Working With Rhizobia

Working with rhizobia
J.G. Howieson and M.J. Dilworth (Eds.)

Working
with rhizobia
J.G. Howieson and M.J. Dilworth (Eds.)
Centre for Rhizobium Studies

Murdoch University
2016
The Australian Centre for International Agricultural Research (ACIAR) was established in
June 1982 by an Act of the Australian Parliament. ACIAR operates as part of Australias
international development cooperation program, with a mission to achieve more
productive and sustainable agricultural systems, for the benefit of developing countries
and Australia. It commissions collaborative research between Australian and developing-
country researchers in areas where Australia has special research competence. It also
administers Australias contribution to the International Agricultural Research Centres.
Where trade names are used this constitutes neither endorsement of nor discrimination
against any product by ACIAR.
ACIAR MONOGRAPH SERIES

This series contains the results of original research supported by ACIAR, or material
deemed relevant to ACIARs research and development objectives. The series is
distributed internationally, with an emphasis on developing countries.
Australian Centre for International Agricultural Research (ACIAR) 2016

This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no
part may be reproduced by any process without prior written permission from ACIAR, GPO
Box 1571, Canberra ACT 2601, Australia, [email protected].
Howieson J.G. and Dilworth M.J. (Eds.). 2016. Working with rhizobia. Australian Centre for
International Agricultural Research: Canberra.
ACIAR Monograph No. 173
ACIAR Monographs ISSN 1031-8194 (print), ISSN 1447-090X (online)
ISBN 978 1 925436 17 4 (print)
ISBN 978 1 925436 18 1 (PDF)
Editing by Kate Langford
Text design by Peter Nolan, Canberra
Cover design by CRS and Joke Hollants
Printing by Bytes n Colours, Canberra

Foreword
The legume symbiosis with rhizobia (root nodule bacteria) has been acknowl-
edged as fundamental to sustainable agriculture because this intimate relationship
between soil bacteria and flowering plants can alleviate the need to provide manu-
factured nitrogen for farming systems. Prior to the invention of the HaberBosch
process, which led to the manufacture of fertiliser nitrogen early last century,
rhizobial nitrogen fixation was the dominant source of nitrogen in agriculture.
We now understand that much of the change to our climate has resulted from the
burning of fossil fuels, which is essential for generating the high temperature and
pressure for the HaberBosch process. Any anthropomorphic activity that can
limit consumption of fossil fuels must therefore be embraced. This has brought a
renewed focus to the science of biological nitrogen fixation because the reduct-
ant and metabolic energy needed for the key enzyme nitrogenase to function in
rhizobia are instead derived from solar radiation.
The Australian Centre for International Agricultural Research (ACIAR) has a

fundamental commitment to the development of sustainable agricultural prac-
tices. Further, many of the ACIAR aid projects are directed at landscapes that are
infertile. Nitrogen infertility is a global production constraint for farmers who
cannot afford manufactured fertiliser, and thus adoption of legumes inoculated
with appropriate rhizobia in their farming systems can lead to greatly increased
food production.
This manual provides scientists and technicians with modern guidelines for en-
suring that the legume symbiosis with rhizobia is optimised for nitrogen fixation
in their environments. It builds upon similar manuals produced over the last hun-
dred years and updates our knowledge of this fundamental biological process and
our ability to use it in agriculture.
Nick Austin
Chief Executive Officer, ACIAR
Contents
Foreword.............................................................................................................3
Preface.................................................................................................................9
About this manual.....................................................................................................9
About the authors and compilation of the manual...............................................9
Correct citation for the manual...............................................................................9
Acknowledgements.................................................................................................10
Authors.....................................................................................................................10
1 The legume-rhizobia symbiosis and assessing the need to inoculate ........15

1.1 Legumes in agriculture, society and the environment............................15
1.2 Rhizobia and nodules..................................................................................16
1.3 The legumes and their nodule characteristics..........................................17
1.4 Assessing the need to inoculate..................................................................20
1.5 Selecting inoculant quality strains.............................................................24
1.6 References.....................................................................................................24
2 Collecting nodules for isolation of rhizobia...............................................25

2.1 Introduction..................................................................................................25
2.2 Collecting new strains of rhizobia.............................................................26
2.3 Respecting the international biodiversity convention.............................27
2.4 Collecting nodules.......................................................................................27
2.5 Many nodules from one plant or a few nodules from many plants?.....31
2.6 Storage of nodules........................................................................................31
2.7 Trapping rhizobia from soil or ruptured nodules....................................34
2.8 Specific host in situ trapping......................................................................34
2.9 Collection of passport data.........................................................................35
2.10 Collection of seed (for authentication).....................................................36
2.11 References.....................................................................................................37
3 Isolation and growth of rhizobia.................................................................39

3.1 Introduction..................................................................................................39
3.2 Preparing solid growth media....................................................................39
3.3 Preparing the nodules..................................................................................40
3.4 Isolating bacteria from nodules..................................................................40
3.5 Recognising nodule bacteria growing on solid media............................42
3.6 Some useful diagnostic features of species of nodule bacteria...............46
3.7 Routine culture media for rhizobia............................................................47

3.8 Antibiotics and indicators...........................................................................54
3.9 Avoiding contamination..............................................................................57
3.10 Authentication of isolates and Kochs postulates.....................................58
3.11 References.....................................................................................................60
4 Preservation of rhizobia..............................................................................61
4.1 The need to preserve cultures.....................................................................61
4.2 Options for preservation.............................................................................61
4.3 Longer-term preservation...........................................................................63
4.4 Quality control..............................................................................................68
4.5 Cryopreservation..........................................................................................69
4.6 Developing a parent/working-lot system..................................................70
4.7 References.....................................................................................................71
5 Authentication of rhizobia and assessment of the legume symbiosis

in controlled plant growth systems.............................................................73
5.1 Introduction..................................................................................................73
5.2 Host-range, cross nodulation and effectiveness.......................................74
5.3 Plant growth systems with which to assess nodulation and
effectiveness..................................................................................................74
5.4 Screening for N2 fixation.............................................................................78
5.5 Substrates for effectiveness experiments...................................................84
5.6 Assessment of N2-fixing potential in non-sterile soil cores....................87
5.7 Nutrient media for legume cultivation in glasshouse experiments.......89
5.8 General procedures for handling seed in preparation for
authentication or for an effectiveness experiment...................................92
5.9 General methods for inoculation...............................................................95
5.10 Growing conditions and facilities..............................................................96
5.11 Data acquisition for quantification of effectiveness.................................97
5.12 Outcomes from authentication and effectiveness experiments...........106
5.13 References...................................................................................................107
6 Counting rhizobia......................................................................................109
6.1 Introduction................................................................................................109
6.2 Serial dilution.............................................................................................109
6.3 Plate counts of rhizobia in sterile diluent................................................111
6.4 Evaluating rhizobial survival on seed......................................................114
6.5 Indirect counts by plant infection to estimate Most Probable
Number (MPN)..........................................................................................115
6.6 Estimate of cell number by optical density.............................................119
6.7 Direct counts under the microscope.......................................................120
6.8 References...................................................................................................124
7 Taxonomy and physiology of rhizobia......................................................125

7.1 Taxonomy of rhizobia................................................................................125
7.2 Physiology of root nodule bacteria..........................................................129
7.3 Studying membrane transport systems...................................................138
7.4 Final comments..........................................................................................142
7.5 References...................................................................................................142
8 Field experiments with rhizobia................................................................145

8.1 Introduction................................................................................................145
8.2 Defining the aim of the field evaluation..................................................145
8.3 Experimental design..................................................................................146
8.4 Site selection...............................................................................................147
8.5 Selecting the appropriate carrier and adhesive for application of
rhizobia to the seed....................................................................................150
8.6 Adhesives for application of inoculants to legume seeds......................151
8.7 General procedures for application of inoculants..................................151
8.8 Laying out plot dimensions for field experiments.................................153
8.9 Plots established with precision seeding machinery.............................160
8.10 Data collection and assessment of nodulation.......................................161
8.11 Long-term experiments.............................................................................164
8.12 References...................................................................................................165
9 Inoculant production and quality control................................................167

9.1 Introduction................................................................................................167
9.2 Inoculant production.................................................................................168
9.3 Carrier materials for inoculants...............................................................171
9.4 Peat inoculants............................................................................................175
9.5 Granular inoculants...................................................................................176
9.6 Storage of inoculants..................................................................................178
9.7 Inoculant quality control...........................................................................178
9.8 Determination of shelf life........................................................................182
9.9 Criteria for the selection of strains for use in commercial
manufacture of legume inoculants..........................................................184
9.10 References...................................................................................................186
10 Measurement of nitrogen fixation.............................................................187

10.1 Introduction................................................................................................187
10.2 Methods currently available......................................................................188
10.3 Which method to use?...............................................................................188
10.4 Sampling for biomass and biomass N......................................................189
10.5 Preparing and analysing plant samples for %N and 15N.......................191
10.6 Accounting for below-ground N..............................................................193
10.7 Nitrogen difference method......................................................................194

10.8 15
N dilution (and incorporation) methods.............................................195
10.9 Ureide (N solute) method.........................................................................202
10.10 Nitrogen balance method..........................................................................216
10.11 Acetylene reduction method (assaying nitrogenase activity)..............217
10.12 Measuring N2 fixation at the farm and beyond......................................217
10.13 Conclusions.................................................................................................218
10.14 References...................................................................................................218
11 Fundamental molecular techniques for rhizobia.....................................221

11.1 Introduction to nucleic acid purification................................................221
11.2 Plasmid purification...................................................................................228
11.3 PCR techniques..........................................................................................236
11.4 References...................................................................................................242
12 Specialised genetic techniques for rhizobia..............................................245

12.1 Introduction................................................................................................245
12.2 Transposon mutagenesis...........................................................................245
12.3 Protocol for transduction .........................................................................250
12.4 Protocol for conjugation...........................................................................253
12.5 Reporter assays ..........................................................................................256
12.6 Protocol for nucleic acid sequencing.......................................................267
12.9 References...................................................................................................280
13 Methods for isolation of RNA from rhizobia............................................283

13.1 Introduction................................................................................................283
13.2 Isolation of total and small RNA (sRNA) from free-living rhizobia...284
13.3 Protocol for isolation of total and small RNA (sRNA) from the
rhizosphere of legumes .............................................................................288
13.4 Protocol for isolation of total and small RNA (sRNA) from
bacteroids of rhizobia................................................................................292
13.5 Protocol for preparation of DNA-free RNA for use in Q-RT-
PCR by treatment with Ambion Turbo DNA free.................................293
13.6 Protocol for SenseAmplinear RNA amplification..............................294
13.7 Protocol for indirect labelling of firststrand cDNA synthesis
using GE healthcare Cyscribe Post-Labelling Kit..................................298
13.8 Protocol for Rubicon WTA amplification of RNA to produce
double-stranded cDNA (SIGMA CAT NO WTA2)..............................302
13.9 Protocol for direct labelling of double-stranded cDNA made by
Rubicon WTA kit.......................................................................................304
13.10 Appendices..................................................................................................305
Index................................................................................................................307
8
About this manual
Preface
About this manual
Study of the legume/rhizobium symbiosis necessitates an understanding of meth-

ods to isolate and characterise the bacteria. Since the publication of A manual
for the practical study of root-nodule bacteria by Jim Vincent (1970) a number of
sequels have been published, such as the NifTAL, CIAT and CIMMYT manuals,
which are now out of date and out of print. Discoveries of a much wider range
of root-nodulating bacteria than previously known means that even simple iso-
lation methods need revisiting to ensure unusual types of bacteria are not dis-
carded. Drawing on the rich experience from earlier publications, this manual
brings together state-of-the-art methods for the study of root-nodule bacteria,
both in the free-living state and in symbiosis with legumes. In each chapter, we
introduce the topic and provide guidance on how study of the symbiosis might
best be tackled. We then provide a detailed description of protocols that need to
be followed and highlight potential problems and pitfalls. Topics covered include
acquiring, recognising, growing and storing rhizobia, experimenting with strains
in the laboratory, glasshouse and field, and applying contemporary molecular and
genetic methodologies to assist in the study of rhizobia. We include a chapter that
describes the current taxonomy and physiological understanding of rhizobia, and
another on the production of inoculants and quality control in the supply chain.
About the authors and compilation of the manual
The lead authors for each chapter were selected on the basis of their current exper-
tise in working with rhizobia. They were invited to an initial meeting to conceive
the contents of the manual at Rottnest Island, Western Australia in March 2011.
Each authorship group provided early drafts which were then circulated to select-
ed co-authors for revision (listed next page) before being finalised by the editors.
Correct citation for the manual
Howieson J.G. and Dilworth M.J. (Eds.). 2016. Working with rhizobia. Australian
Centre for International Agricultural Research: Canberra.
9
Acknowledgements
Acknowledgements
We thank ACIAR (through the ECCAL project http://aciar.gov.au/project/

lps/2004/022 and SIMLESA), the Crawford Fund and the Bill & Melinda Gates
Foundation (through the N2Africa project http://www.n2africa.org/), Murdoch
University, Australian Wool Innovation and the Grains Research and Development
Corporation (through the National Rhizobium Program) for research support.
Authors
The authors who contributed to this manual, in alphabetical order, and their affili-
ated institutions are as follows:
Ricardo Silva Araujo

Total Biotecnologia Indstria e Comrcio Ltda
Rua Emlio Romani 1190, CEP 81460-020 Curitiba, Paran, Brazil
[email protected]
Julie Ardley
Centre for Rhizobium Studies (CRS) Murdoch University
90 South St. Murdoch, WA, Australia
[email protected]
Robert Abaidoo
International Institute of Tropical Agriculture (IITA) CSIR-Crops Research
Institute
Box 3785, Kumasi, Ghana
[email protected]
Abdullahi Bala
Federal University of Technology
Minna PMB69 Minna, Nigeria
[email protected]
Roz Deaker
Faculty of Agriculture and Environment, University of Sydney
Level 4 Biomedical Building 1 Central Ave, Australian Technology Park,
Eveleigh, NSW 2015, Australia
[email protected]
10
Authors
Michael Dilworth
Centre for Rhizobium Studies (CRS), Murdoch University
[email protected]
Greg Gemell
Australian Inoculants Research Group
NSW Department of Primary Industries Locked Bag 26, Gosford, NSW 2250,
Australia
[email protected]
Ken Giller
Plant Production Systems Group Wageningen University, 6700 AK, Box 430,
Wageningen, The Netherlands
[email protected]
Elizabeth Hartley
Australian Inoculants Research Group
NSW Department of Primary Industries
Locked Bag 26, Gosford, NSW 2250, Australia
[email protected].
David Herridge
University of New England Primary Industries Innovation Centre
20A Dodds Street, Redhead NSW 2290 Australia
[email protected]
John Howieson
[email protected]
Mariangela Hungria
EMBRAPA-CNPSo, Cx.
Londrina, PR, Brazil Cx. Postal 231, 86001-970
[email protected]
11
Authors
Nancy Karanja
Department of Land Resource Management and Agricultural Technology,
College of Agriculture and Veterinary Sciences, University of Nairobi
PO Box 29053-00625, Nairobi, Kenya
[email protected]
Karunakaran Ramakrishnan
Cell and Developmental Biology
John Innes Centre, Norwich, UK
[email protected]
Vanessa Melino
Australian Centre for Plant Functional Genomics (ACPFG), University of
Adelaide
PMB 1, Glen Osmond, SA, 5064, Australia
[email protected]
Sofie De Meyer
[email protected]
Graham OHara
[email protected]
Philip Poole
Department of Plant Sciences, University of Oxford
South Parks Road, Oxford, OX1 3RB, UK
[email protected]
Wayne Reeve
[email protected]
12
Authors
Janet Sprent
Division of Plant Sciences at JHI, University of Dundee
Dundee DD2 5DA Scotland, UK
[email protected]
Jason Terpolilli
[email protected]
Ravi Tiwari
[email protected]
Paul Woomer
Forum for Organic Resource Management and Agricultural Technology,
Nairobi, Kenya (FORMAT)
[email protected]
Ron Yates
Department of Agriculture and Food
Baron Hay Crt., South Perth 6015, WA, Australia
[email protected]
Jerry Zilli
EMBRAPA Roraima
Rodova BR 174 km 08, Brazil
[email protected]
13
1.1 Legumes in agriculture, society and the environment
CHAPTER1
The legume-rhizobia
symbiosis and assessing
the need to inoculate
K.E. Giller, D.F. Herridge and J.I. Sprent
1.1Legumes in agriculture, society and the

environment
Legumes are a major component of all agrarian systems throughout the world.
They are particularly attractive to low input systems of agriculture because they
take inert nitrogen from the air and (through rhizobia) transform it into proteins
in a process that leaves no carbon footprint. But beware, legumes differ in their
adaptation to infertile soils, and the right legume must be chosen for each envi-
ronment. Legumes are many and varied. The grain legumes provide protein-rich
food, and soybean and groundnut are also important oilseed crops (Table1.1).
Pasture or fodder legumes are important for livestock feed in various forms: in
grazed systems, as feed concentrates made from their grains, or in cut-and-car-
ry systems where animals are kept in stalls. Woody or tree legumes produce a
number of useful products apart from poles and construction materials; they are
important sources of feed and browse for livestock and several of them produce
edible fruits. The other major uses of legumes are for soil fertility improvement,
through cover crop protection of the soil from erosion, and as green manures con-
tributing nitrogen to improve soil fertility.
15
1.2 Rhizobia and nodules
Table1.1 The major uses of legumes, together with some examples

Use Products Examples
Food Grain Chickpea (Cicer arietinum), common bean (Phaseolus vulgaris), cowpea
(Vigna unguiculata), faba bean (Vicia faba), groundnut (Arachis hypogaea),
pea (Pisum sativum), pigeonpea (Cajanus cajan), soybean (Glycine max),
white lupin (Lupinus albus), lablab (Dolichos lablab).
Oil Groundnut (Arachis hypogaea), soybean (Glycine max)
Fodder Grazed pastures White clover (Trifolium repens), sub clover (T. subterraneum), serradella,
(Ornithopus spp.), biserrula (Biserrula pelecinus), medic (Medicago spp.),
Townsville stylo (Stylosanthes humilis), desmodium (Desmodium intortum),
Wynn cassia (Chamaecrista rotundifolia), crown vetch (Onobrychis vicifolia).
Fodder crops Alfalfa (Medicago sativa), birdsfoot trefoil (Lotus corniculatus), red
clover (Trifolium pratense), lebeckia (Lebeckia ambigua), pink serradella
(Ornithopus sativus)
Feed concentrates Soybean (Glycine max) cake, narrow-leaf lupin (Lupinus angustifolius) seed,
Velvet bean (Mucuna pruriens var. utilis) seed
Fruits Edible pods (Inga edulis)
Society Fuel, dyes and Multi-purpose trees (e.g. Calliandra calothyrsus, Leucaena spp.), Acacia
poles mangium, Acacia karroo, Indigofera spp.
Shade and gum Acacia senegal, Erythrina spp.
Environment Cover crops Mixtures of Calopogonium caeruleum, Centrosema pubescens and Pueraria
and soil phaseoloides used in plantations of rubber (Hevea brasiliensis) and oil palm
fertility (Elaeis guineensis), crimson clover (Trifolium incarnatum) in vineyards
Green manures Velvet bean (Mucuna pruriens var. utilis), sunnhemp (Crotalaria
juncea), Lablab (Dolichos purpureus), common vetch (Vicia sativa, Vicia
benghalensis)
1.2 Rhizobia and nodules
The bacteria associated with legumes (and which produce the enzymatic mech-
anisms that reduce atmospheric di-nitrogen (N2) to ammonia) are collectively
termed rhizobia or root-nodulating bacteria. These bacteria are the subject of
this manual. At the time of writing, rhizobia are found in seven bacterial fami-
lies, divided into 15 genera (Table1.2). A recent major advance, since publica-
tion of manuals that preceded this one, is the discovery of nitrogen fixation in
the -Proteobacteria. The regularity of these organisms in nodules of subtropi-
cal Mimosa (although not an indigenous species in Mexico) and in many herba-
ceous legumes of the South African fynbos is notable. These organisms grow very
quickly and have almost certainly previously been ignored, possibly considered as
contaminants by many rhizobiologists. The -Proteobacteria are treated compre-
hensively in this text, particularly in Chapters 3, 6 and 7. All rhizobia are common
Gram-negative soil-inhabiting bacteria containing genes required for nodulation
(e.g. nod, rhi) and N2 fixation (e.g. nif, fix) as described in Chapter7.
16
1.3 The legumes and their nodule characteristics
Table1.2 T
he currently described rhizobia and the number of
species in each genus (see also Chapter7)
Family Genus Number of
described
species
-Proteobacteria
Bradyrhizobiaceae Bradyrhizobium 15
Brucellaceae Ochrobactrum 2
Hyphomicrobiaceae Azorhizobium 3
Devosia 1
Methylobacteriaceae Methylobacterium 1
Microvirga 3
Phyllobacteriaceae Phyllobacterium 1
Aminobacter 1
Mesorhizobium 29
Rhizobiaceae Rhizobium 43
Neorhizobium 3
Sinorhizobium/Ensifer 13
Shinella 1
-Proteobacteria
Burkholderiaceae Burkholderia 6
Cupriavidus 2
1.2.1The nodule
The structure in which the reduction of N2 gas to ammonia takes place is called
the nodule. Nodules are found mostly on legume roots but occasionally on stems.
Nodules vary in shape and size (Sprent 2009) but all have structures and modifica-
tions from normal root cells that protect the oxygen-sensitive rhizobial enzymes
from inactivation. They accomplish this while both delivering an energy supply to
the bacteria and removing N-rich products. The main external features of nodule
morphology are illustrated in Chapter2 (Figure2.2).
Legumes are the third largest family of dicotyledonous plants and have tradition-
ally been divided into three subfamilies associated with distinct flower types Cae-
salpinioideae, Mimosoideae and Papilionoideae (Figure1.1). Few of the Caesal-
pinioideae are able to nodulate and most of the Mimosoideae and an even larger
proportion of Papilionoideae can nodulate; the latter includes most of the familiar
crop and forage plants, such as peas, beans and clovers. Although legume taxono-
my is currently undergoing major revision, at present it is convenient to retain the
traditional subdivisions until the various legume working groups have an agreed
alternative. A general reference that covers the important points noted below is
Sprent et al. (2013). At the time of writing, the total number of genera in the
17
family was 732, with 19,321 species, but by the time this work is published both
these numbers will have increased. In addition to the qualities noted below, many
legumes are important for what may be summarised as ethnobotanical purposes,
such as medicines, poisons and fibres (van Wyk et al. 1997).
1.3.1 Subfamily Caesalpinioideae

Nine genera within this subfamily are currently known to nodulate. One, Chamae-
crista, is the eighth largest genus of legumes, with 330 species. It is one of the few
that has extended into temperate regions and is the only one of agricultural sig-
nificance, as exemplified by Wynn Cassia a forage legume of warm temperate and
subtropical regions. Chamaecrista may represent a separate evolutionary event
for nodulation. All nodules are branched, often woody, and with the exception of
some species of Chamaecrista, house their bacteroids (the N2 fixing form of rhizo-
bia) in fixation threads. A number of nodulating caesalpinioid trees are of great
economic importance for their pigmented wood, for example Erythrophleum for-
dii in South-West China and Vietnam.
1.3.2 Subfamily Mimosoideae

In generic terms, this is the smallest of the subfamilies but it houses the second
(Acacia), fifth (Mimosa) and ninth (Inga) largest legume genera, all of which form
root nodules. A few basal members of the subfamily cannot nodulate. All spe-
cies studied have indeterminate nodules, often branched, with root hair infec-
tion and with the central region containing both infected and uninfected cells.
Most mimosoids, which are predominantly shrubs and trees, are important in
their ecosystems, which range from tropical rainforests to arid areas. Three tribes
have been generally recognised, Acacieae, Ingeae and Mimoseae, but recent work
on acacias has changed the composition of these. The former genus Acacia has
now been divided into six genera (Sprent 2009) with the type genus Acacia be-
ing largely Australian and placed in the tribe Ingeae. Most of the African and
South American species are now in Senegalia and Vachellia genera. These genera
include species important for products such as animal fodder, gum arabic, honey
and many important timbers. Not all of the new genera have been fully accepted
and new specific combinations are still being made. African workers commonly
retain the generic name Acacia. As a result of these changes, the tribe Acacieae
will probably cease to exist.
1.3.3 Subfamily Papilionoideae

This is the largest of the three subfamilies and the most widely studied for its
N2 fixation processes. It contains about 25 tribes, although this number is likely
to change. Fortunately these legumes can generally be grouped in ways that are
consistent with nodule characters. The principal tribes in the genistoid group are
Genisteae and Crotalarieae. Their nodules are characterised by having an epider-
mal infection process and a central region that is uniformly infected. Nodules are
18
indeterminate, often branched and occasionally encircling the subtending root

(lupinoid) as in Lupinus and Listia. Some lupin species are important grain leg-
umes. Species of Genista, Ulex and others are major components of Mediterrane-
an ecosystems as well as being invasive species in countries such as New Zealand.
The dalbergioid group is characterised by having nodules associated with later-
al (occasionally adventitious) roots and rhizobial infection is via cracks where
these roots emerge. Nodules have determinate growth and the central region is
uniformly infected. Included in this group are important grain legumes such
as Arachis hypogaea (groundnut or peanut), forage plants such as Stylosanthes,
highly-prized coloured timbers such as species of Dalbergia, whose wood may be
more valuable that ivory or rhinoceros horn. Many Dalbergia species are on the
CITES endangered list. The Phaseoleae tribe and close relatives, Desmodieae and
Psoraleae (these three tribes will likely be merged) are characterised by nodules
having determinate growth with surface lenticels, root hair infection and central
tissue containing both infected and uninfected cells. Because of the great impor-
tance of soybeans and Phaseolus beans, this group of legumes has been the subject
of extensive research. Unlike most nodules which export the amides glutamine
and asparagine as products of fixation, this group exports ureides which can be
used as an assay for N2 fixation (see Chapter10). The Indigofereae tribe houses the
third largest legume genus Indigofera whose nodules are like an extended soybean
nodule but have been little studied. Some species are important for production
of dyes such as indigo, some are toxic to animals and many are used for fodder,
green manure and cover crops. With the exception of Lotus that has determinate
nodules but does not export ureides, and Sesbania which is in its own tribe and
has many unique features, all other legumes studied, both temperate and tropical,
have indeterminate nodules, root hair infection and central nodule tissue contain-
ing both infected and uninfected cells. Included in this group are important grains
such as peas, lentils and chickpeas, and forages such as clovers and medics. The
largest of all legume genera, Astragalus, is in this group and now gives its name to
the tribe Astragaleae, formerly Galegeae. This name change has occurred because
the type genus Galega is now known not to belong here.
Figure1.1 Flowers formed by Subfamilies of the Leguminosae; a)

Caesalpinioideae, b) Mimosoideae, c) Papilionoideae
19
1.4 Assessing the need to inoculate
1.3.4 Actinorhizal plants

A completely different group of plants, known as actinorhizal plants, is nodulated
not by rhizobia but by the filamentous bacterial genus Frankia. These plants occur
in a number of different dicotyledonous families. Examples are the genera Alnus,
Casuarina and Ceonothus. Nodules are root-like in structure with a central rather
than a peripheral vascular system, as in legumes, and like legumes they vary in
other features such as mode of infection. A detailed comparison of legume and
actinorhizal nodules can be found in Pawlowski and Sprent (2008). So far, it has
not been possible to culture Frankia from all actinorhizal species.
The establishment of a N2-fixing legume/rhizobia symbiosis needs the presence

of the legume in association with a compatible rhizobial strain. However, there
is very substantial specificity between legumes and rhizobia. When an exotic leg-
ume is introduced into an area in which it has not been previously grown, it is
likely that its growth and yield can be improved by inoculating with rhizobia.
Other cases where inoculation may be necessary are where the soil contains a
large population of rhizobia that are able to nodulate but are ineffective (i.e. inca-
pable of N2 fixation) with the legume of interest. So how can we identify whether
or not inoculation is needed?
The only sure way to know if inoculation is necessary is to conduct need-to-inoc-

ulate trials. These are simple trials with a minimum of three treatments:
1. An uninoculated control without N fertiliser (I, N)
2. A treatment inoculated with the best-quality inoculant, without N fertiliser

(+I, N)
3. An uninoculated control treatment with N fertiliser added (I, +N)
The uninoculated treatment without N fertiliser added will reveal the status of the
background rhizobia, if present. The uninoculated treatment with N is needed
to show whether legume growth is N-limited or indeed if other nutrients (such
as P or K) are limiting growth (see Figure1.3). The inoculated treatment shows
whether the N-limitation can be overcome by establishment of an effective sym-
biosis through inoculation with rhizobia. Ideally all three treatments should be
compared at background soil fertility (i.e. that currently used by the farmer) and
with optimal fertility (all limiting nutrients added). The treatments with optimal
fertility are needed to ensure that the full N2-fixation potential of the legume/
rhizobia symbiosis can be expressed, and is not limited due to deficiencies of other
nutrients. Possible observations of the different treatment plots are given in Ta-
ble1.3 together with an explanation of why such effects may occur.
20
Table1.3 P
ossible field observations made in different plots of need-to-inoculate trials and
explanations of the underlying causes, with suggestions for where further research may be
required*.
Plant growth On the roots Explanation Further research
required
Uninoculated Poor growth, No nodules No native rhizobia
control plants yellow capable of infecting the
test legume
Poor growth, Many small Native rhizobia ineffective
plants yellow ineffective in N2 fixation with the test
nodules legume
Good growth, No nodules Soils rich in mineral N, no
plants dark green native rhizobia capable
of nodulating the test
legume
Good growth, Many small Soils rich in mineral N,
plants dark green ineffective native rhizobia may be
nodules effective or ineffective
Good growth, Many large Large soil population
plants dark green nodules, red of effective compatible
inside rhizobia
Inoculated Poor growth, No nodules Inoculum does not Select better rhizobial
plants plants yellow contain a compatible strain, ensure careful
strain or inoculum dead, production, QA,
or strain not adapted to transport and handling
edaphic conditions of inoculants, improved
carrier
Poor growth or Few nodules Poor quality inoculant Select better rhizobial
growth not as or many small not delivering enough strain, ensure careful
vigorous as N-fed ineffective cells, or rhizobial strain production, QA,
plants, plants nodules not highly effective, or transport and handling
light green or not competitive, or strain of inoculants, improved
yellow not adapted to edaphic carrier
conditions
Poor growth, Many nodules, Other nutritional factors Conduct investigations
plants pale red inside limiting growth to identify other limiting
nutrients
Good growth, Many large Highly effective inoculant
plants dark green nodules, red
inside
N fed plants Poor growth, No nodules or Other factor limiting plant Conduct investigations
plants yellow few nodules growth to identify other limiting
nutrients or biotic factors
Good growth, No nodules or Availability of mineral N
plants dark green few nodules inhibits nodule formation
*Modified from Date (1982); further detail may be found in Sessitsch et al. (2002)
21
100 (I, N)
(+I, N)
80
(I, +N)
Plant growth (%)

60
40
20
0
Case 1 Case 2 Case 3 Case 4
Figure1.2 Possible results obtained from need-to-inoculate trials
A combination of treatments is needed to understand whether N is the principal

factor limiting growth and yield of the legume and whether inoculation is neces-
sary (Figure1.2). In all cases, the roots must be carefully examined for nodulation.
In Case 1, there is no response to inoculation or N fertiliser, suggesting that some

other factor is limiting plant growth. This could be limitations of nutrients such
as P, K or other nutrients, and a set of treatments with all nutrients added would
allow understanding of whether this is the case (Figure1.3). If water is limiting
growth, this is likely to be obvious but biotic stresses may be more difficult to di-
agnose.
In Case 2, legume growth is clearly N-limited but inoculation leads to only par-
tial success in N2 fixation. This suggests that a better inoculant strain is required
or that inoculation was only partially successful due, for example, to problems
with the production and application of the inoculant or competition from resi-
dent background rhizobia that fix nitrogen poorly. This is illustrated in Figure1.4.
Case 3 displays successful inoculation with rhizobia, where the inoculant has de-
livered a high-quality strain and it has achieved nodulation in the absence or pres-
ence of a background ineffective strain.
Case 4 illustrates a situation where either the soil can supply sufficient mineral N
for optimal growth or where the indigenous rhizobial population is both large and
effective so that optimal N2 fixation occurs in the unfertilised control.
22
Figure1.3 Inoculated and nodulated

lablab growing on an apatitic soil
without added K in the foreground, and
with 18kg/ha K2SO4 in the background,
illustrating how lack of macro-nutrients
can limit plant growth in the presence of
an effective symbiosis (Case 1)
Legume fix
MAR1495
uninoculated
Legume fix
MAR1495
uninoculated
Figure1.4 The results of a need to inoculate trial on a farm in Ethiopia, where

the inoculant MAR1495 was suboptimal (Case 2) compared with Legumefix, a
commercial inoculant using strain 532c (Case 3), either because of containing an
inferior strain, a poorer carrier or lower numbers of rhizobia. Missing here is a +N
control.
23
1.5 Selecting inoculant quality strains
1.5 Selecting inoculant quality strains
Where this simple trial indicates that inoculation is necessary, yet a suitable in-
oculant is unavailable or the carrier is inadequate, rhizobial strains must be ac-
quired, isolated and purified (Chapters 24), evaluated (Chapters 58) then man-
ufactured in a suitable carrier (Chapter 9). Frequent scenarios that necessitate
research to provide well-adapted strains or suitable carriers include the following:
Acid soils, which are common in subtropical regions of the world, decrease
the survival of many inoculants, either in the year of application or in sub-
sequent seasons. Where a regenerating pasture legume is reliant on this in-
oculant to persist in the soil, research must be undertaken to select adapted
strains. Techniques applicable to this are discussed in Chapter8.
Hot and dry environments where farmers cannot refrigerate peat cultures
of rhizobia. Research into carriers that are stable at ambient temperature, or
strains that do not require refrigeration, is required. Chapter9 discusses some
approaches to this issue.
Ineffective background populations which preclude the desired inoculant

from forming the majority of the nodules. This is a common occurrence where
a legume has been cultivated for many years, and the original inoculant strains
deteriorate in effectiveness, or where new cultivars are developed that have a
more specific rhizobial requirement than the original inoculant.
This manual covers methodologies which will allow those working with rhizo-
bia to achieve strain selection and manufacture to deliver high-quality inocula,
to overcome the above constraints to nitrogen fixation from legumes, and with
which to assess the success of the inoculation procedure.
1.6References
Date R.A. 1982. Collection, isolation, characterization and conservation of Rhizobium. Pp 95109
in Nitrogen fixation in legumes, ed. by J.M. Vincent. Academic Press: New York.
Pawlowski K. and Sprent J.I. 2008. Comparison between actinorhizal and legume symbiosis. Pp
261288 in Nitrogen fixation: origins, applications, and research progress, ed. by K. Pawlowski
and W. Newton. Springer: The Netherlands.
Sessitsch A., Howieson J.G., Perret X., Antoun H. and Martnez-Romero E. 2002 Advances in Rhizo-
bium research. Critical Reviews in Plant Science 21, 32378.
Sprent J.I. 2009. Legume nodulation: a global perspective. Wiley-Blackwell, Oxford: UK.
Sprent J.I., Ardley J.K. and James E.K. 2013. From north to south: A latitudinal look at legume nodu-
lation processes. South African Journal of Botany 89, 3141.
van Wyk B.E., van Outsdhoorn B. and Gericke N. 1997. Medicinal plants of South Africa. CABI.
24
2.1Introduction
CHAPTER2
Collecting nodules for

isolation of rhizobia
J.G. Howieson, R.J. Yates, A. Bala and M. Hungria
2.1Introduction
For the legumes widely used in commerce, rhizobial strains that are well matched
to these for nitrogen fixation (i.e. highly effective at N2 fixation) are usually avail-
able from manufacturers or from gene banks. Some of these sources are listed
in Table 2.1. However, new acquisitions of rhizobia may be required for many
reasons, for example: to overcome poor N2 fixation from an existing symbiotic
relationship as described in Case 2, Chapter1; to select well-adapted strains for a
difficult environment (e.g. acid soil); to assist in legume domestication programs;
to match with sequenced legumes for genetic studies of N2 fixation (e.g. Terpolilli
et al. 2008); or to undertake biodiversity studies.
If researchers feel that currently available strains may not satisfy their research
program, then a broader range of rhizobium germplasm must be sought. Strains
of rhizobia for many legumes have long been collected from their natural envi-
ronments for this purpose. These environments represent in situ repositories of
rhizobium genetic resources (Date 1982). After the collection, isolation and eval-
uation of new strains they should be deposited in curated gene banks. The bulk of
this chapter covers methods to acquire nodule bacteria from in situ sources.
25
2.2 Collecting new strains of rhizobia
Table2.1 Some curated repositories of nodule bacteria

Rhizobia Gene banks Working collections of rhizobia
Ghent University, Belgium (LM) WSM, CB, CC, WU, Centre for Rhizobium Studies
(CRS), Murdoch University, Australia
USDA ARS, USA IITA, Ibadan, Nigeria
Murdoch University, Australia (WSM) SARDI, University of Adelaide, Australia
EMBRAPA, Brazil (SEMIA) ICARDA, Aleppo, Syria
NIAS, Japan SU, Sydney University, Australia
ICRISAT, India NAK, Nairobi University, Kenya
2.2 Collecting new strains of rhizobia
Rhizobia are best sourced from nodules collected directly from the target leg-
ume growing in its natural environment (in situ). However, if this is not possi-
ble, strains may be recovered indirectly from soil collected near the target legume
(preferably in its rhizosphere) using a trap host grown under controlled condi-
tions in the glasshouse (see Chapter5).
2.2.1The in situ repository

If nodules are to be collected in situ (with a view to solving an agricultural chal-
lenge) the target environment for the eventual application of the legume should
influence the locality from which the rhizobial germplasm is sourced. This is be-
cause rhizobial success in nodulation can be greatly influenced by soil proper-
ties such as clay content, pH, cation exchange saturation and climate (Graham
1998). Hence, the collection site should reflect the target site as far as possible in
its edaphic properties. Prior information on the edaphic properties of the poten-
tial collection area may be derived from a geological atlas that identifies parent
rock materials (Figure2.1, left panel) or published soil surveys and rainfall charts.
These may be overlayed with floral keys and knowledge of legume distribution to
help guide and focus the collection expedition.
Often the richest in situ repositories are in rocky regions where cropping cannot
be practiced (Figure2.1, right panel) or in national parks where land disturbance
has been minimised. In both scenarios, permits for exploration must be obtained
in advance.
26
2.3 Respecting the international biodiversity convention
Figure2.1 A soil map (left) illustrates granitic regions within the Cyclades group of Greek islands
that give rise to acidic and infertile soils (in red colour, e.g. Ikaria top right). Surveying the granitic
region on Ikaria (right) for legumes growing on soils in a region which receives less than 400mm
annual rainfall. This approach has produced high-quality inoculants for similar environments in
southern Australia (e.g. Howieson et al. 2000; Loi et al. 2012).
2.3Respecting the international biodiversity

convention
The 1992 United Nations Convention on Biological Diversity (CBD- http://www.

cbd.int/) sets out a series of articles that assign principles of conservation of ge-
netic resources and rights of ownership. CBD assigns sovereignty over natural
resources to States, and also suggests scientific experiments be undertaken within
the country of origin of the genetic resources, where possible. Within the MO-
SAICS framework (http://bccm.belspo.be/projects/mosaics/) a voluntary and
guiding code of conduct exists and covers access to, and circulation of, genet-
ic resources, a pathway that tracks utilisation and potential commercial benefits
arising from their exploitation. Permission must be gained from the local authori-
ties for collection activities. Some legal and policy issues surrounding ownership
of rhizobia have been discussed by Howieson and Fox (2012).
2.4 Collecting nodules
2.4.1Equipment
Where the collection site is remote from the laboratory, the following equipment
should be assembled to accompany the expedition:
small spade or lever (e.g. screwdriver) to prise legume roots from soil
scissors to remove nodules
27
plastic screw-topped vials with desiccant (e.g. silica gel) for long term (> two
days) nodule storage
plastic bags to store whole plants with soil attached to roots

permanent marker pens to label vials or bags
pH kit or pH meter
bulk aqueous diluent for pH measurement (e.g. 0.01MCaCl2)
file containing passport data recording sheets and pens
small envelopes for collecting seeds if they can be found
camera for recording legume and flower parts for identification, nodule mor-
phology and site characteristics (a scrap book for pressing plant parts is also
useful)
50mL vials for collecting soil (if nodules cannot be found)

GPS for recording site location.
2.4.2 Timing of collection

Nodules are easiest to collect when the soil is moist, as dry conditions can cause
the legume to shed its nodules or the nodules to desiccate and rupture. Soils are
usually moist during winter and spring in temperate and Mediterranean climates
or in the wet season(s) in subtropical and tropical environments.
When collecting nodules, the target legume must be identified, at least to the ge-
nus level. A sample of leaf, flower or pod may be pressed for future identifica-
tion of the species and lodged in a local herbarium. This is often logistically dif-
ficult but, at the very least, the general taxonomic indicators (flower, pod, leaf) of
the legume should be photographed and recorded along with other passport data
(see Figure 2.6). The researcher should be able to recognise general taxonomic
characters of the Leguminosae, particularly the different flower types of the three
subfamilies and their variations (Figure1.1, Chapter1). A secondary advantage
of collecting nodules in spring is the presence of flowers (to aid identification).
Healthy, vigorous and green plants are most likely to have fully effective symbi-
oses (see Chapter5) and resultant nodule isolates may be at the upper end of the
effectiveness scale for N2 fixation.
28
2.4.3 Excavation of plant roots

The plant roots must be carefully removed from the soil, as (except in very sandy
soils) nodules will be dislodged easily if the plant is pulled from the soil. A lever
(e.g. screwdriver) or a small trowel can be very useful to loosen the soil, and a
small volume of water can be carried to wash adhering material away from the
exposed roots. For perennial and shrub legumes, where nodules may be very deep
(>1m) excavation of only a portion of the root system may be possible. However,
the researcher must be satisfied that the exposed piece of root originates from
the target legume. Because many perennial legumes have mature roots which are
often devoid of nodules, fine surface roots or lateral roots must sometimes be
sought. It is often easier to look for a nearby young seedling of the same species
because perennial species often form tap root nodules in the first year of growth.
However, for trees, it may be necessary to go much deeper, as there are reports of
nodules found below 4m in species such as Prosopis (mesquite) (Felker and Clark
1982).
2.4.4 Healthy nodules

Functional nodules which are firm, and when cut in half are red or pink inside,
are the best to collect because isolation of rhizobia from these is usually more suc-
cessful. Shrunken, flaccid or desiccated nodules indicate senescence and should
be avoided. Nodules are best excised leaving a piece of root 25mm in length to
facilitate manipulation during the isolation steps. If healthy nodules cannot be
found, techniques for isolation from ruptured nodules are described in Chapter2,
Section 2.8.
2.4.5 Nodule description

A general description of the nodule morphology can later be useful when work-
ing with novel symbioses. Sprent et al. (2013) have defined nodule types as sum-
marised in Figure2.2.
29
Figure2.2 Legume nodule morphology. A: determinate, desmodioid. These

nodules are more or less spherical and have lenticels, usually as stripes,
but occasionally as stars (see C). B: determinate, aeschynomenoid. These
nodules are always associated with lateral or adventitious roots and do
not have lenticels. C: Indigoferoid, rather like an indeterminate desmodioid
nodule, its detailed structure has not yet been examined and has so far only
been reported from Indigofereae. D: lupinoid. E: indeterminate unbranched
nodules, common in mimosoids and other groups. F: indeterminate with
one or few branches, common in many papilionoids. G: indeterminate with
many branches, found in all subfamilies. In some genera, such as Ormosia and
Crotalaria, there may be more branching than shown here. Note that many
nodules of types F and G are unbranched when young. H: woody. Most nodules
that have fixation threads, both caesalpinioid and papilionoid, have a woody
scleroid outer layer when mature. Reproduced from the Interactive Legume
Database of Nodulation (ILDON) with permission.
30
2.5 Many nodules from one plant or a few nodules from many plants?
2.5Many nodules from one plant or a few nodules

from many plants?
Isolation from nodules (see Chapter3) is not always successful, so it is prudent

to collect several nodules from any root system. Tropical pulse legumes can have
more than 500 nodules, so an early decision must be made about the purpose of
the collection. Whether a researcher should source rhizobia from a single plant, a
diversity of plants in a defined region, or from a diversity of regions depends on
the purpose of the collection. A biodiversity study may wish to cover as many re-
gions as possible, but to describe the complete nodulating population in a discrete
niche, all the nodules from a single plant may need to be collected. In practice,
the researcher has to balance obtaining a representative selection of strains for the
target legumes with the time and resources available. It is important to recognise
that legumes do not necessarily form nodules with the most effective strains in
situ (Howieson 1999) although this is disputed theoretically (Denison 2000) and
so a reasonably large pool of nodules should be collected if a strain that fixes ni-
trogen optimally is required.
2.6 Storage of nodules
Nodules can be stored and preserved in screw-capped plastic tubes containing

desiccant material, such as anhydrous calcium chloride or silica gel, with a cotton
plug separating nodules from the desiccant in the bottom (Figure2.3).
Figure2.3 Plastic screw-capped bottles

of 5mL (left) and 20mL (right) with cotton
wool overlying desiccated silica gel for
storage of nodules
Remember to label the tube (not only its cap) with a permanent marker to re-
cord the location of the collection site and link it to the passport information
(Figure2.6). An alternative is to place a piece of paper with pencilled identifica-
tion inside the tube. Vials of 5mL volume are adequate for storing approximately
31
three to six pasture legume nodules and 20mL vials can be used for larger (e.g.
soybean) nodules. Some silica gel desiccants are blue when dry and pink when
moist; others are orange when dry but dark pink when moist. It is best to test
the silica gel by drying a sample at 60C overnight. The desiccant is designed to
quickly remove water from the nodules and keep them dry, preventing the growth
of other microorganisms. Nodules should be kept in this way for a maximum of
three months, however successful isolations have been made from nodules stored
for longer periods. A disadvantage of anhydrous calcium chloride desiccant is that
vials of white powder can be mistaken for illicit drugs when transiting borders.
If the legumes are collected in close proximity to the laboratory and can be han-
dled within a few days, roots may be bagged with soil attached and stored in a
cool box for transport. The plants can then be shaken free of soil, washed and the
nodules removed, again leaving a small portion of root attached. Fresh nodules
excised from roots may be stored for a few days in the refrigerator in a sealed plas-
tic bag but must not be frozen because internal water crystals formed by freezing
can kill the bacteroids.
2.6.1 Collection of soil for trapping nodule bacteria

Sometimes nodules cannot be found on the target legume, or they may be desic-
cated and ruptured, and there is little choice but to collect soil then attempt to
trap the rhizobia when back in the laboratory. A small amount of soil (e.g. 10g)
from the rhizosphere of the legume will suffice. If the soil is to represent a larger
geographic area, as in a biodiversity study, then an unbiased sampling procedure
is needed. Subsamples, randomly and spatially distributed, should be taken to
make a composite sample that is representative of areas of up to several hectares.
As when collecting nodules, the sampling location should be recorded using a ref-
erence map or GPS. Information about vegetation, previous cropping history and
historical application of chemicals are always very useful.
If the researcher elects to trap rhizobia from the soil, a further decision must be
made: with which legume to trap the rhizobia? The target legume for the study is
clearly the most appropriate (assuming seed is available) but a broad host range
legume such as siratro (Macroptilium atropurpureum) might be appropriate for
some studies.
2.6.1.1Do trapped rhizobia resemble those that might form the nodule
in situ?
There is some evidence that trapped strains represent a different population of

nodule bacteria to those isolated directly from nodules. One example is shown
in Figure2.4 from a study with common bean plants (Alberton et al. 2006). The
genetic diversity of strains as assessed by BOX-PCR (see Chapter11, Section 11.3)
varied according to whether they were sourced directly from nodules on the leg-
ume or from soil around the roots of the legume.
32
Figure2.4 Rhizobia isolated from common bean nodules of field-grown plants

(left) where 57% of the strains fit into three BOX-PCR profiles; and (right) from
greenhouse plants inoculated with a soil dilution (104) from the same field site,
and where none of the profiles represented more than 12% of the strains. Adapted
from Alberton et al. (2006).
There is also evidence that trapped strains possess less of the desirable character-
istics (success in nodulation, effectiveness for N2 fixation, ability to colonise soil)
than strains sourced from the nodules of vigorous legumes growing in soils that
resemble the target edaphic characteristics (Howieson 1999). Thus, the source
and method of isolation should be described in each study. For example, more
competitive strains may be more readily obtained from field-grown plants while
a higher diversity of strains may be obtained when inoculating soil dilutions onto
plants grown under greenhouse conditions.
For soils (or ruptured nodules) collected from remote locations, the samples
should be stored in a suitable container for return to the laboratory. If an estimate
of rhizobial number is required (see Chapter6) the samples should be kept cool.
At the laboratory, samples should be stored in the dark in a refrigerator (47C).
When bacterial counts are to be performed, soil samples must be stored for no
longer than 30 days.
33
2.7 Trapping rhizobia from soil or ruptured nodules
2.7 Trapping rhizobia from soil or ruptured nodules
Once back in the laboratory:
1. place the soil (or ruptured nodule) carrying the target rhizobia above a ster-
ile medium (sand or vermiculite) in a sterile pot or vial filled to 60% of its
capacity
2. cover the rhizobia-rich material with a shallow layer of sterile soil(Figure2.5)
3. wet the soil with sterile DI water
4. sow the trap seedling into the sterile layer (see Chapter5)the radicle then
emerges and grows down through the rhizobium-rich layer of soil to nodulate
5. cover with alkathene beads and insert watering tube for nutrient (see
Chapter5)
6. after four to five weeks, remove plants from the soil and isolate rhizobia from
the fresh nodules (see Chapter3).
Sterile beads
Sterile sand
Soil or nodule debris
Sterile sand
Figure2.5 The layering technique used in the CRS laboratories for trapping
rhizobia from small quantities of soil or from ruptured nodules
2.8 Specific host in situ trapping
A variation on trapping rhizobia from soils in the glasshouse is to sow the (surface
sterilised) target legume directly into the field soil, in situ. This might overcome
the tendency for different populations to be accessed from soil dilutions when
34
2.9 Collection of passport data
applied to legume roots in vitro. This technique was successfully applied to a pro-
ject seeking acid soil-tolerant rhizobia from acidic Sardinian soils, specifically for
Pisum sativum. This species does not naturally occur widely in rangeland settings
in Sardinia, hence surface sterilised seed was sown into a wide number of range-
land sites before the onset of seasonal rains, and surviving plants excavated 10
weeks later. One further advantage of this approach is that authentication can be
undertaken on the same source of seed as that which was planted.
2.9 Collection of passport data
A minimum data set, such as that referred to in Figure2.6, is recommended. The

type of soil, the altitude and rainfall, reference to a photo of the host or environ-
ment and latitude and longitude are all valuable parameters when describing the
source of nodule bacteria. An estimation of the land use, grazing pressure, slope,
water run-off and associated plant species is often also very valuable.
Figure2.6 An example of the passport data that should accompany nodule collection
35
2.10 Collection of seed (for authentication)
A laboratory soil analysis that quantifies the environment from which the strains
were collected is often also very useful in later studies that seek to understand ad-
aptation of rhizobia to new environments. For this, a 100g sample of soil will be
required.
An important parameter in rhizobial ecology is pH, and this can be determined at

the collection site with a portable universal indicator dye kit or by carrying small
quantities of diluent (DI water or 0.01M CaCl2) and a portable pH probe attached
to an electronic meter (Figure2.7).
Figure2.7 Soil pH is a valuable parameter to record. Note the yellow colour

produced by the Universal Indicator pH kit which approximates to a soil pH of
4.69 as registered on the portable electronic pH meter after dilution of soil in
0.01MCaCl2. This represents the lower limit for nodulation of annual species of
Medicago, shown here collected in a natural environment on the Cyclades island
of Tinos in Greece.
2.10 Collection of seed (for authentication)
Isolation of rhizobia from nodules and subsequent authentication are described

in Chapter3. However, authentication of isolates is greatly expedited if seed can
be collected from the same plant as the nodule. The conundrum is that if nodules
are collected in spring, when soils are moist, seeds are rarely mature. Conversely,
if nodules are collected when seeds are mature, they are often desiccated and rup-
tured.
Several resolutions are possible. Nodules can be collected in late spring, when
early maturing seeds may be found on the plant, or on the same species growing
36
2.11References
in nearby pockets of dry soil. Alternatively, and perhaps most conveniently, an-
nual legumes often have multiple hard seeds in pods (ungerminated seed) from
which a single soft seed has germinated. The residue pod containing hard seed
can often be found still attached to the legume root, and thus seeds represent-
ing siblings of the plant containing the nodule can be prised from the pod. This
is common for subterranean clover and most annual species of Medicago. Many
other legumes form hard seeds and these can be recovered from the legume pod
residues on the surface soil, particularly where these pods do not fully dehisce
(e.g. Lotus ornithopodioides, Ornithopus compressus, Lebeckia ambigua). Ensure
that collected seed is stored in a manner that maximises its viability. If seed can-
not be collected at the time of nodule collection, a GPS record will guide a subse-
quent site visit. After nodules have been collected, the task is to isolate, grow and
authenticate strains recovered from inside them. These activities are discussed in
the following Chapter.
2.11References
Alberton O., Kaschuk G. and Hungria M. 2006. Sampling effects on the assessment of genetic di-
versity of rhizobia associated with soybean and common bean. Soil Biology and Biochemistry
38, 12981307.
Date R.A. 1982. Collection, isolation, characterization and conservation of Rhizobium. Pp 95109
in Nitrogen fixation in legumes, ed. by J.M. Vincent. Academic Press: New York.
Denison R.F. 2000. Legume sanctions and the evolution of symbiotic cooperation by rhizobia.
American Naturalist 156, 567576.
Felker P. and Clark P.R. 1982. Position of mesquite (Prosopis spp.) nodulation and nitrogen fixation
(acetylene reduction) in 3m-long phraetophytically simulated soil columns. Plant and Soil 64,
297305.
Graham P.H. 1998. Symbiotic nitrogen fixation. Pp 325347 in Principles and applications of soil
microbiology, ed. by D. Sylvia, et al. Prentice Hall.
Howieson J.G. 1999. The host-rhizobia relationship. Pp 96106 in Genetic resources of Mediterra-
nean pasture and forage legumes, ed. by S.J. Bennett and P.S. Cocks. Kluwer Academic Publish-
ers: The Netherlands.
Howieson J.G., OHara G.W. and Carr S.J. 2000. Changing roles for legumes in Mediterranean agri-
culture: developments from an Australian perspective. Field Crops Research 65, 107122.
Howieson J.G. and Fox S.L. 2012 Plant growth promotion with micro-organisms. Pp 138149 in
Beneficial microorganisms in agriculture, food and the environment: safety assessment and
regulation, ed. by I. Sundh, A. Wilcks and M.S. Goettel. CABI.
Loi A., Nutt J., Howieson J.G., Yates R. and Norman H.C. 2012. Preliminary assessment of bladder
clover (Trifolium spumosumn L.) as an annual legume for ley farming systems in southern Aus-
tralia. Crop and Pasture Science 63, 582591.
Sprent J.I., Ardley J.K. and James E.K. 2013. From North to South: a latitudinal look at legume nodu-
lation processes. South African Journal of Botany 89, 3141.
Terpolilli J.T., OHara G.W., Tiwari R.T., Dilworth M.J. and Howieson J.G. 2008. The model legume
Medicago truncatula A17 is poorly matched for N2 fixation with the sequenced microsymbiont
Sinorhizobium meliloti 1021. New Phytologist 179, 6266.
37
3.1Introduction
CHAPTER3
Isolation and growth of

rhizobia
M. Hungria, G.W. OHara, J.E. Zilli, R.S. Araujo, R. Deaker
and J.G. Howieson
3.1Introduction
This chapter describes basic techniques for the isolation and growth of rhizobia,
some of which have been used for more than a century. While these techniques re-
tain their importance, the success of current and future rhizobiology studies and
enterprises will depend on the training, skills and techniques described in this
chapter. A note of caution: nodules (particularly those collected from the field)
are not always occupied by a single rhizobial isolate nor even by a single micro-or-
ganism. Nodules of pea and lupin, for example, have been described as containing
both the nitrogen-fixing symbiont and associative organisms such as Micromono-
spora (Trujillo et al. 2010). Hence, we must be prepared for a range of organisms
to appear on growth plates during isolation procedures. Recognition of rhizobia
when growing on a solid medium is an essential skill in rhizobiology.
3.2 Preparing solid growth media
A wide range of growth media are available for rhizobia and these are listed in
Section 3.7 of this chapter. For isolation from nodules, it is sufficient to choose one
of the routine (undefined) media such as YMA or LA as these are inexpensive,
simple to prepare and able to support a broad range of nodule bacteria. It is rec-
ommended that isolations be made upon solid media as contaminants are more
readily discernible than in liquid media. A further decision should be made as to
incorporation of Congo Red dye, which is readily adsorbed by Gram positive bac-
teria, but also some nodule bacteria (see Section 3.8.1).
39
3.3 Preparing the nodules
3.3 Preparing the nodules
1. Nodules may have been either stored desiccated or still attached to the legume
roots in the cold room as described in Chapter 2. When attached to roots,
these should be washed free of soil then nodules excised from the root (leav-
ing a small root section attached) and left in water until ready for isolation
(within 12hours).
2. Desiccated nodules should be re-hydrated by immersion in DI water for three

to four hours in a labeled Petri dish. Best results have been obtained with des-
iccated nodules stored for less than six months.
3.4 Isolating bacteria from nodules
The nodules must be surface sterilised to remove as many contaminants as pos-

sible because it is difficult to discern rhizobia growing on a plate if there are many
other bacterial contaminants present. A convenient technique to transfer the nod-
ule through the sterilising then rinsing solutions is to use forceps locked onto the
small piece of adjoining root. If many nodules are to be worked with, then they
can be held together in a tea strainer (Figure3.1) for bulk immersion in the solu-
tions.
1. Prepare sterilising solutions of 70% (v/v) ethanol and 4% (v/v) sodium hy-
pochlorite, and six changes of sterile water in vessels that can accommodate
the forceps (e.g. 20mL) or tea strainer (100mL). If working with many nod-
ules (>20) then prepare a second set of rinsing solutions which can be changed
to decrease the chances of transferring contaminants. Tween 80 (10L/L) can
be added to the sodium hypochlorite solution as a wetting agent. NB: com-
mercial hypochlorite can vary in sodium hypochlorite concentration between
2 and 6% (w/v).
2. Surface sterilise nodules to remove microorganisms on the surface by immers-

ing them for one minute in 70% (v/v) ethanol, followed by up to three minutes
in sodium hypochlorite. Nodules vary in their resilience to surface sterilisa-
tion and it may be necessary to vary this treatment. For example, to sterilise
the surface of Arachis nodules, we immerse in ethanol for only 10 seconds and
replace sodium hypochlorite with 3% (v/v) hydrogen peroxide.
3. Carefully rinse the nodules six times in sterile water by transferring the tea
strainer or locked forceps from vessel to vessel, ensuring the solution is drained
from the strainer each time.
4. From the last rinse, aseptically crush individual nodules with blunt-nosed
forceps held directly over the growth medium, allowing the contents to drop
onto the plate. Alternatively, the nodule can be macerated into a drop of sterile
40
3.4 Isolating bacteria from nodules
water in the lid of a Petri dish with the flattened end of a flamed glass rod or
sterile wooden stick.
5. Using a new plate for each nodule, streak the drop, or take a loopful of the
macerate and aseptically streak onto plates containing an appropriate growth
medium. Use a dilution streaking pattern to isolate single colonies (Figure3.2).
6. Incubate the inoculated plates at 28C and check every 24hours to observe
growth of the rhizobia and contaminant bacteria. Young colonies of most spe-
cies of rhizobia are translucent when young (Figure3.3).
7. A high percentage of isolations will result in mixed cultures (Figure3.3) and

care will be needed to identify and purify rhizobia.
8. Select single colonies that resemble rhizobia for subculture and then purify by
touching the edge of the loop to the colony then streak on a fresh plate.
9. At this point, a decision must be made as to how many cultures from a nodule
should be progressed. If more than one, select colony types that have different
morphologies and re-streak.
10. After the single colony subcultures have grown, store them for the short term
at 80C in a glycerol medium (see Chapter4, Section 4.5). A RAPD PCR or
a partial 16S rRNA sequence can later assist their differentiation (Chapters
11 and 12). To expedite this, take one loopful of the pure culture and place it
into 0.1mL sterile 0.89% (w/v) saline, then refrigerate for later creation of the
DNA template.
Figure3.1 A tea strainer can be used to transfer hydrated nodules between

sterilizing agents
41
3.5 Recognising nodule bacteria growing on solid media
Figure3.2 A dilution
streaking pattern for isolation
of single colonies. This pattern
would suit a left-handed
person. The inoculating loop
should be flamed when starting
each new set of strokes to
ensure adequate dilution of the
culture and development of
well-separated colonies.
3.4.1 Isolation from damaged nodules

It is difficult to obtain cultures from a damaged nodule because the sterilants can
enter the nodule tissue and kill the rhizobia. If the nodule contains a potentially
valuable strain, we suggest macerating the nodule in a drop of sterile water and
then inoculating onto a seedling growing under aseptic conditions, as described
in Chapter2, Section 2.8. Of course, this approach cannot preclude rhizobia on
the outside of the damaged nodule from forming any new nodules on the trap
plants.
3.5Recognising nodule bacteria growing on solid

media
Rhizobia are generally slightly raised (although Burkholderia and Bradyrhizobium

elkanii are nearly flat) and with entire margins, but the growth rates of the major
genera differ substantially. More details of the physiology of rhizobia are given
in Chapter7, and their appearance in Table7.3, but here we briefly cover growth
rates and colonial morphology to aid in selection of the correct colonies from the
isolation exercise.
The recognised factors affecting the growth rate of a rhizobial species are culture
medium, pH and temperature. On LA (see Section 3.7.2) at 28C we expect to
see growth from nodule squashes begin within:
one to two days for very fast growers, such as Burkholderia and Cupriavidus
two days for Microvirga
two to fourdays for classic fast growers, such as Rhizobium and Sinorhizobium
three to fivedays for Mesorhizobium
42
six to 14days for Bradyrhizobium

Helpful notes
Growth rates from a nodule are generally slower than from subcultures, pre-
sumably as the bacterial metabolism for growth on artificial media has to be
expressed.
Nearly all colonies of nodule bacteria are translucent when first visible and
viewed with a dissecting microscopic with light from below (Figure 3.3). If
plates are viewed regularly, dominant patches of translucent bacterial growth
can be identified and marked with a felt-tipped pen.
Figure3.3 Young colonies appear translucent for most nodule bacteria; bradyrhizobium 7d (left),
rhizobium 2d (right).
More often than not, a wide diversity of colony types will grow on the isolation
plates, and it is very helpful in diagnosis of the colonies to record growth every
24hours and to circle colonies that appear in each 24-hour period. Figure3.4 il-
lustrates a range of common nodule bacteria growing on a medium containing
Congo Red. Colonies that are not rhizobia tend to absorb the dye strongly.
43
Rhizobium
Bradyrhizobium
Sinorhizobium
Figure3.4 A mixed culture of bacteria growing on Congo Red YMA medium.

Many of these colonies could be rhizobia and rhizobiologists should become
familiar with the appearance of their cultures. Some morphological features to
help differentiate species are listed in Table3.1.
3.5.1 Viewing colonies through a dissecting microscope

It is helpful to visualise bacteria growing on solid medium with the aid of a dis-
secting microscope. With light from below, and with low magnification, an expe-
rienced microbiologist can become familiar with the subtle differences between
the major genera in shape and hue of the colonies. Burkholderia and Cupriavi-
dus often have a light yellow-brown hue in early growth, but this becomes more
opaque with age (Figure 3.5). Burkholderia sprentiae has less of a brown tinge
than Burkholderia dilworthii and grows faster. The beta-rhizobia can have colo-
nies whose margins are slightly irregular.
Figure3.5 Colonies of Burkholderia after 24hour growth on LA. Burkholderia dilworthii (left)
and B. sprentiae (right), as described in De Meyer et al. (2013). Note the brownish tinge of B.
dilworthii when viewed with light from below, and very slightly irregular margins.
44
Sinorhizobium medicae nearly always produces colonies with a doughnut mor-

phology when isolated directly from nodules, but this may disappear after pro-
longed subculture. Figure3.6 illustrates the different colonial morphology of S.
medicae compared with S. meliloti when grown on LA medium.
Other photographs of colonies of different species of rhizobia can be accessed at

the Centre for Rhizobium Studies (CRS) website: www.crs.murdoch.edu.au/
Figure3.6 S. medicae (A, C) and S. meliloti (B, D) colonial morphology and gum production when
grown on LA medium and lit from below with an incandescent globe. The morphology of S.
medicae is referred to as doughnut (Chatel DL pers. comm.)
45
3.6 Some useful diagnostic features of species of nodule bacteria
3.5.2 Pigmented rhizobia

Root nodule bacteria are rarely pigmented but some specific exceptions have
emerged. Methylobacterium isolates from Listia bainesii (formerly Lotononis bai-
nesii) produce red-pigmented colonies (Norris 1958; Godfrey 1972; Kleinig and
Broughton 1982) (Figure3.7). Microvirga, a recently identified genus of nodulat-
ing bacteria, may produce pink or light orange colonies after 48hour incubation
on LA (Ardley et al. 2011) (Figure3.7).
Figure3.7 Red colonies of Methylobacterium spp. from nodules of Listia bainesii (left) and pink-
orange colonies of Microvirga spp. from nodules of Listia angolensis (right)
Pigmented colonies (as for very rapidly growing colonies) should not automati-
cally be discarded.
3.6Some useful diagnostic features of species of

nodule bacteria
The following table lists diagnostic features of various nodule bacteria, (based on
long experience with them) which can assist in the identification of isolates from
nodules. All colonies are non-pigmented, convex (or elevated) and with entire
margins unless stated otherwise.
46
3.7 Routine culture media for rhizobia
Table3.1 S
ome features of the common nodule bacteria that aid in working with them. This table
will be regularly updated at the CRS website: www.crs.murdoch.edu.au/
Species Features
Bradyrhizobium spp. Colonies slow growing and translucent at first appearance
when less than 1mm in diameter (approx. six to ten days) but
then strongly opaque and tending to dark grey with further
maturity. Rarely gummy.
Bradyrhizobium japonicum As above, but may reach 13mm after five to eight days,
immediately opaque and elevated.
Bradyrhizobium elkanii As above, colonies irregular and non-elevated.
Burkholderia spp. Colonies visible in 2436hours, flat, margin regular except B.
dilworthii, colonies may acquire a yellow-brown tinge after
48hours when viewed with light from below (see Figures. 3.5,
3.11a and 3.11b).
Burkholderia sprentiae As above, but colonies more grey and opaque than brown-
yellow, and more raised (see Figures 3.5 and 3.10). Growth
slightly faster that B. dilworthii.
Cupriavidus spp. Colonies visible in 24 hours, flat, margin not entire, grey tinge
when viewed with light from below.
Mesorhizobium spp. Colonies as for Rhizobium but visible 24h later, weakly
opaque after 72h, increasing with age.
Methylobacterium Colonies visible in four to seven days. Strongly opaque and
not gummy.
Microvirga lotononidis (from Listia angolensis) Colonies visible in 24h; light pink develops 4872h later
(Figure3.7), weakly opaque and mildly gummy. Optimal
growth at 41C.
Microvirga zambiensis (from Listia angolensis) Colonies visible in 24h. Weakly opaque and mildly gummy.
Optimal growth at 35C.
Microvirga texensis (from Listia angolensis) Colonies visible in 24h. Light orange coloration develops 72h
later, opaque and not gummy.
Methylobacterium spp. (from Listia spp. other Colonies visible in four to seven days. Colonies develop
than L. angolensis) strong pink pigmentation within 24h of becoming visible
(Figure3.7).
Rhizobium leguminosarum bv. trifolii Colonies visible in 48hours. Weakly opaque at 72hours.
Gummy strands adhere to loop.
Rhizobium leguminosarum bv. viceae Colonies visible in 48hours. Weakly opaque at 72hours.
Gummy strands generally do not adhere to loop but variation
evident according to host genus.
Sinorhizobium medicae Colonies visible in 48hours. Doughnut morphology when first
isolated from nodules (Figure3.6A and 3.6C), weakly opaque
after 72hours and gummy.
Sinorhizobium meliloti Colonies visible in 48hours. Do not form doughnut colonies,
weakly opaque after 72hours (Figures. 3.6B and 3.6D) often
(but not always) less gum than S. medicae.
Of a variety of media described in the literature for growing rhizobia, only the
most commonly used will be listed here. Specific strains or species may require
modifications to each medium.
47
The culture media can be classified as undefined or defined. The composition of

undefined media is not completely known, for example when a generic yeast ex-
tract is added to provide vitamins. Defined media are completely chemically spec-
ified, and are used for growing rhizobia with specific nutrient requirements, or
to determine metabolic behaviour and growth products. Many molecular studies
utilise generic bacterial media such as TY (3.7.2.3) for growing rhizobia, as these
have been developed over the last several decades, but they tend to distort the
unique colonial morphology of individual species.
3.7.1 pH Control
Bacterial growth on either liquid or solid medium can result in large and rapid
changes in pH as a result of metabolism (see Chapter7, Section 7.2.6). In general,
growth on sugars results in acidification, while growth on organic acids or amino
acids results in alkalinisation (Figure3.8). Where a constant pH is important, for
example when acid-tolerant bacteria are being evaluated, addition of an appro-
priate non-toxic buffer is essential to maintain pH. The buffer should be selected
so that the desired pH for the medium is within 0.7pH unit of the pKa listed for
the buffer (see Table3.2). Growth in a medium with the buffer at the desired con-
centration (usually 2040mM) needs to be as good as that in its absence, other-
wise the buffer is having an effect on bacterial growth outside of effects upon pH
control. This usually means testing growth at a range of buffer concentrations at
normal pH. For many rhizobia, growth rate decreases as pH decreases below 6.5,
as shown in Figure3.9.
Figure3.8 The pH change on unbuffered LA medium (with bromthymol blue indicator) caused
by rhizobial metabolism. The control medium is light green (far right). Colonies spotted onto the
plates in four replicates have acidified the medium, turning it from light green to yellow (far left
WSM3556) or have caused an alkaline reaction (central plate CB1809) turning the indicator blue.
The pH change has permeated the whole plate, but it is strongest under the colonies themselves.
Species are Burkholderia dilworthii WSM3556 and Bradyrhizobium japonicum CB1809.
48
Figure3.9 S. medicae grows well at pH 7 (top left) and at pH 6.5 in MES buffered
media (top right) but decreases at pH 6 (bottom right) and barely grows at pH 5.5
(bottom left). Growth on YMA at pH 7 (middle plate) is gummier than on LA.
Table3.2 Buffers used in rhizobial growth media and their pKa

Chemical name Abbreviation pKa at 25C
3-(Cyclohexylamino)-2-hydroxy-1-propansulfonic acid CAPSO 9.6
N-(2-hydroxyethyl)piperazine-N-(2-ethansulfonic acid) HEPES 7.5
2-N-Morpholino-ethansulfonic acid MES 6.1
Homopiperazine-1,4-bis(2-ethansulfonic acid) HOMOPIPES 4.55
49
3.7.2 Undefined media
3.7.2.1 Yeast mannitol agar
The universal culture medium used to grow rhizobia in laboratories worldwide is

YMA, as described by Fred and Waksman (1928) and Vincent (1970) with slight
modifications through the decades.
Component Quantity/L
K2HPO4 0.5g
MgSO4.7H2O 0.2g
NaCl 0.1g
Mannitol1 5g
Yeast extract 2
0.4g
Distilled water to complete 1L
Agar 1215g
For liquid medium (YMB), omit the agar. The pH should be adjusted to 6.5 to 6.8 before
autoclaving and before addition of agar. Autoclave at 121C for 15minutes.
1
The original recipe calls for 10g of mannitol, but growth of the great majority of rhizobia is
adequate with only 4g or 5g. Other cheaper or preferred C-compounds may be used.
2
Can be substituted by the addition of 100mL of yeast water, prepared from 100g of bakers
compressed yeast mixed with 1L of cold water, allowed to stand at room temperature for
one to two hours, autoclaved for 40 to 60 minutes at 121C, allowed to settle (or centrifuged)
and the clear supernatant adjusted to pH 68 (Vincent 1970).
3.7.2.2 LA (Howieson and Ewing 1986)
This is a useful medium for isolation from nodules that allows differential colo-
nial morphology of the different species to be expressed. The concentration of P is
low by normal standards and serves to reduce culture gumminess in genera such
as Sinorhizobium (Figure 3.9). However, some rhizobial species (e.g. those that
nodulate Hedysarum spinosissimum) require a fivefold increase in P concentration
to grow on this medium.
D-glucose 5g
Mannitol 5g
MgSO4.7H2O 0.8g
NaCl 0.1g
Yeast extract 1.25g
CaCl2.2H2O 0.2g
Agar 1215g
Liquid solutions* Final quantity/L

K2HPO4 0.16 mg
KH2PO4 0.13 mg
FeSO4.7H2O 0.1mg
*make stocks at 50 concentration
50
Trace elements solution**

Na2B4O7.10H2O 2.34mg
MnSO4.4H2O 2.03mg
ZnSO4.7H2O 0.22mg
CuSO4.5H2O 0.08mg
Na2MoO4.2H2O 0.126mg
**make stock at 1000 concentration
Weigh out components (except agar) and add into 800mL of distilled water while
stirring.
To this, add the three liquid stock solutions made at 50-fold higher concentration
(20mL of K2HPO4, 20mL of KH2PO4, 10mL of FeSO4) and 1mL of trace element
stock (made at 1000 concentration). Adjust the pH to 6.8 using 0.1M NaOH.
Add agar if required and make volume up to 1L before autoclaving at 121C for
15 minutes.
3.7.2.3Tryptone yeast extractTY

(for molecular studies, Beringer 1974)
Component Quantity
Tryptone 5.0g
Yeast extract 3.0g
CaCl2.6H2O (or 2H2O) 1.3g (0.87g)
pH adjusted to 6.87.0
3.7.2.4 Peptone (Hirsch et al. 1980)
Component Quantity
Peptone 4.0g
MgSO4.7H2O 0.5g
pH adjusted to 6.87.0
3.7.2.5 Luria-Bertani (Hirsch et al. 1980)
Component Quantity
Tryptone 10.0g
Yeast extract 5.0g
NaCl 3.0g
pH adjusted to 7.5
51
3.7.2.6Sucrose glucose broth (for fast-growing

rhizobia, Date and Halliday 1979)
Component Quantity
K2HPO4 0.75g
MgSO4.7H2O 0.4g
CaCO3 0.4g
Sucrose 2.5g
Glucose 2.5g
Yeast extract 3.0g
3.7.2.7Glycerol broth (for slow-growing

rhizobia, Date and Halliday 1979)
Component Quantity
K2HPO4 0.5g
(NH4)2HPO4 0.3g
KNO3 0.8g
MgSO4.7H2O 0.2g
Yeast extract 4.0g
MnSO4 0.1g
FeCl3 0.1g
Glycerol 10g
Distilled water to complete 1 L
3.7.3 Defined media
3.7.3.1 BSM (modified from Bergersen 1961)
Component Quantity
Mannitol 10g
Na2HPO4.12H2O 0.45g
MgSO4.7H2O 0.1g
Fe solution2
0.6mL
Glutamic acid 1.1g
Thiamine solution1 1 mL
Biotin solution1 1 mL
pH adjusted to 7
1
Vitamin stock solutions (1mg/mL) should be prepared in water separately, filter sterilised
through a 0.2m Millipore membrane and mixed when the medium is at a temperature of
4555C).
2
100mL of distilled water, 0.67g FeCl3.6H2O, 0.42mL conc. HCl.
52
3.7.3.2 Defined medium (Brown and Dilworth 1975)
Glucose 2.5g
KH2PO4 1
0.36g
K2HPO4 1
1.4g
N-source2 0.7g
MgSO4. 7H2O 0.25g
CaCl2.2H2O 0.02g
NaCl 0.2g
Liquid stocks 3
Final quantity/L
FeCl3 6.6mg
EDTA 0.15mg
ZnSO4.7H2O 0.16mg
Na2MoO4.2H2O 0.2mg
H3BO3 0.25mg
MnSO4.4H2O 0.2mg
CuSO4.5H2O 0.02mg
CoCl2.6H2O 1.0g
Vitamins Final quantity/L
Thiamine-HCl 1mg
Ca pantothenate 2mg
Biotin 1g
pH adjusted to 7.0
1
Phosphates are sterilised separately to avoid precipitation.
2
N source can be either KNO3 (0.7g), NH4Cl (0.7g) or L-glutamate (1.0g).
3
Make stock solutions at 1000 concentration and then add 1mL per L of medium.
53
3.8 Antibiotics and indicators
3.7.3.3JMMa minimal defined medium

(OHara et al. 1989)
D-galactose 1.8g
L-arabinose 1.5g
L-glutamate 0.51g
MgSO4.7H2O 0.25g
CaCl2.2H2O 0.15g
Na2SO4 0.1g
Liquid stocks 1
Final quantity/L
FeSO4.7H2O 5.5mg
ZnSO4.7H2O 1.1mg
Na2MoO4.2H2O 1.0mg
MnSO4.4H2O 1.1mg
CuSO4.5H2O 0.5mg
KH2PO42 0.22g
K2HPO4 2
0.26g
Biotin2
20g
Thiamine-HCl2 1mg
Ca pantothenate 2
1mg
These liquid stocks are made up at 1000 concentration.

