Brewing Yeast and Fer

Saccharomyces
BIOTECHNOLOGY HANDBOOKS
Series Editors: Tony Atkinson and Roger F. Sherwood
PHLS Centre for Applied Microbiology and Research
Division of Biotechnology
Salisbury, Wiltshire, England
Volume 1 PENICILLIUM ANDACREMONIUM

Edited by John F. Peberdy
Volume 2 BA GILL US
Edited by Colin R. Harwood
Volume 3 CLOSTRIDIA
Edited by Nigel P. Minton and David]. Clarke
Volume 4 SACCHAROMYCES
Edited by Michael F. Tuite and Stephen G. Oliver
A Continuation Order Plan is available for this series. A continuation order will bring
delivery of each new volume immediately upon publication. Volumes are billed only upon
actual shipment. For further information please contact the publisher.
Saccharo"!)Jces
Edited by
Michael F. Tuite
The University of Kent, Canterbury
Kent, England
and
Stephen G. Oliver
Manchester Biotechnology Centre, UMIST
Manchester, England
Springer Science+Business Media, LLC

Ltbrary of Congraee Catalogtng-tn-Publtcatton Data
Saccharo•yces 1 ed1ted by M1chae1 F. Tu1te and Stephen G. 011var.

p. c1. -- <B1otechnology handbooks ; v. 4l
Includas b1b11ograph1ca1 raferancas and Index.
ISBN 978-1-4899-2643-2 ISBN 978-1-4899-2641-8 (eBook)
DOI 10.1007/978-1-4899-2641-8
1. Saccharo•yces--B1otechnology. 2. Saccharoaycas. I. Tu1te,
M1chu1 F. II. 011ver, s. D. <Staphen George>, 1949-
III. Ser tes.
TP248.27.Y43S22 1991
6B0'.62--dc20 90-19442
CIP
ISBN 978-1-4899-2643-2
© 1991 Springer Science+Business Media New York

Originally published by Plenum. Press, New York in 1991
Softcover reprint of the hardcover 1st edition 1991
AII rights reserved
No part of this book may be reproduced, stored in a retrieval system, or transmitted
in any form or by any means, electronic, mechanical, photocopying, microfllming,
recording, or otherwise, without written permission from the Publisher
Contributors
J. R. Dickinson • School of Pure and Applied Biology, University of

Wales College of Cardiff, Cardiff CF1 3TL, Wales
M. J.Dobson • Department of Botany, University of Nottingham,
Nottingham NG7 2RD, England
A. J. Kingsman • Department of Biochemistry, University of Oxford,
Oxford OX1 3QU, England
S. M. Kingsman • Department of Biochemistry, University of Ox-
ford, Oxford OX1 3QU, England
C. Kreutzfeldt • Institut fur Pharmakologie und Toxikologie, Phil-
ipps-Universitat Marburg, Lahnberge, 3550 Marburg, Federal Re-
public of Germany
T. M. Matthews • Department of Chemical Engineering, University
of Manchester, Institute for Science and Technology, Manchester
M60 1QD, England
E. J. Mellor • Department of Biochemistry, University of Oxford,
Oxford OX 1 3QU, England
Stephen G. Oliver • Manchester Biotechnology Centre, University of
Manchester, Institute for Science and Technology, Manchester M60
1QD, England
Michael F. Tuite • Biological Laboratory, University of Kent, Canter-
bury, Kent CT2 7NJ, England
C. Webb • Department of Chemical Engineering, University of Man-
chester, Institute for Science and Technology, Manchester M60
1QD, England
R. B. Wickner • National Institutes of Health, Bethesda, Maryland
20892
W. Witt • Institut fur Pharmakologie und Toxikologie, Philipps-Uni-
versitat Marburg, Lahnberge, 3550 Marburg, Federal Republic of
Germany
v
Preface
In this volume we aim to present an easy-to-read account of the genus

Saccharomyces that we hope will be of value to all students and researchers
wishing to exploit this important genus, be it for academic or commer-
cial purposes. Individual chapters have been commissioned to cover
specific aspects of the biology of Saccharomyces species: growth, genetics,
and metabolism, with the emphasis on methodology. Basic principles are
discussed without an over-detailed, step-by-step breakdown of specific
techniques, and lengthy discussions of standard molecular, biological,
and biochemical techniques (e.g., polyacrylamide gel electrophoresis,
protein purification, DNA sequencing) have been avoided. We hope the
volume will provide a quick reference to the current status of a wide
range of Saccharomyces-specific methodologies without focusing ex-
clusively on recent developments in molecular techniques which can be
found in the ever increasing numbers of "cloning manuals." By necessity,
much of what is described in this volume concentrates on one particular
species of Saccharomyces, namely Saccharomyces cerevisiae. This is not just a
reflection of the authors' interests, but indicates the extent to which this
simple eukaryote has been studied by biologists from all walks of life, for
all sorts of reasons. If this volume can provide a broader knowledge base
to the experienced yeast researcher, or ease the path of someone just
starting work with Saccharomyces, then we will have achieved our aim.
We are grateful to all the authors and to Ken Derham at Plenum
Press for their patience during the preparation of this volume.
Michael F. Tuite
University of Kent, Canterbury
Stephen G. Oliver
University of Manchester
vii
Contents
Chapter 1
Introduction 1
Michael F. Tuite and Stephen G. Oliver
Chapter 2
Structural Biochemistry 5
C. Kreutzfeldt and W. Witt
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2. Basic Macromolecules in Saccharomyces . . . . . . . . . . . . . . . . . . . 5
2.1. Proteins and Peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2. Polysaccharides, Structural Mannoprotein, and
Polyphosphate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3. Lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.4. Ribonucleic Acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3. Cell Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.1. Cell Envelope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.2. Vacuoles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.3. Endoplasmic Reticulum and Secretory and Endocytic
Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.4. Mitochondria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4. Life Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.1. Mating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.2. Meiosis and Sporulation............. .............. 34
4.3. Vegetative Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5. Cell Fractionation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
5.1. Isolation of Organelles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
5.2. Identification of Yeast Cell Subfractions . . . . . . . . . . . . 43
5.3. Autolytic Enzymes in Yeast Cells . . . . . . . . . . . . . . . . . . . 43
6. Methods of Synchronization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
ix
x CONTENTS
6.1. Imposition of a Cell Cycle Block . . . . . . . . . . . . . . . . . . . 44

6.2. Centrifugational Methods . . . . . . . . . . . . . . . . . . . . . . . . . 44
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Chapter 3
Metabolism and Biosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
J. R. Dickinson
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
2. Outline of Carbohydrate Metabolism . . . . . . . . . . . . . . . . . . . . 60
2.1. Catabolite Repression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.2. Catabolite Inactivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.3. Effect of Oxygen .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . 62
2.4. The Pasteur Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
2.5. Catabolite Conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
2.6. Molecular Genetics of Carbon Metabolism . . . . . . . . . . 66
3. Outline of Nitrogen Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.1. Amino Acid Biosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.2. Proteases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.3. Nucleotide Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
3.4. Molecular Genetics of Nitrogen Metabolism . . . . . . . . . 74
4. Transport of Substrates into the Cell . . . . . . . . . . . . . . . . . . . . 74
5. Regulation of Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.1. Genetic Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.2. Physiological and Biochemical Studies . . . . . . . . . . . . . . 83
5.3. Further Aspects of Regulation . . . . . . . . . . . . . . . . . . . . . 87
6. Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Chapter 4
Methods in Classical Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
R. B. Wickner
1. Introduction 101
2. Life Cycle of Saccharomyces Cerevisiae .................... . 101
3. Tetrad Analysis ............ , .......................... . 105
4. Aneuploidy .......................................... . 112
5. Mutant Induction and Isolation ........................ . 115
CONTENTS xi
5.1. Mutagenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

5.2. Mutant Isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
6. Genetic Mapping Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
6.1. Meiotic Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
6.2. Mitotic Recombination . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
6.3. Aneuploid Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
6.4. The spoll Mapping Method . . . . . . . . . . . . . . . . . . . . . . . 120
6.5. Chromosome-Loss Methods . . . . . . . . . . . . . . . . . . . . . . . 121
6.6. 2-~J.m DNA Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
6.7. Overall Mapping Strategies . . . . . . . . . . . . . . . . . . . . . . . . 122
6.8. The Genetic Map............ ................... .. 122
7. Non-Mendelian Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
7.1. The Mitochondrial Genome . . . . . . . . . . . . . . . . . . . . . . . 123
7.2. The Killer Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
7.3. 2-j..l.m DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
8. Appendix I: Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
9. Appendix II: Sample Tetrad Data . . . . . . . . . . . . . . . . . . . . . . . 140
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Chapter 5
Recombinant DNA Techniques 149
A. J. Kingsman, E. J. Mellor, M. J. Dobson, and S.M. Kingsman
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
2. Yeast Transformation .................. ............ :. . . 150
3. Plasmid Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
3.1. ARS-Based Plasmids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
3.2. 2-~J.m-Based Plasmids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
3.3. Yeast Gene Isolation Using Plasmid Vectors . . . . . . . . . 152
3.4. Site-Directed Mutagenesis in Yeast Using Single-
Stranded Plasmids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
4. Minichromosome Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
5. YAC Cloning Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
6. Integrative Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
6.1. Targeted Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
6.2. Allelic Rescue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
6.3. Gene Disruption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
6.4. Refinements of Transplacement . . . . . . . . . . . . . . . . . . . 163
7. Selection Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
xii CONTENTS
Chapter 6
Expression of Heterologous Genes . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Michael F. Tuite
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
2. Introduction of Heterologous DNA into Yeast . . . . . . . . . . . . 171
2.1. Basic Cloning Technology . . . . . . . . . . . . . . . . . . . . . . . . . 171
2.2. Basic Vector Design............................... 171
3. Transcription of Heterologous Genes . . . . . . . . . . . . . . . . . . . . 173
3.1. The Problem of Introns . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
3.2. Encoded Pre-Pro Sequences . . . . . . . . . . . . . . . . . . . . . . 174
3.3. Transcriptional Promoters . . . . . . . . . . . . . . . . . . . . . . . . . 176
3.4. Termination of Transcription . . . . . . . . . . . . . . . . . . . . . . 179
3.5. Regulated Promoters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
4. Translation of Heterologous mRNAs . . . . . . . . . . . . . . . . . . . . 182
4.1. Codon Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
4.2. 5' Untranslated mRNA Leader . . . . . . . . . . . . . . . . . . . . 184
4.3. AUG Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
5. Posttranslational Modifications . . . . . . . . . . . . . . . . . . . . . . . . . . 186
5.1. Yeast Secretion Pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
5.2. Use of Heterologous Signal Sequences . . . . . . . . . . . . . 187
5.3. Homologous Signal Sequences . . . . . . . . . . . . . . . . . . . . . 189
5.4. Posttranslational Modification and Secretion . . . . . . . . 191
5.5. Other Posttranslational Modification Events . . . . . . . . . 192
6. Maximizing Product Yield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
6.1. Stability of Recombinant Plasmids in Yeast . . . . . . . . . . 195
6.2. Host Strains for Heterologous Gene Expression . . . . . 200
7. Summary and Prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Chapter 7
"Classical" Yeast Biotechnology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Stephen G. Oliver
1. History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
2. Baking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
3. Beer Brewing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
4. Sake Brewing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
5. Wine Making . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
5.1. White Wine Production . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
CONTENTS xiii
5.2. Red Wine Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230

6. Strain Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
7. Ethanol Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
8. Flocculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
9. Polysaccharide Utilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
10. Rare Mating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
11. Protoplast Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
12. Recombinant DNA Technology . . . . . . . . . . . . . . . . . . . . . . . . . 240
12.1. Cloning and Expression in Yeast of Genes
Encoding Amylolytic Enzymes . . . . . . . . . . . . . . . . . . . . 240
12.2. Cloning and Expression of Endoglucanase
Genes in Yeast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
12.3. Cloning and Expression of Complete
Metabolic Pathways: The Way Ahead . . . . . . . . . . . . . . 243
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Chapter 8
Culture Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
T. M. Matthews and C. Webb
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
2. Nutritional Requirements of Saccharomyces . . . . . . . . . . . . . . . . 250
2.1. Carbon .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 250
2.2. Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
2.3. Phosphorus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
2.4. Sulfur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
2.5. Trace Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
2.6. Growth Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
3. Process Variables that Influence the Growth
of Saccharomyces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
3.1. Hydrogen Ion Concentration . . . . . . . . . . . . . . . . . . . . . . 254
3.2. Temperature and Ethanol Inhibition Effects . . . . . . . . 255
3.3. Dissolved Oxygen and Substrate Inhibition Effects . . . 256
3.4. Carbon Dioxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
4. The Theory and Practice of Yeast Culture Systems . . . . . . . . 258
4.1. Batch Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
4.2. Continuous Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
4.3. Practical Continuous Culture Systems . . . . . . . . . . . . . . 266
5. Monitoring the Growth of Saccharomyces . . . . . . . . . . . . . . . . . . 271
6. Cell Separation Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
7. Immobilized Cell Systems . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . 272
xiv CONTENTS
7.1. Passive Immobilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272

7.2. Active Immobilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
7.3. Fermenters for Immobilized Saccharomyces Cells . . . . . 275
8. Downstream Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
8.1. Pressed Yeast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
8.2. Dried Yeast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
8.3. Yeast Extract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
8.4. Alcoholic Beverages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
8.5. Ethanol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
Chapter 9
Biochemical Techniques 283

Michael F. Tuite and Stephen G. Oliver
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
2. Cell Disruption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
2.1. Whole Cell Disruption . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
2.2. Protoplast Lysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
3. Radioactive Labeling of Macromolecules . . . . . . . . . . . . . . . . . 286
3.1. RNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
3.2. DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
3.3. Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
4. RNA Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
4.1. Differential Extraction Techniques . . . . . . . . . . . . . . . . . 290
4.2. Messenger RNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
4.3. Transfer RNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
4.4. Ribosomal RNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
4.5. Double-Stranded RNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
5. DNA Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
5.1. Chromosomal DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
5.2. Mitochondrial DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
5.3. Plasmid DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
6. In vitro Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
6.1. In vitro Transcription Systems . . . . . . . . . . . . . . . . . . . . . 298
6.2. In vitro Translation Systems . . . . . . . . . . . . . . . . . . . . . . . 300
6.3. In vitro DNA Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
6.4. Two-Dimensional Gel Electrophoresis in the Analysis
of DNA Replication ............................... 306
7. Pulsed Field Gel Electrophoresis of Yeast Chromosomes . . . 307
71. Theoretical Background . .. . . .. .. .. .. .. .. .. . . . . . . . 308
CONTENTS xv
7.2. Pulsed Field Gradient Gel Electrophoresis . . . . . . . . . . 309

7.3. Orthogonal-Field-Alteration Gel Electrophoresis..... 310
7.4. Contour-Clamped Homogeneous Electric Field . . . . . . 311
7.5. Vertical Pulsed-Field Gradient Gel Electrophoresis
and Constant Electric Field with Rotating Gel
Platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
7.6. Field Inversion Gel Electrophoresis . . . . . . . . . . . . . . . . 312
7. 7. General Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
7.8. Sample Preparations .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 313
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
Index ................................................... 321

Introduction 1
MICHAEL F. TUITE and STEPHEN G. OLIVER
Fungi have been exploited by mankind for many thousaJ;tds of years,

with perhaps the earliest recorded examples being the use of yeast to
make ethanol, a process known to the Sumerians and the Babylonians
before 6000 B.c. In the ensuing 8000 years one particular genus of yeast,
namely, Saccharomyces, has played a central role in the commercial exploi-
tation of fungi by mankind. These facultative anaerobes utilize the Emb-
den-Meyerhof pathway to convert sugars to pyruvic acid, with each
molecule of pyruvic acid then being reductively decarboxylated to give
rise to one molecule each of ethanol and carbon dioxide. This simple,
efficient way of fermenting glucose to ethanol and carbon dioxide has
provided the foundation for two of our major food industries, brewing
and baking (Fig. 1).
Yet from its earliest foundations yeast "biotechnology" has received
little attention from "Biotechnologists," be they from biological or chem-
ical disciplines. The development of brewing and baking strains with
improved fermentation characteristics has occurred largely on an em-
pirical basis using technologies that have remained essentially unaltered
for millennia. Such traditional approaches do have their limitations,
although one has to admit that they have been successful in the develop-
ment of effective strains for use in many traditional processes. However,
it has only been with the advent of modern molecular genetic techniques
that yeast biotechnology has begun to take on a new face in which the
organism itself has been subjected to the trickery that can profoundly
change its genetic constitution. Even 10 years ago the idea that yeasts
would be used commercially to produce effective vaccines against hepati-
tis B virus can hardly have been contemplated, yet it is a testimony to the
flexibility and manipulability of these organisms that this dream is now a
reality.
MICHAEL F. TUITE e Biological Laboratory, University of Kent, Canterbury, Kent

CT2 7NJ, England. STEPHEN G. OLIVER • Manchester Biotechnology Centre,
University of Manchester, Institute for Science and Technology, Manchester M60 lQD,
England.
I
2 MICHAEL F. TUITE and STEPHEN G. OLIVER
Eth•nol--.....~ 11 Brewlng 11
Glucose---_. Pyruvic
•cid
C•rbon
dioxide
Figure 1. The exploitation of yeast metabolism.
Yeasts have also become important model systems for basic research
into the biology of the eukaryotic cell, with one particular species, S.
cerevisiae, being at the forefront of this research. In the early 20th cen-
tury detailed studies by Warburg, Crabtree, and others into the fermen-
tative ability of this species revealed a number of important aspects of
respiration and fermentation, particularly the regulation of metabolism
by glucose. Yet it was not until the pioneering work of Winge and Lin-
degren in the 1940s that S. cerevisiae became the subject of genetic re-
search. Once the limitations of a homothallic lifestyle had been over-
come (by mutation), genetic analysis became routine and increasingly
came to be used to complement biochemical studies. More recently, with
the development of powerful molecular genetic tools, a remarkable up-
surg~ in the exploitation of S. cerevisiae in fundamental research has
occurred, which has relied to a larger extent on the already accumulated
knowledge of the biochemistry, physiology, and genetics of the species.
Today, much of our understanding about eukaryotic gene structure and
function, cytoskeletal organization, and the cell cycle comes from studies
with this simplest of eukaryotes.
For anyone who has not worked with this genus before (and it is
largely for such people that this book has been written) Saccharomyces
yeasts appear not unlike bacterial-they are unicellular and grow rapid-
ly on simple, well-defined media with a population doubling time under
2 hr. Yet they have many of the fundamental characteristics of a higher
eukaryotic cell: a nuclear membrane, cytoplasmic organelles such as mi-
tochondria, receptor and second messenger systems, and so on. The
ability to rationally manipulate all aspects of gene expression by in vitro
genetic techniques offers S. cerevisiae a unique place among eukaryotic
model systems.
INTRODUCTION 3
To allow rational genetic manipulation of the biosynthetic capabilit-

ies of any organism does not just require an ability to clone a gene into
the said organism, but at the very least requires a detailed understanding
of the organism's life cycle, culture requirements, and metabolism. In
this book we aim to provide a series of chapters, each dealing with a
fundamental aspect of the biology of Saccharomyces, but with the empha-
sis on methodology and current academic and commercial exploitation
of members of this genus. It is not our intention, however, to provide a
state-of-the-art "methods" book with highly detailed protocols, but
rather to provide an overview of the wide range of traditional and mod-
ern methodologies available to yeast researchers, with references to orig-
inal publications in which the fine details can be found. Lengthy discus-
sions of standard molecular and biochemical techniques are also avoided
since these can be found in a number of widely available standard texts.
Structural Biochemistry
2
C. KREUTZFELDT and W. WITT
1. INTRODUCTION
Saccharomyces cerevisiae is a unicellular eukaryote in which a combination

of biochemical, genetic, and molecular biological techniques have been
used in the study of cell structure, organelle biogenesis and function,
and the regulation of growth and mating. This chapter provides an
overview of the biochemistry of the structural components of the yeast
cell.
2. BASIC MACROMOLECULES IN SACCHAROMYCES
2.1. Proteins and Peptides

The architecture of the yeast cell has long attracted the interest of
cell biologists, and consequently, many proteins and peptides have been
analyzed in respect to their role in regulation and maintenance of cell
structure and metabolism. A number of topics have been selected for
detailed consideration in this chapter: tubulin and actin are the main
structural proteins in the yeast cell; the ribosomal proteins have been
physically characterized, but their individual functions are largely un-
known; mating factors are peptide hormones which trigger physiologi-
cal and morphological changes. The discussion of glycoproteins concen-
trates on the processing and sorting of proteins and the formation of the
cell wall.
2.1.1. Thbulin
During the cell cycle, cytoplasmic and nuclear microtubules are
formed. This process has been extensively studied by electron microscopy
C. KREUTZFELDT ·• Phase GmbH, D-2410, Molin, Federal Republic of Germany.

W. WITT e Institut fur Pharmakologie und Toxikologie, Philipps-Universitat Mar-
burg, Lahnberge, 3550 Marburg, Federal Republic of Germany.
5
6 C. KREUTZFELDT and W. WITT
(for reviews, see Byers, 1981; King and Hyams, 1982). Tubulin, the major
component of the microtubule proteins, has been purified from S. cere-
visiae (Clayton et al., 1979; Kilmartin, 1981) and these preparations have
been shown to copolymerize with brain tubulin and to self-assemble in
vitro to form microtubules. Two-dimensional electrophoretic analysis sug-
gests that yeast tubulin, like brain tubulin, contains two subunits, alpha
and beta, with mol. wt. 55,000 and 52,000, respectively. Besides sim-
ilarities in molecular weight and polymerization between yeast and brain
tubulins, there are differences concerning the number of protofilaments
of microtubles formed in vitro, the sensitivity to certain drugs, and the
speed of in vitro depolymerization (Kilmartin, 1981). Although eDNA
clones for alpha and beta chicken brain tubulin did not hybridize to yeast
DNA under stringent conditions (Cleveland et al., 1980), it was neverthe-
less possible to use such probes to isolate the yeast beta tubulin gene TUB2
(Neff et al., 1983). The amino acid sequence predicted from the DNA
sequence ofTUB2 gives rise to a protein of mol. wt. 51,073 (457 residues)
which shows sequence homology of more than 70% with chicken beta
tubulin.
Recently, the yeast alpha-tubulin genes TUBJ and TUBJ have been
sequenced (Schatz et al., 1986a). TUBJ and TUBJ give rise to slightly
different polypeptides with mol. wt. 49,701 and 49,694, respectively.
Both alpha-tubulins are highly homologous to those from other species.
Yeast microtubules have been shown to contain both alpha-tubulins.
Null mutation experiments have led to the conclusion that both tubulins
essentially fulfill common qualitative functions (Schatz et al., 1986b).
2.1.2. Actin
Yeast actin has an intracellular concentration of about 0.5 mg/ml
cell volume (Greer and Schekman, 1982a) and its physical and physio-
logical properties have been studied in vitro. The protein has been pu-
rified from the cytosol of S. cerevisiae by column chromatography to yield
a polypeptide of mol. wt. 43,000 (Greer and Schekman, 1982a). Inhibi-
tion of DNasei activity can be used as an assay to identify fractions
containing actin.
KCl and Mg2 + induce polymerization of actin to produce microfila-
ments of 7 nm diameter which can be decorated with heavy muscle
meromyosin (HMM). HMM ATPase activity has been found to be stimu-
lated by yeast actin. Phalloidin binds to yeast actin filaments and protects
them from depolymerization induced by Kl. While cytochalasin B de-
creases the stability of yeast and muscle actin to equal extents, a number
of differences between these two actins have been identified. In contrast
to muscle actin, the polymerization of yeast actin is strongly affected by
low concentrations of Ca2+ (Greer and Schekman, 1982b) by increasing
STRUCTURAL BIOCHEMISTRY 7
the critical polymerization concentration. This effect is rapidly reversed

by complexing Ca2+ with EGTA or by addition of Mg2+. It has been
suggested that Ca2+ interacts directly with yeast actin.
In yeast there is a single gene for actin and this has been cloned and
sequenced (Gallwitz and Sures, 1980; Ng and Abelson, 1980). The pre-
dicted amino acid sequence consists of 374 residues and is about 90%
homologous to mammalian cytoplasmic gamma actin and Physarum actin.
Disruption of the yeast actin gene has confirmed that it encodes an
essential function (Shortie et al., 1982). Conditionally expressed actin
gene mutations have been constructed and these have revealed that actin
is involved in the organization, assembly, and function of the yeast cell
surface (Shortie et al., 1984; Novick and Botstein, 1985; see also Section
4.3).
2.1.3. Ribosomal Proteins

As in other eukaryotes, the yeast cytoplasmic ribosome consists of a
large (60S) and a small subunit (40S). The precise number of the yeast
ribosomal proteins (r-proteins) found in each subunit is still under con-
tention, since different results have been obtained depending on condi-
tions of preparation and the two-dimensional (2-D) polyacrylamide gel
electrophoresis (PAGE) system used. However, there is agreement that
the small subunit of Saccharomyces contains 30 ± 5 and the large one 40 ±
5 proteins (Warner, 1982). The r-proteins comprise about 15% of total
cell proteins. The numbering of r-proteins largely depends on the 2-D
PAGE system used for resolution. Therefore, different nomenclatures
exist in the literature. Most of them designate the proteins derived from
the small subunit with the prefix S to a number and those from the large
one with a L. In a comparative study, Otaka and Osawa (1981) have
correlated the most common nomenclatures for r-proteins of S. cere-
visiae. In more recent publications, the nomenclature of Kruiswijk et al.
(1978) is preferred.
Almost all r-proteins have been purified by preparative column
chromatography. Their amino acid composition and, in some cases, the
N-terminal amino acid sequence have been determined (ltoh et al., 1979;
Otaka et al., 1982, 1983) by protein chemical methods. Like r-proteins
from other sources, most yeast r-proteins are basic and only a few are
acidic (S2, S5, S19, S26, Ll, L24, L44/45; nomenclature of Kruiswijk et
al., 1978).
Several phospho-r-proteins (S2, S10, L44/45; nomenclature of
Kruiswijk et al., 1978) have been identified in Saccharomyces. S 10 seems to
be the analog of mammalian S6 (Zinker and Warner, 1976; Leer et al.,
1982).
Resistance to trichodermin, cycloheximide, and cryptopleurine has
been found to be conferred by altered ribosomal proteins L3, L29, and

rp59, respectively, and this phenomenon has been used for molecular
cloning of these proteins (for literature see below).
A considerable number of genes coding for yeast r-proteins S7, S 10,
Sll, S16A, S24, S31, S33, L3(tcml), L4, Ll6, L17A, L25, L29(cyh2), L32,
L34, L36, L46, rp24, rp28, rp29, rp39, rp51, rp52, rp58, rp59(cryl), and
rp63 have been cloned, and several have also been sequenced (the muta-
tions conferring resistance to the antibiotics mentioned are given in
parentheses). These investigations permit the following conclusions
about r-protein gene organization in yeast: (1) the r-protein genes are
normally not clustered, (2) some r-proteins are encoded by two unlinked
genes, and (3) many, but notall,r-proteingenescontainintrons(Leeretal.,
1984, 1985a; Warner et al., 1985; Larkin and Woolford, 1983, and liter-
ature cited therein). Current research is directed to understanding how
the synthesis of about 70 different r-proteins is coordinately regulated.
There are indications that this occurs posttranscriptionally, but the auto-
genous regulation seems to be rather different from that elucidated for
Escherichia coli (Warner et al., 1986; Planta et al., 1986; Gourse et al., 1986;
Nomura et al., 1984). However, according to present knowledge, regula-
tion mechanisms that act on the transcriptional level cannot be excluded.
Comparison of DNA sequences of various yeast r-protein genes has
revealed two different conserved sequences located within a region from
about -200 to -500 upstream of the polypeptide initiation codon
(Teem et al., 1984; Leer et al., 1985b). These sequences, called HOMOLl
[consensus sequence: AACATC(CT)(GA)T(GA)CA] and ribosomal pro-
tein gene (RPG) box [consensus sequence: ACCCATACAT(TC)(TA)],
have been exclusively found in upstream regions of r-protein genes (for
the two known exceptions, see below), but in some of them the HO-
MOLI or RPG box is missing. If both HOMOLI and RPG box are
present, the latter is located 3' to the HOMOLI box. The genes coding
for the r-proteins S33 and L3 seem not to be flanked by any of these
boxes. HOMOLI and RPG box have been suggested to be involved in
the control of r-protein gene transcription as enhancer-like structures
(Leer et al., 1985b). This view has been strengthened by Woudt et al.
(1986), who have shown that deletion of one or both conserved up-
stream sequences (RPG 1, RPG2) of ribosomal protein gene L25
strongly affects the transcription of this gene.
5' upstream sequences which are closely related to HOMOLl and
RPG have been also identified in the TEFJ and TEF2 genes, respectively,
which both code for yeast elongation factor EFl-alpha (Huet et al.,
1985). A protein factor named translational upstream factor (TUF) de-
tected in a partially purified protein fraction has been shown to bind
with varying affinities to HOMOLI and RPG boxes of several ribosomal
protein genes, as well as to the related sequences of the TEF1 and TEF2
genes (Huet et al., 1985; Vignais et al., 1987). The TUF protein may thus
be involved in the coordinate regulation of those genes which code for
the proteins of the translational machinery.
2.1.4. Mating Factors

Haploid S. cerevisiae cells constitutively secrete peptide pheromone
into the medium. They are known as alpha- and a-factors and are pro-
duced by cells of alpha- and a-mating type, respectively. These factors,
which act only on cells of the opposite mating type, induce the mating
process (Thorner, 1981; Sprague et al., 1983), finally yielding alpha/a
diploids insensitive to either pheromone. Alpha-factor, which is now
commercially available (e.g., Sigma, St. Louis), is commonly used for the
synchronization of cell cultures (see also Section 6). The pheromone
peptides have been purified and their amino acid sequences have been
established:
a-factor:
N H 2 - Tyr-Ile-Ile-Lys-Gly-Val-Phe-Trp-Asp-Pro-Ala-(S-Far)Cys-OCH 3
or
NH 2 -Tyr-Ile-Ile-Lys-Gly-Leu-Phe-Trp-Asp-Pro-Ala-(S-Far)Cys-OCH 3
alpha-factor:
NH 2 - Trp-His-Trp-Leu-Gln-Leu-Lys-Pro-Gly-Gln-Pro-Met-Tyr-COOH
(S-Far)Cys-OCH 3 denotes S-farnesyl cysteine methyl ester.Cyx denotes a
cysteine residue modified by an as yet unkonwn group (Stotzler et al.,
1976; Andregg et al., 1988). Most likely, there are two genes in a-cells
each of them coding for a different a-factor. Thus two different factors
are secreted by a single haploid strain (Thorner, 1981; Sprague et al.,
1983). TheN-terminal Trp residue is missing in some alpha-factor prep-
arations, possibly owing to action of an aminopeptidase. However, the
remaining peptide exhibits full biological activity. Almost any other
modification, including oxidation of the Met residue, results in severe ·
reduction of biological activity.
The response of target cells to pheromone (arrest of growth at G 1,
elongation yielding pear-shaped cells) can be directly assayed under the
microscope (Duntze et al., 1973). The action of the pheromone is only
transient, since both factors are rapidly degraded by proteases of the
cells of opposite mating type. The rate of degradation has been esti-
mated to be about 107 alpha-factor molecules/min at a concentration of
106 -10 7 cells/ml (25°C). The usual concentration of pheromones in the
-=
0
::0:::
:=
~
c
;::l
;;1
§
§
Cl.
19 23 57 67 84 90 111 132 153 165
~
::;:
CHO ~
sa spt at sp2 a 2 t24 sp3

II Asp Lys Arg Glu Ala Glu Ala Trp ... Tyr
I I Lys Arg Glu Ala ~!~ Ala Glu Ala I Trp . . . Tyr I Lys Arg -: -: -: - COOH
i
' - - - - - - - - - - - c l e a v a g e by proteinase F - - - - - - - - - - - - - - - - - '
at a2
Ill Glu Ala Glu Ala ITrp ... Tyr I Lys Arg Glu Ala ':;'!~ Ala Glu Ala ITrp . . . Tyr I Lys Arg
DPAPase IV
u u CPase a
t I t
DPAPase IV
U
CPase a
at a2
IV ITrp . Tyr I ITrp . . . Tyr I
Figure 1. Proposed scheme of alpha-factor maturation. This model has emerged from studies by different groups
(Kurjan and Herskowitz, 1982; Julius et al., 1983, 1984; Achstetter and Wolf, 1985a). Since alpha-factor maturation
comprises glycosylation and secretion, we refer also to Sections 2.1.5 and 3.3. Transcription of the alpha-factor gene
finally gives rise to an mRNA which codes for a large polypeptide containing a leader sequence with a terminal signal
sequence and four tandemly arranged sequences, each consisting of a spacer sequence (sp 1, etc.) followed by the alpha-
factor amino acid sequence (stage 1). The prepro-alpha-factor is suggested to be cotranslationally inserted into the
endoplasmic reticulum where the polypeptide backbone (moL wt. 18,600) is core-glycosylated at sites indicated (CHO).
Core glycosylation increases the moL wt. to 26,000, which is consistent with the addition of three (GlcNAc)2 (Man)9 units.
Unlike other secreted proteins (e.g., invertase), the signal sequence is not removed at this stage. It has been suggested that
the signal sequence is required to immobilize the prepro-alpha-factor on membrane structures to ensure proper process-
ing. There are some indications that in the Golgi bodies, mannose-rich outer chains are polymerized onto the core
oligosaccharides. Also, proteolytic processing seems to begin at this stage and continues in the secretory vesicles which
appear later in the secretory pathway. The first proteolytic cleavages in the prepro-alpha-factor molecule are performed,
most likely, by proteinase F which has been found to be associated with membranes (stage II). This enzyme recognizes the
first two residues (Lys Arg) at the N-termini of each of the spacer sequences. Spacer sequence 2 is identical to that of
spacers 3 and 4 except that the residue marked by "***"is replaced by Asp in spacers 3 and 4 (the amino acid sequence of
the last two tandem sequences is not shown; the processing of alpha 3 and 4 happens exactly as shown for alpha 1 and 2).
Proteinase F is believed to be the product of the KEX2 gene, since mutants carrying the lcex2 mutation are devoid of this
enzyme activity and defective in production of active alpha-factor and killer toxin. After proteinase F action, the four
pro-alpha-factor molecules are further processed by enzymatic removal of additional residues. The N-terminal residues
are cleaved off step by step most likely by dipeptidyl aminopeptidase IV (DPAPase) and those at the carboxyl terminus by ~
~
carboxypeptidase alpha (CPase; stage Ill), fmally yielding four mature alpha-factor molecules (stage IV; for complete c:
amino acid sequence of alpha-factor, see text). Both enzymes have been found to be attached to membranes. It should be
noted that these processing intermediates cannot be detected in wild-type cells but only in mutants, i.e., those belonging ~
to the sec family. Recently, synthesis and glycosylation of prepro-alpha-factor have been performed in homologous yeast ~
cell-free extracts (Rothblatt and Meyer, 1986a; Waters and Blobel, 1986). t"'
t=
0
C')
::c
tol
::::
~
~
......
medium is in the range from I0- 7 to I0-8 M. Investigations using mut-

ants (sstl or barl) deficient in degradation of alpha-factor have shown
that the ratio of pheromone concentration and titer of wild-type re-
sponding cells is important for the biological effects elicited.
The genes coding for the a- and alpha-factor have been sequenced
(Brake et al., 1985; Kurjan and Herskowitz, 1982). One alpha-factor
gene codes for a large prepropolypeptide (165 residues) from which
four copies of mature alpha-factor are excised during protein process-
ing (Fig. 1). Another alpha-factor gene codes for a precursor (120 amino
acids) that contains two copies of mature alpha-factor (Singh et al., 1983).
Receptors for pheromones on haploid cells have been inferred from
kinetic and genetic investigations (Jenness et al., 1983). The products of
the STE2 and STEJ genes are believed to be constituents of the receptors
for alpha-factor and a-factor, respectively. This suggestion has recently
been further confirmed (Nakayama et al., 1987; Hagen et al., 1986, and
references therein). The number of alpha-factor binding sites and the
dissociation constant have been estimated to be 8000 sites/cell and 6 x
I0- 9 M, respectively (Jenness et al., 1986).
2.1.5. Glycoproteins
Many of the yeast proteins undergo posttranslational modification
by covalent binding of nonprotein groups, e.g., phosphorylation (Mc-
Donough and Mahler, 1982; Castellanos and Mazon, 1985) and acyla-
tion (Wen and Schlesinger, 1984; Towler and Glaser, 1986). In addition,
about 3% of the membrane protein and less than 2% of the total cell
protein are glycosylated (Tanner, 1984).
Generally, the carbohydrate moiety of yeast glycoproteins is less
variable than those of mammalian cells. Oligosaccharide chains bound to
serine or threonine (0-linked) are not longer than five carbohydrate
residues. Those bound to asparagine (N-linked) are composed of a large
number of mannose units forming branched oligosaccharides, which are
attached via two N-acetylglucosamine residues to the protein backbone.
Synthesis and structure of glycoproteins in yeast and related topics are
covered in reviews (Tanner, 1984; Schekman, 1985; Tanner and Lehle,
11987).
The synthesis mechanism of the 0-linked oligosaccharides in yeast
and other fungi differs from that of mammalian cells in that the first
mannosyl residue directly attached to serine or threonine is transferred
from dolichyl-P-Man in the endoplasmic reticulum (ER). GDP-Man is the
donor for the outer carbohydrate units. However, the synthesis of theN-
linked carbohydrate chains proceeds in essentially the same way as in
mammalian cells. Briefly, a core oligosaccharide of the composition
(GlcNAc) 2 Man~:Pk 3 linked to dolichol via a pyrophosphate bridge is
transferred to a target asparagine residue of a protein chain inserted

into the rough ER. Target asparagines are those in a -Asn-X-Ser/Thr
motif. The large part of this oligosaccharide is assembletl by the subse-
quent transfer of two GlcNAc and five Man residues from nucleotide
donors to dolicholphosphate. This heptasaccharide is then elongated by
attaching four more Man and three Glc units from dolicholphosphate-
Man or ~Glc as donors. Three Glc and one Man residue are cut off rapid-
ly (Byrd et al., 1982) and up to seven mannosyl units may be transferred
after transport of the glycoprotein to the Golgi to yield the modified
core oligosaccharide of yeast mannoproteins (Tanner and Lehle, 1987).
An outer chain consisting of 50 or more Man residues may be attached
to the core region of secretory glycoproteins.
The structure of the carbohydrate moiety of theseS. cerevisiae high-
mannose mannoproteins is shown in Fig. 2 (Tanner and Lehle, 1987).
The assembly of carbohydrate in yeast glycoproteins have been analyzed
using numerous mutants defective in glycoprotein processing (desig-
nated mnn), and structural details have been elucidated by selective
chemical and enzymatic degradations and by nuclear magnetic reso-
nance (NMR) spectroscopy (see, e.g., Byrd et al., 1982; Tsai et al., 1984;
Trimble and Atkinson, 1986, and references therein). This topic is also
covered in reviews (Ballou, 1982; Catley, 1983; Tanner and Lehle, 1987).
Many glycoproteins with enzymatic properties have been described
(Tanner and Lehle, 1987). A few of them have been analyzed in detail
and are discussed below.
Invertase occurs in two mature forms in yeast: one is secreted and
glycosylated; the second, a minor form, is found in the cytoplasm and is
not glycosylated. A single invertase structural gene (SUC2) specifies two
mRNA transcripts for both forms (Carlson and Botstein, 1982). The
larger transcript includes a coding sequence for the signal peptide that
determines the secretion of the glycosylated form of invertase (Carlson et
al., 1983). Saccharomyces cerevisiae external invertase is composed of two
indistinguishable subunits each of mol. wt. about 60,000. Six core-sized
oligosaccharide chains and three polymannose chains ranging in size to
over 50 mannosyl residues are attached to each of the polypeptides,
resulting in a mol. wt. of about 135,000 in a single subunit (Trimble et al.,
1983, and references therein). Multiple forms of external invertase have
been separated by isoelectric focusing. The different charges of the
glycoproteins have been found to be mainly due to variable contents of
mannosylphosphodiester groups in the polymannose outer chains (Fre-
vert and Ballou, 1982). The higher functional stability of the secreted
invertase compared to the internal form has been attributed to the car-
bohydrate moiety, which facilitates refolding of the denaturated poly-
peptide and the formation of enzymatically active oligomers (Chu et al.,
1985).
....
..
0
~
~
~;;!
~
I [
Man -&a.an -&a.an .lSMan - 6 Man - - 6 Man -~ - 6 Man - 6 Man - 6 Man ..!'GicNAc-4 Gic:NAc - Asn
12 12 12 12 12 12 13 13 I ~
Man Man Man Man6- P Man Man Man Man ~ ~
12 12 12 I 13 13 13 12 ii:
Man Man Man Man Man Man Man Man w
~
13 13 13 12 ~
Man Man Man Man
X 13
Man
OUTER CHAIN CORE
Figure 2. Structure of N-linked carbohydrate chains of mannoproteins, according to Tanner and Lehle
(1987). The linkages between both GlcNAc residues, and between GlcNAc and Man are of p type. All other
Man residues are (al-6)-, (al-2)-, or (cd-3)-linked, as indicated here.
In contrast to invertase, yeast carboxypeptidase Y, a vacuolar glyco-

protein with a mol. wt. of about 60,000 and a carbohydrate content of
about 17%, contains only core oligomannose units with an average chain
length of 13 mannosyl residues (Hasilik and Tanner, 1978). Some of
these chains are phosphorylated with all of the phosphate diesterified
(Hashimoto et al., 1981) or in phosphomonoester groups (Schwaiger et
al., 1982).
Only three of the four oligosaccharide chains in the native enzyme
are suspectible to hydrolysis by endoglycosidase H, an enzyme that splits
the chitibiosyl group attached to asparagine in glycoproteins (Tarentino
et al., 1974). The fourth group located at Asn becomes accessible after
denaturation of the glycoprotein with SDS. Only seven of the nine
chains of native yeast external invertase are readily released by endo-
glycosidase H. In both enzymes, all of the carbohydrate-bound phos-
phate is removed with the accessible oligosaccharides. These findings
have been explained by the hypothesis that some of the carbohydrate
chains in both enzymes become inaccessible during glycoprotein matu-
ration to mannose elongation and phosphate addition (Trimble et al.,
1983).
2.2. Polysaccharides, Structural Mannoprotein, and Polyphosphate

Chemically heterogenous compounds are discussed together in this
section because of their structural and functional relationship. Glycogen
and polyphosphate function as storage or reserve substances in the yeast
cell; mannan (now preferentially called mannoprotein), glucan, and
chitin are the main structural constituents of the cell wall.
Polyphosphate is composed of inorganic phosphate in pyrophos-
phate linkages with chain lengths of three to several hundred. Polyphos-
phate is found mainly in the vacuole (Urech et al., 1978; Wiemken et al.,
1979), but has been also detected on the outside of the yeast plasma
membrane by toluidine-binding (Tijssen et al., 1981). In addition to its
function as phosphate reserve in yeast cells, polyphosphate seems to be
involved in the sequestration of basic amino acids in the vacuole (Diirr et
al., 1979).
Glycogen has about the same structure as in mammalian cells (see
Catley, 1983, and references therein). The involvement of this storage
polysaccharide in the carbohydrate metabolism is discussed in Chapter 3
of this volume.
Glucan is one of the main components of the yeast cell wall. De-
pending on the extraction procedure and the yeast strains used, differ-
ent classes of this heterogenous group of polysaccharides have been
described. The major alkali-insoluble component consists of (~ 1-3)-
linked D-glycopyranose units with occasional branching via (~ l-6)-
linked glucose. A mol. wt. of at least 240,000 has been found. A minor
component with a minimum mol. wt. of 22,000 contains mainly glucose
in (13 1-6)-linkages. A polysaccharide with an extended (13 1-3)-glucose
chain having a low degree of branching and side chains with (1-3)- and
(1-6)-linked glucose residues constitutes the main component of the al-
kali-soluble glucan fraction (see Duffus et al., 1982, and Catley, 1983, for
reviews and references).
Glucan synthetase catalyzing the formation of (1-3)-13-o-glucan
from UDP-glucose is probably localized at the inner surface of the plas-
ma membrane (Shematek et al., 1980). Shematek and Cabib ( 1980) have
proposed a complex model for the regulation of this enzyme: GTP and a
heat-labile "supernatant factor" which is phosphorylated by a kinase
activate, while Mg2 + inhibits the glucan synthesis. Mg2 + is a cofactor for
endogenous phosphatases, which dephosphorylate the "supernatant
factor."
A variety of activating as well as inhibitory and so far uncharac-
terized endogenous factors have been extracted from different yeast cell
fractions (Guillen et al., 1985).
Chitin, a linear polysaccharide of (13 1-4)-linked GlcNAc, is found
only in the primary septum and in the bud scars of the cell wall. The
synthesis also proceeds at the inner surface of the plasma membrane and
is strictly regulated by the cell cycle events (Cabib et al., 1984). Two chitin
synthases (called I and II) have been found in yeast (Orlean, 1987).
Chitin synthase I is present in the cell as a zymogen, which can be
converted to the active form by a variety of proteases including trypsin
and yeast protease B (Duran and Cabib, 1978), but the responsible pro-
tease activating in vivo remains to be characterized. Chitin synthase I has
been isolated by trapping it in the soluble chitin formed in the reaction
with UDP-GlcNAc as donor (Kang et al., 1984). Chitin synthase II is
highest in the logarithmic growth phase and can be measured without
protease activation. It has been proposed that chitin synthase II is re-
sponsible for chitin formation in vivo, while chitin synthase I may nor-
mally be inactive (Orlean, 1987). The insertion of the enzyme into the
plasma membrane bilayer (Sentandreu et al., 1984) and the restriction of
the active enzyme(s) to certain sites in the membrane (Cabib et al., 1984)
have been also discussed as possible regulatory factors of chitin deposi-
tion and, consequently, of septum formation.
Mannoprotein is another major structural component of the yeast
cell wall. The bulk mannoprotein is a heterogenous group of pro-
teoglycans with different molecular weights and carbohydrate contents.
More than 60 bands have been separated by SDS-PAGE of SDS-extracts
purified from S. cerevisiae cell walls (Valentin et al., 1984). The protein
content of these compounds usually does not exceed 10%. The carbohy-
drate moiety of cell wall mannoprotein is well analyzed (see Cabib et al.,
1982; Ballou, 1982; Catley, 1983; Tanner and Lehle, 1987, for reviews
and references). It is synthesized via the same lipid intermediates as the
mannoprotein enzymes described earlier. Branched polysaccharide
chains consisting of up to 150 Man units are linked to Asn and short
oligomannan chains to serine or threonine.
In contrast to some mannoprotein enzymes, the protein moieties of
the structural mannoproteins are largely unknown. Only a few elec-
trophoretically homogenous compounds have been isolated. A fraction
called "structural cell wall mannoprotein" has been extracted from S.
cerevisiae mnn9 mutant cell wall (Frevert and Ballou, 1985). This uni-
formly sized (mol. wt. 180,000) mannoprotein consists of 88% carbohy-
drate, which is evenly distributed between asparagine and hydroxyami-
no acids. The protein moiety shows an unusually high proline content of
11%. Attempts to further characterize the protein moiety and to prove
its homogeneity have failed since it has not been possible to remove the
carbohydrate completely without degradation of the protein. Another
glycoprotein with mol. wt. 33,000 which contains only one N-linked
core-sized oligosaccharide has been released from the cell wall by treat-
ment with zymolyase (Pastor et al., 1984), and its secretory pathway has
been analyzed by serological procedures (Sanz et al., 1987). An 0-
glycosylated glycoprotein with mol. wt. 22,000 has been extracted from
the surface of intact cells with mercaptoethanol. Evidence has been pre-
sented that this is a mating-type-specific a-cell agglutinin (Orlean et al.,
1986).
2.3. Lipids
The lipid content of yeast cells can vary considerably as a conse-
quence of a variety of influences. Nevertheless, yeast strains have been
classified in respect to the percentage of lipids in the dry weight as low-
lipid strains (<5%), medium (5-15%), and high-lipid strains (> 15%).
Saccharomyces cerevisiae strains belong to the medium and high-lipid
group (Rattray et al., 1975).
The function and synthesis of lipids as membrane constituents and
the phospholipid composition of S. cerevisiae cells and subcellular mem-
brane fractions have been reviewed (Henry, 1982; Prasad, 1985). A more
general overview on yeast lipids (Rattray et al., 1975) also includes lipid-
~elated substances and minor compounds. In a comparative study, the
lipid composition of 30 species of yeast has been analyzed (Kaneko et al.,
1976). The efficiency of different lipid extraction methods suitable for
yeast cells has been compared. Initial heating in 80% ethanol has been
recommended for complete extraction of polar lipids and inactivation of
lipases (Hanson and Lester, 1980).
An array of glycerophospholipids typical for eukaryotic cells is found
in the group of the polar lipids, comprising phosphatidylcholine, -eth-

anolamine, -monomethylethanolamine, dimethylethanolamine, -glycerol
and -inositol, cardiolipin, phosphatidic acid, and lyso compounds. An
unusual high content of phosphatidylinositol (almost 40% of the com-
plete phospholipid fraction) has been detected in certain mutant strains
(Henry, 1982). Phosphatidylinositol biphosphate is also present in yeast
cells, and evidence has been presented that RAS-encoded GTP-binding
proteins are involved in regulation of the release of inositol monophos-
phate and triphosphate from inositol phospholipids (Kaibuchi et al.,
1986).
Sphingolipids are represented in yeast cells by a group of inositol
phosphoceramides with the structure mannose-(inositolphosphate)._2 -
ceramide. In vitro studies have shown that transfer of phosphoinositol
from phosphatidylinositol is involved in the biosynthesis of these com-
pounds (see Becker and Lester, 1980, for references). C-16 and C-18
fatty acids predominate in the pattern of yeast polar lipids, oleic acid
(18:1) being the major component in many yeasts (Rattray et al., 1975).
Yeast cells are capable of fatty acid elongation and, consequently, the
occurrence of fatty acids with chain lengths of up to C-30 has been
reported (Welch and Burlingame, 1973).
As in all typical eukaryotic cells, phospholipids represent the major
membrane lipid constituents, together with free sterols. Ergosterol is the
main sterol component in yeast cells (Rattray et al., 1975). Sterol esters
are one of the main constituents of lipid particles. These particles also
contain triglycerides and small amounts of free fatty acids, phospholi-
pids, and protein. They function as lipid storage vesicles which may
serve as a reservoir of building blocks for membrane lipids (Clausen et
al., 1974). A significant increase in the content of unsaturated free fatty
acids in baker's yeast cells at the end of the growth phase in batch culture
has been described. Based on the finding that valyl-tRNA synthetase was
inhibited by unsaturated free fatty acids, a regulatory role of these com-
pounds in growth control and protein synthesis has been suggested
(Black, 1985).
The lipid content and the composition of membrane lipids are more
variable in yeast than in any other eukaryotic organism. Factors influenc-
ing lipid composition include carbon source and other nutrients, tem-
perature, and lipid precursors in the medium. Growth at anaerobic
conditions prevents the synthesis of ergosterol and unsaturated fatty
acids. A growing number of mutants defective in synthesis of fatty acids,
phospholipids, and sterols has been isolated (see Henry, 1982, and
Prasad, 1985, for references). Since reliable procedures for isolation of
organelle membranes (e.g., mitochondrial membranes and plasma mem-
brane) are available, the yeast cell represents an excellent subject for
studying the influence of individual lipid components on membrane
structure, function, and biogenesis. Examples of recent work in this field

are the investigation of the influence of sterols on the lipid phase transi-
tions by use of fluorescence polarization (Low et al., 1985 ), the correlation
oflipid synthesis and invertase secretion (Letts and Dawes, 1983), and the
effect of plasma membrane phospholipid unsaturation on the general
amino acid permease (Calderbank et al., 1985). The involvement of lipids
in transport processes and some aspects of the coordination of membrane
phospholipid synthesis have been reviewed (Henry et al., 1984; Prasad
and Rose, 1986).
2.4. Ribonucleic Acids

In this section we refer to nuclear encoded mRNA, rRNA, and
tRNA. Nuclear DNA and related topics are covered in Chapter 9 of this
volume. Some aspects of the mitochondrial nucleic acids are discussed in
Section 3.4.
2.4.1. mRNA
As in other eukaryotes, yeast mRN As are synthesized as precursor
molecules which are posttranscriptionally processed to yield finally ma-
ture translatable mRN As. The monocistronic coding region of mature
yeast mRNA is flanked by noncoding sequences, as in other eukaryotes.
At the 5' end there is a leader sequence with a terminal cap structure
and at the 3' end most mRNA species carry a poly(A) tail of variable
length. The functional significance of leader and trailer sequences of
eukaryotic mRNA has been reviewed by Baralle (1983) and it will not be
discussed here.
Introns appear to be comparatively rare in yeast protein-encoding
genes, except for ribosomal protein genes (see also Section 2.1.3). In-
trons have also been detected in the actin, Matal, TUBJ, TUBJ, K/N28,
and COX4 genes (Gallwitz and Sures, 1980; Ng and Abelson, 1980;
Miller, 1984; Schatz et al., 1986a; Simon et al., 1986; Schneider and
Guarente, 1987). It has been found that the mechanism of yeast pre-
mRNA splicing parallels that in the mammalian system (reviewed by
Padgett et al., 1985, 1986), since formation of lariat RNAs could be
demonstrated during splicing of pre-mRNAs whose mature forms code
for the yeast ribosomal proteins L29 and rp51A and for actin (Newman
et al., 1985; Rodriguez et al., 1984; Domdey et al., 1984). For lariat RNA
formation, a branching site within the intron is important besides the
conserved 5' terminal splice junction. In Saccharomyces this branching
site (UACUAAC), which is located about 20-60 nucleotides upstream of
the 3' splice site, is highly conserved and has been found in all introns
sequenced thus far (Langford and Gallwitz, 1983). In mammals, how-
ever, this branch site does not reveal such a high degree of conservation.
There is considerable evidence that U 1 snRNPs participate in splic-
ing in the mammalian system (Mattaj, 1984). In S. cerevisiae, there are at
least 24 unique snRNAs (Guthrie, 1986). Some of them are clearly non-
essential for growth, in contrast to, at least, the giant LSRI RNA (1175
nucleotides). The LSRI RNA, which contains homologies to vertebrate
U2 and U4-U6 snRNAs (Guthrie, 1986; Ares, 1986, and references
cited therein), has also been suggested to be the large trimethyl-capped
RNA component in the yeast spliceosome (Ares, 1986; Pikielny et at.,
1986; Brody and Abelson, 1985). Besides the LSRI RNA, three different
snRNAs of 160-215 nt in length have been reported also to be present
in the splice complex (Pikielny et al., 1986).
The importance of splicing in gene regulation is currently under
study, and Warner et al. (1985) have suggested that, in Saccharomyces,
certain ribosomal proteins may regulate their own synthesis by binding
to pre-mRNA molecules and inhibiting the splice reaction.
Mature mRNA molecules appear mainly in the cytosolic fraction,
but have also been detected in the rough ER and in the mitochondrial
fractions, respectively (Swida et al., 1982; Ades and Butow, 1980). In the
latter fraction, cytoplasmic polysomes have been found to be associated
with the mitochondrial outer membrane (Ades and Butow, 1980). The
ER-bound mRNA has been demonstrated to be translatable in vitro
(Swida et at., 1982).
The half-lives of mature mRNAs have been estimated by use of the
temperature-sensitive mutant rna] in which the transport of RNA from
the nucleus to the cytoplasm is blocked at the restrictive temperature
(Koch and Friesen, 1979; Chia and McLaughlin, 1979). A typical experi-
mental design permitting an estimate of mRNA half-lives is described
next:
The cellular proteins of rnal are pulse-labeled (e.g., 3 min, 35 S-
methionine) after shift to the nonpermissive temperature. An equal
amount of long-term labeled cells (3 H-methionine) grown at the per-
missive temperature is added to the pulse-labeled cells for purpose of
internal standardization. Subsequently, the proteins are extracted, sub-
mitted to 2-D PAGE, and the radioactivity (35 S: 3 H ratio) of single spots is
determined. This experiment is carried out several times, whereby the
temporal interval between temperature shift and start of the pulse-label-
ing procedure is progressively increased. It can be observed that the de
novo synthesis of individual proteins decreases with time. For calculation
of the mRNA half-lives, it is assumed that the synthesis rate of individual
proteins is proportional to the amount of fuctional mRNA.
The half-lives deduced from these studies have been found to be
within the range from 3.5 min to more than 70 min, with an average
value around 20 min. In similar experiments, Gorenstein and Warner
Table I. Half-Lives of mRNAs Coding

for Ribosomal Proteins as Measured by
Gorenstein and Warner (1976)a
Ribosomal protein Half-life time (min)
S7 10.5
L5 11.1
L6 9.6
rp18 11.1
S13 8.7
rp28 8.5
rp29 11.7
S12 12.2
L22 12.1
S28 8.7
S27 10.2
• Whenever possible, the ribosomal proteins are des-

ignated according to the nomenclature of Kruiswijk
et al. (1978).
(1976) have measured the half-lives of mRNAs of several ribosomal

proteins (Table 1).
2.4.2. rRNA
Each yeast ribosome contains four molecules of rRNA. The small
ribosomal subunit contains a 17S (ISS) rRNA and the large subunit has
three RNA molecules with a size of 5S, 26S (25S), and 5.SS (the values
given in parentheses are alternatively used in the literature). The latter is
associated with the 26S (25S) rRNA by hydrogen bonding. The 26S
(25S) and 17S (ISS) rRNA of Saccharomyces have already been sequenced
(Veldman et al., 19Sla; Georgiev et al., 19S1; Rubtsov et al., 19SO), and
the 5S and 5.SS rRNA sequences of different yeasts and other organisms
have been compiled (Erdmann et al., 19S5). Synthesis and maturation of
rRNAs are quite similar in all eukaryotes. For S. cerevisiae these processes
have been reviewed by Warner (19S2). Hadjiolov (19S5) has made a
detailed survey of the nucleolar events and ribosome biogenesis in the
eukaryotic kingdom. We provide only a short summary of the basic
processes that have been elucidated for S. cerevisiae.
The rDNA is organized in approximately 100-140 repeat units
located on chromosome XII. Transcription of a repeat unit yields two
transcripts. One transcript is the 5S rRNA, which does not undergo
posttranscriptional alterations. The other is a 37S (35S) transcript, which
contains the sequences for 17S (ISS), 5.SS, and 26S (25S) rRNA. This
precursor is methylated (about 60 methyl groups per molecule). Termi-
5' 3'
I C~~~~3:~[E~§~§§§~ 375 transcript
~ t
II I E1 I I
l
18S 29S
t
l
III 131
29S
IV I I I
t
1
185
265
v
175 5.85
Figure 3. Processing of the 37S rRNA primary transcript yielding 26S, 178, and 5.8S
rRNA. This model has been suggested by Veldman et al. (198lb). Although it fits the
available experimental data, direct evidence is missing. Stages 1-V represent a time scale.
Arrows in upward direction indicate cleavage sites.
nal and internal spacer sequences are cleaved off step by step. Thus
intermediates are formed as depicted in Fig. 3. The processing occurs in
the nucleus, but the 18S (20S) precursor, which is already packaged into
a 40S ribonucleoprotein complex, is trimmed at the 3' end in the cyto-
plasm.
2.4.3. tRNA
In Saccharomyces there are about 400 tRNA genes which appear to be
randomly distributed throughout the yeast genome (reviewed by Guthrie
and Abelson, 1982). Mutations yielding suppressor tRNAs have been
observed in Saccharomyces (Guthrie and Abelson, 1982; Sherman, 1982).
The sequences of numerous Saccharomyces tRNA species are known and
are periodically published (Sprinzl et al., 1985).
The tRNA genes code for precursors that are posttranscriptionally
processed. This comprises trimming at the 3' and 5' termini, including
addition of CCA to the 3' end, base modifications, and, in some cases,
intron removal (Guthrie and Abelson, 1982, and references therein).
Approximately .40 yeast tRNA genes contain introns. These are 14-60
nucleotides in length and, in contrast to mRNA introns, do not have

conserved sequences at their boundaries. However, the intron location is
conserved since it always starts one nucleotide to the· 3' side of the
anticodon. An endonuclease and a ligase which are required for the
tRNA splicing have been partially purified (Peebles et al., 1983; Greer et
al., 1983). The following splicing mechanism has been proposed: The
endonuclease, which recognizes the intron sequence, precisely excises it
to yield tRNA halves, one with a 2',3' cyclic phosphate and the other
with a 5' hydroxyl terminus at the cleavage sites. Prior to enzymatic
ligation of the tRNA halves, phosphorylation and activation by adenyla-
tion of the 5' terminus and opening of the 2' ,3' cyclic phosphate termi-
nus to the 2' phosphate are essential intermediate steps. This splicing
mechanism is similar to that found in the wheat germ system but it is
different from that in mammals (for a comparison see Filipowicz and
Gross, 1984; Cech, 1983; Padgett et al., 1986).
3. CELL STRUCTURE
Yeast cells contain a set of subcellular compartments typical of eu-

karyotic cells. Morphologically, or at least functionally, defined organel-
les include the nucleus, mitochondria, endoplasmic reticulum, Golgi
complex, secretory vesicles, and vacuoles. The cell envelope consists of a
rigid cell wall separated from the plasma membrane by the periplasmic
space. Not all of the subcellular organelles are completely independent
from each other, but are specialized structures derived from an ex-
tended intramembranous system (see, e.g., Matile and Wiemken, 1976).
Nevertheless, isolation procedures for intact organelles or at least mem-
brane fragments exist (see Section 5.1 ).
In the subsequent discussion of the cell organelles, the nucleus has
been omitted. However, this topic is covered in Sections 4.2 and 4.3.
3.1. Cell Envelope

The arrangement of the structures in the yeast cell envelope is
schematically shown in Fig. 4. The plasma membrane forms a diffusion
barrier for most low-molecular-weight substances between cytoplasm
and the culture medium. Some enzymes secreted by the yeast cells, typ-
ically glycoproteins with hydrolase activities, do not penetrate the cell
wall and are trapped between plasma membrane and cell wall in the
periplasmic space. The cell wall is mostly composed of mannoprotein
and glucan. Although apparently rigid in structure, it participates in a
number of morphological changes, including budding, mating, and spo-
was
CP
Figure 4. Scheme ofthe yeast cell surface, according to Zlotnik et al. (1984) and Schekman
and Novick (1982). CP, cytoplasm; IB, 08, inner and outer halves, respectively, of the
plasma membrane bilayer; PMP, peripheral membrane protein; IMP, integral meJVbrane
protein; PPE, periplasmic enzyme; PP, periplasmic space; G, glucan; M, mannoprotein;
WOS, wall outer surface; (=) S-S thioester bond; (-)covalent bond.
rulation. The carbohydrate moieties of mannoprotein on its surface de-

termine the immunochemical properties of the yeast cell.
3.1.1. Plasma Membrane

The plasma membrane is presented in Fig. 4 in accordance with the
current membrane model. Intrinsic membrane proteins are inserted in a
lipid bilayer consisting mostly of phospholipids and sterols. Extrinsic
proteins (interacting with membrane lipids and proteins by polar bind-
ing) cover part of the bilayer surface. The carbohydrate moieties of
membrane-bound glycoproteins extend only from the external surface
of the membrane.
The phospholipid composition of the plasma membrane is similar
to the phospholipid pattern found in complete yeast lipid extracts (see
Henry, 1982, for reviews and references), while the sterol content seems
to be higher than in intracellular membranes (Rank et al., 1978). The
lipid composition of yeast cells and their membranes exhibits great varia-
tion and can easily be manipulated. Therefore, yeasts are an ideal model
for studying the properties and functions of 'lipids in eukaryotic cell

membranes and are now being increasingly used for this purpose. Main-
ly the plasma membrane and mitochondria have been analyzed in this
respect. Among other questions, lipid protein interactions, effects of
lipid fluidity and composition on membrane enzymes and transport pro-
cesses, and the fluidity and other physical properties of the lipid matrix
itself in relation to fatty acid composition and sterol content have been
investigated. These topics will not be discussed in detail and, therefore,
the reader is referred to the reviews of Prasad (1985) and Prasad and
Rose (1986).
The proteins of the plasma membrane have been analyzed by high-
resolution 2D-PAGE to find marker polypeptides (Robertson et al., 1980)
or to study the membrane assembly in relation to secretory processes
(Tschopp et al., 1984). The efficiency of different detergents to solubilize
plasma membrane proteins has been compared (Navarette and Serrano,
1983). Currently, only a few enzymes have been released in a functional
state from the plasma membrane and identified in the pattern of the
membrane polypeptides. Some of them are listed in Table II together
with other enzymes assumed to be associated with the plasma membrane
of S. cerevisiae. The numerous permeases of the yeast cell are not in-
cluded. This topic is discussed in Chapter 3 of this volume.
It is important to realize that the procedures for plasma membrane
Table II. Enzymes Associated with the Plasma Membrane of S. cerevisiae
Analytical procedure Reference
Chitin synthetase Purification by entrap- Kang et al., 1984

ment in the reaction
product
Glucan synthetase Activity test Shematek et al., 1980
Proton-translocating Solubilization, purifica- Malpartida and Serrano,
ATPase tion, reconstitution 1981a
Phosphorylated ATPase- Solubilization and purifi· Malpartida and Serrano,
intermediate cation 1981b
Tyrosine kinase Activity test and separa- Castellanos and Mazon,
tion of phosphopro- 1985
teins
Adeny1ate cyclase Activity test, inhibition by Liao and Thorner, 1980
a-factor
Phospholipase B and Solubilization and purifi· Witt et al., 1984
lysophospholipase cation
NAD(P)H ferricyanide Activity test, difference Ramirez et al., 1984
reductase, flavins spectroscopy
Lipid-mediated glycosy- Activity test Welten-Verstegen et al.,
lation of proteins 1980
separation are not perfect and that contamination from organelle mem-
branes cannot be excluded, leading to erroneous results in respect to
subcellular distribution of enzyme activities. One of the reasons is that
the endoplasmic reticulum is closely associated with the plasma mem-
brane and connections between both membranes may exist (Schekman
and Novick, 1982).
Plasma membrane fractions isolated by different procedures may
vary greatly in protein and lipid composition (see Section 5). This may be
due to artifacts of preparation, but it has also been explained by the
assumption that certain membrane domains with characteristic proper-
ties coexist in the plasma membrane (Rank et al., 1978). Membrane frac-
tions with different protein patterns and densities have been separated
(Tschopp and Schekman, 1983). Evidence for a laterally heterologous
distribution of membrane components has also been gained by freeze-
fracture electron microscopy, e.g., particles detectable on the cytoplas-
mic surface of the membrane bilayer arranged in semicrystalline arrays
when the cells are grown to stationary phase (Steere et al., 1980; Kramer
et al., 1978).
A growing field of research is evaluation of the processes involved
in the plasma membrane assembly using mutant strains defective in lipid
biosynthesis or secretion (e.g., Tschopp et al., 1984; Ramirez et al., 1983;
Ferro-Novick et al., 1984; Klig et al., 1985; Letts and Henry, 1985). Both
topics have been reviewed (Henry, 1982; Schekman and Novick, 1982;
Novick, 1985; Schekman, 1985). Apparently, cell surface proteins in the
plasma membrane, as well as in the cell wall, are transported to the cell
periphery by the same or very similar mechanisms as secreted proteins
(see Novick, 1985, for references).
3.1.2. Cell Wall

The cell wall structure and the synthesis and assembly of its constit-
uents as a simple and versatile model for morphogenesis have been
studied intensively. A variety of methods have been used for this pur-
pose, including separation and chemical analysis of wall constituents;
serological procedures; binding of dyes, antibodies, and lectins; selective
degradation by enzymes; cell wall mutants; fungicides and other antibi-
otics (e.g., tunicamycin); and electron microscopic procedures. These
topics have been reviewed (Ballou, 1982; Cabib et al., 1982, 1984; Schek-
man and Novick, 1982; Sentandreu et al., 1984; Catley, 1983).
Glucan and mannoprotein, the main structural constituents of the
S. cerevi.siae cell wall, are found in roughly equal amounts. Chitin is
almost entirely restricted to the bud scar (Cabib et al., 1982, 1984) and
does not amount to more than about 4% of the cell wall material
(Roberts et al., 1983). Lipid contents between 3% and 9% have been
reported (see Rattray et al., 1975, for review and references). Glucan is
regarded as the main structural determinant component of the wall

(Cabib et al., 1982; Gatley, 1983). The shape of(~ 1-3)glucan has been
described as a long, twisted chain with a hydrophilic side due to hydroxyl
groups and a hydrophobic side due to methine groups. The interaction
of the hydrophobic surfaces is leading to a double helix with very high
rigidity (Ballou, 1982). At least a part of the cell wall glucan seems to be
arranged in a microfibrillar network. Chitin is polymorphic and exists in
three crystalline forms called a, ~. and -y-chitin. a-Chitin, the form with
maximal hydrophobic binding, has been detected in yeast bud scars (see
Ballou, 1982; Cabib et al., 1982; Gatley, 1983, for details and references).
Manno proteins do not seem to form any crystalline or fibrillar structures
in the yeast cell wall (Cabib et al., 1982).
The precise arrangement of these components in the cell wall of
Saccharomyces is not known, but the available experimental evidence has
been combined in a model (Zlotnik et al., 1984): the outer layer of the
wall consists of mannoproteins linked to each other by S-S or thioester
bonds and hydrophobic interaction, and linked by covalent bonds of
unknown nature to glucan, which forms the inner cell wall layer. Both
layers are not completely separated; mannoproteins may penetrate the
glucan layer to a certain, but unknown extent. Selective degradation
with proteases has shown that the mannoproteins in the external layer
determine the cell wall porosity (Zlotnik et al., 1984).
3.2. Vacuoles
In contrast to other organelles such as mitochondria, vacuoles do
not represent a clearly separated organelle fraction, but are part of an
intramembranous system including the endoplasmic reticulum and
other intracellular spaces separated by membranes from the cytoplasm.
Therefore, it has been recommended that the term vacuole be used in a
purely descriptive sense (Matile and Wiemken, 1976; Matile, 1978).
Form and size of the vacuole may change considerably during the
cell cycle. Large vacuoles, which represent by far the most voluminous
organelles of the yeast cell, shrink and fragment into small vesicles at the
stage of bud initiation. These vesicles fuse to form one or two large
vacuoles when the budding process proceeds (Wiemken et al., 1970). At
this stage, relatively pure and undamaged vacuoles can be isolated whose
functions and structural properties have been investigated.
Many proteases, including carboxypeptidase Y and proteases A and
B, and several other hydrolases, have been detected in the yeast vacuole.
It has been claimed that many of these enzymes are exclusively located in
the vacuolar space or at least absent from the cytoplasm (Wiemken et al.,
1979). The characteristic content of proteases and other hydrolases has
led to the suggestion that vacuoles play a similar role in yeast cells to that
of lysosomes in animal cells (see Matile, 1978, for references).
The second principal function of yeast vacuoles is storage of metab-

olites. One-fourth of the amino acid pool and an even higher percentage
of arginine has been found in the vacuolar fraction of S. cerevisiae pro-
toplasts (Wiemken et al., 1979). The role of vacuolar polyphosphate in
trapping basic amino acids has already been mentioned in Section 2.2.
Polyphosphate has been found to be located exclusively in the vacuolar
sap of protoplasts and has been used as a marker during the preparation
of vacuoles (Wiemken et al., 1979).
Right-side-out vacuolar membranes have been used to demonstrate
an active transport of arginine driven by an electrochemical potential of
protons generated by ATP hydrolysis. A potential of 180 mV, with con-
tribution of 1. 7 pH units, interior acid, has been determined. The
Mg2 +-activated, proton-translocating ATPase of the vacuolar mem-
brane has clearly different properties compared to the mitochondrial
and plasma membrane ATPases and thus may also serve as vacuolar
marker enzyme (Kakinuma et al., 1981, and references therein).
The chemical composition of the vacuolar membrane (tonoplast)
has also been studied in some detail. In comparative investigations of
tonoplast and plasma membrane from S. cerevisiae, a higher content of
phospholipids and unsaturated fatty acids has been found along with a
lower sterol content in the vacuolar membrane. Differences in the pro-
tein-especially the glycoprotein-pattern have also been described.
These findings have been discussed in relation to the higher "elasticity"
of the tonoplast in comparison to the plasma membrane (Kramer et al.,
1978; Niedermeyer, 1976; Schwencke, 1977). Regions of the vacuolar
membrane depleted of intramembranous particles have been detected
by electron microscopy in freeze-fractured tonoplasts isolated from cells
entering the stationary phase. The lipids of these vacuoles have been
extracted and analyzed by differential scanning calorimetry, and an en-
dothermic transition around the growth temperature has been found.
These results have been interpreted as indication for patches of gel
phase lipid in the tonoplast, which exclude intramembranous particles
(Moeller et al., 1981).
The transversally asymmetrical distribution of tonoplast carbohy-
drate, probably mannan chains of membrane glycoproteins, has been
analyzed by concanavalin A binding. Carbohydrate is bound to the
membrane surface opposite to the cytoplasm in both tonoplast and plas-
ma membrane (Boller et al., 1976).
3.3. Endoplasmic Reticulum and Secretory and Endocytic Apparatus

When thin sections of whole wild-type Saccharomyces cells are in-
spected electron microscopically for secretory organelles, vesicles that
obviously function in secretion can be unequivocally detected in the
growth region (see also Section 4.3). The Golgi seems to be totally absent,
and normally only a few profiles may account for the ER. Our present
knowledge about the secretory apparatus and secretion in yeast comes
almost exclusively from studies with several conditionally lethal, temper-
ature-sensitive mutants of the sec family defective in different steps of
the secretory pathway, but not in overall protein synthesis (reviewed by
Schekman, 1985; Novick, 1985; Ferro-Novick, 1985).
Tracing of the intracellular level of secretory enzymes (e.g., inver-
tase, acidic phosphatase) has revealed that sec mutants can be divided
into two classes. Mutants of both classes are blocked in protein export,
but members of class A accumulate intracellular!y active secretory en-
zymes whereas those of class B do not. Later it was found that class A
and class B mutants are blocked in secretion at late and early steps,
respectively.
The organelles and the order of their participation in secretion have
been elucidated by double mutant analysis, by histochemical means, and
by inspection of the phenotype of different sec mutants at the ultrastruc-
tural level. The latter studies have revealed that due to the block in
secretion, certain mutants accumulate ER, others Golgi structures, and
again others vesicles (for electron micrographs, see Novick et al., 1980,
1983; Esmon et al., 1981). It is now firmly established that in Sac-
charomyces the general pathway of secretion is in the order ER--+ Golgi--+
vesicles --+ cell surface, and the functions of ER and Golgi in glycosyla-
tion are well documented (see also Section 2.1.5). This pathway is similar
to that in mammals (Walter and Lingappa, 1986). However, in contrast
to the mammalian ER, the existence of a receptor for the signal recogni-
tion protein (Meyer, 1982) has not been proved for yeast ER. In a ho-
mologous yeast cell-free system, the prepro-alpha-factor has been dem-
onstrated to cross ER membranes uncoupled from translation in an
ATP-dependent fashion, whereas other secretory proteins tested did not
reveal this behavior (Rothblatt and Meyer, 1986b; Waters and Blobel,
1986; for review, see Zimmermann and Meyer, 1986). Whether post-
translational translocation also happens in vivo remains to be established.
Later steps in secretion have also been studied in cell-free systems.
Glycosylation and transfer of invertase from ER to Golgi have been
performed in vitro (Haselbeck and Schekman, 1986).
Recently, evidence has been obtained that endocytosis occurs in
yeast (Riezman, 1985; Makarow, 1985a,b; for a recent review, see
Riezman et al., 1986). Riezman (1985) has shown that the fluorescent dye
lucifer yellow carbohydrazide (LY) is taken up from the medium and
accumulated in the vacuole of S. cerevisiae. This happens in an energy-
and temperature-dependent, but not saturable, fashion. Certain sec mu-
tants (class A) which accumulates vesicles have been demonstrated to be
defective in these functions. Thus gene products required in the late
steps of secretion may also be essential for endocytosis.
Two mutants defective in the endocytic pathway have been isolated
(Chvatchko et al., 1986). The endl mutant is defective in the internaliza-

tion of LY and alpha-factor, whereas end2, most probably, is unable to
transport the endocytic contents into the vacuole. These studies and
those from another laboratory suggest that alpha-factor uptake occurs
via receptor-mediated endocytosis (Jenness and Spartick, 1986).
Makarow ( 1985a) has reported the internalization of enveloped
viruses into spheroplasts. Moreover, bacterial alpha-amylase and FITC-
dextran, which are taken up even by intact cells in a temperature-
dependent fashion, have been shown to be transported into vacuoles
(Makarow, 1985b). Studies with FITC-dextran have revealed that lack
of ATP does not inhibit internalization of substrate but inhibits its trans-
fer from an internal intermediate compartment into the vacuole. This in-
termediate compartment exhibits parallels to the mammalian endosomes
(Makarow and Nevalainen, 1987).
Coated vesicle and clathrin triskelions have been identified in yeast
by electron microscopy but their intracellular source is still unknown
(Mueller and Branton, 1984). Moreover, clathrin seems not to be re-
quired for cell growth and protein secretion (Payne and Schekman,
1985).
3.4. Mitochondria
S. cerevisiae is a facultative anaerobe, which means it possesses the
property to switch from respiration to fermentation, and vice versa.
Mitochondria of fully respiring cells (aerobic growth on a nonfermenta-
ble carbon source) are rich in cristae structures. Under these conditions,
50 or more mitochondria per cell have been observed (Stevens, 1981).
Glucose-repressed cells contain only a few mitochondria which are poor
in cristae and whose respiratory enzymes exhibit only low activity (Ste-
vens, 1981). A large number of respiratory-deficient mutants (petites),
which are viable when grown on a fermentable carbon source, are known.
These mutants have stimulated much research on yeast mitochondrial
genetics, gene expression, and organelle biogenesis (for reviews, see
Stevens, 1981; Dujon, 1981; Rosamond, 1982; Schatz and Butow, 1983;
Hay et al., 1984; Douglas and Takeda, 1985; Trivedi et al., 1985; see also
the volume Mitochondria '83, in which Yaffe, 1983, has given a summary of
protein import into mitochondria). We will restrict ourselves to a descrip-
tion of the major mitochondrial components. Any further information is
enclosed in the reviews cited.
Mitochondria are formed by an inner and outer membrane. The
former may be strongly evaginated into the matrix space to form cristae.
Both membranes are rich in phospholipids (Henry, 1982). Cardiolipin
(diphosphatidylglycerol) is prominent in. the inner membrane (Prasad,
1985). The sterol:phospholipid ratio is about 1:30 in both membranes
(Bottema and Parks, 1980).
Many proteins and enzymes are associated with the mitochondrial

membranes (Dujon, 1981). Some of them are used as marker enzymes
allowing discrimination between outer and inner mitochondrial mem-
brane (see also Table III). Most mitochondrial proteins (approximately
95%) are encoded by nuclear genes, synthesized on free cytoplasmic
polysomes, and finally imported into mitochondria (a detailed descrip-
Table III. Marker Enzymes for Yeast Subcellular Compartments
Organelle or
compartment Marker References for assays
Cytoplasm Glucose-6-P dehydrogenase Kato et al., 1979

Vacuole
Membrane a-Mannosidase van der Wilden et al., 1973
Vacuolar sap Protease A Wiemken et al., 1979
Protease B
Carboxypeptidase Y
Endoplasmic NADPH: cytochrome c Schatz and Klima, 1964
reticulum Oxidoreductase Dow et al., 1981
UDP-GicNAc-N-acetylglu- Lehle and Tanner, 1978
cosaminyl dolichylphos-
phate transfer reactions
Golgi Mannoprotein outer-chain Nakajima and Ballou, 1975
oligosaccharide man-
nosyltransferase
Nucleus
Nucleoplasm DNA-dependent RNA Jerome and Jaehning,
polymerase 1986
Aspartate transcar- Nagy et al., 1982
bamoylase
Nuclear envelope Identification by electron Mann and Mecke, 1982
microscopy
Mitochondria
Matrix space Aconitase Racker, 1950
Fumarase
Intermembrane Cytochrome c Djavadi-Ohaniance et al.,
space Peroxidase 1978
Flavocytochrome b2 Appleby and Morton, 1959
Inner membrane Cytochrome c oxidase Mason et al., 1973
Outer membrane Kynurenine hydroxylase Bandlow, 1972
Porin Daum et al., 1982
Plasma membrane Vanadate-sensitive Borst-Pauwels and Peters,
Mg2+-ATPase 1981
Tschopp and Scheckman,
1983
Cell wall
Periplasmic space Invertase Goldstein and Lampen,
1975
Secretory vesicles Acidic phosphatase Haguenauer-Tsapis and
Hinnen, 1984
tion of this process is given by Schatz and Butow, 1983; Yaffe, 1983; Hay
et al., 1984; Douglas and Takeda, 1985; the latter two references contain
tables with imported proteins studied thus far).
Mitochondria have their own translational machinery. Synthesis of
mitochondrial proteins with high fidelity has been demonstrated in vitro
with isolated mitochondria (McKee et al., 1984). The components of the
mitochondrial protein synthesis apparatus coded by mitochondrial
genes are the two rRNAs (15S and 21S, constituents of the small and
large ribosomal subunits, respectively), a complete set of tRNAs (25
genes), and a few mRNA species. The mitochondrial ribosomal proteins,
which are all coded by nuclear genes (except for the [varl] polypeptide,
which seems to be associated with the small ribosomal subunit), have
been analyzed (Faye and Sor, 1977). The small ribosomal subunit (37S)
contains 33 proteins (mol. wt. ranging from 9500 to 60,000) and the
large subunit (50S) has 38 proteins (mol. wt. ranging from 10,000 to
41 ,000). The ribosomal proteins derived from mitochondria are less
basic than those from the cytoplasm.
The circular mitochondrial DNA (75,000 base pairs, 25 f.Lm in cir-
cumference), which is extremely rich in AT base pairs (82%), codes for
only a few polypeptides, certainly no more than 15: subunits I, II, and
III of cytochrome c oxidase; subunits 6 and 9 of the ATPase complex;
apocytochrome b; [varl] polypeptide and at least eight polypeptides as
inferred from genetic data, e.g., maturases required for intron removal.
The physical identification of two maturases coded by the cob-box re-
gion of the mitochondrial genome has been reported (Jacq et al., 1984;
Guiso et al., 1984).
4. LIFE CYCLE
Saccharomyces is able to grow vegetatively as either a haploid or

diploid and many industrial strains have a polyploid or aneuploid chro-
mosome constitution. In the normal sexual cycle a diploid cell results
from a fusion of two haploid cells of opposite mating type (alpha and a).
This diploid cell, which is now heterozygous with respect to mating type
(alpha/a), has lost the ability to mate and neither secretes mating factors
nor responds to them, but it undergoes meiosis when exposed to certain
starvation conditions (e.g., lack of nitrogen) and forms an ascus which
normally contains four haploid spores, two of each mating type. Trans-
fer of the spores to nutrient medium results in germination and rapid
fusion within the ascus to form two diploid alpha/a-cells. Frequently,
single germinating spores from heterothallic strains escape from the
ascus and give rise to colonies of vegetatively growing haploid cells when
kept isolated from cells of opposite mating type. The vegetative growth
is characterized by budding and mitosis.
Cell cycle events and those just mentioned are accompanied by
changes in cell structure which are described below.
4.1. Mating
For mating, haploid cells have to be conditioned by mating factors.
In response to the action of mating factors, an increase of agglutination
can be microscopically observed, normally about 30-60 min after mix-
ing the cells of opposite mating type. Increase of agglutination happens
not only in G1 of the cell cycle. By about 90-120 min, large cell aggre-
gates are formed from which single unbudded cells arrested at G 1 can
be released by sonication. In addition, some free elongated pear-shaped
cells (shmoos) appear. The occurrence of shmoos on solid medium has
been used as a bioassay to estimate the concentration of pheromone in
the medium (Duntze et al., 1970; Betz et al., 1977). Mating happens only
between cells arrested in G 1. This behavior of mating cells requires
alterations mainly at the cell surface (reviewed by Sprague et al., 1983;
Thorner, 1981; Betz et al., 1981) which are summarized below.
In most strains the drastic increase of agglutination is a function
inducible by the opposite mating factor. However, there are also strains
that are constitutive in this respect when grown at 23°-26°C, but behave
like inducible strains when grown around 36°C. From such constitutive
strains, sex-specific agglutination substances have been extracted (To-
hoyama et al., 1979; Yamaguchi et al., 1984). Obviously, these agglutina-
tion substances, which turned out to be glycoproteins, are only found in
a functional state on the cell surface of haploids and zygotes, but not on
diploids and spores (Tohoyama et al., 1979; Tohoyama and Yanagishima,
1981 ). The agglutination substance extracted from a-cells could mask
alpha-cells, thus inhibiting their normal property to agglutinate with a-
cells; the opposite is also true. The biological activity of the sex-specific
agglutination substance derived from a-cells can be abolished by mer-
captoethanol, which has little effect on the agglutination substance de-
rived from alpha-cells.
Increased agglutination as a result of pheromone induction re-
quires the presence of both a carbon and a nitrogen source and it is also
dependent on RNA and protein synthesis, but, interestingly, not on the
intact glycosylation pathway, since inhibition of glycosylation by drugs
(e.g., tunicamycin) does not inhibit alpha-factor-induced agglutination.
In response to mating factor, the synthesis of cell wall carbohydrates
changes such that the glucan/mannan ratio increases. Furthermore,
shorter oligosaccharide side chains are more frequently found in cell
wall mannan (Lipke et al., 1976).
Cells treated with mating factors become elongated and form tips
which function as a conjugation tube. The cell wall of this tip accumu-
lates chitin, a minor component of the vegetative cell wall. Accordingly,
the chitin synthetase activity of the plasma membrane fraction from
alpha-factor-treated a-cells increases (Schekman and Brawley, 1979).
However, inhibition of chitin synthetase by Polyoxin D (500 J.Lg/ml) fails
to block morphogenesis (Lipke et al., 1976). Bud scars-wall structures
rich in chitin-seem to influence the mating efficiency since cells with a
high number of scars show reduced mating potential. The reason may
be that tip formation, like bud initiation, is impeded by too many scars.
In electron micrographs, the cell wall appears diffuse after alpha-
factor exposure and vesicles have been shown to accumulate in the tip
region (Lipke et al., 1976; Betz et al., 1981 ). These vesicles are believed to
transport lytic enzymes capable of digesting the plasma membrane and
cell wall between conjugating cells, thus allowing cytoplasmic continuity.
Very little is known about the nature of these enzymes and their regula-
tion. En do- and exo-1 ,3-(3-glucanases, which might digest the cell wall,
have been reported to be produced in Saccharomyces constitutively, but
mutants deficient in the exoglucanase activity have not lost the ability to
mate (Del Rey et al., 1979).
The biochemical processes involved in nuclear fusion are largely
unknown. It has been observed that nuclear fusion is initiated at the
spindle pole bodies (SPB), which eventually yield a single, large SPB.
Regulation of the cell recognition process is also not fully under-
stood. It has been proposed that the mating sites, which seem to be
determined by the cytoplasmic microtubules, have to be in proper orien-
tation to each other. This would strongly diminish the number of pos-
sible mating partners within mating aggregates (Sprague et al., 1983)
and would explain that mating between more than two cells of opposite
mating type rarely occurs (approximately one per 104 zygotes formed).
The intracellular signal triggered by the pheromones is another
problem that needs further investigation. It is known that the intra-
cellular level of cyclic AMP (cAMP) is lowered most likely due to the
inhibition of the membrane-bound cAMP synthetase (Liao and Thorner,
1980). Any further discussion is beyond the scope of this chapter. We
therefore refer to the review by Thorner ( 1982).
4.2. Meiosis and Sporulation

When diploid alpha/a-cells are shifted to nitrogen-free acetate me-
dium (sporulation medium), they undergo meiosis and form ascospores
(for media and procedures, see Haber and Halvorson, 1975). The cytol-
ogy and the metabolic processes occurring during meiosis have been
reviewed by Byers (1981) and by Esposito and Klapholz (1981). When
cells are shifted to sporulation medium, they will complete their cell
cycle to G 1 stage and enter the meiotic pathway. As the sporulation med-
ium seems to depress bud growth, a considerable asynchrony of spore
formation is caused by daughter cells derived from those mother cells
which have started budding just at the time of medium shift.
Asynchrony can be reduced by the use of stationary cells in G 1 stage;
however, this may diminish the efficiency of sporulation.
At the ultrastructural level, the progress of meiosis has been exten-
sively studied (e.g., Guth et al., 1972; Byers and Goetsch, 1975a). In
general, the early events of yeast meiosis are similar to those in other
eukaryotes (Fig. 5). Before cells enter the meiotic prophase, the DNA is
replicated (premeiotic S-phase) about 4-10 hr after medium shift. By
the time of DNA replication, the SPB, which is localized at the nuclear
envelope, is duplicated yet remains paired throughout the meiotic pro-
phase (Fig. 5a,b; Moens and Rapport, 1971). Prior to the onset of pre-
meiotic S-phase, dense material, termed dense body (DB; Fig. 5b), ap-
pears in the nucleus. In the DB, central elements of synaptonemal
complexes (SC) are frequently visible. At leptotene, unpaired lateral
elements and in zygotene short SCs become detectable (Zickler and
Olson, 1975). At pachytene, the SCs reach maximum length and DB
structures have vanished or have shrunken (Fig. 5c). At this stage, the
yeast hryotype is accessible to analysis (the yeast karyotype has been also
resolved by gel electrophoresis; see Schwartz and Cantor, 1984). SC
inspection normally reveals approximately 17 bivalents (in diploids),
which coincides with the number of known genetic linkage groups
SPS
sc
DB
a b c d
g h
Figure 5. Stages in yeast meiosis. For details, see text.

