRNA synthetase 3
hisH Glutamine amidotransferase subunit (with 2 cysS Cysteine tRNA synthetase 8
HisF)
metG Methionine tRNA synthetase 2
hisI Phosphoribosyl-AMP cyclohydrolase and 2
serS Serine tRNA synthetase 3
phosphoribosyl-ATP pyrophosphatase
thrS Threonine tRNA synthetase 10
7. Branched-chain family
trmE tRNA methyltransferase 1
ilvC Acetohydroxy acid isomeroreductase 1
7. DNA replication, restriction/modification, and recombination
ilvD Dihydroxyacid dehydratase 1
dnaA DNA biosynthesis, initiation of 1
ilvH Acetohydroxyacid synthase, small subunit 4
chromosome replication, global
ilvI Acetohydroxyacid synthase, large subunit 4 transcription regulator
leuA (p)e 2-Isopropylmalate synthase 15 dnaG DNA biosynthesis, DNA primase 7
leuB (p)e 3-Isopropylmalate dehydrogenase 15 dnaN DNA polymerase III holoenzyme, b-subunit 1
leuC (p)e Isopropylmalate isomerase subunit 15 dnaQ DNA polymerase III holoenzyme, e-subunit 11
leuD (p)e Isopropylmalate isomerase subunit 15 gidA Chromosome replication? 1
F. Purines, pyrimidines, nucleosides, and nucleotides gyrB DNA gyrase subunit B 1
3. 20 -Deoxyribonucleotide metabolism himD Integration host factor, b-subunit 3
dcd 20 -Deoxycytidine 50 -triosephosphate 2 rep Rep helicase, ssDNA-dependent ATPase 1
deaminase
8. Protein translation and modification
trxB Thioredoxin reductase 3
efpf Elongation factor EF-P
G. Biosynthesis of cofactors, prosthetic groups, and carriers
infC Initiation factor IF-3 10
9. Riboflavin
tufg Elongation factor EF-Tu
ribEd Riboflavin synthase, b-chain
9. RNA synthesis, RNA modification, and DNA transcription
10. Thioredoxin, glutaredoxin, and glutathione
rho Transcription termination factor Rho 1
trxA Thioredoxin 1
rpoA RNA polymerase, a-subunit 14
II. Broad Regulatory Functions
rpoB RNA polymerase, b-subunit 9
rpoD RNA polymerase, s70 subunit 7 0
rpoC RNA polymerase, b -subunit 9
rpoHd RNA polymerase, s32 subunit, regulation of
11. Phospholipids
proteins induced at high temperature
clhh Cardiolipin synthase
III. Macromolecule Metabolism
B. Degradation of macromolecules
A. Synthesis and modification of macromolecules
1. RNA
1. rRNA and ‘‘stable’’ RNAs
rnh RNase H 11
rrf 5S rRNA 8
rnpA RNase P 1
. Table 19.4 (continued) . Table 19.4 (continued)
Gene Linkage Gene Linkage

symbol Gene product description groupb symbol Gene product description groupb
3. Proteins, peptides, and glycopeptides ORF194 1788860 1
hslU Heat shock protein, protease? 12 ORF217 1787362 13
htrA Periplasmic serine protease and heat shock 4 ORF235 1786354 6
protein ORF312 1786270 4
sohi Periplasmic protease ORF340 1788543 4
C. Cell envelope ORF453 1788858 1
2. Surface polysaccharides, lipopolysaccharides, and antigens ORFI 1786406 11
kdtB Putative enzyme of lipopolysaccharide 12 ORFV 1787508 5
synthesis
ORFVI 1787507 5
4. Murein sacculus and peptidoglycan
P14 1787506 5
ddlB D-alanine-D-alanine ligase 6
ORFA 1788269 Transmembrane protein? 5
murC L-alanine-adding enzyme, UDP-N- 6
ORFB 1787524 5
acetylmuramate-alanine ligase
ORFC 1787361 ABC transporter protein? ATP- 13
murE meso-Diaminopimelate-adding enzyme 4
binding site?
IV. Cell Processes
ORFD 1789158 13
B. Chaperones
pLeu
dnaJd Heat shock protein
repA1 Related to RepA protein of IncFII plasmids 15
dnaKd Heat shock protein (Hsp 70), DNA (p)e
biosynthesis
repA2 Related to RepA protein of IncFII plasmids 15
groEL Heat shock protein (Hsp 60) 1 (p)e
groES Heat shock protein (Hsp 10) 1 ORF1 Related to E. coli 1789376 15
hscA Cold shock protein (Hsp 70) 1 (p)e
hscB Cold shock protein 1 a
Gene list arranged according to the classification of gene products of Riley
C. Cell division and Labedan (1996). Genes from Buchnera from other species of aphids are
indicated
ftsA Cell division protein, complexes with FtsZ 6 b
See > Table 19.5
ftsI Septum formation; penicillin binding 4 c
From Buchnera (Usn)
protein 3; peptidoglycan synthase d
From Buchnera (Ap) (Nakabachi and Ishikawa 1997, 1999; Sato and Ishikawa
ftsL Cell division protein; ingrowth of wall at 4 1997a, b)
e
septum p, plasmid-associated gene
f
From Buchnera (Mp) (Hogenhout et al. 1998)
ftsZ Cell division, forms circumferential ring 6 g
From Brynnel et al. (1998)
h
E. Protein and peptide secretion From Buchnera (Dn, Usn)
i
From Buchnera (Sc) (Lai et al. 1995)
secB Protein export, molecular chaperone 7
V. Other
F. Adaptations and atypical conditions rubrum ribulose-1,5-bisphosphate carboxylase (Kakeda and
ibp Heat shock protein, HSP20 family 12
Ishikawa 1991).
The Buchnera groESL operon organization resembles that of
Additional ORFs
E. coli (Hassan et al. 1996; Hogenhout et al. 1998; Ohtaka et al.
bcp E. coli homolog Bacterioferritin comigratory 2
1992). Upstream of groES are nucleotide sequences characteris-
protein 1788825
tic of the 35 and 10 regions of s32 promoters. A message of
nifS 1788879 1 2.1 kb (containing both groES and groEL) is made by the endo-
gibN 1790040 7 symbiont using only this promoter (Sato and Ishikawa 1997a). It
gjEA Hypothetical lysine tRNA synthase 12 is not understood why GroES is low in the endosymbiont, in
homolog 1790599 contrast to the high quantities of GroEL (Kakeda and Ishikawa
10 kDa YIDD_ECOLI 1 1991). The genes for s32 (rpoH) as well as dnaKJ have been
39 kDa 1790589 1 cloned and sequenced (Sato and Ishikawa 1997a, b). The latter
60 kDa 1790140 1 also are transcribed solely from a s32 promoter. In E. coli as well
ORF113 1786351 6 as other organisms, transcription of the groESL operon and
the dnaKJ operon is part of the s32 regulon, and their
ORF128 1788878 1
synthesis is increased by heat shock (Gross 1996). It would
ORF177 1788671 2
appear that this mode of regulation is modified in Buchnera
. Table 19.5 van den Heuvel et al. 1994; Hogenhout et al. 1998). These viruses
Order of genes on DNA fragments of Buchnera from the aphid replicate in the plant and are ingested by aphids when they feed
S. graminuma on phloem sap. Subsequently they are transported from the
digestive tract into the hemolymph and from there into the
Chromosomal genes
salivary gland for transmission to plants via salivary secretions.
1. (34.7 kb, AF008210) 39 kDa-groEL-groES-tRNAPhe-trmE- The viruses are retained in an infective form (without replica-
60 kDa-rnpA-rpmH-dnaA-dnaN-gyrB-atpCDGAHFEB-gidA- tion) in the hemolymph throughout the life span of the aphid.
ORF194-0RF453-fdx-hscA-hscB-ORF128-nifS-tRNATrp-ilvD-ilvC- There is evidence that the GroEL that is found in the hemo-
rep-trxA-rho lymph coats the virus particles and protects them from host
2. (12.8 kb, AF067228) bcp-aroC-ORF177-hisG-hisD-hisC-hisB- defenses. A region in Buchnera GroEL has been identified
hisH-hisA-hisF-hisI-gnd-dcd-metG which is essential for binding to the virus (Hogenhout et al.
3. (11.5 kb, L43549) aspS-trxB-serS-serC-aroA-rpsA-himD-tpiA 1998), and similarly a portion of a viral capsid protein has been
4. (9.7 kb, AF060492) dapD-htrA-ORF340-IlvI-ilvH-ORF312-ftsL- identified as the region to which the endosymbiont GroEL binds
ftsI-murE (van den Heuvel et al. 1997). Transmission of plant viruses may
5. (8.4 kb, Z19055) ORFB-ORFA-trpD-trpC(F)-trpB-trpA-ORFV- be advantageous to the aphid because infected plants have
ORFVI-P14 higher levels of nutrients in their sap (Blua et al. 1994).
6. (6.8 kb, AF012886) murC-ddlB-ftsA-ftsZ-ORF113-ORF235 An additional reason for the constitutive synthesis of high
7. (6.5 kb, M90644) dnaG-rpoD-cysE-secB-yibN-argH amounts of GroEL by Buchnera may be to compensate for the
8. (6.1 kb, U09230) aroE-tRNAglu-rrl-rrf-cysS accumulated amino acid substitutions which have occurred at
a high rate in this endosymbiont (Moran 1996; > Table 19.3).
9. (5.0 kb, Z11913) rplL-rpoB-rpoC
These slightly deleterious changes may result in proteins of
10. (4.5 kb, U11066) aroH-thrS-infC-rpmI-rplT
decreased stability, and the high levels of GroEL may compen-
11. (4.4 kb, L18927) argS-rrs-ORF1-rnh-dnaQ sate for these changes, allowing proper folding and retention of
12. (4.1 kb, AF108665) hslU-ibp-fpr-yjeA-kdtB function. A similar role for chaperones in masking deleterious
13. (3.9 kb, U11045) ORFC-ORF217-gapA-ORFD mutations has been recently suggested on the basis of work on
14. (0.9 kb, M74510) rpsK-rpsD-rpoA Drosophila heat shock protein (Pennisi 1998; Rutherford and
Plasmid-associated genes Lindquist 1998).
15. (8.0 kb, AF041836) leuA-leuB-leuC-leuD-repA1-ORF1-repA2
16. (3.6 kb, Z21938) trpEG
Genetics
a
Numbers followed by parenthesis indicate linkage groups, numbers within
parentheses indicate size of fragment and GenBank number. See > Table 19.4
Genome Analysis
for description of gene products
The guanine + cytosine (G + C) content of Buchnera is about
28 mol% (Clark et al. 1998a; Ishikawa 1987). The genome size of
the endosymbiont from the aphid Ap has been found to be
(Sato and Ishikawa 1997a, b). Synthesis of groESL and dnaKJ 0.657 Mb (Charles and Ishikawa 1999). This is considerably
mRNA is constitutive and is not increased by heat shock. This below such free-living organisms as E. coli (4.6 Mb; Blattner
conclusion is supported by the observation that there is no et al. 1997) and Haemophilus influenzae (1.8 Mb; Fleischmann
increase in the level of total GroEL in aphids shifted from et al. 1995) and the intracellular pathogens, Chlamydia
23 C to 33 C for one day (Baumann et al. 1996). trachomatis (1.0 Mb; Stephens et al. 1998) and Rickettsia
Baumann et al. (1996) arrived at an estimate of the amount prowazekii (1.1 Mb; Andersson et al. 1998). The Buchnera
of GroEL per Buchnera cell, based on the premise that the genome is somewhat larger than that of the pathogen Myco-
endosymbiont contained only one genome copy. Recently it plasma genitalium (0.58 Mb; Fraser et al. 1995). An unusual
has been shown that Buchnera is polyploid, containing an aver- property of the Buchnera genome is that it appears to be present
age of 120 genome copies per cell (Komaki and Ishikawa 1999). as about 120 copies per cell (Komaki and Ishikawa 1999).
If this result is incorporated into the calculation, an impossible Approximately 130 kb of DNA have been sequenced from
excess of GroEL would be present in each endosymbiont cell. Buchnera (from Sg) (Baumann and Baumann 1998; Baumann
Since protein extracts of whole aphids were used in the estima- et al. 1995; Clark et al. 1998a, b; Thao and Baumann 1998). The
tion of GroEL content, this result could be explained by the choice of this aphid was predicated on the fact that it contains
finding that GroEL is also present in the hemolymph (van den only one endosymbiont as indicated by morphological exami-
Heuvel et al. 1994). nations as well as extensive restriction enzyme and Southern blot
Perhaps the major economic effect of aphids on agriculture analysis of whole aphid DNA, using probes for many different
is their ability to transmit plant viruses (Blackman and Eastop genes. The latter results indicated that, with the exception of
1984; Gray and Banerjee 1999; Sylvester 1985). Buchnera- plasmid-amplified DNA, only one copy of the targeted genes was
derived GroEL has been implicated in the survival of present in the endosymbiont genome. In many cases, the DNA
luteoviruses in the hemolymph (Filichkin et al. 1997; that was used for cloning was also endosymbiont-enriched.
There were many independently cloned DNA fragments with . Table 19.6
identical overlapping sequences, indicating that the aphids did Codon usage of Buchnera from the aphid S. graminuma
not contain several closely related endosymbionts. At least 20 kb
AA Codon Fraction AA Codon Fraction
of DNA also have been sequenced from Buchnera of each of the
aphids Dn, Sc, and Mr (Baumann et al. 1998a; Clark et al. 1999b; Phe UUU - 0.933 Ala GCA - 0.470
Lai et al. 1995, 1996). > Table 19.4 lists, under different func- Phe UUC - 0.067 Ala GCG - 0.052
tional categories, the genes found in Buchnera, primarily in (Sg). Leu UUA - 0.663 Tyr UAU - 0.854
The order of these genes in the DNA fragments is presented in Leu UUG - 0.091 Tyr UAC - 0.146
> Table 19.5. A total of 126 open reading frames were detected,
Leu CUU - 0.132 His CAU - 0.865
of which 101 corresponded to E. coli genes with known function.
Leu CUC - 0.011 His CAC - 0.135
The remaining 25 open reading frames all had homologs of no
known function in the E. coli chromosome (Blattner et al. 1997). Leu CUA - 0.087 Gln CAA - 0.887
> Table 19.6 presents the codon usage of the Buchnera structural Leu CUG - 0.016 Gln CAG - 0.113
genes. As expected from the G + C content, there is a strong bias Ile AUU - 0.576 Asn AAU - 0.863
for A and T, especially in the third codon position. This bias also Ile AUC - 0.075 Asn AAC - 0.137
affects the composition of proteins, favoring amino acids for Ile AUA - 0.349 Lys AAA - 0.918
which codons contain more A and T (Moran 1996; Clark et al. Met AUG - 1.000 Lys AAG - 0.082
1999b).
Val GUU - 0.474 Asp GAU - 0.875
Buchnera was found to contain dnaA, encoding a protein
Val GUC - 0.054 Asp GAC - 0.125
which initiates bidirectional chromosome replication, and ftsZ,
encoding a protein involved in septum formation during cell Val GUA - 0.407 Glu GAA - 0.913
division (Baumann and Baumann 1998; Lai et al. 1992a). Val GUG - 0.065 Glu GAG - 0.087
Among other genes that were found are those encoding proteins Ser UCU - 0.448 Cys UGU - 0.826
for peptidoglycan synthesis, cell division, DNA replication, DNA Ser UCC - 0.039 Cys UGC - 0.174
transcription, ribosomal proteins, amino acid tRNA synthases, Ser UCA - 0.273 Trp UGG - 1.000
ATP synthase, electron transport, protein secretion, and glycol- Ser UCG - 0.034 Arg CGU - 0.348
ysis. In addition, genes for three tRNAs were detected. Genes
Ser AGU - 0.177 Arg CGC - 0.045
encoding homologs of proteins involved in the E. coli heat shock
Ser AGC - 0.029 Arg CGA - 0.148
response (groEL, groES, htrA, dnaK, dnaJ) and the cold shock
response (hscA, hscB) were also detected (Clark et al. 1998a; Pro CCU - 0.456 Arg CGG - 0.013
Hassan et al. 1996; Ohtaka et al. 1992; Sato and Ishikawa Pro CCC - 0.069 Arg AGA - 0.416
1997b). Nakabachi and Ishikawa (1999) detected a gene (ribE) Pro CCA - 0.420 Arg AGG - 0.030
encoding a protein involved in riboflavin biosynthesis. In addi- Pro CCG - 0.055 Gly GGU - 0.464
tion, some of the genes encoding enzymes for the biosynthesis of Thr ACU - 0.474 Gly GGC - 0.042
aromatic amino acids (shikimate pathway, tryptophan branch), Thr ACC - 0.041 Gly GGA - 0.446
branched-chain amino acids (isoleucine, valine, leucine), lysine,
Thr ACA - 0.433 Gly GGG - 0.048
cysteine, and serine as well as genes for the complete pathway of
Thr ACG - 0.052 Ter UAA - 0.887
histidine biosynthesis were found (Clark et al. 1998a, b; Thao
and Baumann 1998). The presence of genes for enzymes of Ala GCU - 0.426 Ter UAG - 0.094
amino acid biosynthesis is in marked contrast to the obligate Ala GCC - 0.052 Ter UGA - 0.019
intracellular pathogens Rickettsia prowazekii and Chlamydia a
Based on 19,037 codons (Clark et al. 1998a)
trachomatis, as well as such fastidious organisms as Mycoplasma
genitalium and Borrelia burgdorferi, all of which lack genes
encoding enzymes of amino acid biosynthesis (Andersson et al.
1998; Fraser et al. 1995, 1997; Stephens et al. 1998). Retention of Compared to the sequenced bacterial genomes, this organism
amino acid biosynthetic genes by Buchnera probably reflects the is unusual in that 24 % of its DNA is noncoding. In addition it
role of these pathways in the mutualistic association with the has a number of pseudogenes. These findings are interpreted as
host aphids. Overall, these results on gene content indicate that a stage in the adaptation of R. prowazekii to an intracellular
Buchnera has many of the properties of free-living bacteria and lifestyle, involving the loss of genes encoding metabolic path-
would appear to be, in many respects, a self-contained, physio- ways for products that are provided by the host. Currently about
logically autonomous unit enclosed within bacteriocyte-derived 20 % of the Buchnera genome has been sequenced, and the
vesicles. organization of the genes in these genome fragments is highly
Currently the most interesting comparison of the Buchnera compact and similar to that of other bacteria (Clark et al. 1998a).
genome is with the recently sequenced obligate intracellular These findings indicate that if, as seems probable, Buchnera
pathogen R. prowazekii, an organism which is a member of the originated from an organism with a larger genome (Charles
a-subdivision of the Proteobacteria (Andersson et al. 1998). and Ishikawa 1999), then the reduction in the genome size has
. Table 19.7 Amplification has been interpreted as a means of increasing the

Designations and characteristics of Buchnera plasmidsa rate of synthesis of end products. These include genes of purine
biosynthesis in Borrelia (Margolis et al. 1994), cysteine biosyn-
1. pTrpEG. Two or more DnaA boxes in a putative origin of
replication (> Fig. 19.5a–h). Arrangement of DnaA boxes varies, thesis in Synechococcus (Nicholson et al. 1995), and histamine
one conserved pattern is recognized and designated at ori-3.6. biosynthesis in Vibrio (Barancin et al. 1998).
(> Fig. 19.5a–c). The plasmids usually consist of tandem
repeats of identical or similar units containing gene(s) for The trpEG-Containing Plasmids and Gene Silencing > Figure 19.4
a putative anthranilate synthase (TrpEG), the first enzyme of the is an outline of the aromatic amino acid biosynthetic pathway. It
tryptophan biosynthetic pathway consists of a common portion leading to chorismate (shikimate
2. pTrpEG-R. Plasmid contains genes for putative replication pathway) and branches to (1) phenylalanine and tyrosine as well
initiation proteins (repAC1, repAC2) which are related to as (2) tryptophan. In the shikimate pathway, arrows that have
replication initiation proteins of broad host range plasmids of designations correspond to genes detected in Buchnera (Sg). The
the IncA/C group (> Fig. 19.5i, > 19.7c). Within the DNA activity of the tryptophan biosynthetic pathway is regulated by
encoding the C-terminal end of the replication initiation anthranilate synthase (TrpEG) which is feedback inhibited by
proteins and/or downstream of this gene are 19 nt-long
tryptophan (Crawford 1989). In Buchnera from 11 species of
repeats which correspond to putative interons (> Fig. 19.5i).
aphids, trpEG has been found to be plasmid-associated
Plasmid consists of tandem repeats of similar units and
contains genes for a putative anthranilate synthase (TrpEG), the
(> Fig. 19.5; Baumann et al. 1997b; Lai et al. 1994, 1996;
first enzyme of the tryptophan biosynthetic pathway Rouhbakhsh et al. 1996, 1997; van Ham et al. 1999). The
remaining genes of the pathway [trpDC(F)BA] have been
3. pLeu. Plasmid contains genes for putative replication initiation
proteins (repA1, repA2) which are related to replication found to have a chromosomal location in all cases examined
initiation proteins of plasmids of the IncFII group [Buchnera (Sg, Dn, Sc, Mr)] (Baumann et al. 1998a; Clark et al.
(> Fig. 19.9a–c). Putative origin of replication downstream 1999b; Lai et al. 1995). In contrast to the situation in Aphididae,
of repA1. Plasmids contain one copy of genes encoding for in Buchnera (Sc, Mr), trpEG is not plasmid-associated but is
enzymes of leucine biosynthesis (leuA, leuB, leuC, leuD) present as one copy on the endosymbiont chromosome
a
For references see text
(> Fig. 19.4; Clark et al. 1999b; Lai et al. 1995).
The structure of plasmids of the pTrpEG type usually
consists of tandem repeats of a nearly identical unit
(> Fig. 19.5a–h). In Buchnera (Sg, Rp), the plasmids contain
already been accomplished and what has been retained is the four tandem repeats of a 3.6 kb unit; in Buchnera (Rm), the
essential gene complement required for the endosymbiotic plasmid consists of one 3.6 kb unit; while in Buchnera (Ap)
association. plasmids containing 5, 6, or 10 tandem repeats are found.
Buchnera (Sg) contains about 4 plasmids per endosymbiont
Plasmid-Associated Amplification of Biosynthetic Genes chromosome resulting in a 16-fold trpEG amplification. In
Some species of aphids have Buchnera in which genes for Buchnera (Sc, Mr), in which trpEG is chromosomal, these two
enzymes of amino acid biosynthesis are amplified on plasmids genes are found between fpr and hslU (> Fig. 19.6). In Buchnera
(Bracho et al. 1995; Lai et al. 1994; van Ham et al. 1999). This (Sg), in which trpEG is plasmid-associated, trpEG is absent from
plasmid-associated gene amplification has been interpreted as this chromosomal location, consistent with its transfer to the
an adaptation of Buchnera to an endosymbiotic association in plasmid (> Fig. 19.6). Instead ibp is present at this position,
which one of its functions is the overproduction of amino acids. suggesting a concomitant or a subsequent acquisition of this
This interpretation is based on analogies with other prokaryotic gene (Clark et al. 1999a).
systems in which gene amplification is viewed as an attribute of The plasmids in > Fig. 19.5a–h all share in common the
prokaryotic genome plasticity allowing the organism to adapt to presence of 2–5 DnaA boxes which are 9 nt-long sequences to
new environments (for recent reviews see Romero and Palacios which the DnaA protein binds, thereby initiating chromosomal
1997; Roth et al. 1996). Currently Buchnera has been found to replication (Messer and Weigl 1996). There is also considerable
contain three different types of plasmids. The properties of these variation in the length of the repeated units (2.6–3.6 kb). Within
plasmids are summarized in > Table 19.7. The plasmid- this plasmid group, a readily recognized subset contains
amplified genes encode the first enzyme of the tryptophan a unique arrangement of three DnaA boxes and a conserved
biosynthetic pathway (TrpEG) and four enzymes (LeuA, LeuB, region upstream of trpEG which has been designated as ori-3.6
LeuC, LeuD) of the leucine portion of the branched-chain (> Fig. 19.5a–c; Lai et al. 1996; Rouhbakhsh et al. 1996).
amino acid biosynthetic pathway. For purposes of grouping of Buchnera in aphids of the genus Uroleucon (which is derived
plasmid types and ease of presentation, we have used the plas- from Buchnera within the cluster that has ori-3.6-containing
mid designations given in > Table 19.7 followed by the abbrevi- plasmids, > Fig. 19.2b, c) have trpEG units which show a consid-
ation corresponding to the aphid species (> Table 19.2). When erable size range and substantial differences in the arrangement
first used, the designation is followed by the original plasmid of the DnaA boxes (> Fig. 19.5d–g; Baumann et al. 1997b;
name given in parentheses (if applicable). There are a number of Rouhbakhsh et al. 1997). All of these Buchnera are from
other examples of plasmid amplification of biosynthetic genes. aphid species of the family Aphididae (> Fig. 19.2a). Plasmid
aroE aroH Erythrose 4-phosphate

DAHP +
aroA Phosphoenolpyruvate
Folate
aroC
Phenylalanine
Chorismate
Tyrosine
trpEG Anthranilate Tryptophan
Anthranilate trpD trpC(F) trpB trpA

synthase Sg, Dn, Sc, Mr
Tryptophan
trpE synthase
Sc, Mr
trpG
. Fig. 19.4
Outline of the pathway for the biosynthesis of aromatic amino acids. Arrows, single enzymatic reactions; dashed arrows, several
enzymatic reactions; striped arrow, feedback inhibition; genetic designations above striped line, genes detected in Buchnera (Sg); other
genes, detected in Buchnera from the designated host aphids; circle, plasmid containing one or multiple copies of trpEG; DAHP, 3-deoxy-
D-arabinoheptulosonate 7-phosphate. For a description of genes, see > Table 19.4; for references see text
Aphid species
ori-3.6 (units/plasmid)
pTrpEG-Sg,-Rp*
trpE
trpG (4)
a 3.6 kb pTrpEG-Ap
repAC trpE (5, 6, 10)
trpG
b 3.6 kb
pTrpEG-Rm
(1)
trpE trpG
c pTrpEG-Dn
3.2 kb (1; 6,7,8 Ψ trpEG)
trpE trpG
d pTrpEG-Urr
2.6 kb
trpE trpG
e 2.6 kb
pTrpEG-Ue
trpE trpG
f pTrpEG-Uc
3.0 kb
trpE trpG
g 3.2 kb
pTrpEG-Usn
(1; 8,9,13 ΨΔtrpE-trpG)
trpE trpG
h pTrpEG-Tc
3.0 kb (1, 2)
repAC2 trpE trpG

i pTrpEG-R-Ps
3.0–3.8 kb (1; 5,6 repAC-trpG)
(4–12)
. Fig. 19.5
Genetic maps of the repeated units which constitute trpEG-containing plasmids. Filled arrowheads, position and direction of DnaA boxes
which are components of a putative origin of replication; circle on stem in (b) and (i), position of a 19 nt sequence similar to the interon of
the broad host range plasmid RA1; ori-3.6, putative origin of replication found primarily in plasmids consisting of 3.6 kb repeat units
(boxed); striped line, conserved sequence; dashed line, DNA that has not been sequenced; arrow in (i), 19 nt repeated sequence
corresponding to a putative interon; Rp*, TrpEG-Rp also contains a remnant of repAC. For references see text
pTrpEG-Tc (pBTc2; > Fig. 19.5h) is from an aphid within the DnaA boxes between trpE and trpG (> Fig. 19.5h) and not
family Pemphigidae (van Ham et al. 1999). The predominant upstream of trpEG as is the case of the other plasmids
form of this plasmid consists of one 3.0-kb unit; a minor form (> Fig. 19.5a–g). However, since there is considerable
consists of two units. A distinctive feature is the presence of rearrangement of the DnaA boxes within plasmids in Buchnera
of Uroleucon (> Fig. 19.5d–g), it is plausible that the arrange- plasmid RA1, while pTrpEG-Rp from a closely related aphid has
ment of DnaA boxes within pTrpEG-Tc is not a fundamental a remnant of repAC (van Ham et al. 1999).
difference but a variation on the arrangement observed in the In several pTrpEG plasmids, the expression of most of the
other trpEG-containing plasmids (> Fig. 19.5a–g). trpEG copies appears to be silenced (Baumann et al. 1997b; Lai
A totally different trpEG-containing plasmid is pTrpEG-R-Ps et al. 1996; van Ham et al. 1999). In Buchnera (Dn), there are
(pBPs2; > Fig. 19.5i), which does not contain DnaA boxes but about two copies of pTrpEG-Dn for each endosymbiont genome
instead has putative replication initiation proteins (RepAC) (Lai et al. 1996). Plasmid pTrpEG-Dn consists of a single 3.2-kb
which are related to those of plasmids of the broad host range unit containing an open reading frame corresponding to the
group IncA/C (van Ham et al. 1999). Within the DNA encoding putative protein TrpEG (> Fig. 19.7a). This is followed
the C-terminal portion of repAC and/or downstream of it are by a 2.6-kb unit containing trpEG pseudogenes and 5–7 repeats
4–12 repeats of a 19 nt-long sequence corresponding to of a 3.2-kb unit also containing trpEG pseudogenes.
a putative interon. In addition, there is a single copy of a (By pseudogenes, we mean segments of DNA which are clearly
19 nt-long sequence similar to the interon sequence of IncAC recognizable as trpEG but which contain numerous frameshifts
plasmid RA1 (Llanes et al. 1996). Curiously pTrpEG-Rm and stop codons preventing the synthesis of an intact protein.)
(> Fig. 19.5b) contains in its DNA a gene for a putative RepAC A comparison of the sequences between the 3.2 kb fragments
protein and the 19 nt-long sequence similar to that found in with and without pseudogenes indicated 244 differences of
which 93 % were localized in an approximately 900-bp DNA
segment which included the putative promoter and the
kdtB yjeA fpr ibp hslU N-terminus of trpE (> Fig. 19.7a). These changes should result
Sg 4.1 kb in the reduction or elimination of mRNA synthesis; if messenger
is made, it would be translated into short peptides because of the
numerous frame shifts and stop codons in the region of the
N-terminus of trpE.
trpE trpG yiiU Another instance of trpEG silencing is illustrated by pTrpEG-
Sc, Mr 3.9 kb Usn (> Fig. 19.7b). This plasmid consists of a 3.2-kb unit of
fpr hslU
trpEG followed by 10–14 2.1 kb units consisting of DNA with
. Fig. 19.6 a deletion of about 56 % of the N-terminal region of trpE and an
Genetic map of similar chromosomal DNA fragments from intact trpG. A more remarkable example of gene silencing is
Buchnera (Sg) as well as (Sc, Mr). Buchnera (Sg) lacks trpEG due to found in pTrpEG-R-Ps (> Fig. 19.7c), which consists of
its presence on a plasmid, while in Buchnera (Sc, Mr), trpEG is a 3.6–3.8-kb unit containing repAC2-trpEG followed by 1.8 kb
chromosomal. For a description of genes, see > Table 19.4; for units consisting of repAC1, a deletion of trpE, and an intact trpG
references, see text (van Ham et al. 1999). Preceding the latter is a short DNA
a pTrpEG-Dn
3.2 kb 2.6 kb (3.2 kb)5–7
trpE YtrpE Y trpE
trpG YtrpG Y trpG
b pTrpEG-Usn
(2.1 kb)10–14
3.2 kb
trpE trpDY E
trpG trpG
c pTrpEG-R-Ps
(3.6–3.8 kb) (1.8 kb)5–6
trpE repAC1
repAC2 trpG trpG

DtrpE
. Fig. 19.7
Genetic map of plasmids which contain silenced trpEG. Stippled line in (a), region which in the pseudogene containing fragments
has most of the changes; striped line in (a) and (b), pseudogenes (C); d in (b), deletion of the N-terminal portion of trpE; striped line in
(c), sequence homologous to the end of trpE. For references see text
segment which appears to be a remnant of DNA encoding the for industrial purposes; an increase in allosterically inhibitable
C-terminus of trpE. The structure of this plasmid suggests that, TrpEG is the primary means of achieving excretion of high
in Buchnera (Ps) of the family Pemphigidae, trpEG amplification amounts of tryptophan into the medium (Katsumata and
had an origin independent of that of the remaining plasmids Ikeda 1993).
shown in > Fig. 19.5. The initial plasmid probably contained In free-living bacteria, gene amplification is frequently tran-
tandem repeats of repAC-trpEG. Subsequently there was selec- sient. Its persistence depends on a constant selective pressure,
tion pressure for gene silencing resulting in a plasmid with one the absence of which leads to a rapid decrease in number of
intact copy of repAC-trpEG and tandem repeats of repAC-trpG repeats primarily by means of RecA-mediated homologous
which contain a deletion of the putative promoter region and recombination (Roth et al. 1996). There may be differences or
most of trpE. fluctuations in the levels of tryptophan in aphid diets; further-
more, the high level of TrpEG protein may impose an energy
Speculation Concerning trpEG Amplification The following is burden on the endosymbiont. Both of these conditions could
a summary of the results obtained from studies of trpEG in provide short-term selective pressure for the elimination of
Buchnera. (1) trpEG amplification is widespread in Buchnera plasmids from Buchnera (Baumann et al. 1997a). Consequently
within the family Aphididae and is also present in at least two some mechanism of stabilization may be necessary for the
members of the family Pemphigidae. In Buchnera from two maintenance of trpEG-containing plasmids. The gene recA has
aphid species of the latter family, trpEG is chromosomal and is been cloned and sequenced from many bacterial taxa, and ade-
found in the same location. In most cases, trpEG amplification is quate oligonucleotide primers are available for its amplification
affected by plasmids consisting of tandem repeats of the same or by PCR (Eisen 1995). We have made extensive unsuccessful
similar unit. (2) Evolution of plasmid-associated trpEG is verti- attempts to detect this gene by PCR, suggesting that it may be
cal, that is, Buchnera from different aphid species do not absent from Buchnera or is greatly modified. Once plasmid
exchange plasmids. (3) The trpEG-containing plasmids consti- stabilization occurs, the aphid may encounter conditions in
tute at least two replicon types, one of which is based on the which trpEG amplification is no longer necessary due to avail-
presence of DnaA boxes, while the other is based on the presence ability of tryptophan in the diet. If the usual mechanisms
of interons and replication initiation proteins related to plas- (homologous recombination?) which effect a decrease in trpEG
mids of the IncA/C group. Buchnera from the related aphids Rm amplification are absent, then one way of reducing the poten-
and Rp which belong to the first replicon type also have a gene tially wasteful synthesis of TrpEG is gene silencing.
and/or the remnant of a gene for a replication initiation protein In the past, we have speculated that trpEG amplification is
related to plasmids of the IncA/C group. (4) In trpEG plasmids a property of rapidly growing aphids and that gene silencing may
of both replicon types, gene silencing of some of the trpEG occur when following plasmid stabilization, the diet of aphids is
tandem repeats is observed. (5) In Buchnera (Sg), the sequence nutritionally enriched (Baumann et al. 1997a; Lai et al. 1994,
of two of the 3.6-kb units is virtually identical, as is the sequence 1995, 1996). Recent studies have indicated that these specula-
of the 2.6- and 3.2-kb trpEG pseudogene-containing units of tions are overly simplistic. It has been suggested that the lack of
Buchnera (Dn). While results above are derived from genetic trpEG amplification in Buchnera (Sc, Mr) which are in the
analyses, the interpretations that follow are speculative and Pemphigidae is due to the slow growth rate of these aphids
based in large part on analogies with other prokaryotic systems. compared to that of aphids within the Aphididae (Lai et al.
Gene amplification is currently viewed as a reversible aspect 1995). There are, however, few studies on the growth rate of
of genome plasticity which occurs at a frequency considerably aphids, and the finding of amplification in Buchnera (Ps, Tc)
higher than that of mutation in structural genes (Romero and which are in the family Pemphigidae and presumably also have
Palacios 1997; Roth et al. 1996). Gene amplification is frequently a slow growth rate makes this explanation questionable. Dn
used by an organism as a means of increasing the amount of causes major tissue histolysis of plants, and it has been suggested
a growth-limiting enzyme to levels beyond that achieved by gene that the presence of higher amounts of tryptophan in the diet
regulation of expression. One of the functions of Buchnera is the may be the explanation for gene silencing (Lai et al. 1996).
synthesis of essential amino acids (including tryptophan) for the Recent studies on amino acid composition of ingested phloem
aphid host. In almost all prokaryotes, the limiting enzyme sap are not strongly supportive of this hypothesis as a sole
TrpEG (which is feedback inhibited by tryptophan) regulates explanation for the presence of pseudogenes (Sandström and
the activity of the tryptophan biosynthetic pathway. The Moran 1999; Sandström et al. 2000; Telang et al. 1999). Buchnera
Buchnera enzyme is probably also feedback inhibited by trypto- (Rp) and Buchnera (Sg) are similar in that both have pTrpEG
phan since trpE has the conserved amino acid residues that are consisting of four tandem repeats of a 3.6-kb unit (> Fig. 19.5a).
involved in feedback inhibition (Lai et al. 1994). To overproduce In spite of this similarity, Rp causes essentially no modification
tryptophan, the activity of TrpEG must be increased, the poten- of the amino acid composition of plant phloem, while Sg causes
tial effect of tryptophan accumulation on activity must be over- substantial increases (Sandström et al. 2000). Dn contains
come, or both. Since even in high concentrations of tryptophan a plasmid with trpEG pseudogenes (> Fig. 19.7a). The ingested
the activity of TrpEG is not fully inhibited, an increase in enzyme diet of Dn has approximately doubled concentrations of tryp-
protein will result in increased tryptophan production. This has tophan, yet the changes it causes in the amino acid composition
been the case in experiments on overproduction of tryptophan of phloem are less than those caused by Sg. This suggests that
pseudogene formation is not solely the result of increased die- Threonine thrA
Aspartate
tary tryptophan (Sandström et al. 2000; Telang et al. 1999).
However, in the case of Usn, the presence of trpEG pseudogenes α-Ketobutyrate + Pyruvate + Pyruvate Leucine
is consistent with the finding of unusually high levels of essential ilvlH
amino acids in the phloem diet (Sandström and Moran 2000).
ilvC
Thus, the availability of nutrients in plant sap may be a partial leuA
explanation for trpEG amplification and pseudogene formation, ilvD
leuB leuC
but other factors must also be involved. α-Ketoisovalerate
In this connection, it should be mentioned that the past leuD
speculations attempt to correlate adaptations of Buchnera with Isoleucine Valine
properties of the aphid host (growth rate, modification of nutri- . Fig. 19.8
ent content of plant sap). The environment of the endosymbiont Outline of the pathway for branched-chain amino acid
is the bacteriocyte vesicle, which harbors the endosymbiont. biosynthesis. Arrows, single enzymatic reactions; horizontal lines,
This environment is a reflection of the activities of the aphid enzymatic activities functional in both isoleucine and valine
host and is a function of its ability to obtain nutrients from the biosynthesis; dashed arrow, four enzymatic reactions; circle,
plant as well as its demands on the biosynthetic attributes of the plasmid containing genes for leucine biosynthesis. thrA was
endosymbiont. Therefore, host properties, such as the efficiency detected in Buchnera (Sg); ilvIH, ilvC, and ilvD were detected in
of nutrient uptake from the plant and their transformation and Buchnera (Sg, Dn, Sc, Rm); leuACBD were detected in Buchnera (Sg,
delivery, may determine the nutritional parameters within the Dn, Rm). For a description of genes, see > Table 19.4; for
bacteriocyte vesicles and impose the selective pressure resulting references, see text
in Buchnera adaptation to the endosymbiotic association.
Some of the phenomena encountered in pTrpEG from
Buchnera also have been found in other systems. Promoter
inactivation by multiple sequence changes is the mechanism explain the presence of repAC and its remnant in Buchnera
used for silencing the expression of the Bordetella pertussis (Rm, Rp). One possibility is that it is the result of another
toxin gene (Gross and Rappuoli 1988) and the expression of invasion by a similar plasmid. However, organization of ori-3.6
the Bordetella urease gene cluster (McMillan et al. 1998). The in Buchnera (Rm, Rp) closely resembles that of Buchnera of
changes resemble those observed in gene silencing of trpEG in related aphids, suggesting a common plasmid origin for this
Buchnera (Dn; > Fig. 19.7a). Multiple copies of nearly identical group (> Fig. 19.5a–c). Alternatively, van Ham et al. (1999)
chromosomal enzyme-encoding genes have been found in suggested that a plasmid of the pTrpEG-R type is the ancestor
Thiobacillus ferrooxidans (Kusano et al. 1991) and Nitrosospira of all trpEG amplification plasmids. This hypothesis requires the
sp. (Norton et al. 1996). This situation is similar to that found subsequent occurrence of multiple losses of repAC and the
with the repeats of trpEG and trpEG pseudogenes and has led to interons in pTrpEG-R type plasmids and the acquisition of
the postulation of mechanisms for the preservation of sequence DnaA boxes in Buchnera of Aphididae.
identity of the repeated units (Klotz and Norton 1998).
The phylogenetic trees constructed on the basis of plasmid- pLeu Plasmids > Figure19.8 is an outline of the pathway of
associated genes are congruent with the phylogenetic trees based branched-chain amino acid biosynthesis. The gene for
on Buchnera chromosomal genes (> Fig. 19.2a–c). One excep- aspartokinase (thrA) has been found in Buchnera (Sg). The
tion, the basal position of Buchnera (Tc; > Fig. 19.2c), is prob- genes ilvIH, ilvC, and ilvD have been found in Buchnera (Sg,
ably an artifact arising from the more rapid change of trpE in this Dn, Sc, Mr) and encode three enzymes which function in both
lineage (van Ham et al. 1999). These results strongly suggest that the isoleucine and valine biosynthetic pathways (Clark et al.
there is no exchange of trpEG-containing plasmids between 1998c, 1999b; Thao and Baumann 1998). The pathway of leucine
endosymbionts of different aphids. Plasmids of the pTrpEG biosynthesis is a branch off the valine pathway (> Fig. 19.8).
type (> Fig. 19.5a–h) could have an endogenous origin. DnaA Bracho et al. (1995) found that in Buchnera (Rp), the genes for
boxes are found in other locations of the Buchnera genome leucine biosynthesis (leuABCD) were present on a plasmid
(Clark et al. 1998c), and their assembly with trpEG could gen- (> Fig. 19.9a). This plasmid represents a third type
erate a separate replicon. In contrast to these plasmids, the (> Table 19.7), designated pLeu, which is characterized by the
repAC genes of pTrpEG-R are related to replication initiation presence of genes (repA1, repA2) encoding putative replication
proteins of IncA/C plasmids. Thus, this plasmid may be the initiation proteins related to those of plasmids of the IncFII
result of an invasion of Buchnera by an exogenous plasmid that incompatibility group. Besides these genes, pLeu also contains
recombined with endosymbiont genes, resulting in their ampli- ORF1 encoding a putative membrane-associated protein. The
fication. It has been established that some bacteria may persist closely related Buchnera (Sg, Rp, Dn) all contain very similar
for a long time in insects; conceivably, such organisms trans- plasmids of 7.8–8.0 kb in which the genes are arranged in the
ferred their plasmids to Buchnera during the infection of same order (> Fig. 19.9a; Baumann et al. 1999a; Bracho et al.
embryos or eggs at a stage at which the endosymbionts are not 1995). In Buchnera from the more distantly related pLeu-Ppo
sequestered within bacteriocytes. It is, however, difficult to (> Fig. 19.2e), there are rearrangements of the repA genes and
ori
leuA leuC repA1 repA2
a pLeu-Sg,
7.8–8.0 kb
-Rp, -Dn leuB leuD ORF1
ori
leuA leuC ORF1 repA2
b pLeu-Ppo
7.5 kb
leuB leuD repA1
ori
leuC leuA ORF1 repA1
c pLeu-Ts
8.5 kb
leuB leuD ibp repA2
ori
repA1
d pBTc1 1.7 kb
ORF1
. Fig. 19.9
Genetic maps of pLeu plasmids. Unless indicated by an arrow, transcription is left to right. Ori, putative origin of replication; striped line,
repA1 downstream of which is ori. For a description of genes, see > Table 19.4; for references, see text
ORF1 (> Fig. 19.9b; Silva et al. 1998). All of these aphids are chromosomal and plasmid genes. These results indicate that
within the family Aphididae. In pLeu-Ts (pBTs1; > Fig. 19.9c), the pLeu plasmids are not exchanged among endosymbionts
which is from an aphid of the family Thelaxidae, there is also from different aphid species and that their evolution is vertical,
a rearrangement of the leu genes as well as an acquisition of ibp as is the case with the trpEG-containing plasmids.
which encodes a heat shock protein (van Ham et al. 1997). All of
these plasmids have a conserved region, downstream of repA1, Unanswered Questions, Other Possible Adaptations One unan-
which is probably an origin of replication (ori) (Baumann et al. swered question is why only the trp and leu genes are amplified.
1999; Bracho et al. 1995; van Ham et al. 1997). Remarkably The endosymbiont produces other essential amino acids for the
pBTc1 (> Fig. 19.9d), a 1.7-kb plasmid from Buchnera (Tc), aphid host, and their overproduction would in principle also be
which is found in an aphid belonging to the family Pemphigidae, enhanced by plasmid amplification. It has been speculated that
contains only ori, repA1, and ORF1 and probably constitutes aphids make indole acetic acid which is involved in gall forma-
a minimal replicon. tion (Forrest 1987). In many plant pathogens that cause gall
In Buchnera (Sg), there are about 24 copies of pLeu per formation, tryptophan is the precursor of indole acetic acid
endosymbiont genome, while in Buchnera (Dn), there are only (Patten and Glick 1996). Amplification of trpEG is, however,
two copies (Thao et al. 1998). This difference in functional gene found in aphids that do not produce galls (Sg, Dn, Rp, Usn) and
copy number parallels that observed with pTrpEG in the endo- is absent in some that do (Sc, Mr). Consequently this does not
symbionts of these two aphid species. In the case of pLeu, in appear to be a probable explanation for trpEG amplification.
which only one copy of the genes is present, the reduction in Leucine, lysine, valine, arginine, and threonine are the most
amplification in Buchnera (Dn) is achieved by means common amino acid in aphids (Sandström and Moran 1999),
of a reduction of copy number. In pTrpEG, which contains yet only genes for leucine biosynthesis have been detected on
tandem repeats of the same unit, the reduction in amplification plasmids. It is possible that, in other cases where amino acids are
is accomplished by means of pseudogene formation overproduced for the host, an increase in enzyme activity is
(> Fig. 19.7a). obtained by increasing the expression of the gene(s) by pro-
The similarities of pLeu plasmids suggest a single origin with moter modification. Alternatively changes of the allosteric prop-
pBTc1 (> Fig. 19.9d) being the ancestral state (Baumann et al. erties of regulated enzymes may allow retention of activity in the
1999). In the lineage common to the Aphididae and the presence of end products.
Thelaxidae, there was probably a duplication of repA1 and the A possible example of the latter is cysE of Buchnera (Sg; Lai
acquisition of leu genes. This was followed by a rearrangement of and Baumann 1992b). This gene encodes an enzyme of the
the genes and in one lineage the acquisition of ibp [for another biosynthetic pathway of cysteine, and its activity is regulated
interpretation, see van Ham et al. (1997)]. Silva et al. (1998) by cysteine feedback inhibition (Kreditch 1996). It has been
sequenced repA2 from Buchnera of six additional aphids. established that the amino acids at the C-terminus of the
A phylogeny based on this gene (> Fig. 19.2e) as well as a more E. coli enzyme are involved in cysteine feedback inhibition
limited analysis based on leu genes (> Fig. 19.2d) is congruent (Denk and Böck 1987). The Buchnera (Sg) enzyme lacks these
with trees established on the basis of other Buchnera C-terminal amino acids and consequently is probably not
subject to feedback inhibition by the end product; this change related species, Acyrthosiphon kondoi, was found to be
would result in cysteine overproduction (Lai et al. 1992b). pathogenic. Although the rate of maternal transmission of
both the S-endosymbiont and the Rickettsia sp. was high, one
instance of S-endosymbiont loss was observed (Chen and
Secondary Endosymbionts of Aphids Purcell 1997).
Studies based on light and electron microscopy have
Besides Buchnera, many aphids have additional endosymbionts suggested that some S-endosymbionts in some aphid species
usually called secondary (S-) endosymbionts (Buchner 1965; may inhabit syncytial cells or, possibly, bacteriocytes that appear
Houk and Griffiths 1980; Moran and Baumann 1994). In similar to those containing Buchnera (Buchner 1965; Hinde
many cases, these endosymbionts are spheres or rod-shaped 1971b; Iaccarino and Tremblay 1973). Fukatsu and Ishikawa
with different width and length; they were initially recognized (1993) surveyed 61 aphids for the presence of S-endosymbionts.
by differences in size and shape from the round or oval Buchnera. Previously it was found that Buchnera (Ap) overproduced
The S-endosymbionts are also maternally inherited. They have GroEL (Kakeda and Ishikawa 1991). Using anti-E. coli GroES
not been extensively studied, and most of the available informa- for immunoprecipitation of Buchnera (Ap) GroES, it was con-
tion is for the S-endosymbionts of the aphid Ap. Electron cluded that Buchnera produced low levels of this protein
microscopic studies have shown that the rod-shaped (Kakeda and Ishikawa 1991). This antiserum as well as anti-E.
S-endosymbionts are located within vesicles found in the flat- coli GroEL was used to detect GroEL and GroES in immunoblots
tened, syncytial, sheath cells which surround the bacteriome of whole aphid extracts as well as for histochemical detection in
(Griffiths and Beck 1973). Using a probe derived from E. coli thin sections of aphids. From these experiments, it was con-
16S rDNA and restriction enzyme and Southern blot analysis of cluded that the synthesis of substantial amounts of GroES by
total aphid DNA, it was found that the S-endosymbiont from Ap S-endosymbionts distinguishes them from Buchnera and that
contained a single copy of the 16S rDNA gene (Unterman et al. this property can be used for the identification of
1989). DNA obtained from dissected bacteriocytes gave the same S-endosymbionts (Fukatsu and Ishikawa 1993). These studies
restriction pattern, indicating that the S-endosymbiont was have a number of problems that limit their general applicability
located in the bacteriome. Two DNA fragments of 2.3 kb each to the survey of S-endosymbionts. The principal one is the use of
were cloned, and the 16S rDNA sequence determined. antisera against E. coli GroES for the detection of cross-reactivity
A phylogenetic analysis indicated that the S-endosymbiont was of GroES from organisms that have an unknown relationship to
a member of the Enterobacteriaceae (> Fig. 19.1). As in other E. coli. Since it is probable that many of the S-endosymbionts are
members of this family, the 16S rDNA gene of the members of the Enterobacteriaceae, a stronger cross-reaction
S-endosymbiont was directly upstream of 23S rDNA (Unterman would be expected with their proteins than with the proteins
and Baumann 1990). from Buchnera; consequently, an increased reactivity need not
Chen and Purcell (1997) found that 88 % of the strains of Ap indicate a major difference in the amount of the protein. Con-
had the S-endosymbiont. In addition, the S-endosymbionts of versely, in those cases where the S-endosymbiont is not
Ap and Macrosiphum rosae were identical, suggesting recent a member of the Enterobacteriaceae, the distant relationship
infection or horizontal transmission. Interestingly, it was also may preclude a strong cross-reaction (Eremeeva et al. 1998).
found that the hemolymph of 48 % of Ap strains contained The cross-reactivity of the Buchnera and S-endosymbiont pro-
a rod-shaped organism which had a 16S rDNA sequence nearly teins with the anti-E. coli protein antisera has not been com-
identical to that of Rickettsia bellii, an organism found in ticks pared. Finally, the relative production of GroES may vary among
(Chen et al. 1996). As a result of these studies, strains of Ap S-endosymbionts of different types.
became available which had (1) only the S-endosymbiont, The studies of Fukatsu and Ishikawa (1993, 1998), in which
(2) only the Rickettsia sp., as well as (3) neither of these two thin sections of aphids were stained by immunohistochemistry
organisms. Chen et al. (1996) injected one or both of these and examined by light microscopy, do suggest that in many
organisms into Ap, which originally lacked both, and observed aphids the S-endosymbionts occupy bacteriocytes distinct
their effect on fecundity, longevity, and the length of the repro- from those containing Buchnera. The authors also state that
ductive period. The results were complex in that they were the S-endosymbionts have a variety of different shapes. The
affected by the plant on which the aphids grew and by the methods used and the photographs presented do not, however,
temperature of growth. At 20 C, both the S-endosymbiont allow adequate visualization of cell shape and the resolution of
and the Rickettsia sp. reduced the fecundity, longevity, and bacteriocyte structure. In addition, the designation of some of
reproductive period of Ap on clover but had no significant the endosymbionts as Buchnera or S-endosymbionts appears to
effects on Ap grown on alfalfa or sweet pea (Chen et al. 1996). be arbitrary.
In some cases, both of these organisms appeared to cause an Fukatsu et al. (1998) used group-specific oligonucleotide
increase in the fitness of Ap when grown at 25 C. These results probes for in situ detection of aphid P- and S-endosymbionts.
suggest that both the S-endosymbiont and the Rickettsia sp. can A universal eubacterial 16S rRNA probe was used as well as
have either a deleterious or a beneficial effect on the host, probes specific for the 23S rRNA g- and b-subdivision
depending on the environmental conditions (Chen et al. Proteobacteria. The sequences of the latter two probes differ by
1996). The S-endosymbiont, upon injection into the closely only one nucleotide. Buchnera and S-endosymbionts all
hybridized with the 16S eubacterial probe, although the inten- from three distinct bacterial groups (> Table 19.1, > Fig. 19.1).
sities of the signal differed considerably. Curiously the putative These insects have obligatory sexual reproduction with the
P-endosymbionts of two out of seven aphids did not hybridize young hatching from eggs (Borror et al. 1989). The endosymbi-
with the 23S g-subdivision probe. Using the total DNA prepa- onts are housed within bacteriocytes, and at least during some
ration from these aphids, the 16S rDNA was amplified, cloned, stage of the insect’s life cycle, the bacteriocytes are associated
and sequenced. Two sequences were detected in each aphid DNA with the ovarioles, resulting in the transmission of endosymbi-
preparation, and one of these was related to Buchnera 16S rDNA. onts to the eggs (Buchner 1965).
Based on this result, it was concluded that these aphids
contained Buchnera but that their 23S rDNA gene was changed
to such an extent that hybridization with the g-subdivision Psyllid Endosymbionts
probe no longer occurred. This conclusion is questionable
since the 23S rDNA of Buchnera (Sg, Dn, Sc, Mr), which span Fukatsu and Nikoh (1998) sequenced the 16S rDNA of endo-
the diversity of aphid hosts (> Fig. 19.2), contains the exact symbionts from Anomoneura mori, while Spaulding and von
sequence complementary to the probe used (Clark et al. 1999b; Dohlen (1998) performed a similar study of the endosymbionts
Rouhbakhsh and Baumann 1995). The S-endosymbiont of of Blastopsylla occidentalis, Pachypsylla venusta, and Trioza
Tetraneura radicicola hybridized with the probe to the magnoliae. These authors came to the same conclusion, namely,
b-subdivision but not to the g-subdivision, suggesting that this that the P-endosymbionts of psyllids constitute a distinct lineage
endosymbiont is a member of the former group. Since there is within the g-subdivision of the Proteobacteria (> Fig. 19.1).
only a single nucleotide difference between these two probes, These endosymbionts have an unusual property, namely, the
a confirmation of this conclusion by sequencing the rDNA from lowest known G + C content of any 16S rDNA (36.4 mol%).
this organism seems desirable. The S-endosymbiont from two In addition, this lineage appeared to have a substantial acceler-
other aphid species did not hybridize with either the g- or ation of the rate of evolutionary change within the 16S sequence.
b-subdivision probe. In view of the technical difficulties encoun- A. mori, B. occidentalis, and T. magnoliae all had different
tered with some of the specimens, these conclusions cannot be S-endosymbionts, which belong in the g-subdivision. Only the
interpreted as indicating that the S-endosymbionts of these P-endosymbiont was found in P. venusta (Spaulding and von
aphids belong to different bacterial groups. Dohlen 1998). The studies were limited to few taxa but were
consistent with cospeciation of the P-endosymbiont and the
psyllid host and multiple acquisitions of S-endosymbionts
Absence of a Stable Bacterial Flora in Aphid Guts (Fukatsu and Nikoh 1998; Spaulding and von Dohlen 1998).
Psyllids contain bilobed bacteriomes made up of round
Aphids maintained under clean conditions do not appear to uninucleate bacteriocytes and a multinucleate syncytial region
have a bacterial gut flora (Douglas 1990; Grenier et al. 1994; (Buchner 1965). Many psyllids have endosymbionts in both the
Harada and Ishikawa 1993). Older aphids as well as aphids bacteriocytes and the syncytium; some have endosymbionts
reared under crowded conditions may acquire a gut flora only within the bacteriocytes (Buchner 1965). Using electron
which appears to consist of members of the Enterobacteriaceae microscopy, Chang and Musgrave (1969) and Waku and Endo
(Serratia, Erwinia), Pseudomonas, Staphylococcus, and Bacillus (1987) found that two psyllid species have endosymbionts in the
(Grenier et al. 1994; Harada and Ishikawa 1993). These organ- bacteriocytes which are distinguishable from those found in the
isms are frequently associated with plant surfaces. None of these syncytium and that both endosymbiont types have a Gram-
organisms has a close relationship to Buchnera, precluding negative cell wall. In general, the bacteriocyte-associated endo-
a recent common ancestor from which they and Buchnera are symbionts are more numerous than the syncytium-associated
descended. The presence of this bacterial flora has a deleterious endosymbionts (Buchner 1965). Fukatsu and Nikoh (1998),
effect on aphid performance (Grenier et al. 1994). The actual using an oligonucleotide probe specific for the P- or S-endo-
bacterial numbers have not been established. Harada et al. symbiont, showed by means of in situ hybridization that the
(1996) have isolated 38 bacterial strains from the guts of former was localized in the bacteriocytes while the latter was in
20 aphids. This hardly suggests the presence of an indigenous the syncytium (> Fig. 19.10). There is currently no information
bacterial flora and the numbers are insignificant compared to on the requirement of the endosymbiont(s) by the psyllid host
the numbers of Buchnera or the S-endosymbionts. These studies or on their function. The similarity in diet between psyllids and
indicate that the guts of aphids are generally sterile, but under aphids raises the possibility that psyllid endosymbionts may
certain conditions, a transient bacterial gut flora may be present. provide nutrients as do Buchnera.
Endosymbionts of Other Plant Sap-Utilizing Whitefly Endosymbionts

Insects
16S rDNA sequences have been obtained for endosymbionts
Psyllids, whiteflies and mealybugs are three separate lineages of of Bemisia tabaci, B. argentifolii (previously B. tabaci
the suborder Sternorrhyncha and contain P-endosymbionts B biotype), Siphoninus phillyreae, and Trialeurodes vaporariorum
prokaryote-specific antibiotics affect the growth and develop-

ment of whiteflies, indicating a requirement for the endosymbi-
ont(s) by the host (Costa et al. 1993a, 1997).
Mealybug Endosymbionts
16S rDNA sequences have been obtained for endosymbionts of

Pseudococcus longispinus, P. maritimus, and Dysmicoccus
neobrevipes (Munson et al. 1992), and these organisms were
found to be a distinct lineage within the b-subdivision of the
Proteobacteria (> Fig. 19.1). The morphology of endosymbi-
onts from several mealybug species has been studied by means of
electron microscopy (Tremblay 1989). Within the bacteriocytes
the endosymbionts appear to be embedded in mucous spherules
of unknown composition. There is no information on the func-
tion or the requirement for the endosymbionts by the host.
Tsetse Fly Endosymbionts
Tsetse flies (genus Glossina) are important vectors of trypano-

somes, which are causative agents of African sleeping sickness
and various diseases of animals (Harwood and James 1979).
They have a somewhat unusual reproductive cycle in that the
female gives birth to fully grown larvae. Only one larva is carried
at a time within the uterus. During this stage the larva is fed
nutritive fluids from special glands, commonly known as ‘‘milk
glands.’’ The female requires several blood meals to complete the
development period of each larva, and it is these blood meals
. Fig. 19.10 which result in the transmission of trypanosomes (Harwood
Light micrographs of a bacteriome of the psyllid Anomoneura and James 1979).
mori. (a) In situ hybridization using an oligonucleotide probe Tsetse may be associated with three prokaryotes:
specific for the P-endosymbiont 16S rRNA, which reacts with (1) Wigglesworthia (P-endosymbionts), (2) Sodalis (S-endosym-
bacteriocytes containing these endosymbionts. (b) In situ bionts), and (3) Wolbachia. The last are parasites found in
hybridization using an oligonucleotide probe specific for the 16S reproductive tissue and causing reproductive disorders
rRNA, of the S-endosymbiont of this psyllid, which reacts with the (O’Neill et al. 1997) and will not be considered here.
endosymbiont located in the syncytium. Bar = 20 mm (From
Fukatsu and Nikoh (1998) with permission from the authors and
ASM Press) Wigglesworthia: The Primary Endosymbiont of
Tsetse Flies
Phylogeny
(Clark et al. 1992). The P-endosymbionts of these insects are Based on 16S rDNA, Wigglesworthia was found to constitute
a lineage within the g-subdivision of the Proteobacteria a distinct lineage within the g-3 subgroup of the Proteobacteria
(> Fig. 19.1). B. tabaci and B. argentifolii have an (> Fig. 19.1; Aksoy et al. 1995; Chen et al. 1999). These organ-
S-endosymbiont which is a member of the Enterobacteriaceae. isms are related to but distinct from Buchnera of aphids and the
The P-endosymbionts and the S-endosymbionts from these two P-endosymbionts of carpenter ants (> Fig. 19.1). Using the host
species have identical 16S rDNA sequences, consistent with their rDNA transcribed spacer-2, it was found that the phylogeny of
close relationship (Brown et al. 1995; Clark et al. 1992). the host was the same as that of Wigglesworthia, indicating
The ultrastructure of the endosymbionts of B. tabaci, cospeciation of the host and the endosymbiont (Chen et al.
B. argentifolii, and T. vaporariorum has been studied by electron 1999). These results suggest a single infection of a tsetse ancestor
microcopy (Costa et al. 1993b, 1995). There is evidence for at with a bacterium followed by long-term vertical transmission of
least two morphological types. Whiteflies are unusual in that the endosymbiont, that is, a lack of exchange of Wigglesworthia
they transmit an entire bacteriocyte containing endosymbionts between different tsetse fly species. The age of this
to the egg (Buchner 1965; Costa et al. 1996). Some association has been estimated to be at least 50 million years
. Table 19.8 transcription unit. The presence of one copy of the rRNA operon
Species of tsetse (Glossina) for which the 16S rDNA sequence of is characteristic of slow-growing bacteria and also is found in
Wigglesworthia has been determined several other endosymbionts (Baumann et al. 1995).
G. austeni
G. brevipalpis
Sodalis: The Secondary Endosymbiont
G. fuscipes of Tsetse Flies
G. morsitans centralis
G. m. morsitans Tsetse flies may also contain S-endosymbionts. These are pri-
G. palpalis gambiensis marily found within midgut cells but also have been detected in
G. p. palpalis hemolymph and in a variety of other tissues excluding ovaries
G. tachinoides (Aksoy et al. 1997; Beard et al. 1993b; Cheng and Aksoy 1999).
Their numbers are age-dependent, being higher in older insects
Chen et al. (1999)
(Cheng and Aksoy 1999). The 16S rDNA has been sequenced
from the S-endosymbionts of five different tsetse fly species, and
it was found that they are members of the Enterobacteriaceae
(Aksoy et al. 1997). A list of the species of tsetse flies for which (Aksoy et al. 1997; Beard et al. 1993b). The sequences were found
the 16S rDNA of Wigglesworthia has been sequenced is presented to be virtually identical, indicating either multiple recent infec-
in > Table 19.8. tions with the same organism or horizontal transmission of the
S-endosymbiont. The S-endosymbiont is maternally transmit-
Taxonomy ted via the ‘‘milk gland’’ secretions to developing larvae (Aksoy
The genus Wigglesworthia contains one species, W. glossinidia, et al. 1997).
which designates the lineage consisting of the P-endosymbionts The S-endosymbionts have been cultivated in cell-free liquid
of tsetse flies (Aksoy 1995b). The type strain of this species is the media (Beard et al. 1993b) and recently on solid media (Dale and
P-endosymbiont of G. morsitans morsitans. Maudlin 1999). The latter allowed a phenotypic characterization
of this organism and led to its assignment into a new genus and
Habitat species, Sodalis glossinidius (Dale and Maudlin 1999). This spe-
Tsetse flies contain a U-shaped bacteriome located in the ante- cies consists of Gram-negative rods 1–1.5 m in diameter and
rior region of the gut, which is made up of bacteriocytes 2–12 m in length. It is microaerophilic, lacking catalase, and has
containing Wigglesworthia (Aksoy 1995b; Aksoy et al. 1995). a relatively limited capacity for carbohydrate utilization.
These endosymbionts have a Gram-negative cell wall and are The S-endosymbiont has seven copies of 16S rDNA,
somewhat pleomorphic, occurring mostly as 4–5-m-long rods. a number which is similar to that found in rapidly growing
They are found free (not enclosed within host-derived vesicles) free-living organisms (Aksoy 1995a). Plasmids of 80 kb and
in the bacteriocyte cytoplasm. Wigglesworthia is maternally about 130 kb have been detected in these organisms (Beard
transmitted. Since neither the milk gland nor the developing et al. 1993b). The S-endosymbiont has been transformed with
eggs contain Wigglesworthia, the mechanism of their transmis- pSUP204, and plasmid-encoded resistance to ampicillin, tetra-
sion is not known (Aksoy et al. 1997). cycline, and chloramphenicol was expressed (Beard et al. 1993b).
Similarly the S-endosymbiont has been transformed with
Physiology a pSUP204 derivative, which expressed the green fluorescent
The feeding of tsetse flies on animals immunized with protein, allowing ready visualization of this organism in insect
Wigglesworthia results in elimination of the P-endosymbiont tissues (Cheng and Aksoy 1999).
and sterility of the flies (Nogge 1976). A similar effect is observed In one case, the S-endosymbiont from one tsetse fly species
upon treatment of tsetse with prokaryote-specific antibiotics when microinjected into another species became pathogenic,
(Aksoy et al. 1995; Nogge 1976, 1982). These results indicate killing the flies within 48 h (Cheng and Aksoy 1999). This result
that the P-endosymbiont is essential for reproduction. There is is similar to the observations made with the S-endosymbiont of
evidence that one of the functions of the Wigglesworthia is the aphids (Chen and Purcell 1997). The variation in the number of
production of B-complex vitamins (Nogge 1982). S-endosymbionts and their possible absence from some insects
Wigglesworthia produces a high level of GroEL (Aksoy 1995a). suggest that they do not perform a function essential for the
In this respect, it is similar to a number of other endosymbionts survival of tsetse flies.
as well as other intracellular organisms (Hogenhout et al. 1998).
Genetics Sitophilus (Weevils) Endosymbionts

The Wigglesworthia genome has one copy of the 16S rRNA gene
(Aksoy 1995a). In this organism, 16S rRNA gene is directly Weevils of the genus Sitophilus are major pests of stored grain
upstream of 23S rRNA gene, suggesting that, as in the case of (Borror et al. 1989). The female bores a hole in kernels and
many other bacteria, these genes are a part of a single deposits an egg. The larva develops inside the grain from
which the young adults emerge. Three related species have been Buchnera and Wigglesworthia, the endosymbionts of S. oryzae
studied with respect to their endosymbionts, S. oryzae, S. overproduce GroEL (Charles et al. 1997b). Unlike Buchnera, the
granarius, and S. zeamais (Dasch et al. 1984). Of these three endosymbionts of S. oryzae have a heat shock response, as is
species, the most extensive studies deal with S. oryzae (Nardon indicated by an increase of groEL mRNA (Charles et al. 1997b).
and Grenier 1988). In addition, weevils may harbor the patho-
gen, Wolbachia (O’Neill et al. 1997).
Genetics
Phylogeny The genome of the endosymbiont of S. oryzae is 3.0 Mb and has

two copies of the rRNA operon (Charles et al. 1997a). In addi-
Early studies of Sitophilus-endosymbiont morphology and the tion, the endosymbiont contains a plasmid of about 138 kb. This
G + C content of its DNA suggested that weevils have different genome size puts the endosymbiont within the range of many
endosymbionts (Dasch 1975; Dasch et al. 1984; Grinyer and free-living bacteria (Heddi et al. 1998).
Musgrave 1966; Musgrave and Grinyer 1968). S. oryzae has one
endosymbiont with a G + C content of about 54 mol%,
S. granarius has one with a G + C content of 50 mol%, while Comparisons with Other Associations
S. zeamais appears to have both endosymbionts (Dasch et al.
1984; Heddi et al. 1998). One type of endosymbiont 16S rDNA Although the information is somewhat limited, comparison of
sequence was detected in S. oryzae, and two types were detected the Sitophilus-endosymbiont association with other insect-
in S. zeamais (Campbell et al. 1992; Heddi et al. 1998). Phylo- endosymbiont associations suggests that the former has some
genetic analysis indicated that all of these endosymbionts are unique features. In several insect endosymbiotic associations,
members of the family Enterobacteriaceae (> Fig. 19.1). the S-endosymbionts are members of the Enterobacteriaceae,
suggesting that organisms within this lineage have an enhanced
capacity to enter into such associations. An interesting feature of
Habitat the Sitophilus association is that these organisms are the sole
endosymbionts. The morphological diversity of the endosymbi-
In larvae, endosymbionts are present in bacteriocytes which onts, the large endosymbiont genome size, and the fact that
make up a bacteriome located at the junction of the foregut aposymbiotic weevils are viable suggest that the associations
and the midgut as well as in the rudimentary ovaries (Charles arose through multiple recent infections and that major adap-
et al. 1995; Nardon and Grenier 1988). The endosymbionts are tations resulting in obligatory mutual interdependence of both
transmitted via the eggs. S. oryzae endosymbionts are rod- partners have not evolved as yet. This association may conse-
shaped, 5–15 m long, and free (not within host-derived vesicles) quently be an example of an endosymbiosis at an early stage of
in the cytoplasm (Dasch et al. 1984; Nardon and Grenier 1988). its development.
Physiology Carpenter Ant Endosymbionts
S. oryzae may be cured of endosymbionts by treatment with heat Ants feed on complex diets, and the presence of endosymbionts
or antibiotics (Baker and Lum 1973; Nardon and Grenier 1988). has been reported in only two groups. These two groups are the
Such aposymbiotic weevils are softer and paler, have an genus Formica and the genus Camponotus, commonly known as
increased development time, and the fertility of their eggs is the carpenter ants (Borror et al. 1989; Buchner 1965; Dasch et al.
reduced (Nardon and Grenier 1988). On some nutrient-rich 1984). Both groups can use a broad range of food types but
grains, they can grow indefinitely. Aposymbiotic weevils lack typically utilize plant nectar and honeydew (the liquid feces of
bacteriomes, indicating that the endosymbiont triggers their sap-feeding Homoptera) as major components of their diet.
development. Naturally occurring aposymbiotic weevils also Only in the carpenter ants have symbionts been studied using
may be found. There is evidence that one of the functions of modern methods, and we focus on these. The contribution of
the endosymbiont is the synthesis of vitamins as well as possibly ant endosymbionts to host nutrition is not clear (Dasch et al.
phenylalanine or tyrosine (Baker 1975, 1979; Wicker and 1984).
Nardon 1982). An additional function is the conversion of excess
methionine in the diet to methionine sulfoxide (Gasnier-
Fauchet and Nardon 1986). Aposymbiotic weevils also have Phylogeny
mitochondria with reduced levels of enzymes involved in respi-
ration (Heddi et al. 1991). Isolated endosymbionts of S. oryzae The G + C content of the DNA of carpenter ants is 30–32 mol%
do not consume oxygen and lack a number of enzymes of (Dasch 1975; Dasch et al. 1984). The sequence of the 16S rDNA
respiratory metabolism (Heddi et al. 1991, 1993). These results has been determined for endosymbionts of the species
suggest that they have an anaerobic metabolism. As in the case of Camponotus floridanus, C. rufipes, C. ligniperdus, and
C. herculeanus (Schröder et al. 1996). A phylogenetic analysis

indicated that the endosymbionts constitute a distinct, mono-
phyletic group related to, but different from, the endosymbionts
of aphids, tsetse flies, and the members of the Enterobacteriaceae
(> Fig. 19.1). The order of branching reflects the relationships
between the carpenter ant species and is consistent with a single
infection and subsequent vertical evolution of the endosymbi-
onts. The age of the association is estimated at over 100 million
years (Schröder et al. 1996). In endosymbionts of carpenter ants,
the 16S rRNA gene is not directly upstream of the 23S rRNA
gene, suggesting that these genes are organized in two transcrip-
tion units, as is the case in Buchnera (C. Sauer and R. Gross,
personal communication).
Habitat
In both workers and queens of Camponotus, the endosymbionts

are located in bacteriocytes, which are intercalated between
epithelial cells of the midgut (> Fig. 19.11a; Buchner 1965;
Schröder et al. 1996). The endosymbionts are rods of 1 mm in
width to 5–15 mm in length (> Fig. 19.11b). They have a Gram-
negative type cell wall and are free (not enclosed in host-derived
vesicles) in the cytoplasm (Schröder et al. 1996). Transmission is
via infection of the ovaries and incorporation into the eggs.
Blattabacterium-Endosymbionts of Cockroaches
and Termites
Cockroaches (order Blattaria) utilize a complex diet and harbor

prokaryotic endosymbionts (Dasch et al. 1984). It has been
. Fig. 19.11
hypothesized that cockroaches and termites (order Isoptera)
Electron micrographs of the endosymbionts of the carpenter ant
are phylogenetically related (Kambhampati 1995). Common
Camponotus floridanus. (a) Bacteriocyte containing the
ancestry is suggested from the fact that the wood-eating cock-
endosymbionts, bar = 3.0 mm; (b) ultrastructure of the
roach, Cryptocercus punctulatus, has a cellulose-digesting proto-
endosymbionts showing the Gram-negative cell wall and the
zoal gut flora which is similar to that of termites (reviewed in
absence of a vesicular membrane, bar = 0.3 mm (Photos courtesy
Bandi and Sacchi 1999). In addition, Mastotermes darwiniensis,
of C. Sauer and R. Gross)
a primitive termite, lays eggs in rows resembling those made by
cockroaches (Borror et al. 1989; Sacchi et al. 1998b).
Blattabacterium, suggesting that, with the exception of this ter-
mite, the endosymbionts were eliminated in the lineage leading
Phylogeny to the present termite species (Bandi and Sacchi 1999; Bandi
et al. 1997). The 16S rDNA sequence has been determined
The G + C content of the DNA of Blattabacterium is 26–28 mol% for endosymbionts of the cockroach species Periplaneta
(Dasch 1975; Dasch et al. 1984). Phylogenetic analysis of the 16S australasiae, P. americana, Blatella germanica, Pycnoscelus
rDNA from Blattabacterium of cockroaches and the termite M. surinamensis, Nauphoeta cinerea, and C. punctulatus, as well as
darwiniensis indicates that the endosymbionts form a distinct the termite species M. darwiniensis (Bandi et al. 1994, 1995). The
lineage within the Flavobacterium-Bacteroides group of bacteria association between Blattabacterium and termites is estimated to
(Bandi et al. 1994, 1995; > Fig. 19.1). The phylogenetic tree be 135–300 million years old (Bandi et al. 1995).
obtained on the basis of endosymbiont 16S rDNA is the same
as that derived from host taxonomy. This result is consistent
with a single infection in an ancestor of cockroaches and ter- Taxonomy
mites and vertical evolution of the endosymbionts, that is,
a lack of endosymbiont exchange among different species. The genus Blattabacterium contains one species, B. cuenoti,
M. darwiniensis is the only termite known to harbor and currently designates the lineage consisting of the
bacteriocyte-associated endosymbionts of cockroaches and one

termite (Bandi et al. 1995; Dasch et al. 1984). The type strain is
the endosymbiont of Blatta orientalis (Dasch et al. 1984).
Habitat
Blattabacterium is found in bacteriocytes, polyploid cells which

are found within the abdominal fat bodies of cockroaches and
the termite M. darwiniensis (> Fig. 19.12a, Bigliardi et al. 1995;
Sacchi et al. 1996, 1998a, b). The endosymbionts are rods of
1 mm in width and 1.6–9 mm in length (> Fig. 19.12b; Dasch et al.
1984). They have a Gram-negative cell wall and are located
within vesicles derived from the host cell. The bacteriocytes
originate from plasmatocytes, which are phagocytic cells present
in the hemolymph. There are extensive electron microscopic
studies that document the infection and differentiation of
plasmatocytes into bacteriocytes and the transmission of the
endosymbionts to oocytes and eggs (Bigliardi et al. 1995; Sacchi
et al. 1996, 1998a, b).
Physiology
Aposymbiotic cockroaches can be obtained by rearing the

insects on antibiotic-containing foods (Bandi and Sacchi 1999;
Dasch et al. 1984). The resulting insects can be propagated on
enriched diets. The aposymbiotic insects are smaller in size,
light in color, and have a reduced fecundity and an increased
development time. They also have an increased level of uric
acid in fat bodies, suggesting that one function of
Blattabacterium is nitrogen recycling (Cochoran 1985). The
proximity of bacteriocytes and urate cells (which store uric
acid) in the fat body and the presence of adhesion sites between
their plasma membranes suggest direct metabolic interactions
between these cells (Sacchi et al. 1998a). There is also evidence
that the bacteria provide essential amino acids for the host
(Henry 1962).
Isolation . Fig. 19.12