1
The phosphates and vitamin liquid stocks are prepared as separate solutions, filter sterilised
2
and added to the cooled medium after autoclaving to give the required final concentration.
Buffers: Depending on the desired pH an appropriate buffer (HEPES, MES,

HOMOPIPES, CAPSO) is added at 2040mM (see Section 3.7.1)
Adjust pH to required level. Autoclaving does not usually alter the pH of buffered
JMM medium but it can significantly alter the pH of unbuffered JMM medium.
Add 1215g/L of agar for solid medium.
All media described in this chapter should be autoclaved at 121C at 1atm pres-
sure for 20 minutes. Ideally, wait at least two days before using any medium to be
sure that the sterilisation process was successful.
Antibiotics and indicators can be very useful for avoiding contaminants and for
helping in the authentication of strains. The antibiotics can be added through a
sterile filter from stock solutions after autoclaving the medium, while the dyes can
generally be added before autoclaving.
54
3.8.1Preparation
Actidione (cycloheximide) can be used to suppress fungal growth. For the stock
solution, add 25mg of cycloheximide in 300L of ethanol. Add 200L of the
stock solution per 300mL of culture medium (final concentration, 55mg/L).
Vancomycin is useful for inhibiting growth of Gram-positive bacteria. For the

stock solution, dissolve 9mg of vancomycin chloride in 3mL of distilled water.
Add 100L of stock solution per 300mL of culture medium (final concentration,
1mg/L).
Congo Red is an indicator that may be very useful for differentiating rhizobia
from contaminants, especially Gram-positive bacteria, as nodule bacteria do not
tend to adsorb the dye. Thus, it is a useful addition to media for isolation of rhizo-
bia from nodules, where contaminants are common. Routine addition of Congo
Red to media is not recommended. Stock solution should be prepared by adding
0.25g/100mL. Add 10mL of stock solution per L of culture medium (final con-
centration: 25mg/L) immediately before autoclaving.
Some nodule bacteria, such as Sinorhizobium, do adsorb Congo Red, and others
will over time, so growth must be carefully monitored. Figure3.10 shows a culture
of Burkholderia sprentiae which adsorbs Congo Red (right plate) with white colo-
nies on the left in the medium without Congo Red.
Figure3.10 A culture of rhizobia (Burkholderia sprentiae strain WSM3618) on LA medium with

Congo Red (right) where colonies have absorbed the dye. Opaque colonies are seen on the LA
medium without Congo Red (left). See also Chapter7.
55
Because rhizobial strains generally do not absorb Congo Red, the use of this dye
has been a practical way of identifying contaminants on agar. However, the species
of nodulating Burkholderia can be separated by their slightly differential adsorp-
tion of Congo Red and the reaction of the medium to their growth (Figure3.11a
and 3.11b)
Figure3.11a Variable reactions of Burkholderia spp. on LA with Congo Red. B. sprentiae (left)
adsorbs the dye strongly while B. tuberum (centre) less so. B. dilworthii (right) causes purpling of
the medium and adsorbs the dye after approximately eight days. A close up of B. tuberum WSM4180
is reproduced below.
Figure3.11b B. tuberum
strain WSM4180 grown on
LA with Congo Red and
showing the older growth
which does not strongly
adsorb the dye
Bromothymol blue is a broadly used indicator of acid or alkaline reaction in cul-

ture medium.
Prepare a stock solution of 0.5% (w/v) bromothymol blue in 0.2M KOH. Add
5 mL/L of the stock solution to obtain a green coloration just before autoclav-
ing. Figure3.8 shows species of rhizobia giving acid and alkaline reactions when
grown on LA containing bromothymol blue.
56
3.9 Avoiding contamination
Morphological characterisation should be undertaken with freshly grown cul-

tures, i.e. after one to two days for the Burkholderia, three to five days for fast
growers, six to nine days for slow growers and more than 10 days for extra-slow
growers.
3.9 Avoiding contamination
Basic microbiological principles should be followed to exclude contaminants

when pouring agar plates or inoculating them. The most important precautions
are:
rinse hands with ethanol or wear sterile gloves

work in a laminar flow cabinet or adjacent to a flame
flame inoculating loops and needles to red heat but allow to cool before use
flame the necks of flasks, bottles and vials
keep the room clean and free of dust
avoid strong air currents, such as air conditioning fans.
The sterilisation performance of the autoclave should be checked periodically by
commercially-available spore test strips (containing spores from Bacillus). Alter-
natively, plates or small volumes of broth cultures of sporulating bacteria (Bacillus
subtilis is a useful species) can be placed at several points in the autoclave, to verify
if adequate sterilisation is occurring.
Flow hoods facilitate microbiological work under sterile conditions but if the fil-
ters are not changed according to the manufacturers recommendation they can
represent a source of contaminants. The hoods should be cleaned with 70% (v/v)
ethanol and if UV lights are fitted, they should be turned on for about 1520 min-
utes, always screening the opening with paper, appropriate glass or a dark plastic
to protect the operator against UV damage to the eyes. Hoods should be allowed
to run for about 10 minutes before use and for about two to three minutes after
each stage of manipulation so that aerosols can be flushed from the system.
Verification of air filtration in cabinets should be checked periodically by leaving

exposed plates with culture media in several positions throughout the hood while
in operation. Colony growth on media indicates the presence of contaminants in
the air flow.
If access to a commercial hood is not possible, a cabinet can be built (Figure3.12)

or microbiological work can be carried out on a bench adjacent to a Bunsen burn-
er where convection causes air to rise away from the work area.
57
3.10 Authentication of isolates and Kochs postulates
Figure3.12 A cross sectional view of a laminar flow cabinet that is simple

to construct and that protects microbiological work from contamination.
Approximate dimensions are 1m wide 1.25m high 1.5m long (from
Somasegaran and Hoben 1994).
3.10Authentication of isolates and Kochs

postulates
Authentication is the term given to obtaining proof that the selected isolate is
indeed a nodulating bacterium. There have been many examples where substan-
tial research has been undertaken on bacteria whose origin is uncertain, and the
authentication process avoids some of this confusion. However, non-nodulating
variants of rhizobia can always emerge or some rhizobia can lose the symbiotic
plasmid and stop nodulating. There is also increasing evidence of dual nodule
occupancy, sometimes with bacteria that require assistance to enter the nodule.
A recent example of this is the non-nodulating strains of Burkholderia associated
with Lebeckia ambigua (Howieson et al. 2013).
Techniques for preparing plants for authentication are discussed in Chapter5.
There is always a compromise between storing isolates permanently (to avoid pro-
longed subculture on media) while awaiting evidence that the cultures are nodule
bacteria, and the time and resources required for acquiring this evidence. A useful
approach when dealing with large numbers of strains is as follows.
58
3.10 Authentication of isolates and Kochs postulates
Keep a set of non-inoculated plants of the host legume always available, grow-
ing in vials or pots (Chapter5) with supplied N to keep them healthy. Unusual
colonies can be immediately applied to the legume and nodulation can be ob-
served relatively rapidly.
Be prepared to preserve unauthenticated material in glycerol vials for the me-

dium term.
Methods for short-term preservation are detailed in Chapter4, and for plant in-
fection tests in Chapter5.
3.10.1 Kochs postulates

The fundamental principles for isolation of rhizobia are the same as for the iso-
lation of any infective bacterium (pathogenic or beneficial) as outlined by Koch
(1884). Kochs postulates need to be fulfilled to provide the evidence that rhizobia
have been isolated. The postulates recommend the following steps:
1. the infecting organism (rhizobia) should always be present in the host when
the disease (nodulation) occurs
2. the probable cause of the infection (rhizobia) must be isolated from the host
(nodule) and grown in pure culture
3. the organism obtained in pure culture, when inoculated back on the host,
must infect (nodulate) the host
4. the organism believed to be the cause of the infection (nodule) must be iso-
lated again, grown in pure culture and compared with the initial isolate.
Since Koch wrote these guidelines, we have obtained a slightly wider view in the
legume symbiosis.
Pseudo nodules can arise as legume roots perceive nod factor from rhizo-
bia and in response form nodule-like growths. These nodules do not contain
rhizobia and may thus be in conflict with step1.
Nodules may be crushed and the occupants subjected to a PCR-based iden-

tification technique, such as RAPD or ERIC that can provide convincing evi-
dence of the identity of the strain in the nodule without the need to culture the
organism. This is somewhat in conflict with step3.
Similarly, utilizing knowledge from whole genome sequences, a mass spec-

trometry profile of proteins through MALDI-TOF can identify some rhizobial
species to strain level from a nodule crush. The same approach can give cor-
roboration to the nodulating organism, as required in step4.
A variety of techniques can add value to the authentication process, and most
of these are discussed in this manual. Depending on the laboratory equipment
and expertise available, these can range from a careful morpho-physiological
characterisation through assay of serological properties to a variety of molecular
59
3.11References
techniques, such as DNA analysis by rep-PCR (e.g. Menna et al. 2009) or sequence
analysis. The full characterisation of strains is described in other chapters of this
manual; morpho-physiology (Chapter7) and genetic characterisation (Chapters
11 and 12).
Methods to infect legumes with the isolated nodule bacteria and to confirm nodu-
lation are covered in Chapter5.
3.11References
Ardley J.K., Parker M.A., De Meyer S., Trengove R.D., OHara G.W., Reeve W.G., Yates R.J., Dilworth
M.J., Willems A. and Howieson J.G. 2011. Microvirga lupine sp. nov., Microvirga lotonoides sp.
nov. and Microvirga zambesiensis sp. nov. are Alphaprotobacterial root nodule bacteria that spe-
cifically nodulate and fix nitrogen with geographically and taxonomically separate legume hosts.
International Journal of Systematic and Evolutionary Microbiology 62, 25792588.
Bergersen F.J. 1961. The growth of Rhizobium in synthetic medium. Australian Journal of Biological
Sciences 14, 349360.
Beringer J.E. 1974. R factor transfer in Rhizobium leguminosarum. Journal of General Microbiology
84, 188198.
Brown C.M. and Dilworth M.J. 1975. Ammonia assimilation by Rhizobium cultures and bacteroids.
Journal of General Microbiology 86, 3948.
Date R.A. and Halliday J. 1979. Selecting Rhizobium for acid infertile soils of the tropics. Nature
277, 6264.
Fred E.B. and Waksman S.A. 1928. Laboratory manual of general microbiology. McGraw-Hill Book
Company: New York.
Godfrey C.A. 1972. The carotenoid pigment and deoxyribonucleic acid base ratio of a Rhizobium
which nodulates Lotononis bainesii Baker. Journal of General Microbiology 72, 399402.
Hirsch P.R., Van Montagu M., Johnston A.W.B., Brewin N.J. and Schell J. 1980. Physical identifica-
tion of bacteriocinogenic, nodulation and other plasmids in strains of Rhizobium legumino-
sarum. Journal of General Microbiology 120, 403412.
Howieson J.G. and Ewing M.A. 1986. Acid tolerance in the Rhizobium meliloti Medicago symbio-
sis. Australian Journal of Agricultural Research 37, 5564.
Howieson J.G., de Meyer S., Vivas-Marfisi A., Ratnayake S., Ardley J.K. and Yates R.J. 2013. Novel
Burkholderia bacteria isolated from Lebeckia ambiguaa perennial suffruticose legume of the
fynbos. Soil Biology and Biochemistry 60, 5564.
Kleinig H. and Broughton W.J. 1982. Carotenoid pigments in a red strain of Rhizobium from Loton-
onis bainesii Baker. Archives of Microbiology 133, 164.
Koch R. 1884. Die Aetiologie der Tuberculose. Mittheilungen aus dem Kaiserlichen Gesundheit-
samt 2, 188.
Menna P., Pereira A.A., Bangel E.V. and Hungria M. 2009. rep-PCR of tropical rhizobia for strain
fingerprinting, biodiversity appraisal and as a taxonomic and phylogenetic tool. Symbiosis 48,
120130.
De Meyer S.E., Cnockaert M., Ardley J.K., Maker G., Yates R.J., Howieson J.G. and Vandamme P.
2013. Burkholderia sprentiae sp. nov., isolated from Lebeckia ambigua root nodules. Interna-
tional Journal of Systematic and Evolutionary Microbiology 63, 39503957.
Norris D.O. 1958. A red strain of Rhizobium from Lotononis bainesii Baker. Australian Journal of
Agricultural Research 9, 629632.
OHara G.W., Goss T.J., Dilworth M.J. and Glenn A.R. 1989. Maintenance of intracellular pH and
acid-tolerance in Rhizobium meliloti. Applied and Environmental Microbiology 55, 18701876.
Somasegaran P. and Hoben H. 1994. Handbook for rhizobia. Springer-Verlag: New York.
Trujillo M.E., Alonso-Vega P., Rodrguez R., Carro L., Cerda E., Alonso P. and Martnez-Molina E.
2010. The genus Micromonospora is widespread in legume root nodules: the example of Lupinus
angustifolius. The ISME Journal 4, 12651281.
Vincent J.M. 1970. A manual for the practical study of root-nodule bacteria. (International Biologi-
cal Programme Handbook 15). Blackwell: Oxford, U.K.
60
4.1 The need to preserve cultures
CHAPTER 4
Preservation of rhizobia
M. Hungria, J. Ardley, G.W. OHara and J.G. Howieson
4.1 The need to preserve cultures
Experimentation with strains of rhizobia can last for many decades; hence there
must be a reliable and efficient means of storing the bacteria. While many re-
positories of rhizobia have been developed since the symbiosis was scientifically
understood, few remain available for exploitation. This is because strains were
commonly stored on agar and remained the responsibility of an enthusiast, who
may not necessarily have been replaced by his institution upon retirement. Agar
slope-borne cultures have a relatively finite life. For this reason we recommend
long-term preservation of valuable cultures lyophilized in glass tubes which will
ensure survival over long periods of inattention.
4.2 Options for preservation
There is a trade-off, however, between the requirement for short-term storage af-
ter strains have been isolated from a nodule and passed through the purification
process, and long-term storage to preserve the integrity of the strain. One way to
manage this logistically is to have a short-term system based upon agar or glycerol
storage (with strains labeled in a temporary code) and then long-term preserva-
tion after authentication (Chapter5). Storage methods to keep rhizobia alive need
to minimise the opportunity for variation or mutation, because strains may lose
desirable properties during storage, or after repeated subculture. Careful main-
tenance of stock cultures, and periodic testing of their symbiotic efficiency, are
recommended.
Several methods for short- and long-term storage are listed in Table 4.1, with
notes on their characteristics. It is recommended that the survival of strains be
examined earlier rather than later.
61
4.2 Options for preservation
Table4.1 M
ethod of maintenance, main characteristics and cell viability related to
each method (Hungria et al. 2005).
Method Main characteristics Viability
Agar slopes Medium usually undefined, kept at 57C for periodic 1 year
transfer; simple and low-cost
Agar slopes As above, covered with sterilised mineral or paraffin oil, 2 years
kept at 57C; simple and low-cost
Porcelain beads Dry suspension of cells on sterilised porcelain beads, 2 years
kept in a tube with dehydrated silica at 57C
Soil, peat or clays Preferably material with high water activity, ground, 24 years
corrected for chemical properties and sterilised at
57C.
Lyophilization Viability depends on the physiological state of the Several
culture, cell concentration, medium and lyophilization decades
rate; can be kept at room temperature for decades, but
little information is available
Freezing Storage in temperatures ranging from 70C to 190C Months to
in deep freeze or liquid nitrogen. Viability depends several years
on the culture medium, freezing speed, freezing
temperature and type of cryoprotectant used; good
viability has been shown in a number of collections
after 1520 years.
4.2.1 Storage on agar slopes

1. Select preferred growth medium (Chapter3).
2. Add 1520g agar per litre and stir vigorously with heating to dissolve the agar
and ensure it is thoroughly mixed.
3. Dispense mixture into a screw-capped container to fill 33% of the volume (i.e.
10mL into a 30mL McCartney bottle or 1mL into a 3mL plastic vial).
4. Place vials into a rack. After autoclaving, rest the rack at an angle of 4560
degrees until set.
5. Take a loopful of culture and streak across the surface of the agar slope; allow
to grow until visible.
6. Store at 57C.
7. Cover with sterile paraffin or mineral oil to decrease the rate of desiccation.
4.2.2 Storage on porcelain beads

1. Pack (autoclavable) vial to 30% of volume with desiccated silica gel covered
by cotton wool, top with cleansed porcelain beads, screw cap on loosely then
autoclave.
2. Grow culture in selected liquid medium (Chapter 3) to visible turbidity or

wash from solid medium with diluent.
62
4.3 Longer-term preservation
3. Add drops of culture to porcelain beads under sterile conditions, screw top
tight and store at 57C
4. When needed, one bead is removed and dropped into a broth medium. This
can be allowed to grow or can be shaken then streaked on solid medium to
check for purity.
4.3.1 Lyophilization (vacuum drying)

The main objective of the lyophilization process is to preserve living organisms
without modifications in their physiological, biochemical or genetic properties.
The principle of the process is to remove moisture from the culture under a strong
vacuum, with the evaporation of moisture producing a drop in temperature. The
culture becomes desiccated but enough cells survive, protected by the low tem-
perature and preserving medium. Cultures are then sealed in glass ampoules
which maintain the very low vacuum. Several cryoprotectors have been employed
including dimethylsulfoxide, glycerol, albumin, skim milk and peptone (Hublek
2003; Day and Stacey 2007). We recommend a 50:50 mixture of 10% (w/v) pep-
tone with 10% (w/v) Na-glutamate.
Glass ampoules are typically of the dimensions 4mm 50mm, made of high-
quality glass and open at one end (Figure4.1).
4.3.1.1 Preparation of ampoules
1. Place a small swab of cotton wool at the bottom of the tube by pushing it down
the tube with a wooden stick or the handle of an inoculating loop.
2. Prepare labels, written with a pen or printed on copy paper using a laser print-
er or photocopier, and indicating strain information, batch number and date.
Place the label into the ampoules above the cotton wool.
3. Form a plug with non-absorbent cotton wool by rolling it tightly between the
fingers, and place in the top of the tube. Do not push it too far down because
it must be removed before inoculation. Alternatively, a cigarette filter may be
used in place of the cotton wool plug.
4. Place prepared ampoules in a glass beaker or similar vessel and cover with alu-
minium foil prior to autoclaving.
5. Sterilise the ampoules at 121C, 1atm., for 30 minutes.
63
Non-absorbent cotton wool
Glass ampoule
Label
Absorbent cotton wool
Figure4.1 Preparation of ampoules for freeze-drying cultures
4.3.1.2 Preparation of the cultures
1. Grow the strain on the preferred medium (Chapter3). We use a solid medium
at this point to allow us to check for contamination.
2. All steps are to be performed under sterile conditions in a laminar flow cabi-
net. Dispense 2mL of the lyophilisation mixture (e.g. a 50:50 mixture of 10%
peptone and 10% Na-glutamate) into screw-topped vials and autoclave.
3. Take a loopful of the rhizobial strain and place it into the cooled lyophilisa-
tion mixture, replace lid and shake well or vortex. Alternatively, take 1mL of
a broth suspension of rhizobia and mix well into the lyophilisation medium.
4. Check the label before transferring the suspension to the ampoule.
5. Using a sterile Pasteur pipette, transfer 0.1mL of suspension to the ampoule.

We make 10 ampoules per strain, requiring 12mL. It is important not to add
too much fluid to each ampoule as this can lead to problems with removal of
the liquid during lyophilisation. If less than 0.1mL is added, the primary dry-
ing process (below) can be omitted.
6. Push the cotton wool plug halfway down the ampoule, below the nominated
constriction point, to close the vial and protect the culture during the lyophi-
lization process.
4.3.1.3 Primary drying process
1. Turn on the lyophilizer (Figure4.2), introduce the ampoules and follow the
manufacturers instructions. The first drying stage (90% to 95% of dehydra-
tion) usually takes about 90 to 120 minutes in most lyophilizers.
2. Pressure at the end of primary drying should be approximately 13.3 Pascals.
64
Vacuum chamber for primary drying
Centrifuge
Freeze dryer unit with condenser
Digital vacuum pressure gauge
Vacuum pump
Figure4.2 Components of a standard lyophilizer
After the first drying stage, constrict the ampoules with the aid of a burner to draw
out a heavy-walled capillary section on the ampoules.
4.3.1.4 Ampoule constriction
1. Before ampoule constriction, cotton wool plugs should be trimmed and

pushed down the tubes to a position above the label using a sterile stick.
2. Turn on the gas and light the torch. Open the valve to incorporate air into the
flame and adjust both gas and air mixture to produce a small, fine-pointed
flame which allows the heat to be concentrated on a section of the ampoule.
3. Ampoules are then constricted in the middle, above the inserted plug, by gen-
tly turning ampoules while holding the centre point over a flame, allowing the
glass to melt.
4. Ampoules should be held horizontally by the ends, resting on the index and
second fingers. Slowly roll between the thumb and first finger on both hands,
using thumbs on top. Care should be taken at this point as the tendency is for
one hand to roll more quickly than the other. As the glass melts, the two halves
of the ampoule become only weakly connected. Rolling should allow the glass
to flow into the middle of the ampoule at the point where the flame is directed.
As the glass flows into the middle, remove the ampoule from the flame while
continuing to roll and gently stretch the middle section to 11.5cm. It is im-
portant not to stretch the narrow neck too far as the glass walls can become
thin and brittle. Care should be taken not to seal ampoules during this process.
65
5. Mechanical ampoule constrictors (Figure 4.3) provide an easier solution if

available. Ampoule constrictors have angled rollers that apply a gentle out-
wards pressure while the ampoules are turned over a fixed flame.
Figure4.3 Ampoule constrictor

with butane-or propane-fueled
flame and angled rollers to
assist with stretching of tube for
constriction
4.3.1.5 Secondary drying
1. Constricted ampoules are applied to a manifold by pushing over rubber adapt-

ers or into a rubber sleeve (Figure4.4).
2. When the vacuum is re-started, the ampoules will be held tightly, although it
is important that the rubber sleeves have not perished.
3. It is also important that ampoules make good contact with rubber adapters
and that there are no cracks or chips in the glass allowing leaks to occur.
4. Secondary drying should continue for approximately one hour until a vac-
uum of approximately 6.7Pascals is reached. Drying over a desiccant, such
as phosphorus pentoxide, is particularly effective for the removal of any re-
maining moisture but care should be taken not to inhale or touch phosphorus
pentoxide.
5. Once the vacuum has been achieved, ampoules may be sealed by holding a
flame to the constricted part of the ampoule. When melting this narrow neck,
care must be taken to avoid burning a hole in the ampoule, which can occur if
the glass walls have been stretched too thin during the constriction process. If
this happens, there will be a loss of vacuum. The damaged ampoule needs to
be removed and replaced. The vacuum pump is then re-started until a vacuum
of 6.7Pascals is once again reached.
6. The ampoule is held at the end and twisted, after the narrow neck has melted,
until it can be separated from the top half of the ampoule (Figure4.5).
7. Remember to burn off any remaining needles of glass at the point of constric-
tion to avoid injury.
66
Figure4.4 Constricted ampoules attached to a manifold that can evacuate four rows of 12
ampoules (left) and showing the insertion inside rubber sleeves (right panel). In Figure4.5, the
ampoule is placed over a rubber sleeve. Either system works but the rubber can perish over time
and must be periodically replaced.
3 2 1 5
Figure4.5 Description of parts and flow of operations for freeze drying after primary drying:
1. cotton wool, 2. narrow neck, 3. rubber nipple, 4. flame, 5. constriction melted above the cotton
wool, 1 and 6.
67
4.4 Quality control
4.3.1.6 Checking vacuum and survival of dried cells
1. For long-term survival of strains, ampoules need to retain a vacuum.
2. A high-frequency spark tester or generator can be used to check if a vacuum

has been maintained within the sealed ampoule and that there are no leaks.
3. If a vacuum is present, the inside of the ampoule will show a faint glow like a
neon light (Figure4.6).
Figure4.6 Checking for a vacuum with a high-frequency spark tester
4.4 Quality control
Cell viability should be checked in at least one ampoule of each culture per batch
soon after the lyophilization process.
4.4.1 Recovery from lyophilized ampoules

1. In a laminar flow cabinet, score the glass above the cotton plug with a steel file,
heat the score mark, then gently bend the glass allowing it to crack across the
score.
2. Alternatively, and more safely, wet the score mark and touch a softened hot
glass rod across the wet score to crack the ampoule.
3. Remove the fragmented part and the cotton with sterile forceps. Add
0.10.3mL growth medium or lyophilizing mixture with a Pasteur pipette, or
a sterile tip on a micro-pipette.
4. Homogenise by sucking up and expelling through pipette.
5. Transfer a drop or two to a Petri plate containing growth medium.
6. Spread and allow bacteria to grow (Chapter3).
68
4.5Cryopreservation
4.5Cryopreservation
Cryopreservation refers to the biological maintenance of living organisms at low

temperature (80C or below) in a way that allows them to survive after thawing.
Electric freezers allow storage at 80C to 150C. Because there are some reports
that the viability of cultures maintained at 80C decreases over time, the culture
should ideally be checked every five years. For the preservation of bacteria, the
OECD (2007) recommends they be stored at temperatures below 140C.
4.5.1 Preparation
Select an appropriate preservation medium, such as the glycerol-peptone medium
of Gerhardt et al. (1981) modified in Table4.2 to be suitable for rhizobia by re-
moval of a meat extract component.
Table4.2 A glycerol-peptone medium for use with rhizobia

Component Amount/100 mL
Peptone 1.0g
Yeast extract 0.5g
NaCl 0.5g
Glycerol 12.5g
pH adjusted to 7.2
4.5.2Procedure
The following is a basic description of the CRS method for cryopreservation for
rhizobia.
1. Prepare a standard culture medium without indicator or dyes, mix 85:15 with
80% (v/v) glycerol (final concentration=12% glycerol) and autoclave.
2. In a laminar flow cabinet, dispense the culture medium + glycerol mix into
sterile cryotubes (12mL).
3. After confirming purity of the culture to be preserved, take the plates to the
laminar flow hood, add a loopful of culture to the cryotube and vortex to re-
suspend the culture.
4. Alternatively, if the culture is a broth, add 150L of sterile 80% (v/v) glycerol
to the cryotube, make up to 1mL with broth culture and shake to mix.
5. Decrease the temperature at a rate of 1.0C/minute until it reaches 50C and

then leave at the final temperature.
Some laboratories have better cell recoveries by plunging cryotubes into liquid ni-
trogen, allowing a rapid drop of temperature. Glycerol cell suspensions may also
be added to vials containing small sterile glass beads, removing excess liquid so
69
4.6 Developing a parent/working-lot system
that beads are coated with cell suspensions. Coated beads can then be removed
individually from frozen cultures without the need to thaw the whole suspension
before re-culturing.
4.5.3 Recovery of cultures from freezing

Preserved cultures should be checked at least every five years. The CRS uses the
following approach.
1. If frozen cell suspensions are to be returned to cold storage, they should be

kept on ice while being handled and not allowed to thaw.
2. A warm sterile loop is used to transfer a small amount of frozen culture to an

agar plate.
3. Once thawed, the culture is then streaked onto the plate.
To check the purity of the frozen culture, it is advisable to streak to obtain single
colonies (Chapter3).
When rhizobia are being recovered, either from ampoules or from cultures stored
at low temperatures, growth may be much slower than that of cultures streaked
out from routine slopes or plates, and longer incubation times may be needed.
4.6 Developing a parent/working-lot system
With commonly used cultures, genetic drift can be minimised by implementing

a parent/working-lot system (Figure4.7). This allows early cultures to be avail-
able for production of new working stock. The first set of working cultures can be
made from the initial culture. A set of parent cultures is made from the same ini-
tial culture, and these can be used subsequently to make further parent and work-
ing cultures. If the culture collection is to be stored frozen, it is recommended that
parent lots are stored in separate refrigerators to minimize losses from failure of
either the electricity supply or the compressor.
70
4.7References
Initial Working
culture cultures
set 1
Parent Working
cultures cultures
set 1 set 2
Working
cultures
set 3
Parent further
cultures working
set 2 cultures Working
cultures
set 4
Working
further parent cultures cultures
set 5
Figure4.7 A parent/working-lot system for culture preservation
The OECD best practices guidelines for biological resource centers (BRCs) rec-
ommends that bacteria should be kept in at least two different ways, and cop-
ies should be kept in two different places, preferably in two different buildings
(OECD 2007).
4.7References
Day J.G. and Stacey G.N (eds) 2007. Cryopreservation and freeze-drying protocols, 2nd edition.
Methods in molecular biology series 368. Humana Press: Totowa, New Jersey.
Gerhardt P., Murray R.G.E., Costilow R.N., Nester E.W., Wood W.A., Krieg N.R. and Phillips G.B.
(eds) 1981. Manual of methods for general bacteriology. American Society of Microbiology:
Washington.
Hublek Z. 2003. Protectants used in the cryopreservation of microorganisms. Cryobiology 46,
205229.
Hungria M., Loureiro M.F., Mendes I.C., Campo R.J. and Graham P.H. 2005. Inoculant preparation,
production and application. Nitrogen fixation in agriculture, forestry, ecology and the environ-
ment. Nitrogen Fixation: Origins, Applications and Research Progress 4, 223263.
OECD 2007. OECD best practice guidelines for biological resource renters. (Organization for Eco-
nomic Cooperation and Development (OECD): Paris.
71
5.1Introduction
CHAPTER5
Authentication of
rhizobia and assessment
of the legume symbiosis
in controlled plant
growth systems
R.J. Yates, J.G. Howieson, M. Hungria, A. Bala, G.W.
OHara and J. Terpolilli
5.1Introduction
After strains of rhizobia have been isolated from nodules (Chapter3), and (ide-
ally) before long-term preservation (Chapter4), the strains should be examined
to ensure they retain the essential features of nodule bacteria. The first step in this
process is termed authentication, which examines the ability of the strain to in-
fect a legume to form a nodule. Following this, strains may be evaluated for their
ability to fix nitrogen. This latter characteristic is sometimes termed effectiveness;
it is an assessment of the genetic compatibility between the host plant and the
rhizobium strain for nitrogen fixation. If a strain can nodulate a legume and fix
N2 effectively in the glasshouse environment, the researcher may wish to proceed
further, to assessment in the field. However, if the strain is to be released to the
field, then duty of care requires that we have an understanding of its host-range
(Section 5.2). This is because releasing strains into the general environment that
might be detrimental to existing legumes either agricultural or natural would be
negligent. The techniques described in this chapter allow a researcher to compare
strain symbiotic performance across a spectrum of plant genotypes to fulfil this
duty of care.
73
5.2 Host-range, cross nodulation and effectiveness
5.2 Host-range, cross nodulation and effectiveness
The group of legumes with which a strain of rhizobia is able to nodulate is often
described as its host-range. Cross-nodulation within this host-range was a con-
cept developed to record nodulation relationships and thus became a tool with
which to group strains (Fred et al. 1932; Vincent 1970). This has been very useful,
for example, when developing commercial inoculant groups.
In describing host-range it is important to distinguish between specificity for nod-

ulation and effectiveness for N2 fixation. Whenever a strain achieves nodulation
with a legume, the association may have one of several possible outcomes for N2
fixation; varying anywhere from no N2 fixation to maximum N2 fixation (Terpolilli
et al. 2008). Too often researchers ignore the fact that many symbioses are not op-
timal for N2 fixation. This becomes very important when developing commercial
inoculant recommendations, or investigating regulation of genes for N2 fixation.
5.3Plant growth systems with which to assess

nodulation and effectiveness
A wide range of growth systems are available with which to evaluate strain nod-
ulation and effectiveness under controlled conditions (Table 5.1). Selecting the
most appropriate system is a critical decision. It will depend on seed size, the
duration of the experiment and the type of information required. For the larger-
seeded pulses, the legume may need to grow for up to seven or more weeks to ex-
haust cotyledon nitrogen before the expression of the symbiosis becomes evident.
For very large seeds, one cotyledon can be excised after seedling emergence to
expedite the reliance of the legume upon N2 fixation.
A summary of the advantages and disadvantages of a range of apparatus for

screening strains for nodulation or effectiveness is provided in Table5.1.
74
5.3 Plant growth systems with which to assess nodulation and effectiveness
Table5.1 A
pparatus for screening rhizobia for nodulation and effectiveness with a guide to their
advantages and disadvantages
Assembly and growth medium Advantages Disadvantages
Glass test tubes with a vermiculite, Can be sterilised as a whole unit. Not all legumes fix N2 in glass or
sand or agar support medium. Strong control of hygiene. Space polycarbonate tubes.
efficient.
Plastic growth pouches provided Space efficient. Can easily visualise Can be expensive. Less control of
with a liquid medium. root systems. airborne contaminants.
Leonard jars containing sand or Can screen large seeded legumes Time and labour intensive.
vermiculite, with added nutrient (e.g. soybean).
solution.
Enclosed, polycarbonate OHara Easy to set-up. Can be sterilised Not all legumes fix N2 in enclosed
vials. as a unit. Recyclable and space vials.
efficient.
Sand with nutrient solution in Can screen large or small Requires adequate space and a
plastic pots, with surface-applied legumes, with several species per clean glasshouse.
beads. vessel. Sand better reflects field
conditions, allows deep drainage
and is cheap to procure.
Intact soil cores. Can mimic soil physical and Time and labour intensive in core
chemical conditions. collection.
Hydroponic solution Can closely monitor nutrient flows. Difficult to control contamination.
Space efficient and roots are easily Limited rhizosphere development
examined.
5.3.1Growing seedlings in glass tubes for authentication of

rhizobia
For small seeds, such as clovers (Trifolium spp.) and alfalfa (Medicago sativa), it
is possible to conduct authentication tests in glass or polycarbonate test tubes of
200mm 25mm dimensions, with one plant per tube.
5.3.1.1 Procedure for setting up the glass tubes containing agar slants
1. Select the appropriate nutrient solution (see Section 5.7) and add agar powder
at 1520g/L. Adjust the pH to 67, otherwise the agar does not set solidly.
2. Microwave the mixture to melt the agar, stir to homogenise and then dispense
into each tube, to approximately one-third of its volume before it cools and
sets. If a microwave is not available, evenly mix the agar powder into the min-
eral solution by vigorous stirring then dispense evenly to each tube.
3. Allow sufficient tubes for a 10% failure rate, and for both positive (+N) and
negative (-I) controls.
4. Close tubes with individual plastic caps, or by inserting a cotton wool bung
(in this case, cover with a sheet of aluminium foil or moisture-proof paper to
prevent the cotton wool from becoming wet) then place in a rack and sterilise
at 121C for 20minutes.
75
5. Remove from the autoclave before the agar solidifies and place the rack at an
angle of approximately 60degrees to let the agar set.
6. In the laminar flow cabinet, and using sterile tweezers, gently grasp the pre-
germinated seedling around the seed (see Figure5.6) and plant one into each
tube, with the radicle penetrating into the agar. Ensure the radicles do not dry
out and are oriented downwards. NB: sometimes it helps to push the tip of the
tweezers into the agar to make a crack prior to inserting the radicle.
7. Inoculate the seedlings (see Section 5.9) as soon as possible with 1mL of sus-
pension as this also prevents drying. Replace the cap or cotton wool.
8. Place the tubes in a shallow box or cover the base with aluminium foil to pro-
tect the roots from light.
9. Place the tubes in the glasshouse or near a window to allow sufficient light for
photosynthesis.
10. Evaluation of nodulation can be performed 25 to 35 days after inoculation.
11. Authentication is positive if isolates nodulate the roots of inoculated plants

while uninoculated plants remain nodule free.
12. A measure of effectiveness* can be obtained by cutting the tops, drying at 60C
then weighing (NB: see Section 5.12.1 for qualifications).
*NB: some legumes (e.g. Biserrula pelecinus) do not readily fix N2 when enclosed
in glass tubes and therefore caution should be exercised if assessing N2 fixation
with this approach. The reasons for this are unclear.
5.3.2Growing seedlings in OHara vials for authentication of

rhizobia
OHara vials (Figure5.1) are essentially mini-pots. They are screw-topped poly-
carbonate (100 mL) vials that may be autoclaved after the addition of growth
medium (usually sand) and nutrients (Section 5.7) to approximately 50% of ca-
pacity. Germinated seedlings are lowered into the medium with forceps in the
laminar flow cabinet and the lid rested on the top to allow exchange of gases dur-
ing the growth of the seedling. This system is suitable for small-seeded legumes
(120mg) such as siratro (Macroptilium atropurpureum) but larger vials (500mL)
have been used for larger seeded legumes (2050mg). If necessary, sterile water or
nutrient solution can be added to the vials during plant growth.
The advantages of the OHara vials include ease of preparation, cleaning and stor-
age, and they can be recycled. They also represent a closed system that minimises
contamination. As for glass tubes, not all legumes fix N2 to the maximum of their
ability when enclosed in polycarbonate.
76
Figure5.1 Sterilised OHara growth vials containing a sand mixture, with

adhesive labels
5.3.3 Growth pouches for authentication of rhizobia

Growth pouches are made of plastic (polyethylene) bags in which it is possible
to grow legumes of small and medium seed size, such as mungbean (Vigna ra-
diata), Trifolium, Medicago and vetch (Vicia sativa). Although the pouches are not
completely sealed after inoculation, with care, cross-contamination is rare and
they may be manufactured in the laboratory. The pouches have a thickness of ap-
proximately 0.12mm, must be autoclavable and are self-supporting. The average
dimensions of the base and height are of 130mm 140mm, respectively. Ger-
mination paper supports the seedlings in the top of the pouch and also provides
a wick that draws up moisture. The roots grow into the pouch to further access
nutrients. Light entering the root zone is reduced by an inner brown paper lining
and by filing the pouches close together (Figure5.2).
Figure5.2 Plant growth pouches are adequate for observing nodulation and authenticating
strains, but not all legumes will fix N2 optimally in these conditions so alternative growth
assemblies may be needed if N2 fixation is to be quantified
77
5.4 Screening for N2 fixation
5.3.3.1 Procedure for setting up the growth pouches
1. Place pouches into a grid or holder (Figure5.2).
2. Add 80100mL of sterile nutrient solution (Section 5.7) to the pouches.
3. The germination paper is then folded and perforated, and added to each pouch
to support the seedlings.
4. Place one pre-germinated seedling per pouch, with the roots immersed in nu-
trient solution and properly oriented.
5. Inoculate at the base of the seedling after two days.
6. Positive (+N) and negative (I) control pouches should always be included.
7. If necessary, sterile water or nutrient solution can be added to the pouches

during plant growth.
8. The evaluation can be performed between 25 and 35 days after inoculation,

with nodules visible through the bag (Figure5.2).
In our experience, and that of others, diffusion barriers exist in glass tubes where
the plant is supported in either agar or vermiculite, and this can affect N2 fixation.
In extreme cases, N2 fixation is completely suppressed. Important assumptions
governing the screening of rhizobial strains for effectiveness under controlled
conditions are as follows.
1. No plant-available mineral nitrogen enters the system, except that contained

in the seed. This ensures that N2 fixation can be monitored as a direct function
of plant growth. This precludes the necessity for costly analysis of N in the tops
to assess symbiotic performance.
2. The system must be free of rhizobia that are capable of nodulating the legume,
other than that applied as inoculum, implying that hygiene must be main-
tained throughout the duration of the experiment.
3. Control treatments of uninoculated (I) and uninoculated but with mineral N

(+N) must be applied.
4. The +N control treatment must have N available at a non-limiting rate, such

that growth can be compared with the N2-fixing treatments and relative effec-
tiveness of strains can be assessed by comparison. Obtaining the right rate for
the +N control may necessitate a preliminary experiment.
5. Other nutrients required for plant growth must be non-limiting. This may
also require a preliminary experiment as legumes differ in their nutrient
requirements.
78
6. The growth substrate must be non-limiting for plant development (i.e. does
not limit diffusion of gases, light or nutrients, become water-logged, too dry,
salty, anoxic, toxic or compacted).
7. The container must not limit plant growth in any way (e.g. by constricting the
roots).
8. The external environment must not limit plant growth through inappropriate
temperature or day-length, or amount of light.
5.4.1 Leonard jars

Leonard jars are suitable for larger-seeded legumes, such as the tropical pulses
that cannot be grown for long periods in pouches or glass tubes. Leonard jars are
suitable for quantification of N2 fixation but not recommended for competition
experiments as a rhizosphere does not establish in the normal manner. Leonard
jars consist of two separate parts: the base providing a reservoir for nutrient solu-
tion, and the top supporting the plants. Leonard jars can be built from common
commercial bottles (Figure5.3).
Figure5.3 A schematic diagram of a Leonard jar (from the NifTAL manual,

Somasegaran and Hoben 1994).
79
5.4.1.1 Procedures for setting up Leonard jars
1. The top half of the assembly consists of an inverted glass bottle (round beer
or spirit bottle) of about 700mL capacity that has had its bottom removed.
The CIAT manual (SylvesterBradley and Kipe-Nolt 1988) suggests the bot-
tom may be removed by heating at the appropriate place with a wire resistor
then plunging into cool water if necessary.
2. The inverted glass bottle sits on the lower half, which is in effect a glass jar
that holds the reservoir for N-free nutrient solution. The assembly is such that
the inverted bottle sits snugly on the rim of the jar and the neck of the bottle
comes to within 24cm of the bottom of the jar.
3. A wick is placed in the neck of the inverted bottle so that the nutrient solution
in the jar is brought up to the top of the growth vessel by capillarity. The wick
is secured in the bottle neck by cotton wool.
4. Add coarse, well-washed river sand or other substrate (see Section 5.5.1) to
the bottle units to within 2cm of the top. In doing so, keep the wick near the
centre of the vessel and have it reaching almost to the surface.
5. Moisten the jar from the top with N-free nutrient solution (see Section 5.7)
until it begins to drain into the reservoir. Fill the reservoir with the nutrient
solution to within 2cm of the junction of the two parts.
6. Cover the open end of the inverted bottle with aluminium foil (or a glass Petri
dish cover) and the whole unit with moisture-proof paper secured with rubber
bands or heat-resistant tape.
7. Autoclave complete units at 121C for two hours. Do not vent the autoclave af-
ter sterilisation but let it come down to zero pressure before opening the door.
Remove the units to a clean place and keep the covering intact until planting.
8. Sterilise sufficient gravel (34mm diameter size) or alkathene beads to cover

the substrate in the vessels to a depth of 2cm by dry heating in an oven at
200C for one hour.
9. Place the assembly in a laminar flow cabinet or on a clean bench and remove
the foil cover or Petri lid. Using sterile forceps make holes in the substrate into
which pre-germinated seeds are placed. If the treatment is to be inoculated, do
so before covering with sand using forceps and tamp the sand in place over the
seed to provide a firm seed bed.
10. Cover the vessel opening with half a Petri dish or clear plastic film to protect
against contamination, and set the units in a greenhouse or growth chamber.
Remove the covers when the seedlings reach a height of about 2cm and cover
the soil surface with sterile gravel or beads. Inoculation (see Section 5.9) can
be carried out just prior to this step, directly onto the surface of the sand at
the base of the plant, if it has not previously been done. The units are now left
open.
80
11. Uninoculated and mineral N controls are provided as previously described.

The assembly is maintained for five to eight weeks during which N-free nu-
trient solution in the reservoir may be depleted, depending on plant size and
growth conditions. If necessary, new nutrient solution (1:4 dilution) may be
transferred aseptically into the reservoir.
12. At the end of the growth period, the plant shoot is harvested and measured
for dry weight while the nodules are counted and the dry weights of roots and
nodules is measured.
5.4.2 Draining pots with the surface covered by sterile beads

If the researcher has access to coarse sand, vermiculite or other substrate and
a steaming assembly to pasteurise it, then draining pots can be adopted for the
growth of legumes of any seed size. Up to six small-seeded legumes may be grown
in a small plastic pot (5cm radius, 1.5kg sand) which represents an experimen-
tal unit. Larger pots (10cm radius, up to 5kg sand) are required for large-seeded
pulse legumes, such as pea, soybean, lupin and beans, which must be grown for
five or six weeks before seed reserves of nitrogen are exhausted. For these, a sup-
port assembly to keep the plants upright is usually required (Figure5.4).
A feature of this assembly is that more than one species of legume can be grown in
each pot (e.g. Figure5.7) which enables the researcher to assess the host range of
a single rhizobial strain across genotypes, species or genera, in a split-plot design
(Howieson et al. 1995). Hygiene is maximised by a layer of sterile alkathene beads
applied to the surface of the pot, with sterile nutrients added through a capped
watering tube (see Section 5.5.4).
Figure5.4 Free-draining pots

containing coarse sand in a system
for assessing effectiveness. Small
1kg pot (left) growing Siratro and
Acacia saligna, and large 5kg pot
(right) growing both black-eyed
bean and climbing bean.
81
5.4.3 Hydroponic systems

Some researchers have eliminated the substrate as a support structure for growing
legumes, and instead suspend the legumes immediately above hydroponic solu-
tions that provide all the nutrients required for legume growth (except N). Mecha-
nisms to suspend the legumes include foam (which floats on the surface of the
nutrient solution, with holes for the roots to extend through) or lids that fit over
the vessels and that contain slits into which foam sleeves holding the roots are in-
serted (Figure5.5). Contamination by rhizobia can be controlled by altering the
pH, according to the principles described by Munns (1968) and refined by Ewing
and Robson (1990). Briefly, as infection of legumes by rhizobia is pH sensitive,
the solution may be kept at a pH that precludes infection but does not affect root
growth. The pH can be raised to allow infection at the time of inoculation, and
lowered again two days post-inoculation to preclude further infection. Vessels
must be aerated, and this is usually achieved with an aquarium pump.
5.4.3.1Procedure
1. Determine appropriate nutrient solution for growing the legume (Section 5.7).
The CRS uses one-quarter strength CRS solution (Section 5.7.3).
2. Make sufficient concentrated solution for all vessels, ensuring adequate CaSO4
is available (CRS maintain a separate 20 L concentrated (10) solution of
CaSO4 continuously aerated).
3. Adjust pH to below that where infection will occur for the target symbiosis but
that does not damage plant roots. CRS add 1mM MES buffer to provide some
management of pH control that results from plant metabolism (see Chapter3,
Table3.2).
4. Thoroughly cleanse all equipment in 4% (w/v) hypochlorite solution and

rinse. Note that chlorine residues dissipate in full sunshine so residuals are not
damaging if left exposed to the sun for several days.
5. Fill vessels to the required volume with deionised water (DIW) allowing space
for the addition of concentrated nutrients.
6. Insert aeration device (usually an aquaponic aerator with plastic pipes to each
vessel running from a manifold; Figure5.5) and ensure each pot has visible
bubbles.
7. Add nutrients and check pH. Adjust with 0.1N NaOH or HCl solution if buff-
er present or 0.01N solution if not.
8. Germinate seedlings after appropriate measures for ensuring seed viability

and hygiene (Section 5. 8). It is convenient for hydroponic experiments to
germinate seeds in bubbling solution by suspending them on the surface of
the solution on fibreglass wire, down through which the radicles penetrate.
Ensure pH is below that required for infection during this process.
82
9. Several legume cultivars or species can be assessed in a single vessel by having

a different set of plants held in each sleeve (Figure5.5).
10. When roots are 23 cm long gently place seedlings in foam sleeves, insert
sleeves into slits in the lid of the apparatus and suspend over the hydroponic
solution (Figure5.5), ensuring roots are fully immersed.
11. Raise the pH to that required for infection of roots by rhizobia for the par-
ticular symbiosis under study (see Chapter7 for discussion of pH sensitivity
in rhizobia).
12. Inoculate with appropriate rhizobial treatments (Section 5.9). CRS add 10mL
of a 109 suspension per 10L vessel, providing 106 cells per mL.
13. After 48hours reduce the pH to that below which infection can occur, and
monitor daily.
14. Change the nutrient solution every week, taking precautions to ensure pH is
optimised. At this point, it is convenient to have a second set of vessels pre-
pared, and to transfer the lids carrying the seedlings to this set.
15. Maintain the experiment until sufficient growth is achieved to examine the
symbiotic effectiveness of the legumerhizobial association.
NB: pH and calcium are somewhat interchangeable in their effects on nodulation

(Ewing and Robson 1990) so attention must be paid to levels of both.
Figure5.5 An hydroponic assembly for

assessing effectiveness in Medicago spp.
This arrangement has 12 sleeves and two
aeration manifolds per vessel (above). Each
foam sleeve holds five plants (below). Note
the presence of abundant young nodules at
many infection points on the roots.
83
5.5 Substrates for effectiveness experiments
5.5.1 Coarse sand

The key requirement for sand to be used as a medium to support legume growth
is that it is coarse and does not compact or become waterlogged once in the ves-
sel. This is because compaction can affect diffusion of gases and nutrient uptake.
Coarse river sand containing less than 3% clay is often a good choice, if available.
Sand should be free of N or any element that may be toxic to plants. Manganese
and aluminium toxicity has been observed in sands of some origins, and in these
cases it should be washed with hydrochloric acid (HCl) and then rinsed with run-
ning water until it is clear. To complete the process, the sand is then dried in an
oven or in the open air. Good quality coarse sand seldom requires washing and
it is worth the effort to locate a reliable and clean source. The sand, washed and
dried, can be stored in plastic bags. Ensure pH is near neutrality.
5.5.2Vermiculite
Vermiculite is a 2:1 clay mineral with the capacity for expansion and contraction,
conferring high plasticity and stickiness. Depending on the source, vermiculite
may be very alkaline. Therefore it is good advice to wash then soak the vermiculite
for about eight hours in water, then dry at 90100C for 24hours at room tem-
perature for about a week. The vermiculite must be screened with mesh between
1.0mm and 3.0mm dimension. The pH should be checked and if necessary cor-
rected to approximately 6.5.
5.5.3Mixtures
Legumes can be grown in a mixture of sands with differing clay content, or sand
and vermiculite (1:1, v:v), or sand and charcoal (3:1, v:v). It is recommended that
the researcher makes a preliminary assessment of the substrate that most suits
their biological material and needs to ensure that plant growth is optimal.
5.5.4 Procedure for setting up draining pots with coarse sand

1. Prepare substrate, such as a mix of coarse river sand and yellow sand, with pH
adjusted to neutrality. Select a mix of sand that allows free drainage and does
not compact.
2. Steam sand for four hours to remove live cells of rhizobia and transfer to ster-
ile bulk bags.
3. Soak pots of required dimension in 4% hypochlorite solution overnight then

rinse and dry in a clean room.
84
4. Autoclave paper towels and pad the bottom of the pot to completely cover the
drainage holes.
5. Fill pots to the rim with steamed sand mix (1.5 or 5kg) and place in a clean
glasshouse or growth room. The sand settles to leave a small lip of the pot
exposed.
6. Wash twice with boiled water to leach N, allow to drain and cool (this step
may be omitted if soil has very low N).
7. Cover with a plastic sheet to avoid aerosol landing on soil surface.
8. Immediately before planting, wet the soil with a liberal application of sterile
water until it begins to drain.
9. When seeds are germinated (Section 5.8.2) transfer pots for each experimen-
tal treatment to the sterilised laminar flow bench.
10. Make holes of the required number, dimension and depth in the soil surface
around the perimeter of the pot with a sterilised rod.
11. Gently grasp the pre-germinated seed (not the radicle) with sterile forceps and
place into the holes with the radicle facing down (Figure5.6). Ensure there is
excellent root-soil contact to avoid drying. At this stage, it may be necessary to
spray sterile DI water over the seedling to slightly collapse the hole.
12. If the experiment is large, sow only sufficient pots for one rhizobial treatment,
spray the remaining seedlings with sterile water and close the Petri dish to
avoid further drying.
13. Inoculate planted seedlings with 0.5mL of rhizobial suspension (Section 5.9)
then cover the seed with sand using a sterilised rod and bed down with a small
quantity of sterile water.
It is critical that the radicle is not damaged or allowed to dry during this process
14. Cover the pot with clear plastic wrap and place in the glasshouse.
15. Thoroughly sterilise the laminar flow bench and rod before moving to the next
inoculant treatment. Repeat steps 814 for each rhizobial strain.
16. When the legume has emerged and grown 12 cm high, remove the plas-
tic wrap, and insert a sterilised polyvinyl-chloride watering tube (20 mm
150mm) into the centre of the pot, approximately 30mm deep and loosely
cap (Figure5.7).
17. Cover the surface of the pot with sterilised alkathene beads in a layer 1cm
deep, taking care not to bury the cotyledons of small-seeded legumes (Fig-
ure5.7). For large-seeded legumes, the beads can be applied immediately after
step13, eliminating step14.
85
18. Provide water and nutrients as required, including N for the control treatment
(see Section 5.7) through the watering tube.
19. Two weeks after emergence, thin seedlings to the required number by care-
fully removing excess plants with scissors or forceps, sterilising utensils from
one pot to the next. Particular attention should be paid to removing variants.
20. Water sufficiently for pots to begin to drain, or to a measured weight (if pots
become hard to squeeze, then the soil is likely to be too dry). As a guide, 30mL
of water is added to a 1.5 kg pot every second day once the seedlings have
emerged and begun to transpire. Daily inspection is recommended as the
plants become larger, as it is likely that more water will be required, e.g. after
six weeks of growth Phaseolus vulgaris may require 5080mL every second
day.
21. Randomise the pots after inoculation.
Figure5.6 Placing the

pre-germinated seed
gently into holes in the
sand growth medium
86
5.6 Assessment of N2-fixing potential in non-sterile soil cores
Figure5.7 The sterile coarse sand mixture in these pots, topped with alkathene beads, allowed
the expression of differential N2 fixation in eight species of Trifolium by strain WU95 (left) and the
same clover species by strain WSM1325 (right). The photos show all four replicates (front to back)
with four species of Trifolium in each pot, in a split-pot design.
NB: when grown in sand culture, small-seeded pasture legumes and Phaseolus
vulgaris need a pulse of N (5mL per pot of 1% (w/v) KNO3) at the first (and per-
haps second) watering because the growth medium often lacks sufficient N to sus-
tain the legume after seed reserves are exhausted, and before N2 fixation begins.
5.6Assessment of N2-fixing potential in non-sterile

soil cores
After strains have been screened for symbiotic effectiveness under sterile con-
ditions, their performance can be assessed in non-sterile soils in the glasshouse
before final evaluation in the field. At this stage, biological control is lost but the
experimenter is still able to control other environmental influences, such as soil
moisture and temperature. Glasshouse evaluation of the strains in non-sterile
soils takes account of the compounding effects of other factors, such as micro-
bial flora and soil herbicide status, and soil chemical status, such as pH, and can
provide useful information about the symbiotic potential of the test strains under
these conditions. However, collection of the cores is labour-intensive.
87
5.6 Assessment of N2-fixing potential in non-sterile soil cores
5.6.1Collecting intact soil cores

1. Clean and sterilise the corer (Figure5.8) prior to travelling to the field.
2. Select the site and drive the core into the soil to just below the surface by ap-
plying weight to the step and slowly rotating the core.
3. Remove the corer by lifting the whole apparatus out of the soil. With the
plunger, push the core out into a ziplock plastic bag.
4. Place the bags into a cooler box for transfer to the glasshouse.
5. In the glasshouse, place the ziplock bag into an appropriate-sized plastic pot
(Figure5.9)
6. Sow directly into the undisturbed cores utilising seed and inoculant appropri-
ate for the experimental purpose (see Sections 5.8 and 5.9).
7. Acquire data appropriate for the experiment (e.g. top dry weights, nodulation)
(Section 5.10). For example, the effect of herbicide residues on reducing nodu-
lation in clover is also reflected in growth of the plant tops (Figure5.9) so tops
should be dried then weighed.
For intact cores, attention to the field capacity of the soil is vital to ensure the pots
are neither over- nor under-watered. Depending on the fertility of the soil, apply
nutrients as for glasshouse experiments with soil (Section 5.7.3).
Figure5.8 A metal soil corer with cross-bar for stepping on, and showing extracted core
88
5.7 Nutrient media for legume cultivation in glasshouse experiments
Figure5.9 Soil cores in ziplock bags placed into pots with three replicates. Experiment showing
effects of residual herbicides on clover growth, left to right: residues of trisulfuron, pyroxasulfone,
control, chlorsulfuron and clopyralid.
5.7Nutrient media for legume cultivation in

glasshouse experiments
Several different nutrient solutions have been used to provide mineral nutrition to
legumes grown in effectiveness experiments. Although they differ in the amounts
and types of nutrient sources, they all have a similar composition and can be ap-
plied to most legumes. However, over the years the different media have tended
to be applied in different systems. For example, Norris and Date (1976) nutri-
ent formulation works well in aquaponic systems (e.g. Leonard jars) with tropi-
cal legumes, while Jensens medium is suited to small-seeded legumes grown in
agar slopes, and the CRS medium has been successful with Mediterranean and
temperate legumes grown in sand or in hydroponics. It is prudent to conduct a
preliminary experiment to assess the adequacy of the nutrients supplied, and if
necessary amend the constituents to suit the circumstances of the research. For
instance, Phaseolus vulgaris has a higher nutrient requirement than many other
legumes but 300 mL per week of the N-free Broughton and Dilworth medium
(Section 5.7.4) has been successful for growing this legume in a vermiculite sys-
tem (Mwenda 2016). The composition of some of the common nutrient solutions
is given below. Unless otherwise stated, all solutions are prepared and diluted in
boiled DIW
89
5.7.1Norris and Date (1976), e.g. for tropical

legumes grown in Leonard jars
Solution Reagent stock Quantity (g/L)
A KCl 29.8
B K2HPO4 69.6
C MgSO4.7H20 98.6
D Micronutrients
CuSO4.5H2O 0.078
ZnSO4.7H2O 0.22
MnSO4.4H2O 2.03
(NH4)6Mo7024.4H2O 0.01
H3BO3 1.43
E Ferric citrate 1.795
Take 2.5mL each of stock solutions: A, B and C; 0.5mL of D and 1.0mL of E to

prepare a 1L solution. Add 0.344g of CaSO4.2H2O per litre of medium. It does
not dissolve so mix well when dispensing.
5.7.2 Jensens medium (1942), e.g. for growing

small seedlings in agar in test tubes
Reagents Quantity (g/L)
CaHPO4 1.0
K2HPO4 0.2
MgSO4.7H2O 0.2
NaCl 0.2
FeCl3 0.1
Trace element solution*
H3BO3 0.5
Na2MoO4.2H2O 0.05
CuSO4.5H20 0.02
MnSO4.4H2O 0.5
ZnSO4.7H2O 0.05
Add stock trace element solution*(5 mL) and distilled water to make up to
1000mL. Add 1g agar for solid medium.
90
5.7.3CRS Plant Growth Nutrient Solution, e.g. for plants

grown in sand
The advantage of this solution is that it provides adequate calcium with low chlo-
ride content and chelated iron. Make five stock solutions, the fifth being for trace
elements, and a sixth for calcium.
Reagent Quantity (g/L)

1 MgSO4.7H2O 12.3
2 KH2PO4 6.8
3 K2SO4 17.5
4 Fe-EDTA 2.5
5 trace element solution (store at 4C)
H3BO3 0.464
Na2MoO4.2H2O 0.018
ZnSO4.7H2O 0.539
MnSO4.4H2O 0.042
CoSO4.7H2O 0.141
CuSO4.5H2O 0.125
6. CaSO4 agitated solution 2.04
+N control 5 mL weekly of 10 g/L KNO3
5.7.3.1Procedure
1. Combine equal parts of the stock solutions (14) to make up a nutrient con-
centrate of 200mL in a 2L glass screw-capped bottle.
2. Add 0.5mL of trace element solution (5) to this, and make the volume up to
1600mL with de-ionised water, then autoclave.
3. After cooling, add 400mL of autoclaved and well-agitated CaSO4 solution (6)
to make to a final volume of 2L (this limits precipitation of the Ca and pro-
vides adequate Ca without the chloride salt).
4. Add weekly to the sand culture systems for pulse and forage legumes in 20mL
aliquots (for 1.5kg pots) or 40mL for larger pots.
NB: some vigorous legumes respond to bi-weekly applications of CRS nutrients in

their fifth week of growth and onwards.
91
5.8 General procedures for handling seed in preparation for authentication or for an effectiveness experiment
5.7.4Broughton and Dilworth (1970), e.g. for

Phaseolus vulgaris grown in vermiculite
Reagent groups Quantity (g/L)
1. CaCl2.2H2O 294.1
2. KH2PO4 136.1
3. FeC6H5O7.3H20 6.7
MgSO4.7H2O 123.3
K2SO4 87
MnSO4.H2O 0.338
4. H3BO3 0.247
ZnSO4.7H2O 0.288
CuSO4.5H2O 0.10
CoSO4.7H2O 0.056
Na2MoO4.2H2O 0.048
+ N control (KNO3) 5.0 g per L in full strength solution
Prepare stock solutions of reagent groups 14 using warm water to dissolve the
ferric citrate. To make 10L of full-strength solution, add 5mL of each stock to 5L
DI water and mix. Dilute to 10L by adding another 5L of DI water. Adjust pH to
6.66.8 with 1N NaOH.
5.8General procedures for handling seed in

preparation for authentication or for an
effectiveness experiment
Seed preparation to ensure viability and uniformity is very important for effec-
tiveness experiments because seed size and seedling vigour greatly affect plant
development. Ensure fresh seed is acquired and test the level of germination. Be-
cause of seed dormancy mechanisms in some legumes, different species may re-
quire different treatments before their seeds germinate, often refined by trial and
error. Examples are provided below.
5.8.1 Enhancement of germination for impermeable seeds
5.8.1.1Scarification
For species with an impermeable (hard) seed coat:
Place seeds on a sheet of sandpaper in a plastic tray.

Strap a second sheet of sandpaper around a wooden block and rub firmly over
the seeds until they are visibly scratched (test different grades of sand paper to
find the most efficient for the different species of legume). The thickness of the
seed coat can vary 10-fold between legume species, so the degree of rubbing to
92
achieve scarification varies for different species. This must be determined for
the species under examination.
Assess the seeds for imbibition by placing on water agar or wet filter paper for
two hours, and if a high proportion imbibe then they are ready for germina-
tion. Proceed to step1 in General Procedures (Section 5.8.2) for germination.
5.8.1.2Boiling
As an alternative to scarification:
Place seeds in metal or polycarbonate beaker.

Pour boiling water over hard seeds and allow to cool.
Check for imbibition and if a satisfactory proportion imbibe proceed to sur-
face sterilisation and germination (Section 5.8.2).
5.8.1.3 Acid treatment
As a second alternative to disrupting hard seed coats:
Place seeds in a glass or polycarbonate vessel.

