Advances in bacterial pathogen biology 1st

Edition Poole pdf download

https://ebookgate.com/product/advances-in-bacterial-pathogen-biology-1st-edition-poole/

★★★★★ 4.8/5.0 (43 reviews) ✓ 205 downloads ■ TOP RATED

"Perfect download, no issues at all. Highly recommend!" - Mike D.

DOWNLOAD EBOOK
 Advances in bacterial pathogen biology 1st Edition Poole pdf
download

TEXTBOOK EBOOK EBOOK GATE

Available Formats

■ PDF eBook Study Guide TextBook

EXCLUSIVE 2025 EDUCATIONAL COLLECTION - LIMITED TIME

INSTANT DOWNLOAD VIEW LIBRARY

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Advances in Microbial Physiology Vol 46 1st Edition Robert

K. Poole

https://ebookgate.com/product/advances-in-microbial-physiology-
vol-46-1st-edition-robert-k-poole/

ebookgate.com

Advances in Microbial Physiology Vol 54 1st Edition Robert

K. Poole (Ed.)

https://ebookgate.com/product/advances-in-microbial-physiology-
vol-54-1st-edition-robert-k-poole-ed/

ebookgate.com

Advances in Microbial Physiology Vol 55 1st Edition Robert

K. Poole (Ed.)

https://ebookgate.com/product/advances-in-microbial-physiology-
vol-55-1st-edition-robert-k-poole-ed/

ebookgate.com

Bacterial Invasion of Host Cells Advances in Molecular and

Cellular Microbiology 1st Edition Richard J. Lamont
(Editor)
https://ebookgate.com/product/bacterial-invasion-of-host-cells-
advances-in-molecular-and-cellular-microbiology-1st-edition-richard-j-
lamont-editor/
ebookgate.com
Advances in Marine Biology Vol 51 1st Edition Alan J.
Southward

https://ebookgate.com/product/advances-in-marine-biology-vol-51-1st-
edition-alan-j-southward/

ebookgate.com

Population Biology and Criticality From Critical Birth

Death Processes to Self Organized Criticality in Mutation
Pathogen Systems 1st Edition Nico Stollenwerk
https://ebookgate.com/product/population-biology-and-criticality-from-
critical-birth-death-processes-to-self-organized-criticality-in-
mutation-pathogen-systems-1st-edition-nico-stollenwerk/
ebookgate.com

Concepts In Bacterial Virulence Contributions to

Microbiology 1st Edition W. Russell

https://ebookgate.com/product/concepts-in-bacterial-virulence-
contributions-to-microbiology-1st-edition-w-russell/

ebookgate.com

Catastrophes Hilary W. Poole

https://ebookgate.com/product/catastrophes-hilary-w-poole/

ebookgate.com

Advances in Enzymology and Related Areas of Molecular

Biology Volume 77 Eric J. Toone

https://ebookgate.com/product/advances-in-enzymology-and-related-
areas-of-molecular-biology-volume-77-eric-j-toone/

ebookgate.com
Academic Press is an imprint of Elsevier
525 B Street, Suite 1800, San Diego, CA 92101-4495, USA
225 Wyman Street, Waltham, MA 02451, USA
32 Jamestown Road, London NW1 7BY, UK
The Boulevard, Langford Lane, Kidlington, Oxford, OX5 1GB, UK

First edition 2014

Copyright © 2014 Elsevier Ltd. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage and
retrieval system, without permission in writing from the publisher. Details on how to seek
permission, further information about the Publisher’s permissions policies and our
arrangements with organizations such as the Copyright Clearance Center and the Copyright
Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by
the Publisher (other than as may be noted herein).

Notices
Knowledge and best practice in this field are constantly changing. As new research and
experience broaden our understanding, changes in research methods, professional practices,
or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described
herein. In using such information or methods they should be mindful of their own safety and
the safety of others, including parties for whom they have a professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of
products liability, negligence or otherwise, or from any use or operation of any methods,
products, instructions, or ideas contained in the material herein.

ISBN: 978-0-12-800142-4
ISSN: 0065-2911

For information on all Academic Press publications

visit our website at store.elsevier.com
CONTRIBUTORS

Jody Aberdein
Department of Infection and Immunity, University of Sheffield Medical School and Sheffield
Teaching Hospitals, Sheffield, United Kingdom
Elaine Allan
Department of Microbial Diseases, UCL Eastman Dental Institute, University College
London, London, United Kingdom
Sabela Balboa
Department of Infection and Immunity, University of Sheffield, Sheffield, United Kingdom,
and Departamento de Microbiologı́a y Parasitologı́a, Universidad de Santiago de
Compostela, Santiago de Compostela, Spain
Michael Berney
Department of Microbiology and Immunology, Albert Einstein College of Medicine, Bronx,
New York, USA
Annelie Brauner
Department of Microbiology, Tumor and Cell Biology, Division of Clinical Microbiology,
Karolinska Institutet and Karolinska University Hospital, Stockholm, Sweden
Jennifer S. Cavet
Faculty of Life Sciences, University of Manchester, Manchester, United Kingdom
Joby Cole
Department of Infection and Immunity, University of Sheffield Medical School and Sheffield
Teaching Hospitals, Sheffield, United Kingdom
Gregory M. Cook
Department of Microbiology and Immunology, Otago School of Medical Sciences,
University of Otago, Dunedin, and Maurice Wilkins Centre for Molecular Biodiscovery,
The University of Auckland, Auckland, New Zealand
David H. Dockrell
Department of Infection and Immunity, University of Sheffield Medical School and Sheffield
Teaching Hospitals, Sheffield, United Kingdom
C.W. Ian Douglas
Integrated BioSciences Group, School of Clinical Dentistry, University of Sheffield,
Sheffield, United Kingdom
Thomas Farmilo
Integrated BioSciences Group, School of Clinical Dentistry, University of Sheffield,
Sheffield, United Kingdom
Andrew M. Frey
Integrated BioSciences Group, School of Clinical Dentistry, University of Sheffield,
Sheffield, United Kingdom

ix
x Contributors

Chris Greening
Department of Microbiology and Immunology, Otago School of Medical Sciences,
University of Otago, Dunedin, New Zealand
Kiel Hards
Department of Microbiology and Immunology, Otago School of Medical Sciences,
University of Otago, Dunedin, New Zealand
Helen E. Jesse
Faculty of Life Sciences, University of Manchester, Manchester, United Kingdom
Jamil Jubrail
Department of Infection and Immunity, University of Sheffield Medical School and Sheffield
Teaching Hospitals, Sheffield, United Kingdom
Rebecca Lowry
Department of Infection and Immunity, University of Sheffield, Sheffield, United Kingdom
Petra Lüthje
Department of Microbiology, Tumor and Cell Biology, Division of Clinical Microbiology,
Karolinska Institutet and Karolinska University Hospital, Stockholm, Sweden
Peter Mullany
Department of Microbial Diseases, UCL Eastman Dental Institute, University College
London, London, United Kingdom
Kathryn Naylor
Integrated BioSciences Group, School of Clinical Dentistry, University of Sheffield,
Sheffield, United Kingdom
Jennifer L. Parker
Department of Infection and Immunity, University of Sheffield, Sheffield, United Kingdom
Chatchawal Phansopa
Integrated BioSciences Group, School of Clinical Dentistry, University of Sheffield,
Sheffield, United Kingdom
Adam P. Roberts
Department of Microbial Diseases, UCL Eastman Dental Institute, University College
London, London, United Kingdom
Ian S. Roberts
Faculty of Life Sciences, University of Manchester, Manchester, United Kingdom
Jonathan G. Shaw
Department of Infection and Immunity, University of Sheffield, Sheffield, United Kingdom
Graham P. Stafford
Integrated BioSciences Group, School of Clinical Dentistry, University of Sheffield,
Sheffield, United Kingdom
 CHAPTER ONE

Energetics of Pathogenic Bacteria

and Opportunities for Drug
Development
Gregory M. Cook*,†,1,2, Chris Greening*,1, Kiel Hards*,1,
Michael Berney{,1
*Department of Microbiology and Immunology, Otago School of Medical Sciences, University of Otago,
Dunedin, New Zealand
†
Maurice Wilkins Centre for Molecular Biodiscovery, The University of Auckland, Auckland, New Zealand
{
Department of Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, New York, USA
1
These authors contributed equally to this work.
2
Corresponding author: e-mail address: [email protected]

Contents
1. Introduction 2
2. Bacterial Energetics as a Target Space for Drug Development 2
2.1 Generation of the proton motive force: An essential property of all
bacterial cells 3
2.2 Diversity and flexibility of electron transport chains in bacteria 5
2.3 Primary respiratory dehydrogenases 7
2.4 Terminal respiratory reductases 18
2.5 Generators of sodium motive force in bacterial pathogens 32
2.6 ATP homeostasis and the F1Fo-ATP synthase: A clinically approved drug target 35
3. Conclusions and Future Perspectives 40
Acknowledgements 43
References 44

Abstract
The emergence and spread of drug-resistant pathogens and our inability to develop new
antimicrobials to overcome resistance has inspired scientists to consider new targets for
drug development. Cellular bioenergetics is an area showing promise for the develop-
ment of new antimicrobials, particularly in the discovery of new anti-tuberculosis drugs
where several new compounds have entered clinical trials. In this review, we have exam-
ined the bioenergetics of various bacterial pathogens, highlighting the versatility of elec-
tron donor and acceptor utilisation and the modularity of electron transport chain
components in bacteria. In addition to re-examining classical concepts, we explore
new literature that reveals the intricacies of pathogen energetics, for example, how Sal-
monella enterica and Campylobacter jejuni exploit host and microbiota to derive powerful
electron donors and sinks; the strategies Mycobacterium tuberculosis and Pseudomonas

Advances in Microbial Physiology, Volume 65 # 2014 Elsevier Ltd 1

ISSN 0065-2911 All rights reserved.
http://dx.doi.org/10.1016/bs.ampbs.2014.08.001
2 Gregory M. Cook et al.

aeruginosa use to persist in lung tissues; and the importance of sodium energetics and
electron bifurcation in the chemiosmotic anaerobe Fusobacterium nucleatum.
A combination of physiological, biochemical, and pharmacological data suggests that,
in addition to the clinically-approved target F1Fo-ATP synthase, NADH dehydrogenase
type II, succinate dehydrogenase, hydrogenase, cytochrome bd oxidase, and men-
aquinone biosynthesis pathways are particularly promising next-generation drug tar-
gets. The realisation of cellular energetics as a rich target space for the development
of new antimicrobials will be dependent upon gaining increased understanding of
the energetic processes utilised by pathogens in host environments and the ability to
design bacterial-specific inhibitors of these processes.