Electron micrographs of Blattabacterium cuenoti, the
P-endosymbionts have been isolated from aphids and also from endosymbiont of cockroaches and termites. (a) Endosymbiont
Sitophilus. In the case of aphids, the starting material is usually within bacteriocytes of the cockroach Periplaneta americana,
whole insects; in the case of Sitophilus, the starting material is bar = 3.0 mm (Photo courtesy of L. Sacchi); (b) ultrastructure of the
dissected bacteriomes. Since the two methods are similar, only endosymbiont of Cryptocercus punctulatus showing the Gram-
the method for the isolation of Buchnera from aphids will be negative cell wall (large arrow) and the vesicular membrane (small
considered. Details of the isolation of the endosymbionts from arrow), bar = 0.3 mm (From Sacchi et al. (1998) with permission of
Sitophilus are described by Heddi et al. (1991). the authors and Balaban Publishers)
The methods for the isolation of Buchnera have been devel-
oped by Ishikawa (1982), Sasaki and Ishikawa (1995), and
Charles and Ishikawa (1999). The resulting preparations are the endosymbionts are still within host-derived vesicles. Both
suitable for isolation of high molecular weight DNA for genome the aphids and the endosymbionts have a similar mol% G + C in
analysis as well as for physiological studies. The best criteria of their DNA (Ishikawa 1987; Unterman and Baumann 1990).
purity have involved examination of the preparations by elec- Consequently, endosymbiont DNA cannot be separated from
tron microscopy, which also allows determination of whether host DNA by CsCl density gradient centrifugation.
As starting material, it is best to use an aphid strain that has morphologically distinct endosymbionts (Unterman et al.
only Buchnera and lacks S-endosymbionts. All of the reagents 1989). Using total aphid DNA, a probe to E. coli 16S rDNA,
and equipment are kept on ice, and the procedures are and restriction enzyme and Southern blot analysis, only two
performed as rapidly as possible. Approximately 2–3 g (wet restriction maps could be constructed corresponding to the
weight) aphids are transferred to a 1.5-cm-diameter tissue two endosymbionts. There were differences in the intensities of
grinder. Ten milliliters of buffer A of Ishikawa (1982) is added, the bands, indicating that one of the endosymbionts was present
and the aphids are ground with a loose fitting plunger for 5 min. in larger numbers than the other. In addition, the results were
[Buffer A contains 0.25 M sucrose, 35 mM Tris–HCl (pH 7.6), consistent with the presence of only one copy of the 16S rRNA
25 mM KCl, 10 mM MgCl2, and 1 mM dithiothreitol.] The gene per endosymbiont genome. Based on the restriction maps,
preparation is then passed through a double layer of a nylon DNA fragments containing 16S rRNA genes of both endosym-
mesh to remove large particulate material. The filtrate is brought bionts were cloned and sequenced. In addition, DNA was iso-
to a volume of about 100 mL with buffer A and then quickly lated from dissected bacteriomes, and restriction enzyme and
passed through a 100-mm nylon filter followed by filtration Southern blot analysis indicated the association of both endo-
through 20- and 10-mm nylon filters (in some cases filtration symbionts with the bacteriome. Subsequently, restriction
through 5- and 3-mm filters is also performed). Only slight enzyme and Southern blot analysis using DNA from other
vacuum pressure is applied during the last two filtration steps. aphid species indicated the presence of only one copy of the
The suspension is centrifuged in a swinging bucket rotor for 16S rRNA gene, a finding consistent with only one or a single
6–10 min at 1,500 g. The pellets are gently resuspended in predominant endosymbiont (Munson et al. 1991b). Upon
1–2 mL buffer A and centrifuged through a Percoll gradient amplification of the 16S rDNA by PCR, the sequences detected
(12,000 g, 15 min). The gradient consists of 27–70 % Percoll in were all related to Buchnera. This approach is suitable for
buffer A, 5 % PEG 6000, 1 % Ficoll, and 1 % bovine serum the study of insects in which there is one predominant
albumin (Pharmacia Biotech, Uppsala, Sweden; Charles and P-endosymbiont and does not exclude the possibility of not
Ishikawa 1999). Mitochondria are in the upper phase, host detecting S-endosymbionts that are present in lower numbers.
nuclei are in the pellet, and Buchnera appears as a green band. Instead of initial studies involving restriction enzyme and
Southern blot analysis, it is much more convenient to use oli-
gonucleotides complementary to the front and back of 16S
Identification rDNA and PCR to amplify DNA fragments for cloning, sequenc-
ing, or both. Localization of the endosymbiont may be
Currently none of the P-endosymbionts has been cultured, and performed by dissection of different tissues, purification of the
consequently identification is based primarily on sequence anal- DNA, and amplification of the 16S rDNA by PCR (Aksoy et al.
ysis of their 16S rDNAs. Since these studies are just beginning, 1995; Bandi et al. 1994, 1995; Schröder et al. 1996). Differences
we will briefly discuss the methods used for both their charac- in the types of rDNA amplified may be established by restriction
terization and identification. The success of the approach used fragment polymorphism (Clark et al. 1992; Fukatsu and Nikoh
in most of the studies has been dependent on the use of fresh or 1998). Once a pattern of relationship is established, this method
frozen insect samples which meet the following criteria: (1) the may be applied using whole insect DNA. It should be noted that
predominant bacterial flora of the insect consists of one or a few there are potential problems associated with PCR such as selec-
endosymbiont types, (2) there is no significant gut flora, and tive amplification of some sequences and hybrid formation,
(3) the samples are relatively clean. which are reviewed by von Wintzingerode et al. (1997).
An ideal study would utilize the full-cycle rRNA analysis One additional approach to the detection of different kinds
formulated by Amann et al. (1995). In this approach, the 16S of organisms in the DNA samples is to use oligonucleotide
rDNA is amplified by PCR and sequenced. Based on compari- primers that are complementary to the front of the 16S rDNA
sons with 16S rDNA(s) in databases, the organism(s) is identi- and the middle of the 23S rDNA. In bacteria in which the order
fied. Specific oligonucleotide probes are designed and used in in of genes is 16S–23S (most species), there is a spacer region
situ hybridization to identify the endosymbiont associated with between the genes the length of which is labile and usually differs
the sequence. This approach is extremely useful when more than among species. If, after PCR, several bands are observed, they
one endosymbiont is present and, in the case of insects, has been probably correspond to different bacteria. This approach has
applied to the identification of two types of psyllid endosymbi- been applied to the cloning of P-endosymbionts of psyllids and
onts (Fukatsu and Nikoh 1998). to the putative S-endosymbionts which differ in the size of their
In studies in which the primary goal was a phylogenetic 16S–23S spacer (Thao, M. L. and Baumann, P., unpublished
characterization of the P-endosymbiont, one approach used is observations).
an initial thorough study of one or more representative insects Specific oligonucleotide primers also can be made comple-
that can be cultivated or are readily available. Then characteri- mentary to unique sequences of the endosymbiont 16S and 23S
zation is extended to taxa obtained as field collections and rDNA. This has been done for Buchnera 16S rDNA
preserved in dry ice or absolute ethanol. For example, in the (Rouhbakhsh et al. 1994). Another approach is to take advan-
initial study of aphid endosymbionts, the insect specimen cho- tage of unique linkage relationships. Buchnera-specific probes
sen (Ap) was known from electron microscopy to harbor two have been made which span the region argS-16SrDNA and
aroE-23SrDNA (Munson et al. 1993; Rouhbakhsh and Baumann Bandi C, Anderson TJC, Genchi C, Blaxter ML (1998) Phylogeny of Wolbachia in
filarial nematodes. Proc R Soc Lond B Biol Sci 265:2407–2414
1995; Rouhbakhsh et al. 1994). The latter is especially useful,
Bandi C, McCall JW, Genchi C, Corona S, Venco L, Sacchi L (1999) Effects of
since most bacteria have 16S rDNA directly upstream of 23S tetracycline on the filarial worms Brugia pahangi and Dirofilaria immitis and
rDNA. their bacterial endosymbionts Wolbachia. Int J Parasitol 29:357–364
Bandi C, Sacchi L. Intracellular symbiosis. In: Abe T, Higashi M, Bignel D (eds)
Termites: their symbiosis, behavior, and global diversification. Kluwer, Dor-
Application drecht (in press)
Barancin CE, Smoot JC, Findlay RH, Actis LA (1998) Plasmid-mediated hista-
Aphids, psyllids, whiteflies, and mealybugs, which utilize plant mine biosynthesis in the bacterial fish pathogen Vibrio anguillarum. Plasmid
39:235–244
sap as food, are of major economic importance in that they may
Baumann L, Baumann P (1994) Growth kinetics of the endosymbiont Buchnera
cause plant debilitation and the transmission of a variety of plant aphidicola in the aphid Schizaphis graminum. Appl Environ Microbiol
pathogens (Gray and Banerjee 1999; Sylvester 1985). Tsetse flies, 60:3440–3443
which suck blood, are important in the transmission of human Baumann L, Baumann P (1998) Characterization of ftsZ, the cell division gene of
and animal disease, especially in tropical regions of Africa Buchnera aphidicola (endosymbiont of aphids) and detection of the product.
Curr Microbiol 36:85–89
(Harwood and James 1979). Since these organisms are depen-
Baumann P, Moran NA (1997) Non-cultivable microorganisms from symbiotic
dent on their P-endosymbionts for survival, an understanding of associations of insects and other hosts. Antonie Van Leeuwenhoek 72:39–48
the genetics and physiology of the endosymbionts may be of use Baumann P, Baumann L, Lai CY, Roubakhsh D, Moran NA, Clark MA
in devising strategies, which are based on interference with their (1995) Genetics, physiology, and evolutionary relationships of the genus
endosymbionts, for the control of these insect pests. An addi- Buchnera: intracellular symbionts of aphids. Annu Rev Microbiol 49:55–94
Baumann P, Baumann L, Clark MA (1996) Levels of Buchnera aphidicola
tional approach, which is currently being developed, is the
chaperonin GroEL during growth of the aphid Schizaphis graminum. Curr
potential use of endosymbionts to produce compounds that Microbiol 32:279–285
would result in modification of the vector competence of an Baumann P, Moran NA, Baumann L (1997a) The evolution and genetics of aphid
insect. Such an insect would be unable to harbor and/or transmit endosymbionts. Bioscience 47:12–20
the disease agent. This approach has been discussed by Beard Baumann L, Clark MA, Rouhbakhsh D, Baumann P, Moran NA, Voegtlin DJ
(1997b) Endosymbionts (Buchnera) of the aphid Uroleucon sonchi contain
et al. (1993a) and Durvasula et al. (1997).
plasmids with trpEG and remnants of trpE pseudogenes. Curr Microbiol
35:18–21
Baumann L, Baumann P, Moran NA (1998a) The endosymbiont (Buchnera) of
References the aphid Diuraphis noxia contains all the genes of the tryptophan biosyn-
thetic pathway. Curr Microbiol 37:58–59
Akhtar S, van Emden HF (1994) Ultrastructure of the symbionts and mycetocytes Baumann P, Baumann L, Clark MA, Thao ML (1998b) Buchnera aphidicola: the
of bird cherry aphid (Rhopalosiphum padi). Tissue Cell 26:513–522 endosymbiont of aphids. ASM News 64:203–209
Aksoy S (1995a) Molecular analysis of the endosymbionts of tsetse flies: 16S Baumann L, Baumann P, Thao ML (1999a) Detection of messenger RNA tran-
rDNA locus and over-expression of a chaperonin. Insect Mol Biol 4:23–29 scribed from genes encoding enzymes of amino acid biosynthesis in
Aksoy S (1995b) Wigglesworthia gen. nov. and Wigglesworthia glossinidia sp. nov., Buchnera aphidicola (endosymbiont of aphids). Curr Microbiol 38:135–136
taxa consisting of the mycetocyte-associated, primary endosymbionts of Baumann L, Baumann P, Moran MA, Sandström J, Thao ML (1999b) Genetic
tsetse flies. Int J Syst Bacteriol 45:848–851 characterization of plasmids containing genes encoding enzymes of leucine
Aksoy S, Pourhosseini AA, Chow A (1995) Mycetome endosymbionts of tsetse biosynthesis in endosymbionts (Buchnera) of aphids. J Mol Evol 48:77–85
flies constitute a distinct lineage related to the Enterobacteriaceae. Insect Mol Beard CB, O’Neill SL, Tesh RB, Richards FF, Aksoy S (1993a) Modification
Biol 4:15–22 of arthropod vector competence via symbiotic bacteria. Parasitol Today
Aksoy S, Chen X, Hypsa V (1997) Phylogeny and potential transmission routes of 9:179–183
midgut-associated endosymbionts of tsetse (Diptera: Glossinidae). Insect Beard CB, O’Neill SK, Mason P, Mandelco L, Woese CR, Tesh RB, Richards FF,
Mol Biol 6:183–190 Aksoy S (1993b) Genetic transformation and phylogeny of bacterial symbi-
Amann RI, Ludwig W, Schleifer KH (1995) Phylogenetic identification and in situ onts from tsetse. Insect Mol Biol 1:123–131
detection of individual microbial cells without cultivation. Microbiol Rev Bensaadi-Merchermek N, Salvado JC, Cagnon C, Karama S, Mouches C (1995)
59:143–169 Characterization of the unlinked 16S rDNA and 23S-5S rRNA operon
Andersson SGE, Zomorodipour A, Andersson JO, Sicheritz-Ponten T, Alsmark of Wolbachia pipientis, a prokaryotic parasite of insect gonads. Gene
UCM, Podowski RM, Naslund AK, Eriksson AS, Winkler HH, Kurland CG 165:81–86
(1998) The genome sequence of Rickettsia prowazekii and the origin of Berlyn MKB (1998) Linkage map of Escherichia coli K-12, edition 10: the
mitochondria. Nature 396:133–140 traditional map. Microbiol Mol Biol Rev 62:814–984
Baker JE (1975) Vitamin requirements of larvae of Sitophilus oryzae. J Insect Bigliardi E, Selmi MG, Corona S, Bandi CA, Sacchi L (1995) Membrane systems
Physiol 21:1337–1342 in endocytobiosis. III. Ultrastructural features of symbionts and vacuolar
Baker JE (1979) Requirements for the essential dietary amino acids of larvae of the membrane in bacteriocytes of the wood-eating cockroach Cryptocercus
rice weevil. Environ Entomol 8:451–453 punctulatus (Dictyoptera, Cryptocercidae). Boll Zool 62:235–238
Baker JE, Lum PTM (1973) Development of aposymbiosis in larvae of Sitophilus Blackman RL (1984) Reproduction, cytogenetics and development. In: Minks
oryzae by dietary treatment with antibiotics. J Stored Prod Res 9:241–245 AK, Harrewijn P (eds) Aphids: their biology, natural enemies and control,
Bandi C, Damiani G, Magrassi L, Grigolo A, Fani R, Sacchi L (1994) Flavobacteria as vol 2A. Elsevier Biomedical Press, Amsterdam, pp 163–195
intracellular symbionts in cockroaches. Proc R Soc Lond B Biol Sci 257:43–48 Blackman RL, Eastop VF (1984) Aphids on the world’s crops. Wiley, Chichester,
Bandi C, Sironi M, Damiani G, Magrassi L, Nalepa CA, Laudani U, Sacchi L UK
(1995) The establishment of intracellular symbiosis in an ancestor of cock- Blattner FR, Plunkett G, Bloch CA, Perna NT, Burland V, Riley M, ColladoVides J,
roaches and termites. Proc R Soc Lond B Biol Sci 259:293–299 Glasner JD, Rode CK, Mayhew GF, Gregor J, Davis NW, Kirkpatrick HA,
Bandi C, Sironi M, Nalepa CA, Corona S, Sacchi L (1997) Phylogenetically distant Goeden MA, Rose DJ, Mau B, Shao Y (1997) The complete genome sequence
intracellular symbionts in termites. Parassitologia 39:71–75 of Escherichia coli K-12. Science 277:1453–1462
Blua MJ, Perring TM, Madore MA (1994) Plant virus-induced changes in aphid Clark MA, Baumann L, Baumann P (1998b) Buchnera aphidicola (aphid endo-
population development and temporal fluctuations in plant nutrients. symbiont) contains genes encoding enzymes of histidine biosynthesis. Curr
J Chem Ecol 20:691–707 Microbiol 37:356–358
Boman HG, Hultmark D (1987) Cell-free immunity in insects. Annu Rev Clark MA, Baumann P, Moran MA (1999a) Buchnera plasmid-associated trpEG
Microbiol 41:103–126 probably originated from a chromosomal location between hslU and fpr.
Borror DJ, Triplehorn CA, Johnson NF (1989) An introduction to the study of Curr Microbiol 38:309–311
insects. Harcourt Brace College, Fort Worth Clark MA, Moran NA, Baumann P (1999b) Sequence evolution in bacterial
Bracho AM, Martı́nez-Torres D, Moya A, Latorre A (1995) Discovery and molec- endosymbionts having extreme base compositions. Mol Biol Evol 16:1586–
ular characterization of a plasmid localized in Buchnera sp., bacterial endo- 1598
symbiont of the aphid Rhopalosiphum padi. J Mol Evol 41:67–73 Clark MA, Moran NA, Baumann P (2000) Cospeciation between bacterial endo-
Brenner DJ (1984) Enterobacteriaceae. In: Krieg NR, Holt JG (eds) Bergey’s symbionts (Buchnera) and a recent radiation of aphids (Uroleucon) and
manual of systematic bacteriology, vol 1. Williams and Wilkins, Baltimore, pitfalls of testing for phylogenetic congruence. Evolution 54(2):517–525
pp 408–506 Cochoran DG (1985) Nitrogen excretion in cockroaches. Annu Rev Entomol
Brough CN, Dixon AFG (1990) Ultrastructural features of egg development in 30:29–49
oviparae of vetch aphid, Megoura viciae. Tissue Cell 22:51–63 Costa HS, Ullman DE, Johnson MW, Tabashnik BE (1993a) Antibiotic oxytetra-
Brown JK, Frohlich DR, Rosell RC (1995) The sweetpotato or silverleaf white- cycline interferes with Bemisia tabaci (Homoptera, Aleyrodidae) oviposi-
flies—biotypes of Bemisia tabaci or a species complex. Annu Rev Entomol tion, development, and ability to induce squash silverleaf. Ann Entomol Soc
40:511–534 Am 86:740–748
Brynnel EU, Kurland CG, Moran NA, Andersson SGE (1998) Evolutionary rates Costa HS, Westcot DM, Ullman DE, Johnson MW (1993b) Ultrastructure of the
for tuf genes in endosymbionts of aphids. Mol Biol Evol 15:574–582 endosymbionts of the whitefly, Bemisia tabaci and Trialeurodes
Buchner P (1965) Endosymbiosis of animals with plant microorganisms. vaporariorum. Protoplasma 176:106–115
Interscience, New York Costa HS, Westcot DM, Ullman DE, Rosell R, Brown JK, Johnson MW
Campbell BC, Bragg TS, Turner CE (1992) Phylogeny of symbiotic bacteria of (1995) Morphological variation in Bemisia endosymbionts. Protoplasma
four weevil species (Coleoptera, Curculionidae) based on analysis of 16S 189:194–202
ribosomal DNA. Insect Biochem Mol Biol 22:415–421 Costa HS, Toscano NC, Henneberry TJ (1996) Mycetocyte inclusion in the
Campbell BC, Stefen-Campbell JD, Gill RJ (1994) Evolutionary origin of white- oocytes of Bemisia argentifolii (Homoptera, Aleyrodidae). Ann Entomol
flies (Hemiptera: Sternorrhyncha: Aleyrodidae) inferred from 18S rDNA Soc Am 89:694–699
sequences. Insect Mol Biol 3:73–88 Costa HS, Henneberry TJ, Toscano NC (1997) Effects of antibacterial materials on
Chang KP, Musgrave AJ (1969) Histochemistry and ultrastructure of the Bemisia argentifolii (Homoptera: Aleyrodidae) oviposition, growth, survival,
mycetome and its ‘‘symbiotes’’ in the pear psylla, Psylla pyricola Foerster and sex ratio. J Econ Entomol 90:333–339
(Homoptera). Tissue Cell 1:597–606 Crawford IP (1989) Evolution of a biosynthetic pathway—the tryptophan para-
Charles H, Ishikawa H (1999) Physical and genetic map of the genome of digm. Annu Rev Microbiol 43:567–600
Buchnera, the primary endosymbiont of the pea aphid Acyrthosiphon Dadd RH (1985) Nutrition: organisms. In: Kerkut GA, Gilbert LI (eds) Compre-
pisum. J Mol Evol 48:142–150 hensive insect physiology, biochemistry, and pharmacology, vol 4.
Charles H, Ishikawa H, Nardon P (1995) Presence of a protein specific of Pergamon, Elmsford, pp 315–319
endocytobiosis (symbionin) in the weevil Sitophilus. C R Acad Sci Paris Dale C, Maudlin I (1999) Sodalis gen. nov. and Sodalis glossinidius sp. nov.,
Ser III 318:35–41 a microaerophilic secondary endosymbiont of the tsetse fly Glossina
Charles H, Condemine G, Nardon C, Nardon P (1997a) Genome size character- morsitans morsitans. Int J Syst Bacteriol 49:267–275
ization of the principal endocellular symbiotic bacteria of the weevil Dasch GA (1975) Morphological and molecular studies on intracellular bacterial
Sitophilus oryzae, using pulsed field gel electrophoresis. Insect Biochem symbiotes of insects. Yale University, New Haven
Mol Biol 27:345–350 Dasch GA, Weiss E, Chang K-P (1984) Endosymbionts of insects. In: Krieg NR,
Charles H, Heddi A, Guillaud J, Nardon C, Nardon P (1997b) A molecular aspect Holt JG (eds) Bergey’s manual of systematic bacteriology, vol 1. Williams and
of symbiotic interactions between the weevil Sitophilus oryzae and its endo- Wilkins, Baltimore, pp 811–833
symbiotic bacteria: over-expression of a chaperonin. Biochem Biophys Res Denk D, Böck A (1987) L-Cysteine biosynthesis in Escherichia coli: nucleotide
Commun 239:769–774 sequence and expression of the serine acetyltransferase (cysE) gene from
Chen D-Q, Purcell AH (1997) Occurrence and transmission of facultative endo- wild-type and a cysteine-excreting mutant. J Gen Microbiol 133:515–525
symbionts in aphids. Curr Microbiol 34:220–225 Dixon AFG (1973) Biology of aphids. Edward Arnold, London
Chen DQ, Campbell BC, Purcell AH (1996) A new rickettsia from a herbivorous Dixon AFG (1992) Constraints on the rate of parthenogenetic reproduction and
insect, the pea aphid Acyrthosiphon pisum (Harris). Curr Microbiol pest status of aphid. Invertebr Reprod Dev 22:159–163
33:123–128 Douglas AE (1988) Sulfate utilization in an aphid symbiosis. Insect Biochem
Chen XA, Li S, Aksoy S (1999) Concordant evolution of a symbiont with its host 18:159–163
insect species: molecular phylogeny of genus Glossina and its bacteriome- Douglas AE (1989) Mycetocyte symbiosis in insects. Biol Rev Camb Philos Soc
associated endosymbiont, Wigglesworthia glossinidia. J Mol Evol 48:49–58 64:409–434
Cheng Q, Aksoy S (1999) Tissue tropism, transmission and expression of Douglas AE (1990) Nutritional interactions between Myzus persicae and its
foreign genes in vivo in midgut symbionts of tsetse flies. Insect Mol Biol symbionts. In: Campbell RK, Eikenbary RD (eds) Aphid-plant genotype
8:125–132 interactions. Elsevier Biomedical Press, Amsterdam, pp 319–327
Clark MA, Baumann L, Munson MA, Baumann P, Campbell BC, Duffus JE, Douglas AE (1997) Parallels and contrasts between symbiotic bacteria and
Osborne LS, Moran NA (1992) The eubacterial endosymbionts of whiteflies bacterial-derived organelles: evidence from Buchnera, the bacterial symbiont
(Homoptera, Aleyrodoidea) constitute a lineage distinct from the endosym- of aphids. FEMS Microbiol Ecol 24:1–9
bionts of aphids and mealybugs. Curr Microbiol 25:119–123 Douglas AE (1998) Nutritional interactions in insect-microbial symbioses: aphids
Clark MA, Baumann L, Baumann P, Rouhbakhsh D (1996) Ribosomal protein and their symbiotic bacteria Buchnera. Annu Rev Entomol 43:17–37
S1(RpsA) of Buchnera aphidicola, the endosymbiont of aphids: characteri- Douglas AE, Dixon AFG (1987) The mycetocyte symbiosis of aphids: variation
zation of the gene and detection of the product. Curr Microbiol 32:89–94 with age and morph in virginoparae of Megoura viciae and Acyrthosiphon
Clark MA, Baumann L, Baumann P (1998a) Sequence analysis of a 34.7-kb DNA pisum. J Insect Physiol 33:109–113
segment from the genome of Buchnera aphidicola (endosymbiont of aphids) Douglas AE, Prosser WA (1992) Synthesis of the essential amino acid tryptophan
containing groEL, dnaA, the atp operon, gidA, and rho. Curr Microbiol in the pea aphid (Acyrthosiphon pisum) symbiosis. J Insect Physiol
36:158–163 38:565–568
Douglas AE, Smith DC (1989) Are endosymbioses mutualistic? Trends Ecol Fukatsu T, Watanabe K, Sekiguchi Y (1998) Specific detection of intracellular
Evol 4:350–352 symbiotic bacteria of aphids by oligonucleotide-probed in situ hybridiza-
Durvasula RV, Gumbs A, Panackal A, Kruglov O, Aksoy S, Merrifield RB, tion. Appl Entomol Zool 33:461–472
Richards FF, Beard CB (1997) Prevention of insect-borne disease: an Gasnier-Fauchet F, Nardon P (1986) Comparison of methionine metabolism in
approach using transgenic symbiotic bacteria. Proc Natl Acad Sci USA symbiotic and aposymbiotic larvae of Sitophilus oryzae L. (Coleoptera:
94:3274–3278 curculionidae). II. Involvement of the symbiotic bacteria in the oxidation
Eisen JA (1995) The RecA Protein as a model molecule for molecular systematic of methionine. Comp Biochem Physiol 85B:251–254
studies of bacteria—comparison of trees of RecAs and 16S rRNAs from the Gasnier-Fauchet F, Gharib A, Nardon P (1986) Comparison of methionine
same species. J Mol Evol 41:1105–1123 metabolism in symbiotic and aposymbiotic larvae of Sitophilus oryzae L.
Eremeeva ME, Ching WM, Wu YL, Silverman DJ, Dasch GA (1998) Western (Coleoptera: Curculionidae) I. Evidence for a glycine N-methyltransferase-
blotting analysis of heat shock proteins of Rickettsiales and other eubacteria. like activity in the aposymbiotic larvae. Comp Biochem Physiol 85B:245–250
FEMS Microbiol Lett 167:229–237 Gray SM, Banerjee N (1999) Mechanisms of arthropod transmission of plant and
Faye I (1978) Insect immunity: early fate of bacteria injected in a Saturniid pupae. animal viruses. Microbiol Mol Biol Rev 63:128–148
J Invertebr Pathol 31:19–26 Grenier AM, Nardon C, Rahbe Y (1994) Observations on the micro-organisms
Filichkin SA, Brumfield S, Filichkin TP, Young MJ (1997) In vitro interactions of occurring in the gut of the pea aphid Acyrthosiphon pisum. Entomol Exp
the aphid endosymbiotic SymL chaperonin with barley yellow dwarf virus. Appl 70:91–96
J Virol 71:569–577 Griffiths GW, Beck SD (1973) Intracellular symbiotes of the pea aphid,
Fleischmann RD, Adams MD, White O, Clayton RA, Kirkness EF, Kerlavage AR, Acyrthosiphon pisum. J Insect Physiol 19:75–84
Bult CJ, Tomb JF, Dougherty BA, Merrick JM, McKenney K, Sutton G, Griffiths GW, Beck SD (1974) Effect of antibiotics on intracellular symbiotes in
Fitzhugh W, Fields C, Gocayne JD, Scott J, Shirley R, Liu LI, Glodek A, Kelley the pea aphid, Acyrthosiphon pisum. Cell Tissue Res 148:287–300
JM, Weidman JF, Phillips CA, Spriggs T, Hedblom E, Cotton MD, Utterback Grinyer I, Musgrave AJ (1966) Ultrastructure and peripheral membranes of the
TR, Hanna MC, Nguyen DT, Saudek DM, Brandon RC, Fine LD, Fritchman mycetomal microorganism of Sitophilus granarius (L. coleoptera). J Cell Sci
JL, Fuhrmann JL, Geoghagen NSM, Gnehm CL, McDonald LA, Small KV, 1:181–186
Fraser CM, Smith HO, Venter JC (1995) Whole-genome random sequencing Gross CA (1996) Function and regulation of heat shock proteins. In: Neidhard FC
and assembly of Haemophilus influenzae Rd. Science 269:469–512 (ed) Escherichia coli and Salmonella, vol 1. ASM Press, Washington, DC,
Forrest JMS (1987) Galling aphids. In: Minks AK, Harrewijn P (eds) Aphids: their pp 1382–1399
biology, natural enemies and control, vol 2A. Elsevier, Amsterdam, Gross R, Rappuoli R (1988) Positive regulation of pertussis toxin expression. Proc
pp 341–353 Natl Acad Sci USA 85:3913–3917
Fraser CM, Gocayne JD, White O, Adams MD, Clayton RA, Fleischmann RD, Bult Hara E, Ishikawa H (1990) Purification and partial characterization of
CJ, Kerlavage AR, Sutton G, Kelley JM, Fritchman JL, Weidman JF, Small KV, symbionin, an aphid endosymbiont-specific protein. Insect Biochem
Sandusky M, Fuhrmann J, Nguyen D, Utterback TR, Saudek DM, Phillips 20:421–427
CA, Merrick JM, Tomb JF, Dougherty BA, Bott KF, Hu PC, Lucier TS, Harada H, Ishikawa H (1993) Gut microbe of aphid closely related to its intra-
Peterson SN, Smith HO, Hutchison CA, Venter JC (1995) The minimal cellular symbiont. Biosystems 31:185–191
gene complement of Mycoplasma genitalium. Science 270:397–403 Harada H, Oyaizu H, Ishikawa H (1996) A consideration about the origin of
Fraser CM, Casjens S, Huang WM, Sutton GG, Clayton R, Lathigra R, White O, aphid intracellular symbiont in connection with gut bacterial flora. J Gen
Ketchum KA, Dodson R, Hickey EK, Gwinn M, Dougherty B, Tomb JF, Appl Microbiol 42:17–26
Fleischmann RD, Richardson D, Peterson J, Kerlavage AR, Quackenbush J, Harrison CP, Douglas AE, Dixon AFG (1989) A rapid method to isolate symbiotic
Salzberg S, Hanson M, vanVugt R, Palmer N, Adams MD, Gocayne J, bacteria from aphids. J Invertebr Pathol 53:427–428
Weidman J, Utterback T, Watthey L, McDonald L, Artiach P, Bowman C, Harwood RF, James MT (1979) Entomology in human and animal health.
Garland S, Fujii C, Cotton MD, Horst K, Roberts K, Hatch B, Smith HO, Macmillan, New York
Venter JC (1997) Genomic sequence of a Lyme disease spirochaete, Borrelia Hassan AKM, Moriya S, Baumann P, Yoshikawa H, Ogasawara N (1996) Structure
burgdorferi. Nature 390:580–586 of the dnaA region of the endosymbiont, Buchnera aphidicola, of the aphid
Fukatsu T, Ishikawa H (1992a) A novel eukaryotic extracellular symbiont in an Schizaphis graminum. DNA Res 3:415–419
aphid, Astegopteryx styraci (Homoptera, Aphididae, Hormaphidinae). Heddi A, Lefebvre F, Nardon P (1991) The influence of symbiosis on the respi-
J Insect Physiol 38:765–773 ratory control ratio (RCR) and the ADP/O Ratio in the adult weevil-
Fukatsu T, Ishikawa H (1992b) Soldier and male of an eusocial aphid Colophina Sitophilus oryzae (Coleoptera, Curculionidae). Endocytobiosis Cell Res
arma lack endosymbiont: implications for physiological and evolutionary 8:61–73
interaction between host and symbiont. J Insect Physiol 38:1033–1042 Heddi A, Lefebvre F, Nardon P (1993) Effect of endocytobiotic bacteria on
Fukatsu T, Ishikawa H (1992c) Synthesis and localization of symbionin, an aphid mitochondrial enzymatic activities in the weevil Sitophilus oryzae (Coleop-
endosymbiont protein. Insect Biochem Mol Biol 22:167–174 tera, Curculionidae). Insect Biochem Mol Biol 23:403–411
Fukatsu T, Ishikawa H (1993) Occurrence of chaperonin-60 and chaperonin-10 Heddi A, Charles H, Khatchadourian C, Bonnot G, Nardon P (1998) Molecular
in primary and secondary bacterial symbionts of aphids—implications characterization of the principal symbiotic bacteria of the weevil Sitophilus
for the evolution of an endosymbiotic system in aphids. J Mol oryzae: a peculiar G–C content of an endocytobiotic DNA. J Mol
Evol 36:568–577 Evol 47:52–61
Fukatsu T, Ishikawa H (1996) Phylogenetic position of yeast-like symbiont of Henry SM (1962) The significance of microorganisms in the nutrition of insects.
Hamiltonaphis styraci (Homoptera, Aphididae) based on 18S rDNA Trans N Y Acad Sci 24:676–683
sequence. Insect Biochem Mol Biol 26:383–388 Hinde R (1971a) The control of mycetome symbiotes of the aphids Brevicoryne
Fukatsu T, Ishikawa H (1998) Differential immunohistochemical visualization of brassicae, Myzus persicae, and Macrosiphon rosae. J Insect Physiol 17:1971–
the primary and secondary intracellular symbiotic bacteria of aphids. Appl 1800
Entomol Zool 33:321–326 Hinde R (1971b) The fine structure of mycetome symbiotes of the aphids
Fukatsu T, Nikoh N (1998) Two intracellular symbiotic bacteria from the mul- Brevicoryne brassicae, Myzus persicae, and Macrosiphum rosae. J Insect
berry psyllid Anomoneura mori (Insecta, Homoptera). Appl Environ Physiol 17:2035–2050
Microbiol 64:3599–3606 Hogenhout SA, van derWilk F, Verbeek M, Goldbach RW, van den Heuvel JFJM
Fukatsu T, Aoki S, Kurosu U, Ishikawa H (1994) Phylogeny of Cerataphidini (1998) Potato leafroll virus binds to the equatorial domain of the aphid
aphids revealed by their symbiotic microorganisms and basic structure of endosymbiotic GroEL homolog. J Virol 72:358–365
their galls: implications for host-symbiont coevolution and evolution of Houk EJ, Griffiths GW (1980) Intracellular symbiotes of the Homoptera. Annu
sterile soldier castes. Zool Sci 11:613–623 Rev Entomol 25:161–187
Houk EJ, Griffiths GW, Hadjokas NE, Beck SD (1977) Peptidoglycan in the cell McMillan DJ, Mau M, Walker MJ (1998) Characterisation of the urease gene
wall of the primary intracellular symbiote of the pea aphid. Science cluster in Bordetella bronchiseptica. Gene 208:243–251
198:401–403 Messer W, Weigl C (1996) Initiation of chromosome replication. In: Neidhardt
Humphreys NJ, Douglas AE (1997) Partitioning of symbiotic bacteria between FC (ed) Escherichia coli and Salmonella, vol 2. ASM Press, Washington, DC,
generations of insect: a quantitative study of a Buchnera sp. in the pea aphid pp 1579–1601
(Acyrthosiphon pisum) reared at different temperatures. Appl Environ Moran NA (1996) Accelerated evolution and Muller’s ratchet in endosymbiotic
Microbiol 63:3294–3296 bacteria. Proc Natl Acad Sci USA 93:2873–2878
Iaccarino FM, Tremblay E (1973) Comparazione ultrastrutturale della disimbiosi Moran N, Baumann P (1994) Phylogenetics of cytoplasmically inherited micro-
di Macrosiphum rosae (L.) e Dactynotus jaceae (L.) (Homoptera, Aphididae). organisms of arthropods. Trends Ecol Evol 9:15–20
Boll Lab Entomol Agrar Filipo Silvestri 30:319–335 Moran NA, Telang A (1998) Bacteriocyte-associated symbionts of insects:
Ishikawa H (1982) Isolation of the intracellular symbionts and partial character- a variety of insect groups harbor ancient prokaryotic endosymbionts. Bio-
izations of their RNA species of the elder aphid, Acyrthosiphon magnoliae. science 48:295–304
Comp Biochem Physiol 72B:239–247 Moran NA, Munson MA, Baumann P, Ishikawa H (1993) A molecular clock in
Ishikawa H (1987) Nucleotide composition and kinetic complexity of the geno- endosymbiotic bacteria is calibrated using the insect hosts. Proc R Soc Lond
mic DNA of an intracellular symbiont in the pea aphid Acyrthosiphon pisum. B Biol Sci 253:167–171
J Mol Evol 24:205–211 Moran NA, von Dohlen CD, Baumann P (1995) Faster evolutionary rates in endo-
Kakeda K, Ishikawa H (1991) Molecular chaperon produced by an intracellular symbiotic bacteria than in cospeciating insect hosts. J Mol Evol 41:727–731
symbiont. J Biochem 110:583–587 Moran NA, Kaplan ME, Gelsey MJ, Murphy TG, Scholes EA (1999) Phylogeny
Kambhampati S (1995) A phylogeny of cockroaches and related insects based on and evolution of the aphid genus Uroleucon based on nuclear and mito-
DNA sequence of mitochondrial ribosomal RNA genes. Proc Natl Acad Sci chondrial DNA sequences. Syst Entomol 24:85–93
USA 92:2017–2020 Munson MA, Baumann P, Kinsey MG (1991a) Buchnera gen. nov. and Buchnera
Katsumata R, Ikeda M (1993) Hyperproduction of tryptophan in Corynebacte- aphidicola sp. nov., a taxon consisting of the mycetocyte-associated, primary
rium glutamicum by pathway engineering. Biotechnology 11:921–925 endosymbionts of aphids. Int J Syst Bacteriol 41:566–568
Klotz MG, Norton JM (1998) Multiple copies of ammonia monooxygenase Munson MA, Baumann P, Clark MA, Baumann L, Moran NA, Voegtlin DJ,
(amo) operons have evolved under biased AT/GC mutational pressure Campbell BC (1991b) Evidence for the establishment of aphid-eubacterium
in ammonia-oxidizing autotrophic bacteria. FEMS Microbiol Lett endosymbiosis in an ancestor of four aphid families. J Bacteriol
168:303–311 173:6321–6324
Komaki K, Ishikawa H (1999) Intracellular symbionts of aphids are bacteria with Munson MA, Baumannn P, Moran MA (1992) Phylogenetic relationships of the
numerous genomic copies. J Mol Evol 48:717–722 endosymbionts of mealybugs (Homoptera: Pseudococcidae) based on 16S
Kreditch NM (1996) Biosynthesis of cysteine. In: Neidhardt FC (ed) Escherichia rDNA sequences. Mol Phylogenet Evol 1:26–30
coli and Salmonella, vol 1. ASM Press, Washington, DC, pp 514–527 Munson MA, Baumann L, Baumann P (1993) Buchnera aphidicola (a prokary-
Kusano T, Takeshima T, Inoue C, Sugawara K (1991) Evidence for two sets of otic endosymbiont of aphids) contains a putative 16S rRNA operon unlinked
structural genes coding for ribulose bisphosphate carboxylase in Thiobacillus to the 23S rRNA-encoding gene: sequence determination, and promoter and
ferrooxidans. J Bacteriol 173:7313–7323 terminator analysis. Gene 137:171–178
Lai CY, Baumann P (1992a) Genetic analysis of an aphid endosymbiont DNA Musgrave AJ, Grinyer I (1968) Membranes associated with the disintegration of
fragment homologous to the rnpA-rpmH-dnaA-dnaN-gyrB region of mycetomal micro-organisms in Sitophilus zeamais (Mots. Coleoptera). J Cell
eubacteria. Gene 113:175–181 Sci 3:65–70
Lai CY, Baumann P (1992b) Sequence analysis of a DNA fragment from Buchnera Nakabachi A, Ishikawa H (1997) Differential display of mRNAs related to amino
aphidicola (an endosymbiont of aphids) containing genes homologous to acid metabolism in the endosymbiotic system of aphids. Insect Biochem Mol
dnaG, rpoD, cysE andsecB. Gene 119:113–118 Biol 27:1057–1062
Lai CY, Baumann L, Baumann P (1994) Amplification of trpEG: adaptation of Nakabachi A, Ishikawa H (1999) Provision of riboflavin to the host aphid,
Buchnera aphidicola to an endosymbiotic association with aphids. Proc Natl Acyrthosiphon pisum, by endosymbiotic bacteria, Buchnera. J Insect Physiol
Acad Sci USA 91:3819–3823 45:1–6
Lai C-Y, Baumann P, Moran NA (1995) Genetics of the tryptophan biosynthetic Nardon P, Grenier AM (1988) Genetical and biochemical interactions between
pathway of the prokaryotic endosymbiont (Buchnera) of the aphid the host and its endocytobiotes in the weevil Sitophilus (Coleoptera,
Schlechtendalia chinensis. Insect Mol Biol 4:47–59 Curculionidae) and other related species. In: Scannerini S (ed) Cell to cell
Lai CY, Baumann P, Moran N (1996) The endosymbiont (Buchnera sp.) of the signals in plant, animal and microbial symbiosis. Springer, Heidelberg,
aphid Diuraphis noxia contains plasmids consisting of trpEG and tandem pp 255–270
repeats of trpEG pseudogenes. Appl Environ Microbiol 62:332–339 Nicholson ML, Gaasenbeek M, Laudenbach DE (1995) Two enzymes together
Lambert JD, Moran NA (1998) Deleterious mutations destabilize ribosomal RNA capable of cysteine biosynthesis are encoded on a cyanobacterial plasmid.
in endosymbiotic bacteria. Proc Natl Acad Sci USA 95:4458–4462 Mol Gen Genet 247:623–632
Liadouze I, Febvay G, Guillaud J, Bonnot G (1996) Metabolic fate of energetic Nogge G (1976) Aposymbiotic tsetse flies. Glossina morsitans morsitans obtained
amino acids in the aposymbiotic pea aphid Acyrthosiphon pisum (Harris) by feeding adults on rabbits immunized specifically with symbionts. J Insect
(Homoptera, Aphididae). Symbiosis 21:115–127 Physiol 24:299–304
Lipsitch M, Nowak MA, Ebert D, May RM (1995) The population dynamics of Nogge G (1982) Significance of symbionts for the maintenance of an optimal
vertically and horizontally transmitted parasites. Proc R Soc Lond B Biol Sci nutritional state of successful reproduction in hematophagous arthropods.
260:321–327 Parasitology 82:299–304
Llanes C, Gabant P, Couturier M, Bayer L, Plesiat P (1996) Molecular analysis of Norton JM, Low JM, Martin G (1996) The gene encoding ammonia
the replication elements of the broad-host-range RepA/C replicon. Plasmid monooxygenase subunit A exists in three nearly identical copies in
36:26–35 Nitrosospira sp. NpAV. FEMS Microbiol Lett 139:181–188
Margolis N, Hogan D, Tilly K, Rosa PA (1994) Plasmid location of Borrelia purine O’Neill A, Hoffman A, Werren JH (1997) Influential passengers; inherited micro-
biosynthesis gene homologs. J Bacteriol 176:6427–6432 organisms and arthropod reproduction. Oxford University Press, Oxford
Maynard Smith J, Szathmáry E (1995) The major transitions in evolution. W. H. Ohtaka C, Nakamura H, Ishikawa H (1992) Structures of chaperonins from an
Freeman Spektrum, Oxford/New York intracellular symbiont and their functional expression in Escherichia coli
McLean DL, Houk EJ (1973) Phase contrast and electron microscopy pf the groE mutants. J Bacteriol 174:1869–1874
mycetocytes and symbiotes of the pea aphid, Acyrthosiphon pisum. J Insect Patten CL, Glick BR (1996) Bacterial biosynthesis of indole-3-acetic acid. Can
Physiol 19:625–633 J Microbiol 42:207–220
Pennisi E (1998) Evolution—heat shock protein mutes genetic changes. Science Smith DC, Douglas AE (1987) The biology of symbiosis. Edward Arnold, London
282:1796 Smith OH, Yanofsky C (1962) Enzymes involved in the biosynthesis of trypto-
Remaudière G, Remaudière M (1997) Catalogue des aphididae du monde. phan. Methods Enzymol 5:794–806
Institut National de la Recherche Agronomique, Paris Spaulding AW, von Dohlen CD (1998) Phylogenetic characterization and molec-
Riley M, Labedan B (1996) Escherichia coli gene products: physiological functions ular evolution of bacterial endosymbionts in psyllids (Hemiptera:
and common ancestries. In: Neidhardt FC (ed) Escherichia coli and Salmo- Sternorrhyncha). Mol Biol Evol 15:1506–1513
nella, vol 2. ASM Press, Washington, DC, pp 2118–2202 Stephens RS, Kalman S, Lammel C, Fan J, Marathe R, Aravind L, Mitchell W,
Romero D, Palacios R (1997) Gene amplification and genomic plasticity in pro- Olinger L, Tatusov RL, Zhao QX, Koonin EV, Davis RW (1998) Genome
karyotes. Annu Rev Genet 31:91–111 sequence of an obligate intracellular pathogen of humans: Chlamydia
Roth JR, Benson N, Galitski T, Haack K, Lawrence JG, Miesel L (1996) trachomatis. Science 282:754–759
Rearrangements of the bacterial chromosome: formation and applications. Sylvester ES (1985) Multiple acquisition of viruses and vector-dependent pro-
In: Neidhardt FC (ed) Escherichia coli and Salmonella, vol 2. ASM Press, karyotes: consequences on transmission. Annu Rev Entomol 30:71–88
Washington, DC, pp 2256–2276 Telang A, Sandström J, Dyreson E, Moran NA (1999) Feeding damage by
Rouhbakhsh D, Baumann P (1995) Characterization of a putative 23S-5S rRNA Diuraphis noxia results in nutritionally enhanced phloem diet. Entomol
operon of Buchnera aphidicola (endosymbiont of aphids) unlinked to the Exp Appl 91:403–412
16S rRNA-encoding gene. Gene 155:107–112 Thao ML, Baumann P (1998) Sequence analysis of a DNA fragment from
Rouhbakhsh D, Moran NA, Baumann L, Voegtlin DJ, Baumann P (1994) Detec- Buchnera aphidicola (aphid endosymbiont) containing the genes dapD-
tion of Buchnera, the primary prokaryotic endosymbiont of aphids, using htrA-ilvI-ilvH-ftsL-ftsI-murE. Curr Microbiol 37:214–216
the polymerase chain reaction. Insect Mol Biol 3:213–217 Thao ML, Baumann L, Baumann P, Moran NA (1998) Endosymbionts
Rouhbakhsh D, Lai CY, von Dohlen CD, Clark MA, Baumann L, Baumann P, (Buchnera) from the aphids Schizaphis graminum and Diuraphis noxia have
Moran NA, Voegtlin DJ (1996) The tryptophan biosynthetic pathway of aphid different copy numbers of the plasmid containing the leucine biosynthetic
endosymbionts (Buchnera): genetics and evolution of plasmid-associated genes. Curr Microbiol 36:238–240
anthranilate synthase (trpEG) within the Aphididae. J Mol Evol 42:414–421 Tremblay E (1989) Coccoidea endosymbiosis. In: Schwemmler W, Gassner G
Rouhbakhsh D, Clark MA, Baumann L, Moran NA, Baumann P (1997) Evolution (eds) Insect endocytobiosis: morphology, physiology, genetics, evolution.
of the tryptophan biosynthetic pathway in Buchnera (aphid endosymbionts): CRC Press, Boca Raton, pp 145–173
studies of plasmid-associated trpEG within the genus Uroleucon. Mol Unterman BM, Baumann P (1990) Partial characterization of ribosomal RNA
Phylogenet Evol 8:167–176 operons of the pea-aphid endosymbionts: evolutionary and physiological
Rutherford SL, Lindquist S (1998) Hsp90 as a capacitor for morphological implications. In: Campbell RK, Eikenbary RD (eds) Aphid-plant genotype
evolution. Nature 396:336–342 interactions. Elsevier Biomedical Press, Amsterdam, pp 329–350
Sacchi L, Corona S, Grigolo A, Laudani U, Selmi MG, Bigliardi E (1996) The fate of Unterman BM, Baumann P, McLean DL (1989) Pea aphid symbiont relationships
the endocytobionts of Blattella germanica (Blattaria, Blattellidae) and established by analysis of 16S rRNAs. J Bacteriol 171:2970–2974
Periplaneta americana (Blattaria, Blattidae) during embryo development. van den Heuvel JFJM, Verbeek M, van der Wilk F (1994) Endosymbiotic bacteria
Ital J Zool 63:1–11 associated with circulative transmission of potato leafroll virus by Myzus
Sacchi L, Nalepa CA, Bigliardi E, Corona S, Grigolo A, Laudani U, Bandi C persicae. J Gen Virol 75:2559–2565
(1998a) Ultrastructural studies of the fat body and bacterial endosymbionts van den Heuvel JFJM, Bruyere A, Hogenhout A, ZieglerGraff V, Brault V, Verbeek
of Cryptocercus punctulatus Scudder (Blattaria: Cryptocercidae). Symbiosis M, van derWilk F, Richards K (1997) The N-terminal region of the luteovirus
25:251–269 readthrough domain determines virus binding to Buchnera GroEL and is
Sacchi L, Nalepa CA, Bigliardi E, Lenz M, Bandi C, Corona S, Grigolo A, Lambiase essential for virus persistence in the aphid. J Virol 71:7258–7265
S, Laudani U (1998b) Some aspects of intracellular symbiosis during embryo van Ham RCHJ, Moya A, Latorre A (1997) Putative evolutionary origin of
development of Mastotermes darwiniensis (Isoptera: Mastotermitidae). plasmids carrying the genes involved in leucine biosynthesis in Buchnera
Parassitologia 40:309–316 aphidicola (endosymbiont of aphids). J Bacteriol 179:4768–4777
Sandström J, Moran N (1999) How nutritionally imbalanced is phloem sap for van Ham RCHJ, Martı́nez-Torres D, Moya A, Latorre A (1999) Plasmid-encoded
aphids? Entomol Exp Appl 91:203–210 anthranilate synthase (TrpEG) in Buchnera aphidicola from the family
Sandström J, Pettersson J (1994) Amino acid composition of phloem sap and the Pemphigidae. Appl Environ Microbiol 65:117–125
relation to intraspecific variation in pea aphid (Acyrthosiphon pisum) per- von Dohlen CD, Moran NA (1995) Molecular phylogeny of the Homoptera—a
formance. J Insect Physiol 40:947–955 paraphyletic taxon. J Mol Evol 41:211–223
Sandström J, Telang A, Moran NA (2000) Nutritional enhancement of host plants by von Wintzingerode F, Gobel UB, Stackebrandt E (1997) Determination of micro-
aphids-a comparison of three aphid species on grasses. J Insect Physiol 46:33–40 bial diversity in environmental samples: pitfalls of PCR-based rRNA analysis.
Sasaki T, Ishikawa H (1995) Production of essential amino acids from glutamate FEMS Microbiol Rev 21:213–229
by mycetocyte symbionts of the pea aphid, Acyrthosiphon pisum. J Insect Waku Y, Endo Y (1987) Ultrastructure and life cycle of the symbionts in
Physiol 41:41–46 a Homopteran insect, Anomoneura mori Schwartz (Psyllidae). Appl Entomol
Sasaki T, Aoki T, Hayashi H, Ishikawa H (1990) Amino acid composition of the Zool 22:630–637
honeydew of symbiotic and aposymbiotic pea aphids Acyrthosiphon pisum. Wernegreen JJ, Moran NA (1999) Evidence for genetic drift in endosymbionts
J Insect Physiol 36:35–40 (Buchnera): analyses of protein-coding genes. Mol Biol Evol 16:83–97
Sato S, Ishikawa H (1997a) Expression and control of an operon from an Werren JH, O’Neill SL (1997) The evolution of heritable symbionts. In: O’Neill SL,
intracellular symbiont which is homologous to the groE operon. Hoffman AA, Werren JH (eds) Influential passengers; inherited microorgan-
J Bacteriol 179:2300–2304 isms and arthropod reproduction. Oxford University Press, Oxford, pp 1–41
Sato S, Ishikawa H (1997b) Structure and expression of the dnaKJ operon of Whitehead LF, Douglas AE (1993) A metabolic study of Buchnera, the intracel-
Buchnera, an intracellular symbiotic bacteria of aphid. J Biochem 122:41–48 lular bacterial symbionts of the pea aphid Acyrthosiphon pisum. J Gen
Schröder D, Deppisch H, Obermayer M, Krohne G, Stackebrandt E, Holldobler B, Microbiol 139:821–826
Goebel W, Gross R (1996) Intracellular endosymbiotic bacteria of Wicker C, Nardon P (1982) Development responses of symbiotic and aposymbiotic
Camponotus species (carpenter ants): systematics, evolution and ultrastruc- weevils Sitophilus oryzae L. (Coleoptera, Curculionidae) to a diet
tural characterization. Mol Microbiol 21:479–489 supplemented with aromatic amino-acids. J Insect Physiol 28:1021–1024
Silva FJ, van Ham RCHJ, Sabater B, Latorre A (1998) Structure and evolution of Wilkinson TL (1998) The elimination of intracellular microorganisms from
the leucine plasmids carried by the endosymbiont (Buchnera aphidicola) insects: an analysis of antibiotic-treatment in the pea aphid (Acyrthosiphon
from aphids of the family Aphididae. FEMS Microbiol Lett 168:43–49 pisum). Comp Biochem Physiol A Comp Physiol 119:871–881
20 Vibrio fisheri: Squid Symbiosis
Eric V. Stabb1 . Karen L. Visick2
1
Department of Microbiology, University of Georgia, Athens, GA, USA
2
Department of Microbiology and Immunology, Loyola University Chicago, Maywood, IL, USA
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497 and this natural light production provides a noninvasive means
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497 of monitoring colonization. Furthermore, tools such as green
Ecology of V. fischeri and Its Squid Host E. scolopes . . . . 497 fluorescent protein (GFP) have been engineered to permit visu-
alization of bacteria at all stages of colonization in the transparent
Structure of Light Organ, Dynamics of Colonization, and symbiotic tissue of juvenile squid. The bacterium can be readily
Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500 manipulated genetically, and the genome sequences of multiple
Structure of Light Organ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500 strains are known, making it feasible to test specific genes for their
Dynamics of Symbiosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502 roles in bacteria-host interactions. Finally, investigations into the
Specificity and Host Defenses . . . . . . . . . . . . . . . . . . . . . . . . . . . 505 biology of the squid and its symbiotic organ, the light organ,
provide a framework for developing hypotheses to be tested. The
Host Development and Bacterial Signals . . . . . . . . . . . . . . . . . . 508 result is a robust model that is continually yielding novel insights.
Host Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508 In this chapter, we describe in detail the biology of V. fischeri
Bacterial Signals That Influence Host Development . . . . 509 as it relates to the ability of this microbe to form a specific
symbiosis with E. scolopes. We begin with an introduction
Bacterial Genes and Phenotypes Involved in to the ecology of V. fischeri and its squid host. Then, to provide
Colonization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511 a basis for understanding the symbiosis, we describe the struc-
Bioluminescence and Pheromone-Dependent ture of the symbiotic light organ and give an overview of what
Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511 is known about the dynamics of colonization and specificity in
Biofilm Formation and Biofilm Regulators . . . . . . . . . . . . . 514 the symbiosis. We then discuss host development and the known
Motility, Chemotaxis, and Their Regulation . . . . . . . . . . . . 516 roles of the bacteria and bacterial signals in developmental
Iron Uptake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519 processes. With this foundation, we then describe a number
GacA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519 of bacterial genes and phenotypes whose roles in symbiosis
N-Acetyl D-Glucosamine Repressor NagC . . . . . . . . . . . . . . 520 have been investigated, including the processes of bacterial bio-
Possible Role for Pili in Symbiosis . . . . . . . . . . . . . . . . . . . . . . 521 luminescence, biofilm formation, motility, and iron uptake.
Cellobiose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521 Finally, we conclude with a brief discussion of evolution and
Other Genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522 our perspectives on the field and its future.
Evolution of the V. fischeri-E. scolopes Symbiosis . . . . . . . . . 522

Ecology of V. fischeri and Its Squid Host E. scolopes
Perspectives and Future of the Field . . . . . . . . . . . . . . . . . . . . . . . 525
It is not possible to appreciate the biology of V. fischeri fully
without first understanding the environments that it experiences
Introduction during its life cycle. As a marine bacterium, V. fischeri primarily
resides in seawater, which contains dissolved salts at
a concentration of 3.5%. The dissolved salts include sodium,
Overview chloride, magnesium, sulfate, calcium, and potassium. In this
environment, V. fischeri can be found free-living in seawater and
The marine bioluminescent bacterium Vibrio fischeri forms also associated with sediment (Lee and Ruby 1992, 1994b).
a highly specific mutualistic symbiosis with the Hawaiian bob- It also can be found colonizing animal hosts.
tail squid Euprymna scolopes. The study of this symbiosis over The best known of these animal associations are exquisitely
the past 20 years has been aided by the nature of the interaction evolved light-organ symbioses in which V. fischeri colonizes the
itself: the squid hatch without V. fischeri but rapidly acquire them light-emitting organs of certain fishes and squids, generating
from the seawater, and thus experimentally, wild-type or mutant bioluminescence used by the host in exchange for nutrients
bacteria can be added to the seawater and the process of coloni- and a privileged niche. For example, V. fischeri colonizes light
zation examined. Once the bacteria colonize, they bioluminesce, organs in monocentrid ‘‘pinecone’’ fishes of the genera Cleidopus
498 20 Vibrio fisheri: Squid Symbiosis
or Monocentris, both found in the Pacific Ocean (Fitzgerald Among its specialized light-organ symbioses, the best stud-
1977; Ruby and Nealson 1976), and in sepiolid ‘‘bobtail’’ squids ied is that between V. fischeri and E. scolopes, the Hawaiian
of the genera Sepiola or Euprymna, which are found in the bobtail squid (> Fig. 20.1a). E. scolopes is a nocturnal predator
Mediterranean Sea and Pacific Ocean, respectively (Fidopiastis that feeds on polychaetes and shrimp. The hatchlings
et al. 1998; Jones et al. 2006; Nishiguchi 2002; Nishiguchi et al. (> Fig. 20.2a) are typically only ~3–4 mm long but can grow
1998; Wei and Young 1989). Although its role as a biolumines- to be ten times that length (Moynihan 1983; Shears 1988). The
cent symbiont is well studied, the association of V. fischeri with adults likely live less than a year in the wild (Hanlon et al. 1997;
hosts is not restricted to monospecific light-organ symbioses. It Singley 1983) and are found near the Hawaiian coast in shallow,
has also been isolated from multi-species gut consortia of fish sandy reef areas, in a few meters or even as little as a few
(Ramesh and Venugopalan 1989; Ruby and Morin 1979; Sugita centimeters of water. It is unclear the extent to which this habitat
and Ito 2006) and from chitinous structures on the invertebrate reflects where they live versus where researchers typically search
hydrozoans Aglaophenia tubiformis and Halopteris diaphana for them. There are reports of E. scolopes being found well
(Stabili et al. 2008). outside the reefs and even at depths of 200 m (Berry 1912), but
Although it is a close associate of marine animals, the genetic for convenience, researchers typically have stayed closer to shore.
and physiological capacity of V. fischeri is unlike that of obligate These animals appear to be solitary except when they are mating,
symbionts (Ochman and Moran 2001; Ruby et al. 2005). Both its which has been observed both in shallow water and in captivity
demonstrated metabolic flexibility and its genomic content (> Fig. 20.1b). The ability to maintain and mate E. scolopes in
indicate that V. fischeri is able to grow under a range of condi- laboratory aquaria has underpinned its development as a model
tions using any of several substrates, unlike many obligate sym- experimental system.
bionts that evolve reduced genomes adapted to a relatively After mating, E. scolopes females lay clutches of eggs that lack
simple and constant host environment. The metabolic diversity V. fischeri symbionts. Upon hatching, each new generation must
and genomic content of V. fischeri suggest that an important acquire V. fischeri from the surrounding seawater (Wei and
component of this bacterium’s life history occurs outside of Young 1989), a phenomenon that allows researchers in con-
specific symbioses, consistent with the observation that trolled laboratory environments to compare animals infected
V. fischeri has been found free-living in different marine envi- with different strains or with no V. fischeri symbionts at all. If
ronments, both aerobic and anaerobic, in the water column and V. fischeri is present, infection occurs within hours, and it is so
in sediments (Garcia-Amado et al. 2011; Jones et al. 2007; Lee efficient that no uninfected E. scolopes has ever been found in the
and Ruby 1994b; Orndorff and Colwell 1980; Ruby et al. 1980). wild. The animal maintains a monospecific culture of V. fischeri
While it is possible that V. fischeri isolated from sediments and in its light organ. As discussed below, this specific infection with
the water column had recently cycled through a host, it also has V. fischeri triggers a developmental program in the light-organ
been isolated in regions far from any known light-organ symbi- tissue. Although the light organ undergoes large morphological
oses, such as the Sargasso Sea and coastal waters off the north- changes, the animals maintain a monospecific culture of
eastern United States. V. fischeri throughout their life, allowing the squid to exploit
Despite its frequent association with hosts and its phyloge- the symbionts’ bioluminescence.
netic relationship to known pathogens such as V. cholerae, Whether to hide from predators or prey, E. scolopes makes
V. parahaemolyticus, and V. vulnificus (Tantillo et al. 2004), to extensive use of camouflaging (Anderson and Mather 1996;
our knowledge, V. fischeri has never been documented as Shears 1988), a general strategy that appears to include their
a pathogen. Its inability to grow at 37 C certainly restricts it use of V. fischeri bioluminescence. The animals cover themselves
from causing human infections, and it has not been observed to with sand and use chromatophores to change colors among
cause disease in marine organisms even at permissive temperatures. a natural-looking palette (> Fig. 20.1c). They can even be
The apparent nonpathogenicity of V. fischeri contrasts with its observed swimming with a sand coat, which they can discard
relative Vibrio salmonicida, which causes cold-water vibriosis in quickly (Shears 1988). Similarly, the V. fischeri symbionts are
salmonids, and with another common bioluminescent marine apparently used in a strategy referred to as ‘‘counterillu-
bacterium, Vibrio harveyi, which apparently is responsible for mination,’’ where ventrally directed bioluminescence is used to
some die-off events in aquacultured shrimp. V. fischeri has been obscure the squid’s silhouette from organisms beneath it in the
isolated from aquaculture tanks in fish-rearing facilities (Alcaide water column. The strongest evidence that this is the function of
2003; Montes et al. 2006), and in one instance, it was isolated symbiotic bioluminescence includes the architecture of the
from organs of diseased aquacultured fish (Lamas et al. 1990); organ (McFall-Ngai and Montgomery 1990) and the observa-
however, it was not shown to be causal to morbidity or mortal- tion that the light emitted is controlled to directly correlate
ity. It seems plausible that in this isolated incident, V. fischeri may with the ambient downwelling light (Jones and Nishiguchi
have been a secondary opportunist flourishing in a host already 2004). Although a nutritional or other benefit of the symbionts
compromised by another microbe. Overall, although the V. fischeri cannot be ruled out, E. scolopes raised through a complete
genome encodes homologs of virulence factors found in other life cycle without V. fischeri did not appear compromised
members of the Vibrionaceae (Ruby et al. 2005), this species (Claes and Dunlap 2000), further reinforcing the idea that
appears to enter benign or beneficial associations with hosts. symbiont bioluminescence is the main advantage to the host.
Vibrio fisheri: Squid Symbiosis 20 499
. Fig. 20.1
Adult E. scolopes. (a) Adult E. scolopes sitting on coral sand; (b) a mating pair of E. scolopes; (c) sand-covered E. scolopes; (d) ventrally
dissected E. scolopes; (e) a close-up view of the adult light organ (Image from panel a was taken from Stabb and Millikan (2009), while
images from b and c are courtesy of Kati Geszvain)
The counterillumination model above and other possible expla- Given the evidence for mutualism, it is not surprising
nations for the use of bioluminescent symbionts by the squid that V. fischeri isolates from E. scolopes appear to have coevolved
were reviewed recently (Stabb and Millikan 2009). with this host. There is considerable evidence that certain
The advantage of the symbiosis for V. fischeri appears clearer. strains of V. fischeri have adapted to be especially proficient
V. fischeri is provided nutrients in the E. scolopes light organ, and colonizers of E. scolopes (Lee and Ruby 1994a; Mandel et al.
this supports its rapid growth (Ruby and Asato 1993). Further- 2009; Nishiguchi 2002; Nishiguchi et al. 1998; Schuster et al.
more, V. fischeri cells also appear to benefit from the host 2010), and repeated passage of strains through E. scolopes in the
immune system, which maintains an exclusive relationship laboratory has shown that less proficient colonizers can
with the bacteria, protecting them from predation or competi- evolve into more effective symbionts (Schuster et al. 2010).
tion by other microbes. Each morning, the squid expel most Interestingly, the gene encoding the regulatory sensor RscS,
of the V. fischeri cells in their light organ out into the discussed further below, appears to be a key genetic
environment and then support regrowth of the remaining acquisition in the evolution of V. fischeri, leading to more pro-
symbionts throughout the day (Boettcher et al. 1996) ficient colonization of Euprymna hosts in the Pacific
(> Fig. 20.3). Given this daily venting and re-culturing, one (Mandel et al. 2009).
would expect to find relatively high populations of V. fischeri in As a coevolved mutualism, the V. fischeri-E. scolopes symbi-
habitats occupied by E. scolopes, which indeed has been observed osis resembles many specific bacterium-host interactions found
(Lee and Ruby 1994b). Other ecological studies (Lee and Ruby in nature. Given several features that make it experimentally
1994a) support the idea that in shallow, sandy Hawaiian reefs tractable, it serves as a powerful natural model for such associ-
occupied by E. scolopes, the ability to colonize this squid is ations. Although V. fischeri may not require E. scolopes or other
advantageous for V. fischeri. Taken together, the evidence light-organ symbioses to survive and grow in the environment,
suggests that this symbiosis is a mutualism that benefits in locales where hosts are available, light-organ colonization
both partners. appears to be very important in its ecology. Thus, studies of
a b
~100 μm
c d
e pore antechamber f
crypt ep
duct
ciliated mv
epithelial
appendages 3 1
2
. Fig. 20.2
Juvenile E. scolopes. (a) Juvenile E. scolopes, with the light organ prominent as a black shape in the center of the mantle; (b) an image,
generated with scanning electron microscopy, in which the ciliated epithelial appendages are seen extending from the surface and
the underlying light-organ tissues are apparent; (c) a cross section of one of the appendages revealing a blood sinus; (d) the surface
of part of the light organ depicting three pores, in/near one of which is an aggregate of V. fischeri cells; (e) cartoon depicting the
structure of one-half of the juvenile light organ with three crypts labeled (1, 2, and 3); (f) cross section of a region of a colonized crypt.
mv microvilli, ep epithelium (The image from panel a was previously published (Dunn and Stabb 2008a), and the image in panel d is
cropped from a version that was previously published (Yip et al. 2006))
V. fischeri infecting E. scolopes address issues directly significant 1011 cells per ml of light-organ fluid (Boettcher and Ruby 1990;
to the ecology of the bacterium in nature, and they can elucidate Nyholm and McFall-Ngai 1998) (> Fig. 20.1d). It is at these high
our understanding of bacterium-host symbioses in general. cell densities that the bacterial contribution to the symbiosis,
bioluminescence, is produced. The adult organ contains several
tissues, including lens and reflector tissues that direct and modu-
Structure of Light Organ, Dynamics of late the light (> Fig. 20.1e). Of note, the light organ occupies
Colonization, and Development a significant portion of the space within the squid’s body cavity
(mantle), a feature that suggests the relative importance of this
Structure of Light Organ organ and the symbiosis to the life cycle of the animal.
Juvenile E. scolopes, which hatch without symbionts
In adult E. scolopes, V. fischeri cells reside within a complex, (aposymbiotic), are first exposed to V. fischeri cells when the
bilobed organ at a level in excess of 109 bacteria or approximately animal ventilates seawater into its mantle cavity. Derived from
McFall-Ngai 1999; Small and McFall-Ngai 1999). To progress to

Night Morning colonization, V. fischeri must be able to overcome these chal-
lenges and others, as described in greater detail below.
Each of the six ducts leads to an antechamber (Sycuro et al.
2006), a small chamber outside of the larger deep crypt where
most colonizing cells eventually reside (> Fig. 20.2e). Each set of
antechambers has an average size and complexity that corre-
sponds to that of the deep crypt with which it is associated. For
example, the antechamber of crypt 1 has an average cross-
sectional area of 1,380 mm2, while the antechamber of crypt 3
has its largest cross-sectional dimension in the range of 510 mm2
(Sycuro et al. 2006) (> Fig. 20.2e). The antechambers are not
permissive to persistent colonization, likely because this is
. Fig. 20.3 a region of the light organ with extensive antimicrobial activities,
Daily behavior of E. scolopes. The cartoon depicts the behavior of such as nitric oxide (NO) production (Davidson et al. 2004).
E. scolopes, which forages for food in the water column at night, Little else is known about the antechambers. To reach their
and in the morning, expels 90 % of its bacteria and buries in the respective deep crypts, the bacteria must exit the antechamber
sand for the day through a bottleneck region that has small dimensions (between
5 and 9 mM in width) that limit passage.
Like the antechambers, the three sets of deep crypts have
different characteristics with respect to size and complexity. In
an outgrowth of the digestive tract (Montgomery and McFall- the newly hatched juvenile, the largest and most developed is
Ngai 1993), the juvenile light organ (> Fig. 20.2b & e) features known as deep crypt 1 (or simply crypt 1), while the smallest and
two sets of ciliated surface appendages that project into the least developed is crypt 3. Crypt 2 is intermediate between the
mantle cavity (McFall-Ngai and Ruby 1991). The cilia on these other two. Because of the complexity of the deep crypt tissues,
appendages, along with the cilia decorating ridges on either side no estimate has been made of the area or volume of these spaces.
of the organ, entrain the bacteria-containing ambient seawater Each of the deep crypts in the uncolonized juvenile is lined with
toward pores that serve as the entrances to the light organ columnar epithelial cells. The microvilli on the surfaces of these
(McFall-Ngai and Ruby 1998) (> Fig. 20.2d & e). In addition cells in colonized animals provide points of direct contact with
to the cilia, the surface of the light organ is coated with mucus the bacterial symbionts (> Fig. 20.2f). It is within these confined
that is secreted from the epithelial cells that line the appendages deep crypt spaces that multiplication to high cell density ensues,
(Nyholm et al. 2000, 2002). The directed movement of the cilia resulting in colonization of the host by V. fischeri. Sycuro et al.
and surface-secreted mucus are thought to promote attachment (2006) reported that there is no particular order to which the six
of V. fischeri carried into the squid with the ventilated seawater. deep crypts become colonized. Similarly, Dunn et al. (2006)
Thus, at a very early stage in colonization, V. fischeri experiences noted that, while bacterial gene expression was altered in crypt
a mucus-coated surface, an environment that is vastly different 3 relative to the other crypts, the timing of colonization did not
from seawater. appear different. Thus, there appear to be differences in crypt
A small aggregate of V. fischeri cells accumulates on the structure and maturity in hatchlings, but the differences do not
surface of the symbiotic light organ, then ultimately individual interfere with colonization. Rapid growth of the bacteria is
cells track into pores to enter the organ (> Fig. 20.2d) (Nyholm supported by host-provided nutrients, including amino acids
et al. 2000). A total of six pores exist, three on each side of presented in the form of peptides (Graf and Ruby 1998), and
the organ. They range in size from 5 to 15 mm in diameter likely oxygen. When a sufficiently high cell density is achieved,
(Montgomery and McFall-Ngai 1993). Thus, V. fischeri cells, the production of bioluminescence is induced (Ruby and
which are approximately 1–2 mm in length (Millikan and Ruby Asato 1993).
2003; Ruby and Asato 1993), are substantially smaller than the It is readily apparent from this brief description that
pores through which they enter. Bacterial motility appears to be V. fischeri experiences a variety of environments during its
important for entry, as nonmotile bacteria appear to aggregate passage into the deep crypts where multiplication and coloniza-
but do not migrate to the pores (Nyholm et al. 2000). tion occurs. V. fischeri must transit from the nutrient-limited
The six pores open into six ducts or tube-like extensions seawater to the mucus-lined surfaces of the light organ, through
from the surface (> Fig. 20.2e) (Montgomery and McFall-Ngai the apparently challenging environments of the ducts and ante-
1993). Passage through the ducts appears to be a challenge: chambers, to reach the nutrient-rich, hospitable environment of
although heterologous species such as V. parahaemolyticus the deep crypts where rapid growth is possible and a generation
appear competent to reach the duct, they fail to colonize time as low as 30 min is estimated (Ruby and Asato 1993). Each
(Nyholm et al. 2000). The duct contains mucus, cilia that stage likely requires the expression of a distinct set of traits
appear to beat outward toward the pores, and antimicrobial that permit V. fischeri—and no other bacteria—to successfully
molecules (Davidson et al. 2004; McFall-Ngai and Ruby 1998; navigate these challenges.
a 0 0.5 2 3 12 48
h
. Fig. 20.4
Initiation of colonization. (a) An approximate time line of some of the known events in colonization during the first 48 h is depicted.
Following hatching (0 h), the light organ is transiently permissive to entry by bacteria (0.5 h). Mucus shedding (represented by fuzzy
shading on the surface of the organ) is induced (2 h), promoting surface aggregation by V. fischeri (ovals) (3 h). Colonization occurs
(darker shading) (12 h) and triggers apoptosis (represented by dots), regression of the appendages, and cessation of mucus secretion
(48 h). (b) Images of light organs clonally or dually colonized by GFP- or RFP-expressing V. fischeri cells. The segregation of initiating cells
is apparent by the separation of colors; in the last panel, mixing of cells can be seen (white arrow) (Images from panel b were previously
published Dunn et al. (2006))
Dynamics of Symbiosis observed in the deep crypts, but not beads with larger diameters
(2 mm or 10 mm) or cells (Bacillus cereus) with a diameter of
One of the most interesting facets of the V. fischeri-squid sym- 5 mm. This phenomenon was labeled the permissive period,
biosis is its dynamic nature. The squid are nocturnal animals, since even nonsymbiotic bacteria can gain entry. However, no
and many of their behaviors are cued to the day/night cycle. For viable bacteria could be recovered from light organs at this time
example, the animals forage for food at night, but bury in the using plating techniques, and furthermore, no particles (bacteria
sand during the day (> Fig. 20.3) (Moynihan 1983). In addition, or beads) could be detected 2 h after inoculation (Nyholm et al.
juveniles hatch from eggs at dusk; thus, these newly hatched 2002). Thus, following entry, the bacteria and beads appear to be
squid become colonized at night. That event begins a cycle of removed, likely by host defense cells (hemocytes) (Nyholm et al.
colonization, expulsion, and regrowth: every dawn, colonized 2009), and the time between 1 and 2 h after hatching represents
squid expel 90% of their bacterial symbionts by means of a nonpermissive period.
a muscle-induced contraction (> Fig. 20.3) (Boettcher et al. Within 1–2 h following hatching, mucus secretion occurs
1996; Graf and Ruby 1998; Lee and Ruby 1994b; Nyholm and from cells within the ciliated epithelial appendages in animals
McFall-Ngai 1998; Ruby and Asato 1993). In the adult, the result exposed to bacteria, but not in response to beads (> Fig. 20.4a)
is the release of a toothpaste-like gel of acellular matrix along (Nyholm et al. 2002). Sialomucin is the predominant mucin
with bacteria and host cells (Nyholm and McFall-Ngai 1998). type found on the surface of the epithelial fields, but neutral
The remaining 5–10% of the bacterial population repopulates the mucins can also be detected. Although sialomucin can be found
light organ. The consequence of this phenomenon is that both the on the surfaces of the appendages of squid maintained in filter-
bacteria and their host experience a changing environment daily. sterilized seawater, no shedding of mucus occurs. Mucus shed-
In this section, we describe some of the known molecular details ding could be induced by the addition of peptidoglycan (PG)
that correspond to the rhythm of the squid’s biology, including but not lipopolysaccharide (LPS) (Nyholm et al. 2002). Mucus
the early events specifying colonization and subsequent daily secretion also could be induced by PG-coated beads too large to
events that influence the interaction between the partners. enter the light organ, indicating that entry is not necessary and
Detailed studies of the initiation of colonization revealed that a receptor for PG must be present on the surface of the light
a surprising fact: within the first hour following hatching (in organ. These data indicate that the early permissive period is not
fact, as early as 20 min), the light organ is permissive to entry by essential for induction of mucus secretion, if PG is present in the
both V. fischeri and non-V. fischeri bacteria, as well as other seawater. In contrast to several developmental events discussed
similarly sized particles (> Fig. 20.4a) (Nyholm et al. 2002). below that require V. fischeri and also involve PG, this induction
Both Gram-negative and Gram-positive bacteria with sizes of of mucus secretion occurs even if V. fischeri is absent from the
1 mm in diameter could be observed in the crypt spaces in the seawater, indicating that the squid are simply sensing the pres-
first hour. Fluorescent beads with a 1-mm diameter could also be ence of bacteria.
Because the onset of mucus secretion coincides with the begin- This tight control over initiation of symbiosis begs the ques-
ning of the nonpermissive period, experiments were undertaken to tions, how many cells are necessary to initiate colonization and
determine if mucus secretion causes the block to particle entry. how many different V. fischeri cells can successfully colonize the
Newly hatched juvenile squid were exposed to PG for 3 h to induce light organ of a single squid? These questions were experimen-
mucus secretion. These animals were then exposed to GFP-labeled tally addressed in a series of studies. Early studies indicated that
nonsymbiont V. parahaemolyticus, and the number of animals squid could contain more than one strain. For example, bacteria
with these bacteria in their crypts determined. The results indi- with different plasmid profiles could be isolated from the light
cated that reduced numbers of animals with bacteria could be organ of a single field-caught adult animal (Boettcher and Ruby
detected relative to animals that had not been exposed to PG and 1994). In addition, when a mixture of two strains was used to
thus not shedding mucus. The authors of this study concluded inoculate juvenile squid, some colonized animals contained
that mucus secretion contributes to, but is not wholly responsi- both strains, indicating that multiple strains could colonize
ble for, the block to the permissive period (Nyholm et al. 2002). (Lee and Ruby 1994a). Furthermore, a marked strain could be
The production of mucus promotes the ability of V. fischeri introduced into a colonized animal, albeit at a very low fre-
to aggregate on the surface of the light organ and subsequently quency (Lee and Ruby 1994a); likely the poor efficiency of
to enter and colonize (> Fig. 20.4a). Once V. fischeri has colo- superinfection resulted from decreased mucus shedding and
nized, however, mucus shedding is downregulated: the amount loss of the ciliated surface appendages.
of mucus secreted from 72-h aposymbiotic animals was signif- A subsequent study evaluated how different numbers of
icantly greater than that of 48-h colonized animals (Nyholm bacteria impacted colonization proficiency. Generally, at differ-
et al. 2002). Furthermore, V. fischeri could aggregate on the ent inoculation dosages, the percentage of squid that had asso-
light organs of aposymbiotic animals upon exposure at any ciated bacteria at an early time point (3 h) was similar to the
point during the first 4 days, but could not aggregate on the percentage of squid that ultimately became colonized (McCann
light organs of 48-h symbiotic animals, which shed relatively et al. 2003). Inoculation with as few as 250 V. fischeri cells in 4 ml
little mucus (> Fig. 20.4a). Presumably, once colonization is during a short (3-h) period of time was sufficient to promote
achieved, there is no longer a need for mucus to promote colonization of 50% of the animals, indicating that colonization
bacterial aggregation, and thus it is downregulated; this decrease by V. fischeri is an efficient process (McCann et al. 2003). At
in mucus production likely also restricts colonization by inoculation levels above 1,000 bacteria, colonization occurred
undesired species. In further support of the relationship between 100% of the time. With increasing doses of bacteria, the effi-
colonization and mucus shedding, symbiotic animals that were ciency of colonization increased, as determined by the decreas-
cured of their bacteria through antibiotic treatment exhibited an ing time to onset of luminescence, a trait that is governed by cell
increase in mucus shedding, which once again promoted density. However, beyond a certain point, no further increase in
V. fischeri aggregation. It should be noted that cured animals, efficiency was obtained with increasing numbers, suggesting that
while able to shed more mucus and permit aggregation, the process rather than the number of bacteria becomes limiting.
exhibited reduced levels of aggregation, likely due to the reduced The same study asked whether three strains of V. fischeri
numbers of mucus-secreting epithelial cells resulting from apo- could simultaneously colonize a single squid (McCann et al.
ptosis and regression of the ciliated appendages (> Fig. 20.4a), 2003). The strains differed only by distinct antibiotic markers
developmental events that occur in the same time frame that did not substantially impact colonization proficiency of
(described below) (Nyholm et al. 2002). Finally, the deep crypt a single strain. At a low inoculation dose (500 cells), all colonized
spaces of colonized animals contained increased mucus com- squid contained a single strain. When the dosage was increased
pared to uncolonized animals (Nyholm et al. 2002). This effect to 5,000 cells, most squid contained two strains but not three.
was opposite to the downregulation that occurs on the light- Finally, at high doses (16,000 and 27,000), a significant percent-
organ surface, indicating that mucus secretion in the crypts may age of squid contained all three strains. Thus, although multiple
promote symbiosis. strains can co-colonize, it appears that, at levels similar to those
V. fischeri also downregulates the production by the host of found in nature (Lee and Ruby 1992), only one or two bacteria
nitric oxide (NO), which is relatively high in uncolonized hatch- generally colonize a single animal.
lings (Davidson et al. 2004). NO synthesis carries both a cost to A subsequent study evaluated co-colonization using two
produce and a risk (of oxidative damage) to the host. By shutting strains that differed by a fluorescent tag (red or green fluorescent
down mucus and regressing the appendages, the squid has protein) and visual examination of crypt colonization using
greatly reduced the opportunity for further colonization and, epifluorescence microscopy (Dunn et al. 2006). When squid
potentially, can now relax its defenses. Sustained activation were inoculated with moderate doses of the two strains (for
might ultimately jeopardize maintenance of symbiont coloniza- a total of 2,000–7,000 CFU/ml), most animals were colonized
tion, and thus, this change in NO may reflect an accommodation with both strains. Interestingly, however, most animals with
for V. fischeri. Alternatively, it might reflect a signaling function mixed infections contained pockets of either red or green fluo-
for NO (see below). Regardless, it is clear that the result rescence, and only rarely did light organs contain a region with
of colonization is to decrease subsequent attachment and super- a mixture of red and green cells (> Fig. 20.4b). These data
infection. These phenotypes demonstrate the influence of further support the idea that a few cells initiate colonization,
V. fischeri on the biology of its host. and even when multiple strains colonize, they often become
segregated within the light organ, presumably because it is often contributes to a drop in light intensity, it is insufficient to fully
only a single cell that initially reaches each deep crypt and account for the observed patterns, as light emission decreases
becomes the dominant colonist there. steadily over a number of hours prior to the expulsion event.
These two experimental studies, which suggest that low Another mechanism by which the squid controls light emis-
numbers of V. fischeri cells initiate colonization, are consistent sion is by physically blocking it. It is clear that adult animals can
with a subsequent analysis examining the population structure direct, control, and conceal the light produced by their bacterial
of V. fischeri. Wollenberg and Ruby (2009) used an extensive partner using a muscle-controlled ink sac, along with lens and
PCR-based analysis along with phenotypic analyses and model- reflector tissues (McFall-Ngai and Montgomery 1990). These
ing to evaluate the number of different strains of V. fischeri in tissues direct the light downward to match the downwelling
light organs of field-caught animals. These analyses support the moonlight and modulate light emission under a variety of nat-
prediction that one or two bacteria colonize each crypt, resulting ural environmental conditions (e.g., full moon and new moon),
in the mixed population structure found within the adult light thus permitting the squid to use counterillumination as
organ (Wollenberg and Ruby 2009). a defense mechanism (Jones and Nishiguchi 2004). However,
As mentioned earlier, E. scolopes expel ~90% of their bacte- the light organs of newly hatched juvenile squid do not seem to
rial symbionts every dawn (Boettcher et al. 1996; Graf and Ruby have sufficiently developed accessory tissues to account for the
1998; Lee and Ruby 1994b; Ruby and Asato 1993). Intriguingly, observed diel rhythm (Boettcher et al. 1996; McFall-Ngai and
for juveniles, different crypts are emptied with different efficien- Montgomery 1990; Montgomery and McFall-Ngai 1993). Thus,
cies: bacteria are expelled from deep crypt 1 to a greater extent it was speculated that another mechanism must be in place to
than crypt 2, with crypt 3 being least effectively emptied (Sycuro account for the daily modulation of bioluminescence.
et al. 2006). Sycuro et al. (2006) suggest two possibilities for The currently favored model is that the level of specific
these results: the expulsion efficiency (1) reflects the relative luminescence varies over the daily cycle due to changes in
developmental maturity of the crypts or (2) is due to position oxygen levels provided to the symbiont by the host (Boettcher
effects (the muscle contraction is less effective in the area where et al. 1996). Luminescence is an oxygen-dependent reaction, and
crypt 3 resides). These features of the juvenile could impact the thus decreased oxygen availability could lead to decreased lumi-
identity and number of specific symbiont strains over time. nescence. The finding that newly released bacteria exhibit a spike
Indeed, one study determined that a specific mutant strain of in their levels of light production is consistent with the idea that
V. fischeri was preferentially expelled from the light organ in oxygen is more readily available in seawater and, as a limiting
mixed colonization experiments (Millikan and Ruby 2004). reagent in the light organ, can rapidly change the specific lumi-
Understanding the dynamics of crypt expulsion and how it nescence level. Regardless of the cause, changes in luminescence
relates to colonization competitiveness and persistence is an are part of the daily rhythm of the symbiosis.
important area of future investigation. A series of studies has investigated the influence of V. fischeri
Linked to but separate from the expulsion event is another and the daily rhythm on host transcription and protein produc-
rhythm established in E. scolopes, the variation of bacterial tion. One of these studies generated a set of 11 expressed tag
bioluminescence over the day/night cycle. Luminescence sequence (EST) cDNA libraries from juvenile animals at differ-
increases and is highest in the hours preceding darkness; light ent times and either with or without V. fischeri cells (colonized or
production is as much as 100-fold lower at other times uncolonized) (Chun et al. 2006). A large number (46%) of the
(Boettcher et al. 1996). This rhythm of increasing and decreasing non-redundant set of sequences appeared to represent unique
light emission is disrupted if animals are kept in either constant sequences, as they had no matches in the database at the time.
light or constant darkness, indicating that it is a diel rhythm Perhaps the most significant finding from the initial description
rather than a circadian rhythm (Boettcher et al. 1996). Further- of this tool was that the biggest set of unique transcript frag-
more, the amount of specific bioluminescence, or the amount of ments was obtained from animals that had been colonized for
light produced on a per-cell basis, varies over the day/night cycle: 2 days (48 h). This finding agrees with previous proteomic work
when symbiotic luminescence levels are high prior to the onset showing that 48 h was the earliest time point at which differences
of darkness, the amount of symbiotic light per cell matches that in protein profiles between colonized and uncolonized animals
produced by newly released bacteria (Boettcher et al. 1996). could be detected (Doino Lemus and McFall-Ngai 2000). At this
However, in periods of low symbiotic luminescence, the specific point in colonization, numerous developmental changes are
luminescence is lower in symbiotic cells relative to newly occurring or are being induced by the bacteria, including apo-
released cells. These data indicate that the squid controls ptosis and regression, and crypt cell swelling (Nyholm and
(inhibits) light production or emission. The daily expulsion of McFall-Ngai 2004). The proteomic study also found that some
bacteria at dawn contributes to the change in light emission over proteins that appeared specifically in 48-h symbiotic animals but
the daily cycle: a peak of luminescence can be observed at the not in uncolonized animals were not symbiosis-specific, as the
transition from dark to light that is not observed for animals same proteins could be detected in uncolonized animals at
kept in constant light or dark conditions. The peak of light a later time point (96 h) (Doino Lemus and McFall-Ngai
corresponds to the release of bacteria into the seawater; follow- 2000). These data indicate that the bacteria accelerate certain
ing this expulsion event, the levels of luminescence by the squid host developmental events in addition to inducing symbiosis-
are clearly decreased. However, although expulsion clearly specific gene expression.
The EST database permitted the construction of a microar- In addition to light production and, likely, metabolism,
ray, which was used to probe changes in host gene expression in V. fischeri undergoes developmental changes that are reflected
response to the symbiosis. Specifically, a subsequent study inves- in the daily rhythm. Motility, which is necessary for V. fischeri to
tigated how host transcription was affected by colonization, enter and reach the deep crypts, appears unnecessary upon
luminescence, and the LuxI-produced autoinducer pheromone colonization as deep-crypt-localized V. fischeri have no flagella
that induces luminescence (as described in greater detail below) (Ruby and Asato 1993). In addition, V. fischeri cells undergo
(Chun et al. 2008). Perhaps not surprisingly, the biggest influ- a change in morphology: within 24 h of symbiosis, the cells have
ence on the response of the host was the presence of the bacteria, become smaller and rounder than culture-grown cells or sym-
with hundreds of transcripts altered in response to colonization. bionts within the first 12 h of symbiosis (Ruby and Asato 1993).
The host also responded to light production and, to a much The signals and genes responsible for inducing these changes
lesser extent, the presence of autoinducer. Further investigation remain unknown.
of the transcriptional responses and the impact of those changes In summary, the light-organ symbiosis is dynamic. The bac-
on host activities will provide important insights into host teria and their host influence each other’s gene expression and
colonization. induce developmental events, promoting, among other things,
A second microarray study analyzed gene expression of an exquisite specificity in partner selection and reducing com-
both V. fischeri and its host at 6-h intervals to capture the petition by others, including late-arriving V. fischeri. For the
changes that drive or derive from the squid’s daily rhythm host, some changes, once initiated, are irreversible, while others
(Wier et al. 2010). For the host, about 10% of genes present in require the continuous presence of the bacteria. Numerous
the array (a ~14,000 gene EST library) exhibited changes in events occur on a daily cycle driven by a number of factors,
expression. The 6-h intervals on either side of dawn showed including bacterial expulsion, available nutrients and (probably)
the greatest differences in host gene expression, notably available oxygen, and bacterial growth. This dynamic nature
including changes in transcripts encoding proteins with cyto- must be considered when assessing the requirements for specific
skeletal functions. In the interval prior to dawn, there was a bacterial traits, which may be important at one stage or one
substantial upregulation in expression of this suite of genes, temporal period, but not another. This is one example where the
while in the period after dawn, they were substantially laboratory ‘‘batch culture’’ may actually somewhat reflect
downregulated, suggesting cytoskeletal remodeling was induced a natural process in the wild.
then decreased in the intervals bracketing dawn. An examination
of the crypt epithelium revealed that structural changes indeed
occurred right around dawn (Wier et al. 2010). Throughout Specificity and Host Defenses
most of the 24-h period, the host epithelial cells appear ‘‘nor-
mal’’: the cells are highly polarized and have microvilli on the The discrete localization of the V. fischeri light-organ infection
surfaces that interact with the symbionts. However, around and the inability of other bacteria to colonize this tissue dem-
dawn, the cell surfaces appear effaced, and portions appear onstrate an exquisite level of control by the host. As further
to be released into the crypts as vesicles (Nyholm and McFall- evidence that host tolerance of the symbiont is regulated, we
Ngai 1998; Wier et al. 2010). Within a few hours, however, the have observed that when juvenile squid that are colonized by
normal appearance is restored. Thus, there is a daily cycle of V. fischeri become nutritionally stressed, the animals are able to
structural change in the host epithelium coincident with sym- entirely clear the V. fischeri infection (Stabb, unpublished data).
biont expulsion. Thus, while the light organ is receptive to infection and able to
The same microarray experiment confirmed that the symbi- support the rapid growth of V. fischeri, the squid are able both to
onts also changed their gene expression: 17% of genes changed prevent other bacteria from colonizing and to keep the V. fischeri
during one of the four intervals, with the greatest period of symbionts themselves in check. The mechanisms of specificity
regulatory change occurring in the interval after dawn (Wier and control of the infection maintained by the host remain
et al. 2010). A deeper analysis indicated that genes involved in intriguing and somewhat mysterious, although a great deal is
the catabolism of chitin were upregulated in the period prior to now known. It appears that specificity is achieved through
dawn and downregulated after dawn and throughout the day. In multiple layers of enrichment, eventually ‘‘winnowing’’
contrast, genes involved in glycerol metabolism were V. fischeri symbionts away from unwanted interlopers (Nyholm
upregulated after dawn. Wier et al. (2010) proposed that host and McFall-Ngai 2004). The underlying mechanisms involve
vesicles, which become abundant around dawn, would be a rich physical barriers, physiological constraints, and immune func-
source of glycerophospholipids, from which glycerol could be tions that include both broad-spectrum antimicrobial com-
released (Wier et al. 2010). In support of the idea that the pounds and potentially a more microbe-specific population of
symbionts could incorporate fatty acids from host vesicles into macrophage-like hemocyte cells (> Fig. 20.5).
their membranes, symbiont cells contained dramatically differ- Cilia and mucus in the ducts may constitute an impediment
ent lipid profiles compared to cultured V. fischeri (Wier et al. to infection (Nyholm and McFall-Ngai 1998). A sticky mucus
2010). Together, these data suggest that the metabolism of also facilitates adherence to the outside of the light organ by
V. fischeri undergoes dynamic changes over the course of each planktonic cells, but such adherence could constitute a barrier to
day, depending on the availability of nutrients. the movement required to traverse the ducts. Once the
other marine bacteria have similar iron-scavenging systems.