(Byers and Goetsch, 1975a). Classical condensation of chromosomes has

never been unequivocally demonstrated in either meiosis or mitosis.
In diplotene, a single polycomplex body (PCB) is formed and indi-
vidual SCs are no longer visible (Fig. 5d). At about this time, the SPBs
separate and a spindle is formed which spreads between the SPBs. When
the spindle is maximally extended, the SPBs are in opposition and their
cytoplasmic sides-the outer plaques-enlarge (Fig. 5e) and the PCB is
no longer detectable. This also indicates the final stage of meiosis I. Cells
stained for DNA appear binucleated under the light microscope.
Subsequently, each of the SPBs duplicates. The SPB pairs separate
by moving along the still intact nuclear envelope, which has now gained
a lobed shape (Moens and Rapport, 1971). Two spindles are formed
each running between two SPBs which are finally located in opposition
(Fig. 5f,g). The prospore wall develops by coalescence of vesicles at the
cytoplasmic surface of the outer plaques of the SPBs (Fig. 5g). The
prospore wall initially clings to the nuclear membrane during enlarge-
ment but later both structures are found separated. The space between
the developing prospore wall and the nuclear envelope fills up with
cytoplasmic material, including mitochondrial particles. When the sec-
ond nuclear division is complete, the spindles disintegrate, the pros pore
wall closes, and individual ascospores become visible. The wall growth
continues by deposition of material between the two unit membranes of
the spore wall (Fig. 5h) and the inner layer of the spore wall becomes the
plasmalemma of the ascospore. The wall of the mature ascospore con-
tains glucan and mannan. Studies with mutants auxotrophic for glucosa-
mine suggest that the outer layer of the ascosporal wall is enriched in
glucosamine. This is probably deposited as chitin and may be the cause
of the hydrophobicity of the spore surface. In contrast to normal spores,
such mutants are sensitive to ~-glucanases (Ballou et al., 1977).
When cells have entered the meiotic pathway, only a single mito-
chondrion per cell is detectable by electron microscopy (Stevens, 1981 ).
When the meiosis is complete, each spore within an ascus normally con-
tains a single mitochondrion. However, this does not represent a com-
plete packing of the organelle into the spores. Quantitative evaluation of
thin sections has revealed that, on a volume basis, only 52% of the
mitochondrial material, 32% of the cytoplasm, 10% of the vacuole(s),
and 96% of the nucleus are enveloped into spores; the rest remains in
the ascal epiplasm (Brewer and Fangman, 1980). Since the spores occupy
about 30% of the total ascal space, only the cytoplasm is randomly enve-
loped and all other organelles mentioned are nonrandomly enveloped.
4.3. Vegetative Growth

In nutrient medium Saccharomyces cells grow by budding. The emer-
gence of a bud indicates that a cell has entered a new division cycle.
@'"'"""
Figure 6. The cell cycle of Saccharomyces cerevisiae. For details, see text.
During the cell cycle, there is an increase in cellular components, includ-

ing duplication of genetic material. Increase in total cell mass and cell
volume is almost exclusively due to bud growth. The segregation of
chromosomes happens mitotically within an intact nuclear membrane
(Fig. 6). After completion of cytokinesis and formation of wall septa, cell
separation gives rise to a daughter cell with all the cellular components
and functions necessary for vegetative growth. An ordered sequence of
morphological changes during mitotic division has been defined by ex-
tensive electron microscopic studies (Byers and Goetsch, 197 5b; Peter-
son and Ris, 1976; Byers, 1981). In the following discussion of changes
in cell structure during cell cycle, we lay stress on the involvement of
microtubules and actin filaments whose role has been investigated by
immunochemical methods in mutant and wild-type strains of S. cerevisiae
and S. uvarum* (Kilmartin and Adams, 1984; Adams and Pringle, 1984;
Novick and Botstein, 1985).
*Taxonomists agree that S. cerevisiae and S. uvarum are identical. Kilmartin and Adams
(1984), however, have emphasized that their S. uvarum strain was more suitable for the
cytoimmmunochemical studies because of its more rodlike shape compared to S.
cerevisiae.
Budding occurs only in some yeasts, but not in other eukaryotes. A

bud is formed by a localized outgrowth of the cell wall at an obviously
nonrandom site (see below). Electron microscopy has revealed that the
SPB embedded in the nuclear envelope is duplicated as soon as a bud
emerges, but remains paired until the bud has grown out to about one-
third in diameter of the mother cell (reviewed by Byers, 1981; King and
Hyams, 1982). Two subsets of microtubules emanate from the SPBs,
designated as (1) nucleoplasmic and (2) cytoplasmic microtubules ac-
cording to their subcellular distribution. The cytoplasmic microtubules
are orientated to the budding region of the cell wall, which overlays the
paired SPBs. The cytoplasmic microtubules may be involved in the lo-
calization of vesicles detectable near the budding site. These vesicle are
thought to be engaged in the remodeling of the cell wall. The nucleo-
plasmic microtubules form the mitotic spindle. In haploid cells, initially
5-10 continuous microtubules interconnect the now separated SPBs. In
addition, 17 discontinuous nucleoplasmic microtubules (per SPB) are
detectable. In diploids 34 discontinuous microtubules per SPB are seen
while the same number of continuous microtubules as in haploids (5-1 0)
are observed (King and Hyams, 1982; Peterson and Ris, 1976). The
number of the discontinuous microtubules is coincident with the
number of chromosomes. Therefore, it is generally assumed that the
discontinuous microtubules attach to the centromeres of the chromo-
somes, although there is no direct ultrastructural evidence for this. The
number of continuous microtubules decreases as the cell proceeds in its
division cycle until only one microtubule along with the nuclear axis with
a length of about 8 ..-.m is left (King et al., 1982).
A filamentous ring in the neck region develops soon after bud
emergence. It persists for most of the cell cycle but finally disappears at
cytokinesis (Byers and Goetsch, 1976). The function and chemical
nature of these filaments (10 nm in diameter) are still unknown. Figure
6 shows schematically the localization of the SPBs, the orientation of the
mitotic spindle, and the movement of the nucleus during bud enlarge-
ment, based on electron microscopic evidence.
Immunofluorescence studies have extended knowledge on the lo-
calization of tubulin and actin during the cell cycle (Kilmartin and
Adams, 1984; Adams and Pringle, 1984; Novick and Botstein, 1985).
Immunofluorescent labeling of tubulin revealed that the SPBs nucleate
microtubules throughout the cell cycle. In the period during spindle
completion, and before movement of the elongated spindle into the
enlarged bud, long bundles of cytoplasmic microtubules emanate from
both SPBs. One bundle of cytoplasmic microtubules always extends to-
ward or into the bud.
When actin is labeled by fluorescent antibodies or fluorescent phal-
loidin conjugate (which binds to F-actin with high affinity), a ring of actin
dots on apparently unbudded cells, and around the base of small buds,
becomes visible besides faintly stained filaments obviously connected

with the dots. The development of the actin dot ring is not dependent on
the presence of the chitin ring which is formed at the budding site of the
mother cell and remains on the mother cell as a bud scar after cell separ-
ation (see also Section 3.1.2). During spindle elongation, actin dots are
found primarily in the tip region of the growing bud. The bud tip has
been shown to be the growth region (Adams and Pringle, 1984). When
the spindle is fully elongated, actin dots are almost equally distributed
between mother cell and bud. In addition, filamentous structures have
been observed. By the time of cytokinesis, actin dots appear to be con-
centrated in the neck region on both mother cell and bud. Actin dots in
other parts of the cell are also visible (see also scheme depicted in Fig. 6).
The actin dot rings which appear at the beginning and end of the cell
cycle most likely do not correspond to the filamentous ring in the neck
observed electron microscopically (see above), since the filamentous ring
has been found to be incompletely developed or totally absent in those
stages when the actin dot rings have been identified. There seem to be
no pronounced interactions between microtubules and actin filaments as
judged by double fluorescent labeling, but it cannot fully be ruled out
that thin interacting filaments have not been resolved by this method.
The role of actin during the cell cycle is not fully understood, but
there is some evidence that actin filaments may take part in the localiza-
tion of secretory vesicles into the growth region of the bud. Tempera-
ture-sensitive actin mutants are partially blocked in secrection of inver-
tase at the restricted temperature. The secreted form of invertase is
known to be transported to the cell surface in vesicles (see also Section
2.1.5). Similarly, the deposit of chitin (bud scar) may also result from the
participation of actin filaments in directing the flow of vesicle containing
chitin synthetase (zymogen) to the budding site. Vesicular transport of
chitin synthetase (zymogen) was suggested several years ago to be re-
sponsible for the incorporation of chitin into the growth zone of alpha-
factor-treated a-cells (Schekman and Brawley, 1979; see also Section 4.1).
This view is further supported by the finding that the actin mutants
mentioned earlier fail to deposit chitin in a localized way at the nonper-
missive temperature; instead chitin appears to be distributed over most
of the cell surface (Novick and Botstein, 1985). Other interpretations are
possible, but there is strong evidence that actin plays a major role in
polarized growth and cell surface organization. Further investigations
are necessary to confirm the suggested functions.
5. CELL FRACTIONATION
The rigid cell wall of yeast is the main obstacle to the preparation of
cell-free extracts and organelles. The disintegration of the wall can be
achieved mechanically or by enzymatic digestion of its constituents. The

drastic treatment required to disrupt yeast cells by mechanical means
results in severe artifacts; organelles are damaged, and membrane frag-
ments of different organelles may fuse or form enclosures. These effects
may interfere with fractionation and purification of these organelle's and
membranes.
The enzymatic digestion of wall components, while cell integrity is
maintained by an appropriate osmotic stabilizer, results in the formation
of protoplasts. Digestive juice from Helix pomatia and extracellular en-
zymes from bacteria and fungi have been widely used for this purpose in
most cases. Several enzyme preparations are commercially available
(e.g., Helicase, Reactifs IBF; Zymolyase-100T, Kirin Brewery; Novozym
234, Novo Industri; Lyticase, Sigma). The utilization of some other pu-
rified and crude enzymes for cell wall digestion in several yeasts has been
compared (Mann and Jeffery, 1986). Methods for preparation of pro-
toplasts and applications for studying cell wall regeneration, fusion, and
transformation have been reviewed (Peberdy, 1979; Freeman and Peber-
by, 1983).
5.1. Isolation of Organelles

A large number of protocols have been published aiming at the
separation of several membranous organelles from the same cell homog-
enate in parallel (e.g., Martinez et al., 1984) or the purification of only
one organelle or membrane fraction. In many cases the fractionation
scheme has been optimized simply to obtain maximal yield of mem-
brane-associated functions, e.g., enzymes with high specific activities.
5.1.1. Mitochondria
Mechanical as well as enzymatic digestion of the cell wall has been
used to isolate mitochondria from S. cerevisiae. Cell disruption with a
Braun homogenizer MSK has been recommended for small-scale prepa-
rations (Guerin et al., 1979; Trembath and Tzagoloff, 1979). Up to 30
strains can be processed simultanously (Trembath and Tzagoloff, 1979).
For the large-scale preparation of mitochondria, cells have been homog-
enized by continuous grinding in a Dyno mill (Guerin et al., 1979) or
frozen in liquid nitrogen and subsequently disrupted in a Waring blen-
dor at top speed (Trembath and Tzagoloff, 1979). Mechanical proce-
dures, hand shaking with glass beads, and careful use of a Dyno mill are
also recommended for preparation of pure mitochondria (without con-
tamination of nuclear DNA) for transcriptional studies (Rickwood and
Hayes, 1984).
Mitochondria of higher quality with respect to respiratory control,
Qo2 value, and membrane integrity are usually obtained by enzymatic

digestion of the cell wall and gentle mechanical desintegration of the
protoplast membrane (Guerin et al., 1979). Snail gut juice or Zymolyase
ha~ been used (Guerin et al., 1979; Daum et al., 1982) to prepare the
protoplasts, and the resulting cell homogenate is usually fractionated by
differential centrifugation; mitochondria are pelleted and washed at
around 10,000 g. Discontinuous density gradient centrifugation has
been used for further purification (Riezman et al., 1983). The separation
of inner and outer mitochondrial membranes has been achieved by lin-
ear density gradient centrifugation after homogenization with a Dounce
homogenizer or sonication (Daum et al., 1982; Riezman et al., 1983).
5.1.2. Plasma Membrane

Mechanical disruption (in most cases with glass beads in a Braun cell
homogenizer MSK) and enzymatic digestion have also been used as first
steps in plasma membrane preparation. After differential centrifugation
of the cell homogenate, separation of plasma membrane fragments is
usually achieved by discontinuous or continuous density gradient cen-
trifugation. Mitochondrial membranes can be removed by aggregation
at their isoelectric point at pH 4.5. Protoplasts are directly subjected to
osmotical lysis or are stabilized first by coating with concanavalin A. In
the latter case, plasma membrane sheets are obtained, which can be
separated by low-speed centrifugation. These sheets form vesicles upon
removal of concanavalin A. These procedures have been modified and
combined in many ways (see Goffeau and Slayman, 1981, for discussion
and references).
In another approach, protoplasts that naturally carry negative sur-
face charges at acidic pH have been attached to cationic silica micro beads
(Schmidt et al., 1983). After saturation of surplus positive charges by
polyacrylic acid and subsequent osmotic lysis at pH 8, microbead-coated
plasma membranes are separated by low-speed centrifugation ( 1000 g)
from the lysate. This procedure yields at least 85% of the plasma mem-
brane and the fraction appears to be essentially free from contaminating
organelle membranes. The plasma membrane ATPase is recovered in
highly active condition.
Two highly purified plasma membrane fractions, differing in their
protein pattern and slightly in density, have been separated in the follow-
ing way: after cell wall digestion, concanavalin A coating, lysis in the
presence of DNase and RNase, alpha-methylmannoside treatment, and
elimination of mitochondrial contamination on a sucrose gradient, the
membranes are sonicated at high salt concentration to remove tightly
bound ER from the plasma membranes. Finally, gradient centrifugation
using a shallow Renografin gradient results in separation of two frac-
tions identified as plasma membranes by marker enzymes and surface

labeling (Tschopp and Schekman, 1983).
5.1.3. Vacuoles
Procedures for preparation of vacuoles inevitably start with en-
zymatic digestion of the cell wall because mechanical disruption of intact
cells results in fragmentation of the vacuolar membranes. Isoosmotic
lysis of protoplasts by exposure to polybases (e.g., DEAE dextran) has
been used by Wiemken and co-workers (e.g., Wiemken et al., 1979) to
release intact vacuoles, which retain enclosed enzymes and low-molecu-
lar-weight compounds. Purification of the vacuolar fraction is usually
achieved by several density gradient centrifugations (Wiemken et al.,
1979). Isoosmotic rupture of protoplasts by DEAE dextran has been also
used to avoid contamination by vacuolar proteases in preparations of
cytosol and mitochondria (Schwencke et al., 1983).
5.1.4. Nucleus
Protocols for preparation of nuclei from yeast cells generally in-
clude osmotic lysis and gentle mechanical homogenization of protoplasts
in the presence of detergents (e.g., Silver et al., 1984) or low concentra-
tions of the gradient medium (e.g., Nagy et al., 1982; Schultz, 1978) used
for the subsequent continuous or discontinuous density gradient cen-
trifugation (e.g., Percoll). Yeast nuclei obtained in these or similar ways
retain the activity of the different RNA polymerases and can be used for
in vitro transcription experiments (e.g., Ide, 1981; jerome andjaehning,
1986). Nuclear membranes have been isolated by suspending purified
nuclei with a tight-fitting Dounce homogenizer, extensive treatment with
RNase and DNase, and density gradient centrifugation on Ficoll (Mann
and Mecke, 1982).
5.1.5. Microsomes
The microsomal fraction is usually isolated from yeast cells by cell
wall digestion followed by differential and density gradient centrifuga-
tion (see Kappeli et al., 1982, for references), but other procedures have
also been reported. Precipitation of microsomes by addition of CaC12 has
been used with a recovery of cytochrome P-450 greater than 80% (Kap-
peli et al., 1982). Initial cell homogenization in a French Press and isola-
tion of rough and smooth endoplasmic reticulum by repeated sucrose
density gradient centrifugation have been described (Swida et al., 1982).
5.2. Identification of Yeast Cell Subfractions

The isolation of cellular membranous organelles is the main pro-
cedure to analyze the subcellular distribution of cell constituents and
functions. This requires identification of cell fractions and evaluation of
the extent of contamination by other organelles. The activities of enzymes
being exclusively bound to certain organelle membranes or enclosed
therein (marker enzymes) have been tested for this purpose. Marker
enzymes should be used very carefully for several reasons: enzymes
selected as markers for contaminating fractions may be inactivated or
washed out during the purification procedure; the localization of some
enzymes frequently used as markers is not precisely known; no marker
enzyme has been found to be present only in the plasma membrane and
the nuclear envelope; soluble enzymes may be unspecifically attached to
membranes. Nevertheless, a list of marker enzymes suitable for yeast cell
fractionation is presented in Table III. It is partly based on a similar list
compiled by Schekman ( 1982). References are included in which the assay
conditions are described.
In addition, subcellular fractionation can be followed by other pro-
cedures; e.g., typical membrane invaginations detectable by electron mi-
croscopy have been used to identify plasma membranes (Kramer et al.,
1978) and the purification of the same fraction has been evaluated by
iodination of proteins at the surface of protoplasts (Tschopp and Schek-
man, 1983).
5.3. Autolytic Enzymes in Yeast Cells

Another obstacle in analyzing yeast cell constituents in general, and
especially in lengthy cell fractionations, is the occurrence of a large
number of hydrolytic enzymes capable of digesting intra- and extra-
cellular compounds (Babayan and Bezrukov, 1985). A variety of pro-
teases has been found (Achstetter and Wolf, 1985b; Wolf, 1986). They
are not restricted to the vacuole, but seem to occur in nearly all cell
compartments. Membrane phospholipids are hydrolyzed to fatty acids
as soon as the yeast is homogenized (see Witt et al., 1984, for references).
Cell wall-degrading enzymes-mostly glucanases, but also chitinases and
mannanases-have been found in many yeast species (Fleet, 1984).
The older literature on degradation of enzymes and other proteins
during their isolation from yeast by endogenous proteases has been
reviewed and discussed (Pringle, 1975). Pringle (1975) includes a list of
known cases of proteolytic artifacts in yeast studies and a useful chapter
on antiartifact methodology. The problem of proteolytic digestion can
now be handled by using different protease inhibitors which became
commercially available in large number in recent years; e.g., phe-

nylmethyl sulfonylfluoride, p-hydroxymercuribenzoate, pepstatin, and
chymostatin have been used together to prevent proteolytic artifacts in a
study on intracellular protein processing and sorting (Mechler et al.,
1982).
6. METHODS OF SYNCHRONIZATION
Synchronously dividing cells are desirable when studying cell cycle

or related phenomena. There are different methods to induce cell syn-
chrony:
6.1. Imposition of a Cell Cycle Block

The basic premise of this method is that due to the blocking pro-
cedure, the cells of a culture reach a common stage in the cell cycle.
From this point, synchronous growth continues for two or three genera-
tions when the block is released. The simplest way to impose a cell cycle
block is to starve the cells. The cells then arrest at G 1 but continue
division when transferred to fresh medium (Mitchison and Carter,
1975). G1 arrest may also be achieved by inhibiting DNA synthesis, by
exposing cells with mating type a to alpha-factor (this is more convenient
than the opposite procedure, since alpha-factor is easier to purify and is
also commercially available), and by use of certain cell-division-cycle (cdc)
mutants, e.g., cdc28 (Biicking-Throm et al., 1973; Elliott and McLaugh-
lin, 1983; Mitchison and Carter, 1975). Dozens of cdc mutants are now
available so that blockage at distinct stages is possible. A critical survey of
these mutants is given by Pringle and Hartwell ( 1981 ).
6.2. Centrifugational Methods

Cells at different stages of the cell cycle can be selected by cen-
trifugational methods from an asynchronously growing culture. Sepa-
rated cell cycle fractions may be used directly for analysis or may serve as
starting material for a culture to be grown synchronously.
During the cell cycle, the density of cells changes periodically
(Schulman, 1978). In Saccharomyces the maximum density is reached
shortly after bud emergence and thereafter drops to a minimum during
cytokinesis (Shulman, 1978). High- and low-density cells can be sepa-
rated by isopycnic equilibrium centrifugation in density gradients pre-
pared with Ludox, Renografin, or Percoll (Mitchison and Carter, 1975;
Shulman, 1978; Baldwin and Kubitschek, 1984).
Since cell size also varies during the cell cycle, sorting according to
size yields cell cycle fractions. Centrifugal elutration has been successful-
ly applied for this purpose (Gordon and Elliott, 1977). In principle, this
procedure is simply a counterflow centrifugation method. At constant
rotor speed, a cell suspension is pumped opposite to the direction of the
centrifugal forces through the rotor. Fractionation is carried out by in-
crementally increasing the flow rate.
Velocity sedimentation in sucrose gradients can be applied to obtain
small cells which are normally at the start of the cell cycle. The advantage
is that separations can be carried out at low speed in a simple laboratory
centrifuge equipped with a swinging bucket rotor. If large amounts of
cells have to be processed, the method can be scaled up in a zonal rotor
(Mitchison and Carter, 1975).
REFERENCES
Achstetter, T., and Wolf, D. H., 1985a, Hormone processing and membrane bound pro-
teinases in yeast, EMBO]. 4:173-177.
Achstetter, T., and Wolf, D. H., 1985b, Proteinases, proteolysis and biological control in the
yeast Saccharomyces cerevisiae, Yeast 1: 139-15 7.
Adams, A. E. M., and Pringle, J. R., 1984, Relationship of actin and tubulin distribution to
bud growth in wild-type and morphogenetic-mutant Saccharomyces cerevisiae, ]. Cell
Biol. 98:934-945.
Ades, 1., and Butow, R., 1980, The products of mitochondria-bound cytoplasmic poly-
somes in yeast,]. Bioi. Chern. 255:9918-9924.
Andregg, R. J., Beta, R., Carr, S. A., Crabb, J. W., and Dontze, W., 1988, Structure
of Saccharomyces cerevisiae mating hormone a-factor, ]. Bioi. Chern. 263:18236-
18240.
Appleby, C. A., and Morton, R. K., 1959, Lactic dehydrogenase and cytochrome b2 of
baker's yeast, Biochem. ]. 71:492-499.
Ares, M., 1986, U2 RNA from yeast is unexspectedly large and contains homology to
vertebrate U4, U5, and U6 small nuclear RNAs, Cell 47:49-59.
Babayan, T. L., and Bezrukov, M.G., 1985, Autolysis in yeast, Acta Biotechnol. 5:129-136.
Baldwin, W. W., and Kubitschek, H. E., 1984, Bouyant density variation during the cell
cycle of Saccharomyces cerevisiae, ]. Bacteriol. 158:701-704.
Ballou, C. E., 1982, Yeast cell wall and cell surface, in: The Molecular Biology of the Yeast
Saccharomyces, Metabolism and Gene Expression U· N. Strathern, E. W. Jones, and J. R.
Broach, eds.), Cold Spring Harbor Monograph Series, Cold Spring Harbor, NY, pp.
335-360.
Ballou, C. E., Maitra, S. K., Walter,J. W., and Whelan, W. L., 1977, Developmental defects
associated with glucosamine auxotrophie in Saccharomyces cerevisiae, Proc. Natl. Acad.
Sci. USA 74:4351-4355.
Bandlow, W., 1972, Membrane separation and biogenesis of the outer membrane of yeast
mitochondria, Biochim. Biophys. Acta 282:105-122.
Baralle, F. E., 1983, The functional significance of leader and trailer sequences in eu-
caryotic mRNAs, Int. Rev. Cytol. 81:71-106.
Becker, G. W., and Lester, R. L., 1980, Biosynthesis of phospoinositol-containing
sphingolipids from phosphatidylinositol by a membrane preparation from Saccharo-
myces cerevisiae,J. Bacteriol. 142:747-754.
Betz, R., MacKay, V. L., and Duntze, W., 1977, a-Factor from S. cerevisiae: Partial charac-
terization of a mating hormone produced by cells of mating type a, ]. Bacteriol.
132:462-472.
Betz, R., Manney, T. R., and Duntze, W., 1981, Hormonal control of gametogenesis in the
yeast Saccharomyces cerevisiae, Gamete Res. 4:571-584.
Black, S., 1985, Inhibition of the reductive activation of valyl-tRNA synthetase from yeast
by unsaturated fatty acids and associated observations on newly found lipophilic sub-
stances from yeast, ]. Biol. Chem. 260:433-439.
Boller, T., Diirr, M., and Wiemken, A., 1976, Asymmetric distribution of concanavalin A
binding sites on yeast plasmalemma and vacuolar membrane, Arch. Microbial. 109:115-
118.
Borst-Pauwels, G. W. F. H., and Peters, P. H. j., 1981, Factors affecting the inhibition of
yeast plasma membrane ATPase by vanadate, Biochim. Biophys. Acta 642:173-181.
Bottema, C. K., and Parks, L. W., 1980, Sterol analysis of the inner and outer mitochondri-
al membranes in yeast, Lipids 15:987-992.
Brake, A., Brenner, C., Najarian, R., Laybourn, P., and Merryweather,]., 1985, Structure
of genes encoding precursors of the yeast peptide mating pheromone a-factor, in:
Transport and Secretion tf proteins(M.-j.Gething, ed.), Cold Spring Harbor Laboratory,
Cold Spring Harbor, NY, pp. 103-108.
Brewer, B. J., and Fangman, W. L., 1980, Preferential inclusion of extrachromosomal
genetic elements in yeast meiotic spores, Proc. Natl. Acad. Sci. USA 77:5380-5384.
Brody, E., and Abelson,J., 1985, The "spliceosome": Yeast premessenger RNA associates
with a 40S complex in a splicing dependent reaction, Science 228:963-967.
Biicking-Throm, E., Duntze, W., Hartwell, L. H., and Manney, T. R., 1973, Reversible
arrest of haploid yeast cells at the initiation of DNA synthesis by a diffusible sex factor,
Exp. Cell Res. 76:99-110.
Byers, B., 1981, Cytology of the yeast life cycle, in: The Molecular Biology if the Yeast Sac-
charomyces, Life Cycle and Inheritance (J. N. Strathern, E. W. Jones, and J. R. Broach,
eds.), Cold Spring Harbor Monograph Series, Cold Spring Harbor, NY, pp. 59-97.
Byers, B., and Goetsch, L., 1975a, Electron microscopic observations on the meiotic ka-
ryotype of diploid and tetraploid Saccharomyces cerevi.siae, Proc. Natl. Acad. Sci. USA
72:5056-5060.
Byers, B., and Goetsch, L., 1975b, Behaviour of spindles and spindle plaques in the cell
cycle and conjugation of Saccharomyces cerevisiae,]. Bacterial. 124:511-523,
Byers, B., and Goetsch, L., 1976, A highly ordered ring of membrane-associated filaments
in budding yeast,]. Cell Bioi. 69:717-721.
Byrd, j. C., Tarentino, A. L., Maley, F., Atkinson, P. H., and Trimble, R. B., 1982,
Glycoprotein synthesis in yeast: Identification of Man8G1cNAc 2 as an essential inter-
mediate in oligosaccharide processing,]. Biol. Chem. 257:14657-14666.
Cabib, E., Roberts, R., and Bowers, B., 1982, Synthesis of the yeast cell wall and its
regulation, Annu. &v. Biochem. 51:763-793.
Cabib, E., Kang, M.S., Bowers, B., Elango, N., Mattia, E., Slater, M. L., and Au-Young,j.,
1984, Chitin synthesis in yeast, a vectorial process in the plasma membrane, FEMS
Symp. 27:91-100.
Calderbank, J., Keenan, M. H. J., and Rose, A. H., 1985, Plasma-membrane phospholipid
unsaturation affects expression of the general amino-acid permease in Saccharomyces
cerevi.siae Y185,]. Gen. Microbial. 131:57-65.
Carlson, M., and Botstein, D., 1982, Two differentially regulated mRNAs with different 5'
ends encode secreted and intracellular forms of yeast invertase, Cell 28:145-154.
Carlson, M., Taussig, R., Kustu, S., and Botstein, D., 1983, The secreted form of invertase
in Saccharomyces cerevi.siae is synthesized from mRNA encoding a signal sequence, Mol.
Cell. Biol. 3:439-447.
Castellanos, R. M. P., and Mazon, M. J., 1985, Identification of phosphotyrosine in yeast

proteins and of a protein tyrosine kinase associated with the plasma membrane,]. Biol.
Chem. 260:8240-8242.
Catley, B. J., 1983, Regulation of yeast and fungal polysaccharides excluding chitin and
cellulose, Prog. Indu.st. Microbiol. 18:129-200.
Cech, T. R., 1983, RNA splicing: Three themes with variations, Cell54:713-716.
Chia, L.-L., and McLaughlin, C., 1979, The half-life of mRNA in Saccharomyces cerevisiae,
Mol. Gen. Genet. 170:137-144.
Chu, F. K., Takase, K., Guarino, D., and Maley, F., 1985, Diverse properties of external and
internal forms of yeast invertase derived from the same gene, Biochemistry 24:6125-
6132.
Chvatchko, Y., Howald, 1., and Riezman, H., 1986, Two yeast mutants defective in endo-
cytosis are defective in pheromone response, Cell 46:355-364.
Clausen, M. K., Christiansen, K., Jensen, P. K., and Behnke, 0., 1974, Isolation of lipid
particles from baker's yeast, FEBS Lett. 45:176-179.
Clayton, L., Pogson, C. I., and Gull, K., 1979, Microtubule proteins in yeast Saccharomyces
cerevisiae, FEBS Lett. 106:67-70.
Cleveland, D. W., Lopata, M. A., MacDonald, R. J., Cowan, N.J., Rutter, W. J., and
Kirschner, M. W., 1980, Number and evolutionary conservation of alpha- and beta-
tubulin and cytoplasmic beta- and gamma-actin genes using specific cloned eDNA
probes, Cell 20:95-105.
Daum, G., Bohni, P. C., and Schatz, G., 1982, Import of proteins into mitochondria:
Cytochrome b and cyrochrome c peroxidase are located in the intermembrane space
of yeast mitochondria,]. Biol. Chem. 257:13028-13033.
Del Rey, F., Santos, T., Garcia-Acha, 1., and Nombela, C., 1979, Synthesis of 1,3-~
glucanases in Saccharomyces cerevisiae during the mitotic cycle, mating and sporulation,
]. Bateriol. 139:924-931.
Djavadi-Ohaniance, L., Rudin, Y., and Schatz, G., 1978, Identification of enzymatically
inactive apocytochrome c peroxidase in aerobically grown Saccharomyces cerevisiae, ].
Biol. Chem. 253:4402-4407.
Domdey, H., Apostol, B., Lin, R.-J., Newmann, A., Brody, E., and Abelson, J., 1984, Lariat
structures are in vivo intermediates in yeast pre mRNA splicing, Cell 39:611-621.
Douglas, M., and Takeda, M., 1985, Nuclear genes encoding mitochondrial proteins in
yeast, Trends Biochem. Sci. 10:192-194.
Dow, J. M., Carron, R. R., and Villa, V. D., 1981, Role of membranes of mycelial Mucor
rouxii in synthesis and secretion of cell wall matrix polymers,]. Bacteriol. 145:272-279.
Duffus, J. H., Levi, C., and Manners, D. J., 1982, Yeast cell-wall glucans, Adv. Microb.
Physiol. 23:151-181.
Dujon, B., 1981, Mitochondrial genetics and functions, in: The Molecular Biology of the Yeast
Saccharomyces, Life Cycle and Inheritance 0· N. Strathern, E. W. Jones, and J. R. Broach,
Duntze, W., MacKay, V., and Manney, T. R., 1970, Saccharomyces cerevisiae: A diffusible sex
factor, Science 168:1472-1473.
Duntze, W., Stotzler, D., Biicking-Throm, E., and Kalbitzer, S., 1973, Purification and
partial characterization of alpha-factor, a mating-type specific inhibitor of cell re-
production from Saccharomyces cerevisiae, Eur.]. Biochem. 55:357-365.
Duran, A., and Cabib, E., 1978, Solubilization and partial purification of yeast chitin
synthase: Confirmation of the zymogenic nature of the enzyme, ]. Biol. Chem.
253:4419-4425.
Diirr, M., Urech, K., Boller, T., Wiemken, A., Schwencke, J., and Nagy, M., 1979, Se-
questration of arginine by polyphosphate in vacuoles of yeast (Saccharomyces cerevisiae),
Arch. Microbiol. 121: 169-17 5.
Elliott, S. G., and McLaughlin, C. S., 1983, The yeast cell cycle: Coordination of growth
and division rates, Prog. Nucl. Acid Res. Mol. Biol. 28:143-176.
Erdmann, V. A., Wolters, J., Huysmans, E., and DeWachter, R., 1985, Collection of pub-
lished 5S, 5.8S and 4.5S ribosomal RNA sequences, Nucl. Acids Res. 13 (Suppl.):rl05-
rl53.
Esmon, B., Novick, P., and Schekman, R., 1981, Compartmentalized assembly of oligosac-
charides on exported glycoproteins in yeast, Cell 25:451-460.
Esposito, R. E., and Klapholz, S., 1981, Meiosis and Ascospore development, in: The
Molecular Biology tf the Yeast Saccharomyces, Life Cycle and Inheritance (J. N. Strathern,
E. W. jones, and J. R. Broach, eds.), Cold Spring Harbor Monograph Series, Cold
Spring Harbor, NY, pp. 211-287.
Faye, G., and Sor, F., 1977, Analysis of mitochondrial ribosomal proteins of Saccharomyces
cerevisiae by two dimensional polyacrylamide gel electrophoresis, Mol. Gen. Genet.
155:27-35.
Ferro-Novick, S., 1985, Genetic approaches to the study of protein targeting across a lipid
bilayer, Trends Biochem. Sci. 10:425-427,
Ferro-Novick, S., Novick, P., Field, C., and Schekman, R., 1984, Yeast secretory mutants
that block the formation of active cell surface enzymes,]. Cell Biol. 98:35-43.
Filipowicz, W., and Gross, H. j., 1984, RNA ligation in eukaryotes, Trends Biochem. Sci.
9:68-71.
Fleet, G. H., 1984, The occurrence and function of endogenous wall-degrading enzymes
in yeasts, FEMS Symp. 27:227-238.
Freeman, R. F., and Peberby,J. F., 1983, Protoplast fusion in yeasts, in: Yeast genetics. Funda-
mental and applied aspects (J. F. T. Spencer, D. M. S~ncer, and A. R. W. Smith, eds.),
Springer, New York, pp. 243-253.
Frevert, J., and Ballou, C. E., 1982, Yeast invertase polymorphism is correlated with variable
states of oligosaccharide chain phosphorylation, Proc. Natl. Acad. Sci. USA 79:6147-
6150.
Frevert, j., and Ballou, C. E., 1985, Saccharomyces cerevisiae structural cell wall mannopro-
tein, Biochemistry 24:753-759.
Gallwitz, D., and Sures, R., 1980, Structure of a split gene: Complete nucleotide sequence
of the actin gene in Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. USA 77:2546-2550.
Georgiev, 0. I., Nikolaev, N., Hadjiolov, A. A., Skryabin, K. G., Zakharyev, V. M., and
Bayev, A. A., 1981, The structure of the yeast ribosomal RNA genes. 4. Complete
sequence of the 25S rRNA gene from Saccharomyces cerevisiae, Nucl. Acids Res. 9:6953-
6973.
Goffeau, A., and Slayman, C. W., 1981, The proton-translocating ATPase of the fungal
plasma membrane, Biochim. Biophys. Acta 639:197-223.
Goldstein, A., and Lampen,J. 0., 1975, ~-o-Fructofuranoside fructohydrolase from yeast,
Meth. Enzymol. 42:504-511.
Gordon, C. N ., and Elliott, S. G., 1977, Fractionation of Saccharomyces cerevisiae cell popula-
tions by centrifugal elutration,]. Bacterial. 129:97-100.
Gorenstein, C., and Warner, J. R., 1976, Coordinate regulation of the synthesis of eu-
karyotic ribosomal proteins, Proc. Natl. Acad. Sci. USA 73:1547-1551.
Gourse, R. L., Sharrock, R. A., and Nomura, M., 1986, Control of ribosome synthesis in
Escherichia coli, in: Structure, Function and Genetics tf Ribosomes (B. Hardesty and G.
Kramer, eds.), Springer-Verlag, New Yor:k. pp. 766-788.
Greer, C., and Schekman, R., 1982a, Actin from Saccharomyces cerevisiae, Mol. Cell. Biol.
2:1270-1278.
Greer, C., and Schekman, R., 1982b, Calcium control of Saccharomyces cerevisiae actin as-
sembly, Mol. Cell. Biol. 2:1279-1286.
Greer, C. L., Peebles, C. L., Gegenheimer, P., and Abelson,J., 1983, Mechanism of action
of a yeast RNA ligase in tRNA splicing, Cell 32:537-546.
Guerin, B., Labbe, P., and Somlo, M., 1979, Preparation of yeast mitochondria (Sac-
charomyces cerevisiae) with good P/0 and respiratory control ratios, Meth. Enzymol.
55:149-159.
Guillen, A., Leal, F., Andaluz, E., and Larriba, G., 1985, Endogenous factors that modula .e
yeast glucan synthetase in cell-free extracts, Biochim. Biophys. Acta 842:151-161.
Guiso, N., Dreyfus, M., Siffert, 0., Danchin, A., Spyridakis, A., Gargouri, A., Claisse, M.,
and Slonimski, P. P., 1984, Antibodies against synthetic oligopeptides allow identifica-
tion of the mRN A-maturase encoded by the second intron of the yeast cob-box gene,
EMBO]. 3:1769-1772.
Guth, E., Hasimoto, T., and Conti, S. F., 1972, Morphogenesis of ascospores in Sac-
charomyces cerevisiae,]. Bacteriol. 109:869-880.
Guthrie, C., 1986, Finding functions for small nuclear RNAs in yeast, Trends Biochem. Sci.
11:430-434.
Guthrie, C., and Abelson,]., 1982, Organization oftRNA genes in Saccharomyces cerevisiae,
in: The Molecular Biology if the Yeast Saccharomyces, Metabolism and Gene Expression (J. N.
Strathern, E. W. Jones, and J. R. Broach, eds.), Cold Spring Harbor Monograph
Series, Cold Spring Harbor, NY, pp. 487-528.
Haber, J. E., and Halvorson, H. 0., 1975, Methods in sporulation and germination of
yeasts, in: Methods in Cell Biology, Vol. 11 (D. M. Prescott, ed.), Academic Press, New
York, pp. 46-69.
Hadjiolov, A. A., 1985, The Nucleolus and Ribosome Biogenesis, Springer-Verlag, Vienna, New
York, pp. 1-263.
Hagen, D. C., McCaffrey, G., and Sprague, G. F., 1986, Evidence the yeast STE3 gene
encodes a receptor for the peptide pheromone a-factor: Gene sequence and implica-
tions for the structure of the presumed receptor, Proc. Natl. Acad. Sci. USA 83:1418-
1422.
Haguenauer-Tsapis, R., and Hinnen, A., 1984, A deletion that induces the signal pep-
tidase cleavage site impairs processing, glycosylation, and secretion of cell surface
yeast acid phosphatase, Mol. Cell. Biol. 4:2668-2675.
Hanson, B. A., and Lester, R. L., 1980, The extraction of inositol-containing phospholipids
and phosphatidylcholine from Saccharomyces cerevisiae and Neurospora crassa, ]. Lipid
Res. 21:309-315.
Haselbeck, A., and Schekman, R., 1986, Interorganelle transfer and glycosylation of yeast
invertase in vitro, Proc. Natl. Acad. Sci. USA 83:2017-2021.
Hashimoto, C., Cohen, R. E., Zhang, W. -].,and Ballou, C. E., 1981, Glycoprotein chains on
yeast carboxypeptidase Y are phosphorylated, Proc. Natl. Acad. Sci. USA 78:2244-
2248.
Hasilik, A., and Tanner, W., 1978, Carbohydrate moiety of carboxypeptidase Y and pertur-
bation of its biosynthesis, Eur.]. Biochem. 91:567-575.
Hay, R., Bohni, P., and Gasser, S., 1984, How mitochondria import proteins, Biochim.
Biophys. Acta 779:65-87.
Henry, S. A., 1982, Membrane lipids of yeast: Biochemical and genetic studies, in: The
Molecular Biology if the Yeast Saccharomyces, Metabolism and Gene Expression (J. N. Strat-
hern, E. W. Jones, and J. R. Broach, eds.), Cold Spring Harbor Monograph Series,
Henry, S. A., Klig, L. S., and Loewy, B.S., 1984, The genetic regulation and coordination
of biosynthetic pathways in yeast: Amino acid and phospholipid synthesis, Annu. Rev.
Genet. 18:207-31.
Huet, J., Cottrelle, P., Cool, M., Vignais, M.-L., Thiele, D., Marek, C., Buhler, J. M.,
Sentenac, A., and Fromageot, P., 1985, A general upstream binding factor for genes
of the yeast translational apparatus, EMBO]. 4:3539-3547.
Ide, G. J., 1981, Nucleoside 5',-I'Y-Sltriphosphates will initiate transcription in isolated
yeast nuclei, Biochemistry 20:2633-2638.
Itoh, T., Higo, K., and Otaka, E., 1979, Isolation and characterization of twenty-three
ribosomal proteins from large subunits of yeast, Biochemistry 18:5787-5793.
Jacq, C., Banroques, J., Becam, A.M., Slonimski, P. P. Guiso, N., and Danchin, A., 1984,
Antibodies against a fused "lacZ-yeast mitochondrial intron" gene product allow iden-
tification of the mRNA maturase encoded by the fourth intron of the yeast cob-box
gene, EMBO]. 3:1567-1572.
Jenness, D. D., and Spatrick, P., 1986, Down regulation of the alpha-factor peromone in
Saccharomyces cerevisiae, Cell 46:345-353.
Jenness, D., Burkholder, A. C., and Hartwell, H., 1983, Binding of alpha-factor phe-
romone to yeast a-cells: Chemical and genetic evidence for an alpha-factor receptor,
Cell 35:521-529.
Jenness, D. D., Burkholder, A. C., and Hartwell, L. H., 1986, Binding of alpha-factor
pheromone to Saccharomyces cerevisiae a-cells: Dissociation constant and number of
binding sites, Mol. Cell. Biol. 6:318-320.
Jerome, J. F., and Jaehning, J. A., 1986, mRNA transcription in nuclei isolated from
Saccharomyces cerevisiae, Mol. Cell. Biol. 6:1633-1639.
Julius, D., Blair, L., Brake, A., Sprague, G., and Thorner, J., 1983, Yeast alpha factor is
processed from a large precursor polypepide: The essential role of a membrane-
bound dipeptidyl aminopeptidase, Cell 32:839-852.
Julius, D., Schekman, R., and Thorner, J., 1984, Glycosy1ation and processing of prepro-
alpha-factor through the yeast secretory pathway, Cell36:309-318.
Kaibuchi, K., Miyajima, A., Arai, K.-1., and Matsumoto, K., 1986, Possible involvement of
RAS-encoded proteins in glucose-induced inositolphospholipid turnover in Sac-
charomyces cerevisiae, Proc. Natl. Acad. Sci. USA 83:8172-8176.
Kakinuma, Y., Ohsumi, Y., and Anraku, Y., 1981, Properties of H+-translocating ade-
nosine triphosphatase in vacuolar membranes of Saccharomyces cerevisiae,j. Biol. Chem.
256: 10859-10863.
Kaneko, H., Hosohara, M., Tanaka, M., and Itoh, T., 1976, Lipid composition of 30 species
of yeast, Lipids 11:837-844.
Kang, M. S., Elago, N., Mattia, E., Au-Young, J., Robbins, P. W., and Cabib, E., 1984,
Isolation of chitin synthase from Saccharomyces cerevisiae, purification of an enzyme by
entrapment in the reaction product,]. Biol. Chem. 259:14966-14972.
Kappeli, 0., Sauer, M., and Fiechter, A., 1982, Convenient procedure for the isolation of
highly enriched cytochrome P-450-containing microsomal fraction from Candida trop-
icalis, Anal. Biochem. 126:179-182.
Kato, N., Sahm, H., Schuette, H., and Wagner, F., 1979, Purification and properties of
glucose-6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase 'from a
methanol-utilizing yeast, Candida boidinii, Biochem. Biophys. Acta 566:1-11.
Kilmartin, J. V., 1981, Purification of yeast tubulin by self-assembly in vitro, Biochemistry
20:3629-3633.
Kilmartin, J. V., and Adams, A. E. M., 1984, Structural rearrangements of tubulin and
actin during the cell cycle of the yeast Saccharomyces,]. Cell Biol. 98:922-933.
King, S. M., and Hyams, J. S., 1982, The mitotic spindle of Saccharomyces cerevisiae: Assem-
bly, structure, and function, Micron 13:93-117.
King, S. M., Hyams, J. S., and Luba, A., 1982, Absence of microtubule sliding and an
analysis of spindle formation from the yeast Saccharomyces cerevisiae, ]. Cell. Biol.
94:341-349.
Klig, L. S., Homann, M. J., Carman, G. M., and Henry, S. A., 1985, Coordinate regulation
of phospholipid biosynthesis in Saccharomyces cerevisiae: Pleiotropically constitutive opil
mutant,]. Bacterial. 162:1135-1141.
Koch, H., and Friesen, J.D., 1979, Individual messenger RNA half lives in Saccharomyces
cerevisiae, Mol. Gen. Genet. 170:129-135.
Kramer, R., Kopp, F., Niedermeyer, W., and Fuhrmann, G. F., 1978, Comparative studies
of the structure and composition of the plasmalemma and the tonoplast in Sac-
charomyces cerevisiae, Biochim. Biophys. Acta 507:369-380.
Kruiswijk, T., Planta, R. J., and Mager, W. H., 1978, Quantitative analysis of the protein
composition of yeast ribosomes, Eur.]. Biochem. 83:245-252.
Kurjan, j., and Herskowitz, 1., 1982, Structure of a yeast pheromone gene (MFalpha): A
putative alpha-factor precursor contains four tandem copies of mature alpha-factor,
Cell 30:933-943.
Langford, C. J., and Gallwitz, D., 1983, Evidence for an intron-contained sequence re-
quired for the splicing of yeast RNA polymerase II transcripts, Cell 33:519-527.
Larkin,J. C., and Woolford,J. L., 1983, Molecular cloning and analysis of the CRYJ gene:
A yeast ribosomal protein gene, Nucl. Acids Res. 11:403-420.
Leer, R. J., van Raamsdonk-Duin, M. C., Molenaar, M. T., Cohen, L. H., Mager, W., and
Planta, R. j., 1982, The structure of the gene coding for the phosphorylated ribosomal
protein SlO in yeast, Nucl. Acids Res. 10:5869-5878.
Leer, R. j., van Raamsdonk-Duin, M. C., Hagendoorn, J. M., Mager, W. H., and Planta, R.
J., 1984, Structural comparison of yeast ribosomal protein genes, Nucl. Acids Res.
12:6685-6700.
Leer, R. J., van Raamsdonk-Duin, M. C., Kraakman, P., Mager, W. H., and Planta, R. J.,
1985a, The genes of yeast ribosomal proteins S24 and L46 are adjacent and diver-
gently transcribed, Nucl. Acids Res. 13:701-709.
Leer, R. j., Van Ramsdonk-Duin, M. M. C., Mager, W. H., and Planta, R. j., 1985b,
Conserved sequences upstream of yeast ribosomal protein genes, Curr. Gen. 9:273-
277.
Lehle, L., and Tanner, W., 1978, Glycosyltransfer from dolichylphosphate sugars to en-
dogenous and exogenous glycoprotein acceptors in yeast, Eur.]. Biochem. 83:563-570.
Letts, V. A., and Dawes, I. W., 1983, Temperature-sensitive Saccharomyces cerevisiae mutant
defective in lipid biosynthesis,]. Bacterial. 156:212-221.
Letts, V. A., and Henry, S. A., 1985, Regulation of phospholipid synthesis in phosphatidyl
serine synthase-deficient (chol) mutants of Saccharomyces cerevisiae,j. Bacteriol. 163:560-
567.
Liao, H., and Thorner,j., 1980, Yeast mating pheromone alpha-factor inhibits adenylate
cyclase, Proc. Natl. Acad. Sci. USA 77:1898-1902.
Lipke, P. N., Taylor, A., and Ballou, C. E., 1976, Morphogenic effects of a-factor on
Saccharomyces cerevisiae a cells,]. Bacterial. 127:610-618.
Low, C., Rodriguez, R. J., and Parks, L. W., 1985, Modulation of yeast plasma membrane
composition of a yeast sterol auxotroph as a function of exogenous sterol, Arch. Bio-
chem. Biophys 240:530-538.
Makarow, M., 1985a, Endocytosis in Saccharomyces cerevisiae: Internalization of enveloped
virus into sheroplasts, EMBO]. 4: 1855-1860.
Makarow, M., 1985b, Endocytosis in Saccharomyces cerevisiae: Internalization of a-amylase
and fluorescent dextran into cells, EMBO]. 4:1861-1866.
Makarow, M., and Nevalainen, T., 1987, Transport of a fluorescent macromolecule via
endosomes to a vacuole in Saccharomyces cerevisiae,]. Cell Biol. 104:67-75.
Malpartida, F., and Serrano, R., 1981a, Reconstitution of the proton-translocating ade-
nosine triphosphatase of yeast plasma membranes,]. Biol. Chem. 256:4157-4177.
Malpartida, F., and Serrano, R., 1981b, Phosphorylated intermediate of the ATPase from
the plasma membrane of yeast. Eur.]. Biochem. 116:413-417.
Mann, W., and Jeffery, J., 1986, Yeasts in molecular biology. Spheroplast preparation with
Candida utilis, Schizosaccharomyces pombe and Saccharomyces cerevisiae, Biosci. Rep. 6:597-
602.
Mann, K., and Mecke, D., 1982, The isolation of Saccharomyces cerevisiae nuclear membranes
with nuclease and high-salt treatment, Biochim. Biophys. Acta 687:57-62.
Martinez, J. P., Murqui, A., Flores, A., and Sentandreu, R., 1984, Subcellular fractionation
of actively growing protoplasts of Saccharomyces cerevisiae, Biochim. Biophys. Acta

805:59-71.
Mason, T. L., Poyton, R. 0., Wharton, D. C., and Schatz, B., 1973, Cytochrome c oxidase
from baker's yeast. I. Isolation and properties,]. Biol. Chem. 248:1346-1354.
Matile, P., 1978, Biochemistry and function of vacuoles, Annu. Rev. Plant Physiol. 29:193-
223.
Matile, P., and Wiemken, A., 1976, Interaction between cytoplasm and vacuole, in: Trans-
port in Plants, Vol. 3 (C. R. Stocking and U. Heber, eds.), Springer, Berlin, pp. 255-
287.
Mattaj, I. W., 1984, SnRNAs: From gene architecture to RNA processing, Trends Biochem.
Sci. 9:435-437.
McDonough, J. P., and Mahler, H. P., 1982, Covalent phosphorylation of the Mg2 +-
dependent ATPase of yeast plasma membrane,]. Biol. Chem. 257:14579-14581.
McKee, E. E., McEwen, J. E., and Poyton, R. 0., 1984, Mitochondrial gene expression in
Saccharomyces cerevisiae. II. Fidelity of translation in isolated mitochondria from wild
type and respiratory-deficient mutant cells,]. Biol. Chem. 259:9332-9338.
Mechler, B., Muller, M., Muller, H., Meussdoerffer, F., and Wolf, D. H., 1982, In vivo
biosynthesis of the vacuolar proteinases A and B in the yeast Saccharomyces cerevisiae,].
Biol. Chem. 257:11203-11206.
Meyer, D. I., 1982, The signal hypothesis-A working model, Trends Biochem. Sci. 7:320-
321.
Miller, A. M., 1984, The yeast Mata1 gene contains two introns, EMBO]. 3:1061-1065.
Mitchison, J. M., and Carter, B. L. A., 1975, Cell cycle analysis, in: Methods in Cell Biology,
Vol. 11 (D. M. Prescott, ed.), Academic Press, New York, pp. 201-219.
Moeller, C. H., Mudd, J. B., and Thomson, W. W., 1981, Lipid phase separations and
intramembranous particle movements in the yeast tonoplast, Biochim. Biophys. Acta
643:376-386.
Moens, P. B., and Rapport, E., 1971, Spindles, spindle plaques, and meiosis in the yeast
Saccharomyces cerevisiae (Hansen),]. Cell Biol. 50:344-361.
Mueller, S. C., and Branton, D., 1984, Identification of coated vesicles in Saccharomyces
cerevisiae,]. Cell Biol. 98:341-346.
Nagy, M., Laporte, J., Pen verne, B., and Herne, G., 1982, Nuclear localization of aspartate
transcarbamoylase in Saccharomyces cerevisiae,]. Cell Biol. 92:790-794.
Nakajima, T., and Ballou, C. E., 1975, Yeast manno-protein biosynthesis: Solubilization
and selective assay of four mannosyltransferases, Proc. Natl. Acad. Sci. USA 72:3912-
3916.
Nakayama, N., Miyajima, A., and Arai, K., 1987, Common signal transduction system by
STE2 and STE3 in haploid cells of Saccharomyces cerevisiae: Autocrine cell-cycle arrest
results from forced expression of STE2, EMBO]. 6:249-254.
Navarrete, R., and Serrano, R., 1983, Solubilization of yeast plasma membranes and mito-
chondria by different types of nondenaturating detergents, Biochim. Biophys. Acta
728:403-408.
Neff, N. F., Thomas, J. H., Grisafi, P., and Botstein, D., 1983, Isolation of the ~-tubulin
gene from yeast and demonstration of its essential function in vivo, Cell33:211-215.
Newman, A. J., Lin, R., Cheng, S., and Abelson, J., 1985, Molecular consequences of
specific intron mutations on yeast mRNA splicing in vivo and in vitro, Cell42:335-344.
Ng, R., and Abelson,]., 1980, Isolation of the gene for actin in Saccharomyces cerevisiae, Proc.
Natl. Acad. Sci. USA 77:3912-3916.
Niedermeyer, W., 1976, The elasticity of the yeast cell tonoplast related to its ultrastructure
and chemical composition. II. Chemical and cytochemical investigations, Cytobiology
13:380-393.
Nomura, M., Gourse, R., and Baughman, G., 1984, Regulation of the synthesis of ribo-
somes and ribosomal components Annu. Rev. Biochem. 53:75-117.
Novick, P., 1985, Intracellular transport mutants of yeast, Trends Biochem. Sci. 10:432-434.
Novick, P., and Botstein, D., 1985, Phenotypic analysis of temperature-sensitive yeast actin
mutants, Cell40:405-416.
Novick, P., Field, C., and Schekman, R., 1980, Identification of 23 complementation
groups required for post-translational events in the yeast secretory pathway, Cell
21:205-215.
Novick, P., Ferro, S., and Schekman, R., 1981, Order of events in the yeast secretory
pathway, Cell25:461-469.
Orlean, P., 1987, Two chitin synthases in Saccharomyces cerevisiat,]. Biol. Chem. 262:5732-
5739.
Orlean, P., Ammer, H., Watzele, M., and Tanner, W., 1986, Synthesis of an 0-glycosylated
cell surface protein induced in yeast by a-factor, Proc. Natl. Acad. Sci. USA 85:6263-
6266.
Otaka, E., and Osawa, S., 1981, Yeast ribosomal proteins. V. Correlation of several nomen-
clatures and proposal of a standard nomenclature, Mol. Gen. Genet. 181:176-182.
Otaka, E., Higo, K., and Osawa, S., 1982, Isolation of seventeen proteins and amino-
terminal amino sequences of eight proteins from cytoplasmic ribosomes of yeast,
Biochemistry 21:4545-4550.
Otaka, E., Higo, K., and Itoh, T., 1983, Yeast ribosomal proteins. VII. Cytoplasmic
ribosomal proteins from Schizosaccharomyces pombe, Mol. Gen. Genet. 191:519-524.
Padgett, R. A., Grabowski, P. j., Konarska, M. M., and Sharp, P., 1985, Splicing messenger
RNA precursors: Branch sites and lariat RNAs, Trends Biochem. Sci. 10:154-157.
Padgett, R. A., Grabowski, P.J., Konarska, M. M., Seiler, S., and Sharp, P., 1986, Splicing of
messenger RNA and RNA precursors,Annu. Rev. Biochem. 55:1119-1150.
Pastor, F. I. J., Valentin, E., Herrero, E., and Sentandreu, R., 1984, Structure of the
Saccharomyces cerevisiae cell wall. Mannoproteins released by zymolyase and their con-
tribution to wall architecture, Biochim. Biophys. Acta 802:292-300.
Payne, G. S., and Schekman, R., 1985, A test of clathrin function in protein secretion and
cell growth, Science 250: 1009-1014.
Peberdy,j. F., 1979, Fungal protoplasts: Isolation, reversion, and fusion, Annu. Rev. Micro-
bioi. 35:21-39.
Peebles, C. L., Gegenheimer, P., and Abelson, j., 1983, Precise excision of intervening
sequences from precursor tRNAs by a membrane-associated yeast endonuclease, Cell
52:525-536.
Peterson, J. B., and Ris, H., 1976, Electron-microscopic study of the spindle and chromo-
some movement in the yeast Saccharomyces cerevisiae,]. Cell Sci. 22:219-242.
Pikielny, C. W., Rymond, B. C., and Rosbash, M., 1986, Electrophoresis of ribonucleopro-
teins reveals an ordered assembly pathway of yeast splicing complexes, Nature
524:341-345.
Planta, R.j., Mager, W. H., Leer, R.j., Woudt, L. P., Raue, H. A., and El-Baradi, T. T. A. L.,
1986, Structure and expression of ribosomal protein genes in yeast, in: Structure,
Function, and Genetic rf Ribosomes (B. Hardesty and G. Kramer, eds.), Springer-Verlag,
New York, pp. 699-718.
Prasad, R., 1985, Lipids in the structure and function of yeast membrane, Adv. Lipid Res.
21:187-242.
Prasad, R., and Rose, A. H., 1986, Involvement of lipids in solute transport in yeasts, Yeast
2:205-220.
Pringle, J. R., 1975, Methods for avoiding proteolytic artefacts in studies of enzymes and
other proteins from yeast, Meth. Cell Biol. 12:149-184.
Pringle, J. R., and Hartwell, L. H., 1981, The Saccharomyces cerevisiae cell cycle, in: The
Molecular Biology rf the Yeast Saccharomyces, Life Cycle and Inheritance (J. N. Strathern, E.
W. Jones, and J. R. Broach, eds.), Cold Spring Harbor Monograph Series, Cold Spring
Harbor, NY, pp. 97-142.
Racker, E., 1950, Spectrophotometric measurements of the enzymatic formation of

fumaric and cis-aconitic acids, Biochim. Biophys. Acta 4:211-214.
Ramirez, R. M., Ishida-Schick, T., Krilowicz, B. L., Leish, B. A., and Atkinson, K. D., 1983,
Plasma membrane expansion terminates in Saccharumyces cerevisiae secretion-defective
mutants while phospholipid synthesis continues,]. Bacterial. 154:1276-1283.
Ramirez, J. M., Gallego, G. G., and Serrano, R., 1984, Electron transfer constituents in
plasma membrane fractions of Avena sativa and Saccharumyces cerevisiae, Plant Sci. Lett.
34:103-110.
Rank, G. H., Gerlach,J. H., Robertson, A.J., and Van Hoeven, R. P., 1978, High viscosity
vesicles of yeast separated at pH 4 have surface glycoproteins, Nature 273:682-684.
Rattray, J. B. M., Schibeci, A., and Kidby, D. K., 1975, Lipids of yeast, Bacterial. Rev.
39:197-231.
Rickwood, D., and Hayes, A., 1984, An evaluation of methods used to prepare yeast
mitochondria for transcriptional studies, Prep. Biochem. 14:163-171.
Riezman, H., 1985, Endocytosis in yeast: Several of the yeast secretory mutants are defec-
tive in endocytosis, Cell 40:1001-1009.
Riezman, H., Hay, R., Gasser, S., Daum, G., Schneider, G., Witte, C., and Schatz, G., 1983,
The outer membrane of yeast mitochondria: Isolation of outside-out sealed vesicles,
EMBO]. 2:1105-1111.
Riezman, H., Chvatchko, Y., and Dulic, V., 1986, Endocytosis in yeast, Trends Biochem. Sci.
11:325-328.
Roberts, R. L., Bowers, B., Slater, M. L., and Cabib, E., 1983, Chitin synthesis and localiza-
tion in cell division cycle mutants of S. cerevisiae, Mol. Cell. Biol. 3:922-930.
Robertson, A. J., Gerlach, J. H., Rank. G. H., and Fowke, L. C., 1980, Yeast cell wall,
membrane, and soluble marker polypeptides identified by comparative two-dimen-
sional electrophoresis, Can.]. Biochem. 58:567-572.
Rodriguez, J. R., Pikielny, C. W., and Rosbash, M., 1984, In vivo characterization of yeast
mRNA processing intermediates, Cell39:603-610.
Rosamond, J., 1982, The molecular biology of the mitochondrion, Biochem. ]. 202:1-8.
Roth blatt, j. A., and Meyer, D. 1., 1986a, Secretion in yeast: Reconstitution of the transloca-
tion and glycosylation of a-factor and invertase in a homologous cell-free system, Cell
44:619-628.
Rothblatt, J. A., and Meyer, D. E., 1986b, Secretion in yeast: Translocation and glycosyla-
tion of prepro-a-factor in vitro can occur via an ATP-dependent post-translational
mechanism, EMBO ]. 5:1031-1036.
Rubtsov, P. M., Musakhanov, M. M., Zakharyev, V. M., Krayev, A. S., Skryabin, K. G., and
Bayev, A. A., 1980, The structure of the yeast ribisomal RNA genes. I. The complete
nucleotide sequence of the ISS ribosomal RNA gene from Saccharomyces cerevisiae,
Nucl. Acids Res. 8:5779-5794.
Sanz, P., Herrero, E., and Sentandreu, R., 1987, Secretory pattern of a major integral
mannoprotein of the yeast cell wall, Biochim. Biophys. Acta 924:193-203.
Schatz, G., and Butow, R. A., 1983, How are proteins imported into mitochondria? Cell
32:316-318.
Schatz, G., and Klima, j., .1964, Triphosphopyridine nucleotide: Cytochrome c reductase
of Saccharumyces cerevisiae: a microsomal enzyme, Biochim. Biophys. Acta 81:448-461.
Schatz, P. J., Pillus, L., Grisafi, P., Solomon, F., and Botstein, D., 1986a, Two functional a-
tubulin genes of the yeast Saccharumyces cerevisiae encode divergent proteins, Mol. Cell
Biol. 6:3711-3721.
Schatz, P. J., Solomon, F., and Botstein, D., 1986b, Genetically essential and non-essential
a-tubulin genes specify functionally interchangeable proteins. Mol. Cell Biol. 6:3722-
3733.
Schekman, R., 1982, Biochemical markers for yeast organelles, in: The Molecular Biology of
the Yeast Saccharpmyces, Metabolism and Gene Expression (J. N. Strathern, E. W. jones, and
J. R. Broach, eds.), Cold Spring Harbor Monograph Series, Cold Spring Harbor, NY,
pp. 651-652.
Schekman, R., I985, Protein localization and membrane traffic in yeast, Annu. Rev. Cell
Biol. 1:167-195.
Schekman, R., and Brawley, V. L., 1979, Localized deposition of chitin on the yeast cell
surface in response to pheromone, Proc. Natl. Acad. Sci. USA 76:645-649.
Schekman, R., and Novick, P., 1982, The secretory process and yeast cell surface assembly,
in.; The Molecular Biology rf the Yeast Saccharomyces, Metabolism and Gene Expression (J. N.
Strathern, E. W. Jones, and J. R. Broach, eds.), Cold Spring Harbor Monograph
Series, Cold Spring Harbor, NY, pp. 361-393.
Schmidt, R., Ackermann, R., Kratky, Z., Wassermann, B., and Jacobson, B., 1983, Fast and
efficient purification of yeast plasma membranes using cationic silica microbeads,
Biochim. Biophys. Acta 782:421-427.
Schneider, J. C., and Guarente, L., 1987, The untranslated leader of nuclear COX4 gene
of Saccharomyces cerevisiae contains an intron, Nucl. Acids Res. 15:3515-3529.
Schultz, L. D., 1978, Transcriptional role of yeast deoxyribonucleic acid dependent
ribonucleic acid polymerase Ill, Biochemistry 17:750-758.
Schwaiger, H., Hasilik, A., von Figura, K. Wiemken, A., and Tanner, W., 1982, Carbohy-
drate-free carboxypeptidase Y is transferred into the lysosome-like yeast vacuole,
Biochem. Biophys. Res. Commun. 104:950-956.
Schwartz, D. C., and Cantor, C. R., 1984, Separation of yeast chromosome-sized DNAs by
pulse field gradient gel electrophoresis, Cell 87:67-75.
Schwencke, J., 1977, Characteristics and integration of the yeast vacuole with cellular
functiop, Physiol. Veg. 15:491-517.
Schwenke, J., Canut, H., and Flores, A., 1983, Simultaneous isolation of the yeast cytosol
and well-preserved mitochondria with negligible contamination by vacuolar pro-
teinases, FEBS Lett. 156:274-280.
Sentandreu, R., Herrero, E., and Elorza, M. V., 1984, The assembly of wall polymers in
yeast, FEMS Symp. 27:51-61.
Shematek, E. M., and Cabib, E., 1980, Biosynthesis of the yeast cell wall. II. Regulation of
13-(1->3) glucan synthetase by ATP and GTP,]. Biol. Chem. 255:895-902.
Shematek, E. M., Braatz, J. A., and Cabib, E., 1980, Biosynthesis of the yeast cell wall:
I. Preparation and properties of (3-(1->3) glucan synthetase,]. Biol. Chem. 255:888-
894.
Sherman, F., 1982, Suppression in the yeast Saccharomyces cerevisiae, in: The Molecular
Biology rfthe Yeast Saccharomyces, Metabolism and Gene Expression (J. N. Strathern, E. W.
Jones, and J. R. Broach, eds.), Cold Spring Harbor Monograph Series, Cold Spring
Shortie, D., Harber, J., and Botstein, D., 1982, Lethal disruption of the yeast actin gene by
DNA transformation, Science 217:371-373.
Shortie, D., Novick, P., and Botstein, D., 1984, Construction and genetic characterization
of temperature-sensitive alleles of the yeast actin gene, Proc. Natl. Acad. Sci. USA
81:4889-4893.
Shulman, R. W., 1978, Yeast cell selection and synchrony: Density gradients and mating
factor block, in: Methods in Cell Biology, Vol. 20 (D. M. Prescott, ed.), Academic Press,
New York, pp. 43-45.
Silver, P. A., Keegan, L. P., and Ptashne, M., 1984, Amino terminus of the yeast GAL4 gene
product is sufficient for nuclear localization, Proc. Natl. Acad. Sci. USA 81:5951-5955.
Simon, M., Seraphin, B., and Faye, G., 1986, KIN28, a yeast split gene coding for a
putative protein kinase homologous to CDC28, EMBO f. 5:2697-2701.
Singh, A., Chen, E. Y., Lugovoy,J. M., Chang, C. N., Hitzman, R. A., and Seeburg, P. H.,
1983, Saccharomyces cerevisiae contains two discrete genes coding for the a-factor phe-
romone, Nucl. Acids Res. 11:4049-4063.
Sprague, G. F., Blair, L. C., and Thorner, J., 1983, Cell interactions and regulation of cell
type in the yeast Saccharomyces cerevisiae, Annu. Rev. Microbiol. 37:623-660.
Sprinzl, M., Moll, J., Meissner, F., and Hartmann, T., 1985, Compilation of tRNA se-
quences, Nucl. Acids Res. 15 (Suppl.):r1-r50.
Steere, R. L., and Erbe, E. F., and Moseley, J. M., 1980, Prefracture and cold-fracture
images of yeast plasma membranes,]. Cell Biol. 86:113-122. .
Stevens, B., 1981, Mitochondrial structure, in: The Molecular Biology cf the Yeast Sac-
charomyces, Life Cycle and Inheritance (J. N. Strathern, E. W. Jones, and J. R. Broach,
Stotzler, D., Kiltz, H.-H., and Duntze, W., 1976.,Priinary structure of a-factor peptides
from Saccharomyces cerevisiae, Eur.]. Biochem. 69:397-400.
Swida, U., Kreutzfeldt, C., Ramezani-Rad, M., and KAufer, N., 1982, Isolation and charac-
terization of rough and smooth endoplasmic reticulum from Saccharomyces cerevisiae,
FEMS Microbiol. Lett. 15:313-318.
Tanner, W., 1984, Regulation of glycoprotein synthesis in yeast by mating pheromones, in:
Structure, Function, and Biosynthesis cf Plant Cell WalU, Proc. Annu. Symp. Bot., 7th, (M. W.
Dugger and S. Bartinicki-Garcia, eds.), Waverly Press, Baltimore, pp. 20-32.
Tanner, W., and Lehle, L., 1987, Protein glycosylation in yeast, Biochim. Biophys. Acta
906:81-99.
Tarentino, A. L., Plummer, T. H., and Maley, F., 1974, The release of intact oligosac-
charides from specific glycoproteins by endo-~-N-acetylglucosaminidase H, ]. Biol.
Chem. 249:818-824.
Teem, J. L. Abovich, N., KAufer, N. F., Schwindinger, W. F., Warner, J. R., Levy, A.,
Wooldford, J., Leer, R. J., van Ramsdonk-Duin, M. M. C., Mager, W. H., Planta, R. J.,
Schultz, L., Friesen, J.D., Fried, H., and Rosbash, M., 1984, A comparison of yeast
ribosomal protein gene DNA sequences, Nucl. Acids JUs. 12:8295-8312.
Thorner, j., 1981, Pheromonal regulation of development in Saccharomyces cerevisiae, in:
The Molecular Biology rf the Yeast Saccharomyces, Life Cycle and Inheritance (J. N. Strathern,
E. W. jones, and J. R. Borach, eds.), Cold Spring Harbor Monograph Series, Cold
Thorner, J., 1982, An essential role of cyclic AMP in growth control: The case of yeast, Cell
30:5-6.
Tijssen, J. P. F., Beekes, H. W., and van Steveninck, J., 1981, Localization of polyphos-
phates at the outside of the yeast cell plasma membrane, Biochim. Biophys. Acta
649:529-532.
Tohoyama, H., and Yanagishima, N., 1981, Changes in the sexual agglutination ability
during the formation of vegetative cells from spores in Saccharomyces cerevisiae, Mol.
Gen. Genet. 183:205-208.
Tohoyama, H., Yoshida, M. H. K., and Yanagishima, N., 1979, Regulation of the produc-
tion of the agglutination substances responsible for sexual agglutination in Sac-
charomyces cerevisiae: Changes associated with conjugation and temperature shift, Mol.
Gen. Genet. 174:269-280.
Towler, D., and Glaser, L., 1986, Protein fatty acid acylation: Enzymatic synthesis of an N-
myristoylglycyl peptide, Proc. Natl. Acad. Sci. USA 85:2812-2816.
Trembath, M. ~ .• and Tzagoloff, A., 1979, Large- and small-scale prepf!rations of yeast
mitochondria, Meth. Enzymol. 55:160-170.
Trimble, R. B., and Atkinson, P. H., 1986, Structure of yeast external invertase
Mans. 14GlcNAc processing intermediates by 500-megahertz 'H NMR spectroscopy,].
Biol. Chem. 261:9815-9824.
Trimble, R. B., Maley, F., and Chu, F. L., 1983, Glycoprotein biosynthesis in yeast: Protein
conformation affects processing of high mannose oligosaccharides and carboxypep-
tidase Y and invertase,]. Biol. Chem. 258:2562-2567.
Trivedi, A., Fantin, D. J., and Tustanoff, E. R., 1985, Saccharomyces cerevisiae as a model
eukaryote for studies on mitochondriogenesis, Microbiol. Sci. 2:10-13.
Tsai, P. -K., Frevert, J., and Ballou, C. E., 1984, Carbohydrate structure of Saccharomyces
cerevisiae mnn9 mannoprotein,]. Biol. Chem. 259:3805-3811.
Tschopp, J., and Schekman, R., 1983, Two distinct subfractions in isolated Saccharomyces
cerevisiae plasma membranes,]. Bacteriol. 156:222-229.
Tschopp. J., Esmon, P. C., and Schekman, R., 1984, Defective plasma membrane assembly
in yeast secretory mutants,]. Bacteriol. 160:966-970.
Urech, K., Diirr, M., Boller, T., and Wiemken, A., 1978, Localization of polyphosphate in
vacuoles of Saccharomyces cerevisiae, Arch. Microbiol. 116:275-278.
Valentin, E., Herrero, E., Pastor, F. I. J., and Sentandreu, R., 1984, Solubilization and
analysis of mannoprotein molecules from the cell wall of Saccharomyces cerevisiae, ].
Gen. Microbial. 150:1419-1428.
Van der Wilden, W., Matile, P., Schellenberg, M., Meyer, J., and Wiemken, A., 1973,
Vacuolar membranes: Isolation from yeast cells, Z. Naturforsch. 28c:416-42l.
Veldman, G. M., Klootwijk, J., de Regt, V. C. H. F., and Planta, R. J., 198la, The primary
and secondary structure of yeast 26S rRNA, Nucl. Acids Res. 9:6936-6952.
Veldman, G. M., Klootwijk, J. van Heerikhuizen, H., and Planta, R. J., 1981b, The nu-
cleotide sequence of the intergenic region between the 5.8S and 26S rRNA genes of
the yeast ribosomal RNA operon. Possible implications for the interactions between
5.8S and 26S rRNA and the processing of the primary transript, Nucl. Acids Res.
9:4847-4862.
Vignais, M-L., Woudt, L. P., Wassenar, G. M., Mager, W. H., Sentenac, A., and Planta, R.J.,
1987, Specific binding of TUF factor to upstream activation sites of yeast ribosomal
protein genes, EMBO]. 6:1451-1457.
Walter, P., and Lingappa, V. R., 1986, Mechanism of protein translocation across endo-
plasmic reticulum membrane, Annu. Rev. Cell Biol. 2:499-516.
Warner, J. R., 1982, The yeast ribosome: Structure, function, and synthesis, in: The Mo-
lecular Biology rfthe Yeast Saccharomyces, Metabolism and Gene Expression (J. N. Strathern,
E. W. Jones, and J. R. Broach, eds.), Cold Spring Harbor Monograph Series, Cold
Warner, J. R., Mitra, G., Schwindinger, W. F., Studeng, M., and Fried, H. W., 1985, Sac-
charomyces cerevisiae coordinates accumulation of yeast ribosomal proteins by modulat-
ing mRNA splicing, translational initiation, and protein turnover, Mol. Cell Biol.
5:1512-1521.
Warner, J. R., Elion, E. A., Dabeva, M. D., and Schwindinger, W. F., 1986, The ribosomal
genes of yeast and their regulation, in: Structure, Function, and Genetic rf Ribosomes (B.
Hardesty and G. Kramer, eds.), Springer-Verlag, New York, pp. 719-732.
Waters, M. G., and Blobel, G., 1986, Secretory protein translocation in a yeast cell-free
system can occur posttranslationally and requires ATP hydrolysis, ]. Cell Biol.
102:1543-1550.
Welch,J. W., and Burlingame, A. L., 1973, Very long chain fatty acids in yeast,]. Bacterial.
115:464-466.
Welten-Verstegen, G. W., Boer, P., and Stern-Parve, E. P., 1980, Lipid-mediated glycosyla-
tion of endogenous proteins in isolated plasma membranes of Saccharomyces cerevisiae,
]. Bacteriol. 141:342-349.
Wen, D., and Schlesinger, M. J., 1984, Fatty acid-acylated proteins in secretory mutants of
Saccharomyces cerevisiae, Mol. Cell Biol. 4:688-694.
Wiemken, A., Matile, P., and Moor, H., 1970, Vacuolar dynamics in synchronously bud-
ding yeast, Arch. Microbial. 70:89-103.
Wiemken, A., Schellenberg, M., and Urech, K., 1979, Vacuoles: The sole compartments of
digestive enzymes in yeast (Saccharomyces cerevisiae), Arch. Microbial. 123:23-35.
Witt, W., Schweingruber, M. E., and Mertsching, A., 1984, Phospholipase B from the
plasma membrane of Saccharomyces cerevisiae: Separation of two forms with different
carbohydrate content, Biochim. Biophys. Acta 795: 108-116.
Wolf, D. H., 1986, Cellular control in the eukaryotic cell through action of proteinases:
The yeast Saccharomyces cerevisiae as a model organism, Microbial. Sci. 5:107-114.
Woudt, L. P., Smit, A. B., Mager, W. H., and Planta, R. J., 1986, Conserved sequence
elements upstream of the gene encoding yeast ribosomal protein L25 are involved in
transcription activation, EMBO ]. 5:1037-1040.
Yaffe, M., 1983, Import of proteins into mitochondria, a survey, in: Mitochondria '83,
Nucleo-mitochrondriallnteractions (R. J. Schweijen, K. Wolf, and F. Kaudewitz, eds.),
Walter deGruyter Verlag, Berlin, pp. 47-55.
Yamaguchi, M., Yoshida, K., and Yanagishima, N., 1984, Isolation and biochemical and
biological characterization of an a-mating-type-specific glycoprotein for sexual ag-
glutination from the cytoplasm of a-cells in the yeast Saccharomyces cerevisiae, Arch.
Microbial. 140: 113-119.
Zickler, D., and Olson, L. W., 1975, The synaptonemal complex and the spindle plaque
during meiosis in yeast, Chromosoma 50:1-23.
Zimmerman, R., and Meyer, D. 1., 1986, 1986: A year of new insights into how proteins
cross membranes, Trends Biochem. Sci. 11:512-515.
Zinker, S., and Warner, J. R., 1976, The ribosomal proteins of Saccharomyces cerevisiae,
phosphorylated and exchangeable proteins,]. Biol. Chem. 251: 1799-1807.
Z1otnik, H., Fernandez, M. P., Bowers, B., and Cabib, E., 1984, Saccharomyces cerevisiae
mannoproteins form in external cell wall layer that determines wall porosity,]. Bac-
terial. 159:1018-1026.
Metabolism and Biosynthesis
3
J. R. DICKINSON
1. INTRODUCTION
A recent Symposium on Yeast Cell Biology was opened with the state-
ment "In ten years yeast has changed from E[scherichia] coli with a nu-
cleus to the universal cell" (Hartwell, 1985). This bold assertion under-
lines the current importance of yeast as an experimental system. In
terms of our understanding of metabolism and biosynthesis, the extent
of our knowledge of Saccharomyces cerevisiae is surpassed only in the case
of E. coli. Nevertheless, despite this mass of information we are still
ignorant of the actual steps in a number of primary metabolic pathways.
In addition, many important details about the regulation of individual
pathways (the means whereby the cell integrates the many separate path-
ways to effect growth and proliferation) remain undiscovered. Further-
more, the greater structural complexity of the eukaryotic yeast cell (com-
pared with the prokaryotic E. coli) means that compartmentalization of
metabolites is a further complication, the significance of which is now
starting to be seriously considered by more and more researchers.
At this point it is appropriate to give a note of caution. The new-
comer to yeast metabolism is urged not to survey the general schemes of
metabolic pathways that appear in biochemistry textbooks and as wall-
charts. These are only compilations, the information having been de-
rived from numerous different systems. Not all of these pathways exist
in yeast. Furthermore, yeast does some things differently from E. coli
(the model system on which many metabolic pathways are based). For
example, the biosynthesis of lysine is accomplished by entirely different
pathways in yeast and bacteria. The reviews on carbohydrate metabolism
by Fraenkel ( 1982) and on amino acid and nucleotide biosynthesis by
Jones and Fink ( 1982) are excellent starting points for newcomer and
expert alike.
J. R.
DICKINSON • School of Pure and Applied Biology, University of Wales College
of Cardiff, Cardiff CFI 3TL Wales.
59
60 J. R. DICKINSON
The genetics of S. cerevisiae are better worked out than for any other
eukaryote. In addition, the organism is amenable to genetic manipu-
lation. As a result, metabolic questions can be examined far more
extensively and in greater detail than would otherwise be possible. For
example, we are not just limited to examining the effects of individual
mutations. Using "conventional" yeast genetic methods (described else-
where in this volume) it is possible to combine different mutations into a
single cell. The potential for examining control of a metabolic pathway
then becomes apparent especially where gene-enzyme relationships in a
pathway are known. Comparisons between haploids, homozygous, and
heterozygous diploids, and the use of cloned genes maintained in the
yeast cell at defined copy number, enable an understanding of impor-
tance of gene dosage in various aspects of metabolic control.
Probably the best-known feature of Saccharomyces is its facultative
nature; that is, it can be grown aerobically or anaerobically. Thus, the
effects of a mutation affecting glycolysis can be studied by growing
the cells aerobically on (say) ethanol, whereas if one wants to examine the
consequences of defects in oxidative (mitochondrial) functions, the cells
can be grown by fermentation of glucose. This convenient feature of yeast
has been exploited fully by those interested in the metabolism, structure,
and biogenesis of mitochondria. Undoubtedly our knowledge of mito-
chondria would be greatly diminished without work done on yeast.
2. OUTLINE OF CARBOHYDRATE METABOLISM
Carbohydrate metabolism in Saccharomyces has been reviewed by Fraen-

kel ( 1982). Two of the most pertinent points made in that review are: ( 1)
strains of Saccharom,ces used for biochemical and genetic study grow best
by fermentation, even aerobically, and (2) the enzymes of glycolysis rep-
resent 30-65% (depending on estimate) of soluble protein: thus they are
a major component of total cellular protein. There are five physiological
phenomena in Saccharomyces which need to be considered every time we
grow a culture. These are catabolite repression (usually only glucose
repression is considered), catabolite inactivation, the induction of respi-
ratory enzymes by oxygen, the Pasteur effect, and catabolite conversion.
2.1. Catabolite Repression

Catabolite repression is the lowering of the levels of cellular func-
tions involved in respiration by a carbon source such as glucose. In batch
culture at the initially high substrate concentrations growth will be ac-
complished by fermentation. As the substrate is consumed, its con-
centration falls and the cells become derepressed; i.e., induction of res-
METABOLISM AND BIOSYNTHESIS 61
piratory enzymes and components of electron transport occurs. It has

been shown that the hexokinase PII isozyme is bifunctional and has, in
addition to its "normal" catalytic function, a regulatory domain for trig-
gering glucose repression (Entian and Frohlich, 1984; Entian et al.,
1984; Frohlich et al., 1984).
Other sugars are less repressing than glucose, and the means where-
by the cell monitors the concentration and identity of the carbon source
have still to be worked out. However, some progress should result from
the finding that the SNF 1 gene product is a protein kinase (Celenza and
Carlson, 1986). Upon glucose deprivation the SNF1 function is required
in the expression of glucose-repressible genes involved in the metabo-
lism of sucrose, galactose, maltose, melibiose, and nonfermentable car-
bon sources. It is now known that SNF 1 is the same gene as CCR1 (also
known as CAT1) (Celenza and Carlson, 1986). Further clues are likely to
come from the identification of upstream activation sites (UAS) of glu-
cose-repressible genes. These sequences have been identified by deletion
analysis, deletion within the UAS of a structural gene resulting in in-
ability to derepress synthesis of the respective enzyme. This allows iden-
tification of a formal genetic model in which derepression by a positive
effector binding to DNA permits transcription. Clearly, a complete bio-
chemical description of the phenomenon of catabolite repression is still
some way off but now seems a realistic possibility.
2.2. Catabolite Inactivation

Catabolite inactivation refers to the following sequence of events:
Addition of glucose or a metabolically related sugar to cells growing on
acetate or ethanol results in a loss of activity of the key gluconeogenic
enzymes fructose 1,6-bisphosphatase, cytoplasmic malate dehydrogen-
ase, and phosphoenolpyruvate carboxy kinase (Gancedo, 1971; Witt et al.,
1966; Ferguson et al., 1967; Haarasilta and Oura, 1975). The advantages
of such a mechanism to the cell are obvious: it no longer needs to perform
gluconeogenesis in the presence of glucose. Glycolysis using the newly
added glucose with simultaneous gluconeogenesis from the original ace-
tate (or ethanol) would be metabolically futile. In the case of fructose 1,6-
bisphosphatase inactivation occurs in two stages. The first stage is rapid,
with a decrease of activity of 60% occurring within about 3 min; this is
accomplished by phosphorylation of the enzyme (Tortora et al., 1981;
Muller and Holzer, 1981; Mazon et al., 1982a). The second stage, a
proteolytic degradation, takes about 1 hr (Funayama et al., 1980; Tortora
et al., 1981). It has been proposed that addition of glucose causes an
increase in adenosine 3':5'-cyclic monophosphate (cAMP) and that the
phosphorylation occurs by the action of a cAMP-dependent protein
kinase (Purwinetal., 1982; Vander Platt and Van Solingen, 1974; Tortora
62 J. R. DICKINSON
et al., 1982; Mazon et al., 1982b). A study of the effect of glucose on the
three gluconeogenic enzymes in a mutant deficient in the regulatory
subunit of cAMP-dependent protein kinase (bcyl mutation), and in a
double mutant that also carried the cyrl mutation (CYRJ is the gene
coding for adenylate cyclase, the enzyme responsible for the synthesis of
cAMP from ATP), confirmed the involvement of cAMP in catabolite
inactivation but indicated that other mechanisms are also involved (Tor-
tora et al., 1984).
2.3. Effect of Oxygen

The formation of intact functional mitochondria with fully active
respiratory enzymes requires molecular oxygen. Despite its importance,
virtually nothing is known about the way this occurs. It is well known
that utilization of substrates such as glycerol, ethanol, pyruvate, and
lactate occurs oxidatively using the tricarboxylic acid (TCA) cycle and
mitochondrial electron transport. Growth on acetate requires the glyox-
ylate cycle in addition. However, it is often not appreciated that TCA
cycle enzyme activities are still present in anaerobic conditions. TCA
cycle activity is needed for the provision of intermediates for biosynthe-
sis. Reference to schemes of metabolic pathways will confirm that, for
example, the TCA cycle intermediate 2-oxoglutarate is essential in the
biosynthesis of glutamate, glutamine, proline, lysine, and arginine Qones
and Fink, 1982). Furthermore, glutamate, which is derived from 2-ox-
oglutarate (via NADP-dependent glutamate dehydrogenase), is recog-
nized as a nitrogen source in Saccharomyces and is itself used in the
biosynthesis of aspartate, and hence threonine, methionine, and cys-
teine, and also tyrosine, phenylalanine, serine, leucine, isoleucine, and
histidine. Clearly, we are now in the realms of nitrogen anabolic path-
ways. Thus, although carbon and nitrogen metabolism are frequently
considered as separate aspects (e.g., in reviews and by researchers in
their investigations), it is apparent that this division is artificial, made
only for convenience, and that the yeast cell does not recognize such a
separation. Accordingly, after the introduction to nitrogen metabolism
(Section 3), this chapter will take an holistic view to emphasize the unity
of cellular metabolism.
2.4. The Pasteur Effect

The Pasteur effect is the control exerted by respiration over glycoly-
tic rate. When cells are growing fermentatively under anaerobic condi-
tions, the rate of glucose utilization will be high. A switch to aerobic
conditions results in the onset of respiration with a consequent rapid
decrease in the rate of fermentation (glycolysis). The advantage of such a
mechanism to the cell is explained in terms of the far greater yield of

ATP available from complete oxidation of its carbon source via respira-
tion compared with glycolysis. Canonical wisdom states that the Pasteur
effect is mediated primarily by the allosteric regulation of phosphofruc-
tokinase, which is considered to be the chief "pacemaker" enzyme of
glycolysis. Among the components that have been shown to stimulate
phosphofructokinase are AMP, ADP, inorganic phosphate, and fruct-
ose-6-phosphate (fructose-6-P). ATP and citrate are known to inhibit the
enzyme. Thus when respiration is proceeding at high rates, oxidative
phosphorylation will be taking place, producing large amounts of ATP
and consuming ADP, AMP, and phosphate, and increased levels of TCA
cycle activity will give rise to higher concentrations of citrate than would
occur during anaerobic glycolysis. This will result in inhibition of phos-
phofructokinase (by ATP and citrate) and reduced stimulation from
AMP, ADP, and phosphate. Conversely, when mitochondrial activity is
low, the ATP concentration will be low, so there will be little or no
inhibition of phosphofructokinase. In addition, ADP and AMP con-
centrations will be high, thus stimulating phosphofructokinase and
thereby increasing glycolytic flux. Notice also that phosphofructokinase
is stimulated by its substrate (fructose-6-P).
This view of the Pasteur effect cannot be the complete story, how-
ever. Just one of the questions in this area is the difficulty in reconciling
in vivo glycolytic flux through phosphofructokinase with in vitro kinetic
evidence. A typical intracellular concentration of fructose-6-P is 0.2 mM
whereas maximal activity of phosphofructokinase (required to produce
the in vivo glycolytic fluxes observed is only attained in vitro as substrate
concentrations of around 5 mM. Fructose-2,6-bisphosphate, not the
product of the phosphofructokinase (fructose-6-P 1-kinase) reaction,
but produced by fructose-6-P 2-kinase, is a potent stimulator of phos-
phofructokinase (Clifton and Fraenkel, 1983). It seems likely, therefore,
that this compound is also important in regulating phosphofructokinase
activity in vivo. Studies in wild-type strains and in mutants affected in
aspects of glycolysis have revealed there are also other controls operat-
ing on the overall glycolytic rate. Some of these are mentioned in subse-
quent sections of this chapter.
2.5. Catabolite Conversion

The term catabolite conversion (Frohlich and Entian, 1982) is not
generally well known. This effect is seen in the conversion of enolase
isoenzymes after addition of glucose to derepressed cells. Enolase is the
first gluconeogenic enzyme that is also required for glycolysis, and it has
been demonstrated that conversion of gluconeogenic enolase I to glyco-
lytic enolase II is an additional regulatory mechanism to prevent
64 J. R. DICKINSON
gluconeogenesis in the presence of glucose. A further control accom-

panying the coordinated decrease in activity of enolase I and increase in
activity of enolase II is the simultaneous repression of isoenzyme I and
induction of isoenzyme II (Entian et al., 1987).
Table I. List of Saccharomyces Genes Involved in Carbon Metabolism

That Have Been Cloneda
Gene
designation Gene product Reference
ADHJ (ADCJ) (Fermentative) alcohol dehydro- Williamson et al. (1980); Young

genase I et al. ( 1982)
ADH2 (ADR2) (Oxidative) alcohol dehydro- Young et al. (1982)
genase II
ADHJ (ADM) (Mitochondrial) alcohol dehydro- Young et al. (1982)
genase
ADRI [Regulator of alcohol dehydro- Denis and Young (1983)
genase II synthesis)
ADR6 Required for derepression of a!- Taguchi and Young (1987)
coho! dehydrogenase II
CHOJ Phosphatidyl serine synthase Letts et al. ( 1983)
CITJ (CLUJ) Citrate synthase Suissa et al. (1984)
ENOl Enolase I Holland et al. (1981)
EN02 Enolase II Holland et al. (1981)
FASJ Fatty acid synthetase Kuziora et al. (1983); Roberts
FAS2 Fatty acid synthetase and Schewizer (1984)
FBPI Fructose bisphosphatase Sedivy and Fraenkel (1985)
GALl Galactokinase St. John and Davis (1979);
Miyajima et al. (1984)
GAL2 Membrane-bound subunit of Tschopp et al. (1986)
galactose permease
GAL4 (ga/81) [Transcriptional activator of Laughon and Gesteland (1982);
GALl, GAL2, GAL7, GALlO, Johnston and Hopper ( 1982)
and MELJ)
GAL7 Galactose-1-phosphate uridyl- St. John and Davis ( 1979)
transferase
GALlO UDP-glucose-4-epimerase St. John and Davis ( 1979)
GALBO [Regulator of GAL4 gene prod- Nogi et al. (1984)
uct activity)
GCRI Positive transcriptional regulator Holland et al. ( 1987)
of enolase and glyceralde-
hyde-3-P-dehydrogenase genes
GLKI Glucokinase Walsh et al. (1983)
GPMI Phosphoglycerate mutase Kawasaki and Fraenkel ( 1982)
HEX2 [Needed for glucose repression, Niederacher and Entian (1987)
regulates hexokinase PII syn-
thesis)
Table I. (Continued)
Gene Reference
designation Gene product
HXKJ Hexokinase PI (A) Walsh et al. (I 983); Entian et

al. (1984)
HXK2 (HEXJ) Hexokinase PII (B) Walsh et al. (1983); Frohlich et
al. (1984)
JNOJ Inositol-1-phosphate synthase Klig and Henry (1984)
LPDJ Lipoamide dehydrogenase Roy and Dawes (1987)
MAL/1 Maltose permease
MAL/2 Maltase Charron et al. (1986)
MAL/3 Positive regulator of MAL/2
gene
MAL2-8C Rodicio and Zimmerman
(1985a)
[Regulator of a-glucosidase syn-
MAL3 Michels and Needleman (1983)
thesis]
MAL4 Rodicio and Zimmerman
(1985b)
MAL61 Maltose permease
MAL62 Maltase Dubin et al. ( 1986)
MAL63 Positive regulator of MAL/2 gene
PDCJ Pyruvate decarboxylase Schmitt et al. ( 1983)
PFKJ Phosphofructokinase o: subunit Clifton and Fraenkel (1982);
Heinisch ( 1986)
PFK2 Phosphofructokinase ~ (regula- Heinisch (1986)
tory?) subunit
PGll Phosphoglucose isomerase Kawasaki and Fraenkel ( 1982);
Aguilera and Zimmerman
(1986)
PGKJ Phosphoglycerate kinase Hitzeman et al. (1980); Dobson
et al. (1982); Kawasaki and
Fraenkel (1982)
PGMJ (GPMJ) Phosphoglycerate mutase Rodicio and Heinisch (1987)
PJSJ Phosphatidyl inositol synthase Nikawa and Yamashita (1984)
PYKJ (CDCJ9) Pyruvate kinase Kawasaki and Fraenkel (1982);
Burke et al. ( 1983)
SNFJ [Derepressor of many glucose- Celenza and Carlson (1984)
repressible genes)
SNF3 "High-affinity" glucose Bisson et al. ( 1987)
transporter
SUC2 Invertase Hohmann and Zimmermann
(1986)
TDHJ Glyceraldehyde-3-P-dehydro- Holland and Holland (1980);
genase I McAlister and Holland
TDH2 Glyceraldehyde-3-P-dehydro- (1985)
genase II
TDH3 Glyceraldehyde-3-P-dehydro-
genase III
TPll Triose phosphate isomerase Kawasaki and Fraenkel (1982)
•Former or alternative gene designations are given in parentheses.

66 J. R. DICKINSON
2.6. Molecular Genetics of Carbon Metabolism

The S. cerevisiae glycolysis genes and other genes involved in carbon
metabolism that have been cloned are shown in Table I.
3. OUTLINE OF NITROGEN METABOLISM
Nitrogen catabolism has been reviewed by Cooper (1982a). Amino

acid and nucleotide biosynthesis have been reviewed by Jones and Fink
(1982) and by Henry et al. (1984). Many compounds will serve as nitro-
gen source for S. cerevisiae though some are obviously better than others
(see Cooper, 1982a).
3.1. Amino Acid Biosynthesis

The pathways of amino acid biosynthesis are under complex regula-
tion at both genetic and metabolic levels. The so-called general control
of amino acid biosynthesis is involved in the coordinate repression and
derepression of 30 or more enzymes in different pathways. Derepres-
sion occurs upon amino acid starvation, repression takes place in condi-
tions of medium sufficiency. Each biosynthetic pathway is also charac-
terized by its own specific control mechanisms which operate with
respect to the end products and/or intermediates of the particular path-
way. The two levels of control coexist. The general control does not
derepress the synthesis of every enzyme in every pathway. For example,
all of the enzymes of the tryptophan biosynthetic pathway except phos-
phoribosyl-anthranilate isomerase are under general control, whereas in
the leucine pathway only a-isopropylmalate synthase is under general
control (Hsu et al., 1982). So far, the biosynthetic pathways for arginine,
histidine, isoleucine, leucine, lysine, tryptophan, and valine have been
shown to be regulated by the general control mechanism (Henry et al.,
1984).
Cloning of the structural genes encoding the amino acid biosynthet-
ic enzymes has been a major impetus toward unraveling the mechanism
of general control. Other useful genetic approaches have been the isola-
tion of mutants affected in general control of amino acid biosynthesis
and the subsequent cloning of the regulatory genes involved. Bio-
chemical and physiological studies have involved starving a given amino
acid auxotroph for the requisite amino acid or provision of an analog to
a wild-type strain that will cause inhibition of an enzyme in a particular
biosynthetic pathway. The end result is the same, for the cell becomes
depleted of the end product of the pathway (the amino acid) and subse-
quently derepresses the enzymes of the pathway along with all other
enzymes under general control.
METABOLISM AND.BIOSYNTHESIS 67
Analysis of pathway-specific regulation, which might be thought to

be a simpler problem, is not always straightforward. Obviously, study of
those pathways subject to general control will be more difficult than
study of those which are not. An additional complication occurs when an
amino acid can serve as a nitrogen source. Another possible snag (the
consideration of which has general application and is not limited solely
to questions concerning amino acid metabolism) involves the frequently
made assumption that provision of an intermediate to a mutant unable
to synthesize a particular compound will result in a pattern of metabo-
lism similar to that occurring in the wild type. Clearly, this approach
takes no account of the possible regulatory effects on either the immedi-
ate pathway concerned (e.g., feedback inhibition or controls of tran-
scription or translation of the remaining intact enzymes in that pathway),
effects on other related pathways (e.g., where cross-pathway controls are
exerted), or, assuming none of the foregoing gross disturbances occur,
the metabolic inbalance resulting from abnormal concentrations of an
intermediate in the wrong subcellular compartment.
3.2. Proteases
Saccharomyces cerevisiae possesses an extensive range of pro teases. An
excellent review on the synthesis and function of these enzymes has been
published by Jones ( 1984). Table II summarizes our current knowledge
of the properties, roles, cellular location, and genetics of these important
enzymes. More and more yeast proteases are being discovered each year,
though the physiological role of many of these remains obscure. Wolf
has commented that the "conventional" view of proteinase action of a
limited number of proteases being responsible for the degradation of
many intracellular proteins may have to be replaced by the idea of many
proteases being responsible for degradation, with each protease cutting
only a few cellular proteins (Wolf, 1982). If this view is correct, then
biochemical explanations of the specificity of protease action are sim-
plified, but the problem of the cellular regulation of many proteases
becomes more complex.
Proteases are considered to be involved in a wide variety of functions
both general and specific inS. cerevisiae. These include general protein
turnover and degradation, sporulation, protein degradation in the cell
cycle, catabolite inactivation, degradation of NAD-dependent glutamate
dehydrogenase upon nitrogen starvation and NADP-dependent gluta-
mate dehydrogenase following carbon starvation, zymogen processing,
formation and destn~ction of a-factor, removal of the products of mis-
sense and nonsense mutations and heterologous gene expression, "killer"
toxins, and secretion processes (see Jones, 1984, for an extensive list of
references). The major outstanding problems in this area are that there
Table II. Proteases of Saccharomyces cerevisiaea
Glyco- Cellular Structural

Enzyme Characteristics sylation location Cellular role(s) gene Other genes
Proteinase A Acid endoproteinase, + Vacuole N starvation-induced PRAJ PEP2-PEP15, ABM6,

1 X 41-80 kDa protein protein deg- ABMB
radation; protein
turnover in station-
ary phase
Proteinase B Serine sulfhydryl en- + Vacuole N starvation-induced PRBJ PEP2-PEP8, PEPll-
doproteinase, 1 X protein degradation; PEP16, PRB2-PRB4,
31-44 kDa spore dispersal RAD6, PS02
Carboxypeptidase Y Serine sulfhydryl ex- + Vacuole N starvation-induced PRCJ PEPJ-PEPJ6, PEP21,
opeptidase, 1 x protein degradation; ABM6,ABM8
59-63 kDa metabolism of pep-
tides
Aminopeptidase I Exopeptidase (Zn2+) + Vacuole Metabolism of pep- LAP4? PEP4
(LAPIV) 600 kDa tides
(12 X 53 kd)
Mitochondrial protease Metallo endopro- n.d. Mitochondrial Processing of imported Unknown, but
teinase, 115 kDa matrix precursors to nuclear
mitochondrial pro-
teins
Dipeptidyl aminopep- Cleaves X-Pro-Y and n.d. Membrane Processing of pher- STE13?
tidase A X-Ala-Y to yield omone precursors
X-Pro and X-Ala; (a-factor)
heat stable
Dipeptidyl aminopep- Cleaves X-Pro-Y and n.d. Membrane Unknown
tidase B X-Ala-Y to yield
X-Pro and X-ala;
heat labile
Lysine-arginine- Cleave C-terminal to n.d. Membrane Processing of a-factor KEX2?
cleaving endopep- Lys-Arg and Arg- and killer toxin pre-
tidase Arg pairs cursors
Proteinase M Cleaves oligopeptides n.d. Membrane Unknown
with blocked
a-amino groups
Proteinase P As for proteinase M; n.d. Membrane Unknown
cleavage catalyzed
may be the same as
that catalyzed by
pheromone pep-
tidase
Pheromone peptidase Cleaves a-factor be- n.d. Membrane? or Recovery from SST! (BAR!)?
tween Leu-6 and secreted into a-factor-indu ced
Lys-7 (see proteinase medium cell cycle arrest
P)
Aminopeptida se II Exopeptidase (Me2+ ), n.d. Periplasm and/or Unknown LAP!? RADI ?, PS02?,
(LAP I) 85 kDa cytoplasm? RAD6?
Aminopeptida se II Exopeptidase (Me2+ ), n.d. Periplasm and/or Unknown LAP2? RADI?, PS02?,
(LAPII) 85 kDa cytoplasm? RAD6?
Leucine aminopep- Exopeptidase, 94 kDa n.d. n.d. Unknown LAPJ?
tidase III (LAPIII)
Aminopeptida se Co Exopeptidase (Co2+), n.d. n.d. Unknown
100 kDa
Carboxypeptid ase S Exopeptidase (Zn2+) n.d. n.d. Metabolism of pep- CPS!?
tides (DUTJ?)
•Reproduced, with permission, from the Annual Review of Genetics, Volume 18, © 1984 by Annual Reviews Inc.
70 J. R. DICKINSON
are proteases with no known function and cellular processes which ob-
viously involve proteolytic action where the protease concerned has not
been identified. Much of this is due to methodology. Proteases are gener-
ally characterized in terms of their activity toward particular classes of
substrates and with respect to their pH optima. The substrates used in
protease assays are frequently either radioactively labeled peptides or
proteins coupled to chromogenic moieties. The reaction is terminated by
addition of cold trichloroacetic acid. Enzyme activity is measured in terms
of the amount of radioactivity or chromophore that has been rendered
trichloroacetic acid-soluble (unconsumed substrate remains trichloroa-
cetic acid-precipitable). Other methods (e.g., that described by Saheki
and Holzer, 1975) simply rely on colorimetric estimation (Biuret, Folio,
etc.) of trichloroacetic acid-soluble pep tides released by protease action. It
is thus apparent why it is difficult to tie known in vitro activities to observed
in vivo proteolytic events. Genetic attempts at answering these questions
have not, in general, been productive. A more useful approach might be
to first determine the precise subcellular location of the process pre-
sumed to be controlled by protease action and then to investigate which
protease(s) are active there. Subcellular fractionation coupled with
cytochemical techniques would seem to be appropriate. One can then
address the question "Does a given protease attack the protein or peptide
of interest in vitro and is its mode of catalysis consistent with events
observed in vivo?" Finally, one would need to demonstrate that the kinetics
of in vivo regulation of the protease activity correlates with the kinetics of
the original cellular process under investigation.
3.3. Nucleotide Metabolism

The biochemistry and genetics of nucleotide biosynthesis have been
comprehensively reviewed by Jones and Fink (1982). This section will
deal solely with those aspects of nucleotide metabolism that deserve
special mention because of physiological, biochemical, or genetic conse-
quences of naturally occurring features or mutations in nucleotide me-
tabolism and the way we handle yeast as a result of these. In addition,
some recent findings will be mentioned especially with respect to the
metabolism of the cyclic nucleotide cAMP.
3.3.1. Purine Nucleotides

De novo synthesis of the purine nucleotides AMP and GMP starts
with phosphoribosyl pyrophosphate. A sequence of 10 reactions which
involves the products of ADE4, ADE5, ADE8, ADE6, ADE7, ADE2,
ADEJ, and ADE13 genes, phosphoribosylaminoimidazolecarboxamide
formyltransferase, and IMP cyclohydrolase leads to the formation of
IMP where the AMP and GMP pathways branch Uones and Fink, 1982).
AMP and GMP can also be formed from supplied adenine and guanine
by the action of adenine phosphoribosyltransferase and guanine phos-
phoribosyltransferase. In addition, IMP can be formed from hypox-
anthine. adel and ade2 mutations are frequently used genetic markers
because of a strong red coloration upon adenine starvation which en-
ables convenient marker assignment in such strains. The pigment is
believed to derive from the intermediate phosphoribosylaminoimidazole
which accumulates in cells bearing adel and ade2 mutations.
3.3.2. Pyrimidine Nucleotides

De novo synthesis of pyrimidine nucleotides proceeds from gluta-
mine through six enzyme-catalyzed steps to UMP. After UMP the details
are very vague. Figure 1 outlines the pathways that might exist in S.
cerevisiae. The implications for the paucity of our knowledge in this area
are quite serious. For example, the compound hydroxyurea (HU) is
frequently used as an inhibitor of DNA synthesis. It is often stated that
HU is presumed to inhibit DNA replication by inhibiting the enzyme
ribonucleoside diphosphate reductase, thus preventing the synthesis of
(the DNA precursors) deoxyribonucleoside triphosphates, but its defi-
nite mode of action and any possible side effects are not known.
UMP can be derived from exogenous uracil via uracil permease and
uracil phosphoribosyl transferase or from uridine, in which case uptake
is by uridine permease whence the uridine may be metabolized via either
uridine kinase to UMP or uridine ribohydrolase, which results in uracil.
Cytosine can be taken up by cytosine permease with subsequent deamin-
ation to uracil, and cytidine can be taken up by cytidine permease with
cytidine deaminase catalyzing its conversion to uridine (Cooper, 1982b;
Jones and Fink, 1982).
Yeasts, in common with most other fungi, lack thymidine kinase
(Grivell and Jackson, 1968), so it is not possible to specifically label thy-
mine nucleotides or DNA in yeast by the use of radioactively labeled
thymine or thymidine. Consequently we have to employ indirect meth-
ods for this. Typical approaches for isotopic labeling of DNA inS. cere-
visiae include those of Williamson (1965), Hartwell (1967), and Hatzfeld
(1973), in which a less specific precursor such as adenine or uracil is used
to label total nucleic acid, followed by elimination of the radioactivity
due to RNA by solubilizing in alkali (e.g., IM NaOH); RNAase could also
be used. For measurements of RNA an estimate of total nucleic acid is
usually sufficient as RNA is roughly 98% of total cellular nucleic acid
(Hartwell, 1967).
Disturbance of the levels of thymine nucleotides can have profound
genetic effects on the cell. Such effects are achieved by inhibition of
72 J. R. DICKINSON
l
UJIP
Huol.adcle
UDP _ _..:di=p..,ho=ap,..llo;1&1=DU=e.___~) 1ITP ------~CTP
1 Ribollucl.adde
diphoaphate
reduotaae
1
dUDP
Nuoleoe1de R1bolluoleoe1de
d1phoapho- triphoaphate
k11laae reductaae
1
~·~------------~ c
:::phoapha taae
(DUfl)
=~:!::!!:uaael R1boDuoleoa1de
dlphoaphate
dCMP reductaae
deaaillaae <DCDl)
dDIIP ~------------- dCIIP ~-----.;:-dCDP
ThJ111dyl.ata
arntlletaae
CDC3l (TMPl)
1
dTJIP
TMP3
ThJIIidylate
1 ld.uae
CDCB
dTDP
1 lfUoleoaide
d1phoaphok1DUe
Figure 1. The possible pathways of pyrimidine nucleotide biosynthesis in S. cerevisiae.