1. INTRODUCTION
The majority of current antimicrobials were developed during the
golden era of antimicrobial discovery. These compounds target a number of
essential processes for the growth of microbial cells, including peptidoglycan
biosynthesis, RNA and protein synthesis, DNA replication, and folic acid
metabolism. During this period, antimicrobial use became widespread, not
only in hospitals but also in agricultural environments. As quickly as new anti-
microbials were developed, however, resistance followed increasing the
demand for new derivatives through optimisation of existing molecular scaf-
folds. The burden of antimicrobial resistance was further compounded by
the lack of new drugs with unique targets to overcome resistance and by the
increasing cost of antimicrobial discovery and development. The number of
new antibiotic approvals by the FDA continues to decline contributing to
the withdrawal of pharmaceutical companies in this area (Boucher et al., 2013).
To address the emergence and spread of drug-resistant bacterial pathogens,
new drug targets and drugs with a novel mode of action are urgently required
to expand our antimicrobial armoury. The development of narrow spectrum
agents to prevent widespread resistance developing remains a priority. A key
to the development of the next generation of antimicrobials will be increased
understanding of how new targets function in the physiological context of the
pathogen. Deciphering the essential and non-essential roles of these targets in
response to the host environment will be an important question to address.

2. BACTERIAL ENERGETICS AS A TARGET SPACE FOR

DRUG DEVELOPMENT
A major structural component of bacterial cells is the cytoplasmic
membrane made up of a lipid bilayer that forms a continuous barrier around
Energetics of Pathogenic Bacteria 3

the cell. The cytoplasmic membrane imparts structure to the cell and allows
for the selective (filter) passage of nutrients and wastes into and out of
the cell. The membrane also plays an essential role in cellular homeostasis
and energy transduction. Several new antimicrobials have been developed
that target the bacterial membrane (e.g. daptomycin and lipoglycopeptides),
leading to disruption of bacterial membrane integrity, but this target
space has remained largely unexplored (Hurdle, O’Neill, Chopra, &
Lee, 2011).

2.1. Generation of the proton motive force: An essential

property of all bacterial cells
All bacteria require a proton motive force (pmf ) to grow and remain viable
under replicating and non-replicating conditions. During respiration,
energy is conserved by the generation of a pmf across a proton-impermeable
membrane. The electron transport chain components are membrane-bound
and asymmetrically arranged across the membrane to achieve net consump-
tion of protons from the cytoplasm and net release of protons on the outside
the cell. The pmf (electrochemical potential) consists of two gradients: an
electrical potential (△ψ), due to the charge separation across the membrane
(positiveoutside/negativeinside) and a chemical transmembrane gradient of
protons (△pH, acidicoutside/alkalineinside). At neutral pH, the pmf is predom-
inantly in the form of a △ψ, but as the external pH drops, the △pH increases,
and the △ψ decreases to maintain a constant pmf. Dissipation of the pmf leads
to a rapid loss of cell viability and cell death.
A variety of mechanisms are used to generate the pmf in bacteria (Fig. 1).
In obligately aerobic bacteria, the generation of a pmf is mediated primarily
by the proton-pumping components of the electron transport chain (Fig. 1,
mechanism 4). In facultative anaerobes, when alternative electron acceptors
are available (e.g. nitrate and fumarate), proton release is coupled to a ter-
minal reductase (e.g. nitrate reductase) via a pmf redox-loop mechanism
( Jormakka, Byrne, & Iwata, 2003b; Fig. 1, mechanism 2). Under strictly fer-
mentative conditions, the F1Fo-ATP synthase can operate as a reversible
ATP-driven proton pump to generate the pmf (Dimroth & Cook, 2004;
Fig. 1, mechanism 3). Furthermore, in some bacteria, end-products (e.g. lac-
tate) efflux can generate a pmf (Otto, Sonnenberg, Veldkamp, & Konings,
1980; Fig. 1, mechanism 1). The flexibility of respiration in bacteria under
anaerobic conditions is further highlighted by the discovery that endoge-
nous phenazine production by Pseudomonas aeruginosa enhances anaerobic
4 Gregory M. Cook et al.

Figure 1 Mechanisms (1–4) by which a proton motive force can be generated in bacteria.
(1) Co-transport of protons driven by solute (lactate) symport into the periplasm. (2)
Redox-loop separation of charge; quinol oxidation results in proton release into the peri-
plasm by virtue of quinol site proximity to the periplasm, while electrons are transferred
to reduce a terminal electron acceptor in the cytoplasm that results in neutralisation of
charge. (3) Proton export driven by ATP hydrolysis, i.e., ATP synthase working in the
reverse direction. (4) Proton translocation mediated by primary proton-pumping
complexes.

survival through maintenance of the pmf (and ATP production) via a redox
homeostasis mechanism (Glasser, Kern, & Newman, 2014).
There are a wide range of compounds that target the pmf in bacteria
(Fig. 2), including agents that inhibit the major proton pumps (e.g. rote-
none) (Fig. 2, mechanism 3) and those that facilitate proton transport
through the cytoplasmic membrane (protonophores, e.g. carbonyl cyanide
m-chlorophenyl hydrazine—CCCP) (Fig. 2, mechanism 5). The majority of
protonophores are non-specific and functional in both prokaryotic and
eukaryotic cell membranes. Individual components of the pmf can be col-
lapsed using specific inhibitors. For example, the △ψ can be collapsed by
compounds that catalyse electrogenic cation transport across the cell mem-
brane (e.g. valinomycin) Valinomycin is a dodecadepsipeptide that forms a
macrocyclic molecule allowing for rapid K+ movement down its electro-
chemical gradient (Fig. 2, mechanism 1). The chemical transmembrane gra-
dient of protons (△pH) can be collapsed by nigericin through its K+/H+
antiporter (electroneutral) activity (Fig. 2, mechanism 2). Nigericin has sim-
ilar properties to monensin, a Na+/H+ exchanger widely used in livestock as
a feed additive. Gramicidin is a channel-forming ionophore, making the
membrane more permeable to ions (Fig. 2, mechanism 3).
Some bacterial pathogens generate a considerable △pH in response to
acidification of host tissues (e.g. Helicobacter pylori, Salmonella enterica, and
Streptococcus pneumoniae), and collapsing the pH gradient would be an
Energetics of Pathogenic Bacteria 5

Figure 2 Traditional inhibitors of proton motive force generation. (1) Valinomycin is an

ionophore, selective for potassium ions, which equilibrates the potassium gradient—
dissipating the Δψ (electrogenic). (2) Nigericin is a hydrophobic weak carboxylic acid,
which can traverse the membrane as its either protonated acid or neutral salt. It dissi-
pates chemical gradients (i.e. ΔpH) but maintains the charge (one positive charge
exchanged for one positive charge—electroneutral). (3) Carbonyl cyanide
m-chlorophenyl hydrazine (CCCP) is an electrogenic protonophore. CCCP
is driven
to the periplasm by the Δψ, while CCCPH is driven to the cytoplasm by the ΔpH. It
can equilibrate both Δψ and ΔpH. (4) Gramicidin is a channel-forming ionophore, mak-
ing the membrane more permeable to ions. (5) Rotenone inhibits primary proton
pumping—preventing the initial generation of a proton motive force.

effective strategy in acidic tissues to eradicate these bacteria (Hall, Karem, &
Foster, 1995; Matin, Zychlinsky, Keyhan, & Sachs, 1996). The pmf has
recently been screened as a target for methicillin-resistant Staphylococcus
aureus using high-throughput screening to identify compounds that dissipate
individual components of the pmf, i.e., the △ψ or △pH and synergistic com-
binations thereof (Farha, Verschoor, Bowdish, & Brown, 2013).

2.2. Diversity and flexibility of electron transport chains in

bacteria
The main pathogens discussed in this review are summarised in Table 1. The
electron transport chains both within and between these bacteria show a
remarkable diversity with regard to both electron donor and electron accep-
tor utilisation, enabling growth and persistence in a wide variety of environ-
mental niches (Fig. 3). Bacteria are able to use a range of primary
dehydrogenases to deliver electrons from central metabolism into the respi-
ratory chain to generate energy. These electrons pass through various redox
carriers to the quinone/quinol pool. In bacteria, the electron transport chain
is often branched with multiple routes to terminal respiratory oxidases or
reductases (Fig. 3). For example, Escherichia coli uses a low-affinity (μM
for oxygen) proton-pumping cytochrome bo3 (haem–copper) oxidase
Table 1 Classification and characteristics of the pathogens discussed in this review
Organism Classification Metabolism Major diseases Primary tissues
Escherichia coli (pathogenic strains) γ-Proteobacteria Heterotroph Gastroenteritis Gastrointestinal tract
Enterobacteriales Facultative anaerobe Urinary tract infections Urinary tract
Salmonella enterica γ-Proteobacteria Heterotroph Gastroenteritis Gastrointestinal tract
Enterobacteriales Facultative anaerobe Typhoid fever
Pseudomonas aeruginosa γ-Proteobacteria Heterotroph Opportunistic infections (e.g. pneumonia) Cystic fibrosis lungs
Pseudomonadales Facultative anaerobe
Neisseria gonorrhoeae β-Proteobacteria Heterotroph Gonorrhoea Urinary tract
Neisseria meningtidis Neisserales Facultative aerobe Meningitis Meninges
Campylobacter jejuni ε-Proteobacteria Heterotroph Gastroenteritis Gastrointestinal tract
Campylobacterales Microaerobe
Helicobacter pylori ε-Proteobacteria Heterotroph Stomach ulcers Stomach
Campylobacterales Microaerobe Stomach cancer
Staphylococcus aureus Firmicutes Heterotroph Opportunistic infections (e.g. skin infections) Skin
Bacilliales Facultative anaerobe Respiratory tract
Mycobacterium tuberculosis Actinobacteria Heterotroph Tuberculosis Lungs
Actinomycetales Obligate aerobe
Fusobacterium nucleatum Fusobacteria Heterotroph Periodontitis Oral cavity
Fusobacteriales Obligate anaerobe Lemierre’s syndrome
Treponema pallidum Spirochaetes Heterotroph Syphilis Urinary tract
Spirochaetales Obligate anaerobe Yaws Skin
It only lists those pathogens where a relatively complete overview of their energetics is provided.
Energetics of Pathogenic Bacteria 7