Thus, nutrient control is likely, at best, a mechanism to enrich
for ongoing V. fischeri colonization.
Instead, it seems likely that the major control over the
symbiont population and specificity is exerted by the host
immune system. Naturally, this topic has been an area of great
research interest, and the role of the E. scolopes innate immune
system in the symbiosis was recently reviewed (McFall-Ngai
et al. 2010). Discovery and elucidation of E. scolopes immune
functions have been accelerated by the generation of EST librar-
ies (described above) that profile the host transcriptome (Chun
et al. 2008) and the characterization of the host light-organ
proteome (Schleicher and Nyholm 2011). For example, analysis
of ESTs led to the discovery of components of an immunological
complement system, followed by experimental determination
that complement C3 protein is expressed on the apical surface
of light-organ epithelial cells (Castillo et al. 2009). Research can
now address whether the complement system helps direct
immunological responses that maintain specificity or constrain
V. fischeri. There is also evidence that the squid are capable of
. Fig. 20.5 producing antimicrobial peptides (Nyholm and McFall-Ngai
Host defense by E. scolopes. Hemocytes, part of the innate defense 2004), which may have an immune function that is either
of E. scolopes, are shown binding to V. fischeri bacteria (This image broad-spectrum or weighted toward the control of non-
was taken by Andrew Collins and generously provided by Spencer V. fischeri.
Nyholm) Among the potential antimicrobial components of the squid
innate immune response, the most thoroughly studied to date
have been reactive oxygen species (ROS). Recent analysis of the
symbionts are inside the crypts, junctions between epithelial host proteome suggested a number of highly expressed proteins
cells probably form a physical barrier that keeps them constrained are involved in producing ROS (Schleicher and Nyholm 2011).
to the light organ (Nyholm and McFall-Ngai 1998). Although For example, the host apparently encodes a number of putative
the V. fischeri genome appears to encode a Zot toxin that ROS-generating peroxidases, including the halide peroxidase
theoretically could disrupt tight junctions (Ruby et al. 2005), (HPO) described below. In the same study, it was found that
gfp-labeled V. fischeri have not been observed escaping the episymbiotic host and V. fischeri cells contained numerous putative
thelium-lined crypts of the light organ. The squid’s active daily antioxidant proteins. For example, one of the most abundant
expulsion of most light-organ lumen contents provides proteins in symbiotic V. fischeri cells was AhpC, a predicted alkyl
a physical limitation on bacterial overgrowth otherwise hydroperoxide reductase (Schleicher and Nyholm 2011). These
breaching the crypt epithelium, and it would presumably also recent data are consistent with the longstanding hypothesis that
enrich for bacteria that have the ability to avoid being expelled. oxidative stress is a hallmark of the light-organ environment,
As described below, V. fischeri appears to have a variety of pili and they provide new targets for future investigations.
and adhesins that may have coevolved with the host for this Interest in E. scolopes ROS production began with the dis-
purpose (Browne-Silva and Nishiguchi 2008; Ruby et al. 2005). covery that the squid expresses HPO in the light organ, presum-
The daily cycle of regrowth of bacteria in the light organ may ably producing the antimicrobial ROS hypochlorous acid
give the host an additional mechanism for maintaining the (HOCL). Peroxidase-encoding transcripts were among the
specificity of its symbionts. For example, by providing or with- most abundant found in early cDNA libraries from E. scolopes,
holding certain nutrients, the growth of V. fischeri may be and the HPO transcript was among the first expressed genes
favored over other bacteria that gain access to the light organ. discovered from the light organ (Tomarev et al. 1993). Further
Such a mechanism remains speculative, and it would be difficult studies confirmed the biochemical similarity of the squid-
if not impossible to test the growth rate of different bacterial encoded halide peroxidase to mammalian peroxidases, includ-
species within the physiological parameters of the light organ in ing production of hypohalous acid from halide ions and H2O2,
the absence of other (e.g., immunological) specificity factors. It as well as its presence in the light organ (Weis et al. 1996; Small
seems unlikely, however, that this could be the major factor in and McFall-Ngai 1999). Given the high level of chloride ions in
maintenance of specificity: the contents of the light organ appear seawater, it seems likely that the main relevant product of HPO
complex and include substrates (e.g., peptides and amino acids) is HOCl, an effective broad-spectrum antimicrobial used com-
that could support the growth of numerous other bacteria. mercially as a disinfectant. The HPO gene in E. scolopes is more
Although iron appears to be limiting, and iron-uptake machin- highly expressed in tissues that are exposed to bacteria, includ-
ery may be an important colonization factor for V. fischeri, many ing gills as well as the light organ (Small and McFall-Ngai 1999).
Like other bacteria, V. fischeri is sensitive to HOCl, suggesting By limiting iron uptake, symbionts might limit the oxidative
that HPO represents a broad-spectrum antimicrobial that keeps stress generated by Fenton chemistry when H2O2 and iron
bacteria in check rather than a light-organ-specific mechanism are combined (Wang et al. 2010a). Before leaving the topic of
for selecting V. fischeri. Interestingly, recent studies show that NO, it should be noted that it may play an additional symbiotic
HPO is localized in the host hemocyte cells (Heath-Heckman role in addition to functioning as an antimicrobial oxidant: NO
and McFall-Ngai 2011). has signaling functions in many higher organisms, and as
Given that H2O2 is a substrate for HPO, it seems likely that described below, it may be part of the morphogenic develop-
this ROS is produced in the light organ as well, perhaps via a host mental program in the light organ stimulated by V. fischeri
respiratory burst. Like many bacteria, V. fischeri encodes (Altura et al. 2011).
a catalase (KatA) that converts H2O2 to water and oxygen, but In addition to producing broadly antimicrobial molecules,
the periplasmic location of V. fischeri is unusual and suggests E. scolopes has a cellular immune response involving hemocyte
a key role in detoxifying H2O2 originating from an external cells (> Fig. 20.5). These immune cells may play a key role in
source (Visick and Ruby 1998). The relatively high catalase maintaining the specificity of the interaction with V. fischeri
activity in V. fischeri likewise points to a role beyond coping (Nyholm and McFall-Ngai 1998; Nyholm et al. 2009). As is the
with endogenous metabolic production of H2O2 (Visick and case in other cephalopods, E. scolopes appears to have a single
Ruby 1998). A V. fischeri katA mutant was able to colonize the type of hemocyte, which circulates in the blood and moves
E. scolopes light organ to wild-type levels when presented as a throughout the animal. Like mammalian macrophages, these
clonal inoculum but was outcompeted by the wild type in mixed hemocytes can bind, engulf, and kill bacterial cells. E. scolopes
competitive infections, indicating that catalase contributes to, hemocytes have been found in the blood as well as in the light-
but is not required for, colonization of the host (Visick and Ruby organ crypts and the sinuses of the ciliated epithelial appendages
1998). Analysis of the V. fischeri genome suggests that while katA (> Fig. 20.2c) (Koropatnick et al. 2004; Nyholm and McFall-
encodes the bacterium’s only catalase, there are other mecha- Ngai 1998). Within the light-organ crypts of newly hatched
nisms for coping with oxidative damage. For example, there are juveniles, the hemocytes have been observed with internal bac-
three methionine sulfoxide reductase genes that presumably repair teria, presumably engulfed; however, in adult animals, the
proteins damaged by H2O2 (Flores and Stabb, unpublished data). hemocytes were seen surrounded by densely packed V. fischeri
Redundancy in the V. fischeri response to this and other ROS cells but had not phagocytosed them (Nyholm and McFall-Ngai
may explain why mutants lacking single oxidative-response 1998; Nyholm et al. 2009). These data suggested that the hemo-
enzymes are not more severely attenuated in colonization. cyte cells can ignore V. fischeri and that this may be a trait
Another oxidant produced by E. scolopes is nitric oxide acquired as the animals develop. When removed from the
(NO). Both NO and NO synthases (NOS) are produced squid, the macrophage-like hemocytes bound and phagocytosed
throughout light-organ tissues, and both are found in mucus V. fischeri less frequently than they did other marine bacteria
secretions that contact V. fischeri symbionts (Davidson et al. (Nyholm et al. 2009). Binding seemed to be the key rate-limiting
2004). Although both V. fischeri and other species aggregate in step in this process, as bound V. fischeri were as likely to be
host-derived mucus on the surface of the light organ, V. fischeri is phagocytosed as another bacterium (Nyholm et al. 2009).
somehow enriched (Nyholm et al. 2000, 2002; Nyholm and In an interesting twist, Nyholm et al. (2009) also found that
McFall-Ngai 2003). NO-scavenging molecules increased the exposure to V. fischeri was critical for maintaining hemocyte
number of either V. fischeri or nonsymbionts in such aggregates specificity (Nyholm et al. 2009). When squid were cured of
on the light-organ surface (Davidson et al. 2004). Although their V. fischeri symbionts with antibiotics, hemocytes from
nonnative bacteria were still unable to colonize the light-organ these symbiont-free animals became five times more effective
crypts, the results suggest that NO limits bacterial growth in the at binding V. fischeri while their affinity for V. harveyi or Vibrio
aggregates and could play some role in the enrichment seen at parahaemolyticus was unchanged (Nyholm et al. 2009). More-
this stage (Davidson et al. 2004). over, hemocyte binding to V. fischeri was similarly high in hemo-
The transcriptional response of V. fischeri to NO was recently cytes isolated from colonized or naı̈ve animals when the target
elucidated (Wang et al. 2010a), leading to intriguing discoveries V. fischeri strain was a mutant lacking the major outer membrane
and new research directions. For example, a heme-independent protein OmpU (Nyholm et al. 2009). These data suggest that
and NO-resistant alternative oxidase gene (aox) is controlled by E. scolopes hemocytes learn to discriminate V. fischeri from other
the NO-responsive regulator NsrR and is upregulated in bacteria and adapt to preferentially bind nonsymbionts through
response to NO (Dunn et al. 2010). Aox allows aerobic respira- an OmpU-dependent mechanism. Interestingly, it was recently
tion to continue when other oxidases in the cell are inhibited by reported that Vibrio splendidus OmpU mediates adhesion to and
NO, which may impart a competitive advantage to V. fischeri invasion of oyster hemocytes (Duperthuy et al. 2011). While the
over the majority of other Vibrio species, which lack aox (Dunn two OmpU-mediated phenomena seem opposite to each other,
et al. 2010; Spiro 2010). NO production also results in the the underlying processes involved may reveal quite parallel
H-NOX-dependent downregulation of iron acquisition (Wang mechanisms once they are understood.
et al. 2010a). While this response may not relate directly to NO Another observation of E. scolopes hemocytes that may
resistance, it could reflect NO being used by V. fischeri as an have widespread importance is the discovery of chitin and
indicator of other ROS it will soon encounter, specifically H2O2. endogenous chitin synthesis within these immune cells
Stage 0 Stage 1 Stage 2 Stage 3
aa
p
Epi
pa
cr
SEM
aa
pa
. Fig. 20.6
Development of the juvenile light organ. Epifluorescence (Epi) and scanning electron microscopy (SEM) images of the stages of
regression of ciliated epithelial appendages. Upon exposure to peptidoglycan or symbiosis-competent V. fischeri, the ciliated
appendages present on newly hatched squid (stage 0) undergo thinning (stage 1) and progressive shortening (stages 2 and 3), until they
are lost (stage 4, not shown). aa anterior appendage, pa posterior appendage, p pore, and cr ciliated ridge (These images were previously
published Adin et al. (2009))
(Heath-Heckman and McFall-Ngai 2011). This was found to be mimicked using bacterially derived molecules. If kept in water
a common property of invertebrate hemocytes that is lacking in free of V. fischeri, E. scolopes can be raised to adulthood without
their vertebrate counterparts (Heath-Heckman and McFall- the light organ becoming infected or bioluminescent (Hanlon
Ngai 2011). V. fischeri can metabolize chitin and ferment its et al. 1997). Such aposymbiotic animals are healthy and develop
N-acetylglucosamine monomers, but it remains to be determined normally in most respects, including the lens and reflective
what, if any, role hemocyte-derived chitin plays in E. scolopes tissue of the light organ (Claes and Dunlap 2000). However,
immunity or support of V. fischeri growth. From multiple per- specific developmental events fail to occur in the absence of
spectives, the biology of E. scolopes hemocytes appears worthy of V. fischeri infection, including, most dramatically, the regression
further investigation likely to reveal elements unique to the of the ciliated fields on the light organ. At the time of hatching,
V. fischeri-E. scolopes symbiosis as well as phenomena more the ciliated fields begin shedding mucus, which as described
broadly applicable to invertebrate-bacteria interactions. above helps facilitate infection; however, once the light organ is
infected, mucus shedding ceases (Nyholm et al. 2002), and this
cessation is followed by regression of the structures themselves.
Host Development and Bacterial Signals Over the course of 4–5 days postinfection, these infection-
promoting structures are completely lost in infected animals
Host Development (> Fig. 20.6) (Montgomery and McFall-Ngai 1994; Foster and
McFall-Ngai 1998; McFall-Ngai and Ruby 1991). This morpho-
V. fischeri is the lone bacterial species colonizing the E. scolopes logical change is accompanied by apoptosis in the epithelial cells
light organ, and infection by V. fischeri is required to trigger of the ciliated fields and infiltration of host hemocytes into the
developmental changes in the host, some of which can be sinuses of the ciliated appendages (Koropatnick et al. 2004,
2007). None of these developmental events take place in does reappear in animals that are infected and then cured of
aposymbiotic animals. V. fischeri (Nyholm et al. 2002).
V. fischeri also triggers more subtle developmental effects in Taken together, the results of several early experiments indi-
the light organ. For example, the ducts leading from the light- cate a complex pattern of light-organ development that includes
organ surface to the crypts constrict (Kimbell and McFall-Ngai symbiont-dependent and symbiont-independent programs,
2004), and the cells lining the duct become more homogenous some of which require only transient exposure to V. fischeri.
and filled with inclusions (Claes and Dunlap 2000). The epithe- Moreover, although Claes and Dunlap (2000) noted that the
lial cells lining the crypts swell (Montgomery and McFall-Ngai tissues developmentally affected by V. fischeri all come in contact
1994) with a proliferation of microvilli on their surfaces with symbiotic cells, at least some developmental events appear
(Lamarcq and McFall-Ngai 1998), and there is an apparent to involve symbiont induction remotely. Specifically, the ciliated
increase in mucus secretion inside the crypts themselves fields on the light-organ surface are exposed to V. fischeri, but
(Nyholm et al. 2002). Additional V. fischeri-dependent molecu- mutants unable to colonize the crypts still contact these cells on
lar events, including a downregulation of NO synthase the surface of the light organ without triggering their regression.
(Davidson et al. 2004), have been observed but not yet clearly Once the crypts become packed with V. fischeri cells, regression
linked to developmental and physiological processes (Doino of the ciliated appendages advances, even though symbionts in
Lemus and McFall-Ngai 2000; Kimbell and McFall-Ngai 2003; the crypts are several cell layers away.
Chun et al. 2008). Many of these developmental events triggered
by V. fischeri, as well as their timing, were reviewed by Nyholm
and McFall-Ngai (Nyholm and McFall-Ngai 2004). Bacterial Signals That Influence Host
In general, most symbiont-induced developmental changes Development
in the E. scolopes light organ can be rationalized in terms of this
organ’s two temporally distinct functions—first, to become The observation that infection with V. fischeri triggers develop-
infected with V. fischeri and then later to support and control mental programs and morphological changes in the E. scolopes
symbiotic bioluminescence. For newly hatched aposymbiotic light organ has prompted interest in understanding the specific
squid, acquiring symbionts from a dilute environment is mechanisms and bacterial signals involved in these processes.
a numerically daunting task that is facilitated by the ciliated Most research has focused on the involvement of three bacterial
cells and the mucus they secrete (Nyholm et al. 2000, 2002). factors in stimulating host development: bioluminescence, LPS,
However, once the squid are colonized by appropriate V. fischeri and PG. Both LPS and PG can be categorized as microbe-
symbionts, the infection-promoting properties of the ciliated associated molecular patterns (MAMPs), and they have intrigu-
fields become unnecessary and could perhaps be a liability if ingly parallel roles in several host-microbe interactions, both
pathogenic infections were facilitated. Thus, programmed cell pathogenic and mutualistic. MAMPs are relatively conserved
death and regression of the ciliated fields and constriction of the among bacteria, and hosts ranging from plants to animals have
ducts may serve to prevent further infection beyond the initial evolved mechanisms for MAMP recognition. Although the sig-
colonization with a mutualistic V. fischeri symbiont. The devel- naling roles of LPS, PG, and bioluminescence were discovered
opmental events inside the crypts, cell swelling and microvillar separately, and they each have distinct influences on the host,
proliferation, may serve to increase the surface area at the their effects appear intertwined and difficult to deconvolute.
symbiont-host interface and promote the efficient exchange of This is well illustrated by their combined influence on the
metabolites. developmental program associated with regression of the
The symbiont-triggered developmental events in E. scolopes ciliated fields.
appear to result from multiple distinct signaling pathways. Some It was first discovered that LPS from the bacteria triggers
changes are reversible if E. scolopes is cured of its symbionts with apoptosis in the ciliated field but not regression of this structure
antibiotics (Lamarcq and McFall-Ngai 1998; Nyholm et al. (Foster et al. 2000). Specifically, the lipid A moiety of LPS had
2002), whereas other events cannot be stopped once they are this effect—an intriguing finding, given that lipid A triggers
set in motion (Doino and McFall-Ngai 1995). For example, the responses in other host-microbe systems. Lipid A from V. fischeri
swelling of crypt epithelial cells can be reversed by curing the has several modifications including a novel acylated
symbionts, and microvillar proliferation on these cells does not phosphoglycerol moiety (Phillips et al. 2011) which could con-
progress and may even reverse somewhat if symbionts are cured tribute to specificity, and at least one of the lipid A modifying
(Lamarcq and McFall-Ngai 1998). In contrast, regression of the enzymes, designated HtrB1, appears to contribute to coloniza-
ciliated epithelial fields does not require persistent colonization tion efficiency early in infection (Adin et al. 2008a). Moreover,
after about 12 h postinoculation (Doino and McFall-Ngai 1995). a mutant of V. fischeri defective for the response regulator GacA
Regression proceeds in animals that have been cured of symbi- has an altered LPS structure and is impaired in stimulating
onts, and the ciliated epithelial structures do not grow back. apoptosis and regression of the ciliated appendages on the
Thus, it seems that the programmed loss of infection-promoting light organ (Whistler et al. 2007). Apoptosis in the ciliated fields
structures is set in motion early during infection and does not of the light organ does not specifically require V. fischeri lipid A,
require constant colonization. In one exception to this, the as LPS and lipid A from other bacterial species will also elicit
infection-promoting mucus secretion of the ciliated structures apoptosis in these cells; however, the structure of lipid A does
appear to influence its bioactivity in such assays (Foster et al. the apoptotic developmental program, but the combined effect
2000). It is tempting to speculate that a distinctive lipid of TCT and LPS likely represents the genuine symbiotic signal in
A structure is recognized by a host receptor(s), but it should be a natural infection.
kept in mind that in natural infections, alterations in LPS struc- The recognition of MAMPs, including PG and LPS, by
ture could also influence colonization levels and membrane various hosts has been the subject of many studies, and homo-
integrity, thereby affecting how much lipid A is presented to logs of known MAMP receptors and MAMP-responsive proteins
the host. Thus, structural elements of lipid A could affect either have been found in E. scolopes. These MAMP-associated host
direct interactions with a host receptor or the amount of lipid genes include components of the NF-kB pathway (Goodson
A presented to host receptors. et al. 2005), specific PG recognition proteins (PGRPs) (Troll
In reporting the effects of LPS on the ciliated fields, Foster et al. 2009a, b), and LPS-binding proteins (Krasity et al. 2011).
et al. (2000) noted that there must be at least one other signal Although these MAMP-associated genes and proteins are
and suggested potential candidates. One of these, PG, proved to orthologs of those in other organisms, they appear to have
be a second critical MAMP. PG stimulates mucus shedding by novel functions in E. scolopes. Notably, EsPGRP1 has an unprec-
the ciliated epithelial fields (Nyholm et al. 2002), and it dramat- edented nuclear localization, and infection by V. fischeri or
ically affects morphogenesis and regression of the ciliated fields treatment with TCT triggers loss of EsPGRP1 from host nuclei
(Koropatnick et al. 2004). Interestingly, it was also discovered (Troll et al. 2009a). In contrast, EsPGRP2 is exported from host
that V. fischeri sheds a particular PG monomer that is usually cells in association with light-organ mucus shedding; although
recycled and kept within cells (Koropatnick et al. 2004). In two similar to the dynamics of EsPGRP1, the export of EsPGRP2 is
other Gram-negative bacteria, Bordetella pertussis and Neisseria stimulated by TCT or infection with V. fischeri (Troll et al.
gonorrhoeae, the same molecule is released from cells, and in 2009b). EsPGRP2 has an amidase activity capable of degrading
each case, this molecule affects ciliated host cells. Indeed, this PG TCT, and it is exported into the light-organ crypts colonized by
monomer is called tracheal cytotoxin (TCT), because it triggers V. fischeri, leading to the suggestion that it may attenuate this
the death of ciliated host airway cells in Bordetella infections, potentially toxic bacterial MAMP signal after it is received (Troll
giving rise to its ‘‘whooping cough’’ symptoms. Although et al. 2009b).
V. fischeri appears capable of TCT recycling, the combined activ- The combined effects of TCT and LPS mimic many aspects
ity of lytic transglycosylases apparently results in relatively large of natural V. fischeri infection but as Troll et al. noted (2009a),
amounts of free TCT being released from cells (Adin et al. 2009). they fail to completely duplicate it. An intriguing third signal
Using a mutant with decreased TCT release, Adin et al. (2009) may be the bioluminescence produced by V. fischeri. The squid
provided evidence that the advantage of TCT shedding appear capable of perceiving bioluminescence in the light organ,
for V. fischeri may be that by triggering regression of the infec- and components of photoreceptor-mediated signaling are pre-
tion-promoting ciliated fields, an infecting strain thereby min- sent in light-organ tissue (Tong et al. 2009). Moreover, the host
imizes the chances of competition from later infecting V. fischeri transcriptional profile varies depending on whether or not
strains. V. fischeri symbionts are bioluminescent (Chun et al. 2008).
Developmental changes appear to require the combined Although it is difficult to know to what extent these transcrip-
effect of the TCT and LPS signals. Koropatnick et al. (2004) tional changes are due to the perception of light itself or some
found that TCT could stimulate hemocyte trafficking into the other physiological change in dark symbionts related to the lack
sinuses of the ciliated appendages as well as their regression, but of luciferase activity, the presence of photoreceptors and the
curiously did not cause apoptosis in these cells. When TCTor PG importance for the animal to match the intensity of its light-
was combined with LPS, however, the two had a synergistic organ luminescence to the environment make a compelling case
effect on apoptosis and regression of the ciliated appendages for light itself being perceived by the host. In any case, regression
that was very similar to that of a natural symbiotic infection of the ciliated fields, export of EsPGRP2, and trafficking of
(Koropatnick et al. 2004). Similarly, TCT and LPS together, but hemocytes to the ciliated appendages early in infection are all
not singly, led to a decrease in NOS activity and NO similar to attenuated relative to wild-type infections when squid are
that seen in animals infected with V. fischeri (Altura et al. 2011). infected by dark mutants (McFall-Ngai et al. 2011). These dif-
Interestingly, experiments with NOS inhibitors and NO donors ferences are apparent even during infection initiation when the
suggested that NO itself may be an additional key signaling dark mutant presumably is colonizing at similar levels as the
molecule in the apoptosis and morphogenesis associated with parent (Visick et al. 2000). Interestingly, among the host genes
TCT and LPS exposure (Altura et al. 2011). Such two- or three- differentially regulated in wild-type and dark mutant infections
part signals may contribute to the specific recognition of the are the PG recognition protein EsPGRP1 and a putative
correct symbiont species. LPS-binding protein (LBP1) (McFall-Ngai et al. 2011). Light
Somewhat in contrast to these synergistic effects of LPS and production induces transcription of the genes for these
TCT, one recent study suggested that TCT alone can in fact predicted MAMP receptors, as demonstrated by four- to fivefold
trigger apoptosis in the absence of LPS (Troll et al. 2009a). lower levels of these transcripts in a (luxA) luminescence
One explanation for this apparent discrepancy may be that mutant (Chun et al. 2008). These data suggest that light pro-
assays measuring different stages of apoptosis were used in duction signals the host to boost production of receptors for PG
these two studies. TCT alone may stimulate initial elements of and LPS.
Some of the developmental events occurring within the light biosynthesis [LuxAB can comprise 5% of the protein in bright
organ and triggered by V. fischeri are not as well understood as cells (Hastings et al. 1965)], ATP hydrolysis during RCHO
those described above. In particular, changes in the morphology recycling, and the consumption of oxygen and reducing equiv-
of epithelial cells lining the crypts and ducts, as well as prolifer- alents, which theoretically competes with energy recovery from
ation of microvilli, have not been studied to the same extent as aerobic respiration.
the developmental program associated with regression of the Primarily for energetic reasons, it has long been postulated
ciliated fields. This research focus probably reflects the relative that bioluminescence should slow the growth of V. fischeri.
difficulty in assaying changes in crypt and duct structure, which Consistent with this idea, some researchers found luminescence
experimentally usually involves fixation, sectioning, and obser- to be negatively correlated with growth in culture (Keynan and
vation by TEM. It is known that crypt epithelial cell swelling Hastings 1961; Hastings and Nealson 1977; Dunlap et al. 1995;
requires V. fischeri to be bioluminescent (Visick et al. 2000), but Pooley et al. 2004), although the undefined nature of the lumi-
microvillar proliferation, which is another process that can be nescence mutants in these experiments raised some doubts as to
reversed by curing symbionts, is unaffected by luminescence whether luminescence was the only variable. Recently, however,
(Lamarcq and McFall-Ngai 1998). The swelling of epithelial it was shown that defined mutants lacking luxCDABEG could
cells resembles a response to hypoxic stress and could be tied outgrow an isogenic parent under certain culture conditions,
to ongoing oxygen consumption by bioluminescence. Cell swell- such as in a carbon-limited chemostat (Bose et al. 2008).
ing could also represent a developmental response of the animal In contrast, bioluminescence is beneficial for V. fischeri col-
to light itself or to metabolic products related to the physiology onizing the E. scolopes light organ. To begin with, the advantage
of bioluminescence. Similarly, the proliferation of microvilli of luminescence for the host (e.g., antipredatory camouflage)
may relate to metabolic exchange. Although this is purely spec- would confer a fitness advantage on the bacteria, because as
ulative, it would be consistent with the requirement for meta- noted above, V. fischeri populations appear augmented by suc-
bolically active symbionts to trigger and maintain an effect. cessful symbiosis with the squid. However, the advantage for
V. fischeri of expressing bioluminescence in the symbiosis goes
deeper, as luminescence actually promotes successful coloniza-
Bacterial Genes and Phenotypes Involved in tion (Visick et al. 2000; Bose et al. 2008). Mutants lacking
Colonization bioluminescence are able to colonize the host, but do not persist
as well as their parental strains.
Bioluminescence and Pheromone-Dependent Several explanations have been proposed for how biolumi-
Regulation nescence might aid bacteria directly (Stabb 2005). For example,
it was proposed that bioluminescence stimulates photolyase-
As noted above, bioluminescence is the central contribution mediated DNA repair (Czyz et al. 2003); however, photolyase
V. fischeri makes to this symbiosis (> Fig. 20.7a, b), and it mutants are unimpaired in colonizing E. scolopes (Walker et al.
appears intimately involved in the host response to the bacteria 2006). As noted above, the host can apparently detect and
as well. V. fischeri has long been a model for studying bacterial respond to luminescence in its light organ (Chun et al. 2008;
bioluminescence, and the biochemistry, genetics, and regulation Tong et al. 2009), and it has been suggested that the squid might
of light production, as well as the symbiotic role of biolumines- impose sanctions on dark infections, thereby ensuring that it
cence, have all been active research topics. These areas are inter- receives bioluminescence in exchange for the nutrients it pro-
related, and together their study has had a widespread influence vides to V. fischeri. Other hypotheses have posited that the real
on our understanding of bacterial gene regulation and host- function of luminescence is either to burn oxygen or to consume
bacterium interactions. For example, luminescence demands excess reductant (Bourgois et al. 2001; Visick et al. 2000). The
a large energetic commitment, which explains why the process latter idea now appears inconsistent with the regulation of
is tightly regulated, and control of luminescence is accomplished luminescence (Bose et al. 2007; Lyell et al. 2010), but the
in part by a pheromone-mediated regulatory pathway, which model that luminescence confers an advantage by consuming
has become an archetype for understanding similar regulatory oxygen is intriguing: by consuming oxygen, V. fischeri might
mechanisms in numerous host-associated bacteria. generate hypoxic stress in the nearby host epithelium, attenuate
V. fischeri produces light when luciferase (LuxA and LuxB) ROS generation by the host, or render itself more resistant to
converts FMNH2, in the presence of O2, and an aliphatic alde- ROS. Luciferase’s high affinity for O2 (Km = ~35 nM) is consis-
hyde (RCHO) to FMN, releasing water and an aliphatic acid tent with a role in driving down ambient or intracellular oxygen
(> Fig. 20.7c) (Hastings and Nealson 1977; Tu and Mager 1995; levels (Bourgois et al. 2001). The ideas that the host sanctions
Ziegler and Baldwin 1981). LuxD produces RCHO, which is also dark infections or that the light-organ environment somehow
generated by LuxC and LuxE through recycling of the makes bioluminescence physiologically beneficial are also both
corresponding aliphatic acid (Boylan et al. 1989). LuxG and plausible and are not mutually exclusive.
other enzymes re-reduce FMN (Zenno and Saigo 1994; The lux genes underlying light production are regulated
Nijvipakul et al. 2008). The obvious energetic costs of this tightly and induced in the symbiosis. Key enzymes and regulators
process have stimulated curiosity in its regulation and useful- for light production in V. fischeri are encoded in an 8-kb DNA
ness for the bacteria. These costs include Lux protein locus (Engebrecht et al. 1983; Engebrecht and Silverman 1984)
a b
c LuxAB
FMNH2 + O2 + RCHO RCOOH + FMN + H2O + light
d
bioluminescence
“lux
box”
luxR luxI luxC luxD luxA luxB luxE luxG
LuxI SAM
acyl-ACP
LuxR N-3-oxo-hexanoyl homoserine lactone

(3-oxo-C6-HSL) autoinducer (AI)
e O O
O N
LuxI: N-3-oxo-hexanoyl homoserine lactone (3-oxo-C6-HSL)
H
O
O
AinS: N-octanoyl homoserine lactone (C8-HSL) O
N
H
HO OH O
B–
O O
LuxS: autoinducer 2 (AI-2)
HO
HO O
. Fig. 20.7
Bioluminescence control in V. fischeri. (a and b) Images of the same juvenile light organ taken with light and in the dark, respectively; the
light emitted by the bacteria within the organ can be seen in the center of the image taken in the dark. (c) The substrates and products in
the reaction carried out by luciferase (LuxAB) to produce light. (d) The lux operon and the control of lux transcription by AI-modified
LuxR. (e) The structures of the three pheromones produced in V. fischeri by the three pheromone synthases LuxI, AinS, and LuxS. Panels
a and b have been published previously (Stabb 2005)
including the luxICDABEG operon and the divergently tran- sequence to activate transcription of luxICDABEG through
scribed luxR gene (> Fig. 20.7d). luxCDABEG encodes the interactions with RNA polymerase that compensate for a poor
enzymes for luminescence, while luxI encodes a pheromone -35 promoter element (Egland and Greenberg 1999; Fuqua et al.
synthase and luxR encodes a corresponding pheromone recep- 1994). The result is that the lux genes are most highly expressed
tor. Specifically, LuxI generates N-3-oxo-hexanoyl homoserine at high cell densities, when the bacteria have reached
lactone (3-oxo-C6-HSL) (> Fig. 20.7e) (Schaefer et al. 1996), a ‘‘quorum’’ (Fuqua et al. 1996). Moreover, like many bacterial
a pheromone that diffuses across the cell membrane (Kaplan pheromone-mediated regulatory systems, the lux operon con-
and Greenberg 1985). As cell density increases, 3-oxo-C6-HSL stitutes a positive feedback circuit, because the 3-oxo-C6-HSL
accumulates until it reaches a critical concentration where it can produced by LuxI leads to increased transcription of
bind LuxR, producing an activator that binds a ‘‘lux box’’ luxICDABEG. Consequently, environmental regulatory inputs
to the lux system can be amplified, and in theory, the response and Ruby 1994a). Most early studies of lux regulation used either
could spread through a population. bright V. fischeri strains such as MJ1, which was isolated from
It has now become clear that pheromone-mediated regula- a pinecone fish (Ruby and Nealson 1976), or the lux region
tion of bioluminescence in V. fischeri is quite a bit more complex cloned in Escherichia coli. The core circuitry of lux in ES114 is
than this canonical model. Two other pheromones also modu- similar to that of MJ1 (Gray and Greenberg 1992b; Gray and
late the lux genes (> Fig. 20.7e); N-octanoyl homoserine lactone Greenberg 1992a), but like most isolates from E. scolopes, ES114
(C8-HSL) produced by AinS (Hanzelka et al. 1999; Kuo et al. is dim in culture.
1994, 1996) and ‘‘AI-2,’’ which is produced by LuxS (Lupp and With the caveat that there could be strain-specific differ-
Ruby 2004) and is presumably a furanosyl borate diester as it is ences, environmental control of lux in V. fischeri has been stud-
in Vibrio harveyi (Chen et al. 2002). AI-2 and C8-HSL appar- ied with respect to different regulators. For example, cAMP
ently function through distinct receptors that both act via LuxU receptor protein (CRP)-mediated activation of lux was
and LuxO, Hfq, and a small RNA called Qrr to increase levels of documented using transgenic lux-containing E. coli (Dunlap
LitR, which in turn increases luxR expression (Fidopiastis et al. and Greenberg 1985, 1988; Dunlap and Ray 1989), and it was
2002; Lupp et al. 2003; Miyamoto et al. 2000; Miyashiro et al. similarly shown that V. fischeri strain MJ1 regulates lumines-
2010). C8-HSL can also bind to and activate LuxR, although it is cence in response to glucose, a phenomenon that may be CRP-
less effective than 3-oxo-C6-HSL and can even inhibit 3-oxo- mediated (Friedrich and Greenberg 1983). Luminescence of MJ1
C6-HSL-mediated stimulation (Egland and Greenberg 2000; is also inhibited by iron (Haygood and Nealson 1985).
Schaefer et al. 1996). As with LuxI and LuxR, there is also Recently, the redox-responsive ArcA/ArcB two-component
positive feedback in the AinS-AinR system (Lupp and Ruby regulatory system was identified as a strong repressor of lumi-
2004). These interconnected regulatory systems were reviewed nescence (Bose et al. 2007; Lyell et al. 2010). Both arcA and arcB
recently elsewhere (Stabb et al. 2008). mutants are ~500-fold brighter in culture than ES114, and
3-oxo-C6-HSL, C8-HSL, and AI-2 all increase in concentra- activated ArcA binds the luxI promoter near the ‘‘lux box’’
tion with increasing cell density, and textbook models of (Bose et al. 2007). In E. coli, the Arc system is activated by
pheromone-mediated regulation typically depict these phero- reducing conditions, and consistent with this mechanism, lumi-
mones as a mechanism of cell-density-dependent regulation called nescence is induced by relatively oxidizing conditions (Bose et al.
quorum sensing. However, in V. fischeri and many other systems, 2007; Stabb et al. 2008). Interestingly, however, ArcA/ArcB
it is clear that pheromone-mediated signaling is not simply does not significantly repress luminescence during symbiotic
a census-taking process (Dunn and Stabb 2007). Both the pher- colonization, and arcA mutants achieve nearly the same bright-
omone synthases and the pheromone receptors are also regulated in ness in culture as wild-type cells do in the host. Thus, regulation
response to environmental conditions, such that high cell density by ArcA/ArcB could account for most of the differences in
may be necessary to elicit a behavior, but a ‘‘quorum’’ is typically luminescence observed between cultured and symbiotic cells
not sufficient for a full response unless other conditions are met. by invoking a model in which Arc represses lux in culture,
Context dependence of pheromone-mediated signaling is but luminescence is derepressed by inactivation of Arc during
obvious in V. fischeri strains isolated from the light organs of initial infection (Bose et al. 2007). On the other hand, the
E. scolopes. Such strains induce pheromone synthesis and lumi- ArcA/ArcB system could be active and repressing lux during
nescence in the host light organ, but they are dim and produce initiation of the symbiosis but simply overpowered by another
less pheromone in culture (Boettcher and Ruby 1990; Lee regulatory effect.
and Ruby 1994a), even at equivalent cell densities (Stabb, Other genes affecting luminescence in V. fischeri have been
unpublished data). Such context-dependent lux expression was reported (Hussa et al. 2007; Visick et al. 2007; Whistler and Ruby
evident even in comparisons of different light-organ microen- 2003). Recently, a mutant screen revealed numerous loci
vironments: when V. fischeri was marked with both a constitu- involved in regulation of the luxICDABEG operon, either
tive red fluorescent protein gene and a luxI promoter-gfp directly or indirectly (Lyell et al. 2010). That study also revealed
reporter, dense populations of bacteria appeared in the three environmental conditions that influence luminescence of wild-
distinct light-organ crypt types in a similar time frame, but lux type V. fischeri. For example, mutants lacking a phosphate-
induction lagged in crypt 3 (Dunn et al. 2006), which is also uptake system or phoQ were brighter than wild type, leading to
the last crypt to mature developmentally (Montgomery and the demonstration that phosphate or Mg2+ levels can influence
McFall-Ngai 1993). These data are consistent with induction of luminescence of wild-type cells (Lyell et al. 2010). Remaining
luminescence by a host-derived environmental cue initially challenges include determining which environmental conditions
absent in crypt 3. Clearly, high cell density may be necessary that influence luminescence in culture also impact symbiotic
for pheromone-mediated induction of luminescence in luminescence and whether information about environmental con-
V. fischeri, but it is not sufficient. ditions can be communicated among V. fischeri cells using pher-
Bioluminescence and pheromone production by V. fischeri omone signaling, as the results thus far would seem to indicate.
respond to the environment; however, the underlying regulatory It should also be noted that luminescence is just one part of
mechanisms are still being elucidated. This is particularly true the pheromone-controlled regulon in V. fischeri, and the role(s)
for V. fischeri strain ES114, which serves as a model strain typical of other genes co-regulated by C8-HSL, 3-oxo-C-HSL, and AI-2
of other E. scolopes symbionts (Boettcher and Ruby 1990; Lee along with luminescence should be enlightening. LuxR controls
several genes besides the luxICDABEG operon (Antunes et al. Mutants defective for syp also exhibit defects in initiating
2007, 2008; Callahan and Dunlap 2000), and some of the symbiotic colonization. The syp locus comprises 18 genes
corresponding proteins were identified in the proteome of sym- (sypA-R), four of which encode regulatory proteins, while the
biotic V. fischeri (Schleicher and Nyholm 2011). Especially note- remaining genes in the cluster encode proteins with predicted
worthy is the novel protein QsrP, which has no known function functions in polysaccharide biosynthesis, modification, and
but is induced by LuxR and was among the most abundant export (Yip et al. 2005). The genes in the syp locus are grouped
proteins in symbionts (Schleicher and Nyholm 2011). Visick into at least four operons (> Fig. 20.8), each of which contains
et al. (2000) demonstrated that the colonization deficiency of promoter sequences for transcription by RNA polymerase car-
luxR mutants could be overcome by expressing luxCDABEG rying the alternative sigma factor s54. Transcription from such
independently of this activator, suggesting that bioluminescence promoters depends upon binding and activation by a s54-
is the major (or only) LuxR-dependent colonization factor; dependent transcriptional activator, which typically binds
however, other LuxR-regulated genes could have subtle influ- upstream of the promoter and provides the energy for open
ences on the symbiosis or major effects on symbiotic coloniza- complex formation (Buck et al. 2000). Indeed, a conserved
tion but subtle degrees of LuxR-mediated regulation. Also, sequence located upstream of each of the syp promoters has
C8-HSL and AI-2 influence the expression of more genes than been identified and may serve as the binding site for this key
luxR, and the AinS/AinR regulon in particular is distinct from activator. The response regulator SypG is likely to serve as the
that of LuxR and important in host colonization (Lupp and activator for transcription of the syp locus, as this protein con-
Ruby 2004, 2005). Understanding the interplay of these three tains a putative s54 interaction domain in addition to its
pheromone systems in the light-organ symbiosis, their control N-terminal receiver (REC) domain (containing the site of phos-
by the symbiotic environment, the dynamics of their signaling, phorylation) and C-terminal DNA-binding domain. In support
and the constituents of their respective regulons should be a rich of this idea, overexpression of SypG promotes syp transcription
area for further research. in a manner that depends upon the rpoN gene, which encodes
s54 (Yip et al. 2005).
The similarity in the symbiotic phenotypes of rscS and syp
Biofilm Formation and Biofilm Regulators mutants led to the hypothesis that these proteins work in the
same pathway, specifically, that RscS controls transcription of
Studies of the initiation of symbiotic colonization revealed that the syp locus. Indeed, overexpression of RscS led to an increase
an early stage involves the attachment of V. fischeri to the surface in syp transcription, and this induction depended on SypG
of the light organ in a biofilm-like aggregate (> Fig. 20.2d). (> Fig. 20.8) (Hussa et al. 2008; Yip et al. 2006). Excitingly,
Subsequent studies began to probe the mechanism of that overexpression of RscS also led to a number of unusual biofilm
attachment. In particular, a search for genes important in the phenotypes (Yip et al. 2006). Whereas wild-type cells form
symbiosis revealed a requirement for rscS (regulatory of symbi- smooth colonies on complex solid media, RscS-overexpressing
otic colonization, sensor) (Visick and Skoufos 2001), which cells form colonies with a wrinkled morphology (i.e., with 3D
encodes a hybrid sensor kinase protein, and syp (Yip et al. architecture), consistent with biofilm formation. In addition,
2005), an 18-gene symbiosis polysaccharide locus. It was subse- RscS-overexpressing cells form a biofilm pellicle at the air-liquid
quently determined that these genes are necessary both for the interface during static growth in a minimal medium, while
production of biofilms in culture and for the formation of the control cells do not. These phenotypes also depend upon
symbiotic aggregate (Yip et al. 2006) (> Fig. 20.8). SypG. Finally, consistent with the idea that RscS and Syp con-
Mutants lacking RscS are defective in initiating symbiotic tribute to colonization through their role in symbiotic aggregate
colonization: most animals inoculated with this mutant remain formation, overexpression of RscS induces the formation of very
uncolonized, while others exhibit a delay in initiating coloniza- large aggregates on the light-organ surface (> Fig. 20.8).
tion (Visick and Skoufos 2001). This defect can be attributed to Increased aggregate formation depended upon an intact syp
the inability of this mutant to aggregate on the surface of the locus: mutation of a representative syp gene, sypN, which
light organ (Yip et al. 2006). rscS encodes a sensor kinase pro- encodes a putative glycosyltransferase predicted to be involved
tein, a fact that immediately suggested a model for its function in in the production of a polysaccharide, disrupted aggregate for-
symbiosis. Sensor kinases work by sensing an environmental mation (Yip et al. 2006). The formation of this large aggregate is
signal, autophosphorylating on a conserved histidine residue, not detrimental to colonization; rather, RscS-overexpressing
and serving as a phospho-donor to a downstream response cells fully outcompete control cells containing an empty vector
regulator. The activated response regulator then carries out for symbiotic colonization.
a response, most commonly acting as a transcription factor to The syp locus is controlled by several regulators besides RscS
increase or decrease transcription of genes. Although in many and SypG. Currently, the best understood of these additional
cases the genes for a sensor kinase/response regulator pair are regulators is SypE, a second response regulator encoded within
physically linked in the chromosome, this was not the case for the syp locus. SypE is an unusual response regulator in that its
rscS. Subsequent studies revealed that RscS acts upstream of the REC domain, containing the putative site of phosphorylation, is
response regulator SypG to control syp transcription and biofilm centrally located, rather than localized to the N-terminus as is
formation (Hussa et al. 2008). typical of other response regulators. The REC domain is flanked
RscS SypF
SypE Cellulose
SypG VpsR
biosynthesis
SypG P
A B C D E F G H I J K L M N O P Q R
Biofilm
formation
. Fig. 20.8
Control over biofilm formation by the syp regulators. The activities of two sensor kinases, RscS and SypF, influence syp-dependent
biofilm formation. Sensor kinases respond to an environmental signal by autophosphorylating and donating a phosphoryl group to
a downstream response regulator. In the case of the syp locus, phosphorylated SypG is predicted to activate transcription from four
promoters, resulting in the production of a Syp-produced polysaccharide that promotes biofilm formation. Biofilm formation can be
visualized in the laboratory by the development of wrinkled colonies and the production of pellicles at the air/liquid interface of static
cultures (which can be strong enough to retain liquid when the test tube is inverted), and in symbiosis by the production of large
aggregates on the surface of the light organ. RscS also appears to inactivate SypE, a response regulator that exerts a minor positive and
a strong negative impact on biofilm formation. The role of SypF is poorly understood, but it appears to act upstream of SypE/SypG and
VpsR to impact syp-dependent biofilm formation and cellulose biosynthesis, respectively (This model, which was previously published
(Visick 2009), has been slightly altered by the addition of an arrow extending from SypE to biofilm formation, indicating the slight
positive effect exerted by SypE on biofilm formation (Morris et al. 2011))
by putative serine kinase (N-terminus) and serine phosphatase phosphorylated (Morris et al. 2011). Interestingly, a sypE mutant
(C-terminus) domains (Morris and Visick 2010). SypE appears does not exhibit a colonization defect, perhaps due to SypE’s
to act as both an inhibitor and an activator of biofilm formation, subtle positive role in biofilm formation. However, expression of
depending on the conditions used to promote biofilm formation the SypE mutant protein that cannot be phosphorylated and
(overexpression of RscS or SypG) (Hussa et al. 2008). Such thus inactivated causes a severe defect in symbiotic colonization.
phenotypes are consistent with opposing activities for the ter- Therefore, it appears that SypE inhibits symbiotic colonization
minal domains. Indeed, it is now known that the putative serine and must be inactivated to permit colonization, presumably by
kinase domain inhibits biofilm formation, while the putative RscS. The symbiotic defect of an rscS mutant could be attrib-
serine phosphatase activates it (Morris et al. 2011). uted, in part, to its failure to inactivate SypE. Consistent with
These data led to a model in which RscS, in addition to this model, an rscS sypE mutant colonized better than an rscS
activating SypG to promote syp transcription, inactivates SypE mutant. Thus, productive symbiosis depends upon inactivation
to permit biofilm formation. Given the nature of these proteins, of a negative regulator.
Morris et al. (2011) predicted that RscS promotes phosphoryla- Another regulator encoded by the syp locus is SypF. Neither
tion of SypE, thereby inactivating it. Consistent with that pre- insertional disruption of sypF nor overexpression of wild-type
diction, a substitution of alanine for aspartate at the predicted SypF impacted biofilm formation (Darnell et al. 2008). How-
site of phosphorylation results in a protein with constitutive ever, overexpression of an allele with increased activity, sypF*,
inhibitory activity, presumably because it is unable to be led to increased biofilm formation, including wrinkled colony
formation and pellicle production. Surprisingly, these pheno- cells to upwards of 16 flagella on the mutants. Mutants in classes
types persisted even when the two syp response regulators, sypG II and III also produced mucoid colonies rather than the smooth
and sypE, were mutated. However, biofilm formation was elim- nonmucoid colony of their parent furthermore; class II mutants
inated when both sypG and a second response regulator, vpsR, had a growth defect, and class III mutants exhibited defects in
were disrupted ( > Fig. 20.8). In V. cholerae, VpsR controls the bioluminescence, 3-oxo-C6-HSL pheromone production, and
vps polysaccharide locus, which is only poorly conserved in V. the ability to hemagglutinate red blood cells. All three classes
fischeri (Yildiz et al. 2001; Yildiz and Visick 2009; Darnell et al. exhibited defects in initiating symbiotic colonization, failing to
2008). Instead, it appears that a locus involved in cellulose reach wild-type levels in 24 h. Class III mutants demonstrated
biosynthesis contributes to biofilm formation induced by the most severe defect, achieving only 0.1–10% of wild-type
overexpression of SypF*. Clearly, control of biofilm formation levels at 24 and 48 h postinoculation. The defect in initiation
in V. fischeri is complex and is not limited to a single regulator. was attributed to deficiencies at the aggregation stage: wild-type
Many questions remain unanswered. How SypF is integrated cells form non-motile aggregates within 4–6 h; in contrast, the
into the regulatory pathway remains unclear, and its role in sym- mutants retained motility. Furthermore, aggregate formation by
biotic colonization has not yet been established. The product of the the mutants was delayed, occurring between 8 and 12 h after
syp locus is thought to be a polysaccharide (Yip et al. 2006), but inoculation, and aggregates were significantly smaller (tens of
the composition of the polysaccharide is as yet undetermined. cells rather than hundreds) than those formed by the wild type
Importantly, the environmental signals that activate RscS and (Millikan and Ruby 2002). Although the mutations in these
the other regulators to induce biofilm formation and initiate strains were never identified, these studies nevertheless indicate
symbiotic colonization are as yet unknown. Moreover, since the importance of regulating motility during entry into the
symbiotic colonization is not impaired by overproduction of symbiosis and beyond.
RscS and enhanced aggregate formation, there must be a yet- Subsequent studies and the availability of the genome
undiscovered mechanism by which aggregation is turned off to sequence have confirmed the importance of motility in this
permit dispersal and migration into the pores for colonization to symbiosis and have identified specific pathways of genetic con-
proceed. The discovery of the syp locus and its role in biofilm trol. Almost all genes predicted to be involved in flagellar assem-
formation and initiating symbiotic colonization has provided bly and regulation are housed together in a single locus of about
important insights into host colonization by V. fischeri and an 60 genes on the larger of V. fischeri’s two chromosomes. The
important model for in vivo biofilm formation and will con- functions of these genes can be readily predicted through com-
tinue to raise new questions to direct future research. parisons with well-studied models such as E. coli and Salmonella
(Chevance and Hughes 2008). Furthermore, comparisons with
other organisms suggest that regulation of flagellar genes in
Motility, Chemotaxis, and Their Regulation V. fischeri is likely hierarchical. Perhaps the best comparison is
with the closely related microbe V. cholerae, which has been
V. fischeri is motile by virtue of a polar tuft of flagella extensively studied (Correa et al. 2000, 2005; Prouty et al.
(> Fig. 20.9a) (Ruby and Asato 1993). Early on in the study of 2001). In that organism, the master flagellar regulator, FlrA,
this symbiosis, it was determined that motility of V. fischeri is controls a number of flagellar genes, including additional
essential for symbiotic colonization; however, the roles of motil- regulators such as FlrBC and FliA. FlrBC is a predicted
ity, flagellation, and chemotaxis continue to be intriguing and two-component signal transduction system that controls the
outstanding questions. Initial studies performed with next level of flagellar genes, including flaA, a flagellin gene;
uncharacterized mutants revealed that both nonflagellated and flagellin protein constitutes the major subunit assembled into
flagellated-but-nonmotile (paralyzed) mutants fail to colonize the external filament of the flagellum. Indeed, a V. fischeri flrC
(Graf et al. 1994). However, motility per se is not required for the mutant is non-motile and, like other non-motile mutants,
aggregation stage (Nyholm et al. 2000; Millikan and Ruby 2002), exhibits a colonization defect (Hussa et al. 2007). FliA, or s28,
and once inside the crypt spaces, most V. fischeri cells become activates another subset of flagellar genes, including four other
nonflagellated (> Fig. 20.9b) (Ruby and Asato 1993). It is not flagellin genes (flaB-E). The working model is that V. fischeri
known whether the loss of flagella is critical for colonization, likely controls its flagella using a hierarchy similar to that of V.
especially since some bacteria do retain motility (Millikan and cholerae (> Fig. 20.9c) (Millikan and Ruby 2003; Prouty et al.
Ruby 2003). However, it is likely that flagella could interfere with 2001).
the ability of the cells to achieve high cell density in the crypt In V. fischeri, the role of FlrA was investigated in detail by
spaces. Finally, V. fischeri cells rapidly regain their flagella upon Millikan and Ruby (2003). flrA mutants are non-motile, as
release into seawater (> Fig. 20.9b) (Ruby and Asato 1993). predicted from FlrA’s putative role as master flagellar regulator,
One study suggested that excessive amounts of flagella are and fail to colonize. Surprisingly, complementation with FlrA
detrimental to initiation of colonization. Millikan and Ruby expressed from a plasmid restored normal motility in
isolated three classes of hypermotile mutants with increased culture but not normal colonization: complemented flrA
motility in both liquid and semisolid media (Millikan and mutants failed to initiate colonization at the rate and with the
Ruby 2002). The increased motility apparently stemmed from success of wild-type cells (only 35–46% of animals became
an increase in the number of flagella, from 1 to 3 on wild-type colonized by the complemented flrA mutant relative to 94% of
a b
Aggregation
c
FlrA Migration
+ Flagellin
+ Topoisoisomerase
σ54
– Halovibrin
– Phosphoglycerate kinase
FliA – Potassium efflux
FlrBC σ28
Colonization
σ54 FlaB
FlaC
FlaA FlaD
FlaE
FlaF?
Release
. Fig. 20.9
Role of motility in symbiosis. (a) V. fischeri contains a tuft of polar flagella. (b) Presence of flagella during different stages of colonization.
Nonmotile cells are competent to aggregate, but motility is required for entry into the light-organ pores and presumably for passage to
the crypts. Upon colonization, many bacteria lack flagella. Finally, following release from the light organ, V. fischeri cells can rapidly
regrow their flagella. (c) Model for regulatory control by FlrA, a s54-dependent regulator. FlrA is predicted to induce transcription of the
genes for the two-component system FlrBC, which, along with s54, controls transcription of the gene for the major flagellin protein FlaA.
FlrA is also predicted to control transcription of the gene for the alternative sigma factor, FliA, which likely directs transcription of the
genes for the other flagellin proteins, FlaB-E and potentially FlaF. Finally, FlrA appears to positively and negatively control the
production of other proteins, although whether this effect is direct is unknown. The figure in panel A was previously published (O’Shea
et al. 2005), and the scale bar represents 200 mM (The regulatory scheme depicted in panel C is modeled from reported data for V. fischeri
and V. cholerae (Millikan and Ruby 2003; Prouty et al. 2001))
wild-type-inoculated animals); they also failed to achieve wild- whether these FlrA-regulated genes play roles in symbiosis,
type levels of colonization (1.6 103 colony-forming units of remains unknown.
flrA mutant cells per squid relative to 9 104 colony-forming The FlrA protein contains a s54 interaction domain and
units of wild-type bacteria per squid). These experiments sug- a DNA-binding domain and thus is predicted to activate tran-
gest that the proper level or location of FlrA is critical for scription of target genes in conjunction with s54. Indeed, rpoN
symbiosis, and indicate that a trait other than motility, which mutants, which lack s54, are non-motile, fail to induce tran-
was restored by multi-copy expression of flrA based on in vitro scription of flaA, and fail to initiate colonization (Millikan and
assays, may be responsible for the symbiotic defect in the flrA Ruby 2004; Wolfe et al. 2004). Like FlrA, s54 controls additional
mutants (Millikan and Ruby 2003). In support of this idea, genes involved in colonization, notably luminescence and the
symbiotic aggregates produced by flrA mutants are diminished syp polysaccharide locus (Wolfe et al. 2004; Yip et al. 2005). The
in size relative to those produced by the wild-type strain (tens of use of this alternative sigma factor is of note, as genes under
cells vs. hundreds of cells), despite the fact that other non-motile the control of s54 tend to be under tighter ‘‘on/off ’’ regulatory
mutants aggregate normally (Nyholm et al. 2000). Millikan and control (Buck et al. 2000); whether or not this regulation is key
Ruby subsequently identified four nonflagellar genes that were for colonization remains to be determined.
controlled by FlrA positively (e.g., genes encoding topoisomer- Of V. fischeri’s six flagellin genes (flaA-F), only flaA appears
ase and halovibrin C) or negatively (e.g., genes encoding to be directly regulated by FlrA and s54. Despite the redundancy
phosphoglycerate kinase and potassium efflux protein) of flagellin genes, flaA mutants show reduced motility with fewer
(> Fig. 20.9c). Whether FlrA controls these genes directly, and motile cells, and fewer flagella per cell. In contrast, a flaC mutant
was seemingly unaffected for motility and flagellar elaboration. prevent it from appropriately migrating toward a stimulus.
The flaA mutant also exhibited a symbiosis defect: loss of flaA Alternatively, it may get stuck in squid-secreted mucus.
delayed symbiotic colonization by about 3 h, and prevented full Distinguishing between these possibilities requires the identifi-
colonization (20–25% of that achieved by the wild-type strain) cation of a specific chemoreceptor(s) required for attractant
(Millikan and Ruby 2004). The mutant was also defective in recognition during symbiotic colonization. However, V. fischeri
competing for colonization with the wild type, and furthermore, contains upwards of 40 such receptors, making this endeavor
it appeared to be preferentially expelled from the light organ. a challenge (Ruby et al. 2005).
GFP-labeled cells revealed that, in contrast to the flrA regulatory One intriguing and outstanding question is: What factors
mutant, the flaA flagellin mutant formed aggregates with the direct the loss of flagella inside the light organ? Another is its
same approximate size as wild-type aggregates; furthermore, the corollary, What promotes flagellar regrowth in the nutrient-
flaA mutant entered the light organ normally. However, while limited seawater? One possibility is suggested by the dependence
the wild-type cells reached and colonized deep crypt spaces of V. fischeri motility on magnesium (O’Shea et al. 2005). Cells
within 16 h postinoculation, flaA mutant cells did not colonize grown in a medium lacking magnesium are poorly motile, as
these sites within 20 h. This inability to reach and colonize the assessed on soft agar plates. Addition of either magnesium
deep crypt spaces could account for both the delay in coloniza- sulfate or magnesium chloride restores full migration. In partic-
tion and the preferential expulsion of the flaA mutant. ular, magnesium sulfate concentrations between 20 and 40 mM,
As suggested by the phenotype of the flaA mutant and similar to that found in seawater (about 50 mM), were optimal
numerous studies in other organisms, one prime reason for in promoting migration through soft agar. Other cations, such
cells to be motile is to permit them to move to optimal locations. as calcium, could promote motility but to a lesser extent at low
Typically, this movement is directed toward nutrients and concentrations (5 mM); at concentrations higher than 10 mM,
attractants and away from toxic molecules and repellents using CaCl2 was inhibitory to growth and thus difficult to assess.
chemotaxis. The role of chemotaxis in symbiotic colonization by Subsequent analysis revealed that cells grown without magne-
V. fischeri is currently an active area of investigation. One study sium elaborated few to no flagella, while those grown with
asked what molecules serve as chemoattractants for V. fischeri magnesium contained on average two to three flagella and as
(DeLoney-Marino et al. 2003). In tryptone-based soft agar many as seven or eight (O’Shea et al. 2005). It is unknown how
media, V. fischeri forms two rings as it migrates from the spot of magnesium impacts flagellar production, although it does not
inoculation, indicating that it senses and responds to two different appear to be at the level of transcription of flagellin genes
molecules that it likely consumes. The inner ring is comprised of (O’Shea et al. 2006).
cells sensing the amino acid serine. Cells in the outer ring sense the A search for mutants that were motile in the absence of
nucleic acid thymidine, which is present in tryptone. In addition magnesium led to the identification of two genes encoding the
to thymidine, V. fischeri can sense and respond to other ribonu- diguanylate cyclases MifA and MifB. These proteins are involved
cleosides (adenosine, guanosine, uridine, and cytidine) as well as in the production of cyclic diguanylate (c-di-GMP), which is
deoxynucleotide triphosphates. This ability of V. fischeri to sense known to influence motility (Wolfe and Visick 2010). Disrup-
nucleosides and nucleotides is unusual, although more recently tion of either mifA or mifB increased motility in the absence of
it has been shown that E. coli and Pseudomonas putida can magnesium. However, neither mutant alone nor a double
perform chemotaxis to pyrimidines (thymine and uracil in the mutant defective for both genes exhibited the same pattern of
case of E. coli, cytosine in the case of P. putida) (Liu and Parales migration in the absence of magnesium as the wild-type strain
2008; Liu et al. 2009). In addition to serine and the building exhibits in the presence of magnesium. Thus, it remains unclear
blocks of DNA and RNA, V. fischeri responds to a number of how magnesium impacts motility.
sugars, including N-acetylneuraminic acid, which is one of the Regardless of how magnesium acts to stimulate the produc-
sugars in the mucus secreted by the squid (DeLoney-Marino tion of flagella, seawater contains sufficient levels of magnesium
et al. 2003; Nyholm et al. 2000). to permit V. fischeri to express flagella and thus be competent to
It remains unknown whether chemotaxis helps direct the colonize its squid host. Once inside, it is possible that decreased
bacteria into the light organ, and if so, which chemoattractants levels of magnesium and/or calcium contribute to the mecha-
are used, but a chemotaxis-defective cheY mutant fails to com- nism that promotes loss of flagella on colonizing bacteria. The
pete effectively against wild-type cells for colonization (Hussa levels of magnesium and calcium inside the light organ of
et al. 2007). In E. coli and Salmonella, CheY is a response regu- E. scolopes are presently unknown.
lator that modulates reversal of flagellar rotation, permitting In summary, research to date has demonstrated the impor-
‘‘tumbles’’ that when interspersed with ‘‘runs’’ permit the cells tance of flagella and motility to symbiosis, but the role of
to reorient and migrate toward nutrients and away from repel- motility is not simple and straightforward. In other systems,
lents. (Baker et al. 2006). Cells with defects in tumbling, such as flagella can serve as mediators of attachment, a possibility that
cheY mutants, fail to perform normal chemotaxis; in addition, is supported by the aberrant aggregation of the hypermotile and
they can become trapped in ‘‘dead-end’’ passages present in flrA mutant strains. In some hosts, flagellin proteins can act as
agar-solidified media, making them appear nonmotile (Wolfe MAMPs, in much the same way as PG and LPS, which is an
and Berg 1989). Like E. coli and Salmonella mutants, the intriguing possibility given the signaling obvious in this symbi-
V. fischeri cheY mutant exhibits smooth swimming, which may osis. Flagellar components can also serve as type III secretion
systems that direct bacterial proteins into the host environment, HutZ, VF_1226-VF_1228). These genes, as well as putative
and this remains a possibility for V. fischeri as well. When hemin receptor genes VF_1234 and VF_A0331-A0333, were
motility or components of the flagellar system are required and downregulated upon exposure of wild-type cells to NO in a man-
when/if they are detrimental, when/if flagella or the flagellar ner that depended upon an intact hnoX gene, which encodes an
apparatus contributes to attachment or secretion, what the NO sensor (Wang et al. 2010a). Regulatory sequences consistent
precise contribution of chemotaxis and where the optimal loca- with control by Fur (Ferric uptake regulator) were identified.
tions of the bacteria are, and how flagella are lost and regained Consistent with the putative role of this locus, its deletion
are all important questions that remain to be pursued. No doubt disrupted the ability of V. fischeri to grow with hemin (Fe3+) as
the answers will provide important insights into host coloniza- a source of iron, but did not impact growth of the cells when
tion and colonization dynamics. ferrous sulfate was supplied as an iron source (Septer et al. 2011;
Wang et al. 2010a). Investigations into the transcriptional con-
trol of this locus revealed that two divergent promoters,
Iron Uptake upstream of VF_1225 and VF_1226, were upregulated in
response to low iron availability. Furthermore, both promoters
Eukaryotic hosts generally limit the availability of iron, which is were repressed by Fur, as disruption of fur led to an increase in
necessary for the growth and metabolism of bacteria, as transcription even in the presence of iron. These data are con-
a defense mechanism (reviewed in (Nairz et al. 2010)). As sistent with a model in which low iron availability causes an
a result, bacteria have evolved numerous systems for the acqui- increase in transcription of this heme-uptake locus; when iron
sition of iron. V. fischeri, for example, encodes a number of levels rise, Fur is activated to decrease transcription of the locus
different proteins predicted to be involved in binding of iron (Septer et al. 2011). Thus, V. fischeri appears to carefully modu-
or iron complexes and their associated transport systems (Ruby late transcription of this locus in response to iron availability. To
et al. 2005). The first study examining the role of iron uptake assess the relevance of this phenomenon and these genes in
during the V. fischeri-squid symbiosis suggested that, indeed, the symbiosis, two approaches were used. First, the transcriptional
ability to acquire iron was an important symbiosis trait (Graf response of this locus in symbiosis was assessed with promoter-
and Ruby 2000). This report relied on a transposon mutagenesis GFP reporters. Transcription of VF_1225 was turned on as early
screen for mutants with altered siderophore production and as 14 h postinoculation, indicating that the light-organ environ-
focused on one that appeared completely defective in its ability ment is low in iron. Second, the competence of the deletion
to sequester iron. Indeed, growth of this strain depended on the mutant to colonize when presented in competition with the
presence of iron in the medium. Surprisingly, this mutant wild-type strain was evaluated. The mutant exhibited a defect in
turned out to contain an insertion not in a siderophore biosyn- colonization, although the defect was not apparent until after
thesis gene, but in glnD, a gene known in E. coli to play a role in several days postinoculation. Thus, these data support and extend
sensing nitrogen status (reviewed in (Arcondeguy et al. 2001)). the study by Graf and Ruby (2000) that the light-organ environ-
Consistent with a similar role for GlnD in V. fischeri, the glnD ment is limiting for iron, and indicate that uptake of heme is an
mutant was deficient in utilizing a variety of nitrogen sources. important mechanism by which V. fischeri acquires iron during
Complementation with a wild-type copy of glnD restored all of symbiosis. In further support of the importance of heme uptake
the observed phenotypes to wild-type levels, supporting the in symbiosis, the transcriptome study by Weir et al. (2010) found
novel role of GlnD in siderophore production. Finally, the that the heme-uptake genes were induced following bacterial
glnD mutant exhibited a defect in colonizing juvenile squid. In venting. At venting, membrane ‘‘blebs’’ (presumably host
support of the idea that the symbiotic phenotype was due to the derived) appear, and it was speculated that these contain a source
iron sequestration defect, the defect was more pronounced when of iron for the symbionts (Septer et al. 2011). Finally, a recent
cells in the starting inoculum were pre-grown under conditions study investigating the proteomes of V. fischeri and E. scolopes
of limiting iron availability. Furthermore, addition of iron to identified host-iron-binding proteins, including transferrin, fer-
seawater alleviated the defect of the glnD mutant (Graf and Ruby ritin, and melanotransferrin, and symbiont heme-binding pro-
2000). The mechanism by which GlnD impacts siderophore teins (Schleicher and Nyholm 2011). Whether heme is the
production remains unknown. primary source of iron during symbiosis, and how/when iron
A more directed study investigated the role of one possible is supplied to the bacteria, will be of interest to determine.
mode of acquiring iron, heme transport, in the V. fischeri-squid
symbiosis. Bioinformatic analyses identified a locus, VF_1220-
VF_1228, similar to heme-uptake loci in other Vibrios (Septer GacA
et al. 2011; Wang et al. 2010a). One operon within this locus
(VF_1225-VF_1220) is predicted to encode a periplasmic heme- Another important regulatory protein involved in symbiosis is
binding protein (HutB), an inner membrane permease (HutC), the response regulator GacA. A gacA mutant was defective in
an ABC-transporter ATPase (HutD), and proteins that function initiating symbiosis: only about 50% of animals became colo-
in providing the energy for transport (TonB, ExbB, ExbD). nized under conditions in which 100% of wild-type-inoculated
Additional proteins predicted to be involved in optimal heme animals became colonized, and animals that became colonized
utilization were encoded in a divergent operon (HutW, HutX, with the gacA mutant contained on average 100X fewer bacteria
(Whistler and Ruby 2003). Moreover, as mentioned above, in of gacA mutants suggests the GacS/GacA system is active
squid that were colonized by gacA mutants, the symbionts did during colonization, although it cannot be ruled out that
not trigger some of the normal developmental events in the host unphosphorylated GacA has some activity or that GacA is phos-
that are elicited by wild-type V. fischeri, including apoptosis and phorylated by some means other than GacS (e.g., by cross talk
cessation of mucus shedding in light-organ epithelial cells from another sensor kinase). The stimulus for GacS homologs in
(Whistler et al. 2007). To take into account the possibility that other systems has been elusive, although it may involve meta-
low colonization by the gacA mutant might underlie these bolic products of the bacteria themselves (Takeuchi et al. 2009),
apparent non-effects on the host, comparisons were made to and could perhaps be sensed from the cytoplasmic rather than
a lysA mutant, which is a lysine auxotroph that is also attenuated periplasmic side of GacS (Zuber et al. 2003). If so, the Gac
in colonization (Whistler et al. 2007). Using this control, the regulon could reflect a regulatory response to symbiont physi-
authors concluded that low colonization alone could not explain ology as it is constrained by the host environment. This possi-
the altered signaling capacity of the gacA mutant. bility underscores the importance of understanding symbiont
In culture, V. fischeri GacA controls bioluminescence, motil- physiology and how V. fischeri metabolism is supported by the
ity, LPS structure, siderophore production, and nutrient acqui- host. The V. fischeri-E. scolopes model system promises to be a
sition (Whistler et al. 2007; Whistler and Ruby 2003). In rather rich and tractable experimental system for deciphering the wide-
broad terms, the gacA mutant is also impaired in growth, spread Gac-Csr system.
although the severity of this defect depends considerably on
the culture medium (Whistler and Ruby 2003). In short, the
gacA mutant has a pleiotropic phenotype, it is affected N-Acetyl D-Glucosamine Repressor NagC
in a number factors known to (or likely to) affect colonization
of the host, and its inability to grow to wild-type levels in the The bacterial transcriptome analysis of Wier et al. (2010)
light organ is not a symbiosis-specific defect. Nonetheless, its described above revealed that genes involved in chitin catabo-
connection to so many colonization-promoting phenotypes lism were upregulated prior to dawn (Wier et al. 2010). This
marks it as a global regulator of importance in the symbiosis, finding stimulated an interest in understanding the regulation of
and the careful use of another attenuated (lysA) mutant as a these genes. Specifically, Miyashiro et al. (2011) searched for
control strongly suggests a critical symbiosis-specific signaling regulators of the exochitinase gene, VF_1598, and found that
defect. the repressor NagC negatively controls its expression and also
GacA is conserved in numerous and diverse g- represses the nagA operon (which includes nagC itself,
proteobacteria (although its orthologs have distinct names), VF_0806). Loss of nagC increased expression of these and sev-
where it controls a myriad of functions, including carbon-flow eral other genes involved in the utilization of chitin or its
physiology and virulence (Lapouge et al. 2008). Certain themes derivatives, the monomer N-acetyl-D-glucosamine (GlcNAc)
relevant to a potential symbiotic role have emerged, including or the dimer chitobiose (GlcNAc2). Excitingly, in colonization
the control of factors involved in colonizing hosts and pathways assays, a nagC null mutant exhibited a severe defect: when the
for social behavior and intraspecies signaling (e.g., quorum nagC mutant was used to inoculate squid, most squid remained
sensing) (Lapouge et al. 2008). Also conserved is the mechanism uncolonized (6 of 15 became colonized within 48 h compared to
of its associated regulatory cascade (Lapouge et al. 2008). The 90% colonized with the wild-type control) (Miyashiro et al.
sensor kinase GacS is responsible for phosphorylating and thus 2011). In competition assays, the nagC mutant neither impaired
activating the cognate response regulator GacA. Then GacA-P wild-type colonization nor was complemented by the presence
stimulates transcription of CsrB, regulatory RNAs that bind to of the wild-type strain. In contrast to the nagC result, no defect
CsrA, preventing it from binding target mRNAs and affecting was observed for a nagB mutant, which lacks a deaminase
their stability and/or translation. In V. fischeri, bioinformatic involved in releasing the amino group from GlcNAc-6P; this
analyses revealed two such regulatory RNAs, named CsrB1 and mutant failed to grow preferentially on GlcNAc as a carbon and
CsrB2 (Kulkarni et al. 2006), and putative GacA-binding sites are nitrogen source. Thus, the symbiotic defect of the nagC mutant
present upstream of each of these genes. V. fischeri also possesses is unlikely to stem from altered metabolism of GlcNAc. How-
clear homologs of GacS and CsrA. Unpublished results from ever, the generation and evaluation of a nagC nagB double
multiple laboratories suggest that, as predicted, CsrB1, CsrB2, mutant would better address the symbiotic consequences of
CsrA, and GacS have regulatory functions in V. fischeri, at least the overexpression of GlcNAc metabolism genes resulting from
some of which are interconnected. This regulatory circuit is an nagC disruption. Finally, exposure to GlcNAc during inocula-
area of active research, particularly with respect to its effect on tion eliminated the competitive advantage of the wild type over
bioluminescence from the lux system and a global perspective of the nagC mutant (Miyashiro et al. 2011). Together, these data
its regulon, which is likely to include genes involved in signals indicate that NagC represses something that must be turned off
perceived by the host. for colonization to proceed normally; in the absence of NagC-
Important questions that remain include the signal or envi- mediated repression, colonization is severely impaired. It will be of
ronmental cue perceived by GacS and whether this is present great interest to determine the regulon of NagC and which regulon
in the light-organ environment. The altered symbiotic phenotype member(s) impair colonization when inappropriately expressed.
Possible Role for Pili in Symbiosis will be necessary to determine the role of pilA in attachment and
colonization.
Many pathogenic associations depend upon a pilus-mediated Limited characterization of the genes encoding the
adherence of the bacteria to the host (Pizarro-Cerda and Cossart conjugative pilus has also been performed. These genes
2006). For example, type I pili play critical roles in attachment (VF_B38-55) are carried on the large plasmid pES100, and
and uptake in urinary tract infections by uropathogenic E. coli they are clustered with other components of a putative
(Mulvey et al. 1998; Wright et al. 2007). To date, however, conjugative transfer apparatus (Dunn et al. 2005). Indeed,
a critical role for pili in the V. fischeri-E. scolopes symbiosis has experiments indicated that pES100 can direct conjugative trans-
yet to be determined, although several studies have investigated fer of other plasmids and is almost certainly self-transmissible
this possibility. Interestingly, V. fischeri contains ten loci (Dunn et al. 2005). Efficient transfer also required the chromo-
encoding putative pili (Ruby et al. 2005). Eight loci encode somally encoded RecA protein to be present in the donor, and
putative type IV pili, the most common type of bacterial pilus this requirement was not due to homologous recombination
(reviewed in (Burrows 2005; Pelicic 2008; Pizarro-Cerda and between pES100 and the plasmid it mobilized. An exciting
Cossart 2006)). Due in part to their retractability, type IV pili future direction is determining whether conjugation occurs or
play a variety of important roles in other bacteria, including is even induced during colonization.
adherence to various surfaces and hosts, twitching motility, and
secretion of a variety of factors. Intriguingly, one of the type IV
pilus loci is similar to that of the V. cholerae TCP (toxin co- Cellobiose
regulated pilus), an important virulence factor in that bacte-
rium. Although some of the V. fischeri genes are not clustered but V. fischeri exhibits the ability to grow on cellobiose, a feature that
are spread throughout the genome, most of the TCP genes distinguishes it from a number of other Vibrio species, including
appear present (Ruby et al. 2005). One of the other pilin loci V. parahaemolyticus, V. hollisae, V. mimicus, and V. cholerae. This
includes a gene for MshA (mannose-sensitive hemagglutin), an ability is conferred by the cel locus (VF_0603-VF_0608),
adhesin located at the tip of the pilus that recognizes mannose- consisting of three genes that comprise a phosphotransferase
based receptors. Intriguingly, when a mannose analog was added system (PTS) II (celABC), a putative glucokinase gene CelK,
to seawater, it inhibited colonization by V. fischeri, suggesting and a 6-phospho-b-glucosidase gene celG (Adin et al. 2008b).
that a bacteria-mannose interaction may be important during Negatively controlling this locus is a LacI-family regulator, CelI,
initiation (McFall-Ngai et al. 1998). Of the two non-TCP loci, which is encoded at the end of the operon. Loss of celI or the
one appears to encode curli, an adherence factor found in the addition of cellobiose (but not other sugars) resulted in colonies
Enterobacteriaceae (Barnhart and Chapman 2006). Interest- that turned blue on the colorimetric substrate X-gal,
ingly, the curli locus is absent in other well-studied Vibrios, a phenotype that depended upon an intact celG. Further investi-
including V. cholerae, V. parahaemolyticus, and V. vulnificus gations revealed that CelG could cleave a variety of substrates with
(Ruby et al. 2005). Lastly, the large plasmid (pES100) of a b-1,4 linkage and could serve both as a b-glucosidase (to cleave
V. fischeri encodes conjugative pili (Ruby et al. 2005; Dunn substrates such as cellobiose) and as a b-galactosidase (consistent
et al. 2005). with its ability to cleave X-gal); its highest activity was as a b-
Of the 10 loci, two have been characterized in some detail. glucosidase, as that activity was 50-fold higher than its b-
One of the type IV pilus loci contains a single pilus gene galactosidase activity. Although not a preferred substrate, the ability
(VF_A0148), encoding a PilA-like pilin protein (PilA2) of CelG to cleave X-gal has implications for the use of lacZ as
predicted to be the major external subunit of the pilus (Stabb a reporter in V. fischeri (Adin et al. 2008b). Despite the somewhat
and Ruby 2003; Ruby et al. 2005). Disruption of this gene caused unique ability of V. fischeri to use cellobiose, this ability was not
a small but statistically significant defect in colonization when required for colonization: competition experiments of various
the mutant was presented to the squid in a mixture with the cel mutants with wild type revealed no competitive disadvantage
wild-type strain. Potentially, the presence of one or both of the of the lack of cellobiose catabolism; the exception to this was the
two other pilA genes compensates for the loss of VF_A0148. In celI mutant, which constitutive expressed the cel genes and
support of this idea is the lack of adjacent pilus structural and exhibited a twofold defect in competitions with the wild-type
assembly genes, suggesting that PilA2 gets assembled into a pilus strain for colonization. This inability of the celI mutant to
that is associated with a distinct pilin protein (Ruby et al. 2005). compete could be due to the overexpression of the cel locus,
An examination of the presence of pilA2 in V. fischeri isolates but could not be attributed to the cellobiose catabolism, as a celI
revealed that it is conserved in strains isolated from other light- celG mutant exhibited a similar competitive defect (Adin et al.
organ symbioses, including strains from E. tasmanica (from 2008b). Finally, a mutation in the unlinked gene ptsI (VF_1895/
Australia) and those from Sepiola species (from France) 1896), which encodes the E1 component of the PTS system,
(Browne-Silva and Nishiguchi 2008), further suggesting that exhibited a defect in cellobiose uptake as well as a severe com-
this protein may be important in bacteria-host associations. Addi- petitive defect in colonization. However, because this strain also
tional genetic characterization, including the generation of exhibited a growth defect, its specific role in colonization
mutants that lack all three pilA genes and combinations thereof, remains uncertain. These studies thus add to our knowledge
and understanding of the physiology of V. fischeri important to dehydrogenase) in the bacteria (Nishiguchi et al. 1998). How-
its ability to colonize its host E. scolopes and provide a cautionary ever, a subsequent study determined that the bacterial gene
note for those who rely on lacZ as a reporter. sequenced was not gapA, but rather epd, and reanalysis of the
data failed to support the original conclusions (Dunlap et al.
2007). Indeed, the latter study found no evidence from phylo-
Other Genes genetic analyses for parallel evolution in either squid/bacteria or
fish/bacteria symbioses. This may reflect the facultative nature of
The roles in symbiosis of a number of additional genes have been this symbiosis: V. fischeri is an extracellular symbiont acquired
investigated. Some, such as phosphoglucomutase (pgm), play anew from the environment in each generation of squid.
significant but poorly understood roles in symbiosis, while In a more recent study, Wollenberg and Ruby investigated
others have been shown to make small contributions, and yet whether the symbionts from the two geographically separated
others have an impact on symbiosis that cannot be separated populations of E. scolopes squid (in Maunalua Bay and Kaneohe
from their requirement for normal growth of V. fischeri. A select Bay, Oahu, Hawaii) could be phenotypically distinguished
set of these genes is included in > Table 20.1. As the study of (Wollenberg and Ruby 2009). Bacteria from the two populations
V. fischeri symbiosis continues and the interactions between of squid were collected over several nights and subsequently
V. fischeri and its host become better defined, the roles of these cultured to determine their phenotypes with respect to biolu-
genes may become better understood. minescence level, colony pigmentation, motility, growth rate,
and siderophore production. Except for siderophore produc-
tion, these characteristics could be used to distinguish the organ-
Evolution of the V. fischeri-E. scolopes isms isolated from the two squid populations in a manner that
Symbiosis was statistically significant. These data support the idea that the
squid host influences the generation of genetically distinct
Due to its experimental tractability, the V. fischeri-E. scolopes populations of V. fischeri (Wollenberg and Ruby 2009).
model has great potential to inform our understanding of evo- A subsequent study, using four bacterial housekeeping genes
lutionary relationships between bacteria and their specific hosts. (recA, mdh, katA, and pyrC), identified a monophyletic group
In the Hawaiian archipelago, at least two geographically isolated (‘‘group A’’) of strains that could be found at a higher frequency
and genetically distinct populations of E. scolopes exist than other strains in Maunalua squid (Wollenberg and Ruby
(Maunalua Bay and Kaneohe Bay, Oahu) (Jones et al. 2006; 2012). Group A was able to outcompete non-group A strains in
Kimbell et al. 2002). Furthermore, numerous related Vibrio- colonizing Maunalua squid hosts; however, this group appeared
containing squids can be found around the world, including to be at a disadvantage when free-living in the Maunalua Bay
those of the same genus (e.g., Euprymna tasmanica from Aus- environment (Wollenberg and Ruby 2012). These findings
tralia and Euprymna morsei from Japan) as well as those from the suggested a ‘‘fitness trade-off ’’ for growth in the host versus
distinct genus Sepiola (e.g., Sepiola affinis and Sepiola robusta survival in the environment.
from France). A wealth of questions can be addressed about Studies of Vibrio isolates obtained elsewhere, particularly in
symbiont specificity and evolution using phylogenetic and com- the Mediterranean where multiple sympatric squid species
parative analyses and experimental manipulation of established reside, suggest that they are genetically diverse (Jones et al.
symbiosis models. 2006; Nyholm and Nishiguchi 2008). Thus, factors other than
An early study analyzed coevolution of the bacteria and host availability may influence the population structure of the
squid. Specifically, Nishiguchi et al. (1998) asked whether bacteria (Nyholm and Nishiguchi 2008). Such factors may
V. fischeri strains isolated from E. scolopes (‘‘native’’ bacteria) include water flow, which can impact the distribution of the
were more successful at colonizing E. scolopes than strains iso- bacteria; salinity and temperature, which may exert selective
lated from other squid species (‘‘nonnative’’ bacteria). In com- pressures on the bacteria (Jones et al. 2007; Nishiguchi 2000;
petition assays in which a mixture of strains was used to Soto et al. 2009); and competition from other species
inoculate juvenile E. scolopes, the squid predominantly became (Wollenberg and Ruby 2012).
colonized with native strains rather than nonnative strains The phylogenetic tree constructed by Wollenberg and Ruby
(Nishiguchi et al. 1998). Furthermore, when two nonnative (2012) revealed that E. scolopes symbionts form a polyphyletic
strains competed, the strain that was more closely related to clade within V. fischeri (Wollenberg and Ruby 2012). These data
the native symbiont was also the better colonizer of juvenile represent an expansion of a similar data set generated by Mandel
E. scolopes. Parallel results were obtained from competition of et al. (2009). The Mandel study investigated host specificity by
strains from E. tasmanica and E. hyllebergi (Nishiguchi 2002). comparing the genomes of two V. fischeri strains, one isolated
These studies suggest that the bacteria have evolved to better from the squid (ES114) and the other isolated from the fish
colonize a specific host. The 1998 study also described evidence Monocentris japonica (MJ11); the latter strain fails to colonize
of parallel cladogenesis, as assessed using sequence divergence at squid proficiently (Mandel et al. 2009; McFall-Ngai and Ruby
two loci in related squids, the ITS (internal transcribed spacer 1991; Schuster et al. 2010). Genomic comparisons revealed that
between the 18S and 25S rRNA genes) and COI (cytochrome the two strains were quite similar: 91% of the genes encoded
oxidase subunit I) and gapA (glyceraldehyde phosphate proteins with a median amino acid identity of at least 99%
. Table 20.1
Genes whose roles in symbiosis have been tested
Gene(s) Description Sym phenotypea References

hmp Flavohaemoglobin, protects against NO stress Initiation (Wang et al. 2010b)
norV Flavohaemoglobin Initiation (Wang et al. 2010b)
htrB1 Lipid A acetyltransferase Initiation (Adin et al. 2008a)
rscS Sensor kinase, promotes biofilms Initiation (Visick and Skoufos 2001; Yip et al. 2006)
sypG Response regulator, promotes biofilms Initiation (Hussa et al. 2008; Hussa et al. 2007)
syp Polysaccharide biosynthesis, biofilms Initiation (Yip et al. 2006; Yip et al. 2005)
flab Flagella biosynthesis Initiation (Graf et al. 1994)
flrA Flagella biosynthesis Initiation (Millikan and Ruby 2003)
54
rpoN Alternative sigma factor s Initiation (Wolfe et al. 2004)
cheA, cheY Chemotaxis Initiation (DeLoney-Marino and Visick 2012; Mandel et
al. 2012)
sypE Serine kinase/phosphatase, inhibits biofilms Initiationc (Morris et al. 2011)
ompU Outer membrane protein Initiationd (Aeckersberg et al. 2001; Nyholm et al. 2009)
nagC N-acetyl-D-glucosamine repressor Initiation (Miyashiro et al. 2011)
hnoX NO sensor Initiatione (Wang et al. 2010a)
sapA Required for normal growth Initiation, (Lupp et al. 2002)
accommodation
ainS Pheromone signal synthase, luminescence Initiation, (Lupp et al. 2003; Lupp and Ruby 2004;
accommodation Lupp and Ruby 2005)
gacA Response regulator, various cellular functions Initiation, (Whistler et al. 2007; Whistler and Ruby 2003)
accommodationf
flaA Flagellin Initiation, (Millikan and Ruby 2004)
accommodationg
pgm Phosphoglucomutase, required for normal LPS Accommodation (DeLoney et al. 2002)
lysA Lysine biosynthesis Accommodation (Graf and Ruby 1998; Whistler et al. 2007)
thr Threonine biosynthesis Accommodation (Graf and Ruby 1998)
luxI, R, C-G Bioluminescence Persistence (Visick et al. 2000; Bose et al. 2008)
glnD Uridylyl-removing/uridylyltransferase Persistence (Graf and Ruby 2000)
luxS Pheromone signal synthase, luminescence Persistenceh (Lupp and Ruby 2004)
ltgA ltgD ltgY Lytic transglycosylase, promotes release of PG Secondary (Adin et al. 2009)
infections
argG Arginine biosynthesis Colonization (Graf and Ruby 1998)
cys Cysteine biosynthesis Colonization (Graf and Ruby 1998)
leu Leucine biosynthesis Colonization (Graf and Ruby 1998)
met Methionine biosynthesis Colonization (Graf and Ruby 1998)
pro Proline biosynthesis Colonization (Graf and Ruby 1998)
ser Serine biosynthesis Colonization (Graf and Ruby 1998)
tatABC Twin arginine translocation Competition (Dunn and Stabb 2008b)
luxO Response regulator, Luminescence Competition (Hussa et al. 2007; Lupp and Ruby 2005;
Miyashiro et al. 2010)
flrC Response regulator, motility Competition (Hussa et al. 2007)
cheY Response regulator, chemotaxis Competition (Hussa et al. 2007)
VF1909 Response regulator NarP Competition (Hussa et al. 2007)
vpsR Response regulator Competition (Hussa et al. 2007)
VFA0698 Response regulator CheV Competition (Hussa et al. 2007)
VFA0179 Response regulator Competition (Hussa et al. 2007)
VF1988 Response regulator PhoB Competition (Hussa et al. 2007)
Gene(s) Description Sym phenotypea References

