C-Bacterial Metabolism-Springer-Verlag New York (1986)

Bacterial Metabolism
Second Edition
Springer-Verlag Berlin Heidelberg GmbH
Gerhard Gottschal k
Bacterial Metabolism
Second Edition
With 204 Figures
, Springer
Gerhard Gottschalk
Georg-August-Universitt Gtittingen
Institut fUr Mikrobiologie
37077 Gtittingen
Gennany
Library of Congress Cataloging in Publication Data

Gottschalk, Gerhard
Bacterial metabolism
Bibliography: p.
IncJudes indexes.
1. Microbial metabolism. 2. Bacteria-Physiology.
1. Title. II. Series.
QR88.G67 1985 589.9'0133 85-10002
Printed on acid-free paper.
1979, 1986 Springer-Verlag Berlin Heide1berg

Originally published by Springer-Verlag Berlin Heidelberg New York in 1986
Softcover reprint of the hardcover 2nd edition 1986
All rights reserved. No part of this book may be translated or reproduced in any fonn
without written pennission from Springer-Verlag, 175 Fifth Avenue, New York, New
York 10010, U.S.A.
The use of general descriptive names, trade names, trademarks, etc., in this
publication, even if the fonner are not especially identified, is not to be taken as a
sign that such names, as understood by the Trade Marks and Merchandise Marks Act, may
accordingly be used freely by anyone.
Typeset by Polyglot Pte, Ltd., Singapore.

9876
ISBN 978-1-4612-7003-4 ISBN 978-1-4612-1072-6 (eBook)

DOI 10.1007/978-1-4612-1072-6
To Ellen
Preface to Second Edition
Progress in certain areas of bacterial metabolism has been rapid since the first
edition of this book was published. Consequently, large parts of it had to be
rewritten or extensively revised for a second edition. Some new material has
also been included, sections on chemotaxis, bioluminescence, and catabolic
plasmids. The use of NAD, NADP, NADH 2 , and NADPH 2 as abbreviations
throughout the first edition of the book has been criticized by some reviewers.
The author has taken this to heart; the abbreviations for these coenzymes have
been changed to NAD+, NADH + H+, etc.
Many thanks are due to J. R. Andreesen, B. Bowien, B. Friedrich (Gottingen),
L. Ettlinger (Zurich), H.-J. Knackmuss (Wuppertal), H. Mayer (Freiburg), K.-
H. Schleifer (Munchen), R. Thauer (Marburg), A. Trebst (Bochum), and W.
Zumft (Karlsruhe) who read certain sections of the book and made valuable
suggestions; to Claudia Bechtel, Helga Grupe, and Ute Meyer for typing the
manuscript and preparing the figures; to Garabed Antranikian, Michael Blaut,
Armin Quentmeier, and Bernhard Moller for proofreading; and finally to the
publishers for their patience and the pleasant cooperation.
It is the hope of the author that the second edition of Bacterial Metabolism
will be as well-received as the first edition.
Gottingen, 1985 GERHARD GOTISCHALK

Preface to First Edition
This book has been written for students who are taking a course in bacterial
metabolism. I hope, however, that scholars will also find it useful either as a
help in teaching bacterial metabolism or as a review on the special aspects of
metabolism in bacteria.
The concept of this book results from my experience in teaching bacterial
metabolism. In the first chapters the principal reactions of the energy and bio-
synthetic metabolism have been discussed using Escherichia coli as a model
organism. Then the diversity of aerobic metabolism has been outlined. Following
a brief description of the regulation of the level and the activity of enzymes in
bacteria the characteristic features of fermentative, chemolithotrophic and pho-
totrophic metabolism have been discussed. Finally, the last chapter has been
devoted to nitrogen fixation. Throughout the text I have tried not only to describe
metabolic pathways and enzyme reactions but also to elucidate the physiology
of the microorganisms which carry out all these metabolic reactions.
Two comments regarding the formulas used in this book are necessary. Organic
acids are usually called after the names of their salts which are shorter (formate
for formic acid, pyruvate for pyruvic acid). However, in schemes and figures
the formulas of the free acids are given. Furthermore, it should be pointed out
that NADH 2 and NADPH 2 and not NADH and NADPH are used as abbreviations
for reduced nicotinamide-adenine dinucleotide and reduced nicotinamide-adenine
dinucleotide phosphate, respectively. This has been done as these compounds
are two electron carriers and redox reactions involving these carriers are thus
easier to formulate.
x Preface
I am particularly indebted to Joan Macy, Lynne Quandt, Jan Andreesen and

Peter Hillmer for reading the manuscript, for their criticisms and their suggestions,
and I thank Ute Gnass for typing the manuscript and for her invaluable help with
the indexing and with the preparation of the figures. Finally, I am grateful to
the publishers for their patience, willing help, and cooperation.
G6ttingen, 1978 GERHARD GOTTSCHALK

Contents
CHAPTER I
Nutrition of Bacteria
I. Major and Minor Bioelements I
II. The Two Basic Mechanisms of ATP Synthesis 4
III. Nutrients as Energy Sources 6
IV. Growth Factor Requirements of Bacteria 9
V. Summary 10
CHAPTER 2
How Escherichia coli Synthesizes ATP during Aerobic Growth on Glucose
I. Transport of D-Glucose into the E. coli Cell 13
II. Degradation of Glucose-6-Phosphate to Pyruvate via the Embden-Meyerhof-
Parnas (EMP) Pathway 15
III. Oxidative Decarboxylation of Pyruvate to Acetyl-Coenzyme A 18
IV. Oxidation of Acetyl-CoA via the Tricarboxylic Acid Cycle 20
V. The Formation of ATP in the Respiratory Chain 22
VI. Summary 35
CHAPTER 3
Biosynthesis of Escherichia coli Cells from Glucose
I. Composition of E. coli Cells 38
II. Assimilation of Ammonia 40
Ill. Assimilatory Reduction of Sulfate 42
IV. Biosynthesis of Amino Acids 43
V. How Pentose Phosphates and NADPH are Formed 55
xii Contents
VI. Ribonucleotides and Deoxyribonucleotides 59

VII. Biosynthesis of Lipids 65
VIII. Formation of Carbohydrates 71
IX. Synthesis of Polymers 73
X. The Requirement for an Anaplerotic Sequence 92
XI. Summary 93
CHAPTER 4
Aerobic Growth of Escherichia coli on Substrates Other Than Glucose
I. Fructose and Lactose as Substrates 96
II. Pentoses as Substrates 98
Ill. Acetate, Pyruvate, and L-Malate as Substrates 99
IV. Summary 103
CHAPTER 5
Metabolic Diversity of Aerobic Heterotrophs
I. The Different Mechanisms for the Uptake of Substrates 104
II. The Entner-Doudoroff Pathway 114
III. Sugar Degradation via the Pentose Phosphate Cycle 118
IV. The Methylglyoxal Bypass 119
V. Diversity in Energy Metabolism 120
VI. Dissimilatory Reduction of Nitrate 122
VII. Bacterial Bioluminescence 126
VIII. Alternate Anaplerotic Sequences 127
IX. Biosynthesis of Monomers and Polymers 129
X. Summary 138
CHAPTER 6
Catabolic Activities of Aerobic Heterotrophs
I. Degradation of Polymers by Exoenzymes 141
II. Growth with Amino Acids 145
Ill. Growth with Organic Acids 149
IV. Growth with Aliphatic Hydrocarbons 154
V. Growth with Aromatic Compounds 157
VI. Growth with C, Compounds 162
VII. Incomplete Oxidations 169
VIII. Plasmid-Encoded Catabolic Activities 174
IX. Summary 176
CHAPTER 7
Regulation of Bacterial Metabolism
I. Regulation of Enzyme Synthesis by Induction and Repression 178
Contents xiii
II. Regulation of Enzyme Activity 194

Ill. Summary 206
CHAPTER 8
Bacterial Fermentations
I. Alcohol Fermentation 210
II. Lactate Fermentation 214
Ill. Butyrate and Butanol-Acetone Fermentation 224
IV. Mixed Acici and Butanediol Fermentation 237
V. Propionate and Succinate Fermentation 242
VI. Acetate Fermentation 249
VII. Methane Fermentation 252
VIII. Sulfide Fermentation (Desulfurication) 260
IX. The Anaerobic Food Chain 265
X. Fermentation of Nitrogenous Compounds 269
XI. Summary 280
CHAPTER 9
Chemolithotrophic and Phototrophic Metabolism
I. Chemolithotrophic Metabolism 283
II. Assimilation of CO, 295
Ill. Phototrophic Metabolism 300
IV. Summary 316
CHAPTER 10
Fixation of Molecular Nitrogen
I. Nitrogen-fixing Organisms 318
II. Biochemistry of Nitrogen Fixation 319
III. Regulation of Nitrogenase 324
IV. Summary 326
Further Reading 327
Index of Organisms 337
Subject Index 341

Chapter 1
Nutrition of Bacteria
Bacteria, like all other living organisms, require certain nutrients for
growth. These nutrients must contain those chemical elements that are
constituents of the cellular materials and that are necessary for the activity
of enzyme and transport systems. In addition, the nutrients must provide
the organisms with materials for the production of biologically utilizable
energy.
I. Major and Minor Bioelements

Only a small number of the elements of the periodic system are required
by organisms in relatively high concentrations (> 10- 4 M). These 12 major
bioelements and some of their functions are presented in Table 1.1.
Carbon, oxygen, hydrogen, and nitrogen are the main constituents of the
organic compounds occurring in organisms. Sulfur is required for the
synthesis of the amino acids cysteine and methionine and of a number of
coenzymes. Phosphorus is present in nucleic acids, phospholipids, teichoic
acids, and in nucleotides such as ATP, GTP, NAD+, and FAD. Potassium
ions are the principal inorganic cations in the cell.
The next three major bioelements in Table 1.1 are metal ions, which are
required as cofactors for enzyme activity and as components of metal
complexes. Most of the biologically active phosphate esters are, for
instance, present in the cell as magnesium complexes. The phospholi-
poproteins of bacterial cell walls and membranes are also chelated with
magnesium ions. Exoenzymes such as amylases and proteases are calcium
proteins, and calcium dipicolinate is an important component of endo-
spores. Ferrous and ferric ions are present in redox carriers such as
2 1: Nutrition of Bacteria
Table 1.1. The 12 major bioelements, their sources, and some of their functions in
microorganisms
element source function in metabolism
C
o
organic compounds, CO 2
O 2, H 20, organic compounds, CO 2
1 main constituents of cellular
H H 2 , H 20, organic compounds material
N NHt, N03 , N 2 , organic compounds
S SO~-, HS-, So, S20~-, constituent of cysteine,
organic sulfur compounds methionine, thiamine
pyrophosphate, coenzyme
A, biotin, and a-lipoic acid
p HPO~- constituent of nucleic acids,
phospholipids, and nucle-
otides
K principal inorganic cation in
the cell, cofactor of some
enzymes, e.g., pyruvate
kinase
Mg cofactor of many enzymes
(e.g., kinases); present in
cell walls, membranes,
ribosomes, and phosphate
esters
Ca present in exoenzymes
(amylases, proteases) and
cell walls; Ca-dipicolinate
is an important component
of endospores
Fe present in cytochromes,
ferredoxins, and other
iron-sulfur proteins;
cofactor of enzymes (some
dehydratases)
Na involved in various transport
processes
Cl important inorganic anion in
the cell
cytochromes and iron-sulfur proteins. Sodium ions, in relatively high

concentrations, are required by halophilic microorganisms, by methane-
producing bacteria, most rumen bacteria, and many others. Na+ seems to
be involved in various transport processes. Chloride is an important
inorganic anion in the cell.
Major and Minor Bioelements 3
Table 1.2. Minor bioelements, their sources, and some of their functions in
microorganisms
element source function in metabolism
Zn Zn H present in alcohol dehydrogenase, alkaline phosphatase,

aldolase, RNA and DNA polymerase
Mn Mn H present in bacterial and mitochondrial superoxide dismutase
and in photosystem II; cofactor of some enzymes (PEP
carboxykinase, re-citrate synthase)
Mo MoO~- present in nitrate reductase, nitrogenase, xanthine
dehydrogenase, and formate dehydrogenase
Se SeOj- present in glycine reductase and formate dehydrogenase
Co Co 2 + present in coenzyme B l2 -containing enzymes (glutamate
mutase, methylmalonyl-CoA mutase)
Cu Cu 2 + present in cytochrome oxidase, in nitrite reductase of
denitrifying bacteria and oxygenases
Ni NiH present in urease, hydrogenase and in factor F 430
W WO~- present in some formate dehydrogenases
Besides these 12 major bioelements, organisms require a number of

others in small amounts (Table 1.2). Zinc and manganese ions are essential
for all microorganisms. Zinc is especially important because RNA and
DNA polymerase are zinc-metalloproteins. Manganese is present in bacte-
rial and mitochondrial superoxide dismutase. It is a component of photo-
system II in plants and cyanobacteria and functions as cofactor in some
enzymes. Specific functions can be assigned to the other metals listed in
Table 1.2. Molybdoproteins play an important role in nitrogen metabolism
and in formate oxidation. Xanthine dehydrogenase also contains molybde-
num. Of the selenoproteins listed in Table 1.2, the glycine reductase
contains selenium in the form of selenocysteine. Cobalt is required by all
organisms that perform B 12-dependent reactions. Copper is present in a
number of enzymes transferring electrons from substrates to oxygen and in
enzymes involved in denitrification. Nickel has recently been found in
hydrogenase and as a Ni-tetrapyrrole compound (factor F 43o ) in metha-
nogenic bacteria. Finally, tungsten is needed by microorganisms in some
rare cases.
In nature, most of the bioelements occur as salts, and they are taken up
by the organisms as cations and anions, respectively. A greater diversity of
compounds utilized by microorganisms is observed only with respect to the
first five elements shown in Table 1.1: sulfur, nitrogen, oxygen, hydrogen,
and carbon.
Sulfur is normally taken up as sulfate, reduced to the level of sulfide,
and then used for biosynthetic purposes. Certain groups of bacteria,
however, depend on the availability of reduced sulfur compounds. Metha-

nogenic bacteria grow only in the presence of hydrogen sulfide as sulfur
source. Thiobacilli and a number of phototrophic bacteria require sulfide,
elemental sulfur, or thiosulfate as electron donor.
Nitrogen is required in large quantities because it amounts to approx-
imately 10% of the dry weight of bacteria. It occurs naturally in the form of
ammonia, nitrate, nitrite, nitrogen-containing organic compounds, and
molecular nitrogen. The preferred source of nitrogen is ammonia, which
can be utilized by practically all microorganisms. Nitrate is also taken up
and used by many microorganisms but not by all. Before it can be
incorporated into organic substances it has to be reduced to ammonia.
Nitrite is the product of the nitrate-nitrite respiration and of the metabolic
activities of Nitrosomonas and related species. A number of organisms
reduce it to ammonia or Nz. Alternatively, nitrite can be oxidized to nitrate
by Nitrobacter species. Several bacteria are able to fix molecular nitrogen
and to reduce it to ammonia. This capacity is found only in certain
prokaryotes but not in eukaryotes. Finally, organic compounds serve as
nitrogen sources for many microorganisms. Usually these compounds are
degraded such that ammonia becomes available for biosyntheses.
Carbon, hydrogen, and oxygen can be utilized by bacteria in the form of
organic and inorganic compounds. Among the inorganic compounds used
are COz, CO, Hz, HzS, NH 3 , HzO, 0z, NO), and SO~-. The number of
organic compounds utilized by microorganisms is large. On earth, not a
single organic compound formed by organisms is accumulated. This
implies that all of them are degradable. Microorganisms play an important
role in this degradation. Their versatility has led to the formulation of the
"doctrine of microbial catabolic infallibility," meaning that every naturally
occurring carbon compound is used by some microbe.
The metabolism of carbon-, hydrogen-, and oxygen-containing com-
pounds is important not only because these elements are the main
constituents of the cell. These compounds are important substrates for the
energy production in microorganisms.
II. The Two Basic Mechanisms of

ATP Synthesis
The principal carrier of biologically utilizable energy is adenosine-5 ' -
triphosphate (ATP), and all energy-requiring processes in living cells are
directly or indirectly coupled to the conversion of ATP to adenosine-5 '-
diphosphate (ADP) and inorganic phosphate (Pi):
cellular processes
ATP + HzO \" /' ) ADP + Pi

The Two Basic Mechanisms of ATP Synthesis 5
ATP contains two phosphate bonds with a high free energy of hydrolysis.
These bonds are often symbolized by the squiggle "-":
NH 2
o 0- 0 i
II I II jN~N
-0-r-0-r:-0-r-0-r:20~--t'N.~J
0- 0 0- \--.(
, I
OHOH
Because of the high-energy phosphoryl bonds, ATP is an excellent

phosphorylating agent, and it is used as such in a large number of reactions
by all organisms. At the expense of ATP, intermediates of cell metabolism
are activated for further reactions, such as condensations, reductions, and
cleavages. Glutamine, for instance, can be synthesized from glutamate and
ammonia only if a phosphorylated intermediate is formed. The reaction is,
therefore, connected with the formation of ADP and Pi from ATP:
glutamate + NH 3 + ATP - glutamine + ADP + Pi
The high potential of group transfer of the AMP and the ADP group is
also taken advantage of in a number of reactions; amino acids are activated
by their conversion into the corresponding AMP derivatives with ATP,
and AMP is released in the formation of aminoacyl-transfer RNA:
amino acid + ATP - aminoacyl-AMP + pyrophosphate
aminoacyl-AMP + tRNA - aminoacyl-tRNA + AMP
The enzyme adenylate kinase catalyzes the phosphorylation of AMP to
ADP with ATP; pyrophosphate (PP j ) is hydrolyzed to inorganic phosphate
by pyrophosphatase so that the end products of this reaction series are also
ADP and Pi:
AMP + ATP adenylatc kinase) 2A D P
PP + H 0 pyrophosphatasc) 2p.
t 2 I
ADP and Pi are thus the principal products of the energy expenditure
in metabolism, and the generation of ATP from ADP and Pi is a vital
process of all living organisms. There are two basic mechanisms of ATP
generation: electron transport phosphorylation and substrate-level
phosphorylation.
Electron transport phosphorylation refers to a mechanism in which the
flow of electrons from donors with a negative redox potential to acceptors
with a more positive redox potential is coupled to the synthesis of ATP
from ADP and Pi' Systems in which electron transport phosphorylation
occurs are the respiratory chains and the photosynthetic apparatus, they
are principally membrane-bound.
Substrate-level phosphorylation is the second mechanism of ATP gen-

eration. During the degradation of organic substrates a small number of
intermediates is formed containing high-energy phosphoryl bonds. In-
termediates of this kind are:
o
II
C-O-PO,H 2
I 1.3-bisphosphoglycerate
HC-OH
I
CH 2 -O-PO)H 2
CH 2
II
c-o
I - PO,H
. 2 phosphoenolpyruvate
COOH
II
C-O - PO)H 2 acetyl phosphate
I
CH)
The further metabolism of such organic - P compounds is coupled to the

transfer of the phosphate group to ADP, and this kind of ATP synthesis is
called substrate-level phosphorylation:
H 2C=C-COOH
I + ADP ~ CH 3-CO-COOH + ATP
O~P03H2
III. Nutrients as Energy Sources

It has already been mentioned that the function of the nutrients is not only
to provide the organisms with the bioelements. Nutrients are also required
as energy sources-as fuel for the production of ATP. Various energy
sources are available in nature and are taken advantage of by microorgan-
isms, but they cannot be used by every bacterium, and it has become useful
to group bacteria on the basis of their characteristic energy source.
Organisms using light as energy source are called phototrophs (Greek
phos = light, trophe = nutrition). If ATP comes from chemical reactions,
the organisms that carry out this type of energy metabolism are called
chemotrophs.
A. Phototrophy
Phototrophs contain a photosynthetic apparatus that enables them to
convert light energy into the high-energy phosphate bonds of ATP:
Nutrients as Energy Sources 7
Table 1.3. The two types of phototrophic metabolism
type electron donor carbon source examples
photolithotrophy CO 2 plants, cyanobacteria

Chromatiaceae,
Chlorobiaceae
photoorganotrophy organic organic Rhodospirillaceae
substrates substrates
The carbon source commonly used by phototrophs is CO 2 , and organisms

that derive most of their cell carbon from CO 2 are called C-autotrophs
(Greek autos = self; autotroph = self-feeding). Thus, phototrophic bac-
teria are C-autotrophic organisms, and when growing in the presence of
CO 2 , they require an electron donor for the reduction of CO 2 to the level
of cell material. It is apparent from Table 1.3 that, as in plants and in cyano-
bacteria, the electron donors used are frequently inorganic compounds.
Phototrophic bacteria that employ molecular hydrogen or reduced sulfur
compounds are called photolithotrophs (Greek lithos = stone). A number
of phototrophic bacteria can also grow in the light with organic substrates
such as succinate or acetate. Under these conditions the source of any
electrons used in reduction reactions is an organic substrate, and the
bacteria grow then as photoorganotrophs. Clearly, the main source of cell
carbon is then the organic substrate and not CO 2 , and the organisms grow
as C-heterotrophs (Greek heteros = the other; heterotroph = feeding on
others). It follows that the terms C-autotroph and C-heterotroph reflect
the nature of the carbon source, whereas the terms lithotroph and
organotroph describe the nature of the electron donor used.
B. Chemotrophy
Most bacteria gain ATP by chemical reactions. These are commonly
oxidation-reduction reactions, which means that one substrate is reduced
at the expense of a second one:
X red + A ox T\ X ox + A red
ADP + Pi ATP + H2 0
Higher organisms can only use organic substrates as electron donors (Xred )
and oxygen as electron acceptor (A ox ) and it is a specialty of the bacterial
energy metabolism that, alternatively, other donors and acceptors can
be employed. Here, A ox may stand for oxygen, nitrate, sulfate, CO 2 ,
or an organic compound, and Xred for an inorganic or an organic com-
pound. By analogy to the nutritional classification of the phototrophs,
bacteria that employ an organic compound as electron donor are called
8 I: Nutrition of Bacteria
Table 1.4. The two types of chemotrophic metabolism
electron electron carbon

type donor acceptor source examples
chemoorganotrophy organic Oz organic pseudomonads,

substrate substrate bacilli
organic NO; organic Bacillus
substrate substrate licheniformis
organic SO~- organic sulfate reducers
substrate substrate
organic organic organic clostridia, lactic
substrate substrate substrate acid bacteria
lchemolithotrophy Hz Oz COz hydrogen-oxidizing
bacteria
HzS Oz COz thiobacilli
HzS NO; COz Th. denitrificans
Fe z+ Oz COz Th. ferrooxidans
NH 3 Oz COz Nitrosomonas
NOi O2 COz Nitrobacter
Hz COz COz methanogenic
bacteria
Hz COz COz acetogenic bacteria
chemoorganotrophs. This group includes aerobes and anaerobes. The

anaerobes, as is apparent from Table 1.4, use either nitrate, sulfate, or
organic substrates as electron acceptors. Thus, organisms carrying out
fermentations-such as the clostridia and lactic acid bacteria-belong to
this group.
Chemolithotrophs use inorganic electron donors such as hydrogen,
hydrogen sulfide, ferrous ions, nitrite, or ammonia. These compounds are
oxidized with oxygen to water, sulfate, ferric ions, and nitrate, respec-
tively, and these exergonic reactions are coupled to the formation of ATP
from ADP and Pi, e.g.:
Hz + 1-oz ~ HzO
ADP + Pi ATP + HzO
Some organisms, e.g., Thiobacillus denitrificans, can replace oxygen by
nitrate.
Although the methanogenic and acetogenic bacteria are quite different
from all other chemolithotrophs, because they are strictly anaerobic
organisms, they belong to this nutritional group. They gain ATP by
reduction of CO z to methane or acetate with molecular hydrogen. Thus,
only inorganic substrates are used for energy production.
Clearly, chemolithotrophs gain ATP without metabolizing an organic
compound. Their carbon source is usually CO z , and they are, therefore,
Growth Factor Requirements of Bacteria 9
C-autotrophs. However, many chemolithotrophs are not restricted to this

kind of energy metabolism. Hydrogen-oxidizing bacteria, for instance, can
grow as chemoorganotrophs (aerobically with carbohydrates or organic
acids) under appropriate conditions. They are, therefore, called facultative
chemolithotrophs (all hydrogen-oxidizing bacteria, some thiobacilli). Spe-
cies (Nitrosomonas, Thiobacillus thiooxidans) , which are unable to grow in
the absence of their inorganic electron donors, are called obligate chemo-
lithotrophs.
IV. Growth Factor Requirements

of Bacteria
So far it has been presumed that microorganisms themselves are able to
synthesize all organic compounds required for growth. In fact, there are
C-autotrophic bacteria that derive their cell carbon from CO 2 exclusively
(e.g., Alcaligenes eutrophus and Nitrobacter winogradskyi) and there are
also C-heterotrophs that grow on simple carbon sources such as glucose
(e.g., Escherichia coli, Bacillus megaterium, and Clostridium pas-
teurianum). However, a number of bacteria lack the ability to synthesize
all the organic compounds needed for growth and depend on certain
growth factors. These factors can be combined to form three groups:
1. vitamins and related compounds, required in small amounts
2. amino acids
3. purines and pyrimidines
Number and kind of growth factors, which must be present in the
growth medium, differ among bacteria. Lactic acid bacteria require
practically all amino acids, purines, pyrimidines, and vitamins for growth.
Their biosynthetic capacity is rather limited. Common among microorgan-
isms are requirements for vitamins and related compounds. Some of them
and their functions in metabolism are summarized in Table 1.5.
The exact growth factor requirements are not known for all microorgan-
isms, and microbiologists add yeast extract and peptone to the growth
media as complex and cheap sources of these factors. Synthetic media-
media of known composition-are used for special purposes provided the
requirements of a particular organism are known. Clostridium kluyveri
grows in a medium supplemented with biotin and p-aminobenzoic acid. To
the media for phototrophic bacteria, a vitamin solution is added containing
nicotinic acid, thiamine, p-aminobenzoic acid, biotin, and vitamin B l2 .
Some organisms exhibit special requirements. A medium for Haemophilus
species must contain hemin for cytochrome biosynthesis and also NAD+.
Bacteroides species require hemin. Methanobacterium ruminantium grows
only if coenzyme M (2-mercaptoethanesulfonic acid) and 2-methyl-n-
butyric acid are present. These few examples document that microogan-
isms may exhibit various defects in their biosynthetic machinery, and that
growth factors are important for many of them.
Table 1.5. Vitamins and related compounds and their functions in metabolism
compound function in metabolism
p-aminobenzoic acid precursor of tetrahydrofolic acid, a coenzyme involved in

transfer of one-carbon units
biotin prosthetic group of enzymes catalyzing carboxylation
reactions
coenzyme M coenzyme involved in methane formation
folic acid tetrahydrofolic acid is a coenzyme involved in transfer of
one-carbon units
hemin precursor of cytochromes
lipoic acid prosthetic group of the pyruvate dehydrogenase complex
(dithiooctanic acid)
nicotinic acid precursor ofNAD+ and NADP+, which are coenzymes
of many dehydrogenases
pantothenic acid precursor of coenzyme A and of the prosthetic group of
acyl carrier proteins
pyridoxine pyridoxal phosphate is a coenzyme for transaminases and
(vitamin B6 ) amino acid decarboxylases
riboflavin precursor of flavin mononucleotide (FMN) and flavin
(vitamin B z) adenine dinucleotide (FAD), which are the prosthetic
groups of flavoproteins
thiamine (vitamin B I ) thiamine pyrophosphate is the prosthetic group of
decarboxylases and transketolases
vitamin B 12 coenzyme BIZ is involved in rearrangement reactions
(cyanocobalamin) (e.g., glutamate mutase)
vitamin K precursor of menaquinone, which functions as electron
carrier in respiratory chains
V. Summary
1. Twelve chemical elements are required by organisms in relatively
high concentrations: C, 0, H, N, S, P, K, Mg, Ca, Fe, Na, CI.
2. The minor bioelements comprise some that are essential for all
microorganisms (Zn, Mn) and others that are required only in connection
with special metabolic activities (e.g., Se, Mo, Co, Cu, Ni, W).
3. ATP is synthesized from ADP and Pi either by electron transport
phosphorylation or by substrate-level phosphorylation. The energy for
ATP synthesis is provided either as physical (light) or as chemical energy.
Summary 11
4. Phototrophic bacteria, which use inorganic electron donors such as

Hz or HzS for the reduction of CO z to cell carbon, are called photolitho-
trophs. Organisms that grow on organic substrates in the light are called
photoorganotrophs.
5. Chemotrophic organisms derive energy from chemical reactions, in
most cases from oxidation reactions. Chemoorganotrophs metabolize
organic substrates. If only inorganic compounds are involved in energy
production, the organisms are called chemolithotrophs.
6. With respect to the origin of the cell carbon, C-heterotrophs and
C-autotrophs are distinguished, the former use organic compounds and the
latter CO z as the main carbon source.
7. In addition to their simple carbon sources, many microorganisms
require one or several growth factors for growth. These factors are
vitamins and related compounds, amino acids, and purines and pyrimi-
dines. Very common is a requirement for vitamins such as biotin,
p-aminobenzoic acid, thiamine, nicotinic acid, and vitamin B 12 .
Chapter 2
How Escherichia coli
Synthesizes AlP
during Aerobic Growth
on Glucose
Escherichia coli belongs to the group of facultatively anaerobic bacteria. It

is able to grow with a number of substrates in the presence of oxygen or in
its absence. Under aerobic conditions part of the substrate is oxidized to
CO 2 with oxygen as the terminal electron acceptor. This process is
exergonic and allows for the formation of ATP, which is required for the
biosynthesis of cellular constituents.
If D-glucose is the substrate about 50% is oxidized to CO 2 ; this results in
enough ATP to convert the other 50% into cell material:
aGO,= -2870 kJ
(-686 kcal)
cell material
In order to utilize the energy released during glucose oxidation effec-

tively for the formation of ATP from ADP and inorganic phosphate, the
glucose molecule must undergo a series of reactions, which at first sight
appear fairly complicated but nevertheless are very economical. All the
reactions involved in the oxidation of glucose to CO 2 can be divided into a
number of functional blocks of reactions; in E. coli these are:
Transport of D-Glucose into the E. coli Cell 13
glucose
(outside)
V PEP A. Transport of D-glucose into the cell
1'- pyruvate
by the phosphoenolpyruvate phos-
photransferase system.
glucose-6-phosphate
(inside)
~
B. Degradation of D-glucose-6-
phosphate to pyruvate via the
(4H)
Embden-Meyerhof-Parnas pathway.
+
2 pyruvate C. Oxidative decarboxylation of pyru-
vate to acetyl-coenzyme A by pyru-
vate dehydrogenase.
lCD, (4H)
2 acetyl-CoA D. Oxidation of the acetyl moiety of

acetyl-coenzyme A to CO 2 via the
~
tricarboxylic acid cycle.
4C0 2 (16H)
(l4H)2< E. Oxidation of the reduced coenzymes
formed in steps B to D in the respira-
ATPV 12H 2 0 tory chain.
These reactions together accomplish the oxidation of glucose to CO 2
and water, with the conservation of part of the energy released as
phosphate-bond energy of ATP.
I. Transport of D-Glucose into the

E. coli Cell
The cytoplasmic membrane of E. coli is not simply permeable to glucose;
there is no free diffusion of glucose in and out of the bacterial cells which
would allow the concentration of the sugar inside and outside to become
equal. Instead E. coli possesses a transport system that recognizes glucose
specifically. It "picks up" glucose at the medium side of the membrane and
releases it on the cytoplasm side of the membrane. This transport process is
coupled to a chemical conversion of the substrate, i.e., the phosphoryla-
tion of glucose to glucose-6-phosphate.
14 2: How Escherichia coli Synthesizes ATP during Aerobic Growth on Glucose
The phosphate donor in this reaction is phosphoenolpyruvate (PEP)

and the enzyme complex catalyzing this transport process is called phos-
phoenolpyruvate: glucose phosphotransferase system. It is composed of
three reactions:
1. A small protein, designated HPr, is phosphorylated by PEP:
enzyme I
PEP + HPr ( ) phospho-HPr + pyruvate
This reaction is not specific to the glucose transport system but is involved
in sugar transport of E. coli in general. Enzyme I and HPr serve as catalysts
in transport systems for glucose, mannose, fructose, and other hexoses;
both proteins are soluble.
2. The phosphoryl group of phospho-HPr is transferred to enzyme III:
phospho-HPr + enzyme III GIe ( l HPr + phosphoenzyme III Gle
Phosphorylation of enzyme III GIe brings about an increase of the exposure
of lipophilic areas of the enzyme molecule and leads to an increase of its
membrane solubility. Thus phosphoenzyme III GIe can enter the cytoplas-
mic membrane. Enzyme III is substrate specific, and the glucose-specific
protein is designated enzyme III GIe
3. The phosphoryl group of enzyme III GIe is transferred to o-glucose and
the o-glucose-6-phosphate is released into the cytoplasm. This reaction
is catalyzed by enzyme II, which is speci5c for o-glucose and is
membrane-bound:
Gle enzyme II
phosphoenzyme III + glucose )
glucose-6-phosphate + enzyme III GIe
Figure 2.1 summarizes the reactions. Transport processes, which are
membrane
I glucose-6-P I
~"':;"------l~ EIII GIc
J'O
HPr-::XPEP
E,
phospho-Em
Glc
-4j---- phospho-HPr pyruvate
Figure 2.1. Transport of glucose by the PEP: glucose phosphotransferase system.

Note that free glucose does not appear inside the cell. It is phosphorylated by the
action of enzyme II and of phosphoenzyme III and released into the cytoplasm as
glucose-6-phosphate.
Degradation of 0Iucose-6-Phosphate to Pyruvate 15
coupled to a conversion of the substrate (glucose) into a derivative of the

substrate (glucose-6-phosphate), are called translocation processes.
II. Degradation of Glucose-6-Phosphate

to Pyruvate via the
Embden-Meyerhof-Parnas
(EMP) Pathway
The EMP pathway was first discovered in muscle tissues. It is the most
commonly used sequence of reactions for the conversion of glucose-6-
phosphate into pyruvate and occurs in animals, plants, and many bacteria.
E. coli contains high activities of the necessary enzymes (Table 2.1). In
the first two reactions catalyzed by the enzymes glucose phosphate isom-
erase and phosphofructokinase, glucose-6-phosphate is converted into
o-fructose.l,6.bisphosphate, the characteristic intermediate of this path-
way (Fig. 2.2). Fructose bisphosphate aldolase splits fructose-1,6-bis-
phosphate into two C3 fragments-dihydroxyacetonephosphate and
D-glyceraldehyde-3-phosphate. Both compounds are in equilibrium with
each other due to the presence of the enzyme triosephosphate isomerase.
The equilibrium constant of the isomerase reaction favors the dihydroxy-
acetonephosphate but the enzyme triosephosphate isomerase is so active
(compared to other enzymes of the EMP pathway) that immediate
conversion of the ketose derivative into the aldose isomer occurs.
Table 2.1. Activity of the enzymes of the Embden-Meyerhof-Parnas

pathway in cell extracts of E. coli
average specific activity

enzyme (units/mg proteint
glucose phosphate isomerase 1

phosphofructokinase 0.3
fructose bisphosphate aldolase 0.1
triose phosphate isomerase 4
glyceraldehyde-3-phosphate dehydrogenase 1.2
3-phosphoglycerate kinase 2.2
phosphoglycerate mutase 2
enolase 1
pyruvate kinase 0.7
a One enzyme unit is defined as that amount of enzyme that catalyzes the
conversion of 1 fLmol of substrate(s) to the product(s) in 1 min at 25C.
Data obtained as personal communication from D.O. Fraenkel (Boston, MA,
U.S.A.).
CH,OP03H,
I
c=o H,03POH2~0~CH20P03H2
I
HO-C-H
I
H-C-OH
I
H-C-OH
I
~H~H
OH H
CH,OP0 3H,
a b
Figure 2.2. Structure of fructose-l,6-bisphosphate. a: Fischer projection of the
open form. b: Haworth projection of the furanose form.
Consequently, if only little dihydroxyacetonephosphate is needed by the

cells, most of the C 3 fragments originating from fructose-l,6-bisphosphate
can be further metabolized via glyceraldehyde-3-phosphate. This is what
normally happens (Fig. 2.3).
The oxidation of glyceraldehyde-3-phosphate to pyruvate is initiated by
two enzymes, which together accomplish the oxidation of the aldehyde
group to a carboxyl group. The first of these is D-glyceraldehyde-3-
phosphate dehydrogenase. It contains bound NAD+, and the reaction
proceeds as shown in Fig. 2.4.
The l,3-bisphosphoglycerate formed is a mixed anhydride of D-3-
phosphoglycerate and phosphate, and the free energy of hydrolysis of this
anhydride is higher than that for ATP H 20) ADP + Pi' Therefore, the
conversion of l,3-bisphosphoglycerate to 3-phosphoglycerate can be cou-
pled to the phosphorylation of ADP to ATP. The enzyme catalyzing this
reaction is 3-phosphoglycerate kinase. This is one of the two sites of the
EMP pathway that yield ATP by substrate-level phosphorylation:
D-l,3-bisphosphoglycerate + ADP 3-phosphoglyeerate kinase)
D-3-phosphoglycerate + ATP
In the next two steps 3-phosphoglycerate is converted to phosphoenol-
pyruvate. First phosphoglycerate mutase transfers the phosphoryl group
from position three to position two of glycerate. The enzyme requires
2,3-bisphosphoglycerate as cofactor. Enolase (phosphopyruvate hydra-
tase) removes water to yield PEP. This is another compound containing a
phosphoryl bond with a high free energy of hydrolysis, and during the
formation of pyruvate from PEP it is transferred to ADP to yield ATP.
The enzyme catalyzing this reaction is pyruvate kinase (second site of ATP
formation in the EMP pathway):
pyruvate kinase ATP
PEP + ADP l pyruvate +
Since PEP is required for the glucose transport system, only half of it is
available for the pyruvate kinase reaction.
Degradation of Glucose-6-Phosphate to Pyruvate 17
@ @> ~~~O
"'-~
I I
I HO-C-H fructose-I, 6-
- - - - : :.......""'::.....--.~ L - - - t -,
3 H-C-O:H bisphosphate
I L _
H-C-OH
I
CH 2 0
fructose-6-phosphate
H-C=O dihydroxyacetone-
.----Ll
CH 2 0H
t=O-~-.,
HC=O
I
H-C-OH glyceraldehyde-3-
I phosphate I - I I phosphate
H-C-OH CH 2 0 5 : CH 20
I I
HO-C-H
l
I
H-C-OH
Pi+INAD+I~ 'h-~+Pi
I
H-C-OH
INADH+H1~
~J
+---INADH + wi
~~/
I
CH 20 COO COO
I I 1, 3-bisphospho-
glucose-6-phosphate HC-OH HC-OH
I glycerate
CH 20
I
CH 2 00
7tf:~ 7t~@)
~@)
COOH
I 3-phospho-
HC-OH
I glycerate
CH 2 0
8 tt
COOH
I 2-phospho-
HC-Oev
I glycerate
CH 2 0H
COOH COOH
I
'----------i- O I
C-O
II
phosphoenol-
pyruvate
CH 2 CH 2
lO~~
~~
~9Q1f eopa
- - - - - - - -.. ;C~O'!~~o
.. pymvate
I ~+kf I, ,~.,~;
CH3 CH3
Figure 2.3. Uptake of glucose and breakdown of glucose-6-phosphate to 2 pyruvate

via the Embden-Meyerhof-Parnas pathway. 1, PEP: glucose phosphotransferase;
2, glucose phosphate isomerase; 3, phosphofructokinase; 4, fructose bisphosphate
aldolase; 5, triose phosphate isomerase; 6, glyceraldehyde-3-phosphate dehydro-
genase; 7, 3-phosphoglycerate kinase; 8, phosphoglycerate mutase; 9, enolase;
10, pyruvate kinase.
+
H-C=O
H-t-OH ~
@
NADH +H+
E s-c=o
I I
CH 20 H-C-OH
I
CH20
NADH+W
0- s-1=0
H-C-OH
I
CH 2 0
NAD+
0- s-1=0
H-C-OH
O=C-O-
I
H-C-OH
I
CH20
I
CH 2 0
Figure 2.4. The conversion of glyceraldehyde-3-phosphate to 1,3-bisphospho-
glycerate. In the initial reaction the aldehyde is oxidized to a thioester and the
reducing power is transferred to the enzyme-bound NAD+. An exchange reaction
then takes place with soluble NAD+. Finally, the acyl group on the enzyme is
transferred to inorganic phosphate to yield the 1,3-bisphosphate.
The sum of the reactions discussed thus far is:

glucose + PEP ----+ glucose-6-phosphate + pyruvate
glucose-6-phosphate + ADP + 2NAD+ + 2P j ----+
2PEP + ATP + 2NADH + 2H+
PEP + ADP ----+ pyruvate + ATP
glucose + 2ADP + 2NAD+ + 2P j ----+
2 pyruvate + 2ATP + 2NADH + 2H+
NADH is formed in the glyceraldehyde-3-phosphate dehydrogenase re-
action and ATP by 3-phosphoglycerate kinase and pyruvate kinase. One
ATP is consumed in the phosphofructokinase reaction.
III. Oxidative Decarboxylation of Pyruvate

to Acetyl-Coenzyme A
As in most aerobic microorganisms the formation of acetyl-coenzyme A
(acetyl-CoA) from pyruvate in E. coli is catalyzed by the pyruvate
dehydrogenase complex. This complex consists of three enzymes: 24 mole-
cules each of pyruvate dehydrogenase (E 1) and dihydrolipoate trans-
acetylase (E 2 ) and 12 molecules of dihydrolipoate dehydrogenase (E 3 )
The E 2 molecules form the core of the complex, and the E 1 and E 3
molecules are bound to the outside of it. E 1 contains thiamine pyrophos-
Oxidative Decarboxylation of Pyruvate to Acetyl-Coenzyme A 19
phate (TPP) , and the first step in the oxidative decarboxylation is the
addition of pyruvate to C-2 of the thiazolium ring of TPP to form
lactyl-TPP-E 1
NH CH,
2 )=i(CH2h- - -enzyme
NY" I
CH2-~\ I
H,C~N
:)- S
+ Hl-
o
H C-C-C,;/
3 II "'0-
o
The further reactions acting upon the lactyl residue are summarized in
Fig. 2.5. Decarboxylation yields hydroxyethyl-TPP-E t . The hydroxyethyl
moiety is then transferred from TPP to the lipoate group of E z. Concom-
itantly the disulfide bond of lipoate is reduced. The acetyl group thus
TPP-CH-CH 3 + cO 2
I
S OH
I
S
FAD
TPP
SH
SH + CH 3 -CO-SCoA
FAD
sum: CH 3 -CO-COOH + CoA + NAD+ .... CH 3 -CO-CoA + CO 2 + NADH + W

Figure 2.5. Reactions catalyzed by the pyruvate dehydrogenase complex. EJ,
pyruvate dehydrogenase; E 2 , dihydrolipoate transacetylase; E 3 , dihydrolipoate
dehydrogenase; TPP, thiamine pyrophosphate; the disulfide compound linked to E 2
is the oxidized form of lipoate.
formed is then released as acetyl-CoA and under catalysis of E 3 the

sulfhydryl form of lipoate is oxidized by NAD+ to the disulfide form. The
enzyme complex is then ready for the oxidation of another molecule of
pyruvate to acetyl-CoA.
The combined action of the enzymes of the EMP pathway and the
pyruvate dehydrogenase complex leads to the degradation of glucose to
COz and acetyl-CoA according to the equation:
glucose + 2ADP + 4NAD+ + 2P j + 2CoA----
2acetyl-CoA + 2CO z + 2ATP + 4NADH + 4H+
IV. Oxidation of Acetyl-CoA via the

Tricarboxylic Acid Cycle
This cycle was discovered by Eggleston and Krebs in animal tissues. It is
often referred to as the Krebs or citric acid cycle. That it is present in
E. coli has been demonstrated with enzymatic methods and with experi-
ments using radioactive substrates. The cycle carries out the oxidation of
the acetyl moiety of acetyl-CoA to CO 2 with transfer of the reducing equi-
valents to NAD+, NADP+, and FAD. Acetyl-CoA enters the cycle by
the citrate synthase reaction in which oxaloacetate and acetyl-CoA are
condensed to give citrate. In the subsequent reactions of the cycle the
C 4 -acceptor for the next molecule of acetyl-CoA, oxaloacetate, is rather
elegantly regenerated.
As shown in Fig. 2.6 citrate is first isomerized to isocitrate. This is
accomplished by a dehydration to enzyme-bound cis-aconitate, which
subsequently is hydrated to give isocitrate. Next, isocitrate undergoes
oxidation to oxalosuccinate, which is decarboxylated by the same enzyme,
isocitrate dehydrogenase, to yield a-oxoglutarate. Like the corresponding
enzyme of most other bacteria the isocitrate dehydrogenase of E. coli is
NADP+ -specific. Oxidation of a-oxoglutarate to succinyl-CoA is catalyzed
by an enzyme complex analogous to the one involved in the oxidation of
pyruvate to acetyl-CoA. It also contains three enzyme species, one cata-
lyzing the decarboxylation of TPP-bound a-oxoglutarate, one accepting
the succinyl moiety and releasing succinyl-CoA, and one with dihydrolipoate
dehydrogenase activity. The dihydrolipoate dehydrogenase is identical
to the corresponding component of the pyruvate dehydrogenase complex,
and it has been shown that E. coli contains a single gene locus for this
enzyme. Thus, newly synthesized dihydrolipoate dehydrogenase will com-
bine with two enzyme components to yield either the pyruvate or the
a-oxoglutarate dehydrogenase complex.
In the next step the energy of the thioester bond of succinyl-CoA is used
to synthesize ATP from ADP and inorganic phosphate. This is another
reaction in which ATP is generated by substrate-level phosphorylation.
Oxidation of Acetyl-CoA via the Tricarboxylic Acid Cycle 21
HO-CH-COOH
I
H20 TH-COOH
J
~CH2-COOH 4
~
CH-COOH isocitrate
~ - COOH oxalo-
X
cis-aconitate
HOi succinate
2 CH 2-COOH
TH2-COOH
HO-C-COOH citrate a-oxoglu tarate
I
CH 2-COOH
CoASH
NADH+W
TO-COOH oxaloacetate yO-SCOA
.-- ..., CH 2- COOH succinyl-CoA yH 2
~IO p.~CH2-COOH
~~ L-malate succinate 7' ADP
HO-CH-COOH ATP
I CH 2-COOH
CH 2-COOH I
' - - 9 ~20 fumarate CH - COOH CoASH
~ 8 2
nH-COOH~
HOOC-CH ~I~
Figure 2.6. Oxidation of acetyl-CoA via the tricarboxylic acid cycle. 1, citrate
synthase; 2 and 3, cis-aconitate hydratase; 4 and 5, isocitrate dehydrogenase; 6,
a-oxoglutarate dehydrogenase complex; 7, succinate thiokinase; 8, succinate
dehydrogenase; 9, fumarase; 10, malate dehydrogenase.
succinate
succinyl-CoA + Pi + ADP lhiokinase) succinate + ATP + CoA

The corresponding mammalian enzyme phosphorylates GDP or IDP
but not ADP. Succinate is oxidized to fumarate by succinate dehydro-
genase. The enzyme resides in particles associated with the cytoplasmic
membrane and transfers electrons from succinate to bound FAD. As
will be discussed later the electrons are then channeled from FAD into
the respiratory chain. That NAD+ is not used as electron acceptor in
this oxidation reaction is because the fumarate/succinate system
(Eo = +0.03 V) has a more positive oxidation-reduction potential as
compared to that of NAD+ /NADH + H+ (Eo = -0.32 V). Thus, succi-
nate is a very weak reductant for NAD+. Protein-bound FAD with Eo of
approximately -0.06 V is much more suitable.
Two additional enzymes are necessary to generate oxaloacetate. First
fumarase hydrates fumarate to L-malate, then L-malate dehydrogenase
oxidizes L-malate to oxaloacetate with NAD+ as H-acceptor.
With the oxidation of two acetyl-CoA via the tricarboxylic acid cycle
glucose is completely oxidized to CO 2 , The hydrogen which theoretically is
available in this oxidation is conserved in the form of reduced coenzymes.

The number of available hydrogens is usually calculated by oxidizing the
substrate (on paper) to CO z with water; for glucose it is 24(H):
C 6 H 1Z0 6 + 6H zO ~ 24(H) + 6CO z
glucose available hydrogen
glucose + 8NAD+ + 2NADP+ + 2FAD + 4ADP + 4P j ~
8NADH + 2NADPH + lOH+ + 2FADH z + 4ATP + 6CO z

, I
24(H)
It is clear that the process of glucose oxidation soon would come to a
standstill if there were not reactions to regenerate the oxidized forms of the
coenzymes. As in other aerobic organisms the principal H-acceptor during
aerobic growth of E. coli is oxygen and the apparatus used to react the
reduced forms of the coenzymes with oxygen is the respiratory chain.
However, in this connection it must be mentioned that E. coli is a
facultative anaerobe and that even under aerobic conditions a part of the
glucose is catabolized via fermentative pathways not involving oxygen. For
the sake of simplicity the simultaneously occurring fermentative metabo-
lism of E. coli will be neglected here and will be discussed in a later
chapter.
v. The Formation of ATP In the

Respiratory Chain
A. Oxidation-reduction potential
An oxidation-reduction (OR) reaction may be written as follows:
A red ~ A ox + n electrons
Box + n electrons ~ Bred
If an equimolar solution of A red / A ox is added to an equimolar solution

of Bred/Box, the direction of any reaction depends on the tendency of the
A red / A ox system to donate electrons to the Bred/Box system and vice versa.
A quantitative measure of this "tendency" in OR systems is their redox
potential.
To measure redox potentials the hydrogen electrode-a solution con-
taining H+ of unit activity (pH = 0) and an inert metal electrode in
equilibrium with Hz at 1 atm-is normally used as reference electrode. Its
oxidation-reduction potential is arbitrarily taken to be zero (Eo = 0 V).
Spontaneously or in the presence of the appropriate catalysts, OR systems
with negative redox potentials reduce H+ to hydrogen. OR systems with a
positive Eo are reduced by Hz. The dependency of the oxidation-reduction
The Formation of ATP in the Respiratory Chain 23
potential on the concentration of the components of the OR system is

expressed by the Nernst equation:
R T [ox]
E = Eo + - - ' In - -
n F [red]
(R, gas constant; T, absolute temperature; n, number of electrons; F,
Faraday constant).
In all reactions involving protons the standard oxidation-reduction
potential refers to pH = O. In that most biological reactions proceed at pH
values near 7 it is more practical to calculate the standard oxidation-
reduction potential of biological systems when the pH is 7.
At pH 7 and 30C the potential of the hydrogen electrode becomes
-0.42 V:
RT
Eo
(pH=7)
Eo
(pH=O)
+ - - . In 10- 7
n F
8.314 x 303
Eo = 0 + 1 x 96494 x 2.303 x -7 = -0.42 V
The Eo values of NAD+ and O 2 are -0.32 and +0.82 V, respectively, and
the difference between them is the potential span of the respiratory chain.
B. Components of the respiratory chain

The major components of the respiratory chain are proteins bearing
prosthetic groups with oxidation-reduction potentials lying between those
of NAD+ and oxygen. In the mitochondrial membrane of eukaryotic
organisms and the cytoplasmic membrane of bacteria, these proteins are
arranged in such a way that the reducing power of NADH can flow to
oxygen via carriers of increasing oxidation-reduction potentials as if over
cascades. However, the composition of the E. coli respiratory chain is not
identical to that of mitochondria (Fig. 2.7). As is indicated by the
rectangles the mitochondrial chain contains four complexes, which can be
isolated as such. Complex 1 is the NADH dehydrogenase; it contains flavin
mononucleotide (FMN) and iron-sulfur proteins and transfers hydrogen
from NADH to coenzyme Q. Succinate dehydrogenase (complex 2) also
feeds hydrogen into the respiratory chain at the coenzyme Q level. This
enzyme is a flavin adenine dinucleotide (FAD)-containing protein.
Cytochrome c is then reduced by coenzyme Q via complex 3 and finally,
complex 4 catalyzes the transfer of reducing power to oxygen.
The composition of the E. coli respiratory chain is different from that of
mitochondria; cytochrome c is not involved, and, most noteworthy, the
E. coli chain is branched. In cells growing under fully aerobic conditions
reducing power flows preferentially via coenzyme Q, cytochrome b 556 and
cytochrome a to oxygen. Oxygen-limited cells employ coenzyme Q or
menaquinone and cytochromes b 558 , a, and d as carriers.
The structural formulas of the four types of carriers present in respira-

tory chains are given in Figs. 2.8 to 2.11. Flavoproteins, coenzyme Q, and
menaquinone are hydrogen carriers. During reduction two hydrogens are
succinate fumarate
coenzyme Q coenzyme Q
( menaquinone )
cyt b SS8
cyt C
cyto
1
cyta,d
i02 H2 0
A
mitochondria I:'. culi
Figure 2.7. Components of the respiratory chains of mitochondria and of E. coli.
The E. coli chain is branched; a dominates in cells growing under real aerobic
conditions, b dominates in oxygen-limited cells. FeS-protein, iron-sulfur protein;
cyt, cytochrome.
o
II
C
"NH
I
c=o
N
r
I
H2
CHOH
I
CHOH
I
CHOH
I
CH 20PO l H2
Figure 2.8. Flavin mononucleotide (FMN), the prosthetic group of the NADH
dehydrogenase of the respiratory chain. Circles indicate where reduction takes
place. Many other enzymes including succinate dehydrogenase contain flavin
adenine dinucleotide (FAD). The oxidation-reduction potential of flavoproteins is
not identical with the potentials of FMN and FAD (Eo = -0.19 and -0.22 V,
respectively). Due to interaction of the protein with its prosthetic groups Eo can be
either more negative or more positive.
@
COi @
l CH3
CH2
/CH"",
C
tH
/CH2+ H
3
J n
b
Figure 2.9. Coenzyme Q (ubiquinone) (8) and menaquinone (b). Circles indicate
where reduction takes place. n varies from 4 to 10; in E. coli, n = 8 for both
quinones.
O=C
I
H
C d
Figure 2.10. The prosthetic groups of the cytochrome types. The prosthetic group
of the a-type cytochromes contain a formyl group as side chain (heme a). The
prosthetic group of the b-type cytochromes is heme b. In cytochromes of the c-type
the prosthetic group is covalently linked to the protein via cysteine residues. The
prosthetic group of the d-type cytochromes is a derivative of dihydroporphyrine.
The structures of R I , R 2 , and R 3 are unknown.
1---------1 [2Fe: 2S 1 center

Figure 2.11. The [2Fe: 2S] center of an iron-sulfur protein. Iron is bound to
cysteine residues of the peptide chain and to sulfide.
transferred to one carrier molecule:

carrier + 2H ( ) carrier - H 2
Cytochromes and iron-sulfur proteins are electron carriers. Four classes
of cytochromes are distinguished: cytochromes a, b, c, and d. The
structural differences between them are indicated in Fig. 2.10. These
differences have an effect on the redox potential of the cytochromes which,
however, is also influenced by the electron affinity of the protein ligands
interacting with the central Fe atom. The redox potential can vary between
Eo = -0.2 V (low-potential cytochromes) and Eo
= +0.4 V (cytochrome
a3)' Cytochrome 0 does not represent a fifth class of cytochromes; it rather
is a b-type cytochrome with oxidase activity. It is oxidized by O 2 like
cytochrome a3 and is present in the E. coli respiratory chain under aerobic
conditions (see Fig. 2.7). In the iron-sulfur proteins the iron is not bound
to a heme group. It is bound to the sulfhydryl groups of cysteine residues of
the protein. In addition, iron atoms are linked to one another by sulfur
bridges so that iron-sulfur centers are formed. Most of these are of the
[2Fe :2S] type (Fig. 2.11) or of the [4Fe :4S] type (see Chapter 8).
During reduction of a cytochrome or an iron-sulfur protein one electron is
transferred to the central iron of the cytochrome and to one of the irons of the
FeS center of the iron-sulfur protein, respectively:
carrier - Fe3 + + H. ) carrier - Fe2+ + H+
Thus, if electron carriers are reduced by hydrogen carriers, protons are
released. Conversely, the reduction of hydrogen carriers by electron
carriers requires protons. This is important, because hydrogen and elec-
tron carriers arranged in alternating sequence in a membrane may cause
proton translocations; protons released may be excreted at one side of the
membrane and protons required may be taken up from the other side:
2H+
reduced t _ reduced iron- reduced
flavoprotein ~ sulfur proteinr coenzyme Q
2H+
Proton translocations are the basis for the chemiosmotic hypothesis of
ATP formation during respiration.
c. Electron transport phosphorylation

The redox potentials of the components involved in the respiratory chain
are given in Table 2.2. It can be seen that the values for the carriers are
such that electrons can flow from NADH via these carriers to 2 , A 0.1 V
increase of the Eo
values during electron transport corresponds to a free

energy change of IlGo, = -19.2 kJ (-4.6 kcal) per mol. The value for the
oxidation ofNADH with oxygen (!:!.E~ = 1.14 V) is then !:!.Go, = -218.9 kJ
(-52.4 kcal) per mol. If the respiratory chain contained only those
components discussed thus far, for instance, as soluble proteins in the
cytoplasm, this energy would have to be released as heat. Yet it has been
known for more than 30 years from the work with mitochondrial systems
that part of this energy is used to synthesize ATP from ADP and inorganic
Table 2.2. Standard oxidation-reduction potentials of respiratory chain

components at pH 7.0 and 30C
component Eo (V)
NAD+ + 2H+ + 2e- ( 'NADH + H+ -0.320
FAD + 2H+ + 2e- ( FADH z -0.220
FMN + 2H+ + 2e- ( FMNH z -0.190
fumarate + 2H+ + 2e- ( 'succinate +0.033
f1avoproteins + 2H+ + 2e- ( 'red. f1avoproteins -0.450-0.0
FeS-proteins + 2e- ( red. FeS-proteins -0.400-+0.200
menaquinone + 2H+ + 2e- (--2. red. menaquinone -0.074
ubiquinone + 2H+ + 2e- ( 'red. ubiquinone +0.113
2 cyt box + 2e- ( '2 cyt bred +0.070
2 cyt Cox + 2e- ( 2 cyt Cred +0.254
2 cyt a ox + 2e- ( '2 cyt arcd +0.384
1/20 z + 2H+ + 2e- ( HzO +0.818
a Due to interaction of the prosthetic groups with the protein, the Eo value of a
certain flavoprotein or FeS-protein can be somewhere within this range.
phosphate. This process is called oxidative or electron transport phosphory-

lation and was first demonstrated by Kalckar and Belitser with minced liver
and muscle tissue. It is a prerequisite of electron transport phosphorylation
to occur that the components involved in the redox reactions are membrane-
bound. In eukaryotic organisms they are localized in the inner membrane
of the mitochondria and in prokaryotes in the cytoplasmic membrane.
To demonstrate electron transport phosphorylation in vitro, e.g., in
subcellular fractions, is very difficult. Some components of the chain might
become soluble during preparation of subcellular fractions. Thus sophisti-
cated methods are necessary to reconstitute a chain able to catalyze
simultaneously the oxidation of NADH and an efficient electron transport
phosphorylation. Through the work of Green, Racker, Lehninger,
Chance, and others it is now known that there are three phosphorylation
sites at the respiratory chain of mitochondria. Thus, three ADP can be
phosphorylated per two electrons transferred from NADH to oxygen. The
number of ADPs phosphorylated per atom oxygen is frequently expressed
as the P:O ratio; in the mitochondrial chain it is 3 for NADH as electron
donor and 2 for FADH z as electron donor. Even the sites of ATP
formation are known: the dehydrogenation of NADH, the oxidation of
cytochrome b, and the oxidation of cytochrome a (see Fig. 2.7).
The bacterial respiratory chain seems to be even more sensitive to
isolation procedures, and subcellular fractions often have completely lost
the ability to couple electron transport with the phosphorylation of ADP.
Fractions most frequently studied are vesicle preparations. Bacterial cells
are disrupted by ultrasonication or other methods. Under appropriate con-
ditions the membrane fragments produced stabilize themselves by forming
vesicles. The membrane of the vesicles can have the same or the opposite
orientation as the membrane in whole cells. The two types of vesicles are
called right-side-out or inside-out vesicles (Fig. 2.12). The inside-out
vesicles are very suited for studying electron transport phosphorylation.
wall
cytoplasmic membrane
wall fragments
---- -0-
-- V
inside-out
vesicles
@ @ @_ right-side-out
@ vesicles
Figure 2.12. Inside-out and right-side-out vesicles of E. coli. Vesicles are prepared
either by ultrasonication of whole cells or by conversion of the cells into
sphaeroplasts with lysozyme-EDTA (destabilization of the wall) and subsequent
breakage of the sphaeroplasts by osmotic shock. The first procedure yields a
preponderance of inside-out vesicles and the second one a preponderance of
right-side-out vesicles.
All sites at the membrane that normally react with components of the
cytoplasm are exposed now to the medium in which the vesicles are
suspended. Appropriate reagents can be added and experiments can be
done. The first P: 0 ratios reported from studies of such vesicle prepara-
tions were generally below 1 and it was assumed that electron transport
phosphorylation in bacteria is less efficient than in mitochondria. With
improved methods, which largely prevented structural damage of the
vesicles, it has been possible to obtain P: 0 ratios greater than 2 for a
number of bacterial species. Thus it appears safe to conclude that at least
some aerobic bacteria are able to oxidize NADH via the respiratory chain
with a P: 0 ratio of 3; for E. coli this value is probably 2, and the sites of
ATP formation are the dehydrogenation of NADH and the oxidation of
one of the cytochromes.
How then is ATP synthesized? Great efforts have been made in many
laboratories over the years to understand the mechanism of electron
transport-coupled ATP synthesis. For some time it was assumed that an
energy~rich intermediate is generated in the redox reactions (as in the
glyceraldehyde dehydrogenase reaction; see Fig. 2.4). Such an inter-
mediate could subsequently give rise to the formation of ATP from ADP.
Experimental evidence for such a mechanism could not be obtained. In
1961 Mitchell proposed the chemiosmotic hypothesis of ATP formation by
electron transport phosphorylation. This hypothesis has been confirmed in
recent years, and it is not exaggerated to state that the chemiosmotic
hypothesis is one of the most fundamental contributions made in the area
of bioenergetics.
The chemiosmotic theory presumes that:
1. The cytoplasmic membrane is impermeable to OH- and H+.
2. The respiratory chain is localized in the membrane in such a way that a
pH gradient and a membrane potential are formed by vectorial extrac-
tion and excretion of protons during electron transport (proton
translocation) ;
3. The ATP synthase is so ingeniously constructed that it can take
advantage of the pH gradient and the membrane potential for ATP
synthesis from ADP and Pi.
The cytoplasmic membrane is composed of a phospholipid bilayer in
which a number of proteins are embedded. Details about its composition
will be given in Chapter 3. Here it is sufficient to state that the membrane
per se is impermeable to charged and uncharged hydrophilic compounds.
Only uncharged lipophilic or charged highly lipophilic compounds can pass
the membrane without carrier systems. Acetic acid or butyric acid can pass
the membrane but the acetate or butyrate anions cannot. Thus, the
membrane is permeable to these compounds only at low pH values. A
charged highly lipophilic compound is, for instance, the tetra-
phenylphosphonium cation [(phenyl)4 - p+)). If there is a difference in pH
between the cytoplasm and the medium surrounding the cells, this
difference will not be evened out across the membrane in the absence of
the appropriate carrier systems; the cytoplasmic membrane per se is
impermeable to H+ or OH-; it has a very low conductance.
Proton translocation across the cytoplasmic membrane is achieved by
different mechanisms. The loop mechanism is the most plausible one. It
takes advantage of the fact that electron and hydrogen carriers are
involved in the redox reactions of the respiratory chain. These carriers are
arranged such that two hydrogens are taken up by a hydrogen carrier from
the cytoplasm and that an electron carrier facing the outer surface of the
membrane is reduced by a hydrogen carrier. The two protons not needed
for this reduction are then released at the outer surface (Fig. 2.13a). A
second proton-translocating mechanism is represented by the coenzyme Q
cycle (Fig. 2.13b). Its principle is that the reduced form of coenzyme Q
(ubiquinone) releases one electron and two protons at the outer surface of
the membrane. The resulting semiquinone anion can then take up two
out in
NADH+ H+
NAD+
2W
2H
2W
Q- b
le-
2e-
xH+ xH+
Figure 2.13. Mechanisms for proton translocation. a: Loop mechanism, protons are
extracted by alternating sequences of hydrogen and electron carriers. b: Coenzyme
Q cycle. The reduced form of coenzyme Q releases two protons and transfers one
electron to the adjacent carrier. c: Proton pump, such a mechanism is assumed to
occur in complex 4 of the mitochondrial respiratory chain.
membrntl\? cytoplasm
(NADH+W
fbvoprotein ' - NAD+
2H+
FeS-protein
---------...
flavo-
2H+
protein
2H+
coenzyme Q
----
---...
cyt bS56
cyt 0
Figure 2.14. Functional organization of the components of the E. coli respiratory

chain. [B. A. Haddock and C. W. Jones, Bacterial. Rev. 41, 47-99 (1977)]. Not
shown in this scheme is the less ATP-yielding branch with the cytochromes b SS8
and d (see Figure 2.7).
further protons and one electron at the inner surface. Thus, two protons
are translocated per one electron and not only one proton as by the loop
mechanism. Complex IV of the respiratory chain of mitochondria is
composed of cytochromes only. A hydrogen carrier is lacking, and it is
assumed that a proton pumping mechanism is responsible for proton
translocation in this complex (Fig. 2.13c).
A scheme of the functional organization of the redox carriers in the
respiratory chain of E. coli, as has been proposed by Haddock and Jones,
is shown in Fig. 2.14. It indicates that this chain comprises two proton-
translocating sites. At the second site additional protons may be extracted
by a coenzyme Q cycle. The mitochondrial chain would comprise a third
site (proton pump) in the cytochrome a/cytochrome a3 complex (see
Fig. 2.7).
Through proton translocation a protonmotive force (i1P) is generated
across the membrane. It consists of two components:
(i) the membrane potential i1l/!, protons are positively charged, and the
inner surface of the membrane becomes negatively charged by proton
extraction, (ii) the proton gradient between outside and inside, it is usually
expressed as .lpH. The corresponding equation is:

.lP = iltf; - Z ilpH
where Z = 2.3 RT/ F and is equal to 59 at 25C. .lP can be related to the
electrochemical gradient of H+ (.lJLw) through the Faraday constant:
.lp. F = .lJLw
E. coli growing on a medium with a pH of 6 has an internal pH of 7.8 and a
membrane potential of.ltf; = -95 mY. The protonmotive force amounts to
approximately -200 mY.
AlP synthesis at the AlP synthase

The protonmotive force represents the energy form in which part of the
free energy change of NADH oxidation is conserved. It is taken advantage
of by the ATP synthase for ATP formation. The enzyme ATP synthase
comprises a very complex structure (Fig. 2.15). It consists of two parts, Fo
and Fl' The Fo part is made of membrane-integrated proteins which form a
pore in the membrane through which protons can pass under certain
conditions. Five different proteins (subunits) make up the F 1 part: a, f3, ')',
8, c. Their structural organization is apparent from Fig. 2.15. Three a and
three f3 subunits form a hexarneric structure that is connected with F o via
the ')', 8, and c subunits. When cytoplasmic membrane preparations are
investigated by electron microscopy the inner surface can be seen to be
covered by white spots. The spots are the F 1 particles. They can be
removed from the membrane by treatment with EDTA. The ATP
synthase, that is the FIFo complex, allows protons to flow back into the cell
if ADP and inorganic phosphate are available. By a mechanism not fully
understood ATP is formed at the expense of the electrochemical gradient
pore
Figure 2.15. Structure of ATP synthase. The a and f3 subunits alone exhibit activity
of ATP hydrolysis. For ATP synthesis they have to be attached to Fo via the y, e,
and 0 subunits.
NADH+W
4W
+
+
2-4W -+_F-'o:...-_I-_....:..._f-....
ATP
Figure 2.16. Chemiosmotic theory of ATP synthesis. Oxidation of NADH via the
respiratory chain leads to the extraction of protons from the cytoplasm. The loop
mechanism and the Q cycle are indicated in the figure. A ~pH and a ~l/J are
established. The protonmotive force is then used by the FoF 1 complex for ATP
synthesis.
of H+. As indicated in Fig. 2.16, per molecule of ATP synthesized two to

four protons enter the cell; the most likely number is three.
This is the mechanism of electron transport phosphorylation. It now can
be said that ATP is synthesized by a chemiosmotic mechanism.
D. Uncouplers and inhibitors

In growing cells of E. coli the oxidation of NADH and the phosphorylation
of ADP are coupled. This coupling is largely lost when subcellular fractions
are prepared. As a result, the P: 0 ratio is low and only little ATP is
formed during oxidation of NADH. In vivo this uncoupling can also be
achieved by the addition of certain compounds to the cells. The classic
uncoupling agent introduced by Loomis and Lipman is 2,4-dinitrophenol.
NADH
rotenone, amy tal
coenzyme Q
or
menaquinone
t-
cy tochrome b
HQNO
f-
cytochrome 0
antimycin A
or
f-
cytochrome d
CW,CO
2
Figure 2.17. Sites of action of electron transport inhibitors in E. coli.
Glucose is oxidized almost completely in the presence of 2,4-

dinitrophenol, and cell material is not synthesized. In the absence of
2,4-dinitrophenol about 50% of the carbon of glucose is released as CO 2
with the meaning 50% being converted into cellular material:
growth .
C 6 H 12 0 6 + 302 ~ cell matenal + 3C0 2 + 3H zO
C H 0
6 I2 6
+ 602 dinitrophenol) 6CO + 6H 0
z 2
Further uncouplers are carbonylcyanide-p-trifluoromethoxyphenyl hyd-

razone (FCCP) or tetrachlorosalicylanilide (TCS). All these agents make
the cytoplasmic membrane permeable for protons. As a consequence a
ApH cannot be established, and ATP cannot be synthesized by electron
transport phosphorylation. Because of their mode of action uncouplers are
now also called protonophores.
The respiratory chain can also be impaired by inhibitors of electron
transport. Such inhibitors are rotenone, amytal, antimycin, 2-n-heptyl-4-
hydroxyquinoline-N-oxide (HQNO), potassium cyanide, and carbon
monoxide. Their sites of action are indicated in Fig. 2.17.
Some compounds specifically inhibit the ATP synthase, for example,
the antibiotic oligomycin and dicyclohexylcarbodiimide (DCCD). In the
presence of these inhibitors electron transport can only be observed when a
protonophore is also present.
E. The importance of superoxide dismutase

Concentrations of oxygen higher than that found in air are toxic to many
aerobic microorganisms. Strict anaerobes such as clostridia and
methanogenic bacteria die when they are exposed to air. For some time
this deleterious effect of oxygen was related exclusively to the accumula-
tion of its reduction product, hydrogen peroxide (H 2 0 2 ). The latter is
formed whenever reduced flavoproteins or reduced iron-sulfur proteins
Summary 35
come together with oxygen and oxidases that are present in all organisms.
FADH z ( FAD + 2H+ + 2e-
Oz + 2e- + 2H+ ~ HzO z

Aerobes, therefore, contain catalase, which converts HzO z to oxygen
and water:
2H zO z ~ 2H zO + Oz
Catalase is indeed very important. However, it is now apparent that
a compound more toxic than HzO z is produced from oxygen in biological
systems. This compound is the superoxide radical (0;-), which is formed
by a univalent reduction of oxygen with reduced flavins, quinones, or other
electron carriers:
20 z + 2e- + 2H+ ~ 20;- + 2H+

E. coli and all aerobic and aerotolerant microorganisms contain the
enzyme superoxide dismutase, which converts the radical to HzO z and Oz.
superoxide
0;- + 0;- + 2H+ dismutase) HzO z + Oz
E. coli is able to form two superoxide dismutases, one is an iron-protein

and the other one is a mangano-protein, whereas the enzyme of mammals
contains copper or zinc.
VI. Summary
1. Escherichia coli contains a PEP phosphotransferase system, which is
responsible for the uptake of glucose. Transport of the sugar into the cell is
coupled to its phosphorylation to glucose-6-phosphate.
2. Glucose-6-phosphate is degraded to pyruvate via the Embden-
Meyerhof-Parnas pathway. Key intermediate of this pathway is fructose-
1,6-bisphosphate. In the conversion of glucose-6-phosphate to pyruvate
ATP is required for the phosphofructokinase reaction and ATP is pro-
duced in the 3-phosphoglycerate kinase and pyruvate kinase reactions.
3. The oxidative decarboxylation of pyruvate to acetyl-CoA is accom-
plished by the pyruvate dehydrogenase complex. This complex consists
of three enzymes: a pyruvate dehydrogenase, a dihydrolipoate trans-
acetylase, which transfers the acetyl residue to coenzyme A, and a
dihydrolipoate dehydrogenase, which transfers hydrogen from enzyme 2 to
NAD+.
4. Acetyl-CoA is oxidized in the tricarboxylic acid cycle. The oxidation
yields NADPH in the isocitrate dehydrogenase reaction, NADH in the
a-oxoglutarate and malate dehydrogenase reactions, and FADH z in the

succinate dehydrogenase reaction. ATP is formed in the conversion of
succinyl-CoA to succinate.
5. The respiratory chains of E. coli and of mitochondria are different.
The E. coli chain does not contain a cytochrome of the c-type; further-
more, it is branched. The transport of two electrons from NADH to
oxygen is coupled to the phosphorylation of probably two ADPs. For the
mitochondrial chain and the chain of some other bacteria this value is
three.
6. ATP synthesis by electron transport phosphorylation can now be
readily explained by the chemiosmotic theory of Mitchell. In the course of
electron transport, protons are translocated from the cytoplasm to the
space outside of the cytoplasmic membrane. The resulting protonmotive
force (AP) which consists of a proton gradient (ApH) and a membrane
potential (AljJ) drives ATP synthesis at the membrane-bound enzyme ATP
synthase (FoFd.
7. During biological oxidations the toxic superoxide radical O~ - is
formed in small concentrations. Superoxide dismutase converts this com-
pound into HzO z and Oz. Catalase splits HzO z into HzO + 1/20z .
8. The overall equation for glucose oxidation by E. coli is:
glucose + 8NAD+ + 2NADP+ + 2FAD + 4ADP + 4P j ~
8NADH + 2NADPH + lOH+ + 2FADH z + 4ATP + 6CO z
The oxidation of lONAD(P)H via the respiratory chain yields maximally
20ATP and the oxidation of 2FADH z yields maximally 2ATP.
Chapter 3
Biosynthesis of
Escherichia coli Cells
from Glucose
It has already been mentioned that E. coli-when growing aerobically on

glucose-oxidizes about 50% of the glucose to CO 2 for the production of
ATP. The remaining 50% is converted into cellular material. It is this
conversion that consumes most of the ATP formed in oxidation. In this
chapter the principal biosynthetic reactions involved in synthesis of cellular
material and the main sites of ATP consumption will be outlined.
I. Composition of E. coli Cells

More than 95% of the cellular material of E. coli and of other micro-
organisms consists of macromolecules. A typical analysis of microbial cells
is given in Table 3.1. Proteins account for approximately 52% and nucleic
acids for 19% of the dry weight. About 3% of the dry weight of the cells
includes low-molecular-weight organic compounds and salts.
Figure 3.1 presents a general scheme for the formation of the cellular
constituents. Intermediates of glucose breakdown (hexose phosphates,
PEP, pyruvate, acetyl-CoA, oxaloacetate, a-oxoglutarate) are used as
starting material to make all the required amino acids, vitamins, sugar
phosphates, fatty acids, ribo-, and deoxyribonucleotides. Polymerization
reactions then lead to the formation of macromolecules.
Most of the biosynthetic reactions of E. coli are now known, and it is
thus possible to calculate the amount of ATP required for the formation of
the macromolecules (Table 3.2). At first glance it is surprising to learn that
from 34.8 mmol of ATP necessary for the synthesis of 1 g of cell material,
19.1 mmol (56%) have to be invested in polymerization of amino acids. A
38 3: Biosynthesis of Escherichia coli Cells from Glucose
Table 3.1. Content of macromolecules of

microbial cells Q
amount
macromolecule (g/lOO g of dried cells)
protein 52.4
polysaccharide 16.6
lipid 9.4
RNA 15.7
DNA 3.2
total 97.3
a A. H. Stouthamer, Antonie van Leeuwenhoek

39, 545-565 (1973).
Note: These figures of composition are repre-
sentative but the actual composition changes
markedly among different (e.g., Gram-positive
versus Gram-negative) bacteria and varies with
differing growth conditions (e.g., RNA con-
tent).
I glucose I
r~;
monomers macromolecules
amino acids ----t.~ proteins
vitamins
pyruvate sugar phosphates _ polysaccharides
tetYI,coA~
fatty acids .. lipids
ribonucleotides _ RNA
deoxyribonucleotides ... DNA
",oxoglutarate
Figure 3.1. General scheme of the biosynthesis of cell material from glucose.
Composition of E. coli Cells 39
Table 3.2. ATP requirement for the formation of microbial cells from glucose and
inorganic salts a
ATP (mmol) required for the

synthesis of the macromolecule
macromolecule content of 1 g of dried cells
polysaccharide 2.1
protein
glucose ~ amino acids 1.4
polymerization 19.1
lipid 0.1
RNA
glucose ~ nucleoside monophosphates 3.5
polymerization 0.9
DNA
glucose ~ deoxynucleoside monophosphates 0.9
polymerization 0.2
5.2
ATP required for transport processes (salts)
1.4
ATP required for RNA turnover
34.8
total ATP requirement
a A. H. Stouthamer, Antonie van Leeuwenhoek 39, 545-565 (1973).
comparatively small percentage of the available ATP is required for the

formation of the amino acids from glucose. During RNA and DNA
synthesis, the reactions leading from glucose to the appropriate nucleoside
monophosphates are more ATP-consuming than the final polymerization
reactions. A considerable amount of ATP is also required to compensate
for RNA turnover. The half-life of messenger RNA is approximately 3 min
and during the time the cell mass of a culture doubles, the mRNA content
of the cells has to be synthesized several times. Finally, about one-seventh
of the ATP is invested in transport processes; of this the greatest
proportion is required for the uptake of ammonium ions.
In conclusion, a culture of E. coli growing aerobically on glucose has to
invest the major amount of ATP gained by respiration in the polymeriza-
tion of amino acids. Smaller amounts of ATP flow into the biosynthesis of
monomers, other polymerization reactions, and transport processes. Not
considered here are the energy of maintenance, which depends very much
on the growth conditions and may represent 10-20% of the ATP-
consuming reactions in growing cells and the energy of flagellar movement,
which is comparatively small.
In the following sections some of the biosynthetic reactions of E. coli
will be discussed.
II. Assimilation of Ammonia

If the concentration of ammonia in the environment of the E. coli cells is
high, it is assimilated by reductive amination of an intermediate of the
tricarboxylic acid cycle, a-oxoglutarate. The enzyme catalyzing this reac-
tion is a NADP+ -specific L-glutamate dehydrogenase. From the L-glutamate
thus formed the amino group can then be transferred to a-oxoacids.
E. coli contains three enzymes, transaminase A, Band C, responsible for
catalysis of this amino group transfer; their low specificity allows the
formation of more than 10 amino acids from the corresponding a-
oxoacids. Transaminase A is preferentially involved in L-asparatate and
L-alanine synthesis, transaminase B in the synthesis of the aromatic amino
acids, and transaminase C in the synthesis of the branched-chain amino
acids.
Figure 3.2 shows the glutamate dehydrogenase reaction and the forma-
tion of L-valine from a -oxoisovalerate as an example of a transaminase
reaction.
A second important reaction for the assimilation of ammonia is the
L-glutamine synthetase reaction:
COOH COOH
I I
H 2 N-CH H2 N-CH
I I
CH 2 + ATP + I NH 3 1 CH 2 + ADP + Pi
I Lglutamine synthetase
I
CH 2 CH z
I I
COOH CO~
Glutamine serves as NH z donor in the biosynthesis of a number of
a-oxoglutarate
HOOC- CO-CH 2 - CH 2-COOH
L-valine
CH
3:: CH-CO-COOH
a
CH 3
HOOC-CH-CH 2 -CH 2-COOH
a-oxoisovalerate
L-glutamate
Figure 3.2. Assimilation of ammonia by L-glutamate dehydrogenase (1) and

transfer of the amino group in a transamination reaction (2).
Assimilation of Ammonia 41
compounds such as purines, tryptophan, histidine, glucosamine-6-

phosphate, and also in the formation of carbamoyl phosphate, which is one
of the precursors of the pyrimidine compounds:
COOH
I
H2N-CH HC0 3
I
c~ + ,
I carbamoyl phosphate synthetase
CH 2 2ATP
I
CO--{E!Q
COOH
I
H2N-CH ~CO-OP03H2
I
CH 2 +
I
CH 2 + 2ADP + P;
I
COOH
In recent years it has become clear that the glutamate dehydrogenase is

not involved in the primary assimilation of ammonia at low concentrations
of ammonia 1 mM). Its K m value for ammonia is high (-10- 1 M) and
the enzyme cannot function efficiently under these conditions. It has been
shown, first by Tempest and later by other investigators, that at low NH 3
concentrations a combination of glutamine synthetase and glutamate
synthase is responsible for glutamate formation from a-oxoglutarate and
ammonia. The latter enzyme-which is frequently called GOGAT (gluta-
mine: a-oxoglutarate aminotransferase)-catalyzes the reductive transfer
of the amido group of glutamine to a-oxoglutarate. The pathway, which is
widespread among bacteria, is illustrated in Fig. 3.3. Its advantage in
EiJ + ATP L-glutamate IL-glutamate I

NADP'
L-glutamine synthetase L-glutamate synthase
NADPH +H+
ADP + Pi L-glutamine
I O'-oxoglutarak I
Figure 3.3. Assimilation of ammonia by glutamine synthetase/glutamate synthase
(GS/GOGAT).
comparison to the L-glutamate dehydrogenase pathway is apparent.