Figure 3 Generalised schematic overview of relevant electron transfer components. Com-

plexes indicated in blue oxidise various substrates to reduce quinones. The resulting
quinol molecules can be oxidised to result in reduction of various terminal electron
acceptors, mediated by the complexes indicated in green. For some electron transfer
pathways intermediate complexes and molecules exist, for example, complex III will
generally reduce cytochrome c, which will serve as the electron donor for complex
IV. The complexes used, types of quinones, and intermediates thereof are highly vari-
able between genera. Only complexes relevant to this review are indicated.

growing at high oxygen tensions, but switches to a high-affinity (nM for

oxygen) non-proton-translocating cytochrome bd oxidase when growing
at low oxygen tensions (Cotter, Chepuri, Gennis, & Gunsalus, 1990;
D’Mello, Hill, & Poole, 1995, 1996; Fig. 3). In S. enterica, the electron trans-
port chain shows considerable diversity in response to oxygen tension and
will be highlighted throughout this review.
A number of compounds have been shown to inhibit the major compo-
nents of mitochondrial and bacterial electron transport chains (Fig. 4). How-
ever, few if any studies have assessed how specific these compounds are
across different bacterial genera.

2.3. Primary respiratory dehydrogenases

2.3.1 NADH dehydrogenases: The roles of bacterial NDH-1 and NDH-2
In many bacterial pathogens, the major entry point to the electron transport
chain is the transfer of electrons from reduced nicotinamide dinucleotide
(NADH) (reduced by the oxidation of organic carbon) to quinones (e.g.
Figure 4 Diversity of electron transport chain inhibitors. Structural surface representations of electron transport chain components are indi-
cated where possible. Selected inhibitors of these complexes are indicated with flathead arrows and do not reflect the binding site of the
inhibitors. Crystal structures were obtained from RCSB protein data bank from the following accession numbers: complex I, 3M9S; Ndh-2,
4NWZ; complex II, 2WDV; complex III, 3H1J; complex IV, 3ASN; and complex V, 4B2Q. Models were generated using the PyMOL molecular
graphics system.
Energetics of Pathogenic Bacteria 9

ubiquinone or menaquinone; Fig. 3). Three different types of respiratory

NADH dehydrogenases have been identified and characterised on the basis
of reaction mechanism, subunit composition, and protein architecture
(Kerscher, Drose, Zickermann, & Brandt, 2008): the proton-pumping type
I NADH dehydrogenase (NDH-1, complex I), the non-proton-pumping
type II NADH dehydrogenase (NDH-2; Fig. 4), and the sodium-pumping
NADH dehydrogenase (NQR, discussed in Section 2.5). Homologous to
mitochondrial complex I, bacterial NDH-1 is encoded by the nuo operon
and transfers electrons to quinone, conserving energy by translocating pro-
tons across the membrane to generate a pmf. This multimeric enzyme uses
flavin adenine dinucleotide (FAD) and nine iron–sulphur clusters to trans-
port electrons from NADH to the quinone pool. The release of the two
electrons during the NADH oxidation produces enough energy to pump
four protons across the membrane to generate a pmf (Baradaran,
Berrisford, Minhas, & Sazanov, 2013).
NDH-2 is more relevant to drug discovery. This small cytoplasmically
oriented monotopic membrane protein (40–60 kDa; Fig. 4) catalyses
electron transfer from NADH via the flavin cofactor to quinone (Heikal
et al., 2014). NDH-2 enzymes are widespread in bacteria and, while also
encoded in some eukaryotes (Melo, Bandeiras, & Teixeira, 2004), have
not been reported in mammalian mitochondria. This has resulted in the pro-
posal that they may represent a potential drug target for the treatment of path-
ogenic bacteria (Rao, Alonso, Rand, Dick, & Pethe, 2008; Teh, Yano, &
Rubin, 2007; Warman et al., 2013; Weinstein et al., 2005; Yano, Li,
Weinstein, Teh, & Rubin, 2006), as well as protozoa (Biagini,
Viriyavejakul, O’Neill, Bray, & Ward, 2006; Warman et al., 2013).
In many pathogens, there are copies of both types NDH-1 and NDH-2
in the genome (Melo et al., 2004). In the enteric pathogens E. coli and
S. enterica, these enzymes are differentially expressed, with NDH-2 primarily
being synthesised aerobically and NDH-1 being active during anaerobic res-
piration (Calhoun, Oden, Gennis, de Mattos, & Neijssel, 1993; Unden &
Bongaerts, 1997). One potential explanation for the dominant role of
NDH-2, even in the presence of NDH-1, is that lack of proton translocation
may be desirable during some conditions. NADH oxidation by NDH-2
would not be impeded by a high pmf, as would be the case with NDH-1,
which could ultimately slow metabolic flux due to back-pressure on the sys-
tem; NDH-2-mediated NADH oxidation would therefore allow for a
higher metabolic flux and increased carbon flow into biosynthetic pathways
and ultimately higher rates of ATP synthesis, at the expense of low energetic
10 Gregory M. Cook et al.

efficiency of the respiratory chain. The use of NDH-1 could also be dele-
terious during aerobic conditions due to the increased production of reactive
oxygen species (ROS) generated via this complex (Knuuti, Belevich,
Sharma, Bloch, & Verkhovskaya, 2013). By contrast, when the oxidant
for terminal oxidases is unavailable under anoxic conditions, alternative
mechanisms to generate pmf are required and hence enterobacteria select
for the energetic efficiency of NDH-1 (Unden & Bongaerts, 1997).
There is significant variety in the physiological roles of the NADH dehy-
drogenases in different pathogens. Mycobacterium tuberculosis encodes an
NDH-1 complex and two copies of NDH-2 (Ndh and NdhA; Cook,
Hards, Vilchèze, Hartman, & Berney, 2014). Transcriptome and biochem-
ical studies indicate that NDH-2 activity is responsible for the majority of
NADH oxidation in mycobacteria (Shirude et al., 2012; Weinstein et al.,
2005; Yano et al., 2006). Mutagenesis screens suggest that, whereas NdhA
is dispensable, Ndh is essential for growth in vitro and cannot be compensated
for (Griffin et al., 2011; Sassetti, Boyd, & Rubin, 2003). The role of NDH-1
is more enigmatic. In contrast to S. enterica, the complex is downregulated
during infection, hypoxia, and starvation in M. tuberculosis (Betts, Lukey,
Robb, McAdam, & Duncan, 2002; Schnappinger et al., 2003; Shi et al.,
2005). The enzyme is dispensable for growth and survival in vitro (Griffin
et al., 2011), though nuoG has nevertheless been shown to be critical for vir-
ulence in murine and macrophage models (Velmurugan et al., 2007).
S. aureus also encodes two copies of NDH-2, but lacks NDH-1; the iso-
zymes are kinetically distinct, though the differential roles of the individual
enzymes are currently unclear (Schurig-Briccio, Yano, Rubin, & Gennis,
2014). Among other pathogens, the minimal genome of Mycobacterium leprae
encodes a single NDH-2, which is predicted to be the major input into the
respiratory chain of this organism (Cole et al., 2001). A contrasting scenario
is presented by Campylobacter jejuni and H. pylori, which lack NDH-2 and
instead encode NDH-1 variants (Chen, Andersen, Zhai, & Kharazmi, 1999;
Smith, Finel, Korolik, & Mendz, 2000). These organisms have low rates of
NADH oxidation (Hoffman & Goodman, 1982) and recent studies suggest
that flavodoxin is in fact the reductant for their NDH-1 variants
(Weerakoon & Olson, 2008). Consistently, the electron-input subunits
of these complexes are novel in both bacteria and have been shown to
be essential for growth of C. jejuni (Smith et al., 2000; Weerakoon &
Olson, 2008).
Several classes of compounds are proposed to target NDH-2 at micro-
molar concentrations (e.g. phenothiazine analogues, platanetin, and
Energetics of Pathogenic Bacteria 11

quinolinyl pyrimidines; Fig. 4; Shirude et al., 2012; Weinstein et al., 2005;

Yano et al., 2006). Despite poor activity against NDH-2, drugs of the phe-
nothiazine family (trifluoroperazine and chlorpromazine) have potent activ-
ity in vitro against drug-susceptible and drug-resistant M. tuberculosis strains
(Amaral, Kristiansen, Abebe, & Millett, 1996; Amaral, Viveiros, &
Kristiansen, 2006; Ordway et al., 2003). A phenothiazine analogue was also
tested in a mouse model of acute M. tuberculosis infection and was found to
reduce by 90% the M. tuberculosis bacterial load in the lungs after 11 days of
treatment compared to a 3- to 4-log reduction in colony forming units with
the isoniazid (INH) or rifampin control (Weinstein et al., 2005). From a
library of microbial products, two compounds, scopafungin and gramicidin
S, were identified as inhibitors of Mycobacterium smegmatis NDH-2, with
IC50 values better than trifluoperazine (Mogi, Matsushita, et al., 2009).
The mechanism of phenothiazine action remains unclear, but does not
appear to involve competitive inhibition of NADH or quinone binding
(Weinstein et al., 2005; Yano et al., 2006). Phenothiazines have been shown
to inhibit the NDH-2 isozymes of S. aureus (Amaral et al., 2006; Ordway,
Viveiros, Leandro, Arroz, & Amaral, 2002; Ordway et al., 2002; Schurig-
Briccio et al., 2014). The authors propose that, in addition to inhibition
of NDH-2 activity at low micromolar concentrations, phenothiazines are
uncouplers of oxidative phosphorylation (Schurig-Briccio et al., 2014).
The mechanism of uncoupling remains unclear.
Despite the potential of NDH-2 as a drug target, no potent nanomolar
inhibitors of NDH-2 have been reported. The first high-resolution structure
of a bacterial NDH-2 was recently solved, revealing unique binding sites for
quinone and NADH, allowing concomitant oxidation of NADH from the
aqueous cytoplasm and reduction of hydrophobic quinone in the membrane
with both substrates accessing the FAD cofactor sequentially (Heikal et al.,
2014). The structure shows a homodimeric organisation with a unique
dimer interface, but whether the enzyme forms a higher molecular weight
species physiologically remains to be ascertained. The enzyme is localised to
the cytoplasmic membrane by two separated C-terminal membrane-
anchoring regions that are essential for membrane localisation and FAD
binding, but not dimerisation. This structure of bacterial NDH-2 provides
for the first time a molecular framework for the development of inhibitors of
the bacterial enzyme (Heikal et al., 2014). That said, while NDH-2 appears a
very attractive target for drug development, this may not be the case in all
instances. For example, reduction in NDH-2 activity has been linked to iso-
niazid (INH) and ethionamide resistance in both slow- and fast-growing
12 Gregory M. Cook et al.