VFA0181 Response regulator Competition (Hussa et al. 2007)
ntrC Response regulator Competition (Hussa et al. 2007)
VF1689 Response regulator ExpM Competition (Hussa et al. 2007)
arcA Response regulator, represses luminescence Competition (Bose et al. 2007)
cheR Chemotaxis regulator Competition (Deloney-Marino and Visick 2012)
qrr Regulatory mRNA, destabilizes litR transcript Competition (Miyashiro et al. 2010)
ptsI PTS enzyme EI Competition (Adin et al. 2008b)
acs Acetyl coenzyme A synthetase Competition (Studer et al. 2008)
VF1220-1228 Haemin uptake Competition (Septer et al. 2011)
katA Catalase, degrades hydrogen peroxide Competition (Visick and Ruby 1998)
pilA Pilin adhesin Competition (Stabb and Ruby 2003)
celI Cellobiose regulator Competition (Adin et al. 2008b)
pepN aminopeptidase Competition (Fidopiastis et al. 2012)
litR Transcriptional activator of luminescence Competitioni (Fidopiastis et al. 2002; Miyashiro et al. 2010)
fnr Regulator of anaerobic respiration None detected (Septer et al. 2010)
torECA torYZ Trimethylamine N-oxide reductase None detected (Dunn and Stabb 2008a)
dmsABC
nsrR NO-responsive regulator None detected (Wang et al. 2010b)
hvnA, hvnB NAD+ glycohydrolase (Halovibrin) None detected (Stabb et al. 2001)
phr Photolyase None detected (Walker et al. 2006)
flaC Flagellin None detected (Millikan and Ruby 2004)
celGKABC Cellobiose utilization None detected (Adin et al. 2008b)
htrB2 Lipid A acetyltransferase None detected (Adin et al. 2008a)
msbB Lipid A acetyltransferase None detected (Adin et al. 2008a)
glyA Glycine biosynthesis None detected (Graf and Ruby 1998)
ampG PG permease None detected (Adin et al. 2009)
a
Sym phenotype refers to the symbiosis phenotype. The main categories of symbiosis phenotypes include defects in initiation, accommodation, persistence, and
competition with wild-type cells. Where the defects have not been assigned to a category, the more general term of colonization is used. In some cases, multiple
categories are impacted by a mutation, and only the primary defect is listed. Specific notations are made for cases in which the mutation impacts other aspects of
the symbiosis, such as making the squid prone to secondary infection
b
Uncharacterized nonmotile transposon insertion mutants
c
sypE mutants that cannot be inactivated inhibit symbiotic initiation
d
ompU mutants also exhibit a defect in resisting binding and uptake by hemocytes
e
hnoX mutants exhibit a colonization advantage
f
gacA mutants are also defective in inducing host development
g
flaA mutants are also preferentially expelled from the light organ
h
luxS mutants only exhibit a phenotype when the mutation is combined with an ainS mutation
i
litR mutants exhibit a competitive advantage
(Mandel et al. 2009). Thus, there is limited diversity within this colonize E. scolopes (Mandel et al. 2009). This hypothesis was
group of bacteria, unlike in other organisms like some species experimentally addressed by introducing rscS into the fish sym-
within the Enterobacteriaceae, which contain large, diverse biont. This genetically modified strain became competent to
genomic islands (Welch et al. 2002). colonize squid. These experiments supported the conclusion
These studies also revealed the absence of the sensor kinase that rscS serves as a specificity factor that promotes colonization
gene rscS from the fish symbiont. Mandel et al. (2009) proposed of E. scolopes by the subset of strains that contain it.
that the absence of rscS, required for initiation of symbiosis by Mandel et al. (2009) subsequently investigated the origin of
ES114 (Visick and Skoufos 2001; Yip et al. 2006), could be rscS in V. fischeri strains and determined that rscS likely arose
sufficient to account for the inability of the fish symbiont to in a single horizontal acquisition event from an unknown donor
prior to the introduction of V. fischeri into squid in the North mimicked the natural cycle of the symbiosis through initiation,
Pacific Ocean. Subsequent transmission of rscS resulted in two growth and persistence, venting, and persistence in the seawater.
major alleles of rscS, termed rscSA and rscSB. Squid symbionts Importantly, this approach selected for improved traits at
contained rscSA, while half of the fish symbionts lacked rscS a number of different stages, rather than simply one, such as
altogether. The half of the fish symbionts that contained rscS initiation. It was estimated that the bacteria in these experiments
largely carried the rscSB allele; the rscSB allele was not sufficient underwent a total of between 290 and 360 symbiotic generations.
to promote colonization of squid (Mandel et al. 2009). These During this limited experimental evolution, descendents of
data suggested that these rscS-containing fish symbionts the two nonnative strains, WH1 and MJ11, exhibited clear
descended from squid symbionts, and these bacteria either dis- phenotypic differences from their parents, notably with respect
pensed with or modified rscS. These experiments highlight the to luminescence output (Schuster et al. 2010). In contrast to the
power of whole-genome comparisons and the utility of combin- native symbiont ES114, which is non-visibly luminescent, WH1
ing phylogenetic analyses with experimental measurements of and MJ11 produce high levels of visible light. Evolved derivatives
colonization. of WH1 and MJ11 showed convergent evolution toward reduced
One major exception to the striking similarity between the light production. Specifically, in each of six independently
ES114 and MJ11 genomes was found in the bioluminescence evolved lines of MJ11, non-visibly luminescent isolates were
genes (Bose et al. 2011; Mandel et al. 2009). Although the genes observed; in four of these lines, all of the isolates examined
and their arrangement were conserved, the proteins encoded by were non-visibly luminescent. For WH1, four of the six lines
luxR/I and luxCDABE had only between 75% and 89.5% identity yielded strains with decreased bioluminescence. Intriguingly,
at the amino acid level, whereas most other proteins, including although the decrease in light production of the evolved WH1
those whose genes flank the lux operon, were more than 95% strains was not substantial (approximately twofold) when light
identical. Moreover, the intergenic region between luxR and the production was measured in vitro, the same strains exhibited
lux operon showed a much greater divergence than intergenic about a 50-fold decrease in symbiotic (in vivo) bioluminescence
regions of nearby genes (Bose et al. 2011). A comparison of the per cell. These data supported the hypothesis that the decrease in
luxRI intergenic region of 18 V. fischeri strains revealed that only light production was a meaningful change in the context of
104 of the 222 base-pair positions were conserved in all of the symbiosis. To test this idea further, Schuster et al. (2010) asked
strains. When these sequences were used to generate a gene tree, whether decreased luminescence was an advantage or disadvan-
two distinct clades were revealed, one of which encompassed all tage by exposing squid to two strains (from the same evolved
the strains classified as having highly visible luminescence. The population) that exhibited different luminescence levels (ances-
second clade included all of the isolates from E. scolopes, which tral or decreased luminescence). They found that the strain with
are non-visibly luminescent in culture, as well as isolates from decreased luminescence outcompeted the one with ancestral
E. tasmanica and E. morsei and a few planktonic isolates. To luminescence levels, suggesting that this evolved phenotype
understand the significance of the diversity within this may in fact be advantageous during the symbiotic life cycle of
intergenic region, Bose et al. (2011) compared it to another E. scolopes.
intergenic region, that between glpA and fdhA. The latter Together, these data indicate that decreased luminescence is
intergenic region was conserved at 97% of the positions (289 one adaptation V. fischeri makes rapidly to become a proficient
of 298 bp). These data are consistent with the conclusion that the colonizer of the E. scolopes host. Furthermore, although it is clear
lux intergenic region is under different selective pressure— that the squid select for strains that are proficient to produce
presumably the host environment—that has led to a relatively light (Visick et al. 2000; Wollenberg et al. 2012), it seems likely
rapid evolution of bioluminescence genes and regulatory regions that there is additional selection to maintain a luminescence
(Bose et al. 2011). level that is not excessive (Schuster et al. 2010). Future
The above studies have been nicely complemented by an work that assesses the full complement of changes that occur
experimental evolution approach reported by Schuster et al. during the experimental evolution, and a determination of
(2010). In this study, juvenile E. scolopes were exposed to two which changes make important contributions to colonization
nonnative strains of V. fischeri: MJ11, the fish symbiont proficiency, will provide great insights into the mechanisms of
described above, and WH1, a free-living strain isolated from symbiont evolution and the selective forces that direct these
a region of the USA that does not harbor these squid. Although processes.
not native to E. scolopes, these V. fischeri isolates could colonize
given a high inoculation dose and enough time. These strains
then were serially passaged through 14 additional hatchling Perspectives and Future of the Field
squid as follows. Every day, squid were rinsed and moved to
fresh seawater. On the third day following inoculation, the sea- The V. fischeri-E. scolopes symbiosis is a powerful experimental
water into which bacteria were vented by colonized animals was model for elucidating bacteria-host interactions and the impact
used as the inoculum for the next newly hatched squid. Thus, of bacteria on host development. One strength of the system
with the exception of the initial inoculation with cultured bac- comes from the fact that research into squid biology and how
teria, these experiments were culture-independent and bacteria influence host development informs investigations into
the roles of bacteria in symbiosis and vice versa. The result is

a synergy of inquiry that has resulted in great advances in our
understanding.
Another reason that work on the system has progressed so
readily is the continued development of genetic tools that greatly
facilitate the manipulation of the bacteria. Early work focused
on the generation of V. fischeri mutants using such blunt instru-
ments as transposon mutagenesis, gene replacement (with
corresponding insertion of antibiotic resistance cassettes), and
insertional mutagenesis. These tools were critical at the time and
are still used to great advantage. However, it is now also possible
to readily and rapidly generate in-frame deletions or incorporate
targeted point mutations into the V. fischeri chromosome.
Refinement of the ES114 genome database combined with
whole-genome resequencing also now allows researchers to pin-
point spontaneous mutations and to track the genetic changes in
. Fig. 20.10
evolved derivatives of the wild type. This new technology could
A molecular geneticist’s perspective. The tube containing purified
solve old puzzles. For example, genomic sequencing of the
ES114 DNA (pellet at the bottom of the tube), used for generating
hypermotile strains isolated by Millikan and Ruby (2002) may
the first V. fischeri genome, is shown with Hawaiian palm trees
lead us to identify new regulators of motility and symbiosis,
visible in the background
while sequencing the visibly bioluminescent derivative of ES114
(Dunlap et al. 1995) may uncover novel pathways of lumines-
cence regulation. Taken together, this expanding genetic toolbox plasmids has also raised the possibility that genetic information
will facilitate ever more refined investigations of the roles of is being exchanged through conjugation between different
specific genes and their products during symbiosis. Similarly, strains during symbiosis. Similarly, the development of natural
in the future, the current construction of a comprehensive transformation as a tool not only provided another method for
library of catalogued mutants, encompassing disruptions of genetic manipulation of V. fischeri but also suggested the possi-
nonessential genes, will provide an invaluable resource for bility of DNA uptake and recombination during symbiosis
mutant analyses. (Pollack-Berti et al. 2010). Taken together, one wonders whether
Another critical advance has been the development of GFP future investigations will find that the packed confines of the
both as a marker, to visualize where the bacterial cells are in the light organ could be a hot spot for horizontal gene transfer.
symbiosis, and as a reporter, to provide information on gene As in many bacterial systems over the last decades, genomic
expression in symbiotic cells in real time. Our understanding of approaches have transformed our lens into this symbiosis. The
the establishment of this symbiosis stems largely from the sem- availability of the genome sequence of V. fischeri strain ES114
inal work of Nyholm et al. (2000), who first used GFP-labeled (> Fig. 20.10), and subsequently that of the fish symbiont strain
cells to visually discover aggregation and other early infection MJ11, made it possible to use bioinformatics to make predic-
events. Subsequent work with V. fischeri expressing proteins with tions about the importance of genes in the symbiosis and to
red fluorescence enabled the visualization of two infecting readily clone wild-type and mutant alleles of specific genes.
strains at once (Dunn et al. 2006) and provides a fluorescent Comparative genomic studies will become even more important
standard against which GFP expression can be normalized as the sequences of additional genomes become available, and as
(Miyashiro et al. 2010). Similarly, researchers studying the the studies of symbiotic associations of V. fischeri with other
squid have pioneered a number of fluorescence-based micro- squid species (as well as studies of other bacteria-light-organ
scopic methods to investigate processes such as apoptosis or symbioses) become more developed (Guerrero-Ferreira and
specific proteins in the host. The transparent tissue of the light Nishiguchi 2009; Nyholm and Nishiguchi 2008). Likewise, the
organ affords a rare opportunity to use fluorescence-based squid ESTand proteome databases have already provided insight
methods to visualize symbiotic processes in intact tissues, and into how the host detects and responds to V. fischeri, and future
the development of other fluorophores, destabilized GFP deriv- work is likely to expand these resources. Additional molecular
atives, and various animal-tissue stains should continue to be technology grounded in genomics, bioinformatics, and high-
exploited to great effect. throughput methods—from arrays to RNAseq to single-cell ana-
An increased basic understanding of V. fischeri genetic pro- lyses—holds great promise for elucidating bacterial responses to
cesses has both provided tools and led to new intriguing ques- the colonization process (and the corresponding responses of the
tions. For example, the identification of a replication origin from squid). We anticipate that the next decade will see a great expan-
a plasmid endogenous to V. fischeri permitted the construction sion in the availability and productive use of these approaches.
of stable shuttle plasmids that could be readily used for a variety In some cases, knowledge from the system has been slow to
of experiments such as complementation or the strain-tagging impact the conduct of current research. For example, it has been
with GFP, described above. The description of native V. fischeri known for a long time that juvenile squid have six pores and six
distinct crypts, and we now know that it is rare for those crypts approaches, and knowledge of the system, it seems likely that
to be highly colonized by multiple strains: in competition exper- many of the current secrets shortly will be revealed, but also that
iments using two differently labeled strains, most crypts contain many more will present themselves.
one or the other strain but usually not both (> Fig. 20.4) (Dunn
et al. 2006). Furthermore, the extent to which the different
crypts expel their symbionts in the morning varies. For the Acknowledgments
most part, this information has not yet impacted our straight-
forward interpretation of competition experiments: if more cells We thank Jonathan Visick, Andrew Weir, and Mark Mandel for
of one strain are obtained following a competitive colonization reading and advising on parts of the manuscript, and members
experiment, then we conclude that that strain had a colonization of our labs for feedback on the manuscript. We also thank
advantage. But, did this advantage flow from superior entry into individuals who contributed unpublished images, including
multiple crypts, superior binding within a single crypt relative to Spencer Nyholm, Andrew Collins, Jeffrey Bose, Kati Geszvain,
another strain, being lucky enough to colonize the right crypt, or Satoshi Shibata, Emily Yip, and Cindy DeLoney-Marino. Work
some other reason? Follow-up experiments that visualize the on the Vibrio fischeri-Euprymna scolopes symbiosis in the labo-
colonized animals using fluorescently labeled strains will ratories of the authors was funded by NSF grant IOS-1121106 to
provide better insights into the requirements for a particular EVS and by NIH grant GM59690 awarded to KLV.
gene/trait.
In the early days of investigating this symbiosis, insights
from other systems—and especially pathogenic systems—were References
used to develop models for how V. fischeri and E. scolopes
interact. For example, it made sense to ask whether flagella Adin DM, Engle JT, Goldman WE, McFall-Ngai MJ, Stabb EV (2009) Mutations
were important for colonization, as this was the case for many in ampG and lytic transglycosylase genes affect the net release of peptidogly-
can monomers from Vibrio fischeri. J Bacteriol 191:2012–2022
pathogenic associations. As the field has developed, however,
Adin DM, Phillips NJ, Gibson BW, Apicella MA, Ruby EG, McFall-Ngai MJ,
our studies have begun to provide new insights that can inform Hall DB, Stabb EV (2008a) Characterization of htrB and msbB mutants
the work of others. Notably, the identification of H-NOX as of the light organ symbiont Vibrio fischeri. Appl Environ Microbiol
a NO sensor and NO as a potential signal has clear implications 74:633–644
for pathogenic associations (Wang et al. 2010a). Similarly, the Adin DM, Visick KL, Stabb EV (2008b) Identification of a cellobiose utilization
gene cluster with cryptic b-galactosidase activity in Vibrio fischeri. Appl
novel nuclear localization of EsPGRP1 and the role of the sym-
Environ Microbiol 74:4059–4069
bionts in changing the localization of receptors for MAMPS have Aeckersberg F, Lupp C, Feliciano B, Ruby EG (2001) Vibrio fischeri outer mem-
potential to impact studies of the immune response (Troll et al. brane protein OmpU plays a role in normal symbiotic colonization.
2009a, b). A third example lies in the paucity of good animal J Bacteriol 183:6590–6597
models for studying bacterial biofilms; the squid symbiosis Alcaide E (2003) Numerical taxonomy of Vibrionaceae isolated from cultured
amberjack (Seriola dumerili) and surrounding water. Curr Microbiol
makes a robust model, as the parallels between phenotypes
46:184–189
observed in culture and during symbiosis are very strong, and Altura MA, Stabb E, Goldman W, Apicella M, McFall-Ngai MJ (2011) Attenuation
thus the potential for informing the biofilm field is high (Morris of host NO production by MAMPs potentiates development of the host in
et al. 2011; Yip et al. 2006). the squid-vibrio symbiosis. Cell Microbiol 13:527–537
Many questions remain. Although much is known about Anderson RC, Mather JA (1996) Escape responses of Euprymna scolopes Berry,
1911 (Cephalopoda: Sepiolidae). J Moll Stud 62:543–545
conditions and molecules that influence the activities of these
Antunes LC, Ferreira RB, Lostroh CP, Greenberg EP (2008) A mutational analysis
organisms and their interactions, including TCT, PG, phero- defines Vibrio fischeri LuxR binding sites. J Bacteriol 190:4392–4397
mones, and light itself, numerous bacterial regulators have Antunes LC, Schaefer AL, Ferreira RB, Qin N, Stevens AM, Ruby EG, Greenberg
been identified for which the signal transduction cascade EP (2007) Transcriptome analysis of the Vibrio fischeri LuxR-LuxI regulon.
remains to be elucidated. For example, the sensor kinase RscS J Bacteriol 189:8387–8391
Arcondeguy T, Jack R, Merrick M (2001) P(II) signal transduction proteins,
was identified over 10 years ago, but the signal to which it
pivotal players in microbial nitrogen control. Microbiol Mol Biol Rev:
responds during symbiosis remains elusive. Whether V. fischeri MMBR 65:80–105
recognizes a chemotactic signal to guide it into the light organ Baker MD, Wolanin PM, Stock JB (2006) Signal transduction in bacterial che-
and what drives the loss of flagella in colonizing bacteria are motaxis. BioEssays 28:9–22
among the longstanding intriguing questions about coloniza- Barnhart MM, Chapman MR (2006) Curli biogenesis and function. Annu Rev
Microbiol 60:131–147
tion. Indeed, we have yet to truly understand how specificity in
Berry SS (1912) The Cephalopoda of the Hawaiian islands. In: Bulletin of the
this symbiosis is achieved; perhaps there is no one single factor United States Bureau of Fisheries. Government Printing Office, Washington,
but a multitude of them, such as the ability to form a biofilm, pp 255–362
sense a chemotactic signal, resist host defenses, and induce the Boettcher KJ, Ruby EG (1990) Depressed light emission by symbiotic Vibrio
right set of responses, that together result in the successful fischeri of the sepiolid squid Euprymna scolopes. J Bacteriol 172:3701–3706
Boettcher KJ, Ruby EG (1994) Occurrence of plasmid DNA in the sepiolid squid
symbiont. The opportunity to be the sole occupant of
symbiont Vibrio fischeri. Curr Microbiol 29:279–286
a ‘‘privileged’’ niche is a powerful incentive, and V. fischeri has Boettcher KJ, Ruby EG, McFall-Ngai MJ (1996) Bioluminescence in the symbiotic
clearly evolved to make use of this prime opportunity for self- squid Euprymna scolopes is controlled by a daily biological rhythm. J Comp
advancement. With the increasing sophistication of the tools, Physiol 179:65–73
Bose JL, Kim U, Bartkowski W, Gunsalus RP, Overley AM, Lyell NL, Visick KL, DeLoney-Marino CR, Wolfe AJ, Visick KL (2003) Chemoattraction of Vibrio
Stabb EV (2007) Bioluminescence in Vibrio fischeri is controlled by the fischeri to serine, nucleosides and N-acetylneuraminic acid, a component
redox-responsive regulator ArcA. Mol Microbiol 65:538–553 of squid light organ mucus. Appl Environ Microbiol 69:7527–7530
Bose JL, Rosenberg CS, Stabb EV (2008) Effects of luxCDABEG induction in DeLoney CR, Bartley TM, Visick KL (2002) Role for phosphoglucomutase in
Vibrio fischeri: enhancement of symbiotic colonization and conditional Vibrio fischeri-Euprymna scolopes symbiosis. J Bacteriol 184:5121–5129
attenuation of growth in culture. Arch Microbiol 190(2):169–183 Doino JA, McFall-Ngai MJ (1995) A transient exposure to symbiosis-competent
Bose JL, Wollenberg MS, Colton DM, Mandel MJ, Septer AN, Dunn AK, Stabb EV bacteria induces light organ morphogenesis in the host squid. Biol Bull
(2011) Contribution of rapid evolution of the luxR-luxI intergenic region to 189:347–355
the diverse bioluminescence outputs of Vibrio fischeri strains isolated from Doino Lemus J, McFall-Ngai MJ (2000) Alterations in the proteome of the
different environments. Appl Environ Microbiol 77:2445–2457 Euprymna scolopes light organ in response to symbiotic Vibrio fischeri.
Bourgois JJ, Sluse FE, Baguet F, Mallefet J (2001) Kinetics of light emission and Appl Environ Microbiol 66:4091–4097
oxygen consumption by bioluminescent bacteria. J Bioenerg Biomembr Dunlap PV, Ast JC, Kimura S, Fukui A, Yoshino T, Endo H (2007) Phylogenetic
33:353–363 analysis of host-symbiont specificity and codivergence in bioluminescent
Boylan M, Miyamoto C, Wall L, Grahm A, Meighen E (1989) Lux C, D, and symbioses. Cladistics 23:507–532
E genes of the Vibrio fischeri luminescence operon code for the reductase, Dunlap PV, Greenberg EP (1985) Control of Vibrio fischeri luminescence gene
transferase, and synthetase enzymes involved in aldehyde biosyntheses. expression in Escherichia coli by cyclic AMP and cyclic AMP receptor protein.
Photochem Photobiol 49:681–688 J Bacteriol 164:45–50
Browne-Silva J, Nishiguchi MK (2008) Gene sequences of the pil operon reveal Dunlap PV, Greenberg EP (1988) Control of Vibrio fischeri lux gene transcription
relationships between symbiotic strains of Vibrio fischeri. Int J Syst by a cyclic AMP receptor protein-LuxR protein regulatory circuit. J Bacteriol
Evol Microbiol 58:1292–1299 170:4040–4046
Buck M, Gallegos M-T, Studholme DJ, Guo Y, Gralla JD (2000) The bacterial Dunlap PV, Kita-Tsukamoto K, Waterbury JB, Callahan SM (1995) Isolation and
enhancer-dependent s54 (sN) transcription factor. J Bacteriol 182: characterization of a visibly luminous variant of Vibrio fischeri strain ES114
4129–4136 from the sepiolid squid Euprymna scolopes. Arch Microbiol 164:194–202
Burrows LL (2005) Weapons of mass retraction. Mol Microbiol 57:878–888 Dunlap PV, Ray JM (1989) Requirement for autoinducer in transcriptional
Callahan SM, Dunlap PV (2000) LuxR- and acyl-homoserine-lactone-controlled negative autoregulation of the Vibrio fischeri luxR gene in Escherichia coli.
non-lux genes define a quorum sensing regulon in Vibrio fischeri. J Bacteriol J Bacteriol 171:3549–3552
182:2811–2822 Dunn AK, Karr EA, Wang Y, Batton AR, Ruby EG, Stabb EV (2010) The alterna-
Castillo MG, Goodson MS, McFall-Ngai M (2009) Identification and molecular tive oxidase (AOX) gene in Vibrio fischeri is controlled by NsrR and
characterization of a complement C3 molecule in a lophotrochozoan, the upregulated in response to nitric oxide. Mol Microbiol 77:44–55
Hawaiian bobtail squid Euprymna scolopes. Dev Comp Immunol 33:69–76 Dunn AK, Martin MO, Stabb E (2005) Characterization of pES213, a small
Chen X, Schauder S, Potier N, Van Dorsselaer A, Pelczer I, Bassler BL, Hughson mobilizable plasmid from Vibrio fischeri. Plasmid 54:114–134
FM (2002) Structural identification of a bacterial quorum-sensing signal Dunn AK, Millikan DS, Adin DM, Bose JL, Stabb EV (2006) New rfp- and
containing boron. Nature 415:545–549 pES213-derived tools for analyzing symbiotic Vibrio fischeri reveal
Chevance FF, Hughes KT (2008) Coordinating assembly of a bacterial macromo- patterns of infection and lux expression in situ. Appl Environ Microbiol
lecular machine. Nat Rev Microbiol 6:455–465 72:802–810
Chun CK, Scheetz TE, Bonaldo Mde F, Brown B, Clemens A, Crookes-Goodson Dunn AK, Stabb EV (2007) Beyond quorum sensing: the complexities of pro-
WJ, Crouch K, DeMartini T, Eyestone M, Goodson MS, Janssens B, Kimbell karyotic parliamentary procedures. Anal Bioanal Chem 387:391–398
JL, Koropatnick TA, Kucaba T, Smith C, Stewart JJ, Tong D, Troll JV, Webster Dunn AK, Stabb EV (2008a) Genetic analysis of trimethylamine N-oxide
S, Winhall-Rice J, Yap C, Casavant TL, McFall-Ngai MJ, Soares MB reductases in the light organ symbiont Vibrio fischeri ES114. J Bacteriol
(2006) An annotated cDNA library of juvenile Euprymna scolopes with 190:5814–5823
and without colonization by the symbiont Vibrio fischeri. BMC Genomics Dunn AK, Stabb EV (2008b) The twin arginine translocation system contributes
7:154 to symbiotic colonization of Euprymna scolopes by Vibrio fischeri. FEMS
Chun CK, Troll JV, Koroleva I, Brown B, Manzella L, Snir E, Almabrazi H, Scheetz Microbiol Lett 279:251–258
TE, Bonaldo Mde F, Casavant TL, Soares MB, Ruby EG, McFall-Ngai MJ Duperthuy M, Schmitt P, Garzon E, Caro A, Rosa RD, Le Roux F, Lautredou-
(2008) Effects of colonization, luminescence, and autoinducer on host tran- Audouy N, Got P, Romestand B, de Lorgeril J, Kieffer-Jaquinod S, Bachere E,
scription during development of the squid-vibrio association. Proc Natl Destoumieux-Garzon D (2011) Use of OmpU porins for attachment and
Acad Sci USA 105:11323–11328 invasion of Crassostrea gigas immune cells by the oyster pathogen Vibrio
Claes MF, Dunlap PV (2000) Aposymbiotic culture of the sepiolid squid splendidus. Proc Natl Acad Sci USA 108:2993–2998
Euprymna scolopes: role of the symbiotic bacterium Vibrio fischeri in host Egland KA, Greenberg EP (1999) Quorum sensing in Vibrio fischeri: elements of
animal growth, development, and light organ morphogenesis. J Exp Zool the luxl promoter. Mol Microbiol 31:1197–1204
286:280–296 Egland KA, Greenberg EP (2000) Conversion of the Vibrio fischeri transcriptional
Correa NE, Lauriano CM, McGee R, Klose KE (2000) Phosphorylation of the activator, LuxR, to a repressor. J Bacteriol 182:805–811
flagellar regulatory protein FlrC is necessary for Vibrio cholerae motility and Engebrecht J, Nealson K, Silverman M (1983) Bacterial bioluminescence: isola-
enhanced colonization. Mol Microbiol 35:743–755 tion and genetic analysis of functions from Vibrio fischeri. Cell 32:773–781
Correa NE, Peng F, Klose KE (2005) Roles of the regulatory proteins FlhF and Engebrecht J, Silverman M (1984) Identification of genes and gene products
FlhG in the Vibrio cholerae flagellar transcription hierarchy. J Bacteriol necessary for bacterial bioluminescence. Proc Natl Acad Sci USA
187:6324–6332 81:4154–4158
Czyz A, Plata K, Wegrzyn G (2003) Stimulation of DNA repair as an evolutionary Fidopiastis PM, Miyamoto CM, Jobling MG, Meighen EA, Ruby EG (2002) LitR,
drive for bacterial luminescence. Lumin: J Biol Chem Lumin 18:140–144 a new transcriptional activator in Vibrio fischeri, regulates luminescence and
Darnell CL, Hussa EA, Visick KL (2008) The putative hybrid sensor kinase SypF symbiotic light organ colonization. Mol Microbiol 45:131–143
coordinates biofilm formation in Vibrio fischeri by acting upstream of two Fidopiastis PM, von Boletzky S, Ruby EG (1998) A new niche for Vibrio logei, the
response regulators, SypG and VpsR. J Bacteriol 190:4941–4950 predominant light organ symbiont of squids in the genus Sepiola. J Bacteriol
Davidson SK, Koropatnick TA, Kossmehl R, Sycuro L, McFall-Ngai MJ (2004) 180:59–64
NO means ‘yes’ in the squid-vibrio symbiosis: nitric oxide (NO) during the Fidopiastis PM, Rader BA, Gerling DG, Gutierrez NA, Watkins KH, Frey MW,
initial stages of a beneficial association. Cell Microbiol 6:1139–1151 Nyholm SV, Whistler CA (2012) Characterization of a Vibrio fischeri ami-
Deloney-Marino CR, Visick KL (2012) Role for cheR of Vibrio fischeri in the nopeptidase and evidence for its influence on an early stage of squid colo-
Vibrio-squid symbiosis. Can J Microbiol 58:29–38 nization. J Bacteriol 194:3995–4002
Fitzgerald JM (1977) Classification of luminous bacteria from the light organ of the Jones BW, Nishiguchi MK (2004) Counterillumination in the Hawaiian bobtail
Australian pinecone fish, Cleidopus gloriamaris. Arch Microbiol 112:153–156 squid, Euprymna scolopes Berry (Mollusca: Cephalopoda). Mar Biol
Foster JS, Apicella MA, McFall-Ngai MJ (2000) Vibrio fischeri lipopolysaccharide 144:1151–1155
induces developmental apoptosis, but not complete morphogenesis, of the Kaplan HB, Greenberg EP (1985) Diffusion of autoinducer is involved in regu-
Euprymna scolopes symbiotic light organ. Dev Biol 226:242–254 lation of the Vibrio fischeri luminescence system. J Bacteriol 163:1210–1214
Foster JS, McFall-Ngai MJ (1998) Induction of apoptosis by cooperative bacteria Keynan A, Hastings JW (1961) The isolation and characterization of dark
in the morphogenesis of host epithelial tissues. Dev Genes Evol 208:295–303 mutants of luminous bacteria. Biol Bull 121:375
Friedrich WF, Greenberg EP (1983) Glucose repression of luminescence and Kimbell JR, McFall-Ngai MJ (2003) The squid-Vibrio symbiosis: From demes to
luciferase in Vibrio fischeri. Arch Microbiol 134:87–91 genes. Integr Comp Biol 43:254–260
Fuqua C, Winans SC, Greenberg EP (1996) Census and consensus in bacterial Kimbell JR, McFall-Ngai MJ (2004) Symbiont-induced changes in host actin
ecosystems: the LuxR-LuxI family of quorum-sensing transcriptional regu- during the onset of a beneficial animal-bacterial association. Appl Environ
lators. Annu Rev Microbiol 50:727–751 Microbiol 70:1434–1441
Fuqua WC, Winans SC, Greenberg EP (1994) Quorum sensing in bacteria: the Kimbell JR, McFall-Ngai MJ, Roderick GK (2002) Two genetically distinct
LuxR-LuxI family of cell density-responsive transcriptional regulators. populations of bobtail squid, Euprymna scolopes, exist on the island of
J Bacteriol 176:269–275 Oahu. Pacific Sci 56:347–355
Garcia-Amado MA, Bozo-Hurtado L, Astor Y, Suarez P, Chistoserdov A (2011) Koropatnick TA, Engle JT, Apicella MA, Stabb EV, Goldman WE, McFall-Ngai MJ
Denaturing gradient gel electrophoresis analyses of the vertical distribution (2004) Microbial factor-mediated development in a host-bacterial mutual-
and diversity of Vibrio spp. populations in the Cariaco Basin. FEMS ism. Science 306:1186–1187
Microbiol Ecol 77:347–356 Koropatnick TA, Kimbell JR, McFall-Ngai MJ (2007) Responses of host hemo-
Goodson MS, Kojadinovic M, Troll JV, Scheetz TE, Casavant TL, Soares MB, cytes during the initiation of the squid-Vibrio symbiosis. Biol Bull 212:29–39
McFall-Ngai MJ (2005) Identifying components of the NF-kB pathway in Krasity BC, Troll JV, Weiss JP, McFall-Ngai MJ (2011) LBP/BPI proteins and their
the beneficial Euprymna scolopes-Vibrio fischeri light organ symbiosis. Appl relatives: conservation over evolution and roles in mutualism. Biochem Soc
Environ Microbiol 71:6934–6946 Trans 39:1039–1044
Graf J, Dunlap PV, Ruby EG (1994) Effect of transposon-induced motility Kulkarni PR, Cui X, Williams JW, Stevens AM, Kulkarni RV (2006) Prediction of
mutations on colonization of the host light organ by Vibrio fischeri. CsrA-regulating small RNAs in bacteria and their experimental verification
J Bacteriol 176:6986–6991 in Vibrio fischeri. Nucleic Acids Res 34:3361–3369
Graf J, Ruby EG (1998) Host-derived amino acids support the proliferation of Kuo A, Blough NV, Dunlap PV (1994) Multiple N-acyl-L-homoserine lactone
symbiotic bacteria. Proc Natl Acad Sci USA 95:1818–1822 autoinducers of luminescence in the marine symbiotic bacterium Vibrio
Graf J, Ruby EG (2000) Novel effects of a transposon insertion in the Vibrio fischeri. J Bacteriol 176:7558–7565
fischeri glnD gene: defects in iron uptake and symbiotic persistence in Kuo A, Callahan SM, Dunlap PV (1996) Modulation of luminescence operon
addition to nitrogen utilization. Mol Microbiol 37:168–179 expression by N-octanoyl-L-homoserine lactone in ainS mutants of Vibrio
Gray KM, Greenberg EP (1992a) Physical and functional maps of the lumines- fischeri. J Bacteriol 178:971–976
cence gene cluster in an autoinducer-deficient Vibrio fischeri strain isolated Lamarcq LH, McFall-Ngai MJ (1998) Induction of a gradual, reversible morpho-
from a squid light organ. J Bacteriol 174:4384–4390 genesis of its host’s epithelial brush border by Vibrio fischeri. Infect Immun
Gray KM, Greenberg EP (1992b) Sequencing and analysis of luxR and luxI, the 66:777–785
luminescence regulatory genes from the squid light organ symbiont Vibrio Lamas J, Anadon R, Devesa S, Toranzo AE (1990) Visceral neoplasia and epider-
fischeri ES114. Mol Mar Biol Biotechnol 1:414–419 mal papillomas in cultured turbot Scophthalmus maximus. Dis Aquat Organ
Guerrero-Ferreira RC, Nishiguchi MK (2009) Ultrastructure of light organs of 8:179–187
loliginid squids and their bacterial symbionts: a novel model system for the Lapouge K, Schubert M, Allain FH, Haas D (2008) Gac/Rsm signal transduction
study of marine symbioses. Vie Milieu Paris 59:307–313 pathway of gamma-proteobacteria: from RNA recognition to regulation of
Hanlon RT, Claes MF, Ashcraft SE, Dunlap PV (1997) Laboratory culture of the social behaviour. Mol Microbiol 67:241–253
sepiolid squid Euprymna scolopes: a model system for bacteria-animal sym- Lee K-H, Ruby EG (1992) Detection of the light organ symbiont, Vibrio fischeri, in
biosis. Biol Bull 192:364–374 Hawaiian seawater by using lux gene probes. Appl Environ Microbiol
Hanzelka BL, Parsek MR, Val DL, Dunlap PV, Cronan JE Jr, Greenberg EP 58:942–947
(1999) Acylhomoserine lactone synthase activity of the Vibrio fischeri AinS Lee K-H, Ruby EG (1994a) Competition between Vibrio fischeri strains during
protein. J Bacteriol 181:5766–5770 initiation and maintenance of a light organ symbiosis. J Bacteriol 176:
Hastings JW, Nealson KH (1977) Bacterial bioluminescence. Annu Rev Microbiol 1985–1991
31:549–595 Lee K-H, Ruby EG (1994b) Effect of the squid host on the abundance and
Hastings JW, Riley WH, Massa J (1965) The purification, properties, and distribution of symbiotic Vibrio fischeri in nature. Appl Environ Microbiol
chemiluminescent quantum yield of bacterial luciferase. J Biol Chem 240: 60:1565–1571
1473–1481 Liu X, Parales RE (2008) Chemotaxis of Escherichia coli to pyrimidines: a new role
Haygood MG, Nealson KH (1985) Mechanisms of iron regulation of lumines- for the signal transducer Tap. J Bacteriol 190:972–979
cence in Vibrio fischeri. J Bacteriol 162:209–216 Liu X, Wood PL, Parales JV, Parales RE (2009) Chemotaxis to pyrimidines and
Heath-Heckman EA, McFall-Ngai MJ (2011) The occurrence of chitin in the identification of a cytosine chemoreceptor in Pseudomonas putida.
hemocytes of invertebrates. Zoology (Jena) 114:191–198 J Bacteriol 191:2909–2916
Hussa EA, Darnell CL, Visick KL (2008) RscS functions upstream of SypG to Lupp C, Hancock RE, Ruby EG (2002) The Vibrio fischeri sapABCDF locus is
control the syp locus and biofilm formation in Vibrio fischeri. J Bacteriol required for normal growth, both in culture and in symbiosis. Arch
190:4576–4583 Microbiol 179:57–65
Hussa EA, O’Shea TM, Darnell CL, Ruby EG, Visick KL (2007) Two-component Lupp C, Ruby EG (2004) Vibrio fischeri LuxS and AinS: Comparative study of two
response regulators of Vibrio fischeri: identification, mutagenesis and char- signal synthases. J Bacteriol 186:3873–3881
acterization. J Bacteriol 189:5825–5838 Lupp C, Ruby EG (2005) Vibrio fischeri utilizes two quorum-sensing systems
Jones BW, Lopez JE, Huttenburg J, Nishiguchi MK (2006) Population structure for the regulation of early and late colonization factors. J Bacteriol 187:
between environmentally transmitted vibrios and bobtail squids using 3620–3629
nested clade analysis. Mol Ecol 15:4317–4329 Lupp C, Urbanowski M, Greenberg EP, Ruby EG (2003) The Vibrio fischeri
Jones BW, Maruyama A, Ouverney CC, Nishiguchi MK (2007) Spatial and quorum-sensing systems ain and lux sequentially induce luminescence
temporal distribution of the vibrionaceae in coastal waters of Hawaii, Aus- gene expression and are important for persistence in the squid host. Mol
tralia, and France. Microb Ecol 54:314–323 Microbiol 50:319–331
Lyell NL, Dunn AK, Bose JL, Stabb EV (2010) Bright mutants of Vibrio fischeri Nairz M, Schroll A, Sonnweber T, Weiss G (2010) The struggle for iron - a metal at
ES114 reveal conditions and regulators that control bioluminescence and the host-pathogen interface. Cell Microbiol 12:1691–1702
expression of the lux operon. J Bacteriol 192:5103–5114 Nijvipakul S, Wongratana J, Suadee C, Entsch B, Ballou DP, Chaiyen P (2008)
Mandel MJ, Schaefer AL, Brennan CA, Heath-Heckman EA, Deloney-Marino LuxG is a functioning flavin reductase for bacterial luminescence. J Bacteriol
CR, McFall-Ngai MJ, Ruby EG (2012) Squid-derived chitin oligosaccharides 190:1531–1538
are a chemotactic signal during colonization by Vibrio fischeri. Appl Environ Nishiguchi MK (2000) Temperature affects species distribution in symbiotic
Microbiol 78:4620–4626 populations of Vibrio spp. Appl Environ Microbiol 66:3550–3555
Mandel MJ, Wollenberg MS, Stabb EV, Visick KL, Ruby EG (2009) A single Nishiguchi MK (2002) Host-symbiont recognition in the environmentally trans-
regulatory gene is sufficient to alter bacterial host range. Nature 458:215–218 mitted sepiolid squid-Vibrio mutualism. Microb Ecol 44:10–18
McCann J, Stabb EV, Millikan DS, Ruby EG (2003) Population dynamics of Vibrio Nishiguchi MK, Ruby EG, McFall-Ngai MJ (1998) Competitive dominance
fischeri during infection of Euprymna scolopes. Appl Environ Microbiol among strains of luminous bacteria provides an unusual form of evidence
69:5928–5934 for parallel evolution in sepiolid squid-vibrio symbioses. Appl Environ
McFall-Ngai M, Brennan C, Weis V, Lamarcq L (1998) Mannose adhesin-glycan Microbiol 64:3209–3213
interactions in the Euprymna scolopes-Vibrio fischeri symbiosis. In: Le Gal Y, Nyholm SV, Deplancke B, Gaskins HR, Apicella MA, McFall-Ngai MJ
Halvorson HO (eds) New developments in marine biotechnology. Plenum (2002) Roles of Vibrio fischeri and nonsymbiotic bacteria in the dynamics
Press, New York, pp 273–276 of mucus secretion during symbiont colonization of the Euprymna scolopes
McFall-Ngai M, Heath-Heckman EA, Gillette AA, Peyer SM, Harvie EA light organ. Appl Environ Microbiol 68:5113–5122
(2011) The secret languages of coevolved symbioses: Insights from the Nyholm SV, McFall-Ngai MJ (1998) Sampling the light-organ microenvironment
Euprymna scolopes-Vibrio fischeri symbiosis. Semin Immunol 24:3–8 of Euprymna scolopes: description of a population of host cells in association
McFall-Ngai M, Montgomery MK (1990) The anatomy and morphology of the with the bacterial symbiont Vibrio fischeri. Biol Bull 195:89–97
adult bacterial light organ of Euprymna scolopes Berry (Cephalopoda: Nyholm SV, McFall-Ngai MJ (2003) Dominance of Vibrio fischeri in secreted
Sepiolidae). Biol Bull 179:332–339 mucus outside the light organ of Euprymna scolopes: the first site of symbiont
McFall-Ngai M, Nyholm SV, Castillo MG (2010) The role of the immune system specificity. Appl Environ Microbiol 69:3932–3937
in the initiation and persistence of the Euprymna scolopes–Vibrio fischeri Nyholm SV, McFall-Ngai MJ (2004) The winnowing: establishing the squid-
symbiosis. Semin Immunol 22:48–53 Vibrio symbiosis. Nat Rev Microbiol 2:632–642
McFall-Ngai MJ (1999) Consequences of evolving with bacterial symbionts: Nyholm SV, Nishiguchi MK (2008) The evolutionary ecology of a sepiolid squid-
insights from the squid-vibrio associations. Annu Rev Ecol Syst 30:235–256 Vibrio association: from cell to environment. Vie Milieu Paris 58:175–184
McFall-Ngai MJ, Ruby EG (1991) Symbiont recognition and subsequent morpho- Nyholm SV, Stabb EV, Ruby EG, McFall-Ngai MJ (2000) Establishment of an
genesis as early events in an animal-bacterial mutualism. Science 254:1491–1494 animal-bacterial association: recruiting symbiotic vibrios from the environ-
McFall-Ngai MJ, Ruby EG (1998) Sepiolids and vibrios: when first they meet. ment. Proc Natl Acad Sci USA 97:10231–10235
Reciprocal interactions between host and symbiont lead to the creation of Nyholm SV, Stewart JJ, Ruby EG, McFall-Ngai MJ (2009) Recognition between
a complex light-emitting organ. BioScience 48:257–265 symbiotic Vibrio fischeri and the haemocytes of Euprymna scolopes. Environ
Millikan DS, Ruby EG (2002) Alterations in Vibrio fischeri motility correlate with Microbiol 11:483–493
a delay in symbiosis initiation and are associated with additional symbiotic O’Shea TM, DeLoney-Marino CR, Shibata S, Aizawa S-I, Wolfe AJ, Visick KL
colonization defects. Appl Environ Microbiol 68:2519–2528 (2005) Magnesium promotes flagellation of Vibrio fischeri. J Bacteriol
Millikan DS, Ruby EG (2003) FlrA, a s54-dependent transcriptional activator in 187:2058–2065
Vibrio fischeri, is required for motility and symbiotic light-organ coloniza- O’Shea TM, Klein AH, Geszvain K, Wolfe AJ, Visick KL (2006) Diguanylate
tion. J Bacteriol 185:3547–3557 cyclases control magnesium-dependent motility of Vibrio fischeri.
Millikan DS, Ruby EG (2004) Vibrio fischeri flagellin A is essential for normal J Bacteriol 188:8196–8205
motility and for symbiotic competence during initial squid light organ Ochman H, Moran NA (2001) Genes lost and genes found: evolution of bacterial
colonization. J Bacteriol 186:4315–4325 pathogenesis and symbiosis. Science 292:1096–1099
Miyamoto CM, Lin YH, Meighen EA (2000) Control of bioluminescence in Vibrio Orndorff SA, Colwell RR (1980) Distribution and identification of luminous
fischeri by the LuxO signal response regulator. Mol Microbiol 36:594–607 bacteria from the Sargasso Sea. Appl Environ Microbiol 39:983–987
Miyashiro T, Klein W, Oehlert D, Cao X, Schwartzman J, Ruby EG (2011) The Pelicic V (2008) Type IV pili: e pluribus unum? Mol Microbiol 68:827–837
N-acetyl-D-glucosamine repressor NagC of Vibrio fischeri facilitates coloni- Phillips NJ, Adin DM, Stabb EV, McFall-Ngai MJ, Apicella MA, Gibson BW
zation of Euprymna scolopes. Mol Microbiol 82:894–903 (2011) The lipid A from Vibrio fischeri lipopolysaccharide: a unique structure
Miyashiro T, Wollenberg MS, Cao X, Oehlert D, Ruby EG (2010) A single qrr gene bearing a phosphoglycerol moiety. J Biol Chem 286:21203–21219
is necessary and sufficient for LuxO-mediated regulation in Vibrio fischeri. Pizarro-Cerda J, Cossart P (2006) Bacterial adhesion and entry into host cells. Cell
Mol Microbiol 77:1556–1567 124:715–727
Montes M, Farto R, Perez MJ, Armada SP, Nieto TP (2006) Genotypic diversity of Pollack-Berti A, Wollenberg MS, Ruby EG (2010) Natural transformation of
Vibrio isolates associated with turbot (Scophthalmus maximus) culture. Res Vibrio fischeri requires tfoX and tfoY. Environ Microbiol 12:2302–2311
Microbiol 157:487–495 Pooley DT, Larsson J, Jones G, Rayner-Brandes MH, Lloyd D, Gibson C, Stewart
Montgomery MK, McFall-Ngai M (1993) Embryonic development of the light WR (2004) Continuous culture of photobacterium. Biosens Bioelectron
organ of the sepiolid squid Euprymna scolopes Berry. Biol Bull 184:296–308 19:1457–1463
Montgomery MK, McFall-Ngai M (1994) Bacterial symbionts induce host organ Prouty MG, Correa NE, Klose KE (2001) The novel s54- and s28-dependent
morphogenesis during early postembryonic development of the squid flagellar gene transcription hierarchy of Vibrio cholerae. Mol Microbiol
Euprymna scolopes. Development 120:1719–1729 39:1595–1609
Morris AR, Darnell CL, Visick KL (2011) Inactivation of a novel response Ramesh A, Venugopalan VK (1989) Response of enteric luminous bacteria to
regulator is necessary for biofilm formation and host colonization by Vibrio environmental conditions in the gut of the fish. J Appl Bacteriol 66:529–533
fischeri. Mol Microbiol 82:114–130 Ruby EG, Asato LM (1993) Growth and flagellation of Vibrio fischeri during
Morris AR, Visick KL (2010) Control of biofilm formation and colonization in initiation of the sepiolid squid light organ symbiosis. Arch Microbiol
Vibrio fischeri: a role for partner switching? Environ Microbiol 12:2051–2059 159:160–167
Moynihan M (1983) Notes on the behavior of Euprymna scolopes (Cephalopoda: Ruby EG, Greenberg EP, Hastings JW (1980) Planktonic marine luminous bac-
Sepiolidae). Behaviour 85:25–41 teria: species distribution in the water column. Appl Environ Microbiol
Mulvey MA, Lopez-Boado YS, Wilson CL, Roth R, Parks WC, Heuser J, Hultgren 39:302–306
SJ (1998) Induction and evasion of host defenses by type 1-piliated Ruby EG, Morin JG (1979) Luminous enteric bacteria of marine fishes: a study of
uropathogenic Escherichia coli. Science 282:1494–1497 their distribution, densities, and dispersion. Appl Environ Microbiol 38:406–411
Ruby EG, Nealson KH (1976) Symbiotic association of Photobacterium fischeri Tomarev SI, Zinovieva RD, Weis VM, Chepelinsky AB, Piatigorsky J, McFall-Ngai
with the marine luminous fish Monocentris japonica: a model of symbiosis MJ (1993) Abundant mRNAs in the squid light organ encode proteins with
based on bacterial studies. Biol Bull 151:574–586 a high similarity to mammalian peroxidases. Gene 132:219–226
Ruby EG, Urbanowski M, Campbell J, Dunn A, Faini M, Gunsalus R, Lostroh P, Tong D, Rozas NS, Oakley TH, Mitchell J, Colley NJ, McFall-Ngai MJ (2009) Evi-
Lupp C, McCann J, Millikan D, Schaefer A, Stabb E, Stevens A, Visick K, dence for light perception in a bioluminescent organ. Proc Natl Acad Sci
Whistler C, Greenberg EP (2005) Complete genome sequence of Vibrio USA 106:9836–9841
fischeri: A symbiotic bacterium with pathogenic congeners. Proc Natl Acad Troll JV, Adin DM, Wier AM, Paquette N, Silverman N, Goldman WE, Stadermann
Sci USA 102:3004–3009 FJ, Stabb EV, McFall-Ngai MJ (2009a) Peptidoglycan induces loss of a nuclear
Schaefer AL, Val DL, Hanzelka BL, Cronan JE Jr, Greenberg EP (1996) Generation peptidoglycan recognition protein during host tissue development in
of cell-to-cell signals in quorum sensing: acyl homoserine lactone synthase a beneficial animal-bacterial symbiosis. Cell Microbiol 11:1114–1127
activity of a purified Vibrio fischeri LuxI protein. Proc Natl Acad Sci USA Troll JV, Bent EH, Pacquette N, Wier AM, Goldman WE, Silverman N, McFall-
93:9505–9509 Ngai MJ (2009b) Taming the symbiont for coexistence: a host PGRP neu-
Schleicher TR, Nyholm SV (2011) Characterizing the host and symbiont tralizes a bacterial symbiont toxin. Environ Microbiol 12:2190–2203
proteomes in the association between the Bobtail squid, Euprymna scolopes, Tu S-C, Mager HIX (1995) Biochemistry of bacterial bioluminescence.
and the bacterium Vibrio fischeri. PloS One 6:e25649 Photochem Photobiol 62:615–624
Schuster BM, Perry LA, Cooper VS, Whistler CA (2010) Breaking the language Visick KL (2009) An intricate network of regulators controls biofilm formation
barrier: experimental evolution of non-native Vibrio fischeri in squid tailors and colonization by Vibrio fischeri. Mol Microbiol 74:782–789
luminescence to the host. Symbiosis 51:85–96 Visick KL, Foster J, Doino J, McFall-Ngai M, Ruby EG (2000) Vibrio fischeri lux
Septer AN, Bose JL, Dunn AK, Stabb EV (2010) FNR-mediated regulation of genes play an important role in colonization and development of the host
bioluminescence and anaerobic respiration in the light-organ symbiont light organ. J Bacteriol 182:4578–4586
Vibrio fischeri. FEMS Microbiol Lett 306:72–81 Visick KL, O’Shea TM, Klein A, Geszvain K, Wolfe AJ (2007) The sugar
Septer AN, Wang Y, Ruby EG, Stabb EV, Dunn AK (2011) The haem-uptake gene phosphotransferase system of Vibrio fischeri inhibits both motility and bio-
cluster in Vibrio fischeri is regulated by Fur and contributes to symbiotic luminescence. J Bacteriol 189:2571–2574
colonization. Environ Microbiol 13(11):2855–64 Visick KL, Ruby EG (1998) The periplasmic, group III catalase of Vibrio fischeri is
Shears JS (1988) The use of a sand-coat in relation to feeding and diel activity in required for normal symbiotic competence and is induced both by oxidative
the sepiolid squid Euprymna scolopes. Malacologia 29:121–133 stress and by approach to stationary phase. J Bacteriol 180:2087–2092
Singley CT (1983) Euprymna scolopes. In: Boyle PR (ed) Cephalopod life cycles. Visick KL, Skoufos LM (2001) Two-component sensor required for normal
Academic, London, pp 69–74 symbiotic colonization of Euprymna scolopes by Vibrio fischeri. J Bacteriol
Small A, McFall-Ngai M (1999) Halide peroxidase in tissues that interact 183:835–842
with bacteria in the host squid Euprymna scolopes. J Cell Biochem Walker EL, Bose JL, Stabb EV (2006) Photolyase confers resistance to UV light but
72:445–457 does not contribute to the symbiotic benefit of bioluminescence in Vibrio
Soto W, Gutierrez J, Remmenga MD, Nishiguchi MK (2009) Salinity and tem- fischeri ES114. Appl Environ Microbiol 72:6600–6606
perature effects on physiological responses of Vibrio fischeri from diverse Wang Y, Dufour YS, Carlson HK, Donohue TJ, Marletta MA, Ruby EG (2010a)
ecological niches. Microb Ecol 57:140–150 H-NOX-mediated nitric oxide sensing modulates symbiotic colonization by
Spiro S (2010) An alternative route to nitric oxide resistance. Mol Microbiol 77:6–10 Vibrio fischeri. Proc Natl Acad Sci USA 107:8375–8380
Stabb EV (2005) Shedding light on the bioluminescence ‘‘paradox’’. ASM News Wang Y, Dunn AK, Wilneff J, McFall-Ngai MJ, Spiro S, Ruby EG (2010b) Vibrio
71:223–229 fischeri flavohaemoglobin protects against nitric oxide during initiation of
Stabb EV, Millikan DS (2009) Is the Vibrio fischeri-Euprymna scolopes symbiosis the squid-Vibrio symbiosis. Mol Microbiol 78:903–915
a defensive mutualism? In: White JF Jr, Torres MS (eds) Defensive mutualism Wei SL, Young RE (1989) Development of symbiotic bacterial bioluminescence in
in microbial symbiosis. Taylor and Francis, Boca Raton, pp 85–98 a nearshore cephalopod, Euprymna scolopes. Mar Biol 103:541–546
Stabb EV, Reich KA, Ruby EG (2001) Vibrio fischeri genes hvnA and hvnB encode Weis VM, Small AL, McFall-Ngai MJ (1996) A peroxidase related to the mam-
secreted NAD(+)-glycohydrolases. J Bacteriol 183:309–317 malian antimicrobial protein myeloperoxidase in the Euprymna-Vibrio
Stabb EV, Ruby EG (2003) Contribution of pilA to competitive colonization of mutualism. Proc Natl Acad Sci USA 93:13683–13688
the squid Euprymna scolopes by Vibrio fischeri. Appl Environ Microbiol Welch RA, Burland V, Plunkett G 3rd, Redford P, Roesch P, Rasko D, Buckles EL,
69:820–826 Liou SR, Boutin A, Hackett J, Stroud D, Mayhew GF, Rose DJ, Zhou S,
Stabb EV, Schaefer A, Bose JL, Ruby EG (2008) Quorum signaling and symbiosis Schwartz DC, Perna NT, Mobley HL, Donnenberg MS, Blattner FR
in the marine luminous bacterium Vibrio fischeri. In: Winans SC, Bassler BL (2002) Extensive mosaic structure revealed by the complete genome sequence
(eds) Chemical communication among microbes. ASM Press, Washington, of uropathogenic Escherichia coli. Proc Natl Acad Sci USA 99:17020–17024
DC, pp 233–250 Whistler CA, Koropatnick TA, Pollack A, McFall-Ngai MJ, Ruby EG (2007) The GacA
Stabili L, Gravili C, Tredici SM, Piraino S, Tala A, Boero F, Alifano P (2008) global regulator of Vibrio fischeri is required for normal host tissue responses
Epibiotic Vibrio luminous bacteria isolated from some hydrozoa and that limit subsequent bacterial colonization. Cell Microbiol 9:766–778
bryozoa species. Microb Ecol 56:625–636 Whistler CA, Ruby EG (2003) GacA regulates symbiotic colonization traits of
Studer SV, Mandel MJ, Ruby EG (2008) AinS quorum sensing regulates the Vibrio Vibrio fischeri and facilitates a beneficial association with an animal host.
fischeri acetate switch. J Bacteriol 190:5915–5923 J Bacteriol 185:7202–7212
Sugita H, Ito Y (2006) Identification of intestinal bacteria from Japanese flounder Wier AM, Nyholm SV, Mandel MJ, Massengo-Tiasse RP, Schaefer AL, Koroleva I,
(Paralichthys olivaceus) and their ability to digest chitin. Lett Appl Microbiol Splinter-Bondurant S, Brown B, Manzella L, Snir E, Almabrazi H, Scheetz
43:336–342 TE, Bonaldo Mde F, Casavant TL, Soares MB, Cronan JE, Reed JL, Ruby EG,
Sycuro L, Ruby EG, McFall-Ngai MJ (2006) Confocal microscopy of the light McFall-Ngai MJ (2010) Transcriptional patterns in both host and bacterium
organ crypts in juvenile Euprymna scolopes reveals their mophological com- underlie a daily rhythm of anatomical and metabolic change in a beneficial
plexity and dynamic function in symbiosis. J Morphol 267:555–568 symbiosis. Proc Natl Acad Sci USA 107:2259–2264
Takeuchi K, Kiefer P, Reimmann C, Keel C, Dubuis C, Rolli J, Vorholt JA, Haas Wolfe AJ, Berg HC (1989) Migration of bacteria in semisolid agar. Proc Natl Acad
D (2009) Small RNA-dependent expression of secondary metabolism is Sci USA 86:6973–6977
controlled by Krebs cycle function in Pseudomonas fluorescens. J Biol Chem Wolfe AJ, Millikan DS, Campbell JM, Visick KL (2004) Vibrio fischeri s54 controls
284:34976–34985 motility, biofilm formation, luminescence and colonization. Appl Environ
Tantillo GM, Fontanarosa M, Di Pinto A, Musti M (2004) Updated perspectives Microbiol 70:2520–2524
on emerging vibrios associated with human infections. Lett Appl Microbiol Wolfe AJ, Visick KL (2010) The second messenger cyclic di-GMP. ASM Press,
39:117–126 Washington, DC
Wollenberg MS, Preheim SP, Polz MF, Ruby EG (2011) Polyphyly of non- Yip ES, Geszvain K, DeLoney-Marino CR, Visick KL (2006) The symbiosis
bioluminescent Vibrio fischeri sharing a lux-locus deletion. Environ regulator RscS controls the syp gene locus, biofilm formation and symbiotic
Microbiol 14(3):655–68 aggregation by Vibrio fischeri. Mol Microbiol 62:1586–1600
Wollenberg MS, Ruby EG (2009) Population structure of Vibrio fischeri within the Yip ES, Grublesky BT, Hussa EA, Visick KL (2005) A novel, conserved cluster of
light organs of Euprymna scolopes squid from Two Oahu (Hawaii) genes promotes symbiotic colonization and s54-dependent biofilm forma-
populations. Appl Environ Microbiol 75:193–202 tion by Vibrio fischeri. Mol Microbiol 57:1485–1498
Wollenberg MS, Ruby EG (2012) Phylogeny and fitness of Vibrio fischeri from the Zenno S, Saigo K (1994) Identification of the genes encoding
light organs of Euprymna scolopes in two Oahu, Hawaii populations. Isme J NAD(P)H-flavin oxidoreductases that are similar in sequence to Escherichia
6(2):352–62 coli Fre in four species of luminous bacteria: Photorhabdus luminescens,
Wright KJ, Seed PC, Hultgren SJ (2007) Development of intracellular bacterial Vibrio fischeri, Vibrio harveyi, and Vibrio orientalis. J Bacteriol
communities of uropathogenic Escherichia coli depends on type 1 pili. Cell 176:3544–3551
Microbiol 9:2230–2241 Ziegler MM, Baldwin TO (1981) Biochemistry of bacterial bioluminescence. In:
Yildiz FH, Dolganov NA, Schoolnik GK (2001) VpsR, a member of the response Current topics in bioenergetics. Academic, Chestnut Hill, pp 65–113
regulators of the two-component regulatory systems, is required for expres- Zuber S, Carruthers F, Keel C, Mattart A, Blumer C, Pessi G, Gigot-Bonnefoy C,
sion of vps biosynthesis genes and EPS(ETr)-associated phenotypes in Vibrio Schnider-Keel U, Heeb S, Reimmann C, Haas D (2003) GacS sensor domains
cholerae O1 El Tor. J Bacteriol 183:1716–1726 pertinent to the regulation of exoproduct formation and to the biocontrol
Yildiz FH, Visick KL (2009) Vibrio biofilms: so much the same yet so different. potential of Pseudomonas fluorescens CHA0. Mol Plant Microbe Interact:
Trends Microbiol 17:109–118 MPMI 16:634–644
21 Rumen Symbioses
Itzhak Mizrahi
Institute of Animal Science, ARO, Volcani Research Center, Bet Dagan, Israel
Ruminant Digestive System and the Rumen the reticulorumen. The reticulorumen environment is basically
Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533 anaerobic, with a gas composition of approximately 65% CO2,
27% CH4, 7% N2, and 0.2% H2; the remaining gases include CO,
Rumen Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534 H2S, and transient traces of O2 that are consumed very rapidly
by facultative anaerobic microorganisms. The composition of
Rumen Microbial Composition and Interactions . . . . . . . . . . 534 this gas mixture is the outcome of the intensive microbial fer-
Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534 mentation taking place in this compartment by the
Archaea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535 reticulorumen-resident microorganisms, which are mainly
Protozoa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 536 responsible for breaking down and fermenting the plant fibers
Fungi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 538 (see further on) (Stewart 1986; Dehority 1991; McAllister et al.
1994; Miron et al. 2001). One of the tactics used by ruminants to
Acquisition of Microbial Populations After Birth . . . . . . . . . 538 better support microbial metabolism and fermentation is to
regurgitate the partially digested feed (cud) and chew it again
The Microbiota’s Role in Energy-Utilization Efficiency in (Hoover and Miller 1991). This process, also known as rumina-
Ruminants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539 tion, decreases the size and increases the surface area of the plant
fiber particles interacting with the rumen microbes. The time
Rumen Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 540 spent by the animal ruminating is considered to be controlled by
the particle size of its plant feed: the larger the particle size, the
longer the rumination period (Welch and Smith 1969; Welch
Ruminant Digestive System and the Rumen 1986). The regurgitation process also mixes together the micro-
Environment organisms, digesta and saliva, further increasing fiber break-
down (McAllister et al. 1994; Merchen et al. 1997). The saliva
There are 155 ruminant species, six of which are domesticated: is composed of a bicarbonate and phosphate buffer at a pH of
cattle, sheep, goats, buffaloes, reindeer and yaks. Chief among around 8 which assists in keeping the reticulorumen pH stable
these are the dairy cattle and sheep, and consequently most of (Kay 1966). Salivary production increases when the animal is
our knowledge is based on studies performed on these species. eating, and this in turn improves the buffering capacity of the
Ruminants are herbivores, and a fermentative digestive system rumen environment which has to cope with volatile fatty acids
in their foregut allows them to absorb and digest large amounts (VFAs)—the products of microbial fermentation (Counotte
of plant material. This method of digesting feed is considered to et al. 1979). As a result of this buffering system, the rumen pH
be a cooperative interaction between the host and its resident range is normally between 5.5 and 7.0, reaching a minimum of
microorganisms: the ruminant digestive system relies on its 6 h after feeding in animals that are fed once a day (Dehority and
resident microorganisms to digest the feed and subsequently Tirabasso 2001; Welkie et al. 2009). The rumen temperature and
absorbs their fermentation products. Therefore, the partners oxidation-reduction potential are also maintained in a constant
are in a cooperation in which the animal provides the food range of 38–40 C and 0.15–0.4 V, respectively, thus providing
and a suitable environment and the microorganisms digest the a stable environment for the resident microbiota (Church 1969).
food, making it accessible to the animal (Mackie 2002). Hence, After the plant fibers in the feed have been degraded, the digesta
to fulfill its part of the partnership, the ruminant’s digestive passes into the next stomach chamber, the omasum, which acts
system has to provide optimal accommodations for its resident as a filtering device through which only particles less than 2 mm
microorganisms and still be able to extract nutritional products in size can readily pass (Weimer et al. 2009). In the omasum,
from the food. water and many of the inorganic mineral elements are absorbed
The architecture and physiology of this system have evolved into the bloodstream. Then the digesta moves to the true stom-
over millions of years, dating back to the Jurassic era, to allow ach, the abomasum. The abomasum is the direct equivalent of
efficient digestion of plant materials (Mackie 2002). This sys- the human monogastric stomach but contains a unique enzy-
tem’s effectiveness is conferred by its design, which prolongs and matic feature that is adapted to the foregut fermentation system:
maximizes plant biomass exposure to specialized microorgan- it secretes lysozyme—an enzyme that degrades bacterial cell
isms that degrade the plant fibers, while providing stable and walls via hydrolysis of the b-1,4 glycosidic bonds of the pepti-
favorable conditions for their growth. The consumed feed is doglycan components. Lysozyme is usually found in tears,
packed into the first two stomach chambers, collectively termed mucus, and saliva, protecting the tissues from bacterial attack.
534 21 Rumen Symbioses
In the abomasum, it acts as a digestive enzyme, lysing the metagenomic study in which approximately 95% of the coding
bacteria coming from the rumen and enabling proper digestion sequences were bacteria, 0.6–4% archaea, and 1.5% eukarya
of the bacterial protein synthesized by the rumen bacteria to (Brulc et al. 2009). Each of these domains is represented in the
build their cells (Jolles et al. 1984, 1989). In the abomasum, the rumen microbial ecosystem by a vast array of taxa. The general
bacterial protein and digesta are digested in much the same way members of this microbial consortium are the subject of this
as in nonruminant mammals (Merchen et al. 1997). section. Due to space limitations, the complex taxon diversity
within each domain is not covered in this chapter. Here, we review
the common traits and features of each domain, their interactions
Rumen Metabolism and their effects on the rumen ecosystem and ruminant host.
The reticulorumen functions as a pre-gastric anaerobic fermen-

tation chamber. It is inhabited by a high density of resident Bacteria
microbiota, consisting of bacteria, protozoa, archaea and
fungi, which degrade the consumed plant materials (Flint Bacteria make up the largest domain in the rumen ecosystem
1997). In general, plant materials are composed of organic mat- and their presence is vital for the animal’s well-being. They are
ter such as proteins, lipids, and nucleic acids and a large propor- responsible for most of the degradation and fermentation of
tion of carbohydrate polymers such as celluloses, hemicelluloses, plant mass in the rumen (Windham and Akin 1984; Akin and
pectins, fructosans, starches and other polysaccharides which are Benner 1988). A concentration of approximately 1,011 bacteria/
mostly indigestible by the animal (Flint et al. 2008). The ml exists in the rumen, and their populations are considered
reticulorumen microorganisms utilize specialized enzymes and highly complex in terms of taxon identity and functionality. The
enzyme complexes to degrade these polysaccharides into mono- coordinated and complex metabolic interactions between the
meric or dimeric sugars and subsequently ferment them (Flint different bacterial taxa enable efficient utilization of the con-
et al. 2008). The degradation and fermentation take place in sumed plant fiber and maximal energy consumption by the
a coordinated and complex manner, in which the substrate for overall rumen microbiota. The composition of these bacterial
one microorganism is the product of another (Dehority 1991; populations has been the subject of intensive research using
Mackie 2002). This enables the degradation of all plant materials classical microbiology, culture-free methods, and next-
except lignin, which is a phenolic compound present in the cell generation sequencing techniques which have helped character-
wall. The biodegradation and metabolism of plant feed by the ize their complexity by allowing a broad overview of the bacterial
reticulorumen microorganisms make these otherwise indigest- 16S rRNA gene as well as their coding capacity. As revealed by
ible polymers accessible for absorption and utilization. these studies, the dominant phyla in the rumen are similar to
These indigestible polymers are converted by the those in other enteric bacterial ecosystems, consisting of
reticulorumen microorganisms into two main digestible forms. Bacteroidetes, Firmicutes, and Proteobacteria (Brulc et al.
The first is the microbial proteins synthesized by these microor- 2009; Jami and Mizrahi 2012). In a study by Stevenson and
ganisms to build their cells—these proteins are digested in the Weimer (2007), quantitative real-time PCR was used to charac-
abomasum (true stomach) as the microbes travel along the terize the bacterial species in two lactating cows. The dominant
animal’s digestive tract (Kay 1969). This digested microbial bacterial genus was found to be Prevotella, comprising 42–60%
protein makes up 50–80% of the total protein absorbed by the of the bacterial rRNA genes.
animal (Storm and Ørskov 1983). The second is VFAs, which are The rumen’s bacterial populations have evolved coordinated
fermentation end products, along with methane and carbon metabolic functions, in which some taxa depend on others for
dioxide. VFAs are short-chain fatty acids that are absorbed by their growth, complicating their cultivation and study. Thus,
the animal in the reticulorumen and serve in important bio- because most of the rumen bacteria have not been cultured,
chemical processes, such as gluconeogenesis in the case of pro- our knowledge of their metabolism is limited. Due to the high
pionate and fatty acid synthesis in the case of acetate and complexity of the bacterial composition and the unique func-
butyrate (Russell and Wilson 1996). Hence, the VFAs serve as tional distribution of the rumen bacterial species, only the dom-
an energy source for the animal. In a recent study, Weimer et al. inant bacteria and their metabolic functions are covered in this
(2009) estimated that VFAs retain approximately 70% of the section. Basically, the plant fibers are degraded and fermented by
energy content of the plant material’s polymers. a consortium of bacterial taxa that interact metabolically such
that the product of one taxon is utilized by another, thereby
enabling sequential degradation of the plant biomass. This
Rumen Microbial Composition and allows functional specialization of different bacteria for different
Interactions plant mass compounds and degradation stages.
It appears that the plant cell wall, which is composed of
As we have seen, the energy and protein used by the ruminant cellulose fibers embedded in a hemicellulose matrix, is initially
are a direct outcome of its resident ruminal microbes’ activity. broken down by only a small group of cellulolytic bacteria which
All domains of life are represented in these complex communi- are capable of degrading the highly ordered and insoluble forms
ties, and their distribution was demonstrated in a recent of cellulose. These bacteria include Ruminococcus flavefaciens,
Rumen Symbioses 21 535
Ruminococcus albus, and Fibrobacter succinogenes, which are Nevertheless, some bacterial taxa can also utilize protein, includ-
considered to be the dominant cellulolytic species, and to ing R. amylophilus, S. ruminantium, Prevotella ruminicola, and
a lesser extent Butyrivibrio fibrisolvens (Flint and Bayer 2008; B. fibrisolvens. The highly abundant P. ruminicola, which is also
Flint et al. 2008). Although these species are the best character- a hemicellulolytic species, is considered to be the most active
ized ones, other uncultured bacteria might also take part in this producer of ammonia (NH3) in the rumen due to its high
process, such as the recently discovered and described bacterium abundance and proteolytic and deaminase activities. This func-
Cellulosilyticum ruminicola H1 isolated from yak (Bos grunniens) tion is highly important as the rumen environment is not rich in
rumen, which degrades lignocellulose with multiple carbohy- protein and ammonia serves as an important nitrogen source for
drate-borne fibrolytic enzymes (Cai et al. 2010). Some of these the anabolism of amino acids and proteins (Stevenson and
species have been shown to negatively interact with each other, Weimer 2007).
with cellulose digestion decreasing when they are grown Therefore, the ruminal bacterial populations are fine-tuned
together in culture (Saluzzi et al. 1993; Odenyo et al. 1994b). to their host ruminant’s diet and change accordingly, thereby
This inhibitory effect is thought to be the outcome of the enabling fermentation of the different dietary ingredients. They
bacteriocins produced by some of these bacteria, such as also interact metabolically with each other, sometimes compet-
R. albus and R. flavefaciens, to better compete for the mutual ing for recourses and inhabiting each other and sometimes
resource—cellulose (Odenyo et al. 1994a; Kalmokoff and cooperating to enable maximum energy utilization of the feed,
Teather 1997; Rychlik and Russell 2002; Chen et al. 2004). providing the host with bacterial proteins and VFAs.
Bacteriocins produced by other non-cellulolytic ruminal bacte-
ria have been found and are thought to serve for competition in
the various niches within the rumen ecosystem (Laukova 1993; Archaea
Morovsky et al. 2001; Whitford et al. 2001; Cookson et al. 2004;
Xavier and Russell 2009). It is important to note that evidence of A large part of the Archaea population in the rumen is made up
quorum sensing, for both intraspecies and interspecies commu- of anaerobic methanogens, as revealed by studies characterizing
nication, has been documented for rumen bacteria but its exact the microbial small subunit rRNA gene in the rumen, as well as
role in the rumen ecosystem was not determined (Erickson et al. by the aforementioned recent metagenomic study which esti-
2002; Mitsumori et al. 2003). mated the archaea population at 0.6–3.3% of total rumen
After the initial breakdown and solubilization of the plant microbes (Janssen and Kirs 2008; Brulc et al. 2009). In a study
fibers, the polymers are utilized by a vast array of taxa specific to characterizing methanogenic properties, Yanagita et al. (2000)
the various downstream degradation stages of these polymers. calculated that 3.6% of ruminal microorganisms display
This also holds true for other less recalcitrant polymers such as autofluorescence characteristics of F420, an enzymatic cofactor
starch, which is utilized by Ruminobacter amylophilus, Strepto- found in all methanogens (Gorris and van der Drift 1994).
coccus bovis, Succinomonas amylolytica, several Prevotella spe- Methanogens produce CH4 by either reducing CO2 or dissimi-
cies, Butyrivibrio fibrisolvens and Selenomonas ruminantium, and lating acetate to CH4 and CO2 (Thauer et al. 2008). In the rumen
pectin, which is utilized by Succinivibrio dextrinosolvens and ecosystem, the latter reaction is negligible (Hungate et al. 1970),
Lachnospira multiparus (Krause et al. 2003). Bacterial species and therefore most of the CH4 is produced via reduction of CO2.
composition changes according to the rations and dietary ingre- The electrons for the reduction of CO2 can come from several
dients consumed by the animal. This is best exemplified by the chemical compounds, such as H2, formate, methanol, and
presence of starch rather than plant fiber in high-grain diets methylamine. In the rumen, most of the methane production
compared to the high-forage diets widely used as part of differ- results from utilization of hydrogen for carbon dioxide reduc-
ent husbandry protocols, which causes dramatic changes in the tion, although formate, methanol and methylamines are also
bacterial taxa. These changes are apparent at all taxonomic used by some methanogenic species (Hook et al. 2011).
levels, from phylum to species (Callaway et al. 2010; Fernando Phylogenetically, methanogenic archaea belong to the phy-
et al. 2010). The energy from the plant mass is maximally lum Euryarchaeota and are divided into five taxonomic orders—
extracted as even the bacterial fermentation end product lactate Methanopyrales, Methanococcales, Methanobacteriales,
is utilized by other bacteria, such as Megasphaera elsdenii (Elsden Methanomicrobiales, and Methanosarcinales (Balch et al.
et al. 1956) and Veillonella alcalescens (Johns 1951), as are other 1979). Janssen and Kirs (2008) analyzed large data sets from
end products such as hydrogen, acetate, and succinate. several culture-free studies of different ruminants, revealing that
A recent metagenomic study of mammalian gut microbiota most rumen methanogens belong to three genus-level groups:
has demonstrated that in herbivore and ruminant microbiota, Methanobrevibacter (61.6%) order Methanobacteriales,
metabolic pathways that involve the anabolism of amino acids Methanomicrobium (14.9%) order Methanomicrobiales, and
are more common than in carnivores. The latter’s diets are much a large group of uncultured rumen archaea labeled as rumen
higher in protein and therefore their microbiota metagenomes cluster C, or RCC (15.8%) (Janssen and Kirs 2008). In a later
are rich in protein catabolism pathways (Muegge et al. 2011). study performed in New Zealand using denaturing gradient gel
Indeed, the food entering the rumen is composed mainly of electrophoresis (DGGE) of archaeal 16S rRNA genes, the com-
carbohydrates, and therefore most of the bacteria occupying position of archaeal communities in the rumens of farmed
the rumen rely on carbohydrate as their main energy source. sheep, cattle, and red deer was investigated. Total archaeal
communities were relatively constant across species and diets coculture with methanogens (Bauchop and Mountfort 1981).
and were less variable and less diverse than the bacterial com- Interspecies hydrogen transfer between methanogens and other
munities. There were diet- and ruminant-species-based differ- rumen microbes is best demonstrated by the physical association
ences in archaeal community structure, but the same dominant of methanogen species with rumen protozoa. The basis of this
archaea were present in all rumens. In that study, species from association is the large amount of hydrogen produced by the
the genus Methanobrevibacter were also found dominant in all latter eukaryotic microorganisms (Lloyd et al. 1989; Muller
rumens. Members of the RCC archaeal group of unknown 1993; Hackstein and Vogels 1997). Although protozoa prey on
physiology were also present, accounting for an average 26.5% bacteria and Archaea, methanogens have been reported to
of the total archaea (Jeyanathan et al. 2011). adhere to or reside within them, establishing a symbiotic rela-
Some methanogenic archaea use cytochromes for the pro- tionship in which the protozoa degrade and ferment the plant
duction of methane, whereas others use alternative complexes material and the methanogens thermodynamically promote the
for this reaction. The order Methanosarcinales contains protozoa’s energy exploitation by utilizing the hydrogen pro-
methanogens with cytochromes and can grow on the broadest duced in these processes (Vogels et al. 1980; Krumholz et al.
range of substrates. Methanogens with cytochromes have 1983; Ushida and Jouany 1996; Tokura et al. 1999; Irbis and
a growth yield of 7 g/mol methane on hydrogen and carbon Ushida 2004; Tothova et al. 2008). Therefore, the utilization of
dioxide and have a doubling time of greater than 10 h, while hydrogen by methanogens is the basis of syntrophic interactions
methanogens without cytochromes have a growth yield of only with the rumen microbial communities to maximize overall
3 g/mol methane on hydrogen and carbon dioxide and a dou- energy recovery. This notion has led to the postulation that
bling time of 1 h (Thauer et al. 2008). Methanogens without these interactions increase overall plant fiber degradation in
cytochromes grow on a much lower concentration of hydrogen the rumen (Wolin 1979; Holter and Young 1992; McAllister
compared to methanogens with cytochromes, the latter needing and Newbold 2008; Janssen 2010). Feed intake and hemicellu-
a tenfold higher hydrogen concentration for growth (Thauer lose digestibility are also positively correlated with methane
et al. 2008). This might explain the higher abundance of the production (Holter and Young 1992; Ellis et al. 2007). Never-
former in the rumen, as they can better compete for the hydro- theless, the production of methane in the rumen is correlated
gen produced, lowering its concentration and preventing the with energy loss which can reach up to 19% of the total energy
methanogens with cytochromes from utilizing it. The consump- stored in the feed (Czerkawski 1969; Johnson and Johnson
tion of hydrogen by methanogens is highly important in the 1995). Hence, by symbiotically interacting with other rumen
rumen as it lowers the partial pressure of rumen hydrogen, microbes, methanogens enable maximal energy harvest from
thereby enabling some endergonic metabolic reactions to the feed, but by producing methane, they decrease the ruminant
become exergonic, consequently increasing energy conservation host’s ability to extract energy from its feed.
by bacterial fermentation pathways that become energetically
feasible (Stams and Plugge 2009). This is the basis for the
syntrophic relationship between methanogens and some of the Protozoa
anaerobic rumen microbes, such as those that produce hydrogen
as their fermentation end product. Such is the case for Protozoa are a diverse group of single-celled eukaryotic organ-
R. flavefaciens and R. albus, two dominant cellulolytic rumen isms, bound by a cuticle or pellicle. This group is often consid-
bacteria that have a facultative syntrophic relationship with ered the simplest form of animal life (Dehority 2003).
methanogens: in the presence of methanogens, they alter their Rumen protozoa were first observed in 1843 by Gruby and
fermentation pathways to conserve the maximum amount of Delafond. They are unable to grow under laboratory conditions
energy and end products (Latham and Wolin 1977; Pavlostathis in the absence of bacteria (Fondevila and Dehority 2001a, b).
et al. 1988; Stams and Plugge 2009). An effect of methanogens Most of the protozoa in the rumen are ciliate species belonging
on the fermentation of other non-cellulolytic rumen bacteria has to the phylum Ciliophora, although several flagellated species
also been reported (Chen and Wolin 1977). Ruminal fungus belonging to the phylum Sarcomastigophora have been
fermentation and cellulolytic activities have also been reported described in ruminant animals as well, albeit in much lower
to change in the presence of methanogens, where fermentation abundance than ciliates (Ogimoto and Imai 1981). The rumen
pathways have changed and cellulolytic activities have increased ciliates are subdivided into the orders Entodiniomorphida and
when the ruminal fungus isolates were cocultured with rumen Vestibuliferida, which contain 25 genera (Dehority 2003).
methanogens (Joblin et al. 1990; Marvin-Sikkema et al. 1990; The rumen protozoa are highly specialized and adapted to
Teunissen et al. 1992). This was also demonstrated by Bauchop the rumen environment. They are mostly anaerobic, although
and Mountfort (1981) who studied the degradation of cellulose some species are known to scavenge oxygen. This latter trait is
by rumen fungi in the presence and absence of methanogens. In considered beneficial for ruminal communities as it protects the
that study, it was observed that after 100-h incubation, only 10% strictly anaerobic microorganisms, such as the cellulolytic bac-
of the total cellulose was degraded by fungi alone, compared to terial species, and consequently increases fiber degradation (Ellis
70% when cocultured with methanogens; at the end of the et al. 1989). The rumen protozoa can utilize a vast array of
incubation period (200 h), the overall degradation of cellulose carbohydrate compounds, such as soluble sugar, starch and
was 53% for fungi alone compared with 82% for those in lignocellulose. Much of these abilities, particularly the
lignocellulolytic one, are thought to have been acquired by fermentation parameters and microbial composition were
lateral gene transfer from their close association with the bacte- reported. These aspects have been studied extensively for years,
ria and archaea on which they prey (Ricard et al. 2006; Flint and as well as more recently in studies investigating the effect of
Bayer 2008). Although the ruminal protozoa’s nutritional presence or absence of protozoa on rumen fermentation, micro-
requirement for bacteria is well established, the reasons for this bial composition, efficiency of microbial protein synthesis, and
requirement and the degree of specificity for the bacterial species even on the host tissue. Due to their agricultural importance,
on which they prey are not well understood. There have been these studies have been performed mostly with cows and sheep.
several reports on protozoan species’ selectivity for the bacterial The effect of the presence of protozoa on the composition of
species that they ingest and that grow on them (Gutierrez 1958; rumen bacteria was investigated in cattle by Ozutsumi et al.
Gutierrez and Davis 1959; Mah 1964); however, there have also (2005). In a comparison of 16S rRNA gene (rDNA) clonal
been contradictory reports showing less selective uptake and libraries from faunated and defaunated animals, no extreme
digestion of a variety of bacterial species, including nonruminal differences were found in bacterial phyla; however, a computer
ones, by rumen protozoa, including protozoan species from the analysis revealed that the presence of ruminal protozoa mark-
same genera as in the former reports (Coleman 1962, 1964, edly affects the composition of rumen bacteria (Ozutsumi et al.
1967a, b). These discrepancies may result from the different 2005). Belanche et al. (2011) examined the effect of the absence
species and strains used in the studies. The reasons for ruminal of rumen protozoa on rumen fermentation and efficiency of
protozoa’s nutritional need for bacteria are also not entirely microbial protein synthesis in lambs on different diets and
clear, except for requiring bacteria strictly as a source of nutri- reported that the protozoa modified rumen fermentation pat-
ents. Some reports show a need for live bacteria for protozoon tern and reduced digestibility of organic matter and total cell
growth and development, which might imply an active depen- wall (Neutral Detergent Fiber) by 2.0 and 5.1 percentage points
dence on the bacteria’s metabolism (Coleman 1967b; Fondevila respectively, while having only a mild effect on nitrogen flow
and Dehority 2001b). To settle these issues, better experimental (Belanche et al. 2011). Sultana et al. (2011) investigated the effect
systems need to be developed: currently, antibiotic compounds of protozoa on fatty acid composition profile in the rumen of
and boiling are used to kill the bacteria fed to protozoa, and cattle, indicating that protozoa contribute greatly to trans-
these treatments can have an effect on the protozoa in the case of vaccenic acid and conjugated linoleic acid production in the
antibiotics and on the nutritional gain from the bacteria in the rumen (Sultana et al. 2011). Mosoni et al. (2011) examined the
case of boiling. effect of long-term defaunation on the structure of the
Some rumen ciliates are considered to be associated with microbiota, particularly methanogenic archaea and fibrolytic
rumen methanogenic archaea: much of this interaction is bacteria. Total rumen bacterial density showed an estimated
assumed to be based on the generation of hydrogen gas by increased response to long- and short-term defaunation but
protozoan fermentation which, as mentioned above, the without noticeable shifts in diversity. Defaunation increased
methanogens use as an electron donor to reduce carbon dioxide the content of R. albus and R. flavefaciens but did not affect
to methane, an association that results in increasing methane that of F. succinogenes. Despite a 20% reduction in methane
emission to an estimated 9–25% of total rumen methane emis- emission, the methanogen content increased in the absence of
sion (Vogels et al. 1980; Newbold et al. 1995). The nature of this protozoa, while the diversity of the dominant methanogenic
association has been the subject of intensive study aimed at community was not modified (Mosoni et al. 2011). Hegarty
determining whether the methanogenic archaea are ecto- or et al. (2008) studied the effects of the absence of protozoa from
endosymbionts and whether the associations involve specific birth or weaning on growth and methane production in lambs.
species of ciliates and methanogenic archaea (Vogels et al. Single lambs born to defaunated ewes were heavier at birth and
1980; Ushida and Jouany 1996; Tokura et al. 1999; Irbis and at weaning than lambs born to faunated ewes. Wool growth rate
Ushida 2004; Tothova et al. 2008). Aside from the increase in of defaunated lambs was 10% higher than that of faunated lambs
methane emission, specific ruminal ciliate species have been and was increased a further 9% by a high-protein diet. There was
shown to alter the balance between the different VFAs and to no main effect of protozoon treatment on enteric methane
increase ammonia concentrations in vitro (Ranilla et al. 2007). production. These data indicated that while lambs without
The characteristics of these organisms as predators of other rumen protozoa have greater protein availability than faunated
microbes, hydrogen producers, and carbohydrate degraders ruminants, there is no major effect of rumen protozoa on enteric
and fermenters raise intriguing questions regarding their role methane production by lambs fed either a concentrate or rough-
in overall rumen metabolism and their effects on the ruminant age diet (Hegarty et al. 2008). Yanez-Ruiz et al. (2007) studied
host. These effects were examined using experimental tech- the effects of the absence of protozoa in the lamb rumen on
niques that enable working with animals that are free of pro- animal growth, rumen fermentation, microbial diversity, and
tozoa, termed defaunated animals. These studies showed that fatty acid profiles in abomasal fluid and intramuscular fat.
the rumen ciliates are not essential for viability and growth of the Bacterial diversity was higher in control lambs than in their
host animals (Williams and Coleman 1992); this is why rumen protozoon-free counterparts. Abomasal content was different
protozoa are often considered commensal eukaryotes in the with respect to fatty acid composition in the defaunated lambs.
herbivore rumen. Nevertheless, in studies comparing defaunated Differences were also seen in the fatty acid composition of the
animals to their faunated counterparts, several effects on rumen intramuscular fat (Yanez-Ruiz et al. 2007). These studies show
that the presence of protozoa affects rumen fermentation increasing lignocellulose accessibility to other rumen microbes
parameters, microbial composition, and host physiology. There- and contributing to overall fiber utilization (Ho et al. 1988; Akin
fore, the definition of commensalism—often used to refer to et al. 1989).
rumen protozoa—in which one organism benefits but the other
is not affected, does not appear to apply.
Acquisition of Microbial Populations After
Birth
Fungi
Although microbial colonization of the rumen after birth is
The rumen fungi are all anaerobic, belonging to the family important for understanding this ecological niche, little is
Neocallimasticaceae from the class Chytridiomycetes (Dehority known about it. In the first weeks of life, the rumen is not
2003). Chytridiomycetes fungi have been recorded in various functional: the suckled milk does not pass through the rumen
kinds of ruminants and herbivores, including sheep, cattle, because of the esophageal groove, which is closed by reflex
camel, vicuna, deer, impala, goat, reindeer, muskox, gaur, ante- action, although some milk might nevertheless enter the
lope, and greater kudu (Dehority 2003), but they are only rumen and provide nutrients for microbial populations. The
considered to occur in large numbers (up to 8% of total ruminal occurrence of microbial communities in the rumen of young
content) when the animals feed on low-quality forage, which animals was researched in the early days of rumen microbiology,
increases retention time in the rumen; their numbers decrease providing the observation of rapid colonization of the rumen by
when the animals eat high-grain diets, which reduce ruminal microbial taxa (Bryant et al. 1958; Bryant and Small 1960).
retention time (Russell and Rychlik 2001). These differences are Fonty et al. (1987) also observed that the rumen ecosystem is
thought to be due to the long life cycle of these fungi (8–32 h), quickly established after birth, before the rumen itself becomes
which consists of a motile zoospore stage and a vegetative thallus functional. In that study, the authors investigated microbial
stage (Orpin 1975) and results in their being washed out from acquisition and colonization steps in the lamb rumen. The
the host’s digestive tract. authors used five test groups which were kept with or without
In general, the metabolic activities of rumen fungi include their mothers and with different combinations of accessibility to
the degradation and utilization of cellulose, hemicellulose, and plant fiber and milk. Using classical microbiology methods, the
starch, as well as some proteolytic activity (Dehority 2003). The authors divided the rumen microbiota into strictly anaerobic,
metabolic interaction of ruminal fungi with other rumen aerobic, facultatively anaerobic, cellulolytic bacteria,
microbes was studied in vitro, and they were found to be methanogenic archaea and anaerobic fungi. They found that
enhanced or reduced as a function of the microbial partner microbial colonization of the lamb’s rumen is characterized by
with which they were cocultured. The ability of some rumen the dominance of strictly anaerobic species very soon after birth
fungi to degrade cellulose increased in the presence of (2 days) and by the early appearance (1 week after birth) of
methanogens and some bacterial species, such as Selenomonas populations of cellulolytic and methanogenic bacteria and
ruminantium, Veillonella alcalescens and Megasphaera elsdenii anaerobic fungi; the aerobic and facultatively anaerobic micro-
(Marvin-Sikkema et al. 1990; Dehority 1991). Cocultivation of flora in the lambs declined rapidly as the animal matured.
the rumen fungus Neocallimastix frontalis with Prevotella Surprisingly, the authors found that rather than first appearing
ruminicola, Succinivibrio dextrinosolvens, or Selenomonas at weaning, the cellulolytic bacteria are present in large numbers
ruminantium showed a synergistic interaction with regard to before the ingestion of solid feed, as well as in lambs fed exclu-
xylan utilization; xylan utilization decreased in cocultures sively milk. In an earlier work, Bryant et al. (1958) also observed
containing Lachnospira multiparus or Streptococcus bovis cellulolytic bacteria in calves at the end of their first week of life.
(Williams et al. 1991). A negative interaction with respect to Bryant et al. (1958) also suggested that contact between the
cellulose hydrolysis was also observed in coculture with the newborn lamb and its mother or other adult ruminants during
cellulolytic bacterial species R. flavefaciens and R. albus, which the first days of life is necessary for the establishment of these
was thought to involve a polypeptide released into the superna- cellulolytic species, since in lambs which were kept in isolation,
tant that inhibits fungal cellulose hydrolysis without having any the cellulolytic flora did not become established. The appearance
cellulolytic activity of its own (Stewart et al. 1992; Bernalier et al. of anaerobic fungi toward the end of the first week of life was also
1993). Although the ruminal fungi are capable of plant mass reported, possibly because the retention time in the still
degradation and utilization, their effect on overall ruminal nonfunctional rumen is long and allows the development of
fermentation is considered minor, possibly due to bacterial fungi before the ingestion of solid food.
inhibition of their growth and activity as concluded from studies Minato et al. (1992) used classical microbiology methods to
examining their in vitro effect on cellulose digestion and fer- investigate the colonization of microbial populations in the
mentation using antibiotics that inhibit the bacteria and fungi rumen of calves under normal husbandry growth conditions.
separately (Windham and Akin 1984; Akin and Benner 1988; They also observed rapid colonization of the rumen immedi-
Dehority and Tirabasso 2000). Nevertheless, their presence on ately after birth. The first bacteria to develop in abundance were
the plant mass is thought to disrupt the lignocellulose tissues Escherichia coli and Streptococci. The number of E. coli, which
because their rhizoids penetrate the plant cell wall, thus was high in 1-day-old calves, decreased gradually to a constant
level at between 6 and 8 weeks of age, which also concurs with and Jenkins 1985; Thiessen et al. 1985; Taylor et al. 1986; Solis
studies by Fonty et al. (1987) and Bryant et al. (1958) who et al. 1988; Aharoni et al. 2006). The nature of this variation is
reported a decrease in facultative anaerobic bacteria with age. not entirely clear, but various factors are thought to affect the
The number of streptococci, which was high for the first 8 weeks animal’s RFI and hence its energy utilization (Johnson et al.
of life, showed a decrease at 10 weeks of age. The number of 2003). The presence of a moderate genetic component affecting
lactobacilli, which was high in 1-day-old calves, increased until energy utilization was demonstrated by elevating feed efficiency
2 weeks of age and remained constant thereafter. Amylolytic via successful selection of animals according to their RFI in
bacteria, sulfate reducers, lactate utilizers, xylan fermenters and combination with specific genomic markers (Hotovy et al.
pectin fermenters, which were scarce in 1-day-old calves, 1991; Archer et al. 1999; Herd et al. 2003; Richardson et al.
increased within 3 days after birth and then remained constant. 2004; Crews 2005). Differences between high- and low-RFI
The cellulolytic bacteria appeared in 3–5-day-old animals, and animals have also been reported in terms of metabolic activity,
became abundant in 2–3-week-olds. The methanogenic digestibility, and methane production (Richardson et al. 1996;
populations appeared in 1–2-week-old calves and became abun- Arthur et al. 2001; Basarab et al. 2003; Nkrumah et al. 2004,
dant when the animals were approximately 3 weeks old. 2006). These differences can be related to various factors, one of
Protozoa have also been reported to rapidly colonize the which might be the reticulorumen microbiota, upon which the
rumen in the first days after birth (Bryant et al. 1958; Bryant animal’s digestion and absorption of feed are largely dependent.
and Small 1960; Eadie and Hobson 1962). Because these micro- Hence, the reticulorumen microbiota may play an important
organisms are not essential to the animal’s existence, their mode role in energy harvesting from the feed, affecting the animal’s
of transmission could be studied in more detail. Apparently, they energy utilization.
are transmitted from animal to animal, as when calves were Some of the energy lost during the conversion can be attrib-
separated from other ruminants they did not become faunated uted to methane production by methanogenic archaea which, as
(Bryant et al. 1958; Bryant and Small 1960; Eadie and Hobson already mentioned, is one of the end products of the
1962). Transmission can occur from a mother ruminant reticulorumen’s overall metabolism. The methane gas is
grooming its young and passing protozoa via the saliva or via eructated into the atmosphere along with its retained energy,
salivation in the feed or pasture (Dehority 2003). It is speculated which is lost from the cow’s reticulorumen. This process results
that rumen anaerobic fungi are transmitted via the same mech- in a loss of 5–19% of the energy content of the feed (Johnson and
anism, as they have been reported in both the feces and saliva of Johnson 1995) and has wide environmental implications: meth-
ruminants (Trinci et al. 1988). ane is a very potent greenhouse gas (23 times more potent than
carbon dioxide) and in some countries, ruminants are respon-
sible for up to 60% of its emission (Wuebbles and Hayhoe 2002).
The Microbiota’s Role in Energy-Utilization A recent study in which animals with different RFI values were
Efficiency in Ruminants compared reported that animals with low energetic efficiency
(high RFI) exhibit significantly more methane emission (Nkru-
Cattle’s ability to utilize and divert the energy stored in its feed to mah et al. 2006). The authors speculated that the differences in
milk and meat production is governed by its energetic efficiency. methane emission are the outcome of differences in the
Energetic efficiency is defined as the ratio between the energetic methanogenic populations.
value of the animal and its products and the energetic value of In the last few years, the notion of resident microbiota
the feed or diet; it thus includes the energy invested in mainte- affecting a host’s ability to harvest energy from its food has
nance, as well as the energy invested in milk and meat produc- been the topic of intense research with respect to the human
tion (Johnson et al. 2003). In general, the energy is preferentially gut (Ley et al. 2006; Turnbaugh et al. 2006; Turnbaugh and
utilized for maintenance over production (Tolkamp 2010). The Gordon 2009). Those studies established a strong correlation
energy from the consumed feeds that is not retained in the body between bacterial gut divisions and the ability to harvest energy
or milk is lost in the form of feces, urine, heat, and combustion from food, and consequently increase body weight and fat. The
of gases, such as methane. Different methods are used to evalu- relationship between rumen microbiota and host feed efficiency
ate an animal’s energetic efficiency, among them the residual has been examined in a few recent studies in beef cattle (Guan
feed intake (RFI) method (also known as net feed efficiency et al. 2008; Zhou et al. 2009). In one study, the authors used
method) (Koch et al. 1963). This method evaluates energetic PCR-DGGE to compare bacterial reticulorumen population
efficiency according to the difference between the animal’s actual composition in 18 steers from different breeds with different
feed intake and its estimated feed intake over a specified period RFI values. The authors reported the clustering of specific bac-
of time (Koch et al. 1963; Archer et al. 1999). Animals that have terial populations according to the animal’s energy efficiency,
low RFI values are considered to be more energetically efficient accompanied by a correlation with VFA concentration. How-
than those with high values. This method is independent of ever, the identity, activity and coding capacity of those
growth and body size, making it suitable for comparisons populations—a significant portion of the factors determining
between animals (Archer et al. 1999; Moore et al. 2009). the microbial-feed efficiency relationship—were not studied
Energetic efficiency varies considerably among breeds, as (Guan et al. 2008). In a second study, the methanogenic
well as among different individuals from the same breed (Ferrell populations of animals with different RFIs were compared
using 16S rDNA clone libraries. Efficient animals showed less interactions have been documented and researched in in vitro
species diversity, and there was a correlation between the com- cocultures, and they are therefore only assumed to exist in the
position of methanogens and host efficiency (Zhou et al. 2009). rumen ecosystem. Competition is likely to occur in this envi-
In two subsequent studies, the ruminal methanogenic and bac- ronment, as some species compete for the same resources, such
terial populations of 56 beef cattle which differed in feed effi- as cellulose. Therefore, as already noted, inhibition—a negative
ciency and diet were analyzed using PCR-DGGE interaction between the cellulolytic species—is observed when
profiles (Hernandez-Sanabria et al. 2010; Zhou et al. 2010). they are cocultured in vitro. Mutualism has also been
The results indicated that the methanogenic PCR-DGGE pattern documented between bacterial species, whereby some of the
is associated with the feed efficiency of the host, with diet- and species enjoy the cellulose-degradation products, fermentation
feed-efficiency-related bands being identified. However, the products, or secondary metabolites of other species without
size of the methanogenic population did not correlate with contributing to or affecting these ‘‘producers’’: this can be
differences in feed efficiency, diet, or metabolic measurements defined as a one-way positive interaction or commensalism.
(Zhou et al. 2010). Bacterial PCR-DGGE bands were only Such is the case with cellodextrins, which are cellulose-
related to feed-efficiency-associated traits and metabolites degradation products that are utilized by several non-cellulolytic
(Hernandez-Sanabria et al. 2010). Taken together, these species such as S. ruminantium and P. ruminicola (Russell 1985).
suggest a link between rumen microbiota and host feed Mutual interactions in which both partners benefit are also
efficiency. documented for bacteria with cross-feeding or utilization of
fermentation end products that allow better energy utilization
(Bryant and Wolin 1975; Miura et al. 1980, 1983). Such a two-
Rumen Interactions way positive interaction is nicely demonstrated by the interac-
tion between P. ruminicola, which has proteolytic capabilities,
The complexity of the rumen ecosystem is also expressed in the and R. albus, which is a cellulolytic species, in which the former
complicated interactions among its resident microbes and degrades protein and supplies NH3+ and the latter degrades
between those microbes and the ruminant host. The complexity cellulose and provides hydrolyzed cellulolytic products (Bryant
of these cooperative interactions finds its roots in the conflicting and Wolin 1975). Thus, within the bacterial domain, all possible
interests of the partners. The animal cannot digest its feed and, interactions have been documented.
therefore, needs its resident microbes to perform this task. The As noted above, bacteria have been reported to inhibit fungal
animal’s interest lies only in its resident microbes transforming cellulose degradation and growth, exhibiting a one-way negative
the feed to a more digestible form with minimal effect on its interaction, but synergism has also been shown between these
nutritional and energetic value. The microbes, on the other two domains, suggesting a two-way positive interaction. Pro-
hand, are interested in the plant fibers and their components, tozoa interact mainly with the other rumen microbes, and
which are continually being supplied by the animal. Conse- sometimes with each other, as predators exhibiting a one-way
quently, their primary interest is to use this organic material to negative interaction, although cellulose degradation and the
its fullest by extracting as much energy as possible. The common production of fermentation products are also part of their rumi-
plant feed resource also serves as a basis for conflict among the nal effect. Methanogenic archaea utilize the hydrogen produced
resident rumen microbes which are competing for it. These by other rumen microbes as a fermentation end product for
conflicting interests, in which each of the partners aims to use energy, which allows better energy utilization for other rumen
the resources to satisfy their own selfish interests, are expected to microbes. This is a two-way positive interaction with all other
interfere with possible cooperation between the partners. This domains which consequently benefits the ruminant host by
paradox, also known as ‘‘the tragedy of the commons’’ (Hardin improving fiber degradation within the rumen. Nevertheless,
1968), is thought to shape intraspecies and interspecies cooper- by producing methane which is eructated to the atmosphere
ative relationships (Frank 1996; Herre et al. 1999). The conflicts with its retained energy, methanogenic archaea cause energy
are resolved by several strategies and interactions which include losses for the ruminant host. Therefore, the methanogenic rela-
inhibition, predation, commensalism, and synergism. These tionship with the ruminant host is a combination of these
interactions, described earlier in this chapter, are discussed positive and negative interactions.
here in terms of their effects on all involved partners, i.e., As mentioned, the interactions of the resident microbes with
positive, negative or none. the ruminant host harbor a fundamental conflict in terms of
The bacteria, representing the dominant and most impor- energy harvest from the feed, which is resolved by a remarkably
tant domain in the rumen ecosystem, make up the only domain creative solution. The ruminant host provides stable conditions
that can single-handedly sustain the ruminant host’s viability. and a continuous supply of nutrients on which the bacteria and
The composition of the taxa in this domain changes according other microbes can prosper, and extracts the maximum possible
to the animal’s diet, as different bacterial species utilize different amount of energy from the feed within the limitations of the
substrates, and this in turn provides the remarkable ability to anaerobic conditions prevailing in the rumen ecosystem. The
maximize and fine-tune the host’s digestion and energy host, which is capable of aerobic respiration and gluconeogen-
harvesting from its feed. The interactions within the bacterial esis, makes maximal use of the end products of fermentation
domain are also highly complex and versatile. Most of these by the bacteria and other rumen microbes by oxidizing them to
carbon dioxide. The energy invested in building microbes’ rumen and feces of cattle fed different levels of dried distillers grains plus
solubles using bacterial tag-encoded FLX amplicon pyrosequencing. J Anim
cell components (mainly bacteria), which are mainly composed
Sci 88(12):3977–3983
of protein, is later retrieved in the abomasum where the Chen M, Wolin MJ (1977) Influence of CH4 production by Methanobacterium
microbial cells are digested, serving as the animal’s primary ruminantium on the fermentation of glucose and lactate by Selenomonas
source of protein. At the same time, the rumen microbes mul- ruminantium. Appl Environ Microbiol 34(6):756–759
tiply in the rumen compartment where they continue to prosper Chen J, Stevenson DM, Weimer PJ (2004) Albusin B, a bacteriocin from the
ruminal bacterium Ruminococcus albus 7 that inhibits growth of
(with the limitation of flow rate as a function of microbe gen-
Ruminococcus flavefaciens. Appl Environ Microbiol 70(5):3167–3170
eration time). This architecture enables all partners to satisfy Church DC (1969) Digestive physiology and nutrition of ruminants, 2nd edn.
their own selfish interests while conducting mutually beneficial O. S. U Book Stores, Corvallis, OR
associations which can be defined as a two-way positive Coleman GS (1962) The preparation and survival of almost bacteria-free suspen-
interaction. sions of Entodinium caudatum. J Gen Microbiol 28:271–281
Coleman GS (1964) The metabolism of Escherichia coli and other bacteria by
Entodinium caudatum. J Gen Microbiol 37:209–223
Coleman GS (1967a) The metabolism of free amino acids by washed suspensions
References of the rumen ciliate Entodinium caudatum. J Gen Microbiol 47(3):433–447
Coleman GS (1967b) The metabolism of the amino acids of Escherichia coli and
Aharoni Y, Brosh A, Kafchuk E (2006) The efficiency of utilization of metaboliz- other bacteria by the rumen ciliate Entodinium caudatum. J Gen Microbiol
able energy for milk production: a comparison of Holstein with F1 47(3):449–464
Montbeliarde 3 Holstein cows. Anim Sci 82:101–109 Cookson AL, Noel SJ, Kelly WJ, Attwood GT (2004) The use of PCR for the
Akin DE, Benner R (1988) Degradation of polysaccharides and lignin by ruminal identification and characterisation of bacteriocin genes from bacterial strains
bacteria and fungi. Appl Environ Microbiol 54(5):1117–1125 isolated from rumen or caecal contents of cattle and sheep. FEMS Microbiol
Akin DE, Lyon CE, Windham WR, Rigsby LL (1989) Physical degradation of Ecol 48(2):199–207
lignified stem tissues by ruminal fungi. Appl Environ Microbiol Counotte GH, van’t Klooster AT, van der Kuilen J, Prins RA (1979) An analysis of
55(3):611–616 the buffer system in the rumen of dairy cattle. J Anim Sci 49(6):1536–1544
Archer JA, Richardson EC, Herd RM, Arthur PF (1999) Potential for selection to Crews DH Jr (2005) Genetics of efficient feed utilization and national cattle
improve efficiency of feed use in beef cattle. Aust J Agric Res 50(2):147–162 evaluation: a review. Genet Mol Res 4(2):152–165
Arthur PF, Archer JA, Johnston DJ, Herd RM, Richardson EC, Parnell PF Czerkawski JW (1969) Methane production in ruminants and its significance.
(2001) Genetic and phenotypic variance and covariance components for World Rev Nutr Diet 11:240–282
feed intake, feed efficiency, and other postweaning traits in Angus cattle. Dehority BA (1991) Effects of microbial synergism on fibre digestion in the
J Anim Sci 79(11):2805–2811 rumen. Proc Nutr Soc 50(2):149–159
Balch WE, Fox GE, Magrum LJ, Woese CR, Wolfe RS (1979) Methanogens: Dehority BA (2003) Rumen microbiology. Nottingham University Press,
reevaluation of a unique biological group. Microbiol Rev 43(2):260–269 Nottingham, UK
Basarab JA, Price MA, Aalhus JL, Okine EK, Snelling WM, Lyle KL (2003) Residual Dehority BA, Tirabasso PA (2000) Antibiosis between ruminal bacteria and
feed intake and body composition in young growing cattle. Can J Anim Sci ruminal fungi. Appl Environ Microbiol 66(7):2921–2927
83:189–204 Dehority BA, Tirabasso PA (2001) Effect of feeding frequency on bacterial and
Bauchop T, Mountfort DO (1981) Cellulose fermentation by a rumen anaerobic fungal concentrations, pH, and other parameters in the rumen. J Anim Sci
fungus in both the absence and the presence of rumen methanogens. Appl 79(11):2908–2912
Environ Microbiol 42(6):1103–1110 Eadie JM, Hobson PN (1962) Effect of the presence or absence of rumen ciliate
Belanche A, Abecia L, Holtrop G, Guada JA, Castrillo C, de la Fuente G, Balcells J protozoa on the total rumen bacterial count in lambs. Nature 193:503–505
(2011) Study of the effect of presence or absence of protozoa on rumen Ellis JE, Williams AG, Lloyd D (1989) Oxygen consumption by ruminal micro-
fermentation and microbial protein contribution to the chyme. J Anim Sci. organisms: protozoal and bacterial contributions. Appl Environ Microbiol
doi:10.2527/jas.2010-3703 55(10):2583–2587
Bernalier A, Fonty G, Bonnemoy F, Gouet P (1993) Inhibition of the cellulolytic Ellis JL, Kebreab E, Odongo NE, McBride BW, Okine EK, France J (2007)
activity of Neocallimastix frontalis by Ruminococcus flavefaciens. J Gen Prediction of methane production from dairy and beef cattle. J Dairy Sci
Microbiol 139(4):873–880 90(7):3456–3466
Brulc JM, Antonopoulos DA, Miller ME, Wilson MK, Yannarell AC, Dinsdale EA, Elsden SR, Gilchrist FM, Lewis D, Volcani BE (1956) Properties of a fatty acid
Edwards RE, Frank ED, Emerson JB, Wacklin P, Coutinho PM, Henrissat B, forming organism isolated from the rumen of sheep. J Bacteriol 72(5):681–
Nelson KE, White BA (2009) Gene-centric metagenomics of the fiber- 689
adherent bovine rumen microbiome reveals forage specific glycoside hydro- Erickson DL, Nsereko VL, Morgavi DP, Selinger LB, Rode LM, Beauchemin KA
lases. Proc Natl Acad Sci USA 106(6):1948–1953 (2002) Evidence of quorum sensing in the rumen ecosystem: detection of
Bryant MP, Small N (1960) Observations on the ruminal microorganisms of N-acyl homoserine lactone autoinducers in ruminal contents. Can
isolated and inoculated calves. J Dairy Sci 43(5):654–667 J Microbiol 48(4):374–378
Bryant MP, Wolin MJ (1975) Rumen bacteria and their metabolic interactions. In: Fernando SC, Purvis HT II, Najar FZ, Sukharnikov LO, Krehbiel CR, Nagaraja
Proceedings of the first intersectional congress of the International Associ- TG, Roe BA, Desilva U (2010) Rumen microbial population dynamics
ation of Microbiology. Societies Science Council of Japan, Tokyo, vol 2, during adaptation to a high-grain diet. Appl Environ Microbiol
pp 297–306 76(22):7482–7490
Bryant MP, Small N, Bouma C, Robinsona I (1958) Studies on the composition of Ferrell CL, Jenkins TG (1985) Energy utilization by Hereford and Simmental
the ruminal flora and fauna of young calves. J Dairy Sci 41(12):1747–1767 males and females. Anim Prod 41:53–61
Cai S, Li J, Hu FZ, Zhang K, Luo Y, Janto B, Boissy R, Ehrlich G, Dong X (2010) Flint HJ (1997) The rumen microbial ecosystem—some recent developments.
Cellulosilyticum ruminicola, a newly described rumen bacterium that pos- Trends Microbiol 5(12):483–488
sesses redundant fibrolytic-protein-encoding genes and degrades lignocellu- Flint HJ, Bayer EA (2008) Plant cell wall breakdown by anaerobic microorganisms
lose with multiple carbohydrate-borne fibrolytic enzymes. Appl Environ from the mammalian digestive tract. Ann N Y Acad Sci 1125:280–288
Microbiol 76(12):3818–3824 Flint HJ, Bayer EA, Rincon MT, Lamed R, White BA (2008) Polysaccharide
Callaway TR, Dowd SE, Edrington TS, Anderson RC, Krueger N, Bauer N, utilization by gut bacteria: potential for new insights from genomic analysis.
Kononoff PJ, Nisbet DJ (2010) Evaluation of bacterial diversity in the Nat Rev Microbiol 6(2):121–131
Fondevila M, Dehority BA (2001a) In vitro growth and starch digestion by Johns AT (1951) Isolation of a bacterium, producing propionic acid, from the
Entodinium exiguum as influenced by the presence or absence of live bacteria. rumen of sheep. J Gen Microbiol 5(2):317–325
J Anim Sci 79(9):2465–2471 Johnson KA, Johnson DE (1995) Methane emissions from cattle. J Anim Sci
Fondevila M, Dehority BA (2001b) Preliminary study on the requirements of 73(8):2483–2492
Entodinium exiguum and E. caudatum for live or dead bacteria when cultured Johnson DE, Ferrell CL, Jenkins TG (2003) The history of energetic efficiency
in vitro. Reprod Nutr Dev 41(1):41–46 research: where have we been and where are we going? J Anim Sci 81:27–38
Fonty G, Gouet P, Jouany J-P, Senaud J (1987) Establishment of the microflora Jolles P, Schoentgen F, Jolles J, Dobson DE, Prager EM, Wilson AC (1984) Stomach
and anaerobic fungi in the rumen of lambs. J Gen Microbiol 133:1835–1843 lysozymes of ruminants. II. Amino acid sequence of cow lysozyme 2 and
Frank SA (1996) Host-symbiont conflict over the mixing of symbiotic lineages. immunological comparisons with other lysozymes. J Biol Chem
Proc Biol Sci 263(1368):339–344 259(18):11617–11625
Gorris LG, van der Drift C (1994) Cofactor contents of methanogenic bacteria Jolles J, Jolles P, Bowman BH, Prager EM, Stewart CB, Wilson AC (1989) Episodic
reviewed. Biofactors 4(3–4):139–145 evolution in the stomach lysozymes of ruminants. J Mol Evol 28(6):528–535
Guan LL, Nkrumah JD, Basarab JA, Moore SS (2008) Linkage of microbial Kalmokoff ML, Teather RM (1997) Isolation and characterization of a bacteriocin
ecology to phenotype: correlation of rumen microbial ecology to cattle’s (Butyrivibriocin AR10) from the ruminal anaerobe Butyrivibrio fibrisolvens
feed efficiency. FEMS Microbiol Lett 288(1):85–91 AR10: evidence in support of the widespread occurrence of bacteriocin-like
Gutierrez J (1958) Observations on bacterial feeding by the rumen ciliate Isotricha activity among ruminal isolates of B. fibrisolvens. Appl Environ Microbiol
prostoma. J Protozool 5(2):122–126 63(2):394–402
Gutierrez J, Davis RE (1959) Bacterial ingestion by the rumen ciliates Entodinium Kay RN (1966) The influence of saliva on digestion in ruminants. World Rev Nutr
and Diplodiniztm. J Protozool 6:222–226 Diet 6:292–325
Hackstein JH, Vogels GD (1997) Endosymbiotic interactions in anaerobic pro- Kay RN (1969) Digestion of protein in the intestines of adult ruminants. Proc
tozoa. Antonie Van Leeuwenhoek 71(1–2):151–158 Nutr Soc 28(1):140–151
Hardin G (1968) The tragedy of the commons: the population problem has no Koch RM, Swiger LA, Chambers D, Gregory KE (1963) Efficiency of feed use in
technical solution; it requires a fundamental extension in morality. Science beef cattle. J Anim Sci 22:486–494
162(3859):1243–1248 Krause DO, Denman SE, Mackie RI, Morrison M, Rae AL, Attwood GT,
Hegarty RS, Bird SH, Vanselow BA, Woodgate R (2008) Effects of the absence of McSweeney CS (2003) Opportunities to improve fiber degradation in the
protozoa from birth or from weaning on the growth and methane produc- rumen: microbiology, ecology, and genomics. FEMS Microbiol Rev
tion of lambs. Br J Nutr 100(6):1220–1227 27(5):663–693
Herd RM, Archer JA, Arthur PF (2003) Reducing the cost of beef production Krumholz LR, Forsberg CW, Veira DM (1983) Association of methanogenic
through genetic improvement in residual feed intake: opportunity and bacteria with rumen protozoa. Can J Microbiol 29(6):676–680
challenges to application. J Anim Sci 81:9–17 Latham MJ, Wolin MJ (1977) Fermentation of cellulose by Ruminococcus
Hernandez-Sanabria E, Guan LL, Goonewardene LA, Li M, Mujibi DF, Stothard flavefaciens in the presence and absence of Methanobacterium ruminantium.
P, Moore SS, Leon-Quintero MC (2010) Correlation of particular bacterial Appl Environ Microbiol 34(3):297–301
PCR-denaturing gradient gel electrophoresis patterns with bovine ruminal Laukova A (1993) Antagonistic activity of the rumen bacteria, Enterococcus
fermentation parameters and feed efficiency traits. Appl Environ Microbiol faecium and Staphylococcus warneri. Vet Med (Praha) 38(5):267–274
76(19):6338–6350 Ley RE, Turnbaugh PJ, Klein S, Gordon JI (2006) Microbial ecology: human gut
Herre EA, Knowlton N, Mueller U, Rehener S (1999) The evolution of mutual- microbes associated with obesity. Nature 444(7122):1022–1023
isms: exploring the paths between conflicts and cooperation. Trends Ecol Lloyd D, Hillman K, Yarlett N, Williams AG (1989) Hydrogen production by
Evol 14:49–53 rumen holotrich protozoa: effects of oxygen and implications for metabolic
Ho YW, Abdullah N, Jalaludin S (1988) Penetrating structure of anaerobic rumen control by in situ conditions. J Protozool 36(2):205–213
fungi in cattle and swamp buffalo. J Gen Microbiol 134:177–181 Mackie RI (2002) Mutualistic fermentative digestion in the gastrointestinal tract:
Holter JB, Young AJ (1992) Methane prediction in dry and lactating Holstein diversity and evolution. Integr Comp Biol 42(2):319–326
cows. J Dairy Sci 75(8):2165–2175 Mah RA (1964) Factors influencing the in vitro culture of the rumen ciliate
Hook SE, Wright AD, McBride BW (2011) Methanogens: methane producers of Ophryoscolex purkynei Stein. J Protozool 11:546–552
the rumen and mitigation strategies. Archaea 2010:945785 Marvin-Sikkema FD, Richardson AJ, Stewart CS, Gottschal JC, Prins RA
Hoover WH, Miller TK (1991) Rumen digestive physiology and microbial ecol- (1990) Influence of hydrogen-consuming bacteria on cellulose degradation
ogy. Vet Clin North Am Food Anim Pract 7(2):311–325 by anaerobic fungi. Appl Environ Microbiol 56(12):3793–3797
Hotovy SK, Johnson KA, Johnson DE, Carstens GE, Bourdon RM, Seidel GE Jr McAllister TA, Newbold CJ (2008) Redirecting rumen methane to reduce
(1991) Variation among twin beef cattle in maintenance energy require- methanogenesis. Aust J Exp Agric 48:7–13
ments. J Anim Sci 69(3):940–946 McAllister TA, Bae HD, Jones GA, Cheng KJ (1994) Microbial attachment and
Hungate RE, Smith W, Bauchop T, Yu I, Rabinowitz JC (1970) Formate as an feed digestion in the rumen. J Anim Sci 72(11):3004–3018
intermediate in the bovine rumen fermentation. J Bacteriol 102(2):389–397 Merchen NR, Elizalde JC, Drackley JK (1997) Current perspective on assessing
Irbis C, Ushida K (2004) Detection of methanogens and proteobacteria site of digestion in ruminants. J Anim Sci 75(8):2223–2234
from a single cell of rumen ciliate protozoa. J Gen Appl Microbiol 50(4): Minato H, Otsuka M, Shirasaka S, Itabashi H, Mitsumori M (1992) Colonization
203–212 of microorganisms in the rumen of young calves. J Gen Appl Microbiol
Jami E, Mizrahi I (2012) Composition and similarity of bovine rumen microbiota 38(5):447–456
across individual animals. PLoS One 7:e33306 Miron J, Ben-Ghedalia D, Morrison M (2001) Invited review: adhesion mecha-
Janssen PH (2010) Influence of hydrogen on rumen methane formation and nisms of rumen cellulolytic bacteria. J Dairy Sci 84(6):1294–1309
fermentation balances through microbial growth kinetics and fermentation Mitsumori M, Xu L, Kajikawa H, Kurihara M, Tajima K, Hai J, Takenaka A (2003)
thermodynamics. Anim Feed Sci Technol 160:1–22 Possible quorum sensing in the rumen microbial community: detection of
Janssen PH, Kirs M (2008) Structure of the archaeal community of the rumen. quorum-sensing signal molecules from rumen bacteria. FEMS Microbiol
Appl Environ Microbiol 74(12):3619–3625 Lett 219(1):47–52
Jeyanathan J, Kirs M, Ronimus RS, Hoskin SO, Janssen PH (2011) Methanogen Miura H, Horiguchi M, Matsumoto T (1980) Nutritional interdependence
community structure in the rumens of farmed sheep, cattle and red deer fed among rumen bacteria, Bacteroides amylophilus, Megasphaera elsdenii, and
different diets. FEMS Microbiol Ecol 76(2):311–326 Ruminococcus albus. Appl Environ Microbiol 40(2):294–300
Joblin KN, Naylor GE, Williams AG (1990) Effect of Methanobrevibacter smithii Miura H, Horiguchi M, Ogimoto K, Matsumoto T (1983) Nutritional
on xylanolytic activity of anaerobic ruminal fungi. Appl Environ Microbiol interdependence among rumen bacteria during cellulose digestion in vitro.
56(8):2287–2295 Appl Environ Microbiol 45(2):726–729
Moore SS, Mujibi FD, Sherman EL (2009) Molecular basis for residual feed intake Saluzzi L, Smith A, Stewart CS (1993) Analysis of bacterial phospholipid markers
in beef cattle. J Anim Sci 87(14 Suppl):E41–E47 and plant monosaccharides during forage degradation by Ruminococcus
Morovsky M, Pristas P, Javorsky P (2001) Bacteriocins of ruminal bacteria. Folia flavefaciens and Fibrobacter succinogenes in co-culture. J Gen Microbiol
Microbiol (Praha) 46(1):61–62 139(11):2865–2873
Mosoni P, Martin C, Forano E, Morgavi DP (2011) Long-term defaunation Solis JC, Byers FM, Schelling GT, Long CR, Greene LW (1988) Maintenance
increases the abundance of cellulolytic ruminococci and methanogens but requirements and energetic efficiency of cows of different breed types.
does not affect the bacterial and methanogen diversity in the rumen of sheep. J Anim Sci 66(3):764–773
J Anim Sci 89(3):783–791 Stams AJ, Plugge CM (2009) Electron transfer in syntrophic communities of
Muegge BD, Kuczynski J, Knights D, Clemente JC, Gonzalez A, Fontana L, anaerobic bacteria and archaea. Nat Rev Microbiol 7(8):568–577
Henrissat B, Knight R, Gordon JI (2011) Diet drives convergence in gut Stevenson DM, Weimer PJ (2007) Dominance of Prevotella and low abundance
microbiome functions across mammalian phylogeny and within humans. of classical ruminal bacterial species in the bovine rumen revealed by
Science 332(6032):970–974 relative quantification real-time PCR. Appl Microbiol Biotechnol 75(1):
Muller M (1993) The hydrogenosome. J Gen Microbiol 139(12):2879–2889 165–174
Newbold CJ, Lassalas B, Jouany JP (1995) The importance of methanogens Stewart CS (1986) Rumen function with special reference to fibre digestion. Soc
associated with ciliate protozoa in ruminal methane production in vitro. Appl Bacteriol Symp Ser 13:263–286
Lett Appl Microbiol 21(4):230–234 Stewart CS, Duncan SH, Richardson AJ, Backwell C, Begbie R (1992) The
Nkrumah JD, Basarab JA, Price MA, Okine EK, Ammoura A, Guercio S, inhibition of fungal cellulolysis by cell-free preparations from ruminococci.
Hansen C, Li C, Benkel B, Murdoch B, Moore SS (2004) Different measures FEMS Microbiol Lett 76(1–2):83–87
of energetic efficiency and their phenotypic relationships with growth, feed Storm E, Ørskov ER (1983) The nutritive value of rumen microorganisms in
intake, and ultrasound and carcass merit in hybrid cattle. J Anim Sci ruminant. 1. Large-scale isolation and chemical composition of rumen
82(8):2451–2459 microorganisms. Br J Nutr 50:463–470
Nkrumah JD, Okine EK, Mathison GW, Schmid K, Li C, Basarab JA, Price MA, Sultana H, Miyazawa K, Kanda S, Itabashi H (2011) Fatty acid composition of
Wang Z, Moore SS (2006) Relationships of feedlot feed efficiency, perfor- ruminal bacteria and protozoa, and effect of defaunation on fatty acid profile
mance, and feeding behavior with metabolic rate, methane production, and in the rumen with special reference to conjugated linoleic acid in cattle. Anim
energy partitioning in beef cattle. J Anim Sci 84(1):145–153 Sci J 82(3):434–440
Odenyo AA, Mackie RI, Stahl DA, White BA (1994a) The use of 16S rRNA-targeted Taylor SCS, Thiessen RB, Murray J (1986) Inter-breed relationship of mainte-
oligonucleotide probes to study competition between ruminal fibrolytic bac- nance efficiency to milk yield in cattle. Anim Prod 43:37–61
teria: development of probes for Ruminococcus species and evidence for Teunissen MJ, Kets EP, Op den Camp HJ, Huis in’t Veld JH, Vogels GD
bacteriocin production. Appl Environ Microbiol 60(10):3688–3696 (1992) Effect of coculture of anaerobic fungi isolated from ruminants and
Odenyo AA, Mackie RI, Stahl DA, White BA (1994b) The use of 16 S rRNA- non-ruminants with methanogenic bacteria on cellulolytic and xylanolytic
targeted oligonucleotide probes to study competition between ruminal enzyme activities. Arch Microbiol 157(2):176–182
fibrolytic bacteria: pure-culture studies with cellulose and alkaline Thauer RK, Kaster AK, Seedorf H, Buckel W, Hedderich R (2008) Methanogenic
peroxide-treated wheat straw. Appl Environ Microbiol 60(10):3697–3703 archaea: ecologically relevant differences in energy conservation. Nat Rev
Ogimoto K, Imai S (1981) Atlas of rumen microbiology. Japan Scientific Societies Microbiol 6(8):579–591
Press, Tokyo, Japan Thiessen RB, Taylor SCS, Murray J (1985) Multibreed comparisons of British
Orpin CG (1975) Studies on the rumen flagellate Neocallimastix frontalis. J Gen cattle. Variation in relative growth rate, relative food intake and food con-
Microbiol 91(2):249–262 version efficiency. Anim Prod 41:193–199
Ozutsumi Y, Tajima K, Takenaka A, Itabashi H (2005) The effect of protozoa on Tokura M, Chagan I, Ushida K, Kojima Y (1999) Phylogenetic study of
the composition of rumen bacteria in cattle using 16S rRNA gene clone methanogens associated with rumen ciliates. Curr Microbiol 39(3):123–128
libraries. Biosci Biotechnol Biochem 69(3):499–506 Tolkamp BJ (2010) Efficiency of energy utilisation and voluntary feed intake in
Pavlostathis SG, Miller TL, Wolin MJ (1988) Fermentation of insoluble cellulose ruminants. Animal 4(7):1084–1092
by continuous cultures of Ruminococcus albus. Appl Environ Microbiol Tothova T, Piknova M, Kisidayova S, Javorsky P, Pristas P (2008) Distinctive
54(11):2655–2659 archaebacterial species associated with anaerobic rumen protozoan
Ranilla MJ, Jouany JP, Morgavi DP (2007) Methane production and substrate Entodinium caudatum. Folia Microbiol (Praha) 53(3):259–262
degradation by rumen microbial communities containing single protozoal Trinci AP, Lowe SE, Milne A, Theodorou MK (1988) Growth and survival of
species in vitro. Lett Appl Microbiol 45(6):675–680 rumen fungi. Biosystems 21(3–4):357–363
Ricard G, McEwan NR, Dutilh BE, Jouany JP, Macheboeuf D, Mitsumori M, Turnbaugh PJ, Gordon JI (2009) The core gut microbiome, energy balance and
McIntosh FM, Michalowski T, Nagamine T, Nelson N, Newbold CJ, obesity. J Physiol 587(17):4153–4158
Nsabimana E, Takenaka A, Thomas NA, Ushida K, Hackstein JH, Huynen Turnbaugh PJ, Ley RE, Mahowald MA, Magrini V, Mardis ER, Gordon JI
MA (2006) Horizontal gene transfer from bacteria to rumen ciliates indicates (2006) An obesity-associated gut microbiome with increased capacity for
adaptation to their anaerobic, carbohydrates-rich environment. BMC energy harvest. Nature 444(7122):1027–1031
Genomics 7:22 Ushida K, Jouany JP (1996) Methane production associated with rumen-ciliated
Richardson EC, Herd RM, Arthur PF, Wright J, Xu G, Dibley K, Oddy VH protozoa and its effect on protozoan activity. Lett Appl Microbiol 23(2):129–132
(1996) Possible physiological indicators of net feed conversion efficiency. Vogels GD, Hoppe WF, Stumm CK (1980) Association of methanogenic bacteria
Proc Austral Soc Anim Prod 21:901–908 with rumen ciliates. Appl Environ Microbiol 40(3):608–612
Richardson EC, Herd RM, Archer JA, Arthur PF (2004) Metabolic differences in Weimer PJ, Russell JB, Muck RE (2009) Lessons from the cow: what the ruminant
Angus steers divergently selected for residual feed intake. Aust J Exp Agric animal can teach us about consolidated bioprocessing of cellulosic biomass.
44:441–452 Bioresour Technol 100(21):5323–5331
Russell JB (1985) Fermentation of cellodextrins by cellulolytic and noncellulolytic Welch JG (1986) Physical parameters of fiber affecting passage from the rumen.
rumen bacteria. Appl Environ Microbiol 49(3):572–576 J Dairy Sci 69(10):2750–2754
Russell JB, Rychlik JL (2001) Factors that alter rumen microbial ecology. Science Welch JG, Smith AM (1969) Influence of forage quality on rumination time in
292(5519):1119–1122 sheep. J Anim Sci 28(6):813–818
Russell JB, Wilson DB (1996) Why are ruminal cellulolytic bacteria unable to Welkie DG, Stevenson DM, Weimer PJ (2009) ARISA analysis of ruminal bacte-
digest cellulose at low pH? J Dairy Sci 79(8):1503–1509 rial community dynamics in lactating dairy cows during the feeding cycle.
Rychlik JL, Russell JB (2002) Bacteriocin-like activity of Butyrivibrio fibrisolvens Anaerobe 16(2):94–100
JL5 and its effect on other ruminal bacteria and ammonia production. Appl Whitford MF, McPherson MA, Forster RJ, Teather RM (2001) Identification
Environ Microbiol 68(3):1040–1046 of bacteriocin-like inhibitors from rumen Streptococcus spp. and
isolation and characterization of bovicin 255. Appl Environ Microbiol Yanagita K, Kamagata Y, Kawaharasaki M, Suzuki T, Nakamura Y, Minato H
67(2):569–574 (2000) Phylogenetic analysis of methanogens in sheep rumen ecosystem and
Williams AG, Coleman GS (1992) The rumen protozoa. Springer, New York detection of Methanomicrobium mobile by fluorescence in situ hybridiza-
Williams AG, Withers SE, Joblin KN (1991) Xylanolysis by cocultures of the tion. Biosci Biotechnol Biochem 64(8):1737–1742
rumen fungus Neocallimastix frontalis and ruminal bacteria. Lett Appl Yanez-Ruiz DR, Williams S, Newbold CJ (2007) The effect of absence of protozoa
Microbiol 12(6):232–235 on rumen biohydrogenation and the fatty acid composition of lamb muscle.
Windham WR, Akin DE (1984) Rumen fungi and forage fiber degradation. Appl Br J Nutr 97(5):938–948
Environ Microbiol 48(3):473–476 Zhou M, Hernandez-Sanabria E, Guan LL (2009) Assessment of the microbial
Wolin MJ (1979) The rumen fermentation: a model for microbial interactions in ecology of ruminal methanogens in cattle with different feed efficiencies.
anaerobic ecosystems. Adv Microb Ecol 3:49–77 Appl Environ Microbiol 75(20):6524–6533
Wuebbles DJ, Hayhoe K (2002) Atmospheric methane and global change. Earth Zhou M, Hernandez-Sanabria E, Guan LL (2010) Characterization of varia-
Sci Rev 57:117–210 tion in rumen methanogenic communities under different dietary and
Xavier BM, Russell JB (2009) The ability of non-bacteriocin producing Strepto- host feed efficiency conditions, as determined by PCR-denaturing
coccus bovis strains to bind and transfer bovicin HC5 to other sensitive gradient gel electrophoresis analysis. Appl Environ Microbiol 76(12):
bacteria. Anaerobe 15(4):168–172 3776–3786
22 Symbiotic Associations Between Termites
and Prokaryotes
Andreas Brune
Max Planck Institute for Terrestrial Microbiology, Marburg, Germany
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545 Lignin-Degrading Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . 561

