You are on page 1of 696

Free ebooks ==> www.ebook777.


Physiology and
of Prokaryotes

David White
Indiana University

James Drummond
Indiana University

Clay Fuqua
Indiana University

New York Oxford

Free ebooks ==>

Oxford University Press, Inc., publishes works that further Oxford University’s
objective of excellence in research, scholarship, and education.

Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto

With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright © 2012, 2007, 2000, 1995 by Oxford University Press, Inc.

For titles covered by Section 112 of the US Higher Education Opportunity Act,
please visit for the latest information about pricing and
alternate formats.

Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York 10016

Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data

White, David, 1936-

The physiology and biochemistry of prokaryotes / by David White,
James T. Drummond, Clay Fuqua. — 4th ed.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-0-19-539304-0 (alk. paper)
I. Drummond, James T. II. Fuqua, Clay. III. Title.
[DNLM: 1. Bacteria—metabolism. 2. Archaea—physiology.
3. Prokaryotic Cells—physiology. QW 52]


Printing number: 9 8 7 6 5 4 3 2 1

Printed in the United States of America

on acid-free paper
Free ebooks ==>


Introduction physiology of specific groups of prokaryotes is

The fourth edition of The Physiology and emphasized. This pattern of organization lends
Biochemistry of Prokaryotes, designed for use itself to the elucidation of general principles of
in advanced undergraduate and beginning physiology, metabolism, responses to environ-
graduate-level biology courses, provides the mental challenges, and cellular/multicellular
most current, authoritative, and relevant presen- development.
tation of prokaryotic physiology and biochem- Topics include cellular structure and func-
istry. It presents microbial metabolism in the tion, growth and cell division, chromosome
context of the chemical and physical problems replication and partitioning of chromosomes,
that cells must solve in order to grow. The text membrane bioenergetics and the proton poten-
is organized by topic rather than by organism, tial, electron transport , photosynthesis, the reg-
therefore helping students understand the gen- ulation of metabolic pathways, bioenergetics in
eral principles of physiology and metabolism. the cytosol, central metabolic pathways, RNA
This new edition builds in comprehensive cov- and protein synthesis, cell wall and capsule
erage of energetics. It also adds broad coverage biosynthesis, inorganic metabolism, C1 metab-
of molecular machinery, applied throughout the olism, fermentations, responses to environmen-
text to help create a unifying narrative across tal stress, solute transport, protein transport
biological principles. Also added is broader cov- and secretion, responses to environmental cues,
erage of chromosomes, macromolecular synthe- chemotaxis, photoresponses, aerotaxis, micro-
sis, biofilms, and cell–cell communications. bial biofilms, cell–cell communication mecha-
The prokaryotes are a diverse assemblage of nisms, and bacterial development.
organisms that consists of the Bacteria (also called
eubacteria) and the Archaea (also called archae- Distinctive Features
bacteria). This text provides an updated descrip- • Topical organization fosters understanding
tion of the major aspects of the prokaryotes, such of the general principles and concepts of the
as cell structure, biochemistry, bacterial devel- biochemistry and physiology of prokaryotes,
opment, adaptation to environmental changes, in addition to providing detailed data and con-
and signaling interactions between the cells that clusions about specific groups of prokaryotes.
occur in bacterial populations such as those living • End-of-chapter summaries and study ques-
in biofilms. The text highlights signaling mecha- tions help students synthesize material and
nisms that allow individual bacterial cells to sense prepare for examinations.
and respond to the environment, and also to sig- • An extensive references and notes section
nal each other so that they can respond as a coop- is available in the chapters to aid further
erating population of organisms. research and to provide access to the data that
supports the conclusions made in the text.
Organization • Boxed sections call out topics of special
The organization of the text is according to interest, adding historical information about
topics rather than organisms, although the earlier discoveries, covering experiments
Free ebooks ==>
xvi preface

that established the central tenets of micro- Acknowledgments

biology and biochemistry, or exploring in
We would like to express gratitude to the many
detail especially engaging or important top-
individuals named below, and those who remain
ics discussed in the text.
anonymous, who have read sample chapters and
have made helpful suggestions for the fourth
New to the Fourth Edition edition.
• New coauthors: David White is joined by Jennifer Anthony, University of the Sciences in
two colleagues from Indiana University: Jim Philadelphia
Drummond adds expertise on biological Theodore C. Crusberg, Worcester Polytechnic
macromolecules, while Clay Fuqua contrib- Institute
utes authority on microbial interaction. Both John E. Gustafson, New Mexico State University
are well-known authorities in their respec- Shannon Hinsa-Leasure, Grinnell College
tive fields, as well as experienced educators: Adam J. Houlihan, Wagner College
• New chapters: The fourth edition adds a Carol R. Lauzon, California State University–
more detailed account of chromosome rep- East Bay
lication, protein and RNA synthesis, and a Paul W. Lepp, Minot State University
more complete description of the biology of Robert J. C. McLean, Texas State University–
biofilms and of intercellular communication San Marcos
between bacteria: Tina Salmassi, California State University–Los
Chapter 11, RNA and Protein Synthesis Angeles
Chapter 21, Microbial Biofilms—Structured Kathleen Scott, University of South Florida
Multicellular Assemblies Timothy Secott, Minnesota State University
Chapter 22, Cell–Cell Communication Louis Sherman, Purdue University
Mechanisms Teri Shors, University of Wisconsin–Oshkosh
All chapters from the previous edition have been Om V. Singh, University of Pittsburgh
thoroughly reviewed and revised to incorporate John G. Steiert, Missouri State University
the most recent research. Ann M. Stevens, Virginia Tech
Sonia M. Tiquia, University of Michigan–
• New themes: Two new themes have been
incorporated across the text. A comprehen-
Thomas M Walter, Purdue University
sive coverage of energetics adds another per-
Hwan Youn, California State University–Fresno
spective to the physiological and biochemical
topics covered. A thorough incorporation of We also thank the reviewers of the third edition:
molecular machinery helps create a unifying Carl Bauer, Yves Brun, Jim Drummond, Martin
narrative across biological principles. Dworkin, Pat Foster, Heidi Kaplan, Larry
Shimkets, and Ashley Williams, as well as the
A note on chemical notation of acidic and basic
many individuals who helped by reviewing
groups: Most of the carboxyl groups are drawn
the first two editions. Thanks also to the team
as nonionized and the primary amino groups as
at Oxford University Press, including Jason
nonprotonated. However, at physiological pH
Noe, senior editor for the life sciences; Katie
these groups are ionized and protonated, respec-
Naughton and Caitlin Kleinschmidt, editorial
tively. The names of the organic acids indicate
assistants; Jason Kramer, marketing manager;
that they are ionized (e.g., acetate rather than
Frank Mortimer; director of marketing; Patrick
acetic acid).
Lynch, editorial director; and John Challice, vice
president and publisher. The excellent efforts of
Ancillary Materials the Oxford University Press production team
• Instructor’s Resource Companion Website: are gratefully acknowledged: David Bradley, All images are production editor; Steven Cestaro, production
available to instructors for classroom director; Lisa Grzan, production team leader;
use on a password-protected instructor’s Betty Lew, art director; and Brenda Griffing,
website. Please contact your Oxford sales copy editor. Much of this edition, like the first
representative for access. three editions, was illustrated by Eric J. White.
Free ebooks ==>


Boxed Material xiii

Preface xv
Symbols xvii
Conversion Factors, Equations, and Units of Energy xix
Definitions xxi

Chapter 1. Structure and Function 1

Chapter 2. Growth and Cell Division 55

Chapter 3. Chromosome Replication and Partitioning of Chromosomes 77

Chapter 4. Membrane Bioenergetics: The Proton Potential 111

Chapter 5. Electron Transport 146

Chapter 6. Photosynthesis 175

Chapter 7. The Regulation of Metabolic Pathways 199

Chapter 8. Bioenergetics in the Cytosol 207

Chapter 9. Central Metabolic Pathways 222

Chapter 10. Metabolism of Lipids, Nucleotides, Amino Acids, and Hydrocarbons 255

Chapter 11. RNA and Protein Synthesis 281

Chapter 12. Cell Wall and Capsule Biosynthesis 316

Chapter 13. Inorganic Metabolism 335

Chapter 14. C1 Metabolism 358

Chapter 15. Fermentations 383

Chapter 16. Responses to Environmental Stress 403

Chapter 17. Solute Transport 432

Free ebooks ==>
vi brief contents

Chapter 18. Protein Transport and Secretion 452

Chapter 19. Responses to Environmental Cues 482

Chapter 20. Chemotaxis, Photoresponses, Aerotaxis 534

Chapter 21. Microbial Biofilms—Structured Multicellular Assemblies 551

Chapter 22. Cell–Cell Communication Mechanisms 566

Chapter 23. Bacterial Development 587

Index 613
Free ebooks ==>


Boxed Material xiii

Preface xv
Symbols xvii
Conversion Factors, Equations, and Units of Energy xix
Definitions xxi

Chapter 1. Structure and Function 1

1.1 Phylogeny 3
1.2 Cell Structure 6
1.3 Summary 44
Study Questions 46
Reference and Notes 47

Chapter 2. Growth and Cell Division 55

2.1 Measurement of Growth 55
2.2 Growth Physiology 57
2.3 Growth Yields 65
2.4 Growth Kinetics 66
2.5 Steady State Growth and Continuous Growth 68
2.6 Cell Division 69
2.7 Summary 71
Study Questions 72
References and Notes 72

Chapter 3. Chromosome Replication and Partitioning of Chromosomes 77

3.1 DNA Replication, Chromosome Separation, and Chromosome Partitioning 77
3.2 Summary 103
Study Questions 104
References and Notes 104

Chapter 4. Membrane Bioenergetics: The Proton Potential 111

4.1 The Chemiosmotic Theory 111
4.2 Electrochemical Energy 112
4.3 The Contributions of the DΨ and the DpH to the Overall Dp in Neutrophiles,
Acidophiles, and Alkaliphiles 118
4.4 Ionophores 119
4.5 Measurement of the Dp 120
4.6 Use of the Dp to Do Work 122
4.7 Exergonic Reactions That Generate a Dp 124
4.8 Other Mechanisms for Creating a DΨ or a Dp 129

Free ebooks ==>
viii contents

4.9 Halorhodopsin, a Light-Driven Chloride Pump 137

4.10 The Dp and ATP Synthesis in Alkaliphiles 138
4.11 Summary 139
Study Questions 140
References and Notes 141

Chapter 5. Electron Transport 146

5.1 Aerobic and Anaerobic Respiration 147
5.2 The Electron Carriers 147
5.3 Organization of the Electron Carriers in Mitochondria 152
5.4 Organization of the Electron Carriers in Bacteria 152
5.5 Coupling Sites 156
5.6 How a Proton Potential Might Be Created at the Coupling Sites:
Q Loops, Q Cycles, and Proton Pumps 159
5.7 Patterns of Electron Flow in Individual Bacterial Species 163
5.8 Summary 170
Study Questions 171
References and Notes 171

Chapter 6. Photosynthesis 175

6.1 The Phototrophic Prokaryotes 175
6.2 The Purple Photosynthetic Bacteria 178
6.3 The Green Sulfur Bacteria (Chlorobiaceae) 183
6.4 Cyanobacteria and Chloroplasts 184
6.5 Efficiency of Photosynthesis 186
6.6 Photosynthetic Pigments 187
6.7 The Transfer of Energy from the Light-Harvesting Pigments
to the Reaction Center 193
6.8 The Structure of Photosynthetic Membranes in Bacteria 194
6.9 Summary 194
Study Questions 196
References and Notes 196

Chapter 7. The Regulation of Metabolic Pathways 199

7.1 Patterns of Regulation of Metabolic Pathways 199
7.2 Kinetics of Regulatory and Nonregulatory Enzymes 201
7.3 Conformational Changes in Regulatory Enzymes 204
7.4 Regulation by Covalent Modification 204
7.5 Summary 205
Study Questions 205
References and Notes 206

Chapter 8. Bioenergetics in the Cytosol 207

8.1 High-Energy Molecules and Group Transfer Potential 207
8.2 The Central Role of Group Transfer Reactions in Biosynthesis 213
8.3 ATP Synthesis by Substrate-Level Phosphorylation 215
8.4 Summary 220
Study Questions 221
References and Notes 221

Chapter 9. Central Metabolic Pathways 222

9.1 Glycolysis 224
9.2 The Fate of NADH 230
9.3 Why Write NAD+ instead of NAD, and NADH instead of NADH2? 230
9.4 A Modified EMP Pathway in the Hyperthermophilic Archaeon
Pyrococcus furiosus 231
9.5 The Pentose Phosphate Pathway 231
Free ebooks ==>
contents ix

9.6 The Entner–Doudoroff Pathway 236

9.7 The Oxidation of Pyruvate to Acetyl–CoA:
The Pyruvate Dehydrogenase Reaction 238
9.8 The Citric Acid Cycle 241
9.9 Carboxylations That Replenish Oxaloacetate: The Pyruvate and
Phosphoenolpyruvate Carboxylases 245
9.10 Modification of the Citric Acid Cycle into a Reductive (Incomplete)
Cycle during Fermentative Growth 246
9.11 Chemistry of Some of the Reactions in the Citric Acid Cycle 247
9.12 The Glyoxylate Cycle 248
9.13 Formation of Phosphoenolpyruvate 249
9.14 Formation of Pyruvate from Malate 251
9.15 Summary of the Relationships between the Pathways 251
9.16 Summary 252
Study Questions 253
References and Notes 254

Chapter 10. Metabolism of Lipids, Nucleotides, Amino Acids, and Hydrocarbons 255
10.1 Lipids 255
10.2 Nucleotides 264
10.3 Amino Acids 267
10.4 Aliphatic Hydrocarbons 273
10.5 Summary 275
Study Questions 278
References and Notes 279

Chapter 11. RNA and Protein Synthesis 281

11.1 RNA Synthesis 282
11.2 Protein Synthesis 296
11.3 Summary 310
Study Questions 311
References and Notes 312

Chapter 12. Cell Wall and Capsule Biosynthesis 316

12.1 Peptidoglycan 316
12.2 Lipopolysaccharide 321
12.3 Extracellular Polysaccharide Synthesis and
Export in Gram-Negative Bacteria 326
12.4 Levan and Dextran Synthesis 331
12.5 Glycogen Synthesis 332
12.6 Summary 332
Study Questions 332
References and Notes 333

Chapter 13. Inorganic Metabolism 335

13.1 Assimilation of Nitrate and Sulfate 335
13.2 Dissimilation of Nitrate and Sulfate 337
13.3 Nitrogen Fixation 339
13.4 Lithotrophy 344
13.5 Summary 353
Study Questions 354
Reference and Notes 355

Chapter 14. C1 Metabolism 358

14.1 Carbon Dioxide Fixation Systems 358
14.2 Growth on C1 Compounds Other than CO2: The Methylotrophs 374
14.3 Summary 378
Free ebooks ==>
x contents

Study Questions 380

References and Notes 380

Chapter 15. Fermentations 383

15.1 Oxygen Toxicity 383
15.2 Energy Conservation by Anaerobic Bacteria 384
15.3 Electron Sinks 385
15.4 The Anaerobic Food Chain 386
15.5 How to Balance a Fermentation 387
15.6 Propionate Fermentation via the Acrylate Pathway 388
15.7 Propionate Fermentation via the Succinate–Propionate Pathway 389
15.8 Acetate Fermentation ( Acetogenesis) 391
15.9 Lactate Fermentation 392
15.10 Mixed-Acid and Butanediol Fermentation 394
15.11 Butyrate Fermentation 397
15.12 Ruminococcus albus 400
15.13 Summary 400
Study Questions 401
References and Notes 402

Chapter 16. Responses to Environmental Stress 403

16.1 Maintaining a DpH 403
16.2 Osmotic Pressure and Osmotic Potential 406
16.3 Heat-Shock Response (HSR) 412
16.4 Repairing Damaged DNA 415
16.5 The SOS Response 421
16.6 Oxidative Stress 423
16.7 Summary 425
Study Questions 427
References and Notes 427

Chapter 17. Solute Transport 432

17.1 The Use of Proteoliposomes to Study Solute Transport 432
17.2 Kinetics of Solute Uptake 433
17.3 Energy-Dependent Transport 434
17.4 How to Determine the Source of Energy for Transport 444
17.5 Drug-Export Systems 445
17.6 Bacterial Transport Systems in Summary 446
17.7 Summary 446
Study Questions 448
References and Notes 448

Chapter 18. Protein Transport and Secretion 452

18.1 The Sec System 453
18.2 The Translocation of Membrane-Bound Proteins 457
18.3 The E. coli SRP 459
18.4 Protein Translocation of Folded Proteins: The Tat System 459
18.5 Extracellular Protein Secretion 461
18.6 Folding of Periplasmic Proteins 473
18.7 Summary 474
Study Questions 474
References and Notes 475

Chapter 19. Responses to Environmental Cues 482

19.1 Introduction to Two-Component Signaling Systems 483
Free ebooks ==>
contents xi

19.2 Responses by Facultative Anaerobes to Anaerobiosis 488

19.3 Response to Nitrate and Nitrite: The Nar Regulatory System 494
19.4 Response to Nitrogen Supply: The Ntr Regulon 498
19.5 Response to Inorganic Phosphate Supply: The PHO Regulon 503
19.6 Effect of Oxygen and Light on the Expression of Photosynthetic Genes
in the Purple Photosynthetic Bacterium Rhodobacter capsulatus 504
19.7 Response to Osmotic Pressure and Temperature:
Regulation of Porin Synthesis 506
19.8 Response to Potassium Ion and External Osmolarity:
Stimulation of Transcription of the kdpABC Operon by a Two-Component
Regulatory System 507
19.9 Acetyl Phosphate Is a Possible Global Signal in Certain
Two-component Systems 508
19.10 Response to Carbon Sources: Catabolite Repression, Inducer Expulsion,
and Permease Synthesis 510
19.11 Virulence Factors: Synthesis in Response to Temperature, pH,
Nutrient Osmolarity, and Quorum Sensors 515
19.12 Summary 522
Study Questions 524
References and Notes 524

Chapter 20. Chemotaxis, Photoresponses, Aerotaxis 534

20.1 Bacteria Measure Changes in Concentration over Time 534
20.2 Tumbling 535
20.3 Adaptation 536
20.4 Proteins Required for Chemotaxis 536
20.5 A Model for Chemotaxis 537
20.6 Mechanism of Repellent Action 541
20.7 Chemotaxis That Does Not Use MCPs: The Phosphotransferase System
Is Involved in Chemotaxis toward PTS Sugars 541
20.8 Chemotaxis That Is Not Identical with the Model Proposed for
the Enteric Bacteria 541
20.9 Photoresponses 543
20.10 Halobacteria 544
20.11 Photosynthetic Bacteria 544
20.12 Aerotaxis 546
20.13 Summary 546
Study Questions 546
References and Notes 547

Chapter 21. Microbial Biofilms—Structured Multicellular Assemblies 551

21.1 Bacterial Multicellular Structures 551
21.2 Prevalence and Importance of Biofilms 552
21.3 Properties of Biofilms 554
21.4 Progression of Biofilm Formation and Dissolution 557
21.5 Regulation of Biofilm Formation 559
21.6 Inhibition of Biofilm Formation 560
21.7 Evolutionary Processes in Biofilms 561
21.8 Summary 562
Study Questions 563
References and Notes 563

Chapter 22. Cell–Cell Communication Mechanisms 566

22.1 Diversity of Diffusible Signal Molecules Produced by Bacteria 566
22.2 Specific Signaling Systems 566
22.3 Cell–Cell Signaling That Requires Contact 581
Free ebooks ==>
xii contents

22.4 Summary 583

Study Questions 583
References and Notes 583

Chapter 23. Bacterial Development 587

23.1 Myxobacteria 587
23.2 Caulobacter Development: Control of DNA Replication and Cell Cycle Genes 598
23.3 Sporulation in Bacillus subtilis 601
23.4 Summary 610
Study Questions 610
References and Note 611

Index 613
Free ebooks ==>

Structure and Function

The prokaryote (procaryote) domains of life are movement on solid surfaces via a form of motil-
the Bacteria and the Archaea. Prokaryotes are ity called twitching, for gliding motility among
defined as organisms that have no membrane- the myxobaceria, and for mating. Flagella (sing.
bound nucleus. (The prefix pro, borrowed from flagellum) are used by single cells to swim in liq-
Greek pro, means earlier than or before; kary- uid; they are also used in swarming, a form of
ote, borrowed from Greek káryon, means kernel group swimming on moist solid surfaces.
or nut.) Besides lacking a nucleus, prokaryotes In addition, it turns out that the poles of
are devoid of organelles such as mitochondria, nonspherical bacteria cells are physiologically
chloroplasts, and Golgi vesicles. However, as and structurally different from the rest of the
you will learn from reading this chapter, their cell, and sometimes the cell poles in the same
cell structure is far from simple and reflects the cell differ from each other. This is obvious, for
evolution of prokaryotes into quite sophisti- example, when flagella or pili protrude from
cated organisms that are remarkably successful one or both cell poles, or when a cell pole is dis-
in inhabiting diverse ecological niches. tinguished by having a stalk, as is the case for
Despite the absence of organelles comparable Caulobacter, discussed in Chapter 23. In other
to those in eukaryotic cells, the activities within cases, the differences between poles are less
prokaryotic cells are compartmentalized. For obvious, as shown, for example, in the discus-
example, as this chapter points out, compart- sion of the polar localization of the chemotaxis
mentalization occurs within multienzyme gran- proteins in Chapter 20.
ules that house enzymes for specific metabolic It has become clear that prokaryotes even
pathways, in intracellular membranes within have an internal protein cytoskeleton, a prop-
the cytosol, within the cell membrane, within erty previously thought to be restricted to
a special compartment called the periplasm eukaryotic cells. As described in this chapter
in gram-negative bacteria, within the cell wall and in Chapter 2, the cytoskeleton is critical
itself, and within various inclusion bodies that for determining and maintaining cell shape as
house specific enzymes, storage products, or well as for cell division. For a general review of
photosynthetic pigments. As this chapter also subjects covered in this chapter, the reader is
points out, the prokaryotes also have special- referred to ref. 1.
ized external structures called appendages. Thus, as has been pointed out many times in
The most notable appendages are pili and fla- recent years, the prokaryotic cell is not simply
gella. Different types of pili (sing., pilus) serve a “bag of enzymes,” in accordance with ear-
different functions. Depending upon the type, lier descriptions, but rather, a sophisticated
pili are used for adhesion to other cells when entity, that is dynamic both structurally and
that becomes necessary for colonization, for physiologically. Furthermore, as Chapter 22

Free ebooks ==>
2 the physiology and biochemistry of prokaryotes

Table 1.1 Major subdivisions of prokaryotes

Bacteria and their subdivisions B. Flavobacterium group

Purple bacteria (now referred to as the division or Flavobacterium, Cytophaga, Saprospira,
phylum Proteobacteria) Flexibacter
α subdivision
Purple nonsulfur bacteria (Rhodobacter, Planctomyces and relatives
Rhodopseudomonas), rhizobacteria, agrobacteria, A. Planctomyces group
rickettsiae, Nitrobacter, Thiobacillus (some), Planctomyces, Pasteuria
Azospirillum, Caulobacter B. Thermophiles
β subdivision Isocystis pallida
Rhodocyclus (some), Thiobacillus (some), Chlamydiae
Alcaligenes, Bordetella, Spirillum, Nitrosovibrio, Chlamydia psittaci, C. trachomatis
γ subdivision Radio-resistant micrococci and relatives
Enterics (Acinetobacter, Erwinia, Escherichia, A. Deinococcus group
Klebsiella, Salmonella, Serratia, Shigella, Deinococcus radiodurans
Yersinia), vibrios, fluorescent pseudomonads, B. Thermophiles
purple sulfur bacteria, Legionella (some), Thermus aquaticus
Azotobacter, Beggiatoa, Thiobacillus (some),
Photobacterium, Xanthomonas Green nonsulfur bacteria and relatives
δ subdivision A. Chloroflexus group
Sulfur and sulfate reducers (Desulfovibrio), Chloroflexus, Herpetosiphon
myxobacteria, bdellovibrios B. Thermomicrobium group
Thermomicrobium roseum
Gram-positive eubacteria
A. High (G + C) species Archaea subdivisions
Actinomyces, Streptomyces, Actinoplanes, Extreme halophiles
Arthrobacter, Micrococcus, Bifidobacterium, Halobacterium, Halococcus morrhuae
Frankia, Mycobacterium, Corynebacterium
B. Low (G + C) species Methanobacter group
Clostridium, Bacillus, Staphylococcus, Methanobacterium, Methanobrevibacter,
Streptococcus, mycoplasmas, lactic acid bacteria Methanosphaera stadtmaniae,
C. Photosynthetic species Methanothermus fervidus
Heliobacterium Methanococcus group
D. Species with gram-negative walls Methanococcus
Megasphaera, Sporomusa
“Methanosarcina” group
Cyanobacteria and chloroplasts Methanosarcina barkeri, Methanococcoides
Oscillatoria, Nostoc, Synechococcus, Prochloron, methylutens, Methanothrix soehngenii
Anabaena, Anacystis, Calothrix
Methospirillum group
Spirochaetes and relatives Methanospirillum hungatei, Methanomicrobium,
A. Spirochaetes Methanogenium, Methanoplanus limicola
Spirochaeta, Treponema, Borrelia
B. Leptospiras Thermoplasma group
Leptospira, Leptonema Thermoplasma acidophilum

Green sulfur bacteria Thermococcus group

Chlorobium, Chloroherpeton Thermococcus celer
Bacteroides; flavobacteria and relatives Extreme thermophiles
A. Bacteroides group Sulfolobus, Thermoproteus tenax,
Bacteroides, Fusobacterium Desulfurococcus mobilis, Pyrodictium occultum

Source: Hodgson, D. A. 1989. Bacterial diversity: the range of interesting things that bacteria do, pp. 4–22. In: Genetics of
Bacterial Diversity. D. A. Hopwood and K. F. Chater (Eds.). Academic Press, London.

explains, depending upon the growth condi- the investigator to experimentally learn how
tions, prokaryotes are capable of living either the organisms are temporally and spatially
as single cells, when suspended in liquid, or as organized so that myriad activities can take
interacting cells in multicellular populations at place within the cells and between the cells to
physical interfaces, such as solid surfaces. The enable survival in the face of environmental
study of prokaryotes presents a challenge to challenges.
Free ebooks ==>
structure and function 3

1.1 Phylogeny Korarchaeota, has been found in hot environ-

Figure 1.1 shows a current phylogeny of life- ments such as hot springs, and much less is
forms based upon a comparison of ribosomal known about it, or about archaea belonging to
RNA (rRNA) nucleotide sequences. For a more the newly discovered phylum Nanoarchaeota,
complete explanation of Fig. 1.1 and how it which has been cultivated from a submarine hot
was derived, see Box 1.1. Notice that there are vent.
three lines (domains) of evolutionary descent—
Bacteria, Eucarya, and Archaea—that diverged Phenotypes
in the distant past from a common ancestor.2-4 For a comprehensive description of the Archaea,
(The term archaeon may be used to describe the reader is referred to ref. 5. For a review of the
particular Archaea.) Archaea differ from bac- cell surface structures of the Archaea, see ref. 6
teria in ribosomal RNA nucleotide sequences, and also Box 1.6. Archaea are found not only
in cell chemistry, and in certain physiological in extreme environments such as hot springs,
aspects, described in Sections 1.1.1 and 1.2. and alkaline and acid waters, but also in non-
Table 1.1 lists examples of prokaryotes in extreme environments that exist in the oceans,
the different subdivisions within the domains lakes, soil, sewage, swamps, and the animal
Bacteria and Archaea. Notice that the gram- intestinal tract. They are thus widespread in the
positive bacteria are a tight grouping. Although environment.
there is no single grouping of gram-negative From a morphological point of view (size
bacteria, most of the well-known gram-negative and shape), Archaea are similar to the typical
bacteria are in the Proteobacteria division. Bacteria lineage. However, they form a group
of organisms phylogenetically distinct from
1.1.1 Archaea both bacteria and eukaryotes. The Archaea that
Phylogenetic lineages have been best studied commonly manifest one
The two major archaeal phyla are the Euryar- of three phenotypes: methanogenic, extremely
chaeota and the Crenarchaeota. A third phylum, halophilic, and extremely thermophilic.

Fig. 1.1 Phylogenetic relationships among life-forms based upon rRNA sequences. The line lengths are pro-
portional to the evolutionary differences. The position of the root in the tree is approximate. The “purple
bacteria” are now referred to as the phylum Proteobacteria and, as summarized in Table 1.1, comprise a wide
variety of gram-negative organisms including phototrophic, chemoheterotrophic, and chemolithotrophic
bacteria. Source: Woese, C. R., and N. R. Pace. 1993. Probing RNA structure, function, and history by com-
parative analysis. The RNA World. Cold Spring Harbor Press, Cold Spring Harbor, NY.
Free ebooks ==>
4 the physiology and biochemistry of prokaryotes

The methanogenic archaea (Euryarchaeota), Their highly unusual metabolism is explained

also referred to as methanogens, produce meth- in Chapter 14. Methanogens are obligate anaer-
ane. This ability is important for the organisms’ obes that grow in environments such as anaero-
survival, because they derive energy from the bic groundwaters, swamps, and sewage, as well
process. The methanogens produce methane by as part of the digestive tract (rumen) of animals
reducing carbon dioxide to methane or by con- such as cattle and sheep, where they produce
verting acetate to carbon dioxide and methane. methane.


The evolutionary relationships among all can explain the differences in nucleotide
living organisms have been deduced by sequences. The simplest tree is the one that
comparing the ribosomal RNA sequences requires the smallest number of nucleotide
of modern organisms. If the structures of changes to evolve the collection of extant
the ribosomal RNA molecules from differ- sequences from a postulated ancestral
ent living organisms are sufficiently con- sequence.
served in certain segments of the molecule, It has been pointed out that it is diffi-
conserved sequences and secondary struc- cult to assess the validity of a phylogenetic
tures can be aligned to permit comparison tree. There are several reasons for this. For
of the differences in base sequences between example, none of the methods accounts for
RNA molecules. Researchers analyzed the the possibility that multiple changes may
number of positions that differ between have taken place at any single position.
pairs of sequences, as well as other features This shortcoming is likely to result in an
(e.g., which positions vary, the number of underestimate of the distances.
changes that have been made in going from Underestimation of the distances, in
one sequence to another). The number of turn, might bring two distantly related lin-
nucleotide differences between homolo- eages much closer, and in fact might make
gous sequences is used to calculate the evo- them appear to be specifically related to
lutionary distance between the organisms each other when they are not. Other ambi-
and to construct a phylogenetic tree. Most guities in phylogenetic trees might arise if
recently published phylogenetic trees are different positions in the sequence align-
based upon 16S rRNA sequences. ments have changed at very different rates.
There are two major ways in which phy- Furthermore, the branching pattern itself
logenetic trees are constructed from the can differ if a different algorithm is used,
nucleotide differences. In the evolutionary even with the same sequences. The out-
distance method, the number of nucleotide come is that there exist a variety of phyloge-
differences is used as a measure of the evo- netic trees, and no single published tree has
lutionary distance between the organisms. been accepted as perfectly representing the
The second method, the maximum par- 4 billion years or so of bacterial evolution.
simony method, is more complicated. It Despite these problems, there is consistency
takes into account not only the nucleotide in the trees that are based upon 16S rRNA
differences but also the positions at which sequences, and it is believed that most of the
the differences occur and the nature of the phylogenetic trees derived from sequenc-
differences. Parsimony means “less is bet- ing bacterial 16S rRNA are at least plau-
ter” or “stinginess,” going to extreme, for sible. However, the assumption remains
the sake of economy; the method attempts unproven that the relationships between
to find the simplest evolutionary tree that the 16S rRNAs represent the phylogenies
Free ebooks ==>
structure and function 5

of the organisms rather than simply the M. Dworkin, W. Harder, and K.-H. Schleifer
phylogeny of a given 16S rRNA. Because of (Eds.). Springer-Verlag, Berlin.
this, it is imperative that the relationships 2. Stackebrandt, E. 1991. Unifying phylogeny
between the different organisms be tested and phenotypic diversity, pp. 19–47. In: The
by means of other characters (e.g., other Prokaryotes, Vol. I. A, Balows, H. G. Trüper,
M. Dworkin, W. Harder, and K.-H. Schleifer
appropriate molecules besides the 16S (Eds.). Springer-Verlag, Berlin.
rRNA, and various phenotypic character-
3. Felsenstein, J. 1982. Numerical methods
istics of the organisms). See refs 1–4.
for inferring evolutionary trees. Q. Rev. Biol.
REFERENCES 4. Boucher, Y., C. J. Douady, R. T. Papke, D.
A. Walsh, M. E. R. Boudreau, C. I. Nesbo, R.
1. Woese, C. R. 1991. Prokaryotic systemat- J. Case, and W. F. Doolittle. 2003. Lateral gene
ics: the evolution of a science, pp. 3–11. In: The transfer and the origins of prokaryotic groups.
Prokaryotes, Vol. I. A, Balows, H. G. Trüper, Annu. Rev. Gen. 37:283–328.

The extremely halophilic archaea (also in acceptor to oxidize hydrogen gas. These include
the kingdom Euryarchaeota) require very Thermoproteus, Pyrobaculum, Pyrodictium,
high sodium chloride concentrations (at least and Archaeoglobus. [Pyrobaculum and
3–5 M) for growth. They grow in salt lakes Pyrodictium use elemental sulfur (S°) as an elec-
and solar evaporation ponds. The halophilic tron acceptor during autotrophic growth, i.e.,
archaea have light-driven proton and chloride growth on CO2 as the carbon source, whereas
pumps called bacteriorhodopsin and halorho- Archaeglobus uses S2O32−.] Archaea belong-
dopsin, respectively, Bacteriorhodopsin creates ing to the genus Pyrodictium have the highest
an electrochemical proton gradient that is used growth temperature known, with the ability to
to drive ATP synthesis, and halorhodopsin is grow at 110 °C. A few of the sulfur-oxidizing
used to accumulate chloride intracellularly to archaea are acidophiles, growing in hot sulfuric
maintain osmotic stability. These pumps are acid at pH values as low as 1.0. They are called
described in more detail in Chapter 4 (Sections thermoacidophiles, indicating that they grow
4.8.4 and 4.9). optimally in hot acid. For example, Sulfolobus
The extremely thermophilic archaea grow grows at pH values of 1 to 5 and at tempera-
in thermophilic environments (generally tures up to 90 °C in hot sulfur springs, where it
55–100 °C).7 Some of these have an optimal oxidizes H2S (hydrogen sulfide) or So to H2SO4
growth temperature near the boiling point (sulfuric acid). Although most of the extreme
of water! Many use inorganic sulfur either thermophiles are obligately sulfur dependent,
as an electron donor or as an electron accep- some are facultative. For example, Sulfolobus
tor in energy-yielding oxidation–reduction can be grown heterotrophically on organic
(redox) reactions. [The pathways for sulfate carbon and O2 as well as autotrophically on
reduction and sulfur oxidation are described H2S or S°, O2, and CO2. Interestingly, some of
in Sections 13.2.2 and 13.4.1 (see subsection the sulfur-dependent archaea have metabolic
entitled Sulfur-oxidizing prokaryotes), respec- pathways for sugar degradation not found
tively.] Some of these archaea, which are also among the Bacteria. These are mentioned in
called sulfur dependent, oxidize inorganic sul- Section 9.4.
fur compounds such as elemental sulfur and Recently, a new genus of thermoacido-
sulfide, by using oxygen as the electron accep- philic archaea was discovered.8 Picrophilus, a
tor, deriving ATP from the process. Examples member of the kingdom Euryarchaeota, order
include Sulfolobus and Acidianus. Other Thermoplasmales, is an obligately aerobic,
extreme thermophiles are anaerobes that use heterotrophic archaeon that grows at tempera-
elemental sulfur or thiosulfate as the electron tures between 45 and 65 °C at a pH of 0 with
Free ebooks ==>
6 the physiology and biochemistry of prokaryotes

an optimum pH of 0.7. It was isolated from tor, EF-2, that can be ADP-ribosylated by
hot geothermally heated acidic solfataras9 in diphtheria toxin. In contrast, bacteria use
Japan. Despite the sulfur content of the habitat, formylmethionine to initiate protein synthe-
Picrophilus is not sulfur dependent. sis, and their EF-2 is not sensitive to diphthe-
ria toxin. However, the archaeal ribosomes
Comparison of domains Archaea, are similar to bacterial ribosomes in having
Bacteria, and Eucarya an overall size described as 70S; that is, they
Bacteria can be distinguished from Archaea on sediment in a centrifugal field of a velocity of
the basis of the following structural and physi- 70 svedberg units.
ological differences. 6. The methanogenic archaea have several
coenzymes that are unique to Archaea. The
1. Whereas the lipids in the membranes of bac- coenzymes are used in the pathway for the
teria and eukaryotes are fatty acids, which reduction of carbon dioxide to methane and
are esterlinked to glycerol (see later: Figs. in the synthesis of acetyl–CoA from H2 and
1.16 and 10.5), the archaeal lipids are meth- CO2. These coenzymes and their biochemi-
yl-branched, isoprenoid alcohols, ether-
cal roles are described in Section 14.1.5.
linked to glycerol (Fig. 1.18).10 Archaeal
membranes are discussed in Section 1.2.5, In addition to the differences just listed, there
and the biosynthesis of archaeal lipids is dis- are several similarities between archaeal and
cussed in Section 10.1.3. eukaryotic RNA polymerase, and archaeal and
2. Archaea lack peptidoglycan, a universal eukaryotic ribosomes (Table 1.2).
component of bacterial cell walls. The cell
walls of some archaea contain pseudopepti- 1.2 Cell Structure
doglycan, a component absent from bacte-
There are well-known differences between
rial cell walls (Section 1.2.3).
Archaea and Bacteria with respect to cell walls,
3. Archaea contain histones that resemble
cell membranes, and flagella, and these will be
eukaryal histones and bind archaeal DNA
into compact structures resembling eukaryal pointed out. The structure and function of
nucleosomes.11,12 the major cell components will be described,
beginning with cellular appendages (pili and
4. The archaeal RNA polymerase differs from
flagella) and working our way into the interior
bacterial RNA polymerase by having 8 to
of the cell.
10 subunits, rather than 4 subunits, and by
not being sensitive to the antibiotic rifampi-
cin. The difference in sensitivity to rifampi- 1.2.1 Appendages
cin reflects differences in the proteins of the Numerous appendages, each designed for a spe-
RNA polymerase. In fact, the archaeal RNA cific task, can extend from bacterial surfaces.
polymerase resembles the eukaryotic RNA We shall describe two classes: flagella and pili.
polymerase, which also has many subunits The flagella are used by single cells for swim-
(10–12) and is not sensitive to rifampicin. ming in liquid and swarming, a type of group
5. Some protein components of the archaeal swimming on moist solid surfaces. Depending
protein synthesis machinery differ from upon the type, the pili (sometimes called fim-
those found in the bacteria. Archaeal ribo- briae) are used for adhesion to surfaces, includ-
somes are not sensitive to certain inhibitors ing adhesion to the surfaces of animal cells; in
of bacterial ribosomes (e.g., erythromycin, certain cases, described in Box 1.2, they serve
streptomycin, chloramphenicol, and tetracy- for movement on solid surfaces (type IV pili),
cline). These differences in sensitivity to anti- and for conjugal transfer or mating (sex pili).
biotics reflect differences in the ribosomal
proteins. In this respect, archaeal ribosomes Flagella
resemble cytosolic ribosomes from eukary- For a review of all known prokaryotic motil-
otic cells. Other resemblances to eukaryotic ity structures, see ref. 13. For a review of bac-
ribosomes are the use of methionine rather terial flagella, see refs. 14 and 15. Swimming
than formylmethionine to initiate protein bacteria have one or more flagella, which are
synthesis, and a translation elongation fac- organelles of locomotion that protrude from
Free ebooks ==>
structure and function 7

Table 1.2 Comparison between Bacteria, Archaea, and Eucarya

Characteristic Bacteria Archaea Eucarya

Peptidoglycan Yes No No
Lipids Ester linked Ether linked Ester linked
Ribosomes 70S 70S 80S
Initiator tRNA Formylmethionine Methionine Methionine
Introns in tRNA No Yes Yes
Ribosomes sensitive to No Yes Yes
diphtheria toxin
RNA polymerase One (5 subunits) Several (8–12 Three (12–14
subunits each) subunits each)
Ribosomes sensitive to
chloramphenicol, streptomycin,
kanamycin Yes No No

the cell surface. Flagella allow single cells to coli and S. typhimurium. Archaeal flagella are
swim in liquid and can also be used for swarm- somewhat different, as will be described.
ing on a solid surface. Swarming, a means of
group swimming by which bacterial colonies 2. General structure
can spread in limiting liquid, is discussed later The proteins are named after genes found in
in this section. Escherichia coli and Salmonella typhimurium.17
Mutations in the mot (motility) genes result in
1. An overview paralyzed flagella, and as we shall see later, the
The bacterial flagellum is a stiff, helical fila- Mot proteins provide the torque that causes
ment approximately 20 nm in diameter that the flagellum to rotate. The flagellum consists
rotates like a screw-type propeller; it can be of a basal body, a hook, a filament, a motor, a
either a left-handed helix or a right-handed switch, an export apparatus, capping proteins,
helix, depending upon the species. The bac- and junction proteins (typically referred to as
terial flagellum is unrelated to the eukary- fla, fli, flg, flh, and flb). We discuss some of these
otic flagellum in composition, structure, and now.
mechanism of action. (For a description of
eukaryotic flagella and cilia, see note 16 in The basal body. Examine Fig. 1.2. At the base
the section References and Notes, at the end of the flagellum there is a basal body embedded
of the chapter.) The word flagellum, which in the membrane. In gram-negative bacteria, the
means whip in Latin, was first used to describe basal body consists of three stacked rings (C,
the bacterial filament in 1852. The bacterial M, and S rings) and a central rod. The M and S
organelle, however, is more like a stiff propel- rings are actually joined as a single ring called
ler than like a whip, which is a flexible rod. the MS ring, made from different domains of
When the bacterial flagellum rotates, a helical the FliF protein. This scheme is supported by
wave travels from the proximal to the distal electron microscopic evidence. The term “MS”
end (outward from the cell), and as a con- indicates the ring’s location: membrane and
sequence the cell is pushed forward as illus- supramembranous. The P ring (FlgI protein)
trated later (see Fig. 20.2). The flagellum is a and the L ring (FlgH protein) are also named
very complex machine driven by a tiny rotat- according to their location: peptidoglycan and
ing motor embedded in the membrane. Both lipopolysaccharide. The L ring in S. typhimu-
its structure and the mechanism of its motility rium has been shown to be a lipoprotein.18
will be described. The flagella that are studied Presumably the lipid portion helps anchor the
in most detail are those of Escherichia coli and protein in the lipid regions of the outer enve-
Salmonella typhimurium, and we will begin lope. A central rod, made from the FlgB, FliC,
with a discussion of their flagella. The flagella and FlgF proteins, passes through the rings and
of other bacteria, except for the spirochaetes, leads to the hook portion (H) of the flagellum
are similar in general structure to those of E. on the outside of the cell. The outermost two
Free ebooks ==>
8 the physiology and biochemistry of prokaryotes


Many bacteria do not have flagella and yet short, intermittent jerks (hence the name
are capable of motility. Depending upon the “twitching”) of the cells.
bacterium, the motility may take place on a
solid surface (twitching or gliding) or in liq-
uid (swimming). The mechanistic bases for Pore complexes, slime extrusion,
these movements differ and are related to and gliding motility
specific cell structure features present in the
particular cell. See refs. 1–4. Gliding motility by filamentous cyanobac-
teria is apparently due to the secretion of
slime through pores near the septa that
Type IV pili, twitching motility, separate the cells in the filament The model
and gliding motility proposes that as a consequence of the slime
secretion the filament of cells is pushed for-
Type IV pili are fibrillar protein append- ward through the so-called junction pores.
ages located at either pole of certain
gram-negative pathogenic bacteria, such
as Pseudomonas aeruginosa, Bacteroides Gliding in the green fluorescent
ureolyticus, Legionella pneumophila, bacteria (GFP) group
Neisseria meningitidis, Ralstonia solan-
acearum, and Vibrio cholerae, that infect Bacteria in the GFP group glide rapidly on
animals, plants, and fungi. They are also solid surfaces. Some species actually rotate
found in many nonpathogenic gram- as they glide. The cells suspended in liquid
negative bacteria such as Myxococcus can propel adsorbed latex beads in multiple
xanthus, which is a social bacterium that paths around the cell, indicating movement
constructs multicellular fruiting bodies by cell surface molecules, perhaps poly-
(discussed in Chapter 23), and in the uni- mers or fibers, to which the beads attach.
cellular cyanobacterium Synechocystis. Perhaps the same molecules are part of the
Type IV pili comprise the mechanosystem machinery that propels the cells when they
for twitching motility, which is a form of are on a solid surface.
gliding that takes place on moist surfaces.
Twitching motility in M. xanthus is called REFERENCES
social motility (S-motility). It is a form of
smooth motility of cells gliding as groups. 1. McBride, M. J. 2001. Bacterial gliding motil-
ity: multiple mechanisms for cell movement over
Surfaces upon which twitching motility surfaces. Annu. Rev. Microbiol. 55:49–75.
can occur include agar, epithelial tissue,
plant tissue, and various other surfaces 2. Mattick, J. S. 2002. Type IV and twitching.
Annu. Rev. Microbiol. 56:289–314.
such as glass, plastics, and metal. Motility
on such surfaces is important for rapid 3. Semmler, A. B., C. B. Whitchurch, and J. S.
Mattick. 1999. A re-examination of twitch-
colonization, for the formation of bio- ing in Pseudomonas aeruginosa. Microbiology
films (discussed in Chapter 21), and for the 145:2863–2873.
building of fruiting bodies by myxobacte-
4. McBride, M. J., T. F. Braun, and J. L. Brust.
ria (Chapter 23). 2003. Flavobacterium johnsoniae GldH is a
Twitching motility in P. aeruginosa and lipoprotein that is required for gliding and chitin
several other bacteria is characterized by utilization. J. Bacteriol. 185:6648
Free ebooks ==>
structure and function 9

C ring/switch

Fig. 1.2 Bacterial flagellum in a gram-negative envelope (based on Salmonella). The basal body itself, to
which the hook–flagellum assembly is attached, is 22.5 nm × 24 nm in size and is composed of four rings, L, P,
S, and M, connected by a central rod. The M ring is embedded in the cell membrane and the S ring appears to
lie on the surface of the membrane. The S and M rings are actually one ring, the MS ring. The P ring may be in
the peptidoglycan layer, and the L ring seems to be in the outer membrane. The P and L rings may act as bush-
ings that allow the central rod to turn. Gram-positive bacteria have similar flagella but lack the L and P rings.
The MotA and MotB proteins form complexes (Mot) that couple the influx of protons to the rotation of the
rotor. The rotor consists of the MS ring with the FliG protein attached to its cytoplasmic surface and the C ring
attached to the cytoplasmic surface of the MS ring. The switch complex consists of three peripheral membrane
proteins, FliG (also part of the rotor), FliM, and FliN, which probably are closely apposed to the cytoplasmic
face of the M ring. Not shown are hook accessory or adaptor proteins (HAP1 and HAP3) between the hook
and filament, and a protein cap (HAP2) on the end of the filament. The flagellum is assembled from the proxi-
mal to the distal end, with the filament being assembled last. It appears that the HAP1 and HAP3 proteins are
required for the proper assembly of the filament onto the hook. Abbreviations: OM, outer membrane; pg,
peptidoglycan; CM, cell membrane; R, central rod; MOT, MotA and MotB; H, hook; F, filament; L, L-ring;
P, P-ring.

rings (P and L) may act as bushings, allowing and surround the MS ring. The number of such
the central rod to rotate in the peptidoglycan particles in various bacteria has been reported to
and outer membrane. Gram-positive bacteria be 12 to 16. Such arrays of particles surround-
do not have P and L rings. The basal body trans- ing the MS ring were seen in electron micro-
mits torque to the hook and filament, causing graphs of freeze-fractured cell membranes, but
them to rotate. 14 not when either MotA or MotB was missing.20
It has been suggested that the large periplasmic
The motor. One of the fascinating components domain of MotB is attached noncovalently to
of the flagellar apparatus is the tiny motor, the peptidoglycan, explaining why the MotA/
approximately 50 nm in diameter, that lies at MotB complex does not rotate when the motor
the base of the flagellum and causes it to rotate. turns.
For a review of the flagellar motor, see ref. 19. In agreement with the conclusion that MotA
The motor consists of two parts: a nonrotating and MotB are part of the motor, mutations in
part called the stator (made of the Mot pro- the motA and motB genes result in paralyzed

teins) and a rotating part called the rotor, which flagella (Mot phenotype). The MotA/MotB
includes the FliG proteins that transmit torque complexes are believed to conduct protons from
to the MS ring. outside the cell to inside, across the membrane,
The stator consists of two different proteins, and to use this proton movement to provide the
MotA and MotB, indicated as Mot in Fig. 1.2. torque to rotate the rotor. Extreme alkaliphiles
These exist as particles of multiple complexes, and some marine bacteria use a sodium ion cur-
(MotA)4(MotB)2, that span the cell membrane rent. This is reviewed in ref. 21. (As discussed in
Free ebooks ==>
10 the physiology and biochemistry of prokaryotes

Chapter 4, bacteria use a proton current across chemotaxis (Chapter. 20), binds to FliM. (For
the cell membrane to do other kinds of work, more information about these proteins, includ-
e.g., ATP synthesis, in addition to rotating fla- ing their function, see note 28.)
gella.) For an explanation of the conclusion that
The hook and the HAP proteins. The central
MotA and MotB form a complex, see note 22.
flagellar rod is attached to an external curved
Membrane vesicles prepared from strains syn-
flexible hook made of multiple copies of a spe-
thesizing wild-type MotA were more permeable
cial protein called the hook protein, FlgE, the
to protons than were vesicles prepared from
product of the flgE gene. There are also two
strains synthesizing mutant MotA, indicating
hook-associated proteins: HAP1, also called
that MotA is likely to be part of a proton chan-
FlgK (the flgK gene), and HAP3, also called
nel. 23 Some of the evidence in support of the
FlgL (the flgL gene). These proteins are nec-
conclusion that the Mot proteins form a com-
essary to form the junction between the hook
plex that is a transmembrane proton channel is
and the filament. Mutants that lack these HAPs
in ref. 24.
secrete flagellin into the medium. According to
How does passage of protons through the
the model, the hook subunits, consisting of the
Mot complex generate torque? One model pro-
FlgE protein, fill the C ring and then are trans-
poses that as protons pass through the MotA/
ferred en masse to the growing hook through
MotB complexes, conformational changes
the export apparatus, described later, which is
occur in the complexes that drive stepwise rota-
in the middle of the C ring. (See later subsec-
tion of the attached rotor. For a model of how
tion entitled The export apparatus for flagellar
this might occur, consult ref. 25.
What is the rotor? Sometimes the C ring is
referred to as the rotor, but others consider the The filament and the capping proteins. Attached
rotor to be the MS and C rings together. (See to the hook is a rigid, hollow, helical filament
note 26 for references.) An essential rotor com- that, along with the hook, protrudes from the
ponent is FliG. FliG interacts with the Mot pro- cell. When it rotates, it acts as a propeller and
teins and transmits torque generated by the Mot pushes the cell forward. The protein in the fila-
complex to rotate the rotor. The FliG proteins ment is called flagellin (which in Escherichia and
are part of the C ring and attach the C ring to the Salmonella is known as FliC) and is present in
MS ring. As explained next, the C ring functions thousands of copies. Flagellin is not identical in
as a switch that reverses the direction of rota- all bacteria. For example, the protein can vary
tion of the rotor. in size from 20 kDa to 65 kDa depending upon
the bacterial species. Furthermore, although
The switch. In certain bacteria, such as E. coli
there is homology between the C-terminal and
and S. typhimurium, the motor spontaneously
N-terminal ends of most flagellins, the central
changes its direction of rotation periodically.
part can vary considerably and is distinguished
The frequency of switching is influenced by
immunologically in different bacteria. In some
attractants and repellents that the bacterium
cases, there is no homology at all. For example,
might encounter, and as such, is important for
nucleotide-derived amino acid sequences for
chemotaxis, described in Chapter 20. Mutations
the flagellins from Sinorhizobium meliloti show
that cause failure to change flagellar rotation
almost no relationship to flagellins from E. coli,
map in three genes, fliG, fliM, and fliN, which
S. typhimurium, or Bacillus subtilis, but are
code for the complex of switch proteins (FliG,
60% similar to the N and C termini of flagellin
FliM, and FliN).27 It appears that the three
from Caulobacter crescentus.29 Finally, a third
switch proteins form a complex of peripheral
HAP, HAP2, also called FliD (the fliD gene),
membrane proteins closely associated with the
caps the flagellar filament.
cytoplasmic side of the MS ring. In particular,
FliG seems to be bound to the MS ring itself; it 3. Brief summary of assembly
is also bound to FliM and FliN. The latter two of the flagellum
proteins form the cup-shaped C ring, located The flagellum and its associated components
directly beneath the basal body attached to the are assembled in a precise order, beginning with
MS ring. The regulator protein CheY, which is the components closest to the cell membrane.
important in switching flagellar rotation during The first components assembled, probably in a
Free ebooks ==>
structure and function 11

coordinated fashion, are the MS and C rings. How can growth at the tip of the filament
This step is followed by construction of the be demonstrated? Growth at the tip has been
transport apparatus for export of the remain- visualized by the use of fluorophenylanine,
der of the flagellar components through a chan- a phenylalanine analogue, or radioisotopes.
nel in the center of the MS ring. Then the rod Incorporation of fluorophenylalanine by
is assembled, followed by the hook. The hook Salmonella resulted in curly flagella that had
is not completed until the P and L rings have only half the normal wavelength. When the
been assembled. When the hook is complete, analogue was introduced to bacteria that had
the flagellin monomers are exported and the partially synthesized flagella, the completed fla-
filament is assembled. MotA and MotB are gella were normal at the ends next to the cell
assembled late in the assembly process, to coin- and curly at the distal ends, implying that the
cide with the appearance of the filament. As fluorophenylalanine was incorporated at the
described next, the timing of assembly of the tips during flagellar growth.32 When Bacillus
components is at least partially controlled at the flagella were sheared off the cells and allowed
level of the transcription for the genes encoding to regenerate for 40 min before the addition of
these components. radioactive amino acid ([3H]leucine), radioau-
tography showed that all the radioactivity was
4. Brief summary of gene expression control at the distal region of the completed flagella.33
assembly regulating the timing and assembly The flagellin monomers are presumed to be
of flagellum and its associated components transported through the central channel in the
As we have seen, the flagellum and its com- filament to the tip, where they are assembled.
ponents are assembled in a precise order. For more information about the biosynthesis of
Interestingly, the genes for the flagellum and flagella, consult the review by Macnab.14
its associated components are expressed in the
order that the gene products are used. How is The export apparatus for flagellar components.
this done? In E. coli and Salmonella the genes are In the center of the C ring on the cytoplasmic
in three groups that are sequentially expressed. side is a knob composed of several Flh and Fli
The groups are denoted as belonging to class 1, proteins (FlhA, FlhB, FliH, FliI, FliO, FliP, FliQ,
2, or 3. The genes in class 1 (flhDC) comprise FliR). The ATPase FliI is negatively regulated
an operon that encodes transcriptional activa- by FliH, and its activity is required for flagellar
tors of the class 2 genes. Class 2 gene products assembly. (See note 34.) FliI and FliH are solu-
comprise the basal body and hook, as well as a ble proteins, as is a chaperone protein, FliJ, and
sigma factor (FliA) required for transcription of the others are integral membrane proteins. The
class 3 genes. Class 3 genes are required for the knob is the transport apparatus used for export
synthesis of the flagellin monomers (FliC) and of all the flagellar components (basal body, rod,
the torque-generating unit (MotA and MotB). hook, filament) through a central channel in the
MS ring, to be assembled beyond the cytoplas-
5. Growth of the flagellum mic membrane. One of the exported proteins
Although one might expect new flagellin sub- is a muramidase called FlgJ, and it presumably
units to be added to the growing filament at the functions to allow the nascent rod to penetrate
base next to the cell surface, this is not the case. the peptidoglycan. The transport system for the
The flagellin subunits actually travel through a bulk of flagellar materials is a flagellum-specific
hollow core in the basal body, hook, and fila- type III export system. (See the description of
ment and are added at the distal tip. The cap- type III export systems in Chapter 18) The P and
ping protein (HAP2) is important in this regard L rings are assembled using the Sec system of
because it prevents the flagellin monomers from transport, described in Section 18.1.
leaking out into the medium. Somehow, the cap-
ping protein must be able to move out from the 6. Differences in flagellar structure
growing filament to allow extension of the fila- Although the basic structure of flagella is simi-
ment, neither becoming detached nor allowing lar in all bacteria thus far studied, there are
flagellin subunits to leak out into the medium. important species-dependent differences. For
Models by which this may be done have been example, some bacteria have sheathed flagella,
suggested.30, 31 whereas others do not. In some species (e.g.,
Free ebooks ==>
12 the physiology and biochemistry of prokaryotes

Vibrio cholerae) the sheath contains lipopoly- brane, and closely surrounded by an outer lipid
saccharide and appears to be an extension of the bilayer membrane. The spirochaete flagella are
outer membrane.35 In spirochaetes the sheath is called axial filaments. For a review, see ref. 40.
made of protein. (See later.) For a list of diseases caused by spirochaetes, see
Bacteria also vary with respect to the number note 41.
of different flagellins in the filaments. Depending Most spirochaete cells are helically coiled and
upon the species, there may be only one type have two or more flagella (some have 30 or 40 or
of flagellin, or two or more different flagellins more) inserted subterminally near each cell pole
in the same filament. For example, E. coli has and extending along the cell body. Some species
one flagellin; Bacillus pumilis has two differ- have a flat meandering waveform, rather than
ent flagellins, and C. crescentus has three and being helically coiled. The number of flagella
S. meliloti has four.36–38 The data for B. pumilis inserted at opposite poles is the same. The fla-
and C. crescentus support the conclusion that gella are usually more than half the length of the
the different flagellins reside in the same fila- cell and overlap in the middle. The spirochaete
ment. It has been proposed that the filaments of flagellum is often surrounded by a proteina-
S. meliloti are composed of heterodimers of the ceous sheath. Borrelia burgdorferi, the spiro-
two different flagellins. The flagella of many bac- chaete that causes Lyme disease, is somewhat
teria that have been studied (e.g., E. coli) show different. Its axial filaments are not surrounded
a smooth surface under electron microscopy by proteinaceous sheaths. Furthermore, this
and are called plain filaments. However certain organism has a planar waveform shape rather
bacteria (e.g., Rhizobium lupini, S. meliloti) than the corkscrew type.42 The rotation of the
have “complex” filaments with obvious helical rigid periplasmic flagella between the outer
patterns of ridges and grooves on the surface. membrane sheath and the cell cylinder is thought
Flagella with plain filaments rotate either clock- to move the cell by propagating a helical wave
wise or counterclockwise, whereas flagella with backward down the length of the highly flexible
complex filaments rotate only clockwise, with cell cylinder, propelling the cell forward, which
intermittent stops or slowing of rotation. It is allows the cells to “corkscrew” through viscous
thought that because complex filaments are media, such as mud, sediments, and connective
brittle, they are more rigid than plain filaments, tissue in animals. (For a more detailed model,
hence are better suited for propelling bacteria see note 43.) There are five antigenically related
in viscous media such as the gelatinous matrix flagellins in the axial filaments, but it is not
through which R. lupini and S. meliloti must known whether individual filaments contain
swim to infect root hairs of leguminous plants.39 more than one type of flagellin.44 Despite these
A discussion of the role of bacterial flagella in differences, the spirochaete flagella and those
pathogenicity can be found in the review by found in other bacteria are structurally similar
Moens and Vanderleyden.39 insofar as they have a basal body composed of a
Although E. coli and S. typhimurium have series of rings surrounding a central rod, plus a
three rings (MS, P, and L) through which the cen- hook and a filament.
tral rod passes, other bacteria may have fewer,
or more. For example, gram-positive bacteria 7. Site of insertion of flagella and the
lack the outer two rings (P and L). Additional number of flagella
structural elements of unknown function (e.g., The site of insertion of the flagellum and the
additional rings or arrays of particles surround- number of flagella vary with the bacterium.
ing the basal body) have been observed in cer- Some rod-shaped or curved cells have flagella
tain bacteria. that protrude from one or both of the cell poles.
A bacterium with a single, polar flagellum is said
Spirochaete flagella. A major difference between to be monotrichous. Bacteria with a bundle of
spirochaete flagella and those found in other flagella at a single pole are lophotrichous, from
bacteria is that the flagella in spirochaetes do the Greek words lophos (crest or tuft) and trichos
not protrude from the cell; rather, they are in (hair). Bacteria with flagella at both poles are
the periplasm, wrapped around the length of said to be bipolar. They may have either single
the protoplasmic cylinder next to the cell mem- or bundled flagella at the poles. Amphitrichous
Free ebooks ==>
structure and function 13

refers to flagella at both poles. The prefix amphi, more than 100 years ago (1885). Now it is rec-
in Greek, means on both sides.) ognized to be a widespread phenomenon among
Some bacteria (e.g., spirochaetes) have sub- many bacterial genera, both gram-positive and
polar flagella, which are inserted near but not gram-negative.
exactly at the cell poles, and some curved bacte- The cells in swarming populations are called
ria (e.g., Vibrio) have a single, medial flagellum. swarmer cells, and they are often morphologi-
If the flagella are arranged laterally all around cally different from swimmer cells grown in
the cell (e.g., as in Escherichia and Salmonella) liquid. The morphological changes that occur
they are said to be peritrichous. (The prefix when swimming cells from a liquid culture are
peri, in Greek, means around.) Peritrichous inoculated onto an agar plate and convert to
flagella coalesce into a trailing bundle during swarmer cells can be more or less pronounced
swimming. depending upon the bacterium and the concen-
tration of the swarming agar. In general, cells
8. Role of flagella in tactic responses and
that convert from swimming cells to swarming
in virulence
cells become nonseptate filaments, multinucle-
Many swimming bacteria are capable of tactic
oid, and hyperflagellated with lateral flagella.
responses: that is, they swim toward environ-
(Bacillus subtilis swarmer cells differ less dra-
ments more favorable with respect to nutrient,
matically from the nonswarmer cells in being
light, and electron acceptors, and away from
only slightly larger and having only two nuclei.)
toxic or less favorable environments. These
The lateral flagella are critical because to migrate
swimming responses occur because some bac-
as populations of cells, the cells physically inter-
teria can sense environmental signals, transfer
act with each other via the lateral flagella.
these signals to the flagellum motor, and modify
Swarming can be facilitated by the produc-
the rotation of their flagella to swim in a particu-
tion of a surfactant that reduces surface tension
lar direction. How this occurs for bacteria of dif-
at the aqueous periphery of the colony on hydro-
ferent types is discussed in detail in Chapter 20,
philic surfaces such as agar, allowing the colony
which describes chemotaxis. Flagella can also
to expand. The surfactant is one of the compo-
make important contributions to the virulence
nents in the extracellular slime produced by the
of pathogenic bacteria.45 For example, it has
bacteria. For example, Serratia marcescens pro-
been suggested that the ability of spirochaetes
duces a cyclic lipopeptide (3-hydroxydecanoic
(spiral-shaped bacteria) such as Treponema pal-
acid attached to five amino acids), and B. sub-
lidum, the causative agent of syphilis, to swim in
tilis produces a lipopeptide surfactant called
a corkscrew fashion through viscous liquid aids
surfactin.48,49 Isolated swarmer cells rarely
in their dissemination (e.g., through connec-
move. One of the differences between isolated
tive tissue or the junctions between endothelial
cells and cells in a group is that the cells in the
group are encased in much more slime than is
9. Flagella and swarming found around single cells. It could be that wet-
The swarming of flagella is a type of social ting agents in the slime hydrate the external
swimming in which cells move on solid surfaces medium sufficiently for the flagella to rotate,
in groups (called rafts) of physically interact- allowing swarming to take place on the agar.
ing cells. (For a review of the different motility To demonstrate swarming on agar, one
systems that allow bacteria to move on solid inoculates petri plates containing nutrient agar
surfaces, read ref. 46; for a review of swarming at a concentration that is wet enough to allow
behavior, and to learn how swarmer cells dif- swarming to take place (e.g., 0.5–1%, or some-
fer from cells grown in liquid, consult ref. 47.) times 2%), but not so dilute that cells can swim
Swarming allows bacterial populations to rap- as individuals. For example, if the agar concen-
idly spread as multicellular populations, rather tration is between 0.2 and 0.3%, individual bac-
than as single cells, over a wet, solid surface, for teria can swim through water-filled pores in the
example in biofilms discussed in Chapter 21. (As agar. Such very dilute agar plates are referred to
discussed in Box 1.2, gliding represents another as “swim plates.” Swarming is easily recognized
means for bacteria to spread on a solid surface.) because a population of swarming cells spreads
Swarming was first described in Proteus species over the agar plate, whereas nonswarmers grow
Free ebooks ==>
14 the physiology and biochemistry of prokaryotes

as discrete colonies when inoculated in the cen- are assembled by having the subunits pushed
ter of the plate. As reported for B. subtilis, some through the central channel and assembled at
bacteria may lose the ability to swarm when the distal tip. (See earlier subsection entitled 5.
they are maintained as laboratory cultures.49 Growth of the flagellum.) It has been suggested
that archaeal flagella are assembled by adding
A different flagellar system is sometimes pro- subunits at the base.
duced for swarming. Although many bacteria,
such as Escherichia, Proteus, Salmonella, and Fimbriae, pili, filaments, and fibrils
Serratia, use the same lateral flagella system Protein fibrils extending from the cell surface
both for swimming in liquid and for swarming have been called various names, including fim-
on a solid surface (albeit they make many more briae, pili, filaments, and simply fibrils. They
flagella for swarming), other bacteria, such extend from the surface of most gram-negative
as species of Aeromonas, and Azospirillum, bacteria (Fig. 1.3).54-57 Similar fibrils are only
Rhodospirillum centenum, Vibrio alginolyti- occasionally observed in gram-positive bac-
cus, and V. parahaemolyticus, make a spe- teria, such as for Corynebacterium renale and
cial lateral flagella system for swarming and Actinomyces viscosus.58,59 (C. renale is the caus-
growth in viscous environments. (Reviewed in ative agent of bovine pyelonephritis and cystitis,
ref. 50.) These bacteria are polarly flagellated and A. viscosus is part of the normal flora of the
when grown in liquid but become both polar human mouth, where it adheres to other bacte-
and peritrichously flagellated when grown on ria in plaque and to teeth.) All such appendages
surfaces. For example, V. alginolyticus and V. are not alike, nor are their functions known
parahaemolyticus produce many unsheathed in all instances. The size varies considerably.
lateral flagella that aid in swarming when these They can be quite short (0.2 μm) or very long
microorganisms are grown on a solid surface. (20 μm), and they differ in width from 3 nm to
For some bacteria, such as Vibrio species, the 14 nm or greater. Some originate from anchor-
polar and lateral flagella are determined by ing structures in the cell membrane. However,
different genes. The increase in production of most seem not to originate in the cell membrane
lateral flagella may be necessary to overcome at all, and how they are attached to the cell is
surface friction and/or the viscosity of the slime
produced by a swarming population. Perhaps
some sort of mechanosensory system, activated
by surface friction and/or slime viscosity, sig-
nals the expression of the genes for lateral fla-
gella. It has been suggested that the inhibition
of flagellar rotation on the solid surface might
be part of the mechanosensory signaling system
in Vibrio.