mycobacteria (Vilcheze et al., 2005), and phenothiazines have been shown

to be antagonistic with INH (Warman et al., 2013). Therefore, the devel-
opment of NDH-2 inhibitors will have to ascertain that interference with
current drug therapy does not occur.
An alternative approach is to target the quinones, the membrane-
diffusible redox carriers that accept the electrons from NADH dehydroge-
nases and other primary dehydrogenases; this would theoretically prevent
compensation between respiratory complexes when a specific complex is
targeted. There is particular potential for the development of
antimycobacterials that target the maturation factors encoding men-
aquinone; as the sole quinone in M. tuberculosis, synthesis of this cofactor
is essential for growth and recovery from persistence (Dhiman et al.,
2009). There is promise in developing inhibitors against MenA, which catal-
yses the prenylation step in this mycobacterial menaquinone biogenesis
pathway (Kurosu & Crick, 2009); several derivatives of the natural product
aurachin RE have recently been shown to be bactericidal against drug-
resistant M. tuberculosis in vitro (Debnath et al., 2012). There may also be
merit in targeting the novel menaquinone biosynthesis pathways of
H. pylori and C. jejuni (Hiratsuka et al., 2008; Seto et al., 2008).

2.3.2 Succinate dehydrogenase: Enzyme variation and essentiality

in bacterial pathogens
Succinate dehydrogenases (succinate:quinone reductases, also known as
complex II or SQR) catalyse the reaction succinate + Q Ð fumarate +
QH2, which serves as a vital link between the tricarboxylic acid cycle and
oxidative phosphorylation. The reverse reaction can be performed by fuma-
rate reductases (quinol:fumarate reductases, QFR), in which fumarate
would serve as a terminal electron acceptor. QFR and SQR are closely
related enzymes and the reaction catalysed cannot be predicted based on
the primary amino acid sequence alone. In E. coli, which encodes both a
SQR and a QFR, one enzyme can in fact partially compensate for the loss
of the other (Guest, 1981; Maklashina, Berthold, & Cecchini, 1998), imply-
ing functional redundancy between the two enzymes. All complexes con-
tain the A subunit, where a flavoprotein that acts as the site of dicarboxylate
catalysis, and the B subunit, containing iron–sulphur clusters that facilitate
electron movement. Additionally, either one or two transmembrane sub-
units are present, which bind varying amounts of haems depending on
the class of enzyme. Due to the existence of eukaryotic complex II, these
enzymes are often overlooked for their role in human pathogens, as the
Energetics of Pathogenic Bacteria 13

potential for useful drug discovery would seem unlikely. However, human
complex II is a phylogenetically distinct class of enzyme to the majority of
the succinate dehydrogenases and fumarate reductases essential for bacterial
pathogenesis (Hägerhäll & Hederstedt, 1996; Hederstedt, 1999; Lancaster,
2001; Lancaster & Kr€ oger, 2000; Lemos, Fernandes, Pereira, Gomes, &
Teixeira, 2002).
There are several examples of succinate dehydrogenases that are essential
for virulence and can be discriminated for drug targeting. S. enterica, for
instance, encodes an annotated succinate dehydrogenase and a fumarate
reductase (McClelland et al., 2001; Spector et al., 1999). The succinate
dehydrogenase is expressed under aerobic conditions, where it is proposed
to input electrons metabolised through the tricarboxylic acid cycle into the
ubiquinone pool. Consistent with this central role, the deletion of the suc-
cinate dehydrogenase in S. enterica resulted in a defect in its ability to colonise
mice (Mercado-Lubo, Gauger, Leatham, Conway, & Cohen, 2008; Yimga
et al., 2006). However, the defect was not nearly as extreme in a deletion in
succinyl-CoA synthetase, suggesting functional complementation by the
fumarate reductase also encoded in this organism. A single deletion in fuma-
rate reductase gave no substantial difference in mouse survival, but a double
mutant was completely attenuated (Mercado-Lubo et al., 2008). As such, the
succinate oxidation activity of this organism is important for its viability in
host tissues. It is energetically favourable to transfer electrons from succinate
0 0
(E0 ¼ + 33 mV) to ubiquinone (E0 ¼ + 113 mV), as in S. enterica, but not
0
menaquinone (E0 ¼
74 mV). Hence, organisms that use menaquinone
as the sole electron acceptor, for example, C. jejuni, M. tuberculosis, and
S. aureus (Collins & Jones, 1981; Unden & Schirawski, 1997), must energise
this reaction. Studies in environmental organisms suggest that the pmf drives
this reaction, as reviewed in Lancaster (2013) and Lancaster, Haas, Madej,
and Mileni (2006). It seems therefore that driving succinate oxidation to
menaquinone reduction is an ‘Achilles heel’; perhaps targeting the men-
aquinone reduction site in these succinate dehydrogenases, in concert with
inhibitors of menaquinone synthesis and energy generation, would be lethal.
Siccarin serves as an example that selectivity can be achieved in succinate
dehydrogenase inhibitors (Mogi, Kawakami, et al., 2009). This compound
inhibits the succinate dehydrogenases from P. aeruginosa and murine mito-
chondria, but not those from E. coli and porcine mitochondria. Siccarin is
believed to be a ubiquinone analogue specific to succinate dehydrogenases
(Mogi, Kawakami, et al., 2009); while this compound is unsuitable as a spe-
cific antimicrobial, it raises the possibility that menaquinone analogues can
14 Gregory M. Cook et al.

be developed against succinate dehydrogenase specific for the bacterial

menaquinone-reducing enzymes. Caution should be exercised in an
organism-dependent manner, however, as deletion of succinate dehydroge-
nase in S. aureus resulted in a small colony variant phenotype (Gaupp, Schlag,
Liebeke, Lalk, & G€ otz, 2010) and such cells are known to become more
drug-resistant because of their slowed growth (McNamara & Proctor,
2000; Proctor et al., 1998; von Eiff, 2008). As such succinate dehydrogenase
inhibitors against S. aureus may in fact be deleterious.

2.3.3 Formate dehydrogenase: A major electron donor for anaerobic

respiratory chains
Formate serves as a major electron donor during the anaerobic respiration of
certain pathogens (e.g. Salmonella). Formate dehydrogenases are
metalloenzymes capable of oxidising formate to carbon dioxide; with a
mid-point potential of
410 mV, formate yields low-energy electrons that
drive reduction of anaerobic acceptors such as nitrate, concomitant with pmf
generation through a redox-loop mechanism ( Jormakka, Byrne, & Iwata,
2003a, Jormakka et al., 2003b; Jormakka, Tornroth, Byrne, & Iwata,
2002). E. coli and S. enterica encode three molybdopterin-containing formate
dehydrogenases: Fdh-N donates electrons for nitrate respiration; Fdh-O is
aerobically expressed and may serve as a redox valve; and Fdh-H is part
of a formate hydrogenlyase complex that fermentatively evolves hydrogen.
Their activities are particularly important following onset of anaerobiosis,
during which glycolytic pyruvate is converted to formate by pyruvate for-
mate lyase; depending on electron acceptor availability, the formate can be
respired by Fdh-N, decomposed by Fdh-H, or excreted from the cell
(Sawers, 1994). While these enzymes have not been extensively studied
in the context of enterobacterial pathogenesis, it is probable that formate
serves as a major electron donor in the anaerobic environment of the gas-
trointestinal tract. Phenotypic studies on S. enterica indicate that formate
has a protective role during stress (Barker, Kinsella, Jaspe, Friedrich, &
O’Connor, 2000) and is an important substrate for host invasion (Huang,
Suyemoto, Garner, Cicconi, & Altier, 2008). However, the role of formate
dehydrogenases in these processes is still ambiguous and the individual
enzymes can be deleted without causing significant phenotypes in vivo
(Becker et al., 2006; Huang et al., 2008).
In vivo studies have shown that formate is also one of the main electron
donors of the food-borne pathogen C. jejuni. The organism displays chemo-
tactic behaviour towards formate (Tareen, Dasti, Zautner, Gross, & Lugert,
Energetics of Pathogenic Bacteria 15

2010; Vegge, Brondsted, Li, Bang, & Ingmer, 2009) and rapidly consumes it
(Hoffman & Goodman, 1982) using the membrane-bound tungsten-
containing formate dehydrogenase FdhABCD (Shaw et al., 2012; Smart,
Cliff, & Kelly, 2009). Mutants of the enzyme had abnormal immunopathol-
ogy in a mouse model (Bereswill et al., 2011) and, in combination with a
hydrogenase mutant, were unable to colonise its preferred host
(Weerakoon, Borden, Goodson, Grimes, & Olson, 2009). Formate along-
side hydrogen and sulphite is therefore predicted to be a major electron
input in its intestinal niche (Hoffman & Goodman, 1982; Myers & Kelly,
2005; Weerakoon et al., 2009). However, redundancy in respiratory chains
has rendered individual formate-oxidising enzymes dispensable in C. jejuni,
as with S. enterica (Becker et al., 2006; Weerakoon et al., 2009). Formate
oxidation is also crucial for the pathogenesis of Aggregatibacter
actinomycetemcomitans, an opportunistic pathogen associated with periodon-
titis. RNA-seq analysis showed that formate dehydrogenase H was
upregulated in vivo, together with other anaerobic metabolic enzymes
( Jorth, Trivedi, Rumbaugh, & Whiteley, 2013). The mutant strain was
attenuated in a murine model ( Jorth et al., 2013). More fundamental
research is clearly needed into the in vivo roles of formate and formate dehy-
drogenases to elucidate whether there is any therapeutic potential in
targeting these enzymes.