Oxygen Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
Symbiotic Digestion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 546 Microbial Fermentations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
Lactobacillales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
Fiber Degradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 546 Clostridiales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
Role of Intestinal Protozoa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547 Bacteroidetes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
The Role of Fungi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 548 Hydrogen Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
The Role of Host Enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 548 Hydrogen-Producing Microorganisms . . . . . . . . . . . . . . 562
Soil-Feeding Termites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 548 Homoacetogenic Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . 563
Methanogenic Archaea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564
Host Nutrition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549 Sulfate-Reducing Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . 564
Nitrogen Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564
The Gut Microenvironment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549 Nitrogen Recycling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564
Nitrogen Fixation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 565
Physicochemical Gradients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549
Redox Conditions and Oxygen Status . . . . . . . . . . . . . . . . . . . 549 Symbiotic Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 565
Hydrogen Partial Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 550
Intestinal pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551 Microbe–Microbe Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 566
Interactions Among Prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . 566
Gut Compartmentation and Microhabitats . . . . . . . . . . . . . . . 552 Interactions Between Prokaryotes and Protozoa . . . . . . . . 566
Midgut Epithelium, Ectoperitrophic Space . . . . . . . . . . . . . 552
Hindgut Cuticle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552 Microbe–Host Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567
Hindgut Protozoa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552 Pathogens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 568
Intracellular Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 568
Prokaryotic Gut Symbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552
Mutualists or Commensals? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 568
Morphological Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 570
Phylogenetic Diversity and Community Structure . . . . . . . . 553
Bacterial Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553
Spirochaetes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554 Introduction
Endomicrobia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554
Firmicutes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555 The symbiotic associations of termites with microorganisms
Bacteroidetes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555 comprise different levels of interaction, ranging from the extra-
Other Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555 corporeal cultivation of fungus gardens to the most intimate
Archaeal Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555 associations, where bacteria reside intracellularly in dedicated
Methanogenic Archaea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 556 bacteriocytes. However, the majority of prokaryotic symbionts
Non-methanogenic Archaea . . . . . . . . . . . . . . . . . . . . . . . . . 556 of termites are located in the intestinal tract, where they are free-
Spatial Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 557 swimming, attached to the gut epithelium, or associated with
the intestinal protozoa (> Fig. 22.1). Although it is suggestive
Isolates and Major Metabolic Activities . . . . . . . . . . . . . . . . . . . 557 that the gut microbiota of termites is directly or indirectly
Numerically Predominant Isolates . . . . . . . . . . . . . . . . . . . . . . 557 involved in the digestion of lignocellulose or has other nutri-
Cultivation Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 558 tional implications, the exact nature of the associations and
Lignocellulose Degradation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 558 possible benefits for the partners of each particular symbiosis
Cellulolytic Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 559 are often far from clear. Therefore, this chapter will use the term
Hemicellulolytic Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . 560 ‘‘symbiosis’’ in its broader sense, as originally defined by Anton
546 22 Symbiotic Associations Between Termites and Prokaryotes
a b
100 µm 2 µm
5 µm
. Fig. 22.1
Examples of microbial symbionts in the hindgut of Reticulitermes flavipes (Isoptera: Rhinotermitidae), a wood-feeding lower termite. (a)
Preparation of anaerobic protozoa from the hindgut of a worker larva, showing the large hypermastigote flagellate Trichonympha agilis,
filled with wood particles, and numerous smaller flagellates (mainly oxymonads, Dinenympha spp.). Differential interference contrast
photomicrograph taken by U. Stingl. (b) Transverse section through the peripheral hindgut, showing the diverse bacterial microbiota
associated with the thin cuticle of the hindgut wall (bottom left). Transmission electron micrograph provided by J. A. Breznak.
(c) Preparation of the hindgut wall, showing the dense colonization of the cuticle by numerous rod-shaped and filamentous bacterial
morphotypes. Scanning electron micrograph provided by J. A. Breznak (Reproduced from Brune 2003)
de Bary (1878). A definitive classification of the associations into the intestinal symbioses provide metabolic capacities that are
the different categories of symbiosis, such as mutualism, para- otherwise not available to the host. For reviews, see Breznak and
sitism, or commensalism, would require a level of understanding Brune (1994), Kane (1997), Brune (1998, 2003), Bignell (2000),
that is yet to be reached. Breznak (2000), Brune and Friedrich (2000), and Ohkuma
In view of the enormous body of literature on the intestinal (2003).
microbiota of termites and its role in lignocellulose digestion, The symbiotic digestion of lignocellulose by termites is
the subject cannot be covered exhaustively. This chapter will a complex series of events involving both the host and its gut
attempt to summarize the current state of knowledge on the microbiota (> Fig. 22.3). While the events in the foregut and
prokaryotic communities within the intestinal tracts of termites, midgut seem to be mainly due to host activities, the digestive
the major populations and their metabolic activities, and their processes in the hindgut are largely controlled by the symbionts.
interactions. In addition, it will focus on the gut as a microbial Many aspects of lignocellulose digestion are common to all
habitat. The chapter will touch only briefly on the intestinal termites, but there are also several noteworthy differences
flagellates, which are most important in the phylogenetically between the phylogenetically lower and higher taxa.
lower termites, the exosymbiotic fungi in fungus-cultivating
Macrotermitinae, and the intracellular bacteria in termite tis-
sues. For details on these subjects and for many other aspects of Fiber Degradation
the termite gut symbiosis, the reader will be referred to the
pertinent review articles. The degradation of plant cell walls requires the synergistic action
of many different enzymes and, in the case of lignified substrates,
also a mechanism to break up the lignocellulose complex
Symbiotic Digestion (Breznak and Brune 1994). Microorganisms, i.e., bacteria, pro-
tozoa, and fungi, are the most efficient cellulose and hemicellu-
Termites, like other insects thriving on a lignocellulosic diet, lose degraders in nature, and fungi and certain actinomycetes are
possess a pronounced gut microbiota housed in specially also the only organisms that have developed a strategy for the
adapted regions of the alimentary tract (> Fig. 22.2). The sym- chemical breakdown of lignin (Béguin and Aubert 1994; Jeffries
bionts convert a substantial portion of the dietary components 1994). Not surprisingly, termites and other animals have made
to microbial fermentation products, which are then eventually use of these capacities by employing microbial symbionts in the
resorbed by the intestinal epithelia. It is generally assumed that digestion of lignocellulosic food (Martin 1983).
Symbiotic Associations Between Termites and Prokaryotes 22 547
a b Role of Intestinal Protozoa

C C
M The presence of protozoa in termite guts was recognized very
M early, although they were initially considered to be parasites
(e.g., Leidy 1881; Koidzumi 1921). The American protozoologist
L. R. Cleveland recognized that the wood-feeding lifestyle in the
ms
evolutionarily ‘‘lower’’ termites is based on a mutualistic asso-
P1 ciation with their intestinal protozoa (Cleveland 1925a, 1926).
In a series of elegant experiments, he established that the ability
Pa
of lower termites to live on a diet of wood or cellulose depends
on the digestive capacities of their intestinal flagellates. His
P3 pioneering work, published in 1923–1928, paved the way for
many later studies (see reviews by Honigberg (1970), Inoue et al.
(2000), and Brune and Stingl (2005)).
About 15 years later, Hungate elucidated the biochemical
2 mm
basis for this symbiosis (Hungate 1939, 1943). He showed that
P4
the gut flagellates depolymerize and ferment lignocellulose to
short-chain fatty acids, which are resorbed and oxidized by the
R P5 host [reviewed by Hungate (1955) and Breznak and Brune
(1994)]. The importance of protozoa for the metabolic processes
in the hindgut of lower termites is most impressively evidenced
. Fig. 22.2
by their enormous numbers, which may constitute more than
Structure of the intestinal tract of Reticulitermes species (a) and
one-third of the body mass in Zootermopsis species (Katzin and
Cubitermes species (b). All lower termites harbor the gut
Kirby 1939).
microbiota in a single, strongly dilated hindgut paunch (Pa)
There is a large body of literature on the decomposition of
that tapers out via the colon into the rectum (R). In most
wood and cellulose by termite gut flagellates [for references, see
higher termites, especially the soil-feeding species, the hindgut
O’Brien and Slaytor (1982), Breznak and Brune (1994), and
is more elongated and has additional dilatations. Abbreviations:
Inoue et al. (2000)]. Apparently, the different flagellate species
crop C, midgut M, mixed segment ms, and proctodeal segments
are nutritionally specialized, and each species might fill a specific
P1–P5
niche in lignocellulose digestion (Yoshimura et al. 1996;
Hindgut
Foregut Midgut G 6 p
a
L F
b
7
1 2 3 4 C 5 c d
N
Fermentation products
1 Mandibles 6 Phagosomes L Lignocellulose

2 Salivary glands 7 Proctodeal feeding C Carbohydrates
3 Proventriculus N Nitrogenous compounds
4 Midgut epithelium G Gut lumen F Fecal matter
5 Malplghlan tubules P Protozoa
. Fig. 22.3
Major events in the symbiotic digestion of lignocellulose by wood-feeding lower termites. The black lines show the path of the insoluble
material whose lignin-rich residues are released as feces, whereas the red lines represent soluble degradation products, which are
eventually resorbed by the host. The green lines indicate the cycling of nitrogenous compounds. Blue arrows mark the sites where
cellulolytic enzymes are secreted. Lower-case letters refer to the different groups of bacteria, which are either endobionts (a) or epibionts
(b) of the protozoa, suspended in the gut lumen (c) or attached to the gut wall (d). The scheme has been simplified for the sake of clarity;
not all possible interactions are shown (Adapted from Brune 2003)
Inoue et al. 2000). Most of the endoxylanase and b-xylosidase 1978, 1979) gave rise to the acquired enzyme hypothesis of Martin
activity in the lower termite Reticulitermes speratus is located in (1983). However, claims that the fungal cellulases are essential
the anterior hindgut and is lost upon defaunation (removal of for cellulose digestion in the termite gut remain controversial
protozoa by ultraviolet irradiation; Inoue et al. 1997), and the (Slaytor 1992; Bignell et al. 1994; Crosland et al. 1996), especially
effects of artificial diets on the composition of the protozoan in view of the recently discovered ability of termites to produce
community corroborate that different gut flagellates are their own cellulases (see the section > ‘‘The Role of Host
involved in xylan and cellulose degradation (Inoue et al. 1997; Enzymes’’ in this chapter).
Cook and Gold 2000).
The protozoa possess their own cellulase genes, which fall
into different glycosyl hydrolase families (Ohtoko et al. 2000; The Role of Host Enzymes
Nakashima et al. 2002a; Watanabe et al. 2002; Li et al. 2003;
Inoue et al. 2005) and may even exploit host cellulases that are Since phylogenetically higher termites (family: Termitidae) lost
secreted in the anterior gut regions (Li et al. 2003). There is no their gut flagellates in the course of termite evolution, it was
evidence that the prokaryotic symbionts of the gut flagellates initially assumed that either ingested fungal enzymes (see the
(see the section > ‘‘Interactions Between Prokaryotes and Pro- section > ‘‘The Role of Fungi’’ in this chapter) or prokaryotic
tozoa’’ in this chapter) confer cellulolytic activity to their hosts. symbionts took over the function of the cellulolytic protozoa.
However, there is still no clear evidence that bacteria play
a major role in cellulose degradation in any of the termites
The Role of Fungi investigated (see the section > ‘‘Cellulolytic Bacteria’’ in this
chapter), which may find its explanation in the recently discov-
The role of fungi in the digestion of lignocellulose by termites is ered ability of termites to produce their own cellulases.
less clear. There are termites that can thrive on sound wood, but In all insects, the digesta are exposed to a variety of digestive
many species show a strong preference for decaying wood colo- enzymes secreted by the salivary glands and the midgut epithe-
nized by saprophytic fungi, which may either precondition the lium (Terra 1990). Cook (1943) had already demonstrated that
wood for digestion or provide metabolic products important for termites are able to absorb sugars directly, and evidence is
termite nutrition (Sands 1969; Rouland 2000; Cornelius et al. accumulating that termites secrete a full complement of
2002). Schäfer et al. suggested that the yeasts and other fungi enzymes necessary for the digestion of plant structural poly-
present in the guts of the lower termites Zootermopsis saccharides, including cellulose, into the midgut (e.g., Rouland
angusticollis and Neotermes castaneus are involved in et al. 1989; Slaytor 1992; Rouland 2000). The presence of prote-
hemicellulolytic degradation (Schäfer et al. 1996). ase and lysozyme activities has been documented for several
Higher termites of the subfamily Macrotermitinae have termites, which indicates that also microbial cells can be digested
established a unique exosymbiosis with basidiomycetes of the (Rohrmann and Rossman 1980; Fujita et al. 2001; Fujita and Abe
genus Termitomyces, which are maintained on predigested plant 2002; Fujita et al. 2002). Although experimental evidence is
litter in so-called fungus gardens within the nests. The symbiosis scarce, one can safely assume that – as in other insects – most
has rendered fungus-cultivating termites independent of the of the easily digestible material has been mobilized and resorbed
intestinal protozoa, which probably allowed for the obvious by the time the digesta reach the end of the midgut (> Fig. 22.3).
diversification in their diet (Sands 1969; Rouland 2000). The The persisting dogma that higher animals do not possess
specificity of this symbiosis, whose enormous evolutionary suc- their own cellulases has been unequivocally refuted by the dem-
cess is impressively evidenced by the huge nests of fungus- onstration of endoglucanase genes in the termite genome and
cultivating termites populating the grasslands of Africa, is their expression in the cells of the midgut epithelium and in the
documented by several instances of coevolution between the salivary glands (reviewed by Watanabe and Tokuda 2001). Even
termites and their fungal partners, indicating both horizontal in lower termites, host cellulases secreted by the salivary glands
and vertical transmission of the fungal symbionts (Aanen et al. and complement (and surpass) the cellulolytic activities of the
2002; Katoh 2002; Rouland-Lefevre et al. 2002; Taprab et al. intestinal protozoa in the hindgut (Nakashima et al. 2002b;
2002). Tokuda et al. 2004). There is evidence that glycosyl hydrolase
The association with the lignin-degrading fungus enables the family 9 (GHF9) cellulases present in the genomes of termites
fungus-cultivating termites to utilize lignocellulose nearly are ancient and widespread in Metazoa (Lo et al. 2003b; Davison
completely, as reflected in the small volume of their final feces and Blaxter 2005).
(Darlington 1994). The key activities attributed to the fungal
partner in this mutualistic symbiosis are extensive
delignification of the substrate (Hyodo et al. 1999, 2000; Johjima Soil-Feeding Termites
et al. 2003) and the conversion of plant fiber to fungal biomass,
as proposed earlier in the lignin degradation hypothesis of Grassé The majority of termite species are humivorous, yet little is
and Noirot (1958). Evidence for an activity within the gut of known about the exact nature of the dietary components
fungal cellulases ingested by the termites together with the exploited by these ecologically important soil macroinver-
fungus comb material (Abo-Khatwa 1978; Martin and Martin tebrates (Brauman et al. 2000). Besides fragments of plant tissue,
fungal hyphae, and numerous microorganisms, their diet con- the digestion of microbial biomass acquired in the course of anal
sists largely of undefined humic material intimately associated trophallaxis supplies them with high-quality nutrients (Machida
with the mineral soil matrix (Donovan et al. 2001). While the et al. 2001). The gut microbiota supplies essential precursors for
aromatic component of humus was initially assumed to be the the biosynthesis, e.g., of methyl-branched hydrocarbons (Guo
principal substrate of soil-feeding termites (Noirot 1992; Bignell et al. 1991), and might play a role in nestmate recognition
1994), feeding trials with soil-feeding Cubitermes spp. have (Matsuura 2001). To date, the lack of knowledge on the individ-
shown that peptidic soil components – free or polymerized ual components of the prokaryotic microbiota and their meta-
into humic model compounds – are preferentially digested and bolic capacities and activities in situ still makes it difficult to
mineralized (Ji et al. 2000; Ji and Brune 2001, 2005). define the essential functions and understand the complex
The anterior hindgut of soil-feeding termites is extremely interactions.
alkaline (Bignell and Eggleton 1995; Brune and Kühl 1996),
which favors the extraction of organic matter from the soil
(Brune 1998; Kappler and Brune 1999). Microbial biomass and The Gut Microenvironment
its structural components are assimilated more efficiently than
cellulose, which supports the hypothesis that soil microorgan- The intestinal tract of insects is organized into three major gut
isms and the nitrogenous components of humus are an impor- regions: a short foregut, a midgut (which is the main site of
tant dietary resource for humivorous soil macroinvertebrates digestion), and a usually short hindgut (proctodeum). The
(Ji et al. 2000; Ji and Brune 2001). hindguts of all termites, however, have immensely increased in
length and volume over the course of evolution (> Fig. 22.2). In
the more primitive, lower termites, the hindgut is still relatively
Host Nutrition simple, consisting of a dilated ‘‘hindgut paunch’’ that tapers out
into the colon and ends in the rectal compartment (Noirot
Irrespective of their contribution to polymer degradation, the 1995). While this organization has been retained in the fun-
majority of prokaryotes in termite guts are probably involved in gus-cultivating termites (Termitidae: Macrotermitinae), all
the fermentation of the soluble products released into the gut other lineages of higher termites show a trend toward a further
(see the section > ‘‘Microbial Fermentations’’ in this chapter), elongation and additional compartmentalization of the hindgut
which are derived either directly from the food by the digestive (Noirot 2001), which is most pronounced in the soil-feeding
enzymes (see the section > ‘‘The Role of Host Enzymes’’ in this representatives. The gut morphologies of lower and higher ter-
chapter) or by the fermentative activity of the intestinal pro- mites and the significance of the adaptations for the digestive
tozoa (see the section > ‘‘Role of Intestinal Protozoa’’ in this process have been reviewed exhaustively (Noirot 1995, 2001).
chapter). The major products of the hindgut metabolism are The proctodeal dilatations increase the residence time of the
acetate and, to a smaller extent, other short-chain fatty acids digesta, thereby prolonging the exposure to the activities of the
(mainly propionate and butyrate), which accumulate in the intestinal microbiota. Moreover, host factors and microbial
hindgut fluid and are eventually resorbed by the hindgut epi- activities give rise to physicochemical gradients that create dis-
thelium (> Fig. 22.3). Termites – like other insects – cannot use tinct microenvironmental conditions in each gut compartment.
acetate as a substrate for gluconeogenesis, but as long as the This has been shown for oxygen, hydrogen, redox potential, and
digestive processes in the midgut release sufficient amounts of intestinal pH and has to be expected also for any other metab-
soluble sugars and amino acids, this is not a problem. Breznak olite when source and sink are spatially separated (Brune and
calculated that acetate produced by the hindgut microbiota of Friedrich 2000), especially since the microbiota is not randomly
Reticulitermes flavipes would suffice to support the respiratory distributed within the gut (see the section > ‘‘Spatial Organiza-
activity of the host (Breznak 2000). tion’’ in this chapter).
Besides being difficult to degrade, lignocellulose is also an
extremely nutrient-poor substrate. While non-lignified plant
cells are usually rich in protein and other nitrogenous com- Physicochemical Gradients
pounds, the C-to-N ratio of sound wood is up to 100-fold higher
than that of the insect body (La Fage and Nutting 1978). More- Redox Conditions and Oxygen Status
over, a lignocellulosic diet lacks most of the essential nutrients
required by animals, such as amino acids, vitamins, and sterols. The general concept of termite guts as anoxic habitats had been
Many microorganisms are capable of fixing dinitrogen, assimi- based on several pieces of circumstantial evidence (outlined by
lating nitrate and ammonia, or synthesizing those amino acids Veivers et al. 1980): (1) the oxygen sensitivity of the intestinal
and vitamins essential for the host. Animals, including termites, protozoa already recognized by Cleveland (1925a, b); (2) the
have developed means of exploiting these biosynthetic capaci- demonstration of a fermentative metabolism of cellulose by
ties, which include – in the simplest case – the digestion of the these flagellates (Hungate 1939, 1943) and the high concentra-
intestinal symbionts. tions of microbial fermentation products in the hindgut of all
Wood-feeding termites, especially those feeding on sound termites investigated; and (3) the presence of oxygen-sensitive or
wood, have an extreme shortage of nutrients in their diet, and strictly anaerobic processes, such as nitrogen fixation and
methanogenesis (Breznak et al. 1973; Breznak 1975). Also, the Partial pressure of gases (kPa)
subsequent isolation of anaerobic bacteria from termite guts (see 0 2 4 6 8 10
0
the section > ‘‘Isolates and Major Metabolic Activities’’ in this
chapter) supported the general assumption that the principle of
symbiotic digestion in termite guts was analogous to that in the 300
rumen. Agarose
Slaytor and coworkers were the first to question the anoxic 600 O2
status of termite guts. Following the color reaction in the hind-
Depth (µm)
gut of redox indicator dyes fed to Nasutitermes exitiosus and 900
Coptotermes lacteus, they initially claimed that the hindgut
paunch was ‘‘aerobic,’’ since methylene blue remained oxidized
(Eutick et al. 1976). In a later study, however, using a more 1200 H2
refined technique, they obtained Eh values between –230 mV Anoxic Gut
and –270 mV in the hindgut paunch of these and seven other 1500 Oxic
termite species (Veivers et al. 1980). The initial error had been
caused by the color of the reduced dye within the gut being 1800
obscured by that of the oxidized dye, which had also impreg-
nated the oxic gut epithelium.
. Fig. 22.4
Moreover, Slaytor and coworkers demonstrated that the
Radial gradients of oxygen (●) and hydrogen (○) in an agarose-
vitality of Nasutitermes exitiosus and Coptotermes lacteus
embedded hindgut (paunch region) of Reticulitermes flavipes
depended on the presence of their prokaryotic gut microbiota
worker larva. A schematic cross-section through the paunch
(Eutick et al. 1978b) and that bacteria play an important role in
illustrates the relative sizes of the oxic and anoxic zones (Modified
maintaining the low redox potential of the hindgut paunch
after Brune 1998)
(Veivers et al. 1982), which led to the postulation that they
maintain anoxic conditions by removing oxygen from the
hindgut.
Bignell (1984) pointed out that arthropods are relatively and also the absolute values are in good agreement with the
small animals with surface-to-volume ratios higher than those previously published data based on feeding experiments with
in practically all vertebrates and that they are likely to reach redox dyes (Veivers et al. 1980). The most negative redox poten-
equilibrium with their environment unless efficient permeability tials are found in the regions of high hydrogen partial pressure,
barriers for oxygen are established or oxygen is sequestered by although in soil-feeding termites of the genus Cubitermes,
the animal or by intestinal microorganisms. In a series of studies parameters other than oxygen or hydrogen partial pressure
employing oxygen microsensors, Brune and coworkers clarified seem to control the redox status of the intestinal contents
the situation (Brune et al. 1995a; Ebert and Brune 1997; (Kappler and Brune 2002). There is evidence for ferric iron
Schmitt-Wagner and Brune 1999) by demonstrating that the reduction in the gut of soil-feeding and wood-feeding species,
steep gradient in oxygen partial pressure between the oxic gut which may be a microbial process, possibly mediated by the
epithelium and the anoxic gut contents drives a continuous presence of humic acids or other phenolic polymers (Kappler
influx of oxygen into the hindgut (> Fig. 22.4). In all termites and Brune 2002; Vu et al. 2004).
investigated, oxygen penetrated 50–200 mm into the periphery of
the hindgut lumen, leaving only the central portion of the
dilated compartments anoxic (Ebert and Brune 1997; Brune Hydrogen Partial Pressure
1998; Kappler and Brune 1999; Schmitt-Wagner and Brune
1999). Despite the massive hydrogen production by the intestinal pro-
The maintenance of anoxia in the termite hindgut is not tozoa (Odelson and Breznak 1985b), the hydrogen emission
a trivial issue. Since the removal of oxygen in the gut periphery is rates of termites are relatively low (Odelson and Breznak 1983;
fueled by the fermentative processes in the hindgut lumen, there Ebert and Brune 1997; Sugimoto et al. 1998; Schmitt-Wagner
must be a lower size limit for arthropods with a symbiotic and Brune 1999). Originally, it had been assumed that the
digestion. However, even the smallest of all termites investigated situation in the intestinal tract of termites was similar to that
to date (Anoplotermes pacificus, Termitidae: Apicotermitinae) in other methanogenic habitats, where low hydrogen partial
seems to possess a symbiotic gut microbiota, although the spec- pressures result from a tight coupling between hydrogen-
trum of fermentation products in the hindgut differs from that producing and hydrogen-consuming processes (Breznak 1994;
of other termites (Bauer et al. 2000). Breznak and Brune 1994). However, hydrogen microsensor
Fine-scaled redox measurements with platinum microelec- measurements revealed that the situation in termites is quite
trodes in wood- and soil-feeding termites (Ebert and Brune different, giving rise to steep radial gradients of hydrogen toward
1997; Kappler and Brune 2002) have shown that the redox the gut epithelium and enormous differences in hydrogen
potential in the gut mirrors the oxygen gradients (> Fig. 22.5), partial pressure along the gut axis (Ebert and Brune 1997;
a
C M Pa Co R
10 8
+400
Partial pressure (kPa) 8 6
Redox potential (mV)

+200
pH
6 4
0
4
–200
2 –400
0 –600
0 2 4 6 8
Length (mm)
b C M ms P1 P3 P4 P5
10
+400
8 12
Partial pressure (kPa)
Redox potential (mV)

+200
10
6
0
pH
8
4
–200
6
2 –400
4
0 –600
0 2 4 6 8 10 12 14 16 18
Length (mm)
. Fig. 22.5
Profiles of physicochemical conditions along the gut axis of Reticulitermes flavipes (a) and Cubitermes orthognathus (b). Oxygen (●) and
hydrogen (○) partial pressures, intestinal pH (■), and apparent redox potential (□) (against a standard hydrogen reference electrode
[SHE]) were measured with the respective microsensors. Guts were embedded in agarose-solidified Ringer’s solution. The borders
between the different gut regions (see legend to > Fig. 22.2) are indicated by the vertical lines (Data from Brune et al. (1995a),
Brune and Kühl (1996), Ebert and Brune (1997), Schmitt-Wagner and Brune (1999), and Kappler and Brune (2002))
Schmitt-Wagner and Brune 1999; Kappler and Brune 2002), and Anderson 1980; Veivers et al. 1980; Brune et al. 1995a).
which gave rise to the hypothesis that the spatial organization of Kovoor (1967) was the first to report an alkaline region in the
the hydrogen-producing and hydrogen-consuming populations anterior hindgut of wood-feeding termites; this observation was
controls the hydrogen partial pressure in different gut regions. later extended also to soil-feeding species (Bignell and Anderson
1980). Since then, a large body of data has accumulated (Bignell
1994; Bignell and Eggleton 1995), documenting a tendency
Intestinal pH toward strong alkalinity in the anterior hindgut of all higher ter-
mites except the Macrotermitinae (Anklin-Mühlemann et al. 1995).
The intestinal pH in the hindgut of most phylogenetically lower While the initial measurements (performed mostly by spot-
termites seems to be around neutral (Eutick et al. 1976; Bignell ting pooled, disrupted samples of individual gut regions on pH
indicator paper) still lacked accuracy and resolution, studies confirmed their close association with the mesenteric epithe-
with pH microsensors allowed alkaline regions to be precisely lium, suggesting that there is some kind of interaction with the
located (Brune et al. 1995a; Brune and Kühl 1996). The latter gut tissue (Tokuda et al. 2001).
measurements are not biased by homogenization and
documented that guts of soil-feeding termites are even more
alkaline than reported previously. The most alkaline values (pH Hindgut Cuticle
11–12.5) were found among soil-feeding Termitinae and repre-
sent the highest values ever encountered in biological systems Electron-microscopy studies revealed intimate associations of
(Brune and Kühl 1996). In all species tested, the pH of the gut microorganisms with the cuticle of the hindgut epithelium in
contents increases sharply from circumneutral in the midgut to all termites investigated (Breznak and Pankratz 1977; To et al.
highly alkaline between the midgut–hindgut junction and the 1980; Czolij et al. 1985; Yara et al. 1989; > Fig. 22.1). Bacteria are
first proctodeal dilation (P1), which coincides exactly with the associated with cup-like indentations on the epithelial surface of
location of the mixed segment (> Fig. 22.5), a morphologically the hindgut in Reticulitermes flavipes (Breznak and Pankratz
unique gut region present in all higher termites except the 1977). Although Mastotermes darwiniensis and Nasutitermes
Macrotermitinae (Noirot 2001). exitiosus possess similar structures, they seem not to be associ-
ated with microorganisms (Czolij et al. 1984). In certain soil-
feeding termites, cuticular spines protrude from the hindgut
Gut Compartmentation and Microhabitats wall into the lumen of the P4 compartment and form additional
attachment sites for the gut microbiota (Procubitermes
Each gut compartment provides various microhabitats differing aburiensis; Bignell et al. 1980b).
in many environmental parameters (see the section > ‘‘Physico-
chemical Gradients’’ in this chapter). The small size of the guts
results in large surface-to-volume ratios (Brune 1998), and the Hindgut Protozoa
epithelial surfaces provide ample attachment sites for gut micro-
organisms, which are thus protected from washout (Bignell The protozoa in the hindgut of lower termites occupy the bulk of
1984). Additional compartmentalization is created by the pro- the hindgut volume (Katzin and Kirby 1939) and represent an
tozoa inhabiting the hindgut lumen of lower termites. enormous surface area in the hindgut (Berchtold et al. 1999).
Pierantoni (1936) was first to point out the association of gut
flagellates with bacteria; since then, ectobiotic and
Midgut Epithelium, Ectoperitrophic Space endocytobiotic bacteria have been found on and in almost
every flagellate investigated. For example, the hypermastigote
As in other insects, the midgut epithelium is not protected by flagellate Joenia annectens, a symbiont in the hindgut of
a cuticle, but a peritrophic membrane separates the epithelial Kalotermes flavicollis, is densely colonized by prokaryotic micro-
surface from the digesta (Terra 1990). The ectoperitrophic space organisms (Hollande and Valentin 1969). The body is covered
harbors a distinct bacterial microbiota, which can be intimately with rod-shaped bacteria, and the nucleus and the cytoplasm
associated with the microvilli of the brush border (Breznak and contain various types of endocytobiotic bacteria (Radek et al.
Pankratz 1977). In addition, the so-called mixed segment in 1992; Patricolo et al. 2001). Also the oxymonadid flagellate
many Termitidae, a region where midgut and hindgut epithelia Streblomastix strix, a hindgut symbiont of Zootermopsis species,
overlap (Noirot 2001), is a microhabitat that harbors a specific is associated with several, morphologically distinct types of
bacterial microbiota. Kovoor (1968) described a ‘‘pure culture of bacteria that are orderly arranged end-to-end on six or seven
spore-forming fusiform bacteria’’ in the mixed segment of longitudinal vanes, lending S. strix a stellate appearance in
Microcerotermes edentatus, located outside of the peritrophic transverse section (Leander and Keeling 2004). Adhesion of
membrane in a posterior pocket formed by the mesenteric bacteria to the flagellate surfaces is based on different mecha-
side. Also, Potts and Hewitt (1973) observed a prominent flora nisms and facilitated by special surface structures (e.g., Radek
of ‘‘thin long filaments with terminal spores’’ in the mixed et al. 1996; Radek and Tischendorf 1999; Rother et al. 1999;
segment of the harvester termite, Trinervitermes trinervoides Patricolo et al. 2001). The literature has been recently reviewed
(Nasutitermitidae), located in the ectoperitrophic space poste- by Radek (1999). Possible significance of these associations of
rior to the Malpighian tubules. Other authors described dense prokaryotes with hindgut flagellates is discussed in a different
populations of different but also relatively uniform microorgan- section (see the section > ‘‘Interactions Between Prokaryotes
isms in the mixed segment of Nasutitermes exitiosus (Czolij et al. and Protozoa’’ in this chapter).
1985) and of soil-feeding Termitinae (Procubitermes aburiensis
and Cubitermes severus; Bignell et al. 1980a, 1983). Recently,
Tokuda et al. (2000) demonstrated that the bacteria populating Prokaryotic Gut Symbionts
the mixed segment of Nasutitermes takasagoensis are phyloge-
netically within the radiation of the Clostridiales (see the section In view of the variety of microhabitats and microenvironmental
> ‘‘Clostridiales’’ in this chapter). Electron microscopy conditions in the intestinal tracts of termites (see the section
> ‘‘The Gut Microenvironment’’ in this chapter), it is not aston- Ohkuma (2002) for a review). Most studies employed the 16S
ishing to find an equally large diversity among the microorgan- rRNA gene as a molecular marker. As in most other environ-
isms colonizing the gut. The amount of diversity indicated ments, phylogenetic diversity of the intestinal microbiota is
already by the morphological and ultrastructural features of enormous, and there is still little overlap between the phylotypes
the microbiota is greatly exceeded by that encountered at the recovered with cultivation-independent techniques and the iso-
phylogenetic level. lates obtained by cultivation (see the section > ‘‘Isolates and
Major Metabolic Activities’’ in this chapter). Molecular finger-
printing has been used to compare the structure of gut commu-
Morphological Diversity nities and to follow temporal changes. The bias inherent in all
polymerase chain reaction (PCR)-based approaches has been
Already the observation of gut preparations with a phase-contrast addressed by backing the results with independent methods,
light microscope reveals a wide variety of prokaryotic life forms. such as fluorescence in situ hybridization (FISH).
Several comprehensive studies of the bacterial gut microbiota of
termites using transmission electron microscopy have provided
detailed accounts of the morphological diversity of gut microor- Bacterial Diversity
ganisms for several termite species from different families. In
addition to the abundant protozoan fauna in all so-called lower König and coworkers were among the first to use the 16S rRNA-
termites (Yamin 1979), at least 20–30 different bacterial based approach to identify the phylogenetic position of
morphotypes have been distinguished among the microorgan- uncultivated spirochetes in the gut of Mastotermes darwiniensis
isms colonizing the intestinal tract of Reticulitermes flavipes and (Berchtold et al. 1994; Berchtold and König 1996); a parallel
Coptotermes formosanus (Rhinotermitidae; Breznak and study of Paster et al. (1996) was aimed at characterizing the
Pankratz 1977) and Pterotermes occidentis (Kalotermitidae; spirochetes in the gut of Nasutitermes lujae. At about the same
To et al. 1980). time, Ohkuma and coworkers attempted to characterize the full
The so-called higher termites (Termitidae) lack intestinal diversity of archaea and bacteria in the intestinal tract of
flagellates, but the morphology of their prokaryotic microbiota Reticulitermes speratus and Cryptotermes domesticus using
appears to be equally diverse. The hindgut of the wood-feeding a similar strategy (Ohkuma et al. 1995; Ohkuma and Kudo
Nasutitermes exitiosus (Termitidae: Nasutitermitinae) contains 1996, 1998).
almost 30 different bacterial morphotypes (Czolij et al. 1985), Although the cultivation-independent approach
and also the hindgut microbiota of the fungus-growing termite documented the presence of many new, hitherto uncultivated
Odontotermes formosanus (Termitidae: Macrotermitinae) phylotypes in the intestinal tracts of termites (for a review, see
comprises at least 20 different morphotypes (Yara et al. 1989). Kudo et al. 1998), these early studies lacked resolution since only
Numerous bacterial morphotypes, including many filamentous small numbers of clones were investigated. Later studies
forms, colonize the intestinal epithelia and the ectoperitrophic documented that diversity coverage of the clone libraries was
space of the soil-feeding termite Procubitermes aburiensis far from exhaustive, even if larger numbers of clones were used.
(Termitidae: Termitinae; Bignell et al. 1980a). The most comprehensive assessment of molecular diversity and
Although the morphological features usually do not allow bacterial community structure in termite guts to date involved
the affiliation of a bacterium to a specific taxon, members of the the gut microbiota of Reticulitermes species. Hongoh et al. ana-
termite gut microbiota have conspicuous forms or other mor- lyzed 14 clone libraries (96 clones each) of the bacterial 16S
phological features that are of (albeit limited) taxonomic rRNA genes in the hindgut of the Japanese termite species
value or conspicuous forms that seem to occur in different Reticulitermes speratus to characterize phylogenetic diversity
species of termites. One example is the thin, spore-forming and to address the bias introduced by different primer combi-
filaments described by Leidy (1881) as ‘‘Arthromitus’’ species, nations and PCR conditions (Hongoh et al. 2003a, b). Yang et al.
which occur in many invertebrates, including termites (Leidy (2005) performed a similar analysis with more than 500 clones
1881; Margulis et al. 1990). Also, many of the spirochetal forms from the European termite species Reticulitermes santonensis,
are so large and conspicuous that they can be morphologically focusing on the differences between the bacterial communities
distinguished (Breznak 1984). On the basis of the detailed morin the four major intestinal habitats: the midgut, the wall of the
phological features visualized by transmission electron micros- hindgut paunch, the hindgut fluid, and the intestinal protozoa
copy, Margulis and coworkers (Bermudes et al. 1988; Wier et al. (see the section > ‘‘Gut Compartmentation and Microhabitats’’
2000) proposed a number of new species for the larger in this chapter).
spirochetes. The intestinal community of the two Reticulitermes species is
quite similar, comprising representatives of several bacterial
Phylogenetic Diversity and Community phyla (> Fig. 22.6). Both termite species harbor Gram-positive
Structure bacteria (mainly clostridia, streptococci, and Mycoplasmatales-
related clones), members of the Bacteroidetes, spirochetes,
In the recent years, the microbiota in the intestinal tracts of and a number of Proteobacteria, albeit at slightly different
termites has been investigated also with molecular tools (see ratios. A large number of clones fall into the so-called termite
Reticulitermes speratus Cubitermes orthognathus
Bacteroidetes Clostridiales Bacieroidetes
TG-1 phylum Spirochaetes
Others
Others Proteobacteria
Clostridiales
Spirochaetes
. Fig. 22.6
Relative abundance of the major bacterial phyla in clone libraries of 16S rRNA genes from the hindgut of the wood-feeding, lower
termite Reticulitermes speratus and the soil-feeding, higher termite Cubitermes orthognathus (Data from Hongoh et al. (2003a) and
Schmitt-Wagner et al. (2003b))
group 1 (TG-1) phylum, which were most abundant in Spirochaetes

Reticulitermes santonensis (Yang et al. 2005); spirochetal clones
were less abundant in this termite but accounted for approxi- Not astonishingly, in view of the numerical predominance of
mately half of the analyzed clones in Reticulitermes speratus this morphologically diverse and conspicuous group in most
(Hongoh et al. 2003b). wood-feeding termites (Breznak 2002), the majority of the
The situation in soil-feeding termites is quite different. In clones obtained with bacteria-specific primers from the hindgut
a study analyzing bacterial diversity in the highly compartmen- of Reticulitermes speratus represent spirochetes (Hongoh et al.
talized intestinal tract of Cubitermes orthognathus (Schmitt- 2003a). Already the first molecular studies had indicated that
Wagner et al. 2003a, b), the authors combined clone analysis termite gut spirochetes represent a lineage phylogenetically dis-
with FISH and a molecular fingerprinting analysis, which not tinct from other Spirochaetes (Berchtold and König 1996;
only substantiated the data obtained by the individual Ohkuma and Kudo 1996; Paster et al. 1996). Better diversity
approaches but also allowed differences in community structure coverage was achieved by Lilburn et al. (1999), who targeted the
between the different gut compartments to be investigated. In intestinal spirochetes of Reticulitermes flavipes with spirochete-
contrast to the situation in Reticulitermes speratus, the bacterial specific primers. They demonstrated that the 12–15 spirochete
clone libraries contained no clones from the TG-1 phylum and morphotypes in Reticulitermes flavipes (Breznak and Pankratz
only few spirochetal clones. In the anterior gut sections, most 1977) were paralleled by 21 different spirochete phylotypes, which
clones represented Firmicutes. In the posterior gut sections, formed two major clusters of treponemes, one of them containing
clones belonging to the Bacteroidetes and different subgroups only clones of termite origin (Lilburn et al. 1999). Treponema-
of the Proteobacteria gained some numerical significance. related clones have also been recovered from a variety of other
A study of the bacterial microbiota in the P1 compartment of termite species (Lilburn et al. 1999; Ohkuma et al. 1999a). In
several higher termites extended the presence of a compartment- several cases, the ectosymbiotic association of certain phylotypes
specific microbiota and a predominance of Firmicutes in the with flagellate protozoa has been documented using FISH with
highly alkaline gut regions also to representatives of other feed- group-specific oligonucleotide probes (Berchtold and König
ing guilds (Thongaram et al. 2005). 1996; Iida et al. 2000; Noda et al. 2003).
All PCR bias notwithstanding, these large datasets,
together with the numerous clones obtained from other ter-
mite species (for references, see Ohkuma 2003), allow Endomicrobia
a reasonably accurate picture of the dominant phylogenetic
groups to be drawn. Most clones obtained in the different studies In their first survey of the bacterial diversity in Reticulitermes
represent lineages of microorganisms that were exclusively recov- speratus, Ohkuma and Kudo (1996) obtained a number of
ered from the intestinal tract of termites. The termite specificity of clones whose sequences were only distantly related to other
these lineages was underscored by the finding that the closest bacteria and which were subsequently recognized as a novel
relatives of the bacterial clones within each lineage were usually bacterial phylum (Hugenholtz et al. 1998). Also many clones in
derived also from the most-closely related termites, supporting the comprehensive libraries subsequently obtained with
the concept of coevolution between gut microbiota and host Reticulitermes speratus (Hongoh et al. 2003a) and Reticulitermes
(Yang et al. 2005). santonensis (Yang et al. 2005) belong to this lineage, indicating
that members of the TG-1 phylum represent a hitherto of such phylotypes in Reticulitermes species (Hongoh et al.
uncultivated but numerically dominant group of prokaryotes 2003a; Yang et al. 2005). Most clones are only distantly related
in the gut of Reticulitermes species. to described taxa and often form monophyletic clusters with
Using a full-cycle molecular approach, combined with trans- clones recovered from the gut of other termite species. While
mission electron microscopy, Stingl et al. (2005) showed that the some of the phylotypes seem to be associated with the hindgut
TG-1 bacteria in Reticulitermes species are endosymbionts that cuticle (Yang et al. 2005), others represent epibionts of protozoa
colonize – exclusively and in high abundance – the cytoplasm of (see the section > ‘‘Interactions Between Prokaryotes and
the larger flagellate species. The symbionts were specific for their Protozoa’’ in this chapter).
respective host flagellate and were provisionally classified in the
candidate genus ‘‘Endomicrobium.’’ Members of the TG-1 phy-
lum, for which the name ‘‘Endomicrobia’’ has been proposed, Other Groups
are phylogenetically quite diverse and seem to be present in and
also restricted to the guts of those insects (lower termites and Clone libraries of Reticulitermes species contained clones related
wood-feeding cockroaches of the genus Cryptocercus) that are in to the Mycoplasmatales in a distinct and apparently termite-
mutualistic association with such cellulose-fermenting flagel- specific lineage (Hongoh et al. 2003a; Yang et al. 2005) that were
lates (Stingl et al. 2005). abundant in the protozoan fraction of Reticulitermes santonensis
(Yang et al. 2005) and comprised also a clone obtained from
a symbiont of the termite gut flagellate Koruga bonita from
Firmicutes Mastotermes darwiniensis by single-cell PCR (Fröhlich and
König 1999). Among the Lactobacillales, most clones were affil-
In clone libraries of lower termites, clones affiliated with the iated with the genus Streptococcus and were mainly from the
Clostridia are abundant and fall into apparently termite-specific midgut clone library. Most clones affiliated with the
lineages (Hongoh et al. 2003a; Yang et al. 2005). In the higher Proteobacteria formed distinct, termite-specific lineages in
termite Cubitermes orthognathus, they dominated the clone the a-subgroup (only distantly related to other lineages of the
library of the alkaline hindgut sections (Schmitt-Wagner et al. Rickettsiales) or in the b-subgroup (most closely related to
2003b), which was confirmed using FISH with cluster-specific Dechlorimonas agitatus or to fermenting bacteria of the genus
probes and supported by molecular fingerprints of the different Propionivibrio; Brune et al. 2002). Clones belonging to the
gut compartments (Schmitt-Wagner et al. 2003a). d-subgroup were rare but virtually identical to the sequences
One of the clostridial clusters from the Cubitermes of Desulfovibrio desulfuricans and of a sulfate-reducing isolate
orthognathus clone libraries falls into the Clostridium from Reticulitermes santonensis (Kuhnigk et al. 1996).
propionicum group (Schmitt-Wagner et al. 2003b). Interestingly, Only a single clone among the >100 clones retrieved from
this cluster comprises also clones from the termite Nasutitermes the hindgut of the soil-feeding termite Cubitermes orthognathus
takasagoensis, which were localized in the mixed segment was affiliated with the Planctomycetales (Schmitt-Wagner et al.
between the midgut epithelium and the peritrophic membrane 2003b). However, a large fraction of the cells in the posterior
using FISH (Tokuda et al. 2000). Other clones are affiliated with hindgut of the closely related Cubitermes ugandensis hybridized
homoacetogenic isolates (see the section > ‘‘Homoacetogenic with a mixture of FISH probes targeting this phylum. This severe
Bacteria’’ in this chapter). underestimation of this phylum in the clone libraries is probably
Clone libraries and molecular fingerprints indicated that caused by the inadequacy of the commonly used Bacteria-
Clostridia dominate also the bacterial microbiota in the most specific PCR primers to amplify the 16S rRNA genes of
alkaline hindgut compartment (P1) of Termes comis, planctomycetes (Derakshani et al. 2001) and underlines the
Pericapritermes latignathus, and a Microcerotermes species (all importance of backing the results of PCR-based analyses with
Termitinae), whereas Bacilli dominate the P1 of a Speculitermes an independent method. Although FISH analysis indicates that
species (Apicotermitinae) (Thongaram et al. 2005). Many of the more than one-third of the bacteria in the second hindgut
clones derived from the P1 region form phylogenetic clusters compartment (P3 segment) of Cubitermes ugandensis may be
that are unique to termites and are often related to clones planctomycetes (Schmitt-Wagner et al. 2003b), their metabolic
obtained from the other insects with alkaline digestive tracts, function remains obscure.
which suggests that they represent lineages of alkaliphilic bacte-
ria (Schmitt-Wanger et al. 2003; Thongaram et al. 2005).
Archaeal Diversity
Bacteroidetes Molecular phylogenetic profiling of the microbial communities

by dot-blot hybridization with domain-specific probes has indi-
Clones affiliated with the Bacteroidetes were recovered from cated that archaea represent 0.1–2.6 % of small subunit (SSU)
the guts of numerous termite species (Ohkuma et al. 2002; rRNA extracted from the guts of 24 nutritionally and taxonom-
Schmitt-Wagner et al. 2003b). There is an enormous diversity ically diverse termite species (Brauman et al. 2001).
Interestingly, the relative abundance of archaea seems to be between the methanoarchaeal communities of congeneric ter-
related to the host diet. The percentage of archaeal 16S rRNA mites are substantial, and many clones obtained from the intes-
among prokaryotic 16S rRNA in the gut of soil-feeding termite tinal tract of termites cluster with clones retrieved from other
species (1.4–3.1 %) was significantly higher than in wood- insects. However, a purely vertical transmission of the
feeding and litter-feeding termite species (0.1–1.7 %). This is methanogenic gut microbiota is not supported (Donovan et al.
in agreement with the methane emission rates, which are gen- 2004).
erally higher among soil-feeding termite species (Brauman et al. Methanobrevibacter spp. in Reticulitermes are associated with
1992), and it has been speculated that the majority of the archaea the hindgut wall (Leadbetter and Breznak 1996; Leadbetter et al.
in termite guts are methanogens (Brauman et al. 2001). 1998) and, in the gut of Reticulitermes speratus and
Hodotermopsis sjoestedti, attached to the flagellated protist spe-
cies Dinenympha and Microjoenia (Tokura et al. 2000): there are
Methanogenic Archaea indications that the lineages attached to the flagellates are phy-
logenetically different from those associated with the gut epithe-
Partial sequences of the genes encoding for 16S rRNA and for lium. Fröhlich and König (1999) retrieved single cells of
subunit A of the methyl coenzyme M reductase (mcrA) of endosymbiotic methanogens from the anaerobic flagellate
methanogens indicated that the methanogens in Reticulitermes Pentatrichomonoides scroa occurring in the hindgut of
speratus belong to the order Methanobacteriales (Ohkuma et al. Mastotermes darwiniensis that were affiliated with the genus
1995) and are closely related but not identical to the Methanobrevibacter.
Methanobrevibacter species isolated from Reticulitermes flavipes Methanogens are among the few groups of organisms for
(Leadbetter and Breznak 1996; Leadbetter et al. 1998; also, which one can infer metabolic information from the 16S rRNA
see the subsection > ‘‘Methanogenic Archaea’’ in section gene sequence. FISH with an archaea-specific probe revealed
> ‘‘Hydrogen Metabolism’’). Later studies concentrated on the that archaea are largely restricted to the gut sections P3 and P4
16S rRNA genes and confirmed the presence of Methanobac- in Cubitermes ugandensis, which is in agreement with the distri-
teriales in Reticulitermes speratus (Shinzato et al. 1999), bution of F420-fluorescent cells and methane emission rates
Cryptotermes domesticus (Ohkuma and Kudo 1998; Shinzato along the gut axis of Cubitermes species (Schmitt-Wagner and
et al. 2001), Hodotermopsis sjöstedti (Ohkuma et al. 1999c), Brune 1999). Cells hybridizing with the archaea-specific probe
Neotermes koshunensis, Reticulitermes kanmonensis, Coptotermes presented 1.6 % and 3.8 % of the DAPI-stained cells in the P3
formosanus, and Mastotermes darwiniensis (Shinzato et al. 2001). and P4 section, respectively (Schmitt-Wagner et al. 2003b), but
All sequences cluster within the radiation of the genus since many of the abundant and morphologically diverse F420-
Methanobrevibacter, but the sequences from termites differ fluorescent microorganisms in these gut sections were filamen-
from those of known methanogens, forming unique lineages tous forms and appeared to be fragmented or destroyed during
in the phylogenetic trees. A single clone related to Methanomi- homogenization, they were likely underestimated by the FISH
crobiales was recovered from Reticulitermes speratus (Shinzato analysis.
et al. 1999).
Dot-blot hybridization indicated that Methanobacteriales
constitute one-third to more than one-half of the archaea in Non-methanogenic Archaea
the guts of almost all termite species studied (Brauman et al.
2001). By contrast, Methanosarcinales seem to be present only in In a dot-blot analyses of many termite species, the total com-
the guts of about half of the termite species, apparently forming bined value of the subgroup-specific probes was much lower
the dominant group of methanogens in 4 of the 24 species than that of the Archaea domain probe, indicating that termite
studied and accounting for the total archaeal signal in the guts may contain (possibly non-methanogenic) archaeal
fungus-growing species Macrotermes subhyalinus (Brauman populations whose 16S-like rRNAs do not hybridize with probes
et al. 2001). Additionally, the clones retrieved from the guts of for methanoarchaeal subgroups employed in this study
the phylogenetically higher termites Nasutitermes takasagoensis, (Brauman et al. 2001).
Odontotermes formosanus, and Pericapritermes nitobei clustered Shinzato et al. (1999) provided the first evidence for the
mostly among the Methanomicrobiales and Methanosarcinales presence of Thermoplasmales in the intestinal tract of
(Ohkuma et al. 1999c). In a detailed study of archaeal diversity Reticulitermes speratus. Friedrich et al. (2001) obtained numer-
in the gut of the soil-feeding higher termite, Cubitermes ous clones of Thermoplasmales and a few clones related to the
orthognathus, most archaeal clones were affiliated with Thermococcales from the gut of the soil-feeding termite
Methanobacteriales, Methanomicrobiales, and Cubitermes orthognathus and also documented for the first
Methanosarcinales, and a few clones had their closest relatives time the presence of Crenarchaeota in an intestinal tract.
among the Methanococcales (Friedrich et al. 2001). Similar Donovan et al. (2004) retrieved several clones related to the
results were obtained in a study with Cubitermes fungifaber, haloalkaliphile genus Natronococcus from Cubitermes fungifaber,
which also corroborated that there is little overlap between the which is quite intriguing in view of the extreme alkalinity of
communities of methanoarchaea present in the gut and in the the anterior hindgut of Cubitermes species (Brune and
food soil (Donovan et al. 2004). In contrast, the similarities Kühl 1996).
Spatial Organization Isolates and Major Metabolic Activities
As mentioned above, termite guts are axially and radially The major metabolic activities of the gut microbiota have been
structured, providing numerous microhabitats with different outlined (Breznak 2000; Slaytor 2000; Tholen and Brune 2000;
physicochemical microenvironments (see the section > Fig. 22.7), but there are still considerable gaps in our knowl-
> ‘‘The Gut Microenvironment’’ in this chapter). Not astonish- edge that underline the need for a refined concept (Tholen and
ingly, therefore, the distribution of gut microbiota is not Brune 2000). Numerous attempts have been made to character-
random but seems to be spatially organized. Detailed ize the prokaryotic gut microbiota of termites by isolating
descriptions of the spatial arrangement of the intestinal microorganisms that either were numerically abundant or pos-
prokaryotes in situ (Breznak and Pankratz 1977; To et al. 1978, sessed a metabolic potential considered important in the metab-
1980; Czolij et al. 1985; Yara et al. 1989), together with olism of the hindgut. Many of these efforts have yielded results
numerous other observations of certain morphotypes in partic- of uncertain significance, either because the methods were not
ular regions of the gut (see the section > ‘‘Gut Compartmenta- fully described or because no quantitation of bacteria was made
tion and Microhabitats’’ in this chapter), indicate that relative to the total number of microorganisms present. As there is
many microhabitats harbor characteristic microbial little overlap between the existing isolates and the 16S rRNA genes
populations. obtained in the molecular studies (see the section > ‘‘Phyloge-
Until recently, most of such evidence was based purely on netic Diversity and Community Structure’’ in this chapter), it is
morphological data, and the line of evidence is far from apparent that many of the microorganisms responsible for the
complete. With the advent of molecular tools, however, it major metabolic activities remain to be cultivated.
became possible to address not only the diversity of the termite
gut microbiota but also the spatial distribution of individual
phylotypes or phylogenetic groups. A study employing Numerically Predominant Isolates
whole-cell hybridization in homogenates of different gut regions
and in situ hybridization of gut cryosections with group- Most cultivable heterotrophic bacteria in the hindgut of
specific oligonucleotide probes provided the first results Reticulitermes flavipes are Streptococcus and Enterococcus species,
documenting differences in microbial community structure followed by Bacteroides species and representatives of the
between different regions of the hindgut of Mastotermes Enterobacteriaceae, mostly Citrobacter species and Enterobacter
darwiniensis at the level of phylogenetically defined microbial cloacae (Schultz and Breznak 1978; Tholen et al. 1997). Coccoid
groups (Berchtold et al. 1999). Other studies used FISH to lactic acid bacteria also dominated among the isolates obtained
document the association of gut protozoa with certain from the hindguts of the lower termites Mastotermes
phylotypes of hitherto uncultivated spirochetes (Berchtold and darwiniensis and Cryptotermes primus (Eutick et al. 1978a) and
König 1996; Iida et al. 2000; Noda et al. 2003), Bacteroidetes the higher termites Nasutitermes arborum, Thoracotermes
(Wenzel et al. 2003; Stingl et al. 2004), members of the TG-1 macrothorax, and Anoplotermes pacificus (Bauer et al. 2000).
phylum (Stingl et al. 2005), or methanogenic archaea (Tokura While Enterobacter species were found to dominate among the
et al. 2000). isolates from the rhinotermitid species Heterotermes ferox,
Also, PCR-based approaches allow one to resolve differences Coptotermes acinaciformis, Coptotermes lacteus, and Schedorhi-
in the community structure of different gut regions or gut notermes intermedius (Eutick et al. 1978a), most isolates from
compartments (Friedrich et al. 2001; Schmitt-Wagner et al. termitid species Nasutitermes exitiosus, Nasutitermes graveolus,
2003b; Yang et al. 2005). Costs and effort involved in sequencing and Nasutitermes walkeri were staphylococci. Isolates from these
and phylogenetic analysis limit investigations based on clone genera have been recovered also in earlier studies (e.g.,
analysis, but molecular fingerprinting allows expansion of the Mannesmann and Piechowski 1989).
investigation of diversity and – observing the necessary Although some of these studies were at best semiquantitative
cautions inherent to all PCR-based techniques – community and many employed only aerobic techniques, the pattern of
structure to include a larger number of samples. Terminal- bacterial species cultivated from each host species is remarkably
restriction-fragment-length polymorphism (T-RFLP) analysis constant. Most importantly, the majority of the isolates obtained
in the higher, soil-feeding termite Cubitermes orthognathus from termite guts are either aerobes or aerotolerant anaerobes.
documented that the different archaeal and bacterial The absence of obligate anaerobes among the isolates in those
populations are not randomly distributed along the gut studies that did not attempt to apply a methodology appropriate
and that the prokaryotic communities in the individual for the successful cultivation of such bacteria is not astonishing.
gut segments differ considerably with respect to diversity However, even in the studies that explicitly used the Hungate
and abundance (Friedrich et al. 2001; Schmitt-Wagner et al. technique and employed reduced media for cultivation, the
2003a). By contrast, the bacterial community structure fraction of obligate anaerobes was always smaller than that of
in homologous compartments in three different species aerotolerant anaerobes and aerobes (Schultz and Breznak 1978;
of Cubitermes was quite similar, indicating the existence Tholen et al. 1997). It is not clear whether this phenomenon is
of gut-segment-specific communities (Schmitt-Wagner caused by the oxygen status of the termite gut (see the section
et al. 2003a). > ‘‘Redox Conditions and Oxygen Status’’ in this chapter) or the
Wood of nine species of termites indicate an even larger cultivation bias

polysaccharides
(Eutick et al. 1978a).
When the isolates obtained in these studies are compared to
1 the results of the cultivation-independent characterization (see
the section > ‘‘Phylogenetic Diversity and Community Struc-
ture’’ in this chapter), there are enormous discrepancies between
the frequencies of the phylogenetic groups dominating the clone
Sugars libraries (> Fig. 22.6) and the species recovered by cultivation.
Nevertheless, many isolates are unique to the termite gut habitat,
2
and their characterization has provided valuable information on
? metabolic properties and other physiological features relevant
for the colonization of this particular habitat (> Table 22.1).
Lactate To increase cultivation efficiency, it will be necessary to develop
(Succinate) new cultivation strategies that take into account the special
(Ethanol) environmental conditions within the gut, in particular, the
steep physicochemical gradients (see the section > ‘‘The Gut
Microenvironment’’ in this chapter) and the metabolic interac-
tions among the microbiota (see the section > ‘‘Microbe–
3 Microbe Interactions’’ in this chapter).
Like many arthropods, termites harbor filamentous bacterial
morphotypes with refractile inclusions resembling endospores.
Acetate
(Propionate)
H2, CO2 These bacteria are usually attached to the hindgut wall
(Butyrate) 4
(Formate) (> Fig. 22.8) and were first described in 1849 as ‘‘Arthromitus’’
by Leidy (1849, 1881). The filaments have not been cultivated,
5 but Margulis et al. (1998) have proposed that they represent
6
a different life stage of aerobic, rod-shaped bacteria closely
related to Bacillus cereus, a species group that occurs ubiqui-
tously in soil. However, their conclusions were based merely on
CO2 CH4 the isolation of such bacteria from the boiled intestines of ten
species of soil arthropods containing ‘‘Arthromitus’’-like
. Fig. 22.7
filaments and on earlier reports on the isolation of B. cereus
Schematic presentation of the metabolic processes involved in
from arthropod guts. A phylogenetic identity of the filaments
the fermentative degradation of polysaccharides in the hindgut of
with the isolates has not been confirmed with molecular
Reticulitermes flavipes. The dashed lines indicate metabolic fluxes
methods. In this context, it should be noted that other authors
which seem to be strongly influenced by the continuous influx of
had previously identified the segmented filamentous bacteria in
oxygen into the gut periphery. Line thickness indicates the relative
the gut of mice, rats, and chickens as a new lineage of
importance of the process. The major metabolic groups are gut
Clostridiales, based on a 16S rRNA gene sequence analysis
flagellates (1), primary (2) and secondary (3) fermenting bacteria,
(Snel et al. 1994). The same group has proposed ‘‘Candidatus
homoacetogens (4), and methanogens (5); it remains to be
Arthromitus’’ as a provisional generic name for the segmented
clarified whether the flagellates are also a major source of lactate
filamentous bacteria falling into this lineage (Snel et al. 1995) –
(?). Short-chain fatty acids are oxidized by the host (6) (Adapted
unfortunately without verifying whether they are indeed related
from Brune 2003)
to the morphologically similar filaments in arthropods origi-
nally described by Leidy.
strong cultivation bias against certain groups of bacteria (see the

section > ‘‘Cultivation Bias’’ in this chapter). Lignocellulose Degradation
There are numerous reports on the presence in termite guts of