10. Archaeal flagella

Archaeal flagella are different from bacterial
flagella.51 The primary sequences of the flagel-
lins from several archaea show no homology
at all to bacterial flagellins, although they do
show sequence homology to each other at the
N-terminal end.52 Also, the basal body struc-
ture in the archaeon Methanospirillum hun-
gatei appears to be simpler than for bacterial
flagella in that a simple knob is present rather
than rings.53 Also, as opposed to bacterial fla- Fig. 1.3 Electron micrograph of a metal-shadowed
gella, there is no central channel in archaeal preparation of a dividing Salmonella typhimurium
flagella. This means that the assembly mecha- showing flagella and fimbriae (or pili). The cell is
nism is different from that of bacterial flagella. about 0.9 μm in diameter. (Reprinted with permis-
The external portions of the bacterial flagellum sion of J. P. Duguid.)
Free ebooks ==>
structure and function 15

not known. Furthermore, they are not always glycoside residues on the cell surface receptors.
present. Although freshly isolated strains fre- They are called common pili, or mannose-sensi-
quently have fibrils, filaments are often lost dur- tive pili, or (more frequently) type 1 pili. Other
ing extended growth in the laboratory. They are pili recognize different cell surface receptors and
apparently useful in the natural habitat but dis- are sometimes called mannose-resistant pili.
pensable in laboratory cultures. What do they
do? Medical significance. Pilus proteins (and other
adhesins) can have great medical significance.
1. Pili (fimbriae) An example of attachment to host cells via
Many of the fibrils can be observed to mediate bacterial pili is the attachment of Neisseria
attachment of the bacteria to other cells (e.g., gonorrhoeae, the causative agent of the sexu-
other bacteria, or animal, plant, or fungal cells) ally transmitted disease (STD) gonorrhea, to
and to abiotic surfaces. Although adhesive epithelial cells of the urogenital tract. Another
fibrils are sometimes referred to as fimbriae to example is E. coli. Most clinical isolates of E.
distinguish them from the sex pili that are used coli from the urinary and gastrointestinal tracts
in bacterial conjugation, in this discussion, all possess type 1 pili. Many also (or instead) have
such fibrils will be referred to as pili. Pili are galactose-sensitive pili, called P-type pili.60,61
important for colonization because they help the P-type pili bind the α-D-galactopyranosyl-(1–
bacteria stick to surfaces. In nature most bacte- 4)-β-D-galactopyranoside in the glycolipids on
ria grow while attached to surfaces, where the cells lining the upper urinary tract.62
concentration of nutrients is frequently highest. Other known pili include the type 4 pili
Attachment can be via adhesive pili and/or non- (type IV pili) produced by N. gonorrhoeae,
fibrillar material that may be part of the cell wall Pseudomonas aeruginosa, Bacteroides nodo-
or glycocalyx (Section 1.2.2). Adhesive pili have sus, and Moraxella bovis (see note 63). These
adhesins, that is, molecules that cause bacteria pili mediate the form of surface motility known
to stick to surfaces. The adhesins are commonly as twitching. A different type of pili are the Tcp
proteins in the pili, often minor proteins at the (toxin-coregulated pili) pili produced by Vibrio
tip, that recognize and bind to specific receptors cholerae. (See note 64 for the diseases caused by
on the surfaces of cells. Adhesion is studied in these bacteria.) The Tcp pili are necessary for
the laboratory under experimental situations of V. cholerae to colonize the intestinal mucosa of
two types: (1) attachment of bacteria to eryth- mammals, and mutants that do not make these
rocytes, causing them to clump (hemagglutina- pili do not cause disease. Thus it is evident that
tion), and (2) attachment of bacteria in vitro to pili of different types are specialized for attach-
host cells to which their attachment normally ment to specific receptors and can account for
occurs in vivo. Often researchers who study specificity of bacterial attachment to hosts and
such attachments have found that when spe- tissues. Pili can be distinguished by inhibition
cific monosaccharides and oligosaccharides are of binding by mono- and oligosaccharides, as
added to the suspension, they inhibit attach- well as by a variety of other methods, including
ment or hemagglutination. The implications are morphology, antigenicity, molecular weight of
that at least some, if not all, pili bind to oligosac- the protein subunit, isoelectric point of the pro-
charides in cell surface receptors, and that the tein subunit, and amino acid composition and
added monosaccharides and oligosaccharides sequence. Since gram-positive bacteria gener-
are inhibitory because they compete with the ally do not have pili, their adhesins are part of
receptor for the adhesin. Receptors on animal other cell surface components (e.g., the glycoca-
cell surfaces include glycolipids and glycopro- lyx, to be described next, in Section 1.2.2).
teins, which are embedded in animal cell mem-
branes via the lipid and protein portions, and 2. Sex pili
present their oligosaccharide moieties to the Bacteria are capable of attaching to each other
outer surface. It has been reported that many for the purpose of transmitting DNA from a
strains of E. coli, Salmonella, and Shigella carry donor cell to a recipient cell. This is called con-
pili whose hemagglutinin activity is prevented jugation or mating. (See Fig. 1.4.) Some bacteria
by D-mannose and methyl-α-mannoside. These (not all) use pili for mating attachments. The pili
pili, therefore, are believed to attach to mannose that mediate attachment between mating cells
Free ebooks ==>
16 the physiology and biochemistry of prokaryotes

1.2.2 The glycocalyx

The term “glycocalyx” is often used to describe
all extracellular material that is external to
donor recipient
the cell wall.68–71 (The word calyx is from the
2 Greek kályx, meaning husk or outer covering.)
Such material includes the extracellular matrix
(ECM) in biofilms, described in Chapter 21. The
3 polymers in glycocalyces are predominantly
polysaccharides and/or proteins. All bacteria
are probably surrounded by glycocalyces as
they grow in their natural habitat, although they
Fig. 1.4 F-pilus-mediated conjugation. Transfer often lose these external layers when cultivated
of a sex plasmid. The donor cell has a plasmid and in the laboratory. The extracellular polymers
an F pilus encoded by plasmid genes. 1, The F pilus may be in the form of S layers, capsules, slime,
binds to the recipient cell. 2, In this model a depo- or a loose network of fibrils.
lymerization of the pilus subunits causes the pilus to The S layers, which are found on the cell wall
retract, bringing the two cells together. 3, The plas-
surfaces of a wide range of gram-positive and
mid is transferred as it replicates so that when the
gram-negative eubacteria, are arrays of protein
cells separate, each has a copy of the plasmid and is
a potential donor. An alternative model [explained or glycoprotein subunits on the cell wall. They
in Babic, A., A. B. Lindner, M. Vulic, E. J. Stewart, are also present in the Archaea, where the S layer
and M. Radman. Direct visualization of horizon- sometimes tightly covers the cell membrane and
tal gene transfer. 2008. Science 319:1533–1536] is serves as the cell wall itself. If the S layer is the
that DNA transfer can occur between separated E. wall itself, it is not considered a glycocalyx.
coli cells through the F-pilus from F+ donor cells to Capsules are composed of fibrous material at
F− recipient cells. the cell surface and generally ensheath the cell
(Figs. 1.5 and 1.6). They may be rigid, flexible,
are different from the other pili just discussed integral (i.e., very closely associated with the
and are called sex pili. The requirement of sex cell surface), or peripheral (i.e., loose) material
pili for mating is found for a variety of gram- that is sometimes shed into the medium.
negative bacteria such as E. coli, but gram-pos- Material that loosely adheres to the cell wall
itive bacteria do not have sex pili. The sex pilus is sometimes called slime or slime capsule. If the
grows on “male” strains that donate DNA to loosely adhering or shed material is polysac-
recipient (“female”) strains. In E. coli, the sex charide, it is often referred to as extracellular
pilus is encoded by a conjugative transmissible polysaccharide (EPS), such as exists in biofilm
plasmid, the F plasmid, that resides in the donor matrices. (In discussing bioflims, “EPS” also
strains and can be delivered to recipients. (See serves as a general term to refer to extracellular
note 65 for an explanation of plasmids.) polymeric substance, not restrictive to polysac-
The gram-positive Enterococcus faecalis charide.) The capsular polysaccharide is usually
forms efficient mating aggregates in liquid sus- covalently bound to phospholipid or phospho-
pension and does not have sex pili. In other lipid A, which is embedded in the surface of the
words, the adhesins are located on the cell sur- cell. This is not the case for extracellular poly-
face rather than on pili. Mating in E. faecalis is saccharide. The synthesis of extracellular poly-
of additional interest because it is induced by saccharides is discussed later, in Section 12.3.
a sex pheromone secreted into the medium by For a review of bacterial capsules and their
recipient cells. The sex pheromone signals the medical significance, see ref. 72.
donor cells to synthesize cell surface adhesins
that promote cell aggregate formation (clump- Chemical composition
ing) and subsequent DNA transfer.66 Mating The glycocalyces from several bacteria have
interactions mediated by surface adhesins is been isolated and characterized. Although
widespread in the microbial world. Other well- many are polysaccharides, some are proteins.
studied examples include Chlamydomonas For example, some Bacillus species form a
mating and yeast mating.67 glycocalyx that is a polypeptide capsule. Also,
Free ebooks ==>
structure and function 17

Fig. 1.5 Electron micrograph of a thin section of E. Fig. 1.6 Electron micrograph of a thin section of bac-
coli stained with a ruthenium red dye. The cells are teria adhering to a rock surface in a subalpine stream.
adhering to neonatal calf ileum. Source: Costerton, J. The sample, showing fibrous glycocalyx, was stained
W., T. J. Marrie, and K.-J. Cheng. 1985. Phenomena of with ruthenium red dye. Source: Costerton, J. W.,
bacterial adhesion, pp. 3–43. In: Bacterial Adhesion. T. J. Marrie, and K.-J. Cheng. 1985. Phenomena of
D. C. Savage and M. Fletcher (Eds.). Plenum Press, bacterial adhesion, pp. 3–43. In: Bacterial Adhesion.
New York and London. With kind permission from D. C. Savage and M. Fletcher (Eds.). Plenum Press,
Springer Science + Business Media B.V. New York and London. With kind permission from
Springer Science + Business Media B.V.

pathogenic Streptococcus species have a fibril- species. For example, the K1 polysaccharide of
lar (hairlike) protein layer, the M protein, on E. coli is identical to the group B capsular poly-
the external face of the cell wall. (See note 73 saccharide of Neisseria meningitidis.
for a description of the role of M protein in S.
pyogenes pathogenesis.) Role
The capsular polysaccharides are extremely An important role for the glycocalyx can be
diverse in their chemical composition and adhesion to the surfaces of other cells or to inan-
structure.74 Some consist of one type of mono- imate objects to form a biofilm. Such adhesion
saccharide (homopolymers), whereas some are is necessary for colonization of solid surfaces or
composed of more than one type of monosac- growth in biofilms (Figs. 1.5 and 1.6). The bac-
charide (heteropolymers). The monosaccha- teria can adhere to a nonbacterial surface or to
rides are linked together via glycosidic linkages each other via these attachments. An example
to form straight-chain or branched molecules. is the complex ecosystem of several different
They can be substituted with organic or inor- bacteria maintained by intercellular adherence
ganic molecules, further increasing their diver- in human oral plaque.75 Advantages to such
sity. For example, E. coli strains make more adherence include the higher concentrations of
than 80 different capsular polysaccharides, nutrients that may be found on the surfaces.
called K antigens. Sometimes the same capsular Another role of the glycocalyx is protection
polysaccharide is made by different bacterial from phagocytosis. For example, mutants of
Free ebooks ==>
18 the physiology and biochemistry of prokaryotes

pathogenic strains of bacteria that no longer and gram-negative walls, responsible for the
synthesize a capsule, such as unencapsulated strength of bacterial cell walls. Then we con-
strains of Streptococcus pneumoniae, are more sider, in turn, the gram-positive and gram-neg-
easily phagocytized by white blood cells, mak- ative cell walls.
ing the pathogens less virulent.
The glycocalyx can also prevent dehydration The Gram stain
of the bacterial cell, an important role in the The Gram stain was invented in 1884 by a
soil. This is because polyanionic polysaccha- Danish physician, Christian Gram, to allow the
rides, including polysaccharide capsules, are use of ordinary bright-field microscopy in the
heavily hydrated. visualization of bacteria in tissues. For more
about Christian Gram, see Box 1.3.
1.2.3 Cell walls When appropriately stained, bacteria with
Most bacteria are surrounded by a cell wall that cell walls can usually be divided into two
lies over the external face of the cell membrane groups, depending upon whether they retain
and protects the cell from bursting due to the a crystal violet–iodine stain complex (gram-
cell’s internal turgor pressure.76,77 The turgor positive) or do not (gram-negative) (Fig. 1.7).
pressure exists because bacteria generally live Gram-negative bacteria are visualized with a
in environments that are more dilute than the pink counterstain called safranin. During the
cytoplasm. As a consequence, there is a net staining procedure the complex of crystal violet
influx of water that results in a large hydro- and iodine can be readily removed with alco-
static pressure (turgor) of several atmospheres hol or acetone from gram-negative cells, but
directed out against the cell membrane. Two less readily from gram-positive cells. This result
major types of wall exist among the bacteria. is thought to be related to the thickness of the
One type of wall can be stained by using the peptidoglycan layer in the gram-positive wall in
Gram stain procedure, and relatively restricted comparison to the much thinner peptidoglycan
groups of bacteria having walls of this type are layer in the gram-negative wall. During the alco-
called gram-positive. Many more bacteria pos- hol/acetone wash, the outer lipopolysaccharide
sess the second wall type, do not stain, and are layer of the gram-negative bacteria is disrupted
called gram-negative. and the dye leaks out through the thin, porous
We introduce the Gram stain next, follow- peptidoglycan layer.
ing up with a description of peptidoglycan, a The difference in Gram staining is also related
cell wall polymer found in both gram-positive to the outer wall layer of gram-negative bacteria


Christian Gram was a Danish physician different bacteria, which later would be
who worked at the morgue of the City characterized as gram-positive and gram-
Hospital of Berlin. In 1894 he published a negative, but in 1886 Carl Flügge pub-
procedure he had devised for staining bac- lished a textbook in which he pointed out
teria. The method allowed the staining of that Gram’s staining procedure is useful for
all bacteria in tissue preparations to dif- the differential staining of bacteria in tis-
ferentiate them from nuclei. Some bacteria sues and for diagnostic purposes. Christian
did not retain the stain, however, making Gram died in 1935.
the procedure less effective than Gram had
desired. Source: Lechevalier, H. A., and M. Solotorovsky.
It is not clear who realized that the Gram 1965. Three Centuries of Microbiology.
stain could be used to distinguish between McGraw-Hill Book Company, New York.
Free ebooks ==>
structure and function 19

(also called the outer envelope or outer mem- Although Fig. 1.8 depicts the glycan chains
brane), which is rich in phospholipids and thus running parallel to the cell membrane, this is a
made leaky by the lipid solvents, alcohol, and controversial point. Another model, the scaf-
acetone. The Archaea can stain either gram-pos- fold model, proposes that the glycan chains run
itive or gram-negative, but their cell wall com- perpendicular to the cell membrane.78 Others
position is different from that of the Bacteria, as have argued that the parallel model is more
will be described later. consistent with experimental data than the scaf-
fold model.79 The glycan consists of alternating
Peptidoglycan residues of N-acetylglucosamine (GlcNAc or
The strength and rigidity of bacterial cell walls G) and N-acetylmuramic acid (MurNAc or M)
is due to molecules called peptidoglycan or attached to each other via β-1,4 sugar linkages.
murein, which consist of glycan chains cross- Attached to the MurNAc is a tetrapeptide that
linked by peptides (see later: Figs. 1.18, 12.3). cross-links the glycan chains via peptide bonds
(black dotted stems on M residues, Fig. 1.8).
The tetrapeptide usually consists of the follow-
ing amino acids in the order that they occur
outer membrane
from MurNac: L-alanine, D-glutamate, L-R3
peptidoglycan (the R indicates that the amino acid can vary
cell between species: see note 80), and D-alanine.
membrane The structural diversity of the peptidoglycan, as
well its synthesis, is discussed in more detail in
Section 12.1. The peptidoglycan forms a three-
dimensional network surrounding the cell mem-
brane, covalently bonded throughout by the
glycosidic and peptide linkages. It is the cova-
lent bonding that gives the peptidoglycan its
strength. The shape of the peptidoglycan helps
maintain the shape of the cell as well as shape
changes that occur during cellular morphogen-
esis (e.g., during the formation of round cysts
gram- gram-
negative positive
from rod-shaped vegetative cells). In addition
to peptidoglycan, certain cytoskeletal compo-
Fig. 1.7 Schematic illustration of a gram-negative nents such as MreB and crescentin are involved
and a gram-positive bacterial cell wall. Note the pres- in shape determination. This is discussed later
ence of an outer membrane (also called outer enve- in this chapter, in Section 1.2.7. Destruction
lope) in the gram-negative wall and the much thicker of the peptidoglycan by the enzyme lysozyme
peptidoglycan layer in the gram-positive wall. (which hydrolyzes the glycosidic linkages) or




Fig. 1.8 Schematic drawing of a peptidoglycan layer. The peptidoglycan surrounds the cell membrane and
consists of glycan chains (–G–M–) cross-linked by tetrapeptides (solid circles). The peptidoglycan is approxi-
mately one monomolecular layer thick in gram-negative bacteria and several layers thick in gram-positive
bacteria. The direction in which the glycan chains are depicted to be running is not to be taken literally. It has
been suggested to run in a helical pattern in rod-shaped cells. Abbreviations: M, N-acetylmuramic acid; G,
Free ebooks ==>
20 the physiology and biochemistry of prokaryotes

interference in its synthesis by antibiotics (such organic solvents, the cell membrane adheres to
as penicillin, vancomycin, or bacitracin) results the outer envelope in numerous places, while
in the inability of the cell wall to restrain the generally shrinking away in other locations. It is
turgor pressures. Under these circumstances, reasonable to suggest that these zones of adhe-
the influx of water in dilute media causes the sion may be areas in which the peptidoglycan
cell to swell and burst. Since the strength of the bonds the cell membrane to the outer membrane
peptidoglycan is due to its covalent bonding because the peptidoglycan lies between the two
throughout, changes in its shape, or expansion membranes.
during growth of the cell, must be accompanied
by hydrolysis of some of the covalent bonds and Gram-positive walls
synthesis of new bonds. However, because of 1. Chemical composition
the high intracellular turgor pressure, this must The gram-positive cell wall in bacteria is a thick
be under tight control; otherwise a lethal weak- structure approximately 15 to 30 nm wide and
ness in the peptidoglycan structure will be pro- consists of several polymers, the major one
duced when its bonds are cleaved. This problem being peptidoglycan (Figs. 1.8 and 1.9). The
is discussed by Koch.81 kinds and amounts of other polymers in the
There are some important differences wall vary according to the bacterial taxa. The
between the peptidoglycans in gram-positive nonpeptidoglycan polymers can comprise up to
and gram-negative bacteria. The peptidogly- 60% of the dry weight of the wall. Most of these
can of gram-negative bacteria can be isolated are commonly found covalently linked to the
as a sac of pure peptidoglycan that surrounds glycan chain of the peptidoglycan. These poly-
the cell membrane in the living cell. This recep- mers include teichoic acids, teichuronic acids,
tacle, called the murein sacculus, is elastic and neutral polysaccharides, lipoteichoic acids, and
is believed to be under stress in vivo because of glycolipids of different types (Fig. 1.10).
the expansion due to turgor pressure against the
cell membrane.82 In contrast, the peptidogly- Teichoic acids. For a review, see ref. 83. Teichoic
can from gram-positive bacteria is covalently acids are polyanionic polymers of either ribi-
bonded to various polysaccharides and teichoic tol phosphate or glycerol phosphate joined by
acids and cannot be isolated as a pure murein anionic phosphodiester bonds (Fig. 1.10). They
sacculus (See the discussion of the gram-posi- can comprise 30 to 60% of the dry weight of the
tive wall in the next subsection.) wall and may have multiple roles, as described
The peptidoglycan from gram-negative in ref. 84. Teichoic acids vary structurally with
bacteria differs from the peptidoglycan from respect to the extent and type of molecules
gram-positive bacteria in two other ways: (1) covalently bonded to the hydroxyl groups of
diaminopimelic acid is generally the diamino the glycerol phosphate or ribitol phosphate
acid in gram-negative bacteria, whereas this in the backbone. Frequently, the amino acid
D-alanine is attached, as well as the sugar glu-
position is much more variable among gram-
positive bacteria; and (2) the cross-linking is cose or N-acetylglucosamine. The teichoic acids
generally direct in gram-negative bacteria, are attached to the peptidoglycan via covalent
whereas there is usually an additional peptide bonds between the phosphate of glycerol phos-
bridge between the tetrapeptide stems in gram- phate or ribitol phosphate to the C6 hydroxyl of
positive bacteria (see later: Fig. 12.3). N-acetylmuramic acid (MurNac).
As described in Chapter 12, the monolayer Teichuronic acids. Teichuronic acids are polyan-
of peptidoglycan in gram-negative bacteria is ionic acidic polysaccharides containing uronic
attached noncovalently to the outer envelope acids (e.g., some have N-acetylgalactosamine
via lipoprotein. Whether it is also attached to and D-glucuronic acid). When Bacillus subtilis is
the cell membrane is a matter of controversy. grown in phosphate-limited media, the teichoic
The evidence for attachment to the cell mem- acid is replaced by teichuronic acid (reviewed
brane is that when E. coli is plasmolyzed (placed in ref. 84).
in hypertonic solutions so that water exits the
cells), and prepared for electron microscopy by Neutral polysaccharides. Neutral polysaccharides
chemical fixation followed by dehydration with are particularly important for the classification of
Free ebooks ==>
structure and function 21

Fig. 1.9 Schematic drawing of the gram-positive wall. Components, starting from the bottom, are as follows:
PM, plasma cell membrane consisting of protein (Pr), phospholipid (Pl), and glycolipid (Gl). Overlying the
cell membrane is highly cross-linked peptidoglycan (PG) to which are covalently bound teichoic acids, teichu-
ronic acids, and other polysaccharides (SP). Acylated lipoteichoic acid (aLTA) is bound to the cell membrane
and extends into the peptidoglycan. Some of the LTA is in the process of being secreted (LTAt). LTA that
already has been secreted into the glycocalyx is symbolized as aLTAx. Some of the LTA in the glycocalyx is
symbolized as dLTAx because it has been deacylated (i.e., the fatty acids have been removed from the lipid
moiety). Within the glycocalyx can also be found lipids (L), proteins (Pe), pieces of cell wall (W), and polymers
that are part of the glycocalyx proper (G). Also shown is the basal body and hook structures of a flagellum (B).
Source: Wicken, A. 1985. Bacterial cell walls and surfaces, pp. 45–70. In: Bacterial Adhesion. D. C. Savage
and M. Fletcher (Eds.). Plenum Press, New York and London. With kind permission from Springer Science +
Business Media B.V.

streptococci and lactobacilli, where they are used hydrophobic lipid, the molecule is amphipathic
to divide the bacteria into serological groups (e.g., (i.e., it has both a polar and a nonpolar end).
groups A, B, and C streptococci). The lipid portion is bound hydrophobically to
the cell membrane, whereas the polyglycerol
Lipoteichoic acids. A polyanionic teichoic acid phosphate portion extends into the cell wall.
found in most gram-positive walls is lipote- Unlike the other polymers thus far discussed,
ichoic acid (LTA), which is a linear polymer LTA is not covalently bound to the peptido-
of phosphodiester-linked glycerol phosphate glycan. Its biological role at this location is not
covalently bound to lipid. The C2 position of understood, but in some bacteria it is secreted
the glycerol phosphate is usually glycosylated and can be found at the cell surface, where it is
and/or D-alanylated. Because of the negatively thought to act as an adhesin. For example, LTA
charged backbone of glycerol phosphate and the is secreted by S. pyogenes, where it binds with
Free ebooks ==>
22 the physiology and biochemistry of prokaryotes

Fig. 1.10 Some teichoic and teichuronic acids found in different gram-positive bacteria. (A) Glycerol
phosphate teichoic acid with D-alanine esterified to the C2 of glycerol. (B) Glycerol phosphate teichoic
acid with D-alanine esterified to the C3 of glycerol. (C) Glycerol phosphate teichoic acid with glucose and
N-acetylglucosamine in the backbone subunit. D-Alanine is esterified to the C6 of N-acetylglucosamine.
(D) Ribitol phosphate teichoic acid, with D-glucose attached in a glycosidic linkage to the C4 of ribitol. (E)
Teichuronic acid with N-acetylmannuronic acid and D-glucose. (F) Teichuronic acid with glucuronic acid
and N-acetylgalactosamine. It is believed that teichoic and teichuronic acids are covalently bound to the
peptidoglycan through a phosphodiester bond between the polymer and a C6 hydroxyl of muramic acid in
the peptidoglycan.

the M protein and acts as a bridge to receptors Bacteria belonging to the genus Mycobacterium
on host tissues. (See note 85 for a further expla- include the causative agents of tuberculosis (M.
nation.) Under these circumstances, one should tuberculosis) and leprosy (M. leprae). M. tuber-
consider the secreted LTA to be part of the gly- culosis infects approximately one-third of the
cocalyx along with the M protein. world’s population and causes about 3 million
deaths annually. Between 12 million and 13
Other glycolipids. There is a growing list of million people worldwide are infected with M.
gram-positive bacteria that do not contain leprae. For more complete descriptions of tuber-
LTA but have instead other amphiphilic gly- culosis and leprosy, see Boxes 1.4 and 1.5, respec-
colipids that might substitute for some of the tively. The cells walls of mycobacteria consist of
LTA functions, whatever they might be.86 waxy lipids that can comprise up to 40% of the
Bacteria having these cell surface glycolip- dry weight of the cell wall. The waxy lipids are
ids (also called macroamphiphiles or lipo- responsible for resistance to dehydration, acids,
glycans) belong to various genera, including and alkalies. Indeed, treatment of sputum with
Micrococcus, Streptococcus, Mycobacterium, dilute sulfuric acid or dilute sodium hydroxide
Corynebacterium, Propionibacterium, Actino- enriches for mycobacteria, as opposed to other
myces, and Bifidobacterium. bacteria in the respiratory flora.
Free ebooks ==>
structure and function 23


Tuberculosis is a lung disease caused by verb “to caseate” comes from the Latin
Mycobacterium tuberculosis, but the bac- caseus, which means cheese.) The caseated
teria can spread to other tissues. The lung lesions may heal, becoming infiltrated with
tissue is slowly destroyed by a complex pro- fibrous tissue and, often, acquiring depos-
cess involving macrophages activated by its of calcium, which form nodules (large
the immune system. The symptoms are the tubercles) that can be seen in X-ray images.
coughing up of mucus, which may be bloody, The walled-off tubercles remain in the lung
low-grade fever, night sweats, and weight for the rest of the individual’s life and con-
loss (wasting). Tuberculosis used to be called tain viable bacteria. Most active cases of
consumption and also the white plague. tuberculosis are due to the reactivation of
tubercles. This may happen in individuals
with weakened immune systems, such as
Pathogenesis people with AIDS, or in the elderly.
During an early stage in the progress of
The bacteria enter the respiratory tract the disease—that is, during the primary
via inhalation of nasopharyngeal secre- lesion stage—the bacteria may spread into
tions from individuals with active disease. regional lymph nodes, and from there into
Once bacteria have entered the lungs, small the bloodstream. The bacteria may also
lesions, frequently called primary lesions, spread into the bloodstream from case-
are produced, where the bacteria reproduce. ated lung tissue. The circulatory system can
At this stage of the infection, the bacteria bring the bacteria to various organs and tis-
are engulfed by nonactivated alveolar mac- sues, including the bone marrow, spleen,
rophages. A certain fraction of the bacteria liver, meninges, and kidneys. The bacteria
survive and grow inside the macrophages can even return from these organs, to the
at the site of the lesions. When they grow lungs via the blood circulatory system,
inside the macrophages, the macrophages and this can contribute to the reactivation
are killed and release the bacteria. Since, of the disease in the lungs years after the
however, the cellular immune system reacts initial infection. An asymptomatic period
against the bacteria, the alveolar mac- of years frequently precedes the appear-
rophages are activated, becoming better ance of symptoms in various body organs.
killers of the mycobacteria. This stops the Symptoms, when they do occur, are caused
spread of the infection. The lesions heal and by malfunction of the organs or tissues
form granulomas, also called tubercles, con- due to necrotizing lesions, or to damage of
taining large activated macrophages called blood vessels that feed the tissues.
epithelioid cells because they resemble epi-
thelial cells, and multinucleated giant cells
derived from the fusion of epitheloid cells The role of the cellular immune
(Langhans giant cells). The tubercles also response in damaging infected tissue
contain lymphocytes and fibroblasts, the
latter forming fibrous tissue in the tubercles. The macrophages are activated by lympho-
There are relatively few bacteria present in cytes that are activated during the immune
the tubercles, and these are mostly outside response to the bacteria. These are CD4+
the macrophages. However, the activated (T-helper cells) and CD8+ (T-cytolytic)
macrophages eventually destroy the lung lymphocytes. The T-helper (TH) cells acti-
tissue in the center of the tubercles, caus- vate the other lymphocytes, namely, B
ing it to become semisolid or “cheesy.” The lymphocytes, which mature into antibody-
“cheesy” tissue is said to be caseated. (The secreting cells, TC cells, and macrophages.
Free ebooks ==>
24 the physiology and biochemistry of prokaryotes

The B lymphocytes are part of the humoral A more rapid test is to use the polymerase
response, and they mature to make antibod- chain reaction (PCR) to amplify the bacte-
ies. The T-cytolytic lymphocytes (TC cells) rial DNA. The PCR assay can be applied to
are part of the cellular immune response, sputum material, and specific DNA probes
and their major role is to attach to and can be used to detect M. tuberculosis DNA
kill virus-infected cells and foreign tissues. sequences. The reagents can be purchased
Both the TH and TC cells produce interferon in a kit.
gamma (IFN-γ), which is a major activa-
tor of macrophages. The macrophages are
said to be active when they become capable Who gets sick
of new activities, such as the engulfment
and killing of bacteria. Hence, activated Tuberculosis is a major worldwide killer,
macrophages are also part of the cellular certainly one of the most important of
immune response and limit the spread of the lethal infectious diseases. The World
the infection. Tissue necrosis (destruction) Health Organization (WHO) estimates
is not due to toxins produced by the myco- that about 10 million new active cases of
bacteria. Rather, it results when hydrolytic tuberculosis occur each year, and 3 mil-
enzymes and toxic forms of oxygen are lion people die. In the twentieth century
released by the activated macrophages in alone there were about 1 billion deaths
the granulomas. These factors are made due to tuberculosis. It has been estimated
by macrophages to kill the pathogens they that about half the world’s population of
have engulfed; but when leaked into the approximately 6 billion people is infected,
interstitial fluid, they kill surrounding tis- but that only 30 million of infected people
sue. Activated macrophages also release have active cases. This means that about
factors that cause blood clotting (proco- 99% of people who become infected
agulant factors), and when the clots form with Mycobacterium tuberculosis do
in local blood vessels, blood supply to the not become ill and are healthy carri-
tissues is depressed, resulting in death of the ers, although as noted, the disease can
tissues. become activated later in life. Estimates
are that approximately 15 million people
in the United States are infected. A skin
Laboratory diagnosis test called the tuberculin testt can detect
whether a person has been infected with
The identification of acid-fast rods in M. tuberculosis.
smears of sputum is a tentative diagnosis of
tuberculosis. To identify acid-fast rods, the
Ziehl–Neelson or Kinyoun method is used How they get sick
to stain the smears. The cells in the smears
can also be stained with a fluorescent rho- The infection is transmitted from some-
damine–auramine dye and viewed with a one with an active case of tuberculosis
fluorescence microscope. In part because who coughs or sneezes, producing respira-
there may be only very few organisms tory droplets that the next victim inhales.
present in the smears, culturing the speci- Some people who have been infected are
mens is more definitive. The cultures can free of symptoms for many years and may
be grown on medium containing egg yolk die without developing active tuberculo-
or oleic acid and albumin. However, the sis. Activation of the disease is associated
bacteria grow very slowly (the cells divide with malnutrition, old age, and a sup-
appproximately once a day), and it may pressed immune system, as in people with
be 3 to 6 weeks before growth is visible. AIDS.
Free ebooks ==>
structure and function 25

Treatment from a live, attenuated bovine strain of

tubercle bacillus (M. bovis). “Attenuated”
Antituberculosis drugs must be taken for at means that the strain has been grown in cul-
least 6 months. The drugs that can be taken ture for a long time and has lost so much
are isoniazid (Nydrazid), rifampin (Rifadin virulence that it cannot cause tuberculosis.
and Rimactane), ethambutol (Myambutol), BCG is widely used worldwide; in develop-
streptomycin, and pyrazinamide. Usually ing countries it is routinely given to chil-
more than one medication is taken. dren at birth. The WHO has estimated that
85% of children born in 1990 received the
BCG vaccination during the first year of
Tuberculin skin test life. However, there is a great deal of uncer-
tainty about BCG’s efficiency, which has
Tuberculin (called old tuberculin) refers to reportedly ranged from 0% (one study in
material isolated from M. tuberculosis by India) to 90%. The results of studies con-
boiling the cells. A purified protein deriva- ducted in the United Kingdom indicate
tive (PPD) of old tuberculin is used for the protection against infection of approxi-
skin test. It is a mixure of cell wall proteins mately 70% of vaccinated individuals. All
and polysaccharides prepared by frac- tuberculin-negative children in the United
tionation with trichloroacetic acid (TCA), Kingdom receive the BCG vaccination at
ammonium sulfate, and alcohol. The PPD is 10 to 13 years of age.
injected under the skin. If the person being The BCG vaccine is not routinely used in
tested has not been exposed to M. tubercu- the United States. The reasons for this in the
losis, there is no immediate reaction, and a past have been the relatively low incidence
positive skin reaction occurs 3 to 5 weeks of the disease (<1% of children and young
after inoculation. If the individual has been adults give a positive tuberculin skin test)
infected, a delayed-type hypersensitivity and the fact that any person who has been
reaction occurs at the site of injection vaccinated will test positive in the tubercu-
within a couple of days. This is manifested lin test. Under the latter circumstances, it
as an indurated (raised, hard) area that is would not be possible to use the tubercu-
often red (erythematous). The diameter lin test to ascertain the spread of the infec-
of the induration, measured within 2 to 3 tion. Nor would it be possible to ascertain
days, is interpreted according to the follow- whether an individual has become recently
ing list of ranges: infected and therefore should receive pre-
ventive therapy with isoniazid.
0–4 mm = negative
> 10 mm = positive
5–10 mm = uncertain; must be retested
For an immunocompromised person (e.g., The people who have died from tuberculo-
an AIDS patient), a 5 mm diameter is con- sis between 1805 and 1967 include many
sidered to be a positive test. who have been very important in music,
writing, and the theater: the English poet
John Keats (died in 1821), the English
BCG vaccine author D. H. Lawrence (1930), the English
actress Vivien Leigh (1967), the English
There exists a vaccine for tuberculosis: the novelist and essayist George Orwell (1950),
bacille Calmette–Guérin (BCG) vaccine, the American playwright Eugene O’Neill
named after the French microbiologists who (1953), the Polish pianist and composer
developed the drug. The vaccine is made Frédéric Chopin (1849), the Scottish novelist
Free ebooks ==>
26 the physiology and biochemistry of prokaryotes

and poet Walter Scott (1832), the Italian Russian playwright and novelist Anton
violinist and composer Nicolò Paganini Pavlovich Chekhov (1904), and the German
(1840), the German poet and playwright poet and playwright Johann Christoph
Johann Wolfgang von Goethe (1832), the Friedrich von Schiller (1805).


Leprosy, a chronic disease caused by the lungs in response to M. tuberculosis
Mycobacterium leprae, can manifest itself infection and are partly attributable to the
in cutaneous lesions, nerve impairment ability of the bacteria to grow inside the
resulting in sensory loss, and tissue destruc- macrophages. The cell-mediated immune
tion. The bacteria infect the skin, periph- response against the bacteria results in
eral nerves, eyes, and mucous membranes, nerve damage due to inflammation. As a
and cartilage, muscle, and bone may be consequence, there is loss of sensation in
destroyed. Humans are the only known the area of the lesions.
source of M. leprae. The bacteria are obli-
Lepromatous leprosy. The skin lesions are
gate intracellular parasites and grow well
anesthetic, extensive, large, nodular, and
in the peripheral nerves, specifically in the
disfiguring. The face becomes disfigured
Schwann cells, which make up the myelin
(lion’s face) owing to thickening, nodules,
and plaques. Eyebrows and eyelashes are
lost. Lepromatous leprosy results when the
immune system fails to mount an effective
cell-mediated defense. It is a progressive
disease that is fatal if not treated. The
There are two forms of leprosy: leproma-
bacteria disseminate throughout the body
tous and tuberculoid, but these are two
and can be found in all the body organs.
extremes, and infected people can display
The bacteria also grow in the mucous
symptoms intermediate between them.
membranes of the nose. The cartilaginous
Tuberculoid leprosy. This is a self-limiting septum is destroyed, resulting in severe nasal
disease that may regress. Skin lesions are deformities, including collapse of the nose.
present but contain very few organisms. There is also damage to underlying cartilage
The lesions, which can occur anywhere and bone. Fingers or toes may be lost.
on the body, are usually one or two anes-
thetic (without sensation), hypopigmented
macules (flat spots) with raised reddish Laboratory diagnosis
or purple edges. The lesions vary in size
from a few millimeters to larger ones that Diagnosis depends upon detecting the
that may cover the entire trunk. There is bacteria in skin smears and skin biopsies.
palpable thickening of peripheral nerves Generally such tests are positive only for
due to the growth of bacteria in nerve lepromatous leprosy. To make a skin smear,
sheaths. Skin samples subjected to biopsy the dermis is cut and tissue fluid is scraped
show tuberculoid granulomas consisting of from the slit and smeared on a microscope
activated macrophages and lymphocytes. slide. The smears are stained for acid-
These resemble the granulomas formed in fast bacilli by using a modification of the
Free ebooks ==>
structure and function 27

Ziehl–Neelson stain. For the skin biopsy, as contagious as once thought. Spread by
material is taken from the lesion and fixed skin contact would probably require bro-
in 10% formalin and stained for acid-fast ken skin on the recipient for the bacteria to
bacilli. The bacteria cannot be cultured in penetrate.
cell-free cultures (i.e., in vitro). They can
be grown in mouse footpads, however, by
injecting a suspension of bacteria derived Treatment
from a skin biopsy sample. It takes about
6 months for maximum growth to occur. The drug Dapsone, which is related to
Growing the bacteria in mouse footpads the sulfonamides, will successfully treat
allows diagnosticians to test the sensitiv- leprosy. For strains that show some resis-
ity of the infecting organism to therapeutic tance to Dapsone, a combination of drugs,
drugs such as Dapsone and rifampin. M. including Dapsone, rifampin, and clofaz-
leprae can also be similarly cultivated by imine may be used.
inoculation into armadillos.

Who gets sick
Leprosy is an ancient and infamous disease
Between 12 million and 13 million people often referred to as Hansen’s disease, after
are infected worldwide, mostly in develop- the Norwegian scientist Gerhard Hansen,
ing countries, especially those in tropical who discovered the causative agent in
or semitropical areas. In the past, how- 1874. The disease apparently started in the
ever, leprosy has not been confined to the Far East prior to 600 BCE and spread to the
warmer climates. It is most prevalent in Near East, Africa, and finally to Europe,
India, where in 1994 the WHO put the where incidence peaked during the Middle
number of estimated cases at 1,167,900. Ages. It was probably introduced into the
In the United States about 100 to 200 new Americas with the early Spanish explora-
cases a year are reported each year to the tions and the slave trade. The first reference
Centers for Disease Control and Prevention to leprosy in the United States was in the
(CDC) (mostly among immigrants). Floridas in 1758. A hospital for the treat-
ment of leprosy was established in New
Orleans in 1785. The Louisiana Leper
How they get sick Home was established in 1894, near the
village of Carville, and in 1921 this facility
Leprosy is thought to be transmitted via became the National Leprosarium, a fed-
nasal secretions and contact with skin eral facility. It now exists as the Gillis W.
lesions, although the disease is not nearly Long Hansen’s Disease Center at Carville.

For the following discussion, refer to galactose, or to trehalose, which is a disac-

Fig. 1.11. The main waxy lipids are branched- charide of D-glucose. If the layer is bound to
chain hydroxy fatty acids called mycolic acids, trehalose, the compound, which is described
also found in Nocardia, Corynebacterium, next, is called cord factor. The arabinogalactan
and Rhodococcus. A major class of lipid, the itself is covalently bonded to acetyl or glycolyl
mycolic acids form a hydrophobic layer on the groups in the peptidoglycan via phosphodiester
external face of the cell wall. This layer can be bonds. There are variations in the size of indi-
esterified to a polysaccharide called arabinoga- vidual mycolic acids among the different spe-
lactan, which is a copolymer of arabinose and cies of Mycobacterium. In M. tuberculosis the
Free ebooks ==>
28 the physiology and biochemistry of prokaryotes

OH O C–CH–CH–C60H120(OH) mycolic acid
O C24H49
R2 R1 O
OH trehalose
mycolic acid HO
O C–CH–CH–C60H120(OH) mycolic acid

cord factor

Fig. 1.11 Chemical structures of mycolic acid and cord factor. Mycolic acid is a β-hydroxy fatty acid with
alkyl groups R1 attached to the α carbon and R2 attached to the β carbon. In Mycobacterium tuberculosis R1
is C24H49 and R2 is C60H120(OH). Cord factor is a glycolipid in the cell walls of M. tuberculosis. It is a disac-
charide of D-glucose [Glc(α1–α1)Glc], called trehalose, to which mycolic acid is attached via ester linkage to
the C6 hydroxyl of the sugar.

mycolic acid is a 3-hydroxy fatty acid with two take the Gram stain unless the wall lipids have
alkyl branches. The first branch (R1) is at C2 been removed with alkaline ethanol.
and is a hydrocarbon side chain of 24 carbons In summary, gram-positive cell walls have
(–C24H49). The second branch (R2) is at C3 and diverse types of neutral and acidic polysaccha-
is a hydroxylated hydrocarbon with 60 carbons rides, glycolipids, lipids, and other compounds
in the chain (–C60H120–OH). either free in the wall or covalently bound to the
In addition to mycolic acid–arabinogalactan, peptidoglycan. The functions of most of these
the M. tuberculosis cell wall has a mycolic acid– polymers are largely unknown, although some
containing cell surface glycolipid called cord may act as adhesins and/or presumably affect
factor (Fig. 1.11). Cord factor consists of the the permeability characteristics of the cell wall,
glucose disaccharide trehalose, to which are whereas others, such as mycolic acid deriva-
covalently bonded two molecules of mycolic tives in M. tuberculosis, contribute to resistance
acid. The characteristic M. tuberculosis colony properties and virulence.
appearance of long, intertwining cords, due
to the side-by-side interactions of long chains Gram-negative wall
of cells forming serpentine, ropelike rods, is The gram-negative cell wall is structurally and
generally attributed to cord factor—hence the chemically complex. It consists of an outer
name. As reviewed in ref. 87, cord factor is also membrane composed of lipopolysaccharide,
found in noncording mycobacteria. It has been phospholipid, and protein, and an underlying
suggested that cord factor is responsible for the peptidoglycan layer (Fig. 1.12). Between the
wasting, fever, and lung damage symptomatic outer and inner membrane (the cell membrane)
of tuberculosis. Cord factor is required for viru- is a compartment called the periplasm, wherein
lence, and it is lacking in avirulent mycobacte- the peptidoglycan lies.
ria. When injected intravenously into rabbits,
cord factor causes weight loss and granulomas 1. Lipopolysaccharide structure
in the liver and lungs.88 and function
When mycobacteria are appropriately Lipopolysaccharide (LPS) consists of three
stained, the dye (basic fuschin) is not removed regions: lipid A, core, and a repeating oligosac-
by dilute hydrochloric acid in ethanol (acid charide, sometimes called o-antigen or somatic
alcohol) because of the presence of the waxy antigen. The chemical structure and synthe-
lipids. Therefore these microorganisms are sis of LPS are described in Section 12.2. In the
called acid-fast bacteria. Acid-fast staining is an Enterobacteriaceae (e.g., E. coli), the lipopoly-
important diagnostic feature for the identifica- saccharide is confined to the outer leaflet of the
tion of Mycobacterium in clinical specimens, outer membrane and is arranged so that the
such as sputum. Acid-fast bacteria also do not lipid A portion is embedded in the membrane as
Free ebooks ==>
structure and function 29

Fig. 1.12 Schematic drawing of the gram-negative envelope. The outer membrane consists of lipopolysac-
charide, phospholipid, and proteins (most of which are porins). Underneath the outer membrane is the pep-
tidoglycan layer, which is noncovalently bonded to the outer membrane via murein lipoproteins, the lipid
portion of which is integrated into the inner leaflet of the outer membrane, and the protein portion is cova-
lently attached to the peptidoglycan. The cell membrane is composed of phospholipid and protein. The area
between the outer membrane and the cell membrane is called the periplasm. The wavy lines are fatty acid
residues, which anchor the phospholipids and lipid A into the membrane. Abbreviations: LPS, lipopolysac-
charide; O, oligosaccharide; C, core; A, lipid A; P, porin; PL, phospholipid; MLP, murein lipoprotein; pg,
peptidoglycan; Pr, protein; om, outer membrane; cm, cell membrane.

part of the lipid layer, and the core and oligosac- the enteric bacteria to have such a permeability
charide extend into the medium (Fig. 1.12). barrier because they live in the presence of bile
Mutants of E. coli that lack the oligosac- salts in the intestine. In fact, a common basis
charide experience loss of virulence, and it is for selective media for gram-negative bacteria
believed that the LPS can increase pathoge- is the resistance of these bacteria to bile salts
nicity. The lipid A portion of the LPS, which and/or hydrophobic dyes. The bile salts and
can have toxic effects when released from dyes inhibit the growth of gram-positive bac-
bacterial cells, is described as an endotoxin as teria but, because of the LPS, not gram-nega-
explained in note 89. Loss of most of the core tive bacteria. An example of such a selective
and oligosaccharide in E. coli and related bac- medium is eosin–methylene blue (EMB) agar,
teria is associated with increased sensitivity which is used for the isolation of gram-nega-
to hydrophobic compounds (e.g., antibiotics, tive bacteria because it inhibits the growth of
bile salts, and hydrophobic dyes such as eosin gram-positive bacteria. As explained in note
and methylene blue). This is because the LPS 90 and summarized in Section 17.5, the pres-
provides a permeability barrier to hydropho- ence of drug efflux pumps is another impor-
bic compounds. (A model for how this might tant reason that bacteria are resistant to many
occur is described later.) It is advantageous for antibiotics.
Free ebooks ==>
30 the physiology and biochemistry of prokaryotes

Apparently the outer membrane has low membrane is protein, of which there are several
permeability to hydrophobic compounds: the different kinds. One of the proteins is called the
phospholipids are confined primarily to the murein lipoprotein. This is a small protein with
inner leaflet of the outer envelope, whereas lipid attached to the amino-terminal end (Figs.
the LPS layer, which is a permeability bar- 1.13 and 1.14). The lipid end of the molecule
rier to hydrophobic substances, is in the outer extends into and binds hydrophobically with
leaflet. Since lipid A contains only saturated the lipids in the outer envelope. The protein end
fatty acids, the LPS presents a somewhat rigid of some of the molecules is covalently bound
matrix, and it has been suggested that this to the peptidoglycan, thus anchoring the outer
property, plus the tendency of the large LPS envelope to the peptidoglycan. In E. coli, about
molecules to engage in lateral noncovalent one-third of the murein lipoprotein is bound to
interactions, makes it difficult for hydropho- the peptidoglycan. Mutants unable to synthe-
bic molecules to penetrate between the LPS size the murein lipoprotein have unstable outer
molecules to the phospholipid layer. Mutants envelopes that bleb off into the medium at the
that lack a major region of the oligosaccharide cell poles and septation sites. Therefore, the
and core are more permeable to hydrophobic murein lipoprotein may play a structural role in
compounds because there is more phospho- keeping the outer membrane attached to the cell
lipid in the outer leaflet of these mutants; it is surface.
likely, as well, that there is less lateral interac- There are a small number of other outer mem-
tion among these incomplete LPS molecules. brane or cell membrane lipoproteins.91 These
The asymmetric distribution of phospholipid were discovered by chemically cross-linking the
to the inner leaflet of the outer envelope seems peptidoglycan in whole cells to closely associ-
to be an adaptive evolutionary response to ated proteins with a bifunctional cross-linking
hydrophobic toxic substances in the environ- reagent. (Note 92 explains how cross-linking
ment, such as the intestine of animals. Not all reagents work.) There is no evidence that these
gram-negative bacteria have an outer envelope additional lipoproteins are covalently bonded
with an asymmetric distribution of lipopoly- to the peptidoglycan, and in most cases their
saccharide and phospholipids, but many non- functions remain to be elucidated.
enterics as well as enterics do.
3. Porins and other proteins
2. Lipoproteins The major proteins in the outer envelope are
In addition to lipopolysaccharide and phos- called porins. The porins form small nonspe-
pholipid, a major component of the outer cific hydrophilic channels through the outer

Fig. 1.13 Murein lipoprotein. Attached to the amino-terminal cysteine is a diacylglyceride in thioether link-
age and a fatty acid in amide linkage. The lipid portion extends into the outer envelope and binds hydrophobi-
cally with the fatty acids in the phospholipids and lipopolysaccharide. The carboxy-terminal amino acid is
lysine, which can be attached via an amide bond to the carboxyl group of diaminopimelic acid (DAP) in the
peptidoglycan. The murein lipoprotein therefore holds the outer envelope to the peptidoglycan.
Free ebooks ==>
structure and function 31

Fig. 1.14 Pseudopeptidoglycan as found in Archaea. Pseudopeptidoglycan resembles peptidoglycan in being

a cross-linked glycopeptide. It differs from peptidoglycan in the following ways: (1) N-acetyltalosaminuronic
acid replaces N-acetylmuramic acid. (2) The glycosidic linkage is β-1,3 instead of β-1,4. (3) There are no
D-amino acids. Abbreviations: G, N-acetylglucosamine; T, N-acetyltalosaminuronic acid. The peptide sub-
unit is enclosed in dashed lines.

envelope, allowing the diffusion of low molecu- bacteria, although they are not all identical. As
lar weight (<600 Da) neutral and charged sol- mentioned earlier, E. coli regulates the amounts
utes, such as sugars and ions. The channels are of the various porins according to growth con-
necessary to allow passage of small molecules ditions. This is discussed in Chapter 18.
into and out of the cell. E. coli has three major Since the porins exclude molecules with
porins: OmpF, OmpC, and PhoE. Each porin molecular weights larger than 600 Da, one
makes a separate channel. Thus, there are would expect to find other proteins in the outer
OmpF, OmpC, and PhoE channels. membrane that facilitate the translocation of
The OmpC channel is approximately 7% larger solutes across the outer membrane. This
smaller than the OmpF channel and is expected is the case. E. coli has an outer membrane pro-
to make the outer envelope less permeable to tein called the LamB protein that forms channels
larger molecules. Both OmpF and OmpC are for the disaccharide maltose and for maltodex-
present under all growth conditions, although trins. Other examples in the outer membrane
the ratio of the smaller OmpC to the larger of E. coli facilitate the transport of vitamin B12
OmpF increases in high-osmolarity media and (BtuB), nucleosides (Tsx), and several other
at high temperature. The increased amounts of solutes.
OmpC relative to OmpF presumably also occur
in the intestine, where osmolarity and tempera- Archaeal cell walls
ture are higher than in lakes and streams that Archaeal cell walls are not all alike and are very
also harbor E. coli. This may confer an advan- different from bacterial cell walls. For example,
tage to the enterics because the smaller OmpC no archaeal cell wall contains peptidoglycan.
channel should present a diffusion barrier to Archaeal cell walls may be either pseudopeptido-
toxic substances in the intestine, whereas the glycan, polysaccharide, or protein (the S layer).
larger OmpF channel should be advantageous Pseudopeptidoglycan (also called pseudo-
in more dilute environments outside the body. murein) resembles peptidoglycan in consist-
The protein PhoE is produced only under ing of glycan chains cross-linked by peptides
conditions of inorganic phosphate limitation. (Fig. 1.14). However, the resemblance stops
This is because PhoE is a channel for phosphate here. In pseudopeptidoglycan, although one of
(and other anions) whose synthesis under lim- the sugars is N-acetylglucosamine as in pepti-
iting phosphate conditions reflects the need doglycan, the other is N-acetyltalosaminuronic
to bring more phosphate into the cell. Porins acid instead of N-acetylmuramic acid.
appear to be widespread among gram-negative Furthermore, the sugars are linked by a β-1,3
Free ebooks ==>
32 the physiology and biochemistry of prokaryotes

glycosidic linkage rather than a β-1,4, and the The literature discusses whether the periplasm
amino acids in the peptides are L-amino acids has zones of adhesion (called Bayer’s patches)
rather than D-amino acids. (The latter are pres- between the inner and outer membranes.94,95
ent in peptidoglycan; compare Figs. 1.14 and Whether zones of adhesion are seen depends
12.2.) Thus, pseudopeptidoglycan is very dif- upon how the cells are prepared for electron
ferent from peptidoglycan. microscopy. One school of thought holds that
the zones are artifacts of fixation; the counter-
1.2.4 Periplasm argument is that the zones are seen only if the
Lying between the cell membrane and the proper techniques are employed to preserve the
outer membrane of gram-negative bacteria is rather fragile adhesion sites. This is an impor-
a separate compartment called the periplasm93 tant issue because the zones of adhesion have
(Fig. 1.15). It appears as a space in electron been postulated to be sites of the translocation
micrographs of thin sections of cells but should apparatuses that export lipopolysacharide,
be considered to be an aqueous compartment polysaccharide (capsule), and protein through
containing protein, oligosaccharides, salts, the inner and outer membranes.
and the peptidoglycan. It seems that the pep- The periplasm should be considered to be a
tidoglycan and oligosaccharides may exist in a cellular compartment with specialized activities.
hydrated state, forming a periplasmic gel.94 These activities include oxidation–reduction

Fig. 1.15 A model of the periplasm in E. coli. The outer region of the periplasm is thought to consist of
cross-linked peptidoglycan attached to the outer envelope via lipoprotein (LP) covalently bound to the pep-
tidoglycan. The inner region of the periplasm (approaching the cell membrane) is believed to consist of less
cross-linked peptidoglycan chains and oligosaccharides that are hydrated and form a gel. The gel phase is
thought to contain periplasmic proteins (e.g., A and B). Thus A might be a periplasmic enzyme and B a solute-
binding protein that interacts with a membrane transporter (C); D and E are integral membrane proteins, per-
haps enzymes. The outer envelope is depicted as consisting of lipopolysaccharide, porins, and specific solute
transporters (receptors), which require a second protein (TonB) for uptake. Source: Ferguson, S. J. 1991. The
periplasm, pp. 311–339. In: Prokaryotic Structure and Function, A New Perspective. S. Mohan, C. Dow, and
J. A. Coles (Eds.). Cambridge University Press, Cambridge.
Free ebooks ==>
structure and function 33

reactions (Chapters 5 and 13), osmotic regu- carried into the cell via specific inorganic phos-
lation (Chapter 16), solute transport (Chapter phate transporters.
17), protein secretion (Chapter 18), and hydro-
5. Detoxifying agents
lytic activities such as those mediated by phos-
Some periplasmic enzymes are detoxifying
phatases and nucleases. The phosphatases
agents. For example, the enzyme to degrade pen-
and nucleases degrade organophosphates and
icillin (β-lactamase) is a periplasmic protein.
nucleic acids that might enter the periplasm
from the medium and transport the hydrolytic 6. TonB protein96
products into the cell. An interesting periplasmic protein anchored to
the cell membrane in E. coli is the TonB pro-
Periplasmic components tein. (See Fig. 1.15.) It is known that the pro-
The periplasm is chemically complex and car- tein is required for the uptake of several solutes
ries out diverse functions. The following list of that do not diffuse through the porins; rather,
components and their functions, which reflects they require specific transport systems (also
the importance of the periplasm, is simply a called receptors or TonB-dependent trans-
partial inventory emphasizing the periplasmic porters, or TBDTs) in the outer envelope. All
functions about which most is known. these solutes have molecular weights larger
than 600 Da, which is the upper limit for mol-
1. Oligosaccharides
ecules that enter via the porins. Examples of
The oligosaccharides in the periplasm are
solutes with specific outer membrane receptors
thought by some to be involved in osmotic reg-
that require TonB for uptake are iron sidero-
ulation of the periplasm because their amounts
phores and vitamin B12 (cobalamin). (For an
decrease when the cells are grown in media of
explanation of iron siderophores, see note 97.)
high osmolarity. This complex subject is dis-
Interestingly, these solutes are brought into
cussed more fully in Section 16.2.
the periplasm against a large concentration
2. Solute-binding proteins gradient, sometimes 103 times higher than the
Solute-binding proteins in the periplasm assist concentration outside the cell. In a way that
in solute transport by binding to solutes (e.g., is not understood, the TonB protein couples
sugars and amino acids that have entered the the electrochemical energy (the proton motive
periplasm through the outer envelope) and force, Δp) in the cell membrane to the uptake
delivering the solutes to specific transporters of certain solutes through the outer envelope
(carriers) in the cell membrane. and into the periplasm.98-101 TonB is thought
to be an energy transducer. One suggestion
3. Cytochromes c
is that TonB is energized by the electrochemi-
Some of the enzymes in the periplasm are cyto-
cal potential that exists across bacterial cell
chromes c that oxidize carbon compounds or
membranes and that in the energized state,
inorganic compounds and deliver the electrons
TonB causes a conformational change in the
to the electron transport chain in the cell mem-
outer membrane receptor protein that results
brane. These oxidations are called periplasmic
in translocation of the solute (e.g., vitamin B12)
oxidations. There are other oxidoreductases
or ligand-bound solute (e.g., iron–siderophore
in the periplasm as well, but the various cyto-
complexes) through the receptor channel into
chromes c are very common.
the periplasm.101 Accessory proteins in the cell
4. Hydrolytic enzymes membrane, which in E. coli are called ExbB
Hydrolytic enzymes in the periplasm degrade and ExbD, interact with TonB and may use
nutrients to smaller molecules that can be trans- the proton motive force Δp (i.e., uptake of
ported across the cell membrane by specific H+ through ExbB/D) to convert TonB into an
transporters. For example, the enzyme amylase energized conformation that somehow drives
is a periplasmic enzyme that degrades oligo- the uptake of material through transport-
saccharides to simple sugars. Another exam- ers in the outer membrane. This concept was
ple is alkaline phosphatase, which removes reviewed in 2003, in a paper that also discusses
phosphate from simple organic phosphate two experimentally based models designed to
monoesters. The inorganic phosphate is then explain the mechanism by which TonB acts as
Free ebooks ==>
34 the physiology and biochemistry of prokaryotes

an energy transducer.101 One model proposes 1.2.5 Cell membrane

that TonB in an energized conformation leaves We now come to what is certainly the most
the ExbB/D complex in the cell membrane and functionally complex of the cell structures, the
moves to the transporter in the outer mem- cell membrane. The cell membrane is responsi-
brane. It then returns to the ExbB/D complex ble for a broad range of physiological activities
to be reenergized. This has been called the including solute transport, electron transport,
“shuttle” model. An alternative model pro- photosynthetic electron transport, the estab-
poses that one part of TonB remains associated lishment of electrochemical gradients, ATP
with the ExbB/D complex in the cell membrane synthesis, biosynthesis of lipids, biosynthesis
and one part reaches across to the transporter of cell wall polymers, secretion of proteins, the
in the outer membrane. secretion and uptake of intercellular signals,
and responses to environmental signals. To
Is there a periplasm in gram-positive refer to the cell membrane simply as a lipopro-
bacteria? tein bilayer does not do justice to the machinery
In the past it has been assumed that gram- embedded in the lipid matrix, a complex mosaic
positive bacteria lack a space between the cell of parts whose structure and interactions at the
membrane and the cell wall equivalent to the molecular level are not well understood.
periplasm of gram-negative bacteria. This is As expected, the protein composition of cell
because thin sections of gram-positive cells do membranes is complex. There can be more than
not indicate that such a space exists, and of 100 different proteins. Many of the proteins
course gram-positive bacteria do not have an are clustered in functional aggregates (e.g.,
outer envelope. Evidence suggests, however, the proton translocating ATPase, the flagella
that perhaps gram-positive bacteria do have a motor, electron transport complexes, certain
compartment analogous to the periplasm found of the solute transporters). At the molecular
in gram-negative bacteria.102,103 level, the membrane is certainly a complex and
The evidence in favor of the existence of busy place. What follows is a general descrip-
a periplasm in Bacillus subtilis includes the tion of the membrane, without reference to its
release of putative periplasmic proteins after microheterogeneity.
protoplasts of the cells have been made with
lysozyme. (For a definition of protoplasts and Bacterial cell membranes
how they are stabilized, read note 104.) Proteins Bacterial cell membranes consist primarily of
solubilized by removing the cell wall during pro- phospholipids and protein in a fluid mosaic
toplast formation may include proteins resident structure in which the phosphlipids form a
in the area between the cell membrane and the bilayer (Fig. 1.16). The structure is said to be
cell wall. The proteins released by protoplast fluid because there is extensive lateral mobility of
formation include nucleases, which are distinct bulk proteins and phospholipids. Nevertheless,
from cytoplasmic nucleases. certain protein aggregates (e.g., complex solute
Additional evidence has been obtained by transporters and electron transport aggregates)
using ultrarapid freezing and electron micros- remain as aggregates within which the proteins
copy (cryo–transmission electron microscopy, interact to catalyze sequential reactions.
or cryo-TEM) of frozen–hydrated thin sections
of B. subtilis. When this was done, the area out- 1. The lipids
side the cell membrane was seen to be bipartite, The phospholipids are fatty acids esterified to
consisting of a low-density 22 nm region sur- two of the hydroxyl groups of phosphoglycer-
rounded by a 33 mm high-density outer wall ides (Fig. 1.17). The structure and synthesis of
zone, which is thought to consist of peptido- phospholipids is described in detail in Section
glycan, teichoic acid, and cell wall proteins.105 10.1.2. The third hydroxyl group in the glyc-
Further studies must be undertaken to elucidate erol backbone of the phospholipid is covalently
the exact nature of this putative periplasm, bound to a substituted phosphate group, which
including its contents and the relationships of makes one end of the molecule very polar owing
the proteins found therein with the cell mem- to a negative charge on the ionized phosphate
brane and cell wall, as well as the similarities to group. Because the phospholipids are polar at
the gram-negative periplasm. one end and nonpolar at the other end (the end
Free ebooks ==>
structure and function 35

Fig. 1.17 Phospholipids have both a polar and a non-

polar end. (A) Phospholipid with two fatty acids (R)
esterified to glycerol. The phosphate is conjugated
to X, which determines the type of phospholipid. In
Fig. 1.16 Model of the cell membrane showing bacteria, X is usually serine, ethanolamine, a deriva-
bimolecular lipid leaflets and embedded proteins; tive of glycerol, or a carbohydrate derivative. See
the phospholipid molecules are interacting with one Section 10.1.3 for a more complete description of
another via their hydrophobic (apolar) “tails.” The bacterial phospholipids. (B) Schematic drawing of a
hydrophilic (polar) “heads” of the phospholipids face phospholipid showing the polar (circle) and nonpo-
the outside of the membrane, where they interact with lar (straight lines) regions.
proteins and ions. Proteins can span the membrane
or be partially embedded. Source: Singer, S. J., and
G. L. Nicolson. 1972. The fluid mosaic model of the
structure of cell membranes. Science 175:720–731. across the membrane through or on special pro-
Copyright 1972 by the Association of Academies of tein transporters that bridge the phospholipid
Science. Reprinted with permission from AAAS. bilayer. These modes are discussed in the context
of solute transport in Chapter 17. (However,
the lipid bilayer is permeable to water mole-
cules, gases, and small hydrophobic molecules.)
with the fatty acids), they are said to be amphi-
An important consequence of the lipid matrix is
pathic, able to spontaneously aggregate while
that ions do not freely diffuse across the mem-
their nonpolar fatty acid regions interact with
brane unless they are carried on or through
each other by hydrophobic bonding; their polar
protein transporters. Because of this, the mem-
phosphorylated regions, on the other hand, face
brane is capable of holding a charge that is due
the aqueous phase, where ionic interactions
to the unequal transmembrane distribution of
occur with cations, water, and polar groups on
ions. This is discussed in Chapter 5.
proteins. Phospholipids accomplish all this by
spontaneously forming lipid bilayers in water 4. Aquaporins (water channels)
solutions or in cell membranes. Although the lipid bilayer allows rapid equili-
bration of water, there do exist in E. coli and
2. The proteins
other bacteria water channels that are similar
There are two classes of proteins in membranes,
to the aquaporins found in eukaryotes; these
integral and peripheral. Integral proteins are
water channels, called aquaporins, enhance
embedded in the membrane and bound to the
the rapid equilibration of water across the cell
fatty acids of the phospholipids via hydrophobic
membrane. (Reviewed in ref. 105.) The gene
bonding. They can be removed only with deter-
coding for the water channel protein in E.
gents or solvents. Peripheral proteins, attached
coli is aqpZ. The expression of the aqpZ gene
at membrane surfaces to the phospholipids by
under different extracellular osmolarity con-
ionic interactions, can be removed by washing
ditions and its requirement for viability have
the membrane with salt solutions. The insertion
been investigated.106 Null mutants of aqpZ
of the proteins into the membrane during mem-
are viable, although the colonies are smaller
brane synthesis is discussed in Section 18.2.
than the wild-type strain. Interestingly, when
3. Permeability E. coli is grown in media of high osmolarity,
The phospholipid bilayer acts as a permeabil- the synthesis of the aquaporin channels is
ity barrier to virtually all water-soluble mol- repressed. This may help protect the cell from
ecules. Thus most solutes diffuse or are carried hypo-osmotic stress in the event of a sharp
Free ebooks ==>
36 the physiology and biochemistry of prokaryotes

decrease in the osmolarity of the external facilitating the uptake of organic osmolytes such
medium. as proline and betaine. Recently it was reported
that ProP localizes to the membrane at the poles
5. Mechanosensitive (MS) channels
of cells because of the presence there of a lipid
For reviews of mechanosensitive channels,
called cardiolipin, which is more highly concen-
read refs. 105 and 107 through 109. As will be
trated in the cytoplasmic membrane near the
explained in Section 16.2, bacteria adjust the
cell poles.111 This is very interesting because it
internal osmotic pressure so that it is always
points to an important role of certain lipids in
higher than the external osmotic pressure. This
localizing membrane proteins to specific mem-
keeps water flowing into the cell via osmosis and
brane locations.
maintains the high internal turgor pressure that
is important for growth. The internal osmotic
Archaeal cell membranes
pressure is kept high by the accumulation of cer-
1. The lipids
tain solutes such as K+, glutamate, glutamine,
Archaeal membrane lipids differ from those
proline, trehalose, and betaine.
found in bacterial membranes.112-114 The
One consequence of maintaining a high inter-
archaeal lipids consist of isoprenoid alcohols
nal osmotic pressure is that a sudden decrease
(either 20 or 40 carbons long), linked via ether
in the external osmotic pressure can endanger
bonds, in contrast to the ester linkages in eubac-
the cell by promoting a sudden increase in the
teria, to one glycerol to form monoglycerol
influx of water, leading to overexpansion of the
diethers or to two glycerols to form diglycerol
cell wall and subsequent lysis of the cell. This
tetraethers. These are illustrated in Fig. 1.18.
is sometimes referred to as hypo-osmotic stress
The synthesis of these lipids is described later,
or hypo-osmotic shock. The cells are protected
in Section 10.1.3. (Recall that bacterial glyc-
from this form of destruction by mechanosensi-
erides are fatty acids esterified to glycerol.
tive (MS) channels that open under conditions
Refer to Fig. 1.16 for a comparison.) The C20
of hypo-osmotic stress and provide a means for
alcohol is a fully saturated hydrocarbon called
internal solutes to rapidly exit the cell, thus low-
phytanol. The C40 molecules are two phytanols
ering the internal osmotic pressure. (See note
linked together head to head in the diglycerol
110 for an explanation.) Mutants that do not
tetraether lipids. Thus, the lipids are either
have MS channels lyse when subjected to hypo-
phytanylglycerol diethers or diphytanyl diglyc-
osmotic stress.
erol tetraethers. The diethers and tetraethers
Mechanosensitive channels are present in
occur in varying ratios depending upon the bac-
most bacteria as well as archaea. E. coli has
terium. For example, there may be from 5 to 25
three such channels: a large channel called
different lipids in any one cell. This is really quite
MscL, a small channel called MscS, and a
a diverse mix, and it can be contrasted to the
“mini” channel called MscM (L refers to large,
lipid complement of a typical bacterium, which
S to small, and M to mini).
has only four or five different phospholipids.
6. Osmosensory transporters The diversity of the archaeal lipids is due to the
As explained in Chapter 16, bacteria adapt different polar head groups that exist, as well
to high-osmolarity media by increasing the as to the mix of core lipids to which the head
intracellular concentrations of certain solutes groups are attached (Fig. 1.18). Although the
called osmolytes. This raises the cytoplasmic polar head group is responsible for the polarity
osmolarity so that water does not rush out of of most phospholipids, there is some polarity at
the cell into the more concentrated solution. As one end of the archaeal lipids without a polar
stated in Section 16.2.3 (Turgor pressure and head group because of the free hydroxyl group
its importance for growth), it is very important on the glycerol. (Recall that hydroxyl groups are
for bacteria to maintain an internal osmolar- capable of forming hydrogen bonds with water
ity higher than the external osmolarity so that and proteins.) It is usually stated that the ether
turgor pressure within the cell is maintained. linkages, which are more stable to hydrolytic
E. coli has an osmosensory transporter called cleavage, are an advantage over ester linkages
ProP; this membrane protein senses increas- in the acidic and thermophilic environments in
ing osmolarity in the medium and responds by which many archaea live.
Free ebooks ==>
structure and function 37

Fig. 1.18 Major lipids of Methanobacterium thermoautotrophicum: (A) glycerol diether (archaeol),
(B) diglycerol tetraether (caldarchaeol), (C) a glycolipid (gentiobiosyl archaeol), (E) a phospholipid (archaeti-
dyl–X), where X can be inositol, serine, or ethanolamine, (F) a phospholipid (caldarchaetidyl–X), (G) a
phosphoglycolipid (gentiobiosyl caldarchatidyl–X). Source: Nishihara, M., H. Morii, and Y. Koga. 1989.
Heptads of polar ether lipids of an archaebacterium, Methanobacterium thermoautotrophicum: structure
and biosynthetic relationship. Biochemistry 28:95–102. Reprinted with permission from Nishihara, M., H.
Morii, and Y. Koga. Copyright 1989 American Chemical Society.
Free ebooks ==>
38 the physiology and biochemistry of prokaryotes

2. The proteins generally believed to be derived from invagi-

There is little information regarding archaeal nations of chemically modified areas of the
membrane proteins. It is known, however, that cell membrane. Connections to the cell mem-
in bacteriorhodopsin and halorhodopsin in brane are not always seen, however, and it is
Halobacterium, the conformational array in the unknown whether the intracytoplasmic mem-
cell membrane is dependent upon interaction branes are derived from an invagination of the
with polar membrane lipids.113 The functions cell membrane or are synthesized independently
of these two proteins are discussed in Sections of the cell membrane (e.g., the thylakoids of
4.8.4 and 4.9. cyanobacteria). A few prokaryotes with intra-
cytoplasmic membranes and their physiological
3. The membrane
roles are listed.
The thermoacidophilic archaea and some
methanogens have tetraether glycerolipids in 1. Methanotrophs
the cell membrane. These lipids have a polar Bacteria that grow on methane as their sole
head group at both ends and span the mem- source of carbon (methanotrophs) possess
brane, forming a lipid monolayer (Fig. 1.19). intracytoplasmic membranes that are suggested
This is the only known example of a membrane to function in methane oxidation. Methane oxi-
having no midplane region. Since there is no dation is discussed later, in Section 14.2.1.
midplane region, the lipid monolayer is more
2. Nitrogen fixers
resistant to levels of heat that would disrupt the
Bacteria for which nitrogen gas serves as a source
hydrophobic bonds holding the two lipids in
of nitrogen use an oxygen-sensitive enzyme
the lipid bilayer together. The increased resis-
called nitrogenase to reduce the nitrogen to
tance to heat of the lipid monolayer may confer
ammonia, which is subsequently incorporated
an advantage to organisms living at high tem-
into cell material. Many of these organisms have
peratures. However, it cannot be claimed that
extensive intracytoplasmic membranes. One
diether lipids or tetraether lipids are a specific
such nitrogen-fixing bacterium is Azotobacter
adaptation to high temperatures, although they
vinelandii, whose intracytoplasmic membranes
may be advantageous in these environments.
increase with the degree of aeration of the
This is because some mesophilic methanogens
have tetraether lipids, whereas two extremely
thermophilic archaea, Methanopyrus kandleri
and Thermococcus celer, do not have tetraether
1.2.6 Cytoplasm
The cytoplasm is defined as everything enclosed
by the cell membrane. Cytoplasm is a viscous
material containing a heavy concentration of
protein (100–300 mg/mL),115 salts, and metab-
olites. In addition, there are large aggregates of
protein complexes designed for specific meta-
bolic functions, various inclusions, and highly
condensed DNA. Intracytoplasmic membranes
are also present in many prokaryotes. The
soluble part of the cytoplasm is called the cyto-
sol. We will begin with the intracytoplasmic Fig. 1.19 Lipid layer with membrane proteins (shaded
areas) in archaebacteria membranes. The glycerol
diethers form a lipid bilayer, and the tetraethers
form a monolayer. Some archaea (e.g., the extreme
Intracytoplasmic membranes halophiles) contain only the diethers. Most of the
Many prokaryotes have intracytoplasmic sulfur-dependent thermophiles have primarily the
membranes that have specialized physiological tetraethers, with only trace amounts of the diethers.
functions.116 Intracytoplasmic membranes are Many methanogens have significant amounts of both
often connected to the cell membrane and are the di- and tetraethers.
Free ebooks ==>
structure and function 39

culture. Since respiratory activity is localized in organisms to float in lakes and ponds at depths
the membranes, it is probable that an important that support growth because of favorable light,
role for Azotobacter intracytoplasmic mem- temperature, or nutrients. For example, the
branes is to increase the cellular respiratory green sulfur bacterium Pelodictyon phaeoclath-
activity, to provide more ATP for nitrogen fixa- ratiforme forms gas vesicles only at low light
tion and to remove oxygen from the vicinity of intensities.117 Perhaps this allows the bacteria
the nitrogenase. Nitrogen fixation is discussed to float at depths where the light is optimal for
in Section 13.3. photosynthesis. Many bacteria and cyanobac-
teria with gas vesicles are plentiful in stratified
3. Nitrifiers
freshwater lakes, but they are not as abundant
Intracellular membranes are also found in nitri-
in isothermally mixed waters. Other prokary-
fying bacteria (i.e., bacteria that oxidize ammo-
otes containing gas vesicles (e.g., the halophilic
nia and nitrite as the sole source of electrons:
archaeon Halobacterium) live in hypersaline
Nitrosomonas, Nitrobacter, Nitrococcus).
waters, and a few marine species of cyanobacte-
Several of the enzymes that catalyze ammonia
ria belonging to the genus Trichodesmium have
and nitrite oxidation are in the membranes.
gas vesicles. When gas vesicles are collapsed by
This is discussed in Section 13.4.
experimentally subjecting cells to high hydro-
4. Phototrophs static pressure or turgor pressure, the cells are
In bacteria that use light as a source of energy no longer buoyant and sink. Collapsed vesicles
for growth (phototrophs), the intracytoplasmic do not recover, and the cells acquire gas-filled
membranes are the sites of the photosynthetic vesicles only by de novo synthesis of new vesi-
apparatus. The membrane structure varies: flat cles. Thus, during synthesis of the vesicles, water
membranes, vesicles, flat sacs (thylakoids in is somehow excluded, presumably because of
cyanobacteria), and tubular invaginations of the hydrophobic nature of the inner protein
the cell membrane (photosynthetic bacteria). surface.
See the discussion of photosynthesis and pho-
tosynthetic membranes in Chapter 6, especially 2. Carboxysomes
Fig. 6.16. Bacteria that obligately grow on CO2 as their
sole or major source of carbon (strict autotro-
Inclusion bodies, multienzyme aggregates, phs) sometimes have large (100 nm) polyhedral
and granules protein-walled microcompartments called car-
boxysomes.118, 119 These inclusions have been
Certain bacteria contain specialized compart-
observed in nitrifying bacteria, sulfur oxidizers,
ments in the cytoplasm. Some researchers
and cyanobacteria. The distribution appears to
refer to these entities as inclusion bodies. They
be species specific (e.g., not all sulfur oxidizers
are not surrounded by a lipid bilayer–protein
have carboxysomes). However, it should be
membrane as are organelles in eukaryotic cells,
pointed out that most oceanic microorganisms
although they can have a membrane or coat. In
addition, there are numerous large aggregates that fix carbon dioxide, including all cyanobac-
teria, do so with carboxysomes. In these organ-
and multienzyme complexes in all bacteria.
isms ribulose-1,5-bisphosphate carboxylase
1. Gas vesicles (RuBP carboxylase), the enzyme in the Calvin
Aquatic bacteria such as cyanobacteria, certain cycle that incorporates CO2 into organic car-
photosynthetic bacteria, some nonphotosyn- bon, is stored in carboxysomes. The enzyme
thetic bacteria, and certain archaea have gas is discussed in Section 14.1.1. Although many
vesicles surrounded by a simple protein coat autotrophs lack carboxysomes, it has been
consisting primarily of gas vesicle protein A suggested that an advantage to having them is
(GsvA), a small hydrophobic protein that is that the RuBP carboxylase is sequestered there,
highly conserved among the diverse groups of where the concentration of CO2 is kept high.
organisms. Gas vesicles are hollow, spindle- This may be due to the carbonic anhydrase
shaped structures about 100 nm long, filled associated with the shell of the carboxysome.
with gas in equilibrium with the gases dissolved The carbonic anhydrase catalyzes the conver-
in the cytoplasm. The gas vesicles allow the sion of HCO–3 to CO2.120
Free ebooks ==>
40 the physiology and biochemistry of prokaryotes

3. Chlorosomes Magnetosomes should be thought of as a

Green sulfur photosynthetic bacteria (e.g., navigational device or a magnetic compass that
Chlorobium) have ellipsoid inclusions called orients the bacteria with the Earth’s magnetic
chlorosomes (formerly called chlorobium field so that they swim in a particular direction,
vesicles) that lie immediately underneath the a behavior called magnetotaxis. In the Northern
cytoplasmic membrane. The chlorosomes are Hemisphere the geomagnetic north points
surrounded by a nonunit membrane of galac- down at an angle, and magnetotactic bacteria in
tolipid, with perhaps some protein. At one time the Northern Hemisphere that swim toward the
it was believed that such vesicles were found geomagnetic north swim deeper into the water.
only in the green sulfur photosynthetic bacteria. In the Southern Hemisphere the geomagnetic
However, similar vesicles have been found in the north points up at an angle, and magnetotactic
green photosynthetic bacterium Chloroflexus. bacteria in the Southern Hemisphere are “south
The major light-harvesting photopigments are seeking.” (At the equator, north-seeking and
located in the chlorosomes, whereas the photo- south-seeking magnetotactic bacteria coexist.)
synthetic reaction centers are in the cell mem- One way to demonstrate this in the laboratory
brane. This means that during photosynthesis in is to place a small drop of water on a microscope
these organisms, light is absorbed by pigments coverslip and place the south pole of a small bar
in the chlorosomes and energy is transmitted to magnet 1–2 cm away from the drop. The south
the reaction centers in the cell membrane, where pole of the magnet corresponds to geomagnetic
photosynthesis takes place. In photosynthetic north. It is clear that swimming is important for
bacteria that do not have chlorosomes, the this to occur, because dead cells are not pulled by
light-harvesting pigments surround the reac- a magnetic field. A generally accepted model is
tion centers in the cell membrane. The structure that the earth’s magnetic field is such that when
and function of chlorosomes are discussed in the bacteria are swimming, the magnetosomes
Section 6.6. allow orientation in the magnetic field and guide
them to swim downward in their natural aque-
4. Magnetosomes
ous habitat. All the magnetotactic bacteria are
For a review of magnetosomes in bacteria, see
microaerophilic or anaerobic, and it is thought
refs. 121 through 125.
that magnetotaxis to lower levels is beneficial
Magnetosomes and magnetotactic bacteria. because there is less oxygen at greater depths.
Certain marine and freshwater bacteria have However, although this is the case for most MB
chains of membrane-bound organelles called that have been observed, exceptions do occur.
magnetosomes, which originate as invagina- There has been a report of a population of mag-
tions of the cell membrane. The bacteria are netotactic bacteria in the Northern Hemisphere
referred to as magnetotactic bacteria (MB) or that responds to high oxygen levels by swim-
(MTB) because the magnetosomes influence ming toward geomagnetic south in the drop
the direction of swimming with respect to the assay just described, rather than in the direction
earth’s magnetic field. of geomagnetic north. Bacteria displaying south
Magnetosomes are strings of crystals of iron polarity co-occurred in water samples from the
oxide, magnetite (Fe3O4), or in some cases iron same aquatic redox environment with bacteria
sulfide, greigite (Fe3S4), each one of which is sur- displaying north polarity. Thus, more has to be
rounded by a specialized, complex membrane learned about magnetotactic behavior.
containing a lipid bilayer of phospholipids, and The magnetotactic bacteria are a very diverse
proteins. For a review of the proteins associ- group of motile organisms and belong to a
ated with magnetosomes and their proposed wide range of phylogenetic groups, including
functions, refer to ref. 119. The magnetosomes α-Proteobacteria, β-Proteobacteria, δ-Proteo-
are positioned in the cell by cytoskeletal fila- bacteria, and Nitrospira. At this time, magne-
ments composed of a protein called MamK, tosomes have not been found in gram-positive
which flanks the magnetosomes. MamK is a bacteria or archaea.
homologue of the bacterial actinlike cytoskel-
etal protein called MreB. (See the discussion 5. Granules and globules
of cytoskeletal proteins and their functions in Bacteria often contain cytoplasmic granules
Section 1.2.7.) whose content varies with the bacterium. Many
Free ebooks ==>
structure and function 41

bacteria store a lipoidal substance called poly- specific sequences and also bends DNA. The
β-hydroxybutyric acid (PHB) in the granules DNA-binding proteins, HU, H-NS, and Fis,
as a carbon and energy reserve. Other bacteria are called histonelike because they resemble
may store glycogen for the same purpose. Other histone in their electrostatic charge, binding
granules found in bacterial cells can include to DNA, and because they have a low molecu-
polyphosphate and, in some sulfur-oxidizing lar weight and a high copy number. However,
bacteria, elemental sulfur globules. unlike HU itself, the other histonelike proteins
are not similar to histone in their amino acid
6. Ribosomes composition or structure. For example, the
Ribosomes are the sites of protein synthesis. histonelike protein H-NS binds to curved or
They are ribonucleoprotein particles, approxi- bent DNA. The structure of H-NS, its role in
mately 22 nm by about 30 nm, or about the DNA superstructure, and its role in gene regu-
size of the smallest viruses. Ribosomes consist lation is reviewed in ref. 127. It is believed that
of over 50 different proteins and three differ- the DNA-binding structural proteins facilitate
ent types of RNA (23S, 16S, and 5S). Bacterial bending of DNA and/or have a high affinity for
ribosomes sediment in a centrifugal field at a bent DNA, or restraining of DNA supercoils in
characteristic velocity of 70S, as opposed to the nucleoid, and for the compact structure of
eukaryotic cytosolic ribosomes, which are 80S. the nucleoid.128
Bacterial ribosomes are very similar regardless The DNA-binding proteins are important
of the bacterium. However, there are some dif- not only for nucleoid structure but also for the
ferences from archaeal ribosomes, which are regulation of expression of certain genes. See
also 70S. These differences were described in the discussion of the H-NS and Fis proteins in
Section 1.1.1. Section 2.2.2 (ribosomal RNA genes), the dis-
7. The nucleoid cussion of H-NS and IHF in Section 19.11.5
The site of DNA and RNA synthesis is the nucle- (virulence genes), and the discussion of the
oid, an amorphous mass of DNA unbounded by regulatory role of IHF in Section 19.3.5, as well
a membrane, lying approximately in the center as Section 20.4 (Ntr regulon) and Section 19.7
of the cell. Faster-growing bacteria may contain (porin biosynthesis).
more than one nucleoid, but each nucleoid has Another protein important for DNA struc-
but one chromosome, and all the chromosomes ture is DNA gyrase (topoisomerase II), which
are identical. The DNA is very tightly coiled. is responsible for negative supercoiling, that is,
Indeed, if the DNA from E. coli were stretched the twisting of the double helix about its axis in
out, it would be 500 times longer than the cell! the direction opposite to the right-handed dou-
ble helix. Cellular DNA is mostly in a negative
DNA-binding proteins influence nucleoid supercoil and otherwise would be much more
structure and gene regulation. The DNA in the difficult to unwind to obtain single strands for
nucleoids is bound to several proteins as well replication and transcription. There is also an
as to nascent chains of RNA. Of course, among enzyme, topoisomerase I, that removes negative
the proteins bound to the DNA are RNA poly- supercoils.
merase molecules engaged in transcription; but
several other proteins can be present as well. 8. Multienzyme complexes
When nucleoids are isolated from E. coli It should not be thought that the enzymes in
under low salt conditions (thus preserving the the cytoplasm are a random mixture of pro-
attachment of proteins to the DNA), membrane teins. There are many examples of enzymes in
proteins are present as well as the DNA-binding the same pathway forming stable multienzyme
proteins HU, IHF, H-NS, and Fis. HU, often complexes, reflecting strong intermolecular
referred to as a histonelike protein because it has bonding.129 For example, pyruvate dehydroge-
physical properties and an amino acid composi- nase from E. coli is a complex of three different
tion similar to eukaryotic histones, is the major enzymes, each present in multiple copies (50
DNA-binding protein present in E. coli.126 It proteins total), that oxidizes pyruvic acid to
binds to DNA without any apparent sequence acetyl–CoA and CO2. The size of the pyruvate
specificity, wrapping the DNA and causing it dehydrogenase complex is (4.6–4.8) × 106 Da.
to bend. Integration host factor (IHF) binds to Contrast this with the size of a 70S ribosome
Free ebooks ==>
42 the physiology and biochemistry of prokaryotes