Enzymes are named and genes assigned where they have been definitely identified.
dTMP synthesis using either a thymidylate synthetase mutant (tmpl),

folate antagonists to block the methylation of dUMP to dTMP, or 5-
fluoro dUMP. All three methods produce starvation for thymine nu-
cleotides. This causes gene conversion and crossing over, unequal sister
chromatid crossing over, and mating-type switching (Kunz et al., 1980,
1984, 1985; Barclay et at., 1982; Kunz and Haynes, 1982; Eckardt et·al.,
1983). The earlier isolation of tupl mutants (which incorporate dTMP)
had shown that while some alleles produce pleiotropic defects, there was
a general effect on mating efficiency. Strains that were MATa tupl had
reduced efficiency of mating with MATa strains but had higher than
normal mating efficiency with MATa strains such that they mated almost
equally well with a and a tester strains (Wickner, 1974).
3.3.3. Cyclic Nucleotides

Cyclic nucleotides have attracted a lot of attention recently. cAMP
has been implicated in the regulation of a number of processes in S.
cerevisiae, including the cell cycle, sporulation, mating, and catabolite
inactivation (discussed in Section 2.2.). cAMP is formed from ATP by
adenylate cyclase. It is degraded by the enzyme phosphodiesterase to
produce 5' AMP. cAMP acts via cAMP-dependent protein kinase, which
is composed of catalytic and cAMP-binding (regulatory) subunits. Bind-
ing of cAMP to the regulatory subunit causes dissociation of the sub-
units, thus rendering the catalytic subunit active.
The involvement of cAMP in mating has been shown by the fact that
the mating pheromone a-factor inhibits adenylate cyclase in vitro and
causes arrest at "start" and that exogenous cAMP can reduce a-factor-
induced G 1 arrest. Furthermore, ste5 mutants which do not mate have
an a-factor-insensitive adenylate cyclase (Liao and Thorner, 1980). Iso-
butylmethylxanthine inhibits phosphodiesterase in vitro and antagonizes
a-factor arrest (Liao and Thorner, 1981). Temperature-sensitive cyrl
mutants (defective in adenylate cyclase) cannot mate at the restrictive
temperature (Matsumoto et al., 1983a). Evidence for a cAMP regulatory
system controlling cell cycle progression comes from the fact that cyr1
mutants arrest as class II "start" mutants in the absence of cAMP (Mat-
sumoto et al., 1982). The cell division cycle mutation cdc35 which causes
cell cycle arrest at start is allelic to cyrl (Casperson et al., 1985). The
mutations cyrl and CYR3 (the latter results in an altered regulatory
subunit of cAMP-dependent protein kinase requiring elevated levels of
cAMP for dissociation), are suppressed by the mutation bcyl, which re-
sults in a cAMP-independent protein kinase due to deficiency of the
regulatory subunit (Matsumoto et al., 1982). The probably involvement
of cAMP in initiation of sporulation is shown by the fact that diploids
homozygous for cyrl or CYR3 can sporulate in rich media when wild-
type strains would not (Matsumoto et al., 1983b) and that glucose repres-
sion of sporulation can be alleviated by cAMP (Tsuboi et al., 1972;
Tsuboi and Yanagishima, 1973). Other elements of the cAMP regulatory
system that have been discovered include the following mutants; the
defects these mutations confer are given in parentheses: cyr2 (altered
catalytic subunit of cAMP-dependent protein kinase) (Uno et al., 1984);
lAC (increased adenylate cyclase) (Uno et al., 1983); pdel (reduced ac-
tivity of "high Km form" of cAMP phosphodiesterase) (Uno et al., 1983),
and ppdl (deficient in phosphoprotein phosphatase, which de-
phosphorylates phosphorylated proteins-i.e., works in the reverse di-
rection of protein kinase) (Matsumoto et al., 1985). In addition to these
mutants which were identified by classical mutational analyses, direct
cloning has indicated there are three different genes (TPKJ, TPK2, and
TPK3) encoding the catalytic subunits of cAMP-dependent protein
74 J. R. DICKINSON
kinase (Toda et al., 1987a) and a low Km ("high affinity") cAMP phos-
phodiesterase encoded by PDE2 (Sass et al., 1986).
Further interest has been generated by the discovery that yeast has
RAS genes. Two such genes have been identified (RASl and RAS2)
which have structural and functional homologies to mammalian ras on-
cogenes. Cells containing disruptions in both RAS genes do not prolif-
erate and have lowered cAMP. RAS2 val19, an allele with a missense
mutation analogous to one that activates the transforming potential of
mammalian ras genes, results in increased cAMP levels within the cell
and renders the cell constitutive for proliferation (Toda et al., 1985;
Kataoka et al., 1985; Broek et al., 1985). The bcyl mutation suppresses
rasl ras2 mutants but RAS2 val19 does not suppress cyrl (Toda et al.,
1985), thus the CYRl and BCYl gene products must be dependent on
the action of the RAS proteins. The mammalian adenylate cyclase is
membrane bound and has a hormone receptor which upon binding
hormone stimulates GTP binding of the G-protein subunit, rendering it
active and stimulating the catalytic subunit. Slow hydrolysis of the GTP
to GDP by the GTPase activity of the G protein returns it to the inactive
form. Yeast RAS and human ras genes having "oncogenic" alterations
result in a loss of GTPase activity (Toda et al., 1985 ). Other elements that
have to be incorporated into our understanding of this complex system
are the gene products of CDC25, which regulates adenylate cyclase possi-
bly by regulating guanine nucleotide bound to RAS proteins (Robinson et
al., 1987; Broek et al., 1987); RAMl, which is required for processing of
RAS proteins (Powers et al., 1986); RHOl and RH02 (Ras homologs) and
the various SRA (suppressor of ras) genes (Cannon et al., 1986). The
situation may not be as complicated as it first seems, as several of these
apparently different genes may be identical as in the case of SRA4
(= CYRl = CDC35), sral (= BCYl = REGl), and SRA3 = CYR2. Thus,
once again the biochemical details are incomplete; nevertheless this area
is currently under intense investigation in many laboratories throughout
the world. The information that is sought will enable an understanding
of the integrative network of controls involving mating, proliferation,
sporulation, and aspects of carbon metabolism.
3.4, Molecular Genetics of Nitrogen Metabolism

Table III lists those genes of S. cerevisiae involved in nitrogen catabolism
and anabolism that have been cloned.
4. TRANSPORT OF SUBSTRATES INTO THE CELL
Transport processes inS. cerevisiae have been reviewed by Cooper

( 1982b). There have been two ways of attempting to identify the transpor-
Table III. List of Saccharomyces Genes Involved in Nitrogen

Anabolism and Catabolism"
Gene designation Gene product Reference
ADEI Phosphoribosylaminoimidazole Crowley and Kaback (1984)

succinocarboxyamide syn-
thetase
ADE2 Phosphoribosylaminoimidazole Sasnauskas et al. ( 1982)
carboxylase
ADE4 Glutamine phosphoribosylpyro- Mantsala and Zalkin (1984)
phosphate amidotransferase
ARGJ Ornithine carbamoyltransferase Crabeel et al. (1981)
ARG4 Arginosuccinate lyase Hsiao and Carbon ( 1979)
ARGRI (ARG80)
[Specific regulation of arginine Dubois et al. ( 1987)
ARGRII (ARG81)
metabolism]
ARGRlll (ARG82) Dubois and Messenguy ( 1985)
AROIA 3-Enol pyruvoylshikimate-5-
phosphate synthase
AROIB Shikimate kinase
Larimer et al. (1983)
AROIC 3-Dehydroquinate synthase
AROID Shikimate dehydrogenase
AROIE 3-Dehydroquinate dehydratase
AROJ Deoxy-o-arabino-o- Teshiba et al. ( 1986)
heptulosonate phosphate syn-
thase
AR07 Chorismate mutase Ball et al. ( 1986)
CANI Arginine permease Broach et al. ( 1979)
CARl Arginase Jauniaux et al. (1982);
Sumrada and Cooper
(1982)
CAR2 Ornithine transaminase Jauniaux et al. (1984);
Sumrada and Cooper
(1985)
CDC8 Thymidylate kinase Sclafani and Fangman (1984)
CPAI Carbamoylphosphate synthetase Messenguy et al. (1983)
CPA2 Carbamoylphosphate synthetase Lusty and jeffrey (1982);
Messenguy et al. (1983)
GYRI Adeny1ate cyclase Masson et al. (1984); Casper-
son et al. (1985)
DAL2 Allantoicase
Yoo et al. (1985)
DALJ Ureidoglycolate hydrase
DAL5 [Component of allantoate trans- Rai et al. ( 1987)
port system]
DCDI dCMP deaminase Mcintosh and Haynes (1986)
DFRJ Dihydrofolate reductase Mcintosh et al. (1986)
DUR1,2 Urea amidolyase Genbauffe and Cooper (1986)
DUTI dUTP pyrophosphatase Mcintosh et al. (1986)
FUR4 (URA-P) Uracil permease Chevallier ( 1982)
GCD4 Negative transcriptional regula- Skvirsky et al. ( 1986)
tor ofGCN4
(continued)
76 J. R. DICKINSON
Table III. (Continued)

GCNJ (NDRJ)
GCN2 (AASJ)
[Involved in general control of
GCNJ (AAS2) Hinnebusch and Fink (1983b)
amino acid biosynthesis]
GCN4 (AASJ)
GCN5
H/SJ ATP phosphoribosyltransferase Hinnebusch and Fink ( 1983a)
HISJ Imidazoleglycerolphosphate de- Struhl et al. (1979)
hydratase
HIS4A Phosphoribosyl-AMP cyclo-
hydrolase
HIS4B Phosphoribosyl-ATP-pyrophos- Donahue et al. (1982)
phohydrolase
HIS4C Histidinol dehydrogenase
HIS5 Histidinol phosphate amino- Harashima et al. (1981); Sa-
transferase toshi et al. (1981)
/LVI Threonine deaminase Petersen et al. (1983);
Holmberg (1984)
JLV2 Acetohydroxyacid synthase Holmberg (1984); Polaina
(1984)
ILVJ Dihydroxyacid dehydratase Holmberg (1984); Polaina
(1984)
ILV4 Acetohydroxyacid isomero- Holmberg (1984)
reductase
ILV5 Acetohydroxyacid isomero- Polaina (1984)
reductase
KEX2 Endopeptidase processes David et al. (1984)
prepro-a-factor
LEUJ a-Isopropylmalate dehydratase Hsu et al. (1982); Hsu and
Schimmel (1984)
LEU2 13-Isopropylmalate dehydratase Ratzkin and Carbon (1977)
LEUJ [Regulator of LEUJ and LEU2 Brisco et al. (1987)
expression)
LEU4 a-Isopropylmalate synthase Chang et al. ( 1984)
LEU5 [Nonessential regulator of leu- Drain and Schimmel (1986)
cine biosynthetic pathway]
LYS2 2-Aminoadipate reductase Eibel and Philippsen (1983)
MET2 Homoserine acetyltransferase Langin et al. (1986)
METJ ATP: sulfate adenylyl trans- Cherest et al. ( 1985)
ferase
MET6 Homocysteine methyltrans- Csaikl and Csaikl (1986)
ferase
MET25 Homocysteine synthase Sangsoda et al. (1985)
PDE2 "High affinity" cAMP phospho- Sass et al. ( 1986)
diesterase
PPRJ [Regulator of expression of Losson and Lacroute ( 1982)
URAl and URAJ products]
PPR2 Regulator of URA4 gene ex- Hubert et al. (1983)
pression
PRAJ (PEP4) Vacuolar aspartyl protease Woolford et al. (1986); Am-
merer et al. ( 1986)
Table III. (Continued)
PUT2 Pyrroline-5-carboxylate Brandriss (1983)

dehydrogenase
RNR2 Ribonucleotide reductase (small Elledge and Davis (1987)
subunit)
SPEJ Ornithine decarboxylase Fonzi and Sypherd (1985)
SPT2 (SPM2) [Negative regulator of Ty- Roeder et al. ( 1985)
controlled gene expression)
TMPJ Thymidylate synthetase Taylor et al. (1982)
TMP2 Dihydrofolate reductase Nath and Baptist (1984)
TRPJ Phosphoribosylanthranilate iso- Struhl et al. ( 1979)
merase
TRP2 Anthranilate synthase
TRPJ Anthranilate synthase: (indole- Aebi et al. ( 1982)
3-glycerol phosphate synthase
TRP4 Anthranilate phosphoribosyl-
Niederberger et al. (1984)
transferase
TRP5 Tryptophan synthase Zalkin and Yanofsky (1982)
URAl Dihidroorotate dehydrogenase Guerry-Kopecko and Wickner
(1980)
URA2B Aspartate carbamoyltransferase
Souciet et al. ( 1982)
URA2C (CPUJ) Carbamoylphosphate synthetase
URA3 Orotidine-5' -phosphate decar- Bach et al. (1979)
boxylase
URA6 Uridine monophosphokinase Liljelund and Lacroute (1986)
•Former or alternative gene designations are given in parentheses.
ters and their specificity. Biochemical studies center on identification of

the kinetics of substrate uptake. Usually Michaelis-Menten kinetics are
assumed and the data obtained are fitted to a model employing equations
of this type. Competitive inhibition by substrates is adduced to have
demonstrated that the two substrates are transported by the same carrier.
Genetic methods for probing transport use mutants defective in uptake
of one substrate to test whether uptake of other substrates is also affected.
Neither approach is satisfactory alone. Both biochemical and genetic
studies are required with measurements made under appropriate physio-
logical conditions.
The dual biochemical and genetic approach is elegantly demon-
strated in a series of experiments by Bisson and Fraenkel (1983a,b, 1984),
who used both wild-type strains and strains with one or more of the
following mutations-hxkJ (lacking hexokinase A or PI), hxk2 (lacking
hexokinase B or P-II), and glkl (lacking glucokinase)-to study glucose
and fructose uptake. They used 14 C-labeled glucose, fructose, mannitol,
and 6-deoxy-o-glucose (an analog of glucose that is taken up by the cell
but not metabolized). After incubation of cells and labeled sugar, uptake
78 J. R. DICKINSON
was terminated by addition of a large excess of cold water followed by

rapid filtration to collect the cells and further washing with cold water.
Radioactivity due to hexose uptake was measured by liquid scintillation
counting. These workers demonstrated that there are both low- and high-
affinity mechanisms for glucose uptake. The high-affinity uptake system
was low when cells were grown in high concentrations (100 mM) of
glucose and high when cells were in low concentrations (0.5 mM) of
glucose, in stationary phase of growth on glucose, and in other carbon
sources such as galactose and lactate plus glycerol or ethanol. The activity
of the high-affinity transport system is dependent on the activity of the
three glucose phosphorylating enzymes. Hence, triple kinase mutants
(hxkl hxk2 glkl) lacking all three kinase activities do not show high-affinity
uptake of either glucose or fructose (Bisson and Fraenkel, 1983a), where-
as the hxkl hxk2 double mutant lacks only high-affinity fructose uptake
but retains high-affinity glucose uptake.
A number of important questions concerning the mechanism of
hexose uptake remain unanswered (Bisson and Fraenkel, 1984). It is not
apparent from this work whether the high- and low-affinity uptake sys-
tems are different systems or part of the same system. Nor is it clear
whether the uptake of glucose is an energy-linked process or involves
facilitated diffusion. Experiments using analogs of glucose such as 6-
deoxyglucose indicate that both uptake systems involve facilitated diffu-
sion (Kotyk, 1967; Cirillo, 1968; Heredia et al., 1968; Bisson and Fraen-
kel, 1983b, 1984). However, the problem with studying "native" glucose
arises from the fact that it is very rapidly phosphorylated either while
entering the cell or immediately thereafter. Bisson and Fraenkel have
suggested that the hexokinases and/or glucokinase may be involved in
derepression of a component of the transporter system, possibly the
membrane carrier itself (Bisson and Fraenkel, 1984).
One possible way to resolve this problem might be to use 13C nu-
clear magnetic resonance (NMR). 13C NMR spectroscopy has already
been usefully employed to study the uptake and metabolism of glucose
by S. cerevisiae in vivo (den Hollander et al., 1979). One of the advantages
of this approach is that it is possible to discriminate between anomeric
forms of sugars. Figure 2 shows a typical experiment, and in this case
preferential use of the P-anomer of glucose is shown. Using NMR to
study glucose transport mutants should elucidate this long-standing
mystery. Until recently glucose transport mutants had not been ob-
tained, however it has now been shown that the SNF3 gene is required
for high-affinity glucose transport (Bisson et al., 1987). Notice that the
glucose transport mutant was originally isolated as a snf (sucrose non-fer-
menting) mutant; i.e. the phenotype is more subtle than one might have
guessed. This general point is considered again later (Section 5.1 ). The
use of NMR has potential in many other areas of yeast central metabo-
lism, as outlined in Section 5.2.
~
.
E
T
Ia
-
30
l
1
1
1 -20 limo
( mlnutool
I
J I I
l. J -10
I i
100 80 80 40 20 ppm
Figure 2. Time course of utilization of [1-ISC] glucose by Saccharomyces cerevisiae. In vivo

time-elapsed proton decoupled NMR spectra are shown. Each spectrum is the result of
1000 pulses. o:, C-1 of o:-o-glucose; ~. C-1 of ~-o-glucose; T, C-1, 1' of trehalose; E, C-1 of
ethanol; G, C-1 of glycerol.
If we wish to examine carefully the rate of synthesis of macromole-

cules (e.g., protein or nucleic acid), the only reliable way is to use a pulse
labeling method with an appropriate precursor (in these cases an amino
acid and, say, uridine, respectively). Thus, we rely on transport (uptake)
of the labeled precursor even though we are not directly concerned with
the transport process itself. Such "uptake and incorporation" studies
make certain assumptions which need careful examination in each spe-
cific case. The basic assumption is that the amount of precursor incorpo-
rated (into protein or nucleic acid) during the short period of the pulse is
a measure of the rate of synthesis of the macromolecule. The precursor
chosen should be present in low concentrations in the cell (said to have a
"small pool"). Pool sizes vary in different strains and in different media.
The presence of multiple pools should be considered. It can often be
shown that the cell has a "metabolic pool" which is supplied by endoge-
nous biosynthesis and uptake from the exogenous label and is used for
synthesis of the macromolecule. In addition, there may be two kinds of
sequestered pool: a "storage pool" due to uptake of exogenous precur-
80 J. R. DICKINSON
sor and possibly also supplied from the metabolic pool and an "ex-
changeable pool" involved in exchange of exogenous and endogenous
precursor. With a small metabolic pool incorporation will be a linear
function with time. If the pool is large the rate of incorporation will
increase until the pool reaches a constant specific activity. The termi-
nology employed here is that of Creanor and Mitchison ( 1982), whose
precise work demonstrates the considerations just outlined.
The importance of precursor pool sizes cannot be overemphasized,
especially in relation to the cell cycle. The amount of radiolabel incorpo-
rated may not be a true measure of macromolecular biosynthesis through
the cell cycle for a variety of reasons. For example, the size of the
metabolic pool may vary through the cell cycle. The contributions from
exogenous and endogenous precursor may vary through the cell cycle
(i.e., transport of the precursor may be cell cycle modulated) and the
various storage pools may vary in their contributions to biosynthesis
through the cell cycle. It thus becomes important to determine whether
changes in specific activity of the metabolic pool occur. It is virtually
impossible in most cases to extract the metabolic pool on its own. The use
of a mutant auxotrophic for the precursor to be used is an obvious way of
abolishing its endogenous synthesis, but this has other inherent problems
discussed previously (see Section 3.1). It is possible, however, to arrange
conditions in which a storage pool is stable and hence its effect on
incorporation into macromolecules can be ignored. That such conditions
exist (arrived at by trial and error) can be demonstrated by a pulse-chase
type of experiment. Uptake, measured as total cellular radioactivity,
should cease immediately following the chase or transfer to fresh medi-
um. Incorporation into macromolecules from the metabolic pool may
continue increasing for a given period of time before it too becomes
constant. The difference between total radioactivity and incorporated
radioactivity is a measure of the size of the storage pool(s).
5. REGULATION OF METABOLISM
5.1. Genetic Studies

From the preceding material in this chapter the reader will already
have recognized this author's opinion as to the power of a combined
biochemical-genetic approach to questions of metabolism and its reg-
ulation. The genetic studies of recent years involving the isolation of
mutants, examination of their physiology and biochemistry, the cloning
of genes, and subsequent analyses of gene expression have provided
knowledge on regulation at the genetic level and have also furnished
important information on pathways and their control at the biochemical
level. For example, in the pathway of arginine biosynthesis gene-en-

zyme relationships have been established for the complete pathway, reg-
ulatory mutations have been investigated, and the principal properties
and intracellular location of all the enzymes are known Uones and Fink,
1982; Henry et al., 1984).
Genetic approaches have not been without their shortcomings, how-
ever. For example, the usefulness of having mutants for each stage of
glycolysis has long been recognised. Prior to the isolation of aldolase
(jbal) mutants (Lobo, 1984), it had been suggested that aldolase mutants
might have unexpected properties, and evidence of isozymes of aldolase
was cited in explanation for the lack of such mutants (Fraenkel, 1982).
However, the single-gene mutants were apparently obtained readily and
were typical of the majority of glycolysis mutants in that they were un-
able to grow on glucose and showed growth inhibition by glucose when
growing on ethanol. In other words, it is not always clear why attempts to
obtain particular mutants fail even when one knows beforehand the
biochemical lesion that is sought and one might be able to predict some
phenotypes on the basis of this.
The biochemical-genetic approach for analyzing the control over
glycolysis by phosphofructokinase has produced some startling results.
It now seems clear that S. cerevisiae possesses two phosphofructokinase
isoenzymes controlled by at least five genes. The gene PFKJ codes for
the catalytic subunit of the familiar soluble allosteric enzyme. PFK2 is a
common element coding for both the regulatory subunit of the soluble
enzyme and also a particulate phosphofructokinase. Mutants lacking the
particulate enzyme define at least three other genes PFK3, PFK4, and
PFK5 (Parmer et al., 1984). Phosphofructokinase mutants have been
isolated in a number of different laboratories (Clifton et al., 1978; Ciriacy
and Breitenbach, 1979; Navon et al., 1979; Clifton and Fraenkel, 1982;
Lobo and Maitra, 1982, 1983; Breitenbach-Schmitt et al., 1984a; Parmar
et al., 1984). This has led to confusion on the behalf of some authors as to
whether the mutation designated pfkl by Fraenkel's group should be
regarded as the same as the pfkl or the pfk2 mutation of Maitra's group
(Breitenbach-Schmitt et al., 1984a). Such distinctions need not concern
us here; the confusion is more apparent than real. Maitra's group have
investigated this area extensively and provided convincing evidence dis-
pelling any confusion (Nadkarni et al., 1984).
The discovery of a particulate phosphofructokinase is quite unusual
because all the other glycolytic enzymes of yeast are soluble. But an even
more unexpected finding is that of Breitenbach-Schmitt et al. ( 1984b),
who have presented evidence for the existence of a "bypass" pathway
parallel to the phosphofructokinase-aldolase reaction sequence. These
workers examined the metabolism of pfk mutants growing on glucose by
analyzing the fate of 14 C-labeled glucose. The "classical" view of glycoly-
82 J. R. DICKINSON
sis states that only C-3 and C-4 of glucose should contribute to carbon
dioxide production. However, by using U- 14C- glucose and 3,4-14C-
glucose it was shown that in wild-type cells only two-thirds of the carbon
atoms released as C02 derive from the C-3 and C-4 positions of glucose.
The remainder derive from other C atoms in glucose. In pfkl or pfk2
mutants all six positions of glucose contributed equally to C02 produc-
tion (Breitenbach-Schmitt et al., 1984a). These findings suggest that sol-
uble phosphofructokinase catalyzes only part of the total flow of carbon
in glycolysis and that the enzyme is dispensable in glycolysis. Mutants
were isolated in three different genes (BYPJ, BYP2, BYP3) which block
glycolysis in single pfkl and pfk2 mutants (Breitenbach-Schmitt et al.,
1984b). In the bypl mutant C02 was derived almost exclusively from the
C-3 and C-4 positions of glucose (i.e., according to the "classical" pattern
of glycolysis). From accompanying enzyme assays and identification of
accumulating metabolites in pfk and byp mutants it appears that the
nonoxidative portion of the pentose phosphate cycle has a significant
role in hexose fermentation. This is in contrast to use of the oxidative
portion of the cycle via glucose-6-P dehydrogenase, which is known to
make a very small contribution (around 2.5%) to carbon flux from glu-
cose (Lagunas and Gancedo, 1973). The implications of this work are
clearly very important. No doubt many of us think we know all there is to
be known about a pathway as fundamental as glycolysis in yeast. It is
interesting to speculate whether the BYPJ, BYP2, and BYP3 genes iden-
tified by the Darmstadt group are the same as the PFK3, PFK4, and
PFK5 genes described by workers in Bombay.
Genetic methods have been slow in yielding biochemical information
on cell cycle control in yeast. Approximately 80 genes are known to be
involved in cell cycle progression, and much has been learned in physio-
logical and genetic terms especially of the temporal order of gene ex-
pression and dependency relations between the largely unidentified gene
products involved (Pringle and Hartwell, 1981 ). A great number of cell
cycle genes have been both cloned and sequenced. However, the bio-
chemical defect in specific cell cycle mutants is known in only a very few
cases. Those mutations where the biochemical defect is known are algl
(protein N-glycosylation) (Kiehl et al., 1984); bcyl * (regulatory subunit of
cAMP-dependent protein kinase) (Matsumoto et al., 1983a; Toda et al.,
1987b); cdc7* (Patterson et al., 1986); cdc8* (thymidylate kinase) (Jong et
al., 1984; Sclafani and Fangman, 1984); cdc9* (DNA ligase) (Johnston and
Nasmyth, 1978; Barker et al., 1985); cdc19 (pykl)* (pyruvate kinase)
(Kawasaki, 1979; Kawasaki and Fraenkel, 1982; Burke et al., 1983); cdc21
(tmpl)* (thymidylate synthetase) (Game, 1976; Taylor et al., 1982, 1987);
cdc28* (protein kinase) (Reed etal., 1985); cdc30 (pgi2)* (phosphoglucose
isomerase I) (Dickinson and Williams, 1987, unpublished results); cdc35
(cyrl)* (adenylate cyclase) (Masson et al., 1984; Casperson et al., 1985);
cyr2 (catalytic subunit of cAMP-dependent protein kinase) (Uno et al.,

1984); CYR3 (affects the regulatory subunit of cAMP-dependent protein
kinase) (Uno et al., 1982); pdel * pde2* ("low affinity" and "high affinity"
cAMP phospodiesterases, respectively) (Sass et al., 1986; Toda et al.,
1987a); ppdl (glcl) (phosphoprotein phosphatase) (Uno et al., 1983); and
top2* (DNA topoisomerase II) (DiNardo et al., 1984). (The asterisk de-
notes where the mutation has been shown to affect the structural gene.)
Clearly, biochemical understanding of regulation and control of the cell
cycle lags behind genetic and physiological descriptions. This is especially
true in the case of mutants originally isolated as cell division cycle (cdc)
mutants. This gap in knowledge has become a pressing problem in this
area and hinders attempts to understand the integration of metabolism
involved in the cell's control of growth and cell division.
Attempts at identification of CDC gene products have proceeded in
a somewhat piecemeal fashion. Those cdc mutants affected in DNA rep-
lication are relatively easy to work out because one knows which overall
process to inspect. By an ordered sequence of experiments in which one
examines whether initiation of DNA synthesis occurs, if ligation is ef-
fected, or if DNA replication is terminated, it is possible to pinpoint the
defect. Similarly, we might guess that mutants affected in mitosis should
yield those with defective tubulins, as has been the case in other eu-
karyotic systems. However, in the majority of cell cycle mutants there is
no such convenient "pointer" as DNA replication or mitosis. The essen-
tial problem remains: How do we start to guess which biochemical func-
tion (from among the thousands of possibilities within the cell) is af-
fected in a particular mutant?
It has now become a fairly standard procedure to clone the gene of
interest and then to sequence it. The base sequence can then be com-
pared with that of other cloned genes to inspect for homologies and can
also be used to predict the primary structure (amino acid sequence) of
the protein encoded. In some cases this exercise allows identification of
the gene product on the basis of convincing homology with known pro-
teins that have been sequenced previously. One should be prepared to
expect that this approach may be rather limited because an insufficient
number of proteins have been sequenced. The essential problem re-
mains: we need a technique capable of "seeing" the whole metabolism in
the cell.
5.2. Physiological and Biochemical Studies

Generally physiological and biochemical methods do not make in-
teresting reading and so this section will concentrate on selected newer
and novel approaches.
When, late in 1981, I turned my attention to the metabolism of
84 J. R. DICKINSON
sporulation, it seemed that despite previous studies on many separate and

diverse aspects of the phenomenon, there was no clear and integrated
view of the metabolic changes associated with the initiation of sporulation
and the subsequent events of the process itself. Indeed, many of the
previous biochemical studies had presented conflicting views. In vivo 13 C
NMR was presented as an holistic technique capable of providing the
overview of metabolism that was required (Dickinson et al., 1983). By
reference to this particular problem (the metabolism of sporulation), we
can see the utility of the NMR method to other aspects of central inter-
mediary metabolism. It is, of course, nondestructive and noninvasive.
To those unfamiliar with NMR it is a technique that has the reputa-
tion of being far too insensitive to be any use in in vivo analysis of
metabolism. It is true, for example, that the low ( 1.1%) relative natural
abundance of 13C (the nucleus of carbon actually observed in the NMR
experiment) means that we need to employ artificially enriched sub-
strates for most analyses of carbon metabolism. Nevertheless, there are
cellular components we can observe due to natural abundance of 13C.
These are obviously those compounds present in large amounts within
the cell, one example being the storage carbohydrate trehalose, which is
present in quite high amounts under most conditions. Another popular
misconception about NMR for this type of study is that one needs a
superpowerful spectrometer. Nothing could be further from the truth.
A considerable amount of work has been done using very powerful
machines, sometimes with so-called "wide-bore" magnets. (A "wide-
bore" magnet allows the use of a wide sample tube, thus enabling a
greater quantity of cells to be analyzed than would otherwise be pos-
sible.) Perfectly adequate spectra can be obtained from yeast in normal-
sized (10 mm) tubes in machines of modest field strength.
Sporulation in the presence of [2- 13 C] acetate was studied by 13C
NMR spectroscopy. Immediately after transfer to sporulation medium
the major metabolite produced was glutamate. From the labeling pattern
observed it was concluded that both the tricarboxylic acid cycle and the
glyoxylate cycle were operating. At about 4 hr net glutamate synthesis
ceased and trehalose synthesis (gluconeogenesis) was initiated via glyoxy-
late cycle-derived intermediates. Subsequently, beginning at 4 hr the
synthesis of large amounts of fatty acid began, concomitant with onset of
operation of the pentose phosphate pathway. The involvement of the
pentose phosphate pathway in sporulation was previously unsuspected
and was discovered on the basis of the distribution of 13C-labeled carbon
atoms in trehalose. We now know that the NADPH provided by the
pentose phosphate reaction sequence is used to fuel the fatty acid bio-
synthesis (Dickinson and Hewlins, 1988).
These NMR studies provided some important new clues. For exam-
ple, when cells enter sporulation there is a shift in the operation of the
TCA cycle leading to glutamate accumulation. An increase of glutamate

during sporulation had been reported previously (Rousseau, 1972), but
it was believed that amino acids were derived from proteolysis (Tingle et
al., 1973). Clearly, the 13 C NMR studies gave a fresh interpretation of
this, namely, that initiation of sporulation involves a reduction in the
activity of the 2-oxoglutarate dehydrogenase complex. This has been
confirmed (Dickinson et al., 1985). A further consequence of this would
be to reduce GTP synthesis via succinyl CoA synthetase because of the
reduced carbon flux through this portion of the TCA cycle after the
down-regulation of the 2-oxoglutarate dehydrogenase. Freese et al.,
(1985) have confirmed that GTP levels always fall upon initiation of
sporulation. Furthermore, spd mutants that are derepressed for sporula-
tion (i.e., sporulate in rich media) and have greatly reduced sensitivity to
NH.t repression of sporulation (Dawes, 1975; Vezinhet et al., 1979) have
reduced levels of 2-oxoglutarate dehydrogenase activity.
The use of 13 C-enriched substrates allows a view of the whole of
carbon metabolism that is proceeding. By analyzing the labeling pattern
in a particular intermediate it is possible to work out the relative contri-
bution of each pathway involving that intermediate. This is something
that cannot usually be done using radioactive labeling without recourse
to very sophisticated chemistry as well as tedious separation techniques.
We can get information on phosphorylated intermediates, energy me-
tabolism, and phospholipids by 31 P NMR. (31 P has 100% natural abun-
dance, so artificially enriched substrates are not required here.) The use
of 31 P NMR is well demonstrated in a study of wild-type and glycolysis
mutants of S. cerevisiae (Navon et al., 1979). The mutant strains were
shown to accumulate characteristic sugar phosphates upon incubation
with glucose. An additional piece of information that can be gained from
31 P NMR spectra is the intracellular pH of the environment of indi-
vidual metabolites (Moon and Richards, 1973). Intracellular pH is con-
veniently determined on the basis of the chemical shifts of inorganic
phosphate and/or other phosphosphorylated intermediates. It is even
possible to observe the different pH values of different organelles. Ob-
viously this type of information can be particularly valuable when at-
tempting to correlate enzyme activities (measured in vitro) with in vivo
metabolic fluxes. Analogous studies using 15 N labeling can be devised to
reveal aspects of cellular nitrogen metabolism.
Another new technique is the application of membrane inlet mass
spectrometry to measurements of dissolved gas concentrations. This al-
lows a considerable improvement over former methods which have usu-
ally relied on gas phase measurements. It is obviously preferable to be
able to make measurements in the actual cellular environment. The
principles of operation and design of such equipment have been de-
scribed (Lloyd and Scott, 1983). The technique has been used to observe
86 J. R. DICKINSON
the Pasteur effect in yeast (Lloyd et al., 1983a) (as shown in Fig. 3), and to
study the effects of NH,t on glycolysis and respiration (Lloyd et al.,
1983b). The instrumentation has much to commend it, including the
facility to make continuous measurements of numerous gases with a
single probe and the possibility of monitoring many fermentations si-
multaneously. In Cardiff we have built a probe suitable for use in an
NMR tube giving the possibility of simultaneous monitoring of dissolved
gas metabolism along with in vivo NMR analysis of metabolic pathways.
An additional feature of the instrumentation is its reliability over long
periods, which, with its proven use in batch cultures and chemostats,
would seem to be purpose-built for industrial use in computer-con-
trolled fermentation processes.
An interesting report appeared describing the construction of a
complete yeast glycolysis system in vitro (Welch and Scopes, 1985). Indi-
vidually purified enzymes from S. cerevisiae were combined in propor-
tions and at concentrations comparable to those existing in vivo. This
cell-free glycolysis system produced ethanol and carbon dioxide from
glucose. The authors point out that this demonstrates the there is noth-
ing in principle that prevents the use of purified enzymes for other
pathwaylike transformations or possibly the use of semipurified prepa-
rations that allow production of more valuable products. There also
seems to be considerable potential here for analyzing control of a meta-
bolic pathway (at the level of enzyme activity only).
300 N2
S.uvarum I
S06SPa 02
10130Pao2 : •,
'
lcccp ~M S06SPaO
6·613-2
I I
I 60
=
r 200
S06SPG02
I Nz
I
If :
c
~
Oz \11mM
'
glucose i-:
...cc
0
u
\I
~100 COz __:1·········-i
'
,-·--- _____ j
'
i i
'
20 ~
N
0
'\
I
:,
/
I
0 ._ ___ -- ----~
t ______......... 0
0 10 20 30 40 so 60
Time (mini
Figure 3. The Pasteur effect in Saccharomyces uvarum observed by mass spectrometry. The
cells were continuously stirred in the reaction vessel. At the marker lines the composition
of the gas stream entering the open system was altered, or additions were made as shown.
••••••• Dissolved 02 concentration; --dissolved C02 concentration. (Reprinted by per·
mission from Biochemical journal, 212:749-754. Copyright Cl 1983 the Biochemical Soci-
ety, London.)
5.3. Further Aspects of Regulation

Thus far we have examined metabolism as a self-contained entity,
but three additional types of control must be considered: mating-type
effects, the cell cycle, and sporulation.
5.3.1. Mating-Type Effects

Mating-type effect (ala suppression) was observed in overproducing
mutants of S. cerevisiae affected in arginine catabolism and urea utiliza-
tion. The cargA + Oh, cargB + Oh, and durOh mutations resulted in mas-
sive overproduction of arginase, ornithine transminase, and urea carbox-
ylase-allophanate hydrolase (respectively) in haploid strains of either
mating type, and in diploids homozygous at the mating-type locus. How-
.
ever, matmg-type h eterozygos1ty
. ('I.e., m
. MATa
MATo. d'1pl01'd s) caused sup-
pression of the otherwise constitutive overproduction (Dubois et al., 197 8;
Lemoine et al., 1978; Deschamps and Wiame, 1979). A less dramatic
overproduction of iso-2-cytochrome c involving a similar mating-type-
regulated mutation, CYC7-H2, has been demonstrated by Rothstein and
Sherman ( 1980). In the case of the CYC7 -H2 mutation the transposable
element Ty 1 was found to be inserted in the 5' region of the gene coding
for iso-2-cytochrome c (Errede et al., 1980a,b). A Ty element was also
found to be inserted in the cargA region in the cargA + Qh mutant
Uauniaux et al., 1981 ). Taguchi et al. (1984) have reported on Ty-mediated
overproduction of the glucose-repressible alcohol dehydrogenase iso-
zyme ADHII. In this case the mating-type effect was greatest in the
presence of glucose, a repressing carbon source. Thus, Ty-mediated gene
expression is regulated by mating-type heterozygosity and carbon cata-
bolite control.
5.3.2. The Cell Cycle

It is not the purpose of this section to consider the cell cycle per se, but
we must take account of it. Any observations made on a population of
asynchronous cells will necessarily be time-averaged. This is not merely
an academic abstraction. As each cell progresses through the processes
involved in growth, DNA replication, mitosis, and cell division, many
aspects of temporal regulation of metabolism are invoked. Whether this
temporal regulation reflects the circadian clock or a cellular clock, such as
size control inS. cerevisiae (Sudbery et al., 1980), is unimportant in this
context. It now seems clear that the overall pattern of protein synthesis
throughout the cell cycle is exponential and only 2% of proteins show
periodic or modulated patterns of synthesis, and these are not among the
88 J. R. DICKINSON
abundant proteins (Lorincz et al., 1982). It is presumed these proteins

include regulators. Furthermore, in most cases where enzyme activity has
been analyzed carefully, it follows a linear pattern with a rate change once
per cycle (Creanor et al., 1983). However, exceptions are known, and
these appear to involve enzymes and proteins that are temporally and
functionally linked to DNA replication. These include enzymes produc-
ing DNA precursors and the histones (Storms et al., 1984; Moll and
Wintersberger, 1976). Thus if we are in the business of observing, prepar-
ing, or attempting to increase the yield of proteins that are cell cycle
modulated (or their metabolic products), then we may choose to ignore
the cell cycle at our cost.
If it became essential to make synchronous cultures in production-
scale quantities, it should be possible to obtain reasonable synchrony by
means of an appropriate fed-batch culture rendering it analogous to
laboratory-scale "starvation-refeeding" regimes.
5.3.3. Sporulation
The switch from vegetative growth and mitosis to the alternative
development pathway of meiosis and sporulation that can occur in
diploid strains of S. cerevisiae is under complex regulation (Esposito and
Klapholz, 1981). Cells must be heterozygous at the mating-type locus.
The switch can take place only during the G 1 phase of the cell cycle and
is also governed by nutritional conditions. Induction of sporulation re-
quires a shiftdown from a rich growth medium to one containing a
poorly utilized carbon source such as acetate, usually in the absence of
nitrogen. NH.t in particular is a strong repressor of sporulation. Thus
the process is simultaneously subject to mating-type control, cell cycle
control, and carbon- and nitrogen-source regulation. Hence, in a consid-
eration of sporulation we can start to appreciate the significance to the
cell of the array of controls that can be exercised over metabolic pro-
cesses in terms of its life cycle.
6. CONCLUDING REMARKS
In this chapter I have outlined some approaches that have provided

us with our current knowledge of metabolism and biosynthesis. Metabo-
lism inS. cerevisiae is regulated at many different levels, with interactions
occurring between levels. There are a number of conspicuous gaps in
our knowledge, including the nature of intermediates in pathways as
well as many more details concerning regulation. This biochemical infor-
mation is essential, for without it exploitation of the organism must
remain empirical rather than logical. For example, the use of a cloned
gene in construction of overproducing strains or for other aspects of

process optimization will be seriously hindered where the biochemical
identity of a gene product remains unknown. Without a complete bio-
chemical description of the pathway involved, including regulation, the
likely result of such an exercise would be the shifting of a "bottleneck"
from one part of a pathway to another. Thus biochemistry becomes an
even more important foundation to our understanding of the many
features of this fascinating organism.
AcKNOWLEDGMENT. I am grateful to Professor P. K. Maitra for helpful

comments.
REFERENCES
Aebi, M., Niederberger, P., and Hutter, R., 1982, Isolation of the TRP2 and the TRPJ
genes of Saccharomyces cerevisiae by functional complementation in yeast, Curr. Genet.
5:39-46.
Aguilera, A., and Zimmerman, F. K., 1986, Isolation and molecular analysis of the phos-
phoglucose isomerase structural gene of Saccharomyces cerevisiae, Mol. Gen. Genet.
202:83-89.
Ammerer, G., Hunter, C. P., Rothman, J. H., Saari, G. C., Valls, L. A., and Stevens, T. H.,
1986, PEP4 gene of Saccharomyces cerevisiae encodes proteinase A, a vacuolar enzyme
required for processing of vacuolar precursors, Mol. Cell. Bioi. 6:2490-2499.
Bach, M-L., Lacroute, F., and Botstein, D., 1979, Evidence for transcriptional regulation of
OMP-decarboxylase in yeast by hybridization of mRNA to the yeast structural gene
cloned in E. coli., Proc. Natl. Acad. Sci. USA. 76:386-390.
Ball, S. G., Wickner, R. B., Cottarel, G., Schaus, M., and Tirtiaux, C., 1986, Molecular
cloning and characterisation of AR07-0SM2 a single yeast gene necessary for choris-
mate mutase activity and growth in hypertonic medium, Mol. Gen. Genet. 205:326-
330.
Barclay, B.J., Kunz, B. A., Little,j. G., and Haynes, R. H., 1982, Genetic and biochemical
consequences of thymidylate stress, Can.]. Biochem. 60:172-194.
Barker, D. G., White,J. H. M., and johnston, L. H., 1985, The nucleotide sequence of the
DNA ligase gene (CDC9) from Saccharomyces cerevisiae: A gene which is cell cycle
regulated and induced in response to DNA damage, Nucl. Acid Res. 15:8323-8337.
Bisson, L. F., and Fraenkel, D. G., 1983a, Involvement of kinases in glucose and fructose
uptake by Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. USA. 80:1730-1734.
Bisson, L. F., and Fraenkel, D. G., 1983b, Transport of 6-deoxyglucose in Saccharomyces
cerevisiae,]. Bacteriol. 155:995-1000.
Bisson, L. F., and Fraenkel, D. G., 1984, Expression of kinase-dependent glucose uptake in
Saccharomyces cerevisiae,]. Bacteriol. 159:1013-1017.
Bisson, L. F., Neigeborn, L., Carlson, M., and Fraenkel, D. G., 1987, The SNFJ gene is
required for high affinity glucose transport in Saccharomyces cerevisiae, ]. Bacteriol.
169:1656-1662.
Brandriss, M. C., 1983, Proline utilisation in Saccharomyces cerevisiae: Analysis of the cloned
PUT2 gene, Mol. Cell. Biol. 3:1846-1856.
Breitenbach-Schmitt, I., Heinisch,j., Schmitt, H. D., and Zimmerman, F. K., 1984a, Yeast
mutants without phosphofructokinase activity can still perform glycolysis and alcohol
fermentation, Mol. Gen. Genet. 195:530-535.
90 J. R. DICKINSON
Breitenbach-Schmitt, I., Schmitt, H. D., Heinisch, j., and Zimmerman, F. K., 1984b, Ge-
netic and physiological evidence for the existence of a second glycolytic pathway in
yeast parallel to the phosphofructokinase-aldolase reaction sequence, Mol. Gen. Genet.
195:536-540.
Brisco, P.R. G., Cunningham, T. S., and Kohlhaw, G. B., 1987, Cloning, disruption, and
chromosomal mapping of yeast LEUJ, a putative regulatory gene, Genetics 115:91-99.
Broach, J. R., Strathern, J. N., and Hicks, J. B., 1979, Transformation of yeast: develop-
ment of a hybrid cloning vector and isolation of the CANJ gene, Gene 8:121-133.
Broek, D., Samiy, N., Fasano, 0., Fujiyama, A., Tamanoi, F., Northup, j., and Wigler, M.,
1985, Differential activation of yeast adenylate cyclase by wild-type and mutant RAS
proteins, Cell41:763-769.
Broek, D., Toda, T., Michaeli, T., Levin, L., Birchmeier, C., Zoller, M., Powers, S., and
Wigler, M., 1987, The S. cerevisiae CDC25 gene product regulates the RAS/adenylate
cyclase pathway, Cell48:789-799.
Burke, R. L., Tekamp-Olson, P., and Najarian, R., 1983, Isolation characterization and
sequencing of the pyruvate kinase gene of Saccharomyces cerevisiae, ]. Biol. Chem.
258:2193-2201.
Cannon, J. F., Gibbs, J. B., and Tatchell, K., 1986, Suppressors of the ras2 mutation of
Saccharomyces cerevisiae, Genetics 113:247-264.
Casperson, G. F., Walker, N., and Bourne, H. R., 1985, Isolation of the gene coding for
adenylate cyclase in Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. USA. 82:5060-5063.
Celenza, j. L., and Carlson, M., 1984, Cloning and genetic mapping of SNF-1 a gene
required for expression of glucose repressible genes in Saccharomyces cerevisiae, Mol.
Cell. Biol. 4:49-53.
Celenza, j. L., and Carlson, M., 1986, A yeast gene that is essential for release from glucose
repression encodes a protein kinase, Science 233:1175-1180.
Chang, L. F. L., Cunningham, T. S., Gatzek, P.R., Chen, W.j., and Kohlhaw, G. B., 1984,
Cloning and characterization of yeast LEU4, one of 2 genes responsible for a-iso-
propylamalate synthesis, Genetics 108:91-106.
Charron, M.J., Dubin, R. A., and Michels, C. A., 1986, Structural and functional analysis
of the MALl locus of Saccharomyces cerevisiae, Mol. Cell. Biol. 6:3891-3899.
Cherest, H., Nguyen, N. T., and Surdin-Kerjan, Y., 1985, Transcriptional regulation of the
METJ gene of Saccharomyces cerevisiae, Gene 34:269-281.
Chevallier, M-R., 1982, Cloning and transcriptional control of a eukaryotic permease gene,
Mol. Cell. Biol. 2:977-984.
Ciriacy, M., and Breitenbach, I., 1979, Physiological effects of seven different blocks in
glycolysis in Saccharomyces cerevisiae,]. Bacteriol. 139:152-168.
Cirillo, V. P., 1968, Relationship between sugar structure and competition for the sugar
transport system in bakers' yeast,]. Bacteriol. 95:603-611.
Clifton, D., and Fraenkel, D. G., 1982, Mutant studies of yeast phosphofructokinase,
Biochemistry 21:1935-1942.
Clifton, D., and Fraenkel, D. G., 1983, Fructose 2,6-bisphosphate and fructose-6-P 2-kinase
in Saccharomyces cerevisiae in relation to metabolic state in wild type and fructose-6-P !-
kinase mutant strains,]. Biol. Chem. 258:9245-9249.
Clifton, D., Weinstock, S. G., and Fraenkel, D. B., 1978, Glycolysis mutants in Saccharomyces
cerevisiae, Genetics 88: 1-11.
Cooper, T. G., 1982a, Nitrogen metabolism in Saccharomyces cerevisiae, in: The Molecular
Biology of the Yeast Saccharomyces. Metabolism and Gene Expression U· N. Strathern, E. W.
jones, and J. R. Broach, eds.), Cold Spring Harbor Laboratory, Cold Spring Harbor,
NY, pp. 39-99.
Cooper, T. G., 1982b, Transport in Saccharomyces cerevisiae, in: The Molecular Biology of the
Yeast Saccharomyces. Metabolism and Gene Expression U· N. Strathern, E. W. jones, and J.
R. Broach, eds.), Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, pp.
399-461.
Crabeel, M., Messenguy, F., Lacroute, F., and Glansdorff, 1981, Cloning ARGJ, the gene
for ornithine carbamoyltransferase from S. cerevisiae. Expression in E. coli requires
secondary mutations. Production of plasmid ~-lactamase in yeast, Proc. Natl. Acad. Sci.
USA 78:5026-5030.
Creanor,J., and Mitchison,J. M., 1982, Patterns of protein synthesis during the cell cycle of
the fission yeast Schir.osaccharomyces pombe,]. CeU. Sci. 58:263-285.
Creanor, J., Elliott, S. G., Bisset, Y. C., and Mitchison, J. M., 1983, Absence of step changes
in activity of certain enzymes during the cell cycle of budding and fission yeasts in
synchronous cultures,]. CeU. Sci. 61:339-349.
Crowley, J. C., and Kaback, D. B., 1984, Molecular cloning of chromosome 1 DNA from
Saccharomyces cerevisiae: Isolation of the ADEJ gene,]. Bacterial. 159:413-417.
Csaikl, V., and Csaikl, F., 1986, Molecular cloning and characterisation of the MET6 gene
of Saccharomyces cerevi.siae, Gene 46:207-214.
David,J., Brake, E., Blair, L., Kunisawa, R., and Thorner,]., 1984, Isolation ofthe putative
structural gene for the lysine-arginine-cleaving endopeptidase required for process-
ing of yeast prepro-cx-factor, CeU 57:1075-1090.
Dawes, I. W., 1975, Study of cell development using derepressed mutations, Nature
255:707-708.
den Hollander,]. A., Brown, T. R., Ugurbil, K., and Shulman, R. G., 1979, ISC nuclear
magnetic resonance studies of anaerobic glycolysis in suspensions of yeast cells, Proc.
Natl. Acad. Sci. USA. 76:6069-6100.
Denis, C. L., and Young, E. T., 1983, Isolation and characterization of the positive reg-
ulatory gene ADRJ from Saccharomyces cerevisiae, Mol. CeU. Biol. 3:360-370.
Deschamps, J., and Wiame, J. M., 1979, Mating-type effect on cis mutations leading to
constitutivity of ornithine transminase in diploid cells of Saccharomyces cerevisiae, Genet-
ics 92:749-758.
Dickinson, J. R., and Hewlins, M. J. E., 1988, A study of the role of the hexose monophos-
phate pathway with respect to fatty acid biosynthesis in sporulation of Saccharomyces
cerevisiae,]. Gen. Microbial. 154:333-337.
Dickinson, J. R., and Williams, A. S., 1987, The cdcJO mutation in Saccharomyces cerevisiae
results in a temperature-sensitive isoenzyme of phosphoglucose isomerase, ]. Gen.
Microbial. 135:135-140.
Dickinson,]. R., Dawes, I. W., Boyd, A. S. F., and Baxter, R. L., 1983, ISC NMR studies of
acetate metabolism during sporulation of Saccharomyces cerevisiae, Proc. Natl. Acad. Sci.
USA 80:5847-5851.
Dickinson, J. R., Ambler, R. P., and Dawes, I. W., 1985, Abnormal amino acid metabolism
in mutants of Saccharomyces cerevisiae affected in the initiation of sporulation, Eur.].
Biochem. 148:405-406.
DiNardo, S., Voelkel, K., and Sternglanz, R., 1984, DNA topoisomerase II mutant of
Saccharomyces cerevisiae: Topoisomerase II is required for segregation of daughter
molecules at the termination of DNA replication, Proc. Natl. Acad. Sci. USA 61:2616-
2620.
Dobson, M. J., Tuite, M. F., Roberts, N. A., Kingsman, A. J., Kingsman, S.M., Perkins, R.
E., Conroy, S. C., Dunbar, B., and Fothergill, L. A., 1982, Conservation of high
efficiency promoter sequences in Saccharomyces cerevisiae, Nucl. Acid Res. 10:2625-
2637.
Donahue, T. F., Farabaugh, P. J., and Fink, G. R., 1982, The nucleotide sequence of the
Hi.s4 region of yeast, Gene 16:47-59.
Drain, P., and Schimmel, P., 1986, Yeast LEU5 is a PET-like gene that is not essential for
leucine biosynthesis, Mol. Gen. Genet. 204:397-403.
92 J. R. DICKINSON
Dubin, R. A., Perkins, E. L., Needleman, R. B., and Michels, C. A., 1986, Identification of a
second trans-acting gene controlling maltose fermentation in Saccharomyces carlsbergen-
sis, Mol. Cell. Biol. 6:2757-2765.
Dubois, E., and Messenguy, F., 1985, Isolation and characterization of the yeast ARGRIII
gene involved in regulating both anabolism and catabolism of arginine, Mol. Gen.
Genet. 198:283-289.
Dubois, E., Hiernaux, D., Grenson, M., and Wiame, J. M., 1978, Specific induction of
catabolism and its relation to repression of biosynthesis in arginine metabolism of
Saccharomyces cerevisiae,J. Mol. Biol. 122:383-406.
Dubois, E., Bercy, J., and Messenguy, F., 1987, Characterisation of two genes ARGRI and
ARGRIII required for specific regulation of arginine metabolism in yeast, Mol. Gen.
Genet. 207:142-148.
Eckerdt, F., Kunz, B. A., and Haynes, R. H., 1983, Variation of mutation and recombina-
tion frequencies over a range of thymidylate concentrations in a diploid thymidylate
auxotroph of yeast, Curr. Genet. 7:399-402.
Eibel, H., and Philippsen, P., 1983, Identification of the cloned Saccharomyces cerevisiae
ll'S2 gene by an integrative transformation approach, Mol. Gen. Genet. 191:66-73.
Elledge, S. J., and Davis, R. W., 1987, Identification and isolation of the gene encoding the
small subunit of ribonucleotide reductase from Saccharomyces cerevisiae: DNA damage-
inducible gene required for mitotic viability, Mol. Cell. Biol. 7:2783-2793.
Entian, K-D., and Frohlich, K-U., 1984, Saccharomyces cerevisiae mutants provide evidence
of hexokinase PII as a bifunctional enzyme with catalytic and regulatory domains for
triggering carbon catabolite repression,]. Bacteriol. 158:29-35.
Entian, K-D., Kopetzki, E., Frohlich, K. U., and Mecke, D., 1984, Cloning of hexokinase
isoenzyme PI from Saccharomyces cerevisiae. PI transformants confirm the unique role
of hexokinase isoenzyme PII for glucose repression in yeasts, Mol. Gen. Genet. 198:50-
54.
Entian, K-D., Meurer, B., KOhler, H., Mann, K-H., and Mecke, D., 1987, Studies on the
regulation of enolases and compartmentation of cytosolic enzymes in Saccharomyces
cerevisiae, Biochem. Biophys. Acta 92!1:214-221.
Errede, B., Cardillo, T. S., Sherman, F., Dubois, E., Deschamps, j., and Wiame, J. M.,
1980a, Mating signals control expression of mutations resulting from insertion of a
transposable repetitive element adjacent to diverse yeast genes, Cell22:427-436.
Errede, B., Cardillo, T. S., Wever, G., and Sherman, F., 1980b, Studies on transposable
elements in yeast. I. ROAM mutations causing increased expression of yeast genes:
their activation by signals directed toward conjugation functions and their formation
by insertion of Ty I repetitive elements, Cold Spring Harbor Symp. Quant. Biol. 45:593-
602.
Esposito, R. E., and Kalapholz, S., 1981, Meiosis and ascospore development, in: The
Molecular Biology rfthe Yeast Saccharomyces. Life Cycle and Inheritance (J. N. Strathern,
E. W. jones, and J. R. Broach, eds.), Cold Spring Harbor Laboratory, Cold Spring
Ferguson, J. J., Boll, M., and Holzer, H., 1967, Yeast malate dehydrogenase: Enzyme
inactivation in catabolite repression, Eur. ]. Biochem. 1:21-25.
Fonzi, W. A., and Sypherd, P. S., 1985, Expression ofthe gene for ornithine decarboxylase
of Saccharomyces cerevisiae in Escherichia coli, Mol. Cell. Biol. 5:161-166.
Fraenkel, D. G., 1982, Carbohydrate metabolism, in: The Molecular Biology rf the Yeast
Saccharomyces. Metabolism and Gene Expression (J. N. Strathern, E. W. jones, and J. R.
Broach, eds.), Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, pp. 1-37.
Freese, E. B., Olempska-Beer, Z., Hartig, A., and Freese, E., 1985, Initiation of meiosis and
sporulation of Saccharomyces cerevisiae by sulfur or guanine deprivation, Dev. Biol.
102:438-451.
Frohlich,. K-U., and Entian, K-D., 1982, Regulation of gluconeogenesis in the yeast Sac-
charomyces cerevisiae. Evidence for conversion of enolase isoenzymes, FEBS Lett.

139: 164-166.
Frohlich, K-U ., Entian, K-D., and Mecke, D., 1984, Cloning and restriction analysis of the
hexokinase Pll gene of the yeast Saccharomyces cerevisiae, Mol. Gen. Genet. 194:144-
148.
Funayama, S., Gancedo, J. M., and Gancedo, C., 1980, Turnover of yeast fructose bisphos-
phatase in different metabolic conditions, Eur.]. Biochem. 109:61-66.
Game, J. C., 1976, Yeast cell cycle mutant cdc21 is a temperature-sensitive thymidylate
auxotroph, Mol. Gen. Genet. 146:313-315.
Gancedo, C., 1971, Inactivation of fructose-! ,6-diphosphatase by glucose in yeast,]. Bac-
teriol. 107:401-405.
Genbauffe, F. S., and Cooper, T. G., 1986, Induction and repression of the urea amidoly-
ase gene in Saccharomyces cerevisiae, Mol. Cell. Biol. 6:3954-3964.
Grivell, A. R., and Jackson, J. F., 1968, Thymidine kinase: Evidence for its absence from
Neurospora crassa and some other microorganisms and the relevance of this to the
specific labelling of deoxyribonucleic acid,]. Gen. Microbiol. 54:307-317.
Guerry-Kopecko, P., and Wickner, R. B., 1980, Cloning of the URAl gene of Saccharomyces
cerevisiae,]. Bacteriol. 143:1530-1533.
Haarasilta, S., and Oura, E., 1975, On the activity and regulation of anaplerotic and
gluconeogenic enzymes during the growth processes of baker's yeast, Eur.]. Biochem.
52:1-7.
Harashima, S., Sidhu, R. S., Tohe, A., and Oshima Y., 1981, Cloning of the HJS5 gene of
Saccharomyces cerevisiae by yeast transformation, Gene 16:1-3.
Hartwell, L. H., 1967, Macromolecular synthesis in temperature-sensitive mutants of yeast,
]. Bacteriol. 93:1662-1670.
Hartwell, L. H., 1985, Cetus-UCLA Symposium on Yeast Cell Biology, Keystone, CO.
Hatzfeld, J., 1973, DNA labelling and its assay in yeast, Biochim. Biophys. Acta 299:34-42.
Heinisch, J., 1986, Isolation and characterisation of the two structural genes coding for
phosphofructokinase in yeast, Mol. Gen. Genet. 202:75-82.
Henry, S. A., Klig, L. S., and Loewy, B.S., 1984, The genetic regulation and coordination
of biosynthetic pathways in yeast: Amino acid and phospholipid synthesis, Annu. Rev.
Genet. 18:207-231.
Heredia, C. F., Sols, A., and DelaFuente, 1968, Specificity of the constitutive hexose trans-
port in yeast, Eur. ]. Biochem. 5:324-329.
Hinnebusch, A. G., and Fink. G. R., l983a, Repeated DNA sequences upstream from H!Sl
also occur at several other co-regulated genes in Saccharomyces cerevisiae,]. Biol. Chem.
258:5238-5247.
Hinnebusch, A. G., and Fink, G. R., l983b, Positive regulation in the general amino acid
control of Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. USA 80:5374-5378.
Hitzeman, R. A., Clarke, L., and Carbon, J., 1980, Isolation and characterisation of the
yeast 3-phosphoglycerokinase gene (PGK) by an immunological screening technique,
]. Biol. Chem. 255:12,073-12,080.
Hohmann, S., and Zimmermann, F. K., 1986, Cloning and expression on a multicopy
vector of five invertase genes of Saccharomyces cerevisiae, Curr. Genet. ll:217-225.
Holland, J. P., and Holland, M. J., 1980, Structural comparisons of two nontandemly
repeated yeast glyceraldehyde-3-P dehydrogenase genes, ]. Biol. Chem. 255:2596-
2605.
Holland, M. J., Holland,J. P., Thill, G. P., and Jackson, K. A., 1981, The primary structure
of two yeast enolase genes. Homology between the 5' noncoding flanking regions of
yeast enolase and glyceraldehyde-3-phosphate dehydrogenase genes, ]. Biol. Chem.
256:1385-1395.
Holland, M. J., Yokoi, T., Holland,]. P., Myambo, K., and Innis, M.A., 1987, The GCRI
gene encodes a positive transcriptional regulator of the enolase and glyceraldehyde-3-
94 J. R. DICKINSON
phosphate dehydrogenase gene families in Saccharomyces cerevisiae, Mol. .Cell. Biol.

7:813-820.
Holmberg, S., 1984, Genetic improvement of brewer's yeast, Trends Biotechnol. 2:98-102.
Hsiao, C-L., and Carbon, J., 1979, High frequency transformation of yeast by plasmids
containing the cloned yeast ARG4 gene, Proc. Natl. Acad. Sci. USA 76:3829-3833.
Hsu, Y. P., and Schimmel, P., 1984, Yeast Leul repression of mRNA levels and relationship
of 5' noncoding region to that of Lev2,]. Biol. Chem. 259:3714-3719.
Hsu, Y. P., Kohlhaw, G., and Niederberger, P., 1982, Evidence that a-isopropylmalate
synthase of Saccharomyces cerevisiae is under the "general" control of amino acid bio-
synthesis,]. Bacterial. 150:969-972.
Hubert,]. B., Guyonvarch, A., Kammerer, B., Exinger, Fl., Liljelund, P., and Lacroute, F.,
1983, Complete sequence of an eukaryotic regulatory gene, EMBO]. 2:2071-2073.
Jauniaux, J. C., Dubois, E., Crabeel, M., and Wiame,J. M., 1981, DNA and RNA analysis of
the arginase mutants in Saccharomyces cerevisiae, Arch. lnt. Physiol. Biochem. 89:B111-
B112.
Jauniaux, J. C., Dubois, E., Vissers, S., Crabeel, M., and Wiame, J. A., 1982, Molecular
cloning, DNA structure and RNA analysis of the arginase gene in Saccharomyces cere-
visiae. A study of cis-dominant regulatory mutations, EMBO]. 1: 1125-1131.
Jauniaux, J. C., Degols, G., Vissers, S., and Wiame, J. M., 1984, Regulation of L-ornithine
transaminase synthesis in Saccharomyces cerevisiae: Gene cloning, RNA and cis-domi-
nant regulatory mutations analysis. Abstract of paper presented at Twelfth International
Conference on Yeast Genetics and Molecular Biology), Edinburgh, U. K., p. 168.
Johnston, L. H., and Nasmyth, K. A., 1978, Saccharomyces cerevisiae cell cycle mutant cdc9 is
defective in DNA ligase, Nature 274:891...;,893.
Johnston, S. A., and Hopper, J. E., 1982, Isolation of the yeast regulatory gene GAL4 and
analysis of its dosage effects on the galactose-melibiose regulon, Proc. Natl. Acad. Sci.
USA 79:6971-6975.
Jones, E. W., 1984, The synthesis and function of proteases in Saccharomyces: Genetic
approaches, Annu. Rev. Genet. 18:233-270.
Jones, E. W., and Fink, G. R., 1982, Regulation ofamino acid and nucleotide biosynthesis
in yeast, in: The Molecular Biology if' the Yeast Saccharomyces. Metabolism and Gene
Expression (J. N. Strathern, E. W. Jones, and J. R. Broach, eds.), Cold Spring Harbor
Laboratory, Cold Spring Harbor, NY, pp. 181-299.
Jong, A. Y. S., Kuo, C. L., and Campbell, J. L., 1984, The CDC8 gene of yeast encodes
thymidylate kinase,]. Biol. Chem. 259: 1052-1059.
Kataoka, T., Powers, S., Cameron, S., Fasano, 0., Goldfarg, M., Broach,]., and Wigler, M.,
1985, Functional homology of mammalian and yeast RAS genes, Cell 40:19-26.
Kawasaki, G. H., 1979, The cdc19 mutation of Saccharomyces cerevisiae is a temperature-
sensitive defect for the glycolytic enzyme, pyruvate kinase, in: The Molecular Biology if'
Yeast (Abstract of paper presented at Cold Spring Harbor Meeting, 1979), Cold Spring
Harbor Laboratory, Cold Spring Harbor, NY, p. 35.
Kawasaki, G. H., and Fraenkel, D. G., 1982, Cloning of yeast glycolysis genes by comple-
mentation, Biochem. Biophys. Res. Commun. 108:1107-1112.
Kiehl, F., Huffaker, T., and Tanner, W., 1984, A temperature-sensitive N-glycosyla-
tion mutant of S. cerevisiae that behaves like a cell-cycle mutant, Exp. Cell Res. 150:309-
313.
Klig, L. S., and Henry, S. A., 1984, Isolation of the yeast JNOJ gene: located on an
autonomously replicating plasmid the gene is fully regulated, Proc. Natl. Acad. Sci. USA
81:3816-3820.
Kotyk, A., 1967, Properties of the sugar carrier in bakers' yeast II, Specificity of transport,
Folia Microbial. 12:121-131.
Kunz, B. A., and Haynes, R. H., 1982, DNA repair and the genetic effects of thymidylate
stress in yeast, Mutat. Res. 93:353-375.
Kunz, B. A., Barclay, B.J., Game,J. C., Little,J. G., and Haynes, R. H., 1980, Induction of
mitotic recombination in yeast by starvation for thymine nucleotides, Proc. Natl. Acad.
Sci. USA 77:6057-6061.
Kunz, B. A., Taylor, G. R., Konforti, B., Glickman, B. W., and Haynes, R. H., 1984, Inhibi-
tion of thymidylate biosynthesis induces mitotic unequal sister chromatid recombina-
tion in Saccharomyces cerevisiae, Curr. Genet. 8:211-217.
Kunz, B. A., Taylor, G. R., and Haynes, R. H., 1985, Mating-type switching in yeast is
induced by thymine nucleotide depletion, Mol. Gen. Genet. 199:540-542.
Kuziora, M. A., Chalmers, J. H. Jr., Douglas, M. S., Hitzeman, R. A., Mattick, J. S., and
Wakil, S. J., 1983, Molecular cloning of fatty acid synthetase genes from Saccharomyces
cerevisiae,j. Biol. Chern. 258:11648-11653.
Lagunas, R., and Gancedo,J. M., 1973, Reduced pyridine nucleotides balance in glucose-
growing Saccharomyces cerevisiae, Eur. ]. Biochern. 57:90-94.
Langin, T., Faugeron, G., Goyon, C., Nicolas, A., and Rossignol, J-L., 1986, The MET2
gene of Saccharomyces cerevisiae: Molecular cloning and nucleotide sequence, Gene
49:283-293.
Larimer, F. W., Morse, C. C., and Beck, A. K., 1983, Isolation of the ARO1 cluster gene of
Laughon, A., and Gesteland, R. F., 1982, Isolation and preliminary characterization of the
GAL4 gene, a positive regulator of transcription in yeast, Proc. Natl. Acad. Sci. USA
79:6827-6831.
Lemoine, Y., Dubois, E., and Wiame, J. M., 1978, The regulation of urea amidolyase of
Saccharomyces cerevisiae, Mol. Gen. Genet. 166:251-258.
Letts, V. A., Klig, L. S., Bae-Lee, M., Carman, G. M., and Henry, S. A., 1983, Isolation of
the yeast structural gene for the membrane-associated enzyme phosphatidyl serine
synthetase, Proc. Natl. Acad. Sci. USA 80:7279-7283.
Liao, H., and Thorner, J., 1980, Yeast mating pheromone a-factor inhibits adenylate
cyclase, Proc. Natl. Acad. Sci. USA 77:1898-1902.
Liao, H., and Thorner, J., 1981, Adenosine 3',5' phosphate phosphodiesterase and phe-
romone response in the yeast Saccharomyces cerevisiae,j. Bacterial. 148:919-925.
Liljelund, P., and Lacroute, F., 1986, Genetic characterisation and isolation of the Sac-
charomyces cerevisiae gene coding for uridine monophosphokinase, Mol. Gen. Genet.
205:74-81.
Lloyd, D., and Scott, R. I., 1983, Direct measurement of dissolved gases in microbiological
systems using membrane inlet mass spectrometry,]. Microbial. Meth. 1:313-328.
Lloyd, D., Kristensen, B., and Degn, H., 1983a, Glycolysis and respiration in yeasts. The
Pasteur effect studied by mass spectrometry, Biochem.]. 212:749-754.
Lloyd, D., Kristensen, B., and Degn, H., 1983b, Glycolysis and respiration in yeasts: The
effect of ammonium ions studied by mass spectrometry,]. Gen. Microbial. 129:2125-
2127.
Lobo, Z., 1984, Saccharomyces cerevisiae aldolase mutants,]. Bacterial. 160:222-226.
Lobo, Z., and Maitra, P. K., 1982, A particulate phosphofrutokinase from yeast, FEBS Lett.
157:279-282.
Lobo, z., and Maitra, P. K., 1983, Phosphofructokinase mutants of yeast, ]. Biol. Chern.
258:1444-1449.
Lorincz, A. T., Miller, M. J., Xuong, N-H., and Geiduschek, E. P., 1982, Identification of
proteins whose synthesis is modulated during the cell cycle of Saccharomyces cerevisiae,
Mol. CeU. Biol. 2:1532-1549.
Losson, R., and Lacroute, F., 1982, Cloning of a eukaryote regulatory gene, Mol. Gen.
Genet. 184:394-399.
Lusty, C. J., and Jeffrey, L. U., 1982, Cloning of a yeast gene coding for arginine-specific
carbamoyl-phosphate synthetase, Proc. Natl. Acad. Sci. USA 79:2240-2244.
Mantsala, P., and Zalkin, H., 1984, Nucleotide sequence of Saccharomyces cerevisiae ADE4
encoding glutamine phosphoribosylpyrophosphate amidotransferase, ]. Biol. Chern.
259:8478-8484.
96 J. R. DICKINSON
Masson, P.,Jacquemin,J. M., and Culot, M., 1984, Molecular cloning of the tsm0185 gene
responsible for adenylate cyclase activity in Saccharomyces cerevisiae, Ann. Microbial.
(Inst. Pasteur) 155:343-351.
Matsumoto, K., Uno, I., Oshima, Y., and Ishikawa, T., 1982, Isolation and characterization
of yeast mutants deficient in adenylate cyclase and cAMP-dependent protein kinase,
Proc. Natl. Acad. Sci. USA 79:2355-2359.
Matsumoto, K., Uno, I., and Ishikawa, T., 1983a, Control of cell division in Saccharomyces
cerevisiae mutants defective in adenylate cyclase and cyclic AMP-dependent protein
kinase, Exp. Cell. Res. 146:151-161.
Matsumoto, K., Uno, I., and Ishikawa, T., 1983b, Initiation of meiosis in yeast mutants
defective in adenylate cyclase and cAMP-dependent protein kinase, Cell32:417-423.
Matsumoto, K., Uno, I., Kato, K., and Ishikawa, T., 1985, Isolation and characterization of
a phosphoprotein phosphatase-deficient mutant in yeast, Yeast 1:25-38.
Mazon, M. J., Gancedo, J. M., and Gancedo, C., 1982a, Inactivation of yeast fructose-1,6-
bisphosphatase.ln vivo phosphorylation of the enzyme,]. Biol. Chem. 257:1128-1130.
Mazon, M.J., Gancedo,J. M., and Gancedo, C., 1982b, Phosphorylation and inactivation of
yeast fructose-bisphosphatase in vivo by glucose and proton ionophores. A possible
role for cAMP, Eur.]. Biochem. 127:605-608.
McAlister, L., and Holland, M. J., 1985, Isolation and characterisation of yeast strains
carrying mutations in the glyceraldehyde-3-phosphate dehydrogenase genes,]. Biol.
Chem. 260:15013-15018.
Mcintosh, E. M., and Haynes, R. H., 1986, Sequence and expression of the dCMP de-
aminase gene (DCDJ) of Saccharomyces cerevisiae, Mol. Cell. Biol. 6:1711-1721.
Mcintosh, E. M., Gadsden, M. H., and Haynes, R. H., 1986, Transcription of genes
encoding enzymes involved in DNA synthesis during the cell cycle of Saccharomyces
Messenguy, F., Feller, A., Crabeel, M., and Pierard, A., 1983, Control mechanisms acting at
the transcriptional and post-transcriptionallevels are involved in the synthesis of the
arginine pathway carbamoyl-phosphate synthase of yeast, EMBO ]. 2: 1249-1254.
Michels, C. A., and Needleman, R. B., 1983, A genetic and physical analysis of the MALl
and MALJ genes in standard strains of Saccharomyces cerevisiae, Mol. Gen. Genet.
191:225-230.
Miyajima, A., Nakayama, N., Miyajima, I., Arai, N., Okayama, H., and Arai, K., 1984,
Analysis of full-length eDNA clones carrying GALl of Saccharomyces cereviaiae-A
model system for eDNA expression, Nucl. Acid Res. 12:6397-6414.
Moll, R., and Wintersberger, E., 1976, Synthesis of yeast histones in the cell cycle, Proc.
Moon, R. B., and Richards, J. H., 1973, Determination of intracellular pH by P magnetic
resonance,]. Biol. Chem. 248:7276-7278.
Muller, D., and Holzer, H., 1981, Regulation of fructose-1,6-bisphosphatase in yeast by
phosphorylation/dephosphorylation, Biochem. Biophys. Res. Commun. 103:926-933.
Nadkarni, M., Parmar, L., Lobo, Z., and Maitra, P. K., 1984, Mutations in the regulatory
subunit of soluble phosphofructokinase from yeast, FEBS Lett. 175:294-298.
Nath, K., and Baptist, E. W., 1984, Cloning of a yeast dihydrofolate reductase gene in E.
coli, Curr. Genet. 8:265-270.
Navon, G., Shulman, R. G., Yamane, T., Eccleshall, T. R., Lam, K-B., Baronofsky,J.J., and
Marmur, J., 1979, Phosphorus-31 nuclear magnetic resonance studies of wild-type
and glycolytic pathway mutants of Saccharomyces cerevisiae, Biochemistry 18:4487-4499.
Niederacher, D., and Entian, K-D., 1987, Isolation and characterisation of the regulatory
HEX2 gene necessary for glucose repression in yeast, Mol. Gen. Genet. 206:505-509.
Niederberger, P., Furter, R., Prantl, F., Braus, G., and Huetter, R., 1984, Cloning of yeast
TRP genes, Riv. Biol. 77:600-603.
Nikawa,J., and Yamashita, S., 1984, Molecular cloning of the gene encoding CDPdiacylgly-
cerol-inositol 3-phosphatidyl transferase in Saccharomyces cerevisiae, Eur. ]. Biochem.

143:251-256.
Nogi, Y., Shimada, H., Matsuzaki, Y., Hashimoto, H., and Fukasawa, T., 1984, Regulation
and expression of the galactose gene cluster in Saccharomyces cerevisiae. 2. The isolation
and dosage effect of the regulatory gene GALBO, Mol. Gen. Genet. 195:29-34.
Parmar, L., Lobo, Z., Nadkarni, M., and Maitra, P. K., 1984, Multiple genes control partic-
ulate phosphofructokinase of yeast, Mol. Gen. Genet. 197:515-516.
Patterson, M., Sclafani, R. A., Fangman, W. L., and Rosamond,]., 1986, Molecular charac-
terisation of the cell cyle gene CDC7 from Saccharomyces cerevisiae, Mol. Cell. Biol.
6:1590-1598.
Petersen, J. G. L., Holmberg, S., Nilsson-Tillgren, T., and Kielland-Brandt, M. C., 1983,
Molecular cloning and characterization of the threonine deaminase (1 LVI) gene of
Saccharomyces cerevisiae, Carlsberg Res. Commun. 48: 149-160.
Polaina, J., 1984, Cloning of the ILV2, ILVJ and ILV5 genes of Saccharomyces cerevisiae,
Carlsberg Res. Commun. 49:577-584.
Powers, S., Michaelis, S., Broek, D., Anna-A., S. S., Field,J., Herskowitz, 1., and Wigler, M.,
1986, RAM, a gene of yeast required for a functional modification of RAS proteins
and for production of mating pheromone a-factor, Cell47:413-422.
Molecular Biology of the Yeast Saccharomyces. Life Cycle and Inheritance (J. N. Strathern,
E. W. Jones, and J. R. Broach, eds.), Cold Spring Harbor Laboratory, Cold Spring
Purwin, C., Leidig, F., and Holzer, H., 1982, Cyclic AMP-dependent phosphorylation
of fructose-1,6-bisphosphatase in yeast, Biochem. Biophys. Res. Commun. 107:1482-
1489.
Rai, R., Genbauffe, F. S., Lea, H. Z., and Cooper, T. G., 1987, Transcriptional regulation of
the DAL5 gene inS. cerevisiae,J. Bacteriol. 169:3521-3524.
Ratzkin, B., and Carbon, J., 1977, Functional expression of cloned yeast DNA in E. coli,
Reed, S. 1., Hadwiger, J. A., and Lorincz, A. T., 1985, Protein kinase activity of the yeast
cell division cycle gene CDC28, Proc. Natl. Acad. Sci. USA 82:4055-4059.
Roberts, L. M., and Schewizer, E., 1984, Molecular cloning ofthe yeast fatty acid synthetase
genes FASJ and FAS2. Illustrating the structure of the FASJ cluster gene by transcrip-
tion mapping and transcription studies, Mol. Gen. Genet. 194:457-465.
Robinson, L. C., Gibbs,J. B., Marshall, M.S., Sigal, I. S., and Tatchell, K., 1987, CDC25: A
component of the RAS-adenylate cyclase pathway in Saccharomyces cerevisiae, Science
235:1218-1221.
Rodicio, R., and Heinisch,]., 1987, Isolation of the yeast phosphoglyceromutase gene and
construction of deletion mutants, Mol. Gen. Genet. 206:133-140.
Rodicio, R., and Zimmerman, F. K., 1985a, Cloning of maltase regulatory genes in Sac-
charomyces cerevisiae. I. Isolation of the MAL2-8C regulatory gene, Curr. Genet. 9:539-
545.
Rodicio, R., and Zimmerman, F. K., 1985b, Cloning of maltase regulatory genes in Sac-
charomyces cerevisiae. 2. Isolation of the MAL4 regulatory gene, Curr. Genet. 9:54 7-551.
Roeder, G. S., Beard, C., Smith, M., and Keranen, S., 1985, Isolation and characterisation
of the SPT2 gene, a negative regulator of Ty-controlled yeast gene expression, Mol.
Cell. Biol. 5:1543-1553.
Rothstein, R. J., and Sherman, F., 1980, Dependence on mating type for the overproduc-
tion of iso-2-cytochrome c in the yeast mutant CYC7-H2, Genetics 94:891-898.
Rousseau, P., 1972, Germination of yeast spores in Saccharomyces cerevisiae, Ph.D. thesis,
University of Wisconsin, Madison.
Roy, D. J., and Dawes, I. W., 1987, Cloning and characterisation of the gene encoding
lipoamide dehydrogenase in Saccharomyces cerevisiae, ]. Gen. Microbiol. 133:925-933.
98 J. R. DICKINSON
Saheki, T., and Holzer, H., 1975, Proteolytic activities in yeast, Biochim. Biophys. Acta
384:203-214.
Sangsoda, S., Cherest, H., and Surdin-Kerjan, Y., 1984, The expression of the MET25
gene of Saccharomyces cerevisiae is regulated transcriptionally, Mol. Gen. Genet. 200:407-
414.
Sasnauskas, K., Gedminieno, G., Naktinis, V., Citavicius, D., and Janulaitis, A., 1982,
Cloning of the ADE2 gene of Saccharomyces cerevisiae, DOKJ. Adad. Nauk. SSSR
263:224-227.
Sass, P., Field, J., Nikawa, J., Toda, T., and Wigler, M., 1986, Cloning and characterisation
of the high-affinity cAMP phosphodiesterase of Saccharomyces cerevisiae, Proc. Natl.
Acad. Sci. USA. 83:9303-9307.
Satoshi, H., Sidhu, R. S., Tohe, A., and Oshima, Y., 1981, Cloning of the HJS5 gene of
Saccharomyces cerevisiae by yeast transformation, Gene 16:335-342.
Schmitt, H. D., Ciriacy, M., and Zimmerman, F. K., 1983, The synthesis of yeast pyruvate
decarboxylase is regulated by large variations in the messenger RNA level, Mol. Gen.
Genet. 192:247-252.
Sclafani, R. A., and Fangman, W. L., 1984, Yeast gene CDCB encodes thymidylate kinase
and is complemented by herpes thymidine kinase gene TK, Proc. Natl. Acad. Sci. USA
81:5821-5825.
Sedivy, j. M., and Fraenkel, D. G., 1985, Fructose bisphosphatase of Saccharomyces cerevisiae.
Cloning, disruption and regulation of the FBP 1 structural gene,]. Mol. Biol. 186:307-
319.
Skvirsky, R. C., Greenberg, M. L., Myers, P. L., and Greer, H., 1986, A new negative control
gene for amino acid biosynthesis in Saccharomyces cerevisiae, Curr. Genet. 10:495-501.
Souciet, J. L., Hubert, J. C., and Lacroute, F., 1982, Cloning and restriction mapping ofthe
yeast URA2 gene coding for the carbamoyl phosphate synthetase aspartate transcar-
bamoylase complex, Mol. Gen. Genet. 186:385-390.
St.John, T. P., and Davis, R. W., 1979, Isolation of glactose inducible DNA sequences from
Saccharomyces cerevisiae by differential plaque filter hydridization, Cell 16:443-452.
Storms, R. K., Ord, R. W., Greenwood, M. T., Mirdamadi, B., Chu, F. K., and Belfort, M.,
1984, Cell cycle dependent expression of thymidylate synthase in Saccharomyces cere-
visiae, Mol. Cell. Biol. 4:2858-2864.
Struhl, K., Stinchcomb, D. T., Scherer, S., and David, R. W., 1979, High-frequency trans-
formation of yeast: Autonomous replication of hybrid DNA molecules, Proc. Natl.
Acad. Sci. USA 76:1035-1039.
Sudbery, P. E., Goodey, A. R., and Carter, B. L.A., 1980, Genes which control cell pro-
liferation in the yeast Saccharomyces cerevisiae, Nature 288:401-404.
Suissa, M., Suda, K., and Schatz, G., 1984, Isolation of the nuclear Saccharomyces cerevisiae
gene for citrate synthase and 15 other mitochondrial proteins by a new screening
method, EMBO]. 3:1773-1782.
Sumrada, R. A., and Cooper, T. G., 1982, Isolation of the CARl gene from Saccharomyces
cerevisiae and analysis of its expression, Mol. Cell. Biol. 2:1514-1523.
Sumrada, R. A., and Cooper, T. G., 1985, Target sites for positive and negative regulatory
elements controlling expression of an inducible eukaryotic gene, UCLA Symp. Mol.
Cell. Biol. New Ser. 30:291-301.
Taguchi, A. K. W., and Young, E. T., 1987, The cloning and mapping of ADR6, a gene
required for sporulation and for expression of the alcohol dehydrogenase II iso-
enzyme from Saccharomyces cerevisiae, Genetics 116:531-540.
Taguchi, A. K. W., Ciriacy, M., and Young, E. T., 1984, Carbon source dependence of
transposable element-associated gene activation in Saccharomyces cerevisiae, Mol. Cell.
Biol. 4:61-68.
Taylor, G. R., Barclay, B.J., Storms, R. K., Friesen,J. D., and Haynes, R. M., 1982, Isolation
of the thymidylate synthetase gene (TMP 1) by genetic complementation in yeast, Mol.
Cell. Biol. 2:437-442.
Taylor, G. R., Lagosky, P. A., Storms, R. K., and Haynes, R. H., 1987, Molecular charac-
terisation of the cell cycle regulated thymidylate synthase gene of Saccharomyces cere-
visiae,]. Bioi. Chern. 262:5298-5307.
Teshiba, S., Furter, R., Niederberger, P., Braus, G., Paravicini, G., and Hutter, R., 1986,
Cloning of the AROJ gene of Saccharomyces cerevisiae and its regulation, Mol. Gen. Genet.
205:353-357.
Tingle, M., Singh Klar, A.J., Henry, S. A., and Halvorson, H. 0., 1973, Ascospore forma-
tion in yeast, in: Microbial Differentiation, 2Jrd Symposium of the Society for General Micro-
biology 0· M. Ashworth andj. E. Smith, eds.), Cambridge University Press, Cambridge,
pp. 209-243.
Toda, T., Uno, I., Ishikawa, T., Powers, S., Kataoka, T., Broek, D., Cameron, S., Broach,J.,
Matsumoto, K., and Wigler, M., 1985, In yeast, RAS proteins are controlling elements
of adenylate cyclase, Cell 40:27-36.
Toda, T., Cameron, S., Sass, P., Zoller, M., and Wigler, M., 1987a, Three different genes in
S. cerevisiae encode the catalytic subunits of the cAMP-dependent protein kinase, Cell
50:277-287.
Toda, T., Cameron, S., Sass, P., Zoller, M., Scott,J. D., McMullen, B., Hurwitz, M., Krebs,
E. G., and Wigler, M., 1987b, Cloning and characterisation of BCYJ, a locus encoding
a regulatory subunit of the cAMP dependent protein kinase in yeast, Mol. Cell Bioi.
7:1371-1377.
Tortora, P., Birtel, M., Lenz, A. G., and Holzer, H., 1981, Glucose-dependent metabolic
interconversion of fructose- I ,6-bisphosphatase in yeast, Biochem. Biophys. Res. Commun.
100:688-695.
Tortora, P., Burlini, N., Hanozet, G. M., and Guerritore, A., 1982, Effect of caffeine on
glucose-induced inactivation of gluconeogenetic enzymes in Saccharomyces cerevisiae. A
possible role of cyclic AMP, Eur.]. Biochem. 126:617-622.
Tortora, P., Burlini, N., Caspani, G., and Guerretore, A., 1984, Studies on glucose-induced
inactivation of gluconeogenetic enzymes in adenylate cyclase and cAMP-dependent
protein kinase mutants, Eur.]. Biochem. 145:543-548.
Tschopp,J. F., Emr, S.D., Field, C., and Schekman, R., 1986, GAL2 codes for a membrane-
bound subunit of the galactose permease in Saccharomyces cerevisiae, ]. Bacterial.
166:313-318.
Tsuboi, M., and Yanagishima, N., 1973, Effects of cAMP, theophilline and caffeine on the
glucose repression of sporulation in Saccharomyces cerevisiae, Arch. Mikrobiol. 93:1-12.
Tsuboi, M., Kamisaka, S., and Yanagishima, N., 1972, Effects of cyclic 3'-5'-adenosine
monophosphate on the sporulation of Saccharomyces cerevisiae, Plant CellPhysiol. 13:585-
588.
Uno, 1., Matsumoto, K., and Ishikawa, T., 1982, Characterization of cAMP requiring yeast
mutants altered in the regulatory subunit of protein kinase,]. Bioi. Chern. 257:14110-
14115.
Uno, I., Matsumoto, K., and Ishikawa, T., 1983, Characterization of a cyclic nucleotide
phosphodiesterase-deficient mutant in yeast, f. Biol. Chern. 258:3539-3542.
Uno, 1., Matsumoto, K., Adachi, K., and Ishikawa, T., 1984, Characterization of cyclic
AMP requiring yeast mutants altered in the catalytic subunit of protein kinase,]. Biol.
Chern. 259:12508-12513.
Vander Plaat, J. B., and Van Solingen, P., 1974, Cyclic 3',5'-adenosine monophosphate
stimulates trehalose degradation in baker's yeast, Biochern. Biophys. Res. Commun.
56:580-587.
Vezinhet, F., Kinnaird, J. H., and Dawes, I. W., 1979, The physiology of mutants de-
repressed for sporulation in Saccharomyces cerevisiae, ]. Gen. Microbial. 115:391-402.
Walsh, R. B., Kawasaki, G., and Fraenkel, D. G., 1983, Cloning of genes that complement
yeast hexokinase and glucokinase mutants,]. Bacterial. 154:1002-1004.
Welch, P., and Scopes, R. K., 1985, Studies on cell-free metabolism: Ethanol production by a
yeast glycolytic system reconstituted from purified enzymes,]. Biotechnol. 2:257-273.
100 J. R. DICKINSON
Wickner, R. B., 1974, Mutants of Saccharomyces cereVISwe that incorporate deox-

ythymidine-5'-monophosphate into deoxyribonucleic acid in vivo, ]. Bacterial.
117:252-260.
Williamson, D. H., 1965, The timing of DNA synthesis in the cell cycle of Saccharomyces
cerevisiae,]. Cell. Biol. 25:517-528.
Williamson, V. M., Benetzen,J., Young, E. T., Nasmyth, K., and Hall, B.J., 1980, Isolation
of the structural gene for alcohol dehydrogenase by genetic complementation in yeast,
Nature 283:214-216.
Witt, I., Kronau, R., and Holzer, H., 1966, Repression von alkoholdehydrogenase, iso-
citratlyase und malatsynthase in hefe durch glucose, Biochim. Biophys. Acta 118:522-
537.
Wolf, D. H., 1982, Proteinase action in vitro versus proteinase function in vivo: Mutants
shed light on intracellular proteolysis in yeast, Trends Biochem. Sci. 7:35-37.
Woolford, C. A., Daniels, L. B., Park, F.J.,Jones, E. W., Van Arsdeli,J. N., and Innis, M.A.,
1986, The PEP4 gene encodes an aspartyl protease implicated in the posttranslational
regulation of Saccharomyces cerevisiae vacuolar hydrolases, Mol. Cell. Biol. 6:2500-2510.
Yoo, H-S., Genbauffe, F. S., and Cooper, T. G., 1985, Identification of the ureidoglycolate
hydrolase gene in the DAL gene cluster of Saccharomyces cerevisiae, Mol. Cell Biol.
5:2279-2288.
Young, E. T., Williamson, V. M., Taguchi, A. K. W., Smith, M., Sledziewski, A., Russel, D.,
Osterman, J., Denis, C. L., Cox, D., and Beier, D., 1982, The alcohol dehydrogenase
genes of the yeast Saccharomyces cerevisiae isolation, structure and regulation, Basic Life
Sci. 19:335-361.
Zalkin, H., and Yanofsky, C., 1982, Yeast gene TRP5: Structure, function, regulation,].
Biol. Chem. 257:1491-1450.
Methods in Classical Genetics
4
R. B. WICKNER
I. INTRODUCTION
This chapter deals with the classical genetics of Saccharomyces cerevisiae

largely from the methods viewpoint. Some of the interesting areas that
have been studied in this way will be mentioned, particularly in the
section on cytoplasmic genetics, which is the author's particular interest.
Many previous reviews have dealt with the subjects treated here, but
the reader is especially referred to the two volumes entitled The Mq-
lecular Biology of the Yeast Saccharomyces (Strathern et al., 1981), in which
the results of studies of yeast genetics are summarized. Reviews on near-
ly all aspects of yeast genetics and molecular biology by leading workers
in the field can be found in these volumes.
2. LIFE CYCLE OF S. CEREVISIAE
While most wild strains are diploid and most strains used for fermen-
tation are diploid or of even higher ploidy, most laboratory strains are
haploid. Yeast of any ploidy up to at least tetraploid can be easily con-
structed and relatively stably maintained. The usual life cycle is shown
diagramatically in Fig. 1.
Haploid cells can grow mitotically indefinitely with a doubling time
as low as 90 min on rich medium. But yeast has two sexes, or mating
types, called a and a. They are not called male and female or + and -
because the mating process is symmetrical. Insofar as is known, all the
genetic information from both parents, both chromosomal and cytoplas-
mic (nonchromosomal), is contributed to the zygotes. When cells of op-
posite mating type come in contact on rich medium, each arrests the cell
*This chapter was completed in 1985.
R. B. WICKNER • National Institutes of Health, Bethesda, Maryland 20892.
101
102 R. B. WICKNER
,1.mating type ~
haploid (nl ~
Mating
X
.!! mating typa l'fJ)"\ (call fusionl
haploid (nl ~
·N source
"-----,®
Meiosis
& -Dextrose
Sporulation +Acetate
!!.®
®
haploid
spore clones .1.
Dissect
Germinate
4 haploid
spores
in ascus
·®
Figure 1. Life cycle of S. cerevisiae. Mating of cells of opposite mating type occurs under
growth conditions. Diploids can grow mitotically indefinitely or, under the conditions
indicated, be induced to undergo meiosis and spore formation. Mating type segregates 2 a
: 2 Ot in the usual crosses. It is determined by alleles at a locus on chromosome III.
cycle of the other near the start point via a constitutively produced
polypeptide mating pheromone. That produced by a cells is called a
hormone and has the sequence (Trp)-His-Trp-Leu-Gln-Leu-Lys-Pro-
Gly-Gln-Pro-Met)Tyr. That produced by a cells is called a hormone and
has the sequence Lys-Gly-Val-Phe-Trp-Ala-Asx-Pro. Interestingly, the a
pheromone, that has a sex-related function in yeast, is similar in se-
quence to the human hypothalamic hormone, gonadotropin-releasing
hormone (Glu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-Gly-NH 2 ), and can act
in place of the latter (Loumaye et al., 1982).
The expressed mating type of a yeast strain depends on what infor-
mation is present at the MAT locus on chromosome III. While most
laboratory strains stably maintain their mating type, most wild-type
strains rapidly alternate between a and a. This phenomenon is due to the
presence of extra copies of a and a mating-type information at sites on
chromosome III distant from the MAT locus. These sequences are fre-
quently transposed to the MAT locus with elimination of the sequences
previously residing there (Herskowitz and Oshima, 1981). Expression of
the extra copies of a and a information is repressed. The HO gene,
necessary for the transposition reaction, is missing or defective in most
laboratory strains, so that these strains have a stable mating type. This
METHODS IN CLASSICAL GENETICS 103
control of cell "differentiation" by DNA rearrangement or transposition

is analogous to similar events in higher eukaryotes. Yeast also has a
transposon, called Ty 1, whose properties resemble both bacterial trans-
posons and RNA tumor viruses.
The haploid cells of opposite mating type, arrested near their start
points, agglutinate in a specific reaction and fuse to form a transient
heterokaryon, that is, a single cell with two nuclei. In the normal mating
process, this is followed quickly by karyogamy (nuclear fusion) to pro-
duce a diploid cell (the zygote). The zygotes can be recognized under the
microscope and directly picked out of a mating mixture with a micro-
manipulator needle. The first bud of a zygote usually emerges from the
equatorial region of the dumbbell-shaped zygote cell, giving an easily
recognized appearance. More commonly, one selects prototrophic dip-
loid cells from the mating mixture using the complementary nutritional
markers of the parents.
The diploid cells formed can grow indefinitely by mitosis and, at
any point, be induced to undergo meiosis and spore formation by altera-
tion of the culture conditions. Specifically, acetate is used as the carbon
source (usually 1% potassium acetate), and little or no dextrose (0.05%
or less) or nitrogen source (0.1% yeast extract or less) is included.
Meiosis and sporulation are inhibited by the usual levels of dextrose or
nitrogen source, and these processes, which generate diverse offspring
in place of the uniform offspring resulting from mitotic reproduction,
can be viewed as a response to adversity. The whole process takes 12-24
hr in liquid, but 24-48 hr on plates, and each diploid nucleus undergoes
a typical meiosis producing four haploid nuclei, each of which is encap-
sulated in a spore coat. These four spores remain within what was the
diploid cell's wall. That wall is called the ascus(= sac), and the spores are
called ascospores. For this reason, S. cerevisiae is a member of the As-
comycetes. The four spores, or the four haploid "spore" clones they pro-
duce on germination, are called a tetrad.
The nutritional control of the mitosis-meiosis decision is mediated
by the adenylate cyclase-cAMP-dependent protein kinase system (Mat-
sumoto et al., 1982, 1983a,b; Uno et al., 1982; reviewed by Matsumoto et
al., 1985). Specifically, cyclic AMP is needed to begin the cell cycle and
acts at a point in GI even before the point of action of a-hormone. On
the other hand, cyclic AMP must be absent in order for cells to enter
meiOSIS.
This mitosis-meiosis decision for yeast is quite similar to a growth-
differentiation decision in higher eukaryotes. Perhaps it should thus not
be surprising that yeast has proteins homologous to the mammalian
"Ras" oncogenes whose overproduction or malfunction in mammalian
cells results in cancer (Gallwitz et al., 1983; DeFeo-Jones et al., 1983).
These yeast RAS proteins are involved in controlling cAMP production
104 R. B. WICKNER
and thus in modulating the growfh-differentiation decision in yeast