Ammonia is assimilated in an ATP-consuming reaction; the equilibrium
constant of the glutamine synthetase reaction is 103 in favor of the
products, and the K m value for ammonia is approximately 1O- 4 M. The
GS/GOGAT pathway is, therefore, well suited for the assimilation of
ammonia in low concentration.
III. Assimilatory Reduction of Sulfate

As in most bacteria E. coli uses sulfate as the principal sulfur source.
However, this sulfate must be reduced since sulfur of the majority of
sulfur-containing cellular compounds is at the oxidation level of H 2 S. The
path of sulfate reduction is summarized in Fig. 3.4. First, sulfate is actively
transported into the cell and then activated by the ATP sulfurylase
reaction. Of the adenosine-5'-phosphosulfate (APS) and pyrophosphate
thus formed, the latter may be hydrolyzed by a pyrophosphatase, thereby
Isulfate I
(Outside)~
I L-cysteine I sulfate
(inside)
acetate ~
O-acetyl- 6
=t
L-serme
Isulfide I APS
V--
t'---
3NADP' 3H 2 0 3 ATP
3NADPH ADP
+3H' AMP-3'-phosphate H2 0
sulfite .....I----"!::=-..::::::;;
...- ..-.-<E---- PAPS
/S-;:::- 4 -=-::;::::
R,~ R-(SH)2
Figure 3.4. Assimilatory reduction of sulfate and formation of cysteine. 1, Active

transport of sulfate; 2, ATP sulfurylase; 3, APS phosphokinase; 4, PAPS
reductase; 5, sulfite reductase; 6, O-acetylserine sulfhydrylase. R(SH)z stands for
thioredoxin, its reduced form is regenerated from the oxidized form with NADPH.
AMP-3'-phosphate is hydrolyzed to AMP and Pi'
Biosynthesis of Amino Acids 43
favoring APS synthesis. The subsequent phosphorylation of APS in the

3' -position yields adenosine-3' -phosphate-S'-phosphosnlfate (PAPS).
R=H, adenosine-5'-phosphosulfate (APS)

R=P0 3 H z, adenosine-3'-phosphate-5'-phosphosulfate (PAPS)
The reduction of PAPS requires involvement of a dithiol compound

(thioredoxin) and leads to the release of sulfite. The enzyme sulfite
reductase then catalyzes a 6-electron transfer from 3NADPH to sulfite with
the formation of HzS. Since this is very toxic to many microorganisms, the
HzS is immediately incorporated into O-acetylserine. The product formed,
L-cysteine, is the most important precursor of sulfur-containing compounds
of the cell.
The pathway just outlined is referred to as assimilatory sulfate reduc-
tion. A mechanistically different dissimilatory type of sulfate reduction is
found in certain anaerobic bacteria (see Chapter 8).
IV. Biosynthesis of Amino Acids

Twenty amino acids are required for the biosynthesis of proteins, and they
are formed from the metabolic precursors listed in Table 3.3. It is evident
from this table that only a few compounds serve as substrates in amino acid
synthesis. Oxaloacetate, for instance, is the starting point for the synthesis
of six amino acids, a-oxoglutarate is the precursor of four, and pyruvate of
three amino acids.
A. The oxaloacetate and pyruvate families of

amino acids
An inspection of the metabolic routes leading to the nine amino acids of
these families reveals that common pathways are frequently employed.
This and the important role of oxaloacetate and of pyruvate as building
blocks are apparent from Fig. 3.5. It also can be seen from this figure that
L-asparate and L-threonine are both products and intermediates.
Table 3.3. Precursors for amino acid biosynthesis
precursor amino acid
pyruvate L-alanine
L-valine
L-leucine
oxaloacetate L-aspartate
L-asparagine
L-methionine
L-lysine
L-threonine
L-isoleucine
a-oxoglutarate L-glutamate
L-glutamine
L-arginine
L-proline
3-phosphoglycerate L-serine
glycine
L-cysteine
phospho-enolpyruvate L-phenylalanine
erythrose-4-phosphate L-tyrosine
L-tryptophan
5-phosphoribosyl-l-pyrophosphate + ATP L-histidine
Simple transamination reactions are employed for the formation of

L-aspartate from oxaloacetate and of L-alanine from pyruvate:
COOH COOH
I I
co L-g)utamale a-oxoglutarate
H2 N-CH
I ">-- / I
CH 2 transaminase CH 2
I I
COOH COOH
COOH l-glulamatc ,.-oxoglutaratc COOH

I __">---::::_..::L:::"",,_-+) H 2 N -tH
CO
I transaminase I
CH 3 CH 3
Iaspartate 1--1 asparagine I

t
aspartate- 4-phosphate
t
~~
dihydrodipicolinate homoserine
*t
diaminopimelate
h~C~
phosphate homocysteine
t t t
Ithreonine I Imethionine I
J
r
a-oxobu tyrate -----,. I alanine I
a-aceto-
~<! '"
a-aceto-
a-hydroxy bu tyrate lactate
t ~ ~ t
t
2-oxo-3-methyl-
~ ~
a-oxo-
t acetyl-CoA
\ ,. a-isopropyl-
valerate isovalera te malate
< > ~
I isoleucine I
Figure 3.5. Biosynthesis of amino acids derived from oxaloacetate and pyruvate.
L-Aspartate is converted into L-asparagine by L-asparagine synthetase:
COOH COOH
I I
HzN-CH HzN-CH
I + NH 3 + ATP ~ I + ADP + P;
CH z synthelasc CH z
I I
COOH CO-NH z
Aspartate semialdehyde, which is synthesized in two steps from L-

aspartate, is a central intermediate for the synthesis of L-methionine,
COOH NADH + H+ NAD+ fOOH

I
HzN-fH -_'>..':::"'_<~--l'~ HzN-fH aspartate
semialdehyde
fH z 2 Pi fH Z
COOH CHO
~ NADH+W
L-aspartate aspartate- 4-phosphate
3
~NAD+
COOH
I
HzN -fH homoserine
fH z
COOH COOH ADP ATP CHzOH
I I
HzN-fH HzN-TH
.\.. ./
CHOH
I
CH 3
T Hz
CH ZOP0 3 Hz
4
succinyl-CoA
L-threonine homoserine phosphate 6

CoA
COOH COOH
I I
HzN-TH 7Hz O-succinyl
Hz Hz homoserine
CHz-O-CO
COOH
I
HzN-yH
CH,
I -
CH zSCH 3
L-methionine homocysteine cystathionine

Figure 3.6. Synthesis of L-threonine and L-methionine from L-aspartate. 1,
Aspartate kinase; 2, aspartate semialdehyde dehydrogenase; 3, homoserine de-
hydrogenase; 4, homoserine kinase; 5, threonine synthase; 6, homoserine acyl-
transferase; 7, cystathionine synthase; 8, cystathionine lyase; 9, homocysteine:
N 5 - methyItetrahydrofolate methyltransferase .
L-threonine, L-isoleucine, and L-lysine. As shown in Fig. 3.6, it is reduced

to homoserine by homoserine dehydrogenase. Via a phosphorylated
intermediate the hydroxyl-group is shifted from carbon 4 to carbon 3, and
L-threonine is thus formed. Another enzyme acting on homoserine is
homoserine acyltransferase. Its reaction product is succinylhomoserine.
The succinyl residue is then exchanged with a cysteinyl residue, and

elimination of ammonia and pyruvate yields homocysteine. The final step
in L-methionine synthesis is the methylation with N 5-methyltetrahydro-
folate as donor of the methyl group.
For the synthesis of lysine, aspartate semialdehyde is condensed with
pyruvate to yield 2,3-dihydrodipicolinate (Fig. 3.7). Reduction and ring
NADPH+W
0
HOOC N""" COOH
aspartate 2, 3-dihydrodi- tetrahydrodi-

semialdehyde picolinate picolinate
succinyl-CoA
COOH COOH
I I
TO <[Hz
(CH Z)3 CH z
I I
COOH CH-NH-CO
I I
HzN-CH COOH
I
(CHzh
I N-succinyl-2-
CHz-NH z amin0-6-oxo-
L-Iysine pimelate
meso-a, - L, L-Q. - N-succinyl-2,6-

diamino- diamino- diaminopimelate
pimelate pimelate
Figure 3.7. Synthesis of L-Iysine. 1, Dihydrodipicolinate synthase; 2, dihydro-
dipicolinate reductase; 3, tetrahydrodipicolinate succinylase; 4, glutamate: suc-
cinyl-diaminopimelate aminotransferase; 5, succinyl-diaminopimelate desuccinyl-
ase; 6, diaminopimelate epimerase; 7, diaminopimelate decarboxylase,
opening by succinylation gives a, e-diaminopimelate. The L,L form of this

compound is converted into the meso form which is finally decarboxylated
to L-lysine.
The synthesis of the remaining three amino acids of this family starts
with pyruvate and a-oxobutyrate. The latter is formed from threonine by
the threonine deaminase (Fig. 3.8). Interestingly enough both compounds
L-threonine
I'CH] - CH,-CO-COOH I CfH] ~H,

CH,- CH-CH,- CH-COOH
o:-oxobutyrate pyruvate L-Ieucine
~2 ~
-1
ClI-OXOglutarate....J 5
CO'1
pyruvate pyruvate 2
CO'1 glutamate
~H,-CH] ~H] ?H,

CH]-CO-?-COOH CH]-CO-?-COOH CH]- CH-CH,-CO-COOH
OH OH
a-acelo-(k-hydroxy a-Qxoisocaproa te
Q-3cetolactate
l3
butYra~te
NADPH + W NADPH+W
N A D H + w t C01
3
NADP' NADP'1 NAD'
~H,-CH] ?H] CfH, ?H

CH'-7-7H-COOH CH,-C-CH-COOH CH,-CH-CH-CH-COOH
OH OH 6H 6H I
COOH
ClI. P-dihydroxy- ClI. P-dihydroxy-
(j-isopropylmalate
:~.:'.." "'~~.:. 1,
acetyl-{;oA ('oA
?H] ?H] J_
"--
~H, ?H
CH,- CH-C-CH2- COOH
CH]- CH,- CH -CO -COOH CH]-CH-CO-COOH ~ I
ClI-oxo-il-melhyl- COOH
H,O
valerate o:-oxoisovalerate
a-isopropylmaJate
glutamate ~ glutamate ~
ClI-Oxoglutarate15 ClI-oxogiutarate ~ 5
CH] NH, TH] ~H,

CH]-CH 2-tH-tH-COOH CH,- CH-CH-COOH
L-isoleucine L-valinc
Figure 3.8. Synthesis of L-isoleucine, L-leucine, and L-valine. 1, Threonine

deaminase; 2, acetohydroxy acid synthase; 3, acetohydroxy acid isomeroreductase,
the alkyl group (methyl- or ethyl-) is shifted from the a-carbon to the f3-carbon
with simultaneous reduction of the oxo group; 4, dihydroxy acid dehydratase; 5,
transaminase C; 6, a-isopropylmaiate synthase; 7, isopropylmalate isomerase; 8,
f3-isopropylmalate dehydrogenase.
are converted by the same set of four enzymes into L-isoleucine and
L-valine, respectively. a-Oxoisovalerate, the precursor of L-valine, is the
starting point of the L-leucine branch. Addition of acetyl eoA yields
a-isopropylmalate which is isomerized to l3-isopropylmalate. Dehydro-
genation and amination leads to L-leucine.
B. The phosphoglycerate family of amino acids

The amino-acids L-serine, glycine, and L-cysteine are derived from phos-
phoglycerate (Fig. 3.9). Dehydrogenation of 3-phosphoglycerate and
subsequent action of a transaminase and a phosphatase yields L-serine.
r
NAD+ NADH+W
OOH COOH
HC-OH
I .. '>-. ./ I
c=o
I
CH20P03H2 CH 20P0 3H2
3-phosphoglycerate 3-phosphohydroxy-
pyruvate
,,-oxoglutarate 1--...../
3-phosphoserine L-serine
ace ty l-{:oA
COOH
I
CHcNH2
glycine
CoA
COOH
.. (
6 I
H 2N-fH
CH 2-O-CO-CH 3
CH 3-COOH
L-cysteine O-acetylserine
Figure 3.9. Biosynthesis of amino acids derived from 3-phosphoglycerate.
1, Phosphoglycerate dehydrogenase; 2, phosphoserine aminotransferase; 3, phos-
phoserine phosphatase; 4, serine hydroxymethyltransferase; 5, serine trans-
acetylase; 6, O-acetylserine sulfhydrylase.
This amino acid can be converted into L-cysteine as mentioned in connec-

tion with the assimilation of sulfate (Fig. 3.4), or into glycine by the
enzyme serine hydroxymethyltransferase. In the course of this reaction
carbon 3 of serine is transferred to tetrahydrofolic acid. Methylene
tetrahydrofolic acid can be reduced to N 5 -methyltetrahydrofolic acid,
which functions as donor of methyl groups in a number of biosynthetic
reactions, for instance in the synthesis of L-methionine from homocysteine
(see Fig. 3.6).
N I
7
OH (5)
NH
'TH-CHz-NH
(10)-0-
II ~ CO-NH-TH
COOH
I
HzN
A :)
N
NH-CHz CHz
I
lelrahydrofolic acid CH z
I
COOH
IK
serine
serine hydroxymelhyltransferase
glycine + H20
COOH
OH ~ N ~~-O- I
II CO-NH-CH
N, I I
N
7
I CH-CHz CHz
I: ) :
HzN/'::::, N NH'
tH z I
T Hz
COOH
N 5 ,N1o.methylene tetrahydrofolic acid
C. The a-oxoglutarate family of amino acids

a-Oxoglutarate is the precursor of L-glutamate and L-glutamine, com-
pounds that serve not only as building blocks in protein synthesis, but
also as important intermediates in nitrogen metabolism, as already out-
lined. In addition, L-glutamate is the precursor of L-proline and L-arginine
(Fig. 3.10).
L-Proline is synthesized along a pathway consisting only of a few
reactions. First glutamate semialdehyde is formed which spontaneously
cyciisizes to ~ I-pyrroline-5-carboxylate. This compound is then reduced to
L-proline. The synthesis of L-arginine is more complicated. Glutamate is
acetylated and converted to the semialdehyde afterwards. Blockage of the
NH 2 group by an acetyl group prevents spontaneous cyciization and the
aldehyde group is available for a transamination reaction. Removal of
the acetyl group yields L-ornithine and addition of a carbamoyl group
L-citrulline. Finally, the oxo group is replaced by an imino group to give
L-arginine.
a-oxoglutarate
+
__L-_g_lu_t,--am.,--at_e_..JI_1 L-glutamine
ATP+NADH + W acetyl-CoA
L-arginine
ADP + Pi + NAD+ 3 CoA
'~f,m,""
COOH COOH yOOH
I I
HzN-\H Ac-NH-CH HzN-yH
I
CHZ yH z yH z ~Hz yOOH
I
\H Z yH z CHz
I /
C=N-CH
I
CHO COOH CHz-NH CHz
I
COOH
glutamate-
-y-semialdehyde N-acetylglutamate L-argininosuccinate
(spontaneous) 4 ~ATP+NADH+W t-- AMP+ PP;
COOH
~ ADP + P; + NAD+
8
1 aspartate + ATP
I
Ac-NH-CH
I
CHZ
I
CHZ
I
CHO
tl. I-pyrroline- N-acetylglutamate- L-citrulline

5-carboxylate -y-semialdehyde
2 C
!_ NADPH+W
NADP'
~ glutamate
'1'- "~"",""" 7
F P.
ca~bamOYl-
phosphate
L-proline N-acetylornithine L-ornithine

Figure 3.10. Synthesis of L-proline and L-arginine from a-oxoglutarate via
L-glutamate. 1, Glutamate kinase + glutamate semialdehyde dehydrogenase (see
homologous reactions in Fig. 3.6); 2, A1-pyrroline-5-carboxylate reductase; 3,
amino acid acetyltransferase; 4, N-acetylglutamate kinase + N-acetylglutamate
semialdehyde dehydrogenase; 5, N-acetylornithine transaminase; 6, N-acetyl-
ornithine deacetylase; 7, ornithine transcarbamoylase; 8, argininosuccinate synthe-
tase; 9, argininosuccinate lyase.
erythrose- 3-deoxy-o-arabino- 5-dehydroquinate

4-phosphate heptulosonate-7-
phosphate
NH,
CH,-~H-COOH
6
COOH
o""c) OH
OH
L-phenylalanine L-tyrosine
S-dehydroshikima te
a-ogl=i
12
a-Ogl~
12
glu glu NADPH +W
4
NADH+W
phenylpyruvale
p-hydroxyphenyl
:a
pyruvate
r-------..,
: COOH I
NH
I
OH
,:
I I
IL ~ ..J:
shikimate
anthranilate
5 ~ ATP
9
~ADP
A"
Yo-C-COOH
I"' . . ._\....:i"""--__
-OVO-C-COOH
A" ,",;:".;, A -OYOH
OH OH OH
chorismate 3-enolpyruvylshikimale shikimate-S-phosphate

S-phosphate
Figure 3.11. Synthesis of aromatic amino acids. 1, 3-Deoxy-o-arabino-

heptulosonate (DAHP) synthase; 2, 5-dehydroquinate synthase; 3, 5-
dehydroquinate dehydratase; 4, shikimate dehydrogenase; 5, shikimate kinase; 6,
enolpyruvylshikimate-5-phosphate synthase; 7, chorismate synthetase; 8, anthrani-
late synthase; 9, chorismate mutase; 10, prephenate dehydrogenase; 11, prephen-
ate dehydratase; 12, transaminase B. PEP, Phosphoenolpyruvate; gin, glutamine;
glu, glutamate; a-ogl, a-oxoglutarate.
D. Aromatic amino acids

Biosynthesis of the aromatic amino acids is very complicated; however, it
has been of special interest to investigators since it involves the biosynth-
esis of the aromatic nucleus from aliphatic precursors. As illustrated in
Fig. 3.11 erythrose-4-phosphate and PEP are condensed to yield a CT
compound, which undergoes cyclization (enzyme-catalyzed) to 5-
dehydroquinate. The latter is converted via shikimate into chorismate, a
compound identified in 1964 as a common intermediate in the biosynthesis
of aromatic amino acids. At this point the pathway branches off into two,
one leading to L-tryptophan via anthranilate (Fig. 3.12), and a second
yielding prephenate, which is the precursor of both L-tyrosine and L-
phenylalanine. When examining the scheme of aromatic amino acid
synthesis it is possible to understand how Davis was able to isolate a
mutant of E. coli requiring L-phenylalanine, L-tyrosine, and L-tryptophan
HOOCY')
PRPP
/
PP j
_OCQH;J
OH OH
anthranilate
N-( 5'-phosphori bosyl)-an thranila te
O
COOH
I I' ,
OH OH OH
~
NH-CH=C-CH-CH-CHzO-
enol-I-{o-carboxyphenylarnino)-
I-deoxyri bulose-5-phosphate
L-tryptophan indole-3-glycerol phosphate

Figure 3.12. Synthesis of L-tryptophan from anthranilate. 1, Anthranilate phos-
phoribosy! transferase; 2, phosphoribosyl-anthranilate isomerase; 3, indole-3-
glycerol phosphate synthase; 4, tryptophan synthase. PRPP, 5-Phosphoribosy!-1-
pyrophosphate; ser, serine; GAP, glyceraldehyde-3-phosphate.
N 1-( 5'-phosphoribosyl)-
AMP
N'-15'-phosphoribosyl)-ATP
NAO'
NAOH + H'
phosphoribosyl-fonnimino-S amino-
r
N imidazole (arboxamide ribonucleotide
histidinal HC.-- ",
II CH
NAOIl + H' l------N~
CH, ,Lo,",
I'I
10
NAO'
I -
HC-NH,
I -
COOH H,C-NH N>-
-I I
HC.--N~ c=o CH
L-hislidine
II CH Ht -OH "'N N
l-Ni', Ht-OH I
1,
H I
CH,O-
nhose-
HC-NH,
I - phosphoribulosyl-formimino-5-amino-
CH,OH imidazole carhoxamide ribonucleotide
histidinol CONH,
f-.- P;
I[N> - - J g l U l a m i n e
I N glutamate
<) ~H,O NH, I 6
N - ribose- N
HC--N~ HC'-- \ HC'-- "-
II 'ell n
C I
CH II
C I
CH
T------~I ....1------:::-_""""'::::-_ _ 1
I--NH
H
7
I--NH
H~-oH
~H2
H<j-NH 2
7'" '"""""
O'-Qxoglutarak
,
glutamate c=o

H,O HC-OH
CH 20- tH,O- tH,O-
histidinol phosphate imidazoleaeetol phosphate imidazoleglycerol phosphate
Figure 3.13. Synthesis of L-histidine. 1, Phosphoribosyl: ATP pyrophosphorylase;

2, phosphoribosyl-ATP pyrophosphohydrolase; 3, phosphoribosyl-AMP cyclo-
hydrolase; 4, phosphoribosyl formimino-5-aminoimidazole carboxamide ribonu-
cleotide isomerase; 5, phosphoribulosyl-formimino-5-aminoimidazole carboxamide
ribonucleotide: glutamine aminotransferase; 6, cyclase, the name of the compound
released (circled) is 5-aminoimidazole-4-carboxamide ribonucleotide; 7, imidazole-
glycerol phosphate dehydratase; 8, histidinol phosphate transaminase; 9, histidinol
phosphatase; 10, 11, histidinol dehydrogenase (enzyme catalyses the two-step
dehydrogenation of histidinol to L-histidine). Reactions 7 and 9 are catalyzed by
the same enzyme.
How Pentose Phosphates and NADPH are Formed 55
(p-aminobenzoate in addition) and why this mutant could grow in the

presence of shikimate. This compound is an intermediate in the formation
of all three amino acids; thus the mutant was obviously defective in one of
the enzymes involved in the formation of shikimate from erythrose-4-
phosphate and PEP.
An interesting reaction is catalyzed by tryptophan synthase. This
enzyme is a tetramer consisting of two a-subunits and two ,B-subunits. The
conversion of indole-glycerol phosphate and serine to tryptophan and
glyceraldehyde-3-phosphate is pyridoxal phosphate-dependent. Isolated a
and ,B subunits catalyze partial reactions: a subunits the reversible reaction
of indole-glycerol phosphate to indole and glyceraldehyde-3-phosphate
and ,B subunits the condensation of indole and serine to tryptophan.
E. L-Histidine synthesis
A pathway completely independent of the reactions discussed thus far is
employed for L-histidine biosynthesis. It is composed of nine enzymes,
which manage the formation of histidine from 5-phosphoribosyl-l-
pyrophosphate (PRPP), ATP, and glutamine in 11 reactions (Fig. 3.13).
Six reactions are needed to make a compound, imidazole-glycerol phos-
phate, having the skeleton of L-histidine. Its Cs-carbon chain stems from
5-phosphoribosyl-l-pyrophosphate. One nitrogen and the carbon between
the two nitrogens originate from the pyrimidine ring of ATP, and the
second nitrogen from the amido-nitrogen of glutamine (transfered in
reaction 5). In the final reactions of L-histidine biosynthesis the a-amino
group is introduced and the hydroxymethyl group is oxidized to the
carboxyl group. Reactions 7 and 9 as well as reactions 10 and 11 are
catalyzed by one enzyme, respectively.
With the path of L-histidine formation the presentation of the biosynth-
esis of amino acids is concluded. It should be mentioned that in addition to
the 20 amino acids involved in protein synthesis, two amino acids were
discussed which are constituents of bacterial cell walls: ornithine and a,
E-diaminopimelate.
V. How Pentose Phosphates and NADPH

are Formed
Many hydrogenation reactions in biosynthetic pathways require NADPH
as reducing agent and not NADH; the latter, however, is formed in most of
the redox reactions involved in the oxidation of glucose to CO 2 , Only the
isocitrate dehydrogenase reaction yields NADPH. As in animals and a
number of other microorganisms, E. coli regenerates additional NADPH
by the oxidative pentose phosphate cycle. This same pathway also serves as
the source of pentose phosphates for the cell, one of which (5-phospho-
ribosyl-I-pyrophosphate) is required for the biosynthesis of nucleotides.
In the oxidative pentose phosphate cycle three series of reactions can be

distinguished:
1. Oxidation of glucose-6-phosphate to ribulose-5-phosphate and for-
mation of NADPH and CO 2 . This is accomplished by the enzymes
glucose-6-phosphate dehydrogenase, a lactonase, and 6-phospho-
gluconate dehydrogenase.
2. Enzyme reactions that allow the formation of ribose-5-phosphate and
xylulose-5-phosphate from ribulose-5-phosphate. The enzymes in-
volved are ribose-5-phosphate isomerase and ribulose-5-phosphate-3-
epimerase.
3. Transaldolase and transketolase reactions, which catalyze the formation
of hexose-6-phosphates from pentose-5-phosphates.
Figure 3.14 summarizes all of the reactions of the oxidative pentose
phosphate cycle. First glucose-6-phosphate is oxidized to 6-phospho-
gluconate by a NADP+-specific glucose-6-phosphate dehydrogenase.
Then the action of 6-phosphogluconate dehydrogenase leads to the reduc-
tion of a second NADP+ and to the formation of ribulose-5-phosphate
and CO 2 with 3-keto-6-phosphogluconate as an intermediate in the reac-
tion. Ribulose-5-phosphate can be isomerized to yield ribose-5-phosphate
and epimerized to yield xylulose-5-phosphate. If E. coli were to need
NADPH and pentose phosphates in a ratio of 2: 1 additional reactions
would not be required. The sum would, therefore, be:
3 glucose-6-phosphate + 6NADP+ - - - 4
3 pentose-5-phosphate + 3C0 2 + 6NADPH + 6H+
Now, consider a situation where there is a demand for more than
2NADPH for each pentose phosphate made; the excess pentose phos-
phate would have to be removed. This is accomplished by the enzymes
transketolase and transaldolase, which convert pentose phosphates back
into hexose phosphates. Transketolase contains tightly bound thiamine
pyrophosphate as coenzyme.
Transketolase catalyzes the transfer of a glycolaldehyde group from
xylulose-5-phosphate to ribose-5-phosphate or erythrose-4-phosphate:
CHzOH
I
c=o
I
CHzOH CHO HOCH
I I I
c=o HCOH HCOH
I I
+ HCOH HCOH + CHO
I I I
HCOH HCOH HCOH
I I I
CHZOP03Hz CHZOP03Hz CH ZOP0 3 H z
xylulose-5-P ribose-5P sedoheptulose-?-P glyceraldehyde-3-P
transketolase reaction
How Pentose Phosphates and NADPH are Formed 57
glucose-6-P 6-P-gluconolactone 6-P-gluconate
HCO
I
HC-OH
I
HC-OH 3NADPH+ 3W
I
HC-OH
I
H2 CO-
ribose-S-P 3
xylulose-S-P
xylulose-S-P ~ glyceraldehyde-3-P ~ fructose-6-P
ob,"-5-P ~. "dOO",,"'o.-7-P ~,~ fru",,,-6-P
xylulose-S-P L ~;~eraldehYde-3-P
Figure 3.14. Oxidation of glucose-6-phosphate to ribulose-5-phosphate and con-
version of pentose phosphate into hexose phosphate. 1, Glucose-6-phosphate
dehydrogenase; 2, lactonase; 3, 6-phosphogluconate dehydrogenase; 4, ribose-5-
phosphate isomerase; 5, ribulose-5-phosphate-3-epimerase. The formed ribulose-
5-phosphate is in equilibrium with ribose-5-phosphate and xylulose-5-phosphate
if enzymes 4 and 5 are present. TK, transketolase; TA, transaldolase.
Transaldolase catalyzes the transfer of a dihydroxyacetone group from

sedoheptulose-7-phosphate to glyceraldehyde-3-phosphate:
CH 20H
I
C=O
I
CHO HOCH
I I
CHO HCOH HCOH
+ I I + I
HCOH HCOH HCOH
I I I
CH20P03H2 CH 20P0 3H2 CH 20P03H2
sedoheptulose- 7-P glyceraldehyde-3-P erythrosc-4-P fruclosc-6P
transaldolase reaction
Both enzymes together carry out the conversion of pentose-5-

phosphates into fructose-6-phosphate and glyceraldehyde-3-phosphate
according to the equation:
2 xylulose-5-P + ribose-5-P ( ) 2 fructose-6-P + glyceraldehyde-3-P
Consequently, any pentose phosphate in excess can be conveniently
channeled into the Embden-Meyerhof-Parnas pathway. Since the reactions
involved are reversible, the synthesis of pentose phosphates from hexose
phosphates is also possible under conditions under which the formation of
NADPH by the oxidative portion of the cycle is not desired. Moreover, the
reactions of the pentose phosphate pathway provide the cells with
erythrose-4-phosphate, which, for instance, is required for aromatic amino
acid biosynthesis.
E. coli channels about 28% of the glucose that is degraded into the oxi-
dative pentose phosphate pathway and 72% into the Embden-Meyer-
hof-Parnas pathway. This can be calculated from radiorespirometric
experiments similar to that shown in Fig. 3.15, which demonstrates the rate
of release of radioactive carbon dioxide from cultures of E. coli growing
with radioactive glucose labeled in various positions. As can be seen,
carbon atoms 3 and 4 are the preferential sources of CO 2; in the degra-
dation of glucose by the Embden-Meyerhof-Parnas pathway C-3 and
C-4 of glucose are converted to carboxyl groups of pyruvate and released
as CO 2 during acetyl-CoA formation. The Embden-Meyerhof-Parnas
pathway is thus primarily involved in glucose breakdown. Radioactivity
from [l)4C]glucose appears also early in CO 2, This indicates that part of
the glucose is oxidized to ribulose-5-phosphate via 6-phosphogluconate.
The second peak of radioactive CO 2 results from the oxidation of
radioactive acetyl-CoA to CO 2 via the tricarboxylic acid cycle. Its propor-
tion on the totally released 14C02 is small with [3,4- 14 C]glucose but
predominant with [2_ 14 C]glucose.
Ribonucleotides and Deoxyribonucleotides 59
Figure 3.15. Formation of radioactive CO 2 by E. coli cells from glucose labeled

with 14C in various positions. The amount of radioactive CO 2 formed per hour is
plotted against the time. [CO H. Wang, I. Stern, C. M. Gilmour, S. Klungsoyr,
D. J. Reed, J. J. Bialy, B. E. Christensen, and V. H. Cheldelin: J. Bacterial. 76,
207-216 (1958).]
E. coli and most other microorganisms have another possibility at their

disposal to produce NADPH, via a transhydrogenase:
NADH + NADP+ lran,hydrogen,,"> NAD+ + NADPH
The transhydrogenase is membrane-bound and transhydrogenation is

protonmotive force driven; it thus can only proceed at an energized
membrane.
VI. Ribonucleotides and Deoxyribonucleotides

Ribonucleotides consist of a purine or pyrimidine base, ribose, and a
phosphate, pyrophosphate, or triphosphate group:
purine

pyrimidine
ribose -
--
Table 3.4 summarizes the structures of important purines and pyrimi-
dines and the corresponding nucleotides.
A purine or pyrimidine base attached to ribose is called a ribonucleoside
(trivial names: adenosine, guanosine, cytidine, thymidine, and uridine).
Esterification of the ribonucleosides with phosphoric acid yields the
corresponding ribonucleoside monophosphates. Their linkage to a second
and a third phosphate residue results in the ribonucleoside di- and
triphosphates. All these phosphate esters are referred to as nucleotides.
Table 3.4. Important pyrimidines, purines, and corresponding nucleotides
base nucleotide
0 0
HNJ
o=lN o=lN
HNJ
H
uracil -o-~
OH OH uridine monophospha.e (UMP)
NH 2 NH 2
o.t~ N
H
o . tNJ
cytosine -o-~
OH OH cytidine monophospha.e (eMP)
HN
O=lN
yCH'
I
0
HN
O=lN
:yCH'
0
I
H
thymine -o-~
OH OH thymidine monophospha.e (TMP)
0=)
NH 2 NH 2
Cr)N H
adenine -o-~
OH OH adenosine monophosphate (AMP)
0 0
HN~~C>
HN~N
2 N
H
HN:C>
H2N~N N
guanine -o-~
OH OH guanosine monophosphate (GMP)
A. Pyrimidine nucleotides
The precursors of the pyrimidine nucleotides are carbamoyl phosphate and
aspartate (Fig. 3.16). The enzyme aspartate transcarbamoylase condenses
these compounds to yield carbamoyl aspartate, which undergoes cycliza-
tion to give 4,5-dihydroorotate. Dehydrogenation then leads to orotate,
the first intermediate containing the pyrimidine ring. Before orotate is
converted to one of the important physiological pyrimidine bases it is
linked to ribose-5-phosphate to form the corresponding ribonucleotide.
Ribose-5-phosphate itself cannot function as substrate in this condensation
reaction; it must be activated by conversion into 5-phosphoribosyl-l-
pyrophosphate (PRPP). The reaction of 5-phosphoribosyl-l-pyrophosphate
carbamoyl- aspartate carbamoyl aspartate

phosphate
ribose-Sop + ATP
midine monophosphale 4
(UMP) AMP
4. S-dihydroorotate
~ NAD+
N~
OH -O~0'JI
)-(0- 0 - 0
3 t- NADH+W
O""lN~COOH
OH
"~I
OH OH
0-0~o~ .__---."...""'~'-R_PP------
)-( (
HOAN
orotatt;:
COOH
OH OH PP i
orotidine monophosphate
Figure 3.16. Biosynthesis of UMP from aspartate, carbamoyl phosphate, and

5-phosphoribosyl-l-pyrophosphate (PRPP). 1, Aspartate transcarbamoylase; 2,
dihydroorotase; 3, dihydroorotate dehydrogenase; 4, PRPP synthetase; 5, orotate
phosphoribosyl transferase; 6, orotidine-5-phosphate decarboxylase. Note that the
reaction of PRPP with orotate is connected with a change of the configuration from
a to f3 at C\ of ribose.
AT\' AT\'
--~'-"'"--.,---_..UTI'
'"
UMP - - - ' - - " - - - . . . . . , - - - UDI'
'\
ADP ADP
NA
O~N)
ATP 3 ---O~C2 0
\.
+ NH) --"""--,..----.~
ADP+P, OH OH
uridine triphosphate (UTP) cytidine triphosphate (CTP)
Figure 3.17. Biosynthesis of CTP. 1, Nucleoside monophosphate kinase; 2,

nucleoside diphosphate kinase; 3, CTP synthetase.
with orotate then yields orotidine monophosphate, which is subsequently

decarboxylated to uridine monophosphate (UMP).
UMP can be converted into UTP by kinase reactions; a subsequent
amination reaction yields cytidine triphosphate (CTP) (Fig. 3.17). Amina-
tion proceeds via a phosphorylated intermediate and is, therefore, associ-
ated with ATP hydrolysis.
B. Purine nucleotides
The synthesis of the purine nucleotides is more complicated. Starting with
PRPP first an imidazole nucleotide is formed. It can be seen from Fig. 3.18
that the imidazole ring is built up step by step starting with the amido group
of glutamine. Subsequent addition of glycine, the formyl group of
methenyl-tetrahydrofolic acid, and another amido group from glutamine
and ATP-mediated cyclization yields 5'-phosphoribosyl-5-amino-
imidazole.
The three pyrimidine ring atoms still required for the formation of the
purine ring from 5-aminoimidazole are furnished by bicarbonate, aspar-
tate, and formyl H 4 -folate. Ring closure then yields inosinic acid (Fig.
3.19). From IMP a number of additional reactions lead either to AMP or
GMP. The enzyme adenylate kinase (myokinase in animal tissues) then
converts AMP into ADP:
adenylatc kinase
AMP + ATP ( ) 2ADP
Finally, ATP can be formed by electron transport or substrate-level
phosphorylation.
Q
-OCQ0 H2 glutamine + H20 -OC H2 0 NH 2
1--\...---""'--.....::---...
0--0 ""\
OH OH glutamate + PP j OH OH
5-phosphoribosyl-l- 5-phosphoribosylamine
pyrophosphate
~Yl~~~
--1
2
ADP+Pi
'=H
HC-N
~ /NH 2
/
C-N/
I yH 2
NH 2 ribose-
~C,
5'-phosphoribosyl- o NH
5-aminoimidazole
I
ribose-0
5'-phosphoribosyl-
glycineamide
ATP
ADP + Pi 3
H
......-N"'-... ......-N,
H
iH2 CHO fH 2 CHO

4
<.?C,
NH NH
I
7
glu + ADP + Pi
o""C" NH
I
ribose-0
ribose- 0
5'-phosphoribosyl- 5'-phosphoribosyl-
N-formylglycine- N-formylglycine-
amidine amide
Figure 3.18. First reactions of purine nucleotide synthesis. 1, PRPP amido-
transferase; 2, phosphoribosylglycineamide synthetase; 3, phosphoribosyl-
glycineamide formyl transferase; 4, phosphoribosyl-formylglycineamidine syn-
thetase; 5, phosphoribosyl-aminoimidazole synthetase.
GMP is phosphorylated to GDP and GTP by the nucleoside mono-

phosphate and nucleoside diphosphate kinases mentioned in Fig. 3.17.
The role of purine and pyrimidine nucleotides in cell metabolism is
diverse. ATP, GTP, UTP, and CTP are the substrates for RNA synthesis.
In addition, numerous enzymatic reactions depend on these ribonucle-
otides as sources of either high-energy phosphate or groups with a high
COOH
I
\ ( HOOC N\ aspatte
H2r ~
HC-NH-C N
HN:> ---"----
I H NX N
1"7
ATP
""".
ADP + P,
~OOH XN>
2 I 2
1 H2 N I
ribose- ribose-(V ribose-
5'-phosphoribosyl- 5'-phosphoribosyl-5- 5'-phosphoribosyl-4-

5-aminoimidazole aminoimidazole-4- (N-succinocarboxamide)-
carboxylic acid 5-aminoimidazole
5'-phosphoribosyl-4- 5' -phosphoribosyl-4-

carboxamide-5-form- carboxamide-5-amino-
aminoimidazole imidazole
inosinic acid (IMP)
aspartate + GTP
1
HOOC- H-CH 2 -COOH 6~
NH Ar.np
(X)N N
GO'"
I
ribose-
adenylosuccinate xanthylic acid (XMP)
7 ~ r,m.,," ATP + NH l + H 2 0 J
~}- AMP + PP i
~ ~
Figure 3.19. Synthesis of AMP and GMP from 5'-phosphoribosyl-5-amino-
imidazole_ 1, Phosphoribosylaminoimidazole carboxylase; 2, phosphoribosylamino-
imidazole succinocarboxamide synthetase; 3, adenylosuccinate lyase (enzyme is
identical with the one catalyzing step 7); 4, phosphoribosylaminoimidazole-
carboxamide furmyltransferase; 5, IMP cyclohydrolase; 6, adenylosuccinate syn-
thetase; 7, adenylosuccinate lyase; 8, IMP dehydrogenase; 9, GMP synthetase.
Biosynthesis of Lipids 65
potential of group transfer. ATP especially is used as source of high-

energy phosphate, for instance, in the numerous kinase reactions
(sugar + ATP -- sugar phosphate + ADP). GTP is the energy source in
protein synthesis. The high potential of group transfer of ATP is taken
advantage of in amino acid or fatty acid activation (amino
acid + ATP -- aminoacyl-AMP + PPj) and in sugar activation for poly-
saccharide synthesis (glucose + ATP -- ADP-glucose + Pi). Corres-
ponding functions are carried out by UTP in activation of N-acetylmuramic
acid and by CTP in phospholipid synthesis. Ribonucleotides are also the
precursors of the deoxyribonucleotides and, therefore, of DNA.
C. Synthesis of deoxyribonucleotides
Reduction of ribonucleotides to deoxyribonucleotides takes place in
E. coli at the diphosphate level (Fig. 3.20a). The reducing agent in this
reaction is thioredoxin, a flavoprotein; its reduced form is regenerated with
NADPH.
Four nucleoside diphosphates are reduced according to this scheme:
UDP CDP ADP GDP
dUDP dCDP dADP dGDP

Deoxyuridine phosphate is not a major constituent of DNA. However,
dUDP is the precursor of thymine-containing deoxynucleotides; it is
hydrolyzed to dUMP and then methylated (Fig. 3.20b).
In the methylation of dUMP to dTMP, methylene-H 4 -folate plays a
unique role; it is not only the donor of the C\-unit but serves also as
reducing agent. The formed dihydrofolate (H 2 F) is converted back to
tetrahydrofolate by a reductase with NADPH as electron donor. There are
kinases present in E. coli that catalyze the formation of the corresponding
deoxynucleoside triphosphates with ATP as donor of high-energy phos-
phate.
VII. Biosynthesis of Lipids

Bacteria do not accumulate lipids as reserve material but they contain
considerable amounts of lipids as constituents of their membrane systems,
especially phospholipids and glycolipids. The general structure of these
lipids is shown in Fig. 3.21.
The following is an outline of the synthesis of the major constituents of
lipids: fatty acids, glycerol, and choline.
--o~o~ase --o~o~ase
'1-(. ----------.. '1-(. +H 20
OH OH OH
thioredoxine thioredoxine
(reduced) (oxidized)
NADp _. \~)- NADPH + W
dUMP dTMP
Figure 3.20. Synthesis of deoxyribonucleotides. a: The reaction of the ribonu-
cleoside diphosphate reductase. b: The synthesis of deoxythymidylic acid (dTMP).
1, dUDP phosphatase; 2, thymidylate synthetase. H 2 F is dihydrofolate, which is
converted to tetrahydrofolate by a reductase with NADPH as electron donor.
CH 2 -O-CO-R
I
CH-O-CO-R lipid (neutral fats)
I
CH 2 -O-CO-R
CfH 2 -O-CO-R
glycolipid
CH-O-CO-R
I CH 2 0H
CH2-0~:~
HOH
Figure 3.21. General structure of lipids, phospholipids, and glycolipids. R. Carbon
chains of fatty acids esterified with glycerol. Choline (shown here) is one of the
alcohols that can be present in phospholipids; others are ethanolamine, serine, and
inositol. In the glycolipid shown D-mannose is the sugar moiety. Other glycolipids
contain galactose, glucose, or oligosaccharides.
A. Fatty acids
Most of the fatty acids occurring in lipids contain 16 or 18 carbon atoms;
they are saturated or have one (seldom more than one) double bond. The
precursor of fatty acids is acetyl-CoA. However, chain elongation is not
achieved by condensation of two acetyl-CoA molecules followed by further
condensation of the C 4 -compound with acetyl-CoA, a reaction sequence
accomplished by clostridia when forming butyrate and caproate. Two
variations are important:
1. CoA-derivatives are not substrates of the enzymes involved in fatty acid
synthesis. Instead E. coli employs an acyl carrier protein (ACP) of a
molecular weight of 10,000; its prosthetic group is 4'-phospho-
pantetheine, and it thus resembles CoA (Fig. 3.22).
The first reaction in fatty acid synthesis is the formation of acetyl-ACP:
acetyl transacetylase
acetyl-CoA + ACP ( ) acetyl-ACP + CoA
2. Acetyl-ACP functions as primer in fatty acid synthesis, and the Cz-units
are added to this primer in the form of malonyl-ACP. The latter is
synthesized from acetyl-CoA in two steps:
acetyl-CoA carboxylase
CH3 -CO-CoA + A TP + COZ )
CHz-CO-CoA + ADP + Pi
I
COOH
malonyl transacylase
malonyl-CoA + ACP ' malonyl-ACP + CoA
Acetyl-CoA carboxylase is a biotin-containing enzyme.
~H-CH1~CHl SH
c=o
4' -phospho- I
CHl
pantetheine I
CH,
I -
NH
I pantothenate
c=o
I
CHOH
I
CH 3 - C- CH 3
I
CH 1
I
o
I
O=P-OH
/1
r--------- j "o~ /NH~
: adenine~ribose-' /
l..
CHl -CH
/ I peptide chain
C~
11
o
CoA ACP
Figure 3.22. Structure of the prosthetic group of ACP and CoA. In ACP the
4'phosphophantetheine is linked to the peptide chain via a serine residue.
3-ketoacyl-ACP synthase
acetyl-ACP + malonyl-ACP ::;:=:::!: acetoacetyl-ACP + CO 2 + ACP
3-ketoacyl-ACP reductase
acetoacetyl-ACP + NADPH + W::;:=:::!: l3-hydroxybutyryl-ACP + NADp
l3-hydroxyacyl-ACP dehydratase
l3-hydroxybutyryl-ACP ~ crotonyl-ACP + H2 0
enoyl-ACP reductase
crotonyl-ACP + NADPH + W ~ butyryl-ACP + NADp
Figure 3.23. Reactions involved in the formation of butyryl-ACP from acetyl-ACP

and malonyl-ACP.
Four enzymes are required to elongate an acyl-ACP by two carbon

atoms. First acetyl-ACP reacts with malonyl-ACP to form acetoacetyl-
ACP, which is then reduced to o(-)-j3-hydroxybutyryl-ACP. Dehydration
yields crotonyl-ACP and subsequent reduction results in butyryl-ACP
(Fig. 3.23).
The names of the enzymes shown in Fig. 3.23 suggest that they may be
relatively unspecific with respect to the chain length of their substrates.
Indeed, they also catalyze the conversion of butyryl-ACP and malonyl-
ACP to caproyl-ACP as well as all subsequent reactions until palmi-
tyl-ACP is formed. The latter is the predominant saturated acyl-ACP
produced in E. coli. This is because the 3-keto-acyl-ACP synthase is
active with substrates having a maximum chain length of C 14 (Table 3.5).
Thus, with regard to the saturated acyl derivatives, palmityl-ACP (hexa-
decanoyl-ACP) is the final product.
Some unsaturated fatty acids are also important constituents of the
phospholipids. Those predominating in E. coli are palmitoleate (cis-9-
hexadecenoate) and cis-vaccenate (cis-ll-octadecenoate). Their synthesis
involves the enzyme systems discussed above and it is evident from Table
3.5 that the synthase is still active with the unsaturated C 16 -ACP so that
cis-vaccenyl-ACP can be formed. The branch point of saturated and
unsaturated fatty acid synthesis in E. coli is at j3-hydroxydecanoyl-ACP.
CH,-(CH 2 h-CH 2-CHOH-CH 2 -CO-ACP
{3hydroxydecanoylACP
H H H
I I I
CH,-(CH 2 h-C=C-CH 2-CO-ACP CH,-(CH 2 h-CH 2-C=C-CO-ACP
1
unsaturated acids
1
saturated acids
~
A special dehydratase removes water to yield a compound with a cis-
Table 3.5. Activity of 3-ketoacyl-ACP synthase

with various acyl-ACP compounds Q
velocity of reaction
(V max)
/-L mol Product)
substrate ( mm'mg
acetyl-ACP 2.8
decanoyl-A CP 2.8
dodecanoyl-ACP 0.97
tetradecanoyl-ACP 0.31
hexadecanoyl-ACP zero
cis-5-dodecenoyl-ACP 1.7
cis-9-hexadecenoyl-ACP 0.37
cis-11-octadecenoyl-ACP zero
aD. J. Prescott and P. R. Vagelos, Adv. Enzymol. 36,

269-311 (1972).
double bond between carbon atoms 3 and 4; the resulting compound does
not function as a substrate for enoyl-ACP reductase but is subject to
further elongation reactions, which yield C l6 and CIS monounsaturated
acyl-ACP.
It should be mentioned that all the enzymes involved in fatty acid
synthesis in E. coli are soluble and readily separable from one another in
vitro. In higher organisms all the reactions leading from acetyl-CoA and
malonyl-CoA to long-chain fatty acids are catalyzed by a multienzyme
complex: fatty acid synthase.
B. Phosphatidic acid
The principal substrates for the formation of phosphatidic acids are
glycerol-3-phosphate and acyl-ACP The former is readily available from
dihydroxyacetone phosphate-an intermediate of the Embden-Meyer-
hof-Parnas pathway:
glyeerol3-phosphate
HOCH -CO-CH 2 0e) + NADPH + H+ dehydrogenase I

2
HOCH 2-CHOH-CH 20e) + NADP+
Phosphatidic acids are then synthesized as follows:
CH 2 0H CH 20-CO-R
I glycerolphosphate I
CHOH + 2R-CO-ACP .eyltransferase I CHO-CO- R
I \ I
CH 2 0e) 2ACP CH 2 0e)
phosphatidic acid
After hydrolytic removal of phosphate from the phosphatidic acid,

neutral fats may be formed by reaction with a third acyl-ACP. Most of the
phosphatidic acid, however, is used for the synthesis of phospholipids.
c. Phospholipids
The phosphate group of phosphatidic acid is prepared for esterification
with an alcohol by the reaction with CTP:
iHZO-CO-R
C,HzO-CO-R phosphatida,e
cytidylyltransferase
CHO-CO-R + CTP , CHO-CO-R + PP j
I ~
CHzO-P-OH
I ~ ~
CHzO-P-O-P-O-cytidine
I I I
OH OH OH
phosphatidic acid CDP-diacylglyce,ol
Specific enzymes then catalyze the displacement of CMP by alcohols,

like serine, inositol, and glycerol.
CDPdiacylglycerol:serine
COP-diacylglycerol + serine. O.phosphatidyl,ransferas\
phosphatidylserine + CMP
Decarboxylation of phosphatidylserine yields phosphatidylethanol-
amine, which can be methylated with S-adenosylmethionine to yield
phosphatidylcholine.
iHZO-CO-R
COO-
I
CHO-CO-R HC-NH z
I
+3 CH z
I
CHz
I
H 3C-S-adenosine
+
phosphatidylethanolamine Sadenosylmethionine
1
phosphatidylethanolamine
methyltransferase
iHZO-CO-R
+ 3 S-adenosyl homocysteine + 2H+

C1HO-COO_R
I +/CH 3
CHz-O-P-O-CHz-CHz-N, CH 3
II CH 3
o
phosphatidylcholine
Formation of Carbohydrates 71
Analogous pathways are employed for the synthesis of other phos-

pholipids.
VIII. Formation of Carbohydrates

When glucose serves as growth substrate for E. coli a number of hexose
and pentose phosphates are intermediates in the breakdown of this
substrate by the Embden-Meyerhof-Parnas pathway and the oxidative
pentose phosphate pathway: glucose-6-phosphate, fructose-6-phosphate,
fructose-I,6-bisphosphate, ribulose-5-phosphate, xylulose-5-phosphate,
and ribose-5-phosphate. The following is an outline of the synthesis of
some other sugars and sugar derivatives. Mannose-6-phosphate can be
made from fructose-6-phosphate by mannose-6-phosphate isomerase:
mannose6~phosphate isomerase
fructose-6-phosphate , ) mannose-6-phosphate
For the synthesis of galactose esters, glucose is first linked to UDP by the
following reactions:
phosphoglucomutase
glucose-6-phosphate , ) glucose-I-phosphate
UDPglucose
pyrophosphorylase
glucose-I-phosphate + UTP , ) UDP-glucose + PP j
A change of the configuration at carbon atom 4 of the glucose moiety is
then accomplished by a specific epimerase, which has an absolute require-
ment for NAD+. A 4-ketoglucose derivative is an intermediate in this
reaction:
CH 20H
~
epimerase
H0CHPHO
(NAD+)
OH OH
HO o---uridine o---uridine
OH OH
UDPglucose UDPgalactose
UDP-galactose is a precursor of E. coli wall lipopolysaccharide. Other

precursors of this important structural element are hexose amines; they
originate from fructose-6-phosphate, which is converted into glucos-
amine-6-phosphate with glutamine as NH z donor and further into N-
acetylglucosamine-6-phosphate (Fig. 3.24). The latter can be used to
synthesize UDP-N-acetylglucosamine which is the activated form of
N-acetylglucosamine and which is used in lipopolysaccharide and peptido-
glycan synthesis. Another precursor of peptidoglycan is UDP-N-
acetylmuramate. It is the 3-lactylether derivative and is formed by the
addition of phosphoenolpyruvate at carbon atom 3 of the glucose skeleton
of UDP-N-acetylglucosamine and subsequent reduction (Fig. 3.24).
CH 2 0H CHO
I I
C=O glutamine glutamate H<r- NH2
I
HO-CH
I \.. ) HO-CH
I
HC-OH HC-OH
I
Ht-OH HC-OH
I
CH 20- J
CH 2 0-
fructose-6-P glucosamine-6-P
~
t--
2 acetyl-CoA
COA
~
CH20-~
OH
OH OH
NH-CO-CH)
CH 20H N-acetylglucosamine-6-P
4=d--u
3~
H C-CH
3 I
NH-CO-CH) ~:'OH ~
COOH
OH O-
UDP-N-acetylmuramate NH-CO-CH)
,~:::~~'H'
N-acetylglucosamine- J-P
~H'OH ~ EP
. . .I - -_ _\........._..c.)c...___
OH O---U
NH-CO-CH)
H2 C=C
I
COOH UDP-N-acetylglucosamine
UDP-N-acetylg]ucosamine-
3-enolpyruvylether
Figure 3.24. Formation of UDP-N-acetylmuramate from fructose-6-phosphate. 1,
Glutamine: fructose-6-phosphate aminotransferase; 2, glucosamine phosphate
transacetylase; 3, N-acetylglucosaminephosphomutase; 4, UDP-N-acetylglucos-
amine pyrophosphorylase; 5, UDP-N-acetylglucosamine-3-enolpyruvylether synth-
ase; 6, UDP-N-acetylenolpyruvylglucosamine reductase.
Synthesis of Polymers 73
Figure 3.25. UDP-N-acetylmuramylpentapeptide, a building block of the peptido-

glycan of E. coli.
The lactyl residue of UDP-N-acetylmuramic acid is very important; the
building block of the peptidoglycan (murein) of the cell wall is a muramyl..
pentapeptide, and the peptide chain is attached to the carboxyl group of
this lactyl residue (Fig. 3.25). The E. coli pentapeptide is synthesized from
UDP-N-acetylmuramic acid by subsequent reaction with L-alanine, 0-
glutamate, meso-diaminopimelate, and o-alanyl-o-alanine. The addition
of each amino acid to the molecule requires ATP (Fig. 3.26). The
formation of o-alanyl-o-alanine from o-alanine is specifically inhibited by
the antibiotic cycloserine.
IX. Synthesis of Polymers

It has been mentioned that about 97% of the cellular material are
macromolecules. Three groups can be distinguished as lipids, periodic
macromolecules such as peptidoglycan and polysaccharides, and informa-
tional macromolecules such as nucleic acids and proteins.
A. Lipids
Lipids are not true macromolecules, as the monomers are not linked to one
another by covalent bonds. However, in an aqueous environment phos-
pholipid molecules such as phosphatidylcholine associate in such a way that
UDP-N-acetylmuramate
ATP+L-Ala
ADP+Pi~
UDP-N-acetylmuramyl-L- Ala
ATP+D-GIU
ADP+Pi~
UDP-N-acetylmuramyl-L- Ala-D-Glu
A T P + meso-A 2 pm
ADP+Pi~
UDP-N-acetylmuramyl-L- Ala-D- Glu-meso-A 2 pm
A T P + D-Ala-D-Ala
ADP+Pi~
UDP-N-acetylmuramyl-L- Ala -D- Glu -meso-A 2 pm-D-Ala -D- Ala
Figure 3.26. Synthesis of UDP-N-acetylmuramylpentapeptide. The four reactions

are catalyzed by specific synthetases. D-alanyl-D-alanine is synthesized accordingly
from two molecules of o-alanine.
a double-layered structure is formed, as shown in Fig. 3.27. This structure

is stabilized by interaction of the hydrophobic "tails" and by hydrogen
bonding between the hydrophilic "heads" of the phospholipid. Together
with proteins, lipids associate to membranes; their general structure is
shown in Fig. 3.27. Membranes contain up to 60% of proteins; some are
embedded in the lipid layers and have specific functions in the various
transport processes. A membrane of this general structure surrounds the
cytoplasm of the bacterial cell and is called cell or cytoplasmic membrane.
Figure 3.27. Model of the cytoplasmic membrane showing the double-layered

arrangement of phospholipid molecules and various embedded proteins.
B. Periodic macromolecules
Unlike the informational macromolecules, the periodic molecules are not
synthesized along a template. Rather, they are formed by specific enzymes
from one (glycogen), two (peptidoglycan), or several (lipopolysaccharides)
types of building blocks.
1. Glycogen. When E. coli grows on glucose it stores some of the glucose

taken up into the cell as glycogen. Synthesis of glycogen involves two steps:
activation of glucose and transfer of the glucose moiety to a primer
molecule. The activated glucose derivative is ADP-glucose, and it is formed
from glucose-I-phosphate by ADP-glucose pyrophosphorylase:
glucose-I-P + ATP ---+ ADP-glucose + PP j
There is then a transfer of the glucose moiety to the nonreducing end of

an oligosaccharide (primer), which must consist of more than four glucose
residues (Fig. 3.28).
E. coli glycogen is a highly branched polysaccharide and contains, in
addition to the a-(1 ~ 4) linkages between glucose molecules, a large
number of a-(1 ~ 6) linkages. The latter are formed by transglucosyla-
tion; an oligosaccharide is transferred from the nonreducing end (C4
position) to the OH-group at C6 of a glucose residue in the chain (Fig.
3.29).
It should be mentioned that in animals UDP-glucose and not ADP-
glucose is the substrate for glycogen synthesis.
O- O-
CH 0H CH20HO
2 0
OH + OH
HO ? HO O-(glucose)n
OH OH
I
-adenine
'-----0 O-(glucose)"
OH OH
+ ADP
Figure 3.28. Elongation of a polysaccharide chain of glycogen. The enzyme is called
glycogen synthetase.
V
1/ OM
/
1/
~ ~
PERI LP
r
PG
~ eM
Figure 3.30. Surface layers of E. coli in thin section. CM, Cell membrane; PG,
peptidoglycan; PERI, periplasmic space; LP, Braun's lipoprotein; OM, outer
membrane.
2. Cell wall. Figure 3.30 presents a schematic view of the macromolecular
structures surrounding the cytoplasm of E. coli. It can be seen that the cell
membrane is covered by a rather thin peptidoglycan layer, as is characteris-
tic of Gram-negative bacteria. The final layer is the outer membrane. It
consists of lipopolysaccharides, lipids, and proteins. Embedded are
Braun's lipoprotein molecules that anchor the outer membrane to the
peptidoglycan layer.
Peptidoglycan (murein) is synthesized from two building blocks: UDP-
N-acetylmuramyl-pentapeptide and UDP-N-acetylglucosamine. The for-
mation of these compounds in the cytoplasm was outlined. As shown in
Fig. 3.31, N-acetylmuramyl-pentapeptide is first transferred to a membrane
lipid carrier-undecaprenyl phosphate (bactoprenol):
H 3C", TH3 TH3

/C=CH-CH2-(CH2-C=CH-CH2)~-CH2-C=CH-CH20
HJC bactoprenol or undecaprenyl phosphate
It is then linked to N-acetylglucosamine. This disaccharide pentapeptide

is transported by the lipid carrier to the extracellular site of the cell
membrane and used as a building block in peptidoglycan chain elongation.
Presumably E. coli peptidoglycan consists of a monomolecular layer of
chains with the polysaccharide backbone facing the outer membrane and
the peptide residues facing the cell membrane. Cross-linkage between
adjacent chains is brought about by the closure of bridges between some of
the peptide units. This is accomplished by a transpeptidation reaction. As
shown in Fig. 3.32 the peptide bond between the two o-alanine residues is
replaced by a peptide bond between o-alanine and a meso-diaminopime-
late residue of an adjacent chain. The o-alanine released is transported
back into the cytoplasm.
Cross-linkage by transpeptidation outside the cytoplasm seems to be
advantageous since it uses the energy of the peptide bond and, therefore,
M-G-M-G-M-G-M-G-M-G
f f
M-G-M-G-M-G-M-G-M-G
pep tidoglycan ~ {
M-G-M-G-M-G-M-G-M-G
~ f
M-G-M-G-M-G-M-G-M-G- MurNAc-GIcNAc
I
pen tapep ti de
3
undecaprenyl-PP
MurNAc-GIcNAc
I
pentapeptide
cell
membrane
UDP-
cytoplasm
UMP UDP
Figure 3.31. Reactions involved in the elongation of a peptidoglycan chain by a
disaccharide peptide. GlcNAc (G), N-acetylglucosamine; MurNAc (M), N-acetyl-
muramic acid; 7, peptide bridges. 1, transfer of MurNAc-penta-peptide to unde-
caprenyl-P; 2, formation of disaccharide pentapeptide; 3, transfer to peptidoglycan
chain.
ollter membrane
cell membrane
Figure 3.32. Cross-linkage by transpeptidation. o-Alanine is released from one

peptide unit and a peptide bond is formed between the remaining o-alanine residue
and the free amino group of the diaminopimelate residue of an adjacent peptide
unit. Terminal o-alanine residues, which do not participate in transpeptidation
reactions, are removed by a o-alanine carboxypeptidase so that the final peptido-
glycan contains only tetrapeptides.
does not require ATP. It also is apparent that for cross-bridging, an amino
acid with a free functional group must be present in position 3 of the
peptide unit. In E. coli peptidoglycan this position is occupied by diamino-
pimelate while the peptidoglycan of other microorganisms contains lysine
or ornithine (Chapter 5).
Several antibiotics interfere with peptidoglycan synthesis. Penicillin and
cephalosporin (so-called ,I3-lactam antibiotics) inhibit the transpeptidation
reaction. Since these cross-bridging reactions occur in growing, but not
in resting cells it is understandable that penicillin and cephalosporin
preferentially kill growing cells. Penicillin is not a very potent drug
against Gram-negative bacteria such as E. coli because in order to reach
the peptidoglycan layer the penicillin must penetrate the rather thick
lipopolysaccharide layer and other components of the outer membrane.
Gram-positive bacteria are more severely affected by penicillin than
Gram-negative bacteria. Recently developed semisynthetic ,B-Iactam anti-
biotics are also very effective against Gram-negative microorganisms;
because of their chemical structure they accumulate in the outer membrane
and are present in high concentrations in the neighborhood of the
peptidoglycan layer.
3. Outer membrane layer. This layer consists of about 40% lipopolysac-
charide, 20% phospholipids, and 40% protein. The arrangement of these
components is shown schematically in Fig. 3.33.
Lipid A contains fatty acids different from those present in phos-
pholipids; most notably 3-hydroxytetradecanoic acid is found. Four mole-
cules of this acid are linked to the glucosamine disaccharide of lipid A
through ester and amide linkages. Two of these hydroxy fatty acids carry in
addition one lauric or myristic acid esterified with their OH-groups.
repeating oligosaccharide chain
core oligosaccharide
lipid A
phospholipid
t - - - - - - - protein
- - - - periplasmatic space
Figure 3.33. Arrangement of components of the outer membrane layer.

[B. Lugtenberg, Trends Biochem Sci 6, 262-265 (1981)].
repeating oligosaccharide
!
glucose _ ;V-Jcetylglucosallllne
!
glucose
!
glucose _ galactose
!
core
oligo-
saccharide
heptose _ heptose-
!
heptose--- ethanolamine
!
dOclA _ dOdA-0
ethanoLlllllllt'
I
!
0 - glucosamine gluc'osamine-0
IOllOl 1" 1"

lipid A
Figure 3.34. Composition of the lipid A and the RI-core oligosaccharide of E. coli.
dOclA, 3-DeoxY-D-manno-octulosonate (formerly abbreviated as KDO). Fatty
acids: 4 mol 3-hydroxytetradecanoic acid and 2 mol saturated fatty acids esterified
with the OH-groups of 3-hydroxytetradecanoic acid. Arrows indicate direction of
glycosidic linkage from reducing position to non-reducing position.
Furthermore. the disaccharide carries two phosphate groups in I and 4'-

position (Fig. 3.34). Lipid A is integrated in the outer leaflet of the outer
membrane and carries the core oligosaccharide. The latter contains two
unusual components, L-glycero-D-manno-heptose and 3-deoxY-D-manno-
octulosonate (formerly called 2-keto-3-deoxyoctonate. KDO). Present in the
core oligosaccharide are also glucose. galactose. and N-acetylglucosamine.
Hlf\'
H ~ /CHzOH HO~/CH~OH
Ho~L o HO
HO/'-\.... ----OH\ /
HO\~\- 1 .... H~ \ OH
\'COOH
OH
heptose
N-acetylgJucosamine
t
galactose _ mannose
!
rhamnose
! n
lipid A
Figure 3.35. Composition of a repeating oligosaccharide chain of E. coli.
Synthesis of the core oligosaccharide occurs stepwise. First UDP-

derivatives of the appropriate sugars travel by unknown mechanisms to the
lipopolysaccharide area and are added to the growing chain by specific
transferases. The composition of the core oligosaccharide is similar in all
Enterobacteriaceae (in the case of E. coli five core types are known).
However, great differences exist in the composition of the repeating
oligosaccharide chains. Usually four to six sugars form a repeating unit;
within one species, such as E. coli or especially Salmonella typhimurium,
there is a large variation in the nature of these sugars and their sequence.
One example of a repeating unit found in E. coli strain 075 is shown in
Fig. 3.35. The repeating oligosaccharide chains which consist of up to 30
units are primarily responsible for the somatic (0) antigenic specificity of
Gram-negative bacteria.
Biosynthesis of the repeating units proceeds by reactions analogous to
those of peptidoglycan formation. The enzymes involved are membrane-
bound and undecaprenyl phosphate serves as lipid carrier. Nucleoside
diphosphate-activated sugars are transferred to undecaprenyl phosphate in
the right order to form an undecaprenyl-PP-oligosaccharide. At the outer
surface of the cell membrane these units are polymerized and then
transferred to the core oligosaccharide.
C. Synthesis of informational macromolecules

The macromolecules we have discussed thus far are composed of repeating
sequences of building blocks. Glycogen, for instance, consists of glucose
molecules connected to one another by 0'-(1 ~ 4) or 0'-(1 ~ 6) linkages.
Peptidoglycan is a polymer with an alternating sequence of N-acetyl-

glucosamine and N-acetylmuramic acid. The latter bears a tetrapeptide
chain, which for a particular microorganism, is always composed of
the same four amino acids. Consequently it is understandable that these
macromolecules can be synthesized by a small number of specific enzymes
and that there is no requirement for a template to bring the building blocks
together in the appropriate sequence. This becomes different when we turn
to the synthesis of nucleic acids and proteins. Here, regular repeating units
are absent and the synthesis of these macromolecules must proceed at a
template containing the necessary information. With the exception of
certain viruses the same hierarchy in template function is established in all
organisms: DNA serves as template for DNA and RNA synthesis; RNA
serves as template for protein synthesis. The processes involved will be
discussed here only briefly.
I.....--20A----+1-1
34A
Figure 3.36. Double helix structure of DNA. D, Deoxyribose; P, phosphate ester

bridge; A, adenine; C, cytosine; G, guanine; T, thymine.
1. DNA and RNA synthesis. DNA is a double helix (Watson-Crick

model). The two polynucleotide strands are held together by hydrogen
bonds, which can be most effectively formed only between two pairs of the
four bases present in DNA, between adenine and thymine and between
guanine and cytosine. Figure 3.36 shows a short piece of a DNA double
helix and Fig. 3.37 a duplex of two pentanucleotides. It can be seen that
within one strand the nucleosides are connected to one another by
phosphate bridges between the 3' and 5' carbons of deoxyribose. Both
strands run antiparallel, i.e., in the left-hand strand the chain proceeds
from a terminalS' -phosphate group to a 3' -hydroxyl group, while the
opposite is true in the right-hand strand.
Figure 3.38 shows the first two base pairs in more detail. The hydrogen
bonds allow pairing between A and T and between G and C so perfectly
that other combinations of the bases are not possible. Since the genetic
information stored in DNA is in the form of defined base sequences, the
following mechanism lends itself to the duplication of this information: The
two strands become separated; two new strands are formed along the
parental strands, and the nature of each nucleotide added to the growing
chain is determined by the nature of the base at a similar position in the
parental strand. The above duplication is catalyzed by DNA polymerase
and the process is called replication (illustrated in Fig. 3.39).
r---------------~
I OH:
II15t>-
3 A T 5I
-q3 I
'b-
I
I)
5' I 3
I
L_
p
G= C
I
5II 3'
_____________ p_J
{j3:
:t>- T A-q;
-q; 5'
p p
c - G
3' :t>-
C-q:
p p
:D-
OH
G-

q = deoxyribose
Figure 3.37. A duplex of two pentanucleotides.

l
H
N H
OH
/
-)-(N-H-------O CH 3
N~N------1I-0H
N /' ~ '\
deoxyribose H 0 '-----deoxyribOSe

H_: H
H
H\N~O
/
N-{ r-H-------n
N==<f-H-------OJ-N\
deoxyribose deoxyribose
H
Figure 3.38. AT and GC base pairs held together by hydrogen bonds.
A B
5' 3'
single-strand binding protein- , - single-strand binding protein
'\"
I-.. t-
5'
vv~~o
3'
5'
,.::;~' '0<,
.:, b
Figure 3.39. DNA replication. a: Replication fork formation by helicase and
single-strand binding proteins; action of DNA polymerase III and of ligase. JVV,
RNA primer. b: Strand replication indicating base recognition, addition of the
incoming nuc1eotides at the 3'-hydroxyl group, and release of pyrophosphate.
In detail, replication is quite complicated; approximately 30 proteins are

involved in this process in E. coli. The two strands have to become
separated; this is accomplished by a protein called helicase and by DNA
binding proteins. A complication arises from the fact that DNA
polymerase requires a primer nucleic acid. So beginning at the 3'-end of
one parental strand (strand A in Fig. 3.39) a short piece of RNA is
synthesized by a RNA polymerase, then DNA synthesis is started by
DNA polymerase III, which adds monomer after monomer to the 3'-
hydroxyl group of the growing strand. Chain elongation thus proceeds in
the 5' ~ 3' direction. There are about 10 molecules of DNA polymerase
III present in one E. coli cell. Each molecule can handle 15,000 nucleotide
molecules per minute. Since chain elongation is possible only in the 5' ~ 3'
direction, the replication of the second parental strand (strand B in Fig.
3.39) has to proceed in the direction opposite that of replication fork
formation. Apparently, the enzymes "wait" until a strand about 1000
nucleotides long is available, then a small RNA primer is formed and a
DNA fragment is synthesized afterwards (Fig. 3.39). These fragments are
called Okazaki fragments after their discoverer. Their conversion into an
intact strand requires the action of two enzymes. DNA polymerase I
removes the RNA primers by hydrolysis and replaces them' by DNA
pieces. This enzyme was the first known DNA polymerase; it was
discovered by A. Kornberg and associates in 1955. The second enzyme
involved is DNA ligase; it connects the Okazaki fragments with one
another by forming phosphodiester bridges between the 3' -hydroxyl
groups and the 5'-triphosphoester groups with the release of pyrophos-
phate. The overall reaction catalyzed by the DNA polymerases is:
. . lemplate
n deoxynucleoslde trtphosphate ----. DNA n + n PP i
The pyrophosphate is hydrolyzed to phosphate by a pyrophosphatase.
Consequently, DNA formation costs the cell two ATP per monomer. In
addition, ATP is required for the formation of the complex between the
template and the polymerase (initiation complex). Since the DNA content
of cells is approximately 3%, only a small percentage of the ATP required
for growth is invested in DNA synthesis (see Table 3.2).
The synthesis of RNA proceeds principally by a similar mechanism.
Cells contain a DNA-dependent RNA polymerase, which requires GTP,
CTP, ATP, and UTP as substrates. Thus RNA is different from DNA
because uracil is present in place of thymine. This does not affect pairing
with adenine since the structures of uracil and thymine differ by only one
methyl group (see Table 3.4). RNA polymerase transcribes only one of the
DNA strands and chain elongation proceeds in the 5' ~ 3' direction.
Furthermore, the entire DNA genome is not transcribed into RNA
sequences; genes that need not be expressed because of the physiological
conditions of the cell remain silent. Also, genes that code for a protein
required in large amounts are transcribed more frequently than others.
Thus the RNA polymerase must receive signals "telling" it which DNA
strand to transcribe and at what point transcription is to be initiated and
terminated. Transcription is, therefore, also very complicated.
Likewise, RNA polymerase has a very complicated structure; it consists
of five subunits, two a-subunits and one each of {3, {3' , and a subunits. Thus
four different polypeptide chains are present in the enzyme molecule. The
so-called core enzyme a2{3{3' (without u) exhibits catalytic activity but it is
unable to recognize start signals at the DNA. Yet aggregation of the
a-factor with the core enzyme yields the physiologically competent RNA
polymerase, and the u-factor, therefore, causes recognition of the regions
at the DNA where the initiation of RNA synthesis is to take place. These
regions are called promotor regions. Recognition of the stop signal for
RNA synthesis, at least in some instances, involves a protein accessory
factor to trigger RNA polymerase to stop at defined points. Figure 3.40
summarizes the DNA-dependent RNA synthesis.
RNA synthesis requires the same amount of ATP per monomer that is
needed for DNA synthesis. From the three classes of RNA formed,
transfer RNA, ribosomal RNA, and messenger RNA, the latter has only a
short life (half-life in E. coli: 3 min) and is rapidly degraded into
nucleoside monophosphates; following their phosphorylation to the corres-
ponding triphosphates, they can be used again for RNA synthesis. When a
culture of E. coli doubles its cell mass in 30 min it has synthesized its
original messenger RNA content approximately 10 times! In terms of ATP
requirement, RNA synthesis is about 5 times more expensive for the cells
than DNA synthesis (see Table 3.2).
a-factor_
3Q: DNA
/
RNA
RNA
Figure 3.40. Synthesis of RNA. The core enzyme of RNA polymerase combines
with the u-factor. The latter recognizes the start signal at the DNA. During RNA
synthesis the a-factor is released. RNA synthesis is terminated by the p-factor.
2. Protein synthesis. In replication and transcription, a defined sequence

of bases in DNA functions as a template for the synthesis of DNA and
RNA molecules, which also have a defined sequence of bases. Both
processes can be understood on the basis of specific base pairing. It is
obvious that the synthesis of proteins consisting of a certain sequence of
amino acids is more difficult to understand, as it is impossible for an amino
acid itself to recognize a certain succession of bases in RNA. Thus the
process of conversion of a sequence of bases into a sequence of amino acids
(called translation) must comprise a machinery different from those used
for replication or transcription.
All protein functions are determined by their amino acid sequence.
Figure 3.41 is a model of an enzyme protein. The proper position of certain
amino acids is responsible for correct folding of the polypeptide chain, for
the formation of interchain covalent bonds (for instance S-S-bridges), and
the function of a certain region of the enzyme molecule as active site.
Clearly, the synthesis of biologically active proteins has to proceed with
great precision. It is possible that the incorporation of glutamate instead of
lysine at a certain position of the polypeptide chain could result in a protein
lacking enzyme activity.
In 1961 it was shown by Matthaei and Nierenberg that the formation of
polyphenylalanine could be catalyzed in a cell-free system when polyuri-
dylic acid was added as template; other amino acids were not polymerized.
Since it was already known by then that units of three nucleotides of RNA
(triplets) code for one amino acid, it was concluded that the triplet UUU is
36 54
1234567
Ala-Tyr- Val-Ile-Asn-Asp-Ser-
8 9 10 11 12 13 14
Cys-lle-Ala-Cys-G1y-Ala-Cys-
33 I5 16 I 7 I 8 19 20 21
Lys-Pro-Glu-Cys-Pro- Val- Asn-
22 23 24 25 26 27
lle-Glu -Glu -Gly -Ser-lle-
28 29 30 3 I 32 33 34
46 Tyr-Ala-lle-Asp-Ala-Asp-Ser-
35 36 37 38 39 40 41
Cys-lle-Asp-Cys-Gly-Ser-Cys-
9
II'Y 42 43 44 45 46 47 48
Ala-Ser-Val-Cys-Pro-Val-Gly-
49 50 51 52 S3 54
12 Ala-Pro-Asn-Pro-Glu-Asn-
Figure 3.41. Folded polypeptide chain of a protein. The amino acid sequence and
the structure of a bacterial ferredoxin (Peptostreptococcus asacchacolyticus) is
shown. Cube-shaped structures represent iron-sulfur clusters [E. T. Adman, L. C.
Sieker, and L. H. Jensen, J. Bioi. Chern. 248, 3987-3996 (1973)].
Table 3.6. The meaning of some base

triplets
DNA mRNA amino acid
AAA UUU phenylalanine

TTT AAA lysine
GGG CCC proline
TAC AUG methionine
the code word for phenylalanine. With the help of synthetically prepared
ribonucleotides of known base sequences it was then possible to unravel
the meaning of all 64 (4 3 ) possible combinations of triplets. Some code
words easy to remember are given in Table 3.6.
How does a base triplet bring about the incorporation of a particular
amino acid into a growing polypeptide chain? The principle is indicated in
Fig. 3.42. Several components are involved: messenger-RNA (mRNA),
which functions as the template; 70S ribosomes attached to it, which
catalyze the formation of peptide bonds and consist of 30S and 50S
subunits; transfer-RNAs (tRNAs), which are linked to the corresponding
amino acids by specific enzymes-the aminoacyl-tRNA synthetases. In the
example shown in Fig. 3.42 phenylalanine-specific synthetase links this
particular amino acid to the phenylalanine-specific tRNA. The latter
recognizes with its anticodon (AAA) region, by base pairing, the phenyl-
alanine-codon (UUU) of the mRNA. The peptide bond is formed by
transferring the polypeptide chain from the adjacent tRNA (proline-
specific-tRNA in Fig. 3.42) to the amino group of phenylalanine.
/ ' " 30S subunit of ribosome

~COOON-:l
L
~~----- mRNA
tRNA
\
. (\
AMP
synthetase ,AA~A
ATP
+
pp. +
, phenylalanine
peptide chain
Figure 3.42. Principle of translation.
The formation of aminoacyl-tRNA is catalyzed by aminoacyl-tRNA

synthetases according to the following equations:
R-CH-COOH + ATP + enzyme --+ [R-CH-CO-AMPj-enzyme + PPi
I I
NH 2 NH 2
[R-CH-CO-AMPj-enzyme + tRNA --+ R-CH-CO-tRNA + enzyme + AMP
I I
NH 2 NH 2
The aminoacyl group is attached to the 3'-OH group of the terminal

adenosine of tRNA. E. coli contains more than 50 different tRNAs, so that
for each amino acid one to three specific tRNAs are present. Also present
are at least 20 different aminoacyl-tRNA synthetases, and they are very
specific. Each of them is able to pick "its" amino acid out of the 20 around
in the cytoplasm, to form the correspondingadenylate and to transfer the
aminoacyl residue to the "correct" tRNA. The tRNAs consist of about 80
ribonucleotides. In addition to the usual bases U, A, G, and C they contain
unusual bases such as inosine, dihydrouridine, pseudouridine, or meth-
ylated derivatives of the usual bases. The molecule can be drawn in a
cloverleaf form (Fig. 3.43) from which the anticodon region and the
3'-hydroxyl group that binds the amino acid are apparent.
6 3' hydroxyl group
<r:
u
u
Anticodon
Figure 3.43. Cloverleaf of tRNA. Crystallographic studies have shown that the
molecule is more L-shaped. The anticodon region represents one end of the Land
the 3'-hydroxyl group for amino acid binding the other one.
Table 3.7. Composition of ribosomes
subunit rRNA proteins
30S 16S rRNA 21 proteins

(approx. 1600 ribonucleotides)
50S 23S rRNA 32 proteins

5S rRNA
The catalytic machinery for polypeptide synthesis is harbored in the

ribosomes. Their composition is given in Table 3.7. It can be seen that the
two subunits contain different ribosomal RNAs and a large number of
different protein molecules. The molecular weight of the subunits is quite
high (900,000 for the 30S subunit and 1.8 million for the 50S subunit).
The synthesis of all proteins of E. coli and other bacteria is initiated by
the binding of N-formylmethionyl-tRNA at the 30S subunit which then
becomes attached to the start signal AUG (or GUG) of mRNA. Three
proteins, called initiation factors, are required for these reactions. Subse-
quently the 50S subunit is bound and the 70S initiation complex is ready.
This binding is associated with the hydrolysis of GTP to GDP and Pi' It is
apparent from Fig. 3.44 that the N-formylmethionyl-tRNA is situated at
the so-called peptidyl (P) site of the ribosome. Peptide synthesis is now
started by binding of the proper aminoacyl-tRNA at the aminoacyl (A)
site. This reaction requires the elongation factors Tu and Ts and GTP. In
Fig. 3.44 the codon is CCC so that prolyl-tRNA becomes accepted. In the
next step the N-formylmethionine is transferred to the proline residue.
This is followed by the translocation step, which comprises the release of
the uncharged tRNA, movement of the peptidyl-tRNA to the P-site,
movement of the ribosome so that the next codon can be recognized at the
A-site, and GTP hydrolysis. Then the next aminoacyl-tRNA (phenyl-
alanyl-tRNA in Fig. 3.44) can be bound.
Termination of a polypeptide chain is signaled by special codons of the
m-RNA (UAG, UGA or UAA).
Most proteins do not contain N-formylmethionine as N-terminal amino
acid. The formyl group and in many cases the methionine residue are
removed from the finished proteins by specific enzymes.
Much energy is required for protein synthesis. The activation of an
amino acid is accomplished by the conversion of lATP to AMP + PP j ;
in one elongation cycle 2GTP are hydrolyzed to 2GDP + 2P j Thus the
equivalent of 4ATP is required for the formation of one peptide bond. This
makes protein biosynthesis the most expensive process in the cell. E. coli
growing on glucose has to invest about 50% of its metabolic energy into
protein synthesis.
~ f
A f;#=- pro
p~f-mel
pro-f-mel
u
u
"<
::J
Figure 3.44. Initiation of protein synthesis and peptide chain elongation. a: Binding
of N-formylmethionyl-tRNA to the 30S subunit. b: Association of the 50S subunit
and thereby formation of the initiation complex. c: Binding of prolyl-tRNA.
d: Formation of the first peptide bond. e: Translocation. f: Binding of phenylalanyl
tRNA. g: Peptide bond formation. h: Translocation. i: Binding of lysyl-tRNA.
X. The Requirement for an Anaplerotic

Sequence
It has been pointed out that intermediates of the tricarboxylic acid cycle,
such as a-oxoglutarate, serve as precursors in the biosynthesis of several
cellular constituents. Consequently, the tricarboxylic acid cycle has a dual
function: it regenerates reduced pyridine and flavine nucleotides for the
respiratory chain and supplies starting material for biosyntheses. This
latter function creates a problem, as the principle of the cycle is that a
C4 -dicarboxylic acid condenses with a Cz compound to yield a C6 com-
pound; the latter is oxidized in such a way that the C4 -dicarboxylic acid is
regenerated. If an intermediate such as a-oxoglutarate is drained off for
biosynthetic purposes, the pool of C4 -dicarboxylic acids must collapse.
Clearly, additional reactions have to be present, which form C4 -dicar-
boxylic acids directly from intermediates of glucose breakdown. According
to H. L. Kornberg, reactions of this type are called anaplerotic sequences
(replenishing reactions). The situation in E. coli growing with glucose is
illustrated in Fig. 3.45. Cycle intermediates are used for the synthesis of
glutamate, porphyrins, and aspartate; the oxaloacetate needed is supplied
by the PEP-carboxylase reaction:
PEP carboxylase. I P
PEP + C02 oxa oacetate + i
Mutants of E. coli lacking this enzyme only grow on glucose when the
medium is supplemented with utilizable intermediates of the tricarboxylic
acid cycle or precursors thereof. As will be discussed later other organisms
aspartate = oxaloacetate
Figure 3.45. The function of PEP carboxylase as anaplerotic enzyme during growth
of E. coli on glucose.
Summary 93
employ pyruvate carboxylase as anaplerotic enzyme during growth on

carbohydrates, and during growth on substrates like acetate or pyruvate
different anaplerotic sequences are effective.
XI. Summary
1. Escherichia coli is able to synthesize all its cellular constituents from
glucose and minerals. The strategy of biosynthesis is that relatively few
intermediates of the energy metabolism (glucose-6-phosphate, phospho-
enolpyruvate, a-oxoglutarate, etc.) are used to synthesize monomers,
which subsequently are polymerized to give macromolecules; the latter
comprise approximately 97% of the dry weight of the cells.
2. At low concentrations of ammonia, it is assimilated by glutamine
synthetase. Glutamate is formed by the transfer of the amido group to
a-oxoglutarate (glutamate synthase reaction). At high concentrations of
ammonia, glutamate is formed from a-oxoglutarate also by glutamate
dehydrogenase.
3. The principal sulfur source is sulfate. It is reduced to HzS via APS and
PAPS. HzS reacts with O-acetylserine to yield cysteine.
4. Twenty amino acids are required for protein synthesis. Alanine and
aspartate are synthesized from pyruvate and oxaloacetate by transamina-
tion with glutamate as NH z donor. Asparagine is formed in a reaction
analogous to the glutamine synthetase reaction. Reduction of aspartate
yields aspartic semialdehyde-the precursor of lysine, threonine, and
methionine. Deamination of threonine gives a-oxobutyrate, which by the
successive action of four enzymes is converted into isoleucine. The same
four enzymes convert pyruvate into valine; an intermediate in valine
synthesis serves also as precursor in leucine formation. Serine, glycine, and
cysteine are formed from 3-phosphoglycerate, and proline and arginine
from glutamate. The three aromatic amino acids, tyrosine, phenylalanine,
and tryptophan, are synthesized from erythrose-4-phosphate and PEP;
shikimate and chorismate are common intermediates. Histidine is formed
from 5-phosphoribosyl-l-pyrophosphate and ATP in a complex series of
reactions by the action of nine enzymes.
5. Pentose phosphates and NADPH are formed in the oxidative
pentose phosphate cycle. Reactions yielding NADPH are the glucose-6-
phosphate and 6-phosphogluconate dehydrogenase reactions. The first
pentose phosphate formed is ribulose-5-phosphate; it can be isomerized to
yield ribose-5-phosphate. Pentose phosphates can be converted back into
hexose phosphates by the transketolase and transaldolase reactions.
6. Aspartate, carbamoyl phosphate, and 5-phosphoribosyl-l-pyro-
phosphate are the precursors of the pyrimidine nucleotides. First orotic
acid is formed; it is converted to orotidine monophosphate and finally
to UTP and CTP. An imidazole ribonucleotide is an intermediate in purine
nucleotide synthesis. Subsequent carboxylation, formylation, and amina-

tion reactions yield inosinic acid (IMP), which is the ultimate precursor of
ATP and GTP. Reduction of ribonucleotides to deoxyribonucleotides
takes place at the diphosphate level.
7. Escherichia coli contains phospholipids and glycolipids as consti-
tuents of membrane systems. Neutral fats are not stored. Long-chain fatty
acids are synthesized by soluble enzyme systems from acetyl-ACP (acyl
carrier protein) and malonyl-ACP. From the finished acyl-ACP the acyl
residue is transferred to glycerophosphate to yield phosphatidic acids.
Esterification of the latter with alcohols yields the various phospholipids.
8. UDP-N-acetylglucosamine and its 3-lactyl derivative, UDP-N-
acetylmuramic acid, are precursors of the peptidoglycan. They are synthe-
sized from fructose-6-phosphate. UDP-N-acetylmuramic acid is further
converted to UDP-N-acetylmuramyl-pentapeptide. The pentapeptide
chain of E. coli consists of L-alanine, o-glutamate, meso-diaminopime-
late, o-alanine, and o-alanine.
9. Membranes such as the cytoplasmic membrane consist of a double
layer of phospholipid molecules; they contain up to 60% protein. Mem-
branes have specific functions in transport processes.
10. Periodic macromolecules such as glycogen, peptidoglycan, and the
constituents of the outer membrane layer are synthesized from their
building blocks by specific enzymes. Glycogen is formed by transfer of
glucose from ADP-glucose to a primer oligosaccharide. The two building
blocks of peptidoglycan are first transferred to the membrane lipid carrier
undecaprenyl phosphate (bactoprenol) and then transported to the elonga-
tion site of the peptidoglycan. Polysaccharide bonds are formed and cross-
linkage between the polysaccharide chains is accomplished by peptide
bonds between the meso-diaminopimelate residue of one chain and the
o-alanine residue of another chain. The outer membrane layer consists of
lipid A, the core oligosaccharide, and repeating oligosaccharide chains.
11. The synthesis of informational macromolecules proceeds at tem-
plates. DNA serves as template for DNA and RNA synthesis and mRNA
serves as template for protein synthesis. Information is stored in DNA in
the form of defined base sequences; these sequences are recognized by
base pairing between adenine and thymine (uracil in the case of RNA) and
between guanine and cytosine. Substrates in DNA and RNA synthesis are
the ribonucleoside and deoxyribonucleoside triphosphates; pyrophosphate
is formed in the polymerization reactions.
In the translation of base sequences into amino acid sequences base
triplets of mRNA serve as code words for the amino acids. The base
triplets are recognized by base pairing with the anticodon region of tRNA.
For instance, the triplet UUU is recognized by the AAA-anticodon of
phenylalanine-specific transfer-RNA. Thus phenylalanine is incorporated
into a growing peptide chain when UUU appears at the mRNA. Protein
synthesis proceeds at the ribosomes, which consist of 50S and 30S subunits.
Summary 95
Protein synthesis is initiated by the binding of N-formylmethionyl-tRNA at

the 30S ribosomal subunit, which is attached to the initiation triplet of
mRNA. Subsequently the 50S subunit is bound and the formation of
peptide bonds begins. The equivalence of 4ATP is required for the
formation of one peptide bond.
12. When E. coli grows on glucose the PEP-carboxylase reaction serves
as an anaplerotic reaction.
Chapter 4
Aerobic Growth of
Escherichia coli on
Substrates Other Than
Glucose
Escherichia coli is able to grow on carbohydrates other than glucose and on

a number of organic acids and amino acids. The additional enzyme systems
required for the utilization of some of these substrates will be discussed in
this chapter.
I. Fructose and Lactose as Substrates

Fructose is an excellent substrate for the growth of many bacteria. As in
the case with glucose, E. coli employs a PEP-phosphotransferase system to
bring fructose into the cell. However, the product appearing inside the cell
is fructose-I-phosphate and not the 6-phosphate:
enzyme I
PEP + HPr ( l phospho-HPr + pyruvate
enzymes II + III
phospho-HPr + fructose I fructose-I-phosphate + HPr
Consequently, in addition to the fructose-specific enzymes II and III
of the transport system, a third enzyme is required for the utilization of
fructose: I-phosphofructokinase. The synthesis of this enzyme is speci-
fically induced by fructose. Figure 4.1 illustrates the difference in the first
steps of glucose and fructose metabolism. From this it becomes clear why
Kornberg and associates were able to isolate mutants that grew on glucose
but not on fructose. These mutants lacked I-phosphofructokinase.
Glucose-grown cells of E. coli are not able to utilize lactose imme-
diately. To do so several additional enzymes are required, and their
formation is induced when lactose becomes available to cells.
Fructose and Lactose as Substrates 97
fructose glucose
I I
r-
PEP-phosphotransferase systems
t fructose-I-
!
glucose-6-
~
TP glucose-6-phosphate isomerase
Ay
I-phospho- ADP
fructokinase ATP fructose-6- P
6-phosphofructokinase
fructose-l,6-bisphosphate
1
Figure 4.1. The first steps of glucose and fructose metabolism in E. coli.
Three enzymes are synthesized following the addition of lactose to the

growth medium: lactose permease, p-galactosidase, and p-galactoside
acetyltransferase. These enzymes are of great importance, not only
because they enable E. coli to grow on lactose but also because the
investigation of their coordinate formation has led to the operon model of
Jacob and Monod and to an understanding of the regulation of enzyme
synthesis. This will be discussed in Chapter 7.
Lactose permease catalyzes an energy-dependent transport of lactose
into the cell and f3-galactosidase hydrolyses lactose to yield glucose and
galactose (Fig. 4.2). The biological function of the f3-galactoside acetyl-
transferase, which acetylates lactose and other f3-galactosides with acetyl-
lactose
I
permease (energy-dependent) membrane
0"
lactose CH20H
HO~C"'~"
OH
OH
H2 0
OH
J3-galactosidase
galactose glucose
Figure 4.2. Conversion of lactose to galactose and glucose.
98 4: Aerobic Growth of Escherichia coli on Substrates Other Than Glucose
UDP-glucose
galactose-l-
( UD'-",,,.=
Figure 4.3. Conversion of galactose into glucose-I-phosphate. 1, Galactokinase; 2,
glucose: galactose-I-phosphate uridylyltransferase; 3, uridine diphosphate glucose
epimerase.
CoA, is probably that of a detoxification enzyme: nonmetabolizable

structural analogues of lactose are acetylated and excreted.
The glucose formed from lactose is phosphorylated to glucose-6-
phosphate by hexokinase. It should be recalled that hexokinase is not
required by E. coli growing on glucose or fructose (the PEP-phos-
photransferase systems yield hexose monophosphate directly); how-
ever, hexokinase is employed for phosphorylation of glucose or fructose
formed inside the cells by oligosaccharide hydrolysis.
The galactose formed from lactose induces the synthesis of three
enzymes involved in its further metabolism: galactokinase, glucose: galac-
tose-I-phosphate uridylyItransferase, and UDPglucose epimerase. To-
gether they catalyze the following reactions: Galactose is phosphory-
lated at carbon atom 1; UDP-glucose reacts with galactose-I-phosphate to
yield glucose-I-phosphate and UDP-galactose; the latter is epimerized to
UDP-glucose (Fig. 4.3).
With the induction of all these enzymes it then becomes possible to feed
lactose into the Embden-Meyerhof-Parnas and subsequent pathways.
II. Pentoses as Substrates

Pentoses such as o-arabinose, o-xylose, and o-ribose are widespread in
nature, and E. coli is able to grow on them. The strategy of pentose
utilization is that these compounds are actively taken up by the cells and
converted into those pentose-5-phosphates which are substrates of the
transaldolase or transketolase (Fig. 4.4). o-Xylulose-5-phosphate and
o-ribose-5-phosphate are then converted by these enzymes to fructose-6-
phosphate and glyceraldehyde-3-phosphate, which finally are degraded via
the Embden-Meyerhof-Parnas pathway (see Fig. 3.14). It is apparent
from Fig. 4.4 that kinases specific for three different o-pentoses are
available in E. coli. Other o-pentoses have to be isomerized first to the
substrates of these kinases before phosphorylation.
Acetate, Pyruvate, and L-Malate as Substrates 99
CHO CHO CHO

I I I
HC-OH HO-CH HC-OH
I I I
HC-OH HC-oH HO-CH
I I I
HC-OH HC-OH HT-OH
I I
CH 2 0H CH 2 0H CH 2 0H
o-ribose o-arabinose o-xylose
11, 11,
yH 2 0H CH 2 0H
I
c=o c=o
I I
ATP HC-OH HO-CH
I I
3 HC-OH HC-OH
I I
ADP CH 2 0H CH 2 0H
o-ribulose o-xylulose
fAT!' ADP FAn ADP
o-ribose-5-P ::;;.i=====~.. o-ribulose-5-P .. o-xylulose-5-P
Figure 4.4. Reactions in E. coli leading from pentoses to o-ribose-5-phosphate and

D-xylulose-5-phosphate. 1, o-Arabinose isomerase; 2, o-xylose isomerase; 3,
ribokinase; 4, ribulokinase; 5, xylulokinase.
III. Acetate, Pyruvate, and L-Malate as

Substrates
When E. coli is transferred from a glucose or fructose medium to a medium
containing acetate as carbon and energy source, this change is rather
drastic as all cellular constituents must then be synthesized starting from
Cz-units. Obviously this will require a number of new enzyme systems, a
description of which now follows. The formation of NADH to be used for
ATP production in the respiratory chain is very simple. Acetate is actively
transported into the cell and subsequently converted into acetyl-CoA by
acetyl-CoA synthetase:
acetate + HS-CoA + ATP ---. acetyl-SCoA + AMP + PP j
This activation reaction proceeds at the expense of ATP hydrolysis to
AMP and pyrophosphate; acetyl-AMP is an intermediate.
Acetyl-CoA then is fed into the tricarboxylic acid cycle, thus yielding
NADH. The problem during growth on acetate, however, is how inter-
mediates of the cycle that serve as starting materials in biosyntheses are
regenerated. Clearly, the anaplerotic sequence yielding oxaloacetate from

PEP, which is important during growth on carbohydrates, cannot be
functional; there is no direct way to synthesize PEP from acetate. Two
other anaplerotic enzymes are employed instead: isocitrate lyase and
malate synthase. The first of these catalyzes the cleavage of isocitrate to
succinate and glyoxylate:
CH 2-COOH CH 2-COOH
I isocitrate lyase I
CH-COOH ( )
CH 2-COOH
I +
HO-CH-COOH O=CH-COOH
Malate synthase catalyzes the condensation of acetyl-CoA with gly-
oxylate to yield L-malate-a reaction analogous to the citrate synthase
reaction.
CH)-CO-SCoA CH 2-COOH
+ + H 20
malate synthase
)
I + HSCoA
O=CH-COOH HO-CH-COOH
Together with enzymes of the tricarboxylic acid cycle these reactions
form the so-called glyoxylate cycle. The significance of this cycle is
illustrated in Fig. 4.5. Reactions of the TCA cycle plus isocitrate lyase
catalyze the oxidation of acetyl-CoA to glyoxylate. The latter can then
condense with another molecule of acetyl-CoA to yield L-malate. Thus,
extra oxaloacetate becomes available (actually formed from two acetyl-
CoA), which can be used to form L-glutamate (via citrate and a-
oxoglutarate) and other compounds derived from cycle intermediates.
Moreover, oxaloacetate serves as starting material for all the carbohy-
drates required for polymer synthesis (Fig. 4.6). The ultimate precursor of
gluconeogenesis is phosphoenolpyruvate, and it is formed from oxaloace-
tate and ATP by PEP carboxykinase:
PEP carboxykinase
oxaloacetate + ATP , ' phosphoenolpyruvate + ADP + CO 2
All the reactions of the Embden-Meyerhof-Parnas pathway between
PEP and fructose-l,6-bisphosphate are reversible so that under the
appropriate conditions the latter can be synthesized from PEP. Fructose-
1,6-bisphosphate is taken out of equilibrium by the action of a specific
phosphatase, which irreversibly hydrolyzes the bisphosphate to fructose-6-
phosphate.
The glyoxylate cycle reactions serve as an anaplerotic sequence when
E. coli grows on acetate or on compounds that are degraded via acetyl-
CoA. Surprisingly, E. coli employs another anaplerotic sequence when
growing on pyruvate. This substrate is oxidized to acetyl-CoA by the
pyruvate dehydrogenase complex, which is subsequently fed into the TCA
cycle. In the synthesis of PEP a novel enzyme is involved: PEP synthetase.
In this connection it should be recalled that in the Embden-Meyerhof-
Acetate, Pyruvate, and L-Malate as Substrates 101
2 eitrate~
o
3 oxaloaeelale
2 isoeitrale
3 malale
Figure 4.5. The function of the tricarboxylic acid cycle and the glyoxylate cycle
reactions during growth of E. coli on acetate. For illustration, 2 oxaloacetates are
allowed to react with 2 acetyl-CoAs to yield 2 citrate. These are converted into
2 isocitrates, one of which is oxidized through the tricarboxylic acid cycle and
regenerates 1 oxaloacetate. The second is cleaved into succinate and glyoxylate;
oxidation of succinate yields a second molecule of oxaloacetate. Glyoxylate
condenses with acetyl-CoA to form L-malate, from which a third molecule of
oxaloacetate can be formed. Thus one oxaloacetate is available for biosynthetic
purposes.
aeelate
g1yoxylale
cycle
fruelose-6-P
oxaloaeelate
phosphatase PEP
carboxykinase
Em bden-Meyerhof-Parnas
pathway
lj
fructose-I. 6-
bisphosphate
.. PEP
Figure 4.6. Reaction sequence leading from acetate to hexosemonophosphate.