(another multienzyme complex), which is of glucose to pyruvic acid or lactic acid) cannot
about 2.7 × 106 Da. Other enzyme complexes be isolated as complexes. It has generally been
that catalyze a consecutive series of biochemi- assumed that these enzymes exist either as inde-
cal reactions include the α-ketoglutarate dehy- pendent proteins or in loose associations that
drogenase complex, which consists of three are easily disrupted during cell breakage and the
different enzymes present in multiple copies accompanying dilution of the proteins. (Bear in
(i.e., 48 proteins), 2.5 × 106 Da (in E. coli), it mind that because of dilution of the extracts by
oxidizes α-ketoglutarate to succinyl–coenzyme the addition of buffers during the washing and
A (acetyl–SCoA) and CO2. Another example is suspension of the cells prior to breakage, the
fatty acid synthase in yeast. It consists of seven concentration of proteins in broken cell extracts
different enzymes (2.4 × 106 Da) and synthe- is orders of magnitude lower than in the aque-
sizes fatty acids from acetyl–SCoA. (For com- ous portion of the unbroken cell.)
parison to bacteria, see note 130.) One of the
advantages of a multienzyme complex is that 1.2.7 Cell shape
it facilitates the channeling of metabolites, This section is devoted to a discussion of pro-
thus increasing the efficiency of catalysis. For teins that influence the final shape that bacteria
example, in these stable enzyme complexes the take. A central point is the role of components
biochemical intermediates in the pathway are in determining the sites of synthesis and shape
transferred directly from one enzyme to the of the peptidoglycan sacculus that holds the
next without entering the bulk phase. Thus, cell in its particular shape. The argument will
there is no dilution of intermediates, nor is be made that bacteria have several cytoskel-
there reliance on random diffusion to reach a etal components that provide a scaffolding to
second enzyme. which peptidoglycan-synthesizing machinery is
attached, and the placement of the peptidogly-
Cytosol can machinery along the scaffolding determines
The liquid portion of the cytoplasm, the cytosol, the ultimate shape of the sacculus. We will
can be isolated in diluted form as the superna- begin with background information that briefly
tant fraction obtained after broken cell extracts describes cytoskeletal components found in
have been centrifuged at 105,000 × g for 1 to eukaryotic cells and compare these to cytoskel-
2 h, which should result in sedimentation of etal components present in prokaryotic cells.
the membranes, the DNA, the ribosomes, very
large protein aggregates, and other intracellular Eukaryotic cytoskeletal proteins
inclusions. In the cytosol are found the enzymes Eukaryotic cells have a network of protein fila-
that catalyze a major portion of the biochemi- ments (fibers), collectively called a cytoskeleton,
cal reactions in the cell, such as the enzymes of extending throughout the cytoplasm. Three
the central pathways for the metabolism of car- types of protein filament are present in the
bohydrates (glycolysis, the pentose phosphate eukaryotic cytoskeleton, and they are distin-
pathway, and the Entner–Doudoroff pathway) guished in part by their size. In increasing diam-
as well as the central pathways for organic acid eter, they are actin filaments (7 nm diameter),
metabolism (citric acid cycle and glyoxylate intermediate filaments (10–12 nm diameter),
pathways) and enzymes for other pathways, and microtubules (25 nm diameter). The fila-
such as the biosynthesis and degradation of ments are made from different protein subunits:
amino acids, lipids, and nucleotides. actin (actin filaments), tubulin (microtubules),
The concentration of proteins in the cytosol vimentin, lamin (intermediate filaments), and
is very high, making the material viscous, and so on. The eukaryotic cytoskeleton, along with
it is expected that extensive protein–protein motor proteins such as myosin, dynein, and
interactions occur among the enzymes, even kinesin, has many functions, briefly reviewed
those that do not exist in tight complexes. If in note 131. It is responsible for cell shape, the
such interactions exist, however, they must be movements of cilia and flagella, muscle contrac-
weak, since most of the enzymes that catalyze tion, endocytosis, and the movements of com-
metabolic pathways in the cytosol (e.g., the gly- ponents such as vesicles from one part of the cell
colytic enzymes that catalyze the degradation to another.
Free ebooks ==>
structure and function 43

Prokaryotic cytoskeleton proteins the synthesis of peptidoglycan in the septum,

Prokaryotes possess proteins that show some necessary for cell division. The Z ring is also
similarity to the eukaryotic cytoskeletal compo- important for cell separation because it recruits
nents actin (bacterial MreB), tubulin (bacterial peptidoglycan hydrolases that are required for
FtsZ), and intermediate filament components the septal peptidoglycan to split into two to
(bacterial IFs, crescentin), suggesting a prokary- allow the daughter cells to separate.
otic origin to these cytoskeletal components.. The Z ring also is required for the relocal-
These bacterial cytoskeletal components play ization of enzymes that are necessary for lat-
important roles in cell shape, cell division, chro- eral peptidoglycan expansion. These enzymes
mosome segregation, plasmid segregation, and include a transglycosylase called PBP1B. It
the maintenance of cell polarity. For reviews, should be mentioned that at least two other pro-
see refs. 132 through 136. teins, RodA and PBP2, are also required for lat-
eral peptidoglycan synthesis and are important
for cells to elongate as rods. See note 137 for an
How the prokaryotic cytoskeletal proteins
explanation of PBPs and their roles.
might determine cell shape
A description of the individual prokaryotic
Mre proteins. These are cytoskeletal proteins
cytoskeletal proteins is usefully preceded by a
that are required for cells to elongate. One
review of some ideas about how these proteine
of them is a protein called MreB. MreB has
determine cell shape. In this regard it is impor-
sequence patterns similar to that of actin and
tant that localized synthesis of peptidoglycan
is a member of the actin superfamily. It can be
reflects whether cells grow as rods or cocci and
found in many rod-shaped, filamentous, and
that the cytoskeletal proteins have been sug-
helical bacteria. It has been reported to form
gested to act as a scaffolding that determines
a cytoskeleton in some bacteria, encircling the
the location of the peptidoglycan biosynthetic
cell as spirals (also referred to as helical cables)
enzymes. Much of the experimental evidence
under the cell membrane along the longitudinal
showing the sites of peptidoglycan synthesis
axis from cell pole to cell pole, and it contributes
in rod-shaped cells and coccoid cells is due to
toward determining the shapes of nonspheri-
the use in gram-positive bacteria of fluorescent
cal bacteria.138 Mutations in the gene encod-
antibiotics that label peptidoglycan precursors
ing MreB cause E. coli to lose its rod shape and
and, in gram-negative bacteria, whose outer
become spherical. Mutants of MreB cause rod-
membrane is not permeable to fluorescent anti-
shaped B. subtilis cells to round up and lyse.
biotics, procedures the use of D-cysteine label-
Examination of genome sequences indicates
ing. What has been learned is that in E. coli and
that mreB is not present in coccoid bacteria.
B. subtilis the insertion of new peptidoglycan
One hypothesis is that the Mre proteins orga-
components occurs along the lateral cell wall in
nize the peptidoglycan biosynthetic machinery
a helical manner, but not at the cell poles. On
so that the cells grow as rods. Another way of
the other hand, new peptidoglycan is inserted
stating this is that the Mre proteins may be a
only in the septal region of cocci, as well as in E.
scaffolding that might recruit peptidoglycan-
coli round mutants (pbpA, rodA) that have lost
synthesizing enzymes such as transglycosylases
their ability to grow as rods. As mentioned ear-
for elongation of the glycan chains, cross-link-
lier, the sites of peptidoglycan synthesis appear
ing enzymes that form peptide bonds between
to be due to certain cytoskeletal protein compo-
the glycan chains, and hydrolases that introduce
nents that act as a scaffolding to which the pep-
gaps in the glycan chains so that additional com-
tidoglycan-synthesizing machinery is attached.
ponents can be added for elongation of the pep-
FtsZ. FtsZ is a cytoskeletal component related tidoglycan. It is the placement of these enzymes
to tubulin. At the site of cell division, it assem- that might determine cell shape. MreB is present
bles as a ring, called the Z ring, and recruits in both E. coli and Caulobacter crescentus, but
other proteins to form a contractile septal ring Bacillus subtilis has three paralogues (MreB,
that constricts the cell during division. In addi- Mbl, and MreBH). Anchored in the cell mem-
tion, two of the proteins recruited to the Z brane and having a major helical portion in the
ring are FtsI and FtsW, which are required for periplasm, MreC might serve in the periplasm
Free ebooks ==>
44 the physiology and biochemistry of prokaryotes

as a scaffolding for the spatial localization of that of a bundle of 10 to 12 flagella. Swimming

peptidoglycan-synthesizing machinery. is by means of helical contractions and exten-
sions of the cell, and rotation about the helical
Crescentin. In addition to MreB, the loss of
axis. For a discussion of this type of motility,
which leads to a loss of the rod shape, resulting in
including the forces and energetics, the student
lemon-shaped cells, Caulobacter crescentus has
is referred to refs. 142 through 144. Movement
a cytoskeleton protein called crescentin, which
is more rapid in viscous media.
is responsible for its vibrioid shape.136,139,140
What causes movements in Spiroplasma?
Crescentin has certain characteristics similar to
Inside each cell is a single flat, cytoskeletal rib-
those of intermediate filaments. The screening of
bon made of seven paired parallel protein fibers,
transposon-insertion libraries revealed mutants
attached to the inner surface of the cell mem-
that grow as rod-shaped bacteria rather than
brane along its innermost (and shortest) helical
vibrios. This led to the discovery of crescentin.
side. The ribbon is continuous and spans the
Immunofluorescence microscopy revealed that
entire length of the cell. Nonmotile mutants do
crescentin exists as a helical filament along the
not have the cytoskeletal ribbon. The cytoskel-
concave side of the cell membrane.
etal ribbon with its associated proteins is a lin-
ear motor, whereas flagella have rotary motors.
Cytoskeletal components unrelated to What this means is that the cytoskeletal ribbon
eukaryotic cytoskeletal components are undergoes linear contractions and extensions;
present in Mollicutes and because it is attached to the cell membrane,
The Mollicutes (Spiroplasma, Mycoplasma, the cell moves forward in the fluid. It has been
Acholeplasma) are a class of very small, bacte- proposed that the length changes in the ribbon
ria without cell walls. They may be free living, are due to conformational changes in the mono-
but many are parasitic on animals (including meric cytoskeletal proteins.
humans) and plants, and can cause disease.
Mollicutes can be found as part of the natural
flora in the mouth, throat, and genitourinary
Mycoplasma cells are very small, somewhat
tract of mammals and birds. (See note 141 for
spherical cells. One end is extended as a nar-
more information about diseases caused by
row neck, giving the cell a flasklike appear-
Mollicutes.) Several strains of Mycoplasma can
ance. Many are nonmotile. However, some
glide, and Spiroplasma can swim. These bacte-
are capable of movement on a solid surface.
ria are also chemotactic. However, they do not
Mycoplasma has an internal cytoskeleton com-
have analogues to genes responsible for chemot-
posed of protein fibers that originate at the tip
axis or in other bacteria. All Mollicutes stain
of the “neck” region. These fibers are present
gram-negative; however, 16S rRNA analysis
in both motile and nonmotile Mycoplasma.
indicates that they are closely related to gram-
Mycoplasma adheres to host tissues at the tip
positive bacteria, most notably Clostridium,
of the “neck” region, and it has been suggested
from which they are believed to have evolved.
that the cytoskeletal fibers play a role in adhe-
1. Cytoskeleton sion. How the fibers might be related to motility
Mollicutes have an internal protein cytoskel- is not clear.
eton that determines the shape of the cell and in
some cases (e.g., Spiroplasma) can function in 1.3 Summary
motility. They do not have homologues to any
of the eukaryotic genes for cytoskeletal pro- There are two evolutionary lines of prokary-
teins or motor proteins, such as actin, micro- otes, the Bacteria and the Archaea. Archaea are
tubule proteins, myosin, or dynein. Thus, their similar to bacteria, but they differ in certain fun-
cytoskeleton is made from protein unrelated to damental aspects of structure and biochemistry.
cytoskeletal proteins found in eukaryotes. These include differences in ribosomes, cell wall
chemistry, membrane lipids, and coenzymes.
2. Spiroplasma In addition, certain archaea have metabolic
Spiroplasma cells are very thin, round, helical, pathways (e.g., methanogenesis) not found in
tubelike cells. The cell diameter is approximately bacteria.
Free ebooks ==>
structure and function 45

Surrounding many bacteria are pili and extra- certain Mollicutes, driven by a cytoskeletal lin-
cellular material called a glycocalyx. The pili ear motor.
are protein fibrils. The glycocalyx encompasses There are two types of bacterial cell wall. One
all extracellular polymers beyond the cell wall, kind has a thick peptidoglycan layer to which
polysaccharide or protein, including capsules there are covalently bonded polysaccharides,
and slime layers. The pili and glycocalyx anchor teichoic acids, and teichuronic acids. Bacteria
the cell to specific animal and plant cell surfaces, with such walls stain gram-positive. A second
as well as to inanimate objects, and in several type of wall has a thin peptidoglycan layer and
cases to each other. an outer envelope consisting of lipopolysac-
Some bacteria possess a fibril called a sex charide, phospholipid, and protein. Cells with
pilus that attaches the cell to a mating part- such walls do not stain gram-positive and are
ner and retracts to draw the cells into intimate called gram-negative. There is also a separate
contact. When the cells are in contact, DNA is compartment in gram-negative bacteria called
transferred unilaterally from the donor cell (i.e., a periplasm, between the outer envelope and the
the one with the sex pilus) to the recipient. cell membrane. The periplasm is the location of
Protruding from many bacteria are flagella numerous proteins and enzymes, including pro-
filaments that aid the bacterium in swimming. teins required for solute transport and enzymes
At the base of the flagellum is a basal body that function in the oxidation and degradation
that is embedded in the cell membrane. At the of nutrients in the periplasm. Archaea have very
base of the flagellum is a rotary motor that in different cell walls. No archaeon has peptido-
most bacteria runs on a current of protons. glycan, although some have a similar polymer
Extreme halophiles and some marine bacteria called pseudopeptidoglycan or pseudomurein.
use a sodium ion current instead of a proton The cell membranes of the bacteria are all
current. A membranous sheath covers the fla- similar. They consist primarily of phosphoglyc-
gella of some bacteria (e.g., Vibrio cholerae, V. erides in a bilayer and protein. The phospho-
parahaemolyticus, Helicobacter, Bdellovibrio glycerides are generally fatty acids esterified to
bacteriovorus). There is a proteinaceous glycerol phosphate. Archaeal cell membranes
sheath covering the flagella (axial filaments) have lipids that are long-chain alcohols, ether-
of most spirochaetes (an exception is Borrelia linked to glycerol. Archaeal membranes can be
burgdorferi, the causative agent of Lyme dis- a bilayer or a monolayer, or perhaps a mixture
ease) which is actually located in the periplasm of the two.
rather than outside the cell. Spirochaetes are The cytoplasm of prokaryotes is a viscous
shaped like a coil, and when the flagellum solution of protein. Salts, sugars, amino acids,
turns, the cell moves in a corkscrewlike fash- and other metabolites are dissolved in the pro-
ion, which apparently is suited for movement teinaceous cytoplasm. In many bacteria intra-
through high-viscosity media. (Again, B. burg- cytoplasmic membranes are present. In many
dorferi is an exception and does not move in cases these membranes are invaginations of
a corkscrewlike fashion.144) Bacterial flagella the cell membrane. They are sites for electron
are important for the pathogenicity of certain transport and specialized biochemical activi-
bacteria and this is discussed in the review by ties. Numerous inclusion bodies are also pres-
Moens and Vanderleyden.39 Many genera of ent. These include ribosomes, which are the
bacteria contain species that swim as interact- sites of protein synthesis, as well as large aggre-
ing populations of cells on moist solid surfaces gates of enzymes that catalyze short metabolic
such as dilute agar, and other surfaces where pathways, gas vesicles, and carbon and energy
biofilms form. The movement is called swarm- reserves (e.g., glycogen particles). Some bacteria
ing and is due to lateral flagella. have magnetic particles called magnetosomes
There exist several other mechanisms of in the cytoplasm. Also present in the cytoplasm
besides flagella-driven motility. These include is DNA, very tightly packed, and referred to
“twitching,” a form of gliding motility pow- as a nucleoid. There is evidence that a protein
ered by type IV pili; gliding motility, which is cytoskeleton in the cytoplasm and periplasm
driven by extracellular slime secretion through plays a role in determining cell shape, as well as
pores in the cell surface; and swimming in serving other functions. It is clear that although
Free ebooks ==>
46 the physiology and biochemistry of prokaryotes


The microorganisms in the domain Archaea the exchange of nutrients. Other cell sur-
have cell surface structures called cannulae face appendages found on certain archaea
and hami not present in the Bacteria. (See are the hami (sing. hamus). These have been
Ng, S. Y. M., et al. 2008. Cell surface struc- described as a “halo” of filamentous protein
tures of Archaea. J. Bacteriol. 190:6039– appendages that surround each cell. The
6047). Cannulae are found in members of hami are morphologically complex struc-
the genus Pyrodictium that live in hydro- tures used for adhesion of the archaeal cells
thermal marine environments as anaero- to surfaces as well as to each other. They
bic chemoautotrophs. They derive energy have been called “grappling hooks” (hamus
from oxidizing hydrogen gas and reduc- is Latin for hook or barb) and have been
ing sulfur: H2 + S° → H2S. The cannulae found on the surface of SM1 cocci grow-
are hollow tubules made of glycoprotein, ing closely with Thiothrix, a filamentous
which connect the cells with one another. sulfur-oxidizing γ-proteobacterium grow-
When the cells divide, the daughter cells ing in cold sulfurous marshes. The coccoid
stay connected with the growing cannulae archaeal cells bearing the hami are found in
so that the end result is a dense network of “macroscopically visible string-of pearls-
cells that interconnect with one an other via like arrangement” with the Thiothrix. The
the cannulae. Some cannulae can enter the cell arrangement is very interesting, with the
periplasmic spaces of the cells but not the filamentous Thiothrix forming the “whit-
cytoplasm, whereas other cannulae simply ish” exterior of each “pearl” (SM1 coccus)
touch the outer cell wall layers (S layers) of plus the white interconnecting thread. Each
the cells. It may be that the cannulae allow “pearl” consists of up to 107 euryarchaeal
for cell–cell communication or perhaps even SM1 cells.

the cytoplasm of prokaryotes is not compart- (ribosomes, enzyme aggregates, etc.), and
mentalized into membrane-bound organelles membrane.
like the eukaryotic cytoplasm, it nonetheless
represents a very complex pattern.
Study Questions
The central metabolic pathways that are
described in the ensuing chapters (viz., glycoly-
1. What chemical and structural differences
sis, the Entner–Doudoroff pathway, the pentose
distinguish the archaea from the bacteria?
phosphate pathway, and the citric acid cycle) all
take place in the cytosol, the liquid part of the 2. What are the differences between gram-
cytoplasm. Also found in the cytosol are most of positive and gram-negative cell walls?
the other pathways, including the enzymatic reac-
3. Which cell wall polymers use D-alanine (as
tions for the synthesis and degradation of amino
opposed to L-alanine) as part of their struc-
acids, fatty acids, purines, and pyrimidines.
ture? (Hint: Polymers of one type are found
The cell membrane is the site of numerous
in both gram-positive and gram-negative
other metabolic pathways, including phos-
walls; those of the other are present only in
pholipid biosynthesis, protein secretion, solute
gram-positive walls.)
transport, electron transport, cell wall biosyn-
thesis, the generation of electrochemical ion 4. What structures enable bacteria to adhere
gradients, and ATP synthesis. The prokaryotic to surfaces? What is known about the
cell can therefore be considered to have three chemistry, location, and receptors of the
major metabolic domains: cytosol, particulate molecules that mediate the adhesion?
Free ebooks ==>
structure and function 47

5. Contrast the functions and cellular loca- 3. Woese, C. R. 1987. Bacterial evolution. Microbiol.
tion of peptidoglycan, phospholipid, and Rev. 51:221–271.
lipopolysaccharide. What is it about the 4. Pace, N. R., D. A. Stahl, D. J. Lane, and G. J.
chemical structure of these three classes of Olsen. 1985. Analyzing natural microbial popula-
compounds that is suitable for their func- tions by rRNA sequences. ASM News 51:4–12.
tions and/or cellular location? 5. Cavicchioli, R. (Ed.). Archaea, Molecular and
Cellular Biology. 2007. ASM Press, Washington,
6. What are porins, and what do they do? Are DC.
they necessary for gram-positive bacteria?
6. Ng, S. Y. M., B. Zolghadr, A. J. M. Driessen, S.-V.
Explain. Albers, and K. F. Jarrell. 2008. Cell surface structures
7. What are the physiological and enzymatic of Archaea. J. Bacteriol. 190:6039–6047.
functions associated with the periplasm? 7. Stetter, K. O. 1989. Extremely thermophilic
chemolithoautotrophic archaebacteria, pp. 167–
8. The protein TonB is an energy transducer 176. In: Autotrophic Bacteria. H. G. Schlegel and B.
in gram-negative bacteria that transfers Bowien (Eds.). Springer-Verlag, Berlin.
energy from the cell membrane to spe-
8. Schleper, C., G. Puehler, I. Holtz, A. Gambacorta,
cific transporters in the outer membrane. D. Janekovic, U. Santarius, H.-P., Klenk, and W.
Speculate about how this might occur. The Zillig. 1995. Picrophilus gen. nov., fam. nov.: a novel
subject is discussed in ref. 101. aerobic, heterotrophic, thermoacidophilic genus and
family compromising archaea capable of growth
9. Summarize the cell compartmentalization around pH 0. J. Bacteriol. 177:7050–7059.
of metabolic activities in prokaryotes.
9. Solfataras, which are acidic, geothermally heated
10. What are some multienzyme complexes? habitats containing inorganic sulfur compounds,
What advantage do they confer? including sulfuric acid, may be volcanic fissures,
springs, basins, or dried soil that previously con-
11. Which metabolic pathways or activities tained solfataric water. They have an acidic pH in the
are found in the cytosol and which in the aerobic portions because sulfuric acid arises from the
oxidation of hydrogen sulfide, which occurs sponta-
neously and also as a result of the growth of sulfur-
12. What distinguishes twitching from swim- oxidizing bacteria.
ming? Compare and contrast the mecha- 10. Koga, Y., M. Nishihara, H. Morii, and M.
nisms and structures. Akagawa-Matsushita. 1993. Ether polar lipids
of methanogenic bacteria: structures, compara-
13. What is the evidence that bacteria possess tive aspects, and biosynthesis. Microbiol. Rev. 57:
a cytoskeleton? Is there more than one type 164–182.
of cytoskeleton? What is the role of the 11. Grayling, R. A., K. Sandman, and J. N. Reeve.
cytoskeleton? Explain. 1996. Histones and chromatin structure in hyper-
thermophilic archaea. FEMS Microbiol. Rev.
14. What is the experimental basis for the con-
clusion that MotA and MotB form a com-
plex? What is the experimental basis for the 12. Pereira, S. L., R. A. Grayling, R. Lurz, and J.
N. Reeve. 1997. Archaeal nucleosomes. Proc. Natl.
conclusion that MotA conducts protons? Acad. Sci. USA 94:12633–12637.
Why has it been concluded that FliG is the
rotor portion of the flagellar motor? Why 13. Bardy, S. L., S. Y. M. Ng, and K. F. Jarrell.
2003. Prokaryotic motility structures. Microbiology
are the MS and C rings sometimes included 149:295–304.
as part of the rotor?
14. Macnab, R. M. 2003. How bacteria assemble
flagella. Annu. Rev. Microbiol. 57:77–100.
REFERENCES AND NOTES 15. Chevance, F., and K. T. Hughes. 2008.
Coordinating assembly of a bacterial macromolecu-
1. Morris, D. M., and G. J. Jensen. 2008. Toward lar machine. Nat. Rev. Microbiol. 6:455–465.
a biomechanical understanding of whole bacterial
16. Eukaryotic flagella differ from bacterial flagella
cells. Annu. Rev. Biochem. 77:583–613.
in many ways. Rather than being stiff, rotating fila-
2. Woese, C. R., O. Kandler, and M. L. Wheelis. ments, they are flexible and have an undulating wave
1990. Towards a natural system of organisms: motion from base to tip that pushes the cell forward.
proposal for the domains Archaea, Bacteria, and The movement is driven by ATP rather than by a
Eucarya. Proc. Natl. Acad. Sci. USA 87:4576–4579. proton current. The structure is very different from
Free ebooks ==>
48 the physiology and biochemistry of prokaryotes

that of prokaryotic flagella. The core of the filament A. Stoebner, and M. D. Manson. 1995. Protein inter-
is called an axoneme. The axoneme, which has an actions in the bacterial flagellar motor. Proc. Natl.
internal system of rods called microtubules, com- Acad. Sci. USA 92:1970–1974.
posed of tubulin subunits and associated proteins,
is covered by a sheath that is an extension of the 23. Blair, D. F., and H. C. Berg. 1990. The MotA
cell membrane. Each axoneme has nine fused pairs protein of E. coli is a proton-conducting component
of microtubules that are attached to each other as a of the flagellar motor. Cell 60:439–449.
bundle, surrounding two central microtubules, all of 24. Garza, A. G., L. W. Harris-Haller, R. A.
which extend the length of the filament. This is called Stoebner, and M. D. Manson. 1995. Motility protein
the 9 + 2 arrangement. Numerous proteins project interactions in the bacterial flagellar motor. Proc.
along the length of the microtubule doublets. Some Natl. Acad. Sci. USA 92:1970–1974.
of these are cross-links that attach the doublets to
each other. Others, called dynein, cause the bending. 25. Kojima, S., and D. F. Blair. 2001. Conformational
Dynein hydrolyzes ATP as a source of energy and change in the stator of the bacterial flagellar motor.
moves along the length of one microtubule doublet, Biochemistry 40:13041–13050.
pulling the attached doublet so that it bends. Other 26. Some researchers refer to the FliG proteins,
attached proteins constitute a relay system that con- which are bound to the peripheral inner surface of
trols the bending so that the appropriate waveform the MS ring, as well as to the C ring, as the rotor.
is produced. Some researchers refer to the C ring as the rotor, and
Cilia are constructed in a similar fashion, and some researchers refer to both the MS and C rings
the molecular basis for their movement is the same, as the rotor. See Bardy, S. L., S. Y. M. Ng, and K.
although their movement is more whiplike than the F. Jarrell. 2003. Prokaryotic motility structures.
movement of flagella. Microbiology 149:295–304. Berg, H. C. 2003. The
17. Iino, T., Y. Komeda, K. Kutsukake, R. M. rotary motor of the bacterial flagella. Annu. Rev.
Macnab, P. Matsumura, J. S. Parkinson, M. I. Simon, Biochem. 72:19–54. Grünenfelder, B., S. Gehrig, and
and S. Yamaguchi. 1998. New unified nomencla- U. Jenal. 2003. Role of the cytoplasmic C terminus of
ture for the flagellar genes of Escherichia coli and the FliF motor protein in flagellar assembly and rota-
Salmonella typhimurium. Microbiol. Rev. 52:533– tion. J. Bacteriol. 185:1624–1633.
27. Francis, N. R., G. E. Sosinsky, D. Thomas, and
18. Schoenhals, G. J., and R. M. Macnab. 1996. D. J. DeRosier. 1994. Isolation, characterization and
Physiological and biochemical analyses of FlgH, a structure of bacterial flagellar motors containing the
lipoprotein forming the outer membrane L ring of switch complex. J. Mol. Biol. 235:1261–1270.
the flagellar basal body of Salmonella typhimurium.
J. Bacteriol. 178:4200–4207. 28. FliG, FliM, and FliN are thought to be involved
in the turning of the flagellar motor (generating
19. Berg, H. C. 2003. The rotary motor of the bacte- torque) because certain mutations result in paraly-
rial flagella. Annu. Rev. Biochem. 72:19–54. sis. However, the evidence suggests that only FliG
20. Khan, S., M. Dapice, and T. S. Reese. 1988. is directly involved in generating torque. It has been
Effects of mot gene expression on the structure of the concluded that FliG, FliM, and FliN interact with one
flagellar motor. J. Mol. Biol. 202:575–584. another in a complex because certain mutations in
fliG, fliM, and fliN can be suppressed by other muta-
21. Ivey, D. M., M. Ito, R. Gilmour, J. Zemsky, tions in any of the fli genes. CheY-P (phosphorylated
A. A. Guffanti, M. G. Sturr, D. B. Hicks, and T. A. CheY), the molecule that causes the flagellar motor
Krulwich. 1998. Alkaliphile bioenergetics, pp. 181– to reverse its direction of rotation, binds to FliM
210. In: Extremophiles: Microbial Life in Extreme (Sections 20.4–20.7). FliG, FliM, and FliN are prob-
Environments. K. Horikoshi and W. D. Grant (Eds.). ably also involved in flagellar assembly, since null
John Wiley & Sons, New York. mutants do not have flagella. This may be because
22. Suppressor mutations can indicate whether pro- the switch complex must be made prior to more dis-
teins interact in a complex. With respect to MotA and tal portions of the flagellum.
MotB, suppressor mutations in E. coli were isolated 29. Pleier, E., and R. Schmitt. 1989. Identification
that suppressed motB missense mutations that oth- and sequence analysis of two related flagellin genes
erwise would have resulted in paralyzed or partially in Rhizobium meliloti. J. Bacteriol. 171:1467–1475.
paralyzed flagella. DNA sequencing of the suppres-
sor mutants showed that they caused single amino 30. Yonekura, K., S. Maki, D. G. Morgan, D. J.
acid changes in MotA. The conclusion is that MotA DeRosier, F. Vonderviszt, K. Imada, and K. Namba.
and MotB interact as a complex. Electron micro- 2000. The bacterial flagellar cap as a rotary promoter
graphic evidence also indicates that MotA and MotB of flagellin self-assembly. Science 290:2148–2152.
form a complex. Electron micrographs show 10 to 12 31. Hughes, K. T., and P. D. Aldridge. 2001. Putting
particles surrounding the M ring in the cytoplasmic a lid on it. Nat. Struct. Biol. 8:96–97.
membrane. The particles are not present when either
MotA or MotB is absent in mutants. For more infor- 32. Iino, T. 1969. Polarity of flagellar growth in
mation, read Garza, A. G., L. W. Harris-Haller, R. Salmonella. J. Gen. Microbiol. 56:227–239.
Free ebooks ==>
structure and function 49

33. Emerson, S. U., K. Tokuyasu, and M. I. Simon. consider Borrelia burgdorferi., which has 7 to 11 fla-
1970. Bacterial flagella: polarity of elongation. gella attached to each end of the cell; the flagella over-
Science 169:190–192. lap at the cell center to form a continuous bundle. The
cells themselves have a flat-wave shape, rather than
34. It has been proposed that ATP hydrolysis by
being “corkscrew” helical coils. Mutant cells that lack
FliI may drive the export of proteins through the
flagella owing to an insertion in flaB, which is the gene
translocation apparatus, as well as being necessary
that encodes the major flagellar filament protein, no
in the process of bringing proteins to the transloca-
longer resemble a flat wave but are rod shaped. Thus,
tion site. Since FliH binds to FliI and is a negative
the flagella influence the shape of the cell. Bearing this
regulator of its activity, the ATPase activity functions
in mind, recall that the flagellum rotates between the
only when necessary. See Minamino, T., and R. M.
outer membrane sheath and the flexible protoplasmic
Macnab. 2000. Interactions among components of
cell cylinder. One model is based on the configuration
the Salmonella flagellar export apparatus and its sub-
of the flagella: side by side within the flexible proto-
strates. Mol. Microbiol. 35:1052–1064. Minamino,
plasmic cylinder and held there owing to containment
T., and R. M. Macnab. 2000. FliH, a soluble com-
by the outer membrane sheath. The model proposes
ponent of the type III flagellar export apparatus of
the following: (1) the flagella have a left-handed heli-
Salmonella, forms a complex with FliI and inhibits its
cal shape (going away from the observer) wrapped
ATPase activity. Mol. Microbiol. 37:1494–1503.
around the protoplasmic cylinder; (2) the flagella
35. Fuerst, J. A., and J. W. Perry. 1988. Demonstration rotate counterclockwise as viewed from the back of
of lipopolysaccharide on sheathed flagella of Vibrio the cell; (3) as a result of flagella rotation, backward-
cholerae 0:1 by protein A–gold immunoelectron moving waves are propagated down the length of the
microscopy. J. Bacteriol. 170:1488–1494. flexible protoplasmic cylinder, and the cell is pushed
forward in the viscous medium. For a more detailed
36. Weissborn, A., H. M. Steinman, and L. Shapiro.
discussion of the model, see ref. 40.
1982. Characterization of the proteins of the
Caulobacter crescentus flagellar filament. J. Biol. 44. Parales, J., Jr., and E. P. Greenberg. 1991.
Chem. 257:2066–2074. N-Terminal amino acid sequences and amino acid
37. Pleier, E., and R. Schmitt. 1989. Identification compositions of the Spirochaeta aurantia flagellar
and sequence analysis of two related flagellin genes filament polypeptides. J. Bacteriol. 173:1357–1359.
in Rhizobium meliloti. J. Bacteriol. 171:1467–1475. 45. Moens, S., and J. Vanderleyden. 1996. Functions
38. Driks, A., R. Bryan, L. Shapiro, and D. J. of bacterial flagella. Crit. Rev. Microbiol. 22:67–
DeRosier. 1989. The organization of the Caulobacter 100.
crescentus flagellar filament. J. Mol. Biol. 206:627– 46. Harshey, R. M. 2003. Bacterial motility on a
636. surface: many ways to a common goal. Annu. Rev.
39. Moens, S., and J. Vanderleyden. 1996. Functions Microbiol. 57:249–273.
of bacterial flagella. Crit. Rev. Microbiol. 22:67– 47. Belas, R. 1997. Proteus mirabilis and other
100. swarming bacteria, pp. 183–219. In: Bacteria as
40. Charon, N. W., and S. F. Goldstein. 2002. Multicellular Organisms. J. A. Shapiro and M.
Genetics of motility and chemotaxis of a fascinat- Dworkin (Eds.). Oxford University Press, New
ing group of bacteria: the spirochaetes. Annu. Rev. York.
Genet. 36:47–73. 48. Matsuyama, T., K. Kaneda, Y. Nakagawa, K.
41. Several spirochaetes cause disease. Diseases Isa, H. Hara-Hotta, and I. Yano. 1992. A novel extra-
caused by spirochaetes include syphilis, caused cellular cyclic lipopeptide which promotes flagellum-
by Treponema pallidum; Lyme disease, caused by dependent and -independent spreading growth of
Borrelia burgdorferi; relapsing fever, caused by Serratia marcescens. J. Bacteriol. 174:1769–1776.
Borrelia spp.; leptospirosis, caused by Leptospira 49. Kearns, D. B., and R. Losick. 2003. Swarming
spp., and periodontal disease, caused by Treponema in undomesticated Bacillus subtilis. Mol. Microbiol.
denticola. In addition to living in animals, spiro- 49:581–590.
chaetes live in soil, fresh water, and salt water, and
attached to protozoa in the termite gut. The spiro- 50. McCarter, L. M. 2005. Multiple modes of motil-
chaetes attached to the protozoa in the termite gut ity: a second flagellar system in Escherichia coli. J.
enable the protozoa to swim. Bacteriol. 187:1207–1209.
42. Goldstein, S. F., N. W. Charon, and J. A. Kreiling. 51. Jarrell, K. F., D. P. Bayley, and A. S. Kostyukova.
1994. Borrelia burgdorferi swims with a planar 1996. The archaeal flagellum: a unique structure. J.
waveform similar to that of eukaryotic flagella. Proc. Bacteriol. 178:5057–5064.
Natl. Acad. Sci. USA 91:3433–3437.
52. Faguy, D. M., K. F. Jarrell, J. Kuzio, and M. L.
43. One model for spirochaete motility is based upon Kalmokoff. 1994. Molecular analysis of archaeal
the finding that the periplasmic flagella influence the flagellins: similarity to the type IV n-transport super-
shape of the cell. In other words, the periplasmic fla- family widespread in bacteria. Can. J. Microbiol.
gella act as a periplasmic skeleton. As an example, 40:67–71.
Free ebooks ==>
50 the physiology and biochemistry of prokaryotes

53. Faguy, D. M., S. F. Koval, and K. F. Jarrell. 1994. recipient cells (transmissible plasmids). Other genes
Physical characterization of the flagella and flagel- that may be carried by plasmids include genes for
lins from Methanospirillum hungatei. J. Bacteriol. antibiotic resistance (carried by resistance transfer
176:7491–7498. factors or R-factors) and genes for the catabolism
of certain nutrients. For example, Pseudomonas
54. Pearce, W. A., and T. M. Buchanan. 1980.
carries plasmids for the degradation of camphor,
Structure and cell membrane-binding properties
octane, salicylate, and naphthalene. Some plasmids
of bacterial fimbriae, pp. 289–344. In: Bacterial
confer virulence because the plasmids carry genes for
Adherence. E. H. Beachey (Ed.). Chapman & Hall,
toxins or other virulence factors, such as pili. The F
plasmid in E. coli carries genes for self-transmission,
55. Ottow, J. C. G. 1975. Ecology, physiology, and including genes for the F pilus, and is also capable of
genetics of fimbriae and pili. Annu. Rev. Microbiol. integrating into the chromosome and promoting the
29:79–108. transfer of chromosomal DNA.
56. Jones, G. W., and R. E. Isaacson. 1983. 66. Dunny, G. M. 1991. Mating interactions in
Proteinaceous bacterial adhesins and their receptors. gram-positive bacteria. pp. 9–33. In: Microbial Cell–
Crit. Rev. Microbiol. 10:229–260. Cell Interactions. M. Dworkin (Ed.). ASM Press,
Washington, DC.
57. Hultgren, S. J., S. Abraham, M. Caparon, P.
Falk, J. W. St. Geme III, and S. Normark. 1993. Pilus 67. An excellent reference for microbial cell–cell
and nonpilus bacterial adhesins: assembly and func- interactions is the book edited by Dworkin: Microbial
tion in cell recognition. Cell 73:887–901. Cell–Cell Interactions, 1991. M. Dworkin (Ed.).
ASM Press, Washington, DC.
58. Yanagawa, R., and K. Otsuki. 1970. Some
properties of Corynebacterium renale. J. Bacteriol. 68. Costerton, J. W., T. J. Marrie, and K.-J. Cheng.
101:1063–1069. 1985. Phenomena of bacterial adhesion, pp. 3–43. In:
Bacterial Adhesion: Mechanisms and Physiological
59. Cisar, J. O., and A. E. Vatter. 1979. Surface
Significance. D. C. Savage and M. Fletcher (Eds.).
fibrils (fimbriae) of Actinomyces viscosus T14V.
Plenum Press, New York.
Infect. Immun. 24:523–531.
69. Sleytr, U. B., and P. Messner. 1983. Crystalline
60. Eisenstein, B. I. 1987. Fimbriae, pp. 84–90.
surface layers on bacteria. Annu. Rev. Microbiol.
In: Escherichia coli and Salmonella typhimurium:
Cellular and Molecular Biology. Vol. 1. F. C.
Neidhardt et al. (Eds.). ASM Press, Washington, 70. Costerton, J. W., R. T. Irvin, and K.-J. Cheng.
DC. 1981. The bacterial glycocalyx in nature and disease.
Annu. Rev. Microbiol. 35:299–324.
61. Hultgren, S. J., S. Normark, and S. N. Abraham.
1991. Chaperone-assisted assembly and molecu- 71. Roberts, I. S. 1995. Bacterial polysaccharides
lar architecture of adhesive. Annu. Rev. Microbiol. in sickness and in health. Microbiology 141:2023–
45:383–415. 2031.
62. Hultgren, S. J., S. Abraham, M. Caparon, P. 72. Bacterial capsules. 1990. In: Current Topics in
Falk, J. W. St. Geme III, and S. Normark. 1993. Pilus Microbiology and Immunology. K. Jann and B. Jann
and nonpilus bacterial adhesins: assembly and func- (Eds.). Springer-Verlag, Berlin.
tion in cell recognition. Cell 73:887–901.
73. Streptococcus pyogenes (also called group A
63. Type IV pili are also used by certain bacteria, streptococcus) is a human pathogen that causes a
such as Pseudomonas, to form biofilms, and by wide range of diseases such as pharyngitis, impe-
myxobacteria for gliding motility. This is discussed tigo, erysipelas (acute infection of the skin), cellulitis
in Sections 23.1.1 and 23.1.2. (infection of subcutaneous tissues), and necrotizing
fasciitis (the strain that causes necrotizing fasciitis,
64. Pathogenic strains of Escherichia coli cause gas-
called the “flesh-eating bacteria,” also causes subcu-
trointestinal and urinary tract infections, Neisseria
taneous infection resulting in muscle destruction and
gonorrhoeae causes gonorrhea, Bacteroides nodo-
organ failure) and toxic shock syndrome (a drastic
sus causes bovine foot rot, Moraxella bovis causes
decrease in blood pressure, i.e., shock, followed by
bovine keratoconjunctivitis, Vibrio cholerae causes
organ failure). S. pyogenes produces several viru-
cholera, and Pseudomonas aeruginosa causes infec-
lence factors, among which is M protein. This cell
tions of the urinary tract and wounds. The adherence
surface protein is released from the bacteria via a
of these bacteria to their host tissue is the first stage
proteolytic enzyme secreted by S. pyogenes. M pro-
in pathogenesis.
tein binds a plasma protein called fibrinogen, and
65. In addition to the bacterial chromosome, most the complex activates neutrophils (polymorphonu-
bacteria carry extrachromosomal DNA molecules clear leukocytes, or PMNs). The activated PMNs
called plasmids. There are many types of plasmid. release heparin-binding protein, which causes vas-
They all carry genes for self-replication. Conjugative cular leakage resulting in the release of plasma and
plasmids also carry genes for DNA transfer into blood cells from the blood vessels, resulting in shock
Free ebooks ==>
structure and function 51

that contributes to subsequent organ failure. For adhesin that is thought to bind to the LTA, which in
more information, read: Herwald, H., H. Cramer, turn binds to host cell surfaces.
M. Mörgelin, W. Russel, U. Sollenberg, A. Norrby-
Teglund, H. Flodgaard, L. Lindbom, and L. Björck. 86. Sutcliffe, I. C., and N. Shaw. 1991. Atypical
2004. M protein, a classical bacterial virulence deter- lipoteichoic acids of gram-positive bacteria. J.
minant, forms complexes with fibrinogen that induce Bacteriol. 173:7065–7069.
vascular leakage. Cell 116:367–379. 87. Karakousis, P. C., W. R. Bishai, and S. E.
74. Roberts, I. S. 1996. The biochemistry and genet- Dorman. 2004. Mycobacterium tuberculosis cell
ics of capsular polysaccharide production in bacte- envelope lipids and the host immune response. Cell.
ria. Annu. Rev. Microbiol. 50:285–315. Micobiol. 6:105–116.
75. Kolenbrander, P. E., N. Ganeshkumar, F. J. 88. Hamasaki, N., K. Isowa, K. Kamada, Y. Terano,
Cassels, and C. V. Hughes. 1993. Coaggregation: T. Matsumoto, T. Arakawa, K. Kobayashi, and I.
specific adherence among human oral plaque bacte- Yano. 2000. In vivo administration of mycobacterial
ria. FASEB J. 7:406–409. cord factor (trehalose 6,6′-dimycolate) can induce
lung and liver granulomas and thymic atrophy in
76. Nikaido, H., and M. Vaara. 1987. Outer mem-
rabbits. Infect. Immun. 68:3704–3709.
brane, pp. 7–22. In: Escherichia coli and Salmonella
typhimurium: Cellular and Molecular Biology, 89. Virulence is defined as a quantitative measure
Vol. 1. F. C. Neidhardt et al. (Eds.). ASM Press, of a pathogen’s ability to cause disease. Bacterial
Washington, DC. pathogens produce virulence factors of several types.
77. Weidel, W., and H. Pelzer. 1964. Bagshaped These include antiphagocytic capsules, enzymes that
macromolecules—a new outlook on bacterial cell degrade host tissues, and toxins. The lipopolysaccha-
walls. Adv. Enzymol. 26:193–232. ride (LPS) is a toxin that can also contribute toward
virulence. LPS is referred to as an endotoxin because
78. Dmitriev, B. A., F. V. Toukach, K.-J. Schaper, O. it is not secreted by the bacteria into the extracellular
Holst, E. T. Rietschel, and S. Ehlers. 2003. Tertiary medium. It is released from dying bacteria and can
structure of bacterial murein: the scaffold model. J. be shed as blebs from living bacteria. The endotoxic
Bacteriol. 185:3458–3468. activity of the LPS resides in the lipid A portion.
79. Vollmer, W., and J.-V. Höltje. 2004. The archi- Endotoxin causes macrophages to produce tumor
tecture of the murein (peptidoglycan) in gram-nega- necrosis factor (TNF-α), which accumulates in the
tive bacteria: vertical scaffold or horizontal layer(s)? bloodstream. The increased TNF-α in the blood-
J. Bacteriol. 186:5978–5987. stream causes septic shock, which is a drastic drop in
blood pressure due to leakage of blood vessels. This is
80. The amino acid at position three in the peptido- a very dangerous condition because the drop in blood
glycan peptide subunit is referred to as L-R3. It var- pressure results in organ failure.
ies with the species. Gram-negative bacteria usually
have meso-diaminopimelic acid (meso-DAP), or in 90. Although the outer membrane is a permeabil-
the case of some spirochaetes, L-ornithine. There is ity barrier to antibiotics and other toxic substances,
much more variability in gram-positive bacteria, and gram-negative bacteria resist these substances for
L-R3 can be L-alanine, L-homoserine, L-diaminobu- other reasons, as well. A second mechanism is the
tyric acid, L-glutamic acid, L-ornithine, L-lysine, or presence of periplasmic β-lactamases that hydro-
DAP. lyze the earlier β-lactam antibiotics that were used
to treat infections. A third mechanism, and one that
81. Koch, A. L. 1983. The surface stress theory of makes very important contributions to the resistance
microbial morphogenesis. Adv. Microb. Physiol. of pathogenic bacteria to antimicrobial agents, is the
24:301–366. presence of drug efflux pumps. Gram-positive bac-
82. Koch, A. L., and S. Woeste. 1992. Elasticity teria also have drug efflux pumps. A wide range of
of the sacculus of Escherichia coli. J. Bacteriol. toxic substances can be removed from the bacterial
174:4811–4819. cytoplasm of both gram-positive and gram-negative
bacteria by using these pumps. Some pumps transport
83. Weidenmaier, C., and A. Peschels. 2008. a single drug, whereas others transport several unre-
Teichoic acids and related cell-wall glycopolymers in lated drugs. Antimicrobial agents that are pumped
gram-positive physiology and host interactions. Nat. out of cells include basic dyes, quaternary ammo-
Rev. Microbiol. 6:276–287 nium compounds, and a large number of different
84. Neuhaus, F. C., and J. Baddiley. 2003. A con- antibiotics including puromycin, chloramphenicol,
tinuum of anionic charge: structures and functions tetracycline, and erythromycin. For a review of this
of D-alanyl-teichoic acids in gram-positive bacteria. subject, read Section 17.5.
Microbiol. Mol. Biol. Rev. 67:686–723.
91. Leduc, M., K. Ishidate, N. Shakibai, and L.
85. S. pyogenes is an important pathogen that causes Rothfield. 1992. Interactions of Escherichia coli
most streptococcal infections in humans, includ- membrane lipoproteins with the murein sacculus. J.
ing “strep throat.” The M protein is a cell surface Bacteriol. 174:7982–7988.
Free ebooks ==>
52 the physiology and biochemistry of prokaryotes

92. Cross-linking reagents are useful tools for deter- 101. Postle, K., and R. J. Kadner. 2003. Touch
mining the physical association of other molecules. and go: tying TonB to transport. Mol. Microbiol.
These are bifunctional reagents (i.e., they have a 49:869–882.
chemical group at both ends of the molecule that can
102. Merchante, R., H. M. Pooley, and D. Karamata.
form a covalent bond to functional groups on proteins
1995. A periplasm in Bacillus subtilis. J. Bacteriol.
or other molecules such as peptidoglycan); thus they
are able to cross-link two molecules that are within
the length of the cross-linking reagent. For example, 103. Matias, V. R. F., and T. J. Beveridge. 2005.
the reagent dithio-bis-succinimidylpropionate (DSP) Cryo–electron microscopy reveals native polymeric
forms covalent cross-links between amino groups cell wall structure in Bacillus subtilis 168 and the
that are less than 12 angstrom units (1.2 nm) apart. existence of a periplasmic space. Mol. Microbiol.
Therefore, DSP can cross-link proteins that are 56:240–251.
physically associated with the peptidoglycan. After
104. Protoplasts are cells from which the cell wall
treatment with DSP, the peptidoglycan and its cross-
has been removed. They can be stabilized by suspen-
linked proteins can be isolated. DSP has two arms
sion in isotonic buffer (i.e., buffer that is iso-osmolar
held together by a disulfide bond that can be cleaved
with the cell contents, hence preventing water from
with a cleavable cross-linking reagent called mercap-
rushing in and lysing the protoplasts).
toethanol. The cross-linked proteins that were held to
the peptidoglycan by the DSP are released after mer- 105. Booth, I. R., and P. Louis. 1999. Managing
captoethanol treatment and can be analyzed by gel hypoosmotic stress: aquaporins and mechanosen-
electrophoresis. Thus, the proteins cross-linked to the sitive channels in Escherichia coli. Curr. Opin.
peptidoglycan can be identified. In addition to vari- Microbiol. 2:166–169.
ous lipoproteins, including the murein lipoprotein, an
106. Calamita, G., B. Kempf, M. Bonhivers, W. R.
outer membrane protein, OmpA, can be cross-linked
Bishai, E. Bremer, and P. Agre. 1998. Regulation of
to the peptidoglycan, indicating close association. The
the Escherichia coli water channel gene aqpZ. Proc.
cross-linking of OmpA to the peptidoglycan reflects
Natl. Acad. Sci. USA 95:3627–3631.
its extension through the outer envelope.
107. Edwards, M. D., I. R. Booth, and S. Miller.
93. Ferguson, S. J. 1992. The periplasm, pp. 311–
2004. Gating the bacterial mechanosensitive chan-
339. In: Procaryotic Structure and Function: A New
nels: MscS a new paradigm? Curr. Opin. Microbiol.
Perspective. S. Mohan, C. Dow, and J. A. Cole (Eds.).
Society for General Microbiology Symposium 47,
Cambridge University Press, Cambridge. 108. Kung, C., B. Martinac, and S. Sukharev. 2010.
Mechanosensitive channels in microbes. Annu. Rev.
94. Hobot, J. A., E. Carleman, W. Villiger, and E.
Microbiol. 64:313–329.
Kellenberger. 1984. Periplasmic gel: new concept
resulting from the reinvestigation of bacterial cell 109. Booth, I. R., M. D. Edwards, S. Black, U.
envelope ultrastructure by new methods. J. Bacteriol. Schumann, and S. Miller. 2007. Mechanosensitive
160:143–152. channels in bacteria: signs of closure? Nat. Rev.
Microbiol. 5:431–440.
95. Bayer, M. E. 1991. Zones of membrane adhe-
sion in the cryofixed envelope of Escherichia coli. J. 110. When a cell is subjected to sudden hypo-os-
Struct. Biol. 107:268–280. motic stress, the rapid flow of incoming water exerts
additional outward pressure against the cell mem-
96. Noina, N., M. Guillier, J. B.Travis, and S. K.
brane, causing distortion of the membrane, resulting
Buchanan. 2010. TonB-dependent transporters:
in the immediate opening of the mechanosensitive
regulation, structure, and function. Annu. Rev.
channels. (Mechanosensitive channels are sometimes
Microbiol. 64:43–60.
referred to as “gated” channels.) This results in out-
97. Bacteria secrete specific iron chelators called ward efflux of internal solutes along their diffusion
siderophores for scavenging iron from the environ- gradients, thus lowering the internal osmotic pres-
ment. Special transport systems in the outer mem- sure and the influx of water.
brane bind the iron siderophores and bring the
111. Romantsov, T., S. Helbig, D. E. Culham, C.
ligand–iron complex into the periplasm. These are
Gill, L. Stalker, and J. M. Wood. 2007. Cardiolipin
activated by TonB.
promotes polar localization of osmosensory trans-
98. Postle, K. 1990. TonB and the gram-negative porter ProP in Escherichia coli. Mol. Microbiol. 64:
dilemma. Mol. Microbiol. 4:2019–2025. 1455–1465.
99. Braun, V. 1995. Energy-coupled transport and 112. De Rosa, M., A. Gambacorta, and A. Gliozzi.
signal transduction through the gram-negative outer 1986. Structure, biosynthesis, and physiochemical
membrane via TonB-ExbB-ExbD-dependent recep- properties of archaebacterial lipids. Microbiol. Rev.
tor proteins. FEMS Microbiol. Rev. 16:295–307. 50:70–80.
100. Bradbeer, C. 1993. The proton motive force 113. Sternberg, B., C. L’Hostis, C. A. Whiteway,
drives the outer membrane transport of cobalamin in and A. Watts. 1992. The essential role of specific
Escherichia coli. J. Bacteriol. 175:3146–3150. Halobacterium halobium polar lipids in 2D-array
Free ebooks ==>
structure and function 53

formation of bacteriorhodopsin. Biochim. Biophys. DNA-binding protein from Escherichia coli. Proc.
Acta 1108:21–30. Natl. Acad. Sci. USA 72:3428–3432.
114. Gambacorta, A., A. Gliozzi, and M. De Rosa. 127. Rimsky, S. 2004. Structure of the histone-like
1995. Archaeal lipids and their biotechnological protein H-NS and its role in regulation and genome
applications. World J. Microbiol. 11:115–131. superstructure. Curr. Opin. Microbiol. 7:109–114.
115. Westerhoff, H. V., and G. R. Welch. 1992. 128. Pettijohn, D. E. 1996. The nucleoid, pp. 158–
Enzyme organization and the direction of metabolic 166. In: Escherichia coli and Salmonella: Cellular
flow: physiochemical considerations, pp. 361–390. and Molecular Biology. F. C. Neidhardt et al. (Eds.).
In: Current Topics in Cellular Regulation, Vol. 33. E. ASM Press, Washington, DC.
R. Stadtman and P. B. Chock (Eds.). Academic Press,
New York. 129. Molecular interactions among enzymes is
reviewed by Srivastava, D. K., and S. A. Bernhard.
116. Reviewed by Drews, G. 1991. Intracytoplasmic 1986. Enzyme–enzyme interactions and the regula-
membranes in bacterial cells, pp. 249–274. In: tion of metabolic reaction pathways, pp. 1–68. In:
Prokaryotic Structure and Function, A New Current Topics in Cellular Regulation, Vol. 28. B. L.
Perspective. S. Mohan, D. Dow, and J. A. Coles Horecker and E. R. Stadtman (Eds.). Academic Press,
(Eds.). Cambridge University Press, Cambridge. New York.
117. Overmann, J., S. Lehmann, and N. Pfennig. 130. In bacteria, fatty acid synthesis is catalyzed by
1991. Gas vesicle formation and buoyancy regula- enzymatic reactions similar to those found in yeast
tion in Pelodictyon phaeoclathratiforme (green sul- and mammals, but the enzymes cannot be isolated
fur bacteria). Arch. Microbiol. 157:29–37. as a complex.
118. Codd, G. A. 1988. Carboxysomes and ribulose 131. Motor proteins hydrolyze ATP and use the
bisphosphate carboxylase/oxygenase, pp. 115–164. released energy for internal conformational changes
In: Advances in Microbial Physiology, Vol. 29. A. that allow the proteins to move (“walk”) in one
H. Rose and D. W. Tempest (Eds.). Academic Press, direction along linear macromolecules such as pro-
New York. tein filaments. Motor proteins in eukaryotic cells
include myosin, dynein, and kinesin. Myosin moves
119. Yeates, T. O., C. A. Kerfeld, S. Heinhorst, G.
along protein microfilaments (actin filaments) and
C. Cannon, and J. M. Shively. 2008. Protein-based
plays several roles in doing so. These roles include
organelles in bacteria: carboxysomes and related
(a) the contraction of skeletal muscle cells, (b) the
microcompartments. Nat. Rev. Microbiol. 6:681–
contraction of one cell into two during cytokinesis,
(c) ameboid movement, and (d) the movement of
120. Sawaya, M. R., G. C. Cannon, S. Heinhorst, organelles and vesicles along actin microfilaments
S. Tanaka, E. B. Williams, T. O. Yeates, and C. A. in the cytoplasm during cytoplasmic streaming.
Kerfeld. 2006. The structure of β-carbonic anhy- Dynein is a motor protein that moves along micro-
drase from the carboxysomal shell reveals a distinct tubules in cilia and flagella, and as a consequence,
subclass with one active site for the price of two. J. the cilia and flagella bend. Kinesin, like cytoplasmic
Biol. Chem. 281:7546–7555. dynein, is a motor protein that binds to organelles
and vesicles and moves along cytoplasmic microtu-
121. Jogler, C., and D. Schuler. 2009. Genomics, bules, pulling organelles and vesicles through cyto-
genetics, and cell biology of magnetosome forma- plasm.
tion. Annu. Rev. Microbiol. 63:501–521.
132. Graumann, P. L. 2007. Cytoskeletal elements
122. Komeili, A., Z. Li, D. K. Newman, and G. J.
in bacteria. Annu. Rev. Microbiol. 61:589–618.
Jensen. 2006. Magnetosomes are cell membrane
invaginations organized by the actin-like protein 133. Pichoff, S., and J. Lutkenhaus. 2007. Overview
MamK. Science. 311:242–245. of cell shape: cytoskeletons shape bacterial cells.
Curr. Opin. Microbiol. 10:601–605.
123. Komeili, A. 2007. Molecular mechanisms
of magnetosome formation. Annu. Rev. Biochem. 134. Norris, V., G. Turnock, and D. Sigee. 1996.
76:351–366. The Escherichia coli endoskeleton. Mol. Microbiol.
124. Simmons, S. L., D. A. Bazylinski, and K. J.
Edwards. 2006. South-seeking magnetotactic bac- 135. Lewis, P. J. 2004. Bacterial subcellular archi-
teria in the Northern Hemisphere. Science 311:371– tecture: recent advances and future prospects. Mol.
374. Microbiol. 54:1135–1150.
125. Schüler, D. 2003. Molecular analysis of a 136. Lutkenhaus, J. 2003. Another cytoskeleton in
subcellular compartment: the magnetosome mem- the closet. Cell 115:648–650.
brane in Magnetospirillum gryphiswaldense. Arch.
Microbiol. 181:1–7. 137. PBPs refer to penicillin-binding proteins
that form and cross-link the peptidoglycan glycan
126. Rouviere-Yaniv, J., and F. Gros. 1975. strands that comprise the murein sacculus. The PBP
Characterization of a novel, low-molecular-weight enzymes are transglycosylases and transpeptidases.
Free ebooks ==>
54 the physiology and biochemistry of prokaryotes

In addition to the transglycosylases and transpep- 141. Mollicutes are part of the normal flora in
tidases, murein hydrolases are required for the healthy people. However, they can cause disease.
growth of the murein sacculus. The murein hydro- Diseases in humans caused by Mollicutes include
lases create gaps in the sacculus into which new primary atypical pneumonia (Mycoplasma pneumo-
peptidoglycan strands are inserted. For a review of niae), pelvic inflammatory disease (M. hominis), and
peptidoglycan synthesis, see Section 12.1. RodA nongonococcal urethritis (Ureaplasma urealyticum).
and FtsW might be involved in transporting pep- Spiroplasma causes diseases in insects and plants.
tidoglycan precursors to the sites of peptidoglycan 142. Gilad, R., Porat, A., and S. Trachtenberg. 2003.
synthesis. Modes of Spiroplasma melliferum BC3: a helical,
138. van den Ent, F., L. A. Amos, and J. Löwe. 2001. wall-less bacterium driven by a linear motor. Mol.
Prokaryotic origin of the actin cytoskeleton. Nature Microbiol. 47:657–669.
413:39–44. 143. Trachtenberg, S., R. Gilad, and N. Geffen.
139. Ausmees, N., J. R. Kuhn, and C. Jacobs- 2003. The bacterial linear motor of Spiroplasma
Wagner. 2003. The bacterial cytoskeleton: an inter- melliferum BC3: From single molecules to swimming
mediate filament-like function in cell shape. Cell cells. Mol. Microbiol. 47:671–697.
115:705–713. 144. Graumann, P. L. 2007. Cytoskeletal elements in
140. Figge, R. M., A. V. Divakaruni, and J. W. bacteria. Annu. Rev. Microbiol. 61:589–618.
Gobel. 2004. MreB, the cell shape-determining bacte- 145. Goldstein, S. F., N. W. Charon, and J. A.
rial actin homologue, co-ordinates cell wall morpho- Kreiling. 1994. Borrelia burgdorferi swims with a
genesis in Caulobacter crescentus. Mol. Microbiol. planar waveform similar to that of eukaryotic fla-
51:1321–1332. gella. Proc. Natl. Acad. Sci. USA 91:3433–3437.
Free ebooks ==>

Growth and Cell Division

The study of the growth of microbial popula- 2.1.1 Turbidity

tions is at the heart of microbial cell physiol- Turbidity measurements are routinely used
ogy because population growth characteristics to ascertain bacterial growth because of the
reflect underlying physiological events in the simplicity of the procedure. Within limits, the
individual cells. In fact, these cellular events are amount of light scattered by a bacterial cell is
frequently manifested to the investigator only proportional to its mass. Therefore, light scat-
by changes in the growth of populations. It is tering is proportional to the total mass. (The
therefore important to understand the changes relationship between total mass and light scat-
in growth that microbial populations undergo tering deviates from linearity at very high cell
and to be able to measure population growth densities.) If the average mass per cell remains
accurately. Additionally, to investigate cer- a constant, one can also use light scattering to
tain aspects of cell physiology (e.g., the inter- measure changes in cell number. What is actu-
dependencies between the rates of synthesis of ally measured is the fraction of incident light
the individual classes of macromolecules such transmitted through the culture (i.e., not the
as DNA, RNA, and protein), the investigator scattered light). The more dense the culture,
must be able to manipulate population growth the less light is transmitted. This relationship is
(e.g., by placing the culture in balanced growth given by the Beer–Lambert law (eq. 2.1). Thus
or continuous growth). This chapter describes if I0 is the incident light and I is the transmitted
methods to measure growth, presents an analy- light, then the Beer–Lambert law states that in a
sis of exponential growth kinetics, discusses population whose cell density is x, the fraction
growth in batch culture and in continuous cul- of light that is transmitted (I/I0) will decrease as
ture, and introduces some important aspects of the logarithm of x to the base 10.
growth physiology under certain nutritional
conditions. I/I0 = 10−xL (2.1)

That is, if one takes the log to the base 10 of

2.1 Measurement of Growth both sides of eq. 2.1, then log (I/I0) = –xL, where
Growth is defined as an increase in mass, and L is the light path in centimeters.
one can measure any growth parameter pro- The situation can be drawn schematically as
vided it increases proportionally to the mass of shown in Fig. 2.1. Note that the fraction of light
the culture. The most commonly used measure- that is transmitted decreases as a logarithmic
ments of growth are turbidity, total and viable function (base 10); the fraction decreases expo-
cell counts, dry weight, and protein. nentially, that is, with the density of the culture.

Free ebooks ==>
56 the physiology and biochemistry of prokaryotes

Fig. 2.1 Illustration of light scattering. The incident Fig. 2.2 Standard curve of optical density versus cells
light is I0 and the transmitted light is I. The spectropho- per milliliter. The line deviates from linearity at very
tometer usually provides the logarithm of the recipro- high cell densities.
cal of the fraction of transmitted light, that is, log(I0/I),
and this is called the optical density or turbidity.

Bacteriologists, however, prefer to measure counting chamber. This is simply a glass slide
something that increases with cell density. divided into tiny square wells of known area
They therefore use the reciprocal of log(I/I0), and depth. A drop of culture is placed on a slide,
which is log(I0/I). The log(I0/I) is called turbid- and a coverslip is applied. Each square well in
ity or absorbancy or optical density and has the the grid then holds a known volume of liquid.
symbol OD (for optical density) or A (for absor- The number of cells is counted microscopically,
bancy). Thus turbidity or OD is directly propor- and the number of cells per milliliter is obtained
tional to cell mass in the culture (or cell density, by using a conversion factor based on the total
if the ratio of mass to cell remains constant) and number of square wells counted and their vol-
is written as follows: ume. Although this is a simple procedure, and
one used widely, it has two limitations:
OD = A = xl (2.2)
1. It does not distinguish between live and dead
Turbidity is measured with a colorimeter or a
spectrophotometer. In practice, one constructs
2. It cannot be performed on populations
a standard curve by measuring the turbidity of
whose cell density is too low to count micro-
several different cell suspensions, counting the
scopically (<106 cells/mL).
cell number independently and using it as a mea-
sure of cell mass. A straight line is generated, Electronic cell counting is a more sophisticated
as shown in Fig. 2.2. However, the line devi- method. The bacteria, suspended in a saline solu-
ates from linearity at high cell densities. This is tion, are placed in a chamber with an electrode,
because when the cell density is too high, some separated by a second chamber filled with the
of the light is rescattered and directed toward the same saline solution and also provided with an
phototube, thus lowering the turbidity reading. electrode. A microscopic pore separates the two
Whenever turbidity measurements are made chambers. The bacterial suspension is pumped
on an unknown sample, a standard curve must through the pore into the second chamber.
be constructed to determine the cell density. Whenever a bacterium passes through the pore,
Turbidity measurements are the simplest way to the electrical conductivity of the circuit decreases
measure growth. Methods to determine the cell because the electrical conductivity of a bacterial
density by direct cell counts are described next. cell is less than that of the saline solution. This
change in conductivity results in a voltage pulse
2.1.2 Total cell counts that is counted electronically. The electronic
If the mass per cell is constant, one can measure counter has the advantage of allowing one also
growth by sampling the culture over a period to measure the size of the bacterium. This is
of time and counting the cells. There are two because the size of the pulse is proportional to
ways to do this: total cell counts and viable the size of the bacterium. Thus, one can obtain
cell counts. For total cell counts one can use a both a total cell count and a size distribution.
Free ebooks ==>
growth and cell division 57

2.1.3 Viable cell counts molecules, the regulation of the timing of DNA
A viable cell count is one in which the cells are synthesis and cell division, and such adaptive
serially diluted and then deposited on a solid physiological responses to nutrient availabil-
growth medium. Each viable cell grows into a ity as changes in gene expression, homeostasis
colony and the colonies are counted. Viable cell adaptations to the external environment, and
counts are routinely performed in microbiology the coupling of the rates of biosynthetic path-
laboratories. However, it is important to recog- ways to the rates of utilization of products for
nize that they may underestimate the number of growth (metabolic regulation). From this point
viable cells for the following reasons: of view, most of this text is concerned with the
physiology of growth, and we therefore return to
1. A viable cell count will underestimate the the topic in subsequent chapters. What follows
number of live cells if the cells are clumped, here is an introduction to growth physiology as
since each clump of cells will give rise to a it pertains to the synthesis of macromolecules
single colony. during steady state growth, and very gen-
2. Some bacteria plate with poor efficiency; eral adaptations of cells to nutrient depletion.
that is, single cells give rise to colonies with We begin with a discussion of the sequence of
low frequency. growth phases through which a population of
bacteria progresses after being inoculated into a
2.1.4 Dry weight and protein flask of fresh media.
The most direct way to measure growth is to
quantify the dry weight of cells in a culture. The
2.2.1 Phases of population growth
cells are harvested by either centrifugation or
When one measures the growth of populations
filtration, dried to a constant weight, and care-
of bacteria grown in batch culture, a progres-
fully weighed. Since protein increases parallel
sion through a series of phases can be observed
with growth, one can also measure growth by
(Fig. 2.3). The first phase is frequently a lag
doing protein measurements of the cells. The
phase, in which no net growth occurs (i.e., no
cells are harvested by either centrifugation or
increase in cell mass). This is followed by a
filtration or (more usually) precipitated first
phase of exponential growth, in which cell mass
with acid or alcohol, then recovered by centrif-
increases exponentially with time. Following
ugation or filtration. A simple colorimetric test
exponential growth, the culture enters the sta-
for protein is then performed.
tionary phase, a phase of no net growth. After
a stationary period, cell death occurs in a final
2.2 Growth Physiology stage called the death phase. Notice in Fig. 2.3
The vast topic of growth physiology includes that, prior to the exponential phase of growth,
the regulation of rates of synthesis of macro- when cell division occurs, the cells increase in

Fig. 2.3 Growth kinetics in batch culture: solid line, mass; dashed line, viable cells. Note that if we define
growth as an increase in mass, then only the solid line accurately reflects the growth of the culture. The dashed
line reflects growth only when it is parallel to the solid line.
Free ebooks ==>
58 the physiology and biochemistry of prokaryotes

mass (solid line); that is, they grow larger. At and accumulation of toxic products (e.g., alco-
the end of exponential growth, the cells con- hols, solvents, bases, acids). The accumulation
tinue to divide after growth has ceased (dashed of toxic products is frequently a problem for
line), but they become smaller. It would seem fermenting cells because instead of being con-
to be an advantage to a population of bacteria verted to cell material, most of the nutrient is
to continue to divide after growth has ceased in excreted as waste products. The excretion
the population, since in this way more cells can of fermentation end products is discussed in
be produced for distribution to new sites, where Chapter 15. For a review of the physiology of
conditions for continued growth may be better. stationary phase cells, read ref. 1.
One consequence of the uncoupling of growth
from cell division during the lag and late log Death
phases is that the size of the cell varies during Cell death can result from several factors.
growth in batch culture. For the investigator, Common causes of death include the depletion
the noncoincidence of growth and cell division of cellular energy and the activity of autolytic
during stages of batch growth has a practical (self-destructive) enzymes. Some bacteria begin
consequence: namely, cell counts are not always to die within hours of entering the stationary
a valid measurement of growth. phase. However, many bacteria remain viable
for longer periods. For example, some bacte-
Lag phase ria sporulate or form cysts when exponential
When cells in the stationary phase of growth growth ceases. The spores and cysts are resting
are transferred to fresh media, a lag phase cells that remain viable and will germinate in
often occurs. In this situation, the lag phase is fresh media. As discussed next, even nonsporu-
due basically to the time required for the physi- lating bacteria can adapt to nutrient depletion
ological adaptation of stationary phase cells in and remain viable for long periods in stationary
preparation for growth. Usually the longer the phase.
cells are kept in the stationary phase, the lon-
ger is the lag phase when they are transferred to
fresh media. Some of the physiological changes 2.2.2 Adaptive responses to
that occur in the stationary phase are described nutrient limitation
in Section 2.2.2. In the natural environment bacteria are fre-
There are several possible reasons for the lag quently faced with starvation conditions and
phase. In some cases the lag phase accommo- enter intermittent periods of no growth or very
dates the requirement for time for the cells to slow growth. In some environments the gen-
recover from toxic products of metabolism, such eration time may be many days or even months
as acids, bases, alcohols, or solvents, that may because the nutrient levels are so dilute. When a
accumulate in the external medium. Sometimes, bacterial culture faces the depletion of an essen-
new enzymes or coenzymes must be synthesized tial nutrient, the culture may keep growing by
before growth can resume. Such synthesis will be inducing specific uptake systems to scavenge
necessary if, for example, the fresh medium is dif- the environment for the nutrient or a source
ferent from the inoculum medium and requires thereof, and/or it may induce the synthesis of
a change in the enzyme composition of the cells. enzymes that can use an alternative source of
If significant cell death occurs in the stationary the nutrient. For example, bacteria starved for
phase, an apparent lag phase will be measured inorganic phosphate may induce the synthesis
because the inoculum includes dead cells that of a high-affinity inorganic phosphate uptake
contribute to the turbidity. The lag phase can be system as well as the synthesis of enzymes capa-
avoided if the inoculum is taken from the expo- ble of degrading organic phosphates to release
nential phase of growth and transferred to fresh inorganic phosphate (phosphatases). (See the
medium of the same composition. discussion of the Ntr and PHO regulons in
Sections 19.4 and 19.5, respectively.) If the
Stationary phase bacteria cannot bring into the cells the essential
Cells stop growing and enter the stationary nutrient, then the population will stop growing
phase for various reasons. Among these are and dividing. Under these circumstances, the
exhaustion of nutrients, limitation of oxygen, population enters stationary phase or, in the
Free ebooks ==>
growth and cell division 59

case of certain bacteria, the cells may sporulate from self-digestion of cell material, including
or encyst. (See Section 23.3 for a description of the cell envelope. Some gram-negative bacteria
sporulation in B. subtilis.) (e.g., Pseudomonas) bleb off outer membrane
Lately, increased attention is being given to vesicles as they become smaller. Frequently, the
physiological changes that occur in bacteria decrease in size can be accompanied by a change
that enter the stationary phase when they are from rod-shaped cells to coccoid-shaped cells,
experimentally subjected to starvation.2–5 These as discussed next.
bacteria undergo physiological changes that
result in metabolically less active cells that are Morphological changes
more resistant to environmental hazards. This Along with a reduction in size, some bacteria
property has been associated for a long time (e.g., Arthrobacter) change from rod-shaped
with bacteria that sporulate or form desicca- cells to coccoid cells in stationary phase. Similar
tion-resistant cysts upon nutrient deprivation. morphological changes from rods to small
The spores and cysts represent metabolically coccoid shapes can occur upon starvation in
inactive or less active stages of the life cycle of several other bacteria, including Klebsiella,
the organism. When nutrients become avail- Escherichia, Vibrio, and Pseudomonas.
able once more, the spores and cysts germi-
nate into vegetative cells that grow and divide. Changes in surface properties
More recently, it has been discovered that when When certain marine bacteria are starved, the
faced with nutrient deprivation, even some cell surface becomes hydrophobic and the cells
bacteria that do not form spores (e.g., E. coli, are more adhesive.8 Vibrio synthesizes surface
Salmonella, Vibrio, Pseudomonas) undergo fibrils and forms cell aggregates when starved
adaptive changes and thereupon enter station- for a long time. These changes in the surfaces
ary phase. Some of these changes also result of starved cells make them more adhesive.
in resistant, metabolically less active cells, and Presumably, it is advantageous for nutrient-
such responses may be common in most if not limited cells to adhere to particles that have
all bacteria. However, not all the effects can be adsorbed nutrient on the surfaces.
rationalized in terms of survivability, and their
physiological role is not yet known. Some of
Changes in membrane phospholipids
the changes that occur in cells that are starved
In E. coli all the unsaturated fatty acids in the
are described next. For a review of starvation in
membrane phospholipids become converted
bacteria, see ref. 6.
to the cyclopropyl derivatives by means of the
methylation of the double bonds. The advan-
Changes in cell size tage to the cyclopropane fatty acids is not
As discussed earlier, cells that enter stationary known, since it does not appear that mutants
phase upon carbon exhaustion generally become unable to synthesize cyclopropane fatty acids
smaller because they undergo reductive divi- are at a survival disadvantage when faced with
sion; that is, they keep dividing for 1 to 2 h after environmental stresses such as starvation and
growth has ceased. Reductive division results high or low oxygen tension.
in the production of more cells, which may be
advantageous for dispersing the population. In Changes in metabolic activity
some cases there may be several cell divisions in When bacteria enter stationary phase, their
the absence of growth so that the cell size differs overall metabolic rate slows. In addition, during
radically from what is found in the growing cell. starvation many bacteria experience a signifi-
Some bacteria decrease in length from several cant increase in the turnover (metabolic break-
micrometers (e.g. 5–10 μm) to approximately down and resynthesis) of protein and RNA.
1 to 2 μm or even less. (See note 7.) As discussed Presumably, in starved cells the protein and
in ref. 1, in addition to undergoing size reduc- ribosomal RNA can serve as an energy source
tion due to continued division in the absence to maintain viability and crucial cell functions.
of growth, bacteria can become smaller during The latter would include solute transport sys-
starvation after reductive division has been com- tems, an energized membrane, and ATP pool
pleted. The additional decrease in size results levels.
Free ebooks ==>
60 the physiology and biochemistry of prokaryotes