(Toda et al., 1985; Tatchell et al., 1985). They are GTP-dependent stim-
ulators of adenylate cyclase (Uno et al., personal communication; Toda et
al., 1985).
The four ascospores from each diploid cell are routinely dissected
free from each other using a needle attached to a micromanipulator while
viewing the process under the microscope. As will be described in detail in
Section 3, the information available by being able to analyze all four
products of each single meiotic event (tetrad analysis) is sufficiently valu-
able to justify the effort involved in mastering the technique of tetrad
dissection. This process is carried out under a magnification of around
250 x. In order to have plenty of room between the objective and the
surface of the agar slab on which the dissection is done (Fig. 2), it is
desirable to use a lOX or 15X objective with a 25X or 20x or 15X pair of
eyepieces. The stage has to be moved around a lot, and it is useful to be
able to return to a place where you were previously dissecting. This can be
done without taking your eyes away from the eyepieces if the stage has
handles such that one turn = 1 mm. This is the case for the stage designed
by Fogel and Mortimer, available from Lawrence Precision Machine in
Hayward, California. A micromanipulator attached directly to the stage
of the microscope reduces the effects of nearby vibrations and reduces the
likelihood of accidentally moving the needle out of the field. A large
variety of micromanipulators are available, and price, ease of repair,
frequency of need of repair, and ease of learning are all worth consider-
ing. Those for which the needle is controlled in two or three dimensions
by the motions of a single lever are probably the easiest on which to learn,
Vertical Adjust
Microscope Dissecting Aglir Slab on Underside

Stage Needle of Glass Slide.
Figure 2. A dissecting apparatus. Schematic cross-section through a dissecting apparatus

of the Mortimer-Fogel type. The sporulated culture is streaked on a thin agar slab, which
rests on a large glass slide. The slide with agar slab is inverted and placed on a rectangular
plexiglass holder which is open above (for the agar slab) and to the left (to admit the
dissecting needle). The needle's down motion is controlled by the apparatus to the left
attached to the stage of the microscope. The other two dimensions are controlled by
moving the plexiglass slide holder.
but are significantly more expensive and somewhat more complex to

repair.
The four spores of each tetrad are dissected apart and placed in a
defined pattern on a thin slab of agar or agar containing rich medium
(YPAD). This can be done (1) directly on a Petri dish inverted over the
dissecting needle (Fogel et al., 1981) or (2) on a slab of agar on a glass
plate. In the latter case, the slab must then be transferred to a YPAD
plate for germination and growth. The spore clones can be then [case
(1)] directly replicated to various media for testing or [(1) or (2)] picked
and inoculated onto a YPAD master plate for growing and testing. The
author uses a master plate and places the colonies in an array made by
gently pressing the surface of the agar with a 48-point inoculator [avail-
able from West Coast Scientific, Inc., Emeryville, CA (catalog No.
WC48)], suitable for inoculating half the wells of a 96-well microtiter
plate (e.g., Falcon catalog No. 3040). This facilitates multiple matings
done in analyzing the tetrads, for example, mating-type and comple-
mentation tests. The 48-prong inoculator is dipped into an H 2 0 suspen-
sion of the strain(s) used in the mating, and then the adhering cells are
simultaneously delivered to each of the 48 spots on a fresh YPAD replica
of the master plate.
All the markers in the cross are generally analyzed (you'll be sorry
later if you don't) by replica-plating to suitable media and by doing
complementation tests. Mating-type testing is done by mating YPAD
replicas of the master plate separately with known auxotrophic a and a
tester strains. The formation of prototrophs able to grow on SO (syn-
thetic dextrose) plates shows that mating occurred. Simple methods for
carrying out large numbers of complementation tests have been de-
scribed (e.g., Wickner, 1979). For example, a mixture of a and a hisl adel
lysl met4 strains can be mated with the spore clones. This one plate can
then be replica-plated to -his, -ade, -lys, and -met plates to score these
four markers.
After all the spore clones have been suitably analyzed, these strains
can become parents for further crosses.
3. TETRAD ANALYSIS
If the haploid parents of a cross differ in some trait determined by a

single chromosomal gene, the diploids formed by mating will be hetero-
zygous at that locus. The mechanics of meiosis (Fig. 3A, 3B, and 3C)
then require that two of the four nuclei formed will have the genotype of
one parent and two will have the genotype of the other parent. This is
called 2:2 segregation (or 2+: 2-, etc.). Deviations from 2:2 segrega-
tion can be due to polygenic control of the trait examined or to gene
106 R. B. WICKNER
A
Haploid
X Parents
n=2
Centromere l Mating
Diploid
! DNA Replication
Figure 3. The mechanics of meiosis. (A) Haploid (n = 2) cells of opposite mating-type

mate. One parent has marker A on chromosome I and markers B and Con chromosome
II. Markers A and B are close to their respective centromeres. The diploid is induced to
undergo meiosis and begins by replicating its DNA. (B) Homologous chromosomes pair
and recombination occurs. (Following recombination, if cells are returned to growth medi-
um, recombinant diploid cells can be detected that have not completed sporulation.)
Meiosis proceeds. The first meiotic division is shown here. Homologous chromosomes
separate. (C) The second meiotic division then occurs without intervening DNA replication
so that haploid progeny are produced. A single recombination between markers B and C
has produced a tetratype ascus with respect to these two genes. A and B are nonparental
ditype with respect to each other because of their chromosomes' assortment at division I.
Note that a crossover centromere distal to marker A did not alter the distribution of
marker A in the haploid spores.
conversion, mitotic recombination during growth of the diploid, cyto-

plasmic inheritance, or aneuploidy (see below).
The centromere is the site on a chromosome that is attached to the
mitotic or meiotic apparatus and by which a chromosome is pulled to
one or the other daughter nucleus. Thus, it is essential to remember that
the factor determining the segregation of a marker into the daughter
nuclei is the co!lnection of the copies of that marker to their respective
B MEIOSIS I
Homologous Chromosomes
pair and recombine
A +C
A ++
+ B+
+ BC
Homologues remained
paired in metaphase I
Homologous centromeres
separate
Figure 3. (continued)
centromeres. Recombinations occurring between a marker and its cen-

tromere can change that connection, but recombinations occurring cen-
tromere-distal to the marker do not affect that connection.
If the parents in a cross differ in two single chromosomal genes
(e.g., A B X + +) and if these two genes are completely unlinked to each
other or to their respective centromeres, then the distributions of mu-
tant and wild-type alleles of the two genes are random with respect to
each other. All the possible relative segregation patterns are as follows:
AB AB AB A+ A+ A+
AB A+ A+ AB AB A+
++ +B ++ +B ++ +B
++ ++ +B ++ +B +B
PD T T T T NPD
108 R. B. WICKNER
c MEIOSIS II
A+C
A++
+B+
+BC
The first of these tetrad types has only two types of spore clones,
namely, A B and + +. It is thus called a ditype tetrad. Since these two
spore types have the same genotypes as the parent strains with respect to
these two markers, this is called a parental ditype (PD) tetrad. The last
tetrad also has only two types of spore clone genotypes with respect to
markers A and B, namely, A+ and +B. Since both of these are recombi-
nant genotypes when compared to the parents, such a tetrad is called a
non parental ditype (NPD) tetrad with respect to these two markers. Each
of the other tetrads has four genotypes, two parental (A B and + +) and
two recombinant (A+ and +B), and these are thus called tetratype (T)
tetrads. As can be seen, the ratio of PD : NPD : T tetrads is 1 : 1 : 4 in this
case. As will be shown, deviations from this ratio indicate either (1) linkage
of A and B, in which case PD > NPD, or (2) linkage of both A and B to
their respective centromere, in which case PD : NPD : T :: 1 : 1 : < 4.
If the two markers B and C are very closely linked to each other,
most of the tetrads will be PD. As can be seen in Fig. 3, a single crossover
between B and C changes the PD into a T ascus. The crossover shown is
between strands (chromatids) 2 and 3, but the reader can satisfy himself
that the same result would be found for a single crossover between any
pair of nonsister chromatids (1 and 3, 2 and 3, 1 and 4, or 2 and 4).
For a second recombination event between B and C, the result de-
pends on which pair of chromatids are involved relative to the first
crossover. In Fig. 4C, the second exchange involves the pair of chro-
matids not involved in the first. Only this kind of double crossover (four-
strand double crossover) results in an NPD tetrad. (Conversely, observ-
ing an NPD tetrad implies that at least two crossovers have occurred
between B and C, assuming they are on the same chromosome; see
~~ -
8C
2&3 8+ Tetratype 3-Strand Double
1&3 +C Crossover
++
8 B c
~ -
8C
2&3 8C Parental ; 2-Strand Double
2&3 ++ Ditype Crossover
++
c B c
~ -
8+
2&3 8+ Non parental 4-Strand Double
1&4 +C Ditype Crossover
+C
D B c
~ -
8C
2&3 8+ Tetratype 3-Strand Double
2&4 +C Crossover
++
Figure 4. Double crossovers. Given that one crossover in a particular interval involves a
particular pair of strands (2 and 3 in this diagram), the second crossover in that interval can
involve the same pair of strands (result: PD tetrad), the two strands not involved in the first
crossover (result: NPD tetrad), or (for the other two possibilities) one strand involved in the
first crossover and one not involved (results: T tetrads). Thus, one double crossover in four
would be expected to result in an NPD tetrade.
110 R. B. WICKNER
below.) The other possible second exchanges are 1 and 3 or 2 and 4,

producing tetratype tetrads (three-strand double crossover), and 2 and
3, reversing the effect of the first crossover producing a parental ditype
(two-strand double crossover). To arrive at a simple formula for genetic
distances, one assumes that no more than two crossovers occur in a
particular (short) interval and uses the facts that (1) the NPD tetrads are
one-fourth of the total double crossovers and (2) among the T tetrads
are half of the double crossover tetrads, a number equal to twice the
NPD tetrads.
Thus, in the interval between B and C, the total crossovers =
1 x single crossovers + 2 X double crossovers =

1 x (T- 2 NPD) + 2 X (4 NPD) = T + 6 NPD
Since one crossover produces only half recombinant and half paren-
tal spores, the percent recombination (i.e., the percent of progeny that
are recombinant in the limiting case of very short distances) = 100 x
total crossovers/2 (total tetrads) = 50 (T + 6 NPD)/(PD + NPD + T)
(Perkins, 1949). The units are centiMorgans (eM).
This formula was derived assuming that only one or two crossovers
occur in the B-C interval. It is thus increasingly inaccurate with increas-
ing genetic distance. More accurate formulae have been published
(Snow, 1979), as has a curve relating eM to linearly additive map units
(Mortimer and Schild, 1980; Ma and Mortimer, 1983).
If two markers are on different chromosomes, but each is quite
close to its respective centromere, the two markers will be found to
produce fewer tetratype asci than expected. Each marker will be seen to
segregate from its wild-type allele at the first meiotic division. This is
called first-division segregation and is illustrated in Fig. 5. Note that for
the two centromere-linked genes A and Bin Fig. 5, the crossover (which
will only rarely come between A and its centromere or between B and its
centromere) has no effect on their segregation relative to each other.
There are thus only two ways A and B are likely to segregate, depending
only on which centromeres go to which daughter cell at the first meiotic
division, namely, PD and NPD (Fig. 5). If A, for example, is known to be
essentially at its centromere, then B's distance from its centromere can
. T X 100
be calculated simply by the formula eM = 2 (total tetrads) . Centro-
mere linkage can, by definition, only be detected in the range where
single crossovers (tetratype tetrads) are relatively rare, i.e., up to about
30 eM from the centromere. Since direct linkage of two markers on the
same chromosome is detected by a deficiency of double crossovers (NPD
tetrads), it is detectable up to a separation of about 60 eM.
®·®~
1 Mating
Diploid Zygote
®a- 1 Pairing
DNA Replication,
Homologous
of
l Crossing Over
/.llllt ® ~
®© ~(f}©
Meiotic DMslon
+
+ ~
+
'
I 2nd Meiotic
t DMsion
Figure 5. Centromere linkage. The events here are the same as in Fig. 3. Markers A and B,
unlike marker C, are unaffected by the crossing-over stage because they are very dose to
their respective centromeres. The segregation of A and B relative to each other depends
only on whether the A's and B's go to the same pole in the first meiotic division (on the left)
to produce a PD tetrad with respect to A and B, or the A's go with the B+'s and the B's go
with the A +'s (on the right) to produce an NPD tetrad. Tetratype tetrads for A and B can
occur only if, as shown here for marker C, there is at least one crossover between A and its
centromere or between B and its centromere.
112 R. B. WICKNER
The reader who wants to become familiar with tetrad ~nalysis

should ( 1) spend some time with pencil and paper working out the
consequences of various crossovers and segregations as shown in the
figures for this section and (2) apply the analysis described here to some
actual data such as that shown in Appendix II.
4. ANEUPLOIDY
An aneuploid strain is one whose chromosome number is not a

multiple of the basic haploid number, n. An otherwise haploid strain car-
rying one extra chromosome III, for example, is said to be "disomic for
chromosome III." An otherwise diploid strain carrying three chromo-
Disomic for
Chromosome II
-
Trisomic for II
Figure 6. An extra chromosome II, as shown here, results in deviation of segregation from
the normal 2+ : 2- pattern for markers on that chromosome as long as two of the
chromosomes have the dominant allele and one has the recessive allele. This phenomenon
is used in certain mapping methods (see text). A single crossover between B and C is shown
here. Centromere-linked genes on chromosome II (like B) segregate 4+ : 0 and 2+ : 2B.
Centromere-distant genes on chromosome II (like C) segregate mostly 3+ : IC with some
4+ : 0 and 2+ : 2C tetrads (see text).
Chr I 2 + :2A Chr I 2 + :2A

Chrll 4+:08 Chr II 2 + :28
3 + :1C 3 + :1C
somes III is said to be "trisomic for chromosome III." An otherwise

diploid strain that lacks one of the pair of chromosomes II, for example,
is said to be "monosomic for chromosome II." ·
Yeast is quite tolerant of extra chromosomes, and a single extra
chromosome may be relatively stably propagated in mitosis. Disomy for
chromosome XI is particularly widespread among laboratory strains.
However, many single extra chromosomes are unstable, and multiple
extra chromosomes generally cause slow growth or no growth. This
selects for loss of some of the extra chromosomes.
We shall examine first the effect of one extra chromosome on
meiotic segregation and later describe mapping methods based on the
use of aneuploid strains.
As shown in Fig. 6, mating a normal haploid with markers A, B, and
114 R. B. WICKNER
C (A B C) with a strain disomic for a chromosome carrying the wild-type

alleles of markers B and C (+ ++I++), results in a trisomic strain of
genotype A/+ BC/ ++I++. The marker, A, on the chromosome for
which an extra copy is not present segregates 2 A : 2 + as for a normal
cross. However, the extra copies of chromosome II carrying the wild-
type alleles of B and C mask the mutant alleles in many of the spore
clones. For marker B, which is close to its centromere, a mixture of 4
+ : 0 (two-thirds of all tetrads) and 2 + : 2 B (one-third) is observed. This
ratio is expected since one of the three chromosomes II is presumably
chosen at random to segregate by itself to one of the daughter cells at the
first meiotic division. That will be the chromosome carrying B (there-
cessive mutation) one-third of the time and will produce a 2 + : 2 B
tetrad. The other outcomes will produce 4 + : 0 tetrads. For marker C,
which is assumed to be far from its centromere, a mixture of 2 + : 2 C, 3
+: 1 C, and 4 +: 0 tetrads will be produced in the ratio of 1: 10:4 (see
Table 1). A detailed discussion of these ratios and other quantitative
Table I. Theoretical Distribution of Tetrad and Spore Types from -/+I+
Tetrad types
Frequency 2/3 1/3 Spore

Phenotype 4+:0 2+:2- types Frequency
Tightly cen- trpll+ trpl += 113

tromere- trpll+ trpl +!+ = 1/6
linked + +I+ trpl = 116
gene a + +I+ trpll+ = 113
Frequency 4/15 1115 8/15 2/15 Spore

Phenotype 4+:0 2+:2- 3+:1- 3+:1- types Frequency
Centromere- mall+ rnal mal mal/mal += 10/30

unlinked mall+ mal rnall+ +I+ +I+ = 6/30
gene 6 + +I+ +I+ + rnal = 5/30
+ +I+ + + mall+ = 8/30
mal/mal = 1130
•The segregation of trpl is independent of recombination events because of its tight centromere link-
age. At the first meiotic division, one of the three homologs will migrate alone to one pole. If this is the
one carrying the trpi mutation, a 2 trpl : 2 +I+ tetrad will result. If it is one of the two chromosomes
carrying trp+, a 2 trpll + : 2 + tetrad will result. Therefore, the ratio of the former to the latter tetrad
type is 1 : 2, as shown. .
•we assume that rnal is so far from its centromere that many crossovers occur between it and its
centromere and the distribution of the two rnal mutant alleles among the six chromatids is random.
There are 15 possible distinct arrangements. If they are equally likely, the indicated distribution
results. For intermediate distances, the calculations of Shaffer et al. (1971) and Riley and Manney
( 1978) predict intermediate frequencies of the -/- segregants. Our calculation assumes completely
trivalent pairing. Any bivalent-univalent pairing will decrease the overall frequency of -/-
segregants.
aspects has been published (Shaffer et al., 1971; Culbertson and Henry,
1973; Riley and Manney, 1978).
This phenomenon is the basis for certain mapping methods (see
Section 6.3).
5. MUTANT INDUCTION AND ISOLATION
5.1. Mutagenesis
Saccharomyces cerevisiae can be mutagenized by any of the agents
useful in bacteria, e.g., ethylmethanesulfonic acid (EMS) (Lindegren et
al., 1965), nitrosoguanidine, UV or X irradiation, and frameshift muta-
gens (reviewed by Mortimer and Manney, 1971), or using in vitro meth-
ods to specifically modify a gene of interest (see Chapter 5). One must
generally strike a balance between cell killing and mutation induction.
Usually at very high percent killing, the yield of desired mutants actually
falls as a percent of survivors. We usually use 4% EMS in 0.1 M po-
tassium phosphate, pH 7.6, at 20°C for 2 hr. This gives about 50% killing
with our strains, but strain variation is to be expected.
What is to be done with the mutagenized cells depends on what kind
of mutation is sought. If it is necessary to know that one's mutations are
independent, then the mutagenized cells must be divided before growth,
either by plating out or by growing cells in many separate tubes if muta-
tion expression prior to exposure to selection conditions is necessary.
5.2. Mutant Isolation

Selection for nongrowth under some conditions can be made using
nystatin (Snow, 1966) or inositol-less death (Henry et al., 1975) in a man-
ner quite like penicillin selection for Escherichia coli. This can be used for
selecting auxotrophs or other mutants whose growth arrest is not lethal.
Often temperature-sensitive or cold-sensitive mutants in essential mac-
romolecular processes are isolated by plating mutagenized cells on rich
medium at the permissive temperature, replica-plating to a plate at the
nonpermissive temperature, and then subscreening clones that did not
grow for some more specific trait. Examples include (1) cell cycle mutants
isolated by subscreening for a uniform bud size of initially asynchronous
cells whose growth was arrested by shifting to the nonpermissive tem-
perature (reviewed by Pringle and Hartwell, 1981) and (2) protein, RNA,
or DNA synthesis mutants screened by rapid measurements of differen-
tial arrest of macromolecular synthesis at the nonpermissive tempera-
ture.
Selection of new mutants as revertants of other mutations is a stan-
dard method for studying the interactions of genes or their products.
116 R. B. WICKNER
Revertants can be intragenic (including true reversions to wild type) or

extragenic. Any mutation that reverses a mutant phenotype without
eliminating the original mutation is called a suppressor. Extragenic sup-
pressors can be informational suppressors, such as amber, ocher, or
missense tRNA suppressors, or functional suppressors. Informational
suppressors are generally allele-specific, suppressing only certain alleles
of a particular gene, and locus nonspecific, suppressing some mutant
alleles of each of many different genes. Functional suppressors may
suppress all mutant alleles of a given locus by supplying an alternate
means of accomplishing what the original gene normally does. In other
cases, a functional suppressor may be a mutation in a gene whose prod-
uct normally interacts with the original mutant protein and reverses the
original mutant phenotype by stabilizing that protein.
Methods of mutant isolation are extremely diverse and depend too
much on the type of mutant to be isolated, to be listed in detail here.
Suffice it to say that essentially all methods used with any microorganism
can be used with S. cerevisiae.
6. GENETIC MAPPING METHODS
Genetic mapping is a useful way of characterizing a mutant or set of

mutants. Its primary value is in identifying the mutated gene by a meth-
od independent of the phenotype and the actual possession of the mu-
tant strain itself. It demonstrates that the gene being studied is distinct
from all genes previously mapped to other locations-an otherwise im-
practical task. As the S. cerevisiae map is becoming denser, mapping, with
increasing frequency, leads to the finding of unexpected identify of the
newly mapped gene and a gene previously identified and mapped by
another phenotype. This is commonplace in E. coli work. Mapping can
facilitate later efforts to clone a gene if it is found near an already cloned
gene. Since yeast chromosomes have been separated by gel electropho-
resis, we can also expect to have chromosome-specific banks in the near
future.
6.1. Meiotic Mapping

The primary mapping method is simply meiotic mapping by tetrad
analysis. Experience with all other methods has led to pitfalls in even the
most expert hands. Thus, while other methods generally lead more
quickly to the assignment of a chromosome, only meiotic mapping is
reliable for the final conclusion. In view of this, some workers have
constructed sets of strains with large numbers of markers scattered at
intervals throughout the genome. These are then simply crossed to the
new mutants, tetrads are dissected and analyzed, and the results are
examined for meiotic linkage (Contopoulou and Mortimer, cited in Mor-
timer and Schild, 1981; Gaber et al., 1983). The only limitation of this
method is that not quite all of the yeast genome has been sufficiently
mapped with standard markers to assure that a location will be found in
this way, but experience shows that for over 80% of mutants one is
successful.
Gene order can be derived assuming that distances between genes
are additive. If the A-B distance is 10 eM, the A-C distance is 6 eM, and
the C-B distance is 4 eM, the gene order should be ACB. As a check, one
can analyze individual tetrads. For short distances like this, double
crossovers should be rare. For most tetrads, at least two of the markers
A, B, and C should be in the PD (parental ditype) configuration. The
third marker may be tetratype with respect to the other two if a single
crossover has occurred. A crossover in the A-C interval in the example
cited would produce tetrads of the following type:
A c B
+ c B
A + +
+ + +
A crossover between C and B would yield:
A c B
A c +
+ + B
+ + +
But a tetrad of the following type
A c B
A + B
+ c +
+ + +
would be very rare relative to the other two types because it requires two
crossovers (one between A and C, one between C and B) if ACB is the
correct gene order. Table II shows the rare tetrad type for each of the
three possible gene orders of the markers in a cross of the type A B C X
+ + +.
118 R. B. WICKNER
Table II. Ordering Genes

by Examination of Individual Tetrads
Parents: A B C x + + +
Gene order Rare tetrad type
ABC ABC
A+C
+B+
+++
ACB ACB ABC
A+B AB+
+C+ ++C
+++ +++
CAB CAB ABC
C+B +BC
+A+ A++
+++ +++
6.2. Mitotic Recombination

Mitotic recombination has been used similarly to survey chromo-
some arms for the location of a gene (e.g., Mortimer and Hawthorne,
1973). Here, one constructs diploids of the type a-b-c-x-1+ + + +,
where a-, b-, and c- are standard markers and x- is the unmapped
gene. Mitotic recombination is induced by UV or other mutagen treat-
ment.
A correlation between homozygosis of one of the standard markers
and of the unmapped markers means that they are located on the same
chromosome arm. Thus, one isolates a- and b- and c- clones and tests
whether one of these classes has become x-. If, for example, most b-
clones are x-, then x is on the some chromosome arm as b. To under-
stand this, one must look at the details of the consequences of mitotic
recombination.
A recombination between nonsister chromatids (Fig. 7) produces two
possible outcomes depending on how the product nonsister chromatids
segregate into the daughter cells. In 50% of cases the two daughter cells
have each become homozygous, one for certain alleles on one homolog (C
in the figure) and one for those on the other homolog ( + ). It is important
to keep in mind here, as in meiotic recombination, that only the configu-
ration of markers centromere-distal to the site of the recombination event
is changed. Thus, marker B does not become homozygous in Fig. 7
because it is centromere-proximal to the crossover site. The other 50% of
cases produces one daughter cell that is unchanged from the parent
A
+
A
+
Daughter diploid cells

homozygous at C locus
Figure 7. Mitotic recombination can result in homozygosis of genes centromere-distal to
the site of the recombination event.
diploid cell and one that is still heterozygous for all markers, but with the
linkage relationships reversed at the site of the recombination.
The use of mitotic recombination as a mapping method relies on the
fact that, in Fig. 7, for example, marker C signals the homozygosis of all
markers distal to marker C. Mitotic gene conversion is an important
exception, so that a number of mitotic recombinants for each marker
must be examined.
6.3. Aneuploidy Methods

A general method used to locate many of the genes now on the S.
cerevisiae map relies on the effects of extra chromosomes on the pattern
of segregation. As can be seen in Fig. 6, if a normal haploid carrying
markers on various chromosomes is crossed with a strain carrying an
extra chromosome, markers on the chromosome for which an extra
copy is present will not always segregate 2 + : 2-. Instead, the tetrads will
120 R. B. WICKNER
be a mixture of 2+: 2-, 3+: 1-, and 4+: 0 ("aneuploid segregation").

The proportions of these tetrad types vary according to the distance
from its centromere of the marker in the +I+ I- configuration. For a
marker very close to its centromere, two-thirds of the tetrads are 4+: 0
and one-third are 2+: 2-. For a marker far from its centromere, the
majority of tetrads are 3 + : 1-, and a smaller number of 2 + : 2-, 4 + : 0
and even a few 1+ : 3- tetrads will be observed. In contrast, markers on
other chromosomes will be segregating 2 + : 2-.
Use of this phenomenon for mapping new genes relies on the as-
sumption that the extra chromosome(s) is an entire chromosome and
not just one arm or other fragment of a chromosome. This is almost
always true, but because exceptions are known, a chromosome assign-
ment must always be checked by testing for meiotic linkage.
One approach used is to have, along with the unmapped marker,
many standard markers (Mortimer and Hawthorne, 1973). Strains of
ploidy varying around n + nl2 can be generated by sporulating a triploid
strain. Most of the spore clones from a triploid are inviable, but the few
survivors are multiply aneuploid and can be mated with a series of
strains each carrying the unmapped and several standard markers. If, in
a particular cross, the unmapped marker segregates 2+: 2-, then it is
eliminated (provisionally) from each chromosome whose standard
marker showed "aneuploid segregation" (4+ :0, 3+: 1-, and 2+ :2-)
in that cross. Contrariwise, if the unmapped marker showed aneuploid
segregation in the cross, then it is eliminated from each chromosome
whose standard marker showed 2+: 2- segregation.
A variation of this method incorporates the standard markers into
the triploid strain, eliminating the need to construct a lot of strains
(Wickner, 1979). Another variation uses a set of fairly stable multiply
aneuploid strains whose extra chromosomes are known (Hilger et al.,
1982). Another variation uses complementing alleles at the locus to be
mapped to specifically select a strain that is aneuploid only for the chro-
mosome on which the locus to be mapped resides (Hilger and Mortimer,
1980).
6.4. The spoil Mapping Method (Klapholz and Esposito, 1982)

The spoll gene is necessary for meiotic recombination. Diploids
that are homozygous for a defect in this gene undergo a defective meio-
sis, most of whose products are inviable. Most of those which are viable
are haploids but contain an assortment of parental unrecombined chromo-
somes. This phenomenon can be used to map a gene. A spoll canl (or
cyh2) strain carrying the mutation to be mapped is mated with a series of
spoll CAN+ CYH+ tester strains carrying markers on known chromo-
somes, and the diploids are sporulated. The rare viable products of
meiosis are selected using the canl (recessive resistance to canavanine) or

cyh2 (recessive resistance to cycloheximide) marker. These are then
scored for the unmapped marker and the standard marker. In this type
of cross, because of the spoll defect, two markers anywhere on the same
chromosome show nearly perfect linkage to each other even if they
would be unlinked in a normal cross. Analysis of about 200 spore clones
usually suffices to locate on which chromosome a marker is located. A
clone of a gene or other chromosomal DNA sequence can likewise be
located by putting it in a URAJ-carrying integrating vector (e.g., Ylp5),
integrating it by homology with the insert and then mapping the URA3
gene on the vector. The spoll strains all carry a nonreverting allele of
ura3 for this purpose. Note that it must first be checked that the plasmid
integrated by its homology with the unmapped gene. The most com-
monly used ura3 hosts are not deleted for URAJ, and integration can
also occur there.
6.5. Chromosome-Loss Methods

The loss of chromosomes from a diploid strain can be induced by
growth of the strain in the presence of methyl benzimidazole-2-y !-carba-
mate (MCB) (Wood, 1982). Likewise, treating a rad52/rad52 diploid strain
with x-rays or methylmethanesulfonate (MMS) induces loss of chromo-
somes (Mortimer et al., 1981 ). Diploid strains homozygous for chll (chro-
mosome loss) undergo spontaneous, nonrandom chromosome loss that
could be used for mapping, but there are some chromosomes which are
virtually never lost, limiting the usefulness of this method (Liras et al.,
1978). Diploids homozygous for the temperature-sensitive cell division
cycle mutants cdc6 or cdc14 undergo frequent loss of chromosomes when
maintained for several hours at the nonpermissive temperature (Ka-
wasaki, 1979). All chromosomes tested are lost with some frequency, and
this method has been used to map several genes. The logic here is the
same as for the spoll method, except that the degree of haploidy achieved
is not usually complete. If, for example, the recessive allele of the unmap-
ped marker is on the same chromosome as a particular standard marker,
these two markers will show virtually complete linkage.
These chromosome-loss methods and the spoll method represent a
substantial advance in mapping methods. They should make genetic
mapping a routine part of the characterization of any new mutant.
6.6. 2-pm DNA Mapping (Falco and Botstein, 1983)

If an integration vector carries a part of the 2-j.l.m DNA plasmid that
does not include its origin of replication, it remains unable to replicate
unless it integrates into the chromosomal DNA. Such a vector can be
122 R. B. WICKNER
used to map a cloned sequence. The vector integrates at the chromosom-

al site homologous to the cloned sequence. For some as yet unknown
reason, this makes the chromosomal arm carrying the integrated seg-
ment of 2-~J.m DNA unstable. When the haploid strain carrying the
2-~J.m DNA integrated at the site of the unmapped cloned sequence is
mated with a normal multiply marked haploid strain, the 2-~J.m DNA-
carrying chromosome arm is lost at rather high frequency. This can
occur by mitotic recombination or by chromosome loss. This can be used
to map the cloned sequence even if one does not have a mutant for the
gene cloned. Of course, as described earlier, the spoll method can also
be used for this purpose.
Loss of the entire chromosome represents only a minority of these
events, but this minority class of events makes it possible to map a muta-
tion located on the opposite side of the centromere from the standard
marker for that chromosome.
6. 7. Overall Mapping Strategies

All mutants to be mapped should first be tested for centromere
linkage. For those which are not centromere-linked, two overall strate-
gies are feasible. As the yeast map is becoming more well defined, with
fewer uncharted areas, the brute force meiotic mapping approach with
multiply marked strains (see Section 6.1) is increasingly attractive· for a
typical mutant. For mutations selected as suppressors of other muta-
tions, this is more difficult because it involves reconstructing many
strains to introduce the suppressed mutant. Of course, any method used
will involve some of this.
The other alternative for initial mapping is to use one of the meth-
ods described earlier to locate on which chromosome the mutant !.i
located.
For mapping cloned fragments, the preceding method (Gaber et al. 's
strains are allleu2- ), the spoll method (Klapholz and Esposito's strains
are all ura3- ), and the 2-~J.m mapping method (Falco and Botstein,
1983) have all been used with success.
6.8. The Genetic Map

Genetic linkage studies have defined 17 chromosomes for S. cere-
visiae (Fig. 8; reviewed by Mortimer and Schild, 1985). Of these, chromo-
somes I-XVI each have many genes and have even been demonstrated
biochemically as distinct DNA molecules (Carle and Olson, 1985). Chro-
mosome XVII is defined by a single tightly centromere-linked gene,
KRBJ, whose unusual genetic properties suggest the possibility of an
unusual structure (Wickner and Leibowitz, 1977; Wickner et al., 1983).
The classes of mutants available include those affecting nearly all

aspects of the yeast life cycle and metabolism (see examples in Table III).
The primary source for standard markers for mapping purposes is the
Yeast Genetic Stock Center (YGSC), Department of Biophysics and Med-
ical Physics, University of California, Berkeley, California 94 720, super-
vised by R. K. Mortimer, whose work has established most of the map.
While many specialized mutant types are available from YGSC, some
others must be obtained from individual investigators. YGSC also carries
special strains for mapping, as hosts for transformation, and as tools for
special types of genetic study.
7. NON-MENDELIAN GENETICS
Most phenotypic differences between yeast strains are determined by

chromosomal genes. If two haploid strains, A and B, differ in some trait
that is determined by a single chromosomal gene difference, then mating
these two strains and inducing meiosis in the diploid strain formed
generally produces two spore clones with the phenotype of one parent
and two with the phenotype of the other parent, that is, 2 : 2 segregation
.In . .
meiOSIS.
A nonchromosomal gene, however, if present in high copy number
in one parent and absent from the other parent, will be present in high
copy number in the diploid cells formed by mating the two parents, and
all four of the spores formed by meiosis will get one or more copies of
this gene. Thus, instead of the 2+: 2- segregation seen with the chro-
mosomally determined phenotypic difference, a phenotypic difference
between the parents determined by a nonchromosomal gene generally
produces 4+ : 0 segregation in meiosis, with all of the spore clones show-
ing the phenotype determined by possession of the nonchromosomal
gene. Often even a low copy number plasmid will be inherited by all the
offspring because of a specific segregation mechanism. On the other
hand, some high-copy-number plasmids can be unstably inherited be-
cause of the absence of a proper segregation mechanism. This is the case
for many 2-f.Lm DNA-based and ARS-containing cloning vectors (Kiku-
chi, 1983).
A second method to distinguish chromosomal from nonchromosom-
al genes is mitotic segregation. The diploid strain formed between A and
B, if grown mitotically, will generally not show segregation of the parental
phenotypic difference if that difference is chromosomally determined.
Both chromosomal alleles are transmitted to both daughter cells at each
generation, and the dominant allele will control the phenotype. If A has a
nonchromosomal genome and Blacks that genome completely, the A X B
diploids will all have the phenotype of strain A. But if A and B both have
II m: l[ Jill :m
ptl3
cdc24
cyc3
pyll,cdcl9
sue~
t-·~
mok t6
HO
moll7
c-rsl SUFZ~
opo7 !IPQit
1111
odo l
cdc I~
.... dc9
ole7
cdc36
fLO I
pho2
cdc2
,,.,ozzs
SUP22
N1~
lr'lll
SUPI7
ftol
""'0139
SUP76
fro2
hto6
SUP77 bot l, llf1
SUF4 cdc-29
lrfj'4
SUF3 ....,
do l81
doll
odo8 ol4
pur~
,,,
do:J
c:upl4
rnol
- - - -Mitot ic linllo9e
f. .
I
·······Trisomic lin.oge
( l Sequence not utoblilhtd
with qenu outs ide por,.nthuis ,3
ltfl orm is on top
Figure 8. The genetic map of S. cerevisiae (reviewed by Mortimer and Schild, 1985).
r·
X :m :lilt = m
fmok l7 tmok22 or98
I
I
I SUFI
cdc6 l'f'I01cl2
pep4
toot
I
uro2 II
olc4
orq3
pita
suff4
mall26
RON I 01014
c~cl
····~r
tom4512
SUP4 SUP26
cdc8 SUP86
cdc II '""8740
cup3 rod~6
odt2
cor2 SU F~
live
SUFIO
opol
potS
,no2 to3(~1o8)
XI :III! SUP19 ::llZII

SUF22
sun
lyo9 +KRBI
rod 52
dc5
fr
F6
r.,
tmOIII
sec-59 UP 50
tsm0800
•"
d•2
lyt7
orq80 FB
SUP~
prt I
r.
llv2
o ro 8 1
lett f.~
F ll
I'I
cdcl6
tnOir.ll
met14 I
I
r
I
do lBO
spo l4
mut2
SUP18
mole 27
""'20
mtl l rnol
F12
SUP25
t""
mol15
lfltt6
lo1AL4
126 R. B. WICKNER
Table III. Examples of Types of Yeast Mutants Availablea
Auxotrophs-adenine, arginine, his- Chromosomal mutants affecting

tidine, leucine, tryptophan, tyrosine, mitochondrial function (petite)
phenylalanine, threonine, homoserine, Acid and alkaline phosphatase structural
aspartate, isoleucine-valine, lysine, and regulatory mutants
methionine, uracil, inositol, thiamine, Recombinant-deficient and chemical
cysteine, fatty acids, dTMP mutagen-sensitive
Fermentation-sucrose, maltose, Radiation-sensitive cAMP metabolism
melibiose, a-methylglucoside, galac- Sporulation-deficient
tose, glycolytic pathway Steroid synthesis
General control of amino acid Secretion-defective
biosynthesis Heme synthesis
Amber and ocher suppressible alleles Nuclear fusion-defective
and suppressors Nucleotide synthesis
Nitrogen assimilation Sterile (nonmating)
Mannan mutants Mating-type interconversion system
"Killer" (dsRNA replication and expres- Protein synthesis
sion) mutants (nuclear and Cell division cycle
cytoplasmic) Cell lysis, osmotic-sensitive RNA
Permeability mutants synthesis
Drug resistance (nuclear and Flocculation, smooth colony
mitochrondrial) Low temperature-sensitive
Proteinase-deficient mutants Glycogen and glucosamine accumulation
Mitochrondrial genome mutants
•Many of these strains of available from the YGSC. Others must be obtained from the individual
investigators. YGSC publishes a useful catalog of strains that includes helpful hints, a map, references,
and other information.
the cytoplasmic genome, and that genome is differently marked in A than

in B, then, on mitotic growth of the A x B diploids, eventually some cells
will get only the genome from A while others will get only that from B,
and the two phenotypes will be seen to segregate in mitosis. This phe-
nomenon is due to the absence of a regular segregation mechanism and to
the statistical fluctuations in the copies of the genome from A vs. the
copies from B. Although mitotic recombination does produce segrega-
tion (homozygosis) of chromosomal heteroalleles, this process is rare and
produces only a fraction of a percent of homozygous progeny in the
population. Mitotic segregation of nonchromosomal genes generally re-
sults in virtually the entire population becoming homozygous. The native
2-~-tm DNA does have a regular segregation mechanism, and so differ-
ently marked molecules do not necessarily segregate even after extremely
prolonged growth (Livingston, 1977).
The third method for distinguishing chromosomal from non-
chromosomal genes uses a mutant described by Conde and Fink (1976)
which is defective for nuclear fusion. Normally, when two haploid cells
mate, the cellular fusion is quickly followed by nuclear fusion to form a
diploid cell. But if one of the parent cells carries the karl mutation, its
nuclei are somehow not able to fuse efficiently. The cells proceed to
divide, and since the two nuclei did not fuse, they are separated into
separate daughter cells. These daughter cells thus each have the same
haploid nuclei as the two parents, but since cytoplasmic mixing occurred
at the binucleate (or heterokaryon) stage, a cytoplasmic genome present
in only one of the parent cells is present in both daughter cells (Fig. 9).
The transfer of a phenotype, in a cross of this kind (called "cytoduction"
or a "kar cross") from cells with one nucleus to cells with the other
nucleus, is evidence that the phenotype is determined by a nonchromo-
somal gene.
A fourth method relies on the fact that most nonchromosomal ge-
nomes can be eliminated (or "cured") by one or another growth condition
Heterokaryon
-Nuclei fail to
fuse because
of KAR1
-Cytoplasm mixes
I
Nuclei separate
Grows on glycerol plates (Q•)

containing cycloheximide (cyh2)
Figure 9. The karl-1 mutation results in failure of nuclei to fuse after cell mating. The
cytoplasm of both parents mix at the heterokaryon stage, and the daughter cells, each
carrying one of the parental nuclei, have a mixture of the parental cytoplasms.
128 R. B. WICKNER
or chemical treatment. The mitochondrial DNA (p) is eliminated by

growth in the presence of 50 f..Lg/ml of ethidium bromide (Goldring et al.,
1970). lrA and M double-stranded RNAs (dsRNAs) are eliminated by
growth at elevated temperature (37-39°C) (Wickner, 1974; Sommer and
Wickner, 1982), and M is eliminated by growth in the presence of low
concentrations of cycloheximide (Fink and Styles, 1972). [PSI](see below)
is eliminated by growth in high osmotic strength media (Singh et al., 1979)
or in the presence of 5 mM guanidine hydrochloride (Tuite et al., 1981 ).
The curing method of distinguishing chromosomal from non-
chromosomal genes is limited by the fact that curing methods are initially
hard to find and are not available for all known cytoplasmic elements.
These four methods of distinguishing chromosomal from non-
chromosomal traits are summarized in Table IV.
The molecules and phenotypes showing non-Mendelian (non-
chromosomal) inheritance in S. cerevisiae include mitochondrial DNA
(respiration and resistance to choramphenicol, erythromycin, antimycin,
and oligomycin), various dsRNAs [the killer phenotypes, [HOK], [NEX],
[EXL] (defined in Table VI)], 2-~J.m DNA (see also Chapter 5), [PSI]
(increased ocher suppression by chromosomal ocher suppressors),
[URE3] (ability to use ureidosuccinate to bypass a ura2 mutation in the
presence of ammonia), and 20S RNA (production of 20S RNA on trans-
fer to acetate medium). The genomes responsible for [URE3] and 20S
RNA are not yet known. Recent evidence indicates that [PSI] is carried on
Table IV. Chromosomal vs. Nonchromosomal Traits
Result for:
Single chromosomal Nonchromosomal
Method gene gene Exceptions
Meiotic segregation 2+: 2- 4+: 0 [URE3)

Mitotic segregation + 1- heterozygosity + and - segregateb If"-" means no
stably maintaineda genome, the"+"
is stable
Cytoduction (kar Not transferred< Efficiently transferred 2-~~om DNA is
crosses) transferred with
50% efficiency
Curing Rarely mutated Mutated at close to No method known
100% efficiency for URE3, 20S
RNA, T, W, and
L-(BC)
•Rarely, mitotic recombination results in separation of+ and - alleles.

6 Mitotic segregation requires that the parents each have the same cytoplasmic genome, but that that
genome be marked differently,"+" in one parent and"-" in the other.
<Rarely, chromosoma! markers are transferred.
3-t.J.m DNA, the ribosomal DNA repeat sequence present as a covalent

circle in most strains. It is presumably generated by recombination be-
tween two tandemly repeated copies of the chromosomal ribosomal DNA
repeats. We will discuss some aspects of the mitochondrial and killer
systems.
7.1. The Mitochondrial Genoine

The mitochondrion is a complex organelle whose main function is
to produce ATP by oxidative phosphorylation. Of the many mitochon-
drial proteins, it is estimated that 95% are encoded by nuclear genes
(called pet genes), and only 5% by the mitochondrial DNA. Mitochondri-
al DNA (mitDNA or p) is a double-stranded circular molecule about 70
kb in length. Each haploid cell has about 50 copies of mitDNA. Growth
of p + cells in the presence of high levels of ethidium or acridines results
in rapid and complete loss of mitDNA to produce p0 cells. p+ X p0
crosses yield 4 p + : 0 segregation in meiosis. Oligomycin blocks respira-
tion by acting on the mitochondrial ATPase complex, two subunits of
which are encoded by mitochondrial DNA. Oligomycin-resistant (OR)
mutants include mutants in each of these two mitochondrial genes. In a
cross of the type QR x 0 8 (oligomycin-sensitive), the diploids formed
show rapid segregation of QR from 0 8 • p can also be transferred by kar
crosses (cytoduction). Mating a p+ karl strain and a p0 strain results in p+
cytoductants.
The mitochondrial genome has been studied in great detail (for
excellent reviews, see Dujon, 1981, and Slonimskiet al., 1982). Hundreds
of mutants have been analyzed, and the entire genome has been se-
quenced. Figure 10 shows a map of the mitochondrial DNA, and Table
V lists the genes. The mitochondrion has its own protein synthesis appa-
ratus with different ribosomes, tRNA, and other elements, quite distinct
from the usual cytoplasmic ribosome system. The large and small rRN A
are coded by mitDNA, as are the tRNAs used, but the ribosomal pro-
teins, tRNA charging enzymes, and other protein synthesis factors are
coded by nuclear genes. The mitochondrial ribosomes are insensitive to
cycloheximide, which acts on the cytoplasmic (nuclear-coded) ribosomes,
but they are sensitive to many of the antibiotics that affect bacterial ribo-
somes, such as chloramphenicol, erythromycin, and spiramycin. These
antibiotics were very useful in distinguishing mitochondrially synthe-
sized from cytoplasmically synthesized proteins. Also, mutants resistant
to these drugs played a large role in the early studies of mitochondrial
genetics.
While many mature proteins synthesized in the cytoplasm are im-
ported into the mitochondria, as far as is known, no nuclear-encoded
messenger RNAs (mRNAs) are translated on mitochondrial ribosomes,
130 R. B. WICKNER
RNA-
Maturase
exons RNA MATURASE
Transcription and Translation are Clockwise for All Mitochon-

drial Genes (Including the lntron-Encoded Open Reading
Frames) Except for the tRNA 1 thr Gene.
• Exons of Genes.
rllA Open Reading Frames in lntrons.
s~r Location of Serine - tRNA.
Figure 10. The mitochondrial genome is about 50 kb. The genes are described in more
detail in Table V.
and all proteins encoded by mitDNA are translated in the mitochondria.

Examination of the sequences encoding those proteins has revealed that
mitochondria have a slightly different genetic code than the code that
chromosomal genes use. The major difference is that the UGA codon,
read as a stop codon by the cytoplasmic ribosomes, is read as tryptophan
by the mitochondrial ribosomes. Human mitochondria do the same
thing. Also, at least one CUA codon is translated as threonine in mito-
chondria instead of leucine.
Another striking aspect of mitochondrial DNA is the presence of
intervening sequences in three mitochondrial genes: the large rRNA
gene, the structural gene for cytochrome b ("cob-box" gene), and the gene
for subunit I of cytochrome oxidase (oxi3). Genetic studies, combined
with sequencing and analysis of transcripts and protein products, have
Table V. Mitochondrial Genes
Gene
name Gene product Phenotype of mutants Special features
oxil Subunit II of Respiratory-deficient No intervening se-

cytochrome oxidase quences
oxi2 Subunit III of Respiratory-deficient No intervening se-
cytochrome oxidase quences
oxi3 Subunit I of Respiratory-deficient 5 introns; long open
cytochrome reading frames in
first 4 introns; dif-
ferent strains have
more or fewer in-
trons
"cob-box" Cytochrome b Respiratory-deficient 5 introns; 2 with long
resistance to di- open reading
uron, antimycin, frames; ORFs
mucidin, or needed to excise in-
funiculoson tron
olil Subunit 9 of ATPase Phosphorylation- No intervening se-
deficient resistance quences
to oligomycin, ven-
turicidin, or os-
samycin
oli2 Subunit 6 of ATPase Phosphorylation- No intervening se-
deficient resistance quences
to ollgomycin
varl Mitoribosomal poly- Variable size of Part of varl product
peptide mitoribosomal pro- is coded by varl
tein locus; part (?) nu-
clear
"lg-rRNA" 21S rRNA Resistance to An intervening se-
chloramphenicol, quence; (w+) in-
erythromycin, or duces preferential
spiromycin; polarity inheritance of itself
of recombination
(w)
"sm-rRNA" 15S rRNA Resistance to par-
omomycin
"tRNA" Transfer RNAs Deficient in UGA = tryptophan
mitochondrial pro- instead of amber
tein synthesis
repl Origins of replication Extreme suppressive- Each has a similar
rep2 ness when in a p- 300-bp sequence;
rep3 mutant but rep• p- mutants
can still replicate
132 R. B. WICKNER
revealed functions for some of these intervening sequences (introns).

Other reading frames have been found in the intron in the large rRNA
gene, in two of the five cob-box introns, and in four of the seven introns
of the oxi3 gene.
Frameshift and chain-termination mutants in the open reading
frames of the second intron of the cob-box gene result in failure to excise
that very intron from the primary gene transcript. These same intron
mutants in the cob-box gene fail to express the oxi3 gene properly, owing
to a failure to properly excise introns from the oxi3 primary transcript.
These results indicate that the protein encoded by an intron of the
cytochrome b structural gene (cob-box gene) plays a necessary role in the
excision of introns in both the cob-box gene and the oxi3 gene. An oxi3
intron also appears to encode an "RNA maturase."
The intron in the gene for the large rRNA is present in some strains
(called w) and absent in others (called w-). In crosses of the type w+ x
w-, almost all of the progeny are w + ; that is, they have the intervening
sequence. This "selfish DNA" role for the intervening sequence is vague-
ly reminiscent of the "selfish RNA" role that Cech (1983) has suggested
for a self-excising intron in the Tetrahymena rDNA gene.
Before we leave the mitochondria, mention should be made of the
p- or petite mutation, by which Ephrussi et al. (1949) first recognized
the mitochondrial genome. If a single p+ colony is replated, around 1%
of the subclones will be found to be unable to grow using ethanol or
glycerol as a carbon source; that is, they will be unable to respire, al-
though they can use dextrose. Most of these p- mutants are due to
extensive deletions of mitDNA, leaving only a small sequence which is
duplicated in tandem to produce a molecule of about the same size as
normal mitDNA.
7.2. The Killer Systems

Certain strains of S. cerevisiae secrete a protein toxin that kills other
strains, but secreting strains are immune to the action of the toxin.
Ability to secrete the toxin (K + phenotype) and to resist its lethal effects
(R + phenotype) is determined by a linear dsRNA called M. In fact, there
are several distinct killer toxins (K 1, K2 , etc.) determined by distinct M
dsRNAs (M 1, M2 , etc.). The M1 dsRNA has been shown to code for the
protein precursor to the K 1 toxin (for reviews of the killer systems, see
Wickner, 1983, 1985, and Tipper and Bostian, 1984. (See Fig. 11.)
In addition to M1 or M2 , various yeast strains carry a variety of other
dsRNAs (Table VI). J.....A dsRNA is the only other variety known to be
essential for the killer phenomenon, although all killer strains also have
the minor species, W. J.....A codes for the major coat protein of the intra-
cellular viruslike particles (VLP) in which both J.....A and M are found.
M ds RNA
y? Immunity
Protein
leader a Toxin oo {J Toxin Poly A· Poly U No Open Reading Frames
(Sig~al) Subunit 50 5 5 ~ Subunit
Region
Peptide ~ ? ~ .~ • • Function Unknown
. . AAAAAAAAAA
~pG=---------------------------------------_:~~~~~----------------------- CA-OH
HO-AC--------------------------------~----------------==========
uuuuuuuuuu
Gppp :s::
100 bp
t----t
~
:t
0
~ Potential KEX2 t::l
[ll
Processing Site
z
-
• Potential Glycosylation Site C')
Figure 11. Most of the sequence of M1 dsRNA is known from a combination of DNA and RNA sequencing.
!:
~
C')
->
t""'
C'J
tol
ztol
...,
C')
[ll
-
Clll
Clll
-
134 R. B. WICKNER
Table VI. Components of the S. cereviliae Killer Systems
dsRNAs
M• A 1.8-kb dsRNA that codes for polpeptide toxin and a resistance
function. Strains carrying M1 have the genotype [KIL-k.J. Se-
cretion of an active K1 toxin is the K1+ phenotype. Resistance
to the K1 toxin is the R1+ phenotype.
A 1.5-kb dsRNA that codes for a second toxin which kills M-o-
and M1-containing cells. M2 , [KIL-k2 ), K2 +, and R2 + are analo-
gous to M1, [KIL-k.J, K1+, and R 1 +.
L-A A 4.5-kb dsRNA that is noninfectious and stably maintained like a
plasmid, but encapsidated like a virus. It encodes its major cap-
sid protein (81 kDa). The genes [EXL), [NEX), and [HOK) (see
below) are present in various combinations on various forms of
L-A, denoted L-A-E ([EXL] alone), L-A-HN ([HOK) and
[NEX); found in all K1 killers), L-A-HE ([HOK) and [EXL)), or
L-A-H ([HOK) only; found in certain K2 strains). L-A depends
on MAK3, MAKIO, and PET18 for replication.
L-B and L-C 4.5-kb dsRNAs unrelated to L-A and present in VLPs with differ-
ent major proteins. L-B and L-A or L-C and L-A are compat-
ible. L-B and L-C show some sequence homology. L-(BC)
means an L dsRNA of the L-B or L-C type, which, because it
has not been fully characterized, could be either.
Tand W 2. 76-kb and 2.25-kb minor dsRNAs. They do not cross-hybridize
with each other, other dsRNAs, or cell DNA. They are
cytoplasmically inherited. The copy number ofT and W is in-
duced 10-fold by growth at 37°C.
Cytoplasmic genes
[HOK) helper of killer. This non-Mendelian gene supplies the helper
function needed by M1 for replication in a wild-type strain. It is
located on certain forms of L-A dsRNA, namely, L-A-HN, L-A-
HE, and L-A-H. To test for [HOK), the combination of L-A-E
+ M1 is introduced from a ski2 strain into the SKI+ M-o strain
to be tested. L-A-E cannot support M1 replication in a SKJ+
host, but if [HOK) is present, M1 will be maintained
[EXL) excluder of M2 dsRNA. This non-Mendelian trait prevents the
replication of [KIL-k2 ) if [NEX) is absent, but not if [NEX) is
present. [EXL) is located on certain forms of L-A, namely, L-A-
E and L-A-HE.
[NEX) M2 nonexcludable by [EXL], but does not prevent exclusion of M2
by strains carrying M1• [NEX) is located on L-A-HN, the form
of L-A found in wild-type K1 killer strains.
[KIL-b) A frequently encountered natural variant of [KIL-k.J in which M1
does not need many of the MAK gene products, the copy num-
ber of M1 is higher than usual, the strain is a superkiller, and
the degree of repression of L-A copy number by M1 is greater
than usual.
[KIL-d) A mutant form of [KIL-k.J in which haploid strains unstably
maintain and incompletely express M1, but diploid strains stably
maintain and normally express M1•
[KIL-sk) A mutant form of M1 dsRNA in which toxin stability is increased
compared to the parent, resulting in the superkiller phenotype.
Table VI. (Continued)
[KIL-i] A mutant form of M1 dsRNA in which no active toxin is pro-

duced, but cells are immune to toxin action.
[KIL-s] Deletion mutants of M1 which prevent the replication of (sup-
press) the parent M 1 dsRNA. These mutants are analogous to
defective interfering particles of animal viruses.
Chromosomal genes
MAK maintenance of [KIL-kd. MAK genes comprise at lease 29
chromosomal genes necessary to maintain M 1 dsRNA. Mutants
carrying the recessive alleles, mak, are K-R- M-o. At least some
of the genes are also required for the maintenance of M2 •
MAKJ, MAKJO, and PETJB are needed by all forms of L-A.
MAK27 is needed by L-A-E.
clo A complex chromosomal defect resulting in loss of L-B or L-C.
SKI superkiller. Mutants carrying the recessive alleles of ski2, skiJ, ski4,
ski6, ski7, or skiB produce more killer toxin and have increased
copy number of M1, M2 , L-A, and L-(BC). ski- mutations elimi-
nate the need for some of the MAK genes and prevent the ex-
clusion of M2 by L-A-HN in mkt strains. The ski5 mutant has
none of these traits and lacks a cell surface protease that nor-
mally degrades the toxin.
MKT maintenance of [KIL-k2 ] in the presence of [NEX]. Strains having
a recessive allele, mkt (about 80% of laboratory strains), cannot
maintain [KIL-k 2 ] at 30°C if [NEX] (L-A-HN) is present.
MKS mkt suppressor. Like most ski- mutations, mutations in mksl, mks2,
or MKS50 prevent exclusion of M2 by L-A-HN in mktl hosts,
but do not show other characteristics of ski- mutations.
KEX killer expression. Two chromosomal genes needed to process the
toxin precursor. Mutants are K-R+. The kex mutants also do
not properly process the a-hormone precursor, and many other
secreted proteins are affected.
REX resistance expression. One chromosomal gene needed to express
M 1-determined resistance to the toxin.
KRE killer resistant. Three chromosomal genes needed for normal tox-
in action on sensitive cells. KREJ affects ~(l-6)glucan, the nor-
mal toxin cell wall receptor. KRE2 also affects toxin binding,
but KREJ does not.
SEC secretion. Chromosomal genes for general protein secretion; mu-
tants in these genes are conditional lethal and also affect killer
toxin secretion.
These VLPs are found only inside the cell and are not infectious, so that
their status as viruses is questionable. However, their structural re-
semblance to virus particles and the existence of a particle RNA poly-
merase producing full-length, message strand, single-stranded RNA
(ssRNA) is certainly reminiscent of a virus. lrB and lrC determine dif-
ferent VLP major coat proteins from that determined by lrA, but in
normal strains, M requires lrA, and neither lrB nor lrC will suffice.
136 R. B. WICKNER
The T and W dsRNAs are usually present at very low copy number, but
when cells are grown at 37°C, their copy number increases 10-fold or
more to reach copy numbers similar to that of M. T and W are not
present in the same VLPs as are the lrA, lrB, or lrC molecules, but
where they are located in the cell is not yet known.
While the study of most viruses and plasmids has centered on the
viral or plasmid genome itself, the study of the killer systems has been
somewhat more global, with equal emphasis on the chromosomal genes
involved in replication, expression, and regulation of the dsRNA ge-
nomes. Table VI summarizes the chromosomal genes involved in the
killer systems.
Those involved in expression of the killer phenotype are the kexl
and kex2 genes and the sec genes. The kexl and kex2 genes are relatively
specific for the killer toxin, although kex2 mutations also affect secretion
of a pheromone and prevent meiotic spore formation. KEX2 is now
known to encode a protease that cleaves after a pair of basic residues.
The sec mutants are defective in various stages of general protein secre-
tion and are all temperature-sensitive (ts) for growth. The secretion of
the killer toxin apparently follows the general protein secretory pathway,
but, for some reason, needs two extra genes, kexl and kex2.
Resistance to the toxin produced by M dsRNA requires, in addition,
the REXJ product. The KRE products are involved in normal toxin
action. krel mutants lack the toxin receptor-J3(1-6)glucan-in the cell
wall.
The mak- mutants (maintenance of killer) lose M 1 dsRNA. Thirty
MAK genes have been identified and mapped. Biochemical components
shown by these mutations to be involved in M replication are the poly-
amines spermidine and spermine, ribosomal protein L3 (MAK8), and
topoisomerase I (MAKJ). How these components are involved in the
replication process is unknown. Some MAK genes are essential for cell
growth (MAK16, PET18, MAKJO); one (PET18) is essential as well for
mitDNA replication. The MAKJ, MAKJO, and PET18 genes are needed
for lrA replication as well as M replication. Since M requires lrA, per-
haps M's requirement for these three chromosomal genes is via its re-
quirement for lrA. The MKT genes are needed for M2 (but not M 1) if lr
A-HN, a particular form of lrA, is present. As may be seen in Table VII,
different dsRNAs have very different requirements for chromosomal
genes for their replication.
The SKI gene products lower the levels of lr(BC), lrA, and M
dsRNAs. The ski- mutants were first identified by their superkiller phe-
notypes, but they have several other phenotypes. The ski- mutations
suppress many mak- mutations as well as the mkt- mutations, and if M is
present, ski- mutants are cold-sensitive (cs) for growth. This may be due
to the elevated amounts of M dsRNA utilizing all of some cell compo-
nent needed for growth at the low temperature.
Table VII. Different Chromosomal Gene Requirements for dsRNAs
dsRNAs MAl() MAKJO PETJ8a MAXI MAK8 SPE2 SPEIO MAK27 MKTI CLO
MI +b + + + + + + + +
M2 + + + + + + + + +
L-A-HN + + +
L-A-HE + + +
L-A-E + + + - - - + +
-C -C
~
L-B - - - -C -c -C
+
L-C - - - +
!:l::c
T 0
w ~
a Mitochondrial DNA also requires PETI8 for its replication. zn
h+ = gene required for replication or maintenance; - = gene not required for replication or maintenance.
t""'
cRequirement determined for an undifferentiated member of the L-B and L-C class of molecules [L-(BC)]. >
~
n
-
>
t""'
C'l
1:"1
z
1:"1
--l
c:i
en
~
...,
-
138 R. B. WICKNER
A number of dsRNA exclusion phenomena have been described.

M1 excludes M2 ; L-A-E (a form of lrA) excludes M2 if lrA-HN is absent;
lrA-HN excludes M2 if the host is mkt-; and deletion mutants of M1 (S
dsRNAs) exclude the wild-type M1 dsRNA. Another interaction among
the dsRNAs is the ability of M 1 or M2 to lower the copy number of lrA
1O-fold or more. The basis of these interactions is unclear but could
involve competition among dsRNAs for replication proteins.
7.3. 2-pm DNA (Reviewed by Broach, 1981)

2-JLm DNA is a 6300-bp, circular, double-stranded molecule present
in most Saccharomyces strains at 50-100 copies per cell. It has a pair of
599 inverted repeat sequences, frequent recombination between which
is catalyzed by a 2-JLm DNA-coded, site-specific recombinase called FLP.
Thus, two intracellular forms of 2-JLm DNA are in equilibrium. 2-JLm
DNA also encodes two complementable replication functions and a site
(called STB) needed for proper segregation. Crossing strains with and
without 2-JLm DNA results in 4:0 segregation, but 2-JLm DNA is only
transmitted at a 50% rate by cytoduction. The DNA is found in histone-
containing chromatin, and each molecule replicates exactly once per cell
cycle (in S phase), making it seem likely that 2-JLm DNA is actually
located in the nucleus. 2-JLm DNA acts like a multicopy minichromo-
some, with an origin of replication, a centromere, chromatin structure,
and cell-cycle-controlled replication. It has no essential function for the
cell. 2-JLm DNA can be cured by transformation with pJDB219, a
LEU2-carrying vector which is itself unstable. After pJDB219 is lost, the
cells remain healthy.
The origin of replication of 2-JLm DNA is used in YEp vectors (see
Chapter 5).
8. APPENDIX I: MEDIA
To prepare solid media, the ingredients listed are autoclaved to-

gether for 15 min, except as indicated below, and, after cooling to 50-
600C, poured into Petri plates. One liter of medium is autoclaved in a 2-
liter flask. About 30 plates are poured per liter of medium. Plates are left
to dry at room temperature for a few days and are then stored at 4°C in
the plastic bags that the empty Petri dishes came in. The plates are good
indefinitely unless they dry out or become contaminated.
YPAD-Rich Medium
1% yeast e.xtract
2% peptone
2% dextrose
2% agar
0.04% adenine sulfate (prevents adel and ade2 mutants from ac-
cumulating red pigment and dying)
SD Synthetic Minimal Medium with Dextrose

0.67% yeast nitrogen base without amino acids (available from Difco
or from Gibco)
2% dextrose
2% agar
YPAG Glycerol Medium-Only Respiration-Competent

Cells Can Grow
1% yeast extract
2% peptone
3% vollvol glycerol
2% agar
0.04% adenine sulfate
H Synthetic Complete Medium

0.04% adenine sulfate
0.67% yeast nitrogen base without amino acids
2% dextrose
2% agar
10 mllliter 1 M K2 HP04
50 ml/liter of appropriate mix
Mix Composition
Uracil 0.5 g/liter
L-Tryptophan 0.5 g/liter
L-Histine 0.5 g/liter
L-Arginine 0.5 g/liter
L-Methionine 0.5 g/liter
L-Tyrosine 0.7 g/liter
L-Isoleucine 0.7 g/liter
L-Lysine 0.7 g/liter
L-Phenylalanine 1.2 g/liter
L-Valine 3.0 g/liter
140 R. B. WICKNER
After autoclaving, add 5 ml of sterile 5% L-threonine, 5 ml of 2%

sodium aspartate, and 5 ml of 1% L-leucine per liter of medium.
Mixes with the above composition lacking one or more nutrients are
prepared and kept at room temperature. The author uses H-ade, H-ura,
H-trp, H-his, H-arg, H-tyr, H-lys, H-phe, H-ilv (lacking isoleucine and
valine), and H-MAT (lacking methionine, aspartate, and threonine). The
H-MAT plates make scoring of most methionine, threonine, and aspar-
tate auxotrophs more clear-cut than if only the single amino acid was left
out.
Sporulation Medium
I% potassium acetate
0.1% yeast extract
0.05% dextrose
2% agar
9. APPENDIX II: SAMPLE TETRAD DATA
These data were collected from a cross of the following parent

strains:
1653-12C = MAT a trpl [KIL-kd

X
2011-33A =MAT a mak21 arolD petl9 ade8 rna3 [KIL-o]
mak21 results in loss of M 1 dsRNA and thus the K1 - phenotype. arolD

results in a requirement for both tryptophan and phenylalanine, while
trpl strains need only tryptophan. To score trpl, all segregants were
mated with a trpl and a trpl tester strains. The formation of prototro-
phic diploids is indicated by a"+" in the appropriate column and implies
that the spore clone was TRP 1 +. The petl9 segregants are unable to use
glycerol as a carbon source and thus cannot grow on YPG plates. The
ade8 mutation makes cells auxotrophic for adenine. The rna3 mutation
results in temperature sensitivity for cell growth. 37°C is the nonper-
missive temperature, and growth on YPAD plates at 37°C is indicated
under the heading "37°." Mating type is scored by mating the spore
clones on YPAD plates with standard a ura4 and a ura4 tester strains.
Ability to mate with the a ura4 strain is seen as the formation of pro-
totrophic diploids (a"+" in the column headed "X a tester") and implies
that the spore clone tested has the a mating type.
With the data shown here, the reader can construct a partial map of
chromosome IV. What can be said about the location of the centromere?
Of the mating type locus? Hint: the data have been arranged in a rather
suggestive way. Look at the segregation of individual tetrads.
Tetrad X a Xcx x a trpl X cxtrpl

number -trp K, -phe YPG -ade 37° tester tester tester tester
A + + + +
B + + +
c + + + + + + + + + +
D + +
2 A + +
B + + + + +
c + + +
D + + + + + +
3 A + + + + + +
B + + +
c + +
D + + + + + + +
4 A +
B + + + + + + +
c + + + +
D + + + + +
5 A + +
B + + +
c + + + +
D + + + + + + + + + +
6 A + + +
B + + + +
c + + +
D + + + + + +
7 A + + + +
B + +
c + + + + +
D + + + + + +
8 A + +
B + + +
c + + + + +
D + + + + + +
9 A + +
B + + + + +
c + + + +
D + + + + + + + + +
10 A + + +
B + + + + + +
c + + + + +
D + + +
(continued)
142 R. B. WICKNER
Tetrad Xa Xa x a trpl x a trpl

number -trp K• -phe YPG -ade 37° tester tester tester tester
11 A +
B + + + + + +
c + + + +
D + + + + + +
12 A +
B + + + + + + +
c + +
D + + + + + + + +
13 A + +
B + + + +
c + + +
D + + + + + + + +
14 A + +
B + + + + +
c + + + + + +
D + + + +
15 A + + + + + + + + +
B + + + +
c + +
D + + + +
16 A + +
B + + +
c + + + + + +
D + + + + + +
17 A + +
B + + + + +
c + + + +
D + + + + +
18 A + + +
B + +
c + + + + + + + + + +
D + + + +
19 A + + +
B + + +
c + + + + +
D + + + + + +
20 A + +
B + + +
c + + + +
D + + + + + + + + + +
21 A + + +
B + + + +
c + + + +
D + + + + + +
22 A + + +
Tetrad Xa Xa x a trpl x a trpl

B + + + + + + +
c + + + + +
D + + +
23 A + + + + + + +
B + + +
c + + +
D + + + +
24 A + + + +
B + + + +
c + + +
D + + + + +
25 A + +
B + + + + +
c + + + + +
D + + + + +
26 A + +
B + + + + +
c + +
D + + + + + + + +
27 A + + +
B + + + + + +
c + +
D + + + + + +
28 A + + + +
B +
c + + + + +
D + + + + + + +
29 A + + + +
B + + + +
c + + + + +
D + + + + +
30 A + +
B + + +
c + + + + + + + + +
D + + + + +
31 A + +
B + + +
c + + + + + +
D + + + + + +
32 A + + + + +
B + +
c + + + +
D + + + + + +
(continued)
144 R. B. WICKNER
Tetrad xa X a x a trpl x a trpl

33 A + + + + + + +
B + + +
c + + + +
D + + +
34 A + +
B + + + + + + +
c + + +
D + + + + + +
35 A + + + + + + + + +
B +
c + + +
D + + + + + + +
36 A + + + +
B + +
c + + + + + +
D + + + + +
REFERENCES
Broach, J. R., 1981, The yeast plasmid 2 11. circle, in: The Molecular Biology of the Yeast
Saccharomyces: Life Cycle and Inheritance (J. N. Strathern, E. W. Jones, and J. R. Broach,
eds.), Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, pp. 445-470.
Carle, G. F., and Olson, M. V., 1985, An electrophoretic karyotype for yeast, Proc. Natl.
Acad. Sci. USA 82:3756-3760.
Cech, T. R., 1983, RNA splicing: Three themes with variations, Cel/34:713-716.
Conde, J., and Fink, G. R., 1976, A mutant of Saccharomyces cerevisiae defective for nuclear
fusion, Proc. Natl. Acad. Sci. USA 73:3651-3654.
Culbertson, M. R., and Henry, S. A., 1973, Genetic analysis of hybrid strains trisomic for
the chromosome containing a fatty acid synthetase gene complex (fasl) in yeast, Genet-
ics 75:441-458.
DeFeo-Jones, D., Scolnick, E. M., Koller, R., and Dhar, R., 1983, ras-Related gene se-
quences identified and isolated from Saccharomyces cerevisiae, Nature 306:707-709.
Dujon, B., 1981, Mitochondrial genetics and functions, in: The Molecular Biology of the Yeast
Saccharomyces: Life Cycle and Inheritance (J. N. Strathern, E. W. Jones, and J. R. Broach,
Ephrussi, B., Hottinguer, H., and Tavlitzki,J., 1949, Action de !'acriflavine sur les levures.
II. Etude genetique du mutant "petite colonie," Ann. Inst. Pasteur (Paris) 76:419.
Falco, S. C., and Botstein, D., 1983, A rapid chromosome mapping method for cloned
fragments of yeast DNA, Genetics 105:857-872.
Fink, G. R., and Styles, C. A., 1972, Curing of a killer factor in Saccharomyces cerevisiae, Proc.
Fogel, S., Mortimer, R. K., and Lusnak, K., 1981, Mechanism of meiotic gene conversion or
"wanderings on a foreign strand," in: The Molecular Biology of the Yeast Saccharomyces:
Life Cycle and Inheritance (J. N. Strathern, E. W. Jones, and J. R. Broach, eds.), Cold
Spring Harbor Laboratory, Cold Spring Harbor, NY, pp. 289-339.
Gaber, R., Mathison, L., Edelman, 1., and Culbertson, M. R., 1983, Frameshift suppression
in Saccharomyces cerevisiae. VI. Complete genetic map of twenty-five suppressor genes,
Genetics 103:389-407.
Gallwitz, D., Donath, C., and Sander, C., 1983, A yeast gene encoding a protein homolo-
gous to the human C-has/bas protooncogene product, Nature 306:704-707.
Goldring, E. S., Grossman, L. E., Krupnick, D., Cryer, D. R., and Marmur, J., 1970, The
petite mutation of yeast: Loss of mitochondrial DNA during induction of petites with
ethidium bromide,]. Mol. Bioi. 52:323-335.
Henry, S. A., Donahue, T., and Culbertson, M., 1975, Selection of spontaneous mutants by
inositol starvation in Saccharomyces cerevisiae, Mol. Gen. Genet. 143:5-ll.
Herskowitz, I., and Oshima, Y., 1981, Control of cell type in Saccharomyces cerevisiae: Mating
type and mating-type interconversion. in: The Molecular Biology of the Yeast Sac-
charomyces: Life Cycle and Inheritance Q. N., Strathern, E. W. Jones, and J. R. Broach,
Hilger, F., and Mortimer, R. K., 1980, Genetic mapping of argl and arg8 in Saccharomyces
cerevisiae by trisomic analysis combined with interallelic complementation.]. Bacteriol.
141:270-274.
Hilger, F., Prevot, M., and Mortimer, R. K., 1982, Genetic mapping of arg, cpa, car, and tsm
genes in Saccharomyces cerevisiae by trisomic analysis, Curr. Genet. 6:93-98.
Kawasaki, G., 1979, Karyotypic instability and carbon source effect in cell cycle mutants of
Saccharomyces cerevisiae, Ph.D. thesis, University of Washington, Seattle.
Kikuchi, Y., 1983, Yeast plasmid requires a cis-acting locus and two plasmid proteins for its
stable maintenance, Cell 35:487-493.
Klapholz, S., and Esposito, R. E., 1982, A new mapping method employing a meiotic rec-
mutant of yeast, Genetics 100:387-412.
Lindegren, G., Hwang, Y. L., Oshima, Y., and Lindegren, C. C., 1965, Genetical mutants
induced by ethyl methane sulfonate in Saccharomyces, Can.]. Genet. Cytol. 7:491-499.
Liras, P., McCusker,]., Mascioli, S., and Haber, J. E., 1978, Characterization of a mutation
in yeast causing nonrandom chromosome loss during mitosis, Genetics 88:651-671.
Livingston, D. M., 1977, Inheritance of 2 f.L mDNA plasmid from Saccharomyces, Genetics
86:73-84.
Loumaye, E., Thorner, J., and Catt, K. ]., 1982, Yeast mating pheromone activates mam-
malian gonadotrophs: Evolutionary conservation of a reproductive hormone? Science
218:1323-1325.
Ma, C., and Mortimer, R. K., 1983, Empirical equation that can be used to determine
genetic map distances from tetrad data, Mol. Cell. Bioi. 3:1886-1887.
Matsumoto, K., Uno, I., Oshima, Y., and Ishikawa, T., 1982, Isolation and characterization
of yeast mutants deficient in adenylate cyclase and cAMP-dependent protein kinase,
Matsumoto, K., Uno, 1., and Ishikawa, T., 1983a, Initiation of meiosis in yeast mutants
defective in adenylate cyclase and cyclic AMP-dependent protein kinase, Cell 32:417-
432.
Matsumoto, K., Uno, 1., and Ishikawa, T., 1983b, Control of cell division in Saccharomyces
cerevisiae mutants defective in adenylate cyclase and cAMP-dependent protein kinase,
Exp. Cell Res. 146:151-161.
Matsumoto, K., Uno, 1., and Ishikawa, T., 1985, Genetic analysis of the role of cAMP in
yeast, Yeast 1:15-24.
Mortimer, R. K., and Hawthorne, D. C., 1973, Genetic mapping in Saccharomyces. IV. Map-
ping of temperature-sensitive genes and use of disomic strains in localizing genes,
Genetics 74:33-54.
Mortimer, R. K., and Manney, T. R., 1971, Mutation induction in yeast, in: Chemical
Mutagens, Vol. 1 (A. Hollaender, ed.), Plenum Press, New York, pp. 289-310.
Mortimer, R. K., and Schild, D., 1980, Genetic map of Saccharomyces cerevisiae, Microbiol.
Rev. 44:519-571.
146 R. B. WICKNER
Mortimer, R. K., and Schild, D., 1981, Genetic mapping in Saccharomyces cerevisiae, in: The
Molecular Biology of the Yeast Saccharomyces: Life Cycle and Inheritance O. N. Strathern, E.
W. Jones, and J. R. Broach, eds.), Cold Spring Harbor Laboratory, Cold Spring Har-
bor, NY, pp. 11-26.
Mortimer, R. K., and Schild, D., 1985, Genetic map of Saccharomyces cerevisiae, Microbiol.
Rev. 49:519-571.
Mortimer, R. K., Contopoulou, R., and Schild, D., 1981, Mitotic chromosome loss in a
radiation-sensitive strain of the yeast Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. USA
78:5778-5782.
Perkins, D. D., 1949, Biochemical mutants in the smut fungus Ustilago maydis, Genetics
34:607-626.
Molecular Biology of the Yeast Saccharomyces: Life Cycle and Inheritance 0· N. Strathern, E.
W. Jones, and J. R. Broach, eds.), Cold Spring Harbor Laboratory, Cold Spring Har-
bor, NY: pp. 97-142.
Riley, M. 1., and Manney, T. R., 1978, Tetraploid strains of Saccharomyces cerevisiae that are
trisomic for chromosome III, Genetics 89:667-684.
Shaffer, B., Brearley, R., Littlewood, R., and Fink, G. R., 1971, A stable aneuploid of
Saccharomyces cerevisiae, Genetics 67:483-495.
Singh, A., Helms, C., and Sherman, F., 1979, Mutation of the non- Mendelian suppressor,
~f~+, in yeast by hypertonic media, Proc. Natl. Acad. Sci. USA 76:1952-1956.
Slonimski, P. P., Borst, P., and Attardi, G. (eds.), 1982, Mitochondrial Genes, Cold Spring
Harbor Laboratory, Cold Spring Harbor, NY.
Snow, R., 1966, An enrichment method for auxotrophic yeast mutants using the antibiotic
"nystatin," Nature 211:206-207.
Snow, R., 1979, Maximum likelihood estimation of linkage and interference from tetrad
data, Genetics 92:231-245.
Sommer, S. S., and Wickner, R. B., 1982, Co-curing of plasmids affecting killer dsRNAs of
Saccharomyces cerevisiae: [HOK), [NEX), and the abundance of L are related and fur-
ther evidence that M1 requires L,]. Bacteriol. 150:545-551.
Strathern,J. N.,Jones, E. W., and Broach,J. R., eds., 1981, The Molecular Biology of the Yeast
Saccharomyces, 2 vols., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
Tatchell, K., Robison, L. C., and Breitenbach, M., 1985, RAS2 of Saccharomyces cerevisiae is
required for gluconeogenic growth and proper response to nutrient limitation, Proc.
Tipper, D. J., and Bostian, K. A., 1984, Double-stranded RNA killer systems in yeasts,
Microbiol. Rev. 48:125-156.
Toda, T., Uno, I., Ishikawa, T., Powers, S., Kataoka, T., Broek, D., Cameron, S., Broach,J.,
Matsumoto, K., and Wigler, M., 1985, In yeast, RAS proteins are controlling elements
of adenylate cyclase, Cell40:27-36.
Tuite, M. F., Mundy, C. R., and Cox, B.S., 1981, Agents that cause a high frequency of
genetic changes from [psi+) to [psi- I in Saccharomyces cerevisiae, Genetics 98:691-"711.
Uno, I., Matsumoto, K., and Ishikawa, T., 1982, Characterization of cyclic AMP-requiring
yeast mutants altered in the regulatory subunit of protein kinase, ]. Biol. Chem.
257:14110-14115.
Wickner, R. B., 1974, "Killer character" of Saccharomyces cerevisiae: Curing by growth at
elevated temperature,]. Bacteriol. 117:1356-1357.
Wickner, R. B., 1979, Mapping chromosomal genes of Saccharomyces cerevisiae using an
improved genetic mapping method, Genetics 92:803-821.
Wickner, R. B., 1983, Genetic control of replication of the double-stranded RNA segments
of the killer systems in Saccharomyces cerevisiae, Arch. Biochem. Biophys. 222:1-11.
Wickner, R. B., 1985, Killer systems in yeast, in: Current Topics in Medical Mycology, Vol. 1
(M. R. McGinnis, ed.), Springer-Verlag, New York (in press).
Wickner, R. B., and Leibowitz, M. J., 1977, Dominant chromosomal mutation bypassing
chromosomal genes needed for killer RNA plasmid replication in yeast, Genetics
87:453-469.
Wickner, R. B., Boutelet, F., and Hilger, F., 1983, Evidence for a new chromosome in
Wood,J. S., 1982, Mitotic chromosome loss induced by methyl benzimidazole-2-y1-carba-
mate as a rapid mapping method in Saccharomyces cerevisiae, Mol. Cell Biol. 2:1080-
1087.
Recombinant DNA Techniques· 5
A. J. KINGSMAN, E. J. MELLOR, M. J. DOBSON, and
S. M. KINGSMAN
1. INTRODUCTION
The aim of this chapter is to introduce the procedures currently

used to isolate and manipulate yeast genes. We restrict our discussion to
those procedures that are, at least in part, peculiar to the yeast Saccharo-
myces cerevisiae. We do not deal with procedures that are common to all
other organisms, such as eDNA bank construction, hybrid-selected or
arrested translation systems, nucleic acid probe strategies, or radioim-
mune detection strategies, as these have been covered in a variety of
general texts (e.g., Maniatis et al., 1982; Glover, 1985).
One of the earliest procedures for the isolation of yeast genes used
an Escherichia coli complementation system (Ratzkin and Carbon, 1977;
Struhl et al., 1976). A genome representative collection of yeast DNA
fragments was inserted into an E. coli plasmid and then used to trans-
form a variety of E. coli auxotrophs. Some of the E. coli mutations were
complemented by yeast DNA and most of these fragments were found
to contain the corresponding yeast gene. In this way the yeast LE U2
gene, which encodes isopropylmalate dehydrogenase, was isolated by
complementation of an E. coli leuB mutation (Ratzkin and Carbon, 1977)
and the yeast TRP 1 gene, encoding phosphoribosyl anthranilate iso-
merase, was isolated by complementation of an E. coli trpc mutation
(Struhl et al., 1976). This procedure was unreliable because it depended
on expression of the yeast genes in E. coli via the fortuitous presence in
the yeast DNA of sequences that resemble E. coli expression signals. In
fact, it was successful in only about 25% of cases. However, it did provide
the first cloned selectable yeast genes and led directly to the develop-
ment of yeast transformation techniques that used these genes as select-
able markers. With the advent of yeast transformation and plasmid
A. j. KINGSMAN, E. J. MELWR, M. J. DOBSON, and S. M. KINGSMAN • De-

partment of Biochemistry, University of Oxford, Oxford OX 1 3QU, England.
149
150 A. J. KINGSMAN et al.
vectors, yeast gene isolation is no longer achieved by E. coli complemen-

tation, but rather by direct yeast complementation.
2. YEAST TRANSFORMATION
Two major procedures are used for the introduction of DNA into
yeast cells. In the first (Hinnen et al., 1978; Beggs, 1978) exponentially
growing cells are treated with an enzyme, either glusulase, zymolyase, or
lyticase, to remove the cell wall (see Chapter 9). The resulting spher-
oplasts are then exposed to DNA in the presence of Ca2 + ions followed
by treatment with polyethylene glycol of molecular weight about 4000
Da (PEG 4000). The spheroplasts are then embedded in a regeneration
agar that usually imposes selection for a selectable marker and acts as a
physical support for the regeneration of the wall. The mechanism of
DNA uptake is not understood, but it may involve DNA entrapment
during PEG-mediated spheroplast fusion. However, it is worth noting
that diploids are not created at high frequencies (<5%) (M. J. Dobson,
unpublished data) during the transformation process with haploid
strains. Using this technique with anautonomous plasmid vector, trans-
formation frequencies as high as 106 /JJ.g DNA can be obtained.
The second procedure (Ito et al., 1983) does not require the removal
of the cell wall. Whole cells are subjected to prolonged (2-8 hr) exposure
to Li + ions at 4°C followed by treatment with DNA and PEG 4000. Like
the E. coli techniques, ·the cells are then given a heat shock and plated
directly onto the surface of selective plates. This procedure gives trans-
formation frequencies about 100-fold lower than the spheroplast pro-
cedure, but it does have the advantage that transformants grow on the
surface of the selection plates rather than embedded in them. This
makes subsequent processing (e.g., replica plating) easier. The reduction
in transformation frequencies with the Li + technique is not so pro-
nounced (i.e., 10-fold) for linear DNA fragments (see Section 6).
3. PLASMID VECTORS
Almost all yeast plasmid vectors are, in fact, both yeast and E. coli
vectors. They usually comprise all or part of an K coli vector such as
pBR322 and a yeast relication and selection system. This arrangement
allows E. coli to be used as the "preparative" organism for the various
manipulations associated with recombinant DNA procedures. Yeast au-
tonomous plasmid vectors are based on two types of replication system.
The first is dependent on chromosomal DNA fragments that are able to
replicate autonomously (Struhl et al., 1979; Stinchcomb et al., 1979;
RECOMBINANT DNA TECHNIQUES 151
Kingsman et al., 1979). These are the autonomously replicating sequen-

ces (ARSs) and may be chromosomal origins of replication (Williamson,
1985 ). The second makes use of sequences derived from the endoge-
nous yeast 2 ,_..plasmid (e.g., Beggs, 1978; Struhl et al., 1979; Broach,
1982).
3.1. ARS-Based Plasmids

A commonly used set of ARS-based plasmids is based on the plas-
mid YRp7 (Struhl et al., 1979). This is a pBR322 derivative containing a
1.45-kb EcoRI fragment from the yeast genome. This fragment contains
the TRP 1 gene, which can be used as a selectable marker, and the ARSJ
sequence. YRp7 transforms yeast at frequencies up to 105 /,_..g DNA; it is
maintained in the nucleus but, like all recombinant ARS plasmids, it is
extremely unstable (Kingsman et al., 1979). Under selective conditions
only 20-60% of cells contain plasmid, and under nonselective conditions
more than 95% of cells lose the plasmid after 15 generations in haploids
but in diploids ARS plasmids are more stable. Interestingly, in those cells
that contain plasmid, copy numbers are high (20-80/cell). The in-
stability appears to be due to a segregation defect that results in a bias
toward segregation to mother cells (Murray and Szostak, 1983a). The
instability of ARS-based plasmids can be overcome by the addition of a
segment of DNA from a yeast centromere (Clarke and Carbon, 1980).
These ARS-CEN plasmids also transform at frequencies round 105 I ,_..g
DNA, but in contrast to the ARS plasmids, they are almost entirely stable
and they are maintained at a copy number of 1-2/cell. Examples of ARS
and ARS-CEN plasmids are given in Fig. 2 and a comprehensive list can
be found in Parent et al. ( 1985 ).
ARS plasmids are of limited usefulness because of their instability
although there are some circumstances where this can be exploited to
identify a plasmid-encoded function by coloss with a selectable marker.
ARS-CEN plasmids are favored when a low-copy-number, stable system
is required without the use of an integrative vector (Williamson, 1985)
(see Section 6).
3.2. 2-J&m-Based Plasmids

The 2-j.Lm plasmid is present in most laboratory strains of S. cere-
visiae at copy numbers from 20 to 200/cell (Broach, 1982). It is a circle of
6318 bp and contains two inverted repeats of 599 bp (Fig. 1). 2-j.Lm
replication and maintenance are dependent on two cis-acting and two
trans-acting functions (Jayaram et al., 1983). The cis-acting functions are
the origin of replication and a series of 62-bp repeated sequences,
termed REPJ or STB, that seem to be involved in the fidelity of plasmid
E
H
Figure I. Organization of the 2-~J.m plasmid. The A-form of the plasmid is shown. Recom-
bination across the inverted repeats would invert one side of the "dumbbell" with respect
to the other to produce the B-form. Open boxes = large open reading frames. Closed box
= origin of replication. IVR = inverted repeat. Restriction sites: H =Hindi II; E = EcoRI;
P = Pstl.
segregation. The trans-acting functions are encoded by REP1 and REP2.