Parnas pathway pyruvate kinase catalyzes the conversion of PEP into

pyruvate, a reaction accompanied by a decrease of the standard free
energy at pH 7 of 23.8 kJ Imol (5.7 kcal/mol)
PEP + ADP pyruvate kinase, pyruvate + ATP IlGO' = -23.8 kJ Imol

(-5.7 kcal/mol)
Thus, this reaction is irreversible and cannot provide the PEP for
gluconeogenesis. In the PEP synthetase reaction a negative free energy
change is achieved by the hydrolysis of one of the energy-rich pyrophos-
phate bonds of ATP.
PEP
synthetase
pyruvate + ATP ) PEP + AMP + Pi IlGO, = -8.4 kJ Imol
(-2.0 kcal/mol)
A pyrophosphorylated enzyme is an intermediate in this reaction:
enzyme + ATP ( ) enzyme-PP + AMP
enzyme-PP + H 2 0 ( ) enzyme-P + Pi
enzyme-P + pyruvate ( ) PEP + enzyme
The oxaloacetate required to replenish the pool of TCA cycle in-
termediates is formed, as it is during growth on glucose and other
carbohydrates, from PEP by PEP carboxylase. The importance of PEP
synthetase during growth of E. coli on pyruvate or lactate is well
documented. Mutants were isolated, which grew on glucose and acetate
but not on pyruvate and lactate; they lacked PEP synthetase.
C4 -dicarboxylic acids are good substrates for aerobic growth of E. coli.
L-Malate, for instance, is taken up and oxidized to oxaloacetate by malate
dehydrogenase. Oxaloacetate thus formed can serve as substrate in
anabolic reactions and in the tricarboxylic acid cycle. However, it is rather
obvious that an enzyme system is required that converts malate into
pyruvate; otherwise acetyl-CoA and the various metabolites derived from
pyruvate could not be formed. The synthesis of pyruvate is accomplished
by malic enzyme:
malic enzyme
L-malate + NAD+ ( ) pyruvate + NADH + H+ + CO 2
In addition to the NAD+ -specific enzyme, an NADP+ -specific malic
enzyme is present in E. coli cells growing with malate; it provides NADPH
for biosyntheses. PEP is formed from oxaloacetate by PEP carboxykinase
and from pyruvate by PEP synthetase.
The fraction of ATP that has to be invested in the biosynthesis of
monomers during growth depends very much on the nature of the growth
substrate; it is small when glucose is the substrate because the formation of
the precursors in monomer synthesis (PEP, pyruvate, acetyl-CoA) is
connected with a gain of ATP. On the other hand, the formation of the
Summary 103
same compounds from acetate requires ATP. Stouthamer calculated that

34.8 mmol of ATP are required for the formation of 1 g of cells from
glucose and 99.5 mmol of ATP for the formation of 1 g of cells from
acetate.
IV. Summary
1. The change of the growth substrate necessitates changes in the
enzyme equipment of the cells: (1) synthesis of the appropriate catabolic
enzymes and the substrate-specific transport systems; (2) formation of the
appropriate anaplerotic sequences.
2. Fructose transport by the PEP-phosphotransferase system yields
fructose-1-phosphate, which is phosphorylated by 1-phosphofructokinase
to fructose-1 ,6-bisphosphate. Lactose is transported by lactose permease
into the cells and subsequently cleaved by ~-galactosidase to glucose and
galactose.
3. Pentoses are converted by specific kinases and isomerases to xylu-
lose-5-phosphate and ribose-5-phosphate. These pentose phosphates are
then converted to fructose-6-phosphate plus glyceraldehyde-3-phosphate
and degraded via the Embden-Meyerhof-Parnas pathway.
4. During growth on acetate E. coli requires the glyoxylate cycle as
anaplerotic sequence. It consists of enzymes of the tricarboxylic acid cycle
plus isocitrate lyase and malate synthase and accomplishes the synthesis of
C4-dicarboxylic acids from acetate. In addition, PEP carboxykinase is
required for PEP formation in gluconeogenesis.
5. During growth of E. coli on pyruvate, PEP synthetase and PEP
carboxylase serve as anaplerotic enzymes; during growth on L-malate,
malic enzyme is required for the formation of pyruvate.
6. The fraction of ATP which has to be invested in the biosynthesis of
monomers during growth is dependent on the nature of the growth
substrate.
Chapter 5
Metabolic Diversity of
Aerobic Heterotrophs
In the preceding chapters we have used E. coli as a model organism in

order to become acquainted with the principal reactions participating in
the ATP production during aerobic growth and in the biosynthesis of
cellular material. Furthermore, we have discussed the changes in the
enzymatic machinery of the cell that must occur when glucose is replaced
by other substrates.
This relatively simple picture becomes much more complicated when we
consider other bacteria. Many of them resemble E. coli in their nutritive
requirements but use modified catabolic pathways and alternate anaplero-
tic sequences or form different storage materials and different cell wall
components. Some of this metabolic diversity among aerobic heterotrophs
will be discussed in this chapter.
I. The Different Mechanisms for

the Uptake of Substrates
E. coli cells take up glucose and fructose by phosphotransferase systems
and lactose, in contrast, by active transport. When comparing the mode in
which substances from the environment cross the cell membrane into the
cytoplasm four different mechanisms can be distinguished: passive diffu-
sion, facilitated diffusion, active transport, and group translocation.
A. Passive diffusion
The transported substance does not specifically interact with components
of the cell membrane. It crosses the membrane until equilibrium is reached
between the concentration inside and that outside. As the concentration of
The Different Mechanisms for the Uptake of Substrates 105
most of the metabolites in nature is higher inside the cell than outside, it is
clear that transport by passive diffusion must be restricted to a small group
of substances, i.e., gases such as oxygen and carbon dioxide or water.
B. Facilitated diffusion
This process is similar to passive diffusion in that neither one requires
metabolic energy, and both are freely reversible such that equal concentra-
tions of the substance are found inside and outside the cell. However,
unlike passive diffusion, facilitated diffusion involves transport of a sub-
stance via a specific membrane carrier; the substance is bound to the car-
rier on the outside and released on the inside of the cell. Transport is,
therefore, substrate-specific. If, for instance, a membrane contains only a
glucose-specific carrier, other carbohydrates will not be transported into
the cell. Erythrocytes and yeast cells take up sugars by facilitated diffusion,
while in bacteria, this transport mechanism does not seem to be very
important. It might be involved in the excretion of certain fermentation
products by anaerobic bacteria.
C. Active transport
Figure 5.1 illustrates the effectiveness of active transport as compared to
the diffusion processes.
Clearly, active transport allows substrate saturation of the cell's en-
zymatic machinery at much lower concentrations of the substrate in the
medium than do the diffusion processes. A several hundredfold concentra-
tion of compounds has been observed; thus active transport enables the
cells to live nicely in environments containing substrates in low concentra-
tions, a situation common in nature. An uptake process is defined as active
:.;
active transport
~
"
"0
.~ 100
.2
'" 10
e
c::
0
u
OJ
~
.J
10 100
Lactose concentration ou{sidt' (arbitrary units)
Figure 5.1. Internal concentration of lactose in E. coli as dependent on the lactose

concentration in the medium and the transport process involved.
106 5: Metabolic Diversity of Aerobic Heterotrophs
transport according to the following parameters:

1. It is substrate-specific; a carrier-substrate complex is formed on the
outside surface of the membrane.
2. It requires metabolic energy; the carrier has a high affinity for the
substrate when facing the outside surface of the membrane and a low
affinity for it when facing the inside surface. This modification of the
carrier requires energy.
3. It allows transport of its substrate against a concentration gradient; this
results from the change in substrate affinity of the carrier when turning
from outside to inside.
4. It releases the substrate unmodified into the cytoplasm (different from
group translocation).
Active transport is schematically illustrated in Fig. 5.2.
What is the energy source of active transport? In most systems it
apparently is the protonmotive force, which can be generated either by
electron transport or by ATP hydrolysis as catalyzed by the membrane-
bound ATPase (Fig. 5.3a and b). The protonmotive force is taken
advantage of by carriers having binding sites for protons and a particular
substrate, and both (protons and substrate molecules) are transported into
the cells (Fig. 5.3c). This type of active transport is called symport; for
example, lactose or K+ is taken up by E. coli with such a transport system.
high affinity
Figure 5.2. Uptake of a substrate molecule by active transport. a: The carrier,

consisting of two subunits, faces the outside surface of the membrane such that it
has a high affinity for substrate binding; b: At the expense of metabolic energy the
carrier-substrate complex is modified. The substrate binding site is now exposed to
the inside of the cell; its substrate affinity is drastically lowered and the substrate is
released even at high intracellular substrate concentrations.
Some compounds (e.g., L-glutamate by E. coli) are taken up by a sodium

ion-dependent symport mechanism (Fig. 5.3d). This system can function
only in combination with a Na +jH+ -antiport system that maintains a
concentration gradient between Na+(outside) and Na+(inside).
The fact that electron transport or ATP can provide the energy for
protonmotive force-driven active transport is important when we consider
that not only aerobes and phototrophs but also anaerobes require such
transport systems. Many of the latter organisms gain ATP only by
substrate-level phosphorylation and do not contain electron transport
chains.
There is a second active transport system present in E. coli and other
microorganisms (particularly in Gram-negative bacteria). It is involved in
the transport of amino acids, peptides, some sugars, organic acids, and in-
organic cations and anions and consists of a substrate-specific binding
protein and an ATP-requiring transport unit. The binding proteins reside
+ +
+ +
AH z
ATP
A
H'
H'
JW +120 2 H'
- ADP + Pi
H2O +
+ +
+
+ +
+ + substrate
substrate
Na' Na'
H'
H'
+ +
+ +
c d
Figure 5.3. The protonmotive force as the energy source for active transport.
a: Generation of the protonmotive force by respiration; b: Generation of the
protonmotive force by ATP hydrolysis; c: Active transport by a carrier with
binding sites for a particular substrate and for protons; d: Active transport by a
carrier with binding sites for a particular substrate and for Na +, subsequent
excretion of Na + by H+ INa +-antiport.
binding protein
/ substrate
Qj)J/
"
f~ IN
ATP ADP+Pi
Figure 5.4. Binding protein-mediated transport through the cytoplasmic mem-

brane. [Redrawn from G. Ferro-Luzzi Ames and C. F. Higgins; Trends Biochem.
Sci. 8, 97-100 (1983)]. P, Q, and M are subunits of the transport system.
in the periplasmic space of Gram-negative bacteria. Substrate molecules

approaching will be bound by them. Subsequently they interact with the
membrane-embedded components of the transport system, and at the
expense of the hydrolysis of ATP or of another high-energy phosphate
compound the substrate is actively taken up. A model for this type of
transport is presented in Figure 5.4. The binding proteins of Gram-
negative bacteria can be released from the cells by a so-called osmotic
shock: the bacteria are suspended in a hypertonic solution (a 20% solution
of glucose with buffer and EDTA) and then diluted in a cold solution of
low ionic strength (5 mM MgCI 2 solution). Under the first condition the
protoplasts shrink; under the second condition they expand against the cell
wall. This sudden change brings about the release of binding proteins.
It is not so that a certain bacterial species contains only one transport
system for a certain ion or substrate. E. coli contains more than three
different L-glutamate transport systems and two K + transport systems, a
protonmotive-force-driven one and an ATP-driven one. The latter is better
suited for extremely low K+ concentrations.
D. Group translocation
This process differs from active transport in that the substrate appears
inside the cell in a chemica!ly modified form-usually as phosphate ester.
Sugars are transported in many microorganisms by group translocation.
The driving force of transport in this case is that the sugar is trapped within
the membrane by reaction with a phosphorylated enzyme (enzyme III) and
that the phosphate ester formed is released into the cytoplasm. The
phosphorylated enzyme is generated using PEP as the principal source of
phosphate-bound energy. The mechanism of the PEP phosphotransferase
systems has been outlined in Chapter 2, Section 1 (glucose) and Chapter 4,
Section 1 (fructose). In E. coli, glucose, fructose, mannose, and mannitol
are transported by group translocation and in Staphylococcus aureus
lactose and other disaccharides also are thus transported.
Table 5.1. PEP phosphotransferase and active transport systems

for glucose in bacteria Q
active
organism phosphotransferase transport
Arthrobacter species +
Azotobacter vinelandii +
Bacillus megaterium +
B. subtilis +
Brucella abortus +
Clostridium butyricum +
C. pasteurianum +
Enterobacteria +
Micrococcus luteus +
Mycobacterium smegmatis +
Pseudomonas aeruginosa +
Staphylococcus aureus +
" Most data are from A. H. Romano, S. J. Eberhard, S. L. Dingle, and
T. D. McDowell, 1. BaCieriol, 104, ROR-RI3 (1970).
A phosphotransferase system is, however, not present in all bacteria

that are able to grow on glucose, as can be seen in Table 5.1. It appears
that this system occurs predominantly in facultatively and strictly anaerobic
bacteria. Aerobes such as Azotobacter vinelandii and Pseudomonas aeru-
ginosa employ an active transport system for the uptake of glucose. There
are, of course, exceptions: Arthrobacter species, Bacillus megaterium, and
B. subtilis, which are aerobes, transport glucose by a phosphotransferase
system into the cell. For anaerobes, the phosphotransferase system is of
great importance as it helps to save ATP.
E. Iron transport, a special problem

Aqueous environments that are in contact with air contain iron in the ferric
state as a ferric oxyhydroxide polymer of the general composition FeOOH.
The solubility product of this compound is in the order of 10- 20 M and is
very low as compared to the iron concentration required by micro-
organisms (0.1-1 JJ.M). In order to get enough of this important bioele-
ment for growth, aerobic organisms excrete chelators that form very stable
iron complexes. These complexes are taken up by the cells and are
degraded inside the cells with concomitant reduction of Fe 3 + to Fe 2 +. So
iron becomes available for cellular purposes (Fig. 5.5a).
E. coli develops four different Fe 3 + -chelator transport systems. They
are specific for Fe 3 + -ferrichrome, Fe3+ -enterochelin, Fe3+ -citrate, and
Fe 3 + -aerobactin.
OM _"'-----"_.....
CM====~
Fe 3+ -complex
FeOOH
b
Figure 5.5. Iron uptake with the aid of enterochelin. a: Model of uptake mechan-
ism. Shaded areas, receptor proteins in the outer and the cytoplasmic membrane.
b: Formula of enterochelin. OM, Outer membrane; eM, cytoplasmic membrane.
Interestingly, ferrichrome is not synthesized by E. coli itself; it is

produced by certain fungi. Both E. coli and Salmonella typhimurium, are,
however, able to take it up. The structural formular of enterochelin is
given in Fig. 5.5b. Fe3 + is bound by the six phenolic hydroxyl groups.
Fe3 + -chelating compounds are now termed siderophores; they occur in
practically all aerobic and facultative anaerobic microorganisms. In Gram-
negative microorganisms, specific receptor proteins are present for the
siderophores in the outer membrane. From there the complexes are
channeled to receptors on the cytoplasmic membrane. The receptors at the
outer membrane serve also as receptors for certain phages and for colicins
that are toxins.
F. Chemotaxis and flagellar movement

The binding proteins that have been described as constituents of the
ATP-driven active transport systems have an additional function in flagel-
lated microorganisms. They function as chemoreceptors in a sensing
system called chemotaxis. Peritrichously flagellated microorganisms such
as the enterobacteria are able to swim smoothly if their flagellae rotate
counterclockwise. A change to a clockwise rotation of the flagellae results
in an interruption of swimming of the organisms and in tumbling.
Organisms have now a device at their disposal to regulate the tumbling
frequency. This frequency is decreased if they swim in the direction of
increasing nutrient concentrations or decreasing repellent concentrations
(e.g., phenol). Swimming away from nutrients leads to an increase of
tumbling frequency: organisms stop swimming, tumble, and start swim-
ming into another direction which by chance may be more favourable than
the former one.
This very interesting sensing mechanism functions as follows: binding
proteins occupied by their substrate bind to a methyl-accepting chemotaxis
protein (MCP) which is embedded in the cytoplasmic membrane. There are
three different MCPs present in E. coli that can interact with several
binding proteins. The MCPs contain L-glutamate residues within their
polypeptide chain that are subject to methylation at their free carboxyl
groups with S-adenosylmethionine (SAM) under catalysis of a methyl-
transferase.
MCP-(COOH)n + SAM methyltransferase, MCP-(COOCH 3 )n +
S-adenosylhomocysteine
Also present in the cells is a methylesterase that hydrolyzes the methylated
MCP:
MCP-(COOCH 3 )n + nH 2 0 methylesterase, MCP-(COOH)n + nCH 3 0H
The combined action of these antagonistic enzymes results in a certain
methylation grade of MCP. This methylation grade is now increased if
substrate-loaded binding protein interacts with MCP (Fig. 5.6). Thus,
increasing nutrient concentrations in the neighborhood of the bacteria
increasing [C 1
methylation
more cGMP
swimming
resting activated
flagellum
"motor"
decreasing [C J
demethylation
less cGMP
more tumbling
Figure 5.6. Chemotaxis model. [C], Nutrient concentration; number of arrows, rate
of mediator production or measure of swimming/tumbling ratio. Increasing
nutrient concentration results in an increase of the swimming/tumbling ratio
(activated state).
result in an increase of the number of methyl groups in MCP. This then

results in an enhanced rate of the formation of the mediator cyclic
guanosine monophosphate that stimulates flagellar rotation. Under condi-
tions in which the substrate concentration becomes low, the binding
proteins dissociate from MCP and the methylation grade is slowly de-
creased. This finally is followed by more frequent tumbling.
G. The outer membrane is a barrier for nutrients

The rather thick peptidoglycan layer of the Gram-positive organisms is not
a barrier for hydrophilic molecules. They penetrate this layer and reach the
cytoplasmic membrane where "the decision is made" as to their transport
into the cell. Gram-negative organisms are surrounded by an additional
membrane, the outer membrane. This membrane protects the cells against
hydrophobic compounds such as the bile salts in the colon, because its
surface layer is built up of lipid A-bound polysaccharides and of proteins in
which these compounds are not soluble (see Fig. 3.33). Hydrophilic
molecules such as sugars or organic and inorganic salts also cannot pass
because the inner layer is made of phospholipids. To allow the nutrients to
reach the periplasmic space, the outer membrane contains waterfilled
pores that consist of trimers of the protein called porin (Fig. 5.7). Each
porin has the capacity to form a channel but the channels are "opened"
only if the porins are inserted into the outer membrane as trimers.
protein catalysing specific ----,.l:z::::..

facilitated diffusion
trimers of porin protein
Figure 5.7. Porin trimers and special transport proteins in the outer membrane.
rHo Nikaido. Angew. Chern. 91, 394-407 (1979)].
Molecules up to a molecular weight of approximately 600 can pass the

pores and reach the periplasmic space. Consequently, mono-, di-, and
trisaccharides can pass but tetrasaccharides cannot. Three different porins
with molecular weights around 35,000 per monomer are formed by E. coli.
The porins function also as phage receptor proteins.
In addition to the general pores for hydrophilic compounds the outer
membrane contains special transport proteins that allow facilitated diffu-
sion of certain substrates. The uptake of Fe-enterochelin has already been
mentioned. There are also pores for the uptake of vitamin B 12 , for maltose
and maltodextrins, and for other compounds that are useful to the cell.
H. Assembly of proteins into membranes and

protein export
We have seen that many proteins reside in the cytoplasmic membrane of
microorganisms and also in the periplasmic space and outer membrane of
Gram-negative bacteria. Exoenzymes are excreted by several micro-
organisms (see Chapter 6). The question is how these proteins reach their
destination. A mechanism that seems to operate for transport of many
proteins functions as follows: A protein destined for secretion contains an
NHrterminal leader peptide that is 15-30 amino acid residues long. This
signal peptide exhibits a high affinity to certain regions of the cytoplasmic
membrane; it enters the membranes and "pulls" the actual peptide
through it. Then the leader peptide is removed by action of a specific
"leader peptidase" (Fig. 5.8a).
mRNA
ribosome
!
Leader peptide
docking
protein
Figure 5.8. Transport of proteins into and through the cytoplasmic membrane. a: A
polypeptide with a N-terminal leader peptide is synthesized and pulled into the
membrane. b: A ribosome at which the leader peptide has just been synthesized
becomes associated with the membrane via a docking protein. The growing
polypeptide is pulled through the membrane. The leader peptide is apparently
degraded after cleavage by leader peptidase. Binding of the docking protein to the
ribosome proceeds via a protein that is associated only with those ribosomes
synthesizing leader peptides.
It is also possible that ribosomes carrying the leader peptide and the yet
incomplete polypeptide associate with the membrane at sites where a
docking protein is inserted into the membrane. The polypeptide is then
completed while it is already pulled through the membrane (Fig. 5.8b).
Assembly of proteins into membranes or protein export requires that
the membrane is in an energized state. These processes are proton-
motive-force-dependent. For the insertion of some proteins, the simul-
taneous insertion of phospholipid molecules is necessary. Transport of
proteins into membranes is a very complex process, and there are still
many questions to be answered.
II. The Entner-Doudoroff Pathway

Besides the Embden-Meyerhof-Parnas pathway, there is a second im-
portant pathway used for carbohydrate breakdown, which is widely dis-
tributed among bacteria. It waS first discovered by Entner and Doudo-
roff in Pseudomonas saccharophila. As in the pentose phosphate cycle,
glucose-6-phosphate is first dehydrogenated to yield 6-phosphogluconate
(Fig. 5.9). This is converted by a dehydratase and an aldolase reaction into
1 molecule of 3-phosphoglyceraldehyde and 1 molecule of pyruvate. The
3-phosphoglyceraldehyde can then be oxidized to pyruvate by the enzymes
of the Embden-Meyerhof-Parnas pathway.
What are the differences between the Embden-Meyerhof-Parnas
(EMP) and the Entner-Doudoroff (ED) pathways and how is it possible to
differentiate between them? The key enzymes of the EMP path-
way-meaning the enzymes common only to this pathway-are 1- and
6-phosphofructokinase, respectively. All the other enzymes of the EMP
pathway may be part of other metabolic sequences as well, Fru-P z aldolase,
for instance, in gluconeogenesis or in the pentose phosphate cycle. The key
enzymes of the ED pathway are 6-phosphogluconate dehydratase and
2-keto-3-deoxy-6-phosphogluconate (KDPG) aldolase. Thus, these path-
COOH
I
HC=O COOH COOH
c=o
I
I NADPH+W I I
N):+2
CH]
HC-OH HC-OH c=o pyruvate
I
I
HO-CH
I
I
HOCH
I 1' H2
HC--OH I HC-OH HC OH HC=O
I I
Ht-OH H 20 Ht -OH HCOH HC-OH
I
I
CH20- 0 CH 2 0 - 0 tH 2 0-0
I
CH 2 0 - 0
g1ucose-6-0 6-0-gluconate 2-~o--3-deoxy 3-0 -glyceral-
6-\fl-gluconate dehyde
(KDPGl
Figure 5.9. Reactions of the Entner-Doudoroff pathway. 1, Glucose-6-phosphate

dehydrogenase; 2, 6-phosphogluconate dehydratase; 3, KDPG aldolase.
The Entner-Doudoroff Pathway 115
HC=O
I
I CH)
I
I 1'H)
2 HC--OH 2C=O -- 2COSCoA
I I
3 HO-CH 3COOH 13 COz
I
4 HC-OH
I
4COOH
I
14 COz
5 HC-OH 5 C=O -- 5COSCoA
I I I
6 CHzOH 6 CH) 6 CH)
Hy=O 'yOOH I 'COz

2 HC-OH 2 C=O -- 2 COSCoA
I I I
3 HO-yH ~ 3 CH) 3 CH J
4 Hy-OH 4 yOOH I 4 COz
5 HC-OH 5 C=O -- 5 COSCoA
I I I
6 CHzOH 6 CH) 6 CH)
Figure 5.10. Origin of the COrcarbon formed in the pyruvate dehydrogenase

reaction after degradation of glucose by either in the Embden-Meyerhof-Parnas
(EMP) or the Entner-Doudoroff (ED) pathway,
ways can be distinguished in the following way: cells of a particular

bacterium are grown on glucose, cell extracts are prepared, and the key
enzymes of the two pathways are assayed, If high levels of the dehydratase
and KDPG aldolase are present and phosphofructokinases are not detect-
able, it is likely that the ED pathway is involved in glucose degradation.
A second method, radiorespirometry, might also be used to determine
whether the EMP or ED pathway exists in a particular organism. The basis
for this is illustrated in Fig. 5.10. It is obvious that both pathways yield
"different" pyruvate from the first 3 carbon atoms of glucose. In one case
(EMP) the carboxyl group originates from carbon 3 of glucose, in the other
(ED) from carbon 1 of glucose. During aerobic oxidation of glucose the
first CO 2 appearing comes from the oxidative decarboxylation of pyruvate;
it thus originates from glucose carbon atoms 3 and 4 (EMP) or 1 and 4
(ED). This difference is detectable by radiorespirometric methods. For
instance, glucose labeled with 14C in C 1 , C3 , or C4 is added to cell
suspensions of Bacillus subtilis and Xanthomonas phaseoli. The CO 2
evolved is collected every hour and its radioactivity is determined. It is
evident from Fig. 5.11 that B. subtilis releases radioactivity faster from C4
and C3 of glucose and X. phaseoli from C 1 and C4; thus, the first bacterium
employs the Embden-Meyerhof-Parnas and the second the Entner-
Doudoroff pathway.
Which of these two pathways is mainly involved in the breakdown of
hexoses by various bacteria is summarized in Table 5.2. As can be seen, the
Entner- Doudoroff pathway is fairly widespread, particularly among
Gram-negative bacteria, while it is not very common in anaerobes. In view
of the fact that anaerobes do not have a respiratory chain and gain their
ATP usually by substrate-level phosphorylation, the EMP pathway is more
economical for them; the EMP pathway yields 2ATP in the conversion
116 5: Metabolic Diversity of Aerohic Heterotrophs
Bacillus subti/is Xanthomonas phaseuli
Figure 5.11. Evolution of radioactive CO 2 from glucose labeled in positions 1, 3, or

4 by cell suspensions of B. subtilis and X. phaseoli. [Data from V. H. Cheldelin,
Metabolic Pathways in Microorganisms. John Wiley and Sons, Inc., New York,
pp. 30-63 (1961) and A. C. Zagallo and C. H. Wang. 1. Bacteriol, 93, 970-975
(1967).]
of 1 glucose into 2 pyruvates; the ED pathway yields only 1ATP per 2

pyruvates.
The ED pathway is also important when compounds such as gluconate,
mannonate, or hexuronates serve as substrates. If E. coli is transferred
from a glucose medium to a gluconate medium, the following enzymes are
formed; gluconokinase, which phosphorylates gluconate in position 6, and
the two key enzymes of the Entner-Doudoroff pathway. Thus, in E. coli
glucose and other hexoses are degraded via the EMP pathway, but
gluconate and related compounds via the ED pathway (Fig. 5.12).
Clearly, the EMP pathway is not very useful for gluconate degradation.
A reaction sequence leading directly from 6-phosphogluconate to
Table 5.2. Pathways involved in sugar degradation by bacteria
microorganism EMP ED
Arthrobacter species +
Azotobacter chroococcum +
Alcaligenes eutrophus +
Bacillus subtilis, B. cereus + other species +
Escherichia coli + other enterobacteria +
Pseudomonas saccharophila,
P. fluorescens + other species +
Rhizobium japonicum + other species +
Thiobacillus intermedius,
Th. ferrooxidans +
Xanthomonas phaseoli + other species +
The Entner-Doudoroff Pathway 117
glucose glueonate
t-l-- PEP
pyruvate t-~ ATP
ADP
glueonokinase
j j ~"
glueose-6-0 &--phosphoglueonate
EMP " ". ., ED . .,
Figure 5.12. Degradation of glucose and gluconate by E. coli.
glucose-6-phosphate is not known, and many bacteria utilize gluconate,

mannonate, glucuronate, and related compounds via the ED pathway.
It should be mentioned in this connection that a number of pseudomo-
nads also degrade glucose via gluconate. P. aeruginosa, P. fluorescens, and
P. putida contain a glucose dehydrogenase, which converts glucose to
gluconate. The latter can be alternatively phosphorylated to 6-phospho-
gluconate or further oxidized to 2-ketogluconate, which is subsequently
phosphorylated and reduced to yield also 6-phosphogluconate (Fig. 5.13).
Glucose and gluconate dehydrogenases are membrane-bound and feed
electrons directly into the respiratory chain. These enzymes allow the
extracellular conversion of glucose into compounds that are less utilizable
by other microbes than glucose. Thus, these enzymes benefit the pseudo-
monads, which can transport and degrade glucose, gluconate, and 2-
ketogluconate. Fructose is metabolized by these organisms via the con-
ventional Entner-Doudoroff route.
glucose 2-ketogluconate
membrane
glucose gilicona te 2-ketoglliconate
~:: ~:::
cytoplasm ~ ATP
~ADP NAD+ NADH + W NADp NADPH.W
CD "'4/. \,,5/
glucose-6- CD
6- P -giliconate' 2-keto-&-- -glueonate
1ED pathway
Figure 5.13. Extracellular oxidation of glucose by pseudomonads and intracellular

conversion of glucose, gluconate, and 2-ketogluconate into 6-phosphogluconate.
1, Hexokinase; 2, gluconate kinase; 3, 2-ketogluconate kinase; 4, glucose-6-
phosphate dehydrogenase; 5, 2-keto-6-phosphogluconate reductase. [Po H.
Whiting, M. Midgley, and E. A. Dawes,J. Gen. Microbial, 92, 304-310 (1976).]
III. Sugar Degradation via the Pentose

Phosphate Cycle
As has been outlined in Chapter 3, E. coli oxidizes part of its substrate
glucose via 6-phosphogluconate to pentose phosphates in order to fulfil the
requirements of the growing cell for NADPH and precursors of nucleotide
biosynthesis. The majority of aerobic microorganisms do the same, as it
can be estimated that about 20% of the glucose being degraded flows into
this pathway. It also has been pointed out that pentose phosphates can be
converted by a series of reversible transaldolase and transketolase reac-
tions into fructose-6-phosphate and 3-phosphoglyceraldehyde. A combina-
tion of these reactions results in a cycle that can bring about the
degradation of sugars (Fig. 5.14). Glucose can be oxidized to 3CO z and
glyceraldehyde-3-phosphate, and the latter can be channeled into the
tricarboxylic acid cycle via pyruvate. Thus, glucose can be oxidized without
participation of either the EMP or the ED pathway. This oxidative pentose
phosphate cycle is found in Thiobacillus novellus and Brucella abortus,
which lack the key enzymes of the EMP and ED pathways but do grow on
glucose.
The oxidative pentose phosphate cycle also can allow aerobic growth on
carbohydrates without participation of the tricarboxylic acid cycle. If
f
2 glucose-6-
3NAD(P)H + 3W
2 fructose-6- 3 gluconate-6-
3NAD(P)H + 3W
3 ribulose-5-
Figure 5.14. Oxidative pentose phosphate cycle. Glucose is oxidized to 3CO z and
glyceraldehyde-3-phosphate. Depending on the enzyme equipment of the organism
the latter may be oxidized further or converted into 1/2 hexose phosphate.
1, Glucose-6-phosphate dehydrogenase; 2, gluconate-6-phosphate dehydrogenase;
3, transaldolase and transketolase reactions; 4, glucose phosphate isomerase.
NAD+(P+) means that the dehydrogenase uses either NAD+ or NADP+ as
coenzymes.
The Methylglyoxal Bypass 119
glyceraldehyde-3-phosphate flows back into the hexose-monophosphate

pool (via fructose-1 ,6-bisphosphate) , the total oxidation of glucose via this
cycle is possible. Alternatively, the C3 unit can be oxidized to acetate,
which is then excreted. This is the situation in some acetic acid bacteria
that lack a complete tricarboxylic acid cycle. (e.g., Gluconobacter species).
Finally it should be mentioned that the nonoxidative reactions of the
pentose phosphate cycle participate in the breakdown of pentoses in many
bacteria as has been outlined in Chapter 4.
IV. The Methylglyoxal Bypass

In 1923 Harden considered methylglyoxal as an intermediate in glucose
catabolism. Following the identification of phosphorylated compounds as
intermediates, the accumulation of methylglyoxal observed under certain
conditions was considered to result from nonenzymatic reactions. In recent
years, however, Cooper and associates discovered a methylglyoxal bypass
which converts dihydroxyacetonephosphate into pyruvate.
This bypass (Fig. 5.15) is present in Escherichia coli, Pseudomonas
saccharophila, and possibly in a number of other bacteria. Its physiological
significance lies in the fact that it makes the formation of acetyl-CoA from
dihydroxyacetonephosphate possible under conditions where low phos-
phate concentrations limit the activity of glyceraldehyde-3-phosphate
dehydrogenase. In accordance with this role methylglyoxal synthase is
inhibited by inorganic phosphate. The methylglyoxal bypass is apparently
fructose-I, 6-bisphosphate
~
HOH 2 C-CO-CH 2 0-() -
___ glyceraldehyde-
3-phosphate
Pi11
dihydroxyacetone phosphate
H)C-CO-CHO
methylglyoxal
H20i 2
H) C-CHOH-COOH
D-Iactate
2H ~3
pyruvate
Figure 5.15. The methylglyoxal bypass 1, Methylglyoxal synthase; 2, glyoxalase I
and II, 3, D-Iactate oxidase (flavin-linked).
also involved in the formation of D-Iactate, if microorganisms such as

Enterobacter aerogenes or Clostridium sphenoides grow anaerobically
with sugars under conditions of phosphate limitation.
V. Diversity in Energy Metabolism

After having seen that there are alternate routes of sugar degradation
present in certain bacteria besides the Embden-Meyerhof-Parnas path-
way, one might then ask whether all aerobic heterotrophs use the same
type of enzyme systems to synthesize ATP. As will be seen, there are
differences here also.
A. Pyruvate dehydrogenase multienzyme complex

Pyruvate dehydrogenase is usually not present in anaerobes. It is a
characteristic enzyme of aerobic heterotrophs, occurring in practically all
aerobic bacteria that catabolize substrates via pyruvate. However, certain
aerobes are very much restricted with respect to the nature of substrates
they are able to utilize. Hyphomicrobium species, for instance, utilize C 1
and C2 compounds only (methanol, methylamine, acetate); they lack
pyruvate dehydrogenase. The same is probably true for other micro-
organisms limited to growth on C 1 compounds.
B. Tricarboxylic acid cycle (TCA cycle)

There is no doubt that the great majority of aerobic heterotrophs use the
TCA cycle in order to provide NADH for the respiratory chain and
precursors for the biosynthesis of cellular constituents. So the cycle
commonly fulfills a dual function. In some aerobes its function to oxidize
the acetyl moiety of acetyl-CoA to CO 2 is abandoned, but the synthesis
of precursors for the biosynthesis of cellular precursors via cycle inter-
mediates is still possible. These organisms lack a complete cycle, often only
one enzyme is missing. This is, for instance, true for bacteria that are
limited to growth on C 1 compounds (see Chapter 6). These organisms are
devoid of a-oxoglutarate dehydrogenase. The cycle is thus interrupted but
glutamate can still be synthesized via citrate, isocitrate, and a-
oxoglutarate. Succinyl-CoA and compounds derived thereof (e.g., L-
lysine, L-methionine, porphyrin skeleton) are formed by these organisms
from oxaloacetate via L-malate fumarate and succinate. Whereas the
reduction of oxaloacetate to L-malate and the dehydration of the latter to
fumarate are catalyzed by the corresponding TCA cycle enzymes, malate
dehydrogenase and fumarase, a special enzyme is employed for succinate
formation: fumarate reductase. It uses NADH as hydrogen donor and thus
is different from succinate dehydrogenase. Finally, succinyl-CoA is formed
Diversity in Energy Metabolism 121
by succinate thiokinase. A number of acetic acid bacteria (Chapter 6) lack

succinate thiokinase. Therefore they synthesize a-oxoglutarate and suc-
cinyl-CoA via citrate and succinate (if needed) via L-malate.
C. Respiratory chain
Although the possession of a respiratory chain is a characteristic feature of
all aerobes, differences have been encountered when the chains of a
number of bacteria were analyzed for their components. This is in contrast
to the respiratory chain of mitochondria, which does not show species-
specific variations in its general composition. A few examples are given in
Fig. 5.16. It can be seen that the chain of E. coli, which lacks a cytochrome
of the c type, is not characteristic of all bacteria. The chains of Paracoccus
denitrificans and Micrococcus luteus contain in addition cytochromes of
the c and a3 type. Practically all bacterial respiratory chains are branched.
Two branch points have been observed: at the coenzyme Q level and at the
cytochrome c level (Fig. 5.16). It is not yet certain that the P: 0 ratio is the
same in all bacteria. Experimental data indicate that the chain of E. coli
contains two phosphorylation sites, but those of P. denitrificans and M.
luteus have three (at least the branch via cytochrome a, a3)'
substrate substrate su bSlrate
t
NADH
1
NADH
1
NADH
1
flavoprotein
t
flavoprotein
1
flavoprotein
1 1 t
7\ 7\
menaquinone
!
' 'r "T' "'j"' "'t cyt b
!
'T T "r' T
cyt c
/\
Escherichia coli
cyt~a3
!
ParacoccliS
O2
"r T
Micrococcus
denirri[icans lufeus
Figure 5.16. Components of the respiratory chain of three bacterial species.

A remarkable property of the electron transport system of many

bacteria is that under certain conditions nitrate can be used as H-acceptor
instead of oxygen. This will be discussed in the next section.
VI. Dissimilatory Reduction of Nitrate

There are a number of species among the aerobic bacteria that strictly
depend on oxygen for their energy metabolism. Examples are members of
the genera Azotobacter, Arthrobacter, Acetobacter, and Gluconobacter.
Three groups of bacteria, however, are able to make a living under aerobic
as well as under anaerobic conditions:
1. Several species including the enterobacteria are facultative aerobes,
meaning that they gain energy by an oxygen-dependent respiration
under aerobic conditions or by a fermentative metabolism under
anaerobic conditions (see Chapter 8).
2. Some phototrophs (e.g., the Rhodospirillaceae) switch over to oxygen-
coupled respiration if transferred from an anaerobic to an aerobic
environment (see Chapter 9).
3. A third group of organisms is formed by those species that can replace
oxygen as terminal electron acceptor by nitrate. These organisms are
able to carry out an anaerobic respiration. Depending on the products
formed from nitrate, different processes of dissimilatory nitrate reduc-
tion can be envisaged.
A. Denitrification
If molecular nitrogen is the major product of dissimilatory nitrate reduc-
tion, the process is called denitrification. The energetics of this anaerobic
respiration compares well with aerobic respiration as is apparent from
Eqs. 5.1 and 5.2.
glucose + 60 2 ~ 6C0 2 + 6H 2 0 (5.1)
6.Go, = -2870 kJ (-686 kcal)
glucose + 4.8N0 3 + 4.8H+ ~ 6C0 2 + 2.4N 2 + 8.4H 2 0 (5.2)
6.Go, = -2669 kJ (-638 kcal)
Organisms capable of carrying out a denitrification encompass species of
various genera. Some of them are listed in Table 5.3. Many bacilli and
pseudomonads in particular are denitrifiers.
The enzyme machinery for denitrification is formed only under anaero-
bic conditions or conditions of low oxygen tension. In most cases nitrate is
required as inducer. Also, the activity of the enzymes involved in nitrate
reduction to N 2 is strongly inhibited by oxygen. Thus, denitrification takes
place only when oxygen is absent or available in insufficient amounts. All
bacteria capable of performing this kind of respiration prefer oxygen
Dissimilatory Reduction of Nitrate 123
Table 5.3. Denitrifying bacteria
Alcaligenes faecalis
Bacillus licheniformis
Hyphomicrobium vulgare
Paracoccus denitrificans
Pseudomonas stutzeri
Spirillum itersonii
Thiobacillus denitrificans
respiration if possible. It should be mentioned in this connection that some

denitrifiers are microaerophilic; they tolerate only low oxygen tensions
(p02 below 2%). An example is the obligate sulfur oxidizer (see Chapter
9) Thiomicrospira denitrificans.
Like oxygen respiration, denitrification allows a complete oxidation of
the organic substrate to CO 2 (see Eq. 5.2). For instance, if Bacillus
licheniformis grows with glucose and nitrate under anaerobic conditions,
the substrate is degraded via the Embden-Meyerhof-Parnas pathway; the
acetyl-CoA formed from pyruvate is oxidized via the tricarboxylic acid
cycle, and NADH and FADH 2 thus formed serve as electron donors for
the respiratory chain. Nitrate, however, does not simply replace oxygen;
special types of cytochromes and membrane-bound enzyme systems are
formed, which reduce nitrate to nitrite and further to nitrogen. The
electron flow in denitrification is illustrated in Fig. 5.17. Three reduction
steps can be distinguished in denitrification. They are depicted in Fig. 5.18
in greater detail.
Nitrate reductase is a molybdenum-containing membrane-bound en-
zyme which reduces nitrate to nitrite. Molybdenum is present in this
enzyme in the form of the so-called molybdenum cofactor (Moco) in which
molybdenum is bound to a pterin moiety, the so-called molybdopterin.
Moco is noncovalently bound to the protein and its function is that of a
prosthetic group. It is interesting that this cofactor occurs in all molybdo-
r
proteins except the nitrogenase in which one type of subunit is a molybde-
num-iron-sulfur-protein (see Chapter 10). Besides nitrate reductase,
cyt 0
NADH~ flavoproteins
( FeS-proteins
ubiquinone
b 562 _
565
.~
nitrate -----..
Figure 5.17. Electron flow in denitrification. As an example the scheme for

Paracoccus denitrificans is shown. [A. H. Stouthamer, F. C. Boogerd, and
H. W. van Verseveld. Antonie van Leeuwenhoek 48, 545-553 (1982)].
r------- ------------~
:[ 2H+ I 2H+ X-Fe 2+

NO:; -\.~I--\..-"'--"""-". NO; I\.\. .. X-Fe2+ _ 50+
H20 :I "H 0
2
I
I
I
I
I
I
I
I
I
I------------..j
I
I
I
X-Fe2+-N~O
I
-O-N=O
I
2H+ 2e- ~
X-Fe 2+
4H' 2e-
2<-
i I
I
I
1:-" I I X-Fe 2+-N-O- :
f
N ~f--.,......LI-_I,,--- N20
3 f -o-~-o-
L__ ~~~ JI
2 I
H 20
Sum: 2NO:; + 10e- + 12H' -----i.~ N 2 + 6H 20
Figure 5.18. Tentative scheme of the reactions involved in the reduction of nitrate
to molecular nitrogen. 1, Nitrate reductase; 2, nitrite reductase; 3, nitrous oxide
reductase. The nitrite reductase reaction is formulated according to Averill and
Tiedje, FEBS Lett. 138, 8-12 (1982).
molybdenum cofactor was detected in formate dehydrogenase, sulfite

oxidase, xanthine dehydrogenase, trimethylamine-N-oxide reductase, and
CO oxidase.
Nitrite reductase. It is obvious from Fig. 5.18 that the reduction of
nitrite to N 2 0 is very complex. It comprises two 2-electron reduction steps.
First an enzyme-ferrous-nitrosyl complex is formed which a second NOi is
added to. Reduction yields coordinately bound oxyhyponitrate that is
reduced further to N2 0. Nitrite reductase is membrane-bound; it contains
haem of the c- and of the d-type. Some nitrite reductases contain copper.
Nitrous oxide reductase is also membrane-bound and linked to the respira-
tion chain via a cytochrome of the c-type.
It should be noted that all reactions in Fig. 5.18 are 2-electron reactions
and that NO is not an intermediate. Small amounts of NO are, however,
formed by denitrifiers, probably by decomposition of one of the inter-
mediates of the nitrite reductase reaction.
Denitrifying bacteria can use nitrate-and many of them also nitrite or
N2 0 as electron acceptors. Evidence has been obtained that electron flow
to all three compounds is coupled to ATP formation.
Dissimilatory Reduction of Nitrate 125
B. Nitrate/nitrite respiration and

nitrate/ammonia respiration
A number of bacteria, for instance most enterobacteria and staphylococci,
perform a different mode of dissimilatory nitrate reduction: Nitrate is
reduced only to the level of nitrite and the latter is excreted (nitrate / nitrite
respiration). Eq. 5.3 gives the stoichiometry of glucose oxidation coupled
to the reduction of nitrate to nitrite.
glucose + l2NO;- ~ 6CO z + 6H zO + 12NOi (5.3)
AGo, = -1766 kJ (-422 kcal)
As in denitrification a membrane-bound nitrate reductase is involved

here. Several organisms do not excrete nitrite but reduce it further to
ammonia (Eq. 5.4).
glucose + 3NO;- + 3H+ ~ 6CO z + 3NH 3 + 3H zO (5.4)
AGo, = -1796 kJ (-429 kcal)
There are indications that the reduction of nitrite to ammonia is also

coupled to ATP synthesis by a chemiosmotic mechanism. In this case the
enzyme system reducing nitrite to ammonia must be linked to the
respiratory chain.
What has been said so far relates to aerobic and facultative aerobic
bacteria. A dissimilatory reduction of nitrate to ammonia, however, has
also been observed in several obligate anaerobic bacteria, e.g., in Clostri-
dium perfringens, Veillonella alcalescens, and propionibacteria.
Here, nitrate is used as a "hydrogen sink"; NAOH is consumed, and
NAO+ is generated for the further fermentative breakdown of substrates.
In most cases, nitrate reduction is not associated with ATP formation by a
chemiosmotic mechanism. This process may be called fermentative nitrate
reduction.
In this connection it should be mentioned that a number of nondenitrify-
ing bacteria are also able to reduce nitrate to ammonia by soluble enzyme
systems including a soluble nitrate reductase. These systems are not
oxygen-sensitive and are not coupled to ATP-yielding reactions. This
process is called assimilatory nitrate reduction, and it has the purpose of
supplying the cell with ammonia for the biosynthesis of cellular material. It
is present in all bacteria that grow with nitrate as nitrogen source. In
accordance with this, the formation of the assimilatory enzymes is repres-
sed by ammonia.
C. Related processes
Enterobacteria and phototrophs such as Rhodopseudomonas capsulata
have been shown to use trimethylamine-N-oxide in a similar fashion as
nitrate. It is reduced via an electron transport chain to trimethylamine.

(CH3h-N~O + 2H+ + 2e- ~ (CH3h-N + H 2 0 (5.5)
The N-oxide is present in marine fish in substantial amounts, and it allows
an anaerobic respiration in decaying fish.
Processes such as methane formation from CO 2 and H 2 S formation
from SO~- also can be considered as anaerobic respirations.
Since the organisms carrying out these processes are obligate anaerobes
and have so much in common with other obligate anaerobes in regard to
their physiology, these processes will be discussed together with other
fermentations (see Chapter 8).
VII. Bacterial Bioluminescence

Luminous bacteria are found within two genera: Vibrio and Photobac-
terium. Members of the first genus (e.g., V. fischeri) emit light only as
free-living organisms whereas members of the latter, e.g., P. phosphoreum
function also as symbionts in the light organs of fish. Light emission is
usually observed only in dense cultures. This is explained on the basis that
NAD+ NADH + H+
"---~_ _./oc::::.- FMN _ - - - - - _ .
(
_--------L .....------.. . .
R-CHO
~
----~-- L'FMNH1'Ol'RCHO - - - - _.......)
Figure 5.19. Reactions involved in bacterial bioluminescence. L, Bacterial

luciferase [Modified from K. H. Nealson. Trends Biochem. Sci. 4, 105-110
(1979)).
Alternate Anaplerotic Sequences 127
cells excrete a small molecule and that the synthesis of the components of
the bioluminescent reaction is induced only above a threshold concentra-
tion of this molecule in the environment of the bacteria.
The biochemical mechanism of light emission by bacteria is different
from the one in fireflies or in those luminous fish that have photophores
and produce their own light. In addition to the enzyme called bacterial
luciferase, FMNH2, 02, and a long-chain aliphatic aldehyde such as
hexadecanal (palmitaldehyde) are required. Bacterial bioluminescence is
strictly 02-dependent. The reactions involved in light emission are de-
picted in Fig. 5.19. Substrate oxidation via the tricarboxylic acid cycle
yields NADH that is used to reduce FMN. Bacterialluciferase then forms a
complex with all the components involved. An excited state of this
complex decays and, coupled to oxidation reactions, light is emitted.
Bacterial bioluminescence is a very interesting process. Its advantage for
those organisms that live in symbiosis with fish is apparent; the advantage
is less apparent for free-living bacteria.
VIII. Alternate Anaplerotic Sequences

We have seen that the kind of anaplerotic sequence employed by E. coli
depends on the growth substrate; on glucose, oxaloacetate is synthesized
from PEP with PEP carboxylase; on acetate, the glyoxylate cycle is
induced and PEP is formed with PEP carboxykinase; on pyruvate, the
enzyme PEP synthetase is responsible for the provision of PEP for
gluconeogenesis. Although not many bacteria have been studied in such
detail as E. coli, the experimental results show that alternate anaplerotic
sequences occur among bacteria also.
For substrates that are catabolized via PEP (glucose, other carbohy-
drates) the following two anaplerotic sequences are widespread among
bacteria.
A. PEP carboxylase
PEP + HCO) ~ oxaloacetate + Pi
This enzyme is responsible for the replenishment of oxaloacetate in
enterobacteria, Bacillus anthracis, Acetobacter xylinum, Thiobacillus
no vellus , and Azotobacter vinelandii.
B. Pyruvate carboxylase
pyruvate + HCO) + ATP. ) oxaloacetate + ADP + Pi
Like mammalian systems a number of bacteria employ pyruvate car-
boxylase for oxaloacetate synthesis. This enzyme contains biotin, and the
latter is directly involved in this carboxylation reaction:

acetylCoA
Mg'< )
enzyme-biotin + ATP + HC0 3 (
enzyme-biotin-C02" + ADP + Pi
enzyme-biotin-C02" + pyruvate ( ) enzyme-biotin + oxaloacetate-
enzyme-biotin -C0 2
Pyruvate carboxylase from mammalian sources requires acetyl-CoA for

maximal activity. With the exception of Pseudomonas citronellolis and P.
aeruginosa, this is also true for the enzyme from a number of bacteria
including Arthrobacter globiformis, Bacillus coagulans, Acinetobacter cal-
coaceticus, and Rhodopseudomonas sphaeroides.
Bacteria which grow on acetate or on compounds channeled into the
intermediary metabolism via acetyl-CoA (butyrate, aliphatic hydro-
carbons) usually employ the glyoxylate cycle to form C4 -dicarboxylic acids
and PEP carboxykinase to synthesize PEP for gluconeogenesis. Certain
phototrophic bacteria (Rhodospirillum rubrum, Rhodopseudomonas
sphaeroides), however, contain malate synthase during aerobic growth on
acetate but lack isocitrate lyase. It is not yet known how these micro-
organisms form glyoxylate.
Many bacteria are able to grow on pyruvate and lactate. Those that
contain pyruvate carboxylase do not have problems with respect to
oxaloacetate synthesis. The others-as we have already seen for Esche-
richia coli-may use PEP syntetase to form PEP from pyruvate and PEP
carboxylase to replenish the oxaloacetate pool. Not much is known about
the distribution of the synthetase. It is present in E. coli and Salmonella
typhimurium and presumably in other enterobacteria. Acetobacter xyli-
num and the anaerobe Propionibacterium shermanii contain another
enzyme that also allows the direct synthesis of PEP from pyruvate:
pyruvate-phosphate dikinase. The reactions catalyzed by PEP synthetase
and the dikinase are given in the following equations:
PEP synthetase
pyruvate + ATP ( ) PEP + AMP + P j
pyruvate-phosphate dikinase
pyruvate + ATP + P j ( ) PEP + AMP + PP j
The differences are apparent. If pyrophosphorylated enzymes are
intermediates, the PEP synthetase releases one of these phosphate groups
Biosynthesis of Monomers and Polymers 129
Table 5.4. Kind of enzyme systems involved in oxaloacetate and PEP synthesis,
as dependent on the nature of the growth substrate
synthesis of
catabolic sequence oxaloacetate PEP
substrate PEP carboxylase

!
PEP or
!
pyruvate pyruvate carboxylase
!
acetyl-CoA
substrate pyruvate carboxylase PEP carboxykinase

!
pyruvate or
!
acetyl-CoA PEP carboxylase PEP synthetase
or
pyruvate-phosphate dikinase
substrate glyoxylate cycle PEP carboxykinase

!
acetyl-CoA
as phosphate and transfers the second to pyruvate; in the dikinase reaction

one of the phosphate groups is transferred to phosphate to yield pyro-
phosphate and the second to pyruvate.
Table 5.4 summarizes the function of the enzyme systems discussed in
this section.
IX. Biosynthesis of Monomers and

Polymers
The pathways used by microorganisms to synthesize amino acids, purine
and pyrimidine bases, the various lipids, and carbohydrates are usually the
same as those outlined for Escherichia coli. There is not much diversity as
to the nature of the intermediates involved in the synthesis of these
compounds. However, this does not mean that the enzyme catalyzing the
same type of reaction in different bacteria are identical. For example,
chorismate mutase and aspartate kinase of E. coli are different from the
corresponding enzymes of Bacillus subtilis. Their molecular weight and
chemical and physical properties are different, but in both cases the
enzymes are indeed either chorismate mutase or aspartate kinase. The

same is true for many enzymes of different bacteria. It is noteworthy that
enzymes specific for the same reaction very often have regulatory prop-
erties that differ from one bacterial species to the other. This will be
discussed in Chapter 7.
We can therefore conclude that there is a relatively constant set of
anabolic reactions in bacteria, on the one hand, and much variation in the
structure and the properties of the enzyme systems involved, on the other.
Many bacteria are, however, devoid of enzyme systems required for the
synthesis of one or more precursors of polymers. E. coli, some other
enterobacteria, Bacillus megaterium, and Azotobacter vinelandii-to men-
tion just a few-are able to grow with glucose and minerals, that is, they
contain the entire enzymatic machinery necessary to make all compounds
required for growth. A number of bacilli and pseudomonads depend for
growth on the presence of certain vitamins and amino acids in their
environment. In the laboratory they are grown in media supplemented
with peptone and yeast or beef extract as sources of these compounds.
Particularly dependent on preformed monomers are bacteria that normally
grow on complex organic material (juice, milk, decaying plants, etc.).
The anaerobic lactic acid bacteria, for instance, are so poorly equipped
with anabolic enzyme systems that they require practically all the mono-
mers for growth.
As far as is known, the synthesis of informational macromolecules
(DNA, RNA, and proteins) proceeds in all bacteria in the same fashion.
This is also true for periodic macromolecules, although one has to consider
here that not all bacteria form periodic macromolecules of the same
composition and structure.
A. Reserve materials
Bacteria differ in the type of reserve material they accumulate under
certain conditions. Table 5.5. lists a small selection of bacteria that
accumulate one or more of the typical storage materials of micro-
organisms: glycogen, polY-I3-hydroxybutyrate, or polyphosphate. There
seem to be some microorganisms that do not form any reserve material;
one of these is Pseudomonas aeruginosa.
Polyphosphate accumulation is widespread among bacteria. It functions
as a phosphorus storage material and is utilized for nucleic acid and
phospholipid synthesis under conditions of phosphorus starvation.
Glycogen and poly-jJ-hydroxybutyrate serve as energy storage com-
pounds. In the absence of an exogenous energy source they are slowly
degraded and supply the cells with ATP for maintenance metabolism. The
enzymes involved in glycogen synthesis and degradation were described in
Chapter 3.
Table 5.5. Occurrence of glycogen, poly-{3-hydroxybutyrate (PHB)

and polyphosphate in bacteria
organism glycogen PHB polyphosphate
Enterobacter aerogenes + +
Alcaligenes eutrophus + +
Azotobacter vinelandii + +
Bacillus megaterium + +
Escherichia coli + +
Mycobacterium phlei + +
Pseudomonas aeruginosa
Pseudomonas multivorans + ?
Rhodospirillum rub rum -t- + ?
Sphaerotilus natans + ?
PolY-I3-hydroxybutyrate (PHB) is a typical prokaryotic storage mate-

rial. It is widespread in bacilli, in chemolithotrophic and phototrophic
bacteria, and in pseudomonads (but does not occur in the fluorescent
group). PHB is a polymer of D( - )-I3-hydroxybutyrate and has a molecular
weight between 60,000 and 250,000.
HO-CH-H C-C-[-O-CH-CH -C-J-O-CH-CH2-COOH
I 2 II I 2 II I
CH 3 0 CH 3 0 n CH 3
poIY-J3-hydroxybutyric acid
It is accumulated in the cells as granules surrounded by membranes. Under

appropriate conditions (no nitrogen source but plenty of energy and
organic substrates) bacterial cells may accumulate so much PHB that it
accounts for approximately 60% of their dry weight.
Synthesis of PHB in Azotobacter beijerinckii proceeds as shown in
Figure 5.20. The polymerase appears to be granule-bound. An alternate
route is employed by Rhodospirillum rubrum to synthesize the D( - )-
monomer (also Fig. 5.20). This microorganism contains a L( + )-13-
hydroxybutyryl-CoA dehydrogenase but in addition two crotonases that
bring about the conversion of the L( +) form into the D( -) form.
Utilization of PHB starts when an exogenous energy source is no
longer available to the cells. A depolymerase then releases D( - )-13-
hydroxybutyrate from the granules, which is subsequently oxidized to
acetoacetate by a NAD+ -specific D( - )-I3-hydroxybutyrate dehydro-
genase. Acetoacetate is then channeled into the intermediary metabolism
by a CoA transferase reaction (Fig. 5.21). Thus, CoA esters are the
substrates for PHB synthesis and acids are the products of PHB degrada-
tion; this separation facilitates the regulation of both processes.
2CH 3-CO-SCoA
, ~HSCOA NADH+W NAD'
CH 3-CO-CH 2 -CO-SCoA :;;"i==="--==~L==.!!::: CH -yH -

3 CH 2-CO-SCoA
tr- 4 OH
t
2 NADPH + W
OH
NADP
'
H2 0
5~H'O
I ./
CH 3 -CH- CH 2 - CO-SCoA ::!I..t::===~==1.j;: CH 3- CH=CH-CO-SCoA
6
3tHSCOA
PHB
Figure 5.20. Synthesis of PHB in Azotobacter beijerinckii and in Rhodospirillum
rubrum. 1, f3-Ketothiolase; 2, D(-)-f3-hydroxybutyryl-CoA dehydrogenase;
3, D(-)-f3-hydroxybutyryl-CoA polymerase; 4, L( + )-f3-hydroxybutyryl-CoA de-
hydrogenase; 5, crotonase I; 6, crotonase II. A. beijerinckii, reactions 1, 2, 3;
R. rubrum, reactions 1, 4, 5, 6, 3.
PHB o(-)-j3-hydroxybulyrale
~NAD'
2 t - NADH + H'
3
acetoacetat? ~cetoacetyi-eoA
succinyi-CoA succinate
Figure 5.21. Degradation of PHB. 1, Depolymerase; 2, D( - )-f3-hydroxybutyrate

dehydrogenase; 3, CoA transferase.
B. Cell wall components

In a discussion of the diversity of pathways involved in periodic
macromolecule biosynthesis, it is also necessary to consider that bacteria
vary with respect to the composition of their cell walls. This implies, of
course, that different enzyme systems are involved in the formation of
these structures in different bacteria.
The synthesis of the layers of E. coli cell wall has been outlined in
Chapter 3. Approximately 10% of the wall of E. coli and of other
Gram-negative bacteria consists of peptidoglycan. The investigation of the
walls of a fairly large number of Gram-negative bacteria has shown that the
composition of their peptidoglycan (murein) layer is very similar. They
contain N-acetylmuramic acid, N-acetylglucosamine, and a tetrapeptide
consisting of L-alanine, D-glutamate, meso-diaminopimelate and D-alanine
(except the spirochetes, see below). Cross-linkage of the peptide chains

usually occurs between the amino group of diaminopimelate and o-alanine
(see Fig. 3.32).
The peptidoglycan layer of Gram-negative bacteria is usually mono-
molecular; only the one of the cyanobacteria is rather thick and multi-
layered. Variations of the composition of the tetrapeptide occur in a few
species. Many spirochetes contain ornithine instead of meso-
diaminopimelate. In Treponena pallidum cross-linkage between ornithine
and o-alanine is not direct but via a glycine residue.
The outer membrane of Gram-negative bacteria usually consists of the
three components lipid A, core saccharide and repeating oligosaccharide
chains (see Fig. 3.33). Lipid A is the most conservative part. Variations
occur in some phototrophic bacteria in which o-glucosamine is replacect by
2,3-diamino-2,3-dideoxy-o-glucose. Although aldoheptose and 2-keto-3-
deoxy-octonate (KDO, dOclA) are characteristic components of the core
region, they are absent in some outer membranes, e.g., in the one of
Bacteroides species, of Anabaena and of Thermus aquaticus. Only KDO is
lacking in Vibrio cholerae and only aldoheptose in Rhodopseudomonas
capsulata, Acinetobacter calcoaceticus, and in most cyanobacteria. The
composition of the repeating oligosaccharide chains is most variable. These
chains represent the determinants of 0 antigenic specificity and are the basis
for serological classification. The chains consist of up to 30 repeating
oligosaccharides, that contain hexoses, 6-deoxy- or 3,6-dideoxyhexoses.
O-Methylated sugars occur in the outer membrane of phototrophs and of
cyanobacteria. Three to six sugar molecules form one repeating unit.
Gram-positive bacteria contain peptidoglycan as the major component
of the cell wall. In many of these microorganisms it accounts for 80 to 90%
of the cell wall components. As in the peptidoglycan of Gram-negative
bacteria, its backbone consists of alternating sequences of N-acetyl-
muramic acid and N-acetylglucosamine.
However, studies of Kandler, Schleifer, and others have shown that the
peptide moiety is subject to considerable variation:
1. Cross-linkage between adjacent peptides extends from the distal amino

group of the diamino acid in position 3 of one peptide to the carboxyl
group of o-alanine in position 4 of the other peptide (group A
peptidoglycans) or from the a-carboxyl group of o-glutamate in
position 2 to the carboxyl group of o-alanine (group B peptidoglycans).
A diamino acid is required as interpeptide bridge in group B peptido-
glycan. The two groups of peptidoglycans are depicted in Fig. 5.22.
Group A peptidoglycan is very common among Gram-negative as well
as Gram-positive bacteria, whereas group B occurs in microbacteria,
coryneform bacteria, and in Eubacterium limosum.
2. The amino acids L-alanine and meso-diaminopimelate, present in
positions 1 and 3 are frequently exchanged against others. Examples are
.....
.J>.
'"
G M G M
CH20H 0 fJ(I-4) CH 20H CH20H 0 fJ(I-4) CH 20H
" H H 0 H H
0_ 0-..
o OH H H -"0 OH H H 0
H NHAc H NHAc H NHAc

~ ~
CH 3 -C-H CH -C-H
3 ~O
Jo
t
Ala +
Gly <J>
l t c> w
D-Glu-(NH 2) D-Glll~D-Lys"D-Ala ~
t 0;
Pw ,1' cr'
L-DA.;o.(l)_D-Ala L-AA L-AA :2-
ri'
t +'Y
D-At. L-!A D-Ala D-Glu o
+ <'
<1>
D-Gt'-(NH 2) Ala fa.
Ala +
-G-M-
-<
t g,
-G-M- ;I-
<1>
A-type B-type <3
cr'
ri'
Figure 5.22. The two types of peptidoglycan found in Gram-positive bacteria. (NH z) means that the +-carboxyl groups of the D-glutamate
moieties are partly amidated; (I) means that cross-linkage may be by interpeptide bridges. L-DA, L-diamino acid; L-AA, L-amino acid.
::r:
<1>
;;
S
<3
"0
a-
Table 5.6. Variations of the peptidoglycan of gram-positive

bacteria
amino acid in cross-linkage example

position
1 3
position 3 to 4 (group A)
L-Ala m-A 2 pm direct bacilli, clostridia
L-Ala L-Lys (Gly)s Staphylococcus
aureus
L-Ala L-Lys (L-Alah Streptococcus
thermophilus
L-Ala L-Lys (L-Alah-Thr Arthrobacter
citreus
L-Ala LL-A 2 pm Gly most streptomyces
L-Ala L-Lys o-Asp most lactobacilli
L-Ala L-Orn o-Lys Actinomyces israeli
position 2 to 4 (group B)
Gly L-Lys L-Lys Microbacterium
lacticum
L-Ser L-Orn o-Lys Eubacterium
limosum
Gly L-Hsr o-Orn Corynebacterium
poinsettiae
[Modified from K. H. Schleifer and E. Stackebrandt. Ann. Rev. Microbial.

37, 143-187 (1983) reproduced with permission].
-Azpm, meso-diaminopimelate; Hsr, homoserine.
presented in Table 5.6. Note that position 4 is always occupied by

o-alanine. The a-carboxyl group of o-glutamate in position 2 may be
amidated.
3. Cross-linkage between the peptides is accomplished either directly as in
Gram-negative organisms or via interpeptide bridges. Interpeptide
bridges consist of up to eight amino acids; glycine, L-alanine, 0-
aspartate, L- or o-lysine, or o-ornithine may be present. In group B
peptidoglycans an interpeptide bridge with a diamino acid has to be
present.
It is apparent that there is a great variability within the group of

Gram-positive bacteria as to the composition of their peptidoglycans.
The cell wall of Gram-positive bacteria does not contain components
comparable to the outer membrane fraction of the Gram-negative bacte-
ria. However, in addition to the peptidoglycan layer the wall contains
various proteins, polysaccharides and/or the so-called teichoic acids,
which account for about 10-20% of its dry weight. Teichoic acids are
polymers of glycerol or ribitol phosphate substituted with various sugars
and with o-alanine.
HO-CH z O-CH z O-CH z
H-t-O Ala O / H - t - O Ala O/H-t-O Ala
I II I II I
H-C-OH P H-C-OH P H-C-OH
H-{-osug/6- H - { - 0 76 - H-{-07
CHzJ CHzO CHzO
l.5-poly(ribitol phosphale)teichoic acid
They are covalently bound to the peptidoglycan layer by phosphodiester

bridges. One function of the teichoic acids is to maintain a high concentra-
tion of divalent cations in the vicinity of the cells. In addition to the cell
wall teichoic acids, many Gram-positive bacteria contain Iipoteichoic acids
which are associated with the membrane; they are of the glycerol phos-
phate type.
C. Wall components and other interesting

components of archaebacteria
On the basis of 16 S ribosomal RNA homology studies, it was concluded
that a few groups of bacteria are genealogically only distantly related to the
eubacteria. For these organisms the name archaebacteria was introduced
by Woese. The archaebacteria comprise all known methanogenic bacteria,
the extremely halophilic organisms such as Halobacterium or Halococcus,
thermoacidophilic organisms such as Sulfolobus or Thermoplasma and the
extremely thermophilic organisms Thermoproteus and Pyrodictium. It is a
characteristic feature of the archaebacteria that they lack peptidoglycan.
They are, therefore, resistant to ,B-lactam antibiotics. Cell wall structures
of some of the archaebacteria are summarized in Table 5.7. An interesting
structure is the pseudomurein in Methanobacterium and in Methanobrevi-
bacter. It contains L-talosaminuronic acid instead of muramic acid
(Fig. 5.23); the peptides are made of L- and not of o-amino acids.
It should also be mentioned that the membranes of archaebacteria
contain a characteristic lipid, biphytanyldiglycerol tetraether glycolipid
which is unique for these organisms (Fig. 5.24). In Methanospirillum
hungatei it accounts for 35% of the membrane, other components being
proteins and carbohydrates. Differences as compared to eubacteria have
also been encountered in the transcriptional machinery. The RNA
polymerase consists of 8-10 subunits but lacks a u factor.
Archaebacteria are still a promising source to search for new meta-
bolically active compounds. Recently, 2,3-cyclopyrophosphoglycerate
Table 5.7. Cell wall structures in archae bacteria
organism sacculus a envelope polymer
Methanobacterium + pseudomurein
Methanobrevibacter + pseudomurein
Methanosarcina + heteropolysaccharide
Methanococcus + glycoprotein
Methanogenium + glycoprotein
Methanomicrobium + glycoprotein
Methanospirillum + protein
Halococcus + sulfatized
heteropolysaccharide
Halobacterium + glycoprotein
Sulfolobus + glycoprotein
Thermoplasma none
Pyrodictium + glycoprotein
[0. Kandler, Naturwiss. 68, 183-192 (1981)].

aA sacculus is rigid, an envelope flexible
(cyclic 2,3-diphosphoglycerate) (Fig. 5.24) was isolated from a metha-

nogenic bacterium; this compound may function as a storage material for
phosphate .. The unique coenzymes involved in methanogenesis will be
discussed in Chapter 8.
o-GlcNAc L-NAcTaINU
H H
H
/co
<"'>":+---0-
(3-1)13
Glu-(NH 2 )
t'Y
Ala
t<
Lys_Glu
t t 'Y
.<
(Glu) Lys-Glu
b
(Ala) Ala
t'Y
Glu-(NH 2 )
-o-GlcNAc-- L-NAcTaINU--
/
Figure 5.23. Structure of pseudomurein. D-GIcNAc, N-acetyl-D-glucosamine;
L-NAcTaINU, N-acetyl-L-talosaminuronic acid.
IH2-0~ coo-
I ~/O-
HC-O-P
TH-O~
CH 20H
I
H (-0-1'
)0
2 R' 0 -
o
diphytanylglycerol diether 2,3-cyclopyrophospho-
g]ycerate (cyclic 2, 3-
bisphosphoglycerate)
dibiphytanyldiglycerol tetraether
Figure 5.24. Novel compounds found in archaebacteria. Phytanyl is C2o H 41 - and

biphytanyl is C40 H so .
X. Summary
1. When comparing the mode in which substances from the environ-
ment cross the cell membrane into the cytoplasm, four different mecha-
nisms can be distinguished: passive diffusion, facilitated diffusion, active
transport, and group translocation. The latter two processes are important
for the uptake of nutrients by bacteria.
Active transport allows transport against a concentration gradient. The
energy required is provided either by the protonmotive force or by ATP.
The process of group translocation is employed by anaerobes or facultative
anaerobes to transport sugars into the cells. The sugars are trapped by
phosphorylation.
Many bacteria excrete siderophores that complex Fe 3 +. The complexes
are taken up and serve as iron source. The chemotactic responses of
bacteria are regulated by the methylation grade of special membrane
proteins. Signals from these proteins to the flagellar motor are mediated by
cyclic GMP. Transport of nutrients, through the outer membrane of
Gram-negative bacteria proceeds via pores that are made up of trimers of a
protein called porin. Transport of proteins into membranes occurs either
directly from membrane-attached ribosomes to their destination or with
the aid of "membrane soluble" leader peptides.
2. Pseudomonads and many other Gram-negative bacteria employ the
Entner-Doudoroff pathway for the breakdown of hexoses. The key
enzymes of this pathway are 6-phosphogluconate dehydratase and
2-keto-3-deoxy-6-phosphogluconate (KDPG) aldolase. The conversion of
Summary 139
glucose into 2 pyruvate via the ED pathway yields lATP whereas 2ATP
are formed in the EM pathway. The ED pathway is also important for
bacteria growing with gluconate, mannonate, and hexuronates.
3. In the oxidative pentose phosphate cycle, glucose is oxidized to 3CO z
and glyceraldehyde-3-phosphate. Subsequent oxidation of the latter in the
tricarboxylic acid cycle results in the breakdown of glucose without
participation of the ED and EMP pathways.
4. At low concentrations of phosphate, a number of bacteria employ
the methylglyoxal bypass to form acetyl-CoA from dihydroxyacetone-
phosphate.
5. The tricarboxylic acid cycle is employed by most aerobic hetero-
trophs for the generation of reducing power. Organisms limited to the
metabolism of C 1 compounds and some acetic acid bacteria do not contain
a complete cycle; they are devoid of a-oxoglutarate dehydrogenase or
succinyl-CoA thiokinase. These bacteria can synthesize L-glutamate via
citrate.
The analysis of several bacterial speci~s as to the composition of
their respiratory chains has revealed differences in the number and type
of cytochromes involved and in the occurrence of branched chains.
6. A number of bacteria are able to respire with nitrate instead of
oxygen as H-acceptor. Nitrate respiration proceeds only under anaerobic
conditions. Two processes are distinguished; denitrification in which
nitrate is reduced to N z via nitrite and NzO and nitrate/nitrite respiration
in which nitrate is reduced to nitrite. The latter is excreted or reduced to
ammonia.
Bacterial bioluminescence results from a decay of a complex consisting

of the enzyme luciferase, palmitaldehyde, FMN H 2 and 2 ,
7. Bacteria which metabolize substrates via PEP use either PEP car-
boxylase or pyruvate carboxylase in order to synthesize oxaloacetate. The
glyoxylate cycle is generally employed for oxaloacetate synthesis when
aerobes grow on acetate. PEP synthetase and pyruvate-phosphate dikinase
serve several microorganisms for PEP synthesis from pyruvate.
8. Reserve materials accumulated by bacteria under certain conditions
are polyphosphate and the energy-storage compounds glycogen and poly-
f3-hydroxybutyrate.
9. Bacteria vary with respect to the composition of their cell walls. In
Gram-negative bacteria, the composition of the peptidoglycan layer is very
similar but the outer membrane layer shows much variation as to the
nature of phospholipids and sugars present.
In Gram-positive bacteria, the peptidoglycan layer accounts for 80-90%
of the cell wall components. The composition of the tetrapeptide chains
and the type of cross-linkages vary among the Gram-positive bacteria.
L-Lysine and L-ornithine are found instead of meso-diaminopimelate at
position 3 of the peptide chains; besides the direct cross-linkage between
positions 3 and 4 of two adjacent peptides, interpeptide bridges such as
pentaglycyl bridges are found. Cell walls of Gram-positve bacteria also

contain proteins and teichoic acids.
10. Archaebacteria are devoid of peptidoglycan. Some of them contain
pseudomurein, others glycoproteins or heteropolysaccharides as principal
wall component. Membranes of methanogens do not contain phos-
pholipids as major constituent but a novel type of glycerol ether lipid. The
subunit composition of archaebacterial RNA polymerase is different from
the one of eubacteria.
Chapter 6
Catabol ic Activities of
Aerobic Heterotrophs
Thus far we have discussed aerobic growth on very common substrates,

such as glucose, lactose, or some organic acids. Clearly, mineralization of
organic matter requires that microorganisms have the ability to degrade a
vast number of organic compounds. When animals and plants die, a
number of low-molecular-weight compounds becomes available to the
microorganisms, as well as polymers, such as starch, cellulose, other
polysaccharides, nucleic acids, and proteins. Furthermore, end products of
bacterial fermentations (methane, propionate, butyrate, etc.) diffuse into
aerobic zones and can serve as growth substrates.
Many bacteria exhibit an enormous flexibility as to the nature and the
number of substrates they can utilize. Den Dooren de long found that 200
different organic compounds can serve as sole source of carbon and energy
for Pseudomonas putida. Other pseudomonads, Bacillus, Azotobacter, and
Acinetobacter species are also remarkable in that they can grow with many
substrates. On the other hand, there are organisms that are more "special-
ized", Bacillus fastidiosus, for instance, can utilize only uric acid and
related purine compounds. Bacteria growing with methane are restricted
to C,-compounds as growth substrates. Some of the catabolic activities of
aerobic microorganisms will be discussed in this chapter.
I. Degradation of Polymers by Exoenzymes

Polymers as such cannot penetrate the cell membrane. Therefore, bacteria
excrete enzymes (in most cases hydrolases) that degrade the polymers to
small transportable molecules. These enzymes are either set free by the
organisms or remain associated with them.
142 6: Catabolic Activities of Aerobic Heterotrophs
Starch consists of D-glucose residues that are connected by 0'(1 ~ 4)

linkages and at branch points by 0'(1 ~ 6) linkages. Starch is a storage
material, and as such is structured to be degradable. It is, therefore, not
surprising that powerful starch-decomposing enzymes are produced and
also excreted by microorganisms. Four types of starch decomposing
enzymes are of importance: a-amylase, ,B-amylase, pullulanase or isoamy-
lase (debranching enzymes), and glucoamylase.
a-Amylase is produced by bacteria and fungi, notably by Bacillus
species, pseudomonads, and lactobacilli and by Aspergillus species. It
cleaves 0'(1 ~ 4)-glycosidic linkages in the starch molecule at random and
can be designated as an endoglucanase; 0'(1 ~ 6) linkages are left intact.
Starch is degraded to dextrins, maltose, and glucose.
p-Amylase is common in plants and is not so frequently produced by
bacteria. It successively removes maltose residues from the nonreducing
end of starch and can be designated as an exoglucanase.
Glucoamylase is produced by fungi (Aspergillus and Rhizopus species).
The enzyme removes glucose residues from the nonreducing end of starch.
Pullulanase and isoamylase are so-called debranching enzymes, because
they hydrolyze the 0'(1 ~ 6) linkages in starch. Pullulan is a polysaccharide
formed by the fungus Pullularia pullulans. It consists of D-glucose and
contains 0'(1 ~ 4) and (1 ~ 6) linkages. Both enzymes differ slightly in
specificity. In industrial starch saccharification the addition of a debran-
ching enzyme is necessary for the degradation of dextrins that cannot be
degraded further by 0'- or ,B-amylase.
The mode of action of the various enzymes attacking starch is depicted
in Fig. 6.1.
Cellulose is a polymer of glucose in which the molecules are linked to
one another by ,B-(1 ~ 4) linkages. It is degraded rather slowly by enzymes,
especially slowly when it is encrusted with lignin, because its structure is
very compact, with crystalline and amorphous stretches in the macro-
molecule (Fig. 6.2).
Cellulose can be hydrolyzed and utilized by several bacteria and fungi:
Aerobic bacteria such as Cellulomonas species, gliding bacteria
.. l3-amylase
>+Of---O'-amylase
pullulanase
glucoamylase
Figure 6.1. Mode of action of starch-decomposing enzymes.
Degradation of Polymers by Exoenzymes 143
(Cytophaga and Sporocytophaga) and streptomycetes; anaerobes such as

rumen organisms (Bacteroides succinogenes, Butyrivibrio fibrisolvens) and
Clostridium thermocellum; and various fungi of which only Trichoderma
reesei is mentioned here. Organisms growing with celIulose excrete the
enzyme complex calIed cellulase. It consists of an endo-,8-(l ~ 4)-glucanase
(Cx-celIulase) that in its mode of action is comparable to a-amylase and an
exo-,8-(l ~ 4)-glucanase, which removes cellobiose from nonreducing
ends of the macromolecule (like ,8-amylase). The action of these enzymes
is illustrated in Fig. 6.2. The primary attack is carried out by the
cellulase enzymes
Crystalline~ ~ amorphous
regIOns ~ regIOns
~
~
j endo-j3-( 1.... 4)-glucanase

(ex-cellulase)
______" ' - -0.,
! exo-j3-{ ' .... 4)-glucanase

(cellobiohydrolase)
! combined action of
the two enzymes
-~~':"ib
.-:--.---
-
...... . . ..-.
~ .-.
~
13-(1 .... 4 )-glucosidase

(cellobiase)
cellotriosefcello biose glucose
Figure 6.2. Mode of action of the cellulase components. [Redrawn from S. M.

Cuskey, E. M. Frein, B. S. Montenecourt, and D.E. Eveleigh, in Overproduction
of Microbial Products (V. Krumphanzl, B. Sikyta, and Z. Vanek, eds), Academic
Press, (1982) p. 407.]
Cx-cellulase that breaks bonds in amorphous regions of the macro-

molecule. Subsequently, the combined action of the two enzymes brings
about the decomposition of the polymer. Fungi and probably some aerobic
cellulolytic bacteria excrete in addition a cellobiase [/3(1- 4)-glucosidase]
that hydrolyzes cellobiose or cellotriose to two or three molecules of
glucose. Thus, the cellulase complex of these organisms consists of up to
three enzymes.
However, many bacteria have uptake systems for disaccharides (cello-
biose, maltose). In these organisms the intracellular breakdown of di-
saccharides is often initiated by a phosphorylytic cleavage:
cellobiose + Pi cellobiose phosphorylase) glucose-l-P + glucose
P
maltose + Pi
maltose phosphorylase
) glucose-l- + gIucose
sucrose + Pi sucrose phosphorylase) glucose-loP + fructose
The phosphorylases are, of course, more economical than hydrolases

such as /3-galactosidase and invertase (which hydrolyzes sucrose to
fructose + glucose). The energy of the glycosidic link is saved and ATP is
not required for the formation of sugar-I-phosphate.
Sucrose phosphorylase was first discovered in Pseudomonas saccha-
rophila. Maltose and cellobiose phosphorylases occur in bacteria that
decompose starch and cellulose.
Other polysaccharides and related compounds are also hydrolyzed by
specific exoenzymes. Many bacteria are able to produce pectinase and to
degrade pectin (e.g., Bacillus polymyxa, Erwinia carotovora, Clostridium
felsineum). Pectin is the methyl ester of 0'-(1- 4)-polygalacturonic acid,
and pectinase is a mixture of a methylesterase that produces methanol and
of endo- and exoenzyme (analogous to amylase and cellulase). The pec-
tinase used in food technology is from Aspergillus niger. Corynebacteria,
Chromobacterium violaceum, Pseudomonas chitinovorans, and other soil
bacteria excrete chitinase. Other exoenzymes of bacterial origin are
hyaluronidase, neuraminidase, xylanase, and agarase.
In order to be able to utilize the proteins of dead organisms many
bacteria excrete proteases; these include bacilli, pseudomonads, Proteus
vulgaris, Clostridium species, many other anaerobes, and fungi. On the
basis of their pH optima, alkaline, neutral, and acid proteases are
distinguished. A well-characterized and widely used alkaline protease is
subtilisin which is produced by bacilli (B. subtilis, B. licheniformis, B.
amyloliquefaciens). It is active in the pH range 8-11. The neutral proteases
excreted by B. megaterium and P. aeruginasa are unstable and are
inactivated under acidic or alkaline conditions. Acid proteases, with a pH
optimum between 4 and 6, are produced by Aspergillus species. Like the
alkaline proteases they are unspecific and cleave "every" peptide bond.
Well-known acid proteases that cleave only peptide bonds between certain
Growth with Amino Acids 145
amino acids are pepsin and the milk-clotting rennin. The latter is tradi-
tionally produced from the abomasum (the fourth stomach) of milk-fed
calves. Acid proteases of the rennin type are formed by lactobacilli,
streptococci, bacilli, and a number of fungi (e.g., Mucor pusillus).
Nucleic acids are hydrolyzed by ribonucleases (RNases) and deoxyribo-
nucleases (DNases). The excretion of the latter type of enzyme has been
reported for hemolytic streptococci, Staphylococcus aureus, and clostri-
dial species. A powerful ribonuclease is excreted by Bacillus subtilis in the
late log phase of growth and in the stationary phase. An RNase produced
on a large scale is nuclease P from Penicillium citrinum.
It should also be mentioned here that lipids are hydrolyzed as well by
exoenzymes before uptake. Many organisms excrete Iipases that hydrolyze
triglycerides (triacylglycerols) to fatty acids and glycerol.
II. Growth with Amino Acids

Amino acids and low-molecular-weight peptides produced by proteases are
actively taken up and utilized for growth by many microorganisms. Some
of the amino acids are structurally so related to central intermediates of cell
metabolism that their degradation is very easy. In most cases the amino
acid is first converted to the corresponding keto acid.
glutamate - 2-oxoglutarate
aspartate - oxaloacetate
alanine -pyruvate .
valine - 2-oxoisovalerate
leucine - 2-oxoisocaproate
isoleucine - 2-oxo-3-methylvalerate
This oxidative deamination can be accomplished in the following dif-
ferent ways.
(a) Oxidation by cytochrome-linked oxidases
I via respiratory chain
R-CH-COOH + 202 R-CO-COOH + NH 3
I
NH 2
Several bacteria contain L-amino and o-amino acid oxidases. The

latter are important because of the presence of o-amino acids in some
polymers (e.g., peptidoglycan) and because they work together with
racemaces, which catalyze the conversion of L-amino acids into
o-amino acids. The oxidases are flavoproteins and feed the electrons
into the respiratory chain. They are relatively unspecific, and a
particular oxidase may attack 10 different amino acids.
(b) Oxidation by NAD(Pt -linked dehydrogenases

alanine dehydrogenase
CH 3-CH-COOH + NAO+ + H20
I
NH,
CH 3 -CO-COOH + NAOH + NH. +
Alanine dehydrogenase occurs in a number of bacilli and clostridia.