Changes in protein composition starvation sigma factor. It is also known as σ38

Bacteria may synthesize 50 to 70 or more new because it has a molecular weight of 38 kDa.
proteins under conditions of carbon, nitrogen, Sometimes called the “master regulator of the
or phosphate starvation. Many of the proteins stationary phase response,” RpoS has been
serve specific functions related to the nutrient found in other gram-negative bacteria in addi-
that is in low concentration. For example, phos- tion to E. coli, but not in gram-positive bacte-
phate starvation induces the synthesis of PhoE ria.12 Gram-positive bacteria use a different
porin, which is an outer membrane channel for sigma factor (σB) to regulate the transcription of
anions, including phosphate. This helps the genes homologous to some of the σs -dependent
bacterium bring in more phosphate. Another genes in E. coli.13
example is the nitrogen fixation genes that are Proteins whose synthesis depends upon σs
induced when the cells are starved for nitrogen. include a catalase made during stationary
In addition to the proteins of known function, phase (HP II), which presumably accounts for
there are many proteins made under all condi- resistance to H2O2, an exonuclease III that can
tions of starvation for which there is presently repair DNA damage due to H2O2 and near-UV
no known function, although many of the pro- radiation, and an acid phosphatase. Because
teins are presumed to be involved in resistance wild-type rpoS mutants do not have station-
to stress. ary phase resistance properties (measured as
percentage survival) to heat, high salt (osmotic
Changes in resistance to shock), near-UV, H2O2, or prolonged starva-
environmental stress tion (e.g., several days of starvation for carbon
Cells entering stationary phase also become or nitrogen), it has been concluded that the
more resistant to environmental stresses such products of rpoS-dependent genes are necessary
as high temperatures, osmotic stress, and cer- for survival under stressful conditions during
tain chemicals. For example, E. coli reportedly this phase.8 The transcription of the rpoS gene
becomes more resistant in stationary phase to (measured by using a rpoS–lacZ fusion plas-
high temperature, hydrogen peroxide, high salt, mid) increases dramatically in a nutrient-rich
ethanol, solvents such as acetone and toluene, medium as cells approach and enter stationary
and acidic or basic pH. These resistant prop- phase and during starvation for specific nutri-
erties are due to the synthesis of a starvation ents such as phosphate.14,15 (See note 16 for a
sigma factor that has four names: RpoS, σs, σ38, description of lacZ gene fusions.) Homologues
and KatF. to rpoS exist in other gram-negative bacteria,
including pathogens. Mutations in rpoS in the
1. The rpoS gene in E. coli encodes a intestinal pathogen Salmonella result in the
sigma factor required for transcription of attenuation of virulence in mice.17,18 One might
genes expressed during stationary phase suppose that RpoS aids pathogens to survive
and starvation stress such as oxidative stress, which they might
The reasons for concluding that rpoS encodes encounter inside host cells, or pH stress (e.g., in
a sigma factor are as follows: (1) the predicted the stomach).
amino acid sequence of the rpoS gene product,
based upon nucleotide sequence data, suggests 2. RpoS is a global regulator that is important
that it is a sigma factor, and (2) studies with for stress responses in exponentially growing
purified protein have confirmed that the protein cells as well as in stationary phase
binds to RNA polymerase and directs transcrip- The RpoS subunit of RNA polymerase should be
tion of rpoS-dependent genes. RpoS is required viewed as a global regulator that also functions
for the transcription of a regulon that includes during responses to stress during exponential
at least 50 genes that encode proteins induced growth.19 Increased RpoS levels during expo-
by carbon starvation when cells enter station- nential growth can be caused by slow growth,
ary phase; it is also necessary for the transcrip- temperature upshift from 30 °C to 42 °C, and
tion of several genes whose activities increase high osmolarity. For example, during osmotic
during phosphate and nitrogen starvation.9–12 upshifts in exponentially growing cultures
Accordingly, the protein is also called σs, for RpoS levels increase. Such cultures become not
Free ebooks ==>
growth and cell division 61

only osmotolerant but also thermotolerant and limitations of carbon, phosphate, or nitrogen
resistant to H2O2. sources. How bacteria survive such nutritional
stress is fundamental to an understanding of
3. Regulation of RpoS levels can be at the their physiology in natural ecosystems.
transcriptional, post-transcriptional,
and post-translational levels
The control of ribosomal RNA synthesis
Interestingly, the reason for the increased lev-
There is much controversy surrounding the
els of RpoS during stress responses is not sim-
question of what regulates the rate of synthesis
ply increased transcription of the rpoS gene.
of ribosomes, although it is clear that ribosome
The levels of RpoS are regulated not only at the
synthesis is coupled to growth rates (Section
transcriptional level but also at the post-tran-
2.2.3).20 As shown in Fig. 2.4, faster growing
scriptional and post-translational levels. Thus,
cells have more ribosomes (reflected in cellular
the regulation of RpoS levels is very complex.
RNA) per unit cell mass. For example, the num-
For example, the rate of degradation of RpoS
ber of ribosomes in an E. coli cell can vary from
decreases when cells enter stationary phase.
fewer than 20,000 to about 70,000 depending
The half-life of RpoS in exponential cultures
upon the growth rate. In addition to growth
of E. coli growing at 37 °C is less than 2 min.
rate control, amino acid starvation results in the
However, the half-life is greater than 30 min
inhibition of ribosome synthesis. Probably sev-
when cells enter stationary phase. See Box 2.1
eral factors are involved in regulating rates of
and the references therein for discussion of how
ribosome synthesis. For example, Fis, a DNA-
the levels of RpoS are adjusted, including tran-
binding protein whose synthesis is increased in
scriptional regulation of the rpoS gene, regula-
rich media, positively regulates transcription of
tion of translation of the RpoS messenger RNA,
rRNA genes. Another DNA-binding protein,
and the role of proteases in the degradation of
H-NS, antagonizes Fis stimulation (see later).
A very important regulator is guanosine tetra-
phosphate (ppGpp), which increases when cells
Other regulators for stationary phase are subjected to amino acid starvation or carbon
gene expression and energy limitation, and slows the synthesis
Although σs is certainly a major transcription of rRNA (and tRNA). The global regulator
factor for gene expression during station- ppGpp is best known as the effector that inhib-
ary phase, additional regulatory factors exist its rRNA and tRNA synthesis during the strin-
(reviewed in ref. 5). This conclusion is sug- gent response, an effect originally discovered as
gested partly because the variation in the tim- the inhibition of rRNA and tRNA synthesis due
ing of expression during stationary phase of to amino acid starvation.21 See note 22 for other
σs-dependent genes, depending upon the gene physiological effects of ppGpp and ref. 23 for a
in question, clearly points to regulatory fac- recent review.
tors in addition to σs. In fact, there are several
σs-dependent genes that are also controlled 1. The stringent response
by other global regulators, both positive and The stringent response in E. coli is a temporary
negative. One of these, cyclic AMP (cAMP) inhibition in the synthesis of ribosomal RNA
receptor protein (CRP) complex, stimulates and transfer RNA when the cells are shifted to a
the expression of approximately two-thirds medium in which they are starved for an amino
of the genes expressed during carbon starva- acid. It works in the following way. When
tion. Furthermore, not all genes whose expres- cells are shifted from a medium rich in amino
sion increases during stationary phase require acids to a minimal medium, “uncharged”
σs. It is clear from the available data that gene tRNA accumulates, causing an insufficiency of
expression during stationary phase is complex aminoacylated tRNA, which in turn results in
and poorly understood at the present time. “idling” or “stalled” ribsomes. An enzyme on
Nevertheless, expression during stationary the “stalled” ribosomes called RelA [(p)ppGpp
phase is extremely important because bacteria synthetase I, or PSI], which is the product of
in their natural habitat are frequently in this the relA gene, becomes activated by uncharged
phase or growing extremely slowly owing to tRNA that binds to the A site. Activated RelA
Free ebooks ==>
62 the physiology and biochemistry of prokaryotes


The regulation of transcription of rpoS ClpXP. This, then, is an example of trans-
(katF) appears to be complicated and is lational and post-translational control over
not well understood. There are probably the levels of a protein.
several signaling pathways that influence The stability of the RpoS protein increases
the transcription of the rpoS gene, allow- during stationary phase. Apparently RpoS is
ing the cells to respond to a variety of envi- sensitive to proteolytic digestion by ClpXP
ronmental challenges. The transcription during exponential growth but less so dur-
of rpoS does not always increase mark- ing stationary phase. The evidence for this
edly when cells enter stationary phase is that RpoS is stabilized during exponen-
because expression during growth may be tial growth in clpP and clpX X mutants. For
high. What happens during transcription a discussion of the role of RpoS as well as
depends upon how the cells are grown. how the levels of RpoS can be regulated, see
Anaerobiosis has been reported to stimu- refs. 1 through 5.
late rpoS expression in exponential cells
grown in a rich medium. Several other fac-
tors appear to influence the levels of rpoS
expression. For example, depending upon 1. Loewen, P. C., and R. Wenge-Aronis. 1994.
the strain of E. coli, rpoS expression may The role of the sigma factor σs (KatF) in bacte-
be high during exponential growth in a rial global regulation. Annu. Rev. Microbiol.
minimal medium and increase only slightly
during stationary phase. Additionally, 2. Wilmes-Riesenberg, M. R., J. W. Foster, and
starvation induced by transferring expo- R. Curtiss III. 1997. An altered rpoS allele con-
nential phase cells to a medium lacking tributes to the avirulence of Salmonella typh-
imurium LT2. Infect. Immun. 65:203–210.
a specific nutrient might stimulate rpoS
expression significantly or only margin- 3. Schweder, T., K.-H. Lee, O. Lomovskaya, and
ally, depending on which component is A. Matin. 1996. Regulation of Escherichia coli
starvation sigma factor (σs) by ClpXP protease.
missing from the medium. J. Bacteriol. 178:470–476.
The amounts of RpoS also increase
4. Zhou, Y., and S. Gottesman. 1998. Regulation
in response to an upshift in osmolarity, of proteolysis of the stationary-phase sigma fac-
even in exponentially growing cells. The tor RpoS. J. Bacteriol. 180:1154–1158.
increase is due to an increase in translation 5. Venturi, V. 2003. Control of rpoS transcrip-
of the RpoS mRNA as well as to a decrease tion in Escherichia coli and Pseudomonas: why
in the turnover of RpoS due to the protease so different? Mol. Microbiol. 49:1–9.

synthesizes the signaling molecule ppGpp and, 2. Carbon and energy limitation cause
to a lesser extent, the pentaphosphate (i.e., an increase in the levels of ppGpp in a
pppGpp). Together they are referred to as (p) pathway independent of RelA
ppGpp. The ppGpp somehow slows the tran- The regulator (p)ppGpp also accumulates dur-
scription of ribosomal RNA (and transfer ing carbon and energy limitation on a pathway
RNA). This leads to an inhibition of ribosomal that occurs in relA mutants, hence is indepen-
protein synthesis by an interesting mechanism dent of RelA, and requires the product of the
called translational coupling, as explained in spoT gene [(p)ppGpp synthetase II, or PSII]. The
note 24. SpoT protein is most probably a bifunctional
Free ebooks ==>
growth and cell division 63

Fig. 2.4 Macromolecular composition of E. coli grown at different growth rates. The values are amounts
per cell and have been normalized to values at a doubling time of 0.6 doubling/h. Source: Neidhardt,
F. C., J. L. Ingraham, and M. Schaechter. 1990. Physiology of the Bacterial Cell. Sinauer Associates,
Sunderland, MA.

enzyme that both synthesizes and degrades (p) medium and when stationary phase cells, which
ppGpp. The ratio of biosynthetic to degradative have very low Fis levels, are subcultured into a
activity is somehow regulated by carbon and nutrient-rich medium.28 One would expect an
energy limitation. This may account at least in increase in the rate of rRNA synthesis under
part for the lower number of ribosomes in cells these conditions, and Fis accounts in part for
growing slowly in poor medium. It is therefore the increased synthesis.
clear that the inhibition of rRNA and tRNA syn- Fis-dependent activation of ribosomal RNA
thesis coincides with essentially any nutritional genes is antagonized by another protein called
condition (not simply amino acid starvation) H-NS (also called H1), which can therefore act
that slows the growth rate. See note 25 for a as a repressor of rRNA transcription.29 (H-NS,
further discussion of the synthesis of (p)ppGpp discussed in Section 1.2.6 as part of the nucleoid,
and the contributions of RelA and SpoT. does not bind to specific nucleotide sequences
but does bind to curved or bent DNA.) The cel-
3. Other factors that control rRNA lular levels of H-NS are highest under conditions
synthesis: Fis, H-NS of slow rRNA synthesis (e.g., stationary phase).
It must be emphasized that the control of ribo- H-NS proteins belong to a remarkable group of
somal RNA synthesis is complex and that DNA-binding proteins that have other cellular
ppGpp is only one factor in the regulation. functions besides the regulation of rRNA gene
Other proteins involved in transcriptional transcription. (See Section 1.2.6 and note 30.)
regulation of the ribosomal RNA genes include So, although ppGpp somehow slows ribosomal
the Fis and H-NS proteins.26 (See note 27 for RNA synthesis, it appears to be only one com-
a further discussion of Fis.) Fis binds to DNA ponent of one or more complex regulatory
sequences upstream of the rRNA promoter and systems that must control ribosomal RNA syn-
activates the transcription of the operon. (Fis thesis under a variety of growth conditions.
regulates the expression, both positively and
negatively, of many genes besides rRNA genes.) 2.2.3 Macromolecular composition as a
Importantly, Fis synthesis is under growth rate function of growth rate
control. Synthesis increases when exponentially Much has been learned about growth physiol-
growing cells are shifted to a nutritionally richer ogy by observing changes in macromolecular
Free ebooks ==>
64 the physiology and biochemistry of prokaryotes

composition of cells whose growth rates have Why faster growing cells have more mass
been altered by changing the nutrient composi- As seen in Fig. 2.4, cells that are grown at a
tion of the medium, or by manipulation of the faster growth rate have a greater mass (i.e., they
dilution rates of cells growing in continuous are larger). For example, E. coli when growing
culture (Section 2.4). at a doubling rate of 2.5 doublings per hour is
Figure 2.4 illustrates the changes that are due approximately six times larger in mass than
to differences in the nutrient composition of the when it is growing at 0.6 doubling per hour.
growth media as seen in the ratios of cell mass, Why is that? It is because, first, the initiation of
RNA, protein, and DNA of E. coli grown at DNA replication requires a minimum cell mass
increasingly faster growth rates. The ordinate per replication origin, called the critical cell
gives relative amounts of macromolecules per mass (Mi); in addition, faster growing cells have
cell, normalized to cells undergoing 0.6 doubling several replication origins, and for the shorter
per hour. The abscissa gives the growth rate μ, in doubling times (<60 min), replication must
units of doublings per hour. Notice that the mass begin in an earlier cell cycle; see note 33 for a
of the cell as well as the relative concentrations more complete explanation. (The significance
of the macromolecules increase as exponential of Mi was reported by Donachie in 1968.34 See
functions of the growth rate. As shown in Fig. note 35 for how this was done. Whether Mi is
2.4, faster growing cells have a higher ratio of an absolute constant or whether it decreases
RNA to protein, a larger mass, and more DNA. slightly with increasing growth rate is not
known with certainty.36,37 For this discussion,
Why faster growing cells have a larger
we will assume that it is constant.)
RNA-to-protein ratio
Faster growing cells are enriched for RNA 2.2.4 Diauxic growth
with respect to the other cell components.
Many bacteria, including E. coli, will grow
This reflects a higher proportion of ribosomes
preferentially on glucose when presented with
in faster growing cells, as well as the composi-
mixtures of glucose and other carbon sources.
tion of ribosomes: approximately 65% RNA
The result is a biphasic growth curve as shown
and 35% protein by weight. The reason for the
in Fig. 2.5 for glucose and lactose. Preferential
increase in ribosomes is that, over a wide range
growth on one carbon source before growth
of rapid growth rates, a ribosome polymerizes
on a second carbon source is called diauxic
amino acids at an approximately constant rate.
growth. The bacteria first grow exponentially
When a cell increases or decreases its growth
on glucose, then enter a lag period, and finally
rate, and therefore the number of proteins it
grow exponentially on the second carbon
must make per unit time, it adjusts the number
of ribosomes rather than making each ribosome
work substantially slower or faster.
2.2.5 Catabolite repression by glucose
Why faster growing cells have more DNA It is now known that in many bacteria, glucose
Figure 2.4 also shows that faster growing cells represses the synthesis of enzymes required to
have more DNA per cell. The increased DNA grow on certain alternative carbon sources such
per cell is rationalized in the following way. In as lactose, and this can account for the diauxic
E. coli, the minimum time required between the growth discussed.
initiation of DNA replication and completion of Glucose also inhibits the uptake of certain
cell division (cell separation) for rapidly grow- other sugars into the cell. The result is that
ing cells (i.e., those with doubling times between growth on the second carbon source does not
20 and 60 min) is about 60 min.31 (See note 32 proceed until the glucose has been exhausted
for an explanation of the C and D periods.) If from the medium. During the lag period follow-
the generation time is shorter than 60 min, then ing the exponential growth on glucose, the bac-
DNA replication must begin in an earlier cell teria synthesize the enzymes required to grow
cycle and cells are born with partially replicated on the second carbon source. The repression of
DNA, which increases the average amount of genes by glucose is called catabolite repression,
DNA per cell. This is illustrated in Fig. 2.4, or glucose repression, and is indeed widespread
which shows that cells with a generation time of among bacteria. The inhibition of uptake of
less than 60 min have more DNA. alternative sugars is called inducer exclusion.
Free ebooks ==>
growth and cell division 65

Fig. 2.5 Diauxic growth on glucose and lactose: x, bacterial mass, y, glucose or lactose concentration in the
medium. There are two phases of growth. Initially the cells grow on the glucose. When the glucose is suffi-
ciently depleted, a lag phase occurs followed by growth on lactose. During the lag phase the cells induce the
synthesis of enzymes necessary for growth on lactose. Glucose prevents the induction of the genes necessary
for growth on lactose. The inhibition by glucose is called catabolite repression and is discussed in Sections
19.10.1 and 19.10.2.

A rationale for glucose repression points out of succinate and glucose, Rhizobium shows
that glucose is one of the most common car- diauxic growth by growing on the succinate
bon sources in the environment and that bac- first and the glucose second. This is because
teria frequently grow more rapidly on glucose in these bacteria, succinate represses key
than on other carbon sources, especially other glucose-degrading enzymes of the Entner–
sugars. Therefore, bacteria in certain ecologi- Doudoroff and Embden–Meyerhof–Parnas
cal habitats that preferentially utilize glucose pathways.38 Pseudomonas aeruginosa offers
may be at a competitive advantage with respect another example of a species in which diauxic
to the other cells that depend upon glucose for growth occurs first on organic acids and then
carbon and energy. Additionally, the enzymes on carbohydrates (e.g., glucose). If presented
required to metabolize glucose are generally with glucose or other carbohydrates and any
constitutive (i.e., present under all growth con- one of several organic acids such as acetate,
dition), and the cell is prepared to grow on glu- or one of the intermediates of the citric acid
cose at any time. cycle, P. aeruginosa will grow on the organic
acid first and will repress the enzymes required
2.2.6 Catabolite repression by molecules to degrade the carbohydrate.39
other than glucose
However, not all bacteria preferentially grow
on glucose, nor are the glucose catabolic 2.3 Growth Yields
enzymes necessarily constitutive. Several When a single nutrient (e.g., glucose) is the
obligately aerobic bacteria, when given a sole source of carbon and energy, and when
mixture of glucose and an organic acid, will its quantity limits the production of bacteria,
grow first on the organic acid. This is also it is possible to define a growth yield constant,
called catabolite repression except that it is Y. The growth yield constant is the amount
the organic acid that is the repressor of glu- of dry weight of cells produced per weight of
cose utilization. One example is presented by nutrient used (i.e., Y is the weight of cells made
the nitrogen-fixing bacteria belonging to the divided by the weight of nutrient used); Y has no
genus Rhizobium. These bacteria, which live units, since the weights cancel out. [The molar
symbiotically in root nodules of leguminous growth yield constant is Ym, which is the dry
plants, grow most rapidly in laboratory cul- weight of cells produced (in grams) per mole
tures on C4 carboxylic acids (e.g., succinate, of substrate used.] For example, the Yglucose for
malate, and fumarate) that are part of the cit- aerobically growing cells is about 0.5, which
ric acid cycle. When presented with a mixture means that about 50% of the sugar is converted
Free ebooks ==>
66 the physiology and biochemistry of prokaryotes

to cell material, and 50% is oxidized to CO2. X = x02Y (2.3)

For certain sugars and bacteria, the efficiency of
where x is anything that doubles each genera-
conversion to cell material can be much lower
tion, x0 is the starting value, and Y is the num-
(e.g., 20%). These differences are thought to be
ber of generations; x can be cell number, mass,
related to the amount of ATP generated from
or some cell component (e.g., protein, DNA).
unit weight of the carbon source.
Taking log10 of both sides,40 we write
The more ATP made, the greater the growth
yields. This makes sense if you consider that log x = log x0 + 0.301Y (2.4)
growth is, after all, an increase in dry weight, and
Let us define g as the generation time (i.e., the
a certain number of ATP molecules is required
time per generation). Therefore, g = t/Y, and
to synthesize each cell component. A value of
thus Y = t/g, where t is time elapsed.
10.5 g of cells per mole of ATP (called the YATP
Therefore, eq. 2.4 becomes
or molar growth yield) has been determined
for fermenting bacterial cultures growing on log x = log x0 + (0.301/g)(t) (2.5)
glucose as the energy source. The experiments
You will recognize this as an equation for a
are performed in media in which all the precur-
straight line. When x is plotted against t on
sors to the macromolecules (e.g., amino acids,
semilog paper (which is to base 10), the slope is
purines, pyrimidines) are supplied, so that the
0.301/g and the intercept is x0 (Fig. 2.6).
glucose serves only as the energy source, and
essentially all the glucose carbon is accounted The generation time, g
for as fermentation end products. The generation time, the time needed for the
Knowing the amount of ATP produced in population of cells to double, is an important
the fermentation pathways, it is possible to cal- parameter of growth and is probably the one
culate the YATP from the Yglucose. For example, most widely used by microbiologists. For this
22 g of Streptococcus faecalis cells is produced reason, the student should learn how to find the
per mole of glucose. This organism ferments a generation time. The simplest way is graphi-
mole of glucose to fermentation end products cally. In practice, one determines the generation
using the Embden–Meyerhof–Parnas pathway, time from inspection of a plot of x versus t on
which produces two moles of ATP (Section 9.1). semilog paper (Fig. 2.3 or 2.5) and simply reads
Thus, the YATP is 22/2 or 11. On the other hand, off the time it takes for x to double. Or, if x, x0,
Zymomonas mobilis produces only 8.6 g of and t are known accurately for the exponential
cells per mole of glucose fermented. These phase of growth, one can use eq. 2.3.
organisms use a different pathway for glucose
The growth rate constant, k
fermentation, the Entner–Doudoroff pathway,
The growth rate constant, k, is a measure of the
which yields only 1 ATP per mole of glucose
instantaneous growth rate and has the units of
fermented (Section 9.6). Thus, the YATP is 8.6.
reciprocal time. Another widely used parameter
As mentioned, a comparison of several differ-
ent fermentations by bacteria and yeast has
produced an average YATP of 10.5. Although
this value can vary, knowledge of the growth
yields and the YATP has allowed some deduc-
tions regarding which fermentation pathway
might be operating. Also, an unexpectedly high
growth yield in fermenting bacteria can point
to unrecognized sources of metabolic energy
(Section 4.8.3).

2.4 Growth Kinetics

2.4.1 The equation for exponential growth
During exponential growth, the mass in the cul-
ture doubles in each generation. The equation Fig. 2.6 Exponential growth plotted on a semilog
for exponential growth is scale: g, generation time (time per generation).
Free ebooks ==>
growth and cell division 67

of growth, k is not to be confused with the gen- find 108 = x028 or x0 = 108/28 = 3.9 × 105 cells/
eration time, g, which is the average time needed mL. This means that the initial cell density in
for the population of cells to double. To clarify the growth flask should be 3.9 × 105 cells/mL.
the distinction, recall that since each bacterial Since, however, the cell density of the inoculum
cell gives rise to two cells, the rate at which the is 108/mL, the inoculum must be diluted 108/3.9
population is growing at any instant, that is, × 105 or 2.6 × 102 times. To grow 1 liter of cells,
the instantaneous rate of growth (dx/dt), must the inoculum size would have to be 3.8 mL (see
be equal to the number of cells at that time (x) note 42). Most routine growth problems can be
times a growth rate constant (k). Thus: solved with the following equations:
dx/dt = kx or dx/x = k dt (2.6) x = x02Y
Upon integration, we have Y = t/g
ln x = kt + ln x0 g(k) = 0.693
x = x0ekt (2.7) 2.4.2 The relationship between the
growth rate (k) and the nutrient
Taking log10 of both sides of eq. 2.7 converts concentration (S)
the expression from the natural log (loge) to log In the natural environment the concentration
base 10, so that x can be plotted against t on of nutrients is so low that the growth rates are
semilog paper: limited by the rates of nutrient uptake or the
log x = 0.4342(kt) + log x0 rate at which a stored nutrient is used. At very
low nutrient concentrations, the growth rate
or can be shown to be a function of the nutrient
log x = kt/2.303 + log x0 (2.8) (substrate) concentration (Fig. 2.7). The curve
approximates saturation kinetics and is proba-
Note that this gives the same line as in Fig. 2.6. bly due to the saturation, in the membrane, of a
However, the slope is equal to k/2.303. Since the transporter that brings the nutrient into the cell.
slope in Fig. 2.6 is equal to 0.301/g, it follows The curve can be compared with the kinetics of
that k/2.303 must be equal to 0.301/g. Thus solute uptake on a transporter, as described in
k = 0.693/g (or kg = 0.693). Therefore, k and Section 17.2.
g vary as the reciprocal of each other with the One can rationalize the kinetics by assuming
proportionality constant of 0.693. The slope of that at very low concentrations of nutrient, the
the curve in Fig. 2.6 can give you either k (the
instantaneous growth rate) or g (the doubling
time for the population).41
Using the growth equations
The growth equations can be used for finding g
and k. In another, more routine, circumstance,
one is growing a culture to be used at a later
time and must choose the proper size inocu-
lum. Convenient approaches to remember for
this calculation have been given (eq. 2.3 or 2.5).
Suppose you would like to subculture an expo-
nentially growing culture at a density of 108
cells/mL so that 16 h later the density of the new
Fig. 2.7 Variation of growth rate as a function of
culture will also be 108 cells/mL. If g = 2 h, what
substrate concentration: kmax, maximal growth rate
should x0 be? constant; k, specific growth rate constant; ks, nutri-
One way to do this problem is to estimate the ent concentration that gives ½kmax; S, nutrient con-
number of generations (Y) and then use eq. 2.3 centration. The concentrations at which growth rate
or 2.5. For example, since Y = t/g, the number is proportional to nutrient concentration are very
of generations is 16/2, or 8. By using eq. 2.3, we low, in the micromolar range.
Free ebooks ==>
68 the physiology and biochemistry of prokaryotes

growth rate is proportional to the percentage growth chamber. For example, if the flow rate
of transporter that has bound the nutrient. At (F) is 10 mL/h and the volume (V) in the growth
high nutrient concentrations, the rate of nutri- chamber is 1 liter, then the dilution rate (D) is
ent entry approaches its maximal value because 10 mL/h per thousand milliliters, or 0.01 h–1.
all of the transporter has bound the nutrient. Notice that the units of D are reciprocal time. If
Therefore the growth rate becomes independent one multiplies the dilution rate D by x, the num-
of the nutrient concentration and is now limited ber of cells in the growth chamber, the product
by some other factor. The curve in Fig. 2.7 can Dx represents the rate of loss of cells from the
be described by the following equation, where outflow and has the units of cells/time.
k is the specific growth rate constant, kmax is the
Relationship of dilution rate (D) to
maximal growth rate constant, S is the nutrient
growth rate constant (k)
concentration, and Ks is the nutrient concentra-
In the steady state, k = D. To show this, consider
tion that gives 0.5kmax:
that the rate of growth is dx/dt = kx. When a
k = kmaxS/(Ks+ S) (2.9) steady state is reached in the growth chamber,
the rate of formation of new cells (kx) equals
These are the same kinetics used to describe the the rate of loss of cells from the outflow (Dx).
saturation of an enzyme by its substrate, that That is, kx = Dx, and therefore, k = D. Thus in
is Michaelis–Menten kinetics (Section 7.2.1). a chemostat, the growth rate of a culture can
Equation 2.9 can be used to calculate growth be changed merely by changing the flow rate, F,
yields during continuous growth (Section 2.3). which determines the dilution rate, D.
Dependence of cell yield, x, on concentration
2.5 Steady State Growth and of limiting nutrient in reservoir, Sr
Continuous Growth The actual concentration of cells in the growth
A culture is said to be in steady state growth (bal- chamber is manipulated by changing the con-
anced growth) when all its components double centration of limiting nutrient in the reservoir.
at each division and maintain a constant ratio
with respect to one another. Steady state growth
is usually achieved when a culture is maintained
in exponential growth by subculturing (i.e.,
when it is not allowed to enter stationary phase) medium reservoir
or during continuous growth.

2.5.1 The chemostat flow rate regulator

Continuous growth takes place in a device called
a chemostat (Fig. 2.8). A reservoir continuously air
feeds fresh medium into a growth chamber at a
flow rate (F) set by the operator. The chemostat
volume is kept constant. The concentration of
the limiting nutrient in the reservoir is kept very siphon overflow
low so that growth is limited by the availability growth chamber
of the nutrient. When a drop of fresh medium
enters the growth chamber, the growth-limiting
nutrient is immediately used up and the cells Fig. 2.8 A chemostat. The device works because the
cannot continue to grow until the next drop of growth of the cells in the growth chamber is limited
medium enters. This set of conditions allows by the rate at which a particular limiting nutrient
one to manipulate the growth rate by adjust- (e.g., glucose) is supplied. That is, the concentration
ing the rate at which fresh medium enters the of the limiting nutrient in the reservoir is low enough
growth chamber. to ensure that there is never an excess in the growth
chamber. At each moment when fresh medium
The dilution rate, D comes into the growth chamber, the incoming lim-
The dilution rate is equal to F/V, where F is the iting nutrient is rapidly utilized and growth cannot
flow rate and V is the volume of medium in the continue until fresh nutrient enters.
Free ebooks ==>
growth and cell division 69

Let Sr be the concentration of limiting nutrient a very complex process, which must be care-
in the reservoir and S be the concentration of the fully coordinated. Most research attention has
nutrient in the growth chamber. The difference, been focused on the gram-negative bacteria E.
Sr – S, must be equal to the amount of nutrient coli (and the closely related Salmonella typh-
used up by the growing bacteria. If we define imurium), and Caulobacter crescentus, and the
the growth yield constant Y as being equal to gram-positive bacterium Bacillus subtilis.
the mass of cells (x) produced per amount of
nutrient used up (Sr – S), then Y = x/(Sr – S) or, 2.6.1 The period before septum
rearranging, formation
x = Y(Sr − S) (2.10) When E. coli reaches a critical cell mass during
growth, DNA synthesis is initiated. During rep-
where S can be calculated from Eq. 2.9 by lication, the sister chromosomes move toward
substituting D for k. Since S is much smaller opposite halves of the cell, so that when cell divi-
than Sr, it can be conveniently ignored in most sion takes place, each daughter cell receives a
calculations. chromosome. Rapidly growing E. coli requires
Varying the cell density (x) and growth rate approximately 40 min (the C period) to repli-
constant (k) independently cate its chromosome, followed by a period (the
Equation 2.10 predicts that if the concentra- D period) of about 20 min before cell division
tion of rate-limiting nutrient in the reservoir (Sr) is complete.
increases, the cell density in the growth cham-
ber will increase. When Sr increases, more nutri- Septum formation
ent enters the growth chamber per unit time. The first morphological sign of cell division is
Initially, S must increase. Since the growth rate the centripetal synthesis of a septum at mid-
is limited by the concentration of S (Fig. 2.7), cell, which forms by inward growth of the cell
the cells will respond initially by growing faster membrane and peptidoglycan layers. In E. coli,
and the cell density will increase. A new steady this begins soon after chromosome replication
state will be reached in which the increased is complete (i.e., at the end of the C period).
number of cells can and does use up all the avail- In many gram-negative bacteria, including E.
able nutrient as it enters. Thus, even though Sr coli, the outer envelope also invaginates at this
is increased, the dilution rate still controls the time (i.e., cell separation and septation occur
growth rate. What has been accomplished, at approximately the same time) so that septa
therefore, when Sr is increased, is a higher steady usually are not seen in thin sections of cells pre-
state value of x, according to eq. 2.10, but no pared for electron microscopy. Invagination is
change in the growth rate constant (k). There manifested as a constriction, which becomes
are two conclusions to be drawn: narrower as septation proceeds (Fig. 2.9A).
Initially there is a double layer of peptidoglycan
1. The only way to change the steady state
in the septum, but this is enzymatically split in
growth rate constant is to change the dilu-
half as constriction proceeds and the cells sepa-
tion rate (D), because this changes the rate
rate. Septation and peptidoglycan synthesis are
of supply of S.
coupled; that is, inhibitors of septal peptidogly-
2. The only way to change the steady state
can synthesis prevent septation.
growth yield is to change the concentration
In E. coli, septation occurs during approxi-
of limiting nutrient in the reservoir, Sr.
mately the first 10 to 13 min of the D period,
depending upon the growth rate (slower grow-
2.6 Cell Division ing cells have a longer period of septation).
Cell division is the splitting of a mother cell into Daughter cells separate within 7 min of comple-
two daughter cells separated by a septum. For tion of the septum.
bacteria that grow as single cells, cell division Cell division in B. subtilis differs from that in
is accompanied by, or followed by, cell separa- E. coli in that septation precedes cell separation,
tion. There are many proteins involved in the although cell separation may begin before sep-
division of a mother cell into two daughter cells tation is complete (Fig. 2.9B). The result of all
and their subsequent separation. This is clearly of this is that under electron microscopy, thin
Free ebooks ==>
70 the physiology and biochemistry of prokaryotes

A E. coli outer membrane

cell membrane

B B. subtilis cell wall

cell membrane

Fig. 2.9 Cell division in gram-negative and gram-positive bacteria. (A) Many gram-negative bacteria divide
by constriction as the outer membrane invaginates with the cell membrane. Thus, septation and cell separa-
tion occur together. (B) In gram-positive bacteria, a septum consisting of cell membrane and peptidoglycan
forms, but there is generally little constriction in the early stages. Thus, septation precedes cell separation.

sections of dividing B. subtilis cells are seen to also important for septum formation. The FtsZ
have partially completed septa, and there is no ring with the other proteins is called a septal
visible constriction at the beginning of septation. ring. Figure 2.10 is a model of how the cell divi-
As in gram-negative bacteria, cell separation in sion proteins might be arranged at the site of
B. subtilis (and other gram-positive bacteria) is septation.
effected by the enzymatic splitting of the sep-
tum. Under certain conditions of rapid growth, 2.6.3 A model for the sequence of events in the
septum formation in B. subitilis can actually be assembly of the proteins at the division site
completed before cell separation begins, and as A model for the sequence of events in the
a consequence, chains of cells grow attached to assembly of the cell division proteins at the
each other via their septa. division site for E. coli postulates an ordered
sequence of events during which the prior
2.6.2 Proteins required for septum
formation and cell division
Most research in the area has focused on E. coli. LP
Several of the cell division genes are called fts OM

genes, for filamentation temperature-sensitive N PG

phenotype. At the restrictive temperature, the PT
cells do not divide but grow into long filaments N N FtsA
that eventually lyse. There are several proteins CP

in E. coli known to be associated with the septal


FtsZ ring
ring and essential for cell division. They include C
FtsZ, FtsA, ZipA, FtsK, FtsQ, FtsL, FtsB (also
called YgbQ), FtsW, FtsI (also called PBP3),
and FtsN. Fig. 2.10 Septal ring model in which a ring of FtsZ
FtsZ and FtsA are cytoplasmic, and the oth- proteins is connected to the cell membrane by ZipA.
Other abbreviations as follows: PT, FtsK and FtsW
ers are inner membrane proteins associated at
division proteins; BT, FtsL, FtsN, and FtsQ division
the constriction site or the septum. Importantly, proteins; PP, periplasm; CP, cytoplasm; LP, lipopro-
FtsZ forms a ring in the cell center. The FtsZ tein; OM, outer membrane; PG, peptidoglycan; IM,
ring is made before invagination and is believed inner (cell) membrane. Source: Hale, C. A., and P. A.
to contract, leading to the constriction of the J. de Boer. 1997. Direct binding of FtsZ to ZipA, an
cell in the center. As we shall see shortly, other essential component of the septal ring structure that
cell division proteins join the FtsZ ring and are mediates cell division in E. coli. Cell 88:175–185.
Free ebooks ==>
growth and cell division 71

localization of the earlier added proteins is A variety of physiological and morphologi-

necessary for the recruitment of the later added cal changes take place in bacteria when they
proteins. First FtsZ forms the ring, with the are subjected to nutrient limitation and enter
help of ZipA.43 Then FtsA is recruited by FtsZ stationary phase. These may include the induc-
and joins the ring, followed by FtsK. Next to tion of specific uptake systems to scavenge the
join is FtsQ, followed by FtsL and FtsB; then environment for limiting nutrients or ions,
FtsW is added, followed by FtsI, and last FtsN. sporulation, or encystment. Even bacteria that
The assembly of FtsL and FtsB is codependent. do not sporulate or encyst undergo significant
FtsQ is found only in bacteria with a cell wall, physiological changes when they are starved.
and it may be involved with cell wall synthesis. Cells may become metabolically less active and
Shortly after or around the time of the addi- more resistant to environmental stresses such
tion of FtsN, constriction begins at the septal as heat, osmotic stress, and certain chemicals
site. In B. subtilis the situation is different, (e.g., hydrogen peroxide). Bacteria starved for a
inasmuch as the proteins are recruited to the required amino acid, or for a carbon and energy
septal ring in a concerted manner, rather than source (stringent response), may also undergo
sequentially. inhibition of ribosomal RNA synthesis, an out-
come that is correlated with increased synthe-
sis of guanosine tetra- and pentaphosphates.
2.7 Summary The response to starvation is mediated in part
Growth is defined as an increase in mass and by an increase in the transcription of certain
can be conveniently measured turbidometri- genes. The transcription of many of these genes
cally. Other methods can be used, provided they requires σs, also called σ38. Cells become smaller
measure something that parallels mass increase, and may change their morphology when they
such as the rate of increase in cell number (via- are starved. There may also be changes in the
ble or total) or specific macromolecules (e.g., chemistry of the cell surface.
protein, DNA, RNA). Steady state (continuous) When cells are grown at different growth rates
growth is defined as the growth of a population because of nutritional alterations or chemo-
of cells during which all the components of the stat growth, the macromolecular composition
cell double at each division. changes. Faster growing cells are larger, have
Growth in batch culture can progress through proportionally more ribosomes, and have more
a lag and a log to stationary phase, where net DNA per cell. This observation can be rational-
growth of the population (measured as mass or ized in terms of an approximate constancy of
its equivalent) has ceased. Eventually, the viable ribosome efficiency in protein synthesis, and an
cells may decrease in number, and this is referred almost constant period between the initiation
to as the death phase. The availability of nutri- of DNA replication and cell division.
ents, the need to synthesize specific enzymes to Diauxic growth is characterized by two
metabolize newly encountered nutrients, and phases of population growth separated by a
the accumulation of toxic end products in the stationary phase during which the cells are
medium can explain the different stages of the incubated with certain pairs of carbon and
growth curve. In addition, when cells enter the energy sources. In diauxic growth with glu-
stationary phase, nutrient depletion causes spe- cose, for example, the cells grow on the glucose
cific adaptive physiological changes to occur. first because it represses the expression of the
Growth during the log phase can be described genes necessary to grow on the second carbon
by a simple exponential equation depicting a source and because it prevents the uptake of
first-order autocatalytic process. That is, the other sugars. Repression under these condi-
mass doubles at each generation. From this tions is called catabolite repression or glucose
equation one can derive a generation time and repression. Prevention of sugar uptake is called
an instantaneous growth rate constant. These inducer exclusion. There are at least two pos-
constants are used to characterize growth under sible rationales for this. Glucose is the most
different physiological situations and to predict widely used carbon and energy source, and cells
growth yields at specific times for experimental are better able to outgrow their neighbors if
purposes. they use the glucose first. In doing so, moreover,
Free ebooks ==>
72 the physiology and biochemistry of prokaryotes

they lower the supply of glucose to other cells. density will be 2 × 108/mL. Assume a genera-
Furthermore, many bacteria always express the tion time of 3.5 h. What should be the dilu-
genes to metabolize glucose (i.e., they are con- tion? What size inoculum should be used?
stitutive); the genes to metabolize other carbon
ans. 1/500 or 2 × 10–3; 2 mL
sources, however, are often not expressed unless
the carbon source is present in the medium. This 5. Assume the yield coefficient for glucose
lowers the energy burden of expressing genetic (Yglucose) is 0.5 g of cells per gram of glucose
information. However, glucose is certainly not consumed. In a glucose-limited chemostat,
a universal catabolite repressor among the bac- what should be the concentration of glu-
teria. For example, several obligately aerobic cose in the reservoir to produce a mass of
bacteria (e.g., Rhizobium) are known to grow cells in the growth chamber (x) of 0.1 mg/
preferentially on organic acids such as succinate, mL? For this problem, ignore S in eq. 2.9
malate, and fumarate, when given a mixture because it is much smaller than Sr.
of glucose and one of these acids. This makes
ans. 0.2 mg/mL
physiological sense, since they grow faster on
the C4 carboxylic acids than on glucose. 6. In a 500 mL chemostat, what should be the
Cells can be grown in continuous culture by flow rate (F) in milliliters per minute for a
using a chemostat. Two advantages to growing generation time (g) of 6 h?
cells this way are as follows: (1) the cells can
ans. 0.95 mL/min
be maintained in balanced growth and (2) the
growth rate constant can be easily changed sim- 7. Suppose you were operating a chemostat
ply by altering the flow rate. The growth yields with S as the limiting nutrient. Assume that
can be separately manipulated by changing the D is 0.2 h–1, Ks is 1 × 10–6 M, and kmax is
concentration of limiting nutrient in the reser- 0.4 h–1.
voir. Growth of continuous cultures is possible
a. What is the concentration of S in the
when the growth rate of the culture is limited by
growth chamber?
the supply of a nutrient that is continuously fed
into the growth chamber. b. If the cell density were 0.25 mg/mL,
what would be the concentration of S
in the reservoir for YS = 0.5?
Study Questions
c. What is the concentration of S in the
1. What is the generation time (g) of a culture growth chamber when the cells are grow-
with a growth rate constant (k) of 0.01 ing only half as fast (i.e., D = 0.1 h–1)?
min–1? ans. a. 1 × 10–6 M; b. 0.5 mg/mL; c. 3.3
ans. 69.3 min × 10–7 M
2. Assume you want to grow a culture to 108 8. During which phases of population growth
cells/mL in 3 h. The generation time is 30 would you not use cell number as an indica-
min. What should be the starting cell den- tor of growth?
sity (x0)? 9. What is the physiological role of RpoS?
ans. 1.6 × 106 When is it made?
3. Assume an inoculum whose cell density is
108/mL. The generation time is 30 min. If REFERENCES AND NOTES
you started with a 10–2 dilution, how many
hours would you have to grow the culture 1. Nyström, T. 2004. Stationary-phase physiology.
to reach 108/mL? Annu. Rev. Microbiol. 58:161–181.

ans. 3.3 h 2. Siegele, D. A., and R. Kolter. 1992. Life after log.
J. Bacteriol. 174:345–348.
4. Assume you have a stock culture at 5 × 109 3. Matin, A. 1991. The molecular basis of carbon-
cells per milliliter and you wish to inoculate starvation-induced general resistance in Escherichia
1 liter of fresh medium so that in 15 h the cell coli. Mol. Microbiol. 5:3–10.
Free ebooks ==>
growth and cell division 73

4. Kolter, R., D. A. Siegele, and A. Tormo. 1993. the carboxy-terminal end from β-galactosidase.
The stationary phase of the bacterial life cycle. Annu. The hybrid protein has β-galactosidase activity.
Rev. Microbiol. 47:855–874. Therefore, an assay for β-galactosidase is a measure
of the expression of the target gene. Thus, one can
5. Hengge-Aronis, R. 1996. Regulation of gene
measure the expression of virtually any gene simply
expression during entry into stationary phase, pp.
by constructing the proper gene fusion and perform-
1497–1512. In: Escherichia coli and Salmonella:
ing an assay for β-galactosidase. One can construct
Cellular and Molecular Biology. F. C. Neidhardt et
gene fusions in vitro or in vivo. In vitro construction
al. (Eds.). ASM Press, Washington, DC.
involves using restriction endonucleases to cut from
6. Kjelleberg, S. 1993. Starvation in Bacteria. a plasmid containing the cloned DNA a portion of
Plenum Press, New York and London. the gene with its promoter region. The excised por-
tion is ligated to a lacZ gene, without its promoter,
7. This is especially obvious among the cytophages
or ribosome-binding site, in a second plasmid. The
and flexibacteria, the latter sometimes decreasing in
plasmid containing the fused gene is then intro-
length from over 100 μm to approximately 10 to
duced into the bacterium, and transformants are
30 μm.
selected on the basis of resistance to an antibiotic-
8. Kjelleberg, S., M. Hermansson, and P. Marden. resistant marker on the plasmid and the production
1987. The transient phase between growth and non- of β-galactosidase.
growth of heterotrophic bacteria, with emphasis
17. Fang, F. C., S. J. Libby, N. A. Buchmeier, P. G.
on the marine environment. Annu. Rev. Microbiol.
Loewen, J. Switala, J. Harwood, and D. G. Guiney.
1992. The alternative sigma factor KatF (RpoS) reg-
9. McCann, M. P., J. P. Kidwell, and A. Matin. 1991. ulates Salmonella virulence. Proc. Natl. Acad. Sci.
The putative σ factor KatF has a central role in devel- USA 89:11978–11982.
opment of starvation-mediated general resistance in
18. Wilmes-Riesenberg, M. R., J. W. Foster, and R.
Escherichia coli. J. Bacteriol. 173:4188–4194.
Curtiss III. 1997. An altered rpoS allele contributes
10. Loewen, P. C., and R. Hengge-Aronis. 1994. The to the avirulence of Salmonella typhimurium LT2.
role of the sigma factor σs (KatF) in bacterial global Infect. Immun. 65:203–210.
regulation. Annu. Rev. Microbiol. 48:53–80.
19. Hengge-Aronis, R. 1996. Back to log phase: σs
11. O’Neal, C. R., W. M. Gabriel, A. K. Turk, S. J. as a global regulator in the osmotic control of gene
Libby, F. C. Fang, and M. P. Spector. 1994. RpoS is expression in Escherichia coli. Mol. Microbiol.
necessary for both the positive and negative regula- 21:887–893.
tion of starvation survival genes during phosphate,
20. Gourse, R. L., T. Gaal, M. S. Bartlett, J. A.
carbon, and nitrogen starvation in Salmonella typh-
Appleman, and W. Ross. 1996. rRNA transcription
imurium. J. Bacteriol. 176:4610–4616.
and growth rate-dependent regulation of ribosome
12. Hengge-Aronis, R. 2000. The general stress synthesis in Escherichia coli. Annu. Rev. Microbiol.
response in Escherichia coli, pp. 161–178. In: 50:645–677.
Bacterial Stress Responses. G. Storz and R. Hengge-
21. Cashel, M., D. R. Gentry, V. J. Hernandez, and
Aronis (Eds.). ASM Press, Washington, DC.
D. Vinella. 1996. The stringent response, pp. 1458–
13. Price, C. W. 2000. Protective function and regu- 1496. In: Escherichia coli and Salmonella: Cellular
lation of the general stress response in Bacillus subtilis and Molecular Biology, Vol. 1. F. C. Neidhardt et al.
and related gram-positive bacteria, pp. 179–197. In: (Eds.). ASM Press, Washington, DC.
Bacterial Stress Responses. G. Storz and R. Hengge-
22. In addition to an inhibition of ribosomal RNA
Aronis (Eds.). ASM Press, Washington, DC.
synthesis, ppGpp seems to be responsible for the fol-
14. Mulvey, M. R., J. Switala, A. Borys, and P. lowing:
C. Loewen. 1990. Regulation of transcription of
a. A large decrease in the rate of synthesis of pro-
katE and katF in Escherichia coli. J. Bacteriol. 172:
b. A temporary cessation in the initiation of new
15. Gentry, D. R., V. J. Hernandez, L. H. Nguyen,
rounds of DNA replication.
D. B. Jensen, and M. Cashel. 1993. Synthesis of the
stationary-phase sigma factor σs is positively regu- c. An increase in the biosynthesis of amino acids.
lated by ppGpp. J. Bacteriol. 175:7982–7989.
d. A decrease in the rates of synthesis of phospho-
16. Gene fusions are valuable probes to monitor lipids, nucleotides, peptidoglycan, and carbo-
the expression of genes of interest. Consider a lacZ hydrates.
fusion. The fused gene has the promoter region of
e. A stimulation of development in the myxobac-
the target gene but not the promoter for the lacZ
terium Myxococcus xanthus.
gene. Expression of the fused gene is therefore under
control of the promoter region of the target gene. f. A stimulation of expression of virulence genes
The fusions produce a hybrid protein; its amino- in Salmonella. (See Pizzaro-Cerdá, J., and K.
terminal end is derived from the target gene and Tedin. 2004. The bacterial signal molecule,
Free ebooks ==>
74 the physiology and biochemistry of prokaryotes

ppGpp, regulates Salmonella virulence gene is believed to be able to prevent translation of the
expression. Mol. Microbiol. 52:1827–1844.) entire operon when the protein is present in excess. It
is therefore a translational repressor.
g. Stimulation of the synthesis of sigma factor
σs, which is induced during starvation. (See The repressor ribosomal protein can bind not
Mulvey, M. R., J. Switala, A. Borys, and P. C. only to rRNA but also to the initiating region of the
Loewen. 1990. Regulation of transcription of mRNA of one of the genes (usually the first one) in
katE and katF in Escherichia coli. J. Bacteriol. the operon. It appears that in at least some instances
172:6713–6720.) the sequence in the mRNA to which the repressor
protein binds is similar to the sequence in the rRNA,
One must therefore conclude that the increase in lev-
thus accounting for the ability of the repressor ribo-
els of ppGpp and the responses due to those increases
somal protein to bind to both the ribosomal RNA
can be viewed as general responses to conditions
and the mRNA. As an example of such translational
that limit growth, not simply as results of amino acid
autoregulation, consider translation of the rplK–
starvation. This is because ppGpp levels rise as a con-
rplA operon, which encodes the 50S ribosomal sub-
sequence of any nutrient or energy limitation (e.g.,
unit proteins L11 (rplK) and L1 (rplA). The rplA gene
nitrogen limitation) or of a shift to a poorer carbon
is downstream of the rplK gene, and its translation
or energy source. Despite the obvious importance of
requires prior translation of the rplK gene. The L1
ppGpp or one or more metabolically related com-
protein regulates the translation of the entire operon.
pounds in mediating the stringent response, little is
It can bind to either rRNA or the mRNA in the trans-
known concerning how ppGpp is responsible for
lational initiation region of the rplK gene, and it can
the myriad effects that appear to be correlated with
block translation not only of the rplK gene but also of
increase of this molecule. One might think of ppGpp
the rplA gene that follows it. All the ribosomal pro-
as a second messenger that receives a signal from the
tein operons are regulated in a similar manner by one
environment (i.e., nutrient or energy depletion) and
of the protein products.
transfers the signal to the genome, either activating or
inhibiting transcription of relevant genes, or affect- 25. The ppGpp can be synthesized on ribosomes that
ing some other cellular process such as the activity of are “stalled” because of a restriction in the supply of
an enzyme. Of course, ppGpp need not act directly aminoacylated tRNA. This occurs during amino acid
on the target, and perhaps it is one component in a starvation (e.g., by restricting amino acids to aux-
longer signal transduction sequence. otrophs) and requires the product of the relA gene.
The relA gene was discovered in a search for mutants
23. Potrykus, K., and M. Cashel. 2008. (p)ppGpp:
that failed to respond to amino acid starvation with
Still magical? Annu. Rev. Microbiol. 62:35–51.
the stringent response. These mutants were termed
24. Ribosomal RNA and ribosomal proteins are “relaxed,” hence the name of the gene. The RelA
synthesized independently of one another and then protein [also called (p)ppGpp synthetase I or PSI] is
assembled into ribosomes. However, the synthesis of a ribosome-associated protein that synthesizes either
ribosomal proteins depends on continued synthesis ppGpp or pppGpp by displacing AMP and transfer-
of rRNA. Hence, a decrease in the synthesis of rRNA ring a pyrophosphoryl group from ATP to the 3′-OH
leads to an inhibition of ribosomal protein synthe- of either GDP or GTP:
sis. This is because certain ribosomal proteins, in the
absence of rRNA to which they normally bind, will GDP + ATP → ppGpp + AMP
bind to ribosomal mRNA and inhibit translation.
Because the pool size for GDP in E. coli is small
To understand this, it is important to recognize that
relative to GTP, it is believed that the product is
ribosomal proteins are encoded in operons in which
pppGpp, which is then dephosphorylated by a
the translation of the genes is coupled. E. coli has
5′-phosphohydrolyase (gpp) to ppGpp. This sugges-
at least 20 such operons that encode the ribosomal
tion is in agreement with the kinetics of appearance
proteins, and in certain instances also DNA primase,
of pppGpp and ppGpp. The synthetase is activated
RNA polymerase, and elongation factors.
in starved (stalled) ribosomes when aminoacylated
In translational coupling, the operon produces a tRNA becomes limiting during amino acid starvation.
polycistronic mRNA, and the translation of the gene The levels of (p)ppGpp may also rise upon carbon
immediately upstream is required for translation of and energy starvation (e.g., glucose starvation) even
the downstream gene. A model of translational cou- in null relA mutants, indicating alternative ways to
pling for the downstream gene postulates that the synthesize (p)ppGpp. Thus, even in a relaxed strain,
translational start region, including the start codon, RNA accumulation stops when the cells are shifted
is inside a hairpin loop in the mRNA and thus can- from a good carbon and energy source to a poorer
not bind to a ribosome. However, when the ribo- medium (but not when starved for an amino acid).
some translating the upstream gene comes to the stop This constitutes the evidence for a relA-independent
codon of the upstream gene, the ribosome disrupts (p)ppGpp synthetase, called PS II. PS II is thought to
the secondary structure of the mRNA and “opens be encoded by the spoT gene and is believed to be
it up,” making the translational start region for the a 3′-pyrophosphotransferase that uses GTP as the
downstream gene accessible to a second ribosome. acceptor and synthesizes pppGpp. SpoT is also a
One of the ribosomal proteins encoded by the operon pyrophosphohydrolase that degrades (p)ppGpp to
Free ebooks ==>
growth and cell division 75

GDP and GTP. In other words, it has been proposed decreases (but less so) as the growth rate increases:
that SpoT is a bifunctional enzyme that takes part from 30 min at 0.6 doubling per hour to 23 min at 2.5
in either the synthesis or the degradation of ppGpp, doublings per hour. See: Bipatnath, M., P. P. Dennis,
and the ratio of biosynthetic to degradative activity is and H. Bremer. 1998. Initiation and velocity of
somehow controlled by energy starvation. The result chromosome replication in Escherichia coli B/r and
is an increase during energy starvation of ppGpp. For K–12. J. Bacteriol. 180:265–273. Bremer, H., and P.
a discussion of these points, see ref. 21. P. Dennis. 1996. Modulation of chemical composi-
tion and other parameters of the cell by growth rate.
26. Xu, J., and R. C. Johnson. 1995. Identification
In: Escherichia coli and Salmonella typhimurium:
of genes negatively regulated by Fis: Fis and RpoS
Cellular and Molecular Biology, 2nd ed., Vol. 2, pp.
comodulate growth-dependent gene expression in
1553–1569. F. C. Neidhardt et al. (Eds.). ASM Press,
Escherichia coli. J. Bacteriol. 177:938–947.
Washington, DC.
27. Fis negatively regulates the expression of the
fis gene and positively or negatively regulates the 33. For illustrative purposes, let us assume that the C
expression of several other genes. Two-dimensional and D periods are constants at 40 and 20 min, respec-
gel electrophoresis of fis mutants has revealed over tively, so that a cell divides 60 min after it has initi-
20 proteins whose synthesis is decreased (indicating ated DNA replication. We will also assume that each
Fis-dependent transcription) and a similar number of cell grows exponentially and that x = x02Y, where x0
proteins whose synthesis has increased (indicating Fis is the mass at birth and x is the mass at a point in the
repression). The evidence suggests that Fis binds to cell cycle where Y is the fraction of cell cycle time. For
both the rRNA DNA and the RNA polymerase and example, if the cell is halfway through the cell cycle,
stimulates transcription by interacting with the poly- x = x020.5. Let us call a cell dividing every 60 min a 60
merase rather than by causing an alteration in DNA min cell, a cell dividing every 40 min a 40 min cell,
structure such as bringing upstream DNA closer to and a cell that divides every 30 min a 30 min cell.
the polymerase. See refs. 20 and 26. The 60 min cell must be born with one chromosome
(and one replication origin) and will begin replicat-
28. Ball, C. A., R. Osuna, K. C. Ferguson, and R. C. ing that chromosome at the beginning of the cell cycle
Johnson. 1992. Dramatic changes in Fis levels upon so that 60 min later it divides. Let us say that the mass
nutrient upshift in Escherichia coli. J. Bacteriol. of the 60 min cell at the beginning of the cell cycle
174:8043–8056. when it begins replicating the chromosome is x0. This
29. Reviewed in: Wagner, R. 1994. The regula- means that the mass (x) 20 min later (when, as we
tion of ribosomal RNA synthesis and bacterial cell shall see, the 40 min cell begins DNA replication) is
growth. Arch. Microbiol. 161:100–109. about 1.15x0 (x = x020.2). The 40 min cell also needs
60 min between the initiation of a round of DNA
30. In addition to being a transcriptional activator replication and cell division. This means that the 40
for rRNA genes, Fis stimulates Hin-, Gin-, and Cin- min mother cell must initiate DNA synthesis at two
mediated inversion and site-specific recombination replication origins (one for each of the daughter cell
of phage λ (both excision and integration) with the chromosomes) 20 min into the cell cycle. Now if you
bacterial chromosome. Fis also binds to the E. coli compare the 40 min cell with the 60 min cell, you
origin of replication (oriC) and possibly plays a role will realize that the 40 min cell must on average be
in the initiation of DNA replication. H-NS is involved larger. This is because by 20 min, the 40 min cell must
in transcriptional regulation of several genes whose grow to a mass that will initiate replication at two
activity is modulated by cellular stress (e.g., gene reg- origins of replication; that is, it must be 2x0, where x0
ulation in stationary phase, osmotic shock). See the is the size of the 60 min cell at birth. Since that size is
discussion of the nucleoid in Section 1.2.6. reached by the 40 min cell halfway into the cell cycle,
31. Cooper, S., and C. R. Helmstetter. 1968. the 40 min cell must be 1.41 times larger than its mass
Chromosome replication and the division cycle of at birth (x = x020.5 = x01.41). That is, 2x0 is 1.41 times
Escherichia coli B/r. J. Mol. Biol. 31:519–539. larger than the mass at birth of the 40 min cell. The
mass at birth must therefore be 2x0/1.41, or about
32. The length of time between the onset of DNA 1.41x0. This can be contrasted to the 60 min cell that
replication and the completion of cell division is the was born with a mass of x0. Now let’s look at the 30
sum of two time periods [i.e., the time it takes for the min cell. To ensure that the daughter cells can divide
chromosome to replicate (the C period) and the inter- 60 min later, the 30 min cell must begin replicating
val between the end of a round of DNA replication the daughter chromosomes at the beginning of the
and the completion of cell division (the D period)]. cell cycle. Thus, replication must begin at two origins
The shortest C and D periods are approximately 40 (one for each of the daughter chromosomes), and the
and 20 min, respectively, in rapidly growing E. coli. mass at the beginning of the cell cycle must be 2x0 as
Hence 60 min is the minimum time required between opposed to 1.41x0 for the 40 min cell and x0 for the
the onset of a round of DNA replication and the com- 60 min cell.
pletion of cell division. These times do change with
growth rate. The C period decreases from approxi- 34. Donachie, W. D. 1968. Relationship between
mately 67 min at 0.6 doubling per hour to about 42 cell size and time of initiation of DNA replication.
min at 2.5 doublings per hour. The D period also Nature 219:1077–1079.
Free ebooks ==>
76 the physiology and biochemistry of prokaryotes

35. Donachie assumed that each cell grew expo- 39. Collier, D. N., P. W. Hager, and P. V. Phibbs
nentially and doubled in mass during the cell cycle. Jr. 1996. Catabolite repression control in the
Knowing the size of the cell at the time of cell division, pseudomonads. Res. Microbiol. 147:551–561.
he was able to calculate the size at any time during the
cell cycle. With this information and the knowledge 40. Suppose you want to convert log2 to log10. First
that DNA initiation occurred 60 min before cell divi- you would write the exponential equation (e.g., x =
sion, he was able to compute the mass of the cell at the x02Y). Now take log10 of both sides of the equation:
time of initiation of DNA initiation. This mass, divided log10x = log10x0 + log102(Y). Since log102 = 0.301, the
by the number of replication origins, was a constant. equation becomes log x = log x0 + 0.301Y. Note that
log10 is usually written simply as log.
36. Wold, S., K. Skarstad, H. B. Steen, T. Stokke,
and E. Boye. 1994. The initiation mass for DNA 41. Some investigators use the symbol μ for the
replication in Escherichia coli K-12 is dependent on instantaneous growth rate constant and k for the
growth rate. EMBO J. 13:2097–2102. reciprocal of the generation time: k = 1/g.
37. Cooper, S. 1997. Does the initiation mass for 42. The dilution is 2.6 × 102. Assume the volume of
DNA replication in Escherichia coli vary with the inoculum is x. Therefore, 2.6 × 102(x) must equal the
growth rate? Mol. Microbiol. 26:1138–1143. final volume, which is 1000 mL + x. Solving for x
gives 3.8 mL.
38. Chandra, N. M., and P. K. Chakrabartty. 1993.
Succinate-mediated catabolite repression of enzymes 43. Davjkovic, A. Pirchoff, S. J. Lutkenhaus, and
of glucose metabolism in root-nodule bacteria. Curr. D. Wirtz. 2010. Cross-linking FtsZ polymers into
Microbiol. 26:247–251. coherent Z rings. Mol. Microbiol. 78: 651–668.
Free ebooks ==>

Chromosome Replication and
Partitioning of Chromosomes

From a biochemical point of view, nucleic acids The DNA strands are held together by
are polymerized by donating the subunit from base pairing between A–T and G–C residues
a donor with a high group transfer potential to (Fig. 3.2). DNA is replicated via semiconserva-
the growing chain via a nucleophilic displace- tive replication. This means that each strand
ment reaction. This condensation reaction is acts as a template for the synthesis of a daugh-
generally referred to as chain elongation. For ter strand. Semiconservative replication was
nucleic acids, the donors of the nucleotide suggested by Watson and Crick in their paper
monophosphates, (d)NMP, are the respective on the structure of DNA published in Nature
nucleotide triphosphates, (d)NTP. During the in 1953, and was later experimentally demon-
condensation reaction in nucleic acid biosyn- strated by Meselson and Stahl, as described in
thesis, the α phosphate in (d)NTP is attacked by Fig. 3.3.
the 3′-hydroxyl group on ribose or deoxyribose
(at the 3′ end of the growing polynucleotide)
and the released pyrophosphate (PPi) is subse-
quently hydrolyzed by a pyrophosphatase. This
chapter describes the steps in chromosome rep-
lication, the separation of chromosomes, and
the deliverance of chromosomes to daughter

3.1 DNA Replication, Chromosome

Separation, and Chromosome
3.1.1 Semiconservative replication
DNA consists of two strands wound around
each other in a right-handed double helix
(Fig. 3.1). For a review of the history of DNA
research that led to the realization that it was
the genetic material, see Box 3.1; for a summary Fig. 3.1 The direction of helical turns. To determine
of the contributions of Francis Crick, Rosalind whether the molecule is in a right- or left-handed
Franklin, James Watson, and Maurice Wilkins helix, sight down the molecule from one end. The
toward the elucidation of the three-dimensional right-handed helix turns clockwise, and the left-
structure of DNA, see Box 3.2. handed helix turns counterclockwise.

Free ebooks ==>
78 the physiology and biochemistry of prokaryotes



The discovery of DNA (The S stands for smooth, which is how the
colonies look when grown on an agar plate,
The realization that DNA is the genetic and the R stands for rough, which describes
material was preceded by decades of patient the colonies formed by unencapsulated
investigation. After its identification in strains.)
1869 in the nuclei of human white blood Griffith found that if he heat-killed the
cells, DNA was later found in cells of many S bacteria, they did not cause pneumonia
types. Its chemical composition, as well unless they were injected with live R bac-
as that of RNA, was learned in 1910. By teria. Furthermore, he saw that all the bac-
the 1920s, microscopic studies of stained teria recovered from the dead mice had
preparations demonstrated that DNA, capsules. Thus, the R bacteria had been
along with protein, is contained in chro- transformed into S bacteria by dead S bac-
mosomes. However, since DNA is made teria. This is because genes to make capsule
of only four different nucleotides, and its were transferred from the dead S bacteria
enormous diversity was not suspected, this to the live R bacteria, a phenomenon now
nucleic acid was not then considered to be called transformation (since the R bacteria
the genetic material. In fact, it was protein are transformed into S bacteria).
that was thought to be the genetic material Avery, MacLeod, and McCarty were
because of the great diversity in proteins. It able to transform R bacteria into S bacteria
was thought that the DNA played a struc- in a test tube by adding DNA purified from
tural role in the chromosomes. S bacteria. They thus provided the first evi-
dence that genes were made of DNA. The
DNA preparations had small amounts of
DNA is the genetic material protein, and to prove the genetic material
was the DNA and not the protein, Avery
It was not until 1944 that experiments were and his coworkers showed that the trans-
published indicating that the genetic mate- forming activity was not destroyed by pro-
rial was indeed DNA. These experiments, teases, which are enzymes that degrade
aimed at investigating a report published protein, nor was the activity destroyed by
in 1928 by Fred Griffith, a British scien- an enzyme that degrades RNA. However,
tist, were done at the Rockefeller Institute the transforming activity was destroyed
in New York by Oswald Avery, Colin by deoxyribonuclease, an enzyme that
MacLeod, and Maclyn McCarty. Griffith, destroys DNA.
who had studied pneumonia caused by A second experiment, which also showed
the bacterium Streptococcus pneumoniae that the genetic material was made of
(at that time called Pneumococcus), had DNA, was published in 1952 by Alfred
described two strains of S. pneumoniae, Hershey and Martha Chase. Hershey and
called R and S. One difference between the Chase studied the infection of the bacte-
two strains was that when the S strain was rium Escherichia coli by the bacterial virus
injected into mice, the mice died of pneu- T2. When the virus infects the bacterium,
monia. However, when the R strain was the viral genes are injected into the bacte-
injected, the mice lived. Another difference rial cell, and those genes direct the synthesis
between the R and S strains was that the S of new viruses. Hershey and Chase demon-
strain had a capsule and the R strain did not. strated, via radioisotopes that labeled the
Free ebooks ==>
chromosome replication and partitioning of chromosomes 79

phosphorus in DNA and the sulfur in pro- Together, the experiments of Avery,
teins, that viral DNA and not viral protein MacLeod, and McCarty, and of Hershey
was injected into the cell, thus providing and Chase, convinced everyone that genes
evidence that its genes were made of DNA are made of DNA. (We now know that
and not protein. genes in some viruses are made of RNA.)


In 1953 James Watson and Francis Crick called the “A” form, which exists in solu-
at Cambridge University in England pub- tions that have relatively little water to
lished their model of DNA as a double hydrate the DNA. It was immediately clear
helix held together by hydrogen bonding to Watson from Franklin’s photograph that
between the bases. Based upon their model the B form was due to a helical structure.
of the structure of DNA, Watson and In 2004, in an article on the history of
Crick suggested that one of the two strands the discovery of the DNA double helix,2
serves as a template for a new complemen- Aaron Klug revealed that one of Rosalind
tary strand during DNA replication, and Franklin’s laboratory notebooks, written in
that each daughter cell receives a duplex 1952, showed that she also thought that the
consisting of a parental strand and a new B form was helical. Franklin’s photograph
strand. This process is known as semicon- helped Watson and Crick to construct the
servative replication, and its occurrence double-helix model of DNA that was pub-
was later demonstrated experimentally. lished in Nature in 1953. In 1962 Watson,
The Watson–Crick model of the structure Crick, and Wilkins received the Nobel Prize
of DNA was based in part upon the exami- for their contributions in deciphering the
nation of X-ray diffraction photographs of structure of DNA. Rosalind Franklin had
DNA taken by Rosalind Franklin, a research died of cancer in 1958, at the age of 37. For
fellow in the laboratory of Maurice Wilkins more information about the fascinating
at King’s College in London, as well as her history of the elucidation of the structure of
presentation at a colloquium in 1952 that DNA, read refs. 1 through 3.
showed that the phosphate groups were
on the outside of the DNA molecule and
the bases faced inward toward each other. REFERENCES
According to the account by Watson, in his 1. Watson, J. D. 1968. The Double Helix.
book The Double Helix,1 Wilkins showed Penguin Books Ltd., London.
Watson one of Franklin’s recent X-ray pho- 2. Klug, A. 2004. The discovery of the DNA
tographs of DNA molecules surrounded by double helix. J. Mol. Biol. 335:3–26.
a large amount of water. The DNA was in a 3. Kass-Simon, G., and P. Farnes (Eds.). 1990.
new form called the “B” form, which is far Women of Science. Indiana University Press,
simpler than the pattern obtained earlier, Bloomington.
Free ebooks ==>
80 the physiology and biochemistry of prokaryotes

Fig. 3.2 Base pairing in DNA. (A) Two complementary strands of the double helix are in opposite polarity
and are held together by hydrogen bonds between A–T and G–C pairs: A, adenine; T, thymine; G, guanine;
C, cytosine; P, phosphate. The deoxyribose moieties are attached via phosphodiester linkages between the C3
hydroxyl of one sugar and the C5 hydroxyl of another. (B) A more detailed examination showing the structure
of the deoxyribose connected by phosphodiester linkages. The phosphate–oxygen double bond is drawn as a
semipolar bond because of the high electronegativity of oxygen and the low propensity of phosphorus to form
double bonds. (C) The structures of the bases, showing the hydrogen bonds. Note that two hydrogen bonds
hold the A–T base pairs together, whereas three hydrogen bonds hold the G–C base pairs together.

3.1.2 The topological problem in and more tightly coiled into a positive super-
DNA replication coil (Fig. 3.5). Unless something was done to
DNA replication is a complex topological prob- prevent overwinding, the unreplicated portion
lem because the DNA in a typical bacterium of the DNA helix would become bunched up
exists as a covalently closed circle of a right- in positive supercoils and further unwinding
handed double helix that may be 500 to 600 would stop. As described shortly, an important
times longer than the cell and is tightly folded enzyme, DNA gyrase, solves the problem.
into supercoiled loops. (However, not all bacte-
ria have circular chromosomes. See note 1.) Positive versus negative supercoiling
Supercoiling refers to the twisting of the DNA What does this mean? Positively supercoiled
double helix around its central axis (Fig. 3.4). DNA has more than 10.5 base pairs per heli-
To visualize supercoiling, think of a telephone cal turn (i.e., it is overwound); negatively super-
cord wound into secondary coils. (See note 2 for coiled DNA has fewer than 10.5 base pairs per
a further explanation of supercoiling.) helical turn (i.e., it is underwound). The DNA
Pulling apart of the duplex strands at the duplex itself is a right-handed helix, and there-
replication fork in the closed circle makes the fore positive supercoils are twisted in the same
unreplicated portion ahead of the replication direction as the helix (overwound). If the twist
fork twist so that the helix becomes overwound of the coil is counterclockwise (left-handed),
Free ebooks ==>
chromosome replication and partitioning of chromosomes 81

15 15


grow in 14N for one


15 14 15 14

B N/14N

growth in 14N

15 14 14 14 15 14 14 14
C N/14N

Fig. 3.3. The Meselson–Stahl experiment (Meselson, M. and F. W. Stahl, The replication of DNA in
Escherichia coli, Proc. Natl. Acad. Sci. USA 44:671–82, 1958) that demonstrated that DNA is replicated
semiconservatively. (A) Cells are grown in 15N for many generations so that all the nitrogen in the DNA is
heavy (15N). The DNA is isolated and centrifuged to equilibrium in a cesium chloride (CsCl) density gra-
dient that separates the molecules according to their density. If both strands have 15N, the duplex DNA
sediments to a position near the bottom of the tube. (B) The 15N cells are then grown in 14N (light) media,
and after one generation the DNA is isolated and centrifuged in the CsCl gradient. Semiconservative rep-
lication predicts that “hybrid” DNA would be formed, one strand being labeled with 15N and the other
strand with 14N. The 15N/14N DNA occupies a position in the gradient higher than the 15N/15N DNA. (C)
Further growth in 14N results in the formation of “light” DNA (i.e., 14N/14N), which occupies the highest
position in the gradient.

the coil is negatively supercoiled (opposite to DNA ahead of the replication fork and converts
the helix), hence is underwound. them into negative supercoils (underwound
DNA). This activity is advantageous, since the
DNA gyrase solves the problem of duplex must unwind if DNA replication and
overwinding RNA transcription are to take place. As shown in
The problem of overwinding the double helix Fig. 3.6, DNA gyrase works by making a double-
during DNA replication is solved by using DNA stranded break in the DNA, passing an unbro-
gyrase, also called topoisomerase II. This enzyme ken portion through the gap, and resealing the
continuously removes the positive supercoils break. Note 3 provides more information about
(overwound DNA) that form in the unreplicated topoisomerases and how they work.
Free ebooks ==>
82 the physiology and biochemistry of prokaryotes

Fig. 3.4 Supercoiled DNA. (A) When there are 10.5 base pairs per helical turn, the DNA is relaxed. Native
DNA is generally underwound (i.e., >10.5 base pairs per helical turn). Underwinding introduces a strain in the
molecule; and to reduce the strain, the molecule twists upon itself to form supercoils (B). Supercoiling result-
ing from underwound DNA is referred to as negative supercoiling. Supercoils will also form if DNA is over-
wound (i.e., >10.5 base pairs per turn). Supercoiling due to overwound DNA is called positive supercoiling. In
negative supercoiling the DNA is twisted in a direction opposite to that of the right-handed double helix, and
in positive supercoiling the DNA is twisted in the same direction as the right-handed double helix.

More problems to be solved so that each strand can act as a template.

Not only must the DNA be unwound and cop- However, the unwinding does not begin at just
ied rapidly (to replicate a chromosome in E. coli, any place in the DNA, but rather at a particu-
about 1,000 nucleotides per second must be lar site in the DNA duplex termed the origin,
polymerized, in about 40 min), but the strands which should be remembered as the oriC locus.
must be separated from each other without get- When the duplex is unwound, a Y is created.
ting entangled. Finally, the strands must be par- The arms of the Y are single stranded because
titioned into the daughter cells. We will begin the duplex has become unwound there; but the
with the “replication fork,” which serves in the region downstream of the juncture, where the
initiation of DNA replication. The first ques- two arms come together, is still double stranded
tion is, how is it created? This turns out to be (Fig. 3.7). The juncture should be remembered
a very complicated process indeed, involving as the replication fork. DNA is usually repli-
many different proteins. cated bidirectionally; that is, there are actu-
ally two replication forks (Fig. 3.8). (Note 12
3.1.3 Creating the replication fork explains how directionality can be determined.)
At least 30 different proteins are required to ini- Bidirectional replication, which halves the time
tiate replication and to replicate the DNA in E. needed to replicate the DNA molecule, gener-
coli. For more information, the student should ally takes place with phages, plasmids, bacteria,
consult refs. 4 through 11. and eukaryotic cells. The replication forks are
DNA replication takes place in a complex thought to be stationary, and the unreplicated
DNA-synthesizing “factory” called a repli- DNA appears to thread through them. (This is
some, which consists of many enzymes and discussed in Section 3.1.4.) The detailed steps in
proteins that will soon be described. Within the creation of a replication fork are described
the replisome there are replication forks cre- next. Before DNA synthesis can begin, a pre-
ated on the DNA, and this is where replication priming complex must form. The first step is the
takes place. (Clearly, DNA synthesis must be formation of an “open complex,” which devel-
related to the growth rate of the cell. Review ops into the “prepriming complex.”
Section 2.2.3 for a discussion of the relation-
ship between the timing of initiation of DNA Unwinding the duplex: Creation of the
replication and the growth rate.) So how is the prepriming complex
replication fork made? To make the replica- The prepriming complex is formed first and
tion fork, the DNA strands must first unwind it is created in two stages.9 In stage 1 the open
Free ebooks ==>
chromosome replication and partitioning of chromosomes 83

Fig. 3.5 Supercoiling ahead of the replication fork. As the template strands in the closed circle are pulled
apart, the duplex ahead of the replication fork overwinds as positive supercoils are formed. The twists in the
overwound regions are removed by DNA gyrase, which produces negative supercoils and underwinds the

complex is formed, and in stage 2 the open influence nucleoid structure and gene expres-
complex develops into the prepriming complex. sion in Section 1.2.6. Within the origin (oriC)
DNA synthesis, which will be described later, there are multiple sites to which DnaA binds.
actually begins with the prepriming complex. Approximately 30 DnaA molecules bind to
the sites as the DNA wraps around a core of
1. Formation of the open complex DnaA molecules. Then, in an ATP-dependent
Creation of the open complex is initiated at reaction that is aided by HU, the adjacent A–T-
the origin of replication (oriC) with ATP and rich region at the 5′ end of the origin sequence
two DNA-binding proteins: DnaA and a his- unwinds to form the 45 bp open complex.
tonelike protein called HU (Fig. 3.9). See the However, something must be done to prevent
subsection entitled DNA-binding proteins the single strands from coming together again
Free ebooks ==>
84 the physiology and biochemistry of prokaryotes

Fig. 3.6 The action of DNA gyrase: converting positive supercoils into negative supercoils. (A) A positively
supercoiled node; that is, the duplex is twisted around its central axis. For example, this happens during DNA
replication as the duplex is being unwound by helicase at the replication fork and the duplex ahead of the rep-
lication fork spontaneously becomes overwound by being twisted in a clockwise direction. (B) Both strands
are cut by DNA gyrase, and then the gyrase passes the uncut portion through the gap and the gap is sealed. (C)
The duplex is now twisted in the opposite direction; that is, it is negative supercoiled and underwound. DNA
gyrase, an ATP-dependent enzyme, is sometimes referred to as providing a “swivel” that allows the replica-
tion fork to continue.