Cover seeds with concentrated sulphuric acid and stir (cautiously) for three
to five minutes.
Carefully drain off the bulk of the acid and wash the seeds at least six times
with sterile water.
Leave some of the seeds in the final change of sterile water for two hours to
assess imbibition. If a satisfactory proportion imbibe, these seeds are now sur-
face sterilised. Proceed with germination (Section 5.8.2).
5.8.1.4 Smoke treatment
For some legumes, mainly from Australia and South Africa, germination lev-
els are increased when seeds are exposed to chemicals contained in smoke from
burnt plant residues (Staden et al. 2000). Smoke water containing these chemi-
cals is now available commercially. Soak the seeds in smoke water overnight prior
to germination.
5.8.1.5Vernalisation
Some legumes of Mediterranean and temperate climate origin require exposure to

a period of cold temperature before germination. Place non-scarified seeds into a
cool room set at between 15C for two weeks prior to the experiment.
93
5.8.2General procedures for surface sterilisation and

germination of seeds
1. From a fresh batch of germinable seed, select the appropriate number or
weight required and remove any damaged or irregular variants.
2. Sort seeds to an even size by rejecting excessively small or large seeds and
pre-test the level of germination. Enhance germination by the techniques de-
scribed above if necessary.
3. Remove live cells of rhizobia from the seeds by immersion in 70% ethanol for
1min then wash twice in sterile water. Some labs transfer from ethanol to a
34% sodium hypochlorite solution for 1min followed by six changes in ster-
ile water. NB: some legumes, such as Arachis, are sensitive to this procedure
and the surface sterilisation regime must be diminished.
4. Germinate small seeds (120 mg) on sterilised water agar in Petri dishes
wrapped in aluminium foil to exclude light. After overnight imbibition, invert
the plate to allow downward development of the radicles (Figure5.10). This
assists in the planting step.
5. Some legume seeds carry pathogens below their seed coats and these cannot
be reached by sterilant. For these species, place surface-sterilised seed on a
sterile moist filter in Petri dishes or sterile paper towel, and after germination
sow the seed before the radicals reach 5mm. This minimises fungal and bacte-
rial development.
6. Germinate larger-seeded species, such as soybean or lupin, on sterilised, mois-

tened paper towel in sterilised plastic containers. Ensure there is adequate free
water in the container for seeds to imbibe.
7. When the radicle is fully emerged, plant the seedlings into the target growth
medium or system. Take great care not to damage the radicle while sowing
and ensure that the radicle is projecting downwards and the seed head is just
below the surface then covered by the growing medium.
Seedlings should not be allowed to dry at any stage during

germination or in transfer to the experimental vessel.
8. If a large number of plants are required, and the time to sow them is extensive,
then the seedlings can be regularly sprayed with a fine mist of sterile water
during the sowing period. Similarly, in warm weather the surface of the sub-
strate may need additional moisture.
94
5.9 General methods for inoculation
Figure5.10 An inverted Petri plate with germinated seedlings. At this stage of germination it is
easy to transfer a seedling to the chosen growth medium.
5.9 General methods for inoculation
1. Assemble the full range of root-nodule bacteria to be tested and ensure they
are equally fresh and well grown. If broth cultures, keep agitated or on ice.
2. If inocula are agar-grown cultures, in the laminar flow cabinet add 0.51mL
sterile 1% (w/v) sucrose solution or DI water to agar plate or slope and scrape
the colonies off with a flamed spatula to produce a suspension. Phosphate
buffer or saline are not used as they can affect plant growth.
3. Transfer the suspension to a sealable bottle, close the lid, label and shake vig-
orously or vortex to ensure uniformity. Keep on ice. The volume of the bot-
tle can be adjusted to between 10 and 100mL (according to how many seeds
are to be inoculated) with either DI water or 1% (w/v) sucrose solution. This
will dilute the inocula, but each mL should contain between 106 and 108 cells,
which is sufficient for nodulation. If an accurate count of cell number is re-
quired, see Chapter6.
4. With a sterile syringe or pipette tip place 0.51mL of the shaken suspension
directly over the seedlings (Figure5.11). Inoculate every replicate of the par-
ticular rhizobium treatment.
5. Change the syringe or pipette tip between each rhizobium treatment and take
care not to suck inoculant into the body of the pipette.
6. After completion of inoculation, pipette 0.5mL of sterile 1% (w/v) sucrose or

DI water to the seedlings of the I and +N controls. If these controls are nodu-
lated at harvest then the microbiological procedures have been defective and
the results are invalid.
Inoculant can also be supplied as liquid broth culture (Chapter3, Section 3.1).
The advantage of liquid culture is that strains can be produced in a more uniform
95
5.10 Growing conditions and facilities
concentration, adjusted by optical density, for example to OD0.1. The disadvan-

tage is that contaminants are more difficult to detect as they are not as visible as
when they are growing on nutrient agar plates.
Figure5.11 Inoculation of the

planted seed using a disposable tip
attached to a pipette gun
5.10 Growing conditions and facilities
Typical facilities for housing the growth vessels include:
1. open glasshouses where contamination is managed at the individual container

level, such as described in Section 5.5.4
2. temperature-controlled glasshouses and phytotrons, where opportunities for

contamination are lessened, but control of contamination must remain at the
container level
3. closed growth chambers that permit little air exchange and maximum control
of hygiene.
In facilities where the temperature can be regulated, optimal temperature settings

for temperate legume species range from 1525C and for tropical legume species
from 2530C. Nodulation and N2 fixation can be markedly depressed in some
symbioses by temperatures higher than 28C, hence a glasshouse must have an
appropriate cooling system, such as evaporative air blowing through it, or a refrig-
erated air recycling system. Painting of windows with water-soluble whitener and
use of shading (e.g. shade cloth) reduces the problem of high temperature during
the summer and allows enough brightness for plant growth. On the other hand,
in temperate regions in the winter it may be necessary to heat the glasshouse and
supply extra lighting to grow subtropical species.
96
5.11 Data acquisition for quantification of effectiveness
A closed growth chamber often provides a superior facility for controlled condi-
tions. It is an aseptic small room with automatic temperature control that can
supply consistent brightness from 100 to 1,000 lumens, with a timer to control the
photoperiod (corresponding to the legume requirements) and automatic mainte-
nance of humidity to about 70%.
5.10.1 Hygiene and preparation

Cleanliness of the facility is crucial to the success of the experiment and a high
level of hygiene must be practised at all times. A clean glasshouse floor and bench
are essential; the facility must be free of excessive dust, old potting mixture, soil
and other debris. Fixed structures should be regularly washed with bleach (4%
NaClO) or ethanol (70%, v/v). The bench top is ideally constructed of a wire mesh
to allow free-draining systems to drain excess solution onto the floor and avoid
contamination from pot to pot through the base.
All equipment, such as test tubes, vials, jars, pots, tubes, lids, alkathene beads, fil-
ter paper, syringes and containers must be cleansed and sterilised or disposed of
between experiments. Common practices include soaking equipment in ethanol
or bleach followed by rinsing in sterile water or autoclaving where appropriate.
5.11Data acquisition for quantification of

effectiveness
5.11.1 Top dry weight

In most scenarios, if the I control plants have characteristic symptoms of N-
deficiency (viz. small plants with pale green or yellow leaves) and are not nodulat-
ed, and the inoculated plants are two or three times as large, have formed nodules
and produced healthy green leaves, it can be assumed that the symbiosis is effec-
tive. However, this must be quantified.
The shoot dry weight of plants harvested after significant plant biomass accumu-
lation is an accepted criterion for N2-fixing effectiveness in systems that are free
of mineral nitrogen. After four to 10 weeks (depending upon growth conditions)
tops are cut at the pot surface, or at the hypocotyl, placed in a paper bag, labelled
then dried (6070C) for three to four days then weighed.
If I controls are not nodulated, the capacity for N2 fixation can be assessed by
comparing yields of inoculated plants with the +N controls as well as with the
commercial inoculant strains. Strains can be ranked by comparing yield as a per-
centage of that achieved by the + N treatment or by the best strain or by the com-
mercial strain, as required.
97
This approach requires some qualifications, however. Firstly, when working with
seeds from wild plants, or even outcrossing plants such as lucerne, there can be
substantial plant to plant variation in top growth even though the seeds have been
sorted for uniformity. Further, many woody legumes put more resources into
their roots than into their tops. The researcher must be aware of these anomalies
and should determine an appropriate parameter to suit their circumstances.
Researchers can consider the symbiosis effective (E) where plant weight exceeds
75% of the +N treatment. Where plant weight is less than 20% of the +N treat-
ment, the symbiosis is deemed ineffective (I). Between these parameters, the sym-
biosis is considered partially effective (P). However, one must be aware that the
+N controls must be at an acceptable level of growth to make such comparisons.
In some species it is difficult to estimate adequate rates of mineral N for good
growth and a preliminary experiment is sometimes required.
5.11.2 Root nodulation

It is often a very valuable exercise to look at the nodules and assess their size, po-
sition on the root system, colour and shape. Carefully remove the roots from the
pots, wash them free of the growth medium and score them for nodulation based
on the number, size and position of nodules on the root system (Figure 5.12).
Plants collected from field experiments can also be scored with this chart but it
works best where there are less than 50 nodules per root system. Once the re-
searcher has become familiar with the root systems in the experiment, the scoring
process can be much more time efficient than counting the number of nodules,
and 20 to 30 plants can be scored quickly to provide an average score per treat-
ment. This approach works well for agricultural legumes but many woody root
systems form very few nodules, and hence it may be better for the researcher to
simply count the number present on woody plants and shrubs. Nodule dry weight
may also be assessed but, as for nodule scoring, it is not a reliable indicator of
strain effectiveness.
For detailed descriptions of nodule morphology (see Chapter2, Figure2.2) it is

valuable to have a camera and lighting system set-up to capture images of the
nodules.
The researcher may also wish to assess nodule occupancy as a final part of adher-
ing to Kochs postulates. This is commonly achieved with molecular tools such as
RAPD or RFLP PCR techniques (Chapter11).
98
0 0.5 1 2 3 4
Absent Ineffective Rare (Effective) Scarce Moderate Adequate
No nodules White ineffective 5 small < 10 small 1020 small > 20 small
nodules pink nodules pink nodules pink nodules and/or large
pink nodules
Figure5.12 A nodule scoring chart than can be applied to both pulse and pasture legumes where
the root systems have less than 50 nodules per plant
5.11.3 Acetylene reduction by nitrogenase

Reduction of acetylene to ethylene can provide an assay for the activity of the
rhizobial enzyme nitrogenase at a point in time. It is particularly useful to assess
whether a strain is genetically compatible for N2 fixation with a given legume host.
In the nodule environment it is the host that protects nitrogenase from denatura-
tion in the presence of oxygen through production of leghemoglobin (Section
5.11.4 and Figure5.14). Acetylene reduction is not particularly useful for quantifi-
cation of N2 fixation, although if handling procedures are optimised it can provide
relative quantification data for pot experiments.
The protocol below is used at the CRS and is a simplified routine because it uses
modern columns that allow a more clear separation of acetylene and ethylene
peaks than when the technique was first developed in the 1960s (Dilworth 1966).
SAFETY INFORMATION
Since explosive gases and high-pressure cylinders are used in this protocol, you
must comply with all local safety rules before commencing. You must also ensure
that your experimental set-up and the location of gas cylinders has been inspected
and approved by the relevant health and safety authority in your institution.
1. Grow plants in a system where they can be harvested with minimal root dis-
turbance. Ideally, a system using fine vermiculite is best as this can be easily
removed from roots at time of harvest without washing the root extensively.
99
2. Once harvested, transfer intact plants into an appropriately-sized bottle

that can comfortably contain the plant without damage. For approximate-
ly four-week-old pea plants, 250 mL Duran bottles are of an adequate size
(Figure5.13).
3. Place a moist piece of paper towel at the bottom of the bottle to ensure that
plants do not dry out during the assay.
4. Seal the Duran bottle with rings (GL-45, 34mm aperture size, fit all standard
Duran tops, catalogue number 292271007) and seals (Silicone rubber seals
VMQ for piercing GL-45 cat # 292461002).
5. Within one hour add a total of 2% acetylene gas to each bottle using a timed
staggered-start method.
NB: acetylene should be taken from the gas storage cylinder in a way that ad-
equately vents the gas into a fume cupboard. One way is to have a piece of sili-
cone tubing attached to the cylinder regulator which then vents directly into
the fume cupboard. The unattached end of the silicone tubing is placed into a
1L plastic beaker which is approximately half-filled with water.
6. Procedure for adding 2% acetylene to a 250mL Duran bottle.
The total volume of a standard 250mL Duran bottle (i.e. when filled to the
brim) is 320mL, which is the value from which the 2% calculation is taken.
To ascertain the total volume, fill the bottle to the brim with water and
measure the volume of the water by transferring it to a measuring cylinder.
If you use a bottle with a different volume, all subsequent calculations will
need to be scaled accordingly.
It is important to check that the integrator on your gas chromatograph

(GC) can measure the acetylene and ethylene peaks.
The procedure is as follows:
a. open the acetylene cylinder lid fully allowing the acetylene to bubble
through the water in the 1L beaker
b. using a 22G syringe, remove 8mL air from the bottle
c. take 6.4 mL of acetylene from silicone tubing and add it to the bottle
(Figure5.13)
d. the difference in the volume of air removed and acetylene added creates a
small amount of negative pressure on the bottle.
7. Incubate the bottles for a known amount of time at a known, constant tem-
perature which is usually 2025C. When running the assay for the first time,
it is important to perform a time course experiment to ensure that you are ac-
cumulating ethylene in the sample bottles and that the rate of accumulation
is linear. To begin with, you could sample at 30minute, 60minute, 90minute
100
and 120minute intervals. Sampling beyond 120minutes is not recommend-

ed as longer exposure of nitrogenase to acetylene ultimately inactivates the
enzyme.
8. At each time point, take 1 mL samples from bottles using a 12.7 mm 29 G

(0.33 mm) syringe, which is the correct size to adequately transfer samples
into a GC. Once sampled, the syringe can be placed into a labelled rubber
bung to prevent any loss of gas.
9. Once all bottles have been sampled they can then be run on a GC.
Figure5.13 Whole plants stored in Duran bottles (left) and acetylene being added to start the
assay (right)
5.11.3.1 Protocol for gas chromatography
The precise step by step protocol for separating and quantifying ethylene and
acetylene in gas samples will vary widely between GCs and operating systems.
However, a set of parameters that work well to separate these gases on a Varian
CP-3800 gas chromatograph with a Flame Ionisation Detector (FID) are given
below.
A Restek RT-U-BOND 15m 0.32mm internal diameter 10m column

(catalogue number 19751).
Oven temperature at 40C.

Carrier gas is H2 set at a constant flow rate of 1mL/minute.
Compressed air supply for FID.
FID set at 190C with a supply of 30mL/minute H2 and 300mL/minute air.
1mL samples are injected with a 5:1 split ratio (i.e. 20% of the sample is loaded
onto the column).
Program runs for 48sec.
101
These parameters allow for clear peak resolution of ethylene and acetylene peaks,
with ethylene and acetylene retention times of 24sec and 32sec respectively.
NB: other columns could also be used to separate ethylene and acetylene peaks,
such as the HaySep N range of columns available from Agilent, Restek or Perkin
Elmer. Check with a technician to ensure your column and operating parameters
are adequate to clearly separate these two gases.
5.11.3.2 Data handling
To calculate the rate of acetylene reduction as amount of ethylene produced/plant/

unit time, the following information is required:
1. the amount of acetylene present in the bottle at the start of the assay (A0)
2. the ratio of the ethylene peak area to the total peak area (ER)
3. time of incubation with acetylene (t).
With this information, rate of acetylene reduction (AR) can then be calculated as
follows:
Below, we give an example of how to derive the rate of acetylene reduction using
the information given in the above protocol, where 6.4mL (0.0064L) of acetylene
was added to a 250mL Duran (of total volume 320mL) at a standard pressure of
1atm and a temperature of 293.15K (i.e. 20C). This approach can be easily modi-
fied by substituting in alternative values for temperature and volume of acetylene,
as appropriate.
1. Determining the amount of acetylene at the start of the assay (A0)
To calculate the initial amount of acetylene added at the start of the assay you
can use the ideal gas law which states:
PV=nRT
Where P = Pressure, V = volume (L), n = moles, R = gas constant defined

as 0.0821atm.Lmol1K1 and T= Temperature (in Kelvin). In this instance,
P=1atm and V=0.016L, so rearranging for n and substituting in values:
n=2.66 104moles
n=266moles
Shows that A0=266moles of acetylene
102
2. Determining the ratio of the ethylene peak area to the total peak area (ER)
Derive the ratio of the ethylene peak area to the total peak area (ER) using the
data obtained from the GC chromatogram in the following way.
a. Firstly, integrate the area under the ethylene and acetylene peaks. This can
be expressed in a number of ways, either area under the curve or percent-
age area, and should be done using a GC analysis software package specific
for your GC (such as the Varian MS Data Review software package).
b. Sum the area under the ethylene and acetylene curves to yield a value for
the total area
c. Calculate the ratio of the ethylene peak area to the total area to yield ER.
So, if the percentage area for:
ethylene=0.4508 units
acetylene=16.0036 units
total area=16.4544 units
Then the ratio of the ethylene peak to total area (ER) would be:
ER=0.0274
3. Time of incubation
Determining the time of incubation was discussed above. We have found that
for peas, incubation at 20C for one hour generates measurable quantities of
ethylene and that accumulation is linear.
Using these values derived for A0, ER and t, the rate of acetylene reduction in
this case would be:
AR=7.29moles/hour/plant
5.11.4 Estimation of leghemoglobin in legume root nodules

Many a field researcher has sliced open a nodule to seek evidence that it contains
leghemoglobin (Figure5.14) and indeed the leghemoglobin content of nodules
is correlated with the ability of the nodule to fix N2 (Riley and Dilworth 1985).
Sometimes researchers like to quantify the amount of leghaemoglobin in nodules
and this is best expressed per unit of nodule or bacteroid tissue.
103
Figure5.14 A cross section through multiple lupinoid nodules on the root of Lupinus
angustifolius showing strong leghemoglobin content
Leghemoglobin can be most conveniently measured by the pyridine hemochro-

mogen assay (Appleby and Bergersen 1980; Dilworth 1980). The methods de-
scribed below are used in the CRS for extraction of leghemoglobin from nodules
(Dilworth 1980; Riley and Dilworth 1985; Melino et al. 2012), and estimation of
leghemoglobin concentration (Dilworth 1980; Bergersen et al. 1973; Bisseling et
al. 1978).
5.11.4.1 Extraction of leghemoglobin from nodules
1. Mix fresh root nodules (usually 100200g) with about four volumes of cold
(4C) 100mM sodium phosphate buffer (6.8 pH).
2. Macerate in a mortar.
3. For extraction from soybean and lupin nodules, the addition of polyvinylpyr-
rolidone in a ratio of 0.3g per g nodules is recommended to remove polyphe-
nols (Dilworth 1980).
4. For extraction from faba bean (Vicia faba) nodules, the addition of 2.5mM
EDTA is recommended to inhibit polyphenol oxidase activity (Dilworth 1980).
5. Filter the macerate through two layers of muslin (miracloth or cheesecloth).

This step is useful to remove a lot of plant debris when extracting from large
nodules (e.g. lupin, soybean, pea, faba bean) but may not be necessary when
extracting from small nodules (e.g. clovers, medics, serradella).
104
6. Centrifuge the macerate at 100g for 15minutes to pellet plant debris.
7. Discard the nodule debris and collect the supernatant then centrifuge at
20,000 g for 20 minutes at 5C to remove bacteroids and other suspended
material.
8. Measure the heme concentration of the resulting supernatant by the hemo-

chrome method as described below.
5.11.4.2 Measurement of leghemoglobin concentration
1. To a volume of the supernatant add an equal volume of 4.2 M pyridine in

0.2M NaOH and mix. NB: Pyridine is very toxic
2. The resulting mixture is divided between two 1mL cuvettes.
3. The hemochrome in one cuvette sample is reduced by adding a few crystals of

sodium dithionite (Na2S2O4). Very little dithionite is needed.
4. The hemochrome in the other cuvette sample is oxidised by adding a few crystals
of potassium hexacyanoferrate (III) aka. potassium ferricyanide K3[Fe(CN)6 ].
Very little is needed. Alternatively, add 0.01mL of 3M K3[Fe(CN)6 ].
5. The differential absorbance at 556nm (A556nm ) and 539nm (A539nm ) is de-

termined and the heme concentration (mM) is calculated using a millimolar
extinction coefficient of 23.4.
Leghemoglobin concentration (mM) of the supernatant is calculated using the

following formula:
5.11.4.3 Purified preparations of leghemoglobin
Specific methods are described in Dilworth (1980) for isolation of purified leghe-
moglobin from soybean, broad bean, kidney bean and yellow lupin. These prepa-
rations can be assayed without needing to use an oxidised sample. Simply, equal
volumes of 4.2M pyridine in 0.2MNaOH and purified leghemoglobin solution
are mixed and the resulting hemochrome reduced by adding a few crystals of so-
dium dithionite (Na2S2O4).
Absorbance at 556nm is measured against a reagent blank, the heme concentra-

tion (mM) is calculated using a millimolar extinction coefficient of 34.6:
105
5.12 Outcomes from authentication and effectiveness experiments
5.12Outcomes from authentication and

effectiveness experiments
There can be a range of outcomes in any authentication experiment. There may be

specificity exhibited for nodulation or for effectiveness and this may be expressed
within a legume species (i.e. between cultivars) between legume species or between
legume genera.
5.12.1 Specificity between cultivars within a legume species

A trait often sought in strains for commercial application in agriculture is an abil-
ity to fix N2 optimally with a range of cultivars of the target host legume spe-
cies. For example, USDA110 was considered broadly effective across a wide range
of soybean breeding lines and cultivars and was sown widely on a number of
continents (Hungria et al. 2006). However, strains like USDA110 are the excep-
tion rather than the rule. Clover strain TA1 was used commercially in Australia
for many years but did not nodulate clover cv. Woogenellup effectively (Gibson
1968). We must be aware that legume cultivars within species may differ in their
reaction to rhizobial strains. This is clearly illustrated below; in the early phase of
the domestication of Biserrula pelecinus (Howieson et al. 1995) it was discovered
that some strains of mesorhizobia could fix N2 with the white flowered form, but
not with the blue flowered form of this legume (Figure5.15).
Figure5.15 Specificity for N2 fixation within a legume species. A contaminant seed of the white
flowered form of Biserrula pelecinus (on the right of the pen in this row of plants) has nodulated
and grown well when inoculated with a single strain of Mesorhizobium ciceri bv biserrulae whereas
the many seedlings of the blue flowered form (all other plants) are small and N-deficient, indicating
lack of effective nodulation.
106
5.13References
5.12.2 Broad specificity across legume species within a genus

Some legume genera are widely exploited commercially, and the clovers and med-
ics are good examples. Valuable rhizobial strains for these genera are those that
fix optimally across the wide range of the species that have been exploited in agri-
culture. Figure5.7 illustrates a clover nodulating strain (WSM1325) that is highly
effective for N2 fixation across a wide range of species of Trifolium. It is compared
with a former inoculant quality strain (WU95) that was adequate in an era prior
to the wider domestication of the Trifolium spp. in Australia (Howieson et al.
2000).
5.12.3 Broad specificity across legume genera

Researchers and manufacturers may also require strains that are effective across a
range of legume genera. A relevant example here is that Lens, Vicia, Lathyrus and
Pisum all nodulate with Rhizobium leguminosarum bv viceae and manufacturers
of inoculants prefer to produce one strain to suit all these genera wherever pos-
sible. Such strains are rare but do exist. Similarly, certain strains of bradyrhizobia
may effectively nodulate Cajanus, Lablab and Macrotyloma. Selecting inoculant
strains with a broad effective host-range, which thus reduces the cost of manufac-
ture, is always an important consideration in any selection program.
5.13References
Appleby C.A. and Bergersen F.J. 1980 Preparation and experimental use of leghemoglobin. Pp
315335 in Methods for evaluating biological nitrogen fixation, ed. by F.J. Bergersen. Wiley-
Chichester.
Bergersen F.J., Turner G.L. and Appleby C.A. 1973. Studies of the physiological role of leghaemo-
globin in soybean root nodules. Biochimica et Biophysica Acta 292, 271282.
Bisseling T., Van Den Bos R.C. and Van Kammen A. 1978. The effect of ammonium nitrate on the
synthesis of nitrogenase and the concentration of leghemoglobin in pea root nodules induced
byRhizobium leguminosarum. Biochimica et Biophysica Acta 539, 111.
Broughton W.J. and Dilworth M.J. 1970. Methods in legume-rhizobium technology: plant nutri-
ent solutions. Pp 245249 in Handbook for rhizobia, ed. by P. Somasegaran and H.J Hoben.
Springer-Verlag: New York.
Dilworth M.J. 1966. Acetylene reduction by nitrogen-fixing preparations from Clostridium pasteu-
rianum. Biochimica et Biophysica Acta 127, 28594.
Dilworth M.J. 1980. Leghemoglobins. Methods in Enzymology 69, 812823.
Ewing M.A. and Robson A.D. 1990. The effect of solution pH and external calcium concentration
on the early growth and nodulation of several annual Medicago species. Australian Journal of
Agricultural Research 41, 933939.
Fred E.B., Baldwin I.L. and McCoy E. 1932. Root nodule bacteria and leguminous plants. University
of Wisconsin Press: Madison, WI.
Gibson A.H. 1968. Nodulation failure in Trifolium subterraneum L. cv. Woogenellup (Sum. Marrar).
Australian Journal of Agricultural Research 19, 907918.
Howieson J.G., Loi A. and Carr S.J. 1995. Biserrula pelecinus L.a legume pasture species with
potential for acid, duplex soils which is nodulated by unique root-nodule bacteria. Australian
Journal of Agricultural Research 46, 9971009.
107
5.13References
Hungria M., Franchini J.C., Campo R.J., Crispino C.C., Morales J.Z., Sibaldelli R.N.R., Mendes I.C.
and Arihara J. 2006. Nitrogen nutrition of soybean in Brazil: contributions of biological N2 fixa-
tion and N fertilizer to grain yield. Canadian Journal of Plant Science 86, 927939.
Jensen H.L. 1942 Nitrogen fixation in leguminous plants. I. General characteristics of root nodule
bacteria isolated from species of Medicago and Trifolium in Australia. Proceedings of the Lin-
nean Society NSW 67, 98108.
Munns D.N. 1968. Nodulation of Medicago sativa in nutrient solution. I Acid sensitive steps. Plant
and Soil 28, 129146.
Norris D.O. and Date R.A. 1976. Legume Bacteriolgy. Pp 13474 in Tropical Pasture Research
Principles and Methods, ed. by N.H. Shaw and W.W. Bryan. Commonwealth Bureau Pastures &
Field Crops Bulletin No.51. Tropical Pasture ResearchPrinciples and Methods.
Melino V.J., Drew E.A., Ballard R.A., Reeve W.G., Thomson G., White R.G. and OHara G.W. 2012
Identifying abnormalities in symbiotic development between Trifolium spp. and Rhizobium le-
guminosarum bv. trifolii leading to suboptimal and ineffective nodule phenotypes. Annals of
Botany 110, 15591572.
Mwenda G. 2016. Diversity, effectiveness and competitiveness of Phaseolus vulgaris nodulating bac-
teria from Kenya. PhD thesis, Murdoch University.
Riley I.T. and Dilworth M.J. 1985. Cobalt requirement for nodule development and function in Lu-
pinus angustifolius L. New Phytologist 100, 347359.
Staden J.V., Brown N.A.C., Jger A.K. and Johnson A. 2000. Smoke as a germination cue. Plant Spe-
cies Biology 15, 167178.
SylvesterBradley R. and Kipe-Nolt J. 1988. LegumeRhizobium Symbiosis. International Center
for Tropical Agriculture (CIAT): Colombia.
Terpolilli J., OHara G.W., Tiwari R.P., Dilworth M.J. and Howieson J.G. 2008. The model legume
Medicago truncatula A17 is poorly matched for N2 fixation with the sequenced microsymbiont
Sinorhizobium meliloti 1021. New Phytologist 179, 6266.
Vincent J.M. 1970. A manual for the practical study of root-nodule bacteria. IBP handbook 15.
Blackwell Scientific Publications: Oxford.
108
6.1Introduction
CHAPTER6
Counting rhizobia
G.W. OHara, M. Hungria, P. Woomer and J.G. Howieson
6.1Introduction
The enumeration of rhizobia is valuable for the assessment of rhizobial populations

in soil and how they vary, to follow the growth of cultures in the laboratory or to
assess the number and viability of rhizobia in commercial inoculants for quality
control. The number of rhizobia in soil is dynamic and varies within and between
seasons, so any enumeration must be set in context. Enumeration of rhizobia is la-
bour and resource intensive and should only be undertaken when the information
is vital.
There are few new techniques with which to quantify rhizobia. Despite the mo-
lecular and genomic eras, there has emerged no robust DNA-based technology
to replace serial dilutions and direct or indirect quantification. The qPCR and
MISEQ methodologies cannot currently differentiate rhizobial strains reliably nor
distinguish sufficiently between DNA from live cells and DNA from dead cells to
make them effective substitutes for the standard viable counting techniques dis-
cussed below, although this may change. This chapter, therefore, reiterates appro-
priate and well-utilised approaches to enumeration of rhizobia in soil, inoculants
and in vitro.
6.2 Serial dilution
The principle in counting bacteria by dilution is to serially dilute them to reduce

the bacterial density to the level where individual cells can be differentiated. This
may be, for example, as live cells under the microscope, as colonies that grow on
plates from single cells, or estimated in the plant-infection technique (with the
principle that a single cell can multiply to initiate an infection). Serial dilution can
be applied to liquids, solids or soils. A 10-fold serial dilution is most often used
109
6.2 Serial dilution
(Figure6.1) but if the number of rhizobia is expected to be low then a five-fold

dilution can be adopted.
For dilution of soils or solids (such as a clay inoculant) a large mass (e.g. 10g) is
usually selected for initial dilution as quite often this will represent a subsample
from a larger source. Diluents must be sterile and can be high-purity water, rhizo-
bial growth medium (Chapter 3) or a more osmotically-friendly solution such
as 0.85% (w/v) NaCl or dilute phosphate buffer (e.g. Chapter13, Section 13.4) as
required.
6.2.1 Procedure for serial dilution

1. Autoclave chosen diluent in bulk (e.g. 2L in a 5L bottle) and allow to cool.
2. Aseptically dispense 90mL diluent into a pre-sterilised 250mL flask or bottle

and five to eight lots of 9mL in a 30mL McCartney (pre-sterilised) for each
sample that is to be enumerated*.
3. Place a 10g subsample or 10mL of rhizobial suspension into the 90mL to

form a 101 dilution and shake for five to 30 minutes on a shaker. Glass beads
can be added to the vial or flask to help the release of rhizobia from the solid
matrix.
4. With a sterile pipette or micro-pipette with a sterile tip, transfer 1.0mL of the
101 dilution to a bottle containing 9.0mL of diluent, forming the 102 dilution.
Shake well or use a vortex mixer for 20sec to mix the suspension. It is some-
times convenient to cut off the tip of the pipette for this initial dilution if solids
are present that might block the opening.
5. With a new pipette or pipette tip, transfer 1.0mL of the 102 dilution to another
bottle containing 9.0mL of diluent, to form a 103 dilution, and so on.
6. For soils, it is rare to encounter populations of rhizobia >107 cells/g. For in-
oculants, this concentration may be 109 cells/g. If counting a broth culture, a
slightly turbid suspension contains approximately 107 cells/mL.
7. At the completion of the dilution, and starting with the most dilute sample,
apply the dilutions to the medium that provides enumeration. If the dilutions
contain only rhizobia, they can be directly plate counted (see Section 6.3 and
Figure6.2). For counting non-sterile systems, the diluents may be inoculated
onto a seedling growing on an agar slant, OHara vial or growth pouch (see
Chapter 5 and Figure 6.1). This method of enumeration is called the MPN
(Most Probable Number) count.
*Some laboratories use smaller volumes of sample and diluent, especially for labo-
ratory cultures. A 10-fold decrease will be obtained by using Eppendorf tubes
containing 0.9mL of diluents to which 0.1mL samples are added. However, the
procedure should be verified before adoption, as micro-pipettes must be checked
regularly for accuracy. The step of mixing the samples is even more critical under
110
6.3 Plate counts of rhizobia in sterile diluent
those circumstances and it is also recommended that the sterile diluent should be
distributed into sterile Eppendorf tubes under the hood, as any evaporation will
significantly affect the dilution.
Carrier suspended in sterile diluent
Prepare dilution series to 106 Inoculate 2 plants

from each of the
105and 106dilutions
and check for nodules
to confirm colonies
are rhizobia on the
102 103 104 105 106 corresponding plates
Spread 0.1 mL on the surface of duplicate CRYMA

plates and count colonies after growth taking note
of dilutions with contamination
Figure6.1 Scheme of a serial dilution series as might be applied to an inoculant carrier or

to inoculated seed (see Chapters 8 and 9). A sample of the carrier (e.g. peat, granule or seed) is
shaken in a bulk diluent which represents the first level of dilution. This is then serially diluted
with a sample at each level of dilution directly plated or inoculated onto the legume to achieve
enumeration via plant infection.
The counting of microorganisms on plates, following dilution, is also called direct

counting. Two main methods can be used: the drop-plate method and the spread-
plate method. For rhizobia, the usual culture medium for counting viable cells
is the traditional YMA (Chapter3) with 1520mL of medium per plate. Other
media can be used, e.g. containing specific antibiotics to suppress contaminants,
Congo Red to observe contaminants, or LA to reduce gum production (Chap-
ter3). The latter may be important when colonies overgrow and thus obscure each
other in lower dilution plates.
6.3.1 Spread-plate counting

1. Inoculate 0.1mL of the serial dilutions desired (Section 6.2.1) (e.g. 105, 106
and 107) on the surface of the culture medium in Petri dishes (Figure6.1).
111
2. Spread the 0.1mL aliquot over the culture medium with a sterilised L-shaped
glass spreader (or equivalent, e.g. a Drigalski loop). There should be at least
three separate replicate plates for each dilution.
3. After inoculation and absorption of the inoculum into the agar, the plates are
placed in an incubator at approximately 28C, inverted and allowed to grow
for a period of two to eight days (according to the growth rate of the species;
see Chapter3).
4. Count the number of colonies on plates where colonies are well separated. If
colony numbers are low, variation between plates and errors may be large. If
colony numbers are too high, overcrowding may result in an underestima-
tion of numbers. Many texts recommend counting between 30 and 300 CFU
(Colony Forming Units) per plate to give statistical robustness.
Calculation: Multiply the average number of CFU on the three Petri dishes by the
inverse of the dilution that gave the reading in the range 30 to 300 CFU by 10 (to
correct for the 0.1mL used).
Example: Assuming that the average of three plates was 75 CFU and that the dilu-
tion that gave this reading was 106, calculate:
Average=75
Correction factor=10
Dilution of the suspension=106
CFU=75 10 106 7.5 108CFU/mL
6.3.2 Drop-plate counting

A drop of approximately 25L of the desired serial dilutions (e.g. 105, 106 and
107) is deposited on a sector of a Petri dish containing solid culture medium.
NB: if the plate is too wet, the drop will spread uncontrollably and if the plate is
too dry, the drop may bounce. To avoid these scenarios, cool the agar growth me-
dium to approximately 50C before pouring plates. This reduces condensation on
the lid which can then fall onto the agar surface. Pre-incubate plates for one hour
at 28C before use.
1. As a suggestion, plates can be divided into six sectors, using two sectors of each
plate as replicates per dilution. In this way, one plate is sufficient to count three
dilutions with two replicates, each with two drops per replicate (Figure6.2).
2. Calculate the number of plates required for the exercise.
3. Label the base of the plates before inoculation.
4. Using a sterile micro-pipette tip, or a Pasteur pipette, and working from the
most dilute suspension, place a drop from a height of approximately 12cm
112
onto the appropriate sector of the plate. Be careful to avoid splashing from
drops that can cause cross contamination between dilutions.
5. Repeat for replicate plates then move to the next lower dilution step and repeat
the procedure.
6. There is no need to change the pipette, or tip, but the weight of the drop should
be calibrated on a sensitive scale before the tip is discarded.
7. After absorption of the inoculum, the plates are inverted and placed in an
incubator at approximately 28C and allowed to grow for one to eight days,
depending on the strain (see Chapter 3 for growth rates). Inspect daily for
emergence of colonies.
8. Count the colonies in the sector that produces 550CFU per drop (sectors
2 or 3 in Figure6.2) either directly or through a dissecting microscope. It is
convenient to mark each colony with a fine-tipped marker pen while counting
to avoid confusion.
Figure6.2a An outcome of counting using the drop-plate method. The fresh

drops applied to the plate (left) illustrate how to position the drops for maximum
efficiency. After incubation (right) showing the expected dilution in number of
colonies from sector 3 through to sector 1.
Figure6.2b The emergence

of colonies in a drop at
approximately the correct
stage for counting. Colonies
often concentrate on the
circumference of the drop
with this technique.
113
6.4 Evaluating rhizobial survival on seed
Calculation: Multiply the average number of CFU per drop the inverse of the
dilution of the two sectors that provided these readings the correction factor
for volume of the drop (40, for the aliquot of 25L), to give live cells per mL of
original suspension.
Example: Assume that the average of the two drops in sector 2 above was 6.25CFU/
drop for the 106 dilution:
Average=6.25
Correction factor=40 (for a drop weight of 0.025g=25L)
Dilution factor=106
CFU=6.25 106 40 2.5 108CFU/mL
6.4 Evaluating rhizobial survival on seed

Survival of rhizobia on seed may be assessed by applying inoculant to seed at
the recommended rate (see Chapter8) then counting viable cells over time. For
this purpose, seeds should be surface-sterilised prior to inoculation (Chapter5,
Section 5.8.2) to reduce the effect of competition or antagonism by microorgan-
isms already present on the seed surface. The recommended rate of inoculant ap-
plied to seed will vary according to the number of rhizobia per unit weight of
inoculant and the desired number per seed. As an example, the recommended
rate of application of Australian peat inoculant for subterranean clover containing
>109CFU/g is 250g moist peat inoculant to 50kg seed.
Survival on seed of rhizobia from peat cultures may be tested in the laboratory
using the following protocol (also see Figure6.1).
1. Prepare a slurry by mixing 25g peat inoculant with 100mL adhesive solution
(rate will vary depending on manufacturers recommendations and rhizobial
number per g of inoculant). The peat culture should be homogenised before
sampling by rolling the packet between the fingers. The quantity of peat used
should be sufficient to adequately represent the total rhizobial population
within the packet.
2. Apply 1mL of slurry to 50g seed (or equivalent proportions) in a small beaker,
container or plastic bag and mix or shake thoroughly to distribute inoculant
over the seed surface. Note that if the adhesive solution is viscous and difficult
to pipette, it is better to weigh the quantity of slurry onto the seeds.
3. If coating inoculated seed with limestone or another dry powder, the powder
should be applied while seeds are wet and stirring or shaking should be con-
tinued for another 510secs until the seeds are evenly coated.
114
6.5 Indirect counts by plant infection to estimate Most Probable Number (MPN)
4. Sample seeds as soon as possible after coating, noting the time, and count the
viable number of rhizobia on the seed surface by dilution.
5. Remaining seeds should be spread on a clean surface to air dry and samples
taken at intervals to count viable rhizobia during drying and storage (e.g. one
hour to 72hours).
In general, the number of cells per seed counted immediately after coating pro-
vides an actual inoculum number, and counts after one hour may indicate natu-
ral death of rhizobia, incompatibilities with water-soluble toxins from the natu-
ral seed coat, seed applied chemicals (e.g. fungicide) or polymer adhesive. Some
strains of Burkholderia can lose viability completely within four hours after coat-
ing onto their legume host or onto glass beads (Howieson et al. 2013) so timing
of measurements for survival on seed may vary for each strainseed combination.
6.5Indirect counts by plant infection to estimate

Most Probable Number (MPN)
The Most Probable Number (MPN) method is an indirect way of counting rhizo-
bia in any non-sterile medium such as soil, inoculants or on seed. It involves in-
oculation of a compatible legume grown under aseptic conditions with the in-
creasing dilutions. Nodulation indicates that infective rhizobia were present in
the inoculum; no nodulation indicates they were absent. The MPN relies on the
assumption that one viable rhizobial cell can give rise to a nodule but it is implicit
that this cell must multiply after the inoculation event to produce sufficient nod-
factor to begin the nodulation process. For the method to give valid results, all
replicates of the last dilution should be negative.
After the addition of the inoculum, the test requires 10 to 30 days until the nod-
ules form. With basic microbiological skills, three replicates give adequate results,
although many laboratories use four replicate plants at each dilution step.
The MPN assays can be conducted in test tubes, plastic bags, pots of sand or Leon-
ard jars (Chapter5) or adaptations of these vessels that allow cell multiplication in
the rhizosphere, with one seedling growing per unit and at least three replicates.
The level of dilution required for enumeration of viable rhizobia will depend on
the number of microorganisms per g of sample. For example, if counting cells in a
high-quality peat containing 109CFU/g peat, the 107 and 108 dilutions should be
sufficient to reduce colony numbers to 10 and one respectively when 1mL is used
to inoculate plants. If a pre-inoculated seed product contains 103 CFU/seed then
the 101 and 102 dilutions should reduce colony numbers to 10 and one respec-
tively (where initial suspension of 100 seeds in 100mL of diluent is considered
the 100 dilution).
115
6.5.1 Counting multiple species of rhizobia simultaneously

More often than not soils contain several different species/genera of rhizobia and
these may be of interest to the researcher. An adaptation in the CRS laboratories
at Murdoch University is that the MPN is often used to count two or three species
of rhizobia in the soil simultaneously (e.g. S. medicae, M. ciceri bv biserrulae, R.
leguminosarum bv trifolii). The three host seedlings are planted in the one vessel.
One mL of inoculant is placed onto the roots of each seedling which are sown as
spaced plants in a 5cm pot containing steamed sand (Chapter5).
1. Sow seedlings into chosen vessel (Chapter5) one week before commencing
the MPN allowing 10% extra for mortality and for controls.
2. Prepare diluents and undertake dilution series (Section 6.2).
3. Inoculate 1mL from each dilution onto the roots of each seedling(s), repeat-
ing for each plant replicate.
4. Place inoculated plants into glasshouse and take steps to avoid contamination
(Chapter5).
5. After three to six weeks, inspect the plants in each unit for the presence or
absence of nodules.
6. Record nodulation on each plant in every dilution. The presence of at least one
nodule indicates a positive result.
7. Utilising MPN tables or the MPNES computer program* (Woomer 1994) cal-
culate the most probable number of cells in the original sample.
Figure6.3 A multiple MPN with each unit containing a seedling of biserrula and sub-clover
(left). The labels a-e, represent soil dilutions of 102106 with four replicates of each. The units
are watered through the pipette tip and further protected from contamination by Perspex shields
(right).
116
Example
An example of a calculation for a three replicate experiment of the type shown in

Figure6.3 is given below. For purposes of calculation, three dilutions inoculated
onto sub-clover (a: 102; b: 103; and c: 104) are nodulated; the next dilutions (d:
105 and e: 106) have resulted in no nodules.
Search Table 6.1 columns D1 (a), D2 (b) and D3 (c) with the number of units
(reps) testing positive for each dilution to determine factor (f). This factor from
Table6.1 is then multiplied by the lowest dilution before all units were negative
for nodulation.
Number of rhizobia=f d
Where:
f=factor from Table6.1
d=lowest dilution before all units were negative
Assuming that a soil gave three positive units at D1 (a), three at D2 (b), one at D3
(c) and none at d or e (105 or 106)
Factor (f; Table6.1) for 331=46.208
Lowest dilution=104
Number of cells=46.208 104=4.62 105 cells per g of soil
However, the biserrula in the example above was nodulated at d (the 105 dilution)
giving three positive reps in b, two in c and one in d.
The calculation for biserrula rhizobia in the soil is:
Factor (f, Table6.1) for 321=14.938
Lowest dilution=105
Number of cells=14.938 105=1.5 106 cells /g soil
NB: there are further tables for estimating the number of viable cells in Vincent
(1970), Brockwell et al. (1975) and Woomer et al. (1990). Table6.1 is from An-
drade and Hamakawa (1994) estimated according to Cochran (1950). Andrade
and Hamakawa (1994) also provide tables for three, four and five replicates.
117
Table6.1 E
stimation of number of cells of rhizobia evaluated by the Most Probable
Number infection method in plants. Adapted from Andrade and
Hamakawa (1994).
Dilution Factor Prob.% Interval of confidence
D1 D2 D3 MPN minimum maximum
0 0 1 0.300 0.332 0.073 1.675
0 0 2 0.601 0.001 0.186 2.176
0 0 3 0.904 0.000 0.329 2.645
0 1 0 0.305 3.365 0.074 1.698
0 1 1 0.611 0.045 0.189 2.208
0 1 2 0.917 0.000 0.334 2.684
0 1 3 1.224 0.000 0.498 3.140
0 2 0 0.620 0.156 0.192 2.239
0 2 1 0.930 0.004 0.338 2.722
0 2 2 1.242 0.000 0.505 3.185
0 2 3 1.555 0.000 0.686 3.636
0 3 0 0.944 0.004 0.343 2.762
0 3 1 1.261 0.000 0.512 3.233
0 3 2 1.579 0.000 0.695 3.693
0 3 3 1.898 0.000 0.890 4.141
1 0 0 0.357 39.203 0.087 2.058
1 0 1 0.723 0.624 0.225 2.711
1 0 2 1.098 0.006 0.399 3.336
1 0 3 1.482 0.000 0.601 3.948
1 1 0 0.736 6.445 0.228 2.762
1 1 1 1.118 0.179 0.407 3.401
1 1 2 1.510 0.002 0.612 4.026
1 1 3 1.911 0.000 0.838 4.644
1 2 0 1.138 0.627 0.414 3.468
1 2 1 1.538 0.025 0.623 4.108
1 2 2 1.950 0.000 0.854 4.740
1 2 3 2.370 0.000 1.104 5.356
1 3 0 1.568 0.030 0.635 4.194
1 3 1 1.988 0.002 0.871 4.842
1 3 2 2.418 0.000 1.126 5.486
1 3 3 2.860 0.000 1.397 6.126
2 0 0 0.917 31.927 0.288 3.773
2 0 1 1.432 1.120 0.522 4.791
2 0 2 1.990 0.019 0.802 5.832
2 0 3 2.600 0.000 1.123 6.887
2 1 0 1.469 11.963 0.535 4.951
2 1 1 2.046 0.632 0.823 6.037
2 1 2 2.680 0.015 1.156 7.143
2 1 3 3.359 0.000 1.529 8.261
2 2 0 2.106 2.318 0.847 6.261
continued
118
6.6 Estimate of cell number by optical density
Table6.1 (contd) E
stimation of number of cells of rhizobia evaluated by the Most
Probable Number infection method in plants. Adapted from
Andrade and Hamakawa (1994).
Dilution Factor Prob.% Interval of confidence
D1 D2 D3 MPN minimum maximum
2 2 1 2.763 0.170 1.191 7.420
2 2 2 3.478 0.005 1.579 8.598
2 2 3 4.240 0.000 2.009 9.782
2 3 0 2.855 0.216 1.229 7.726
2 3 1 3.602 0.021 1.632 8.966
2 3 2 4.408 0.001 2.082 10.217
2 3 3 5.254 0.000 2.574 11.478
3 0 0 2.312 34.098 0.871 12.822
3 0 1 3.850 3.099 1.511 17.662
3 0 2 6.348 0.158 2.434 22.749
3 0 3 9.538 0.004 3.706 27.927
3 1 0 4.272 37.433 1.664 21.327
3 1 1 7.488 6.579 2.777 28.088
3 1 2 11.520 0.649 4.380 35.159
3 1 3 15.878 0.032 6.485 42.502
3 2 0 9.324 32.817 3.331 38.555
3 2 1 14.938 12.507 5.569 50.581
3 2 2 21.470 2.470 8.652 64.067
3 2 3 29.170 0.235 12.543 79.220
3 3 0 23.970 36.594 9.128 139.550
3 3 1 46.208 42.767 17.836 240.763
3 3 2 109.849 44.442 38.227 478.767
6.6 Estimate of cell number by optical density
It is possible to determine the cell density of a suspension of a pure culture of

rhizobia spectrophotometrically but this method does not allow an assessment of
cell viability nor does it distinguish cell types. The advantage of measurement of
cell growth by optical density is that it is very rapid. An example is the standardi-
sation of suspensions of bacteria for competitiveness studies. However, a calibra-
tion curve has to be constructed for each strain by measuring optical density and
determining CFU per mL by plate counts. Calibration curves may vary between
strains, particularly if they produce exopolysaccharides.
1. Select 1cm path length cuvettes (plastic or glass).
2. Set spectrophotometer to 600nm and allow five minutes to warm up.
3. Insert cuvette containing 3mL growth medium in the correct orientation and
adjust to zero.
119
6.7 Direct counts under the microscope
4. Replace growth medium with suspension to be assayed, preferably a log-phase

culture.
5. Record OD, repeat with a second replication.
6. If OD is above 0.2*, dilute the sample to read in the range 0.05 to 0.2.
7. Construct calibration curve by diluting original culture with growth medium

to produce at least five points between 0.01 and 0.4, and determine plate count
at each dilution.
* Because the measured optical density of a suspension is due to light scatter-

ing by the suspended bacteria rather than light absorption, optical density figures
above 0.2 at 600nm are not linear with cell number and cultures should be diluted
so that optical densities are less than 0.2. It is important to note that cell death
starting from the stationary phase will reduce the relationship between viable cell
number obtained in plate counts and the optical density which will include live
and dead cells.
It is possible to enumerate rhizobia directly in pure cultures and high-standard

inoculants. There are some modern specialised techniques and equipment (e.g.
Coulter counters) for counting cells directly from a culture medium but this sec-
tion only describes standard laboratory methods.
Counts under the microscope produce total rhizobial numbers but do not dif-
ferentiate between viable and non-viable cells. Results from microscope counts
can be very useful to make quick decisions without having to wait for bacterial
growth.
The use of calibrated chambers is well described in several textbooks (Vincent

1970; Somasegaran and Hoben 1994). The Petroff-Hausser chamber has a depth
of 0.002cm, with intermediate (4 104cm2) and small (2.5 105cm2) squares
(Figure6.4). With a known volume derived from the depth and the area of the
squares, the number of rhizobia per mL can be calculated. The shallow depth al-
lows a direct count of cells with normal bright field illumination. However, phase
contrast microscopy gives better definition because unstained cells in bright field
microscopy have essentially the same refractive index as the counting medium. A
special coverslip is required for the use of 20 to 40 objectives; the 40 magnifi-
cation phase contrast objective usually gives the better results. The counting limit
in these chambers is eight to ten cells per intermediate square, corresponding to
106cells/mL.
120
Figure6.4 A representation
of the components of a Petroff-
Hausser counting chamber
(a) Cross section showing location
of the sunken platform etched
with the grid system.
(b) Top view of complete grid.
(c) Magnified view of an
intermediate square containing 16
small squares.
(Image from Somasegaran and
Hoben 1985)
121
6.7.1 Petroff-Hausser chamber

There are several precautions to be observed in the use of a Petroff-Hausser cham-
ber.
1. The chamber and cover glass should be washed in detergent, rinsed several
times in distilled water and dried in a dust-free environment. This will ensure
that the liquid spreads evenly and prevents the formation of air bubbles dur-
ing counting.
2. Bacterial cultures should be diluted several times (Section 6.2.1) to determine

the best dilution to be used during the tests at each stage of growth. After dilu-
tion, the chamber should be washed again.
3. The diluted culture is added to the slide (after the coverslip is in place) with
the aid of a micro-pipette or Pasteur pipette. The pipette should only touch the
chamber and the liquid will flow by capillarity, thus avoiding leakage. If excess
culture does escape, it should not be dried-off with filter paper. The coverslip
should be removed and, along with the chamber, washed again, dried and the
whole process restarted.
4. The image should be observed at a 40 magnification under phase contrast

and the number of cells counted.
5. For cells that overlap a border, a criterion should be established, e.g. count a
cell as in if it overlaps the top or right border, and out if it overlaps the bot-
tom or left border.
Logarithmic growth phase cultures, where few cells are dead, give good results
for counting viable cells; the advantage being the speed of obtaining results. It is
only appropriate for liquid media without particles and requires trained people
to make the meticulous observations needed for quick and reproducible counts.
Calculation: Table 6.2 can be used for planning counts of rhizobia in Petroff-
Hausser chambers.
Table6.2 N
umber of cells that can be counted without and with dilution (Vincent
1970)
Expected number of bacteria/mL (106) Dilution
<10 No need
10100 No need
1001,000 10-fold dilution
1,00010,000 100-fold dilution
Calculation: Bacteria per mL of the initial suspension: Multiply d (dilution of the

counted suspension) n (number of cells per square) f (factor for the size of
square).
122
Example: Assume an average of 33 bacteria are counted in an intermediate square

from 102 dilution.
Dilution (d)=102
Number of cells (n)=33
Factor (f)=1.25 106
Number of cells=33 102 1.25 106 4.125 109cells/mL
Other useful information for counting in Petroff-Housser chambers is given in

Tables 6.3 and 6.4.
Table6.3 A
rea, corresponding volume and factor for sizes of squares in the Petroff-
Hausser chamber
Area Corresponding Factor (f)
volume (mL) (1/vol.)
Total enclosed squares (25 16=1 102cm2) 2 105 5 104
Intermediate square (=16 smallest squares=4 104cm2) 8 107 1.25 106
Smallest square (2.5 105cm2) 5 108 2 107
Table6.4 N
umber of fields that have to be counted
according to the number of bacteria per field
of the Petroff-Hausser chamber (Vincent
1970)
Average number Number of fields to be counted for the
per square coefficient of variability
10% 5%
10 10
20 5 20
40 3 10
80 2 5
100 1 4
6.7.2 Neubauer haemocytometer

For more dilute samples, a haemocytometer (0.01cm depth) with Neubauer rul-
ings can be used. In this methodology, bacteria can be stained with methylene blue
(1mL of 1% (w/v) methylene blue, freshly prepared) and counted after approxi-
mately 20 minutes. The main divisions of the Neubauer haemocytometer separate
the grid into nine large squares, each of which has a surface area of 1.0mm2 and
the depth of the chamber is 0.1mm. Thus, the entire counting grid lies under a
volume of 0.9mm3. In this chamber, cells in squares have to be selected so that
the total count is around 100 cells; the number needed for statistical significance
if several cells are counted. Table6.5 gives the parameters to estimate cell number
in the haemocytometer.
123
6.8References
Table6.5 A
rea, corresponding volume and factor for size of square in the Neubauer
haemocytometer
Area Volume Factor (f)
(mL)1 (1/vol.)
Total squares9 large squares 1 mm2 (area) 0.1 mm (depth)=0.9 mm3 9 104 1.1 103
Large square1 mm2 (area) 0.1 mm (depth)=0.1 mm3 1 104 1 104
Small squares1/25 mm2 0.04 mm2 (area) 0.1 mm (depth)=0.004 mm3 4 106 2.5 105
1,000mm3=1cm3=1mL
The same formula used for the Petroff-Hausser can be also used for the haemo-
cytometer.
Calculation: To determine the number of bacteria per mL for the initial suspen-
sion, multiply d (dilution of the counted suspension) n (number of cells per
square) f (factor for the size of square).
Example: Assuming that 1mL of a rhizobia culture of the 102 dilution gave an
average of 38 bacteria in the small squares (e.g. five small squares resulting in 190
bacteria, therefore higher than 100 and statistically valid).
Dilution (d)=102
Number of cells (n)=38
Factor (f)=2.5 105
Number of cells=38 102 2.5 105 9.5 108cells/mL
6.8References
Andrade D.S. and Hamakawa P.J. 1994. Estimativa do nmero de clulas viveis de rizbio no solo
e em inoculantes por infeco em plantas. Pp 6394 in Manual de mtodos empregados em
estudos de microbiologia agrcola, ed. By M. Hungria and R.S. Araujo. EMBRAPA-SPI: Braslia,
Brazil.
Brockwell J., Diatloff A., Grassia A. and Robinson A.C. 1975 Use of wild soybean (Glycine ussuriensis
Regel and Maack) as a test plant in dilution-nodulation frequency tests for counting Rhizobium
japonicum. Soil Biology and Biochemistry 7, 305311.
Cochran W.X. 1950. Estimation of bacterial density by means of the most probable number. Bi-
ometrics 6, 105116.
Howieson J.G., De Meyer S., Vivas-Marfisi A., Ratnayake S., Ardley J.K. and Yates R.J. 2013. Novel
Burkholderia bacteria isolated from Lebeckia ambiguaa perennial suffruticose legume of the
Somasegaran P. and Hoben H.J. 1994. Handbook for rhizobiamethods in legume rhizobium tech-
nology. Springer-Verlag: New York, USA.
Vincent J.M. 1970 Manual for the Practical Study of Root Nodule Bacteria. (International Biological
Programme Handbook 15). Blackwell: Oxford, UK.
Woomer P.L. 1994. Most probable number counts. Pp 5979 in Methods of soil analysis. Part 2.
Microbiological and biochemical properties, ed. by R.W. Weaver, S. Angle, P. Bottomley, D.
Bezdicek, S. Smith, A. Tabatabai and A. Wollum. (SSSA Book Series 5). Soil Science Society of
America: Madison.
Woomer P. Bennett J. and Yost R. 1990. Overcoming the inflexibility of most-probable-number pro-
cedures. Agronomy Journal 82, 349353.
124
7.1 Taxonomy of rhizobia
CHAPTER7
Taxonomy and
physiology of rhizobia
G.W. OHara, J.E. Zilli, P.S. Poole and M. Hungria
Rhizobia are common Gram-negative soil-inhabiting bacteria distinguished by

the feature that they contain genes required for nodulation (e.g. nod, rhi) and ni-
trogen fixation (e.g. nif, fix). These genes enable them to form a symbiotic asso-
ciation with legumes. Currently there are 15 genera of root nodule bacteria (Ta-
ble7.1) containing more than 120 described species.
7.1.1 Species concept for bacteria

The current species concept used for bacteria is necessarily rather subjective, with
a species being considered as a distinct group of strains that share many stable
distinguishing features and differ significantly from other groups of strains (Bren-
ner et al. 2005a). The reader is referred to Bergeys manual of systematic bacteri-
ology, in particular Volume 2, Part A Introductory Essays (Brenner et al. 2005b)
for detailed information on current understanding and principles of prokaryotic
taxonomy.
For prokaryotic organisms, the International Committee on Systematics of

Prokaryotes (ICSP) defined in 2002 the Gold Standard for species assignment
based on a polyphasic approach using the following four characteristics.
1. Phenotypic and/or morphological similarity.
2. Genome similarity.
3. G+C content similarity.
4. 16S rRNA sequence similarity.
125
A prokaryotic species is thus recognised as a collection of strains that have: phe-

notypic and/or morphological similarity; genome similarity, as shown by DNA-
DNA hybridisation >70%; G+C content similarity, as indicated by melting tem-
peratures of DNA within 5C; and 16S rRNA sequence similarity as indicated by
<3% divergence. Established practice is that new species of prokaryotic organisms
are formally published in the International Journal of Systematic and Evolution-
ary Microbiology (http://ijs.sgmjournals.org/), the official journal of record for
bacterial names of the ICSP.
A further taxonomic sub-division of biovar has been used for some species of
rhizobia, where the biovar is defined by the host range for nodulation as deter-
mined by nod and related genes being carried on the accessory genome (Jordan
1984). More recently the term symbiotic variant or symbiovar has been proposed
to replace biovar, as a parallel term to pathovar in pathogenic bacteria (Rogel et
al. 2011). At present, there are numerous recognised symbiovars in four genera
of nodule bacteria: Sinorhizobium (e.g. S. meliloti sv. ciceri, S. meliloti sv. meliloti);
Rhizobium (e.g. R. leguminosarum sv. viciae, R. leguminosarum sv. trifolii); Mes-
orhizobium (e.g. M. ciceri sv. ciceri, M. ciceri sv. biserrulae); and Bradyrhizobium
(e.g. B. japonicum sv. genistearum, B. japonicum sv. glycinearum).
7.1.2 Polyphasic approach for taxonomy of rhizobia

The polyphasic taxonomic approach (Graham et al. 1991; Vandamme et al. 1996)
used for determining species of rhizobia is overseen by the ICSP Subcommittee
on the taxonomy of Rhizobium and Agrobacterium (Lindstrom and Young 2011).
The subcommittees website (http://edzna.ccg.unam.mx/rhizobial-taxonomy)
provides useful information on:
1. up-to-date taxonomic information about all validly published and forthcom-

ing rhizobial/agrobacterial taxa
2. hands-on tutorials on how to perform rigorous phylogenetic analyses using

sequence data, from SSU rRNA sequence alignment and model fitting up to
ML tree searches
3. information on current issues in rhizobial taxonomy, systematics and genom-

ics, including scientific meetings and workshops
4. advice and recommendations from the subcommittee for microbiologists on

good taxonomic/systematic practice.
The designation of species of rhizobia is based on the combination of classical

microbiological characteristics (morphology, physiology, biochemistry) with the
genome characteristics of DNA base composition (G + C content), DNA-DNA
hybridization, sequences of the 16S ribosomal RNA genes and Fatty Acid Methyl
Ester (FAME) profiles.
126
The usefulness of the 16S rRNA gene as a molecular marker for assessing phyloge-
ny and taxonomy of prokaryotes has been broadly demonstrated, and the gene has
also been applied to rhizobial taxonomy. Information about amplification of these
genes is given in Chapter11. However, there are problems with solely relying on
16S rRNA sequences, including: the issues of plesiomorphy; multiple copies of
the 16S rRNA gene with small intragenomic differences (up to 5%) (Kampfer and
Glaeser 2012); the conserved structure of the 16S rRNA gene sequence limiting
resolution power below genus level (e.g. Willems et al. 2001; Gevers et al. 2005);
and reports that genetic recombination and horizontal gene transfer occur among
16S rRNA genes (e.g. Gevers et al. 2005). Other regions of DNA have been pro-
posed as alternative phylogenetic markers (Stackebrandt et al. 2002), such as the
16S-23S rRNA ITS region and genes (termed housekeeping genes) located in the
core genome. These have a faster evolution rate than 16S rRNA but are conserved
enough to retain genetic information. These genes need to be both broadly dis-
tributed among taxa and also be present in single copies within a given genome;
the current consensus is that at least five genes are necessary for reliable taxonom-
ic classification (Stackebrandt et al. 2002). Among the most used genes are dnaK,
dnaJ, glnA, gyrB, recA, gltA, glnII, rpoA, rpoB, and atpD. There is also increasing
discussion about the adoption, together with the 16S rRNA, of MLSA (Multi Lo-
cus Sequence Analysis) of housekeeping genes (Martens et al. 2007; Kampfer and
Glaeser 2012) to replace the high cost and intensive work required for the DNA-
DNA hybridization still required for species definition. Increased use of MLSA
with rhizobia suggests that it may soon be an accepted tool to define new species;
identities of 9496% have been proposed to replace the 70% of DNA-DNA ho-
mology.
With wider availability of whole genome sequences, their use for species defini-
tion has been proposed. A first proposal was based on the ANI (Average Nucleo-
tide Identity) of the whole genome by Konstantinidis and Tiedje (2005a,b) who
suggested that the level of 94% identity of the shared genes between two strains
would be equivalent to the 70% level of DNA-DNA hybridization. Further devel-
opments will follow as more rhizobia genomes are sequenced and available for
analysis.
Further information on the current taxonomy of rhizobia can be found on Bev-

an Weirs New Zealand rhizobia website (http://www.rhizobia.co.nz/taxonomy/
rhizobia).
7.1.3Current rhizobia are in the Proteobacteria (Gram-

negative bacteria)
All the current known rhizobia are in the Proteobacteria (the Gram-negative bac-
teria), a large phylum of more than 500 genera containing more than 2000 species.
Prior to 2001, rhizobia were not known to occur outside the -Proteobacteria, but
127
Moulin and collaborators (Moulin et al. 2001) reported the isolation from Mimo-
sa spp. of rhizobia of the genus Burkholderia in the -Proteobacteria, a result con-
firmed more broadly in Mimosa spp. by others (Gyaneshwar et al. 2011; Bournaud
et al. 2013). The host range of beta-rhizobia has been extended further with their
isolation from root nodules of South African shrub (Elliott et al. 2007) and forage
legumes (Garau et al. 2009; Howieson et al. 2013; De Meyer et al. 2013a,b,c). It
is important to be aware that the genera and species containing rhizobial strains
also contain non-rhizobial strains and species (non-symbiotic bacteria lacking
symbiotic genes) that can be isolated from soils and plants. To avoid confusion
and misunderstandings, the status of a strain as rhizobial must be assessed by
nodulation tests (Chapter3).
The Alpha-Proteobacteria contains about 20 families of predominantly aerobic

bacteria; they are characteristically oligotrophic (able to grow at very low sub-
strate concentrations) and characterised by having some unusual metabolic
modes such as methylotrophy, chemolithotrophy and N2 fixation. There are cur-
rently 13 genera of Alpha-Proteobacteria containing legume-nodulating species:
Aminobacter, Azorhizobium, Bradyrhizobium, Devosia, Ensifer (Sinorhizobium),
Mesorhizobium, Methylobacterium, Microvirga, Neorhizobium, Ochrobactrum,
Phyllobacterium, Rhizobium and Shinella.
The Beta-Proteobacteria contains about 12 families of bacteria that are somewhat

similar to the Alpha-Proteobacteria but tend to use substrates that diffuse from
organic decomposition in anoxic zones in soils/sediments. Bacteria in this group
are often characterised by having greater catabolic versatility than the Alpha-Pro-
teobacteria. There are currently two nodulating genera of Beta-Proteobacteria:
Burkholderia and Cupriavidus.
Table7.1 F amilies and genera of root nodule bacteria and the approximate
number of described species (Also shown are genera containing species
used as commercial legume inoculants.)
Family Genus Number of described species
-Proteobacteria
Bradyrhizobiaceae Bradyrhizobium# 15
Brucellaceae Ochrobactrum 2
Hyphomicrobiaceae Azorhizobium 3
Devosia 1
Methylobacteriaceae Methylobacterium# 1
Microvirga 3
Phyllobacteriaceae Phyllobacterium 1
Aminobacter 1
Mesorhizobium# 29
Rhizobiaceae Rhizobium# 43
Neorhizobium# 3
Sinorhizobium/Ensifer# 13
Shinella 1
continued
128
7.2 Physiology of root nodule bacteria
Table7.1 (contd) F amilies and genera of root nodule bacteria and the approximate
number of described species (Also shown are genera containing
species used as commercial legume inoculants.)
Family Genus Number of described species
-Proteobacteria
Burkholderiaceae Burkholderia# 6
Cupriavidus 2
# Indicates genera that contain species with strains used as commercial inoculants in
agriculture.
Currently the number of formally described species of rhizobia is increasing at a

rate of more than 10 species per year. However it is important to note that strains
in only five genera of rhizobia are currently used as inoculants in agriculture, as
indicated in Table7.1.

Root nodule bacteria are predominantly aerobic chemoorganotrophic organisms
with an oxidative metabolism. They are relatively easy to culture and grow very
well in the presence of oxygen when utilizing simple carbohydrates and amino
acids. Optimal growth for most strains occurs at a temperature range of 2530C
and at a pH of 67 (Table7.2).
7.2.1 Metabolic diversity

Being predominantly chemoheterotrophic organisms, most rhizobia are not par-
ticularly fastidious in their nutritional requirements. However, there are excep-
tions and some species do not grow on particular classes of carbon compounds;
many species of Bradyrhizobium fail to grow on disaccharides. It is also impor-
tant to recognise that some strains have quite particular metabolic activities, such
as facultative autotrophy (e.g. B. japonicum, Boursier et al. 1988), denitrification
(OHara and Daniel 1985), phototrophy (Evans et al. 1990; Young et al. 1991) and
in a few strains an ability to fix N2 ex planta (e.g. Azorhizobium, Dreyfus et al.
1988; Bradyrhizobium, McComb et al. 1975).
7.2.2Growth rates in laboratory culture vary with genus and

species
The growth rates of rhizobia in laboratory culture vary widely and can be used to
differentiate them from contaminants, as outlined in Chapter3. Table7.2 below
provides a guide to common growth characteristics of each genus.
129
Table7.2 G
rowth rate, generation time, temperature and pH ranges for the genera of nodule
bacteria
Genus Growth rate Mean Optimal Minimum Maximum Common pH
generation temp. range growth growth range
time (h) (C) temp. (C) temp. (C)
-Proteobacteria
Aminobacter fast 35 2530 nr1 3739 511
Azorhizobium fast 35 2530 12 43 58
Bradyrhizobium slow 818 2530 3335 3335 48
Devosia fast 24 2530 nr 1
nr1
5.57
Mesorhizobium moderate 415 2530 410 3742 410
Methylobacterium moderate 57 3037 nr 1
3739 57
Microvirga fast 24 3035 10 43 5.59.5
Neorhizobium fast 24 2530 10 3740 510
Ochrobactrum fast 34 2530 20 39 510
Phyllobacterium fast 24 2530 4 37 68
Rhizobium fast 24 2530 10 40 410
Sinorhizobium fast 24 2530 10 44 510.5
Shinella fast 24 2530 nr 1
nr1
511
-Proteobacteria
Burkholderia very fast 1 2530 10 40 4.59
Cupriavidus very fast 1 2530 nr 1
39 nr1
nr1=not recorded
7.2.3 General description of rhizobia

Most colonies of nodule bacteria are white or creamy and opaque, are rarely
translucent (except when young) rarely pigmented and do not adsorb Congo
Red. However there are rare exceptions and some species of Burkholderia and
Sinorhizobium can adsorb the dye, while other species are pigmented (Chapter3,
Section 3.5) so this general description is equivocal.
Colony characteristics change with media, time and conditions of incubation. The
colony texture may be buttery or elastic, and can be determined by touching the
colony surface with a loop. Colony appearance may be gelatinous, dry or wet. In
general, colonies are flat or rounded, with a few strains producing colonies with a
conical or fried egg-shaped elevation. Some strains of S. medicae will produce col-
onies with a doughnut appearance (Howieson et al. 1988) as shown in Chapter3,
Figure3.6. Table7.3 provides general descriptions of the colony characteristics of
different genera and species of nodule bacteria.
130
Table7.3 C
olony morphology and characteristic features for nodule bacteria grown on routine
media (usually YMA unless otherwise noted) at 28C
Genus Colony morphology and characteristic features
Aminobacter Colonies appearing on YMA within two to three days of incubation at 28C are circular,
opaque, convex, have a creamy colour and are usually a diameter of 23mm.
Azorhizobium Colonies on agar are circular and have a creamy colour.
Bradyrhizobium Colonies are circular, convex and often translucent at first appearance, and when
less than 1mm in diameter (approx. five to eightdays), but then strongly opaque
and tending to dark grey with further maturity. Colonies of some strains are 3mm in
diameter or greater after four to seven days. Rarely gummy.
Devosia Colonies are mucoid and pearl white on YMA medium; typical of fast growing
rhizobia.
Mesorhizobium Colonies as for Rhizobium but visible 2448hours later, weakly opaque after 72hours,
increasing with age.
Methylobacterium Colonies visible in four to seven days. Colonies are glistening, smooth, raised, with
entire margins, 0.51mm in diameter after four to seven days incubation. After
7days, colonies of some strains from species of Listia are pale pink to bright red-
orange.
Microvirga Colonies visible in 24hours. Many colonies are light pink or pale orange to brown,
convex, smooth, round and colouration develops 72hours after visible colonies
emerge, weakly opaque and mildly gummy. Some strains may show brown spot after
seven to-10days.
Neorhizobium Colonies are circular, convex, white or cream coloured, semi-translucent or opaque,
with a diameter of 14mm within two to four days on YMA at 28C.
Ochrobactrum Cells form white mucoid colonies on YMA medium.
Phyllobacterium Colonies are small, punctiform or circular, pearl white with a regular edge and colony
diameter <1mm after one to two days growth on YMA at 28C.
Rhizobium Colonies usually visible in 48hours. Colonies are usually white or beige, circular,
convex, semi-translucent or opaque, raised and mucilaginous, 24mm in diameter
within three to five days on YMA.
Sinorhizobium/Ensifer Colonies visible in 48hours, weakly opaque after 72hours and gummy. Doughnut
morphology (Figure3.6) in S. medicae when first isolated from nodules.
Shinella Colonies are circular, cream coloured, semi-translucent and 24mm in diameter after
three days incubation at 28C on YMA medium
Burkholderia Colonies visible in 24hours. Colonies are white, smooth, round, convex, mostly with
entire margins, 0.52.0mm in diameter after 24hours incubation. Colonies may
acquire a yellow-brown tinge after 48hours when viewed from below.
Cupriavidus Colonies visible in 24hours. Colonies are white, smooth, round, convex with entire
margins, 0.54.0mm in diameter after 24hours incubation. Colonies may acquire a
grey tinge after 48hours when viewed from below (Figure 3.5).
7.2.4 Colony dimorphism

Some strains of rhizobia show two colony types (e.g. Bradyrhizobium, Sylvester-
Bradley et al. 1988; Sinorhizobium, Bloem et al. 2002). In these situations it is
important to take special care to eliminate the possibility of contamination. Sub-
culturing of each colony type several times may be required to verify purity; plant
testing to confirm authenticity is recommended (Chapter4).
131
7.2.5 Purity tests for rhizobial cultures

Rhizobial cultures can be easily contaminated and it is essential in all stages of
strain management that purity checks, as described below, are regularly used.
7.2.5.1 Colony characteristics
The easiest purity check is the recognition of typical colony morphology of a par-
ticular strain. A very important practical skill for a rhizobiologist to develop is
an ability to recognise what are rhizobia and what are contaminants. With regu-
lar practice, and by repeated examination, individual strain characteristics can be
readily recognised by the experienced researcher, and the presence of contami-
nants easily distinguished. It is essential that cultures are incubated for the appro-
priate period for the particular strain (e.g. incubation for more than seven days
for Bradyrhizobium colonies to develop, especially those isolated from nodules).
Table7.4 lists a series of observations that can assist in differentiating common
species of rhizobia.
Table7.4 C
olony morphology and characteristic features useful for distinguishing between species
of some common for nodule bacteria
Genus and species Colony morphology and characteristic features
Rhizobium leguminosarum Colonies usually visible in 48hours. Colonies are usually white or beige,
bv trifolii circular, convex, semi-translucent or opaque, raised and mucilaginous, 24mm
in diameter within three to five days on YMA. When grown on LA gummy
strands from colony adhere to a loop.
Rhizobium leguminosarum bv As for Rlt described above, and weakly opaque at 72hours. When grown on
viceae LA gummy strands generally do not adhere to a loop.
Burkholderia dilworthii Colonies visible in 24hours. Colonies are white, smooth, round, convex with
entire to slightly irregular margins, 0.52.0mm in diameter after 24hours.
incubation. Colonies acquire a yellow-brown tinge with a brown centre after
48hours when viewed from below. When grown on LA containing Congo Red
the agar shows a purple colour after eight to 10 days incubation (Figure 3.11).
Burkholderia sprentiae As for Bd described above. Colonies are creamier and do not develop brown
centre after 48hours. When grown on LA medium containing Congo Red
Burkholderia phymatum
colonies show dark red colour after eight to 10 days incubation.
Methylobacterium spp. (from Colonies visible in four to seven days. Colonies are glistening, smooth, raised,
Listia bainesii) with entire margins, 0.51mm in diameter after four to seven days incubation.
Colonies develop pink pigmentation within 24hours. of becoming visible. After
seven days colonies are 13mm in diameter and pale pink to bright red-orange.
Methylobacterium spp. (from As above, except colonies are strongly opaque, not gummy and do not develop
other legumes) pink or orange colour.
Sinorhizobium medicae Colonies visible in 48hours. Doughnut morphology when first isolated from
nodules, weakly opaque after 72hours and gummy.
Sinorhizobium meliloti Colonies visible in 48hours. Do not form doughnut colonies, weakly opaque
after 72hours, often (but not always) drier than S. medicae.
Bradyrhizobium spp. Colonies slow growing and translucent at first appearance when less than 1mm
in diameter (approx. seven to 10 days) but then strongly opaque and tending to
dark grey with further maturity. Rarely gummy.
Bradyrhizobium japonicum As above, but may reach 13mm after five to eight days, immediately opaque,
and elevated.
Bradyrhizobium elkanii As for B spp., colonies irregular and non-elevated.
132
7.2.5.2 Gram stain and motility
Several characteristics are important and discriminating. Firstly, cells are usual-
ly motile in fresh broth cultures which can be examined under the microscope.
While bacterial flagella are not visible with the light microscope unless stained,
motility caused by movement of flagella is seen by direct microscopy of wet prepa-
rations. A fresh broth culture is required for best observations of motility. Freshly
prepared slides should be viewed promptly since motility can be easily lost with
prolonged exposure on a slide.
1. Using a sterile Pasteur pipette, take a sample of the broth culture.
2. Transfer a drop of the culture to the centre of a slide, cover with a coverslip.
3. Examine under the microscope with the 40 objective (you may need to re-
duce the illumination of your specimen in order to see the unstained bacteria).
4. Note that cells showing simple vibrationBrownian motionare NOT mo-

tile; they need to show long swims across the field.
Secondly, all rhizobia are Gram-negative, rod-shaped bacteria and do not produce
spores. The Gram stain devised by the Danish scholar, Christian Gram, in 1884 is
the most widely used differential stain in bacteriology and divides common bacte-
ria into two groups: Gram-negative (which appear pink after Gram staining) and
Gram-positive (which appear purple).
Protocol for the Gram stain
NB: use forceps to hold the slide for all manipulations involving dyes.
1. Take a clean slide out of the alcohol in the storage tub, flame off any alcohol
and allow the slide to cool. Label the slide with pencil in the frosted area so
that you know which sample it is.
2. Using a sterile loop, take a small amount of a bacterial colony from the agar
plate and place it on the slide. Emulsify it with a small drop of sterile saline and
spread the liquid over an area of about 1cm square.
3. Air dry the smear or dry it gently high above a Bunsen burner flame.
4. Heat fix the dried slide by passing it three times through the hot part of the
Bunsen flame with the smear side up. Use forceps to hold the slide. Allow the
slide to cool.
5. At the staining rack, cover the smear with a few drops of crystal violet for
60seconds.
6. Pour off the excess crystal violet and rinse with a gentle stream of running tap
water.
7. Cover the smear with iodine and leave for 45seconds.
133
8. Pour off the excess iodine, wash down the front and back of the slide with the
acetone/alcohol solution until purple stain stops eluting and then immediately
wash under a gentle stream of running tap water for 510seconds.
9. Counterstain with safranin for 30seconds.
10. Wash off any excess stain with water and gently blot the slide dry between
sheets of blotting paper. Always remember which side of the slide is up.
11. View your slide under the microscope (40 magnification) to describe the
Gram stain and morphology of the bacterial cells.
Gram-positive bacteria retain the crystal violet stain and appear purple. Gram-
negative bacteria, such as rhizobia, do not, and are thus colourless until counter
stained with a basic dye such as safranin, which stains colourless bacteria pink but
does not change the purple colour of the Gram-positive bacteria. It is essential to
use fresh broth or agar cultures, as older cultures of Gram-positive bacteria may
fail to retain the crystal violet and falsely appear to be Gram-negative.
Crystal violet stain
Crystal violet 10g

Ammonium oxalate 4g
Ethanol 100mL
Water 400mL
Iodine solution
Iodine 1g
Potassium iodide 2g
Ethanol 25mL
Water 100mL
Counterstain
2.5% (w/v) Safranin in ethanol 10mL

Make up to 100mL with water
There are also a range of molecular methods which can aid in the identification of
rhizobia. These are covered in Chapters 11 to 13.
7.2.6 Physiology of rhizobia

Rhizobia commonly use the Entner-Doudoroff and Pentose Phosphate pathways
for sugar metabolism, and normally have a fully functional TCA cycle. They can
134
often use a variety of nitrogen sources (see below), and ammonia assimilation is
via the GS/GOGAT system.
In general, rhizobia can catabolise a wide array of carbon compounds and obtain
iron from a range of sources; some strains use a variety of siderophores for iron
uptake (Poole et al. 2008). However, some species of rhizobia can be auxotrophic
for specific vitamins such as biotin, thiamin, and pantothenate (e.g. M. loti, Sul-
livan et al. 2001; Graham 1963; S. meliloti, Watson et al. 2001) and these need to
be included in defined media for growth of these strains.
7.2.6.1 Physiology of rhizobia in relation to carbon sources
Summarising the carbon metabolism of such a diverse range of organisms as the

root nodule bacteria is obviously difficult and individual exceptions to many of
the generalisations listed below can be readily found. Nevertheless, these gener-
alisations are useful as background to devising media for specific organisms and
purposes.
The three major physiological groupings are essentially the fast- and slow-growing
rhizobia, and the very fast-growing -rhizobia. Growth of the root nodule bacte-
ria is generally heterotrophic; in a few cases autotrophic growth on H2/CO2 has
been observed in the slow-growing group. Genes for C1 metabolsim, like ribulose
bis-phosphate carboxylase, are nevertheless conserved, and autotrophic growth in
the soil under particular conditions may be possible.
While there may be occasional differences with particular strains, the following
generalisations about carbon nutrition can be made for the -rhizobia.
1. In very broad terms, the range of substrates on which the fast-growing group
will grow is wider than that for the slow-growing group.
2. Both groups metabolise a wide range of C5- and C6- sugars and sugar alco-
hols, often resulting in acidification of solid and liquid media due to organic
acid production. The major sugar degradation route appears to be the Entner-
Doudoroff pathway.
3. Only the fast-growing groups metabolise C12 sugars like sucrose; the slow-
growing groups apparently lack uptake systems and the appropriate disaccha-
ridases. Metabolism of C12 sugars by fast-growers also results in acidification
of media (see Chapter4 for pH control).
4. The metabolic pathways for the metabolism of the C5- sugar L-arabinose differ
between the fast- and slow-growing groups.
5. C4-Dicarboxylic acids (malate, fumarate, succinate) are usually good sub-

strates for both fast- and slow-growing groups. Malic enzyme is an essential
enzyme for degradation of these dicarboxylic acid substrates, both in labo-
ratory cultures and in N2-fixing nodule bacteroids. Growth on organic acids
135
as sole carbon sources results in alkalinisation of unbuffered solid or liquid

media.
6. Growth on amino-acids as sole carbon sources is often strain-specific, so

growth of particular strains has to be checked individually. Growth is often
inhibited by concentrations normally used for sugars or organic acids like
1020mM but growth may occur at lower concentrations. Their metabolism
results in marked alkalinisation of unbuffered media and significant ammo-
nia (NH3) accumulation. Apart from its alkalinising effect, this ammonia can
be tolerated to quite high concentrations (50mM or more). Growth on some
amino-acids as sole source of carbon can cause marked changes in cell mor-
phology. For example, growth of Rhizobium leguminosarum on histidine pro-
duces almost coccoid cells.
7. The ability of root nodule bacteria to utilise aromatic compounds (those con-
taining the benzene ring) varies widely, though the range of compounds used
is usually wider with the fast-growing than with the slow-growing groups. The
catechol and/or the protocatechuate branches of the 3-oxoadipate pathway
may occur in a particular strain. In most cases, the enzymes of the 3-oxoadi-
pate pathway are inducible in the fast-growing groups and constitutive in the
slow-growing, though in the latter one key enzyme is usually inducible.
8. Terminal oxidation of substrates typically involves the tricarboxylic acid cycle.