Cultivation Bias enzyme activities acting on different cellulose and hemicellulose
preparations. The enzyme activities in foregut and midgut are
A comparison of the viable counts of heterotrophic bacteria to most likely of host origin or are ingested fungal cellulases that
the direct microscopic counts of the microorganisms in the remain active within the gut (see the section > ‘‘Fiber Degrada-
hindgut of Reticulitermes flavipes indicates that about 90 % of tion’’ in this chapter), whereas most activities in the hindgut are
the microbial cells have escaped cultivation (Schultz and probably due to the microbiota. In the lower termites, it is
Breznak 1978; Tholen et al. 1997). Viable counts obtained in also important to differentiate between protozoan and bacterial
a similar study attempting to characterize the major gut bacteria origin.
. Table 22.1
Described species of prokaryotes unique to the intestinal tract of termites
Group species Termite speciesa Unusual feature References

Firmicutes
Acetonema longum Pterotermes occidentis (K) Homoacetogenic Kane and Breznak (1991)
Bacillus oleronius Reticulitermes santonensis (R) Degrades aromatic Kuhnigk et al. (1995)
compounds
Clostridium mayombei Cubitermes speciosus (T) Homoacetogenic Kane et al. (1991)
Clostridium termitidis Nasutitermes lujae (T) Cellulolytic Hethener et al. (1992)
Isoptericola variabilis (formerly Mastotermes darwiniensis (M) Cellulolytic Bakalidou et al. (2002),
Cellulosimicrobium variabile) Stackebrandt et al. (2004)
Sporobacter termitidis Nasutitermes lujae (T) Homoacetogenic Grech-Mora et al. (1996)
Sporomusa aerivorans Thoracotermes macrothorax (T) Homoacetogenic Boga and Brune (2003)
Sporomusa termitida Nasutitermes nigriceps (T) Homoacetogenic Breznak et al. (1988)
Sporotomaculum hydroxybenzoicum Cubitermes speciosus (T) Degrades aromatic Brauman et al. (1998)
compounds
Proteobacteria
Desulfovibrio intestinalis Mastotermes darwiniensis (M) Sulfate-reducing Fröhlich et al. (1999)
Desulfovibrio termitidis Heterotermes indicola (R) Sulfate-reducing Trinkerl et al. (1990)
Fusobacteria
Sebaldella termitidis (formerly Bacteroides Reticulitermes lucifugus (R) Uricolytic Collins and Shah (1986)
termitidis)
Bacteroidetes
Candidatus Vestibaculum illigatumb Neotermes cubanus (K) Epibiont of Stingl et al. (2004)
Staurojoenina sp.
Spirochaetes
Treponema azotonutricium Zootermopsis angusticollis (Z) Nitrogen-fixing Graber et al. (2004)
Treponema primitia Zootermopsis angusticollis (Z) Homoacetogenic Graber et al. (2004)
‘‘Endomicrobia’’
Candidatus Endomicrobium Reticulitermes santonensis (R) Endobiont of Stingl et al. (2005)
trichonymphaeb Trichonympha agilis
Candidatus Endomicrobium Reticulitermes santonensis (R) Endobiont of Stingl et al. (2005)
pyrsonymphaeb Pyrsonympha vertens
Methanobacteriales
Methanobrevibacter curvatus Reticulitermes flavipes (R) Methanogenic Leadbetter and Breznak (1996)
Methanobrevibacter cuticularis Reticulitermes flavipes (R) Methanogenic Leadbetter and Breznak (1996)
Methanobrevibacter filiformis Reticulitermes flavipes (R) Methanogenic Leadbetter et al. (1998)
a
Termites belong to the families Kalotermitidae K, Mastotermitidae M, Rhinotermitidae R, Termitidae T, and Termopsidae Z
b
Candidatus taxon: not cultivated but well characterized with respect to morphology, ultrastructure, phylogeny, and specific location
Cellulolytic Bacteria of debate (see the section > ‘‘Fiber Degradation’’ in this
chapter).
Cellulolytic prokaryotes have been isolated from the guts of Interestingly, successful attempts to isolate cellulolytic bac-
lower and higher termites on numerous occasions (for refer- teria usually employed oxic cultivation conditions; only a few
ences, see Breznak (1975), Breznak and Brune (1994), and anaerobic strains have been reported. Hungate isolated an anaer-
Wenzel et al. (2002)). However, many of these efforts either obic actinomycete, ‘‘Micromonospora propionici,’’ from
were unsuccessful or have yielded positive results of uncertain Amitermes minimus (Hungate 1946). Clostridium termitidis
significance, either because the methods were not fully described was isolated from the gut of the wood-feeding termite
or because the population size had not been established (Schultz Nasutitermes lujae (Hethener et al. 1992). All attempts to isolate
and Breznak 1978). Therefore, the contribution of bacteria to anaerobic cellulolytic bacteria from the gut of Reticulitermes
cellulose degradation in the termite gut has always been a matter flavipes were negative (Schultz and Breznak 1978). No
various other termites (Bakalidou et al. 2002) and form signif-

icant populations also in the gut of other insects (Cazemier et al.
2003).
Cellulolytic activity has been detected also among numerous
actinomycetes isolated from the gut of several soil-feeding
Termitidae (Pasti and Belli 1985); some of the isolates were
also lignolytic (Pasti et al. 1990). Although an unusual associa-
tion of soil-feeding termites (Termitidae, Termitinae) with acti-
nomycete-like bacteria has been documented (Bignell et al.
1979, 1980b) and facultatively aerobic actinomycetes have been
isolated from the gut, parent soil, and mound materials of the
termites Procubitermes aburiensis and Cubitermes severus
(Bignell et al. 1991), their significance in the degradation of
lignocellulose remains to be established.
. Fig. 22.8 Hemicellulolytic Bacteria

Phase-contrast photomicrograph of the hindgut wall of
Reticulitermes santonensis, colonized by Pyrsonympha flagellates Compared to the information on cellulose degradation, our
and ‘‘Arthromitus’’-like filaments (arrows) (Reprinted from Yang understanding of the degradation of the hemicellulose compo-
et al. 2005) nent of lignocellulose in termite guts is rather meager. Xylanase
activity has been observed in midgut and hindgut of several
termites, including wood-feeding, soil-feeding, and fungus-
cellulose-degrading bacteria were present among the numeri- cultivating species (Breznak and Brune 1994; Rouland 2000).
cally predominant isolates recovered from nine species of ter- In the fungus-cultivating Macrotermes bellicosus, there is evi-
mites representing all major families, using both aerobic and dence that the source of xylanase activity may be a symbiotic
anaerobic techniques (Eutick et al. 1978a). No anaerobic cellu- fungus (Matoub and Rouland 1995), but analogous to the host
lolytic bacteria were isolated from the soil-feeding termite cellulases, a host origin of the midgut xylanases remains to be
Cubitermes speciosus using serial dilutions of gut homogenates scrutinized.
and the Hungate technique (Brauman et al. 1990b). In lower termites, the gut flagellates also seem to play a major
By contrast, numerous studies have yielded aerobic bacteria, role in the degradation of xylan (see the section > ‘‘Role of
albeit of often dubious numerical significance. Frequent isolates, Intestinal Protozoa’’ in this chapter). However, also bacteria
found in numerous studies, were assigned to Serratia marcescens and yeasts might be involved in the metabolism of hemicellu-
and Bacillus cereus (e.g., Thayer 1976). Strains of both species, loses. Aerobic and facultatively anaerobic hemicellulose-
isolated from Reticulitermes hesperus, formed soluble, and in the degrading bacteria and yeasts were isolated from the guts of
case of Bacillus cereus, also cell-bound, carboxymethylcellulases several wood-feeding termites, with xylan-degrading bacteria
(Thayer 1978). Serratia marcescens seems to be a common but (106–107 per ml) dominating in Mastotermes darwiniensis and
minor inhabitant of the intestinal tract of insects that on occa- xylan-degrading yeasts (107–5 108 cells per ml) in
sion can become pathogenic (see the section > ‘‘Pathogens’’ in Zootermopsis angusticollis and Neotermes castaneus (Schäfer
this chapter). et al. 1996). Gram-positive isolates belonged to the genera
On the basis of the few studies where the number or density Bacillus and Paenibacillus or to the Actinobacteria,
of cellulolytic bacteria in the gut had been determined (e.g., Paul while Gram-negative strains were affiliated with the genera
et al. 1993; Wenzel et al. 2002), it appears that bacteria are not Pseudomonas, Acinetobacter, and Ochrobactrum or with the
very relevant for cellulose digestion. The most abundant among enterobacteria.
23 groups of cellulolytic isolates from Zootermopsis angusticollis Several strains of alkaliphilic bacteria, which were isolated
were closely affiliated with Bacillus megaterium, Bacillus cereus, from the extremely alkaline P1 compartment the soil-feeding
or Paenibacillus polymyxa (Wenzel et al. 2002). Many of the Sinocapritermes mushae and Amitermes longignathus and repre-
cellulolytic isolates are species known to occur also in soil and sent a novel lineage of Paenibacillus, express alkalitolerant
other habitats and may represent transient permanent xylanase activity with a high pH optimum (Ohkuma et al.
populations of microorganisms ingested with the food, while 2003; Thongaram et al. 2003). An unusual xylanase, distantly
others are apparently specific for the intestinal tracts of insects related to xylanases of bacteria and fungi colonizing the bovine
(> Table 22.1). rumen (GHF11), has been discovered in a genomic library to
Isoptericola variabilis (formerly Cellulosimicrobium variabile; microbial DNA extracted from the intestinal tract of an
Stackebrandt et al. 2004) and other closely related bacteria were unspecified higher termite belonging to the subfamily
isolated from the hindgut of Mastotermes darwiniensis and Nasutiermitinae (Brennan et al. 2004).
Lignin-Degrading Bacteria Lignin-solubilizing actinomycetes have been isolated from

the gut of soil-feeding termites (Pasti et al. 1990), and all isolates
The degradation of lignin involves the ‘‘enzymatic combustion’’ were Streptomyces strains. Moreover, in lower termites, the only
of the highly recalcitrant polyphenolic resin by peroxidases. isolate decolorizing Remazol Brilliant Blue and Azure B was
Despite a few claims to the contrary, there is currently no a Streptomyces strain (Kuhnigk and König 1997). Since isolation
conclusive evidence that bacteria lacking extracellular and phylogenetic characterization of actinomycetes from ter-
peroxidative activity solubilize or degrade polymeric core lignin mite guts indicated that the actinomycete flora of termites
significantly, and lignin degradation by the gut microbiota has depends largely on their geographical origin (Watanabe et al.
been questioned (Breznak and Brune 1994; Kuhnigk et al. 1994). 2003), it is not clear whether such isolates simply represent
A 13C-nuclear magnetic resonance (NMR) analysis of the spores ingested with the food.
feces of the wood-feeding termites Microcerotermes parvus and
Coptotermes formosanus indicated a preferential loss of
polysaccharide during gut passage, whereas lignin accumulated Oxygen Reduction
and was modified only when its O-aromatic-C and O-methyl-C
components were modified (Hopkins et al. 1998; Hyodo The bacteria and protozoa colonizing the gut periphery, espe-
et al. 1999). cially those directly associated with the gut epithelium
However, extensive delignification occurs in the fungal gar- (> Fig. 22.8), are exposed to the continuous influx of oxygen
dens of fungus-cultivating termites (Hyodo et al. 2000) and has (Brune et al. 2000). Oxygen microprofiles indicate that the
to be expected also in any wood colonized by lignolytic fungi. anoxic status of the termite hindguts must be attributed to the
Many lignin-derived aromatic compounds (representing the oxygen consumption of the microbiota located in the gut
major subunits and lignin–carbohydrate linkages found in lig- periphery (see the section > ‘‘Redox Conditions and Oxygen
nins and their depolymerization products) can be degraded by Status’’ in this chapter), which can represent a substantial frac-
aerobic bacteria, and numerous bacteria degrading lignin tion of the respiratory activity of the host (Brune et al. 1995a).
monomers or certain dimeric lignin model compounds have This conclusion is supported by the predominance of faculta-
been isolated from various wood-feeding termites (Kuhnigk tively or even obligately aerobic isolates in all cultivation-based
et al. 1994; Kuhnigk and König 1997; Harazono et al. 2003). studies.
They comprised a wide range of Gram-positive bacteria and However, there are several indications that not all oxygen-
Proteobacteria, with strict aerobes dominating in Nasutitermes consuming activities in termite guts are of a respiratory nature.
nigriceps and enterobacteria in the lower termites. An aerobic In the extremely alkaline gut regions, the high rates of oxygen
isolate able to degrade hydroxybenzoic acids was described as consumption might be partly attributable also to chemical reac-
a new species, Bacillus oleronius (Kuhnigk et al. 1995). tions, such as the autoxidation of phenolic residues in lignin or
All attempts to enumerate bacteria degrading aromatic com- humic substances (see Kappler and Brune 1999). On the other
pounds indicated that termite guts contain significant numbers hand, radiotracer analysis of the in situ metabolism in the
of aerobic bacteria capable of degrading aromatic rings, whereas hindgut of Reticulitermes flavipes demonstrated that the high
anaerobic degradation of the aromatic nucleus appears to be of oxygen fluxes also significantly influence fermentation processes
little significance (Brune et al. 1995b; Kuhnigk and König 1997). in the hindgut (Tholen and Brune 2000). This is supported by
Under anoxic conditions, only ring and side-chain modification the oxygen reduction potential found in obligate anaerobes
seem to be of importance (Kuhnigk et al. 1994; Brune et al. such as lactic acid bacteria (see the section Lactic Acid Bacteria
1995b). Some isolates from termite guts involved in anaerobic in this chapter), homoacetogenic bacteria (see the section
modification or degradation of aromatic compounds were novel > ‘‘Homoacetogenic Bacteria’’ in this chapter), and sulfate-
taxa (> Table 22.1). reducing bacteria (see the section > ‘‘Sulfate-reducing Bacteria’’
The digestibility of lignocellulose is improved if the in this chapter).
phenylpropanoic acid residues esterified to the hemicellulose
chains and the diferulic ester linkages between the hemicellulose
chains in plant cell walls are hydrolyzed, since this increases the Microbial Fermentations
solubility of the macromolecules and reduces steric hindrance of
hydrolytic enzymes (Jeffries 1994). A strain of Clostridium In lower termites, the bulk of the fermentative activity should be
xylanolyticum (producing hydroxycinnamoyl esterases) that caused by the intestinal protozoa, which immediately phagocy-
hydrolyzes and then hydrogenates ferulic and p-coumaric acid tize the wood particles entering the hindgut. The identity of the
residues has been isolated from the gut of the grass-feeding substrates of the prokaryotic microbiota is far from clear. Since
termite Tumulitermes pastinator (McSweeney et al. 1999). Sev- most of the prokaryotic symbionts probably do not participate
eral termites have been shown to possess also activities cleaving in fiber degradation, presumably they ferment either soluble
non-phenolic b-O-4 type-lignin model compounds products entering the hindgut from the midgut or intermediates
representing the major linkage within the lignin macromolecule released by the protozoa (> Fig. 22.7). The composition of
(Hirai et al. 2000). short-chain fatty acids and other fermentation products in the
hindgut fluid of termites indicates that a variety of microbial first exception, since they were represented also in a clone library
fermentations occur simultaneously (Schultz and Breznak 1979; obtained from the P1 compartment (Thongaram et al. 2005).
Odelson and Breznak 1983; Anklin-Mühlemann et al. 1995;
Tholen and Brune 2000). However, the exact nature and amount
of the monomeric substrates entering the hindgut are not Bacteroidetes
known, and the spectrum of fermentation products formed by
the protozoa might be much larger than initially assumed Besides the coccoid lactic acid bacteria, a relatively large propor-
(Tholen and Brune 2000). tion of the isolates from the hindgut of Reticulitermes flavipes
The apparent absence of pyruvate dehydrogenase activity in have been identified as Bacteroides species (Schultz and Breznak
termite tissues (O’Brien and Breznak 1984) and several other 1978). The strains fall into two groups: aeroduric anaerobes
observations prompted the suggestion that the termite releases fermenting lactate to propionate and acetate (Schultz and
pyruvate into the hindgut to be converted to acetate and Breznak 1979) and anaerobic strains forming butyrate and
reabsorbed by the host (Slaytor et al. 1997; Slaytor 2000). How- isobutyrate as characteristic products from complex media
ever, recent findings documented that termite mitochondria (Schultz and Breznak 1978). The only described Bacteroides
possess a pyruvate dehydrogenase complex, although the differ- species isolated from termite guts, Bacteroides termitidis (Sebald)
ence in activity suggests differences in pyruvate metabolism from Reticulitermes lucifugus, was phylogenetically misplaced
between lower and higher termites (Itakura et al. 1999, 2003). (Paster et al. 1985) and recently reclassified among the
Fusobacteria in the genus Sebaldella (Collins and Shah 1986).
A recent isolate, obtained from the hindgut of Reticulitermes
Lactobacillales speratus by dilution plating of gut suspensions, has a 16S rRNA
gene sequence identical to that of a group of clones recovered
Coccoid lactic acid bacteria predominate among the carbohy- exclusively from termite guts, representing a lineage of bacteria
drate-utilizing bacteria isolated from the hindguts of termites of within the radiation of the Bacteroides subgroup but only dis-
the families Mastotermitidae, Kalotermitidae, and tantly related to the genus Bacteroides (Ohkuma et al. 2002).
Rhinotermitidae (Eutick et al. 1978a; Schultz and Breznak The uricolytic activity of Bacteroides isolates from termite
1978; Tholen et al. 1997; Bauer et al. 2000) and have been guts (Potrikus and Breznak 1980) suggests a potential role in
isolated also from several representatives of the Termitidae nitrogen cycling (see the section > ‘‘Nitrogen Recycling’’ in this
(Eutick et al. 1978a; Bauer et al. 2000). The most abundant chapter). Many of the clones from termite guts clustering among
isolates obtained from the hindguts of Reticulitermes flavipes the Bacteroidetes fall into a lineage comprising epibionts of
and Thoracotermes macrothorax belong to the genera Enterococ- protozoa (see the section > ‘‘Interactions Between Prokaryotes
cus and Lactococcus (Bauer et al. 2000). However, the low fre- and Protozoa’’ in this chapter).
quency of such clones in cultivation-independent studies
indicates that lactic acid bacteria might be present only in
moderate numbers (Hongoh et al. 2003a; Schmitt-Wagner Hydrogen Metabolism
et al. 2003b; Yang et al. 2005).
Physiological characterization of the isolates has revealed The enormous accumulation of hydrogen at the gut center and
high potential rates of O2 reduction in the presence of ferment- the steep radial hydrogen gradients in the gut periphery of
able substrates (Tholen et al. 1997; Bauer et al. 2000), which several termites (Ebert and Brune 1997; Schmitt-Wagner and
might represent an adaptation to variable oxygen tensions and Brune 1999) indicate that molecular hydrogen is a key interme-
could explain why lactococci and enterococci are regularly diate in the microbial food chain. Hydrogen-dependent CO2
encountered in the intestinal tracts of termites and other insects reduction by methanogens and homoacetogens is probably the
and possibly restricted to specific compartments of the intestinal most important hydrogen sink in termite hindguts. To under-
tract (Yang et al. 2005). stand the metabolism in termite guts, it is important to identify
the hydrogen-producing and hydrogen-consuming populations
and their functional interactions (see the section > ‘‘Microbe–
Clostridiales Microbe Interactions’’ in this chapter).
Compared to their abundance in bacterial clone libraries derived

from termite guts, clostridial isolates are underrepresented in Hydrogen-Producing Microorganisms
cultivation-based studies. Several isolates have been described as
new species and seem to be unique to termite guts (> Table 22.1), On the basis of the few pure culture studies available, it is
but their numerical significance is either low or has not been generally assumed that the polysaccharides of the wood particles
established. Bacteria distantly related to Clostridium oroticum, taken up by the gut flagellates are oxidized to acetate and CO2,
detected in an alkaline enrichment culture derived from gut and the reducing equivalents are released as H2 (see Breznak and
homogenates of Pericapritermes latignathus, may prove to be Brune (1994) and Brune and Stingl (2005); > Fig. 22.7). This is
in agreement with the enormous accumulation of hydrogen in lituseburense group. A very unusual homoacetogen
the hindgut of Reticulitermes flavipes (Ebert and Brune 1997; (Sporobacter termitidis, isolated from Nasutitermes lujae) clus-
> Fig. 22.4). In view of the wide variety of clostridial clones ters in the Clostridium leptum group together with numerous
retrieved from Reticulitermes species (Hongoh et al. 2003a; clones of 16S rRNA genes retrieved from the anterior hindgut of
Yang et al. 2005), bacterial fermentations may also contribute Cubitermes orthognathus (Schmitt-Wagner et al. 2003b). It grows
to the intestinal H2 production. Such assumptions are substan- exclusively by the disproportionation of methyl groups derived
tiated by the considerable amounts of H2 produced in certain from methoxylated aromatic compounds but not on H2 + CO2
gut regions of soil-feeding Cubitermes species (Schmitt-Wagner or other typical substrates of homoacetogens, and it methylates
and Brune 1999), a group of termites that lack gut flagellates but sulfide and cysteine if these compounds are present in the
contain a similar assemblage of clostridial clones (Schmitt- medium (Grech-Mora et al. 1996).
Wagner et al. 2003b). The cellulolytic bacterium Clostridium Recently, numerous spirochetal strains have been isolated
termitidis isolated from Nasutitermes lujae (Hethener et al. from the gut of Zootermopsis angusticollis (Leadbetter et al.
1992) is affiliated with a number of clones from Cubitermes 1999). They belong to the Treponema branch of Spirochaetes
orthognathus guts (Schmitt-Wagner et al. 2003b), but nothing that contains mainly sequences obtained from termite guts
is known about the physiology of the bacteria represented by (Lilburn et al. 1999), and many are capable of H2-dependent
these 16S rRNA genes. acetogenesis from CO2 (Leadbetter et al. 1999). One of the
homoacetogenic isolates, which has been described as a new
species, Treponema primitia (Graber et al. 2004), utilizes a wide
Homoacetogenic Bacteria range of substrates and is capable of mixotrophic growth, i.e.,
the simultaneous utilization of H2 and organic substrates
Reductive acetogenesis from H2 and CO2 in termite gut homog- (Graber and Breznak 2004). These represent important clues
enates was first demonstrated by Breznak and Switzer (1986). In for the unknown metabolic function of the enormous
the following years, the presence of homoacetogenic activities populations of spirochetes colonizing the hindgut lumen of
was established for a large number of termite species from all termites and the surfaces of many intestinal protozoa (Breznak
major feeding guilds, including representatives of wood-feeding, 2002).
fungus-cultivating, and soil-feeding termites (Breznak and Kane Generally, there seems to be a strong cultivation bias against
1990; Brauman et al. 1992; Williams et al. 1994). Although homoacetogens from termite guts, and their numerical abun-
reductive acetogenesis in gut homogenates of soil-feeding ter- dance and contribution to reductive acetogenesis in situ are not
mites was always outcompeted as a hydrogen sink by clear. Recently, sequences clustering with the formyltetrahy-
methanogenesis (Brauman et al. 1992; Breznak 1994), microin- drofolate synthase (FTHFS) homologues of termite gut
jection of H14CO3– into intact hindguts of soil-feeding spirochetes were found to dominate the diversity of genes in
Cubitermes spp. has identified a high potential for the hindgut of Zootermopsis angusticollis coding for FTHFS
reductive acetogenesis (Tholen and Brune 1999), which indi- (Salmassi and Leadbetter 2003). FTHFS is a key enzyme in
cates that the contribution of reductive acetogenesis to the reductive acetogenesis, which makes the FTHFS gene a good
overall electron flow in the guts of soil-feeding termites functional marker for homoacetogenic bacteria (Leaphart
might be larger than expected. Possibly, part of the explanation and Lovell 2001; Leaphart et al. 2003). Quite likely
lies in the ability of homoacetogens to grow mixotrophically on therefore homoacetogenic spirochetes are responsible for
H2 and other substrates (Breznak and Switzer Blum 1991; the large potential activities of H2-dependent acetogenesis
Breznak 1994). encountered in most wood-feeding termites (Breznak 1994,
Although reductive acetogenesis has been demonstrated to 2002).
occur in a large number of termite species from all major feeding All investigated homoacetogens isolated from termite guts
guilds (Brauman et al. 1992), only seven species of catalyzed hydrogen-dependent oxygen reduction (Boga and
homoacetogenic bacteria from termites have been described to Brune 2003); the isolate Sporomusa aerivorans has the highest
date (> Table 22.1). Sporomusa termitida (isolated from known oxygen-reducing capacity among obligately anaerobic
Nasutitermes nigriceps; Breznak et al. 1988), Sporomusa bacteria other than sulfate-reducing bacteria of the genus
aerivorans (isolated from Thoracotermes macrothorax; Boga Desulfovibrio (see the section > ‘‘Sulfate-Reducing Bacteria’’ in
et al. 2003), and Acetonema longum (isolated from Pterotermes this chapter), which possibly indicates an adaptation to
occidentis; Kane and Breznak 1991) are three spore-forming the oxygen status of the gut environment. Although the
representatives in the Sporomusa group of Gram-positive bacte- location of homoacetogens relative to the oxygen gradient
ria that are characterized by a Gram-negative cell wall. All iso- remains to be established, it is rather unlikely that the
lates are capable of H2-dependent reduction of CO2 to acetate homoacetogens in Reticulitermes flavipes contribute substan-
and possess a large potential for hydrogen-dependent oxygen tially to the removal of oxygen diffusing into the gut: in contrast
reduction (Boga and Brune 2003). to methanogenesis, the in situ rates of reductive acetogenesis in
Clostridium mayombei has been isolated from Cubitermes this termite species are not affected by oxygen (Tholen and
speciosus (Kane et al. 1991) and belongs to the Clostridium Brune 2000).
Already the location in the microoxic gut periphery and the

presence of catalase activity in Methanobrevibacter cuticularis
indicated a considerable oxygen tolerance (Leadbetter and
Breznak 1996). This was substantiated by the finding that –
despite a general inability among Methanobacteriales to synthe-
size heme – the wetwood isolate Methanobrevibacter arboriphilus
expresses a catalase when the medium is supplemented with
hemin (Shima et al. 2001). The documentation of a F420H2
oxidase in M. arboriphilus (Seedorf et al. 2004) finally provided
also a biochemical basis for the high rates of H2-dependent O2
reduction exhibited by Methanobrevibacter species colonizing
the periphery of the termite hindgut (A. Tholen and A. Brune,
unpublished results) and their apparent ability to cope with the
continuous influx of oxygen (Leadbetter and Breznak 1996;
. Fig. 22.9 Ebert and Brune 1997).
Termites entrapped in a block of copal (young amber) from the
Andean uplift region of Boyaca Province, Colombia (Pleistocene).
The photograph shows gas bubbles around the termites, Sulfate-Reducing Bacteria
presumably methane that escaped from the body after the
termites had been engulfed by the resin Sulfate-reducing bacteria have been isolated from the intestinal
tracts of many different termite species (Brauman et al. 1990a, b;
Trinkerl et al. 1990; Kuhnigk et al. 1996; Fröhlich et al. 1999). All
isolates are members of the genus Desulfovibrio and seem to
Methanogenic Archaea form substantial populations in the gut of certain termites.
The relevance of sulfate reduction under in situ conditions is
Methane formation by termites, first documented by Breznak not clear, but the sulfate-reducing bacteria in termite guts might
(Breznak 1975; Odelson and Breznak 1983), has received con- partake also in the removal of oxygen or play a role in syntrophic
siderable attention owing to its implication for the global meth- fermentations (Kuhnigk et al. 1996). Like other Desulfovibrio
ane budget (Sanderson 1996). Methane formation is restricted species isolated from sediments (for references, see Cypionka
to the archaea; therefore, the methane production by all termites 2000), all strains isolated from termite guts show extremely high
investigated indicates that methanogenic archaea are present in rates of O2 reduction in the presence of H2 (Kuhnigk et al. 1996;
virtually all termites (Brauman et al. 1992). Furthermore, the gas Fröhlich et al. 1999).
bubbles around termites entrapped in amber (> Fig. 22.9),
which have been attributed to continuing methanogenesis in
the fresh resin, are paleontological evidence for the presence of Nitrogen Transformations
methanogenic archaea in termite guts.
The autofluorescence of coenzyme F420 allows easy visuali- The diet of the termites ranges from sound wood to lignocellu-
zation of methanogens by epifluorescence microscopy. They are losic plant materials in various stages of humification, including
either located free in the hindgut lumen, attached to the hindgut soil and animal dung. Owing to the high C-to-N ratio of sound
cuticle, or associated with other microorganisms (Breznak and wood, many xylophagous termites are strongly limited by nitro-
Brune 1994). In Zootermopsis, methanogens are almost exclu- gen (Collins 1983). They conserve nitrogen by recycling –
sively associated with the gut flagellates (Lee et al. 1987). On the a strategy that has been termed ‘‘carbon elimination’’ (Higashi
basis of their F420-like autofluorescence, filamentous bacteria et al. 1992). Another way to improve the C/N balance of the
colonizing the cuticular spines in the fourth proctodeal segment colony is through nitrogen fixation.
of soil-feeding termites (Bignell et al. 1980a) also appear to be This section will summarize the present knowledge on these
methanogens (Schmitt-Wagner et al. 2003b). processes and the microorganisms involved. For details, the
Despite their relative abundance and phylogenetic diversity reader is referred to the comprehensive review of Breznak
in termite guts (see the section > ‘‘Archaeal Diversity’’ in this (2000).
chapter), to date, only a few methanogens from Reticulitermes
flavipes have been isolated in pure culture. The isolates represent
new species in the genus Methanobrevibacter (> Table 22.1), Nitrogen Recycling
grow best with H2 and CO2, and form large populations that
are attached to the luminal side of the gut epithelium or adhere The best way to deal with a rare resource is conservation and
to other prokaryotes colonizing the hindgut wall (Leadbetter recycling. Termites, like other insects, secrete uric acid and urea,
and Breznak 1996; Leadbetter et al. 1998). the waste products of nucleic acid and protein metabolism, via
the Malpighian tubules into the intestinal tract (Terra 1990; content of a colony within a few years (Breznak 2000), and stable
> Fig. 22.3). Potrikus and Breznak (1981) demonstrated that isotope analysis has revealed that 30–60 % of the nitrogen in
termites lack uricase and that uric acid is recycled by anaerobic Neotermes koshunensis workers is derived via this pathway
bacteria in the hindgut of Reticulitermes flavipes, including (Tayasu et al. 1994).
Streptococcus and Citrobacter species and Bacteroides (now However, the nitrogen content of the diet increases with
Sebaldella) termitidis (Potrikus and Breznak 1980). humification of the organic matter, and peptides, amino sugars,
Despite the low nitrogen content of the diet, nitrogen and microbial biomass are potential sources of nutrition for soil-
recycling creates high ammonia concentrations in the hindgut feeding Cubitermes spp. (Ji et al. 2000; Ji and Brune 2001). The
of wood-feeding termites, which allows the maintenance of an natural abundance of the 15N isotope indicates that dietary
active gut microbiota and thus ensures high rates of carbon nitrogen and not atmospheric nitrogen fixation is an important
mineralization. The concentration of ammonia in the paunch nitrogen source in humivorous termite species (Tayasu et al.
of Nasutitermes walkeri is in the range of 3 mM (Slaytor and 1997; Tayasu 1998). The enormous ammonium concentrations
Chappell 1994). The efficient assimilation of ammonium into in feces and nests (constructed from feces) of soil-feeding ter-
microbial biomass (and possibly also resorption of ammonium mites (Ndiaye et al. 2004; R. Ji and A. Brune, unpublished
in the posterior hindgut) avoids the loss of nitrogen via the feces results) suggest that termites in this feeding guild are not nitro-
(Breznak 2000). The transformation to high-quality microbial gen limited.
protein also leads to an upgrading of any low-quality nitrogen in There are also many indications that the nitrogen fixation
the diet. rates cannot easily be extrapolated to the colony level or ecosys-
The nitrogen cycle is closed by the digestion of microbial tem level. Apart from methodological aspects, there are inter-
cells. Since termites cannot access the microbes in the hindgut specific differences, intraspecific variations, seasonal patterns,
directly, worker larvae solicit hindgut contents from their and effects of laboratory maintenance (Curtis and Waller 1995,
nestmates. This behavior, termed ‘‘proctodeal trophallaxis,’’ is 1998) to consider. Additionally, the reduced oxygen partial pres-
unique to this group of social insects and increases in frequency sure in subterraneous nests and galleries has to be taken into
with the level of nitrogen limitation (Machida et al. 2001). account (Curtis and Waller 1996). The problems are discussed in
Digestion of the hindgut contents and resorption of the nitrog- detail by Breznak (2000).
enous products probably take place in the foregut and midgut Dinitrogen-fixing bacteria isolated from termite guts include
(> Fig. 22.3), as indicated by lysozyme and protease activities in enterobacteria (French et al. 1976; Potrikus and Breznak 1977;
these gut regions (Fujita et al. 2001; Fujita and Abe 2002; Fujita Eutick et al. 1978a), sulfate-reducing bacteria (Kuhnigk et al.
2004). The efficiency of nitrogen conservation within the colony 1996), and spirochetes (Leadbetter et al. 1999; Graber et al.
is increased further by the consumption of exuviae and dead 2004). Molecular approaches, using the nifH gene as
nestmates. a functional marker, have revealed a much wider spectrum of
potentially N2-fixing microorganisms, including clostridia,
Proteobacteria, and methanogenic archaea, comprising several
Nitrogen Fixation sequence clusters unique to termites (Ohkuma et al. 1996,
1999b). However, a gene-expression study at the community
Though nitrogen cycling is efficient in termites, termite colony level has revealed that only a few nifH homologs, probably
growth is limited by the net nitrogen taken in with food representing alternative nitrogenases (anf genes), were actually
(Breznak 2000). Therefore, many termites show a preference transcribed under in situ conditions in the gut of Neotermes
for lignocellulosic substrates that are colonized by fungi and koshunensis (Noda et al. 1999). The preferential expression of
therefore have a decreased C-to-N ratio (Amburgey et al. 1980; anf genes might be related to an insufficient molybdenum (Mo)
Cornelius et al. 2002). Termites living on sound wood, however, supply with the lignocellulosic diet (Ohkuma 2002). The num-
rely on the exclusive capacity of their prokaryotic gut microbiota ber of unrelated nifH homologs in a single spirochete species and
to fix atmospheric nitrogen. the variety of apparently spirochete-related nifH homologs in
After the first convincing demonstrations of N2 fixation hindgut clone libraries indicate that spirochetes are potentially
in termites using the acetylene reduction assay (Benemann important nitrogen-fixing microorganisms in termite guts
1973; Breznak et al. 1973) and the incorporation of 15N (Lilburn et al. 2001).
into biomass by termites incubated under an 15N2-enriched
atmosphere (Bentley 1984), a large body of data on
nitrogen fixation in termite guts has accumulated. A few Symbiotic Interactions
aspects will be mentioned; for more details, the reader is
again referred to the excellent review of this subject by The enormous complexity of this subject emerges from the large
Breznak (2000). number of species present within the gut, the different organi-
Nitrogen fixation is widespread among termites, although zational levels of the potential partners, their metabolic capaci-
the rates differ considerably. Nitrogenase activity in certain ties and topological orientation, and the numerous possibilities
Nasutitermes spp. would be sufficient to double the nitrogen for metabolic interactions especially within anaerobic
communities. In addition, symbiotic interactions can be coexistence of different populations (Ebert and Brune 1997;
extremely specific with respect to the partners involved or may Tholen and Brune 2000).
simply consist of a cross-feeding of metabolites or nutrients Although reductive acetogenesis is outcompeted as
between populations that are not even in direct contact with a hydrogen sink by methanogenesis in gut homogenates of
each other. soil-feeding termites (Brauman et al. 1992), intact guts of soil-
feeding Cubitermes spp. show a large potential for reductive
acetogenesis (Tholen and Brune 1999). Taking into account
Microbe–Microbe Interactions the possibility of an intercompartmental transfer of H2
(Schmitt-Wagner and Brune 1999), as has been demonstrated
Trophic cascades and cross-feeding of metabolites can form between the midgut and hindgut compartments of cockroaches
a network of interactions between the individual populations (Lemke et al. 2001), the contribution of reductive acetogenesis
of any microbial community. In termite guts, the resulting to the overall electron flow in the guts of soil-feeding termites
uneven distribution of sources and sinks of microbial metabo- might be larger than expected.
lites within the system gives rise to metabolic gradients (see the A cross-feeding of hydrogen could also be the basis for the
section > ‘‘Physicochemical Gradients’’ in this chapter), which interaction between the prokaryotic epibionts on filamentous
are indicators of the location of microbial activities in situ. bacteria in the gut of Reticulitermes flavipes (Breznak and
Additionally, the spatial organization of different populations, Pankratz 1977), some of which seem to be methanogenic
including intimate associations between prokaryotes and pro- archaea (Leadbetter and Breznak 1996). Schultz and Breznak
tozoa, underlines the importance of studying the nature of the (1979) had demonstrated a cross-feeding of lactate between
respective interactions. Lactococcus lactis and a propionigenic Bacteroides species, to
consolidate the predominance of these species among the culti-
vable bacteria in the hindgut fluid of Reticulitermes flavipes with
Interactions Among Prokaryotes the absence of lactate accumulation. Microinjection of 14C-
labeled lactate confirmed that lactate is an important interme-
At least three metabolically different groups of microorganisms diate under in vivo conditions but raised several new questions
are involved in the metabolism of hydrogen in the gut of regarding the source of lactate and the role of oxygen in shifting
lower termites: the protozoa, which produce H2 as a product the product spectrum from propionate to acetate (Tholen and
of their fermentative metabolism, and methanogens and Brune 2000).
homoacetogens, which reduce CO2 to methane or acetate, An example for cross-feeding of nutrients has been provided
respectively (see the section > ‘‘Hydrogen Metabolism’’ in this by Graber and Breznak (2005), who made a convincing case that
chapter). It is generally assumed that hydrogenotrophic Treponema primitia, a homoacetogenic spirochete in the gut of
methanogens outcompete homoacetogens owing to their higher Zootermopsis angusticollis that requires folate for growth, bene-
affinity for the common substrate (Cord-Ruwisch et al. 1988). fits from the excretion of 5-formyltetrahydrofolate by other
Nevertheless, both metabolic groups occur simultaneously in community members (Lactococcus lactis and Serratia grimesii).
the hindgut of termites. In Reticulitermes flavipes, where
methanogens are mostly restricted to the gut periphery
(Leadbetter and Breznak 1996), in situ rates of reductive Interactions Between Prokaryotes and Protozoa
acetogenesis surpass those of methanogenesis considerably
(Tholen and Brune 2000), whereas in Zootermopsis angusticollis, Most of the flagellate protozoa in the guts of lower termites are
where methanogens are located mostly inside the gut protozoa associated with prokaryotic microorganisms, either as epibionts
(Lee et al. 1987), methanogenesis appears to be the major on their cell surface or as endobionts in the cytoplasm and in the
hydrogenotrophic process (Brauman et al. 1992). nucleus. The abundance and variety of such associations are
The methanogens in termites seem to be hydrogen limited in documented by numerous electron-microscopy studies (for
situ, as indicated by the stimulation of methane emission by references, see Honigberg (1970), Radek (1999), Inoue et al.
externally supplied H2 (Odelson and Breznak 1983; Messer and (2000), and Dolan (2001)), but virtually, nothing is known
Lee 1989; Ebert and Brune 1997; Schmitt-Wagner and Brune about the physiological basis of these associations, especially
1999). Also, the per cell rates of methanogenesis determined in the metabolic interactions (Brune and Stingl 2005).
vitro with Methanobrevibacter cuticularis, multiplied by the via- Cook (1932) was the first to observe the emission of a gas
ble counts of methanogens in Reticulitermes flavipes, are much other than CO2 by Termopsis nevadensis (syn. Zootermopsis
higher than the methane emission rate of living termites nevadensis) and speculated that intestinal protozoa could pro-
(Leadbetter and Breznak 1996). The relative rates of methane duce hydrogen, methane, or possibly a mixture of both. Many
and hydrogen emission by different termites vary considerably decades later, termites were shown to emit both hydrogen and
(Sugimoto et al. 1998), and evidence is accumulating that the methane (Breznak 1975; Odelson and Breznak 1983), and axenic
spatial organization of the hydrogen-producing and hydrogen- cultures of Trichomitopsis termopsidis (a gut flagellate from
consuming microorganisms determines the competition and Zootermopsis) were shown to produce only hydrogen after
a b
10 mm
. Fig. 22.10
Phase-contrast (a) and epifluorescence (b) photomicrographs of a small trichomonad in Schedorhinotermes lamanianus and its attached
epibionts, showing the typical F420-autofluorescence of methanogenic archaea (Photomicrographs taken by M. Pester)
methanogenic symbionts were eliminated by treating the cul- by specific attachment sites (> Fig. 22.11); it is affiliated with
tures with bromoethanesulfonate, an inhibitor of other uncultivated Bacteroidales.
methanogenesis (Odelson and Breznak 1985a, b). The devescovinid flagellate Caduceia versatilis (d’Ambrosio
Methanogenic symbionts of protozoa can be easily visual- et al. 1999) carries two different, rod-shaped and filamentous,
ized by epifluorescence microscopy (> Fig. 22.10). Lee et al. epibionts. In this case, the host is propelled by the self-
(1987) reported that methanogenic bacteria in the hindgut of synchronizing movement of the flagella of several thousand
Zootermopsis angusticollis are associated only with the rod-shaped bacteria, which are deeply embedded into its cell
small flagellated protozoa Trichomitopsis termopsidis, surface (Tamm 1982). The epibionts of Staurojoenina flagellates
Tricercomitus termopsidis, and Hexamastix termopsidis, whereas (> Fig. 22.12), recently assigned to the candidate taxon
they were not observed in the large, hypermastigid protozoa. On ‘‘Vestibaculum illigatum’’ (Stingl et al. 2004), have a similar
the basis of the results of various treatments that selectively morphology, although in this flagellate, motility is not due to
eliminated or affected certain protozoa from the gut of this the bacteria but to their own flagella. Nevertheless,
termite, Messer and Lee (1989) concluded that the large pro- ‘‘Vestibaculum illigatum’’ falls into the same cluster of
tozoa of the genus Trichonympha were the most important Bacteroidales as the rod-shaped epibiont of Mixotricha paradoxa
hydrogen source in the hindgut, whereas the methanogenic (Stingl et al. 2004), a lineage that also contains numerous clones
symbionts of Trichomitopsis termopsidis produced most of obtained from other termites (Ohkuma et al. 2002; Yang et al.
the methane. 2005).
The large gut flagellates are often colonized by epibiotic The larger flagellate species in the gut of Reticulitermes spe-
bacteria. The presence of special attachment sites on the cell cies harbor numerous prokaryotic endobionts within their cyto-
envelope of the flagellates (e.g., Tamm 1980; Radek et al. 1992, plasm (Bloodgood et al. 1974; Bloodgood and Fitzharris 1976).
1996; Dolan and Margulis 1997; Stingl et al. 2004) indicates The endobionts of Trichonympha and Pyrsonympha species in
a tight association. In some cases, the epibionts seem to be Reticulitermes santonensis (> Fig. 22.13) belong to the candidate
responsible for the locomotion of the host. phylum ‘‘Endomicrobia,’’ which seem to occur exclusively in
The polymastigote flagellate Mixotricha paradoxa, which termite gut flagellates (Stingl et al. 2005). To date, there is no
occurs exclusively in the gut of Mastotermes darwiniensis, uses indication of their possible function. An involvement in hydro-
its four flagella only for steering. The cell is propelled by the gen metabolism is unlikely since the oxymonadid Pyrsonympha
many thousands of spirochetes that cover most of the body species are phylogenetically distant from the hypermastigid fla-
surface. The spirochetes are attached to projecting brackets of gellates (Moriya et al. 2003; Stingl and Brune 2003) and do not
the cell surface in a manner that allows the helical movement of seem to possess hydrogenosomes (see Brune and Stingl 2005).
the individual cells to travel in metachronal waves along the cell
surface of the host, resulting in locomotion (Cleveland and
Grimstone 1964). The epibiotic spirochetes were recently iden- Microbe–Host Interactions
tified as members of the Treponema cluster by 16S rRNA gene
sequence analysis (Wenzel et al. 2003). In addition, a second, Although the host provides a favorable habitat for the intestinal
rod-shaped epibiont is intimately associated with the cell surface microbiota and thrives on their metabolic products, there is not
from the guts of Reticulitermes hesperus (Thayer 1976) and

Mastotermes darwiniensis (Kuhnigk et al. 1994), causes septice-
mia in Coptotermes formosanus (Osbrink et al. 2001). Serratia
marcescens seems to form part of the normal microflora of
insects since it can be isolated from both healthy and diseased
specimens. Usually nonpathogenic when present in the digestive
tract in small numbers, it multiplies rapidly once it enters the
hemocoel and causes death in 1–3 days [for references, see
Lysenko (1985) and Sikorowski and Lawrence (1998)]. Also,
the fungi associated with Reticulitermes flavipes include both
cellulolytic species and potential pathogens (Zoberi and Grace
1990). It is not clear whether entomopathogenic fungi form
a part of the natural gut community or are only found among
diseased insects (Rath 2000).
Intracellular Symbionts
Termites, like many other insects, are also associated with intra-
cellular prokaryotes that are vertically transmitted via the ova-
ries (Breznak 2004). Flavobacteria of the genus Blattabacterium
reside in specialized cells (bacteriocytes) of cockroaches and the
most primitive termite, Mastotermes darwiniensis (Bandi et al.
1995). The close phylogenetic relationship between endosymbi-
onts from Mastotermes darwiniensis and members of the wood-
feeding cockroach genus Cryptocercus supports the hypothesis
that termites evolved from subsocial, wood-dwelling cock-
roaches (Lo et al. 2003). All other termites examined carried
endocytoplasmic bacteria that are affiliated with the Wolbachia
. Fig. 22.11 group and are located in nonspecialized fat body cells (Bandi
Transmission electron micrograph (a) of the epibiotic bacteria on et al. 1997). Endonuclear bacteria have been observed in the
the cell surface of Mixotricha paradoxa, a large flagellate occurring trophocytes of Reticulitermes lucifugus and Kalotermes flavicollis
exclusively in the gut of Mastotermes darwiniensis that lives in (Grandi et al. 1997).
a motility symbiosis with prokaryotes. The schematic presentation
(b) illustrates the regular arrangement of the spirochetal (S) and
rod-shaped (B) epibionts and the special attachment brackets Mutualists or Commensals?
(br) at the cell surface (Reproduced from Cleveland and
Grimstone 1964) In the hindgut symbiosis, the host creates a rather constant
environment for its symbionts; provides a continuous supply
of substrates; and, by the transfer of the microorganisms to other
always a necessity or clear indication for any specific interac- nestmates, ensures their propagation within the ecosystem. In
tions. However, the intimate associations of bacteria with micro- the symbiosis between lower termites and fiber-digesting flagel-
villi of the midgut epithelium or with the epithelial cups in the lates, both partners are indispensable and the mutual advantage
hindgut cuticle (Breznak and Pankratz 1977) indicate a closer is obvious. However, in the case of most prokaryotic gut sym-
integration of certain bacteria with the host tissues. Another bionts, the symbionts appear to be of little advantage to the host.
example suggesting interactions with the gut tissue are the If the benefit of the association is unidirectional, a symbiont
clostridia-related bacteria located in the ectoperitrophic space would be classified as a commensal, and the host might even
between the midgut wall and the peritrophic membrane in the benefit from its elimination (Brune 2003). Unfortunately, it is
mixed segment, closely associated with the mesenteric epithe- not easy to eliminate specific prokaryotes from the intestinal
lium (Tokuda et al. 2000, 2001). microbial community of termites selectively, and it is difficult to
distinguish between the direct and indirect consequences of
their elimination. Removal of the spirochetes from the gut of
Pathogens Nasutitermes exitiosus by feeding metronidazole or exposing the
termites to pure oxygen kills the termites almost as rapidly as the
A few reports indicate that the gut microbiota of termites is not complete removal of all bacteria by antibiotics (Eutick et al.
always beneficial. Serratia marcescens, which has been isolated 1978b).
a b
f
f
s
10 µm 1 µm
c dl
om
im
0,3 µm
. Fig. 22.12
Scanning (a, b) and transmission (c) electron micrographs of Staurojoenina sp. from Neotermes cubanus. (a) Overview of a cell, showing
two of the four flagellar tufts (f), numerous bacterial rods (b), and occasional spirochetes (s) attached to the surface. (b) Close-up of the
cell surface, showing single spirochetes between the ectobiotic rods. Arrows point to early stages of cell division. (c) Cross-section of the
epibiotic rods. In addition to an inner membrane (im) and outer membrane (om), the cell is surrounded by a diffuse layer (dl). Electron-
dense material supports the plasma membrane of the flagellate below the attachment sites (arrowhead). The arrow points to
a phagocytized rod-shaped bacterium with remnants of attachment complex (Reprinted from Stingl et al. 2004)
a b
. Fig. 22.13
Transmission electron micrographs of ultrathin sections of Trichonympha agilis (a) and Pyrsonympha vertens (b) showing the
ultrastructure of the endosymbiotic bacteria in the candidate genus ‘‘Endomicrobium,’’ which are very abundant in the cytoplasm of
these flagellate species. Cells are surrounded by two membranes; the outermost membrane forms tube-like elongations at the tapered
cell poles (arrow in a). g glycogen. Bars = 0.2 mm (Reprinted from Stingl et al. 2005)
In view of the constant and massive inoculation with micro- occur exclusively in termites (Yang et al. 2005). The absence of
organisms in their diet, it is clear that termites cannot keep their a ‘‘normal gut microflora’’ would allow the uncontrolled prolif-
gut sterile. The molecular diversity studies have provided ample eration of ingested foreign microorganisms and increase the
evidence for a specific microbiota, composed of lineages that danger of colonization by potential pathogens. Serratia
. Table 22.2 proctodeal trophallaxis (Machida et al. 2001), a behavioral

Free energy of important reactions involved in symbiotic trait that was probably fundamental to both the establishment
digestiona of the gut microbial community and the evolution of sociality in
termites (Nalepa et al. 2001).
DG0 (kJ/ Relative
Reaction mol)b changec (%)
(1) Glucose + 6 O2 ! 6 CO2 + 6 H2O 2,872 100 Conclusions

(2) Glucose + 2 H2O ! 2 Acetate + 216 7.5
2 H+ + 2 CO2 + 4 H2 Termite guts are minute but efficient bioreactors for the conver-
(3) 4 H2 + 2 CO2 ! Acetate + H+ 95 3.3 sion of lignocellulose to short-chain fatty acids and microbial
+ 2 H2O biomass. However, termite guts are not simply anoxic fermen-
(4) 3 Acetate + 3 H+ + 6 O2 ! 6 CO2 + 2,561 89.2 tors but axially and radially structured environments with phys-
6 H2O icochemically distinct microhabitats, and we are just beginning
(5) 4 H2 + CO2 ! CH4 + 2 H2O 131 4.6 to understand the complex interactions within the intestinal

(6) 2 Acetate + 2 H + 4 O2 ! 4 CO2 +
+
1,707 59.5 microbial communities. Microbial diversity in the termite gut
4 H2O is enormous, and the existing isolates represent only a negligible
(7) CH4 + 2 O2 ! CO2 + 2 H2O 818 28.4 fraction of the untapped diversity of prokaryotes in the guts of
a
the more than 2,600 described species of termites
For explanations, see text
b (Kambhampati and Eggleton 2000).
Gibbs free energy under standard conditions at pH 7 is calculated after
Thauer et al. (1977)
c
Free energy change relative to the aerobic oxidation of glucose (Reaction 1)
References
marcescens failed to colonize the gut of normal Coptotermes Aanen DK, Eggleton P, Rouland-Lefevre C, Guldberg-Froslev T, Rosendahl S,
lacteus, but transient colonization occurred when the protozoa Boomsma JJ (2002) The evolution of fungus-growing termites
and spirochetes were killed by exposure to pure oxygen (Veivers and their mutualistic fungal symbionts. Proc Natl Acad Sci USA
et al. 1982). 99:14887–14892
Abo-Khatwa N (1978) Cellulase of fungus-growing termites: a new hypothesis on
Under oxygen-limited conditions, the energy loss caused by
its origin. Experientia 34:559–60
the activity of the gut microbiota is relatively small Amburgey TL, Johnson GN, Etheridge JL (1980) A method to mass-produce
(> Table 22.2). The hypothetical, nonsymbiotic oxidation of decayed-wood termite bait blocks. J Georgia Entomol Soc 16:112–115
cellulose to CO2 would allow the termite to exploit 100 % of Anklin-Mühlemann R, Bignell DE, Veivers PC, Leuthold RH, Slaytor M (1995)
the free energy of the reaction (1). A homoacetogenic degrada- Morphological, microbiological and biochemical studies of the gut flora in
the fungus-growing termite Macrotermes subhyalinus. J Insect Physiol
tion of one glucose equivalent to three acetate molecules by the
41:929–940
gut microbiota (Reactions 2, 3) releases only 10.8 % of the free Bakalidou A, Kämpfer P, Berchtold M, Kuhnigk T, Wenzel M, König H (2002)
energy contained in the substrate; the rest is still available for the Cellulosimicrobium variabile sp. nov., a cellulolytic bacterium from the hind-
host (Reaction 4), which also benefits from the nutritionally gut of the termite Mastotermes darwiniensis. Int J Syst Evol Microbiol
valuable microbial biomass. However, the situation is quite 52:1185–1192
Bandi C, Sironi M, Damiani G, Magrassi L, Nalepa CA, Laudani U, Sacchi
different if the H2 formed in the microbial fermentations is
L (1995) The establishment of intracellular symbiosis in an ancestor of
converted to methane (Reactions 2, 5). Although the amount cockroaches and termites. Proc R Soc Lond Ser B Biol Sci 259:293–299
of free energy released in methanogenesis is not much larger Bandi C, Sironi M, Nalepa CA, Corona S, Sacchi L (1997) Phylogenetically distant
than that released in reductive acetogenesis, the host receives intracellular symbionts in termites. Parassitologia 39:71–75
only two-thirds of the acetate available under homoacetogenic Bauer S, Tholen A, Overmann J, Brune A (2000) Characterization of abundance
and diversity of lactic acid bacteria in the hindgut of wood-and soil-feeding
conditions (Reaction 6); 28.4 % of the free energy in the sub-
termites by molecular and culture-dependent techniques. Arch Microbiol
strate remains in methane, which cannot be exploited by the 173:126–173
host (Reaction 7) and is lost to the environment. Béguin P, Aubert J-P (1994) The biological degradation of cellulose. FEMS
Such energetic considerations suggest an obvious advantage Microbiol Rev 13:25–58
if a termite uses reductive acetogenesis as the hydrogen- Benemann JR (1973) Nitrogen fixation in termites. Science 181:164–165
Bentley BL (1984) Nitrogen fixation in termites: fate of newly fixed nitrogen.
consuming process, and why especially the more highly evolved
J Insect Physiol 30:653–655
termites (Termitidae) show a tendency toward increasing meth- Berchtold M, König H (1996) Phylogenetic analysis and in-situ identification of
ane emission rates is still a mystery (Brauman et al. 1992; Bignell uncultivated spirochetes from the hindgut of the termite Mastotermes
et al. 1997). Nevertheless, considering the added value of meta- darwiniensis. Syst Appl Microbiol 19:66–73
bolic properties such as nitrogen fixation, ammonia assimila- Berchtold M, Ludwig W, Koenig H (1994) 16S rDNA sequence and phylogenetic
position of an uncultivated spirochete from the hindgut of the termite
tion, and provision with vitamins, the advantages for the host
Mastotermes darwiniensis Froggatt. FEMS Microbiol Lett 123:269–273
may be well worth the investment. The exploitation of the Berchtold M, Chatzinotas A, Schönhuber W, Brune A, Amann R, Hahn D,
nutritive resources contained in the microbial biomass requires König H (1999) Differential enumeration and in situ localization of
the digestion of the gut symbionts, which is realized by micro-organisms in the hindgut of the lower termite Mastotermes
darwiniensis by hybridization with rRNA-targeted probes. Arch Microbiol Brauman A, Müller JA, Garcia JL, Brune A, Schink B (1998) Fermentative
172:407–416 degradation of 3-hydroxybenzoate in pure culture by a novel strictly anaer-
Bermudes D, Chase D, Margulis L (1988) Morphology as a basis for taxonomy of obic bacterium, Sporotomaculum hydroxybenzoicum gen. nov., sp. nov. Int
large spirochetes symbiotic in wood-eating cockroaches and termites: J Syst Bacteriol 48:215–221
pillotina gen. nov., nom. rev.; Pillotina calotermitidis sp. nov., nom. rev.; Brauman A, Bignell DE, Tayasu I (2000) Soil-feeding termites: biology, microbial
diplocalyx gen. nov., nom. rev.; diplocalyx calotermitidis sp. nov., nom. rev.; associations and digestive mechanisms. In: Abe T, Bignell DE, Higashi
hollandina gen. nov., nom. rev.; Hollandina pterotermitidis sp. nov., nom. M (eds) Termites: evolution, sociality, symbiosis, ecology. Kluwer, Dor-
rev.; and Clevelandina reticulitermitidis gen. nov., sp. nov. Int J Syst Bacteriol drecht, pp 233–259
38:291–302 Brauman A, Dore J, Eggleton P, Bignell D, Breznak JA, Kane MD (2001) Molecular
Bignell DE (1984) The arthropod gut as an environment for microorganisms. In: phylogenetic profiling of prokaryotic communities in guts of termites with
Anderson JM, Rayner ADM, Walton DWH (eds) Invertebrate-microbial different feeding habits. FEMS Microbiol Ecol 35:27–36
interactions. Cambridge University Press, Cambridge, UK, pp 205–227 Brennan Y, Callen WN, Christoffersen L, Dupree P, Goubet F, Healey S,
Bignell DE (1994) Soil-feeding and gut morphology in higher termites. In: Hunt Hernandez M, Keller M, Li K, Palackal N, Sittenfeld A, Tamayo G, Wells S,
JH, Nalepa CA (eds) Nourishment and evolution in insect societies. Hazlewood GP, Mathur EJ, Short JM, Robertson DE, Steer BA
Westview Press, Boulder, pp 131–158 (2004) Unusual microbial xylanases from insect guts. Appl Environ
Bignell DE (2000) Introduction to symbiosis. In: Abe T, Bignell DE, Higashi Microbiol 70:3609–3617
M (eds) Termites: evolution, sociality, symbiosis, ecology. Kluwer, Dor- Breznak JA (1975) Symbiotic relationships between termites and their intestinal
drecht, pp 189–208 microbiota. Symp Soc Exp Biol 29:559–580
Bignell DE, Anderson JM (1980) Determination of pH and oxygen status in the Breznak JA (1984) Hindgut spirochetes of termites and Cryptocercus punctulatus.
guts of lower and higher termites. J Insect Physiol 26:183–188 In: Krieg NR, Holt JG (eds) Bergey’s manual of systematic bacteriolog, vol 1.
Bignell DE, Eggleton P (1995) On the elevated intestinal pH of higher termites Williams and Wilkins, Baltimore, pp 67–70
(Isoptera: Termitidae). Insect Soc 42:57–69 Breznak JA (1994) Acetogenesis from carbon dioxide in termite guts. In: Drake
Bignell DE, Oskarsson H, Anderson JM (1979) Association of actinomycete-like HL (ed) Acetogenesis. Chapman and Hall, New York, pp 303–330
bacteria with soil-feeding termites (Termitidae, Termitinae). Appl Environ Breznak JA (2000) Ecology of prokaryotic microbes in the guts of wood-
Microbiol 37:339–342 and litter-feeding termites. In: Abe T, Bignell DE, Higashi M (eds)
Bignell DE, Oskarsson H, Anderson JM (1980a) Distribution and abundance of Termites: evolution, sociality, symbiosis, ecology. Kluwer, Dordrecht,
bacteria in the gut of a soil-feeding termite Procubitermes aburiensis pp 209–231
(Termitidae, Termitinae). J Gen Microbiol 117:393–403 Breznak JA (2002) Phylogenetic diversity and physiology of termite gut spiro-
Bignell DE, Oskarsson H, Anderson JM (1980b) Specialization of the hindgut wall chetes. Integr Comp Biol 42:313–318
for the attachment of symbiotic microorganisms in a termite Procubitermes Breznak JA (2004) Invertebrates – insects. In: Bull AT (ed) Microbial biodiversity
aburiensis. Zoomorphology 96:103–112 and bioprospecting. ASM Press, Washington, DC, pp 191–203
Bignell DE, Oskarsson H, Anderson JM, Ineson P, Wood TG (1983) Structure, Breznak JA, Brune A (1994) Role of microorganisms in the digestion of lignocel-
microbial associations and function of the so-called ‘‘mixed segment’’ of the lulose by termites. Ann Rev Entomol 39:453–487
gut in two soil-feeding termites, Procubitermes aburiensis and Cubitermes Breznak JA, Kane MD (1990) Microbial H2/CO2 acetogenesis in animal guts:
severus (Termitidae, Termitinae). J Zool Lond 201:445–480 nature and nutritional significance. FEMS Microbiol Rev 87:309–314
Bignell DE, Anderson JM, Crosse R (1991) Isolation of facultatively aerobic Breznak JA, Pankratz HS (1977) In situ morphology of the gut microbiota of
actinomycetes from the gut, parent soil and mound materials of the termites wood-eating termites [Reticulitermes flavipes (Kollar) and Coptotermes
Procubitermes aburiensis and Cubitermes severus. FEMS Microbiol Ecol formosanus Shiraki]. Appl Environ Microbiol 33:406–426
85:151–160 Breznak JA, Switzer JM (1986) Acetate synthesis from H2 plus CO2 by termite gut
Bignell DE, Slaytor M, Veivers PC, Mühlemann R, Leuthold RH (1994) Functions microbes. Appl Environ Microbiol 52:623–630
of symbiotic fungus gardens in higher termites of the genus Macrotermes: Breznak JA, Switzer Blum J (1991) Mixotrophy in the termite gut acetogen,
evidence against the acquired enzyme hypothesis. Acta Microbiol Immunol Sporomusa termitida. Arch Microbiol 156:105–110
Hung 41:391–401 Breznak JA, Brill WJ, Mertins JW, Coppel HC (1973) Nitrogen fixation in
Bignell DE, Eggleton P, Nunes L, Thomas KL (1997) Termites as mediators of termites. Nature 244:577–580
carbon fluxes in tropical forests: budgets for carbon dioxide and methane Breznak JA, Switzer JM, Seitz HJ (1988) Sporomusa termitida sp. nov., an H2/
emissions. In: Watt AB, Stork NE, Hunter MD (eds) Forests and insects. CO2-utilizing acetogen isolated from termites. Arch Microbiol 150:282–288
Chapman and Hall, London, pp 109–134 Brune A (1998) Termite guts: the world’s smallest bioreactors. Trends Biotechnol
Bloodgood RA, Fitzharris TP (1976) Specific association of prokaryotes with 16:16–21
symbiotic flagellate protozoa from the hindgut of the termite Reticulitermes Brune A (2003) Symbionts aiding digestion. In: Resh VH, Cardé RT (eds)
and the wood-eating roach. Cryptocercus Cytobios 17:103–122 Encyclopedia of insects. Academic, New York, pp 1102–1107
Bloodgood RA, Miller KR, Fitzharris TP, Mcintosh JR (1974) The ultrastructure Brune A, Friedrich M (2000) Microecology of the termite gut: structure and
of Pyrsonympha and its associated microorganisms. J Morphol 143:77–106 function on a microscale. Curr Opin Microbiol 3:263–269
Boga HI, Brune A (2003) Hydrogen-dependent oxygen reduction by Brune A, Kühl M (1996) pH profiles of the extremely alkaline hindguts of soil-
homoacetogenic bacteria isolated from termite guts. Appl Environ Microbiol feeding termites (Isoptera: Termitidae) determined with microelectrodes.
69:779–786 J Insect Physiol 42:1121–1127
Boga HI, Ludwig W, Brune A (2003) Sporomusa aerivorans sp. nov., an oxygen- Brune A, Stingl U (2005) Prokaryotic symbionts of termite gut flagellates: phy-
reducing homoacetogenic bacterium from the gut of a soil-feeding termite. logenetic and metabolic implications of a tripartite symbiosis. In: Overmann
Int J Syst Evol Microbiol 53:1397–1404 J (ed) Molecular basis of symbiosis. Springer, New York
Brauman A, Koenig JF, Dutreix J, Garcia JK (1990a) Characterization of two Brune A, Emerson D, Breznak JA (1995a) The termite gut microflora as an oxygen
sulfate-reducing bacteria from the gut of the soil-feeding termite, Cubitermes sink: microelectrode determination of oxygen and pH gradients in guts of
speciosus. Ant v Leeuwenhoek 58:271–275 lower and higher termites. Appl Environ Microbiol 61:2681–2687
Brauman A, Labat M, Garcia JL (1990b) Preliminary studies on the gut Brune A, Miambi E, Breznak JA (1995b) Roles of oxygen and the intestinal
microbiota of the soil-feeding termite: Cubitermes speciosus. In: Lésel microflora in the metabolism of lignin-derived phenylpropanoids and
R (ed) Microbiology in poecilotherms. Elsevier, Amsterdam, pp 73–77 other monoaromatic compounds by termites. Appl Environ Microbiol
Brauman A, Kane MD, Labat M, Breznak JA (1992) Genesis of acetate and 61:2688–2695
methane by gut bacteria of nutritionally diverse termites. Science Brune A, Frenzel P, Cypionka H (2000) Life at the oxic-anoxic interface: microbial
257:1384–1387 activities and adaptations. FEMS Microbiol Rev 24:691–710
Brune A, Ludwig W, Schink B (2002) Propionivibrio limicola sp. nov., de Bary A (1878) Ueber Symbiose Bericht der Versammlung. Deutscher
a fermentative bacterium specialized in the degradation of hydroaromatic Naturforscher und Aerzte, pp 121–126
compounds, reclassification of Propionibacter pelophilus as Propionivibrio Derakshani M, Lukow L, Liesack W (2001) Novel bacterial lineages at the (sub)
pelophilus comb. nov. and amended description of the genus Propionivibrio. division level as detected by signature nucleotide-targeted recovery of 16S
Int J Syst Evol Microbiol 52:441–444 rRNA genes from bulk soil and rice roots of flooded rice microcosms. Appl
Cazemier AE, Verdoes JC, Reubsaet FAG, Hackstein JHP, van der Drift C, Op den Environ Microbiol 67:623–631
Camp HJM (2003) Promicromonospora pachnodae sp. nov., a member of Dolan MF (2001) Speciation of termite gut protists: the role of bacterial symbi-
the (hemi)cellulolytic hindgut flora of larvae of the scarab beetle Pachnoda onts. Int Microbiol 4:203–208
marginata. Ant v Leeuwenhoek 83:135–148 Dolan M, Margulis L (1997) Staurojoenia and other symbionts in Neotermes
Cleveland LR (1925a) The effects of oxygenation and starvation on the symbiosis from San Salvador Island. Bahamas Symbiosis 22:229–239
between the termite, Termopsis, and its intestinal flagellates. Biol Bull Donovan SE, Eggleton P, Bignell DE (2001) Gut content analysis and a new
48:309–327 feeding group classification of termites. Ecol Entomol 26:356–366
Cleveland LR (1925b) Toxicity of oxygen for protozoa in vivo and in vitro: Donovan SE, Purdy KJ, Kane MD, Eggleton P (2004) Comparison of Euryarchaea
animals defaunated without injury. Biol Bull 48:455–468 strains in the guts and food-soil of the soil-feeding termite
Cleveland LR (1926) Symbiosis among animals with special reference to termites Cubitermes fungifaber across different soil types. Appl Environ Microbiol
and their intestinal flagellates. Quart Rev Biol 1:51–64 70:3884–3892
Cleveland LR, Grimstone AV (1964) The fine structure of the flagellate Mixotricha Ebert A, Brune A (1997) Hydrogen concentration profiles at the oxic-anoxic
paradoxa and its associated micro-organisms. Proc R Soc Lond Ser B Biol Sci interface: a microsensor study of the hindgut of the wood-feeding lower
159:668–686 termite Reticulitermes flavipes (Kollar). Appl Environ Microbiol 63:4039–
Collins NM (1983) The utilization of nitrogen resources by termites (Isoptera). 4046
In: Lee JA, McNeill S, Rorison ICH (eds) Nitrogen as an ecological factor. Eutick ML, O’Brien RW, Slaytor M (1976) Aerobic state of gut of Nasutitermes
Blackwell Scientific, Oxford, UK, pp 381–412 exitiosus and Coptotermes lacteus, high and low caste termites. J Insect
Collins MD, Shah HN (1986) Reclassification of Bacteroides termitidis Sebald Physiol 22:1377–1380
(Holdeman and Moore) in a new genus Sebaldella, as Sebaldella termitidis Eutick ML, O’Brien RW, Slaytor M (1978a) Bacteria from the gut of Australian
comb. nov. Int J Syst Bacteriol 36:349–350 termites. Appl Environ Microbiol 35:823–828
Cook SF (1932) The respiratory gas exchange in Termopsis nevadensis. Biol Bull Eutick ML, Veivers PC, O’Brien RW, Slaytor M (1978b) Dependence of the higher
63:246–257 termite Nasutitermes exitiosus and the lower termite Coptotermes lacteus on
Cook SF (1943) Nonsymbiotic utilization of carbohydrates by the termite their gut flora. J Insect Physiol 24:363–368
Zootermopsis angusticollis. Physiol Zool 16:123–128 French JRJ, Turner GL, Bradbury JF (1976) Nitrogen fixation by bacteria from the
Cook TJ, Gold RE (2000) Effects of different cellulose sources on the structure of hindgut of termites. J Gen Microbiol 95:202–206
the hindgut flagellate community in Reticulitermes virginicus (Isoptera: Friedrich MW, Schmitt-Wagner D, Lueders T, Brune A (2001) Axial differences in
Rhinotermitidae). Sociobiology 35:119–130 community structure of Crenarchaeota and Euryarchaeota in the highly
Cord-Ruwisch R, Seitz HJ, Conrad R (1988) The capacity of hydrogenotrophic compartmentalized gut of the soil-feeding termite Cubitermes orthognathus.
anaerobic bacteria to compete for traces of hydrogen depends on the Appl Environ Microbiol 67:4880–4890
redox potential of the terminal electron acceptor. Arch Microbiol Fröhlich J, König H (1999) Rapid isolation of single microbial cells from mixed
149:350–357 natural and laboratory populations with the aid of a micromanipulator. Syst
Cornelius ML, Daigle DJ, Connick WJ Jr, Parker A, Wunch K (2002) Responses of Appl Microbiol 22:249–257
Coptotermes formosanus and Reticulitermes flavipes (Isoptera: Fröhlich J, Sass H, Babenzien HD, Kuhnigk T, Varma A, Saxena S, Nalepa C,
Rhinotermitidae) to three types of wood rot fungi cultured on different Pfeiffer P, König H (1999) Isolation of Desulfovibrio intestinalis sp. nov. from
substrates. J Econ Entomol 95:121–128 the hindgut of the lower termite Mastotermes darwiniensis. Can J Microbiol
Crosland MWJ, Chan LK, Buswell JA (1996) Symbiotic fungus and enzymatic 45:145–152
digestion in the gut of the termite, Macrotermes barneyi (Light) (Isoptera: Fujita A (2004) Lysozymes in insects: what role do they play in nitrogen metab-
Termitidae). J Entomol Sci 31:132–137 olism? Physiol Entomol 299:305–310
Curtis AD, Waller DA (1995) Changes in nitrogen fixation rates in termites Fujita A, Abe T (2002) Amino acid concentration and distribution of lysozyme
(Isoptera: Rhinotermitidae) maintained in the laboratory. Ann Entomol and protease activities in the guts of higher termites. Physiol Entomol
Soc Am USA 88:764–767 27:76–78
Curtis AD, Waller DA (1996) The effects of decreased pO2 and increased pCO2 on Fujita A, Shimizu I, Abe T (2001) Distribution of lysozyme and protease, and
nitrogen fixation rates in termites (Isoptera: Rhinotermitidae). J Insect amino acid concentration in the guts of a wood-feeding termite,
Physiol 42:867–872 Reticulitermes speratus (Kolbe): possible digestion of symbiont bacteria
Curtis AD, Waller DA (1998) Seasonal patterns of nitrogen fixation in termites. transferred by trophallaxis. Physiol Entomol 26:116–123
Funct Ecol 12:803–807 Fujita A, Minamoto T, Shimizu I, Abe T (2002) Molecular cloning of
Cypionka H (2000) Oxygen respiration by Desulfovibrio species. Ann Rev lysozyme-encoding cDNAs expressed in the salivary gland of a wood-
Microbiol 54:827–848 feeding termite, Reticulitermes speratus. Insect Biochem Molec Biol
Czolij R, Slaytor M, Veivers PC, O’Brien RW (1984) Gut morphology of 32:1615–1624
Mastotermes darwiniensis Froggatt (Isoptera: Mastotermitidae). Int Graber JR, Breznak JA (2004) Physiology and nutrition of Treponema primitia, an
J Insect Morphol Embryol 13:337–355 H2/CO2-acetogenic spirochete from termite hindguts. Appl Environ
Czolij R, Slaytor M, O’Brien RW (1985) Bacterial flora of the mixed segment and Microbiol 70:1307–1314
the hindgut of the higher termite Nasutitermes exitiosus Hill (Termitidae, Graber JR, Breznak JA (2005) Folate cross-feeding supports symbiotic
Nasutitermitinae). Appl Environ Microbiol 49:1226–1236 homoacetogenic spirochetes. Appl Environ Microbiol 71:1883–1889
D’Ambrosio U, Dolan M, Wier AM, Margulis L (1999) Devescovinid trichomo- Graber JR, Leadbetter JR, Breznak JA (2004) Description of Treponema
nad with axostyle-based rotary motor (‘‘Rubberneckia’’): taxonomic assign- azotonutricium sp. nov. and Treponema primitia sp. nov., the first spirochetes
ment as Caduceia versatilis sp. nov. Eur J Protistol 35:327–337 isolated from termite guts. Appl Environ Microbiol 70:1307–1314
Darlington JPEC (1994) Nutrition and evolution in fungus-growing termites. In: Grandi G, Guidi L, Chicca M (1997) Endonuclear bacterial symbionts in two
Hunt JH, Nalepa CA (eds) Nourishment and evolution in insect societies. termite species: an ultrastructural study. J Submicrosc Cytol Pathol
Westview Press, Boulder, pp 105–130 29:281–292
Davison A, Blaxter M (2005) Ancient origin of glycosyl hydrolase family 9 Grassé P-P, Noirot C (1958) Le meule des termites champignonnistes et sa
cellulase genes. Mol Biol Evol 22:1273–1284 signification symbiotique. Ann Sci Nat Ser 11, Zool Biol Animale 11:113–128
Grech-Mora I, Fardeau ML, Patel BKC, Ollivier B, Rimbault A, Prensier G, Garcia Itakura S, Tanaka H, Enoki A (1999) Occurrence and metabolic role of the
JL, Garnier-Sillam E (1996) Isolation and characterization of Sporobacter pyruvate dehydrogenase complex in the lower termite Coptotermes
termitidis gen. nov., sp. nov., from the digestive tract of the wood-feeding formosanus (Shiraki). Insect Biochem Molec Biol 29:625–633
termite Nasutitermes lujae. Int J Syst Bacteriol 46:512–518 Itakura S, Tanaka H, Enoki A, Chappell DJ, Slaytor M (2003) Pyruvate and acetate
Guo L, Quicili DR, Chase J, Blomquist GJ (1991) Gut tract microorganisms metabolism in termite mitochondria. J Insect Physiol 49:917–926
supply the precursors for methyl-branched hydrocarbon biosynthesis in Jeffries TW (1994) Biodegradation of lignin and hemicelluloses. In: Ratledge
the termite, Zootermopsis nevadensis. Insect Biochem 21:327–333 C (ed) Biochemistry of microbial degradation. Kluwer, Dordrecht,
Harazono K, Yamashita N, Shinzato N, Watanabe Y, Fukatsu T, Kurane R (2003) pp 233–277
Isolation and characterization of aromatics-degrading microorganisms from Ji R, Brune A (2001) Transformation and mineralization of 14C-labeled cellulose,
the gut of the lower termite Coptotermes formosanus. Biosci Biotechnol peptidoglycan, and protein by the soil-feeding termite Cubitermes
Biochem 67:889–892 orthognathus. Biol Fertil Soils 33:166–174
Hethener P, Brauman A, Garcia JL (1992) Clostridium termitidis sp. nov., Ji R, Brune A (2005) Digestion of peptidic residues in humic substances by an
a cellulolytic bacterium from the gut of the wood-feeding termite, alkali-stable and humic-acid tolerant proteolytic activity in the gut of soil-
Nasutitermes lujae. Syst Appl Microbiol 5:52–58 feeding termites. Soil Biol Biochem 37:1648–1655
Higashi M, Abe T, Burns TP (1992) Carbon-nitrogen balance and termite ecology. Ji R, Kappler A, Brune A (2000) Transformation and mineralization of synthetic
14
Proc R Soc Lond Ser B Biol Sci 249:303–308 C-labeled humic model compounds by soil-feeding termites. Soil Biol
Hirai H, Shinzato N, Nakagawa A, Watanabe Y, Kurane R (2000) Degradation of Biochem 32:1281–1291
lignin model compounds by various termites [in Japanese]. Mokuzai Johjima T, Ohkuma M, Kudo T (2003) Isolation and cDNA cloning of novel
Gakkaishi 46:63–67 hydrogen peroxide-dependent phenol oxidase from the basidiomycete
Hollande A, Valentin J (1969) Appareil de Golgi, pinocytose, lysosomes, Termitomyces albuminosus. Appl Microbiol Biotechnol 61:220–225
mitochondries, bactéries symbiontiques, atractophores et pleuromitose Kambhampati S, Eggleton P (2000) Taxonomy and phylogenetics of Isoptera. In:
chez les Hypermastigines du genre Joenia: Affinités entre. Joenidae et Abe T, Bignell DE, Higashi M (eds) Termites: evolution, sociality, symbiosis,
Trichomonadines Protistologica 5:39–86 ecology. Kluwer, Dordrecht, pp 1–23
Hongoh Y, Ohkuma M, Kudo T (2003a) Molecular analysis of bacterial Kane MD (1997) Microbial fermentation in insect guts. In: Mackie RI, White BA
microbiota in the gut of the termite Reticulitermes speratus (Isoptera: (eds) Gastrointestinal microbiology, vol 1. Chapman and Hall, New York,
Rhinotermitidae). FEMS Microbiol Ecol 44:231–242 pp 231–265
Hongoh Y, Yuzawa H, Ohkuma M, Kudo T (2003b) Evaluation of primers and Kane MD, Breznak JA (1991) Acetonema longum gen. nov. sp.nov., an H2/CO2
PCR conditions for the analysis of 16S rRNA genes from a natural environ- acetogenic bacterium from the termite, Pterotermes occidentis. Arch
ment. FEMS Microbiol Lett 221:299–304 Microbiol 156:91–98
Honigberg BM (1970) Protozoa associated with termites and their role in diges- Kane MD, Brauman A, Breznak JA (1991) Clostridium mayombei sp. nov., an
tion. In: Krishna K, Weesner FM (eds) Biology of termites, vol 2. Academic, H2/CO2 acetogenic bacterium from the gut of the African soil-feeding
New York, pp 1–36 termite, Cubitermes speciosus. Arch Microbiol 156:99–104
Hopkins DW, Chudek JA, Bignell DE, Frouz J, Webster EA, Lawson T (1998) Kappler A, Brune A (1999) Influence of gut alkalinity and oxygen status on
Application of 13C NMR to investigate the transformations and biodegrada- mobilization and size-class distribution of humic acids in the hindgut of
tion of organic materials by wood-and soil-feeding termites, and soil-feeding termites. Appl Soil Ecol 13:219–229
a coprophagous litter-dwelling dipteran larva. Biodegradation 9:423–431 Kappler A, Brune A (2002) Dynamics of redox potential and changes in redox
Hugenholtz P, Goebel BM, Pace NR (1998) Impact of culture-independent state of iron and humic acids during gut passage in soil-feeding termites
studies on the emerging phylogenetic view of bacterial diversity. J Bacteriol (Cubitermes spp.). Soil Biol Biochem 34:221–227
180:4765–4774 Katoh H, Miura T, Maekawa K, Shinzato N, Matsumoto T (2002) Genetic
Hungate RE (1939) Experiments on the nutrition of Zootermopsis. III: The anaerobic variation of symbiotic fungi cultivated by the macrotermitine termite
carbohydrate dissimilation by the intestinal protozoa. Ecology 20:230–245 Odontotermes formosanus (Isoptera: Termitidae) in the Ryukyu Archipelago.
Hungate RE (1943) Quantitative analyses of the cellulose fermentation by termite Molec Ecol 11:1565–1572
protozoa. Ann Entomol Soc Am 36:730–739 Katzin LI, Kirby H (1939) The relative weight of termites and their protozoa.
Hungate RE (1946) Studies on cellulose fermentation. II: an anaerobic cellulose- J Parasitol 25:444–445
decomposing actinomycete, Micromonospora propionici, n. sp. J Bacteriol Koidzumi M (1921) Studies on the intestinal protozoa found in the termites of
51:51–56 Japan. Parasitology 13:235–305
Hungate RE (1955) Mutualistic intestinal protozoa. In: Hutner SH, Lwoff A (eds) Kovoor J (1967) Le pH intestinal d’un termite supérieur, Microcerotermes
Biochemistry and physiology of protozoa, vol 2. Academic, New York, edentatus (Was., Amitermitinae). Insect Soc 14:157–160
pp 159–199 Kovoor J (1968) L’intestin d’un termite supérieur (Microcerotermes edentatus,
Hyodo F, Azuma J, Abe T (1999) Estimation of effect of passage through the gut of Wasman, Amitermitinae). Histophysiologie et flore bacterienne
a lower termite, Coptotermes formosanus Shiraki, on lignin by solid-state CP symbiotique. Bull Biol Fr Belg 102:45–84
MAS 13C-NMR. Holzforschung 53:244–246 Kudo T, Ohkuma M, Moriya S, Noda S, Ohtoko K (1998) Molecular phylogenetic
Hyodo F, Inoue T, Azuma JI, Tayasu I, Abe T (2000) Role of the mutualistic identification of the intestinal anaerobic microbial community in the hind-
fungus in lignin degradation in the fungus-growing termite Macrotermes gut of the termite, Reticulitermes speratus, without cultivation.
gilvus (Isoptera: Macrotermitinae). Soil Biol Biochem 32:653–658 Extremophiles 2:155–161
Iida T, Ohkuma M, Ohtoko K, Kudo T (2000) Symbiotic spirochetes in the Kuhnigk T, König H (1997) Degradation of dimeric lignin model compounds by
termite hindgut: Phylogenetic identification of ectosymbiotic spirochetes aerobic bacteria isolated from the hindgut of xylophagous termites. J Basic
of oxymonad protists. FEMS Microbiol Ecol 34:17–26 Microbiol 37:205–211
Inoue T, Murashima K, Azuma JI, Sugimoto A, Slaytor M (1997) Cellulose and Kuhnigk T, Borst E-M, Ritter A, Kämpfer P, Graf A, Hertel H, König H (1994)
xylan utilization in the lower termite Reticulitermes speratus. J Insect Physiol Degradation of lignin monomers by the hindgut flora of xylophagous ter-
43:235–242 mites. Syst Appl Microbiol 17:76–85
Inoue T, Kitade O, Yoshimura T, Yamaoka I (2000) Symbiotic associations with Kuhnigk T, Borst EM, Breunig A (1995) Bacillus oleronius sp. nov., a member of
protists. In: Abe T, Bignell DE, Higashi M (eds) Termites: evolution, sociality, the hindgut flora of the termite Reticulitermes santonensis (Feytaud). Can
symbiosis, ecology. Kluwer, Dordrecht, pp 275–288 J Microbiol 41:699–706
Inoue T, Moriya S, Ohkuma M, Kudo T (2005) Molecular cloning and charac- Kuhnigk T, Branke J, Krekeler D, Cypionka H, König H (1996) A feasible role
terization of a cellulase gene from a symbiotic protist of the lower termite of sulfate-reducing bacteria in the termite gut. Syst Appl Microbiol
Coptotermes formosanus. Gene 349:67–75 19:139–149
La Fage JP, Nutting WL (1978) Nutrient dynamics of termites. In: Brian MV (ed) Matsuura K (2001) Nestmate recognition mediated by intestinal bacteria in
Production ecology of ants and termites. Cambridge University Press, a termite, Reticulitermes speratus. Oikos 92:20–26
Cambridge, UK, pp 165–232 McSweeney CS, Dulieu A, Webb RI, Del Dot T, Blackall LL (1999) Isolation and
Leadbetter JR, Breznak JA (1996) Physiological ecology of Methanobrevibacter characterization of a Clostridium sp. with cinnamoyl esterase activity and
cuticularis sp. nov. and Methanobrevibacter curvatus sp. nov., isolated from unusual cell envelope ultrastructure. Arch Microbiol 172:139–149
the hindgut of the termite Reticulitermes flavipes. Appl Environ Microbiol Messer AC, Lee MJ (1989) Effect of chemical treatments on methane emission by
62:3620–3631 the hindgut microbiota in the termite Zootermopsis angusticollis. Microb
Leadbetter JR, Crosby LD, Breznak JA (1998) Methanobrevibacter filiformis sp. Ecol 18:275–284
nov., a filamentous methanogen from termite hindguts. Arch Microbiol Moriya S, Dacks JB, Takagi A, Noda S, Ohkuma M, Doolittle WF, Kudo T (2003)
169:287–292 Molecular phylogeny of three oxymonad genera: Pyrsonympha,
Leadbetter JR, Schmidt TM, Graber JR, Breznak JA (1999) Acetogenesis from H2 Dinenympha and Oxymonas. J Eukar Microbiol 50:190–197
plus CO2 by spirochetes from termite guts. Science 283:686–689 Nakashima K, Watanabe H, Azuma JI (2002a) Cellulase genes from the
Leander BS, Keeling PJ (2004) Symbiotic innovation in the oxymonad parabasalian symbiont Pseudotrichonympha grassii in the hindgut of the
Streblomastix strix. J Euk Microbiol 51:291–300 wood-feeding termite Coptotermes formosanus. Cell Mol Life Sci
Leaphart AB, Lovell CR (2001) Recovery and analysis of formyltetrahydrofolate 59:1554–1560
synthetase gene sequences from natural populations of acetogenic bacteria. Nakashima K, Watanabe H, Saitoh H, Tokuda G, Azuma J-I (2002b) Dual
Appl Environ Microbiol 67:1392–1395 cellulose-digesting system of the wood-feeding termite, Coptotermes
Leaphart AB, Friez MJ, Lovell CR (2003) Formyltetrahydrofolate synthetase formosanus Shiraki. Insect Biochem Molec Biol 32:777–784
sequences from salt marsh plant roots reveal a diversity of acetogenic Nalepa CA, Bignell DE, Bandi C (2001) Detritivory, coprophagy, and the evolu-
bacteria and other bacterial functional groups. Appl Environ Microbiol tion of digestive mutualisms in Dictyoptera. Insect Soc 48:194–201
69:693–696 Ndiaye D, Lensi R, Lepage M, Brauman A (2004) The effect of the soil-feeding
Lee MJ, Schreurs PJ, Messer AC, Zinder SH (1987) Association of methanogenic termite Cubitermes niokoloensis on soil microbial activity in a semi-arid
bacteria with flagellated protozoa from a termite hindgut. Curr Microbiol savanna in West Africa. Plant Soil 259:277–286
15:337–341 Noda S, Ohkuma M, Usami R, Horikoshi K, Kudo T (1999) Culture-independent
Leidy J (1849) Proc Acad Nat Sci (Phila) 4:225–233 characterization of a gene responsible for nitrogen fixation in the symbiotic
Leidy J (1881) The parasites of the termites. J Acad Nat Sci (Phila), 2nd Ser microbial community in the gut of the termite Neotermes koshunensis. Appl
8:425–447 Environ Microbiol 65:4935–4942
Lemke T, van Alen T, Hackstein JHP, Brune A (2001) Cross-epithelial hydrogen Noda S, Ohkuma M, Yamada A, Hongoh Y, Kudo T (2003) Phylogenetic position
transfer from the midgut compartment drives methanogenesis in the hind- and in situ identification of ectosymbiotic spirochetes on protists in the
gut of cockroaches. Appl Environ Microbiol 67:4657–4661 termite gut. Appl Environ Microbiol 69:625–633
Li L, Fröhlich J, Pfeiffer P, König H (2003) Termite gut symbiotic archaezoa are Noirot C (1992) From wood-to humus-feeding: an important trend in termite
becoming living metabolic fossils. Eukar Cell 2:1091–1098 evolution. In: Billen J (ed) Biology and evolution of social insects. Leuven
Lilburn TG, Schmidt TM, Breznak JA (1999) Phylogenetic diversity of termite gut University Press, Leuven, pp 107–119
spirochaetes. Environ Microbiol 1:331–345 Noirot C (1995) The gut of termites (isoptera): comparative anatomy, systemat-
Lilburn TG, Kim KS, Ostrom NE, Byzek KR, Leadbetter JR, Breznak JA ics, phylogeny I: lower termites. Ann Soc Entomol Fr (NS) 31:197–226
(2001) Nitrogen fixation by symbiotic and free-living spirochetes. Science Noirot C (2001) The gut of termites (isoptera) Comparative anatomy, systemat-
292:2495–2498 ics, phylogeny. II: higher termites (termitidae). Ann Soc Entomol Fr (NS)
Lo N, Bandi C, Watanabe H, Nalepa C, Beninati T (2003a) Evidence for 37:431–471
co-cladogenesis between diverse dictyopteran lineages and their intracellular O’Brien RW, Breznak JA (1984) Enzymes of acetate and glucose metabolism in
endosymbionts. Molec Biol Evol 20:907–913 termites. Insect Biochem 14:639–643
Lo N, Watanabe H, Sugimura M (2003b) Evidence for the presence of a cellulase O’Brien RW, Slaytor M (1982) Role of microorganisms in the metabolism of
gene in the last common ancestor of bilaterian animals. Proc R Soc Lond Ser termites Australian. J Biol Sci 35:239–262
B Biol Sci 270(Suppl 1):S69–S72 Odelson DA, Breznak JA (1983) Volatile fatty acid production by the hindgut
Lysenko O (1985) Non-sporeforming bacteria pathogenic to insects: incidence microbiota of xylophagous termites. Appl Environ Microbiol 45:1602–1613
and mechanisms. Ann Rev Microbiol 39:673–695 Odelson DA, Breznak JA (1985a) Cellulase and other polymer-hydrolyzing activ-
Machida M, Kitade O, Miura T, Matsumoto T (2001) Nitrogen recycling through ities of Trichomitopsis termopsidis, a symbiotic protozoan from termites.
proctodeal trophallaxis in the Japanese damp-wood termite Hodotermopsis Appl Environ Microbiol 49:622–626
japonica (Isoptera, Termopsidae). Insect Soc 48:52–56 Odelson DA, Breznak JA (1985b) Nutrition and growth characteristics of
Mannesmann R, Piechowski B (1989) Verteilungsmuster von Gärkammer- Trichomitopsis termopsidis, a cellulolytic protozoan from termites. Appl
bakterien einiger Termitenarten. Mater Org 24:161–178 Environ Microbiol 49:614–621
Margulis L, Olendzenski A, Afzelius BA (1990) Endospore-forming filamentous Ohkuma M (2002) Symbiosis in the termite gut: culture-independent molecular
bacteria symbiotic in termites: ultrastructure and growth in culture of approaches. In: Seckbach J (ed) Symbiosis: mechanisms and model systems.
Arthromitus. Symbiosis 8:95–116 Kluwer, Dordrecht, pp 717–730
Margulis L, Jorgensen JZ, Dolan S, Kolchinsky R, Rainey FA, Lo SC (1998) The Ohkuma M (2003) Termite symbiotic systems: efficient bio-recycling of lignocel-
Arthromitus stage of Bacillus cereus: intestinal symbionts of animals. Proc lulose. Appl Microbiol Biotechnol 61:1–9
Natl Acad Sci USA 95:1236–1241 Ohkuma M, Kudo T (1996) Phylogenetic diversity of the intestinal bacterial
Martin MM (1983) Cellulose digestion in insects. Comp Biochem Physiol community in the termite Reticulitermes speratus. Appl Environ Microbiol
75A:313–324 62:461–468
Martin MM, Martin JS (1978) Cellulose digestion in the midgut of the fungus- Ohkuma M, Kudo T (1998) Phylogenetic analysis of the symbiotic intestinal
growing termite Macrotermes natalensis: the role of acquired digestive microflora of the termite Cryptotermes domesticus. FEMS Microbiol Lett
enzymes. Science 199:1453–1455 164:389–395
Martin MM, Martin JS (1979) The distribution and origins of the cellulolytic Ohkuma M, Noda S, Horikoshi K, Kudo T (1995) Phylogeny of symbiotic
enzymes of the higher termite, Macrotermes natalensis. Physiol Zool methanogens in the gut of the termite Reticulitermes speratus. FEMS
52:11–21 Microbiol Lett 134:45–50
Matoub M, Rouland C (1995) Purification and properties of the xylanases from Ohkuma M, Noda S, Usami R, Horikoshi K, Kudo T (1996) Diversity of nitrogen-
the termite Macrotermes bellicosus and its symbiotic fungus Termitomyces sp. fixation genes in the symbiotic intestinal microflora of the termite
Comp Biochem Physiol 112B:629–635 Reticulitermes speratus. Appl Environ Microbiol 62:2747–2752
Ohkuma M, Iida T, Kudo T (1999a) Phylogenetic relationships of Rath AC (2000) The use of entomopathogenic fungi for control of termites.
symbiotic spirochetes in the gut of diverse termites. FEMS Microbiol Lett Biocontrol Sci Technol 10:563–581
181:123–129 Rohrmann GF, Rossman AY (1980) Nutrient strategies of Macrotermes ukuzii
Ohkuma M, Noda S, Kudo T (1999b) Phylogenetic diversity of nitrogen fixation (Isoptera: Termitidae). Pedobiologia 20:61–73
genes in the symbiotic microbial community in the gut of diverse termites. Rother A, Radek R, Hausmann K (1999) Characterization of surface structures
Appl Environ Microbiol 65:4926–4934 covering termite flagellates of the family Oxymonadidae and ultrastructure
Ohkuma M, Noda S, Kudo T (1999c) Phylogenetic relationships of symbiotic of two oxymonad species, Microrhopalodina multinucleata and Oxymonas
methanogens in diverse termites. FEMS Microbiol Lett 171:147–153 sp. Eur J Protistol 35:1–16
Ohkuma M, Noda S, Hongoh Y, Kudo T (2002) Diverse bacteria related to the Rouland C (2000) Symbiosis with fungi. In: Abe T, Bignell DE, Higashi M (eds)
bacteroides subgroup of the CFB phylum within the gut symbiotic commu- Termites: evolution, sociality, symbiosis, ecology. Kluwer, Dordrecht,
nities of various termites. Biosci Biotechnol Biochem 66:78–84 pp 289–306
Ohkuma M, Shimizu H, Thongaram T, Kosono S, Moriya K, Trakulnaleamsai S, Rouland C, Chararas C, Renoux J (1989) Les osidases digestives présentes dans
Noparatnaraporn N, Kudo T (2003) An alkaliphilic and xylanolytic l’intestin moyen, l’intestin postérieur et les glandes salivaires du
Paenibacillus species isolated from the gut of a soil-feeding termite. Microb termite humivore Crenetermes albotarsalis. CR Acad Sci Paris Série III
Environ 18:145–151 308:281–285
Ohtoko K, Ohkuma M, Moriya S, Inoue T, Usami R, Kudo T (2000) Diverse genes Rouland-Lefevre C, Diouf MN, Brauman A, Neyra M (2002) Phylogenetic rela-
of cellulase homologues of glycosyl hydrolase family 45 from the symbiotic tionships in Termitomyces (family Agaricaceae) based on the nucleotide
protists in the hindgut of the termite Reticulitermes speratus. Extremophiles sequence of ITS: a first approach to elucidate the evolutionary history of
4:343–349 the symbiosis between fungus-growing termites and their fungi. Molec
Osbrink WLA, Williams KS, Connick WJ Jr, Wright MS, Lax AR (2001) Virulence Phylogenet Evol 22:423–429
of bacteria associated with the formosan subterranean termite (Isoptera: Sacchi L, Nalepa CA, Lenz M, Bandi C, Corona S, Grigolo A, Bigliardi E (2000)
Rhinotermitidae) in New Orleans, La. Environ Entomol 30:443–448 Transovarial transmission of symbiotic bacteria in Mastotermes darwiniensis
Paster BJ, Ludwig W, Weisburg WG, Stackebrandt E, Hespell RB, Hahn CM, (Isoptera: Mastotermitidae): ultrastructural aspects and phylogenetic impli-
Reichenbach H, Stetter KO, Woese CR (1985) A phylogenetic grouping of the cations. Ann Entomol Soc Am 93:1308–1313
bacteroides, cytophagas, and certain flavobacteria. Syst Appl Microbiol Salmassi TM, Leadbetter JR (2003) Molecular aspects of CO2-reductive
6:34–42 acetogenesis in cultivated spirochetes and the gut community of the termite
Paster BJ, Dewhirst FE, Cooke SM, Fussing V, Poulsen LK, Breznak JA (1996) - Zootermopsis angusticollis. Microbiology 149:2529–2537
Phylogeny of not-yet-cultured spirochetes from termite guts. Appl Environ Sanderson MG (1996) Biomass of termites and their emissions of methane and
Microbiol 62:347–352 carbon dioxide: a global database. Global Biogeochem Cycles 10:543–557
Pasti MB, Belli ML (1985) Cellulolytic activity of actinomycetes isolated from Sands WA (1969) The association of termites and fungi. In: Krishna K,
termite (Termitidae) gut. FEMS Microbiol Lett 26:107–112 Weesner FM (eds) Biology of termites, vol 1. Academic, New York,
Pasti MB, Pometto AL III, Nuti MP, Crawford DL (1990) Lignin-solubilizing pp 495–524
ability of actinomycetes isolated from termite (Termitidae) gut. Appl Envi- Schäfer A, Konrad R, Kuhnigk T, Kämpfer P, Hertel H, König H (1996) Hemi-
ron Microbiol 56:2213–2218 cellulose-degrading bacteria and yeasts from the termite gut. J Appl Bacteriol
Patricolo E, Villa L, Arizzi M (2001) TEM observations on symbionts of Joenia 80:471–478
annectens (Flagellata Hypermastigida). J Nat Hist 35:471–480 Schmitt-Wagner D, Brune A (1999) Hydrogen profiles and localization of
Paul J, Sarkar A, Varma AK (1986) In vitro studies of cellulose digesting properties methanogenic activities in the highly compartmentalized hindgut of soil-
of Staphylococcus saprophyticus isolated from termite gut. Curr Sci feeding higher termites (Cubitermes spp.). Appl Environ Microbiol
55:710–714 65:4490–4496
Paul J, Saxena S, Varma A (1993) Ultrastructural studies of the termite Schmitt-Wagner D, Friedrich M, Wagner B, Brune A (2003a) Axial dynamics,
(Odontotermes obesus) gut microflora and its cellulolytic properties. World stability, and interspecies similarity of bacterial community structure in the
J Microbiol Biotechnol 9:108–112 highly compartmentalized gut of soil-feeding termites (Cubitermes spp.).
Pierantoni U (1936) La simbiosi fisiologica nei termitidi xilophagi e nei loro Appl Environ Microbiol 69:6018–6024
flagellati intestinali. Arch Zool Ital 22:135–173 Schmitt-Wagner D, Friedrich M, Wagner B, Brune A (2003b) Phylogenetic
Potrikus CJ, Breznak JA (1977) Nitrogen-fixing Enterobacter agglomerans isolated diversity, abundance, and axial distribution of bacteria in the intestinal
from guts of wood-eating termites. Appl Environ Microbiol 33:392–399 tract of two soil-feeding termites (Cubitermes spp.). Appl Environ Microbiol
Potrikus CJ, Breznak JA (1980) Anaerobic degradation of uric acid by gut bacteria 69:6007–6017
of termites. Appl Environ Microbiol 40:125–132 Schultz JE, Breznak JA (1978) Heterotrophic bacteria present in hindguts of
Potrikus CJ, Breznak JA (1981) Gut bacteria recycle uric acid nitrogen in wood-eating termites [Reticulitermes flavipes (Kollar)]; Appl Environ
termites: a strategy for nutrient conservation. Proc Natl Acad Sci USA Microbiol 35:930–936
78:4601–4605 Schultz JE, Breznak JA (1979) Cross-feeding of lactate between Streptococcus lactis
Potts RC, Hewitt PH (1973) The distribution of intestinal bacteria and cellulase and Bacteroides sp. isolated from termite hindguts. Appl Environ Microbiol
activity in the harvester termite Trinervitermes trinervoides 37:1206–1210
(Nasutitermitidae). Insect Soc 20:215–220 Seedorf H, Dreisbach A, Hedderich R, Shima S, Thauer RK (2004) F420H2
Radek R (1999) Flagellates, bacteria, and fungi associated with termites: diversity oxidase (FprA) from Methanobrevibacter arboriphilus, a coenzyme
and function in nutrition – a review. Ecotropica 5:183–196 F420-dependent enzyme involved in O2 detoxification. Arch Microbiol
Radek R, Tischendorf G (1999) Bacterial adhesion to different termite flagellates: 182:126–137
ultrastructural and functional evidence for distinct molecular attachment Shima S, Sordel-Klippert M, Brioukhanov A, Netrusov A, Linder D, Thauer RK
modes. Protoplasma 207:43–53 (2001) Characterization of a heme-dependent catalase from Methanobre-
Radek R, Hausmann K, Breunig A (1992) Ectobiotic and endocytobiotic bacteria vibacter arboriphilus. Appl Environ Microbiol 67:3041–3045
associated with the termite flagellate Joenia annectens. Acta Protozool Shinzato N, Matsumoto T, Yamaoka I, Oshima T, Yamagishi A (1999) Phyloge-
31:93–107 netic diversity of symbiotic methanogens living in the hindgut of the lower
Radek R, Rösel J, Hausmann K (1996) Light and electron microscopic study of the termite Reticulitermes speratus analyzed by PCR and in situ hybridization.
bacterial adhesion to termite flagellates applying lectin cytochemistry. Appl Environ Microbiol 65:837–840
Protoplasma 193:105–122 Shinzato N, Matsumoto T, Yamaoka I, Oshima T, Yamagishi A (2001)
Rasmussen RA, Khalil MAK (1983) Global production of methane by termites. Methanogenic symbionts and the locality of their host lower termites.
Nature 301:704–705 Microb Environ 16:43–47
Sikorowski PP, Lawrence AM (1998) Transmission of Serratia marcescens Tholen A, Brune A (2000) Impact of oxygen on metabolic fluxes and in situ rates
(Enterobacteriaceae) in adult Heliothis virescens (Lepidoptera: Noctuidae) of reductive acetogenesis in the hindgut of the wood-feeding termite
laboratory colonies. Biol Control 12:50–55 Reticulitermes flavipes. Environ Microbiol 2:436–449
Slaytor M (1992) Cellulose digestion in termites and cockroaches: what role do Tholen A, Schink B, Brune A (1997) The gut microflora of Reticulitermes
symbionts play? Comp Biochem Physiol 103B:775–784 flavipes, its relation to oxygen, and evidence for oxygen-dependent
Slaytor M (2000) Energy metabolism in the termite gut and its gut microbiota. In: acetogenesis by the most abundant Enterococcus sp. FEMS Microbiol Ecol
Abe T, Bignell DE, Higashi M (eds) Termites: evolution, sociality, symbiosis, 24:137–149
ecology. Kluwer, Dordrecht, pp 307–332 Thongaram T, Kosono S, Ohkuma M, Hongoh Y, Kitada M, Yoshinaka T,
Slaytor M, Chappell DJ (1994) Nitrogen metabolism in termites. Comp Biochem Trakulnaleamsai S, Noparatnaraporn N, Kudo T (2003) Gut of higher
Physiol 107:1–10 termites as a niche for alkaliphiles as shown by culture based and culture-
Slaytor M, Veivers PC, Lo N (1997) Aerobic and anaerobic metabolism in the independent studies. Microb Environ 18:152–159
higher termite Nasutitermes walkeri (Hill). Insect Biochem Molec Biol Thongaram T, Hongoh Y, Kosono S, Ohkuma M, Trakulnaleamsai S,
27:291–303 Noparatnaraporn N, Kudo T (2005) Comparison of bacterial communities
Snel J, Blok HJ, Kengen HMP, Ludwig W, Poelma FGJ, Koopman JP, Akkermans in the alkaline gut segment among various species of higher termites.
ADL (1994) Phylogenetic characterization of Clostridium related segmented Extremophiles 9:229–238
filamentous bacteria in mice based on 16S ribosomal RNA analysis. Syst Appl To L, Margulis L, Cheung ATW (1978) Pillotinas and hollandinas: distribution
Microbiol 17:172–179 and behaviour of large spirochaetes symbiotic in termites. Microbios
Snel J, Heinen PP, Blok HJ, Carman RJ, Duncan AJ, Allen PC, Collins MD 22:103–133
(1995) Comparison of 16S rRNA sequences of segmented filamentous bac- To LP, Margulis L, Chase D, Nutting WL (1980) The symbiotic microbial com-
teria isolated from mice, rats, and chickens and proposal of ‘‘Candidatus munity of the Sonoran desert termite: Pterotermes occidentis. Biosystems
Arthromitus’’. Int J Syst Bacteriol 45:780–782 13:109–137
Stackebrandt E, Schumann P, Cui XL (2004) Reclassification of Cellulosi- Tokuda G, Yamaoka I, Noda H (2000) Localization of symbiotic clostridia in the
microbium variabile Bakalidou et al. 2002 as Isoptericola variabilis gen. mixed segment of the termite Nasutitermes takasagoensis (Shiraki). Appl
nov., comb. nov. Int J System Evol Microbiol 54:685–688 Environ Microbiol 66:2199–2207
Stingl U, Brune A (2003) Phylogenetic diversity and whole-cell hybridization of Tokuda G, Nakamura T, Murakami R, Yamaoka I (2001) Morphology of the
oxymonad flagellates from the hindgut of the wood-feeding lower termite digestive system in the wood-feeding termite Nasutitermes takasagoensis
Reticulitermes flavipes. Protist 154:147–155 (Shiraki) [Isoptera: Termitidae]; Zool Sci 18:869–877
Stingl U, Maass A, Radek R, Brune A (2004) Symbionts of the gut flagellate Tokuda G, Lo N, Watanabe H, Arakawa G, Matsumoto T, Noda H (2004) Major
Staurojoenina sp. from Neotermes cubanus represent a novel, termite- alteration of the expression site of endogenous cellulases in members of an
associated lineage of Bacteroidales: description of ‘‘Candidatus Vestibaculum apical termite lineage. Mol Ecol 13:3219–3228
illigatum. Microbiology 150:2229–2235 Tokura M, Ohkuma M, Kudo T (2000) Molecular phylogeny of methanogens
Stingl U, Radek R, Brune A (2005) ‘‘Endomicrobia’’: cytoplasmic symbionts of associated with flagellated protists in the gut and with the gut epithelium of
termite gut protozoa form a separate phylum of prokaryotes. Appl Environ termites. FEMS Microbiol Ecol 33:233–240
Microbiol 71:1473–1479 Trinkerl M, Breunig A, Schauder R, König H (1990) Desulfovibrio termitidis sp.
Sugimoto A, Inoue T, Tayasu I, Miller L, Takeichi S, Abe T (1998) Methane nov., a carbohydrate-degrading sulfate-reducing bacterium from the hind-
and hydrogen production in a termite-symbiont system. Ecol Res gut of a termite. System Appl Microbiol 13:372–377
13:241–257 Veivers PC, O’Brien RW, Slaytor M (1980) The redox state of the gut of termites.
Tamm SL (1980) The ultrastructure of prokaryotic-eukaryotic cell junctions. J Insect Physiol 26:75–77
J Cell Sci 44:335–352 Veivers PC, O’Brien RW, Slaytor M (1982) Role of bacteria in maintaining the
Tamm SL (1982) Flagellated epibiotic bacteria propel a eucaryotic cell. J Cell Biol redox potential in the hindgut of termites and preventing entry of foreign
94:697–709 bacteria. J Insect Physiol 28:947–951
Taprab Y, Ohkuma M, Johjima T, Maeda Y, Moriya S, Inoue T, Suwanarit P, Vu AT, Ngyen NC, Leadbetter JR (2004) Iron reduction in the metal-rich guts of
Noparatnaraporn N, Kudo T (2002) Molecular phylogeny of symbiotic wood-feeding termites. Geobiology 2:239–247
basidiomycetes of fungus-growing termites in Thailand and their relation- Watanabe H, Tokuda G (2001) Animal cellulases. Cell Mol Life Sci
ship with the host. Biosci Biotechnol Biochem 66:1159–1163 58:1167–1178
Tayasu I (1998) The use of carbon and nitrogen isotope ratios in termite research. Watanabe H, Nakashima K, Saito H, Slaytor M (2002) New endo-beta-1,4-
Ecol Res 13:377–387 glucanases from the parabasalian symbionts, Pseudotrichonympha grassii
Tayasu I, Sugimoto A, Wada E, Abe T (1994) Xylophagous termites depending on and Holomastigotoides mirabile of Coptotermes termites. Cell Mol Life Sci
atmospheric nitrogen. Naturwissenschaften 81:229–231 59:1983–1992
Tayasu I, Abe T, Eggleton P, Bignell DE (1997) Nitrogen and carbon isotope ratios Watanabe Y, Shinzato N, Fukatsu T (2003) Isolation of actinomycetes from
in termites: an indicator of trophic habit along the gradient from wood- termites’ guts. Biosci Biotechnol Biochem 67:1797–1801
feeding to soil-feeding. Ecol Entomol 22:343–351 Wenzel M, Schönig I, Berchtold M, Kämpfer P, König H (2002) Aerobic and
Terra WR (1990) Evolution of digestive systems of insects. Ann Rev Entomol facultatively anaerobic cellulolytic bacteria from the gut of the termite
35:181–200 Zootermopsis angusticollis. J Appl Microbiol 92:32–40
Thauer RK, Jungermann K, Decker K (1977) Energy conservation in Wenzel M, Radek R, Brugerolle G, König H (2003) Identification of the
chemotrophic anaerobic bacteria. Bacteriol Rev 41:100–180 ectosymbiotic bacteria of Mixotricha paradoxa involved in movement sym-
Thayer DW (1976) Facultative wood-digesting bacteria from the hind-gut of the biosis. Eur J Protistol 39:11–24
termite Reticulitermes hesperus. J Gen Microbiol 95:287–296 Wier A, Ashen J, Margulis L (2000) Canaleparolina darwiniensis, gen. nov., sp.
Thayer DW (1978) Carboxymethylcellulase produced by facultative bacteria from nov., and other pillotinaceous spirochetes from insects. Int Microbiol
the hind-gut of the termite Reticulitermes hesperus. J Gen Microbiol 3:213–223
106:13–18 Williams CM, Veivers PC, Slaytor M, Cleland SV (1994) Atmospheric carbon
Tholen A, Brune A (1999) Localization and in situ activities of homoacetogenic dioxide and acetogenesis in the termite Nasutitermes walkeri (Hill). Comp
bacteria in the highly compartmentalized hindgut of soil-feeding higher Biochem Physiol 107A:113–118
termites (Cubitermes spp.). Appl Environ Microbiol 65:4497–4505 Yamin MA (1979) Termite flagellates. Sociobiology 4:1–119
Yang H, Schmitt-Wagner D, Stingl U, Brune A (2005) Niche heterogeneity Yoshimura T, Fujino T, Ito T, Tsunoda K, Takahashi M (1996) Ingestion and
determines bacterial community structure in the in the termite gut decomposition of wood and cellulose by the protozoa in the hindgut of
(Reticulitermes santonensis). Environ Microbiol 7(7):916–932 Coptotermes formosanus Shiraki (Isoptera: Rhinotermitidae) as evidenced by
Yara K, Jahana K, Hayashi H (1989) In situ morphology of the gut microbiota of polarizing and transmission electron microscopy. Holzforschung 50:99–104
the fungus-growing termite Odontotermes formosanus (Termitidae: Zoberi MH, Grace JK (1990) Fungi associated with the subterranean termite
Macrotermitinae). Sociobiology 15:247–260 Reticulitermes flavipes in Ontario. Mycologia 82:289–294
23 Marine Chemosynthetic Symbioses
Colleen M. Cavanaugh1 . Zoe P. McKiness2 . Irene L. G. Newton3 . Frank J. Stewart4
1
Biological Laboratories 4081, Cambridge, MA, USA
2
Wildwood, MO, USA
3
Department of Biology, Indiana University Bloomington, Bloomington, IN, USA
4
School of Biology, Georgia Institute of Technology, Atlanta, GA, USA
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579 methanotrophs. In both chemoautotrophic and methanotrophic