Fig. 3.7. A replication fork. The two template strands on the left have become unwound and are being copied.
The unwinding is caused by DnaB protein, which is the replication fork helicase (not shown) and requires
ATP. Note that there are single-stranded regions near the fork. A DNA-binding protein, called SSB (not
shown), binds to the single-stranded regions, preventing them from coming together. Now (right) the duplex
has both unwound and double-stranded components.

to re-form the duplex. This task is accomplished bind to the DNA. However, DnaB does not bind
by single-stranded binding (SSB) proteins that to the DNA on its own but must be transferred
coat the strands. to the open complex from a DnaB:DnaC:ATP
complex. After binding to the DNA, DnaB
2. Formation of the prepriming complex further unwinds the strands bidirectionally to
The open complex unwinds into the preprim- form the prepriming complex, with two rep-
ing complex. The unwinding is performed by a lication forks ready for the initiation of DNA
protein called helicase (DnaB), which must first replication.
Free ebooks ==>
chromosome replication and partitioning of chromosomes 85

(Pol II), encoded by polB; DNA polymerase III

(Pol III), encoded by polC (dnaE); DNA poly-
merase IV (Pol IV), encoded by dinB; and DNA
polymerase V (Pol V), encoded by umuC. Here
is what they do.
1. DNA polymerase I has a role in DNA rep-
lication, which will be discussed later. (See
subsection entitled Attaching the Okazaki
fragments to each other.) DNA polymerase I
is also important for DNA repair.
2. DNA polymerase II functions in the repair
Fig. 3.8. Bidirectional replication of DNA: solid of damaged DNA. After UV damage, DNA
lines, template strands; broken lines, daughter polymerase II catalyzes a fast reinitiation of
strands. In most organisms and viruses, the DNA is synthesis called “replication restart,” and
replicated bidirectionally, as indicated by the arrows. then DNA polymerase III takes over.
Bidirectionality reduces the time required to replicate 3. DNA polymerase III replicates DNA at the
the DNA. Available data indicate that the replication fork. See note 14 for a description of the sub-
forks are stationary in a replication factory called units in DNA polymerase III.
a replisome, and the unreplicated DNA threads 4. DNA polymerase IV functions in repairing
through them.
damaged DNA.
5. DNA polymerase V functions during DNA
repair. DNA polymerase V inserts nucle-
Timing of initiation otides nonspecifically opposite lesions, such
Precisely what determines the timing of ini- as abasic sites and UV-induced thymine
tiation of bacterial DNA replication is still a dimers, in the template strand. The process
matter of speculation. As discussed in Section is called translesion synthesis, and it is asso-
2.2.3, the cell mass per chromosomal replica- ciated with increased mutations because
tion origin (as opposed to plasmid replication the nucleotides are inserted nonspecifically.
origins) at the time of initiation is constant, Because of the increase in mutation rate,
and all oriC origins (even plasmid origins), translesion synthesis is referred to as SOS
begin replication at the same time. This mass mutagenesis or error-prone repair.
is called the initiation mass or the initiation
volume, and some have suggested that it cor- The problem in synthesizing
responds to the activity of DnaA. Another strands of opposite polarity at the
question is, what prevents the newly repli- replication fork
cated origins from reinitiating in the same At the replication fork the DNA templates are
cell cycle? For a discussion of what regulates antiparallel; that is, one strand is 5′ to 3′ and
the initiation of DNA replication at oriC, see the other strand is 3′ to 5′. In other words,
note 13. the 5′-phosphate end of one strand is paired
with the 3′-hydroxyl end of the other strand
3.1.4 Replicating the DNA (Fig. 3.7). Since the DNA copy strand is antipar-
DNA polymerases allel to the template strand, this means that the
The enzymes that synthesize DNA by using a new strands of DNA must be of opposite polar-
strand of DNA as a template are called DNA ity too; that is, one is 3′–5′ whereas the other is
polymerases. Two important characteristics 5′–3′. That is, at the replication fork one of the
of DNA polymerases are (1) they can only copy strands has its 3′ end pointed toward the
extend DNA chains (i.e., they cannot initiate fork, whereas the other copy strand has its 3′
new ones), and (2) they add mononucleotides end pointed away from the fork. This presents a
to the 3′ hydroxyl of deoxyribose and therefore problem because the DNA polymerase remains
elongate nucleic acid only at the 3′ end. E. coli at the fork, whereas the growing end (the 3′ end)
has five DNA polymerases: DNA polymerase of one of the strands keeps moving further and
I (Pol I) encoded by polA; DNA polymerase II further away from the fork.
Free ebooks ==>
86 the physiology and biochemistry of prokaryotes


binding site
A+T for DnaA


open complex

DnaB: DnaC: ATP

DnaC, ADP, Pi

prepriming complex

DnaB (helicase) SSB

Fig. 3.9. Creation of the prepriming complex. A multimer of about 20 DnaA proteins recognizes specific
sequences and binds to the duplex DNA at the origin adjacent to an A–T-rich region. The A–T region adja-
cent to the DnaA proteins then unwinds to form the open complex. This step is aided by the HU protein and
requires ATP. DnaB protein (helicase) then binds in a DnaC-dependent reaction and further unwinds the
duplex. Single-stranded DNA-binding proteins (SSBs) bind to the single strands behind the helicase, prevent-
ing the single strands from coming together again.

Okazaki fragments 3′ end then returns to the fork to initiate rep-

We have seen that the DNA polymerase man- lication of another short fragment. (See note
ages to remain at the replication fork and yet 15 for a description of the Okazaki experi-
synthesize a strand of DNA whose 3′ end keeps ments.) The strand copied in short fragments
moving further and further away from the rep- is called the lagging-strand template. The other
lication fork. How is this accomplished? As will strand, which is copied in one piece, is called the
be explained next, the answer lies in synthesiz- leading-strand template. As discussed later, in
ing the strand whose 3′ end faces away from the the subsection entitled Model explaining how
fork. The synthesis proceeds in short fragments the DNA polymerase III synthesizing the lag-
(about 1,000 nucleotides long), called Okazaki ging strand can stay with the replication fork,
fragments after two of the investigators who the DNA polymerases that synthesize the lead-
discovered them. As implied in Fig. 3.10, the ing and lagging strands are actually associated
Free ebooks ==>
chromosome replication and partitioning of chromosomes 87

Fig. 3.10 Synthesis of leading and lagging strands. Both template strands must be copied in the 3′-to-5′
direction so that the copy strands, which grow at the 3′ ends, are of opposite polarity to their respective
templates. Since the template strands are of opposite polarity, the copy strands must be synthesized in
opposite directions: one strand in the direction of replication fork movement and the other strand in the
opposite direction. However, it is known that the DNA polymerase, which is at the 3′ end of the growing
strand, remains at the replication fork. The strand whose 3′ end faces the replication fork is synthesized
by a polymerase that stays attached to the template and moves with the replication fork. This template
is called the leading-strand template, and the strand being synthesized is called the leading strand. The
strand whose 3′ end faces away from the replication fork is synthesized in short fragments called Okazaki
fragments, which are about 1,000 nucleotides long. Upon completion of an Okazaki fragment, the DNA
polymerase leaves the template and returns to the replication fork to synthesize another fragment. The
template that is copied in short fragments is called the lagging-strand template, and the strand being
synthesized is called the lagging strand. Each Okazaki fragment begins with the synthesis of an RNA
primer that is subsequently elongated by the DNA polymerase. The RNA primers are drawn as wavy
lines. Eventually, the RNA primers are removed and the fragments are elongated by DNA polymerase and
sealed. See text for details.

with each other as a dimer and do not leave the tracks. However, as we shall describe shortly,
replication fork. in Section 3.1.5, it now appears that the rep-
lication fork and the replicative DNA poly-
Synthesis of leading strand merase reside in a replication “factory” that
Recall that DNA polymerase cannot initi- does not move along the DNA. The replication
ate new polynucleotide strands but can only factory is called a replisome, and the unrepli-
elongate the 3′ end of an existing strand. This cated DNA threads through it; the terminus is
means that the initiation of replication of the the last part to enter. In connection with topics
leading strand at oriC (as well as the lagging covered later, we describe the experiments that
strand) must begin with the synthesis of a rela- support this model (see note 34, which relates
tively short RNA primer (5–60 nucleotides) primarily to Section 3.1.6).
by an RNA polymerase. An RNA polymerase
called primase (or DnaG) synthesizes the Synthesis of the lagging strand
RNA primer for the Okazaki fragments, and The opposite template strand is called the lag-
perhaps also the RNA primer for the leading ging template strand. It cannot be copied in the
strand. DNA polymerase III then extends the same way as the leading template strand is cop-
RNA primer at its 3′ end and continues the ied because it is of opposite polarity, and the 3′
elongation of the DNA (Fig. 3.10). The DNA end of the RNA primer (i.e., the growing end) is
polymerase that synthesizes the leading strand facing away from the replication fork. Instead,
stays with the replication fork as it adds nucle- the polymerase periodically disengages from the
otides; it dissociates rarely, if at all. At one template strand to return to a newly synthesized
time it was thought that the DNA polymerase primer at the replication fork, thus synthesizing
moved along the DNA with the replication fork short polynucleotide fragments, the Okazaki
to the terminus, much as a train moves along fragments (Fig. 3.10).
Free ebooks ==>
88 the physiology and biochemistry of prokaryotes

The DnaB and DnaG complex unwinds the of DNA in the lagging strand, it stops and leaves
DNA at the replication fork and synthesizes the 3′ end of the newly synthesized DNA. The
the RNA primer for the lagging strand result is a nick in the DNA strand between the
As replication proceeds, RNA primers for the DNA and RNA. The very short RNA primer
Okazaki fragments are synthesized by the spe- fragments (10–12 nucleotides long) in the
cial RNA polymerase called the primase or Okazaki fragments are removed by the 5′- to-3′-
DnaG (Fig. 3.11). The primase is associated exonuclease activity of DNA polymerase I, and
with the helicase (DnaB) to form a complex each ribonucleotide is simultaneously replaced
that stays with the replication fork as the DNA with a deoxyribonucleotide polymerized by
is replicated. The primase must synthesize the DNA polymerase I. When this happens, the nick
Okazaki primers in the direction away from the moves in the direction of DNA synthesis and the
replication fork. The DnaB helicase unwinds process is called nick translation. The forma-
the duplex. Then the DNA polymerase III elon- tion of the phosphodiester bond between the 3′
gates the RNA primer at its 3′ end to synthesize hydroxyl at the end of one strand of DNA and
the Okazaki fragment. the 5′ phosphate of the other strand has been is
catalyzed by DNA ligase. The overall scheme is
Model explaining how the DNA polymerase shown in Fig. 3.12. The ligase makes the cova-
III synthesizing the lagging strand lent bond between the 3′ phosphate of the newly
can stay with the replication fork synthesized DNA and the 3′ hydroxyl of the pre-
The two DNA polymerases III (Pol III) that viously synthesized DNA by transferring AMP
synthesize the leading and lagging strands are from either ATP or NAD+ (depending upon the
physically associated with each other as a dimer
that does not leave the fork. This ensures that
both strands are replicated simultaneously and
DNA polymerase III
at the same rate. However, there is a problem.
SSB DNA gyrase
The template strands are of opposite polarity,
and this implies that the two DNA polymerases 5'

III must move in opposite directions. Yet, they (1) 3'

stay together as a dimer at the replication fork.
primosome (helicase, primase)
It may be that the dimer stays with the replica-
tion fork and still synthesizes the lagging strand, RNA primer

whose growing 3′ end extends away from the

fork because the lagging-strand template loops Fig. 3.11 Replication fork. The model explains
around one of the polymerases at the replica- how DNA polymerase might synthesize the lagging
strand and stay with the replication fork. It is sug-
tion fork so that its 5′ end faces the replication
gested that the lagging-strand template loops around
fork rather than away from it. This suggested a dimeric polymerase (two DNA polymerases physi-
mechanism is illustrated in Fig. 3.11. If the pos- cally associated with each other) so that the poly-
tulated scheme is accurate, then the Okazaki merase bound to the lagging strand moves along it
fragments will be elongated at their 3′ ends at in the direction of the replication fork. Thus, one of
the replication fork and the dimeric DNA poly- the polymerases of the dimer extends an RNA primer
merase III can elongate both the leading and (wavy line) on the lagging strand, and the other poly-
lagging strands at the same time. According to merase elongates the leading strand. Another RNA
the model, after polymerizing approximately primer, labeled (1), is made by primase ahead of the
1,000 nucleotides, or one Okazaki fragment, polymerase at the replication fork. The polymerase
the polymerase synthesizing the lagging strand keeps moving on the template until it encounters the
5′ end of a previously made Okazaki fragment (wavy
would disengage from the template and a new
line). Upon encountering the Okazaki fragment, the
loop would form. polymerase that is synthesizing the lagging strand
disengages from the template and reinitiates at the
Attaching the Okazaki fragments to new RNA primer with a new loop. The arrows point
each other in the direction of strand elongation at a stationary
When DNA polymerase III reaches the RNA at replication fork. See Section 3.1.3 for a discussion of
the 5′ end of the previously synthesized segment the stationary replication fork.
Free ebooks ==>
chromosome replication and partitioning of chromosomes 89

Fig. 3.12 Attaching DNA fragments to each other in the lagging strand. When the growing Okazaki fragment
reaches the 5′ end of the previously synthesized Okazaki fragment, DNA polymerase III that is synthesizing
the lagging strand dissociates from the template DNA. DNA polymerase I then removes the RNA via 5′- to
3′-exonuclease activity from the 5′ end of the previously synthesized Okazaki fragment and replaces the RNA
(wavy line) with DNA (straight line). DNA ligase then seals the break in the DNA.

Fig. 3.13 DNA ligase reaction. The enzyme catalyzes the adenylylation of the 5′ phosphate. Depending upon
the ligase, the AMP can be derived either from ATP, in which case pyrophosphate is displaced, or NAD+, in
which case nicotinamide monophosphate (NMN) is displaced. E. coli DNA ligase uses NAD+ as the AMP
donor. Then the 3′ hydroxyl of the ribose attacks the AMP derivative and displaces the AMP. The result is a
phosphodiester bond.

organism) to the phosphate (Fig. 3.13). This and a clamp loader. The sliding β clamp has a
produces an AMP derivative with a high group large central pore through which the DNA tem-
transfer potential. The ligase then catalyzes the plate is threaded. Very importantly, the sliding
displacement of the AMP by the 3′ hydroxyl, clamp complex binds to the DNA polymerase
and the phosphodiester bond is formed. III core, which consists of the α, ε, and θ sub-
units, and keeps it tethered to the DNA, allow-
Sliding clamps ing the processing capacity of the polymerase to
DNA polymerase III, which is the replicative increase from approximately 12 nucleotides to
polymerase in E. coli, consists of three subassem- thousands of nucleotides before the DNA poly-
blies. The total complex is referred to as DNA merase dissociates from the template. Thus, the
polymerase III holoenzyme. The subassemblies sliding clamp is very important for the efficiency
are the polymerase itself, a sliding β clamp, of DNA replication. Since DNA is synthesized
Free ebooks ==>
90 the physiology and biochemistry of prokaryotes

continuously on the leading strand, the clamp inhibiting the replicative helicase (DnaB).17–19
is loaded only once. However, on the lagging Both Ter sites and the Ter-binding protein
strand, where DNA is synthesized in Okazaki have been well studied in B. subtilis. See note
fragments of approximately 1,000 nucleotides, 20 for similarities and differences with respect
the clamp is unloaded and loaded frequently. to E. coli.
For a discussion of the clamps and clamp load-
ers, see note 16.

3.1.5 Termination
Ter sites and Tus protein
In E. coli the two replication forks that begin
at oriC and polymerize bidirectionally stop
in a region of the chromosome opposite oriC,
called Ter, the termination region (Fig. 3.14).
After termination, the sister chromosomes
separate and partition (segregate) to opposite
halves of the cell. The termination region con-
tains specific sequences of nucleotides called
Ter sequences.
The Ter sites are quite unusual because they
allow replication to proceed in only one direc-
tion! (Note that in Fig. 3.14 the arrows refer to
the direction of replication, not the movement Fig. 3.14 Termination of DNA replication in E. coli.
The arrows refer to the direction of replication,
of the replication forks, which are station-
rather than to movement of the replication forks. A
ary within the replisome.) This property can
terminator region exists 180° opposite the origin. In
account for termination at the Ter sites. For the terminator region there are nucleotide sequences
example, assume that there are two Ter sites, that allow replication in only one direction. Thus, the
which are located next to each other, with per- terminator sequences are said to be polar. E. coli has
haps a short segment of DNA between them. six terminator (Ter) sites. These are, in the sequence
Suppose Ter site 1 allows replication to proceed in which they exist in the DNA, TerE, TerD, TerA,
clockwise but not counterclockwise, and Ter TerC, TerB, and TerF. On one side of the terminator
site 2 allows counterclockwise replication but region are TerE, TerD, and TerA, and TerC, TerB,
not clockwise replication. Thus, if clockwise and TerF occupy the opposite side. The figure shows
replication arrives at site 2, it will stall; and if two replication forks. Data discussed earlier in this
section support a model in which the replication
counterclockwise replication arrives at site 1, it
forks are stationary within a DNA synthesizing “fac-
will stall. Consequently, bidirectional replica-
tory” called the replisome. Unreplicated DNA enters
tion will meet at site 1 or site 2 or somewhere the replisome and passes through the stationary rep-
between them. lication forks. TerC, TerB, and TerF inhibit clock-
As shown in Fig. 3.14, the situation in wise replication, and TerA, TerD, and TerE inhibit
E. coli is a little bit more complicated because counterclockwise replication. Therefore, there are
it has six Ter sites, sites are divided into two three chances to stop replication in the termination
groups of three: one group (TerA, TerD, and region. For example, if counterclockwise replica-
TerE) prevents counterclockwise replication, tion moves past TerA, it will stop at TerD or TerE.
and the second group (TerC, TerB, and TerF), Consequently, replication from both directions will
prevents clockwise replication. As explained stop at one of the termination sites or a site between
them. A protein, called the Tus protein in E. coli,
in the legend to Fig. 3.14, the presence of
binds to Ter and prevents the helicase from proceed-
three Ter sites per replication fork direction
ing in one direction, thus accounting for the polar-
provides a backup in case the replication fork ity of inhibition of movement of the replication fork.
gets by one of these sites. A protein, which in Notice the site marked dif. It is here that recombina-
E. coli is called the Tus (terminator utilization tion takes place to separate chromosomes that have
substance) protein binds to the Ter site and undergone an unequal number of recombinations
imposes the one-way travel. Tus does this by during replication.
Free ebooks ==>
chromosome replication and partitioning of chromosomes 91

3.1.6 Chromosome separation and reflecting the fact that because the sister chro-
partitioning; some general principles mosomes do not partition to opposite cell poles,
First, some definitions: chromosome separation they can both be in the same daughter cell upon
refers to the detachment of the newly replicated cell division. Under these conditions, the nucle-
chromosomes from one another, and chromo- oids are single nucleoids and are not larger than
some partitioning refers to the segregation of normal.
sister chromosomes to opposite halves of the
cell prior to cell division. Chromosome separation
Chromosome separation can take place in Chromosome separation requires the separa-
one of two ways. If the linkage between the tion of the two sister chromosome dimers, said
daughter chromosomes is noncovalent, then to be catenated, by topoisomerase IV. The pro-
topoisomerases such as topoisomerase IV sepa- cess is called decatenation. If there has been an
rate the sister chromosomes, as we shall discuss unequal number of crossover events during
shortly. If an unequal number of recombina- DNA replication, then the sister chromosomes
tions have occurred between the sister chro- are covalently linked and must be separated by
mosomes, then the sister chromosomes are recombination by XerC and XerD.
covalently linked and a site-specific recombina- Recombination occurs at dif sites. A key pro-
tion must occur. (See later subsection entitled 2. tein here is the cell division protein FtsK, which
Site-specific recombination at dif.) is localized to the septum and recruits other
Chromosome partitioning is not a well-un- cell division proteins before cell division. FtsK
derstood process in prokaryotes because a spin- ensures that XerCD and topoisomerase IV act
dle apparatus similar to that which separates on the DNA at midcell during the final stages
sister chromosomes in eukaryotes is not pres- of cell division. A discussion of how this might
ent. This section discusses current ideas about occur can be found later (see subsection 3, which
how separation and partitioning of daughter considers the catalytic activity on this enzyme).
chromosomes occur in prokaryotes. Section
1. Topoisomerase IV
3.1.7 describes chromosome partitioning in B.
When replication is complete (at about 180°
subtilis, and Section 3.1.8 describes chromo-
from the start site), the two completed DNA
some partitioning in C. crescentus.
molecules are linked as two monomeric circles
in a chain; that is, they are catenated and must
Phenotypes of partition mutants be separated from each other (Fig. 3.15). The
The genes involved in chromosome partition- process is called decatenation. In E. coli and
ing have been discovered by analyzing mutants. Salmonella typhimurium, the decatenation
Mutants defective in chromosome partitioning of two monomeric interlinked chromosomes
can be defective in any one of several genes: some requires the action of a DNA gyrase–like enzyme
are involved in detaching the chromosomes called topoisomerase IV.21 Topoisomerase IV
from each other, some are involved in the par- mutants are temperature sensitive for growth;
tition of the detached chromosomes, and some that is, they grow at 30 °C but die at 42 °C.
are even involved in the replication of DNA, At the restrictive temperature, cell division
because only fully replicated DNA molecules is inhibited and the mutants form elongated
can detach from one another. Depending upon cells, often with unusually large nucleoids
whether the defect is in detachment or partition (revealed by DNA staining) in midcell or as
of chromosomes, the phenotype differs. If the several nucleoid masses unequally distributed
defect is in detaching the sister chromosomes in the elongated cells.22 The nucleoids are large
from each other, then the phenotype includes because the two daughter nucleoids are not able
elongated cells with an unusually large nucleoid to separate. The filamentous cells may divide
mass positioned near the cell center. Anucleate in regions where there is no DNA to produce
cells may also form, as discussed next. anucleate cells. Inhibition of cell division result-
On the other hand, if the sister chromosomes ing in elongated or filamentous cells is typical
can detach but cannot partition to opposite of mutants that have a block in DNA replica-
poles, the phenotype is an increased production tion or separation of daughter chromosomes.
of anucleate cells of the size of a newborn cell, This reflects the coupling between cell division
Free ebooks ==>
92 the physiology and biochemistry of prokaryotes

they chromosomes are covalently linked as a cir-

cular dimer to each other and decatenation can-
not separate them. Under these circumstances,
the DNA molecules must be separated by site-
specific recombination by using enzymes called
recombinases. E. coli has a locus called dif (dele-
tion-induced filamentation), which lies in the
replication terminus (Ter) region.25 The recom-
binational event that separates the two sister
chromosomes that have undergone an unequal
number of recombinations during replication
takes place in this Ter region (Fig. 3.16).
The site-specific recombination is catalyzed
by two recombinase proteins, XerC and XerD,
both of which are required.26 The recombinases
are activated by an E. coli cell division protein
called FtsK, which is localized at midcell with
other cell division proteins. (See Section 2.6.2
for a discussion of FtsK and the other cell divi-
sion proteins in E. coli, and ref. 27 for a discus-
sion of the multiple roles of FtsK in cell division
and chromosome segregation.) Homologues
of XerC and XerD have been found in a wide
variety of bacteria. Mutations in dif are viable,
and most of the cells are normal. This is to be
Fig. 3.15 Separating catenated daughter circles after expected, since dif is required only when there
DNA replication by using type IV topoisomerase. has been an unequal number of recombinations.
The reaction is catalyzed by type IV topoisomerase However, approximately 10% of the cells are
(similar to DNA gyrase). The enzyme catalyzes a filamentous with unusually large nucleoids,
double-stranded cut in one of the duplexes. Then the indicative of a failure of the newly replicated
unbroken duplex passes through the gap. The gap is chromosomes to separate. As expected, xerC
sealed, and the DNA molecules have become sepa- and xerD mutants show the same phenotype as
rated. Mutants defective in topoisomerase IV are dif– mutants.
temperature sensitive for growth and show abnor-
mal nucleoid segration. The phenotype is called Par–.
See ref. 21. 3. Topoisomerase IV catalyzes decatenation
primarily at dif in the presence of XerC,
XerD, and FtsK
The activity of topoisomerase IV can take place
and DNA metabolism as described in note 23.
at several chromosomal sites; but in the presence
(Topoisomerase IV mutants have been isolated
of FtsK, XerC, and XerD, decatenation takes
in C. crescentus, but the phenotype differs from
place preferentially at midcell at the dif site.28
that of E. coli and S. typhimurium mutants
However, unlike site-specific recombination,
because checkpoints exist in the Caulobacter
which requires XerC and XerD, decatenation at
cell cycle.24) Topoisomerase IV makes a double-
dif does not require the cell division protein FtsK,
stranded break in one of the DNA molecules,
but is simply favored at dif because of FtsK. Why
and the unbroken molecule is passed through
should topoisomerase IV preferentially cleave
the break (Fig. 3.15). This is followed by seal-
at dif? There have been two suggestions.28 One
ing the break (see section entitled Attaching the
suggestion is that the dif/XerC/XerD complex
Okazaki fragments to each other).
might increase the affinity of topoisomerase IV
2. Site-specific recombination at dif for the dif site. A second suggestion is that topoi-
If an unequal number of recombinations have somerase IV is part of a decatenation/resolution
occurred between sister circular chromosomes, complex that forms at dif.
Free ebooks ==>
chromosome replication and partitioning of chromosomes 93

Chromosome partitioning in eukaryotes

It is worthwhile to summarize how chromosome
partitioning takes place in eukaryotes before
discussing how the process occurs in prokary-
otes. In eukaryotes, chromosome replication is
completed to form two sister chromatids bound
along their length in the S phase of the cell cycle.
The sister chromatids remain together dur-
ing the G2 phase, which follows the S phase.
Following the G2 phase is the M phase (mito-
sis), and this is when the sister chromatids parti-
tion. The M phase consists of nuclear division
and cytokinesis. Partitioning occurs because
the sister chromatids are attached to microtu-
bule spindle fibers that pull them apart. This
occurs during the anaphase portion of mitosis.
Specifically, the sister chromatids are separated
(and are now called chromosomes) and pulled
to opposite cell poles by the shortening of the
microtubular spindle fibers in the mitotic spin-
dle. Each chromatid is attached to the microtu-
bule spindle fibers via a protein complex called
a kinetochore, which binds to a subset of micro-
tubules in the mitotic spindle. The kinetochore
is located at a highly condensed, constricted
region called a centromere.

Chromosome partitioning in prokaryotes

In prokaryotes, “chromosome partitioning”
refers to the segregation of sister chromosomes
toward opposite cell poles in preparation for
cell division. (Bacterial chromosomes are often
referred to as nucleoids.) There is no evidence
in bacteria for a device similar to the eukaryotic
mitotic spindle to partition sister chromosomes.
So how do bacteria partition sister chromo-
Fig. 3.16 Site-specific recombination at termina- somes faithfully? This is a very active area of
tion to separate chromosomes that have undergone research, and the literature provides detailed
unequal numbers of recombinations. Simplified information about the similarities and differ-
drawing of recombination at dif to resolve a DNA ences with respect to what has been learned
dimer into monomers. As discussed in Section 3.1.5, about B. subtilis, E. coli, and C. crescentus.29–33
the replication forks are stationary and remain close Certain generalizations can be made. First, it is
to each other in the replisome, a DNA-synthesizing
“factory.” The unreplicated DNA is fed through the
forks. The arrows depict the direction of polymer- recombination event at dif can proceed in both direc-
ization of DNA. (A) Unequal numbers of recom- tions. However, monomer formation is favored.
binations have taken place covalently, linking the Perhaps the segregation of the monomers ensures
two circles to form a dimer. (B) Recombination is that the newly replicated dif sites cannot recombine.
catalyzed by recombinases acting at the dif sites in For recombination at dif to resolve the dimers, dif
the termination (Ter) regions. The chromosomes are must be at its original site in the terminus region. If a
then segregated to opposite cell poles. A dif− mutant copy of dif is reinserted in a different chromosomal
produces filaments and anucleate cells as well as nor- site in a dif– strain, the cells still show the dif– phe-
mal cells. As the figure indicates, it is believed that the notype.
Free ebooks ==>
94 the physiology and biochemistry of prokaryotes

noted that partitioning (segregation) takes place replication in the previous cell cycle, the student
concurrent with replication of the DNA. is referred to ref. 36.
The DNA is replicated in a stationary “fac- There are important differences between
tory,” called the replisome, which in E. coli or the positioning of the replisome in E. coli and
B. subtilis is assembled at or near midcell when B. subtilis on the one hand, and C. crescentus
DNA replication is initiated and is thought to on the other hand. These differences, which
be anchored there. For a review of replisomes, are related to differences in the cell cycle, are
see ref. 34. When DNA synthesis is finished, the described later. (See Sections 3.1.7 and 3.1.8.)
replisome is disassembled. The unreplicated
DNA threads through the replisome, and the What moves the sister chromosomes apart,
newly duplicated portions move away from the and what directs their movement?
anchored replication forks within the replisome Clearly, prokaryotic cells do not have a spindle
and partition toward opposite poles of the cell. apparatus. What then is responsible for the
Since the origins (ori sites) are replicated first, movement of sister chromosomes to opposite
they leave the replisome first and very quickly poles of the cell prior to cell division? Recently,
are moved to opposite halves of the cell near the there has been a great deal of research and
cell poles. much speculation. Several proteins have been
As will be discussed later, the “extrusion– implicated. Some of the proteins are thought
capture” model proposes a mechanism(s) that to comprise the “motor” that moves the chro-
provides the force that “extrudes” the newly mosomes, some of the proteins may attach the
replicated origins from the replisome and “cap- oriC regions to the opposite cell pole regions,
tures” the origins at or near opposite cell poles. and some of the proteins may compact the
As will also be discussed, for this to occur suc- chromosomes and “guide” the chromosomes
cessfully, certain proteins must compact the toward the opposite poles. Which proteins are
chromosomes and guide the origin regions these? We present an overview of current ideas,
toward the cell poles. Eventually, the terminus is followed by a more detailed examination of the
drawn into the replisome, where it becomes rep- proteins thought to be involved in chromosome
licated; then the sister chromosomes separate, partitioning.
and segregation is completed as the duplicated
termini remain near the cell center. As a result, 1. Sister chromosomes might be pushed
newborn cells in slowly growing cultures of E. out by the replisome to opposite halves
coli, B. subtilis, and C. crescentus have a termi- of the cell
nus near the new cell pole and an origin near Because of the absence of a spindle apparatus
the old cell pole (Fig. 3.17). Finally, the termi- similar to that found in eukaryotes, the mecha-
nus in newborn cells is drawn toward the center nism for the partitioning of sister chromosomes
of the cell by DNA replication in the centrally toward opposite poles of the cell in prokaryotes
located replisome factory. There can be small is not well understood. Various models have
differences among bacteria in the cellular loca- been proposed. One model postulates the repli-
tion of chromosome regions. For example, in some as a DNA-synthesizing “factory” that has
E. coli the origin regions localize at quarter-cell motor proteins generating force that pushes
positions instead of at the cell poles, whereas in the newly replicated sister chromosomes apart
B. subtilis growing vegetatively or sporulating, toward opposite cell poles. According to this
they localize at the cell poles. model, referred to as the “extrusion–capture”
For a description of how the replisome as well model (Fig. 3.17), in E. coli and B. subtilis the
as specific regions of the chromosome such as replisome is anchored to the membrane at the
origins, termini, and regions between them, can cell center; that unreplicated DNA enters the
be microscopically visualized, see note 35, as replisome and threads through it, whereupon
well as the references cited earlier concerning the newly replicated sister chromosomes exit the
chromosome segregation in B. subtilis, E. coli, replisome and partition toward opposite poles
and C. crescentus. For more information about of the cell. In the slightly different situation of
the location of the origins and termini, specifi- C. crescentus, (discussed later), the replisome
cally for fast-growing cells that initiate DNA is initially at the stalked pole of the cell, where
Free ebooks ==>
chromosome replication and partitioning of chromosomes 95

A newborn cell
origin terminus

origin, terminus move to

midcell position

DNA replication initiated,

C duplicated origins move to opposite
sides of the cell

D terminus duplicated, sister

chromosomes separate
and segregate to
opposite sides of cell, cell
begins to divide

origin origin

:::: replisome
replication fork


Fig. 3.17 General model for duplication, position, and partitioning of bacterial chromosomes in slowly
growing cells. (A) Newborn cell showing chromosome with single origin (circle) and single terminus (square)
at opposite cell poles. (B) Origin and terminus move and locate to the midcell position, where the replisome
is assembled and DNA replication is initiated. (C) DNA replication is initiated, and copies of origin segregate
to opposite halves of cell. In E. coli and C. crescentus, the origins are near opposite cell poles for most of the
cell cycle and move to the midcell position in the newborn cell. In B. subtilis, origin copies segregate to cell
quarter-positions that will become midcell following cell division. (D) Cell division begins. (E) Stationary
replisome at midcell within which are two replication forks. DNA is pulled into the replisome, origin first and
terminus last. The sister chromosomes are pushed toward opposite cell halves. (F) Cell division completed.

DNA synthesis is initiated, and it moves toward mechanism has been proposed whereby energy
midcell during DNA replication. As mentioned released by the combined action of the DNA
earlier, the replisome is assembled when DNA polymerase and the DNA helicase, both of
replication is initiated and disassembled when which can be considered to be force-generating
DNA replication is finished. motors, “pulls” the unreplicated DNA into
the stationary replisome and “pushes” the
2. Candidate proteins that might be involved replicated sister chromosomes, origin regions
in providing the force to move the DNA first, out of the replisome into opposite halves
What are the postulated motor proteins that seg- of the cell. It has been suggested that because
regate the sister chromosomes? A cooperative the inhibition of transcription also inhibits
Free ebooks ==>
96 the physiology and biochemistry of prokaryotes

sister chromosome partitioning, the force-gen- A more detailed description of proteins that
erating motor may also include stationary RNA play a role in chromosome partitioning
polymerases. As mentioned, several different proteins appear
to have a role in chromosome partitioning.
3. What confers directionality to
This is clearly not a well-understood process.
chromosome movements during partitioning?
However, the following discussion and the ref-
In addition to the “motor” that moves sister
erences cited should spark the student’s curi-
chromosomes apart, one may consider what
osity to explore this very important area of
confers directionality to chromosome move-
ment. One may ask, in other words, why sis-
ter chromosomes move to opposite poles. 1. RacA
One region of the chromosome that is directed RacA (remodeling and anchoring of the
toward opposite cell poles is the oriC region, Chromosome A) is a DNA-binding protein in
located at the origin of replication. In addi- B. subtilis that functions in chromosome par-
tion to oriC, the chromosomes of B. subtilis titioning only during sporulation.39 It binds
have another region near the origin, the polar preferentially at the PLR region, near oriC, and
localization region (PLR), which directs the less specifically throughout the DNA. RacA is
sister chromosomes to cell poles. It may work required for two reasons: (1) for the extension
independently of oriC, or it may be responsible of the nucleoid into an axial filament during spo-
for the migration of the nearby oriC to the cell rulation and (2) for the anchoring of the chro-
poles when chromosomes are partitioned dur- mosome near its origin region to the cell pole
ing sporulation.37 during sporulation (Fig. 3.18A). RacA works in
The question now is: What might “drag” the conjunction with Soj in moving the Spo0J–oriC
localization regions to the cell poles? One pos- complex to the cell pole via interaction with
sibility is that proteins bind to the DNA at the DivIVA at the pole. (See Section 3.1.7.) During
localization regions and that these proteins vegetative growth of B. subtilis, the two chro-
tether the chromosome to receptor proteins mosomal origins become anchored at cell quar-
in the membrane near or at the cell poles. Two ter-positions rather than at the extreme poles,
proteins mentioned in this regard are the RacA and RacA is not involved.
protein in B. subtilis, which binds to the PLR as
well as to regions near oriC, and the Par proteins 2. The Par proteins
(partitioning proteins) in many bacteria, includ- In most bacteria, and in low copy number plas-
ing B. subtilis and C. crescentus (but not E. coli or mids, an early stage of the partitioning process
Haemophilus influenzae), which bind to oriC. involves the Par proteins, which bind to the
In addition to proteins that have been sug- chromosome near the origin of replication and
gested to tether the localization regions to the may contribute toward the directionality of
pole, other proteins function in chromosome movement of the chromosome origins to oppo-
partitioning. These include SMC (structural site cell poles. (For a review, see ref. 40, and for
maintenance of chromosomes) proteins in B. a discussion of the role of Par proteins in plas-
subtilis and C. crescentus, and MukB proteins mid partitioning, see note 41.)
in E. coli. The SMC and Muk proteins condense The Par proteins discussed here that are
the chromosome. It appears that condensation encoded by the bacterial chromosome are called
of the chromosome might be important in mov- ParA and ParB. ParB is a DNA-binding protein
ing it toward the cell pole. In addition, there are that binds to specific sequences, called parS
the SetB and MreB proteins. SetB is an integral sequences, near the origin of replication. The
membrane protein that physically interacts binding is cooperative with ParA, an ATPase that
with MreB, a cytoplasmic actinlike filament. It interacts with ParB and results in a large nucleo-
has been suggested that MreB forms dynamic protein complex that is important for the posi-
filament cables that might be a mitotic-like tioning of sister chromosome and plasmid copies
DNA segregation apparatus.38 Exactly how this in opposite halves of the cell. As reviewed in ref.
works is not known. However, mutations in 21, ParA is membrane associated and may help
mreB in E. coli result in severe effects on nucle- tether the ParB–parS complex to the membrane.
oid segregation. Proteins homologous to the chromosomally
Free ebooks ==>
chromosome replication and partitioning of chromosomes 97

i ii

iv iii

B ii

i iii


Fig. 3.18 Partitioning of chromosomes in Bacillus subtilis. (A) Sporulating cell. (i) Stage 0. The two chromo-
somes are a single axial filament attached by their origins to opposite poles of the cell. (ii) A septum forms,
trapping the region proximal to the origin in the forespore. (iii) DNA translocation moves the rest of the
chromosome into the forespore. (iv) After engulfment, the origins detach from the poles. (B) Vegetative cell.
(i) Prior to DNA replication there is one origin. (ii) The DNA replicates, and both origins move toward the
opposite cell halves. (iii) The chromosomes condense, moving away from each other toward the poles. (iv) A
septum forms in the middle, dividing the cell into two daughter cells. Source: Adapted from Webb, C. D., A.
Teleman, S. Gordon, A. Straight, A. Belmont, D. C.-H. Lin, A. D. Grossman, A. Wright, and R. Losick. 1997.
Bipolar localization of the replication origin regions of chromosomes in vegetative and sporulating cells of B.
subtilis. Cell 88:667–674.

encoded ParA and ParB proteins have been 3. Bacterial SMC and Muk proteins
detected in many bacterial species, including Regardless of the mechanism of partitioning,
B. subtilis, where they are called Soj and Spo0J, the chromosome partitioning cannot take place
respectively, as well as in Caulobacter crescentus unless the chromosomes are compacted and
and Pseudomonas putida. However, E. coli and maintained in a higher order structure. One
H. influenzae do not have chromosomal parABS speculation is that condensation of the DNA
genes. Other proteins are presumably present in contributes toward a “pulling force” that com-
these bacteria that serve a function similar to plements the “pushing force” generated in the
that of the Par proteins. replisome.27 Proteins important for compaction
Free ebooks ==>
98 the physiology and biochemistry of prokaryotes

and higher order structure are called structural MukB forms a complex with two other Muk
maintenance of chromosome (SMC) proteins, proteins, MukE and MukF, and the three act
which are analogues of similar proteins in together. For more information about the phe-
eukaryotes. Not every bacterium has an SMC notype of mutations in the muk and smc genes,
protein. Bacteria such as E. coli, which lack see note 45.
an SMC protein, usually have a protein called
MukB, which is similar in structure and function 4. SetB and MreB
to SMC and sometimes is referred to as a “dis- Another protein that appears to be involved in
tant relative.” MukB (see Fig. 3.19) has thus far chromosome partitioning is SetB, an integral
been detected in E. coli and H. influenzae.42–44 membrane protein in E. coli.46 The deletion of

Fig. 3.19 Model of MukB. Based upon amino acid sequence studies and electron microscopic data, the model
postulates that MukB is a homodimer with a rod-and-hinge structure. At the C-terminal end lies a pair of large
globular domains. The N-terminal end has a pair of smaller globular domains. Electron micrographs suggest
that MukB can bend at a hinge site in the middle of the rod section. MukB binds to DNA, ATP, and GTP. The
amino acid sequence of MukB indicates homology with dynamin, a microtubule-associated protein from rats.
It has been speculated that MukB attaches to the DNA and to some cellular structure and moves the DNA to
opposite poles. Source: Niki, H., R. Imamura, M. Kitaoka, K. Yamanaka, T. Ogura, and S. Hiraga. 1992.
E. coli MukB protein involved in chromosome partition forms a homodimer with a rod-and-hinge structure
having DNA binding and ATP/GTP binding activities. EMBO J. 11:5101–5109.
Free ebooks ==>
chromosome replication and partitioning of chromosomes 99

setB results in a delay in sister chromosome sep- fork has been visualized by means of replicative
aration from the cell center, and overproduc- DNA polymerase fused in frame to the green
tion causes stretching and breaking up of the fluorescent protein (GFP).
nucleoid. Interestingly, SetB physically interacts
with MreB, a member of the family of actinlike Some of the factors important for
proteins that form cytoplasmic helical filaments chromosome structure and partitioning
along the inner cytoplasmic membrane of cylin- in B. subtilis
drical cells and are believed to be important Some of the proteins previously discussed are
for maintaining cell shape.47 (One reason for important for chromosome orientation and
believing that MreB is important for cell shape partitioning during sporulation and vegetative
is that deletion of the gene mreB causes the rod- cell division.
shaped E. coli to become spherical.) The MreB
protein, which is discussed in Section 1.2.7, has 1. ParA (Soj) and ParB (Spo0J)
been found in several different bacteria, includ- The ParA homologue in B. subtilis is Soj, and
ing E. coli, B. subtilis, Thermotoga maritima, the ParB homologue is Spo0J. Spo0J binds to
and Caulobacter crescentus.48 at least eight parS sites located on either side of
oriC, and in the presence of Soj the Spo0J–parS
3.1.7 Chromosome partitioning in complexes are organized into a condensed,
Bacillus subtilis higher order oriC supercomplex, which persists
throughout the cell cycle (growth, sporulation,
Sister chromosome partitioning in
germination) and is one of the factors that is
Bacillus subtilis during sporulation
required for chromosome partitioning dur-
(Fig. 3.18A)
ing vegetative growth as well as sporulation.
It is proposed that at stage 0 (Spo0), prior to
Mutations in soj–spo0J result in cells that are
the initiation of sporulation, the two chromo-
deficient in prespore chromosome partitioning
somes form a single axial filament with their
(segregation), as well as a small fraction of anu-
origin regions attached to opposite poles of the
cleate cells during vegetative growth. The Soj–
cell (Fig. 3.18Ai). Septation occurs close to one
Spo0J system and RacA (described next) appear
pole of the predivisional cell, trapping approxi-
to have partially redundant roles in partition-
mately a third of the forespore chromosome in
ing the prespore chromosome in that they both
the forespore (Fig. 3.18Aii). The origin of the
function to move oriC to the prespore cell pole
chromosome must be very close to the pole of
via an interaction with DivIVA at the pole.
the cell to ensure that it is captured in the pre-
spore when the septum forms. DNA transloca- 2. RacA
tion across the forespore septum takes place, RacA, known to be important in B. subtilis,
resulting in the movement of the forespore chro- is synthesized for a period during the begin-
mosome being entirely into the forespore (Fig. ning of sporulation but not during vegetative
3.18Aiii). Translocation of the DNA requires growth.39,50 This protein binds nonspecifically
a transporter called SpoIIIE DNA translocase, throughout the chromosome and is required
which forms a pore in the septum. Engulfment for formation of the axial filament. Mutants of
occurs, and the origins are detached from the racA fail to form an extended axial filament and
poles (Fig. 3.18Aiv). instead form a “stubby” nucleoid. RacA also
binds preferentially near oriC and is important
Sister chromosome partitioning during for efficient trapping of the oriC region at the
vegetative growth of Bacillus subtilis cell pole in the prespore. Mutations in racA
Prior to replication, a single chromosome with result in approximately 50% of prespores with-
one origin exists (Fig. 3.18Bi). After replica- out DNA. RacA also appears to be important
tion, the origins (called oriC, as in E. coli) are for the formation of the septum at the cell pole
moved to opposite halves of the cell, where during sporulation. Formation of the polar sep-
the origin regions become attached at the cell tum is significantly delayed in racA mutants.
quarter-positions (Fig. 3.18Bii).49 The chromo- It has been suggested that the binding of RacA
somes condense (Fig. 3.18Biii), and a septum to DivIVA at the cell pole displaces the divi-
forms (Fig. 3.18Biv). The stationary replicative sion inhibitor, MinCD, thus allowing septum
Free ebooks ==>
100 the physiology and biochemistry of prokaryotes

formation at the pole.51 The suggestion is based (called Cori or ori) and can be seen to localize
upon an earlier publication showing that to both cell poles when chromosomal replica-
DivIVA sequesters MinCD at the poles so that tion is complete in the stalked cell.53 This can
cell division occurs at midcell during vegetative be demonstrated by using immunofluorescence
growth.52 (See Chapter 23, Box 23.1.) microscopy with anti-ParA or anti-ParB anti-
body. (See note 54 for an explanation of how
3. DivIVA
these experiments were done.)
DivIVA is an anchor protein at the cell poles.
It probably binds directly or indirectly to RacA
and to Spo0J, which themselves are bound to A moving replisome
the chromosome in the area of the oriC region. Rather than being initiated in a centrally located,
Thus, DivIVA, RacA, and Spo0J recruit the stationary replisome as in B. subtilis and E. coli,
oriC region to the cell poles during sporula- DNA synthesis in C. crescentus is initiated in
tion. When RacA is absent, anucleate prespore a replisome assembled at the stalked-cell pole.
compartments form, resulting in a 50% drop in Soon after replication has been initiated, one of
the frequency of sporulation. In racA mutant the origin copies rapidly moves to the opposite
cells that do sporulate, the DivIVA–Spo0J sys- cell pole. During DNA replication, the repli-
tem recruits the chromosome origins to the some slowly moves toward the midcell posi-
cell poles. Thus, the functions of RacA and the tion, where DNA replication, including the
Spo0J system are partially redundant. replication of the terminus, continues.55 It has
been suggested that perhaps the replisome is
3.1.8 Chromosome partitioning in moved passively from the pole to midcell by the
Caulobacter crescentus newly replicated DNA accumulating at the cell
Chromosome partitioning in C. crescentus is pole.55 When DNA replication is complete, the
similar in some ways to the same procedure in B. replisome disassembles. The result is that prior
subtilis and E. coli, and different in other ways. to cell division there is a copy of each origin near
In all three species, when DNA replication is the opposite cell poles.
complete, the origins have moved toward oppo-
site cell poles and the termini copies are segre- 3.1.9 Inhibitors of DNA replication
gated from each other closer to the cell center. A variety of antibiotics made by microorgan-
In C. crescentus, however, replication begins at isms, as well as chemically synthesized anti-
a cell pole (the stalked cell pole) rather than at microbial agents, inhibit DNA replication
midcell, and the replisome is not stationary but itself or inhibit the synthesis of substrates for
moves from the pole to the cell center during DNA replication. Two antibiotics produced by
DNA replication. Before this is aspect of parti- Streptomyces include novobiocin, which inhib-
tioning described further, a brief account of the its DNA gyrase, and mitomycin C, which cross-
cell cycle in C. crescentus will be presented. links the guanine bases in DNA, sometimes in the
same strand and sometimes in opposite strands.
The Caulobacter cell cycle Cross-linking guanine in opposite strands pre-
Normal cell division in C. crescentus produces vents strand separation. Another antibacterial
two cell types, a motile swarmer cell with pili agent is nalidixic acid, a synthetic quinolone
and a polar flagellum, and a stalked cell. (See compound that inhibits DNA gyrase. Its more
Section 23.2 for a detailed description.) The potent fluoroquinolone derivatives (norfloxa-
stalked cell synthesizes DNA and undergoes cin, ciprofloxacin) are used to treat urinary tract
repeated divisions to give rise to swarmer cells, infections—for example, those caused by uro-
which do not synthesize DNA and do not divide. pathogenic E. coli.
After a while the swarmer cell sheds the flagel- There are several other chemically produced
lum, grows a stalk, begins synthesizing DNA, drugs that are widely used to inhibit DNA
and becomes a dividing cell. synthesis. These include acridine dyes (e.g.,
ethidium, proflavin, chloroquine), which insert
ParA and ParB between the bases in the same strand of DNA
As with B. subtilis, ParA and ParB form a com- and distort the double helix. At low concentra-
plex attached to the Caulobacter origin region tions the dyes inhibit plasmid DNA replication
Free ebooks ==>
chromosome replication and partitioning of chromosomes 101

but not chromosomal replication, although a methyl (–CH3) and thus convert dUMP to
they can cause frameshift mutations. (See note dTMP. As a consequence of losing the hydride
56 for an explanation.) If the concentration is ion, the tetrahydrofolate is oxidized to dihydro-
sufficiently high, chromosomal replication is folate. The THF is regenerated by reducing the
inhibited. Chemicals that inhibit the synthesis dihydrofolate with NADPH in a reaction cata-
of precursors to DNA include the monophos- lyzed by dihydrofolate reductase. The dihydro-
phate of 5-fluorodeoxyuridine, aminopterin, folate analogues aminopterin and methotrexate
and methotrexate. inhibit dihydrofolate reductase.
5-Fluorodeoxyuridine-5′-phosphate [FdUMP,
made by cells from fluorouracil (FU)] inhibits 3.1.10 Repairing errors during
thymidylate synthase, which is the enzyme that replication
converts dUMP to dTMP. (See Fig. 3.20.) In this All cells have repair mechanisms that fix DNA
reaction a methylene group (–CH2) and hydride that has been damaged or has suffered errors
(H:) are transferred from methylenetetra- during replication. Repair systems in bacteria
hydrofolate (methylene–THF) to dUMP to form that operate during DNA replication and cor-
rect erroneous insertions of nucleotides are
described next. Such errors cause relatively
minor topological changes in the DNA.

Editing repair
Occasionally the wrong base is inserted in the
copy strand (e.g., a T rather than a C opposite
a G). This can occur if a base tautomerizes to a
form that allows hydrogen bonding to the wrong
partner. See Fig. 3.21 for an explanation. If the
wrong nucleotide is added to the growing chain
(e.g., a T opposite a G), the T must be removed;
otherwise, when the strand containing the G–T
pair is replicated, one of its progeny duplexes
will have an A–T pair in place of the G–C pair:
that is, a mutation will result. When the wrong
base is inserted in the growing chain so that a
mismatched pair results (e.g., G–T), DNA repli-
cation stops: that is, the DNA polymerase does
not continue to the next position. Presumably
this occurs because of the resultant topological
change in the DNA (distortion in the double-
stranded helix). The incorrect nucleotide is
Fig. 3.20 Inhibition of dTMP synthesis by anti- then removed via a 3′-exonuclease. DNA poly-
cancer drugs. 5-Fluorouracil is converted to merases have 3′-exonuclease activity that func-
5-fluorodeoxyuridine-5′-phosphate (fluorodeoxyu- tions in proofreading and removes mismatched
ridylate, or FdUMP), which inhibits thymidylate nucleotides when they are added. Mutations
synthase (1), the enzyme that converts dUMP to in the dnaQ gene (mutD), which codes for a
dTMP. A methylene group (–CH2) and a hydride 3′-exonuclease subunit in the DNA polymerase
(H:) are transferred from methylenetetrahydrofolate III holoenzyme, result in greatly increased rates
(methylene-THF) to dUMP to form a methyl (–CH3), of spontaneous mutations.
which converts dUMP to dTMP. As a consequence
of losing the hydride, the THF is oxidized to dihy-
drofolate. The dihydrofolate is reduced back to THF
Methyl-directed mismatch repair
by NADPH in a reaction catalyzed by dihydrofolate The proofreading system is not perfect, and a
reductase (2). The dihydrofolate reductase is inhib- certain number of wrong bases do get inserted
ited by dihydrofolate analogues such as methotrex- into the newly replicated DNA. These can be
ate and aminopterin. removed, and the fidelity of DNA replication
Free ebooks ==>
102 the physiology and biochemistry of prokaryotes

region of DNA containing the GATC sequence.

Thus, for a few seconds or minutes after replica-
tion, the template strand is fully methylated but
the copy strand is undermethylated. During this
brief period, the copy strand can be repaired.
What happens is that the copy strand with the
incorrect base is cut at the 5′ side of the G in the
unmethylated GATC and the newly synthesized
DNA is removed by an exonuclease to a point
just beyond the mismatch (Fig. 3.22). The gap is
then filled with DNA polymerase III and sealed
with DNA ligase.
Several proteins are involved in mismatch
repair. In E. coli, these include the products
of mutH, mutL, and mutS. MutH, MutL, and
MutS are thought to form a complex with the
Fig. 3.21 Tautomerization can lead to incorrect base DNA. In this complex, MutH is the endonu-
pairing. Two isomers in equilibrium that differ in the clease. According to the model, MutS binds
arrangements of their atoms are called tautomers. to the single base pair mismatch. Then MutL
Commonly, tautomers differ in the placement of binds to MutS. MutH binds to a nearby GATC
hydrogen. For example, one isomer might exist in sequence, and its endonuclease activity is stimu-
the enol form (–OH attached to a carbon–carbon
lated by the MutS/MutL complex. (Another mut
double bond), whereas the other isomer might exist
in the keto form (contains a C=O group). Keto–enol
gene, mutD, encodes the DNA polymerase III
tautomerizations greatly favor the keto form. When 3′ exonuclease, which is important for editing,
thymine tautomerizes from the keto to the enol form, as described earlier in the subsection entitled
a proton dissociates from the nitrogen in the ring and Editing repair.) DNA helicase II (the product of
moves to the oxygen in the keto group to form the the mutU gene) is also required, and it is thought
hydroxyl. When this happens, two electrons shift in that its role is to unwind the cut strand so that it
from the nitrogen to form the C=N. (A) Adenine (A) can be degraded by exonuclease.
correctly base pairs with the keto form of thymine More than one exonuclease can be used. If
(T). (B) Thymine has tautomerized to the enol form the endonucleolytic cut by MutH is made 3′
and base pairs with guanine (G) to form a mismatch. to the mismatch, then exonuclease I (ExoI) or
ExoX, which is a 3′- to 5′-exonuclease is used.
If the cut is made 5′ to the mismatch, then either
improved 102- to 103-fold (Fig. 3.22), by what exonuclease VII (ExoVII) or RecJ, which are
is called the methyl-directed mismatch repair 5′- to 3′-exonucleases, are used. DNA poly-
(MMR) system (reviewed in refs. 57 and 58). merase III fills in the gap, and DNA ligase seals
The nucleotide that is removed is the incorrect the gap.
nucleotide in the copy strand rather than the Mismatch repair can also be used to repair
template strand; thus the template strand is not DNA damaged by the incorporation of base
changed, and a mutation does not occur. analogues or by certain types of alkylating agent
The repair system can distinguish the copy as long as the distortion of the double helix
strand from the template strand because the is not severe. Otherwise, repair mechanisms
template strand is marked by methylation. described in Section 19.2.1 are used. Null muta-
E. coli has an enzyme called deoxyadenosine tions in any of the mut genes involved in mis-
methylase (Dam methylase) that methylates all match repair result in a 102- to 103-fold increase
adenines at the N6 position within 5′-GATC-3′ in mutation frequency in E. coli. Homologues
sequences. (The sequence is a palindrome and of the mut genes can be found in eukaryotes
therefore is present in both strands but in oppo- such as yeast, mice, and humans, where loss of
site orientation, i.e., 3′-CTAG-5′.) However, function also results in increased mutation rates
the enzyme does not begin to methylate the and, in the case of humans, is correlated with
DNA until a short period after replication of the cancer in tumor cell lines from certain tissues
Free ebooks ==>
chromosome replication and partitioning of chromosomes 103

Fig. 3.22 Mismatch repair can occur when the wrong nucleotide has been inserted. For example, a T instead
of a C might be inserted opposite a G. The template strand is methylated at an adenine in a CTAGsequence.
The newly synthesized strand is not yet methylated, and this aids the repair enzymes in distin-guishing between
the newly synthesized strand and the template strand. (There is a slight delay in the methy-lation of newly syn-
thesized DNA. However, the A in GATC eventually becomes methylated.) (A) Anendonucleolytic cut is made
either on the 5′side or the 3′side of the mismatch in the GATC sequence in thenonmethylated strand. (B) An
exonuclease then removes the newly synthesized DNA past the point of themismatch. This requires helicase
II (MutU). (Helicase II is not identical to DnaB, which unwinds the strandsduring DNA replication.) If the
mismatch is on the 5′side of the cut, then exonuclease I degrades the DNA 3′to 5′through the mismatch. If the
mismatch is on the 3′side of the cut, then exonuclease VII or RecJ proteindegrades the DNA 5′to 3′through
the mismatch. (C) DNA polymerase III then fills in the missing DNA, andthe gap is sealed with DNA ligase.
Single-stranded DNA-binding protein is also required. The proteinsinvolved in recognizing the mismatch and
making the cuts are MutH, MutS, and MutL. The MutS proteinrecognizes the mismatch. The MutH protein
is the endonuclease that makes the cut. It recognizes the GATCsequence and cleaves the unmethylated DNA
on the 5′side of the G in the GATC. MutS and MutH form acomplex in which they are linked by MutL.

(e.g., colorectal carcinomas). (For reviews, see in a positive supercoil in the unreplicated por-
refs. 59–61.) tion, creating a primer that can be extended
by DNA polymerase III, and keeping the poly-
merase that synthesizes the lagging strand at
3.2 Summary the replication fork. These and other factors
To replicate DNA, several problems must necessitate using about 30 different proteins to
be attended to. These include unwinding the replicate the DNA. The replication fork is cre-
double helix without causing it to overwind ated at a specific sequence in the DNA (called
Free ebooks ==>
104 the physiology and biochemistry of prokaryotes

the origin) when DNaA, HU, and ATP begin Muk, Par, SetB, and MreB proteins. The MreB
to unwind the helix. Further unwinding of the proteins are related to actin.
DNA during replication requires replication Errors made during replication are repaired
fork helicase (DnaB). Single-stranded binding in two ways. Editing repair involves the imme-
protein binds to the single strands to prevent diate removal of the incorrect nucleotide by the
them from coming together again, thus ensur- 3′-exonuclease activity of DNA polymerase III.
ing that they can serve as templates for new The second method of repair, called mismatch
DNA. Meanwhile, DNA gyrase binds to the repair, can occur if the polymerase fails to
DNA downstream of the replication fork and remove the incorrect nucleotide and proceeds
continuously converts the positive supercoils beyond the error site. During mismatch repair
into negative supercoils so that the DNA does an exonuclease removes the segment of copy
not become overwound. DNA that includes the mismatched nucleotide
Replication of the DNA is initiated when the pair, and the gap is filled with DNA polymerase
enzyme primase synthesizes a short RNA oligo- III and sealed with DNA ligase. The copy strand
nucleotide at the replication fork on both the is recognized because it is undermethylated for
single strands. One copy strand, called the lead- a short period after synthesis.
ing strand, is synthesized continuously by elon-
gating the 3′ end of the RNA primer, the end
Study Questions
that faces the replication fork. The other copy
strand, called the lagging strand, is synthesized
1. Describe the roles of the following proteins
in short fragments of about 1,000 nucleotides
in DNA replication: helicase, DNA gyrase,
called Okazaki fragments, growing in the direc-
DNA ligase, DNA polymerase III, DNA
tion opposite to the movement of the replica-
polymerase I, primase, DnaA, SSB.
tion fork. Each fragment begins with a short
oligonucleotide RNA primer. The 3′ end of the 2. Compare the biochemical reactions cata-
lagging strand faces away from the replication lyzed by DNA polymerase and DNA ligase.
fork. The same DNA polymerase would be How do they differ?
able to synthesize both copy strands if the poly- 3. Describe the Meselson–Stahl experiment,
merase were dimeric and the template for the and explain how it proves that DNA is rep-
lagging strand looped around the polymerase licated semiconservatively.
so that its 5′ end faced the replication fork. The
RNA primers at the 5′ end of the copy strands 4. DNA exists in a negative supercoil. How
are removed by DNA polymerase I, which also might this be advantageous to DNA repli-
fills the gap. The break is sealed by DNA ligase. cation and RNA synthesis? During DNA
Replication is usually bidirectional. replication the unreplicated portion of the
Termination occurs at ter sites and requires DNA winds tighter. Why is this the case?
Tus protein. The enzyme topoisomerase IV 5. What are Okazaki fragments, and why are
separates the newly synthesized duplexes. In they necessary?
the event that an unequal number of recombi-
nations have taken place between sister chro- 6. Outline an experimental approach to
mosomes during replication, the daughter determine whether DNA synthesis takes
chromosomes are covalently linked and a site- place in a centrally located “factory,” and
specific recombination must take place at the dif how the origins and termini move during
site, which is in the Ter region. partitioning.
The process of partitioning nucleoids to 7. Why is the 3′-exonuclease activity of DNA
opposite poles of the cells is not understood. polymerase III important?
One model posits a stationary, centrally located
replicon that provides the force to partition sister
chromosomes to opposite poles. Additionally, REFERENCES AND NOTES
several different proteins outside the replisome
1. Most bacteria have circular chromosomes.
appear to be involved in chromosome position- Exceptions include Borrelia burgdorferi, which
ing and partitioning. These include the SMC, causes Lyme disease, Rhodococcus fasciens, and
Free ebooks ==>
chromosome replication and partitioning of chromosomes 105

Streptomyces lividans. See: Hinnebursch, J., and the number of times two strands are crossed per cata-
K. Tilly. 1993. Linear plasmids and chromosomes in lytic event, they alter the linking number in multiples
bacteria. Mol. Microbiol. 10:917–922. of 2. An example of a type II topoisomerase is DNA
gyrase from E. coli. It produces negative supercoils,
2. Supercoiling of DNA is an important aspect of unwinding the double helix in the following way: (1)
its structure in part because it makes the DNA more It binds to the DNA, wrapping approximately 120
compact. Consider DNA as consisting of two strands base pairs of duplex around itself. (2) It cleaves both
coiled around a central axis. Supercoiling refers to strands with a four-base-pair stagger. When it does
the twisting or coiling of the central axis. This is anal- this, the enzyme covalently bonds to the 5′ ends of
ogous to a coiled telephone cord that twists on itself each strand. (3) It moves the uncut complementary
to form additional coils. Suppose the DNA exists as strands through the break. The region that is moved
a closed double-helical circle. If one strand is cut and is either within the region that is wrapped around the
unwound while the other strand is prevented from enzyme or close by. (4) It reseals the breaks. Passing
turning, and then the cut is sealed, the circular DNA the unbroken portion through the broken portion to
will have fewer helical turns. Moreover, the DNA produce negative supercoils unwinds the DNA. ATP
will then be under structural strain and, as a conse- is required for DNA gyrase activity.
quence, will supercoil to relieve the strain. If the DNA
is underwound (i.e., fewer helical turns), it will twist To see how negative supercoiling can unwind a
on itself into negative (left-handed) supercoils to helix, take two pieces of rubber tubing and wrap them
relieve the stress. If the DNA is overwound (i.e., more around each other in a right-hand helix. This would
helical turns are introduced before the cut is sealed), be clockwise, sighting down the helix, away from you.
it will twist upon itself to produce positive (right- Clamp both ends. Now twist the linear helix so that the
handed) supercoils to relieve the strain. Most cellular central axis twists counterclockwise. This produces
DNA is underwound and negatively supercoiled. It negative supercoils. The helix will unwind. Twisting
is believed that the strands of the underwound DNA the helix so that the central axis turns clockwise (posi-
are more readily separated for replication and tran- tive supercoils) makes the helix wind tighter.
scription. The production of negative supercoils is Topoisomerase I and DNA gyrase adjust the super-
accomplished by a topoisomerase II enzyme called coiling of DNA. Gyrase adds negative supercoils and
DNA gyrase and does not occur by unwinding the topoisomerase I removes negative supercoils. For a
cut strands before rejoining them. (See text for how review of topoisomerases, see: Luttinger, A. 1995.
DNA gyrase introduces negative supercoils.) The twisted “life” of DNA in the cell: bacterial topoi-
3. Topoisomerases change the linking number somerases. Mol. Microbiol. 15:601–608.
of a covalently closed duplex of DNA. The link- 4. Marians, K. J. 1992. Prokaryotic DNA replica-
ing number is the number of times that the chains tion. Annu. Rev. Biochem. 61:673–719.
in the duplex of a covalently closed circle cross one
another; that is, it is the number of helical turns. As 5. Baker, T. A., and S. H. Wickner. 1992. Genetics
noted in the text, DNA is said to be overwound when and enzymology of DNA replication in Escherichia
it has more than 10.5 base pairs per helical turn and coli. Annu. Rev. Genet. 26:447–477.
underwound when it has fewer than 10.5 base pairs 6. Messer, W., and C. Weigel. 1996. Initiation
per helical turn. The original model proposed that of chromosome replication, pp. 1579–1601. In:
the topoisomerase cut one strand of the duplex and, Escherichia coli and Salmonella: Cellular and
while holding both ends of the nicked strands, rotated Molecular Biology, Vol. 1. F. C. Neidhardt et al.
a free end around the intact strand, thus changing the (Eds.). ASM Press, Washington, DC.
number of helical turns. This was followed by sealing
the break. A more recent model suggests that rota- 7. Marians, K. J. 1996. Replication fork propa-
tion is not necessary and that the number of helical gation, pp. 749–763. In: Escherichia coli and
turns can be changed simply by passing the unbro- Salmonella: Cellular and Molecular Biology, Vol. 1.
ken strand through the break in the complementary F. C. Neidhardt et al. (Eds.). ASM Press, Washington,
strand. An example is type I topoisomerase. The DC.
type I topoisomerases catalyze a break in one of the 8. Ryan, V. T., J. E. Grimwade, J. E. Camara, E.
strands, which allows the passage through the break Crooke, and A. C. Leonard. 2004. Escherichia coli
of the unbroken partner strand. The break is then pre-replication complex assembly is regulated by
sealed. This results in changing the linking number dynamic interplay among Fis, IHF, and DnaA. Mol.
(number of helical turns) by multiples of one. Type Microbiol. 51:1347–1359.
II topoisomerases do it somewhat differently and
9. Leonard, A. C., and J. E. Grimwade. 2005.
change the linking number by multiples of 2. Type
Building a bacterial orisome: emergence of new regu-
II topoisomerases catalyze breakage of both strands
lator features for replication origin unwinding. Mol.
of the duplex and allow the passage of two unbro-
Microbiol. 55:1365–2958.
ken complementary strands from a nearby region of
the molecule through the breaks before sealing. (The 10. Lopez de Saro, F. J. 2009. Regulation of inter-
breaks are staggered breaks, i.e., not directly opposite actions with sliding clamps during DNA replication
each other.) Because type II topoisomerases change and repair. Curr. Genom. 10:206–215.
Free ebooks ==>
106 the physiology and biochemistry of prokaryotes

11. Johnson, A., and M. O’Donnell. 2005. Cellular boxes are nucleotide repeats containing 9 base pairs;
DNA replicases: components and dynamics at accordingly, they are called 9-mers. The three non-R
the replication fork. Annu. Rev. Biochem. 2005. boxes are called I boxes (because, as we shall see, the
74:283–315. binding of DnaA to these sites is stimulated by integra-
tion host factors), and these are 11-mers. These sites
12. Bidirectional replication can be demonstrated in
differ in their affinity for DnaA. In vitro studies indi-
two ways. One way is to insert into the chromosome
cate that binding initiates at the tighter binding sites
prophage I at the att site and prophage Mu within vari-
and progresses to the weaker binding sites in the fol-
ous mapped genes around the chromosome. The Mu
lowing order: R4,R1 > R2 > R5(M) > I3 > I2 > I1 > R3.
can be located because when it inserts into a gene, it
Studies indicate that DnaA is bound to R4, R1, and
causes a mutation in that gene, and the location of the
R2 throughout most of the cell cycle, and binds to the
gene is known. If such mutants are also auxotrophic
weaker sites just before DNA replication is initiated.
for certain amino acids, the experimenter can stop the
The binding of DnaA at the weaker sites (R5, R3, and
initiation of new rounds of replication by removing the
the integration sites) results in the unwinding of DNA
required amino acid. By adding the amino acid back
to form an open complex within three contiguous
again, the experimenter can initiate new rounds of rep-
13-mers at the left boundary of oriC; all these nucle-
lication at the replication origin. If new rounds of repli-
otide repeats are rich in A–T. Is there is a mechanism
cation are begun with bromouracil in the medium, then
that ensures that DnaA–ATP binds to the weaker sites
bromouracil is incorporated into the DNA in place
at the same time in all the oriC copies? The answer is
of thymine. The presence of the bromouracil makes
yes. Two histonelike DNA-bending proteins, Fis (fac-
the newly synthesized DNA denser than the parental
tor for inversion stimulation) and IHF (integration
strand. The newly synthesized strands can then be sep-
host factor) are involved. These have binding sites in
arated at various times after the initiation of replica-
the right and left half of oriC, respectively.
tion by density gradient centrifugation and hybridized
with both prophages, I and Mu, whereupon the ratio Fis and IHF bind at oriC and bend the DNA.
of Mu DNA to I DNA can be measured. Knowing the Presumably this is important for the mechanism of
map position of the various Mu inserts, it is possible to their action at oriC. Mutants that do not make either
demonstrate that replication proceeds bidirectionally of these proteins are viable, but they cannot initiate
from the origin of replication. Bidirectional replication replication synchronously at multiple origins in fast-
forks have also been observed by using radioautogra- growing E. coli. In vitro studies have shed light on the
phy and electron microscopy after a pulse of tritiated mechanism of action of IHF and Fis. It was reported
thymidine. Both replication forks can be seen to be that Fis binds to oriC in vitro and forms a complex
labeled when replication is examined in several bacte- that blocks the binding of DnaA to low-affinity sites
ria. If DNA replication were unidirectional, only one in oriC, as well as the binding of IHF to oriC, possi-
replication fork would be labeled. bly inhibiting oriC unwinding. It has been suggested
13. This note summarizes what is known about the that increased levels of active DnaA displace Fis and
regulation of the initiation of DNA replication in allow IHF to bind to DNA, resulting in the redistribu-
E. coli. tion of DnaA from high-affinity sites to low-affinity
sites. Whereas Fis binds to oriC during most of the
Several DNA-binding proteins are involved. One cell cycle, IHF binds to oriC only at the beginning of
critical protein is DnaA, which binds to the origin DNA replication. The regulation of the binding of
first, begins the process of strand separation, and IHF to DNA is not understood. As mentioned, the
then recruits helicase. The active form of DnaA that binding of DnaA to the low-affinity sites is necessary
initiates unwinding is DnaA–ATP. The binding of for unwinding and, as a consequence, IHF promotes
ATP to DnaA promotes an allosteric modification unwinding at oriC. Thus, Fis and IHF have opposite
in DnaA, allowing it to function. Nonhydrolyzable effects and help to regulate the initiation and syn-
analogues of ATP also work. After the initiation of chrony of DNA replication.
DNA replication, DnaA hydrolyzes the bound ATP
to bound ADP. DnaA–ADP is inactive in initiating It has been suggested that because Fis is growth
DNA replication. When the activity of DnaA in the rate regulated (lower levels at slow growth rates), the
cell is sufficiently high, DNA replication is initiated. protein functions only in rapidly growing E. coli with
One line of evidence for the role of DnaA in initia- more than one copy of oriC. The levels of IHF are not
tion is derived from studying a temperature-sensitive growth rate related, and it has been suggested that
dnaA mutant harboring a plasmid containing wild- IHF functions under all growth conditions to pro-
type dnaA under the control of the inducible lac mote DnaA binding to the weaker sites in oriC. For
promoter. When the cells are grown at 42 °C, replica- more discussion, see ref. 8.
tion is dependent upon the expression of the plasmid Interestingly, the levels of DnaA do not change very
dnaA gene. The timing of initiation varies with the much during the cell cycle, and part of the regulation
levels of expression of the gene. When DnaA levels are of the initiation of DNA replication is due to the reg-
increased, initiation occurs earlier in the cell cycle. ulation of the activity of existing DnaA. One model
What does DnaA do? DnaA binds to eight sites proposes that the phospholipids in the membrane
in oriC, five of which are called R boxes. The five R keep sequestered DnaA in an inactive form until it is
Free ebooks ==>
chromosome replication and partitioning of chromosomes 107

bound to oriC. In support of this model is the obser- 14. DNA polymerase III has 10 different protein
vation that a fraction of the cellular DnaA can be subunits coded for by separate genes. One of these,
isolated with the membrane fractions. Furthermore, the α subunit, is encoded by the polC (dnaE) gene
DnaA forms complexes with phospholipids and the and is responsible for polymerization. The subunit
phospholipid can inhibit DnaA activity. responsible for the 3′-exonuclease activity is the ε
subunit encoded by the dnaQ (mutD) gene. These
Another factor that blocks the reinitiation of new
two subunits, along with the ϕ subunit, comprise
rounds of DNA replication is sequestration of oriC,
the core of the polymerase. The τ subunit dimerizes
which refers to any mechanism that prevents reinitia-
the core. The other subunits form a clamp that keeps
tion at oriC. To discuss sequestration, it must be pointed
the polymerase on the template. The subunits that
out that the origin region is rich in GATC/CTAG
are not part of the core are sometimes called acces-
sequences and that these sequences are methylated at
sory proteins. The core plus the accessory proteins is
the adenine residues by deoxyadenosine (Dam) methy-
referred to as the holoenzyme.
lase shortly after a strand is copied. For reinitiation to
occur, both strands at oriC must be methylated. 15. The realization that DNA copied from one
template is made in short fragments came from
However, for a period of time after initiation (about
experiments in which bacteriophage T4 that was
one-third of the cell cycle), the origin remains hemi-
infecting E. coli was labeled with tritiated thymidine
methylated; that is, the template strand is methylated
for various lengths of time (2–60 s). The DNA was
but the copy strand is not. This is because newly rep-
denatured with base, and the intermediates were sep-
licated origins are sequestered from Dam methylase.
arated from the template DNA by centrifugation in
The sequestration is due to a protein called SeqA,
a sucrose gradient. It was found that when the DNA
which binds to oriC and prevents methylation of the
was labeled for a short period of time, substantial
new strand. One of the lines of evidence for this is
radioactivity was recovered in small pieces of DNA
that purified SeqA binds to oriC and prevents DNA
(about 1,500 nucleotides long). As the labeling time
replication in vitro. There is also some evidence that
was increased, the radioactivity in the smaller pieces
in E. coli, sequestration of oriC may also involve
remained constant but accumulated in high molecu-
binding of the SeqA/DNA complex to outer mem-
lar weight DNA, as would be expected if the smaller
brane components. Cell fractionation studies have
pieces were precursors to the high molecular weight
revealed that certain subcellular fractions containing
DNA. Okazaki fragments are made during the syn-
outer membrane components are enriched for hemi-
thesis of all DNA molecules in viruses, prokaryotes,
methylated oriC sequences and that hemimethylated
and eukaryotes.
oriC regions bind more readily to membrane frac-
tions than do fully methylated oriC regions. 16. The E. coli sliding β clamp, which is part of the
DNA polymerase III holoenzyme, is actually a ring of
Eventually however, newly replicated DNA
protein with a hollow core through which the DNA
becomes fully methylated by Dam methylase. This
template is threaded. The β clamp tethers the DNA
takes place well before the next round of initiation,
polymerase to the template, thus increasing the pro-
and therefore other controls besides sequestration
cessivity of replication. It does this by binding to the
by SeqA prevent reinitiation during the cell cycle.
catalytic subunit (α subunit) of the core polymerase,
The nature of these other controls in E. coli is a mat-
tethering it to the template DNA. The clamp itself
ter of speculation. However, there is evidence for a
is assembled by a clamp loader, also a part of the
repressor (CtrA) of the initiation of replication in
holoenzyme, which consists of seven different pro-
Caulobacter crescentus. See Section 23.2.1.
tein subunits: three copies of the dnaX gene product,
For reviews, see: Bravo, A., G. Serrano-Heras, and and a single copy each of four other. The dnaX gene
M. Salas. 2005. Compartmentalization of prokary- encodes two proteins, a full-length protein and a
otic DNA replication. FEMS Microbiol. Rev. truncated version formed by a translational frame-
29:25–47, and Funnel, B. E. 1996. The Role of the shift. A subunit of the clamp loader binds to the DNA
Bacterial Membrane in Chromosome Replication and polymerase, tethering it to the clamp loader, and also
Partition. Chapman & Hall, London. Sequestration interacts with DnaB helicase to coordinate helicase
of the origin also plays a role in regulating transcrip- activity with fork progression. In addition, the sub-
tion in the oriC region, including the transcription of unit in the clamp loader stimulates the release of the
dnaA. For a discussion of this point, see: Bogan, J. A., polymerase from the Okazaki fragment, allowing the
and C. E. Helmstetter. 1997. DNA sequestration and lagging-strand polymerase to be used for the exten-
transcription in the oriC region of Escherichia coli. sion of the next primer.
Mol. Microbiol. 26:889–896. For information on
17. Kuempel, P. L., A. J. Pelletier, and T. M. Hill.
the role of SeqA in sequestering oriC, read: Lu, M., J.
1989. Tus and the terminators: the arrest of replica-
L. Campbell, E. Boye, and N. Kleckner. 1994. SeqA:
tion in prokaryotes. Cell 59:581–583.
a negative modulator of replication initiation in E.
coli. Cell 77:413–426, and Bach, T., M. A. Krekling, 18. Khatri, G. S., T. MacAllister, P. R. Sista, and D.
and K. Skarstad. 2003. Excess SeqA prolongs seques- Bastia. 1989. The replication terminator protein of
tration of oriC and delays nucleoid segregation and E. coli is a DNA sequence-specific contrahelicase.
cell division. EMBO J. 22:315–323. Cell 59:667–674.
Free ebooks ==>
108 the physiology and biochemistry of prokaryotes