Glutamate dehydrogenase is very widespread and catalyzes the
analogous reaction with glutamate as substrate.
(c) Transamination with pyruvate or 2-oxoglutarate as acceptor of the
amino group and subsequent regeneration of the acceptor by a
dehydrogenation reaction (as in b).
R CH 3 R CH 3
I I transaminase I I
HCNH 2 + CO ~ J C=O + HCNH,
I I I I
COOH COOH COOH COOH
,..~
""... a/a . aSC ,."
"
' ....... _~~~~- ~:_~~~:?~~.!'-~,...'"
Whereas 2-oxoglutarate, oxaloacetate, and pyruvate can easily be
handled by bacterial cells, specific catabolic routes are required to channel
2-oxoisovalerate or 2-oxoisocaproate into the intermediary metabolism.
The routes found in bacteria are the same as those found in animals, and
lead to the formation of acetyl-CoA and propionyl-CoA.
Another reaction used to initiate the breakdown of amino acids is
deamination. A prerequisite for this reaction is that the removal of
ammonia from the a-amino carboxylic acid is facilitated by substituents at
the .a-carbon atom. Thus, amino acids, such as serine, threonine, aspar-
tate, and histidine are subject to deamination. Serine and threonine
deamination yields pyruvate and 2-oxobutyrate, respectively. 2-
Oxobutyrate can be oxidized by a multienzyme complex resembling the
pyruvate dehydrogenase complex to yield propionyl-CoA; the metabolic
fate of the latter will be outlined later.
serine
dehydratase
-'------+. CH 3 -CO-COOH + NH 3
pyruvate
serine
threonine
dehydratase
H 3 C-CH-CH-COOH -'------+. CH 3 -CH,-CO-COOH + NH 3
I I 2oxobutyrate
OH NH 2
These reactions start with the removal of water:

CHpH CH, CH 3
I H,O II I
HC-NH 2 ~ C-NH, C=O
I I I
COOH COOH COOH
Growth with Amino Acids 147
For this reason the enzymes are called dehydratases (also common is
deaminases).
Catabolic threonine dehydratase of E. coli differs in its properties,
especially in its regulatory aspects, from the anabolic threonine dehydra-
tase present in the same microorganism; the latter catalyzes the first step of
a reaction sequence leading from 2-oxobutyrate to isoleucine.
Many aerobes synthesize aspartase if they grow on aspartate:
aspartase
aspartate E fumarate + NH 3
A similar reaction initiates the breakdown of histidine (Fig. 6.3).
Several different pathways for the degradation of arginine have been
described. They are depicted in Fig. 6.4. It is apparent that degradation is
initiated by four different enzymes: arginase, arginine deiminase, arginine
oxidase, and arginine decarboxylase. The enzyme arginase is present in
Bacillus subtilis, B. licheniformis, and Proteus vulgaris, the deiminase in P.
aeruginosa, Clostridium perfringens, and in streptococci and lactobacilli.
L-Citrulline formed by the deiminase is converted into L-ornithine and
carbamoyl phosphate by ornithine carbamoyltransferase. The latter pro-
duct can then be used, notably by the anaerobes, for ATP synthesis in the
carbamate kinase reaction. Arginine oxidase occurs in P. putida and the
decarboxylase in most enterobacteria including E. coli. Several microor-
ganisms (e.g., pseudomonads) contain more than one pathway for arginine
utilization which is so important for organisms under N-limitation because
of the four nitrogens present in the molecule.
Aerobic breakdown of the aromatic amino acids is also feasible for
many microorganisms. Thus, it is understandable that many aerobes grow
t> ~H,
COOH COOH COOH
L> of)
I I I
H 2 NCH histidase CH urocanase CH2
I II I
;'0
urocanate 4-imidazolone-
5-propionate
COOH COOH
r r
I formimino- I
imidazolone H2 glutamate H2
propionase
r H2 hydrolase
rH2
7'
H 20
HC-NH-CH=NH - - - -..
I
COOH
, H -NH 2
COOH
r + HCONH 2
N-formiminO-L- L-glutamate + formamide

glutamate
Figure 6.3. Breakdown of histidine.
11
agmatine
10 2-oxoarginine
r-
II
NH2
12
citrulline
NH
(r
I
H2 )]
p_
i "7 \.
ADP ATP
9 ) CO 2
CH 2 -NH 2
f'-- ~-NH2 .. + NH
a
N-carbamoyl-
putrescine
0-0
I
NH] f- NH2
NH
carbamoyl
H20~, phosphate d::H 2)]
tHO
~ NH] +C0 2
NH 2
({H 4-guanidino-
2 )] butyraldehyde
(r
~H2
H2h
1H- NH2
COOH
CH 2-NH 2
ornithine
putrescine
~2t
2H
NH] Iglutamate I urea

NH
~H2 ~_ NH
(CH ) H20 H20 I 2
CHO 12]~~NH
5 14 (tH )
I 2]
4-aminobutyraldehyde NH 2 COOH
I urea
2H (<[H 2h 4-guanidinobutyrate
COOH
4-aminobutyrate
~ 15
~
I succinate I
Figure 6.4. Pathways of arglOme degradation. 1, Arginine decarboxylase; 2,
agmatine deiminase; 3, N-carbamoylputrescine hydrolase; 4, putrescine oxidase; 5,
aminobutyraldehyde dehydrogenase; 6, arginine deiminase; 7, ornithine carba-
moyltransferase; 8, reaction sequence via N 2 -acetylornithine and N-acetyl-
glutamate semialdehyde is identical with the pathway of arginine biosynthesis;
9, carbamate kinase; 10, arginase; 11, arginine oxidase; 12, 2-oxoarginine decar-
boxylase; 13, 4-guanidinobutyraldehyde oxidoreductase; 14, guanidinobutyrase;
15, a transaminase yields succinate semialdehyde which is oxidized to succinate by a
dehydrogenase [Modified from: V. Stalon and A. Mercenier; J. Gen. Microbial.
130, 69-76 (1984)].
Growth with Organic Acids 149
well on nutrient broth and other media containing mostly amino acids and
peptides.
III. Growth with Organic Acids

In connection with the general metabolism of E. coli, we have already
discussed the metabolic implications of growth on acids such as pyruvate
and acetate, particularly the requirement of special anaplerotic sequences
for growth on these compounds. Some further remarks about growth on
organic acids are necessary. A number of aerobes can grow on long-chain
fatty acids (pseudomonads, Acinetobacter, bacilli, E. coli); they are
usually degraded by {3-oxidation. The fatty acid, for instance palmitic a.:::id,
is first converted to the corresponding CoA ester by acyl-CoA synthetase.
Such synthetases have a low specificity. The enzyme from Bacillus
megaterium, for instance, reacts with acids from C 6 to C zo . The CoA ester
is then oxidized in the {3-position and subsequently cleaved to yield
acetyl-CoA and the CoA ester of the fatty acid shortened by two carbon
atoms. One l3-oxidation cycle requires the action of four enzymes, as is
illustrated in Fig. 6.5. In the first dehydrogenase reaction the CoA ester of
the saturated fatty acid is converted into the CoA ester of an a,
l3-unsaturated fatty acid. The redox potential of this type of redox reaction
lies in the order of Eb = zero V, and this reaction cannot be coupled with
the reduction of NAD+. Enzymes of this type are f1avoproteins, and FAD
is reduced. The addition of water as catalyzed by a dehydratase is followed
by a NAD +-specific l3-hydroxyacyl-CoA dehydrogenase reaction. Finally,
cleavage of the l3-oxoacyl-CoA with CoA proceeds under catalysis of
l3-ketothiolase. Even-numbered fatty acids yield only acetyl-CoA and it is,
therefore, apparent that microorganisms growing on these substrates
require the glyoxylate cycle as anaplerotic sequence and PEP carboxyki-
nase to form PEP for gluconeogenesis.
With odd-numbered fatty acids, the final l3-oxidation cycle yields
acetyl-CoA and propionyl-CoA. The latter is also formed in L-valine and
L-isoleucine catabolism, and, in fact, a number of bacteria are able to grow
with propionate as energy and carbon source.
The metabolism of propionate is initiated by its activation to propionyl-
CoA. For the further metabolism of propionyl-CoA several different
pathways have been found that might be involved. The conversion of
propionyl-CoA to succinyl-CoA is widespread; it occurs in animal tissues,
Paracoccus denitrificans, and rhizobia. The first enzyme carboxylates
propionyl-CoA to yield methylmalonyl-CoA. The carboxylase contains
biotin and the reaction resembles the one catalyzed by pyruvate carboxy-
lase. Methylmalonyl-CoA undergoes a rearrangement reaction leading to
succinyl-CoA. The mutase catalyzing the last reaction contains a derivative
of vitamin BIz as an essential coenzyme. The formation of succinyl-CoA is
CH) -CHl -CHl-CHl-CHl-CHz-C1ll CHl-CH2 -CHl ~CHl CH z -CH 2 -CH l ~CHl-COOH
CH)-CHl-CH1
f: COA, AT_P
AMP, PP,
CHl-CHl-CHl~Hl CHl-CHz-CHz-CH1-CHl-CHl-CHl~CHl-CO-CoA
acyl-CoA syn thetasc
l---
FAD fatty acyl-CoA dehydrogenase
r
t-FADH 2
CH) -CHl -C1ll-CHl-CHl CIl 2 -CH l --CHl . Cli l --CH 2 - CHl-CHl-CHl-CH=CH-CO--CoA
0
Hl 3-hydroxyacyl-CoA hydrolyase
CIl) -CHl-CHl-CHl CHl -CHl -CH 2 -CHl CH l - Clll-CH z --CH l -CH 2 -~H-CH2 -CO-CoA
~ NAD+ OH
}-- NADH + H+ l-3-hydroxyacyl-CoA dehydrogenase
CH) -CH 2-CH 2 -CHl- CHl - CHl CHl C~'~; --CH 2 -CHl-CHl--CHl- ~-CHl-CO-COA
t ~-ketothiolase
~--=-.,----:,.--,
CH) -CH l --CH 2 - CH l CH 2 -CO-CoA + ICH3 --eo-,coAI
ICH)-CO-COAI
!
leH) eo- coAl + ICH) CO CoA I !
Figure 6.5. ,a-Oxidation of palmitic acid.
shown in Fig. 6.6. Since these reactions are so important in the propionate
fermentation, more details will be given in Chapter 8.
E. coli converts propionyl-CoA to pyruvate by the following reactions:
CH 3 CH2 CH 3 CH3
I II I I
+---
2H H 2O CoA
CH 2 ~CH ~ CHOH c=o
I I I 2H I
CO-SCoA CO-SCoA CO-SCoA COOH
propionyl-CoA acrylyt-CoA lactyl-CoA pyruvate
COOH
I
<fH2-CH2
CO-SCoA
succinyl-CoA propionyl-CoA
..
COOH COOH
I 2 I
H]Cl'H Hy-CH]
CO-SCoA CO-SCoA
R -methylmalonyl-CoA s-methylmalonyl-CoA
Figure 6.6. Formation of succinyl-CoA from propionyl-CoA. 1, Propionyl-CoA
carboxylase, contains biotin; 2, methylmalonyl-CoA racemase; 3, methylmalonyl-
CoA mutase, contains vitamin B t2
The same intermediates occur in Moraxella lwoffi, but they are enzyme-
bound and are not found as CoA esters.
Glyoxylate appears in nature as a degradation product of the purine
bases. Under aerobic conditions these bases are oxidized by xanthine
oxidase to uric acid. The latter is oxidized further to allantoin by the
enzyme uricase. Hydrolysis of allantoin yields allantoate, which in different
microorganisms is metabolized further by one of the two reaction se-
quences illustrated in Fig. 6.7. Pseudomonas aeruginosa, P. fiuorescens,
and Penicillium species form first 2 mol of urea and 1 mol of glyoxylate.
P. acidovorans, Streptococcus allantoicus, Arthrobacter allantoicus, and E.
coli, on the other hand, first remove ammonia and CO 2 from allantoate so
that finally 1 mol of urea per mol of glyoxylate is formed. These differences
are probably not important for the microorganisms because they synthesize
the enzyme urease if ammonia becomes growth-limiting. Urease cleaves
urea to ammonia and carbon dioxide:
urease
NH 2 -C-NH 2 + H 2 0 ~ 2NH) + CO 2
II

Growth of E. coli and pseudomonads on glyoxylate requires some
enzymes that have not been discussed as yet. The production of reducing
power for the respiratory chain precludes that a reaction sequence is
present in the organisms by which glyoxylate can be converted into an
intermediate of the central metabolism. This pathway consists of three
enzymes and is known as the glycerate pathway (Fig. 6.8). The first
enzyme, glyoxylate carboligase, catalyzes the condensation of two mole-
cules of glyoxylate to CO 2 and tartronate semialdehyde. The latter is
urate allantoin
H20 NHj,C0 2
~./ NHl <[OOH
--'------"--"'--.... oJ; W".cH- NH2
IP. acidovorans I H
allantoate ureidoglycine
I P. aeruginosa I
~Hl
COOH C=O
7H2
C=O COOH
I I I I
urea + HO-CH-NH NH--CH-OH
(-) ureidoglycollate (+) ureidoglycollate
COOH
I + urea
HC=O
glyoxylate
Figure 6.7. Degradation of urate to glyoxylate by P. aeruginosa and P. acidovor-

ans. 1, Uricase; 2, allantoinase; 3, allantoicase; 4, allantoate amidohydrolase; 5,
ureidoglycine aminohydrolase; 6, ureidoglycolase.
C02
20=CH-COOH I ~. O=CH-CHOH-COOH
glyoxylate tartronate semialdehyde
2~ NADH+H+
~NAD+
HOCH2-CHOH-COOH
glycerate
3~ATP
r-
OCH2 -CHOH-COOH
3-phosphoglycerate
ADP
Figure 6.8. The glycerate pathway. 1, Glyoxylate carboligase; 2, tartronate

semialdehyde reductase; 3, glycerate kinase.
IO=CH-COOH I------...,..~ H2 N-CH 2 -COOH
H2 N-CH-COOH
I
HO-CH-COOH
ery t hro -I3-hydroxyaspa rtate
Ioxaloacetate I
Figure 6.9. The ,B-hydroxyaspartate pathway. 1, Transaminase; 2, erythro-,B-
hydroxyaspartate aldolase; 3, erythro-,B-hydroxyaspartate dehydratase.
reduced to glycerate, which is phosphorylated to yield 3-phospho-
glycerate-an intermediate of the Embden-Meyerhof-Parnas pathway. It
can be oxidized via pyruvate and the tricarboxylic acid cycle. Microorgan-
isms that have the glycerate pathway at their disposal contain malate
synthase in addition. It functions as an anaplerotic enzyme to replenish the
pool of C4 -dicarboxylic acids in the organisms.
The metabolism of glyoxylate in Paracoccus denitrificans is different.
Glycine is formed from glyoxylate by transamination. It then condenses
with another molecule of glyoxylate to form erythro-j3-hydroxyaspartate,
which is converted to oxaloacetate and ammonia (Fig. 6.9). Glycollate is
metabolized by P. denitrificans by a similar route.
The most highly oxidized Cz-compound is oxalate, and it is rather
surprising that some microorganisms are able to utilize it as carbon and
I \
HOOC-COOH / " ~ .. HOOC-CO-SCoA
oxalate oxalyl-CoA
HCO-SCoA HCOOH
formyl-CoA format; l----
NAD+
~NADH+W 2
CO 2
cO 2
Figure 6.10. Oxidation of oxalate by P. oxalaticus. 1, Oxalate is converted to
oxalyl-CoA in a CoA transferase reaction using formyl-CoA as donor; 2, oxalyl-
CoA is decarboxylated to yield formyl-CoA, which is converted to formate by the
transferase reaction; 3, formate is oxidized by formate dehydrogenase.
energy source. One of these is Pseudomonas oxalaticus, which was studied

in detail by Quayle. It gains reducing power by the oxidation of oxalate to
CO 2 via oxalyl-CoA and formate (Fig. 6.10).
Synthesis of cell constituents begins with the reduction of oxalyl-CoA to
glyoxylate:
oxalyl-SCoA + NADPH + H+ ~ glyoxylate + NADP+ + HSCoA
Glyoxylate is then metabolized further via the glycerate pathway.
IV. Growth with Aliphatic Hydrocarbons

Because of the inertness of hydrocarbons it is interesting that many
microorganisms are able to utilize these compounds for growth. The
primary attack on hydrocarbons requires oxygen, so that growth on them is
an obligately aerobic process. From the nature of the substrates utilized
two groups of organisms can be envisaged.
(a) Bacteria growing with methane. Since growth on a C t compound
requires special pathways, these bacteria deserve special attention and
will be discussed in a later section of this chapter. Methane-oxidizing
bacteria do not grow with other hydrocarbons.
(b) Organisms growing on hydrocarbons other than methane. This is a
rather heterogeneous group. Relatively few bacterial species (mycobacte-
ria, flavobacteria, Nocardia), are able to grow on ethane, propane, butane,
and hydrocarbons up to Cg. However, utilization of long-chain hydrocar-
bons is widespread among microorganisms, and n-alkanes with 10-18
carbons are utilized with the greatest frequency and rapidity. Table 6.1
shows a small selection of organisms able to grow on long-chain hydrocar-
bons. It is noteworthy that yeasts and fungi also utilize such substrates.
. coli, Enterobacter aerogenes, and Bacillus subtilis are unable to grow
with hydrocarbons.
Table 6.1. Organisms capable
of growing with long-chain
hydrocarbons
Pseudomonas fiuorescens
P. aeruginosa
Acinetobacter calcoaceticus
Mycobacterium smegmatis
Nocardia petroleophila
Candida lipolytica
Torulopsis colliculosa
Cephalosporium roseum
Arthrobacter paraffineus
Arthrobacter simplex
Corynebacterium glutamicum
Growth with Aliphatic Hydrocarbons 155
Hydrocarbons are water-insoluble compounds and their uptake and

their enzymatic attack are difficult tasks. In times in which oil catastrophies
threaten our water resources, the activities of the above-mentioned
organisms are especially important. Since organisms live in an aqueous
environment and cannot live in oil, a large oil-water boundary layer is a
prerequisite for a rapid degradation of oil. Attack of the hydrocarbon is
possible only in the boundary layer.
Hydrocarbon utilization is initiated by its uptake. Yeasts have been
shown to contain a rather thick and glycolipid-rich cell wall when growing
on hydrocarbons. Bacteria form trehalolipids, rhamnolipids, or analogous
structures as constituents of their cell walls in which the hydrocarbons are
dissolved and transported to the cytoplasmic membrane. Acinetobacter
strains have been shown to excrete particles with the composition of the
outer membrane, in which hydrocarbons are soluble. On reassociation of
the particles with the outer membrane, the hydrocarbons can be transfer-
red to the membrane. Thus, the general strategy is to produce an emulsifier
that mediates transfer of the hydrocarbon from the oil layer to the
cytoplasmic membrane.
The first three reactions on the hydrocarbon proceed within the
membrane, and the enzymes responsible for them are membrane-bound.
The primary attack of the hydrocarbon is brought about by a monooxyge-
nase. Enzymes of this type catalyze reactions according to the following
equation:
substrate-H + Oz + AH z monooxygenase, substrate-OH + A + HzO
In addition to the substrate to be oxidized, an oxidizable cosubstrate
(AH z) is required. In hydrocarbon oxidation this function is fulfilled by
reduced rubredoxin and the formation of the primary alcohol proceeds as
depicted in Fig. 6.11. Rubredoxin, a small protein that contains iron (Fe z+
NADH + W NAD+
~
"d""~,"" I"dl
rubre!d:~J\\L'i:::~"'1
o
- ICHO+
NAD+ NADH ICOOH
22
H 0 NAD+ NADH + W
'If
,,/
.. i
'ZL
H 0 +W
.,
3 4
Figure 6.11. Terminal oxidation of an n-alkane by Pseudomonas oleovorans. 1,

Rubredoxin: NADH oxidoreductase; 2, n-alkane monooxygenase; 3, alcohol
dehydrogenase; 4, aldehyde dehydrogenase.
or Fe3+) linked to cysteine residues of the polypeptide chain, is reduced by

a soluble NADH-dependent enzyme system. It then functions as a cosub-
strate in the monooxygenase reaction. In Corynebacterium species and in
yeasts, reduced cytochrome P450 is the cosubstrate. It receives reducing
power from NADH via a flavoprotein and iron-sulfur protein. The primary
alcohol formed is then oxidized further via the corresponding aldehyde to
the corresponding organic acid by NAD+ -dependent dehydrogenases.
Finally, the fatty acid is degraded by f3-oxidation, as previously described.
It should be mentioned that a diterminal attack of hydrocarbons is also
observed. As a result a dicarboxylic acid is formed which is also degraded
by f3-oxidation.
A different mechanism of hydrocarbon oxidation has been found in
certain organisms (Nocardia species). There, a subterminal attack of the
hydrocarbon chain is employed that leads to the formation of a secondary
alcohol (Fig. 6.12). This alcohol is oxidized to the corresponding ketone,
which is subjected to a second monooxygenase reaction. The product is an
acetylester that on hydrolysis yields acetate and a long-chain alcohol that is
degraded further as outlined above. Evidence for the occurrence of this
mechanism in the organisms mentioned comes from the isolation of the
acetylester and from labelling experiments. Terminal or subterminal
oxidation results in differences in the origin of the carbon atoms of the
,CH 3 /CH 3 /CH 3
~H2 q'lOH NAD+ NADH + W c,=o
CH
2
A~ I { CH
2
\.. 2L . CH 2
02 H 20
Figure 6.12. Subterminal oxidation of an n-alkane. 1 and 3, Monooxygenase

reactions; 2, secondary alcohol dehydrogenase; 4, acetylesterase.
Growth with Aromatic Compounds 157
acetate produced. C j of the hydroxycarbon becomes C 1 of acetate (termi-

nal attack) or C2 (subterminal attack).
Alkenes are also degradable. Either water is added across the double
bond or an epoxide is formed first, which is subsequently hydrolyzed to
yield a dihydroxy alkane.
V. Growth with Aromatic Compounds

Large amounts of compounds containing aromatic rings are produced by
plants; the most predominant of these is lignin. Aromatic amino acids and
vitamins are constituents of every organism. All these compounds become
available when organisms die and organic matter is decomposed; they are
degraded by bacteria and fungi.
0
The pathway used by animals and bacteria for phenylalanine and
tyrosine degradation is outlined in Fig. 6.13. A key intermediate involved
Q
OH OH
0
0 2 + NADPH + W NADP+ + H2
~ '\--L. ~ I
I ,
CH, eH, CII>

I - t - I -
HC-NH,
I -
HC-NH,
I -
c=o
I
eOOH eOOH COO II
tyrosine I'-hydroxyphenyl-
pyruvate
~CH'-COO"
0,
(
3 4
eOOH
homogentisate 4-maleylacetoace late
4-furn arylacetoace tate fumarate + acetoacetate
Figure 6.13. Degradation of phenylalanine and tyrosine by the homogentisate

pathway. 1, Phenylalanine hydroxylase; 2, transamination reaction; 3, p-
hydroxyphenylpyruvate oxidase, which catalyzes decarboxylation, migration of the
side chain, and hydroxylation of the ring; 4, homogentisate oxidase; 5,
maleylacetoacetate isomerase (requires glutathione); 6, fumarylacetoacetate hy-
drolase.
is homogentisate, which is oxidized to maleylacetoacetate and metabolized

further to fumarate and acetoacetate. A number of bacteria are able to
cleave gentisate in a similar fashion, with fumarate and pyruvate being
formed. Gentisate can be formed in Pseudomonas acidovorans, for
instance, by hydroxylation of m-hydroxybenzoate.
OH
C;COOH
OH
gentisate
Gentisate and homogentisate, however, are not the only compounds

that can oxidatively be cleaved to nonaromatic products. In fact, the
6
COOH CHOH-COOH
o - (1 p-hydroxY-L-
mandelate
CH 3 OH OH
'-h'jro"b,",,,,lfom",
,~~.::: 'f1
Y OHyOH
;:;H HO:O:::
o
OH OH OH
COOH p-hydroxy-
benzaldehyde shikimate quinate
t t
b enzoate
\ COOH
COOH
OOH
OH OH
m-hydroxy-
benzoate
p-hydroxy-
benzoate
5-dehydroshikimate
/" QOCH'OH
vanillate
OH
protocatechuate
Figure 6.14. Aromatic and hydro-aromatic compounds that can be converted to
protocatechuate. [R. Y. Stanier and L. N. Omston, Adv. Microbial Physiol. 9,
89-151 (1973).]
majority of aromatic compounds are converted by bacteria into catechol

and protocatechuate, which-as discovered by Stanier and collaborators-
are the "starting substrates" in the subsequent oxidative cleavage reac-
tions. Figures 6.14 and 6.15 show catabolic routes leading to catechol and
protocatechuate formation. It is apparent that pathways exist by which
compounds such as anthracene, mandelate, trytophan, or quinate can be
metabolized to the "starting substrates" mentioned. In fact, Figs. 6.14 and
6.15 show only a small selection of compounds.
Two types of cleavage reactions are known for catechol and pro-
tocatechuate: artha-cleavage and meta-cleavage.
6 6
CHOH-COOH
L -mandelate t oluene
! !
anthracene 6 cooll
benzoyl formate
611
benzyl alcohol
~ CO-CH z --CHNH 2 -COOH
6
CHO/
QNHCHO
phenan threne formylkynurenine

benzaldehyde
~ ~
COOH CO-CH z --CHNH z -COOH
naphthalene 6
benzoate
0
L-kynurenine
NH2
1
j O
6-
COOH COOH
0H NH
':.
2
~nthranilate
'.""'."~
0--- 6
0"/ Oil
1/---6
benzene catechol phenol
Figure 6.15. Aromatic compounds that can be converted to catechol. [R. Y.
Stanier and L. N. Omston, Adv. Microbial Physiol. 9, 89-151 (1973).]
A. Ortho-c1eavage or 3-oxoadipate pathway

As illustrated in Fig. 6.16 the aromatic rings of catechol and protocate-
chuate are cleaved, via dioxygenase reactions, between the two hy-
droxyl groups. The products-cis,cis-muconate and ,B-carboxy-cis ,cis-
muconate-then yield in two reactions, the first common intermediate of
these pathways-4-oxoadipate enol-lactone. This compound is degraded
further to yield succinate and acetyl-eoA. The reactions of the catechol
and the protocatechuate branch are catalyzed by different sets of enzymes;
there is, for instance, a catechol 1,2-dioxygenase and a protocatechuate
O
OH
1-I0oeoIoH
I
~ OH ~ OH
4f0
catechol protocatechuate
HOOC(:
r COOH
~ COOH
cis, cis-muconate (1-carboxy-cis, cis-muconate
51
HOOCtS C.OOH
0,
~ c=o
muconolactone ")'-carboxymuconolactone
,Y(~
~I
0 ('0011
~('=o
4-oxoadlpate
enol-lactone
7!
,OC~:~ll"",t-SCOAI
O~(,OOH
~(,OOH
HSCoA
3-oxoadipate
, . . - - - - " " - - - - - - succinyl-SCoA
Figure 6.16. Reactions of the 3-oxoadipate pathway. 1, Catechol 1,2-dioxygenase;

2, muconate-lactonizing enzyme; 3, muconolactone isomerase; 4, protocatechuate
3,4-dioxygenase; 5, ,B-carboxymuconate-lactonizing enzyme; 6, y-carboxymucono-
lactone decarboxylase; 7, 4-oxoadipate enol-lactone hydrolase; 8, 3-oxoadipate
succinyl-CoA transferase; 9, 3-oxoadipate-CoA thiolase.
3,4-dioxygenase. These enzymes catalyze the actual ring-cleavage reac-

tions, e.g.:
0,
----+
catechol 1,2dioxygenase reaction
In contrast to the hydroxylase, or monooxygenase (see oxidation of

hydrocarbons), an additional hydrogen donor such as NADPH is not
O I '"
OH HOOCl)
OH
OH / OH
catechol protocalechuale
HOOC~OH
l~. ~OOH
CHO
2-hydroxymuconic semialdehyde 2-hydroxy-4-carboxymllconic semialdehyde
r
HOOCyCH"
o
II
CH, COOH
2-oxopent-4-enoate 2-oxo-4-carboxypenl-4-enoale
.....CH"
HCOH c=o
I I
CH 3 COOH
4-hydroxy-2-oxovalerate 4-hydroxy-4-carboxy-
41 2-oxovale:r
e
Ipyruvalel+lacetaldehydel 12 pyruvate I
Figure 6.17. Dissimilation of catechol and protocatechuate by the pathways
involving meta-cleavage. 1, Catechol 2,3-oxygenase; 2, 2-hydroxymuconic semi-
aldehyde hydrolase; 3, 2-oxopent-4-enoic acid hydrolase; 4, 4-hydroxy-2-
oxovalerate aldolase; 5, protocatechuate 4,5-oxygenase; 6, 2-hydroxy-4-
carboxymuconic semialdehyde hydrolase; 7, 2-oxo-4-carboxypent-4-enoic acid
hydrolase; 8, 4-hydroxy-4-carboxy-2-oxovalerate aldolase.
required in the dioxygenase reaction; both oxygen atoms appear in the

product. An analogous dioxygenase initiates the oxidation of benzene to
catechol. Here, dihydroxy-dihydrobenzene and (very likely) a dioxetane
structure function as intermediates.
(XI
NAD' NADH + H'
a~
~H ----'U""""""-- .
H
OH ~
OH
OH
B. Meta-cleavage
In 1959 Dagley and Stopher found that a soil pseudomonad contained
enzyme systems that catalyze the breakdown of catechol and protocatechu-
ate in a different way. The ring is opened adjacent to the hydroxyl groups,
so that 2-hydroxymuconic semialdehyde or 2-hydroxy-4-carboxymuconic
semialdehyde is formed. Their further metabolism leads to the formation
of pyruvate, formate, and acetaldehyde (Fig. 6.17).
The distribution of the various pathways among the bacteria is very
complex. Pseudomonas acidovorans and P. testosteroni metabolize pro-
tocatechuate-yielding substrates via the meta-cleavage pathway and
catechol-yielding substrates via the ortho-cleavage pathway. In other
pseudomonads the ortho-cleavage pathway dominates; this might be true
for many bacteria that grow with aromatic substrates.
VI. Growth with C1 Compounds

In connection with the energy metabolism of aerobes it was mentioned that
a few groups of bacteria cannot employ the tricarboxylic acid cycle for the
production of reducing power. One of these groups comprises the microor-
ganisms growing on C 1 compounds. Two subgroups can be envisaged:
A. Obligate methylotrophs, which grow only at the expense of compounds
containing no carbon-carbon bonds (methane, methanol, etc.). This
group includes all methanotrophs.
B. Facultative methylotrophs, which grow on a variety of carbon sources
including C 1 compounds. These organisms utilize methanol and methy-
lamine but not methane.
A. Obligate methylotrophs
The first methylotroph isolated was Bacillus methanicus (S6hngen, 1906).
Fifty years later it was reisolated and named Pseudomonas methanica. During
the last 10 years a large number of methylotrophic bacteria became known
through the work of Whittenbury and Wilkinson; they are distinguished as
Methylosinus, Methylocystis, Methylomonas, Methylobacter, and Methylo-
Growth with C 1 Compounds 163
coccus. Ultrathin sections of cells of these species revealed that they contain
extensive membrane structures. It is also noteworthy that all these species
form resting stages, in most cases cysts, but some species form thermo-
stable exospores.
Clearly, during growth on methane, reducing power for the respiratory
chain can be produced only by oxidation of methane to CO 2 , There is no
acetyl-CoA available to be oxidized via the tricarboxylic acid cycle. Oxi-
dation of methane proceeds as shown in Fig. 6.18. Methane is first oxi-
dized by a monooxygenase to methanol. NADH functions as cosubstrate
in this reaction. Methanol dehydrogenase is not coupled to the reduction of
NAD+ but to electron transfer to a novel coenzyme-methoxatin (PQQ).
It is evident from its chemical structure that this compound is an artha-
quinone. By reduction the corresponding hydroquinone is formed. The
redox potential of PQQ is Eo = + 120 mV, so it is more suitable as
H-acceptor in a number of dehydrogenation reactions than NAD+. That is
the reason why it also functions as coenzyme in the glucose dehydrogenase
reaction (see Chapter 5) and in other systems.
The oxidation of formaldehyde to CO 2 proceeds in two NAD+ -linked
steps. One of the NADH formed has to be invested by the organism into
the monooxygenase reaction. The second NADH is fed into the respira-
tory chain, as is PQQH z, the latter, however, at the level of cytochrome c.
It is noteworthy that methane monooxygenase does not exhibit an absolute
substrate specificity. For instance, it oxidizes ammonia to hydroxylamine,
but is definitely different from the true ammonia monooxygenase (see
Chapter 9).
The anabolic metabolism of the methylotrophs diverges from the energy
metabolism at the level of formaldehyde. Three formaldehyde fixation
cycles have been elucidated, largely due to work done in the laboratories of
Quayle, Hersh, and Harder.
1. The serine-isocitrate lyase pathway. The formation of acetyl-CoA from
formaldehyde and COz in the serine pathway is shown in Fig. 6.19. The
acceptor for formaldehyde is glycine, and the serine formed in the
hydroxymethylase reaction is converted to the corresponding Q'-oxo acid
- - -... CH)O + PQQH)
_ _3_~. HCOOH + NADH + H'

4
HCOOH + NAD - - _ . CO 2 + NADH + H'
o
PQQ
Figure 6.18. Oxidation of met~ane to CO 2 . 1, Monooxygenase reaction; 2,
methanol dehydrogenase; 3, formaldehyde dehydrogenase; 4, formate dehy-
drogenase. PQQ, methoxatin (2,7,9-tricarboxy-1H-pyrrolo (2,3-f) quinoline-4,5-
dione).
NADH+W
hydroxypyruvate ~ . _ 1
3~AD+
2
glycerate
yATP
4 \ADP
"'~:f~
X
oxaloacetate
NADH+W
glyoxylate 7
~9 NAD+
,..--_fL-~_9_ malyl-CoA ~~;te

I acetyl-CoA I ."; CoA
ADP + Pi
Figure 6.19. Formation of acetyl-CoA from formaldehyde and CO 2 in the serine
pathway. 1, Serine hydroxymethylase; 2, a transaminase that converts serine into
hydroxypyruvate and glyoxylate into glycine; 3, hydroxypyruvate reductase; 4,
glycerate kinase; 5, phosphoglycerate mutase and enolase; 6, PEP carboxylase; 7,
malate dehydrogenase; 8, malyl-CoA synthetase; 9, malyl-CoA lyase.
via a transaminase reaction. Hydroxypyruvate is then further converted to
PEP by the action of several enzymes. Oxaloacetate is formed through the
PEP carboxylase reaction. The key enzyme for the generation of glyoxy-
late-malyl-CoA lyase-came under study only recently, despite the fact
that this cleavage reaction was discovered 15 years ago in Rhodopseudo-
monas sphaeroides. It converts malyl-CoA into acetyl-CoA and glyoxylate.
Before this reaction can proceed, L-malate has to be activated. This is done
by the action of a specific synthetase, malyl-CoA synthetase. The {3-
carboxyl group of L-malate is converted into a CoA-thioester group. This
conversion is associated with the formation of ADP and Pi from ATP.
Besides malyl-CoA synthetase and lyase two other enzymes can be
considered as key enzymes of this cycle, serine-glyoxylate transaminase
and hydroxypyruvate reductase.
COOH COOH
I I
HCOH malylCoA lyase HC=O
I +
CH z CH 3
I I
CO-SCoA CO-SCoA
Growth with C, Compounds 165
2 serine
~( 2 glyoxylatc
2oxaloacetalc
malyl-CoA
acetyl-CoA =====::::
I succinate I ' - . r - - - - :===== citrate
Figure 6.20. Net formation of succinate from formaldehyde and CO 2 by the

serine-isocitrate lyase pathway.
The result of the serine pathway is the formation of acetyl-CoA from

CHzO and COz. The cycle is insufficient to supply the cell with PEP or
oxaloacetate for biosynthetic purposes. This, however, is achieved if it is
combined with the citrate synthase, cis-aconitase, and isocitrate lyase
reactions (Fig. 6.20). In essence, this accomplishes the net formation of
succinate from 2CO z and 2 formaldehyde. The serine-isocitrate lyase
pathway is present in Methylosinus and Methylocystis and a number of
facultative C 1 utilizers.
2. The ribulose-monophosphate cycle. The key reactions of this cycle are
the condensation of ribulose-5-phosphate and formaldehyde by hexulose-
6-phosphate synthase and the isomerization of the product to fructose-6-
phosphate.
CH 2 0H CH2 0H CH 2 0H
I I I
c=o HOCH c=o
I I I
HCOH + CH 2 0 ----+ c=o ----+ HOCH
I I I
HCOH HCOH HCOH
I I I
CH2 0 HCOH HCOH
ribulose-5- I I
CH 2 0 CH 2 0
o-erythro- L-glycero-3- frUClose-6-
hexulose-6-
The acceptor-ribulose-5-phosphate-is regenerated through a series

of reactions depicted in Fig. 6.21. Three molecules of formaldehyde finally
end up in one molecule of dihydroxyacetonephosphate, which can be
utilized for biosynthetic purposes. The ribulose-monophosphate cycle
occurs in Methylococcus and Methylomonas species.
3eH zo
\ 3 hexulose-6-
~----------,
II aldolase
:I ....f - - - - fructose-6-
:
I
phosphatase
transketolase
: ...4 1 - - - - dihydroxyacetone-
I t
L
I reactions :
J ....4 1 - - - - 2 glyceraldehyde-
2 fructose-6-
2ATP~
2ADP"t1
I dihydroxyacetone- I 4
2 fructose- t, 6- @
Figure 6.21. The ribulose-monophosphate cycle. 1, Hexulose-6-phosphate syn-
thase; 2, hexulose-6-phosphate isomerase; 3, phosphofructokinase; 4, fructose-
1,6-bisphosphate aldolase.
In detail, the cycle operates as follows: Two of the three molecules of

fructose-6-phosphate are phosphorylated and the resulting fructose-l,6-
bisphosphate is cleaved by aldolase into dihydroxyacetonephosphate and
I xylulose-S- 1....41-----~ glyceraldehyde-3-
~~ fructose-6-
erythrose-4-
l'
dihydroxyacetone-
sedohePtulose-I,7-@
~P
sedoheptulose-7-
I xylulose-S- 1.. 4------.,~1---- glyceraldehyde-3-
Iribose-S- I..f--_~/l
Figure 6.22. The conversion of fructose-6-phosphate, glyceraldehyde-3-phosphate,
and dihydroxyacetonephosphate into pentose phosphate. 1, Transketolase; 2,
aldolase; 3, phosphatase.
Growth with C, Compounds 167
glyceraldehyde-3-phosphate. Two glyceraldehyde phosphate, one dihy-

droxyacetonephosphate, and one fructose-6-phosphate undergo a series of
reactions that resemble the reactions of the pentose phosphate cycle (see
Fig. 3.14). The difference is that the reaction sequence here is unidirec-
tional; it proceeds only in the direction: hexosephosphates ~
pentosephosphates, but not vice versa. The reason for this is that
sedoheptulose-7-phosphate is not formed from erythrose-4-phosphate in a
transaldolase reaction but in a true aldolase reaction with dihydroxyace-
tonephosphate as the second substrate (Fig. 6.22). Sedoheptulose-I,
7-bisphosphate is converted to the 7-phosphate by a phosphatase. This
reaction is irreversible and determines the direction of the entire sequence.
3. Xylulose-monophosphate cycle. In yeast, but not in bacteria, a third
formaldehyde fixation cycle has been found to occur (Fig. 6.23). Its key
reaction is a specific transketolase reaction by which the first two carbon
atoms of xylulose-5-phosphate are transferred to formaldehyde:
CH 2 0H
I
c=o
I
HOCH + CH 2 0 -->
I
HCOH
I
CH 2 0
Dihydroxyacetone is formed which is phosphorylated to dihydroxyacetone

phosphate by a specific kinase that can be regarded as the second key
I 3CH 0 I
2
3 xylulose-5- ------\-.....;:",,-l....'\:::-----~. 3 dihydroxyacetone
3 glyceraldehyde-
3ATP -+-+-_1
2
3ADP ~-+----'I
+
:1---------
- - - - - - - - - ~ .... ../ 3 dihydroxyacetone-
aldolase
phosphatase
I
:
/
3
/ / J
transketolase : : fructose-6- P;
I reactions I _ .
~ ~ ....- - - - - - - - - - - - - ; : ,::-.__--.L
dihydroxyacetone- 1
Figure 6.23. The xylulose-monophosphate cycle. 1, Dihydroxyacetone synthase; 2,
dihydroxyacetone kinase (triakinase); 3, fructose-1,6-bisphosphate aldolase +
fructose-1,6-bisphosphatase.
enzyme of this cycle. All further reactions are very much similar to the
ones of the ribulose-monophosphate cycle.
It should be mentioned that formaldehyde is a very suitable C 1
precursor for the synthesis of cellular material. Its redox grade is the same
as the one of sugar. Its conversion to glyceraldehyde-3-phosphate requires
2-4 ATP (depending on the pathway). For comparison, the fixation and
reduction of 3CO z to glyceraldehyde-3-phosphate requires 9ATP (see
Chapter 9).
A few additional remarks with respect to the methylotrophic yeasts are
necessary. The ability to grow on methanol is not very widespread among
yeasts. Only some Candida, Hansenula, and Torulopsis species are able to
do so. They employ the xylulose-monophosphate cycle for formaldehyde
fixation. It is also noteworthy that these yeasts do not contain a PQQ-
dependent methanol dehydrogenase. Instead, a methanol oxidase is acting
on methanol:
A catalase takes care of the HzO z formed in this reaction. Both enzymes,
methanol oxidase and catalase, reside in special organelles, the peroxi-
somes. Formaldehyde produced diffuses out of these peroxisomes and is
oxidized further in the cytoplasm.
B. Facultative methylotrophs
A number of microorganisms besides the obligate methylotrophs are able
to grow with C 1 compounds and more complex organic compounds.
Among these are yeasts, Hyphomicrobium species, pseudomonads such as
P. oxalaticus and P. AMI, and Protaminobacter species. These organisms
utilize methanol, formate, trimethylamine, or methylamine, but are unable
to grow with methane. Consequently, growth on methane is the domain of
the obligate methylotrophs.
Methylamines are important substrates in nature. Several enzyme
systems have been characterized that oxidize these compounds. A
flavoprotein acting on trimethylamine catalyses the following reaction:
The reduced flavoprotein is reoxidized by reaction with components of the

respiratory chain. Further oxidation of dimethylamine and monomethyla-
mine also yields formaldehyde as the principal product.
When growing with methanol, prokaryotic facultative methylotrophs
employ either the serine-isocitrate lyase pathway or the ribulose-
monophosphate cycle. Formate is assimilated by P. oxalaticus via COz and
the ribulose-I,5-bisphosphate cycle (Chapter 9). Yeasts use, as outlined,
the xylulose-monophosphate pathway.
Incomplete Oxidations 169
C. Methanol as substrate for the production of

single-cell protein
The use of methanol as substrate for single-cell protein production has
several advantages; it is cheap and available in high purity; it is water
soluble and allows faster growth of methylotrophs than methane; Or and
NADH-consuming monooxygenase reactions are not involved in substrate
oxidation. The commercial processes for the production of single-cell
protein are based on the use of Methylomonas or Melhylophilus species.
Yeasts play only a minor role.
VII. Incomplete Oxidations

During growth of aerobic heterotrophs, a part of the substrate is normally
oxidized to CO 2 and the remainder is used to synthesize cellular material.
In some organisms a complete oxidation of organic substrate is not possible
and partially oxidized compounds-usually organic acids-are formed
and excreted. The acetic acid bacteria, for instance, excrete large amounts
of acetic acid and decrease the pH of their environment to values below 4.
A. Acetic acid bacteria

Acetic acid bacteria are obligate aerobic organisms; they live on the
surfaces of plants and fruits. The members of the two genera Gluconobac-
ter and Acetobacter are collectively called acetic acid bacteria. G. oxydans
prefers glucose and other sugars and does not grow on ethanol. Glucose is
degraded by Gluconobacter via the pentose phosphate cycle (see
Fig. 5.14). The glyceraldehyde-3-phosphate formed is then converted via
pyruvate to acetate, which is excreted.
Organisms such as A. aceti and A. pasteurianus utilize ethanol and
oxidize it to acetic acid. A. aceti is especially important for the production
of vinegar. The energy metabolism of these organisms is very easy to
describe. Ethanol is oxidized with alcohol and acetaldehyde dehy-
drogenases to yield acetate and reducing power in the form of POOH 2 . It
seems now that POO is the preferred H-acceptor of these dehydrogenases
of acetic acid bacteria and not NAD + :
alcohol
CH3 -CH 2 0H + POO dehydrogenase, CHJ-CHO + POOH 2

aldehyde
CH 3 -CHO + POO + H 2 0 dehydrogenas\ CH 3 -COOH + POOH 2

The POOH 2 produced reacts then with a c-type cytochrome of the
respiratory chain, and electron transport gives rise to the generation of a
protonmotive force and finally to ATP synthesis.
Acetic acid is excreted because its activated form, acetyl-CoA, cannot

be oxidized via the tricarboxylic acid cycle, since in Gluconobacter a
distinct enzyme is lacking, the succinate dehydrogenase. Therefore, a-
oxoglutarate and succinyl-CoA (precursors of L-glutamate and porphyrins)
can be synthesized via reactions of the cycle, but an oxidation of acetyl-
CoA is not possible. Acetobacter species produce large amounts of acetic
acid from ethanol. The final acetic acid concentration may reach values of
0.4 M. These organisms are not totally unable to oxidize acetate via the
cycle. Strains belonging to A. aceti or A. pasteurianus establish a complete
tricarboxylic acid cycle if ethanol is no longer available. They then take up
acetate and oxidize it further.
When growing on ethanol, the acetic acid bacteria actually require
anaplerotic sequences that lead to PEP formation. Many of them grow
ethanol
PQQ
I biosynthesis
li ",0
'U=e===========b "7" /
excreted
"
oxaloacetate ~-).,../_-----.~,...-_.
.. pymvate
--<
,,-- '-, , ~ATP+Pi
" I cO 2
/ I
~alate citrate AMP+ PP i
succinate "--- \ IPEP I

~ isocitrate
I'o~",o", 1- ""'"'~ )
,,-oxoglutarate
I glutamate
Figure 6.24. Metabolic processes of A. aceti growing on ethanol in mineral media.
Conversion of acetate into acetyl-eoA is catalyzed by acetyl-CoA synthetase. The
activity of the tricarboxylic acid cycle is largely inhibited if ethanol is present. The
oxaloacetate formed is decarboxylated to pyruvate and then phosphorylated to
PEP.
only in complex media and apparently take up the precursors required for
biosynthesis. Others can live in mineral media; they induce the key
enzymes of the glyoxylate cycle during growth on ethanol. Part of the
net-synthesized oxaloacetate is decarboxylated to pyruvate by an oxaloace-
tate decarboxylase. Finally, PEP is produced from pyruvate by pyruvate:
phosphate dikinase. These metabolic relationships are summarized in
Fig. 6.24.
In addition to ethanol the acetic acid bacteria oxidize a large number of
other alcohols to the corresponding acids and ketones. Examples are:
propanol - propionate
isopropanol - acetone
glycerol - dihydroxyacetone
gluconate - 5-ketogluconate
Again, the organisms are unable to synthesize the catabolic enzymes for
the degradation of these substrates in high activity and excrete acids and
ketones. Of special interest is the oxidation of o-sorbitol by acetic acid
bacteria to L-sorbose. The latter is required in large amounts for the
synthesis of vitamin C (Fig. 6.25).
B. Bacilli
Most bacilli carry out incomplete oxidations when growing on carbohy-
drates. Under these conditions the synthesis of the enzymes of the
tricarboxylic acid cycle is repressed; the substrate is partly converted to
acetate, pyruvate, acetoin, and 2,3-butanediol. The pathway leading to the
CHO CH 20H H20H

I
H-C-OH
I
H-C-OH
T
H-C-OH
electrolytic
I I I
HO-C-H reduction G.oxydans
HO-C-H HO-C-H
I I I
H-C-OH H-C-OH H-C-OH
I I I
H-C-OH H-C-OH O=C
I I I
CH 20H CH 20H CH 2 0H
D-glucose D-sorbitol L-sorbose
",,0
TH20H
C=O
C-'-----,
I
I chemical oxidation HO-~ 0
HO-C-H
I
H-C-OH
of diacetone derivative HO- ~
HC
T
I I
HO-C-H HO-C-H
I I
CH 20H CH 20H
L-sorbose L-ascorbate
Figure 6.25. Synthesis of vitamin C. Electrolytic reduction of D-glucose yields
D-sorbitol, which is oxidized by G. oxydans to L-sorbose. Chemical oxidation leads
to formation of the vitamin (Curved arrow indicates that formula is turned around
by 180).
glucose
I
IF
2,3-butanediol pyruvate
co,~,
NAO <
3 NAOH+W
CO 2
CH 3- CO-CHOH-CH 3 ~--'---.:::_--
...... CH 3- f(OH)- COOH
2 CO-CH 3
acetoin
a-acetolactate
Figure 6.26. Formation of acetoin and 2,3-butanediol during growth of bacilli on
glucose. 1, a-Acetolactate synthase, a thiamine pyrophosphate containing enzyme;
2, a-acetolactate decarboxylase; 3, 2,3-butanediol dehydrogenase.
latter two products is depicted in Fig. 6.26. During sporulation the
enzymes of the tricarboxylic acid cycle are formed, and the excreted
products are taken up and oxidized. This oxidation provides the energy for
the sporulation process. Whereas acetate and pyruvate are fed into the
oxidation pathway via conventional routes, a special enzyme system is
necessary for the utilization of acetoin and 2,3-butanediol.
There is evidence that acetoin can be cleaved to acetate and acetal-
dehyde:
~.
x XH,
The natural electron acceptor (X) is not known.
C. Incomplete oxidations under stress conditions

Products such as o-gluconate, citrate, and L-glutamate are of great
economical interest, and processes have been developed in which certain
organisms are forced to excrete them in large amounts. Organisms and
conditions used in these productions are listed in Table 6.2. o-Glucose is
oxidized by Aspergillus niger in the presence of high O 2 concentrations to
o-gluconate. The enzyme glucose oxidase is a flavoprotein and transfers
the reducing equivalents to O 2 with H 2 0 2 formation. The latter is cleaved
by catalase:
o-glucose + O 2 ~ o-gluconolactone + H 2 0 2
o-gluconolactone + H 2 0 ~ o-gluconate
H 2 0 2 ~ H 2 0 + 1/202
Table 6.2. Organisms and conditions under which some incomplete oxidation
products of D-glucose or sucrose are excreted.
product organism conditions
D-gluconate Aspergillus niger high sucrose concentration; high O 2 con-

centration in the medium; pH approx. 4
citrate Aspergillus niger high sucrose concentration; pH < 3; sub-
optimal Fe 3 + concentration (with respect
to growth); Zn2+, Mn2+, Cu2+ in a cer-
tain concentration range
L-glutamate Corynebacterium biotin concentration < 5 JLg/l; addition of
glutamicum long-chain fatty acids; addition of penicil-
lin at the beginning of the production
phase
2NAD+
2NADH + H'
+
glucose
~ 2ATP
2ADP + 2P j
2 pyruvate
lJ
OASH
ADP + Pj A 1
P
/ ,NAD+ NADH + H'
~C02~~
oxaloacetate acetyl-SCoA
citrate
I
I I
I
I
I I
t
L
I
isocitrate
," , - - - - . - NADP+--------, '
/
,
.... , '
~ ..--- -NADPH + W .. - - - - / {
I
./'
;'
I .... ;' ......

NH 3 - ' ".... jt' CO 2
a-oxoglu tarate
Figure 6.27. Oxidation of glucose to citrate by A. niger and further conversion of
citrate to L-glutamate by C. glutamicum.
At low pH values the relative activities of the glycolytic enzymes and the
enzymes of the tricarboxylic acid cycle are such in A. niger that citrate is
accumulated and then excreted into the medium. It is a prerequisite for
citrate production from glucose that oxaloacetate is generated from PEP or
pyruvate by an anaplerotic reaction. The presence of an active pyruvate
carboxylase in A. niger has been demonstrated. The metabolic flux from
glucose to citrate is depicted in Fig. 6.27, which illustrates also the
situation in Corynebacterium glutamicum. Here, L-glutamate is the com-
pound that is accumulated. Its unhindered excretion requires partial
damage of the cytoplasmic membrane. This is achieved by affecting
phospholipid synthesis through biotin deficiency and an excess of saturated
fatty acids.
Incomplete oxidations are a very interesting property of microorgan-
isms. Their importance for the production of chemical compounds will
increase in the future.
VIII. Plasmid-Encoded Catabolic Activities

Pseudomonas putida and related species contain so-called degradative
plasmids. These are circular autonomous DNA molecules replicating
themselves independently of the bacterial genome and comprising the
genetic information for the synthesis of certain catabolic enzymes. Thus,
these enzymes cannot be synthesized by plasmid-free strains. This was first
shown by Gunsalus for enzymes involved in the degradation of camphor. A
number of degradative plasmids that occur in Pseudomonas species is listed
in Table 6.3. The CAM plasmid contains the information for all the
Table 6.3. Degradative plasmids
designation of end product of

plasmid compound degraded plasmid-encoded pathway
CAM camphor isobutyrate

OCT n-octane n-octanol
SAL salicylate acetaldehyde + pyruvate
NAH naphthalene acetaldehyde + pyruvate
TOL xylene/toluene acetaldehyde a + pyruvate
ASL 4-alkyl benzene
sulfonate acetaldehyde a + pyruvate
pJP4 2,4-dichlorophenoxyacetate succinate + acetyl-CoA
pAC21 p-chlorobiphenyl 4-chlorobenzoate
a In the case of xylene or the 4-alkyl benzene sulfonates higher alkanals

(propionaldehyde, etc.) are formed.
Plasmid-Encoded Catabolic Activities 175
I strain m t-2 I strain B-13 I
~I
YC~H y~I~"
f
strain WR-24 I
HOOQC
OH OH
~
I HOOQC
r
t
OH OH
1/ H ! 1/ H
~
I ./
Cl : // CI
I /"
6~'
/y~
0" ~/! QCOO~OOH
(, CHO A'
,..p-
I"
Cl I Cl
--- + .J1
j
acetate + succinate
[
c,6=o J COOH
t
3-oxoadipate
t ~n
O
COOH 6COOH
COOH .. vO-C=O
/ /
Figure 6.28. Degradation of 4-chlorobenzoate by strain WR-241 [H.-]. Knackmuss;
Forum Mikrobiol. 6,311-317 (1979)]. The first two enzymes stem from strain mt-2
and the remaining ones from strain B-13.
enzymes that convert camphor into isobutyrate. The gene for an-octane
oxygenase that requires reduced rubredoxin as cosubstrate (see Fig. 6.11)
and that is membrane-bound resides on the OCT plasmid. An alcohol
dehydrogenase is also encoded there; a second alcohol dehydrogenase,
however, is chromosomal as is the aldehyde dehydrogenase.
The SAL, NAH, TOL, and ASL plasmids contain the information for
the conversion of the corresponding substrates all the way to pyruvate and
acetaldehyde or a homologous aldehyde. First catechol or an alkylcatechol
is formed which is then cleaved by catechol-2,3-dioxygenase. It is interest-
ing that P. putida employs a chromosome-encoded ortho-cleavage pathway
when it grows on compounds such as phenol. With toluene as substrate,
however, this pathway is not induced, but the plasmid-encoded meta-
cleavage enzymes are formed. Herbicides such as 2,4-dichlorophenoxy-
acetate or p-chlorobiphenyl are also degradable with the aid of information
present on certain plasmids. Here, a modified ortho-cleavage pathway is
used.
An interesting Pseudomonas strain has been selected by Knackmus and
coworkers. It rapidly degrades 4-chlorobenzoate and originates from two
strains that do not grow at all on this compound (strain mt-2 and strain
B-13). As is depicted in Fig. 6.28, the conjugant strain WR-241 employs
the very active and unspecific benzoate dioxygenase from mt-2. In contrast
to the enzyme from B-13, which is highly specific and inactive on the
chlorinated substrate, it oxygenates 4-chlorobenzoate. The product cannot
be metabolized via the meta-cleavage pathway of mt-2 (dead-end product
formed) but it can via the ortho-cleavage pathway of B-13. Consequently,
the combination of plasmid-encoded information yielded an organism
(strain WR-241) with very much improved catabolic abilities.
It can be expected that certain catabolic activities in other bacterial
genera will also turn out to be associated with certain plasmids. Thus, the
information on plasmids is not only important for gene transfer and for
resistance against toxins but also for substrate breakdown.
IX. Summary
1. The polymers of dead organic material are hydrolyzed by exoen-
zymes, which are excreted by bacteria and fungi. a-Amylase cleaves starch
into dextrins and subsequently into maltose, glucose, and oligosaccharides.
Cellulase hydrolyzes cellulose to cellobiose. The disaccharides are taken
up by bacteria and cleaved to hexose-I-phosphate and hexose by specific
phosphorylases.
2. Proteases are formed by bacilli, pseudomonads, and many
anaerobes; DNases are excreted by hemolytic streptococci, Staphylococcus
aureus, and clostridia. Bacillus subtilis excretes a powerful RNase.
3. The degradation of many amino acids is initiated by their conversion
Summary 177
to the corresponding a-oxo acids. This is accomplished either by

cytochrome-linked oxidases or by NAD(P)+ -linked dehydrogenases and
transaminases. Serine and threonine are converted into a-oxo acids by
dehydratases. Many bacteria form aspartase, which removes ammonia
from aspartate to yield fumarate.
4. Fatty acids are degraded by f3-oxidation. Bacteria growing with
these compounds employ the glyoxylate cycle for the formation of C4 -
dicarboxylic acids.
Propionate is utilized either by the methylmalonyl-CoA pathway or by
the acrylyl-CoA pathway. Glyoxylate is a degradation product of the
purine bases; it is oxidized via the glycerate pathway and the tricarboxylic
acid cycle. Pseudomonas oxalaticus grows with oxalate, which is oxidized
to CO 2 via formate for ATP formation and reduced to glyoxylate for
synthesis of cell constituents.
5. Aliphatic hydrocarbons are oxidized at one terminal methyl group by
monooxygenases to the corresponding alcohols. The alcohols are further
oxidized to fatty acids by NAD+ -dependent dehydrogenases. Subterminal
attack leads first to a 2-hydroxyalkane which is further oxidized to an acetyl
ester and hydrolyzed to acetate and an alcohol.
6. The majority of aromatic compounds utilized by bacteria is first
converted into catechol and protocatechuate. Two kinds of cleavage
reactions for these compounds occur in bacteria: (1) ortho-cleavage, in
which catechol is oxidized to cis, cis-muconic acid; (2) meta-cleavage, in
which catechol is oxidized to 2-hydroxymuconic semialdehyde.
7. Obligate methylotrophs utilize methane and other C 1 compounds for
growth. Reducing power for the respiratory chain is formed by these
organisms by the oxidation of the C 1 compound to CO 2 , This involves a
methane-monooxygenase and a methoxathin-dependent methanol dehy-
drogenase. Formaldehyde is the starting material for biosynthesis. It is
assimilated either by the ribulose-monophosphate cycle or by the serine-
isocitrate lyase pathway. In methylotrophic yeasts a xylulose-
monophosphate cycle is present.
8. Acetic acid bacteria gain reducing power for the respiratory chain by
the oxidation of ethanol to acetate. The latter is excreted because of the
incompleteness or low capacity of the tricarboxylic acid cycle. During
growth, bacilli oxidize sugars to acetate, pyruvate, and 2,3-butanediol.
During the sporulation process these compounds serve as an energy source
and are oxidized to CO 2 , Compounds such as citrate, L-glutamate, or
o-gluconate are excreted by some organisms in large amounts under
conditions of stress.
9. Pseudomonas species contain so-called degradative plasmids. They
harbor the information for breakdown of camphor, n-octane, naphthaline,
or toluene. Combination of plasmid-encoded catabolic activities has
allowed to develop strains that degrade "so-far persistent" compounds
such as chlorobenzoate.
Chapter 7
Regulation of Bacterial
Metabolism
The discussion of the metabolism of aerobic hettrotrophs has shown that a

number of catabolic and anaplerotic enzymes are needed by microorgan-
isms only under certain growth conditions. ,B-Galactosidase is required by
E. coli during growth on lactose but not if glucose serves as substrate. A
Pseudomonas strain confronted with phenol in its environment requires
several specific enzymes in order to take advantage of this compound.
These enzymes are not needed if organic acids such as malate or succinate
were the substrates utilized by this microorganism. Thus, it is reasonable
and economical that organisms do not synthesize all the time all the
enzymes they are able to make but only those that are necessary for their
metabolism under current physiological conditions. This regulation of
enzyme synthesis is accomplished by induction and repression.
Alternatively, the activity of the enzymes present in the cell has to be
under control. Enzyme-catalyzed reactions must proceed in accordance
with the demands of the cell for energy and for cellular constituents. The
cell must, therefore, contain mechanisms to slow down or to speed up the
synthesis of a particular amino acid or the formation of ATP. This control
is accomplished by the ability of the cell to increase or to decrease the
activity of certain key enzymes of the metabolic pathways.
I. Regulation of Enzyme Synthesis by

Induction and Repression
A. Enzyme induction
Figure 7.1 shows the increase of the level of ,B-galactosidase in E. coli cells
following the addition of lactose to a culture medium. In the absence of
lactose the level of this enzyme in E. coli is so low that it is detectable only
Regulation of Enzyme Synthesis by Induction and Repression 179
addition of
lactose
time
Figure 7.1. Increase of the specific activity of f3-galactosidase in E. coli following

the addition of lactose to a cell suspension.
with special techniques. Lactose increases it by a factor of approximately
1,000. As early as 1.4 min after lactose addition an increase of enzyme
activity is already measurable. Enzyme activity reaches a plateau after
15-180 min depending on the growth conditions. This process of substrate-
mediated enzyme synthesis is called enzyme induction, and the compound
turning on enzyme synthesis is called an inducer. In the above system
this function is not fulfilled directly by lactose but by allolactose (0'-0-
galactosyl-I3-I,6-D-glucose), a compound derived from lactose (0'-0-
galactosyl-I3-I ,4-0-glucose) by l3-galactosidase.
We know already that more than one enzyme is usually required to
channel a substrate into the intermediary metabolism. For the breakdown
of lactose the permease, l3-galactosidase and the enzymes that convert
galactose into glucose-I-phosphate are necessary (see Chapter 4). For
growth on an aromatic compound Pseudomonas putida must synthesize
about 10 enzymes that are not present in cells growing on other substrates.
What is the inducer for all these enzyme syntheses? Two kinds of
mechanisms can be distinguished, coordinate and sequential induction. In
the first case, the growth substrate switches on the synthesis of all enzymes
required for its degradation. This is found for short catabolic sequences.
The enzymes of the Entner-Doudoroff pathway are formed by coordinate
induction, as well as the enzymes for the metabolism of arabinose. In the
presence of lactose a specific permease, l3-galactosidase, and a transacety-
lase are induced. Galactose formed in the l3-galactosidase reaction induces
the enzymes for its conversion into glucose-I-phosphate. Coordinate
induction is indicated when upon addition of the substrate all enzymes are
induced almost simultaneously, as shown in Fig. 7.2.
Sequential induction is found for long catabolic pathways that serve for
the degradation of several substrates.
180 7: Regulation of Bacterial Metabolism
substrate substrate
time
(a) (b)
Figure 7.2. Coordinate (a) and sequential (b) induction of enzymes a, b, and c.
If the substrates A, B, C, and D are catabolized as indicated, a few

enzymes are necessary for the breakdown of all four substrates, some are
required for two substrates, and some for only one-A, B, or C. Here
sequential induction lends itself as an economical principle. One of the best
studied examples of sequential induction is the formation of the enzymes
for the degradation of aromatic compounds. The inducers for the synthesis
of the enzymes of the 3-oxoadipate pathway in Pseudomonas putida are
given in Fig. 7.3. It is obvious that the two branches of this pathway are
regulated in a different manner. Protocatechuate is the inducer of pro-
tocatechuate oxygenase. The four subsequent enzymes are coordinatedly
induced by one of the terminal intermediates, 3-oxoadipate. In the
catechol branch, cis ,cis-muconate acts as productinducer of catechol
1,2-oxygenase and as substrate-inducer of muconate-lactonizing enzyme
and muconolactone isomerase. The fact that catechol l,2-oxygenase and
the terminal enzymes of this pathway are product-induced requires that
catechol be converted to cis ,cis-muconate and muconolactone to 3-
oxoadipate before the induction of the enzymes involved can be initiated.
In uninduced cells low levels of the corresponding enzymes are indeed
detectable (Table 7.1).
It should be noted that the kind of regulation of the 3-oxoadipate
pathway as elucidated in P. putida cannot be generalized. The investiga-
tions of Stanier and collaborators have shown that species-specific differ-
ences are very large. In Acinetobacter calcoaceticus, all enzymes for the
inducer enzyme metabolites enzyme inducer
O ,
IOH
OH
!
protocatechuate protocatechuate catechol cis, cis-
oxygenase oxygenase muconate
-02 C Y" CO i COi
~COi (, COi
(}-carboxy mucona le- muconate-
iactonizing laclonizing
enzyme enzyme
-02CtSC02
,
')'-carboxymuconolaclOne ""-
decarboxylase
0,
'"
C=O
G ,
0,
CO i
C=O
muconolactone
Isomerase
cis, cis-
muconate
3-oxoadipate ~COi
~c=O
!
4-oxoadipate enol-
lactone hydrolase
O~COi
~COi
!
3-oxoadipale succinyl-
CoA transferase
0
O~~-SCOA
~COi
3-oxoadipyl-CoA
j
not studied
thiolase
succinate and acetyl-CoA

Figure 7.3. The inducers for the enzymes of the 3-oxoadipate pathway in
Pseudomonas putida. [R. Y. Stanier and L. N. Omston. Adv. Microbial Physiol.
9, 89-151 (1973).]
degradation of protocatechuate are induced by that compound; cis,cis-

muconate induces all enzymes required if catechol is the substrate. Still
another regulatory map has been found for Alcaligenes eutrophus.
How does the inducer tell the cell to synthesize the enzymes of a
catabolic sequence? In Chapter 3 we discussed the mechanism of protein
synthesis. That glucose-grown E. coli cells do not contain the enzymes for
the catabolism of lactose and many other substrates means that the entire
bacterial genome is not blindly transcribed into RNA sequences and
Table 7.1. Levels of some enzymes of the 3-oxoadipate pathway in

uninduced and induced cells of Pseudomonas putidaa
specific activity
(D / g protein)
uninduced induced
enzyme (succinate) (benzoate)
catechol 1,2-oxygenase <0.2 1,090

cis,cis-muconate-lactonizing enzyme <0.2 390
muconolactone isomerase <20 2,200
4-oxoadipate enol-lactone hydrolase 30 1,490
a R. Y. Stanier and L. N. Omston. Adv. Microbial Physiol. 9, 89-151

(1973).
further translated into proteins. Gene expression is controlled by sophisti-

cated mechanisms. Several genes are always expressed in growing cells, for
bacteria contain a number of enzymes that have to be present under all
conditions (constitutive enzymes). Other genes are switched on or off just
as required in a particular physiological situation. For a number of
inducible enzymes, it is now known how this is accomplished. The best
studied example is the formation of the enzymes of lactose catabolism in
E. coli.
The three genes that code for lactose permease, ~-galactosidase, and the
transacetylase are located on the E. coli chromosome adjacent to one
another. Together with the promotor and the operator region they form a
functional unit, the lac operon. This unit is not transcribed in the absence
of lactose because a repressor blocks the operator region (Fig. 7.4a).
When lactose enters an E. coli cell some allolactose is formed by the action
of ~-galactosidase, which is present in uninduced cells in very small
amounts. Allolactose is bound by the repressor protein; and the latter is
thereby modified in such a way that it can no longer bind to the operator
region; transcription can proceed (Fig.7.4b). The lac repressor was
isolated by Gilbert and Muller-Hill, and it was shown in vitro that it binds
to the operator sequence of the lac operon and that this binding is
abolished by allolactose.
Presumably a number of inducible enzyme systems are regulated in a
similar way. Sequential induction is also understandable on the basis of the
operon model. The structural genes of the enzymes of a branched catabolic
pathway can be located on the chromosome in the form of two, three, or
more functional units which are switched on or off by the appropriate
inducers.
3=C DNA
S'~3~----, mRNA
prolein
a repressor I repressed lac operon I
DNA
5'--3' 5 ' - - - - - 3' mRNA
~ t " (
~.~
prolein
o
inducer
I !ransaCelYlaSe
1 permease
{3-galaclosidase
b induced lac operon
Figure 7.4. Induced synthesis of the enzymes for lactose catabolism. a: The product
of the gene I is a repressor protein that binds to the 0 region of the lac operon and
thus prevents transcription of this operon. b: The inducer (allolactose) binds to the
repressor, which thereby loses its affinity to the 0 region, mRNA is made, and the
three proteins are formed. P, promotor; 0, operator; Z, Y, and A, genes coding
for ,B-galactosidase, permease, and transacetylase. Note that the lengths in base
pairs of the structural genes and the PO region are very different and not identical
as could be deduced from the presentation of the DNA in this figure.
The lac operon is an example of a negative control mechanism. The

operon itself is ready to be transcribed, but transcription is prevented by a
specific repressor. This repressor has to be removed by the inducer. The
operons for the enzymes of arabinose, maltose, and rhamnose catabo-
lism-at least in E. coli-are in addition under positive control. The
operon has to combine with an activator-consisting of a specific protein
and the inducer-in order to be transcribed.
The metabolism of L-arabinose involves the expression of six genes.
Four of them, genes araA through araD, are located on the bacterial
chromosome adjacent to one another. The genes araE and araF are
located on the DNA, each at a different position, far away from the
afore-mentioned genes. AraE and araF code for two proteins that are
involved in the active transport of L-arabinose into the cell, and genes
araA, araB, and araD for the three enzymes that catalyze the conversion of
L-arabinose into D-xylulose-5-phosphate (see Chapter 4). All these five
genes together represent a so-called regulon. The gene product of araC is
the regulator protein. It is apparent from Fig. 7.5 that the gene araC is
separated from the araBAD operon by a control region. In the absence of
L-arabinose, araC is transcribed (from the right to the left in Fig. 7.5) and
protein PI is formed. This protein binds at the araO region and the
transcription of araBAD and also of E and F is prevented. Thus, protein PI
functions as repressor like the lac repressor. However, Engelsberg and
coworkers noted that mutants defective in the synthesis of the araC
product could not grow on L-arabinose. This is different from results
obtained with mutants that no longer produce the lac repressor; these
3'---5' mRNA
DNA
protein
protein PI
I repressed ara regulon I

3 ' - - - 5' mRNA
DNA
- - - - - - 3' 5' - - 3' 5 ' - - 3' mRNA
\) ~C!JtP @ (Jf) protein
r-
t t
1.1).".
L-arabinose
protein P2
permease binding
protein
isomerase
I activated ara regulon I

b
Figure 7.5. Induced synthesis of the enzymes for L-arabinose catabolism. a: The
product of the araC gene functions as repressor for the araBAD operon. b: The
inducer brings about a conformational change of protein PI to protein P2. It
functions as activator for all ara genes. Note that starting in the araOI region, one
DNA strand is transcribed to yield the araC mRNA and the other one to yield the
araBAD mRNA. (mRNA grows from the 5' end to the 3' end). Promotors are not
indicated in the figure.
mutants are, as expected, constitutive producers of {3-galactosidase, per-

mease and transacetylase. Engelsberg concluded that a modified gene
product, protein P2, functions as activator for the transcription of the
araBAD and EF genes (from the left to the right in Fig. 7.5). Thus, the ara
genes are under positive control of the protein P2.
B. Catabolite repression
In 1942 Monod discovered the phenomenon of diauxie. A culture of
Bacillus subtilis supplemented with glucose and arabinose as carbon and
energy sources utilized first glucose and then arabinose. This resulted in a
biphasic growth curve (Fig. 7.6). The same growth behavior is observed
for a great variety of substrate combinations utilized by B. subtilis, . coli,
and many other organisms. In all these cases one readily utilizable
substrate represses the utilization of other substrates, and, therefore,
Magasanik introduced the term catabolite repression for this phenomenon.
. coli, confronted with a mixture of glucose and lactose in its growth
medium, will grow first at the expense of glucose. The three structural
genes of the lac operon are not expressed during this period. Expression
begins when the concentration of glucose has become low. How can
glucose prevent lactose from initiating the formation of the lac enzymes?
80
~ 60
~
.;;;
""
"0
] 40
0-
o
20
4 6 8
hours
Figure 7.6. Diauxic growth of B. subtilis on glucose and arabinose. It was shown by
additional experiments that glucose was utilized during the first period of growth
and arabinose during the second. fJ. Monod. Recherches sur la croissance des
cultures bacteriennes (thesis 1942), Hermann, Paris, p. 145 (1958).]
This question was the subject of intensive research work in many labor-
atories. Finally it was found that catabolite repression is connected to the
level of cyclic AMP in the cells.
The utilization of glucose by E. coli leads to a tremendous decrease of
the intracellular concentration of cyclic AMP. After consumption of most
of the glucose the level of this compound increases again, and it could be
shown that cyclic AMP is required for the transcription of the lac operon.
It forms a complex with the so-called CRP protein (cyclic AMP receptor
protein) and this complex binds to the P region of the lac operon. Initiation
of transcription of the lac operon is possible only if this complex is present.
Thus the mechanism of enzyme induction as described in Fig. 7.4b is
incomplete; Fig. 7.7b accounts for the discovery of the involvement of
cyclic AMP. In accordance with this mechanism, catabolite repression of
glucose toward lactose is overcome if high concentrations of cyclic AMP
are added to the growth medium.
t-::r
NH 2
o_c~~No"
adenosine-3' ,5' -monophosphatc
(cyclic AMP)
N
H H
H H
o=p--o OH
I
OH
What makes the level of cyclic AMP decrease if glucose is being

metabolized? The level of cyclic AMP in the cells is determined by the
activities of the enzyme adenylate cyclase, which forms this compound
from ATP, and of phosphodiesterase, which hydrolyzes cyclic AMP to
AMP:
ATP adenyl ate cyclasc) cyclic AMP + PP j
cyclic AMP + H 0 phosphodiesterase, AMP

2
The phosphodiesterase activity of the cells is low and constant. However,

adenylate cyclase is membrane-bound, and its activity is high only if
components of the sugar transport (phosphotransferase system) are phos-
phorylated. This is the case in the absence of transportable sugars. In their
presence, the degree of phosphorylation decreases because of the phos-
phorylation of the sugar molecules entering the membrane. Adenylate
cyclase thus becomes less active; the level of cyclic AMP goes down, and
the cell is under catabolite repression.
It should be mentioned that the induction of the ara en7ymes is also
under control of catabolite repression. The CRP protein-cyclic AMP
complex binds here in the araD! region (see Fig. 7.5). Its presence there
5'--3' DCRP mRNA
<>
o ~~ct::::J
protein
inducer
lac operon under catabolite repression
JI~DNA
o .
5'--3' DCRP 5' 3' mRNA
+
~.~
cAMP
i .. t protein
I
o
inducer
t tansacetYlaSe
permease
l3-galactosidase
induced lac operon I

Figure 7.7. Induced synthesis of the enzymes for lactose catabolism. a: In the
absence of cyclic AMP, CRP is not bound at the promotor; mRNA is not made.
b: The CRP-cyclic AMP complex is bound at the promotor; the structural genes Z,
Y, and A are transcribed and the corresponding enzymes are synthesized.
is a prerequisite for the transcription of araC and consequently, in the

presence of L-arabinose, of the other ara genes.
The fact that glucose itself does not cause catabolite repression, but the
change in the concentration of a common metabolite, makes it under-
standable that this kind of repression is found for various substrate
combinations. In E. coli glucose represses the induction of several catabo-
lic pathways (for utilization of lactose, sorbitol, xylitol, arabinose, glycerol,
etc.). Fructose or glucose-6-phosphate can replace glucose. As a rule it can
be stated that substrates utilized rapidly by constitutive enzyme systems
cause catabolite repression of inducible pathways. Glucose and fructose do
not dominate in all microorganisms. In Clostridium tetanomorphum,
L-glutamate represses the formation of the enzymes for glucose utilization;
in Alcaligenes eutrophus, a hydrogen-oxidizing chemolithotroph, H 2 pre-
vents the induction of the Entp.er-Doudoroff pathway by fructose.
C. End product repression and attenuation

When E. coli grows in a minimal medium with glucose as carbon and
energy source, all the monomers for the formation of macromolecules have
to be synthesized along the pathways discussed in Chapter 3. Amino acids,
nucleotides, etc., are needed in correct amounts for polymer synthesis, and
an overproduction of some monomers has to be avoided. Sometimes
certain monomers are available from the environment of the bacteria. If,
for instance, histidine or tryptophan can be taken up, the biosynthesis of
these compounds is superfluous. Thus, for economical reasons it is
desirable that organisms are able to adjust the level of anabolic enzymes
such that this level is in correspondence with the needs for the products
synthesized. This adjustment is achieved by end production repression
and/ or attenuation. If tryptophan and histidine are added to a growing
E. coli culture, the cells stop the synthesis of the enzymes making histidine
from phosphoribosylpyrophosphate and tryptophan from chorismate.
For an understanding of this kind of regulation the finding was
important that the structural genes of several anabolic enzymes are also
located on the chromosome as operons. The nine genes responsible for the
enzymes of histidine biosynthesis in Salmonella typhimurium represent one
operon. The five genes coding for the enzymes that catalyze the synthesis
of tryptophan from chorismate also form one operon (Fig. 7.8a). Repres-
sion of enzyme synthesis is then understood as follows: when the product
5'--3' 5' 3' mRNA
.... protein
i
I
repressor
(inactive)
trYPt!Phan synthetase (A + B)
indolglycerol-P synthetase
an thranilate syn thase complex
trp operon under conditions of expression
a
]EC, ,L---l+;;;;;;+.L_---L_---'---_-'-_L---'-' DNA

mRNA
5'- ~ protein
lJ tryptophan
b trp operon under end product repression
Figure 7.8. The tryptophan operon and the mechanism of end product repression.
a: If tryptophan does not accumulate an inactive repressor is present; the structural
genes are transcribed and proteins are synthesized. The products of genes E and D
form the anthranilate synthase complex; gene C gives indolglycerolphosphate
synthetase and A and B, the two components of tryptophan synthetase. b: If
tryptophan accumulates an active repressor is formed and transcription is
repressed.
of a pathway accumulates in the cells, it combines with a repressor protein

to give an active repressor. The latter binds to the operator region and
prevents transcription of the operon (Fig. 7.8b).
Attenuation was discovered by Yanofsky and coworkers in studies on
the regulation of the trp operon. Here, this regulatory mechanism is
explained using the his operon as an example. It became evident through
the investigation of several mutants that this operon is not regulated by end
product repression. An aporepressor and a gene that codes for it could not
be detected. Nevertheless, transcription of the nine genes of the his operon
stops when histidine is present in excess. How are the genes switched off
without a repressor? The answer is, by attenuation. After RNA
polymerase has bound at the promotor site of the his operon, a so-called
leader sequence is transcribed (approximately 60 bases) and the attennator
is passed (approximately 140 bases) before the polymerase reaches the first
structural gene. The leader sequence codes for a leader peptide that,
interestingly enough, contains seven successive histidines. In the presence
of sufficient histidinyl-tRNA, a translating ribosome follows the transcrib-
ing RNA polymerase very closely, and it reaches a stop codon (VAG,
position 149-147 in Fig. 7.9a). Thereby it forces the attenuator RNA into
a terminator hairpin, and the RNA polymerase dissociates from the DNA.
Transcription of the structural genes does not take place. The situation at
limiting concentrations of histidine is depicted in Fig. 7.9b. Here, the
seven his-codons are a real barrier for the ribosome. Stalling of the
ribosome now has an effect on the conformation of the mRNA; a hairpin
conformation is already formed with participation of bases around the stop
codon of the leader peptide. As a result, the terminator structure cannot be
formed, RNA polymerase does not dissociate, and transcription of the
structural genes proceeds. Attenuation is not an "all or nothing" kind of
regulation. In the presence of histidine about 10% of the transcriptions are
successfully continued, at limiting histidine concentrations> 90%.
Attenuation is a very interesting regulatory mechanism for the synthesis
of enzymes involved in amino acid formation, because amino acid starva-
tion has an effect on the conformation of the mRNA similar to that caused
by excess of a certain amino acid. Translation cannot proceed at all, and
the formation of the attenuator stem is favoured again (Fig. 7.9c).
Besides the his operon a number of other amino acid operons have been
shown to be regulated by attenuation; the tryptophan operon (in addition
to regulation by end product repression), the operons for phenylalanine,
threonine, and branched-chain amino acid biosynthesis; and this list will
soon be longer.
End product repression or attenuation stop enzyme synthesis within a
short time because of the short half-life of messenger RNA. It should also
be noted that clustering of the genes of anabolic sequences on the
chromosome is not a prerequisite for regulation by end product repression.
In E. coli, the genes for the enzymes involved in arginine biosynthesis are
altenuafor
t
v A
A V
GC
A-V
n
U C A'V
V A G'C
~' r~
a
hi' stop 1 f AcE F
r-
1
,
,
I
: A:
, -- r
8
,--
,
I
C G'C
V'A
,-
V'A A'V
A-U A:.ll
CAVCAVCACCAVCAVCCVGACVAGVCVVVCAGGCGAVGVGVGC CVGGVG CG A G 'VVVVVVV
ribosome
~:~
r D AG'CG 6:g
i:
I I
AC
I I A G
I I G<
I I V'A
I I A'V
, I V G V G C GAG A
: I VGCGAA
n
A G G.C
b I G V V'C
i
tUG
Blf~ C
D
V'A
E
no attenuator
I
I
t
:
V'A
V'A
_A_ 0:~
I~:C
V'A
C'G
'--f' _
,
CACCACCAVCAVCACCAVCAVCCVGACVAG'CAVVCAG,CAVCVVCCGGGGGCVVVVVVVVV
I
I
allenualor
,,
I
I C
G1J V A
, A
V
C
V
A V
GC
C A'V V A'V
V C E A'V F
,~1{t :'0 II
GC 8
A V'A U A G'C
C'G
c VCG
C
A C' G
V'A
A'V
2:g v G'C
GC
G'C
V A
G'C
A'V
A'V V'A A' U A-U
AUGACACGCGUUCAAUUUAAACACCACCAUCAUC G---C'GVGGVGC,CAGA'UVVVVVV
ribosome
Figure 7.9. Dependence of the secondary structure of the histidine leader mRNA
on the position of the ribosome. a: The ribosome has completed the leader peptide,
the attenuator stem (E + F) is formed, and transcription is discontinued. b: The
ribosome activity is low because of a decrease of his-tRNA. Stems are formed by
stretches B + C and D + E. Transcription is continued. c: Because of a general
shortage in amino acids, protein synthesis is extremely slow. Stems are formed by
stretches A + B, C + D and E + F. The latter functions again as attenuator.
[Modified from H. M. Johnston, J. R. Roth, J. Mol. BioI. 145,735-756 (1981)].
located at various positions of the chromosome; they all are switched off by
the same repressor-arginine complex. The same could be true for attenua-
tion provided that each structural gene is preceded by a similar leader
sequence and an attenuator.
Regulation of enzyme synthesis becomes more difficult when we look at
branched anabolic pathways. If here one end product would completely
repress the formation of enzymes that also serve in the synthesis of other
monomers, this could result in a shortage of these compounds and a
stoppage of growth. An inspection of branched pathways with respect to

the control of enzyme synthesis reveals that two principles are applied in
order to avoid the above-mentioned difficulties (Fig. 7.10):
1. Isoenzymes are formed for common reactions so that each end product
has "its" enzyme whose synthesis can be regulated. E. coli forms three
aspartate kinases and two homoserine dehydrogenases, which are under
the control of different end products.
2. Repression of enzyme synthesis requires that all products synthesized
by these enzymes are present in excess. This kind of regulation was
previously called multivalent repression. Since the underlying mecha-
nism is attenuation it may now be called multivalent attenuation.
It is apparent from Fig. 7.10 that aspartate kinase I and homoserine
dehydrogenase I (hollow arrows) are subject to divalent attenuation by
L-threonine and L-isoleucine. This is reasonable. Since L-threonine is the
precursor of L-isoleucine, repression only by L-threonine could bring the
cell into difficulties. In this case the leading sequence contains successive
...---------v-:---]I F
dihydrodipi- .
. "'1 III II aspartate semlaldehyde
co Imate 6
m 3 t + i )--,--'----------.,
homoserine phosphate
T
tetrahydro- 4
dipicolinate
o-succinyl homoserine
T
N-succiny]--keto
O/-aminopimelate
cystathionine I
I
T I
N-succinyl-LL- I
diaminopimelate homocysteine t
ll-diaminopimelate
Figure 7.10. End product repression and attenuation in the pathway leading from
aspartate to the aspartate family of amino acids in E. coli. Enzymes indicated by
hollow arrows are subject to divalent attenuation by L-threonine + L-isoleucine.
L-Methionine represses the synthesis of the enzymes indicated by solid arrows and
L-Iysine those indicated by broken arrows. 1, Aspartate kinase; 2, aspartate
semialdehyde dehydrogenase; 3, homoserine dehydrogenase; 4, homoserine
kinase; 5, threonine synthase; 6, dihydrodipicolinate cyclohydrolase.
codons for both threonine and isoleucine. Ribosome stalling is, therefore,
still achieved if one aminoacyl-tRNA is present in excess. Only the
abundance of both tRNA derivatives allows rapid traveling of the ribo-
some and formation of the terminator hairpin. The level of aspartate
kinase II and of homoserine dehydrogenase II is controlled by L-
methionine and that of aspartate kinase III by L-Iysine. Here, end product
repression is the mechanism.
Another interesting branched pathway is the one for L-isoleucine,
L-valine, and L-Ieucine synthesis. Here it is necessary to remember that
each of the four steps involved in L-valine and L-isoleucine synthesis is
catalyzed by the same enzyme and that a-oxo-l3-methylbutyrate is not only
the precursor of L-valine but also the starting material for L-leucine
synthesis (Fig. 7.11). In this case a trivalent attenuation mechanism is
present in order to prevent a shortage of one of the three amino acids.
L-Leucine alone, via leu-tRNA, attenuates the synthesis of the enzymes
leading from a-oxo-l3-methylbutyrate to L-Ieucine.
The synthesis of the enzymes involved in aromatic amino acid formation
is also under control. E. coli contains three 3-deoxy-7-phospho-o-
arabinoheptulosonate synthases and two chorismate mutases. Regulation
of enzyme synthesis by the end products phenylalanine, tyrosine, and
o:-oxobutyrate
c"ru.."
o:-aceto-a-hydroxy- o:-acetolactate
butyrate
0:,
i +v +
/l-dihydroxy-
I
( o:,ll-dihydroxy-
/l-methylvalera te /l-methylbutyrate
o:-oxo-Il-methyl-
( o:-oxo-Il-methyl- a-isopropyl-
valerate butyrate malate
)(
5
6
/l-isopropyl-
malate
7
o:-oxoisocaproate
4
L - - - - - L - - - - - - - ' - - - - - - - - - - - - - - I L-Ieucine
Figure 7.11. Regulation of synthesis of the enzymes involved in L-isoieucine,