2.3.4 Hydrogenase: Consuming a dependable reduced gas

While respiratory pathogens primarily rely on organic electron donors,
reduced gases also represent significant energy reservoirs. For example, there
is evidence that the M. tuberculosis complex liberates electrons for aerobic
respiratory chains using a carboxydobacteria-type, molybdenum-containing
carbon monoxide dehydrogenase (King, 2003; Park et al., 2007; Zacharia
et al., 2013). More generally, molecular hydrogen (H2) has proven a partic-
ularly dependable fuel source for pathogens, given its electronegativity
0
(E0 ¼
420 mV), diffusibility, and abundance in animal systems. As a result
of activity of fermentative gut bacteria, this gas is supersaturated by five logs
in animal systems compared to the atmosphere; in murine models, micro-
electrode measurements show that H2 levels vary from 170 (small intestines)
to 40 μM (stomach, liver, and spleen; Maier, 2005; Olson & Maier, 2002). It
is known that some respiratory pathogens can use the electrons derived from
H2 to generate a pmf. This primarily depends on the activity of the
membrane-bound [NiFe]-hydrogenases, which liberate electrons from H2
at a specialised [NiFe] centre and relay them into the quinone pool via a
16 Gregory M. Cook et al.

b-type cytochrome. These electrons can then be relayed to terminal oxidases

during aerobic respiration or fumarate reductase during anaerobic respira-
tion, resulting in vectorial or scalar proton translocation.
The Maier group first illuminated the importance of hydrogen metabo-
lism in pathogenesis by showing that the membrane-bound hydrogenase of
H. pylori is essential for virulence (Olson & Maier, 2002). A strain lacking the
enzyme was viable in vitro, but colonised mice at efficiencies up to 85% lower
than wild-type strains (Olson & Maier, 2002). With a high whole-cell
activity and an apparent Km of 1.8 μM, the hydrogenase is adapted to
rapidly consume the micromolar concentrations of H2 available in the
stomach. Though physiological and biochemical data on this enzyme remain
scarce, it is known that the hydrogenase is optimally expressed during
preferred microaerobic conditions. Maier proposes that H. pylori grows
mixotrophically in the human host through co-metabolising organic carbon
sources with molecular hydrogen; in light of both the nutritional deprivation
of the gastric mucosa and the limited capacity for the pathogen to catabolise
sugars, H2 oxidation may be responsible for a particularly large proportion of
electron flux through the respiratory chain (Maier et al., 1996; Olson &
Maier, 2002). It is thought that the electrons derived from H2 are transferred
to O2, though fumarate may substitute as an electron acceptor (Benoit &
Maier, 2008; Maier et al., 1996). Also in line with a central role in the vir-
ulence, the hydrogenase appears to be under the control of the two central
response regulators in H. pylori, the nickel-sensing NikR, and the iron-
sensing Fur (Contreras, Thiberge, Mandrand-Berthelot, & Labigne, 2003;
Merrell et al., 2003).
Hydrogen oxidation is similarly important but more complex in
S. enterica. Three homologous membrane-bound [NiFe]-hydrogenases,
Hya, Hyb, and Hyd, are collectively required for the organism’s virulence
in the murine model (Maier, Olczak, Maier, Soni, & Gunn, 2004). The aer-
obically expressed, oxygen-tolerant Hyd sustains aerobic respiration by cou-
pling H2 oxidation to oxygen reduction (Bowman et al., 2014; Parkin et al.,
2012). Hya and Hyb, in contrast, are anaerobically expressed, oxygen-
sensitive enzymes; it is proposed that Hyb sustains anaerobic respiration
by coupling H2 oxidation to fumarate reduction (Lamichhane-Khadka,
Kwiatkowski, & Maier, 2010; Sawers, Jamieson, Higgins, & Boxer,
1986), whereas Hya recycles fermentative H2 produced by the hydrogen
formate lyase complex (Sawers et al., 1986; Zbell & Maier, 2009). In line
with these roles, in vitro and in vivo studies show that the enzymes are differ-
entially controlled by the global regulatory systems FNR, ArcBA, and IscR
Energetics of Pathogenic Bacteria 17

in response to oxygen tension and redox state (Greening & Cook, 2014;
Jamieson, Sawers, Rugman, Boxer, & Higgins, 1986; Zbell, Benoit, &
Maier, 2007; Zbell, Maier, & Maier, 2008). The ability to sustain three
modes of energy conservation using hydrogen metabolism may contribute
the metabolic flexibility required for S. enterica to invade and disseminate at a
range of oxygen tensions (Maier et al., 2004). Consistently, researchers have
observed significant reductions in the growth of a Hyb mutant during initial
murine infection (Maier et al., 2013), as well as the survival of a Hya mutant
in macrophages (Zbell et al., 2008). However, Maier’s observation that tri-
ple, but not double, hydrogenase mutants were avirulent in a mouse model
suggests that there is some overlap or compensation between the hydroge-
nases (Maier et al., 2004).
The role of hydrogen metabolism in the energetics of other pathogens
remains unclear. For example, it has also been shown that the opportunistic
pathogen Bilophila wadsworthia can rapidly consume hydrogen, but the clin-
ical significance of this observation has not been investigated (da Silva,
Venceslau, Fernandes, Valente, & Pereira, 2008). Likewise, studies are
resolving the physiological role of the formate hydrogen lyase-related Ehr
complex of M. tuberculosis (Berney, Greening, Hards, Collins, & Cook,
2014; Berney, Greening, Conrad, Jacobs, & Cook, 2014); this putative
complex is regulated by the hypoxic response regulator DosR (Galagan
et al., 2013; He, Bretl, Penoske, Anderson, & Zahrt, 2011) and is predicted
to be essential for growth (Sassetti et al., 2003). BLAST searches reveal that
putative hydrogenases are encoded in numerous other human pathogens,
among them obligate parasites Neisseria meningitidis and Yersinia pestis. How-
ever, no comprehensive study has related hydrogen metabolism to their
pathogenesis to date.
Hydrogenases have also been proposed and patented as drug targets to
treat certain pathogens (Nie et al., 2012). These enzymes have many desir-
able traits as next-generation targets: they can be essential for virulence, are
absent in humans, and are often accessibly membrane-bound. Theoretically,
it would also be challenging for organisms to develop point mutations
resisting inhibitors specifically targeting the [NiFe]-active site. If used in
conjunction with traditional antibiotics, there is a possibility that hydroge-
nase inhibitors may impede the growth or persistence of certain pathogens.
In addition to the hydrogenase structural studies, studies on H. pylori suggest
that there could be merit in targeting the maturation factors (Olson,
Mehta, & Maier, 2001), nickel transporter (Nolan et al., 2002), and the
tat translocase (Benoit & Maier, 2014) required for the assembly of a
18 Gregory M. Cook et al.

functional hydrogenase; targeting such components might also generate the

desired pleiotropic effect on the three hydrogenases of S. enterica. Due to the
strong understanding of the structure and function of membrane-bound
[NiFe]-hydrogenases, there are opportunities to discover hydrogenase
inhibitors through activity-based screening and structure-based design.
However, no significant research or commercial development in this area
has been reported to date.

2.4. Terminal respiratory reductases

2.4.1 Haem–copper terminal oxidases: Proton translocation at a range
of oxygen partial pressures
Terminal oxidases are an essential component of aerobic respiratory chains
(Figs. 3 and 5A). These enzymes catalyse the four-electron reduction of
oxygen to water using electrons derived from organic and inorganic donors.
0
Given the electropositivity of the O2/H2O redox couple (E0 ¼ +825 mV),
O2 is the most energetically favourable electron acceptor available to path-
ogens and has the potential to energise vectorial proton translocation.
However, terminal oxidases must also employ mechanisms to minimise
the release of partially reduced intermediates that cause oxidative stress.
There are two major types of terminal oxidases relevant to bacterial patho-
gens (Fig. 5A). The cytochrome bd oxidases catalyse oxygen reduction at
an active site comprising haem b595 and haem d; these bacterial-specific,
non-proton translocating oxidases are phylogenetically distinct to the mito-
chondrial enzymes (Borisov, Gennis, Hemp, & Verkhovsky, 2011;
Junemann, 1997; Poole & Cook, 2000). In contrast, the haem–copper
oxidases are a diverse group of terminal oxidases that are related to mito-
chondrial complex IV; these complexes couple the reduction of O2, at a cat-
alytic site comprising a high-spin haem and a copper ion, to the translocation
of protons across the cytoplasmic membrane. There is much variation in the
primary sequences and haem content of the haem–copper enzymes, which
can be further-divided into the cytochrome aa3 oxidases, cytochrome bo3
oxidases, and cytochrome bcc3 oxidases (Buschmann et al., 2010; Pereira,
Santana, & Teixeira, 2001). The bd and bo3 oxidases use quinones as their
electron donors, whereas the principal electron donors for the bcc3 and
aa3 oxidases are c-type cytochromes reduced by the cytochrome bc1 complex
(Borisov, Gennis, et al., 2011; Junemann, 1997; Poole & Cook, 2000).
The major respiratory pathogens each encode at least one haem–copper
oxidase. The facultative aerobes Neisseria meningitidis and Neisseria gonorrhoeae
Figure 5 Diversity of electron transport chain composition under different oxygen tensions. The major components of the Salmonella enterica
electron transport chain under (A) oxic and (B) anoxic conditions. As with Fig. 3, not all intermediate steps in electron transport are indicated.
While represented as oxic, cytochrome bd is generally expressed under microoxic conditions. Additionally, nitrate and nitrite reductases are
used preferentially to the other anoxic electron acceptor complexes. This diversity serves as a general model for that which is observed in
other bacteria, culturable under both oxic and anoxic conditions.
20 Gregory M. Cook et al.