19. Manna, A. C., K. S. Pai, D. E. Bussiere, C. 29. Errington, J. 1998. Dramatic new view of bac-
Davies, S. W. White, and D. Bastia. 1996. Helicase– terial chromosome segregation. ASM News 64:
contrahelicase interaction and the mechanism of ter- 210–217.
mination of DNA replication. Cell 87:881–891.
30. Webb, C. D., A. Teleman, S. Gordon, A. Straight,
20. B. subtilis also has 6 Ter sites (TerI–TerVI), and and A. Belmont. 1997. Bipolar localization of the
they are polar as in E. coli. The protein in B. sub- replication origins of chromosomes in vegetative and
tilis that binds to the Ter sequences is called RTP sporulating cells of B. subtilis. Cell 88:667–674.
(replication terminator protein). The B. subtilis Ter
31. Yongfang, L., B. Youngren, K. Serqueev, and S.
sites and RTP do not seem to be related in sequences
Austin. 2003. Segregation of the Escherichia coli chro-
to the E. coli terminators or Tus.
mosome terminus. Mol. Microbiol. 50:825–834.
21. Kato, J.-I., Y. Nishimura, R. Imamura, H. Niki,
32. Mohl, D. A., and J. W. Gober. 1997. Cell cycle–
S. Hiraga, and H. Suzuki. 1990. New topoisomerase
dependent polar localization of chromosome par-
essential for chromosome segregation in E. coli. Cell
titioning proteins in Caulobacter crescentus. Cell
22. Kato, J.-I., Y. Nishimura, M. Yamada, H.
33. Jensen, R. B., S. C. Wang, and L. Shapiro. 2001.
Suzuki, and Y. Hirota. 1988. Gene organization in
A moving DNA replication factory in Caulobacter
the region containing a new gene involved in chro-
crescentus. EMBO J. 20:4952–4963.
mosome partition in Escherichia coli. J. Bacteriol.
170:3967–3977. 34. Reyes-Lamonthe, R., D. J. Sherratt, and M. C.
23. Cell division and DNA metabolism are coupled. Leake. 2010. Stoichiometry and architecture of
That is, cell division is inhibited whenever DNA repli- active DNA replication machinery in Escherichia
cation or partitioning is blocked. This is advantageous coli. Science 328:498–501.
in that it lowers the frequency of formation of anucle- 35. The replisome can be visualized via fluores-
ate cells. When cell division is inhibited, long cells or cence microscopy by fusing the DNA polymerase
filaments form. If chromosome separation or segrega- to the green fluorescent protein (GFP) from the jel-
tion is impaired, then the nucleoids are abnormally lyfish Aequorea victoria. These experiments support
placed in the elongated cells or filaments. There are the idea of a stationary replisome in B. subtilis and
two mechanisms responsible for inhibiting cell divi- E. coli. For more information on how the fusions are
sion. One of these requires SulA. Blocking DNA repli- made and introduced into the cell, review the discus-
cation results in the synthesis of SulA as part of the SOS sion of ZipA in Section 2.6.3.
response. (See Section 16.5.2 for a discussion of the
SOS response and SulA.) SulA inhibits the formation DNA can be visualized by using the fluorescent
of the FtsZ septal ring; hence cell division is inhibited. protein GFP. The DNA can be labeled in specific
There is also a SulA-independent block in cell division regions by inserting into the DNA a cassette of many
whenever DNA replication or segregation is inhibited. copies of lac operators (lacO) and labeling the lacO
copies with a fusion of the Lac repressor (Lacl) to
24. Ward, D., and A. Newton. 1997. Requirement GFP (Lacl–GFP). This has been done to label and fol-
of topoisomerase IV parC and parE genes for cell low unreplicated DNA into the replisome and repli-
cycle progression and developmental regulation in cated DNA out of the replisome. This technique has
Caulobacter crescentus. Mol. Microbiol. 26:897– allowed investigators to demonstrate that the origin
910. is replicated first and the copies move away from the
25. Kuempel, P. L., J. M. Henson, L. Dirks, M. replisome toward opposite cell poles, that chromo-
Tecklenburg, and D. F. Lim. 1991. dif, a recA- somal regions located midway between oriC and
independent recombination site in the terminus terC move toward the centrally located replisome,
region of the chromosome of Escherichia coli. New and that the termini follow last into the replisome
Biol. 3:799–811. and, after replication, become positioned in opposite
halves of the cell, closer to the cell center than the ori-
26. Blakely, G., G. May, R. McCulloch, L. K. gins. The terminus has also been visualized by insert-
Arciszewska, M. Burke, S. T. Lovett, and D. J. ing a single parS site near it and labeling the parS site
Sherratt. 1993. Two related recombinases are with GFP–ParB. ParA and ParB have also been visu-
required for site-specific recombination at dif and cer alized via immunofluorescence microscopy bound as
in E. coli K12. Cell 75:351–361. a complex to the parS site at the origin in C. cres-
27. Pogliano, K., J. Pogliano, and E. Becker. 2003. centus. The bacteria were fixed with glutaraldehyde
Chromosome segregation in eubacteria. Curr. Opin. and formaldehyde and then washed and treated with
Microbiol. 6:586–593. lysozyme. After further treatment with methanol
and acetone, they were incubated with antibody. The
28. Hojgaard, A., H. Szerlong, C. Tabor, and P. antibody was visualized with secondary antibody,
Kuempel. 1999. Norfloxacin-induced cleavage which was goat anti-rabbit IgG conjugated to either
occurs at the dif resolvase locus in Escherichia coli fluorescein or biotin. The biotin-conjugated anti-
and is the result of interaction with topoisomerase body was visualized with streptavidin-conjugated
IV. Mol. Microbiol. 33:1027–1036. Texas red dye, which binds to avidin.
Free ebooks ==>
chromosome replication and partitioning of chromosomes 109

36. Draper, G. C., and J. W. Gober. 2002. Bacterial The par system in the E. coli multidrug-resistant
chromosome segregation. 2002. Annu. Rev. Microbiol. plasmid TP228 consists of (1) an ATPase called ParF,
56:567–597. which is part of a subgroup of the ParA superfamily
of proteins (not related to ParM), (2) a DNA-binding
37. Wu, J. L., and J. Errington. 2002. A large dis- protein called ParG, which is not related to ParB, and
persed chromosomal region required for chromo- (3) a region upstream of the parFG genes to which
some segregation in sporulating cells of Bacillus ParG attaches. References that can be consulted
subtilis. EMBO J. 21:4001–4011. include the following: Golovanov, A. P., D. Barillà,
38. Gerdes, K., J. Møller-Jensen, G. Ebersbach, T. M. Golovanova, F. Hayes, and L.-Y. Lian. 2003.
Kruse, and K. Nordström. 2004. Bacterial mitotic ParG, a protein required for active partition of bacte-
machineries. Cell 116:359–366. ria plasmids, has a dimeric ribbon–helix–helix struc-
ture. Mol. Microbiol. 50:1141, and Barillà, D., and F.
39. Ben-Yehuda, S., D. Z. Rudner, and R. Losick. Hayes. 2003. Architecture of the ParF/ParG protein
2003. RacA, a bacterial protein that anchors chro- complex involved in prokaryotic DNA segregation.
mosomes to the cell poles. Science 299:532–536. Mol. Microbiol. 49:487–499.
40. Wheeler, R. T., and L. Shapiro. 1997. Bacterial 42. Niki, H., A. Jaffé, R. Imamura, T. Ogura, and
chromosome segregation: is there a mitotic appara- S. Hiraga. 1991. The new gene mukB codes for a
tus? Cell 88:577–579. 177 kD protein with coiled-coil domains involved
41. Research with low-copy plasmids has revealed in chromosome partitioning of E. coli. EMBO J.
plasmid segregation systems based upon par systems 10:183–193.
encoded by the plasmid. Generally, the par system 43. Niki, H., R. Imamura, M. Kitaoka, K. Yamanaka,
consists of three components: (1) a Par protein that T. Ogura, and S. Hiraga. 1992. E. coli MukB protein
has ATPase activity, which will be referred to here involved in chromosome partition forms a homodi-
as ParA, (2) a Par DNA-binding protein, referred to mer with a rod-and-hinge structure having DNA
here as ParB, and (3) a centromere-like region in the binding and ATP/GTP binding activities. EMBO J.
DNA to which ParB attaches. As will be explained 11:5101–5109.
later, plasmids can differ in their Par proteins, and
the reference to ParA and ParB here is for the sake of 44. Yamanaka, K., T. Ogura, H. Niki, and S.
convenience. Plasmid pairs move to the midregion of Hiraga. 1996. Identification of two new genes, mukE
the cell prior to septation and then are moved bidirec- and mukF, involved in chromosome partitioning in
tionally (segregated) to opposite cell poles. The Par Escherichia coli. Mol. Gen. Genet. 250:241–251.
proteins are required for the segregation of the plas- 45. Muk is derived from the Japanese word mukaku,
mids. ParA and ParB assemble at the centromere-like which means anucleate. When grown at 22 °C,
region, and the resulting complex of DNA and Par mukB mutants form normal nucleated rods with a
proteins interacts with a bacterial scaffoldlike struc- small percentage (5%) of anucleate cells. At higher
ture that is involved in moving the plasmids bidirec- temperatures (e.g., 42 °C) the cells die, and anucleate
tionally toward the cell poles. Some specific examples cells and multinucleate filaments with abnormal pat-
of plasmid segregation will now be described. terns of nucleoids accumulate (clumps and isolated
The system for the E. coli R1 plasmid consists of the nucleoids). Examination of two paired daughter cells
ParM ATPase, the ParR DNA-binding protein, and revealed that some of the anucleate cells were paired
parC, the putative centromere-like region. The sepa- with cells that appeared to have more than one nucle-
rated paired plasmids at midcell move to opposite cell oid, reflecting the fact that both daughter chromo-
poles. The ParM protein is not related to ParA. Rather, somes remained with one of the daughter cells upon
it is a member of the MreB family and forms actinlike division.
helical filaments at the nucleation point of ParR/parC. Mutations in two other genes also lead to the pro-
(In addition to the ParR/parC complex, polymeriza- duction of anucleate cells. These are the mukF and
tion requires ATP and Mg2+, and depolymerization mukE genes. There probably encode alternatives to
requires ATP hydrolysis.) It has been suggested that the Muk proteins. This conclusion is based upon the
the polymerization of ParM provides the driving force finding that null mutants in mukB, mukF, or mukE
that moves the plasmid bound to the ParM protein show a decreased plating efficiency and a very large
toward a cell pole so that each daughter cell receives at number of anucleate cells only at temperatures
least one copy of the plasmid. This could occur if the higher than 22 °C, indicating that the proteins are
plasmids were located at the tips of the filaments (which not absolutely essential at the lower temperatures.
they are), and if polymerization occurred between Similarly, strains of B. subtilis and C. crescentus
plasmid copies. Another possibility is that the ParM bearing null mutations in smc have a severe plat-
filaments are a scaffolding along which the plasmids ing deficiency in rich media at higher temperatures.
move toward opposite cell halves. For a discussion of The B. subtilis mutants form a significant number
how ParM might be involved in plasmid segregation, of anucleate cells, and the majority of C. crescentus
see: Møller-Jensen, J., R. B. Jensen, J. Löwe, and K. mutants are “stalled” at a predivisional stage, do
Gerdes. 2002. Prokaryotic DNA segregation by an not elongate, and are defective in nucleoid distribu-
actin-like filament. EMBO J. 21:3119–3127. tion.
Free ebooks ==>
110 the physiology and biochemistry of prokaryotes

46. Espeli, O., P. Nurse, C. Levine, C. Lee, and K. J. 54. Immunofluorescence microscopy to visual-
Marians. 2003. SetB: an integral membrane protein ize ParA and ParB was performed with synchro-
that affects chromosome ion in Escherichia coli. Mol. nized cultures and the appropriate antibodies. The
Microbiol. 50:495–509. procedure used is detailed in the last paragraph of
note 34.
47. Van den Ent, F., L. A. Amos, and J. Löwe. 2001.
Prokaryotic origin of the actin cytoskeleton. Nature 55. Jensen, R. B., S. C. Wang, and L. Shapiro. 2001.
413:39–44. A moving DNA replication factory in Caulobacter
crescentus. EMBO J. 20:4952–4963.
48. Figge, R. M., and J. W. Gober. 2003. Cell shape,
division and development: the 2002 American Society 56. Frameshift mutations are mutations due to the
for Microbiology (ASM) Conference on Prokaryotic insertion or removal of one or more (but not three)
Development. Mol. Microbiol. 47:1475–1483. bases from DNA. Acridine dyes such as ethidium
bromide and proflavine cause frameshift mutations.
49. Real, G., S. Autret, J. E. Harry, J. Errington, and They insert between the bases, thus increasing the
A. O. Henriques. 2005. Cell division protein DivlB distance between bases and preventing them from
influences the SpoOJ/Soj system of chromosome aligning correctly. As a consequence, the two strands
segregation in Bacillus subtilis. Mol. Microbiol. of DNA slip with respect to each other. This leads to
55:349–367. the insertion or deletion of a base.
50. Wu, L. J., and J. Errington. 2003. RacA and the 57. Modrich, P. 1991. Mechanisms and biologi-
Soj–SpoOJ systems combine to effect polar chromo- cal effects of mismatch repair. Annu. Rev. Genet.
some segregation in sporulating Bacillus subtilis. 25:229–253.
Mol. Microbiol. 49:1463.
58. Schofield, M. J., and P. Hsieh. 2003. DNA mis-
51. Ben-Yehuda, S., D. Z. Rudner, and R. Losick. match repair: molecular mechanisms and biological
2003. RacA, a bacterial protein that anchors chro- function. Annu. Rev. Microbiol. 57:579–608.
mosomes to the cell poles. Science 299:532–536.
59. Kolodner, R. 1996. Biochemistry and genet-
52. Marston, A. L., H. B. Thomaides, D. H. Edwards, ics of eukaryotic mismatch repair. Genes Dev. 10:
M. E. Sharpe, and J. Errington. 1998. Polar localiza- 1433–1442.
tion of the MinD protein of Bacillus subtilis and its
role in selection of the mid-cell division site. Genes 60. Modrich, P. 1995. Mismatch repair, genetic sta-
Dev. 12:3419–3430. bility and tumour avoidance. Philos. Trans. R. Soc.
Lond. B 347:89–95.
53. Mohl, D. A., and J. W. Gober. 1997. Cell cycle-
dependent polar localization of chromosome par- 61. Modrich, P. 1994. Mismatch repair, genetic sta-
titioning proteins in Caulobacter crescentus. Cell bility and cancer. Science 266:1959–1960.
Free ebooks ==>

Membrane Bioenergetics:
The Proton Potential

A major revolution in our conception of mem- the principles of the chemiosmotic theory will
brane bioenergetics has taken place in the last give the reader a deeper understanding of how
50 years as a result of the theoretical ideas of the bacterial cell uses ion gradients to couple
Peter Mitchell referred to as the chemiosmotic energy-yielding (exergonic) reactions to energy-
theory.1–6 Briefly, the chemiosmotic theory requiring (endergonic) reactions. This chapter
states that energy-transducing membranes explains the principles of the theory. For a brief
(i.e., bacterial cell membranes; mitochondrial historical account, see Box 4.1.
and chloroplast membranes) pump protons
across the membrane, thereby generating an
electrochemical gradient of protons across the 4.1 The Chemiosmotic Theory
membrane (the proton potential) that can be According to the chemiosmotic theory, protons
used to do useful work when the protons return are translocated out of the cell by exergonic
across the membrane to the lower potential. In (energy-producing) driving reactions, which
other words, bacterial, chloroplast, and mito- are usually biochemical reactions (e.g., respira-
chondrial membranes are energized by proton tion, photosynthesis, ATP hydrolysis). Some of
currents. the translocated protons leave behind negative
Of course, the return of the protons across counterions (e.g., hydroxyl ions), thus estab-
the membrane must be through special proton lishing a membrane potential, outside positive.
conductors that couple the translocation of pro- Protons may also accumulate electroneutrally
tons to do useful cellular work. These proton in the extracellular bulk phase, establishing a
conductors are transmembrane proteins. Some proton concentration gradient that is high on
membrane proton conductors are solute trans- the outside (outside acid). When the protons
porters, others synthesize ATP, and others are return to the inside, moving down the concen-
motors that drive flagellar rotation. The proton tration gradient and toward the negative pole of
potential provides the energy for other mem- the membrane potential, work can be done.
brane activities besides ATP synthesis, solute Figure 4.1 illustrates the proton circuit in a
transport, and flagellar motility (e.g., reversed bacterial cell membrane. The cell membrane
electron transport and gliding motility). is similar to a battery in that it maintains a
Because the chemiosmotic theory is central potential difference between the inside and out-
to energy metabolism, it lies at the foundation side, except that the current that flows is one
of all bacterial physiology. This chapter shows of protons rather than electrons. In Fig. 4.1,
how the chemiosmotic theory brings together the potential difference is maintained by reac-
principles of physics and thermodynamics in tions 1, which translocate protons to the out-
explaining membrane bioenergetics. A study of side. Reactions 1 include redox reactions that

Free ebooks ==>
112 the physiology and biochemistry of prokaryotes



The first mechanism suggested to explain electron carrier protein, and this energy is
how electron transport was coupled to then used to drive the synthesis of ATP from
ATP synthesis was published in 1953. It ADP and inorganic phosphate.
was postulated that mitochondria make a Even as late as 1977 there were some
high-energy phosphorylated derivative that reservations. Mitchell’s hypothesis was
donates a phosphoryl group to ADP. The termed “an attractive possibility” that had
model for this mechanism was based upon not been proven.1
the mechanism of ATP synthesis in the cyto- It was, however, accepted as being cor-
sol catalyzed by triosephosphate dehydro- rect and of far-reaching importance, and
genase and phosphoglycerate kinase (see in 1978 Mitchell won the Nobel Prize in
later: Fig. 9.2, reactions 6 and 7). However, Chemistry. For more information about the
no investigators were able to find the postu- prevalent ideas in 1977 regarding electron
lated phosphorylated intermediate despite transport and oxidative phosphorylation,
many attempts. the student is referred to a series of review
A second postulated mechanism was pub- articles.2
lished in 1961 by the English scientist Peter
Mitchell (1920–1992). Mitchell hypoth- REFERENCES
esized that electron transport is coupled to
1. Boyer, P. D. 1977. Coupling mechanisms in
the generation of an electrochemical proton capture, transmission, and use of energy. Annu.
gradient, which in turn drives ATP synthesis Rev. Biochem. 46:957–966.
(the chemiosmotic theory). In 1973 a third
2. Boyer, P. D., B. Chance, L. Ernster, P. Mitchell,
mechanism was suggested, namely, that the E. Racker, and E. C. Slater. 1977. Oxidative
energy released during electron transport is phosphorylation and photophosphorylation.
trapped in a conformational change in an Annu. Rev. Biochem. 46:955–1026.

occur during electron transport (Fig. 4.2, 1) (Fig. 4.2, 2). (See Sections 4.7.1 and 4.8.1.) The
and an ATP-driven proton pump (the ATP syn- reentry of sodium ions can also be coupled to
thase) (Fig. 4.2, 7). These will be discussed later the performance of work (e.g., solute uptake
(Sections 4.7.1 and 4.7.2). The cell membrane Fig. 4.2, 5). Once established, the membrane
does work via reactions 2. The work that is potential can energize the secondary flow of
done by the protons that enter the cell includes other ions. For example, the influx of potassium
the extrusion of sodium ions (Fig. 4.2, 3), sol- ions can be in response to a membrane potential,
ute transport (Fig. 4.2, 4), flagellar rotation inside negative, created by proton extrusion.
(Fig. 4.2, 6), and the synthesis of ATP via the Mitochondrial and chloroplast membranes are
ATP synthase (Fig. 4.2, 7, and Section 4.6.2). also energized by proton gradients. Therefore,
As Mitchell emphasized (see Box 4.1), the mem- this is a widespread phenomenon, not restricted
brane’s low permeability to protons is impor- to prokaryotes.
tant because the major route of proton reentry is
via the energy-transducing proton transporters
rather than by general leakage. This, of course, 4.2 Electrochemical Energy
would be expected of a lipid bilayer that is rela- When bacteria translocate protons across the
tively nonpermeable to protons. Some bacteria membrane to the outside surface, energy is
couple respiration or the decarboxylation of conserved in the proton gradient that has been
carboxylic acids to the extrusion of sodium ions established. The energy in the proton gradient
Free ebooks ==>
membrane bioenergetics: the proton potential 113

H+ The same description applies to chemical

energy. Energy is required to move the proton
against its concentration gradient. This energy
is stored in the concentration gradient. The
energy energy that is stored in a concentration gradi-
ent is called chemical energy. When the proton
returns to the lower concentration, the energy
in the concentration gradient is dissipated and
work can be done.
The sum of the changes in electrical and
chemical energies is called electrochemical
energy. The symbol for electrochemical energy
is ∆μ, which is equal to μin − μout. For the proton,
Fig. 4.1 The proton current. There is a proton circuit it would be ∆μH+ . Electrochemical energy, dis-
traversing the bacterial cell membrane. Protons are cussed in more detail in the sections that follow,
translocated to the cell surface, driven there by either is now expressed in joules per mole.
chemical or light energy through a proton pump
(1) and returned through special proton transport- 4.2.1 The electrochemical energy
ers (2) that do work. The accumulation of protons of protons
on the outside surface of the membrane establishes
a membrane potential, outside positive. A pH gradi-
The proton motive force
ent can also be established, outside acid. In several The electrochemical work that is performed
gram-negative bacteria oxidizing certain inorganic when an ion crosses a membrane is a function of
compounds (lithotrophs), or single-carbon com- both the membrane potential, ∆Ψ, and the dif-
pounds such as methanol, protons that are released ference in concentration between the solutions
into the periplasm via periplasmic oxidations con- separated by the membrane. For example, for
tribute to the proton current (Chapter 13). In some one mole of protons:
cases, periplasmic oxidations are the sole provider of
protons for inward flux. ∆μH+ = F∆Ψ + RT ln([H+]in/[H+]out) joules

is both electrical and chemical. The electrical
energy exists because a positive charge (i.e., In eq. 4.1, F∆Ψ represents the electrical energy
the proton) has been moved to one side of the when one mole of protons moves across a poten-
membrane, creating a charge separation, and tial difference of ∆Ψ volts, and RT ln([H+]in/
therefore a membrane potential. When the [H+]out) represents the chemical energy when
proton moves back into the cell toward the one mole of protons moves across a concentra-
negatively charged surface of the membrane, tion gradient of [H+]in/[H+]out.
the membrane potential is dissipated (i.e., To express eq. 4.1 in millivolts (mV), we sim-
energy has been given up and work can be ply divide by the Faraday constant F (≈96,500
done). The energy dissipated when the proton coulombs/mole; see later). Since RT/F ln([H+]in/
moves to the inside of the cell is equal to the [H+]out) = –60 ∆pH at 30 °C, where ∆pH =
energy required to translocate the proton to pHin – pHout, eq. 4.1 is expressed in millivolts as
the outside. follows:
Stated more precisely, energy is required to
move a charge against the electric field (i.e., ∆p = ∆μH+ /F
to the side of the same charge). This energy is = ∆Ψ − 60 ∆pH mV (at 30 oC) (4.2)
stored in the electric field. The energy that is
stored in the electric field is called electrical Usually, ∆μH+/F is called the proton motive force
energy. Conversely, the electric field gives up and is denoted as ∆p. The ∆p is the potential
energy when a charge moves with the electric energy in the electrochemical proton gradient.
field (i.e., to the opposite pole) and work can When protons move toward the lower electro-
be done. The amount of energy is the same, but chemical potential, the ∆p gives up energy (is
opposite in sign. dissipated) and work can be done (e.g., flagellar
Free ebooks ==>
114 the physiology and biochemistry of prokaryotes

Fig. 4.2 An overview of the proton and sodium ion currents in a generalized bacterial cell Driving reactions
(metabolic reactions) deliver energy to create proton (1) and sodium ion (2) electrochemical gradients, high on
the outside. The major driving reactions encompassed by reaction 1 are the redox reactions that occur during
electron transport. The establishment of sodium ion potentials coupled to metabolic reactions such as respira-
tion is not widespread. When the ions return to the lower electrochemical potentials on the inside, work can
be done. Built into the membrane are various transporters (porters) that translocate protons and sodium ions
back into the cell, completing the circuit, and in the process doing work. There are three classes of porters:
(a) antiporters, which carry two solutes in opposite directions, (b) symporters, which carry two solutes (S) in
the same direction, and (c) uniporters, which carry only a single solute. The Na+/H+ antiporter (3) is the major
mechanism for extruding Na+ in bacteria and also functions to bring protons into the cell for pH homeostasis
in alkaliphilic bacteria. In most bacteria the Na+/H+ antiporter creates the sodium potential necessary for the
Na+/solute symporter because a primary Na+ pump is not present. The antiporter uses the proton electro-
chemical potential as a source of energy. The H+/solute symporter (4) uses the proton potential to accumulate
solutes, and a Na+/solute symporter (5) uses the sodium electrochemical potential to accumulate solutes. Also
shown are a flagellar motor that turns at the expense of the proton electrochemical potential (6) and an ATP
synthase that synthesizes ATP at the expense of the proton electrochemical potential (7). The ATP synthase is
reversible and can create a proton electrochemical potential during ATP hydrolysis.

rotation, ATP synthesis, solute transport). several exergonic reactions (reactions that give
Cells must continuously replenish the ∆p as it up energy), the most widely used being oxida-
is used for doing work. One should view the ∆p tion–reduction reactions in the membrane that
as a force pulling protons across the membrane occur during respiration, and ATP hydrolysis.
into the cell toward their lower electrochemi- Bacteria maintain an average ∆p of approxi-
cal potential. To replenish the ∆p, an equal but mately –140 to –200 mV. The values for respir-
opposite force must be exerted to push protons ing bacteria tend to be a little more positive or
out of the cell toward the higher electrochemical less positive than those for fermenting bacteria.7
potential (against the ∆p). As will be discussed Equation 4.2 is a fundamental equation in cell
later, the force that generates the ∆p (translo- biology. It is derived in more detail later. (See
cates protons out of the cell) can result from eqs. 4.3–4.10.)
Free ebooks ==>
membrane bioenergetics: the proton potential 115

Units: what is meant by volts, as the potential energy in the concentration gra-
electron volts, and joules? dient, 0.180 V (17,400/F). We can also think
It is important to distinguish between volts (V), about volts or millivolts as a force that pushes
electron volts (eV), and joules (J). Potential a molecule down its electrical, electrochemical,
energy differences (e.g., ∆E, ∆Ψ, ∆p) are or chemical gradient. The ∆p, in millivolts, is
expressed as volts or millivolts. As long as the force that pushes protons; thus it is called
charges are not moving, these remain as poten- the proton motive force.
tial energy and work is not done. When charges
are moving, work is being done, either on the A more detailed explanation of and
charges or by the charges, depending upon derivation of eq. 4.2
whether the charges are moving toward a 1. The electrical component of the Δ μH+
higher potential or a lower potential. The quan- A membrane potential, ∆Ψ, exists across the cell
tity of work that is done is proportional to the membrane where ∆Ψ = Ψin – Ψout. By conven-
product of the amount of charge that moves and tion, the ∆Ψ is negative when the inner mem-
the potential difference over which the charge brane surface is negative. The volt is the unit of
moves. ∆Ψ. The work done on or by the electric field
The units of work are either joules or elec- when charges traverse the membrane potential
tron volts. One joule (J) is defined as the energy is equal to the total charges carried by the ions
required to raise a charge of one coulomb (C) or electrons multiplied by the ∆Ψ. If a single
through a potential difference of one volt (i.e., electron or monovalent ion such as the proton
J = C × V). (The older literature used calorie units moves across the membrane, then the work
instead of joules: 1 calorie = 4.184 J.) To calcu- done is ∆Ψ electron volts. (For a divalent ion,
late the change in joules when a mole of mon- the work done would be 2∆Ψ electron volts.)
ovalent ions or electrons travels over a voltage The amount of work that is done per mole of
gradient, one multiplies volts by F, the Faraday protons that traverses the ∆Ψ is
constant, since this is the number of coulombs
∆G = F∆Ψ joules (4.3)
of charge per mole of electrons or monovalent
ions. (See note 8.) One electron volt is (rather Bearing in mind that coulombs × volts = joules
than a coulomb) the increase in energy of a (i.e., V = J/F), eq. 4.3 is often expressed as electri-
single electron or monovalent ion when raised cal potential energy (or force) in volts by divid-
through a potential difference of one volt. The ing both sides of the equation by the Faraday
charge on the electron or monovalent ion is constant:
approximately 1.6 × 10–19 coulomb. Therefore,
one electron volt is equal to 1.6 × 10–19 joule. If ∆G/F = ∆Ψ volts (4.4)
the electrons are moving toward a lower energy 2. The chemical component of the Δ μH+
level (i.e., toward a higher electrode potential), Of course if a concentration gradient of protons
then work (e.g., the generation of a ∆p) can be exists, we must add the chemical energy to the
obtained from the system. The energy available electrical energy in eq. 4.3 to obtain the expres-
from the electron flow is n∆E eV or nF∆E joules sion for the electrochemical energy, ∆μH+. The
(if n refers to moles of electrons). An equivalent chemical energy of the proton (or any solute) as
amount of work must be done to move the elec- a function of its concentration is
trons to a higher energy level.
Similarly, if y protons move over a potential G = G0 + RT ln[H+] joules (4.5)
of ∆p volts, then y∆p electron volts of work is
done. If y moles of protons move, then yF∆p (See note 9 for a more complete discussion.)
joules of work is done. Often work units are The free energy change accompanying the
converted to volts or millivolts when one wishes transfer of one mole of protons between a solu-
to express the potential energy in a concentra- tion of protons outside the cell [H+]out and inside
the cell [H+]in is the difference between the free
tion gradient. For example, 17,400 J is required
energies of the two solutions,
to move one mole of solute against a concentra-
tion gradient of 1,000. This can be expressed ∆G = RT ln[H+]in − RT ln[H+]out joules (4.6)
Free ebooks ==>
116 the physiology and biochemistry of prokaryotes

or motive force, or ∆p) by dividing by the Faraday

constant (or by adding eqs. 4.4 and 4.8):
∆G = RT ln([H+]in /[H+]out) joules
∆p = ∆Ψ − 60 ∆pH millivolts at 30 oC (4.10)
Equation 4.6 refers to the free energy change
when one mole of protons moves from one con- The same equation is used to express the elec-
centration to another, in which the concentra- trochemical potential for any ion (e.g., Na+),
tion gradient does not change (i.e., as applies
∆μNa+ /F = ∆Ψ − 60 ∆pNa millivolts (4.11)
to a steady state). It does not refer to the total
energy released when the concentration of pro- where pNa is −log(Na+).
tons comes to equilibrium. Equation 4.6 can be By convention, the values of the potentials
used for the movement of any solute over a con- are always negative when the cell membrane is
centration gradient, not simply protons, and energized for that particular ion.
we will see this equation again when we discuss
solute transport. (In describing the movement 4.2.2 Generating a ∆Ψ and a ∆pH
of solutes other than protons, the symbol S may We now consider some biophysical aspects of
be substituted for H+. Otherwise, the equation generating the proton motive force. First we
used is identical to eq. 4.6.) will examine the formation of the ∆Ψ, and then
Usually, eq. 4.6 is expressed in electrical units we will consider the establishment of a ∆pH.
of potential (volts). To do this, one substitutes
8.3144 J K–1 mol–1 for R, 303 K (30 °C) for T, Electrogenic flow creates a ∆Ψ
converts ln to log10 by multiplying by 2.303, The movement of an uncompensated charge
and divides by F to convert joules to volts, thus creates the membrane potential. When this
deriving eq. 4.7: happens, the charge movement is said to be
∆G/F = 0.06 log([H+]in/[H+]out) volts electrogenic. The moving charge can be either a
proton or an electron (or in fact any uncompen-
= 60 log([H+]in/[H+]out) millivolts (4.7) sated charge). For example, a membrane poten-
tial is generated when a proton is translocated
Since log([H+]in/[H+]out) = pHout – pHin, eq. 4.7 through the membrane to the outer surface,
can be written as leaving behind a negative charge on the inner
∆G/F = 60 (pHout – pHin)
H+in → H+out
= −60 (pHin – pHout)
A membrane potential can also develop if a
= −60 ∆pH millivolts (4.8)
molecule that has been reduced on the inner
membrane surface, picking up cytoplasmic
3. Proton electrochemical energy, ΔμH+
protons in the process, then diffuses across the
The sum of the electrical (1) and chemical
membrane. This molecule is oxidized on the
energies (2) of the proton is the proton electro-
outer surface, releasing the protons to the exte-
chemical energy. We are now ready to derive
rior, while the electrons return electrogenically
an expression for the proton motive force.
across the membrane to the inner surface:
The total energy change accompanying the
movement of one mole of protons through the A + 2e− + 2H+in → AH2
membrane is the sum of the energy due to the
membrane potential (eq. 4.3) and the energy AH2 → A + 2e− + 2H+out
due to the concentration gradient (eq. 4.6). This
sum is called the electrochemical energy, ∆μH+: This is also electrogenic flow, but here the
electron is the moving charge rather than the
∆μH+ = F∆Ψ + RT ln([H+]in/[H+]out) joules proton. The work done is the same because
the electron carries the same charge as the pro-
ton (i.e., 1.6 × 10–19 C). It makes no difference
One can also express the electrochemical energy from the point of view of calculating the ∆p
as a potential in volts or millivolts (proton whether one thinks in terms of protons moving
Free ebooks ==>
membrane bioenergetics: the proton potential 117

as positive charges across the membrane from limits electrogenic proton pumping and ulti-
the inside to the outside, or simply as being car- mately the size of the membrane potential itself.
ried as hydrogen atoms in a reduced organic This is because the membrane potential, which
compound (e.g., AH2). There is a net transloca- is negative on the inside, pulls protons back into
tion of protons in either case. In both instances, the cell.
the total energy necessary to move y protons
against the ∆p is y∆p electron volts. We will Generating a ∆pH
see that bacteria use both electrogenic electron
When a proton is translocated across the mem-
flow and electrogenic proton flow to create a
brane, a ∆Ψ and a ∆pH cannot be created
membrane potential.
simultaneously. This statement is summarized
The material between the two faces of the
graphically in Fig. 4.3. Let us first discuss cre-
membrane consists in part of long hydrophobic
ating a ∆pH (i.e., accumulating protons in the
regions of the phospholipids in the lipid bilayer
external bulk phase). Remember that the bulk
that prevent the protons from rapidly leaking
external medium can become acidified during
back into the cell, and so the cell membrane
proton translocation only if electrical neutral-
stores positively charged protons on its outside
ity is conserved: that is, each proton in the bulk
surface. It is only when the protons traverse the
phase must have a negative counterion. This
membrane via the appropriate protein machin-
can happen if the proton is pumped out with
ery that the flow of charges can be coupled to
an anion (i.e., H+/R−) or if a cation enters the
doing work (e.g., the synthesis of ATP via the
cell from the external bulk phase in exchange
ATP synthase).
for the proton (i.e., H+ in exchange for R+)
(Fig. 4.3). Since, however, these would be elec-
The membrane potential that develops troneutral events, a charge separation, hence
when even a small number of protons move a ∆Ψ, would not develop. Therefore, in the
across the membrane can be more than
100 mV
The relationship between the positive charge
due to protons that accumulates on one face of
the membrane and the membrane potential is
H+ H+
ΔΨ (volts) = en/c 1 H+ A ΔΨ develops.
where e is the charge per proton (1.6 × 10 A−
coulomb), n is the number of protons, and C is 2 H+ A− A ΔpH develops.
a constant termed the capacitance of the mem- A− H+ H+
brane measured in farads (F). The capacitance is H+ H+ A−
related to the conductivity of material separat- 3 A− R + A
− A ΔpH develops.
ing the two surfaces of the membrane, the area R+ A−
R H+

of the membranes, and the distance between

the membranes. Assuming a membrane area e−
4 A ΔΨ develops.
− +
of about 3 × 10–8 cm2 (for a spherical cell the
size of a typical bacterium), then C ≈ 3 × 10–14 F.
Therefore, the translocation of only 40,000 Fig. 4.3 A ∆pH in the bulk phase and a ∆Ψ can-
protons to the cell surface is sufficient to gener- not develop simultaneously by the movement of a
ate a membrane potential of –200 mV.10,11 (By proton. (1) Electrogenic movement of a proton. (2)
Establishment of a ∆pH by the extrusion of both pro-
convention, the ∆Ψ is said to be negative when
tons and counterions. (3) Establishment of a ∆pH by
the inside potential is negative.)
the exchange of protons for cations in the medium.
The membrane potential that actually devel- Theoretically, a ∆pH could develop if the proton on
ops varies in magnitude from approximately the outer surface of the membrane exchanged with a
–60 mV to about –200 mV, depending upon cation from the bulk phase. But even under these cir-
the bacterium and the growth conditions.6 The cumstances, the membrane potential (positive out-
membrane potential that develops when a rela- side) would limit further efflux of protons. (4) A ∆Ψ
tively small number of protons is translocated can also develop via electrogenic flow of an electron.
Free ebooks ==>
118 the physiology and biochemistry of prokaryotes

absence of compensating ion flow, the protons 4.3 The Contributions of the ∆Ψ
that are pumped out of the cell remain on or and the ∆pH to the Overall ∆ p in
very close to the membrane, and a ∆Ψ rather
Neutrophiles, Acidophiles, and
than a ∆pH (measurable with a pH electrode)
is created. (Theoretically, however, a ∆pH and Alkaliphiles
a ∆Ψ could develop if a cation in the exter- Partly for the reasons stated in Section 4.2.2, the
nal medium exchanged for the proton on the contributions of the ∆Ψ and the ∆pH to the ∆p
membrane.) are never equal. Additionally, the relative con-
Of course, some of the protons could be elec- tributions of the ∆pH and the ∆Ψ to the ∆p vary,
troneutrally released into the bulk phase to cre- depending upon the pH of the environment in
ate a ∆pH, and some might remain at the outer which the bacteria naturally grow. Sections
membrane surface to create a ∆Ψ; but even 4.3.1, 4.3.2, and 4.3.3 summarize, respectively,
under these circumstances, a large ∆pH cannot the relative contributions of the ∆Ψ and the
be generated in the face of a large ∆Ψ. This is ∆pH to the ∆p in neutrophiles, acidophiles, and
because the positive charge on the outside sur- alkaliphiles.12 Notice that in acidophiles and
face inhibits further efflux of protons. In fact, alkaliphiles, the ∆Ψ (acidophiles) or the ∆pH
to demonstrate the formation of a ∆pH experi- (alkaliphiles) has the wrong sign and actually
mentally as a result of proton pumping, a large detracts from the ∆p.
∆Ψ must not be allowed to develop. These con-
ditions are achieved experimentally by making 4.3.1 Neutrophilic bacteria
the membrane permeable either to a cation, so For neutrophilic bacteria (i.e., those that grow
that the incoming cations can compensate elec- with a pH optimum near neutrality), the ∆Ψ
trically for the outgoing protons, or to an anion contributes approximately 70 to 80% to the
that moves in the same direction as the proton. ∆p, with the ∆pH contributing only 20 to 30%.
For example, in many experiments the K+ iono- This is reasonable when one considers that the
phore valinomycin is added to make the mem- intracellular pH is near neutrality and therefore
brane permeable to K+ (Section 4.4). When this the ∆pH cannot be very large.
is done, K+ exchanges for H+, and a ∆pH can
Although bacteria cannot make both a 4.3.2 Acidophilic bacteria
∆Ψ and a ∆pH with the same proton, they For acidophilic bacteria (i.e., those that grow
can create a ∆Ψ during proton translocation between pH 1 and pH 4, and not at neutral pH),
and then convert it to a ∆pH. Suppose a ∆Ψ the ∆Ψ is positive rather than negative (below
is created because a few protons are trans- an external pH of 3, which is where they are usu-
located to the cell membrane outer surface. ally found growing in nature) and thus lowers
Proton translocation cannot proceed for very the ∆p. Under these conditions the force in the
long because a membrane potential devel- ∆p is due entirely to the ∆pH. Let us examine an
ops quickly, which limits further efflux of example of this situation. Thiobacillus ferroox-
protons. However, a cation such as K+ might idans is an aerobic acidophile that can grow at
enter the cell electrogenically. This would pH 2.0 (see Section 13.4.1). Because it has an
result in a lowering of the membrane poten- intracellular pH of 6.5, the ∆pH is 4.5, which
tial because of the positive charge moving in. is very large (remember, ∆pH = pHin – pHout).
Now more protons can be translocated out of When the aerobic acidophiles grow at low pH,
the cell. The protons can leave the outer mem- however, they have a positive ∆Ψ, and for T.
brane surface and accumulate in the external ferrooxidans the ∆Ψ is +10 mV.13 (See note 14.)
phase because they are paired with the anion The contribution of the ∆pH to the ∆p is –60
that was formerly paired with the K+. Thus, a ∆pH or –270 mV. Since ∆p = ∆Ψ – 60 ∆pH, the
membrane potential can form during proton actual ∆p would be –260 mV. In this case, the
translocation and be converted into a ∆pH by ∆Ψ lowered the ∆p by 10 mV. As discussed in
the influx of K+. This is an important way for Section 16.1.3, the inverted membrane poten-
bacteria to maintain a ∆pH, as described later tial is necessary for maintaining the large ∆pH
in Section 16.1.3. in the acidophiles.
Free ebooks ==>
membrane bioenergetics: the proton potential 119

4.3.3 Alkaliphilic bacteria hand, will initially dissipate the ∆Ψ because it

An opposite situation holds for the aerobic electrogenically carries K+ into the cell, thus set-
alkaliphilic bacteria (i.e., those that grow above ting up a potential opposite to the membrane
pH 9, often with optima between pH 10 and potential.
pH 12). For these organisms the ∆pH is one It is also possible to create a ∆Ψ by using
to two units negative (because the cytoplasmic valinomycin. The addition of valinomycin
pH is less than 9.6) and consequently lowers to starved cells or vesicles loaded with K+ will
the ∆p, by 60 to 120 mV.15 Therefore, in the induce a temporary ∆Ψ predicted by the Nernst
alkaliphiles, the potential of the ∆p may come equation, ∆Ψ = –60 log[Sin]/[S]out mV. (This is
entirely from the ∆Ψ. In fact, as we shall explain the concentration gradient expressed as mil-
in Section 4.10, because of the large negative livolts at 30 °C.) What happens is that the K+
∆pH, the calculated ∆p in these organisms is so moves from the high internal concentration to
low that it raises conceptual problems regard- the lower external concentration in the pres-
ing whether there is sufficient energy to synthe- ence of valinomycin (Fig. 4.5). Because the K+
size ATP. moves faster than its counterion, a temporary
diffusion potential, outside positive, is created.
The diffusion potential is temporary because
4.4 Ionophores of the movement of the counterions. The use
Before we continue with the discussion of the of ionophores has helped researchers deter-
proton motive force, we must explain iono- mine which ions are carrying the primary cur-
phores and their use. Ionophores are important rent that is doing the work; it has also assisted
research tools for investigating membrane bio- them investigating the relative importance of
energetics, and their use has contributed to an the membrane potential and ion concentration
understanding of the role of electrochemical gradients in providing the energy for specific
ion gradients in membrane energetics. As men- membrane functions.
tioned earlier, membranes are poorly permeable
to ions, and this is why the membrane can main-
4.4.1 The effect of uncouplers on
tain ion gradients. Ionophores perturb these ion
gradients. Most ionophores are organic com-
pounds that form lipid-soluble complexes with Uncouplers are ionophores that have the fol-
cations (e.g., K+, Na+, H+) and rapidly equilibrate lowing effects:
these across the cell membrane (Fig. 4.4). The 1. They collapse the ∆p and thereby inhibit ATP
incorporation of an ionophore into the mem- synthesis coupled to electron transport.
brane is equivalent to short-circuiting an elec- 2. They stimulate respiration.
trical device with a copper wire. Another way
of saying this is that the ionophore causes the Why should uncouplers stimulate respiration?
electrochemical potential difference of the ion The flow of electrons through electron carriers
to approach zero. Since some ionophores are in the membrane is obligatorily coupled to a
specific for certain ions, it is sometimes possible, flow of protons in a closed circuit (Section 5.5).
by means of the appropriate use of ionophores, This occurs at the coupling sites discussed in
to identify the ion current that is performing Section 4.7.1. Protons are translocated to the
the work. For example, if ATP synthesis is pre- outer surface of the membrane and then reenter
vented by a proton ionophore, this implies that via the ATP synthase. The reentry of the protons
a current of protons carries the energy for ATP through the ATP synthase can be viewed as rate
synthesis. limiting for the circular flow of protons through
One can also preferentially collapse the ∆Ψ the circuit. In the presence of uncouplers such
or ∆pH with judicious use of ionophores, per- as dinitrophenol, protons rapidly enter the cell
haps gaining information about the driving on the uncoupler rather than through the ATP
force for the ion current. For example, nigeri- synthase, thus stimulating electron transport.
cin, which catalyzes an electroneutral exchange Consistent with this model, if proton reentry is
of K+ for H+, will dissipate the pH gradient but blocked by inhibitors of the ATP synthase [e.g.,
not the ∆Ψ.16 Valinomycin plus K+, on the other dicyclohexylcarbodiimide (CDDC)], or slowed
Free ebooks ==>
120 the physiology and biochemistry of prokaryotes






dinitrophenol H+ R–

Fig. 4.4 Examples of ionophores and the transport processes that they catalyze. All the reactions are revers-
ible. Valinomycin, which transports K+ and Rb+, transports only one cation at a time. Since K+ is positively
charged, valinomycin carries out electrogenic transport (i.e., creates a membrane potential). If valinomycin
is added to cells with high intracellular concentrations of K+, then the K+ will rush out of the cell ahead of
counterions and create a transient membrane potential, outside positive, as predicted by the Nernst equation
(given shortly and further discussed later in connection with eq. 4.13). In the presence of excess extracellular
K+ and valinomycin, the K+ will rapidly diffuse into the cell, collapsing the existing potential. Nigericin car-
ries out an electroneutral exchange of K+ for H+. When nigericin is added to cells, one can expect a collapse
of the pH gradient as the internal K+ exchanges for the external H+, but the membrane potential should not
decrease. The combination of nigericin and valinomycin will collapse both the ∆pH and the membrane poten-
tial. Monensin carries out an electroneutral exchange of Na+ or K+ for H+. There is a slight preference for Na+.
Dinitrophenol is an anion (R−) that carries out electroneutral influx of H+ and R− into the cell and returns to
the outside without H+ (i.e., R−). Therefore this anion will collapse both the ∆pH and the ∆Ψ. This is the classic
uncoupler of oxidative phosphorylation. Gramicidin carries out electrogenic transport of H+ > Rb+, K+, Na+.
Gramicidin differs from the other ionophores in that it forms polypeptide channels in the membrane. That is,
it is not a diffusible carrier. Since the addition of gramicidin results in the equilibration of protons across the
cell membrane, it will collapse the ∆Ψ and the ∆pH (i.e., the ∆p). Carbonyl cyanide-p-trifluoromethylhydra-
zone (FCCP) (not shown) is a lipophilic weak acid that exists as the nonprotonated anion (FCCP−) and as the
protonated form (FCCPH), both of which can travel through the membrane. Protons are carried into the cell
in the form of FCCPH. Inside the cell, the FCCPH ionizes to FCCP−, which exits in response to the membrane
potential, outside positive. The result is a collapse of both the ∆pH and the ∆Ψ.

by depletion of ADP, then respiration is slowed. in the overall physiology of the cell. The two
One possible explanation for the slowing of res- components of the ∆p, the ∆Ψ and the ∆pH, are
piration is that, as the protons are translocated to measured separately.17,18
the outside surface, the ∆p rises and approaches
the ∆G of the oxidation−reduction reactions. 4.5.1 Measurement of ∆Ψ
This might slow respiration, since the oxidation–
The membrane potential is measured indirectly
reduction reactions are reversible. One can view
because bacteria are too small for the inser-
this as the ∆p producing a “back pressure.”
tion of electrodes. Suppose a membrane poten-
tial exists and an ion is allowed to come to its
4.5 Measurement of the ∆p electrochemical equilibrium in response to the
Measurements of the size of the ∆p are a neces- potential difference. Then, at equilibrium, the
sary part of analyzing the role that the ∆p plays electrochemical energy of the ion is zero, and we
Free ebooks ==>
membrane bioenergetics: the proton potential 121

Fig. 4.5 Valinomycin-stimulated K+ efflux can impose Fig. 4.6 Measurement of membrane potential by
a temporary membrane potential. When valinomy- the accumulation of a lipophilic cation. The cation
cin is added to cells or membrane vesicles loaded with accumulates in response to the membrane potential
K+, the valinomycin dissolves in the membrane and until the internal concentration reaches equilibrium,
carries the K+ out of the cell. The efflux of K+ creates that is, the point of which efflux equals influx. At this
a diffusion potential, since the negative counterions time the electrochemical energy of the cation is zero
lag behind the K+. The membrane potential is pre- and ∆Ψ = –60 log[(R+)in/(R+)out] mV at 30 °C. Any
dicted by the Nernst equation, ∆Ψ = –60 log(K+)in/ permeant ion can be be used as long as it diffuses pas-
(K+)out. The membrane potential is transient because sively, is used in small amounts, is not metabolized,
it is neutralized by the movement of counterions. and accumulates freely in the bulk phase on either
side of the membrane. Permeant ions that have been
used include lipophilic cations such as tetraphenyl-
can use the equation for electrochemical energy phosphonium (TPP+).
(for a monovalent ion):

∆μS/F = 0 = ∆Ψ + 60 log10([S]in/[S]out) millivolts

the collapse of the membrane potential by the
(4.12) influx of the ion.
Cationic or anionic fluorescent dyes have also
Solving for ΔΨ, we write
been used to measure the ∆Ψ. The dyes partition
∆Ψ = −60 log10([S]in/[S]out) millivolts at 30 °C between the cells and the medium in response to
the membrane potential, and the fluorescence is
(4.13) quenched. The fluorescent dye will monitor rela-
Equation 4.13, a form of the Nernst equa- tive changes in membrane potential (Fig. 4.7).
tion, states that the measurement of the intra- When a fluorescent dye is used to measure the
cellular and extracellular concentrations of absolute membrane potential, it is necessary to
a permeant ion at equilibrium allows one to produce a standard curve. This involves measur-
calculate the membrane potential. The bacte- ing the fluorescence quenching in a sample for
rial cell membrane is relatively nonpermeable which the membrane potential is known—that is,
to ions; therefore, to measure the membrane has been measured by independent means, such
potential, one must use either an ion plus an as the distribution of a permeant ion (Fig. 4.8).
appropriate ionophore (e.g., K+ and valino-
mycin) or a lipophilic ion (i.e., one that can 4.5.2 Measurement of ∆pH
dissolve in the lipid membrane and pass freely A common way to measure ∆pH is to measure
into the cell). When the inside is negative with the distribution of a weak acid or weak base at
respect to the outside, a cation (R+) is cho- equilibrium between the inside and the outside
sen because it accumulates inside the cells or of the cell. The assumption is that the uncharged
vesicles (Fig. 4.6). When the inside is positive, molecule freely diffuses across the membrane but
an anion (R−) is used. It is important to use the ionized molecule cannot. Inside the cell the
a small concentration of the ion to prevent acid becomes deprotonated or the base becomes
Free ebooks ==>
122 the physiology and biochemistry of prokaryotes

Fig. 4.7 The use of fluorescence to measure the mem- Fig. 4.8 Relationship between the membrane poten-
brane potential. If valinomycin were added to bacteria tial and fluorescence change of 1,1′-dihexyl-2,2′-
in buffer containing low concentrations of potassium oxacarbocyanine (CC6). The changes in fluorescence
ion in the presence of a fluorescent cationic lipophilic (ordinate) are plotted as a continuous line corre-
probe, one might expect the potassium inside to rush sponding to different membrane potentials imposed
out, creating a diffusion potential as predicted by by the addition of valinomycin to Streptococcus
the Nernst equation, whereupon the cationic probe lactis cells. The membrane potentials were calcu-
would enter in response to the membrane potential, lated by using the Nernst equation from potassium
and the fluorescence would be quenched. It has been concentration ratios (in/out) in parallel experiments,
suggested that the quenching of fluorescence is due to where the intracellular K+ concentrations were about
the formation of dye aggregates with reduced fluores- 400 mM and the extracellular concentrations were
cence inside the cell. varied. Also shown is the membrane potential caused
by the addition of glucose (rather than valinomycin)
to the cells. Source: Adapted from Maloney, P. C.,
E. R. Kashket, and T. H. Wilson. 1975. Methods for
protonated, the extent of which depends upon studying transport in bacteria, pp. 1–49. In: Methods
the intracellular pH. For example, in Membrane Biology, Vol. 5. E. D. Korn (Ed.).
Plenum Press, New York.
AH → A− + H+ or B + H+ → BH+

On addition of a weak acid (or base) to a cell sus-

pHin − pHout = ∆pH
pension, the charged molecule accumulates in
the cell. One uses a weak acid (e.g., acetic acid) = log10 [A−]in/[A−]ou t (4.14)
when pHin exceeds pHout because the acid will
Thus, the log10 of the ratio of concentrations
ionize more extensively at the higher pH, and a
inside to outside the weak acid is equal to the
weak base (e.g., methylamine) when pHin is less
∆pH. In practice one uses radioactive acids or
that pHout because such a base will become more
bases as probes and measures the amount of
protonated at the lower pH. At equilibrium, for
radioactivity taken up by the cells. A more com-
a weak acid,
plex equation must be used when the concen-
Ka = [H+]in[A−]in /[AH]in tration of the un-ionized acid (AH) cannot be
ignored.19 Cytoplasmic pH is sometimes also
= [H+]out[A−]out/[AH]out measured by 31P nuclear magnetic resonance
(31P NMR) of phosphate whose spectrum is pH
where Ka is the dissociation constant of the
dependent. (See note 20.)
acid, HA.
If pHin and pHout are at least 2 units higher than
the pK, most of the acid is ionized on both sides 4.6 Use of the ∆p to Do Work
of the membrane and there is no need to take The ∆p provides the energy for several mem-
into account the un-ionized acid. Therefore, the brane functions, including solute transport and
∆pH can be calculated by using eq. 4.14. ATP synthesis discussed in this section. (See
Free ebooks ==>
membrane bioenergetics: the proton potential 123

Chapter 17 for a more complete discussion of but more details are provided shortly, when the
solute transport.) mechanism of its function is discussed. (See next
subsection, entitled Model of the mechanism of
4.6.1 Use of the ∆p to drive ATP synthase.) Mitochondria, chloroplasts,
solute uptake and bacteria have a similar ATP synthase with
As an example of how the ∆p can be used for homologous proteins. The ATP synthase is a
doing work, consider symport of an uncharged membrane-embedded rotary engine consisting
solute, S, with protons (Fig. 4.2, 4). The total of two regions, the F0 and F1 (Fig. 4.9).23
driving force on S at 30 °C is The F0 region spans the membrane and serves
as a proton channel through which the protons
y∆p + 60 log([S]in/[S]out) millivolts (4.15) are pumped. The F1 region is located on the
where y is the ratio of H+ to S, and 60 log[S]in/ inner surface of the membrane and is the cata-
[S]out, the force involved in moving S from one lytic subunit responsible for the synthesis and
concentration to another, is the same as eq. 4.7 hydrolysis of ATP. The F1 unit is also called the
but written for S, the solute, rather than for the coupling factor. The F1 subunit from E. coli is
proton. made from five different polypeptides with the
At equilibrium, the sum of the forces is zero, following stoichiometry: α3, β3, γ1, δ1, and ε1.
and therefore, Each of the β-polypeptides contains a single
catalytic site for ATP synthesis and hydrolysis,
y∆p = –60 log([S]in/[S]out ) (4.16) although the sites are not equivalent at any one
time. The F0 portion has three different polypep-
For y = 1 and ∆p = –0.180 V, we write tides, which in E. coli are stoichiometrically a1,
3 = log[S]in/[S]out and [S]in/[S]out = 103 b2, and c10. All in all, E. coli uses polypeptides of
8 different types to construct a 22-polypeptide
Therefore, a ∆p of –180 mV could maintain a machine that acts as a reversible pump driven
103 concentration gradient of Sin/Sout if the ratio by the proton potential or by ATP hydrolysis.
of H+/S were 1. If the ratio of H+/S were 2, then As discussed more completely in Section 4.7.2,
a concentration gradient of 106 could be main- the amount of energy to synthesize one mole of
tained. It can be seen that very large concentra-
tion gradients can be maintained by using the ∆p.

4.6.2 The ATP synthase IN OUT

Built into the cell membranes of prokaryotes
and mitochondrial and chloroplast membranes
is a complex protein complex that couples the ADP + Pi
translocation of protons down an electrochemi- 3H+
cal proton gradient (∆p) to the phosphorylation F0
of ADP to make ATP. It is called the proton-
translocating ATP synthase, or simply ATP
synthase. As discussed later, in Section 4.7.2,
the ATP synthase is reversible and can pump
protons out of the cell, generating a ∆p. As
discussed in Section 4.8.1, some bacteria such
as Propionigenium modestum have a Na+- Fig. 4.9 ATP synthase. The proton channel F0 spans
the membrane. It consists of polypeptides of types a,
dependent ATP synthase and utilize the flow
b, and c. The catalytic subunit F1, on the inner mem-
of Na+ rather than H+ down an electrochemi- brane surface, catalyzes the reversible hydrolysis of
cal gradient through the ATP synthase to make ATP. It consists of polypeptides of types α, β, γ, δ,
ATP. and ε. Under physiological conditions, the ATP syn-
thase reaction is poised to proceed in either direction.
Description of the ATP synthase When ∆p levels decrease relative to ATP, they can
For an overview of the ATP synthase, see refs. 21 be restored by ATP hydrolysis. When the ATP levels
and 22. The structure is briefly described here, decrease relative to ∆p, more ATP can be made.
Free ebooks ==>
124 the physiology and biochemistry of prokaryotes

ATP is given by ∆Gp. Assume that ∆Gp is equal (T, or tight, conformation), and ATP released
to approximately 50,000 joules, or (dividing by from the enzyme (O, or open, conformation).
the Faraday constant) 518 mV. Thus, y∆p must What drives the conformational changes?
be ≥ –518 mV, where y is the number of enter- A remarkable mechanism has come to light.
ing protons. If y = 3, then to synthesize one mole Evidence suggests that the ATP synthase con-
of ATP, the ∆p must be at least –518/3 or –173 tains a centrally located rotary motor driven by
mV. (By convention the ∆p is negative when proton translocation and that the rotation of
energy is available to do work.) (See note 24.) the motor causes the conformational changes at
the catalytic sites in the β subunits. The motor
Model of the mechanism of ATP synthase consists of the cn subunits comprising a ring in
The mechanism of ATP synthesis/hydrolysis by F0, linked to the γ subunit and the ε subunit in
the ATP synthase is complex. One model, the F1. As mentioned, the γ subunit along with the
binding-change mechanism, is based upon the ε subunit forms a central stalk in F1. The idea is
finding that purified F1 will synthesize tightly that proton flux through F0 causes the cn ring to
bound ATP in the absence of a ∆p.25 (The Kd rotate, and this in turn rotates γε in F1. When
for the high-affinity site is 10–12 M.) The equi- the γε rotor turns as a consequence of the influx
librium constant between enzyme-bound ATP of protons, the γ subunit sequentially makes
and bound hydrolysis products (ADP and Pi) is contact with the three different αβ subunits,
approximately one, with the use of soluble F1.26 which are prevented from rotating with γε,
The tight binding of ATP and the equilibrium and force is exerted sequentially on each of the
constant of approximately one suggest that the three β subunits, causing each one to undergo
energy requirement for net ATP synthesis is for three sequential conformational changes (L, T,
the release of ATP from the enzyme rather than or O). As mentioned, each site binds ADP and
for its synthesis. The model proposes that when Pi while in the L conformation, synthesizes ATP
protons move down the electrochemical gradi- while in the T conformation, and releases ATP
ent through the F0F1 complex, a conformational while in the O conformation. It takes approxi-
change occurs in F1 that results in the release of mately three H+ ions traversing the F0 portion of
newly synthesized ATP from the high-affinity the ATP synthase to make one ATP. The molec-
site. This is illustrated in Fig. 4.10A. ular structure of the ATP synthase, as well as
To understand the proposed mechanism of evidence that it contains a rotating motor, can
how ATP synthesis is energized, it is necessary be found in refs. 27 through 30.
to add some more details about the structure As discussed later in Section 4.7.1, certain
of the ATP synthase. (See Fig. 4.10B.) In the F1 bacteria use a sodium motive force, rather than
region the three α and three β subunits alternate a proton motive force, to provide energy to
to form a hollow cylinder that surrounds the. γ the ATP synthase. The rotating motor model
subunit, which interacts with the three α and applies to the sodium motive force as well as
three β subunits. The. γ subunit forms a central the proton motive force. Recall that bacteria
stalk with the ε subunit and protrudes from the have another membrane-bound rotating motor
bottom of the cylinder, where it attaches to the driven by proton influx. It is the flagellum motor
c subunits, which are in the form of a cylindrical described in Section 1.2.1.
ring in the F0 part embedded in the membrane.
The a and b subunits are connected laterally
(peripherally) to the c ring. The b subunits form
4.7 Exergonic Reactions That
a peripheral stalk connecting the a subunit to a Generate a ∆p
complex of subunits α and β. Section 4.6 described the influx of protons down
As Fig. 4.10A points out, the “binding- a proton potential gradient coupled to the per-
change mechanism” postulates that there are formance of work (e.g., solute transport or ATP
three catalytic sites for synthesizing ATP, one synthesis). These activities dissipate the ∆p. Let
on each of the three β subunits, each of which us now consider driving forces that generate a
is paired with an α subunit. Each catalytic site ∆p, that is, forces that move protons out of the
on a β subunit must cycle through three confor- cell toward a higher potential. The major driv-
mational changes for ADP and Pi to be bound ing reactions in most prokaryotes are the redox
(L, or loose, conformation), ATP synthesized reactions in the cell membranes of respiring
Free ebooks ==>
membrane bioenergetics: the proton potential 125

c ring
H+ periplasm

F0 membrane

gamma cytoplasm
beta ADP + Pi

alpha alpha

Fig. 4.10 (A) Binding-change mechanism for ATP synthase. Because there are three β subunits encoded by a
single gene, and because there is only one catalytic site per β subunit, the maximum possible number of cata-
lytic sites is three. However, the three catalytic sites are not functionally equivalent and are thought to cycle
through conformational changes driven by the electrochemical proton gradient. As a result of the conforma-
tional changes, newly synthesized ATP is released from the catalytic site. The three conformational states are
O (open, very low affinity for substrates), L (loose binding,), and T (tight binding, the active catalytic site).
ATP is made spontaneously at site T, which converts to site O in the presence of a ∆p and releases the ATP.
The ∆p also drives the conversion of the preexisting site O to site L, which binds ADP and Pi, as well as the
conversion of the preexisting site L to site T. In this model, the only two catalytic sites that bind substrates
are L and T. Source: Adapted from Cross, L. R., D. Cunningham, and J. K. Tamura. 1984. Curr. Top. Cell
Regul. 24:335–344. (B). Rotary model for E. coli ATP synthase. The ten c subunits in the F0 portion form a
cylindrical ring in the lipid bilayer. The ε and γ subunits interact with each other and form a central stalk in
the F1 portion that is connected to the c ring. Surrounding the εγ stalk are alternating α3β3 subunits. Subunits
a and b are at the periphery of the c ring. The a subunit is thought to be confined to the membrane, and the two
b subunits extend as a second stalk peripherally located between F0 and F1. The b stalk connects to a complex
of subunits, δ and α in the F1 portion. The rotor is considered to be the c ring and the εγ stalk, and α3β3δab2
is proposed as the stator against which the rotor moves. It is thought that protons move through the a and c
subunits across the membrane. The flux of H+ (or Na+ in the case of certain bacteria that use a sodium motive
force) through the F0 portion is believed to cause the rotor to move with respect to the stator. This is believed
to induce conformational changes in the catalytic subunits, α3, resulting in ATP synthesis. Source: Adapted
from Nicholls, D. G., and S. J. Ferguson. 2000. Bioenergetics 3. Academic Press, San Diego, CA.
Free ebooks ==>
126 the physiology and biochemistry of prokaryotes

organisms (electron transport), ATP hydroly- or

sis in fermenting organisms, and several other
Eh = E0 + 2.3[RT/nF] log10([ox]/[red])
less frequently used driving reactions, which we
shall consider in Section 4.8. At this time, let us In these equations n is the number of electrons
consider some general thermodyamic features transferred per molecule, R is the gas constant,
common to all the driving reactions that gen- and F is the Faraday constant. Usually standard
erate the proton gradient. The translocation of potentials at pH 7 are quoted and the symbol is
protons out of cells is an energy-requiring pro- E″0
(See note 31 for more discussion of the rela-
cess that can be written as tionship of E0 to pH.)
You can see that it is important to know the
yH+in → yH+out
concentrations of the oxidized and the reduced
∆G = yF∆p joules or y∆p eV forms to be able to find the actual electrode
potential. Molecules with more negative Eh
That is, the amount of energy required to trans-
values are better donors (i.e., electrons sponta-
locate y moles of protons out of the cell is yF∆p
neously flow to molecules with more positive
joules, or for y protons, y∆p electron volts.
Eh values). It is important to understand that
(This is the same energy, but of opposite sign,
when an electron moves over a ∆Eh (which is
that is released when the protons enter the cell.)
Eh,acceptor – Eh,donor) to a higher electrode poten-
Therefore, the ∆G of the driving reaction must
tial, the electron is actually moving toward a
be equal to or greater than yF∆p joules or y∆p
lower potential energy and, as a consequence,
electron volts. If the reaction is near equilibrium,
energy is released. The energy released is pro-
which is the case for the major driving reactions,
portional to the potential difference (∆Eh volts)
the energy available from the driving reaction
over which the electron travels. When n elec-
is approximately equal to the energy available
trons move over a potential difference of ∆Eh
from the proton gradient and ∆Gdriving reaction =
volts, the energy released is equal to n∆Eh elec-
yF∆p. This relationship allows one to calculate
tron volts. (Recall that potential differences
the ∆p if y and ∆Gdriving reaction are known.
are given in volts but work units are in elec-
The most common classes of driving reac-
tron volts.) Since the total charge carried by
tions, which we shall discuss first, are oxidation–
one mole of electrons is F coulombs, the work
reduction reactions that occur during respiration
done per n moles of electrons is ∆G = –nF∆Eh
and photosynthetic electron transport, and ATP
joules, where ∆G is the Gibbs free energy. The
hydrolysis. In all cases we will write the ∆G for
negative sign in the equation shows that energy
the driving reaction. Then we will equate the
is released (work is done) when the ∆Eh is posi-
∆Gdriving reaction with yF∆p. Finally, we will solve
tive. As described next, the movement of elec-
for ∆p.
trons during an oxidation–reduction reaction
toward a lower potential energy level (higher
4.7.1 Oxidation–reduction reactions electrode potential) can be coupled to the
as driving reactions extrusion of protons toward a higher potential
energy level (higher ∆p), that is, to the genera-
Electrode potentials and energy change during tion of a ∆p.
oxidation–reduction reactions
The tendency of a molecule (A) to accept an Oxidation–reduction reactions can
electron from another molecule is given by its
generate a ∆p
electrode potential E, also called the reduction
An important method of generating a ∆p is
potential, the redox potential, or the oxidation–
to couple proton translocation to oxidation–
reduction potential. Electrons spontaneously
reduction reactions that occur during electron
flow toward molecules with a higher electrode
transport. The details of electron transport are
potential. Under standard conditions (1 M for
discussed in Chapter 5; here we will simply com-
all solutes and gases at 1 atm), the electrode
pute the ∆p that can be created. Energy released
potential has the symbol E0 and is related to the
from oxidation–reductions can generate a ∆p
actual electrode potential Eh as follows:
because the oxidation–reduction reactions are
Eh = E0 + [RT/nF] ln([ox]/[red]) coupled to proton translocation. We postpone
Free ebooks ==>
membrane bioenergetics: the proton potential 127

a description of the mechanism of coupling until approximately +0.2 V (200 mV). Substituting
Section 5.5, concentrating here on the energetic this value for ∆Eh, and by using n = 2 and y = 4,
relationship between the ∆Eh and the ∆p. we obtain a ∆p of –0.1 V (–100 mV). That is,
Let us do a simple calculation to illustrate how when two electrons travel down a ∆Eh of 200
to estimate the size of the ∆p that can be gener- mV and four protons are translocated, –100
ated by an oxidation–reduction reaction that is mV is stored in the ∆p. For another way of
coupled to proton translocation. Consider what looking at this, assume that when two electrons
happens in mitochondria and in many bacteria. travel down a redox gradient of 0.2 V, there is
A common oxidation–reduction reaction in the 0.4 eV available for doing work (2 × 0.2 V). The
respiratory chain is the oxidation of reduced 0.4 eV is used to raise each of four protons to a
ubiquinone, UQH2, by cytochrome c1 (cyt c1) in ∆p of 0.1 V. Oxidation–reduction reactions in
an enzyme complex called the bc1 complex. The the respiratory chain that are coupled to proton
oxidation–reduction is coupled to the translo- translocation are called coupling sites, of which
cation of protons across the membrane, hence the bc1 complex is an example. Coupling sites
the creation of a ∆p. As stated earlier, the total are discussed in more detail in Section 5.5.
energy change during an oxidation–reduction Coupling of redox reactions during electron
reaction is ∆G = –nF∆Eh joules, where n is the transport to proton translocation is the main
number of moles of electrons. The total energy process by which a ∆p is created in respiring
change during proton translocation is yF∆p bacteria, in phototrophic prokaryotes, in chlo-
joules, where y is the number of moles of trans- roplasts, and in mitochondria.
located protons. [One can convert to electrical
potential (volts) and describe these reactions in Reversed electron transport
terms of forces by dividing by the Faraday con- Some of the redox reactions in the respiratory
stant.] This is summarized as follows: pathway are in equilibrium with the ∆p, and
1. UQH2 + 2 cyt c1(ox) this has important physiological consequences.
→ UQ + 2 cyt c1(red) + 2H+ For example, it supports the expectation that
2. yH+in→ yH+out the ∆p can drive electron transport in reverse.
That is, protons driven into the cell by the ∆p
Reaction 1 gives the Gibbs free energy, in joules, at coupling sites can drive electrons to the more
and in reaction 2, for yF∆p, the units are joules, negative electrode potential (e.g., the reversal of
as well. reactions 2 and 1, shown earlier for the oxida-
These reactions (1 and 2) are coupled; that is, tion of UQH2 by cyt c1 coupled to the extrusion
one cannot proceed without the other. What is of protons). In fact, one test for the functioning
the size of the ∆p that can be generated? of reversed electron transport is its inhibition by
Reactions 1 and 2 are close to equilibrium ionophores that collapse the ∆p (Section 4.4).
and can proceed in either direction. Therefore Reversed electron transport commonly occurs
one can write that the total force available from in bacteria that use inorganic compounds (e.g.,
the redox reaction is equal to the total force of ammonia, nitrite, sulfur) as a source of electrons
the proton potential: to reduce NAD+ for biosynthesis of cell material
(chemolithotrophs), since these electron donors
∆G/F = −n∆Eh = y∆p (4.17)
are at a potential higher than NAD+. (A list of
Equation 4.17 summarizes an important relation- electrode potentials of biological molecules is
ship between the ∆Eh of an oxidation−reduction provided later: see Table 5.1.) The chemolitho-
reaction during respiration and the ∆p that can trophs are discussed in Chapter 13.
be generated at a coupling site. We will return
to this point, and to this equation, in Chapter 5 Respiration coupled to sodium ion efflux
when coupling sites are discussed in more Although respiratory chains coupled to pro-
detail. ton translocation appear to be the rule in most
Now we solve for ∆p. Four protons (y = 4) are bacteria, a respiration-linked Na+ pump (a Na+-
translocated for every two electrons (n = 2) that dependent NADH–quinone reductase) has been
travel from reduced quinone to cytochrome c1. reported in several halophilic marine bacteria
The ∆Eh between quinone and cytochrome c1 is that require high concentrations of Na+ (0.5 M)
Free ebooks ==>
128 the physiology and biochemistry of prokaryotes

for optimal growth.32,33 The situation has been is usually created by Na+/H+ antiport rather
well studied with Vibrio alginolyticus, an alka- than by a primary Na+ pump. For example, the
lotolerant marine bacterium that uses a ∆μNa+ marine sulfate reducer Desulfovibrio salexi-
for solute transport, flagellar rotation, and ATP gens, which uses a ∆μNa+ for sulfate accumula-
synthesis (at alkaline pH). tion, generates the ∆μNa+ by electrogenic Na+/
V. alginolyticus creates the ∆μNa+ in two ways. H+ antiport driven by the ∆μH+, which is cre-
At pH 6.5, a respiration-driven H+ pump gener- ated by a respiration-linked proton pump.37
ates a ∆μH+ , which drives a Na+–H+ antiporter Similarly, nonmarine aerobic alkaliphiles
that creates the ∆μNa+. The antiporter creates the belonging to the genus Bacillus, which rely
sodium ion gradient by coupling the influx of on the ∆μNa+ for most solute transport and
protons (down the proton electrochemical gra- for flagella rotation, create the ∆μNa+ by using
dient) with the efflux of sodium ions. However, a Na+/H+ antiporter driven by a ∆μH+ that is
at pH 8.5, the ∆μNa+
is created directly by a res- created by a respiration-linked proton pump
piration-driven Na+ pump, the Na+-dependent (Section 4.10). Furthermore, in D. salexigens
NADH–quinone reductase. In agreement with as well as the alkaliphilic Bacillus species, the
this conclusion, the generation of the membrane membrane-bound ATP synthase is H+ linked
potential at alkaline pH (pH 8.5) is resistant to rather than Na+ linked.
the proton ionophore m-carbonylcyanide phe-
nylhydrazone (CCCP), which short-circuits the 4.7.2 ATP hydrolysis as a driving reaction
proton current (Section 4.4) but is sensitive to for creating a ∆ p
CCCP at pH 6.5.34 Electron transport reactions are the major
It has been suggested that switching to a energy source for creating a ∆p in respir-
Na+-dependent respiration at alkaline pH ing organisms, but not in fermenting bacte-
may be energetically economical.35 The rea- ria. (However, see note 38.) A major energy
soning is that when the external pH is more source for the creation of the ∆p in fermenting
alkaline than the cytoplasmic pH, the only bacteria is ATP hydrolysis catalyzed by the
part of the ∆p that contributes energy to the membrane-bound, proton-translocating ATP
antiporter is the ∆Ψ, and therefore the anti- synthase, which yields considerable energy
porter must be electrogenic. That is, the H+/ (Fig. 4.2, 7). Consider the following coupled
Na+ must be greater than one. If one assumes reactions:
that the antiporter is electroneutral when the
external pH is acidic, then the continued use 1. ATP + H2O → ADP + Pi ∆Gp
of the antiporter at alkaline pH would neces- 2. yH+in → yH+out yF∆p
sitate increased pumping of protons out of the To refer to the energy of ATP hydrolysis or syn-
cell by the primary proton-linked respiration thesis by means of physiological concentrations
pumps. Rather than do this, the cells simply of ADP, Pi, and ATP, the term ∆Gp (phosphory-
switch to a Na+-dependent respiration pump lation potential) is used instead of ∆G:
to generate the ∆μNa+. This argument assumes
that the ratios Na+/e− and H+/e– are identical so ∆Gp = ∆G0′ + 2.303RT log[ATP]/[ADP][Pi]
that the energy economies of the respiration- (4.18)
linked cation pumps are the same. (See note
36 for additional information regarding pro- The ATP synthase reaction is close to equi-
ton and sodium ion pumping in V. alginolyti- librium and can operate in either direction.
cus.) Several other bacteria have been found to Therefore, the total force available from the
generate sodium ion potentials by a primary proton potential equals the force available from
process. For example, primary Na+ pumping ATP hydrolysis:
is also catalyzed by sodium ion translocating
∆Gp/F = y∆p volts 4.19)
decarboxylases in certain anaerobic bacteria,
as described shortly (Section 4.8.1). It must be The value of ∆Gp is about –50,000 joules (518 mV),
pointed out, however, that although the ∆μNa+ and the consensus value for y is 3. Under these
is relied upon for solute transport and motil- circumstances, the hydrolysis of one ATP would
ity in many other Na+-dependent bacteria, it generate a maximum ∆p of –173 mV.
Free ebooks ==>
membrane bioenergetics: the proton potential 129

4.8 Other Mechanisms for Creating a to the electrogenic translocation of two moles
∆Ψ or a ∆ p of sodium ions per mole of methylmalonyl–
CoA decarboxylated46:
Redox reactions and ATP hydrolysis are the
most common driving reactions for creating a Methylmalonyl–CoA + 2Na+in
proton potential. However, other mechanisms → propionyl–CoA + CO2 + 2Na+out
exist for generating proton potentials, and even
P. modestum differs from most known bacte-
sodium ion potentials. These driving reactions
ria in that it has a Na+-dependent ATP synthase
are not as widespread among the prokary-
rather than a H+-dependent ATP synthase, and
otes as the others. Nevertheless, they are very
thus it relies on a Na+ current to make ATP. The
important for certain groups of prokaryotes,
consequence of this is that, when P. modestum
especially anaerobic bacteria and halophilic
grows on succinate, the decarboxylation of
archaea. In some instances, they may be the only
methylmalonyl–CoA is its only source of ATP.
source of ATP. Some of these driving reactions
Another example is V. alcalescens. This is an
are considered next.
anaerobic gram-negative bacterium that grows
in the alimentary canal and mouth of humans
4.8.1 Sodium transport decarboxylases and other animals. Like P. modestum, it is unable
can create a sodium potential to ferment carbohydrates, but it does ferment
Although chemical reactions directly linked to the carboxylic acids lactate, malate, and fumar-
Na+ translocation (primary transport of Na+) ate to propionate, acetate, H2, and CO2. During
are not widespread among the bacteria, since the fermentation, pyruvate and methylmalo-
most primary transport is of the proton, they nyl–CoA are formed as intermediates, as in P.
can be very important for certain bacteria.39–44 modestum.47 Some of the pyruvate is oxidized
An example of primary Na+ transport is the Na+ to acetate, CO2, and H2 with the formation of
pump coupled to respiration in Vibrio alginolyt- one ATP via substrate-level phosphorylation
icus, discussed in Section 4.7.1. There is also (Section 8.3). Pyruvate will then be converted
the decarboxylation of organic acids coupled to methylmalonyl–CoA that is decarboxylated
to sodium ion efflux. This occurs in anaerobic to propionyl–CoA coupled to the extrusion of
bacteria that generate a sodium gradient by cou- sodium ions. The sodium ion gradient that is
pling the decarboxylation of a carboxylic acid created might be used as a source of energy for
to the electrogenic efflux of sodium ions. The solute uptake via a process called Na+-coupled
decarboxylases include methylmalonyl–CoA symport, described in Section 17.3.1.
decarboxylase from Veillonella alcalescens and Acidaminococcus fermentans is an anaer-
Propionigenium modestum, glutaconyl–SCoA obe that ferments the amino acid glutamate to
decarboxylase from Acidaminococcus fermen- acetate and butyrate. During the fermentation,
tans, and oxaloacetate decarboxylase from glutaconyl–CoA is decarboxylated to crotonyl–
Klebsiella pneumoniae and Salmonella typh- CoA by a decarboxylase that is coupled to the
imurium. A description of these bacteria and translocation of sodium ions, as just shown for
their metabolism will explain the importance methylmalonyl–CoA48:
of the sodium-translocating decarboxylases as
Glutaconyl–CoA + yNa+in
a source of energy.
→ crotonyl–CoA + CO2 + yNa+out
Propionigenium modestum is an anaerobe
isolated from marine and freshwater mud, and
from human saliva.45 It grows only on carboxy- The oxaloacetate decarboxylase in
lic acids (i.e., succinate, fumarate, L-aspartate, Klebsiella pneumoniae
L-malate, oxaloacetate, and pyruvate). The car- We will examine the Na+-dependent decar-
boxylic acids are fermented to propionate and boxylation of oxaloacetate by oxaloacetate
acetate, forming methylmalonyl–CoA as an decarboxylase from the facultative anaerobe K.
intermediate. (The description of the propionic pneumoniae because this process has been well
acid fermentation in Section 15.7 explains these studied. A substantial sodium potential develops
reactions.) The methylmalonyl–CoA is decar- because the decarboxylation of oxaloacetate is
boxylated to propionyl–CoA and CO2 coupled coupled to the electrogenic efflux of sodium ion.
Free ebooks ==>
130 the physiology and biochemistry of prokaryotes

(The standard free energy for the decarboxyla- sodium ion was prevented by avidin, an inhibi-
tion reaction is approximately –29,000 J.) The tor of the oxaloacetate decarboxylase.
enzyme from K. pneumoniae translocates two
2. The structure of the oxaloacetate
Na+ ions out of the cell per oxaloacetate decar-
boxylated according to the following reaction:
Oxaloacetate decarboxylase consists of two
2Na+in + oxaloacetate parts, a peripheral catalytic portion attached
→ 2Na+out + pyruvate + CO2 to the inner surface of the membrane and an
integral membrane portion that serves as a Na+
1 Evidence that oxaloacetate decarboxylase channel (see note 36). The integral membrane
is a sodium pump protein also takes part in the decarboxylation
Inverted membrane (inside-out) vesicles were step, along with the peripheral membrane pro-
prepared from Klebsiella and incubated with tein. How the decarboxylase translocates Na+
Na+ and oxaloacetate. (See note 49 for a to the cell surface is not understood.
description of how to prepare the vesicles.)
3. Use of the sodium current
As seen in Fig. 4.11, oxaloacetate-dependent
What does Klebsiella do with the energy con-
sodium ion influx took place. The uptake of
served in the sodium potential? It uses the energy
to actively transport the growth substrate, cit-
rate, into the cell, as well as to drive the reduction
of NAD+ by ubiquinol (i.e., via a Na+-dependent
NADH:ubiquinone oxidoreductase). The latter
is an example of reversed electron transport
driven by the influx of Na+. This all occurs dur-
ing the fermentation of citrate and is depicted
in Fig. 4.12.