L-valine, and L-Ieucine formation in Salmonella typhimurium. i + v + I, trivalent
attenuation by the tRNA derivatives of L-isoleucine + L-valine + L-Ieucine; 1,
acetohydroxy acid synthase; 2, acetohydroxy acid isomero reductase; 3, dihydroxy
acid dehydratase; 4, transaminase; 5, a-isopropylmalate synthase; 6, isopropylma-
late isomerase; 7, l3-isopropylmalate dehydrogenase.
tryptophan is quite similar to that encountered for the pathway of the

aspartate family.
The regulatory maps discussed here have been elaborated for E. coli
and Salmonella typhimurium and they cannot be generalized. In fact,
comparative studies have shown that other solutions for the regulatory
problems in branched pathways have been developed and are employed.
Bacillus polymyxa and Rhodospirillum rubrum possess only one aspar-
tate kinase and one homoserine dehydrogenase; the synthesis of these
enzymes is regulated by multivalent repression or attenuation. Bacillus
subtilis contains one 3-deoxy-7-phospho-o-arabinoheptulosonate synthase
but two chorismate mutases. It can be concluded that the anabolic
pathways remained unchanged in the various microorganisms during
evolution but that the mode of control of these pathways has been modified
considerably.
D. Synthesis of enzymes of central pathways

The term induction is so much associated with catabolic enzymes and the
term repression with anabolic enzymes that it is necessary to remember
that the level of central enzymes also needs to be regulated. The
requirement for anaplerotic sequences changes with the substrate. When
E. coli is transferred from a glucose medium to an acetate medium,
isocitrate lyase and malate synthase are formed. For facultatively anaero-
bic bacteria the change from aerobic to anaerobic growth has consequences
with respect to the level of central enzymes. The complete tricarboxylic
acid cycle is not needed under anaerobic conditions, and the bacteria
stop to synthesize a-oxoglutarate dehydrogenase. Citrate synthase, cis-
aconitase, and isocitrate dehydrogenase are then required only for gluta-
mate synthesis; their level in the cells is, therefore, lower than under
aerobic conditions (Table 7.2). The synthesis of the pyruvate dehy-
drogenase multienzyme complex is completely repressed under anaerobic
Table 7.2. Level of tricarboxylic acid cycle enzymes in E. coli grown

aerobically or anaerobically with glucose a
specific activity (D /g protein)
enzyme aerobic growth anaerobic growth
citrate synthase 51.1 10.5

cis-aconitase 317 16.1
isocitrate dehydrogenase 1,416 138
a-oxoglutarate dehydrogenase 17.4 o
C. T. Gray, J. W. T. Wimpenny, and M. R. Mossman, Biochim. Biophys. Acta
I17, 33-41 (1966).
conditions. Instead, pyruvate-formate lyase is produced and activated (see

Chapter 8). The synthesis of a number of "anaerobic enzymes" requires
the synthesis of a regulatory protein that is the product of the fnr gene. The
formation of this protein is under redox control; it is formed only in the
absence of 02' If present, it promotes the transcription of the genes for
nitrate reductase, fumarate reductase, and formate-hydrogen lyase (a
complex of formate dehydrogenase II and hydrogenase). The degree of
expression of these genes depends on the redox potential of the terminal
electron acceptor present: no expression, as already said, in the presence
of 02; expression of only nitrate reductase in the presence of nitrate;
almost full or full expression of all enzymes in the presence of fumarate or
in the absence of both nitrate and fumarate. It should be mentioned that
E. coli forms a second formate dehydrogenase (FDH r) under anaerobic
conditions; it is linked to nitrate reductase or fumarate reductase (see
Chapter 8). This enzyme is not under redox control-as is the one that is
part of formate-hydrogen lyase-the formation of FDH r is induced by
formate.
II. Regulation of Enzyme Activity
A. Feedback inhibition
In addition to the regulation of the level of enzymes the cell must have
devices to adjust the activity of enzymes to the metabolic requirements. If
a monomer is synthesized in larger amounts than is needed for polymer
synthesis, it is not only necessary to stop the synthesis of the enzymes
involved but also to reduce immediately the synthesis of that monomer.
This is accomplished by feedback inhibition; the end product-when
accumulating in the cell-inhibits the activity of the first enzyme involved
specifically in its formation. If isoleucine is synthesized in excess, it inhibits
the activity of the first enzyme of its biosynthetic pathway, threonine
deaminase:
threonine -------> a-oxobutyrate - - - - isoleucine
t I
I I
I I
--------------------------------
I I
If pyrimidine nucleotides accumulate in the cell, CTP inhibits the

activity of the first enzyme of their biosynthetic pathway, aspartate
transcarbamoylase:
aspartate carbamoyl
+
carbamoyl
7 I
aspartate - - - - - - CTP
I
J
phosphate ~
Regulation of Enzyme Activity 195
As soon as threonine deaminase and aspartate transcarbamoylase are

inhibited, the subsequent enzymes of these pathways run out of substrates,
and monomer synthesis stops. It is now established that the substrate flow
through most biosynthetic pathways is regulated by feedback inhibition.
As it is in the regulation of the enzyme level, branched pathways are more
difficult to be controlled in their activity than unbranched pathways. Figure
7.12 summarizes the inhibitory effects of the amino acids derived from
aspartate. The activities of aspartate kinase I and of homoserine dehy-
drogenase I are controlled by L-threonine. Aspartate kinase III activity is
regulated by L-lysine. Interestingly, the isoenzyme II is not subject to
feedback inhibition by L-methionine; the latter inhibits the first enzyme
specifically involved in methionine synthesis (O-succinylhomoserine
synthetase) .
Figure 7.13 summarizes the control of enzyme activity in the aromatic
amino acid pathway of E. coli. One DAHP synthase is inhibited by
L-phenylalanine and the second one by L-tyrosine. The third enzyme,
which exhibits the lowest activity of the three enzymes, is not inhibited by
tryptophan. The latter, however, inhibits anthranilate synthase. The two
chorismate mutases are also under the control of the corresponding end
products. Most enterobacteria contain three DAHP synthases, and these
enzymes are regulated in a similar fashion. In Bacillus species, there is only
one synthase present; it is not inhibited by the end products of the
pathway but by the branch point intermediates chorismate or prephenate.
aspa rtate-4-phosphate
t
t
dihydrodipi- '411 11111111111111111111111 aspartate semialdehyde
colinate t >------~
homoserine =C>
m
O-succinylhomoserine
t
t O'-oxobutyrate
t
L-methionine f------'
t
t
Figure 7.12. Feedback inhibition in the pathway leading from aspartate to the
aspartate family of amino acids in E. coli. L-1soleucine inhibits threonine deami-
nase; L-threonine inhibits aspartate kinase 1 and homoserine dehydrogenase 1;
L-methionine inhibits O-succinylhomoserine synthetase; L-lysine inhibits aspar-
tate kinase III and dihydrodipicolinate synthetase.
phosphoenolpyruvate
+
eryth rose-4-phosphate
...
p )--------,
3-deoxy-u-arabinoheptulosonate-7-
(DAHPj
t
t
t
t
t
t
chorismate
- prephcnate
Figure 7.13. Feedback inhibition in the aromatic amino acid pathway of E. coli.
L-Phenylalanine inhibits DAHP synthase I and chorismate mutase I; L-tyrosine
inhibits DAHP synthase II and chorismate mutase II; L-tryptophan inhibits
anthranilate synthase.
This type of control of activity is called sequential feedback inhibition. Still

another pattern of regulation of the aromatic amino acid pathway was
found in hydrogen-oxidizing bacteria such as Alcaligenes eutrophus. Here
the DAHP synthase is subject to cumulative inhibition by phenylalanine
and tyrosine. Table 7.3 gives the percent inhibition observed with phenyla-
lanine, tyrosine, and with the mixture of these amino acids. The inhibition
rates are not additive. The residual activity observed in the presence of the
mixture is equal to the product of the residual activities observed with
phenylalanine and tyrosine. This is characteristic for cumulative inhibition.
From the examples of feedback inhibition given it is evident that in most
cases the inhibitor differs significantly in its chemical structure from the
substrates of the enzyme it acts upon. Carbamoyl phosphate and aspartate,
the substrates of aspartate transcarbamoylase, are very different from the
inhibitor CTP. So it is very unlikely that CTP competes with the substrates
of aspartate transcarbamoylase for the active site of the enzyme as does
malonate with succinate at succinate dehydrogenase. Since the inhibitors
are not steric analogues of the substrates, enzymes must contain special
binding sites for them. Monod, Changeux, and Jacob coined the term
allosteric sites for these areas on the enzymes. Compounds that are bound
at these sites and that alter the activity of the enzymes are called allosteric
Table 7.3. Regulation of DAHP synthase activity of Alcaligenes

eutrophus by cumulative inhibition a
residual activity
amino acid present in the (100% = 1)
DAHP synthetase assay inhibition
mixture (%) observed calculated
phenylalanine 25 0.75 0.75

x
tyrosine 47.5 0.525 0.525
--
mixture 58.2 00418 0.394
OR. A. Jensen, D. S. Nasser, and E. W. Nester, J. Bacterial. 94, 1582-1593

(1967).
effectors and the enzymes subject to control by these effectors are called
allosteric enzymes.
B. Properties of allosteric enzymes

Enzymes in general have active sites; these are clefts or crevices to which
the substrate molecules are bound. Then the enzyme-catalyzed reaction
takes place. The topology of the active site may be such that the substrate
fits into it like a key into a lock (Fig. 7.14a). This is not true for all
enzymes. Many of them gain the complementary shape through the
interaction with the substrate (induced-fit model of Koshland; Fig. 7.14b).
This model implies that the active site of enzymes is not a rigid structure
and that it is subject to alteration. It can exist in two forms which represent
a low-affinity and a high-affinity state of the enzyme. This is the basis for an
understanding of the properties of allosteric enzymes.
Allosteric enzymes are oligomers; they consist of 2, 4, 6, or more
subunits which may be identical or not. One of the most frequent
characteristics of allosteric enzymes is their anomalous kinetic behavior. If
the reaction rate is plotted against the substrate concentration, a sigmoid
curve is obtained and not a hyperbola as with enzymes that follow
Michaelis-Menten kinetics. This is shown for aspartate transcarbamoylase
in Fig. 7.15. CTP is a negative effector of this enzyme and it can be seen
that at intermediate concentrations of aspartate the decrease in velocity
caused by 0.2 mM CTP is considerable.
The sigmoidal response of the reaction velocity to an increase of the
substrate concentration is explained by cooperative binding. In the absence
of substrate, the active sites of a tetrameric enzyme are assumed to exist in
forms that have little catalytic activity. However, through interaction with
one substrate molecule, one active site is assumed to be forced into a
conformational state that is catalytically active and brings about the
enzymatic reaction.
lock-and-key model
induced-fit model
c
<="",:==~"": "'" =
<:::xl
. ~
. PC:>
<:::xl
T-state <:::xl
low affinity R-state
high affinity
concerted symmetry model
effect of negative effector
Figure 7.14. Active site models. a: The active site is rigid. b: The high-affinity active
site is formed through interaction with the substrate. c: Cooperative conformation-
al changes in a tetrameric enzyme are caused by binding of one substrate molecule.
d: Conformational changes are prevented by a negative effector (arrow). Small fish,
substrate molecules.
The increase of enzyme activity by cooperative binding is then brought

about by the substrate-loaded subunit that, by conformational changes,
forces the active sites of the other subunits from a low-affinity into a
high-affinity state (Fig. 7.14c). Accordingly, a negative effector (CTP in
Fig. 7.15) bound at the allosteric site of one subunit not only keeps the
active site of this subunit in the low-affinity state but also prevents the
transition of the sites of the other subunits of the enzyme molecule from
the low-affinity to the high-affinity state (Fig. 7.14d). A positive effector
has the opposite effect.
4.0
;:-
u
o
"> 2.0
o 5.0 10.0 15.0

concentration of asparta te (mM)
Figure 7.15. Activity of aspartate transcarbamoylase as a function of the concentra-

tion of aspartate. [J. C. Gerhart and A. B. Pardee, 1. Bioi. Chern. 237, &91-896
(1962).]
Two models for cooperative binding have been proposed: the sequential
interaction model of Koshland and co-workers and the concerted symmetry
model of Monod and co-workers. The latter model assumes that the
oligomeric enzyme exists as an equilibrium mixture of two forms: the T
state in which all subunits are in the low-affinity form and the R state in
which all subunits are in the high-affinity form (Fig. 7.14c). Effectors then
change the equilibrium between these states. The sequential model of
Koshland allows also intermediate states of the enzyme in which, for
instance, two subunits of a tetrameric enzyme have a high affinity for the
substrate and two have a low affinity.
increasing affinity
sequential interaction model
A posItIve effector would then increase the number of subunits in the

high-affinity state and a negative effector would decrease this number.
C. Allosteric control of central pathways

The main objectives of catabolic and central pathways are to provide the
cell with energy and with starting material for biosynthesis, and it is,
therefore, reasonable that the regulatory signals used here for control are
ultimate products of the energy metabolism and central precursors of the
biosynthetic metabolism. Table 7.4 summarizes some allosteric enzymes
involved in central pathways of E. coli and their inhibitors and activators.
An increase of the NADH concentration in the cells signals that the
Table 7.4. Allosteric enzymes involved in central pathways of E. coli a
enzyme inhibitor activator
ADP-glucose pyrophosphorylase AMP pyruvate, F-6-P, F-P 2 ,b

fructose bisphosphatase AMP
phosphofructokinase PEP ADP, GDP
pyruvate kinase F-P 2
pyruvate dehydrogenase NADH, PEP, AMP, GDP
acetyl-CoA
PEP carboxylase aspartate, acetyl-CoA, F-P2 , GTP,
malate CDP
citrate synthase NADH, (l'-
oxoglutarate
malate dehydrogenase NADH
aBo D. Sanwal, Bacterial. Rev. 34, 20-39 (1970).

bF-P2 , fructose-l,6-bisphosphate; F6P, fructose-6-phosphate.
respiratory chain is saturated with NADH and that the tricarboxylic acid
cycle may slow down. Consequently, citrate synthase and also malate
dehydtogenase and the pyruvate dehydrogenase complex are subject to
inhibition by NADH. Moreover, citrate synthase is inhibited by a-
oxoglutarate and the pyruvate dehydrogenase complex by acetyl-CoA.
Not all bacterial citrate synthases are inhibited by NADH. The NADH
inhibitable type of enzyme is found mainly in Gram-negative bacteria;
Gram-positive bacteria contain a synthase that is inhibited by ATP like the
enzyme of eukaryotic organisms. a-Oxoglutarate functions as inhibitor in
enterobacteria only.
PEP carboxylase is inhibited by aspartate and malate. A high level of
the latter compounds signals that C4 -dicarboxylic acids need not to be
synthesized. An increase of the acetyl-CoA concentration, on the other
hand, might indicate a shortage of C4 -dicarboxylic acids. Thus acetyl-CoA
is an activator of PEP carboxylase. Many organisms contain pyruvate
carboxylase instead of PEP carboxylase as anaplerotic enzyme. The
pyruvate carboxylase from most sources is also activated by acetyl-CoA.
Fructose-l,6-bisphosphate is a strategic branch point of glycolysis and of
glycogen formation and its intracellular level is under regulatory control.
An increased AMP concentration, signaling ATP deficiency, results in an
inhibition of glycogen formation, because ADP-glucose pyrophosphory-
lase and fructose-bisphosphatase are inhibited by AMP. Carbohydrates in
excess lead to an increase of the fructose-l ,6-bisphosphate level. This has a
positive effect on glycolysis, as both pyruvate kinase and PEP carboxylase
are activated by fructose-l,6-bisphosphate. A sufficient supply of ATP in
the cell is signaled by an elevated PEP level. As a result phosphofructo-
kinase is inhibited and glycogen formation is favored by activation of
ADP-glucose pyrophosphorylase.
Figure 7.16 summarizes the control of glycolytic and glycogenic reac-

tions in E. coli. It can be seen that the target enzymes for the control of
reaction sequences usually catalyze reactions that are irreversible under
physiological conditions. These enzymes (phosphofructokinase, pyruvate
kinase, etc.) function as pacemakers, and regulation is very efficient at
these points. Furthermore, it is apparent from Fig. 7.16 that antagonistic
enzymes occur in the cells simultaneously, e.g., phosphofructokinase and
fructose-1,6-bisphosphatase. These enzymes particularly have to be under
stringent control, otherwise futile cycles would be established, which bring
about the hydrolysis of ATP as net reaction: fructose-6-phosphate is
phosphorylated by phosphofructokinase, and the bisphosphate is hydro-
lyzed again by the phosphatase. A futile cycle could also be established
fructose-6-phospha te
PEP----[~ CEJ---AMP
ADP----(
H
H
H
1 I = l ' - - - NADH, a-oxoglutarate
Figure 7.16. Schematic representation of glycolytic and glycogenic reactions in E.

coli and their control. F-P2' fructose-l,6-bisphosphate; B-, inhibitor; (f)-,
activator.
with glucose-I-phosphate, ADP-glucose, and glycogen as participating

metabolites.
Figure 7.16 also shows that adenine nucleotides are important effectors.
AMP is formed in many biosynthetic reactions from ATP and so is ADP.
Any increase of the concentration of these nucleotides leads to a stimula-
tion of ATP-yielding reactions. By regulation of ATP-producing and
consuming reactions organisms try to maintain a constant energy status.
According to Atkinson this energy status can be described by the energy
charge, which is defined as:
[ATP] + i[ADP]
ec = ----''----=------::-=-----=---
[ATP] + [ADP] + [AMP]
Systems containing only ATP have an energy charge of 1 and for those
containing only AMP the ec is zero. Measurements have shown that the
energy charge of growing organisms is about 0.8. E. coli cells die at ec
values below 0.5.
D. Covalent modification of enzymes

In recent years the importance of another mechanism for the regulation of
enzyme activity has been recognized more and more: modulation of
enzyme activity by covalent modification of enzymes. Whereas in allosteric
regulation a low-molecular-weight compound (metabolite) is bound to or
released from the enzyme, in the type of regulation discussed now the
enzyme is covalently modified in an enzyme-catalyzed reaction. The
principle of this regulatory mechanism is evident from the following
equations:
enzyme - X -modifying - - - , enzyme + X
- - - enzyme(s) (7.1)
(active) ( (inactive)
or
enzyme + X -----~)
modifying enzyme(s)
enzyme - X (7.2)
(active) ( (inactive)
Equation 7.1 describes enzyme systems that are active only when substi-
tuted with X. Removal of X yields an inactive form of the enzyme.
Enzymes are also known that are inactivated by substitution (Eq. 7.2).
The first enzyme found by Cori and co-workers to exist in two forms was
muscle glycogen phosphorylase. Form b is virtually inactive; it is present in
resting aerobic muscle. When glycogenolysis becomes necessary in work-
ing muscle, form b of phosphorylase is phosphorylated and converted
thereby into the highly active form a. At the expense of ATP, one
phosphoryl group is linked to a serine residue per subunit. Conversion of
form a into form b again is achieved by the hydrolytic removal of the
phosphate groups. Thus for the modification of phosphorylase two en-
zymes are required: phosphorylase kinase and phosphorylase phospha-

tase. These regulatory enzymes, of course, also have to be under control,
so that a highly sensitive cascade-like mechanism is responsible for the
regulation of muscle phosphorylase.
Here it is necessary to mention that the glycogen phosphorylase of
prokaryotes is not regulated by covalent modification and that enzyme
systems modified by phosphorylation/dephosphorylation are generally
more common among eukaryotes. A bacterial enzyme system regulated by
phosphorylation/ dephosphorylation is the NADP+ -isocitrate dehydro-
genase of E. coli and Salmonella typhimurium. It was observed that
addition of acetate to a culture containing limiting glucose resulted in rapid
inactivation of isocitrate dehydrogenase. Later, this inactivation was
shown to be due to the phosphorylation of the enzyme. One serine residue
per subunit is phosphorylated; this brings about the inactivation of the
enzyme:
4ATP 4ADP
isocitrate ~inasz isocitrate
dehydrogenase <===::::;;;:::=:~~==~) dehydrogenase-4 (7.3)
(active) ~hosPha~ (inactive)
4P j 4H zO
There is evidence that the isocitrate dehydrogenase kinase and the isocitrate
dehydrogenase phosphatase are activities that are associated with the same
protein. Why is isocitrate dehydrogenase inactivated on the addition of
acetate to the culture? Isocitrate is a branch point between the tricarboxy-
lic acid cycle and the glyoxylate cycle. Inactivation of isocitrate dehy-
drogenase brings the glyoxylate cycle more into play and favors oxaloacetate
and PEP formation.
Two other interesting systems of covalent modification have been
observed in bacteria: glutamine synthetase of E. coli and many other
microorganisms is regulated by adenylylation/ deadenylylation and citrate
lyase of Rhodopseudomonas gelatinosa is regulated by acetyla-
tion/deacetylation. Figure 7.17 summarizes the control of glutamine
synthetase as unraveled by Stadtman, Holzer, and their co-workers.
This enzyme consists of 12 subunits. The deadenylylated form is very
active; at high levels of glutamine and low levels of a-oxoglutarate the
enzyme is adenylylated by the action of ATP at one tyrosine residue per
subunit. The resulting enzyme form exhibits low activity and, in addition,
is subject to cumulative feedback inhibition by AMP, CMP, tryptophan,
histidine, glucosamine-6-phosphate, and carbamoyl phosphate. In case of a
shortage of glutamine the AMP residues are transferred to inorganic
phosphate and the active enzyme is restored again. The interconversion of
the two forms of glutamine synthetase is catalyzed by an adenylyltrans-
ferase and a regulatory protein (PIIA) which can be uridylylated by UTP.
In the presence of the uridylylated protein (PIID-UMP), the transferase
glutamine
synthetase
most active form
a-oxoglutarate
~
l2ATP 12ADP
PIIA PIID-UMP
12PP j
~, glu tamine'----..l
12P j
glutamine
synthetase-(AMP)12
less active form
Figure 7.17. Regulation of glutamine synthetase in E. coli by covalent modification.

The interconversion is catalyzed by an adenylyltransferase (ATase). Together with
the protein PIIA the ATase exhibits adenylylating activity. When ammonia is low
and glutamine is low, but a-oxoglutarate high, PIIA is uridylylated to yield
PIID-UMP. In its presence the ATase catalyzes the deadenylylation of glutamine
synthetase. UTase is uridylyltransferase and UR is uridylylremoving enzyme. Both
proteins are probably identical.
catalyzes the deadenylylation of the synthetase; otherwise, adenylylation is

favored. The entire system is shifted by ammonia (glutamine) limitation
toward PIID-UMP formation and activation of glutamine synthetase. At
high ammonia (glutamine) concentrations PIIA is formed and the syn-
thetase is inactivated.
Citrate lyase of the phototrophic bacterium Rhodopseudomonas gelati-
nosa has been found to be regulated by acetylation/deacetylation
(Fig. 7.18). Citrate lyase cleaves citrate to oxaloacetate and acetate.
Active enzyme is present if citrate is available to the cells as substrate.
Under these conditions the deacetylating enzyme is inhibited by L-
glutamate for which citrate serves as precursor. In the absence of citrate,
the glutamate concentration in the cells decreases and citrate lyase is
inactivated by a specific deacetylase. If citrate becomes available again to
the cells, the active enzyme is formed by acetylation. The citrate lyase
system differs from the other systems discussed in that the substituent (the
acetyl group) participates in the enzymatic reaction. The mechanism of the
lyase reaction according to Eggerer and co-workers is as follows:
acetyl-S-enzyme + citrate ( citryl-S-enzyme + acetate
citryl-S-enzyme ( Iacetyl-S-enzyme + oxaloacetate
The acetyl group participates in the initial transferase reaction and it is
thus apparent that the HS-enzyme must be inactive.
6AMP+ 6PP j
=+===.~ ligase
inactive
6 acetate 6ATP + 6 acetate
Figure 7.18. Regulation of citrate lyase of Rhodopseudomonas gelatinosa by

acetylation and deacetylation. Citrate lyase deacetylase is inhibited by glutamate.
Citrate lyase ligase occurs in an active and in an inactive form. L-Glutamate is
an inhibitor of the deacetylase. The mechanism of ligase activation is unknown
[G. Gottschalk, F. Giffhorn, and G. Antranikian, Biochem. Soc. Trans. 10,324-
326 (1982)].
Why are several enzymes regulated by covalent modification? Its

advantage is that relatively small changes of the intracellular concentra-
tion of a certain metabolite can provoke either full activation or inactiva-
tion to zero activity or to a very low activity. To give an example: Citrate
lyase is a catabolic enzyme; its presence in an active form is useful only if
citrate is available to the organism in substrate amounts. In the absence of
extracellular citrate, it has to be inactivated, because it would interfere
with L-glutamate synthesis via citrate and would create a futile cycle
together with citrate synthase. Therefore, it makes sense that the citrate
lyase-inactivating enzyme, the deacetylase, becomes active at low in-
tracellular L-glutamate concentrations.
It should be mentioned that some other regulatory mechanisms show
resemblances to the one discussed here. Enzyme activity may be regulated
by dissociation/association. An example is the lactate dehydrogenase of
certain lactic acid bacteria and clostridia. The enzyme is inactive as a
dimer. In the presence of fructose-l ,6-bisphosphate, the dimers associate
to enzymatically active tetramers. Lactate is, therefore, produced only if
there is a build-up of fructose-l ,6-bisphosphate in the organisms. Another
regulatory system was described by Buchanan and co-workers for chloro-
plasts; certain enzymes are activated in the light and inactivated in the dark
by oxidation. The light-driven electron transport to ferredoxin leads to an
increase of the reduced ferredoxin: oxidized ferredoxin ratio. Under
catalysis of the enzyme ferredoxin-thioredoxin reductase this affects the
level of reduced thioredoxin, which is required for the activation of certain

enzymes, e.g., the fructose-1,6-bisphosphatase of chloroplasts. Since
thioredoxin is present in phototrophic bacteria and also in facultative and
strict anaerobes an analogous regulatory mechanism might also be impor-
tant for bacteria.
III. Summary
1. The process of substrate-mediated enzyme synthesis is termed en-
zyme induction. In coordinate induction all enzymes of a pathway are
under the control of one inducer. In sequential induction several inducers
are involved.
2. The genes of inducible enzymes which are coordinately formed are
located on the bacterial genome adjacent to one another. Together with an
operator and a promotor they form a regulatory unit, the operon. The
transcription of operons under negative control (e.g., lac operon) is
repressed by a repressor protein; the inducer prevents binding of the
repressor. Transcription of operons under positive control (e.g., ara
operon) requires binding of activator protein at the DNA.
3. The phenomenon that readily utilizable substrates prevent the
induction of the enzymes of other catabolic pathways is called catabolite
repression. Transport of these substrates into the cells causes a drastic
decrease of the intracellular level of cyclic AMP. The latter, however, is
required for the initiation of RNA synthesis at operons that code for
inducible pathways.
4. The level of the enzymes of anabolic pathways is regulated by end
product repression or attenuation. An end product, which accumulates in
the cell, combines with a specific aporepressor to yield an active repressor.
The latter prevents transcription of the genes involved in the synthesis of
the enzymes of this particular pathway. In attenuation, the RNA
polymerase dissociates again from the DNA, if the product amino acid of
the anabolic pathway to be transcribed is present in excess. Regulation of
the enzyme levels in branched pathways is difficult; isoenzymes and
multivalent attenuation help to avoid that one end product interferes with
the synthesis of another end product of the same pathway.
5. In feedback inhibition, the end product of an anabolic pathway
decreases the activity of the first enzyme specifically involved in its
formation. This allows the rapid adjustment of the rate of product
synthesis to the cellular demands. The target enzymes for feedback
inhibition are allosteric enzymes. Besides their active sites, allosteric
enzymes contain specific binding sites for their effectors (inhibitors or
activators). Most allosteric enzymes show a sigmoidal response of the
reaction velocity to an increase of the substrate concentration. This is
explained by cooperative binding of the substrate and/or the effectors.
Summary 207
6. Regulatory signals used for the control of the activIty of central

pathways are products of energy metabolism (AMP, AD P, ATP, NADH)
and important precursors of biosynthetic metabolism (PEP, acetyl-CoA ,
aspartate). The enzymes under allosteric control usually function as
pacemakers in central pathways.
7. The energy status of cells is described by the energy charge:
ec = [ATP] + -HADP]/[ATP] + [ADP] + [AMP].
8. Some key enzymes are regulated by covalent modification. These
enzymes exist in an active form and an inactive or less active form.
Interconversion of these forms is achieved by phosphorylation/
dephosphorylation (eukaryotic enzymes involved in glycogen metabolism;
isocitrate dehydrogenase in E. coli), adenylylation/ deadenylylation (gluta-
mine synthetase), and acetylation/ deacetylation (citrate lyase). Other
regulatory mechanisms are association/dissociation of proteins and
oxidation / reduction by thioredoxin.
Chapter 8
Bacterial Fermentations
Large areas in the soil, in rivers, lakes, and oceans are devoid of oxygen.
Nevertheless, numerous microorganisms live in these anaerobic environ-
ments. We have already discussed the dissimilatory reduction of nitrate,
which takes place under anaerobic conditions. Besides this process, two
other microbiological processes account to a large extent for the biological
activities in environments devoid of oxygen: bacterial fermentations and
bacterial photosynthesis. The term fermentation was first defined by
Pasteur, to whom we owe the pioneering work in this field; he described
fermentations as life in the absence of oxygen. Today fermentations can be
defined as those biological processes that occur in the dark and that do not
involve respiratory chains with oxygen or nitrate as electron acceptors. In
many fermentations ATP is formed only by substrate-level phosphoryla-
tion. However, in a number of fermentations electron transport phos-
phorylation is also involved in ATP synthesis.
The bacteria carrying out fermentations are either facultative or obli-
gate anaerobes. Facultative anaerobes such as the enterobacteria grow as
aerobic heterotrophs in the presence of oxygen; under anaerobic condi-
tions they carry out a fermentative metabolism. In contrast, obligate
anaerobes are not able to synthesize the components of an oxygen-linked
respiratory chain. Consequently, they cannot grow as aerobes. Moreover,
many of the obligate anaerobes do not even tolerate oxygen and are killed
in air. These organisms are referred to as strict anaerobes.
When reduced flavoproteins or reduced iron-sulfur proteins come
together with oxygen, two toxic compounds are formed: hydrogen perox-
ide and the superoxide radical (see Chapter 2, Section V). Aerobes contain
catalase and superoxide dismutase for destruction of these compounds.
Table 8.1 shows the activity of these enzymes as found in aerobes,
Bacterial Fermentations 209
aerotolerant anaerobes, and strict anaerobes. It is evident that aerobes

contain high levels of catalase and superoxide dismutase. Most aero-
tolerant anaerobes are devoid of catalase but contain superoxide dis-
mutase. Strict anaerobes lack both enzymes. Although other factors might
be involved, it can be concluded that aerointolerant species die in the
presence of oxygen because of the deleterious effects of the superoxide
radical.
For growth of most strict anaerobes it is not sufficient to exclude
molecular oxygen. A low redox potential is required and usually the
growth media have to be supplemented with SH-containing compounds
such as thioglycolate, cysteine, or sodium sulfide. These compounds
establish reducing conditions. Methanogenic bacteria grow only in media
with a redox potential lower than Eo
= -0.3 V.
A comparison of aerobic heterotrophic metabolism with fermentative
metabolism reveals one major difference: aerobic heterotrophs couple the
oxidation of organic substrates to the reduction of oxygen or nitrate. This
involves respiratory chains with high ATP yields. With the exception of
Table 8.1. Activity of superoxide dismutase and catalase in a

variety of microorganisms Q
specific activity (U /mg)
organism superoxide dismutase catalase
Aerobes
Escherichia coli 1.8 6.1
Salmonella typhimurium 1.4 2.4
Rhizobium japonicum 2.6 0.7
Micrococcus radiodurans 7.0 289
Pseudomonas species 2.0 22.5
Aerotolerant anaerobes
Eubacterium limosum 1.6 0
Streptococcus faecalis 0.8 0
Streptococcus laetis 1.4 0
Clostridium oroticum 0.6 0
Lactobacillus plantarum 0 0
Strict anaerobes
Veillon ella alcalescens 0 0
Clostridium pasteurianum 0 0
Clostridium butyricum 0 0
Clostridium sticklandii 0 0
Butyrivibrio fibrisolvens 0 0.1
QJ. M. McCord, B. B. Keele, and 1. Fridovich, Proc. Natl. Ac~d. Sci. 68,
1024-1027 (1971).
210 8: Bacterial Fermentations
some incomplete oxidizers, the substrate is converted into cell material,

carbon dioxide, and water.
Fermentative anaerobes carry out a variety of oxidation-reduction
reactions involving organic compounds, carbon dioxide, molecular hy-
drogen, and sulfur compounds. These reactions are coupled to substrate-
level and/or electron transport phosphorylation. The ATP yield (mol ATP
per mol substrate consumed), however, is very low. Therefore, the amount
of cells obtained per mol of substrate is much smaller than with aerobes
and, in addition to cell material, large amounts of "fermentation end
products" are formed.
There are two fermentative processes that at first appear to be quite
similar to oxygen- and nitrate-dependent respirations: the reduction of
CO 2 to methane and of sulfate to sulfide. Together with the nitrate-
dependent respiration these processes are frequently called anaerobic
respirations. However, they bear little resemblance to the process of
denitrification. The reduction of CO 2 and of sulfate is carried out by strict
anaerobes whereas nitrate reduction is carried out predominantly by
aerobes if oxygen is not available. Furthermore, the energetics of these
processes is very different. Table 8.2 shows that the free energy change of
O 2 and nitrate reduction per two electrons is about the same; the values are
much lower for CO 2 and sulfate reduction. In fact, the values are so low
that the formation of one ATP per H 2 or NADH oxidized cannot be
expected (ATP synthesis from ADP + Pi requires more than 35.6 kJ
(8.5 kcal) per mol. Consequently, not all the reduction steps in methane and
sulfide formation can be coupled to ATP synthesis. Only the reduction of one
or two intermediates may yield ATP by electron transport phosphorylation,
and the ATP gain is therefore small. In conclusion, it seems appropriate to
reserve the term anaerobic respiration for denitrification and nitrate/
nitrite respiration and to consider CO 2 reduction and sulfate reduction
as fermentations.
Fermentations are usually classified according to the main fermentation
end products, and we distinguish alcohol, lactate, propionate, butyrate,
mixed acid, acetate, methane, and sulfide fermentations.
I. Alcohol Fermentation
A. Ethanol fermentation by yeasts

Alcohol fermentation is the domain of yeasts, notably of Saccharomyces
species, and most of the ethanol formed in nature and produced by the
fermentation industry comes from the anaerobic breakdown of glucose and
other hexoses by these organisms. Gay-Lussac had already established in
1815 that hexoses are converted into ethanol and CO 2 according to the
>
r;
o
:0-
Q.
~
3
:l
"
~
o'
:l
Table 8.2. Free energy changes in aerobic and anaerobic respiration and in sulfide and methane fermentation
fJ.GO, (kJ/2e-)
fJ.GO'
redox reaction (kJ /mol acceptor) H2 NADH
2H 2 + O 2 ~ 2H 20 -474.38 (-113.38) -237.19 (-56.69) -219.Q7 (-52.36)

~H2 + NO)' + H+ ~ -rN2 + 3H20 -560.61 (-133.99) -224.24 (-53.60) -206.12 (-49.27)
4H 2 + SO~- + 1H+ ~ -rH2S + -rHS- + 4H 20 -153.46 ( -36.68) -38.36 ( -9.17) -20.24 ( -4.84)
4H 2 + CO 2 ~ CH 4 + 2H 20 -130.79 ( -31.26) -32.70 ( -7.81) -14.58 ( -3.48)
akcal values in parentheses
--
212 8: Bacterial Fermentations'
following equation:
C6 H 12 0 6 -- 2CzH sOH + 2CO z
Figure 8.1 summarizes the alcohol fermentation as carried out by yeasts.
It is apparent that yeasts employ the Embden-Meyerhof-Parnas pathway
for glucose degradation. Thus, 2 mol of pyruvate are formed from 1 mol of
glucose. Pyruvate, however, is not converted to acetyl-CoA as in aerobic
metabolism, but is decarboxylated to acetaldehyde. The enzyme that
catalyzes this reaction is pyruvate decarboxylase; it can be regarded as the.
key enzyme of alcohol fermentation. Pyruvate decarboxylase contains
bound thiamine pyrophosphate; in fact, the function of thiamine pyrophos-
phate was first studied using this enzyme. As in the pyruvate dehy-
drogenase reaction, hydroxyethyl-thiamine pyrophosphate enzyme is an
intermediate. The hydroxyethyl group, however, is not oxidized but is
released as free acetaldehyde. The latter is then reduced to ethanol by
alcohol dehydrogenase.
One important feature of fermentations is apparent from Fig. 8.1: an
even hydrogen balance. Since there is no external H-acceptor such as
I~
~
fructose-I. 6-bisphosphate
12 ethanol I
~ 2 glyceraldehyde-3-
~ 2NAD+ ~ y-2P;
K---
2 acetaldehyde
2NADH + 2H -.
2
/t
I, 3-bisphosphoglycerate
t;-I 2eO I l 3 ~ G~~

[7 t'-@D
4)
2 pyruvate 3-phosphoglyceratc
~6
CM
2 phosphoenolpyruvate 2-phosphoglycerate
1 2H,O~ 5 I
Figure 8.1. Fermentation of glucose to ethanol and CO 2 by yeasts. 1, Initial
enzymes of the Embden-Meyerhof-Parnas pathway; 2, glyceraldehyde-3-
phosphate dehydrogenase; 3, 3-phosphoglycerate kinase; 4, phosphoglycerate
mutase; 5, enolase; 6, pyruvate kinase; 7, pyruvate decarboxylase; 8, alcohol
dehydrogenase.
Alcohol Fermentation 213
oxygen, NADH-producing and NADH-consuming reactions have to be

balanced out. This also implies that NADH formation is avoided by
anaerobes if possible. Therefore, they do not oxidize acetyl-CoA via the
tricarboxylic acid cycle. The cycle is usually interrupted between (1'-
oxoglutarate and succinyl-CoA so that glutamate can still be formed from
oxaloacetate and acetyl-CoA via citrate.
NH2 H 3C
N:. J
~ )=rCCH 2h---enzyme
CH2-~\+ I
H3CAN I lC~
H 3C-C-OIH
I I
H
hydroxyethyl-thiamine pyrophosphate enzyme
B. The Pasteur effect

The net ATP yield of the alcohol fermentation is 2 mol ATP Imol glucose
(Fig. 8.1)-much lower than the ATP yield of aerobic metabolism. Yeast
cells respond to this large difference. When they are transferred from
aerobic to anaerobic conditions they increase the rate of glucose break-
down by a factor of 3 to 4; a change from anaerobic to aerobic metabolism
is accompanied by a reduction of this rate and a stoppage of alcohol
formation. This phenomenon is called the Pasteur effect. Apparently, the
Pasteur effect is the result of differences in the cell's energy charge under
aerobic and anaerobic conditions. In the presence of oxygen the respira-
tory chain and the sites of substrate-level phosphorylation in the glycolytic
pathway compete for ADP. Furthermore, phosphofructokinase activity is
controlled by the level of ATP and citrate. Under anaerobic conditions the
activity of phosphofructokinase increases because it is activated by ADP
and AMP; also, more ADP is available for the enzymes catalyzing
substrate-level phosphorylation reactions. All this allows a greater sub-
strate flow through the glycolytic pathway.
C. Alcohol fermentation by bacteria

It should be mentioned that yeasts are not truly facultatively anaerobic
organisms. They grow only for some generations under these conditions.
There are, however, some bacterial species that carry out an alcohol
fermentation and grow very well under anaerobic conditions. Zymomonas
mobilis isolated from Mexican pulque and the closely related Zymomonas
anaerobica degrade glucose to pyruvate via the Entner-Doudoroff path-
way; they contain pyruvate decarboxylase and form nearly 2 mol each of
ethanol and carbon dioxide from 1 mol glucose. Sarcina ventriculi, a strict
anaerobe capable of growth under extremely acidic conditions, and
Erwinia amylovora, a facultatively anaerobic enterobacterium, ferment

glucose to ethanol and CO 2 via the Embden-Meyerhof-Parnas pathway
and the pyruvate decarboxylase and alcohol dehydrogenase reactions.
Both organisms, however, form small quantities of other products in
addition: acetate and molecular hydrogen (S. ventriculi) or lactate (E.
amylovora). In general, pyruvate decarboxylase is rare in bacteria. Many
lactic acid bacteria, enterobacteria, and clostridia form considerable
amounts of ethanol but do not employ pyruvate decarboxylase for acetal-
dehyde synthesis. In these organisms acetyl-CoA functions as ultimate
precursor of acetaldehyde; it is reduced by acetaldehyde dehydrogenase:
ace' aldehyde
CH3 -CO-SCoA + NADH + H+ dehydrogenase)
CH 3-CHO + CoASH + NAD+

The investigation of the alcohol fermentation has been of inestimable
importance for the development of biological sciences. In 1897 the
Buchner brothers discovered that an extract of macerated yeast fermented
glucose to ethanol. This demonstrated that a complex biological process
could function outside the cell.
In 1905 Harden and Young discovered the importance of phosphate
esters in metabolism; they identified a hexose bisphosphate (fructose
1,6-bisphosphate) as an intermediate in sugar fermentation. Later, the
path of glucose breakdown and the function of the coenzymes in this
process were unraveled by Neuberg, Embden, Meyerhof, Parnas, and
Warburg.
It should be mentioned here that ethanol is now produced on a large
scale by fermentation for industrial purposes. Bargasse (from sugar cane),
molasses (from sugar beets), and corn starch are used as substrates; yeasts
and to a small extent Zymomonas mobilis are employed as organisms.
Processes are currently being developed that allow ethanol production
from various substrates by thermophiles, such as Clostridium thermohydro-
sulfuricum, Thermoanaerobium brockii, or Thermoanaerobacter ethanoli-
cus. These organisms, however, do not contain pyruvate decarboxylase;
they form ethanol via acetyl-CoA and excrete in addition to ethanol
considerable amounts of acetate and lactate.
II. Lactate Fermentation

Lactate is a very common end product of bacterial fermentations. Some
genera-often referred to as lactic acid bacteria-form large amounts of
this product. These microorganisms have in common that they are highly
saccharolytic and that they lack most anabolic pathways. So they exhibit
very complex nutritional requirements, which are met by their environ-
ment such as plant materials, milk, and the intestinal tract of animals. Most
Lactate Fermentation 215
lactic acid bacteria are strictly fermentative but are aerotolerant. Some
streptococci, however, can use oxygen as H-acceptor and even form
cytochromes under certain conditions.
In Table 8.3 species of the genera Lactobacillus, Sporolactobacillus,
Streptococcus, Leuconostoc, Pediococcus, and Bifidobacterium are listed.
Either of three pathways is employed by these microorganisms for the
fermentation of carbohydrates to lactate.
The homofermentative pathway yields 2 mol of lactate per mol of
glucose:
homofermentative pathway 2 I
gIucose ) actate
The heterofermentative pathway yields 1 mol each of lactate, ethanol,
and CO 2 per mol of glucose:
heteroferrnentative pathway CO
glucose ) lactate + ethanol + 2
Table 8.3. Homo- and heterofermentative lactic acid bacteria and configuration of
lactic acid produced
configuration
genera and species homofermentative heterofermentative of lactic acid
Lactobacillus
L. delbrueckii + D (-)
L. lactis + D (-)
L. bulgaricus + D (-)
L. casei + L (+)
L. plantarum + DL
L. curvatus + DL
L. brevis + DL
L. fermentum + DL
Sporolactobacillus
S. inulinus + D (-)
Streptococcus
S. faecalis + L (+)
S. cremoris + L(+)
S. lactis + L (+)
Leuconostoc
L. mesenteroides + D (-)
L. dextranicum + D (-)
Pediococcus
P. damnosus + DL
Bifidobacterium
B. bifidum + L (+)
The bifidum pathway yields acetate and lactate in a ratio of 3 to 2:

bifidum pathway
2 glucose 3 acetate + 2 lactate
These equations are, of course, not followed completely by the lactic
acid bacteria but in homofermenting organisms the yield is frequently 1.8
mol lactate/mol glucose. The heterofermenting bacteria produce 0.8 mol.
lactate and some acetate in addition to ethanol and CO 2 ,
A. Homofermentative pathway
The homofermentative pathway is illustrated in Fig. 8.2. A close rela-
tionship to the alcohol fermentation is apparent. Glucose is degraded via
the Embden-Meyerhof-Parnas pathway to pyruvate. The latter is, how-
ever, not decarboxylated to acetaldehyde as in the alcohol fermentation
but used directly as H-acceptor. The ATP yield in both fermentations is
the same, 2ATP/glucose.
B. Heterofermentative pathway
The heterofermentative pathway is illustrated in Fig. 8.3. As in the
oxidative pentose phosphate cycle, ribulose-5-phosphate is formed via
6-phosphogluconate. Epimerization yields xylulose-5-phosphate, which is
cleaved into glyceraldehyde-3-phosphate and acetyl phosphate by an
I glucose I
~~
Figure 8.2. Formation of lactate from glucose by the homofermentative pathway. 1,
Enzymes of the Embden-Meyerhof-Parnas pathway; 2, lactate dehydrogenase.
I ethanol I
~NAD+
~NADH+W
acetaldehyde
COA~ NAD+
KNADH+W
acetyl-CoA
I glucose I
COA-f; Pi
~
CH) -CO-OPO) H2
acetyl-0
~
--t----- --I
glucose-6-0
k;- NAD+
H)PO.
H)C-C-TPP-E TPP-E
~NADH+W
b I
6-0-gluconate
NAD+
t iI
3
NADH + H+
-------<~ C02
H2C=~-TPP-E
r
I
OH I ribulose-5-0
H20 I
I
I
!4
H 2C-CH-TPP-E
__
I \ I ~H20H
OHOH I C=O
I I
>:::----~I HO-C-H
I
H-C-OH
O=C-H I
I CH2-0-
H-C-OH
tH2-0- xylulose-5-
glyceraldehyde-3-
Pi
NADH+ W
NAD+
Figure 8.3. Formation of CO 2 , lactate, and ethanol from glucose by the heterofer-
mentative pathway. 1, Hexokinase; 2, glucose-6-phosphate dehydrogenase; 3,
6-phosphogluconate dehydrogenase; 4, ribulose-5-phosphate 3-epimerase; 5, phos-
phoketolase. The cleavage reaction yields glyceraldehyde-3-phosphate and en-
zyme-bound a,f3-dihydroxyethyl-thiamine pyrophosphate. This is converted to
acetyl-TPP-E via the a-hydroxyvinyl derivative; phosphorylytic cleavage results in
acetyl phosphate formation. 6, phosphotransacetylase; 7, acetaldehyde dehy-
drogenase; 8, alcohol dehydrogenase; 9, enzymes as in homofermentative pathway.
enzyme not mentioned thus far, phosphoketolase. It contains thiamine

pyrophosphate and the formation of acetyl phosphate can be understood if
enzyme-bound a-hydroxyvinyl-thiamine pyrophosphate is assumed to
occur as an intermediate.
Acetyl phosphate is converted into acetyl-CoA by phosphotransacety-
lase. Subsequent reduction by acetaldehyde and alcohol dehydrogenases
yields ethanol. The glyceraldehyde-3-phosphate formed in the phos-
phoketolase reaction is converted to lactate as in the homofermentative
pathway.
In the course of this fermentation 2NADH are formed and consumed;
the ATP yield is 1 per mol of glucose-i.e., half of that of the homofer-
mentative pathway.
c. Bifidum pathway
In glucose breakdown by Bifidobacterium bifidum two phosphoketolases
are involved: one specific for fructose-6-phosphate and one specific for
xylulose-5-phosphate. The mechanism of both reactions is similar; fruc-
tose-6-phosphate phosphoketolase splits fructose-6-phosphate into acetyl
phosphate and erythrose-4-phosphate.
CHpH CH 3 -CO-OP0 3H 2
I
c=o +
I
HO-r-H ~Pi H-C=O
I
H-C-OH H-C-OH
I I
H-C-OH H-C-OH
I I
CH 2-O-@ CH 2-O-P
The bifidum pathway is illustrated in Fig. 8.4. It exhibits a very

interesting sequence of reactions. Without the participation of hydrogena-
tion and dehydrogenation reactions 2 mol of glucose are converted into
3 mol of acetate and 2 mol of glyceraldehyde-3-phosphate. The conversion
of the latter to lactate involves then glyceraldehyde-3-phosphate and
lactate dehydrogenases. The formation of acetate from acetyl phosphate is
coupled to the formation of ATP from ADP.
acetate kinase
CH 3-C-O-P0 3H 2 + ADP , ' CH 3-COOH + A TP
II
o
This reaction, which is catalyzed by acetate kinase, is of great impor-
tance for all anaerobes that form acetate because it effects ATP synthesis by
substrate-level phosphorylation. With 2.5 mol of ATP per mol of glucose
the ATP yield of the bifidum pathway is higher than that of the homo- and
heterofermentative pathway.
[ 2 glucosc I
fructose-6-0 fructose-6-0
erythose-4-0
): acetyl-0
___ 3~ ~
==x::
glyceraldehyde-3-0 sedoheptulose-7-0 acetate I I
xylulose-5- 0 ribose-5-0
+5
2t'c
ribulose-5-0
t6
xylulose-5-0
2 glyceraldchyde-3-0 2 acctyl-0
2P j 2NAD+ 2ADP
2NADH + 2W _
2NADH + 2H+
9
4ADP I 2 acetate I
2NAD+ @V
I 2 lactalc I
Figure 8.4. Formation of acetate and lactate from glucose by the bifidum pathway.
1, Hexokinase and glucose-6-phosphate isomerase; 2, fructose-6-phosphate phos-
phoketolase; 3, transaldolase; 4, transketolase; 5, ribose-5-phosphate isomerase;
6, ribulose-5-phosphate 3-epimerase; 7, xylulose-5-phosphate phosphoketolase;
8, acetate kinase; 9, enzymes as in homofermentative pathway.
D. Stereospecificity of lactate dehydrogenases

Table 8.3 has shown that lactic acid bacteria form either o( - )-, L( +)-, or
oL-lactic acid. 0(- )-Lactic acid-formers produce this enantiomer exclu-
sively whereas L( + )-lactic acid-formers always produce some o( - )-
enantiomer. Most organisms that excrete oL-lactate contain two lactate
dehydrogenases that differ in their stereospecificity. Some lactobacilli,
however, produce first L( + )-lactic acid, which-while accumulating-
induces a racemase. This enzyme then converts the L( + )-form into the
D( - )-form until equilibrium is reached. L. curvatus belongs to this rather

small group of lactobacilli.
The lactate dehydrogenases of streptococci, of Bifidobacterium bifidum,
and of L. casei have an absolute and specific requirement for fructose
1,6-bisphosphate and manganese ions. The bisphosphate brings about the
association of inactive protein dimers to enzymatically active tetramers.
E. Fermentation of other saccharides

A large number of other sugars besides glucose are fermented by lactic acid
bacteria; they include fructose, galactose, mannose, saccharose, lactose,
maltose, and pentoses. Certain variations of the normal fermentation
schemes have been observed with some of these substrates. Pentoses are,
for instance, fermented by some homofermentative lactobacilli (e.g., L.
casei) using phosphoketolase-the key enzyme of the heterofermentative
pathway. Fructose is fermented by Leuconostoc mesenteroides, but part of
the NADH formed is not used to reduce acetyl-CoA to ethanol; instead
fructose is reduced to mannitol. Thus, the products of fructose fermenta-
tion are lactate, acetate, CO 2 , and mannitol.
F. Malo-lactate fermentation
The decrease of the acidity of wine is partly due to the conversion of
L-malate to L-lactate. This process is called malo-lactate fermentation, and
it is carried out by some lactic acid bacteria, e.g., L. plantarum, L. casei,
and Lc. mesenteroides. In the presence of L-malate and a fermentable
sugar these organisms synthesize a special malic enzyme, which converts
L-malate to L-lactate:
Mn 2 + NAO-+-
L-malate ' L-lactate + CO 2
The purified enzyme does not possess lactate dehydrogenase activity so
that free pyruvate cannot be an intermediate. Presumably enzyme-bound
oxaloacetate and pyruvate function as intermediates in this reaction. It is
not known whether or not this decarboxylation reaction is coupled to the
generation of a chemical gradient across the membrane.
G. Formation of diacetyl and acetoin

In addition to the usual end products of the lactate fermentation Strepto-
coccus lactis subsp. diacetylactis and Leuconostoc cremoris form acetoin
and diacetyl-the characteristic flavor of butter. Citrate, which occurs in
milk in concentrations up to 1.5 g/liter, is the preferred substrate for
acetoin and diacetyl formation by the above organisms. As illustrated in
Fig. 8.5 citrate is cleaved by citrate lyase into acetate and oxaloacetate.
The enzyme citrate lyase is the key enzyme for the anaerobic breakdown of
citrate. It occurs in lactic acid bacteria, in enterobacteria, Veillonella

species, Clostridium sphenoides, and Rhodopseudomonas gelatinosa but is
not involved in citrate degradation under aerobic conditions; its interesting
reaction mechanism and its regulation in R. gelatinosa have been discussed
in Chapter 7.
The acetate formed in the citrate lyase reaction is excreted, and
oxaloacetate is decarboxylated to yield pyruvate. Diacetyl synthesis does
not proceed via a-acetolactate as in bacilli (see Fig. 6.26) or in enterobac-
teria (Fig. 8.15), but is accomplished by reaction of acetyl-CoA with
"active acetaldehyde" (enzyme-bound hydroxyethyl-thiamine pyrophos-
phate). Diacetyl is usually only produced in small amounts (2 mg/I); most
of it is reduced to acetoin by acetoin dehydrogenase:
acetoin
dehydrogenase
CH]-CO-CO-CH] + NADH + H+ , ' CH]-CH-CO-CH] + NAD+
I
OH
CH 2 -COOH
I
3HO-C-COOH
!
CHrCOOH
I 3CH] -(OOH I-{ 1

3 yH 2 -COOH
f
CO-COOH
3C0
1 21
3 CH) -CO-COOH
I CO2 I TPP-E
----..., ...... CoA
4
CH] -yH-TPP-E
Cli] -CO-CoA OH
LFTPP_E
I CH) -CO-CO-CH) I
diacetyl
sum: 3 cilrale_ laclate + 3 acetate + 5C0 2 + diacetyl

Figure 8.5. Formation of diacetyl from citrate by lactic acid bacteria. 1, Citrate
lyase; 2, oxaloacetate decarboxylase; 3, enzyme not characterized; 4, pyruvate
dehydrogenase complex; 5, lactate dehydrogenase; 6, diacetyl synthase. Instead of
pyruvate dehydrogenase the other pyruvate-degrading enzymes mentioned in the
text may be involved as well.
Whereas diacetyl synthesis according to Fig. 8.5 does not yield any
ATP, some ATP can be formed in the formation of acetoin. Because of the
NADH consumption in the last step only 0.5 lactate, but additional 0.5
acetate (via acetyl-CoA), are produced.
Diacetyl synthesis requires the conversion of pyruvate into C 2 com-
pounds. Moreover, all homofermentative lactic acid bacteria have to
synthesize acetyl-CoA for biosynthetic purposes from pyruvate. The
occurrence of three enzyme systems that yield acetyl-CoA or acetyl
phosphate from pyruvate have been demonstrated in lactic acid bacteria.
Streptococci contain the pyruvate dehydrogenase multienzyme complex.
In addition, the pyruvate-formate lyase system, which will be discussed in
Section IV of this chapter, has been detected in S. faecalis, in Bifidobacte-
rium bifidum, and in some lactobacilli (e.g., L. casei). Finally, L. delbriickii
and L. plantarum carry out a dismutation of pyruvate to lactate and acetyl
--
phosphate:
pyruvate
oxidase
+ Pi + FAD + CO2 + FADH2
--
pyruvate acetyl phosphate
lactate
oxidase
FADH2 + pyruvate lactate + FAD
Pyruvate oxidase is a flavoprotein that contains thiamine pyrophos-
phate. Neither lipoate nor CoA is involved in acetyl phosphate formation.
Lactate oxidase is also a flavoprotein, and the hydrogen transfer from one
flavoprotein to the other is mediated by riboflavin.
H. Growth yields and growth in air

It has been outlined in Chapter 3 (see Table 3.2) that approximately 35
mmol of ATP are necessary for the synthesis of 1 g of cells from glucose
and inorganic salts. Thus, about 30 g of cells can be formed per mol of
ATP. This value does not account for the energy of maintenance and for
other ATP expenditures of the living cell. However, if the amount of ATP
required for these processes is constant or if the variations are small, a
given amount of ATP should result in the formation of the same amount of
cells independent of the species. That this is true was shown by Bauchop
and Elsden. They determined a yield of 22 g S. faecalis cells per mol of
glucose fermented (all monomers required for macromolecule synthesis
were supplied with the medium). Since the homofermentative pathway
yields 2ATP per glucose fermented, 11 g of cells were produced per mol of
ATP:
S. faecalis growth yield: Y m = 22 g/mol glucose
YATP = 11 g/mol ATP
Zymomonas mobilis, which ferments glucose to ethanol and CO 2 via the
Entner-Doudoroff pathway (IATP per glucose; see Chapter 8, Section I),
yielded 8.3 g of cells per mol of glucose:

Z. mobilis growth yield: Y m = 8.3 g/mol glucose
YATP = 8.3 g/mol ATP
Thus, an organism gammg half the amount of ATP from glucose
(compared to S. faecalis) formed approximately half the amount of cells.
A YATP value of 10.5 for growth with all the monomers available for
biosynthesis is now generally accepted. It should be indicated, however,
that growth yield determinations in continuous culture are much more
reliable and allow more precise calculations of the amount of ATP
synthesized per mol of substrate consumed.
It has already been mentioned that many lactic acid bacteria are
aerotolerant. These organisms contain superoxide dismutase but lack a
true catalase. Instead, peroxidases are present, which catalyze the oxida-
tion of organic compounds (alcohols, aldehydes) or of NADH with'HzO z:
NADH + H+ + HzO z NADH pcroxidase~ NAD+ + 2H zO
However, some lactic acid bacteria, notably S. faecalis, L. brevis, and

L. plantarum form a true catalase when they are grown in the presence of
hemin. Thus, they are only unable to synthesize the heme prosthetic group
of catalase.
When aerotolerant lactic acid bacteria are grown in air, the growth yield
is usually much higher than under anaerobic conditions. For S. faecalis Ym
increases from 22 to 52. This increase is partly due to a change of the
fermentation pattern:NADH is oxidized with oxygen; pyruvate can be
converted to acetyl-CoA and additional ATP can be formed by the acetate
kinase reaction. Thus, oxygen changes the lactate fermentation into an
acetate fermentation. There are, however, also indications that S. faecalis
is capable of oxidative phosphorylation to a limited extent. In the presence
of hemin and in air cytochromes are formed by this organism and by some
other streptococci. So it appears that some lactic acid bacteria are not truly
obligately anaerobic bacteria.
I. Energetics of lactate export

Lactic acid is a stronger acid than acetic and butyric acids. The pK values of
these three acids are 3.86 (lactic), 4.75 (acetic), and 4.82 (butyric),
respectively. This difference is one reason why lactic acid bacteria reach
lower pH values during growth than anaerobes producing acetic or butyric
acid. This can be explained as follows: The membrane of obligate
anaerobes has to be energized like that of aerobes. This means that a
protonmotive force has to be continuously generated. In lactic acid
bacteria which lack membrane-bound electron transport chains, ATP
hydrolysis as catalyzed by the proton-translocating ATPase is the principal
pH 6 pH 4
H+
laclale- laclale-
CH 3 - COOH
2H+ 2H+
CH 3-COO-
e e
CH 3 -CHOH- COO-
nATP nATP
H+ H+
nADP + nP j nADP +nP j
a b
Figure 8.6. Function of weak acids as uncouplers at low pH (a), coupling of lactate
efflux with !:J.P generation (b).
mechanism for IJ.P generation (see Chapter 2). At a pH of, e.g., 4, acetic
and butyric acids are present in the medium in the undissociated forms.
These undissociated acids are lipophilic and can penetrate the membrane.
At an intracellular pH of, e.g., 6, the acids will dissociate again; more acid
diffuses in, and the D.pH between outside and inside will finally be evened
(Fig. 8.6a). Thus, undissociated organic acids function as uncouplers at pH
values around or below their pK values. Since lactic acid is a stronger acid
than acetic or butyric acids, its uncoupling function comes into play only at
a very low pH (pH - 3.5).
Recently, Konings and co-workers have demonstrated that at least some
lactic acid bacteria (shown for Streptococcus cremoris) have a second
mechanism for IJ.P generation at their disposal. As depicted in Fig. 8.6b,
lactate export is electrogenic; two protons are excreted per one molecule of
lactate, and a D.pH is generated that can be used to drive symport processes
or ATP synthesis. This systems's driving force is the inside/outside lactate
gradient. Thus, it can be employed only at low extracellular lactate
concentrations. Such conditions exist when lactic acid bacteria coexist with
organisms consuming lactate.
III. Butyrate and Butanol-Acetone

Fermentation
The fermentation of sugars to butyric acid was discovered by Pasteur in
1861. Soon after, microorganisms responsible for the formation of butyrate
were isolated, and it was found that several clostridial species carried out
this type of fermentation. Generally, only obligate anaerobes form buty-
rate as a main fermentation product; they belong to the four genera
Clostridium, Butyrivibrio, Eubacterium, and Fusobacterium (Table 8.4).
Butyrate and ButanolAcetone Fermentation 225
Table 8.4. Some species

forming butyrate as a
major fermentation end
product
Clostridium butyricum
C. kluyveri
C. pasteurianum
Butyrivibrio fibrisolvens
Eubacterium limosum
Fusobacterium nucleatum
The mechanism of butyrate formation was not well understood until

Barker and collaborators did their pioneering studies on C. kluyveri and
until ferredoxin was discovered in C. pasteurianum. The clostridia em-
ploy phosphotransferase systems for sugar uptake and the Embden-
Meyerhof-Parnas pathway for degradation of hexose phosphates to pyru-
vate. The conversion of pyruvate to acetyl-CoA involves an enzyme system
not discussed thus far: pyruvate-ferredoxin oxidoreductase.
A. Ferredoxin and pyruvate-ferredoxin

oxidoreductase
When a cell extract of C. pasteurianum is incubated with pyruvate under
anaerobic conditions the decomposition of pyruvate according to the
following equation can be observed:
CHrCO-COOH + H 3 P0 4 -- CH r CO-OP0 3 H 2 + CO 2 + H 2
Acetyl phosphate is formed and hydrogen gas is evolved. This reaction
is known as the phosphoroclastic reaction. A thorough investigation of the
enzymes involved resulted in the finding that this phosphorylytic cleavage
is the sum of several reactions; they are summarized in Fig. 8.7. Pyruvate is
first decarboxylated by pyruvate-ferredoxin oxidoreductase; the remaining
C2 -moiety is covalently bound to the TPP-containing enzyme as in the
pyruvate decarboxylase and the pyruvate dehydrogenase reactions. In the
next step acetyl-CoA is formed; in contrast to pyruvate dehydrogenase
the two hydrogens are not transferred to NAD+, but are used to reduce
ferredoxin. This difference is important because ferredoxin has a very low
redox potential; at pH 7.0 it is about the same as that of the hydrogen
electrode:
H 2 /2H+ Eo = -0.42 V
red. ferredoxin/ox. ferredoxin Eo = -0.41 V
NADH + H+ /NAD+ Eo = -0.32 V
I pyruvate I + TPP-E +===!:" HETPP-E + ICOzl
2 HETPP-E + ferredoxin + CoA ::;.==:::!:" TPP-E + acetyl-CoA + ferredoxin' Hz
hydrogenase
3 ferredoxin . Hz ferredoxin + ~
phosphotransacetyla_se --.,
4 acetyl-CoA + Pi " Iacetyl phosphate I + CoA
Figure 8.7. Steps of phosphoroclastic reaction. Steps 1 and 2 involve the enzyme
pyruvate-ferredoxin oxidoreductase and ferredoxin. TPP-E, thiamine pyrophos-
phate-containing oxidoreductase; HETPP-E, hydroxyethyl-TPP-E. Steps 3 and 4
are catalyzed by hydrogenase and phosphotransacetylase, respectively.
Consequently, even in an environment saturated with hydrogen gas,

reduced ferredoxin can transfer electrons to hydrogenase and hydrogen
can be evolved. This is what is observed when clostridia ferment carbohy-
drates. Moreover, the pyruvate-ferredoxin oxidoreductase is reversible
and under appropriate conditions pyruvate can be synthesized from
acetyl-CoA, reduced ferredoxin, and CO 2 (see C. kluyveri).
NADH is a much weaker reducing agent than reduced ferredoxin. This
is one reason why the pyruvate dehydrogenase reaction is not reversible.
Ferredoxin was discovered in C. pasteurianum by Mortenson, Valen-
tine, and Carnahan in 1962. It belongs to the iron-sulfur proteins (until
recently called nonheme-iron proteins) and it should be remembered that
this class of proteins plays an important role in the respiratory chain (see
Figs. 2.7 and 2.11). Clostridial ferredoxins have a molecular weight of 6000
and contain eight iron atoms per molecule, which are arranged in the form
of two cubanelike iron-sulfur clusters, [4Fe-4S]-clusters. One of these
clusters is shown in Fig. 8.8. The iron atoms are bound to cysteine residues
of the peptide chain and they are interconnected by sulfide bridges. So
ferredoxin contains eight sulfide bridges; upon acidification of ferredoxin
solutions they are liberated as H 2S. This sulfur is therefore called labile
sulfur. Clostridial ferredoxins are 2-electron carriers (one electron per
Figure 8.8. [4Fe-4S]-cluster of ferredoxin . , Fe atoms; 0, sulfur atoms.

Butyrate and Butanol-Acetone Fermentation 227
cluster). Because of their brownish color they are easily recognized in cell
extracts and during fractionation of such extracts.
It should be emphasized that not all ferredoxins have the same general
composition. The span of variation is apparent from Table 8.5. It can be
seen that the C. thermoaceticum ferredoxin differs from the typical
clostridial ones in that it contains only one [4Fe-4S] cluster. The Chroma-
tium ferredoxin is larger and the one of D. gigas harbors even three
clusters. In E. coli, in A. vinelandii, and in many other aerobes we find
iron-sulfur proteins with [2Fe-2S] clusters. The occurrence of even [3Fe-
3S] clusters in sulfate-reducing bacteria has been reported.
All the [4Fe-4S]-ferredoxins mentioned have in common that they are
low-potential electron carriers. The redox state of their reduced form is
3Fe2+ + IFe3+ and that of their oxidized form is correspondingly
2Fe z+ + 2Fe 3 +. Another class of iron-sulfur protein is present in many
bacteria, the high-potential iron proteins called HiPiP. Their redox poten-
tial is in the order of + 350 mV; the HiPiP's reduced form contains
2Fe z+ + 2Fe3 + and the oxidized form IFe z+ + 3Fe3+. This difference
makes the difference in redox potential understandable.
The investigation of composition and structure of purified enzymes in
recent years has shown that many of them contain iron-sulfur centers as
well. Examples are hydrogenase, formate dehydrogenase, sulfite reduc-
tase, pyruvate: ferredoxin oxidoreductase, COz reductase, nitrogenase,
CO dehydrogenase, CO oxidase, trimethylamine dehydrogenase, dihy-
droorotate dehydrogenase, glutamate synthase, and xanthine dehy-
drogenase.
One other protein is known that is also a low-potential electron carrier
but differs in other aspects: ftavodoxin. It is formed by several obligate
anaerobic bacteria when they grow in iron-limited media. Iron and labile
sulfide are absent and flavin mononucleotide functions as redox group. The
Table 8.5. Properties of various ferredoxins
molecular No. of
Organism weight cluster clusters Eb (mY)
Clostridium pasteurianum 6000 4Fe-4S 2 -390

C. acidiurici 6000 4Fe-4S 2 -430
C. thermoaceticum 7500 4Fe-4S 1 ND
Chromatium vinosum 10,000 4Fe-4S 2 -480
Desulfovibrio gigas 18,000 4Fe-4S 3 -455
Escherichia coli 12,000 2Fe-2S 1 -380
Azotobacter vinelandii 21,000 2Fe-2S 1 -225
Chromatium vinosum HiPip a 10,000 4Fe-4S +35
a HiPiP, High-potential iron protein; ND, not determined

redox potential lies around -0.4 V. Finally it should be mentioned that

rubredoxin occurs in many anaerobes. This redox carrier contains per
redox center one iron, which is bound to four cysteine residues of the
polypeptide chain. Thus, rubredoxin does not contain labile sulfide. Its
redox potential lies around -0.06 V. The function of rubredoxin in
anaerobes is unknown; in some aerobes it is involved in alkane oxidation
(see Fig. 6.11).
B. The path of butyrate formation

The reactions involved in butyrate formation from glucose are summarized
in Fig. 8.9. The Embden-Meyerhof-Parnas pathway and pyruvate-
ferredoxin oxidoreductase yield 2 acetyl-CoA, 2C0 2 , and 2 reduced
ferredoxins, which are reconverted to the oxidized form by hydrogenase
under hydrogen evolution. Furthermore, 2NAD+ are reduced and 2ATP
are formed. The advantage of hydrogen evolution in the course of pyruvate
breakdown is apparent; only 2NADH are formed in the degradation of
glucose to acetyl-CoA.
The NADH formed in the glyceraldehyde-3-phosphate dehydrogenase
reaction is oxidized by the conversion of 2 acetyl-CoA into butyrate. In this
pathway acetoacetyl-CoA, L( + )-,8-hydroxybutyryl-CoA, and crotonyl-
CoA are intermediates. It should be remembered that the storage material
poly-,8-hydroxybutyric acid is a polymer of the D( - )-form (see Fig. 5.20)
and that the thioester of the D( - )enantiomer is an intermediate in
long-chain fatty acid synthesis.
Butyrate is not formed from butyryl-CoA by simple hydrolysis; this
would be a waste of energy. Instead butyryl-CoA is converted to butyryl
phosphate, and finally the phosphate group is transferred to ADP. These
reactions proceed under catalysis of the enzymes phosphotransbutyrylase
and butyrate kinase, which are analogous to phosphotransacetylase and
acetate kinase.
Figure 8.9 shows that the ATP yield of the butyrate fermentation is
3ATP jglucose; this is more than was gained in the fermentations discussed
thus far.
C. The formation of acetone and butanol

A number of butyrate producing clostridia form small amounts of n-
butanol. With a few species, however, a real shift from butyrate production
to solvent production (n-butanol and acetone or isopropanol) can be
observed under certain conditions. These species include C. acetobutyli-
cum, C. beijerinckii, C. tetanomorphum, and C. aurantibutyricum. The
most prominent species among these is C. acetobutylicum, which has been
used on an industrial scale for the synthesis of n-butanol and acetone from
molasses. Since n-butanol is very toxic (it interferes with membrane
I CH 3-CH 2 -CH 2 -COOH
butyrate
ATP --'19
ADP --"I
CH 3-CH l -CH 2 -co-o-CD
butyryl phosphate
I glucose I
2NAD+
2NADH+2W
2ADP
()
CH 3 -CH l --CH 2 -co -CoA
butyryl-CoA ~ 2CH 3 -CO-COOH
NAD+~
+w-1 7 k' 2Fd ~~

NADH
CH3-CH=CH-CO-CoA
f- 2Fd . Hl ~ 2CoA
r
crotonyl-CoA
2CH 3 -CO-CoA
H2--i6 COA
CH 3 -yH- CH l-CO-CoA
CH3-CO-CHl-CO-CoA
OH acetoacetyl-CoA
L (+ Hl-hydroxybutyryl-CoA
t ".. . . . . _
("5'\
NAD+ NADH + H+
sum: glucose + 3ADP + 3P j __ butyrate + 2C0 2 + 2H 2 + 3ATP
Figure 8.9. Path of butyrate formation from glucose. 1, Phosphotransferase system
and Embden-Meyerhof-Parnas pathway; 2, pyruvate-ferredoxin oxidoreductase;
3, hydrogenase; 4, acetyl-CoA-acetyltransferase (thiolase); 5, L( + )-,B-hydroxy-
butyryl-CoA dehydrogenase; 6, L-3-hydroxyacyl-CoA hydrolyase (crotonase); 7,
butyryl-CoA dehydrogenase; 8, phosphotransbutyrylase; 9, butyrate kinase.
functions) the final solvent concentration reached is relatively low, in the

order of 2%.
Figure 8.10 illustrates the course of product formation during growth of
C. acetobutylicum. Towards the middle of the fermentation when the pH
had dropped below 5, acids are no longer produced, and n-butanol and
acetone appear as new products. During the solvent phase the butyrate
concentration decreases again and an increase of the pH is recorded.
Acetate stays relatively constant. The shift from acid to solvent production
is not only a response of the cells to the low pH. Cultures growing very
rapidly often "miss" the shift and produce acids only. Recent results
indicate that the shift and the initiation of sporulation are connected to one
another.
7
ISO
6
:I:
0.
5
4
~ 100
~
U
.g
o
:i:
SO
20 30 40 SO 70
Time (h)
Figure 8.10. Course of fermentation in a batch culture of C. acetobutylicum at low
phosphate concentration (0.62 mM). Butanol, .; acetone, A; ethanol, .; buty-
rate, 0; acetate, L".; pH, O. [H. Bahl, W. Andersch, G. Gottschalk. Eur. J. Appl.
Microbiol. Biotechnol. 15, 201-205 (1982).]
The reactions involved in acetone and butanol formation are summa-
rized in Fig. 8.11. Under shift conditions two enzyme activities appear in
the cells that convert acetoacetyl-CoA into acetone: a coenzyme A
transferase and acetoacetate decarboxylase. Likewise, two enzymes are
formed that reduce butyryl-CoA to n-butanol. The ratio in which butanol
and acetone are formed lies in the order of 2: 1.
Table 8.6. Fermentation balances of clostridia a
C. butyricum C. perfringens C. acetobutylicum

products (amounts formed in mol/loa mol glucose fermented)
butyrate 76 34 4
acetate 42 60 14
lactate 33
COz 188 176 221
Hz 235 214 135
ethanol 26 7
butanol 56
acetone 22
aW. A. Wood, In: The Bacteria, I. C. Gunsalus and R. Y. Stanier

(eds.). Academic Press, New York and London, 1961, vol. 2, pp.
59-149.
12 glucose I
4NAD+
4NADH + 4W
t
CH3-CO-CH2-CO-CoA CH3-CO-CH2-CO-CoA
acetoacetyl-CoA acetoacetyl-CoA
k-acetate
2NADH+ 2w
H2 0
2NAD+ 4 r.-acetYI-COA
CH3-CH2-CH2-CO-CoA CH 3 -CO-CH 2 -COOH

butyryl-CoA acetoacetate
NADH+wt
NAD+
CoA
~'
CH 3 --CH 2 -CH 2 -CHO
butyraldehyde
acetone
NADH + H+~6
NAD+-1
!CH3-CHl-CH2-CHlOH I
butanol
sum: 2 glucose + 4ADP + 4P j ------<~
butanol + acetone + 4H 2 + 5C0 2 + 4ATP
(acetyl-CoA formation in step 2 not considered)
Figure 8.11. Formation of acetone and butanol by C. acetobutylicum. 1, Reactions
as in Figure 8.9; 2, acetoacetyl-CoA: acetate coenzyme A transferase; 3, acetoace-
tate decarboxylase; 4, L( + )-,B-hydroxybutyryl-CoA dehydrogenase, crotonase, and
butyryl-CoA dehydrogenase; 5, butyraldehyde dehydrogenase; 6, butanol dehyd-
rogenase.
The acetone-butanol fermentation is an interesting process, and it may

become of economic importance again in the future.
D. Fermentation balances
Table 8.6 gives fermentation balances for C. butyricum, C. perfringens,
and C. acetobutylicum. It can be seen that the fermentation schemes
depicted in Figs. 8.9 and 8.11 are not exactly followed. C. butyricum forms
acetate and some extra hydrogen; C. perfringens forms lactate and ethanol
as do many saccharolytic clostridia. In the C. acetobutylicum fermentation
the amount of H 2 evolved is largely diminished because of the reutilization
of butyrate for butanol synthesis.
In complex fermentations such as the one carried out by C. acetobutyli-
cum it is difficult to judge whether the hydrogen balance is even or not.
Therefore, the oxidation/reduction (O/R) balance of complex fermenta-

tions is determined, as illustrated in Table 8.7. Arbitrarily the O/R value
of formaldehyde and multiples thereof are taken as zero. Each 2H in
excess is expressed as -1, and a lack of 2H is expressed as + 1. Ethanol
has the sum formula C2 H 6 0; addition of H 2 0 gives C2 H s0 2 . In compari-
son to C2 H 4 0 2 , 4H are in excess (O/R value = -2). From CO 2 water is
subtracted; in C(-2H)O four hydrogens are missing as compared to CH 2 0
(O/R value = +2).
Alternatively, such calculations can be done on the basis of the number
of available hydrogen in the substrate and in the products. This number
is determined by oxidizing the compound to CO 2 with water
(C 6 H 12 0 6 + 6H 2 0 -- 24H + 6C02 ; C3 H 6 0 + 5H2 0 -- 16H + 3C02 ).
E. Linear and branched fermentative pathways

Product formation in the alcohol or the homolactate fermentation can be
described by linear pathways. The substrate: product ratios and also the
thermodynamic efficiencies of these fermentations are invariable. Thauer
and colleagues distinguished from them the branched fermentative path-
ways. Here, either a more oxidized product resulting in a comparatively
Table 8.7. Carbon recovery, O/R balance, and balance of available hydrogen of an
acetone-butanol fermentation a
balance of available
O/R balance hydrogen
substrate mol/
and 100 mol mol O/R O/R value available available H
products substrate carbon value (mol/lOO mol) H (mol/100 mol)
glucose 100 600 0 24 2400

butyrate 4 16 -2 -8 20 80
acetate 14 28 0 8 112
COz 221 221 +2 +442 0
Hz 135 -1 -135 2 270
ethanol 7 14 -2 -14 12 84
butanol 56 224 -4 -224 24 1344
acetone 22 66 -2 -44 16 352
total 569 -425 +442 2242
aCarbon recovered: 569/600 x 100 = 94.8%; O/R balance: 442/425 = 1.04; balance of
available H: 2400/2242 = 1.07. The latter two values are different because the carbon
recovery affects the balance of available H but not the 0/ R balance (0/ R value of
glucose = 0).
high ATP yield or a reduced product resulting in a low ATP yield can be
formed. The ratio in which the two products are formed depends on the
portion of NADH that can be oxidized by the organisms to NAD+ under
Hz evolution. A good example for a branched pathway is the formation of
butyrate and acetate by C. butyricum. According to the pathway depicted
in Fig. 8.9, the fermentation balance is even if butyrate only is produced.
Table 8.6, however, teaches us that acetate is among the products and that
more than 2 mol Hz per mol of glucose are produced. The explanation for
this is depicted in Fig. 8.12: Hz evolution from NADH makes acetate
excretion possible. Since 1ATP can be synthesized from 1 acetyl-CoA
through conversion to acetate and only 1/2ATP through conversion to
butyrate, the evolution of more Hz and the excretion of acetate increase
the thermodynamic efficiency of the pathway.
Since the redox potential of NADH is mote positive than the one of Hz
and of ferredoxin, the enzyme NADH: ferredoxin oxidoreductase can
work only in the direction of ferredoxin reduction at a low partial pressure
of Hz (see section IX of this chapter) or when driven by some type of
reverse electron transport. It has been shown that the reverse direction
(reduction of NAD+ by reduced ferredoxin) is strongly inhibited by
NADH. In other words, the system functions like a valve. Hz formation
from NADH is allowed; NAD+ reduction by Hz is prevented:
NADH ( I ) red Fd ( Hz
NADH
Branched fermentative pathways provide organisms with more flexibility
and allow them to adjust the thermodynamic efficiency of their catabolism
according to the possibilities to oxidize reduced coenzymes under Hz
evolution.
glucose
ATP+-NA~
H PYIate 1~
-~
2
H1
ethanol acetyl-CoA
ATP/AT:~
acetate butyrate
Figure 8.12. Alcohol fermentation and butyrate-acetate fermentation as examples
of linear and branched pathways. 1, Reaction catalyzed by NADH: ferredoxin
oxidoreductase; 2, hydrogenase reaction. [Modified from R. K. Thauer, K.
Jungermann, K. Decker; Bacterial. Rev. 41, 100-180 (1977)].
F. Not all saccharolytic clostridia form butyrate

Clostridia are always brought into close connection with the butyrate
fermentation; and it is, therefore, important to stress that several saccha-
rolytic clostridia do not produce butyrate at all. C. sphenoides, for instance,
ferments glucose to ethanol, acetate, COb hydrogen and small amounts of
lactate. Some additional clostridial species represent the same fermenta-
tion type: C. glycolicum, C. indolis, C. thermohydrosulfuricum, and the
cellulolytic species C. cellobioparum and C. thermocellum. Furthermore, a
few clostridial species (c. thermoaceticum and C. formicoaceticum) fer-
ment hexose to almost three acetate; they will be discussed in a separate
section of this chapter.
G. The ethanol-acetate fermentation of

Clostridium kluyveri
c. kluyveri carries out a very interesting fermentation. It produces
butyrate, caproate, and hydrogen from ethanol and acetate. Ethanol alone
is not fermented, but acetate can be replaced by propionate. The ratio in
which butyrate and caproate are formed can vary; an increase of the
ethanol concentration of the medium favors caproate formation. A typical
fermentation balance is:
6 ethanol + 3 acetate - - 3 butyrate + 1 caproate + 2H 2 + 4H 2 0 + H+
Approximately 0.3 mol of hydrogen is evolved per mol of ethanol
fermented.
How C. kluyveri gains ATP for growth is difficult to see. The principle is
that per mol of hydrogen evolved, 0.5 mol of acetyl-CoA is not required as
H-acceptor as has been shown by Decker and collaborators. This acetyl-
CoA serves to form ATP. The reactions involved are summarized in
Fig. 8.13. For the sake of simplicity only the formation of butyrate and H 2
is considered. Reactions involved in caproate formation are analogous
(butyryl-CoA condenses with acetyl-CoA, 3-oxocaproyl-CoA is reduced,
water is removed, and subsequent reduction yields caproyl-CoA).
Alcohol and acetaldehyde dehydrogenases are associated with one
another and are particle bound. The first enzyme is NAD+ -specific and the
second one reacts with either NAD+ and/or NADP+. Hydrogen is
evolved in the course of these reactions by a mechanism not fully
understood. Presumably electrons are transferred from NADH to hy-
drogenase via ferredoxin as outlined in section E. A NADH: ferredoxin
oxidoreductase has been demonstrated in cell extracts of C. kluyveri.
L( + )-.8-Hydroxybutyryl-CoA dehydrogenase is NADP+ -specific, which is
unique among butyrate-forming organisms. It indicates a close coupling
between acetaldehyde oxidation and acetoacetyl-CoA reduction.
4=
NAD+
Gill 3 I
~ NADH + H+ (I) acetate
+ !. ) acetaldehyde
(1
2 ~NADP+ + (I +1)CoA ~
~ 32 ~
~ NADPH+W 10
(1 + I) acetyl-CoA ..!ADP
2
~}l"",e <;lCoA
acetyl-CoA CoA acetoacetyl-CoA
'f
NADPH+W
-
Ibutyrate I
NADP+
l( + )-(3-hydroxybutyryl-CoA
butyryl-CoA
)",0
crotonyl-CoA
a
~W
sum: (I +"2) ethanol + (I
a
-"2) a a
acetate + "2 ADP +"2 Pi---~
a
butyrate + aH 2 + "2ATP
a approx. 0.4
Figure 8.13. The ethanol-acetate fermentation of C. kluyveri. 1, Alcohol dehy-
drogenase; 2, acetaldehyde dehydrogenase; 3, Hz-evolving enzyme system; 4,
thiolase; 5, L( + )-,B-hydroxybutyryl-CoA dehydrogenase (NADP-specific); 6, cro-
tonase; 7, butyryl-CoA dehydrogenase; 8, CoA transferase; 9, phosphotransacety-
lase; 10, acetate kinase.
In Figure 8.13 "a" has a value of about 0.4 so that approximately

0.4 mol Hz are evolved per 1.2 mol ethanol consumed. From the 1.2 mol of
acetyl-CoA formed, 1 mol is required for NADH and NADPH oxidation
in the butyric acid cycle; 0.2 mol is available for ATP formation. Thus,
the ATP yield is 0.2/1.2 or 1ATP per 6 ethanol.
H. Aspects of biosynthetic metabolism of

Clostridium kluyveri.
Although C. kluyveri grows on C z compounds, it does not contain a
glyoxylate cycle for the synthesis of C 4 dicarboxylic acids. Instead, it takes
advantage of the reversibility of the pyruvate-ferredoxin oxidoreductase
reaction. Under the reducing conditions of the ethanol-acetate fermenta-

tion the reductive carboxylation of acetyl-CoA is feasible:
CH3 CO-SCoA + CO 2 + Fd . H 2 E )
CHrCO-COOH + HSCoA + Fd
Pyruvate is trapped by formation of alanine or by further carboxylation
to oxaloacetate by pyruvate carboxylase. The importance of carboxylation
reactions in biosynthetic metabolism of C. kLuyveri is underlined by the
fact that C. kLuyveri requires CO 2 for growth. About 30% of its cellular
material is derived from CO 2 , In agreement with the above-mentioned
carboxylation reactions, the alanine carboxyl group and both carboxyl
groups of aspartate originate quantitatively from CO 2 , Since C. kLuyveri
contains also a pyruvate-formate lyase, the carboxyl group of pyruvate is
also the precursor of formate, which is the starting material for C 1 units in
biosynthesis.
Interestingly, C. kLuyveri uses enzymes of the tricarboxylic acid cycle to
synthesize glutamate; it contains citrate synthase, cis-aconitase, isocitrate
dehydrogenase, and glutamate dehydrogenase, and so do many other
obligate anaerobic bacteria. This emphasizes the dual function of the
tricarboxylic acid portion of the cycle: provision of NADH as part of the
complete cycle (aerobes) and provision of a-oxoglutarate (aerobes and
anaerobes).
The citrate synthase of C. kLuyveri and a few other anaerobes (c.
acidiurici, C. cyLindrosporum, some sulfate-reducing bacteria) differs in
two respects from all other citrate synthases. It has a specific requirement
for Mn 2 + whereas all other citrate synthases do not require metal ions at
all. Moreover, it differs in its stereospecificity (Fig. 8.14). Thus, a very
small group of anaerobic bacteria contain a special enzyme, re-citrate
synthase. All other organisms contain-as far as tested-the si-type
synthase.
-;J
si- HOO~" /H
" I" I 1-'I', , I
HOOC-H,C ,HOOC-H,C OH C,
+,
cUrare synthase ~ 0,.. B
/iI , , I
'1 ~H, -~OOH
~=O CH) -CO-CoA CH,-COOIi, ---
COOH COOH HOO~
re-
citrate synthase
HOOC-H,C,
1
CH,-COOH
1 A I ' 'I
HOOC-H,C OH 1HooC~ ""
'" C ~OH '''',(,~CH, -COOH ....!!.,.. C
tOOH ~OOH HOOC/ ) \'H,-COOH
6 4 s
Figure 8.14. Formation of citrate by si-citrate synthase and by re-citrate synthase.
Note that the carboxymethyl residue derived from acetyl-CoA comes into different
positions. A, Molecule turned 1200 ; B, stereospecific removal of water by
cis-aconitase allows distinction between the two carboxymethyl groups of radioac-
tive citrate prepared. With 14C-labeled acetyl-CoA, 4,5- 14C-cis-aconitate and
1,2- 14C-cis-aconitate are formed, respectively.
Mixed Acid and Butanediol Fermentation 237
IV. Mixed Acid and Butanediol

Fermentation
This type of fermentation is carried out by the enterobacteria. Organisms
belonging to the genera Escherichia, Salmonella, and Shigella ferment
sugars to lactic, acetic, succinic, and formic acids. In addition CO 2 , H 2 , and
ethanol are formed. Species of the genera Enterobacter, Serratia, and
Erwinia produce less acids but more gas (C0 2), ethanol, and above all
large amounts of 2,3-butanediol. Two typical fermentation balances are
given in Table 8.8, and the pathways leading to all these products are
summarized in Fig. 8.15.
Enterobacteria employ the Embden-Meyerhof-Parnas pathway for
hexose breakdown. The pathway leading to succinate branches off at phos-
phoenolpyruvate; all other end products are derived from pyruvate.
Three enzyme systems act on pyruvate, and the amounts in which the fer-
mentation products are formed depend very much on the activity of these
enzyme systems. In the mixed acid fermentation large amounts of lactate
are formed by the action of lactate dehydrogenase. Little lactate only
is produced in the butanediol fermentation. The two other enzyme sys-
tems-pyruvate-formate lyase and a-acetolactate synthase-deserve spe-
cial attention.
A. Pyruvate-formate lyase
The enterobacteria are able to synthesize two enzyme systems for
pyruvate breakdown to acetyl-CoA. The pyruvate dehydrogenase multien-
zyme complex is involved in aerobic metabolism. Under anaerobic con-
ditions it is no longer synthesized, and the enzyme still present is inhibited
Table 8.8. Products formed in the mixed acid and butanediol

fermentation"
Escherichia coli Enterobacter aerogenes

product (mol formed/lOa mol glucose fermented)
formate 2.4 17.0

acetate 36.5 0.5
lactate 79.5 2.9
succinate 10.7
ethanol 49.8 69.5
2,3-butanediol 0.3 66.4
CO 2 88.0 172.0
hydrogen 75.0 35.4
a W.A . Wood, In: The Bacteria, I. C. Gunsalus and R. Y. Stanier