History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 581 symbioses, the hosts, through an astonishing array of physiolog-
Habitat Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 581 ical and behavioral adaptations, provide the symbiont access to
Methods for Studying Chemosynthetic Symbioses . . . . . 583 the substrates (i.e., electron donors and acceptors) necessary for
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 583 the generation of energy and bacterial biomass. In exchange,
a portion of the carbon fixed by the symbiont is used, either
Host Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584 directly or indirectly, for host energy and biosynthesis. These
symbioses thereby increase the metabolic capabilities, and there-
Symbiont Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584 fore the number of ecological niches, of both the host and the
Morphology and Ultrastructure . . . . . . . . . . . . . . . . . . . . . . . . . 584 bacterial symbiont.
Symbiont Phylogeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 587 In those symbioses for which the electron donor has been
Symbiont Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591 explicitly identified, sulfide and other reduced inorganic sulfur
Enzyme Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591 compounds (e.g., thiosulfate) fuel energy generation by the chemo-
Stable Isotope Signatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . 592 autotrophic symbionts, serving as electron sources for oxidative
phosphorylation. In these symbioses, the ATP produced in electron
Ecophysiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 594 transport fuels autotrophic CO2 fixation via the Calvin cycle. In
Spanning the Oxic-Anoxic Interface . . . . . . . . . . . . . . . . . . . . 594 contrast, bacteria in marine invertebrate-methanotroph symbi-
Carbon Uptake and Transport . . . . . . . . . . . . . . . . . . . . . . . . . . 596 oses use methane (CH4) as an energy, electron, and carbon
Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 597 source. Unlike their protist or metazoan hosts, chemoautotrophs
and methanotrophs share the ability to use reduced inorganic
Ecology and Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 597 compounds or methane for energy generation and carbon dioxide
History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 597 or methane for carbon fixation and utilization. On the basis of
Organism Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 597 these unique biosynthetic capacities, notably the ability to syn-
Transmission Strategies and Effects on Symbiosis . . . . . . 598 thesize C3 compounds from C1 compounds, we refer collectively
Biogeography and Population Genetics . . . . . . . . . . . . . . . . . 599 to these bacterial symbionts as ‘‘chemosynthetic.’’
Given the sulfide-rich habitats in which chemoautotrophic
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600 symbioses occur, researchers infer that the bacterial symbionts
oxidize reduced inorganic sulfur compounds to obtain energy
and reducing power for autotrophic carbon fixation. While some
Introduction endosymbionts (such as those in the protobranchs Solemya velum
and S. reidi; Cavanaugh 1983; Anderson et al. 1987) utilize
Bacteria and marine eukaryotes often coexist in symbioses that thiosulfate (S2O3m), an intermediate in sulfide oxidation, hydro-
significantly influence the ecology, physiology, and evolution of gen sulfide is inferred to be the preferred energy source in
both partners. De Bary (1879) defined symbiosis as ‘‘the living a variety of symbioses (see review in Van Dover 2000). But for
together of differently named organisms,’’ implying that the many symbioses, the actual energy source has not been identi-
term encompasses both positive (e.g., mutualism) and negative fied definitively; rather, only an autotrophic metabolism has
(e.g., parasitism) associations. Many researchers now view sym- been confirmed. Indeed, chemosynthetic bacteria utilizing
biotic interactions as those that persist over the majority of the other energy sources (e.g., hydrogen or ammonia) could also
lifespan of the organisms involved and that provide benefits to serve similar nutritional roles in symbiotic associations. In this
each partner beyond those obtained in the absence of associa- review, bacterial symbionts that have been shown to use reduced
tion. This chapter describes such symbioses, specifically those sulfur compounds (H2S, HSm, Sm2, S2O3m, So) for energy metab-
between marine invertebrate and protist hosts and chemosyn- olism are referred to as thioautotrophs, while the more general
thetic bacterial symbionts. term ‘‘chemoautotroph’’ is used to describe symbionts for which
These bacteria, which cluster primarily within the data supporting autotrophic CO2 fixation exist but for which the
Gammaproteobacteria (> Fig. 23.1), are chemoautotrophs or lithotrophic energy source is unknown.
580 23 Marine Chemosynthetic Symbioses
Chemoautotrophic symbionts:
Mussel symbionts
Vesicomyid clam symbionts
Solemyid clam symbionts
Lucinid clam symbionts
Thyasirid clam symbionts
Tubeworm symbionts
Oligochaete symbionts
Escherichia coli Nematode symbionts
Coxiella burnetii
Solemva reidi symbiont Methanotrophic symbionts
GAMMA JTB254 Epsilon symbionts
86 Achromatium oxaliferum
54 Inanidrilus leukodermatus symbiont
100 Olavius algarvensis gamma symbiont
65 Laxus sp symbiont
Olavius loisae gamma symbiont
82 Solemva relum symbiont
Solemva occidentalis symbiont
Lucinella nassula symbiont
97 Codakia orbicularis symbiont
98 Lucina floridana symbiont Chemoautotrophic
57 Codakia costata symbiont symbionts
68 Lucinoma annulata symbiont
Lucinoma aequizonata symbiont
94 100 Ridgeia piscesae symbiont
Rifna packyptria symbiont
100 Escorpia spicata symbiont
Lamembrachia colmnua symbiont
64 93 Anodontia phillipiana symbiont
70 Solemya terraeregina symbiont
69 Thyasira flexuosa symbiont
Solemya pusillu symbiont
Lucina pectinata symbiont
80 Pseudomonas fluorescens
100 Pseudomonas putida
88 Pseudomonas lundensis
81 Pseudomonas fragi
100 Marinobacter sp CAB
Marinobacter aquaeolei
Oceanospirillum kriegii
100 Thiothrix eikelboomii
52 Thiothrix unzii
98
Thiothrix nivea
Thiothrix ramosa
54 Bathymodiolus japonicus symbiont
Bathymodiolus childressi symbiont Methanotrophic
100
Bathymodiolus platifrons symbiont symbionts
100 Bathymodiolus puteoserpentis symbiont 2
100 76 Methylobacter sp BB5.1
60 53 Methylococcus luteus
Methylococcus capsulatus BARH
61 Methyllobacter whittenburyi
100 Methylomonas aurantiaca
Methylomonas rubra
Methylococcus thermophilus
GAMMA JTB35
100 Thiomicrospira sp.
82 Thiomicrospira crunogena
82 Thiomicrospira L12
100 Thiomicrospira milos T2
Thiomicrospira milos T1
69 100 Thiomicrospira thyasirae
Thiomicrospira pelophila
Maorithyas hadalis symbiont 1
Juan de Fuca mussel symbiont
55 Bathymodiolus septemdierum symbiont
52 70 Bathymodiolus thermophilus symbiont
Bathymodiolus puteoserpentis symbiont 1
Bathymodiolus aff. brevior symbiont
62 Calyptogena magnifica symbiont
100 Calyptogena fossajaponica symbiont Chemoautotrophic
82 Calyptogena phaseoliformis symbiont symbionts
Calyptogena FL sp. symbiont
94 Calyptogena pacifica symbiont
86 Vesicomya lepta symbiont
100
87 Vesicomya chordata symbiont
100 Calyptogena elongata symbiont
100 Calyptogena kilmeri symbiont
Vesicomya gigas symbiont
Ectenagena extenta symbiont
Maorithyas hadalis symbiont 2
100 Halothiobacillus kellyi
Halothiobacillus neapolitanus Non-gamma
Olavius loisae alpha symbiont oligochaete
Olavius loisae spirochete symbiont
100 Olavius algarvensis delta symbiont symbionts
Desulfosarcina variabilis
100 Alvinella pompejana epibiont APB13B
84 Alvinella pompejana epibiont APG44B Epsilon
87 Alvinella pompejana epibiont APG5B epibionts
100 Rimicaris exoculata symbiont
Alvinella pompejana epibiont APG56B
100 Sulfurospirillum sp 18.1
Sulfurospirillum Am-N
. Fig. 23.1
Phylogeny showing the strict consensus of 46 trees obtained via parsimony analyses of 16S rRNA gene sequences (1,456 bp) from
symbiotic and free-living bacteria. Results greater than 50 % from a 500 replicate bootstrap analysis are reported above respective
branches. Chemosynthetic symbiont taxa are color coded (see key on Figure)
Marine Chemosynthetic Symbioses 23 581
History
The discovery of deep-sea hydrothermal vents and the

flourishing ecosystems associated with them significantly
advanced scientific understanding of chemosynthetic symbio-
ses. Oceanographers in the research submersible Alvin discov-
ered hydrothermal vents along the Galapagos Rift in 1977. In
stark contrast to perceptions of the deep benthos as a cold, food-
limited habitat incapable of supporting substantial biomass,
hydrothermal vents are oases, characterized by high concentra-
tions of free-living microorganisms and dense aggregations of
invertebrates (Lonsdale 1977; Grassle 1985; Van Dover 2000).
Researchers first argued that vent invertebrates achieved high
densities by filtering organic matter, which was presumably
transported to vent sites in hydrothermally driven convection
cells (Lonsdale 1977). A second hypothesis suggested that the
invertebrate community fed directly on locally dense
populations of free-living chemoautotrophic bacteria (Lonsdale
1977; Corliss et al. 1979).
However, studies of the giant tubeworm, Riftia pachyptila
(> Fig. 23.2), whose lack of mouth and gut precludes suspension
feeding, suggested that sulfide-oxidizing endosymbiotic bacteria
might contribute substantially to the vent food web. Cavanaugh
et al. (1981) proposed that symbiotic chemosynthetic bacteria
occurred in R. pachyptila. Microscopic and biochemical evi-
dence indicated Gram-negative bacteria were present, packed
. Fig. 23.2
within the tubeworm trophosome (Cavanaugh 1981;
Symbiont-containing host organisms from hydrothermal vents
> Fig. 23.3), a highly vascularized organ in the tubeworm
and cold seeps. (a) Calyptogena magnifica shell (Courtesy of Emilio
trunk, in which activities of enzymes involved in sulfide oxida-
Jorge Power). (b) Filamentous bacteria on sulfide deposits from
tion and carbon fixation were also detected (Felbeck 1981). In
the East Pacific Rise. (c) Bathymodiolus childressi from the Gulf of
addition, Rau (1981) used stable isotope signatures to show
Mexico (Courtesy of the National Oceanic and Atmospheric
a nonphotosynthetic carbon source for R. pachyptila, implying
Administration). (d) Riftia pachyptila at the East Pacific Rise
a role for chemoautotrophy in tubeworm metabolism. Follow-
ing confirmation of a chemosynthetic endosymbiosis within the
giant tubeworm, researchers questioned the putative reliance on
filter feeding by other vent invertebrates. Ultimately, anatomical, and in vent and seep mussels, sometimes co-occurring with
enzymological, and isotopic analyses revealed the presence of sulfur-oxidizing chemoautotrophs in a ‘‘dual symbiosis’’
sulfur-oxidizing bacterial symbionts either within the tissues (Childress et al. 1986; Cavanaugh et al. 1987; Cavanaugh 1993).
(endosymbiotic) or attached to the surfaces (episymbiotic) of
most vent taxa, including vesicomyid clams, mytilid mussels,
shrimp, and polychaete worms (Cavanaugh 1994; Nelson and Habitat Chemistry
Fisher 1995; Van Dover 2000).
Recognizing the ubiquity of chemoautotrophic symbioses at The seemingly disparate ecological niches where these symbioses
hydrothermal vents, researchers searched for and discovered are found all possess a chemical gradient, or chemocline, which
similar symbiotic associations in other marine habitats, includ- chemosynthetic bacteria exploit for energy production.
ing coastal and subtidal reducing sediments (e.g., Felbeck et al. Chemoclines form where reduced, high-energy compounds
1981; Southward et al. 1981; Southward 1982; Cavanaugh 1983; such as sulfide or methane (typically produced in anoxic habi-
Giere 1985; Bauer-Nebelsick et al. 1996), brine and hydrocarbon tats, including vent fluids and sediments) underlie an oxic water
seeps (Sibuet and Olu 1998), and whale skeletons (Bennett et al. column. Chemosynthetic microorganisms must bridge the oxic-
1994; Smith and Baco 2003), thereby extending the host taxa to anoxic interface to access both the reduced compounds (e.g.,
include solemyid and lucinid bivalves, pogonophoran H2S) used as an energy source and the oxygen to which electrons
tubeworms, echinoids, and ciliates. In addition, methanotrophic are shuttled in aerobic energy metabolism.
bacteria were detected in a marine sponge (Vacelet et al. 1995), The source of reduced compounds used in chemoautotro-
a pogonophoran tubeworm (Schmaljohann and Flügel 1987), phic energy metabolism differs among habitats. In marine
a d
b m
tc
c e
pm
om
cm
. Fig. 23.3
Riftia pachyptila Jones (a–c: Galapagos Rift; d, e: 21 N, East Pacific Rise). (a) Photograph showing elemental sulfur crystals (arrows)
scattered throughout trophosome (Courtesy of M. L. Jones). (b) Scanning electron micrograph showing lobules of trophosome; arrow
indicates area of c (below) where surface epithelium was removed to reveal symbionts within trophosome. (c) Same, higher
magnification, showing symbionts within trophosome; note spherical cells as well as rod-shaped cells (small arrows); large arrows
indicate likely host cell membranes. (d) Cross section of portion of trophosome lobule showing variable fine structure of symbionts,
including membrane-bound vesicles in many cells; all symbionts contained within membrane-bound vacuoles, either singly or in groups
of two or more; arrow: dividing bacterium; b bacteria, m mitochondria, tc trunk coelomic cavity. (e) Same, higher magnification, showing
cell envelope of symbiont (resembling that of Gram-negative bacteria), intracytoplasmic vesicles, and peribacterial membrane; v vesicle,
cm symbiont cytoplasmic membrane, om symbiont outer membrane, pm peribacterial membrane. Scale bars: a, 1 mm; b, 250 mm;
c, 10 mm; d, 3 mm; and e, 0.2 mm (From Cavanaugh (1985), with permission)
sediments microbial sulfate reduction (in which SO4m2 is used as Methods for Studying Chemosynthetic Symbioses
an electron acceptor during the oxidation of organic matter)
dominates, and the sulfide utilized by thioautotrophic To date, the bacteria involved in these symbiotic associations
symbioses (e.g., involving lucinid clams or solemyid have not yet been isolated and grown in pure culture—perhaps
protobranchs) in these habitats is of biogenic origin. In contrast, because the unique environment encountered in situ by chemo-
sulfide at hydrothermal vents is produced by the geothermal synthetic symbionts has not been recreated or because a reduced
reduction of seawater sulfate and by the interaction between genome, characteristic of many endosymbionts, has precluded
geothermally heated water and sulfur-containing rocks growth outside of the host. Symbiotic bacteria are therefore
(e.g., basalt; Alt 1995; Elderfield and Schultz 1996; Rouxel et al. studied indirectly, using methods to assess their physiology,
2004). Seawater that percolates into the developing crust ecology, and phylogeny within the context of the intact symbi-
becomes heated and reacts with oceanic basalt, becoming osis. Traditionally, researchers identify chemosynthetic symbio-
enriched with metals and sulfide and charged with volcanic ses using a combination of microscopy (light, confocal,
gases such as methane and carbon dioxide. Heated vent water scanning, and transmission electron), which provides visual
then exits with concentrations of reduced compounds orders of information on the location, morphology, and ultrastructure
magnitude higher than in ambient seawater. Hydrothermal of symbionts, and enzyme assays, which detect and quantify
effluent is hot (temperatures up to 350–400 C), acidic the activity of key proteins involved in chemoautotrophic (e.g.,
(pH 3) and anoxic and can contain 3–12 mmol/kg of H2S, ribulose 1,5-bisphosphate carboxylase-oxygenase) or
25–100 mmol/kg of CH4, and 0.05–1 mmol/kg of H2, as well as methanotrophic (e.g., methanol dehydrogenase) metabolism.
360–1,140 mmol/kg of Mn and 750–6,500 mmol/kg of Fe In addition, tracing the incorporation of radiolabeled substrates
(Elderfield and Schultz 1996). As it exits the seafloor (e.g., carbon dioxide, methane, and nitrogen species) within the
and mixes with the ambient bottom oxygenated seawater host helps define the physiology of the host-bacteria partner-
(pH, ca. 8; temperature = 1.8 C; [O2], ca. 110 mM), metallic ship. Such physiological assays, in conjunction with the analyses
sulfides precipitate out resulting in ‘‘black smokers’’ (reviewed in of stable isotope signatures of symbiont-containing and
Elderfield and Schultz 1996). Vent organisms are typically found symbiont-free host tissue, provide valuable insight into the
clustered around more diffuse or low flow vents, which trophic dynamics of symbiont-based communities. Molecular
are caused by ambient seawater mixing in the shallow subsurface approaches, such as polymerase chain reaction (PCR)-based
with vent fluid. These vents are characterized by a higher gene probing, 16S rRNA gene analysis, and fluorescent in
pH (ca. 6), lower temperatures (1.8 to ca. 40 C), and, conse- situ hybridization (FISH), are increasingly used to characterize
quently, lower concentrations of reduced chemicals the systematic relationships of symbiont and host species
(Van Dover 2000). (e.g., Distel et al. 1995; Peek et al. 1998; Dubilier et al. 1999)
The relative acidity of vent fluid (black smokers—pH, ca. 3; and the metabolism and gene flow of the bacterial symbionts
diffuse flow vents—pH, ca. 6) significantly impacts the concen- (Robinson et al. 1998; Lee et al. 1999; Millikan et al. 1999; Podar
tration of inorganic chemicals available to chemosynthetic sym- et al. 2002). Molecular techniques have also been used to detect
bioses. For example, carbon dioxide (CO2), bicarbonate symbiont transmission modes (Cary and Giovannoni 1993;
(HCO3m), and carbonate (CO3m2), the three distinct chemical Krueger 1996) as well as symbiont abundance (Polz and
species of the dissolved inorganic carbon (DIC) pool, vary in Cavanaugh 1995).
relative abundance depending on pH; pKa values for these com-
pounds are 6.4 for CO2:HCO3m and 10.3 for HCO3m:CO3m2 at
25 C (Stumm and Morgan 1996). Thus, CO2, which diffuses Summary
freely across biological membranes and is the DIC species fixed
by chemoautotrophic symbionts utilizing the Calvin cycle, is This chapter reviews symbiotic associations between chemosyn-
readily available at vents (Cavanaugh and Robinson 1996; thetic bacteria and marine invertebrate and protist hosts. A bias
Goffredi et al. 1997b). In addition, sulfide exists at three levels toward symbioses between chemoautotrophic bacteria and
of dissociation (H2S, HSm, and Sm2) depending on pH, with pKa hydrothermal vent invertebrates is evident, primarily because
values of 7.0 for H2S:HSm and 12.9 for HSm:Sm2 at 25 C (Stumm our knowledge of marine bacterial symbioses stems largely from
and Morgan 1996). Therefore, in the relatively acidic vent fluids, studies of vent fauna done in the 27 years following the discovery
sulfide occurs predominantly as H2S. In such effluent, total of these unique organisms. But despite this impressive body of
sulfide (H2S, HSm, and Sm2) concentration correlates positively research, much about these marine symbioses remains to be
with temperature (Johnson et al. 1988); conversely, higher tem- revealed. In conjunction with several earlier reviews that provide
peratures (>30 C) may facilitate reactions between sulfide and a thorough and thoughtful treatment of symbioses occurring at
other dissolved elements (such as iron) that reduce free sulfide hydrothermal vents and cold seeps (Fisher 1990; Felbeck and
availability (Luther et al. 2001). The chemical environment (i.e., Distel 1991; Childress and Fisher 1992; Cavanaugh 1994; Nelson
concentrations of CO2, O2, H2S, CH4, H+, and dissolved metals) and Fisher 1995), the following chapter presents an overview of
is therefore expected to significantly influence the ecology and the ecology, physiology, and evolution of chemosynthetic
evolution of chemosynthetic symbioses. symbioses.
Host Diversity In other families, such as the Mytilidae, chemoautotrophic sym-

bionts have been detected only in members of the subfamily
Chemosynthetic bacteria are known to associate with a diversity Bathymodiolinae, which are found exclusively in the deep sea.
of invertebrate hosts (six phyla) as well as with ciliate protists Further, dual symbioses involving both methanotrophs and
(> Table 23.1 and references therein). To date, the majority of chemoautotrophs are restricted to species of deep-sea
the symbionts characterized via 16S rRNA phylogenetic analyses bathymodioline mussels collected from methane seeps and
fall within the Gammaproteobacteria division (> Fig. 23.1; hydrothermal vents (Cavanaugh 1994; Nelson and Fisher 1995;
discussed further below). The intimacy of these associations Van Dover 2000; > Fig. 23.14).
varies among taxa. The bacterial partners may be episymbionts Chemosynthetic bacteria also occur as episymbionts on
living on the surface of the host (e.g., on shrimp, nematodes, marine invertebrates (> Table 23.1; > Fig. 23.4). These symbi-
sponges, limpets, and ciliates; > Figs. 23.4–> 23.6) or endosym- onts include the Epsilonproteobacteria that cover the cuticle of
bionts living either extracellularly within host tissue (e.g., in Rimicaris shrimp, dominant members of the metazoan fauna at
oligochaetes; > Fig. 23.7) or in specialized host cells and organs vents on the Mid-Atlantic Ridge (MAR; Polz et al. 1998) and the
(e.g., in bivalves and vestimentiferan tubeworms; > Figs. 23.2, Central Indian Ridge (CIR; Van Dover et al. 2001; Van Dover
> 23.3, and > 23.8). In intracellular endosymbioses, the symbi- 2002), and the surfaces of alvinellid polychaetes (Desbruyères
onts are housed within specialized host cells called et al. 1985; Cary et al. 1997). Chemosynthetic episymbionts also
‘‘bacteriocytes’’ and are contained within a host-derived mem- associate with nematodes (Weiser 1959; Ott et al. 1991; Polz et al.
brane-bound vacuole (Cavanaugh 1983, 1994; Fisher 1990). 1992, 1994) and ciliates (e.g., Fenchel and Finlay 1989; Bauer-
Host morphology clearly suggests a nutritional benefit from Nebelsick et al. 1996). In addition, methanotrophic
these intimate associations. Indeed, as in the giant vent episymbionts have been found living on a deep-sea sponge
tubeworms, the digestive system of many endosymbiont- (Vacelet et al. 1995, 1996). Vent limpet-bacteria associations
containing marine invertebrates is either reduced (e.g., in coastal seem to be an intermediate between epi- and endosymbioses;
solemyid protobranchs) or absent altogether (e.g., in oligo- bacteria exist partially embedded in the limpet gill epidermis
chaetes and vestimentiferan and pogonophoran tubeworms), and may be endocytosed or fed on by the host (de Burgh and
consistent with host dependence on the symbiont for part or Singla 1984; Bates et al. 2004; > Fig. 23.5). Some epibiont com-
all of its nutrition. munities, like those residing on the Rimicaris shrimp and the
All of the members of the tubeworm family Siboglinidae nematode Laxus sp., are dominated by a single phylotype (Polz
examined to date, including the vestimentiferan and the smaller et al. 1994; Polz and Cavanaugh 1995), while others are quite
pogonophoran tubeworms, contain intracellular symbionts. diverse (Polz et al. 1999; Campbell et al. 2003). But given that
Most of these symbionts are inferred to be chemoautotrophic, morphological plasticity often belies the phylogenetic identity of
but methanotrophs have been found in one host species symbionts (Polz et al. 1999; Giere and Krieger 2001), symbiont
(Siboglinum poseidoni; Schmaljohann and Flügel 1987). The diversity estimates are only appropriate when putative
vent tubeworm Riftia pachyptila and other vestimentiferan and symbiont phylotypes are confirmed using hybridization
pogonophoran tubeworm species possess a unique morpholog- methods (e.g., FISH).
ical adaptation to accommodate their symbionts. Tubeworm
bacteria reside within a lobular and highly vascularized organ
(the trophosome) that occupies most of the tubeworm trunk Symbiont Diversity
and functions specifically to house bacteria (Cavanaugh 1981;
Felbeck 1981; > Figs. 23.3 and > 23.9). The symbiosis is obligate Morphology and Ultrastructure
for these worms, as they are mouthless and gutless as adults and
depend on their internal bacteria for their nutrition. Symbiont morphology varies among functional types
Chemosynthetic symbioses are widespread within the (chemoautotroph vs. methanotroph), among phylotypes
Mollusca and have been detected in five bivalve and two gastro- within the same functional group, and among individuals in
pod families (> Table 23.1). In mollusk symbioses, the bacteria a population of a single phylotype. The symbionts all
occur only in the gills; bacteria have been found within gill have a Gram-negative cell envelope but range from small
epithelial cells of solemyid protobranchs (Cavanaugh 1983; (ca. 0.25 mm diameter) coccoid endosymbionts within mussel
Krueger et al. 1996; > Figs. 23.10–> 23.13), lucinid clams gills (Cavanaugh 1985; Dubilier et al. 1998) to large (>10 mm
(Cavanaugh 1983; Felbeck 1983), thyasirid clams (Felbeck et al. length) rod-shaped and filamentous episymbionts on
1981; Cavanaugh 1983; Arp et al. 1984), vesicomyid clams (Boss vent shrimp (Hentschel et al. 1993; Polz and Cavanaugh
and Turner 1980; Rau 1981; Arp et al. 1984; > Fig. 23.8), mytilid 1995; > Fig. 23.4). Some bathymodioline mussels host
mussels (Fiala-Médioni 1984; Le Pennec and Hily 1984; two metabolically distinct symbionts: small (<0.5 mm)
> Figs. 23.2 and > 23.14), and provannid gastropods (Stein chemoautotrophs and larger (1.5–2.0 mm) methanotrophic bac-
et al. 1988; Windoffer and Giere 1997). Within certain mollusk teria possessing stacked intracytoplasmic membranes, which
families (e.g., Solemyidae, Lucinidae, and Thyasiridae), all spe- are typical of type I methanotrophs (e.g., Childress et al.
cies examined form symbioses with chemoautotrophic bacteria. 1986; Cavanaugh et al. 1987, 1992; Fiala-Médioni et al. 2002;
. Table 23.1
List of invertebrate taxa hosting chemoautotrophic or methanotrophic bacterial symbiontsa
Symbiont-
Common containing
Group name tissue Location Habitat Symbiont type References
Protozoa
Class Ciliata Ciliate NA Epibiotic Cold seeps, Chemoautotroph Bauer-Nebelsick
mangrove et al. (1996)
swamp Ott et al. (1998)
Buck et al. (2000)
Fenchel and Finlay
(1989)
Porifera
Class Demospongiae Sponge NA Extracellular Cold seeps Methanotroph Vacelet et al. (1995,
1996)
Family Cladorhizidae
Nemata
Subfamily Nematode Cuticle Epibiotic Reducing Chemoautotroph Schiemer et al.
Stilbonematinae sediments (1990)
Polz et al. (1992)
Mollusca
Class Bivalvia
Subclass Protobranchia
Family Solemyidae Clam Gills Intracellular Reducing Chemoautotroph Cavanaugh (1983)
sediments, Fisher and Childress
hydrothermal (1986)
vents,b cold
Conway et al. (1989)
seepsc
Subclass Heterodonta
Family Lucinidae Clam Gills Intracellular Reducing Chemoautotroph Giere (1985)
sediments, Schweimanns and
cold seepsc Felbeck (1985)
Family Thyasiridae Clam Gills Intracellular Reducing Chemoautotroph Dando and
sediments, Southward (1986)
cold seepsc Herry and Le Pennec
(1987)
Family Vesicomyidae Clam Gills Intracellular Hydrothermal Chemoautotroph Cavanaugh (1983)
vents, cold Rau (1981)
seeps
Subclass Pteriomorphia
Family Mytilidae Mussel Gills Extracellular Hydrothermal Chemoautotroph and/or Childress et al.
vents, cold methanotroph (1986)
seeps Cavanaugh et al.
(1987)
Intracellular Fisher et al. (1988)
Cavanaugh et al.
(1992)
Class Gastropoda
Family Provannidae Snail Gills Intracellular Hydrothermal Chemoautotroph Stein et al. (1988)
vents Endow and Ohta
(1989)
Symbiont-
Common containing
Group name tissue Location Habitat Symbiont type References
Family Lepetodrilidae Limpet Gills Epibiotic Hydrothermal Chemoautotroph de Burgh and Singla
vents (1984)
Fox et al. (2002)
Bates et al. (2004)
Annelidad
Class Polychaeta Worm Dorsal surface Epibiotic Hydrothermal Chemoautotroph Desbruyères et al.
vents (1983, 1985)
Family Alvinellidae Cary et al. (2003)
Family Siboglinidae Tubeworm Trophosome Intracellular Deep-sea Chemoautotroph Cavanaugh et al.
(Vestimentiferan and hydrothermal (1981)
pogonophoran) vents, cold Felbeck (1981)
seeps,
Southward et al.
reducing
(1981)
sediments,
fjords Brooks et al. (1987)
Schmaljohann and
Flügel (1987)
Southward and
Southward (1988)
de Burgh et al.
(1989)
Class Clitellata
Subfamily Phallodrilinae Oligochaete Subcuticular Extracellular Coralline sands Chemoautotrophe Felbeck et al. (1983)
Giere (1981, 1985)
Giere and Langheld
(1987)
Arthropoda
Class Crustacea
Family Alvinocarididae Shrimp Carapace Epibiotic Hydrothermal Chemoautotroph Van Dover et al.
vents (1988)
Polz and Cavanaugh
(1995)
Polz et al. (1999)
Echinodermata
Class Echinoidea Sea urchin Gut Extracellular Reducing Chemoautotrophf Temara et al. (1993)
sediments Brigmon and De
Ridder (1998)
Abbreviation: NA not applicable
a
Chemosynthetic status of symbionts inferred from ultrastructural, physiological, enzymatic, and molecular data
b
A solemyid protobranch, Acharax alinae, has been described from the Lau Basin hydrothermal vents, but the symbiosis has not been characterized (Metivier and
Voncosel 1993)
c
Solemyid clams have been collected from cold seeps in the eastern Pacific, and lucinid and thyasirid clams have been collected from cold seeps in the Gulf of
Mexico, Sagami Bay, and Barbados Prism, though the presence of symbionts has not been formally described (Sibuet and Olu 1998)
d
Though it is now accepted that the pogonophoran and vestimentiferan worms are not separate phyla but members of phylum Annelida, they are listed as
separate groups for identification purposes
e
The oligochaete Olavius algarvensis has been shown to have an additional symbiont which is a sulfur-reducing bacterium and member of the delta
Proteobacteria (Dubilier et al. 2001). Olavius loisae has been shown to host an alpha proteobacterium and a spirochete as well as a chemoautotroph (Dubilier
et al. 1999)
f
This symbiont has been described as Thiothrix-like on the basis of morphology, physiology and immunological assays (Temara et al. 1993; Brigmon and de Ridder
1998)
a b
100 µm 10 µm
c d
1000 µm 100 µm
e f
5 µm 1 µm
g h
1 µm 10 µm
. Fig. 23.4
Scanning electron micrographs showing the morphological diversity of ectosymbiotic bacteria on the colonial ciliate Zoothamnium
niveum (a, b), the shrimp Rimicaris exoculata (c, d), and the nematodes within the subfamily Stilbonematinae (e–h). (a) Entire ciliate
colony with zooids attached to a common stem and (b) bacterial epigrowth on an individual zooid. (c) Shrimp appendage covered by
dense arrays of filamentous bacteria and (d) detail of the hairlike bacterial covering. Epigrowth on different species of nematodes
showing (e) irregular epigrowth of two morphological types on Robbea sp.; (f) coccoid bacteria forming multilayers on Stibonema sp.;
(g) upright standing, longitudinally dividing rods on Laxus oneistus; and (h) dense array of nonseptate filaments that can reach up to
100 mm in length on Eubostrichus dianae (From Polz et al. (2000), with permission)
Pimenov et al. 2002; > Fig. 23.14). Morphological diversity can Symbiont Phylogeny
also occur throughout monospecific populations within a single
host animal. For example, populations of sulfur-oxidizing che- While chemoautotrophic symbionts have consistently evaded
moautotrophic symbionts in the tubeworm Riftia pachyptila culture, the suite of cellular and molecular methods used to
contain distinct morphotypes that vary in abundance depending characterize these bacteria has revealed startling evolutionary
on location within the trophosome lobule (Bright et al. 2000); trends. Investigators have successfully sequenced 16S rRNA
small, rod-shaped bacteria occur primarily in the innermost genes from symbiont-containing tissue and subsequently con-
zone of the lobule nearest the host’s axial blood vessel, firmed the symbiont origin of these sequences via hybridization
while small and large cocci (1.6–10.7 mm diameter) occupy with symbiont-specific probes. In contrast to the wide diversity
zones nearer the periphery of the trophosome (Bright et al. of host taxa involved in these symbioses, chemosynthetic
2000). Such variability may relate to differences in the life symbionts cluster primarily within a single bacterial division,
cycle stage and metabolism among symbiont cells (Bright the Gammaproteobacteria, on the basis of 16S rRNA gene
et al. 2000). sequences (Distel and Cavanaugh 1994; Dubilier et al. 1999;
McKiness 2004). Such analyses have also shown that most host methanotrophs, and marine bacterioplankton. Here, an
species typically form a relationship with a unique symbiont expanded phylogenetic analysis is presented that includes the
phylotype. But this clearly is not always the case. For example, Epsilonproteobacteria symbionts of shrimp and alvinellid
while strain level variation may occur, vestimentiferan worms (> Fig. 23.1). This consensus tree illustrates the strong
tubeworms belonging to the genera Riftia, Tevnia, and Oasisia level of resolution afforded by the 16S rRNA gene and shows that
appear to share a single, or very similar, symbiont phylotype almost all of the chemosynthetic symbionts for which sequence
based on 16S rRNA gene sequences (Feldman et al. 1997; Laue data are available cluster into two main clades. The first clade
and Nelson 1997; Di Meo et al. 2000; Nelson and Fisher 2000; includes symbionts of lucinid and thyasirid clams, solemyid
McMullin et al. 2003), as do some species of tropical lucinid protobranchs, tubeworms, nematodes, and oligochaetes, and
clams (Durand and Gros 1996; Durand et al. 1996). the second clade includes the mytilid mussel and vesicomyid
Recently, McKiness (2004) reported phylogenetic analyses of clam symbionts.
16S rRNA gene sequences from chemosynthetic symbionts Though the first clade as a whole has relatively low bootstrap
within the Gammaproteobacteria. This study represented the support, smaller clusters within the first clade are strongly
most comprehensive analysis of chemoautotrophic symbionts to supported. For example, the monophyletic cluster of nematode
date. It included 39 symbiont sequences and over 30 sequences and oligochaete symbionts has 100 % bootstrap support. Simi-
from free-living bacteria representatives of chemoautotrophs, larly, the vestimentiferan tubeworm symbiont clade also has
high bootstrap support, corroborating prior evidence that
these worms share a single or very similar symbiont phylotype,
which is consistent with environmental transmission of symbi-
onts (Feldman et al. 1997; Laue and Nelson 1997; Di Meo et al.
2000; Nelson and Fisher 2000; McMullin et al. 2003; see the
section > ‘‘Ecology and Evolution’’ in this chapter). In contrast,
the clam symbionts exhibit a more complicated relationship.
The lucinid clam symbionts form a paraphyletic group; some
n
are sister to the tubeworm symbionts, while others group with
bs thyasirid and solemyid symbionts. The solemyid symbionts
es show similarly complicated relationships, as they are polyphy-
letic and scattered throughout the first clade. Solemya velum and
b S. occidentalis symbionts cluster with the nematode and oligo-
chaete symbionts, while S. terraeregina and S. pusilla symbionts
10 µm cluster with lucinid and thyasirid clam symbionts. Thus, this
disjointed distribution does not suggest cospeciation between
. Fig. 23.5 host taxa and symbionts in this first clade and indicates that
Lepetodrilus fucensis. Transverse section of gill tissue from the there were multiple initiations of symbiosis within the solemyid
hydrothermal vent limpet showing episymbiotic filamentous and lucinid clams.
bacteria partially embedded in the host epithelium. b bacteria, es The position of the S. reidi symbiont within this first main
extracellular space, n nucleus, bs blood space (Courtesy of group is curious; this symbiont falls at the base of this first main
Amanda Bates) group, clustering with an intracellular pathogen, Coxiella
a b
. Fig. 23.6
Eubostrichus cf. parasitiferus. Scanning electron micrographs showing the symbiotic bacteria on the surface of the nematode. (a) Anterior
(bottom) and posterior (top) end with symbionts arranged in a characteristic helix. (b) Higher magnification. Bacteria are attached with
both ends to the worm’s cuticle. Note the increasing length of the cells from proximal to distal along the worm’s surface. Scale bars: a,
20 mm; b, 2 mm (From Polz et al. (1992), with permission)
a a mv
CU ni
S
m
C
b
nb
b . Fig. 23.8
Calyptogena magnifica Boss and Turner (21 N East Pacific Rise).
CU
(a) Transmission electron micrograph of slightly oblique
transverse section of gill filament, showing coccoid-shaped
bacteria within gill bacteriocyte and intercalary cells lacking
symbionts; b bacteria, mv microvilli (of both cell types), nb nucleus
of bacteriocyte, ni nucleus of intercalary cell. (b) Same, higher
magnification, transverse section of coccoid-shaped symbionts,
showing cell ultrastructure typical of Gram-negative bacteria and
peribacterial membrane (arrow). Scale bars: a, 5 mm; b, 0.25 mm
(From Cavanaugh (1985), with permission)
. Fig. 23.7
Inanidrilus leukodermatus. (a) Light micrograph of a cross section a b
of an oligochaete worm. (b) Transmission electron micrograph of
symbiont-containing region just below the cuticle. Note smaller b
and larger symbiont morphotypes (smaller and larger arrows,
respectively). c coelomic cavity, m muscle tissue, s symbiont-
containing region between cuticle and epidermis, cu cuticle. Scale
nb
bars: a, 20 mm; b, 2 mm (From Dubilier et al. (1995), with
permission)
burnetti, and an environmental clone from a Japanese cold seep, nb

Gamma JTB254 (discussed further below). This symbiont
sequence has held a basal position in other analyses (see Bayesian
analysis in McKiness 2004) and provokes questions concerning . Fig. 23.9
the nature of symbiosis in protobranch bivalves. As additional Escarpia spicata Jones (San Clemente Fault). (a) Transmission
sequences become available, it will be necessary to reassess the electron micrograph, portion of trophosome lobule showing
position of the S. reidi symbiont with respect to other chemo- numerous coccoid- to ovoid-shaped bacterial symbionts, some of
autotrophic symbionts. which appear intracellular. (b) Same, higher magnification,
The second clade of symbionts, which includes the mytilid showing bacterial cell envelope (resembling that of Gram-
mussel and vesicomyid clam symbionts from vents and cold negative bacteria) and intracytoplasmic membrane-bound
seeps, shows 100 % bootstrap support (support for the mussel vesicles; arrow: peribacterial membrane; b bacteria, nb nucleus of
and vesicomyid clades being 70 % and 99 %, respectively). The bacteriocyte. Scale bars: a, 10 mm; b, 1 mm (From Cavanaugh
symbiont from the Central Indian Ridge mussel, Bathymodiolus (1985), with permission)
a
mv
i b
bl
ci
b
. Fig. 23.10
Solemya sp. (right hand) collected from deep-sea vent sites
(2,380 m depth) along the subduction zone off Oregon and cm
om
Solemya velum (left hand) collected from subtidal reducing
sediments (<1 m depth, mean low tide) of Massachusetts eelgrass p
beds (Photo courtesy of Dr. Ruth D. Turner)
. Fig. 23.12
Solemya borealis. (a) Transverse section of gill filaments showing
intracellular rod-shaped bacteria (arrows, rectangle). Bacteriocytes
are confined to the region proximal to the ciliated edge of the gill
and are flanked by symbiont-free intercalary cells that appear to
comprise the microvillar surface of the gill filament. Light
micrograph. b bacteriocyte nucleus, c ciliated cell nucleus, i
intercalary cell nucleus, bl blood space, ci cilia, mv microvilli. (b)
Higher magnification of symbionts showing cell ultrastructure
typical of Gram-negative. Inset: detail of symbiont cell envelope
and peribacterial membrane. p peribacterial membrane, cm cell
membrane, om outer membrane. Scale bars: a, 20 mm; b, 1 mm;
inset, 0.05 mm (From Conway et al. (1992), with permission)
PLATE 3. SOLEMYA
The symbionts of the thyasirid clam Maorhithyas hadalis
. Fig. 23.11
occupy unique positions in this phylogenetic framework. On
Solemya velum. Characteristic Y-shaped burrows dug by the
the basis of 16S rRNA sequence data and in situ hybridization,
coastal protobranch clam to bridge the oxic-anoxic interface and
Fujiwara et al. (2001) described two different symbionts within
access both reduced sulfur (from below) and oxygen (from above)
this clam. One of the symbionts shows evolutionary relatedness
(From Stanley (1970), with permission)
to the bathymodioline mussel and vesicomyid clam symbionts,
occurring basal to the clade containing these bacteria, and the
other symbiont clusters with the free-living Thiomicrospira
aff. brevior, falls at the base of the vesicomyid clam symbionts, spp. This free-living ‘‘symbiont’’ phylotype, however, may be
with 82 % bootstrap support. In contrast to the first symbiont a contaminant. Difficulties with in situ hybridization have pre-
clade (> Fig. 23.1, top), evidence suggests that an ancestral cluded attempts to describe the microdistribution of the two
symbiont initiated symbioses with both the vesicomyid clams symbiont types (Fujiwara and Uematsu 2002), bringing into
and the bathymodioline mussels and predated the split between question the phylogenetic identity of the clam symbionts.
these bivalve lineages. On the basis of 16S rRNA gene sequence The phylogenetic positions of the methanotrophic endo-
data, the divergence of the clam and mussel symbionts has been symbionts and the filamentous epibionts are shown also in
dated to 125–300 million years ago (Mya). This is corroborated > Fig. 23.1. The methanotrophic symbionts characterized to
by the fossil record, which dates the bathymodioline mussel date all belong to the Gammaproteobacteria, forming a clade
hosts to 150 Mya and the vesicomyid clams to 100 Mya (Distel with 100 % bootstrap support. The sister group of this clade
1998). consists of free-living type I methanotrophs (Methylococcus,
a b
m
ni
nb
mv
. Fig. 23.13
Solemya velum. (a) Transmission electron micrograph, transverse section of gill filament, showing rod-shaped bacteria within gill
bacteriocyte and intercalary cells lacking symbionts; b bacteria, mv microvilli, nb nucleus of bacteriocyte, ni nucleus of intercalary cell. (b)
Same, higher magnification, transverse section of rod-shaped bacterium, showing cell ultrastructure typical of Gram-negative bacteria
and peribacterial membrane (arrows). Scale bars: a, 3 mm; b, 0.2 mm (From Cavanaugh (1985), with permission)
Methylobacter, and Methylomonas spp.). Given their monophyly, a freshwater sulfur-oxidizing bacterium, falls out with the nem-
the mussel symbionts apparently arose from a common ances- atode and oligochaete symbionts clade. Achromatium oxaliferum
tor. But the question of whether these symbionts subsequently occurs in freshwater sediments along the redox zone where it has
cospeciated with their hosts remains unanswered. access to sulfide and oxygen (Head et al. 1996; Glockner et al.
The episymbionts in this analysis include the sulfur- 1999; Gray et al. 1999). As cultivation methods improve and
oxidizing Epsilonproteobacteria found on the Mid-Atlantic sequences are added to the 16 rRNA gene database, other free-
Ridge shrimp Rimicaris exoculata and the eastern Pacific poly- living bacteria that are closely related to symbionts will likely be
chaete Alvinella pompejana. Interestingly, the shrimp epibiont identified. Indeed, recent studies incorporating 16S rRNA gene
clusters with the polychaete epibionts despite the fact that R. sequences from unidentified environmental clones into phylo-
exoculata occurs on the Mid-Atlantic Ridge, while alvinellid genetic analyses of free-living and symbiotic bacteria suggest
polychaetes inhabit vents in the eastern Pacific Ocean. that chemosynthetic symbionts may in fact resolve into three
Free-living microorganisms can potentially provide insight distinct clades (N. Dubilier, personal communication;
into the ancestral form of endosymbionts. For instance, the Duperron et al. 2004). Free-living relatives of chemosynthetic
evolution of insect endosymbionts (e.g., Wolbachia and symbionts should reveal much about the ecological and evolu-
Buchnera spp.) is commonly studied by comparative analyses tionary constraints on the symbiont as well as about the poten-
with free-living, closely related microbes (Wernegreen 2002; tial for gene loss during the transition from the free-living to the
Moran 2003). But until recently, the chemoautotrophic symbiotic state.
symbiont clades have not included any free-living bacteria.
> Figure 23.1 includes two species of bacteria that are not che-
moautotrophic symbionts (Coxiella burnetti and Achromatium Symbiont Characterization

oxaliferum) and an environmental clone (JTB254). All three of
these sequences fall within the first symbiont clade. Coxiella Enzyme Activities
burnetii is an intracellular pathogenic bacterium (Woldehiwet
2004) and the Gamma JTB254 clone was recovered from a deep- Researchers routinely demonstrate chemoautotrophy or
sea cold seep in the Japan Trench (Li et al. 1999). They both fall methanotrophy in symbionts by the activity or presence of
in a cluster with the S. reidi symbiont. Both C. burnetii and the S. diagnostic enzymes. Indeed, given the inability to culture che-
reidi symbionts are able to maintain an intracellular existence in moautotrophic symbionts, detection of such enzymes is often
eukaryotic hosts. Coxiella burnetii, however, is capable of growth the only evidence used to infer symbiont metabolism. This
in animal cell lines (e.g., Woldehiwet 2004) and pathogenically characterization often involves physiological assays using tissue
infects a wide range of hosts (Niemczuk and Kondracki 2004; or purified protein extract, immunodetection, or PCR-based
Watanabe 2004; Woldehiwet 2004). In addition, A. oxaliferum, gene probing. Such studies were initially conducted on the
a C B phosphoribulose kinase (PRK) as well as enzymes associated

with the oxidation of reduced inorganic sulfur compounds.
The activity or presence of RubisCO has subsequently been
used to diagnose symbiont autotrophy in a diversity of host
species, including all of the chemoautotroph-harboring inverte-
brates listed in > Table 23.1 (excluding alvinellid polychaetes;
I
S
see below): shallow water solemyid and lucinid bivalves, vent
and seep tubeworms and bivalves (including mussels hosting
both methanotrophic and chemoautotrophic symbionts), nem-
atodes, oligochetes, shrimp, sea urchins, and ciliates (Felbeck
et al. 1981; Cavanaugh 1983; Cavanaugh et al. 1988; Polz et al.
1992; Johnson et al. 1994; Nelson et al. 1995; Bauer-Nebelsick
b et al. 1996; Krieger et al. 2000; Elsaied et al. 2002; Fiala-Médioni
et al. 2002). While RubisCO has not been detected in the
alvinellid epibionts, genes encoding citrate lyase, a key enzyme
of the reductive tricarboxylic acid (TCA) cycle, recently have been
detected via analyses of symbiont DNA sequences, suggesting that
the Epsilonproteobacteria symbionts of alvinellid worms fix
carbon via this pathway (Campbell et al. 2003).
Enzymes involved in chemosynthetic energy generation have
also been used to characterize these symbionts. Although the
sulfur metabolism enzymes are not unique to sulfur oxidation,
certain enzymes such as ATP sulfurylase, when detected in
high activities, have been used to infer sulfur-based
n l chemolithotrophy (Felbeck 1981; Fisher et al. 1993; Laue and
Nelson 1994). As methane monoxygenase, the enzyme that
catalyzes the first step in the oxidation of methane in aerobic
methanotrophs, is notoriously labile (Prior and Dalton 1985;
Cavanaugh 1993), methanol dehydrogenase (MeDH), the
enzyme that catalyzes the second oxidation step (i.e., methanol
to formaldehyde) and is known to occur only in methylotrophs,
has been used extensively to diagnose methanotrophy. MeDH
has been detected in gill extracts of mussels hosting
methanotrophs or both methanotrophs and thioautotrophs
(Cavanaugh et al. 1992; Fisher et al. 1993; Robinson et al.
S 1998; Fiala-Médioni et al. 2002; Pimenov et al. 2002; Barry
et al. 2002) and in a deep-sea sponge (Vacelet et al. 1996).
. Fig. 23.14
Such enzymatic evidence strongly suggests methanotrophy, par-
Bathymodiolus puteoserpentis. Transverse section of Mid-Atlantic
ticularly when coupled with ultrastructural observations show-
Ridge (MAR) mytilid gill filament, showing symbiont-containing
ing symbionts with the complex intracytoplasmic membranes
gill epithelial cells (bacteriocytes). (a) Diagram of gill filament.
that are characteristic of type I methanotrophs.
Bacteriocytes are confined to the region proximal to the ciliated
border of the gill. Small box shows positions of Figs. 21.14b. (b)
symbiont-containing bacteriocyte region and (C) symbiont-free
Stable Isotope Signatures
ciliated region. (b) Transmission electron micrograph. Large and
small symbionts (large and small arrows, respectively) are located
Carbon Isotopes
in the apical region of the cells, while nuclei and lysosomal
In addition to enzymology, stable isotope data provided some of
residual bodies occupy the region closest to the blood sinus. Note
the first evidence in support of chemoautotrophy in marine
centrally stacked intracytoplasmic membranes in large symbionts.
invertebrate-bacteria symbioses (e.g., Rau and Hedges 1979;
1 lysosomal residual body, n bacteriocyte nucleus, and s blood
Spiro et al. 1986) and continue to be useful in assessing symbiont
sinus. Scale bar: 5 mm (From Distel et al. (1995), with permission)
metabolism and tracking energy and carbon transfer in chemo-
synthetic symbioses (Colaco et al. 2002; Levin and Michener
tubeworm Riftia pachyptila. For example, Felbeck (1981) 2002; Van Dover 2002; Robinson et al. 2003; Scott et al. 2004).
assayed tubeworm trophosome tissue for the activities of key Because enzymes involved in distinct carbon fixation pathways
enzymes of the Calvin cycle, the CO2-fixing enzyme, ribulose discriminate differently against the use of the heavier carbon
1,5-bisphosphate carboxylase-oxygenase (RubisCO), and isotope (13C), the stable carbon isotope ratio comparing 13C to
12
C (d13C) can be used to help distinguish different autotrophic significantly heavier, ranging from m9‰ to m16‰ (Childress and
metabolisms. For example, whereas d13C values of marine phy- Fisher 1992; Van Dover and Fry 1994; Robinson and Cavanaugh
toplankton typically vary between m18‰ and m28‰ (Fry and 1995; Cavanaugh and Robinson 1996; Robinson et al. 2003). The
Sherr 1984; Gearing et al. 1984; Goericke et al. 1994), carbon difference between these groups relates to the form of RubisCO
derived chemosynthetically at vents is either considerably lighter used to fix CO2 by the symbionts, with form I RubisCO occur-
(enriched in 12C), with d13C values from m27‰ to m35‰, or ring in most members of the isotopically lighter group and form
heavier (depleted in 12C), with values from m9‰ to m16‰ II in all members of the heavier tubeworm group (Robinson and
(Childress and Fisher 1992; Robinson and Cavanaugh 1995; Cavanaugh 1995). Corroborating this hypothesis, Robinson
Robinson et al. 2003). Depending on the source of methane, et al. (2003) showed that the kinetic isotope effect (E value),
symbioses between mussels and methanotrophic bacteria may the relative rate of 13CO2 to 13CO2 fixation (12k/13k) and
be even more depleted in 13C, with d13C ranging from m37‰ to a measure of discrimination against 13C by the purified RubisCO
m78‰ (Cavanaugh 1993; Nelson and Fisher 1995; Barry et al. enzyme in vitro, is significantly lower for form II RubisCO from
2002). Riftia pachyptila symbionts (e = 19.5‰) than for the form I
Because consumers generally retain the carbon isotopic enzyme (e = 22–30‰). Such variation may have arisen from
signature of their food (i.e., ‘‘you are what you eat’’; DeNiro evolution under differing concentrations of CO2 and O2
and Epstein 1979), comparisons between d13C signatures of (Robinson et al. 2003).
symbiont-containing host tissue and symbiont-free host tissue But isotopic discrimination by RubisCO does not fully
can be used to study the transfer of symbiont-derived carbon to account for 13C-enrichment in these symbioses. For example,
the host. For example, d13C values (m30.8‰–m35.8‰) in sym- to explain the discrepancy in R. pachyptila biomass d13C values,
biont-containing gill tissue from the western Pacific vent mussel Scott (2003) used a mass balance model to show that steep gradients
Bathymodiolus brevior were significantly lower than values from in [CO2] among symbiont, host, and environmental pools may
symbiont-free foot tissue, suggesting that B. brevior supplements drive 13C-enrichment. RubisCO, which occurs in the symbiont
its diet via filter feeding on photosynthetically derived carbon cytoplasm, preferentially fixes 12CO2, leaving 13CO2 behind. If
(Dubilier et al. 1998). In contrast, other studies show a high fixation is rapid, CO2 equilibration between the isotopically
dependence on symbiont carbon by host species, including coastal lighter host cytoplasm and the isotopically heavier symbiont
solemyid protobranchs (Fisher and Childress 1986; Conway and cytoplasm cannot occur, causing RubisCO to draw from
Capuzzo 1991), thyasirid clams (Dando and Spiro 1993; Fiala- a more enriched 13CO2 pool and accounting for the relatively
Médioni et al. 1993), and vestimentiferan tubeworms (Kennicutt heavy d13C of tubeworm biomass. Further, stable carbon isotope
et al. 1992), as well as suggest differences in the contribution of values are also affected by the d13C of the environmental carbon
methanotrophs and chemoautotrophs to host carbon in mussels pool (Fisher 1995; Colaco et al. 2002). Scott et al. (2004) dem-
containing dual symbioses (Cavanaugh 1993; Trask and Van onstrated that the ‘‘light’’ d13C values of the symbionts of the
Dover 1999; Fiala-Médioni et al. 2002; Yamanaka et al. 2003). coastal protobranch clam Solemya velum are explained not
Stable carbon isotope signatures have also been used to only by the kinetic isotope effect of symbiont form I RubisCO
detect chemosynthetic symbioses in fossil bivalves of the clam (e = 24.5‰) but also by the d13C value of the CO2 in the
family Lucinidae, whose extant members all host chemosyn- sediment.
thetic symbionts. CoBabe (1991), by determining the d13C Similarly, the source of methane production can signifi-
values of organic matrix material extracted from lucinid fossils cantly impact the isotopic signature of methanotroph symbioses
dating to ca. 120,000 ya, showed that fossilized lucinid (Fisher 1995; MacAvoy et al. 2002). The d13C of methane varies
(Epilucina sp.) shells (d13C = m25‰) were about 5‰ lighter considerably depending on whether it is produced
than values from other bivalves collected in the same deposit thermogenically (d13C of CH4 > m45‰) or biologically by
(Pt. Loma, CA). Also, the organic matter of lucinid fossils was methanogens (d13C of CH4 < m60‰; Lilley et al. 1993; Fisher
similar to that from modern samples, implying that the fossil 1995), and this variability is reflected in the d13C values of
organic matrix did not decay or change significantly over time. chemosynthetic symbioses (Fisher 1995). Therefore, in instances
These results, along with strong evidence showing that shell d13C where the d13C of the source methane is unknown, conclusions
values are reasonable proxies for tissue values in extant species, about the contribution of methanotrophic symbionts to host
suggest that the fossil lucinid hosted chemosynthetic symbionts d13C values should be interpreted with caution (Fisher 1995). In
ca. 120,000 ya (CoBabe 1991). Thus, stable carbon isotope addition, interpreting d13C signatures may be especially prob-
analysis may be an effective tool for tracing the evolution of lematic for dual symbioses in which both methanotrophic and
chemosynthetic symbioses in the fossil record. thioautotrophic symbionts co-occur in the same host cell. In
One of the main factors affecting stable carbon isotope signa- these symbioses, the d13C values of the symbionts and the host
tures of chemoautotroph-invertebrate symbioses appears to be reflect a mix of methanotrophic and thioautotrophic metabo-
the form of RubisCO used by the symbionts. While d13C values lism (Fisher 1995). These signatures are potentially confounded
for vent bivalves hosting sulfide-oxidizing symbionts cluster in instances when the thioautotrophic symbionts use the CO2
between m27‰ and m35‰ and resemble values for free-living respired by the methanotrophs, resulting in a second discrimi-
chemoautotrophic bacteria, values for vent tubeworms, shrimp nation against an already light pool of CO2 and an anomalously
episymbionts, and many free-living bacterial mats at vents are light tissue d13C value (Fisher 1993).
Thus, while d13C values often provide the first evidence that enzymatic, genetic, and environmental data, caution must be
chemoautotrophic or methanotrophic symbioses occur in cer- used when comparing d15N values from different habitats and
tain animal species, researchers must recognize that d13C is species.
inherently responsive to physical, environmental, and enzymatic
factors. Stable carbon isotope signatures therefore should not
be used apart from other corroborating evidence (e.g., physio- Ecophysiology
logical and enzyme activity assays and genetic characterization)
to identify carbon fixation pathways or methane oxidation Symbioses between chemosynthetic bacteria and marine inver-
in chemosynthetic symbioses (Fisher 1995; Scott 2003; tebrates must acquire all of the substrates necessary for chemo-
Scott et al. 2004). synthetic metabolism: reduced sulfur or methane, oxygen,
dissolved inorganic carbon (DIC, as CO2 or CH4), and other
Sulfur and Nitrogen Isotopes nutrients (e.g., nitrogen and phosphorus) for use in biosynthe-
In addition to carbon isotopes, stable isotopes of sulfur and sis. In particular, to support energy generation, these symbioses
nitrogen are also used to study sources and metabolism of must obtain substrates from both oxic and anoxic environments.
these elements in symbioses. The extent to which different To meet these demands, the host-symbiont association relies on
sources of reduced sulfur—geothermal production in vent specialized biochemistry, physiology, and behavior. These adap-
fluid or microbial sulfate reduction in bottom sediment— tations are best studied in thioautotrophic endosymbioses and
support thioautotrophic metabolism has been inferred from are discussed primarily within this context below.
the d34S value of biological samples. Such analyses revealed
hydrothermally derived sulfide as the dominant sulfide source
for deep-sea vent symbioses (Fry et al. 1983; Yamanaka et al. Spanning the Oxic-Anoxic Interface
2003). In contrast, symbiotic bacteria within a shallow water
vestimentiferan tubeworm, Lamellibrachia satsuma (Miura et al. Access to both oxygen and reduced chemicals is necessary for
2002), and the protobranch, Solemya velum (Conway et al. aerobic respiration by chemosynthetic symbionts. Specifically,
1989), rely predominantly on sulfide derived from microbial thioautotrophs shuttle electrons from reduced sulfur (e.g., sul-
sulfate reduction. fide) to a terminal electron acceptor during oxidative phosphor-
Similarly, d15N values, because they vary predictably and ylation, generating a proton gradient that drives ATP synthesis.
largely between producer and consumer trophic levels (increase Though some thioautotrophic symbionts (such as those in the
of ca. 3.4 per level), are particularly useful markers for studying tubeworm Riftia pachyptila (Hentschel and Felbeck 1993) and
aquatic food web interactions (Minagawa and Wada 1984). In the clam Lucinoma aequizonata (Hentschel et al. 1993)) may use
general, d15N values of chemoautotrophic organisms are signif- nitrate as an electron acceptor during periods of anoxia, most
icantly lighter (<0‰; Van Dover and Fry 1994) than values for thioautotrophic symbionts typically use molecular oxygen for
photosynthetic organisms (>6‰; see Michener and Schell respiration. Similarly, methanotrophs must obtain oxygen for
(1994) and Fisher (1995)). Researchers have used this discrep- respiration as well as methane for both energy generation (via
ancy and the predictable trophic level fractionation of 15N to methane oxidation) and carbon assimilation (Anthony 1982).
show host reliance on symbiont-derived organic matter in This dual requirement for oxygen and reduced compounds
a number of symbioses including the coastal clams Solemya poses unique problems for thioautotrophs and methanotrophs.
velum and S. borealis (Conway et al. 1989, 1992) and in vent First, these organisms must obtain energy substrates from mutu-
mussels from the Mid-Atlantic Ridge (MAR) and the Galapagos ally exclusive environments—oxygen is absent or at very low
Rift (Fisher et al. 1988; Trask and Van Dover 1999). In addition, levels in the anoxic zones from which sulfide or methane is
d15N values have been used extensively in conjunction with d13C typically obtained. Second, sulfide, the predominant energy
values to show the flow of chemosynthetically derived organic source for thioautotrophy, spontaneously reacts with oxygen to
matter through vent food webs, including those on the MAR form less-reduced sulfur compounds (S0, S2O3m2, or SO4m2;
(Vereshchaka et al. 2000; Colaco et al. 2002), the Central Indian Zhang and Millero 1993), thereby decreasing the availability of
Ridge (Van Dover 2002), and the Galapagos Rift (Fisher et al. substrates for thioautotrophy. Though such abiotic oxidation
1994). As with d13C data, d15N values vary considerably among may be several orders of magnitude slower than biological sul-
sites; d15N may depend in part on the d15N of the dissolved fide oxidation (Millero et al. 1987; Johnson et al. 1988),
inorganic nitrogen (DIN) pool, the proportions and d15N values thioautotrophic symbioses must still compete with oxygen for
of different components (NH4+, NO3m2, NO2m, and urea) in the free sulfide. Also, in habitats containing both sulfide and meth-
DIN pool (Waser et al. 1998; Colaco et al. 2002), the uptake ane, abiotic oxidation of sulfide may limit the oxygen available
kinetics of different DIN assimilation pathways (Waser et al. for methanotrophy. These limitations force free-living
1998; Krueger 1996), and, as shown for vent shrimp thioautotrophs and methanotrophs into microaerophilic zones
(Vereshchaka et al. 2000) and mussels (Trask and Van Dover at the interface, or chemocline, between oxic (e.g., water col-
1999), the ontogenetic stage of the host. Therefore, as noted umn) and anoxic (e.g., vent fluid and sediment pore water)
above with stable carbon isotopes, in the absence of additional habitats. Such free-living bacteria demonstrate unique
mechanisms to support life at the oxic-anoxic interface; these . Table 23.2

adaptations may be behavioral (e.g., tracking the chemocline via Adaptations of thioautotrophs and methanotrophs for life at
gliding by Beggiatoa), anatomical (e.g., keeping cells in the oxic-anoxic interfacesa
chemocline via ‘‘veil’’ formation by Thiovulum or creation of
Adaptation Example
a filamentous sulfur matrix by Arcobacter), biochemical (e.g.,
internal or external sulfur deposition that serves as an electron Attachment Thiothrix
source or sink when sulfide or oxygen is limiting, as by Beggiatoa Motility, chemotaxis Beggiatoa and Thioploca
and Arcobacter), or developmental (e.g., resting stage formation Elemental sulfur Beggiatoa and Thiothrix
by methanotrophs; > Table 23.2 and references therein). deposition
Symbiosis thus may be viewed as an adaptation to simulta- Nitrate and sulfur storage Thiomargarita and Thioploca
neously obtain sulfide (or methane) and oxygen from anoxic- Create own interface Thiovulum
oxic interfaces, allowing thioautotroph or methanotroph sym-
Filamentous sulfur Arcobacter sp.
bionts, via association with a eukaryotic host, to circumvent production
many of the problems of sulfide acquisition (Cavanaugh 1985).
Resting cysts Methanotrophs
Similarly to free-living sulfur bacteria, thioautotrophic symbio-
Associate with eukaryote Thioautotroph and methanotroph
ses use specialized behavioral, anatomical, or physiological
symbionts
mechanisms, either to spatially or temporally bridge sulfidic
a
and oxic zones or to simultaneously sequester sulfide and From Anthony (1982), Jørgensen and Postgate (1982), Cavanaugh (1985),
oxygen (Cavanaugh 1994; Fisher 1996; Polz et al. 2000). For Schulz et al. (1999), and Wirsen et al. (2002)
instance, the cold seep vestimentiferan tubeworm