It is not clear, at present, whether these genes control replication, segre-
gation, or both. A third major coding sequence, FLP, encodes a recom-
binase that mediates site-specific recombination between the inverted
repeats. 2-f.Lm-based vectors may contain all four essential functions, in
which case they are able to replicate in the absence of an endogenous 2 f.L
plasmid (i.e., in a 2-f.Lm0 strain), or they may contain only the cis-acting
functions, in which case stable maintenance is dependent on the comple-
menting REP1 and REP2 genes of the endogenous plasmid in a 2-f.Lm +
strain.
2-f.Lm-based vectors are capable of transforming yeast at frequencies
as high as 106 /f.Lg DNA and maintaining copy numbers as high as
400/cell (Dobson et al., unpublished data). However, the precise charac-
teristics of any vector vary with the particular construction used. Copy
numbers as low as 5-l 0 are not uncommon. An example of a commonly
used 2-f.Lm-based plasmid is given in Fig. 2 and a comprehensive list of
2-f.Lm-based vectors can be found in Parent et al. ( 1985 ).
3.3. Yeast Gene Isolation Using Plasmid Vectors

Most yeast genes have been isolated by selecting, from a genome
representative plasmid pool, a molecule that is able to complement the
defect in a known yeast mutant (e.g., Nasmyth and Reed, 1980). In the
case of, for example, an auxotrophic mutation, selection is direct and
requires only that growth is obtained on minimal medium. Alternatively,
in the case of more complex phenotypes, such as those conferred by the
mating-type genes, it may be necessary to obtain sufficient yeast transfor-
mants, using the plasmid selectable marker, to ensure representation of
the entire genome and then screen these transformants for the desired
phenotype (Nasmyth and Tatchell, 1980). The numbers involved can
a b
H H
c E
Figure 2. (a) The ARS plasmid YRp7. (b) The ARS-CEN plasmid pMA9. (c) The 2-tJ.m-
based plasmid pMA3a. Open box = yeast DNA. Thin line = pBR322 DNA with ampicillin
and tetracycline resistance markers shown. Restriction sites: H = Hindlll; E = EcoRl; B =
BamHl; P = Pstl.
easily be calculated from considerations of the genome size and the size
of the fragments inserted into the vector (Clarke and Carbon, 1976).
Plasmid systems are favored for this approach because they give high
frequencies of transformation, and, more important, because they con-
tain bacterial plasmid replicons they can be retrieved from the initial
yeast transformant by preparing DNA from the transformant and using
it to transform E. coli, selecting for a bacterial marker (e.g., ampicillin
resistance). The bacterial clone can then be propagated to produce large
amounts of the yeast DNA fragment.
The vast array of mutations (Mortimer and Schild, 1985) and this
very simple gene isolation procedure maintain yeast as an attractive
molecular genetic system.
3.4. Site-Directed Mutagenesis in Yeast Using

Single-Stranded Plasmids
It is often useful to be able to isolate a cloned gene as a single-
stranded DNA (ssDNA) as well as a double-stranded DNA (dsDNA). A
number of plasmids that can be propagated in E. coli as either ssDNA or
dsDNA which can also replicate in S. cerevisiae as dsDNA have been

developed by incorporating the replication and packaging signals from
filamentous bacteriophage such as fl, into the plasmid (Cesareni and
Murray, 1987). The demonstration that S. cerevisiae can be efficiently
transformed with ssDNA plasmids (Singh et al., 1982) has subsequently
led to the use of such plasmids for carrying out site-directed mutagenesis
of cloned genes directly in yeast (Burke and Olson, 1986; Walder and
Walder, 1986). In this method yeast is transformed directly with a het-
eroduplex molecule comprised of the ssDNA plasmid, carrying the gene
of choice, and a short oligonucleotide, containing the desired nucleotide
alteration, annealed to the ssDNA plasmid. No prior in vitro synthesis of
the complementary DNA strand is necessary. A further development of
strategies for site-directed mutagenesis directly in yeast has come from
the work of Moerschell et al. ( 1988), who have shown that one can mutate
genomic sequences by transforming yeast directly with the mutant
oligonucleotide. Mutants can be generated with oligonucleotides as short
as 20 nucleotides and with amounts as low as 100 ~J.g, but this has the
drawback that a direct selection method is required for isolating the
generated mutant.
4. MINICHROMOSOME VECTORS
Linear autonomous vectors can be created by assembling an ARS, a

CEN, a selectable marker, and a TEL (telemere) (Murray and Szostak,
1983b) (Fig. 3). These vectors resemble minichromosomes, although
they are substantially less stable than authentic chromosomes. However,
they appear to become more stable with increased size, making them
particularly suitable for the introduction of large DNA fragments into
yeast. In this context, the development of yeast artificial chromosome
(YAC) vectors (see Section 5) has provided the opportunity for assem-
bling whole metabolic pathways where multiple genes need to be used.
5. YAC CWNING VECTORS
The cloning of very large DNA fragments is of central importance

to the analysis of the genomes of higher eukaryotes, including the pro-
YL p4
lMill I CEN 3 I
110·7 kb)
u>
Figure 3. The yeast minichromosome YLp4. T = Tetrahymena telomeres. Open boxes =
yeast DNA. Thin line "' pBR322 DNA.
posed DNA sequencing of the human genome. Many genes from such
organisms, e.g., the human factor VIII gene and the Drosophila bithorax
genes, span enormous distances. These large genes have previously had
to be cloned as a series of overlapping cut fragments of less than 40 kb in
order to meet the size constraints that packaging in bacteriophage im-
poses. This is both laborious and unsatisfactory, in that some very large
loci may never be reliably reconstituted. However, Burke et al. ( 1987)
have developed a method using YACs to clone DNA fragments up to 500
kb. This new technique is illustrated in Fig. 4.
The circular plasmid vector pYAC2 may be propagated readily in E.
coli. Circular plasmid molecules, isolated from the bacterium, can be
cleaved into two fragments that constitute chromosome arms. The left
arm contains all the structural and functional requirements of a yeast
chromosome: a centromere, an ARS element, but only one telomere.
The right arm contains only a telomere. Each arm carries a different
yeast gene as a selectable marker; TRP1 for the left arm and URAJ for
ECORI DIGEST
__ , _,
pYAC
cut with B/ E
ligate to partial E digest
select TRP+ URA+
TRP/ARS/CEN ~A3
~~
E Insert E ·:;,·: ':
HIS3
~
Figure 4. YAC cloning vector pYAC. Restriction sites: B = BamHI; E = EcoRl. For detailed
description of the cloning procedure, see text.
the right. The cleavage site that separates the two arms lies within the
yeast SUP4 gene which encodes a tyrosine-inserting, ocher suppressor
tRNA. Large(> 100 kb) fragments of heterologous DNA may be ligated
to the two arms. Transformation of yeast with the ligation mix yields
about 100 transformant colonies per microgram of DNA. By selecting
for TRP1 and URAJ simultaneously, only molecules that have both arms
are obtained. The presence of an insert is easily detected in a host strain
carrying an ade2-ocher mutation. In YACs without an insert, SUP4 will
be functional and transformant colonies white. In YACs carrying hetero-
logous DNA, SUP4 is inactivated, the ade2-ocher allele is not suppressed,
and transformant colonies are red. The recombinant YAC molecules
appear to be stably maintained without structural rearrangement. This,
in addition to their use in analyzing large genomes, YAC vectors have
potential for introducing whole new metabolic pathways into S. cerevisiae
to greatly extend its catabolic and anabolic capabilities.
6. INTEGRATIVE SYSTEMS
The first experiment that resulted in the successful transformation

of S. cerevisiae used the simplest of integrative systems (Hinnen et al.,
1978). The plasmid used was a bacterial plasmid containing the yeast
LEU2 gene and the recipient strain contained a double mutation in
LEU2 to reduce the frequency of spontaneous reversion to LEU+.
When LEU+ transformants were obtained, it was found that, in most
cases, the entire plasmid had integrated into chromosome III via homol-
ogous recombination across the LEU2 regions (Fig. 5). A variety of inte-
CalE-1
X
leu 2
Col E-1
I LEU 2+ I ltu2
Figure 5. Integration of plasmid pYE(LEU2)10 into the yeast genome. Open box= yeast
DNA. Thin line = ColEI DNA.
grative strategies have arisen from these initial experiments. Together

they make yeast unique among eukaryotes in that chromosomal genes
can be accurately replaced by any derivative with almost total reliability.
6.1. Thrgeted Integration

The observation that a double-strand break in a plasmid-borne
yeast DNA fragment is highly recombinogenic (Orr-Weaver et al., 1981)
has led to a system for directing integration of a DNA fragment to a
specific chromosomal site. If a plasmid contains two chromosomal yeast
DNA fragments and is used to transform yeast, integration will occur at
the chromosomal location of either sequence with more or less equal
frequency. However, if a double-strand break is produced, probably by
cleavage with a restriction enzyme, in one of those fragments, then inte-
gration at the corresponding chromosomal locus will be stimulated at
least 100-fold (Fig. 6). Integration can therefore be targeted to a site by
double-strand cleavage. Either transformation procedure can be used. A
minor problem with this technique is that integration of tandem arrays is
relatively common. This can be detected easily by Southern blots (South-
ern, 1975) and can be diminished by using reduced amounts of DNA.
This technique, like the simple integrative vector system described ear-
0
110
Q
iii
Q Q
Figure 6. Targeted integration. Closed box= yeast DNA. Thin line= bacterial vector. a=
any restriction site in the yeast DNA. A cloned piece of yeast DNA containing restriction
site a (i) is cleaved at a (ii) and then used to transform yeast. Integration occurs at the point
of cleavage to generate a duplication of the cloned yeast fragment with the components of
that duplication separated by the bacterial plasmid sequences (iii).
lier, results in a duplication of the target DNA with the duplicates sepa-
rated by the remainder of the plasmid (Fig. 6).
6.2. Allelic Rescue

Having cloned a wild-type yeast gene by complementation, for ex-
ample, it is frequently the case that DNA sequence analysis of mutant
alleles of that gene will be most illuminating. Strategies have been de-
vised, therefore, to isolate mutant alleles by using targeted integration of
a plasmid vector at a site close to the mutation using homologous recom-
Q
b
ii
Q t b
+CUT WITH ENZYME b
iii
f2J
•zzaazrz
I
b
t LIGATE
b
Q
iv b
+TRANSFORM E.COLI
Figure 7. Allelic rescue I. Hatched box= cloned yeast DNA. Open box= yeast genome.
Thin line = bacterial plasmid. Black spot = mutation. a and b = restriction sites within the
cloned yeast DNA. A wild-type gene inserted into a bacterial vector is used to transform a
yeast strain that is mutant at the corresponding locus (i). Integration occurs by recombina-
tion across the yeast homologous sequences to generate a duplication of the cloned frag-
ment flanking the bacterial vector sequences (ii). Total DNA is then prepared and digested
with enzyme a and enzyme b in separate reactions. Digestion with b is shown (iii). Depend-
ing on which side of the mutation the original integration event occurred, the reaction with
either a orb will generate a fragment containing the mutant allele linked to bacterial vector
sequences (iii). The cut DNA is then ligated and used to transform E. coli selecting for a
bacterial plasmic marker (iv). Transformants are then screened, for example, for plasmids
that are unable to transform the mutant yeast strain to wild type.
bination between the wild-type and mutant genes. The resulting struc-
ture has the plasmid sequences flanked by one mutant and one wild-type
gene (Fig. 7), the precise configuration depending oi1 the crossover
point. If DNA is prepared from the resulting transformant and then
cleaved with an appropriate restriction fragment, the plasmid and mu-
tant allele sequences will be on the same fragment (Stiles et al., 1981).
Ligation followed by transformation of E. coli will give bacterial transfor-
mants containing plasmids carrying the mutant allele. Plasmids can be
screened for their inability to correct the mutation in yeast.
A more elegant technique that guarantees that the mutant allele is
recovered exploits the ability of yeast cells to carry out gap repair during
integration of a linearized plasmid (Fig. 6) (Orr-Weaver et al., 1982). In
this procedure integration of a plasmid bearing a selectable marker is
directed to the mutant gene by the presence of a double-strand break in
the plasmid-borne wild-type gene. However, the wild-type gene is not
just broken, it is also gapped at the break by using either one or more
restriction enzymes or a restriction enzyme and an exonuclease. The gap
is designed to run over the region of the mutation. When the plasmid
integrates, the gap is repaired using the mutant gene as template. The
resulting structure has the plasmid vector sequences flanked on both
·0· b
I¢ZT4"'Z0??4
c
ao
a b c d
n/ ~
I(?ZfUZ(ZVf VW{ZV<ZZZ(II
d a~ c ati c d
b c
Figure 8. Allelic rescue by gap repair. Hatched box = cloned yeast DNA. Open box =
yeast genome. Thin line = bacterial vector DNA. Black spot = mutation. a, b, c, and dare
restriction sites in the cloned yeast DNA. A gap is created in the cloned DNA over the
region of the desired mutation by cleavage at b and c (i). The resulting linear molecule is
used to transform a yeast strain carrying the desired mutant allele (i). Gap repair may
occur directly (ii) and if the gapped plasmid is potentially an autonomous yeast plasmid the
mutant allele will be transferred to and maintained on this autonomous molecule, or gap
repair may occur as a function of integration (iii). However, unlike the situation in Fig. 7,
digestion with any of the enzymes a, b, c, or d will give a fragment that carries the mutant
allele. If the digested DNA is ligated and used to transform E. coli, all transformants will
contain the desired allele.
sides by a mutant gene. Appropriate cleavage of the transformant DNA

followed by ligation and E. coli transformation results in recovery of the
mutant allele with I 00% efficiency.
The gap repair strategy can also be used to "transfer" a mutant
allele to an autonomous plasmid that has been gapped (Orr-Weaver and
Szostak, 1983; Orr-Weaver et al., 1982). If an unstable ARS plasmid is
used, the gapped plasmid and therefore its selectable marker can only
survive by gap repair, either via integration or not. Transformants with
autonomous repaired plasmids are easily identified, as the selectable
marker will show marked instability (Fig. 8). This also has the advantage
that recovery of the plasmid carrying the mutant allele is achieved by
transformation of E. coli directly with total yeast DNA preparations.
6.3. Gene Disruption

Using a cloned yeast gene to disrupt its chromosomal counterpart is
an important procedure. It can be used to assess the phenotype of a null
mutation or to confirm that a cloned DNA fragment confers a pre-
viously identified phenotype. A simple procedure is to insert a fragment,
generated by restriction enzymes or exonucleases, that comes from with-
in the cloned gene, into an integrative vector (Shortie et al., 1982). Inte-
s' 3'
0
5' X
::::=,~==j~~~~f~~~~~~~?~~[:::;,::::::::ICHR
3'
+
a b c d e
iii
s' 3'
vzqzz{l4 fV441A I
a b c d b c d e
Figure 9. Gene disruption. Open and hatched boxes = yeast DNA. Hatched box = a gene
within the cloned fragment a-e (i). Thin line = bacterial vector sequences. CHR = chro-
mosome. a, b, c, d, and e = restriction sites in the cloned yeast DNA. The chromosomal
copy of the hatched gene is to be disrupted. The internal fragment b-d is subcloned by
inserting it into a bacterial vector. It is then cleaved at c and used to transform a yeast strain
(ii). Integration occurs at c to create a duplication of fragment b-d but with the 5' and 3'
ends of the hatched gene separated by one copy of b-d and the bacterial vector (iii).
gration is then directed into the chromosomal copy of the gene by mak-
ing a double-strand break in the fragment (Fig. 9). This creates a
duplication of the cloned fragment in the genome, but the 5' and 3' ends
of the gene are separated by the vector sequences. The gene will there-
fore be nonfunctional. The only major disadvantage of this technique is
that the disruption will be unstable because of the duplication.
A more satisfactory procedure involves replacement of the genomic
copy of a gene by mutated derivative. The general name for this is "gene
transplacement" (Scherer and Davis, 1979). For example, a small dele-
tion may be introduced into the cloned gene using restriction enzymes
and an exonuclease. The mutated derivative is then inserted into an
integrative vector and forced to recombine with its genomic comple-
ment. Integration can occur on either side of the deletion to produce a
duplication of the gene-flanking vector sequences. One copy of the gene
will be wild type, the other will carry the deletion. The transformant will
also bear the phenotype determined by the selectable marker. A second
recombination event on the opposite side of the deletion to the inte-
grative event will resolve the duplication, eliminate the selectable mark-
er, and leave the mutant derivative in place of the wild-type gene (Fig.
10). Scoring for loss of the selectable marker permits identification of
clones that have undergone a second recombination event. The position
of that event with respect to the deletion can be determined either by the
phenotype confered by mutant alleles or by appropriate Southern hy-
bridization experiments. URA3 is frequently used as a selectable marker
for this type of approach as its loss can be selected by growth on 5-
fluoroorotic acid (Boeke et al., 1984). The disadvantages of this tech-
nique are that two steps are required and that 5-fluorooritic acid is very
expensive.
A particularly elegant and convenient disruption procedure makes
use of the recombinogenic properties of "ends" to insert a selectable
marker into a gene (Rothstein, 1983). The key feature of this approach is
that the marker provides both selection and disruption. A selectable
marker is first inserted into a cloned gene in vitro. This can be a simple
insertion, where a marker fragment is inserted at a suitable restriction
site, or it can be a replacement, where the marker fragment replaces a
region of the cloned gene. In the former this would create an insertion
mutation; in the latter, an insertion/deletion mutation. A relatively large
fragment comprising ends that are homologous with the target gene
flanking the selectable marker is then isolated, usually from an agarose
gel. This fragment is then used to transform a recipient selecting for the
marker. Recombination with the chromosomal copy takes place across
the ends of the fragment and so the chromosomal copy is replaced, in
one step, by a derivative that is disrupted. This disruption links a selecta-
ble marker to the gene in question and it is nonrevertable (Fig. 11).
a
II
iii
CHR
iv
CHR
I
a b (cldl e
v CHR
+
a b (~) e
vi
I '0(41
a b (cAll e
Figure 10. Gene transplacement. Open and hatched boxes= yeast DNA. Hatched box = a
gene present on a cloned fragment. URA3 = URA3 gene as the selectable marker. Thin
line = bacterial vector sequences. CHR = chromosome. a, b, c, d, e, and f = restriction sites
within the cloned yeast DNA. The hatched gene in the genome of a yeast strain is to be
ii a
iii I
a'\..
•
~URA3~
d"c
I
I ~ CHR
a
iv
~
f??';:l URA 3 ~ CHR
a
Figure 11. Marker-mediated gene disruption. Hatched and open boxes = yeast DNA.
Hatched box = gene present on a cloned yeast DNA fragment. Thin line = bacterial vector
sequences. CHR = chromosome. URA3 = the URA3 gene as a selectable marker. a, b, and
c are restriction sites within the cloned yeast DNA fragment. A disrupted version of the
hatched gene is to replace its genomic copy. The gene is present on an integrative vector (i).
It is disrupted by insertion of the URA3 marker at b (ii). The resulting plasmid is cleaved at
a and c and the fragment a-c is purified. This fragment is used to transform a yeast strain
selecting for URA +. Recombination occurs at a and c (iii) to replace the genomic copy of
the gene by a disrupted derivative (iv).
6.4. Refinements of Transplacement

The procedure described in Section 6.3 is a convenient way to
achieve gross disruption of a yeast gene. However, one of the most
common modern experimental strategies in yeast molecular biology is to
make relatively small, subtle changes to a gene and then ask if that change
replaced by a deleted version. The hatched gene is present on a cloned fragment in an

integrative vector carrying URA3 as the selectable marker (i). The deleted derivative of the
gene is created by removing fragment c-d (ii). The plasmid containing the deletion is then
cleaved at b and used to transform a yeast strain to URA +. Integration occurs at b (iii) to
create a duplication, one component of which carries the mutation and the other does not
(iv). Excision of the vector sequences and one component of the duplication by recombina-
tion (v) can be selected on 5-fluoroorotic acid. Depending on which side of the mutation
the excision event occurs, either the wild-type or the mutant allele will be left in the
genome. The mutant allele is shown (vi). This can be tested either by the resulting phe-
notype or by Southern hybridization (see text).
affects function. The mutations can vary from single nucleotide changes
to relatively small deletions. Clearly, it is common for there to be no easy
way to select for replacement of a wild-type gene by a mutant gene, as
there is when a selectable marker is the mutagenic agent. This is a
particular problem when, as is usually the case, multiple versions of the
mutant gene need to be analyzed.
A variety of conceptually similar tricks have been devised to solve
these problems. They all involve marking the wild-type or resident gene
in such a way that its loss is selectable. One of the first procedures to be
described uses a recessive drug resistance marker (Struhl, 1983). First, the
strain to be used is made resistant to cycloheximide by either crossing in or
isolating a mutation at the cyh2 locus. The gene to be analyzed is then
disrupted, by a method described in the previous section, with the wild-
type CYH2 gene. This strain, now sensitive to cycloheximide, is the
starting strain for all further transplacements. Mutant derivatives of the
,Q• +
ii
,Q· +
iii-
/Q
I ~
J, CYH 5
~ t:l:l:l s ~ 'RR
A ~
iy ~ cyh R
~ CHR
Q
I
~
Figure 12. Selected transplacement. Open and hatched boxes = yeast DNA. Hatched box
= a gene present on a cloned yeast DNA fragment. Thin line = bacterial vector sequences.
CHR = chromosome. CYH = CYH2 gene. Black spot = mutation. a and b = restriction
sites within the cloned yeast DNA fragment. The hatched gene in the genome is to be
replaced by derivative containing a mutation. The hatched gene is present on a cloned
fragment in a bacterial vector (i). A mutation in that gene is created in vitro (ii). The
resulting plasmid is cut at a and band the fragment a-b is purified and used to transform a
yeast strain selecting for cycloheximide resistance (iii). This strain has already undergone a
mating to introduce a cyhr mutation at the CYN2 locus and then a gene disruption at the
hatched locus using the wild-type CYH• gene. The strain is therefore sensitive to cyclohexi-
mide by virtue of this insertion at the hatched gene. When fragment a-b recombines with
its disrupted genomic counterpart, it replaces it and the transformant becomes cyclohexi-
mide resistant (iv).
gene to be analyzed are introduced into this strain by transformation.

They replace the CYH2 marked resident gene by recombination and the
cell becomes resistant to cycloheximide. The transplacement can there-
fore be selected. (See Fig. 12.) Other similar strategies have been devised
using a suppressable canavanine resistance marker (Nasmyth, 1985) or
the ability to select ura3 mutants on 5-fluoroorotic acid (Rothstein, 1985).
A less sophisticated and perhaps more convenient technique makes
use of the high frequency with which cotransformation of two DNA
molecules occurs in yeast (Rudolph et al., 1985). In this procedure a
mutant DNA fragment is used to replace the resident gene directly. Yeast
cells that have taken up DNA in the transformation process are chosen by
selecting for a marker present on a plasmid that is mixed with the mutant
fragment. Typically between 1 and 10% of plasmid-bearing transfor-
mants also carry the trans placement. If the resident gene is marked either
by the phenotype that it confers or by any other marker that can be
scored, then transformants carrying the transplacement are easily
detected.
7. SELECTION SYSTEMS
Many of the procedures described in this chapter depend on the use

of easily selectable markers. Auxotrophic markers are usually used for
laboratory strains as they are very easy and cheap to use. However, there
is increasing interest in manipulating industrial strains that may not have
appropriate mutations because of their history, or because they are of
dubious ploidy, or because the process for which they are used precludes
any other genetic background. A typical example would be strains used
in the food or brewing industries. In these cases it is necessary to mark
any DNA molecule that is to be used with a dominant selectable marker.
Two systems are most commonly used. The first is the copper-thionein
gene of S. cerevisiae, CUP 1, which confers resistance to copper (Butt et
al., 1984). This gene can be included in any of the types of vector de-
scribed and selection is achieved by plating on medium containing cop-
per (Henderson et al., 1985). The second commonly exploited system
uses a bacterial transposon (Tn601) gene encoding aminoglycoside
phosphotransferase (Jiminez and Davies, 1980). This enzyme inactivates
the antibiotic G418 which is active against yeast cells. G418 resistance is
most effective when the gene is expressed from a yeast promoter (see
Chapter 7). The only disadvantage of the system is that relatively high
concentrations of G418 are required and it is expensive.
AcKNOWLEDGMENTS. We thank Frank Caddick for excellent artwork and

photography. M.J.D. is a MRC Senior Research Fellow. Work in the Ox-
ford laboratory was supported by MRC, SERC, AFRC, the Royal Society
of London, the Nuffield Foundation, Celltech Ltd. and Delta Biotech-

nology Ltd.
REFERENCES
Beggs,J. D. 1978, Transformation of yeast by a replicating hybrid plasmid,Nature, 275: I 04-
109.
Boeke, J.D., Lacroute, F., and Fink, G. R. 1984, A positive selection for mutants lacking
orotidine-5'-phosphate decarboxylase activity in yeast: 5-Fluoro-orotic acid resistance,
Mol. Gen. Genet. 197:345-346.
Broach, J. R. 1982, The yeast plasmid 2 J.l. circle, Cell 28:203-204.
Burke, D. T., and Olson, M. V. 1986, Oligo-directed mutagenesis of E. coli and yeast by
simple co-transformation of the primer and template, DNA 5:325-332.
Burke, D. T., Carle, G. F., and Olsen, M. V. 1987, Cloning of large segments of exogenous
DNA into yeast by means of artificial yeast chromosome vectors, Science 236:806-813.
Butt, T. R., Sternberg, E. j., Gorman, J. A., Clark, P., Hamer, D., Rosenberg, M., and
Crooke, S. T. 1984, Copper metallothionein of yeast: Structure of the gene and
regulation of expression, Proc. Natl. Acad. Sci. USA 81:3332-3336.
Cesareni, G., and Murray, J. A. H. 1987, Plasmid vectors carrying the replication origin of
filamentous single-stranded phages, in: Genetic Engineering, Vol. 9 (J. K. Setlow, ed.),
Plenum Press, New York, pp. 137-144.
Clarke, L., and Carbon, J. 1976, A colony bank containing synthetic Col E1 hybrid plas-
mids representative of the entire E. coli genome, Cell9:91-99.
Clarke, L., and Carbon, J. 1980, Isolation of a yeast centromere and construction of
functional small circular chromosomes, Nature 257:504-509.
Glover, D. M. 1985, DNA Cloning. Vol. 1111. A Practical Approach, IRL Press, Oxford.
Henderson, R. C. A., Cox, B.S., and Tubb, R. 1985, The transformation of brewing yeasts
with a plasmid containing the gene for copper resistance, Curr. Genet. 9:133-138.
Hinnen, A., Hicks, J. B., and Fink, G. R. 1978, Transformation of yeast, Proc. Natl. Acad.
Sci. USA 75:1929-1933.
Ito, H., Fukuda, Y., Murata, K., and Kimura, A. 1983, Transformation of intact yeast cells
treated with alkali cations,]. Bacterial. 153:163-168.
jayaram, M., Li, Y.-Y., and Broach, J. R. 1983, The yeast plasmid 2 1.1. circle encodes
components required for its high copy number propagation, Cell 34:95-104.
Jiminez, A., and Davies, J. 1980, Expression of a transposable antibiotic resistance element
in Sacch!Jromyces, Nature 287:869-871.
Kingsman, A. J., Clarke, L., Mortimer, R. K., and Carbon, J. 1979, Replication in Sac-
ch!Jromyces cerevisiae of plasmid pBR313 carrying DNA from the yeast TRPJ region,
Gene 7:141-152.
Maniatis, T., Fritsch, E. F., and Sambrook, J. 1982, Molecular Cloning. A Laboratory Manual,
Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
Moerschell, R. P., Tsunasawa, S., and Sherman, F. 1988, Transformation of yeast with
synthetic oligonucleotides, Proc. Natl. Acad. Sci. USA 85:524-528.
Mortimer, R. K., and Schild, D. 1985, Genetic map of Sacch!Jromyces cerevisiae, edition 9,
Microbial. Rev. 49:181-213.
Murray, A. W., and Swstak, J. W. 1983a, Pedigree analysis of plasmid segregation in yeast,
Cell 34:961-970.
Murray, A. W., and Szostak,j. W. 1983b, Construction of artificial chromosomes in yeast,
Nature 305:189-193.
Nasmyth, K. 1985, At least 1400 base pairs of 5'-flanking DNA is required for the correct
expression of the HO gene in yeast, Cell42:213-223.
Nasmyth, K., and Reed, S. I. 1980, Isolation of genes by complementation in yeast: Mo-
lecular cloning of a cell-cycle gene, Proc. Natl. Acad. Sci. USA 77:2119-2123.
Nasmyth, K., and Tatchell, K. 1980, The structure of transposable yeast mating-type loci,
Cell19:753-764.
Orr-Weaver, T. L., and Szostak, W.J. 1983, Yeast recombination: The association between
double-strand gap repair and crossing-over, Proc. Natl. Acad. Sci. USA 80:4417-4421.
Orr-Weaver, T. L., Szostak, J. W., and Rothstein, R. J. 1981, Yeast transformation: A model
system for the study of recombination, Proc. Natl. Acad. Sci. USA 78:6354-6358.
Orr-Weaver, T. L., Szostak, J. W., and Rothstein, R. J. 1982, Genetic applications of yeast
transformations with linear and gapped plasmids. In: Methods in Enzymology, Vol. 101
(R. Wu, L. Grossman, and K. Moldave, eds.), Academic Press, New York, pp. 228-245.
Parent, S. A., Fenimore, C. M., and Bostian, K. A. 1985, Vector systems for the expression,
analysis and cloning of DNA sequences inS. cerevisiae, Yeast 1:83-138.
Ratzkin, B., and Carbon, J. 1977, Functional expression of cloned yeast DNA in E. coli,
Rothstein, R. J. 1983, One-step gene disruption in yeast, in: Methods in Enzymology, Vol. 101
(R. Wu, L. Grossman, and K. Moldave, eds.), Academic Press, New York, pp. 202-211.
Rothstein, R. J. 1985, Cloning in yeast, in: DNA Cloning. Vol. II. A Practical Approach (D. M.
Glover, ed.), IRL Press, Oxford, pp. 45-66.
Rudolph, H., Koenig-Rauseo, 1., and Hinnen, A. 1985, One-step gene replacement in
yeast by cotransformation, Gene 56:87-95.
Scherer, S., and Davis, R. W. 1979, Replacement of chromosome segments with altered
DNA sequences constructed in vitro, Proc. Natl. Acad. Sci. USA 76:4951-4955.
Shortie, D., Haber, J. E., and Botstein, D. 1982, Lethal disruption of the yeast actin gene by
integrative DNA transformation, Science 217:371-373.
Singh, H., Bieker, J. j., and Dumas, L. B. 1982, Genetic transformation of Saccharomyces
cerevisiae with single-stranded circular DNA vectors, Gene 20:441-449.
Southern, E. M. 1975, Detection of specific sequences among DNA fragments separated by
gel electrophoresis, ]. Mol. Biol. 98:503-517.
Stiles, J. 1., Szostak, J. W., Young, A. T., Wu, R., Consaul, S., and Sherman, F. 1981, DNA
sequence of a mutation in the leader region of the yeast iso-1-cytochrome mRN A, Cell
25:277-284.
Stinchcomb, D. T., Struhl, K., and Davis, R. W. 1979, Isolation and characterisation of a
yeast chromosomal replicator, Nature 282:39-43.
Struhl, K. 1983, The new yeast genetics, Nature 505:391-396.
Struhl, K., Cameron, J. R., and Davis, R. W. 1976, Functional genetic expression of eu-
karyotic DNA in E. coli, Proc. Natl. Acad. Sci. USA 75:1471-1475.
Struh1, K., Stinchcomb, D. T., Scherer, S., and Davis, R. W. 1979, High frequency transfor-
mation of yeast; autonomous replication of hybrid molecules, Proc. Natl. Acad. Sci.
USA 76:1035-1039.
Walder, R. Y., and Walder, J. A. 1986, Oligodeoxynucleotide-directed mutagenesis using
the yeast transformation system, Gene 42:133-139.
Williamson, D. H. 1985, The yeastARS element, six years on: A progress report, Yeast 1:1-
14.
Expression of Heterologous
Genes
6
MICHAEL F. TUITE
1. INTRODUCTION
With the advent of recombinant DNA technology in the mid-1970s a

number of shrewd academics quickly realized the commercial potential
of these tools in producing proteins of high commercial value, which,
until that time, had proven both difficult and expensive to isolate in
quantities sufficient for clinical trials. While Escherichia coli as a host
organism was an obvious choice in the early stages of commercial genetic
engineering (Harris, 1983), it soon became apparent that it was far from
the ideal host. It did not make sense, for example, to expect a eukaryotic
gene product to be expressed in a prokaryotic environment and be cor-
rectly folded and modified at the posttranslationallevel. So by 1980 the
search was underway to find a more suitable host and the budding yeast
Saccharomyces cerevisiae immediately presented itself as an attractive alter-
native. It is a host that is widely accepted for human consumption by
both the consumer and the regulatory authorities, has no associated
toxins, and, with the development of an effective plasmid-based gene
cloning system (see Chapter 5), its exploitation as a host for producing
high-value proteins has progressed rapidly.
Inserting a heterologous gene into a high-copy-number vector and
introducing it by transformation into yeast is a relatively straightforward
exercise, but such a manipulation does not guarantee even minimal
expression of the heterologous gene. To ensure successful and efficient
transcription of the heterologous gene it must be inserted into a plasmid
immediately adjacent to a DNA sequence that can effectively bind yeast
RNA polymerase II, i.e., a yeast promoter. Once transcribed into mRNA,
the next hurdle is to get yeast ribosomes to efficiently initiate upon, and to
MICHAEL F. TUITE • Biological Laboratory, University of Kent, Canterbury, Kent

CT2 7NJ, England.
169
170 MICHAEL F. TUITE
accurately translate, the mRNA into the encoded amino acid sequence.
Yet success at this second stage of expression still does not guarantee the
ultimate goal, namely, to produce high levels of the fully matured hetero-
logous gene product. For example, many eukaryotic proteins need, to a
greater or lesser extent, a degree of posttranslational modification or
processing to ensure biological activity, and we must also ensure that the
polypeptide, once synthesized, is not then rapidly degraded by host
proteolytic activities. We must also minimize or circumvent potential
deleterious effects that the heterologous polypeptide may present to the
general welfare of the yeast host by coaxing the yeast to secrete the
polypeptide out into the surrounding growth medium. Finally, we must
ensure that the plasmid system being exploited (be it high copy number or
low copy number) is stably maintained both under short-term batch
growth conditions and in the more long-term continuous culture systems,
particularly where high-level synthesis of the foreign polypeptide is
desirable.
The aim of this chapter is to provide answers, where possible, to a
number of questions normally.raised when one first sets out to explore
the possibility of expressing a heterologous gene in yeast, namely;
1. What are the best plasmid constructions for introducing a het-
erologous gene into yeast?
2. How can one ensure that the gene will be transcribed and its cor-
responding mRNA translated in its new environment?
3. What steps can one take to ensure maximum yield of the gene
product?
4. What advantages and disadvantages (if any) are associated with
using this microorganism as a host?
Table I can be used as a simple guide to the basic manipulations
Table I. Expression of a Heterologous Gene in Yeast:

A Step-by-Step Guide
Step 1 Clone heterologous gene into an appropriate expres-

sion vector
Step 2 Introduce vector into yeast via transformation
Step 3 Transcription of the gene and transport of mRNA
from the nucleoplasm to the cytoplasm
Step 4 Initiation of translation of mRNA
Step 5 Elongation of polypeptide chain
Step 6 Posttranslational modification of product
Step 7a or Intracellular accumulation of protein
Step 7b Removal of signal sequence during transfer into endo-
plasmic reticulum, followed by
Step 8 Secretion into surrouding medium
EXPRESSION OF HETEROLOGOUS GENES 171
required to achieve successful expression of a heterologous gene m

yeast, which will be discussed in detail in the following sections.
2. INTRODUCTION OF HETEROWGOUS DNA INTO YEAST
2.1. Basic Cloning Technology

The routes that can be adopted in cloning and manipulating both
prokaryotic and eukaryotic genes have been well documented in a
number of "laboratory manuals" (e.g., Maniatis et al., 1982; Dillon et al.,
1985 ). Similarly, the structure and fundamental characteristics of basic
cloning vectors, together with transformation systems for introducing
cloned genes into yeast, have been described elsewhere in this volume
(Chapter 5) and will not be dealt with in detail here.
2.2. Basic Vector Design

A plasmid for introducing heterologous DNA into yeast should ( 1) be
directly selectable in yeast, (2) be replicated autonomously in yeast, (3)
possess a high copy number, (4) be able to replicate in E. coli, and (5) be
directly selectable in E. coli. To suit these properties such a plasmid should
be either of the YRp or the YEp category, depending on the yeast origin of
replication required (see Chapter 5). What are the options available for
the various plasmid encoded sequences? Particularly, which sequences
have proven useful to date?
2.2.1. Selectable Markers

The majority of basic yeast cloning plasmids carry wild-type genes
that are capable of functionally complementing corresponding aux-
otrophic mutations in the host strain (Table II). However, one might
want to introduce vectors into genetically undefined strains (e.g., poly-
ploid sterile brewing strains) or into another Saccharomyces species for
which there is no appropriate complementable genetic marker. In addi-
tion, in large-scale batch culture, where one wants to maintain selective
pressure for retention of the plasmid, it is prohibitively expensive to use
defined medium lacking the required nutrient. These needs have led to
the development of a series of dominant selectable markers (Table II)
that require little or no manipulation of the host genotype and can be
selected for in rich growth medium.
With the economic and regulatory constraints placed on the use of
antibiotic markers, the use of the CUPJ gene as a selectable marker
offers great potential. A multicopy plasmid carrying the CUP 1 gene can
Table II. Selectable Markers That Can Be Used in Yeast Cloning

and Expression Vectors
Host
Selectable gene genotype Selection medium• Reference
Auxotrophic selection
LEU2 leu2 CDM-leu Kingsman et al., this
volume
HIS3 hisJ CDM-his Kingsman et al., this
volume
URA3 uraJ CDM-ura Kingsman et al., this
volume
TRP1 trpl CDM-trp Kingsman et al., this
volume
Dominant selection
CUP I Wild type URM + CuS04 Henderson et al.,
1985
HSV-TK Wild type URM + thymidine + McNeil and Freisen,
(thymidine kinase) amethopterin + 1981; Zhu et al.,
sulfanilamide 1984
R-dhfr Wild type URM + sulfanilamide Miyajima et al.,
(E. coli + amethopterin 1984
dihydrofolate
reductase)
M-dhfr Wild type URM + sulfanilamide Zhu et al., 1985
(mouse + methotrexate
dihydrofolate
reductase)
hph Wild type URM + hygro- Kaster et al., 1984
mycin B
Kmr Wild type URM + G418 Webster and Dick-
son, 1983
M I killer toxin K-: killer URM pH 4.6 Bussey and Mead-
immunity gene sensitive en, 1985
•CDM = complete defined medium; URM = undefined rich medium.
be used to directly select for transformants in copper-sensitive brewing

strains (Henderson et al., 1985) by plating directly into regeneration
medium containing appropriate concentrations of CuS04 , usually 0.5-1
mM. Selection can then be maintained by growing transformants in the
presence of this cheap, relatively nontoxic salt.
2.2.2. Origins of Replication

For a plasmid to be able to replicate autonomously in yeast it must
contain an origin of DNA replication. Such sequences have been ob-
tained from several different sources: (1) the endogenous 2-~J.m circle
plasmid which carries a single origin of replication (designated ORI), (2)
EXPRESSION OF HETEROWGOUS GENES 173
sequences from yeast chromosomal DNA, (3) sequences from yeast mito-
chondrial DNA, and (4) sequences from heterologous DNA.
In cases 2-4 these sequences are defined purely on the criteria that
they confer the ability of autonomous replication in yeast and are re-
ferred to as autonomous replicating sequences (ARS) (see Williamson,
1985). Plasmids using an homologous ARS or the 2-JJ.m ORI have gener-
ally been incorporated in vectors for introducing foreign genes into
yeast. ARS-based plasmids have a high copy number, in the order of 30-
50 copies per cell, as do the majority of ORI-based vectors (but see
Section 6.1 ). A major problem with the majority of 2-JA.m ARS plasmids is
that they are highly unstable even when selective pressure is maintained
(see Section 6.1).
Almost in variably the E. coli origin of replication is of the co 1E 1 type
derived from pBR322 or one of its close relatives (pBR325, pBR328,
pAT153, etc.). Similarly, the selectable marker for transfer and manip-
ulation in E. coli is generally the ampicillin-resistance and/or the tetracy-
cline-resistance determinant derived from the pBR family of plasmid
vectors.
3. TRANSCRIPTION OF HETEROLOGOUS GENES
It is unusual to observe transcription of a heterologous gene in yeast

unless that gene is linked to a yeast promoter. In addition, a number of
other considerations have to be taken into account to ensure that tran-
scription is both efficient and accurate.
3.1. The Problem of Introns

Many of the genes from eukaryotes whose expression is sought in
yeast naturally contain noncoding introns. While S. cerevisiae has few
genes that contain introns, e.g., genes for actin and ribosomal proteins,
recent evidence has indicated that yeast has a strict requirement for a
sequence located within the intron in order to facilitate splicing of the
intron from the transcribed RNA (Langford and Gallwitz, 1983). This
sequence (5' TACTAACA 3') is located between 20 and 55 nudeotides
upstream of the 3' splice site of all yeast introns and is absent from most
introns of higher eukaryotic genes. Consequently, although yeast is
dearly capable of transcribing eukaryotic, DNA sequences, it is unable to
excise the introns posttranscriptionally (Beggs et al., 1980; Langford et
al., 1983; Handa et al., 1985) unless they fortuitously contain this se-
quence (Woudt et al., 1985). A further problem can arise if the hetero-
logous intron contains a sequence capable of functioning as a signal for
transcription termination in yeast, as observed by Beggs et al. ( 1980) in
their efforts to express a genomic (intron-containing) copy of the rabbit

~-globin gene in yeast.
The problem posed by heterologous introns can be effectively over-
come by using eDNA copies of the gene of interest (which therefore
lacks introns), and since the normal route for cloning higher eukaryotic
genes is via eDNA, this generally presents no great difficulty.
3.2. Encoded Pre-Pro Sequences

If a protein whose expression is sought in yeast is a naturally secreted
product, then it will be synthesized with a signal or "pre"-sequence that
will be encoded by a eDNA copy of the gene. In addition, some of these
proteins have additional amino-terminal or carboxy-terminal "pro" -se-
quences that are removed posttranslationally in order to activate the
mature product in yeast; these prosequences are also encoded by the
gene. Do these additional sequences affect the transcription and ultimate
expression of the gene product? Detailed analysis of the expression of the
bovine chymosin gene in yeast (Mellor et al., 1983), a naturally secreted
product with an additional prosequence, clearly indicates they can do so.
Using a complete eDNA copy of the chymosin gene, variants were con-
structed in which the presequence had been replaced by a methionine
codon (to give met-prochymosin) or in which both the pre- and prose-
quence had been replaced by a methionine codon (to give met-chymosin)
(Fig. 1). The expression of these two variants was compared to that
achieved with the intact preprochymosin gene, and significant dif-
ferences were found with maximum levels being achieved with the met-
prochymosin gene (5% total cell protein) and minimal levels with the met-
chymosin gene (<0.1% total cell protein). Preprochymosin was expressed
at levels up to 1% of total cell protein.
The need to retain the presequence to allow efficient expression of a
heterologous gene in yeast was further confirmed by the efforts of Edens
et al. ( 1984) to express the gene coding for the sweet-tasting protein
thaumatin. They observed 20-fold higher levels of expression if the
encoded presequence was retained over that observed for an engineered
met-thaumatin gene. Presumably, in the case of both chymosin and
thaumatin, retention of the presequence facilitates its translation on the
ribosomes of the yeast endoplasmic reticulum where the presequence is
removed after passage through the membrane.
In contrast to these observations, however, Hitzeman et al. (1983b)
noted that expression of the human interferon genes IFNal and IFNa2
was significantly less if the associated presequences were retained, but
their data were based on levels of biologically active IFN detectable in or
outside the yeast cell rather than direct measurements of the levels of
EXPRESSION SITE
(Bg/11}
2pm(R] PfiK J
Ap'
b 3&1
A,_ro_ _ _ _ _ _ _ ____. PRE-PRO-
f-------------------~~-
t--------t HATURE·
c
s' 3' -t. TCP
I I
%'= • 0·5-1
ATG
2-5
ATG
' I
< 0·1
ATG
Figure 1. The expression of different derivatives of calf chymosin in yeast: the influence
of the pre- and prosequences on levels of expression. (a) The yeast expression plasmid
pMA91 (Mellor et al., 1983) that contains the LEU2 and 2-tJ.m circle origin of replication
(open box) and the 5' promoter and 3' terminator sequences of the gene encoding phos-
phoglycerate kinase (PGK; closed box). The heterologous gene is "sandwiched" between
these control sequences by insertion into the unique Bglll "expression site." "' indicates
mRNA start and polyadenylation sites. (b) Naturally occurring derivatives of calf
chymosin: the precursor polypeptide preprochymosin, which during secretion loses the
16-amino-acid signal peptide (stippled box) to form the zymogen prochymosin. At low pH
the 42-amino-acid prosequence (hatched box) is removed via an autocatalytic cleavage
process to produce the 323-amino-acid mature chymosin polypeptide that has an aspartyl
proteinase activity. (c) Levels of expression of the three different derivatives of chymosin in
yeast, represented as percentage of total cell protein in transformed strains (% TCP). For
details of construction of the DNA fragments, see Mellor et al. ( 1983).
IFN-related proteins, and, as the authors suggest, pre-IFN could lack

biological activity.
These data suggest that if one wants to express a naturally secreted
heterologous protein intracellularly in yeast, care must be taken to ex-
amine the levels of expression obtained both with the native full-length
gene and with a truncated version lacking any potential amino-terminal

pre- and/ or prosequences.
3.3. Transcriptional Promoters

Optimal transcription of a heterologous gene in yeast requires that
it be coupled to a yeast transcriptional promoter. Yeast promoters con-
tain many of the canonical sequences that are common to higher eu-
karyotic promoters, although the locations of these sequences, namely
the "TATA" box and the "CAAT" box, with respect to the mRNA start
site, are much more heterogeneous than observed in higher eukaryotes
(for review, see Struhl, 1986). Yeast genes also contain upstream cis-
acting elements (upstream activator sequences; UAS) that modulate
transcription and appear to be analogous to higher eukaryotic "en-
hancer" elements (Guarente, 1984). The basic features of a yeast tran-
scriptional promoter and translation sequence requirements are shown
in Fig. 2.
Given the similarity of transcription signals between yeast and high-
er eukaryotes, it is somewhat surprising that S. cerevisiae is largely unable
to efficiently and accurately initiate transcription on a heterologous eu-
karyotic promoter. Examples of aberrant initiation on a heterologous
promoter include the rabbit ~-globin gene (Beggs et al., 1980), the Dic-
tyostelium discoidin I gene Qellinghaus et al., 1982), the gene from Dros-
ophila capable of complementing the yeast ade8 mutation (Henikoff and
Furlong, 1983), the Drosophila hsp70 heat shock gene (Nicholson and
a 5~NtJN.
1-·------ ---- PROHOTtR- ----- ------ --~+-COO/Nfi REGION
40-120 -1
1- 5-14 -+- 25-50 •l
·:~-:·..::=.:.=::: (i':.'.~i ~·
'if,,,.
b
Figure 2. The basic sequence features of an efficient yeast transcriptional promoter and its
corresponding mRNA. (a) The promoter region contains the following elements: UAS =
upstream activating sequence; TATA = consensus sequence common to all eukaryotic
promoters and implicated in controlling the fidelity of transcription initiation; CnT n =
pyrimidine (cytosine + thymine) rich tract between 15 and 30 nucleotides in length; CAAG
= consensus sequence at or adjacent to which mRNA start site maps; ATG = initiation of
translation codon. The approximate distance in nucleotides between these elements in
indicated. (b) The 5' end of the corresponding mRNA m7~ = methylated cap; Pu =
purine; AUG = initiation of translation codon.
Moran, 1984), and the bean phaseolin gene (Cramer et al., 1985). Even
the promoters of a number of genes from the fission yeast Schizosac-
charomyces pombe are not correctly recognized by S. cerevisiae RNA poly-
merase II (Russell, 1983; Losson and Lacroute, 1983).
There are a number of examples of bacterial 5' noncoding se-
quences functioning as promoters in yeast, e.g., those associated with the
bla and lac genes (Breunig et al., 1982), but in both these cases the precise
5' ends of the mRNA transcripts synthesized in yeast were not mapped
and it is therefore not completely clear whether yeast RNA polymerase
II initiates at or near the E. coli promoter.
There are at least three reported examples of yeast RNA poly-
merase II successfully recognizing the natural transcription initiation
site of a heterologous gene, namely, for the maize storage protein genes
coding for the 19-kDa and 21-kDa polypeptides (Langridge et al., 1984),
the French bean phaseolin gene (Cramer et al., 1985), and the soybean
leghemoglobin Lbc3 gene Oensen et al., 1986).
It is therefore unwise to expect that the natural promoter of a
Table III. Homologous Promoter Sequences Used to Direct the Expression

of Foreign Genes in Yeast
Gene Gene product Efficiency• Reference
ADH1 Alcohol dehydrogenase I High Ritzman et al. (1981);

Ammerer (1983)
ENO Enolase High Innis et al. (1985)
PGK 3-Phosphoglycerate kinase High Tuite et al. ( 1982); Hitz-
man et al. (1983a)
TPI Triose phosphate isomerase High Smith et al. (1985); Zhu
et al. ( 1985)
GAP Glyceraldehyde-3-phosphate de- High Urdea et al. (1983)
hydrogenase
GALI,IO Galatokinase/galactose epimerase Medium Goff et al. (1984); Step-
ien et al. (1983)
PH05 Alkaline phosphatase Medium Miyanohara et al. (1983);
Rosenberg et al. ( 1984)
MFal a-Mating factor Medium Bitter et al. (1984); Singh
et al. ( 1984)
ARG3 Ornithine carbamoyl transferase Low Cabezon et al. (1984)
TRPI N (5'-phosphoribosyl)-anthianilate Low Dobson et al. (1983)
isomerase
TRP5 Tryptophan synthase Low Miyajima et al. (1984)
URA3 Orotidine-5'-phosphate decar- Low Smith et al. (1985)
boxy lase
ACT! Actin Low Zhu et al. (1985)
CYCI !so-l-cytochrome C Low Gritz and Davies (1983)
•Defined by relative mRNA concentration; high> 1% total mRNA; medium, 0.1-1.0% total mRNA;
low< 0.1% total mRNA.
heterologous gene will function efficiently in yeast. Consequently, atten-

tion has turned to cloning and identifying yeast promoters and coupling
them to heterologous gene sequences. As shown in Table III, a large
number of yeast promoters have been used for this purpose.
When transcribing a heterologous gene in yeast using a yeast pro-
moter, two strategies are available: (1) that a promoter lacking any adja-
cent coding sequence be fused directly to the intact coding sequence to
be expressed, i.e., to construct a transcriptional fusion-this results in the
synthesis of a native gene product; or (2) that a promoter, together with
a few amino-terminal codons of its corresponding gene, be fused in the
correct translational reading frame to the heterologous gene sequence,
i.e., to construct a translational fusion-this will result in the synthesis of a
hybrid or fusion protein which may or may not retain the required
biological activity.
Almost invariably the option of a transcriptional fusion is chosen,
but to be able to engineer such a "promoter" fragment one must have
detailed structural and functional information with regard to DNA se-
quences important for both transcriptional efficiency and regulation. In
this context perhaps the efficient phosphoglycerate kinase (PGK) pro-
moter is one of the better-understood promoters (Stanway et al., 1987).
As shown in Fig. 3, a series of PGK-promoter fragments have been
isolated and fused to the mature human IFNa2 sequence and levels of
-150 -~ -50I 0 5f'

I I 1i:Mf2
T T CT
a c PGK 0
A1tj
b pMA23o-1 6 }t to6
Alii
c pMA301-t 2 xto6
Alii
d pMA303-1 2 xrf
Alii
Figure 3. Transcriptional and translational fusion vectors based on the yeast phos-
phoglycerate kinase gene (PGK): a comparison of levels of expression of the human
interferon IFNo:2 gene. (a) cPGK: gives the basic structure of the intact PGK gene, giving
nucleotide positions relative to the ATG codon. T = TATA box; CT = pyrimidine-rich
tract. Hatched region = PGK coding sequence. (b) pMA230-I: a translational fusion with
an intact human o:2 interferon (1FNo:2) gene (closed box). The hybrid protein contains 12
amino acids of the yeast PGK protein. The levels of I FNo:2-PGK expression are shown in
terms of molecules per transformed cell. (c) pMA30 1-1: a transcriptional fusion containing
all PGK 5' noncoding DNA up to -1 followed by the intact human I FNa2 gene. (d)
pMA303-I: a transcriptional fusion containing all PGK 5' noncoding DNA up to -48,
followed by the intact human 1FNa2 gene. (Data taken from Kingsman and Kingsman,
1983.)
expression of biologically active interferon assayed in transformed

yeasts. The optimum levels are achieved when a translational fusion is
employed (pMA230), although a transcriptional fusion vector that car-
ries all essential promoter sequences also directs comparatively high lev-
els of interferon synthesis. Removal of promoter sequences near the
coding sequence results in a significant reductions in expression levels.
Analogous experiments using PGK-promoter fragments that have dele-
tions at the distal side of the promoter fragment have shown that a U AS
is necessary for optimum promoter activity (Ogden et al., 1986). It is
important, therefore, that the location of the functional elements of the
chosen promoter be identified and precisely located.
As detailed in Table III, yeast promoters can vary considerably in
their efficiencies. The m~ority of heterologous gene expression studies
to date have used the efficient promoters, particularly those associated
with glycolytic genes (PGK, ENO, TPI, GAP), to ensure that maximal
levels of the heterologous mRNA are synthesized.
3.4. Termination of Transcription

Once initiated, transcription of the adjacent heterologous gene oc-
curs unheeded provided no introns are present (see Section 3.1). The
next problem is how to get RNA polymerase II to terminate transcription
once the required heterologous gene has been transcribed. Yeast genes
contain DNA sequences in their 3' noncoding regions that apparently
signal not only transcription termination, but also polyadenylation (Zaret
and Sherman, 1982, 1984) since the majority of yeast mRNAs have
defined 3' ends with poly(A) tails (McLaughlin et al., 1973). The nature of
these "termination" sequences is still not clear. Zaret and Sherman (1982)
have identified the sequence TTT-TAG(T)-TTT in the 3' ends of a
number of (but not all) yeast genes and shown that loss of this sequence
from the CYCJ gene results in a dramatic drop in the levels of CYCJ
mRNA coupled with an increase in size ofthe mRNA by 1 kb. In contrast,
Henikoff and Cohen ( 1984) have identified a sequence (T5 ATA) that can
act autonomously as a partial termination signal, but which requires
additional downstream sequences for optimal functioning. Again, this
sequence is not found in all yeast genes. Few yeast genes contain the
polyadenylation signal (AATAAA) identified in higher eukaryotic genes
(Proudfoot and Brownlee, 1976).
An effective termination/polyadenylation sequence 3' to the hetero-
logous gene would therefore seem essential for efficient production of
the mRNA transcript, particularly when one is dealing with a circular
template. Consequently, the majority of yeast expression vectors are
constructed such that a homologous 3' noncoding sequence follows the
heterologous gene and a number of such sequences have been so em-
ployed: TRP 1 (Hitzeman et al., 1981, 1983a), ADH1 (V rdea et al., 1983),
SUC2 (Goff et al., 1984), GAP (Rosenberg et al., 1984), ARG3 (Cabezon et
al., 1984), 2-tJ.m FLP (Hitzeman et al., 1983b), PGK (Mellor et al., 1983),
TRP5 (Miyajima etal., 1985), CYC1 (Zhu etal., 1985), MFa1 (Brake etal.,
1984; Bitter et al., 1984), TPI (Thim et al., 1986). In all cases these 3'
sequences ensured efficient termination of transcription irrespective of
"strength" of the promoter to which they were coupled.
Although it was originally suggested that the presence of a homolo-
gous 3' termination sequence in the immediate vicinity of the foreign
gene was an absolute requirement for efficient expression of the gene
(Hitzeman et al., 1983b), it is now clear that heterologous 3' noncoding
sequences often (albeit fortuitously) contain effective transcriptional ter-
mination signals [e.g., human IFNa2 (Tuite et al., 1982); human 'Y-IFN
(Derynck et al., 1983)], while Goff et al. (1984) have demonstrated that
the addition of an extra 1000-2000 nucleotides on the 3' end of calf
chymosin mRNA (synthesized using the yeast GALl promoter) is no less
stable than the same mRNA whose synthesis was terminated by a SUC2
terminator sequence and gave rise to a much shorter, more "authentic"
transcript.
We know very little about the nucleotide sequences important for
termination of transcription and polyadenylation of mRN As in yeast,
and it is not clear as to the necessity for such sequences immediately
a<ljacent to the heterologous gene. Transcription must, however, be ter-
minated effectively at some point during expression of the foreign gene,
which leads to the general recommendation that an identified transcrip-
tional terminator should be included in a yeast expression vector.
3.5. Regulated Promoters

A number of heterologous gene products synthesized at high levels
inS. cerevisiae have a detrimental effect on cell growth; some may even
be toxic. Strong yeast promoters that are constitutive (e.g., ADH1) thus
present a problem when used to direct the synthesis of such a toxic
product; namely, how does one achieve high intracellular levels of the
product without impairing growth or viability of the host?
One possible way to alleviate this problem is to use a promoter
whose expression can be effectively regulated up or down by changes in
growth conditions, be they media modifications or changes in tem-
perature. The first demonstration of the ability to regulate heterologous
gene expression in yeast was with the PGK promoter coupled to the
human IFNa2 gene (Tuite et al., 1982). PGK is one of a number of
glycolytic enzymes whose synthesis is coordinately regulated by the car-
bon source; when cells are grown on a fermentable carbon source such
as glucose, enz.yme levels can be 100-fold higher than when they are
grown on a nonfermentable source such as acetate (Maitra and Lobo,

1971 ). This regulation occurs at the level of transcription (Holland and
Holland, 1978). The synthesis of human IFNa2 couplt!d to the PGK
promoter showed similar regulation, with the intracellular levels of
IFNa2 being some 20-fold higher on glucose medium than on acetate
medium (Tuite et al., 1982). However, the level of IFNa2 in acetate-
grown cells (approximately 8 X 104 molecules/cell) was still quite high
and it would be more desirable if this "uninduced" level were lower.
Currently three tightly regulated yeast promoters are being used
widely in expression plasmids.
3.5.1. GAL1,7,10
The promoters of these three genes, encoding the galactose-util-
izing enzymes in yeast, are tightly repressed in cells grown in the pres-
ence of glucose, but are induced approximately 1000-fold when glucose
is replaced by galactose (St. John and Davis, 1981; Johnston and Davis,
1984). This activation by galactose can be achieved at relatively high cell
densities, thus making these promoters particularly attractive for util-
izing in tightly regulated expression plasmids. To date, only the GALl
and GALlO promoters have been used for this purpose (e.g., Stepien et
al., 1983; Smith et al., 1985; Kataoaka et al., 1985; Kornbluth et al., 1987).
The potential of utilizing the GAL promoters is limited somewhat by the
level of the endogenous GAL4-encoded protein, which becomes rate-
limiting when the cell contains multiple copies of an induced GAL pro-
moter Uohnston and Hopper, 1982); the GAL4 protein is a positive
regulatory factor required for the activation of GAL promoters. A con-
comitant regulated overexpression of the GAL4 protein may overcome
this problem.
3.5.2. PH05
This gene encodes the yeast-repressible acid phosphatase and its
transcription is tightly repressed in medium containing inorganic phos-
phate. PH05 transcription can be activated by transferring cells to medi-
um depleted of inorganic phosphate (Bostian et al., 1980), but this means
of regulating heterologous gene expression cannot be efficiently carried
out during large-scale fermentations. Consequently, Kramer et al. ( 1984)
have used a host strain carrying mutations in the PH080 and PH04
regulatory genes such that at 35°C the PH05 promoter is not activated
even in low phosphate medium, but when cells are switched to 23°C the
promoter is activated in medium either containing or lacking inorganic
phosphate. This simple temperature switch has been used to successfully
regulate the expression of human leukocyte interferon (Kramer et al.,
1984).
3.5.3. «·Factor
The expression of mating-type-specific genes, including the gene
coding for a.-factor, i.e., MFal, is tightly regulated by the products of the
SIR genes. Brake et al. (1984) made use of a conditional (temperature-
sensitive) sir3 mutant to develop a highly regulated promoter system for
expressing foreign genes in yeast based on the MFal promoter. They
coupled the gene for human epidermal growth factor (hEGF) to the
MFal promoter and introduced the plasmid into a sir3 mutant. If these
transformed cells were grown at 36°C, the MFal promoter was not
functional, but when the transformed cells were shifted down to the
permissive temperature (24°C), a 105 -fold increase in expression of
hEGF was observed after approximately six generations of growth. An
added advantage to using the MFal system is that the foreign proteins
are secreted if they are provided with the MFal signal sequence (Emr et
al., 1983; Brake et al., 1984, and see Section 5.3).
A number of promoters are thus available which, in conjunction
with appropriate host mutations, provide highly regulated transcription
systems for heterologous gene expression in yeast.
4. TRANSLATION OF HETEROLOGOUS mRNAs
Once transcription of the heterologous gene has been efficiently

and accurately achieved, the heterologous mRNA must then be recog-
nized and translated by yeast ribosomes. As with transcription, there are
a number of sequence features that many yeast mRNAs have in common
(Fig. 2), and these must be taken into consideration if one is to achieve
efficient and accurate translation of a heterologous mRN A.
4.1. Codon Bias

The choice of synonymous codons in both prokaryotic and eu-
karyotic organisms is strongly biased, and in this respect Saccharomyces is
no exception (Bennetzen and Hall, 1982; Sharp et al., 1986; and Table
IV). This nonrandom choice of synonymous codons appears to be di-
rectly related to the availability of tRNA isoacceptors within the cell
(lkemura, 1982). Furthermore, genes that are strongly expressed show a
much greater bias in their codon usage than genes that are expressed at
lower levels; this is believed (but not formally proven) to reflect dif-
ferences in the rate of translation of these mRN As. The concentration of
charged cognate tRNA governs the step time required to add amino
acids opposite each codon, and rapid translation would therefore be
favored by the use of triplets for high-abundance tRN As. Analysis of
Table IV. Relative Synonymous Codon Usage

in Highly Expressed Saccharomyces Genesa
u c A G
u uuu 0.42 ucu 3.17 UAU 0.26 UGU 1.78

uuc 1.58 ucc 2.17 UAC 1.74 UGC 0.22
UUA 0.80 UCA 0.23 UAA 2.61 UGA 0.24
UUG 4.50 UCG 0.09 VAG 0.16 UGG 1.00
c cuu 0.13 CCV 0.50 CAU 0.52 CGU 0.64
cue 0.02 CCC 0.03 CAC 1.48 CGC 0.01
CUA 0.42 CCA 3.44 CAA 1.89 CGA 0.00
CUG 0.13 CCG 0.03 CAG 0.11 CGG 0.00
A AUU 1.36 ACU 1.94 AAU 0.28 AGU 0.17
AUC 1.58 ACC 1.78 AAC 1.72 AGC 0.16
AVA 0.06 ACA 0.22 AAA 0.38 AGA 5.20
AUG 1.00 ACG 0.06 AAG 1.62 AGG 0.14
G GUU 2.18 GCU 2.72 GAU 0.84 GGU 3.80
GUC 1.65 GCC 1.13 GAC 1.16 GGC 0.15
GUA 0.04 GCA 0.12 GAA 1.83 GGA 0.02
GUG 0.13 GCG 0.04 GAG 0.17 GGG 0.03
•The data are taken from Sharp tl al. (1986), who analyzed the synonymous codon usage in 32
Saccharomyces genes showing a codon bias index characteristic of a highly expressed gene, i.e., greater
than 0.6.
110 yeast genes (Sharp et al., 1986) reveals that while yeast is capable
of translating all61 amino acid-specifying codons, there are a number
of codons that are very rarely used in yeast mRNAs, e.g., CGA/CGG,arg.
Codon bias in higher eukaryotic mRNAs is markedly different from
that observed for yeast mRNAs, yet it is clear that yeast can efficiently
translate a wide range of heterologous mRNAs. There are at least three
reported instances of heterologous genes, successfully transcribed in
yeast, yet whose mRNAs are not, for whatever reason, translated into
protein: human synthetic proinsulin (Stepien et al., 1983), rat growth
hormone (Ammerer, 1983), and the Drosophilia hsp70 gene (Nicholson
and Moran, 1984). If one determines the codon bias index (Bennetzen
and Hall, 1982) for heterologous mRNAs successfully translated, it is
clear that this appears to have minimal influence. For example, the
codon bias index for hepatitis B surface antigen is 0.02 (Bitter and Egan,
1984), while highly expressed yeast genes such as ADH1 and PGK have
indices of 0.90 or greater.
With the advanced technology of oligonucleotide synthesis it has now
become feasible to examine directly the influence of codon bias by chem-
ically synthesizing a gene using only those codons that are used at high
frequency in yeast mRNAs. This indeed has been done already; for
example, Urdea et al. ( 1983) synthesized a gene coding for human epider-
mal growth factor, a 53-amino polypeptide, using only these codons

found in highly expressed yeast genes. Not surprisingly, this synthetic
gene was expressed in yeast when linked to the yeast GAP promoter
although the levels of expression reported (30 ~Jog/liter) were low. Levels
of expression with the native gene for hEGF were not reported. Similarly,
Bitter et al. (1984) have reported the expression of synthetic genes for P-
endorphin and a-interferon in yeast. Bitter and Egan (1984) constructed
a partly synthetic Hepatitis B surface antigen (HBsAg) gene in which the
first 30 codons were chemically synthesized utilizing "ideal" yeast codons
while the remainder of the gene contained native sequences. While this
hybrid gene was expressed 10- to 15-fold more efficiently than the native
HBsAg gene, this is probably the result of concommitant modification of
the 5' noncoding region of the HBsAg mRNA (see Section 4.2).
From the preceding discussion it is clear that there is little evidence
that codon bias plays an important role in determining the steady-state
levels of a heterologous protein in yeast although it may influence the
fidelity with which a heterologous mRNA is translated. On the other
hand, successful expression in yeast of two synthetic genes constructed
using codons preferred by E. coli, namely, those for human growth hor-
mone (Tokunaga et al., 1985) and human proinsulin (Stepien et al.,
1983), has been achieved.
4.2. 5'-Untranslated mRNA Leader

The translation initiation codon AUG is invariably preceded by a 5'
untranslated leader sequence of up to 100 bases in length. In eukary-
otes, including yeast, this includes a methylated "cap" at the 5' end of the
mRNA. The 40S subunit ofthe ribosome is believed to recognize the 5'
cap and subsequently migrate along the mRNA until it encounters the
first AUG codon (Kozak, 1981, 1984). Analysis of the nucleotide se-
quences of cloned yeast genes encoding abundant proteins reveals that
the occurrence of bases in the 5' untranslated leader of these genes is not
random (Table V). In particular, there is a strong bias for A residues
(51.9%) and against G residues (4.8%). In addition, these 5' leader re-
gions are comparatively short for highly expressed genes, being on aver-
age 41 nucleotides (Table V).
Are those features of the 5' noncoding region important for transla-
tion of heterologous mRNAs in yeast? The answer is almost certainly
yes. For example, Kniskern etal. (1986), by replacing the heterologous 5'
nontranslated sequence of hepatitis B core antigen (HBcAg) with an
authentic yeast (GAP) 5' sequence, increased the levels of detectable
protein by almost two orders of magnitude without significantly chang-
ing the steady-state levels of RNA. Similarly, Bitter and Egan (1984)
replaced the 5' nontranslated region ofthe HBsAg gene (which includes
Table V. Nucleotide Composition of 5' Untranslated mRNA

Leader Sequences in I 0 Highly Expressed Yeast Genes
No. of nucleotides
Length of 5' leader
Gene (nucleotides) A u G c
GAPll 82 39 19 6 18
GAP491 44 24 8 3 9
GAP63 53 30 10 2 11
ENOS 36 20 7 I 8
EN046 41 21 6 2 12
PGKI 39 20 12 2 5
TPII 20 13 3 0 4
PYKI 33 19 4 I 9
ADH•
I 37 16 9 2 10
II 27 12 7 1 7
AVE 41 51.9% 20.6% 4.8% 22.6%
•There are two major ADHJ transcripts.
10 G residues and only six A residues in the 25 nucleotides immediately

preceding the translational initiation codon) with a chemically synthe-
sized fragment containing 27 nucleotides immediately 5' to the initiation
codon of the efficiently expressed yeast GAP491 gene (containing 17 A
residues and only one G residue). A 10- to 15-fold increase in the levels
of HBsAg synthesized was achieved by this manipulation.
The work of Sherman and his colleagues (Bairn et al., 1985; Laz et
al., 1988) using in vitro modifications of the yeast CYCJ mRNA have
confirmed that sequences 5' to the AUG will inhibit translation if they
contain a high proportion of G residues or if they can form hairpin loop
structures.
Sequences within the 5' untranslated leader therefore have pro-
found effects on expression of both homoglous and heterologus mRN As
in yeast and must be taken into consideration when engineering a gene
for high-level expression in yeast.
4.3. AUG Context

In higher eukaryotic mRNAs Kozak (1984, 1986) has shown that the
nucleotide context within which the initiator AUG codon is found can
have a profound effect on translational efficiency of the corresponding
mRNA, and that the nudeotides immediately flanking the AUG codon
are not random. A recent analysis of 131 yeast mRNA sequences clearly
indicates a nonrandom distribution of nucleotides around the AUG
codon (Cigan and Donahue, 1987). In particular, the -3 position is

occupied by an adenine residue in 75% of the mRNAs examined. How-
ever, the bias in nucleotide distribution around the yeast AUG codon (5'-
AAAAUGUCU-3'; Cigan and Donahue, 1987) is different from the
higher eukaryotic consensus (5'-CACCAUGG-3'; Kozak, 1984, 1986).
The nucleotides adjacent to the AUG may be important in dictating the
efficiency with which a particular AUG codon is recognized as a "stop"
signal for the migrating 40S ribosomal subunit (Kozak, 1981 ). The im-
portance of the nucleotide at -3 for maximal translational efficiency has
been confirmed by Bairn et al. (1985), with CYCJ mRNA with a purine,
preferably A, at the -3 position being required for highest efficiencies
of translation of this mRNA. The sequence context is not absolutely
required for translation, however.
It therefore seems necessary to have the AUG initiator codon of a
heterologous mRNA in a favorable nucleotide environment if one is
to achieve maximum efficiency of translation of that mRN A, namely, to
have a purine (preferably A) at the -3 position and preceded by a 5'
untranslated leader that is less than 50 nucleotides in length, rich in A
residues, and lacking any G residues or stable secondary structure po-
tential.
5. POSTIRANSLATIONAL MODIFICATIONS
Once a heterologous mRNA has been successfully translated into a

polypeptide, subsequent modification is often necessary to produce a
mature and functional polypeptide. This is particularly true when the
polypeptide is a secretory or membrane protein which is subjected to
three principal types of modification: formation of disulfide bonds, spe-
cific proteolytic cleavages, and addition of carbohydrate residues. While
little is known about how correct disulfide bond formation is catalyzed in
yeast, a great deal is known about the latter two, particularly the steps
one takes to ensure they take place.
5.1. Yeast Secretion Pathway
High intracellular levels of expression of naturally secreted hetero-

logous proteins in yeast might not be desirable for a number of reasons:
the high levels might prove toxic to the host cell, the product may
accumulate as an insoluble aggregate, or the product may not be fully
posttranslationally processed. In principle, getting yeast to secrete pro-
teins into the surrounding medium should overcome these problems and
furthermore should ease the downstream processing requirements since
a secreted protein will be contaminated with few other proteins.
Yeast is capable of efficiently secreting a number of its own polypep-

tides, provided the nascent polypeptide has a short (15-30 amino acid)
hydrophobic amino-terminai region or "signal sequence." This signal
sequence facilitates contact with the membrane of the endoplasmic re-
ticulum (ER). The nascent polypeptide is then cotranslationally translo-
cated across the membrane of the ER at which point the signal peptide is
cleaved by a membrane-bound protease. The secretory pathway then
involves a series of membrane-bound structures that transfer the mature
polypeptides to their site of release at the plasma membrane.
As in higher eukaryotes, protein secretion in yeast is often linked to
protein glycosylation (for review, see Tanner and Lehle, 1987; and see
Chapter 2). Initially N-linked asparagine core glycosylation of protein
occurs probably during translocation of the polypeptide through the
membrane of the ER. Core oligosaccharides are then extended in a
Golgi-like organelle by addition of outer-chain sugar branches. Finally,
fully glycosylated proteins are packaged into vesicles which are trans-
ported to the growing bud where they fuse with the plasma membrane
and discharge their contents into the periplasmic space (Schekman and
Novick, 1982).
The yeast cell surface consists of at least three layers, the cell wall, a
peri plasm, and a plasma membrane (Ballou, 1982). Most of the enzymes
secreted naturally by yeast are located in the peri plasm or in the cell wall;
examples include invertase, acid phosphatase, and L-asparaginase. Some
smaller nonglycosylated proteins, such as a-factor and killer toxin, are
efficiently secreted through the cell wall and into the surrounding
culture medium, yet the total amount of protein yeast cells secrete into
the medium represents less than 1% of the total cell-associated protein.
The factors that determine periplasmic or extracellular location of yeast
proteins are unknown although, as in other organisms, they are all syn-
thesized with a cleavable hydrophobic amino-terminal signal sequence
(see Section 5.3).
For the reasons just outlined it would be most appropriate to en-
gineer yeast to secrete the foreign protein out into the medium rather
than into the cell-associated periplasmic space. This approach has been
undertaken with remarkable success using both homologous and hetero-
logous signal sequences.
5.2. Use of Heterologous Signal Sequences

As catalogued by Perlman and Halvorson (1983), signal sequences
found in both prokaryotes and eukaryotes, although varying in length
and amino acid composition, all possess a hydrophobic core of approx-
imately 12 amino acids followed by a three-amino-acid peptidase recog-
nition sequence: A-X-B, with A being a large aliphatic amino acid (such
as leu, val, or ile) as are those found at B (principally ala, gly, and ser).
The most frequently occurring peptidase recognition sequence found is

ala X ala.
Reasoning that secretion signals may be functionally similar in all
eukaryotes, a number of groups have examined the ability of hetero-
logous signal sequences to direct the secretion of foreign proteins from
yeast, with varying degrees of success (Table VI). The results obtained
clearly demonstrate that heterologous signal sequences can be recog-
nized and processed in the yeast secretory pathway, but their ap-
plicability is far from satisfactory because ( 1) the efficiency of secretion is
relatively low and, with the exception of mouse a-amylase (Thomsen,
1983) and Aspergillus awamori glucoamylase (Innis et al., 1985), the major-
ity of the protein remains associated with the cell, and (2) a number of
aberrantly processed proteins are often synthesized, both intracellularly
and extracellularly, where the signal peptide has only been partially
removed. For example, Hitzeman et al. (l983b) have shown that, of the
IFNa2 secreted from yeast using its associated signal sequence, only 64%
was. properly processed, with the remaining 36% containing an addi-
tional three amino acids at the amino terminus. Intracellular forms of
IFNa2 contained, in addition, a third species with an additionall4 ami-
no terminal amino acids. That is not to say yeast cannot correctly process
heterologous signal sequences-Sato et al. ( 1986) have shown that the
15-amino-acid signal of human salivary a-amylase is cleaved correctly
Table VI. Heterologous Signal Sequences Examined in Yeast:

Location of the Gene Product
Location of gene product (% total)

Gene Reference Intracellular Periplasmic In medium
Human IFNal Hitzeman et al., 1983b 96 4

Human IFNa2 Hitzeman et al., 1983b 92-98 ~-8
Human IFN'Y Hitzeman et al., 1983b 79-80 10-21
Wheat a-amylase Rothstein et al., 1984 40-70 30-60
Human immuno- Wood et al., 1985 60-90 10-40
globulin A chain
Human immuno- Wood et al., 1985 85-95 5-15
globulin 1.1. chain
Mouse a-amylase Thomsen, 1983 10 90
Human salivary Nakamura et al., 1986 34 21 45
a-amylase
French bean phaseolin Cramer et al., 1985 100 0
Chicken egg white Oberto and Davison, 33 66
lysozyme 1985
A. awamori Innis et al., 1985 10 90
glucoamylase
and theN-terminal glutamine residue correctly modified into pyroglu-

tamate.
Because of the unpredictability of heterologous signal sequence
function in yeast it is therefore more appropriate to couple signal se-
quences from native yeast genes onto a heterologous coding sequence.
5.3. Homologous Signal Sequences

For optimal product purification it would be advisable to use a signal
sequence for a truly extracellular protein. The major candidate for this
is the signal sequence of a-factor.
The a-factor locus (MFal) (Fig. 4) encodes a 165-amino-acid pre-
propolyprotein precursor containing a 22-amino-acid hydrophobic,
amino-terminal signal sequence. Only the mature 13-amino-acid peptide
with a-factor activity is secreted from the cell. MFal-based secretion
vectors generally retain both the signal sequence and the adjacent 61
amino acid prosequence (Fig. 4) and have been shown to efficiently
direct the secretion of a number of foreign proteins, for example, ~-
iJ
ATU TAA
+1 +498bp
+1
pYE/0(-X
e I
f x ~ pHFrJ..S-X
Figure 4. Secretion vectors based on the yeast a-factor gene MFal. (a) Schematic represen-
tation of the MFal locus and the a-factor precursor polypeptide showing the postulated
20-amino-acid signal sequence (closed box), the 61-amino-acid prosegment (open box), the
e
spacer peptides (stippled box), and the a-factor peptides (hatched box). = N-glycosyla-
tion sites (asn x thr); K = KEX2-encoded endopeptidase site (lys arg ~ ). For details of
processing see Chapter 2. (h) Examples of two secretion vectors utilizing fusions with the
MFal gene and heterologous gene X. pYE/a-X (Bitter et al., 1984) contains the signal
peptide, the prosegment, and the first spacer peptide (arg glu ala glu ala). The first amino
acid of X occupies the same position in the hybrid protein as does the first amino acid of
the first a-factor peptide in the native precursor. pMFa8-X (Miyajima et al., 1985) contains
just the lys arg of the first spacer peptide, thus retaining the KEX2 processing site. In both
vectors the MFal promoter is used to direct transcription.
endorphin (Bitter et al., 1984), a-interferon (Bitter et al., 1984; Singh et

al., 1984), human epidermal growth factor (Brake et al., 1984), bovine
prochymosin (Smith et al., 1985), mouse interleukin-2 (Miyajima et al.,
1985), rat carboxypeptidase A (Cardell et al., 1985), and somatomedin-C
(Ernst, 1986). Although such constructs efficiently secrete mature for-
eign proteins into the medium, both aberrant amino terminal processing
and internal proteolytic cleavages are often observed. In particular, in-
complete removal of the (glu-ala) repeats from the amino terminal re-
gion, due to rate-limiting quantities of the dipeptidyl aminopeptidase
encoded by the STE13 gene (Julius et al., 1983), has been observed
. (Brake et al., 1984; Bitter et al., 1984; Thim et al., 1986). This problem
can be overcome by simultaneously introducing into the transformed
cell extra copies of the STE13 gene either on the same plasmid or di-
rectly integrated into the chromosome. Brake et al. (1984) have demon-
strated that removal of the (glu-ala) repeats from the vector by in vitro
mutagenesis results in all the foreign proteins being secreted with au-
thentic amino terminal sequences-this construct retained the lys-arg
KEX2 protease-cleavage sequence immediately adjacent to the coding
sequence and was therefore sufficient for the necessary amino-terminal
processing.
A second, more serious problem is internal proteolytic cleavage of
the secreted foreign proteins with, in the extreme case of P-endorphin
(Bitter et al., 1984), no complete mature protein being found outside the
cell. These cleavages occur exclusively on the carboxyl side of certain
(but not all) lys residues within the protein. The extent to which these
internal cleavages occur is not determined by adjacent residues, but is
rather determined by the conformation-dependent accessibility of sus-
ceptible bonds in the foreign protein. It is clear that not all proteins are
equally susceptible to this proteolytic cleavage; for example, 95% of a-
interferon secreted from yeast using the MFal signal remains intact
(Bitter et al., 1984). The enzyme responsible for this internal cleavage is
almost certainly the endopeptidase encoded by the KEX2 gene, and
since the enzyme is required for secretion per se, it is difficult to envisage
how one might overcome this potentially deleterious activity. Using a
kex2 mutant is clearly not suitable although perhaps the activity of the
enzyme can be modified by in vitro mutagenic techniques, particularly
since the KEX2 gene has been isolated (Julius et al., 1984).
Much work is still required in optimizing yeast secretion vectors
both at the level of signal sequence design and in identifying genetic and
physiological parameters important for achieving maximal secretion.
Yeast signal sequences other than MFal are currently under investiga-
tion, particularly the signal for the yeast killer toxin (Skipper et al., 1985)
and for the extracellular glucoamylase gene (STAI, sometimes desig-
nated DEXI) th.at has been isolated from Saccharomyces diastaticus (Yama-
shita et al., 1985). The use of the signal sequences from the PH05 and
SUC2 genes (coding for repressible acid phosphatase and invertase, re-
spectively) is somewhat limited in that their gene products are not truly
extracellular, but rather are secreted into the periplasmic space and thus
remain, to a large extent, cell-associated.
The growth phase of the culture is important if one is trying to
achieve optimal levels of secretion. A number of reports have indicated
that maximal levels of secretion are achieved in stationary phase cultures;
Brake et al. (1984) for example reported that only 33% of human epider-
mal growth factor (linked to the MFal signal) was secreted from log-
arithmically growing cells, yet 24 hr after the cells reached stationary
phase less than 1% ofhEGF remained intracellular, with the recombinant
product representing greater than 90% of the secreted proteins in the
culture supernatant. The physiological or biochemical basis for these
differences is unknown but dearly requires more extensive investigation,
as does the observation of Rothstein et al. (1984) that levels of secreted
wheat a-amylase are much higher in rich medium, although in this case
secretion was directed by a heterologous secretion signal.
In yeast there seems to be no size limitation on the proteins that can
be effectively secreted from the cell; for example, Schultz et al. ( 1987)
have reported the secretion of the 400-kDa major envelope glycoprotein
(gp350) of Epstein-Barr virus.
5.4. Posttranslational Modification and Secretion

Protein secretion in yeast, as in other eukaryotic cells, appears to be
tightly coupled with posttranslational modification of the nascent poly-
peptide. Apart from cleavage of the amino-terminal signal sequence by a
"signal peptidase," at least four additional modifications can occur to
protein naturally secreted from yeast:
1. Addition of a core oligosaccharide to asparagine residues via an
N-linked glycosylation event, generally at the sequence asn-X-
ser/thr (Lehle and Bause, 1984)
2. Formation of an 0-glycosidic linkage to the seryl or threonyl resi-
dues
3. Covalent attachment of long-chain fatty acid
4. Formation of disulfide bridges between cysteine residues
The addition of core oligosaccharides occurs in the endoplasmic
reticulum as does the fatty acid acylation, although these two modifica-
tions are not coupled. Oligosaccharide side chains are extended in the
Golgi-like apparatus prior to transfer to the periplasmic space. Little, if
anything, is known about disulfide bond formation in yeast-there is
currently no evidence for the presence of a protein disulfide isomerase
activity (Freedman, 1984) although at least one naturally secreted pro-

tein from yeast (the killer toxin) has at least three disulfide bridges (Bos-
tian et al., 1984).
Use of the antibiotic tunicamycin allows one to determine whether
or not a protein synthesized in yeast is subject to N-linked glycosylation
since this antibiotic is a specific inhibitor of such reactions and can act in
vivo. For example, Wood et al. (1985) used tunicamycin to demonstrate
that a significant proportion of immunoglobulin f.L chain produced in
yeast is N-glycosylated. Alternatively, putative glycoproteins can be de-
glycosylated in vitro using either endoglycosidase H or trifluoromethane
sulfonic acid.
To ensure the biological activity of a secreted heterologous protein
posttranslational modification must be both efficient and correct. A
number of investigations have reported the accumulation of nonauthen-
tic forms of several heterologous proteins secreted from yeast that are
immunologically related to the authentic product; for example, human
salivary a-amylase (Sato et al., 1986), Epstein-Barr virus major envelope
glycoprotein gp350 (Schultz et al., 1987), and human T-cell leukemia
virus type I envelope protein (Kuga et al., 1986) are all found to be
synthesized in more than one form due to incomplete glycosylation of
the proteins. There is also evidence that yeast can fail to add any carbo-
hydrate to a naturally glycosylated protein, e.g., hepatitis B surface anti-
gen (Hitzeman et al., 1983a).
The core structure of yeast oligosaccharides is identical to that pres-
ent on mammalian glycoproteins (Ballou, 1982) but atypically the outer
chains are mannose-rich. It is therefore to be expected that the
glycosylation patterns of recombinant mammalian glycoproteins will
not, by and large, be authentic.
5.5. Other Posttranslational Modification Events

Viral and mammalian proteins can also require other posttransla-
tional modifications in addition to glycosylation and disulfide bond for-
mation to generate biological activity, and recent evidence suggests that
yeast can carry out several of the required modifications; for example,
phosphorylation and myristoylation (p60v-src; Kornbluth et al., 1987) and
amino terminal acetylation (human Cu, Zn superoxide dismutase; Hal-
lewell et al., 1987) have been directly demonstrated in yeast. In addition,
yeast, like other eukaryotes, has an enzyme (methionine aminopeptidase)
that can cleave off the N-terminal methionine residue cotranslationally.
This enzyme is inhibited by the presence of a charged residue in the
penultimate residue (Huang et al., 1987), and therefore the possibility
exists that theN-terminal methionine will not be removed from a hetero-
logous protein which may effect its biological activity.
6. MAXIMIZING PRODUCT YIELD
Previous sections have dealt with the question "How can one get a
heterologous gene expressed authentically in yeast?" Once initial ex-
pression has been achieved, attention is then often turned toward max-
imizing the levels of recoverable, fully processed, biologically active pro-
tein. Potentially, given a high-copy-number plasmid (50-1 00 copies/cell)
carrying an efficiently transcribed gene, one should obtain levels in ex-
cess of 50% total cell protein. This potential synthesis output has indeed
been achieved with homologous gene products; for example, if the en-
tire PGKJ coding sequence (coding for the abundant glycolytic enzyme
phosphoglycerate kinase) and corresponding transcription signals are
introduced into yeast on an autonomous high-copy-number YEp vector,
more than 50% of the accumulated total cell protein is phosphoglycerate
kinase (PGK) (Fig. 5). This compares to a normal level of 1-2% in
nontransformed wild-type yeast cells (Mellor et al., 1985). Yet replacing
the PGK coding region with a heterologous PGK coding sequence,
a. b,
pMA27 pMA826
Figure 5. Overexpression of homologous and heterologous phosphoglycerate kinases

(PGK) inS. cerevisiae transformants. (a) The structure of expression plasm ids for high-level
synthesis of yeast PGK (pMA27) and human PGK (pMA826). In each case the PGK coding
sequence is transcribed from the yeast PGK promoter and both plasmids contain the LE U2
selectable marker and the 2-f.Lm plasmid origin of replication (ORI) and REPJ (STB)
sequences. (b) One-dimensional SDS polyacrylamide gel stained with Coomassie blue to
detect proteins. Lane 1: Pure yeast PGK. Lane 2: Total soluble protein from a pMA826
transformant (human PGK is indicated by the arrow). Lane 3: Total soluble protein from a
pMA27 transformant. Lane 4: Total soluble protein from an untransformed strain. Lane
5: Molecular weight markers. Yeast PGK is indicated by "'·
namely, that coding for human PGK, does not result in 50% of total cell
protein as human PGK, but rather only 1-2% total cell protein (Fig. 5).
The discrepancy between homologous and heterologous gene ex-
pression in yeast is a cause for some concern, and Table VII indicates the
possible reasons for such a discrepancy and the consequences each would
have on the copy number of the plasmid and the steady-state levels of the
mRNA and protein. All three parameters can be assayed routinely (see
Chapter 9).
The possible causes for these discrepancies in expression levels in
the PGK-based expression system have been investigated in detail (Mel-
lor et al., 1985; Chen et al., 1984). The major observation is that there is
at least a 10-fold reduction in the steady-state levels of the heterologous
mRNA compared to PGK mRNA. This indicates there is either a reduc-
tion in transcription initiation and/or elongation on the heterologous
gene, or the heterologous mRNA has a short half-life. The recent evi-
dence that the PGK gene of yeast contains a transcriptional "enhancer"
element within its coding region (and is thus missing from the hetero-
logous expression plasmid; Mellor et al., 1987) suggests a possible reason
for these results; namely, transcription initiation from the PGK 5' pro-
moter region is not optimal in the absence of coding and/or 3' se-
quences. It should be stressed that this may be unique to the PGK gene
and this comparison is indeed further complicated by the relatively long
half-life of both PGK mRNA and protein (Chen et al., 1984). It does,
however, illustrate that placing a heterologous gene between efficient 5'
and 3' transcription signals may not be sufficient to guarantee maximal
transcription of the heterologous gene.
The efficiency of gene expression initiation is not the only para-
Table VII. Possible Reasons for Differences in Efficiency

of Expression of Homologous vs. Heterologous Genes
in Yeast
Consequences
Plasmid Steady state
Copy mRNA Protein

Cause no. levels levels
1. Decreased plasmid copy Low Low Low

number
2. Structural instability of plasmid High Low Low
3. Increased protein turnover High High Low
4. Decreased translation of mRNA High High Low
5. Decreased transcription of gene High Low Low
6. Increased mRNA turnover High Low Low
meter to which careful attention must be paid. Two other parameters

should also be considered: the stability of the recombinant plasmid and
the genotype of the host strain into which the recombinant plasmid
has been introduced.
6.1. Stability of Recombinant Plasmids in Yeast

Recombinant plasmids, present at high copy number in yeast, show
varying degrees of stability both in the absence and, to a lesser extent, in
the presence of selective pressure. There are two types of plasmid in-
stability: segregational instability, giving rise to plasmid-free cells, and
structural instability, which generates plasmids, structurally related to the
parental plasmid, but having either deletions, insertions, or rearrange-
ments. The majority of studies to date have concentrated on the former
phenomenon and can be monitored by the loss of a plasmid-encoded
phenotypic marker, e.g., auxotrophic requirement. It is, however,
important to be aware of the latter class of instability, because it can lead
to either loss or reduction in the levels of expression of the plasmid-
borne heterologous gene even though that plasmid may still retain the
"scorable" phenotypic marker used to monitor plasmid retention.
The parameters that influence the stability of a recombinant plas-
mid in yeast can be broadly subdivided as follows:
1. The source of the origin of replication
2. The copy number of the plasmid
3. The mechanism for partitioning the plasmid during mitotic cell
division
4. The nature of the gene product whose synthesis is being directed
by the heterologous gene present on the plasmid
5. The genotype of the host strain employed
The majority of recombinant plasmids utilized in heterologous gene
expression systems carry either the 2-JA.m circle origin of replication
(ORI) or a homologous ARS sequence (see Section 2.2.2). It is generally
accepted that plasmids utilizing the 2-J~.m circle sequence show greater
stability (Futcher and Cox, 1984).
6.1.1. Stability of 2·J&m·Based Vectors

The stability of the 2-JA.m plasmid is generally attributed to ( 1) its
ability to increase its copy number when its copy number is low (Futcher,
1986), and (2) an efficient, plasmid-encoded partitioning system Qay-
aram et al., 1983; Kikuchi, 1983) that ensures equal distribution of plas-
mid molecules to mother and daughter cells after mitosis. The latter
partition system operates without modulating the copy number of the
plasmid and consists of two trans-acting functions (encoded by REP1 and

REP2) and two cis-acting functions (REPJ and ORI). Invariably manip-
ulation of 2-JJ.m sequences during the construction of a chimaeric yeast
plasmid leads to disruption or total loss of one or more of these func-
tions. The consequences of such losses on general segregational stability
of a constructed plasmid are summarized in Table VIII.
2-JJ.m-based vectors lacking the ORI sequence dearly cannot be
used to transform yeast unless an alternative origin of replication (i.e.,
an ARS sequence) is also present on the plasmid. Removal of any one of
the other REP components can significantly reduce the stability of the
plasmid without impairing its transforming ability (Table VIII). For the
two trans-acting functions (REP 1 and REP2) absence of one or both of
these sequences from the constructed plasmid can be overcome by using
a host strain that already carries the native 2-JJ.m plasmid which can
supply the missing trans-acting functions. Note that the majority of labo-
ratory yeast strains are [cir+]).
Loss of REPJ from the vector has more serious consequences on
plasmid stability; a REPJ-defective plasmid carrying the 2-JJ.m OR/ is as
unstable as an ARS-containing vector. The stabilizing activity of REPJ is
independent of its distance from, or orientation with respect to, the OR/
sequence (Kikuchi, 1983). The stabilizing property of REPJ is not re-
stricted to the 2-JJ.m ORI, since it will also stabilize a vector carrying an
ARS (Kikuchi, 1983). Thus any vector, to be maximally stable and retain
high copy number, should contain the REP3 sequence, with the REP1
and REP2 functions being supplied by the same plasmid or via the
endogenous 2-JJ.m population. The origin of the ORII ARS sequence per
Table VIII. Summary of the Stability Characteristics of a Range

of Recombinant Yeast Plasmids
2-f.l.m REP %Plasmid-

function bearing cells Segregation Mitotic stability
Plasmid present (selective) bias• (nonselective)
YRp [cir+] None 10-30% M Very low

YRp [ciro] None 10-30% M Very low
YEp None n.d. M Very low
YEp [cir+J REP3 >90% None High
YEP (ciro) REP3 n.d. M Very low
YCp + ARS None >90% None High
YRp REP3 n.d. None High
YEp [cir+fo] REP1, REP2, n.d. n.d. Moderate
REP3
n.d. = not determined.
•M = segregation bias toward mother cell.
se seems less important for stability. No 2-~J.m-based vector has yet been
described that retains the high degree of segregational stability of the
unmodified 2·1J.m circle under nonselective conditions, clearly indicating
that other, as yet unidentified, parameters are involved.
6.1.2. ARS-Based Plasmids

ARS-based plasmids are generally unstable even under selective
conditions. For example, in a population transformed with a
pBR322-TRPl-ARS1 plasmid (YRp7) and grown under selective condi-
tions (for TRP+ phenotype) less than 20% of the cells actually carry the
plasmid (Struhl et al., 1979; Kingsman et al., 1979). During growth under
non-selective conditions there is an exponential loss of the plasmid from
the population such that after only 10 generations less than 1% of the cells
still retain the plasmid (Struhl et al., 1979; Kingsman et al., 1979). The use
of ARS-based plasmids should therefore be avoided when constructing
expression vectors since applying selective pressure clearly does not guar-
antee plasmid retention.
Why ARS-based plasmids are so unstable is still a point of conten-
tion (see Williamson, 1985). Since their copy number is as high as, if not
higher than, 2-~J.m-based plasmids, their instability appears to be due to
an inability to efficiently segregate during mitotic growth, with plasmid
invariably segregating to the parental (mother) cell (Murray and Szostak,
1983).
The degree of segregational instability of both ARS and 2-~J.m ORI-
based plasmids can be greatly reduced by the incorporation of a CEN
sequence (CEN = a DNA fragment from a yeast centromere; Clarke and
Carbon, 1980), but this is at the expense of lowering the copy number to
one to two plasmid molecules per cell.
6.1.3. Instabilities Induced by Overexpression of a Cloned Gene

The overexpression of a cloned gene in yeast, as in E. coli (Carrier et
al., 1983) can lead to a reduction in cell growth rate. Consequently, a cell
in which either the expression plasmid has been lost or in which it has
undergone a structural rearrangement such that the cloned gene is no
longer expressed will have a selective growth advantage over a cell carry-
ing a functional expression system. Cells with such a selective growth
advantage quickly overtake and predominate in a culture. In the absence
of selection for a plasmid-borne marker such as LEU2, the majority of
"nonexpressors" would be plasmid-free, while if such selection was ap-
plied for plasmid retention the observed instability would arise by struc-
tural rearrangement. The former is shown in Fig. 6; while the standard
cloning plasmid pMA3a and a derivative carrying and overexpressing
100
80 ~~~
... 60
ill
....
.~
40
20
0 20 40 60 80 100
no. of gener4fions (non-seltctivl/
Figure 6. Relative segregational stabilities of expression plasmids in yeast during nonselec-

tive growth. All plasmids are based on the high-copy-number LEU212-JJ.m origin plasmid
pMA3a that is relatively stable during nonselective growth. Plasmid loss is monitored by
loss of the plasmid-encoded LE U2 marker following growth through a number of genera-
tions in medium containing leucine. pMA3a ( 0 ); pMA27 (e) = pMA3a carrying the intact
yeast PGK gene; pMA9l-CH <•> = pMA27 with the PGK coding sequence replaced by the
calf prochymosin gene (for details see Fig. 1); pMA9l-HA (0) = pMA27 with the PGKJ
coding sequence replaced by the hemagglutinin gene of influenza virus. (Data from K.
Walters and M. F. Tuite, unpublished.)
the homologous PGK gene (see Fig. 5) are fairly stable under nonselec-
tive growth conditions, a derivative in which the homologous PGK coding
sequence has been replaced either by the calf prochymosin gene
(pMA91-CH) or by the influenza HA gene (pMA91-HA) is lost rapidly
from the culture (M. F. Tuite, and K. Walters, unpublished data). If
selection is retained (via the LEU2 gene), then both pMA91-CH and
pMA91-HA undergo structural rearrangements at high frequency (K.
Walters, Akhamaloka, and M. F. Tuite, unpublished data).
A range of structural rearrangements have been noted by us to occur
with expression plasmids when selective pressure is maintained, and they
appear to generally arise via a homologous recombination event between
plasmid-borne sequences and chr~mosomal sequences. For an expression
vector this invariably involves the 5' and/or 3' sequences of the promoter.
The use of strains blocked in mitotic recombination (e.g., rad52; Malone
and Esposito, 1980) should reduce or even eliminate such structural
instability, as would the use of hosts in which the potential substrates for
homologous recombination had been deleted from the chromosome by
gene disruption (see Chapter 5).
homologous recombination had been deleted from the chromosome by

gene disruption (see Chapter 5).
There is as yet no fully effective means of eliminating the growth
disadvantage of actively expressing cells, although by using a tightly
regulated promoter system (see Section 3.5) expression during the active
growth phase of the culture, i.e., exponential phase, can be turned off
and switched on at a point late in the growth cycle as the cells enter
stationary phase where biomass is almost maximum and cells have fin-
ished dividing, giving little chance for segregational loss to occur. An
alternative approach would be to use an effective secretion system that
ensures no deleterious intracellular buildup of the heterologous gene
product (see Section 5.3).
Another consequence of introducing a high-copy-number plasmid
into yeast is that their replication may impose a heavy energetic drain on
the host cell. One option would be to use plasmids whose copy number
can be switched from low to high during growth. No plasmids analogous
to the temperature-sensitive runaway, plasmid replication mutants de-
scribed in E. coli (Uhlin et al., 1979; Yasuda and Takagi, 1983) have yet
been described for yeast.
6.1.4. Use of Integrated Expression Systems

The pioneering work of Hinnen et al. ( 1978) demonstrated that
heterologous DNA sequences can be stably integrated into the yeast
genome and integrating might offer an alternative approach to ensure
the mitotic stability of an expression system. The technology for inte-
grating plasmid DNA sequences into yeast chromosomes is discussed at
length in Chapter 5. The feasibility of this approach for heterologous
gene expression has been demonstrated by Hitzeman et al. ( 1984), who
integrated a TRP 1 vector, lacking an ARS but carrying the human
IFNal gene coupled to the yeast ADHJ promoter. They found that
individual TRP+ transformants obtained showed great variability in lev-
els of expression of IFNa1 and this was a consequence of varying num-
bers of the plasmid tandemly integrated into the genome. All the trans-
formants obtained were stable during nonselective growth, yet in no case
were the levels of IFNa1 expression obtained (<0.2% total cell protein)
comparable with the levels achieved with the same expression system
located on an autonomous high-copy ARS-based plasmid (1-2%). Hitze-
man et al. ( 1984) also reported similar findings for an integrated "ex-
pression unit" containing the PGK promoter coupled to the HBsAg gene
and the TRP 1 terminator.
The use of integrated expression systems therefore is a promising
alternative to the more traditional autonomous plasmid-based systems
which have been the main focus of this chapter. They have the important
advantage of long-term stability during nonselective growth conditions

(Zhu et al., 1986), but have the disadvantage that the levels of expression
achieved are somewhat lower than achieved with an autonomous system.
Yet, there may be situations where one wants to manipulate an endoge-
nous biosynthetic pathway without requiring large amounts of the gene
product, but where stability is of paramount importance.
6.2. Host Strains for Heterologous Gene Expression

Relatively few laboratory yeast strains have been used as hosts for
expressing heterologous genes and generally contain one, or more, of the
required auxotrophic mutations (leu2, trpl, ura3, his3) to allow selection
and maintenance of the expression vector, but little else in the way of
potentially beneficial mutations. Little attempt has been made to system-
atically modify the host strain characteristics to see if one can create a
more suitable host for heterologous gene expression. Modifications to
the host strain genotype that influence either plasmid copy number,
plasmid integrity, plasmid segregation, or heterologous protein stability
would be of immense value and, as will be discussed in the following
Sections, such modifications can be made. The host strain as well as the
expression system should therefore be carefully considered in parallel at
the onset of a heterologous gene expression program.
6.2.1. Plasmid Stability and Copy Number

The stable propagation of a plasmid population inS. cerevisiae re-
quires DNA replication functions, copy number control, and an effective
mitotic segregation mechanism to be operational at all phases of the
growth cycle. To what extent do plasmids in yeast rely on host-encoded
functions, as opposed to autonomy in these functions? The detailed
studies by Broach and others (see Section 6.1.1) have shown that the
2-j..t.m circle encodes functions that allow DNA replication, that regulate
copy number, and that ensure effective partition at mitosis. Conse-
quently, the 2-j..t.m circle in a very stable plasmid maintaining an effective
copy number of approximately 50-100 molecules/cell. Yet, of course,
replication does require host-encoded polymerases, and plasmid DNA
replication, like chromosomal DNA replication, is under strict, nuclear-
encoded cell cycle control (Zakian et al., 1979).
The involvement of host-encoded functions in maintenance of the
2-j..t.m circle population has recently been confirmed by the isolation of a
mutation (mapl) that causes a significant reduction in the copy number
of the 2-j..t.m circle (Kikuchi and Toh-E, 1986). This mutation also mark-
edly decreases the stability of both YEp and YCp plasmids.
The ploidy, of the host strain can have a marked effect on the segre-
gational stability of yeast chimaeric plasmids (Mead et al., 1986; Spalding

and Tuite, 1989). For example, the rate of loss of a TRP 1-ARSJ plasmid
during nonselective growth is dramatically reduced as the ploidy of the
host strain is increased from haploid to tetraploid (Spalding and Tuite,
1989). In addition to the increase in segregational stability, the copy
number of both chimeric plasmids and the endogenous 2-J.Lm circle is
also affected by the ploidy of the host strain; for example, a tetraploid
strain has four times as many plasmid molecules per plasmid-bearing
cell as a haploid strain (Mead et al., 1986; Spalding and Tuite, 1989).
The constitution of the mating-type system may also play an as yet
undefined role in determining YRp plasmid copy number (Gullov and
Friis, 1985).
Another approach to optimizing plasmid stability is to incorporate
onto the expression plasmid a gene whose product is essential for cell
growth and where the endogenous chromosomal copy of that gene has
been inactivated. This means that the recombinant strain can be grown
in a cheap complex medium. Loison et al. ( 1986) have recently described
just such a system employing a host strain carrying the ura3 furl muta-
tions. Such strains are normally inviable, but can be grown provided they
contain a plasmid with a functional VRAJ (oritidine-5' -phosphate decar-
boxylase) gene. Such an "autoselective" vector can be used to express
high levels of human a-antitrypsin in various complex media (Loison et
al., 1986). In our hands, such vectors, based on a range of essential
genes, show high levels of structural instability (A. Spalding et al., un-
published). One possible way around this problem is to also introduce
the rad52 mutation into the host strain by gene disruption techniques
and thus significantly reduce the likelihood of both inter- and intra-
molecular recombination events (see Section 6.1.3).
6.2.2. Protein Degradation

Saccharomyces cerevisiae contains more than 40 proteolytic enzymes
(Achstetter and Wolf, 1985) and all the compartments of the cell repre-
sent possible locations for proteinases and possible sites for proteolysis.
A number of these proteinases have been implicated in the degradation
of aberrant and/or nonfunctional proteins, namely, proteinases yscA
and yscB and carboxypeptidases yscY and yscS (Achstetter and Wolf,
1985). One might predict that mutant strains that lack one or more of
these proteolytic enzymes might prove suitable hosts for the stable ac-
cumulation of heterologous proteins. Such mutants exist and are listed
in Table IX, yet few data have yet appeared to confirm this notion. While
a number of laboratories have carried out expression work in pep4-3
strains (e.g., Mellor et al., 1983; Smith et al., 1985; Miy~ima et al., 1985),
no dramatic enhancement of yields has yet been reported by use of these
Table IX. Mutations of S. cerevisiae That Result in Lack of

One or More Proteolytic Enzymes
Mutation Deficiency
pral Deficient in proteinase yscA, a vacuolar carboxylic

endopeptidase
prbl Deficient in proteinase yscB, a vacuolar serine sulfy-
dral endopeptidase
prel Deficient in carboxypeptidase yscY, a vacuolar serine
sulfydral exopeptidase
pep4-3 Defective in a variety of vacuolar hydrolase activi-
ties; accumulates precursor molecules
kex2 Defective in proteinase yscF, a metallo sulfydral en-
dopeptidase (Ca2+ ); inability to process a-factor
and killer factor
cpsl Defective in carboxypeptidase yscS, a vacuolar metallo
exopeptidase (Zn2+)
strains. The increasing tendency to secrete heterologous proteins from

yeast (see Section 5) to a large extent overcomes any problems with
protein degradation that arise as a consequence of intracellular accumul-
ation.
6.2.3. Supersecretion Mutants

Irrespective of the strength of the promoter utilized in a secretion
vector, there is apparently a limitation in the amount of protein the yeast
secretion pathway can successfully process and efficiently secrete. While
the rate-limiting step(s) have yet to be identified, Smith et al. (1985) have
increased the capacity of the yeast secretion system by isolating host mut-
ants that secrete up to 10 times the amount of calf prochymosin com-
pared to the wild-type host even though these so-called "supersecreting"
mutants synthesize the same total amount of prochymosin. Two genes
have been identified-SSCl and SSC2-and sccl scc2 double mutants
proved to be more efficient secreters than either sscl or ssc2 single mu-
tants. These host genetic modifications will be of immense value in op-
timizing secretion of heterologous proteins from yeast.
7. SUMMARY AND PROSPECTS
The yeastS. cerevisiae is an ideal biotechnological resource made all

the more useful in the modern biotechnology era by the wide range of
gene manipulation techniques to which it can be subjected. There can be
little argument that S. cerevisiae is an (almost) ideal host in which to

express heterologous genes and largely satisfies a wide range of neces-
sary properties that are needed for universal acceptability in the recom-
binant DNA industry. Table X lists the necessary properties and com-
pares yeast's performance with' the other three "popular" host organisms
in the recombinant DNA industry, namely, E. coli, Bacillus subtilis, and
mammalian cells. While it is clear that the latter probably represents the
ideal (primarily because the majority of genes we seek to express are
generally of mammalian origin), a large number of hurdles remain to be
overcome in their development as cost-effective, large-scale producers of
heterologous gene products, not the least being transfer of culture tech-
nology to the 1000-liter industrial fermentor.
Yeast has the greatest advantage in its wide acceptability as an indus-
trial production strain, and the ease with which it can be genetically
manipulated. While currently achievable levels of heterologous gene
expression are generally lower than the maximal levels achievable in E.
coli, this is due largely to our poor understanding of what DNA signals
are required in an expression plasmid for optimal levels of transcription
of the cloned gene and the critical parameters for efficient translation of
its mRNA. The recent reports of high levels of expression of two recom-
binant proteins, hepatitis B core antigen (Kniskern et al., 1986) and
Table X. Comparison of Yeast with Other Hosts

Utilized in Heterologous Gene Expression Systems
Properties examined Yeast E. coli Bacillus Mammalian
1. Acceptibility as production strain +++ ± ++ ++f±a