each encode single cytochrome bcc3 oxidases that are essential for growth
(Aspholm et al., 2010). Enzymes of this class typically have nanomolar affin-
ities for O2 and hence may confer Neisseria the ability to colonise microoxic
tissues (Pitcher & Watmough, 2004). Likewise, the microaerophile H. pylori
is predicted to solely rely on a cytochrome bcc3 oxidase (Nagata, Tsukita,
Tamura, & Sone, 1996; Smith et al., 2000; Tomb et al., 1997). Its close rel-
ative C. jejuni combines a cyanide-insensitive cytochrome bd oxidase (cioAB)
with a cytochrome bcc3 oxidase (ccoNOQP; Jackson et al., 2007); whereas
cioA was shown to be dispensable, Olson showed that a ccoN mutant
was unable to colonise chicken caeca (Guccione et al., 2010; Weingarten,
Taveirne, & Olson, 2009). Given bcc3-type oxidases are essential for
several pathogens, yet distantly related to human complex IV, there may
be pharmaceutical potential in specifically targeting them. Among other
pathogens, enterobacteria switch between the micromolar-affinity,
proton-translocating cytochrome bo3 oxidase, and the nanomolar-affinity
cytochrome bd oxidase during shifts in oxygen partial pressure (Cotter
et al., 1990; D’Mello et al., 1995, 1996); in vivo studies have shown that
mutations in either oxidase attenuates virulence of S. enterica in galline
and murine models (Turner et al., 2003; Zhang-Barber et al., 1997).
A recent study has also shown that S. aureus strains containing deletions
in its cytochrome aa3 oxidase or cytochrome bd oxidase were unable to dis-
seminate between tissues in mice (Hammer et al., 2013).
The ability of P. aeruginosa to use some five terminal oxidases appears to
be particularly important for its adaptation in the cystic fibrosis lung (Arai,
2011; Williams, Zlosnik, & Ryall, 2007). Three of the terminal oxidases
accept electrons from cytochrome c, namely, a cytochrome aa3 oxidase
(coxABC) and two cytochrome cbb3 oxidases (ccoNOQP1 and ccoNOQP2).
Whereas Cox has the highest efficiency of the oxidases (H+/O ratio ¼ 6),
Cco1 and Cco2 have the highest affinities for O2 (Km < 40 nM) of the five
enzymes (Arai, 2011). Consistent with this biochemistry, these enzymes
are known to be differentially expressed in vitro; whereas Cco1 is constitu-
tively active, the ANR-controlled Cco2 (Ray & Williams, 1997) scavenges
oxygen during hypoxia and the RpoS-activated Cox (Kawakami, Kuroki,
Ishii, Igarashi, & Arai, 2010) maximises proton translocation during nutrient
starvation (Comolli & Donohue, 2004). The organism also encodes a stress-
induced cytochrome bo3 oxidase (cyoABCD; Matsushita, Shinagawa,
Adachi, & Ameyama, 1982) and a cyanide-insensitive cytochrome bd
oxidase variant (cioAB; Cunningham, Pitt, & Williams, 1997), which
directly accept electrons from quinones. Despite having lower affinities
Energetics of Pathogenic Bacteria 21

and efficiencies than the cytochrome c oxidases, these enzymes are proposed
to be important during iron starvation and cyanogenesis, respectively (Arai,
2011). In vivo, Arai and Williams postulate that this multitude of oxidases
facilitates the adaptation of P. aeruginosa to decreasing oxygen availability
during the invasion of mucosal tissues (Arai, 2011; Jewes & Spencer,
1990; Williams et al., 2007; Worlitzsch et al., 2002). Consistently, despite
some overlap and redundancy between the oxidases, a triple oxidase mutant
was defective in microaerophilic growth and biofilm formation (Alvarez-
Ortega & Harwood, 2007).
Terminal oxidases are also central to the virulence of the obligate aerobe
M. tuberculosis. Mycobacteria combine a cytochrome bd oxidase with a fused
cytochrome bcc–aa3 supercomplex (Cook et al., 2014; Megehee, Hosler, &
Lundrigan, 2006; Shi et al., 2005). The constitutively expressed bcc–aa3 path-
way serves as the major respiratory route in mycobacteria under standard
culturing conditions; consistently, groups have not been able to disrupt
either component of this supercomplex in M. tuberculosis (Matsoso et al.,
2005). A recent high-throughput screen (600,000 compounds) targeting
ATP homeostasis in M. tuberculosis has uncovered inhibitors belonging to
imidazopyridines (Mak et al., 2012). A representative of this class of inhib-
itors (i.e. imidazo[1,2-a]pyridine) has been shown to target the bcc complex
encoded by the qcrCAB operon of M. tuberculosis (Abrahams et al., 2012;
Kang et al., 2014; Pethe et al., 2013). The lead compound from these studies
Q203 is antimycobacterial at low micromolar concentrations against pure
cultures, in macrophages, and in the murine model of TB infection
(Kang et al., 2014; Pethe et al., 2013).

2.4.2 Cytochrome bd oxidase: A bacterial-specific next-generation

drug target
Cytochrome bd oxidases are a unique component of prokaryotic electron
transport chains. In general, these enzymes have a higher affinity for oxygen
but a lower energetic efficiency than the cytochrome aa3 oxidases and cyto-
chrome bo3 oxidases (Borisov, Gennis, et al., 2011; Junemann, 1997;
Poole & Cook, 2000). The structure of cytochrome bd oxidase has yet to
be determined, but it is known that the enzyme is a heterodimeric complex
composed of a large (CydA) and small subunit (CydB) that hold together the
haems b558, b595, and d. A third protein CydX, which is important for
enzyme activity of the bd oxidase, was proposed to be a third subunit of
the complex in E. coli (Sun et al., 2012; VanOrsdel et al., 2013). It is gen-
erally thought that haems b595 and d form a di-haem site for the reduction of
22 Gregory M. Cook et al.

O2 (Borisov, Gennis, et al., 2011). Cytochrome bd oxidase does not pump

protons like cytochrome c oxidases, but generates a pmf by transmembrane
charge separation at an H+/e
ratio of 1 (Borisov, Forte, Sarti, & Giuffre,
2011). Ubiquinone and menaquinone are accepted as electron donors
depending on the species. For example, CydAB of E. coli accepts both
quinols (Unden & Bongaerts, 1997), whereas in other species like
mycobacteria, only menaquinone is present.
While the physiological role of cytochrome bd oxidase has not yet been
fully elucidated, the major function is thought to be the reduction of oxygen
under low oxygen conditions. This contributes to the maintenance of pmf
and allows the scavenging of oxygen to colonise oxygen poor niches or
to protect oxygen labile enzymes. It is therefore of no surprise that many
human pathogens harbour this enzyme because bacteria residing in human
tissues are faced with low oxygen conditions either due to the immune
response or competition with other bacteria (e.g. in the gut). Many enteric
pathogens like S. enterica (Fig. 5A), Shigella flexneri, E. coli, and Enterococcus
faecalis express the enzyme, as do other infectious agents of the genera
Brucella, Bacteroides, Streptococcus, Listeria, and Mycobacterium (Borisov,
Gennis, et al., 2011). Consistent with these physiological roles, E. coli and
S. enterica differentially express the cydAB and cyoABCD operons according
to the oxygen tension and redox state of the cell. To facilitate oxygen scav-
enging under hypoxic conditions, the two-component system ArcBA acti-
vates cydAB but represses cyoABCD transcription (Cotter, Melville,
Albrecht, & Gunsalus, 1997; Iuchi, Chepuri, Fu, Gennis, & Lin, 1990).
By contrast, both enzymes are repressed by FNR at the onset of anaerobiosis
in favour of nitrate reductase and fumarate reductase (Cotter et al., 1990,
1997; Fink et al., 2007; Sellars, Hall, & Kelly, 2002). The cydAB genes in
mycobacteria likewise responds to oxygen depletion (Berney & Cook,
2010; Kana et al., 2001); though no redox or oxygen-sensing regulator
has been found to control cydAB transcription thus far, SenX3–RegX3
has been shown to be a regulator (Roberts, Vadrevu, Madiraju, & Parish,
2011). Nanaerobes also express cytochrome bd oxidase when O2 is available
at nanomolar partial pressures. The term nanaerobes was created when the
opportunistic pathogen Bacteroides fragilis, a so-called strict anaerobe, was
found to grow in the presence of nanomolar O2 concentrations facilitated
by cytochrome bd oxidase (Baughn & Malamy, 2004). Subsequently, a
plethora of so-called strict anaerobes were found to harbour this enzyme,
suggesting that many of these are capable of growth under nano-aerobic
conditions (Baughn & Malamy, 2004).
Energetics of Pathogenic Bacteria 23

Several additional functions have been proposed for this enzyme. Increas-
ing evidence suggests the involvement of CydAB in protecting against the
host’s oxidative stress and nitrosative stress defences (Giuffre, Borisov,
Arese, Sarti, & Forte, 2014; Poole & Cook, 2000). For example, bacterial
pathogens face NO stress in the phagosome of macrophages triggered by
the innate immune system. In E. coli, the enzyme shows lower sensitivity
to NO inhibition than cytochrome bo and this could be of particular advan-
tage under low oxygen conditions where many NO detoxification mecha-
nisms are less efficient (Giuffre, Borisov, Mastronicola, Sarti, & Forte, 2012;
Mason et al., 2009). Several reports also indicate that CydAB is involved in
the detoxification of hydrogen peroxide (Borisov, Gennis, & Konstantinov,
1995; Giuffre et al., 2014; Poole & Williams, 1988). E. coli cytochrome bd
oxidase expression increases in response to H2O2 and a cydAB mutant dis-
played enhanced expression of catalase, indicating an increase in intracellular
H2O2 concentration (Lindqvist, Membrillo-Hernandez, Poole, & Cook,
2000). The pleiotropic effects of cytochrome bd oxidase deletion in E. coli
and Brucella abortus could be rescued by the addition of catalase and superox-
ide dismutase (SOD; Endley, McMurray, & Ficht, 2001; Goldman,
Gabbert, & Kranz, 1996). Superoxide is a by-product of several electron
transport chain components such as NDH-1 and the cytochrome bc1 com-
plex and is converted to H2O2 by SOD (Turrens, 1997). Hence, it is tempt-
ing to speculate that cytochrome bd oxidase also scavenges ROS produced
from the respiratory chain when imbalances in the electron flux occur. Con-
sistently, transcription of the cydAB operon responds to other environmental
triggers like high pH, high pressure, high temperature, and oxidative stress, as
well as respiratory poisons and uncouplers (Borisov, Gennis, et al., 2011;
Poole & Cook, 2000). It is also of note that the cytochrome bd oxidase var-
iants of P. aeruginosa and C. jejuni appear to be unusual in that they are low-
affinity enzymes, but are exceptionally cyanide-insensitive (Cunningham
et al., 1997; Jackson et al., 2007); though their physiological roles are still
incomplete-understood, they may be particularly important during stress-
inducing conditions.
An additional role that has been proposed for CydAB is supplying
oxidising power for biosynthetic processes. For example, haem biosynthesis
and disulphide-bond formation have been shown to depend on a functional
electron transport chain under oxic, microoxic, and anoxic conditions
(Bader, Muse, Ballou, Gassner, & Bardwell, 1999; Mobius et al., 2010).
Quinols are good electron acceptors and are available as long as terminal oxi-
dases are functional. Cytochrome bd oxidase was shown to be required
24 Gregory M. Cook et al.