Oxidation of C2- substrates like acetate via the glyoxylate cycle is well docu-
mented but in many organisms, the activities of the key enzymes, isocitrate
lyase and malate synthetase, are low and growth on acetate is poor.
9. Terminal electron transport is generally to O2 via a variety of cytochrome-

based systems. Some strains can use nitrate as a terminal electron acceptor
and achieve limited anaerobic growth.
10. Metabolic regulation of the utilisation of mixed sources of carbon is more like
that in Pseudomonas than in the enterobacteria. Thus, root nodule bacteria
will co-utilise mixed carbon sources (though obviously the rates may be wide-
ly different) rather than fully utilise a preferred carbon source before starting
on a less-favoured one. The classical catabolite repression typical of substrate
utilisation in Escherichia coli is not a major aspect of regulation in root nodule
bacteria.
7.2.6.2 Physiology in relation to nitrogen source
A very limited range of root nodule bacteria show nitrogenase activity in the labo-
ratory and only under highly selective conditions; only a few strains have been
shown to fix enough N2 (usually only under very stringent conditions) to meet
their growth requirements. A fixed nitrogen source is therefore necessary for
growth in laboratory culture.
136
The ability to use particular compounds as nitrogen sources varies widely. Most
complex media contain a sufficiently wide range of compounds to meet growth
requirements. Trying to grow an unstudied strain in defined medium may there-
fore require assessment of whether it will grow using urea, ammonium salts, ni-
trates or amino-acids as sources of N. Many strains utilise all these forms of N, but
some do not. Growth may also be faster with a particular form and this can only
be determined experimentally.
Because a dense culture (A600 nm = 1 to 2) may use all the N from a 510 mM
N-source, pH changes associated with N utilisation may also be significant, i.e.
acidification with ammonium salts or alkalinisation with nitrate or amino-acids.
Accumulation of nitrite may be a (not commonly recognised) consequence of
growth on nitrates.
7.2.6.3 Physiology in relation to pH change
As mentioned above, catabolism of carbon substrates and utilisation of N-com-

pounds during growth results in pH changes. Growth on compounds like sugars
(e.g. glucose or sucrose) or sugar alcohols (e.g. mannitol) results in acidification
(for fast growers) with pH in unbuffered media commonly falling into the pH 45
range as a result of organic acid production. Growth on organic acids (e.g. fuma-
rate or succinate) can in unbuffered media lead to pH rises to pH 8.5 or higher as a
result of what is essentially a release of hydroxide ions coupled to carboxylate up-
take, usually terminating growth. Growth on amino-acids (e.g. histidine or gluta-
mate) as sole source of carbon results in alkalinisation, partly from hydroxide ex-
change as with the organic acids and partly from release of NH3 into the medium.
Since the quantitative demand for C is much greater than that for N, very high
concentrations of NH3 may be released; for an N-rich amino-acid like histidine,
growth on this amino-acid alone can result in up to 100mM ammonia release.
If growth at a particular pH is to be assessed, adequate pH maintenance through

buffering is essential (Howieson and Ewing 1986). Buffers which generally do
not effect growth rate can be selected from the so-called good biological buff-
ers, though each must be tested at the concentration range required (usually
2030mM).
Each buffer has a particular pH range of usefulness, depending on the pKa value(s)
of the buffer chemical. It is important to recognise that the ability to buffer de-
pends on the ratio of the two forms of the buffer; where the pH is the same as the
pKa, the ratio is one and buffering is maximal. At a pH one unit removed from the
pKa (either up or down) ca. 90% of the buffer is in a form incapable of contributing
to further buffering. Accordingly, this represents an extreme departure from the
pKa for using that buffer effectively. Ideally, the pH at which a buffer is to be used
should be less than one unit from its pKa.
Four buffers which are extremely useful for pH control in media were given in
Chapter3 and these are listed again in Table7.5.
137
7.3 Studying membrane transport systems
Table7.5 B
uffers recommended for studies of rhizobia where pH control is
important.
Chemical name Abbreviation pKa at 25C
3-(Cyclohexylamino)-2-hydroxy-1-propansulfonic acid CAPSO 9.6
N-(2-hydroxyethyl)piperazine-N-(2-ethansulfonic acid HEPES 7.5
2-[N-morpholino] ethansulfonic acid MES 6.1
Homopiperazine-1,4-bis(2-ethansulfonic acid) HOMOPIPES 4.55
7.2.6.4 Physiology in relation to mineral nutrient deficiencies
There are a wide range of specialised techniques and procedures that have been
developed for studying the physiology and growth of rhizobia, and it is beyond
the purpose of this manual to provide full details. The reader is referred to the
rich scientific literature and previous manuals and publications, such as Bergersen
(1980) for guidance and information. Nutritional studies require the use of puri-
fied defined media (Abreu et al. 2012; OHara et al. 1987) and specialised culture
techniques (Cassman et al. 1981; Smart et al. 1984) as discussed in OHara et al.
(1988). In particular, the use of specific anionic or cationic resins to purify media
components of the nutrient being studied, and the addition of metal chelators,
such as EDTA and nitrilotriacetate (NTA) to remove residual nutrient contami-
nants from media and solutions, have proven essential for studies determining
roles of micronutrients in nodule bacteria.
Nodule bacteria must acquire all substrates and nutrients from external sources
and consequently membrane transport is essential in both free-living and sym-
biotic life styles. Studies of membrane transport systems have been essential for
revealing both the common metabolic features and differences rhizobia have de-
veloped to deal with the challenges associated with nutrient acquisition in com-
plex soil and plant environments (Udvardi and Poole 2013). This protocol de-
scribes how to undertake a membrane transport assay using the rapid filtration
technique. It applies to the use of 14C and 3H labelled compounds, although others
can be used. You will need to comply with local regulations about the use of ra-
dioactivity. The assay is based on bacteria accumulating a radioactive compound
inside cells which are then collected on a membrane filter. Radioactivity not in-
side cells is washed off with a salts solution. The filter is then removed and added
to a vial with scintillation fluid.
Preparation
Previous day:
1. Set up 50mL of growth medium with appropriate carbon and nitrogen sourc-
es in a 250mL conical flask. (If strain contains a plasmid then add required
138
antibiotic). Inoculate with 600 L of freshly washed slope culture that is

23days old. This may vary with different rhizobial strains as this protocol was
developed for R. leguminosarum that has a mean generation time on minimal
medium of approximately 4hours;
Day of assay:
2. Fill wash bottle with transport wash.
3. Take 4050 mL of overnight culture and centrifuge cells at 4000 rpm for
20min in a bench centrifuge.
4. Wash cells by re-suspending in the same volume of transport wash. Spin at

4000rpm for 20min.
5. Re-suspend cells in 5 mL of transport wash. Take OD600 of 1/10 dilution.

Re-suspend in transport wash to OD600 1. Take OD600 and note for final
calculation.
6. Leave culture to starve for 1hour in a large shaking water bath at 28C.
7. Prepare 25 M working concentration solutions of the cold (unlabelled)

compounds in the assay. Make up a 0.1M stock (to store at 4C for future as-
says). From this, make up 5mL of 0.5mM solution (25L of 0.1M in 5mL
HEPES buffer pH 7.0).
8. Dilute hot labelled compound to a specific activity of 4.625kBq (0.125Ci)

in 50L. i.e. for six assays you will need 300L hot solute.
Procedure
1. Put tips on three Gilson type pipettes (100200L sample volumes). To the
first pipette, add 25L cold substrate (0.5mM). To the second, add 50L hot
substrate (4.625kBq). The third is a sampling pipette.
2. Assemble a vacuum manifold with a filter disc that will trap bacteria (see be-
low). Filters such as nitrocellulose are used or you can save costs by cutting
your own using Whatman GFF glass fibre filters. Use a standard cork borer to
cut sheets of Whatman GFF into the appropriate-sized discs.
3. Add 225L of 28C TRANSPORT WASH to a sampling tube (Universal or

Falcon as appropriate). Add 200L cells.
4. Start a stop-watch.
5. At 55sec on the clock, add 25L cold substrate (0.5mM) and mix.
6. At 60sec, add 50L hot substrate (4.625kBq) and mix.
7. At 1min 10sec, take 100L of the mix and squirt into a vacuum manifold.
Wash twice with transport wash, remove filter to scintillation tube and cover
with scintillation fluid (this will be the time zero value).
139
8. Take further samples at 2min 10sec, 3min 10sec and 4min 10sec (to give
1min, 2min and 3min samples). Take 50L standard from the sampling tube
without filtration. When this standard is counted, you will be able to check
that the correct level of radioactivity was added to the sample.
9. Count all samples in a liquid scintillation counter. The radioactivity in the bac-
teria enables the calculation of a rate of uptake as follows:
10. nmole transported=[CPM in sample/(standard 2)] 2.5 nmole
11. This can be converted to a rate by dividing by the sample time. Use the OD600
reading to convert the rate to per unit OD600. By preparing a standard curve of
bacterial OD600 versus protein or dry weight it is also possible to express rates
per unit bacterial protein or dry weight.
Notes
There are a number of different vacuum manifolds that can be used. Companies
such as Millipore produce a vacuum manifold that can take several 25mm nitro-
cellulose filters. You can make your own filtration device out of polycarbonate. This
trap contains a Sintered Disc from VWR (10mm diameter, P40 (232/0005/14)). It
is wise to periodically (weekly or monthly depending on use) invert the sintered
glass disc as it becomes clogged with glass fibres. Scintillation vials are typically
polyethylene tubes/ polypropylene cap and are available from a number of suppli-
ers (e.g. 215/0092/02 from VWR). Scintillation fluid should be bought commer-
cially and is available from a number of suppliers.
The above rate calculation is based on 2.5 nmole of cold solute being in each
100L sample. The hot sample will add some extra solute and for greatest pre-
cision this amount should be calculated from the specific activity on the stock
bottle. All radiochemical suppliers will specify KBq mmole1 (or equivalent) and
from this it is possible to calculate how many nmoles of solute are added in the
4.625kBq of hot solute. This amount should be added to the 2.5nmole above e.g.
if 0.1nmole is present in 4.625kBq of hot solute then 2.6 rather than 2.5 should be
used in the above formula. You may wish to alter the starting solute concentration
or the assay time and these will require appropriate alterations to the calculations.
Transport wash
Ingredients:
1.4g K2HPO4
0.36g KH2PO4
0.25g MgSO4.7H2O
0.2g NaCl
1mL Solution A (below)
2mL Solution B
1mL Solution C
1000mL FINAL VOLUME
140
Method:
Dissolve ingredients in slightly less than final volume; adjust pH to 7.0, make up
to 1L.
DO NOT autoclave but keep in dark bottle at 4C. It should store for several weeks.
Rhizobium Solution A
Ingredients:
15g EDTA-Na2
0.16g ZnSO4.7H2O
0.2g Na2MoO4.2H2O
0.25g H3BO3
0.2g MnSO4.4H2O
0.02g CuSO4.5H2O
1mg CoCl2.6H2O[Dissolve 100 mg in 100 mL GDW (Glass Distilled
Water) and add 1mL.]
Method:
Dissolve each ingredient in turn before adding the next.
Make up to 1L with GDW.
Store at 4C.
Rhizobium Solution B
Ingredients:
1.28g CaCl2.2H2O
0.33g FeSO4.7H2O
Method:
Dissolve each ingredient in 50mL water then combine.
Store at 4C for no more than one week.
Rhizobium Solution C
Ingredients:
1g Thiamine hydrochloride
2g D-Pantothenic acid Ca salt
1mg Biotin[Dissolve 100mg in 1000mL GDW and add 10mL.
Store rest of biotin at 20C]
Method:
Dissolve each ingredient in turn before adding the next.
Make up to 1L with GDW.
141
7.4 Final comments
Filter, sterilise and store at 4C
Add aseptically to media at 1mL per litre.
7.4 Final comments
The description of the nodule bacteria is a rapidly expanding field as scientists

continue to examine roles for legumes in sustainable production systems, and
understand legumes in their natural environments. Fundamental to working ef-
ficiently with legumes is being aware of the varied forms of their symbionts, and
recognising them in the laboratory. The unique capacity of rhizobia to harness
inert dinitrogen will mean their physiology and biochemistry will always be the
subject of intensive study. The material in this chapter will expedite these pursuits.
7.5References
Abreu I., Cerda M.E., Perez de Nanclares M., Baena I., Lloret J., Bonilla I., Bolanos L. and Reguera
M. 2012. Boron deficiency affects rhizobia cell surface polysaccharides important for suppres-
sion of plant defense mechanisms during legume recognition and for development of nitrogen-
fixing symbiosis. Plant and Soil 361, 385395.
Bergersen F.J. 1980. Methods for Evaluating Biological Nitrogen Fixation. John Wiley & Sons:
Chichester.
Bloem J.F., Botha W.J., Law I.J. and Steyn P.L. 2002. Colony variation in Sinorhizobium meliloti in-
oculant strain U45. Microbiological Research 157, 283292.
Bournaud C., de Faria S.M., dos Santos J.M.F., Tisseyre P., Silva M., Chaintreuil C., Gross E., James
E.K., Prin Y. and Moulin L. 2013. Burkholderia species are the most common and preferred
nodulating symbionts of the Piptadenia Group (Tribe Mimoseae). PLoS ONE 8(5): e63478.
doi:10.1371/journal.pone.0063478
Boursier P., Hanus F.J., Papen H., Becker M.M., Russell S.A. and Evans H. 1988. Selenium increases
hydrogenase expression in autotrophically cultured Bradyrhizobium japonicum and is a con-
stituent of the purified enzyme. Journal of Bacteriology 170(12), 55945600.
Brenner D.J., Staley J.T. and Krieg N.R. 2005a. Classification of prokaryotic organisms and the con-
cept of bacterial speciation. Pp 2732 in Bergeys Manual of Systematic Bacteriology. Volume
2 The Proteobacteria, Part A Introductory essays, ed. by G. Garrity. Springer: New York, USA.
Brenner D.J., Krieg N.R. and Staley J.T. 2005b. Bergeys Manual of Systematic Bacteriology. Volume
2 The Proteobacteria, Part A Introductory essays, ed. by G. Garrity. Springer: New York, USA.
Cassman K.G., Munns D.N. and Beck D.P. 1981. Growth of Rhizobium strains at low concentrations
of phosphate. Journal Soil Science Society America 45, 520523.
De Meyer S.E., Cnockaert M., Ardley J.K., Trengove R.D., Garau G., Howieson J.G. and Vandamme
P. 2013a. Burkholderia rhynchosiae sp. nov., isolated from Rhynchosia ferulifolia root nodules.
International Journal of Systematic and Evolutionary Microbiology 63, 39443949.
De Meyer S.E., Cnockaert M., Ardley J.K., Maker G., Yates R., Howieson J.G. and Vandamme P.
2013b. Burkholderia sprentiae sp. nov., isolated from Lebeckia ambigua root nodules. Interna-
tional Journal of Systematic and Evolutionary Microbiology 63, 39503957.
De Meyer S.E., Cnockaert M., Ardley J.K., Van Wyk B-E., Vandamme P. and Howieson J.G. 2013c.
Burkholderia dilworthii sp. nov., isolated from Lebeckia ambigua root nodules. International
Journal of Systematic and Evolutionary Microbiology 64, 10901095.
Dreyfus B., Garcia J.L. and Gillis M. 1988. Characterization of Azorhizobium caulinodans gen. nov.
sp. nov., a stem-nodulating nitrogen-fixing bacterium isolated from Sesbania rostrata. Interna-
tional Journal of Systematic and Bacteriology 38, 8998.
Elliott G.N., Chen W.M., Chou J.H., Wang H.C., Sheu S.Y., Perin L., Reis V.N., Moulin L., Simon
M.F., Bontemps C., Sutherland J.M., Bessi R., de Faria S.M., Trinick M.J., Prescott A.R., Sprent
142
7.5References
J.I. and James E.K. 2007. Burkholderia phymatum is a highly effective nitrogen-fixing symbiont
of Mimosa spp. and fixes nitrogen ex planta. New Phytologist 173(1), 168180.
Evans W.R., Fleischman D.E., Calvert H.E., Pyati P.V., Alter G.M. and Subba Rao N.S. 1990. Bacte-
riochlorophyll and photosynthetic reaction centers in Rhizobium sp. strain BTAi1. Applied and
Environmental Microbiology 56, 34453449.
Garau G., Yates R.J., Deiana P. and Howieson J.G. 2009. Novel strains of nodulating Burkholderia
have a role in nitrogen fixation with papilionoid herbaceous legumes adapted to acid, infertile
soils. Soil Biology & Biochemistry 41, 125134.
Graham P.H. 1963. Vitamin requirements of root nodule bacteria. Journal of General Microbiology
30, 245248.
Graham P.H., Sadowsky M.J., Kersters H.H., Barnet Y.M., Bradley R.S., Cooper J.E., De Ley D.J.,
Jarvis B.D.W., Roslycky E.B., Strijdom B.W. and Young J.P.W. 1991. Proposed minimal standards
for the description of new genera and species of root-and stem-nodulating bacteria. Interna-
tional Journal of Systematic Bacteriology 41, 582587.
Gevers D., Cohan F.M., Lawrence J.G., Spratt B.G., Coenye T., Feil E.J., Stackebrandt E., de Peer Y.V.,
Vandamme P., Thompson F.L. and Swings J. 2005. Re-evaluating prokaryotic species. Nature
Reviews Microbiology 3, 733739.
Gyaneshwar P., Hirsch A.M., Moulin L., Chen W.M., Elliott G.N., Bontemps C., Estrada-de Los San-
tos P., Gross E., Dos Reis F.B., Sprent J.I., Young J.P. and James E.K. 2011. Legume-nodulating
betaproteobacteria: diversity, host range, and future prospects. Molecular Plant Microbe Inter-
actions 24(11), 12761288.
Howieson J.G. and Ewing M.A. 1986. Acid tolerance in the Rhizobium meliloti Medicago symbio-
sis. Australian Journal of Agricultural Research 37, 5564
Howieson J.G. Ewing M.A. DAntuono M. 1988. Selection for acid tolerance in Rhizobium meliloti.
Plant and Soil 105, 179188.
Howieson J.G., De Meyer S.E., Vivas-Marfisi A., Ratnayake S., Ardley J.K. and Yates R.J. 2013. Novel
Burkholderia bacteria isolated from Lebeckia ambiguaa perennial suffrutescent legume of the
fynbos. Soil Biology & Biochemistry 60, 5564.
Jordan D.C. 1984. Genus I. Rhizobium. Pp. 235242 in Bergeys Manual of Systematic Bacteriology,
1st Ed., Vol. 1, ed. by Kreig and Holt. The Williams & Wilkins Co.: Baltimore.
Kampfer P. and Glaeser S.P. 2012. Prokaryotic taxonomy in the sequencing erathe polyphasic ap-
proach revisited. Environmental Microbiology 14(2), 291317.
Konstantinidis K. and Tiedje J.M. 2005a. Genomic insights that advance the species definition for
prokaryotes. Proceedings National Academy Science USA 102, 256772.
Konstantinidis K.T. and Tiedje J.M. 2005b. Towards a genome-based taxonomy for prokaryotes.
Journal of Bacteriology 187, 62586264.
Lindstrm K. and Young J.P. 2011. International Committee on Systematics of Prokaryotes Subcom-
mittee on the taxonomy of Agrobacterium and Rhizobium: minutes of the meeting, 7 Septem-
ber 2010, Geneva, Switzerland International Journal of Systematic and Evolutionary Microbiol-
ogy 61(12), 308993.
McComb J.A., Elliot J. and Dilworth M.J. 1975. Acetylene reduction by Rhizobium in pure culture.
Nature (London) 256, 409410.
Martens M., Delaere M., Coopman R., de Vos P. Gillis M. and Willems A. 2007. Multilocus sequence
analysis of Ensifer and related taxa. International Journal of Systematic and Evolutionary Micro-
biology 57, 489503.
Moulin L., Munive A., Dreyfus B. and Boivin-Masson C. 2001. Nodulation of legumes by members
of the -subclass of Proteobacteria. Nature 411, 948950.
OHara G.W. and Daniel R.M. 1985. Rhizobial denitrification: a review. Soil Biology & Biochemistry
17, 19.
OHara G.W., Franklin M. and Dilworth M.J. 1987. Effect of sulfur supply on sulfate uptake, and
alkaline sulfatase activity in free-living and symbiotic bradyrhizobia. Archives of Microbiology
149, 163167.
OHara G.W., Boonkerd N. and Dilworth M.J. 1988 Mineral constraints to nitrogen fixation. Plant
and Soil 108, 93110.
Poole P.S., Hynes M.F., Johnston A.W.B., Tiwari R.P., Reeve W.G. and Downie J.A. 2008. Physiology
of root nodule bacteria. Pp 241292 in Nitrogen fixing Legume Symbioses, ed. by W.E. Newton
and M.J. Dilworth. Springer Sciences.
Rogel M.A., Ormeno-Orrillo E. and Martinez Romero E. 2011. Symbiovars in rhizobia reflect bacte-
rial adaptation to legumes. Systematic and Applied Microbiology 34, 96104.
143
7.5References
Smart J.B., Robson A.D. and Dilworth M.J. 1984. A continuous culture study of the phosphorus nu-
trition of Rhizobium trifolii WU96, Rhizobium NGR234 and Bradyrhizobium CB756. Archives
of Microbiology 140, 276280.
Stackebrandt E., Frederiksen W., Garrity G.M., Grimont P.A.D., Kampfer P., Maiden X., Nesme
M.C.J., Rossello-Mora R., Swings J., Truper H.G., Vauterin L., Ward A.C. and Whitman W.B.
2002. Report of the ad hoc committee for the re-evaluation of the species definition in bacteriol-
ogy. International Journal of Systematic and Evolutionary Microbiology 52, 10431047.
Sullivan J.T., Brown S.D., Rogers Yocum R. and Ronson C.W. 2001. The bio operon on the acquired
symbiosis island of Mesorhizobium sp. Strain R7A includes a novel gene involved in pimeloyl-
CoA synthesis. Microbiology 147, 13151322.
Sylvester-Bradley R., Thornton P. and Jones P. 1988. Colony dimorphism in Bradyrhizobium strains.
Applied and Environmental Microbiology 54, 10331038.
Udvardi M. and Poole P.S. 2013. Transport and metabolism in legume-rhizobia symbioses. Annual
Review of Plant Biology 64, 781805.
Vandamme P., Pot B., Gillis M., De Vos P., Kersters K. and Swings J. 1996. Polyphasic taxonomy; a
consensus approach to bacterial systematics. Microbiology and Molecular Biology Reviews 60,
407438.
Watson R.J., Heyes R., Martin T. and Savard M. 2001. Sinorhizobium meliloti cells require biotin
and either cobalt or methionine for growth. Applied and Environmental Microbiology 67,
37673770.
Willems A., Coopman R. and Gillis M. 2001. Comparison of sequence analysis of 16S23S rDNA
spacer regions, AFLP analysis and DNADNA hybridizations in Bradyrhizobium. International
Journal of Systematic and Evolutionary Microbiology 51, 623632.
Young J.P.W., Downer H.L. and Eardly B.D. 1991. Phylogeny of the phototrophic rhizobium strain
BTAil by polymerase chain reaction-based sequencing of a16S rRNA gene segment. Journal of
Bacteriology 173, 22712277.
144
8.1Introduction
CHAPTER8
Field experiments with

rhizobia
R.J. Yates, R. Abaidoo and J.G. Howieson
8.1Introduction
In this chapter, we discuss methodologies to select rhizobial strains for their ad-
aptation to field conditions as this is an important trait when developing inocu-
lant-quality strains. Techniques to assess nitrogen fixation per se are described in
Chapters 5 and 10, and traits related to manufacturing in Chapter9.
For many legume symbioses, the greatest challenge in developing inoculants is to

select strains that confer consistent nodulation upon the legume in the target soil
environment. Edaphic stresses after inoculation, such as dry heat and acidity, can
be detrimental to rhizobial survival, so selecting strains tolerant of these stresses
is very useful. Some legumes regenerate annually from hard (impermeable) seeds
in the soil and therefore, like their rhizobia, are only sown once in several dec-
ades. Hence, assessment of the ability of strains to persist in the field environment
through several seasons is required. This trait cannot be reliably predicted without
field experimentation (Howieson et al. 2000). Many legume evaluation programs
over the years have been compromised because inoculants suitable to the field
conditions have not been available.
8.2 Defining the aim of the field evaluation
It is important to clearly define the aim of the activity as this will influence the site
selection and the design of the field experiment. Examples of common aims in
field experiments are to select rhizobial strains that:
are competitive with resident (but less-effective) nodulating bacteria (see need
to inoculate experiment in Chapter1)
145
8.3 Experimental design
have limited persistence (these strains would rely heavily on a successful in-
oculation procedure but ultimately provide the opportunity to be displaced if
a superior strain were to be developed)
colonise difficult soils and persist through several seasons, such that legumes
regenerating from hard seed banks after cropping can promptly nodulate
tolerate difficult soil conditions, such as dry and hot soil, long enough to nod-
ulate the host
have optimal nodulation performance across different delivery systems (e.g.

achieve high cell number, long shelf life and compatibility with other bacteria
in multiple strain inoculants).
8.3 Experimental design
For the data from field experiments to be statistically robust, several fundamental
aspects of experimental design must be followed.
Treatments must be randomly assigned to plots.

Treatments must be replicated.
Trial layout must take account of (and try to minimise) natural variation in
the site.
Controls must be included in the treatments.

A simple design that is often applicable to field experiments is the randomised
block. In this form of design, treatments are allocated into replicates, which are
sown in blocks, and the blocks are arranged such that any evident variation in the
field is exposed to a discrete replicate. An Excel spreadsheet can be designed to
record the treatment randomisation and allocate these to plots, as shown in Fig-
ure8.1.
Figure8.1 A randomised complete block layout for 10 treatments (A-J) with four replicates, to
guide the layout of a field experiment as shown in Figure8.2.
146
8.4 Site selection
8.4 Site selection
One of the most critical aspects of field work, often neglected, is to select appro-
priate sites for the experiment. The key elements of site selection are to:
maintain relevance to the ultimate aim of the experiment (e.g. an understand-

ing of farming systems and soil types likely to be encountered by the symbiosis)
ensure that previous management of the site (e.g. herbicide history) will not
compromise rhizobia or legume performance
reduce variability in plant or microbial performance across the plots

choose sites with suitable texture to avoid damage to the roots and loss of
nodules when collecting data (i.e. avoid fine textured, hard-setting clay soils
if possible)
minimise rhizobial contamination through water movement and soil erosion

by choosing sites that are on a slope gradient of less than 1:100 and where
good water drainage is retained
avoid sections of the field that may have been subjected to heavy machinery or
animal traffic as resulting compaction may result in uneven plant performance
avoid specific locations in the field where header-trails, burnt residues, hay,
fertiliser, gypsum or lime have been accumulated, as residues of these can af-
fect the consistency of plant performance
avoid sites where top soil depth is variable.
8.4.1 Land and seed preparation

Particular attention should be given to preparing the site in such a way that there
is a minimal requirement for physical intervention (e.g. human, animal or ma-
chinery movement) once the experiment has been sown. This is essential to re-
duce cross-plot contamination. The site must be prepared to be agronomically
suitable for growth of the legume.
Weeds should be controlled in the season prior to the experiment, and again
before planting, by herbicide or cultivated by machinery, or by manual raking.
The site should be fertilised with all necessary macro- and micronutrients,
except nitrogen.
The site should be fenced to ensure that there is no traffic movement on the
site, particularly from farm or wild animals.
Sites that have produced cereal crops (e.g. wheat in temperate/Mediterranean cli-
mates, sorghum or maize in subtropical areas) in the preceding season are useful,
as they often have reduced weed burden, may homogenise the soil and deplete
soil nitrogen. However, the historical herbicide regime must be clear. Breakdown
147
8.4 Site selection
of the sulfonyl-urea herbicides occurs more rapidly in acid soils, wet soils and
at higher temperatures (Koopman et al. 1995). There is mounting evidence that
minute residues of these herbicides can reduce legume nodulation (Figure 5.9)
and this can be problematic after drought (Farquarson 2010).
If biological yield of the legume is to be an important parameter, then the depth

of the soil profiles and its variability can be gauged from soil sampling across a
grid, or by using ground-penetrating radar (GPR) to map changes in soil texture
with depth. Data from GPR can be a useful covariate for statistical analysis of yield
(Burgos et al. 2009).
High-quality seed needs to be sourced for experiments, and scarification of hard

seeds undertaken to ensure they germinate (Chapter5). A germination test should
precede the trial; many field experiments have failed due to poor-quality seed.
Figure8.2 Land preparation and layout; in this example, the site is relatively flat, uniform and
fenced, with individual plots marked with pegs.
8.4.2Matching site characteristics to the aim of the

experiment
There are a few key considerations when selecting the site for a field experiment.
For low numbers of background rhizobia. Avoid sites where the inoculated
legume has previously been sown. An MPN estimate (Chapter6) prior to sow-
ing the experiment is often indicative, but for some symbioses even low back-
ground counts (<20 cells of infecting rhizobia per gram of soil) can be very
148
8.4 Site selection
competitive with seed applied inoculants, and thus remove any response to
inoculation (Thies et al. 1991).
For high numbers of background rhizobia. To research selective nodulation,

the impact of different inoculant carriers or the capacity to introduce superior
strains on seed, seek sites where the nodulating rhizobia have previously been
established. It is possible to introduce rhizobial strains on unrelated legumes
the previous season (e.g. Ornithopus-nodulating rhizobia may be introduced
as inoculants on Lupinus; Pisum rhizobia may be introduced in association
with Vicia, Lathyrus or Lens; many strains of bradyrhizobia can be introduced
as inocula on Macroptilium purpureum) to avoid plant disease pressures from
continuous cultivation of the legume under investigation.
To assess abiotic stress. The range of stressful soils to be encountered (e.g.

waterlogged, dry or desiccating, acid, alkaline, low clay content) should form
the basis of the selection criteria. If seeking acid tolerance, then be aware that
acidity combined with low clay content and low organic matter content is
more stressful than low pH in isolation (Howieson and Ballard 2004). Acidity
can also be associated with toxicity due to soluble Al3+ and Mn2+ ions (Helyar
and Porter 1989). If resources permit, a control set of plots should be estab-
lished where the stress is ameliorated (e.g. by liming an acidic soil).
To assess biotic stress. If acting directly upon the nodule bacteria, the agents
of stress, such as predators or antibiotic-producing agents, need to be encour-
aged, perhaps by reduced tillage. The level of organic matter and the microbial
biodiversity of the site will be important parameters. Biotic stress associated
directly with nodulation or nodule function may need prior history of the
legume to encourage appropriate levels of the stress parameter. For example,
nodule predation by sitona weevil larvae is very specific. Medicago but not Tri-
folium nodules are predated by Sitona discoideus in Australia. A healthy stand
of medic would need to be developed, and the Sitona encouraged to colonise
the site prior to introducing an experiment to control the effects of this biotic
stress.
To assess strain persistence through rotations. Annually regenerating pas-

ture legumes are iconic in Australian agriculture. If selecting rhizobia that need
to persist in the absence of the host legume, ensure the field site is suitable for
both the target legume and the rotational species, and that a long-term lease
of the site is possible. For example, rhizobia that effectively nodulate Biser-
rula pelecinus often need to persist through three to five successive crops (e.g.
wheat, canola, barley, wheat, canola) before the legume (regenerating from
hard seed buried in the soil) requires them for nodulation (Loi et al. 2005).
149
8.5 Selecting the appropriate carrier and adhesive for application of rhizobia to the seed
8.5Selecting the appropriate carrier and adhesive

for application of rhizobia to the seed
The aim of many field experiments with rhizobia is to introduce a spectrum of

strains, or species, into the soil to select one or two that are well adapted to the
conditions. A decision must be made on which form of inoculant carrier is to be
used as it may affect the outcome. The carrier supports live cells of rhizobia and
protects them from stress, such as desiccation that can lead to death (Vincent et al.
1962, see Figure8.3). There are many inoculant carriers available (Table8.1) and
the characteristics of these are discussed in more detail in Chapter9. However, it
is important that the carrier selected must deliver a known quantity of live cells to
the target legume under the prevailing conditions.
Figure8.3 Unthrifty, yellow and poorly-nodulated seedlings of Lebeckia ambigua (left) contrast
with stronger, nodulated plants at six weeks of age (right). The poor growth and yellowing was due
to N deficiency as a result of the death of the peat-based inoculant after application to the seed,
and prior to nodulation in this sandy soil (Howieson et al. 2013).
Table8.1 C
arriers available to deliver rhizobial strains in field experiments (see also
Chapter9)
Carrier Characteristics
Finely-ground organic substrates, such Moist (3045% water) can deliver high cell count (109 cells/g) but
as peat, coir dust, filter or mud residues must be applied to wet soils as death rate is 90% per day, and
applied to the legume seed coat as a must not contact fungicides or other seed-applied chemicals.
slurry with an adhesive (Table8.2)
Peat-based granulessown alongside As above, but can be used where the inoculant must be separated
seed at planting from the seed to avoid chemical toxicity.
Clay granulessown alongside seed at Dry (510% water) can deliver lower cell count (105107 cells/g) but
planting, or mixed with the fertiliser stable with slow death rate, and can be used to sow legumes into
dry soil before rains.
Liquid inoculantssprayed onto seed or Broth cultures stabilised in their growth phase (at approx. 109
in furrow cells/mL) by low temperature or nutrient exhaustion, but generally
less robust than peats or clays.
Freeze dried inoculants Very high count (1012 cells/mL) that are diluted before sowing and
used in the same manner as liquid inoculants.
Pre-coated seedsusually a peat-based Can deliver stable high count live cells of Sinorhizobium (103/seed)
matrix applied to seed with polymers to but have the same death rate as peat inoculants for other rhizobial
aid longevity species (Deaker et al. 2004)
150
8.6 Adhesives for application of inoculants to legume seeds
Any of these carriers can establish the experiment in small plots sown by hand (as
swards or lines) but all may contaminate machinery when used to sow larger plots
(>20m2). Dry clay carriers have a large dust component that can cause contami-
nation in machine parts, such as the seed bin or tubes that carry the inoculants to
the sowing tynes. Peat carriers can loosen from the surface of seed as they dry, but
coating with lime or dolomite can minimise this. When coating seeds, adhesive
concentrations can be increased to ensure high pellet integrity and minimise for-
mation of dust during sowing. To minimise contamination, machine parts may be
swabbed with ethanol (70%) and fertilisers can be passed through the machinery
between rhizobial treatments to dislodge dust.
8.6Adhesives for application of inoculants to

legume seeds
Several additives are available to increase adherence of rhizobial carriers to seed,

to protect them from desiccation and toxic chemicals, and to allow a longer win-
dow for sowing (Table8.2).
Table8.2 Adhesives for applying rhizobial carriers to seed for field experiments
Adhesive Form Characteristics
Sugars Sucrose, maltose often used as 10% Not as effective as some other
(w/v) solutions adhesives, but can aid survival of
rhizobia.
Natural Gum Arabic, xanthan gum at 25% Usually very effective, but can be
polymers (w/v) solutions variable. Require pre-preparation to
solubilise.
Semi- methyl cellulose used at 12% (w/v) Prepared the day before use by
synthetic dissolving in hot water. Avoid
polymers preparations with fungicides.
Synthetic polyvinylpyrrolidone, polyvinyl Have been shown to assist cell
polymers acetate, vinylacetate/ survival. Often used at 1020% but
vinylpyrrolidone co-polymer used at may become expensive if used on a
1020% (w/v) broad scale.
The properties of these adhesives or additives can be species-specific, hence some

pre-experimentation is recommended to assess their efficacy against the rhizobial
species under evaluation.
8.7General procedures for application of

inoculants
1. Grow rhizobial strains to be investigated on appropriate medium (Chapter3)

and introduce strains individually into the chosen carrier. Mix well and incu-
bate if required.
151
8.7 General procedures for application of inoculants
2. Enumerate inoculant product after period of maturation (Chapter6).
3. Ensure similar numbers of each strain are compared by varying the amount of
carrier applied to seed or soil. For seed-applied inoculants, refine the inocula-
tion procedure by testing trial batches.
4. For peat and similar carriers, mix inoculant with a pre-sterilised adhesive (at
a rate recommended for the product) in a sterilised vessel, such as a 20mL
plastic McCartney vial for small seeded legumes or a 500mL beaker for larger
seeds. Use a new vial for each rhizobial strain and sterilise utensils (spatula,
spoon) in ethanol and flame (allow to cool) between each strain.
5. Surface-sterilise seed (Chapter5).
6. Mix seed thoroughly with the inoculant /sticker slurry in the vial to deliver
approximately commercial rates of inoculant per gram of seed (see below and
Chapter9).
7. It is important to ensure the seeds are not so moist after inoculation that they
clump together, as this makes sowing very difficult. Surface-inoculated seeds,
particularly small pasture seeds, may be dried and separated into single units
by application of any benign substance, such as lime, clay or talc to the vial
with gentle stirring or shaking.
8. Weigh inoculated seed into aliquots for each replicate and arrange packets
into sowing order. Small seed envelopes are useful.
9. If using a granular inoculant, weigh granules into aliquots for each replicate,
again taking care to ensure similar numbers of cells are compared for each
strain and avoid contamination. Add appropriate weight of surface-sterilised
seed to the packet if granules and seed are to be sown in the one operation.
10. Store aliquots of inoculated seed immediately in a secure cool, dark and dry
area to ensure survival of inoculant.
11. Sow as soon as possible (unless measuring cell deaths over time) with chosen
apparatus and approach (see below).
12. In the laboratory, keep a sample of inoculated seed and/or carrier and store
at 15C for later assessment of rhizobial numbers at time of inoculation
(Chapter6).
13. In the field, collect a sample of inoculated seed and/or carrier at the time of
sowing, store in cold box and enumerate when back at the laboratory to esti-
mate rhizobial numbers at the time of sowing.
152
8.8 Laying out plot dimensions for field experiments
8.8Laying out plot dimensions for field

experiments
Determine the experimental design and calculate the number of plots per repli-
cate, the number of replicates, their shape and the dimensions of each individual
plot.
Develop a printed trial plan showing the plots and the preferred layout (Figure8.1).
At the field site, calculate the full dimensions of the trial, survey the site and select
a space sufficient for the trial.
1. Using a tape measure or notched string, lay out a baseline for one dimension
of the trial site.
2. Place pegs at intervals along the tape, appropriate to the plot size, leaving a
buffer between each plot.
3. Select one corner of the baseline and form a right angle (using an optical
square, or calculate using a 345 triangle).
4. Lay out the plot sizes along this second dimension to delineate blocks or rep-
licates. Peg for all replicates.
5. Ensure space for a buffer is provided between each replicate to minimise the
risk of contamination and to allow the passage of sowing and spraying ma-
chinery, and operators.
6. Repeat the right angle from the other end of the baseline and peg all replicates.
7. Lay the tape or notched string along a line parallel with the baseline delineat-
ing the second replicate and place pegs to form each individual plot.
8. Repeat by moving the tape or string in parallel blocks for the number of rep-
licates required.
9. Number each plot in each replicate on the plot pegs with a permanent marker.
8.8.1 Sowing swards

In the simplest scenario, inoculated seed can be sprinkled by hand evenly across a
small sward (e.g. dimensions of 2m 2m). A significant advantage of hand-sown
plots is that hygiene can be optimised, whereas for machine-sown plots, strains
can transfer from plot to plot through dust or loose carrier, adhering to machin-
ery parts.
1. Calculate the weight of seed required, surface sterilise, inoculate and weigh
out the seed required for each plot into a small envelope as outlined in Sec-
tion 8.7.
2. Lay out the field experiment as described in Section 8.8 (Figure8.4).
153
3. Select the first treatment and mark its plot in each replicate*.
4. Distribute seeds evenly over the plot surface in each replicate.
5. Gently cover by light raking (treatment by treatment) if the species needs to be

buried to assist germination.
6. Rinse hands and any implements in 70% ethanol and repeat for all subsequent
treatments.
7. For large seeds to be sown in lines, trenches can be established to the required
depth in the sward with a simple tyned implement, if not provided earlier dur-
ing the land preparation.
* The un-inoculated treatment is often sown first, with additional un-inoculated

controls sown last.
Figure8.4 Hand-sowing seed into lines in small plots delineated by strings, and
covering with a rake. Note that both scientists are standing in the buffers between
plots.
154
8.8.2 The cross-row technique

This technique has been very successful in selecting inoculant-quality strains. It
requires two seasons to complete. The cross-row approach identifies strains that
are saprophytically competent in the target soil (Chatel and Parker 1973) and
that can colonise the soil away from the point of inoculation with some reliability
(Howieson and Ewing 1986; Yates et al. 2005).
Year 1
1. Plots are sown as 1.5m lines, with seed sown at a rate of 0.5g/m for small
seeds and 20g/m for large seeds. Calculate the weight of seed required, surface
sterilise and inoculate as described in Section 8.6.
2. Weigh out the seed required for each plot (line) into a small envelope and
place the envelopes in treatment order in a tray. Store in cold room.
3. Experimental design should be as randomised blocks, as described in Section

8.7. Individual lines are separated by 1m buffers and banks are separated by
1.5m buffers.
4. After pegging the plots, make a shallow indentation (2cm deep) in the soil,
usually with a length of angle iron or a vegetable planter (Figure 8.5). It is
helpful if a length of chain is welded to each end of the iron to facilitate lifting
between plots.
5. Lay out the envelopes containing the appropriate treatment for each plot and
cross check with the randomisation chart (Figure8.6).
6. To sow the seed, open the top flap or alternatively tear the bottom off the ap-
propriate envelope then place a crease in the main surface of the envelope to
guide the seed out.
7. Apply a gentle tapping motion to the envelope with the forefinger and encour-
age the seed to flow from the envelope evenly along the row. If the sowing is
not uniform, seed may be spread along the row using the tip of the envelope,
which is then discarded.
8. Cover the seeds by collapsing the side of the indentation with the envelope or
an inverted rake, making sure not to make contact with the seed.
9. Allow the plants to grow through the season. Above-ground production pa-
rameters may be scored or measured, if required, but the root system and soil
should not be disturbed. If a first year assessment of nodulation is required,
samples can be taken from plants at the end of the row.
10. Annual legumes are allowed to senesce naturally between the wet seasons
while crop legumes are carefully cut at 1 cm height and herbage removed,
dried and weighed. Perennial legumes may be sprayed with herbicide at the
completion of the first year growing season.
155
Figure8.5 Angle iron with chains attached to aid lifting (above left) and vegetable planter (right)
have been used to make indentations into soil for a cross-row trial.
Figure8.6 Packet envelopes containing seed laid out ready for sowing into the
furrow.
Year 2
1. Allow weeds to germinate and control these with herbicide without disturb-
ing the soil.
2. Package the seed for each cross-row (pasture 0.4g/m, pulses 8g/m) into indi-
vidual envelopes. Allow for 10% extra packets.
3. Before sowing, with a short (40cm) piece of angle iron, create the indenta-
tions for the cross rows (Figure8.7). Equipment to prepare the indentation
in the soil, to a depth of 12cm, must be cleansed in 70% ethanol between
treatments.
156
4. Sow un-inoculated seed (surface sterilised) across the original row in a per-
pendicular pattern at two or three points, with the cross row extending 20cm
each side of the original row (Figure8.8).
5. Ensure a relatively constant number of seeds are placed into each plot. If a
cross row is accidentally sown unevenly, seed cannot be moved with the tip
of the envelope as described in year 1, as this may move rhizobia laterally. It
is better to apply extra seed to cross rows under these circumstances, so it is
necessary to pack extra packets of fresh seed for this eventuality.
Figure8.7 Making the cross row furrow. On this coarse textured soil, the sand
has been pre-wet to aid this step, but the implement must be cleaned in ethanol
between plots.
Figure8.8 Sowing the un-inoculated seed, taking care not to disturb the soil
laterally.
157
8.8.3 Data collection from a cross-row experiment

1. In the first year of the cross-row, it is particularly successful if you see plots of
yellow plants from un-inoculated treatments (Figure8.9) but the major pa-
rameter to assess the colonisation of the soil by rhizobia is the proportion of
nodulated seedlings in each of the cross rows in year 2.
2. After allowing sufficient time for seedlings to nodulate (six to10 weeks) divide
the cross-rows into sections (010cm and 1120cm) from the original row as
shown in Figure8.10.
3. Excavate plants from each section using a plastic T to guide the sections
as shown in Figure8.10. The whole 10cm section can often be lifted in one
spade-full to a depth required to access the main part of the root section. For
small seeded legumes, this is often approximately 15cm deep after six weeks
of growth.
4. Shake the plants to remove excess soil then bag for transportation. Sections
(e.g. the 010cm group) within a plot may be bulked and then plants select-
ed randomly from them for assessment. At least 15 plants from each section
should be evaluated (see Section 8.10.1).
5. Strains that have persisted over the dry period then achieved nodulation in
the cross-rows in the second year (particularly in the 1120cm sections) are
considered suitable for further development.
Figure8.9 The first year of a cross-row experiment with Pisum sativum and R. leguminosarum bv
viceae after five weeks of growth. The un-inoculated treatment (left row) lacks nitrogen and is pale
in comparison to the surrounding inoculated rows, and to the inter-row weeds.
158
12 m
10 cm 10 cm
Inner
Outer Outer
10 cm 10 cm
Figure8.10 Schematic
diagram of the cross-row
Year 1 Year 2 technique describing the
layout in years 1 and 2.
Where strains have colonised the soil away from the original placement, and sur-
vived over the dry period, good nodulation should be reflected in strong legume
growth across the cross-row, as illustrated in Figure8.11.
Figure8.11 Possible outcomes

in the second season of a cross-
row experiment with Medicago
polymorpha and Sinorhizobium
medicae. In the top plate, the
rhizobial strain has colonised
at least 20cm either side of the
original row (along the handle
of the poly pipe). In the middle
plate, the strain has survived
in its original position but not
moved laterally. In the bottom
plate, there has been no survival
and hence plant growth is greatly
compromised by N deficiency.
Note the high seeding rate
required to generate sufficient
plants for nodulation assessment
in this example.
159
8.9 Plots established with precision seeding machinery
8.9Plots established with precision seeding

machinery
Where the aim of the experiment is to investigate response to inoculation in an

on-farm scenario, it may be unrealistic to sow plots by hand. A reliable estimate
of seed production requires a larger plot size than can be uniformly sown by hand,
and destructive sampling for nodule parameters and early growth can reduce the
amount of material available. Further, many grain legumes (e.g. soybean, com-
mon bean, lupin) must be sown at 5cm depth or deeper which is sometimes dif-
ficult to achieve uniformly by hand. Under these circumstances, machine-drawn
precision seeders may be the best option for establishment of the plots. Precision
seeding machines (Figure8.12) have been designed to sow small plots (525m
length) uniformly, with defined seeding rates per hectare and defined distances
between seeds and between rows to ensure plant to plant competition is uniform.
Machines typically have a sowing width of 1.21.8m, with row spacing around
1525cm apart. If replicates are to be sown in blocks, ensure there is sufficient
buffer area between blocks for turning the machinery.
Figure8.12 A
precision seeding
machine for
sowing larger
plots, such as
for grain legume
experiments.
8.9.1 Inoculation of large-seeded legumes for plots to be

sown by machine
For sowing rates of 100kg per ha, standard plots sown with a precision seeder (e.g.
10m long) require up to 100g of seed.
1. Surface sterilise the bulk seed (Chapter5).
2. Split seed into treatments for each plot and pack into cardboard boxes (e.g.
unused milk containers) or waterproofed paper bags.
3. For inoculation with a liquid (e.g. peat slurry or broth) it is possible to inject
the required volume (15mL) into the container using a disposable syringe.
160
8.10 Data collection and assessment of nodulation
4. For granular inoculants, a measured amount (usually at 10kg/ha) can be de-

livered to each container, calculated for the plot size.
5. After inoculation, shake the container to ensure even distribution of the in-
oculant, allow to dry enough that the individual seeds flow, and then place in
sowing order.
6. As for hand-sown experiments, each inoculant treatment must be prepared in a

separate container and all equipment should be sterilised between treatments.
7. Once inoculation has been completed, sowing should occur within 24hrs.
8. Mark out and peg plots as described in Section 8.7.
8.9.2 Hygiene at sowing

As for hand-sown plots, all replicates for a given treatment should be sown be-
fore moving to the next treatment. However, for the un-inoculated control, four
replicates may be sown first to enable assessment of nodulation and plant growth
achieved by the resident soil population of rhizobia, and then additional un-in-
oculated treatments may be sown last. The latter plots can aid in understanding
whether there has been contamination in the sowing procedure, by assessing nod-
ule occupancy (Chapter11). Thus, if logistics permit, eight replicates of the un-
inoculated control can be sown. The machinery should be thoroughly cleansed
between treatments to avoid transfer of inoculant strains between plots. Parts
should be wiped with ethanol (70% v/v) and allowed to air dry. Ethanol can also
be washed down sowing tubes and collected in a bucket for reuse. Take care when
swabbing any Perspex sections of the seeder with ethanol as this can cause parts
to crack. Fertiliser granules can also be dropped through the seeding tubes after
each treatment and before cleaning with ethanol to dislodge material that may
have been caught in the machine.
Field experiments are large undertakings so it is important to gather the appro-

priate information from them. This does not infer the need to gather maximum
information but rather to understand the purpose of the experiment and sample
accordingly. For example, biomass accumulation is not particularly relevant in
the second year of the cross-row experiments described previously, as these are
designed to assess the adaptation of rhizobial strains to the soil.
8.10.1 Nodule counts and scores

Nodulation is evident within three weeks for many symbioses, but for field experi-
ments it is advisable to wait four to eight weeks before the first assessment of nod-
ulation, depending upon the temperature. Representative samples for nodulation
161
assessment (e.g. 40 plants) can be removed from random positions within the
plots, avoiding the edges. In the cross-row experiments, all plants in the 010cm
and 1120 cm region are excavated. Twenty to 30 plants are randomly scored
from both areas and the average score recorded. In machine-sown plots, a second
nodulation assessment may be taken after 12 weeks to give an indication of the
dynamics of nodulation. Nodule abundance may be either counted or scored (see
Figure8.13). Scores can be analysed statistically if they conform to rules of conti-
nuity in data. For nodule scores, continuous data is acquired if a number of plants
are sampled in each replicate, and the mean score then developed. Ultimately, the
decision on what data to accumulate must be made based on the quality of the
data required and the resources available.
Nodule score is closely correlated with nodule weight, which can be assessed if
nodules are picked from the root system, cleaned, dried then weighed. Where soil
cannot be fully removed from the nodules, it may not be possible to generate ac-
curate weight data.
0 0.5 1 2 3 4 5 6 7 8
Absent Ineffective Rare Scarce Moderate Adequate Ample Abundant Very Extremely
(Effective) Abundant Abundant
No nodules White 05 small 610 small 1120 2140 < 40 10 large Crown Crown
ineffective pink pink small pink small pink small pink pink nodulation nodulation
nodules nodules nodules and/or and/or and/or nodules <5 mm 5 mm
12 large 34 large 59 large diameter diameter
pink pink pink
nodules nodules nodules
Figure8.13 A nodule-scoring chart that can be applied to both pulse and pasture legumes that
have grown more than 12 weeks in the field
8.10.2 Nodule occupancy

Nodule occupancy is an important parameter when assessing the impact of inoc-
ulation in the face of competition by strains resident in the soil. The identification
of strain(s) within nodules can be determined by isolation from a representative
number of these nodules (Chapter3) and identification by an appropriate tech-
nique (e.g. agglutination, ELIZA, PCR RAPD). Molecular techniques to identify
strains are discussed in Chapter11. MALDI-TOF analysis of ribosomal proteins is
rapid and does not require culturing of bacteria. This has been applied to nodules
to determine occupancy, and can be discriminative at strain level for some species
of rhizobia (Zeigler et al. 2015).
162
8.10.3 Rhizosphere counts by MPN

The colonisation of the rhizosphere by inoculant strains can be an important pa-
rameter with which to judge strain success, yet counts by MPN (Chapter6) are a
time-consuming exercise and results vary both temporally and spatially. The value
of the knowledge gained by an MPN estimation needs to be balanced against the
effort and resources required, and the likely nature of its variability. For fresh root
samples, rhizosphere counts of nodule bacteria may be undertaken by immersion
of the roots in a volume of liquid and using this as the primary source of material
to begin the dilution series (Chapter6). The rapid development in next genera-
tion sequencing technologies, combined with the growing availability of whole
genome sequences for rhizobia, holds promise for the development of a quantita-
tive sequencing-based enumeration method. This would greatly expedite studies
of rhizobial ecology.
8.10.4 Plant yield

Plant yield can be very valuable to demonstrate the benefits of inoculation to in-
creased growth, or the benefits of an improved strain relative to controls (Chapter
10, Section 10.4.1). Biomass can be estimated by removing plants within a defined
quadrat (e.g. 0.25 m2) placed randomly several times within each plot, usually
across the tillage lines. Plants can be cut at ground level using a sharp knife or sec-
ateurs, dried in loosely-packed bags at 6580C then weighed. Roots from within
the quadrat can also be excavated, washed free of soil then dried and weighed, al-
though root biomass is usually highly correlated with plant biomass (Russell and
Fillery 1996; McNeill et al. 1997). An exception may be in un-inoculated plots, or
nitrogen-fed plots, where root systems may be more substantial if they need to
explore soil for uptake of mineral nitrogen.
8.10.5 Seed yield

Seed yield can demonstrate the value of an effective symbiosis. However, under
conditions of terminal drought, a plant that grows larger because of an improved
symbiosis can sometimes yield less seed because of premature senescence. For
larger plots, a machine harvester can accurately determine yield, and this can be
sub-sampled for analysis. Alternatively, yield on a sub-set of the plot can be de-
termined by quadrats cut at random within the plot. Several quadrats of at least
0.5m2 are recommended per 10m of plot length.
8.10.6 Nitrogen fixation

Plant nitrogen at peak biomass is often a useful parameter to fully understand and
describe the benefits of inoculation. N analysis, including measurements on 15N
natural abundance, is covered fully in Chapter10. Plant parts must be harvested,
dried at 6070C then stored before analysis. Note that in many legume crops,
seed protein is not a good indication of N2 fixation per se, as legumes are very
163
8.11 Long-term experiments
efficient at translocating N to the seeds. In addition, nodule number is not always

a good indicator of N2 fixation. Where molybdenum is deficient, or the strain is
poorly effective, nodule number can be greatly increased.
The above parameters (Sections 8.10.1 to 8.10.6) may all be measured to aid in in-
terpretation of the impact of inoculation. Sample containers, such as plastic bags
for fresh roots and tops, screw capped vials for MPN counts, and paper bags for
dry plant parts and seed, must be clearly labelled with the experiment number,
date, treatment and plot number.
8.11 Long-term experiments
Inoculants have several fates: they may remain in a stable form and become part
of the soil microflora, such as for soybean in Brazil (Boddey and Hungria 1997);
they may be displaced and eventually disappear altogether (Graham 1992); or
they may mutate and be altered in their symbiotic properties (Kucey and Hynes
1989; Louvrier et al. 1996; Ochman et al. 2000). Inoculants may also donate
their symbiotic genes to resident (but non-nodulating) bacteria, which can then
emerge as competitive, nodulating organisms. Nandasena et al. (2009) described
the emergence of new species of mesorhizobia within five years of sowing Biser-
rula pelecinus in Western Australia, after their acceptance of a symbiotic plasmid
from the inoculant strain. Long-term experiments that wish to follow the fate of
introduced strains need to be well marked, preferably with a GPS coordinate. The
corners of the main plots can be pegged or metal can be buried beneath the cor-
ners for later detection with a metal detector or with GPR.
8.11.1 Following the fate of fixed N and free N farming

In many situations, the profit from incorporating legumes into farming systems is
realised in the rotational crop, and much of this benefit is in the fixed N. Where
the legume growth has been successful, there is often sufficient fixed N to grow
one or more subsequent non-legume crops without additional inorganic N. This
we have termed free N farming. The economic benefit can be estimated by carry-
ing the experiments described in Section 8.9 into a second season, with the plots
sown to the target non-legume. The benefit of fixed N to a subsequent crop can be
estimated as follows.
1. Ensure legume rhizobium treatments sown in the first year are in plots of
sufficient size to split for N rates in a second season. Plots 4m 20m are usu-
ally sufficient.
2. Include an un-inoculated treatment, as well as several rhizobia treatments of

reduced effectiveness relative to the best strain, as this will provide a set of
comparisons of N fixed.
164
8.12References
3. Carefully mark the corner of each plot in the first year (legume rhizobia) of
the experiment.
4. In the next season, manage weeds with a knockdown herbicide, then over-sow
the plots with the non-legume crop (cereal or canola) and within each plot,
split for levels of inorganic N. Usually zero, 25 and 50 units of N will suffice.
5. Thus, the original 4m wide plot is split into three sub-plots, each of 1m
20m with 30cm gaps between sub-plots.
6. The rotational crop can be sampled for biomass yield and N concentration
through the season, as well as seed yield, N concentration and protein when
the crop has matured.
7. By comparing the response to applied N, both within plots and between

strains, the value of the fixed N from the most effective strain of rhizobia can
be estimated.
8.12References
Burgos F.A., OHara G.W., Kobryn H. and Howieson J.G. 2009. The application of remote sensing
to estimate nitrogen levels across field pea genotypes. In Proceedings 15th Australian Nitrogen
Fixation Conference, Margaret River, ed. by M.J. Dilworth and L. Brau.
Boddey L.H. and Hungria M. 1997. Phenotypic grouping of Brazilian Bradyrhizobium strains which
nodulate soybean. Biology Fertility Soils 25, 407415.
Chatel D.L. and Parker C.A. 1973. Survival of field-grown rhizobia over dry summer period in West-
ern Australia. Soil Biology & Biochemistry 5, 415423.
Deaker R., Roughley R. and Kennedy I.R. 2004. Legume seed inoculation technology- a review. Soil
Biology and Biochemistry 36, 12751288.
Farquarson R. 2010. The impact of acetohydroxyacid synthase inhibiting herbicides on symbiotic
nitrogen fixation of grain and pasture legumes. PhD thesis. University of Adelaide, Australia.
Graham P.H. 1992. Stress tolerance in Rhizobium and Bradyrhizobium, and nodulation under ad-
verse soil conditions. Canadian Journal of Microbiology 38, 47584.
Helyar K.R. and Porter W.M. 1989. Soil acidification, its measurement and the processes involved.
Pp. 61101 in Soil Acidity and Plant Growth, ed. by A.D. Robson. Academic Press: Sydney.
Howieson J.G. and Ewing M.A. 1986. Acid tolerance in the Rhizobium meliloti-Medicago symbiosis.
Australian Journal of Agricultural Research 36, 5564.
Howieson J.G. and Ballard R. 2004. Optimising the legume symbiosis in stressful and competitive
environments within southern Australiasome contemporary thoughts. Soil Biology & Bio-
chemistry 36, 12611273.
Howieson J.G., De Meyer S.E., Vivas-Marfisi A., Ratnayake S., Ardley J.K. and Yates R.J. 2013. Novel
Koopman D.J., Tow P.G., Reeves T.G. and Gibson A.H. 1995. Soil acidification, chlorsulfuron ap-
plication and Rhizobium meliloti as factors in lucerne yield decline. Soil Biology & Biochemistry
27, 673677.
Kucey R.M.N. and Hynes M.F. 1989. Populations of Rhizobium leguminosarum biovars phaseoli and
viciae in fields after bean or pea in rotation with non-legumes. Canadian Journal of Microbiol-
ogy 35, 661667.
Loi A., Howieson J.G., Nutt B.J. and Carr S.J. 2005. A second generation of annual pasture legumes
and their potential for inclusion in Mediterranean-type farming systems. Australian Journal of
Experimental Agriculture 45, 289299.
165
8.12References
Louvrier P., Laguerre G. and Amarger N. 1996. Distribution of symbiotic genotypes in Rhizobium
leguminosarum biovar viciae populations isolated directly from soils. Applied Environmental
Microbiology 62, 42024205.
McNeill A.M., Zhu C. and Fillery I.R.F. 1997. Use of in situ 15N-labelling to estimate the total below-
ground nitrogen of pasture legumes in intact soil-plant systems. Australian Journal of Agricul-
tural Research 48, 295304.
Nandasena K.G., OHara G.W., Tiwari R.P., Willems A. and Howieson J.G. 2009. Mesorhizobium
australicum sp. nov. and Mesorhizobium opportunistum sp. nov., isolated from Biserrula peleci-
nus L. in Australia. International Journal Systematic Evolutionary Microbiology 59, 21402147.
Ochman H., Lawrence J.G. and Groisman E.A. 2000. Lateral gene transfer and the nature of bacte-
rial innovation. Nature 405, 299304.
Russell C.A. and Fillery I.R.P. 1996. In situ labelling of lupin below-ground biomass N. Australian
Thies J.E., Cook S.E. and Corner R.J. 1991. Influence of the size of indigenous rhizobial populations
on establishment and symbiotic performance of introduced rhizobia on field-grown legumes.
Applied Environmental Microbiology 57, 1928.
Vincent J.M., Thompson J. and Donovan K. 1962. Death of root-nodule bacteria on drying. Austral-
ian Journal of Agricultural Research 13, 258270.
Yates R.J., Howieson J.G., Real D., Reeve W.G., Vivas-Marfisi A. and OHara G.W. 2005. Evidence
of selection for effective nodulation in the Trifolium spp. symbiosis with Rhizobium legumino-
sarum biovar trifolii. Australian Journal of Experimental Agriculture 45, 189198.
Ziegler D., Pothier J.F., Ardley J., Fossou R.K., Pflger V., Meyer S.E., De Vogel D., Tonolla M.,
Howieson J., Reeve W. and Perret X. 2015. Ribosomal protein biomarkers provide root nodule
bacterial identification by MALDI-TOF MS. Applied Microbiology and Biotechnology (DOI
10.1007/s00253-015-6515-3).
166
9.1Introduction
CHAPTER9
Inoculant production
and quality control
R. Deaker, E. Hartley, G. Gemell, D.F. Herridge and
N. Karanja
9.1Introduction
The manufacture of high-quality inoculants increases the potential for maximum

nitrogen fixation in inoculated legumes (see Chapters 1, 5, 8 and 10). If legume
inoculation is successful, the inoculant strain will colonise the rhizosphere and
compete with resident soil rhizobia for nodulation sites on the host legume root.
Researchers too must prepare high quality inoculants for experimental field trials,
similar to those described in Chapter8.
Large-scale production of high-quality legume inoculants is complex. It requires

expert skills in aseptic handling of rhizobia, an understanding of the conditions
in which rhizobia grow and survive, research and development to find suitable
carrier materials for inoculant formulation, and a program of quality control and
quality assurance to maintain the manufacturing process and product integrity.
There are several other publications describing methods involved in legume inoc-
ulant quality control; comprehensive texts have been published by Vincent (1970)
and Somasegaran and Hoben (1994).
The inoculant supply chain can be summarised as:
rhizobial strains are selected that are highly effective in fixing nitrogen with
the target legume host (Chapters 2, 5 and 8)
batch cultures of these strains are grown in liquid medium (Chapter3)

the cells are maintained either as liquids, concentrated cell masses or incorpo-
rated into solid carriers to form inoculant products
the inoculants are distributed to farmers
167
9.2 Inoculant production
the farmer inoculates the legume, by coating directly onto the seed, or apply-
ing as liquids or granules to the soil in close proximity to the seed.
At each point along this chain, quality control protocols should be implemented
to ensure strain purity, growth, survival and functional stability (Figure9.1).
Strain selection (Chapters 2, 3) Starter culture
Strain preservation (Chapter 4)

Large-scale fermentation
Grandmother culture
Mother cultures
Cell concentration Injection of solid carrier
Solid-state fermentation
Liquid inoculant Frozen free-dried inoculant Solid inoculant
Storage Distribution Application
Figure9.1 Stages in production of legume inoculants, their distribution and application. Colours
indicate different stages in the process: 1. strain selection and preservation (blue); 2. Large-scale
production of batch cultures (pink); 3. formulation of inoculants (green); 4. storage, distribution
and application (orange). Red stars indicate points for quality control to determine strain purity,
growth, survival and functional stability.
The first step in inoculant production is the growth of liquid starter cultures from
a mother culture.
9.2.1 Growth in batch culture

The most widely adopted method for growing cultures of 20mL to 1,000mL is
the regular batch or discontinuous culture. The key requirements are maintenance
of culture sterility during inoculation and provision of oxygen for respiration of
168
the rhizobial cells. Volumes less than 1L are aerated on a shaking platform; larger
volumes require filtered air or oxygen.
1. Select culture medium (Chapter3) and dispense into appropriate-sized vessels

to no more than 20% of the vessel volume. For laboratory purposes this might
be a 250mL Erlenmeyer flask containing 50mL of liquid.
2. Close with screw cap (not fully tightened) or cotton bung covered with alu-
minium foil to keep the bung dry during sterilisation in the autoclave.
3. Autoclave at 121C (15psi) for 20minutes. Larger volumes (>1L) take longer
to reach the required temperature so the sterilisation time must be increased.
4. Place media in a laminar flow cabinet to cool and close screw caps when ves-
sels reach room temperature.
5. Inoculate small volumes (<1L) in the laminar flow cabinet. Remove the cap
or cotton bung and pass the neck of the vessel through a Bunsen flame. Tilt
to one side to bring the liquid near the neck, then using a sterile inoculating
loop, transfer a loopful of culture from an agar slope or plate into the liquid, or
insert a porcelain bead coated with rhizobia. Take care not to touch the top of
the open vessel. Replace cap or cotton bung and aluminium cover.
6. Inoculate larger volumes by pipette with a broth culture at log phase to a final
volume of 10% of the growth medium.
7. Place small vessels on a platform shaker set at 150rpm, or provide filtered oxy-
gen / air to large vessels (e.g. 5L/L/hour) and monitor pO2.
8. Monitor the growth characteristics of the strain by regular inspection. There

are about 107 cells per mL when the growth medium is slightly opaque
(OD=0.1).
After reaching the target cell density (usually 109 cells/mL), strains can be intro-
duced directly into the carrier of choice to make small-scale inoculants for re-
searchers, or as a starter culture in a scale-up process, with the batch culture pro-
viding a proportion (e.g. 10%) of the volume to the next vessel.
There are alternatives to preparing a starter culture in liquid broth. One such
method uses two plugged Buchner flasks with side arms joined by silicon tubing
(described in Thompson 1983). One contains agar medium which can be sloped
for culturing cells, the other has liquid medium, and the whole system is sterilised.
After growing a culture on the agar slope, the liquid is decanted through the sili-
con tubing to the sloped agar culture. The cells are then washed from the slope
and transferred directly to a fermentation unit without opening any bungs. The
advantage of this method is that growth of any contaminant microorganism on
the slope is easily observed.
169
9.2.2 Scaling up broth cultures for manufacture

Vessels to produce large volumes of liquid cultures (>5L) are generically termed
fermenters. They vary from simple to highly complex in design, but it is essential
that they are readily sterilised, allow access for inoculation and sampling, and pro-
vide aeration for the culture. Because large fermentors are bulky and heavy, they
usually have capacity for sterilisation in situ.
Continuing with the example above, the 50mL broth culture can provide the in-
oculant for 450mL held in a 2.5L flask that can be constructed from readily avail-
able laboratory materials (Figure9.2a). Fermentors larger than this generally have
the same main components (Figure9.2b). The broth is aerated via an inlet tube
fitted with a sterile air filter. This filtered air pumped through the medium acts to
both aerate and mix the broth culture. Where the fermentor volume exceeds the
capacity of the air to provide adequate mixing, mechanical paddles can be added
to the design to stir the medium. Fermentors also require a gas outlet and a port
for sampling broth during the growth period. All components and fittings shown
in Figure9.2b must be sterilised to ensure the broth does not become contami-
nated. This unit can be sterilised as a whole in an industrial autoclave, or it can be
set on a gas ring burner and the contents steamed for at least 30minutes, with all
ports vented and exposed to the steam.
After inoculation, the cultures should be monitored for cell number and pH dur-
ing growth, taking great care not to introduce contaminants in the process. Unu-
sual changes in pH in complex media (i.e. < 6 or > 8) can be a convenient way of
detecting possible contamination, however, as is microscopic visualisation of a
sample from the fermentor in the early phases of growth (see Chapter7, Section
7.2.5.2). Common contaminants may be detected in broth samples by plating a
sample on peptone agar (Chapter3, Section 3.7.2.4) and observing growth and
pH change after 2448hours at 30C (Chapter7).
E
D C
C B
B Figure9.2a A simple
F
fermentor adapted from
Somasegaran and Hoben
G (1994). A: aquarium pump;
B: non-absorbent cotton
A
wool; C: plastic syringe
barrel; D: clamp; E: silicon
H and glass tubing air inlet; F:
D I
silicon and glass tubing air
outlet; G: glass Erlenmyer
flask; H: broth culture; and
I: broth sampling tube.
170
9.3 Carrier materials for inoculants
Air inlet port
Air outlet port
Stainless steel drum
Large threaded
Inoculation bung to allow
port access for
Solid copper tubing cleaning
Filters
Air outlet
Rubber bung
Flask
Sampling
port Sparger Rubber tubing
To air
pump
Figure9.2b A simple stainless steel fermentor (reproduced from Thompson 1983).