2. Science of fermentation +++ + ++ +
3. Posttranslational modification of eukaryo- ++ +++
tic proteins
4. Effective extracellular secretion systems ++ + +++
5. Levels of heterologous gene expression ++ +++ ?
6. Stability of heterologous proteins ++b ++' ?
7. Inducible expression systems ++ +++ ++
8. Plasmid stability and integrity ++ ++ + ?
9. Positive selection systems +++ +++ +++ +++
10. Genetic modification of host +++ +++ +++ ±
II. Production of soluble heterologous ++ + ?
protein
12. Ease of cell disruption ± +++ +++ ++
13. Viral contamination and induced cell lysis +++ ± ± +
14. mRNA splicing +< +++d
•± if tumour or semitransformed cell lines used.

blf appropriate host mutants are used.
•Only its own introns.
d All except yeast.
human Cu/Zn superoxide dismutase (Hallewell et al., 1987), do suggest

that there is no a priori reason why high expression levels cannot be
achieved inS. cerevisiae. The major selling point for yeast over E. coli is
the availability of effective secretion systems that can give high extra-
cellular levels of posttranslationally modified mammalian gene prod-
ucts.
More than 80 different heterologous genes have now been reported
as being successfully expressed in S. cerevisiae using the strategies out-
lined in this chapter. It would, however, be dishonest to suggest that
there are no problems involved in expressing heterologous genes in
yeast. Some of these have already been touched on here; for example,
the inability of yeast to efficiently and correctly splice out introns from
higher eukaryotic mRNAs that lack the critical "TACTAAC" box, while
an inconvenience, is readily overcome by using standard eDNA cloning
techniques. On the other hand, plasmid stability is still a major problem,
particularly since regulatory authorities are not enamoured by the inclu-
sion of "selective agents" in industrial-scale fermentations, especially if
the gene product is for human therapeutic use. The development of a
truly stable, plasmid-based expression system, which can be maintained
in the absence of an exogenously supplied selective agent, is eagerly
awaited. Stability in this context is so often viewed purely from the
segregational point of view, yet structural stability has to be given equal
attention; a segregationally stable plasmid may still be subject to intra-
molecular rearrangements that reduce, even eliminate, heterologous
gene expression. Stable integration of the expression system into the
host chromosome clearly offers one route which has yet to be fully ex-
plored, although the problem here may well be achieving an appropriate
copy number of the basic expression system. It is clear from the work of
Smith et al. (1985) that integration of the transcriptional unit into the
chromosome can lead to an increase in the efficiency of secretion irre-
spective of the copy number.
Major advances await to be made in developing a suitable host strain
and require a greater understanding of host-plasmid and host-product
interactions to improve both plasmid and product stability. The power of
traditional genetics appears to have been largely ignored at the expense
of recombinant genetics within this context. Another area that has been
overlooked is the fidelity of heterologous gene expression in yeast, par-
ticularly at the level of translation; can the yeast protein synthesis ma-
chinery efficiently translate mRNAs that contain codons for which the
corresponding isoacceptor tRNA may be in very short supply without
making an error? In addition to such projected missense errors, there is
also evidence that a significant degree of error at the level of transla-
tional termination occurs in yeast (Tuite and McLaughlin, 1982; M. F.
Tuite et al., unpublished). A consequence of a termination error is the
production of a polypeptide with a C-terminal extension due to transla-

tion of the normally nontranslated 3' region. The isolation of mutants of
yeast that show increased translational fidelity could well prove valuable
in ensuring maximal fidelity of heterologous gene expression.
The proof that S. cerevisiae will be a commercially viable choice for
expressing recombinant proteins comes from the recent licensing of a
hepatitis B subunit vaccine to Merck, Sharp, and Dohme (USA) for hu-
man use. The HBsAg particles assembled in yeast (Valenzuela et al.,
1982; Hitzeman et al., 1983a) have proven to be highly immunogenic
and safe in both animals and humans (McAleer et al., 1984; Scolnick et
al., 1984; Hollinger et al., 1986). In addition, further steps have been
taken to improve this yeast-derived vaccine by engineering a modified
HBsAg particle containing a polyalbumin receptor to improve its
efficacy (Valenzuela et al., 1985a) and by including a second immunogen,
that for herpes simplex virus surface antigen, to produce a potential
polyvalent vaccine (Valenzuela et al., 1985b). A very recent and exciting
further development in this context is the exploitation of the yeast Ty-
like virus particle to generate novel subunit vaccines (Adams et al., 1987).
The use of yeast to produce safe, efficacious subunit vaccines is but one
of a number of the new developments in yeast biotechnology that have
come from exploiting genetic engineering technology.
REFERENCES
Achstetter, T., and Wolf, D. H., 1985, Proteinases, proteolysis and biological control in the
yeast Saccharomyces cerevisiae, Yeast 1:139-157.
Adams, S. E., Dawson, K. M., Gull, K., Kingsman, S.M., and Kingsman, A.J., 1987, The
expression of hybrid HIV:Ty virus-like particles in yeast, Nature 329:68-70.
Ammerer, G., 1983, Expression of genes in yeast using the ADCJ promoter, Methods
Enzymol. 101:192-201.
Bairn, S. B., Goodhue, C. T., Pietras, D. F., Eustice, D. C., Labhard, M., Freidman, L. R.,
Hampsey, D. M., Stiles, J. 1., and Sherman, F., 1985, Rules of translation: Studies with
altered forms of the yeast CYC I gene, in: Sequence Specificity in Transcription and Trans-
lation: ICN-UCLA Symposium, Alan R. Liss, pp. 351-362.
Ballou, C. E., 1982, Yeast cell wall and cell surface, in: The Molecular Biology of the Yeast
Saccharomyces. Vol. 2: Metabolism and Gene Expression(]. Strathern, E. W. Jones, and J. R.
Beggs, J. D., van den Berg, J., van Ooyen, A., and Weissman, C., 1980, Abnormal ex-
pression of chromosomal rabbit (3-globin gene in Saccharomyces cerevisiae, Nature
283:835-840.
Bennetzen,J. L., and Hall, B. D., 1982, Codon selection in yeast,]. Bioi. Chern. 257:3026-
3031.
Bitter, G. A., and Egan, K. M., 1984, Expression of heterologous genes in Saccharomyces
cerevisiae from vectors utilising the glyceraldehyde-3-phosphate dehydrogenase gene
promoter, Gene 32:263-274.
Bitter, G. A., Chen, K. K., Banks, A. R., and Lai, P-H., 1984, Secretion offoreign proteins
from Saccharomyces cerevisiae directed by a-factor gene fusions, Proc. Natl. Acad. Sci.
USA 81:5330-5334.
Bostian, K. A., Hemire, J. M., Cannon, L. E., and Halvorson, H. 0., 1980, In vitro synthesis
of repressible yeast acid phosphatase: Identification of multiple mRNAs and prod-
ucts, Proc. Natl. Acad. Sci. USA 77:4504-4508.
Bostian, K. A., Elliott, Q., Bussey, H., Burn, V., Smith, A., and Tipper, D. J., 1984, Se-
quence of the preprotoxin dsRNA gene of type I killer yeast: Multiple processing
events produce a two-component toxin, Cell36:741-751.
Brake, A.J., Merryweather,]. P., Coit, D. G., Heberlein, U. A., Masiarz, F. R., Mullenbach,
G. T., Urdea, M.S., Valenzuela, P., and Barr, P. j., 1984, a-Factor-directed synthesis
and secretion of mature foreign proteins in Saccharomyces cerevisiae, Proc. Natl. Acad.
Sci. USA 81:4642-4646.
Breunig, K. D., Malcedonski, V., and Hollenberg, C. P., 1982, Transcription of the bacte-
rial p-lactamase gene in Saccharomyces cerevisiae, Gene 20:1-10.
Bussey, H., and Meaden, P., 1985, Selection and stability of yeast transformants expressing
eDNA of an M1 killer toxin-immunity gene, Curr. Genet. 9:285-291.
Cabezon, T., de Wilde, M., Herion, P., Loriau, R., and Bollen, A. 1984, Expression of
human a 1-antitrypsin eDNA in the yeast Saccharomyces cerevisiae, Proc. Natl. Acad. Sci.
USA 81:6594-6598.
Cardell, S. j., Craik, C. S., Hilvert, D., Urdea, M. S., and Rutter, W. J., 1985, Site-directed
mutagenesis shows that tyrosine 248 of carboxy-peptidase A does not play a crucial
role in catalysis, Nature 317:551-556.
Carrier, M. J., Nugent, M. E., Tacon, W. C. A., and Primrose, S. B., 1983, High expression
of cloned genes in E .. coli and its consequences, Trends Biotechnol. 1:109-113.
Chen, C. Y., Oppermann, H., and Hitzeman, R. A., 1984, Homologous and heterologous
gene expression in the yeast Saccharomyces cerevisiae, Nucleic Acids Res. 12:8951-8970.
Cigan, A.M., and Donahue, T. D., 1987, Sequence and structural features associated with
translational initiator regions in yeast-a review, Gene 59:1-18.
Clarke, L., and Carbon, J., 1980, Isolation of a yeast centromere and construction of
functional small circular chromosomes, Nature 257:504-509.
Cramer, J. H., Lea, K., and Slightom, j. L., 1985, Expression of phaseolin eDNA genes in
yeast under control of natural plant DNA sequences, Proc. Natl. Acad. Sci. USA
82:334-338.
Derynck, R., Singh, A., and Goedde!, D. V., 1983, Expression of human interferon-'Y
eDNA in yeast, Nucleic Acids Res. 11:1819-1837.
Dillon, J-A. R., Nasim, A., and Nestmann, E. R. (eds.), 1985, Recombinant DNA Methodology,
Wiley, New York.
Dobson, M.J., Tuite, M. F., Mellor,j., Roberts, N. A., King, R. M., Burke, D. C., Kingsman,
A. J., and Kingsman, S. M., 1983, Expression in Saccharomyces cerevisiae of human
interferon-alpha directed by the TRP 5' region, Nucleic Acids Res. 11:2287-2302.
Edens, L., Born, I., Ledeboer, A. M., Maat, j., Toonen, M. Y., Visser, C., and Verrips, C. T.,
1984, Synthesis and processing of the plant protein thaumatin in yeast, Cell 37:629-
633.
Emr, S.D., Schekman, R., Flessel, M. C., and Thorner,j., 1983, AnMFal-SUC2 (a-factor-
invertase) gene fusion for study of protein localization and gene expression in yeast,
Ernst, J. F., 1986, Improved secretion of heterologous proteins by Saccharomyces cerevisiae:
Effects of promoter substitution in alpha-factor fusions, DNA 5:483-491.
Freedman, R. B., 1984, Native disulphide bond formation in protein biosynthesis: Evi-
dence for the role of protein disulphide isomerase, Trends Biochem. Sci. 6:438-441.
Futcher, A. B., 1986, Copy number amplification of the 2JJ.m circle plasmid of Sac-
charomyces cerevisiae,j. Theoret. Biol. 119:197-204.
Futcher, A. B., and Cox, B.S., 1984, Copy number and the stability of 2~J.m circle-based
artificial plasmids of Saccharomyces cerevisiae,]. Bacteriol. 157:283-290.
Goff, C. G., Moir, D. T., Kohno, T., Gravins, T. C., Smith, R. A., Yamasaki, E., and
Taunton-Rigby, A., 1984, Expression of calf prochymosin in Saccharomyces cerevisiae,
Gene 27:35-46.
Gritz, L., and Davies, J., 1983, Plasmid-encoded hygromycin B resistance: The sequence of
hygromycin B phosphotransferase gene and its expression in Escherichia coli and Sac-
charomyces cerevisiae, Gene 25:179-188.
Guarente, L., 1984, Yeast promoters: Positive and negative elements, Cell 36:799-800.
Gullov, K., and Friis, J., 1985, Maintenance and copy number control of ARSI plasmids in
Saccharomyces cerevisiae: Evidence of a mating type effect, Curr. Gent. 10:21-27.
Hallewell, R. A., Mills, R., Tekamp-Olson, P., Blacher, R., Rosenberg, R., Otting, F., Mas-
iarz, F. R., and Scandella, C. J., 1987, Amino terminal acetylation of authentic human
Cu, Zn superoxide dismutase produced in yeast, Rio/Technology 5:363-366.
Handa, H., Mizumoto, K., Oda, K., Okamoto, T., and Fukasawa, T., 1985, Transcription of
the human adenovirus Ela gene in Saccharomyces cerevisiae, Gene 33:159-168.
Harris, T. J. R., 1983, Expression of eukaryotic genes in E. coli, in: Genetic Engineering, Vol.
4 (R. Williamson, ed.) Academic Press, New York, pp. 128-175.
Henderson, R. C. A., Cox, B.S., and Tubb, R., 1985, The transformation of brewing yeasts
with a plasmid containing the gene for copper resistance, Curr. Genet. 9:133-138.
Henikoff, S., and Cohen, E. H., 1984, Sequence responsible for transcription termination
on a gene segment in Saccharomyces cerevisiae, Mol. Cell. Biol. 4:1515-1520.
Henikoff, S., and Furlong, C. E., 1983, Sequence of a Drosophila DNA segment that func-
tions in Saccharomyces cerevisiae with its regulation by a yeast promoter, Nucleic Acids.
Res. 11:789-800.
Hinnen, A., Hicks,J. B., and Fink, G. R., 1978, Transformation of yeast, Proc. Natl. Acad.
Sci. USA 75:1929-1933.
Hitzeman, R. A., Hagie, F. E., Levine, H. L., Goedde], D. V., Ammerer, G., and Hall, B. D.,
1981, Expression of a human gene for interferon in yeast, Nature 293:717-722.
Hitzeman, R. A., Chen, C. Y., Hagie, F. E., Patzer, E.J., Liu, C-C., Estell, D. A., Miller,]. V.,
Yaffe, A., Keleid, D. G., Levinson, A. D., and Oppermann, H., 1983a, Expression of
hepatitis B virus surface antigen in yeast, Nucleic Acids Res. 11:2745-2763.
Hitzeman, R. A., Leung, D. W., Perry, L. J., Kohr, W. J., Levine, H. L., and Goedde!, D. V.,
1983b, Secretion of human interferons by yeast, Science 219:6~0-625.
Hitzeman, R. A., Chen, C. Y., Hagie, F. E., Lugovoy,J. M., and Singh, A., 1984, Yeast: An
alternative organism for foreign protein production, in: Recombinant DNA Products:
Insulin, Interferon and Growth Hormone (A. P. Bollon, ed.), CRC Press, Boca Raton, FL,
pp. 47-65.
Holland, M. J., and Holland, J. P., 1978, Isolation and identification of yeast mRNAs
coding for enolase, glyceraldehyde-3-phosphate dehydrogenase and phosphoglyce-
rate kinase, Biochemistry 17:4900-4907.
Hollinger, F. B., Troisi, C. L., and Pepe, P. E., 1986, Anti-HBs responses to vaccination
with a human hepatitis B vaccine made by recombinant DNA technology in yeast,].
Infect. Dis. 153:156-159.
Huang, S., Elliott, R. C., Liu, P-S., Koduri, R. K., Weickmann,J. L., Lee,J. H., Blair, L. C.,
Ghosh-Dastidar, P., Bradshaw, R. A., Bryan, K. M., Einarson, B., Kendall, R. L.,
Kolacz, K. H. Y., and Saito, K., 1987, Specificity of cotranslational amino-terminal
processing of proteins in yeast, Biochemistry 26:8242-8246.
Ikemura, T., 1982, Correlation between the abundance of yeast transfer RNAs and the
occurrence of the respective codons in protein genes,]. Mol. Biol. 158:573-597.
Innis, M. A., Holland, M. J., McCabe, P. C., Cole, G. E., Whittman, V. P., Tal, R., Watt, K.
W. K., Gelfand, D. H., Holland, J. P., and Meade, J. H., 1985, Expression, glycosyla-
tion and secretion of an Aspergillus glucoamalylase by Saccharomyces cerevisiae, Science

228:21-26.
Jayaram, M., Li, Y. Y., and Broach, J. R., 1983, The yeast plasmid 211- circle encodes
components required for its high copy number propagation, Cell 34:95-104.
Jellinghaus, V., Schatzle, V., Schmid, W., and Roewekamp, W., 1982, Transcription of a
Dictyostelium discoidin-I gene in yeast: Alternative promoter sites used in two different
eukaryotic cells,]. Mol. Biol. 159:623-636.
Jensen, E. 0., Marcker, K. A., and Villadsen, T. S., 1986, Heme regulates the expression in
Saccharomyces cerevisiae of chimaeric genes containing 5' -flanking soybean leghemo-
globin sequences, EMBO, ]. 5:843-847.
Johnston, M., and David, R. W., 1984, Sequences that regulate the divergent GALJ-GALJO
promoter of Saccharomyces cerevisiae, Mol. Cell Biol. 4:1440-1448.
Johnston, S. A., and Hopper, J. E., 1982, Isolation of the yeast regulatory gene GAL4 and
analysis of its dosage effects on the galactose/melibiose regulon, Proc. Nat/. Acad. Sci.
USA 79:6971-6975.
Julius, D., Blair, L., Brake, A., Sprague, G., and Thorner, J., 1983, Yeast a-factor is
processed from a larger precursor polypeptide: The essential role of a membrane
bound dipeptidyl aminopeptidase, Cell 32:839-852.
Julius, D., Brake, A., Blair, L., Kunisawa, R., and Thorner, J., 1984, Isolation of the
putative structural gene for the lys-arg cleavage endopeptidase required for process-
ing of yeast prepro-a-factor, Cell37:1075-1089.
Kaster, K. R., Burgett, S. G., and Ingolia, T. D., 1984, Hygromycin B resistance as a
dominant selectable marker in yeast, Curr. Genet. 8:353-358.
Kataoka, T., Powers, S., Cameron, S., Fasano, 0., Goldfarb, M., Broach,J., and Wigler, M.,
1985, Functional homology of mammalian and yeast RAS genes, Cell 40:19-26.
Kikuchi, Y., 1983, Yeast plasmid requires a cis-acting locus and two plasmid proteins for its
stable maintenance, Cell35:487-493.
Kikuchi, Y., and Toh-E., A., 1986, A nuclear gene of Saccharomyces cerevisiae needed for
stable maintenance of plasmids, Mol. Cell Bioi. 6:4053-4059.
Kingsman, S.M., and Kingsman, A.J., 1983, The production of interferon in bacteria and
yeast, in: Interferons, Society for General Microbiology Symposium, Vol. 35 (D. C. Burke and
A. Morris, eds.), Cambridge University Press, Cambridge, pp. 211-254.
Kingsman, A. J., Clarke, L., Mortimer, R. K., and Carbon, J., 1979, Replication in Sac-
charomyces cerevisiae of plasmid pBR313 carrying DNA from the yeast TRPJ region,
Gene 7:141-152.
Kniskern, P. J., Hagopian, A. Montgomery, D. L., Burke, P., Dunn, N. R., Hofmann, K. J.,
Miller, W. J., and Ellis, R. W., 1986, Unusually high-level expression of a foreign gene
(hepatitis B virus core antigen) in Saccharomyces cerevisiae, Gene 46:135-141.
Kornbluth, S.,Jove, R., and Hanafusa, H., 1987, Characterization of avian and viral p60••c
proteins expressed in yeast, Proc. Natl. Acad. Sci. USA 84:4455-4459.
Kozak, M., 1981, Possible role of flanking nucleotides in recognition of the AUG initiator
codon by eukaryotic ribosomes, Nucleic Acids. Res. 9:5233-5252.
Kozak, M., 1984, Compilation and analysis of sequences upstream from the translational
start in eukaryotic mRNAs, Nucleic Acids Res. 12:857-879.
Kozak, M., 1986, Point mutations define a sequence flanking the AUG initiator codon that
modulates translation by eukaryotic ribosomes, Cell44:283-292.
Kramer, R. A., de Chiara, T. M., Schaber, M. D., and Hilliker, S., 1984, Regulated ex-
pression of a human interferon gene in yeast: Control by phosphate concentration or
temperature, Proc. Nat/. Acad. Sci. USA 81:367-370.
Kuga, T., Hattori, S., Yoshida, M., and Taniguchi, T., 1986, Expression of human T-cell
leukemia virus type I envelope protein in Saccharomyces cerevisiae, Gene 44:337-
340.
Langford, C., and Galiwitz, D., 1983, Evidence for an intron-contained sequence required
for the splicing of yeast RNA polymerase II transcripts, Cell 53:519-527.
Langford, C., Nellen, N., Niessing, J., and Gallwitz, D., 1983, Yeast is unable to excise
foreign intervening sequences from hybrid gene transcripts, Proc. Natl. Acad. Sci. USA
80:1496-1500.
Langridge, P., Eibel, H., Brown, J. W. S., and Feix, G., 1984, Transcription from maize
storage protein gene promoters in yeast, EMBO ]. 5:2467-2471.
Laz, T., Clements, J., and Sherman, F., 1988, Structural features of mRNA that affect
translation in yeast, in: Genetics rf Translation: New Approaches (M. F. Tuite, M. Picard,
and M. Bolotin-Fukuhara, eds.), Springer Verlag, New York, 1988.
Lehle, L., and Bause, E., 1984, Primary structural requirements from N- and 0-glycosyla-
tion of yeast mannoproteins, Biochim. Biophys. Acta 799:246-251.
Loison, G., Nguyen-Juilleret, M., Alouani, S., and Marquet, M., 1986, Plasmid-trans-
formed uraJ furl double mutants of S. cerevisiae: An autoselection system applicable to
the production of foreign proteins, BioiTechnology 4:433-437.
Losson, R., and Lacroute, F., 1983, Plasmids carrying the yeast OMP decarboxylase struc-
tural and regulatory genes: Transcription regulation in a foreign environment, Cell
52:371-377.
Maitra, P. K., and Lobo, Z., 1971, A kinetic study of glycolytic enzyme synthesis on yeast,].
Biol. Chem. 246:475-488.
Malone, R. E., and Esposito, R. E., 1980, The RAD52 gene is required for homothallic
interconversion of mating types and spontaneous mitotic recombination in yeast, Proc.
Maniatis, T., Fritsch, E. F., and Sam brook, J., 1982, Molecular Cloning. A Laboratory Manual,
Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
McAleer, W.J., Buyak, E. B., Maigetter, R. Z., Wampter, D. E., Miller, W.J., and Hilleman,
M. R., 1984, Human hepatitis B vaccine from recombinant yeast, Nature 307: 178-
180.
McLaughlin, C. S., Warner,]. R., Edmonds, M., Nakazato, H., and Vaughan, M. H., 1973,
Polyadenylic acid sequences in yeast messenger ribonucleic acid, ]. Biol. Chem.
248:1466-1471.
McNeil, J. B., and Friesen, J. D., 1981, Expression of the herpes simplex virus thymidine
kinase gene in Saccharomyces cerevisiae, Mol. Gen. Genet. 184:386-393.
Mead, D. J., Gardner, D. C. J., and Oliver, S. G., 1986, Enhanced stability of a 2tJ.-based
recombinant plasmid in diploid yeast, Biotechnol. Lett. 8:391-396.
Mellor, J., Dobson, M. J., Roberts, N. A., Tuite, M. F., Emtage, J. S., Lowe, P. A., Patel, T.,
Kingsman, A. J., and Kingsman, S. M., 1983, Efficient synthesis of enzymatically active
calf chymosin in Saccharomyces cerevisiae, Gene 24:1-14.
Mellor, J., Dobson, M. J., Roberts, N. A., Kingsman, A. J., and Kingsman, S. M., 1985,
Factors affecting heterologus gene expression in Saccharomyces cerevisiae, Gene 33:215-
·226.
Mellor, J., Pobson, M. J., Kingsman, A. J., and Kingsman, S.M., 1987, A transcriptional
activator is located in the coding region of the yeast PGK gene, Nucleic Acids Res.
15:6243-6259.
Miyajima, A., Miyajima, 1., Arai, K-1., and Arai, N., 1984, Expression of plasmid R388-
encoded type II dihydrofolate reductase as a dominant selective marker in Sac-
charomyces cerevisiae, Mol. Cell. Biol. 4:407-414.
Miyajima, A., Bond, M. W., Otsu, K., Arai, K-1., and Arai, N., 1985, Secretion of mature
mouse interleukin-2 by Saccharomyces cerevisiae: Use of a general secretion vector con-
taining promoter and leader sequences of the mating pheromone a-factor, Gene
37:155-161.
Miyanohara, A., Toh-E., Nozaki, C., Hamada, F., Ohtomo, N., and Matsubara, K., 1983,
Expression of hepatitis B surface antigen gene in yeast, Proc. Natl. Acad. Sci. USA
80:1-5.
Murray, A. W., and Szostak, j. W., 1983, Pedigree analysis of plasmid segregation in yeast,
Cell34:9ll-917.
Nakamura, Y., Sto, T., Emi, M., Miyanohara, A., Nishide, T., and Matsubara, K., 1986,
Expression of human salivary a-amylase gene in Saccharomyces cerevisiae and its secre-
tion using the mammalian signal sequence, Gene 50:239-245.
Nicholson, R. C., and Moran, L. A., 1984, Expression of a Drosophila heat-shock gene in
cells of the yeast Saccharomyces, Biosci. Rep. 4:963-972.
Oberto, j., and Davison, j., 1985, Expression of chicken egg white lysozyme by Sac-
charomyces cerevisiae, Gene 40:57-65.
Ogden,J. E., Stanway, C., Kim, S., Mellor,j., Kingsman, A.J., and Kingsman, S.M., 1986,
Efficient expression of the yeast PGK gene depends on an upstream activation se-
quence but does not require "TATA" sequences, Mol. Cell Biol. 6:5335-4343.
Perlman, D., and Halvorson, H. 0., 1983, A putative signal peptidase recognition site in
eukaryotic and prokaryotic signal peptides,J. Mol. Biol. 167:391-409.
Proudfoot, N.J., and Brownlee, G. G., 1976, 3' non-coding region sequences in eukaryotic
mRNA, Nature 263:211-214.
Rosenberg, S., Barr, P. j., Najarian, R. C., and Hallewell, R. A., 1984, Synthesis in yeast of a
functional oxidation-resistant mutant of human a 1-antitrypsin, Nature 312:77-80.
Rothstein, S. j., Lazarus, C. M., Smith, W. E., Baulcombe, D. C., and Gatenby, A. A., 1984,
Secretion of a wheat a-amylase expressed in yeast, Nature 308:662-665.
Russell, P. R., 1983, Evolutionary divergence of the mRNA transcription initiation mecha-
nism in yeast, Nature 301:167-169.
Sato, T., Tsunasawa, S., Nakamura, Y., Emi, M., Sakiyama, F., and Matsubara, K., 1986,
Expression of the human salivary a-amylase gene in yeast and characterisation of the
secreted protein, Gene 50:247-257.
Schekman, R., and Novick, P., 1982, The secretory process and yeast cell surface assembly,
in: The Molecular Biology rfthe Yeast Saccharomyces. Vol. 2: Metabolism and Gene Expression
U· Strathern, E. W. Jones, and J. R. Broach, eds.), Cold Spring Harbor Laboratory,
Schultz, L. D., Tanner, j., Hofmann, K. j., Emini, E. A., Condra, j. H., Jones, R. E., Kieff,
E., and Ellis, R. W., 1987, Expression and secretion in yeast of a 400-kDa envelope
glycoprotein derived from Epstein-Barr virus, Gene 54:113-123.
Scolnick, E. M., McLean, A. A., West, D. j., McAleer, W. j., Miller, W. J., and Buyak, E. B.,
1984, Clinical evaluation in healthy adults of a hepatitis B vaccine made by recombi-
nant DNA,]AMA 251:2812-2815.
Sharp, P.M., Tuohy, T. M. F., and Mosurski, K. R., 1986, Codon usage in yeast: Cluster
analysis dearly differentiates highly and lowly expressed genes, Nucleic Acids Res.
14:5125-5143.
Singh, A., Lugovoy, J. M., Kohr, W. j., and Perry, L. j., 1984, Synthesis, secretion and
processing of a-factor-interferon fusion proteins in yeast, Nucleic Acids Res. 12:8927-
8938.
Skipper, N., Sutherland, M., Davies, R. W., Kilburn, D., Miller, R. C., Warren, A., and
Wong, R., 1985, Secretion of a bacterial cellulase by yeast, Science 230:958-960.
Smith, R. A., Duncan, M.J., and Moir, D. T., 1985, Heterologous protein secretion from
yeast, Science 229:1219-1224.
Spalding, A., and Tuite, M. F., 1989, Host-plasmid interactions in Saccharomyces cerevisiae:
effect of host-ploidy on plasmid stability and copy number, ]. Gen. Microbiol.
135:1037-1045.
Stanway, C., Kingsman, A. j., and Kingsman, S. M., 1987, The control of transcription in
Saccharomyces cerevisiae, Bioessays 7:62-67.
Stepien, P. P., Brousseau, R., Wu, R., Narang, S., and Thomas, D. Y., 1983, Synthesis of a
human insulin gene. VI. Expression of the synthetic proinsulin gene in yeast, Gene
24:289-297.
St. John, T. P., and Davis, R. W., 1981, The organisation and transcription of the galactose
gene cluster of Saccharomyces,]. Mol. Biol. 152:285-316.
Struhi,'K., 1986, Yeast promoters, in: Maximizing Gene Expression (W. Reznikoff, W. Gold,
and L. Gold, eds.), Butterworths, London, pp. 35-78.
Struhl, K., Stinchcomb, D. T., Scherer, S., and Davis, R. W., 1979, High-frequency trans-
formation of yeast: Autonomous replication of hybrid DNA molecules, Proc. Natl.
Acad. Sci. USA 76:1035-1039.
Tanner, W., and Lehle, L., 1987, Protein glycosylation in yeast, Biochim. Biophys. Acta
906:81-99.
Thim, L., Hansen, M. G., Norris, K., Hoegh, 1., Boel, E., Forstrom, J., Ammerer, G., and
Fiil, N. P., 1986, Secretion and processing of insulin precursors in yeast, Proc. Natl.
Acad. Sci. USA 83:6766-6770.
Thomsen, K. K., 1983, Mouse a-amylase synthesised by Saccharomyces cerevisiae is released
into the culture medium, Carlsberg Res. Commun. 48:545-555.
Tokunaga, T., lwai, S., Gomi, H., Kodama, K., Ohtsuka, E., Ikehara, M., Chisaka, 0., and
Matsubara, K., 1985, Expression of a synthetic human growth hormone gene in yeast,
Gene 39:117-120.
Tuite, M. F., and McLaughlin, C. S., 1982, Natural readthrough of a UGA termination
codon in a yeast cell-free system: Evidence for involvement of both a mitochondrial
and nuclear tRNA, Mol. Cell. Biol. 2:490-497.
Tuite, M. F., Dobson, M.J., Roberts, N. A., King, R. M., Burke, D. C., Kingsman, S.M., and
Kingsman, A. J., 1982, Regulated high efficiency expression of human interferon-
alpha in Saccharomyces cerevisiae, EMBO ]. 1:603-608.
Uhlin, B. E., Molin, S., Gustafsson, P., and Nordstrom, K., 1979, Plasmids with tem-
perature-dependent copy number for amplification of cloned genes and their prod-
ucts, Gene 6:91-106.
Urdea, M.S., Merryweather,]. P., Mullenback, G. T., Coit, D., Hebertein, U., Valenzuela,
P., and Barr, P. J., 1983, Chemical synthesis of a gene for human epidermal growth
factor urogastrone and its expression in yeast, Proc. Natl. Acad. Sci. USA 80:7461-
7465.
Valenzuela, P., Medina, A., Rutter, W. j., Ammerer, G., and Hall, B. D., 1982, Synthesis
and assembly of hepatitis B virus surface antigen particles in yeast, Nature 298:347-
350.
Valenzuela, P., Coit, D., and Kuo, C. H., 1985a, Synthesis and assembly in yeast of hepatitis
B surface antigen particles containing the polyalbumin receptor, Bioi Technology 3:317-
322.
Valenzuela, P., Coit, D., Medina-Selby, M.A., Kuo, C. H., Nest, G. V., Burke, R. L., Urdea,
M.S., and Graves, P. V., 1985b, Antigen engineering in yeast: synthesis and assembly
of hybrid hepatitis B surface antigen-herpes simplex 1 gD particles, Rio/Technology
3:323-328.
Webster, T. D., and Dickson, R. C., 1983, Direct selection of Saccharomyces cerevisiae resistant
to the antibiotic G418 following transformation with a DNA vector carrying the kan-
amycin-resistance gene of Tn903, Gene 26:243-252.
Williamson, D. H., 1985, The yeastARS element, six years on: A progress report Yeast 1:1-
14.
Wood, C. R., Boss, M.A., Kenton, J. M., Calvert, J. E., Roberts, N. A., and Emtage, J. S.,
1985, The synthesis and in vivo assembly of functional antibodies in yeast, Nature
314:446-449.
Woudt, L-P., van den Heuvel, J., van Raamsdonk-Duin, M. M. C., and Mager, W. H., and
Planta, R. j., 1985, Correct removal by splicing of a Neurospora intron in yeast, Nucleic
Acids Res. 13:7729-7739.
Yamashita, I., Suzuki, K., and Fukui, S., 1985, Nucleotide sequence of the extracellular
glucoamylase gene STAJ in the yeast Saccharomyces diastaticus, ]. Bacterial. 161:567-
573.
Yasuda, S., and Takagi, T., 1983, Overproduction of Escherichia coli replication proteins by
use of runaway-replication plasmids,j. Bacterial. 154:1153-1161.
Zakian, V. A., Brewer, B. J., and Fangman, W. L., 1979, Replication of each copy of the
yeast 2 micron DNA plasmid occurs during the S phase, Cell17:923-934.
Zaret, K. S., and Sherman, F., 1982, DNA sequences required for efficient transcription
termination in yeast, Cell 28:563-573.
Zaret, K. S., and Sherman, F., 1984, Mutationally altered 3' ends of yeast CYCJ mRNA
affect transcript stability and translational efficiency,]. Mol. Biol. 177:107-135.
Zhu, X-L., Ward, C., and Weissbach, A., 1984, Control of herpes simplex virus thymidine
kinase gene expression in Saccharomyces cerevisiae by a yeast promoter sequence, Mol.
Gen. Genet. 194:31-41.
Zhu, J., Contreras, R., Gheysen, D., Ernst, J., and Fiers, W., 1985, A system for dominant
transformation and plasmid amplification in Saccharomyces cerevisiae, Bio/Technology
3:451-456.
Zhu, J., Contreras, R., and Fiers, W., 1986, Construction of stable laboratory and industrial
yeast strains expressing a foreign gene by integrative transformation using a dominant
selection scheme, Gene 50:225-237.
"Classical" Yeast Biotechnology
7
STEPHEN G. OLIVER
1. HISTORY
Yeast was probably the first microorganism to be exploited by humans,

and the brewing of alcoholic beverages and the leavening of bread using
yeast represent the oldest, and still the largest, biotechnologies. The
earliest records of the use of yeast date to before 6000 B.c. when the
Sumerians and Babylonians exploited the organism, albeit uncon-
sciously, in the production of beer (Corran, 1975). The use of yeast to
leaven bread by the production of carbon dioxide was developed at a
much later date, in Egypt at ca. 4000 B.c. The ancient Egyptians ex-
tended their bread-making technology to the production of an acidic
beer called boza or boozah from a lightly baked "loaf' of germinated
grain. The production of distilled liquor probably originated in either
China or the Middle East and was common throughout the known world
by the 14th century A.D.
2. BAKING
In the 18th and 19th centuries distiller's or brewer's yeast was sup-
plied to the bakers, and it was not until the introduction of the Vienna
process in 1846 that yeast was deliberately grown for use in baking
(Oura et al., 1979). Today, the provision of baker's yeast is a major indus-
try, with some 1.5 million tonnes of fresh yeast produced annually. The
same companies are frequently involved in supplying both the baking
and distilling industries. Although the two processes demand yeast with
rather different characteristics, the use of baker's yeast for the produc-
tion of alcohol is extensive. For instance, virtually the whole of the Bra-
zilian ethanol industry relies on Fleischmann's baking yeast for its
fermentations.
STEPHEN G. OLIVER • Manchester Biotechnology Centre, University of Manchester,

Institute for Science and Technology, Manchester M60 IQD, England.
213
214 STEPHEN G. OLIVER
The requirements for a good baker's yeast are that it should have a
fast growth rate under aerobic conditions to reduce production time and
that it should also have a high fermentation rate on bread dough. The
development of suitable strains has been greatly facilitated by the fact
that baker's yeasts, unlike brewing yeasts, are diploid organisms which
produce a high proportion of viable spores (Hough, 1985).
The methods involved in the production of baker's yeast contrast
greatly with those employed in alcohol fermentations since the rate of
production and yield of biomass, rather than ethanol, must be op-
timized. Although continuous culture might appear the obvious way of
achieving this, baker's yeast is usually produced in a fed-batch process.
Thus the production of baker's yeast, unlike that of ethanol, is per-
formed under aerobic conditions in substrate-limited conditions so as to
maximize the conversion of assimilable carbon into biomass. Molasses is
the most commonly used carbon source, although there is now some
interest in exploiting starch products and whey as substrates. This would
necessitate genetic manipulation of the yeast and is discussed more fully
in Section 12. Feedback control systems for the addition of carbon sub-
strate have been designed; however, nutrient is most commonly supplied
to the culture according to a predetermined pattern. Frequently this
involves a simple exponential increase in the rate of addition of mo-
lasses, but regimes in which the rate of feed is kept constant, or even
reduced, as stationary phase is approached have been found to improve
both the yield and keeping quality of the yeast (Burrows, 1979).
The sugars available to yeast in the fermentation of dough come
from two sources. First there is the small amount of free sugars that the
flour itself contains (Oura et al., 1979; Burrows, 1979). In wheat flour
these sugars amount to 1-2% by weight according to the particular type
of flour; they consist of glucose, fructose, maltose, sucrose, and a variety
of glucofructans. Typical figures for the sugar composition of wheat
flour are given in Table I. The second source of fermentable sugar is the
maltose produced by the hydrolysis of starch by wheat amylases which
are activated during the making of dough. This maltose released from
starch amounts to some 3% by weight (Burrows, 1979) and thus repre-
sents the principal source of fermentable carbon for the panary fermen-
tation. Therefore, any good baker's yeast must be able to rapidly fer-
ment maltose, yet there is no direct correlation between maltose
fermentation and dough-raising ability (Lovgren and Hautera, 1977).
Baker's yeast must also have high levels of invertase and related enzyme
activities in order to hydrolyze sucrose and the glucofructans. Finally,
although baker's yeast is produced in highly aerobic conditions, it is used
in bread dough in circumstances in which oxygen is limiting or absent.
For this reason the characteristics of a good baker's yeast must represent
a compromise between the contrasting demands of production and use.
"CLASSICAL" YEAST BIOTECHNOWGY 215
Table I. Sugar Content of Wheat Flour
Sugar %w/w
Glucose 0.02
Fructose 0.04
Sucrose 0.26
Glu 1 -fru2 0.40
Glu 1 -fru3 0.26
Higher glucofructans 0.72
Maltose 0.12
3. BEER BREWING
The brewing of beer is an enormous enterprise and probably the

largest biotechnological industry in the world. The amount of beer pro-
duced commercially each year in the world is ca. 10 11 liters. For most of
this production the principal source of fermentable sugars is malted
barley, although a number of African beers use other cereals such as
sorghum. In the United States it is common to supplement malt with
other, cheaper, carbohydrate sources which may readily be converted to
fermentable sugars since American malts have particularly high enzyme
activity and are also rich in nitrogeneous compounds. The most com-
mon solid adjuncts used are maize and rice grits. Grits are produced by
abrading cereal grains to remove the husk, embryo, and aleurone layer.
The remaining endosperm, which is rich in starch, but contains little
lipid or protein, is then milled to produce the grits. Liquid adjuncts may
take the form of corn syrups (produced by the enzymic hydrolysis of
grits) or solutions of cane sugar or invert sugar.
The basic process of beer brewing is outlined in Fig. 1. In the malt-
ing process, barley grains are germinated under controlled conditions
such that enzymes are produced that begin the conversion of the starch
and protein reserves of the seed's endosperm to fermentable sugars and
amino acids which may later be used by yeast. This germination process
is arrested by drying ("kilning") and the temperature employed at this
stage determines the principal characteristics of the malt. If moderate
temperatures are used, then the malt will be pale in color and high in
enzyme activity. Kilning at high temperatures produces dark malts, suit-
able for ales and stouts, which are low in enzyme activity. American malts
are usually of the former class and this permits the use of starch adjuncts
described above, a practice forbidden by law in West Germany.
The next stage in the brewing process involves milling the malted
barley into a coarse flour called grist. This is then steeped in water and
incubated at ca. 65°C, a process known as mashing. The hydrolysis of
Nl
....
01
~
t>l
"tj
:I:
t>l
z
0
0
r
<:t>l
"'
Figure 1. Flow diagram of the brewing process. (From Hough, 1985.)

"CLASSICAL" YEAST BIOTECHNOLOGY 217
starch to fermentable sugars continues in the mash and this is mainly

due to the action of the barley a-amylase. This enzyme is highly ther-
motolerant and only 4% of its activity is lost during kilning (Preece,
1954). This contrasts markedly with ~-amylase and endo-~-glucanase
whose activities decline by 66% and 33%, respectively, as a result of
kilning (Preece, 1954; Preece and Hoggan, 1957). There is also consider-
able conversion of protein and peptides to amino acids during mashing.
This is largely due to the action of carboxypeptidase, which, again, is
relatively heat tolerant (Mikola et al., 1972). During mashing solid ad-
juncts to malt may be added. In order to aid the digestion of this supple-
mentary starch, preparations of amylases and other carbohydrate-
degrading enzymes, such as pullulanase, may be added. These commer-
cial enzyme preparations are usually derived from filamentous fungi or
bacteria, particularly Aspergillus or Bacillus species. Enzyme addition is
especially prevalent in Western Europe (outside West Germany), where
the use of dark malts means that there is insufficient enzyme. activity to
degrade additional starch.
At the end of the mash, the aqueous extract is separated from the
spent grains and other solid material. The spent grains, which still con-
tain about 20% protein by weight, are frequently used in cattle feed. The
clear liquid, called wort or sweet wort (to emphasize its high content of
fermentable sugars) is then boiled in a vessel called a wort kettle or
copper (although it is now commonly constructed of stainless steel). The
purposes of boiling the wort are manifold:
1. Inactivation of enzymes
2. Sterilization of the wort prior to inoculation with yeast
3. Precipitation of residual proteins
4. Reduction in pH through the precipitation of calcium phosphate
(this favors yeast growth and militates against bacterial infec-
tions)
5. Distillation of volatile compounds which might impart off-flavors
to the beer
6. Further development of the color originally imparted by the
kilning of malt
7. Concentration of the wort which determines the starting gravity
and therefore the ultimate alcohol content of the beer
Liquid adjuncts such as sugar solutions or corn syrup may be added
to the wort kettle to provide a further source of fermentable carbon.
Traditionally hops are added at this stage and boiling serves to isomerize
the a-acids in the hops. However, hops are expensive and it is becoming
increasingly common to use a process of postfermentation bittering,
where hop extracts are added as a final step before conditioning of the
beer. This practice is more efficient than the traditional method since
bittering substances are frequently lost with the removal of the foam
head after fermentation.
On completion of boiling, which may take up to 2 hr, the liquid wort
is separated from spent hops and precipitated matter ("trub") either by
filtration through a bed of spent hops or a stainless steel strainer or by
use of a centrifuge or a whirlpool tank. The clarified wort is then cooled
by passage through a heat exchanger and is finally ready for fermenta-
tion. (It is a chastening experience for a microbiologist to walk round a
large modern brewery and see what a small part of the whole enterprise
the fermentation vessels represent.)
Classically, two kinds of yeasts are used in the production of beer.
Yeasts of the species Saccharomyces cerevisiae are used in top fermenta-
tions to produce the beers and ales traditionally drunk in the United
Kingdom; the fermentation is conducted at a temperature of 15-22°C.
Lagers are brewed at a low temperature, 8-l5°C, using the bottom-
fermenting yeast species, S. uvarum. This yeast was previously known as
S. carlsbergensis in honor of the famous Danish beer maker. The distinc-
tion between these two supposed species of yeast has become in-
creasingly blurred, and since S. cerevisiae and S. uvarum can readily in-
terbreed to produce diploid yeasts which yield viable spores, it is
probably best to regard them as varieties of the same species. For diag-
nostic purposes, S. cerevisiae and S. uvarum can easily be distinguished by
their ability to use mellibiose (S. uvarum can and S. cerevisiae cannot).
However, there are now bottom-fermenting ale yeasts and even, in the
United Kingdom at least, top-fermenting lager yeasts.
The fermentation of wort to beer may be represented by the follow-
ing equation (Hough, 1985):
Maltose + amino acid -+ yeast + ethanol + C02 + heat

(from hydrolysis of barley starch)
100 g + 0.5 g-+ 5 g + 48.8 g + 46.8 g + 50k Cal
The low amount of biomass produced, and the high amount of ethanol,
contrasts markedly with the production of baker's yeast under aerobic
conditions:
Sucrose (from molasses) + ammonia + oxygen -+ yeast + water

+ carbon dioxide + heat
100 g + 5 g + 51 g-+ 48 g + 35 g + 73 g
In addition to ethanol, yeast may produce a number of higher alcohols,

the so-called fusel alcohols. It is the reaction between acyl CoA moieties
and those various alcohols which produce esters and fats; the fo:rmer are
important contributors to the flavor of beer. Most of these other alcohols
are produced by the transamination of an amino acid, followed by decar-

boxylation and reduction:
oxoacid amino acid
(CH3) 2CHCHO + C02
isobutyraldehyde
y- NADH 2
~NAD+
(CH 3)2CHCH 20H
isobutanol
A mixture of isobutyraldehyde and methylglyoxal gives (Palamund and

Hardwick, 1969) an earthy, musty aroma to the beer. Other important
flavor compounds in beer include diketones and diacetyl. The latter
imparts a "butterscotch" flavor and results from the chemical decar-
boxylation and oxidation of acetolactate excreted into the beer by yeast
during the fermentation.
OH
CH 3CO - C - COOH --+ CH 3CO - C =0
CH 3 CH 3
acetetolactate diacetyl
Sulfur compounds are other major flavor determinants in beer, the
sulfur being derived either from sulfate or from the sulfur-containing
amino acids methionine and cysteine. Two simple compounds that have
very potent flavors are hydrogen sulfide (H 2S) and dimethyl sulfide
[(CH 3 ) 2 S].
Figures 2 and 3 contrast the course of a typical top fermentation
10 •p 1040
7.5 •p 1030 20
SG •c pH fa and e
5 •p 1020 4.2 50
15 e
2.5 •p 1010 ' ~
~,_ .....SG____ 3.7
24 36 48 60 72 ·o
Time(h)
Figure 2. Time course of a top fermentation with an ale yeast. SG, specific gravity of wort;
t, temperature; fa, fusel alcohol (mg I -I); e, ester content (mg I- I). (From Hough, 1985.)
Krausan Cooling and

SG High Krausen falling setting
pH fa and e
1050 Low krausen Medium krausen
6.0
1040 10 ............ ..------, ..

t
... .
Temperature ·······
I°CI '
.....
--- --
1030 7.5
______ ...
.
1020 5
- .-·- .
fa
.
1010 •
-·- 7 8 10
Days
Figure 3. Time course of a bottom fermentation with a lager yeast. (Symbols as for Fig. 2.)
(From Hough, 1985.)
with S. cerevisiae at a temperature of 15-20°C with that of a bottom

fermentation by S. uvarum at 6-l0°C. A major difference in the two
processes is the time involved, the lager fermentation taking three times
longer than the ale one. This is one reason why a conventionally brewed
lager is more expensive than an equivalent ale. In the traditional ale
fermentation, the yeast rises to the surface during the last 10-12 hr of
the process and may be skimmed or sucked off. A yeasty foam also
develops in the lager process; this is referred to as the "Krausen". In
contrast to the ale fermentation, however, the Krausen falls to the base
of the vessel toward the end of the lager process, taking the yeast with it.
In traditional breweries, a large variety of fermenter designs with
different systems for the retrieval of yeast was employed; these had
evocative names such as Yorkshire Stone Squares and Burton Unions.
Even lager was traditionally brewed in open vessels. Today, closed fer-
mentation vessels are the rule. In large, modern breweries these vessels
are equipped with automatic cleaning systems and have controlled mix-
ing and cooling devices. The most common design is the cylindroconical
vessel (Fig. 4), where the flocculent yeast accumulates in the basal cone
whose temperature may be controlled separately from that of the bulk of
the vessel. Fermentation times are reduced in such vessels and little foam
is formed and thus the amount of bittering substances lost during the
fermentation is minimized. These two features make considerable con-
tributions to the efficiency, and hence the profitability, of the brewing
process.
Alcohol production is, of course, a growth-linked characteristic and
therefore continuous culture commends itself as the optimal system of
Top
H 22' -ro·
26·3 Brl.
Middle
12'-9.
1-4 Brl.
Figure 4. Production tower fermenter and ancillary equipment. A, wort collecting vessels;
B, impeller-type pump; C, flowmeter; D, control valve; E, flash pasteurizer; F, tower; G,
yeast separator; H, beer receiver; J, C0 2-collecting vessel. (From Ault et al., 1969.)
fermentation. In the brewing industry, as in baker's yeast production,

continuous culture has not caught on (at least, not in the Northern
hemisphere). Attempts to introduce continuous beer production in Brit-
ain have failed, falling victim to problems of strain stability (Thorne,
1970; Hall, 1970) and infection. In one, now famous, case a continuous
fermenter became infected with a strain of "killer" yeast (Maule and
Thomas, 1973). Although sake brewers have guarded against this prob-
lem by developing production strains that are themselves killers, no such
innovations have been made in British brewing practice. One suspects
that the radical changes in working practices which continuous fermen-
tation incurs had much to do with the system's demise in the United
Kingdom.
In the Southern hemisphere, continuous culture has been more
successful. A number of alcohol factories in Brazil employ a continuous
fermentation system developed by the Alfa-Laval company (Carioca,
1984). The continuous brewing of beer has been very successful in New
Zealand, where sophisticated systems of "cascade" fermentation (see Fig.
5), with a number of fermenters connected in series and recycling of
yeast cells, have been employed (Coutts, 1961, 1966). A single-vessel,
continuous system was employed in the United Kingdom; the pressure-
cycle tower fermenters used (Ault et al., 1969) were similar in concept, if
not in scale, to the ill-fated ICI "pruteen" fermenter. A critical param-
eter for brewing beer in these tower fermenters is that a sufficient supply
of oxygen to the wort is maintained. Although the alcoholic ferll)enta-
tion is an anaerobic process, small amounts of oxygen are required to
sustain fatty acid and sterol biosynthesis by the yeast (Andreasen and
Stier, 1953; 1954).
Freshly fermented beer is not ready for consumption and must
undergo further treatment before it can be distributed and sold. In
traditional brewing practice, postfermentation treatment, or condition-
ing, is a relatively simple process, but in modern beer manufacture, as in
the other areas of biotechnology, downstream processing is assuming
increasing importance.
In traditional cask conditioning of ale, freshly fermented beer is
"racked ofr' into wooden or (nowadays) stainless steel casks. The beer is
usually held in the cask under pressure and undergoes a secondary
fermentation. The freshly fermented beer introduced into the cask
should contain about 106 yeast cells ml- 1 and these are responsible for
the secondary fermentation. In traditional breweries, however, wild
yeasts (particularly Brettanomyces species) also invaded the casks and im-
parted characteristic flavors to the beer. The secondary fermentation
may rely on the presence of residual sugars in the fresh beer. Alter-
natively, the secondary fermentation may be primed by the addition of
sugar syrups to. the fresh beer. The use of caramel at this stage permits
r--
HOPS
FROM
LAUTER TUN
....
..............
INTERI'IEOIATE
0 I COOLING AN,2..
BOILING 'WORT BOILER ~ CHILLING
VESSEL DJ
......... IAAAA
-
HOP STIWNEA
i : ~•• __..:=:•:."\,. _.:
~
I
HOrTA~
FILTEI\EO WORT STORAGE STABILIZING VESSEL n•F

H•F
OIAtOooiAC[Q.J~n
-
[ARTH FILTDI
1-
.
.--+.....
! i I
HOL'o·UP
!
C. F. I.
i I
!
C.F.2.
i
VES;SEL
i ! ! i:
..._ !
!
:a-LQ
i
i
..i~,
:
~-c
I
I
I
o.J..,
~
l YUST roo ~OO~•GATIOII

UR .. fMT[O BtU Of UR" rOR ~H RfOUI;JIOM
I Y(AST IN SUIP!NSION CONTitOI,.
----------------~
I
I
I
I
AIR FEP.MENT£0 WORT
OG. DILUTION
~
,.... YEA'S
I
SUCROS( H~ CARAMELIIED
I
co, PIUYING UTRACT SUGAR rLAV, FiliNGS
Q Qt Q Q
\ L j _)I
T0 FINISHED BEER
COLLECTION l'rJ:'TTI I
'I '
'
I
I
C1 IUR~UI
-CI~ITATCO
Y(AIT AND
MATUIAL
Figure 5. Cascade fermenter system for continuous beer production. (From Coutts, 1966.)
the color of the finished product to be controlled and the addition of

isomerized hop extracts determines bitterness. Traditional lager produc-
tion involves extensive postfermentation conditioning of the beer in en-
closed lagering tanks which are held at 2°C for up to 9 months. Under
these conditions the residual sugars are slowly fermented and the yeast
settles out together with precipitated polyphenols (tannins) and poly-
peptides (proteins). This lagering yields a haze-free beer with a very long
shelf life, but it is a very costly process.
Modern brewing practice has greatly reduced the time required for
postfermentation conditioning of both ales and lagers. The primary
fermentation is usually allowed to proceed to completion and then the
beer is held at 14-l8°C for 2 days (the diacetyl rest) to allow the vol-
atilization of undesirable products of fermentation. The beer may then
be chilled and carbonated artifically. Special treatments may be em-
ployed to prevent beer haze developing. Haze formation is catalyzed by
heavy metals and is the result of polyphenols and polypeptides coming
out of solution. Two types of haze may be distinguished, chill haze and
permanent haze. Both appear on chilling the beer, but in the latter the
aggregates formed do not redissolve on subsequent warming.
4. SAKt BREWING
Sake is often referred to as rice wine but, in fact, its production is

more similar to that of beer than wine since starch is the primary carbo-
hydrate source and must first be converted to sugars before yeast may
undertake the alcoholic fermentation. The sake fermentation is not only
very interesting, involving the action of two microorganisms in succes-
sion, but also gives rise to a beverage that has the highest alcohol content
(ca. 25% vol./vol.) obtained by direct fermentation. It is believed that the
original way of converting rice starch into fermentable sugars for sake
fermentation was to chew steamed rice so that salivary amylases became
mixed with it. The method of using filamentous fungi to degrade the
starch was introduced from China at the beginning of the 5th century
A.D. (Kodama and Yoshikawa, 1977). An outline of the modern process
of sake brewing is given in Fig. 6.
The rice is "polished" before use by passage through a mill. This has
the effect of increasing the relative carbohydrate content of the grains
from ca. 72% to ca. 78% and results from a fall in the protein, lipid, and
ash contents. Polishing increases the capacity of the grains to absorb
water, which they do to the extent of 25-30% in the next step of the
process-steeping. The steeped grains are then drained of excess water
before steaming for 30-60 min. Steaming has a number of effects: it
sterilizes the rice, permits further absorption of water, denatures pro-
"CLASSICAL" YEAST Blai'ECHNOLOGY 225
IRi~e I
Polishing
riRi'ii~ce;t;b;;ro;-;;n~--~
Spores of
Aspergillus
oryzoe
Main fermentation (about three weeks)

t
Filtration
Saki coke
Settling
~
Filtration
t . .
POStiUriZOIJOn
t
Storage
t
Blending
Difution~
Bo,tling
• . .
Posteur•zotJon
I s!t,q
Figure 6. Flow diagram of the sake process. (From Kodama and Yoshikawa, 1977.)
teins, and converts starch to the a-form in which it is more susceptible to

the action of the Aspergillus amylases.
In parallel with the pretreatment of the rice, the "koji" is prepared.
Koji is a culture of A. oryzae grown on steamed rice (Murakami et al.,
1968). In all, about 50 kinds of enzymes have been reported as present in
koji, but the most important to the sake process are a-amylase,
amyloglucosidase, acid protease, and alkaline protease. The tem-
perature of the koji fermentation influences the proportions of these
enzymes, with high culture temperatures (ca. 42°C) favoring the

amylases and lower ones (ca. 30°C) resulting in higher protease activity.
The final component required for sake fermentation is the yeast
starter culture, or moto. There are two main methods by which this
inoculum is prepared. In the traditional technique, referred to as ki-
moto and yamahai-moto, the culture is naturally acidified by the action
oflactic acid bacteria, which produce conditions under which the growth
of sake yeast is selected over that of the wild yeasts previously found in
the culture. Modern practice, sukujo-moto, demands that the culture is
artificially acidified by adding lactic acid at the start of the sake fermen-
tation, which is then pitched with pressed yeast grown in pure culture by
procedures similar to those described for the generation of baker's yeast
(see Section 2). A good sake yeast needs to exhibit tolerance to low pH,
high osmotic pressure, and high ethanol concentrations. Sake producers
are so convinced of the peculiar properties of their yeast that they have
been led, in the past, to distinguish it by the specific name Saccharomyces
sake. There is no doubt, however, that sake yeast is Saccharomyces cere-
visiae, and its peculiar properties, especially in regard to ethanol toler-
ance, need to be carefully defined. Brown et al. (1981) compared the
ethanol tolerance of the commercial sake yeast, Kyokai No.7, with that
of a laboratory strain, 5D-cyc. They found that the two yeasts were
indistinguishable in terms of ethanol tolerance of net growth, but that
fermentation by the sake yeast was more tolerant to the inhibitory effects
of ethanol than was that of the laboratory strain (Fig. 7). The inhibition
constant, K;, for the effect of ethanol on fermentation rate was found to
be 0.97 M for 5D-cyc but 1.32 M for Kyokai No.7. The ability of sake
yeast to withstand high ethanol concentrations during sake brewing may
not simply be due to the properties of the yeast itself. Hayashida et al.
(1975) have reported that a proteolipid produced by Aspergillus oryzae
enhances the ethanol tolerance of sake yeast.
The main fermentation of the sake process, the "moromi," is carried
out in a large, open vessel. The moromi is a fed-batch process in which
successive additions of fresh substrate, in the form of steamed rice and
koji, are made. In the moromi, the conversion of starch to fermentable
sugars, by the action of A. oryzae, and the fermentation of those sugars to
alcohol, by S. cerevisiae, occur simultaneously. This again distinguishes
the brewing of sake from that of beer, where saccharification (malting
and mashing) and fermentation occur sequentially. Since the final con-
centration of alcohol reached (ca. 20% vol.lvol.) requires nearly 40%
wt./vol. sugar to be supplied, it is evident that this can only be achieved
by such a sequential process. Yeast would not be able to ferment such a
high concentration of sugar if it were all provided at the start of the
process.
..-
1
"7- 1
r::r
~
'
Jz ~\
3 \~,
Figure 7. Effect of ethanol on fermentation in two yeast \\
strains: plot for noncompetitive inhibition, -6-, a !abo- 4-'------.-...,.;;;"'...,
ratory haploid strain of S. cerevisiae; -A-, a commercial 0 10 20 30
sake yeast. (From Brown et al., 1981.) "'o Wfv ethanol
The fermentation takes about 25 days to complete, and while the

temperature varies considerably through the process, it never exceeds
l8°C and at the early stages is as low as 7-8°C. It is often considered that
this low temperature contributes to the ethanol tolerance of sake yeast.
However, the effects of temperature on ethanol tolerance are complex
and differ for growth and fermentation (Brown and Oliver, 1982a). A
particularly low temperature is not necessarily beneficial.
At the end of the fermentation, when the alcohol content may be
enhanced by the deliberate addition of ethanol diluted in a salts solution,
the sake is harvested by filtration under pressure. The cake left behind
on the filter, 'sake-kasu', contains starch, protein, and various enzymes
as well as yeast cells. It is sold to the food industry for use in other
processes such as pickle making. One tonne of polished rice yields 3000
liters of sake (at ca. 20% vol/vol ethanol) and 200-250 kg of sake-kasu.
The sake is held at low temperature for 5-10 days, during which time it
further matures owing to the action of enzymes both released from the
Aspergillus and contained within the, now largely dead, yeast. It is impor-
tant to avoid autolysis of the yeast cells at this stage. Much solid material
settles out during storage and the sake is finally clarified by filtration
through activated charcoal. A further period of maturation may then be
allowed before pasteurization and bottling. Pasteurization was intro-
duced into the sake process in the 16th century, thus considerably pre-
dating Louis Pasteur. Given the very high alcohol content of the product,
it is perhaps surprising that it is needed at all. However, infections with
highly ethanol-tolerant lactic-acid bacteria ("hiochi" bacteria) do occur
and produce off-flavors which are mainly attributable to diacetyl forma-
tion (see Section 3). Deterioration of bottled sake can also result from
photooxidation reactions, which are readily prevented by the use of
opaque bottles.
5. WINE MAKING
The fermentation of grape must into wine is, at a superficial level at

least, a much simpler process than the production of beer or sak~. This is
because the starting material contains fermentable sugars and there is no
problem of converting nonfermentable polysaccharides into fermenta-
ble monomers. Grapes contain four principal sugars-glucose, fructose,
arabinose, and rhamnose. In the unripe grapes, glucose is by far the
most abundant sugar, but during the ripening process the proportion of
fructose increases until it equals or even exceeds that of glucose (Lafon-
Lafourcade, 1983). The total concentration of reducing sugars varies
according to the strain of grapes employed as well as the conditions
(particularly climatic) of their cultivation. However, the usual range is
120-150 g/liter, and of this total, the pentoses represent less than 1%
(Vogt et al., 1974). In addition to these monomers, polysaccharides are
also present as pectins and gums. The principal acids present in grape
must are tartaric and malic and their content varies from ca. 2 g/liter to 7
g/liter according to climate. This gives the must an average pH of 2.8-
3.9. In hot, dry regions, such as Argentina, the content of these acids can
be neglible. The citric acid content of grapes is 10% the level of malic or
tartaric acids. Nitrogen is present to the extent of 0.1-1 g/liter and
assumes a variety of forms, including free ammonia (3-1 0% of the total),
amino acids (25-30%), and polypeptides (30-50%). The principal ami-
no acids are arginine, glutamate, proline, and threonine and the propor-
tion of total nitrogen present as amino acids is found to increase during
ripening.
Grape must, therefore, contain the essential requirements for
growth and fermentation by yeasts, and in traditional wine-making prac-
tice, the natural flora of the grape is relied upon to initiate the fermenta-
tion. A number of nutritional supplements will accelerate the process,
however, and are permitted by the regulatory authorities. These supple-
ments include phosphate, potassium, and thiamin (Lafon-Lafourcade,
1983). The latter, in addition to activating the fermentation, is also
useful in preventing the accumulation of keto compounds which will
combine with the S02 used to arrest the fermentation and prevent the
microbial degradation of the wine. The requirement for small amounts
of oxygen to permit fatty acid and sterol biosynthesis (Andreasen and
Stier, 1953, 1954) for yeast growth may be fulfilled by a process known
as "pumping over" (see Section 5.2), which is used exclusively for the
production of red wines.
The performance of the yeasts present on the grapes is very variable
and this frequently results in a long lag before the start of fermentation.
In modern practice, therefore, it is usual to inoculate with ca. 10 g/hl of
active dried yeast. The natural flora is often eliminated by sulfiting

before this inoculum is added. Specially selected strains of S. cerevisiae or
S. bayanus are used for this purpose and each winery jealously guards its
own strains. As with S. sake, it is not clear that there is any genuine
taxonomic distinction between S. cerevisiae and S. bayanus.
5.1. White Wine Production

White wine is produced from pure juice obtained from white
grapes. The extraction of juice is performed rapidly to avoid oxidation
and maceration of the skins. The use of perfectly sound grapes is pre-
ferred in order to avoid accidental contamination of the juice. The ex-
traction process has four phases: crushing, draining, pressing, and
clarification.
Crushing breaks the skins of the grapes and lets the juice run out of
the pulp. Fragmentation or grinding of the grapes is avoided since this
results in the production oflarge amounts of sediment. Vibrating screens
are then used to drain off the juice since this accelerates the process and
avoids oxidation. Following crushing, pressing permits extraction of the
remaining juice from the pulp, and continuous, pneumatically driven
presses are employed in large-scale operations. The problem with such
equipment is that it may pulverize the press cake, leading to contamina-
tion with skin fragments. Oxidation can also be a problem at this stage. An
alternate approach is to press intact grapes without crushing. This yields
musts that are readily clarified and highly fermentable.
Clarification is the removal of suspended solids from the must and
may be achieved by filtration, centrifugation, or, more traditionally, set-
tling and decantation. Addition of sulfite to the settling tanks will pre-
vent both fermentation by the indigenous flora and oxidative damage.
Pectinolytic enzymes may be added at this stage to reduce the polysac-
charide content and accelerate the process. Industrial enzyme prepara-
tions commonly contain cellulases, hemicellulases, xylanases, and pro-
teases in addition to pectinases. Removal of protein from the must may
also be achieved by addition of the absorbent clay bentonite. This pro-
cedure is familiar to biochemists who need to reduce nuclease activity in
their preparations.
The clarified must is stored in casks or, these days, in large tanks
before fermentation, which is carried out under anaerobic conditions at
a temperature of 20°C. Fermentation is arrested by sulfiting, and this
may be done early if it is desired to produce a sweet white wine. An
alternate strategy is to add fresh must, often in concentrated form, to the
wine at the time of bottling. Such a procedure is said to "mellow" a dry
wme.
After fermentation, white wine should be rapidly clarified by filtra-

tion or centrifugation. Prolonged storage of the wine on the "lees" (yeast
and other solid residues of fermentation) can cause the development of
disagreeable odors due to the formation of sulfhydryl compounds.
Nevertheless, some classes of wine are deliberately left on the lees to
promote the malolactic fermentation.
5.2. Red Wine Production

The major distinction between the production of red and white
wines is that the former requires maceration of the grapes while the
latter avoids it. Maceration is the process whereby the must is left in
prolonged contact with the solid parts of the grape-the skins, the seeds,
and even, for some varieties, the stems. These impart color to the wine
and also increase its tannin content. Crushing and destemming the
grapes are often carried out simultaneously. Removal of stems gives the
wine a more subtle, "refined" flavor. If the stems are left on the berries,
the product has an astringent, "herby" taste.
After the crushing stage, production of red wine follows a course
similar to that of the white varieties. However, the fermentation is car-
ried out at a higher temperature, 25-30°C, since this facilitates color
extraction. Even at this temperature, cooling of the fermenter will often
be required since red wine fermentation is more rapid than that of
white, as the must is rich in nitrogen and other "activating" substances.
During the fermentation the practices of skin, seed, and pulp are carried
to the top of the fermentation vessel by the rising bubbles of C02 and
form a "cap" on the surface. The fermenter may b~ drained into a
second vessel and the liquor "pumped over" into the first vessel again.
This stimulates yeast growth by the addition of small quantities of oxy-
gen and also promotes color extraction due to the passage of liquor
through the cap. The liquor is finally drawn off into large casks where
the conversion of residual sugars into alcohol is completed. This may
also permit the malolactic fermentation. The wine is finally stabilized by
the addition of so2, before a period of storage which may be prolonged
to produce a mature red wine.
Storage of wine in barrels for the purposes of maturation involves
decanting quarterly and the frequent addition of wine to replace evap-
orative losses. It is essential that the headspace of the barrel be kept full
in order to prevent the growth of acetic acid bacteria. Many changes in
the color and flavor of wine occur during maturation, some of which are
acquired from the wooden casks themselves.
The last stage of the wine-making process before bottling is fining.
In this clarification procedure, suspended impurities are precipitated by
the addition of bentonite or protein (gelatin, globulin, casein, or egg
white). The protein reacts with tannin to produce a precipitate which

carries down other suspended particles. Filtration usually follows fining,
although this stage is omitted for some varieties. The wine must then be
bottled under conditions that prevent microbial contamination. In the
hot-bottling process, this is achieved by heating the wine to 50°C and
bottling at the same temperature. An alternative is to "flash"-pasteurize
the wine at 80-85°C for several seconds immediately before filling the
bottles.
6. STRAIN DEVEIDPMENT
In all industrial processes involving microorganisms, considerable

improvement in product yield or process efficiency may be achieved by
genetic development of the production organism. This is true of yeast in
the traditional biotechnological processes described here. Indeed, much
unconscious selection of desirable traits must have been carried out by
practitioners of the arts of brewing and bread making over the millennia
in which they have been in use. Modern strain development has a wide
range of techniques available for use with yeast, and I will not duplicate
here the descriptions provided in other chapters in this volume but will,
instead, ·provide some specific examples of their use. The endeavors of
the geneticist in yeast strain development are often hampered by the fact
that production strains have ill-defined genetics and are frequently un-
balanced polyploids. The factors whose improvement is sought, such as
flavor, flocculation ability, or alcohol tolerance, may themselves be
poorly defined and controlled (or modified) by large numbers of genes.
Our ignorance of the biochemical nature and genetic control of these
traits may mean that, in some cases, the application of gene cloning
techniques would be premature and wholly inappropriate. In other
cases, such as the extension of substrate range, recombinant DNA tech-
nology holds great promise since it is hoped that new potentialities can
be recruited in a specific manner that does not compromise the desirable
traits the strain already possesses.
7. ETHANOL TOLERANCE
The life cycle of S. cerevisiae provides both haploid and diploid

vegetative phases and permits isolation of all four products of meiosis.
This makes it ideal for strain development by conventional genetic tech-
niques (see Chapter 4 by R. B. Wickner). Such techniques usually involve
the screening or selection of mutants on agar plates, and the geneticist
designs conditions such that only the desired mutant will grow or not
grow. For some characteristics such aU-or-none selection or screening

techniques are not practicable. This may be because the desired phe-
notype is determined by a large number of genes or because only rela-
tively small, quantitative, improvements may be achieved. Both these
problems are encountered when attempting to isolate yeast mutants with
improved tolerance to ethanol.
Yeast is highly tolerant of ethanol when compared to other micro-
organisms, probably because of unconscious selection on the part of the
producers of alcoholic beverages throughout history (Oliver, 1984).
Thus only small improvements to the organism's tolerance are likely to
be achieved in any strain improvement program. However, owing to the
energy consumption profile of alcohol distillation, even relatively mod-
est increases in final product concentration may be economically signifi-
cant (Righelato, 1980). The likelihood that a large number of genes are
involved in determining ethanol tolerance is indicated by the complexity
of the effect that the alcohol has on its producer organism. Fermentation
rate (Rubner, 1912), glucose consumption (Gray, 1941), biomass yield
(Troyer, 1953), growth rate (Troyer, 1953; Holzberg et al., 1967; Aiba et
al., 1968), and cell viability (Thomas et al., 1978) have all been used as
indicators of the relative tolerance of yeast strains to the toxic effects of
ethanol. In fact, different inhibition constants may be determined for
the inhibitory effect of ethanol on cell viability, true growth rate, and
fermentation rate (Brown et al., 1981). That this complexity at the phe-
notypic level is mirrored at the genotypic level is confirmed by the iden-
tification of both nuclear (Aguilera et al., 1982; Sugden and Oliver,
1983) and cytoplasmic (Brown et al., 1984) mutations which produce an
ethanol-sensitive phenotype. The above description of the physiological
and genetic characteristics of yeast ethanol tolerance makes it unlikely
that conventional agar plate screening techniques will succeed in isolat-
ing tolerant mutants, and this has been found to be the case (Ismail and
Ali, 197la,b).
In this situation when only small quantitative increases are likely to be
obtained as a result of multiple mutations, the use of continuous selection
is indicated (Harder et al., 1977; Dykhuizen and Hartl, 1983). However,
the very complexity of the inhibitory effects of ethanol on yeast makes it
difficult to design a suitable selection regime. For this reason, a system in
which the intensity of selection was determined by the culture itself via a
feedback control circuit was adopted by Brown and Oliver (1982b). The
feedback selection system employed is diagrammed in Fig. 8. An infrared
analyzer was used to monitor continuously the fermentation activity of
the culture by determining the concentration of carbon dioxide in the exit
gas. When this concentration equaled or exceeded the value set on the
potentiometric controller, a relay closed. This switched on a peristaltic
pump and intr~duced ethanol into the culture vessel. When the addition
FRESH
MEDIUM
Figure 8. Continuous culture apparatus for feedback selection of ethanol-tolerant yeast

mutants. (From Brown and Oliver, 1982b.)
of the alcohol had reduced the carbon dioxide concentration of the exit
gas to below the value set by the controller, the relay opened and the
ethanol pump was switched off. The ethanol was then gradually diluted
out of the fermenter vessel and the rate of carbon dioxide production
increased. When the carbon dioxide concentration in the exit gas exceed-
ed the value set by the controller once more, the pump was switched on
again. Therefore, the system was functionally analogous to a turbidostat.
However, it was the rate of carbon dioxide production by the culture,
60
?.:
!i
li:
50
40
•
•
•
•
• .
•••
•
~ 30
!;
z
20 _,~
'-!~'\
~~
t-UJ ~
00 00 200 :m 400 600 700 eoo
RUN TIME (HR.)
Figure 9. Improvement in ethanol tolerance of the culture. The frequency of switching of
the ethanol pump in response to an increase in C0 2 concentration of the exit gas is shown.
Each point represents a 7-hr moving average of the interval between operations of the
pump. (From Brown and Oliver, 1982b.)
rather than its turbidity, which was held constant and the supply of
inhibitor, rather than nutrients, which was regulatory.
The system indicated an improvement in the ethanol tolerance of
the culture by an increase in the frequency with which the ethanol pump
was switched on by the control system. The pump did not switch on at all
for the first 12 days of the experiment. After that, it switched on occa-
sionally until, 27 days (650 hr) into the experiment, there was a dramatic
increase in the frequency with which the pump switched on (Fig. 9). This
0·65
0·45
50
45
c 40
3 35
0
j 30
it; 25
20
15
t=]~"\
~.~!l ~
~90
~
:s 80
~ /
0
70 -'-o-'sr-'......to-..,-t's-2....
~
0;....,2,.--S_,.3'o-3...,...s-40.-
• ....,4s_so...,...,.....,s's
Time (days)
Figure 10. Changes in the performance of the culture during feedback continuous selec-
tion. x-x cell viability; 0-0, dry weight; D.-D., ethanol concentration; \1-\1, specific
rate of production of C02 • These parameters were measured every 2 days; the graph
shows mean values for successive 6-day periods. (From Brown and Oliver, 1982b.)
increase in switching coincided with a number of changes in the physio-

logical state of the culture (Fig. 10). The specific rate of carbon dioxide
production and the concentration of ethanol in the growth medium
increased, as did the viability of the culture. These changes were accom-
panied by a fall in the concentration of biomass in the culture vessel, as
predicted by analogy with the turbidostat. The abruptness of these
changes might be interpreted as being the result of some particular
mutational event. This seems unlikely; mutants with a range of phe-
notypes were obtained and genetic analysis showed that these carried
mutations in more than one gene.
In the first instance, mutants were isolated by selecting cells that
formed colonies on 12% (wt./vol.) ethanol plates. This concentration of
alcohol was lethal to the parent strain used in the experiment. The
fermentation performance of these mutants was subsequently investi-
gated and the best were found to ferment at twice the rate of the parent
strain in the presence of 10% (wt./vol.) ethanol (Table II), were selected
on the basis of an enhanced overal fermentation rate, the Ki of ethanol
inhibition of fermentation being little changed from that in the wild
type. Others, such as SB154 and SB155, had an overall fermentation
rate no better that that of the wild-type strain, but the Ki of ethanol for
fermentation inhibition was significantly increased. Mutants of the first
class may find immediate application in the commercial production of
ethanol. Indeed, SB 160 has been shown to reduce significantly attenua-
tion times in batch fermentation on a pilot-plant scale. However, it is
likely that mutants of the second class will be more valuable in the long
run if they are used as starting material for further strain development.
Although the mutants isolated could survive and ferment at en-
hanced rates in the presence of high concentrations of alcohol, the selec-
tion regime never exposed the culture to more than 5% {wt./vol.) eth-
anol. This illustrates the advantage of allowing the culture to determine
the intensity of selection via the feedback system. If the selection had
Table II. Comparison of the Fermentation Performance

of Mutants of Saccharomyces uvarum Isolated by
Feed-Back Continuous Selection
Fermentation Rate Ki (fermentation)

Strain (QC02; gg-1 h- 1) (% w/v ethanol)
5d-cyc (wild type) 24.7 4.4

SB154 20.4 8.2
SB155 19.1 5.1
SB159 30.5 4.6
SB160 43.8 4.6
been imposed externally, then ethanol concentrations higher than 5%

(wt./vol.) would certainly have been used.
8. FWCCULATION
The ability of yeast to flocculate, that is, to form large aggregates of
cells that will fall to the bottom of the culture vessel toward the end of
the flocculation process, has long been considered a desirable trait for
brewing yeasts to possess (Rose, 1984). Both the biochemistry and genet-
ics of the process are complex and poorly understood. This situation has
not been helped by the fact that flocculation ability has been bred out of
many genetically defined laboratory strains, and it has only been in
comparatively recent years that isogenic flocculent and nonflocculent
pairs of haploid strains have been available (Russell et al., 1980). A fuller
understanding of the process of flocculation is required to enable its
exploitation as a cheap, easy means of separating cells and culture liquor
in modern biotechnological processes. It is a highly desirable trait, for
instance, in strains used for ethanol production when continuous culture
with cell recycle is employed (Cysewski and Wilke, 1976). If vectors
containing a cloned flocculation gene could be constructed, they might
provide a self-selecting system to ensure plasmid stability, or, at least, the
preferential retention of plasmid-containing cells in continuous culture
U· R. johnston and D. R. Berry, personal communication).
The physical process of flocculation is complex, and it appears that
an initial input of mechanical energy is required to overcome the mutual
repulsion between individual yeast cells (Stratford and Keenan, 1987).
Once this barrier is overcome, it is evident that floc formation requires
the presence of divalent cations, particularly Ca2 +. Most theories of
flocculation view Ca2+ as providing a bridge between anionic groups on
the walls of yeast cells (Mill, 1964).
The source of these groups is far from dear, and both proteins
Uayatissa and Rose, '1976; Beavan et al., 1979; Stewart et al., 1975) and
polysaccharides (Stratford and Keenan, 1987; Lyons and Hough, 1970,
1971) have been suggested as candidates. A completely different class of
hypothesis is that glycoproteins act to cross-link yeast cells in a manner
analogous to that of plant lectins (Taylor and Orton, 1978; Miki et al.,
1982). The role of calcium would then be seen as promoting conforma-
tional changes in these lectinlike molecules. The idea that wall proteins
are involved is supported by experiments demonstrating that treatn1ent
of whole cells with proteolytic enzymes can remove their flocculation
ability (Nishihara et al., 1977, 1982; Hodgson et al., 1985). Moreover,
Holmberg ( 1978) has identified cell wall proteins that are flocculation-
specific by comparing extracts from isogenic flocculent and nonfloc-
culent strains.
The investigation of the genetic control of flocculation has a con-
"CLASSICAL" YEAST BlafECHNOWGY 237
fused history owing to the identification of a number of genes in differ-

ent laboratories that subsequently proved to be allelic (Russell et al.,
1980). Currently, six genes are believed to be able, separately, to confer
flocculation ability on yeast (Hodgson et al., 1985; Johnston and Reader,
1983). Three of these genes are dominant and three recessive, although
this may be conditioned by the genetic background. A nuclear gene
designated fsul (Holmberg and Kielland-Brandt, 1978) acts as a sup-
pressor of flocculation ability and, in general, the cytoplasmic petite
mutation prevents flocculation (Holmberg and Kielland-Brandt, 1978;
Wilkie and Mudd, 1981). The gene FWJ has been mapped to chromo-
some I, some 37 eM (centimorgans) from ADEJ; the genes previously
identified as FW2 and FW4 are now considered to be alleles of FWJ
(Russell et al., 1980). The second dominant gene, FW5, confers stronger
flocculation ability than does FWJ and is not linked to the latter gene.
Yamashita and Fukui (1983a and b) have identified a third dominant
flocculence gene, FW8, which maps to chromosome VIII, being linked
to ARG4. This gene and its recessive allele, flo8, show an interesting
behavior in respect to dominance. All FW8/FW8 homozygotes are floc-
culent but FW8/flo8 heterozygotes are flocculent only in diploids that
are homozygous at the mating-type locus (ala or a/a diploids); ala
diploids do not flocculate. It may be that the recessive allele is the result
of the insertion of the yeast Ty transposon adjacent to the gene, thus
creating a ROAM mutation (Errede et al., 1980) which puts flo8 ex-
pression under mating-type control.
The fact that flocculation is a dominant character when expressed
by the FWJ, FW4, or FW8 gene should facilitate its transfer to indus-
trial strains. Stewart (1981) attempted such a transfer via the technique
of protoplast fusion. The fusion of a flocculent laboratory haploid to a
nonflocculent brewing yeast (or vice versa) produced flocculent hybrids,
but these hybrids failed to produce palatable beer. This illustrates a
major problem with the technique of protoplast fusion. Complete ge-
nomes are combined and undesirable traits may be recruited into the
hybrid in addition to the desired trait. Recombinant DNA technology
offers the promise of converting nonflocculent strains to flocculent ones
by the specific recruitment of a single gene. The announcement that a
DNA fragment conferring flocculation ability has been cloned (Watari et
al., 1987) suggests that this promise will be fulfilled. However, the fact
that the cloned fragment does not map to chromosome I, in spite of the
fact that the gene bank was made from a FWJ strain, gives some cause
for concern.
9. POLYSACCHARIDE UTILIZATION
The extension of the substrate range of S. cerevisiae to include poly-
saccharides such as starch and cellulose has excited much interest in
recent years. In the brewing industry, amylolytic enzymes are added to

the mash in the production of "lite" beers and to supplement the sac-
charification capacity of malt when starch adjuncts are used (see Section
3). More significant is probably the use of starch as a substrate for the
production of industrial ethanol by fermentation. In the United States,
the low-grade starch residues left from the enzymic processing of grain
into high-fructose syrups offers a cheap substrate for ethanol produc-
tion (Tubb, 1986). Enzymic conversion of starch to fermentable sugars
has now largely replaced acid hydrolysis (Fogarty and Kelly, 1980), and if
yeast could be given the ability to produce amylases itself, then the
production of ethanol from starch would be much simplified. The cost
of sufficient glucoamylase to saccharify 1 ton of starch is ca. $8 at 1986
prices (Schenberg Frascino and Da Costa, 1987). The cost savings would
be increased to ca. $10/ton if yeast were able to process raw starch itself
and thereby remove the need for precooking (Tubb, 1986). While these
savings are fairly modest, the total sums involved are huge on an indus-
try-wide basis. Further economic benefits might also accrue if the
glucoamylase produced by the yeast could be retrieved and sold to corn
syrup producers.
In Brazil, there are a number of advantages to the use of starch,
rather than sugar cane, as a substrate for ethanol production. If cassava
(Manihot esculenta) as well as sugar cane could be used for ethanol pro-
duction, then a number of problems related to soil exhaustion and dis-
ease which stem from crop monoculture should be relieved (Carioca,
1984). Moreover, cassava can grow on poorer soils and over a greater
climatic range than sugar cane. Its use would therefore release rich
agricultural land for food production and permit the spread of the
ethanol industry into poorer regions of Brazil where new sources of
employment are badly needed. Although cassava is an annual crop, its
use is less seasonal than sugar cane since roots may be stored in the soil.
Thus cassava, whether used on its own or together with sugar cane,
would relieve the seasonal fluctuations in ethanol production and pro-
vide some insulation against variations in the world price of sugar. It is
estimated that 180 liters of ethanol can be obtained from 1 tonne of
cassava using existing technologies (Carioca, 1984).
Since S. cerevisiae is unable to hydrolyze starch, new genes must be
recruited into the organism to give it this ability. While this is an obvious
candidate for the exploitation of recombinant DNA technology, a
number of nonconventional genetic routes have been used to achieve
this end.
10. RARE MATING

Most industrial strains of yeast are not capable of mating since they
have a polyploid genetic constitution. However, mass matings of labora-
tory haploids with industrial strains produce occasional hybrid organ-

isms as a result of rare-mating events (Gunge and Nakatomi, 1972;
Spencer and Spencer, 1977). These matings probably result from a sin-
gle mating-type transposition event at the MAT locus of one of the chro-
mosomes III of the industrial strains since commercial lager yeasts, at
least, appear to be triploid (Kielland-Brandt et al., 1983). A major prob-
lem with the rare-mating technique, as with protoplast fusion (see Sec-
tion 11), is to develop some way of selecting the hybrid organisms pro-
duced. A convenient method is to mate an auxotrophic haploid strain
with a cytoplasmic petite mutant of the commercial yeast; prototrophic
grande may then be selected on minimal medium plus glycerol (Spencer
et al., 1981).
Tubb and his colleagues (1981) exploited this technique to mate a
starch-utilizing haploid strain of S. diastaticus with a commercial lager
yeast. The initial hybrids were unsuitable for use in brewing since they
gave a phenolic off-flavor due to the production of 4-vinylguaiacol. This
undesirable characteristic was found to be determined by a single nu-
clear gene (POF), and it was found possible, in subsequent experiments,
to isolate pof- segregants (Goodey and Tubb, 1982). Nevertheless, the
hybrid strain reduced the amount of dextrin in beer to only a limited
extent since the S. diastaticus enzyme is unable to degrade the 13-1,6
branch points in the starch molecule.
11. PROTOPLAST FUSION
Protoplast fusion is a more efficient way of producing hybrids be-

tween strains that do not normally mate since it does not rely on mating-
type switching events. It is essential to have some way of selecting for the
desired hybrid and against the parent strains, and the utility of the
cytoplasmic petite mutation (which may be induced in polyploid com-
mercial strains) has already been referred to. An alternate approach is
the so-called "dead donor" technique in which the mating partner which
carries no genetic markers is killed by UV irradiation or heat treatment
prior to the fusion (Hockney and Freeman, 1980). Ferenczy and Kucsera
( 1985) have taken this technique one step further and developed a meth-
od whereby the hybrid products of a protoplast fusion event may be
selected even when neither of the parental strains contains a genetic
marker. This method, which I call the Lazarus technique, kills both
parental strains by use of metabolic poisons. Poisons that attack a differ-
ent class of enzymes are used for each parent, e.g., N-ethylmaleimide for
one, and myconazole for the other. This results in the death of both
parents, which may be brought back to life on fusion by phenotypic
complementation. Each parent supplies a functional set of enzymes
which the other lacks.
A number of workers have employed protoplast fusion to confer

amylolytic activity on either brewery or distillery yeasts. Hockney and
Freeman ( 1980) exploited both the dead-donor technique and the natu-
ral biotin auxotrophy of brewing strains to hybridizeS. diastaticus and S.
cerevisiae. Wilson et al. (1982) used complementation of auxotrophies
between two haploid strains in the fusion of Schwanniomyces alluvius and
Saccharomyces uvarum. Two groups of Brazilian researchers (Galembeck
et al., 1982; Echeverrigaray, 1983) have used protoplast fusion to form
hybrids between the starch-degrading yeast Lipomyces konoenkoae and S.
cerevisiae. Hybrids were selected as being able to grown on starch at 37°C
since S. cerevisiae cannot utilize starch and L. kononenkoae is unable to
survive at elevated temperatures. While successful fusants, capable of
converting starch to ethanol, were obtained in all cases, the instability of
the hybrids in the absence of selection was a universal problem.
12. RECOMBINANT DNA TECHNOLOGY
The techniques of gene cloning and transformation have been used

to transfer genes encoding amylolytic and cellulolytic enzymes into yeast
from a variety of sources. A problem arises when the recipient is a
commercial yeast strain since the complementation of an auxotrophic
marker by a yeast gene cloned in the vector may not be used as a way of
selecting transformanants. What is required is a dominant selectable
marker such as the drug resistances commonly used in the genetic ma-
nipulation of bacteria. Unfortunately, genes conferring resistance to
drugs in yeast and other eukaryotes are frequently recessive since re-
sistance is due to a structural alteration of the drug's target, rather than
to destruction of the drug itself. An exception to this is resistance to the
aminoglycoside antibiotic G418 which may be conferred upon yeast by
the bacterial transposon Tn5 Uimenez and Davies, 1980). A dominant
selectable marker obtainable from yeast itself is the CUP I gene, which
confers resistance to high levels of Cu2 + ions in a dosage-dependent
manner (Fogel and Welch, 1982). Either of these genes may be incorpo-
rated into yeast cloning vectors and used to select transformants of com-
mercial strains (Henderson et al., 1985).
12.1. Cloning and Expression in Yeast of Genes Encoding

Amylolytic Enzymes
A wide range of yeast species have the ability to utilize starch, but
these do not include the varieties of S. cerevisiae used currently in the
production of beer or industrial alcohol. In addition to these amylolytic
yeasts, genes encoding suitable enzymes may also be sought in the fila-
mentous fungi, bacteria, plants, and mammals.
Much work has been invested in the cloning and expression of
amyloglucosidase genes from S. diastaticus. This amylolytic yeast is now
regarded as a variety of S. cerevisiae and the transfer of the required
genes from S. diastaticus has the twin advantages that there should be no
barriers to their expression and no problems with the regulatory au-
thorities. The difference between S. cerevisiae and S. diastaticus in their
production of amylolytic enzymes appears to be mainly a function of
gene regulation. S. cerevisiae produces an amyloglucosidase enzyme dur-
ing sporulation, where it is believed to function in glycogen utilization
(Colonna and Magee, 1978). In contrast, S. diastaticus produces an
amylo-1 ,4-glucosidase throughout vegetative growth as the result of the
action of one or more of a series of genes denoted DEX or STA (Erratt
and Stewart, 1978). There is no evidence that the S. diastaticus enzyme
has any ability to degrade past the ~-1 ,6-linkage at the branch points of
the starch molecule, and this severely limits its usefulness. Two groups
have cloned genes from S. diastaticus which they separately call DEXJ
(Tamaki, 1978) and STAJ (Meaden et al., 1985); a comparison of their
restriction maps indicates that they are probably allelic. In addition to its
direct use in starch conversion, the cloned DEX gene may be used as a
positively selectable marker inS. cerevisiae transformations and also has
potential in secretion vectors.
A number of form-species of the fungus Aspergillus produce large
amounts of extracellular amyloglucosidase. Industrial strains of A. niger
or A. awamori, for instance, produce as much as 20 g/1- 1 of the enzyme
(Van Brunt, 1986). The successful expression of S. cerevisiae genes in
Escherichia coli encouraged the belief that the expression of fungal genes
in yeast should be relatively simple. This turned out not to be the case
since the Aspergillus amyloglucosidase gene contains four introns, which
S. cerevisiae is unable to splice, and this prevents expression of the ge-
nomic copy of the gene in yeast. The intron problem has been solved by
two groups (Boel et al., 1984; Innis et al., 1985) by cloning a eDNA copy
identified using an oligonucleotide probe. The eDNA gene may be ex-
pressed in yeast from a high-efficiency promoter, such as ENOl (Innis et
al., 1985). The product is glyosylated in yeast to a similar extent to the
native enzyme, and more than 90% of the activity is found in the culture
medium. The form of the Aspergillus enzyme produced is capable of
degrading raw starch since it has debranching activity and thus offers a
distinct advantage over the S. diastaticus enzyme. A merit of the latter, for
the production of beer, if not industrial ethanol, is that it is readily
inactivated by heat (Tubb, 1986).
The ubiquity of signal peptide sequences has permitted the ex-
pression in, and secretion from, yeast of a-amylase enzymes encoded by
eDNA clones from either wheat (Rothstein et al., 1984) or mouse

(Thomsen, 1983). The coding sequence of the mouse a-amylase has also
been fused to the signal peptide of the yeast pheromone (Astolfi Filho et
al., 1986). Very efficient secretion of the amylase is achieved by this
route, although a direct comparison of the efficiency of the yeast and
mouse signal sequences has yet to appear.
12.2. Cloning and Expression of Endoglucanase Genes in Yeast

Lignocellulosic agricultural wastes represent an enormous, and
largely untapped, carbon and energy source for biotechnological pro-
cesses. In Brazil alone more than 71 million tonnes of waste is produced
every year from the cultivation of sugar cane, cassava, corn, rice, and
soya. More than half of this waste is cellulose, and a number of genes
encoding enzymes involved in cellulose degradation have been cloned
and expressed in yeast (see Van Brunt, 1986). However, the accessibility
of the cellulose is a problem both biochemically, due to its association
with lignin, and physically. The latter problem is likely to favor filamen-
tous organisms, rather than yeasts, for the initial processing of these
agricultural wastes. However, the synthesis by yeast of one class of en-
zymes involved in cellulose degration, the endoglucanases, has impor-
tant implications for traditional yeast biotechnologies such as beer brew-
ing. Excess p-glucans in beer may lead to the formation of gels, hazes,
and precipitates which compromise the quality of the product and lead
to problems with its filterability (Cantwell et al., 1986; Knowles et al.,
1986).
The construction of a yeast strain that is able to degrade ~-glucans
requires not only expression of the ~-glucanase gene in yeast, but effi-
cient secretion of the enzyme into the medium. A number of groups
have cloned and expressed bacterial genes encoding endo-~-1,3-1,4-
glucanase in yeast. Both Bacillus subtilis (Hinchliffe and Box, 1984; Cant-
well et al., 1986) and Clostridium thermocellum (Sacco et al., 1984) have
been used as sources of this gene, but there is no indication that the
enzyme is excreted from S. cerevisiae. An alternative approach is to link
the coding sequence of the bacterial gene to the signal sequence of a
yeast gene that specifies a secreted product. This strategy was adopted
by the Allelix group (Skipper et al., 1985), who used the signal sequence
of the yeast killer peptide to obtain the secretion of a bacterial endo-
glucanase specified by a gene cloned from Cellulomonas fimi.
The preceding account suggests that prokaryotic signal sequences
are not functional in yeast. Eukaryotic ones, on the other hand, appear
to be ubiquitous. Jackson et al. ( 1986) exploited the ability of the mouse
a-amylase signal peptide to direct protein excretion in yeast (see Section
12.1) to achiev~ the excretion of the barley endo-P-1,3,-1,4-glucanase.
Two groups have cloned eDNA copies of two endo-p-1,4-glucanase

genes, egll and egl2, from. the cellulolytic fungus Trichoderma reesei
(Arsdell et al., 1987; Pentilla et al., 1987). Both these enzymes are
glycoslyated and excreted by yeast, the native signal sequence being
used. When expressed at high level using the yeast PGK promoter, the
Trichoderma enzymes alter the morphology of the host ye~st cells, making
them larger and of irregular shape (Pentilla et al., 1987)., This effect
could be due to a weak effect of endoglucanases on the chitin of the yeast
bud scars.
12.3. Cloning and Expression of Complete Metabolic Pathways:

The Way Ahead
The use ofrecombinant DNA technology to improve the range and
efficiency of yeast's activities in classical biotechnological processes has so
far been limited to the recruitment of just one or two genes which
specify novel activities. However, the technology currently available for
yeast should permit the wholesale genetic engineering of the organism
by addition of complete, novel metabolic pathways. A start in this direc-
tion has been made by the assembly of all the genes required for the
complete yeast tryptophan biosynthetic pathway on a single plasmid
(Niederberger et al., 1984; Prasad et al., 1987). It should be possible to
extend this approach to recruit complete pathways from heterologous
sources and also to construct novel pathways that do not presently exist
in nature. The ability to construct artificial yeast chromosomes (Murray
and Szostak, 1983) means that there is no obvious limit to the length of
such pathways or to our conceptual horizons in yeast strain develop-
ment.
REFERENCES
Aguilera, A., del Castillo, L., and Benitez, T., 1982, Alcohol and sucrose tolerant wine yeast
strains, Cien. Biol. (Portugal) 7:89-94.
Aiba, S., Shoda, M., and Nagatani, M., 1968, Kinetics of product inhibition in alcohol
fermentation, Biotechnol. Bioeng. 10:845-864.
Andreasen, A. A., and Stier, T. J. B., 1953, Anaerobic nutrition of Saccharomyces cerevisiae.
I. Ergosterol requirement for growth in defined medium,]. Cell. Comp. Physiol. 41:23-
27.
Andreasen, A. A., and Stier, T. J. B., 1954, Anaerobic nutrition of Saccharomyces cerevisiae.
II. Unsaturated fatty acid requirement for growth in defined medium,]. Cell. Comp.
Physiol. 43:271-277.
Arsdeli,J. N., Kivok, S., Schweickart, V. L., Ladner, M. B., Gelfand, D. H., and Innis, M.
A., 1987, Cloning, characterization and expression in Saccharomyces cerevisiae of endo-
glucanase I from Trichoderma reesei, Rio/Technology 5:60-64.
Astolfi Filho, S., Galembeck, E. V., Faria,J. B., and Schenberg Frascino, A. C., 1986, Stable
yeast transformants that secrete functional a-amylase encoded by cloned mouse pan-
creatic eDNA, Bio/Technology 4:311-315.
Ault, R. G., Hampton, A. N., Newton, R., and Roberts, R. H., 1966. Biological and bio-
chemical aspects of tower fermentation,]. lnst. Brew. 75:260-277.
Beavan, M. J., Belk, D. M., Stewart, G. G., and Rose, A. H., 1979, Changes in elec-
trophoretic mobility and lytic enzyme activity associated with development of floc-
culating in Saccharomyces cerevisiae, Can.]. Microbiol. 25:68-74.
Boel, E., Hjort, I., Svensson, B., Norris, F., Norris, K. E., and Fiil, N. P., 1984,
Glucoamylases E I and E2 from Aspergillus niger are synthesised from two different but
closely related mRNAs, EMBO]. 3:1097-1102.
Brown, S. W., 1983, Ethanol tolerance in the yeast Saccharomyces. Ph.D. thesis, University of
Kent at Canterbury.
Brown, S. W., and Oliver, S. G., 1982a, The effect of temperature on the ethanol tolerance
of the yeast Saccharomyces uvarum, Biotechnol. LeU. 4:269-274.
Brown, S. W., and Oliver, S. G., 1982b, Isolation of ethanol tolerant mutants of yeast by
continuous selection, Eur.J. Appl. Microbiol. Biotechnol. 16:119-122.
Brown, S. W., Oliver, S. G., Harrison, D. E. F., and Righelato, R. C., 1981, Ethanol
inhibition of yeast growth and fermentation: Differences in the magnitude and com-
plexity of the effect, Eur.]. Appl. Microbiol. Biotechnol. 11:151-155.
Brown, S. W., Sugden, D. A., and Oliver, S. G., 1984, Ethanol production and tolerance in
grade and petite yeast,]. Chem. Technol. Biotechnol. 34(13):116-120.
Burrows, S., 1979, Bakers' yeast, Econ. Microbiol. 4:31-64.
Cantwell, B. A., Brazil, G., Murphy, N., and McConnell, 1986, Comparison of expression
of the endo-p-1,3-1,4-glucanase gene from Bracillus subtilis in Sacchromyces cerevisiae
· from the CYCI and ADHl promoters, Curr. Genet. 11:65-70.
Carioca, J. 0. B., 1984, Potencial da biomassa, in: Biomassa, Fundamentos and Aplicaeoes
technologias (J. 0. B. Carioca and H. L. Arora, eds.), Universidade Federal do Ceara e
Ministerio do Interior, Fortazela, Brazil, pp. 65-77.
Colonna, W. J., and Magee, P. T., 1978, Glycogenolytic enzymes in sporulating yeast,].
Bacteriol. 134:844-853.
Corrao, H. S., 1975, A History rf Brewing, David and Charles, London.
Coutts, M. W., 1961, British patent 872,391-872,400.
Coutts, M. W., 1966, The many facets of continuous fermentation, Proc. Australian Section
lnst. Brew. (Ninth Convention), pp. 1-8, Institute of Brewing, Sydney.
Cysewski, G. R., and Wilke, C. R., 1976, Utilization of cellulosic materials through en-
zymatic hydrolysis. I. Fermentation of hydrolysate to ethanol and single-cell protein,
Biotechnol. Bioeng. 18:1297-1313.
Dykhuizen, D. E., and Hartl, D. L., 1983, Selection in chemostats, Microbiol. Rev. 47:150-
168.
Echeverrigaray, S. L., 1983, Estabilidade Gen~tica e Heterose em Hybridos Interspecfficos
de Leveduras. M.Sc. thesis, Universidade de Sao Paulo, Piracicaba, Brazil.
Erratt, J. A., and Stewart, G. G., 1978, Genetic and biochemical studies of yeast strains able
to utilise dextrins,J. Am. Soc. Brew. Chem. 36:151-161. '
Errede, B., Cardillo, T. S., Wever, G., and Sherman, F., 1980, ROAM mutations causing
increased expression of yeast genes: Their activation by signals directed toward con-
jugation functions and their formation by insertion of Tyl repetitive elements, Cell
45:593-602.
Ferenczy, L., and Kucsera, J., 1985, Gene transfer via chemically inactivated protoplasts ot
yeasts, Proc. 1Oth Internotional Specialized Symposium on Yeasts (Varna), p. I 03, Institute of
Molecular Biology, Sofia.
Fogarty, W. M., and Kelly, C. T., 1980, Amylases, amyloglucosidases and related glucanases,
in: Microbial Enzymes and Bioconversions (A. H. Rose, ed.), Academic Press, New York,
pp. 115-165.
Fogel, S., and Welch,J. W., 1982, Tandem gene amplification mediates copper resistance in
yeast, Proc. Natl. Acad. Sci. USA. 79:5342-5346.
Galembeck, E. V., Fernandes, B. L., Costa, S. 0. P., and Schenberg Frascino, A. C., 1982,
Fusion of protoplasts of different yeast genera: Saccharomyces and Lipomyces, Microb.
Genet. Bull. p. 524.
Goodey, A. R., and Tubb, R. S.. , 1982, Genetic and biochemical analysis of the ability of
Saccharomyces cerevisiae to decarboxylate cinnamic acids,]. Gen. Microbial. 128:2615-
2620.
Gray, W. D., 1941, Studies on alcohol tolerance of yeasts,]. Bacterial. 42:561-574.
Gunge, N., and Nakatomi, Y., 1972, Genetic mechanisms of rare matings of the yeast
Saccharomyces cerevisiae heterozygous for mating type, Genetics 70:41-58.
Hall, J. F., 1970, The use of Difco WLN agar for demonstration of the instability of strains
of Sacch. carlsbergensis,]. lnst. Brew. 76:522-523.
Harder, W., Kuenen, J. G., and Matin, A. A., 1977, Microbial selection in continuous
culture,]. Appl. Bacterial. 43:1-24.
Hayashida, S., Der Fong, D., and Hongo, M., 1975, Mechanism of formation of high
concentration alcohol in sake brewing. X. Physiological properties of yeast cells grown
in proteolipid-supplemented medium, Agr. Biol. Chern. 39: 1025-1031.
Henderson, R. C. A., Cox, B. S., and Tubb, R. S., 1985, Transformation of brewing
yeasts with a plasmid containing the gene for copper resistance, Curr. Genet. 9:133-
138.
Hinchliffe, E., and Box, W. G., 1984, Expression of the cloned endo-1,3-1,4-B-glucanase
gene of Bacillus subtilisin Saccharomyces cerevisiae, Curr. Genet. 8:471-475.
Hockney, R. C., and Freeman, R. F., 1980, Construction of polysaccharide-degrading
brewing yeast by protoplast fusion, in: Advances in Protoplast Research (L. Ferenczy and
G. Farkas, eds.), Pergamon Press, Oxford, pp. 139-144.
Hodgson, J. A., Berry, D. R., and Johnston, J. R., 1985, Discrimination by heat and pro-
teinase treatments between flocculent phenotypes conferred on Saccharomyces cerevisiae
by the gene FWI and FL05,]. Gen. Microbial. 131:3219-3227.
Holmberg, S., 1978, Isolation and characterisation of a polypeptide absent from non-
flocculent mutants of Saccharomyces cerevisiae, Carlsberg Res. Commun. 43:401-413.
Holmberg, S., and Kielland-Brandt, M. C., 1978, A mutant of Saccharomyces cerevisiae
temperature sensitive for flocculation: Influence of oxygen and respiratory deficiency
on flocculence, Carlsberg Res. Commun. 43:401-413.
Holzberg, 1., Finn, R. K., and Steinkraus, K. H., 1967, A kinetic study of the alcohol
fermentation of grape juice, Biotechnol. Bioeng. 9:413-427.
Hough, J. S., 1985, The Biotechnology rf Malting and Brewing, Cambridge University Press,
Cambridge.
Innis, M. A., Holland, M. J., McCabe, P. C., Cole, G. E., Wittmann, V. P., Tal, R., Watt, K.
W. K., Gelfand, D. H., Holland,]. P., and Meade,]. H., 1985, Expression, glycosyla-
tion and secretion of an Aspergillus glucoamylase by Saccharomyces cerevisiae, Science
228:21-26.
Ismail, A. A., and Ali, A. M. M., 1971a, Selection of high ethanol-yielding Saccharomyces. I.
Ethanol tolerance and the effect of training in Saccharomyces cerevisiae Hansen, Folia
Microbial. 16:350-354.
Ismail, A. A., and Ali, A.M. M., 1971b, selection of high ethanol-yielding Saccharomyces. II.
Genetics of ethanol tolerance, Folia Microbioll16:350-354.
Jackson, E. A., Ballance, G. M., and Thomsen, K. K., 1986, Construction of a yeast vector
directing the synthesis and release of barley ( 1-3, 1-4)-~-glucanase, Carlsberg Res. Com-
mun. 51:445-458.
jayatissa, P. M., and Rose, A. H., 1976, Role of wall phosphomannan in flocculation of
Saccharomyces cereivisiae,]. Gen. Microbial. 35:61-68.
jimenez, A., and Davies, j., 1980, Expression of a transposable antibiotic resistance ele-
ment in Saccharomyces, Nature 287:869-871.
Johnston, J. R., and Reader, H. P., 1983, Genetic control of flocculation, in: Yeast Genetics:
Fundamental and Applied Aspects U· F. T. Spencer, D. M. Spencer, and H. R. W. Smith,
eds.), Springer Verlag, New York, pp. 205-224.
Kielland-Brandt, M. C., Nilsson-Tillgren, N., Petersen, j. G. L., Holmberg, S., and Gjer-
mansen, C., 1983, Approaches to the genetic analysis and breeding of brewer's yeast,
in: Yeast Genetics. Fundamental and Applied Aspects U· F. T. Spencer, D. M. Spencer, and
A. R. W. Smith, eds.), Academic Press, New York, pp. 421-455.
Knowles, j., Lehtinen, U., Nikkola, M., Pentilla, M., Suihko, M-L., Home, S., Vilpola, A.,
and Enari, T-M., 1986, Glucanolytic brewer's yeast, Proc. 21st Eur. Brew. Conv. (Madrid),
p. 37, Institute of Brewing, Madrid.
Kodama, K., and Yoshikawa, K., 1977, Sake, Econ. Microbial. 1:423-475.
Lafon-Lafourcade, S., 1983, Wine and brandy, Biotechnology 5:81-163.
Lovgren, T., and Hautera, P., 1977, Maltose fermentation and leavening ability of bakers'
yeast, Eur. ]. Appl. Microbial. 4:37-43.
Lyons, T. P., and Hough, J. S., 1970, Flocculation of brewer's yeast,]. Inst. Brew. 76:564-
571.
Lyons, T. P., and Hough, j. S., 1971, Further evidence for the cross-bridging hypothesis
for flocculation of brewer's yeast,]. lnst. Brew. 77:300-305.
Maule, A. P., and Thomas, P. D., 1973, Strains of yeast lethal to brewery yeasts,]. Insi. Brew.
79:137-141.
Meaden, P., Ogden, K., Bussey, H., and Tubb, R. S., 1985, ADEX gene conferring produc-
tion of extracellular amyloglucosidase in yeast, Gene 34:325-334.
Miki, B. L.A., Poon, N. H., James, A. P., and Seligy, V. L., 1982, Possible mechanism for
flocculation interactions governed by gene FWI in Saccharomyces cerevisiae,J. Bacteriol.
150:878-889.
Mikola, J., Pietila, K., and Enari, T-M., 1972, Inactivation of malt peptidases during mask-
ing, J. Inst. Brew. 78:384-388.
Mill, P. J., 1964, The nature of interactions between flocculent cells in the flocculation of
Saccharomycn cerevisiae,]. Gen. Microbial. 96:165-174.
Murakami, H., Sagama, H., and Takase, S., 1968, Non-productivity of aflatoxin by Japa-
nese industrial strains of AspergiUus. Ill. Common characteristics of aflatoxin-produc-
ing strains,]. Gen. Appl. Microbial. (Tokyo) 14:251-262.
Murray, A. W., and Szostak, J. W., 1983, Construction of artificial chromosomes in yeast,
Nature 305:189-193.
Niederberger, P., Aebi, M., Furter, R., Prantl, F., and Hutter, R., 1984, Expression of an
artificial yeast TRP-gene duster in yeast and Escherichia coli, Mol. Gen. Genet: 195:481-
486.
Nishihara, H., Toraya, T., and Fukui, S., 1977, Effect of chemical modification of cell
surface components of a brewer's yeast on the floc-forming ability, Arch. Microbial.
150:890-899.
Nishihara, H., Toraya, T., and Fukui, S., 1982, Flocculation of all walls of brewer's yeast
and effects of metal ions, protein denaturants and enzyme treatments, Arch. Microbial.
131:112-115.
Oliver, S. G., 1984, Biological limits to ethanol production, Chem. Indust. 14:425-427.
Oura, E., Soumalainen, H., and Viskari, R.,1979, Breadmaking,Econ. Microbial. 4:88-146.
Palamund, S. R., and Hardwick, W. A., 1969, Studies on the relative flavour importance of
some beer constituents, Tech. Q. Master Brew. Assoc. Amer. 6:117-128.
Pentilla, M. E., Andre, L., Saloheimo, M., Lehtovaara, P., and Knowles, J. K. C., 1987,
Expression of Trichoderma reesei endoglucanases in the yeast Saccharomyces cerevisiae,

Yeast lJ: 175-187.
Prasad, R., Niederberger, P., and Hutter, R., 1987, Tryptophan accumulation in Sac-
charomyces cerevisiae under the influence of an artificial yeast TRP gene cluster, Yeast
3:95-105.
Preece, I. A., 1954, The Biochemistry of Brewing, Oliver and Boyd, Edinburgh.
Preece, I. A., and Hoggan, J., 1957, Carbohydrate modification during malting, Proceed-
ings of the European Brewery Convention Congress, Elsevier, London, pp. 72-83.
Righelato, R. C., 1980, Anaerobic fermentation: Alcohol production, Phil. Trans. Roy. Soc. B
295:491-500.
Rose, A. H., 1984, Physiology of cell aggregation; flocculation by Saccharomyces cerevisiae as
a model system, in: Microbial Adhesion and Aggregation (K. C. Marshall, ed.), Spring-
Verlag, Berlin, pp. 323-335.
Rothstein, S. J., Lazurus, C. M., Smith, W. E., Baulcombe, D. C., and Gatenby, A. A., 1984,
Secretion of a wheat a-amylase expressed in yeast, Nature 508:662-665.
Rubner, M., 1912, The physiology and nutrition of yeast during alcoholic fermentation,
Arch. Physiol. 1:1-392.
Russell, 1., Stewart, G. G., Reader, H. P.,Johnston,J. R., and Martin, P. A., 1980, Revised
nomenclature of genes that control yeast flocculation,]. lnst. Brew. 86:120-121.
Sacco, M., Millet, J., and Aubert, J. P., 1984, Cloning and expression in Saccharomyces
cerevisiae of a cellulase gene from Clostridium thermoceUum, Ann. Microbiol. lnst. Pasteur
1lJ5A:485-488.
Schenberg Frascino, A. C., and DaCosta, S. 0. P ., 1987, Molecular approaches to Alcohol
Biotechnology in Brazil, in: Critical Reviews in Biotechnology, Vol. 6 (G. G. Stewart and I.
Russell, eds.), CRC Press, Boca Raton, FL, pp. 323-355.
Skipper, N., Sutherland, M., Davies, R. W., Kilburn, D., Miller, R. C., Jr., Warren, A., and
Wong, R., 1985, Secretion of a bacterial cellulase from yeast, Science 250:958-960.
Spencer, J. F. T., and Spencer, D. M., 1977, Hybridisation of non-sporulating strains of
brewer's and distiller's yeasts,]. lnst. Brew. 83:287-289.
Spencer, J. F. T., Land, P., and Spencer, D. M., 1981, The use of mitochondrial mutants
in the isolation of hybrids obtained by protoplast fusion, Mol. Gen. Genet. 178:651-
654.
Stewart, G. G., 1981, The genetic manipulation of industrial yeast strains, Can.]. Microbial.
27:973-990.
Stewart, G. G., Russell, 1., and Garrison, I. F., 1975, Some considerations on the floccula-
tion characteristics of ale and lager yeast strains,]. Int. Brew. 81:248-257.
Stratford, M., and Keenan, M. H.J., 1987, Yeast flocculation: Kinetics and collision theory,
Yeast 5:201-206.
Sugden, D. A., and Oliver, S. G., 1983, Reduced ethanol tolerance: One of the pleiotropic
effects of the pep4.3 mutation in Saccharomyces cerevisiae, Biotechnol. Lett. 5:419-422.
Tamaki, H., 1978, Genetic studies of ability to ferment starch in Saccharomyces gene poly-
morphism, Mol. Gen. Genet. 164:205-209.
Taylor, N. W., and Orton, W. L., 1978, Aromatic compounds and sugars in flocculation of
Saccharomyces cerevisiae, ]. lnst. Brew. 84: 113-114.
Thomas, D. S., Hossack, J. A., and Rose, A. H., 1978, Plasma-membrane lipid composition
and ethanol tolerance in Saccharomyces cerevisiae, Arch. Microbiol. 117:239-245.
Thomsen, K. K., 1983, Mouse a-amylase synthesised by Saccharomyces cerevisiae is released
into the culture medium, Carlsberg Lab. Res. Commun. 48:545-555.
Thorne, R. S. W., 1970, Yeast mutation during continuous culture,]. lnst. Brew. 76:555-
563.
Troyer, J. R., 1953, A relation between cell multiplication and alcohol tolerance in yeasts,
Mycologia 45:20-39.
Tubb, R. S., 1986, Amylolytic yeasts for commercial applications, Trends Biotechnol. 4:96-
104.
Tubb, R. S., Brown, A. J.P., Searle, B. A., and Goodey, A. R., 1981, Development of new
techniques for the genetic manipulation of brewing yeasts, in: Current Developments in
Yeast Research (G. G. Stewart and I. Russell, eds.), Pergamon Press, Oxford, pp. 775-
779.
Van Brunt, J., 1986, Fungi: The perfect host? Rio/Technology 4:1057-1062.
Vogt, E., Jakob, L., Lemperle, E., and Weiss, E., 1974, Wein, Eugen Ulmer Verlag,
Stuttgart.
Watari, j., Takata, Y., Nishikawa, N ., and Kamada, K., 1987, Cloning of a gene controlling
yeast flocculence, Proceedings cf the 21st European Brewing Convention (Madrid), pp. 537-
544, Institute of Brewing, Madrid.
Wilkie, D., and Mudd, R. C., 1981, Aspects of mitochondrial control of cell surface charac-
teristics in Saccharomyces cerevisiae, in: Advances in Biotechnology: Current Developments in
Yeast Research (G. G. Stewart and I. Russell, eds.), Pergamon Press, Toronto, pp. 345-
349.
Wilson, j. j., Khachatourian, G. G., and Ingledew, W. M., 1982, Protoplast fusion in the
yeast, Saccharomyces alluvius, Mol. Gen. Genet. 186:95-100.
Yamashita, 1., and Fukui, S., 1983a, Molecular cloning of a glucoamylase gene in the yeast
Saccharomyces, Agric. Biol. Chem. 47:2689-2692.
Yamashita, S., and Fukui, S., 1983b, Mating signals control expression of both starch
fermentation genes and a novel flocculation gene FW8 in the yeast Saccharomyces,
Agric. Biol. Chem. 47:2889-2896.
Culture Systems 8
T. M. MATTHEWS and C. WEBB
1. INTRODUCTION
Many yeasts of the genus Saccharomyces have a long history of industrial

usage where the substrates are generally either agricultural or process-
ing wastes, or crops grown for the purpose. Principle examples of the
production of food yeast areS. uvarum, which is recovered from malt
wort in beer production, S. cerevisiae (baker's yeast), usually grown on
molasses, and S. fragilis, which is cultured on cheese whey. Molasses,
surplus grain, and sulfite waste liquor are the main raw materials used
for fuel ethanol production, usually by fermentation with S. uvarum or S.
cerevisiae. Cellulosic waste materials may also be used as a substrate after
preliminary acid hydrolysis to break down long-chain polysaccharides to
simple sugars. Many Saccharomyces species also play a large part in the
production of alcoholic beverages, where substrates are usually provided
by fruit, grain, or cereal crops.
The Saccharomyces yeasts can be grown in solutions of simple chem-
ical substances but growth is more rapid and economically viable when
the culture medium contains certain supplements.
Once a suitable culture medium satisfying the nutritional require-
ments of the Saccharomyces has been established, the physical and chem-
ical conditions for growth must be optimized for maximum cell yield or
growth rate. The aeration intensity, dissolved oxygen, and carbon diox-
ide concentrations in the medium can affect yeast growth and the tem-
perature and pH can also influence the fermentation. In addition, high
sugar and high ethanol concentrations can have inhibitory effects on
yeast growth. The intensity of these effects is generally strain depen-
dent; hence each parameter must be considered for each Saccharomyces
species to be cultured.
T. M. MATTHEWS and C. WEBB • Department of Chemical Engineering, University

of Manchester Institute for Science and Technology, Manchester M60 1QD, England
249
250 T. M. MATTHEWS and C. WEBB
2. NUTRITIONAL REQUIREMENTS OF SACCHAROMYCES
2.1. Carbon
Saccharomyces species of yeast are able to ferment the hexose sugars
of d-glucose, d-fructose, and d-mannose. The rate of fermentation of
mannose is less than that of glucose, and with some species, fructose is
fermented more slowly than glucose (Menzinsky, 1943).
The main sugar constituent of cane and beet molasses is sucrose, but
as the rate of inversion of sucrose by Saccharomyces yeasts is far in excess
of the rate of fermentation, a sucrose fermentation can be considered a
glucose and fructose fermentation (Atkin et al., 1946).
Maltose, the sugar present in wort, is not fermentable by all mem-
bers of the genus. However, fermentation of maltose by brewer's and
baker's yeasts will proceed and the fermentation rate may be increased
by the addition of a small quantity of glucose or oxygen to the culture
medium (Schultz et al., 1940; Leibowitz and Hestrin, 1939).
Other sugars that can be used as a carbon source by some, but not
all, Saccharomyces species include d-galactose, which can be fermented by
brewer's and baker's yeasts, and lactose, which cannot, however, be fer-
mented by S. cerevisiae.
All l-sugars and all pentoses can be considered unfermentable by
Saccharomyces yeasts, and there is evidence that some pentoses can retard
the rate of fermentation of hexoses by certain species (Sobotka et al.,
1936).
The carbohydrate present in some of the raw materials used indus-
trially in Saccharomyces fermentation, and their hydrolysis products, are
shown in Table I.
2.2. Nitrogen
Yeasts have a nitrogen content of around 10% of their dry weight;
hence nitrogen is an important constituent of any growth medium.
Many inorganic ammonium salts have been found to promote the
growth of Saccharomyces species and the most efficient are the following:
Ammonium acetate
Ammonium carbonate
Ammonium bicarbonate
Ammonium lactate
Mono-, di-, and triammonium phosphate
Ammonium sulfate
Ammonium tartrate
Ammonium sulfate is the most widely used nitrogen source as it also
provides a readily assimilable source of sulfur.
CULTURE SYSTEMS 251
Table I. The Carbohydrate Content of Raw Materials

Used in Saccharomyces Fermentation
Raw material Carbohydrate present Hydrolysis products
Cane sugar Sucrose Glucose

Beet sugar Fructose
Molasses
Cannery waste
Grains Starch Glucose
Cassava Maltose
Potatoes Mal to triose
Jerusalem artichokes Higher-molecular-weight
Sago dextrins•
Wood and wood Cellulose Glucose
waste Mannose
Agricultural waste Galactose
Municipal waste Xylose•
Arabinosea
Whey Lactose Glucose
Galactose
•Not fermentable by most Saccharomyces species.
Both baker's and brewer's yeasts are unable to assimilate nitrates, so

nitrates are generally not suitable as nitrogen sources in culture media
for the Saccharomyces.
Many amino acids provide suitable sources of nitrogen for yeast
growth. In a survey (Schultz and Pomper, 1948) of 40 strains of yeast all
classified as Saccharomyces and with an amino acid as the sole nitrogen
source, seven amino acids were found to promote good growth in all 40
yeast strains and a further seven were found to support medium to good
growth in around 90% of the yeasts tested. Five amino acids were found
to be inadequate nitrogen sources for the growth of these by Sac-
charomyces species. The amino acids tested are listed in Table II.
A balanced amino acid mixture has been shown to be a more effi-
cient nitrogen source than ammonium ions (Thorne, 1944; Hoggan,
1977). This could be due to the ability of yeasts to assimilate complete
mixtures of amino acids without preliminary deamination. Amino acid
takeup relieves the cell of a synthetic carbon demand which would other-
wise have to be met from sugar metabolism (Jones et al., 1969).
Urea can also be metabolized by Saccharomyces species as a nitrogen
source and is used as such industrially in India and South America;
however, for yeast growth comparable to that with ammonium sulfate as
nitrogen source, the vitamin biotin is usually added to the culture medi-
um to assist assimilation.
Table II. Suitability of Individual Amino Acids as Sole

Nitrogen Source for Growth of Saccharomyces Strains
Group 1: Good growth of all strains Alanine

Arginine
Asparagine
Aspartic acid
Glutanic acid
Leucine
Valine
Group 2: Good to moderate growth Isoleuaine
Methionine
Phenylalanine
Serine
Tryptophan
Group 3: Poor growth Cystine
Glycine
Histidine
Lysine
Proline
Threonine
2.3. Phosphorus
Phosphorus is essential for the growth of all yeasts, controlling the
synthesis of lipids and carbohydrates and maintaining the integrity of
the cell wall. Baker's and brewer's yeasts can grow on a medium without
phosphorus for a short time but the phosphate reserves within the yeast
cells are used for the growth (Markham and Bryne, 1968). Yeasts take up
phosphate as the monovalent anion H 2 P04 but the divalent anion
HPO~- is not absorbable (Rothstein, 1961). The amount of phosphate
assimilated depends on the quantity supplied. Potassium dihydrogen
phosphate is normally added in a concentration of around 0.6 mM/g
cells to culture media for optimum fermentation rates.
2.4. Sulfur
Saccharomyces species can obtain the sulfur they require from in-
organic sulfate, sulfite, or thiosulfite (Schultz and McManus, 1950),
which are reduced to the amino acid methionine within the cell (Lewis
and Wildenradt, 1969). Although methionine is the preferred sulfur
source, ammonium sulfate is generally chosen for industrial fermenta-
tions on the basis of cost. Sulfur constitutes about 0.4% of the dry weight
of yeast cells.
CULTURE SYSTEMS 253
Table III. Trace Elements and

the Sacclaaromyces
Group 1: Group 2: Group 3:

macroelements microelements inhibitors
(0.1-1 mM) (0.1-100 ,...M) (>100 ,...M)
Potassium Cobalt Silver

Magnesium Boron Arsenic
Calcium Cadmium Barium
Zinc Chromium Mercury
Iron Copper Lithium
Manganese Iodine Nickel
Chlorine Molybdenum Osmium
Nickel Lead
Vanadium Selenium
Tellurium
2.5. Trace Elements

The trace elements required by Saccharomyces for active growth have
been well documented and can be divided into three categories (Rose,
1976):
1. The macro elements, which are required in concentrations of
0.1-1.0 mM
2. The micro elements, which are required in concentrations of
0.1-100 .,._M
3. The inhibitors, which can adversely affect yeast growth when
present in concentrations greater than 100 ...,M
The elements in these groups are listed in Table III. The presence
of some of the elements in groups 1 and 2 in greater concentrations than
those stated has been found to be inhibitory to cell growth in some cases.
Growth media that contain organic compounds, for example, molasses
and wort, may protect the yeast from the influence of otherwise inhibito-
ry elements by precipitating them or forming complexes.
2.6. Growth Factors

Various growth factors are taken up during biosynthesis which re-
lieve the cell of the need to synthesize the particular compound and
hence save energy. These growth factors include vitamins, amino acids,
nucleic acids, fatty acids, and sterols.
One key growth factor throughout the Saccharomyces is inositol, and
a lack of this can lead to less efficient cell division (Smith, 1951) and
morphological changes within the cell wall (Power and Challinor, 1969;
Kirsop and Brown, 1972). Beet molasses contain sufficient concentra-
tions of inositol for optimum growth of the yeast (White and Munns,
1950), but with other substrates a minimum of 2 mg/liter is required.
Biotin and pantothenate are essential for all strains of Saccharomyces
(Williams et al., 1940) and some strains also require thiamin, pyridoxine,
p-aminobenzoic acid, niacin, folic acid, and riboflavin. The concentra-
tions of these vitamins required for optimum growth rates are strain
dependent, and with some species of the Saccharomyces the presence of
thiamin in the absence of pyridoxine can reduce growth (Williams et al.,
1940; Gordon and Stewart, 1972).
Environmental conditions may also dictate a requirement for one or
more growth-promoting substances; for example, ergosterol and un-
saturated fatty acids are essential for yeast growth under anaerobic con-
ditions (Nes et al., 1978) and choline, carnitine, or leucine is required by
some thermophilic yeast strains (Travassos and Cury, 1971 ).
Addition of all the necessary growth-promoting compounds to a
culture medium can be costly, and hence lower-than-optimal growth
rates become acceptable on consideration of the economics of the sys-
tem. Alternatively, adequate growth-promoting factors can be provided
by the addition of yeast extract to the fermentation medium.
3. PROCESS VARIABLES THAT INFLUENCE

THE GROWTH OF SACCHAROMYCES
3.1. Hydrogen Ion Concentration

Most yeast of the genus Saccharomyces will grow at pH values in the
range 2.4-8.6. The intracellular pH of S. cerevisiae is controlled within
the range 5.8-6.3 and has been shown to be independent of external pH
values ranging from 3 to 7 (Martiny, 1972). Yeast growth and fermenta-
tion rates do not appear to be affected by pH values varying between 3.5
and 6 in the bulk medium, and this is probably due to the tight control
over intracellular pH.
When sucrose is used as a carbon source, the system is more pH
sensitive than with glucose because yeast invertase activity is affected by
low pH values.
Most laboratory-scale and industrial Saccharomyces fermentations are
controlled at a pH between 4 and 5, which also helps reduce the risk of
bacterial contamination within the reactor.
CULTURE SYSTEMS 255
3.2. Temperature and Ethanol Inhibition Effects

Most strains of Saccharomyces will grow at any temperature between
oo and 40°C, though optimum temperature for maximum growth rate is
strain dependent and generally in the range 28-35°C (Walsh and Mar-
tin, 1977). However, when considering cell yield rather than rate of
growth, optimum temperatures tend to be somewhat lower, and in yeast
manufacture a temperature of 26°C is maintained for the first 6 hr of
the fermentation (Walter, 1940).
Growth of S. cerevisiae at temperatures near its tolerance limit (40°C)
results in disruption of sterl and fatty acid synthesis (Thompson and
Parks, 1974), destabilization of the plasma membrane, and a decrease in
cell viability (Krouwel and Braber, 1979). It has also been observed that
temperature has a marked effect on the mutation rate to respiratory
deficiency in S. cerevisiae, which can produce significant quantities of
such "petite" mutant cells at 40°C, though very few are detected at tem-
peratures below 35°C Uones et al., 1981 ).
In fermentations for ethanol production, slightly higher tem-
peratures are employed than for yeast cell production. Increasing the
temperature of a fermentation of S. cerevisiae on a glucose-based medi-
um from 30° to 39°C can result in a decrease in cell growth and an
increase in ethanol productivity and eventual cell death at about 39°C. In
general, the optimum temperature for ethanol production by Sac-
charomyces strains in 5-10°C higher than the optimum temperature for
growth (White and Munns, 1951). However, ethanol produced during
the fermentation has an inhibitory effect on yeast growth. In one study
addition of ethanol to cultures of S. cerevisiae and S. uvarum growing
exponentially at 23°C produced a 10% decrease in the growth rate at 2%
(wt./vol.) ethanol, a 50% decrease in 6% (wt./vol.) ethanol, and a 90%
decrease in 10% (wt./vol.) ethanol (Brown et al., 1981). This inhibition of
cell growth by high ethanol concentrations is greater at higher tem-
peratures (Gray, 1941; Nagodawithana et al., 1974). Table IV shows the
Table IV. Inhibitory Concen-

trations of Ethanol at Different
Temperatures
Temperature Ethanol
(OC) (wt./vol.)
9 9.5
18 8.3
27 7.5
36 3.8
ethanol concentrations that prevented further fermentation by S. cere-

visiae for a range of temperatures (Ranganathan and Bhat, 1958).
This effect is considered to be due to ethanol being produced within
the cell at a higher rate than it can be transported through the cell
membrane. This leads to enzyme inhibition and is followed by cell death
(Nagodawithana et al., 1974). For batch growth of the Saccharomyces,
lower temperatures will produce a higher cell yield. In continuous
culture or systems where ethanol is removed from the reactor, the eth-
anol inhibition effect is not so severe and higher temperatures can be
employed. Heating culture media can be costly; so for economic reasons
Saccharomyces cultures are generally maintained at temperatures between
25° and 30°C (Walter, 1940; Del Rosario et al., 1979; Eroshin et al., 1976).
3.3. Dissolved Oxygen and Substrate Inhibition Effects

Yeasts are unable to grow for more than four or five generations
under fully anaerobic conditions Oones et al., 1981). Addition of air or
oxygen to Saccharomyces fermentations is essential to maintain cell
viability. Complete oxidation of the sugar to carbon dioxide and water
will give optimum cell production. Oxygen acts as the final electron
acceptor in oxidative phosphorylation, during yeast respiration. Under
conditions of high dissolved oxygen concentrations, fermentation of the
sugar to ethanol is inhibited. This effect, the Pasteur effect, was first
noted by Pasteur in 1867 (Fiechter et al., 1981).
The two extremes of respiration and fermentation of glucose can be
represented by
fully aerobic G = -686 kcal

(respiration)
C6H1206
---fully anaerobic--. 2C2H 5 0H + 2C02 G =-54 kcal
(fermentation)
Respiration releases more energy than fermentation and is therefore the
preferred process. However, these two extremes are not realized in prac-
tice. Many Saccharomyces species are sensitive to glucose (Fiechter et al.,
1981) and their respiration is repressed in the presence of a concentra-
tion of glucose greater than 1.0 g/liter (Rickard and Hogan, 1978). Un-
der such conditions the biomass yield will decrease and ethanol will be
produced. This is known as the Crabtree effect or Contre-Pasteur ef-
fect. In a study of the Crabtree effect in various yeast strains, growing on
a medium containing 30 g/liter glucose, seven of eight Saccharomyces
species tested gave a positive Crabtree effect (de Deken, 1966). However,
CULTURE SYSTEMS 257
Table V. Effect of Initial Sugar Concentration

on the Specific Growth Rate of S. cerevisiae
Initial sugar concentration Specific growth rate

(% wt./vol.) (hr-1)
5 0.6
14 0.5
30 0.3
other research has concluded that the Crabtree effect should be consid-
ered a transient control mechanism and that a gradual derepression of
respiration will occur (Barford et al., 1980).
In the brewing industry the specific growth rate, viability, and yield
of the Saccharomyces species employed have been found to increase with
the level of oxygen concentration in the wort for levels of up to 20%
saturation. Higher dissolved oxygen levels do not affect the fermenta-
tion, indicating that oxygen levels of around 20% of saturation are nec-
essary for yeast cell maintenance and growth (Gordon and Stewart,
1972). More generally, the minimum level of oxygen required is strain
dependent and also can be dependent on environmental conditions.
Lower dissolved oxygen levels are required when unsaturated fatty acids
and sterols are present in the culture media, but higher levels are
needed if serine is present (Harding and Kirsop, 1979).
High concentrations of sugars also have an inhibitory effect on yeast
growth though some Saccharomyces species are more sugar tolerant than
others. In aerobic batch fermentations the specific growth rate of one
strain of S. cerevisiae has been shown to fall as the initial sugar concentra-
tion is increased (Ghose and Tyagi, 1979b), as shown in Table V.
The specific growth rate of a more sugar-tolerant S. uvarum strain
has been reported to be unaffected by sugar concentrations of up to 25%
(wt./vol.) (Del Rosario et al., 1979).
In the cultivation of Saccharomyces yeasts it is important that the
initial sugar concentration in the culture medium not be so high as to
prevent growth of the yeast.
3.4. Carbon Dioxide

Carbon dioxide, a by-product of yeast growth and ethanol produc-
tion, is inhibitory to both these processes under aerobic and anaerobic
conditions (Chen and Gutmanis, 1976; Kunkee and Ough, 1966). Car-
bon dioxide can affect the permeability and composition of yeast cell
membranes and can also shift the equilibrium in carboxylation/ decar-
boxylation reactions in the metabolic pathways of the yeast. The inhibi-

tion effects are greater in high ethanol concentrations and at low pH
values (Kunkee and Ough, 1966).
4. THE THEORY AND PRACTICE OF YEAST

CULTURE SYSTEMS
4.1. Batch Systems

Historically Saccharomyces yeasts have been grown batchwise in sub-
merged cultures, for the production of cells for food and fodder, and in
order to ferment liquid media to alcoholic beverages. When a single-cell
organism such as a Saccharomyces strain is grown in submerged culture, a
plot of the logarithm of the dry weight of cells produced against time
gives a characteristic curve dependent on strain and environmental con-
ditions. A typical growth curve, as shown in Fig. 1, consists of three
distinct sections: (A) the lag phase, (B) the exponential phase, (C) the
stationary phase. A growth acceleration phase may be considered to
follow the lag phase and precede the exponential phase, and similarly a
growth retardation phase may be considered to occur before the station-
ary phase.
The lag phase represents the time period between inoculation of the
culture medium with the organism and a measurable increase in cell
c
0
:;:.
....e
c
G)
0
5
0
1i
u A 8 c
Time
Figure 1. Idealized time course for yeast growth, showing (A) lag phase, (B) exponential
phase, (C) stationary phase.
CULTURE SYSTEMS 259
concentration. During this time the cells are adapting to their new en-
vironment. The lag phase can be shortened by using a large inoculum or
an inoculum culture that is already growing exponentially under similar
conditions. If the culture medium is near the optimum temperature for
the yeast growth and contains all the essential nutritional requirements
for the yeast, this will also decrease the apparent lag phase.
The exponential phase is the time period during which the specific
growth rate (1-L) is constant and at a maximum (1-Lmax) for the given strain
and the environmental conditions. To achieve a constant specific growth
rate individual cells must be budding at regular time intervals. The
increase in cells per unit time is a product of the specific growth rate (1-L)
and the amount of cells present at a given time (x) and can be repre-
sented by
dx
dt = 1-Lmax X or (1)
where x0 is the dry mass of cells present at the start of the exponential
growth phase.
Therefore,
(2)
By taking logarithms of both sides of equation (2)
In X - In X0 = ILrnax t (3)
Hence
_ In x- In X0
1-Lmax- t (4)
The maximum specific growth rate (1-Lmax) is given by the gradient of

section B of the growth curve (Fig. 1).
In general,
(5)
or
(6)
The mean generation time or doubling time (g) of the yeast population
can also be obtained from the growth curve or calculated as
ln 2x - ln x ln2 . 0.69
J.l.max = = - , I.e., g = - - (7)
g g J.l.max
As the nutrients for yeast cell growth in the culture medium become
limited and ethanol and carbon dioxide begin to accumulate, the in-
crease in cells per unit time will no longer be given by J.l.max· In 1950,
Monod (Burrows, 1970) proposed a theory of cell growth based on two
assumptions:
1. Microorganisms grow exponentially with a maximum. specific
growth rate J.l.max until limited by some environmental condition
to a lower specific growth rate J.t.
2. The specific growth rate J.t of any microbial culture is propor-
tional to the limiting substrate concentration S.
Assumption (a) can be written
(8)
Assumption (b) can be written
s (9)
J.l. = J.l.max S + Km
where Km is a constant numerically equal to the substrate concentrations
at which the specific growth rate is half its maximum value, as shown in
Fig. 2. Equation (8) is analogous to the Michaelis-Menten equation for
enzymic reactions and indicates that when S is very large, SIS + Km
tends to unity and J.t tends to J.l.max and when S = 0 then J.t = 0; i.e., if the
limiting substrate has been consumed, no further growth will occur.
Beyond this condition the accumulated cell mass therefore remains con-
stant and growth enters the stationary phase.
The type and size of vessels employed for batch fermentation of
Saccharomyces yeasts vary immensely. The extremes of the range are a
sterile Erlenmeyer flask and an open vat with a volume of hundreds of
cubic meters. A typical batch fermenter of the type that could be used
under sterile conditions in a laboratory (a stirred tank reactor) is shown
in Fig. 3. Temperatures, pH, DO (dissolved oxygen), and foam detecting
probes may be inserted for monitoring and controlling the fermenta-
tion. Ports are available for the addition of alkali, air, antifoam, and the
inoculum culture and for the removal of gases and sampling. Simple
agitation can be achieved by a magnetic stirrer, but more commonly a
CULTURE SYSTEMS 261
,_,
::1.
.......
IJ max
....0Ill
ct::
....
..r::.
J:
...0
{!)
IJ max
u
It=
'ij 2
Ill
a.
en
Km
Limiting Substrate Concentration (S)
Figure 2. Idealized representation of the effect of limiting substrate concentration on

specific growth rate.
mechanical stirrer is employed. More efficient agitation is achieved by

means of shaped baffles.
The commercial manufacture of Saccharomyces yeasts (Pyke, 1957;
Rhodes and Fletcher, 1966) usually begins with the transfer of a single
cell of the preferred Saccharomyces strains to a nutrient agar slope. When
grown, this culture is used to inoculate a flask of liquid medium contain-
ing around 6% (wt./vol.) sugar, which in turn will be transferred follow-
ing a suitable growth period to a larger fermentation vessel containing a
similar growth medium. Further transfers to progressively larger vol-
umes of media take place at intervals of 12-24 hr and aeration is in-
creased at each stage to maintain a steady dissolved oxygen concentra-
tion. The final production stage takes place when the production vessel
containing around 200 m 3 medium is inoculated with yeast suspension
to give an initial concentration in the reactor of around 20% wt./vol.
biomass. The cells in this final inoculum are usually in the exponential
phase of growth in order to minimize the lag phase. Throughout the
entire process the pH is maintained below 4.5 to reduce the risk of
bacterial contamination and above 3.5, since at lower pH values the color
of the yeast produced can be affected and if sucrose is the carbon source,
the yeast invertase activity may be affected. The temperature is initially
maintained at 25°C but is allowed to rise gradually to 30°C by the end of
the fermentation.
Figure 3. Typical laboratory scale stirred tank fermenter. (Courtesy of LH Fermentation,

Stake Pages, UK.)
4.2. Continuous Systems

During most of the lag phase and the stationary phase of a batch
culture no cells are produced. Continuous culture systems have been
devised so that after an initial adaptation period the Saccharomyces cells
can grow at high specific growth rates for long periods of time without a
stationary phase occurring.
A simple continuous culture system, a continuous stirred tank reac-
CULTURE SYSTEMS 263
tor (CSTR) as shown diagramatically in Fig. 4, would comprise a fermen-

tation vessels containing a constant volume (V) of culture medium. A
continuous supply of fresh culture medium would be added at a constant
flow rate (F) and fermented liquor would be removed from the fermenta-
tion vessel at the same flow rate. Cells would leave the vessel with the spent
medium. The dilution rate (D) of such a system is defined as
F
D =V (10)
and the mean liquid residence time = A= ~ (11)
The overall accumulation of cells in the fermentation vessel (per

unit time) is given as the difference between the rate of growth and the
rate of removal of the cells, i.e.,
dx
- = J.lX- Dx (12)
dt
After adapting to the conditions in the fermenter, the growth rate of the
yeast cells remains constant, and the system is said to be at a steady state
until there is a change in the environmental conditions. Since under
steady-state conditions the cell conce~tration remains constant, : =0
and therefore 1-1. = D. As the dilution rate is increased, the specific
growth rate of the yeast cells will increase until D reaches its critical value
De at which 1-1. = 1-"max· If the dilution rate is further increased above De
Medium Fermentation Spent

Reservoir Vessel Fermentation
Liquor
Figure 4. A simple continuous culture system.

the yeast cells will be removed from the fermenter at a faster rate than
they can be produced and hence all cells will eventually be washed out of
the vessel.
At steady state there is a relationship between the substrate con-
centration and the yeast cell concentration. Assuming the rate of sub-
strate uptake ~is proportional to the rate of growth of the cells, then
(13)
where Y, the yield coefficient, is the ratio of the dry weight of cells
produced to the weight of substrate consumed over a given time period.
The substrate concentration in the feed reservoir (SR) will be higher
than in the fermentation vessel (S) as the yeast cells consume the sub-
strate for growth. The rate of change of the substrate concentration in
the fermentation vessel is the difference between the input, the output,
and the rate of consumption:
dS
-=DSR-DS-
ldx
-- (14)
dt y dt
Substituting from equation (8)
(15)
and substituting for fJ. from equation (9)
dS = D (SR - S) - f..LmaxX S (16)

dt Y S + Km
and
dx s
dt = f..LmaxX S + Km - Dx (17)
The behavior of single-cell organisms in continuous culture is defined by

equations (16) and (17).
Under steady-state conditions there is no change in the number of
cells or of the limiting substrate concentration in the fermenter.
Hence
(18)
CULTURE SYSTEMS 265
Equations (16) and (17) will then become, respectively,
D (S - S) = JJ.max X S (19)
R Y S + Km
and
s (20)
D=JJ.maxs+K
m
where x is the steady-state yeast cell concentration and S is the steady-

state limiting substrate concentration in the fermenter. Equations (19)
and (20) can be simplified by substitution to
SR- s = yX (21)
hence
X= y (SR- S) (22)
and
S =K D (23)
m JJ.max- D
At steady state the production of cells per unit volume of culture

medium is Dx and from equations (22) and (23)
D
Dx = DY [SR - Km
JJ.max-
D] (24)
Cell production will increase with dilution rate up to a maximum value

Dmax which can be determined by differentiating equation (24) with
respect to D, i.e.,
d (Dx) =y [S +K (25)
dD R m
_ d(Dx) _ .
and D - Dmax when ---;ny- - 0, so.
at d (Dx) = 0 S + K = -;--_K_,m!!....:..JJ.-';~:?ax,___,:n (26)

---;ny- ' R m (JJ.max - Dmax) 2
or
1-Lmax
_
Dmax -
_ [ Km JJ.~ax
S + K
J 0 ·5
(27)
R m
and
K ] o.5
Dmax = IJ.max [ 1 - S +mK (28)
R m
If SR is large [ K
m
K
S+] o.5
R
will be small and therefore Dmax = 1-Lmax
and hence= De.
4.3. Practical Continuous Culture Systems

The simplest type of vessel for a continuous culture fermentation,
the CSTR, is almost identical in configuration to that described for batch
cultures (Fig. 3), differing only in that for a continuous system an inlet
port and outlet port for the flow of liquid media are needed. Improve-
ments on this basic system have been made for the continuous produc-
tion of food and fodder yeasts at industrial scale by changes in the
stirring and aeration system (Hospodka, 1966). Examples of these are
shown in Fig. 5.
Figure 5a shows the Vogelbusch aeration system. Air flows down the
hollow central stirring shaft and out into the culture medium through
small holes, 0.5-1.0 mm in diameter, in the stirring paddle. Figure 5b
shows the system of Scholler and Seidel, based on the principle of the
airlift pump. ·Fermentation medium is circulated through tubes ar-
ranged outside the fermentation vessel. The liquid and air enter the
vessel through ceramic aeration candles at the bottom of the tubes and
are either recycled through the tubes or removed from the fermenter
via a central overflow cylinder. Figure 5c shows the Phrix aeration sys-
tem, which consists of three stirrer paddles mounted on a central shaft.
Air enters under, arid is dispersed by, the bottom paddle, which is made
from perforated plate. The middle paddle is for breaking up any large
air bubbles and preventing further bubble coalescence, while the top
paddle defoams the culture medium by a suction action as it rotates.
Figure 5d shows the Lafrancois system, which comprises two inlet tubes,
one for air and a narrower one for liquid medium, which extend almost
to the bottom of the fermenter. The ends of these tubes are contained in
a cooling tube to aid circulation.
These systems were devised during or just after World War II, when
there was an increased demand for food and fodder yeast. Baker's yeast
has only been produced by continuous fermentation during the past 25
years. The first successful plant was located at the Distillers Co., Dover-
court, England, and consisted of six fermentation vessels, each with a
working volume of 40 m 3 , connected in series as shown in Fig. 6. The
first vessel was filled with liquid medium and inoculated from the
CULTURE SYSTEMS 267
Air Flow
t
Product Outlet
Inlet
(Liquid and Air)
a b
Liquid-+ + - Air
t
Air
c d
Figure 5. Mixing and aeration systems for continuous production of food and fodder
yeast. (a) Vogelbusch system, (b) Scholler and Seidel system, (c) Phrix system, (d) Lafran-
cois system.
growth vessel. The outflow from vessel 1 entered vessel 2. That from
vessel 2 entered vessel3, and so on in sequence. When the yeast culture
had grown in all six tanks, usually after approximately 27 hr, the opera-
tion became continuous. Liquid media was pumped into the first five
vessels, which were fully aerated. The 6th vessel was only gently aerated,
which allowed the yeast to mature before being separated and dried for
use. Continuous operation of this system varied from 50 to 80 hr, and
.NI
en
OD
:-l
a:
a:
Inoculation ~
....,
Culture ::c
1:>':1
Nutrient
~
Culture "'
8.
Medium
0
~
1:>':1
t:=
t:=
2 3 4 5 6
Yeast
Separator
Figure 6. Distiller's system for continuous baker's yeast production.
CULTURE SYSTEMS 269
the baker's yeast productivity was reported to be 33% higher than in

batch systems.
An alternative smaller continuous production plant for baker's yeast
proved to be successful in Russia. This, the Plevato system, comprised
only two connected fermentation vessels. The first, with a volume of 30
m 3 , was fully aerated and had a complete nutrient medium added con-
tinuously, while the second was only 10 m3 in volume, only mildly ae-
rated, and no nutrients were added. Eighty-five percent of the total
biomass formed was produced in the first vessel. Continuous operation
lasted up to 140 hr and gave a 30-35% increase inS. cerevisiae productiv-
ity over a batch system.
During the 1960s the brewing industry developed tower fermenta-
tion systems for the continuous production of beer. This type of fer-
menter is, in essence, a vertical cylindrical tower with a conical bottom;
one such, developed by the APV Company Ltd. (Royston, 1966), is
shown schematically in Fig. 7. A settling zone containing a yeast sepa-
rator is included above the tower section and in principle allows the yeast
to settle and return to the tower while clear beer is removed from the
fermenter. Culture medium (wort) and air enter at the base of the tower.
For this system to work it is essential that the yeast strain selected is
highly flocculent and will settle back into the tower section, otherwise it
C0 2 Outlet
Beer Outlet
Clarifying
Tube
. Inlet ·
"'-·-·-·-·~
Figure 7. APV tower fermenter.

would be washed out with the fermentation effluent. Initially only one
strain of S. cerevisiae was found to be suitable, but later, tests were devel-
oped for yeast flocculence under tower fermentation conditions and
several more suitable Saccharomyces strains were identified (Greenshields
and Smith, 1971). Wort with an initial sugar concentration of 5-8% has
successfully been converted to beer with an alcohol content of 2-4%,
using such a system (Prince and Barford, 1982).
Attempts to produceS. cerevisiae by either batch or continuous oper-
ation of a tower fermenter system resulted in low yields, but harvesting
of the yeast was facilitated by its flocculent nature (Greenshields and
Smith, 1971 ).
More recently, as interest in power alcohol production has in-
creased, the performance of the tower fermenter with high sugar con-
centrations has been studied. Quantitative conversion of 20% glucose to
ethanol by flocculent strains of S. cerevisiae and S. diastaticus has been
shown to be possible at low dilution rates (Prince and Barford, 1982).
With the increased interest in ethanol production and in order to
obtain higher ethanol productivities many innovative fermentation sys-
tems have been developed since the mid-1970s.
Cysewski and Wilke (1977,1978) designed a fermentation system
for the production of ethanol by S. cerevisiae which incorporated a cell
recycle system whereby cells washed out of the fermentation vessel are
separated from the effluent liquor by sedimentation or centrifugation
and returned to give increased cell densities and hence higher ethanol
productivities. Ethanol inhibition effects were eliminated by running the
fermenter under reduced pressure, allowing the alcohol to be boiled off
from the fermentation liquor at a sufficiently low-temperature (35°C) so
as not to affect the yeast.
More conventional continuous stirred tank reactors with external
cell settling devices have also been considered for the production of
ethanol (Ghose and Tyagi, 1979a; Fricker and Witts, 1981). Another
system, where the fermentation broth is agitated by recycled gas and
incorporating an external separator for the Saccharomyces cells, has been
operated by Bu'Lock and Comberbach (1981).
Ultrafiltration offers an alternative method for cell retention within
a fermenter. The rotofermenter as proposed by Margaritis and Wilke
( 1978) is a pressurized vessel with an internal rotating filter which re-
tained S. cerevisiae cells. A similar system with a static filter has been used
by Chern and Zall ( 1982) with S. fragi.
The economics of operating all fermentation systems for ethanol
production needs serious consideration; recycling of yeast cells which
settle under gravity appears to have a cost advantage over systems em-
ploying centrifuges or filters.
CULTURE SYSTEMS 271
5. MONITORING THE GROWTH OF SACCHAROMYCES
A number of techniques are available to monitor the growth of a

Saccharomyces culture. All require a well-mixed, representative sample of
the culture broth, and the most appropriate should be chosen for a
particular application.
Cell counting, the most direct method, involves the use of a count-
ing chamber (either a hemocytometer or a Petroff-Hauser slide). The
slide is etched with a grid of squares of known size and is recessed to a
specified depth, typically 0.1 mm. The number of cells per square is
determined by microscopic observation and the concentration of cells is
found by dividing by the square area times depth. The viability of the
cells can be determined by staining 1 part liquid culture with 9 parts
methylene blue solution (Lee et al., 1981 ). Viable cells will remain color-
less and nonviable cells will take up the blue dye. The technique is quick
but is inaccurate if cells are flocculated.
Similarly, colonies may be counted. A sample is spread over a nu-
trient agar and incubated before counting the number of colonies
formed, either by naked eye (from a plate) or microscopically (from an
agar-coated slide). Early detection of colonies can be aided by the use of
fluorescent dyes added to the culture. Cell viability may be estimated
from the microscopic examination by comparing the initial number of
cells present with the number of colonies formed. The technique is again
inaccurate if cells are flocculated, since only colony-forming units (not
necessarily single cells) are counted.
Automatic cell counting can be achieved by the use of a Coulter
counter, in which cells suspended in an electrolyte solution pass through
an orifice and the change in current is recorded. Size as well as number
of particles is recorded, enabling flocculated cells to be taken into
account.
The most common method of monitoring cell growth is by measure-
ment of cell mass. This can be achieved by filtering or centrifuging and
drying a sample of cell suspension and then weighing to give the dry
weight. Alternatively, by measuring the optical density, at 600 nm, of the
sample and comparing this against a cell-free sample, then referring to a
standard, an almost instant measure of cell mass can be obtained. The
turbidity of a Saccharomyces culture can also be used to give a measure of
the cell mass.
In some alcoholic fermentations, carried out at industrial scale, on-
line analysis for the evolution of carbon dioxide gas is used to check the
fermentation rates and hence give an indirect measure of the amount of
active cell mass present. NADH, upon UV radiation at 366 nm, emits
fluorescence at about 400 nm. Since the concentration of NADH is
directly related to biomass density, flouresence measurements provide

another possibility for determining cell mass in situ, using on-line probes.
6. CELL SEPARATION TECHNIQUES
Saccharomyces cells are generally separated from the fermented li-

quor at the end of a fermentation. Flocculent yeasts can be allowed to
settle under gravity, but this method is too slow for fine suspensions of
nonflocculent cells. Filtration or centrifugation is more commonly em-
ployed for separating yeast cells from the bulk liquor.
In the industrial processing of baker's and brewer's yeast the cells
are initially separated from the liquor by centrifugation. The yeast con-
centrate obtained is then washed, recentrifuged, and filtered either
through cloth in presses or in rotary vacuum filters to obtain pressed
yeast with a dry matter content of 28-30% wt./vol. (Burrows, 1970).
7. IMMOBILIZED CELL SYSTEMS
The immobilization of whole cells within a fermentation vessel is not

a new idea but has only been applied to the Saccharomyces for alcoholic
fermentations during the past 10 years. Immobilization techniques are
aimed at the aggregation of cells to form some sizable structure which
can be retained within a fermenter. There are several advantages in
using an immobilized cell system over a free cell system for the produc-
tion of ethanol.
• A higher fermenter yeast cell concentration can be achieved than
with free cell batch or continuous systems.
• With free cells the dilution rate is limited by the growth rate of the
cells, but as immobilized yeast cells are retained within the fer-
menter, dilution rates can be optimized for maximum ethanol
productivity.
• Cell recycling is unnecessary, and since fewer yeast cells leave the
fermenter with the product, ethanol extraction is easier than with
free cell fermentations.
• A further advantage of high cell densities and high dilution rates
is that the risk of contamination may be greatly reduced.
In a comparison of an immobilized cell reactor (ICR) and a CSTR
for ethanol production by S. cerevisiae (Sitton et al., 1980), it was found
that the immobilized cell reactor could be operated at dilution rates six
times higher than the washout rate of the CSTR and that the ethanol
CULTURE SYSTEMS 273
productivity of the ICR was better than that of the CSTR by a factor of
nine.
Methods available for the immobilization of Saccharomyces cells can
be categorized as either passive immobilization, which involves the natu-
ral growth of the yeasts onto or within a support, or active immobiliza-
tion, which requires some chemical process to immobilize pregrown
yeast cells.
7.1. Passive Immobilization

7.1.1. Films on Solid Supports
Saccharomyces cells have been immobilized as films on a wide range
of inert solid supports, including PVC flake (Ghose and Bandyopadhy-
ay, 1980), porous brackets (Ghose and Bandyopadhyay, 1980), Raschig
rings (Ghose and Bandyopadhyay, 1980; Sitton and Gaddy, 1980), gela-
tin-coated Raschig rings (Sitton and Gaddy, 1980), ceramic spheres
(Moo-Young et al., 1980), wood chips (Moo-Young et al., 1980; Gencer
and Mutharasan, 1981), spherical mullite (Fricker and Witts, 1981), and
brick (Minier and Gomas, 1982).
However, in some cases the Saccharomyces cells were sloughed off the
supports at superficial liquid velocities above a critical value and hence
were washed out of the fermenter.
7.1.2. Biomass Support Particles
Some Saccharomyces species are not normally flocculent or film form-
ing and hence cannot be immobilized passively onto solid supports. An
alternative immobilization technique, which can be applied to all Sac-
charomyces yeasts and which takes advantage of the sloughing off of over-
growth, is given by biomass support particles (BSPs) (Atkinson et al.,
1978, 1979).
BSPs are open, porous structures which are available in a selection
of sizes and shapes and which have contiguous internal voids accounting
for up to 97% of the volume of the particle. Yeast cells can grow nor-
mally and be retained within the structure and any external growth can
be sheared off.
BSPs constructed from knitted crushed stainless steel wire and flexi-
ble reticulated polyurethane foam have been used to immobilize strains
of S. cerevisiae and S. uvarum for ethanol production (Black et al., 1984).
A fermenter containing BSPs and fermentation medium can be inocu-
lated with a cell suspension, and after an initial batch growth period
during which the cells become immobilized, continuous operation can be
maintained for long periods.
7.2. Active Immobilization

7.2.1. Adsorption to a Charged Surface
This technique depends on the Saccharomyces cell wall surface con-
taining constituents providing the necessary ionic sites for attachment to
a charged support and generally requires more expensive supports such
as ion exchange resins.
7.2.2. Covalent Bonding

The attachment of Saccharomyces cells to a solid support by covalent
bonds is limited in that only a monolayer of yeast can be formed on the
support and the chemical reagents required to assist the formation of
covalent linkages are generally toxic and hence can cause a loss of
viability in the yeast cells.
7.2.3. Entrapment within Gels and Resins

This technique is the most widely used method of Saccharomyces cell
immobilization and all Saccharomyces species can be immobilized in this
way. A wide range of gels and resins are available, including calcium
alginate (Kierstan and Bucke, 1977; White and Portno, 1978; Williams
and Munnecke, 1981; Linko and Linko, 1981; Cho and Choi, 1981),
kappa carrageenan (Wada et al., 1979,1980,1981), agar (Holcberg and
Margalith, 1981), polyacrylamide (Holcberg and Margalith, 1981), gela-
tin (Sivaraman et al., 1982), and epoxy resins (Klein and Kressdorf,
1982).
The size and shape of the gels and resins may vary according to the
method of production. Calcium alginate is generally produced as spher-
ical beads whereas other gels may be extruded and cut to size or pro-
duced as a sheet and then cut to the required shape.
In general, a concentrated suspension of cells is mixed with one or
more of the constituents of the gel prior to gel formation. If a complete
nutrient feed is produced, the yeast cells can continue to grow after
immobilization, and this has been shown to produce higher yeast con-
centrations near the gel surface where there are less diffusion limitations
(Cho and Choi, 1981; Wada et al., 1979).
The strength of gels and resins can be varied by changing the ratio
of their constituents. However, calcium alginate can be weakened in the
presence of phosphate ions, and polyacrylamide and some epoxy resins
can have a toxic effect on Saccharomyces; so care is needed during the
immobilization procedure with these. The most common problem en-
countered with ,gel-immobilized yeast is the evolution of carbon dioxide
CULTURE SYSTEMS 275
gas during fermentation. The gel can be disrupted or caused to float if

the gas remains trapped.
7.3. Fermenters for Immobilized Saccharomyces Cells

CSTRs and tower fermenters are generally not suitable for immo-
bilized Saccharomyces cells. A range of fermenters have been developed
specifically for their required application and these include the follow-
mg.
7.3.1. Packed Beds

These vary in size and shape but all contain a retaining volume in
which the immobilized cells are packed. Fermentation medium is
pumped through this volume and may be recycled. In general, liquid
mixing is poor and the large amounts of C02 generated during fermen-
tation make these fermenters susceptible to gas flooding (the densely
packed particles restrict the movement of the gas phase through the bed,
resulting in a buildup of gas which displaces the liquid medium; Cho
and Choi, 1981); hence sampling and parameter control are difficult.
7.3.2. Sheet Reactors
These are similar to packed beds but contain parallel plates of gel or
support material over which the fermentation broth can be pumped.
Liquid mixing is improved and evolved gases can be removed more
easily than in a packed bed (Larsson and Mosbach, 1979).
7 .3.3. Fluidized Beds

Particles containing immobilized Saccharomyces with a density great-
er than the fermentation medium can be fluidized by the upflow of the
fermentation medium which is usually recycled. In order to achieve
adequate particle movement, the bed is generally expanded by around
20%. Gel beads may be weighted, for example, by the inclusion of mag-
netite (Larsson and Mosbach, 1979), in order that the density may be
increased sufficiently for the beads to be fluidized. A fluidized bed fer-
menter used with stainless steel BSPs is shown in Fig. 8.
7.3.4. Gas Mixed Reactors

Particles of immobilized biomass with a density similar to that of the
fermentation medium can be circulated by the introduction of air or
recycled gases at the base of the fermenter. Liquid recycling is not neces-
sary. Better mixing occurs when gas is introduced only over a segment of
0 0
----e-
0
0 0 Retaining
Plate
• 0 0
·~.0
I .-• a o, •
•• 0
• o o•
••• 0 0
•
•
•
0
0
t •o
• •o
~• o •o •
• • o• •
• 0 0
•~· 1U Offset Air

0 0
Distributor
Air--+--+
Circulating Bed Reactor
Figure 8. Fluidized bed reactor for use with dense particles of immobilized yeast cells.
the fermenter base and when a bed expansion of around 20% is permit-
ted (Black et al., 1984). A circulating bed fermenter used with foam BSPs
is shown in Fig. 9.
In a comparison of a packed bed and a fluidized bed reactor each
containing the same loading of S. cerevisiae immobilized in calcium algi-
nate gel beads (Cho and Choi, 1981), it was found that over a range of
dilution rates the ethanol productivity obtained from the fluidized bed
fermenter was double that of the packed bed fermenter. However, ob-
taining very high ethanol productivities is not always desirable if there is
a corresponding decrease in ethanol concentration, increasing the costs
of downstream processing.
8. DOWNSTREAM PROCESSING
The products of Saccharomyces fermentation are yeast and alcohol.

Processing methods for yeast and alcoholic beverages are well estab-
lished, although fermentations for fuel alcohol production are much
newer processes in competition with established production methods
CULTURE SYSTEMS 277
Liquid
Recycle
i ii
•••••••
Retaining
Plate
•••••• •••••• •
• •• • ••
•••••••
••••••
• •••••• •
•••••• ••••••
•• ••
••
•••••••
•••••• •••••• • Distributor
Fluidised Bed Reactor
Figure 9. Circulating bed reactor for use with immobilized yeast cells where the particles
have a similar density to the liquid medium.
(see Chapter 7). The downstream processing costs of such fermentations

will affect the economic viability of the process.
8.1. Pressed Yeast
Traditionally yeast is processed as pressed yeast. After separation of
the cells from the broth, as described previously, the yeast is removed
from the presses or filter and passes to a hopper. Water and an
emulisifer may be added to assist extrusion and packing, and the yeast is
either granulated or compressed into blocks and packed in waxed pa-
pers or plastic containers. For storage the yeast must be kept below 10°C
to prevent metabolization of storage carbohydrate (Burrows, 1970).
8.2. Dried Yeast

The commercial production of dried yeast has the advantage of a
cost reduction in storage and transport but is generally lower in activity
than pressed yeast. Pressed yeast is extruded through a screen to give

continuous threads which are chopped and dried in a rotary drum drier
or by tunnel drying. Dry air at between 35°C and 90°C is passed over the
yeast but the surface temperature is kept low by evaporative cooling; this
process takes between 6 and 24 hr. Loss of activity occurs at the end of
the drying period (usually between 6 and 24 hr) when the cells are no
longer protected by the evaporative cooling. Where high levels of cell
activity are to be retained, drying agents such as methyl cellulose can be
used to absorb final moisture.
Spray drying is sometimes used but requires higher temperatures
which cause biological damage to the yeast unless protective agents are
used.
8.3. Yeast Extract

The most common method of producing yeast extract is to plas-
molyze pressed yeast with salt, heat to around 65°C for 40 hr, and then
concentrate the extract. Other plasmolyzing agents can be used to re-
duce saltiness. An alternative method is by acid hydrolysis of yeast under
pressure and at high temperatures (Pyke, 1957).
8.4. Alcoholic Beverages

Fermented liquor usually contains some Saccharomyces cells, and
hence in order to prevent further fermentation, the liquid may be fil-
tered, heat-treated, or pasteurized or a chemical antiseptic may be add-
ed. For spirits various methods of distillation are available depending on
the requirements of the final products. Beers, ciders, and wines may be
clarified by racking, filtration, or centrifugation prior to bottling or keg-
ging. Carbon dioxide may be introduced to some alcoholic beverages at
the bottling stage.
8.5. Ethanol
Ethanol is normally required at concentrations about the 95%
wt./vol. which can be achieved by distillation. Very pure ethanol can be
obtained by extractive distillation and azeotropic columns can be in-
cluded in multiple-column stills for absolute alcohol (99.7%). However,
the ethanol content of fermented liquor from a Saccharomyces fermenta-
tion is rarely above 10% and hence the distillation step can be costly.
Research into energy-efficient alternative methods of alcohol recovery is
currently in progress, and it is clear that the development of such eco-
nomic downstream processing methods is essential if Saccharomyces fer-
mentations are to provide alternative fuels for the future.
CULTURE SYSTEMS 279
REFERENCES
Atkin, L., Schultz, S., and Frey, C. N. 1946. Yeast Fermentation in Enzymes and Their Role in
Wheat Technology, lnterscience, New York.
Atkinson, B., Black, G. M., Lewis, P. J. S., and Pinches, A. 1978. Growth of biological
material, British Patent No. 2006 181 B.
Atkinson, B., Black, G. M., Lewis, P. J. S., and Pinches, A. 1979. Biological particles of
given size, shape and density for use in biological reactors, Biotech. Bioeng. 21: 193-
200.
Barford, J. P., Jeffery, P. M., and Hall, R. J. 1980. The Crabtree effect in Saccharomyces
cerevisiae-Primary control method or transient? Adv. Biotechnol. 1:255-274.
Black, G. M., Webb, C., Matthews, T., and Atkinson, B. 1984. Practical reactor systems for
yeast cell immobilisation using biomass support particles, Biotech. Bioeng. 26:134-141.
Brown, S. W., Oliver, S. G., Harrison, D. E. F., and Righelato, R. C. 1981. Ethanol inhibi-
tion of yeast growth and fermentation differences in the magnitude and complexity of
the effect, Eur.J. Appl. Microbial. Biotechnol. 11:151-155.
Bu'Lock,]. D., and Comberbach, D. M. 1981. A practical system for high productivity
ethanol fermentations, Second European Congress on Biotechnology, Abstracts of Commu-
nications, Society of Chemical Industry, London, p. 204.
Burrows, S. 1970. Bakers' Yeasts in: The Yeasts, Vol. 3 (A. H. Rose and]. S. Harrison, eds.),
Academic Press, New York, p. 349.
Chen, H. C., and Zall, R. R. 1982. Continuous fermentations of whey into alcohol using an
attached film expanded bed reactor, Process Biochem. 17:20-25.
Chen, S. L., and Gutmanis, F. 1976. Carbon dioxide inhibition of yeast growth in biomass
production, Biotech. Bioeng. 18:1462.
Cho, G. H., and Choi, C. Y. 1981. Continuous ethanol production by immobilised yeast in a
fluidized reactor, Biotech. Lett. 3:667-671.
Cysewski, G. R., and Wilke, C. R. 1977. Rapid ethanol fermentations using vacuum and
cell recycle, Biotech. Bioeng. 19:ll25-1143.
Cysewski, G. R., and Wilke, C. R. 1978. Process design and economic studies of alternative
fermentation methods for the production of ethanol, Biotech. Bioeng. 20:1421-1444.
de Deken, R. H. 1966. The Crabtree effect: A regulatory system in yeast,]. Gen. Microbial.
44: 149-156.
Del Rosario, E.]., Lee, K. ]., and Rogers, P. L. 1979. Kinetics of alcohol fermentation at
high yeast levels, Biotech. Bioeng. 21:1477-1482.
Eroshin, V. K., Utkin, I. S., Ladynichev, S. V., Samoylov, V. V., Kushinnikov, V. D., and
Skryabin, G. K. 1976. Influence of pH and temperature on the substrate yield coeffi-
cient of yeast growth in a chemostat, Biotech. Bioeng. 18:289-295.
Fiechter, A., Fuhrmann, G. F., and Kappeli, C. 1981. Regulation of glucose metabolism in
growing yeast cells, Adv. Microbial Physiol. 22:123-183.
Fricker, R., and Witts, W. S. 1981. The alcon process for continuous production of ethanol
by fermentation, Second European Congress on Biotechnology, Abstracts of Communications,
Society of Chemical Industry, London, p. 275.
Gencer, M. A., and Mutharasan, R. 1981. Ethanol fermentation in a yeast immobilised
column fermenter, Adv. Biotechnol. 1:627-633.
Ghose, T. K., and Bandyopadhyay, K. K. 1980. Rapid ethanol fermentation in immobilised
yeast cell reactor, Biotech. Bioeng. 22:1489-1496.
Ghose, T. K., and Tyagi, R. D. 1979a. Rapid ethanol fermentation of cellulose hydrolysate.
I. Batch versus continuous systems, Biotech. Bioeng. 21:1387-1400.
Ghose, T. K., and Tyagi, R. D. 1979b. Rapid ethanol fermentation of cellulose hydrolysate.
II. Product and substrate inhibition and optimisation of fermenter design, Biotech.
Bioeng. 21:1401-1420.
Gordon, P. A., and Stewart, P. R. 1972. Effect of lipid status on cytoplasmic and my-
tochondrial protein synthesis in anaerobic cultures of Saccharomyces cerevisiae, ].
Gen. Mirobiol. 72:231.
Gray, W. D. 1941. Studies on the alcohol tolerance of yeasts,]. Bacterial. 42:561-574.
Greenshields, R. N., and Smith, E. L. 1971. Tower fermentation systems and their applica-
tions, Chem. Engineer 249:182-190.
Harding, S. A., and Kirsop, B. H. 1979. Relative significance of oxygen and other nutrients
as fermentation regulators in Saccharomyces cerevisiae,]. lnst. Brew. 85: 174.
Hoggan, J. 1977. Aspects of fermentation in conical vessels, ]. lnst. Brew. 83: 133-
138.
Holcberg, I. B., and Margalith, P. 1981. Alcoholic fermentation by immobilised yeast at
high sugar concentrations, Eur.J. Appl. Microbial. Biotechnol. 15:133-140.
Hospodka, J. 1966. Industrial Application of Continuous Fermentation, in: Theoretical and
Melhodological Basis tf Continuow Culture tf Microorganisms (I. Malek and Z. Fenel, eds.),
Academic Press, New York, p. 535.
jones, M., Pragnell, M. J., and Pierce, J. S. 1969. Adsorption of amino acids by yeasts from
· a semi-defined medium simulating wort,]. lnst. Brew. 75:520-536.
jones, R. P., Pamment, N ., and Greenfield, P. F. 1981. Alcohol fermentation by yeasts-
The effect of environmental and other variables, Process Biochem. 16:42-49.
Kierstan, M., and Bucke, C. 1977. The immobilisation of microbial cells, subcellular
organelles and enzymes in calcium alginate gels, Biotech. Bioeng. 19:387-397.
Kirsop, B. H., and Brown, M. L. 1972. Some effects of wort compositions on rate and
extent of fermentation by brewing yeasts,]. lnst. Brew. 78:51.
Klein, J., and Kressdorf, B. 1982. Immobilisation of living whole cells in an epoxy matrix,
Biotech. Lett. 4:375-380.
Krouwel, P. G., and Braber, L. 1979. Ethanol production by yeast at supraoptimal tem-
peratures, Biotech. Lett. 1:403.
Kunkee, R. E., and Ough, C. S. 1966. Multiplication and fermentation of Saccharomyces
cerevisiae under carbon dioxide pressure in wine, Appl. Microbial. 14:643.
Larsson, P. 0., and Mosbach, K. 1979. Alcohol production by magnetic immobilised yeast,
Biotech. Lett. 1:501-506.
Lee, S. S., Robinson, F. M., and Wang, H. Y. 1981. Rapid determination of yeast viability,
Biotech. Bioeng. Symp. 11:641-649.
Leibowitz, J., and Hestrin, S. 1939. The direct fermentation of maltose by yeast, En-
zymologia 6:15-16.
Lewis, M. J., and Wildenradt, H. L. 1969. Sulphur in brewing, Brewers Digest 44:88.
Linko, Y. Y., and Linko, P. 1981. Continuous ethanol production by immobilised yeast
reactor, Biotech. Lett. 3:21-26.
Margaritis, A., and Wilke, C. R. 1978. The rotorfermentor. II. Application to ethanol
fermentation, Biotech. Bioeng. 20:727-753.
Markham, E., and Byrne, W. J. 1968. Uptake storage and utilisation of phosphate by yeast.
3. Behaviour of phosphate starved yeast,]. lnst. Brew. 44:374.
Martiny, S. C. 1972. Analysis and Simulation tf Biochemical Systems, Vol. 25, Fed. Eur. Bio-
chem. Soc. Meet [Proc.], p. 387-397.
Menzinsky, G. 1943. Application of microorganisms to sugar analysis. I. Quick micro-
bilogical procedures for the determination of fermentable sugar in the sulfite waste
liquor, Biochem. Z. 314:312-326.
Minier, M., and Goma, G. 1982. Ethanol production by extractive fermentation, Biotechnol.
Bioeng. 24:1565-1579.
Moo-Young, M., Lamptey, J., and Robinson, C. W. 1980. Immobilisation of yeast cells on
various supports for ethanol production, Biotech. Lett. 2:541-548.
Nagodawithana, T. W., Castellano, C., and Steinkraus, K. H. 1974. The effect of dissolved
CULTURE SYSTEMS 281
oxygen, temperature, initial cell count and sugar concentration on the viability of S.
cerevisiae in rapid fermentations, Appl. Microbiol. 28: 383-391.
Nes, W. R., Sekula, B. C., Nes, W. D., and Adler, J. L. 1978. The functional importance of
structural features of ergosterol in yeasts,]. Biol. Chem. 255:6218-6225.
Power, D. M., and Challinor, S. W. 1969. Effects of Inositol deficiency on chemical com-
position of yeast cell wall, f. Gen. Microbiol. 55:169-176.
Prince, I. G., and Barford, J. P. 1982. Continuous tower fermentation for power ethanol
production, Biotech. Lett. 4:263-268.
Pyke, M. 1957. Industrial Production in: Yeasts (W. Roman, ed.), Dr. W. Junk Publishers,
Holland, pp. 69-78.
Ranganathan, B., and Bhat, J. V. 1958. Ethanol tolerance of some yeasts,]. Indian Jnst. Sci.
40:105-110.
Rhodes, A., and Fletcher, P. L. 1966. Principles if Industrial Microbiology, Pergamon Press,
Oxford.
Rickard, P. A. D., and Hogan, C. B. J. 1978. Effects of glucose on the activty and synthesis
of fermentative and respiratory pathways of Saccharomyces sp., Biotech. Bioeng.
20:1105-1110.
Rose, A. 1976. Chemical Microbiology, 3rd ed., Plenum Press, New York.
Rothstein, A. 1961. Interrelationships between the ion transporting systems of the yeast
cell in: Membrane Transport and Metabolism (A. Kleinzeller and A. Kotyk, eds.), Academ-
ic Press, Prague, pp. 270-284.
Royston, M.G. 1966. Tower fermentation of beer, Process Biochem. 1:215-221.
Schultz, A. S., and McManus, D. K. 1950. Amino acids and inorganic sulphur as sulphure
source for the growth of yeasts, Arch. Biochem. 25:401-409.
Schultz, A. S., and Pomper, S. 1948. Amino acids as a nitrogen source for growth of yeasts,
Arch. Biochem. 19:184-192.
Schultz, A. S., Atkin, L., and Frey, C. N. 1940. Influence of oxygen on the fermentation of
maltose and galactose,]. Am. Chem. Soc. 62:2271-2272.
Sitton, 0. C., and Gaddy, J. L. 1980. Ethanol production in an immobilised cell reactor,
Biotech. Bioeng. 22:1735-1748.
Sitton, 0. C., Manrider, G. C., Book, N. L., and Gaddy, J. L. 1980. Comparison of immo-
bilised cell reactor and CSTR for ethanol production, Biotech. Bioeng. Symp. 10:213-
235.
Sivaraman, H., Seetarama Rao, B., Pundle, A. U., and Sivaraman, C. 1982. Continuous
ethanol produciton by yeast cells immobilised in open pore gelatin matrix, Biotech. Lett.
4:359-364.
Smith, R. H. 1951. A study of the role of inositol in the nutrition of Nematospora gossypii and
Saccharomyces carlsbergensis, ]. Gen. Microbiol. 5:772-780.
Sobotka, H., Holzman, M., and Reiner, M. 1936. Selective fermentation. III. Fermentation
of hexose-pentose mixtures, Biochem. ]. 50:933-940.
Thompson, E. D., and Parks, L. W. 1974. Effects of altered sterol composition on growth
characteristics of Saccharomyces cerevisiae, f. Bacterial. 120:779-784.
Thorne, R. S. W. 1944. Growth of yeasts in binary mixtures of nitrogen nutrients,). Jnst.
Brew. 50: 186.
Travassos, L. R., and Cury, A. 1971. Thermophilic enteric yeasts, Annu. Rev. Microbiol.
25:49-74.
Wada, M., Kato, J., and Chibata, I. 1979. A new immobilisation of microbial cells. Immo-
bilised growing cells using carrageenan gel and their properties, Eur. f. Appl. Microbiol.
Biotechnol. 8:241-247.
Wada, M., Kato, J., and Chibata, I. 1980. Continuous production of ethanol using immo-
bilized growing yeast cells, Eur. ]. Appl. Microbiol. Biotechnol. 10:275-287.
Wada, M., Kato, J., and Chibata, I. 1981. Continuous production of ethanol in high
concentration using immobilised growing yeast cells, Eur.J. Appl. Microbial. Biotechnol.
11:67-71.
Walsh, R. M., and Martin, P. A. 1977. Growth of Saccharomyces cerevisiae and Saccharomyces
uvarum in a temperature gradient incubator,]. Inst. Brew. 83:169-172.
Walter, F. G. 1940. The Manufacture tf Compressed Yeast, Chapman and Hall, London.
White, F. H., and Portno, A. D. 1978. Continuous fermentation by immobilised brewers
yeast,]. Inst. Brew. 84:228.
White, J., and Munns, D. J. 1950. Nutrilites and the production of pressed yeast,]. lnst.
Brew. 56:194-202.
White, J., and Munns, D. J. 1951. Influence of temperature on yeast growth and fermenta-
tion,]. Inst. Brew. 57:280-284.
Williams, D., and Munnecke, D. M. 1981. The production of ethanol by immobilized yeast
cells, Biotech. Bioeng. 23:1813-1825.
Williams, R. J., Eakin, R. E., and Snell, E. E. 1940. Relationship of inositol thiamin biotin
pantothenic acid and vitamin B6 to growth of yeasts,]. Am. Chem. Soc. 62: 1204-1207.
Biochemical Techniques
9
MICHAEL F. TUITE and STEPHEN G. OLIVER
I. INTRODUCTION
A plethora of biochemical techniques have been developed to study

fundamental cellular processes of yeasts of the Saccharomyces genera. All
of these generally require efficient disruption of the yeast cell followed
either by purification of particular classes of macromolecules (e.g., pro-
teins, nucleic acids, ribonucleoprotein complexes) or by partial frac-
tionation of the lysate to yield subfractions that are particularly enriched
in some biosynthetic activity, be it a given enzyme or a multicomponent
process such as protein synthesis. In this chapter we describe the basic
strategies that have been developed for Saccharomyces species (especially
S. cerevisiae) for both cell disruption and fractionation, together with
consideration of how one radiolabels specific macromolecules in vivo. A
detailed account of strategies for subcellular fractionation of yeast cell
components-mitochondria, membranes, vacuoles, and so forth-can
be found in Chapter 2.
2. CELL DISRUPTION
The cell wall of S. cerevisiae presents a daunting barrier to the bio-

chemist and molecular biologist alike. It comprises a thick, complex
mixture of proteins and polysaccharides (Ballou, 1982). The outer sur-
face is a tight aggregation of mannoproteins lying among the long, thin
polysaccharides glucan and chitin and is largely impenetrable to nucleic
acids, antibiotics, and other macromolecules. It is not readily susceptible
to disruption by classical methods such as sonication and freeze-thaw-
ing, and therefore other strategies have been devised that allow for the
MICHAEL F. TUITE • Biological Laboratory, University of Kent, Canterbury, Kent

CT2 7NJ, England. STEPHEN G. OLIVER • Manchester Biotechnology Centre,
University of Manchester, Institute for Science and Technology, Manchester M60 IQD,
England
283
effective breakage of yeast cells. These strategies can be subdivided into

those employing whole cells and those employing protoplasts.
2.1. Whole Cell Disruption

Traditionally yeast enzymes such as invertase have been extracted
from whole cells by prolonged periods of autolysis, a process of "self·
digestion" that relies on endogenous enzyme activities, particularly pro-
teases, to disrupt the cell and to solubilize at least some of the yeast
protein (Peppler, 1979). In the laboratory the use of autolysis is best
avoided-particularly if one is setting out to purify a single polypeptide
or trying to generate an active cell-free lysate, since most cellular en-
zymes are inactivated by this technique due to proteolysis or thermal
inactivation.
A number of alternative, but no less effective, mechanical methods
for disrupting yeast cells and releasing their proteins intact have been
devised. The French Press has been frequently and widely used, as has
the Eaton Press (for details, see Hughes et al., 1971). More recently,
methods that involve adding small mesh glass beads to a yeast cell slurry
followed by vigorous stirring or shaking have come into favor. This
approach was originally devised for large-scale yeast cell disruption in
specially designed pieces of apparatus such as the Braun homogenizer
(e.g., Schatz, 1967) and the Nossal shaker (e.g., Gregory et al., 1967).
More recently the method has been adapted for small-scale disruption
using a vortex mixer (Mills, 1974) or by carefully choreographed hand
shaking (Land et al., 1977) to agitate the glass bead/yeast slurry. The size
of the glass beads can be critical for achieving efficient breakage, and
usually acid-washed glass beads with a diameter of 0.4-0.45 mm have
proven to be most effective. Glass bead disruption can be used in prepar-
ing DNA, RNA, or protein and, in addition, has proven capable of
generating active cell-free translation systems (see Section 6.2).
In all cell disruption techniques, steps must be taken to minimize the
likelihood of proteolysis, and this generally entails carrying out the pro-
cedures at low temperatures (+2°C to +4°C) and in the presence of
nonspecific protease inhibitors such as phenylmethyl sulfonyl fluoride
(PMSF) and leupeptin (Pringle, 1975). The chances of avoiding pro-
teolysis in an extract can be further increased by employing a strain of
yeast deficient in one or more of the now well-characterized proteinases
(Achstetter and Wolf, 1985). In particular, strains carrying the pep4, prbl,
and prcl mutations are already widely in use (e.g., Rothblatt and Meyer,
1986).
2.2. Protoplast Lysis

A protopl~st can be simply defined as an osmotically sensitive spher-
ical body formed from a fungal cell without regard to completeness of
BIOCHEMICAL TECHNIQUES 285
removal of all cell wall material (Brunner et al., 1958). This term is
frequently used interchangeably with the term spheroplast although the
latter should only be applied to protoplasts completely devoid of cell wall
material.
A variety of lytic enzyme preparations are effective in producing
spheroplasts from S. cerevisiae provided an osmotic stabilizer such as
sorbitol, mannitol, or MgS04 is present to protect the osmotically fragile
spheroplast. Only recently have many of these enzymes become com-
mercially available (Table 1). The majority of these are also effective
against other Saccharomyces species and other yeast genera, e.g., Candida.
Many of the favored lytic enzymes originate from microbial sources,
with the exception of the widely used preparations from the gut of the
snail Helix pomatia (Eddy and Williamson, 1957). These enzyme prepara-
tions contain a large number of, often undefined, enzymatic activities,
although by and large they contain a (3-1 ,3-glucanase activity necessary
to degrade the rigid glucan layer in the cell wall, together with proteases,
chitinases, and deoxyribonucleases (DNases). It has been suggested that
the protease activity may be a necessary component of the lytic activity of
these enzyme preparations (Villanueva et al., 1973; Scott and Schekman,
1980).
A. number of factors affect the efficiency of protoplast formation,
some of which are controllable. Perhaps the most frustrating problem is
the resistance of a number of laboratory strains to most protoplasting
enzymes even though closely related strains are not. While this impli-
cates some genetic component in susceptibility to protoplasting, the
physiological state of the cell can also profoundly affect susceptibility to
these enzymes. Particularly critical is the growth phase from which one
harvests the cells; exponentially growing cells are highly susceptible, but
as cells enter stationary phase they become relatively resistant (Eddy and
Table I. Commercially Available Enzymes Used for Preparing

Protoplast& of S. cerevisiae
Enzyme Source Supplier
Novozym 234 Trichoderma spp. Novo Enzyme Products Ltd.

Lyticase Arthrobacter lutew Sigma
Zymolyase 20Ta Arthrobacter luteu.s ICN Biomedicals Ltd.
Zymolyase lOOP
Glusulase Helix pomatia New England Nuclear (E.I. du
Pont de Nemours and Co.)
Protoplast-forming enzyme Helix pomatia Boehringer Mannheim
Helicase Helix pomatia L'lndustrie Biologique
Francaise
4 20T = 20,000 units/g; lOOT = 100,000 units/g.
Williamson, 1957), probably as a consequence of changes in cell wall

structure. Some enzyme preparations do, however, show an ability to
attack the cell walls of stationary phase cells-zymolyase and lyticase are
particularly effective in this respect (Kitamura and Yamamoto, 1972).
Yeast cells can be rendered more susceptible to protoplasting en-
zymes by the addition of a mercapto compound such as dithiothreitol, ~
mercaptoethanol, or thioglycollate (Sommer and Lewis, 1971), which
presumably destroys disulfide bonds in cell wall mannoproteins making
the glucan layer more accessible to the ~-1 ,3-glucanase activity. Such
treatment is particularly important for preparing protoplasts from sta-
tionary phase cells (Kuo and Yamamoto, 1975) and for preparing pro-
toplasts in the cold.
Prior to lysis of the protoplasts, it is generally necessary to remove
all traces of the lytic enzyme since many of the enzyme preparations are
contaminated with nucleases and proteases. This can be achieved by
multiple washes with an isotonic buffer, based on either sorbitol or
mannitol.
Effective disruption of protoplasts can be achieved by mechanical
means, e.g., Dounce homogenizer, glass beads, cycles of freeze-thaw, or
by lysis with a nonisotonic buffer containing a detergent such as SDS or
Briij. The efficacy of the latter procedures depends on the degree of
protoplasting achieved, since nonprotoplasted cells remain largely intact
in the presence of low concentrations of detergents.
The strategies employed in the purification of specific enzymes
and/or structural proteins from the yeast lysate are generally those ap-
plied for other cells lysates. The reader is referred to a number of
excellent texts on the general subject of protein purification, e.g., Scopes
(1982).
3. RADIOACTIVE LABELING OF MACROMOLECULES
The ability to radioactively label macromolecules to a high specific

activity is a key biochemical tool not only for studying the rate of syn-
thesis of a particular macromolecular species, but also as an accurate
qualitative and quantitative estimate of different species within a particu-
lar group of macromolecules. The addition of radiolabeled precursors
to growing cells, for brief periods (i.e., pulse labeling) or for longer
periods (often many generations), is the most effective means of achiev-
ing this for studies on RNA, DNA, and protein synthesis. Such studies by
and large use whole cells, but spheroplasts, incubated in an osmotically
stabilized medium, can incorporate radiolabeled precursors into RNA,
DNA, and protein for several hours, often at a rate similar to that ob-
served for whole cells in an identical osmotically stabilized medium
(Hutchison and Hartwell, 1967).
3.1. RNA
RNA, which represents greater than 98% of the total nucleic acids
inS. cerevisiae, can be labeled in vivo using either [3H)- or [1 4 C]-purines
or pyrimidines. Alternatively, [32P]orthophosphate can be used (Rubin,
1975), but in this case DNA will also be labeled (see Section 3.2). For
routine analysis of RNA synthesis, pulse labeling with 2-5 IJ.Ci/ml [3H]-
adenine or [3H]-uracil has proven most effective and labeling can be
achieved in a rich medium such as YEPD (Yeast Extract, Peptone, Dex-
trose) (Sogin et al., 1972}, although much higher specific activity labeling
can be achieved in a minimal medium using a mutant defective in ade-
nine or uracil biosynthesis, respectively. As in labeling all mac-
romolecules, there are serious problems in labeling RNA with precur-
sors because of the time needed for equilibration of internal and
external precursor pool sizes. A further complication is that the rate of
uptake of precursors can be drastically altered by changes in cell phys-
iology brought about, for example, by amino acid starvation.
A number of yeast RNA species are methylated, e.g., ribosomal
RNAs. and mRNAs at their 5' methylated cap (m7G}, and these can be
labeled in vivo with [methyPH]-methionine provided the labeling medi-
um is supplemented with sufficient guanine and adenine (approximately
2 mM each) to prevent labeling of the purine ring via the one-carbon
pool (Retel et al., 1969; Sriptai et al., 1976).
To label RNA to a high specific activity with [32 P]orthophosphate
requires a low-phosphate medium. Such a medium can be prepared
from a rich medium (such as YEPD) by selective precipitation of in-
organic phosphate using MgS0 4 (Rubin, 1975), and RNA with a specific
activity in excess of 106 cpm/ JA.g can be obtained using [32 P]orthoph-
osphate added at 0.1-0.5 mCi/ml of growing cells.
3.2. DNA
To specifically label DNA in yeast using precursors is very difficult
because, unlike Escherichia coli, yeast cannot efficiently incorporate either
thymine or thymidine into DNA. The reason for this is that S. cerevisiae,
like many fungi, does not possess the enzyme responsible for phos-
phorylating thymidine monophosphate (TMP}, i.e., thymidine kinase
(Grivell and Jackson, 1968}, thus preventing S. cerevisiae from utilizing
either thymine or thymidine as a precursor for DNA synthesis.
While certain strains can incorporate small amounts of exogenous
dTMP into DNA, the efficiency of incorporation is very low. To achieve
efficient incorporation requires the use of mutants (called tup mutants)
that can efficiently incorporate dTMP into DNA, and these can be read-
ily selected for in most strains after mutagenesis and using inhibitors of
endogenous folate synthesis such as aminopterin and p-aminobenzoic
acid (Wickner, 1975). While the use of tup mutants permits the routine
labeling of DNA in physiological studies, there are still a number of
problems: for example, mitochondrial DNA seems to be preferentially
labeled (Cryer et al., 1973), although this problem can, of course, be
eliminated by use of a mitochrondria-less (petite) mutant.
3.3. Proteins
To label yeast proteins to a high specific activity in vivo requires that
the radioactive precursor (i.e., an amino acid) is rapidly incorporated
both into new polypeptide chains and into the majority of protein spe-
cies. The choice of amino acid as the precursor is critical in this respect
since the intracellular pools of amino acids can differ in molar con-
centrations by as much as 250-fold; e.g., in nitrogen-rich medium the
intracellular concentration of serine was reported to be about 5000
j.Lmoles/ 10 g cells, whereas cysteine was only present at around 20
j.Lmoles/ 10 g cells (Chan and Cossins, 1976). Clearly, the larger the intra-
cellular pool, the longer it takes to reach isotopic equilibrium; for exam-
ple, for the relatively abundant amino acid lysine it takes 70-80 min to
reach isotopic equilibrium, whereas methionine, present at relatively low
intracellular concentrations, takes less than 30 sec (Gross and Pogo,
1974). Care must also be taken in the choice of medium since intra-
cellular amino acid levels can vary considerably depending on the media
composition (Chan and Cossins, 1976). An appropriate labeling medium
is a defined medium, based on yeast nitrogen base and buffered to pH
5.2 with a sodium succinate-sodium hydroxide buffer. To increase the
efficiency of labeling, the cold amino acid precursor can be reduced or
eliminated from such a medium, or an appropriate mutant unable to
synthesize de novo the supplied radiolabeled amino acid can be used.
For routine monitoring of the rate of protein synthesis in vivo, pulse
labeling with a [3H]-amino acid is appropriate using, for example, [3 H)-
leucine or [3 H]-phenylalanine, both of which have low endogenous pool
sizes. To rapidly terminate the incorporation of the radioactive precur-
sor, the simultaneous addition of the potent protein synthesis inhibitor
cycloheximide (to 50 I-Lg/m1) and crushed ice to the culture has proven
very effective in the authors' hands. To label proteins for one-dimension-
al and two-dimensional polyacrylamide gel analysis, methionine is widely
used because of its availability as [35S]-methionine. This fact, coupled
with the rapidity of isotopic equilibration for methionine, means that a
10-min pulse of a 25-m1 culture with 15-25 j.t.Ci of [35S]-methionine
(specific activity > 1000 Ci/mmole) can generate sufficient sample for
several gel analyses. Addition of histidine or tyrosine to the labeling
medium can significantly enhance methonine incorporation, although
the reason for this is unclear (Miller et al., 1979).
The danger in using methionine in this context is that methionine is

used comparatively rarely in yeast proteins and theN-terminal meth-
ionine is generally cleaved by an amino peptidase. An alternative option
is to use a high-specific-activity [14C)-amino acid mixture to generate a
polypeptide electrophoretic pattern that is more quantitative and a truer
reflection of the total number and relative abundance of proteins syn-
thesized by yeast under various physiological conditions. An elec-
trophoretic comparison of total yeast proteins labeled under identical
conditions with either [1 4 C]-amino acid mixture or with [35S]-meth-
ionine is shown in Fig. 1, and a number of differences can be noted.
If one wants to specifically label only mitochondrial proteins, i.e.,
proteins synthesized on mitochondrial ribosomes, this can be achieved
by pulse-labeling cells for 1 hr in the presence of an antibiotic such as
cycloheximide that specifically inhibits protein synthesis on cytoplasmic
ribosomes, but not mitochondrial ribosomes (Douglas et al., 1979). Label-
ing can be achieved with [35S]-methionine, although a higher specific
activity labeling with 35S04 2- requires that the strain is not a meth-
ionine auxotroph and Mg2 + and NH4 + components of the growth me-
dium are in the form of chloride salts.
...
...
• •-·.~ : •.
. ..
•
.
.•··--
Figure 1. Two-dimensional gel electrophoresis of yeast proteins synthesized in vivo. (a)

Total soluble proteins labeled with [S5S]·methionine. (b) Total soluble proteins labeled with
[14C)-amino acid mixture. The first dimension was by nonequilibrium pH gradient gel
electrophoresis (NEPHGE), the second dimension by SDS-polyacrylamide gel electropho-
resis (SDS-PAGE). Shown are autoradiographs exposed for 2 days (a) and 14 days (b). The
arrows in b indicate proteins not detectable by [35S]-methionine labeling. (Photographs
courtesy of Dr. Ian Fitch, University of Kent.)
4. RNA PREPARATION
4.1. Differential Extraction Techniques

It is possible to select for a particular class of RNA molecule by use
of the different extraction techniques described in this chapter. How-
ever, it is relatively simple to separate the different classes of yeast RNA
from a total cell extract by using CF11 cellulose chromatography (Frank-
lin, 1966). The CF11 cellulose (Whatman) column should be equili-
brated with STE buffer (0.1 M NaC1, 1 mM EDTA, 0.05 M Tris, pH
6.85) containing 1% ~-mercaptoethanol and then washed with STE con-
taining 35% (vol./vol.) ethanol. The RNA sample is resuspended in 0.1
column volume of STE plus 35% ethanol and loaded onto the column.
The different RNA species may then be eluted in a stepwise manner: (1)
STE plus 35% ethanol-tRNA; (2) STE plus 15% ethanol-rRNA; (3)
STE without ethanol-dsRNA. The progress of the stepwise process can
be monitored by recording the OD 260 of the eluate.
4.2. Messenger RNA

Messenger RNA (mRNA) represents between 1 and 5% of total cell
RNA inS. cerevisiae and comprises a complex heterogeneous collection
of single-stranded RN As with sizes generally in the range 500-5000
nucleotides. Like most eukaryotic mRNAs, S. cerevisiae mRNAs show a
certain degree of posttranscriptional modification; a methylated 5' ter-
minus (or cap; Sripati et al., 1976) and a polyadenylate [poly(A)] exten-
sion of approximately 50 nucleotides at the 3' terminus (McLaughlin et
al., 1973) are the most obvious. Few yeast cellular mRNAs contain in-
trons, which is in contrast to higher eukaryotic mRNAs.
In studies with S. cerevisiae the isolation of mRNA is usually required
for one of four reasons: (1) for in vitro translation studies (see Section
6.2); (2) for Northern hybridization studies of specific gene transcripts
(3) for mapping the 5' and 3' ends of specific gene transcripts; and (4)
for the generation of eDNA libraries. In each case the mRNA isolated
needs to be intact, but not necessarily free of other RNA species. For
example, translation of yeast mRNAs in a homologous cell-free lysate is
much more efficient in the presence of excess rRNA or tRNA (Gasior et
al., 1979a; Tuite and Plesset, 1986), and the sensitivity of Northern
hydridization methods allows for the detection of even minor mRNAs in
total cellular RNA.
mRNAs are highly susceptible to degradation by endogenous
ribonucleases, and inS. cerevisiae they have half-lives ranging from 3.5 to
70 min, with an average of 22 min (Chia and McLaughlin, 1979; Koch
and Friesen, 1979). Isolation of these RN As therefore presents the ex-
perimenter with a range of problems, not least being the inactivation of

ribonucleases. Generally, ribonuclease degradation can be avoided by
"standard" precautions: use of baked glassware, use of diethylpyrocar-
bonate-treated buffers, and inclusion of ribonuclease inhibitors. In addi-
tion, yeast strains deficient in one or more ribonuclease activities are
available; e.g., the RNase3 mutant (Littlewood et at., 1971) is deficient
in a potassium-activated ribonuclease found associated with 40S ribo-
somal subunits in wild-type strains (Baan et at., 1981).
A range of extraction protocols can be used for the isolation of
intact mRNA. These generally include extraction of either glass bead-
disrupted whole cells or spheroplasts (e.g., Sripati and Warner, 1978)
with organic solvents such as phenol or phenol:chloroform:isoamyl alco-
hol mixtures. Total polysomal RNA, extracted from pelleted polysomes,
has also proven to be an effective source of translatable mRNA for in
vitro studies (Gasior et at., 1979a). More recently, a chaotropic disruption
method, based on the ability of high concentrations of guanidinium
thiocyanate and other guanidinium salts to dissociate ribonucleoprotein
complexes and denature proteins, has been employed in conjunction
with glass bead disruption to obtain intact mRNA (Feinberg and
McLaughlin, 1988). This method has proven of value in obtaining intact
mRNAs from mammalian tissues with high ribonuclease content (Chirg-
win et at., 1979) and may be the method of choice for eDNA work, and
for 5' and 3' end mapping studies.
Should pure mRNA be required, it can be isolated from total RNA
preparations using the standard affinity chromatography method that
exploits the ability of the 3' poly(A) tail of mRNA to bind to oligo (dT)
cellulose or poly(U) sepharose at high salt concentrations (Holland et at.,
1977; Sripati and Warner, 1978). However, because the length of the
poly(A) tail in S. cerevisae mRNAs is somewhat shorter than found in
higher eukaryotic mRNAs (50-as opposed to 200-adenylic acid residues),
the "flowthrough" RNA has to be recycled several times through the
affinity column to ensure maximal yields. The ability to isolate mRNA
from S. cerevisiae in this way becomes increasingly difficult as batch-
grown cells enter stationary phase since this is accompanied by a short-
ening of the poly(A) tail (Sogin and Saunders, 1980).
4.3. Transfer RNA

The low molecular weight of transfer RNA (tRNA) species greatly
simplifies their purification from yeast. Phenol extraction of whole yeast
cells is sufficient to release tRNA (Monier et at., 1960) in a relatively pure
form. Figure 2 shows absorbance scans of both 2.6% and 12% poly-
acrylamide gels on which nucleic extracts from yeast have been run and
demonstrates that tRNA is the only low-molecular-weight species ex-
a
i... <C(
b
• •
VI z
~ 'C
....
VI
OCI
<C(~
zO::
c~
tt
A.
A. B.
B. -
Figure 2. Electrophoretic analysis of different yeast RNA preparations. Shown are absor-
bance scans (A2ao nm) of ethidium bromide stained gels; (a) RNA separated on a 12%
polyacrylamide gel; (b) RNA separated on a 2.6% polyacrylamide gel. Sample A was
prepared by phenol extraction of a total cell lysate; sample B was prepared by phenol
extraction of whole cells. (Adapted from Clare, 1982.)
tracted from whole cells and that subsequent breakage of the cells re-
leases no more tRNA. The whole cell extraction procedure may there-
fore be used for quantitative recovery of tRNA from yeast. The only
RNA species that contaminates such preparations is double-stranded
RNA (dsRNA; see Section 4.5) and this accounts for less than 2.5% of
the total nucleic acid content of the extract. Such small amounts of
dsRNA do not appear to interfere with the use of tRNA in in vitro
translation systems or the determination of tRNA charging levels.
Determination of the proportion of molecules of any tRNA species
that are charged with their cognate amino acid may be performed on
extracts prepared in the manner described earlier. The principle of the
charging assay is outlined in Fig. 3. The tRNA extract, containing both
charged and uncharged species, is divided into two fractions. One frac-
tion is chemically deacylated by mild alkaline hydrolysis (Yang and Nov-
elli, 1971) while the other is treated with periodate. The oxidizing agent
reacts with the ribosyl hydroxyl groups of the terminal adenosine resi-
dues of uncharged tRNA such that they can no longer esterify amino
acids (Schmidt, 1968). The oxidized preparation is then subjected to
mild alkaline hydrolysis to remove amino acids from the charged spe-
cies. Both fractions are then used in separate charging reactions. Incor-
poration into the first fraction gives a measure of the total amount of
tRNA (charged plus uncharged) of a given species present in the extract.
Incorporation into the second fraction measures the charged species
alone. The ratio of counts incorporated thus provides the proportion of
charged tRN A molecules in the extract.
IOij
~ Periodate oxidation
0.3M Tris HCI

pH 8.0
~Alkaline hydrolysis
tRNAO + e tRNAO + e
L-------- •·- ~ ...