under microoxic conditions to sustain the activity of protoporphyrinogen

IX oxidase (HemG) of E. coli (Mobius et al., 2010), as well as for the
reoxidation of DsbB, a protein needed for disulphide-bond formation that
supports protein folding (Bader et al., 1999). The ABC transporter CydCD
was shown to transport glutathione and cysteine in E. coli (Pittman,
Robinson, & Poole, 2005), contributing to the maintenance of a reduced
periplasm. The CydDC proteins are also essential for the assembly of CydAB
(Bebbington & Williams, 1993; Siegele, Imlay, & Imlay, 1996).
Some recent literature points towards an important role of this oxygen
reductase in pathogenesis. Brucella suis and B. abortus, the causative agents of
brucellosis in swine and cattle, were shown to depend on cytochrome bd
oxidase for replication and survival in macrophages and mice (Endley
et al., 2001; Loisel-Meyer, Jimenez de Bagues, Kohler, Liautard, &
Jubier-Maurin, 2005). Other studies have shown that cytochrome bd
mutants of S. typhimurium (Zhang-Barber et al., 1997), S. flexneri (Way,
Sallustio, Magliozzo, & Goldberg, 1999), and M. tuberculosis (Zhang et al.,
2012) are at a competitive disadvantage in murine models when
co-infected with a wild-type strain. M. tuberculosis resides in so-called gran-
uloma in the lungs of its host and this tissue is hypoxic. CydAB was expressed
transiently in a mouse infection model (Shi et al., 2005), but given that mice
do not form hypoxic granulomas, the full importance of cytochrome bd oxi-
dase might be underestimated. Experiments in animal models with hypoxic
granuloma could shed more light on the in vivo importance of this enzyme.
In M. tuberculosis, the bd oxidase cannot sustain growth as the sole terminal
oxidase because the aforementioned cytochrome bcc–aa3 supercomplex is
essential (Matsoso et al., 2005). However, a defect in the cytochrome c bio-
synthesis pathway causes increased expression of cydAB (Small et al., 2013);
this suggests that, while CydAB activity does not yield enough energy to
support growth, it might be sufficient to maintain a membrane potential
when the haem–copper oxidase is downregulated or inhibited.
Cytochrome bd oxidase has not yet been explored as a therapeutic target.
Nevertheless, several compounds are known to inhibit the enzyme and have
been used for studying structural and mechanistic aspects of respiratory
chains in bacteria. Comprehensive lists of known inhibitors have been
published (Borisov, Gennis, et al., 2011; Meunier, Madgwick, Reil,
Oettmeier, & Rich, 1995). However, most inhibitors are unspecific and
inhibit haem–copper oxidases as well. This could be problematic as cyto-
chrome aa3 oxidases are also present in mitochondria. Aurachin D was
the only inhibitor that showed specificity for cytochrome bd oxidase, albeit
only at lower concentrations (Meunier et al., 1995). On the contrary, many
Energetics of Pathogenic Bacteria 25

inhibitors are known that have higher affinity for aa3 oxidases rather than bd
oxidases. For example, the bd oxidase is more resistant to the action of cya-
nide, azide, nitric oxide, and Zn ion stress. Based on the in vitro data available
to date, it is tempting to speculate that cytochrome bd oxidase plays a role in
the adaptation, colonisation, and survival of human pathogens in the host
environment. However, evidence for an essential function of this enzyme
during pathogenesis is still scarce and proof of such would undoubtedly spark
more interest for developing a specific drug. Cytochrome bd oxidase might
be an imperfect single drug target (due to the presence of other terminal oxi-
dases), but an inhibitor could potentially increase killing rates in combina-
tion with other respiratory chain inhibitors due to the enzymes proposed
role in oxidative stress defence. Studies aimed at elucidating the physiolog-
ical role of cytochrome bd oxidase during bacteria–host interactions, as well
as structural information of the enzyme, will guide potential initiatives for
targeted drug development in the future.

2.4.3 Respiratory nitrate and nitrite reductases: Exploiting

host immune defences
Oxygen partial pressures vary by orders of magnitude between tissues, as
a result of the high rates of aerobic respiration by host, commensal, and
pathogen cells. Most pathogens therefore employ anaerobic respiratory
chains to grow or survive in hypoxic or anoxic tissues (Fig. 3). Nitrate is
the most energetically favourable anaerobic electron acceptor given its rel-
0
atively positive mid-point potential (E0 ¼ + 430 mV) compared to alterna-
0
tives such as fumarate (E0 ¼ + 30 mV), trimethylamine N-oxide (TMAO;
0 0
E0 ¼ + 130 mV), and tetrathionate (E0 ¼ +170 mV). As a result, many
pathogens upregulate preferentially nitrate reductases and denitrification
pathways when oxygen is limited and substrate is available (Fig. 5B). During
infection, many pathogens can derive nitrate from the nitric oxide produced
by inducible nitric oxide synthase (iNOS) during the innate immune
response. For example, S. enterica uses the flavohaemoglobin/nitric oxide
dioxygenase Hmp to convert nitric oxide to nitrate in the macrophage;
exploiting the host’s defences, the bacterium both detoxifies a poison and
provides a strong oxidant for anaerobic respiration (Mills, Rowley, Spiro,
Hinton, & Richardson, 2008).
There is growing evidence that denitrification confers a competitive
advantage to S. enterica and pathogenic E. coli. Under anoxic conditions, these
enterobacteria employ the regulatory systems FNR, NarXL, and NarQP to
ensure that nitrate is used above energetically inferior acceptors such as fuma-
rate (Gilberthorpe & Poole, 2008; Stewart, 2003). The reduction of nitrate to
26 Gregory M. Cook et al.

nitrite is catalysed by three nitrate reductases (i.e. respiratory membrane-

bound narGHJI, assimilatory periplasmic napFDAGHBC, and starvation-
induced membrane-bound narZYWV; McClelland et al., 2001; Spector
et al., 1999), and the six-electron reduction of nitrite to ammonium is medi-
ated by two nitrite reductases (i.e. periplasmic cytochrome c-dependent
nrfABCDEFG and cytosolic NADH-dependent nirBDC; Wang &
Gunsalus, 2000). S. enterica can also detoxify macrophage-derived nitric oxide
to either nitrate using flavohaemoglobin Hmp, nitrous oxide using flavor-
ubredoxin NorV, or ammonium using NrfA (Mills et al., 2008; Poole,
2005). The nitrate reductases appear to be individually dispensable, but col-
lectively confer a growth advantage (Knuth, Niesalla, Hueck, & Fuchs,
2004; Winter et al., 2013). Using a murine colitis model, the Bäumler group
recently demonstrated that wild-type E. coli could outcompete a nitrate reduc-
tase triple mutant using iNOS-derived nitrate (Winter et al., 2013). Equiva-
lent studies in S. enterica showed that nitrate enhances growth at least 10-fold
and is also a stimulus for Tsr-mediated energy taxis (Bliska & van der Velden,
2012; Lopez et al., 2012; Rivera-Chavez et al., 2013).
M. tuberculosis is unable to grow using electron acceptors such as nitrate
and nitrite. However, Sohaskey has demonstrated that the organism
enhances its long-term viability by reducing these compounds (Sohaskey,
2008). M. tuberculosis expresses a respiratory nitrate reductase (narGHJI )
and nitrite reductase (nirBD) homologous to those of enterobacteria
(Malm et al., 2009). In addition, it encodes the nitrate importer NarK2
(Sohaskey, 2005; Sohaskey & Wayne, 2003), an unresolved nitrite exporter
(Griffin et al., 2012), and the potentially non-functional fusion protein NarX
(Sohaskey & Wayne, 2003). While narGHJI and nirBD expression is consti-
tutive, nark2x is upregulated during hypoxic and nitrosative stress by DosR
(Shi et al., 2005). In line with increased nitrate import and accumulation of
reduced cofactors, the rate of nitrate reduction increases in hypoxic cells
in vitro (Sohaskey & Wayne, 2003). Though the physiological role of the
nitrate reductase has not been fully resolved, it may generate pmf and
reoxidise cofactors in non-growing hypoxic cells; consistently, the enzyme
enhanced adaptation to sudden anaerobiosis, but not gradual oxygen deple-
tion in vitro (Sohaskey, 2008). While the nitrite produced is partly exported
(Griffin et al., 2012), it can also be reduced to ammonium by NirBD; this
process is essential for survival in both the in vitro Wayne model and in
human macrophages (Akhtar, Khan, Sohaskey, Jagannath, & Sarkar,
2013). Like S. enterica, there is evidence to suggest that denitrification in
M. tuberculosis depends on nitric oxide produced by iNOS in the human
Energetics of Pathogenic Bacteria 27