Lamellibrachia cf. luymesi acquires oxygen via its anterior
plume while extending a posterior section of its tube (the root)
deep into the sediment to acquire sulfide (Julian et al. 1999; disadvantageously with other blood components in the absence
Freytag et al. 2001). Similar burrowing tactics occur in some of sulfide.
species of symbiont-containing thyasirid clams, which possess Extracellular hemoglobins that simultaneously bind sulfide
a superextensile foot (up to 30 times the length of the shell) that and oxygen are absent in most other marine invertebrates that
burrows into the sediment to access hydrogen sulfide (Dufour host sulfide-oxidizing symbionts (Weber and Vinogradov 2001);
and Felbeck 2003), and in protobranchs of the genus Solemya, such organisms have evolved other mechanisms for regulating
which dig Y-shaped burrows in reducing sediments to allow sulfide toxicity and delivery. For instance, the vesicomyid clam
simultaneous pumping of oxygenated water from above and Calyptogena magnifica synthesizes a di-globular, nonheme mol-
sulfide-rich pore water from below (Stanley 1970; Cavanaugh ecule that readily binds free sulfide within the blood serum,
1983; > Fig. 23.11). Also, shrimp, nematodes, and oligochaetes perhaps via zinc residues (Arp et al. 1984; Zal et al. 2000).
migrate vertically along the oxygen-sulfide gradient or Also, several thioautotroph-containing species, including the
between separate oxic and anoxic zones, thereby enabling vent mussel Bathymodiolus thermophilus and the coastal clam
their symbionts to simultaneously access both energy substrates Solemya velum, appear to mediate detoxification in part by
or to store reduced sulfur compounds for later oxidation storage of sulfur in amino acids (e.g., taurine and thiotaurine;
(Polz et al. 2000). Conway and Capuzzo 1992; Pruski et al. 2000a; Joyner et al.
The vent tubeworm Riftia pachyptila possesses a remarkable 2003; Pruski and Fiala-Médioni 2003). Indeed, thiotaurine may
biochemical adaptation to simultaneously acquire sulfide and be used effectively as a biomarker of thioautotrophic symbioses
oxygen. R. pachyptila produces coelomic and vascular hemoglo- (Pruski et al. 2000b).
bins that, in contrast to most invertebrate and vertebrate hemo- Other host organisms, including some bivalve mollusks,
globins, can bind oxygen in the presence of sulfide (Arp et al. apparently avoid sulfide toxicity via mitochondrial oxidation
1985, 1987; Childress et al. 1991; Zal et al. 1996). R. pachyptila of sulfide. Powell and Somero (1986) first demonstrated mito-
appears to preferentially take up HSm from the surrounding chondrial sulfide oxidation in the coastal protobranch S. reidi.
fluid, despite a large H2S gradient from tubeworm blood to the The authors showed that mitochondria isolated from the gill
environment (Goffredi et al. 1997a). The HSm diffuses across the and foot of S. reidi exhibit ADP-stimulated oxygen uptake and
plume of the worm (Goffredi et al. 1997a) and then binds ATP synthesis following the addition of sulfide. On the basis of
reversibly and independently of O2 at two free cysteine the effects of cytochrome and reduced nicotinamide adenine
residues, each located on a distinct globin type (Zal et al. 1997, dinucleotide (NADH) oxidase inhibitors, electrons from sulfide
1998; Bailly et al. 2002). These residues are well conserved in oxidation appear to enter the respiratory chain at cytochrome c
both symbiont-containing and symbiont-free annelids from in S. reidi mitochondria (Powell and Somero 1986). Further
sulfidic environments but are absent in annelids from sulfide- characterization of this system using 35S showed that sulfide is
free habitats (Bailly et al. 2002, 2003). Bailly et al. (2003) suggest oxidized exclusively to thiosulfate (O’Brien and Vetter 1990),
that the sulfide-binding function may have been lost via positive a nontoxic intermediate that can function as the energy source in
selection, if the sulfide-binding cysteine residues react symbiotic carbon fixation. Subsequently, researchers have
demonstrated mitochondrial sulfide oxidation across a wide may, in combination with the relatively low discrimination
range of organisms, including polychaete worms, clams, fishes, of form II RubisCO against 13C, result in a 13C-enriched signa-
and chickens (Grieshaber and Volkel 1998; Yong and Searcy ture of symbiont and host biomass (Robinson et al. 2003;
2001). These data lend credence to the hypothesis that mito- Scott 2003).
chondria evolved from sulfide-oxidizing endosymbiotic bacteria In chemosynthetic endosymbioses, the host benefits by
(Searcy 1992). obtaining part or all of its nutrition from the symbiont, via
Readers should consult several additional reviews (e.g., two potential transfer mechanisms: the host may assimilate
Cavanaugh 1994; Fisher 1996; Polz et al. 2000) for a more autotrophically fixed carbon that has been released by the sym-
extensive discussion of the remarkable adaptations used by biont and translocated to host cells in the form of soluble
chemoautotrophic symbioses to sequester both oxygen and organic molecules, or the host may directly digest bacterial
reduced chemicals across oxic-anoxic zones. cells. Radiotracer analysis and microscopy have proven par-
ticularly useful in studying host nutrition. For example,
Fisher and Childress (1986) showed a rapid (within hours)
Carbon Uptake and Transport appearance of radiolabeled carbon in the symbiont-free
tissues of the host clam Solemya reidi following exposure to
14
In addition to oxygen and reduced sulfur compounds, C-labeled bicarbonate, suggesting release of fixed carbon
thioautotrophic symbionts utilizing the Calvin cycle require by the symbiont population. In contrast, a slow (1–5 days)
CO2 for autotrophic carbon fixation. Acquisition of CO2 is transfer of labeled organic carbon from methanotroph-
not trivial given that relative concentrations of the three containing tissue to symbiont-free tissue of a seep mussel
distinct chemical species (CO2, HCO3m, and CO3m2) in exposed to 14C-labeled methane was inferred to be due to initial
14
the dissolved inorganic carbon (DIC) pool can vary considerably CH4 incorporation by the symbionts with host digestion of
depending on pH (pKa of 6.4 for CO2:HCO3m at 25 C; see the symbionts occurring later (Fisher and Childress 1992). Electron
section > ‘‘Habitat Chemistry’’ in this chapter). In general, microscopy showing symbionts being degraded in the basal
the majority of DIC in seawater (pH 8.0) is HCO3m. region of bacteriocytes in other methane-based and dual
But at vents the typically lower pH of the mixed vent fluid chemoautotroph-methanotroph mussel symbioses supports
and ambient bottom water generates higher concentrations of this interpretation (Cavanaugh et al. 1992; Barry et al. 2002),
CO2, giving organisms that use the Calvin cycle a distinct as does the detection of lysosomal enzymes in the gills of the
advantage. vent bivalves Calyptogena magnifica and Bathymodiolus
The tubeworm Riftia pachyptila provides an interesting thermophilus (Fiala-Médioni et al. 1994; Boetius and Felbeck
model in which to study the uptake and transport of DIC. 1995) and the shallow water clam Lucinoma aequizonata
Goffredi et al. (1997b) demonstrated that for R. pachyptila, pH (Boetius and Felbeck 1995).
plays an important role in DIC uptake. The acidity of diffuse In the R. pachyptila tubeworm symbiosis, the transfer of
vent fluid (pH ca. 6) around tubeworms ensures that CO2 (pKa carbon from symbiont to host appears to occur via both trans-
of 6.1 at in situ temperature and pressure of ca. 10 C and 101.3 location and digestion (Bright et al. 2000). Felbeck (1985) and
kPa; Dickson and Millero 1987) is the dominant DIC form in the Felbeck and Turner (1995) documented a rapid (within seconds)
vent environment. This contrasts with the vascular fluid of the appearance of labeled succinate and malate in trophosome tissue
worm, which has an alkaline pH of 7.1–7.5, apparently because and in vascular and coelomic blood following exposure of whole
of the action of H+-ATPases (Goffredi et al. 1999; Goffredi and worms (in pressure vessels) and plumes to 14C-bicarbonate.
Childress 2001; Girguis et al. 2002). The alkaline pH inside Riftia Subsequently, Felbeck and Jarchow (1998) showed that succi-
results in rapid conversion of CO2 to HCO3m, which, because of nate, malate, and several other organic acids and sugars were
its negative charge, cannot diffuse out of the worm; this in effect excreted by purified suspensions of R. pachyptila symbionts,
creates a bicarbonate ‘‘trap’’ (Childress et al. 1993). Thus, suggesting that these simple organic compounds might be
a gradient of higher external [CO2] to lower internal [CO2] important intermediates in the transfer of fixed carbon from
develops across the tubeworm plume and drives diffusion of symbionts to host. Corroborating these data, Bright et al.
DIC into the blood (Childress et al. 1993; Goffredi et al. 1997b; (2000), using pulse labeling analysis, showed that the bulk of
Scott 2003). Following diffusion into the plume, DIC (as CO2 organic carbon assimilated into R. pachyptila tissue is first
and HCO3m) is transported by the vascular system to the released by metabolically active bacteria at the center of
symbiont-containing trophosome. Here, carbonic anhydrase, a trophosome lobule. However, these authors also showed that
the enzyme that reversibly converts CO2 into HCO3m in both a smaller fraction of host carbon is obtained by digestion of
prokaryotes and eukaryotes, may play a role in converting bacterial cells at the lobule periphery (Bright et al. 2000). This
HCO3m into CO2, the DIC species used by RubisCO (Kochevar evidence for digestion is supported by prior studies showing
and Childress 1996; De Cian et al. 2003a, b). As discussed degenerative stages of bacteria within the R. pachyptila
above for Riftia, DIC incorporation into symbiont biomass trophosome (Bosch and Grassé 1984; Hand 1987). In addition,
occurs via CO2 fixation by a form II RubisCO of the Calvin- relatively high lysozyme activity in Riftia tissue further suggests
Benson cycle. Rapid CO2 fixation rates create steep internal that digestion of symbionts plays a role in tubeworm nutrition
[CO2] gradients between symbiont and host cytoplasm that (Boetius and Felbeck 1995).
Nitrogen morphologies suggest a primitive state, purportedly survived

past extinction events due to the relative isolation of vents
The partners in a symbiosis must also acquire all of the other from the photic zone (McArthur and Tunnicliffe 1998). This
macro- and micronutrients, particularly nitrogen and phosphorus, perception of vents as ancient ecosystems is supported by the
for use in the biosynthesis of organic compounds. Currently, very fossil record, which shows that over 80 % of vent species are
little is known about how various forms (inorganic and organic) of found only at vent sites (Tunnicliffe 1991, 1992; Little et al. 1997;
phosphorus are transferred to and among different pools within Little and Vrijenhoek 2003) and that the oldest vent site dates to
chemoautotrophic endosymbioses. Most studies have focused on the Silurian (430 Mya; Little et al. 2004). But the fossil record
nitrogen metabolism, using a combination of enzyme character- for vents is relatively poor. There are only 19 known fossilized
izations and physiological experiments to elucidate nitrogen assim- vent sites on the planet, perhaps because calcium carbonate
ilation pathways. Nitrate (NO3m2), which is abundant at vents (in structures dissolve relatively quickly in vent fluids (Hunt 1992;
situ concentrations of 40 mM; Johnson et al. 1988), appears to Kennish and Lutz 1999). Also, studying vent fauna evolution
be the predominant nitrogen source for vent symbioses. For based on the morphological characters of fossils is problematic if
example, Lee et al. (1999) demonstrated the activity of nitrate much of the specimen has degraded or if the preserved character
reductase, a bacterial enzyme involved in converting nitrate to is plastic or isomorphic. In particular, vestimentiferan
ammonia for either assimilatory or respiratory purposes, in the tubeworms are known for the phenotypic plasticity of their
vent tubeworms Riftia pachyptila and Tevnia jerichonana and the tubes (Southward et al. 1995; Black et al. 1998).
mussel Bathymodiolus thermophilus. In addition, the ammonia In contrast, molecular evidence suggests that vent taxa
assimilation enzymes glutamine synthetase (GS) and glutamate evolved more recently (22–150 Mya; later-Mesozoic and
dehydrogenase (GDH) were detected in these symbioses, and Cenozoic) and suggests an alternative hypothesis to vent taxa
almost all GS activity in symbiont-containing tissue was shown as living relics: vents were recently populated from shallow
to be due to enzyme produced by the bacterial symbiont and not seeps or whale falls (Van Dover et al. 2002; Hurtado 2002).
the host (Lee et al. 1999). Supporting these data, physiological Indeed, the communities most similar to those of vents
experiments on R. pachyptila kept in pressurized chambers occur at seeps. Compared to the spatially and temporally
showed that the symbiont population reduces nitrate to ammo- patchy distribution of vent fossils (with most being concentrated
nia not for respiratory purposes but for incorporation into both in the Silurian and Devonian rocks of the Ural Mountains),
symbiont and host biomass (Girguis et al. 2000). However, in R. seep fossils are ubiquitous (Little and Vrijenhoek 2003).
pachyptila, high GS activity also occurred in symbiont-free At least 50, and perhaps as many as 200, fossilized seep sites
branchial plume tissue, suggesting that the host may also be dating from the Devonian to the Pleistocene have been
involved in assimilation of ammonia from the vent environment uncovered. These specimens are much better preserved than
(Minic et al. 2001). But further enzymatic characterization of most vent fossils and include extant vent taxa not yet uncovered
Riftia tissues demonstrated that the tubeworm depends on its at fossil vent sites (e.g., vesicomyids, thyasirids, mytilids, and
symbionts for the de novo synthesis of pyrimidine nucleotides solemyids; Little and Vrijenhoek 2003). This greater diversity
(Minic et al. 2001) as well as for the biosynthesis of polyamines supports the seeps-to-vents hypothesis. However, opponents
(Minic and Herve 2003), suggesting that the trophosome is argue that the vent fossil record has been greatly affected
a primary site for nitrogen assimilation and metabolism. by high calcium carbonate dissolution rates (Little and
In contrast, in the thioautotrophic symbiosis involving the Vrijenhoek 2003).
shallow water clam Solemya reidi, inorganic nitrogen is readily While the discrepancy between the evolutionary histories
assimilated in the form of ammonia (Lee and Childress 1994), suggested by the fossil and molecular evidence needs to be
which is abundant in the shallow water, nutrient-rich habitats of resolved, it must also be stressed that these data are not evidence
the clam (e.g., sewage outfalls). Ammonia incorporation rates for chemosynthetic symbioses. In a unique study, CoBabe
are highest in the symbiont-containing gill tissue, and the sulfur- (1991) was able to deduce a chemosynthetic symbiosis by ana-
containing amino acid taurine appears to be a major end prod- lyzing the organic matrix from fossil lucinid shells using stable
uct of ammonia assimilation (Lee et al. 1997). The mechanisms carbon isotopes. This result is encouraging and suggests that
by which chemosynthetic symbionts, particularly those both the age of these organisms and their symbiosis can be
contained within the cells of invertebrate hosts (such as Solemya addressed using current methods.
and Riftia), acquire all of the other macro- and micronutrients
for biosynthesis have yet to be characterized.
Organism Interactions
Ecology and Evolution In the relatively featureless and nutrient poor deep sea, vent and
seep environments are ecological oases (Laubier 1989). Initially,
History free-living chemoautotrophic bacteria were hypothesized to
provide the bulk of primary production in these communities
Prior to the use of molecular techniques, researchers considered (Lonsdale 1977). Indeed, at some vent sites, suspended bacteria
vent taxa to be relic species. These organisms, whose strange or bacteria in surface-attached mats are a large food source for
higher trophic levels (Humes and Lutz 1994; Van Dover 2000). In contrast, symbiont populations that are environmentally
But the dominant strategy for the major vent and seep fauna is transmitted are effectively larger and more genetically heteroge-
symbiosis with chemoautotrophic bacteria (Cavanaugh 1994), neous than populations transmitted vertically. Comparisons of
and these symbioses significantly influence the ecology of 16S rRNA gene evolution between free-living bacteria, in which
the nonsymbiotic community. Not only are chemosynthetic significant recombination occurs (Dykhuizen and Green 1991;
symbioses a major and stable source of organic carbon (Sarrazin Levin and Bergstrom 2000), and symbiotic chemosynthetic bac-
and Juniper 1999), but as biogenic structures, they also provide teria revealed unexpected differences in rates of evolution
living space for a diversity of species in an otherwise two- depending on mode of transmission (Peek et al. 1998). While
dimensional landscape of basalt or sediment (Bergquist et al. chemoautotrophic, maternally transmitted endosymbionts did
2003). For example, the tubes of chemosynthetic vestimen- exhibit rapid evolutionary rates, consistent with their small
tiferans support mussels, sponges, and limpets, many of which population sizes, environmentally transmitted symbionts
host their own chemosynthetic symbionts (Yamamoto et al. evolved more slowly than their free-living counterparts
2002; Bergquist et al. 2003; Bates et al. 2004). (Peek et al. 1998). The authors suggest that this slower rate of
Vent symbioses may also significantly impact the free-living evolution could be caused by purifying selection in a large pop-
bacterial community by providing increased surface area for ulation. These results, however, were based on one gene across
attachment. Free-living bacteria that cluster phylogenetically many lineages; a true genomic analysis of evolution in chemo-
with known chemoautotrophic and heterotrophic groups have synthetic endosymbionts is necessary to extend these findings.
been isolated from tubeworm surfaces (Lopez-Garcia et al. 2002; Because chemosynthetic symbionts have yet to be cultured
Yamamoto et al. 2002). On the Mid-Atlantic Ridge, a single and their hosts are difficult to maintain in the laboratory, the
phylotype of shrimp episymbionts, which appear to be trans- transmission strategy of a symbiosis has been inferred by phy-
mitted among hosts via the environment, represented over 60 % logenetic analysis or PCR-based detection of bacteria in host
of the free-living bacteria (Polz and Cavanaugh 1995). This reproductive tissues or gametes. If the symbionts are maternally
suggests that the host inoculates inanimate surfaces continu- transmitted and the symbioses stable, congruence of host and
ously, increasing the probability of symbiont attachment relative symbiont phylogenies should occur (e.g., Chen et al. 1999; Thao
to the free-living community (Polz and Cavanaugh 1995). Such et al. 2000; Degnan et al. 2004) and bacterial symbionts should
environmental inoculation may also occur in tubeworm and be found in ovaries or oviducts of the host. Using these tech-
lucinid clam symbioses, in which the symbionts also appear to niques, vertical transmission has been proposed for the solemyid
be transmitted environmentally (Durand and Gros 1996; protobranchs (Cary 1994; Krueger et al. 1996) and vesicomyid
Durand et al. 1996; Di Meo et al. 2000; Nelson and Fisher clams (Endow and Ohta 1990; Cary and Giovannoni 1993; Peek
2000; McMullin et al. 2003). et al. 1998; Hurtado et al. 2003). Interestingly, although bacteria
have been detected via PCR in the gonads of female hosts, this
does not necessarily imply direct bacterial endocytic localization
Transmission Strategies and Effects on Symbiosis in host eggs. Indeed, in Solemya reidi, the internal contents of
oocytes do not contain bacteria, and instead the transmission
The transmission strategy of a symbiosis reveals much about the mechanism is thought to occur via ingestion; the larvae ingest
evolutionary dynamics between host and symbiont. Symbiont the bacteria, which are then engulfed by hemocytes in the larval
transmission can occur environmentally (through a free-living perivisceral cavity and transported to the developing gill
population of symbiotic bacteria), horizontally (between (Gustafson and Reid 1988). The oligochaetes also exhibit an
contemporary organisms sharing the same habitat), or vertically interesting mechanism of vertical transmission. During ovipo-
(from parent to offspring). Vertically transmitted endosymbi- sition, the eggs appear to be infected with the symbiotic bacteria
onts are effectively disconnected from their free-living counter- via the adult’s genital pad (Giere and Langheld 1987). During
parts. These symbionts experience elevated rates of mutation the development of the larvae, many of the bacteria exist intra-
and fixation of slightly deleterious alleles because of genetic drift cellularly, but as the animal matures, the symbionts take their
(Wernegreen 2002). For the most part, these evolutionary primarily extracellular form.
effects are due to a vastly different selective regime inside the However, of the putatively vertically transmitted symbioses,
host and a severely depreciated population size (Ohta 1973); only associations involving vesicomyid clams show phylogenetic
endosymbionts undergo a population bottleneck upon host congruence between host and symbiont (Peek et al. 1998;
colonization and another upon transmission (Mira and Moran Hurtado et al. 2003). Cospeciation does not appear to have
2002). But the asexuality and lack of recombination in occurred in the solemyid protobranchs (Durand et al. 1996;
endosymbionts exacerbate these genetic problems through Krueger and Cavanaugh 1997) or in the mytilid mussels
what is known as ‘‘Muller’s rachet’’ (Muller 1964; Moran (McKiness 2004). When evaluating phylogenetic congruence,
1996). In Muller’s ratchet, wild-type recombinants cannot be however, other factors that influence a phylogenetic reconstruc-
introduced into the endosymbiont population (Moran and tion, such as geographic constraints, must be taken into account.
Baumann 1994; Dale et al. 2003); genetic drift therefore occurs Also, robust phylogenies with adequate taxa sampling for both
quickly, and the population cannot recover after fixation of host and symbiont are necessary; incomplete phylogenies may
deleterious alleles. be hindering analyses of the solemyid and mytilid symbioses.
Lack of PCR-based evidence and phylogenetic incongruence gene flow with increasing distance between sites (Black et al.
has been used to infer an environmental mode of transmission 1994), as a stepping-stone model would predict (Kimura and
for several of the chemosynthetic symbioses. For instance, the Weiss 1964), while others show a more widespread gene flux
lucinid clams exhibit environmental transmission (Durand and (Karl et al. 1996) or appear to encounter barriers to dispersal
Gros 1996; Gros et al. 1996, 1998, 2003a, b). Researchers have other than distance (Black et al. 1998). These differences should
even been able to exchange symbionts between lucinid species be resolved with a greater understanding of the major variables
without affecting the development of the juvenile animal (Gros affecting vent biogeography, including larval development and
et al. 2003a). In addition, vent tubeworms appear to acquire dispersal, symbiont distribution, oceanic flow, and past and
their symbionts from the environment (Distel and Cavanaugh current bathymetry. This section focuses predominantly on
1994; Feldman et al. 1997; Laue and Nelson 1997; Di Meo et al. host biogeography because research on the population genetics
2000; Nelson and Fisher 2000; McMullin et al. 2003), as and biogeography of bacterial symbionts is lacking. Understand-
evidenced in part by the presence of functional genes for sensing ing host population dynamics, however, does provide valuable
and responding to the environment as well as a flagellin gene in insight into the distribution of the chemosynthetic symbionts to
the Riftia symbiont (Hughes et al. 1997, 1998; Millikan et al. which most vent fauna are tightly linked.
1999). Indeed, the 16S rRNA phylotype has been detected in vent Vent habitats are highly ephemeral and sensitive to varia-
environments via both PCR and in situ hybridization, tions in tectonic activity, hydrothermal inputs, and geologic
suggesting the vent tubeworm symbionts are environmentally events. Consequently, the persistence of vent organisms, which
transmitted (Harmer et al. 2004). While environmental trans- are predominantly sessile as adults, depends on successful larval
mission of tubeworm symbionts seems to be a potentially risky dispersal to new sites. The dispersal strategy of larvae can sig-
strategy, given the stochastic nature of environmental transmis- nificantly impact the biogeography of the adult organism (Lalou
sion and the complete dependence of the adult tubeworms on and Brichet 1982; Fustec et al. 1987). On the basis of laboratory
their symbionts for nutrition, detection of ‘‘wild’’ symbionts in studies and comparisons with shallow water species, researchers
conjunction with the phylogenetic evidence supports environ- infer that some vent larvae are planktotrophic, while others are
mental transmission in this species. lecithotrophic (Lutz et al. 1980; Turner et al. 1985; Young et al.
The mechanism of transmission for the mytilid mussels 1996; Marsh et al. 2001). Although both forms are pelagic,
remains largely unresolved; on the basis of varying evidence, planktotrophic larvae are positively buoyant and feed in the
researchers have suggested both vertical and environmental water column, while lecithotrophic larvae are negatively buoyant
transmission. Vertical transmission in Bathymodiolus and nonfeeding (Poulin et al. 2001). Larvae of the large
thermophilus, the thioautotroph-hosting mussel, was suggested vesicomyid clam Calyptogena magnifica typify a planktotrophic
in 1993, but evidence supporting this report is not yet published dispersal strategy successfully exploiting the vent plume to carry
(Cary and Giovannoni 1993). In contrast, a recent study based them many kilometers (Pradillon et al. 2001; Mullineaux et al.
on genetic and ultrastructural data of the chemoautotrophic 2002). Although planktotrophic larvae risk being carried off the
symbionts of B. azoricus, a MAR mussel hosting both ridge axis by cross currents, C. magnifica apparently encounters
thioautotrophs and methanotrophs, indicated environmental no significant barriers to dispersal across the equator on the East
acquisition of the chemoautotrophic symbionts (Won et al. Pacific Rise (EPR; Karl et al. 1996). Conversely, lecithotrophic
2003a; DeChaine et al. 2004). In addition, McKiness (2004) larvae are less affected by crosscurrents but, because they are
provided the first assessment of cospeciation between symbiont nonfeeding, have relatively short larval stages and therefore
and host in Bathymodiolus mussels, analyzing molecular data for limited time for dispersal. For example, in laboratory studies,
both methanotrophic and chemoautotrophic symbionts and larval Riftia pachyptila exhibits a lecithotrophic strategy, surviv-
testing phylogenetic congruence with the hosts. The results ing a maximum of 38 days (Marsh et al. 2001). Assuming flow
showed weak support for vertical transmission of the chemoau- rates characteristic of EPR currents, this interval suggests
totrophic symbionts but provided no evidence for vertical trans- a maximum dispersal distance of 100 km (Marsh et al. 2001);
mission of the methanotrophs. however, the in situ dispersal distance is unknown given that
Riftia larvae have not been detected in the wild.
The geology and tectonic activity associated with mid-ocean
Biogeography and Population Genetics ridges also impact the biogeography of vent organisms. For
instance, Iceland, an active site of crust formation, rises out of
The view of hydrothermal vents as deep-sea islands frames the ocean along the northern MAR, forming a barrier that pre-
questions of vent biogeography and population genetics. Com- vents dispersal along the ridge axis (Tyler and Young 2003).
pared to the relatively uniform and stable environment of the Given that Iceland interrupted the MAR approximately 55
abyssal deep sea, hydrothermal vents are ephemeral, dynamic, Mya, the ridge axis north of Iceland constitutes one of the
and geographically fragmented. A chain of vents along a mid- most isolated vent systems on the planet, perhaps representing
ocean ridge resembles a chain of islands in an archipelago. a new biogeographic province (Bilyard and Carey 1980; Dunton
However, genetic data for many vent species do not cleanly fit 1992; Svavarsson et al. 1993). Similar dispersal barriers are seen
an ‘‘island’’ or ‘‘stepping-stone’’ model of biogeography among vent fields abutting the Azores Rise in the Atlantic (Tyler
(Vrijenhoek et al. 1998). Some host taxa do exhibit a decline in and Young 2003) and also evident between the EPR and the
northeast Pacific vent fields (Tunnicliffe 1988; Tunnicliffe and complement traditional methods. These studies provide valu-
Fowler 1996). These barriers are insurmountable and may pro- able insight into the population dynamics, evolutionary history,
vide the conditions for allopatric speciation of both host and and carbon and nutrient metabolism of symbionts. In addition,
symbiont. projects are currently underway to sequence the genomes of
Research on EPR bathymodiolid mussels and their symbi- some of the chemosynthetic symbionts described in this review
onts provides a good example of how larval dispersal strategy, (e.g., symbionts of Riftia pachyptila and Solemya velum). Geno-
current regime, and bathymetry interact to structure biogeogra- mic analysis, in conjunction with new technologies to manipu-
phy (Lonsdale 1977; Corliss et al. 1979). Except at northern late symbioses under in situ conditions (e.g., via vascular
Pacific sites, which are separated from the EPR by the North catheters; Felbeck et al. 2004) and to sample the physical envi-
America landmass, vent communities on the Pacific ridge ronment (e.g., electrochemical sampling; Luther et al. 2001), will
axis appear relatively uniform. For example, the mussel contribute significantly to our understanding of symbiont biol-
Bathymodiolus thermophilus, which undergoes ogy. Scientists are now poised to reveal how interactions with the
a planktotrophic larval stage, occurs over a distance of host and the abiotic environment impact symbiotic chemosyn-
4,900 km (from 13 N to 32 S) on the EPR. Mussel populations thetic bacteria over both ecological and evolutionary timescales.
along 13 N and 11 S are genetically indistinguishable, indicat-
ing no population subdivision (Craddock et al. 1995; Won et al.
2003). Deep ocean currents that flow primarily NNW and SSE Acknowledgments
along the axis (Marsh et al. 2001) may facilitate dispersal of
larval B. thermophilus, contributing to the homogeneity We thank our colleagues and collaborators for active discussions
observed along the EPR. However, there is some genetic struc- on chemosynthetic symbioses and for their scientific contribu-
ture in the EPR mytilid populations; the westward currents tions. Without the chief scientists, captains, and crews of the
across the ridge axis at 15 N and the Easter Microplate are research vessels (including R/V Atlantis II, R/V Atlantis, and R/V
obstacles for planktotrophic larvae. At 15 N, a westward current Knorr), and the expedition leaders and crews of Deep Submer-
flows across the ridge axis, partially isolating the 17 S popula- gence Vehicle (DSV) Alvin and the remotely operated vehicles,
tion from the other northern populations. Further south, at the we could not explore the vast unknown deep sea—to them we
Easter Microplate, mussel populations are severely divergent are grateful. Research in my laboratory (CMC) on chemosyn-
(Won et al. 2003). Although morphologically indistinguishable, thetic symbioses has been supported by grants from NSF (Bio-
mussels north and south of the Microplate are genetically dis- logical Oceanography, RIDGE, Cell Biology), the Office of Naval
tinct (Won et al. 2003). The Easter Microplate therefore appears Research, NOAA National Undersea Research Center for the
to be a significant topographic obstacle for larval dispersal. Such West Coast and Polar Regions, and NASA (Exobiology) and by
a feature can produce cross-axis currents, like those at 15 N, that graduate fellowships from the NIH, NSF (IGERT), and Howard
may sweep bathymodiolid larvae (which are positively buoyant) Hughes Medical Institute, which we gratefully acknowledge.
off the ridge axis (Fujio and Imasato 1991; Mullineaux et al.
1995). The degree to which such barriers also impact the genetic
diversity and biogeography of chemosynthetic symbiont
populations remains an open question. References
Alt JC (1995) Subseafloor processes in mid-ocean ridge hydrothermal systems.
In: Humphris SE, Zierenberg RA, Mullineaux LS, Thomson RE (eds)
Summary Seafloor hydrothermal systems: physical, chemical, biological, and
geological interactions. American Geophysical Union, Geophysical
Scientific understanding of chemosynthetic symbioses con- Monograph 91
tinues to expand. The spectacular discovery of hydrothermal Anderson A, Childress J, Favuzzi J (1987) Net uptake of CO2 driven by sulfide and
thiosulfite oxidation in the bacterial symbiont-containing clam Solemya
vents highlighted the importance of chemosynthetic bacteria
reidi. J Exp Biol 133:1–131
both in food webs and in symbioses with eukaryotes and pro- Anthony C (1982) The biochemistry of methylotrophs. Academic, London
vided the impetus to examine less exotic environments for such Arp AJ, Childress JJ, Fisher CRJ (1984) Metabolic and blood gas transport
associations. As other oxic-anoxic environments (e.g., freshwa- characteristics of the hydrothermal vent bivalve Calyptogena magnifica.
ter) and the invertebrates and protists that inhabit them are Physiol Zool 57:648–662
Arp AJ, Childress JJ, Fisher CR (1985) Blood gas transport in Riftia pachyptila.
explored, new symbioses will undoubtedly be discovered. Fur-
Bull Biol Soc Wash 6:289–300
ther, chemosynthetic bacteria that use other sources of energy Arp AJ, Childress JJ, Vetter RD (1987) The sulphide-binding protein in the blood
(e.g., hydrogen) may be found in similar associations. of the vestimentiferan tube-worm, Riftia Pachyptila, is the extracellular
Current studies of these fascinating symbioses involve haemoglobin. J Exp Biol 128:139–158
a range of experimental and diagnostic tools, including physio- Bailly X, Jollivet D, Vanin S, Deutsch J, Zal F, Lallier F, Toulmond A (2002)
Evolution of the sulfide-binding function within the globin multigenic
logical assays in specialized growth chambers, enzyme charac-
family of the deep-sea hydrothermal vent tubeworm Riftia pachyptila. Mol
terizations, immunodetections, and stable isotope analyses. Biol Evol 19:1421–1433
Increasingly, molecular techniques, such as PCR-based gene Bailly X, Leroy R, Carney S, Collin O, Zal F, Toulmond A, Jollivet D (2003) The
probing, FISH, and 16 rRNA phylogenetic analysis, are used to loss of the hemoglobin H2S-binding function in annelids from sulfide-free
habitats reveals molecular adaptation driven by Darwinian positive selection. Cavanaugh CM (1985) Symbioses of chemoautotrophic bacteria and marine
Proc Natl Acad Sci USA 100:5885–5890 invertebrates from hydrothermal vents and reducing sediments. Biol Soc
Barry JP, Buck KR, Kochevar RK, Nelson DC, Fujiwara Y, Goffredi SK, Hashimoto Wash Bull 6:373–388
J (2002) Methane-based symbiosis in a mussel, Bathymodiolus platifrons, Cavanaugh CM, Levering PR, Maki JS, Mitchell R, Lidstrom ME (1987) Symbiosis
from cold seeps in Sagami Bay, Japan Invertebr. Biol 121:47–54 of methylotrophic bacteria and deep-sea mussels. Nature 325:346–348
Bauer-Nebelsick M, Bardele CF, Ott JA (1996) Electron microscopic studies on Cavanaugh CM, Abbott MS, Veenhuis M (1988) Immunochemical localization
Zoothamnium niveum (Hemprich & Ehrenberg, 1831) Ehrenberg 1838 of ribulose-1,5-bisphosphate carboxylase in the symbiont-containing
(Oligohymenophora, Peritrichida), a ciliate with ectosymbiotic, chemoau- gills of Solemya velum (Bivalvia:Mollusca). Proc Natl Acad Sci USA
totrophic bacteria. Eur J Protistol 32:202–215 85:7786–7789
Bennett BA, Smith CR, Glaser B, Maybaum HL (1994) Faunal community Cavanaugh CM, Wirsen C, Jannasch HJ (1992) Evidence for methylotrophic
structure of a chemoautotrophic assemblage on whale bones in the deep symbionts in a hydrothermal vent mussel (Bivalvia:Mytilidae) from the
northeast Pacific Ocean. Mar Ecol Progr Ser 108:205–223 Mid-Atlantic ridge. Appl Environ Microbiol 58:3799–3803
Bergquist DC, Ward T, Cordes EE, McNelis T, Howlett S, Kosoff R, Hourdez S, Cavanaugh CM (1993) Methanotroph-invertebrate symbioses in the marine
Carney R, Fisher CR (2003) Community structure of vestimentiferan- environment: ultrastructural, biochemical, and molecular studies. In:
generated habitat islands from Gulf of Mexico cold seeps. J Exp Mar Biol Murrell JC, Kelly DP (eds) Microbial growth on C1 compounds intercept
Ecol 289:197–222 Cavanaugh CM (1994) Microbial symbiosis: patterns of diversity in the marine
Bilyard GR, Carey AG (1980) Zoogeography of Western Beaufort Sea Polychaeta environment. Am Zool 34:79–89
(Annelida). Sarsia 65:19–25 Cavanaugh CM, Robinson JJ (1996) CO2 fixation in chemoautotroph-
Black MB, Lutz RA, Vrifenhoek RC (1994) Gene flow among vestimentiferan tube invertebrate symbioses: expression of form I and form II RuBisCO. In:
worm (Riftia pachyptila) populations from hydrothermal vents of the East- Lidstrom ME, Tabita FR (eds) Microbial growth on C1 compounds. Kluwer,
ern Pacific. Mar Biol 120:33–39 The Netherlands
Black MB, Trivedi A, Maas PAY, Lutz RA, Vrijenhoek RC (1998) Population Chen XA, Li S, Aksoy S (1999) Concordant evolution of a symbiont with its host
genetics and biogeography of vestimentiferan tube worms. Deep Sea Res Pt insect species: molecular phylogeny of genus Glossina and its bacteriome-
II Top Stud Oceanogr 45:365–382 associated endosymbiont, Wigglesworthia glossinidia. J Mol Evol 48:49–58
Boetius A, Felbeck H (1995) Digestive enzymes in marine-invertebrates from Childress JJ, Fisher CR, Brooks JM, Kennicutt MC 2nd, Bidigare R, Anderson AE
hydrothermal vents and other reducing environments. Mar Biol (1986) A methanotrophic marine molluscan (Bivalvia, Mytilidae) symbiosis:
122:105–113 mussels fueled by gas. Science 233:1306–1308
Bosch C, Grassé PP (1984) Cycle partiel des bactéries chimioautotrophes Childress JJ, Fisher CR, Favuzzi JA, Kochevar RE, Sanders NK, Alayse AM
symbiotiques et eurs rapports avec les bacteriocytes chez Riftia pachyptila (1991) Sulfide-driven autotrophic balance in the bacterial symbiont-
Jones (Pogonophore Vestimentifére). II: L’evolution des bactéries containing hydrothermal vent tubeworm, Riftia pachyptila Jones. Biol Bull
symbotiques et des bacteriocytes. Comptes Rendus L’Acad. Sci Ser III: 180:135–153
Sciences de la Vie—Life Sciences 299:413–319 Childress JJ, Fisher CR (1992) The biology of hydrothermal vent animals: phys-
Boss KJ, Turner RD (1980) The giant white clam from the Galapagos Rift, iology, biochemistry, and autotrophic symbioses. Oceanogr Mar Biol Annu
Calyptogena magnifica species novum. Malacologia 20:161–194 Rev 30:337–441
Bright M, Keckeis H, Fisher CR (2000) An autoradiographic examination of Childress JJ, Lee RW, Sanders NK, Felbeck H, Oros DR, Toulmond A, Desbruyères
carbon fixation, transfer and utilization in the Riftia pachyptila symbiosis. D, Kennicutt MC, Brooks J (1993) Inorganic carbon uptake in hydrothermal
Mar Biol 136:621–632 vent tubeworms facilitated by high environmental pCO2. Nature
Brigmon RL, De Ridder C (1998) Symbiotic relationship of Thiothrix spp. with an 362:147–169
echinoderm. Appl Environ Microbiol 64:3491–3495 CoBabe EA (1991) Lucinid Bivalve evolution and the detection of
Brooks JM, Kennicutt MC 2nd, Fisher CR, Macko SA, Cole K, Childress JJ, chemosymbiosis in the fossil record (Ph.D. thesis). Harvard University,
Bidigare RR, Vetter RD (1987) Deep-sea hydrocarbon seep Cambridge, MA
communities: evidence for energy and nutritional carbon sources. Science Colaco A, Dehairs F, Desbruyères D (2002) Nutritional relations of deep-sea
238:1138–1142 hydrothermal fields at the Mid-Atlantic ridge: a stable isotope approach.
Buck KR, Barry JP, Simpson AGB (2000) Monterey Bay cold seep biota: Deep Sea Res Pt I Oceanogr Res Pap 49:395–412
Euglenozoa with chemoautotrophic bacterial epibionts. Eur J Protistol Conway N, Capuzzo J, Fry B (1989) The role of endosymbiotic bacteria in the
36:117–126 nutrition of Solemya velum: evidence from a stable isotope analysis of
Campbell BJ, Stein JL, Cary SC (2003) Evidence of chemolithoautotrophy in the endosymbionts and hosts. Limnol Oceanogr 34:249–255
bacterial community associated with Alvinella pompejana, a hydrothermal Conway N, Capuzzo JM (1991) Incorporation and utilization of bacterial lipids in
vent polychaete. Appl Environ Microbiol 69:5070–5078 the Solemya velum symbiosis. Mar Biol 108:277–292
Cary SC, Giovannoni SJ (1993) Transovarial inheritance of endosymbiotic bac- Conway NM, McDowell Capuzzo JE (1992) High taurine levels in the Solemya
teria in clams inhabiting deep-sea hydrothermal vents and cold seeps. Proc velum symbiosis. Comp Biochem Physiol B 102:175–185
Natl Acad Sci USA 90:5695–5699 Conway N, Kennicutt MC, Van Dover CL (1992) Stable isotopes in the study of
Cary SC, Warren W, Anderson E, Giovannoni SJ (1993) Identification and marine chemosynthesis-based ecosystems. In: Lajtha K, Michener R (eds)
localization of bacterial endosymbionts in hydrothermal vent taxa with Stable isotopes in ecology. Blackwell, Oxford, UK
symbiont-specific polymerase chain reaction amplification and in situ Corliss JB, Dymond J, Gordon LI, Edmond JM, Herzen RPV, Ballard RD, Green
hybridization techniques. Mol Mar Biol Biotechnol 2:51–62 K, Williams D, Bainbridge A, Crane K, Van Andel TH (1979) Submarine
Cary SC (1994) Vertical transmission of a chemoautotrophic aymbiont in the thermal springs on the Galapagos Rift. Science 203:1073–1083
protobranch bivalve, Solemya reidi. Mol Mar Biol Biotechnol 3:121–130 Craddock C, Hoeh WR, Lutz RA, Vrijenhoek RC (1995) Extensive gene flow
Cary SC, Cottrell MT, Stein JL, Camacho F, Desbruyères D (1997) Molecular among mytilid (Bathymodiolis thermophilus) populations from hydrother-
identification and localization of filamentous symbiotic bacteria associated mal vents of the Eastern Pacific. Mar Biol 124:137–146
with the hydrothermal vent annelid Alvinella pompejana. Appl Environ Dale C, Wang B, Moran N, Ochman H (2003) Loss of DNA recombinational
Microbiol 63:1124–1130 repair enzymes in the initial stages of genome degeneration. Mol Biol
Cavanaugh CM, Gardiner SL, Jones ML, Jannasch HW, Waterbury JB (1981) Pro- Evol 20:1188–1194
karyotic cells in the hydrothermal vent tube worm Riftia pachyptila Jones: Dando PR, Southward AJ (1986) Chemoautotrophy in the bivalve molluscs of the
possible chemoautotrophic symbionts. Science 213:340–342 genus Thyasira. J Mar Biol Assoc UK 66:915–929
Cavanaugh CM (1983) Symbiotic chemoautotrophic bacteria in marine inverte- Dando PR, Spiro B (1993) Varying nutritional dependence of the thyasirid
brates from sulfide-rich habitats. Nature 302:58–61 bivalves Thyasira sarsi and T equalis on chemoautotrophic symbiotic
bacteria, demonstrated by isotope ratios of tissue carbon and shell carbonate. Durand P, Gros O (1996) Bacterial host specificity of Lucinacea endosymbionts:
Mar Ecol Progr Ser 92:151–158 interspecific variation in 16S rRNA sequences. FEMS Microbiol Lett
De Bary A (1879) Die Erscheinung der Symbiose. Verlag von Karl J Trubner, 140:193–198
Strasburg Durand P, Gros O, Frenkiel L, Prieur D (1996) Phylogenetic characterization of
de Burgh ME, Singla CL (1984) Bacterial colonization and endocytosis on the gill sulfur-oxidizing bacterial endosymbionts in three tropical Lucinidae by 16S
of a new limpet species from a hydrothermal vent. Mar Biol 84:1–6 rDNA sequence analysis. Mol Mar Biol Biotechnol 5:37–42
de Burgh ME, Juniper SK, Singla CL (1989) Bacterial symbiosis in northeast Dykhuizen DE, Green L (1991) Recombination in Escherichia coli and the defi-
Pacific vestimentifera—a TEM study. Mar Biol 101:97–105 nition of biological species. J Bacteriol 173:7257–7268
De Cian MC, Andersen AC, Bailly X, Lallier FH (2003a) Expression and locali- Elderfield H, Schultz A (1996) Mid-ocean ridge hydrothermal fluxes and
zation of carbonic anhydrase and ATPases in the symbiotic tubeworm Riftia the chemical composition of the ocean. Annu Rev Earth Planet Sci
pachyptila. J Exp Biol 206:399–409 24:191–224
De Cian MC, Bailly X, Morales J, Strub JM, Van Dorsselaer A, Lallier FH (2003b) Elsaied H, Kimura H, Naganuma T (2002) Molecular characterization and
Characterization of carbonic anhydrases from Riftia pachyptila, a symbiotic endosymbiotic localization of the gene encoding D-ribulose 1,5-
invertebrate from deep-sea hydrothermal vents. Prot Struct Funct Genet bisphosphate carboxylase-oxygenase (RuBisCO) form II in the deep-sea
51:327–339 vestimentiferan trophosome. Microbiology 148:1947–1957
Degnan PH, Lazarus AB, Brock CD, Wernegreen JJ (2004) Host-symbiont stabil- Endow K, Ohta S (1989) The symbiotic relationship between bacteria and
ity and fast evolutionary rates in an ant-bacterium association: cospeciation the mesogastropod snail, Alviniconcha hessleri, collected from
of Camponotus species and their endosymbionts, Candidatus blochmannia. hydrothermal vents of the Mariana back-arc basin. Bull Jpn Soc Microb
System Biol 53:95–110 Ecol 3:73–82
DeNiro MJ, Epstein S (1979) Relationship between the oxygen isotope Endow K, Ohta S (1990) Occurrence of bacteria in the primary
ratios of terrestrial plant cellulose, carbon dioxide, and water. Science oocytes of vesicomyid clam Calyptogena soyoae. Mar Ecol Progr Ser
204:51–53 64:309–311
Desbruyères D, Gaill F, Laubier L, Prieur D, Rau GH (1983) Unusual nutrition of Felbeck H (1981) Chemoautotrophic potential of the hydrothermal vent tube
the ‘‘Pompeii worm’’ Alvinella pompejana (polychaetous annelid) from worm, Riftia pachyptila Jones (Vestimentifera). Science 213:336–338
a hydrothermal vent environment: SEM, TEM, 13C and 15 N evidence. Mar Felbeck H, Childress JJ, Somero GN (1981) Calvin-Benson cycle and sulfide
Biol 75:201–205 oxidation enzymes in animals from sulfide-rich habitats. Nature
Desbruyères D, Gaill F, Laubier L, Fouquet Y (1985) Polychaetous annelids from 293:291–293
hydrothermal vent ecosystems: an ecological overview. Biol Soc Wash Bull Felbeck H (1983) Sulfide oxidation and carbon fixation by the gutless clam
6:103–116 Solemya reidi: an animal-bacteria symbiosis. J Comp Physiol 152:3–11
Dickson AG, Millero FJ (1987) A comparison of the equilibrium constants for the Felbeck H, Liebezeit G, Dawson R, Giere O (1983) CO2 fixation in tissues of
dissociation of carbonic acid in seawater media. Deep Sea Res Pt A Oceanogr marine oligochaetes (Phallodrilus leukodermatus and P. planus) containing
Res Pap 34:1733–1743 symbiotic, chemoautotrophic bacteria. Mar Biol 75:187–191
Di Meo CA, Wilbur AE, Holben WE, Feldman RA, Vrijenhoek RC, Cary SC Felbeck H (1985) CO2 fixation in the hydrothermal vent tube worm Riftia
(2000) Genetic variation among endosymbionts of widely distributed vesti- pachyptila (Jones). Physiol Zool 58:272–281
mentiferan tubeworms. Appl Environ Microbiol 66:651–658 Felbeck H, Distel DL (1991) Prokaryotic symbionts of marine invertebrates. In:
Distel DL, Cavanaugh CM (1994) Independent phylogenetic origins of Balows A, Truper HG, Dworkin M, Harder W, Schleifer K-H (eds) The
methanotrophic and chemoautotrophic bacterial endosymbioses in marine prokaryotes. Springer, New York
bivalves. J Bacteriol 176:1932–1938 Felbeck H, Turner PJ (1995) CO2 transport in catheterized hydrothermal vent
Distel DL, Lee HKW, Cavanaugh CM (1995) Intracellular coexistence of tubeworms, Riftia pachyptila (Vestimentifera). J Exp Zool 272:95–102
methano-and thioautotrophic bacteria in a hydrothermal vent mussel. Felbeck H, Jarchow J (1998) Carbon release from purified chemoautotrophic
Proc Natl Acad Sci USA 92:9598–9602 bacterial symbionts of the hydrothermal vent tubeworm Riftia pachyptila.
Distel D (1998) Evolution of chemoautotrophic endosymbioses in bivalves. Physiol Zool 71:294–302
Bioscience 48:277–286 Felbeck H, Arndt C, Hentschel U, Childress JJ (2004) Experimental application of
Dubilier N, Giere O, Grieshaber MK (1995) Morphological and ecophysiological vascular and coelomic catheterization to identify vascular transport mecha-
adaptations of the marine oligochaete Tubificoides benedii to sulfidic sedi- nisms for inorganic carbon in the vent tubeworm, Riftia pachyptila. Deep Sea
ments. Am Zool 35:163–173 Res Pt I Oceanogr Res Pap 51:401–411
Dubilier N, Windoffer R, Giere O (1998) Ultrastructure and stable carbon isotope Feldman R, Black M, Cary C, Lutz R, Vrijenhoek R (1997) Molecular phyloge-
composition of the hydrothermal vent mussels Bathymodiolus brevior and netics of bacterial endosymbionts and their vestimentiferan hosts. Mol Mar
B. sp. affinis brevior from the North Fiji Basin, western Pacific. Mar Ecol Biol Biotechnol 6:268–277
Progr Ser 165:187–193 Fenchel T, Finlay BJ (1989) Kentrophoros—a mouthless ciliate with a symbiotic
Dubilier N, Amann R, Erseus C, Muyzer G, Park SY, Giere O, Cavanaugh CM kitchen garden. Ophelia 30:75–93
(1999) Phylogenetic diversity of bacterial endosymbionts in the gutless Fiala-Médioni A (1984) Ultrastructural evidence of abundance of intracellular
marine oligochete Olavius loisae (Annelida). Mar Ecol Progr Ser symbiotic bacteria in the gill of bivalve mollusks of deep hydrothermal vents.
178:271–280 Comptes Rendus L’Acad Sci Ser III Sciences de la Vie—Life Sciences
Dubilier N, Mulders C, Ferdelman T, de Beer D, Pernthaler A, Klein M, Wagner 298:487–492
M, Erseus C, Thiermann F, Krieger J, Giere O, Amann R (2001) Endosym- Fiala-Médioni A, Boulègue J, Ohta S, Felbeck H, Mariotti A (1993) Source of
biotic sulphate-reducing and sulphide-oxidizing bacteria in an oligochaete energy sustaining the Calyptogena populations from deep trenches in sub-
worm. Nature 411:298–302 duction zones off Japan. Deep Sea Res 40:1241–1258
Dufour SC, Felbeck H (2003) Sulphide mining by the superextensile foot of Fiala-Médioni A, Michalski J-C, Jollès J, Alonso C, Montreuil J (1994) Lysosomic
symbiotic thyasirid bivalves. Nature 426:65–67 and lysozyme activities in the gill of bivalves from deep hydrothermal vents.
Dunton K (1992) Arctic biogeography: the paradox of the marine benthic fauna Comptes Rendus L’Acad Sci Ser III Sciences de la Vie—Life Sciences
and flora. Trends Ecol Evol 7:183–189 317:239–244
Duperron S, Nadalig T, Caprais J, Sibuet M, Fiala-Medioni A, Amann R, Dubilier Fiala-Médioni A, McKiness ZP, Dando P, Boulegue J, Mariotti A, Alayse-Danet
N (2004) Dual symbiosis in a Bathymodiolus mussel from a methane seep on AM, Robinson JJ, Cavanaugh CM (2002) Ultrastructural, biochemical, and
the Gabon continental margin (South East Atlantic): 16S rRNA phylogeny immunological characterization of two populations of the mytilid mussel
and distribution of the symbionts in the gills. Appl Environ Microbiol Bathymodiolus azoricus from the Mid-Atlantic Ridge: evidence for a dual
71(4):1694–1700 symbiosis. Mar Biol 141:1035–1043
Fisher CR, Childress JJ (1986) Translocation of fixed carbon from symbiotic vestimentiferan tubeworm, Riftia pachyptila and Lamellibrachia cf luymesi:
bacteria to host tissues in the gutless bivalve, Solemya reidi. Mar Biol proton elimination is a necessary adaptation to sulfide-oxidizing chemoau-
93:59–68 totrophic symbionts. J Exp Biol 205:3055–3066
Fisher C, Childress J, Arp A, Brooks J, Distel D, Favuzzi J, Felbeck H, Hessler R, Glockner FO, Babenzien HD, Wulf J, Amann R (1999) Phylogeny and diversity of
Johnson K, Kennicut M 2nd, Macko S, Newton A, Powell M, Somero G, Soto Achromatium oxaliferum system. Appl Microbiol 22:28–38
T (1988) Microhabitat variation in the hydrothermal vent mussel, Goericke R, Montoya JP, Fry B (1994) Physiology of isotopic fractionation in
Bathymodiolus thermophilus, at the Rose Garden vent on the Galapagos algae and cyanobacteria. In: Lajtha K, Michener RH (eds) Stable isotopes in
Rift. Deep Sea Res 35:1769–1791 ecology and environmental science. Blackwell, Oxford, UK
Fisher CR (1990) Chemoautotrophic and methanotrophic symbioses in marine Goffredi SK, Childress JJ, Desaulniers NT, Lallier FH (1997a) Sulfide acquisition
invertebrates. Rev Aquat Sci 2:399–436 by the hydrothermal vent tubeworm Riftia pachyptila appears to be via
Fisher CR, Childress JJ (1992) Organic carbon transfer from methanotrophic uptake of HS , rather than H2S. J Exp Biol 200:2069–2616
symbionts to the host hydrocarbon seep mussel. Symbiosis 12:221–235 Goffredi SK, Childress JJ, Desaulniers NT, Lee RW, Lallier FH, Hammond
Fisher CR (1993) Oxidation of methane by deep-sea mytilids in the Gulf of D (1997b) Inorganic carbon acquisition by the hydrothermal vent tubeworm
Mexico. In: Oremland RS (ed) Biogeochemistry of global change: radiatively Riftia pachyptila depends upon high external pCO2 and upon proton-
active trace gases. Chapman and Hall, New York equivalent ion transport by the worm. J Exp Biol 200:883–896
Fisher CR, Brooks JM, Vodenichar JS, Zande JM, Childress JJ, Burke RA Jr Goffredi SK, Childress JJ, Lallier FH, Desaulniers NT (1999) The ionic composi-
(1993) The co-occurrence of methanotrophic and chemoautotrophic sul- tion of the hydrothermal vent tube worm Riftia pachyptila: evidence for the
fur-oxidizing bacterial symbionts in a deep-sea mussel. Marine Ecology elimination of SO42 and H+ and for a Cl /HCO3 shift. Physiol Biochem
14:277–289, Pubblicazioni Della Stazione Zoologica Di Napoli I Zool 72:296–306
Fisher CR, Childress JJ, Macko SA, Brooks JM (1994) Nutritional interactions in Goffredi SK, Childress JJ (2001) Activity and inhibitor sensitivity of ATPases in
Galapagos Rift hydrothermal vent communities: inferences from stable car- the hydrothermal vent tubeworm Riftia pachyptila: a comparative approach.
bon and nitrogen isotope analyses. Mar Ecol Progr Ser 103:45–55 Mar Biol 138:259–265
Fisher CR (1995) Toward an appreciation of hydrothermal vent animals: their Grassle JF (1985) Hydrothermal vent animals: distribution and biology. Science
environment, physiological ecology, and tissue stable isotope values. In: 229:713–717
Seafloor hydrothermal systems: physical, chemical, biological, and geological Gray ND, Howarth R, Rowan A, Pickup RW, Jones JG, Head IM (1999) Natural
interactions. American Geophysical Union communities of Achromatium oxaliferum comprise genetically, morpholog-
Fisher CR (1996) Ecophysiology of primary production at deep-sea vents and ically, and ecologically distinct subpopulations. Appl Environ Microbiol
seeps. Biosystem Ecol Ser 11:313–336 65:5089–5099
Fox M, Juniper SK, Vali H (2002) Chemoautotrophy as a possible nutritional Grieshaber MK, Volkel S (1998) Animal adaptations for tolerance and exploita-
source in the hydrothermal vent limpet Lepetodrilus fucensis. Cahiers Biol tion of poisonous sulfide. Annu Rev Physiol 60:33–53
Mar 43:371–376 Gros O, Darrasse A, Durand P, Frenkiel L, Moueza M (1996) Environmental
Freytag JK, Girguis PR, Bergquist DC, Andras JP, Childress JJ, Fisher CR transmission of a sulfur-oxidizing bacterial gill endosymbiont in the
(2001) A paradox resolved: sulfide acquisition by roots of seep tubeworms tropical lucinid bivalve Codakia orbicularis. Appl Environ Microbiol
sustains net chemoautotrophy. Proc Natl Acad Sci USA 98:13408–13413 62:2324–2330
Fry B, Gest H, Hayes JM (1983) Sulfur isotopic compositions of deep-sea hydro- Gros O, De Wulf-Durand P, Frenkiel L, Moueza M (1998) Putative environmental
thermal vent animals. Nature 306:51–52 transmission of sulfur-oxidizing bacterial symbionts in tropical lucinid
Fry B, Sherr EB (1984) d13C measurements as indicators of carbon flow in marine bivalves inhabiting various environments. FEMS Microbiol Lett
and freshwater ecosystems. Contrib Mar Sci 27:13–47 160:257–262
Fujio SZ, Imasato N (1991) Diagnostic calculation for circulation and water mass Gros O, Liberge M, Felbeck H (2003a) Interspecific infection of aposymbiotic
movement in the deep Pacific. J Geophys Res Oceans 96:759–774 juveniles of Codakia orbicularis by various tropical lucinid gill endosymbi-
Fujiwara Y, Kato C, Masui N, Fujikura K, Kojima S (2001) Dual symbiosis in the onts. Mar Biol 142:57–66
cold-seep thyasirid clam Maorithyas hadalis from the hadal zone in the Japan Gros O, Liberge M, Heddi A, Khatchadourian C, Felbeck H (2003b) Detection of
Trench, western Pacific. Mar Ecol Progr Ser 214:151–159 the free-living forms of sulfide-oxidizing gill endosymbionts in the lucinid
Fujiwara Y, Uematsu K (2002) Microdistribution of two endosymbionts in gill habitat (Thalassia testudinum environment). Appl Environ Microbiol
tissue from a hadal thyasirid clam Maorithyas hadalis. Cahiers Biol Mar 69:6264–6267
43:341–343 Gustafson RG, Reid RGB (1988) Association of bacteria with larvae of the gutless
Fustec A, Desbruyères D, Juniper SK (1987) Deep-sea hydrothermal vent com- protobranch bivalve Solemya reidi (Cryptodonta, Solemyidae). Mar Biol
munities at 13 N on the East Pacific Rise: microdistribution and temporal 97:389–401
variations. Biol Oceanogr 4:121–164 Hand SC (1987) Trophosome ultrastructure and the characterization of isolated
Gearing JN, Gearing PJ, Rudnick DT, Requejo AG, Hutchins MJ (1984) Isotopic bacteriocytes from invertebrate-sulfur bacteria symbioses. Biol Bull
variability of organic-carbon in a phytoplankton-based, temperate estuary. 173:260–276
Geochim Cosmochim Acta 48:1089–1098 Head IM, Gray ND, Clarke KJ, Pickup RW, Jones JG (1996) The phylogenetic
Giere O (1981) The gutless marine oligochaete Phallodrilus leukodermatus: struc- position and ultrastructure of the uncultured bacterium Achromatium
tural studies on an aberrant tubificid associated with bacteria. Mar Ecol oxaliferum. Microbiol SGM 142:2341–2354
Progr Ser 5:353–357 Hentschel U, Felbeck H (1993) Nitrate respiration in the hydrothermal vent
Giere O (1985) The gutless marine tubificid Phallodrilus planus, a flattened tubeworm Riftia pachyptila. Nature 366:338–340
oligochaete with symbiotic bacteria. Zool Scripta 14:279–286 Hentschel U, Cary SC, Felbeck H (1993) Nitrate respiration in chemoautotrophic
Giere O, Langheld C (1987) Structural organization, transfer and biological fate symbionts of the bivalve Lucinoma aequizonata. Mar Ecol Progr Ser 94:35–41
of endosymbiotic bacteria in gutless oligochaetes. Mar Biol 93:641–650 Herry A, Le Pennec M (1987) Endosymbiotic bacteria in the gills of the littoral
Giere O, Krieger J (2001) A triple bacterial endosymbiosis in a gutless oligochaete bivalve molluscs Thyasira flexuosa (Thyasiridae) and Lucinella divaricata
(Annelida): ultrastructural and immunocytochemical evidence. Invertebr (Lucinidae). Symbiosis 4:25–36
Biol 120:41–49 Hughes DS, Felbeck H, Stein JL (1997) A histidine protein kinase homolog from
Girguis PR, Lee RW, Desaulniers N, Childress JJ, Pospesel M, Felbeck H, Zal the endosymbiont of the hydrothermal vent tubeworm Riftia pachyptila.
F (2000) Fate of nitrate acquired by the tubeworm Riftia pachyptila. Appl Appl Environ Microbiol 63:3494–3498
Environ Microbiol 66:2783–2790 Hughes DS, Felbeck H, Stein JL (1998) Signal transduction and motility genes
Girguis PR, Childress JJ, Freytag JK, Klose K, Stuber R (2002) Effects of metabolite from the bacterial endosymbionts of Riftia pachyptila. Cahiers Biol Mar
uptake on proton-equivalent elimination by two species of deep-sea 39:305–308
Humes AG, Lutz RA (1994) Aphotopontius acanthinus, new species (Copepoda, Laue BE, Nelson DC (1994) Characterization of the gene encoding the autotro-
Siphonostomatoida), from deep-sea hydrothermal vents on the East Pacific phic ATP sulfurylase from the bacterial endosymbiont of the hydrothermal
Rise. J Crustac Biol 14:337–345 vent tubeworm Riftia pachyptila. J Bacteriol 176:3723–3729
Hunt S (1992) Structure and composition of the shell of the archaeogastropod Laue BE, Nelson DC (1997) Sulfur-oxidizing symbionts have not co-evolved with
limpet Lepetodrilus elevatus-elevatus (Mclean, 1988). Malacologia their hydrothermal vent tubeworm hosts: an RFLP analysis. Mol Mar Biol
34:129–141 Biotechnol 6:180–188
Hurtado LA (2002) Evolution and biogeography of hydrothermal vent organisms Lee R, Childress J (1994) Assimilation of inorganic nitrogen by marine inverte-
in the Eastern Pacific Ocean, PhD thesis, Rutgers University, New Brunswick brates and their chemoautotrophic and methanotrophic symbionts. Appl
Hurtado LA, Mateos M, Lutz RA, Vrijenhoek RC (2003) Coupling of bacterial Environ Microbiol 60:1852–1858
endosymbiont and host mitochondrial genomes in the hydrothermal vent Lee RW, Childress JJ, Desaulniers NT (1997) The effects of exposure to ammonia
clam Calyptogena magnifica. Appl Environ Microbiol 69:2058–2064 on ammonia and taurine pools of the symbiotic clam Solemya reidi. J Exp
Jacobs DK, Lindberg DR (1998) Oxygen and evolutionary patterns in the sea: Biol 200:2797–2805
onshore/offshore trends and recent recruitment of deep-sea faunas. Proc Lee RW, Robinson JJ, Cavanaugh CM (1999) Pathways of inorganic nitrogen
Natl Acad Sci USA 95:9396–9401 assimilation in chemoautotrophic bacteria-marine invertebrate symbioses:
Johnson KS, Childress JJ, Hessler RR, Sakamoto-Arnold CM, Beehler CL expression of host and symbiont glutamine synthetase. J Exp Biol
(1988) Chemical and biological interactions in the Rose Garden hydrother- 202:289–300
mal vent field, Galapagos spreading center. Deep Sea Res 35:1723–1744 Le Pennec M, Hily A (1984) Anatomy, structure and ultrastructure of the gill of
Johnson M, Diouris M, Le Pennec M (1994) Endosymbiotic bacterial contribu- a deep-sea hydrothermal vent mytilid. Oceanol Acta 7:517–523
tion in the carbon nutrition of Loripes lucinalis (Mollusca: Bivalvia). Sym- Levin BR, Bergstrom CT (2000) Bacteria are different: observations, interpreta-
biosis 17:1–13 tions, speculations, and opinions about the mechanisms of adaptive evolu-
Jones ML (1981) Riftia pachyptila Jones: observations on the vestimentiferan tion in prokaryotes. Proc Natl Acad Sci USA 97:6981–6985
worm from the Galapagos Rift. Science 213:333–336 Levin LA, Michener RH (2002) Isotopic evidence for chemosynthesis-based
Jones JL, Gardiner SL (1985) Light and scanning electron microscope studies of nutrition of macrobenthos: the lightness of being at Pacific methane seeps.
spermatogenesis in the vestimentiferan tube worm Riftia pachyptila Limnol Oceanogr 47:1336–1345
(Pogonophora: Obturata). Trans Am Microscop Soc 104:1–18 Li L, Kato C, Horikoshi K (1999) Microbial diversity in sediments collected from
Jørgensen BB, Postgate JR (1982) Ecology of the bacteria of the sulfur cycle with the deepest cold-seep area, the Japan Trench. Mar Biotechnol 1:391–400
special reference to anoxic oxic interface environments. Philos Trans R Soc Lilley MD, Butterfield DA, Olson EJ, Lupton JE, Macko SA, McDuff RE
Lond Ser B Biol Sci 298:543–561 (1993) Anomalous CH4 and NH4+ concentrations at an unsedimented
Joyner JL, Peyer SM, Lee RW (2003) Possible roles of sulfur-containing amino mid-ocean-ridge hydrothermal system. Nature 364:45–47
acids in a chemoautotrophic bacterium-mollusc symbiosis. Biol Bull Little CTS, Herrington RJ, Maslennikov VV, Morris NJ, Zaykov VV (1997) Silu-
205:331–338 rian hydrothermal-vent community from the southern Urals Russia. Nature
Julian D, Gaill F, Wood E, Arp AJ, Fisher CR (1999) Roots as a site of hydrogen 385:146–148
sulfide uptake in the hydrocarbon seep vestimentiferan Lamellibrachia sp. Little CL (2002) The fossil record of hydrothermal vent communities. Cahiers
J Exp Biol 202:2245–2257 Biol Mar 43:313–316
Karl SA, Schutz S, Desbruyères D, Lutz R, Vrijenhoek RC (1996) Molecular Little CTS, Thorseth IH (2002) Hydrothermal vent microbial communities:
analysis of gene flow in the hydrothermal vent clam (Calyptogena magnifica). a fossil perspective. Cahiers Biol Mar 43:317–319
Mol Mar Biol Biotechnol 5:193–202 Little CTS, Vrijenhoek RC (2003) Are hydrothermal vent animals living fossils?
Kennicutt MC, Burke RA, MacDonald IR, Brooks JM, Denoux GJ, Macko SA Trends Ecol Evol 18:582–588
(1992) Stable isotope partitioning in seep and vent organisms: chemical and Little CTS, Danelian T, Herrington RJ, Haymon RM (2004) Early Jurassic hydro-
ecological significance. Chem Geol 101:293–310 thermal vent community from the Franciscan Complex, California.
Kennish MJ, Lutz RA (1999) Calcium carbonate dissolution rates in deep-sea J Paleontol 78:542–559
bivalve shells on the East Pacific Rise at 21 N: results of an 8-year in situ Lonsdale P (1977) Clustering of suspension-feeding macrobenthos near abyssal
experiment. Palaeogeogr Palaeoclimatol Palaeoecol 154:293–299 hydrothermal vents at oceanic spreading centers. Deep Sea Res 24:857
Kimura M, Weiss GH (1964) The stepping stone model of population Lopez-Garcia P, Gaill F, Moreira D (2002) Wide bacterial diversity
structure and the decrease of genetic correlation with distance. Genetics associated with tubes of the vent worm Riftia pachyptila. Environ Microbiol
49:561–576 4:204–215
Kochevar RE, Childress JJ (1996) Carbonic anhydrase in deep-sea chemoauto- Luther GWI, Rozan TF, Taillefert M, Nuzzio DB, Di Meo C, Shank TM, Lutz RA,
trophic symbioses. Mar Biol 125:375–383 Cary SC (2001) Chemical speciation drives hydrothermal vent ecology.
Krieger J, Giere O, Dubilier N (2000) Localization of RubisCO and sulfur in Nature 410:813–816
endosymbiotic bacteria of the gutless marine oligochaete Inanidrilus Lutz RA, Jablonski D, Rhoads DC, Turner RD (1980) Larval dispersal of a deep-sea
leukodermatus (Annelida). Mar Biol 137:239–244 hydrothermal vent bivalve from the Galapagos Rift. Mar Biol 57:127–133
Krueger DM (1996a) Ecology and evolution of Solemya spp., marine bivalves MacAvoy SE, Macko SA, Joye SB (2002) Fatty acid carbon isotope signatures in
living in symbiosis with chemoautotrophic bacteria, PhD thesis, Harvard chemosynthetic mussels and tubeworms from Gulf of Mexico hydrocarbon
University Cambridge, MA seep communities. Chem Geol 185:1–8
Krueger DM, Dubilier N, Cavanaugh CM (1996) Chemoautotrophic symbiosis in Marsh AG, Mullineaux LS, Young CM, Manahan DT (2001) Larval dispersal
the tropical solemyid Solemya occidentalis (Bivalvia: Protobranchia): ultra- potential of the tubeworm Riftia pachyptila at deep-sea hydrothermal
structural and phylogenetic analysis. Mar Biol 126:55–64 vents. Nature 411:77–80
Krueger DM, Cavanaugh CM (1997) Phylogenetic diversity of bacterial symbi- McArthur AG, Tunnicliffe V (1998) Relics and antiquity revisited in the modern
onts of Solemya hosts based on comparative sequence analysis of 16S rRNA vent fauna. In: Mills R, Harrison K (eds) Modern ocean floor processes and
genes. Appl Environ Microbiol 63:91–98 the geological record. Geological Society of London, London
Lalou C, Brichet E (1982) Ages and implications of East Pacific Rise sulfide McKiness ZP (2004) Evolution of endosymbioses in deep-sea bathymodioline
deposits at 21 N. Nature 300:169–171 mussels (Mollusca:Bivalvia) (Ph.D. thesis). Harvard University, Cambridge,
Laubier L (1989) Ecosystèmes benthiques profonds et chimiosynthèse MA
bactérienne: Sources hydrothermales et suintement froids (Deep benthic McMullin ER, Hourdez S, Schaeffer SW, Fisher CR (2003) Phylogeny and bioge-
ecosystems and bacterial chemosynthesys: hydrothermal wells and cold ography of deep-sea vestimentiferan tubeworms and their bacterial symbi-
seepages). In: Denis M (ed) Oceanology: today and in the near future onts. Symbiosis 34:1–41
Metivier B, Voncosel R (1993) Acharax alinae n sp., a giant solemyid (Mollusca, Podar M, Mullineaux L, Huang HR, Perlman PS, Sogin ML (2002) Bacterial
Bivalvia) from the Lau Basin. Comptes Rendus L’Acad Sci Ser III Sciences de group II introns in a deep-sea hydrothermal vent environment. Appl Environ
la Vie—Life Sciences 316:229–237 Microbiol 68:6392–6398
Michener RH, Schell DM (1994) Stable isotope ratios as tracers in marine aquatic Polz MF, Felbeck H, Novak R, Nebelsick M, Ott JA (1992) Chemoautotrophic,
food webs. In: Lajtha K, Michener RH (eds) Stable isotopes in ecology and sulfur-oxidizing symbiotic bacteria on marine nematodes: morphological
environmental science. Blackwell, Oxford, UK and biochemical characterization. Microb Ecol 24:313–329
Millero FJ, Plese T, Fernandez M (1987) The dissociation of hydrogen-sulfide in Polz MF, Distel DL, Zarda B, Amann R, Felbeck H, Ott JA, Cavanaugh CM
seawater. Limnol Oceanogr 33:269–274 (1994) Phylogenetic analysis of a highly specific association between
Millikan DS, Felbeck H, Stein JL (1999) Identification and characterization of ectosymbiotic, sulfur-oxidizing bacteria and a marine nematode. Appl Envi-
a flagellin gene from the endosymbiont of the hydrothermal vent tubeworm ron Microbiol 60:4461–4467
Riftia pachyptila. Appl Environ Microbiol 65:3129–3133 Polz MF, Cavanaugh CM (1995) Dominance of one bacterial phylotype at a Mid-
Minagawa M, Wada E (1984) Stepwise enrichment of 15N along food chains— Atlantic Ridge hydrothermal vent site. Proc Natl Acad Sci USA 92:7232–7236
further evidence and the relationship between delta 15N and animal age. Polz MF, Robinson JJ, Cavanaugh CM, Van Dover CL (1998) Trophic ecology of
Geochim Cosmochim Acta 48:1135–1140 massive shrimp aggregations at a Mid-Atlantic Ridge hydrothermal vent site.
Minic Z, Simon V, Penverne B, Gaill F, Herve G (2001) Contribution of the Limnol Oceanogr 43:1631–1638
bacterial endosymbiont to the biosynthesis of pyrimidine nucleotides in the Polz MF, Harbison C, Cavanaugh CM (1999) Diversity and heterogeneity of
deep-sea tubeworm Riftia pachyptila. J Biol Chem 276:23777–23784 epibiotic bacterial communities on the marine nematode Eubostrichus
Minic Z, Herve G (2003) Arginine metabolism in the deep-sea tube worm Riftia dianae. Appl Environ Microbiol 65:4271–4275
pachyptila and its bacterial endosymbiont. J Biol Chem 278:40527–40533 Polz MF, Ott JA, Bright M, Cavanaugh CM (2000) When bacteria hitch a ride.
Mira A, Moran NA (2002) Estimating population size and transmission bottle- ASM News 66:531–539
necks in maternally transmitted endosymbiotic bacteria. Microb Ecol Poulin E, von Boletzky S, Feral JP (2001) Combined ecological factors permit
44:137–143 classification of developmental patterns in benthic marine invertebrates:
Miura T, Nedachi M, Hashimoto A (2002) Sulphur sources for chemoautotrophic a discussion note. J Exp Mar Biol Ecol 257:109–115
nutrition of shallow water vestimentiferan tubeworms in Kagoshima Bay. Powell MA, Somero GN (1986) Adaptations to sulfide by hydrothermal vent
J Mar Biol Assoc UK 82:537–540 animals: sites and mechanisms of detoxification and metabolism. Biol Bull
Moran N, Baumann P (1994) Phylogenetics of cytoplasmically inherited micro- 171:274–290
organisms of arthropods. Trends Ecol Evol 9:15–20 Pradillon F, Shillito B, Young CM, Gaill F (2001) Deep-sea ecology: developmen-
Moran NA (1996) Accelerated evolution and Muller’s rachet in endosymbiotic tal arrest in vent worm embryos. Nature 413:698–699
bacteria. Proc Natl Acad Sci USA 93:2873–2878 Prior SD, Dalton H (1985) The effect of copper ions on membrane content and
Moran NA (2003) Tracing the evolution of gene loss in obligate bacterial symbi- methane monooxygenase activity in methanol-grown cells of Methylococcus
onts. Curr Opin Microbiol 6:512–518 capsulatus (Bath). J Gen Microbiol 131:155–163
Muller HJ (1964) The relation of recombination to mutational advance. Mut Res Pruski AM, Fiala-Médioni A, Fisher CR, Colomines JC (2000a) Composition
1:2–9 of free amino acids and related compounds in invertebrates with
Mullineaux LS, Wiebe PH, Baker ET (1995) Larvae of benthic invertebrates in symbiotic bacteria at hydrocarbon seeps in the Gulf of Mexico. Mar Biol
hydrothermal vent plumes over Juan de Fuca Ridge. Mar Biol 122:585–596 136:411–420
Mullineaux LS, Speer KG, Thurnherr AM, Maltrud ME, Vangriesheim A (2002) Pruski AM, Fiala-Médioni A, Prodon R, Colomines JC (2000b) Thiotaurine is
Implications of cross-axis flow for larval dispersal along mid-ocean ridges. a biomarker of sulfide-based symbiosis in deep-sea bivalves. Limnol
Cahiers Biol Mar 43:281–284 Oceanogr 45:1860–1867
Nelson D, Fisher C (1995) Chemoautotrophic and methanotrophic endosymbi- Pruski AM, Fiala-Médioni A (2003) Stimulatory effect of sulphide on thiotaurine
otic bacteria at deep-sea vents and seeps. In: Karl D (ed) Microbiology of synthesis in three hydrothermal vent species from the East Pacific Rise. J Exp
deep-sea hydrothermal vents. CRC Press, Boca Raton Biol 206:2923–2930
Nelson DC, Hagen KD, Edwards DB (1995) The gill symbiont of the hydrother- Rau GA, Hedges JI (1979) Carbon-13 depletion in a hydrothermal vent mussel:
mal vent mussel Bathymodiolus thermophilus is a psychrophilic, chemo- suggestion of a chemosynthetic food source. Science 203:648–649
autotrophic, sulfur bacterium. Mar Biol 121:487–495 Rau GH (1981) Hydrothermal vent clam and tube worm 13C/12C: further evi-
Nelson K, Fisher CR (2000) Absence of cospeciation in deep-sea vestimentiferan dence of non photosynthetic food sources. Science 213:338–340
tubeworms and their bacterial endosymbionts. Symbiosis 28:1–15 Robinson JJ, Cavanaugh CM (1995) Expressions of form I and form II RubisCO
Niemczuk K, Kondracki M (2004) Q fever as zoonosis. Med Weter 60:129–131 in chemoautotrophic symbioses: implications for the interpretation of stable
O’Brien J, Vetter RD (1990) Production of thiosulphate during sulphide oxida- carbon isotope values. Limnol Oceanogr 40:1496–1502
tion by mitochondria of the symbiont-containing bivalve Solemya reidi. Robinson JJ, Polz MF, Fiala-Medioni A, Cavanaugh CM (1998) Physiological and
J Exp Biol 149:133–148 immunological evidence for two distinct C-1-utilizing pathways in
Ohta T (1973) Slightly deleterious mutant substitutions in evolution. Nature Bathymodiolus puteoserpentis (Bivalvia: Mytilidae), a dual endosymbiotic
246:96–98 mussel from the Mid-Atlantic Ridge. Mar Biol 132:625–633
Ott JA, Novak R, Schiemer F, Hentschel U, Nebelsick M, Polz M (1991) Tackling Robinson JJ, Scott KM, Swanson ST, O’Leary MH, Horken K, Tabita FR
the sulfide gradient: a novel strategy involving marine nematodes and che- (2003) Kinetic isotope effect and characterization of form II RubisCO
moautotrophic ectosymbionts. Mar Ecol 12:261–279, Pubblicazioni Della from the chemoautotrophic endosymbionts of the hydrothermal vent
Stazione Zoologica Di Napoli I tubeworm Riftia pachyptila. Limnol Oceanogr 48:48–54
Ott JA, Bright M, Schiemer F (1998) The ecology of a novel symbiosis between Rouxel O, Fouquet Y, Ludden JN (2004) Subsurface processes at the lucky strike
a marine peritrich ciliate and chemoautotrophic bacteria. Mar Ecol 19:229– hydrothermal field, Mid-Atlantic Ridge: evidence from sulfur, selenium, and
243, Pubblicazioni Della Stazione Zoologica Di Napoli I iron isotopes. Geochim Cosmochim Acta 68:2295–2311
Peek AS, Vrijenhoek RC, Gaut BS (1998) Accelerated evolutionary rate in sulfur- Sarrazin J, Juniper SK (1999) Biological characteristics of a hydrothermal edifice
oxidizing endosymbiotic bacteria associated with the mode of symbiont mosaic community. Mar Ecol Progr Ser 185:1–19
transmission. Mol Biol Evol 15:1514–1523 Schiemer F, Novak R, Ott J (1990) Metabolic studies on thiobiotic free-
Pimenov NV, Kalyuzhnaya MG, Khmelenina VN, Mityushina LL, Trotsenko YA living nematodes and their symbiotic microorganisms. Mar Biol
(2002) Utilization of methane and carbon dioxide by symbiotrophic bacteria 106:129–137
in gills of mytilidae (Bathymodiolus) from the rainbow and Logachev Schmaljohann R, Flügel HJ (1987) Methane-oxidizing bacteria in pogonophora.
hydrothermal fields on the Mid-Atlantic Ridge. Microbiology 71:587–594 Sarsia 72:91–98
Schweimanns M, Felbeck H (1985) Significance of the occurrence of chemoau- Vacelet J, Fiala-Médioni A, Fisher CR, Boury-Esnault N (1996) Symbiosis
totrophic bacterial endosymbionts in lucinid clams from Bermuda. Mar Ecol between methane-oxidizing bacteria and a deep-sea carnivorous cladorhizid
Progr Ser 24:113–120 sponge. Mar Ecol Progr Ser 145:77–85
Scott KM (2003) A d13C-based carbon flux model for the hydrothermal Van Dover CL, Fry B, Grassle JF, Humphris S, Rona PA (1988) Feeding biology of
vent chemoautotrophic symbiosis Riftia pachyptila predicts the shrimp Rimicaris exoculata at hydrothermal vents on the Mid-Atlantic
sizeable CO2 gradients at the host-symbiont interface. Environ Microbiol Ridge. Mar Biol 98:209–216
5:424–432 Van Dover CL, Fry B (1994) Microorganisms as food resources at deep-sea
Scott KM, Schwedock J, Schrag DP, Cavanaugh CM (2004) Influence of form IA hydrothermal vents. Limnol Oceanogr 39:51–57
RubisCO and environmental dissolved inorganic carbon on the delta 13C of Van Dover C (2000) The ecology of deep-sea hydrothermal vents. Princeton
the clam-chemoautotroph symbiosis Solemya velum. Environ Microbiol University Press, Princeton
6(12):1210–1219 Van Dover CL, Humphris SE, Fornari D, Cavanaugh CM, Collier R, Goffredi SK,
Searcy DG (1992) Origins of mitochondria and chloroplasts from sulfur-based Hashimoto J, Lilley M, Reysenbach AL, Shank TM, Von Damm KL, Banta A,
symbioses. In: Hartman H, Matsuno K (eds) The origin and evolution of the Gallant RM, Götz D, Green D, Hall J, Harmer TL, Hurtado LA, Johnson P,
cell world. Scientific Press, Singapore McKiness ZP, Meredith C, Olson E, Pan IL, Turnipseed M, Won Y, Young CR
Sibuet M, Olu K (1998) Biogeography, biodiversity and fluid dependence of deep- 3rd, Vrijenhoek RC (2001) Biogeography and ecological setting of Indian
sea cold-seep communities at active and passive margins. Deep Sea Res II Ocean hydrothermal vents. Science 294:818–822
45:517–567 Van Dover C (2002) Trophic relationships among invertebrates at the Kairei
Smith CR, Baco AR (2003) Ecology of whale falls at the deep-sea floor. Oceanogr hydrothermal vent field (Central Indian Ridge). Mar Biol 141:761–772
Mar Biol 41:311–354 Van Dover CL, German CR, Speer KG, Parson LM, Vrijenhoek RC (2002) Marine
Southward AJ, Southward EC, Dando PR, Rau GH, Felbeck H, Flugel H (1981) biology: evolution and biogeography of deep-sea vent and seep invertebrates.
Bacterial symbionts and low 13C/12C ratios in tissues of Pogonophora indi- Science 295:1253–1257
cate unusual nutrition and metabolism. Nature 293:616–620 Vereshchaka AL, Vinogradov GM, Lein AY, Dalton S, Dehairs F (2000) Carbon
Southward EC (1982) Bacterial symbionts in Pogonophora. J Mar Biol Assoc UK and nitrogen isotopic composition of the fauna from the Broken Spur
62:889–906 hydrothermal vent field. Mar Biol 136:11–17
Southward AJ, Southward EC (1988) Pogonophora: tube worms dependent on Vrijenhoek R, Shank T, Lutz R (1998) Gene flow and dispersal in deep-sea
endosymbiotic bacteria ISI Atlas of science. Animal & Plant Sciences 1:203– hydrothermal vent animals. Cahiers Biol Mar 39:363–366
207 Waser NAD, Harrison PJ, Nielsen B, Calvert SE, Turpin DH (1998) Nitrogen
Southward EC, Tunnicliffe V, Black M (1995) Revision of the species of Ridgeia isotope fractionation during the uptake and assimilation of nitrate, nitrite,
from North-East Pacific hydrothermal vents, with a redescription of Ridgeia ammonium, and urea by a marine diatom. Limnol Oceanogr 43:215–224
piscesae Jones (Pogonophora: Obturata = Vestimentifera). Can J Zool Watanabe A (2004) Various clinical types of Q-fever disease. Int Med 43:12
73:282–295 Weber RE, Vinogradov SN (2001) Nonvertebrate hemoglobins: functions and
Spiro B, Greenwood PB, Southward AJ, Dando PR (1986) 13C/12C ratios in molecular adaptations. Physiol Rev 81:569–628
marine invertebrates from reducing sediments: confirmation of nutritional Weiser W (1959) Eine ungewöhnliche Assoziation zwischen Blaualagen und
importance of chemoautotrophic endosymbiotic bacteria. Mar Ecol Progr freilebenden marinen. Nematoden Österr Bot Zeitschr 106:81–87
Ser 28:233–240 Wernegreen JJ (2002) Genome evolution in bacterial endosymbionts of insects.
Stanley SM (1970) Shell form and life habits of the Bivalvia. Geol Soc Am Mem Nat Rev Genet 3:850–861
125:119–121 Windoffer R, Giere O (1997) Symbiosis of the hydrothermal vent gastropod
Stein J, Cary S, Hessler R, Ohta S, Vetter R, Childress J, Felbeck H (1988) Ifremeria nautilei (Provannidae) with endobacteria: structural analyses and
Chemoautotrophic symbiosis in a hydrothermal vent gastropod. Biol Bull ecological considerations. Biol Bull 193:381–392
174:373–378 Wirsen CO, Sievert SM, Cavanaugh CM, Molyneaux SJ, Ahmad A, Taylor LT,
Stumm W, Morgan JJ (1996) Aquatic chemistry, 3rd edn. Wiley-Interscience, DeLong EF, Taylor CD (2002) Characterization of an autotrophic sulfide-
New York oxidizing marine Arcobacter sp. that produces filamentous sulfur. Appl
Svavarsson J, Stromberg JO, Brattegard T (1993) The deep-sea asellote (Isopoda, Environ Microbiol 68:316–325
Crustacea) fauna of the Northern Seas: species composition, distributional Woldehiwet Z (2004) Q fever (coxiellosis): epidemiology and pathogenesis. Res
patterns and origin. J Biogeogr 20:537–555 Vet Sci 77:93–100
Temara A, de Ridder C, Kuenen JG, Robertson LA (1993) Sulfide-oxidizing Won YJ, Hallam SJ, O’Mullan GD, Pan IL, Buck KR, Vrijenhoek RC (2003a)
bacteria in the burrowing echinoid, Echinocardium cordatum Environmental acquisition of thiotrophic endosymbionts by deep-sea mus-
(Echinodermata). Mar Biol 115:179–185 sels of the genus Bathymodiolus. Appl Environ Microbiol 69:6785–6792
Thao M, Moran N, Abbot P, Brennan E, Burckhardt D, Baumann P (2000) Won Y, Young CR, Lutz RA, Vrijenhoek RC (2003b) Dispersal barriers and
Cospeciation of psyllids and their primary prokaryotic endosymbionts. isolation among deep-sea mussel populations (Mytilidae:Bathymodiolus)
Appl Environ Microbiol 66:2898–2905 from eastern Pacific hydrothermal vents. Mol Ecol 12:169–184
Trask JL, Van Dover CL (1999) Site-specific and ontogenetic variations in nutri- Yamamoto H, Fujikura K, Hiraishi A, Kato K, Maki Y (2002) Phylogenetic
tion of mussels (Bathymodiolus sp.) from the Lucky Strike hydrothermal vent characterization and biomass estimation of bacterial endosymbionts associ-
field, Mid-Atlantic Ridge. Limnol Oceanogr 44:334–343 ated with invertebrates dwelling in chemosynthetic communities of hydro-
Tunnicliffe V (1988) Biogeography and evolution of hydrothermal-vent fauna in thermal vent and cold seep fields. Mar Ecol Progr Ser 245:61–67
the eastern Pacific Ocean. Proc R Soc Lond B 233:347–366 Yamanaka T, Mizota C, Satake H, Kouzuma F, Gamo T, Tsunogai U, Miwa T,
Tunnicliffe V (1991) The biology of hydrothermal vents: ecology and evolution. Fujioka K (2003) Stable isotope evidence for a putative endosymbiont-based
Oceanogr Mar Biol 29:319–407 lithotrophic Bathymodiolus sp. mussel community atop a serpentine sea-
Tunnicliffe V, Fowler C (1996) Influence of sea-floor spreading on the global mount. Geomicrobiol J 20:185–197
hydrothermal vent fauna. Nature 379:531–533 Yong R, Searcy DG (2001) Sulfide oxidation coupled to ATP synthesis in chicken
Turner RD, Lutz RA, Jablonski D (1985) Modes of molluscan larval development liver mitochondria. Comp Biochem Physiol Pt B 129:129–137
at deep-sea hydrothermal vents. Biol Soc Wash Bull 6:167–184 Young CM, Vazquez E, Metaxas A, Tyler PA (1996) Embryology of vestimentif-
Tyler PA, Young CM (2003) Dispersal at hydrothermal vents: a summary of recent eran tube worms from deep-sea methane/sulphide seeps. Nature
progress. Hydrobiologia 503:9–19 381:514–516
Vacelet J, Boury-Esnault N, Fiala-Medioni A, Fisher CR (1995) A methanotrophic Zal F, Lallier FH, Wall JS, Vinogradov SN, Toulmond A (1996) The multi-
carnivorous sponge. Nature 377:296 hemoglobin system of the hydrothermal vent tube worm Riftia pachyptila.
1: reexamination of the number and masses of its constituents. J Biol Chem binding by Riftia pachyptila hemoglobins. Proc Natl Acad Sci USA
271:8869–8874 95:8997–9002
Zal F, Green BN, Lallier FH, Toulmond A (1997) Investigation by electrospray Zal F, Leize E, Oros DR, Hourdez S, Van Dorsselaer A, Childress JJ
ionization mass spectrometry of the extracellular hemoglobin from the (2000) Haemoglobin structure and biochemical characteristics of the sul-
polychaete annelid Alvinella pompejana: an unusual hexagonal bilayer phide-binding component from the deep-sea clam Calyptogena magnifica.
hemoglobin. Biochemistry 36:11777–11786 Cahiers Biol Mar 41:413–423
Zal F, Leize E, Lallier FH, Toulmond A, Van Dorsselaer A, Childress JJ Zhang JZ, Millero FJ (1993) The products from the oxidation of H2S in seawater.
(1998) S-sulfohemoglobin and disulfide exchange: the mechanisms of sulfide Geochim Cosmochim Acta 57:1705–1718

The Prokaryotes - Prokaryotic Biology and Symbiotic Associations PDF

Uploaded by

Copyright:

Available Formats

The Prokaryotes - Prokaryotic Biology and Symbiotic Associations PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Prokaryotes - Prokaryotic Biology and Symbiotic Associations PDF

Uploaded by

Copyright:

Available Formats

Eugene Rosenberg (Editor-in-Chief)

With 183 Figures and 62 Tables

ISBN 978-3-642-30193-3 ISBN 978-3-642-30194-0 (eBook)

Library of Congress Control Number: 2012950985

3rd edition: © Springer Science+Business Media, LLC 2006

Printed on acid-free paper

Springer is part of Springer ScienceþBusiness Media (www.springer.com)

Eugene Rosenberg (Editor-in-Chief)

Section 1: Biology of Bacteria and Archaea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

3 Prokaryotes and Their Habitats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

4 Origin of Life: RNA World Versus Autocatalytic Anabolist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

5 Morphological and Physiological Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

6 Prokaryote Characterization and Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

7 Principles of Enrichment, Isolation, Cultivation, and Preservation of Prokaryotes . . . . . . . . . . . . . . . . . . . . . . 149

8 Comparative Genomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

9 Defining Taxonomic Ranks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

10 Population Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255

11 Public Service Collections and Biological Resource Centers of Microorganisms . . . . . . . . . . . . . . . . . . . . . . . . 267

12 Repositories for Patented and Safeguarded Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305

13 Biotechnology and Applied Microbiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315

14 Genomes and Post-genome Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329

Section 2: Symbiotic Associations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345

16 Cyanobacterial-Plant Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359

17 Root and Stem Nodule Bacteria of Legumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401

18 Symbiotic Associations Between Ciliates and Prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427

19 Bacteriocyte-Associated Endosymbionts of Insects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465

20 Vibrio fischeri: Squid Symbiosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497

21 Rumen Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533

22 Symbiotic Associations Between Termites and Prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545

23 Marine Chemosynthetic Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579

David G. Adams W. Ford Doolittle

Cristine Chaves Barreto Paula S. Duggan

Linda C. Baumann Martin Dworkin

Hans-Dieter Görtz Nancy A. Moran

Amar N. Rai Frank J. Stewart

Eugene Rosenberg Masayuki Sugawara

Hans G. Schlegel Alessandra B. G. Valladão

Erko Stackebrandt Irene von der Weid

Carl R. Woese Ilana Zilber-Rosenberg

Biology of Bacteria and Archaea

Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 The Organismal Structure of the Archaea . . . . . . . . . . . . . . . . . . 13

Microbiology’s Halting Development . . . . . . . . . . . . . . . . . . . . . . . . 4 The Organismal Structure of the Eubacteria . . . . . . . . . . . . . . . 14

Where to Play? The Sand Castles Have All Background

The Organismal Structure of the Archaea

lex herm oto t

Zea 0.1 changes / site

Paraphyly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 This chapter, then, hopes to explain why such questions

warm-blooded flying things

fishes whales bats therapsids turtles crocodiles pterosaurs birds lizards/snakes

designation, the ‘‘prokaryote/eukaryote dichotomy’’ imposed IA

The Concept of a Bacterium

Jan Sapp, a professional historian unusually well-read in the ORIGIN

plastids plastids plastids

Pace’s second objection, that the ‘‘pro’’ in prokaryotes is mislead-

Number of Gene Families

Residual historical signals can also be extracted in the form of Coda

Ecological Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Gradients as Habitats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Inhabitants of an Ecosystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Gradients of a Microscale: The Sediments . . . . . . . . . . . . . . . . . . 58