(eds.). Academic Press, New York and London, 1961, vol. 2, pp. 59-
149.
I glucose I
~~6p
10 (H) 9 CO 2 f=.ATP
Isuccinate I ... ~- - oxaloacetate ~./ PEP
~ADP
Ilactate I..
2 (H)
<
r---
pyruvate
ATP
XCOA
5 (H)
acetaldehyde ~ I
A "
1(H) acetyl-CoAPi
~ CoA I
ormate
7 4
I ethanol I
acetyl- CO 2 I~
8~ADP
~ATP
Iacetate I
I glucose
a-acetolactate
acetoin
~~ (H)
b 2,3-butanediol
Figure 8.15. Mixed acid (a) and butanediol (b) fermentation. 1, Enzymes of the
Embden-Meyerhof-Parnas pathway; 2, lactate dehydrogenase; 3, pyruvatefor
mate lyase; 4, formate-hydrogen lyase; 5, acetaldehyde dehydrogenase; 6, alcohol
dehydrogenase; 7, phosphotransacetylase; 8, acetate kinase; 9, PEP carboxylase;
10, malate dehydrogenase, fumarase, and fumarate reductase; 11, a-acetolactate
synthase; 12, a-acetolactate decarboxylase; 13, 2,3-butanediol dehydrogenase.
by NADH. Instead, the synthesis of pyruvate-formate lyase is induced

under anaerobic conditions. The reaction catalyzed by this enzyme pro-
ceeds in two steps with an acetyl-enzyme as intermediate and formate and
acetyl-CoA as products:
CHrCO-COOH + enzyme ---- CHrCO-enzyme + HCOOH
CHrCO-enzyme + CoASH ---- enzyme + CHrCO-SCoA
Pyruvate-formate lyase is irreversibly and rapidly inactivated under air
so that it functions only in fermentative metabolism of the enterobacteria.
Apparently even under anaerobic conditions the active enzyme is not very
stable; at low concentrations of pyruvate it changes over to an inactive
form which can again be reactivated. This reactivation requires the
presence of four components; reduced flavodoxin, activating enzyme
(enzyme II), S-adenosylmethionine, and pyruvate. The latter is not
consumed in the activation reaction; thus, it functions as positive effector.
pyruvate-formate lyase pyruvate-formate lyase
(inactive) (active)
+ enzyme II
+
reduced flavodoxin (pyruvate)
) flavodoxin
+ +
S-adenosylmethionine methionine + 5'-deoxyadenosine
S-adenosylmethionine is reductively cleaved in this activation reaction.
How and where the lyase actually is modified is not known.
The advantage of the pyruvate-formate lyase over the pyruvate dehy-
drogenase complex in fermentative metabolism is apparent: the formation
of acetyl-CoA is not accompanied by the reduction of NAD+.
B. a-Acetolactate synthase
The third enzyme system acting upon pyruvate is a-acetolactate synthase.
This enzyme is also involved in 2,3-butanediol formation by bacilli (see
Fig. 6.26). It contains thiamine pyrophosphate. First, enzyme-bound
hy.droxyethyl-thiamine pyrophosphate and CO 2 are formed from pyru-
vate. Active acetaldehyde is then transferred to a second molecule of
pyruvate:
CH)-CO-COOH + (H)TPP-E ----> CH)-CH-TPP-E + CO 2
I
CH)-CH-TPP-E OH
I CH)-CO
OH
+ I + (H)TPP-E
CH)-CO-COOH CH)-C-COOH
I
OH
This synthase is formed and is active under slightly acidic conditions,

and it is referred to as the pH 6 enzyme. Thus, a decrease of pH in the
environment of Enterobacter aerogenes leads to an increase of 2,3-

butanediol formation. Consequently less acids can be produced from
pyruvate.
The pH 6 enzyme is distinct from the anabolic a-acetolactate synthase,
which is involved in valine synthesis. This enzyme is most active at pH 8
(pH 8 enzyme) and is subject to feedback inhibition of L-valine.
C. Formate-hydrogen lyase
Species belonging to the genera Shigella and Erwinia do not contain
formate-hydrogen lyase; they produce considerable amounts of formate.
Escherichia coli and Enterobacter aerogenes contain this activity when
grown on sugars under anaerobic conditions, and formate can be cleaved
into CO z and Hz.
Formate-hydrogen lyase is not one single enzyme entity. The formation
of Hz and COz from formate is the result of the combined activity of a
special formate dehydrogenase (FOH II ) and a hydrogenase. FOH II is
under redox control, and formate cleavage is observed only, if this
compound cannot function as electron donor for the nitrate reductase or
fumarate reductase. In other words, if nitrate or fumarate is present,
formate is oxidized by another formate dehydrogenase (FOH,), and from
there the electrons are channelled to nitrate or fumarate, but not to H+
(Fig. 8.16). Thus, Hz evolution is not observed in the presence of nitrate or
fumarate.
All the enzyme systems mentioned above are membrane-bound. The
formate dehydrogenases are selenoproteins containing iron-sulfur centers,
molybdenum in the form of Moco-factor and cytochrome b.
D. Decarboxylations coupled to
membrane energization
Like several lactic acid bacteria, some enterobacteria are able to grow with
citrate under anaerobic conditions. Citrate is cleaved by citrate lyase, and
the oxaloacetate formed is decarboxylated to yield pyruvate (see section
II.G of this chapter). Stern and collaborators showed a number of years
ago that growth of Enterobacter aerogenes on citrate depended on the
presence of sodium ions. Oimroth demonstrated that Na + was required by
the oxaloacetate decarboxylase which is a biotin-containing and mem-
brane-associated enzyme. Interestingly enough this enzyme manages to
couple the decarboxylation reaction with the generation of an electrochem-
ical gradient of sodium ions as depicted in Fig. 8.17. This gradient can be
transformed into a pH gradient that can be taken advantage of by the ATP
synthase. The decarboxylation of oxaloacetate is associated with a free
energy change of IiGo, = -30 kJ (-7.2 kcal) mol-I, and one could
expect synthesis of 1/3 ATP per 1 CO z formed. The uptake of 3 H+ per
NO:;
r nitrate
reductase
1'------'.. . . NO;:
cyt b S56
t
UQ
t
HCOOH
FOH 1 Fe-S cyt b _ 2e-
t
MK
t
L ,.--------,/
cyt b
fumarate
fumarate
reductase
succinate
HCOOH
2W
FOHn Fe-S cyt b _ 2e- _ hydrogenase
2W + CO " H2/;:/
2
Figure 8.16. Formate-hydrogen lyase reaction and relationship ~ate and
fumarate reductases. Only when NO) or fumarate is not available ~i-i;formed.
oxa!oacelate
a
CO 2 + pyruvate
Na'
b
nAOP + nP i
H' c
nATP
Figure 8.17. Sodium-dependent 0 loacetate decarboxylase. a: Sodium transloca-

tion as coupled to the decar\)oxylation reaction. b: Na + - H+ antiporter.
c: Proton-translocating ATP synthase; n may be in the order of 1/3.
ATP synthesized would be in agreement with this figure. Three remarks

are necessary in this context:
1. Anaerobic citrate degradation in most organisms is not sodium-
dependent. This is true for lactic acid bacteria, phototrophs, and
clostridia. Here, oxaloacetate decarboxylase is a soluble enzyme not
containing biotin. E. coli is able to utilize citrate anaerobically in the
presence of a cosubstrate such as glucose; it does not contain oxaloace-
tate decarboxylase, and the oxaloacetate formed is reduced to succinate
(with the reducing power from the degradation of glucose to acetate).
Thus, the sodium-dependent oxaloacetate decarboxylase does not seem
to be widespread among the anaerobes.
2. Other decarboxylation reactions are coupled to Na + translocation as
well. This has been shown for methylmalonyl-CoA decarboxylase of
Veillonella alcalescens and glutaconyl-CoA decarboxylase of Acidami-
nococcus fermentans and Clostridium symbosium.
methylmalonyl-CoA
---
Na+
propionyl-CoA + CO 2
glutaconyl-CoA
---
Na+
crotonyl-CoA + CO 2
The enzyme mentioned first participates in propionate fermentation,
the second one in a pathway for the anaerobic breakdown of L-
glutamate.
Decarboxylation of methylmalonyl-CoA is the only energy-yielding
reaction, when Propiogenium modestum grows on succinate:
succinate + H+ --- propionate + CO 2 AGO' = -20.6 kllmol
3. We have seen that not only substrate level phosphorylation is used by
anaerobes for ATP synthesis and not only ATP hydrolysis for energiza-
tion of the membrane. Some lactic acid bacteria take advantage of
product efflux for the generation of an electrochemical gradient; here,
anaerobes employ decarboxylation reactions for this purpose.
The next sections of this chapter will show that electron transport is also
used by several groups of obligate anaerobes for the generation of a
protonmotive force across the membrane.
V. Propionate and Succinate Fermentation

Propionate is a major end product of fermentations carried out by a variety
of anaerobic bacteria. Many of them ferment glucose to propionate,
acetate, and CO 2 :
1.5 glucose --- 2 propionate + acetate + CO 2
A preferred substrate of propionate-forming bacteria is lactate, so that
Propionate and Succinate Fermentation 243
these organisms can grow with the major end product of the lactate
fermentation:
3 lactate - 2 propionate + acetate + CO 2
There are two pathways for propionate formation from lactate; in the
acrylate pathway lactate is reduced stepwise to propionate; in the succi-
nate-propionate pathway lactate is converted to propionate via pyruvate
and succinate.
A. The acrylate pathway

This pathway seems to occur only in a few microorganisms, e.g., in
Clostridium propionicum and in Megasphaera (Peptostreptococcus) elsde-
nii. It is shown in Fig. 8.18; L-, D, or DL-Iactate may serve as substrate; a
racemase is present which interconverts the enantiomers. L-Lactate is
converted to L-lactyl-CoA in a CoA transferase reaction. By reactions not
yet established in detail acrylyl-CoA is formed. It is reduced to propionyl-
CoA, and propionate is produced by the above-mentioned CoA trans-
ferase.
OH OH
I 1
(L)2CH J -(,-COOH 2CH J -C-CO-CoA
I 1
H H
Ul t3
H
I
(0) CH J -C -COOH
f- 2H i O
I 2CH 2 =CH-CO-CoA
OH
~ ETF
4
~----~. ETF' H2 , 1
sum: 3 lactate - - - - -......~ 2 propionate + acetate + CO 2 + H 2 0

Figure 8.18. Formation of propionate, acetate, and CO 2 from DL-Iactate by
Megasphaera elsdenii and Clostridium propionicum. 1, Lactate racemase; 2, CoA
transferase; 3, reaction not established; 4, dehydrogenase, which employs reduced
electron-transferring flavoprotein (ETFH 2 ) as H-donor; 5, D-Iactate dehy-
drogenase; 6, pyruvate-ferredoxin oxidoreductase; 7, transhydrogenase; 8, phos-
photransacetylase + acetate kinase.
The H-donor for acrylyl-CoA reduction is reduced electron-transferring

flavoprotein. It is formed from D-lactate and from reduced ferredoxin (or
flavodoxin). The ATP yield of this fermentation is 1 mol/3 mol of
lactate.
C. propionicum also ferments alanine and acrylate to propionate.
B. The succinate-propionate pathway

This pathway is employed by most propionate-producing organisms.
Succinate is an intermediate but is also produced as end product in small or
large amounts. On the other hand, organisms using the acrylate pathway
do not excrete significant amounts of succinate.
The establishment of the succinate-propionate pathway was a rather
difficult task. As is shown in Fig. 8.19 several enzymes are involved. First,
lactate is oxidized to pyruvate in a reaction requiring a flavoprotein as
ICH1-CHOH-COOH I
6 ~2H
CH1-CH2 -CO-CoA CH1-CO-COOH
~ biotin- C0 2
2
9H1 ~ biotin _ - - - /
HOOC-C-CO-CoA(S)
~ ls
I;"
HOOC-C-CO-CoA(R)
I HOOC-CH2-CO-COOH
8
0" ~"ym, ~NADH+W
~NAD+
HOOC-CH 2 -CH 2 -CO-CoA HOOC-CH2-CHOH-COOH
6
2H
~
~4
H2 0
HOOC-CH 2 -CH 2 -COOH ~ HOOC-CH=CH-COOH
@P+H~~
sum: lactate + NADH + H+ + ADP + Pi - propionate + NAD+ + ATP + 2H 20
Figure 8.19. Fermentation of lactate via the succinate-propionate pathway by
propionibacteria. 1, Lactate dehydrogenase (the H-acceptor is probably a
flavoprotein); 2, (S)-methylmalonyl-CoA-pyruvate transcarboxylase; 3, malate
dehydrogenase; 4, fumarase; 5, fumarate reductase; 6, CoA transferase; 7,
(R)-methylmalonyl-CoA mutase; 8, methylmalonyl-Co-A racemase.
H-acceptor. Oxaloacetate is then formed in a transcarboxylation reaction

with (S)-methylmalonyl-CoA as COz-donor and biotin as COz-carrier. The
action of malate dehydrogenase and fumarase yields fumarate, which is
reduced to succinate by fumarate reductase. This reduction reaction is
coupled to ATP formation by electron transport phosphorylation. Suc-
cinyl-CoA is then formed in a CoA transferase reaction and the rearrange-
ment as catalyzed by the coenzyme B 12-containing methylmalonyl-CoA
mutase leads to (R)-methylmalonyl-CoA, which is not a substrate for the
transcarboxylase. Rather, the (S)-enantiomer is formed by a specific
racemase. Then transcarboxylation yields propionyl-CoA and CoA trans-
fer to succinate finally yields propionate.
One NADH is consumed in propionate formation from lactate; it comes
from lactate oxidation to acetate according to the overall fermentation
equation given above.
Besides the transcarboxylase-a biotin-containing enzyme with a high
molecular weight (approximately 800,000 daltons) and a very complex
quarternary structure-two enzymes of the succinate-propionate path-
way deserve special attention: methylmalonyl-CoA mutase and fumarate
reductase.
C. Methylmalonyl-CoA mutase and other

coenzyme B12 -dependent rearrangement
reactions
Rearrangements of this type were discovered by Barker and collaborators
when they investigated the fermentation of glutamate by Clostridium
tetanomorphum. As is apparent from Fig. 8.20 glutamate and succinyl-
CoA are rearranged in analogous reactions to yield l3-methylaspartate and
methylmalonyl-CoA, respectively. The principle of these reactions is that a
substituent group is moved between two adjacent positions of the carbon
skeleton while a hydrogen is moved in the opposite direction. Not only
carbon-carbon bonds are rearranged in coenzyme BIz-dependent reac-
tions. A number of dehydratases, deaminases, and amino mutases are also
B l2 -enzymes and catalyze analogous reactions (Fig. 8.20c-e).Glycerol
dehydrase, which is present in some lactobacilli, converts glycerol into
l3-hydroxypropionaldehyde. Ethanolamine deaminase is present in cho-
line-fermenting clostridia and 13-lysine mutase is the second enzyme in
clostridial L-Iysine fermentation (see Chapter 8, Section X).
Coenzyme B12 is not identical with vitamin B\2. The latter is a corrin ring
system with cobalt 2 + as central metal atom and 5,6-dimethylbenzimidazole
ribonucleotide as characteristic component (Fig. 8.21). The sixth coordina-
tion position of C0 2 + is occupied by hydroxyl or cyanide (hydroxy- or
cyanocobalamin). Coenzyme B l2 contains in addition a 5'-deoxyadenosyl
group, which is covalently bound to cobalt replacing cyanide (5'-
deoxyadenosylcobalamin). In the rearrangement reactions the hydrogen is
a glutamate mutase
r-----:-l
I COOHI
I I I
I H 2N-C-H
.. TH3
..
I
L----l---.J ~
I H-C-H ... HOOC-C-C-H
I I r-il.J I I
L_H_C~H NH 2 COOH
10..:.;..
COOH
L-g1utamate threo-{3-methyl-L-aspartate
b methylmalonyl...(:oA mutase
,.-------,
~--T.9-:.S~o!-'
I H-C-H"',
1,...--1
.. ?H 3
...
I
~H-CtH_J CoAS-OC-C-H
tOOH tOOH
succinyl...(:oA (R )-methylmalonyl...(:oA
c glycerol dehydrase
H
r-,
I
I
H.....C-OH ...
H 20
~_;.J I r---f j ..
~H-C"-OH I
I L ---'
H 2 C-OH
glycerol l3-hydroxypropionaldehyde
d ethanolamine deaminase
H
r-.,I
I H..... C-OH~
~_.J I r--' CHO
I
CH 3
~H2Ct~2J
ethanolamine acetaldehyde
e l3-iysine-5,6-aminomutase
COOH <[OOH
yH
I
2 <[H 2
H 2 N-y-H
.. .. H 2 N-<[-H
CH 2 CH 2
,.--., I I
I H~C-H'" H N-C-H
~-_ .. I ,...--, 2 I
'+- H 2 C.+NH
L __2 ..JI CH 3
13-lysine 3,5-diaminohexanoate
Figure 8.20. Coenzyme B t2 -dependent rearrangement reactions.
transferred from the substrate to coenzyme B 12 (to the C-5' methylene

group) and from there to the product.
D. Fumarate reductase
The reduction of fumarate to succinate is a process that can be coupled by
anaerobes with the generation of an electrochemical proton gradient and
subsequently with ATP synthesis. That ATP is formed in this reaction is
S. 6-dimethyl-
benzimidazole
ribonucleotide
corrin ring
CN
Icyanocobalamin I CH 2
<N'~~
-:P' ~ ..J
l/o~ _ N- - - - - - - ,
S'-deoxyadenosine ~ IS'-deoxyadenOSylcobalamin!
OH OH
Figure 8.21. Structure of vitamin B!2 and coenzyme B!2'
indicated by the high growth yields of propionate-succinate-producing

organisms. Moreover, an ATP synthesis coupled to the reduction of
fumarate could be demonstrated in cell-free systems of some organisms,
first with extracts of Desulfovibrio gigas by Peck, Le Gall, and Barton.
Fumarate reductase is membrane-bound, and it is associated with the
electron carriers menaquinone and cytochrome b. Bacteroides fragi/is
grows only slowly in the absence of hemin (precursor in cytochrome
synthesis) and ferments glucose to lactate, acetate, and malate. In the
presence of hemin, however, a cytochrome b is formed and glucose is
rapidly fermented to propionate and succinate indicating that cytochrome
b is an essential component of the fumarate reductase system. Bacteroides
amylophilus, is the only known exception; it contains a membrane-bound
fumarate reductase but no cytochromes.
The fumarate reductase system is present not only in the classic
propionate-forming microorganisms, the propionibacteria, but also in
several enterobacteria and in various species of the genera Bacteroides,
Veillonella, Peptostreptococcus, Ruminococcus, Succinivibrio, and Seleno-
monas. Even a clostridium (c. formicoaceticum) contains fumarate reduc-
tase. Some of these bacteria (such as E. coli and other enterobacteria)
form only succinate but no propionate.
The redox potential of the fumarate / succinate couple is Eo
= + 33 mV.
Thus, the potential span between donors such as Hz, formate, and NADH
is large enough to be coupled to proton translocation and ATP synthesis.
In addition, a-glycerol phosphate (E. coli) and lactate (most propionic
acid bacteria) have been shown to function as NAD+ -independent donors
for fumarate reductase. The entire system-the H-donating enzymes
(hydrogenase; NADH dehydrogenase, a-glycerol phosphate dehy-

drogenase, etc.), carriers, and the fumarate reductase-is membrane-
bound. The reductase is a flavoprotein containing iron-sulfur centers. The
arrangement of the components in the membrane, as unravelled for
Wolinella (Vibrio) succinogenes by Kroger, is depicted in Fig. 8.22.
In correspondence with the discussed energetics of the fumarate reduc-
tase system, organisms, such as W. succinogenes, E. coli, Citrobacter
freundii, and D. gigas are able to grow with fumarate plus H 2 or formate.
fumarate + H 2 - succinate
fumarate + formate - succinate + CO 2
E. PEP carboxytransphosphorylase
of propionibacteria
Since propionibacteria produce succinate in addition to propionate, the
transcarboxylase alone cannot be responsible for the formation of C4 -
dicarboxylic acids; an enzyme system must be present in these microorgan-
isms that catalyzes the synthesis of C4 -dicarboxylic acids from a C3
compound and CO 2 . It was discovered by Wood and Werkman in 1936.
Since then carboxylating reactions that yield oxaloacetate are called
Wood- Werkman reactions. We have already discussed the role of
Fe-S
Fe-S
cyt b(-200)
MK
2H+ + fumarate
succinate
Figure 8.22. Electron flow from Hz or HCOOH to fumarate. Boxes represent

hydrogenase, formate dehydrogenase, and fumarate reductase, respectively. Num-
bers in parentheses give redox potentials of the two b-type cytochromes; MK,
menaquinone.
Acetate Fermentation 249
pyruvate carboxylase and PEP carboxylase as Crcarboxylating enzymes.

In some Bacteroides species the energy-conserving PEP carboxykinase
reaction is involved in oxaloacetate formation (this enzyme functions
normally in the direction of PEP synthesis; see Table 5.4):
PEP carboxykinase
PEP + CO z + ADP , I oxaloacetate + ATP
Propionibacteria contain an enzyme that catalyzes an analogous reac-
tion:
PEP carboxytransphosphorylase I PP
PEP + CO z + P i I oxa oacetate + i
Here, the phosphoryl group is transferred to inorganic phosphate to

form pyrophosphate. Thus, propionibacteria contain two enzyme systems
that yield pyrophosphate: PEP carboxytransphosphorylase and pyruvate-
phosphate dikinase (see Table 5.4). The pyrophosphate formed is used to
phosphorylate fructose-6-phosphate to fructose-l,6-bisphosphate and
serine to phosphoserine.
It is apparent from the discussion of the enzyme systems in this and
the previous section that a number of variations exist in the propionate-
succinate pathway as depicted in Fig. 8.19. In some organisms (e.g.,
Veillonella alcalescens) the transcarboxylase is replaced by sodium-
dependent decarboxylase. Both enzymes yield propionyl-CoA from
methylmalonyl-CoA.
With respect to oxaloacetate formation from an intermediate of the
glycolytic pathway, one out of three enzymes might be involved: the
transcarboxylase using pyruvate as acceptor or the two enzymes discussed
above that use PEP as acceptor.
VI. Acetate Fermentation

We have seen that acetate is an -important product in a number of
fermentations. There is a group of organisms, however, by which acetate is
formed as the predominant nongaseous product. Butyrate is usually not
detectable in the fermentation broth; ethanol and lactate in very small
amounts at the most. This formerly very small group of acetogenic
organisms has grown very much in recent years. Representative species are
given in Table 8.9. It is obvious that most species are able to live at the
expense of acetate formation from Hz and CO z according to the following
equation:
4H z + 2CO z ----+ CH3 - COOH + 2HzO
e:.Co, = -107.1 kJ (-25.6 kcal) per mol acetate
This type of fermentation was discovered in 1936 by Wieringa who isolated
Clostridium aceticum. Several acetogens grow with carbon monoxide and
Table 8.9. Acetogenic bacteria and some of their properties
growth on
cytochromes
organism thermophilic H 2 + CO 2 sugars present
Clostridium aceticum + + +
C. thermoautotrophicum + + + +
C. formicoaceticum + +
C. thermoaceticum + _, +a + +
Acetobacterium woodii + +
A. wieringae + +
Acetogenium kivui + + + ?
Sporomusa sphaeroides + +
aSome strains are positive.
Eubacterium /imosum is able to produce acetate from CO 2 + H 2 , but it forms considerable
amounts of butyrate when growing on other substrates.
with methanol plus CO z :

4CO + 2H zO - - CH3 - COOH + 2CO z
IlC O' = -484.4 kJ (-115.9 kcal) per reaction
4CH 3 0H + 2CO z - - 3CH3 -COOH + 2H zO
IlCo, = -706.3 kJ (-168.9 kcal) per reaction
Hexoses are converted by acetogenic bacteria to almost 3 mol acetate
per mol:
C6 H 1Z 0 6 -- 3CH 3 -COOH
The recently described genus Sporomusa represents the first genus of
Gram-negative endospore-formers. In addition to Hz + 2CO z , its members
utilize N-methyl compounds (e.g., betaine) and primary alcohols (ethanol,
n-butanol).
How is it possible for the acetogens to make three molecules of acetate
from one molecule of hexose? The pathway is depicted in Fig. 8.23. First,
the hexose is degraded to two pyruvates via the Embden-Meyerhof-
Parnas pathway. Their subsequent degradation by the action of pyruvate;
ferredoxin oxidoreductase and the typical enzymes acting on acetyl-CoA
yield 2 acetates. In addition, COz is produced, and NADH and reduced
ferredoxin are generated from the corresponding oxidized forms. Synthesis
of the third molecule of acetate is now initiated by the reduction of COz all
the way to 5-methyltetrahydrofolate. The first enzyme of this pathway is a
remarkable formate dehydrogenase. It is a tungsten-selenoprotein and is
structured so as to make formate and not only to oxidize it like the
membrane-bound formate dehydrogenases of other organisms (e.g., en-
terobacteria). The electron donor is known for only some acetogenic
organisms; in C. thermoaceticum it is NADPH. Formyltetrahydrofolate is
Acetate Fermentation 251
fructose
2ADP + 2P i ---..... ...-- 2NAD+
2ATP 2NADH + 2H'
pyruvate pyruvate
COA~Fd
,
COA~Fd
2 FdH 2 2 FdH 2
--------., -------.,
acetyl-CoA acetyl-CoA
t
CO 2 CO 2
3 t - ADP+P j XH 2 3~ADP+Pj XH 2
COA+ATP COA~ ATP
10
I acetate I x I acetate I X+H 20
formate ICOl
ATP~ H4 F
ADP+ Pi 4" CoA
I 0-formyl-H 4 F
H0-1 6
acetyl-CoA
2 3 r ADP+ Pi
5, IQ-methenyl-H 4 F COAT"ATP
9
NADH+W~7
NAD+-1
I acetate I
5, 10-methylene-H 4 F
FdH 2 ~8
Fd-t
s-methyl-H 4 F BI2-E __- - J
Figure 8.23. Pathway of the acetate fermentation. 1, Degradation of fructose via
the Embden-Meyerhof-Parnas pathway; 2, pyruvate-ferredoxin oxidoreductase;
3, phosphotransacetylase plus acetate kinase; 4, formate dehydrogenase; 5,
formyl-tetrahydrofolate synthetase; 6, methenyl-tetrahydrofolate cyclohydrolase;
7, methylene-tetrahydrofolate dehydrogenase; 8, methylene-tetrahydrofolate re-
ductase; 9, tetrahydrofolate: B 12 methyltransferase; 10, CO dehydrogenase; 11,
acetyl-CoA-synthesizing enzyme (probably ATP-requiring); [CO), enzyme-bound.
then formed from formate in an ATP-consuming reaction, and subsequent

reduction leads to the previously mentioned methyltetrahydrofolate.
How acetate is synthesized from methyltetrahydrofolate was the subject
to intense research in the laboratories of Wood and of Thauer. The
breakthrough came with the discovery of carbon monoxide dehydro-
genase-a nickel-containing enzyme. It is different from CO oxidase,
the characteristic enzyme of the carboxydobacteria, in that the enzyme-
specific electron carrier has a low redox potential. That reaction is,
therefore, reversible under physiological conditions:
CO 2 + X-H 2 ) CO + X + H 2 0
In the pathway, CO 2 is now reduced to CO, and it is the latter which finally
gives the carboxyl group of acetate: methyltetrahydrofolate is carbony-
lated, and acetyl-CoA and finally acetate are formed.
The pathway as depicted in Fig. 8.23 also allows to outline the routes
used for acetate formation from methanol + CO 2 and from H 2 + CO 2 ,
The strategy is to make CO from CO 2 and to make methyltetrahydrofolate
from H 2 + CO 2 or methanol. Thus, part of the methanol has to be
oxidized to provide the reducing power for CO 2 reduction to CO;
methyltetrahydrofolate and CO finally yield acetate. The energetics of the
latter fermentations is not clear yet. Likewise, the function of the
cytochromes in acetogens is unknown.
VII. Methane Fermentation

Methane is the most reduced organic compound and its formation is the
terminal step of the anaerobic food chain that will be discussed in Section
IX of this chapter. Methanogenesis is a domain of the Archaebacteria; in
other words, all known methanogenic bacteria belong to this kingdom of
organisms. Some of their characteristic features have already been outlined
in Chapter 5.
A. Substrate utilization
Methanogenic bacteria are extremely oxygen-sensitive. This is not a great
disadvantage to them in nature. In habitats rich in degradable organic
compounds, oxygen is consumed rapidly and trapped by organisms in
surface layers. Thus the methanogenic bacteria are particularly abundant
in all sorts of mud and sediments. Other important habitats of these
bacteria are the rumen and the (man-made) anaerobic digesters in sewage
plants. The oxygen sensitivity of the methanogens creates problems when
pure cultures are to be isolated and experiments with pure cultures are to
be carried out. Appropriate methods (the so-called Hungate technique)
have been worked out for this purpose.
Representatives of various genera of methanogenic bacteria, the sub-
strates utilized by them, and the distribution of cytochromes are summa-
rized in Table 8.10. It is apparent that complex organic compounds cannot
be utilized by the methanogens. Substrates are C 1 compounds and as the
only C 2 compound: acetate. Two nutritional groups of organisms can be
envisaged:
Methane Fermentation 253
Table 8.10. Representative species of the methanogenic bacteria, the substrates

utilized by them, and the presence of cytochromes
presence of
organism substrates utilized cytochromes
Methanobacterium
thermoautotrophicum
Methanobrevibacter arboriphilus Hz + COz
Methanococcus vanniellii Hz + CO z , HCOOH
Methanospirillum hungatei Hz + CO z , HCOOH
Methanosarcina barkeri Hz + COz, CH 3 0H, +
CH 3 COOH,
methylamines
Methanosarcina mazei CH 3 0H, CH 3 COOH, +
methylamines
Methanothrix soehngenii CH 3 COOH +
Methanolobus tindarius CH 3 0H, methylamines +
Methanococcoides methylutens CH 3 0H, methylamines +
Methanoplanus limicola Hz + COz, HCOOH
a CO allows only slow growth; it may also be used by other methanogens listed.
1. Obligate chemolithotrophic methanogens that grow with CO z + Hz

according to the equation:
CO z + 4H z ----+ CH 4 + 2H zO AGo, = -136 kJ (- 32.4 kcal)
Some of these organisms also grow with the "quasi-chemolithotrophic"
substrates HCOOH and CO. This term is adequate here because both
substrates are utilized via CO z + Hz:
4HCOOH ----+ 4CO z + 4H z (carrier-bound)
CO z + 4H z ----+ CH 4 + 2H zO
4HCOOH ----+ CH 4 + 3CO z + 2HzO
AGo, = -144.2 kJ (-34.5 kcal) per mol methane
correspondingly:
4CO + 2HzO ----+ CH 4 + 3CO z
AGo, = -211 kJ (-50.4 kcal) per mol methane
2. Methylotrophic methanogens that grow with methyl-group-containing
substrates (methanol, methylamines, acetate). The fermentation equa-
tion for acetate is very simple:
CHrCOOH ----+ CH 4 + CO z
AGo, = -37 kJ (-8.9 kcal) per mol methane
Organisms such as Methanosarcina barkeri grow on methanol or

methylamines. Here, one-fourth of the substrate has to be oxidized to
CO z for reducing power generation:
CHjOH + HzO - - + CO z + 6H
3CHj OH + 6H - - + 3CH4 + 3HzO
4CHj OH - - + 3CH4 + CO z + 2H zO
ilC o, = -319.5 kJ (-76.4 kcal) per reaction
ilC o, (per mol methane) = -106.5 kJ (-25.5 kcal)
or
4(CHj h-N + 6HzO - - + 9CH4 + 3COz + 4NHj
ilC o, = -683.2 kJ (-163.4 kcal) per reaction
ilC o, (per mol methane) = -75.9 kJ (-18.1 kcal)
Group 2 organisms produce methane directly from the methyl groups
and not via CO z . This has been demonstrated with deuterated substrates
that gave methane with three deuteriums in it:
CDjOH fermentation) CDjH
Obligate chemolithotrophic methanogens do not contain cytochromes,
which are, however, present in methylotrophic methanogens. The metha-
nogenic bacteria require Na + in concentrations of about 5 mM for growth;
its function is still unknown. A number of rather unique redox carriers and
coenzymes have been found to occur in all of these organisms; they do not
occur in other organisms (few exceptions) and are specifically involved in
the pathway from CO z to CH 4. It should be mentioned that most of these
were first isolated from Methanobacterium thermoautotrophicum, an
organism that was described by Zeikus and Wolfe and that grows in a
mineral medium with Hz + COz at 55C.
B. Novel coenzymes
The first two novel coenzymes discovered in methanogens by Wolfe and
co-workers were coenzyme M and coenzyme F420. Their chemical struc-
tures are given in Fig. 8.24. Coenzyme M is a simple chemical compound.
Its reactive group is the mercapto group which can be methylated and
methyl-coenzyme M is the ultimate precursor of methane. Coenzyme F 4Z0
is a deazaflavin (ring-N in position 5 of the flavin skeleton is missing); it is a
redox carrier with an Eo = -370 mY, and its role is analogous to the one
of ferredoxin in other anaerobes. It functions as electron acceptor of
hydrogenase and as electron donor in several reduction reactions.
The structure of factor F4jO which is involved in the final step of methane
formation was elucidated by Thauer, Eschenmoser, and co-workers as a
tetrapyrrol with nickel as central metal ion. In addition to hydrogenase and
CO dehydrogenase, both of which are nickel proteins, factor F4jO repre-
coenzyme M
factor F430
5,6,7, 8-tetrahydromethanopterin
Figure 8.24. Novel coenzymes in methanogenic bacteria,
sents the third component of methanogens containing this metal. Two
additional structures are given in Fig. 8.24: that of tetrahydromethano-
pterin determined by Vogels and co-workers, and of methanofuran. Both
are involved in CO 2 reduction. Methanofuran functions as the primary
CO 2 acceptor.
C. Pathway of methane formation

First we shall look at methane formation from CO 2 + H 2 . A sequence
based on Barker's scheme of methane formation from CO 2 is given in
Fig. 8.25. The first carrier molecule known to be involved is methano-
furan. In a reaction that requires CO 2 and reducing equivalents it is
converted to formylmethanofuran with the formyl group residing at the
aminomethyl group of the furan ring. Transfer of the C 1 moiety to
tetrahydromethanopterin and reduction of the formyl to the methyl group
follows. In analogy to the tetrahydrofolate biochemistry the methyl group
resides at N5 of the ring skeleton. It finally is transferred to coenzyme M
and reduced to methane. The methyl-coenzyme M methylreductase reac-
tion has been studied in detail; the responsible enzyme is of a very complex
structure; it contains several proteins, factor F430 and an as yet unidentified
factor B.
MF-H
MF-CHO
~
THMP-H
420
II H2 MF-H
THMP-CH 2 0H
III
H'~H'O
THMP-CH)
THMP-H --t-COM-SH
CoM-S-CH)
IV
ATP
Figure 8.25. Scheme for the reduction of CO 2 to CH 4 and site of ATP synthesis.
MF-H, Methanofuran; THMP-H, tetrahydromethanopterin; 1, methyl-coenzyme
M methylreductase.
The free energy changes and the redox potentials of the four reactions
leading from bicarbonate to methane are as follows:
IiGo, liEn
HCOi + Hz HCOO- + HzO -1.3 kJ ( -0.3 kcal) -432 mV
HCOO- + Hz + H+ CHzO + HzO +23.0 kJ ( +5.5 kcal) -535 mV
CHzO + Hz CH 3 0H -44.8 kJ (-10.7 kcal) -182 mV
CH 3 0H + Hz CH 4 + HzO -112.5 kJ (-26.9 kcal) +169 mV
-135.6 kJ (-32.4 kcal)

(values per mol product)
It is obvious that the first two steps are not very favorable. The third
step is distinctly exergonic; most of the energy, however, is released in the
fourth step. The values given are for the free redox couples (e.g.,
CHzOjCH3 0H). They might be slighly different for the carrier-bound
compounds.
It can be stated now that the fourth reduction step is coupled to the
generation of a protonmotive force at the membrane which in turn is used
by an ATP synthase for the phosphorylation of ADP. The evidence comes
from experiments with Methanosarcina barkeri. Reduction of CH3 0H to
CH 4 by Hz is coupled in this organism to an increase of the intracellular
ATP concentration. This ATP synthesis and also methane formation are
inhibited by dicyclohexylcarbodiimide (DCCD)-the classic inhibitor of
the ATP synthase (Fig. 8.26). The protonmotive force, 6.P, however,
remains unaffected (because it cannot be utilized for ATP synthesis in the
presence of DCCD). The two processes-methane formation and ATP
synthesis-are apparently coupled via 6.P. As expected, this coupling can
be abolished by an uncoupler. In its presence, the protonmotive force is
dissipated and methane is produced again from CH 3 0H + Hz, but ATP is
not synthesized. The sequence of effects is therefore: methane
formation ~ protonmotive force ~ ATP synthesis and not: methane
formation ~ ATP synthesis ~ protonmotive force. It excludes that ATP
is formed by substrate-level phosphorylation. Since methane is produced
from methanol + Hz in one reduction step-in the methylreductase
reaction-this step must be the site of energy conservation by a che-
miosmotic mechanism.
Methanosarcina barkeri is the classic, and representative species for
those methanogenic bacteria that utilize acetate, methanol, and methyla-
mines as substrates. Growth on acetate is much slower as compared to that
on the other substrates. Nevertheless, it is the most important metha-
nogenic substrate in nature (see Section IX). Looking at the fermentation
equation may give the false impression that methanogenesis from acetate is
a simple decarboxylation. After the discovery of high levels of CO
DCCD uncoupler
! !
Figure 8.26. Coupling of ATP synthesis to methane formation from CH 3 0H + Hz

in M. barkeri by a chemiosmotic mechanism [Redrawn from M. Blaut and
G. Gottschalk; Eur. J. Biochem. 141,217-222 (1984)].
dehydrogenase in acetate-grown cells of M. barkeri by Zeikus and co-

workers, it became more and more clear that CO is the primary cleavage
product and not CO z . The fermentation scheme in Fig. 8.27 shows that the
fermentation can then be written as an oxidoreduction process: first CO
and methyl-coenzyme M are produced, the oxidation of the former pro-
vides the reducing equivalents for the reduction of the latter to methane.
From the free energy change (- 31 kJ) it is apparent that the formation of
1 mol methane from acetate can yield only a fraction of a mol ATP.
It has already been mentioned that methanogenesis of methanol and of
the methylamines can be subdivided into two processes: oxidation of
one-fourth of the methyl groups to CO z and reduction of three-fourths of
the methyl groups to CH 4 (Fig. 8.28). A methanol: coenzyme M and a
trimethylamine: coenzyme M methyltransferase have been characterized.
The methyl group is first transferred to protein-bound 5-hydroxy-
benzimidazolylcobamide which is the principal form of BIz in methanogens
(not the dimethyl derivative as in the propionibacteria). Methyl group
transfer proceeds further to coenzyme M. The carrier at which the methyl
group is oxidized to CO z is unknown. Very likely the cytochromes present
in those methanogens that grow on methanol and methylamines are
involved in the first oxidation step (X-CH 3 to X-CHzOH). Coupling
of this oxidation ( 0 = -182 mV) with reduction of factor F 4Z0
(0 = -373 mV) requires reverse electron transfer that might proceed
with the participation of cytochromes.
jCHJ-COOHI
HX~
+
t H 20
CHJ-CO-X
ICOJ----/''''------i~CoM-S-CHJ
CoM-SH
ATP
Figure 8.27. Tentative scheme for the formation of methane and carbon dioxide
from acetate. 1, methyl-coenzyme M methylreductase; 2, CO dehydrogenase.
Details of the other reactions are not yet known.
~H ----.J....---__
XH
3CoM-S-CH)
ATP
Figure 8.28. Tentative scheme for the formation of methane and carbon dioxide
from methanol. 1, Methanol: coenzyme M methyltransferase (transfer via B 12 ; see
text); X, unknown carrier.
D. Synthesis of cell carbon

M. thermoautotrophicum and a number of other species are able to grow in
a mineral medium. Therefore, these organisms must be able to make all
their cellular constituents from COz. Fuchs and co-workers have shown
that the methanogens do not operate a cyclic COz fixation pathway as do,
for example, the phototrophs (see Chapter 9). The strategy is to synthesize
first acetyl-CoA from 2CO z . For this purpose the organisms use a pathway
that resembles the one present in acetogens (Fig. 8.23). Advantage is
taken of the fact that carrier-bound methyl groups are intermediates of
methane formation anyway. Not all of them are used to produce
methane; some are carbonylated to give acetyl-CoA. The latter is reduc-
tively carboxylated to yield pyruvate. Ferredoxin or coenzyme F4zo serve as
H-donor. PEP synthetase and PEP carboxylase lead to oxaloacetate
that is reduced to succinate; in M. thermoautotrophium, finally succinyl-
CoA is reductively carboxylated, and a-oxoglutarate-the precursor of
L-glutamate-is formed:
succinyl-SCoA + COz + FdH z ~ a-oxoglutarate + CoASH + Fd
Thus, in contrast to Clostridium kluyveri which also uses acetyl-Co A as
starting material for biosyntheses, several methanogens lack citrate syn-
thase, and employ the "dicarboxylic acid portion" of the citric acid
cycle for glutamate synthesis rather than the "tricarboxylic acid portion".
M. barkeri, however, has been shown to contain citrate synthase and to
employ the "tricarboxylic acid portion" of the cycle.
VIII. Sulfide Fermentation (Desulfurication)

Most microorganisms use sulfate as the principal sulfur source and contain
enzyme systems for the reduction of sulfate to sulfide. This process of
assimilatory sulfate reduction has been discussed in Chapter 3 (see
Fig. 3.3). In sulfide fermentation, sulfate is used as terminal electron
acceptor, and the hydrogen sulfide formed is excreted. This process is
therefore called dissimilatory sulfate reduction; it is carried out only by
strictly anaerobic bacteria. On the basis of their oxidative abilities the
sulfate-reducing bacteria (sulfidogenic bacteria) can be subdivided into two
groups. Representative species of these groups are given in Table 8.11.
The incomplete oxidizers, such as the Desulfovibrio species and most
Desulfotomaculum species, have been known for a long time; they oxidize
a number of organic acids and alcohols to acetate. The complete oxidizers
were recently discovered by Pfennig and co-workers. Desulfotomaculum
acetoxidans was the first; others, morphologically very different, followed.
Sulfate-reducing bacteria are loaded with electron and hydrogen car-
riers. Cytochrome C3 discovered by Postgate, was the first cytochrome
'J)
Table 8.11. The two physiological groups of sulfate-reducing bacteria and some of their
S;
'"
properties 0-
"
~
growth 3
with "a
species substrates utilized a Hz + CO z cytochromes ~
o'
::l
Group I '8
Desulfovibrio desulfuricans lactate, ethanol, malate +b c "'"
Desulfovibrio vulgaris lactate, ethanol, malate +b c '~"
;:l.
Desulfomonas pigra lactate c n
~
Desulfotomaculum nigrificans lactate, ethanol +b b o'
Desulfobulbus propionicus propionate b, c 2,
Group II
Desulfotomaculum acetoxidans acetate + b
Desulfobacter postgatei acetate b, c
Desulfococcus multivorans acetate, propionate, b, c
benzoate,
fatty acids (C 1 - C 14 )
Desulfonema limicola formate, acetate, + b, c
propionate,
fatty acids (C 1 -C 12)
Desulfosarcina variabilis acetate, propionate, + ND
benzoate,
fatty acids (C 1 - C 14 )
NO, not determined.

Q Substrates utilized in the presence of sulfate;
b Group 1 organisms require acetate for growth with Hz + CO z . tv
~
detected in strict anaerobes. All sulfate reducers contain cytochromes,

which are of the c-type and/or of the b-type. In addition, menaquinone,
several ferredoxins, flavodoxin, rubredoxin, and desulforedoxin (closely
related to rubredoxin) are present. Another remarkable constituent of
sulfate reducers is a siroheme protein, a protein containing two tetrapyrrol
ring systems and iron sulfur centers. This protein exhibits sulfite reductase
activity and may catalyze six-electron-transfer reactions; it is known as
desulfoviridin (Desulfovibrio, Desulfomonas, Desulfococcus, Desulfonema
species) or P582 (Desulfotomaculum, Desulfonema species). First the
fermentation of lactate by the incomplete oxidizers will be discussed.
A. Fermentation of lactate and sulfate

This fermentation can be summarized as follows:
2CH 3-CHOH-COOH + 2H 20 ~ 2CH 3-COOH + 2C0 2 + 8H
SO~- + 8H ~ S2- + 4H 20
Lactate oxidation to acetate proceeds via pyruvate and acetyl-CoA. A
flavoprotein probably functions as H-acceptor in the first oxidation step,
and pyruvate-ferredoxin oxidoreductase is involved in acetyl-CoA forma-
tion. The final acetate production is coupled to ATP synthesis.
The 8H generated in lactate oxidation are used to reduce sulfate to
sulfide. Prior to the first reduction step, sulfate is activated by conversion
to adenosine-5'-phosphosulfate (APS) (Fig. 8.29), and the reduction
products are sulfite and AMP. The further phosphorylated compound
PAPS, which is an intermediate in assimilatory sulfate reduction (see
Fig. 3.4), is not involved here. APS reductase contains FAD and iron-
sulfur. The reaction is reversible; in vitro APS reduction is observed with
reduced methyl viologen as H-donor and APS formation from sulfite and
AMP with ferricyanide as H-acceptor.
The reduction of sulfite is catalyzed by the above-mentioned sulfite
reductase. Until recently, the intermediary formation of trithionate and
thiosulfate was assumed. It seems now that free intermediates do not occur
during reduction of sulfite to sulfide.
Two aspects of dissimilatory reduction of sulfate to sulfide are interest-
ing: the mode of H-transfer from lactate oxidation reactions to sulfate
reduction reactions and the bioenergetics of sulfate reduction.
It has been demonstrated for Desulfovibrio species that one hydrogenase
and its electron acceptor, cytochrome C3' are localized in the periplasmic
space. Other cytochromes, menaquinone, other redox carriers and lactate
dehydrogenase are situated in the membrane whereas all the enzymes for
pyruvate oxidation, for sulfate reduction, and a second hydrogenase are in
the cytoplasm. This enzyme localization led Peck and co-workers to
assume that a "hydrogen cycling" occurs in sulfate reducers. As indicated
in Fig. 8.29, the reducing equivalents are released by the cytoplasmic
Sulfide Fermentation (Desulfurication) 263
membrane
I I
cytoplasm I periplasm
I
I
I
I
2X ....- -...1
9 I
4H z- -........- -.....: - -....
,
2CoA
1 4 Hz
2Fd---__...'
i
:
J
hydrogenase
2CH)-CO-CoA
2Pi~ 3 ! ~8W
t- 2COA
I cyt c)
2CH)-COO ~
I
~~4
I
G1
I
AMP I
I
I
I
!2CH)-COOH I I
I
I
I
I
I
I
ATP balance
'\------..~ 2 acetate + 2CO z

2 lactate - - -........
2ATP
SO~- ----....;\----~.SO~- -------~~Sz-

-~~~~;~;~~~is~----)
3ATP
Figure 8.29. Pathway of dissimilatory sulfate reduction in Desulfovibrio species
and "hydrogen cycling" hypothesis. 1, Lactate dehydrogenase, membrane-bound,
H-acceptor not known; 2, pyruvate-ferredoxin oxidoreductase; 3, phospho-
transacetylase; 4, acetate kinase; 5, ATP sulfurylase; 6, pyrophosphatase; 7,
APS reductase; 8, sulfite reductase; 9, cytoplasmic hydrogenase; 10, periplasmic
hydrogenase.
hydrogenase, picked up by the periplasmic hydrogenase, transferred to

cytochrome C3' and channelled via membrane-bound electron carriers to
APS-reductase and sulfite reductase. It is obvious that hydrogen cycling
allows the generation of a proton gradient at the cytoplasmic membrane.
The reduction of sulfate to sulfite via APS is connected to the hydrolysis
of two high-energy phosphoester bonds. Since most sulfate-reducing
bacteria are able to reduce sulfate with H 2 and since some group II species
even grow with H 2 + SO~- +C0 2 , it is apparent that electron transport to
SO~- must yield at least 3ATP.
B. Fermentation of acetate and propionate by

group II organisms
The conversion of ethanol to acetate and of L-malate to acetate by group I
organisms proceeds along the usual pathways. Propionate is converted by
Desulfobulbus propionicus to acetate via propionyl-CoA, methylmalonyl-
CoA, and succinyl-CoA. Formation of pyruvate from oxaloacetate is
achieved by the action of methylmalonyl-CoA: pyruvate transcarboxylase.
Of course, it is most interesting that organisms such as D. acetoxidans
can oxidize acetate completely to CO 2 , It is clear now that the tricarboxylic
acid cycle is employed for this purpose. The presence of all the enzymes
has been demonstrated. Succinate and malate dehydrogenase are mem-
brane-bound and transfer the reducing equivalents to menaquinone rather
than to FAD or NAD+.
C. Dissimilatory reduction of sulfur

Pfennig and collaborators isolated Desulfuromonas acetoxidans which
grows with acetate and elemental sulfur; acetate is oxidized to CO 2 via the
tricarboxylic acid cycle, and the reducing power generated is used to
reduce sulfur to sulfide:
CH 3 -COOH + 2H 2 0 ----. 2C0 2 + 8H
flC o, + 106.7 kJ (+25.5 kcal)
flC O , = -130.5 kJ (-31.2 kcal)
CH 3 -COOH + 2H 2 0 + 4So ----. 2C0 2 + 4H 2 S

flC o, = -23.8 kJ (-5.7 kcal)/reaction
Desulfuromonas is devoid of normal fermentative activities, but
elemental sulfur can be replaced by fumarate which is reduced to succinate
by a membrane-bound fumarate reductase.
Recently, the order Thermoproteales belonging to the archaebacteria
was described by Zillig, Stetter, and co-workers. Its members, such as
Thermoproteus tenax and Pyrodictium occultum, are extremely thermophi-
lic organisms. Their growth is most interesting, occurring at the expense of
The Anaerobic Food Chain 265
HzS formation from Hz plus elemental sulfur:

Hz + S ----+ HS- + H+ LiCo, = -28.0 kJ (-6.7 kcal)
P. oecu/tum is able to do so at a temperature of I05C! T. tenax is a

facultative chemolithotroph and grows also with organic substrates plus
sulfur; P. occu/tum, however relies on molecular hydrogen and elemental
sulfur as substrates.
IX. The Anaerobic Food Chain

The previous sections of this chapter have made us familiar with the
various fermentations that are carried out by microorganisms. These
fermentations lead to the formation of a number of products which, in
natural habitats, accumulate only temporarily. The relationships between
fermentative organisms are such that the products of one group of
organisms can serve as substrates for another group of organisms. In other
words, the various organisms are lined up with their catabolic activities and
form a so-called anaerobic food chain. If sulfate is present in sufficient
amounts the terminal process of this food chain is the sulfide fermentation;
at limiting sulfate concentrations (fresh water environments) it is the
methane fermentation.
A. Degradation of organic matter to

CH 41 CO 2 I and minerals
We have seen that the substrate spectrum of the methanogenic bacteria is
rather narrow. Only Hz + CO z, formate, methanol, methylamines, and
acetate can be utilized. Therefore, all the other fermentation products
have to be converted to these compounds in the anaerobic food chain. This
is not such a problem with lactate; it can be converted by the propionibac-
teria to propionate and acetate or by Clostridium tyrobutyricum to butyrate
(lactate + acetate - butyrate + CO z), but how are ethanol, propionate,
and butyrate converted to acetate? The thermodynamics of reactions in
which these compounds yield acetate and Hz (and CO z in the case of
propionate) is as follows:
ethanol + HzO ----+ acetate + H+ + 2H z
LiCo, = +9.6 kJ/reaction (+2.3 kcal/reaction)
butyrate + 2H zO ----+ 2 acetate + H + + 2H z
LiCo, = +48.1 kJ/reaction (+ 11.5 kcal/reaction)
propionate + 3H zO - acetate + bicarbonate + H+ + 3H z
LiCo, = +76.1 kJ/reaction (+18.2 kcal/reaction)
The free energy changes are positive, and the reactions will not proceed
from left to right except under conditions in which a product is kept at an
extremely low concentration. Due to the high affinity of the methanogenic

bacteria towards H 2 the partial pressure of H 2 is kept as low as 10- 4 atm in
the presence of these organisms. This is low enough to make the formation
of H 2 from NADH + H+ and also from the above-mentioned substrates
thermodynamically feasible (Fig. 8.30). Microbial associations in which a
Hrproducing organism can grow only in the presence of a Hrconsuming
organism are called syntrophic associations. The coupling of formation and
use of H 2 is called interspecies hydrogen transfer, a term coined by Wolin,
Bryant, and Wolfe.
A well-known syntrophic culture is "Methanobacillus omelianskii." It
consists of the S organism that oxidizes ethanol to acetate + H 2 and a
methanogenic bacterium (Fig. 8.31a). Other syntrophic partners of H 2 -
consuming organisms are Syntrophobacter wolinii and Syntrophomonas
wolfei which oxidize propionate (S. wolinii) or butyrate (S. wolfei)
according to the above equations, but only in the presence of H r
consuming organisms. They are obligate syntrophs.
Another type of relationship between Hrproducing and Hrconsuming
organisms is found with various clostridia, Ruminococcus albus, etc. on
one side and methanogens on the other (Fig. 8.31b). These organisms
grow on carbohydrates and ferment glucose to ethanol/acetate
+ CO2 + H 2 or butyrate/acetate + CO2 + H 2 . In the presence of a
H 2 utilizer the fermentation is shifted towards an acetate fermentation:
glucose + 4H2 0 - 2 acetate - + 2HC0 3 + 4H2 + 4H+
-16
NADH + W _NAD+ + H 2 toGo' = +18.8 kJ( +4.5kcal)
Ireaclion
-8
a
'(,
0
<l
+8
+16
10- 2 10- 4
PH 2 (atmospheres)
Figure 8.30. The free energy change for the oxidation of NADH to NAD+ and H2
as dependent on the partial pressure of hydrogen.
The Anaerobic Food Chain 267
CO 2
...... H 2O
...... H20
*
CH 4
S organism methanogemc bactenum
glucose CO2
~ADP + 2P i 2NAD+
2ATP 2NADH
+ 2H+
r-- H 20
2 pyruvate
lC0
2
--1r-- 2Fd
b
~ 2FdH 2 ~H20
2 acetyl-{:oA
=f
2 Pi
2ADP 2 CoA
2ATP
2 acetate CH4
Ruminucoccus a/bus methanogemc bactenum

Figure 8.31. Examples of interspecies hydrogen transfer. At low PH, ethanol is
oxidized by the S organism to acetate and H 2 (a) and glucose is oxidized by
saccharolytic clostridia or Ruminococcus a/bus to acetate + CO 2 + H 2 (b). The
PH, is kept low by the action of methanogenic bacteria.
Such shifts are possible if organisms are able to evolve H 2 and have
branched fermentation pathways at their disposal (see Fig. 8.12).
The special fermentations under low partial pressures of H 2 make it
understandable that fermentation products others than CH 4 + CO 2 do not
accumulate in the anaerobic food chain. The importance of a low PH, was.
first recognized by Hungate. A scheme of the anaerobic food chain is given
in Fig. 8.32.
B. Food chains in marine environments and

in the rumen
It has already been mentioned that sulfate reducers predominate in marine
environments as compared to methanogenic bacteria. Here the food chain
is largely directed towards formation of CO 2 and H2 S. The reason for this
is that the substrate spectrum of the sulfate-reducing bacteria is much
wider, that they can utilize H 2 very effectively, and that their affinity
polymers
(proteins, polysaccharides,
lipids, nucleic acids, etc.)
j
monomers and oligomers
(pep tides, amino acids, sugars, acids,
glycerol, purines, and pyrimidines)
alcohols, propionate,
butyrate, lactate,
other products
I cO 2 + H 2
A
1 ~ I acetate
I
I formate
I methanol
methylamines
I
Figure 8.32. The anaerobic food chain.
towards acetate is much higher than the one of methanogens. It is apparent

from Fig. 8.32 that acetate is the most important substrate for the terminal
food chain reactions. In the presence of sulfate it is not split to CH 4 + CO 2
but oxidized to CO 2 .
A special situation is met in the rumen. The ruminant is, of course, not
interested in a complete anaerobic food chain. Thus, the partial pressure
of H 2 is kept low in the rumen by methanogenic bacteria, thereby direct-
ing the fermentations towards the formation of organic acids (mainly
acetic and propionic acids). These compounds are then taken up by the
ruminant.
Fermentation of Nitrogenous Compounds 269
x. Fermentation of Nitrogenous
Compounds
Sugars and organic acids are not the only substrates for anaerobes. Amino
acids (formed from proteins by extracellular proteases) and purine and
pyrimidine bases are fermented by a variety of microorganisms.
A. Single amino acids

A number of single amino acids can serve as energy and carbon source for
anaerobes. Alanine is fermented by Clostridium propionicum via the
acrylate pathway (see Fig. 8.18). Peptostreptococcus micros (Diplococcus
glycinophilus) ferments glycine according to the following equation:
Originally it was found that the CO 2 produced in this fermentation was

derived from the carboxyl group of glycine and that both carbons of acetate
originated partly from the methylene carbon of glycine and partly from
CO 2 , This was in agreement with a pathway in which glycine was hydroxy-
methylated to serine and then converted to acetate via pyruvate.
Recently, Andreesen and co-workers showed that the pathway indicated
was only used by the organism in selenium-deficient media and for bio-
synthetic purposes. In the presence of selenite, the organisms grow much
better and the pathway used is different; it is depicted in Fig. 8.33. Part of
the glycine is oxidized via methylenetetrahydrofolate to CO 2 , The reduc-
ing equivalents generated are used to reduce 3 molecules of glycine to
acetate + ammonia. The first reaction, the oxidation of glycine to CO 2 and
methylenetetrahydrofolate, is very complex. Four proteins are involved: a
decarboxylase, a hydrogen carrier protein, a lipoamide dehydrogenase,
and a transferase. All subsequent steps are similar to the ones in acetate
fermentation, but proceed in the reverse reaction (see Fig. 8.23). It should
be noted that the formyltetrahydrofolate synthetase reaction yields ATP
by substrate-level phosphorylation.
The most interesting reaction of this pathway is the glycine reductase
reaction. As shown by Stadtman and co-workers (working with Clostrid-
ium sticklandii), the protein complex responsible for this reaction consists
of three proteins; the so-called protein A contains selenocysteine. Thus it is
clear that the operation of the pathway depicted in Fig. 8.33 depends on
the presence of a selenium compound in the growth medium. The glycine
reductase reaction is apparently coupled to ATP synthesis which proceeds
most likely by substrate-level phosphorylation and not by a chemiosmotic
mechanism. A scheme indicating the involvement of the proteins in the
reaction is depicted in Fig. 8.34.
The fermentation of threonine (Fig. 8.35) is initiated by threonine
lelrahydrofolale + NH 2- CH 2- COOH
~
~ r----.~NADH
----,1
NAD+ ....
+ W ----I
r---
5, IO-methylenelerrahydrofolale
k--- NAD+
NADH + H+ - - - - ' 1 6
5, IO-melhenyllelrahydrofolale
If ADP+P i
1Q-formyllelrahydrofolale
~ATP
~ADP+Pi
i
ATP
formate
I CO 2 I 1 3CH )COOH + 3NH) I
Figure 8.33. Fermentation of glycine by Peptostreptococcus micros. 1, Glycine

decarboxylase complex; 2, methylenetetrahydrofolate dehydrogenase; 3, methenyl-
tetrahydrofolate cyclohydrolase; 4, formyltetrahydrofolate synthetase; 5, formate
dehydrogenase; 6, glycine reductase.
dehydratase (c. propionicum, Peptostreptococcus prevotii). The a-
oxobutyrate formed is further converted to propionate in a reaction
sequence involving an enzyme similar to pyruvate-ferredoxin oxidoreduc-
tase.
Aspartate is fermented by many facultative and some obligate anaerobic
bacteria; it is deaminated to fumarate, which is partly reduced to succinate
and partly oxidized to acetate. The pathways involved are similar to those
of fumarate and malate fermentation. Some clostridial species, e.g., C.
novyi, contain a decarboxylase that converts aspartate into alanine or
l3-alanine. Both are substrates for C. propionicum:
HOOC-CH 2 -CH(NH2 )-COOH ~
CO 2 + CH3 -CH(NH2 )-COOH
HOOC-CH 2 -CH(NH2 )-COOH ~
CO 2 + NH 2 -CH2 -CH2 -COOH
I CH)COOH I
Figure 8.34. Tentative scheme of the glycine reductase mechanism. Only the
possible function of two of the three proteins required is indicated. -CHO at P B
indicates a prosthetic group. [Redrawn from G. F. Barnard and M. Akhtar. Eur. J.
Biochem. 99, 593-603 (1979)].
COOH
I
H,N-C-H
- I
H-C-OH
I
CH)
L-threonine
threonine dehydratase
(P.l'revorii)
COOII
1
NH) C=O propionate + H 2 + C02
CH,
I -
CH j
Figure 8.35. Initial reactions in the fermentation of threonine.
The fermentation of glutamate by obligate anaerobic bacteria has

received considerable attention. This amino acid seems to be the preferred
substrate of Clostridium tetanomorphum, which employs a rather unusual
pathway for its breakdown. The elucidation of this pathway by Barker and
collaborators led to the discovery of the first B 12-dependent enzyme,
glutamate mutase. The rearrangement reaction catalyzed by this enzyme
has already been discussed in connection with other B 12-dependent
enzymes (Chapter 8, Section V). As shown in Fig. 8.36a, the product of
the mutase reaction, J3-methylaspartate, is deaminated to yield mesacon-
ate. Addition of water leads to citramalate, which subsequently is cleaved
to acetate and pyruvate. This reaction resembles the citrate lyase reaction,
and the citramalate and citrate lyases are closely related to one another.
a COOH COOH
I I
HlN-<[H I H 2 N-CH
~ I HOOC-CH
II
TH2
Hl HC-COOH
I C-COOH
T
COOH
CH 3 I
CH3
l-glutamate l-chreo-{3-methyl- mesaconale

aspartate
CH 3 -CO-COOH
>>---4_ COOH
I
CH 2
I
HO-C-COOH
I
pyruvate CH 3
! citramalate
0.2 acetate, 0.4 butyrate, ICO l , O.2H l
HlO NH)
+ +
b COOH NAD+ NADH T OOH NADH + H+ NAD+ TOOH
I
HlN-CH
I
\ c -:
~
H+
T=O \.64 HO-T-H
TH2 THl '

Hl
T
TH2
COOH
T Hl
COOH
T
Hl
eaOH
l-g1utamate O<-Qxoglutarate o<-hydroxyglularate
acetyl-eoA~ATP)
7 HlO
acetate
CO-CoA
I
CH
II
CH
I
THl
COOH
+_co,
glutaconyl-CoA
CO-CoA
I
)>-_9_ CH
II
CH
I
CH)
crotonyl-Co A
Figure 8.36. Pathways of glutamate fermentation by Clostridium tetanomorphum

(a) and by Peptostreptococcus asaccharolyticus (b). 1, Glutamate mutase; 2, f3-
methylaspartase; 3, citramalate dehydratase; 4, citramalate lyase; 5, glutamate
dehydrogenase; 6, a-hydroxyglutarate dehydrogenase; 7, a dehydratase and a CoA
transferase are involved; 8, glutaconyl-CoA decarboxylase (Na+-dependent); 9,
dismutation of crotonyl-CoA.
Pyruvate is then further degraded to acetyl-CoA, CO 2 , and reduced

ferredoxin. Little hydrogen is evolved from the latter. Acetyl-CoA is only
partly converted to acetate; the rest yields butyrate.
Acidaminococcus fermentans and Peptostreptococcus asaccharolyticus
ferment glutamate to the same products as C. tetanomorphum but use
another pathway. As is apparent from Fig. 8.36b, glutamate is first
converted to a-hydroxyglutarate. Then glutaconyl-CoA is formed which is
decarboxylated to crotonyl-CoA-a reaction that is biotin-dependent and
coupled with Na + export (see Fig. 8.17). Finally crotonyl-CoA dismutates
to yield acetate and butyrate: 1 crotonyl-CoA is oxidized to 2 molecules of
acetate (via acetoacetyl-CoA) and 1 crotonyl-CoA is reduced to butyrate
(via butyryl-CoA). The mechanism of a-hydroxyglutarate dehydration is
not fully understood. The hydroxyl group may be removed from the
molecule in a radical reaction.
It should also be mentioned that a glutamate decarboxylase yielding
4-aminobutyrate from glutamate is present in clostridia (e.g., C. perfrin-
gens), in enterobacteria (e.g., E. coli), and others. 4-Aminobutyrate is a
substrate for certain clostridia.
Lysine is fermented by Clostridium sticklandii and by C. subterminale to
acetate and butyrate. The interesting pathway employed is summarized in
Fig. 8.37; it involves two shifts of amino groups. One of them, L-,B-lysine
mutase, requires coenzyme Bu. The C6 carbon skeleton is cleaved in a
very interesting reaction; the terminal-CH 2 -COOH group of 3-oxo-5-
aminohexanoate is transferred to acetyl-CoA so that acetoacetate and
3-aminobutyryl-CoA are formed as products. The final products are
butyrate, acetate, and ammonia.
Arginine is fermented to ornithine, CO 2 , and NH 3 by clostridia,
streptococci, halobacteria, eubacteria, and mycoplasmas. This fermenta-
tion is unusual in that ATP is formed from carbamoyl phosphate.
Ornithine is degraded further by C. sticklandii. As in the fermentation of
lysine, ornithine breakdown also involves a shift of an amino group. First,
L-ornithine is converted to the D-enantiomer by a racemase. Then the
amino group is shifted from Cs to C4 in a coenzyme B t2 and pyridoxal
phosphate depending reaction. 2-Amino-4-oxopentanoate is subsequently
formed which is cleaved with CoA to yield acetyl-CoA and L-alanine. This
is another variation of the arginine degradation pathways that were
outlined in Chapter 6 (see Fig. 6.4).
B. Stickland reaction
Although a number of clostridial species grow with some single amino
acids, many clostridia prefer to ferment mixtures of amino acids. They
carry out coupled oxidation-reductions between pairs of amino acids. One
amino acid, e.g., alanine, is oxidized, and a second one, e.g., glycine, is
~H2
~H2CH2CH2CH2CH-COOH ::;"i=====::::::j.~ CH-CH-CH-CH-CH-COOH
I 2 2 2 I 2
NH 2 NH 2 NH 2
L-lysine 3, 6-diaminohexanoate
(13-lysine)
CHJCH2CH~OOH I
butyrate
CHTCHrCHrCoA
butyryl-CoA
CHJiH-CHrTH-CHz-COOH
NH 2 NH 2
3,5-diaminohexanoate
CHJCH=CH-CO-CoA
crotonyl-CoA
CoA
CH')H-CH~O-COA
NADH+W
NH 2
+INH 3 j
3-aminobutyryl-CoA
4 CHTH-CH 2n -CH 2 COOH
NH 2 0
3-oxo-5-aminohexanoate
acetoacetate
CH'jCO-CHz-CO-CoA CH')CO-CoA
acetoacetyl-CoA acetyl-CoA
CH')CO-CoA
acetyl-CoA
AD:::'iATP
I CHJCOOH I
acetate
Figure 8.37. The clostridial pathway of L-Iysine fermentation. 1, L-Lysine-2,3-
aminomutase, pyridoxal-P and Fe 2 + -dependent; 2, /3-lysine mutase, B12-
dependent; 3, 3,5-diaminohexanoate dehydrogenase; 4, 3-oxo-5-aminohexanoate
cleavage enzyme (acetyl-CoA requiring); 5, L-3-aminobutyryl-CoA deaminase;
6, butyryl-CoA dehydrogenase; 7, CoA transferase; 8, /3-ketothiolase; 9,
phosphotransacetylase + acetate kinase.
reduced:
CH 3 -CH(NH2 )-COOH + 2H 2 0 ----+
CH3 - COOH + CO 2 + NH 3 + 4H
2NH 2 -CH2 -COOH + 4H ----+ 2CH3 -COOH + 2NH 3
This type of fermentation was discovered by Stickland in 1934; it is
carried out by practically all proteolytic clostridia, such as C. sporogenes,
C. sticklandii, C. histolyticum, and C. botulinum. Some amino acids are
preferably used as H-donors and others as H-acceptors. The most suitable
donors and acceptors are given in Table 8.12. Depending on the microor-
ganism, the aromatic amino acids and leucine may function either as
H-donor or as H-acceptor.
Amino acid oxidation proceeds via the corresponding a-oxoacid:
+H20 +H 20
R-CH-COOH ---.::;;-+ NH3 + R-CO-COOH ~ R-COOH + CO 2
I 2H 2H
NH 2
The first step, the oxidative deamination, is accomplished either by an

enzyme of the type of glutamate dehydrogenase or by transamination with
a-oxoglutarate as NHracceptor and subsequent regeneration of a-
oxoglutarate by glutamate dehydrogenase. The oxidative decarboxylation
is catalyzed by enzymes analogous to pyruvate-ferredoxin oxidoreductase.
ATP is formed from the CoA-esters by the action of CoA-transferase,
phosphotransacetylase, and acetate kinase.
Amino acid reduction is a rather complex reaction. The glycine reduc-
tase system has already been discussed in connection with the metabolism
of P. micros (see Fig. 8.34). Reduction of betaine or sarcosine proceeds by
analogous Se-requiring reductases.
The use of L-proline as oxidant is initiated by a racemase, and the
o-enantiomer actually becomes reduced. In contrast to the glycine reduc-
tase, the o-proline reductase does not contain selenium, but it contains a
pyruvate moiety linked to the protein by a peptide bond. This moiety is
essential for catalytic activity. Purified o-proline reductase does not react
with NADH as H-donor; a dithiol compound can be used as artificial
H-donor. A flavoprotein and an iron-sulfur protein are involved in
H-transfer to the enzyme in vivo.
CH 2-CH 2 /SH
I I +R--->
CH 2 CH-COOH "SH
"N/
H
o-proline 5-aminovalerate
Organisms such as C. sporogenes or C. botulinum contain an enzyme-

ornithine cyclase-that converts ornithine via A' -pyrroline-2-carboxylate
IV
-.J
0-
Table 8.12. Amino acids that function as H-donors and as H-acceptors in Stickland reactions.
H-donor H-acceptor
substrate product substrate product
alanine acetate + CO 2 + NH 3 glycine acetate + NH 3

leucine 3-methylbutyrate + CO 2 + NH 3 proline 5-aminovalerate
phenylalanine phenylpropionate + NH 3
isoleucine 2-methylbutyrate + CO 2 + NH 3 tryptophan indolpropionate + NH 3
valine 2-methylpropionate + CO 2 + NH 3 ornithine 5-aminovalerate + NH 3 (via proline)
phenylalanine phenylacetate + CO 2 + NH 3 leucine 4-methylvalerate + NH 3
tryptophan indolacetate + CO 2 + NH 3 betaine acetate + trimethylamine ??
to
sarcosine acetate + monomethylamine 0>
histidine glutamate + CO 2 + NH 3 n
...(;
[
ri1
3
::l
'"
~
o
::l
'"
to proline so that ornithine can also function as oxidant in Stickland

reactions.
It is interesting that the aromatic amino acids and leucine can also
function as oxidants. They apparently are converted directly or via the
a-hydroxy acids to the a,j3-unsaturated acids, which are then reduced by
an enzyme present in many clostridia, enoate reductase. It is not very
specific and acts on many unsaturated acids.
COOH COOH
I I
CH CH 2
II enoate
reductase
I
L-Ieucine - - CH --7~~""'\I""---+) CH 2
I NADH + H+ NAD+ I
CH-CH3 CH-CH3
I I
CH3 CH 3
Since L-Ieucine can be oxidized and reduced it is not surprising that
several proteolytic clostridia can carry out a Stickland reaction with leucine
alone-part of it is oxidized to 3-methylbutyrate and another part is
reduced to 4-methylvalerate.
C. Heterocyclic compounds
Purines and pyrimidines are readily fermented under anaerobic conditions.
The long known species Clostridium acidiurici and C. cylindrosporum
ferment guanine, hypoxanthine, urate, and xanthine; they are so special-
ized that they will not grow with any other substrate except the inter-
mediate glycine.
Recently, C. purinolyticum was described; it grows on a much greater
variety of purines including adenine, and it also grows on glycine. Growth
is strictly selenium-oependent. Four products are formed by these species
from purines: carbon dioxide, ammonia, acetate, and formate. Because of
the high N content of the substrates and the two single carbons in the
purine skeleton, NH 3 and CO 2 are the major fermentation products. The
fermentation of purines is generally initiated by their conversion into
xanthine (Fig. 8.38). In the case of guanine this is accomplished by guanine
deaminase; in the case of adenine by adenine deaminase plus xanthine
dehydrogenase. The latter is an interesting enzyme. It is a selenoflavopro-
tein containing iron-sulfur centers and molybdenum cofactor, and it
catalyzes a variety of oxidation/reduction reactions at the purine ring and
other N-heterocyclic compounds: the oxidation of hypoxanthine to xan-
thine, the reduction of urate to xanthine, the oxidation of purine to
hypoxanthine, and the oxidation of, for instance, pterins and 4-
hydroxypyrimidine.
",ot
adenine hypoxanthine
2H
"'jc>
H2 O 2H
\ 2 .- .. 3 )
t
NH 3
O~NH N
H
(
H2O
guanine xanthine urate
HOOC
::1 "N
H
x> N
N
H
4-ureido-5-imi- 4-amino-S-imi
dazole carboxylate dazole carboxylate
~C02
L)
NH H 20 N
H C/ '----CH
2
1 I
8 )
..
HOOC NH H 2N N
NH 2 H
formiminoglycine 4-imidazolone 4-aminoimidazole

Figure 8.38. Conversion of adenine, guanine, and urate to formiminoglycine by
purinolytic clostridia. 1, Adenine deaminase; 2, guanine deaminase; 3, xanthine
dehydrogenase; 4, xanthine amidohydrolase; 5, 4-ureido-5-imidazole carboxylate
amidohydrolase; 6, 4-amino-5-imidazole carboxylate decarboxylase; 7, 4-amino-
imidazole deaminase; 8, 4-imidazolonase.
In the purine fermentation pathway, xanthine is then converted to

formiminoglycine by a series of deaminations and decarboxylations. The
fate of formiminoglycine is depicted in Fig. 8.39. The formimino group is
converted to formate and further to CO 2 , The reducing equivalents
produced in the formate dehydrogenase reaction are used to drive the
HN=CH - NH- CH 2 -COOH
fonnirninoglycine
tetrahydrofolate
5-formiminotetrahydrofolate
glycine
NH,~2
5, IO-methenyltetrahydrofolate
H,ai,
lO-fonnyltetrahydrofolate
k-
tetrahYdrofolatef ATP
ADP + P
fonnate
1.......- - - - - - X ------,,1.---
6
ADP + Pi
1 ' - - - - - - XH 2 ------"1I'. ATP
~ acetate
Figure 8.39. Conversion of formiminoglycine to acetate, CO 2 , and ammonia.

1, Glycine formimino transferase; 2, formiminotetrahydrofolate cyclodeaminase;
3, methenyltetrahydrofolate cyclohydrolase; 4, formyltetrahydrofolate synthetase;
5, formate dehydrogenase; 6, glycine reductase.
glycine reductase reaction. If reducing equivalents are generated by
oxidation of hypoxanthine to xanthine, formate does not need to be
oxidized to CO 2 and is excreted. If urate serves as substrate, reducing
equivalents have to be generated by an additional pathway, e.g., by
oxidation of glycine to CO 2 as outlined in Fig. 8.33.
A number of bacteria are known that ferment pyrimidines. Uracil is
degraded by C. glycolicum to l3-alanine, CO2 , and NH 3 and orotic acid
by C. oroticum to acetate, CO 2 , and NH 3 . In general pyrimidines are not
fermented as readily as purines. The degradation of nicotinic acid by
Eubacterium barkeri is initiated by a dehydrogenase that resembles
xanthine dehydrogenase. Creatinine is converted by some clostridial
species to N-methylhydantoin; the further degradative pathway of this

compound remains to be elucidated.
It can be concluded that the fate of a large number of nitrogenous
compounds under anaerobic conditions is now known. Details have still to
be worked out.
XI. Summary
1. Fermentations are anaerobic dark processes. ATP is formed by
substrate-level and/or electron transport phosphorylation.
2. Most strictly anaerobic bacteria lack catalase and superoxide dis-
mutase; strictly anaerobic bacteria require a low redox potential for
growth.
3. Yeasts ferment glucose to ethanol and CO 2 , The key enzyme of this
fermentation is pyruvate decarboxylase. Two mol of ATP are formed per
mol of glucose fermented; yeasts increase the rate of glucose breakdown
when transferred from aerobic to anaerobic conditions (Pasteur effect).
Zymomonas species, Sarcina ventriculi, and Erwinia amylovora carry
out alcohol fermentations. Ethanol produced in smaller amounts by lactic
acid bacteria, enterobacteria, and clostridia is formed by reduction of
acetyl-CoA.
4. Lactic acid bacteria employ the homofermentative, the heterofer-
mentative, or the bifidum pathway for the fermentation of hexoses. The
first pathway yields 2 lactate/glucose, the heterofermentative pathway
yields lactate, ethanol, and CO 2 , whereas acetate and lactate are formed in
a ratio of 3: 2 by the bifidum pathway. The key enzyme of the latter two
pathways is phosphoketolase.
Lactic acid is produced in several forms-o( -), L( +), and oL-and
lactic acid bacteria contain o-lactate dehydrogenase, L-Iactate dehy-
drogenase, or a mixture of these enzymes. A few producers of the oL-form
contain L-lactate dehydrogenase plus racemase.
Lactobacillus plantarum and some other species ferment malate to
lactate and CO 2 (malo-lactate fermentation). Streptococcus cremoris and
Leuconostoc cremoris produce diacetyl from citrate. The latter is cleaved
into acetate and oxaloacetate by citrate lyase and the final step in diacetyl
synthesis is the condensation of acetyl-CoA with hydroxyethylthiamine
pyrophosphate.
From growth yield studies with lactic acid bacteria it was deduced that
an average of 10.5 g of cells can be formed per 1 mol ATP produced, pro-
vided that all monomers required in biosynthesis are available to the cells.
5. The main fermentation product of many clostridia, eubacteria,
fusobacteria, and butyrivibrios is butyrate. It is formed from sugars via
pyruvate, acetyl-CoA, acetoacetyl-CoA, and butyryl-CoA. The conver-
sion of pyruvate to acetyl-CoA is catalyzed by pyruvate-ferredoxin ox-
Summary 281
idoreductase. In many fermentations Hz is produced from reduced ferre-

doxin with hydrogenase. Hz production from NADH allows the organisms
to produce less butyrate, more acetate, and more ATP (advantage of
branched pathways).
Some clostridia (e.g., C. acetobutylicum) form acetone and butanol at
pH values below 4.5.