After reaching the required cell density, the inoculant is then formulated with
several possible different carriers and extenders. These are solid or liquid media
that support live bacteria after broth culture substrates are exhausted. The broth
culture from the fermentor is introduced into the carrier, and for most carriers
(after an initial curing period) there is little subsequent active growth of bacteria.
The carrier is then stored, distributed to growers and finally applied to either the
legume seed or the soil.
Finely milled peat is a well-researched and widely adopted carrier in many coun-
tries but inoculants may also be available as liquid, granular, frozen pastes and
freeze-dried preparations (Figure9.3). Suitable peat sources are not always avail-
able and many alternatives have been investigated, including organic carriers
such as compost, bagasse, farmyard manures, sugarcane filter mud and coconut
coir dust; and inorganic carriers such as perlite, polymers and clays (reviewed by
Roughley and Pulsford 1982; Burton 1984; Bashan 1998). Composted materials
171
from the cork industry and perlite were superior to peat in maintaining survival
of different rhizospheric bacteria (Albareda et al. 2008).
Desirable characteristics of solid carrier materials are listed below (from Thomp-
son 1980).
1. A capacity to be finely ground to allow thorough mixing with other compo-

nents and to allow ease of application.
2. Adjustability of pH to 6.57.0.
3. Good moisture-holding capacity.
4. Able to be sterilised to favour survival of rhizobia.
5. Free of toxic materials and low salinity.
(a) (b) (c)
Figure9.3 Alternatives to peat inoculant products. (a) Peat granules (left), bentonite clay
granules (middle), attapulgite clay (right); (b) Liquid; (c) Freeze-dried powders.
9.3.1 General preparation of a solid carrier, e.g. peat

1. Drain any excess moisture from the peat for ease of handling.
2. Grind peat in a hammer mill until particle size is less than 75m.
3. Homogenise the peat.
4. Adjust pH to near neutral with weak acid or CaCO3.
5. Apportion peat into desired volumes in containers.
6. Adjust moisture content to desired level (often 1020%).
7. Reduce resident contaminating microflora by sterilisation, e.g. autoclaving,

heating, radiation or other treatment.
9.3.2 Adjusting the pH of carriers

If CaCO3 is required to raise the pH of the carrier, it should be well mixed and
reacted carefully, allowing time for equilibration. The amount of CaCO3 or acid
required to change the pH will depend upon organic matter and clay content as
well as the buffering capacity of the carrier.
172
9.3.3 Reducing contaminants in the carrier

Ideally, the carrier material should contain very low numbers of contaminant mi-
croorganisms before introducing rhizobial strains, as resident microflora can re-
strict the growth of rhizobia and reduce their final number in the inoculant. The
amount of restriction depends upon the rhizobial strain (Roughley and Vincent
1967). Autoclaving of organic carriers is an effective means of sterilisation but
can result in the release of toxic polyphenols, and this should be assessed. A small
amount of moisture is required for the production of steam within the carrier for
optimum sterilisation. If a carrier is too dry, sterilisation will not be complete.
The efficacy of sterilisation can be measured by injecting the sterile carrier with
sterile growth medium and measuring growth of contaminants over time (e.g. for
one month).
It is better to autoclave organic materials several times at temperatures lower than

121C (e.g. 100C), allowing them to cool between runs, as toxic products from
the breakdown of organic matter may accumulate. Repeated cycles of heating will
ensure spore-forming microorganisms are sufficiently reduced. Otherwise, gam-
ma-radiation (at approximately 75 KGy) is a well-proven means of sterilisation
but is not without limitations; its effectiveness depends on the initial contaminant
load in the carrier which is related to moisture content and packaging conditions
before exposure to radiation (Bullard et al. 2005).
Other means of reducing contaminants in the carrier can be tested. Exposure to

UV light, hydrogen peroxide, ethylene oxide, acidification and solarisation have
all been explored for this purpose.
If sterilisation is not possible, broth culture can be injected at a higher cell concen-
tration resulting in little dependence on further growth. However, carriers must
have a high moisture holding capacity if large volumes of broth are added. After
injecting sterile carriers with broth cultures of rhizobia, the number of viable cells
per gram of carrier should be measured at several time intervals to indicate rela-
tive growth, survival and shelf life. Rhizobial strains can be differentiated from
other microorganisms using techniques described in Chapters 3, 7 and 11.
9.3.4 Moisture content

The moisture content of the carrier is important as it affects rhizobial growth and
survival. Particle size, organic matter and clay content of peat contribute to its
water holding capacity. For example, Vietnamese peat inoculants with low or-
ganic matter content have been amended with materials such as coconut coir dust
and fish pond filter mud to increase moisture holding capacity (unpublished data
from AusAID CARD project 01306VIE). Increasing the water holding capacity
of carriers allows larger volumes of concentrated rhizobial broth culture to be
introduced before incubation while maintaining optimum moisture content for
cell growth.
173
9.3.5Determination of optimum moisture content in solid-

based carriers
Optimum moisture content for good growth and survival of rhizobia can be de-
termined by measuring viable rhizobial numbers in carriers with different mois-
ture contents over time. Examples of the amount of liquid required to adjust peat
to 40, 50 and 60% moisture content are listed in Table9.1. The volumes are based
on an initial moisture content of 20%. Peat carriers that have different moisture
holding characteristics will require different volumes of liquid to adjust to the
same moisture content.
Table9.1 E
xamples of treatments for measuring optimum moisture content for
microbial inoculants. Calculations based on 70g dry peat after adjusting
to 20% moisture content for sterilisation.
Moisture content Total liquid added Volume of broth Volume of sterile
(%) (broth + water, mL) (mL) water
(mL)
40 37.3 37.3 0
50 56.0 37.3 18.7
60 84.0 37.3 46.7
9.3.5.1 Calculating the amount of liquid to add to peat or carrier mixture
The equation below is used to calculate the moisture content of 70 g dry peat,
where x is the amount of liquid added and y is the final percent moisture. The
same equation can be used for any quantity of peat, but as peat will invariably al-
ready contain some moisture, the mass of dry peat must first be determined.
Therefore, to adjust 70 g of dry peat to 20% for sterilisation 17.5 mL of water

should be added.
Sample calculation of liquid added to peat to obtain specific moisture content
How much liquid (e.g. broth) should be added to 70g peat with 20% moisture to
achieve a moisture content of 40%?
Mass of dry peat
Therefore, the mass of dry peat is 56g
Volume of liquid to add to dry peat
174
9.4 Peat inoculants
Therefore, 37.3g moisture should be added to 70g dry peat to achieve 40% after
adjusting to 20% moisture.
9.4 Peat inoculants
Pre-sterilised (gamma-radiated) finely ground peat sealed in plastic packets is

widely utilised as a carrier for rhizobia. Peat inoculants consist of finely ground
peat injected with a single strain of rhizobia at a high concentration. The final
preparation has relatively high moisture content when compared with other solid
formulations. They are usually prepared as a slurry to coat the seed and are most
effective when sown into moist soils. They can also be diluted in potable water and
applied as a liquid directly into the soil.
The preparation procedures described below may also be applied to any finely
ground organic carrier.
1. Prepare a broth culture of rhizobia in a suitable vessel or flask.
2. Estimate the amount of broth culture to be added to the packet to obtain the
correct moisture content for optimal rhizobial growth and survival.
3. At the site of injection, surface sterilise the packet containing pre-sterilised

peat (or other carrier) with a swab containing 70% ethanol.
4. Using a sterile syringe and needle, carefully draw up the required volume of
broth from the flask. Flame the neck of the flask, tilt it to one side and insert
the needle into the broth culture. Draw the broth into the syringe. Take care
not to touch the open neck of the flask.
5. Inject the required amount of broth culture into the packet, ensuring that
none of the culture escapes from the injection site. Eject the last drop of liq-
uid to ensure the needle is not blocked by organic matter if it is to be reused.
Manufacturers can automate this step using an injection system.
6. Swab the injection site with ethanol and allow it to evaporate before covering
the hole with a sticky label displaying the rhizobial strain name and the date
of injection.
7. Multiple peat packets can be injected using the one syringe and needle when
researchers are producing multiple inoculants, but needles and syringes must
be discarded before working with a second or subsequent strain.
8. To check for contaminants in the broth culture prior to injection into the
carrier, streak a drop of the broth culture onto general agar medium (e.g.
175
9.5 Granular inoculants
glucose-peptone medium) in a Petri dish and incubate at 2628C and check

for growth after 2448hours. Check the growth every 24hours to verify purity
of the broth culture.
9. Carefully massage the peat packets until the broth culture is evenly distributed
within the packet, making sure that the broth reaches each of the four corners.
10. Incubate the inoculant for one to two weeks at 2628C.
Some inoculant manufacturers dilute broth cultures (e.g. 1:10 or 1:100 dilutions)
to make peat-based inoculants. However, it is necessary to check growth and sur-
vival of rhizobia with each carrier over time to determine suitable dilutions.
9.4.1 Counting viable rhizobia in peat inoculants

If the carrier is sterile, before injection of broth, the Miles and Misra drop plate
count (Chapter6) can be used as an alternative to the spread plate count method
for enumeration. If numbers differ widely between strains then the quantity of
inoculants to be utilised in the experiments (Chapter8) will have to be varied.
If the peat carrier is non-sterile, it is recommended to use a plant infection fre-

quency of nodulation test that provides a most probable number estimate of vi-
able rhizobia per gram of inoculant (Chapter6).
Granular inoculants are convenient to use as they can be applied directly to soil
and there is no need to prepare adhesive slurries to inoculate the seed. Granules
can be mixed with seed and then distributed from the seed box, or through a
separate fertiliser box available on some planters. Granules generally have lower
moisture contents than peat inoculants and therefore have the advantage of not
requiring refrigeration and can be loaded into seeders in bulk when required.
Some dry granules have proven effective for introducing inoculants to dry soils
by protecting rhizobial cells until rains fall (Loi et al. 2012). However, the require-
ment for high application rates (because of generally lower numbers of rhizobia
per gram) can be a disadvantage.
Granules can be manufactured by:
prilling an existing carrier such as clay (usually dry) or organic matter (usu-
ally moist); this may be done by either extrusion or sorting of naturally formed
granules
impregnating broth cultures onto pre-formed granules (such as peat, alginate,

hydrophilic minerals or polymers) which absorb the inoculant into the granu-
lar matrix.
176
Granular products are graded into a range of particle sizes from 0.5mm to 5mm
in diameter. The smaller particles are better for distribution of rhizobia over a
cropping area. If a large proportion of particles are less than 125m, flowability of
the granules through the seeder or other distribution equipment may be reduced
and cause blockages.
9.5.1 Granulated clays

A recent invention is the dry clay granule which holds rhizobial cells protected
from desiccation within a clay and organic matrix, analogous to how rhizobia ex-
ist naturally in many soils. Survival in the clay carrier has been related to forma-
tion of a protective layer around cells by clay minerals as a result of surface charge
interactions (Marshall 1968, 1969).
Procedures for making dry-clay granules
1. Prepare a quality peat inoculant as described in Section 9.4.
2. Source appropriate clay, and if practicable, reduce numbers of contaminant

microbes.
3. Dilute peat inoculant 1:5 in sterile water to produce a slurry.
4. Add slurry 1:10 to clay and mix thoroughly.
5. Pour the clay slurry into a shallow (15cm) containment vessel and allow to
air dry.
6. Crush or mill the dry product to the required particle size.
7. Bag, label and store at room temperature.
Clays may be alkaline or acidic, and rhizobial species react differently to pH.
Bradyrhizobia in general survive in greater numbers in acidic clays, while the
faster growing species of rhizobia prefer alkaline clays. However, the Burkholderia
inoculants for Lebeckia ambigua did not survive the drying process in manufac-
turing dry clay inoculants based upon bentonite (Howieson et al. 2013); hence the
researcher must assess the suitability of the carrier for the target rhizobial species.
Numbers of rhizobia per gram are generally lower in dry clay granules than in
moist granules or peat, decreasing from approximately 109 per g at mixing to 107
per gram when dry. However, as the granules are dry, further death of the inocu-
lants upon exposure to a drying environment is potentially minimised.
9.5.2 Freeze-dried powders

Freeze-dried inoculants are produced from cell pastes that have been concentrat-
ed from large quantities of broth, usually by centrifugation. The cells are mixed
with cryoprotectant agents (e.g. skim milk powder) before drying in commercial
freeze-dryers. Depending on the volume and capacity of the freeze-dryer, drying
177
9.6 Storage of inoculants
may take several days. Freeze-dried cells are reconstituted with a protective liquid
polymer to form a slurry to inoculate seed and can also be applied as a liquid di-
rectly into the soil.
Not all strains of rhizobia respond well to freeze-drying and rhizobial numbers
may be reduced. However, freeze-dried inoculants generally have very high num-
bers of rhizobia per gram and their shelf life under vacuum over a range of tem-
peratures can be many years.
9.6 Storage of inoculants
The different carriers and rhizobial species have varying requirements for shelf life
and storage depending upon:
the growth characteristics of the rhizobial strain

whether the strain originated from a tropical or temperate climate
rhizobial peat inoculants should be stored in a cool place (ideally 48C)
and should never be frozen
there are exceptions for inoculants prepared for tropical legume species e.g.
Desmodium, Centrosema, Lablab and Stylosanthes which should be stored
between 20 and 25C
the type of carrier (solid, liquid or freeze-dried)

the form of packaging.
Table9.2 below provides a matrix showing recommended storage conditions for
different inoculant formulations and strain origins.
Table9.2 General storage conditions for different rhizobial inoculants

Inoculant Origin of strain
formulation Tropical Temperate
Peat-based Ambient (2025C) Refrigerated (24C)
Granular Ambient (2025C) Ambient (1020C)
Freeze-dried Refrigerated(410C) Refrigerated (410C)
Liquid Refrigerated (46C) Refrigerated (410C)
9.7 Inoculant quality control
9.7.1 What is quality?

Quality refers to the delivery of high numbers of viable and effective rhizobial
cells to nodulate the target legume at sowing. The term quality is often associated
178
with the number of live cells delivered and the ratio of these to contaminants. The
efficacy of inoculants may not be the same at all sites because of the manner in
which the inoculant is used, and the great diversity of environmental, biotic and
edaphic conditions they may confront. For some symbioses, very few inoculant
cells are required to achieve optimal nodulation (less than 10 per seed), whereas
for others this number may be 1,000 times greater. However, the number of live
cells delivered is a convenient descriptor of quality, and one which the industry
has come to promote.
Without adequate practical means of determining the symbiotic interactions at

each site, standards are based largely on the maximum possible inoculum poten-
tial (number of live rhizobia in inoculant application rate) with each formula-
tion. The numerical standards for Australian legume inoculants are listed in Ta-
ble9.3. These standards have been recommended for many years and are sourced
from the Australian National Code of Practice, Quality Trademark for Microbial
Inoculant Products used in Australian Crops and Pastures compiled by the Aus-
tralian Inoculants Research Group (AIRG) and the Legume Inoculant Industry,
and implemented in 2010.
Table9.3 Current numerical standards for Australian inoculants

Product Fresh count Expiry count Expiry (months
from test date)
Peat (cfu/g) 1 109 1 108 1218

Liquid* (cfu/mL) 5 10 9
1 10 9
6
Granules (MPN/g) 1 107 1 106 6

Freeze dried** (cfu/vial) 1 10 12
5 10
11
6
* Standard for liquids based on a 3L bottle used to treat one tonne of seed.
**Standard for freeze-dried product based on vial used to treat 500kg seed
for large-seeded legumes. Contaminants in peat, liquid and freeze-dried
products should be absent at the 106 dilution when counted according to
AIRG protocols.
Despite technical advances in inoculant production systems over the years, and
the expertise of commercial inoculant manufacturers in producing large volumes
of legume inoculants, there have been challenges with inoculant production and
utilisation.
Challenges associated with the production and use of peat inoculants between
1953 and 2003 in Australia were listed by Bullard et al. (2005) and are presented in
Table9.4 to illustrate some of the issues that may occur. Some challenges remain
unresolved, such as poor survival for some genera of rhizobia on pre-inoculated
seed.
179
Table9.4 Problems associated with the manufacture of peat inoculants in Australia 19532003
Problem Description Year
Quality of peat cultures Low numbers and poor survival of some strains of rhizobia in 19531966
peat due to contaminants
Quality of peat carrier Overheated peat caused poor viable counts of clover strains 1963
of rhizobia
Quality of peat carrier Ethylene oxide residue after sterilisation treatment caused 1964
low numbers of rhizobia in peat cultures
Quality of peat carrier High salt content due to changes in climate and water table 1974
levels caused low numbers of rhizobia in peat cultures
Peat culture quality Death in peat culture of some rhizobia for tropical legumes 1978
due to cold storage
Quality of peat carrier Ineffective sterilisation using gamma irradiation caused 19782003
problems with contaminants
Plastaid seed coating Low counts of rhizobia on seed due to high pH of Plastaid 19902002
calcium carbonate (calcium carbonate) material used for coating seeds
Pre-inoculated seed quality A Low viable numbers of clover rhizobia on seed 1959ongoing
Pre-inoculated seed refers to seed that is commercially inoculated and coated prior to sale, and then stored for
A
up to six months. (Table adapted from Bullard et al. 2005).
While there is no global standard, there is general agreement that inoculants

should deliver a recommended number of rhizobia per seed and this varies ac-
cording to seed size (Table9.5).
Table9.5 The number of cells recommended for legumes of different seed size*
Seed size Cells per seed
<2mg e.g. biserrula, white clover 500
25mg e.g. sub clover, lotus, 1,000
510mg e.g. hedysarum, siratro 10,000
>10mg e.g. lupin, pea, soy, bean 100,000
*(from Lupwayi et al. 2000; Herridge et al. 2002)
Field evaluation of the different inoculant products is recommended (Chapter8).

As a guide, in Australia the recommended rate of application of inoculants has
been designed to deliver > 1 1010 live cells per ha within the shelf life of the in-
oculant.
In addition to rhizobial numbers, inoculants should be correctly labelled with

information about the recommended plant host, the rhizobial strain, methods for
application, batch number, conditions for storage and expiry date.
9.7.2 Numbers of contaminating organisms

Ideally, peat, liquid and freeze-dried inoculants should not contain a large num-
ber of any organism (contaminants) other than the selected rhizobial strains.
However, if this is not possible, contaminants should be at least 100 times fewer in
number than the rhizobial strain. Numbers of contaminants and moisture content
180
of peatbased inoculants are good indicators of potential shelf life and should be
checked routinely. If inoculants are retested just prior to expiry and the rhizobial
numbers have been maintained without contaminants, shelf lives can be extended.
9.7.3 Quality control protocols

The objectives of quality control protocols are to determine the number of viable
rhizobia in an inoculant formulation at a point in time and to measure parameters
that indicate the likelihood that quality will be maintained throughout shelf life.
The key activities and parameters are:
monitoring of batch variability

recovery, enumeration and positive identification of rhizobial cells from the
carrier
effectiveness in fixing nitrogen of inoculant strains

enumeration of contaminants
measurement of moisture content or relevant conditions that support growth
and survival in the carrier
assessment of products from different points in the supply chain.
9.7.3.1 Batch variability
Both liquid and solid inoculants are usually produced from single strain fermen-
tation in batches. Each batch represents a different set of conditions that may af-
fect the quality of the resultant products. The number of packets that should be
tested per batch depends on variability of the product. For example, in Australia,
seven packets are submitted from each batch of legume inoculant produced. A
subset of five packets is tested, and if all five pass the standards, the batch is re-
ported as passed. If one of these five packets fails, the other two packets are tested.
If these two additional packets pass the standards, then the batch is passed. How-
ever, the batch is reported to have failed if two or more of the original five packets
fail the set standards, or one of the two extra packets tested fails.
There should be an adequate recording system so that testing of each batch can
be tracked and reported back to manufacturers. Consideration should be given
to the logistics of product recall should product release be based on presumptive
tests only.
It is useful if the QC testing laboratory is notified ahead of time of the number of

samples being sent to enable the necessary materials (media etc.) to be prepared.
After receipt of products in the laboratory, they should be stored in appropriate
conditions (e.g. at 4C) and quality control tests should be carried out promptly.
181
9.8 Determination of shelf life
9.7.3.2 Strain identity
To ensure the correct strain is present in the inoculant, the identity of the strain
must be established. Refer to Chapters 3, 6, 7 and 11 for methods of identification
of rhizobia.
9.8.1 Testing inoculants in a distribution network

The number of live rhizobia in inoculants should be determined from products in
the distribution network (e.g. from retail outlets and storage warehouses). Surviv-
al of inoculant strains is variable and is dependent on environmental factors, such
as temperature fluctuations. Rhizobial numbers in Australian peat inoculants
counted at the point of manufacture (fresh) and throughout the distribution chain
(retail) demonstrates the variable response of some rhizobial species to conditions
during distribution of inoculants (Figure9.4). Results from studies of long-term
cold storage (4C) of peat inoculants revealed less variation in survival, and while
interesting, may not be a true representation of shelf life in the distribution chain.
As well as temperature, shelf life of inoculants is affected by changes in moisture
content of the carrier. Loss of moisture can be reduced by selecting less permeable
packaging materials, but packaging should also allow adequate exchange of gases.
In Australia, standards were originally set by consciously rejecting inoculants

with rhizobial counts in the bottom 20% of the distribution. The boxplots in Fig-
ure9.4 allow visualisation of the distribution of viable rhizobial numbers in in-
oculants above the 25th percentile. With the exception of white clover, 75% of the
inoculants have counts above the minimum standard of log10 8.0 at expiry. In this
dataset there are several inoculants that fall outside the normal distribution, as
indicated by the red and green crosses. Therefore, when setting numerical stand-
ards, an acceptable minimum number should be recommended, bearing in mind
that numerical standards are often a compromise between what is agronomically
desirable and commercially realistic.
182
10.0
9.5
9.0
8.5
8.0
7.5
7.0
6.5
6.0
Annual_medic_Fresh
Annual_medic_Retail
Chickpea_Fresh
Chickpea_Retail
Cowpea_Fresh
Cowpea_Retail
Faba_bean_Fresh
Faba_bean_Retail
Lucerne_Fresh
Lucerne_Retail
Lupin_Fresh
Lupin_Retail
Pea_Fresh
Pea_Retail
Pigeon_pea_Fresh
Pigeon_pea_Retail
Sub_clover_Fresh
Sub_clover_Retail
White_clover_Fresh
White_clover_Retail
Figure9.4 Boxplot of log10 transformed number of rhizobia per gram of peat

inoculant counted at point of sale (retail) and at point of manufacture (fresh).
Data are from the Australian Inoculant Research Group inoculant surveys. The box
represents 50% of data or inter-quartile range (IQR). The line in the box represents
the median and whiskers extend no more than 1.5 IQR. Data points outside
this range may be considered outliers in an analysis. The boxplot gives a useful
summary of the distribution of data, whether it is symmetrical or skewed.
183
9.9 Criteria for the selection of strains for use in commercial manufacture of legume inoculants
9.9Criteria for the selection of strains for use in

commercial manufacture of legume inoculants
The underlying principles for the selection of strains for commercial inoculant
production relate to functional characteristics, such as effective N2 fixation and
the ability to grow and survive during manufacture, distribution and application.
The selection criteria for commercial quality strains are:
1. effectiveness and appropriate host range (Chapter5)
2. genetic stability in culture, storage and manufacture
3. potential for scale up of production
4. ability to survive in carrier and on seed
5. Field performance and (where required) persistence (Chapter8)
Criteria 24 are briefly covered here.
9.9.1 Genetic stability

Rhizobia can alter their characteristics when cultured or stored for long periods
(Bullard et al. 2005; Ochman et al. 2000). Some outcomes from the Australian
quality control evaluations are presented in Table9.6 as examples. Isolates of some
strains may exhibit genetic instability, hence strains recommended for commer-
cial inoculants should be selected to be stable in culture with no loss of symbiotic
capacity.
Table9.6 S
ome genetic stability problems associated with commercial rhizobial strains in Australia
19532003. (Table adapted from Bullard et al. 2005).
Rhizobium Strain description Description of problem Year(s)
strain(s) reported
NA30 Clover strain from 19561960 (to 1962 Loss of effectiveness on white and red 19591960
in WA) clovers
TA1 Current commercial strain for white Produced variants ineffective on white 1963
clover and red clovers
NA34 Clover strain from 19531955 Loss of effectiveness and production of 1956
non-invasive isolates
UNZ29 Clover strain from 19631966 Poor competitive ability and loss of 1966
effectiveness
CB756 Strain for mung bean (19631977), Reduction of, or variation in, 1975
peanut (19631989) and several others effectiveness
WSM826 Lucerne, Medicago littoralis and Variation in colony types 1992
Medicago tornata strain (19922000)
W118 Medicago rugosa strain (19701980) Partial loss of invasiveness 1980

WU425 Lupin (1970present) and serradella Variation in time to form nodules 1971
strain (19701996)
184
9.9 Criteria for the selection of strains for use in commercial manufacture of legume inoculants
Of the problems listed in Table9.6, loss of effectiveness is the most serious. Strains
should be tested every season for their ability to form nodules and effectively fix
nitrogen after being cultured and stored under different conditions (Chapters 3
and 4). To test for this, undertake a glasshouse experiment (Chapter5) and follow
the procedures below.
1. Prepare legume seedlings for nodulation in the preferred authentication me-

dium (Chapter5). If the legume can fix nitrogen optimally in glass tubes or
pouches, these are most convenient assemblies, as the time to nodulation can
be observed.
2. Streak out strain from mother culture to give isolated colonies.
3. Select five to 20 isolated colonies and add each to a diluent (Chapter6) to pro-
vide sufficient inocula for the number of test plants (usually four repetitions
per colony).
4. Use diluents as inocula, changing pipette or sterile tip for each source, and
monitor nodulation, and N2fixation.
As it is likely that loss of effectiveness is related to loss of symbiotic DNA, a well-

equipped laboratory may also choose to assay the single colonies selected above
for known symbiotic genes such as nif, nod and fix, or the presence of the sym-
biosis island using directed primers and PCR or sequencing techniques (Chapters
1113).
9.9.2 Potential for scale-up production

The growth rate of rhizobial strains in broth culture varies and this can affect
their suitability for manufacture. The criteria relating to growth characteristics
can be examined by counting cells at appropriate times in the manufacturing cy-
cle. Methods for counting rhizobia are described in Chapter6.
If strain growth or survival characteristics have been compromised during stor-

age, an alternative inoculant strain should be used.
9.9.3Strain variation in survival on seed, in the carrier, or in

soil
Strains vary considerably in their survival in soil (Chapter8) and there is a large
volume of literature on this subject, particularly in relation to stress tolerance (e.g.
OHara et al. 2002). The cross-row technique (Chapter8) provides a bio-assay for
strain success in soil in which nodulation is the criterion for selection rather than
cell number. The MPN determined by the plant infection test (Chapter6) is an
alternative method to enumerate strains in soil, and this can be applied to assess
spatial and temporal variation in cell number.
There is considerably less assessment of strain variation in survival in carriers or

when applied to seed, but this is still a very important criterion when selecting
185
9.10References
strains for commercial use (Roughley and Vincent 1967). Poor survival on seed
can have serious consequences for legume success (Materon and Weaver 1984)
and strains do vary in their ability to tolerate desiccating conditions on seed and
survive (e.g. Deaker et al. 2007; Howieson et al. 2013). The dilution and enumera-
tion techniques outlined in Chapter6 are appropriate to determine viable num-
bers in carriers or on seed, particularly where there is likely to be a low level of
contamination.
9.10References
Albareda M., Rodrguez-Navarro D.N., Camacho M. and Temprano F.J. 2008. Alternatives to peat as
a carrier for rhizobia inoculants: solid and liquid formulations. Soil Biology and Biochemistry
40, 27712779.
Bashan Y. 1998. Inoculants of plant-growth promoting bacteria for use in agriculture. Biotechnol-
ogy Advances 16, 729770.
Bullard G.K., Roughley R.J. and Pulsford D.J. 2005. The legume inoculant industry and inoculant
quality control in Australia: 19532003. Australian Journal of Experimental Agriculture 45,
127140.
Burton J.C. 1984. Legume inoculant production manual. University of Hawaii, Dept. of Agronomy
and Soil Science, College of Tropical Agriculture and Human Resources: Hawaii.
Deaker R., Roughley R.J. and Kennedy I.R. 2007. Desiccation tolerance of rhizobia when protected
by synthetic polymers. Soil Biology and Biochemistry 39, 573580.
Herridge D., Gemell G. and Hartley E. 2002. Legume Inoculants and Quality Control, Pp 105115
in Inoculants and Nitrogen Fixation of Legumes in Vietnam ed by D. Herridge. ACIAR Pro-
ceedings 109e, Canberra.
Howieson J.G., De Meyer S., Vivas-Marfisi A., Ratnayake S., Ardley J.K. and Yates R.J. 2013. Novel
Loi A., Nutt B., Yates R. and DAntuono M. 2012. Summer sowing: a new alternative sowing tech-
nique to introduce annual legumes into mixed farming systems. Proceedings of the 16th Aus-
tralian Agronomy Conference, Armidale: Australia.
Lupwayi N.Z., Olsen P.E., Sande E.S., Keyser H.H., Collins M.M., Singleton P.W. and Rice W.A.
2000. Inoculant quality and its evaluation. Field Crops Research 65, 259270.
Marshall K. 1968. Interaction between colloidal montmorillonite and cells of rhizobium species
with different ionogenic surfaces. Biochimica et Biophysica Acta 156, 179186.
Marshall K. 1969. Studies by microelectrophoretic and microscopic techniques of the sorption of
illite and montmorillonite to rhizobia. Journal of General Microbiology 56, 301306.
Materon L. and Weaver R. 1984. Survival of Rhizobium on toxic and non-toxic arrowleaf clover
seeds. Soil Biology and Biochemistry 16, 533535.
Ochman H., Lawrence J.G. and Groisman E.A. 2000. Lateral gene transfer and the nature of bacte-
rial innovation. Nature 405, 299304.
OHara G.W., Howieson J.G. and Graham P.H. 2002. Nitrogen fixation and agricultural practise. Pp
391410 in Nitrogen fixation in the millenium, ed by G.J. Leigh. Elsevier.
Roughley R.J. and Vincent J.M. 1967. Growth and survival of Rhizobium spp. in peat culture. Journal
of Applied Bacteriology 30, 362376.
Roughley R. and Pulsford D. 1982. Production and control of legume inoculants. Pp 193209 in
Nitrogen fixation in legumes, ed. by J. Vincent. Academic Press Australia: Sydney.
Thompson J. 1980. Production and quality control of legume inoculants. Pp. 489533 in Methods
for evaluating biological nitrogen fixation, ed. by F.J. Bergersen. John Wiley & Son Ltd.
Thompson J. 1983. Production and quality control of carrier-based legume inoculants. Information
bulletin no. 17. International Crops Research Institute for the Semi Arid Tropics (ICRISAT).
Vincent J. 1970. A Manual for the practical study of root-nodule bacteria. Blackwell Scientific: Ox-
ford.
186
10.1Introduction
CHAPTER10
Measurement of
nitrogen fixation
D.F. Herridge and K.E. Giller
10.1Introduction
Previous chapters have described how to assess the symbiotic effectiveness of

rhizobia with legume genotypes in a range of experimental systems, from growth
pouches and small tubes containing single plants in a growth room, to large rep-
licated field trials. What is important is to measure the extent of effective nodula-
tion and the yield of biomass and/or grain by the legume.
Rhizobial treatments can be evaluated without quantifying the amount of nitro-

gen being fixed. Measuring the amount of N fixed by single plants will not add
much greater insight to that provided by nodulation and yield data. However,
quantifying N2 fixation is achieved by simply multiplying the plant biomass by tis-
sue %N for each treatment, then subtracting the N in the uninoculated control (N
difference method, see later).
In the field, there are many reasons for estimating N2 fixation. These estimates
can provide information that is critical to understanding the cycling of N in the
soil-plant system and what effect the treatments are having on it. Given sufficient
numbers and locations, the individual values can be aggregated to provide in-
formation about N inputs from legumes and N cycling in systems at regional,
country and global scales. Field-based estimates of N2 fixation can also assist in
determining if different rhizobial treatments have improved N2 fixation in the face
of environmental challenges.
The potentially beneficial economic and environmental impacts of effective ex-

ploitation and management of legume N2 fixation can only be assessed by acquir-
ing quantitative data on N2 fixation. As a cautionary note however, measuring
N2 fixation is often a resource-intensive activity and the researchers must satisfy
themselves that it is necessary.
187
10.2 Methods currently available
10.2 Methods currently available
There are five current methodologies for quantifying N2 fixation by nodulated

legumes:
1. N balance
2. N difference
3. N isotope dilution
15
4. ureide concentration
5. acetylene reduction.
The principles behind these methods and their effective use are detailed in Un-
kovich et al. (2008), a book published by the Australian Centre for Internation-
al Agricultural Research (ACIAR) and available free from the Centres website
(http://aciar.gov.au/publication/MN136). Another recent, comprehensive treatise
on measurement of N2 fixation is Peoples et al. (2009). Other useful publications
include Chalk (1985), Shearer and Kohl (1986), Witty and Minchin (1988), Danso
et al. (1993), Vessey (1994), Unkovich and Pate (2000), Giller (2001) and Peoples
et al. (2002).
The N balance method estimates N2 fixation on an area basis, i.e. kgN/ha, while
the N difference method can be used for either single plants or an area. The 15N
and ureide methods provide estimates of the percentage of total N of the plant or
crop that is derived from N2 fixation (%Ndfa). The N fixed per unit area or unit of
production is then calculated as the product of %Ndfa and the total amount of N
accumulated by the legume. Finally, the acetylene reduction technique assays the
activity of nitrogenase, the enzyme catalysing N2 fixation.
10.3 Which method to use?
There is no single correct way to measure N2 fixation and since all current meth-
odologies have limitations, measuring the exact amount of N fixed continues to
be a challenge. Ideally, several different methods should be used simultaneously,
particularly if they do not rely on the same assumptions. Because this is not always
practical, the choice of method should be made carefully and will often depend on
the resources and expertise available.
The particular analytical equipment need not always be in the researchers labo-
ratory. Mass spectrometers for 15N isotope analysis, for example, are costly and
have very high skill and maintenance requirements. It may be wiser to establish
collaborative activities with specialised groups that have suitable equipment and
expertise. The International Atomic Energy Agency (IAEA) in Vienna, Austria of-
fers services and training facilities and has special programs in agriculture (http://
www.iaea.org/OurWork/ST/index.html).
188
10.4 Sampling for biomass and biomass N
Table10.1 C
haracteristics of methods to quantify biological N2 fixation. The signifies that the
particular method has the characteristic. In the last row, the more $ the higher the cost
(adapted from Unkovich et al. 2008).
Characteristics Non-isotopic methods Isotopic methods
N balance N Ureide C2H2 N natural
15 15
N
difference reduction abundance enrichment
Time integrated
Reference plant needed c
Non-destructive a a
%Ndfa measured d

Quantify kg N/ha fixed
Laboratory
Glasshouse
Field
Possibility to assess fate
of fixed N in system
Short-term
Long-term
Precision low low- good low lowgoodb medium-
medium good
Costs $$ $ $ $ $ $$
a
If only %Ndfa is required
b
Depending on natural enrichment of soil
c
Not when cultivated in N-free media
d
Can be calculated indirectly
The characteristics and potential suitability of the various methods for quantify-
ing N2 fixation in different plant-growth systems, i.e. glasshouse or field, are sum-
marised in Table10.1. It is not an absolute guide but an aid to help the researcher
decide which method to use.
10.4.1 Sampling in space

Accurately estimating plant (crop) biomass and N is critical to measuring N2 fixa-
tion. Insufficient and biased sampling of plants will result in errors. Plant samples
should be collected from the field using a predetermined pattern for all plots or
sampling areas. Generally, samples are taken from replicates as a fixed row length
(e.g. 1m) or quadrat area (e.g. 1.00.5m). Quadrat sizes should reflect the size
of the plots or field and the extent of variation within them. Although it may
sometimes be necessary to use individual plants in field studies, reduced accuracy
may result, and if small numbers are involved, (i.e. <10/replicate) bias (Hunt et al.
1987; Swan et al. 2003).
189
Table10.2 E
ffects of sampling area of field-grown common bean and faba bean
on estimates and variability of shoot dry matter (DM) and shoot %N.
Coefficients of variation (CVs) are given in brackets (adapted from
Peoples et al. 2009).
Quadrat size
0.2 m2 0.5 m2 1.0 m2
Shoot DM (t/ha)
Common bean 2.17 (41%) 1.86 (22%) 1.81 (20%)
Faba bean 10.0 (52%) 8.70 (38%) 7.78 (29%)
Shoot %N
Common bean 3.59 (8%) 3.61 (11%) 3.63 (8%)
Faba bean 2.32 (10%) 2.31 (8%) 2.33 (10%)
In Table 10.2, typical sampling strategies are compared. For both legumes, the
largest sampling area (1m2) gave the lowest estimates of biomass, while the small-
est areas (0.2m2) gave the highest estimates. The coefficients of variation (CVs) for
DM (plant dry matter) also increased substantially as the sampling area became
smaller. Ideally, at least 0.5m2 should be harvested from each replicate plot to pro-
vide a reasonable estimate of DM. The number of replicates required depends on
sample size and field variability. With a 1m2 sample, four to six replicates are suf-
ficient, while up to 10 replicates may be necessary for smaller sampling areas. The
shoot %N values did not vary with sampling area and their CVs were uniformly
small (generally 10% or less).
10.4.2 Sampling in time

The researcher must also decide when to sample for maximum crop N accumula-
tion. With grain legumes, sampling should be at late-pod filling, close to physi-
ological maturity. This usually works well with determinate, bush varieties but un-
fortunately not for indeterminate varieties, with climbing beans being an extreme
example. Sampling too early can result in the grain N being greater than N in the
biomass samples thought to have been taken at maximum N accumulation (Kipe-
Nolt and Giller 1993). However, if sampling is left too late, i.e. between physiologi-
cal maturity and grain harvest, substantial leaf fall can lead to an underestimate
of total N. Judging exactly when to sample for peak N content is difficult and the
ideal time for sampling may vary between varieties (Figure10.1). For very long-
duration, indeterminate varieties, one may have to sample at or very near to final
grain harvest and estimate the amount and N content of fallen leaves. In many
experiments, different sampling times may be needed to accommodate different
durations of different legumes/varieties.
190
10.5 Preparing and analysing plant samples for %N and 15N
Indeterminate variety
Total N content (kg N/ha)
Determinate variety
Days after sowing
Figure10.1 N accumulation in a determinate legume variety and an

indeterminate legume variety. The optimum time for sampling for N accumulation
in the determinate variety is 80 days after sowing. In the indeterminate variety it is
at final harvest.
10.5Preparing and analysing plant samples for %N

and 15N
10.5.1 Drying and weighing plant material

Sample bags should be clearly labelled with the date (including year), plot or treat-
ment number or name, the plant species, and the part of the plant harvested. Plant
samples are best kept in paper bags and allowed to breathe prior to drying. Plas-
tic bags are not recommended. Avoid tight packing of samples as this can result in
stewing of material and subsequent changes in plant N forms. Where plants have
been labelled with isotopes it is important to keep labelled and unlabelled mate-
rial in separate boxes to avoid contamination. The same applies for oven drying:
labelled and unlabelled material should be kept separate, especially when dry and
brittle material is easily detached. Many laboratories maintain separate systems
(rooms, ovens, balances, storages) for 15N-labelled and unlabelled samples.
Samples should be oven-dried as soon as possible after collection. If this cannot

be arranged within a few hours, plant samples should be kept in a coolroom or
cool place overnight. The objective of drying is to remove water from the plant
material without loss of organic compounds, and without the growth of bacteria
and fungi. Where there is insufficient oven space for a large amount of sample
material, all samples should receive the same treatment. Preliminary air drying in
a glasshouse may be possible in dry environments. If plant samples have attached
soil or dust, it should be washed off while the plants are fresh as it cannot be sepa-
rated from dried material.
191
10.5 Preparing and analysing plant samples for %N and 15N
Plant material must be dried to constant weight at 6580C, usually for at least
48hours, in an air-circulating oven. Very bulky samples can be spread on metal
trays and returned to the sample bags when dry. Sample bags must not be packed
tightly into an oven as this can cause a fire. Drying plant material at 100C is not
recommended as rapid proteolysis (protein breakdown), thermal decomposition
and loss of N may occur. High temperatures also increase the risk of an oven fire.
The dry weight of the total sample should be recorded immediately after drying,
before any grinding and subsampling. Samples can conveniently be weighed in
their paper bags with the bag weight (the average weight of similarly oven-dried
bags) subtracted from the sample weight. Once dried and weighed, the samples
can be stored indefinitely provided they are kept in dry, airtight conditions. How-
ever, ground-up samples require less storage space and they can be kept in con-
tainers to prevent damage from insects and rodents. Samples should be stored in
paper bags and cardboard boxes for only short periods (days to weeks) to avoid
possible sample damage or loss.
10.5.2 Grinding and subsampling plant material

As the amounts of sample used for %N and 15N analysis are very small with respect
to the sample harvested in the field, great care is needed in preparing subsamples
for analysis. In particular, the relative proportion of the different plant parts (e.g.
stems, leaves and immature pods) in the subsample must be representative. After
chopping, or at different phases of subsampling, quartering can be useful. The
plant sample is spread thinly on a sheet of plastic/paper or other clean surface, one
quarter taken and the rest discarded. This procedure is repeated until the subsam-
ple has the required size, the aim being to prevent the inevitable settling (layering)
of samples according to particle size in upright containers.
Although the amount milled depends on constraints of time and the capacity of
the mill, the larger the sample that can be hammer milled and then subsampled
for analysis the better. To produce a fine powder for 15N analysis, a small amount
of sample needs to be ball milled. Additional details on preparation of samples for
analysis of %N and 15N can be found in Unkovich et al. (2008).
10.5.3 Analysis of plant material for %N and 15N

Plant material can be analysed for %N using one of three standard methods
Kjeldahl digestion, dry combustion or near infra-red spectroscopy (NIR). Analy-
sis of 15N requires an isotope-ratio mass spectrometer or emission spectrometer
for higher-level enrichments, i.e. greater than 0.05 atom % 15N excess.
The Kjeldahl method involves digesting the plant sample with a catalyst in hot
sulfuric acid, converting the organic N to ammonium (NH4+) then determining
the concentration of NH4+. The method uses limited resources and allows analy-
sis of total N and 15N from a single sample distillation. Full details of the meth-
odology are given in Peoples et al. (1989). Unkovich et al. (1993) provide details
192
10.6 Accounting for below-ground N
of modifications required for using the sample distillates for high-precision 15N
analysis.
Dry combustion avoids the wet chemistry of the Kjeldahl method and is the
choice of most laboratories. Its use, however, requires access, directly or via col-
laboration, to a combustion analyser. This instrument provides rapid analysis of
large numbers of samples but is expensive and requires a skilled operator. In re-
cent years, the dry combustion instruments have been coupled to isotope-ratio
mass spectrometers so that both total N and 15N can be determined on a single
sample. A number of commercial services for such analyses are listed in Appendix
9 of Unkovich et al. (2008).
One such service is at the University of California Davis Stable Isotope Facil-
ity http://stableisotopefacility.ucdavis.edu/. Samples for %N and 15N analysis
have to be weighed into tin cups using a balance that can measure accurately to
0.00001g. Samples need to be dried in an oven at 6070C and kept in a desic-
cator prior to weighing. The amount of sample depends on the %N in the sample;
about 2.5mg for legume seeds (4%N) and pods, about 5mg for legume shoots
(2%N) and 10mg for cereal straw (1%N). As the tin cups are very full with the
10mg samples, care must be taken to ensure they are closed well so that they do
not leak in transit.
The third method for %N analysis, NIR, is rapid, inexpensive, non-destructive

and is repeatable in routine use (Williams et al. 1983) but its reliability must be
constantly validated by cross-referencing data against Kjeldahl or dry combustion
analyses. This ability to make multiple determinations (e.g. plant moisture, carbo-
hydrates) further enhances the value of the NIR technique. It cannot be linked to
isotope-ratio mass spectrometry so samples must be analysed separately for 15N.
Additional details (and references) about using the methodology are provided in
Unkovich et al. (2008).
10.6 Accounting for below-ground N
Total plant (crop) biomass and N have usually been determined from measure-
ments of shoot biomass, assuming that N in the roots represents only a small frac-
tion (515%) of the total plant N, and that shoot N provides a reasonable approxi-
mation to total plant N. However, that assumption has been proven incorrect;
below-ground N associated with, or derived from, roots can represent 3050% of
the total plant N of both legumes and cereals (e.g. Russell and Fillery 1996; Mc-
Neill et al. 1997; Rochester et al. 1998; Unkovich and Pate 2000; Khan et al. 2002).
Total inputs of fixed N could therefore be 50100% greater than those determined
from measurements based only on the shoots.
There is no single value for below-ground N as a fraction of total plant N; pub-

lished estimates reflect the influence of species, soil and climate. To account
for below-ground N when calculating total plant or crop N and N2 fixation, we
193
10.7 Nitrogen difference method
suggest multiplying shoot N by 2.0 for chickpea (assumes 50% of plant N is below
ground), 1.5 for soybean (assumes 33% below-ground N) and 1.4 for the remain-
der of the grain legumes (assumes 30% below-ground N). With the pasture/fod-
der legumes, multiply shoot N by 2.0 for lucerne and 1.4 for the remainder. Use
a factor of 1.4 with cereal crops. We believe that the errors associated with these
approximations are far less than those incurred by ignoring below-ground N or by
using values for physically-recovered roots.
10.7 Nitrogen difference method
With this method, total N accumulated by N2-fixing plants is compared with that
of neighbouring non-N2-fixing plants, with the difference in the two assumed to
be due to N2 fixation. Nitrogen difference is a simple, low-cost method usable
when facilities for only DM determination and %N analysis are available. Thus:
N2 fixed=(legume plant N) (non-N2-fixing control plant N)
Where plant N is derived from plant DM and %N:
This method assumes that the N2-fixing plants assimilate the same amount of soil
mineral N as the neighbouring non-N2-fixing plants. In infertile soils with little
mineral N, the method can be highly accurate. It may be less useful in soils with
moderate to high levels of mineral N because differences between N2-fixing and
non-N2-fixing plants in root morphology and rooting depth can result in differ-
ent capacities to capture the soil N (Chalk 1998). The method has little value for
on-farm surveys where discrete areas of appropriate non-N2-fixing plants may not
be present.
A successful application of the N-difference method is shown in Table10.3. Pi-

geonpea genotypes of different maturities were grown on a sandy clay-loam in
Zimbabwe (Mapfumo et al. 1999) with maize plants of matching maturities as the
non-N2-fixing references. As expected, shoot N increased with increasing growth
duration for both the N2-fixing pigeonpea and the non-N2-fixing maize; the esti-
mated shoot N fixed also increased with time.
Table10.3 E
stimations of N2 fixation by five genotypes of pigeonpea grown on
a sandy clay-loam in Zimbabwe using the N difference method with
maize of matching maturity as the non-N2-fixing controls (Mapfumo et
al. 1999).
Maturity of Shoot N pigeonpea Shoot N non-N2- Shoot N fixed
pigeonpea (kg/ha) fixing reference (kg/ha)
(kg/ha)
Short 943 8 135
Medium 145 54 91
Long 172183 71 101112
194
10.8 15N dilution (and incorporation) methods
Unfortunately, it is not always this straightforward. N2 fixation by seven tropi-

cal forage legumes in Brazil was estimated (data for three legumes shown in Ta-
ble10.4). Total N accumulation by the legumes ranged from 102 to 117kgN/ha,
and for the three grass species used as non-N2-fixing reference plants 3356kgN/
ha. Two grass reference species (Brachiaria brizantha and Panicum maximum)
provided very similar estimates of N2 fixation while the estimates using the grass
that accumulated the most total N (Brachiaria arrecta) were around 20kg/ha low-
er. It is not clear which reference plant provided the most correct estimate; ad-
ditional information such as growth duration of the different species would be
needed to determine this.
This method for quantifying N2 fixation is less popular now and has been largely
replaced by 15N and ureide methods. However, the method remains appropriate
and can provide simple, accurate estimates of legume N2 fixation for little cost and
without the need to access sophisticated analytical facilities.
Table10.4 E
stimations of N2 fixation by three legumes using the N difference
method with three grass species as the non-N2-fixing controls (from
Unkovich et al. 2008; data from Viera-Vargas et al. 1995b)
Plant Total shoot Legume N2 fixation estimated using
N (kg/ha) three different reference plants
(kgN/ha)
Brachiaria Panicum Brachiaria
brizantha maximum arrecta
Legumes Centrosema (hybrid) 107 69 74 52
Galactia striata 117 79 84 62
Desmodium ovalifolium 102 64 69 47
Grasses Brachiaria brizantha 38
Panicum maximum 33
Brachiaria arrecta 56
Of the two stable N isotopes, 14N and 15N, the lighter isotope 14N is naturally much
more abundant (>99.6% of the N in the atmosphere). The isotopic abundance of
the minor isotope (15N) is expressed as a percentage of the total N present, atom%
15
N, or as atom% 15N excess (Table10.5):
Atmospheric N2 has a constant abundance of just 0.3663 atom% 15N (Table10.5;

Mariotti 1983). Only small natural variations around this value occur in other N
pools (e.g. soil N) within the biosphere (Hgberg 1997). Natural variations in 15N
195
abundance are usually expressed in terms of units, which are the parts per thou-
sand () deviation relative to the international standard of atmospheric N2, i.e.
0.3663 atom% 15N (Table10.5). The units are calculated from atom% 15N values:
The natural abundance of atmospheric N2 by definition has a 15N of 0 (Ta-

ble10.5). Values for plants and soils can either be positive (more 15N than in at-
mospheric N2), or negative (less 15N than in atmospheric N2).
To illustrate natural abundance calculations, consider a plant sample with 0.36784

atom% 15N. It will have a 15N value of + 4.2:
15
N was first used to detect N2 fixation by bacteria in the 1940s (Burris et al. 1942);
the unavailability of 15N-enriched materials and mass spectrometers to analyse
the samples severely restricted use of 15N methods. Since the 1970s, mass spec-
trometers and isotopes have been more generally available, facilitating greater use
of 15N methodologies.
Experimental protocols involve: (i) 15N2 incorporationnamely labelling the N2

gas in the atmosphere surrounding the N2-fixing plants (Warembourg et al. 1982)
and measuring incorporation of 15N by the plants; and (ii) 15N isotope dilution
growing the plants in 15N-enriched soil or other medium and calculating the ex-
tent of dilution of 15N in the plants by atmospheric (fixed) 14N (McAuliffe et al.
1958; Chalk 1985). A later variation of 15N isotope dilution, known as the natu-
ral 15N abundance method, utilised the natural 15N enrichment of soils, thereby
avoiding the need to add 15N-enriched materials (Shearer and Kohl 1986).
Table10.5 Terms associated with 15N stable isotope methods

Term Definition
Atom% 15N Abundance of 15N atoms as a percentage of the total
Natural abundance Atom% 15N naturally present in materials

15
N abundance of atmospheric N2 0.3663 atom% 15N
N ()
15
Sample natural abundance expressed as parts per
thousand relative to atmospheric N2
15N of atmospheric N2 0
15
N-enriched nitrogen Nitrogen with an atom% 15N greater than atmospheric N2
Atom% N excess
15
A measure of a samples 15N content above the
atmospheric N2: sample atom% 15N 0.3663
Labelled nitrogen Material generated with a specific 15N enrichment
%Ndfa The percentage of plant N derived from atmospheric N2
196
N2 incorporation is the most direct measure of N2 fixation but is limited to short

15
experimental periods in a laboratory or growth chamber.
N isotope dilution with artificially enriched soil was, until a few years ago, used
15
widely to quantify %Ndfa and hence N2 fixation of legumes (Chalk and Ladha
1999). In recent years, the natural abundance method has gained prominence for
work in both experimental plots and in farmers fields.
N isotope dilution provides an estimate of the plants dependence on N2 fixation

15
(%Ndfa) as follows:
Where:
x=15N content of the non-N2-fixing reference plants (surrogate for plant-available

soil N)
y=15N content of the N2-fixing legume
B=a measure of the 15N content of the target legume fully dependent on N2 fixa-
tion for growth. This is generally assumed to be zero in 15N enrichment studies but
needs to be considered when using 15N natural abundance (see below).
The estimate of %Ndfa is then combined with total legume N to calculate plant
N2 fixed. If the aim is to estimate the total amount of N fixed by a legume crop
in a season, the crop should be sampled for biomass and %N at the time of peak
biomass (see Section 10.4):
Where legume N is calculated as legume DM (%N)/100 (see above). Note that

legume N should include the N in the soil derived from the plant, i.e. the N in
roots, nodules and the soil around the roots. Accounting for below-ground N is
discussed in Section 10.6.
Advantages of the method are that: (1) it provides a time-integrated estimate of

%Ndfa for a period of growth; (2) amounts of N2 fixed can be estimated from a
single analysis of the 15N contents of the N2-fixing legume and the non-N2-fixing
reference combined with a measure of legume biomass and %N; and (3) the 15N
natural abundance variation of the method allows N2 fixation to be monitored
almost anywhere if both N2-fixing and non-N2-fixing plants are present, e.g. in
farmers fields.
The limitations of the method are: (1) that the 15N composition of plant-availa-
ble soil N can change with soil depth and with time during the growing season,
particularly where 15N-enriched materials have been applied to soil (Witty 1983;
Chalk 1985); (2) that the efficacy of the non-N2-fixing reference plant to provide
an accurate measure of the isotopic composition of plant-available soil N can be
compromised by differences in the rooting depths and patterns of N uptake of the
197
N2-fixing and reference plants; (3) the high cost of 15N-enriched materials (when
used); and (4) that the 15N abundance of plant-available soil N can be either too
low and/or too variable for the methodology.
10.8.1Natural 15N abundance

The non-N2-fixing reference for this method can be an unnodulated or non-N2-
fixing legume or a non-legume growing in the vicinity. The equation for estimat-
ing %Ndfa using the natural 15N abundance variation of the N isotope dilution
method is as follows:
In a hypothetical example, if the 15N of the legume was +4.2, the 15N of the
non-N2-fixing reference was +6.2 and the B value was 0.0, then %Ndfa is
estimated to be 32%:
As stated above, B values are almost never 0.0, generally ranging between 1.0
and 2.0. Table10.6 presents average B values for a range of crop and fodder/
tree legumes; full data sets and a fuller discussion can be found in the Appendix
section of Unkovich et al. (2008).
Table10.6 A
verage B values for a number of tropical and subtropical crop, forage
and shrub legumes (source: Unkovich et al. 2008)
Tropical and subtropical B value Tropical and subtropical B value
crop legumes Shoot 15N forage, shrub and tree Shoot 15N
() legumes ()
Vigna angularis 0.91 Centrosema 1.65
Vigna mungo 1.75 Desmodium 1.14
Phaseolus vulgaris 2.16 Macrotyloma axillare 1.83
Vigna unguiculata 1.61 Macroptilium atropurpureum 2.35
Vigna radiata 2.05 Calopogonium 0.95
Arachis hypogaea 0.88 Crotalaria 1.08
Cajanus cajan 1.12 Dolichos lablab 1.09
Vigna umbellata 0.91 Mucuna pruriens 1.82
Glycine max 1.83 Pueraria 1.22
Psophocarpus 1.54 Calliandra 0.90
tetragonolobus
With soybean, more substantial 15N:14N discrimination may be associated with

N2 fixation when the plants are nodulated by certain strains of Bradyrhizobium
elkanii (see Unkovich et al. 2008) resulting in shoot B values as low as 4.5.
However, without evidence to the contrary, the average for soybean in Table10.6
of 1.83 is a reasonable value.
198
The choice of non-N2-fixing reference plant(s) to provide an estimate of the 15N

enrichment of the soil is critical. The impact of the reference on the calculations of
%Ndfa is generally most important when the %Ndfa of the legume is low (<30%)
or where the 15N of the reference is less than about 4. Unkovich et al. (2008)
strongly recommend the use of more than one reference species, e.g. a broad-
leaf (dicotyledonous) non-legume plus a cereal or grass. If the 15N values differ
amongst reference species, %Ndfa should be calculated for each species to show
the range and to derive a mean %Ndfa.
Adjei-Nsiah et al. (2008) used this approach to quantify N2 fixation for five cul-
tivars of cowpea in the forest/savannah transitional zone of Ghana (Table10.7).
The average 15N values for the non-N2-fixing references ranged between 3.38 and
6.78; for cowpea, 15N values ranged between 0.21 to 1.34.
Table10.7 R
anges and average %Ndfa estimates for cowpea genotypes in the
forest/savannah transitional zone of Ghana (source: Adjei-Nsiah et al.
2008)
Cultivar Range of %Ndfa estimates Average %Ndfa
Adom 6479 75
Asontem 6378 73
Ayiyi 4668 61
IT810D-1010 6077 71
Legion prolific 6881 77
The ranges of %Ndfa estimates are shown in the second column of Table10.7; the
average %Ndfa estimates, those most likely to be used in further analysis of the
data are shown in the third column.
Additional details of the natural 15N abundance method, particularly related to

experimental design and working in farmers fields, are to be found in Chapter8
of Unkovich et al. (2008). Plant sampling protocols are considered in Section
10.4. Most 15N values in agricultural systems range from 2 to +12 (Letolle
1980); values less than 3 or greater than +15 should be carefully examined
with re-analysis usually warranted.
10.8.2 15N enrichment

The 15N enrichment methodology was most widely used in the 1970s to 1990s,
before improvements in mass spectrometry in the 1980s led to the development of
the 15N natural abundance technique. Where the 15N of plant-available N is <2
and the 15N methodology should not be used, or the researcher cannot access
high-precision mass spectrometers (15N 0.3, 0.0001 atom% 15N), plant-
available soil N can be artificially enriched with 15N-labelled materials.
Legume and non-N2-fixing reference plants are then grown in soil receiving the
same 15N-labelled material, usually fertiliser. Shoots of both sets of plants are
199
analysed for 15N, and the percent of N derived from the atmosphere (%Ndfa) by
the legume is calculated (McAuliffe et al. 1958) as:
The principal assumption is that the 15N enrichment of the non-N2-fixing refer-
ence plants accurately reflects the 15N enrichment of soil N taken up by the leg-
ume. The 15N enrichment of the soil N therefore needs to be relatively constant
over time and space, and the time course and depth of soil N uptake by the refer-
ence and N2-fixing plants identical.
Because materials labelled with 15N are expensive, the amount of 15N used is mini-
mised. There is no single correct method for the addition of 15N to label plant-
available N in the soil. The most common, however, is to add 15N-labelled in-
organic salts (e.g. (NH4)2SO4, NH4Cl, KNO3, NH4NO3) or 15N urea fertiliser as
a liquid using a watering can, by spraying or injecting into soil or as a solid by
broadcasting, banding or mixing with soil. Some possible sources of stable iso-
tope-labelled materials are given in Appendix 8 of Unkovich et al. (2008).
Following addition of 15NH4+ or 15NO3 to soil, the mineralisation of N from unla-

belled soil organic N causes the 15N enrichment of the plant-available soil N pool
to decline. Resulting differences in the 15N enrichment of the N taken up from the
soil by N2-fixing and reference plants may undermine the principal assumption
of this method.
Further, the added 15N is not distributed evenly with regard to depth when ap-
plied to the soil surface. Thus, the 15N enrichment of soil mineral N can change
with both time and depth during plant growth. Practices to reduce the impact of
uneven distribution of 15N in space and in time include:
regular additions of 15N-labelled inorganic N to the soil (Viera-Vargas et al.