4.8.2 Oxalate:formate exchange

can create a ∆ p
Oxalobacter formigenes, an anaerobic bacte-
rium that is part of the normal flora in mamma-
lian intestines, uses dietary oxalic acid as its sole
source of energy for growth. The organism has
evolved a method for creating a proton potential
Fig. 4.11 Sodium ion influx into vesicles driven by at the expense of the free energy released from
oxaloacetate decarboxylation. Inside-out vesicles the decarboxylation of oxalic acid to formic
were prepared from Klebsiella pneumoniae and acid and carbon dioxide.50–52 What is especially
incubated with 22Na+ in the presence (curve A) and interesting is that the enzyme is not in the mem-
absence (curve B) of oxaloacetate. Because of their brane and therefore cannot act as an ion pump,
large diameter, the vesicles could be isolated after one yet a ∆Ψ is created. The reaction catalyzed by
minute of incubation by using size exclusion chroma- oxalate decarboxylase is
tography. The fractions from the Sephadex columns
that contained vesicles were detected by absorbance –OOC–COO− + H+ → CO2 + HC, OO−
at 280 nm (protein). Only when oxaloacetate was
oxalate formate
present did the vesicle fraction contain 22Na+ (curve
A). This demonstrates the dependency of sodium For every mole of oxalate that enters the cell,
uptake upon oxaloacetate. In separate experiments one mole of formate leaves (Fig. 4.13). Since a
it was demonstrated that oxaloacetate decarboxy-
dicarboxylic acid crosses the cell membrane in
lase was present in the vesicles, that the oxaloac-
exchange for a monocarboxylic acid, there is
etate was decarboxylated, and that inhibition of
the oxaloacetate decarboxylase by avidin prevented net movement of negative charge toward the
uptake of 22Na+. Source: Adapted from Dimroth, inside, thus creating a ∆Ψ, inside negative. Also,
P. 1980. A new sodium-transport system energized a proton is consumed in the cytoplasm dur-
by the decarboxylation of oxaloacetate. FEBS Lett. ing the decarboxylation, which can contribute
122:234–236. to a ∆pH, inside alkaline. The antiporter was
Free ebooks ==>
membrane bioenergetics: the proton potential 131

citrate OAA pyruvate formate
acetate Na
acetyl~SCoA H+ + NADH
acetate acetyl~ P PTA

Fig. 4.12 Citrate fermentation and proposed Na+ ion currents in Klebsiella pneumoniae. (A) Na+–citrate
symporter. Citrate enters the cell via symport with Na+. Uptake of citrate is not dependent upon the ∆Ψ
and is therefore electroneutral and dependent upon the ∆pNa+. (It has been suggested that only two of the
carboxyl groups of citrate are neutralized with sodium ions, and the third is protonated.) The citrate is then
cleaved by citrate lyase (CL) to oxaloacetate and acetate. (B) Na+-pumping oxaloacetate decarboxylase. Na+
is pumped out of the cell during the decarboxylation of oxaloacetate to pyruvate and CO2, creating a sodium
ion electrochemical gradient that is used for citrate uptake (Na+–citrate symporter) and NADH production
(Na+-pumping NADH:ubiquinone oxidoreductase). The pyruvate is used to make ATP via substrate-level
phosphorylation. First the pyruvate is cleaved by pyruvate–formate lyase (PFL) to acetyl–CoA and formate.
Then the acetyl–CoA is converted to acetyl–phosphate by phosphotransacetylase (PTA). Finally, the acetyl–
phosphate donates the phosphoryl group to ADP to form ATP and acetate in a reaction catalyzed by acetate
kinase (AK). NADH is produced as described next. Some of the formate is oxidized to CO2 by a membrane-
bound formate dehydrogenase (FDH). The electron acceptor is ubiquinone. Ubiquinol is reoxidized by NAD+
to form NADH + H+ in a reaction catalyzed by a Na+-pumping NADH:ubiquinone oxidoreductase (C). This
is a case of reversed electron transport driven by the electrochemical Na+ gradient. There is therefore a Na+
circuit. Sodium ion is pumped out of the cell via the decarboxylase and enters via the citrate transporter and
the NADH:ubiquinone oxidoreductase. Source: Adapted from Dimroth, P. 1997. Primary sodium ion trans-
locating enzymes. Biochim. Biophys. Acta 1318:11–51.

purified in 1992 and shown to be a 38 kDa hydro- The decarboxylation of other acids
phobic polypeptide that catalyzes the exchange may also create a ∆ p
of oxalate and formate in reconstituted proteo- In principle, the influx and decarboxylation of
liposomes.53 (For a discussion of proteolipo- any dicarboxylic acid coupled to the efflux of the
somes, see note 54 and Section 17.1.) monocarboxylic acid can create a proton poten-
tial (Fig. 4.14). For example, a decarboxylation
ATP synthesis and electrogenic antiport was reported for the
Assuming that the stoichiometry of the ATP decarboxylation of malate to lactate acid dur-
synthase reaction is 3H+/ATP and for oxalate ing malolactate fermentation in Lactobacillus
decarboxylation it is 1H+/oxalate, then a steady lactis.55
proton current requires the decarboxylation of
three moles of oxalate per mole of ATP synthe- 4.8.3 End-product efflux as the
sized. In other words, one-third mole of ATP driving reaction
can be synthesized per mole of oxalate decar- Theoretically, it is possible to couple the excre-
boxylated. This is apparently the only means of tion of fermentation end products down a
ATP synthesis in Oxalobacter formigenes. concentration gradient to the translocation of
Free ebooks ==>
132 the physiology and biochemistry of prokaryotes

Fig. 4.13 The electrogenic oxalate:formate exchange and the synthesis by means of a proton current of ATP
in Oxalobacter formigenes. Oxalate2– enters via an oxalate2–/formate– antiporter. The oxalate2– is decarboxy-
lated to formate– and CO2, while a cytoplasmic proton is consumed. (The oxalate is first converted to oxalyl–
coenzyme A, which is decarboxylated to formyl–coenzyme A. The formyl–CoA transfers the coenzyme A to
incoming oxalate, thus forming formate and oxalyl–CoA.) An electrogenic exchange of oxalate2– for formate−
creates a membrane potential, outside positive. It is suggested that the stoichiometry of the ATP synthase is
3H+/ATP. Since the decarboxylation of one mole of oxalate results in the consumption of one mole of protons,
the incoming current of protons during the synthesis of one mole of ATP is balanced by the decarboxylation
of three moles of oxalate. In other words, a steady state current of protons requires that of a mole of ATP be
made per mole of oxalate decarboxylated. Source: Anantharam, V., M. J. Allison, and P. C. Maloney. 1989.
Oxalate:formate exchange. J. Biol. Chem. 264:7244–725. Also, see refs. 45 and 46.

protons out of the cell, thereby creating a ∆p. gradient; or the efflux of a solute down a con-
This scheme, the reverse of solute uptake by centration gradient can create a ∆p. The driving
proton-coupled transport systems, is called force for solute transport in symport with pro-
the energy recycling model.56,57 In other words, tons is the sum of the proton potential and the
the direction of solute transport depends upon electrochemical potential of the solute, that is,
which is greater: the energy from the proton
potential, ∆p, which drives solute uptake, cre- Driving force (mV) = y∆p + ∆μs/F (4.22)
ating an electrochemical solute gradient ∆μ/F, where y is the number of protons in symport
or the electrochemical solute gradient, which with the solute, S. The term ∆μs/F represents the
drives solute efflux, creating a ∆p. Solute efflux electrochemical potential of S:
coupled to proton translocation can spare ATP
because in fermenting bacteria, the hydrolysis ∆μs/F = m∆Ψ + 60 log[S]in/[S]out millivolts
of ATP (catalyzed by the ATP synthase) is used (4.23)
to pump protons to the outside to create the ∆p. where m is the charge of the solute. Therefore,
As the ∆p rises, ATP hydrolysis is diminished. the overall driving force (substituting for ∆μs/F
and ∆p) is
Consider solute/proton symport to be revers- Driving force (mV)
ible. Under these circumstances, the ∆p can drive = 60 log[S]in/[S]out + (y + m)∆Ψ − y60 ∆pH
the uptake of a solute against a concentration (4.24)
Free ebooks ==>
membrane bioenergetics: the proton potential 133

_ _ _
O2C R CO 2 O2C R CO 2
H+ A
_ _ CO2

_ _
H+ B

ATP + Pi
3H+ C

Fig. 4.14 Scheme for the coupling of the decarboxylation of a dicarboxylic acid to the development of a pro-
ton potential: dicarboxylic acid enters as a divalent anion (A) or a monovalent anion (B). In the cytoplasm,
a decarboxylase cleaves off CO2, consuming a proton in the process and producing a monocarboxylic acid
with one negative charge fewer than the original dicarboxylic acid. Exchange of the dicarboxylic acid and the
monocarboxylic acid via the antiporter is electrogenic and produces a membrane potential, inside negative.
The consumption of the proton during the decarboxylation creates a ∆pH. (C) Influx of protons via the ATP
synthase completes the proton circuit and results in ATP synthesis. The energetics can be understood in terms
of the decarboxylase removing the dicarboxylic acid from the inside, thus maintaining a concentration gradi-
ent that stimulates influx.

Upon substituting m = –1 for lactate, eq. 4.24 Symport of protons and sodium ions with
becomes fermentation end products
Driving force Coupled translocation of protons and lactate
= (y − 1)∆Ψ − y60∆pH + 60 log[S]in/[S]out can be demonstrated in membrane vesicles pre-
pared from lactic acid bacteria belonging to the
genus Streptococcus.58–61 Similarly, coupled
During growth, lactate transport is near equil- translocation of sodium ions and succinate is
brium and the net driving force is close to zero. catalyzed by vesicles prepared from a rumen bac-
Thus by setting eq. 4.25 equal to zero, one can terium belonging to the genus Selenomonas.62 In
solve for y: both cases a transporter exists in the membrane
that simultaneously translocates the organic
∆Ψ − 60 log([S]in/[S]out) acid (R−) with either protons or sodium ions out
y= (4.26)
∆p of the cell (symport) (Fig. 4.15). (See note 63.)
Lactate efflux will be described next and is sum-
Note that eq. 4.25 states that when y, which marized in Fig. 4.16. Succinate efflux is similar
is the number of moles of H+ translocated per and is summarized in Fig. 4.17.
mole of lactate, is one, then the translocation is
electroneutral and a ∆Ψ does not develop; only
a small ∆pH develops, owing to the acidifica- Lactate efflux
tion of the external medium by the lactic acid. L-Lactate-loaded membrane vesicles were pre-
We will return to these points later when we pared from Streptococcus cremoris in the fol-
discuss the physiological significance of energy lowing way. A concentrated suspension of cells
conservation via end-product efflux. Let us now was treated with lysozyme in buffer to degrade
consider some data that support the hypothesis the cell walls. The resulting cell suspension was
that lactate/H+ symport and succinate/Na+ sym- gently lysed (broken) by adding potassium sul-
port can create a membrane potential in mem- fate. The cell membranes spontaneously resealed
brane vesicles derived from cells. into empty vesicles. The vesicles were purified
Free ebooks ==>
134 the physiology and biochemistry of prokaryotes

R nNa+



Fig. 4.15 End-product efflux in symport with protons Fig. 4.16 Lactate efflux can produce a membrane
or sodium ions. The high intracellular concentrations potential. Lactate-loaded membrane vesicles from
of R− may drive the efflux of Na+ or H+ via symport- Streptococcus cremoris were incubated with the
ers. If the ratio of protons or sodium ions to carboxyl lipophilic cation Ph4P+ without lactate (curve 1)
(i.e., n/carboxyl) exceeds 1, then the symport is elec- and with 50 mM lactate in the external medium
trogenic and a membrane potential develops. (curve 2). In the absence of external lactate, the lipo-
philic probe accumulated inside the vesicles, suggest-
ing that a membrane potential developed as lactate
and then incubated for 1 h with 50 mM L-lac- left the cells. The addition of lactate to the external
tate, which equilibrated across the membrane, medium prevented the formation of the membrane
potential because the lactate concentration (in/out)
thus loading the vesicles with L-lactate. Then
was lowered. Additional experiments showed that
the membrane vesicles loaded with L-lactate the electrogenic ion was the proton. Source: Adapted
were incubated with the lipophilic cation tetra- from Otto, R., R. G. Lageveen, H. Veldkamp, and
phenylphosphonium (Ph4P+) in solutions con- W. N. Konings. 1982. Lactate efflux-induced electri-
taining buffer (Fig. 4.16, curve 1) and buffer + cal potential in membrane vesicles of Streptococcus
50 mM L-lactate (Fig. 4.16, curve 2). Samples cremoris. J. Bacteriol. 149:733–738.
were filtered to separate the vesicles from the
external medium, and the amount of Ph4P+ that
accumulated was measured. The Ph4P+ accumu- developed during succinate efflux (Fig. 4.17,
lated inside the cells when the ratio of lactatein to curve 1).
lactateout was high. When the external buffer contained a high
The accumulation of Ph4P+ implies that the concentration of succinate, the membrane
efflux of lactate along its concentration gradient potential did not develop, indicating that the
imposed a membrane potential on the vesicles. energy to establish the potential was derived
The membrane potential was calculated by from the efflux of succinate along its concen-
using the Nernst equation and the Ph4P+ accu- tration gradient, [succ]in/[succ]out (Fig. 4.17,
mulation ratio as explained in Section 4.5.1. curve 2). The molar growth yields of S. rumi-
nantium are higher when succinate production
Succinate efflux is at a maximum, implying that more ATP is
An experiment similar to the one just described available for biosynthesis as a result of the suc-
for lactate efflux was done with vesicles from the cinate efflux.
rumen bacterium Selenomonas ruminantium
loaded with succinate (Fig. 4.17). The Ph4P+ Physiological significance of end-product
was taken up by the vesicles when there was a efflux as a source of cellular energy
concentration gradient of succinate, indicat- To demonstrate the generation of a membrane
ing that a membrane potential, inside negative, potential due to end-product efflux, cells or
Free ebooks ==>
membrane bioenergetics: the proton potential 135

Brink et al., who used experimentally measured

values of lactate concentrations, membrane
potential, and ∆pH.64 Only when the external
lactate concentrations were very low was lac-
tate/proton efflux electrogenic (i.e., y > 1). For
example, when the cells were grown at pH 6.34
and the lactate accumulated in the medium over
time, the value of y decreased from about 1.44
to 0.9 while the external lactate concentrations
increased from 8 mM to 38 mM. This means
that only under certain growth conditions (low
external lactate and external pH high so that the
∆pH is 0 or inverted) would one expect lactate
efflux effectively to generate a ∆p. This may
occur in the natural habitat, where growth of
the producer may be stimulated by a population
of bacteria that utilize lactate, thus keeping the
Fig. 4.17 Succinate efflux can produce a membrane external concentrations low.
potential. Succinate-loaded membrane vesicles from
Selenomonas ruminantium were incubated with the
lipophilic cation Ph4P+ without succinate (curve 1) or 4.8.4 Light absorbed by
with succinate (curve 2) in the external medium. The bacteriorhodopsin can drive the
uptake of Ph4P+ indicates that a membrane potential creation of a ∆ p
developed when the ratio of succinatein to succina-
Certain archaea, namely, the extremely halo-
teout was high. Source: Adapted from Michel, T. A.,
and J. M. Macy. 1990. Generation of a membrane
philic archaea, have evolved a way to produce
potential by sodium-dependent succinate efflux a ∆p by using light energy directly (i.e., without
in Selenomonas ruminantium. J. Bacteriol. 172: the intervention of oxidation–reduction reac-
1430–1435. tions and without chlorophyll).65–67 See note 68
for a more complete discussion of the extreme
halophiles. (It has been reported that under the
membrane vesicles must be “loaded” with high proper conditions, these microorganisms can be
concentrations of the end product by incubat- grown photoheterotrophically, i.e., on organic
ing de-energized cells or membrane vesicles for carbon with light as a source of energy, but such
several hours before dilution into end-product- conclusions have been questioned.69,70)
free media. These are not physiological condi- Halophilic archaea are heterotrophic
tions, and the physiological relevance of the organisms that carry out an ordinary aerobic
conditions under which the electrogenic extru- respiration, creating a ∆p driven by oxidation–
sion of lactate coupled to protons has been reduction reactions during electron transport.
questioned. (See note 71.) The ∆p is used to drive ATP
Assuming a given value of y, one can use synthesis via a membrane ATP synthase. (See
eq. 4.26 to calculate the necessary lactate Sections 4.7.1 and 4.6.2.) However, conditions
concentration gradient from experimentally for respiration are not always optimal and, in
determined values of ∆Ψ and ∆pH. When one the presence of light and low oxygen levels, the
substitutes y = 2 (for electrogenic lactate extru- halophiles adapt by making photopigments
sion), a major problem appears. To substitute a (rhodopsins), one of which (bacteriorhodopsin)
measured external lactate concentration of 30 functions as a proton pump that is energized
mM, a ∆Ψ of –100 mV, a –60 ∆pH of –30 mV directly by light energy. (See note 72.) Whereas
(at pH 7.0), and y = 2, the calculated internal photosynthetic electron flow is an example of
lactate concentration would have to be 14 M an indirect transformation of light energy into
for lactate secretion to occur.64 This is a non- an electrochemical potential (via redox reac-
physiological concentration. (Internal lactate tions), bacteriorhodopsin illustrates the direct
concentrations reach concentrations of about transformation of light energy into an electro-
0.2 M.) In fact, the values of y were calculated by chemical potential. We will first consider data
Free ebooks ==>
136 the physiology and biochemistry of prokaryotes

that demonstrate the light-dependent pumping

of protons. We will then describe the proton
pump, the photocycle, and a model for the mech-
anism of pumping protons. Bacteriorhodopsin
is examined in detail because it is the best char-
acterized ion pump.

Evidence that halophiles can use light energy

to drive a proton pump
Figure 4.18 models the results of subjecting a
suspension of Halobacterium halobium to light
of different intensities and for different periods
of time, during which the pH of the external
medium was monitored. As illustrated in Fig.
4.18A, the light produced an efflux of protons
from the cell. Higher light intensities produced
a greater rate of proton efflux (compare Ic to
Ia.) In Fig. 4.18B, the rate of proton efflux, in
nanograms per second, as determined from
the slopes in Fig. 4.18A, is plotted as a func-
tion of the light intensity in nanoeinsteins per Fig. 4.18 Proton pumping by bacteriorhodopsin.
second (an einstein is equal to a “mole” of pho- Expected results if Halobacterium cells were illu-
tons). From data such as these, a quantum yield minated by light. To measure proton outflow accu-
(i.e., protons ejected per photon absorbed) rately, proton inflow through the ATP synthase must
be blocked either with uncouplers or nigericin, which
was calculated to be 0.52 proton per photon
collapse the proton electrochemical potential, or
absorbed.73 The reported values for the maxi-
with an ATP synthase inhibitor such as DCCD. The
mum quantum yield for proton efflux is a little extruded protons can be quantitated with a pH meter.
higher (0.6–0.7 proton/photon absorbed). The (A) Proton efflux measured as a function of time at
quantum yield for the photocycle (i.e., the frac- increasing light intensities Ia < Ib < Ic, expressed as
tion of bacteriorhodopsin molecules absorbing nanoeinsteins per second. The slope of each line is the
light that undergoes the photocycle described rate of proton efflux. (B) The rate of proton efflux is
in the next subsection) is 0.64 ± 0.04.74,75 These plotted as a function of the light intensity. From these
values suggest that one proton is pumped per data one can calculate the quantum yield (i.e., the
photocycle. number of protons extruded per photon absorbed).
Source: Adapted from data by Bogomolni, R. A.,
R. A. Baker, R. H. Lozier, and W. Stoeckenius. 1980.
Bacteriorhodopsin is the proton pump Action spectrum and quantum efficiency for proton
Built into the cell membrane of the halophilic pumping in Halobacterium halobium. Biochemistry
archaebacteria is a pigment protein called bac- 19:2152–2159.
teriorhodopsin, which is a pump responsible
for the light-driven electrogenic efflux of pro-
tons. It consists of one large polypeptide (248 The photocycle and a model for
amino acids, 26,486 Da) folded into seven α proton pumping
helices that form a transmembrane channel The photoevents occurring in halophiles can
(Fig. 4.19). (See Ref. 76 for a review.) Located be followed spectroscopically because when
in the middle of the channel, and attached to bacteriorhodopsin (bR) absorbs light, it loses
the bacteriorhodopsin, is a pigment called reti- its absorption peak at 568 nm (bleaches) and is
nal (a C20 carotenoid), which is attached via a converted in the dark to a series of pigments that
Schiff base to a lysine residue on the protein. have absorption peaks at different wavelengths.
(Note 77 tells what a Schiff base is.) When the This sequence of events is called the photocycle.
retinal absorbs light, the bacteriorhodopsin Also, site-specific mutagenesis of bacteriorho-
remarkably translocates protons out of the cell, dopsin is being used to identify the amino acid
and a ∆p is created. side chains that transfer protons across the
Free ebooks ==>
membrane bioenergetics: the proton potential 137

N. Aspartate-96 acquires a proton from the

cytoplasm in going from N to O. Thus, a proton
has moved from the cytoplasmic side through
aspartate-96 to the Schiff base to aspartate-85.
From aspartate-85, the proton moves to the
outside membrane surface. (See note 82.)
Precisely how the proton travels through the
bacteriorhodopsin channel from the cytoplas-
mic side to the Schiff base in the center of the
channel, and from there to the external sur-
face of the membrane, is not known. Probably
the proton is passed from one amino acid
side group that can be reversibly protonated
to another. Site-specific mutagenesis experi-
ments have suggested that two of these amino
acids are aspartate-85 and aspartate-96. The
protonatable residues would extend along the
protein from the cytoplasmic surface to the out-
side. In the case of bacteriorhodopsin, where
the three-dimensional structure is known, the
protonatable residues are oriented toward the
center of the channel. Any bound water might
also participate in proton translocation. For
example, there might be a chain of water and
Fig. 4.19 Diagram of bacteriorhodopsin. The seven
hydrogen-bonded protons connecting protona-
helices, shown as rods, form a central channel; the
retinal is attached to a lysine residue on helix G.
table groups. Since the Schiff base gives up its
Source: Henderson, R., J. M. Baldwin, T. A. Ceska, proton to the extracellular side of the channel
F. Zemlin, E. Beckmann, and K. H. Downing. 1990. (transition L to M) and becomes protonated
Model for the structure of bacteriorhodopsin based with a proton from the cytoplasm (transition
on high-resolution electron cryomicroscopy. J. Mol. M to N), it has been assumed that there is a
Biol. 213:899–929. switch that reorients the Schiff base, causing it
to face the cytoplasmic and extracellular sides
of the channel alternately. Logically, the switch
membrane.78–81 The photocycle is shown in Fig. would be at M. The mechanism for the switch
4.20. The retinal is attached via a Schiff base to is unknown, although a conformational change
the ε amino group of lysine-216 (K216). Before in the bacteriorhodopsin has been suggested.80
absorbing light, the retinal is protonated at the It is not obviously correlated with the retinal
Schiff base and exists in the all-trans (13-trans) isomerizations because the retinal is in the cis
configuration (bR568). The subscript refers configuration throughout most of the photocy-
to the absorption maximum, in nanometers. cle, including the protonation and deprotona-
Upon absorbing a photon of light, the retinal tion of the Schiff base. The transfer of protons
isomerizes to the 13-cis form (K625). All subse- along protonatable groups has been suggested
quent steps do not require light and represent for several proton pumps besides bacteriorho-
the de-energization of the bacteriorhodopsin dopsin, including cytochrome oxidase and the
via a series of intermediates that have absor- proton-translocating ATP synthase.83
bance maxima different from those of the origi-
nal unexcited molecule. The transitions are
very fast and occur in the nanosecond and mil- 4.9 Halorhodopsin, a Light-Driven
lisecond ranges. These intermediates are, in the Chloride Pump
order of their appearance, K, L, M, N, and O. The halophiles have a second light-driven
In converting from L to M, the Schiff base electrogenic ion pump, but one that does not
loses its proton to aspartate-85 but regains a energize the membrane. The second pump,
proton from aspartate-96 in going from M to called halorhodopsin, is structurally similar
Free ebooks ==>
138 the physiology and biochemistry of prokaryotes

Fig. 4.20 The photochemical cycle of bacteriorhodopsin. Upon absorption of a photon of light, bR568 under-
goes a trans-to-cis isomerization and is converted to a series of intermediates with different absorbance max-
ima. The Schiff base becomes deprotonated during the L-to-M transition and reprotonated during the M-to-N
transition. A recent photocycle postulates two M states: L to M1 to M2 to N. Although not indicated, some of
the steps are reversible. (See Lanyi, J. K. 1992. Proton transfer and energy coupling in the bacteriorhodpsin
photo-cycle. J. Bioenerg. Biomemb. 24:169–179.) Source: Krebs, M. P., and H. Gobind Khorana. 1993.
Mechanism of light-dependent proton translocation by bacteriorhodopsin. J. Bacteriol. 175:1555–1560.
Reproduced with permission from American Society for Microbiology.

to bacteriorhodopsin and is used to accumu- Bacillus pasteurii, B. firmus, and B. alcalophi-

late Cl– intracellularly, to maintain osmotic lus. The ∆p has been measured in obligate aero-
stability.59,73,84 Recall that the halophiles live bic alkaliphiles that grow optimally at pH 10
in salt water, where the extracellular concen- to 12, and it appears to be too low to drive the
trations of NaCl can be 3 to 5 M. The osmotic synthesis of ATP. The problem is that because
balance is preserved by intracellular concen- the external pH is so basic, the internal pH is
trations of KCl that match the extracellular generally at least 2 pH units more acid. This
Cl− concentrations. (As discussed in Chapter 16, gives the ∆pH a sign opposite to that found in
K+ is important for osmotic homeostasis in the other bacteria, that is, negative. A ∆pH of 2
eubacteria as well.) Since the membrane is nega- is equivalent to 60 × 2 or 120 mV. Thus, the ∆p
tively charged on the inside, energy must be used can be lowered by about 120 mV in these organ-
to bring the Cl− into the cell, and halorhodop- isms. Typical membrane potentials for aerobic
sin accomplishes this purpose. In the dark the alkaliphiles are approximately –70 mV (posi-
halophiles use another energy source, probably tive out). Therefore, the ∆p values can be as low
ATP, to accumulate chloride ions actively.85 as (–170 + 120) or –50 mV.87
The ATPase in the alkaliphiles is a proton-
translocating enzyme, as in most bacteria. Even
4.10 The ∆p and ATP Synthesis if the ATPase translocated as many as four pro-
in Alkaliphiles tons, this would generate only 0.2 eV, far short
Alkaliphilic bacteria grow in habitats having of the approximately 0.4 to 0.5 eV required to
a very basic pH, usually around pH 10. These synthesize an ATP under physiological condi-
habitats include soda lakes, dilute alkaline tions. The energy to synthesize an ATP (i.e.,
springs, and desert soils, where the alkalinity ∆Gp) is 40,000 to 50,000 J. Dividing this num-
is usually due to sodium carbonate. Obligate ber by the Faraday constant gives 0.4 to 0.5 eV.
alkaliphiles are organisms that cannot grow How can this dilemma be resolved? It is pos-
at pH values of 8.5 or less and usually have an sible that the protons in the bulk extracellu-
optimum around 9. (See note 86.) These include lar phase are not as important for the ∆p as
Free ebooks ==>
membrane bioenergetics: the proton potential 139

protons on the membrane or a few angstrom purpose in most bacteria. Thus, although the
units away from the membrane. One sug- major ion circuit is a proton circuit, the sodium
gestion is that random collisons within the circuit is also important.
membrane may put proton pumps in frequent A ∆p is created when an exergonic chemical
contact with the ATPases, and as soon as a reaction is coupled to the electrogenic flow of
proton is pumped out of the cell, it may reen- charge across the cell membrane and the libera-
ter via an adjoining ATPase without entering tion of protons on the outer membrane surface.
the pool of bulk protons.88 This suggestion Energy input of at least yF∆p joules is neces-
emphasizes the activity of protons at the face sary to raise the electrochemical potential of y
of the membrane rather than the ∆pH, which is moles of protons to ∆p volts. The three most
due to the concentration of protons in the bulk widespread reactions that provide the energy to
phase, and raises questions about the details of create the ∆p are oxidation–reduction reactions
the proton circuit. during electron transport in membranes (respi-
One important tenet of the chemiosmotic ration), oxidation–reduction reactions during
theory may not apply in this situation, and that electron transport stimulated by light absorp-
is that a delocalized ∆p is used. That is, the che- tion (photosynthesis), and ATP hydrolysis via
miosmotic theory postulates that proton cur- the membrane ATP synthase. Respiration and
rents couple any exergonic reaction with any ATP hydrolysis are reversible, and the ∆p can
endergonic reaction (i.e., proton circuits are drive reversed electron transport as well as the
delocalized over the entire membrane). In that synthesis of ATP. During reversed electron
way, proton extrusion during respiration can transport, protons enter the cells rather than
provide the energy for several different reac- leave the cells. The ATP synthase is an enzyme
tions, not simply the ATP synthase (Fig. 4.2). complex that reversibly hydrolyzes ATP and
However, it may be that in some alkaliphiles pumps protons out of the cell. When protons
there is direct transfer of protons extruded dur- enter via the ATP synthase, ATP is made.
ing respiration to the ATP synthase (i.e., local- Light energy can also be used directly to
ized proton circuits). create a ∆p without the establishment of a
redox potential. This occurs in the halophilic
archaea, which use a light-driven proton
4.11 Summary pump called bacteriorhodopsin to create a ∆p.
The energetics of bacterial cell membranes can Bacteriorhodopsin, which forms a proton chan-
be understood for most bacteria in terms of an nel through the membrane, is being studied as a
electrochemical proton potential established model system to investigate the mechanism of
by exergonic chemical reactions or light. The ion pumping across membranes.
protons are raised from a low electrochemical Fermenting bacteria have evolved additional
potential inside the cell to a high electrochemi- ways to generate a ∆p. Lactic acid bacteria can
cal potential outside the cell. When the protons create a ∆p via coupled efflux of protons and
circulate back into the cell through appropriate lactate (in addition to ATP hydrolysis) under
carriers, work can be done (e.g., the synthesis certain growth conditions. Another anaerobic
of ATP via the membrane ATP synthase, solute bacterium, Oxalobacter, creates a ∆p by the
transport, flagellar rotation). oxidation of oxalic acid to formic acid coupled
The proton potential is due to a combination with the electrogenic exchange of oxalate for
of a membrane potential (∆Ψ), outside positive, formate and the consumption of protons during
and a ∆pH, outside acid. The membrane poten- the decarboxylation. Other examples similar to
tial seems to be the dominant component in the these will no doubt be discovered in the future.
∆p for most bacteria, except for acidophiles, Sodium potentials are also important in the
which can have a reversed membrane potential. prokaryotes, especially for solute transport.
Other cations, especially sodium ions, can use Although most sodium potentials are created
the established membrane potential for doing secondarily from the proton potential via anti-
work, principally solute accumulation. The porters, there are some prokaryotes that couple
sodium ions must be returned to the outside a chemical reaction to the creation of a sodium
of the cell, and Na+/H+ antiporters serve this potential. For example, some marine bacteria,
Free ebooks ==>
140 the physiology and biochemistry of prokaryotes

as exemplified by Vibrio alginolyticus, couple Study Questions

respiration to the electrogenic translocation
of sodium ions out of the cell at alkaline pH. 1. The E′0 for ubiquinone (ox)/ubiquinone
Because these bacteria can couple the sodium (red) is +100 mV, and for NAD+/NADH it
potential to the membrane ATP synthase, they is –320 mV. What is the ∆E0?
rely on a sodium current rather than a proton
ans. 420 mV
current when growing in basic solutions. When
growing in slightly acidic conditions, they use a 2. In the electron transport chain, oxidation–
proton potential. reduction reactions with a ∆Eh of about
Several different fermenting bacteria can cre- 200 mV appear to be coupled to the extru-
ate a sodium potential by coupling the decarbox- sion of protons. Assume that 100% of the
ylation of organic acids with the translocation oxidation–reduction energy is converted
of sodium ions out of the cell, or by coupling the to the ∆p. For a two-electron transfer and
efflux of end products of fermentation with the the extrusion of two protons, what is the
translocation of sodium ions to the outside. In expected ∆p?
some bacteria (e.g., Klebsiella pneumoniae), the
ans. –200 mV
sodium potential is used to drive the influx of
the growth substrate into the cell but not for the 3. What is the maximum ∆p (i.e., 100%
generation of ATP. Rather, ATP is synthesized energy conversion) when the hydrolysis of
by a substrate-level phosphorylation during the one mole of ATP is coupled to the extru-
conversion of pyruvate to formate and acetate. sion of four moles of protons? Three moles
These bacteria may also generate a membrane of protons? Assume that the free energy of
potential by coupled efflux of the end products hydrolysis of ATP is –50,000 J.
of citrate degradation (formate and acetate)
ans. –130 mV, –173 mV
with protons.
It is emphasized that in fermenting bacteria, 4. Design an experimental approach that
ATP is hydrolyzed via the ATP synthase to cre- can show that the efflux of an organic acid
ate the ∆p that is necessary for membrane activ- along its concentration gradient is coupled
ities (e.g., solute transport, flagellar rotation). to proton translocation and can generate a
A decrease in the ∆p should result in even more membrane potential. (You must be able not
ATP hydrolysis. Thus, reactions are expected only to demonstrate the membrane poten-
to conserve ATP if they create a membrane tial but also to show that the proton is the
potential (e.g., electrogenic efflux of sodium conducting charge.)
ions or protons in symport with end products
5. Explain how the decarboxylation of oxalic
of fermentation, or during decarboxylation
acid by Oxalobacter creates a ∆pH and a
Measurements of the ∆Ψ and the ∆pH are
necessarily indirect because of the small size 6. What is the reason for stating that light cre-
of the bacteria. The ∆Ψ is measured by using ates a ∆p indirectly in photosynthesis but
cationic or anionic fluorescent dyes that equili- directly in the extreme halophiles?
brate across the membrane in response to the
7. Lactate efflux was in symport with pro-
potential. The distribution of the dyes is moni-
tons, whereas succinate efflux was in sym-
tored by fluorescence quenching. A second way
port with sodium ions. Which ionophores
to measure the membrane potential is by the
might you use to distinguish the cations
equilibration of a permeant ion that achieves
electrochemical equilibrium with the membrane
potential. The membrane potential is computed 8. A reasonable figure for the actual free
by using the Nernst equation and the intracel- energy of hydrolysis of ATP inside cells is
lular and extracellular ion concentrations. The –50,000 J/mol. It is believed that the hydro-
∆pH is measured by using a weak acid or weak lysis of ATP is coupled to the extrusion of
base whose log ratio of concentrations inside the three protons in many systems. If a ∆p of
cell to outside the cell is a function of the ∆pH. –150 mV were generated, what would be
Free ebooks ==>
membrane bioenergetics: the proton potential 141

the efficiency of utilization of ATP energy in millivolts, at 30 °C? What is the common
to create the ∆p? expression used for the force due to the con-
centration gradient when S is a proton?
ans. 87%
18. Assume that an ion traverses a charged
9. Assume a reduction potential of +400 mV
membrane with a potential of ∆Ψ volts.
for an oxidant and a potential of –100 mV
Further assume that there is no concentra-
for a reductant. How many joules of energy
tion gradient. What is the expression that
will be released when two moles of electrons
denotes the driving force? What will the
flow from the reductant to the oxidant?
sign be if the ion moves toward the side of
ans. 96,500 J opposite charge?
10. Assume that 45 kJ is required to synthe- 19. What is the expression, in millivolts at 30
size one mole of ATP. What would be the °C, for the ∆p? What is the expression in
required ∆p, assuming that three moles of joules?
H+ entered via the ATPase per mole of ATP
ans. –155 mV
11. Assume that the ∆p is –225 mV. If the 1. Harold, F. M. 1986. The Vital Force: A Study of
∆pH at 30 °C is 1.0, what is the membrane Bioenergetics. W. H. Freeman, New York.
potential? 2. Nicholls, D. G., and S. J. Ferguson. 2002.
Bioenergetics 3. Academic Press, San Diego, CA.
ans. –165 mV
3. Mitchell, P. 1961. Coupling of phosphorylation
12. How much energy in joules is required to to electron and hydrogen transfer by a chemiosmotic
move a mole of uncharged solute into the cell type of mechanism. Nature 191:144–148.
against a concentration gradient of 1,000 at 4. Mitchell, P. 1966. Chemiosmotic coupling in oxi-
30 °C? If transport were driven by the ∆p, dative and photosynthetic phosphorylation. Biol.
what would be the minimal value of the ∆p Rev. Cambridge Philos. Soc. 41:445–502.
required if one H+ were cotransported? 5. Mitchell, P. 1977. Vectorial chemiosmotic pro-
cesses. Annu. Rev. Biochem. 46:996–1005.
ans. 17,370 J, 180 mV
6. Mitchell, P. 1979. Compartmentation and com-
13. Assume membrane vesicles loaded with K+ munication in living systems. Ligand conduction: a
so that Kin+ / K+out = 1,000. The temperature is general catalytic principle in chemical, osmotic and
chemiosmotic reaction systems. Eur. J. Biochem.
30 °C, and valinomycin is added. What is
the predicted initial membrane potential in
millivolts? 7. Kashket, E. R. 1985. The proton motive force
in bacteria: a critical assessment of methods. Annu.
ans. –180 mV Rev. Microbiol. 39:219–242.

14. What is the rationale for adding valino- 8. A single proton (or any monovalent ion, or elec-
tron) carries 1.6 × 10–19 C of charge. If we multiply
mycin and K+ to an experimental system in this by Avogadro’s number, we arrive at the charge
which the number of protons being pumped carried by a mole, which is approximately 96,500
out of the cell is measured? C/mol, or the Faraday constant (F).
15. Briefly, what does the chemiosmotic theory 9. The actual equation is G = G0 + RT ln a, where a
is activity. The activity is a product of the molal con-
centration (c) and the activity coefficient (γ) for the
16. What is meant by the ∆Eh? What is the rela- particular compound, a = γc. In practice, concentra-
tionship between the ∆Eh and the ∆p that is tions are usually used instead of activities, and the
concentrations are in molar units instead of molal.
generated at coupling sites? The symbol T is the absolute temperature in degrees
17. Assume that a concentration gradient of an kelvin (273 + °C). The symbol R is the ideal gas con-
stant, and G0 is the standard free energy (i.e., when
uncharged solute, S, exists across the cell mem- the concentration of all reactants is 1 M). When one
brane. What is the formula that expresses the uses 8.3144 J K–1 mol–1 as the units of R, then the free
driving force in the concentration gradient, energy (G) is given in J/mol.
Free ebooks ==>
142 the physiology and biochemistry of prokaryotes

10. Cecchini, G., and A. L. Koch. 1975. Effect of 22. Ballmoos, C. von, G. M. Cook, and P. Dimroth.
uncouplers on “downhill” β-galactoside transport in 2008. Unique rotary ATP synthase and its biological
energy-depleted cells of Escherichia coli. J. Bacteriol. diversity. Annu. Rev. Biophys. 37:43–64.
23. Cross, R. L. 1992. The reaction mechanism of
11. Gould, J. M., and W. A. Cramer. 1977. F0F1-ATP synthases, pp. 317–330. In: Molecular
Relationship between oxygen-induced proton efflux Mechanisms in Bioenergetics. L. Ernster (Ed.).
and membrane energization in cells of Escherichia Elsevier Science Publishers, Amsterdam.
coli. J. Biol. Chem. 252:5875–5882.
24. Another way of stating this is in terms of the total
12. E. Padan, D. Zilberstein, and S. Schuldiner. force. The total force is equal to y∆p +∆Gp/F, where
1981. pH homeostasis in bacteria. Biochim. Biophys. ∆Gp/F = 518 mV. At equilibrium the total force is 0;
Acta 650:151–166. therefore y∆p = –518 mV, and when y = 3, ∆p = –173
13. Reviewed in: Cobley, J. G., and J. C. Cox.
1983. Energy conservation in acidophilic bacteria. 25. Kandpal, R. P., K. E. Stempel, and P. B. Boyer.
Microbiol. Rev. 47:579–595. 1987. Characteristics of the formation of enzyme-
bound ATP from medium inorganic phosphate
14. The pumping of protons out of the cell or the
by mitochondrial F1 adenosine triphosphatase in
electrogenic influx of electrons will create a mem-
the presence of dimethyl sulfoxide. Biochemistry
brane potential, positive outside. However, in the
aerobic acidophilic bacteria [i.e., bacteria that live
in environments of extremely low pH (pH 1–4)], 26. Grubmeyer, C., R. L. Cross, and H. S. Penefsky.
other events act to reverse the membrane potential. 1982. Mechanism of ATP hydrolysis by beef heart
These bacteria have positive membrane potentials mitochondrial ATPase: rate constants for elemen-
(i.e., inside positive with respect to outside, at low tary steps in catalysis at a single site. J. Biol. Chem.
pH). It is not clear why the aerobic acidophiles have 257:12092–12100.
a positive ∆Ψ. One possibility is that they have an
27. Vonck, J., T. K. von Nidda, T. Meier, U. Matthey,
energy-dependent K+ pump that brings K+ into the
D. J. Mills, W. Kuhlbrandt, and P. Dimroth. 2002.
cells at a rate sufficient to establish a net influx of
Molecular architecture of the undecameric rotor
positive charge, creating an inside positive mem-
of a bacterial Na+-ATP synthase. J. Mol. Biol. 321:
brane potential. This point is discussed further in
Section 17.1.3.
28. Kaim, G., M. Prummer, B. Sick, G. Zumofen, A.
15. Krulwich, T. A., and A. A. Guffanti. 1986.
Renn, U. P. Wild, and P. Dimroth. 2002. Coupled
Regulation of internal pH in acidophilic and alka-
rotation within single F0F1 enzyme complexes dur-
lophilic bacteria, pp. 352–365. In: Methods in
ing ATP synthesis or hydrolysis. FEBS Lett. 525:
Enzymology, Vol. 125. S. Fleischer and B. Fleischer
(Eds.). Academic Press, New York.
29. Capaldi, R. A., and R. Aggeler. 2002. Mechanism
16. Actually, what happens in the presence of nigeri-
of the F1F0 ATP synthase, a biological rotary motor.
cin is that an equalization of the K+ and H+ gradients
Trends Biochem. Sci. 27:154–160.
30. Fillingame, R. H., and O. Y. Dmitriev. 2002.
17. Padan, E., D. Zilberstein, and S. Schuldiner.
Structural model of the transmembrane F0 rotary
1981. pH homeostasis in bacteria. Biochim. Biophys.
sector of H+-transporting ATP synthase derived by
Acta 650:151–166.
solution NMR and intersubunit cross-linking in situ.
18. Rottenberg, H. 1979. The measurement of Biochim. Biophys. Acta 1565:232–245.
membrane potential and ∆pH in cells, organelles,
31. The ∆E0 can be a function of the pH. This is
and vesicles. Methods Enzymol. 55:547–569.
the case for the following reaction, where m is not
19. Bakker, E. P. 1990. The role of alkali–cation zero: in some reactions n = 1 and m = 0 (cytochrome
transport in energy coupling of neutrophilic and aci- redox reactions), n = 2 and m = 1 (NAD+ or NADP+
dophilic bacteria: an assessment of methods and con- redox reactions), n = 2 and m = 2 (fumarate–succi-
cepts. FEMS Microbiol. Rev. 75:319–334. nate redox reactions).
20. This is because the resonance frequency of inor- In reactions involving protons, if the pH is not
ganic phosphate or of the γ-phosphate of ATP in zero, then the ∆E0 is more negative. When m = n, the
a high magnetic field is a function of the degree to ∆E0 is –60 mV/pH. When m = 1 and n = 2, the ∆E0
which the phosphate is protonated. (See: Ferguson, S. is –30 mV/pH. For more discussion of this subject,
J., and M. C. Sorgato. 1982. Proton electrochemical see ref. 2.
gradients and energy-transduction processes. Annu.
32. Reviewed in Unemoto, T., H. Tokuda, and M.
Rev. Biochem. 51:185–217.)
Hayashi. 1990. Primary sodium pumps and their
21. Nicholls, D. G., and S. J. Ferguson. 2002. signficance in bacterial energetics, pp. 33–54. In: T.
Bioenergetics 3, pp. 195–217. Elsevier Science Ltd., A. Krulwich (Ed.). The Bacteria, Vol. XII. Academic
London. Press, New York.
Free ebooks ==>
membrane bioenergetics: the proton potential 143

33. Reviewed in: Skulachev, V. P. 1992. Chemiosmotic 41. Dimroth, P. 1980. A new sodium-transport sys-
systems and the basic principles of cell energetics, pp. tem energized by the decarboxylation of oxaloac-
37–73. In: Molecular Mechanisms in Bioenergetics. etate. FEBS Lett. 122:234–236.
Ernster, L. (Ed.). New Comprehensive Biochemistry:
42. Dimroth, P., and A. Thomer. 1988. Dissociation
Molecular Mechanisms in Bioenergetics, Vol. 23.
of the sodium-ion-translocating oxaloacetate decar-
Elsevier, Amsterdam.
boxylase of Klebsiella pneumoniae and reconstitu-
34. Tokuda, H., and T. Unemoto. 1982. Character- tion of the active complex from the isolated subunits.
ization of the respiration-dependent Na+ pump in Eur. J. Biochem. 175:175–180.
the marine bacterium Vibrio alginolyticus. J. Biol.
Chem. 257:10007–10014. 43. Hilpert, W., and P. Dimroth. 1983. Purification
and characterization of a new sodium transport
35. Maloney, P. C., and F. C. Hansen III. 1982. decarboxylase. Methylmalonyl–CoA decarboxy-
Stoichiometry of proton movements coupled to ATP lase from Veillonella alcalescens. Eur. J. Biochem.
synthesis driven by a pH gradient in Streptococcus 132:579–587.
lactis. J. Membrane Biol. 66:63–75.
44. Buckel, W., and R. Semmler. 1983. Purification,
36. In his review of primary sodium ion translo- characterization and reconstitution of glutaconyl–
cating enzymes, Dimroth points out that V. algi- CoA decarboxylase. Eur. J. Biochem. 136:427–434.
nolyticus has two different NADH:ubiquinone
oxidoreductases: NQR1, which is Na+ dependent 45. Schink, B., and N. Pfennig. 1982. Propionigenium
and functions at pH 8.5 but not at pH 6.5, and modestum gen. nov. sp. nov. A new strictly anaerobic
NQR2, which is Na+ independent and is not a cou- nonsporing bacterium growing on succinate. Arch.
pling site. There is apparently no H+-dependent Microbiol. 133:209–216.
NADH:ubiquinone oxidoreductase. However, these 46. Hilpert, W., B. Schink, and P. Dimroth. 1984.
bacteria do have a cytochrome bo oxidase that oxi- Life by a new decarboxylation-dependent energy
dizes the quinol, is not Na+ dependent, and is believed conservation mechanism with Na+ as coupling ion.
to be a proton pump as in other bacteria. The pres- EMBO J. 3:1665–1670.
ence of both pumps can explain how V. alginolyti-
cus operates a Na+-dependent respiratory pump at 47. De Vries, W., R. Theresia, M. Rietveld-Struijk,
pH 8.5 and a H+-dependent respiratory pump at pH and A. H. Stouthamer. 1977. ATP formation asso-
6.5. The cytochrome bo proton pump must function ciated with fumarate and nitrate reduction in grow-
at both acidic and basic pH values because mutants ing cultures of Veillonella alcalescens. Antonie van
lacking the Na+-dependent NADH:ubiquinone oxi- Leeuwenhoek 43:153–167.
doreductase extrude Na+ at pH 8.5, using a Na+/H+ 48. Buckel, W., and R. Semmler. 1982. A biotin-de-
antiporter in combination with a primary proton pendent sodium pump: glutaconyl–CoA decarboxy-
pump, and the wild type is known to extrude Na+ at lase from Acidaminococcus fermentans. FEBS Lett.
pH 6.5, using the Na+/H+ antiporter in combination 148:35–38.
with the primary proton pump. (Dimroth, P. 1997.
Primary sodium ion translocating enzymes. Biochim. 49. Inside-out vesicles are prepared by sonicating
Biophys. Acta 1318:11–51.) whole cells or shearing them with a French pres-
sure cell. Right-side-out vesicles are prepared by first
37. Kreke, B., and H. Cypionka. 1994. Role of removing the cell wall with lysozyme in a hypertonic
sodium ions for sulfate transport and energy metab- medium, and then osmotically lysing the protoplasts
olism in Desulfovibrio salexigens. Arch. Microbiol. or spheroplasts in hypotonic medium.
50. Anantharam, V., M. J. Allison, and P. C.
38. Many nonfermenting anaerobic bacteria carry Maloney. 1989. Oxalate:formate exchange. J. Biol.
out electron transport by using as electron accep- Chem. 264:7244–7250.
tors either organic compounds such as fumarate or
inorganic compounds such as nitrate. Thus, electron 51. Baetz, A. L., and M. J. Allison. 1990. Purification
flow in these bacteria can be coupled to proton efflux and characterization of oxalyl–coenzyme A decar-
and the establishment of a ∆p. Furthermore, even boxylase from Oxalobacter formigenes. J. Bacteriol.
fermenting bacteria can carry out some fumarate res- 171:2605–2608.
piration generating a ∆p. However, the major source 52. Baetz, A. L., and M. J. Allison. 1990. Purification
of energy for the ∆p in most fermenting bacteria is and characterization of formyl–coenzyme A trans-
ATP hydrolysis. ferase from Oxalobacter formigenes. J. Bacteriol.
39. Dimroth, P. 1997. Primary sodiumion translocat- 171:3537–3540.
ing enzymes. Biochim. Biophys. Acta 1318:11–51.
53. Ruan, Z., V. Anantharam, I. T. Crawford,
40. Dimroth, P. 1990. Energy transductions by an S. V. Ambudkar, S. Y. Rhee, M. J. Allison, and P.
electrochemical gradient of sodium ions, pp. 114– C. Maloney. 1992. Identification, purification, and
127. In: The Molecular Basis of Bacterial Metabolism. reconstitution of OxIT, the oxalate:formate antiport
G. Hauska and R. Thauer (Eds.). Springer-Verlag, protein of Oxalobacter formigenes. J. Biol. Chem.
Berlin. 267:10537–10543.
Free ebooks ==>
144 the physiology and biochemistry of prokaryotes

54. To prepare proteoliposomes, one disperses phos- growing and nongrowing cells of Streptococcus cre-
pholipids (e.g., those isolated from E. coli) in water, moris. J. Bacteriol. 162:383–390.
where they spontaneously aggregate to form spheri-
65. Oesterhelt, D., and J. Tittor. 1989. Two pumps,
cal vesicles consisting of concentric layers of phos-
one principle: light-driven ion transport in halobac-
pholipid. These vesicles, called liposomes, are then
teria. Trends Biochem. Sci. 14:57–61.
subjected to high-frequency sound waves (sonic
oscillation), which break them into smaller vesicles 66. Bogomolni, R. A., R. A. Baker, R. H. Lozier, and
surrounded by a single phospholipid bilayer resem- W. Stoeckenius. 1980. Action spectrum and quan-
bling the lipid bilayer found in natural membranes. tum efficiency for proton pumping in Halobacterium
Then purified protein (e.g., the OxIT antiporter) is halobium. Biochemistry 19:2152–2159.
mixed with the sonicated phospholipids in the pres-
67. Henderson, R., J. M. Baldwin, and T. A. Ceska.
ence of detergent, and the suspension is diluted into
1990. Model for the structure of bacteriorhodopsin
buffer. The protein becomes incorporated into the
based on high-resolution electron cryomicroscopy.
phospholipid bilayer, and membrane vesicles called
J. Mol. Biol. 213:899–929.
proteoliposomes are formed. When the proteoli-
posomes are incubated with solute, they catalyze 68. The extreme halophiles require unusually high
uptake of the solute into the vesicles, provided the external NaCl concentrations [at least 3–5 M (i.e.,
appropriate carrier protein has been incorporated. In 17–28%)] to grow. They inhabit hypersaline envi-
addition, one can “load” the proteoliposomes with ronments such as the solar salt evaporation ponds
solutes (e.g., oxalate) by including these in the dilu- near San Francisco and salt lakes (e.g., the Great Salt
tion buffer. Lake in Utah and the Dead Sea). There are now six
recognized genera, two of them being the well-known
55. Poolman, B., D. Molenaar, E. J. Smid, T. Ubbink,
Halobacterium and Halococcus. The best studied is
T. Abee, P. P. Renault, and W. N. Konings. 1991.
Hb. salinarium (halobium). The other four genera
Malolactic fermentation: electrogenic malate uptake
are Haloarcula, Haloferax, Natronobacterium, and
and malate/lactate antiport generate metabolic
Natronococcus. The majority of the known halo-
energy. J. Bacteriol. 173:6030–6037.
philic archaea are aerobic chemo-organotrophs
56. Konings, W. N. 1985. Generation of metabolic and can grow on simple carbohydrates as well as
energy by end-product efflux. Trends Biochem. Sci. long-chain saturated hydrocarbons. They generally
10:317–319. grow best at pH values between 8 and 9. However,
Natronobacterium and Natronococcus are also alka-
57. Konings, W. N., J. S. Lolkema, and B. Poolman.
liphilic and grow well at pH values up to 11. When
1995. The generation of metabolic energy by solute
oxygen is not present, the halobacteria will grow
transport. Arch. Microbiol. 164:235–242.
anaerobically by using several electron acceptors in
58. Michels, J. P., J. Michel, J. Boonstra, and W. place of oxygen. These include fumarate, dimethyl
N. Konings. 1979. Generation of an electrochemi- sulfoxide (DMSO), and trimethylamine N-oxide
cal proton gradient in bacteria by the excretion of (TMAO). Members of the genera Haloarcula and
metabolic end-products. FEMS Microbiol. Lett. Haloferax can grow on nitrate as the terminal elec-
5:357–364. tron acceptor. Some of the halobacteria reduce the
nitrate to nitrite and some reduce it completely to
59. Otto, R., et al. 1982. Lactate efflux–induced elec-
nitrogen gas. Some halobacteria can also grow fer-
trical potential in membrane vesicles of Streptococcus
mentatively in the absence of oxygen. These include
cremoris. J. Bacteriol. 149:733–738.
Hb. salinarium (halobium), which can ferment argi-
60. Brink, B. T., and W. N. Konings. 1982. nine to citrulline.
Electrochemical proton gradient and lactate con-
69. Oesterhelt, D., and G. Krippahl. 1983.
centration gradient in Streptococcus cremoris cells
Phototrophic growth of halobacteria and its use for
grown in batch culture. J. Bacteriol. 152:682–686.
isolation of photosynthetically deficient mutants.
61. Driessen, A. J. M., and W. N. Konings. 1990. Ann. Microbiol. (Inst. Pasteur). 134B:137–150.
Energetic problems of bacterial fermentations: extru-
70. Gest, H. 1993. Photosynthetic and quasi-photo-
sion of metabolic end products, pp. 449–478. In:
synthetic bacteria. FEMS Microbiol. Lett. 112:1–6.
Bacterial Energetics. T. A. Krulwich (Ed.). Academic
Press, New York. 71. Some halophilic archaea can use nitrate as an
electron acceptor to carry out anaerobic respiration.
62. Michel, T. A., and J. M. Macy. 1990. Generation
of a membrane potential by sodium-dependent succi- 72. Respiration can be severely limited under cer-
nate efflux in Selenomonas ruminantium. J. Bacteriol. tain growth conditions because the oxygen content
172:1430–1435. of hypersaline waters, the normal habitat of these
organisms, is usually 20% or less than is found in
63. The lactic and succinic acids are presumed to be
normal seawater, and in unstirred ponds oxygen
in the ionized form because the intracellular pH is
becomes even more scarce. The halobacteria can
much larger than the pKa values.
derive energy from the fermentation of amino acids;
64. Brink, B. T., R. Otto, U. Hansen, and W. N. but in the absence of a fermentable carbon source
Konings. 1985. Energy recycling by lactate efflux in and respiration, light is the only source of energy.
Free ebooks ==>
membrane bioenergetics: the proton potential 145

73. Bogomolni, R. A., R. A. Baker, R. H. Lozier, and 81. Krebs, M. P., and H. G. Khorana. 1993.
W. Stoeckenius. 1980. Action spectrum and quan- Mechanism of light-dependent proton translocation
tum efficiency for proton pumping in Halobacterium by bacteriorhodopsin. J. Bacteriol. 175:1555–1560.
halobium. Biochemistry 19:2152–2159.
82. Since aspartate-85 remains protonated while a
74. Tittor, J., and D. Oesterhelt. 1990. The quan- proton is released into the aqueous phase, we must
tum yield of bacteriorhodopsin. FEBS Lett. 263: conclude that the immediate source of the released
269–273. proton is a different amino acid residue.
75. Govindjee, R., S. P. Balashov, and T. G. Ebrey. 83. Senior, A. E. 1990. The proton-translocating
1990. Quantum efficiency of the photochemical cycle ATPase of Escherichia coli. Annu. Rev. Biophys.
of bacteriorhodopsin. Biophys. J. 58:597–608. Chem. 19:7–41.
76. Lanyi, J. K. 1997. Mechanism of ion trans- 84. Lanyi, J. K. 1990. Halorhodopsin, a light-driven
port across membranes. J. Biol. Chem. 272: electrogenic chloride-transport system. Physiol. Rev.
31209–31212. 70:319–330.
77. A Schiff base is an imine and has the following 85. Duschl, A., and G. Wagner. 1986. Primary and
structure: R–CH5N–R′. It is formed between a car- secondary chloride transport in Halobacterium halo-
bonyl group and a primary amine, as follows: R–CHO bium. J. Bacteriol. 168:548–552.
+ H2N–R′ → R–CH5N–R′ + H2O. In bacteriorho-
86. There are many nonobligate alkaliphilic bacteria
dopsin, the amino group is donated by lysine in the
whose optimal growth pH is 9 or greater that never-
protein and the carbonyl is donated by the retinal.
theless can grow at pH 7 or below. These microor-
78. Lanyi, J. K. 1992. Proton transfer and energy ganisms, widely distributed among the bacterial
coupling in the bacteriorhodopsin photocycle. J. genera, include both bacteria and archaea. There are
Bioenerg. Biomemb. 24:169–179. also alkaliphilic-tolerant bacteria that can grow at
pH values of 9 or more, but whose optimal growth
79. Oesterhelt, D., J. Tittor, and E. Bamberg. 1992.
pH is around neutrality.
A unifying concept for ion translocation by retinal
proteins. J. Bioenerg. Biomemb. 24:181–191. 87. Reviewed in: Krulwich, T. A., and D. M. Ivey.
1990. Bioenergetics in extreme environments,
80. Fodor, S. P. A., J. B. Ames, R. Gebhard, E. M.
pp. 417–447. In: The Bacteria, Vol. XII. T. A.
M. van den Berg, W. Stoeckenius, J. Lugtenburg,
Krulwich (Ed.). Academic Press, New York.
and R. A. Mathies. 1988. Chromophore structure
in bacteriorhodopsin’s N intermediate: implications 88. Krulwich, T. A., and A. A. Guffanti. 1989.
for the proton-pumping mechanism. Biochemistry Alkalophilic bacteria. Annu. Rev. Microbiol.
27:7097–7101. 43:435–463.
Free ebooks ==>

Electron Transport

Electron transport refers to the current of elec- membrane as part of reduced quinone (QH2).
trons in the cell membranes of prokaryotes, and Sometimes (e.g., Fig. 5.12), reduced quinone is
in mitochondrial and chloroplast membranes. referred to as reduced ubiquinone (UQH2). The
The electrons flow spontaneously down an quinone is picked up as a proton (H+) from the
energy gradient through a series of electron car- cytosol on one side of the membrane, translo-
riers, and at specific sites in the chain of carri- cated through the membrane as part of reduced
ers, the energy that is released is conserved as quinone, and released as H+ on the outside of the
a proton motive force (∆p) that can be used membrane surface. At other coupling sites, the
to make ATP and supply energy for other cel- proton itself (H+) is actually pumped through
lular tasks. This chapter describes the entire the membrane. The net result, regardless of
process as it occurs in bacteria and in mito- the mechanism, is the translocation of protons
chondria. Electron transport in photosynthetic from one side of the membrane to the other side,
membranes is described in Chapter 6. Archaeal as shown in Fig. 5.1. This is important because
electron transport chains appear to be similar a membrane potential is created during proton
in many ways to bacterial electron transport translocation (due either to the transmembrane
chains but are quite diverse. For reviews, see movement of electrons or of H+, as explained
refs. 1 and 2. later), and also because the reentry of protons
In what direction do electrons spontaneously through the membrane-bound ATP synthase
flow? As discussed in Section 4.7.1, electrons results in the synthesis of ATP. As explained in
flow spontaneously down a potential energy Chapter 4, the energy to synthesize the ATP is
gradient toward acceptors that have a more derived from a combination of the membrane
positive electrode potential. The electrons potential and the proton concentration gradi-
flow from primary electron donors to termi- ent, called the ∆p.
nal electron acceptors through a series of elec- Mitochondria have three coupling sites,
tron carrier proteins and a class of lipids called whereas bacteria can have one to three coupling
quinones. An important point is that proton sites depending upon the bacterium and the
translocation across the membranes takes place growth conditions. In other words, prokaryotic
during electron transport at specific sites in the cell membranes (and mitochondria and chloro-
electron transport chain called coupling sites; plasts) convert an electrode potential difference
an electrochemical proton potential, called a ∆Eh into a proton electrochemical potential dif-
proton motive force, or ∆p, is created at these ference (∆p). (As discussed in Section 4.7.1, Eh is
sites. In some cases (i.e., the Q loop and Q the actual electrode potential of a compound at
cycle, explained in Sections 5.6.1 and 5.6.2), the specified concentrations of its oxidized and
the hydrogen is actually transported across the reduced forms.) The proton electrochemical

Free ebooks ==>
electron transport 147

OUT IN the final electron acceptors, the mechanism is

referred to as anaerobic respiration. The final
electron acceptors can be inorganic, such as
oxygen, nitrate, or sulfate. They can also be
yH+ organic, for example, fumarate. The name of
the respiration is derived from the final electron
Box acceptor. Thus, we have oxygen respiration,
Bred nitrate respiration, sulfate respiration, fumar-
ate respiration, and so on.

Fig. 5.1 Electron transport in membranes. Electrons 5.2 The Electron Carriers
flow from A to B, through a series of electron carri-
ers in the membrane, from a low electrode potential The electrons flow through a series of electron
toward a higher potential. The individual reactions carriers. Some of these carry hydrogen as well
that transfer the electrons are referred to as redox as electrons, and some carry only electrons. The
reactions. The intermediate redox reactions between electron carriers are as follows:
A and B are not shown. Certain of the redox reac-
tions in the series are coupled to the translocation of 1. Flavoproteins (hydrogen and electron carriers)
protons across the cell membrane, involving either 2. Quinones (hydrogen and electron carriers)
reduced quinone (UQH2) or H+ pumps. These occur 3. Iron–sulfur proteins (electron carriers)
at coupling sites. In this way a redox potential (∆Eh), 4. Cytochromes (electron carriers)
which in this case refers to the energy that is released
when the electrons move down an energy gradient
The quinones are lipids, whereas the other
from the donor to the acceptor, is converted into a electron carriers are proteins, which exist in
proton potential (∆p): that is, n∆Eh = y∆p, where n multiprotein enzyme complexes called oxi-
is the number of electrons transferred from donor to doreductases. (See note 3 for examples of oxi-
acceptor, and y is the number of protons extruded. doreductases.) The electrons are not carried in
When the protons return across the membrane, work the protein per se, but in a nonprotein molecule
(e.g., the synthesis of ATP or solute transport) can be bound to the protein. The nonprotein portion
done. The type of work that is done depends upon that carries the electron is called a prosthetic
the protein machinery through which the protons group. (See note 4 for useful definitions.) The
return. prosthetic group in iron–sulfur proteins is a
cluster of iron–sulfide, which is abbreviated as
FeS. The prosthetic group in flavoproteins (Fp)
potential is then used to drive solute transport, is a flavin, which can be either flavin adenine
ATP synthesis, flagellar rotation, and other dinucleotide (FAD) or flavin mononucleotide
membrane activities. In mitochondria, all elec- (FMN). The prosthetic group in cytochromes
tron transport pathways are much the same. is heme. The chemistry of the prosthetic groups
However, prokaryotes are diverse creatures, is described in Section 5.2.1. Some of the pros-
and their electron transport pathways differ thetic groups (flavins) carry hydrogen as well as
depending upon the primary donor and termi- electrons, and they are referred to as hydrogen
nal acceptor. This chapter describes electron carriers. The quinones are also hydrogen car-
transport pathways in mitochondria and bacte- riers. Some of the prosthetic groups (FeS and
ria and how they are coupled to the formation heme) carry only electrons, and they are referred
of a ∆p. to as electron carriers.
Each of the electron carriers, described in
Sections 5.2.1 through 5.2.4, has a different
5.1 Aerobic and Anaerobic electrode potential, and the electrons are trans-
Respiration ferred sequentially to a carrier of a higher poten-
Electron transport in membranes is referred tial to the final acceptor, which has the highest
to as respiration. If oxygen is the final electron potential.
acceptor, then the mechanism is referred to as The standard potentials at pH 7 of the elec-
aerobic respiration. If other compounds are tron carriers and other electron donors and
Free ebooks ==>
148 the physiology and biochemistry of prokaryotes

acceptors in metabolic pathways are shown in two ring nitrogens. There are many different
Table 5.1. flavoproteins, and they catalyze diverse oxida-
tion–reduction reactions in the cytoplasm, not
5.2.1 Flavoproteins merely those of the electron transport chain in
A flavoprotein (Fp) is an electron carrier that the membranes. Although all the flavoproteins
has as its prosthetic group an organic molecule have FMN or FAD as their prosthetic group,
called a flavin. The term is derived from the they catalyze different oxidations and have dif-
Latin word flavius, which means yellow, in ref- ferent redox potentials. These differences are
erence to the color of flavins. The flavins FAD due to differences in the protein component of
and FMN are synthesized by cells from the vita- the enzyme, not in the flavin itself.
min riboflavin (vitamin B2).
What is the difference between FMN and 5.2.2 Quinones
FAD? This is shown in Fig. 5.2. Phosphorylation Quinones are lipid electron carriers. Owing
of riboflavin at the ribityl 5′-OH yields FMN, to their hydrophobic lipid nature, some are
and adenylylation (addition of ADP) yields believed to be highly mobile in the lipid phase
FAD. of the membrane, carrying hydrogen and elec-
As Fig. 5.2 illustrates, when flavins are trons to and from the complexes of protein
reduced they carry 2H• (equivalent to two electron carriers that are not mobile. Quinone
electrons and two protons), one on each of structures and oxidation–reduction reactions
are shown in Fig. 5.3. All quinones have hydro-
phobic isoprenoid side chains that contribute
Table 5.1 Standard electrode potentials at pH 7 to their lipid solubility. The number of isoprene
units varies but is typically 6 to 10. Bacteria
Couple Potential (mV)
make two types of quinone that function dur-
Fdox/Fdred (spinach) –432 ing respiration: ubiquinone (UQ), a quinone
CO2/formate –432 also found in mitochondria, and menaquinone
H+/H2 –410 (MQ, or sometimes MK). Menaquinones (Fig.
Fdox/Fdred (Clostridium) –410
NAD+/NADH –320
5.3C), which are derivatives of vitamin K, differ
FeS (ox/red) in mitochondria –305 from ubiquinones in being naphthoquinones in
Lipoic/dihydrolipophilic –290 which the additional benzene ring replaces the
Sº/H2S –270 two methoxy groups present in ubiquinones
FAD/FADH2 –220 (Fig. 5.3A,B). Menaquinones also have a much
Acetaldehyde/ethanol –197
FMN/FMNH2 –190
lower electrode potential than ubiquinones and
Pyruvate/lactate –185 are used predominantly during anaerobic res-
Oxaloacetate/malate –170 piration, where the electron acceptor has a low
Menaquinone (ox/red) –74 potential (e.g., during fumarate respiration). A
cyt b558 (ox/red) –75 to –43 third type of quinone, plastoquinone (Fig. 5.3D),
Fumarate/succinate + 33
Ubiquinone (ox/red) +100
occurs in chloroplasts and cyanobacteria, and
cyt b556 (ox/red) +46 to +129 functions in photosynthetic electron transport.
cyt b562 (ox/red) +125 to +260 In plastoquinones, the two methoxy groups are
cyt d (ox/red) +260 to +280 replaced by methyl groups.
cyt c (ox/red) +250
FeS (ox/red) in mitochondria +280
cyt a (ox/red) +290 5.2.3 Iron–sulfur proteins
cyt c555 (ox/red) +355 Iron–sulfur proteins contain nonheme iron and
cyt a3 (ox/red) in mitochondria +385 usually acid-labile sulfur (Fig. 5.4). The term
NO−3/NO−2 +421 “acid-labile sulfur” means that when the pH is
Fe3+/Fe2+ +771
O2 (1 atm)/H2O +815 lowered to approximately 1, hydrogen sulfide is
released from the protein. This is because there
Sources: Thauer, R. K., K. Jungermann, and K. Decker. is sulfide attached to iron by bonds that are rup-
1977. Energy conservation in chemotrophic anaerobic
tured in acid. Generally, the proteins contain
bacteria. Bacteriol Rev. 41:100–180. Metzler, D. E. 1977.
Biochemistry: The Chemical Reactions of Living Cells. clusters in which iron and acid-labile sulfur are
Academic Press, New York. present in a ratio of 1:1. However, there may
Free ebooks ==>
electron transport 149

Fig. 5.2 Structures of riboflavin (X = H), FMN (X = PO3H2), and FAD (X = ADP). For convenience, the reduc-
tion reaction is drawn as proceeding via a hydride ion even though this is not the actual mechanism in all flavin

Fig. 5.3 The structure of quinones: (A) oxidized ubiquinone, (B) reduced ubiquinone, (C) oxidized menaqui-
none, and (D) oxidized plastoquinone. The value of n can be 4 to 10 and is 8 for both quinones in E. coli. In
E. coli ubiquinone plays a major role in aerobic and nitrate respiration, whereas menaquinone is dominant
during fumarate respiration. One reason for this is that ubiquinone has a potential of +100 mV, versus +30
mV for fumarate. It is therefore at too high a potential to deliver electrons to fumarate. Menaquinone has a
low potential, –74 mV, and is thus able to deliver electrons to fumarate. Plastoquinone is used in chloroplast
and cyanobacterial photosynthetic electron transport.

be more than one iron–sulfur cluster per pro-

tein. For example, in mitochondria the enzyme
complex that oxidizes NADH has at least four
FeS clusters (see later, Fig. 5.9). The FeS clusters
have different Eh values, and the electron trav-
els from one FeS cluster to the next toward the
higher Eh. It appears that the electron may not
be localized on any particular iron atom, and
the entire FeS cluster should be thought of as
carrying one electron, regardless of the number
of Fe atoms. Fig. 5.4 Scheme for FeS clusters: this is a Fe2S2 clus-
Iron–sulfurs proteins also contain cysteine ter. More than one cluster may be present per pro-
sulfur, which is not acid labile, and bonds the tein. The sulfur atoms held only by the iron are acid
iron to the protein. There are several different labile. The iron is bonded to the protein via sulfur in
types of iron–sulfur protein, and these catalyze cysteine residues.
Free ebooks ==>
150 the physiology and biochemistry of prokaryotes

numerous oxidation–reduction reactions in the oxygen. (In naming cytochromes, sometimes

cytoplasm as well as in the membranes. (See the O2-binding heme is given the subscript 3.)7
note 5 for more information on iron–sulfur As the names imply, each cytochrome contains
proteins.) The iron–sulfur proteins have char- two types of heme, one being heme b and the
acteristic electron spin resonance (ESR) spectra other being heme d or o.7 As mentioned previ-
because of an unpaired electron in either the ously, the hemes can be distinguished according
oxidized or reduced form of the FeS cluster in to the side groups that they possess, as summa-
different FeS proteins. (See note 6 for a descrip- rized in Fig. 5.5. For example, heme o differs
tion of electron spin resonance.) The iron–sulfur from heme b in having an hydroxyethylfarnesyl
proteins cover a very wide range of potentials, group substituted for a vinyl group. However,
from approximately –400 mV to +350 mV. the only difference between heme b and heme
They therefore can carry out oxidation–reduc- c is that the latter is covalently bound to pro-
tion reactions at both the low-potential end and tein by thioether linkages between the two vinyl
the high-potential end of the electron transport groups and cysteine residues in the protein.
chain, and indeed are found in several locations. Hemes can usually be distinguished spectro-
In the FeS cluster shown in Fig. 5.4, note that photometrically. When cytochromes are in the
each Fe is bound to two acid-labile sulfurs and reduced state, absorption by the heme produces
two cysteine sulfurs. This would be called an characteristic light absorption bands in the vis-
Fe2S2 cluster. ible range: the α, β, and γ bands. The α bands
absorb light between 500 and 600 nm, the β
5.2.4 Cytochromes bands absorb at a lower wavelength, and the γ
Cytochromes are electron carriers that have bands (also called Soret bands) are in the blue
heme as the prosthetic group. Heme consists region of the spectrum. The spectrum for a cyto-
of four pyrrole rings attached to each other by chrome c is shown in Fig. 5.6. Cytochromes are
methene bridges (Fig. 5.5). Because hemes have distinguished, in part, by the position of the max-
four pyrroles, they are called tetrapyrroles. imum in the α band. For example, cyt b556 has a
Each of the pyrrole rings is substituted by a peak at 556 nm and cyt b558 a peak at 558 nm.
side chain. Substituted tetrapyrroles are called
porphyrins. Therefore, hemes are also called Reduced minus oxidized spectra
porphyrins. (An unsubstituted tetrapyrrole is Because of light scattering and nonspecific
called a porphin.) Hemes are placed in differ- absorption, it is very difficult to resolve the
ent classes, described shortly, on the basis of the different peaks of individual cytochromes in
side chains attached to the pyrrole rings. In the whole cells unless one employs difference spec-
center of each heme there is an iron atom that troscopy. For difference spectroscopy, the cells
is bound to the nitrogen of the pyrrole rings. are placed into two cuvettes in a split-beam
The iron is the electron carrier and is oxidized spectrophotometer, and monochromatic light
to ferric or reduced to ferrous ion during elec- from a single monochromator scan is split to
tron transport. Cytochromes are therefore one- pass through both cuvettes. In one cuvette the
electron carriers. The Eh values of the different cytochromes are oxidized by adding an oxi-
cytochromes vary depending on the protein and dant, and in the second cuvette they are reduced
the molecular interactions with surrounding by adding a reductant. The spectrophotometer
molecules. subtracts the output of one cuvette from the
other to give a reduced minus oxidized differ-
Classes of cytochromes ence spectrum. In this way nonspecific absorp-
Figure 5.5 shows five classes of heme that dis- tion and light scattering are eliminated from the
tinguish the cytochromes: hemes a, b, c, d, and spectrum, and the cytochromes in the prepara-
o. Hemes d and o have been found only in the tion are identified.
prokaryotic cytochrome oxidases. Bacterial
cytochromes include cytochromes bd (some- Dual-beam spectroscopy
times called cytochrome d) and bo (some- A dual-beam spectrophotometer is used to fol-
times called cytochrome bo3 or cytochrome low the kinetics of oxidation or reduction of
o), which are quinol oxidases that reduce a particular cytochrome. The instrument has
Free ebooks ==>
electron transport 151

Fig. 5.5 The prosthetic groups of the different classes of cytochromes. The hemes vary according to their
side groups. Heme c is covalently bound to the protein via a sulfur bridge to a cysteine residue on the protein.
Source: Adapted from Gottschalk, G. Bacterial Metabolism, Springer-Verlag, New York. 1986.

two monochromators: light from one is set at Table 5.1 shows standard electrode potentials
a wavelength at which absorbance will change at pH 7 (E′0) of some electron donors, acceptors,
during oxidation or reduction, and the second and electron carriers. Notice that redox couples
beam of light is at a nearby wavelength for are generally written in the form “oxidized/
which absorbance will not change. The light is reduced.” Many of the oxidation–reduction
sent alternatively from both monochromators reactions in the electron transport chain can be
through the sample cuvette, and the difference reversed by the ∆p as discussed in Section 4.7.1.
in absorbance between the two wavelengths is This means that the ox/red ratio for several of the
automatically plotted as a function of time. electron carriers (flavoproteins, cytochromes,
Free ebooks ==>
152 the physiology and biochemistry of prokaryotes

Four complexes can be recognized in

mitochondria. They are complex I (NADH–
ubiquinone oxidoreductase), complex II
(succinate dehydrogenase), complex III
(ubiquinol–cytochrome c oxidoreductase,
also called the bc1 complex), and complex IV
(cytochrome c oxidase, which is cytochrome
aa3). Complexes I, III, and IV are coupling sites
(Section 5.5). Each complex can have several
proteins. The most intricate is complex I, from
mammalian mitochondria, which has about
40 polypeptide subunits, at least four iron–sul-
fur centers, one flavin mononucleotide (FMN),
and one or two bound ubiquinones. Analogous
complexes have been isolated from bacteria,
Fig. 5.6 Absorption spectra of oxidized (dashed but in some cases (e.g., NADH–ubiquinone
curve) and reduced (solid curve) cytochrome c. The oxidoreductase and the bc1 complex), they
α band in the reduced form is used to identify cyto- have fewer protein components.8–11 (See note
chromes. 12 for a description of complex II and how it
varies with different bacteria.) Note the pat-
tern in the arrangement of the electron carri-
quinones, FeS proteins) must be close to 1. ers in mitochondria; a dehydrogenase complex
Thus, for these reactions the Eh (actual potential accepts electrons from a primary donor and
at pH 7) values of the redox couples are close to transfers the electrons to a quinone. The qui-
their midpoint potentials, which is the potential none then transfers the electrons to an oxi-
at pH 7 when the couple is 50% reduced: that dase complex via intervening cytochromes. As
is, [ox] = [red]. described next, the same general pattern exists
in bacteria.
5.3 Organization of the Electron
Carriers in Mitochondria 5.4 Organization of the Electron
For a historical perspective, see Box 5.1. Carriers in Bacteria
The electron carriers are organized as an For a review, see ref. 13. Bacterial electron
electron transport chain that transfers electrons transport chains vary among the different bac-
from electron donors at a low electrode poten- teria, and also according to the growth condi-
tial to electron acceptors at a higher electrode tions. These variations will be discussed later
potential (Fig. 5.7). Electrons can enter at the (Section 5.7). First, we shall describe the com-
level of flavoprotein, quinone, or cytochrome, mon features in bacterial electron transport
depending upon the potential of the donor. schemes and compare them with systems of
The carriers are organized in the membrane as mitochondrial electron transport. As with the
individual complexes. The complexes can be mitochondrial electron transport chain, the bac-
isolated from each other by appropriate separa- terial chains are organized into dehydrogenase
tion techniques after mild detergent extraction, and oxidase complexes connected by quinones
which removes the lipids but does not destroy (Fig. 5.8). The quinones accept electrons from
the protein–protein interactions. The separated dehydrogenases and transfer them to oxidase
complexes can be analyzed for their components complexes that reduce the terminal electron
and also can be incorporated into proteolipo- acceptor. Bacteria are capable of using electron
somes to facilitate study of the oxidation–reduc- acceptors other than oxygen (e.g., nitrate and
tion reactions that each catalyzes, in addition fumarate) during anaerobic respiration. The
to proton translocation. (Proteoliposomes are enzyme complexes that reduce electron accep-
artificial constructs of purified lipids and pro- tors other than oxygen are called reductases,
teins. They are described in Section 17.1.) rather than oxidases.
Free ebooks ==>
electron transport 153


The elucidation of the pathway by which 1887. MacMunn reported the discovery of
electrons flow from organic compounds a pigment in the course of a spectroscopic
to oxygen was the result of many years of examination of the tissues of various species
research by different investigators. The of vertebrates and invertebrates. He called
realization that iron compounds were the pigment histohaematin, or myohae-
electron carriers came from early work matin, and it had a characteristic absorp-
on the effect of cyanide on respiration. tion spectrum consisting of four bands.
Otto Heinrich Warburg (1883–1970), a MacMunn reported that when he added
German biochemist who studied respi- the oxidizing agent hydrogen peroxide, the
ration, reported the inhibitory effect of bands disappeared, only to reappear upon
cyanide on respiration in sea urchin eggs, reduction. He concluded that since the sub-
yeast, and bacteria before and after World stances were capable of being oxidized and
War I. Because he knew that cyanide inhib- reduced, they were respiratory pigments.
its autoxidation reactions (e.g., the oxida- He also concluded that they were protein.
tion of cysteine to cystine, catalyzed by iron However, MacMunn’s discovery was not
compounds), Warburg concluded that cya- accepted by his contemporaries, especially
nide also inhibits respiration, further rea- the famous biochemist Hoppe-Seyler, who
soning that an iron-containing enzyme that thought MacMunn’s belief that he had dis-
he called Atmungsfermentt (“respiratory covered a cellular respiratory pigment was
ferment”) catalyzes the oxidation of the erroneous.
organic substrates. Warburg also showed, MacMunn was clearly frustrated that
in 1926, that carbon monoxide inhibits the his discovery had not been accepted, and
uptake of oxygen by yeast cells. He knew he wrote in a book that was not published
from earlier studies by others that carbon until after he died, in 1911, that “doubtless
monoxide combines with hemoglobin and in time this pigment will find its way into
that it can be dissociated from the ferro– the textbooks.”1 This indeed did occur, pri-
heme complex with visible light. He was marily through the work of David Keilin in
able to show that visible light also reverses the 1920s.
the inhibition of respiration of yeast by David Keilin, who left Poland to study
carbon monoxide. By measuring the effect first in Belgium, then in France, and emi-
of light of different wavelengths on revers- grated to England in 1915, was unaware
ing the inhibition, Warburg was able to of MacMunn’s work, although later when
determine the absorption spectrum of the he learned of it, he acknowledged it. In
pigment. The absorption spectrum, which 1925 Keilin, using a spectromicroscope,
was that of heme, supported his conclusion rediscovered the pigments in the thoracic
that Atmungsfermentt was a heme pigment. muscles of insects, in Bacillus subtilis, and
(Warburg’s Atmungsferment turned out in yeast, and named them cytochromes
to be cytochrome oxidase, later studied (cellular pigments). Keilin discovered cyto-
by David Keilin.) For his discovery of the chromes a, b, and c, and later cytochrome
respiratory enzyme and its mode of action, oxidase. Keilin published several papers
Warburg was awarded the Nobel Prize in in the late 1920s and early 1930s describ-
Physiology or Medicine in 1931. ing the respiratory chain as a chain of car-
Research on respiratory enzymes riers consisting of dehydrogenases that
had actually begun in the late 1800s. remove hydrogen from organic substrates,
Cytochromes were first described by the as well as oxidized cytochromes that are
Englishman Charles Alexander MacMunn reduced by the dehydrogenases, and cyto-
in three papers published between 1884 and chrome oxidase, an autoxidizable heme
Free ebooks ==>
154 the physiology and biochemistry of prokaryotes

compound that oxidizes the cytochromes For more information about the history
and reduces oxygen. In the early 1930s it of biochemical research, see ref. 2.
was recognized that the respiratory chain
transfers electrons, rather than hydrogen, REFERENCES
from the organic substrate to oxygen, and
1. Quoted in Fruton, J. S., and S. Simmonds.
in 1931 Warburg correctly attributed the 1958. General Biochemistry, 2nd ed. John
oxidations and reductions to a change in Wiley & Sons, New York.
the valency of iron. The realization of the
2. Florkin, M. 1975. A history of biochemistry.
role of pyridine-linked dehydrogenases and In: Comprehensive Biochemistry, Vol. 31. M.
flavin-linked dehydrogenases was due to Florkin and E. H. Stotz (Eds.). Elsevier Scientific
the research of Warburg in the 1930s. Publishing, Amsterdam.