The hydrogen balance of fermentations can be determined on the basis
of the /R values of substrates and products or on the basis of the number
of available hydrogens in substrates and products.
C. kluyveri ferments ethanol and acetate to butyrate, caproate, and
molecular hydrogen. Hz evolution is closely connected with ATP synthesis
by this microorganism. Per mol Hz evolved 0.5 mol of acetyl-CoA
becomes available for ATP synthesis. Pyruvate is synthesized in C.
kluyveri by reductive carboxylation of acetyl-CoA, and oxaloacetate by the
pyruvate carboxylase reaction. Consequently, about 30% of the cellular
material of C. kluyveri is derived from CO z . C. kluyveri contains re-citrate
synthase.
6. Microorganisms belonging to the genera Escherichia, Salmonella,
and Shigella carry out a mixed acid fermentation and produce lactate,
acetate, succinate, formate, CO z, and Hz. Characteristic enzymes of this
fermentation are pyruvate-formate lyase, which cleaves pyruvate into
acetyl-CoA and formate, and formate-hydrogen lyase, which splits formate
into Hz + CO z . Pyruvate-formate lyase is rapidly inactivated by oxygen.
7. Microorganisms belonging to the genera Enterobacter, Serratia, and
Erwinia produce less acids than the above-mentioned enterobacteria but
more CO z , ethanol, and 2,3-butanediol. The first enzyme in 2,3-butanediol
formation is a-acetolactate synthase. An oxaloacetate decarboxylase has
been found in Enterobacter that couples decarboxylation with generation
of a Na + gradient across the membrane.
8. Clostridium propionicum and Megasphaera elsdenii employ the acry-
late pathway for the formation of propionate from lactate. Lactyl-CoA and
acrylyl-CoA are intermediates of this pathway. Electron-transferring
flavoprotein functions as H-carrier.
The propionibacteria and other propionate-forming microorganisms
employ the succinate-propionate pathway in which succinyl-CoA and
methylmalonyl-CoA function as intermediates. The interconversion of
these two thioesters is catalyzed by methylmalonyl-CoA mutase, a coen-
zyme B 12 -containing enzyme.
The reduction of fumarate to succinate by fumarate reductase is a
process by which strict anaerobes gain ATP by electron transport phos-
phorylation. Fumarate reductase is membrane-bound and associated with
menaquinone and in many organisms with a cytochrome of the b type.
9. C. formicoaceticum and C. thermoaceticum ferment 1 mol of hexose
to almost 3 mol of acetate. Acetate is formed by the Embden-Meyerhof-
Parnas pathway and by reduction of CO z to acetate. Clostridium aceticum
and Acetobacterium woodii ferment Hz + CO z to acetate. CO is a

precursor of the carboxyl group of acetate.
10. Methanogenic bacteria ferment CO 2 + Hz, formate, methanol,
methylamines, and acetate. The reduction of CO z to CH 4 proceeds with
methanofuran, tetrahydromethanopterin, and coenzyme M as carriers.
ATP is produced in the last reduction step by a chemiosmotic mechanism.
A nickel-tetrapyrrol is a constituent of the terminal reductase. A B\2-
containing enzyme is involved in transfer of the methyl groups of methanol
and methylamines to coenzyme M. Those methanogens that utilize metha-
nol, methylamines, and acetate (e.g., Methanosarcina barkeri) contain
cytochromes. Coenzyme F4Z0 is an important hydrogen carrier in meth-
anogenic bacteria.
11. In sulfide fermentation, the oxidation of organic compounds is
coupled to the reduction of sulfate to sulfide. The substrate for reduction is
adenosine-Sf -phosphosulfate (APS), and reduction proceeds via sulfite to
sulfide. Electron carriers such as ferredoxin, cytochrome C3, rubredoxin,
and menaquinone are involved in sulfate reduction and in A TP synthesis
by electron transport phosphorylation. Group I sulfidogenic bacteria
oxidize substrates to acetate, group II organisms all the way to CO z.
12. The partial pressure of hydrogen in mud, in anaerobic digesters,
and in the rumen is kept low by the action of methanogenic bacteria. This
favors organisms that produce hydrogen (interspecies hydrogen transfer)
and acetate and is a prerequisite for the functioning of the anaerobic food
chain. In marine environments, sulfidogenic organisms predominate, and
HzS + CO z are produced from acetate, formate, and Hz.
13. Single amino acids are fermented by a number of anaerobic
bacteria: alanine by Clostridium propionicum, glycine by Peptostreptococ-
cus micros, glutamate by C. tetanomorphum, and lysine by C. sticklandii.
The fermentation of glutamate and lysine involves coenzyme B 1Z -
containing enzymes.
Pairs of amino acids are fermented by a number of proteolytic clostridia
(Stickland reaction). The oxidation of one amino acid (e.g., alanine) is
coupled to the reduction of another amino acid (e.g., glycine).
Microorganisms such as C. acidiurici and C. purinolyticum are special-
ized for the fermentation of purine bases. Urate is fermented by C.
acidiurici to acetate, CO z, and ammonia.
Chapter 9
Chemolithotrophic and
Phototrophic Metabolism
Chemolithotrophic and phototrophic bacteria have in common the ability'

to grow in mineral media, deriving their cell carbon from CO 2 , The
reducing power required for CO 2 reduction is obtained from inorganic
compounds and energy is provided either by light-dependent reactions or
by oxidation of inorganic compounds with oxygen or nitrate (aerobic
chemolithotrophs). The anaerobic chemolithotrophs (methanogenic, ace-
togenic and sulfidogenic bacteria) have been discussed in Chapter 8.
I. Chemolithotrophic Metabolism
A. Physiological groups of aerobic

chemol ithotrophs
As already mentioned chemolithotrophs gain energy by oxidation of an
inorganic compound. Depending on the nature of the inorganic compound
oxidized, six groups of aerobic chemolithotrophs are recognized; they are
summarized in Table 9.1. The reactions carried out and their free energy
changes are also given.
Hydrogen-oxidizing bacteria have in common that they use molecular
hydrogen as energy source. In other physiological properties and morpho-
logically, however, they are very diverse. Since they all are facultative
chemolithotrophs, taxonomists have preferred to classify them with their
chemoheterotrophic relatives. The hydrogen-oxidizing bacteria include
Pseudomonas saccharophila, P. fadlis, Alcaligenes eutrophus, Nocardia
N
'{;
Table 9.1. Physiological groups of chemolithotrophs
aGO'
group ATP-yielding reaction kJ / reaction kJ/2e-

'0
hydrogen bacteria H 2 + toz ~ H 20 -237.2 (-56.7) -237.2 (-56.7)
()
carboxydobacteria CO + +0 2 ~ C0 7 - 257.1 (-61.5) -257.1 (-61.5) :r
sulfur bacteria S2- + 20 2 ..... SO;- -794.5 (-189.9) -198.6 (-47.5) "3
SO + 1102 + H20 ~ SO~- + 2H+ - 584. 9 (-139.8) -195.0 (-46.6) g,
;.
iron bacteria Fe 2t + toz + H + ~ Fe Jt + tHzO -44.4 (- 10.6)" -88.8 (- 21.2)"
ammonia oxidizers -270.7 (-64.7) -90.2 (-21.6) <3
NH; + ItOz ~ NO z + 2W + H 2 0 "0
:r
;S.
nitrite oxidizers NO + t02 ~ NO)
z -77.4 (-18.5) -77.4 (-18.5)
0>
:l
Q.
U ilC'" values for pH 0 are given; iron bacteria can grow at acidic pH values only; the ilC'" value for pH 7 would be
'"0
-4 kJ / reaction. Kcal values in parentheses. :r
o

<3
"0
:r
;SO
:s:
":r
g,
'"
3
Chemolithotrophic Metabolism 285
autotrophica, the nitrogen fixing Xanthobacter autotrophicus and Paracoc-

cus denitrificans. The latter is able to use nitrate instead of oxygen as
electron acceptor.
Carbon monoxide-oxidizing bacteria are also found in various genera.
Examples are: Pseudomonas carboxydovorans, Alcaligenes carboxydus,
and the thermophile Bacillus schlegelii. All CO-oxidizing bacteria are
likewise Hz oxidizers (not vice versa!); they are faculative chemolitho-
trophs.
Sulfur oxidizers are the thiobacilli, Thiomicrospira pelophila, a marine,
spiral organism, and Sulfolobus, a thermophilic archae bacterium of
irregular cell form. In addition, filamentous gliding organisms, such as
Beggiatoa and Thiothrix are able to oxidize sulfide to elemental sulfur and
subsequently to sulfate. Most sulfur bacteria are obligate chemolitho-
trophs. Some of them, e.g., Thiobacillus intermedius, can grow as aerobic
heterotrophs. T. denitrificans can utilize nitrate instead of oxygen as
electron acceptor.
The Fe2+ oxidizer Thiobacillus ferrooxidans is able to use reduced sulfur
compounds and ferrous ions alternatively as electron donors. The iron
bacterium. Gallionella probably also gains energy by oxidation of Fe z+ to
Fe 3 +. The free energy change of Fe z+ oxidation at low pH values is large
enough to be coupled to ATP synthesis; it is rather small at neutral pH and
iron bacteria cannot grow at pH values above about 4.
The oxidation of ammonia to nitrite is carried out by Nitrosomonas,
Nitrosospira, Nitrosovibrio, and Nitrosococcus species; they all are obli-
gate chemolithotrophs and so are the nitrite oxidizers Nitrobacter, Nitro-
spina, and Nitrococcus, with the exception of some Nitrobacter strains.
Because nitrite is very toxic to most organisms, the processes of nitrite
production and nitrite oxidation are remarkable.
B. Energy production and generation of reducing

power in H 2 and CO oxidizers
The ratio in which Hz, 0z, and CO z are consumed by a growing culture of
hydrogen-oxidizing bacteria is about the following:
4H z + 20 z ~ 4H zO
2H z + CO z ~ (CHzO) + HzO
6H z + 20 z + CO z ~ (CHzO) + 5HzO
Thus, the oxidation of 4H z to water yields enough ATP to allow the
synthesis to cell material (CHzO) from CO z and Hz.
ATP synthesis proceeds by a chemiosmotic mechanism as in aerobic
respiration. Cytochromes, ubiquinone, and menaquinone have been found
in membrane fractions of hydrogen-oxidizing bacteria. Differences be-
tween hydrogen-oxidizing species have been encountered as to the transfer
of electrons from Hz to the respiratory chain. Nocardia opaca contains a
286 9: Chemolithotrophic and Phototrophic Metabolism
soluble hydrogenase -.vhich catalyzes the reduction of NAD+ by Hz. The

product, NADH, then serves as H-donor for the respiratory chain. This is
the exception. All other hydrogen-oxidizing bacteria studied contain a
particulate (membrane-bound) hydrogenase which feeds electrons directly
into the respiratory chain. This enzyme does not react with NAD+. Very
few hydrogen-oxidizing bacteria (e.g., Alcaligenes eutrophus,) contain, in
addition to the particulate enzyme a soluble hydrogenase which reduces
NAD+ and which is primarily responsible for the provision of NADH for
COz reduction. The function of the two hydrogenases in these organisms is
summarized in Fig. 9.1.
The uptake-hydrogenases discussed here catalyze the oxidation of Hz
to H+ according to the equation
Hz - - 2H+ + 2e-
and the transfer of the electrons to the appropriate acceptors. The enzymes
of the Hz oxidizers have been shown to be nickel-proteins. In addition,
they contain iron-sulfur centers. The K m values for Hz are in the order of
30-40 J-tM, so these enzymes are very effective in oxidizing Hz.
The composition of the membrane-bound and of the soluble hy-
drogenase of A. eutrophus and of hydrogenases from some other micro-
organisms is depicted in Table 9.2. It is obvious that the evolution-
hydrogenases belong to the group of enzymes that have a simpler
composition. They consist of just one subunit; nickel is apparently not
essential. The most complex ones are the uptake-hydrogenases that reduce
NAD+ or F4zo . They consist of several subunits and contain a flavin. The
function of the latter may be the one of a redox-switch from I-electron
carriers to 2-electron acceptors: Iron-sulfur-centers reduce the flavin that
subsequently transfers the electrons to NAD+ or F4zo .
Recently, it has been shown by Friedrich and Schlegel that the genetic
information for hydrogenase synthesis in A. eutrophus resides on a large
plasmid. In other Hz oxidizers this information is chromosome-encoded,
and the advantage of either location is not clear yet.
c, a/a~ _
J~02
-
ATP,
-
.................................
__ :::) biosynthesis
/ ....
NADH
Figure 9.1. The function of the two hydrogenases of A. eutrophus. [H. G. Schlegel.
Antonie van Leeuwenhoek 42, 181-201 (1976).]
()
:r
Table 9.2. Composition of hydrogenases of various bacterial species
...
3
g,
S-
subunits flavin nickel iron-sulfur o
Organism function e- -acceptor (molecular weight) content content centers 3
'0
:r
;:;.
a
A. eutrophus uptake quinone or 67,000 0.7 3Fe - 3S
iron-sulfur 31,000 4Fe - 4S ...~
cr"
A. eutrophus uptake NAD+ 63,000 1FMN 2 2Fe - 2S .
u;.
'"
56,000 3Fe - 3S a 3
30,000 4Fe - 4S (2)
26,000
Methanobacterium uptake F 420 2 x 40,000 2FAD 2.8 ND
thermoautotrophicum 2 x 31,000
26,000
Desulfovibrio gigas uptake cytochrome C3 62,000 0.9 3Fe - 3S a
26,000 4Fe - 4S (2)
Chromatium vinosum uptake unknown 60,000 1.8 4Fe - 4S
Clostridium evolution ferredoxin 60,000 4Fe - 4S (2)
pasteurianum 4Fe - 4S
(HiPiP)
Megasphaera evolution ferredoxin 52,000 0.4 4Fe - 4S (2)
elsdenii 4Fe - 4S
(HiPiP)
Data were collected by K. Schneider, G6ttingen. NO, not determined.

aIt is not clear whether these are real [3Fe - 3S] centers or partly degraded [4Fe - 4S] centers. Number of centers in parentheses.
tv
::s
The compOSItion of the respiratory chain of A. eutrophus and of P.

denitrificans (which is a Hz oxidizer) seems to be identical (see Fig. 5.16).
It is not clear yet at which site the electrons from the particulate
hydrogenase enter the respiratory chain. Since iron-sulfur proteins are
located between NADH dehydrogenase and ubiquinone in the chain and
since hydrogenase contains iron-sulfur centers, the enzyme could feed
electrons into the chain at the ubiquinone or iron-sulfur protein level as
indicated in Fig. 9.2.
Whereas there are no difficulties with the provision of NADH for
COz reduction in Hz oxidizers that contain a soluble NAD+ -dependent
hydrogenase these problems exist in organisms with only a membrane-
bound enzyme. They have to carry out reverse electron transfer for NADH
formation. The basis for NAD+ reduction by reverse electron transfer is
that the redox reactions of the respiratory chain including the proton
translocation are reversible. The portion of the respiratory chain in Fig. 9.2
indicated by dashed lines is oxidized when hydrogen is taken up by the
hydrogenase and fed into the chain. A protonmotive force is built up; it is
partly used for ATP synthesis but also partly consumed to reduce the
portion of the chain enclosed by dashed lines and finally to reduce NAD+.
o
ADP + Pi
ATPsynthase _+-------4---------4- 3W
ATP
NADH dehydrogenase
NADH+ H+_ .....
, ----, I
I
~'- ..........
I
_ _ _ _ .J ................
-2W
hydrogenase
+
iron-sulfur protein
2W
ubiquinone
to
cytochromes
Figure 9.2. Tentative scheme of the electron flow from the membrane-bound
hydrogenase to the components of the respiratory chain. Reverse electron transfer
leading to NADH formation is indicated by dashed lines.
The characteristic enzyme of the carboxydobacteria is carbon monoxide

oxidase, which is membrane-associated and oxidizes CO to CO 2 , The
electrons released are fed into the respiratory chain and travel to oxygen
thereby generating the protonmotive force:
CO + H 2 0 COoxidase, CO 2 + 2H+ + 2e-
2H+ + 2e- + 1/20 respiralorychain) H 0
2 2
Carbon monoxide oxidase is a molybdoprotein; it contains the classic Mo-

cofactor. In addition, iron-sulfur centers and FAD are present in the
protein. The redox potential of the CO / CO 2 couple is very low. Neverthe-
less, the coupling of CO oxidase with carriers in the membrane seems to be
such that NAD+ is reduced by reverse electron transfer as with the
membrane-bound hydrogenase. Carboxydobacteria contain a branched
respiratory chain; one branch is remarkably insensitive to inhibition by
CO, which, of course, is a prerequisite of the ability of these organisms to
grow with CO as energy source.
C. Ammonia, sulfur, and ferrous ion oxidation

Table 9.3 gives the redox potentials of a number of reactions relevant for
chemolithotrophic metabolism. It is obvious that NAD+ reduction is very
difficult for most of the chemolithotrophic organisms. As an example we
consider Nitrobacter species that oxidize nitrite to nitrate:
N0 2 + 1/202 ~ NO;-
The redox potential for the couple N0 2 /NO;- is +0.42 V. Thus, nitrite is
an extremely weak reducing agent for NAD+. The situation is illustrated in

Fig. 9.3. It first becomes clear from this figure that nitrite is not directly
oxidized with 2 , Oxygen from water is incorporated into nitrite. The
Table 9.3. Redox potentials of reactions important

in chemolithotrophic metabolism
Eo(V)
co + HzO~ COz + 2H+ + 2e- -0.54

Hz ~ 2H+ + 2e- -0.41
NADH + H+ ~ NAD+ + 2e- + 2H+ -0.32
HzS ~ S + 2H+ + 2e- -0.25
S + 3HzO ~ SO~- + 6H+ + 4e- +0.05
SO~- + HzO ~ SO~- + 2H+ + 2e- -0.28
NHt + 2H zO ~ N0 2 + 8H+ + 6e- +0.44
N0 2 + HzO ~ N03" + 2H+ + 2e- +0.42
Fe z+ ~ Fe3+ + e- +0.78
Oz + 4H+ + 4e- ~ 2HzO +0.86
NADH+W_,
\
--- ---------
--
- - ----------
- -~~----,- ---
--
J 2e-
L
r---
...J
NOj + 2W
ATP
Figure 9.3. Proton-translocating nitrite oxidase of Nitrobacter. It is assumed that

nitrite oxidase (a cytochrome a I-linked molybdoprotein) reduces cytochrome c
which transfers electrons via cytochrome oxidase to O 2 . The mechanism of proton
translocation is not fully understood.
enzyme nitrite oxidase is located at the inner aspect of the membrane. It is
a molybdo-protein and redox reactions proceed via cytochrome aI, the
more electronegative cytochrome c, and cytochrome aa3 to oxygen. This
transport is associated with proton translocation and further with ATP
synthesis. A large part of the 6,P, however, has to be invested in NAD+
reduction. This presumably is the reason why growth of Nitrobacter is
extremely slow (doubling times in the order of 18 h). To drive the
reduction of 1 NAD+, 5 molecules of nitrite have to be oxidized.
As compared to the oxidation of nitrite to nitrate, which involves one
2-electron transfer, the oxidation of ammonia is more complex. The first
and the only definitely known intermediate in ammonia oxidation is
hydroxylamine; it is formed in a monooxygenase reaction (Fig. 9.4.). The
cosubstrate required by this enzyme (XH 2 in Fig. 9.4) is probably reduced
cytochrome P 460, a specialized b-type cytochrome. The oxidation of
hydroxylamine to nitrite is catalyzed by a membrane-associated enzyme
Oz HzO __ NAD+
NH3~NHZOH __ FeS/FP
_ .. coenzyme Q
XH z X 2~2e-_ _--- cyt b-

ey! c
t I ~OH) --cytu _ _
oz
NO;
Y
Figure 9.4. Scheme of ammonia oxidation to nitrite. 1, Monooxygenase reaction,
XH z is probably reduced cytochrome P460; 2, hydroxylamine-cytochrome c
reductase, which is coupled to a terminal oxidase; 3, nitroxyl is oxidized to nitrite
and regenerates XH z , the cosubstrate in the monooxygenase reaction. Carriers
connected by dashed arrows are involved in reverse electron transfer.
complex hydroxylamine-cytochrome c reductase and an enzyme system

that oxidizes the hypothetical intermediate, nitroxyl (NOH), further to
nitrite and generates XH z . It is apparent from Fig. 9.4 that only the 2
electrons released in the valence change of the nitrogen atom from -1 to
+ 1 are fed into a respiratory chain. Again, reverse electron transfer is
necessary for NAD+ reduction.
Sulfur bacteria may use sulfide, elemental sulfur, or thiosulfate as
electron donors. The routes by which these compounds are oxidized to
sulfate are indicated in Fig. 9.5. Sulfide is oxidized to polysulfide-sulfur in a
reaction that requires glutathione (GSH):
nS 2 - + GSH ~ GSnSH + 2ne-
This is also the form in which elemental sulfur is fed into the oxidation
pathway. The action of sulfur-oxidizing enzyme then leads to sulfite. All
electrons released from the substrate run through the cytochrome section
of the respiratory chain and give rise to the establishment of a proton-
motive force that is used for ATP synthesis or reverse electron transfer to
NAD+. The acceptor of the aerobic thiobacilli is a cytochrome c, and of
the facultative anaerobe T. denitrificans a flavo-protein that is linked to
cytochrome c via cytochrome b. Growth yields of T. denitrificans are
considerably higher than those of T. thiooxidans. The reason could be that
T. denitrificans can take advantage of an additional proton extraction site
between the flavoprotein and cytochrome c.
Thiobacilli usually grow very well on thiosulfate. Several enzyme
systems have been described that feed thiosulfate into the sulfur oxidation
pathway: rhodanese cleaves thiosulfate into elemental sulfur and sulfite;
the action of thiosulfate reducing enzyme yields sulfite and sulfide; a
thiosulfate oxidizing multi-enzyme complex recently described by Kelly
and co-workers gives two sulfates without the accumulation of free
intermediates. This enzyme activity is indicated in Fig. 9.5. The relative
importance of the enzymes acting on thiosulfate is not known yet.
so - - - - - - - lSI
5H 2 0 3 22
lOW + 8e- ~ 4H 2 0 + 2W
8
,~SO~l.",0
APS 5 1/20,
~ 2W+2e- ~H20
8
2S0~- ADP so~-
Figure 9.5. Routes of the oxidation of sulfur compounds in chemolithotrophs. 1,

Oxidation of sulfide to polysulfide sulfur lSI; 2, conversion of elemental sulfur to
polysulfide sulfur; 3, thiosulfate-oxidizing multi-enzyme complex; 4, sulfur oxidase;
5, sulfite oxidase; 6, APS reductase; 7, ADP-sulfurylase; 8, electrons are transfer-
red to oxygen via components of the respiratory chain.
Two enzyme systems have been found to catalyze the oxidation of sulfite
to sulfate. Sulfite oxidase is a membrane-bound molybdo-protein and
transfers electrons to cytochrome c. In addition, the enzyme APS reduc-
tase is present in many sulfur bacteria. Here also a cytochrome functions as
electron acceptor. The APS formed is further converted to sulfate and
ADP. The oxidation of sulfite to sulfate via APS is accompanied by
substrate level phosphorylation (AMP~ ADP).
The oxidation of Fe 2 to Fe 3 by oxygen is taken advantage of
by Thiobacillus ferrooxidans. The redox potential of Fe 2 + /Fe 3 + is
Eo = +780 mY and very close to the one of oxygen. T. ferrooxidans,
however, lives at pH values of about 2 and uses iron oxidation primarily as
a proton-scavenging reaction inside the cell. This is apparent from Fig. 9.6.
The intracellular pH of T. ferrooxidans is about 6 and the organisms have a
~pH of 4 at their disposal. However, acidification by protons flowing in via
the ATP synthase has to be prevented. This is done by Fe 2 + oxidation as
indicated. T. ferrooxidans and T. thiooxidans are of special importance in
leaching of metal ores. Pyritic ores are penetrated by aerated solutions
containing minerals and the organisms. Sulfur is then oxidized to sulfuric
acid. Iron pyrite is oxidized to ferric sulfate and sulfuric acid and
o
pH 6 pH 2
ADP + P;
__- + - - - - - - - + - - - - - - - - t - 3H+
ATP
1/20, + ZH+
H2 0
Figure 9.6. Intracellular H+ consumption in T. ferrooxidans. 1, Rusticyanin, a

high-potential copper-protein; 2, cytochrome oxidase.
iron-copper pyrite to copper sulfate and ferric sulfate:

SO + 1.50z + HzO ---+ H zS0 4
2FeSz + 7.50 z + HzO ---+ FeZ(S04h + H zS0 4
2CuFeSz + 8.50 z + H zS0 4 ---+ FezCS04h + 2CUS04 + HzO
Thus, a copper sulfate solution is resulting from which copper can be
isolated. The basis of uranium leaching is that the insoluble uranium
dioxide reacts with the ferric sulfate produced by the thiobacilli to the
soluble uranyl cation:
UO z + FeZ(S04h ---+ (UO Z)S04 + 2FeS04
Leaching processes work at pH values of 2-3; they can be performed
because of the remarkable acid tolerance of the thiobacilli.
D. Facultative and obligate chemolithotrophs

It has been mentioned that all hydrogen-oxidizing bacteria studied so far
are facultative chemolithotrophs. They grow aerobically with sugars and
amino and organic acids, some of them even with purines and pyrimidines.
Nevertheless, chemolithotrophic metabolism is somewhat dominant in
these organisms. The induction of catabolic enzymes for organic substrate
utilization is thus repressed in a Hz/Oz-atmosphere. An example is given

in Fig. 9.7. Alcaligenes eutrophus grows aerobically on fructose and
employs the Entner-Doudoroff pathway for fructose degradation. Induc-
tion of the enzymes of the Entner-Doudoroff pathway is repressed in a
Hz/Oz-atmosphere, and the organisms do not grow under such conditions.
The mechanism of this repression is not known; it may be related to that of
catabolite repression. In a Hz/Oz-atmosphere ATP is readily available,
and the synthesis of catabolic enzymes is kept repressed irrespective of the
presence or absence of CO z. With some substrates, however, mixotrophic
growth is observed; succinate, for instance, is utilized together with Hz
and Oz.
Several hydrogen-oxidizing bacteria (e.g., Alcaligenes eutrophus) grow
on formate. It is oxidized by formate dehydrogenase to CO z , which is
assimilated as under chemolithotrophic conditions. ATP is formed by
oxidation of NADH in the respiratory chain.
Among the nitrifiers facultative chemolithotrophs are rare. Some Nitro-
bacter strains grow very slowly with acetate. In this connection, however, it
should be mentioned that there is a process called "heterotrophic nitrifica-
tion." It is carried out by a number of Arthobacter and Alcaligenes strains
and others. These organisms produce nitrate by unknown mechanisms.
The occurrence of heterotrophic nitrification in various soils has been
demonstrated. Apparently, organic nitrogen-containing compounds serve
as substrate; the process is very different from the chemolithotrophic
process of nitrification.
Quite a few sulfur oxidizers are facultative chemolithotrophs or even
heterotrophs. T. novel/us, T. intermedius, and Sulfolobus acidocaldarius
are able to grow with organic substrates as well as with sulfur and CO z . T.
perometabolis is a chemolithoheterotroph that is unable to grow auto-
time
Figure 9.7. Aerobic growth of Alcaligenes eutrophus on fructose and repression by
Hz + Oz [G. Gottschalk, Biochem. Z. 341, 260-270 (1965).]
Assimilation of CO 2 295
trophically. With species such as T. novel/us or T. versutus true mixotrophic

growth has been demonstrated: simultaneous utilization of glucose or
acetate and thiosulfate + CO 2 , A number of thiobacilli, however, are
obligate chemolithotrophs. They include T. thioparus, T. neapolitanus, T.
thiooxidans, and T. denitrificans. The question is, what is the basis of
obligate chemolithotrophy? Several explanations can be envisaged.
1. The organisms might be impermeable to organic compounds. However,

the uptake of a number of these compounds and their incorporation
into cellular material has been demonstrated for several obligate
chemolithotrophs.
2. The formation of ATP from organic substrates might be impaired or
impossible. This hypothesis has been advanced by Stanier and collabor-
ators and, in fact, it is very plausible. Chemolithotrophs do not require
a complete tricarboxylic acid cycle, and they have been shown to lack
a-oxoglutarate dehydrogenase (as do the incomplete oxidizers and
several strict anaerobes). Thus, NADH for the respiratory chain cannot
be generated as normally in aerobic heterotrophs. In addition, the
respiratory chains of obligate chemolithotrophs might not be suited to
accept electrons from NADH, as we have seen that the entry of
electrons from the inorganic substrates occurs at the level of
cytochromes.
3. Ammonia oxidizers and sulfur bacteria might suffer from a shortage of
ammonia and reduced sulfur compounds under heterotrophic condi-
tions. If, for instance, Nitrosomonas is incubated with an organic
substrate in a mineral medium under air, ammonia is preferentially
oxidized to nitrite and there is no N source left for biosynthetic
purposes.
The reason for obligate chemolithotrophy might not be the same in all
organisms possessing this property. However, impaired energy metabolism
and/or shortage of ammonia or reduced sulfur compounds offer plausible
explanations.
II. Assimilation of CO 2
Chemolithotrophs use the ATP and the reducing power produced by
oxidation of inorganic substrates to reduce CO 2 and to convert it to cell
material. Since CO 2 functions as sole source of carbon in these organisms,
they are often called C-autotrophs. The mechanism of CO 2 fixation
employed by chemolithotrophs is the Calvin cycle, the same mechanism
that occurs in green plants and was discovered by Calvin, Benson, and
Bassham in green algae.
A. Reactions of the Calvin cycle

The actual CO 2 fixation reaction of the Calvin cycle is the ribulose-I,5-
bisphosphate carboxylase reaction, and the primary product of CO 2
fixation is 3-phosphoglycerate (Fig. 9.8a). One carboxyl group of the 2
molecules of 3-phosphoglycerate formed is derived from CO 2
The next two reactions serve to reduce the carboxyl group of 3-
phosphoglycerate to the aldehyde group. The enzymes involved are
phosphoglycerate kinase and glyceraldehyde-3-phosphate dehydrogenase
(Fig. 9.8b).
The processes of CO 2 fixation and CO 2 reduction bring about the
conversion of 1 molecule of ribulose-1 ,5-bisphosphate into two molecules
of glyceraldehyde-3-phosphate. If the latter is solely used for biosyntheses,
CO 2 fixation would soon come to a standstill because of a shortage of
ribulose-bisphosphate. Thus a third process must follow, the regeneration
of the CO 2 -acceptor. It involves a whole series of reactions; most are
common to us from the discussion of the glycolytic pathways, the pentose
phosphate cycle and the CH 2 0 assimilation (see Figs. 2.3,3.14 and 6.22).
Part of the glyceraldehyde-3-phosphate is first converted into fructose-
6-phosphate by the action of triose-phosphate isomerase, fructose-1,6-
bisphosphate aldolase, and phosphatase (Fig. 9.9). Then, 1 fructose-6-
phosphate, 2 glyceraldehyde-3-phosphates, and 1 dihydroxyacetone phos-
phate react with one another to yield 2 xylulose-5-phosphates and 1
ri bose-5-phosphate.
One important difference in comparison to the pentose phosphate cycle
as it occurs in heterotrophs should be pointed out. In the latter cycle a
transaldolase is involved, which forms sedoheptulose-7-phosphate and
~ ~~20 CHzO
co2 c-o t(OHJ-COOH
I
+ HCOH t=o
I
HtOH HCOH COOH
tHzO tHzO HtOH

I
CHzO
a CO;!ixation
COOH ATP ADP

I \..L
HCOH
2
tHzO
b CO; reduction
Figure 9.8. The first two phases of the Calvin cycle, CO 2 fixation (a) and CO 2
reduction (b). 1, Ribulose-1,5-bisphosphate carboxylase. The unstable intermedi-
ate is 2-carboxy-3-oxoribitol-1,5-bisphosphate; 2, 3-phosphoglycerate kinase; 3,
glyceraldehyde-3-phosphate dehydrogenase.
Assimilation of COz 297
fructose-6-P
OH OH
I I
OHC-C-C-eH 20
~ A
ID~"''''I '"'h'F4-P
H OH OH OH
I I I I
OH 2C-CO-C-C-C-C-CH 2 0
I I I I
OH H H H
sedoheptulose-I,7-P z
HZO~Pi
H OH OH OH
I I I I
HOH 2 C-CO--c-C -C-C-CHzO
I I I I

OH H H H
sedoheptulose-7-P
4
ribose-5-P
xylulose-5-P I
3ATP 3ADP
2 xYlulose-5-P~ \. 9.)
. ~ 3 ribulose-5-P .. 3 ribulose-I, 5-P z
nbose-5-P 8
Figure 9.9. The third phase of the Calvin cycle, acceptor regeneration. 1,
Triosephosphate isomerase; 2, fructose-1,6-bisphosphate aldolase; 3, fructose-1,6-
bisphosphatase; 4, transketolase; 5, sedoheptulose-1,7-bisphosphate aldolase; 6,
reaction is catalyzed by fructose-1,6-bisphosphatase; 7, phosphopentose
epimerase; 8, ribosephosphate isomerase; 9, phosphoribulokinase.
glyceraldehyde-3-phosphate from erythrose-4-phosphate and fructose-6-
phosphate in a reversible reaction. In the Calvin cycle, this reaction is
replaced by the reversible sedoheptulose-l,7-bisphosphate aldolase and
the irreversible phosphatase reaction. This guarantees a unidirectional
substrate flow toward pentose phosphates, which following epimerization
and isomerization reactions, are finally phosphorylated by phosphoribulo-