1995a)
addition of soluble C to immobilise mineral N (Giller and Witty 1987; Boddey

et al. 1995)
the use of 15N-labelled plant residues to provide mineralised N of a more con-

stant value (Watanabe et al. 1990)
using residual 15N fertiliser carried over in soil from a previous experiment
(McNeill et al. 1998)
the use of a series of different reference plants to capture variations in 15N en-
richments in space and time (Viera-Vargas et al. 1995a)
repeated sampling and analysis of extractable soil N over time (Chalk 1996)
combinations of the above.
200
All authors highlight the fact that using non-N2-fixing reference plants to estimate
legume 15N uptake from the soil is the principal weakness of the isotope dilu-
tion methodology. However, access to a yield-independent and time-integrated
estimate of %Ndfa makes it worthwhile to persist with this potentially powerful
technique.
The next question is how much 15N to add? The total amount of N applied in
the 15N-labelled material should be small enough that it does not influence N2
fixation, i.e. <5kgN/ha). It is also much less expensive per gram of 15N to buy
10atom%15N than 99atom%15N; only a few situations would justify enrichments
of >10atom%15N excess.
The areas of crop to be labelled with 15N (microplots) are typically only 12m2
per replicate, and usually much smaller than the areas sampled for crop DM. Steel
boxes are often placed into the soil around the 15N-labelled area to reduce lateral
movement and runoff of 15N and to prevent the scavenging of unlabelled N by
plant roots (Sanchez et al. 1987). Separate plant biomass samples for DM estima-
tion are usually taken from the plot areas surrounding the 15N-labelled microp-
lots, with plant 15N and %N determined from samplings within them.
The non-N2-fixing reference plants should be as similar as possible to the legume

in growth habit, duration etc. Because it is very difficult to find a single species that
fulfils all of the above requirements, we recommend using more than one refer-
ence species in any given experiment.
A typical set of data from a 15N isotope dilution experiment is given in Table10.8
with %Ndfa for the 128-day-old plants calculated as follows:
Note that %Ndfa would be calculated for each replicate, to allow for statistical
analysis of the data.
Table10.8 15
N enrichments of shoots of lupin and wheat sampled from 15N-labelled
microplots and estimates of %Ndfa at 123 and 193 days after sowing
(data of Evans et al. 1987)
Species Days after atom% 15N atom% 15N Lupin
sowing excess %Ndfa
Lupin 128 0.4344 0.0681 79
Wheat 0.6852 0.3189
Lupin 193 0.4112 0.0449 85
Wheat 0.6561 0.2898
The enriched and natural abundance variations of the 15N isotope dilution
method are arguably the benchmark for quantifying N2 fixation by nodulat-
ed legumes. Use of centralised laboratories, e.g. University of California, Davis
(http://stableisotopefacility.ucdavis.edu/) for analysing 15N (and %N) through
201
10.9 Ureide (N solute) method
fee-for-service or collaboration provides a means for all researchers to use these

techniques.
The basis of the ureide method is the export by many agronomically-impor-

tant tropical legumes (e.g. soybean, common bean, in the tribes Phaseoleae and
Desmodieae within the Papilionoideae subfamily) of allantoin and allantoic acid
(collectively known as ureides) as the products of N2 fixation from their nodules
to the shoots (Figure10.2).
Ureide exporting legume Amide exporting legume

Ureides Amino Amino
Nitrate Nitrate
N2 fixation N2 fixation
N2 N2
Nitrate Nitrate
N2 N2
reduction reduction
N2 N2
Soil Soil
nitrate nitrate
Figure10.2 Assimilation and export of N from the nodulated roots of ureide- and amide-
producing legumes. The ureide species have three major groups of N solutes in the xylem stream:
ureides, amino compounds and nitrate, while the amide legumes have only two: amino compounds
and nitrate.
Other legumes produce and export the amides, asparagine and glutamine, from
their nodules. Nitrogen is also taken up from the soil by the legume roots, prin-
cipally as nitrate. Once inside the plant, the N solutes derived from soil mineral
N are transported in the xylem as free nitrate or, after reduction and/or ammo-
nium metabolism in the root, as organic products (principally asparagine and
glutamine).
202
100 100
(a)
(b)
80 Nitrate 80
Amino-N
%N composition
%N composition
Ureides
60 60
40 40
20 20
0 0
5 15 25 35 45 55 65 75 85 95 5 15 25 35 45 55 65 75 85 95
%Ndfa %Ndfa
Figure10.3 Changes in the composition of N solutes in: (A) root-bleeding and (B) vacuum-
extracted xylem sap of nodulated soybean, supplied with different concentrations of 15N-labelled
nitrate to generate different values of %Ndfa (data from Herridge and Peoples 1990).
With the ureide legumes, the ratio of ureide N to total N in xylem sap or stem
segments is highly correlated with %Ndfa (Figure10.3). Quantifying % ureide-N
provides a means of estimating %Ndfa. Although not applicable to all legumes,
or to other N2-fixing associations, the technique has been widely used with both
experimental and non-experimental (farmer) crops.
The relationships between %ureide-N in xylem sap or stem segments and %Ndfa
need to be specified for each legume species. Since the technique was established
about 30 years ago, many experiments have defined legume species as either urei-
de (Table10.9) or amide exporters and calibrated particular species.
The % ureide-N in xylem sap is calculated as:
where a is the molar concentration of ureides (ureides contain four N atoms per
molecule), b is the molar concentration of nitrate-N and c is the molar concentra-
tion of -amino-N (Herridge 1984).
% ureide-N in extracts of whole stems or stem segments is calculated as:
where a and b are, respectively, the molar concentrations of ureides and nitrate.
- Amino-N is not included in this equation, principally because the relationship
between % ureide-N and %Ndfa was sufficiently robust without it (Herridge and
Peoples 1990).
203
Table10.9 S
ome of the legumes that transport the ureides, allantoin and allantoic
acid, as the dominant products of N2 fixation in xylem sap (source:
Unkovich et al. 2008).
Grain legumes Forage and tree legumes
Soybean (Glycine max) Calopogonium caeruleum
Pigeon pea (Cajanus cajan) Centrosema pubescens
Cowpea (Vigna unguiculata) Codariocalyx gyroides
Mung bean (V. radiata) Desmodium ovalifolium.
Black gram (V. mungo) D. rensonii
Adzuki bean (V. angularis) Siratro (Macroptilium atropurpureum)
Rice bean (V. umbellata) Kudzu (Pueraria phaseoloides)
Bambara groundnut (V. subterranea) Hardenbergia spp.
Common bean (Phaseolus vulgaris) Hyacinth bean (Lablab purpureus)
Lima bean (P. lunatus)
Runner bean (P. coccineus)
Winged bean (Psophocarpus tetragonolobus)
Guar (Cyamopsis tetragonoloba)
Kerstings groundnut (Macrotyloma geocarpum)
Relationships between %ureide-N (y) and %Ndfa (x) have now been published
for a number of species of crop, forage and shrub legumes (see Unkovich et al.
2008). To estimate %Ndfa, the %ureide-N values are inserted into the calibration
equations for the particular species. For example, for soybean xylem sap these are:
y=0.64x + 7.7 (for vacuum-extracted sap during vegetative growth and flowering)
y=0.64x + 15.9 (for vacuum-extracted sap during pod-fill)
y=0.83x + 4.8 (for root-bleeding sap during vegetative growth and flowering)
y=0.67x + 21.3 (for root-bleeding sap during pod-fill)
The %Ndfa value, when combined with measures of legume total N, can then be
used to calculate amounts of N fixed during a period of growth (e.g. Herridge et
al. 1990).
10.9.1 Sampling of legumes for N solutes

Samples can be collected from legumes for N solute analysis as vacuum-extract-
ed xylem sap (Herridge 1984, Herridge et al. 1988), root-bleeding xylem sap
(Norhayati et al. 1988) or aqueous extracts from stem segments or petioles (Her-
ridge and Peoples 1990). Procedures for the three methods are described below.
Irrespective of the method used, it is advisable to sample the plants between 0900
and 1600hours to avoid error due to diurnal variations in relative ureide-N (e.g.
Herridge et al. 1988; 1996).
204
10.9.1.1 Vacuum-extracted sap
With most species, sufficient (0.10.3mL/plant) xylem sap can be recovered from
each whole stem or stem section of field-grown crop legumes for complete N-
solute analysis. Thus, the method may also be used as a non-destructive assay on
individual plants (Herridge et al. 1988).
Equipment requirements
Items needed for vacuum-extraction of xylem sap are:
sharp secateurs or pruning shears

syringe needles (19 or 20 gauge)
silicon- or latex-rubber tubing of a range of internal diameters (315mm)
fittings or adaptors (e.g. disposable micropipette tip cut to size)
5mL Vacutainers
a vacuum source, which may be a hand-held vacuum pump (e.g. Nalgene,
Sybron Corp., Rochester, New York, USA, as depicted in Figure10.4), a foot
pump or a laboratory vacuum pump powered by a petrol-run generator or a
car battery (e.g. a Waters/Millipore DOA-V130-BN vacuum pump with a Ka-
wasaki GA1000A portable generator).
Figure10.4 Apparatus for vacuum-

extraction of xylem sap using a hand-
held vacuum pump. The base of a
freshly-detached stem is placed into an
appropriately-sized silicon rubber tubing
sleeve (A) and fitted onto a syringe needle
(using an adaptor (B) if necessary). The
syringe needle is then inserted through
the rubber stopper of a Vacutainer (C)
connected to a vacuum pump (D) via
a second syringe needle connection
(source: Unkovich et al. 2008).
205
The advantage of the laboratory vacuum pump is that a manifold can provide
more than one vacuum line for several operators to simultaneously sample sap
from different plants. Photographs of a foot pump and powered vacuum pump
plus battery are provided in Unkovich et al. (2008).
Procedure
1. For crop legumes, use secateurs to cut a stem >3 mm in diameter close to
ground level. For tree or shrub legumes, 1.0 m long stems or canes are cut
from the main plant and used for sap sampling (Herridge et al. 1996).
2. The detached stem is immediately inserted into a sleeve of silicon- or latex-

rubber tubing with an internal diameter slightly smaller than the stem (Fig-
ure10.4), and fitted onto a syringe needle using an adaptor of appropriate size.
3. The needle is then pushed through the rubber stopper of a 5mL Vacutainer
that has been linked to a vacuum pump via another syringe needle connection
and a flexible plastic-tubing line. NB: the base of the two syringe needles must
not be level or sap may be sucked directly into the vacuum pump line rather
than collecting in the Vacutainer.
4. A vacuum (6070kPa) is applied and 34cm segments of the stem are then
cut with secateurs successively from the top to the bottom of the shoot. Entry
of air at the cut surface displaces the xylem sap from the base of the stem to
be collected within the Vacutainer. Vacuum extraction should commence im-
mediately after detachment of stems from the root. A time delay of more than
five minutes can introduce errors because of changes in concentrations of N
solutes (see Herridge et al. 1988). The technique will not work if the crop is
infested with pests, such as stem borers, since a vacuum on the stem cannot
be maintained. Avoid the needle becoming blocked with debris by clearing it
frequently and changing needles periodically.
5. Sap samples should be kept chilled on ice until frozen at 15C for long-term
storage or, if ice is unavailable, stabilised immediately after extraction by
adding an approximately equal volume of ethanol to the sap collected in the
Vacutainer.
10.9.1.2 Root-bleeding sap
1. The shoot is cut below the first node close to ground level with secateurs or a
very sharp blade.
2. A sleeve of silicon- or latex-rubber tubing, 24cm long with an internal di-

ameter slightly smaller than the stem, is placed over the exposed root stump
(Figure10.5).
3. The sap exuding under root pressure can easily be collected from within the tub-
ing sleeve using a Pasteur pipette or syringe. We recommend that root stumps
be allowed to exude for no more than 20 to 30 minutes. The accumulated sap
206
should be collected about every 10 minutes, placed in a sealable tube or vial

(e.g. Vacutainer, Becton Dickinson, Rutherford, New Jersey, USA) and kept
on ice to minimise potential decomposition or metabolism.
4. Sap samples should be kept chilled on ice until frozen at 15C for long-term
storage or, if ice is unavailable, stabilised immediately after collection by mix-
ing with an equal volume of ethanol in the collection tube.
Figure10.5 Sampling of root-

bleeding xylem sap. A length of
silicon rubber tubing (A) is placed
over the root stump following
decapitation of the shoot. The
sap (B) collects in the tubing to
be drawn out of the tubing with a
Pasteur pipette or syringe (source:
Unkovich et al. 2008)
10.9.1.3 Stems and stem segments
The stem, because it is mainly involved in transport of N solutes, is the most suita-
ble plant part for N-solute extraction and analysis. Leaves are not suitable because
they rapidly metabolise incoming N-compounds and tissue N-solutes can vary
with changes in plant metabolism unrelated to N2 fixation. Relative ureide N in
stems is insensitive to diurnal fluctuations and unchanged by storage at 2030C
for up to 24hours after harvest before being oven dried.
The advantages of using dried and ground stems or stem segments are ease of sam-
pling and the 24-hour stability of the samples. This may also be the only means of
field sampling small plants or legumes that do not readily yield vacuum-extracted
or root-bleeding xylem sap. The disadvantages are the additional steps in drying
and grinding the stem segments and in solute extraction.
Procedure
1. Stems and stem segments are harvested and leaves removed.
2. Samples are placed in clearly labelled bags and dried at 6580C in a forced-
air oven for two days.
3. Tissue is ground to pass through a 60-mesh (1.0mm) screen and stored in a

dry place until extraction.
207
4. Subsamples (0.5g) of dried and ground material are weighed and transferred
to 100mL beakers or Erlenmeyer flasks.
5. Distilled water (25mL) is added to each subsample, which is then boiled for
one to two minutes. An electric frying pan half-filled with sand can be used.
6. Extracts are filtered while hot through 15cm filter paper (Whatman No. 40)
in a funnel into a 50mL volumetric flask. The residue is washed onto the filter
and rinsed with a little distilled water.
7. When contents of flask are cool, the volume is made up to 50mL with distilled
water.
8. The extract can be stored indefinitely in a freezer in small vials or flasks until
analysis of N solutes.
10.9.2 Analysis of xylem sap and aqueous stem extracts

The following items are required for analysis of xylem sap or extracts:
weighing balances (accurate to 0.1 and 1mg)

test tube racks and glass test tubes to match (e.g. 85 15mm)
micropipettes and tips, and/or dispensers to cover 220 L, 50200 L,
0.21mL, 15mL ranges
vortex mixer
boiling-water bath
cold-water ice bath (e.g. ice in a foam box), or refrigerated water bath
spectrophotometer or colorimeter.
10.9.2.1 Ureide assay (Young and Conway 1942)
Reagent preparation
a. Sodium hydroxide (0.5 N NaOH)
Add 20g NaOH to 1L distilled H2O and pour into Dispenser 1.
b. Hydrochloric acid (0.65 N HCl)
Add 6.5mL concentrated HCl to 100mL distilled H2O.
c. Phenylhydrazine hydrochloride (0.33% (w/v))
Add 0.33g phenylhydrazine hydrochloride to 100mL distilled H2O.
Take great care with the phenylhydrazine, as it is toxic. Make fresh daily.
d. The 0.65 N HCl and the phenylhydrazine are now mixed together; total vol-
ume 200mL.
208
(Pour the HCl and phenylhydrazine mix into Dispenser 2.)
e. Potassium ferricyanide (K3Fe(CN)6 1.67% (w/v))
Add 1.67g to 100mL distilled H2O. Make fresh daily.
Caution: K3 Fe(CN)6 is highly toxic.
f. Concentrated hydrochloric acid (10N HCl)
Decant 400mL concentrated HCl from the bottle.
g. The concentrated HCl and 1.67% K3Fe(CN)6 are mixed together; total vol-
ume 500mL, and placed in a freezer (if possible) or a fridge/ice bath. (Pour
the concentrated HCl and K3 Fe(CN)6 mix into Dispenser 3; leave in fridge.)
Make fresh daily.
h. Ureide (allantoin) standard1mM
Add 39.53mg allantoin to 250mL distilled H2O.
i. Distilled H2O (Dispenser 4)
The 1mM ureide standard (h) is used to make the following concentrations for a
standard curve determination:
Concentration 1.0 mM Water (mL)

Typical standard curve for ureides
(mM) allantoin (mL)
1.6 Range 00.25 mol
0.00 0.0 100.0
0.01 1.0 99.0
y = 5.259x + 0.0454
0.02 2.0 98.0 1.2 R2 = 0.9928
O.D. (525 nm)
0.04 4.0 96.0

0.10 10.0 90.0 0.8
0.4
0
0 0.05 0.1 0.15 0.2 0.25
mol ureide
Analysis
1. A standard curve covering the range 00.10 mM allantoin should be con-

structed periodically as part of the laboratorys quality control. Add 2.5mL
of each of the five concentrations to duplicate test tubes corresponding to 0,
0.025, 0.05, 0.1 and 0.25mol ureide.
2. Add 0.5mL 0.5N sodium hydroxide (NaOH) using Dispenser 1 sufficiently

forcefully that additional mixing is unnecessary.
209
3. Place rack of tubes in a boiling water bath for 10minutes. Make sure the level
of the boiling water is above the contents of the tubes.
4. Remove from water bath and place on bench.
5. Add 1.0mL HCl/phenylhydrazine mix using Dispenser 2. Again, the dispens-

er should be used in such a way that additional mixing is unnecessary.
6. Place rack of tubes in the boiling water bath for exactly two minutes. Make
sure the boiling water is higher than the contents of the tubes.
7. Remove from boiling water bath and immediately plunge rack of tubes into an
ice bath (plastic tub containing ice is okay) and leave for 15 minutes.
8. Remove rack of tubes from the ice bath and add 2.5mL cold HCl/K3Fe(CN)6
using Dispenser 3. Thorough mixing of the contents of the test tubes and the
K3Fe(CN)6 is vital for uniform development of colour, and can be achieved
with careful use of the dispenser. The red colour that develops with high ure-
ide concentrations should be uniform; layered colour indicates inadequate
mixing.
9. Leave on the bench for 10 minutes and read the O.D. at 525nm on a spectro-
photometer. The optical densities must be read as quickly as possible because
the colour will fade after a further 15 minutes.
The xylem sap samples together with internal allantoin standards and water
blanks are best analysed in a batch. A convenient batch size is 26 tubes, consist-
ing of:
3 water blanks 2.5 mL
3 internal standards (0.04 mM ureide) 2.5 mL
20 sap samples 0.1 mL + 2.4 mL distilled H2O

(1:25 dilution)
or
20 stem extracts 0.5 mL + 2.0 mL distilled H2O

(1:5 dilution)
10.9.2.2Amino-N (ninhydrin method) (Yemm and Cocking (1955) as

adapted by Herridge 1984)
Reagent preparation
a. Ninhydrin reagent
Add 4.79 g (Merck) ninhydrin plus 167 mg ascorbic acid dissolved in

16mL distilled H2O to 500mL methoxyethanol slowly, without aerating
210
the methoxyethanol. Store in a dark bottle in the fridge (pour into Dis-
penser 1). Can be stored for up to two weeks.
b. Citrate buffer
Add 67.2g citric acid plus 25.6g NaOH to 400mL distilled H2O (pour into
Dispenser 2).
c. Ethanol (60% v/v)
Add 300mL absolute ethanol to 200mL distilled H2O (pour into Dispens-
er 3).
d. 1 mM Asparagine (Asn)/1 mM glutamine (Gln) standard
Add 66mgAsn and 73mgGln to 500mL distilled H2O.
e. Distilled H2O (fill Dispenser 4)
The 2mM standard is now used to make the following concentrations for a stand-
ard curve determination:
Concentration 2.0 mM Water (mL)

(mM) amino-N (mL) Typical standard curve for amino-N
2.5 Range 00.20 nmol
0.00 0.0 100.0
0.10 5.0 95.0
2.0 y = 0.1014x + 0.0634
0.20 10.0 90.0 R2 = 0.9936
O.D. (570 nm)
0.40 20.0 80.0

1.5
1.00 50.0 50.0
1.0
0.5
0
0 4 8 12 16 20
nmol amino-N
Analysis
1. A standard curve covering the range 01.0mM should be constructed peri-

odically as part of the laboratorys quality control. Add 1.0mL of each of the
five concentrations to duplicate test tubes.
2. Using Dispenser 1 add 0.5mL citrate buffer in such a way that additional mix-
ing is unnecessary.
3. Using Dispenser 2 add 1.2mL ninhydrin reagent in such a way that additional
mixing is unnecessary.
4. Place in a boiling water bath for 15 minutes.
5. Remove from water bath and place on bench.
211
6. Using Dispenser 3, add 3.0 mL 60% ethanol in such a way that additional
mixing is unnecessary. The contents of the test tubes and the EtOH must be
thoroughly mixed for uniform colour development. The blue colour should
be uniform; uneven colour development in the test tubes indicates inadequate
mixing.
7. Read O.D. at 570nm on a spectrophotometer.
The xylem sap samples together with internal amino standards and water
blanks are best analysed in a batch. A convenient batch size is 26 tubes,
consisting of:
3 internal standards (0.20 mM) 1.0 mL
20 sap samples 0.1 mL + 0.9 mL H2O

(1:10 dilution)
10.9.2.3Salicylic acid method for nitrate determination (Cataldo et al.

1975)
This method is suitable for all legume xylem sap samples tested to date except pi-
geon pea (colour interference). Other recommended methods include Cu-hydra-
zine reduction (Kamphake et al. 1967) and an automated flow injection technique
(Alves et al. 2000).
Reagent preparation
a. Salicylic acid in concentrated sulfuric acid (H2SO4) (5% w/v).
Add 5g salicylic acid to 100mL concentrated H2SO4 (beaker). Leave for

24hours before use.
b. Sodium hydroxide (2M NaOH)
Add 40g NaOH to 500mL distilled H2O (Dispenser 1).
c. Nitrate standard (25 mM)
Add 632mgKNO3 to 250mL distilled H2O.
212
The 25 mM standard is now used to make the following concentrations for a

standard curve determination:
Concentration 25 mM KNO3 Water (mL)

Typical standard curve for nitrate
(mM) (mL)
1.6 Range 00.75 mol
0.00 0.0 100.0
1.25 5.0 95.0 1.4
y = 1.8649x + 0.0257
2.50 10.0 90.0 1.2 R2 = 0.9949
O.D. (410 nm)

5.00 20.0 80.0 1.0
10.0 40.0 60.0 0.8
15.0 60.0 40.0
0.6
0.4
0.2
0
0 0.2 0.4 0.6
mol nitrate
Analysis
1. A standard curve describing the O.D. response to increasing concentrations

of nitrate should be constructed within the range 015.0mM nitrate. Pipette
0.05mL of each of the six concentrations into duplicate test tubes.
2. Pipette 0.20 mL salicylic/sulfuric acid into the tubes. Mix and leave on the
bench for 20 minutes. Make sure that the solution in the tube is clear. Mix
further if cloudy.
3. Add 4.75mL 2 M NaOH using Dispenser 1. The dispenser should be used in

such a way that additional mixing is unnecessary.
4. Leave on the bench for 10 minutes and read O.D. at 410 nm on a

spectrophotometer.
A white precipitate will sometimes form with the addition of salicylic acid. It is
very important that the test tubes are shaken well to dissolve this precipitate. It
forms when the salicylic acid reagent is freshly made up and often when it is cold.
It is therefore best to make it up the day before analysis and make sure it is at room
temperature.
The xylem sap samples together with internal amino standards and water
blanks are best analysed in a batch. A convenient batch size is 26 tubes,
consisting of:
3 internal standards (2.5 mM) 0.05 mL
20 sap samples or stem extracts 0.05 mL
213
10.9.3Calculating N solute concentrations, %ureide-N and

%Ndfa
10.9.3.1Calculating concentrations of ureides, nitrate and amino-N
For the purposes of this exercise, the hypothetical sample for analysis is xylem sap
vacuum extracted from soybean harvested during early pod-fill, i.e. R3. The sam-
ple was analysed using the methods described above, with the following results.
Analysis Sample O.D. Dilution Internal standard Sample conc.*

Conc. (mM) O.D. (mM)
Ureide 0.50 25 0.04 0.57 0.88

Amino 0.85 10 0.20 0.46 3.69
Nitrate 0.14 1 2.50 0.24 1.46
* sample conc.=standard conc. (O.D. sample/O.D. standard) dilution

Ureide=0.04 (0.50/0.57) 25 = 0.88 mM
Amino=0.2 (0.85/0.46) 10 = 3.69 mM
Nitrate=2.5 (0.14/0.24) 1 = 1.46 mM
10.9.3.2 Calculating %ureide-N and %Ndfa
Now %ureide-N can be calculated as follows:
The standard curve relating %ureide-N to %Ndfa for sap vacuum extracted from
soybean during pod-fill is:
%ureide-N=(0.64 %Ndfa) + 15.9
Thus: %Ndfa=1.56 (%ureide-N 15.9)
=1.56 (41 15.9)
=39%
10.9.4 Sampling frequency and other factors to consider

If the experiment aims to compare the N2-fixing activity of treatments, one or two
samplings of xylem sap or stem segments will be sufficient. If, however, the objec-
tive is to quantify N2 fixation for the full growth cycle of the crop, multiple, i.e.
six to eight, samplings at regular intervals from mid vegetative to late reproduc-
tive stages are recommended. Seasonal profiles of %Ndfa are then calculated from
those of %ureide-N and used to partition accumulated crop N, estimated from
repeated biomass samplings, between fixed N and soil-derived N (see Herridge et
al. 1990, and a worked example in Unkovich et al. 2008).
214
The ureide method would have more appeal if the number of samplings could be
reduced to only one or two. The 15N methods integrate all N2 fixation activity from
the onset of growth until the time of sampling, so total N fixed can be estimated
by sampling only once at the time of maximum biomass N. A single sampling of
xylem sap or stem segments for N-solute analysis would not necessarily coincide
with the sampling for maximum biomass N and 15N, but would be done when the
point-of-time, ureide-determined %Ndfa had a similar value to the integrative,
15
N-determined %Ndfa.
Two studies in Australia (Herridge and Peoples 2002) on soybean, cowpea, mung
bean and black gram showed %Ndfa could be estimated with reasonable accuracy
from a single determination of %ureide-N during early pod-fill. These estimates
were very similar to the 15N-determined %Ndfa values, assessed at the time of
maximum biomass N (late pod-fill).
Other factors to consider when using the method are presented in Table10.10.
Table10.10 F actors to consider when using the ureide technique for quantifying legume N2 fixation
(source: Unkovich et al. 2008).
Variable Comments
Plant species Relationships between %ureide-N and %Ndfa are similar, but not identical, amongst
ureide-exporters. It is therefore recommended that each species under study is calibrated.
Cultivar/genotype Relationships between %ureide-N and %Ndfa appear to be unaffected by cultivar/
genotype within a species.
Strain of rhizobia Conflicting reports on effects of rhizobial strain on relationships between %ureide-N and
%Ndfa. No strain effect for pigeon pea and soybean, except in Brazil. Brazilian data also
suggest a strain effect with common bean.
Plant age Different calibrated relationships between %ureide-N and %Ndfa may need to be used
for vegetative and reproductive stages of development (Herridge and Peoples 1990;
Hansen et al. 1993).
N stress, including Relationships between %ureide-N and %Ndfa appear to be invalid when plants are
drought, and under severe N stress or are in senescence since ureides may also be synthesised from
senescence degradation products of nucleic acids. Indicated by xylem N-solute concentrations of
<12mM.
Ureides not High levels of ureides in vacuum-extracted xylem sap of Gliricidia sepium were not
associated with N2 associated with N2 fixation (Herridge et al. 1996). Need to firmly establish the link
fixation between %ureide-N and N2 fixation.
Sampling of Sampling best between 0900 and 1600hours, because of diurnal effects (see Herridge et
vacuum-extracted al. 1988; Peoples et al. 1989; Herridge et al. 1996);
sap (VES)
%ureide-N for VES unaffected by source or strength of vacuum.
A time delay of >5 minutes between sampling and extraction of xylem sap progressively
increases %ureide-N levels (Herridge et al. 1988; Peoples et al. 1989) creating an artefact.
For non-destructive sampling of single soybean plants for xylem sap, best results in terms
of volume of sap collected and separation of treatments are at late flowering, using the
shoot detached at internode 5 (Herridge et al. 1988).
Storage of xylem Xylem sap is stable at 25C for at least 14 days when diluted 1:1 in ethanol. Undiluted sap
sap stable at 25C for just one day. Freeze for long-term storage.
Errors in analysis of False readings when analysing xylem sap for ureides may occur with certain species (e.g.
xylem sap, such as Sesbania grandiflora; Herridge et al. 1996).
colour interference
215
10.10 Nitrogen balance method
In conclusion, the ureide method is relatively simple, accurate and versatile, and
can be applied in glasshouse and field experiments, or used in farmers fields, to
assess N2 fixation by ureide-exporting legumes. Many samples can be collected
and analysed in a single day; plants can be sampled non-destructively from the
top of the stem, or from lateral branches, with the base left intact to continue to
grow (Herridge et al. 1988). The resulting estimates of N2 fixation (%Ndfa and
total N2 fixed) for field-grown legumes are similar to those using more expensive
techniques.
10.10 Nitrogen balance method
Nitrogen balance compares the total N of a plantsoil system on two separate

occasions, with increases in N attributed to N2 fixation only after other possible
inputs and outputs of N have been quantified. The list of potential inputs and out-
puts is daunting, making this method one of last resort.
Inputs include: fertiliser N; manures and other sources of organic N; N in irriga-

tion water; wet deposition (mineral N in rainfall); dry deposition (mineral and
organic N in dust); runon; lateral subsoil flow and N2 fixation (which includes N2
fixation by associative and free-living bacteria as well as from the legume-rhizobia
symbiosis).
Outputs include: the N in harvested crop and animal products; gaseous N emis-
sions (NH3 via volatilisation, N2O via nitrification and NOx and N2 via denitrifica-
tion); soil erosion, runoff and leaching.
N2fixation will be underestimated or overestimated if N outputs or inputs are not

accurately measured. An accurate N balance therefore requires measurements of
as many potential N inputs and outputs as possible. The time frame is generally
several years because of the need to quantify incremental changes in the N con-
tent of the soil against a large background (Giller and Merckx 2003).
The methodology is technically challenging, requiring substantial commitments

of labour for long periods. Errors in quantifying the N fluxes, and inaccuracies in
sampling and analysing soil for changes in total N and bulk density, can introduce
substantial errors into the final estimates of N2 fixation (Chalk 1998).
Nitrogen balance was more commonly used in the past (e.g. Vallis 1973; Firth
et al. 1973; Wetselaar et al. 1973) but has now been essentially replaced by other
methods.
216
10.11 Acetylene reduction method (assaying nitrogenase activity)
10.11Acetylene reduction method (assaying

nitrogenase activity)
The enzyme nitrogenase, universally responsible for biological N2 fixation, can

also reduce acetylene (C2H2) to ethylene (C2H4), both of which are readily quanti-
fied using gas chromatography (Hardy et al. 1968). Although the C2H2 reduction
assay is a sensitive measure of nitrogenase activity at a point in time, and can be
very useful for detecting N2 fixation activity of, for example, bacterial cultures, it
is less useful for quantifying N2 fixation by legumes.
Acetylene reduction remains a valuable tool, such as when assessing the impacts
on N2 fixation of mutational studies with rhizobia or assessing N2 fixation in free
living cultures (McComb et al. 1975). Methods to undertake an assay are given in
Chapter5.
10.12 Measuring N2 fixation at the farm and beyond
If estimates of N2 fixation at farm, farming system, regional or national scale are

required, the area of legumes at the particular scale needs to be quantified. This
can be done by a walking field survey. Areas can be measured using a hand-held
GPS (geographic positioning system) if the fields are large enough (>0.5ha) in re-
lation to the precision of the GPS (commonly 5m). Another useful tool is to use
GoogleEarth to create a geo-referenced image of the study area in a geographical
information system. This can be used to prepare a field map in advance of a field
survey to allow accurate measurement of field areas. Remote sensing can be used
to map areas of different crops, at small scales (less than 1km2) using sensors lift-
ed using octocopters, kites or balloons, and at larger scales using unmanned aerial
vehicles (UAVs or drones). If remote sensing methods are used at large scales, de-
tailed ground-truthing is needed for accurate results.
Once legume fields are identified, they can be sampled as described above. The
number of fields that need to be sampled depends on the measured differences in
N yield and N2 fixation, and statistical advice is probably needed to develop an op-
timal sampling strategy. Ideally, surveys need to be conducted several times dur-
ing the year, or over several years, to gain insights into the frequency with which
legumes are grown on each field. If this is not always possible, farmers are often
willing to share their records or questionnaires can be used.
Studies at regional or global level usually depend on statistical records. The ac-
curacy of estimates of N2 fixation made with production records depends on the
reliability of the available data used for scaling up. Herridge et al. (2008) estimated
N2 fixation inputs at global scale and discuss the accuracy and pitfalls of such es-
timates.
217
10.13Conclusions
10.13Conclusions
No method for assessing N2 fixation is perfect. Therefore other lines of investiga-

tion should be included that can provide insights into treatment effects, such as
assessment of nodulation and soil mineral N levels. Although all methods have
specific limitations and sources of error (some more than others), the N differ-
ence, 15N (isotope dilution and natural abundance) and ureide methods are argu-
ably the best available. The choice of method needs to be made after consideration
of the question being addressed. Confidence in the answer can be enhanced if two
independent methods are used.
All of the methods depend on quantifying legume biomass N to calculate the

amount of N2 fixed, so none are truly independent. Although much attention
is paid to accurately estimating %Ndfa, an accurate measurement of the total
amount of N accumulated in the legume is of equal importance and deserves care
and attention to achieve the best results.
Finally, legumes are often grown in farming systems to benefit other cash crops,
such as cereals or canola. A discussion of how to estimate the N benefit to rota-
tional crops, and the concept of free N farming is provided in Chapter8.
10.14References
Alves B.F.R., Zotareli L., Resende A.S., Polidoro J.C., Urquiaga S. and Boddey R.M. 2000. Rapid and
sensitive determination of nitrate in plant tissue using flow injection analysis. Communications
in Soil Science and Plant Analysis 31, 27392750.
Adjei-Nsiah S., Kuyper T.W., Leeuwis C., Abekoe M., Cobbinah J., Sakyi-Dawson O. and Giller K.E.
2008. Farmers agronomic and social evaluation of productivity, yield and N2-fixation in differ-
ent cowpea varieties and their subsequent residual effects on a succeeding maize crop. Nutrient
Cycling in Agroecosystems 80, 199209.
Boddey R.M., de Oliveira O., Alves B. and Urquiaga S. 1995. Field application of the 15N isotope di-
lution technique for the reliable quantification of plant-associated biological nitrogen fixation.
Fertilizer Research 42, 7787.
Burris R.H., Eppling F.J., Wahlin H.B. and Wilson P.W. 1942. Studies of biological nitrogen fixation
with isotopic nitrogen. Soil Science Society America Proceedings 7, 258262.
Cataldo D.A., Haroon M., Schrader L.E. and Youngs V.L. 1975. Rapid colorimetric determination of
nitrate in plant tissue by nitration of salicylic acid. Communications in Soil Science and Plant
Analysis 6, 7180.
Chalk P.M. 1985. Review: estimation of N2-fixation by isotope dilution: an appraisal of techniques
involving 15N enrichment and their application. Soil Biology and Biochemistry 17, 389410.
Chalk P.M. 1996. Estimation of N2 fixation by 15N isotope dilution: the A-value approach. Soil Biol-
ogy and Biochemistry 28, 11231130.
Chalk P.M. 1998. Dynamics of biologically fixed N in legumecereal rotations: a review. Australian
Chalk P.M. and Ladha J.K. 1999. Estimation of legume symbiotic dependence: an evaluation based
on 15N dilution. Soil Biology and Biochemistry 31, 19011917.
Danso S.K.A., Hardarson G. and Zapata F. 1993. Misconceptions and practical problems in the use
of 15N soil enrichment techniques for estimating N2 fixation. Plant and Soil 152, 2552.
Evans J., Turner G.L., OConner G.E. and Bergersen F.J. 1987. Nitrogen fixation and accretion of
soil nitrogen by field-grown lupins (Lupinus angustifolius). Field Crops Research 16, 309322.
Firth P., Thitipoca H., Suthipradit S., Wetselaar R. and Beech D.F. 1973. Nitrogen balance studies in
the central plain of Thailand. Soil Biology and Biochemistry 5, 4146.
218
10.14References
Giller K.E. 2001. Nitrogen fixation in tropical cropping systems: second edition. CAB International:
Wallingford.
Giller K.E. and Merckx R. 2003. Exploring the boundaries of N2-fixation in non-legumes: An hypo-
thetical and experimental framework. Symbiosis 35, 317.
Giller K.E. and Witty J.F. 1987. Immobilized 15N-fertilizer sources improve the accuracy of field esti-
mates of N2-fixation by isotope-dilution. Soil Biology and Biochemistry 19, 459463.
Hansen A.P., Rerkasem B., Lordkaew S. and Martin P. 1993. Xylem-solute technique to measure N2
fixation by Phaseolus vulgaris L.: calibration and sources of error. Plant and Soil 150, 223231.
Hardy R.W.F., Burns R.C. and Holsten R.D. 1973. Application of the acetyleneethylene assay for
measurement of nitrogen fixation. Soil Biology and Biochemistry 5, 4781.
Hardy R.W.F., Holsten R.D., Jackson E.K. and Burns R.C. 1968. The acetyleneethylene assay for N2
fixation: laboratory and field evaluation. Plant Physiology 43, 11851207.
Herridge D.F. 1984. Effects of nitrate and plant development on the abundance of nitrogenous sol-
utes in root-bleeding and vacuum extracted exudates of soybean. Crop Science 25, 173179.
Herridge D.F., Bergersen F.J. and Peoples M.B. 1990. Measurement of nitrogen fixation by soybean
in the field using the ureide and natural 15N abundance methods. Plant Physiology 93, 708716.
Herridge D.F., OConnell P. and Donnelly K. 1988. The xylem ureide assay of nitrogen fixation: sam-
pling procedures and sources of error. Journal of Experimental Botany 39, 1222.
Herridge D.F., Palmer B., Nurhayati D.P. and Peoples M.B. 1996. Evaluation of the xylem ureide
method for measuring N2 fixation in six tree legume species. Soil Biology and Biochemistry 28,
281289.
Herridge D.F. and Peoples M.B. 1990. Ureide assay for measuring nitrogen fixation by nodulated
soybean calibrated by 15N methods. Plant Physiology 93, 495503.
Herridge D.F. and Peoples M.B. 2002. Timing of xylem sampling for ureide analysis of nitrogen fixa-
tion. Plant and Soil 238, 5767.
Herridge D., Peoples M. and Boddey R. 2008. Global inputs of biological nitrogen fixation in agri-
cultural systems. Plant and Soil 311, 118.
Hgberg P. 1997. Tansley Review No. 95 15N natural abundance in soil-plant systems. New Phytolo-
gist 137, 179203.
Hunt P., Burnham K. and Matheny T. 1987. Precision and bias of various soybean dry matter sam-
pling techniques. Agronomy Journal 79, 425428.
Kamphake L.J., Hannah S.A. and Cohen J.M. 1967. Automated analysis for nitrate by hydrazine re-
duction. Water Research 1, 205216.
Khan D.F., Peoples M.B., Chalk P.M. and Herridge D.F. 2002. Quantifying below ground nitrogen of
legumes: a comparison of 15N and non-isotopic methods. Plant and Soil 239, 277289.
Kipe-Nolt J.A. and Giller K.E. 1993. A field evaluation using the 15N isotope dilution method of lines
of Phaseolus vulgaris L. bred for increased nitrogen fixation. Plant and Soil 152, 107114.
Letolle R. 1980. Nitrogen-15 in the natural environment. Pp. 407433 in Handbook of environmen-
tal isotope geochemistry, ed. by P. Fritz and J.C. Fontes. Elsevier: Amsterdam.
Mapfumo P., Giller K.E., Mpepereki S. and Mafongoya P.L. 1999. Dinitrogen fixation by pigeonpea
of different maturity types on granitic sandy soils in Zimbabwe. Symbiosis 27, 305318.
Mariotti A. 1983. Atmospheric nitrogen is a reliable standard for 15N natural abundance measure-
ments. Nature 303, 685687.
McAuliffe C., Chamblee D.S., Uribe-Arango H. and Woodhouse W.W. 1958. Influence of inorganic
nitrogen on nitrogen fixation by legumes as revealed by N15. Agronomy Journal 50, 334337.
McComb J.A., Elliot J. and Dilworth M.J. 1975. Acetylene reduction by Rhizobium in pure culture.
Nature (London) 256, 409410.
McNeill A.M., Pilbeam C.J., Harris H.C. and Swift R.S. 1998. Use of residual fertiliser 15N in soil for
isotope dilution estimates of N2 fixation by grain legumes. Australian Journal of Agricultural
Research 49, 821828.
McNeill A.M., Zhu C. and Fillery I.R.F. 1997. Use of in situ 15N-labelling to estimate the total below-
ground nitrogen of pasture legumes in intact soil-plant systems. Australian Journal of Agricul-
tural Research 48, 295304.
Minchin F.R., Witty J. and Sheehy J.E. 1983. A major error in the acetylene reduction assay: de-
creases in nodular nitrogenase activity under assay conditions. Journal of Experimental Botany
34, 641649.
Minchin F.R., Sheehy J.E. and Witty J.F. 1986. Further errors in the acetylene reduction assay: effects
of plant disturbance. Journal of Experimental Botany 37, 15811591.
Norhayati M., Noor S.M., Chong K., Faizah A.W., Herridge D.F., Peoples M.B. and Bergersen F.J.
1988. Adaptation of methods for evaluating N2 fixation in food legumes and legume cover crops.
Plant and Soil 108, 143150.
219
10.14References
Peoples M.B., Boddey R.M. and Herridge D.F. 2002. Quantification of nitrogen fixation. Pp. 357389
in Nitrogen fixation at the millenium, ed. by G.J. Leigh. Elsevier Science: Amsterdam.
Peoples M.B., Unkovich M.J. and Herridge D.F. 2009. Measuring symbiotic nitrogen fixation by
legumes. Pp. 125170 in Nitrogen Fixation in Crop Production, ed. by D.W. Emerich and H.B.
Krishnan. Agronomy Monograph 52, American Society of Agronomy, Crop Science Society of
America and Soil Science Society of America.
Peoples M.B., Faizah A.W., Rerkasem B. and Herridge D.F. 1989. Methods for Evaluating Nitrogen
Fixation by Nodulated Legumes in the Field. ACIAR: Canberra.
Rochester I.J., Peoples M.B., Constable G.A. and Gault R.R. 1998. Faba beans and other legumes add
nitrogen to irrigated cotton cropping systems. Australian Journal of Experimental Agriculture
38, 253260.
Russell C.A. and Fillery I.R.P. 1996. In situ labelling of lupin below-ground biomass N. Australian
Sanchez C., Blackmer A., Horton R. and Timmons D. 1987. Assessment of errors associated with
plot size and lateral movement of nitrogen-15 when studying fertilizer recovery under field con-
ditions. Soil Science 144, 344351.
Shearer G. and Kohl D.H. 1986. N2-Fixation in field settings: estimations based on natural 15N abun-
dance. Australian Journal of Plant Physiology 13, 699756.
Swan A., Peoples M., Gault R., Crews T., Khan D.F., Urquiaga S., Alves B., Boddey R.M. and For-
rester R. 2003. Evaluating sources of error when quantifying N2 fixation in commercial leg-
ume crops. Pp 14 in 11th Australian Agronomy Conference. Geelong: Australian Society of
Agronomy.
Unkovich M., Herridge D., Peoples M., Cadisch G., Boddey B., Giller K., Alves B. and Chalk P. 2008.
Measuring plant-associated N2 fixation in agricultural systems. ACIAR: Canberra.
Unkovich M.J. and Pate J.S. 2000. An appraisal of recent field measurements of symbiotic N2-fixa-
tion by annual legumes. Field Crops Research 65, 211228.
Unkovich M.J., Pate J.S. and Sanford P. 1993. Preparation of plant samples for high precision nitro-
gen isotope ratio analysis. Communications in Soil Science and Plant Analysis 24, 20932106.
Vallis I. 1973. Sampling for soil nitrogen changes in large areas of grazed pasture. Communications
in Soil Science and Plant Analysis 4, 163170.
Vessey K.J. 1994. Measurement of nitrogenase activity in legume root nodules: in defense of the
acetylene reduction assay. Plant and Soil 158, 151162.
Viera-Vargas M.S., Deoliveira O.C., Souto C.M., Cadisch G., Urquiaga S. and Boddey R.M. 1995a.
Use of different 15N labeling techniques to quantify the contribution of biological N2 fixation to
legumes. Soil Biology & Biochemistry 27, 11851192.
Viera-Vargas M.S., Souto C.M., Urquiaga S. and Boddey R.M. 1995b. Quantification of the contri-
bution of N2 fixation to tropical forage legumes and transfer to associated grass. Soil Biology and
Biochemistry 27, 11931200.
Warembourg F.R., Montange D. and Bardin R. 1982. The simultaneous use of 14CO2 and 15N2 label-
ling techniques to study the carbon and nitrogen economy of legumes grown under natural
conditions. Physiologia Plantarum 56, 4655.
Watanabe I., Chiu C. and Yoshida T. 1990. Estimation of N2 fixation in soybean and cowpea by using
soil residual 15N. Soil Science and Plant Nutrition 36, 375381.
Wetselaar R., Jakobsen P. and Chaplin G. 1973. Nitrogen balance in crop systems in tropical Aus-
tralia. Soil Biology and Biochemistry 5, 3540.
Williams P.C., Norris K.H., Gerkhe C.W. and Bernstein K. 1983. Comparison of near infra-red
methods for measuring protein and moisture in wheat. Cereal Foods World 150, 149152.
Witty J.F. 1983. Estimating N2-fixation in the field using 15N-labelled fertilizer: some problems and
solutions. Soil Biology and Biochemistry 15, 631639.
Witty J.F. and Minchin F.R. 1988. Measurement of nitrogen fixation by the acetylene reduction as-
say; myths and mysteries. Pp. 331344 in Nitrogen Fixation by Legumes in Mediterranean Ag-
riculture, ed. by D.P. Beck and L.A. Materon. Martinus Nijhoff: Dordrecht.
Yemm E.W. and Cocking E.C. 1955. The determination of amino acids with ninhydrin. Analyst 80,
209213.
Young E.G. and Conway C.F. 1942. On the estimation of allantoin by the Rimini-Schryver reaction.
Journal of Biological Chemistry 142, 839853.
220
11.1 Introduction to nucleic acid purification
CHAPTER11
Fundamental molecular
techniques for rhizobia
W.G. Reeve, R.P. Tiwari, V. Melino and P.S. Poole
Working with DNA is now a fundamental skill in working with rhizobia. It is

necessary for typing strains using PCR methods and for sequencing activities ap-
plied to understanding genomes; their structure, how they function, and their
taxonomic position.
Nucleic acid purification is the separation of nucleic acids from proteins, cell wall
debris and polysaccharide after lysis of cells. For rhizobia, we provide here a num-
ber of commonly used methods for the extraction of genomic and plasmid DNA.
Methods for extraction of total RNA are presented in Chapter 13. The CTAB
method (Protocol 11.1.1) has been used extensively for extraction of total genom-
ic DNA for DNA sequencing while Protocol 11.1.2 gives higher yields but gener-
ally with slightly lower purity. Plasmid DNA can be differentially displayed using
Protocol 11.2.1 for determination of replicon number. This method allows locali-
sation of genes to replicons, confirmation of genome assemblies and identification
of genetic changes. The plasmids can subsequently be purified from low melting
point gels using GELase (Epicentre, http://www.epibio.com/item.asp?id=297).
Protocol 11.2.2 presents a method to recover introduced plasmids from rhizobia
(i.e. complementing plasmids) for transformation into Escherichia coli prior to
restriction analysis. Protocol 11.2.3 provides an alternative method to the GELase
procedure for purifying plasmids but has not been tested as extensively.
221
11.1.1Protocol for bacterial genomic DNA isolation using

CTAB
This CTAB protocol comes from the DOE Joint Genome Institute Standard Op-
erating Procedures (http://my.jgi.doe.gov/general/protocols.html) and has been
successfully used by the authors of this chapter for the isolation of genomic DNA
from 130 rhizobial strains to date.
11.1.1.1Materials and reagents
Materials and reagents are listed in Table11.1
Table11.1 Materials and reagents for isolation of genomic DNA

Disposables Vendor Stock number
1.5mL microcentrifuge tube Eppendorf 22 36 320-4
50mL Nalgene Oak Ridge polypropylene centrifuge tube VWR 21010-568
10mL pipette Falcon 357551
1mL pipette tips MBP 3781
Reagents: Vendor Stock number

CTAB (cetyltrimethylammonium bromide) Sigma H-6269
NaCl Sigma S-3014
TE buffer (10mM Tris; 1mM EDTA, pH8.0) Ambion 9858
Lysozyme Sigma L-6876
Proteinase K Qiagen 19131
5MNaCl Ambion 9759
10% (w/v) SDS (Sodium dodecyl sulfate) Sigma L-4522
Chloroform Sigma C-2432
iso-Amyl alcohol Sigma I-9392
Phenol Sigma P-4557
Isopropanol VWR PX-1835-14
Ethanol AAPER
RNAse A (100mgmL1) Qiagen 19101
Equipment
Hot plate
250mL glass beaker
Magnetic stirring bar
Thermometer
Automatic pipette dispenser
Sorvall 500 Plus centrifuge (DuPont, Newtown, CT)
65C water bath
37C incubator
222
11.1.1.2 Reagent/stock preparation
CTAB/NaCl (cetyltrimethylammonium bromide)
Dissolve 4.1g NaCl in 80mL of water and slowly add 10g CTAB while heating
(~65C) and stirring. This takes more than three hours to dissolve CTAB. Adjust
final volume to 100mL and sterilise by filtering or autoclaving.
Procedure
Cell preparation and extraction techniques
Notes
In step1 below, do not use too many bacterial cells (OD600nm of not more than
1 is recommended) or DNA will not separate well from the protein.
Most of the time, inverting several times is sufficient to mix well. Avoid vigor-
ous shaking which will shear the DNA.
Use any standard protocol for DNA precipitation (Ausubel et al. 2013).
Total volume....................................................................................................................1.5mL 30mL
1. Grow cells in broth and pellet for five minutes or scrape from plate.
2. Transfer bacterial suspension to the appropriate centrifuge tube.
3. Pellet cells for five minutes.
4. Discard the supernatant.
5. Re-suspend cells in TE buffer (10mM Tris; 1mM EDTA; pH8.0) to an

OD600nm~1.0.
6. Transfer given amount of cell suspension to a clean centrifuge tube................. 740 L 14.8mL
7. Add lysozyme (conc. 100mgmL1). Mix well. ....................................................... 20 L 400 L
This step is necessary for hard-to-lyse bacteria.
8. Incubate for five minutes at room temperature.
9. Add 10% (w/v) SDS. Mix well.................................................................................... 40 L 800 L
10. Add Proteinase K (10mgmL1). Mix well.................................................................. 8 L 160 L
11. Incubate for one hour at 37C
12. Add 5MNaCl. Mix well........................................................................................... 100 L 2 mL
13. Add CTAB/NaCl (heated to 65C). Mix well......................................................... 100 L 2 mL
14. Incubate at 65C for 10 minutes.
15. Add chloroform:isoamyl alcohol (24:1). Mix well.................................................0.5 mL 10 mL
223
16. Spin at maximum speed for 10 minutes at room temperature.
17. Transfer aqueous phase to clean microfuge tube (should not be viscous).
18. Add phenol:chloroform:isoamyl alcohol (25:24:1).

Mix well.......................................................................................................................0.5 mL 10 mL
19. Spin at maximum speed for 10 minutes at room temperature.
20. Transfer aqueous phase and add 0.6 volumes (vol.) of isopropanol (20C).
E.g. if 400L of aqueous phase is transferred, add 240L of isopropanol

21. Incubate at room temp for 30 minutes.
22. Spin at maximum speed for 15 minutes.
23. Wash pellet with 70% (v/v) ethanol, spin at maximum speed for five
minutes.
24. Discard the supernatant and let pellet dry for five to 10 minutes at room
temperature.
25. Re-suspend in TE buffer plus RNAse (99LTE + 1l RNAse (10mgmL1)..... 20 L 400 L
26. Transfer to sterile micro-centrifuge tubes.
28. Run 1L in a 1% (w/v) agarose gel with concentration standards (Figure11.1).
1 2 3 4
Figure11.1 Gel electrophoresis of genomic

DNA extracted from rhizobia using Protocol 11.1.1
described above from: Mesorhizobium sp. WSM3224
(Lane 1); Rhizobium leguminosarum bv trifolii
SRDI565LMG23256 (Lane 2); R. leguminosarum bv
trifolii SRDI943 (Lane 3); and Lambda HindIII cut
DNA marker (Lane 4).
224
11.1.2Protocol for genomic DNA preparation using phenol-

chloroform
11.1.2.1Precautionary measures
All pipette tips should be filter barrier tips.

Benches should be scrupulously clean (wipe with 70% (v/v) ethanol before
use) and gloves must be worn at all times to prevent contamination.
Containers should be disposable plastic ware, and glassware needs to be

soaked in 0.1NHCl and then 0.1NNaOH prior to rinsing with RODI (re-
verse-osmosis, deionised) water.
Genomic DNA should be prepared using pipette tips with the tip end cut off
to prevent shearing of the DNA.
General preparation
TY broth
Tryptone 5g
Yeast-extract 3g
CaCl2.2H2O 0.75 g (reduced from original recipe to avoid precipitation)
Distilled H2O to 1 L
Adjust pH to neutral
Procedure
1. Inoculate a 10mL TY broth in a 50mL conical flask with a loopful of cells

and grow to late log phase (for root nodule bacteria, it usually requires two to
three days at 28C). Subculture the cells if necessary to ensure the cells are in
the log phase.
2. Transfer the log phase culture into a centrifuge tube.
3. Pellet cells at 7,500g for five minutes at room temperature.
4. Remove the supernatant (last drops can be removed using a pipette).
5. Re-suspend pellet in 1 mL TES buffer (30 mM Tris pH 8.0, 50 mM NaCl,

5mM EDTA) by vortexing for one to two minutes.
6. Repeat steps 3 and 4.
7. Re-suspend cell pellet in 1mL TES buffer.
8. Add 0.2mL lysozyme solution (10mgmL1 in TES buffer).
225
10. Add 0.15mL 10% (w/v) SDS and 60l of proteinase K (6mgmL1 in TES).
11. Incubate at 45C for one hour.
12. Add another aliquot of 60l proteinase K and incubate at 55C for one hour.
13. Extract DNA twice using an equal volume of phenol:chloroform:isoamyl al-

cohol (25:24:1).
14. Add 0.1 volume of 3M sodium acetate (pH5.2) to the extracted DNA.
15. Add an equal volume of isopropanol and mix gently.
16. Centrifuge at room temperature at 7,500g for 10 minutes.
17. Remove supernatant carefully without losing the pellet.
18. Add 0.5mL of 70% (v/v) ethanol, close the lid and invert the tube carefully a
few times.
19. Centrifuge at 7,500g at room temperature for five minutes and remove super-
natant carefully without losing the pellet.
20. Invert tubes on a sterile paper towel and drain the last drops of liquid.
21. Leave the tubes to air dry for 15 to 20 minutes.
There is no need to vacuum dry DNA; it makes resuspension of the DNA difficult.
22. Re-suspend pellet in sterile RODI water.
23. Treat with RNAse (final concentration 20g mL1) for one hour at 37C to
remove RNA.
11.1.3Protocol for nucleic acid extraction from soil

This method is modified from Griffiths et al. (2000).
General preparation
5% CTAB/phosphate buffer (120mM, pH8)

mix equal volumes of 10% (w/v) CTAB in 0.7MNaCl and 240mM potas-
sium phosphate buffer (pH8.0) and autoclave
phenol:chloroform:isoamyl alcohol (pH 8.0) (25:24:1)
Dangernever autoclave
chloroform:isoamyl alcohol (24:1)
Dangernever autoclave
226
30% (w/v) PEG (Polyethylene glycol) purchased from Fluka, SIGMA)
Dont autoclave PEG
70% (v/v) ethanol. Keep in the freezer.
No need to autoclave
Molecular grade, sterile water

100mL of240mM potassium phosphate buffer (pH8)
1 M K2HPO4 : 22.56 mL
1 M KH2PO4 : 1.44 mL
Dilute to 100 mL
Elution Buffer (EB)
10 mM Tris-Cl, pH 8.5
Cell homogenisation/lysis equipment (i.e. Fast-Prep beadbeater)
Bead tubes
0.1mm zirconia/silica beads, 0.5mm glass beads (fill this bead mixture up to
20% of the tubes volume and add one of a 2mm glass ball).
Procedure
1. Weigh out 1g of soil.
2. Add 0.5 mL CTAB buffer and 0.5 mL phenol:chloroform:isoamyl alcohol

(25:24:1) into ready-made 2mL bead tubes, tighten lids and keep on ice.
3. Place tubes in Fast-Prep beadbeater and lyse cells for 30 seconds at speed 5.5
(repeat three times with a five minute break in between lysing treatments).
4. Centrifuge at full speed (16,000g) for five minutes at 4C.
5. Extract the top aqueous layer and transfer to a new microfuge tube.
6. Add 0.5mL of chloroform:isoamyl alcohol (24:1) and vortex for five seconds
to form an emulsion.
7. Centrifuge at full speed (16, 000g) for five minutes (4C).
8. Extract the top layer again and precipitate nucleic acids by adding 1 mL of
PEG solution. Mix well.
9. Leave samples for one to two hours at room temperature.
10. Centrifuge at 16,000g for 10 minutes (room temperature).
227
11.2 Plasmid purification
11. Pour off supernatant and wash pellet with 70% (v/v) ethanol (200L). Vortex,
centrifuge for 10 minutes (16, 000g) and discard alcohol.
12. Repeat step11, removing all traces of liquid.
13. Re-suspend pellet in 50L of EB buffer.
Removal of humic acids from DNA samples
One-stepTM PCR inhibitor removal kit (ZYMO RESEARCH cat. No D6030). The
protocol is provided here.
1. Remove the cap and snap off the base of the supplied filter tubes. Place them
into collection tube (provided).
2. Spin down at 8,000g for exactly three minutes.
3. Place the filter into 1.5mL microfuge tube and add 100 to 200L of DNA
sample.
4. Spin down at 8,000g for exactly one minute. The eluate should be free of hu-
mic acid.
11.2.1Protocol for plasmid profiling using a modified

Eckhardt procedure
The original method (Eckhardt 1978) has been subsequently modified (Priefer,
1984; Hynes et al., 1985; and Hynes and McGregor, 1990).
Recipes (stocks and media)
The components of TBE buffer and HP media are given in Tables 11.2 and 11.3.
Table11.2 10 TBE buffer

Component Quantity
Tris 108 g L1
Boric acid 55 g L1
EDTA (0.5 M, pH 8.0) 40 mL L1
Make up to 1L with RODI water, pH to 8.0 with HCl, autoclave.
1 TBE buffer
Dilute from 10 TBE buffer using RODI water. Also maintain a stock of sterile
TBE buffer for preparing the agarose gels, Sarcosyl and lysis solution.
228
10% SDS in 1 TBE buffer (200mL)
Dissolve 20g of SDS in 160mL of pre-warmed 1TBE buffer.

Adjust to pH8.0.
Make up to 200mL with 1TBE buffer.
Autoclave.
NB: if the SDS precipitates, re-dissolve it at 50C in a water bath.
0.3% (w/v) Sarcosyl in 1 TBE buffer
Dissolve 0.15g Sarcosyl (n-lauroylsarcosine) in 50mL of 1 sterile TBE buffer.

Chill at 4C for at least one hour prior to use.
E1 Solution10% (w/v) sucrose in 1 TBE buffer + 10 g mL1 RNase
Dissolve 1g sucrose in 10mL 1 sterile TBE buffer.

Store at 4C.
Lysozyme Stock Solution 100
100mgmL1 lysozyme dissolved in 1 TBE

Store at 20C.
Lysis Solution (100 g mL1 lysozyme in E1 Solution)
Prepare fresh on the day.

Mix 986L of E1 solution with 4L RNAase stock solution (10mgmL1).
Add 10L of 100 lysozyme stock solution .
Table11.3 HP media

Component Quantity
Peptone 4.0 g L1
Yeast extract 0.5 g L1
Tryptone 0.5 g L1
CaCl2.2H2O 0.2 g L1
MgSO4.7H2O 0.2 g L1
Procedure
Step 1
Inoculate strains into prepared 5mL broths (add antibiotics if selection pressure
is required) and grow at 28C with shaking at 200rpm.
229
NB: selection of the broth media depends on which genera of root nodule bacteria
(RNB) are used. Rhizobium leguminosarum lyses well when grown in TY while
Sinorhizobium medicae lyses better when grown in TY with half the normal CaCl2
or in HP media (when using HP media, omit the antibiotics). Agrobacterium sp.
may be grown in LB. Grow overnight (28C, 200rpm) to an OD600nm of 0.3. (OD
is important; try not to grow over 0.3). It is best to do a range of subcultures to
ensure the correct OD; i.e. subculture 2, 5, 10 and 20L into 5mL broths. Faster
growing strains can be further subcultured.
Step 2
Prepare and pour the 0.75% (w/v) agarose plus 1% (w/v) SDS gel:
1. Add 7.5g agarose to 900mL 1TBE (pH8.0) buffer and microwave (stirring
occasionally) until clear.
2. Cool the agarose to 55C, pour 90mL into a conical flask and add10 mL of
10% (w/v) SDS solution. Mix by gentle swirling and pour into the gel tray.
This quantity is sufficient for a 15cm 15cm tray; the gel should be as thin as pos-
sible (3 to 4mm) to give the best results.
3. When the gel is set, place in the gel tank and cover with 1TBE buffer.
Preparation of the cultures
1. Set the centrifuge to 4C (or place centrifuge in a refrigerated room).
2. Pipette 200L of culture (OD600nm 0.3this is the best cell density but it can
be up to 0.5) into a 1.5mL micro-centrifuge tube and place on ice.
You can vary the volume of culture used in this step according to the OD600nm i.e.
300L at OD0.2. Additionally, a series of aliquots i.e. 200, 400, and 800L may be
used. Exopolysaccharide may be removed by pelleting cells and re-suspending them
in sterile RODI water.
3. Add 1.0 mL of cold 0.3% (w/v) Sarcosyl and mix by inversion twice.
4. Place on ice for 10 minutes.
5. Prepare the lysis solution and place on ice.
6. Centrifuge the Sarcosyl-culture mix (4C, 16,000rpm; five minutes).
7. Carefully pour off the supernatant and place the tube back on ice. Using a
P200 pipette, carefully remove the remaining supernatant, leaving the pellet
as dry as possible.
8. Place the tubes in a rack at room temperature, add 20L of lysis solution and
re-suspend the pellet by carefully pipetting up and down twice. Avoid creating
bubbles in the solution.
230
9. Immediately load 20 L of sample into the gel well. The sample should be
evenly re-suspended and slightly cloudy. The loading dye can be added to one
well as a marker.
10. When loading of all samples is complete, initiate electrophoresis at 5 to 10V

for 15 to 30 minutes (maximum) until the samples have lysed and are clear.
Then run the gel electrophoresis at 70 to 80V overnight. LEAVE THE GEL
TANK LID OFF to avoid over-heating the gel tank. DANGER: Indicate with
appropriate signage that the lid off means exposed live wires. Alternatively,
run in the cold room or in an air-conditioned laboratory leaving the gel tank
lid on.
11. After staining with ethidium bromide (for 30 minutes to one hour), de-stain
in sterile RODI water for 30 minutes (longer de-stain times may improve the
resolution of the bands; see Figure11.2).
Figure11.2 Plasmid profile of five Rhizobium

leguminosarum strains compared with control R. l. bv.
viciae VF39 (lane 1) showing bands of sizes 900, 700, 500,
400, 220 and 150 kb (top to bottom).
11.2.2Protocol for plasmid preparation by alkaline lysis

This alkaline lysis method is recommended for rapid extraction and purification
of small (<50kb) plasmids that have been introduced into rhizobia (i.e. via con-
jugation; see Protocol 11.2.1). The following protocol (Ausubel et al. 2013) was
adapted specifically for the isolation of rhizobial megaplasmids. It is important to
mention that although plasmids may be isolated from RNB using this method, the
profile after separation by gel electrophoresis will appear smeared and the authors
have noted inhibition of restriction enzyme digest reactions. Plasmids prepared
from RNB using this method will need to be transformed back into E. coli prior
to extraction (using the original version of this protocol) and restriction enzyme
digestion.
231
General preparation
Prepare Solution I from standard stocks in batches of approximately 100mL;

sterilise by autoclaving.
Prepare Solution II fresh using sterile distilled H20 and use at room tempera-
ture. Warm solution II if a precipitate is observed.
Store Solution III at 4C and use at 4C.

100% (v/v) ethanol at room temperature.
70% (v/v) ethanol at room temperature.
Molecular grade H2O.
Recipes
Alkaline Lysis Solution I
25 mM Tris-Cl (pH 8.0)

10 mM EDTA (pH 8.0)
Alkaline Lysis Solution II
0.2 N NaOH (freshly diluted from a 10N stock)

1% (w/v) SDS (diluted on the day of use from a 10% (v/v) SDS stock)
Alkaline Lysis Solution III
The resulting solution is 3 M with respect to potassium and 5 M with respect to ac-
etate. Prepare100 mL total volume.
5 M potassium acetate, pH 4.8 60.0 mL
glacial acetic acid 11.5 mL
molecular grade H2O 28.5 mL
Procedure
1. Inoculate 2mL of TY containing the appropriate antibiotic with a single col-

ony of bacteria. Incubate the culture overnight at 28C with vigorous shaking.
2. Pour 1.5mL of the culture into a micro-centrifuge tube. Centrifuge at maxi-

mum speed for 30 seconds in a micro-centrifuge at room temperature. Store
the unused portion of the original culture at 4C.
3. Remove the medium by aspiration, leaving the bacterial pellet as dry as

possible.
232
4. Re-suspend the bacterial pellet in 100L of Alkaline Lysis Solution I at room

temperature by vigorous vortexing.
5. Add 200L of freshly prepared Alkaline Lysis Solution II at room temperature

to each bacterial suspension. Close the tube tightly and mix the contents by
inverting the tube rapidly several times. Do not vortex and rapidly proceed to
Step 6.
6. Add 150L of ice-cold Alkaline Lysis Solution III. Close the tube and disperse
Alkaline Lysis Solution III through the viscous bacterial lysate by inverting the
tube several times. Transfer tube to 20C for 15 minutes to promote precipi-
tation of SDS.
7. Centrifuge the bacterial lysate at maximum speed for five minutes at room
temperature in a micro-centrifuge. Transfer the supernatant to a fresh tube
and warm to >15C to solubilise any remaining SDS.
8. Precipitate nucleic acids from the supernatant by adding two volumes of etha-
nol at room temperature (cold ethanol is best avoided as this promotes precipi-
tation of residual SDS). Mix the solution by inverting several times.
9. Collect the precipitated nucleic acids by centrifugation at maximum speed for

five minutes at room temperature in a micro-centrifuge.
10. Remove the supernatant by gentle aspiration as described in Step 3 above.

Stand the tube in an inverted position on a paper towel to allow all of the fluid
to drain away. Use a Kimwipe or disposable pipette tip to remove any drops of
fluid adhering to the walls of the tube.
11. Add 1mL of room temperature 70% (v/v) ethanol to the pellet and invert the
closed tube several times. Recover the DNA by centrifugation at maximum
speed for two minutes at room temperature in a micro-centrifuge.
12. Remove all of the supernatant by gentle aspiration as described in Step 3. Take
care with this step as the pellet sometimes does not adhere tightly to the tube.
13. Remove any beads of ethanol that form on the sides of the tube. Store the open
tube at room temperature until the ethanol has evaporated and no fluid is vis-
ible in the tube (five to 10 minutes).
NB: residual ethanol in the prepared DNA will cause the DNA (visualised
with loading dye) to rise out of the agarose gel well when setting up for gel
electrophoresis.
14. Dissolve the nucleic acids in 50L of sterile H2O. Gently mix by repetitive pi-
petting. Warm solution to 55C for 10 minutes.
15. Store the DNA solution at 20C.
233
11.2.3Protocol for plasmid preparation by cesium chloride
General preparation
Cesium chloride (CsCl) preparation.

Dissolve 11g CsCl in 5mL TE buffer, pH8.0.
Because CsCl will not dissolve at low temperatures, a water bath may be
required. All solutions and equipment must be maintained at >20C during
the procedure to ensure CsCl remains in solution.
Ethidium bromide (EtBr) is a carcinogen and appropriate protective wear

(gloves, lab coat and glasses) should be worn at all times during this proce-
dure. Consult the MSDS for EtBr prior to use.
Preparing salt-saturated isopropanol (flammable solution, avoid contact with

open flame).
Prepare 500mL of a 5MNaCl solution (dissolve the NaCl in TE, pH8.0).

Autoclave in a 1L Schott bottle.
Add 500mL isopropanol to the solution above, mix and allow the phases
to separate.
On the day of use, dispense the top layer of the salt-saturated isopropanol
into 30mL aliquots (1 aliquot per DNA sample).
Prepare the bent needle for the isopropanol extraction steps; the 90 bend
helps to maintain phase separation.
To bend the needle, hold the needle firmly with a sterile implement either
side (i.e. flame-sterilised pliers).
Pass the needle through a flame to bend it into a loop.

Avoid touching the needle tip to minimise DNA contamination.
TE buffer.
10mM Tris and 1mM EDTA, pH8.0.
TY broth (see Protocol 11.1.2).
Procedure
1. Grow overnight culture of bacteria in 400mL TY broth.
2. Spin cultures in sterile 500mL tubes.
3. Remove all growth media; use vacuum if required.
4. Re-suspend pellet in 30mL TE buffer ensuring there are no lumps.
234
5. Prepare lysate using genomic method in 15mL quick-seal (Beckman) centri-

fuge tubes.
Steps 613 must be carried out in dim light to prevent UV/EtBr interaction that can
damage the DNA.
6. Add 5mL of CsCl solution to the cleared lysate in the quick seal tube. Take
up 0.2mL pre-warmed EtBr (10mgmL1) into a syringe with needle attached
and dispense into the quick-seal centrifuge tubes containing the DNA/cesium
chloride solution.
7. Inject the DNA/EtBr solution and mix thoroughly but gently.
There must be no bubbles in the gradient tubes.
8. Seal the tubes with the heat sealer and place in a fixed angle rotor which has
been pre-warmed to 28C.
9. Centrifuge at 45,000g for 48hours to enhance separation of the forms of DNA.
10. Remove the desired DNA from the gradient by drawing off the band (intensely
stained red) with a 20mL syringe (Figure11.3), beveled edge of needle upper-
most. (NB: supercoiled DNA will migrate faster and be located at the bottom
of the gradient.)
The procedure is easier if a needle is inserted in the top of the tube to allow air in, and
a piece of tape is applied to the side of the tube at the point where the second needle
is inserted to prevent blocking up the needle (Figure11.3).
Air
Tape
Figure11.3 Drawing
off the band with a
20mL syringe.
11. Replace the straight needle with a needle pre-bent at 90 to allow ease of isopro-
panol uptake. Using the syringe containing the DNA, take up an equal volume
of salt-saturated isopropanol. Mix gently and allow phases to separate while
keeping the syringe upright. Expel the top phase with the syringe upright.
235
11.3 PCR techniques
Steps 11 and 12 must be carried out in dim light to prevent UV/EtBr interaction that
can damage the DNA.
12. Repeat the above step six times, or as required, until the ethidium bromide is
no longer visible.
13. Add the washed DNA to sterile 50mL tubes then add two volumes of sterile
water and six volumes of 100% (v/v) EtOH (volumes are proportional to the
original volume of DNA solution).
14. Spin for 10 minutes at maximum speed in a micro-centrifuge at 25C. Dry

pellet under vacuum.
15. Re-suspend pellet in 0.45mL TE buffer and transfer solution into a micro-
centrifuge tube.
16. Add 50L of 3MNa acetate (pH5.2) then 1mL of 100% EtOH; mix well.
17. Spin for 10 minutes at 10, 000g.
18. Wash pellet with 70% EtOH, then dry under vacuum.
19. Re-suspend pellet in 0.5mL (high copy number) or 0.1mL (low copy number)
TE buffer, pH8.0 (adjust volume according to pellet size).
20. Determine the concentration using the Dot method (Ausubel et al. 2013).
11.3PCR techniques
Polymerase Chain Reaction (PCR) is a common molecular biology technique

used to amplify DNA from a starting template (i.e. cells, genomic DNA or plas-
mids). Methods for cell preparation are provided in 11.3.1 and can be used in both
single and multi-amplicon generation (11.3.2 and 11.3.3 respectively). Genomic
DNA preparations (Protocol 11.1.1 and 11.1.2) may be used in PCR when prob-
lems arise from the use of cell templates. The amplified product, known as the am-
plicon, can then be used for a variety of purposes, including sequencing, cloning,
probe design. Multi-amplicons can be used for fingerprinting RNB.
11.3.1Protocol for preparation of cell template
A. Basic cell preparation
This procedure is modified from Schneider and deBruijn (1996).
Prepare a pure rhizobial culture by dilution streaking on to LA or TY agar

plates.
Incubate at 28C until growth is visible on final dilution streak.