1 ...... Ch.,glng
•
tRNAe tRNAe
Total of charged plus uncharged Charged only
Figure 3. Protocol used for determination of in vivo tRN A charging levels. tRN A•, charged
tRNA; tRNA0 , uncharged tRNA; 0, amino acid; e* radiolabeled amino acid.
Charging reactions (Hampel et al., 1979) may be conveniently per-

formed in drops of the appropriate cocktail on the surface of a stretched
sheet of Saran Wrap ("Cling Film"). The aminoacyl-tRN A synthetase
preparation which catalyzes the reaction may be prepared by a method
involving differential centrifugation, the precipitation of nucleic acids

with plyethylenediamine (Polymin P), and two ammonium sulphate pre-
cipitations followed by ion exchange chromatography and molecular
sieve chromatography (Hartwell and McLaughlin, 1968; VonderHaar,
1979; Faulhammer and Cramer, 1977). However, a rapid purification
method has been devised (Clare, 1982; Clare and Oliver, 1982) involving
two differential centrifugation steps which on their own are sufficient to
remove endogenous tRNAs. The final dialysis step is important since, in
the absence of ammonium sulfate precipitation, it is the only step that
frees the synthetase preparation from endogenous amino acids. Syn-
thetase preparations may be stored in 50% glycerol at - 80°C with little
or no loss of activity.
4.4. Ribosomal RNA

Yeast, as a fast-growing microorganism, is an abundant source of
ribosomal RNA (rRNA), which may readily be isolated from broken cells
using standard phenol extraction methods (e.g., Kirby, 1965). While the
large 35S rRNA precursor molecule can be isolated from disrupted cells
(Van den Bos and Planta, 1971 ), it is probably better prepared from
lysed spheroplasts (Udem and Warner, 1972). Of the two small mole-
cules of the four mature species, 5.8S rRNA is tightly hydrogen-bonded
to the 5' end of the 25S species. It must therefore be released by either
heat or formamide denaturation before it can be identified. A conve-
nient method of releasing the 5.8S species is simply to freeze and thaw
the total RNA extract two or three times. Rubin (1975) has demon-
strated that both small mature rRNA species, 5.8S and 5S, may be pre-
pared, in quantitative amounts, from intact yeast cells by treatment with
20 mM EDTA, 0.1 M Na acetate, and 1% SDS (pH 5) at 33°C. The use of
phenol extraction to release low-molecular-weight RNAs from intact
yeast cells was dealt with in Section 4.3.
4.5. Double-Stranded RNA

Almost all laboratory strains and many wild-type (nonlaboratory)
strains of S. cerevisiae contain one or more species of dsRNA encapsi-
dated within viruslike particles designated ScV particles (Mitchell and
Bevan, 1987, a review). The major species of dsRNA, designated L, is
approximately 4. 7 kb in length, while a second group of smaller
dsRNAs, designated M, are associated with the killer phenotype origi-
nally described by Bevan and co-workers. M dsRNAs vary in size be-
tween 1.9 and 2.7 kb.
Two general strategies are available for purifying dsRNAs from S.
cerevisiae. In the first, ScV particles are partially purified on a CsCl
gradient followed by final purification by sucrose gradient centrifuga-

tion (Oliver et al., 1977). The dsRNA is subsequently extracted from the
purified particles using simple phenol extraction methods (Kandel,
1979). In the second strategy, the dsRN A species are differentially pre-
cipitated from total RNA samples using LiCl (Diaz-Ruiz and Kaper,
1978). The latter method involves first precipitating rRNA and mRNA
with 2 M LiCl and then precipitating the dsRNA from the 2 M LiCl
supernatant with 4 M LiCl, leaving small single-stranded RNAs (particu-
larly tRNA) and DNA in the supernatant (Fig. 4). This strategy works for
both Land M dsRNAs and is relatively quick and simple. Should further
purification of the dsRNA be necessary, it can be achieved using CF-11
chromatography (Franklin, 1966) or by preparative electrophoresis.
One useful property of dsRN A is that it is very resistant to
ribonucleases; for example, it is insensitive to ribonuclease A at high, but
not low, ionic strength (Wickner and Leibowitz, 1976). This property can
also serve as a simple confirmation that the isolated material is indeed
dsRNA.
5. DNA PREPARATION
5.1. Chromosomal DNA

An ability to isolate intact, high-molecular-weight chromosomal
DNA is an obvious prerequisite if one is to study the yeast genome by
classical or recombinant DNA techniques. Both lengthy large-scale and
more rapid miniscale preparative techniques have been developed for S.
cerevisiae that generate chromosomal DNA of varying yields and degrees
of purity, but which is suitable for a range of analytical purposes, includ-
ing restriction enzyme digestion. To avoid contamination with mitochon-
drial DNA, [rho 0 ] petite mutants that totally lack that species may be
employed.
DNA-
L
Figure 4. Fractionation of yeast RNAs using differ-
ential precipitation with lithium chloride. Lane l, 4 M
M LiCI supernatant; Lane 2, total RNA; Lane 3, 2
M LiCI precipitate; Lane 4, 2 M LiCI supernatant.
25S
The strain used contained both L and M double-
185
stranded RNAs. The position of the ISS and 25S
ribosomal RNAs is also indicated. 1234
The method first described by Cryer et at. (1975) has been widely
adopted as the method of choice, and yields up to 0.25 mg DNA/g cells
can be achieved. The initial step invariably involves removal of the cell
wall by treating cells with an appropriate lytic enzyme (see Section 2.2).
While diethyl oxydiformate (i.e. diethyl pyrocarbonate) can be used to
inactivate any nudeases associated with the enzyme preparation (Davis et
al., 1980), this compound unfortunately also chemically reacts with the
DNA, often rendering it useless for cloning and transformation experi-
ments. To ensure that the chromosomal DNA remains intact, the spher-
oplasts are gently lysed by SDS in the presence of proteinase K, the latter
ensuring that partial proteolysis of the lysate occurs. Deproteinization
and extraction of the lipids from the lysate is achieved by mixing with an
appropriate organic solvent (usually chloroform:isoamylalcohol in a
ratio of 24: 1), and the high-molecular-weight DNA is either "spooled" or
pelleted by centrifugation from the resulting aqueous phase following
ethanol precipitation. Contaminating RNA and polysaccharide often re-
mains at this stage and can be removed by treatment with DNase-free
RNase A. Deproteinization and separation of the DNA from the diges-
tion products can be achieved by selective isopropanol precipitation. In
addition, either before or (more generally) after the RNase/isopropanol
step, further purification can be achieved by employing an isopycnic
cesium chloride gradient, using either ethidium bromide (Radloff et al.,
1967) or 4' ,6-diamidino-2-phenylindole (DAPI) (Williamson and Fen-
nell, 1975) to enhance the separation of chromosomal (buoyant density
1.699 g/cm3 ) and mitochrondrial DNA (buoyant density 1.683 g/cm3 ).
Additionally the ribosomal DNA-containing heavy satellite DNA band
(buoyant density 1.704 g/cm 3 ) can usually also be resolved in these
gradients.
A number of more rapid miniprocedures employing small cultures
(5-50 ml) have been devised to accommodate the need to process more
than one strain at a time, for example, when studying integrated plas-
mids (see Chapter 5). Some are adaptations of the large-scale method
described earlier (e.g., Davis et al., 1980; Winston et al., 1986) while
others use harsh cell disruption techniques to avoid the need to prepare
spheroplasts (e.g., Specht et al., 1982; Weeks et al., 1986), although in the
latter cases a CsCl gradient is almost certainly required to produce DNA
of sufficient purify for restriction enzyme digestion. A more rapid and
efficient protocol has recently been devised that involves lysing spher-
oplasts at 65°C in the presence of the chaotropic agent guanidine hydro-
chloride (GuHCl) followed by deproteination with proteinase K and
pheno-chloroform extraction (Holm et al., 1986). The yields with this
protocol can reach 75% of the theoretical yield, a value reported to be
much higher than with other miniprocedures (Holm et al., 1986). We
have also founp the latter method to be the most reproducible mini-
procedure for a wide variety of laboratory strains (M. F. Tuite et al.,

unpublished). A slightly different approach is taken when preparing
chromosomal DNA for analysis by pulsed-field gradient gel electropho-
resis (Schwartz and Cantor, 1984) or orthogonal-field-alternation gel
electrophoresis (Carle and Olson, 1984) where it is critical that chromo-
somes are isolated intact (see Section 7).
5.2. Mitochondrial DNA
Mitochondrial DNA (mtDNA) represents 10-15% of the total DNA

of S. cerevisiae, depending on carbon source and growth rate, and a
number of protocols have been devised to isolate this fraction in high
yield and free of chromosomal DNA. These protocols do not employ
isolated mitochondria since this is a difficult task and isolated mitochon-
dria are invariably damaged and contaminated with chromosomal DNA.
The strategy is to isolate total DNA from a crude mitochondrial lysate
and rely on the difference in the A + T base content between nuclear
and mitochondrial DNA (61% and 83%, respectively). This difference
generates a buoyant density difference that can be exaggerated by the
use of the fluorescent, DNA-binding agents DAPI (Williamson and Fen-
nell, 1975) and bisbenzimide (Hoechst 33258; Hudspeth et al., 1980).
Advantages to using bisbenzimide include greater separation between
chromosomal and mtDNA and the fact that there is no need to remove
the dye from DNA preparations prior to restriction enzyme digestion
and transformation (Hudspeth et al., 1980). Mitochondrial DNA pre-
pared by this strategy can be in excess of 99.5% pure.
5.3. Plasmid DNA
It is more practicable to propagate and manipulate yeast recombi-

nant plasmids in E. coli since the m~ority of such plasmids are constructed
to allow for both selection and replication in this bacterium (see Chapter
5), and the methodology is both simpler and more reproducible for
producing high yields of pure covalently closed circular DNA (cccDNA).
Techniques do exist, however, for extracting native cccDNA molecules
from yeast, the most successful of which is that described by Devenish and
Newlon (1982). In this case spheroplasts are first lysed in alkaline condi-
tions (pH 12.45) to denature linear DNA (but not cccDNA) and then
subjected to a high salt-phenol extraction to remove single-stranded DNA
(ssDNA), the cccDNA being recovered by ethanol precipitation. This
technique, originally devised as a means of purifying a circular derivative
of chromosome III (the "ring chromosome" Strathern et al., 1979), can
also be used for general plasmid isolation from yeast (Devenish and
Newlon, 1982).
If one is interested in obtaining plasmid DNA from yeast without
destroying its chromatin structure, Shalitin and Vishlizky (1984) have
described a technique that employs nondetergent lysis of purified nuclei
coupled with a metrizamide [2-(3-acetamido-5-N-methylacetamido-
2,4,6-triiodo-benzamido)-2-deoxy-o-glucose] gradient. This technique
allows for the purification of intact 2-f.Lm "minichromosomes," free of
cellular chromatin and ribosomal protein contamination, a problem
often associated with techniques based on separation in sucrose gra-
dients (Shalitin et al., 1983).
6. IN VITRO SYSTEMS
6.1. In Vitro Transcription Systems
Test tube systems that can accurately and efficiently transcribe

cloned yeast genes are an essential component of the armory of tech-
niques required for a complete understanding of gene expression. The
subject has a rather checkered history. Initial enthusiasm following the
purification of the three RNA polymerase complexes from yeast was
tempered by the finding that such pure preparations usually failed to
transcribe accurately in vitro. A retreat was made into the use of isolated
nuclei and of whole celllysates, which were often combined with tech-
niques for selective isolation of transcripts that had been initiated in the
cell-free system. Recently, identification of specific proteins that act as
transcription activating factors has led to a more complete understand-
ing of the components required for in vitro transcription and permitted
the design of more defined systems.
Sawadago et al. (1981) demonstrated that purified RNA polymerase
I (or A) together with the transcription factor "P37 " was able to selec-
tively transcribe a cloned rRNA gene in a plasmid vector. However, it
was found that the in vitro transcript was initiated at least 200 bp up-
stream of the start site which had been identified in vivo. Swanson and
Holland (1983) obtained similar results with a cell-free system in which a
yeast lysate was subjected to (NH 4 ) 2 S04 fractionation and ultracentrifu-
gation. A correspondence between the in vivo and in vitro sites of initia-
tion of synthesis of the 35S pre-rRNA was only obtained using less de-
fined systems. Since a considerable amount of "run-off' synthesis occurs
in these crude systems, it is essential to have some means of selecting, for
further analysis, only those transcripts which have been initiated in vitro.
This is achieved by including 'Y-SH derivatives of ribonucleotide tri-
phosphates in the reaction mix and purifying molecules that carry a 5'
thiol group (and hence were initiated in vitro) by Hg-agarose chro-
matography. This approach was first exploited by Lohr and Ide ( 1983) in
a system based on isolated nuclei. Both RNA polymerase I (or A) and
III (or C) activities may be detected in these nuclear preparations, and
these enzymes accurately initiate the synthesis of 35S pre-rRN A and of 5S
rRNA, respectively. Klootwijk et al. (1984) employed the 5'-thiol selec-
tion system to identify transcripts synthesized by highly concentrated
whole cell lysates. These workers included two ribonuclease inhibitors,
vanadyl ribonucleotide complex and RNAsin, in their reaction mixes
and were able to demonstrate the activity of all three yeast RNA poly-
merase complexes. Accurate initiation of the synthesis of the 35S rRNA
precursor molecule from the start site defined in vivo was recorded.
The development of cell-free systems that faithfully transcribe pro-
tein-encoding genes via the activity of RNA polymerase II (or B) has
been difficult and efficient methods have been developed only recently.
Initial studies using the purified polymerase (Carnevali et al., 1982) dem-
onstrated that initiation occurred in vitro, as could be recognized by the
formation of "ternary transcription complexes," but that elongation was
defective. A fully functional system has now been developed by Lue and
Kornberg ( 1987). The method involves isolation of yeast nuclei and their
subsequent lysis, the lysate being fractionated using two (NH 4 ) 2 S04 pre-
cipitation steps. Three different protease inhibitors [phenylmethyl sul-
fonyl fluoride (PMSF), pepstatin, and leupeptin] are included in the
buffers for both spheroplast and nuclear lysis. However, the most impor-
tant innovation appears to be the incorporation of Ficoll into the spher-
oplast lysis buffer. This polymer is essential as it permits the nuclei to
retain protein transcription factors that would otherwise be lost in the
supernatant when the organelles are collected by centrifugation.
Nuclear extracts formed the basis of the first RNA polymerase III
(or C) in vitro transcription system developed by Klekamp and Weil
(1982). However, they found that the yields were too low, and the prepa-
ration method too tedious, for routine use. These workers then devel-
oped a system based on a 100,000-g supernatant (S100) of a whole cell
lysate prepared by glass bead breakage. The S 100 is subsequently frac-
tionated using (NH 4 ) 2 S04 precipitation, residual nucleic acids being re-
moved by ion exchange chromatography. This system has formed the
basis for most subsequent protocols for the cell-free synthesis of transfer
RNA. Recent developments have come from Geiduschek's laboratory
(Kassovetis et al., 1989). These workers have developed systems in which
either multiple or single rounds of transcription may be studied. In the
former case, a cell extract of the Klekamp and Weil (1982) type is frac-
tionated using DEAE-sephadex chromatography. Such extracts contain
not only RNA polymerase III (or C), but also the transcription factor
TFIIIC and will carry out multiple rounds of transcription on a plasmid

template in the presence of all four ribonucleotide triphosphates. Alter-
natively purified RNA polymerase III, prepared using the heparin-
agarose methodology of Hammond and Holland (1983), may be used
together with affinity-purified TFIIIC. This is the most defined yeast
system yet developed for accurate in vitro transcription. In the single-
round protocol, the initial incubation is performed in the absence of
GTP to obtain stable ternary transcription complexes. GTP is then add-
ed back to the reaction mix in the presence of heparin. This permits
elongation of the nascent RNA chain but prevents any reinitiation or
transcriptional processing.
6.2. In Vitro Translation Systems

The ability to translate natural mRNAs in a cell-free lysate requires
that all three functional steps of the process of translation are occurring
in vitro, namely, initiation, elongation, and termination. On the other
hand, artificial templates such as poly(U) require only a functional
elongation system since the initiation upon this template is "forced" by
using supraoptimallevels of Mg 2 + ions. Both kinds of system are power-
ful tools for studying mRNA-coding potential, fidelity of translation,
and the basic mechanisms of protein synthesis.
A poly(U)-driven system was developed for S. cerevisiae by straight-
forward extrapolation of the techniques used successfully with other
organisms (see Sissons, 1978), yet attempts at establishing an mRNA-
dependent cell-free translation system were numerous, hut singularly
unsuccessful, until 1979, when Gasior et al. (1979a) finally reported the
successful establishment of a cell-free system able to initiate translation
upon natural mRNAs. Since this first report the system has been op-
timized for a wide range of mRNAs and has been used in a number of
studies (reviewed in Tuite and Plesset, 1986). In addition, a number of
modifications have been made, largely in the cell disruption protocols,
that allow researchers to reproducibly generate highly active systems
from a range of yeast strains.
6.2.1. Systems for Translating Synthetic Templates

Translation of a polynucleotide can be achieved in the absence of a
natural initiation mechanism, but generally requires the presence of
supraoptimal Mg2 + concentrations. Two types of systems have been
developed for S. cerevisiae. The first of these is analogous to the E. coli
S30 systems and simply consists of a crude S30 lysate, prepared from
either disrupted whole cells or lysed spheroplasts (Sissons, 1978) and
subjected to either dialysis or gel filtration chromatography to lower the
amount of endogenous amino acids. The second type of system com-

prises highly purified ribosomes and translation factors, particularly
those required for elongation in the yeast, namely, elongation factors
EF-1, EF-2, and EF-3. This highly fractionated system has provided
important insights into the mechanism of elongation in yeast (e.g., the
existence of a novel elongation factor, EF-3; Skogerson and Wakatama,
1976) and the basis of ribosome-mediated antibiotic resistance (e.g., re-
sistance to the trichothecene antibiotic trichodermin is expressed in the
60S subunit; Schindler et al., 1974). Nevertheless, it is far more difficult
to prepare than the S30 system and more prone to failure because of the
increased chance of damaging labile translation components during
their lengthy purification.
The yeast S30 system, capable of efficiently elongating on synthetic
templates such as poly(U) and poly(U ,C), has been useful in studying the
mechanism of action of inhibitors of translation (Vazquez and Jimenez,
1980), the fidelity of protein synthesis (e.g., Masurekar et al., 1981), and
the ionic requirements for elongation (Sissons, 1978), yet it is unable to
initiate translation upon exogenously supplied natural mRNA species. It
has been demonstrated, however, that the system can translate endoge-
nous yeast mRNAs already bound to ribosomes prior to cell disruption
(Gallis and Young, 1975).
6.2.2. Systems for Translating Natural mRNAs

A yeast cell-free system capable of carrying out all phases of protein
synthesis-initiation, elongation, and termination-using exogenously
supplied natural mRNAs was first described by Gasior et al. (1979a). This
system basically consists of an S 100 lysate from which polysomes (but not
monosomes or ribosomal subunits) have been removed by differential
ultracentrifugation and which has been subjected to gel filtration chro-
matography to remove low-molecular-weight components such as amino
acids and nucleoside triphosphates. An unusual characteristic of the
system is that it will only initiate translation at temperatures of 20°C or
less, a characteristic unique to yeast. The lysate can be effectively ren-
dered mRNA-dependent by treating with micrococcal nuclease (Pelham
and Jackson, 1976). Detailed studies of the ionic requirements and cell
disruption strategies (Tuite et al., 1980; Hofbauer et al., 1982; Tuite and
Plesset, 1986) have allowed the system to be optimized for the efficient
and faithful translation of both homologous mRNAs and a wide range
of heterologous mRNAs, be they viral, plant, or mammalian in origin
(Tuite and Plesset, 1986).
The Gasior system (Gasior et al., 1979a) has already proven useful in
studying the basic mechanism of translational initiation in yeast (Gasior
et al., 1979b), in analyzing yeast nonsense suppressors and their modi-
!102 MICHAEL F. TUITE and STEPHEN G. OLIVER
tiers (Tuite et al., 1981, 1983), in identifying the defects in conditional

lethal protein synthesis mutants (review, Moldave and McLaughlin,
1988), and in examining the role of translational control in the yeast heat
shock response (Plesset et al., 1982).
Since the initial report by Gasior et al. (1979a), a number of other
yeast cell-free systems have been reported that are also capable of trans-
lating exogenous mRNAs (Szczesna and Filipowicz, 1980; Hofbauer et
al., 1982; Kreutzfeldt and Lochmann, 1983; Chanda and Kung, 1983;
Hussain and Leibowitz, 1986). While all of these systems show the same
temperature requirement for initiation as reported by Gasior et al.
( 1979a), their preparation is generally less time-consuming, employing
S30 rather than S100 lysates and using whole cells rather than spher-
oplasts to generate the lysate. A number of them suffer from having
high endogenous mRNA activity, although the use of micrococcal nu-
clease (see above) can reduce this problem. One problem with the Gasior
system has been that, because efficient spheroplast formation is an abso-
lute necessity, and not all laboratory strains can be efficiently converted
to spheroplasts, there is a lack of strain-to-strain reproducibility. Em-
ploying whole cell disruption methods such as glass bead lysis (Hofbauer
et al., 1982; Tu'ite and Plesset, 1986; Hussain and Leibowitz, 1986) or the
French Press (Kreutzfeldt and Lochmann, 1983) allows one to overcome
this problem, although it is found to be essential that a protease inhibitor
such as PMSF be included during cell disruption to ensure the high
levels of translational activity of the Gasior system (Chanda and Kung,
1983; Hussain and Leibowitz, 1986).
6.2.3. Systems that Posttranslationally Process Secretory Proteins

Cell-free translation systems have been widely used to study the first
step in the secretion of proteins from eukaryotic cells, namely, transloca-
tion of the secretory polypeptide across the rough endoplasmic re-
ticulum (RER) membrane. Microsomes are a source of translocation
competent membranes. With the increasing use of yeast as a model
organism for studying the secretory process, the need for a yeast cell-
free system that is capable not only of translating natural mRNAs, but
also of translocating a secretory protein across microsomal membranes,
became important. Such a system has now been developed (Rothblatt
and Meyer, 1986; Waters and Blobel, 1986) and this was achieved by
including a microsomal membrane fraction from yeast that contains the
activities necessary for translocation, glycosylation, and cleavage of the
signal sequence from the precursors of such yeast secretory proteins as
invertase and mating factor a. The microsome-containing membrane
fraction is prepared from gently lysed spheroplasts.
Rothblatt and Meyer (1986) reported that active yeast microsome
membranes failed to process the prepro-a-factor secretory protein in a

heterologous cell-free translation system, namely, wheat germ, nor
would pancreatic microsomes process prepro-a-factor in the yeast lysate,
pointing to an inability of yeast translation complexes to interact prop-
erly with membranes from higher eukaryotes, and vice versa. The homol-
ogous yeast-processing system, however, will prove of great value in
dissection of the yeast secretion pathway, which has already been so
effectively defined genetically, and in studying the role of the signal
recognition particle (SRP) and its receptor.
6.2.4. Systems that Translate Mitochondrial mRNAs

Utilization of a different genetic code by mitochondria, both in yeast
and in higher eukaryotes, means that mitochondrial mRN As are gener-
ally not translatable in cytoplasmically derived cell-free systems such as
those described in Section 6.2.2. The major problem in yeast is that the
UGA codon is recognized as specifying tryptophan in mitochondria.
While supplementing a wheat germ cell-free system with yeast UGA
suppressor tRNA allows the faithful translation of yeast mitochondrial
mRNAs (de Ronde et al., 1980), the ideal solution is to prepare an in vitro
translation system composed entirely of yeast mitochondrial translation
components. Two alternative approaches have been taken in the devel-
opment of such systems. In the first, yeast mitochondria are purified,
fractionated into the various necessary components, and then recon-
stituted (Pfisterer and Buetow, 1981). Such a reconstituted system has
proven useful in the definition of the components, and to some extent
the mechanism, of mitochondrial protein synthesis, but, like its
cytoplasmic counterpart (see Section 6.2.2), the system is difficult and
time-consuming to prepare. The second approach has been the develop-
ment of an in vitro mitochondrial protein synthesis system which makes
use of intact yeast mitochondria (e.g., McKee and Poyton, 1984). Such a
system is able to synthesize authentic yeast mitochondrial translation
products and can initiate new polypeptide chains (McKee et al., 1984).
6.3. In Vitro DNA Synthesis

The development of a cell-free synthetic system is essential for elu-
cidating the detailed biochemistry of macromolecular synthesis. DNA
synthesis is no exception to this rule, as has been demonstrated by stud-
ies on bacteria and on animal viruses (see review by Kornberg, 1980).
The first attempts to generate in vitro systems for yeast DNA synthesis,
like those for bacteria, involved the use of permeabilized cells. Hereford
and Hartwell (1971) permeabilized yeast using the nonionic detergent
"Brij 58" and an incubation cocktail which included the four deoxynu-
cleotide triphosphates, ATP and a regenerating system, 13-mercap-

toethanol, and Mg 2 + ions. Using this system, they were able to show that
DNA synthesis in permeabilized cells carrying the cdc8 mutation was
temperature-sensitive. Banks (1973) demonstrated that only mitochon-
drial DNA is synthesized in Brij-permeabilized yeast cells, although this
has been contested by Kuo and Campbell (1982), who claimed that DNA-
zero petite mutants were still active in such a system.
Permeabilized cell systems appear to offer little opportunity for ex-
periments involving reconstruction of the DNA replication complex.
Nevertheless, having demonstrated that thymidylate kinase was the
CDCB gene product, Jong et al. ( 1984) were able to complement the cdc8
temperature-sensitive lesion by adding purified enzyme to permeabil-
ized cells. Thymidylate kinase may be a special case, however, since it is
not clear whether the protein must interact directly with the replication
complex or whether complementation can be effected by precursor mol-
ecules traveling in both directions across the permeabilized cell mem-
branes. More complex reconstruction experiments demand a truly cell-
free DNA synthesis system.
Three such systems are currently in use, developed in the laborato-
ries ofJazwinski Uazwinski and Edelman, 1982), Sugino (Kojo et al., 1978),
and Campbell (Celniker and Campbell, 1982), respectively. All three
involve the preparation of protoplasts using lytic enzyme and their lysis in
the presence Uazwinski; Campbell) or absence (Sugino) of detergent. A
high-speed supernatant is then prepared and fractionated by ammonium
sulfate precipitation. The resuspended precipitate may then be used
without further fractionation (Sugino) or subjected to gel filtration Uaz-
winski) or ion exchange chromatography (Campbell). The incubation
cocktail employed includes the four deoxyribonucleoside triphosphates,
the four ribonucleotide triphosphates, glycerol, and dithiothreitol. The
Jazwinski method includes EDTA and an ATP regenerating system,
whereas both Campbell and Sugino omit these but include spermidine.
The Sugino cocktail also used NAD+. There have been various claims
and counterclaims about the merits of these three systems and they have
been critically evaluated by Jong and Scott ( 1985 ).
The incorporation of radiolabeled triphosphate in all three cases is
dependent on exogenously added DNA and, it is claimed, is absolutely
specific for DNA containing yeast origins of replication and capable of
the de novo initiation of DNA synthesis. Jong and Scott (1985) claim that
the Jazwinski system, at least as originally described Uazwinski and
Edelman, 1982), is not specific for yeast DNA. These workers have fur-
ther suggested that, when recombinant plasmids prepared from bacteria
are used as templates, only elongation rather than initiation is observed
since oligonucleotide primers are copurified with the plasmid molecules
Uong and Scott, 1985 ). The alkaline denaturation of the plasmids, to
remove these primers, followed by renaturation was said to prevent

DNA synthesis in a modified Sugino system. Both Campbell (Celniker
and Campbell, 1982) and Jazwinski Uazwinski and Edelman, 1984; Jaz-
winski, 1987) have reported that alkali pretreatment of template mole-
cules does not prevent DNA synthesis in their systems. This prepriming
effect may account for the very high levels of DNA synthesis found in
the Sugino system (Koko et al., 1978).
The replication activity from such yeast extracts may be isolated as
high-molecular-weight complex of ca. 2 X 106 Da Uazwinski and
Edelman, 1982). A number of enzyme activities that might be expected
to be found in a "replisome" (Kornberg, 1980) are associated with this
complex. These include DNA polymerase I, DNA ligase, DNA primase,
and topoisomerase II Uazwinski and Edelman, 1984). The large size of
the replication complex means that its association with DNA may be
visualized by electron microscopy. Protein "knobs" are found associated
with portions of the 2-JJ,m molecule which have been shown (Newlon et
al., 1981) to function as in vivo origins Uazwinski and Edelman, 1982).
Similarly, these knobs are also found associated with restriction frag-
ments of YRp plasmids, which are known to contain a functional ARS
element Uazwinski et al., 1983).
Plasmid DNA treated with any of the three types of extract has been
found, by electron microscopy, to contain replication bubbles with ap-
propriately positioned centers Uazwinski and Edelman, 1979; Kojo et al.,
1978; Celniker and Campbell, 1982). In the case of the Campbell system
such analysis demonstrated the existence of "bubbles within bubbles,"
indicating that reinitiation events had occurred (Celniker and Campbell,
1982). Reinitiation is prohibited in eukaryotic nuclear DNA in vivo, and
it is not dear whether this phenomenon reflects an intrinsic defect of the
cell-free system or whether it is an artifact produced by the inclusion of
the chain-terminating analog cytosine-P-n-arabinoside-5 '-triphosphate
(ara CTP) in the reaction mix. Celniker and Campbell (1982) added this
analog in order to enrich the system for replication complexes. Its use
enabled them to perform labeling experiments to confirm the site of the
replication origin within the plasmids used and also to demonstrate that
replication was bidirectional. The latter was only obtained in extracts
made from exponential phase cells; when stationary phase cells were
used, then replication appeared to be unidirectional.
The combination of an effective cell-free DNA replication system
and the many temperature-sensitive mutants of yeast available which
have blocks in DNA synthesis is a powerful tool in the analysis of the
biochemistry of eukaryotic DNA replication. However, doubts still re-
main as to the accuracy with which the currently employed cell-free
system represents normal DNA replication. The availability of improved
methods of analyzing the in vivo situation, such as two-dimensional gel
electrophoresis (Brewer and Fangman, 1987; Huberman et al., 1987),

has tended to shift attention away from in vitro techniques, at least for
the present.
6.4. Two-Dimensional Gel Electrophoresis in the Analysis

of DNA Replication
The position, in an electrophoresis gel, of any DNA sequence for
which a suitable restriction fragment or oligonucleotide probe exists may
be accurately determined by the Southern blotting procedure (South-
ern, 197 5). This opens the way to the identification of those sequences
which act as origins of yeast chromosomal DNA replication. To achieve
this it is necessary to be able to select or enrich for replication intermedi-
ates in DNA prepared from cultures of growing yeast cells and to utilize
gel electrophoresis to resolve those intermediates from the bulk of the
chromosomal DNA. Two methods, which both exploit two-dimensional
agarose gel electrophoresis (Brewer and Fangman, 1987; Huberman et
al., 1987), have been developed to achieve this.
In the Brewer and Fangman (1987) approach, DNA preparations
enriched for replication intermediates are obtained by using syn-
chronized cultures. The synchrony technique employed involves hold-
ing cells in G 1 by use of a combined a-factor/cdc7 block (Fangman et al.,
1983) and then releasing the inhibition by proteolytically degrading the
pheromone and shifting to the permissive temperature.
The two-dimensional agarose gel electrophoresis methods em-
ployed separate molecules in the first phase according to their mass and
in the second phase according to their shape (Bell and Byers, 1983). The
first dimension is run in 0.4% agarose at 1 V/cm whereas the second
dimension is run in 1% agarose and 5 V/em in the presence of ethidium
bromide. Nonlinear DNA molecules, such as the "Y" and "eye" forms
expected of replication intermediates, are known to migrate anoma-
lously in agarose gels as compared to linear molecules of equal mass
(Oppenheim, 1981). The increase in voltage and agarose concentration
in the second dimension exaggerates these differences (Bell and Byers,
1983). Figure 5 diagrams possible replication intermediates and shows
the way they are predicted to migrate in the two-dimensional gel system.
The approach of Huberman et al. (1987) differs from that of
Brewer and Fangman (1987) in both the method of DNA preparation
and the electrophoretic conditions employed. Synchronized cultures are
not used, but DNA preparations from random cultures are enriched for
replication intermediates using a biochemical method. Molecules con-
taining single-stranded regions, which will include the replicating forms,
are selectively adsorbed onto benzoylated naphthoylated DEAE-cel-
lulose (BND-c<;llulose; Gamper et al., 1985). The first electrophoresis
dimension employs 0.5% agarose at 40-45 V under neutral conditions
--< -o- >---< --o-

-< -o- >---::. -o-
-< -o
0
>-<
X
-<
-1st
A ~ ~ £\
1
2nd
2k'ii"" .. 2kb"" ..
1kb iiJ 1dJ 'iN!
Figure 5. Two-dimensional gel electrophoresis in the analysis of DNA replication. (Upper
panel) The types of replication intermediate that arise depending on the placement of the
replication origin on an arbitrary 1-kb restriction fragment. (Lower panel) The predicted
migration of the four types of replication intermediate in two-dimensional agarose gels
(for details, see text). The dashed lines mark the locations of linear molecules of various
sizes. (Adapted from Brewer and Fangman, 1987.)
in the presence of ethidium bromide, whereas the second dimension

uses 1.2% agarose gels run at 12 or 17 V under alkaline conditions. The
second, denaturing, phase separates simple linear DNA fragments into
their two single strands. Molecules containing a replication fork, on the
other hand, will produce four strands on denaturation: the two parental
strands and two, smaller, nascent strands. The linear molecules and
parental strands migrate as an arc in the two-dimensional system, with
the smaller nascent strands running ahead of this major arc. Closed and
open circular molecules also migrate ahead of the major arc, and this
complicates the analysis of plasmid replication unless all molecules are
linearized using an appropriate restriction cut. Single strand nicks in the
DNA also complicate the patterns, leading to vertical streaking in the
second dimension. In all, the data obtained from the Huberman et al.
( 1987) method is not as clear or as easily interpreted as that provided by
the Brewer and Fangman (1987) technique.
7. PULSED FIELD GEL ELECTROPHORESIS

OF YEAST CHROMOSOMES
Conventional agarose gel electrophoresis techniques fail to resolve

DNA molecules much greater than 20 kb in size. While large molecules
will enter the gel, different sized species do not migrate at significantly
different rates (Fisher and Dingman, 1971) and so are not resolved from
one another. Undigested DNA extracts from yeast or E. coli commonly
produce a chromosomal smear near the top of the gel. A solution to this
problem has been found in pulsed field gel electrophoresis, in which
DNA molecules are subjected to an electric field which periodically
changes its orientation. The time it takes molecules to realign to the new
field direction is proportional to their size, and molecules up to 1 mega-
base (mb) in size can be resolved one from another. Fortunately for yeast
researchers, the size of yeast chromosomes (240 kb-1.5 mb) falls within
these limits and thus pulsed field techniques may be used to provide a
karyotype of S. cerevisiae in a way never achieved in a convincing manner
by either light or electron microscopy. Moreover, the technique of
Southern blotting may be used to locate any cloned gene or other DNA
fragment to a particular chromosome and so an alternative to conven-
tional mapping procedures by meiotic segregation analysis (see Chapter
4) is provided. Furthermore, individual chromosomes may be isolated
from these agarose gels for restriction analysis or the construction of
gene banks.
A large number of different types of apparatus have been designed
to carry out pulsed field gel electrophoresis and have been applied suc-
cessfully to the resolution of yeast chromosomes. Unfortunately, our
theoretical understanding of the technique is poor and so a large
amount of empiricism is involved. A short discussion of the theory will
be provided before describing the main types of apparatus and the
methods of sample preparation.
7.1. Theoretical Background

All theories on pulsed gel electrophoresis assume that large DNA
molecules move through the gel matrix by a process of reptation. This
term, which literally means "snakelike," implies that the molecules move
through the gel and on, with their long axis parallel to the electric field.
Indeed, large circular molecules, such as circular derivatives of yeast
chromosome III (Strathern et al., 1979), fail to enter the gel. Small cir-
cular molecules, such as the 2-~J.m plasmid, show a quite different rate of
migration relative to their size than the large linear chromosomes.
The pore size of a 1.5% agarose gel is much smaller than the coiled
length of a DNA molecule greater than 30 kb in size (Lumpkin and
Zimm, 1982; Serwer and Hayes, 1986). Thus, when the orientation of
the electric field through the gel is changed, the molecule is lying across
many of the gel pores, which will allow it to move in the new direction of
electrophoresis. The molecule must therefore reorientate before it can
move forward again, and the time taken for this reorientation is propor-
tional to its size (Schwartz and Cantor, 1984). In a pulsed field gel,
according to this theory, large linear DNA molecules are resolved from
one another on the basis of their reorientation time.
An alternative view has been presented by Southern et al. ( 1g87).
Their theory is again based on reptatory movement and postulates that
the leading end of the molecule can never cross the path of the rest of
the DNA trailing behind it. In the simplest form of the theory, when the
direction of the electric field changes only one end or the other is able to
lead the molecule in the new direction. With a field change angle of goo
either end may lead off in the new direction. When the new field is at less
than goo to the previous one, the old leading end continues to lead. But
when the change in field orientation is greater than goo, it is the old
trailing end that becomes the leader. In support of their theory, South-
ern et al. ( 1g87) demonstrated that field change angles of <goo provided
no better resolution of large linear molecules than conventional gel elec-
trophoresis, while angles of >goo produced increasing resolution of
large DNA species. The critical importance of the angle of field reorien-
tation was also noted by Chu et al. ( }g86).
While the heuristic properties of the second model are appealing,
they do not explain how molecules are resolved when the field change
angle is 180°. Such a situation obtains in the field inversion gel electro-
phoresis (FIGE) technique (Carle et al., 1g86) when the direction of
electrophoresis is reversed periodically. In such a situation the old trail-
ing ends should always be chosen to be the new leader, but the new
trailing ends would all be in the same position relative to the new field.
The success of FIGE implies that there is some physical difference be-
tween the two ends of the molecule, and Carle et al. (1 986) talk of DNA
adopting a "wedgelike" configuration. The physical basis of this is not
clear and an equivalent result would obtain if there was, for instance, a
charge gradient along the length of the molecule. We have far to go in
understanding this new tool of pulsed field gradient gel electrophoresis
(PFGGE); however, our theoretical inadequacy does not prevent us from
exploiting the technique.
7.2. Pulsed Field Gradient Gel Electrophoresis

This was the original pulsed field technique developed by Schwartz
and Cantor (1g84) and may now be considered obsolete. The apparatus
employs arrays of small electrodes perpendicular to one another (Fig. 6)
to give a field reorientation angle of goo. Usually, all available electrodes
were used in one orientation, to create a uniform field, but only a single
anode and all of the cathodes in the second orientation. This produced a
gradient of field strength in the second direction and resulted in bands
that had migrated the same distance in adjacent tracks lying on a curved
diagonal across the gel. This makes track-to-track comparisons difficult,
A+
!
lf-o ....-----__..;;.--,o-M
!of-. o-M
!of-. o-M Figure 6. Geometries for pulsed field gradient
!of-. ~ gel electrophoresis (PFGGE). A and B repre-
B-~ --MB+ sent the two alternative electric fields used in
!of-. o-M the system. Their polarities are indicated by
~ o-M plus ( +) and minus (-) signs, the electrodes
lf-4 = = = ... employed being shown as either heavy lines or
~ o-Jolll filled circles. The heavy arrows (~) represent
:t:t:t:l-:t:t:t:t diodes. Three sample slots, indicated by open
A- rectangles, are also shown.
so this method was quickly replaced by the orthogonal-field-alternation

gel electrophoresis (OFAGE) system of Carle and Olson (1984).
7.3. Orthogonal-Field-Alternation Gel Electrophoresis

The OFAGE system uses two sets of continuous electrodes placed at
right angles to one another. In the usual configuration (Fig. 7) the
anodes are much smaller than, and are placed near to one end of, the
cathodes. This means that the field change angle changes continuously
across the gel and produces a field gradient analogous to that obtained
with the discontinuous electrode system of PFGGE. It is unlikely that this
gradient is essential to the resolving power of the system (Chu et al.,
1986; Southern et al., 1987) and causes a distortion of the track path,
particularly for tracks distal to the center of the gel (see Fig. 8). However,
even when cathode and anode are of equal length in both sets of elec-
trodes, similar distortion is observed (Carle and Olson, 1984). This is
because the resistance of the platinum electrodes is less than that of the
running buffer, so an induced current is produced in the two inactive
electrodes. This results in an inhomogeneous field and distorts the mi-
D
Figure 7. Geometries for orthogonal-field-
alternation gel electrophoresis (OFAGE).
For details of symbols see legend to Fig. 6.
IV
VII (U40) , XV
XVI
XIII
II
XIV
XI
v
VIII
IX
Ill
VI
Figure 8. Separation of yeast chromosomes by CHEF pulsed field gel electrophoresis. The
left track presents the electrophoretic karyotype of S. cerevisiae with the bands correspond-
ing to the different chromosomes identified together with their estimated size (in kb). The
right track shows a "ladder" of multimers of phage lambda DNA which were used as size
standards. (The help of Mike Musialowski and Nadia Danhash in the preparation of this
figure is gratefully acknowledged.)
gration pathway. While OFAGE is a great improvement in PFGGE, it is

still difficult to make accurate track-to-track comparisons for a large
number of samples. Various novel electrode configurations have been
employed to obtain homogeneous electric field and obviate track
distortions.
7.4. Contour-Clamped Homogeneous Electric Field

The reasoning behind the electrode configuration in the contour-
damped homogeneous electric field (CHEF) system (Fig. 9) can be de-
scribed as follows: The induced current in the inactive electrodes which
distorts the field in the OFAGE system could be obviated if the electrodes
1~1
Figure 9. Geometries for CHEF electrophoresis. For details of symbols see legend to Fig. 6.
were infinitely long. This is clearly impracticable, but the situation may
be approximated by using multiple electrodes separated by resistors that
"clamp" the potential of each individual electrode to the value it would
have reached in an infinitely long continuous wire. In the system devel-
oped by Chu et al. (1986) a hexagonal array of electrodes is employed
(Fig. 9). The CHEF system is currently the method of choice for yeast
chromosome separations. By the use of low field strengths and low
agarose concentrations, Vollrath and Davis (1987) have succeeded in
resolving DNA molecules of greater than 5 mb in size and in producing
electrophoretic karyotypes of Candida albicans and Schizosaccharomyces
pombe, yeast species that have much larger chromosome than S. cerevisiae.
7.5. Vertical Pulsed-Field Gradient Gel Electrophoresis and

Constant Electric Field with Rotating Gel Platform
Two other solutions to the field distortion problem of OFAGE have
been found. In vertical pulsed-field gradient gel electrophoresis (VPFG-
GR) (Gardiner et al., 1986); lateral distortions are removed by alternating
the field through the thickness of a vertical gel rather than along the
length of a horizontal one. In the method of Southern et al. (1987) a
conventional horizontal gel system with just one set of electrodes of
equal length is employed. However, the gel itself is rotated through an
angle of 110° with respect to the electric field. Good results for the
separation of yeast chromosomes have been published for both systems.
7.6. Field Inversion Gel Electrophoresis

A radically different system of pulsed gel electrophoresis has been
developed by Carle et al. ( 1986). This employs a conventional horizontal
gel electrophoresis system but periodically inverts the electric field such
that the DNA molecules migrate toward the anode in a "two steps for-
A+ (B-)
Figure 10. Geometries for FIGE. For details of symbols see

legend to Fig. 6. A-
D (B+)
ward, one step back" fashion (Fig. 10). Thus the field reorientation angle
is a maximum 180°. The system has the merit of simplicity for its appa-
ratus but has the procedural complication that the migration of DNA
molecules within a certain size range is length-independent at any given
switching interval. Thus molecules of quite disparate size can unexpect-
edly comigrate. This problem may be largely overcome by ramping the
switch interval such that it increases linearly as the run progresses. Such
ramped control of switching may easily be achieved by use of a very
simple microcomputer. The FIGE and OFAGE techniques may be com-
bined to produce a two-dimensional separation of yeast chromosomes
(Carle et al., 1986).
7. 7. General Considerations
All of the systems discussed here are available commercially, but
OFAGE, CHEF, and FIGE systems may readily be constructed in-house.
Any standard power pack that can maintain 350 V at 350 rnA (milli-
amps) may be used and cyclic time delay relays may be purchased cheap-
ly from most suppliers of electronic and electrical components. The
cooling of the running buffer is very important and a temperature of
12-l4°C must be maintained. A suitable heat exchanger and a recir-
culating pump must be supplied. Recirculation of the buffer is most
effectively maintained using a peristaltic pump and wide-bore silicone
tubing.
7.8. Sample Preparation

Systems of sample preparation that avoid shear are essential if intact
chromosomes are to be obtained. This is achieved by embedding yeast
cells in slabs or beads of agarose, removing their cell walls by enzyme
treatment, and then lysing the resultant spheroplasts in situ. The slab
system was originally introduced by Schwartz and Cantor (1984) and is
the most widely used procedure since it is simple and reliable. The
production of agarose beads involves suspending the cells in molten
agarose and then shaking with paraffin oil to form an emulsion. The
main merit of the bead system is that lower enzyme concentrations are
required for spheroplast formation or for the digestion of DNA with
rare-cutting restriction endonucleases.
REFERENCES
Achstetter, T., and Wolf, D. H. 1985. Proteinases, proteolysis and biological control in the
yeast Saccharomyces cerevisiae, Yeast 1:139-158.
Baan, R. A., Keller, P. B., and Dahlberg, A. E. 1981. Isolation of eukaryoticinitiation factor
2 from yeast Saccharomyces cerevisiae, ]. Biol. Chem. 256: I 063-1066.
Ballou, C. E. 1982. Yeast cell wall and cell surface, in: The Molecular Biology rf the Yeast
Saccharomyces: Metabolism and Gene Expression (J. N. Strathern, E. W. Jones, and J. R.
Banks, G. R. 1973. Mitochondrial DNA synthesis in permeable cells, Nature 245:196-199.
Bell, L., and Byers, B. 1983. Separation of branched from linear DNA by two-dimensional
_gel electrophoresis, Anal. Biochem. 130:527-535.
Brewer, B. J., and Fangman, W. L. 1987. The localization of replication origins on ARS
plasmids inS. cerevisiae, Cell51:463-471.
Brunner, S., Dark, F. A., Gerhardt, P., Jeynes, M. H., Kandler, 0., Kellenberger, E.,
Klienberger-Nobel, E., McQuillen, K., Rubio-Huertos, M., Salton, M. R. J., Strange, R.
E., Tomcsik, j., and Weibull, C. 1958. Bacterial protoplasts, Nature 181:1713-1715.
Carle, G. F., and Olson, M. V. 1984. Separation of chromosomal DNA molecules from
yeast by orthogonal-field-alternation gel electrophoresis, Nucleic Acids Res. 12:5647-
5664.
Carle, G. F., Frank, M., and Olson, M. V. 1986. Electrophoretic separations of large DNA
molecules by periodic inversion of the electric field, Science 232:65-68.
Carnevali, F., Caserta, M., and DiMauro, E. 1982.In vitro transcription by purified RNA
polymerase II. Coarse promoter mapping on homologous cloned genes, Nucleic Acids
Res. 10:3195-3209.
Celniker, S. E., and Campbell, J. L. 1982. Yeast DNA replication in vitro: Initiation and
elongation events mimic in vivo processes, Cell 51:203-210.
Chan, P. Y., and Cossins, F. A. 1976. General properties and regulation of argini!te trans-
porting systems in Saccharomyces cerevisiae, Plant Cell Physiol. 17:341-349.
Chanda, P. K., and Kung, H-F. 1983.In vitro synthesis of biologically active human leuko-
cyte inteferon in a RNA-dependent system from Saccharomyces cerevisiae, Proc. Natl.
Acad. Sci. USA 80:2569-2573.
Chia, 1..-L., and McLaughlin, C. S. 1979. The half-life of mRNA in Saccharomyces cerevisiae,
Mol. Gen. Genet. 170:137-144.
Chirgwin, J. M., Przybyla, A. E., MacDonald, R~ J., and Rutter, W. J. 1979. Isolation of
biologically active ribonucleic acid from sources enriched in ribonuclease, Biochemistry
18:5294-5299.
Chu, G., Vollrath, D., and Davis, R. W. 1986. Separation of large DNA molecules by
contour-clamped homogeneous electric fields, Science 234:1582-1585.
Clare, J. j. 1982. Studies on the regulation of RNA synthesis in the yeast, Saccharomyces
cerevisiae. Ph.D. thesis, University of Kent at Canterbury.
Clare, J. J., and Oliver, S. G. 1982. The regulation of RNA synthesis in yeast. V. tRNA
charging studies, Mol. Gen. Genet. 188:86-102.
Cryer, D. R., Goldthwaite, C. D., Zinker, S., Lam, K.-B., Storm, E., Hirschberg, R.,
Blamire, j., Finkelstein, D. B., and Marmur, J. 1973. Studies on nuclear and mitochon-
drial DNA of Saccharomyces ceuvisiae, Cold Spring Harb. Symp. Quant. Biol. 58:17-29.
Cryer, D. R., Ecdeshall, R., and Marmur, J. 1975. Isolation of yeast DNA, Methods Cell Biol.
12:39-44.
Davis, R. W., Thomas, M., Cameron, J., St. John, T. P., Scherer, S., and Padgett, R. A. 1980.
Rapid DNA isolations for enzymatic and hybridisation analysis, Methods En%ymol.
65:404-411.
de Ronde, A., van Loon, A. P. G. M., and Grivell, L. A. 1980. In vitro suppression of UGA
codons in a mitochrondrial mRNA, Nature 287:361-363.
Devenish, R. J., and Newlon, C. S. 1982. Isolation and characterisation of yeast ring
chromosome III by a method applicable to other circular DNAs. Gene 18:277-288.
Diaz-Ruiz, J. R., and Kaper, J. M. 1978. Isolation of viral double-stranded RNAs using LiCl
fractionation procedure, Prep Biochem. 8: 1-17.
Douglas, M., Finkelstein, D., and Butow, R. A. 1979. Analysis of products of mitochondrial
protein synthesis in yeast: Genetic and biochemical aspects, Methods En%ymol. 56:58-
66.
Eddy, A. A., and Williamson, D. H. 1957. A method of isolating protoplasts from yeast,
Nature 179:1252-1253.
Fangman, W. L., Hice, R. H., and Chlebowicz-Sledziewska, E. 1983. ARS replication during
the yeastS-phase, Cell 32:831-838.
Faulhammer, H. G., and Cramer, F. 1977. Tyrosyl-tRNA synthetase from Baker's yeast:
Rapid isolation by affinity elution, molecular weight of the enzyme and determination
of sulfhydryl groups, Eur. ]. Biochem. 75:1664-1671.
Feinberg, B., and McLaughlin, C. S. 1988. Isolation of yeast mRNA and in vitro translation
in a yeast cell-free system, in: Yeast: A Practical Approach (I. Campbell and J. H. Duffus,
eds.), IRL Press, Oxford, pp. 147-161.
Fisher, M. P., and Dingman, C. W. 1971. Role of mr '. cular conformation in determining
the electrophoretic properties of polynucleotide& in agarose-acrylamide composite
gels, Biochemistry 10:1895-1899.
Franklin, R. M. 1966. Purification and properties of the replicative intermediate of the
RNA bacteriop~a_!ie R17, Proc. Natl. Acad. Sci. USA 55:1504-1511.
Gallis, B. M., and Young, E. T. 1975. Endogenous messenger RNA-directed polypeptide
chain elongation in a cell-free system from the yeast Saccharomyces cerevisiae,J. Bacteriol.
122:719-726.
Gamper, H., Lehman, H., Piette, J., and Hearst, J. 1985. Purification of DNA using
benzoylated naphthoylated DEAE-cellulose, Gene 4:157-164.
Gardiner, K., Laas, W., and Patterson, D. 1986. Fractionation of large mammalian DNA
restriction fragments using vertical pulsed-field gel electrophoresis, Somatic Cell Genet.
12: 185-195.
Gasior, E., Herrera, F., Sadnik, I., McLaughlin, C. S., and Moldave, K. 1979a. The prepa-
ration and characterisation of a cell-free system from Saccharomyces cerevisiae that
translates natural messenger ribonucleic acid,]. Biol. Chem. 254:3965-3969.
Gasior, E., Herrera, F., McLaughlin, C. S., and Moldave, K. 1979b. The analysis of inter-
mediary reactions involved in protein synthesis, in a cell-free extract of Saccharomyces
cerevisiae that translates natural messenger ribonucleic acid,]. Biol. Chem. 254:3970-
3976.
Gregory, D. W., Polakis, E. S., and Bartley, W. 1967. The isolation and purification of yeast
mitochondria, in: Methods in En%ymology, Vol. 12 (L. Grossman and K. Moldave, eds.),
Academic Press, New York, pp. 469-475.
Grivell, A. R., and Jackson, J. F. 1968. Thymidine kinase: Evidence for its absence from
Neurospora crassa and some other microorganisms and the relevance of this to the
specific labelling of DNA,]. Gen. Microbiol. 54:307-312.
Gross, K. J., and Pogo, 0. 1974. Control mechanism of ribonucleic acid synthesis in eu-
karyotes,]. Biol. Chem. 249:568-576.
Hammond, C. I., and Holland, M. J. 1983. Purification of yeast RNA polymerase using
heparin agarose affinity chromatography,]. Biol. Chem. 258:3230-3241.
Hampel, A. E., Enger, M.D., and Rider, P. E. 1979. Aminoacyl-tRNA synthetases from
cultured CHO cells, Methods Enz.ymol. 59:229-234.
Hartwell, L. H., and McLaughlin, C. S. 1968. Temperature-sensitive mutants of yeast
exhibiting a rapid inhibition of protein synthesis,]. Bacterial. 96:1664-1671.
Hereford, L. M., and Hartwell, L. H. 1971. Defective DNA synthesis in permeabilized
yeast mutants, Nature 254:171-172.
Hofbauer, R., Fessl, F., Hamilton, B., and Ruis, H. 1982. Preparation of an mRNA-depen-
dent cell-free translation system from whole cells of Saccharomyces cerevisiae, Eur. ].
Biochem. 122:199-203.
Holland, M. j., Hager, G. L., and Rutter, W. J. 1977. Characterisation of purified poly(ade-
nylic acid)-containing messenger RNA from Saccharomyces cerevisiae, Biochemistry 16:8-
16.
Holm, C., Meeks-Wagner, D. W., Fangman, F. L., and Botstein, D. 1986. A rapid, efficient
method for isolating DNA from yeast, Gene 42:169-173.
Huberman,J. A., Spotila, L. D., Nawotka, K. A., El-Assouli, S.M., and Davis, L. R. 1987.
The in vivo replication origin of the yeast 2 11-m plasmid, Cell 51:473-481.
Hudspeth, M. E. S., Shumard, D. S., Tatti, K. M., and Grossman, L. I. 1980. Rapid
purification of yeast mitochondrial DNA in high yield, Biochim. Biophys. Acta 610:221-
228.
Hughes, D. E., Wimpenny, j. W. T., and Lloyd, D. 1971. The disintegration of micro-
organisms, in: Methods in Microbiolog) Vol. 5B O-R. Norris and D. W. Ribbons, eds.),
Academic Press, New York, pp. 1-54.
Hussain, I., and Leibowitz, M. J. 1986. Translation of homologous and heterologous
mRNAs in a yeast cell-free system, Gene 46:13-23.
Hutchison, H. T., and Hartwell, L. H. 1967. Macromolecule synthesis in yeast
sphaeroplasts, J. Bacteriol. 94: 1697-1705.
jazwinski, S. M. 1987. Participation of ATP in the binding of a yeast replicative complex to
DNA, Biochem.]. 246:213-219.
Jazwinski, S. M., and Edelman, G. M. 1982. Protein complexes from active replication
fractions associate in vitro with the replication origins of yeast-211-m DNA plasmid,
Proc. Nat. Acad. Sci. USA 79:3428-3432.
Jazwinski, S.M., and Edelman, G. M. 1984. Evidence for the participation of a multipro-
tein complex in yeast DNA replication in vitro,]. Biol. Chem. 259:6852-6857.
jazwinski, S. M., Niedzwiecka, A., and Edelman, G. M. 1983. In vitro association of a
replication complex with a yeast chromosomal replicator, ]. Biol. Chem. 258:2754-
2757.
Jong, A. Y. S., and Scott,J. F. 1985. DNA synthesis in yeast cell-free extracts dependent on
recombinant plasmid purified from Escherichia coli, Nucleic Acids Res. 15:2943-2949. ·
jong, A. Y. S., Kuo, C-L., and Campbell, j. L. 1984. The CDC8 gene of yeast encodes
thymidylate kinase,]. Biol. Chem. 259:11052-11059.
Kandel, j. 1979. Isolation and detection of double-stranded RNA from fungi, Methods
Enzymol. 60:549-554.
K,assovetis, G. A., Riggs, D. L., Negri, R., Nguayen, L., and Geiduschek, E. P. 1989.
Transcription factor Ill B generates extended DNA interactions in RNA polymerase
Ill transcription complexes on tRNA genes, Mol. Cell Biol. 9:2551-2566.
Kirby, K. S. 1965. Isolation and characterization of ribosomal ribonucleic acid, Biochem.].
96:266-269. .
Kitamura, K., and Yamamoti, Y. 1972. Purification and properties of an enzyme,
zymolyase, which lyses viable yeast cells, Arch. Biochem. Biophys. 155:403-406.
Klekamp, M.S., and Weil, P. A. 1982. Specific transcription of homologous class Ill genes
in yeast soluble cell-free extracts,]. Biol. Chem. 257:8432-8441.
Klootwijk, J., Verbeet, M.Ph., Veldmann, G. M., deRegt, V. C. H. F., van Heerikhuizen, H.,
Bogerd, J., and Planta, R. J. 1984. The in vivo and in vitro initiation site for transcrip-
tion of the RNA operon of Saccharomyces carlsbergensis, Nucleic Acids Res. 12:1377-
1390.
Koch, H., and Friesen, J.D. 1979. Individual messenger RNA half lives in Saccharomyces
Kojo, H., Greenberg, B. D., and Sugino, A., 1978, Yeast 2 tJ.m plasmid replication in vitro:
Origin and direction, Proc. Natl. Acad. Sci. USA 78:7261-7265.
Kornberg, A. 1980. DNA Replication, W. H. Freeman, San Francisco.
Kreutzfeldt, C., and Lochmann, E-R. 1983. Preparation of a cell-free extract from yeast
that is active in protein synthesis, FEMS Microbial. Lett. 16:179-182.
Kuo, C-L., and Campbell, J. L. 1982. Purification of the cdc8 protein of Saccharomyces
cerevisiae by complementation in an aphidocolin-sensitive in vitro DNA replication
system, Proc. Natl. Acad. Sci. USA 79:4243-4247.
Kuo, S-C., and Yamamoto, S. 1975. Preparation and growth of yeast protoplasts, Methods
Cell Biol. 11:169-183.
Lang, B., Burger, G., Doxiadis, I., Thomas, D. Y., Bandlow, W., and Kaudewitz, F. 1977. A
simple method for the large scale preparation of mitochondria from microorganisms,
Anal. Biochem. 77:110-121.
Littlewood, R., Shaffer, B., and Davies, J. 1971. Mutants of Saccharomyces cerevisiae with
reduced levels of ribonuclease activity, Genetics 68:s39.
Lohr, D., and Ide, G. I. 1983. In vitro initiation and termination of ribosomal RNA tran-
scription in isolated yeast nuclei,]. Biol. Chem. 258:4668-4671.
Lue, N. F., and Kornberg, R. D. 1987. Accurate initiation at RNA polymerase II promoters
in extracts from Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. USA 84:8839-8f:.43.
Lumpkin, 0. J., and Zimm, B. H. 1982. Mobility of DNA in gel electrophoresis, Biopolymers
11:2315-2316.
Masurekar, N., Palmer, E., Ono, B-1., Wilhelm,]. M., and Sherman, F. 1981. Misreading of
the ribosomal suppressor SUP46 is due to an altered 40S subunit in yeast,]. Mol. Biol.
147:381-390.
McKee, E. E., and Poyton, R. 0. 1984. Mitochondrial gene expression in Saccharomyces
cerevisiae. I. Optimal conditions for protein synthesis in isolated mitochondria,]. Biol.
Chem. 259:9320-9331.
McKee, E. E., McEwen,J. E., and Poyton, R. 0. 1984. Mitochondrial gene expression in
Saccharomyces cerevisiae. II. Fidelity of translation in isolated mitochondria from wild
type and respiratory-deficient mutant cells,]. Biol. Chem. 259:9332-9336.
McLaughlin, C. S., Warner,]. R., Edmonds, M., Nakazato, H., and Vaughan, M. H. 1973.
Polyadenylic acid sequences in yeast messenger RNA,]. Biol. Chem. 248:1466-1471.
Miller, M.J., Xuong, N., and Geiduschek, E. P. 1979. A response of protein synthesis to a
temperature shift in the yeast Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. USA
76:5222-5225.
Mills, D. 1974. Isolation of polyribosomes from yeast during sporulation and vegative
growth, Appl. Microbial. 27:944-948.
Mitchell, D. J., and Bevan, E. A. 1987. dsRNA killer systems in yeast, in: Yeast Biotechnology
(D. R. Berry, I. Russell, and G. G. Stewart, eds.), Allen and Unwin, London, pp. 104-
155.
Moldave, K., and McLaughlin, C. S. 1988. The analysis of temperature-sensitive mutants
of Saccharomyces cerevisiae altered in components required for protein synthesis, in:
Genetics of Translation: NeU! Approaches (M. F. Tuite, M. Picard, and M. Bolotin-
Fukuhara, eds.), Springer-Verlag, Berlin, pp. 271-282.
Monier, R., Stephenson, M. L., and Zamechnik, P. C. 1960. The preparation and some
properties of a low molecular weight ribonucleic acid from· baker's yeast, Biochim.
Biophys. Acta 43:1-8.
Newlon, C. S., Devenish, R. J., Suci, P. A., and Roffi, C. J. 1981. Replication of small
chromosomal DNAs in yeast,ICN-UCLA Symp. Mol. Biol. 22:501-516.
Oliver, S. G., McCready, S. J., Holm, C., Sutherland, P., McLaughlin, C. S., and Cox, B. S.
1977. Biochemical and physiological studies of the yeast virus-like particle,]. Bacterial.
1!10:1303-1309.
Oppenheim, A. 1981. Separation of closed circular DNA from linear DNA by electropho-
resis in two dimensions in agarose gels, Nucleic Acids Res. 9:6805-6812.
Pelham, H. R. B., and jackson, R. j. 1976. An efficient mRNA-dependent translation
system from reticulocyte lysates, Eur.]. Biochem. 67:247-256.
Peppler, H. J. 1979. Production of yeasts and yeast products, in: Microbial Technology, 2nd
ed., Vol. 1 (H. J. Peppler and D. Perlman, eds.), Academic Press, New York, pp. 157-
185.
Pfisterer,]., and Buetow, D. E. 1981./n vitro reconstitution of the mitochondrial translation
system of yeast, Proc. Natl. Acad. Sci. USA 78:4917-4921.
Plesset, J., Foy, J. J., Chia, L. L., and McLaughlin, C. S. 1982. Heat shock in Saccharomyces
cerevisiae: Quantitation of transcriptional and translational effects, in: Transcrip-
tion/Translational Regulation if' Gene Expression (M. Grunberg-Manago and B. Safer,
eds.), Elsevier, New York, pp. 495-514.
Pringle, J. R. 1975. Methods for avoiding proteolytic artefacts in studies of enzymes and
other proteins from yeasts, Methods Cell Biol. 12:149-184.
Radloff, R., Bauer, W., and Vinograd, J. 1967. A dye-buoyant density method for the
detection and isolation of closed circular duplex DNA: The dosed circular DNA in
HeLa cells, Proc. Natl. Acad. Sci. USA 57:1514-1521.
Rete!, J., Van den 8os, R. C., and Planta, R. J. 1969. Characteristics of the methylation in
vivo of ribosomal RNA in yeast, Biochim. Biophys. Acta 195:370-380.
Rothblatt, J. A., and Meyer, D. I. 1986. Secretion in yeast: Reconstitution of the transloca-
tion and glycosylation of a-factor and invertase in a homologous cell-free system, Cell
44:619-628.
Rubin, G. M. 1975. Preparation of RNA and ribosomes from yeast, Methods Cell Biol.
12:45-64.
Sawadago, M., Sentenac, A., and Fromageot, P. 1981./n vitro transcription of cloned yeast
ribosomal DNA by yeast RNA polymerase A, Biochem. Biophys. Res. Commun. 101:250-
257.
Schantz, G. 1967. Stable phosphorylating submitochondrial particles from baker's yeast,
Methods Enz.ymol. 10:197-202.
Schindler, D., Grant, P. G., and Davies, J. E. 1974. Trichodermin resistance-mutation
affecting eukaryotic ribosomes, Nature 248:535-536.
Schmidt, G. 1968. Periodate oxidation of ribonucleic acids and their derivatives, Methods
Enzymol. 128:230-235.
Schwartz, D. C., and Cantor, C. R. 1984. Separation of yeast chromosome-sized DNAs by
pulse field gradient gel electrophoresis, Cell37:67-75.
Scopes, R. K. 1982. Protein Purification: Principles and Practice, Springer, New York.
Scott, J. H., and Schekman, R. 1980. Lyticase: Endoglucanase and protease activities that
act together in yeast cell lysis,]. Bacterial. 142:414-423.
Serwer, P., and Hayes, S. J. 1986. Exclusion of spheres by agarose gels during agarose gel
electrophoresis: Dependence on the sphere's radius and the gel's concentration, Anal.
Biochem. 158:72-78.
Shalitin, C., and Vishlizky, A. 1984. An improved isolation procedure for yeast two-mi-
crometer minichromosomes, Current Genet. 9: 107-111.
Shalitin, C., Pan, C. J., and Davie, J. R. 1983. Isolation of 2~~om minichromosomes from
Saccharomyces cerevisiae using shallow metrizamide gradients, Exp. Mycol. 7:175-181.
Sissons, C. H. 1978. Methods for yeast protein synthesis in a cell-free system, Methods Cell
Biol. 20:83-99.
Skogerson, L., and Wakatama, E. 1976. A ribosome-dependent GTPase from yeast distinct
from elongation factor 2, Proc. Natl. Acad. Sci. USA 75:73-76.
Sogin, S. J., and Saunders, C. A. 1980. Fluctuation in polyadenylate size and content in
exponential- and stationary-phase cells of Saccharomyces cerevisiae,]. Bacterial. 144:74-
81.
Sogin, S. J., Haber, J. E., and Halvorson, H. 0. 1972. Relationship between sporulation-
specific 20S ribonucleic acid and ribosomal ribonucleic acid processing in Sac-
charomyces cerevisiae,]. Bacterial. 112:806-814.
Sommer, A., and Lewis, M. J. 1971. Effect of dithiothreitol on yeast: Sphaeroplast forma-
tion and invertase release,]. Gen. Microbial. 68:327-335.
Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by
gel electrophoresis,]. Mol. Biol. 98:503-517.
Southern, E. M., Anand, R., and Fletcher, D. S. 1987. A model for the separation oflarge
DNA molecules by crossed field gel electrophoresis, Nucleic Acids Res. 15:5925-5943.
Specht, C. A., Di Russo, C. C., Novotny, C. P., and Ullrich, R. C. 1982. A method for
extracting high molecular weight DNA from fungi, Anal. Biochem. 119:158-163.
Sripati, C. E., and Warner, J. R. 1978. Isolation, characterisation and translation of mRNA
from yeast, Methods Cell Biol. 20:61-81.
Sripati, C. E., Groner, Y., and Warner,J. R. 1976. Methylated blocked 5'termini of yeast
mRNA, ]. Biol. Chem. 251:2898-2904.
Strathern, J. N., Newlon, C. S., Herskowitz, I., and Hicks, J. B. 1979. Isolation of a circular;
derivative of yeast chromosome. III. Implications for the mechanism of mating type
interconversion, Cell18:309-319.
Swanson, M. E., and Holland, M. J. 1983. RNA polymerase !-dependent selective tran-
scription of yeast ribosomal DNA,]. Biol. Chem. 258:3242-3250.
Szczesna, E., and Filipowicz, W. 1980. Faithful and efficient translation of viral and cellular
eukaryotic mRNAs in a cell-free S-27 extract of Saccharomyces cerevisiae, Biochem. Bi-
ophys. Res. Commun, 92:563-569.
Tuite, M. F., and Plesset, J. 1986. mRNA-dependent yeast cell-free translation systems:
Theory and practice, Yeast 2:35-52.
Tuite, M. F., Plesset, J., Moldave, K., and McLaughlin, C. S. 1980. Faithful and efficient
translation of homologous and heterologous mRNAs in an mRNA-dependent cell-
free system from Saccharomyces cerevisiae,]. Bioi. Chem. 255:8761-8766.
Tuite, M. F., Cox, B.S., and McLaughlin, C. S. 1981. An homologous in vitro assay for yeast
nonsense suppressors,]. Biol. Chem. 256:7298-7304.
Tuite, M. F.. Cox, B.S., and McLaughlin, C. S. 1983./nvitrononsense suppression in [psi+)
and [psi-) cell-free lysates of Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. USA
80:2824-2828.
Udem, S. A., and Warner,J. R. 1972. Ribosomal RNA synthesis in Saccharomyces cerevisiae,J.
Mol. Biol. 65:227-242.
Van den Bos, R. C., and Planta, R. J. 1971. Studies on the role of rapidly labelled 20S RNA
in the biosynthesis of ribosomal RNA in yeast, Biochim. Biophys. Acta 247:175-180.
Vazquez, D., and Jimenez, A. 1980. Antibiotic inhibitors of translation in eukaryotes, in:
Ribosomes: Structure, Function and Genetics (G. Chambliss, G. R. Graven, J. E. Davis, K.
Davis, L. Kahan, and M. Nomura, eds.), University Park Press, Baltimore, pp. 847-
869.
Villanueva, J. R., Gacto, M., and Sierva, S. M. 1973. Enzymatic composition of a lytic
system from Micromonospora chalcea, in: Yeast, Mold and Plant Protoplasts (J. R. Vil-
lanueva, I. Garcia-Aeha, S. Gascon, and F. Urburu, eds.), Acaemic Press, New York,
pp. 3-24.
Vollrath, D., and Davis, R. W. 1987. Resolution of DNA molecules greater than 5 mega-
bases by contour-damped homogeneous electric fields, Nucleic Acids Res. 15:7865-
7876.
Von der Haar, F. 1979. Purification of aminoacyl-tRNA synthetases, Methods Enzymol.

59:257-267.
Waters, M. G., and Blobel, G. 1986. Secretory protein translocation in a yeast cell-free
system can occur posttranslationally and require ATP hydrolysis, ]. Cell Biol.
10%:1543-1550.
Weeks, D.P., Beerman, N., and Griffith, 0. M. 1986. A small-scale five hour procedure for
isolating multiple samples of CsC1-purified DNA: Application to isolations from
mammalian, insect, higher plant, algal, yeast and bacterial sources, Anal. Biochem.
15%:376-385.
Wickner, R. B. 1975. Mutants of Saccharomyces cerevisiae that incorporate deoxythymidine
5'-monophosphate into DNA in vivo, Methods Cell Biol. 11:295-302.
Wickner, R. B., and Leibowitz, M. J. 1976. Chromosomal genes essential for replication of
a dsRNA plasmid of Saccharomyces cerevisiae: The killer character of yeast,]. Mol. Biol.
105:427-434.
Williamson, D. H., and Fennell, D. J. 1975. The use of fluorescent DNA-binding agent for
detecting and separating yeast mitochondrial DNA, Methods Cell Biol. 11:335-351.
Winston, F., Chumley, F., and Fink, G. R. 1983. Eviction and transplacement of mutant
genes in yeast, Methods Enzymol. 101:211-227.
Yang, W. K., and Novelli, G. D. 1971. Analysis ofisoaccepting tRNAs in mammalian tissues
and cells, Methods Enzymol. 20:44-55.
Index
a-factor, 9 Bud scars, 34

Actin, 6-7, 38-39 Budding, 36-38
gene, 7
dot, 39 Calcium alginate, 274
Agglutination, 33 Candida albicans, 312
Allelic rescue, 158-160 Carbohydrate
Alpha-factor, 9-12, 30, 44, 182, 303 metabolism, 60-66
Amino acids genes, 64-65
biosynthesis, 66-67 Carbon, 250
nitrogen source, 251-252 Carboxypeptidase Y, 15, 27, 201-202
pools, 288 Cardiolipin, 30
Aminoacyl-tRNA synthetases, 294 Cascade fermentation, 222-223
Amylase Cask conditioning, 222
barley alpha-, 217 Cassava, 238
beta-, 217 Catabolite conversion, 63-64
mouse alpha-, 188, 242 Catabolite inactivation, 61-62
Amyloglucosidase genes, 241 Catabolite repression, 60-61
Amylolytic enzymes, 238, 240-242 Cell counting, 271
Aneuploidy, 112-115 Cell cycle, 37, 80, 87-88
markers, 119-120 mutants, 44, 82-83, 115
Antibiotic resistance, 301 periodic synthesis, 87-88
ARS: see Autonomously replicating synchronization, 44-45, 88
sequence Cell envelope, 23-27
Ascus, 103 Cell separation techniques, 272
Aspergillus awamori, 241 Cell wall, 23-24, 26-27, 39
Aspergillus niger, 241 disruption, 40, 283-286
AUG context, 185-186 Centrifugal elutriation, 45
Autolysis, 43, 284 CentiMorgans (eM), 110
Autonomously replicating sequence, 151, Centromere
173 effect on recombination, 106-107
ARS-based plasmids, 151, 153 linkage, 110-111
ARS-CEN-based plasmids, 151 Chitin, 16, 26-27
Chitin synthase, 16, 34, 39
Bakers yeast, 214, 266 Chromosomes, 308
Baking, 213-214 Chromosomal DNA, preparation, 295-
Batch culture, 297
commercial, 261 Chromosome-loss mapping, 121
process, 258-261 Chymosin, expression in yeast, 174-175
vessels, 260-262 Clarification, 229-230
Beer brewing, 215-224 Clathrin, 30
process, 215-218 Clathrin triskelions, 30
Brettanomyces species, 222 Coated vesicles, 30
Braun homogenizer, 284 Codon bias, 182-184
Brewery, 220 Codon bias index, 183
321
322 INDEX
Competitive inhibition, 77 Endoglycosidase H, I5, I92

Complementation tests, I 05 Endoplasmic reticulum, I, 26, 28-29, 187
Concanavalin A, 28 (3-endorphin, 190
Continuous culture, 222, 262-270 Enolase, 63-64
Continuous stirred tank reactor, 270 Ergosterol, I8
Contour-clamped homogeneous electric Ethanol
field (CHEF), 311-312 downstream processing, 278
Core glycosylation, 19I-192 sensitive mutants, 232
Cotransformation, 165 tolerance, 23I.,..236
Coulter counter, 271 tolerance mutants, 232-235
Crabtree effect, 256-257 Ethylmethane sulphonate, 115
Crossing-over, 109-IIO Exponential phase, 259-260
Cryptopleurine, 7 Extrachromosomal genes, curing, I27-
Cyclic AMP, 34, 51, 73-74 I28
Cyclic AMP-dependent protein kinase, Extragenic revertants, 116
61-62, 103
Cyclic nucleotides, 73-74 Feedback control systems, 214
Cycloheximide, 7, 128, 289 Ficoll gradients, 299
Cytochalasin B, 6 Field inversion gel electrophoresis, 312-
Cytoduction, I27 313
Cytokinesis, 37 FIGE: see Field inversion gel
Cytosine-(3-o-arabinoside-5' -triphosphate, electrophonesis
305 Finings, 230
Flavor compounds, 219
DAPI, 296 Flocculation, 236-237
Dense body, 35 genetic control, 236-237
6-deoxy-o-glucose, 77 Fluidized bed fermenter, 275-276
4' ,6-diamidino-2-phenylindole: see DAPI 5-Fluoroorotic acid, 161
Diethyl pyrocarbonate, 296 French Press, 284
Dilution rate (D), 263-265 Fructose 1,6-biphosphatase, 6I
Dipeptidyl aminopeptide, I90 Fructose 2,6-biphosphate, 63
Disomy, 112-113 Fuel alcohol, 276
Ditype tetrad, I 08 Fuse! alcohols, 2I8
DNA
isotopic labeling, 71 G protein, 74
preparation, 295-298 G4I8 resistance, I65, 240
replication, 83 GAL promoter, 181
Dominant selectable marker, I71-172, Gap repair, 159-160
240 Gas mixed reactor, 275-276
Double-stranded RNA, 128, I32-I38, Gene
294-295 disruption, 160-163
exclusion phenomenon, 138 isolation, 152
preparation, 294-295 order, 117-118
Double crossover, 109-I10 transplacement, 161-165
Dough fermentation, 214 Gene map, 122-125
Downstream processing, 276-278 Genes
CDC25, 74
Eaton press, 284 cob-box, 130-121
Elongation factors, 8, 30I CUPI, 165, 17I-I72, 240
Embden-Meyerhof pathway, I CYCI, 185
Endocytosis, 29-30 CYH2, I64-I65
Endoglucanase genes, 242-243 CYRI, 62,74
Endo-(3-glucanase, 217 DEXJ, 241
INDEX 323
Genes (cont.) Glass bead disruption, 284, 302

FLOJ, 237 Glucans, 15, 26-27, 242
FL05, 237 Glucan synthetase, 16
FL08, 237 (3-Glucanase, 242-243
FLP, 152 Glucoamylase, 188, 238
GALl, 181 Glucose
GAL4, 181 transport, 78
GALlO, 187 uptake, 78
HO, 102 Glusulase, 150, 285
KRBJ, 122 Glycerophospholipids, 17-18
KEX2, 189-190 Glycogen, 15
LEU2, 149 Glycolysis
MAKJ, 136 "by-pass," 81-82
MAK8, 136 cell-free system, 86
MAT, 102 mutants, 81, 85
MFal, 182, 189 Glycolytic flux, 63
PET, 129 Glycoproteins
PFKl-5, 81 N-linked, 12-14
PH05, 181 0-Iinked, 12
PH080, 181 Glycosylation, 187, 191-192
POFJ, 239 Golgi, 28
RAMJ, 74 Grape fermentation, 228
RASJ, 74 Grits, 215
RAS2, 74 Growth
REPJ, 152, 196 acceleration phase, 258
REP2, 152, 196 carbon dioxide, 257-258
REP3, 196 dissolved oxygen, 256-257
RHOJ, 74 ethanol effects, 255-256
RH02, 74 factors, 253-254
ROAM, 237 monitoring, 271-272
SKJ, 136 process variables, 254
SNFJ, 61 substrate inhibition, 256-257
SNFJ, 78 temperature effects, 255-256
SRA1,74 Guanidine hydrochloride, 128
SRA3, 74 Guanidinium thiocyanate, 291
SRA4, 74
STAJ, 190,241
STE2, 12 Haze formation, 224
STE3, 12 Helicase, 40, 285
STE13, 189 Hemocytometer, 271
SUC2, 13, 191 Heparin, 300
SUP4, 156
Hepatitis B surface antigen, 184, 205
TPKJ, 73
Heterologous gene expression, 169-205
TPK2, 73
Hexokinase PII, 61
TPK3, 73
Hexose uptake, 78
TRPJ, 149
HOMOL 1 box, 8
TUBJ, 6
Hsps, 217
TUB2, 6
Human epidermal growth factor, 182, 184
TUB3, 6
Hydroxyurea, 71
URA3, 201
Genetic mapping Immobilized cell reactor, 272-273
methods, 116-123 Immobilized cell systems, 272-276
strategies, 122 active immobilization, 274-275
324 INDEX
Immobilized cell systems (cont.) Metabolic pathways, 243

fermenters, 275-276 Metabolic regulation, 80-88
passive immobilization, 273 Metabolism, genetics, 80-83
In vitro systems Methionine, 252, 289
DNA synthesis, 303-306 Methionine aminopeptidase, 192
transcription, 298-300 Methylbenzimidazole-2-y 1 carbamate,
translation, 300-303 121
Inositol, 253-254 Methylmethane sulphonate, 121
Insertion mutation, 161 Metrizamide gradient, 298
Integrating vectors (Yip), 121, 156-165 Micromanipulation, 104
Interferon, 174, 180-181, 199 2-Micron DNA plasmid, 138, 151-152,
Intracellular pH, 85 172-173,200-201,305
lntragenic revertants, 116 curing, 138
Introns, 8, 19, 22-23, 132, 173-174 mapping, 121-122
Invertase, 13, 15 segregation, 123, 126
lso-2-cytochrome C, 87 YEp plasmids, 151-152
Microsomes, 42, 302
Karyotype,35 Microtubules, 5-6, 38-39
Kex2 protease, 190 Minichromosome vectors, 154
Killer systems, 132-138, 294-295 Mistranslation, 204-205
Killer toxin, 136, 190 Mitochondria, 30-32, 36
Killer yeast, 222 genes, 129-132
Kilning, 215 genetic code, 130
Koji fermentation, 225-226 genome, 32, 129
isolation, 40-41
Lafrancois aeration system, 266-267 membranes, 31
Lag phase, 258-259 mRNA, 303
Lager production, 224 proteins, 289
Lariat RNA, 19 translation system, 32
Life cycle, 32-39, 101-105 Mitochondrial DNA
Linkage analysis, 108-111 curing, 128
Lipids, 18-19, 24-25 labeling, 288
Lipomyces konoenkode, 240 preparation, 297
Lucifer yellow, 29 Mitotic recombination, 118-119
Lytic enzymes, 285-286 Mitotic segregation, 123, 126
Lyticase, 40, 150, 284-285 Molasses, 214
Monosomy, 113
Maltose, 214, 250 Moromi, 226
Mannoproteins, 16-17, 27 Moto, 226
Mating, 33-34, 103 Mutagenesis, 115
Mating factors, 9-12, 33-34, 102 Mutant isolation, 115-116
Mating type Mutants
effects on regulation, 87 adel, 71
testing, 105 ade2, 72
Mean generation time, 260 barl, 12
Media, 138-140 bcyl,62, 73,74
Meiosis, 34-36, 105-106 bypl,82
Meiotic mapping, 116-117 can], 121
Membrane proteins, 24-25 cdc, 44, 83
Messenger RNA (mRNA) cdc6, 121
half life, 20-21, 290-291 cdc8, 304
preparation, 290-291 cdcl4, 121
splicing, 19-20 cdc35, 73
INDEX 325
Mutants (cont.) Oligo (dT)-cellulose, 291

chll, 121 Oligomycin, 129
CYC7-H2, 87 Organelles, isolation, 40-42
cyh2, 121 Origin of replication, I 72-173
cyrl, 62, 73 Orthogonal field alteration gel electropho-
cyr2, 73 resis (OFAGE), 310-3II
CYRJ, 73 2-oxoglutarate, 62
endl, 30 Oxygen, 62
end2, 30
foal, 81 Packed bead fermenter, 275
fsul, 237 Parental ditype (PD) tetrad, 108-111
glkl, 77 Partitioning, 195-196
hxkl' 77 Pasteur effect, 62-63, 86, 256
hxk2, 77 Pasteurization, 227
karl, 126-127 Periplasmic space, 187
kexl, 136 Permeabilized cells, 304
kex2, 136, 190, 204 Petite mutation, 30, 132
krel, 136 PGK promoter, 178-180
mak, 136 pH control, 254
mapl, 200 Phalloidin, 6
mnn, 13, 17 PH05 promoter, 181
pde, 73 Phosphate assimilation, 252
pep4-3, 201, 284 Phosphatidylinositol, 18
pfkl, 82 Phosphofructokinase, 63, 81-82
pfk2, 82 Phosphoglycerate kinase, 193-194
ppdl, 73 Phospholipids, I 8-19
rad52, 121, 198 Phosphorus, 252
rnal, 20 Phosphorylation, 192
sec, 29 Phrix aeration system, 266-267
ski, 130 Plasma membrane, 23-26, 41-42
snf, 78 Plasmid
spd, 85 copy number, 194
spoll, 120-121 DNA preparation, 297-298
sstl, 12 replication, 307
ste5, 73 stability, 195-201
tmpl, 72 structural rearrangements, 197-199
tupl, 72, 287 Plevato continuous fermentation system,
Myristilation, 192 269
Ploidy, 101
Poly(A) tail, 291
Nitrogen, 250-251 Poly(U) translation, 300
Nitrogen metabolism, 66-74 Polyadenylation, 179
genes for, 74-77 Polycomplex body, 36
Non-Mendelian genetics, 123-138 Polyphosphate, 15, 28
Non parental ditype (NPD), 108-111 Polysaccharide utilization, 237-238
Nonsense suppressor, 301 Posttranslational modification, 186-192
Nossal shaker, 284 Precursor pools, 79-80
Novozym, 40, 285 Prepro sequences, 174-176
Nuclear magnetic resonance, 78-79, 84- Pressed yeast, 277-278
85 Promoters, 176- I 79
Nucleotide metabolism, 70-74 regulated, 180-182, 199
Nucleus, 42 Prospore, 36
Nutritional requirements, 250-254 Proteases, 43, 67-70, 201-202
326 INDEX
Protease A, 27 Saccharomyces uvarum, 218, 220, 240, 249,

Protease inhibitors, 43-44, 284, 302 257,273
Protein disulfide isomerase, 191-192 Sake brewing, 224-227
Protoplast fusion, 237, 239-240 Sake yeast, 226-227
Protoplast lysis, 284-288 Scholler and Seidel aeration system, 266-
Proton-translocating ATPase, 28 267
[psi] factor, 128 Schwanniomyces alluvius, 240
Pullanase, 217 Secreted enzymes, 187
Pulse-chase experiments, 80 Secretion, 28-29, 186-187
Pulse-labeling, 79, 288-289 Segregation
Pulsed-field gel electrophoresis, 307-314 2:2, 105, 120, 123
Pulsed-field gradient gel electrophoresis, 4:0, 128, 138
309-310 aneuploid, 114, 120
Selectable markers, 165, 171-172
Radioactive labeling Sheet reactor fermenter, 275
DNA, 287-288 Signal peptidase, 191
proteins, 288-289 Signal recognition particle (SRP), 29, 303
RNA, 287 Signal sequences, 187-191
Rare mating, 238-239 Single-stranded DNA plasmids, 153-154
RAS proteins, 103-104 Small nuclear RNAs, 20
Replication bubbles, 305-306 Specific growth rate (fl.), 259-260
Replisome, 305 Spheroplasts, 150
Reptation, 308-309 Sphingolipids, 18
Respiration, 256 Spindle pole bodies, 34-36, 38
Respiratory-deficient mutants: see Petite Spliceosomes, 20
mutation spoll mapping, 120-122
Ribosomal proteins, 7-9, 20 Sporulation, 35-26, 73, 88, 103
Ribosomal RNAs; 21-22, 294 mating type control, 88
processing, 21-22 medium, 140
Ribosomes, 7 metabolic changes, 83-85
Rice wine, 224 Stationary phase, 260
Rich medium, 138-139 Stirred tank reactor, 260
Ring chromosome, 298 Strain development, 231
RNA maturase, 132 Subcellular fractionation, 31,43
RNA methylation, 287 Submerged culture, 258
RNA polymerase Subunit vaccines, 205
1,298-299 Sulfur, 252
II, 177, 299-300 Super-secretion mutants, 204
III, 299-300 Suppressor tRNAs, 22,116
RNA preparation, 290-295 Synaptonemal complex, 35
chaotropic disruption method, 291 Synchrony methods, 306
differential extraction techniques, 290 Synthetic complete medium, 139-140
mRNA, 290-291 Synthetic minimal medium, 139
tRNA, 291-294
Rotofermenter, 270 Targeted integration, 157-158
RPG box, 8 TATA box, 176
Tetrad analysis, 104-112, 140-144
Saccharomyces bayanus, 229 Tetratype tetrad, 108-111
Saccharomyces carlsbergensis, 218 Thymidine kinase, 287
Saccharomyces diastaticus, 239, 240, 241, Thymidylate kinase, 304
270 Tonoplast, 28
Saccharomyces fragi, 270 Tower fermenter, 221-222, 269-270
Saccharomyces sake, 229 Trace elements, 253
INDEX 327
Transcription UAS: see Upstream activation sites

factor P37, 298 Unsaturated free fatty acids, 18
fusion, 178 5' Untranslated mRNA leader, 184-186
termination, 179-180 Upstream activation sites (UAS), 61, 176
Transfer RNA (tRNA), 22-23, 291-294 [UREJ] factor, 128
charging, 292-294 Urea, 251
splicing, 23
suppressors, 22, 116 Vacuoles, 27-28
Transformation membranes, 28
lithium method, 150 preparation, 42
spheroplast method, 150 Vectors, 150-152
Translation Vertical pulsed-field gradient gel electro-
fusion, 178 phoresis, 312
initiation, 184-186 Virus-like particles
Translational upstream factor (TUF), 8-9 killer, 132, 135, 294
Translocation, 29 Tyl, 205
in vitro, 302-303 Vitamins, 254
Transport, 74, 77-80 Voge1busch aeration system, 266-267
Transposon
Tn601, 165 Wheat flour, 214-215
Tyl, 87-103 Wine making, 228-231
Trehalose, 84 red wine, 230-231
Trichloroacetic acid (TCA) cycle, 62, 85 white wine, 229-230
Trichodermin, 7 Wort, 217-218
Trifluoromethane sulfonic acid, 192 fermentation, 218
Trisomy, 112-112
Tryptophan biosynthesis, 243 X-rays, 121
Tubulin, 5-6, 38
genes, 6 YAC vectors, 154-156
Tunicamycin, 192 Yeast extract, 278
Two dimensional gel electrophoresis Yeast Genetic Stock Centre (YGSC), 123
DNA replication, 306-307 Yield co-efficient (Y), 264
proteins, 289
Ty virus-like particle, 205 Zymolyase, 40, 150, 285-286

Brewing Yeast and Fer

Uploaded by

Copyright:

Available Formats

Brewing Yeast and Fer

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Brewing Yeast and Fer

Uploaded by

Copyright:

Available Formats

Saccharomyces

Volume 1 PENICILLIUM ANDACREMONIUM

Springer Science+Business Media, LLC

Saccharo•yces 1 ed1ted by M1chae1 F. Tu1te and Stephen G. 011var.

© 1991 Springer Science+Business Media New York

J. R. Dickinson • School of Pure and Applied Biology, University of

In this volume we aim to present an easy-to-read account of the genus

Michael F. Tuite and Stephen G. Oliver

C. Kreutzfeldt and W. Witt

6.1. Imposition of a Cell Cycle Block . . . . . . . . . . . . . . . . . . . 44

Metabolism and Biosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

Methods in Classical Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

5.1. Mutagenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

Recombinant DNA Techniques 149

A. J. Kingsman, E. J. Mellor, M. J. Dobson, and S.M. Kingsman

Expression of Heterologous Genes . . . . . . . . . . . . . . . . . . . . . . . . . . 169

"Classical" Yeast Biotechnology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213

5.2. Red Wine Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230

Culture Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

T. M. Matthews and C. Webb

7.1. Passive Immobilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272

Biochemical Techniques 283

7.2. Pulsed Field Gradient Gel Electrophoresis . . . . . . . . . . 309

Index ................................................... 321

Fungi have been exploited by mankind for many thousaJ;tds of years,

MICHAEL F. TUITE e Biological Laboratory, University of Kent, Canterbury, Kent

Figure 1. The exploitation of yeast metabolism.

To allow rational genetic manipulation of the biosynthetic capabilit-

Saccharomyces cerevisiae is a unicellular eukaryote in which a combination

2. BASIC MACROMOLECULES IN SACCHAROMYCES

2.1. Proteins and Peptides

C. KREUTZFELDT ·• Phase GmbH, D-2410, Molin, Federal Republic of Germany.

the critical polymerization concentration. This effect is rapidly reversed

2.1.3. Ribosomal Proteins

been found to be conferred by altered ribosomal proteins L3, L29, and

2.1.4. Mating Factors

sa spt at sp2 a 2 t24 sp3

medium is in the range from I0- 7 to I0-8 M. Investigations using mut-

transferred to a target asparagine residue of a protein chain inserted

In contrast to invertase, yeast carboxypeptidase Y, a vacuolar glyco-

2.2. Polysaccharides, Structural Mannoprotein, and Polyphosphate

in the group of the polar lipids, comprising phosphatidylcholine, -eth-

structure, function, and biogenesis. Examples of recent work in this field

2.4. Ribonucleic Acids

Table I. Half-Lives of mRNAs Coding

Ribosomal protein Half-life time (min)

• Whenever possible, the ribosomal proteins are des-

(1976) have measured the half-lives of mRNAs of several ribosomal

nucleotides in length and, in contrast to mRNA introns, do not have

Yeast cells contain a set of subcellular compartments typical of eu-

3.1. Cell Envelope

rulation. The carbohydrate moieties of mannoprotein on its surface de-

3.1.1. Plasma Membrane

for studying the properties and functions of 'lipids in eukaryotic cell

Table II. Enzymes Associated with the Plasma Membrane of S. cerevisiae

Analytical procedure Reference

Chitin synthetase Purification by entrap- Kang et al., 1984

3.1.2. Cell Wall

regarded as the main structural determinant component of the wall