kinase to yield the CO 2-acceptor.
The carbon balance of the whole cycle can be described schematically as
follows:
3C s + 13C02! - 5C3 + ~
fixation and reduction
5C 3 - 3C s
regeneration
The amounts of ATP and NADH required for the formation of one
glyceraldehyde-3-phosphate (GAP) from CO2 are apparent from the
following equation:
3C0 2 + 9ATP + 6NADH + 6H+ ------> GAP + 9ADP + 8P i + 6NAD+
The entire cycle is represented in Fig. 9.10.
-----
~
~~~
RlI-S-P
RlI-5-P
/ ~R"-5~ PGA~\ IGAP I

!Ri-{r
XlI-5-P
PGA
GAP
GAP
XlI-5-P
GAP
Sll-7-P
\ Sll-P 2
DAP
E-4-P
Figure 9.10. The Calvin cycle. Ru-P z, Ribulose-l,5-bisphosphate; PGA,

3-phosphoglycerate; GAP, glyceraldehyde-3-phosphate; DAP, dihydroxyacetone
phosphate; F-P z , fructose-l,6-bisphosphate; F-6-P, fructose-6-phosphate;
E-4-P, erythrose-4-phosphate; Su-P z , sedoheptulose-l,7-bisphosphate; Su-7-P,
sedoheptulose-7-phosphate; Xu-5-P, xylulose-5-phosphate; Ri-5-P, ribose-5-
phosphate; Ru-5-P, ribulose-5-phosphate.
Assimilation of CO 2 299
B. Phosphoribulokinase and
ribulose-1,S-bisphosphate carboxylase
Two enzymes can be regarded as key enzymes of the Calvin cycle,
phosphoribulokinase and ribulose-l,5-bisphosphate carboxylase. They
occur only in organisms that are able to fix CO 2 via the Calvin cycle, and
the evidence that the chemolithotrophs employ this cycle rests primarily on
the presence of the kinase and the carboxylase in these organisms.
Furthermore, it has been shown that 3-phosphoglycerate is the primary
CO 2 fixation product in chemolithotrophs as it is in green plants.
Phosphoribulokinase from Alcaligenes eutrophus that has been thor-
oughly studied has a molecular weight of 260,000 and is an octameric
enzyme containing identical subunits. NADH functions as positive effector
for the kinase of this organism and other chemolithotrophs. AMP inhibits
this enzyme so that ATP can be conserved in times of energy deprivation.
Interestingly, PEP has also been found to be a strong inhibitor of the
kinase of chemolithotrophs; thus an accumulation of PEP prevents the
further formation of ribulose-l ,5-bisphosphate, and as a consequence the
further operation of the Calvin cycle.
Ribulose-l,5-bisphosphate carboxylase has been isolated from many
phototrophic and chemolithotrophic organisms. The composition of this
enzymes is not identical in all organisms. Most organisms contain a
carboxylase made of eight large and eight small subunits. The molecular
weights of these subunits are 54,000 and 13,000, respectively. As compared
to this large type, the Rhodospirillum rubrum carboxylase is surprisingly
small; it is built up of just two large subunits. This variation in composition
indicates that the large subunit is the catalytic one. The function of the
small subunit is still in debate.
The active ribulose-l,5-bisphosphate carboxylase complex contains
enzyme, Mg2+, and CO 2 . This complex is ready to carboxylate ribulose
1,5-bisphosphate. In the absence of CO 2 and in the presence of oxygen the
carboxylase catalyzes another reaction, the oxygenolytic cleavage of ribu-
lose-l,S-bisphosphate to yield phosphoglycolate and phosphoglycerate (the
ribulose-l,5-bisphosphate oxygenase reaction).
This activity thus competes with the carboxylation and can significantly
reduce net CO 2 fixation in autotrophs. This is particularly evident by the
photorespiratory loss of carbon in plants.
Phosphoglycolate is converted to glycolate by a phosphatase, and

glycolate contributes to photorespiration in plants. Phototrophic and
chemolithotrophic bacteria also produce glycolate, part of which is ex-
creted.
Ribulose-l,5-bisphosphate carboxylase is strongly inhibited by 6-
phosphogluconate. This seems reasonable for facultative chemolitho-
trophs. 6-Phosphogluconate signals that the oxidative pentose phosphate
cycle is operating and that organic substrates are available. An active
carboxylase is not required under such conditions.
Interesting organelles were discovered by Shively and co-workers in
Thiobacillus neapolitanus and later by others in many chemolithotrophic
organisms: the carboxysomes. These polyhedral inclusions consist of
ribulose-l,5-bisphosphate carboxylase; their biological function is not
absolutely clear, but they may serve as a protein storage. Finally it should
be mentioned that ribulose-l,5-bisphosphate carboxylase is the most
abundant enzyme protein in nature. In certain chemolithotrophs it
amounts to up to 20% of the total protein.
III. Phototrophic Metabolism

Photosynthesis as carried out by green plants and cyanobacteria is accom-
panied by the evolution of oxygen; photosynthesis as carried out by
phototrophic bacteria is different: it proceeds only under anaerobic
conditions, and oxygen is not evolved. Since water cannot be utilized as
hydrogen source, a suitable H-donor is required by phototrophic bacteria
for growth. Such hydrogen donors are reduced sulfur compounds, molecu-
lar hydrogen, and organic substrates. Not every H-donor can be used by
every phototrophic species, and on the basis of the compounds preferred
and other properties four families of phototrophic bacteria are distin-
guished: Rhodospirillaceae, Chromatiaceae, Chlorobiaceae and Chloro-
f1exaceae.
A. The four families of phototrophic bacteria

Some properties of the phototrophic bacteria are summarized in Table 9.4.
Reduced sulfur compounds are preferentially used as H-donors by the
Chromatiaceae and the Chlorobiaceae. Only a few species of the Rhodo-
spirillaceae (e.g., Rhodopseudomonas palustris, R. sulfidophila) utilize
thiosulfate and hydrogen sulfide as H-donors readily; none of these species
can oxidize elemental sulfur. Species from all four families are able to
employ H 2 for CO 2 reduction.
The relationship of phototrophs to organic substrates is different. Green
sulfur bacteria are able to photoassimilate only simple organic substrates
such as acetate and butyrate; they will not grow in the absence of CO 2 and
'"0
0-
o
o
a
"0
0-
r;.
Table 9.4. Some properties of the four families of phototrophic bacteria ~
or
cr-
Rhodospirillaceae Chromatiaceae .
Vi
(purple nonsulfur (purple sulfur Chlorobiaceae Chloroftexaceae a
bacteria) bacteria) (green sulfur bacteria) (green gliding bacteria)
bacteriochlorophylls present BCHL a or b BCHL a or b BCHL a and c, d, or e BCHL a and c or d

use of H 2 S as H-donor - (some +) + + (+)
accumulation of SO and use
of SO as H-donor + (it + (e)b
use of H 2 as H-donor + + + +
use of organic substrates as + + +
H-donor
carbon source CO 2 CO 2 CO 2 CO 2
organic organic organic organic
substrates substrates substrates substrates
aerobic dark growth + +
a i Intracellular accumulation except genus EClolhiorhodospira.
be Extracellular accumulation.
l>J
o
.....
R, R2 Table 9.5. Chemical structure and absorption maxima of the various bacteriochlorophylls. In bacterio-
chlorophylls a and b carbon atoms 3 and 4 are linked by a single bond a
(v~~R3
bacteriochlorophyll
}-~" /~=<
R7yC~/M~g,,FH substituent a b c d e
H C N N
4
~ _ I C~ # R R1 -CO-CH 3 -CO-CH 3 -CHOH-CH3 -CHOH-CH 3 CHOH-CH 3
H' \ Rz -CH3 -CH 3 -CH3 -CH 3 -CHO
CH 2 HC-C R3 -CzH s =CH-CH 3 -CzH s -CzH s -CzH s
I :' ~ R4 -CH3 -CH 3 -CzH s
CH 2 Rs 0 -CzH s -CzH s
I Rs -CO-OCH3 -CO-OCH 3 -H -H -H
C=O phytyl phytyl farnesyl farnesyl farnesyl
I R6
o R7 -H -H -CH3 -H -CH 3
I absorp. max. b 850-910 1020-35 745-60 725-45 715-25
Ro
(nm)
a A. Gloe, N. Pfennig, H. Brockmann, and W. Trowitzsch. Arch. Microbial. 102, 103-109 (1975).
b The absorption maxima are from spectra of whole cells and represent the spectral properties of the bacteriochlorophylls
interacting with the components of the light-harvesting centers.
Phototrophic Metabolism 303
reduced sulfur compounds. Many purple sulfur bacteria utilize organic

substrates even in the absence of reduced sulfur compounds and grow
photoheterotrophically . Finally, the purple nonsulfur bacteria and the
green gliding bacteria grow as photoheterotrophs or, in the presence of
oxygen, as chemoheterotrophs, organic acids being the preferred sub-
strates. Only one species of the purple sulfur bacteria grows in the dark
under microaerophilic conditions, Thiocapsa roseopersicina. It should be
stressed that anaerobic conditions are indispensable for bacterial photo-
synthesis. Thus, Rhodospirillaceae will grow as chemoheterotrophs in air
regardless of whether light is available or not. Only under anaerobic
conditions will the components of the photosynthetic apparatus be synthe-
sized and arranged in the membrane.
The various bacteriochlorophylls occurring in phototrophs are listed in
Table 9.4, and the chemical differences between these chlorophylls and
their absorption maxima are summarized in Table 9.5. It is apparent that
five different bacteriochlorophylls (BCHL) occur in phototrophic bacteria.
The purple bacteria contain a or b and the green ones c, d, or e in addition
to a. The absorption maxima of these bacteriochlorophylls as measured
with whole cells are also given. Isolated bacteriochlorophylls absorb at
shorter wavelengths and the shift to longer wavelengths in vivo is attri-
buted to interactions with proteins. The shift of the absorption maximum
to the near infrared enables phototrophic bacteria to perform photo-
synthesis with light that is not absorbed by green plants or cyanobacteria
(Fig. 9.11). This shift is most pronounced in phototrophs such as R. viridis
that contain BCHL b instead of BCHL a.
Cells of phototrophic bacteria exhibit a second region of absorption
between 400 and 500 nm. The bacteriochlorophylls and the carotenoids
contribute to this absorption. Carotenoids are responsible for the red,
purple, and brownish color of phototrophs. They protect the cells from
photooxidative damage. In addition, they are also involved in transfer of
light energy. Various types of carotenoids occur in phototrophs. Green
sulfur bacteria contain predominantly carotenoids bearing aromatic rings;
as an example the formula of isorenieratene is given in Fig. 9.12. A typical
carotenoid of the purple bacteria is spirilloxanthin.
B. Reactions of the photosynthetic apparatus

The components of the photosynthetic apparatus are arranged in mem-
brane systems such that light can be absorbed and the energy be used to
generate a proton motive force across the membrane. Then the 6.P can be
taken advantage of as usual for ATP synthesis or solute transport.
Differences as to the organization of the photosynthetic apparatus and the
components involved have been recognized between the families of purple
bacteria and the families of green bacteria. Photosynthesis by purple and
by green bacteria, therefore, will be discussed separately.
visible spectrum A
,.
"
"
"
," I
I :
I ,
, I
J t
I I
J I
, I
,
\
\
'" ,
200 400 600 800 1000
.
,
B
"\
I \
<-
u
, ,
J I
I \
'""'
, ,
.D I I
, I
~
.D
", I
" 'I'
" , I
I
I IV '
200 400 600 800 1000

wavelength (nm)
Figure 9.11. Absorption spectra of a green alga and of phototrophic bacteria. A:
Chlorella ( ); Chlorobium limicola (-----). B: Rhodopseudomonas viridis
( ) (BCHL b); R. sphaeroides (BCHL a) (-------).
isorenie-ratenr (3)
spirilloxanthin (b\
Figure 9.12. Typical carotenoids of green sulfur bacteria (a) and of purple
bacteria (b).
Purple bacteria
The membrane surface of these organisms is enlarged by several in-

vagi nations to provide enough space for the photosynthetic apparatus.
Light is absorbed by the light harvesting (LH) centers. They consist of
bacteriochlorophyll, carotenoids, and several proteins. Two types of LH
centers designated B850 and B875 are present in most purple bacteria. The
LH centers contain more than 90% of the total BCHL. From these
antennae pigments the energy is transferred to the reaction centers.
Several light harvesting centers (20-40) are linked to one reaction center.
The latter is composed of 4 molecules bacteriochlorophyll, 2 molecules
bacteriopheophytin (Mg_ 2 + free bacteriochlorophyll), and a protein con-
sisting of three different types of subunits. It is referred to as P870. Two of
the bacteriochlorophyll molecules are photochemically active. They are
referred to as the "special pair". Associated with the reaction center is an
iron-ubiquinone complex. Both entities are integrated into the membrane
as a "super complex" with the reaction center facing the outer surface of
the membrane and the Fe-Q complex facing the inner surface of the
membrane (Fig. 9.13). A second membrane-integrated complex is the
one formed by a b- and a crtype cytochrome plus an iron-sulfur protein.
This one is referred to as cytochrome bCI complex. Other components
of the photosynthetic apparatus are ubiquinone and cytochrome C2, both
being mobile carriers within the membrane (see Fig. 9.13). Several mole-
cules of these carriers are present per reaction center (around 12 ubi-
quinones per RC). What happens when the light energy is transferred
to the reaction center is schematically shown in Fig. 9.14. Per light
2H+
o
ADP + Pi ATP
Figure 9.13. Localization of the components of the photosynthetic apparatus in the

membrane. The light harvesting centers and the two membrane-integrated com-
plexes are apparent. Also depicted is the ATP synthase. The Q cycle allows the
translocation of 2 protons per electron.
En(mY)
-400
-200
o
Fe-S
bel
+200
Iight~
LH
+400 P870
LII
Figure 9.14. Cyclic electron transport in purple bacteria. Bph, Bacteriopheophytin;

Fe-Q, iron-ubiquinone complex; Q, ubiquinone; Fe-S, iron-sulfur protein bCI'
cytochrome bc} complex; C2, cytochrome C2. The steep decline from Bph (primary
acceptor) to Fe-Q shall indicate that this reaction is not taken advantage of for
proton extraction (see text).
quantum absorbed one electron is translocated from the "special pair"
via a "secondary" bacteriochlorophyll to a bacteriopheophytin mole-
cule (Eo = -600 mY) and further to the ubiquinone-iron complex
(Eo = -180 mY). At the same time P870 is reduced again by a high
potential cytochrome, cyt C2 (Eo = +300 mY). The potential drop from
-600 to -180 mY is not taken advantage of; it is sacrificed to set the
direction of electron flow. From the ubiquinone-iron complex the electron
flows via ubiquinone to the cytochrome bc 1 complex and from there to cyt
C2. All carriers are now again in the same redox state as before; a cyclic
electron transport has occurred.
The benefit of this cyclic electron transport for the organism performing
it is that it provokes a pH gradient. It has been estimated that per electron
transported two protons are translocated from inside to outside of the cell.
This requires the special organization of the complexes and carriers as
depicted in Fig. 9.13. Ubiquinone is reduced by the P870-Fe-Q-super-
complex; it travels to the bCI complex. Protons are released to the outside,
and the electron is transferred to cyt C2 which travels back and forth
between the reaction center and the bCI complex. A Q cycle is operative
(see Fig. 2.13) so that the stoichiometry is 2 protons per electron.
NADH is generated by the purple bacteria with either organic sub-
strates (Rhodospirillaceae) or HzS (Chromatiaceae) as H-donors. In addi-
tion, Hz can be utilized. Two mechanisms lend themselves for NADH
formation. If allowed by the redox potential, the substrate is oxidized in
NAD+ -depending reactions (e.g., L-malate by malate dehydrogenase or
ethanol by alcohol dehydrogenase). For all other substrates (also Hz)
reverse electron transfer is involved. The electron acceptor of hydrogenase
is a cytochrome c. Succinate feeds in electrons at the level of ubiquinone.
The oxidation of HzS via elemental sulfur to sulfate proceeds as outlined
for the sulfur-oxidizers, but the electrons accepted by a cytochrome care
used to reduce NAD+. In all these reactions light is only indirectly
involved, it provides the 6.P to drive reverse electron transfer.
Green bacteria
The organization of the photosynthetic apparatus in the two families
of green bacteria is different. Here the light harvesting centers are not
localized on invaginations of the cytoplasmic membrane but are present in
special organelles, the so called Chlorobium vesicles or chlorosomes.
These vesicles are surrounded by a non-unit membrane that consists of a
galactolipid and lies just underneath the cytoplasmic membrane. They
contain essentially all bacteriochlorophyll c, d, or e and in addition, most
of the carotenoids and proteins. A' model of the chlorosome is shown in
Fig. 9.15. Light is absorbed by these pigmented organelles and transferred
cytoplasmic membrane
intramembrane
particles, probably
reaction centers
Figure 9.15. Model of a chlorosome and the associated cytoplasmic membrane
[L. Staehelin, J. R. Golecki, and G. Drews. Biochem. Biophys. Acta 589, 30-45
(1980).]
to the reaction centers that are situated in the cytoplasmic membrane. The
ratio, light harvesting center-bacteriochlorophyll: reaction center-bacterio-
chlorophyll is in the order of 1000 and is much higher than for purple
bacteria (40). Thus, the chlorosomes function as some sort of super-
antennae, and the green bacteria are well adapted to take advantage of
weak light intensities.
The photochemically active pair of bacteriochlorophyll molecules re-
sides in a reaction center called P840. The fate of the electron expelled
from it is depicted in Fig. 9.16. The primary acceptor is probably not a
bacteriopheophytin but an iron-sulfur protein with Eo= -540 mY. The
principal difference as compared to the purple bacteria is now that the
electron cannot only travel via carriers to cyt C555 which is the reductant for
P840. It can alternatively be transferred to NAD+ via ferredoxin. In this
case the "electron hole" at cyt C555 has to be filled at the expense of an
external electron donor. This is accomplished by sulfide oxidation which is
mediated by enzymes that couple this process to the reduction of
cytochromes.
Eo(mV)
-600
-400 ----..,
I N!O+ I
-200 e S2--r- so~
I
I
I
Fe-S __ -II
b
o
+200
+400
Figure 9.16. Cyclic electron transport in green bacteria and light-dependent

reduction of NAD+ _MK, Menaquinone; Fe-S, iron-sulfur protein; b, cytochrome
b; C555, cytochrome C555-
The mechanism of proton extraction is analogous to the one in purple

bacteria. The reaction center and the primary acceptor form one complex
and a cytochrome b plus an iron-sulfur protein the other; menaquinone
and cytochrome C555 function as mobile carriers.
C. The photosynthetic apparatus of cyanobacteria

The cyanobacteria-until recently called blue-green algae-comprise a
large group of organisms. They are typical prokaryotes with 70S ribosomes
and with peptidoglycan as cell wall constituent. Unicellular cyanobacteria
are Synechococcus and Gloeocapsa species; well known filamentous forms
are Oscillatoria, A nahaena, and Spirulina species. The latter grow in certain
lakes in large amounts and serve as food in Central Africa.
The cyanobacteria carry out an oxygenic photosynthesis like the green
plants. They play an important role in nature as primary producers
of organic material, because they use light as energy source, water as
electron donor, and CO 2 as carbon source. In addition most of them are
able to use N2 as nitrogen source.
The photosynthetic apparatus of the cyanobacteria differs from the one
of phototrophic bacteria in that other redox carriers are involved and in
that it comprises two light reactions. One light reaction (PS I) covers a span
from approximately +400 mY to -600 mY. It somewhat resembles the
photosystem of the green bacteria. The second system (PS II), however,
pushes electrons up from a reaction center as far positive as +850 mY to
approximately 0 mY. It is this system that allows the use of water as
H-donor.
Cyanobacterial photosynthesis is depicted in Fig. 9.17. It is localized in
intracytoplasmatic membrane systems (thylakoids). These membranes are
covered with phycobilisomes, the light harvesting centers of PS II. The
phycobilisomes consist of three components: phycocyanin, phycoerythrin,
and allophycocyanilI. All three are proteins associated with tetrapyrrol
dyes and absorb around 600 nm. The light harvesting centers of RCI are
built up of chlorophyll a and proteins.
Light above 660 nm is harvested and transferred to P700. An electron
is ejected, an iron-sulfur protein with an Eo = -600 mY function as
acceptor, and a cyclic electron transport is established via ferredoxin,
plastoquinone, the iron-sulfur protein-cytochrome bf complex and
plastocyanin. Again, two membrane-integrated complexes are involved
(RCI-Fe-S complex and Fe-S-cyt bf complex) and mobile carriers
(ferredoxin + plastoquinone and plastocyanin). As discussed for the
phototrophic bacteria, electron transfer from the quinone to the
Fe-S-cytochrome bf complex is then associated with proton export.
It is apparent from Fig. 9.17 that the reduced iron-sulfur protein not
only feeds electrons into the cyclic process, but functions also as electron
donor for NADP+. Under these conditions an external electron donor is
Eo(mV)
-600
-400
-200
+200 Fe-S
bi
+400 e
~ght
+600
(Mn 2+) P680

+800 1/2HP - - : , - - RC II ~ght
"
1/42 + W
Figure 9.17. Photosynthetic apparatus of cyanobacteria. RC I and RC II,
Reaction centers I and II; B, light harvesting centers of RC I consisting of
chlorophyll a-protein complexes; P, phycobilisomes, the light harvesting centers of
RC II; PO A - Fe, plastoquinone-iron complex; PO, plastoquinone pool; Fe-S,
iron-sulfur protein; bI, cytochrome bs63 -cytochrome I complex (cyt I belongs to the
c-type cytochromes); PC, plastocyanin (a copper protein).
required which actually is water. The following view of the relationships

between photosystem I and II can be developed: If NADP+ is reduced
with electrons coming from P700, then the plastoquinone pool becomes
more and more oxidized. PO is present in the membrane in much higher
concentrations than the other components. So it functions as some sort of
electron reservoir for P700. Eventually PO has to be reduced. This is
accomplished by RC II in the second light reaction by which an electron is
ejected from P680 to the primary acceptor PQA - Fe (a plastoquinone

complex). Subsequently PQ is reduced. The redox potential of
P680jP680+ is Eo = +850 mY. This is so positive that P680+ will oxidize
water to oxygen under catalysis of a manganoprotein. The stoichiometry,
H+ translocated per electron is not clear. Both cyclic and noncyclic
electron flows are coupled to proton translocation, and the proton gradient
generated is then used by the membrane-embedded ATP synthase for ATP
formation.
There is at first sight an intriguing analogy between oxygenic photo-
synthesis and photosynthesis as carried out by purple and green sulfur
bacteria:
2H zS + 2NAD+ ~ 2S + 2NADH + 2H+
2H zO + 2NAD(P)+ ~ Oz + 2NAD(P)H + 2H+
We have seen that these processes are mechanistically very different. They
have to be, because of the large difference in the redox potentials of the
electron donors.
The NAD(P)H produced by noncyclic electron flow in photosynthesis is
used by the organisms for the reduction of COz. It should be stressed,
before the relevant COz fixation pathways will be discussed, that a
stringent coupling exists in phototrophic organisms between the generation
of reducing power in the light-dependent processes and its consumption for
COz reduction. In other words, a light-dependent Hz evolution from water
or HzS is not observed, which is a pity. Molecular hydrogen produced from
water in a photoprocess would be the cheapest energy source one can think
of. Therefore, it is understandable that scientists try to modify cyanobacte-
rial and plant systems such that they will evolve Hz.
D. Carbon metabolism
When growing with COz as carbon source, cyanobacteria, Rhodos-
pirillaceae, Chromatiaceae, and Chloroflexaceae will employ the reactions
of the Calvin cycle for COz fixation. The key enzymes of this cycle,
phosphoribulokinase and ribulose-l,5-bisphosphate carboxylase, have
been shown to be present in these organisms. These enzymes could,
however, not be found in the Chlorobiaceae. This family of organisms is
unique in that it uses the reductive tricarboxylic acid cycle for the
conversion of COz into acetyl-coenzyme A. This cycle, which was first
formulated by Arnon, Buchanan, and Evans and confirmed with isotope
experiments by Fuchs, takes advantage of the fact that most of the
tricarboxylic acid cycle reactions are reversible. The following exchanges
of enzymes are necessary to drive the cycle the other way around:
1. Succinate dehydrogenase is not favorable for fumarate reduction and is
replaced by fumarate reductase which probably uses NADH as H-
donor.
2. The reaction catalyzed by the a-oxoglutarate dehydrogenase multien-

zyme complex is irreversible. It is replaced by a-oxoglutarate: ferredox-
in oxidoreductase:
succinyl-CoA + CO 2 + FdHz ---> a-oxoglutarate + Fd + CoASH
3. The citrate synthase reaction is irreversible under physiological condi-
tions. It is replaced by ATP-citrate lyase:
citrate + CoASH + ATP --->
oxoloacetate + acetyl-SCoA + Pi + ADP
It should be noted that ATP-citrate lyase was considered as a typical
eukaryotic enzyme for a long time. The Chlorobiaceae are the only
prokaryotes known to possess it.
The entire cycle is presented in Fig. 9.18. The result is that 2 CO 2 are
_o_xa_lo_ac_et_a_te_1 ~""~'"
7 cO'~'"
oxaloacetate
, I
co,
acetyl..(:oA
!-","
L-malate
ADP + Pi
ATP + CoA
rumarate t:itrate
'"1'
succinate
'~"'O
cis-aconitate
"'"1
A
CoA T P
4 \
ADP + Pi
A
succinyl-CoA isocitrate
CoA 2H
co,
a::-oxog!utarate
Figure 9.18. The reductive tricarboxylic acid cycle of the Chlorobiaceae. 1, Malate
dehydrogenase; 2, fumarase; 3, fumarate reductase; 4, succinyl-CoA synthetase; 5,
a-oxoglutarate synthase; 6, isocitrate dehydrogenase; 7, cis-aconitase; 8, ATP-
citrate lyase; 9, pyruvate synthase; 10, PEP synthetase, 11, PEP carboxylase. Four
CO 2 molecules are fixed and converted into 1 oxaloacetate.
reduced to acetyl-CoA. The latter can be carboxylated to pyruvate as in

Clostridium kluyveri and, correspondingly, converted further to PEP and
oxaloacetate by the action of PEP synthetase and PEP carboxylase.
Growth of the Chlorobiaceae is largely stimulated by acetate, propion-
ate, or pyruvate. Oxaloacetate is inhibitory, and it has been shown to
interfere with ATP-citrate lyase activity.
Rhodospirillaceae grow on acetate anaerobically in the light and also
aerobically. Rhodospirillum tenue, Rhodopseudomonas palustris, and
Rhodomicrobium vannielii contain isocitrate lyase and malate synthase and
employ the glyoxylate cycle as anaplerotic sequence. Species like Rhodo-
spirillum rubrum and Rhodopseudomonas sphaeroides lack isocitrate lyase
but contain malate synthase. How glyoxylate is formed in these organisms
is not known. Nevertheless, it seems that the purple nonsulfur bacteria use
a glyoxylate cycle in order to synthesize C4 dicarboxylic acids from acetate
whereas direct carboxylation reactions (acetyl-CoA ~ pyruvate ~
oxaloacetate) are employed by purple sulfur and green bacteria.
Sugars are not the preferred substrates of phototrophs. Many species do
not utilize them at all. Thiocapsa roseopersicina grows on fructose and so
does Rhodospirillum rubrum. Rhodopseudomonas sphaeroides and Rps.
capsulata utilize fructose and glucose. The former sugar is degraded via the
Embden-Meyerhof-Parnas pathway and the latter via the Entner-
Doudoroff pathway.
Rhodopseudomonas palustris grows aerobically on p-hydroxybenzoate;
the compound is degraded via protocatechuate and the meta-fission
pathway. The key enzymes of this pathway are virtually absent from
extracts of Rps. palustris grown phototrophically on p-hydroxybenzoate or
benzoate. This fact led to the discovery of a new pathway for the
breakdown of aromatic compounds; it does not involve Oz. As shown in
Fig. 9.19 benzoate is reduced to cyclohex-1-ene-1-carboxylate, which is
further degraded to pimelate. How widely this pathway is distributed
among phototrophs is not known.
benzoate cyclohex-I-ene- 2-hydroxycyclohexane

carboxylate carboxylate
COOH
~O COOH
U -\.-"-----
2H H2 0
I
L. (Cf H 2)S
COOH
2-oxocyclohexane pimelate
carboxylate
Figure 9.19. Breakdown of benzoate by Rps. palustris anaerobically in the light.
E. Photoproduction of molecular hydrogen

Gest and Kamen first observed that cultures of RhodospirilLum rubrum
produced hydrogen when grown phototrophically on organic acids with
amino acids (e.g., glutamate) as nitrogen source. Resting cells of such
cultures photometabolized substrates such as acetate, lactate, and malate
completely to COz and Hz:
CH 3COOH + 2H zO ~ 2CO z + 4H z
HOOC-CHz-CHOH-COOH + 3H zO ~ 4CO z + 6H z
Photoproduction of hydrogen is inhibited by NH: and by nitrogen,
which indicates involvement of the enzyme nitrogenase in Hz evolution.
Like a number of other Rhodospirillaceae, RhodospirilLum rubrum is
capable of fixing nitrogen; the formation of nitrogenase (see Chapter 10) is
derepressed at low NH: concentrations (as is the case during growth with
glutamate as N source), and, in the absence of molecular nitrogen, the
hydrogenase activity of the nitrogenase system produces Hz from NADH
in a reaction requiring ATP. Thus, the organic acids are oxidized via the
tricarboxylic cycle, and the reduced coenzymes generated are oxidized by
the nitogenase system. Since the actual H-acceptor, N z , is not available,
the reducing power is released as molecular hydrogen.
F. Photosynthesis in halobacteria
Halobacteria live in salt lakes and are adapted to the high salt concentra-
tion in their environment and to the high light intensity in these usually
shallow lakes. They are not viable at NaCI concentrations lower than
2.5 M and grow best in media containing about 5 M NaCI (sea water is
about 0.6 M). Cell stability requires that the osmotic pressure of the cell's
interior is approximately the same as the one outside. Halobacteria are,
therefore, approximately 1 M in Na + and 3 M in K + .
The membrane of the halobacteria is red and contains carotenoids
which serve to prevent photochemical damage of the cells. In addition,
purple-colored patches are present in the membrane. They are numerous if
the cells are grown at low concentrations of oxygen in the light and they
have a very interesting function. Halobacteria are obligate aerobes that
utilize amino and organic acids. At low concentrations of oxygen (the
solubility of Oz in concentrated salt solutions is low), ATP is not only
generated by oxidative phosphorylation but also by photosynthesis.
Stoeckenius, Oesterhelt and co-workers found that the purple areas of the
membrane contain bacteriorhodopsin. Like rhodopsin, the pigment of the
rod cells of the human eye, bacteriorhodopsin consists of a protein and of
the chromophore retinal, which is linked to a lysine residue of the protein
(Figure 9.20). The purple membrane acts as a proton pump. The Schiff
base in bacteriorhodopsin is protonated and retinal is in the all-trans form.
In a light-dependent reaction this purple pigment is bleached. This is the
all-trans fonn
form
~ -N~C~
13-<:15
< L- .I
'" ""l ( ]
lysJne-N-+H;O+cH . " ' ' ' '
~
polypeptide retinal
Figure 9.20. Composition of bacteriorhodopsin. Retinal, the chromophore, is
linked to the amino group of a lysine residue of the apoprotein (bacterioopsin). The
Schiff base thus formed loses a proton in a photoreaction that converts the trans
into the cis form. The back reaction is a dark reaction.
result of an isomerization of the retinal to the 13-cis form and is
accompanied by a loss of a proton. This proton is released to the outside,
and in a dark reaction a proton is taken up again from the cytoplasmic side
of the membrane. Thus, an electrochemical gradient is established across
the membrane, which can be made use of for ATP synthesis (Fig. 9.21).
purple cell wall

membrane
light
red
membrane
Figure 9.21. Proton transport as mediated by the purple membrane in a light-

powered process. [W. Stoeckenius. Sci. Am. 234, 38-46 (1976).]
inside
Figure 9.22. Location of the bacteriorhodopsin polypeptide chain in the mem-
brane. (e), Tryptophan residues. Retinal is bound to lysine residue 219. The
proton released could be passed via tryptophan residues to the outside. [Redrawn
from K.-S. Huang, R. Radhakrishnan, H. Bayley, H. G. Khorana. J. BioI. Chern.
257, 13616-13623 (1982).]
Recent studies have shed some light on the mechanism of proton

extraction. Bacteriorhodopsin is situated in the membrane in a "zig-zag"
manner as shown in Fig. 9.22. The N-terminus of the polypeptide chain is
outside the membrane and the C-terminus inside. The chain is folded such
that seven helical stretches pass through the membrane. Interestingly, the
tryptophan residues of bacteriorhodopsin are exclusively located towards
the outside. It is assumed that the proton released from retinal is passed to
the outside via these tryptophan residues and that retinal is protonated
again in reactions with the carboxyl groups of aspartate and glutamate
residues. Bacteriorhodopsin is one of the best-studied membrane proteins.
IV. Summary
1. Chemolithotrophic microorganisms gain ATP by oxidation of inorga-
nic compounds with oxygen or in some cases with nitrate. The inorganic
compounds oxidized are molecular hydrogen (hydrogen-oxidizing bacte-
ria); CO (CO-oxidizing bacteria); hydrogen sulfide, elemental sulfur, and
thiosulfate (sulfur-oxidizing bacteria); ferrous ions (iron-oxidizing bacte-
ria); ammonia (ammonia oxidizers) and nitrite (nitrite oxidizers).
2. ATP is synthesized by a chemiosmotic mechanism. The electrons of
hydrogen are fed into a respiratory chain via a membrane-bound hy-
drogenase. The oxidation of nitrite to nitrate is linked to the reduction of
cytochrome at; its oxidation with oxygen via cytochrome c and a31eads to
the generation of a AP. Ammonia is oxidized to hydroxylamine in a
Summary 317
monooxygenase reaction, and the electrons from hydroxylamine oxidation

are fed into a respiratory chain. Hydrogen sulfide is oxidized to sulfate via
elemental sulfur and sulfite. Two routes leading from sulfite to sulfate have
been found; sulfite oxidase and APS reductase + ADP sulfurylase.
3. Cell material is formed from CO 2 , With the exception of some
hydrogen-oxidizing bacteria chemolithotrophs use reverse electron trans-
fer in order to reduce NAO+ with their hydrogen donors (nitrite,
ammonia, elemental sulfur, H 2 , etc).
4. All known hydrogen-oxidizing bacteria, Thiobacil/us novel/us,
T. intermedius, and some Nitrobacter strains are facultative chemo-
lithotrophs. All the others are obligate chemolithotrophs. Reasons for
obligate chemolithotrophy could be an incomplete tricarboxylic acid cycle,
respiratory chains that cannot accept electrons from NADH, or a shortage
of ammonia or reduced sulfur compounds for biosyntheses.
5. Chemolithotrophs use the Calvin cycle for CO 2 fixation. The key
enzymes of this cycle are ribulose-1,5-bisphosphate carboxylase and phos-
phoribulokinase. The reduction of 3C0 2 to glyceraldehyde-3-phosphate
requires 6NADH and 9ATP.
6. In the absence of CO 2 , ribulose-1 ,5-bisphosphate carboxylase cataly-
zes an oxygenolytic cleavage of ribulose-1,5-bisphosphate into phospho-
glycolate and phosphoglycerate.
7. Hydrogen donors used by the four families of phototrophic bacteria
are reduced sulfur compounds, molecular hydrogen, and organic com-
pounds. Phototrophic bacteria contain various bacteriochlorophylls and
carotenoids.
8. The photosynthetic apparatus of the Rhodospirillaceae and the
Chromatiaceae is organized in membrane invaginations. The light energy
transferred to the reaction center P870 gives rise to a cyclic electron
transfer that results in the generation of a !:J.P. NAO+ is reduced via
reverse electron transfer reactions.
9. The photosynthetic apparatus of the Chlorobiaceae and the Chloro-
flexaceae is organized in vesicles (light harvesting centers) and in the
cytoplasmic membrane (reaction centers). Primary acceptor of the electron
from P840 is an iron-sulfur protein, from which an electron can be fed into
a cyclic flow or can be used for NAD+ reduction.
10. Cyanobacteria possess two photosystems and carry out an oxygenic
photosynthesis like the green plants.
11. Purple bacteria, Chloroflexaceae, and cyanobacteria use the Calvin
cycle for CO 2 fixation; Chlorobiaceae use the reductive tricarboxylic acid
cycle.
12. Halobacteria form a purple membrane when grown at low oxygen
tensions. This membrane contains bacteriorhodo.ps'ln, which pumps protons
outward in a light-powered reaction. The proto~ gradient established can
be taken advantage of for ATP synthesis.
Chapter 10
Fixation of Molecular
Nitrogen
That microorganisms present in soil are able to assimilate molecular

nitrogen was recognized many years ago. The nitrogen gain determined by
the Kjeldahl method during growth in a medium free of fixed nitrogen was
usually taken as a measure of nitrogen fixation activity. In 1943 Burris and
Wilson introduced the 15N technique, which allowed the direct demonstra-
tion of fixed nitrogen and ammonia formation from N 2 With 15N it was
also possible to detect the Nrfixing enzyme system-the nitrogenase-in
cell extracts. In recent years a very convenient technique was introduced
for the determination of nitrogenase activity, acetylene reduction. Ni-
trogenase reduces acetylene to ethylene, and the latter can be easily
quantitated by gas chromatography.
N==N + 6H 2NH)
\ !
---
nitrogenase
/ \,
CH=CH + 2H CHz=CH z
This assay has contributed much to the recent progess in understanding
the distribution and the mechanism of nitrogen fixation.
I. Nitrogen-fixing Organisms
Nitrogen fixation is a property found only among prokaryotic organisms.
Bacterial species belonging to practically all known orders and families are
able to fix molecular nitrogen. It is customary to distinguish between
free-living and symbiotic N2 -fixing organisms. The latter exist in part-
nerships with plants, e.g., all members of the genus Rhizobium with
Biochemistry of Nitrogen Fixation 319
Table 10.1. Examples of nitrogen-

fixing organisms
Cyanobacteria
Anabaena azollae
Gloeocapsa species
Phototrophic bacteria
Rhodospirillum rubrum
Rhodopseudomonas capsulata
Chromatium vinosum
Strict anaerobes
Clostridium pasteurianum
Desulfovibrio vulgaris
Obligate and facultative aerobes
Rhizobium japonicum
Frankia alni
Klebsiella pneumoniae
Azotobacter vinelandii
Bacillus polymyxa
Mycobacterium fiavum
Beijerinckia indica
Azospirillum lipoferum
legumes, Anabaena azollae with the aquatic fern Azolla, Azospirillum

lipoferum with tropical grasses, and all members of the genus Frankia
with pioneer trees such as alder. Some Nz-fixing organisms are listed in
Table 10.1.
II. Biochemistry of Nitrogen Fixation

In 1960 Carnahan and collaborators announced the first successful reduc-
tion of N z to ammonia by a cell extract of Clostridium pasteurianum. The
enzyme system which catalyzes this reaction is called nitrogenase. In the
mean time the nitrogenase of a number of Nrfixing organisms has been
studied.
A. Composition of nitrogenase
Nitrogenase consists of two proteins in a ratio of 2: 1: azoferredoxin and
molybdoferredoxin. The properties of these proteins are listed in Table
10.2. Both proteins contain iron-sulfur centers. In addition, molybdoferre-
doxin contains two centers of the general composition Mo-8Fe-6S.
320 10: Fixation of Molecular Nitrogen
Table 10.2. Properties of the components of the nitrogenase

system
property azoferredoxin molybdoferredoxin
molecular weight 70,000 230,000

number of subunits 2 4a
iron atoms 8 b
28
molybdenum atoms o 2
acid-labile sulfide 8 28
aTwo of each of two subunit types, molecular weight 55,000 and

60,000, respectively.
bOne 4Fe-4S center per subunit.
B. The nitrogenase reaction

The formation of ammonia from N2 and H 2 is an exothermic reaction:
6.H = -91.2 kJ (-21.8 kcal)
However, N2 is an extremely stable molecule. It contains a triple bond
and the activation energy required is very high. This is why the chemical
synthesis of ammonia from nitrogen and hydrogen (Haber-Bosch process)
has to be carried out at high temperatures and high pressures. Nitrogenase
ATP
ferredoxin'H 2 -t---+...
ferredoxin ~----l""'-lCi";"""
ferredoxin' H 2
ferredoxin
ADP + Pi
Figure 10.1. N z reduction to ammonia as catalyzed by nitrogenase. The small
circles symbolize azoferredoxin and the large one molybdoferredoxin. H* Mo
symbolizes super-reduced molybdoferredoxin.
works at room temperature. Besides nitrogen gas it requires reduced

ferredoxin or flavodoxin and ATP as substrates (Fig. 10.1).
Reduced ferredoxin or flavodoxin transfers electrons to the azoferredo-
xin. At the expense of the energy of ATP hydrolysis the potential of redox
groups of the enzyme is lowered further, and finally a super-reduced
molybdoferredoxin is formed, which binds N z and reduces it stepwise to
ammonia. The binding of the dinitrogen molecule occurs probably by
insertion into a metal-hydride bond involving molybdenum. A Mo =
N-NH z group functions probably as an intermediate.
Only the azoferredoxin component of nitrogenase has ATP-binding
sites, and ATP hydrolysis is primarily associated with the formation of the
super-reductant. Using cell-free nitrogenase preparations the reduction of
N z has been found to be coupled to the hydrolysis of an enormous amount
of ATP. The figure is in the order of 16 ATP per N z:
N z + 6H + 16ATP - - 2NH 3 + 16ADP + 16P j
C. Sources of reducing power

It has been mentioned that the reducing power is supplied to the
nitrogenase in the form of reduced ferredoxin or flavodoxin. The former
functions as reductant in strict anaerobes and phototrophs. Its reduced
form is generated by the pyruvate-ferredoxin oxidoreductase or hydrog-
enase reactions (strict anaerobes), by noncyclic photophosphorylation
(cyanobacteria, green bacteria), or by reverse electron transport (purple
bacteria). Aerobic Nz-fixing organisms contain ferredoxin and/or flavodo-
xin, and these carriers are reduced with NADH or NADPH as H-donors.
Reverse electron transport might also be involved.
Nitrogenase catalyzes an ATP-dependent evolution of molecular hy-
drogen as some sort of a side reaction.
SATP + 2e- + 2H+ nitrogenas\ Hz + SADP + SP j
Here protons instead of molecular nitrogen serve as electron acceptor.

Hz production by phototrophic bacteria on the basis of this reaction has
already been discussed (Chapter 9). It also takes place in Rhizobia and
other Nz-fixing bacteria. Several of these species contain a special hy-
drogenase that enables the organism to take up the lost Hz again.
D. Oxygen sensitivity
A common property of all nitrogenase preparations is their senSItivity
toward oxygen. The enzyme is irreversibly inactivated by oxygen so that
nitrogen fixation can be regarded as a strictly anaerobic process. This is not
surprising if one considers that a strong reductant is required for N z
fixation. To keep the nitrogenase system anaerobic is no problem for strict
anaerobes and for phototrophs. However, it is a problem for aerobes and

for cyanobacteria.
Aerobes such as Azotobacter vinelandii and Mycobacterium fiavum fix
nitrogen better at oxygen tensions lower than in air. For protection of
nitrogenase against oxygen damage, aerobes such as the Azotobacter
species employ two mechanisms.
1. Respiratory protection. Azotobacter species possess a very active bran-
ched respiratory system. One branch of the electron transport chain is
coupled to three phosphorylation sites; the other two branches are
coupled to only one phosphorylation site. Thus with increasing oxygen
concentrations the rate of respiration can be increased in these organ-
isms by a partial uncoupling. This is a waste of NADH but it does
protect the nitrogenase against oxygen damage.
2. Conformational protection. Following a sudden increase of the oxygen
concentration, the nitrogenase of Azotobacter chroococcum is switched
off, as shown in Fig. 10.2. Nitrogenase activity appears again after
lowering of the oxygen concentration. The enzyme is apparently
protected by a conformational change and by the association with
protective proteins.
In most filamentous cyanobacteria the fixation of nitrogen takes place in
heterocysts. These cells are formed when cultures are deprived of a
nitrogen source other than Nz. Heterocysts are larger than vegetative cells
and are surrounded by a rather thick cell wall; they are the cellular sites of
nitrogen fixation. The heterocysts are devoid of photosystem II and,
600
>.
:~
.~
'"~ 400
~
0-
""
2
c
200
o 10 20 30 40 50
timt' (min)
Figure 10.2. "Switch off" and switch on" of nitrogenase. A culture of A.

chroococcurn was shaken gently under argon + oxygen (9: 1). At A, the shaking
rate was increased and at B, returned to original value. [So Hill, J. W. Drozd, and 1.
R. Postgate. 1. Appl. Chern. Biotechnol. 22, 541-548 (1972).]
vegetative cell
light
light CO 2
Figure 10.3. Nitrogen fixation in heterocysts. Adjacent vegetative cells provide
metabolites for generation of NADPH. ATP is synthesized in the heterocysts by
photophosphorylation. PSI, photosystem 1.
therefore, cannot produce oxygen. Thus, an interference of oxygenic

photosynthesis with the strictly anaerobic nitrogen fixation is not possible.
As illustrated in Fig. 10.3 heterocysts gain ATP by photophosphorylation.
Reducing power for N2 fixation is produced from metabolites provided by
the adjacent vegetative cells, which in turn receive ammonia from the
heterocysts.
This very evident protective mechanism is not found in all filamentous
cyanobacteria. Some of them (e.g., Plectonema strains) do not form
heterocysts. These species and the unicellular N 2 fixers (e.g., Gloeocapsa
species) fix nitrogen only at low partial pressures of oxygen.
E. Symbiotic nitrogen fixation

Nitrogen fixation by the legume-Rhizobium symbiosis is of considerable
agricultural importance. It takes place in the root nodules, which develop
following the infection of root hairs by a Rhizobium species. There is a
marked specificity with respect to the legume species infected; a certain
Rhizobium strain will invade the roots of some legumes but not of others.
After rhizobial infection tetraploid cells of the root are stimulated to
divide, and nodules are formed in which the bacteria multiply rapidly.
Finally, the bacteria are transformed into swollen, irregular cells called
bacteroids. These bacteroids are the sites of nitrogen fixation. The path of
nitrogen fixation is illustrated in Fig. 10.4. N2 is reduced by the bacteroids
to NH 3 , which is trapped by glutamine synthetase present in the plant
cytosol. The amido group is then transferred to aspartate, and the product
asparagine is removed by the root transport system. The amido group can
also be transferred to a-oxoglutarate to form glutamate. In addition to
asparagine the compounds glutamine and glutamate serve as important
nitrogen sources to the plant. a-Oxoglutarate is generated from sucrose via
the tricarboxylic acid cycle.
t
I
f
glutamate yaspargine--- - - - - - - - / : t
glutamine A aspartate - - - - - - - -./ J
sucrose
/----------------/ I
NH I tricarboxylic J
If
3
I acid cycle
I I
R
glutamine <>-oxoglutarate _ - ./
I glutamate glutamate }
bacteroid
root nodule
plant cytosol
/
root transport
system
Figure 10.4. The pathway of nitrogen from the atmosphere into the transport
stream of the plant. 1, Glutamine synthetase; 2, glutamine: aspartate amido-
transferase; 3, glutamine: a-oxoglutarate amidotransferase. [B. J. Miflin and P. J.
Lea; Trends Biochem. Sci. I, 103-106 (1976).]
The plant not only provides the bacteroids with the necessary organic
substrates but also shelters the nitrogenase from oxygen; N 2 -fixing nodules
are reddish in color because of the presence of leghemoglobin. This
hemoglobin-like compound serves as oxygen carrier controlling access of
oxygen to the bacteroids. The rhizobia are aerobes, and oxygen is
transported to the bacteroids by leghemoglobin in amounts sufficient for
ATP generation in the respiratory chain but low enough not to destroy the
nitrogenase. Some Rhizobium species also fix nitrogen when grown outside
their host plant in defined media.
III. Regu lation of Nitrogenase

Up to 16 mol of ATP are hydrolyzed per mol of nitrogen reduced, and it is
necessary for the cell to switch off nitrogen fixation under conditions of
ATP starvation. This is accomplished by ADP, which is an inhibitor of
nitrogenase. A 90% inhibition of N2 fixation has been observed, when
ATP and ADP are present in the cells in a 1 : 1 ratio. In free-living Nrfixing
organisms such as Azotobacter species but not in Rhizobia, a strong
inhibitory effect on nitrogenase activity is also exerted by ammonia.
Especially the synthesis of nitrogenase is under stringent control. The
enzyme is not synthesized in the presence of sufficient amounts of ammonia
Regulation of Nitrogenase 325
in the cells. In other words, the transcription of the nitrogenase (nit) genes
is derepressed only under conditions of ammonia limitation. This control is
achieved in a very complicated manner. Seventeen genes have to be
expressed for the synthesis of a functional nitrogenase in Klebsiella
pneumoniae. These genes are organized in several operons that are
adjacently located on the DNA and form the so-called nit cluster. The
function of some of these genes is given in Fig. 10.5. Genes H, M, and 5
code for azoferredoxin and genes D, K, E, N, V and B for molybdoferre-
doxin. The products of the genes A and L are involved in regulation. The
protein of the nitA gene (A) is absolutely essential for the transcription of
the other nit operons; it is an activator. The nitL product (L) is always
synthesized along with the nitA product. Only in the presence of ammonia
- 1-- nitrogen control

ntrA nlr C nlf B ginA
nif duster
nif B A L M V S NE
Figure 10.5. Organization and control of the nif cluster in K. pneumoniae. The
situation under the condition of derepression is shown. The products of
nlrA + nlrC, designated as nA and nC, together activate the promotor of ginA and
of nifA-nifL. The nifA product (A) then activates the remaining promotors of the
nif cluster. pro-Fe-P, Polypeptide of azoferredoxin; incorporation of the iron-sulfur
centers is controlled by nifM and nifS products; pro-MoFe-P, precursor of
molybdoferredoxin; the iron-molybdenum cofactor results from the products of
genes nifB, nifN, nifV, and nifE. In the presence of ammonia product nB prevents
activation at the nitrogen control level and product L at the nif cluster level. Solid
arrows indicate the promotors at which transcription is de repressed by the gene
products nA + nC. Dotted arrows indicate the promotors at which transcription is
de repressed by the gene product A.
does it prevent the activator function of the nitA gene product and exhibits
the properties of a repressor. This picture becomes now more complicated
because the whole nit cluster is under so-called nitrogen control.
The term nitrogen control stands for a superimposed regulatory
mechanism by which the synthesis of a number of enzymes yielding
ammonia remains repressed as long as ammonia is available to the cells.
Enzymes of this sort in K. pneumoniae are histidase, urease, and
nitrogenase. In addition, glutamine synthetase is under nitrogen control.
All these enzymes are formed in only small amounts at high intracellular
NH: levels, but in large amounts if the NH: concentration is low. It was
assumed for some time that glutamine synthetase itself exhibits an activa-
tor function. Now it is clear through the work of Kustu, Magasanik,
Merrick, and co-workers that three different proteins are responsible for
nitrogen control: the products of the genes ntrA, ntrB, and ntrc. The latter
two are linked to the structural gene for glutamine synthetase, ginA. The
gene ntrA is unlinked (see Fig. 10.5). The ntrC product (ne) is the general
activator of all the operons that are under nitrogen control, but to do so it
requires the ntrA product (nA). The ntrB product (nB) inactivates the ntrC
protein in the presence of ammonia.
It is remarkable that cells make an enormous effort to regulate the
synthesis of enzymes involved in nitrogen metabolism. It may reflect the
fluctuations in nature of the type of nitrogen source that is available to the
microorganisms.
IV. Summary
1. Nitrogen fixation is found only among prokaryotic organisms. It
occurs in strictly anaerobic, aerobic, and phototrophic organisms.
2. The enzyme that catalyzes the reduction of molecular nitrogen to
ammonia is called nitrogenase. It consists of two proteins, azoferredoxin
and molybdoferredoxin. For Nz reduction, ATP and reduced ferredoxin
are required as energy and hydrogen sources.
3. Nitrogenase also catalyzes the reduction of acetylene to ethylene.
This reaction is used as a convenient assay of nitrogenase activity.
4. Nitrogenase is very oxygen-sensitive. Respiratory and conforma-
tional protection and heterocyst formation are used as mechanisms to
prevent inactivation by oxygen. In symbiotic nitrogen fixation, the trans-
port of oxygen to the Nz-fixing bacteroids is controlled by leghemoglobin.
5. The nitrogenase activity is inhibited by ADP. Nitrogenase is only
synthesized if the ammonia concentration in the cells is low. Under these
conditions the promotor of the genes nitA and nitL is activated by
products of the nitrogen control genes. The protein of the nitA gene
subsequently activates the transcription of all other genes of the nit cluster.
Further Reading
It is recommended that standard textbooks of biochemistry, microbiology and

genetics be consulted. The small selection of review articles and books given below
may be useful for further study.
Chapter 1
Guirard, B. M., Snell. E. E.: Nutritional requirements of microorganisms. In: The
BaCTeria, vol. 4, I. C. Gunsalus and R. Y. Stanier (eds.). Academic Press, New
York, 1962, pp. 33-95.
Hutner, S. H.: Inorganic nutrition. Ann. Rev. Microbial. 26,313-346 (1972).
Harder, W., Dijkhuizen, L.: Physiological responses to nutrient limitation. Ann.
Rev. Microbial. 37, 1-23 (1983).
Chapter 2
Kornberg, H. L.: The nature and control of carbohydrate uptake by Escherichia
coli. FEBS Lett. 63,3-9 (1976).
Krampitz, L. 0.: Cyclic mechanisms of terminal oxidation. In: The bacteria, vol. 2,
I. C. Gunsalus and R. Y. Stanier (eds.). Academic Press, New York, 1962,
pp. 209-256.
Ingeiedew, W. J., Poole, R. K.: The respiratory chains of Escherichia coli.
Microbial. Rev. 48, 181-198 (1984).
Harold, F. M.: Vectorial metabolism. In: The BaCTeria, vol. 6, L. N. Omston and
J. R. Sokatch (eds.). Academic Press, New York, San Francisco, London, 1978,
pp. 463-521.
Ferguson, S. 1., Sorgato, M. c.: Proton electrochemical gradients and energy-
transduction processes. Ann Rev. Biochem. SI, 185-217 (1982).
328 Further Reading
Wikstrom, M., Krab, K., Saraste, M.: Proton translocating cytochrome complexes.
Ann. Rev. Biochem. 50, 623-655 (1981).
Amzel, L. M., Pedersen, P. L.: Proton ATPases: Structure and mechanism. Ann.
Rev. Biochem. 52, 801-824 (1983).
Futai, M., Kanazawa, H.: Structure and function of protontranslocating adenosine
triphosphatase (FoF 1): Biochemical and molecular biological approaches. Mi-
crobial. Rev. 47,285-312 (1983).
Schneider, K., Altendorf, K. H.: The protontranslocating portion (Fo) of the E.
coli ATP synthase. Trends Biochem. Sci. 9,51-53 (1984).
Racker, E.: From Pasteur to Mitchell; a hundred years of bioenergetics. Fed. Proc.
39,210-215 (1980).
Fridovich, I.: Superoxide dismutases. Adv. Enzymol. 41, 35-97 (1974).
Packer, L. (ed.): Oxygen radicals in biology systems. Methods in Enzymol. 105,
Sections I and II (1984).
Chapter 3
Dagley, S., Nicholson, D. E.: An Introduction to Metabolic Pathways. Blackwell
Scientific Publications, Oxford and Edinburgh, 1970.
Dalton, H.: Utilization of inorganic nitrogen by microbial cells. In: International
Review of Biochemistry, vol. 21, J. R. Quayle (ed.). 227-266, 1979.
Herrmann, K. M., Sommerville, R. L.: Amino Acids: Biosynthesis and Genetic
Regulation. Addison-Wesley Publishing Company, Reading, Mass., 1983.
Cooper, A. J. L.: Biochemistry of sulfur-containing amino acids. Ann. Rev.
Biochem 52, 187-222 (1983).
Wakil, S. J., Stoops, J. K., Joshi, V. c.: Fatty acid synthesis and its regulation.
Ann. Rev. Biochem. 52, 537-579 (1983).
Preiss, J.: Bacterial glycogen synthesis and its regulation. Ann. Rev. Microbial. 38,
419-458 (1984).
Raetz, C. R. H.: Enzymology, genetics, and regulation of membrane phospholipid
synthesis in Escherichia coli: Microbiol. Rev. 42, 614-659 (1978).
Nikaido, H., Nakae, T.: The outer membrane of gram-negative bacteria. Adv.
Microbial. Physiol. 20, 163-250 (1979).
Ingraham, J. L., Maale, 0., Neidhardt, F. c.: Growth of the Bacterial Cell.
Sinauer Assoc., Inc., Sunderland, Mass., 1983.
Birge, E. A.: Bacterial and Bacteriophage Genetics. Springer-Verlag, New York,
Heidelberg, Berlin, 1981.
Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., Watson, J. D.: Molecular
Biology of the Cell. Garland Publishing, Inc., New York and London, 1983.
Nossal, N. G.: Prokaryotic DNA replication systems. Ann. Rev. Biochem. 52,
581-615 (1983).
Ogawa, T., Okazaki, T.: Discontinuous DNA replication. Ann. Biochem. 49,
421-457 (1980).
v. Hippel, P. H., Bear, D. G., Morgan, W. D., McSwiggen, J. A.: Protein-nucleic
acid interactions in transcription: A molecular analysis. Ann. Rev. Biochem. 53,
389-446 (1984).
Geider, K., Hoffmann-Berling, H.: Proteins controlling the helical structure of
DNA. Ann. Rev. Biochem. 50, 233-260 (1981).
Further Reading 329
Drlica, K.: Biology of bacterial deoxyribonucleic acid topoisomerases. Microbiol.

Rev. 48, 273-289 (1984).
Wittmann, H. G.: Components of bacterial ribosomes. Ann. Rev. Biochem. 51,
155-183 (1982).
Wittmann, H. G.: Architecture of prokaryotic ribosomes. Ann. Rev. Biochem. 52,
35-65 (1983).
Maitra, U., Stringer, E. A., Chaudhuri, A.: Initiation factors in protein biosyn-
thesis. Ann. Rev. Biochem. 51, 869-900 (1982).
Hunt, T.: The intitiation of protein synthesis. Trends Biochem. Sci. 5, 178-181
(1980).
Kozak, M.: Comparison of initiation of protein synthesis in prokaryotes, eukary-
otes and organelles. Microbiol. Rev. 47, 1-45 (1983).
Clark, B.: The elongation step of protein biosynthesis. Trends Biochem. Sci. 5,
207-210 (1980).
Caskey, C. T.: Peptide chain termination. TrendsBiochem. Sci. 5,234-237 (1980).
Rich, A., Kim, S. H.: The three-dimensional structure of transfer RNA. Sci. Ann.
238, 52-62 (1978).
Schimmel, P. R., Soil, D.: Aminoacyl-tRNA synthetases: General features and
recognition of transfer RNAs. Ann. Rev. Biochem. 48, 601-648 (1979).
Chapter 4
Doelle, H. W.: Bacterial Metabolism. Academic Press, New York, San Francisco,
and London, 1975. Chapter 5 is on carbohydrate metabolism.
Kornberg, H. L.: Anaplerotic sequences and their role in metabolism. In: Essays in
Biochemistry, vol. 2, P. N. Campbell and G. D. Greville (eds.). Academic
Press, London and New York, 1966, pp. 1-32.
Chapter 5
Dills, S. S., Apperson, A., Schmidt, M. R., Saier, Jr., M. H.: Carbohydrate
transport in bacteria. Microbiol. Rev. 44, 385-418, 1980.
Reizer, J., Deutscher, J., Sutrina, S., Thompson, J., Saier, Jr., M. H.: Sugar
accumulation in Gram-positive bacteria: Exclusion and expulsion mechanisms.
Trends Biochem. Sci. 10, 32-35 (1985).
Overath. P., Wright, J. K.: Lactose permease: a carrier on the move. Trends
Biochem. Sci. 8, 404-408 (1983).
Konings, W. N., Michels, P. A. M.: Electron-transfer-driven solute translocation
across bacterial membranes. In: Diversity of Bacterial Respiration, vol. I. C. J.
Knowles (ed.). CRC Press, Inc., Boca Raton, Florida, 1980, pp. 33-86.
Postma, P. W., Roseman, S.: The bacterial phosphoenolpyruvate: sugar phospho-
transferase system. Biochim. Biophys. Acta 457, 213-257 (1976).
Gibson, J.: Nutrient transport by anoxygenic and oxygenic photosynthetic bacteria.
Ann. Rev. Microbiol. 38, 135-159 (1984).
Nikaido, H., Vaara, M.: Molecular basis of bacterial outer membrane permeabil-
ity. Microbiol. Rev. 49, 1-32 (1985).
Neilands, J. B.: Microbial envelope proteins related to iron. Ann. Rev. Microbiol.
36, 285-309 (1982).
330 Further Reading
Braun, Y.: The unusual features of the iron transport systems of Escherichia coli.
Epstein, W., Laimins, L.: Potassium transport in Escherichia coli: Diverse systems
with common control by osmotic forces. Trends Biochem. Sci. 5,21-23 (1980).
Randall, L. L., Hardy, S. J. S.: Export of protein in bacteria. Microbiol. Rev. 48,
290-298 (1984).
Pugsley, A. P., Schwartz, M.: Export and secretion of proteins by bacteria. FEMS
Microbiol. Rev. 32, 3-38 (1985).
Taylor, B. L.: Role of proton motive force in sensory transduction in bacteria.
Ann. Rev. Microbiol. 37,551-573 (1983).
Taylor, B. L.: How do bacteria find the optimal concentration of oxygen? Trends
Biochem. Sci. 8, 438-441 (1985).
Fraenkel, D. G., Yinopal, R. T.: Carbohydrate metabolism in bacteria. Ann. Rev.
Microbiol. 27,69-100 (1973).
Mortlock, R. P.: Catabolism of unnatural carbohydrates by microorganisms. Adv.
Microbiol. Physiol. 13, 1-53 (1976).
Lessie, T. G.: Alternative pathways of carbohydrate utilization in Pseudomonas.
Ann. Rev. Microbiol. 38, 359-387 (1984).
Cooper, R. A.: Metabolism of methylglyoxal in microorganisms. Ann. Rev.
Microbiol. 38, 49-68 (1984).
Stouthamer, A. H.: Energy-yielding pathways. In: The Bacteria, vol. 6, L. N.
amston and J. R. Sokatch (eds.). Academic Press, New York, San Francisco,
London, 1978, pp. 389-462.
Haddock, B. A., Jones, C. W.: Bacterial respiration. Bacteriol. Rev. 41,47-99
(1977).
Padan, E., Zilberstein, D., Schuldiner, S.: pH-homeostasis in bacteria. Biochim.
Biophys. Acta 650, 151-166 (1981).
Knowles, R.: Denitrification. Microbiol. Rev. 46, 43-70 (1982).
Coughlan, M. (ed.): Molybdenum and Molybdenum-Containing Enzymes. Perga-
mon Press, Oxford, 1980.
Nealson, K. H., Hastings, J. W.: Bacterial bioluminescence: Its control and
ecological significance. Microbiol. Rev. 43, 496-518 (1979).
Dawes, E. A., Senior, P. J.: The role and regulation of energy reserve polymers in
microorganisms. Adv. Microbial Physiol. 10, 136-297 (1973).
Sutherland, J. W.: Biosynthesis of microbial exopolysaccharides. Adv. Microbial
Physiol.. 23, 79-150 (1982).
Schleifer, K. H., Kandler, 0.: Peptidoglycan types of bacterial cell walls and their
taxonomic implications. Bacterial. Rev. 46, 407-477 (1972).
Rogers, H. J. Biogenesis of the wall in bacterial morphogenesis. Adv. Microbial
Physiol. 19, 1-62 (1979).
Ward, J. B.: Teichoic and teichuronic acids: Biosynthesis, assembly, and location.
Microbiol Rev. 45, 211-243 (1981).
Shockman, G. D., Barrett, J. F.: Structure, function, and assembly of cell walls of
Gram-positive bacteria. Ann. Rev. Microbiol. 37, 501-527 (1983).
Utter, M. F., Barden, R. E., Taylor, B. L.: Pyruvate carboxylase. Adv. Enzymol.
42, 1-72 (1975).
Wood, H. G., Utter, M. F.: The role of CO 2 fixation in metabolism. In: Essays in
Biochemistry, vol. 1, P. N. Campbell and G. D. Greville (eds.). Academic
Press, London and New York, 1965, pp. 1-28.
Further Reading 331
Chapter 6
Takagi. T .. Toda. H .. Isemura. T.: Bacterial and mold amylases. In: The Enzymes,
vol. 5. P. D. Boyer (ed.). Academic Press, New York and London, 1971,
pp.235-271.
French. D.: Amylases: Enzymatic mechanisms. In: Trends in Ihe Biology of
Fermentalions. A. Hollaender (ed.). Plenum Press, New York and London.
1981, pp. 151-180.
Ueda. S.. Saha. B. C, Koba. Y.: Direct hydrolysis of raw starch. Microbiol. Sci. 1,
21-24 (1984).
Whitaker. D. R.: Cellulases. In: The Enzymes. vol. 5. P. D. Boyer (ed.).
Academic Press. New York and London. 1971, pp. 273-289.
Cuskey. S. M., Frein. E. M .. Montenecourt. B. S., Eveleigh. D. E.: Overproduc-
tion of cellulase-screening and selection. In: Overproduclion of Microbial
Producls. V. Krumphanzl. B. Sikyta. and Z. Vanek (eds.). Academic Press.
London and New York. 1982. pp. 405-416.
Eveleigh. D. E .. Montenecourt. B. S.: Increasing yields of extracellular enzymes.
Adv. Appl. Microbiol. 25. 57-74 (1979).
Bellamy. W. D.: Role of thermophiles in cellulose recycling. ASM News 45,
326-331 (1979).
Burns. R. G.: Extracellular enzyme-substrate interactions in soil. In: Microbes in
Iheir Nalural Environmel1ls. J. H. Slater. R. Whittenbury and J. W. T.
Wimpenny (eds.). p. 34. Symp. Soc. Cell. Microbiol. University Press. Cam-
bridge. 1983. pp. 249-298.
Sokatch. J. R.: Baclerial Physiology and Melabolism. Academic Press. London and
New York, 1969. Chapter II is on protein and amino acid catabolism.
Massey. L. K.. Sokatch. J. R .. Conrad. R. S.: Branched-chain amino acid
catabolism in bacteria. BaCieriol. Rev. 40. 42-54 (1976).
Ahdelal. A. T.: Arginine catabolism by microorganisms. Ann. Rev. Microbiol. 33.
139-168 (1979).
Vogels. G. D .. Van der Drift. C: Degradation of purines and pyrimidines by
microorganisms. BaCleriol. Rev. 40. 403-468 (1976).
Dagley. S.: Pathways for the utilization of organic growth substrates. In: The
BaCieria. vol. 6. L. N. Ornston and J. R. Sokatch (eds.). Academic Press. New
York. San Francisco. London. 1978. pp. 305-388.
Klug, M. J., Markovetz. A. J.: Utilization of aliphatic hydrocarbons by microor-
ganisms. Adv. Microbiol. P!zysiol. 5. 1-39 (1971).
Wyatt. J. M.: The microbial degradation of hydrocarbons. Trends Biochem. Sci. 9.
20-23 (1984).
Perry. J. J.: Microbial cooxidations involving hydrocarbons. Microbiol. Rev. 43.
59-72 (1979).
Atlas. R. M.: Microbial degradation of petroleum hydrocarbons: An environmen-
tal perspective. Microbiol. Rev. 45, 180-209 (1981).
Higgins. I. 1.. Best. D. 1.. Hammond, R. C. Scott. D. Methane-oxidizing
microorganisms. Microhiol. Rev. 45.556-590 (1981).
Crawford. R. L.. Hanson. R. S. (eds.): Microbial growth on C, compounds. Amer.
Soc. Microbiol.. Washington. D. C (1984).
Large. P.: Mnhylolrop!zy and Mel!zanogenesis. Van Nostrand Reinhold, Woking-
ham (1983).
332 Further Reading
Stanier, R. Y., amston, L. N.: The ,l3-ketodipate pathway. Adv. Microbial Physiol.
9, 89-149 (1973).
Hardy, K.: Bacterial Plasmids. Thomas Nelson and Sons, Ltd, Walton-on-Thames,
Surrey (1981).
Williams, P. A.: Catabolic plasmids. Trends Biochem. Sci. 6, 23-26 (1981).
Alexander, M.: Biodegradation of chemicals of environmental concern. Science
211, 132-138 (1981).
Chapter 7
Miller, J. H., Reznikoff, W. S. (eds.). The Operon. Cold Spring Harbor Labora-
tory, New York (1978).
Ullmann, A., Danchin, A.: Role of cyclic AMP in regulatory mechanisms in
bacteria. Trends Biochem. Sci. 5, 95-96 (1980).
Botsford, J. L.: Cyclic nucleotides in prokaryotes. Microbiol. Rev. 45, 620-642
(1981).
Watson, M. D. Attenuation: translational control of transcription termination.
Cohen, G.: The Regulation of Cell Metabolism. Hermann, Paris, (1968).
Herrmann, K. M., Somerville, R. L.: Amino acids: Biosynthesis and Genetic
Regulation. Addison-Wesley Publishing Comp., Reading, Mass., 1983.
amston, L. N.: Regulation of catabolic pathways in pseudomonas. Bacteriol. Rev.
35, 87-116 (1971).
Cohen, P.: Control of Enzyme Activiry. Chapman and Hall, London and New York
(1983).
Koshland, Jr., D. E.: Conformational aspects of enzyme regulation. Curro Top.
Cell. Regul. 1, 1-28 (1969).
Wyman, J.: On allosteric models. Curro Top. Cell Regul. 6, 209-226 (1972).
Sanwal. B. D.: Allosteric controls of amphibolic pathways in bacteria. Bacreriol.
Rev. 34, 20-39 (1970).
Kantrowitz, E. R., Pastra-Landis, S. c., Lipscomb, W. N.: E. coli aspartate
transcarbamylase. Trends Biochem. Sci. 5, 124-128 (1980).
Weitzman, P. D. J.: Unity and diversity in some bacterial citric acid-cycle enzymes.
Adv. Microbial Physiol. 22, 186-244 (1981).
Stadtman, E. R.: Mechanisms of enzyme regulation in metabolism. In: The
Enzymes, vol. 1, P. D. Boyer (ed.). Academic Press, New York and London,
1970, pp. 398-459.
Cozzone, A. J.: Protein phosphorylation in bacteria. Trends Biochem. Sci. 9,
400-403 (1984).
Switzer, R. L.: The inactivation of microbial enzymes in vivo. Ann. Rev.
Microbial. 31, 135-157 (1977).
Chapter 8
Wood, W. A.: Fermentation of carbohydrates and related compounds. In: The
Bacreria, vol. 2, I. C. Gunsalus and R. Y. Stanier (eds.). Academic Press, New
York, 1961, pp. 59-149.
Thauer, R. K., Jungermann, K., Decker, K.: Energy conservation in chemotrophic
anaerobic bacteria. Bacterial. Rev. 41. 100-180 (1977).
Further Reading 333
Gottschalk, G., Andreesen, J. R.: Energy metabolism in anaerobes. In: Interna-

tional Review of Biochemistry, vol. 21, J. R. Quayle (ed.), 1979, pp. 5-115.
Zeikus, I. G.: Chemical and fuel production by anaerobic bacteria. Ann. Rev.
Microbial. 34, 423-464 (1980).
Thauer, R. K., Morris, J. G.: Metabolism of chemotrophic anaerobes: Old views
and new aspects. In: The Microbe 1984, D. P. Kelly and N. G. Carr (eds.),
p. 36. Symp. Soc. Gen. Microbiol., 1984, pp. 123-168 (1984). Cambridge
University Press, Cambridge.
Morris, J. G.: The physiology of obligate anaerobiosis. Adv. Microbial Physiol. 12,
169-233 (1975).
Krebs. H. A.: The Pasteur-effect and the relations between respiration and
fermentation. In: Essays in Biochemistry. vol. 8. P. N. Campbell and F. Dickens
(eds.). Academic Press. London and New York, 1972, pp. ]-26.
Ramaiah, A.: Pasteur effect and phosphofructokinase. Curro Top. Cell. Regul. 8,
298-344 (1974).
Wiegel, J.: Formation of ethanol by bacteria. A pledge for the use of extreme
thermophilic anaerobic bacteria in industrial ethanol fermentation processes.
Experientia 36. 1434-1446 (1980),
Rogers, P. L., Goodman. A. E .. Heyes, R. H.: Zymomonas ethanol fermenta-
tions. Microbiol. Sci. I. 133-136 (1984).
O'Neal Ingram, L., Buttke. T. M.: Effects of alcohols on microorganisms. Adv
Microbial Physiol. 25, 253-300 (1984).
Garvie, E. J.: Bacterial lactate dehydrogenases. Microbial. Rev. 44, 106-139
( 1980).
Srere, P. A.: The enzymology of the formation and breakdown of citrate. Adv.
Enzymol. 43, 57-102 (1975).
Gottschalk, G., Bahl. H.: Feasible improvements of the butanol fermentation by
Clostridium acelOburylicum. In: Trends in the Biology of Fermentations, A.
Hollaender (ed.). Plenum Press, New York and London, 1981, pp. 463-470.
Sweeney. W. Y., Rabinowitz. J. c.: Proteins containing 4Fe-4S clusters: An
overview. Ann. Rev. Biochem. 49, 139-161 (1980).
Yoch, D. c.. Carithers. R. P.: Bacterial iron-sulfur proteins. Microbiol. Rev. 43,
384-321 (1979).
Mayhew, S. G., OConnor. M. E.: Structure and mechanism of bacterial hy-
drogenase. Trends Biochem. Sci. 7. ]8-2] (1982).
Kroger. A.: Bacterial electron transport to fumarate. In: Diversity oj' Bacterial
Respiratory Systems, vol. II. C. J. Knowles (ed.). CRC Press, Inc., Boca Raton,
Florida. 1980, pp. 1-17.
Hilpert. W .. Schink, B.. Dimroth. P.: Life by a new decarboxylation-dependent
energy conservation mechanism with Na+ as coupling ion. The EMBO Journal
3. 1665-1670 (1984).
Wood. H. G.: Some reactions in which inorganic pyrophosphate replaces ATP and
serves as a source of energy. Fed. Proc. 36. 2197-2205 (]977).
Wood, H. Goo Drake. H. L.. Hu, S. I.: Studies with Clostridium thermoaceticum
and the resolution of the pathway used by acetogenic bacteria that grow on
carbon monoxide or carbon dioxide and hydrogen. Proc. Biochem. Symp.
pp. 29-56 (1982).
Zeikus. J. G.: Metabolism of one-carbon compounds by chemotrophic anaerobes.
Adv. Microbial Physiol. 24. 215-299 (1983).
334 Further Reading
Barker, H. A.: Coenzyme BIz-dependent mutases causing carbon chain rearrange-

ments. In: The Enzymes, vol. 6, P. D. Boyer (ed.). Academic Press, New York
and London, 1972, pp. 509-537.
Halpern, J.: Mechanisms of coenzyme BIz-dependent rearrangements. Science 227,
869-875 (1984).
Daniels, L., Sparling, R., Sprott, G. D.: The bioenergetics of methanogenesis.
Biochim. Biophys. Acta 768, 113-163 (1984).
Kandler, O. (ed.): Papers presented at the first International Workshop on
Archaebacteria. Zbl. Bakt. Hyg., 1. Abt. Org. C3 (1982).
Wolfe, R. S., Higgins, I. J.: Microbial biochemistry of methane-a study in
contrasts. In: International Review of Biochemistry, vol. 21, J. R. Quale (ed.),
1979, pp. 267-353.
Thauer, R. K., Badziong, W.: Respiration with sulfate as electron acceptor. In:
Diversity of Bacterial Respiration Systems, vol. II, C. J. Knowles (ed.). CRC
Press, Inc., Boca Raton, Florida, 1980, pp. 65-85.
Odom, J. M., Peck Jr., H. D.: Hydrogenase, electron-transfer proteins, and
energy coupling in the sulfate-reducing bacteria Desulfovibrio. Ann. Rev.
Microbial. 38, 551-592 (1984).
Pfennig, N.: Microbial behaviour in natural environments. In: The Microbe 1984,
D. P. Kelly and Carr, N. G. (eds.), p. 36. Symp. Soc. Gen. Microb. Cambridge
University Press, Cambridge, 1984, pp. 23-50.
Hungate, R. E.: The rumen microbial ecosystem. Ann. Rev. Ecol. System. 6,
39-66 (1975).
Wolin, M. J., and Miller, T. L.: Interspecies hydrogen transfer: 15 years later.
ASM News 48, 561-565 (1982).
Zinder, S. H.: Microbiology of anaerobic conversion of organic wastes to methane:
Recent developments. ASM News 50, 294-298 (1984).
Laanbroek, H. J., Veldkamp, H.: Microbial interactions in sediment communities.
Phil. Trans. R. Soc. Lond. 8297, 533-550 (1982).
Sleat, R., Robinson, J. P.: The bacteriology of anaerobic degradation of aromatic
compounds. 1. Appl. Bacterial. 57, 381-394 (1984).
Barker, H. A.: Amino acid degradation by anaerobic bacteria. Ann. Rev.
Biochem. 50, 23-40 (1981).
Barker, H. A.: Fermentation of nitrogenous organic compounds. In: The Bacteria,
vol. 2, 1. C. Gunsalus and R. Y. Stanier (eds.). Academic Press, New York,
1961, pp. 151-207.
Stadtman, T. c.: Selenium-dependent enzymes. Ann. Rev. Biochem. 49,93-110
(1980).
Recsei, P. A., Snell, E. E.: Pyruvoyl enzymes. Ann. Rev. Biochem. 53,357-387
(1984).
Chapter 9
Knowles, C. J. (ed.): Diversity of Bacterial Respiratory Systems, vol. II, CRC
Press, Inc., Boca Raton, Florida, 1980. Some articles on respiration of
chemolithotrophic bacteria.
Jones, C. W.: Bacterial respiration and photosynthesis. Thomas Nelson and Sons,
Ltd., Walton-on-Thames, Surrey (1982).
Bowien, B., Schlegel, H. G.: Physiology and Biochemistry of aerobic hydrogen-
oxidizing bacteria. Ann. Rev. Microbiol. 35, 405-452 (1981).
Further Reading 335
Meyer, 0., Schlegel, H. G.: Biology of aerobic carbon monoxide-oxidizing

bacteria. Ann. Rev. Microbiol. 37,277-310 (1983).
Schimel, J. P., Firestone, M. K., Killham, K. S.: Identification of heterotrophic
nitrification in a Sierran Forest Soil. Appl. Environ. Microbiol. 48, 802-806
(1984).
Cobley, J. G., Cox, J. c.: Energy conservation in acidophilic bacteria. Micro-
bioi. Rev. 47, 579-595 (1983).
Rittenberg, S. c.: The roles of exogenous organic matter in the physiology of
chemolithotrophic bacteria. Adv. Microbial Physiol. 3, 159-193 (1969).
Krulwich, T. A., Guffanti, A. A.: Physiology of acidophilic and alkalophilic
bacteria, Adv. Microbial Physiol. 24, 173-214 (1983).
Suzuki, I.: Mechanisms of inorganic oxidation and energy coupling. Ann. Rev.
Microbiol. 28, 85-102 (1984).
Lundgren, D. G., Silver, M.: Ore leaching by bacteria. Ann. Rev. Microbiol. 34,
263-283 (1980).
Miziorko, H. M., Lorimer, G. H.: Ribulose-1,5-bisphosphate carboxylase-
oxygenase. Ann. Rev. Biochem. 52, 507-535 (1983).
Codd, G. A., Marsden, W. J. N.: The carboxysomes (polyhedral bodies) of
autotrophic prokaryotes. Bioi. Rev. 59, 389-422 (1984).
Leadbeater, L., Bowlen, B.: Control of autotrophic carbon assimilation in
Alcaligenes eutrophus by inactivation and reactivation of phosphoribulokinase.
J. Bacteriol. 157, 95-99 (1984).
Pfennig, N.: Photosynthetic bacteria. Ann. Rev. Microbiol. 21,285-324 (1967).
Gest, H.: Energy conversion and generation of reducing power in bacterial
photosynthesis. Adv. Microbial Physiol. 7, 243-278 (1972).
Clayton, R. K., Sistrom, W. R. (eds.): The Photosynthetic Bacteria. Plenum Press,
New York, 1978.
Ormerod, J. G. (ed.): The Phototrophic Bacteria. Blackwell Scientific Publications,
Oxford, 1983.
Kondratieva, E. N.: Interrelation between modes of carbon assimilation and
energy production in phototrophic purple and green bacteria. In: International
Review of Biochemistry, vol. 21, J. R. Quayle (ed.), 117-175 (1979).
Drews, G.: Structure and functional organization of light-harvesting complexes and
photochemical reaction centers in membranes of phototrophic bacteria. Micro-
bioi. Rev. 49, 59-70 (1985).
Allen, M. M.: Cyanobacterial cell inclusions. Ann. Rev. Microbiol. 38, 1-25
(1984).
Nugent, I. H. A.: Photosynthetic electron transport in plants and bacteria. Trends
Biochem. Sci. 9, 354-357 (1984).
Stoeckenius, W., Bogomolni, R. A.: Bacteriorhodopsin and related pigments of
halobacteria. Ann. Rev. Biochem. 51, 587-616 (1982).
Packer, L. (ed.): Bacteriorhodopsin. Methods Enzymol. 88, part I (1982).
Skulachev, V. P.: Sodium bioenergetics. Trends Biochem. Sci. 9,483-485 (1984).
Chapter 10
Mortenson, L. E., Thorneley, R. N. F.: Structure and function of nitrogenase.
Ann. Rev. Biochem. 48, 387-418 (1979).
Shah, V. K., Ugalde, R. A., Imperial, J., Brill, W. J.: Molybdenum in ni-
trogenases. Ann. Rev. Biochem. 53, 231-257 (1984).
336 Further Reading
Robson, R. L., Postgate, J. R.: Oxygen and hydrogen in biological nitrogen

fixation. Ann. Rev. Microbial. 34, 183-207 (1980).
Magasanik, B.: Genetic control of nitrogen assimilation in bacteria. Ann. Rev.
Genet. 16, 135-168 (1982).
Ausubel, F. M.: Regulation of nitrogen fixation genes. Cell 37, 5-6 (1984).
Index of Organisms
Acetobacter 122, 169, 170 Bacillus 141, 142, 145, 149, 176, 195
aceti 169, 170 amyloliquefaciens 144
pasteurianus 169, 170 anthracis 127
nlinum 127, 128 cereus 116
Acetobacterium wieringae 250 coagulans 128
woodii 250, 281 fastidiosus 141
Acetogenium kivui 250 licheniformis 8, 123, 144, 147
Acidaminococcus fermentans 242, 273 megaterium 9, 109, 130, 131, 144,
Acinetobacter 141, 149, 155 149
calcoaceticus 128, 133, 154, 180 methanicus 162
Actinomyces israeli 135 polymyxa 144, 193, 315
Alcaligenes carboxydus 285 schlegelii 285
eutrophus 9, 116, 131, 181, 187, 196, subtilis 109, 115, 116, 129, 144, 145,
197, 283, 286-288, 294, 299 147, 154, 176, 185, 193
faecalis 123 Bacteroides 133, 249
Anabaena 133,309 amylophilus 247
azollae 319 fragilis 9, 247
Arthrobacter 109, 116, 122, 294 succinogenes 143
allantoicus 151 Beggiatoa 285
citreus 135 Beijerinckia indica 319
paraffineus 154 Bifidobacterium bifidum 215, 218, 220,
simplex, 154 222
Aspergillus niger 142, 144, 172-174 Brucella abortus 109, 119
Azospirillum lipoferum 319 Butyrivibrio fibrisolvens 143, 209, 225
Azotobacter 122, 141, 322
beijerinckii 131, 132
chroococcum 116, 322 Candida 168
vinelandii 109, 127, 130, 131,227, lipolytica 154
319,322 Cellulomonas 142
338 Index of Organisms
Cephalosporium roseum 154 Desulfobacter postgatei 261

Chlorobiaceae 7,300,301,311-313. Desulfobulbus propionicus 261, 264
317 Desulfococcus multivorans 261, 262
Chloroflexaceae 300, 30 I, 311, 317 Desulfomonas pigra 261
Chromatiaceae 7, 300, 30 I, 307, 311. Desulfonema limicola 261,262
317 Desulfosarcina variabilis 261
Chromatium vinosum 227, 287. 319 Desulfotomaculum acetoxidans 260, 262,
Chromobacterium violaceum 144 264
Citrobacter freundii 248 nigrificans 261
Clostridium 144, 176.224 Desulfovibrio africanus 261, 262
aceticum 249, 250. 281 desulfuricans 261, 262
acetobutylicum 228-231. 281 gigas 227, 247, 248, 287
acidiurici 227. 277, 282 vulgaris 261, 319
aurantibutyricum 228 Desulfuromonas acetoxidans 264
beijerinckii 228
botulinum 275
butyricum 109.209,225,230,231, Ectothiorhodospira 30 I
233 Enterobacter 237, 281
cellobioparum 234 aerogenes 120, 131, 154,237,240
cylindrosporum 236, 277 Enterobacteria 109
felsineum 144 Erwinia 237, 240, 281
formicoaceticum 234, 247, 250, amylovora 214,280
281 carotovora 144
glycolicum 234, 279 Escherichia coli 9, 12-93,96-111, 113,
indolis 234 116-119, 121, 127-132, 147, 149-
kluyveri 9, 225, 226, 234-236, 260, 151,154,178, 179, 181-183, 185-
281. 313 189, 191-196, 199-204,207,209.
oroticum 209, 279 227,237,240,242,247,248,273,
pasteurianum 9, 109, 209, 225-227, 281
287,319 Eubacterium barkeri 279
perfringens 125,147,230,231, limosum 133, 135,209,225,250
273
propionicum 243,244,269,270,281,
282 Flavobacterium 154
sphenoides 120, 221, 234 Frankia alni 319
sporogenes 275 Fusobacterium nucleatum 225
sticklandii 209. 269, 273, 275, 282
symbiosum 242
tetanomorphum 187,228,245,271- Gallionella 285
273,282 Gloeocapsa 309,319,323
thermoaceticum 227, 234, 250, Gluconobacter 119, 122
281 oxydans 169, 170
thermoautotrophicum 250
thermocellum 143, 234
thermohydrosulfuricum 250 Halococcus 136, 137
tyrobutyricum 265 Hansenula 168
Corynebacterium 144, 156 Hyphomicrobium 120, 168
glutamicum 154, 173, 174 vulgare 123
poinseltia 135
Cyanobacteria 317, 319, 321-323
Cytophaga 143 Klebsiella pneumoniae 319, 325, 326
Index of Organisms 339
Lactobacillus 145, 147 Nitrobacter 4, 8, 285, 289, 290, 294,

brevis 215, 233 317
bulgaricus 215 winogradskyi 9
casei 215, 220, 222 Nitrococcus 285
curvatus 215, 220 Nitrosococcus 285
delbrueckii 215, 222 Nitrosomonas 4, 8, 285, 295
fermentum 215 Nitrosospira 285
lactis 215 Nitrospina 285
plantarum 209,215,220,222,223, Nocardia 154, 156
280 autotrophica 283-285
Leuconostoc cremoris 220, 280 opaca 285
dextranicum 215 petroleophila 154
mesenteroides 215, 220
OsiciLlatoria 309
Megasphaera elsdenii 243, 281,
287
"Methanobacillus omelianskii" 266 Paracoccus denitrificans 121, 123, 149,
Methanobacterium 136, 137 153, 285, 288
ruminantium 9 Pediococcus cerevisiae 215
thermoautotrophicum 253, 254, 260, Penicillium 151
287 citrinum 145
Methanobrevibacter arboriphiLus 136, Peptostreptococcus asaccharoLyticus 87,
137,253 243, 247, 272, 273
Methanococcoides methylutens 253 micros 269, 270, 275, 282
Methanococcus 137 prevotii 270
vannielii 253 Photobacterium phosphoreum
Methanogenium 137 126
Methanolobus tindarius 253 PLectonema 323
Methanomicrobium 137 Propiogenium modestum 242
Methanoplanus Limicola 253 Propionibacterium shermanii 128
Methanosarcina barkeri 137, 253, 254, Protaminobacter 168
257,258,260,282 Proteus vulgaris 144, 147
mazei 253 Pseudomonas 138, 147, 149, 151, 168,
Methanospirillum hungatei 136, 137, 174, 176-178,209
253 acidovorans 151,152,158, 162
Methanothrix soehngenii 253 aeruginosa 109, 117, 128, 130, 131,
Methylobacter 162 144,147,151,152,154
Methylococcus 162, 163, 165 carboxydovorans 285
Methylocystis 162, 165 chitinovorans 144
Methylomonas 162, 165, 169 citronellolis 128
MethylophiLus 169 facilis 283
Methylosinus 162, 165 fluorescens 116, 117, 151, 154
Microbacterium lacticum 135 methanica 162
Micrococcus luteus 109, 121 muLtivorans 131
radiodurans 209 oLeovorm;s 155
Moraxella lwoffi 151 oxalaticus 153, 154, 168, 177
Mucor pusiLlus 145 putida 117,141,147,174,176,179-
Mycobacterium flavum 319, 322 182
phlei 131 saccharophiLa 114, 116, 119, 144,
smegmatis 109, 154 283
340 Index of Organisms
Pseudomonas (continued) lactis 215, 220, 209

stutzeri 123 thermophilus 135
testosteroni 162 Succinivibrio 247
Pullularia pul/ulans 142 Sulfolobus 136, 137,285
Pyrodictium occultum 136, 137, 264, acidocaldarius 294
265 Synechococcus 309
Syntrophobacter wolinii 266
Syntrophomonas wolfei 266
Rhizobium 149,318,321,323,324
japonicum 116,209,319
Rhizopus 142 Thermoanaerobacter ethanolicus 214
Rhodomicrobium vannielii 313 Thermoanaerobium brockii 214
Rhodopseudomonas capsIllata 125, 133, Thermoplasma 136, 137
313,319 Thermoproteus tenax 264, 265
gelatinosa 203-205, 221 Thermus aquaticlls 133
paillstris 300, 313 Thiobacil/us denitrificans 8, 123, 285,
sphaeroides 128, 164,304,313 291,295
sulfidophila 300 ferrooxidans 8, 116, 285, 292, 293
viridis 303, 304 intermedius 116, 294, 317
Rhodospirillaceae 7, 122,300,301,303, neapolitanus 285, 295, 300
307,311,313,314,317 novel/us 118, 127,294,295,317
Rhodospirillilm rubrum 128, 131, 132, perometabolis 294
193,299,313,314,319 thiooxidans 291, 292, 295
tenue 313 thioparus 295
RuminococcllS 247 versutus 295
albus 266, 267 Thiocapsa roseopersicina 303, 313
Thiomicrospira denitrificans 123
pelophila 285
Saccharomyces 210 Thiothrix 285
Salmonel/a 237, 281 Torulopsis 168
typhimllrium 81, 110, 128, 188, 192, colliculosa 154
193,203,209 Treponema pallidum 133
Sarcina ventriculi 213,214,280 Trichoderma reesei 143
Selenomonas 247
Serratia 237, 281
Shigella 237, 240, 281 Veillonel/a 221, 247
Sphaerotilus natans 131 alcalescens 125, 209, 242, 249
Spirillum itersonii 123 Vibrio cholarae 133
Spirulina 309
Sporocytophaga 143
Sporolactobacillus inulinus 215 Wolinella succinogenes 248
Sporomusa sphaeroides 250
Staphylococcus aureus 108, 109, 135, Xanthobacter autotrophicus 285
145, 176 Xanthomonas phaseoli 115, 116
Streptococcus 145, 147
allantoicus 151
cremoris 215,224,280 Zymomonas 280
diacetylactis 220 anaerobica 213
faecalis 209,215,222,223 mobilis 213,214,222
Subject Index
Acetaldehyde 174,212,214,235 AceLOgenic bacteria 8, 249-252

Acetaldehyde dehydrogenase 169, 214. Acetone 171, 229, 230
217,235,238 Acetyl-ACP 67
Acetate 204,219,220,234,235,237, Acetyl-CoA 99, 128, 129,214,222,
238, 242, 7.52, 259, 263, 268, 272. 226, 234, 260, 274
279 formation from pyruvate 18-20
anaerobic oxidation of 264 in fatty acid synthesis 67-69
fermentation of 234, 249-253, 259, oxidation of 20-22
264 Acetyl-CoA acetyltransferase 229
formation of 169,249,265 Acetyl-CoA carboxylase 67
growth on 99-103, 253, 264 Acetyl-CoA synthetase 99. 170
Acetate kinase 218,21'1,235,238,243, Acetylene, reduction of 318
251, 263, 274, 275 N-Acetylglucosamine 132, 133, 137
Acetic acid bacteria 169-171 N-Acetylglucosamine phosphate 72
Acetoacetate 131, 132, 231, 274 N-Acetylglutamate 51
Acetoacetate decarboxylase 230, 231 N-Acetylglutamate-'Y-semialdehyde 51
Acetoacetyl-ACP 68 N-Acetylglutamate kinase 51
Acetoacetyl-CoA 132, 229-231, 235 N-Acetyl muramic acid 72, 132, 133
Acetohydroxy acid isomeroreductase 48. N-Acetylmuramyl-pentapeptide 74
192 N-Acetylomithine 51
Acetohydroxy acid synthase 48, 192 Acetyl phosphate 217,218,222,226
a-Aceto-a-hydroxybutyrate 48 N-Acetyl-L-talosaminuronic acid 137
Acetoin 172, 220 O-Acetylserine 42
cleavage of 172 O-Acetylserine sulfhydrylase 42, 49
Acetoin dehydrogenase 221 cis-Aconitase 21, 193
a-Acetolactate 48, 238 cis-Aconitate 21
a-Acetolactate decarboxylase 172, 238 ACP (see acyl carrier protein)
a-Acetolactate synthase 172, 238-240 Active site 198
342 Subject Index
Active transport 105-108 Alcohol dehydrogenase 169, 176,212,

Acrylate pathway 243 217.235,238
Acrylyl-CoA 150.244 Alcohol fermentation 210-214
Acyl carrier protein 67-69 Aldehyde dehydrogenase 169
structure of 67 Aliphatic hydrocarbons 154-157
Acyl-CoA synthetase 150 I1-Alkane. oxidation of 155
Adenine 278 Allantoate 152
Adenine. formula of 60 Allantoate amidohydrolase 152
Adenine deaminase 278 Allantoicase 152
Adenosine-5' -diphosphate 4-6 Allantoin 152
Adenosine-3' -monophosphate 42 Allantoinase 152
Adenosine-S' -monophosphate 5, 62-65, Allolactose 179, 182
102. 263 Allosteric effectors 197
Adenosine-3' ,5' -monophosphate 186 Allosteric enzymes, properties of 197
Adenosine-5'-phosphosulfate 43,263 Amino acids, activation of 88
Adenos ine- 3'-phosphate-S' - as growth factors 9
phosphosulfate 43 biosynthesis of 43-55
Adenosine-5' -triphosphate degradation of 145-149
as phosphorylating agent 5 fermentation of 269-277
formula of 5 Amino acid acetyltransferase 51
requirement 39 o-Amino acid oxidase 145
in transport 108 L-Amino acid oxidase 145
synthesis of 4-6, 12-36, 122-126, Aminoacyl site 88
208-280,283-316 Aminoacyl-t-RNA synthetase 88. 89
S-Adenosylmethionine 70, III, 239 p-Aminobenzoic acid 10
Adenylate cyclase 186 4-Aminobutyrate 273
Adenylate kinase 5. 62 L-3-Aminobutyryl-CoA deaminase
Adeny!osuccinate lyase 64 274
Adenylosuccinate synthetase 64 4-Aminoimidazole 278
Adenylylation 204 5-Aminoimidazole-4-carboxamide
Adenylyltransferase 204 ribonucleotide 54
ADP (see adenosine-S' -diphosphate) 4-Amino-5-imidazole carboxylate 278
ADP-glucose 75 4-Amino-5-imidazole carboxylate
ADP-glucose pyrophosphorylase 75, 200 decarboxylase 278
ADP-sulfurylase 292 4-Amino-imidazole deaminase 278
Aerobes, effect of superoxide 35 Aminotransferase (see transaminase)
Aerobactin 109 5-Aminovalerate 275
Aerobic heterotrophs, catabolic activities Ammonia 269,273,277, 320
of, 141-176 assimilation of 40-42
metabolic diversity of 104-138 oxidation of 285, 289-291
Aerobic metabolism of glucose 12-35 Ammonia oxidizing bacteria 284, 285
of inorganic compounds 283-317 AMP (see adenosine-5' -monophosphate)
of organic substrates 96-103, 141-176 a-Amylase 142
Agarase 144 I)-Amylase 142
L-Alanine 44.73,74, 135 Amytal 34
o-Alanine 73, 74. 78 Anaerobes 208-280
Alanine. fermentation of 269 Anaerobic food chain 265-268
Alanine dehydrogenase 146 Anaerobic respiration 122-126. 2 IO
o-Alanyl-o-alanine 73. 74 Anaerobiosis 208
Subject Index 343
Anaplerotic sequences 92, 100-103, 127 ATP requirement, in biosynthesis 39

diversity in 103 in CO 2 -fixation 298
Anthracene 159 in N2 -fixation 321
Anthranilate 52, 53, 159 ATP yields 22,27-33, 122-126,210,
Anthranilate synthase 52, 188, 195 216,218,222-224,228,233,235,
Antibiotics 79 240, 244, 258, 264, 269, 284
Anticodon 88, 89 ATP synthase 29,32,33,257,305,315
Antiport 107,241 ATP sulfurylase 42, 263
APS (see adenosine-5' -phosphosulfate) Attenuation 187
APS-phosphokinase 42 Autotrophy definition 7
APS-reductase 263, 292 Available hydrogen 22, 232
Arabinose 98,99, 184, 185 Azoferredoxin 320
Arabinose isomerase 99
Archaebacteria 136, 252, 264, 285
cell wall components of 136-138 Bacterial bioluminescence 126
Arginase 147 Bacterial fermentations 208-282.
L-Arginine 51 Bacterial nutrition I-I I
fermentation of 273 Bacterial photosynthesis 300-316
Arginine decarboxylase 147, 148 Bacteriochlorophyll 301-304
Arginine deiminase 147, 148 Bacteriopheophytin 305
Arginine oxydase 147, 148 Bacteriorhodopsin 314-316
Argininosuccinate 51 Bacteroids 323
Aromatic amino acids 52, 53 Bactoprenol 77
regulation of 192, 196 Balances of fermentations 231-233
Aromatic compounds 157-162, 181, Base pairing 83, 84
182, 174-176 Bases, purine and pyrimidine 60
degradation of 157-162 Benzaldehyde 159
L-Ascorbate, synthesis of 171 Benzene 159
Asparagine 45 Benzoate 158, 159
Asparagine synthetase 45 degradation by phototrophs 313
Aspartase 147 Benzoylformate 159
Aspartate 44,46, 135 Benzyl alcohol 159
fermentation of 270 Betaine 275, 276
Aspartate family, regulation of 191, BFoF, (see ATPase)
195 Bifidum pathway 216
Aspartate-4-phosphate 46 Binding protein, in transport 107, 108
Aspartate kinase 46, 191, 195 Bioelements 1-4
Aspartate semialdehyde 46, 47 Biosynthesis of cell material 37-93
Aspartate semialdehyde Biotin 10, 127, 128, 173,240,245
dehydrogenase 46, 191 Biphytanyldiglycerol tetraether
Aspartate transcarbamoylase 61, 194, glycolipid 136, 138
199 1,3-Bisphosphoglycerate 17
Assimilation of ammonia 40-42, 324 Branched-chain amino acids 48
Assimilation of CO 2 295-300,311-313 Branched fermentative pathways 232
Assimilatory reduction of nitrate 125 Branched pathways 190-193, 195
Assimilatory reduction of sulfate 42, 43 Braun's lipoprotein 77
ATPase (see ATP synthase) 2,3-Butanediol 172, 237, 238
ATP (see adenosine-5' -triphosphate) 2,3-Butanediol dehydrogenase 172, 238
ATP citrate-lyase 312 Butanediol fermentation 237
344 Subject Index
Butanol 229. 230 Catechol-2,3-oxygenase 161, 176

Butanol-acetone fermentation 224 COP-diacylglycerol 70
Butanol dehydrogenase 231 Cell constituents 37-93
Butyraldehyde 231 Cell membrane 74
Butyraldehyde dehydrogenase 231 Cell wall 77-81. 132-138
Butyrate 224. 234. 272 Cellobiose 143, 144
synthesis of 228. 229 Cellobiose phosphorylase 144
Butyrate fermentation 224 Cellulase 143, 144
Butyrate kinase 229 Cellulose, degradation of 143
Butyryl-CoA 229. 230. 235 Central pathways. regulation of 193
Butyryl-CoA dehydrogenase 229. 235. Cephalosporin 79
274 Chemical elements, in nutrition 1-4
Butyryl phosphate 229 Chemiosmotic theory 29-32
Chemolithotrophy 283-232
definition of 7
Calcium. in nutrition I, 2 facultative 293-295
Calvin cycle 294-300 obligate 293-295
Camphor 174 Chemoorganotrophy, definition of 8
Caproate 234 Chemotaxis 110-112
Carbamoyl aspartate 61 Chitinase 144
Carbamoyl phosphate 41. 51. 61, 273 4-Chlorobenzoate 174, 175
Carbamoyl phosphate synthetase 41 p-Chlorobiphenyl 174
Carbon dioxide (see CO,) Chlorosomes 307
Carbohydrates, degradation of 12-18. Choline 70, 245
96-99. 114-120, 141-145,210-242 Chorismate 52, 188
synthesis of 71-74 Chorismate mutase 52, 192, 196
Carbon monoxide 284 Citramalate 272
Carbon monoxide-oxidizing bacteria 284, Citramalate lyase 272
285 Citrate 21, 173,204,236
Carbonylcyanide-p- fermentation of 220
trifluoromethoxyphenyl hydrazone production of 173
(see FCCP) Citrate lyase 203, 220, 240
Carboxy lases 67,92, 127-129,236, regulation of 203-205
248,249,296-300,312 Citrate synthase 21, 193,200
I3-Carboxy-cis. cis-muconate 160 stereospecificity 236
I3-Carboxymuconate-lactonizing Citric acid cycle (see tricarboxylic acid
enzyme 160, 181 cycle)
-y-Carboxymuconolactone 160 L-Citrulline 51
-y-Carboxymucono lactone Cleavage of aromatic rings 157-162. 313
decarboxylase 160, 181 phosphorylytic 144
Carboxysomes 300 CO 34,285
Carotenoides 303, 304 CO dehydrogenase 227, 251, 258, 259
Carrier, in electron transfer 23-27, 226- CO oxidase 124,227,289
228, 300-303 CO, as carbon source 7, 8, 249-260.
in hydrogen transfer 23-27 295-300, 311. 312
in transport 105-108 CO" carboxylation of acetyl-CoA 67.
Catabolite repression 185.294 236
Catalase 35. 168, 172,209,223 carboxylation of PEP 92, 127
Catechol 159-161 carboxylation of pyruvate 127, 128
Catechol-1.2-oxygenase 160. 181. 182 carboxylation of propionate 149
Subject Index 345
carboxylation of ribulose-I ,5- Cytochrome b 24,34, 121,240,247,

bisphosphate 296 308
CO 2 carrier 245 Cytochrome c, 24, 121
CO, reductase 227 Cytochrome c, 305, 306
CO 2 requirement 236, 248 Cytochrome c, 260
CoA transferase 131, 132, 274 Cytochrome d 24,34, 121
Cobalt, in nutrition 3 Cytochrome f 310
Codon 88, 89 Cytochrome 0 24, 34, 121
Coenzyme A, formula of 67 Cytochrome p 460 291
Coenzyme A transferase 230, 235, 243, Cytochromes, formulas of 25
244,275 Cytoplasmic membrane 74
Coenzyme B" 245-247, 273 Cytosine formula of 60
Coenzyme F4 ,o 254, 260
formula of subscript 255
Coenzyme M 9, 10, 254, 256, 258 DAHP synthase 52, 196
formula of 255 Deacetylation 205
Coenzyme M methyltransferase 258, 259 Deadenylation 204
Coenzyme Q 23, 24, 34, 121, 305, 306 Deamination 146
cycle 30 Decarboxylation 18-20,212,220,221,
formula of 25 226,231,238,240-242
Composition of cell material 37-39 Degradation of (see also fermentation of)
Concerted symmetry model 199 Degradation of acetate 99
Configuration of lactic acid 215 allantoin 152
Conformational protection of amino acids 145
nitrogenase 322 arginine 148
Cooperative binding 197 aromatic compounds 157
Coordinate induction 179 catechol 160
Copper, in nutrition 3 cellulose 142
Core enzyme 86 4-chlorobenzoate 175
Core oligosaccharide 79, 80, 133 fructose 96
Corrin ring 247 galactose 97
Creatinine 279 gluconate 15-18, 114-117
Crotonase 131, 132, 235 glucose 114-119, 172,212
Crotonyl-ACP 68 glyoxylate 152
Crotonyl-CoA 229,235,242,272, 132 histidine 147
CRP protein 186 hydrocarbons 154
CTP (see cytidine-5' -triphosphate) lactose 97, 185
CTP synthetase 62 malate 100
Cummulative feedback inhibition 196 organic acids 99, 149
Cyanide 34 palmitic acid 150
Cyanocobalamin 10, 247 pentoses 98
Cyclase 54 poIY-13-hydroxybutyrate 131
Cyclic AMP 186, 187 polymers 141
Cycl ic AMP receptor protein 186, 187 propionate 149
Cyclic photophosphorylation 306 protocatechuate 160
Cystathionine 46 pyruvate 100
L-Cysteine 42, 49 starch 142
Cytidine-5' -monophosphate 60-62 sugars 114-119
Cytidine-5' -triphosphate 70, 194 urate 152
Cytochrome a, 24, 121 Degradative plasm ids 174-176
346 Subject Index
5-Dehydroshikimate 52. 158 Dihydroxy acid dehydratase 48, 192

5-Dehydroquinate 52 o:-I3-Dihydroxyethylthiamine
5-Dehydroquinate synthase 52 pyrophosphate 2 I7
Denitrification 122-124 0:, I3-Dihydroxyisovalerate 48
Denitrifying bacteria 122 0:, I3-Dihydroxy-l3-methylvalerate 48
5'-Deoxyadenosine 247 5,6-Dimethylbenzimidazole
5' -Deoxyadenosylcobalamin 247 ribonucleotide 247
3-Deoxy-D-arabino-heptulosonate-7- 2.4-Dinitrophenol 34
phosphate (DAHP) 52. 196 Dihydrofolate 66
3-Deox y-7-phospho- D-arabino- Dioxygenase 160-162
heptulosonate synthase 192, 196 Disulfide bond 19, 87
3-DeoxY-D-manno-octulosonate 80 Dissimilatory reduction of nitrate 122
Deoxyribonuclease 145 sulfate 260
Deoxyribonucleotides 59-66 Dithiooctanic acid (see lipoic acid)
Deoxythymidylic acid 66 DNA 38,39,81-86
Desulforedoxin 262 DNA ligase 84
Desulfoviridin 262 DNA polymerase 84
Desulfurication 260-265 DNA replication 83-85
Dextrins 142 DNase 145
Diacetyl 220 Docking protein 113
Diacetyl synthase 221 Double helix structure 82
2,5-Diaminohexanoate 274 Duplex 83
3,5-Diaminohexanoate 246, 274
3,6-Diaminohexanoate 274
3,5-Diaminohexanoate ED pathway (see Entner-Doudoroff
dehydrogenase 274 pathway)
0:, e-Diaminopimelate 47,73,74, 132, Electrochemical gradient of H + 32, 107,
133 224, 246
Diaminopimelate decarboxylase 47 of Na+ 240,241
Diaminopimelate epimerase 47 Electron acceptor 7,8, 122, 126,248,
Diauxie 185 287
Diauxie growth 185 Electron carrier 26, 121, 226, 248, 262,
Dicarboxylic acids. growth on 102 305-310
2.4-Dichlorophenoxyacetate 174 Electron donor 7, 8, 283-285, 30 I, 302
DeeD 34 Electron-transferriJ'lg flavoprotein 243
Dicyclohexylcarbodiimide (see DeeD) Electron transport phosphorylation 5, 27-
Diffusion, facilitated 105. 112 34, 245, 248, 258. 263, 288-293,
passive 104 305-311
Dihydrodipicolinate 47 Elemental sulfur 264, 284, 30 I
Dihydrodipicolinate cyclohydrolase 191 EMP (see Embden-Meyerhof-Parnas
Dihydrodipicolinate synthase 47, 195 pathway)
Dihydrolipoate dehydrogenase 18, 19 Embden-Meyerhof-Parnas pathway 15-
Dihydrolipoate transacetylase 18, 19 18,101,212,228,251
Dihydroorotase 61 End product repression 187
4.5-Dihydroorotate 61 Energy charge 202
Dihydroorotate dehydrogenase 61. 227 Energy metabolism
Dihydroxyacetone 167. 171 aerobic 12-36, 121-126
Dihydroxyacetone synthase 167 anaerobic 208-282
Dihydroxyacetone kinase 167 chemolithotrophic 282-295
Dihydroxyacetone phosphate 17, 119. diversity in 122
166. 167 phototrophic 300-316
Subject Index 347
Energy source 6-9 FCCP 34

Enoate reductase 277 Feedback inhibition 194
Enolase 15, 17,212 Fe'+ oxidizer 284, 285, 293
3-Enolpyruvylshikimate-5-phosphate 52 Fermentation 208-282
Enol-I-(o-carboxyphenylamino)- acetate 249-252
deoxyribulose-5-phosphate 53 alcohol 210
Enoyl-ACP reductase 68 butanediol 237
Enterochelin 110 butanol-acetone 224
Entner-Doudoroff pathway 114-117, butyrate 224
213,294 ethanol-acetate 234
Enzyme activity, regulation of 194 methane 211, 252
Enzyme induction 178 mixed acid 237
Enzymes, allosteric 197-206 propionate 242
regulation by covalent succinate 242
modification 202-206 sulfide 211
Enzymes. plasmid-encoded 174-176 Fermentation balances 231-233
D-Erythro-L-glycero-3-hexulose-6- Fermentation of amino acids 269-277
phosphate 165 heterocyclic compounds 277-280
Erythro-13-hydroxyaspartate aldolase 153 lactate 242-249
Erythro-13-hydroxyaspartate sugars 210-242, 249-252
dehydratase 153 Fermentative pathways
Erythrose-4-phosphate 52, 56-58, 218, branched 232
297 linear 232
ETF (see electron transferring Ferredoxin 87, 225-227, 233, 244, 260,
flavoprotein) 262.287,308.310,312
Ethane 154 Ferredoxin oxidoreductase 227, 233,
Ethanol 169,210,212,214,215,217, 234.251
230. 232. 233, 235, 237. 238 Ferrichrome 109
fermentation of 234 FeS-c1uster 26, 226
oxidation of 169 FeS-protein (see iron-sulfur protein)
Ethanol-acetate fermentation 234 Fixation of carbon dioxide 236, 295-
Ethanolamine 80, 246 300, 311-313
Ethanolamine deaminase 245, 246 molecular nitrogen 249-260, 266,
Ethylene, formation of 318 318-326
Exoenzymes 141-145 Flagellar movement I 10-112
Flavin adenine dinucleotide 21, 27
Flavin mononucleotide 27, 227
F"F, complex 32 formula of 24
Facilitated diffusion 105 Flavodoxin 227, 239, 262
Factor, F.,,, 258 Flavoproteins 35, 121, 244
Factor F.\tI 254 redox potential of 27
formula of 255 Folate, H,_ 66
p-Factor 86 H._ 50, 251, 270
a-Factor 86 H4 _ formula of 50
FAD (see flavin adenine dinucleotide) Folic acid 10
Faraday constant 23 Fom1aldehyde 163-168
Fatty acids 67-69 Formaldehyde dehydrogenase 163
degradation of 149-154. 265 Formamide 147
unsaturated 68 Formate 153, 168.237,238,248,251,
Fatty acid synthetase 68 268, 270, 279
Fatty acyl-CoA dehydrogenase 150 fermentation of 253
348 Subject Index
Formate dehydrogenase 124, 153, 163, Gluconate kinase 117

194,240,251,270,279 Gluconate-6-phosphate 56-58, 114, 117,
Formate-hydrogen lyase 194, 238, 240, 118,217,300
241 Gluconate-6-phosphate
Formimino-H. folate cyclodeaminase 279 dehydrogenase 56-58, 118,217
N-Formimino L-glutamate 147 Gluconeogenesis 100, 119
Formiminoglutamate hydrolase 147 Glucosamine-6-phosphate 72
Formiminoglycine 278 Glucose, degradation of 15-18, 114,
Formyl-CoA 153 118, 172
Formylmethanofuran 256 cell material from 37-93
Formyl-H. folate 250, 279 fermentation of 210-219,229,231,
Formyl-H. folate synthetase 251, 269, 233, 238, 242
270,279 Glucose dehydrogenase 117
Formylkynurenine 159 Glucose: galactose-I-phosphate
N-Formylmethionyl-t-RNA 90, 91 uridylyltransferase 98
Free energy changes 12,122,211,240, Glucose oxidase 172
249, 253, 284 Glucose-PEP phosphotransferase 14, 17,
Fructose, growth on 96, 97 108, 109
Fructose-I,6-bisphosphatase 101, 200, Glucose-I-phosphate 71,98, 144
297 Glucose-6-phosphate 17, 56-58, 114,
Fructose-I,6-bisphosphate 15-17, 20 I 117
Fructose-I ,6-bisphosphate aldolase 15, Glucose-6-phosphate dehydrogenase 57,
17, 165,297 114, 117, 118,217
Fructose-I-phosphate 96, 97 Glucose-6-phosphate isomerase 15, 17,
Fructose-6-phosphate 17, 57, 10 I, 165, 118,219
218,297 Glutaconyl-CoA 242, 272
Fructose-6-phosphate Glutaconyl-CoA decarboxylase 242, 272
phosphoketolase 218, 219 D-Glutamate 73,74, 135
Fumarase 21, 238, 244 L-Glutamate 40,41,51,173,204,205,
Fumarate 21, 24, 246-248, 264 236, 246, 272
Fumarate reductase 194, 238, 241, 244- fermentation of 245, 271, 272
248,264 production of
4-Fumarylacetoacetate 157 Glutamate decarboxylase 273
Fumarylacetoacetate hydrolase 157 Glutamate dehydrogenase 40, 146, 272,
Futile cycle 201 275
Glutamate kinase 51
Glutamate mutase 271, 272
Galactokinase 98 Glutamate semialdehyde 5\
Galactose 97 Glutamate semialdehyde
Galactose-I-phosphate 98 dehydrogenase 51
I3-Galactosidase 97, 178, 182, 183 Glutamate synthase 41, 227
I3-Galactoside acetyltransferase 97, 183 Glutamine 40,41,51,73,74,204,324
GOP (see guanosine-5-diphosphate) Glutamine aspatate amidotransferase 324
Genes 178-194 Glutamine a-oxoglutarate
Genetic code 88 amidotransferase 41, 324
Gentisate 158 Glutamine synthetase 40, 41, 203, 324,
Glucoamylase 142 325
Gluconate 116,117,171,172 Glyceraldehyde-3-phosphate 17, 56-58,
production of 172 118, 166, 169,217,218,269-298
Gluconate dehydrogenase 117 dehydrogenase 15, 17, 18,2\2,296
Subject Index 349
Glycerate 152 2-n-Heptyl-4-hydroxyquinoline-N-oxide

Glycerate kinase 152, 164 (see HQNO)
Glycerate pathway 151, 152 Heterocyclic compounds, fermentation
D-Erythro-L-glycero-3-hexulose-6- of 277-280
phosphate 165 Heterocysts 323
Glycerol 171, 245, 246 Heterofermentative pathway 215
Glycerol dehydrase 245, 246 Heterotrophy, definition of 7
Glycerol phosphate acyltransferase 69 cis-9-Hexadecenoate 68
Glycerol-3-phosphate dehydrogenase Hexokinase 217, 219
69 Hexose amines 71, 72
Glycine 49, 277 Hexulose-6-phosphate 165
fermentation of 269 Hexulose-6-phosphate isomerase 166
Glycine decarboxylase complex 270 Hexulose-6-phosphate synthase 165, 166
Glycine formiminotransferase 279 High-energy bonds 6
Glycine reductase 269-271, 275, 279 High-potential iron protein 227
Glycogen 75, 76, [30 Histidase 147
Glycogen phosphorylase 202 Histidinal 54
Glycogenic reactions, regulation of 200- L-Histidine 54, 55, 188
202 degradation of 147
Glycolipid, formula of 66 Histidine operon 189
Glycollate 153 Histidinol 54
Glycolytic reactions, regulation of 200- Histidinol dehydrogenase 54
202 Histidinol phosphate 54
Glyconeogenesis 101 Homocysteine "46
Glyoxalase 1[9 Homofermentative pathway 215
Glyoxylate 100, 151-154, 164 Homogentisate 157
Glyoxylate cycle 100, 10J, 128, 129, Homoserine 467
149, 171 Homoserine acyltransferase 46
Glyoxylate carboligase 152 Homoserine dehydrogenase 46, 191,
GOGAT (see glutamate synthase) 195
GMP (see guanosine-5' -monophosphate) Homoserine kinase 46, 191
GMP synthetase 64 Homoserine phosphate 46
Gram-negative bacteria 79 HPr 14,96
Gram-positive bacteria 79 HQNO 34
Group translocation 14, IS, 108, 109 Hyaluronidase 144
Growth, diauxie 185 Hydrocarbons, degradation of 154-157
Growth factor requirement 9, 10 Hydrogen 235,237,240,249,252,263,
Growth yields 222 266,284,321
GTP (see guanosine-S' -triphosphate) available 22, 232
Guanine 278 formation of 233, 235, 237, 240, 266-
formula of 60 268
Guanine deaminase 278 in nutrition 4
Guanosine-S' -monophosphate 60, 62-65, oxidation of 283, 301
112 photoproduction of 314
Guanosine-5'-triphosphate 90,91 Hydrogen bonds, in base pairing 83, 84
Hydrogen carriers 24, 26
Hydrogen cycling 262-264
Helicase 84 Hydrogen electrode 22
Hemin 10 Hydrogen-oxidizing bacteria 283
Heptose 80 Hydrogen peroxide 35
350 Subject Index
Hydrogen sulfide 42. 264, 265. 284, IMP (see inosinic acid)
289-293.301, 308 IMP cyclohydrolase 64
Hydrogen transfer, interspecies 266, 267 IMP dehydrogenase 64
Hydrogenase 226,229,233.241, 263, Imidazoleacetol phosphate 54
286-288.321 Imidazoleglycerol phosphate 54
f3-Hydroxyaspartate pathway 153 4-lmidazolonase 278
f3-Hydroxyacetyl-ACP dehydrogenase 68 4-lmidazolone 278
L-3-Hydroxyacyl-CoA Imidazolone propionase 147
dehydrogenase 150 4-lmidazolone-5-propionate 147
L-3-Hydroxyacyl-CoA hydrolyase 150, Incomplete oxidations 169-174
229 Indole-3-glycerol phosphate 53, 55
p-Hydroxybenzaldehyde 158 Indole-3-glycerolphosphate
m-Hydroxybenzoate 158 synthetase 53, 188
p-Hydroxybenzoate 158 Induced-fit model 198
p-Hydroxybenzoylformate 158 Induction 178
D ( - )-f3-Hydroxybutyrate 131, 132 Inhibitors 33, 34, 258
D ( - )-f3-Hydroxybutyrate Inosinic acid 64
dehydrogenase 131, 132 Informational macromolecules, synthesis
f3-Hydroxybutyryl-ACP 68 of 81-93
D ( - )-f3-Hydroxybutyryl-CoA Interconversion of enzymes 202-206
D ( - )-f3-Hydroxybutyryl-CoA Interspecies hydrogen transfer 265-268
dehydrogenase 131, 132 Invertase 144
L (+ )-f3-Hydroxybutyryl-CoA 229, 235 Iron, in nutrition I, 2
dehydrogenase 131, 132, 229, 231, oxidation 293
235 transport 109, I 10
D ( - )-f3-Hydroxybutyryl-CoA Iron-sulfur center 26, 227, 240
polymerase 132 Iron-sulfur protein 26, 225-227, 287,
2-Hydroxy-4-carboxymuconic 305-310
semialdehyde-hydrolase 161 Isoamylase 142
4-Hydroxy-4-carboxy-2-oxovalerate Isoci trate 21, 163
aldolase 161 Isocitrate dehydrogenase 21, 193, 203
f3-Hydroxydecanoyl-ACP in fatty acid Isocitrate lyase 100, 101, 163-165
synthesis 68 Isoenzymes 191
Hydroxyethyl-thiamine pyrophosphate- L-Isoleucine 48, 191, 192, 195
enzyme 19,213,226 Isopropanol 171
a-Hydroxyglutarate 272 a-Isopropylmalate 48, 192
a-Hydroxyglutarate dehydrogenase 272 f3-lsopropylmalate dehydrogenase 48,
Hydroxylamine 290 192
Hydroxylamine-cytochrome c Isopropylmatate isomerase 48, 192
reductase 291 a-Isopropylmalate synthase 48, 192
f3-HydroxY-L-mandelate 158 Isorenieratene 304
2-Hydroxymuconic semialdehyde
hydrolase 161
4-Hydroxy-2-oxovalerate aldolase 161 KDO (see 3-DeoxY-D-manno-
p-Hydroxyphenylpyruvate 52 octulosonate)
p-Hydroxyphenylpyruvate oxidase 157 KDPG (see 2-keto-3-deoxy-6-
f3-Hydroxypropionaldehyde 245, 246 phosphogluconate)
Hydroxypyruvate reductase 164 KDPG aldolase 114, 115
3-Hydroxytetradecanoic acid 80 3-Ketoacyl-ACP reductase 68
Hypoxanthine 277 3-Ketoacyl-ACP synthetase 68, 69
Subject Index 351
~-Ketoadipate (see 3-oxoadipate) o-Lysine 135

2-Keto-3-deoxy octonate 80, 133 L-Lysine 47, 135, 191, 195
2-Keto-3-deoxy-6-phospho- fermentation of 273
g]uconate 114, 115 ~-Lysine-5 ,6-aminomutase 246
2-Ketogluconate 117 L-Lysine-2,3-aminomutase 274
5-Ketogluconate 171 ~-Lysine mutase 245. 274
2-Ketogluconate kinase 117 Lysozyme 28
a-Ketoglutarate (see a-oxoglutarate)
2-Keto-6-phospho-gluconate 117
3-Keto-6-phospho-gluconate 56 Macromolecules 38, 39, 73-91
2-Keto-6-phosphogluconate reductase 117 cell content 38
~-Ketothiolase 132, 150, 274 degradation of 141-145
KREBS cycle (see tricarboxylic acid informational 81-93
cycle) periodic 75-81
L-Kynurenine 159 Magnesium
in nutrition I. 2
Lac operon 182 L-Malate 21. 102, 164
Lactate 214,215,216,217,219,234, fermentation of 220
237, 238, 263 L-Malatedehydrogenase 21, 164,200,
o-Lactate 119,215 238, 244
Lactate export Malate synthase 100,101,153
energetics of 223-224 4-Maleylacetoacetate 157
Lactate, fermentation of 243, 262 Maleylacetoacetate isomerase
Lactate dehydrogenase 205, 221, 238, 157
244,263 Malic enzyme 102, 220
stereospecificity of 219 Malo-lactate fermentation 220
o-Lactate dehydrogenase 216, 243 Malonyl-ACP 67
Lactate fermentation 214-224 Malonyl-CoA 67
Lactate oxidase 119. 222 Malonyl transacylase 67
Lactate racemase 219, 243 Maltose 142, 183
Lactic acid, configuration of 215 Maltose phosphorylase 144
Lactic acid bacteria 214 Malyl-CoA 164
Lactonase 57 Malyl-CoA lyase 164
Lactose 96-98, 178, 182 Malyl-CoA synthase 164
Lactose permease 97, 182 L-Mandelate 159
Lactyl-CoA 150 Manganese, in nutrition 3
Lactyl-TPP-enzyme 19 Mannitol 108, 220
Leaching 293 o-Mannose 108
Leader peptide 113 o-Mannose-6-phosphate 71
Leghemoglobin 324 Mannose-6-phosphate isomerase 71
L-Leucine 48, 192, 277 Membrane 106-109, 123-125, 155,
Light-harvesting center 305,307,310 286-293,304-310
Linear fermentative pathways 232 bound enzymes 28, 32, 240
Lipase 145 carrier 28-31, 105-114
Lipid A 79,80, 133 composition of 74
Lipids 38,39.65-71.73.74 energization 240-242, 288-293, 305-
Lipoic acid 10 311
Lipopolysaccharide 79-81, 133 outer 112
Luciferase 126, 127 vesicles 28
~-Lysine 246 Membrane potential 29-32, 224
352 Subject Index
Menaquinone 24.34,247.262.308 2-Methyl-ll-butyric acid 9

formula of 25 Methyl-H.-folate 49. 250. 251
2-Mercaptoethane sulfonic acid (see Methyl-H.-folate cyclohydrolase 251,
coenzyme M) 279
Mesaconate 272 Methylotrophic bacteria 162
Metabolic diversity 104-138 Methyltransferase I II. 251
Metabolism. chemolithotrophic 283-295 Mitochondrial respiratory chain 24
chcmoorganotrophic 141-176 Mixed acid fermentation 231-240
fermentative 208-280 Mixed cultures 265-268
phototrophic 300-314 Molybdenum 123, 140
regulation of 178-194 in nutrition 3
Meta-cleavage 161 Molybdenum cofactor 123, 124, 277
Metals. in nutrition 1-4 Molybdoferredoxin 320
Methane. free energy change of Monomers. biosynthesis of 38, 39
formation 257 formation from polymers 141-145
formation of 256-259 Mono-oxygenase 155. 156, 163. 291
oxidation of 163 cis. cis-Muconate 160, 181
Methane fermentation 211. 252-260 cis. cis-Muconate-Iactonizing
Methane-oxidizing bacteria 162 enzymes 160, 181, 182
Methanofuran 256 Muconolactone isomerase 160, 181.
formula of 255 182
Methanol. fermentation of 252. 253. 259 Multienzyme complex 17, 18, 20. 32,
oxidation of 163, 168 33. 320
Methanol dehydrogenase 163 Multivalent repression 191
Methanoloxydase 168 Multivalent attenuation 191
Methenyl-H. cyclohydrolase 270 Muramic acid 72
L-Methionine 46, 191, 195 Muramyl-pentapeptide 73, 74
Methoxatin 163. 169 Murein (see peptidoglycan)
Methyl-accepting chemotaxis protein III
Methylamine 168, 253, 258
r3-Methylaspartase 272 NAD+ (see nicotinamide adenine
r3-Methylaspartate, L-threo 246, 272 dinucleotide)
Methylcocnzyme M 259 NADH-dehydrogenase 23. 290
Methylcoenzyme M NADH ferredoxin reductase 233
methyl reductase 256. 259 NADH peroxidase 223
Methylene-H. folate 49. 269 NADH rubredoxin reductase 155
Methylene-H. folate dehydrogenase 251, NADP+ (see nicotinamide adenine
270 dinucleotide phosphate)
Methylene-H. folate reductase 251 Naphthalene 159, 174
Methylesterase I I I Negative control 183
Methylglyoxyal bypass 119 NERNST equation 23
Methylglyoxal synthase 119 Neuraminidase 144
tR)-Methylmalonyl-CoA 149, 242. 245 Nickel. in nutrition 3
Methylmalonyl-CoA decarboxylase 242 Nickel proteins 254
(S)-Methylmalonyl-CoA 245. 249 Nicotinamide adenine dinucleotide. as
(R)-Methylmalonyl-CoA mutase 151. growth factor 9
144. 145, 244. 246 Nicotinamide adenine dinucleotide 18,
(S)-Methylmalonyl-CoA: pyruvate 20-22
transcarboxylase 244 (reduced form). formation by reverse
Methylmalonyl-CoA racemase 151. 244 electron transfer 288
Subject Index 353
generation of 18, 20-22, 286-291, O-acetylserine 42

307,308 Obligate anaerobes 208-280
H, evolution from 233, 235 Obligate chemolithotrophy 293-295
oxidation in fermentations 212, 216, cis-II-Octadecenoate 68
233 n-Octane 174
oxidation with 0, 24, 30-32, 121 n-Octane oxygenase 176
redox potential of 27, 233, 257, 225, Oligomycin 34
289, 304-310 Oligosaccaride core 79, 80
Nicotinamide adenine dinucleotide Oligosaccharide chains, repeating 79, 80
phosphate 21 Operon model 182
(reduced form) formation of 21, 55- O/R balance 232
59,310 OR system 22
Nicotinic acid 10, 279 Organic acid, degradation of 149-154,
Nil' genes 325 265-268
Nitrate, as electron acceptor 122 D-Ornithine 135
assimilatory reduction 125 L-Ornithine 51, 135,273
dissimilatory reduction 122-125 Ornithine cyclase 275
fermentative reduction 125 Ornithine transcarbamoylase 51
formation of 284, 289-291 Orotate 61, 279
Nitrate reductase 123, 125, 194,241 Orotidine monophosphate 61
Nitrate/ammonia respiration 125 Orotidine-5' -phosphate decarboxylase 61
Nitrate/nitrite respiration 125 Ortho-cleavage 160
Nitric oxide reductase 125 O-succinyl homoserine 46
Nitrite, formation from ammonia 284, Osmotic shock 108
289-291 Outer membrane layer 77, 79-81, I 12,
formation from nitrate 124 133
Nitrite oxidizer 284, 285 Oxalate 153
Nitrite reductase 124 Oxaloacetate 21, 171, 204, 220, 312
Nitrogen, control 325 in anaplerotic reactions 100, 101, 127,
in nutrition 4 128, 174
fixation of 318, 326 in succinate-propionate
Nitrogenase 320 fermentation 244, 245, 248-249
components of 227, 320 Oxaloacetate decarboxylase 171, 221,
regulation of 324-326 240-242
Nitrogen fixation, symbiotic 323, 324 Oxaloacetate family 21
Nitrogenous compounds, fermentation Oxalosuccinate 21
of 269-280 Oxalyl-CoA 153
Nitrous oxide 124 13-0xidation 149
Nitrous oxide reductase 124 Oxidation, incomplete 169-174
Noncyclic photophosphorylation Oxidation of, n-alkanes 155-157
307-311 ammonia 285, 289-291
Non-heme iron protein (see iron-sulfur carbon monoxide 285, 289
proteins) ferrous ions 285
Nucleic acids, synthesis of 83-86 glucose 12-35, 114-120
Nucleosides 59 hydrogen 293, 286, 289
nucleoside diphosphate kinase 62 methane 163
nucleoside monophosphate kinase 62 methanol 163, 168
Nucleotides, formulas of 60 nitrite 285, 289-291
sulfide 289, 292, 301, 308
sulfur 264, 285, 289-293
354 Subject Index
Oxidation reduction balance 232 Passive diffusion 104

Oxidation reduction potential 22. 23. 27. Pasteur effect 213
194.209.225-228.257,289.306- Pectinase 144
310 Penicillin 79
Oxidative deamination 145 Pentoses. degradation of 98. 99. 118
Oxidative decarboxylation, mechanism synthesis of 55-59
of 19 Pentose phosphates. synthesis of 55-59
of pyruvate 18-20 Pentose phosphate cycle 55-59, 118. 169
3-0xoadipate 160. 181 PEP 170.249
4-0xoadipate enol-lactone 160 formula of 6
4-0xoadipate enol-lactone hydrolase 160. in anaplerotic reactions 127. 128. 129
181. 182 in biosynthesis 52. 72
3-0xoadipate pathway 160 in gluconeogenesis 101
3-0xoadipate pathway. regulation of 181 in transport 16. 17. 90
3-0xoadipate succinyl-CoA PEP-carboxykinase 100-102. 128. 129.
transferase 160. 181 149. 249
3-0xoadipyl-CoA thiolase 160 PEP-carboxylase 92, 127-129. 164.200.
3-0xo-5-aminohexanoate 274 238. 260
a-Oxobutyrate 48. 192 PEP-carboxytransphosphorylase 248. 249
2-0xo-4-carboxypent-4-enoic acid 161 PEP-phosphotransferase system 14. 96.
hydrolase 161 108. 109
a-Oxoglutarate 21. 40, 41. 312 PEP synthetase 100-102. 128. 129.260.
in fermentation of glutamate 272 312
in regulation 200. 204 Peptidase I 13
a-Oxoglutarate dehydrogenase 21. 193 Pept ide bond. formation of 87-91
a-Oxoisovalerate 42 Peptidoglycan 73.77-79, 133-135
2-0xo-3-methylbutyrate 192 synthesis of 77-79
2-0xo-3-methylvalerate 42. 192 variation of composition 133-135
2-0xopent-4-enoic acid hydrolase 161 Peptidoglycan layer 132, 133
Oxygen. as electron acceptor 23-32 Peptidyl site 88
effect on strict anaerobes 208-210 Peptone 9
evolution in photosynthesis 310 Periodic macromolecules 75-81
in nutrition 4 Periplasmic space 77
product of catalase reaction 35 Peroxisomes 168
sensitivity of nitrogenase 121. 283- pH gradient 29-33
295.321-323 PHB (see poly-f3-hydroxybutyrate)
Oxygenase 155. 156. 160-162. 163. Phenanthrene 159
175. 181. 182.291 Phenol 159
Oxygenolytic cleavage of ribulose-I. L-Phenylalanine 52. 157. 196
5-bisphosphate 299 Phenylalanine hydroxylase 157
Phenylpyruvate 52
Phosphatase 166
Palmitaldehyde 127 Phosphate, in nutrition I
Palmitic acid Phosphatidate cytidylyltransferase 70
degradation of 150 Phosphatidic acid 69. 70
Palmitoleate 68 Phosphatidyl choline 70
Pantothenate 10.67 Phosphatidyl ethanolamine 70
PAPS (see adenosine-3'-phosphate-5'- Phosphatidyl serine 70
phosphosulfate) Phosphodiesterase 186
PAPS reductase 42 Phospho-enol pyruvate (see also PEP)
Subject Index 355
Phospho-enol pyruvate: fructose 5' -Phosphoribosyl-N-

phosphotransferase 96 formylglycineamide 63
Phospho-enol pyruvate: glucose 5-Phosphoribosyl-I-pyrophosphate 53,
phosphotransferase 14 54,61, 188
Phosphoenzyme III, 14, 96 synthetase 61
I-Phosphofructokinase 96, 97 Phosphoribulokinase 298-300
6-Phosphofructokinase 15, 17, 166 Phosphoribulokinase 298-300
in Pasteur effect 213 Phosphoroclastic reaction 225
regulation of 200 Phosphotransacetylase 217, 226, 235,
Phosphoglucomutase 71 238,243,251,263,274,275
6-Phosphogluconate 56---58, 114, 116, Phosphorous, in nutrition I
117 Phosphotransferase, in sugar
6-Phosphogluconate dehydratase 114, transport 14, 108, 225, 229
115 Phosphorylase 144, 202
6-Phosphogluconate dehydrogenase, Phosphorylation, electron transport 5
56-58,217 substrate-level 6
6-Phosphogluconolactone 57 3-Phosphoserine 49
Phosphoglucose isomerase 17 Phosphoserine aminotransferase 49
3-Phosphoglyceraldehyde (see Phosphoserine phosphatase 49
glyceraldehyde-3-phosphate) Photolithotrophy, definition of 7
2-Phosphoglycerate 17 Photoorganotrophy, definition of 7
3-Phosphoglycerate 16, 17, 153, 164 Photophosphorylation, cyclic 306
product of CO 2 fixation 296-298 noncyclic 307-311
3-Phosphoglycerate family 49, 50 Photoproduction of molecular
3- Phosphoglycerate kinase 15, 17, 212, hydrogen 314
296 Photoreduction of NAD 308
Phosphoglycerate mutase 15, 17. 164, Photosynthesis 300-310
212 in halobacteria 314-316
Phospho-HPr 14, 96 Photosynthetic apparatus 303-31 I
3-Phosphohydroxypyruvate 49 Phosphotransferase system 13-15
Phosphoketolase 217, 218, 220 Phototrophy, definition of 6
Phospholipid, formula of 66 Phototrophic bacteria 225, 300-303
in membranes 74 Phycobilisomes 310
synthesis of 65-74 Plastocyanin 310
4' -Phosphopantetheine 67 Plastoquinone 310
Phosphoribose isomerase 57 Phototrophic metabolism 300-316
5-Phosphoriboxylamine 63 Plasmids 174-176
5' -Phosphoribosyl-5-aminoimidazole 63, P/O ratio 28
64 Poly-f3-hydroxybutyrate 130-132
carboxylic acid 64 Polymers (see macromolecules)
5' -Phosphoribosyl-4-carboxamide- Polypeptide synthesis 87-91
5-aminoimidazole 64 Polyphosphate 130, 131
5-formaminoimidazole 64 1,5-Poly (ribitolphosphate) 136
Phosphoribosyl-formimino- Polysaccharide 38, 39, 75, 76
5-aminoimidazole carboxamide Porin 112
ribonucleotide 54 Positive control 183
N-(5' -Phosphoribosyl)-anthranilate 53 Potassium, in nutrition I, 2
N' -(5' -phosphoribosyl)-ATP 54 transport 108
PQQ (see Methoxatin)
5' -Phosphoribosyl-glycineamide 63 Prephenate 52
356 Subject Index
Prephenate dehydrogenase 52 in amino acid synthesis 48

Product-inducer 180 oxidative decarboxylation of 18-20
L-Proline 51. 275 Pyruvate carboxylase 127-129. 174,236
D-Proline reductase 275 Pyruvate decarboxylase 212,213
Propanol 17 I Pyruvate dehydrogenase complex 19, 20,
PRPP (see 5-phosphoribosyl-l- 200, 221
pyrophosphate) distribution of 120
Propionate 17 L 242, 244. 247. 264. 271 regulation of 193
degradation of 149 Pyruvate family 43
Propionate fermentation 244 Pyruvate-ferredoxin oxidoreductase 225.
Propionyl-CoA 242. 249 226,229.235,243,251,263,312
Propionyl-CoA carboxylase 151 Pyruvate formate lyase 222, 236,
Protease 144 237-239
Protein 38, 39, 169 Pyruvate kinase 15-17, 200
Protein synthesis 87-91 Pyruvate oxidase 222
Protein transport I 13, 114 Pyruvate phosphate dikinase 128. 129,
Protocatechuate 158-161, 180 249
Protocatechuate-3 A-d ioxygenase 160 Pyruvate synthase 236, 312
Protocatechuate-4,5-oxygenase 161, 181
Proton gradient 264. 288. 290, 293. 305
Protonmotive force 31, 32, 223, 257. Quinate 158
258, 288, 290, 293. 305
equation 32 Radiorespirometry 58, I 15
in transport 106. 107 Reaction center 305,306,308,310
Protonophores 34 Rearrangement reactions 246
Proton pump 30, 290, 315 Redox control 194. 240
Proton translocation 26, 29-32 Redox potential (see oxidation-reduction
mechanisms of 30 potential)
Pseudomurein 136 Reductive tricarboxylic acid cycle 312
Pullulanase 142 Regulation. of enzyme activity 194-206
Purines. as growth factors 9 of enzyme synthesis 178-194
fermentation of 277-280 of nitrogenase 324-326
formula of 60 of ribulose-I ,5-bisphosphate
synthesis of 62-65 carboxylase 300
Purine nucleotides 60 Regulon 183
Purple membrane of halobacteria 315 Repeating oligosaccharide 80
Purple nonsulfur bacteria 30 I Replication of DNA 82-85
Purple sulfur bacteria 30 I Repression 185-194
Pyridoxine 10 Repressor 188
Pyrimidines. as growth factors 9 Reserve materials 75. 130
fermentation of 277 Respiration 23-35
formula of 60 anaerobic 122, 210
synthesis of 61. 62 with nitrate 122
Pyrimidine nucleotides 60 Respiratory chain 23-32
Pyrophosphatase 5, 263 components of 23-27
Pyrophosphate 5, 249 variation of composition 121
~-I-Pyrroline-5-carboxylate 51 Respiratory protection of nitrogenase 322
Pyruvate 17 Retinal 315
growth on 100. 102 Reverse electron transfer 288
Subject Index 357
Rhamnose 81. 183 Sequential induction 179

Rhodanese 291 Sequential interaction model 199
Riboflavin 10 L-Serine 49
Ribitol phosphate 136 Serine dehydratase 146
Ribokinase 99 Serine-glyoxylate transaminase 164
Ribonuclease (RNase) 145 Serine hydroxy methyl transferase 49, 50,
Ribonucleoside diphosphate reductase 66 164
Ribonucleosides 59 Serine-isocitrate lyase pathway 163-165
Ribonucleotides 59-66 Serine transacetylase 49
Ribose 98. 99 Shikimate 52, 158
Ribose-5-phosphate 56-58, 99, 297 Shikimate kinase 52
Ribose-5-phosphate isomerase 57, 99. Siderophores 110
219. 297 Sodium in nutrition 2
Ribosomes 90 in transpOli 107
Ribulokinase 99 pump 240-242
Ribulose 99 D-Sorbitol 171
Ribulose-I,5-bisphosphate L-Sorbose 171
carboxylase 296, 298-300 Spirilloxanthin 304
oxygenase reaction 299 Starch. degradation of 142
Ribulose-monophosphate cycle 165-167 Stereospecificity of citrate synthase 236
Ribulose-5-phosphate 56-58, 99, 118, lactate dehydrogenase 219
165. 297-300 Stickland reaction 273-277
Ribulose-5-phosphate-3-epimerase 57. Strict anaerobes 34, 208
217,219,297 Substrate-inducer 180
RNA Substrate-level phosphorylation 6, 16,
content of 38. 39 21,218,257,271
messenger 28, 190 Subtilisin 144
synthesis of 86 Succinate 21, 24, 237. 238, 244, 246
transfer 89, 190, 192 Succinate dehydrogenase 21, 33
turnover 39 Succinate fermentation 244
RNA polymerase 85,86. 136, 189 Succinate-propionate pathway 244
RNase 145 Succinate thiokinase 21. 312
Root nodules 323 Succinyl-CoA 21,151,245,246,260
Rotenon 34 N-Succinyl-2.6-diaminopimelate 47
Rubredoxin 155, 176,228,262 O-Succinyl homoserine 46
Rumen bacteria 268 O-Succinyl homoserine synthetase 195
Rusticyanin 293 Sucrose 144. 173
Sucrose phosphorylase 144
Sugars, degradation of 96-99, 114-120
Salicylate 159, 174 Sulfate. as electron acceptor
Sarcosine 275, 276 assimilation of 42, 43
Sedoheptulose-I,7-bisphosphate 166, 297 dissimilatory reduction of 260-265
Sedoheptu lose-I, 7-bisphosphate formation from sulfide 289,292,301,
aldolase 166, 297 308
Sedoheptulose-7-phosphate 56-58, 166. Sulfide, as electron donor 289, 292
297 ox idation of 289. 292, 30 I, 308
Selenium 3, 240, 269 Sulfide fermentation 260-265
Selenoprotein 240. 269, 277 Sulfite 42, 292
Sequential feedback inhibition 196 Sulfite oxidase 142, 292
358 Subject Index
Sulfite reductase 42, 263 Thiosulfate-oxidation enzyme 292

Sulfur. fermentation of 264 L-Threonine 46,48,191, 195,271
in nutrition 3 Threonine, fermentation of, 269
in photosynthesis 30 I Threonine deaminase 48, 194, 195
Sulfur-oxidizing bacteria 284, 285 Threonine dehydratase 146, 270, 271
Sulfur-oxidizing enzyme 292 Threonine synthase 191
Superoxide dismutase 34, 35, 209 Thymine 60
Symbiotic nitrogen fixation 323, 324 Thymidine-5' -monophosphate 60-62
Symport 106 Thymidylate snythetase 66
Synthesis of acetone 231 Toluene 159, 174
amino acids 43-55 Toluate 158
butanol 231 TPP (see thiamine pyrophosphate)
butyrate 229 Transaldolase 57,58, 118,219
carbohydrates 71-74 Transaminase 40, 44, 48, 146, 153,
DNA 83-85 192
diacetyl 221 Transcarboxylase 244, 245
ethanol 212 Transcription 85, 86, 188
lipids 65-71 Transglucosylation 76
macromolecules 73-91 Transhydrogenase 59, 243
pentose phosphates 55-59 Transketolase 56,57, 118, 166,219
proteins 87-91 Translation 87-91
RNA 86 Translocation, group 14, 15, 96, 108,
unsaturated fatty acids 68 109
vitamin C 171 Transpeptidation, in peptidoglycan
Syntrophic associations 266 synthesis 78
Transport 104-114
active 105-108
L-Talosaminuronic acid 136 of fructose 96
Tartronate semialdehyde 152 of glucose 13-15
Tartronate semialdehyde reductase 152 of iron 109,110
TeA cycle (see tricarboxylic acid cycle) of lactose 97
TCS 34 of proteins I 13-1 14
Teichoic acids 135, 136 passive 104
Tetrachlorosalicylanilide (see TCS) Tricarboxylic acid cycle 20-22, 99
Tetrahydrodipicolinate 47 distribution of 120
Tetrahydrofolic acid reductive 312
formula of 50 regulation of 193, 200
Tetrahydromethanopterin 256 Trimethylamine 168
formula of 255 Trimethylamine dehydrogenase 227
Tetrapeptides, of peptidoglycans 132, Trimethylamine-N-oxide 125
133 Trimethylamine-N-oxide reductase 124
Thermodynamic efficiency 233 Triosphosphate isomerase 15-17
Thiamine 10 Trp operon 188, 189
Thiamine pyrophosphate 18, 19,212, L-Tryptophan 53, 159, 196
226,239 Tryptophan, regulation of synthesis 188
formula of 19 Tryptophan synthetase 53, 54, 188
Thiolase 235 Tungsten
Thioredoxin 42, 66, 205 in nutrition 3
Thiosulfate-cleavage enzyme 291 L-Tyrosine 52, 157, 196
Subject Index 359
Ubiquinone (see coenzyme Q) Vitamin B I (see thiamine)

UDP (see uridine-5' -dis phosphate) Vitamin B, (see riboflavin)
UDP-N-acetylglucosamine 72, 78 Vitamin B6 (see pyridoxine)
UDP-N-acetylglucoseamine-3- Vitamin B" 10,245,258,271
enolpyruylether 72 formula of 247
UDP-N-acetylmuramic acid 72 Vitamin C, synthesis of 171
UD P-N -acetyImuramy Ipentapeptide 73, Vitamin K 10
74,78
UDP-galactose 71, 98
UDP-glucose 71, 98 Watson-Crick model 82, 83
UDP-glucose epimerase 71, 98 Wood-Werkman reaction 248
dUDP-phosphatase 66
Uncouplers 33, 34
Undecaprenyl phosphate 77 Xanthine 278
Unsaturated fatty acids 68, 69 Xanthine amidohydrolase 278
Uptake of substrates see transport Xanthine dehydrogenase 124, 227, 278
Uracil 60, 279 Xanthine oxidase 151
Urate 152, 277 Xanthylic acid 64
Urease 151 Xylene 174
Ureidoglycine aminohydrolase 152 Xylanase 144
Ureidoglycolase 152 Xylose 98, 99
4-Ureido-5-imidazole carboxylate 278 Xylose isomerase 99
4-Ureido-5-imidazole carboxylate Xylulokinase 99
amidohydrolase 278 Xylulose 99
Uricase 152 Xylulose-monophosphate cycle 167-168
Uridine-5' -monophosphate 60-62 Xylulose-5-phosphate 56-58,99, 167,
Uridine-5' -diphosphate 65 218,297
Urocanase 147 XyluJose-5-phosphate
Urocanate 147 phosphoketolase 219
cis- Yaccenate 68 YAW 222

Vanillate 158 Y", 222
L-Valine 40,48, 192
Vesicles of membranes 28
Vitamins as growth factors 9 Zinc, in nutrition 3

C-Bacterial Metabolism-Springer-Verlag New York (1986)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

C-Bacterial Metabolism-Springer-Verlag New York (1986)

Uploaded by

Copyright:

Available Formats

Bacterial Metabolism

With 204 Figures

Library of Congress Cataloging in Publication Data

Printed on acid-free paper.

1979, 1986 Springer-Verlag Berlin Heide1berg

Typeset by Polyglot Pte, Ltd., Singapore.

ISBN 978-1-4612-7003-4 ISBN 978-1-4612-1072-6 (eBook)

Gottingen, 1985 GERHARD GOTISCHALK

I am particularly indebted to Joan Macy, Lynne Quandt, Jan Andreesen and

G6ttingen, 1978 GERHARD GOTTSCHALK

VI. Ribonucleotides and Deoxyribonucleotides 59

II. Regulation of Enzyme Activity 194

Further Reading 327

Index of Organisms 337

Subject Index 341

I. Major and Minor Bioelements

element source function in metabolism

cytochromes and iron-sulfur proteins. Sodium ions, in relatively high

element source function in metabolism

Zn Zn H present in alcohol dehydrogenase, alkaline phosphatase,

Besides these 12 major bioelements, organisms require a number of

however, depend on the availability of reduced sulfur compounds. Metha-

II. The Two Basic Mechanisms of

ATP + HzO \" /' ) ADP + Pi

Because of the high-energy phosphoryl bonds, ATP is an excellent

Substrate-level phosphorylation is the second mechanism of ATP gen-

The further metabolism of such organic - P compounds is coupled to the

III. Nutrients as Energy Sources

Table 1.3. The two types of phototrophic metabolism

type electron donor carbon source examples

photolithotrophy CO 2 plants, cyanobacteria

The carbon source commonly used by phototrophs is CO 2 , and organisms

Table 1.4. The two types of chemotrophic metabolism

electron electron carbon

chemoorganotrophy organic Oz organic pseudomonads,

chemoorganotrophs. This group includes aerobes and anaerobes. The

C-autotrophs. However, many chemolithotrophs are not restricted to this

IV. Growth Factor Requirements

compound function in metabolism

p-aminobenzoic acid precursor of tetrahydrofolic acid, a coenzyme involved in

4. Phototrophic bacteria, which use inorganic electron donors such as

Escherichia coli belongs to the group of facultatively anaerobic bacteria. It

In order to utilize the energy released during glucose oxidation effec-

V PEP A. Transport of D-glucose into the cell

2 acetyl-CoA D. Oxidation of the acetyl moiety of

I. Transport of D-Glucose into the

The phosphate donor in this reaction is phosphoenolpyruvate (PEP)

Figure 2.1. Transport of glucose by the PEP: glucose phosphotransferase system.

coupled to a conversion of the substrate (glucose) into a derivative of the

II. Degradation of Glucose-6-Phosphate

Table 2.1. Activity of the enzymes of the Embden-Meyerhof-Parnas

average specific activity

glucose phosphate isomerase 1

Consequently, if only little dihydroxyacetonephosphate is needed by the

Figure 2.3. Uptake of glucose and breakdown of glucose-6-phosphate to 2 pyruvate

The sum of the reactions discussed thus far is:

III. Oxidative Decarboxylation of Pyruvate

sum: CH 3 -CO-COOH + CoA + NAD+ .... CH 3 -CO-CoA + CO 2 + NADH + W

formed is then released as acetyl-CoA and under catalysis of E 3 the

IV. Oxidation of Acetyl-CoA via the

succinyl-CoA + Pi + ADP lhiokinase) succinate + ATP + CoA

available in this oxidation is conserved in the form of reduced coenzymes.

glucose + 8NAD+ + 2NADP+ + 2FAD + 4ADP + 4P j ~