Re-suspend a loopful of culture in 1mL of 0.89% (w/v) saline.
236
11.3 PCR techniques
1. Centrifuge the culture at 10,000 g for 30 seconds and remove the

supernatant.
NB: cultures producing large amounts of EPS may need to be washed with
0.89% (w/v) saline for a second time.
2. Re-suspend the cell pellet in sterile (DNA-free) water to an OD600nm

of 6.0 (NB: an OD600nm of 10 has been used reproducibly for Rhizobium
leguminosarum).
3. Store at 4C for up until one month or use immediately in the PCR reac-
tion below.
NB: it is important that fresh cultures of rhizobia, free of contaminants, are

used to prepare the template.
B. Cell preparation using PEG
Method (Chomczynski and Rymaszewski 2006).
Reagents
1. Alkaline PEG reagent preparation
a. Combine 60g PEG 200 (Sigma-Aldrich or equivalent) with 0.93mL 2M

KOH and 39mL water. If desired, NaOH can substitute for KOH in the
reagent. Note that PEG 200 is measured by mass rather than volume because
of the viscosity of the liquid.
b. Confirm that the pH is 13.3 to 13.5. Due to storage, some batches of PEG
200 have an acidic rather than neutral pH. In this case, add an additional
amount of alkali to reach the target pH range.
2. Scoop a loopful of a RNB colony containing your gene of interest into 500L
of sterile molecular- grade water in a sterile microfuge tube.
3. Spin down at 6000rpm for four minutes.
4. Leave 10L of water in the microfuge tube.
5. Add 100L of alkaline PEG reagent and mix well.
6. Leave at room temperature for 15 minutes.
7. Take 1.0 to 1.5L of the mixture and add into 20L of PCR reaction.
11.3.2Protocol for single amplicon PCR

Targeted single amplicons can be generated and sequenced for the purposes of
typing and taxonomic analysis (see Protocol 12.7.1 Building phylogenetic trees).
A selection of commonly used primers for single amplicon generation is provided
in Tables 11.4 and 11.5. Products used for phylogenetic analysis should be fully
237
11.3 PCR techniques
sequenced on both DNA strands for the largest fragment possible. Cells can be
prepared as described in Protocol 11.3.1A but if the desired amplicon cannot be
generated, an alternative cell preparation method is provided in Protocol 11.3.1.B.
Table11.4 Oligonucleotide primers used for PCR amplification of specific loci (chromosomal and
pSYM localised) from rhizobia
Target Primer Name Primer sequence (5-3)* Reference
gene
nodC nodC-251F AYGTHGTYGAYGACGGTTC (Laguerre et al. 2001)
nodCI-1160R CGYGACAGCCANTCKCTATTG (Laguerre et al. 2001)
nodA nodAF TGCRGTGGAARNTRNNCTGGGAAA (Haukka et al. 1998)
nodAR GNCCGTCRTCRAAWGTCARGTA (Haukka et al. 1998)
nifH nifH1 AAGTGCGTGGAGTCCGGTGG (Eardly et al. 1992)
nifH2 GTTCGGCAAGCATCTGCTCG (Eardly et al. 1992)
dnaK DnaK1468F AAG GAGCAGCAGATCCGCATCCA (Stepkowski et al. 2003)
DnaK1772R GTACATGGCCTCGCCGAGCTTCA (Stepkowski et al. 2003)
recA recAF ATCGAGCGGTCGTTCGGCAAGGG (Gaunt et al. 2001)
recAR TTGCGCAGCGCCTGGCTCAT (Gaunt et al. 2001)
16s 16S rRNA Universal forward (fD1) AGAGTTTGATCCTGGCTCAG (Weisburg et al. 1991)
rRNA1 16S rRNA Universal reverse (rP3) ACGGATACCTTGTTACGACTT (Weisburg et al. 1991)
FGPS1490 TGCGGCTGGATCACCTCCTT (Navarro et al. 1992)
FGPS6 GGAGAGTTAGATCTTGGCTCAG (Laguerre et al. 1996)
FGPS132 CCGGGTTTCCCCATTCGG (Ponsonnet and Nesme 1994)
gyrB gyrB343F TTCGACCAGAAYTCCTAYAAGG (Martens et al. 2008)
gyrB1043 AGCTTGTCCTTSGTCTGCG (Martens et al. 2008)
rpoB rpoB83F CCTSATCGAGGTTCACAGAAGGC (Martens et al. 2008)
rpoB1061R AGCGTGTTGCGGATATAGGCG (Martens et al. 2008)
atpD atpDF ATCGGCGAGCCGGTCGACGA (Gaunt et al. 2001)
atpDR GCCGACACTTCCGAACCNGCCTG (Gaunt et al. 2001)
glnA glnA532F (GSI-1) AAG GGC GGC TAY TTC CCG GT (Turner and Young, 2000)
glnA1124R (GSI-2) GTC GAG ACC GGC CAT CAG CA (Turner and Young, 2000)
dnaJ dnaJF CAGATCGAGGTSACCTTCGAC (Alexandre et al. 2008)
dnaJR CGTCRYCATMGAGATCGGCAC (Alexandre et al. 2008)
* Abbreviations: H=Adenine, Cytosine or Thymine; I=Inosine; N= Adenine, Cytosine, Guanine or Thymine;

Y=Cytosine or Thymine.
1
A single primer pair needs to be selected from the collection provided here
238
11.3 PCR techniques
Table11.5 Common reference genes selected for qRT-PCR amplification from prokaryotic RNA
Gene Bact.* Forward Primer (5-3) Reverse Primer (5-3) Reference
Recombinase A Pa GGTGAGCTGGTTGATCTGGG GCATTCGCTTTACCCTGACC Takle et al.
(recA) (2007)
Malate Ec CTGCGTAACATCCAGGACACTAACG CGACGGTTGGGGTATAAATAACAGG Wang et al.
dehydrogenase (2009)
(mdh)
Malate Rlv GCGATGACCTTCTCGGCATCA CATGGCGTCGAGCGGATTG Karunakaran
dehydrogenase et al. (2009)
(mdh)
Sigma factor Pp GTTAAGGCTTTGCACCAG GATTTCATCGACCTGCTC Chang et al.
54, (rpoN) (2009)
sigma factor Pp CGATGGAAATCACCAGAC GCTGATCGACCTTGAGAC Chang et al.
70 (rpoD) (2009)
16S ribosomal Pp CCGTGTCTCAGTTCCAGT TGAGCCTAGGTCGGATTA Chang et al.
RNA (16s rRNA) (2009)
Cation Pp GTAGTCGGCAAAGGTCTG CAACTTCTGGGTCGACAT Chang et al.
transporting (2009)
ATPase (atkA)
* Pa, Pectobacterium atrosepticum; Ec, Escherichia coli; Rlv, Rhizobium leguminosarum bv. viciae; Pp,
Pseudomonas putida
The products of those reactions can be purified from the reaction components
(i.e. salts, nucleotides and enzymes) or from agarose gel using either a commer-
cially available silica-column based kit or a non-commercial method (Ausubel et
al. 2013). A number of the latter methods have been used by the current authors,
including isolation from low melting temperature agarose gels followed by phenol
extraction or purification of DNA fragments using a Sephacryl S-300 column.
However, commercially available kits are selected when rapid sample processing
is required. The factors to consider before purchasing these kits include both the
DNA yield/recovery, the size of the DNA fragment and the elution buffer volume
and type (for example Tris-EDTA buffer may interfere with downstream reactions
such as ligations). The authors frequently use both the Wizard SV Gel and PCR
Clean-Up System (Promega, USA) and the QIAquick Gel Extraction Kit (QIA-
GEN, Germany). If the final concentration of your purified DNA sample is low,
then a simple ethanol-salt precipitation of that DNA followed by re-suspension in
a smaller volume is an easy solution. Alternatively, the MinElute PCR purification
kit (Qiagen) enables re-suspension of the purified DNA in a small volume.
11.3.3Protocol for multi-amplicon profiling (fingerprinting)

Identification and phylogenetic analysis of rhizobia can be performed using PCR
and gel electrophoresis techniques. Oligonucleotide primers (Table11.6) target-
ing naturally occurring interspersed repetitive DNA elements enable amplifica-
tion of multiple DNA fragments of different sizes simultaneously. The unique
banding pattern (DNA fingerprint) is visualised using gel electrophoresis to type
rhizobia. Primer pairs have been designed from the repetitive extragenic palin-
dromic (REP), enterobacterial repetitive intergenic consensus (ERIC) (Versalovic
239
11.3 PCR techniques
et al. 1991) and BOX sequences. The complete protocols for preparation of cell
templates and repetitive-PCR (see Table 11.7 for PCR reaction conditions) are
provided here, and have frequently been used by the authors for the genera
Bradyrhizobium, Mesorhizobium, Rhizobium and Sinorhizobium. Information and
protocols for additional repetitive-PCR types are available from http://www.msu.
edu/~debruijn/ (Rademaker and De Bruijn 1997).
Table11.6 Oligonucleotide primers used for PCR fingerprint profiling of rhizobia

Primer Sequence (5-3)* Reference
ERIC 1R ATGTAAGCTCCTGGGGATTCAC Versalovic et al. (1991)
ERIC 2 AAGTAAGTGACTGGGGTGAGCG Versalovic et al. (1991)
REP1R-I IIIICGICGICATCIGGC Versalovic et al. (1991)
REP2-I ICGICTTATCIGGCCTAC Versalovic et al. (1991)
BOXAIR CTACGGCAAGGCGACGCTGACG Versalovic et al. (1994)
RPO1 AATTTTCAAGCGTCGTGCCA Richardson et al. (1995)
PucFor GTAAAACGACGGCCAGT Laguerre et al. (1996)
* Abbreviations: I=Inosine.
Table11.7 PCR reactions for ERIC, BOX and rep

Reagent (stock concentration) Volume (L) for 1 reaction
Template cells (OD600nm 6.0) 1
Gitschier buffer (5) (See Table11.8) 5
Bovine Serum Albumin (BSA at 20 mg mL1) 0.2
Dimethyl sulfoxide (100% DMSO) 2.5
Primer 1 (50 M) 1
Primer 2 (50 M)* 1*
dNTP mix (25 mM) 1.25
Taq Polymerase (5 U/L) 0.4
DNase-free water to 25
* Not relevant for BOX, RPO1, pUCF as they are the only primer required.
NB: the authors use Taq Polymerase (Invitrogen).
Use DMSO at room temperature and mix stock if precipitates form.
Use filter-tips when preparing the PCR mixture and all stock solutions.
PCR Cycle
Temperature (C) Time Cycles

95 7 minutes 1
94 1 minutes
52* 1 minutes 35
65 8 minutes
65 16 minutes 1
* 53C for BOX, 40C for REP
240
11.3 PCR techniques
NB: store the reaction tubes at 20C until ready to visualise by gel electrophoresis.
11.3.4Gel electrophoresis
1. Prepare a 1.5% (w/v) agarose gel in TAE buffer (0.75g of molecular-grade aga-
rose in 50mL of 1TAE buffer).
Melt mixture in a microwave or water bath until clear. Cool to 50C and add
SYBR- safe DNA gel stain (Invitrogen; use at a final concentration of 1) or
Ethidium Bromide (final concentration of 0.5gmL1) and mix by swirling.
Pour the gel into a gel tray, slot in the appropriate size comb and allow to set.
2. Mix loading buffer to 1 final concentration with PCR amplified sample.
3. Place the set gel inside the gel tank filled with 1TAE buffer.
4. Load PCR-dye mix prepared in Step 2 into a well. Also load a commercial 1kb
DNA ladder with size ranges from 250 to 10,000bp in a well.
5. Connect the electrodes to the power source and run the gel at 80V (time is de-
pendent on gel size but approximately three to six hours; reproducible results
require this to be standardised between runs).
11.3.5Analysis
Unique banding patterns (e.g. Figure11.4) can be obtained by these repetitive-
PCR methods to assist with rhizobia typing. Banding patterns are analysed by
determination of amplicon size relative to a 1kb ladder. The presence of a control
(mother-culture-isolate) is necessary to validate your samples. Reproducibility
may be enhanced by performing duplicate PCR reactions in independent cycling
runs. The authors also use the Phoretix 1D Advanced software analysis package
(Nonlinear Dynamics, UK) for analysis of banding patterns from uploaded gel
images. Alternative software packages such as AMBIS and GelCompar are men-
tioned at http://www.msu.edu/~debruijn/ (Rademaker and De Bruijn 1997).
Figure11.4 ERIC PCR amplification from Rhizobium

leguminosarum bv. trifolii strain WSM1325 from four
different storage sources (lanes 25) aligned with
Promega 1kb DNA marker (lane 1).
241
11.4References
Reagent stock preparation
Table11.8 5 Gitschier Buffer

Reagent Volume (mL)
Tris (1 M, pH 8.8) 6.7
(NH4)2SO4 (1 M) 1.66
0.005 M EDTA (pH 8.8) 0.13
MgCl2 (1 M) 0.67
2-Mercaptoethanol (14.4 M) 0.21
DNase-free water 10.63
Total Volume 20
NB: prepare 5 Gitschier buffer and store 1mL aliquots at 20C.
6 Loading Dye: 0.4% (w/v) bromophenol blue in 50% (v/v) glycerol.
TAE Buffer (50): 242g Tris base, 57.1mL glacial acetic acid, 100mL of 0.5M
EDTA (pH8.0), distilled water to 1L final volume.
11.4References
Alexandre A., Laranjo M., Young J.P. and Oliveira S. 2008. dnaJ is a useful phylogenetic marker
for alphaproteobacteria. International Journal of Systematic and Evolutionary Biology 58,
28392849.
Ausubel F.M., Brent R., Kingston R.E., Moore D.D., Seidman J.G., Smith J.A. and Struhl K. 2013.
Current Protocols in Molecular Biology. Wiley and Sons.
Chang C., Damiani I., Puppo A. and Frendo P. 2009. Redox changes during the legume-Rhizobium
symbiosis. Molecular Plant 2, 370377.
Chomczynski P. and Rymaszewski M. 2006. Alkaline polyethylene glycol-based method for direct
PCR from bacteria, eukaryotic tissue samples, and whole blood. BioTechniques 40, 454458.
Eardly B.D., Young J.P. and Selander R.K. 1992. Phylogenetic position of Rhizobium sp. strain Or
191, a symbiont of both Medicago sativa and Phaseolus vulgaris, based on partial sequences of
the 16S rRNA and nifH genes. Applied and Environmental Microbiology 58, 18091815.
Eckhardt T. 1978. A rapid method for the identification of plasmid deoxyribonucleic acid in bacte-
ria. Plasmid 1, 584588.
Gaunt M.W., Turner S.L., Rigottier-Gois L., Lloyd-Macgilp S.A. and Young J.P. 2001. Phylogenies of
atpD and recA support the small subunit rRNA-based classification of rhizobia. International
Journal of Systematic and Evolutionary Biology 51, 20372048.
Griffiths R.I., Whiteley A.S., ODonnell A.G. and Bailey M.J. 2000. Rapid method for coextraction
of DNA and RNA from natural environments for analysis of ribosomal DNA- and rRNA-based
microbial community composition. Applied and Environmental Microbiology 66, 54885491.
Haukka K., Lindstrom K. and Young J.P. 1998. Three phylogenetic groups of nodA and nifH genes in
Sinorhizobium and Mesorhizobium isolates from leguminous trees growing in Africa and Latin
America. Applied and Environmental Microbiology 64, 419426.
Hynes M.F. and McGregor N.F. 1990. Two plasmids other than the nodulation plasmid are neces-
sary for formation of nitrogen-fixing nodules by Rhizobium leguminosarum. Molecular Micro-
biology 4, 567574.
Hynes M.F. Simon R. and Phler A. 1985. The development of plasmid-free strains of Agrobacterium
tumefaciens by using incompatibility with a Rhizobium meliloti plasmid to eliminate pAtC58.
Plasmid 13, 99105.
Karunakaran R., Ramachandran V.K., Seaman J.C., East A.K., Mouhsine B., Mauchline T.H., Prell J.,
Skeffington A. and Poole P.S. 2009. Transcriptomic analysis of Rhizobium leguminosarum biovar
242
11.4References
viciae in symbiosis with host plants Pisum sativum and Vicia cracca. Journal of Bacteriology
191, 40024014.
Laguerre G., Mavingui P., Allard M.R., Charnay M.P., Louvrier P., Mazurier S.I., Rigottier-Gois L.
and Amarger N. 1996. Typing of rhizobia by PCR DNA fingerprinting and PCR-restriction frag-
ment length polymorphism analysis of chromosomal and symbiotic gene regions: application to
Rhizobium leguminosarum and its different biovars. Applied and Environmental Microbiology
62, 20292036.
Laguerre G. Nour S.M., Macheret V., Sanjuan J., Drouin P. and Amarger N. 2001. Classification of
rhizobia based on nodC and nifH gene analysis reveal a close phylogenetic relationship among
Phaseolus vulgaris symbionts. Microbiology 147, 981993.
Martens M., Dawyndt P., Coopman R., Gillis M., De Vos P. and Willems A. 2008. Advantages of
multilocus sequence analysis for taxonomic studies: a case study using 10 housekeeping genes
in the genus Ensifer (including former Sinorhizobium). International Journal of Systematic and
Evolutionary Biology 58, 200214.
Navarro E., Simonet P., Normand P. and Bardin R. 1992. Characterization of natural populations of
Nitrobacter spp. using PCR/RFLP analysis of the ribosomal intergenic spacer. Archives of Mi-
crobiology 157, 107115.
Ponsonnet C., Nesme X. 1994. Identification of Agrobacterium strains by PCR-RFLP analysis of pTi
and chromosomal regions. Archives of Microbiology 161, 300309.
Priefer U.B. 1984. Isolation of Plasmid DNA Springer-Verlag: Berlin.
Rademaker J.L.W., andDe Bruijn F.J. 1997. Characterization and classification of microbes by REP-
PCR genomic fingerprinting and computer-assisted pattern analysis. Pp 151171 in DNA
Markers: Protocols, applications and overviews, ed. by G. Caetano-Anolles and P.M. Gresshoff.
John Wiley: New York.
Richardson A.E., Viccars L.A., Watson J.M. and Gibson A.H. 1995. Differentiation of Rhizobium
strains using the polymerase chain reaction with random and directed primers. Soil Biology and
Biochemistry 27, 515524.
Schneider M. and deBruijn F.J. 1996. Rep-PCR mediated genomic fingerprinting of rhizobia and
computer-assisted phylogenetic pattern analysis. World Journal of Microbiology and Biotech-
nology 12, 163174.
Stepkowski T., Czaplinska M., Miedzinska K. and Moulin L. 2003. The variable part of the dnaK
gene as an alternative marker for phylogenetic studies of rhizobia and related alpha Proteobac-
teria. Systematic and Applied Biology 26, 483494.
Takle G.W., Toth I.K. and Brurberg M.B. 2007. Evaluation of reference genes for real-time RT-PCR
expression studies in the plant pathogen Pectobacterium atrosepticum. BMC Plant Biology 7, 50.
Turner S.L. and Young J.P. 2000. The glutamine synthetases of rhizobia: phylogenetics and evolu-
tionary implications. Molecular Biology and Evolution 17, 309319.
Versalovic J., Koeuth T., Lupski J.R. 1991. Distribution of repetitive DNA sequences in eubacteria
and application to fingerprinting of bacterial genomes. Nucleic Acids Research 19, 68236831.
Versalovic J., Schneider M., de Bruijn F.J. and Lupski J.R. 1994. Genomic fingerprinting of bacteria
using repetitive sequence based PCR (rep-PCR). Methods Molecular and Cell Biology 5, 2540.
Wang M., Jacoby G.A., Mills D.M. and Hooper D.C. 2009. SOS regulation of qnrB expression. Anti-
microbial Agents and Chemotherapy 53, 821823.
Weisburg W.G., Barns S.M., Pelletier D.A. and Lane D.J. 1991. 16S ribosomal DNA amplification for
phylogenetic study. Journal of Bacteriology 173, 697703.
243
12.1Introduction
CHAPTER12
Specialised genetic
techniques for rhizobia
W.G. Reeve, R.P. Tiwari, V. Melino, S. De Meyer and
P.S. Poole
12.1Introduction
The aim of this chapter is to provide a selection of specialised genetic techniques

which are widely used to interrogate gene functions in root nodule bacteria
(RNB). The reader is referred to reviews of the applicability of these techniques
to understand biological processes occurring in RNB (i.e. Long 1989; Stanley and
Cervantes 1991). The techniques presented here cover the ability to randomly
mutate genes, select mutants of interest based on their phenotype and identify
the gene affected, verify that the gene mutation caused the observed phenotypic
defect, examine expression of a target gene, perform genome structural and func-
tional studies and conduct comparative analyses with other RNB.
12.2Transposon mutagenesis
Mutations can be introduced into RNB by the use of chemical, physical and bio-
logical mutagens. Transposon mutagenesis is the most widely used technique for
introducing random mutations in RNB. Alternative methods for targeting specific
genes have been described elsewhere (Kokotek and Lotz 1991; Quandt and Hynes
1993; Selbitschka et al. 1993). Random mutations can be introduced by transfer
of a mobilisable, suicidal plasmid carrying a trans-positionally active mobile ele-
ment (transposon, mini-transposon or plasposon). A plasmid carrying the mo-
bile element is conjugally transferred from an RP4 integrant of Escherichia coli,
such as BW20767 (Metcalf et al. 1996) or S17.1 (Simon 1984). Alternatively, trans-
fer can be mediated by a helper E. coli strain, such as MT616 (Finan et al. 1986) or
HB101 (pRK2013) (Figurski and Helinski 1979) to provide RP4 transfer functions
245
12.2 Transposon mutagenesis
to a non-RP4 integrant. A range of transposons, or derivatives, are available and

include: Tn5 and antibiotic-resistant derivatives (De Vos et al. 1986; Quandt et
al. 2004; Selvaraj and Iyer, 1983; Simon 1984); mTn5 constructs (de Lorenzo et
al. 1990; Herrero et al. 1990; Pobigaylo et al. 2006; Reeve et al. 1999; Wilson et al.
1995); mTn10 constructs (Herrero et al. 1990); and plasposons (Dennis and Zyl-
stra, 1998). An appropriate choice of mutagen will be dependent on the antibiotic
sensitivity of the rhizobial strain. The presence of a promoterless reporter in the
mutagen enables gene expression to be monitored at the site of insertion, while
a promoter driven reporter in a mutagen can permit cell localisation, which is
particularly useful for nodule occupancy or infection studies. Reporters that have
been used include: cat (Kim et al. 1986); celB (Sessitsch et al. 1996); gusA (Reeve
et al. 1999); lacZ (de Lorenzo et al. 1990; Herrero et al. 1990); gfp (Xi et al. 1999);
and derivatives, lux (Legocki et al. 1986), nptII (Simon et al. 1989)and phoA (de
Lorenzo et al. 1990; Muller 2004).
Many of the constructs in current use are based on the pUT plasmid delivery sys-
tem. This plasmid is stable only in hosts synthesising PIR proteins, such as E. coli
strain BW20767 (chromosomal gusA:pir), and not in root nodule bacteria devoid
of pir. In this plasmid delivery system, the tnpA gene is located on the plasmid ad-
jacent to the minitransposon effectively disarming the transposon and prevent-
ing further transposition. To illustrate this process (Figure12.1), an approach is
provided to mutagenise Sinorhizobium medicae WSM419 with mTn5-GNm. The
mini-transposon mTn5-GNm contains a kanamycin resistance (nptII) and a pro-
moterless gusA gene (Reeve et al. 1999). This strain is naturally resistant to chlo-
ramphenicol.
When the suicide plasmid is transferred conjugally from the donor to recipient
S. medicae, the plasmid pCRS487 cannot replicate in the new background and
kanamycin-resistant transconjugants arise from the transposition of the mini-
transposon into the rhizobial genome. Insertion of mTn5 will inactivate the target
gene, and if in the correct orientation, the gusA gene will be expressed under the
inactivated genes promoter. GUS expression can be monitored as described in
Protocol 12.5.
Transconjugants are selected on a medium containing kanamycin and chlo-

ramphenicol, preventing growth of the donor E. coli and recipient parent strain
S. medicae WSM419. Kanamycin-resistant Sinorhizobium transconjugants are in-
sertion mutants. The phenotype of the mutated cells can then be examined by
replica plating colonies onto particular types of selective media.
246
Figure12.1 A schematic for minitransposon mutagenesis of S. medicae
12.2.1Frequency of isolation of mutants

The genome of S. medicae WSM419 contains 6,518 protein coding genes, and if a
single gene is involved in the expression of a characteristic, one could expect, on
average, to have to screen 6,518 kanamycin-resistant colonies in order to be sure
of isolating a mutation in a particular gene. In reality, this number can be higher
due to the presence of hotspots for insertion.
12.2.2Protocol for transposon mutagenesis

The overall strategy for isolating minitransposon-induced mutants involves:
1. transfer of mTn5 into rhizobial cells by conjugation
2. selection of rhizobial transconjugants containing mTn5 insertions
3. selection of the desired mutant displaying the expected phenotype.
Strains
E. coli BW20767 (pCRS487) ApR, KmR.

S. medicae WSM419 CmR, NxR.
247
Recipe
TY broth:
Tryptone 5g
Yeast extract 3g
CaCl2.2H2O 0.75 g (reduced from original recipe to avoid precipitation)
Distilled H2O to 1 L
Adjust pH to neutral
For plates: add 1.5 % (w/v) agar and autoclave.
Procedure
Day 1
1. Inoculate one loopful of the Sinorhizobium culture (chloramphenicol resist-

ant) into 5mL TY broth containing chloramphenicol (20gmL1). Place the
inoculated broth on a gyratory shaker set to 200rpm at 28C.
Day 3
2. Inoculate E. coli culture BW20767 (pUT::mTn5-GNm) into 5 mL LB broth

(containing ampicillin 100gmL1and kanamycin 100g mL1) and incubate
on a gyratory shaker set to 200rpm at 37C.
Day 4
3. Subculture E. coli 1:50 into LB broth containing the same antibiotics and place
the culture on a 37C shaker. Allow it to grow for three to four hours.
4. Centrifuge 5mL of the 4d Sinorhizobium culture and the subcultured E. coli

culture. Discard the supernatant and re-suspend pellets in 250L of TY broth.
5. Spot 50L of each parent separately onto TY plates to act later as controls.
Combine E. coli and Sinorhizobium cultures, mix and spot four aliquots of
100L onto TY plates, divided into quadrants. Leave until dry and incubate
the TY plates at 28C.
Day 5
6. Re-suspend each spot (mating mixture) into 1 mL sterile saline containing

15% (v/v) glycerol.
7. Prepare 101 to 108 dilutions of a mating mixture and spread 100L from neat
to 103 dilutions onto TY plates containing chloramphenicol and kanamycin,
and 100L from 105 to 108 on TY plates containing just chloramphenicol. In
248
addition, spread neat to 103 dilutions of each control onto TY plates contain-
ing chloramphenicol and kanamycin. Incubate the plates at 28C.
Day 8
8. Perform counts on plates containing between 30 and 300 colonies. These data
can be used to calculate the efficiency of transposition but are only necessary
if the transposition frequency in your strain needs to be determined.
9. From the counts obtained, spread the appropriate volumes to obtain the de-
sired number of mutants that need to be screened, and incubate for three to
four days at 28C.
Days 11 to 13
10. Replica plate kanamycin-resistant Sinorhizobium transconjugants onto selec-

tive media. For example, mutations in genes expressed in acidic conditions
can be identified by replica patching mutants on minimal media containing X-
Glc (a substrate for -glucuronidase) which has been buffered to pH7 and 5.7.
The random nature of transposition can be verified by analysing auxotrophic
phenotypes in your strain (Selvaraj and Iyer 1983).
Day 14
Observe plates for growth and also for colour development. Plates can be stored at
4C until required. The plates will enable you to select the following.
a. Mutants of Sinorhizobium lacking one of the proteins required to grow at low

pH. These mutants will grow on the pH7 plate but will not grow on the low
pH plate.
b. Reporter gene fusions with low pH up-regulated or low pH down-regulat-

ed genes. Transcription of the target gene will control the expression of the
inserted promoterless gusA gene. Expression of the gusA gene will produce
-glucuronidase enzyme that converts X-Glc to the blue-coloured product.
Thus, mutants containing a reporter gene fusion with a low pH up-regulated
gene will produce colonies with deeper blue colour at pH5.7 than at pH7.0.
Reporter gene fusions in a low pH down-regulated gene will show the reverse
pattern of colour on low and neutral pH plates.
12.2.3Identification of the mutated gene

Once mutants are isolated by phenotype testing, the affected gene in the mutant
can be identified by DNA cloning and sequencing. The presence of an antibiotic
marker in the biological mutagen facilitates cloning of the transposon and asso-
ciated rhizobial flanking sequence. Genomic DNA can be isolated from the mu-
tant, as described in Protocol 11.1 (Chapter11). The selection of a restriction en-
zyme that does not cut inside the mutagen is particularly useful for this purpose.
249
12.3 Protocol for transduction
Restriction enzyme-cut genomic DNA can be ligated with a suitable vector (such
as pUC18) and transformants selected on growth media supplemented with an-
tibiotics that select for the cloning vector and the mutagen. For example, the en-
zyme EcoRI (or HindIII) can be used to cleave out mTn5-GNm and associated
rhizobial DNA from mutant DNA; it can then be ligated into EcoRI (or HindIII)-
cut pUC18 DNA. In this case, transformants can be selected on LB plates supple-
mented with ampicillin and kanamycin.
The plasmid DNA of desired transformants can be purified by conventional meth-

ods and the DNA flanking the transposon sequenced using primers designed at
the ends of the transposon. For example, the primers WIL3 and TAC-105F can be
used to sequence out from the ends of mTn5-GNm (Reeve et al. 1999). The gene
affected can be identified by searching DNA databases (Protocol 12.7) with the
sequence generated.

Generalised transduction is one method that can be used to determine whether a
specific tagged mutation (i.e. antibiotic resistance marker on a transposon) is the
cause of a noted phenotype. The process requires the presence of a suitable gener-
alised transducing phage for the RNB host since the process is very specific. Prior
to the commencement of the transduction protocol, the phage is passaged twice
through the wild-type host followed by the preparation of a mutant lysate at high
titre. The mutant lysate is irradiated and used to infect the wild-type to reintro-
duce the mutation. The phenotype of transduced cells can then be compared to
the expected phenotype for the mutation of interest.
An example of transduction is provided using the generalised transducing phage

RL38 on the host Rhizobium leguminosarum bv. viciae. A host-specific transduc-
ing phage is required for these procedures to be useful.
12.3.1Preparation of the wild-type phage lysate

1. Grow culture of wild-type RNB strain overnight in 5mL TY broth (see recipe
in Section 12.2.2).
2. Subculture 0.5mL into 5mL and grow until OD600nm = 0.3.
This gives approximately 3 108 cells mL1.
3. Infect bacteria with RL38 phage at a ratio of 0.5:1 (phage:cell).
While this ratio is ideal, it can be less. The method works successfully if the
ratio drops to 1:10 phage:cells.
4. Incubate overnight at 28C with shaking.
250
5. Spin cells for 15 minutes at full speed in a microcentrifuge and take superna-
tant into fresh sterile glass tubes. Add a drop of chloroform, mix and store at
4C.
THIS IS THE PRIMARY LYSATE
6. Repeat the procedure (Steps 1 to 5) by infecting wild-type cells with the pri-
mary lysate generated in Step 5. The titre can be established as detailed below.
This step overcomes the restriction mechanism of the host bacteria and pro-
duces a secondary lysate.
Determine the phage titre to calculate yields and volumes to be used for infection
cells. Yields vary from between about 104 to 108 pfu mL1.
12.3.2Improving phage titre

To obtain higher yields of phage, the lysates can be collected from agar plates us-
ing the following procedure.
1. Mix the lysate and bacteria with 5mL 42C top agar (0.7%(w/v) TY agar) in
Falcon 2057 tubes and pour onto pre-warmed TY plates.
2. Incubate at 28C for three to four days.
3. Elute phage with 3mL TY broth (see recipe in Section12.2.2); leave for two
hours.
4. Remove elution buffer to fresh sterile glass tubes, add a drop of chloroform
and store at 4C.
12.3.3Preparation of the mutant phage lysate

1. Grow overnight culture of mutant strain at 28C in 5mL TY broth.
2. Subculture 0.5mL into 5mL and grow until OD600nm=0.3.
This gives approximately 3 108 cells mL1.
3. Infect bacteria with RL38 phage at a ratio of 0.5:1 (phage:cell).
This is an ideal ratio but it can also be less; the method works successfully if the
ratio drops to 1:10 phage:cells
4. Incubate overnight at 28C with shaking.
5. Spin cells for 15 minutes at full speed in a microcentrifuge and remove the
supernatant to fresh sterile glass tubes. Add a drop of chloroform and store at
4C.
There is no need to obtain a secondary lysate from mutant stock.
251
Determine the phage titre to calculate yields. Yields needs to be at least 108 plaque-
forming units per mL (pfumL1).
To improve yields of phage, the mutant lysate can be collected from agar plates
using the procedure Improving phage titre (Section 12.3.2).
12.3.4Calculation of phage titre

1. Add 100L of neat and various dilutions (102 to 108) of lysate to Falcon 2057
tubes containing 100L of log-phase wild-type bacteria (OD600nm=0.3).
2. Incubate at 28C for between 15 minutes and three hours.
3. Add 5mL 42C top agar (0.7% (w/v) TY agar) and pour onto pre-warmed TY
plates.
4. Incubate at 28C for three to four days.
5. Select plates with between 50 to 300 plaques and count. Back calculate titre
using the following formula.
Number of plaques dilution factor 10 (converts to mL)=number of pfu

mL1 of lysate
E.g. If 0.1mL of 104 dilution gives 20 plaques then titre is:
20 104 10=2 106 pfu mL1
12.3.5Phage irradiation
1. Add phage lysate to a Petri plate placed on a shaking platform. A good volume
is 5mL. Set the shaker to gentle shaking. UV irradiate the suspension with
the Petri lid off until the titre has dropped 1000-fold. Establish titre using the
method Calculation of phage titre. (Section 12.3.4).
Optimal irradiation is 2 mJ/m2/s.
12.3.6Transduction
1. Grow overnight culture of wild-type bacteria in 5mL TY broth.
2. Subculture 0.5mL into 5mL and grow until OD600nm=0.3 (approximately 3

108 cells mL1).
3. Infect 0.1mL bacteria with 1, 10 and 100L irradiated RL38 mutant phage
lysate (see above for irradiation protocol).
Washing 0.1mL of bacterial cells in 0.9% (w/v) saline and then re-suspending
in 0.1mL of TY prior to infection may aid in optimising the infection cycle
for highly mucoid strains. However, problems with infection have also been
encountered after performing this step.
252
12.4 Protocol for conjugation
4. Incubate at 28C for between 15 minutes and three hours.
5. Spin mixture, wash in 0.9% (w/v) saline and re-suspend in 0.23mL of 0.9%
(w/v) saline.
This step removes phage that have not attached and injected their DNA into
the bacteria. It also removes calcium, preventing any further infection. Selec-
tion is for infected cells, preferably those infected with non-lysing phage dam-
aged by irradiation.
6. Spread 1, 10 and 100L aliquots of the mixture onto minimal media plates
containing no more than the absolute required amount of calcium for cell
growth.
E.g. 0.25mM CaCl2 in JMM minimal medium works well. Add the appropri-
ate selective agent for the marker to be transduced. For example, if checking
for Tn5 transduction, add kanamycin (an appropriate level for RNB would be
100 g mL1).
7. Check kanamycin-resistant colonies on selective media for co-transduction of

the mutant phenotype. Ensure the cells grow at the expected rate; occasionally
abortive transduction may yield slow-growing colonies. Perform Southern hy-
bridisation or PCR to confirm position of mutation.
Conjugation, also known as mating, is used by the authors to transfer plasmids

from E. coli into rhizobial strains. A variety of plasmids can be mobilised into
rhizobia; a list of commonly used vectors is provided in Table 12.1. Biparental
mating is the transfer of DNA from one E. coli parent to one rhizobial strain,
relying upon the availability of a mob (mobilisable) gene present on the plasmid
of transfer. The authors recommend the use of the E. coli donor BW20767 for
conjugative transfer of R6K oriV and RP4 oriT plasmids (Metcalf et al. 1996).
Triparental mating involves the transfer of DNA from one E. coli parent to one
rhizobial strain using an additional E. coli strain containing a helper plasmid, such
as pRK2013 which has a ColE1 replicon (Figurski et al. 1979). Triparental mating
would be used in the circumstance where the donor E. coli does not carry the tra
genes conferring transfer of the plasmid. Either of the two methods can be used
to mobilise a plasmid into a rhizobial strain for the purpose of complementation
(restoration of a wild-type phenotype), gene expression studies and mutant con-
struction. Conjugation can also be used to mobilise a plasmid from a rhizobial
strain back into E. coli using a triparental mating involving a helper E. coli strain.
Two methods for mating are described (filter and plate methods) and preference
for their use is at the discretion of the researcher.
253
Table12.1 Common mobilisable plasmids and their useful features

Plasmid Replicon Antibiotic Application Features2, 4 Reference
type marker1
pBBR series pBBR1 Varies (Cm Complementation BHR, blue-white (Kovach et al. 1995; Kovach et
or Km or Tc selection al. 1994)
or Ap or Gm)
pJP2series 3 IncP Tc Expression and BHR, (Karunakaran et al. 2005)
cloning autofluorescent
genes and some
carry gusA
pOT series 3 pBBR1 Gm Expression and BHR, (Karunakaran et al. 2005)
cloning autofluorescent
genes and some
carry lacZ
pFUS1 IncP Tc Expression Promoterless (Reeve et al. 1999)
gusA, BHR
pFUS1P IncP Tc Expression Promoterless (Yost et al. 2004)
gusA, BHR,
includes par
stability loci
pMP220 IncP Tc Expression Promoterless (Spaink et al. 1987)
lacZ, BHR
pWR220 IncP Tc Complementation BHR (Reeve et al. 1999)
pUFR009 IncQ Km Complementation BHR (DeFeyter et al. 1990)
pUFR042 IncW Gm, Nm Cloning and BHR, blue white (Defeyter and Gabriel 1991)
Complementation screening, phase
packaging site
(cos)
pUFR047 IncW Ap, Gm Cloning and BHR, blue white (De Feyter et al. 1993)
complementation screening, lacZ
pSUP series ColE1 or Varied Homologous NHR and (Simon et al. 1986)
p15A recombination varied features
or transposon dependent on
mutagenesis construct
pWS233 p15A Gm, Tc Homologous NHR, sucrose (Selbitschka et al. 1993)
recombination sensitive (sacB),
mobilisable
pJQ200 P15A Gm Homologous NHR, sucrose (Quandt and Hynes 1993)
recombination sensitivity
(sacB), blue-
white selection
pUT series R6K Ap Transposon NHR in pir- (de Lorenzo et al. 1990);
mutagenesis strain (Herrero et al. 1990); (Reeve et
al. 1999)
1
Ap: ampicillin; Cm: chloramphenicol; Gm: gentamycin; Km: kanamycin; Tc: tetracycline.
2
BHR: broad-host range; NHR: narrow host-range (suicidal vectors are unable to replicate in rhizobia).
3
Deposited in addgene (www.addgene.com)
4
Reporter fusions can be assayed in multiple copy (BHR plasmid-borne) in different genetic backgrounds or in
single copy as integrants (NHR delivered).
254
12.4.1General preparation
This protocol relies on the use of the recipient rhizobial strain at stationary
phase and the donor E. coli strain at log phase.
rhizobium at stationary phase.

E. coli at log phase after subculture from overnight growth.
Pre-warm 0.89% (w/v) saline (sterile) and TY plates (see recipe in Protocol
12.2.2) to 28C for optimum efficiency.
Day 1: inoculate 10mL TY culture with rhizobium (recipient) and grow at the
appropriate temperature until it reaches saturation (e.g. three-day growth of
Sinorhizobium medicae WSM419).
Day 3: inoculate 10mL LB culture with E. coli (donor carrying plasmid of in-
terest) maintaining selection for the plasmid with the appropriate antibiotics.
Day 4: subculture E. coli (1 in 50 dilution) and grow (approximately three

hours) without antibiotics.
Day 4: prepare TY plates.

Recommended ratio for biparental matings:
log-phase E. coli RP4 integrant (donor) and stationary-phase rhizobium at
1:1 ratio.
Recommended ratios for triparental matings:

log-phase E. coli (donor):log-phase E. coli (helper plasmid):stationary-phase
rhizobium (recipient) 1:1:1 ratio. Alternatively, stationary-phase E. coli can
be mixed with cultures of rhizobium at a ratio of 1:5.
12.4.2Procedure
Protocol 1: Plate mating
1. Separately centrifuge 5mL of each culture at maximum speed for one minute
and re-suspend pellet in 200L of 0.89% (w/v) saline (1:25).
2. Combine the donor and recipient (and helper) cultures together and tap gen-
tly to mix.
3. Dot 100L amounts (mating spots) of the prepared mixed culture onto a pre-
warmed TY plate (no antibiotics) and incubate overnight at 28C.
4. NB: additional mating spots may be made on the same plate well separated
from each other.
255
12.5 Reporter assays
Protocol 2: Filter mating
1. Pass 1mL E. coli culture through a 0.45m filter then pass 1mL rhizobium
culture through (this washes away any residual antibiotics).
2. Place filter cell-side up on pre-warmed TY plate, incubate overnight at 28C.
Protocol continued for Protocols 1 or 2
3. Re-suspend one mating in saline containing glycerol (850L 0.89% (w/v) sa-
line with 150L of 80% (v/v) glycerol).
Cells from Protocol 1 can be scraped off a plate while the entire filter from
Protocol 2 can be immersed in the re-suspension tube.
4. At this stage, the mixture can be stored at 4C or 80C.
5. Plate mating mixture (100, 10 and 1L aliquots) onto desired selective media
using marker on host and selecting for plasmid or transposon.
6. Plate controls:
i. rhizobium on selective media
ii. rhizobium dilutions on TY to determine the rate of conjugation
iii. E. coli on selective media.
The measurement of reporter gene expression enables expression of native RNB

genes to be determined quantitatively or qualitatively under the conditions as-
sayed. Reporters can be mobilised by conjugation (see Protocol 12.4) into rhizo-
bia on various vector backbones (see Table 12.1). There are many protocols to
reveal the level of expression of a reporter and this will ultimately depend on
the type of reporter gene chosen and the relative ability to permeabilise the cells
(to ensure the rapid onset of enzyme activity in the presence of the appropriate
substrate). This protocol has been widely used to determine gusA, phoA or lacZ
expression. Section 12.5.1 describes three different methods for cell permeabili-
sation/lysis. The most appropriate method to be used will be determined by the
degree of permeabilisation/lysis, which is dependent on the cell type. The authors
suggest the use of Method 1 (toluene method) in the first instance. This method
is highly reproducible and results in efficient permeabilisation of a wide range of
rhizobial strains. However, it suffers from the following disadvantages: (i) toluene
is toxic; (ii) the process speed is reduced by the need for vigorous vortexing of
samples after toluene addition; and (iii) an additional 30 minutes incubation is
required to remove toluene. An alternative is to use either Method 2 or Method 3
listed in Section 12.5.1. These latter two methods are amenable to high through-
put but may not lyse/permeabilise all cell types efficiently. The method should be
256
optimised for each cell type to ensure that there is no significant lag in enzyme ac-
tivity resulting from incompletely permeabilised/lysed cells. A procedure is pro-
vided for both low (cuvette) and high throughput (microtitre plate).
12.5.1Quantitative method
Growing the cells
1. Prepare a starter culture by growing cells in 5mL media supplemented with

appropriate antibiotics.
2. Add culture to final choice of media (containing antibiotic if required; do not

add tetracycline at low pH) so that after 16 to 20hours, growth OD595nm reach-
es 0.5 to 0.8 (growth phase of the cells is critical for reproducible results).
3. Harvest cells the next day by centrifugation and adjust OD595nm of cell sus-
pension to 2.0 using normal saline (0.89% w/v NaCl) for Method 1 and 2 OR
100mM Tris (pH 8.0) for Method 3.
12.5.1.1Cell permeabilisation
Reporter enzyme buffers
GusA assay buffer: 50mM Na phosphate buffer (60mM Na2HPO4 + 40mM

NaH2PO4; pH7.0); 50mM dithiothreitol; 1mM EDTA.
PhoA assay buffer: 100mM Tris pH8.0 (pH adjusted using HCl).
LacZ assay buffer: 50mM Na phosphate buffer (60mM Na2HPO4 + 40mM
NaH2PO4; pH7.0), 10mM KCl and 1mM MgSO4, 50mM -mercaptoethanol
(2.7mL L1).
Method 1 (toluene method)
1. Dispense 200L of each re-suspended culture at OD595nm= 2 into a microcen-

trifuge tube.
Perform this step in triplicate. The volume used will depend on the activity of
the fusion and may have to be determined empirically.
2. Add 800L of the appropriate reporter enzyme buffer to each microcentrifuge

tube.
3. Add two drops toluene (addition of two drops has no effect on GUS activity)
to each tube and VORTEX vigorously for a standardised period of time (i.e.
ten seconds/sample).
This is the critical step to ensure that the rate is linear from time zero. Shaking
samples on a microtitre plate shaker is not vigorous enough.
4. Place microfuge tubes at 37C for 30 minutes with the lids open.
Evaporates toxic toluene. Perform in a fume hood.
257
Method 2 (lysozyme and SDS method)
1. Dispense 200L of each re-suspended culture at OD595nm=2 into a microcen-

trifuge tube.
2. Add 690L of the appropriate reporter buffer to each microfuge tube.
3. To each tube, add 100L of lysozyme (5mg mL1 in 10mM phosphate buffer
at pH7.8) giving a final concentration of 0.5mg mL1. Invert several times
and leave for five minutes at room temperature. Add 15L of 0.05M EDTA,
pH8.0. Invert several times and leave for 15 minutes.
4. Add 10L of 1% (w/v) SDS (i.e. final concentration 0.01% (w/v)). Invert sev-
eral times.
5. Equilibrate samples at 30C for at least five minutes.
Method 3 (chloroform and SDS method)
1. Dispense 200L of each re-suspended culture at OD595nm= 2 into a microcen-

trifuge tube.
2. Add 800L of the appropriate reporter enzyme buffer to each microfuge tube.
3. Add 20L of chloroform and 10L of 10% (w/v) SDS and vortex.
12.5.1.2Substrate addition to start the reaction
Reagents required:
GusA substrate: p-Nitrophenyl -D-glucuronide (PNPG) 35 mg mL1 pre-

pared in sterile RODI water
LacZ substrate: o-Nitrophenyl -D-galactoside (ONPG) 35mg mL1 prepared

in sterile RODI water
PhoA substrate: p-Nitrophenylphosphate (PNPP) 35 mg mL1 prepared in

100mM Tris (pH8.0).
258
Assay method in microtitre plates
1. Equilibrate samples at assay temperature (28C or 37C) for five minutes.
Both temperatures have been used in the literature. The enzymes: PhoA, GusA
and LacZ from transcriptional fusions are stable at 37C. However, translation-
al fusions to rhizobial proteins may not necessarily be stable at 37C.
2. Transfer 200L from each microfuge tube into a well of a microtitre plate.
If possible, consider placing the samples in order so that cultures expressing

the least GUS are positioned in the beginning on plate.
3. Add 5L substrate using a multichannel pipette.
4. Position plate in the microplate reader and start reading at 405nm. Read sam-
ples every two minutes with prior shaking over a total of 1.5hours.
Assay method in cuvettes
1. Equilibrate samples at assay temperature (28 C or 37C) for five minutes.
Both temperatures have been used in the literature. The enzymes: alkaline
phosphatase, - glucuronidase and -glactosidase from transcriptional fu-
sions are stable at 37C. However, translational fusions to rhizobial proteins
may not necessarily be stable at 37C.
2. Add 10L of substrate (stagger time of addition for each sample) and monitor
colour development.
3. When the samples begin to turn yellow, transfer 200L of the reaction mix-
ture to a new tube and stop the reaction by adding 800L of 125mM K2HPO4
for PhoA assays or 800L of 0.4M Na2CO3 for lacZ and gusA assays. Record
the assay time for each sample.
Measurement of assay culture OD
1. Add 40L of normalised culture (OD595nm =2.0, cells from 12.5.1. section 3) to
160 L saline into the well of a microtitre plate. Add 200 L saline into 2 wells
as a control.
Perform samples at least in triplicate.
2. Position plate in reader and start OD protocol using wavelength set to 595nm.
A 600nm filter could be used as a substitute.
3. Finally, subtract OD595nm reading of saline from OD595nm readings of cultures

determined in the microtitre plate reader. Multiply these readings by five to
provide the cell OD595nm reading.
NB: OD595nm of cells (from Section 12.5.1, Step 3) can be measured in cuvettes us-
ing standard spectrophotometric technique.
259
Calculating PNP or ONP molar extinction coefficient (EM )
Generate a standard curve by making two-fold serial dilutions from a stock of

1.0mM p-nitrophenol or o-nitrophenol prepared in reporter assay buffer. For the
microtitre method, add 205L of each dilution to microtitre wells and read ab-
sorbances. For the cuvette method, add 200 L of each dilution and 800 L of
125mM K2HPO4 for PhoA assay or 800L of 0.40M Na2CO3 for LacZ or GusA
assays. Do a linear regression of OD405nm versus concentration (molL1). The slope
of the line is the molar extinction coefficient. Absorbance is linear with concentra-
tion up to an OD405nm of approximately 2.0.
NB: the range of molar extinction coefficients will vary depending on the condi-
tions used. For this reason, we recommend determining it empirically according
to the method above. The molar extinction coefficients for PNP determined in a
cuvette method at pH7.0 and pH9.0 falls in the ranges of 4,800 to 5,820 (mean=
5,165 RSD 8.4%) and 14,79018,180 (16,640, RSD 8.4%) M1 cm1 respectively
(http://web.viu.ca/krogh/chem331/PNP extinctions.PDF).
Calculation of specific activity of enzymes in microtitre plate method
Export data from microplate reader software into Excel spreadsheet and use the
following calculation to determine specific activity.
i.e.
Enzyme activity is expressed as nmol PNP or ONP min1 mL1
Calculation of specific activity of enzymes in cuvette method
i.e.
Enzyme activity is expressed as nmol PNP or ONP min1 mL1
260
12.5.2Qualitative determination
Reporter genes provide a useful tool to determine if a particular gene is differ-
entially expressed under specific conditions. Qualitative determination on agar
plates can reveal if a gene is up- or down-regulated. Solid growth media are sup-
plemented with a water-insoluble chromogen which is converted into a coloured
product by the activity of the reporter enzyme. The production of colour is pro-
portional to the amount of reporter enzyme produced. The insoluble nature of
the pigment prevents it from spreading across the plate, maintaining it in the vi-
cinity of the colony. Commonly used chromogens include X-Gal, X-Glc and X-
Phosphate for reporter genes lacZ, gusA and phoA, respectively. In the presence
of the reporter enzyme, an appropriate substrate is cleaved, releasing X (5-bromo-
4-chloro-3-indole) which is a blue-coloured compound. The application of this
qualitative method using the gusA reporter has been described previously in Pro-
tocol 12.1, Step 10.
12.5.2.1Protocol for genomic libraries
The creation of genomic libraries from RNB can be particularly important for a
number of reasons, including complementation, sequencing and expression stud-
ies. This section provides a detailed protocol for the construction of a DNA library
but does not discuss the many types of vectors that can be chosen for this purpose.
The reader is referred to Table 12.1 for a table of vectors which may be appropriate
for the construction of a particular DNA library better suited to your specific ex-
periment. The protocol provided here illustrates one specific example for the con-
struction of a library suitable for functional studies, and which can be sequenced
for genome compilation efforts. The vector pTH1522 (Figure12.2) was used in
this example. Vector pTH1522 is a mobilisable plasmid that is unable to replicate
in RNB but which can be converted to a replicative cointegrant when required to
enable expression to be monitored in different genetic backgrounds (Cowie et al.
2006). This protocol has been successfully used to generate libraries of S. medicae
WSM419 (Reeve et al. 2010a) and R. leguminosarum bv. trifolii strains WSM1325
(Reeve et al. 2010b) and WSM2304 (Reeve et al. 2010c) for genome sequence ef-
forts.
261
Figure12.2 The features of the vector pTH1522 and its replicative cointegrate
262
Precautionary measures
All pipette tips should be filter barrier tips.

Benches should be scrupulously clean (wiped with 70% (v/v) ethanol before
use) and gloves must be worn at all times to prevent contamination.
Containers should be disposable plastic ware. If glassware is used, it needs

to be soaked in 0.1N HCl and then 0.1N NaOH prior to rinsing with sterile
RODI water.
Genomic DNA should be prepared using wide-bore pipette tips to prevent

shearing.
Procedure
a. Vector preparation
1. CsCl purified pTH1522 (50 g) should be digested to completion with

XhoI (typically 5Ug1 DNA for two hours). The XhoI digest should then
be heat inactivated at 65C for 15 minutes.
2. The cut plasmid should then be purified using a Qiagen minelute column
and eluted in sterile, molecular-grade water.
3. The purified cut plasmid should then be half end-filled using dTTP and
dCTP, and the Klenow fragment of DNA polymerase (1Ug1 DNA) in
1 polymerase buffer. The reaction should be incubated at 37C for 30
minutes. Following this, the polymerase should then be heat inactivated
by incubating the mixture at 75C for 15 minutes. The cut half end-filled
plasmid can then be purified on a Qiagen minelute column and eluted with
sterile, molecular-grade water.
4. The XhoI half end-filled plasmid needs to be dephosphorylated using Ant-

arctic shrimp alkaline phosphatase (SAP or equivalent) (0.5Ug1) in 1
phosphatase buffer at 37C for 20 minutes. The SAP should be heat inac-
tivated by incubating the reaction at 65C for 5 minutes. The dephospho-
rylated plasmid should be purified using a Qiagen minelute column and
eluted in sterile, molecular-grade water.
5. The purified dephosphorylated XhoI digested half end-filled DNA should

then be incubated overnight at room temperature in 1 T4 DNA ligase
buffer containing T4 DNA ligase (3Ug1). The reaction mixture is then
electrophoresed on a 1% (w/v) agarose gel and the 10.5kb linear non-reli-
gated pTH1522 excised from the gel and purified using a Qiagen minelute
column. The DNA should be eluted in water.
263
b. Genomic DNA preparation
1. DNA can be extracted from a three-day-old culture of RNB using a suit-

able genomic DNA isolation method, such as the CTAB method (Protocol
11.1.1, Chapter11). Choose the appropriate method for your strain to ob-
tain high molecular weight DNA that provides intact DNA (no smearing
on a gel after electrophoresis).
2. Re-suspend the genomic DNA in sterile, molecular-grade water and treat

with RNAse (10gmL1). The genomic DNA concentration can then be es-
timated by UV spectrophotometry ensuring a 260nm/280nm absorbance
ratio of 1.7 to 2.0.
c. Preparation of insert DNA
1. Partially digest RNB genomic DNA with Sau3AI (at 0.05Ug1) in a 50L
reaction and terminate by the addition of 1L 0.5M EDTA and gel load-
ing buffer. A sample needs to be electrophoresed and checked on a gel for
partial digestion. The approximately 2kb fragments can then be excised
from the gel and purified using a Qiagen minelute column and then elut-
ed in sterile, molecular-grade water. (Expected yield: 100ng from 150g
genomic DNA.)
2. Half end-fill the 2kb purified genomic fragments using dGTP and dATP
and the Klenow fragment of DNA polymerase (1Ug1 DNA) in 1 poly-
merase buffer by incubating the mixture at 37C for 30 minutes. The poly-
merase should be heat inactivated by incubating the mixture at 75C for 15
minutes. The DNA fragments can then be purified using a Qiagen minelute
column and eluted in sterile, molecular-grade water.
3. Incubate the half end-filled 2kb genomic partial Sau3AI DNA fragments
overnight at room temperature in 1 T4 DNA ligase buffer in the pres-
ence of T4 DNA ligase. Electrophorese the ligation mixture on a 1% (w/v)
agarose gel and excise the 2kb linear genomic DNA fragments from the
gel. Purify the DNA from the gel using a Qiagen minelute column and re-
suspend in sterile, molecular-grade water. (Expected yield: 100 ng from
150g genomic DNA.)
d. Ligation
The dephosphorylated XhoI digested, half end-filled, religated, purified 10.5kb

linear pTH1522 should be ligated to partial Sau3AI half end-filled religated
purified 2kb linear genomic DNA fragments overnight at room temperature.
The ligation reaction should contain 90 ng pTH1522 and 10 ng insert. Ap-
propriate controls need to be included, such as religation of vector only and a
control containing the religation of insert.
264
e. Transformation
Dilute the ligated DNA (100 ng in 20 L) in sterile, molecular-grade water

(1:3) and transform 1 L into 40 L electrocompetent DH10B. (A suitable
supplier of eletrocompetent DH10B is Invitrogen, which supplies cells with an
electroporation efficiency of 1 1010 cfu g1 DNA.) The efficiency of the com-
petent cells can be calculated at the time of the experiment but should be at
least 1 108cfug1 of a vector such as pGEM7Zf(+) DNA. A typical number
of transformants obtained would be in the order of 100 gentamicin resistant
(10gmL1 in AM3 media containing X-Gal at 50g mL1) transformants per
250L of transformation mixture spread (Figure12.3). There should be no
growth of the same number of DH10B cells on AM3 containing gentamicin
(10g mL1).
Figure12.3 A 250L aliquot of

DH10B transformed with 1L of
1:3 diluted ligation mix and plated
on AM3 containing gentamicin
and X-Gal and incubated
overnight at 37C (left panel) or
overnight at 37C and then for two
days at room temperature (right
panel).
NB: blue DH10B colonies contain the inserts and this feature is specific to the
construction of recombinants in pTH1522.
Colour distinction is significantly enhanced by leaving the cells at room tem-

perature until the desired size is obtained and then placing the plates at 4C.
f. Insert verification
The presence or absence of an insert in pTH1522 should be verified by PCR

analysis using the following primers:
gfplac3970*F 5-TTAggACAACTCCAgTgAAAAgTTC-3
gusrfp-4106R 5-ATAAgggACTCCTCATTAAgATAAC-3.
The PCR reaction mixture of 25L should contain:
1L of the cell suspension

0.5L of 5UmL1 Taq DNA polymerase (Invitrogen Life Technologies)
0.5L of 50M gfplac3970*F primer
265
0.5L of 50M gusrfp-4106R primer

5L of 5 polymerisation buffer1
15L of sterile, molecular-grade water
2.5L of 10mM MgCl2 (Promega Corp.).
Cycling conditions:
four minutes at 95C;

30 cycles of:
30 seconds at 96C
15 seconds at 58C
two minutes at 70C;
final hold at 15C.
1
67mM Tris.HCl, pH8.8 at 25C; 16.6mM (NH4)2SO4, 0.45% (w/v) Triton X-100, 0.2mgmL1
gelatin, 0.2mM dNTPs (Fisher Biotech, Australia).
Amplified PCR-products should be visualised after electrophoresis on a 1%

(w/v) agarose gel following staining with ethidium bromide (0.5gmL1). The
bands can be visualised after staining using a UV transilluminator connected
to a digital camera. Gel images can be stored using appropriate software, such
as Quantity One (BIORAD). An appropriate DNA ladder should be used as a
reference. One suitable marker is the 1kb Promega DNA ladder.
A range of white and blue colonies are chosen and the recombinant rate can
then be calculated. Typically, around 50 colonies are chosen. The following
gel (Figure 12.4) shows the PCR product profile for five white and 44 blue
colonies selected from AM3 containing gentamicin (10gmL1) and X-Gal
(50gmL1).
1 10 20 30 40 50
Figure12.4 Analysis of E. coli DH10B white (lanes 9, 15, 19, 24, 33, 46, 47) and blue (lanes 28,
1014, 1618, 2023, 2532, 3445, 4950) transformants from a pTH1522 library of S. medicae
WSM419. Lane 48 contains the 100bp amplification product from pTH1522. Lanes 1 and 51 are DNA
size ladders.
266
12.6 Protocol for nucleic acid sequencing
NB: in this example, the percentage of clones that contained an insert was ap-
proximately 92%. As expected, the white colonies did not contain an insert and
are typically deletion derivatives of the parent plasmid.
g. Potential library size
A greater than 10-fold coverage library can easily be obtained from the ligated
mixture. The fold representation can be calculated from the following:
number of transformants in total [insert size (kb)/genome size (kb)]
DNA sequencing involves determining the sequential order of the nucleotide bas-
es (adenine, guanine, cytosine and thymine). Variations of the original chain-ter-
mination method (Sanger method) using radioactively- or fluorescently- labelled
dideoxynucleotide triphosphates (ddNTPs) for detection in automated sequenc-
ing machines are the most popular choices of low-throughput DNA sequencing
methods. DNA sequencers now use capillary electrophoresis for size separation,
detection and recording of dye fluorescence. (Each of the four ddNTP terminators
is labelled with a fluorescent dye [Smith et al. 1986]). These automated sequenc-
ers (e.g. Genetic Analyzer, Applied Biosystems, Life Technologies Corp., US) can
sequence up to 384 DNA samples in a batch, with software programs accurately
calling approximately 700900bp prior to the deterioration of quality of base call-
ing. Due to this limitation, two priming sites in opposite orientations (one in 5-3
and one in 3-5) are initially selected, and upon analysis of the sequencing results,
another two priming sites in opposite orientation are again selected to achieve
accurate double-stranded DNA sequence information of a prokaryotic gene. This
approach is, for example, recommended prior to submitting a 16s rRNA gene se-
quence to Genbank (NCBI) which is a public sequence database. Developments
in high-throughput sequencing technologies now produce thousands or millions
of sequences at once, enabling researchers to sequence whole plasmids and ge-
nomes. The major sequencing centres and their sequencing platforms are pro-
vided in Table12.2.
There are three major international sequence databases that are publicly available:
GenBank in the U.S. (http://www.ncbi.nlm.nih.gov/); EMBL in Europe (http://
www.ebi.ac.uk/); and DDBJ in Japan (http://www.ddbj.nig.ac.jp/). Each day, all
newly submitted sequences are exchanged between the three databases. Every sci-
entific publication in which bacterial strains are identified or described requires the
author to deposit the gene sequences in one of these three public databases. There-
fore, the databases contain the gene sequence of nearly all sequenced bacterial spe-
cies and enable researchers to compare their own sequences with those in the da-
tabases for identification. For example, the nucleic acid sequence of your gene may
be submitted to the National Centre for Biotechnology Information (NCBI) using
the web-based submission tool, BankIt, or the off-line submission tool, Sequin.
267
Information comparing the two options is provided online: http://www.ncbi.nlm.

nih.gov/WebSub/?tool=genbank. Submission of sequences to NCBI relies on the
researcher providing both metadata and evidence of sequence verification; these
requirements should be considered prior to initiating a sequencing project (http://
www.ncbi.nlm.nih.gov/WebSub/html/requirements.html).
Table12.2 M
ajor sequencing centres used by the authors for either routine sequencing of DNA
fragments, sequencing of large inserts, plasmids or whole genomes using a shotgun
approach or high-throughput sequencing of whole genomes and/or transcriptomes
using next generation sequencing technologies
Sequencing centre Location Access information at Sequencing platforms available
as of May 2013
Joint Genome California, USA http://www.jgi.doe.gov/ Roche 454, Illumina GAII and
Institute PacBio
Macrogen Seoul, Korea http://www.macrogen.com/eng/ ABI 3730XL and ABI3700
sequencing/sequence_main.jsp
Australian Genome Australia http://www.agrf.org.au/ AB 3730XL, Roche GS FLX and
Research Facility (multiple Illumina GAII
(AGRF) locations)
BGI Genomics Beijing, China http://en.genomics.cn/ Illumina HiSeq, AB SOLiD and Ion
navigation/index.action Torrent
12.7 Protocol for bioinformatic tools

Comparative genomic tools enable researchers to interrogate the genome of new-
ly sequenced RNB by comparison with genomes of other organisms. Here we have
provided a brief synopsis of two genomic tools used by the authors which are
available for free online.
12.7.1Joint Genome Institute (JGI)

The Integrated Microbial GenomesGenome Encyclopedia of Bacteria and Ar-
chaea Genomes (IMG/GEBA) serves as the preliminary release of genomes from
the United States Department of Energy GEBA project as soon as they are submit-
ted to Genbank. The number of genomes available is constantly growing but as of
August 2015 there were 26,888 bacterial genomes available with 934 Rhizobiales
genomes.
Access at http://img.jgi.doe.gov/cgi-bin/geba/main.cgi
IMG Manual available at http://img.jgi.doe.gov/geba/doc/userGuide.pdf
Tools
Neighbourhood matching/genome comparisons.

Basic Local Alignment Search Tool (BLAST), whereby a nucleotide or amino
acid sequence is submitted and returns sequences with regions of similarity.
268
Gene searchfinds genes using a variety of search criteria. (The gene neigh-
bourhood display is very useful for visualising homologs.)
Function alignmentsearch for COG, KOG or Pfam.

Genome statisticsi.e. number of genes, number of paralogs, number of plas-
mids, size etc.
Chromosome mapscircular view of plasmids coloured according to COG

function.
Synteny viewcompares scaffolds (VISTA), plasmids or whole genomes (Ar-

temis or Dot Plot) based on nucleotide or amino acid identity (Figure12.5).
Figure12.5 Synteny between the nucleotide genome sequences of Ensifer (Sinorhizobium) meliloti
strain 1021 and Ensifer (Sinorhizobium) medicae strain WSM419 using Dot Plot analysis (IMG, JGI).
Abundance profilevisual display of the abundance of all protein families

(COGs or Pfams) or functional families (enzymes).
Functional profilethe number of genes in an organism within a functional

group and link to associated genes.
Genome clusteringgenomes can be clustered based on protein/functional

family.
COG category statisticsstatistics for genomes by COGs (cluster of ortholo-

gous genes) in tabular format, pie chart or bar graph.
269
KEGG category statisticsstatistics for genomes by KEGG (Kyoto Encyclope-

dia of Genes and Genomes) category and link to KEGG pathway maps.
NB: the KEGG database can be accessed directly from http://www.genome.jp/

kegg/. Useful search tools include:
KEGG pathwaysearch for a metabolic pathway map and view all contribut-
ing enzymes via E.C. (Enzyme Commission) number.
KEGG orthology (KO)known functions of genes and proteins (using ex-

perimental evidence) are organised in the KO database
KEGG organismseach KO entry can be taxonomically mapped, enabling

identification of the presence/absence of gene orthologs.
12.7.2NCBI access at http://www.ncbi.nlm.nih.gov/

Access Microbial Genomes at http://www.ncbi.nlm.nih.gov/genomes/lproks.cgi;
8,055 bacterial genomes are available with 2,258 proteobacteria and 688 alpha-
proteobacteria.
Tools
BLAST. Key criteria used to select BLAST type are given in Table12.3.
Table12.3 Key criteria used to select BLAST type

Query Genome BLAST
nucleic acid nucleic acid BLASTN or TBLASTX
Protein nucleic acid TBLASTN
nucleic acid protein BLASTX
Protein protein BLASTP
NCBI access to BLAST with Microbial Genomes http://www.ncbi.nlm.nih.

gov/sutils/genom_table.cgi
NCBI access to concise microbial protein BLAST. (Proteins have been clus-
tered at the genus level.) http://www.ncbi.nlm.nih.gov/genomes/prokhits.cgi
CD Treeanalyse conserved domain data (CDD).

COBALTprotein multiple alignment.
Cn3Dviewing 3D protein structures.
Gene Plotpairwise comparison of two prokaryotic genomes displaying pairs
of protein homologs.
Genetic Codesgenetic code for organism.
270
Bacterial genetic code below accessed at: http://www.ncbi.nlm.nih.gov/Taxono-

my/taxonomyhome.html/index.cgi?chapter=tgencodes#SG11. Initiation of trans-
lation is most efficient at AUG but also occurs at GUG and UUG in bacteria.
TTT F Phe TCT S Ser TAT Y Tyr TGT C Cys

TTC F Phe TCC S Ser TAC Y Tyr TGC C Cys
TTA L Leu TCA S Ser TAA * Ter TGA * Ter
TTG L Leu TCG S Ser TAG * Ter TGG W Trp
CTT L Leu CCT P Pro CAT H His CGT R Arg
CTC L Leu CCC P Pro CAC H His CGC R Arg
CTA L Leu CCA P Pro CAA Q Gln CGA R Arg
CTG L Leu CCG P Pro CAG Q Gln CGG R Arg
ATT I Ile ACT T Thr AAT N Asn AGT S Ser
ATC I Ile ACC T Thr AAC N Asn AGC S Ser
ATA I Ile ACA T Thr AAA K Lys AGA R Arg
ATG M Met ACG T Thr AAG K Lys AGG R Arg
GTT V Val GCT A Ala GAT D Asp GGT G Gly
GTC V Val GCC A Ala GAC D Asp GGC G Gly
GTA V Val GCA A Ala GAA E Glu GGA G Gly
GTG V Val GCG A Ala GAG E Glu GGG G Gly
Genome ProtMapmaps each protein from a Cluster of Orthologous Genes

(COG) back to its genome.
Map Viewerview and search an organisms complete genome and display

maps.
12.8 Protocol for phylogenetic trees

Phylogenetic analysis was once considered too difficult and therefore not widely
utilised by molecular and biochemical biologists. This changed with the arrival
of DNA sequencing and accessible software programs (Hall 2011). Building phy-
logenetic trees is now a four-step process. First, you need to identify and acquire
the sequences to include in the tree. Second, you need to align the sequences and
estimate the tree by one of the given methods. Finally, you draw the tree and edit
according to the journals requirements.
Several free software programs exist to build phylogenetic trees. Here we dis-
cuss in detail how to use MEGA (http://www.megasoftware.net/) but alternative
software can be PHYML (http://www.atgc-montpellier.fr/phyml/) or MrBayes
(http://mrbayes.sourceforge.net/index.php). A short overview of the most com-
monly used phylogenetic tree types is given below.
Neighbor-Joining (NJ) is currently the most widely used tree type of the dis-
tance methods. It produces a single, strictly bifurcate tree, which means that
each internal node has exactly two branches descending from it. In distance
methods, the fraction of sites that differ between two sequences in a multiple
271
alignment are expressed as the distances between them. Another example of

a distance method is UPGMA (Unweighted Pair-Group Method with Arith-
metic Mean). Due to the fact that the tree derived from this method is ultra-
metric (all taxa are equally distant from the root) it is advisable not to use this
for publications.
Maximum Likelihood (ML) is a method that searches for the tree that makes
the data most likely. It uses the log-likelihood to compare the various models
of nucleotide substitution for a particular dataset. The ML program seeks the
tree with the largest log-likelihood. ML trees are gaining popularity since they
are based on more solid calculations, and the latest software makes them more
easily accessible.
For further assistance with phylogenetic trees and more background theory, visit:
http://www.sinauer.com/hall/3e/
12.8.1Steps to build a phylogenetic tree in MEGA

In the MEGA main window go to Align and choose Do BLAST search (Fig-
ure12.6). A new window opens (Figure12.7) where you can paste your sequence
in the sequence box or upload a file. Click the BLAST button and a summary win-
dow will appear (Figures 12.8 and 12.9).
Figure12.6 Undertake a BLAST search
272
Figure12.7 NCBI Standard Nucleotide BLAST showing where to add query

sequence
Figure12.8 NCBI BLAST results summary window (graphic summary)
273
Figure12.9 NCBI BLAST summary results window (results summary)
After checking the results, select the sequences you want to import into MEGA by
clicking the Accession link and click the Add to alignment button (Figure12.10).
In this way, you can select sequences one by one. The program will ask you how to
name your entry (Figure12.11); ensure that you add the accession number under
the sequence label. The accession number is the only part that makes your entry
unique and traceable. If you already have the required sequences in a FASTA file
you just open this file in the MEGA program ( MEGA main window File
Open a file/session and select the correct file).
Figure12.10 NCBI example of gene sequence information
274
Figure12.11 MEGA entry of

sequence label through link with
NCBI
To build a tree, we also want to include reference sequences (sequences published

in the IJSEM journal or on the approved list in IJSEM, see http://ijs.sgmjournals.
org/). An easier way to access reference sequences is to visit StrainInfo (http://
www.straininfo.net/). Type in the genus and the browser gives you a list of all
currently described species. An alternative is the List of Prokaryotic names with
standing in Nomenclature (http://www.bacterio.cict.fr/). This site also provides
currently described species and the accession numbers as given in the new species
description paper. After adding the appropriate sequences to MEGA we can start
to align them. Save the sequence list as a .mas file (Figure12.12). Go to data and
click on save session.
Figure12.12 Saving reference and query sequences in MEGA as a .mas file
275
The next step is to align your sequences; a process designed to introduce gaps into
the sequences to shift the bases back to their corresponding homologous posi-
tions. A phylogenetic tree is only as good as the quality of the alignment, which
makes this step crucial for building trees. Two algorithms are available in MEGA,
including ClustalW and Muscle. Here we will choose align by ClustalW (Fig-
ure12.13), although Muscle is just as powerful. A dialogue box will appear and
remind you that nothing is selected, click OK to select all sequences. A ClustalW
parameters window will appear, click OK and the program will start aligning your
sequences (Figure12.14).
Figure12.13 ClustalW alignment in MEGA
Figure12.14 ClustalW Parameters in MEGA
276
The default values are optimal for aligning DNA sequences. When the program is
finished calculating the alignment, you have to check whether the sequences are
aligned correctly and delete those that are too short (Figure12.15). Subsequently,
the beginning and the end of the alignment must be trimmed (deleted) so that the
tree is based on the actual similarity of the sequences (Figure12.16). When you
are satisfied with the alignment, you need to export it as a .meg file as shown in
Figure12.17.
Figure12.15 Arrows indicate short sequences in MEGA after ClustalW alignment
Figure12.16 Trimmed sequences in MEGA
277
Figure12.17 Export
alignment as a .meg file
To build the actual tree, you have to go back to the home window of MEGA (Fig-
ure12.18). Here you can choose between several trees; currently Neighbor-joining
and Maximum Likelihood are the most frequently used methods. The Maximum
Likelihood method is discussed in Section 12.8. When clicking on the construct
ML tree button, you will have to search for your .meg file. An analysis window
opens and here you can define if you want bootstrap analysis, the number of rep-
lications (which is mostly 500 or 1,000), which model to use (generally General
Time Reversal model is used), gaps treatment (mostly complete deletion) and ML
heuristic method (mostly CNI). The characteristics given in Figure 12.19 work
well for general ML trees; more information about these characteristics is avail-
able (Hall 2011).
Figure12.18 Constructing a phylogenetic tree using Maximum Likelihood
278
Figure12.19 MEGA analysis

preferences for constructing a
phylogenetic tree
Select calculate to begin building the tree. When the program is finished, a new
window will open showing your ML tree. To set the root, go to Subtree and select
the root button, then select the preferred root branch (Figure12.20). When you
are satisfied with the tree, you can export it as a pdf, click Image and save as a pdf
file (Figure12.21). Additionally, you can use graphic analysis software (i.e. Adobe
Photoshop or Adobe Illustrator) to conform your tree to the journals specific re-
quirements.
Figure12.20 Setting the root

in MEGA TreeExplorer
279
12.9References
Figure12.21 Saving the phylogenetic tree output
12.9References
Cowie A., Cheng J., Sibley C.D., Fong Y., Zaheer R., Patten C.L., Morton R.M., Golding G.B. and
Finan T.M. 2006. An integrated approach to functional genomics: construction of a novel re-
porter gene fusion library for Sinorhizobium meliloti. Applied and Environmental Microbiology
72, 71567167.
De Feyter R., Yang Y. and Gabriel D.W. 1993. Gene-for-genes interactions between cotton R genes
and Xanthomonas campestris pv. malvacearum avr genes. Molecular Plant Microbe Interactions
6, 225237.
de Lorenzo V., Herrero M., Jakubzik U. and Timmis K.N. 1990. Mini-Tn5 transposon derivatives for
insertion mutagenesis, promoter probing, and chromosomal insertion of cloned DNA in gram-
negative eubacteria. Journal of Bacteriology 172, 65686572.
De Vos G.F., Walker G.C. and Signer E.R. 1986. Genetic manipulations in Rhizobium meliloti utiliz-
ing two new transposon Tn5 derivatives. Molecular and General Genetics 204, 485491.
Defeyter R. and Gabriel D. 1991. At least six avirulence genes Are clustered on a 90-Kilobase plasmid
in Xanthomonas campestris pv. malvacearum. Molecular Plant Microbe Interactions 4, 423432.
DeFeyter R., Kado C.I. and Gabriel D.W. 1990. Small, stable shuttle vectors for use in Xanthomonas.
Gene 88, 6572.
Dennis J.J. and Zylstra G.J. 1998. Plasposons: modular self-cloning minitransposon derivatives for
rapid genetic analysis of gram-negative bacterial genomes. Applied and Environmental Micro-
biology 64, 27102715.
Figurski D.H. and Helinski D.R. 1979. Replication of an origin-containing derivative of plasmid
RK2 dependent on a plasmid function provided in trans. Proceedings of the National Academy
of Sciences USA 76, 16481652.
Figurski D.H., Meyer R.J. and Helinski D.R. 1979. Suppression of Co1E1 replication properties by
the Inc P-1 plasmid RK2 in hybrid plasmids constructed in vitro. Journal of Molecular Biology
133, 295318.
280
12.9References
Finan T.M., Kunkel B., De Vos G.F. and Signer E.R. 1986. Second symbiotic megaplasmid in Rhizo-
bium meliloti carrying exopolysaccharide and thiamine synthesis genes. Journal of Bacteriology
167, 6672.
Hall B.G. 2011. Phylogenetic trees made easy: A how-to manual. Sinauer Associates Inc.
Herrero M., de Lorenzo V. and Timmis K.N. 1990. Transposon vectors containing non-antibiotic
resistance selection markers for cloning and stable chromosomal insertion of foreign genes in
gram-negative bacteria. Journal of Bacteriology 172, 65576567.
Karunakaran R., Mauchline T.H., Hosie A.H. and Poole P.S. 2005. A family of promoter probe vec-
tors incorporating autofluorescent and chromogenic reporter proteins for studying gene expres-
sion in Gram-negative bacteria. Microbiology 151, 32493256.
Kim C.H., Helinski D.R. and Ditta G. 1986. Overlapping transcription of the nifA regulatory gene in
Rhizobium meliloti. Gene 50, 141148.
Kokotek W. and Lotz W. 1991. Construction of a mobilizable cloning vector for site-directed mu-
tagenesis of gram-negative bacteria: application to Rhizobium leguminosarum. Gene 98, 713.
Kovach M.E., Elzer P.H., Hill D.S., Robertson G.T., Farris M.A., Roop R.M. 2nd and Peterson K.M.
1995. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying differ-
ent antibiotic-resistance cassettes. Gene 166, 175176.
Kovach M.E., Phillips R.W., Elzer P.H., Roop R.M. 2nd and Peterson K.M. 1994. pBBR1MCS: a
broad-host-range cloning vector. Biotechniques 16, 800802.
Legocki R.P., Legocki M., Baldwin T.O. and Szalay A.A. 1986. Bioluminescence in soybean root nod-
ules: Demonstration of a general approach to assay gene expression in vivo by using bacterial
luciferase. Proceedings of the National Academy of Sciences USA 83, 90809084.
Long S.R. 1989. Rhizobium genetics. Annual Reviews in Genetics 23, 483506.
Metcalf W.W., Jiang W., Daniels L.L., Kim S.K., Haldimann A. and Wanner B.L. 1996. Conditionally
replicative and conjugative plasmids carrying lacZ alpha for cloning, mutagenesis, and allele
replacement in bacteria. Plasmid 35, 113.
Muller P. 2004. Use of the multipurpose transposon Tn KPK2 for the mutational analysis of chro-
mosomal regions upstream and downstream of the sipF gene in Bradyrhizobium japonicum.
Molecular Genetics and Genomics 271, 359366.
Pobigaylo N., Wetter D., Szymczak S., Schiller U., Kurtz S., Meyer F., Nattkemper T.W. and Becker
A. 2006. Construction of a large signature-tagged mini-Tn5 transposon library and its applica-
tion to mutagenesis of Sinorhizobium meliloti. Applied and Environmental Microbiology 72,
43294337.
Quandt J., Clark R.G., Venter A.P., Clark S.R., Twelker S. and Hynes M.F. 2004. Modified RP4 and
Tn5-Mob derivatives for facilitated manipulation of large plasmids in Gram-negative bacteria.
Plasmid 52, 112.
Quandt J. and Hynes M.F. 1993. Versatile suicide vectors which allow direct selection for gene re-
placement in gram-negative bacteria. Gene 127, 1521.
Reeve W., Chain P., OHara G., Ardley J., Nandesena K., Brau L., Tiwari R., Malfatti S., Kiss H.,
Lapidus A., Copeland A., Nolan M., Land M., Hauser L., Chang Y.J., Ivanova N., Mavromatis K.,
Markowitz V., Kyrpides N., Gollagher M., Yates R., Dilworth M. and Howieson J. 2010a. Com-
plete genome sequence of the Medicago microsymbiont Ensifer (Sinorhizobium) medicae strain
WSM419. Standards in Genomic Sciences 2, 7786.
Reeve W., OHara G., Chain P., Ardley J., Brau L., Nandesena K., Tiwari R., Copeland A., Nolan M.,
Han C., Brettin T., Land M., Ovchinikova G., Ivanova N., Mavromatis K., Markowitz V., Kyrpi-
des N., Melino V., Denton M., Yates R. and Howieson J. 2010b. Complete genome sequence of
Rhizobium leguminosarum bv. trifolii strain WSM1325, an effective microsymbiont of annual
Mediterranean clovers. Standards in Genomic Sciences 2, 347356.
Reeve W., OHara G., Chain P., Ardley J., Brau L., Nandesena K., Tiwari R., Malfatti S., Kiss H.,
Lapidus A., Copeland A., Nolan M., Land M., Ivanova N., Mavromatis K., Markowitz V., Kyrpi-
des N., Melino V., Denton M., Yates R. and Howieson J. 2010c. Complete genome sequence of
Rhizobium leguminosarum bv trifolii strain WSM2304, an effective microsymbiont of the South
American clover Trifolium polymorphum. Standards in Genomic Sciences 2, 6676.
Reeve W.G., Tiwari R.P., Worsley P.S., Dilworth M.J., Glenn A.R. and Howieson J.G. 1999. Con-
structs for insertional mutagenesis, transcriptional signal localization and gene regulation stud-
ies in root nodule and other bacteria. Microbiology 145, 13071316.
Selbitschka W., Niemann S. and Phler A. 1993. Construction of gene replacement vectors for
Gram- bacteria using a genetically modified sacRB gene as a positive selection marker. Applied
Microbiology and Biotechnology 38, 615618.
Selvaraj G. and Iyer V.N. 1983. Suicide plasmid vehicles for insertion mutagenesis in Rhizobium
meliloti and related bacteria. Journal of Bacteriology 156, 12921300.
281
12.9References
Sessitsch A., Wilson K.J., Akkermans A.D. and de Vos W.M. 1996. Simultaneous detection of dif-
ferent Rhizobium strains marked with either the Escherichia coli gusA gene or the Pyrococcus
furiosus celB gene. Applied and Environmental Microbiology 62, 41914194.
Simon R. 1984. High frequency mobilization of gram-negative bacterial replicons by the in vitro
constructed Tn5-Mob transposon. Molecular and General Genetics 196, 413420.
Simon R., OConnell M., Labes M. and Puhler A. 1986. Plasmid vectors for the genetic analysis
and manipulation of rhizobia and other gram-negative bacteria. Methods in Enzymology 118,
640659.
Simon R., Quandt J. and Klipp W. 1989. New derivatives of transposon Tn5 suitable for mobilization
of replicons, generation of operon fusions and induction of genes in gram-negative bacteria.
Gene 80, 161169.
Smith L.M., Sanders J.Z., Kaiser R.J., Hughes P., Dodd C., Connell C.R., Heiner C., Kent S.B. and
Hood L.E. 1986. Fluorescence detection in automated DNA sequence analysis. Nature 321,
674679.
Spaink H.P., Okker R.J.H., Wijffelman C.A., Pees E., Lugtenberg B.J.J. 1987. Promoters in the nodu-
lation region of the Rhizobium-leguminosarum Sym Plasmid Prl1jI. Plant Molecular Biology 9,
2739.
Stanley J. and Cervantes E. 1991. Biology and genetics of the broad host range Rhizobium NGR234.
Journal of Applied Bacteriology 70, 919.
Wilson K.J., Sessitsch A., Corbo J.C., Giller K.E., Akkermans A.D. and Jefferson R.A. 1995. be-
ta-Glucuronidase (GUS) transposons for ecological and genetic studies of rhizobia and other
gram-negative bacteria. Microbiology 141, 16911705.
Xi C., Lambrecht M., Vanderleyden J. and Michiels J. 1999. Bi-functional gfp- and gusA-containing
mini-Tn5 transposon derivatives for combined gene expression and bacterial localization stud-
ies. Journal of Microbiological Methods 35, 8592.
Yost C.K., Del Bel K.L., Quandt J. and Hynes M.F. 2004. Rhizobium leguminosarum methyl-accept-
ing chemotaxis protein genes are down-regulated in the pea nodule. Archives of Microbiology
182, 505513.
282
13.1Introduction
CHAPTER13
Methods for isolation of

RNA from rhizobia
P.S. Poole and R. Karunakaran
13.1Introduction
This chapter provides methods for the isolation of RNA from rhizobia under a va-
riety of growth conditions. There are many ways in which this RNA can be used,
including Q-RT-PCR, microarray analysis and for RNA sequencing (RNAseq).
We do not provide detailed protocols for these downstream applications as these
are procedure specific and methods are changing very rapidly. However, isolation
of high-quality RNA is key for all applications, representing the transcriptome
of living cells that in many cases will be actively growing. This may be central in
choosing to isolate RNA as well as, or instead of DNA, which may come from liv-
ing or dead cells. Consider for example whether amplification of the 16S rRNA
gene from a complex microbial community (e.g. soil or rhizosphere) will give the
same community profile when DNA or cDNA derived from RNA is used as the
target nucleic acid. Amplification from DNA will give the community of live and
dead bacteria while cDNA will highlight the living community. However, many
of the RNA procedures are prohibitively expensive. Ask whether you need to use
RNA-based procedures or if DNA-based procedures will suffice. Further, proce-
dures that work routinely in the best equipped molecular biology laboratories
may be difficult to establish when setting up a laboratory from scratch. In spite of
these caveats, the power of modern RNA-based methods, which can lead to a me-
tatranscriptome analysis of the rhizosphere or soil, is remarkable. Choose wisely
and consider your hypothesis carefully.
283
13.2 Isolation of total and small RNA (sRNA) from free-living rhizobia
13.2Isolation of total and small RNA (sRNA) from

free-living rhizobia
In this first protocol, a commercial RNA isolation kit is used to produce very high-
quality RNA suitable for almost all downstream applications. It has been adapted
for use with rhizobia but should work with most bacteria and should be used as
your first choice. However, some rhizobia are very resistant to isolation of RNA
and in the following protocol (13.2) an alternative hot SDS procedure is given.
13.2.1General preparation for RNA isolation

All Gilson tips should be filter barrier tips guaranteed to be RNA free. Pipettes
should be wiped with RNA Zap before working and laid on paper towel.
Benches should be scrupulously clean (wipe with 70% ethanol before use) and
gloves must be worn at all times; you are one of the main sources of RNAases.
As a rule, avoid making solutions; wherever possible buy commercial reagent

solutions guaranteed to be RNAase free.
If it is unavoidable to weigh-out a chemical then spatulas must be dry-oven

baked or wiped with RNA Zap to remove RNAases before use. Make sure the
chemical is weighed into an RNAase-free container (e.g. Falcon tube) and add
RNAase-free water (e.g. Sigma water cat. no. (catalogue number) W4502).
Containers should be disposable plastic ware (such as Falcon tubes) but if

equipment or glassware needs to be used then soak it in 0.1NNaOH with
1mM EDTA before washing with RNAase-free water (e.g. Sigma water cat. no.
W4502). Use sterile Falcon tubes to measure out large volumes and dispose of
them after each use.
Use Amber Eppendorf tubes that are supplied sterile and guaranteed RNAase
free (e.g. Eppendorf 0030 120 191).
The recipes for RNAase inhibitor solution and how to prepare Fast Prep lysing
tubes can be found in Appendices 13.10.1 and 13.10.2.
13.2.2Isolation of total and small RNA (sRNA) from free-

living cells using Qiagen RNeasy Plus Micro kit (Cat No:
74034)
Reagent preparation
Add 10L 2-mercaptoethanol per 1mL Buffer RLT Plus (from kit). Dispense
in a fume hood. Buffer RLT Plus is stable at room temperature (1525C) for
one month after addition of 2-mercaptoethanol.
284
Buffer RLT Plus may form a precipitate upon storage. If necessary, re-dissolve
by warming and then place at room temperature.
Buffer RLT Plus and Buffer RW1 contain a guanidine salt and are therefore not
compatible with disinfecting reagents containing bleach.
Perform all steps of the procedure at room temperature. During the proce-
dure, work quickly.
Perform all centrifugation steps at 2025C in a standard microcentrifuge.