Fig. 5.7 Electron transport scheme in mitochondria. Electrons travel in the electron transport chain from a
low to a high electrode potential. Complexes I to IV are enclosed in dashed lines. Complex I is NADH dehydro-
genase, also called NADH–ubiquinone oxidoreductase. Complex II is succinate dehydrogenase, also called
succinate–ubiquinone oxidoreductase. Complex III is the bc1 complex, also called ubiquinol–cytochrome
c oxidoreductase. Complex IV is the cytochrome aa3 oxidase, also called cytochrome c oxidase. There are
several FeS clusters in complexes I and II, and an FeS protein in complex III. Complex II has both peripheral
and integral membrane protein subunits. The flavin (FAD) and FeS centers are in the peripheral membrane
subunits, and hemes (not shown) are in the integral membrane subunits. Electrons flow from FAD through
the FeS centers to quinone, probably via heme. Abbreviations: fp, flavoprotein; FeS, iron–sulfur protein; UQ,
ubiquinone; b, cytochrome b; c1, cytochrome c1; c, cytochrome c; aa3, cytochrome aa3.

Some of the dehydrogenase complexes are synthesis of some dehydrogenases, and stimu-
NADH and succinate dehydrogenase com- lates the synthesis of others. The electron carrier
plexes, analogous to complexes I and II in mito- complexes in bacteria are sometimes referred to
chondria. In addition to these dehydrogenases, as modules, since they can be synthesized and
several others reflect the diversity of substrates “plugged into” the respiratory chain when
oxidized by the bacteria: H2 dehydrogenases needed.
(called hydrogenases), formate dehydrogenase,
lactate dehydrogenase, methanol dehydroge- 5.4.1 The different terminal oxidases
nase, methylamine dehydrogenase, and so on. A word should be said about the many different
Depending upon the source of electrons and terminal oxidases found in bacteria.14 Whereas
electron acceptors, bacteria can synthesize and mitochondria all have the same cytochrome c
substitute one dehydrogenase complex for oxidase (cytochrome aa3), bacteria have a vari-
another, or reductase complexes for oxidase ety of terminal oxidases, often two or three dif-
complexes. For example, when E. coli is grown ferent ones in the same bacterium (i.e., two or
anaerobically, it makes the reductase complexes three branches to oxygen). Some of the bacte-
instead of the oxidase complexes, represses the rial terminal oxidases oxidize quinols (quinol
Free ebooks ==>
electron transport 155

Fig. 5.8 Generalized electron transport pathways found in bacteria. The details will vary depending upon the
bacterium and the growth conditions. (A) Aerobic respiration. A dehydrogenase complex removes electrons
from an electron donor and transfers these to a quinone. The electrons are transferred to an oxidase complex
via a branched pathway. Many bacteria have bc1, cytochrome c, and cytochrome aa3 in one of the branches,
and in this way they resemble mitochondria. Other bacteria do not have a bc1 complex and may or may not
have cytochrome aa3. (B) Anaerobic respiration: Y represents either an inorganic electron acceptor other than
oxygen (e.g., nitrate) or an organic electron acceptor (e.g., fumarate). Under anaerobic conditions the elec-
trons are transferred to reductase complexes, which are synthesized anaerobically. Several reductases exist,
each one specific for the electron acceptor. More than one reductase can simultaneously exist in a bacterium.

oxidases) and some oxidize cytochrome c (cyto- bo3 oxidase (cytochrome o oxidase), and cyto-
chrome c oxidases). The terminal oxidases dif- chrome bb3 oxidase. It should be emphasized
fer in their affinities for oxygen and in whether that some of these are quinol oxidases and some
they are proton pumps, as well as in the types are cytochrome c oxidases.
of hemes and metals they contain. However,
despite these differences, most belong to the 5.4.2 Bacterial electron transport
heme–copper oxidase superfamily of oxidases, chains are branched
to which the mitochondrial cytochrome c oxi- Two major difference between mitochondrial
dase also belongs. An exception is the cyto- and bacterial electron transport chains are as
chrome bd oxidase, discussed later, which is follows: (1) the routes to oxygen in the bacte-
not a member of the heme–copper oxidase ria are branched, the branch point being at the
superfamily. quinone or cytochrome, and (2) many bac-
All members of the heme–copper oxidase teria can alter their electron transport chains
superfamily share a protein subunit that is depending upon growth conditions (Fig. 5.8).
homologous to subunit I of the mitochon- As noted in Section 5.4.1, under aerobic con-
drial cytochrome c oxidase. This subunit has ditions there are often two or three branches
a bimetallic (binuclear) center that binds oxy- leading to different terminal oxidases. For
gen and reduces it to water, and pumps pro- example, a two-branched electron transport
tons. (Several of the heme–copper oxidases are chain might contain a branch leading to cyto-
known to be proton pumps, whereas it is not yet chrome o oxidase (quinol oxidase) and a branch
known whether certain others pump protons.) leading to cytochrome aa3 oxidase (cytochrome
The bimetallic center contains a heme iron and c oxidase). Other bacteria (e.g., E. coli) have
copper. There is a second heme in all these oxi- cytochrome bo and bd oxidase branches (both
dases, but it is not part of the bimetallic center are quinol oxidases) but lack the cytochrome
and probably functions in transferring electrons aa3 branch. The ability to synthesize branched
to the bimetallic center. Heme–copper oxidases electron transport pathways to oxygen confers
that are mentioned later in the context of bac- flexibility on the bacteria, since not only may
terial respiratory systems include the cbb3-type the branches differ in the Δp that can be gener-
oxidases, cytochrome aa3 oxidase, cytochrome ated (because they may differ in the number of
Free ebooks ==>
156 the physiology and biochemistry of prokaryotes

coupling sites) but their terminal oxidases also anaerobes will ferment when a terminal electron
may differ with respect to affinities for oxygen. acceptor is unavailable. Fermentation is the sub-
For example, in E. coli cytochrome bo has a low ject of Chapter 15.)
affinity for oxygen, whereas cytochrome bd has
a higher affinity for oxygen. Switching to an
oxidase with a higher affinity for oxygen allows
5.5 Coupling Sites
the cells to continue to respire even when oxy- For a review, see ref. 17. Sites in the electron
gen tensions fall to very low values. This abil- transport pathway at which redox reactions
ity is important to ensure the reoxidation of the are coupled to proton extrusion creating a ∆p
reduced quinones and NADH so that cellular are called coupling sites.18 Each coupling site
oxidations such as the oxidation of glucose to is also a site for ATP synthesis, since the pro-
pyruvate or the oxidation of pyruvate to CO2 tons extruded reenter via ATP synthase to make
may continue. ATP. The three coupling sites in mitochondria,
It has also been suggested that under sites 1, 2, and 3, were shown in Fig. 5.7, where
microaerophilic conditions the use of an oxi- site 1 is the NADH dehydrogenase complex
dase with a high affinity for oxygen would (complex I), site 2 is the bc1 complex (complex
remove traces of oxygen that might damage III), and site 3 is the cytochrome aa3 complex
oxygen-sensitive enzymes that are made under (complex IV). The succinate dehydrogenase
microaerophilic or anaerobic growth condi- complex (complex II) is not a coupling site.
tions.15 As an example of a protective role that The ratio of protons translocated for every
an oxidase might play, consider the situation two electrons varies, depending upon the com-
with the strict aerobe Azotobacter vinelan- plex. A consensus value is 10 protons extruded
dii. This organism has a branched respiratory for every two electrons that travel from NADH
chain leading to cytochromes bo and bd as the to oxygen. The bc1 complex translocates four
terminal oxidases. It also fixes nitrogen aerobi- protons for every two electrons; depending
cally. However, as with other nitrogenases, the upon the reported value, complexes I and IV
Azotobacter nitrogenase is inactivated by oxy- translocate 2 to 4 protons for every two elec-
gen. Azotobacter employs two mechanisms to trons. (The consensus value for mitochondrial
protect its nitrogenase, respiratory and confor- complex I is 4H+/2e–.) During reversed electron
mational. (See Section 13.3.2 for a discussion of flow, protons enter the cell through coupling
this point.) It is thought that the rapid consump- sites 1 and 2, driven by the ∆p, and the electrons
tion of oxygen by a terminal oxidase maintains are driven toward the lower redox potential.
the intracellular oxygen levels low enough to This creates a positive ∆E at the expense of the
prevent the inactivation of the nitrogenase. In ∆p. (See eq. 4.17.) Coupling site 3 is not physi-
agreement with this suggestion is the report that ologically reversible. Thus, water cannot serve
mutants of A. vinelandii that are deficient in the as a source of electrons for NAD+ reduction
cytochrome bd complex failed to fix nitrogen in by means of reversed electron flow. However,
air, although they did fix nitrogen when the oxy- during oxygenic photosynthesis, light energy
gen tension was sufficiently reduced.16 (Mutants can drive electrons from water to NADP+.
in cytochrome bo can fix nitrogen in air.) The mechanism of photoreduction of NADP+,
The adaptability of the bacteria with respect to which is different from reversed electron flow,
their electron transport chains can also be seen in is discussed in Chapter 6.
many bacteria that can respire either aerobically
or anaerobically. Under anaerobic conditions 5.5.1 The identification of coupling sites
they do not make the oxidase complexes but For an understanding of the physiology of
instead synthesize reductases. For example, energy metabolism during electron transport,
during anaerobic growth, E. coli synthesizes it is necessary to study the mechanism of pro-
fumarate reductase, nitrate reductase, and ton translocation, and for this the coupling
trimethylamine-N-oxide (TMAO) reductase. sites must be identified and isolated. The cou-
The different reductases enable the bacte- pling sites can be identified by the use of elec-
ria to utilize alternative electron acceptors tron donors that feed electrons into the chain
under anaerobic conditions. (Some facultative at different places, followed by measuring the
Free ebooks ==>
electron transport 157

amount of ATP made for every two electrons must occur between ubiquinone and oxygen.
transferred through the respiratory chain. The When electrons enter the respiratory chain
number of ATPs made for every 2e– transfer to after the bc1 complex, the P/O ratio is reduced
oxygen is called the P/O ratio. It is equal to the to 1, indicating that the bc1 complex is the sec-
number of ATP molecules formed per atom of ond coupling site and that site 3 is cytochrome
oxygen taken up. When an electron acceptor aa3 oxidase. Site 3 can be demonstrated by
other than oxygen is used, P/2e– is substituted bypassing the bc1 complex. The bc1 complex
for P/O. The P/O ratio is approximately equal can be bypassed via an artificial electron donor
to the number of coupling sites. (However, see that will reduce cytochrome c [e.g., ascorbate
Section 5.5.2.) and tetramethylphenylenediamine (TMPD)],
In mitochondria, the oxidation of NADH thus channeling electrons from ascorbate to
results in a P/O ratio of about 3, indicating that TMPD to cytochrome c to cytochrome aa3.
three coupling sites exist between NADH and Alternatively, one can simply use reduced cyto-
O2. The use of succinate as an electron donor chrome c as an electron donor to directly reduce
results in a P/O ratio of approximately 2. Since cytochrome aa3.
electrons from succinate enter at the ubiqui- Each coupling site is characterized by a drop
none level, this indicates that coupling site 1 in midpoint potential of about 200 mV, which
occurs between NADH and ubiquinone [i.e., is sufficient for generating the ∆p (Fig. 5.9). The
the NADH–ubiquinone oxidoreductase reac- size of the ∆p that can be generated with respect
tion (Fig. 5.7)]. The other two coupling sites to the ∆E was discussed in Section 4.7.

Fig. 5.9 Average midpoint potentials, E′m, for components of the mitochondrial respiratory chain. The com-
plexes are in dashed boxes. The actual potentials (Eh) for most of the components are not very different from
the midpoint potentials. An exception is cytochrome aa3, whose Eh is much more positive than its midpoint
potential. There are changes in potential of 200 mV or more at three sites, which drive proton translocation.
One site is within complex I, a second within complex III, and the third between complex IV and oxygen.
Source: Adapted from Nicholls, D. G., and S. J. Ferguson. 1992. Bioenergetics 2. Academic Press, London.
Free ebooks ==>
158 the physiology and biochemistry of prokaryotes

5.5.2 The actual number of ATPs that can in mitochondria is 6. Therefore, the maximal
be made for every two electrons traveling P/O ratio for this segment of the electron trans-
through the coupling sites port chain may be 6/4, or 1.5. The P/O ratios
By definition, the P/O ratio is the number of in bacteria can be higher, since a proton is not
moles of ATP made at a coupling site per two required to bring Pi into the cell. One significant
electrons that pass through the coupling site. aspect of branched aerobic respiratory chains
It is called the P/2e− ratio when oxygen is not in bacteria is that the number of coupling sites,
the terminal electron acceptor. The P/2e− ratio and therefore H+/O, can differ in the branched
does not need to be a whole number, since the chains. Thus, the different branches are not
amount of ATP that can be made per coupling equally efficient in generating a ∆p or making
site is equal to the ratio of protons extruded at ATP. We will return to this point later.
the coupling site to protons that reenter via the Of course, an ATP can be made when three
ATP synthase (Fig. 5.10). The ratio of protons protons enter via the ATP synthase only if the
translocated for every two electrons traveling ∆p is sufficiently large. As an exercise, we can
through all the coupling sites to O2 is called ask how large the ∆p must be. Recall that y∆p
the H+/O ratio. It can be measured by admin- is the work that can be done in units of electron
istering a pulse of a known amount of oxygen volts when y protons traverse a proton potential
to an anaerobic suspension of mitochondria or of ∆p volts. If H+/ATP is 3, then the number of
bacteria and measuring the initial efflux of pro- electron volts made available by proton influx
tons with a pH electrode as the small amount of through the ATP synthase is 3∆p. How many
oxygen is used up. The experiment requires that electron volts is needed to synthesize an ATP?
valinomycin plus K+ or a permeant anion such as The free energy of formation of ATP at physi-
thiocyanate, SCN–, be in the medium to prevent ological concentrations of ATP, ADP, and Pi is
a ∆Ψ from developing. (See Section 4.2.2 for a the phosphorylation potential, ∆Gp, which is
discussion of this point.) The reported values approximately 45,000 to 50,000 J. (See note 20
for H+/O for NADH oxidation vary. However, for an explanation of the phosphorylation
there is consensus that the true value is proba- potential.) Dividing by the Faraday constant
bly around 10. The ratio of protons entering via (96,500 C/mol) expresses the energy required
the ATP synthase to ATP made is called the H+/ to synthesize an ATP in electron volts. For
ATP ratio. It can be measured by using inverted 45,000 J this is 0.466 eV. Therefore, 3∆p must
submitochondrial particles prepared by sonic be greater than or equal to 0.466 eV. Thus, the
oscillation. These particles have the ATP syn- minimum ∆p is 0.466/3 or –0.155 V. Values of
thase on the outside and will pump protons ∆p that approximate this are easily generated
into the interior upon addition of ATP. Similar during electron transport. (However, see the
inverted vesicles can be made from bacteria by discussion in Section 4.10 regarding the low ∆p
first enzymatically weakening or removing the in alkaliphiles.)
cell wall and breaking the spheroplasts or pro-
toplasts by passage through a French press at
high pressures. (See note 19 for how to make
right-side-out vesicles.) Values of H+/ATP from
2 to 4 have been reported, and a consensus value
of 3 can be used for calculations. For intact
mitochondria, an additional H+ is required to
bring Pi electroneutrally from the cytosol into
the mitochondrial matrix in symport with H+,
so H+/ATP would be 4. A value of 10 for H+/O
predicts a maximum P/O ratio of 2.5 (10/4) for
mitochondria. This means that the often-stated
value of 3 for a P/O ratio for NADH oxidation
by mitochondria may be too high. Fig. 5.10 The ratio of protons extruded to protons
The number of protons ejected for every translocated through the ATPase determines the
2e– traveling between succinate and oxygen amount of ATP made.
Free ebooks ==>
electron transport 159

5.6 How a Proton Potential Might Be and quinones) and those that carry only elec-
Created at the Coupling Sites: Q Loops, trons (iron–sulfur proteins and cytochromes).
This is illustrated in Fig. 5.11. The electron
Q Cycles, and Proton Pumps
carriers and their sequence are flavoprotein
The preceding discussion points out that pro- (H carrier), FeS protein (e– carrier), quinone
ton translocation takes place at coupling sites (H carrier), and cytochromes (e– carriers). The
when electrons travel “downhill” over a poten- flavoprotein and FeS protein comprise the
tial gradient of at least 200 mV. (See Fig. 5.9.) NADH dehydrogenase, which can be a cou-
However, the mechanism by which the redox pling site (Section 5.6.3). When electrons are
reaction is actually coupled to proton transloca- transferred from the FeS protein to quinone
tion was not explained. This coupling is thought (Q) on the inner side of the membrane, two
to occur (1) in a Q loop or Q cycle, where H+ is protons are acquired from the cytoplasm.
picked up on the cytosolic side of the membrane According to the model, the reduced quinone
and carried as part of UQH2 and released as (QH2), called quinol, then diffuses to the outer
H+ on the outside surface of the opposite side membrane surface and becomes oxidized,
of the membrane, and (2) by means of a proton releasing the two protons. The electrons then
pump where H+ actually is pumped across the return via cytochromes to the inner membrane
membrane. Even when protons (H+) per se are surface, where they reduce oxygen. This would
not translocated as charged molecules, but are create a ∆Ψ because the protons are left on the
translocated at part of UQH2 during the Q loop outer surface of the membrane as the electrons
or Q cycle, a membrane potential is created, as move electrogenically across the membrane to
will be explained next. the inner surface. Thus, quinol oxidation is a
In the Q loop or Q cycle, quinones are reduced second coupling site.
on the inner membrane surface and carry hydro- As mentioned earlier, the energy to create the
gen across the membrane and becomes oxidized, membrane potential is derived from the ∆Eh
releasing protons on the external face of the between the oxidant and the reductant. One
membrane. Then, the electrons return electro- way to view this is that the energy from the ∆Eh
genically via electron transport carriers to the “pushes” the electron to the negative membrane
inner membrane surface creating a membrane potential on the inside surface. Note that the role
potential. Proton translocation in the Q loop or of the quinone is to ferry the hydrogens across
Q cycle is referred to as scalar translocation. the membrane, by diffusing from a reduction
Proton pumps are electron carrier proteins site on the cytoplasmic side of the membrane to
that couple electron transfer to the electrogenic an oxidation site on the outer side. For the pur-
translocation of protons through the membrane pose of calculating the expected ∆p generated at
creating a membrane potential. Such trans- a coupling site, it makes no difference whether
location of protons through the membrane is one postulates the transmembrane movement
referred to as vectorial. of protons in one direction or electrons in the
Although the two mechanisms, Q loop or opposite direction, since both carry the same
Q cycle and proton pump, are fundamentally charge (eq. 4.17).
different, the result is the same: the net trans-
location of protons across the membrane with 5.6.2 The Q cycle
the establishment of a ∆p. The ∆p is the same
For a review, see ref. 22. Although the linear Q
regardless of whether the moving charges are
loop as just described for the oxidation of qui-
electrons or protons, since both carry the same
nol accurately describes quinol oxidation in E.
charge. The relationship between the ∆p and
coli and some other bacteria, it is inconsistent
the ∆Eh was given by eq. 4.17.
with experimental observations of electron
transport in mitochondria, chloroplasts, and
5.6.1 The Q loop many bacteria. For example, the Q loop pre-
For a review, see ref. 21. The essential fea- dicts that the ratio of H+ released per reduced
ture of the Q loop model is that the electron ubiquinone (UQH2 ) oxidized is 2, whereas
carriers alternate between those that carry the measured ratio in mitochondria and many
both hydrogen and electrons (flavoproteins bacteria is actually 4. To account for the extra
Free ebooks ==>
160 the physiology and biochemistry of prokaryotes

Fig. 5.11 Proton translocation showing a Q loop and a proton pump. It is proposed that the electron carriers
exist in an alternating sequence of the following hydrogen [H] and electron [e–] carriers: flavoprotein (fp),
iron–sulfur protein (FeS), quinone (Q), and cytochromes. Oxidation of the flavoprotein deposits protons
on the outer membrane surface. Bacteria contain two types of NADH dehydrogenase. One of these, called
NDH-1, is a proton pump and, as in mitochondria, deposits four protons on the outer membrane surface. See
Section 5.6.3 for a further discussion. The electrons return to the inner membrane surface, where a quinone
is reduced, taking up two protons from the cytoplasm. The reduced quinone diffuses to the outer surface of
the membrane, where it is oxidized, depositing two more protons on the surface. The electrons return to the
cytoplasmic surface via cytochromes, where they reduce oxygen in a reaction that consumes protons. Some
cytochrome oxidases function as proton pumps. During anaerobic respiration the cytochrome oxidase is
replaced by a reductase, and the electrons reduce some other electron acceptor (e.g., nitrate or fumarate). It
should be noted that electrons can also enter at the level of quinone (e.g., from succinate dehydrogenase).

two protons, Peter Mitchell suggested a new anion UQ– during step 1 (Fig. 5.12 stage I), and
pathway for the oxidation of quinol called the the two protons are released to the membrane
Q cycle.23 The Q cycle operates in an enzyme surface. One electron travels to the FeS protein,
complex called the bc1 complex (complex III). and from there to cytochrome c1 on its way to
The bc1 complex from bacteria contains three the terminal electron acceptor. Thus, the ratio
polypeptides. These are cytochrome b with two of protons released at the membrane surface
b-type hemes, an iron–sulfur protein contain- to electrons traveling to the terminal electron
ing a single 2Fe–2S cluster (the Rieske protein), acceptor is 2:1 (stage 1, step 1, Fig. 5.12). The
and cytochrome c1 with one heme. The complex second electron is removed during step 2 and
spans the membrane and has a site for binding transferred via steps 2 and 3 across the mem-
reduced ubiquinone, UQH2, on the outer sur- brane to the N surface (inside surface) via
face of the membrane called site P (for posi- cytochromes bL and bH. This actually creates
tive), and a second site on the inner surface for a membrane potential, inside surface nega-
binding UQ, site N (for negative) (Fig. 5.12). tive, because it is a moving negative charge.
At site P, UQH2 is oxidized to the semiquinone The result is that UQ− becomes UQ, which
Free ebooks ==>
electron transport 161

A 2H
Fe/S c1 terminal
1 2
P site

pool 3 pool

from bH

N site


B 2H
Fe/S c1 terminal
1 2
P site

pool 3
from bH

N site


Fig. 5.12 The bc1 complex. The bc1 complex isolated from bacteria contains three polypeptides, which
are a cytochrome b containing two heme b groups, bH and bL, an iron–sulfur protein, Fe/S (the Rieske pro-
tein), and cytochrome c1. It is widely distributed among the bacteria, including the photosynthetic bacteria.
(Mitochondrial bc1 complexes are similar but contain an additional 6–8 polypeptides without prosthetic
groups.) The iron–sulfur protein and cytochrome c1 are thought to be located on the outside (positive, or P)
surface of the membrane, and the cytochromes bH and bL are believed to span the membrane, acting as an elec-
tron conductor. On the outer surface there is a binding site in the bc1 complex for ubiquinol, UQH2 (the P site).
On the inner surface (negative, or N) there is a binding site in the bc1 complex for UQ and UQ− (the N site).
UQ binds to the P site and one electron is removed, forming the semiquinone anion, UQ−, (reaction 1). At this
time two protons are released on the outer membrane surface. The electron that is removed is transferred to
the iron–sulfur protein and from there to cytochrome c1 (step 5). (Cytochrome c1 transfers the electron to the
terminal electron acceptor via a series of electron carriers in other reactions.) In reaction 2 the second electron
is removed from UQ−, producing the fully oxidized quinone, UQ. The second electron is transferred trans-
membrane via cytochrome b to UQ at site N (steps 3 and 4) on the opposite side of the membrane. Because
the electron travels transmembrane, a membrane potential is created, outside positive (P). The result is that
UQH2 is oxidized to UQ, which enters the UQ pool. The second electron that is transferred transmembrane
via cytochrome b generates a membrane potential, and also reduces a UQ from the UQ pool to UQ−. A second
UQH2 is oxidized at the P site, releasing two more protons, and the electron that is transferred transmem-
brane reduces UQ– at the N site to UQH2. The two protons are acquired from the cytoplasmic surface of the
cell membrane. The net result is that for every electron that travels to the terminal electron acceptor at the P
site, two protons are released to the outer membrane surface so the ratio of protons released to electrons that
travel to the terminal electron acceptor is 2:1 as opposed to 1:1 when UQH2 is oxidized via the Q loop shown
in Fig. 5.11 (site 2).
Free ebooks ==>
162 the physiology and biochemistry of prokaryotes

enters the UQ pool (stage I). In step 4 (stage I) 5.6.3 Pumps

UQ molecules in the UQ pool can accept elec- Proton pumps also exist. These catalyze the elec-
trons that travel across the membrane via cyto- trogenic translocation of protons (H+) across the
chromes b (step 4) and become reduced to UQ−. membrane rather than electrons (Fig. 5.11). For
Now focus on stage II. Each UQ− can accept an example, proton extrusion accompanies the
electron (step 4) on the N side of the membrane cytochrome aa3 oxidase reaction when cyto-
and pick up 2H+ from the cytosol (step 6) to chrome c is oxidized by oxygen in mitochon-
become UQH2, which can then be oxidized in dria and some bacteria. This can be observed
stage 1 (step 1). The net result is that for every by feeding electrons into the respiratory chain
2H+ and 2 electrons removed from UQH2, one at the level of cytochrome c, thus bypassing
electron is recycled across the membrane so the quinone. The experimental procedure is to
that the net result is 2H+ released per electron incubate the cells in lightly buffered anaerobic
that travels through the respiratory chain to the media with a reductant for cytochrome c and a
terminal electron acceptor. But, as mentioned, permeant anion (e.g., SCN– or valinomycin plus
there is also another very important result of K+). Changes in pH are measured with a pH elec-
this pathway. Electron flow from the QP site trode. Upon addition of a pulse of oxygen, given
to the QN site is transmembrane and creates a as an air-saturated salt solution, a sharp, tran-
membrane potential, which contributes most sient acidification of the medium occurs, and the
of the energy to the ∆p at that site. The situ- ratio of H+ to O can be calculated. If the electron
ation with respect to released protons can be donor itself releases protons, these scalar pro-
summarized as follows: tons are subtracted from the total protons trans-
located to determine the number of vectorially
P site: 2UQH2 + 2cyt cox translocated protons (due to proton pumping by
→ 2UQ + 2cyt cred + 4H+out + 2e− the cytochrome c oxidase). For example, when
cytochrome c is reduced by means of ascor-
N site: UQ + 2e− + 2H+in → UQH2 bate plus TMPD, the ascorbate is oxidized to
Sum: UQH2 + 2cyt cox + 2H+in dehydroascorbate with the release of one scalar
proton for every two electrons. Any additional
→ UQ + 2cyt cred + 4H+out protons released are due to proton pumping by
the cytochrome c oxidase. When similar experi-
Since the Q cycle translocates four protons for ments were done with Paracoccus denitrificans,
every two electrons that flow to the terminal it was found that the P. denitrificans cytochrome
electron acceptor, it generates a larger pro- aa3 oxidase translocates protons with a stoichi-
ton current than the Q loop that translocates ometry of 1H+/1e–.24,25
only one proton per electron. This can result Cytochrome oxidase pumping activity can
in more ATP synthesis. Consider the situation also be demonstrated in proteoliposomes made
of an ATP synthase that requires the influx of with purified cytochrome aa3. Because the pro-
three protons to make one ATP. If the trans- ton pumps move a positive charge across the
fer of a mole of electrons through the electron membrane, leaving behind a negative charge, a
transport pathway resulted in the transloca- membrane potential, outside positive, develops.
tion of one mole of protons, then a third of The membrane potential should be the same as
a mole of ATP could be made for each mole that recorded when an electron moves inward,
of electrons if the protons reentered via the since the proton and the electron carry the same
ATP synthase. On the other hand, if a mole charge (i.e., 1.6 × 10–19 C). The mechanism of
of electrons transported resulted in the trans- pumping probably requires conformational
location of two moles of protons, then two- changes in cytochrome oxidase resulting from
thirds of a mole of ATP could be made when its redox activity. Conformational changes
the protons reentered via the ATP synthase. probably also occur in bacteriorhodopsin dur-
In other words, the size of the proton cur- ing proton pumping (Section 4.8.4).
rent generated by respiration determines the The mitochondrial NADH dehydrogenase
upper value of the amount of ATP that can complex (NADH:ubiquinol oxidoreductase)
be made. translocates four protons for every NADH
Free ebooks ==>
electron transport 163

oxidized.26 Two types of NADH:ubiquinol oxi- not complete, but it conveys the diversity of
doreductases exist in bacteria.8–11 One of these, electron transport systems found in bacteria.
called NDH-1, is similar to the mitochondrial The electron transport chains for the respira-
complex I in that it is a multisubunit enzyme tory metabolism of ammonia, nitrite, inorganic
complex (approximately 14 polypeptide sub- sulfur, and iron are described in Chapter 15.
units) consisting of FMN and FeS clusters, and
it translocates protons across the membrane 5.7.1 Escherichia coli
during NADH oxidation (4H+/2e–). The mech- E. coli is a gram-negative heterotrophic faculta-
anism of proton translocation is not known, tive anaerobe. It can be grown aerobically by
but it is referred to as a proton pump. [The using oxygen as an electron acceptor, anaerobi-
marine bacterium Vibrio alginolyticus has a cally (e.g., by using nitrate or fumarate as the
sodium-translocating NDH (Na–NADH). See electron acceptor), or anaerobically via fer-
Section 4.7.1.] mentation (Chapter 15). The bacteria adapt to
A second NADH:ubiquinol oxidoreductase, their surroundings, and the electron transport
called NDH-2, may also be present. NDH-2 system that the cells assemble reflects the elec-
differs from NDH-1 in consisting of a single tron acceptor that is available. (See ref. 29 for
polypeptide and FAD, and in that it is not an a review.)
energy-coupling site. In E. coli, NDH-1 and All the electron transport chains present in
NDH-2 are simultaneously present.27 How E. coli branch at the level of quinone, which
E. coli regulates the partitioning of electrons connects the different dehydrogenases with
between the two NAD dehydrogenases clearly the various terminal reductases and oxidases
has important energetic consequences. that are present under different growth condi-
tions. E. coli makes three quinones, ubiquinone
(UQ), menaquinone (MQ), and demethylme-
5.7 Patterns of Electron Flow in naquinone (DMQ), and their relative amounts
Individual Bacterial Species depend upon the nature of the electron acceptor.
Although the major principles of electron trans- When UQ is growing aerobically, it accounts
port as outlined already apply to bacteria in for 60% of the total quinone, DMQ is 37% of
general, several different patterns of electron the total, and MQ is only about 3%. However,
flow exist in particular bacteria, often within very little UQ is made anaerobically. For anaer-
the same bacterium grown under different con- obic growth on nitrate, the major quinones
ditions.28 The patterns of electron flow reflect are DMQ (70%) and MQ (30%). The major
the different sources of electrons and electron quinone grown anaerobically on fumarate or
acceptors that are used by the bacteria. For DMSO is MQ (74%), with DMQ (16%) and
example, bacteria may synthesize two or three UQ (10%) contributing the rest.28 There is no
different oxidases in the presence of air and bc1 complex and no cytochrome c, which may
several reductases anaerobically. In addition, serve as additional branch points in other bacte-
certain dehydrogenases are made only anaero- rial respiratory chains (e.g., see the discussion of
bically because they are part of an anaerobic Paracoccus denitrificans in Section 5.7.2).
respiratory chain. Furthermore, whereas elec-
tron donors such as NADH and FADH2 gen- Aerobic respiratory chains
erated in the cytoplasm are oxidized inside the When grown aerobically, E. coli makes two dif-
cell, there are many instances of oxidations ferent quinol oxidase complexes, cytochrome
in the periplasm in gram-negative bacteria. bo complex (has heme b and heme o and is also
Substances that are oxidized in the periplasm called bo3 or cytochrome o) and cytochrome
include hydrogen gas, methane, methanol, bd complex (has heme b and heme d and is also
methylamine, formate, perhaps ferrous ion, called cytochrome d), resulting in a branched
reduced inorganic sulfur, and elemental sulfur. respiratory chain to oxygen (Fig. 5.13).30–34 In
In most of these instances periplasmic cyto- these pathways electrons flow from ubiquinol
chromes c accept the electrons from the electron to the terminal oxidase complex, which is why
donor and transfer them to electron carriers in the oxidases are called (ubi)quinol oxidases.
the membrane. The discussion that follows is The cytochrome bo complex from E. coli is a
Free ebooks ==>
164 the physiology and biochemistry of prokaryotes

Fig. 5.13 Aerobic respiratory chain in E. coli. The chain branches at the level of ubiquinone (UQ) to two
alternate quinol oxidases, cytochrome bo and bd. Cytochrome bd complex has a higher affinity for oxygen
and is synthesized under low oxygen tensions, where it becomes the major route to oxygen. Proton transloca-
tion occurs during NADH oxidation, catalyzed by NADH:ubiquinone oxidoreductase (also called NDH-1),
a rotenone-sensitive enzyme complex, and quinol oxidation, catalyzed by cytochrome bo complex or cyto-
chrome bd complex. Additionally, cytochrome bo is a proton pump. E. coli has a second NADH dehydro-
genase, called NADH-2, which does not translocate protons and is not sensitive to rotenone. Therefore, the
number of protons translocated per NADH oxidized can vary from two (NADH-2 and bd complex) to eight
(NADH-1 and bo complex).

proton pump, with a stoichiometry of 1H+/e–. of transcription of the NADH dehydrogenase

Cytochrome bo complex is the predominant genes in E. coli is complex.34,36
oxidase when the oxygen levels are high. When
the oxygen tensions are lowered, E. coli makes Physiological significance of alternate
more cytochrome bd than cytochrome bo. electron routes that differ in the number
The cytochrome bd complex has a higher of coupling sites
affinity for oxygen (lower Km) than does the Since the NDH-1 dehydrogenase may trans-
cytochrome bo complex, which may be why it locate as many as four protons for every two
is the dominant oxidase under low oxygen ten- electrons, whereas the NDH-2 dehydrogenase
sions. Cytochrome bd is not a proton pump. In translocates none, the number of protons trans-
the same sense that the bc1 complex is a cou- located per NADH oxidized can theoretically
pling site because it catalyzes the oxidation of vary from 2 (NDH-2 and bd complex) to 8
quinol resulting in the scalar extrusion of pro- (NDH-1 and bo complex). Assuming a H+/ATP
tons, both the bo and bd complexes are cou- of 3 (i.e., the ATP synthase translocates inwardly
pling sites. The difference between the two, as 3 protons for every ATP made), the ATP yields
mentioned, is that the bo complex is also a pro- per NADH oxidized can vary fourfold from
ton pump and catalyzes the vectorial extrusion 2/3, or 0.67, to 8/3, or 2.7. It also means that
of protons. The oxidation of quinol by the bo E. coli has great latitude in adjusting the ∆p gen-
complex results in two protons translocated for erated during respiration. Since a large ∆p can
every electron (one scalar and one vectorial), drive reversed electron transport and thus slow
whereas the oxidation of quinol by the bd com- down oxidation of NADH and quinol, it may
plex results in only one proton translocated per be an advantage to be able to direct electrons
electron (scalar). along alternate routes that bypass coupling
E. coli has two NADH dehydrogenases, sites and translocate fewer protons. This could
NDH-1 and NDH-2, encoded by the nuo ensure adequate rates of reoxidation of NADH
operon and ndh, respectively. Only NDH-I is and quinol.
a coupling site (proton pump). It is a complex
enzyme consisting of 14 subunits and translo- Anaerobic respiratory chains
cates two protons per electron. As discussed In the absence of oxygen, E. coli can use either
in Section 5.6.3, NDH-1 is similar to com- nitrate or fumarate as an electron acceptor.
plex I of mitochondria. It has been shown that The nitrate is reduced to nitrite, which is fur-
NDH-1 is used during fumarate respiration, ther reduced to ammonia, and the fumarate is
whereas NDH-2 is used primarily during aer- reduced to succinate, all of which are excreted
obic and nitrate respiration.35 The regulation into the medium.37 (See note 38 for a further
Free ebooks ==>
electron transport 165

Fig. 5.14 Anaerobic respiratory chains in E. coli. When oxygen is absent, E. coli synthesizes any one of sev-
eral membrane-bound reductase complexes depending upon the presence of the electron acceptors. Nitrate
induces the synthesis of nitrate reductase and represses the synthesis of the other reductases. Menaquinone
(E0′ = –74 mV) (or demethylmenaquinone, E0′ = –40 mV ) must be used to reduce some of the reductases
(e.g., fumarate reductase, because it has a sufficiently low midpoint potential). Ubiquinol (E0′ = +100 mV )
or menaquinone can reduce nitrate reductase because the E0′ of nitrate is 421 mV. Each reductase may be a
complex of several proteins and prosthetic groups through which the electrons travel to the terminal electron
acceptor. The transfer of electrons from the dehydrogenases to the reductases results in the establishment
of a proton potential. If the dehydrogenase has site 1 activity, there can theoretically be two coupling sites,
one at the dehydrogenase step and one linked to quinol oxidation at the reductase step. Abbreviations: cyt b,
cytochrome b; Fe/S, nonheme iron–sulfur protein; FAD, flavoprotein with flavin adenine dinucleotide as the
prosthetic group: Mo, molybdenum; TMANO, trimethylamine N-oxide; DMSO, dimethyl sulfoxide; MQ,
menaquinone; UQ, ubiquinone.

description of fumarate and nitrate reduction.) is zero, indicating that both the release of pro-
The anaerobic respiratory chains consists of a tons when MQH2 is oxidized and the uptake of
dehydrogenase, a reductase, and a diffusible protons when fumarate is reduced take place
quinone to mediate the transfer of electrons on the cytoplasmic side of the membrane. (It
between the dehydrogenase and the reductase does not appear that either fumarate reductase
(Fig. 5.14). or any of the nitrate reductases are proton
The number of coupling sites depends upon pumps.) Thus, for E. coli to make ATP during
whether the electron acceptor is nitrate or the oxidation of NADH by fumarate, it must
fumarate.28 When nitrate is the electron accep- use NDH-1 rather than NDH-2, since only the
tor, there can be two coupling sites: that is, one former is a coupling site. In agreement with this
at the dehydrogenase step if NDH-1 is used conclusion, it has been found that the genes for
(site 1) and one at the quinol oxidation step [i.e., NDH-1 (nuo genes) are essential for anaerobic
the nitrate reductase (scalar)]. During nitrate respiration with fumarate as the electron accep-
however, respiration, NDH-2, which is not a tor. (Reviewed in ref. 39.)
coupling site, is also used.
Quinol oxidation by nitrate results in a ∆p 5.7.2 Paracoccus denitrificans
because the oxidation of one quinol and release P. denitrificans is a nonfermenting gram-
of two protons takes place on the periplasmic negative facultative anaerobe that can obtain
side of the membrane, and the uptake of two energy from either aerobic respiration or
protons during the reduction of one nitrate to nitrate respiration.20,40,41 It is found primar-
nitrite takes place on the cytoplasmic side. Thus, ily in soil and sewage sludge. This bacterium
the oxidation of quinol by nitrate catalyzed by can grow heterotrophically on a wide variety
nitrate reductase yields a proton-to-electron of carbon sources, or autotrophically on H2
ratio of 1.28 See the discussion of the reaction and CO2 under anaerobic conditions using
for nitrate reductase in Section 5.7.2. nitrate as the electron acceptor. It can be iso-
However, the H+/e– ratio for quinol oxidation lated from soil by anaerobic enrichment with
by fumarate in E. coli or Wolinella succinogenes media containing H2 as the source of energy
Free ebooks ==>
166 the physiology and biochemistry of prokaryotes

and electrons, Na2CO3 as the source of carbon, Bradyrhizobium.42 Cytochrome bb3 formerly
and nitrate as the electron acceptor. Electron was called cytochrome o or b0. But no heme o was
transport in P. denitrificans receives a great detected when the hemes were extracted from
deal of research attention because certain fea- membranes and analyzed by reversed-phase high
tures closely resemble electron transport in performance liquid chromatography24,43,44). The
mitochondria. cytochrome aa3 and cytochrome bb3 are proton
pumps. (Whether cytochrome cbb3 pumps pro-
Aerobic pathway tons apparently depends upon the assay con-
P. denitrificans differs from E. coli in that it ditions.39) The aerobic pathways as well as the
has a bc1 complex and a cytochrome aa3 oxi- sites of proton translocation are shown in Fig.
dase (cytochrome c oxidase), and in this way 5.15A. Electrons traveling from NADH to oxy-
resembles mitochondria. In addition to the gen can pass through as many as three coupling
cytochrome aa3, there are two other terminal sites (NDH-1, bc1 complex, cyt aa3 or perhaps
oxidases in the aerobic pathway.24,39 These are cyt cbb3) or as few as two coupling sites (NDH-1
a different cytochrome c oxidase (cytochrome and cyt bb3). Recall that the bc1 complex and
cbb3) and a ubiquinol oxidase (cytochrome cyt bb3 are coupling sites because they oxidize
bb3). (Cytochromes cbb3 are heme–copper quinol, whereas the NDH-1 and cytochrome c
oxidases found in several bacteria including oxidases are proton pumps. As described later,
Thiobacillus, Rhodobacter, Paracoccus, and P. denitrificans also oxidizes methanol, and in

Fig. 5.15 A model for electron transport pathways in Paracoccus denitrificans. (A) Aerobic. The pathway has
two branch points. One branch is at the level of ubiquinone, leading to one of two ubiquinol oxidases (i.e.,
the cyt bc1 complex or cyt bb3). Both these quinol oxidases are coupling sites. The bc1 complex extrudes two
protons per electron via the Q cycle. Whereas cyt bb3 is a proton pump and extrudes one proton per electron
vectorially, the second proton is extruded via a Q loop. A second branch point occurs at the level of the bc1
complex. Electrons can flow either to cyt aa3, which is a proton pump, or to cyt cbb3, which has been reported
to pump protons under certain experimental conditions. NDH-1 is also a coupling site. (B) Anaerobic. When
the bacteria are grown anaerobically, using nitrate as the electron acceptor, the cytochrome aa3 levels are very
low and the electrons travel from ubiquinone to nitrate reductase and also through the bc1 complex to nitrite
reductase, nitric oxide reductase, and nitrous oxide reductase.
Free ebooks ==>
electron transport 167

this case the electrons enter at a cytochrome c, of the membrane and reduces nitrate on the
thus bypassing the bc1 site. cytoplasmic side. This would create a ∆Ψ as
two electrons from UQH2 flow electrogenically
across the membrane, leaving two protons on
Anaerobic pathway
the outside. In the cytoplasm, protons are taken
P. denitrificans can also grow anaerobically by up during nitrate reduction according to the fol-
using nitrate as an electron acceptor, reducing lowing reaction:
it to nitrogen gas in a process called denitrifica-
tion (Fig. 5.15B).36,45 During anaerobic growth 2H+ + 2e− + NO−3 → NO−2 + H2O
on nitrate, P. denitrificans has a complete citric
acid cycle in which electrons are donated to the The result is the net translocation of protons
electron transport chain; but the electron trans- to the outside, although only electrons moved
port chain is very different from that of aero- across the membrane, not protons. This is the
bic growth. The cells contain cytochrome bb3, same as the Q loop described in Section 5.6.1
nitrate reductase, nitrite reductase, nitric oxide except that the terminal electron acceptor is
reductase, and nitrous oxide reductase. (For a nitrate instead of oxygen. The electrons to the
more complete description of these enzymes, see other reductases flow from UQH2 through the
note 46.) The levels of cytochrome aa3 are very bc1 complex, and a ∆p is generated via the Q
low. The cytochrome c shown in Fig. 5.15 is cycle catalyzed by the bc1 complex which, as
periplasmic, although there is also cytochrome described in Section 5.6.1, is a modification of
c in the membrane associated with some of the the Q loop that results in the translocation of
electron carriers. The nitrate (NO3−) is reduced two protons per electron rather than one.
to nitrite (NO2−) in a two-electron transfer In P. denitrificans, Pseudomonas aeruginosa,
via a membrane-bound nitrate reductase. and many other facultative anaerobes, both the
The NO2− is reduced to nitric oxide (NO) in a synthesis and the activity of the denitrifying
one-electron transfer via a periplasmic nitrite enzymes are prevented by oxygen. However,
reductase. The NO is reduced to a half-mole in certain other facultative anaerobes, includ-
of nitrous acid (1/2N2O) in a one-electron step ing Comamonas spp., certain species of
via a membrane-bound nitric oxide reductase. Pseudomonas, Thiosphaera pantotropha, and
And the 1/2N2O is reduced to a half-mole of Alcaligenes faecalis, denitrifying enzymes are
dinitrogen (1/2N2) in a one-electron step by a made and are active in the presence of oxy-
periplasmic nitrous oxide reductase. Thus to gen.47 In these systems, both oxygen and nitrate
reduce one mole of NO3− to 1/2N2 a total of five are used simultaneously as electron acceptors,
electrons flow in and out of the cell membrane although aeration can significantly decrease the
through membranous electron and periplasmic rate of nitrate reduction. The advantage of co-
electron carriers from ubiquinol to the various respiration using both oxygen and nitrate is not
reductases. obvious.
As shown in Fig. 5.15, the electron trans-
port pathway includes several branches to the Periplasmic oxidation of methanol
individual reductases. The first branch site is at Many gram-negative bacteria oxidize sub-
UQ, where electrons can flow either to nitrate stances in the periplasm and transfer the elec-
reductase or to the bc1 complex. Then there are trons to membrane-bound electron carriers,
three branches to the three other reductases often via periplasmic cytochromes c. (See
after the bc1 complex at the level of cytochrome Section 1.2.4 for a description of the periplasm.)
c. In agreement with the model, electron flow to An example is P. denitrificans, which can grow
nitrate reductase is not sensitive to inhibitors of aerobically on methanol (CH3OH) by oxidiz-
the bc1 complex, whereas electron flow to the ing it to formaldehyde (HCHO) and 2H+ with
other reductases is sensitive to the inhibitors. a periplasmic dehydrogenase, methanol dehy-
The nitrate reductase spans the membrane, and drogenase (Fig. 5.16):
it has been proposed that it creates a ∆p via a Q
CH3OH → HCHO + 2H+
loop, similar to that described in Section 5.6.1.
It is proposed that the nitrate reductase accepts (Growth on methanol is autotrophic, since the
electrons from UQH2 on the periplasmic side formaldehyde is eventually oxidized to CO2,
Free ebooks ==>
168 the physiology and biochemistry of prokaryotes

Periplasm cell membrane cytoplasm 5.7.3 Rhodobacter sphaeroides

Rhodobacter sphaeroides is a purple photo-
CH3OH HCHO + 2H + synthetic bacterium that can be grown photo-
heterotrophically under anaerobic conditions
methanol or aerobically in the dark. When growing aero-
dehydrogenase bically in the dark, the bacteria obtain energy
from aerobic respiration. The respiratory chain
2 x 1e-
resembles the mitochondrial and P. denitri-
cyt c551 1 O + 2H+ ficans respiratory chains in that it consists of
2 x 1e- 2 2 NADH:ubiquinone oxidoreductase (coupling
cyt c
H2O site 1), a bc1 complex (coupling site 2), and a
2H + 2H+ cytochrome aa3 complex (coupling site 3). As in
cyt aa3 E. coli, there are two NADH:ubiquinone oxi-
doreductases: NDH-1 and NDH-2.

Fig. 5.16 Oxidation of methanol by P. denitrificans. 5.7.4 Fumarate respiration in

Methanol is oxidized to formaldehyde by a periplas- Wolinella succinogenes
mic methanol dehydrogenase. The electrons are Fumarate respiration occurs in a wide range of
transferred to periplasmic cytochromes c and to a bacteria growing anaerobically.48 This is prob-
membrane-bound cytochrome aa3, which is also a ably because fumarate itself is formed from
proton pump. A ∆p is created as a result of the elec- carbohydrates and protein during growth.
trogenic influx of electrons and the electrogenic efflux We describe the electron transport pathway in
of protons, accompanied by the release of protons in W. succinogenes, a gram-negative anaerobe iso-
the periplasm (methanol oxidation) and uptake in lated from the rumen, by reference to Fig. 5.17.
the cytoplasm (oxygen reduction).
W. succinogens can grow at the expense of
H2 or formate, both produced in the rumen by
other bacteria. The electron transport pathway
which is assimilated via the Calvin cycle: see is shown in Fig. 5.17A. The active sites for both
Chapter 14.) The electrons are transferred the hydrogenase and the formate dehydroge-
from the dehydrogenase to c-type cytochromes nase are periplasmic, whereas the active site
in the periplasm. The cytochromes c transfer for the fumarate reductase is cytoplasmic. An
the electrons to cytochrome aa3 oxidase in the examination of the topology of the components
membrane, which reduces oxygen to water on of the respiratory chain reveals how a ∆p is
the cytoplasmic surface. A ∆p is established as generated.
a result of the inward flow of electrons and the
outward pumping of protons by the cytochrome Topology of the components of the
aa3 oxidase, as well as the release of protons in electron transport pathway
the periplasm during methanol oxidation and The electron transport chain consists of a
consumption in the cytoplasm during oxygen periplasmic enzyme that oxidizes the elec-
reduction. Since the oxidation of methanol tron donor, a membrane-bound menaquinone
bypasses the bc1 coupling site, the ATP yields (MQ) that serves as an intermediate elec-
are lower. In addition to methanol, the bacteria tron carrier, and membrane-bound fumarate
can grow on methylamine (CH3NH+3), which reductase, which accepts the electrons from
is oxidized by methylamine dehydrogenase to the menaquinone and reduces fumarate on the
formaldehyde, NH+4 , and 2H+: cytoplasmic side of the membrane (Fig. 5.17B).
Both the hydrogenase and the formate dehydro-
CH3NH+3 + H2O → HCHO + NH+4 + 2H+
genase are made of three polypeptide subunits,
Methylamine dehydrogenase is also located in two facing the periplasm and one an integral
the periplasm and donates electrons via cyto- membrane protein (cytochrome b). Note that
chromes c to cytochrome aa3, bypassing the bc1 cytochrome b not only serves as a conduit for
complex. electrons, but also binds the dehydrogenases
Free ebooks ==>
electron transport 169

Fig. 5.17 A model for the electron transport system of Wolinella succinogenes. (A) Electrons flow from H2
and formate through menaquinone (MQ) to fumarate reductase. (B) Illustration that the catalytic portions
of hydrogenase and formate dehydrogenase are periplasmic, whereas fumarate reductase reduces fumarate
on the cytoplasmic side. Electrons flow electrogenically to fumarate. A ∆p is created because of electrogenic
influx of electrons together with the release of protons in the periplasm and their consumption in the cyto-
plasm. Source: Modified from Kroger, A., V. Geisler, E. Lemma, F. Theis, and R. Lenger. 1992. Bacterial
fumarate respiration. Microbiology 158:311–314.

into the membrane. The two periplasmic sub- Electron flow and the establishment of a Δp
units of the hydrogenase are a Ni-containing Electron flow is from the dehydrogenase or
protein subunit and an iron–sulfur protein. The hydrogenase to cytochrome b to menaquinone
two periplasmic subunits of the formate dehy- to fumarate reductase. Two electrons are elec-
drogenase are a Mo-containing protein subunit trogenically transferred across the membrane
and an iron–sulfur protein. to fumarate for every H2 or formate oxidized,
The fumarate reductase is a complex contain- leaving two protons on the outside, thus estab-
ing three subunits. One subunit of the fumar- lishing a ∆p. In a study of whole cells, a value
ate reductase is a flavoprotein with FAD as the of 1.1 was obtained for the ratio of protons
prosthetic group (subunit A). A second subunit to electrons during fumarate reduction, sug-
has several FeS centers (subunit B). And the gesting perhaps that a mechanism of proton
third subunit has two hemes of the b type (sub- translocation through the membrane may also
unit C), which binds the fumarate reductase to exist.48 If one assumes that 1.1 is a correct num-
the membrane. Fumarate reductase is similar in ber, and also assumes a stoichiometry for the
structure to succinate dehydrogenase isolated ATP synthase of 3H+/ATP, then the theoretical
from several different sources, which catalyzes maximum number of ATPs that can be formed
the oxidation of succinate to fumarate in the cit- from the transfer of two electrons to fumarate is
ric acid cycle. 2.2/3, or 0.73. The actual number measured by
Free ebooks ==>
170 the physiology and biochemistry of prokaryotes

experimentation was 0.56. Note that the qui- used when oxygen levels are high, and to cyto-
none functions as an electron carrier between chrome d, which is used when oxygen becomes
the cytochromes b but does not take part in limiting. Other bacteria may have in addition,
hydrogen translocation across the membrane or instead, an electron transport pathway in
as in a Q loop or Q cycle. which electrons travel from reduced quinone
through a bc1 complex, to cytochrome aa3,
which reduces oxygen. This is the same as the
5.8 Summary electron transport pathway found in mitochon-
All electron transport schemes can be viewed as dria, although the carriers may not be identi-
consisting of membrane-bound dehydrogenase cal. The alternative branches may differ in the
complexes, such as NADH dehydrogenase or number of coupling sites, and this could have
succinate dehydrogenase, that remove electrons regulatory significance regarding the rates of
from their substrates and transfer the electrons oxidation of reduced electron carriers, as well
to quinones, which in turn transfer the electrons as ATP yields.
to oxidase or reductase complexes. The latter Another difference from mitochondria is that
complexes reduce the terminal electron accep- bacteria can have either aerobic or anaerobic
tors. In contrast to mitochondria, which all electron transport chains; or, as is the case with
have the same electron transport scheme, bac- facultative anaerobes such as E. coli, either can
teria differ in the details of their electron trans- be present, depending upon the availability
port pathways, although the broad outlines of of oxgyen or alternative electron acceptors. A
all such schemes are similar. In bacteria, the hierarchy of electron acceptors is used. For E.
dehydrogenase, oxidase, and reductase com- coli, oxygen is the preferred acceptor, followed
plexes are sometimes referred to as modules by nitrate, and finally fumarate.
because specific ones are synthesized under cer- With respect to cytochrome c oxidase, there
tain growth conditions and “plugged into” the are two classes. Cytochrome aa3 is the major class
respiratory pathway. For example, in faculta- and has been reported in many bacteria, includ-
tive anaerobes such as E. coli, the oxidase mod- ing Paracoccus denitrificans, Nitrosomonas
ules are synthesized in an aerobic atmosphere europaea, Pseudomonas AM1, Bacillus subti-
and the reductase modules under anaerobic lis, and Rhodobacter sphaeroides. A different
conditions. cytochrome c oxidase has been reported for
Other dehydrogenases besides NADH dehy- Azotobacter vinelandii, Rhodobacter capsulata,
drogenase and succinate dehydrogenase exist. R. sphaeroides, R. palustris, and Pseudomonas
These oxidize various electron donors (e.g., aeruginosa, as well as P. denitrificans.13,49
methanol, hydrogen, formate, H2, glycerol) Apparently, both classes of cytochrome c oxi-
and are located in the periplasm or the cyto- dase coexist in the same organism and serve as
plasm. The coenzyme or prosthetic groups for alternate routes to oxygen.
these soluble dehydrogenases vary (e.g., they The main energetic purpose of the respira-
may be NAD+ or flavin). The electrons from tory electron transport pathways is to convert
the various dehydrogenases are transferred to a redox potential (∆Eh) into a proton potential
one of the electron carriers (e.g., quinone, cyto- (∆p). This is done at coupling sites. A mem-
chrome) and from there to a terminal reductase brane potential is created by electrogenic influx
or oxidase. of electrons, leaving the positively charged
An important difference between elec- proton on the outside, or during electrogenic
tron transport chains in bacteria and those in efflux of protons during proton pumping, leav-
mitochondria is that the former are branched. ing a negative charge on the inside. Influx of
Branching can occur at the level of quinone or electrons occurs when oxidations take place
cytochrome. The branches lead to different oxi- on the periplasmic membrane surface or in
dases or reductases, depending upon whether the periplasm, and electrons move vectorially
the bacterium is growing aerobically or anaer- across the membrane to the cytoplasmic sur-
obically. Many bacteria, including E. coli, face, where reductions take place. This occurs
transfer electrons from reduced quinone to in two situations: (1) when the substrate (e.g.,
cytochrome o, which is the major cytochrome H2, methanol) is oxidized by dehydrogenases
Free ebooks ==>
electron transport 171

in the periplasm and the electrons move across 3. What are two features that distinguish
the membrane to reduce the electron acceptor the Q cycle from linear flow of electrons?
and (2) when quinones are reduced on the cyto- How can you verify that the Q cycle is
plasmic side of the membrane, diffuse across operating?
the membrane, and are oxidized on the outside
4. What is the relationship between H+/O, H+/
membrane surface.
ATP, and the number of ATPs that can be
Quinone oxidation can be via a Q loop or a Q
cycle. A Q loop is an electron transport pathway
in which a reduced quinone carries hydrogen to 5. What is the relationship between the ∆p and
the outside surface of the cell membrane and number of ATPs that can be synthesized?
releases two protons as a result of oxidation.
6. Draw a schematic outline of three ways in
The two electrons return to the inner surface,
which a cell might create a membrane poten-
via cytochromes, where they reduce an electron
tial during respiration. You can include both
acceptor. The inward transfer of the electrons is
periplasmic and cytoplasmic oxidations.
electrogenic (i.e., a membrane potential is cre-
ated). In the Q loop, which is a linear pathway 7. Assuming a ∆p of –200 mV and a H+/ATP of
of electron flow, a proton is released for every 3, calculate what the ∆Gp must be in joules
electron. per mole if ∆Gp is in equilibrium with ∆p.
The Q cycle, a more complicated pathway, Assume that the temperature is 30 °C, and
results in the release of two protons for every that ∆G′p is 37 kcal/mol. What will be the
electron. In the Q cycle, QH2 gives up two pro- ratio of ATP to [ADP][Pi]?
tons and two electrons, but one of the electrons
8. What are the similarities and differences
is recycled back to oxidized quinone. Thus, the
between bacterial respiratory pathways and
ratio of H+ to e– is 2, rather than 1. The Q cycle
the mitochondrial respiratory pathway?
is more efficient, since it results in the availabil-
ity of more protons for influx through the ATP 9. Calculate the maximum number of ATPs
synthase per electron, and thus can increase the that can be made per NADH oxidized by E.
yield of ATP. coli by using the following combinations:
In a second method of coupling oxidation– NDH-1 plus cytochrome bo and NDH-2
reduction reactions to the establishment of a plus cytochrome bd.
proton potential, some of the electron carriers
10. What drives reversed electron transport?
act as proton pumps. A membrane potential is
What experiment can be done to support
created when protons are pumped electrogeni-
this conclusion?
cally out of the cell. A well-established example
is cytochrome aa3 oxidase. However, there is 11. What is it about the sequential arrange-
evidence that other oxidases are also proton ment of electron carriers that makes pro-
pumps, including cytochrome bo in E. coli and ton translocation possible in a Q loop or Q
cytochromes bb3 and cbb3 in P. denitrificans. cycle?
Mitochondrial and certain bacterial NADH
12. What is the definition of the P/O ratio?
dehydrogenases are also coupling sites, and
P/2e− ratio?
these may function as proton pumps.


1. What is it about the solubility properties 1. Archaea—Molecular and Cellular Biology. 2007.
and electrode potentials of quinones that R. Cavicchioli (Ed.). ASM Press, Washington, DC.
make them suitable for their role in electron 2. Lubben, M. 1995. Cytochromes of archaeal
transport? electron transfer chains. Biochim. Biophys. Acta
2. Design an experiment that can quantify the 3. For example, the enzyme complex that oxidizes
H+/O for the cytochrome aa3 reaction in NADH and reduces quinone is an NADH–quinone
proteoliposomes. oxidoreductase. It consists of a flavoprotein and
Free ebooks ==>
172 the physiology and biochemistry of prokaryotes

iron–sulfur proteins. (Reduced quinone is called qui- 10. Ohnishi, T. 1993. NADH–quinone oxi-
nol.) An enzyme complex that oxidizes quinol and doreductase, the most complex complex. J. Bioenerg.
reduces cytochrome c is called a quinol–cytochrome Biomemb. 25:325–329.
c oxidoreductase. It consists of cytochromes and an
iron–sulfur protein. 11. Yagi, T., T. Yano, and A. Matsuno-Yagi. 1993.
Characteristics of the energy-transducing NADH–
4. Here are some definitions that are useful to quinone oxidoreductase of Paracoccus denitrificans
know. Cofactors are nonprotein molecules, either as revealed by biochemical, biophysical, and molecu-
metals or organic, bound with varying degrees of lar biological approaches. J. Bioenerg. Biomemb.
affinity to enzymes and required for enzyme activ- 25:339–345.
ity. Coenzymes are organic cofactors that shuttle
12. Complex II (succinic dehydrogenase) has four
back and forth between enzymes carrying electrons,
polypeptides, two of which are hydrophilic periph-
hydrogen, or organic moieties (e.g., acyl groups).
eral membrane proteins protruding into the matrix
Prosthetic groups are cofactors (organic or inor-
of mitochondria or the cytoplasm of bacteria; the
ganic) that are tightly bound to the protein and do
other two are hydrophobic integral membrane
not dissociate from the protein. Thus, one difference
proteins. The polypeptide furthest from the mem-
between a coenzyme and a prosthetic group is that
brane, called subunit A, is a flavoprotein with FAD
the former shuttles between enzymes and the latter
as its prosthetic group. Subunit A is attached to the
remains tightly bound to the enzyme. Coenzyme
second peripheral polypeptide, called subunit B,
A (carrier of acyl groups) and NAD+ (electron and
which is closest to the membrane. Subunit B con-
hydrogen carrier) are examples of coenzymes,
tains three FeS centers through which electrons
whereas FAD (electron and hydrogen carrier) and
from the flavoprotein are passed. The two integral
heme (electron carrier) are examples of prosthetic
membrane proteins (subunits C and D) are associ-
ated with one or two heme groups in many bacteria
5. The first iron–sulfur protein identified was a sol- depending upon the enzyme. The hemes may trans-
uble protein isolated from bacteria and called ferre- fer electrons from the FeS centers to the quinone
doxin. A protein similar to ferredoxin with a low within the membrane. There are variations of this
redox potential can be isolated from chloroplasts and common theme. For example, the succinic dehy-
mediates electron transfer to NADP+ during noncyc- drogenase from certain bacteria such as E. coli and
lic electron flow. Iron–sulfur proteins that have no some others does not contain heme. In addition,
other prosthetic groups are divided into four classes most gram-positive bacteria as well as ε proteobac-
based upon the number of iron atoms per molecule teria have just one very large integral polypeptide
and whether the sulfur is acid labile. Examples from subunit (called C), and this is believed to result
the four different classes are rubredoxin (no acid- from fusion of the genes for the two smaller sub-
labile sulfur), isolated from bacteria; high-potential units (C and D). The student is referred to a special
iron protein (HiPIP), isolated from photosynthetic issue of Biochimica et Biophysica Acta, volume
bacteria; chloroplast ferredoxin; and various bacte- 1553 (2002), which has a collection of articles on
rial ferredoxins. succinic dehydrogenase.
6. Electron spin resonance spectroscopy detects 13. Anraku, Y. 1988. Bacterial electron trans-
unpaired electrons such as the ones that exist in the port chains, pp. 101–132. In: Annual Review of
FeS cluster. Monochromatic microwave radiation is Biochemistry, Vol. 57. C. C. Richardson, P. D. Boyer,
absorbed by the unpaired electron when a magnetic I. B. Dawid, and A. Meister (Eds.). Annual Reviews,
field is applied. The size of the magnetic field required Palo Alto, CA.
for radiation absorption depends upon the molecular 14. García-Horsman, J. A., B. Barquera, J. Rumbley,
environment of the unpaired electron. A spectrum is J. Ma, and R. B. Gennis. 1994. The superfamily of
obtained by varying the magnetic field and keeping heme–copper respiratory oxidases. J. Bacteriol.
the frequency of the microwave radiation constant. 176:5587–5600.
The spectra differ for different FeS clusters. A spectro-
scopic constant, called the g value, is obtained which 15. Hill, S., S. Viollet, A. T. Smith, and C. Anthony.
is characteristic of the FeS cluster in the protein. 1990. Roles for enteric d-type cytochrome oxidase
in N2 fixation and microaerobiosis. J. Bacteriol.
7. Puustinen, A., and M. Wikstrom. 1991. The heme 172:2071–2078.
groups of cytochrome o from Escherichia coli. Proc.
Natl. Acad. Sci. USA 88:6122–6126. 16. Kelly, M. J. S., R. K. Poole, M. G. Yates, and C.
Kennedy. 1990. Cloning and mutagenesis of genes
8. Yagi, T., X. Xu, and A. Matsuno-Yagi. 1992. The encoding the cytochrome bd terminal oxidase com-
energy-transducing NADH–quinone oxidoreductase plex in Azotobacter vinelandii: mutants deficient in
(NDH-1) of Paracoccus denitrificans. Biochim. the cytochrome d complex are unable to fix nitrogen
Biophys. Acta 1101:181–183. in air. J. Bacteriol. 172:6010–6019.
9. Finel, M. 1993. The proton-translocating 17. Hinkle, P. C. 2005. P/O ratios of mitochon-
NADH:ubiquinone oxidoreductase: a discussion of drial oxidative phosphorylation. Biochim. Biophys.
selected topics. J. Bioenerg. Biomemb. 25:357–366. Acta—Bioenergetics 1706:1–11.
Free ebooks ==>
electron transport 173

18. Reviewed in J. Bioenerg. Biomembr. 25(4), cation of the cytochrome o complex of Escherichia
1993. coli yields a two heme/one copper terminal oxidase
with high specific activity. Biochemistry 31:6917–
19. Right-side-out vesicles are made by osmotic lysis
of the spheroplasts or protoplasts.
32. Gennis, R. B., and V. Stewart. 1996. Respiration,
20. The phosphorylation potential, ∆Gp, is depen- pp. 217–261. In: Escherichia coli and Salmonella,
dent upon the standard free energy of formation Cellular and Molecular Biology, Vol. 1. F. C.
of ATP and the actual ratios of ATP, ADP, and Pi, Neidhardt et al. (Eds.). ASM Press, Washington,
according to the following equation: ΔGp = ΔG′0 + RT DC.
ln[ATP]/[ADP][Pi] J/mol. The free energy of forma-
tion is proportional to the equilibrium constant (i.e., 33. Anraku, Y., and R. B. Dennis. 1987. The aero-
ΔG′0 = −RT ln K′eq ). bic respiratory chain of Escherichia coli. Trends
Biochem. Sci. 12:262–266.
21. Trumpower, B. L. 2003. Proton motive force
generation by a redox loop mechanism. FEBS Lett. 34. Calhoun, M. W., K. L. Oden, R. B. Gennis, M.
545:25–30. J. Teixeira de Mattos, and O. M. Neijssel. 1993.
Energetic efficiency of Escherichia coli: effects of
22. Hunter, C., H. Palsdottir, and B. L. Trumpower. mutations in components of the aerobic respiratory
2003. Proton motive pathways and mechanisms chain. J. Bacteriol. 175:3020–3025.
in the cytochrome bc1 complex. FEBS Lett. 545:
39–46. 35. Tran, Q. H., J. Bongaerts, D. Vlad, and G. Unden.
1997. Requirement for the proton-pumping NADH
23. Reviewed in: Trumpower, B. L. 1990. dehydrogenase I of Escherichia coli in respiration
Cytochrome bc1 complexes of microorganisms. of NADH to fumarate and its bioenergetic implica-
Microbiol. Rev. 54:101–129. tions. Eur. J. Biochem. 244:155–160.
24. van Verseveld, H. W., K. Krab, and A. H. 36. Green, J., M. F. Anjum, and J. R. Guest. 1997.
Stouthamer. 1981. Proton pump coupled to cyto- Regulation of the ndh gene of Escherichia coli by
chrome c oxidase in Paracoccus denitrificans. integration host factor and a novel regulator, Arr.
Biochim. Biophys. Acta 635:525–534. Microbiology 143:2865–2875.
25. de Gier, J.-W. L., M. Lubben, W. N. M. 37. Reviewed in: Berks, B. C., S. J. Ferguson, J. W.
Reijnders, C. A. Tipker, D-J. Slotboom, R. J. M. van B. Moir, and D. J. Richardson. 1995. Enzymes and
Spanning, A. H. Stouthamer, and J. van der Oost, associated electron transport systems that catalyze
1994. The terminal oxidases of Paracoccus denitri- the respiratory reduction of nitrogen oxides and
ficans. Mol. Microbiol. 13:183–196. oxyanions. Biochim. Biophys. Acta 1232:97–173.
26. Hinkle, P. C., M. A. Kumar, A. Resetar, and D. 38. Fumarate is reduced to succinate by using the
L. Harris, 1991. Mechanistic stoichiometry of mito- membrane-bound enzyme