Ensure that the centrifuge does not cool below 20C.
Prepare 10 mM Tris-Cl pH 8.0 (500 L for each RNA preparation) from
1MTris-Cl pH8.0 stock (Sigma RNAase free cat. no. T2694) by diluting with
Sigma (cat. no. W4502) RNAase-free water. Prepare the dilution in a sterile
Falcon tube.
Procedure
1. Grow rhizobia according to your normal laboratory procedure. While the

precise conditions will depend on the experiment, it is important to conduct
preliminary growth experiments. For microarray and RNAseq experiments,
it is quite important to harvest batch-cultured bacteria early in exponential
growth phase to avoid inconsistent gene expression and shut-down of riboso-
mal protein synthesis.
2. Add 24mL of RNAase inhibitor to a 50mL Sorval SS34 centrifuge tube. (The
Sorval tube must have been cleaned and autoclaved from a batch of tubes
kept for RNA work, but the tubes will not be RNAase-free so the addition of
RNAase inhibitor is essential to inactivate RNAases).
3. Add 12mL of cells of 0.40.6OD (600nm), vortex for five seconds and incu-
bate for five minutes at room temperature.
4. Harvest the cells at 10,000rpm at 4C for 10 minutes in the Sorval SS34 rotor.
5. Decant the supernatant and carefully dry the tube on a filter paper.
6. Add 250L of 10mM Tris-HCl pH8.0 and re-suspend the bacteria.
7. Add the re-suspended cells to 350 L of pre-cooled RLT Plus buffer (with
2-mercaptoethanol) in the FastPrep Tube (either use commercial FastPrep lys-
ing matrix B or prepare your own tubes as described above).
8. Lyse the cells in Fast Prep instrument (MP Bio 29525 Fountain Parkway Solon,
OH 44139 United States) at speed 6.5 for 30 seconds. If a Fast Prep is not avail-
able, cells can be lysed with lysozyme at Step 6 modified as follows: add 250L
of 10mM Tris-HCl pH8.0 containing 400g/mL lysozyme. Incubate at room
temperature for 15 minutes. Add 350L of pre-cooled RLT Plus buffer (with
285
2-mercaptoethanol) and proceed to Step 9. However, RNA yields are not as

good with the lysozyme procedure.
9. Incubate the tube on ice for three minutes. Steps 1019 that follow are com-
mon for all RNA isolation procedures.
10. Spin tubes for three minutes at 13000rpm in a microfuge at 4C. Carefully
remove the supernatant by pipetting and transfer 600L to a Qiagen gDNA
Eliminator spin column placed in a 2mL collection tube.
11. Spin for 30 seconds at 10,000rpm in a microfuge. Transfer the flow-through

from the 2 mL collection tube to a sterile 15 mL Falcon tube. Add any re-
maining solution from Step 10 to the gDNA eliminator column and centrifuge
again. Add the flow-through to the 15mL Falcon tube and discard the gDNA
eliminator column. Measure the total flow-through volume.
12. Add one volume (approximately 1500L) of 70% (v/v) ethanol to the flow-
through and mix well by pipetting. Transfer up to 700L of the sample, includ-
ing any precipitate that may have formed, to an RNeasy spin column placed in
a 2mL collection tube. (Columns should be stored in the fridge).
13. Close the lid gently and centrifuge for 15 seconds at 10,000rpm. Discard the
flow-through. Repeat this step, adding 700L of sample from Step 12 until the
entire sample has been added to the column.
14. Add 700L Buffer RW1 to the RNeasy spin column. Close the lid gently and
centrifuge for 15 seconds at 10,000rpm in a microfuge to wash the spin col-
umn membrane. Discard the flow-through.
15. Add 500L Buffer RPE to the RNeasy spin column. Close the lid gently and
centrifuge for 15 seconds at 10,000rpm in a microfuge to wash the spin col-
umn membrane. Discard the flow-through.
16. Add 500L Buffer 80% Ethanol to the RNeasy spin column. Close the lid gen-
tly and centrifuge for two minutes at 10,000rpm in a microfuge to wash the
spin column membrane. Discard the collection tube.
17. Place the column in a new collection tube (provided in the Qiagen kit), open
the lid and centrifuge at 10,000rpm in a microfuge for five minutes.
18. Place the RNeasy spin column in a new 1.5mL collection tube.
19. Add 35L RNase-free water directly to the spin column membrane, Close the
lid gently and centrifuge for one minute at 10,000rpm to elute the RNA.
13.2.3Measuring quantity and integrity of RNA

The RNA prepared above is suitable for use in a variety of applications, includ-
ing Q-RT-PCR, microarray analysis or RNAseq. However, before being used for
any downstream application, it must be quantified and its integrity checked. To
determine the concentration and purity of RNA, obtain A260 and A280 readings in
286
0.1TE Buffer on a spectrophotometer. This should only be attempted for sam-

ples with predicted sufficient yield. For example, for an RNA sample (400500L)
placed in a 1mL cuvette (1cm path length), a concentration of approximately
4g/mL is typically required to obtain A260 readings of 0.1 or higher. It is strongly
recommended that RNA is quantified using the Bio-Rad Experion, Agilent Bio-
analyzer or equivalent system. These systems can also be used to estimate yield
from quantities as small as 10ng. An additional reason for using these systems
is that they will show whether the RNA is intact or degraded (Figure13.1 A, B).
Use the following equation to calculate RNA concentration:
Concentration (g/mL)=A260 DF (dilution factor) 40.
Calculate the A260/A280 ratio to determine RNA purity. A ratio of 2.02.3 is most
desirable and indicates high-quality RNA with little contaminating protein.
A
E coli Control RNA _ RNA control 250 diluted
300
250
200
Fluorescence
150
100 Figure13.1 A An
Experion profile of high-
50
quality RNA from E. coli.
0 Note the very sharp 23S
16S
23S
and 16S rRNA peaks.

Broadening of these
20 25 30 35 40 45 50 55 60 65
Time (seconds) peaks indicates RNA
degradation.
2394 Free-living B
Figure13.1 B An
300 Experion profile of
high-quality RNA from
250
R. leguminosarum.
200 Note the very small 23S
Fluorescence
rRNA peak. Many, but

150 not all, rhizobia have
100 an RNAase processing
site in the middle of the
50 23S peak. Notice there
is a new peak to the left
0
16S
23S
of the 16S peak. It is

actually a doublet of two
20 25 30 35 40 45 50 55 60 65
Time (seconds) 1.3kb peaks formed by
processing of 23S rRNA.
287
13.3 Protocol for isolation of total and small RNA (sRNA) from the rhizosphere of legumes
13.3Protocol for isolation of total and small RNA

(sRNA) from the rhizosphere of legumes
This protocol describes how to set up a rhizosphere colonisation experiment for

small- and large-seeded legumes. It was designed so that microarray or RNAseq
experiments could be conducted on bacteria recovered from the rhizosphere after
a suitable incubation. However, similar procedures can be used to score bacterial
colonisation by culturing bacteria rather than isolating their RNA. Which proto-
col to use will depend on the precise legume and inoculation. The large-seeded
legume protocol is slightly simpler, but either procedure should work for most
legumes if preliminary growth trials are conducted. If RNA recovery is insuffi-
cient, grow the plant for longer, either before or after bacterial inoculation, to en-
able a stronger root system to develop. In these protocols, various legumes are
grown in fine or medium grade vermiculite. The appropriate grade for a particular
legume should be tested in a growth experiment. While vermiculite is used, there
is no fundamental reason why other media, such as sand, perlite or leca could not
be used, although this needs to be tested; i.e. use the best sterile growth medium
for the legume to be grown. However, to isolate bacterial RNA from the rhizos-
phere of plants grown in soil, use soil isolation procedures such as the commercial
Mo Bio Power Soil, which works well.
13.3.1General preparation
Ensure seeds are germinable, surface sterilised and sort them for quality as
described in Chapter5, Section 5.8.
Prepare a nitrogen-free rooting solution, as described in Appendix 13.10.3 or

in Chapter5, Section 5.7.
13.3.2Procedure for large-seeded legumes such as pea

1. Add 10mL of rooting solution to 25mL vermiculite (fine grade, do not pack
tightly) in a 50mL Falcon tube, cap loosely and autoclave.
2. Sow surface sterilised seeds 12 cm below the vermiculite and incubate in

the growth room for seven days (keep the tube open). The caps can be left off,
leaving the Falcon tubes open, but holes can be drilled in the lid to allow the
shoot to grow (see small-seeded legumes). The advantage of capping is that it
reduces contamination with other bacteria.
3. Streak rhizobial strains on TY agar slopes four days after seed germination.
4. On the seventh day after seed germination, re-suspend the slope inoculated
with a strain of rhizobia in 5mL sterile water and wash the cells three times
in sterile water, recovering cells each time by centrifuging at 4,000 rpm for
10 minutes. You may wish to wash in your standard sterile minimal medium
288
rather than water, e.g. JMM (Chapter 3) but omitting nitrogen and carbon
sources from the minimal medium.
5. Measure the OD600 and adjust the cell count to 108 cells mL1 (Chapter 6).
Rhizobium leguminosarum has approximately 109cells/mL at an OD600 of 1.
This should be validated for each laboratory spectrophotometer and strain.
6. Inoculate each seeding with 1mL containing 108 cells. It should be noted that
108 bacteria is a large inoculum. Inocula as low as 103 bacteria can produce
excellent microarray results when peas are harvested seven days post inocula-
tion. However, low inocula will not yield sufficient RNA for successful experi-
ments if plants are harvested after a short time interval (e.g. one day) or if there
is a weak root system. It can also be very challenging to maintain sterility in
long-term plant growth experiments, and if a low inoculum is used, your or-
ganism may be overgrown by contaminants.
7. The standard incubation time of seven days produces excellent results for pea.
However, peas have been harvested at one and three days with excellent re-
sults. The only real limit to the bacterial inoculation and plant growth time is
whether sufficient RNA can be isolated from rhizosphere bacteria.
13.3.3Procedure for small-seeded legumes such as clover

1. After surface sterilisation (Chapter5) place seeds on water agar and wrap plate
in foil. After the seeds have fully imbibed (approximately three hours) invert
the plates.
2. Leave at room temperature in the dark for one to two days, checking for root
development daily.
3. Place plates at 4C to slow growth if required.
4. Wash medium-size (grade 3) vermiculite, then oven dry at 60C overnight.
5. Drill a central hole into the white cap of a McCartney vial (see photo in
Figure13.2).
6. Add nitrogen-free rooting solution to wet the vermiculite in a bucket.
7. Fill McCartney vials (use acid-washed, rinsed and dried vials only) to the very
top with the medium-wet vermiculite. While McCartney or Universal bottles
(name depends on local usage) were used, it should be possible to use other
sterile containers such as Falcon tubes,
8. Place the caps with the central hole on to the vials. Excess solution should
come out of the holes in the cap.
9. Place vials in a rack and into an autoclave bag to sterilise on a fluid cycle
(121C for 15 minutes).
10. Store vials within the closed autoclaved bags in the fridge to cool.
289
11. Starter nitrogen (4mL per vial of 200mg/L NH4NO3) may need to be added
for some small-seeded legumes, but is not required for T. polymorphum or
T.purpureum or usually for large-seeded legumes.
12. Make a hole in the vermiculite through the cap hole using a sterile toothpick.
13. Place pre-germinated seeds so that the top of the seed is just sitting above cap
level.
14. Place KimCare wipes over the vials and dampen with sterile water.
15. Place tray of vials in the autoclave bag and transfer to growth room for two to
three days.
16. Remove vials from bag and remove tissues gently. Seed coats should be re-
moved by this process but any remaining seed coats should be removed with
sterile tweezers.
17. After five days of growth in the growth room (seven days post-plate germina-
tion) select uniform seedlings and inoculate seedlings with 1mL of a culture of
rhizobia via the hole in the cap. Typically 108 cells will be inoculated (see large-
seeded legumes above) but this can be varied depending on the experiment.
18. Vials will need to be watered once in a three-week period (maximum 30mL
water).
19. A Schott bottle with auto dispenser attached with hose and needle will assist
with large-scale watering (sterilise needle with ethanol and flame before each
watering). Alternatively, on a smaller scale, a sterile pipette tip can be used for
each vial.
20. Seedlings are typically grown for seven days in a growth room at 22C with
a 16hour light/eight hour dark cycle before isolation of rhizosphere bacteria.
However, the precise timing depends on the experiment. Growth may need to
be extended for slow-growing legumes.
Figure13.2 Germinated Trifolium polymorphum seedlings in the vial system
290
13.3.4RNA isolation from rhizosphere bacteria using Qiagen

RNeasy PLUS Micro kit
General preparation
Prepare 10mM Tris-Cl pH8.0 from 1M Tris-Cl pH8.0 (Sigma) stock. (Dilute
100L of 1M Tris to 10mL of sterile water).
Aliquot the required amount of RLT Plus buffer (Qiagen RNeasy PLUS Micro
kit) cat. no. 74034 (500l for three plants) and add 10L 2-mercaptoethanol
(2-ME) (Sigma) to 1mL of RLT Plus buffer.
Add 500L RLT Plus-2-ME buffer into the Fast Prep Lysing tube containing
FastPrep matrix B (Thermo) and incubate in ice. Alternatively, you can pre-
pare your own FastPrep lysing tubes as described in Appendix 13.10.2.
Procedure
1. Autoclave muslin cloth funnels and water. Treat funnels with RNase ZAP
(Ambion) before use.
2. Cut and discard the shoot. Add 6mL of sterile water and 12mL of RNase in-
hibitor to the Falcon tube containing roots and vermiculite.
3. Vortex immediately for five minutes.
4. Pool the vortexed solution from three plants, e.g. pea (for small plants this can
be increased to six to eight) and immediately filter the mix through four lay-
ers of muslin cloth. A note of WARNING: Miracloth is often used in place of
muslin but we found that it clogs with vermiculite fines, preventing bacteria
entering the filtrate and resulting in failure to recover significant RNA.
5. Spin the filtrate at 1,000rpm, 4C for one minute to remove any plant debris
and heavy vermiculite.
6. Transfer the supernatant to a 50mL centrifuge tube and spin at 10,000 rpm,
4C for 10 minutes (Sorvall centrifugeSS-34 head) to pellet the bacteria. You
will also see vermiculite sticking to the bottom of the tube with the bacteria
but this is not of concern.
7. Decant the supernatant and dry the tube on a filter paper.
8. Add 1,000L of 10mM Tris-Cl pH8.0, re-suspend (it will look like mud) and
load everything into a Fast Prep lysing tube (either use commercial FastPrep
lysing matrix B or prepare your own tubes as described in Appendix 13.10.2).
9. Lyse vermiculite-containing cells in the FastPrep (Thermo) at level 6.5 for 30

seconds. Place the tubes onto ice for one minute. Lyse again using the Fast
Prep at level 6.5 for 30 seconds.
10. Incubate tubes on ice for three minutes.
11. Go to Step 10 of Protocol 13.2.2 (isolation of RNA from free-living Rhizobium).
291
13.4 Protocol for isolation of total and small RNA (sRNA) from bacteroids of rhizobia
13.4Protocol for isolation of total and small RNA

(sRNA) from bacteroids of rhizobia
The legume nodules are picked into liquid nitrogen, ground and bacteroids isolat-
ed by a simple differential spin. More complex procedures for isolation of bacte-
roids may increase their purity but may slow down the isolation procedure. Speed
is important once the nodules are removed from liquid nitrogen and ground be-
cause changes in the amount of bacteroid RNA could occur. If RNA degradation
is a problem, you should consider using two-parts RNAase inhibitor per part of
isolation buffer for grinding of nodules (Step 3 below).
General preparation
Prepare 10mM Tris-Cl pH8.0 from 1M Tris-Cl pH8.0 (Sigma) stock.
Aliquot the required amount of RLT buffer (Qiagen RNeasy Plus micro kit)
(700L for each plant) and add 10L 2-mercaptoethanol (Sigma) for 1mL of
RLT buffer.
Add the 700L RLT-2-ME buffer to a FastPrep tube containing Lysing matrix
B (Thermo) or prepare your own tubes, as described above, and incubate in
ice.
Sterile 15mL centrifuge tubes.

Autoclaved mortar and pestle.
Prepare Isolation Buffer (10mM phosphate buffer pH7.4, 300mM sucrose
and 2mM MgCl2). (See Table13.1).
Phosphate buffer (Maniatis-A 1.5): Take 80.2L of 1M K2HPO4 and 19.8L

of 1M KH2PO4 and make up to 10mL with sterile distilled water to get phos-
phate buffer pH7.4.
Table13.1 Isolation buffer

Components Amount Amount
(10 mL) (50 mL)
1M K2HPO4 80.2 L 401 L
1M KH2PO4 19.8 L 99 L
Sucrose 1.03 g 5.15 g
MgCl2. 7(H2O) 4.0 mg 20 mg
Make the volume to 10 mL 50 mL
Filter sterilise the isolation buffer through a 0.22m filter and store at room tem-
perature.
292
13.5 Protocol for preparation of DNA-free RNA for use in Q-RT-PCR by treatment with Ambion Turbo DNA free
Procedure
1. Pick nodules (1.0g) into liquid nitrogen.
2. Filter the nodules using a tea strainer and weigh the nodules (do this quickly).
3. Macerate 1.0 g nodules in a mortar with the pestle with 5 mL of isolation

buffer. Macerate with an up and down motion; do not macerate with a circular
motion as it shears the genomic DNA.
4. Add 1mL of isolation buffer if the macerated mix is too thick.
5. Transfer the mix to a 15mL Falcon tube.
6. Spin at 1,000rpm for five minutes (to remove plant debris).
7. Carefully transfer the supernatant to a 1.5mL microfuge tube.
8. Spin at 6,000 rpm for five minutes and discard the supernatant.
9. The pellet contains the bacteroids.
10. Re-suspend the pellet in 250L of 10mM Tris-Cl pH8.0. Go straight to Step
7 of the RNA isolation procedure for free-living bacteria (Protocol 13.2.2). It
is important to go immediately to Step 7 because in this step the bacteroids are
added to RLTplus buffer which prevents RNA degradation.
13.5Protocol for preparation of DNA-free RNA for

use in Q-RT-PCR by treatment with Ambion
Turbo DNA free
You should now have very pure intact RNA which is suitable for Q-RT-PCR.
There are many variations of Q-RT-PCR so you should consult the manuals of
your Q-PCR machine and molecular biology suppliers of standard kits. However,
it is crucial to remove all traces of genomic DNA (gDNA) and Ambions Turbo
DNAase is particularly effective.
Procedure
1. Set up a reaction in a 0.5 mL RNAase-free tube with the additions from
Table13.2.
Table13.2 Components required to prepare DNA-free RNA
Components Volume
Total RNA with contaminating gDNA 144 L
RNase-free water 043 L
0.1 vol of 10 Turbo DNase buffer 5 L
Turbo DNase (2U/L) 1 L
Final volume 50 L
293
13.6 Protocol for SenseAmplinear RNA amplification
3. Add 0.1 volume of DNAase inactivation reagent.
4. Incubate for two minutes at room temperature with occasional mixing.
5. Spin at 11,000 rpm for two minutes at room temperature.
6. Transfer the supernatant to a new tube without disturbing the DNAase inac-
tivation reagent.
13.6Protocol for SenseAmplinear RNA

amplification
This procedure enables small quantities of RNA to be highly amplified by incor-

porating a T7 promoter into the RNA. The final product is a sense strand RNA
that can be used directly to produce fluorescently-labelled cDNA for microarray
analysis. In this book, it is used for the production of single-stranded antisense
cDNA that is chemically labelled with cy dyes for use in microarrays. The proto-
col is based on the commercial kit SenseAmp, available from Genisphere. This kit
is used because it produces sense stranded RNA which can be used directly for
reverse transcription to produce cDNA for use in microarrays. Ambion produce
a similar kit which can be used to produce amplified RNA (aRNA) although it is
in the antisense orientation. Ultimately, which kit you choose to use will depend
on your final application.
Reagent preparation
Thaw Vials 1, 2, 6, 8, 10, 11 and 12 as provided in the Genisphere kit at room

temperature, vortex and briefly microfuge.
Thaw Vials 3, 4, 5, 7, 9, and 13 on ice, briefly microfuge if necessary and keep

on ice at all times. Do not vortex.
Reverse transcribe up to 0.25 g of total RNA using the provided random

primer (Vial 2) in a total volume of 20L.
The random primer should be used at 2 by mass to the input total RNA (for
example, use 2L of Random Primer (0.5g) per 0.25g of total RNA).
Procedure
13.6.1First strand cDNA synthesis with random primers

1. For each RNA sample, prepare the following RNA/primer mix in a tube on ice:
19L total RNA (do not exceed 0.25g).
2. Add 2l Random 9mer RT primer (Vial 2); use RT primer at 2 by mass of

RNA.
294
3. Add Nuclease Free Water (Vial 10) to a volume of 11L.
4. Heat to 80C for 10 minutes. Keep on ice immediately for two minutes. Briefly
microfuge and return to ice.
5. For each reaction, prepare a Master Mix in a separate tube on ice:
4 L 5 First Strand Buffer

2 L 0.1 M DTT
1 L Superase-In (Vial 4)
1 L dNTP Mix (Vial 3)
1 L Superscript II (Invitrogen)
9 L total volume.
6. Combine the Master Mix and the RNA/primer mix for a volume of 20L. Mix
gently and microfuge. Incubate at 42C for two hours. Microfuge briefly after
incubation.
7. Add 80L of 1 TE buffer for a final volume of 100L.
13.6.2Purification of cDNA
Purify the 100L of cDNA produced above using the Qiagen MinElute PCR Pu-
rification Kit (Catalog # 28006) as follows.
1. Add 500L Buffer PB to the 100l cDNA sample and mix.
2. Apply the cDNA mixture to the MinElute column and centrifuge for one min-
ute at 13,000 rpm in a microcentrifuge.
3. Discard the flow-through. Place the MinElute column into the same collec-
tion tube.
4. Add 750L Buffer PE to the MinElute column and centrifuge for one minute.
5. Discard the flow-through. Place the MinElute column back into the same col-
lection tube.
6. Add 500 L 80% ethanol to the MinElute column and centrifuge for two
minutes.
7. Discard the flow-through. Place the MinElute column back into the same col-
lection tube.
8. Open the column caps and place in a microfuge with the cap opposite the di-
rection of the rotation of the rotor to avoid breaking the cap off. Centrifuge
for five minutes.
9. Place the MinElute column into a clean, labelled 1.5mL microfuge tube.
295
10. To elute cDNA, add 10L Buffer EB to the center of the column membrane.
Incubate at room temperature for two minutes. Centrifuge for two minutes.
Discard the column and save the 10L eluted cDNA. If the eluted cDNA is less
than 10L, increase the volume to 10L with Nuclease Free Water (Vial 10).
13.6.3Tailing of first strand cDNA

Caution: do not exceed 2g equivalent of dT-primed total RNA (or 0.25g ran-
dom- or random/dT-primed total RNA) for the tailing reaction.
1. Heat purified cDNA (10L) to 80C for 10 minutes. Add immediately to ice
for one to two minutes. Briefly microfuge and return to ice.
2. For each reaction, prepare a Master Mix in a separate tube on ice:
2 L 10 Reaction Buffer (Vial 6)

2 L Nuclease Free Water (Vial 10)
4 L 10mM dTTP (Vial 5)
2 L TdT Enzyme (Vial 7)
10 L Total volume
3. Combine the Master Mix and the cDNA for a volume of 20L. Mix gently and
microfuge.
4. Incubate in a 37C heat block for three minutes. Do not exceed three minutes.
5. Stop the reaction by heating to 80C for 10 minutes. Briefly microfuge and
cool to room temperature for one to two minutes.
13.6.4T7 Promoter synthesis

NB: when amplifying less than 10ng total RNA, or when performing two rounds
of SenseAmp amplification, diluting the T7 Template Oligo (Vial 8) in the first
amplification round may help to reduce T7 amplification artefacts commonly as-
sociated with amplification of very small sample sizes and with multiple rounds
of amplification.
1. For amplification of total RNA samples greater than 10 ng, add 2 L of T7

Template Oligo (Vial 8) to the tailed cDNA for a volume of 22L. Briefly vor-
tex and microfuge.
2. Incubate at 37C for 10 minutes to anneal the strands.
3. To each reaction, add the following components for a volume of 25L:
1 L 10 Reaction Buffer (Vial 6)

1 L dNTP mix (Vial 3)
1L Klenow Enzyme (Vial 9).
296
4. Mix gently and microfuge. Incubate at room temperature for 30 minutes.
5. Stop the reaction by heating to 65C for 10 minutes. Place on ice.
6. Option 1 (recommended): proceed to the in vitro transcription reaction using

half (12.5L) of the promoter-modified cDNA. Save the remaining modified
cDNA at 20C for future use or for use in a parallel amplification reaction.
Option 2: The entire 25L of promoter-modified cDNA may be used in the in

vitro transcription reaction but it must be purified first. Add 75L of 1 TE
Buffer for a final volume of 100L and purify using the MinElute PCR Purifi-
cation Kit as directed in the cDNA purification step. Add Nuclease Free Water
(Vial 10) to adjust the volume to 11.5L. Add 1L of the T7 Template Oligo
(Vial 8) to the purified cDNA for a volume of 12.5L.
13.6.5In vitro transcription

1. Incubate the 12.5L of cDNA at 37C for 10 minutes to re-anneal the strands.
2. Thaw the T7 Nucleotide Mix (Vial 11) and 10 T7 Reaction Buffer (Vial 12)
at room temperature and keep at room temperature until use. Thoroughly
vortex the 10 T7 Reaction Buffer (Vial 12) to avoid precipitation of buffer
components.
3. For each reaction, add the following components at room temperature, for a
final volume of 25L:
8.0 L T7 Nucleotide Mix (Vial 11)

2.5 L 10 T7 Reaction Buffer (Vial 12)
2.0 L T7 Enzyme Mix (Vial 13).
4. Mix gently and microfuge. Incubate in a thermocycler (with heated lid) at
37C for four to 16 hours. Or, place the reaction in a 37C heat block for
five minutes and then transfer to a 37C air hybridisation oven for four to
16hours. It is essential to avoid evaporation and condensation of the reaction
during this step.
13.6.6Purification of senseRNA
Purify senseRNA using the RNeasy Kit (Qiagen cat. no. 74106) following Qiagens
protocol for RNA Cleanup.
1. Add 75L of RNAase-free water to the reaction mix.
2. Add 350L of RLT buffer (+2-ME) and mix well by pipetting.
3. Add 250L of 95% (v/v) ethanol and mix well by pipetting. Immediately add
to the purification column.
4. Spin the column at 12,000rpm for 30 seconds.
297
13.7 Protocol for indirect labelling of firststrand cDNA synthesis using GE healthcare Cyscribe Post-Labelling Kit
5. Add 500L of RPE buffer and spin at 12,000rpm for 15 seconds; repeat the
step twice.
6. Discard the collection tube and place the column in a new collection tube, and
spin at 12,000rpm for two minutes.
7. Add 50L of RNAase-free water, incubate at room temperature for two min-
utes and spin at 12,000rpm for one minute.
8. Add the eluted 50L RNA to the column, incubate for two minutes and spin
at 12,000rpm for one minute for maximum recovery of senseRNA.
9. Quantitate the RNA as described above. An OD260/OD280 ratio of 2.02.3 is

most desirable. Higher ratios may indicate that an excessive poly (A) tail was
generated during the amplification reaction.
13.7Protocol for indirect labelling of firststrand

cDNA synthesis using GE healthcare Cyscribe
Post-Labelling Kit
This protocol describes how to reverse transcribe RNA into cDNA for microarray
or other experiments that require a DNA template. The protocol uses a secondary
coupling procedure where RNA is reverse transcribed to cDNA with the use of
aminoallyl UTP to which either cy3 or cy5 can be subsequently chemically cou-
pled. Thus, it is optimally adapted for microarray use. It is included here because
cy labelling of DNA is still an important technique, even though the use of micro-
arrays is now waning as people switch to RNAseq.
Reagent preparation
Cyscribe Post-Labelling kit with Cyscribe GFX purification kitGE Health-

care Cat No RPN 5660X.
Cyscribe GFX purification kit Cat No 27-9606-02 (optional).

2.5 M NaOH (1 g/10 mL sterile Milli-Q water; filter sterilise with 0.45 m
filter).
2MHEPES (4.77g/10mL sterile Milli-Q Water; filter sterilise with 0.45m

filter).
0.1MNa2CO3 (Sigma S7795) 0.53g/50mL sterile Milli-Q water.

0.1M sodium bicarbonate pH9.0 (Sigma S6297): 4.2g in 500mL sterile Milli-
Q water; adjust to pH9.0 with 0.1MNa2CO3 (2530mL). Filter sterilise and
dispense into aliquots and store at 20C for up to two months.
Ethanol absolute (Sigma-Aldrich 32221) diluted to 80% (v/v) ethanol.
298
4M Hydroxylamine hydrochloride Aldrich 15941-7 (2.78g in 100mL sterile

Milli-Q Water). Make fresh every time.
Heating block set at 65C, 70C, and water bath at 42C.

Microcentrifuge.
RNAase-free filter tips.
Spectrophotometer.
Microcuvettes (Brand UV-cuvettes) Cat No Fischer SUPC CXA-205-020F.
Nitrile Gloves.
NB: prepare a 100 mM aa-dUTP solution as follows: dissolve 1 mg of aa-
dUTP in 17L DEPC-treated water (Sigma RNAase Free Water) and 0.68L
1MNaOH. Measure pH with a drop on a pH filter paper and immediately
adjust to pH7.0 using 0.10.2L of 1M NaOH. Store aa-(dUTP) solutions at
20C.
Table13.3 Composition of the mix (50X)

Components Volume Final Concentration
100 mM dATP 10.0 L 25 mM
100 mM dGTP 10.0 L 25 mM
100 mM dCTP 10.0 L 25 mM
100 mM dTTP 2.0 L 5 mM
aa-dUTP 8.0 L 20 mM
Procedure
13.7.1Primer annealing
1. Add the following components to a 1.5mL microcentrifuge tube on ice. The
RNA may need to be concentrated in a Speed Vac. Wipe out machine with
RNAZap and set temperature to 37C. To reduce 16.5L to 11.5L will take
about five minutes; to reduce 22L to 11.5L will take about 15 minutes. Do
not dry the sample. Reduce volume to slightly less than 11.5L then make up
to 11.5 L with sterile water.
Total RNA (1015 g) 11.5 L

Random Nanomers (from Cyscribe kit) 1 L
Mix gently by pipetting up and down.
2. Incubate the reaction at 70C for five minutes.
3. Cool reaction at RT for 10 minutes to allow the primers and mRNA template
to anneal.
4. Spin in a microfuge for 10 seconds to collect material.
299
13.7.2Extension reactions
1. Place the cooled annealing reaction on ice and add the following components,
making sure to add the enzyme last.
5 Cyscript Buffer 4 L
0.1M DTT 2 L
50 aa-UTP/Nucleotide mix (see Table13.3) 0.5 L
Cyscript RT 1.0 L
2. Mix by stirring with the pipette tip (Cyscript is sensitive to mechanical disrup-
tion) and spin in a microfuge for 10 seconds to collect material.
3. Incubate the reaction at 42C in a water bath for 5hours.
13.7.3Degradation of mRNA
1. Add 2L 2.5MNaOH to each cDNA reaction.
2. Mix reaction by vortexing and spin for 15 seconds.
3. Incubate reaction at 37C for 15 minutes.
4. Add 10L 2MHEPES (free acid used to neutralise the NaOH added in Step
1) to each cDNA reaction.
5. Mix reaction by vortexing and spin for 15 seconds.
The cDNA reaction is ready for purification or can be stored at 20C.
13.7.4Purification of cDNA with Cyscribe GFX purification kit

1. For every cDNA reaction to be purified, place one Cyscribe GFX column into
a clean collection tube. Add 500L of capture buffer to each Cyscribe GFX
column.
2. Briefly spin down the cDNA reaction and add into Capture Buffer, pipetting
up and down five times.
3. Centrifuge the column at 13,000rpm in a microfuge for 30 seconds.
4. Discard the flow-through.
5. Add 600L of 80% (v/v) ethanol to each column and centrifuge the column at
13,000rpm in a microfuge for 30 seconds.
6. Repeat Step 5 another two times.
7. Centrifuge the column at 13,000rpm for 10 seconds to remove the residual

wash buffer. Discard the collection tube.
8. Transfer the column into a new microfuge tube and add 60L0.1M sodium
bicarbonate (pH9.0) to the top of the glass fibre matrix.
300
9. Incubate at room temperature for one to five minutes.
10. Centrifuge the column at 13,000rpm for one minute to collect the purified
cDNA.
13.7.5Labelling of amino-allyl modified cDNA with cy dyes

Turn the lights off for all subsequent steps
1. Add (60 L) amino-allyl modified cDNA (in 0.1 M sodium bicarbonate
pH9.0) directly into an aliquot of cyDye NHS ester (from the Cyscribe kit)
and re-suspend NHS ester by pipetting several times.
2. Spin down at 13,000rpm for one minute to collect sample at the bottom.
3. Incubate in the dark overnight (preferably in a cupboard).
4. Next day, add 15L 4M hydroxylamine to each coupling reaction;
5. Mix by pipetting up and down and INCUBATE IN DARK at room tempera-

ture for 15 minutes.
13.7.6Purification of Cy DyeLabelled DNA with Cyscribe GFX

purification kit
Keep the Elution Buffer at 65C.
1. For every cDNA reaction to be purified, place one Cyscribe GFX column into
a clean collection tube. Add 500L of capture buffer to each Cyscribe GFX
column.
2. Microfuge pulse the cDNA reaction (from Step 5 above) and add into Capture
Buffer, pipetting up and down five times (work quickly, capture buffer will de-
grade cy dyes).
3. Centrifuge the column at 13,000rpm for 30 seconds (on the membrane, Cy3
will look pink and cy5 blue). Discard the flow-through.
4. Add 600 L of wash buffer to each column and centrifuge the column at
13,000rpm for 30 seconds.
5. Repeat Step 4 another two times.
6. Centrifuge the column at 13,000rpm for 10 seconds to remove the residual

wash buffer. Discard the collection tube.
7. Transfer the column into a new microfuge tube and add 70L elution buffer
(65C) to the top of the glass fibre matrix.
8. Incubate at RT for one to five minutes.
9. Centrifuge the column at 13,000rpm for one minute to collect the purified
cDNA.
301
13.8 Protocol for Rubicon WTA amplification of RNA to produce double-stranded cDNA (SIGMA CAT NO WTA2)
Measure the absorbance of the labelled cDNA (all 70L) on a spectrophotometer

using a microcuvette (100L volume). Use the following formulae:
Calculate the Nucleotide/Dye ratio and required volume for x pmole of both Cy5
and Cy3 labelled cDNA.
Nucleotide/ Dye ratio=pmole cDNA/pmole Cy Dye
(Example values are provided in the Excel sheet below. If you have an electronic
version of this document, you can enter your data after double clicking on the
sheet.)
http://www.corning.com/lifesciences/technical_information/techdocs/calcula-
tor/index.asp
13.8Protocol for Rubicon WTA amplification of RNA

to produce double-stranded cDNA (SIGMA CAT
NO WTA2)
This protocol describes how to convert RNA to cDNA and amplify this by a very
simple two-step process (Rubicon WTA amplification). It can be used instead of
the more complex linear T7 amplification of RNA described in Protocol 13.6 and
has the added advantage of producing double-stranded cDNA. The following Pro-
tocol (13.9) describes how to produce cy-labelled DNA from any double-stranded
302
13.8 Protocol for Rubicon WTA amplification of RNA to produce double-stranded cDNA (SIGMA CAT NO WTA2)
DNA. Protocols 13.8 and 13.9 can therefore be used instead of Protocols 13.6 and
13.7 to produce cy-labelled cDNA.
Before starting the Rubicon WTA procedure, RNA should be treated with Turbo
DNAase (Protocol 13.5).
Procedure
Part A: Library synthesis reaction
1. Thaw the library synthesis buffer, library synthesis solution, library synthesis
enzyme and nuclease-free water.
Thoroughly mix the library synthesis buffer and the library synthesis solu-
tion. Dissolve any precipitate in these solutions by briefly heating at 37C with
mixing.
2. To at least 25ng of total RNA, add 2.5L library synthesis solution and make
up to 16.6L with nuclease-free water.
3. Mix and incubate in a thermocycler programmed for 70C for five minutes
then cool to 18C.
4. To the cooled-primed RNA immediately add the following:

a. 2.5 L library synthesis buffer
b. 3.9 L water
c. 2 L of library synthesis enzyme.
5. Incubate in a thermal cycler using the following parameters (one cycle):

18C for 10 minutes
25C for 10 minutes
37C for 30 minutes
42C for 10 minutes
70C for 20 minutes
Cool to 4C.
6. Mix and centrifuge in a microfuge.
Part B: Amplification reaction
1. Thaw the amplification mix and 10mM dNTP mix.
2. Prepare the following master mix:
301 L of nuclease-free water

37.5 L Amplification Mix
7.5 L WTA dNTP mix
303
13.9 Protocol for direct labelling of double-stranded cDNA made by Rubicon WTA kit
3.75 L Amplification enzyme.

3. Add the entire library synthesis reaction mix from Step 6 of Part A (25L) to
the master mix solution from Step 2 Part B and mix.
4. Divide the sample from Step 3 into five 75L reactions and incubate in a ther-
mal cycler using the following parameters:
94C for two minutes

17 cycles (94C for 30 seconds, 70C for five minutes)
Cool to 4C.
5. After cycling is complete, maintain the reaction at 4C or store at 20C.
6. Pool the samples (approx. 375L) and purify using Qiagen PCR purification
kit (cat no 28104).
7. Elute in 100L of 50C preheated 10mM Tris-EDTA (pH8.0) or EB Buffer.
8. Quantify using Nanodrop or Agilent Bioanalyser.
13.9Protocol for direct labelling of double-

stranded cDNA made by Rubicon WTA kit
This protocol is a very simple way of producing cy-labelled DNA that can be used
for microarrays or to produce biotinylated double-stranded DNA. It can be used
for any double-stranded DNA including genomic DNA or cDNA. It is particularly
easy to use after Rubicon amplification of RNA to produce double-stranded DNA.
Thus, by using Protocols 13.8 and 13.9 together it is possible to use ng quantities
of environmental RNA to produce g quantities of cDNA labelled with various
labels, including fluorescent tags and biotin.
1. In a sterile amber microfuge tube, add 10L of ds cDNA (2g of ds cDNA)

isolated using the Rubicon WTA kit (Sigma Cat Number WTA2) and make up
the volume to 18L with Sigma ultra-pure water (molecular biology reagent,
Cat: W4502).
2. Add 2l of random primer (3g/L) (e.g. Invitrogen SKU# 48190-011). Heat

at 95C for five minutes and then put on ice for five minutes.
3. On ice, add:
a. 3L of 10 Klenow buffer (Fermentas)
b. 5L of 10 dNTP mix (10 dNTP mix: 1.2mM each dATP, dGTP, dTTP;
0.6mM dCTP; 10mM Tris pH8.0; 1mM EDTA. Do NOT use the dNTP
mix from the kit).
304
13.10Appendices
c. 2L of Cy5 dCTP or Cy3 dCTP (1mM stock, GE Healthcare Lifesciences,

Cat: PA55321).
d. 2L of Klenow enzyme (Fermentas);
4. The total reaction volume is 30L.
5. Spin briefly and incubate the reaction mixture at 37C overnight protected
from light.
6. Use a Qia-quick PCR purification kit (Qiagen Cat: 28104) to remove unincor-
porated/quenched Cy dyes. Elute twice using 30L of Sigma water to maxim-
ise recovery.
13.10Appendices
13.10.1RNAase inhibitor composition

The final composition of RNAase inhibitor is:
20mM EDTA disodium salt dihydrate(EDTA Sigma cat no E5134)

25mM trisodium citrate dihydrate (Fisher cat no S/3320/53)
70g/100mL ammonium sulfate (Sigma cat no A4418).
To make RNAase inhibitor, first prepare stock solutions of EDTA and sodium
citrate as follows:
0.5M EDTA pH8.0 (dissolve 18.61g EDTA disodium salt dihydrate/100mL

and adjust the pH to 8.0 with NaOH while stirring)
1M sodium citrate (dissolve 29.4g trisodium citrate dihydrate/100mL).

To prepare the final solution, add the following to a beaker:
93.5mL sterile water

4mL 0.5M EDTA
2.5mL 1M sodium citrate
70g ammonium sulfate.
Stir the reagents on a heated stirring block until the salt completely dissolves. Allow
the solution to cool, with constant stirring. Adjust the pH to 5.2 with 1MH2SO4
Transfer to a screw-capped bottle and store at room temperature or at 4C.
RNAase inhibitor is used at twice the volume of samples to be stabilised.
305
13.10Appendices
13.10.2FastPrep Lysing tubes

Materials required
0.1mm Silica Beads (Biospec Products Inc. cat no: 1107910z).

0.1mm Glass Beads (Biospec Products Inc. cat no: 1109101).
Microtubes 2.0 mL (Sarstedt cat no: 72.694.005) or equivalent.
Preparation procedure
1. Wash the silica beads and glass beads separately in RNAase-free water and
oven dry them.
2. Add 300 mg of washed silica beads and 100 mg of washed glass beads to
micro-tubes.
3. Loosely cap the micro-tubes and autoclave before use. These steps cannot be
guaranteed to produce RNAase-free tubes and lysing matrix. All cell disrup-
tion procedures should therefore be done with either RNA extraction buffer
(e.g. RLT buffer) or RNAase inhibitor present to prevent RNA degradation.
13.10.3A nitrogen-free rooting solution

Chemical Stocks Stock conc. Volume for 1L Final conc.
CaCl2.2H20 73.51 g/500 mL 1M 1 mL 1 mM
KCl 3.73 g/500 mL 100 mM 1 mL 100 M
MgSO4.7H2O 98.59 g/500 mL 800 mM 1 mL 800 M
Fe EDTA 1.84 g/500 mL 10 mM 1 mL 10 M
H3BO3 2.16 g/100 mL 350 mM 0.1 mL 35 M
MnCl2.4H2O 1.78 g/100 mL 90 mM 0.1 mL 9 M
ZnCl2 0.109 g/100 mL 8 mM 0.1 mL 0.8 M
Na2MoO4.2H2O 0.121 g/100 mL 5 mM 0.1 mL 0.5 M
CuSO4.5H2O 0.075 g/100 mL 3 mM 0.1 mL 0.3 M
KH2PO4 25 g/1 L 18 mM 20 mL 368 M
Na2HPO4 28.4 g/1 L 20 mM 20 mL 400 M
Add each chemical in turn to 900mL of water and make up to 1L. It is essential
that the phosphates are added last otherwise they may precipitate and not dis-
solve.
306
Index
A Calopogonium 198
caeruleum 16, 204
Acacia cells needed for different seed sizes 180
karroo 16 Centrosema 195, 198
mangium 16 pubescens 16, 204
saligna 84 CFU (Colony Forming Units) 112
senegal 16 Chamaecrista
actidione 55. Seecycloheximide rotundifolia 16
adzuki bean 204 chloroform:
agarose 230 isoamyl alcohol 226
agarose gel 224 Cicer
Agrobacterium sp 230 arietinum 16
allantoic acid 202 Codariocalyx gyroides 204
allantoin 202, 209, 210 collecting nodules 25
amide producing legumes 202 site 26
Aminobacter 128, 130, 131 timing 28
Arachis 40, 94 common bean 190, 204
hypogaea 16, 198 Congo Red 39, 43, 55, 56, 111, 130
asparagine 211 conjugation
autoclave performance 57 biparental mating 253
avoiding contamination 57 commonly used vectors 253
Azorhizobium 128, 130, 131 filter mating 256
plate mating 255
B protocol 255
triparental mating 253
bacterial genomic DNA isolation
counting rhizobia
cell preparation/extraction 223
by spectrophotometry 119
CTAB method 222
diluents for serial dilutions 110
gel electrophoresis 224
direct counts in Neubauer haemocytometer 123
phenol/chloroform method 225
direct counts in Petroff-Hauser chamber 122
bacterial species
direct counts under the microscope 120
assignment characteristics 125
drop-plate method 111
definition 125
Most Probable Number (MPN) method 115
rhizobia 127
optical density-viable count calibration 120
Bambara groundnut 204
purposes 109
Biserrula pelecinus 16, 76, 106, 149, 164
serial dilutions 109
black gram 204, 215
simulataneous MPN for multiple species 116
Brachiaria arrecta 195
spread-plate counting 111
Brachiaria brizantha 195
survival on seed 114
Bradyrhizobium 128, 130, 131, 132, 240
cowpea 199, 215
elkanii 47, 198
cross-row technique
japonicum 47, 48
data collection/assessment 158
bromophenol blue 242
layout diagram 159
bromothymol blue 56
Crotalaria 198
Burkholderia 115, 129, 130, 131
juncea 16
dilworthii 44, 48, 132
cryopreservation 69
inoculants 177
culture recovery 70
on Congo Red 56
medium 69
phymatum 132
procedure 69
sprentiae 44, 47, 55, 132
crystal violet 133
C CTAB (cetyltrimethylammonium bromide) 222, 223,
227
Cajanus 107 Cupriavidus 47, 129, 130, 131
Cajanus cajan 198, 204 Cyamopsis tetragonoloba 204
Calliandra 198 cycloheximide 55
calothyrsus 16
307
Index
D Qiagen minelute column 263

transformation 265
Desmodium 198 vector preparation 263
intortum 16 genomic tools
ovalifolium 195, 204 Joint Genome Institute (JGI) 268
rensonii 204 National Center for Biotechnology Information
Devosia 128, 130, 131 (NCBI) 270
dimethylsulfoxide 240 germination of impermeable seed
dissecting microscope 113 acid treatment 93
dithiothreitol 257 boiling 93
DNA-free RNA for use in Q-RT-PCR 293 scarification 92
DNA samples smoke treament 93
removal of humic acids 228 vernalisation 93
DNA sequences Gitschier buffer 240, 242
automated sequencing machines 267 Gliricidia sepium 215
international databases 267 glutamine 211
Dolichos Glycine max 16, 198, 204
lablab 16, 198 Gram staining protocol 133
purpureus 16 guar 204
dry combustion method 192
H
E
Hardenbergia spp 204
Ensifer 128, 130, 131 Hedysarum spinosissimum 50
Erythrina spp. 16 herbicide status 87
ethidium bromide 231, 234, 241 host range
effectiveness for N2 fixation 74
F specificity for nodulation 74
fababean 190 hyacinth bean 204
field experiments with rhizobia hydrogen peroxide 40
aims 145 Hydroponic systems 82
background numbers 148 hydroxylamine hydrochloride 299
carrier/adhesive selection 150
cross-row technique 155
I
design 146 indirect labelling of first strand cDNA
long term 164 amino-allyl cDNA labeling with cy dyes 301
nodule count/score 161 cDNA purification 300
nodule occupancy 162 cy-dye labeled cDNA purification 301
plant/seed nitrogen 163 extension reactions 300
plant/seed yield 163 mRNA degradation 300
rhizosphere counts 163 primer annealing 299
seeding machines 160 reagent preparation 298
seed inoculation for machine seeding 160 Inga
site biotic/abiotic stress 149 edulis 16
site layout 153 inoculant production 167
site/seed preparation 147 Australian standards 179
site selection 147 batch culture 169
swards 153 carrier materials 171
free N farming 164 contaminant control 173
desired commercial characters of strains 184
G fermentors 170
Galactia striata 195 freeze-dried powders 177
genomic libraries 261 genetic stability of strains 184
genomic DNA preparation 264 granular inoculants 176
insert DNA prepration 264 inoculating peat packets 175
ligation 264 moisture content control 174, 175
PCR verification of insert presence 265 problems 179
precautions 263 quality control 168, 179, 181
pTH1522 261 shelf life 182
308
Index
storage of inoculants 178 recovery from ampoules 68

supply chain 167 vacuum test 68
survival on seed 185 lysozyme 222, 223, 225, 229
inoculants for field experiments
adhesives for attachment to seed 151 M
application 151
2-mercaptoethanol 242
inoculation
Macroptilium atropurpureum 32, 76, 198, 204
general methods 95
Macrotyloma 107
International Biodiversity Convention 27
axillare 198
K geocarpum 204
malate dehydrogenase 239
Kersting's groundnut 204 media
Kjeldahl digestion 192 antibiotics 54
Koch's postulates 58, 59 indicators 54
kudzu 204 vancomycin 55
mediadefined
L Brown & Dilworth 53
BSM 52
Lablab 107
mediaundefined
Lablab purpureus 204
LA 50, 111
laminar flow hoods 57
glycerol broth 52
Lathyrus 107
Luria-Bertani 51
Lebeckia ambigua 37, 58, 150, 177
peptone 51
leghemoglobin
sucrose-glucose broth 52
chemical assay 105
tryptone yeast extract (TY) 51, 234, 248
extraction from nodules 104
yeast mannitol agar (YMA) 50, 111
purified 105
Medicago 77, 149
legumes
littoralis 185
floral forms 19
rugosa 185
major uses 15
sativa 16, 75
need to inoculate 20
tornata 185
sub-families 19
megaplasmids 231
Lens 107
membrane transport assay
Leonard jars 115
bacterial growth 138
lima bean 204
protocol 139
linear RNA amplification
transport wash 139
cDNA purification 295
Mesorhizobium 47, 128, 130, 131, 224, 240
first strand cDNA synthesis 294
Mesorhizobium cicer bv biserrulae 106, 116
in vitro transcription 297
methylene blue 123
reagent preparation 294
Methylobacterium 46, 128, 130, 131, 132
senseRNA purification 297
Microvirga 128, 130, 131
T7 promoter synthesis 296
lotononidis 47
tailing first strand cDNA 296
texensis 47
Listia angolensis 46, 47
zambiensis 47
Listia bainesii 46
molar extinction coefficient
Lotononis bainesii 46
o-nitrophenol 260
Lotus
p-nitrophenol 260
ornithopioides 37
Mucuna pruriens 16, 198
lucerne 185
mung bean 185, 204, 215
lupin 185, 201
Lupinus N
albus 16
angustifolius 16, 104 near infra-red spectroscopy (NIR) 192
lyophilization Neorhizobium 129, 130, 131
ampoule constriction 66 N-free rooting solution 288, 306
ampoule dryingprimary 64 ninhydrin 210
ampoule dryingsecondary 66 nitrate 202
ampoule preparation 63 nodulation and effectiveness screening systems
ampoule sealing 66 advantages and disadvantages 75
309
Index
assumptions 78 phylogenetic trees

glass test tubes 75 building with MEGA 272
growth pouches 75, 77 four step process for building 271
hydroponics 75 maximum likelihood (ML) 272
intact soil cores 75, 87 neighbour joining (NJ) 271
Leonard jars 75, 79 selecting reference sequences 275
sand culture 75, 81 software programs for building 271
transparent O'Hara vials 75, 76 pigeon pea 194, 204, 215
nodules Pisum 107
aeschynomenoid 30 sativum 16, 35
desiccated 40 plant growth conditions
desmodioid 30 hygiene 97
determinate 30 physical 96
excavation 29 plasmid preparation
indeterminate 30 alkaline lysis 231
indigoferoid 30 cesium chloride 234
lupinoid 30 recovery from CsCl gradients 235
morphology 29 plasmid profiling
rehydration 40 electrophoresis 231
ruptured 33 modified Eckhardt procedure 228
senescence 29 polyethylene glycol 227
storage 31 Polymerase Chain Reaction (PCR) 236
surface sterilisation 40 cell preparationbasic 236
nucleic acid extraction from soil 226 cell preparation using PEG 237
nutrient solutions for growing legumes fingerprinting 239
Broughton and Dilworth for P. vulgaris 92 gel electrophoresis of products 241
CRS nutrient solution 91 oligonucleotide primers16SRNA, atpD, dnaJ,
Jensen 90 dnaK, glnA, gyrB, nodA, nodC, recA, rpoB
Norris and Date 89, 90 238
oligonucleotide primers for fingerprinting 239
O reactions for ERIC, Box and rep 240
single amplicon 237
Ochrobactrum 128, 130, 131
potassium ferricyanide 209
Onobrychus
proteinase K 222, 223, 226
vicifolia 16
proteobacteria 127
Ornithopus
Pseudomonas putida 239
compressus 37
Psophocarpus tetragonolobus 198, 204
sativus 16
Pueraria phaseoloides 16, 198, 204
P Q
Panicum maximum 195
qRT-PCR amplification
passport data 35
reference genes 239
peanut 185
quantification of effectiveness 97
Pectobacterium atrosepticum 239
acetylene reduction 99
pH 36
leghemoglobin content 103
buffers 48, 137
nodulation scoring 98
change 48
nodule morphology 98
control 48, 137
nodule occupancy 98
growth substrate 137
top dry weight 97
Phaseolus
quantifying legume N2 fixation
lunatus 204 15
N abundance method 198
vulgaris 16, 87, 89, 198, 204 15
N dilution and incorporation methods 195
phenol: 15
N enrichment techniques 199
chloroform:
accounting for below ground N 193
isoamyl alcohol 224, 226
acetylene reduction method 217
phenylhydrazine hydrochloride 208
amino-N analysis 210
Phoretix 1D Advanced software analysis 241
biomass and biomass N sampling 190
phosphorus pentoxide 66
calculating concentrations of ureides, nitrate and
Phyllobacterium 128, 130, 131
amino-N 214
310
Index
choosing non-N2-fixing reference plants 199 Mesorhizobium, Methylobacterium,

correlation of %ureide-N to %Ndfa 203 Microvirga, Neorhizobium, Ochrobactrum,
grinding and sub-sampling 192 Phyllobacterium, Rhizobium, Shinella 17, 128
large scale methods 217 rhizobia in beta-Proteobacteria
natural abundance methods 196 Burkholderia, Cupriavidus 17, 128
nitrate analysis 212 rhizobial storage
nitrogen balance method 216 agar slopes 62
nitrogen difference method 194 cryoprotectants 63
sample collection and drying 191 freezing 62
sampling for ureide analysis root bleeding sap 206 lyophilization 62, 63
sampling for ureide analysis stems or stem oil covered agar slopes 62
segments 207 parent/working lot system 70
sampling for ureide analysis ureide assay 208 porcelain beads 62, 63
sampling for ureide analysis vacuum extraction soil, peat or clay 62
204 Rhizobium 128, 130, 131
sampling time 190 leguminosarum 136, 230, 289
selection of method 188 leguminosarum bv trifolii 47, 116, 132, 224, 241
summary of methods 188 leguminosarum bv viciae 47, 107, 231, 239, 250
ureide method 202 ribosomal 16 S rRNA 239
ureide method factors to consider 215 rice bean 204
RNA amplification to produce double-stranded cDNA
R amplification reaction 303
dye labelling 304
recombinase 239
procedure 302
red clover 185
RNAase inhibitor composition 305
reporter gene assays 256
RNA isolation from rhizobia
cell permiabilisation 258
general precautions for 284
specific activity calculations from microtitre plates
reasons for 283
260
RNAse A 222
rhizobia
RODI (reverse osmosis deionised) water 225
amino acid metabolism 136
arabinose catabolism 135 S
aromatic compound use 136
autotrophy/heterotrophy 135 safranin 134
colony characteristics 130 salicylic acid 212
colony dimorphism 131 Sarcosyl (n-lauroylsarcosine 229
curated repositories 26 seed
dicarboxylate metabolism 135 germination 94
disaccharide metabolism 135 preparation for authentication 69
glyoxylate cycle 136 surface sterilization 94
Gram staining 133 seed collection 36
growth rates 42, 129 serradella 185
mean generation times 130 Sesbania grandiflora 215
metabolic diversity 129 Shinella 128, 130, 131
micronutrient studies 138 sigma factor (54) 239
mixed substrate utilization 136 sigma factor (70) 239
motility 133 silica gel 32
N sources for growth 137 Sinorhizobium 128, 130, 131
optimum growth temperature 129 medicae 45, 47, 116, 132, 230, 246
pH range 129 meliloti 45, 47, 132
pigmented 46 siratro 84, 204
range of growth temperatures 129 Sitona discoideus 149
siderophores for iron uptake 135 sodium dodecyl sulfate 222
sugar metabolismEntner-Doudoroff/pentose sodium hypochlorite 40
phosphate pathways 134 soybean 204, 215
tricarboxylic acid cycle 136 strain specificity
vitamin requirements 135 across legume genera 107
rhizobia in alpha-Proteobacteria across legume species in a genus 107
Aminobacter, Azorhizobium, Bradyrhizobium, cultivars within a legume species 106
Devosia, Ensifer (Sinorhizobium), streaking pattern 42
311
Index
Stylosanthes identification of mutated gene 249

humilis 16 promoterless reporter gene 246
substrates for effectiveness experiments protocol 247
sand 84 pUT plasmid delivery system 246
vermiculite 84 transpositionally active mobile element 245
symbiovar trap host 26
in Bradyrhizobium 126 trapped strains
in Mesorhizobium 126 less desirable characteristics 32
in Rhizobium 126 population difference 32
in Sinorhizobium 126 trapping nodule bacteria 32
replacement for biovar 126 Trifolium 75, 77
incarnatum 16
T polymorphum 290
pratense 16
Taq polymerase 240
purpureum 290
tea strainer 40
repens 16
total and small RNA (sRNA) isolation from free-
Tween 80 40
living cells 284
TY broth 225
cell growth and lysis 285
reagent preparation 284 U
RNA analysis 286
total and small RNA (sRNA) isolation from legume ureide-producing legumes 202
rhizospheres
large seeded legumes 288 V
legume growth media 288
Vicia 107
rooting solution 288
benghalensis 16
small seeded legumes 289
faba 16, 104
total and small RNA (sRNA) isolation from rhizobial
sativa 16, 77
bacteroids
Vigna
general preparation 292
angularis 198, 204
isolation buffer 292
mungo 198, 204
transduction
radiata 77, 198, 204
improving phage titre 251
subterranea 204
mutant phage lysate 251
umbellata 198, 204
phage titre calculation 252
unguiculata 16, 198
use 250
wild type phage lysate 250
W
transposon mutagenesis
frequency of isolation of mutants 247 white clover 185
helper E. coli strain 245 winged bean 204
312

Working With Rhizobia

Uploaded by

Copyright:

Available Formats

Working With Rhizobia

Uploaded by

Document Information

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

Working With Rhizobia

Uploaded by

Copyright:

Available Formats

Working with rhizobia

J.G. Howieson and M.J. Dilworth (Eds.)

Centre for Rhizobium Studies

ACIAR MONOGRAPH SERIES

Australian Centre for International Agricultural Research (ACIAR) 2016

The Australian Centre for International Agricultural Research (ACIAR) has a

1 The legume-rhizobia symbiosis and assessing the need to inoculate ........15

2 Collecting nodules for isolation of rhizobia...............................................25

3 Isolation and growth of rhizobia.................................................................39

3.7 Routine culture media for rhizobia............................................................47

5 Authentication of rhizobia and assessment of the legume symbiosis

7 Taxonomy and physiology of rhizobia......................................................125

8 Field experiments with rhizobia................................................................145

9 Inoculant production and quality control................................................167

10 Measurement of nitrogen fixation.............................................................187

10.7 Nitrogen difference method......................................................................194

11 Fundamental molecular techniques for rhizobia.....................................221

12 Specialised genetic techniques for rhizobia..............................................245

13 Methods for isolation of RNA from rhizobia............................................283

About this manual

Study of the legume/rhizobium symbiosis necessitates an understanding of meth-

About the authors and compilation of the manual

Correct citation for the manual

We thank ACIAR (through the ECCAL project http://aciar.gov.au/project/

Ricardo Silva Araujo

1.1Legumes in agriculture, society and the

Table1.1 The major uses of legumes, together with some examples

1.2 Rhizobia and nodules

1.3 The legumes and their nodule characteristics

1.3.1 Subfamily Caesalpinioideae

1.3.2 Subfamily Mimosoideae

1.3.3 Subfamily Papilionoideae

indeterminate, often branched and occasionally encircling the subtending root

Figure1.1 Flowers formed by Subfamilies of the Leguminosae; a)

1.3.4 Actinorhizal plants

1.4 Assessing the need to inoculate

The establishment of a N2-fixing legume/rhizobia symbiosis needs the presence

The only sure way to know if inoculation is necessary is to conduct need-to-inoc-

1. An uninoculated control without N fertiliser (I, N)

2. A treatment inoculated with the best-quality inoculant, without N fertiliser

3. An uninoculated control treatment with N fertiliser added (I, +N)

Plant growth (%)

Figure1.2 Possible results obtained from need-to-inoculate trials

A combination of treatments is needed to understand whether N is the principal

In Case 1, there is no response to inoculation or N fertiliser, suggesting that some

Figure1.3 Inoculated and nodulated

Figure1.4 The results of a need to inoculate trial on a farm in Ethiopia, where

1.5 Selecting inoculant quality strains

Ineffective background populations which preclude the desired inoculant

Collecting nodules for

Table2.1 Some curated repositories of nodule bacteria

2.2 Collecting new strains of rhizobia

2.2.1The in situ repository

2.3Respecting the international biodiversity

The 1992 United Nations Convention on Biological Diversity (CBD- http://www.

2.4 Collecting nodules

plastic bags to store whole plants with soil attached to roots

50mL vials for collecting soil (if nodules cannot be found)

2.4.2 Timing of collection

2.4.3 Excavation of plant roots

1.1Legumes in agriculture, society and the

Table1.1 The major uses of legumes, together with some examples

Table2.1 Some curated repositories of nodule bacteria

2.3Respecting the international biodiversity

2.5Many nodules from one plant or a few nodules

3.5Recognising nodule bacteria growing on solid

3.6Some useful diagnostic features of species of

Table3.2 Buffers used in rhizobial growth media and their pKa

3.7.2.3Tryptone yeast extractTY

3.7.2.6Sucrose glucose broth (for fast-growing

3.7.2.7Glycerol broth (for slow-growing

3.7.3.3JMMa minimal defined medium