macrophage ( Jung et al., 2013); this molecule serves to both provide the
substrate nitrate (Arya et al., 2013) and induce its import through activating
the haem-based histidine kinases for DosR (Voskuil et al., 2003).
There is significant evidence that P. aeruginosa also uses denitrification
during human infection. This opportunistic pathogen chronically infects
the cystic fibrosis lung, forming biofilms (Singh et al., 2000) and harnessing
the amino acids available in the viscous sputum (Ohman & Chakrabarty,
1982; Palmer, Mashburn, Singh, & Whiteley, 2005). Likely as a result of
iNOS activity, the sputum is supersaturated with nitrate and nitrite in cystic
fibrosis versus normal lungs (Linnane et al., 1998). Encoding a complete
denitrification pathway, P. aeruginosa sequentially reduces nitrate to din-
itrogen using a membrane-bound nitrate reductase (NarGJ), a cytochrome
cd1-type nitrite reductase (NirS), a cytochrome bc-type nitric oxide reductase
(NorBC), and a two-copper centre nitrous oxide reductase (NosZ; Williams
et al., 2007). Due to the elevated epithelial oxygen uptake (Stutts, Knowles,
Gatzy, & Boucher, 1986) and high bacterial cell densities (Wessel et al.,
2014) in cystic fibrosis lung tissues, the sputum contains hypoxic or anoxic
microenvironments that may promote nitrate respiration ( Jewes & Spencer,
1990; Worlitzsch et al., 2002). The Whiteley group have also showed that
the nitrate reductase is required for anaerobic growth in an in vitro cystic
fibrosis sputum medium (Palmer, Brown, & Whiteley, 2007). Consistently,
transcription of the denitrifying enzymes is activated by nitrate, the O2 sen-
sor ANR, and the NO sensor DNR (Arai, Kodama, & Igarashi, 1997; Ye
et al., 1995). However, it has become apparent that P. aeruginosa is also capa-
ble of aerobic denitrification. This process may be particularly important in
the context of the host (Alvarez-Ortega & Harwood, 2007; Chen, Xia, & Ju,
2003; Filiatrault et al., 2005); while energetically less efficient, co-respiration
of nitrate and oxygen may provide the flexibility needed to withstand tem-
poral and spatial shifts in the concentrations of electron acceptors.
Contrary to its original classification as an obligate aerobe, the sexually
transmitted pathogen N. gonorrhoeae employs a partial denitrification path-
way to grow anaerobically and microaerobically (Knapp & Clark, 1984).
It harbours the capacity to reduce nitrite to nitrous oxide, but nitrate reduc-
tase is absent and the nitrous oxide reductase genes contain nonsense muta-
tions (Cardinale & Clark, 2005). The copper-containing outer membrane
nitrite reductase AniA reduces nitrite to nitric oxide using electrons donated
from c-type cytochromes in the respiratory chain (Hopper, Tovell, & Cole,
2009; Mellies, Jose, & Meyer, 1997). NorB, a cytoplasmic quinol-
dependent nitrous oxide reductase, in turn maintains nitric oxide at anti-
28 Gregory M. Cook et al.

inflammatory nanomolar concentrations (Cardinale & Clark, 2005;

Householder, Fozo, Cardinale, & Clark, 2000). Both aniA and norB are
induced by nitric oxide as a result of inactivation of the repressor NsrR
and hence activation of NarP; as the most strongly upregulated gene under
anaerobiosis, aniA is also regulated by the oxygen sensor FNR (Householder
et al., 2000; Overton et al., 2006; Whitehead et al., 2007). In the human
host, denitrification may confer the metabolic flexibility for N. gonorrhoeae
to generate pmf and maintain redox balance in anoxic microenvironments
(Householder et al., 2000). Moreover, there is compelling in vivo evidence
that detoxification of nitric oxide at a range of oxygen tensions enables the
bacterium to survive nitrosative stress and suppress inflammatory responses
effected by the host’s iNOS (Cardinale & Clark, 2005; Stevanin, Moir, &
Read, 2005; Tunbridge et al., 2006). Equivalent pathways have been
characterised in causative agent of meningococcal disease, N. meningitidis
(Anjum, Stevanin, Read, & Moir, 2002; Rock et al., 2005), though clinical
isolates suggest that AniA is dispensable for pathogenesis (Stefanelli
et al., 2008).
Denitrification appears to be a widespread mechanism for the anaerobic
adaptation of human pathogens. While H. pylori lacks the capacity for deni-
trification, the periplasm of the related ε-proteobacterium C. jejuni encodes
a nitrate reductase and a nitrite reductase; nitrate is incapable of sustaining
anaerobic growth, though co-respiration with oxygen confers a small advan-
tage in vitro (Sellars et al., 2002). S. aureus can grow using nitrate and nitrite as
alternative electron acceptors (Burke & Lascelles, 1975); microarray analysis
has shown that NarGH and NirBD in this organism is induced under hyp-
oxia (Fuchs, Pane-Farre, Kohler, Hecker, & Engelmann, 2007) by the
oxygen-sensitive two-component system NreBC (Schlag et al., 2008).
Looking beyond the respiratory chain, it is also important to note that nitric
oxide enhances the oxidative stress response and antibiotic resistance of both
S. aureus and Bacillus anthracis in vivo (Gusarov, Shatalin, Starodubtseva, &
Nudler, 2009; Shatalin et al., 2008); this depends on NO derived from
the bacterial nitric oxide synthase (bNOS) reactivating catalase and
suppressing the Fenton reaction (Gusarov & Nudler, 2005).
Given the central role of denitrification in many pathogens, it is conceiv-
able that inhibiting these pathways may have therapeutic outcomes. How-
ever, much research would be required to develop a specific, penetrable, and
potent antimicrobial. Inhibition of nitrate reductase or nitrite reductase may
promote membrane depolarisation, redox imbalance, and, in the latter case,
accumulation of toxic intermediates; this could impair both fast growers
Energetics of Pathogenic Bacteria 29

such as S. enterica and persisters such as M. tuberculosis and P. aeruginosa. In

addition to the structural subunits, there may also be potential in targeting
the proteins involved in the unique pathways for the biosynthesis of their
cofactors, for example, sirohaem, haem d1, or c-type cytochromes. There
may also be potential in modulating nitric oxide metabolism through
inhibiting iNOS, bNOS, or flavohaemoglobin. A new study identified
two lead compounds that specifically inhibited bNOS and reduced the
survival of Bacillus subtilis; there are opportunities for this proof-of-
concept study to be extended to pathogens (Holden et al., 2013).
However, given the multifaceted role of nitric oxide in both the cytotox-
icity and cytoprotection of pathogens, inhibiting nitric oxide production
may also yield deleterious effects. For example, administration of the
iNOS-2 inhibitor N 6-(1-iminoethyl)-L-lysine caused accelerated progres-
sion of M. tuberculosis infection to chronic tuberculosis in murine lungs
(MacMicking et al., 1997).

2.4.4 Fumarate reductase: Anaerobic respiration using an endogenous

electron acceptor
While fumarate reductases cannot be readily distinguished from succinate
dehydrogenases on a sequence level, these enzymes have generally been
selected for distinct purposes in bacterial cells (Fig. 3). For example, whereas
the aforementioned succinate dehydrogenase of S. enterica primarily inputs
electrons from the tricarboxylic acid into the aerobic respiratory chain, its
fumarate reductase has a role in anaerobic respiration (Encheva, Shah, &
Gharbia, 2009; Fig. 4B). Consistently, these enzymes are differentially reg-
ulated by ArcBA and FNR in S. enterica and E. coli, with succinate dehydro-
genase being expressed under oxic conditions and fumarate reductase being
expressed under anoxic conditions in the absence of nitrate (Fink et al.,
2007; Sellars et al., 2002)
The microaerophile H. pylori lacks a succinate dehydrogenase but
encodes a fumarate reductase (Ge, Jiang, Kalisiak, & Taylor, 1997;
Pitson, Mendz, Srinivasan, & Hazell, 1999). In this case, fumarate serves
as a terminal electron acceptor, with succinate more akin to a fermentative
end-product. The central role of fumarate reductase for the growth of this
organism is demonstrated by the fact that it is essential for the colonisation
of the mouse stomach (Ge et al., 2000) and that nizatidine enhances
Helicobacter killing in concert with metronidazole (Chen et al., 2002), in
a manner that is correlated with the ability of both compounds to indepen-
dently inhibit fumarate reductase activity. A high percentage of infected
30 Gregory M. Cook et al.

patients generate antibodies against this protein, validating that targeting

this protein is unlikely to be toxic to humans (Birkholz & Knipp,
1994). Its more flexible relative C. jejuni encodes a distinct type of fumarate
reductase, as well as an enzyme that has the ability to function physiolog-
ically as both a fumarate reductase and a succinate dehydrogenase. Whereas
the unidirectional enzyme seems to have a marginal role in growth
(Guccione et al., 2010), deletion of the bidirectional enzyme results in
major growth defects in microaerobic growth, lethality during oxygen lim-
itation, and the inability to colonise chicken caeca (Guccione et al., 2010;
Weingarten et al., 2009).
M. tuberculosis encodes two annotated succinate dehydrogenases (sdh1
and sdh2) and one fumarate reductase (frd; Cook et al., 2014). These enzymes
must therefore fill roles more complicated than being one dedicated succi-
nate dehydrogenase and one dedicated fumarate reductase, although it is
highly likely that considerable functional redundancy exists between the
three enzymes. Consistently, the deletion of the frd genes resulted in no phe-
notype (Watanabe et al., 2011), likely a response of compensation. Homo-
logues of sdh1 and sdh2 in M. smegmatis have been found to be differentially
expressed, being upregulated under energy-limiting and oxygen-limiting
conditions, respectively (Berney & Cook, 2010). In agreement with this,
transposon site hybridisation screens find that M. tuberculosis sdh1 is essential
under aerobic conditions (Griffin et al., 2011), whereas sdh2 is essential
under hypoxia (Baek, Li, & Sassetti, 2011). It may be that sdh1 is the pre-
ferred succinate dehydrogenase for growth and sdh2 is a bidirectional
enzyme that behaves according to the availability of reduced/oxidised sub-
strates (Pecsi et al., 2014). During hypoxia, studies have found that succinate
is produced and excreted as an end-product and that this may effect mem-
brane potential generation (Eoh & Rhee, 2013; Watanabe et al., 2011).
Rhee proposed that succinate is produced in large quantities as a
multifunctional compound that can maintain succinate dehydrogenase
activity, be used for ATP synthesis, or be excreted to generate membrane
potential when terminal electron acceptors are lacking (Eoh & Rhee,
2013). These results would be consistent with sdh2 being a bidirectional
enzyme and the compensation observed when frd was deleted. The func-
tional redundancy and structural differences within the three enzymes
may make it challenging to develop inhibitors against succinate dehydroge-
nase or fumarate dehydrogenase activity in M. tuberculosis; however, the
development of menaquinone biosynthesis inhibitors or menaquinone ana-
logue inhibitors may be able to overcome this.
Energetics of Pathogenic Bacteria 31

2.4.5 Tetrathionate reductase and other alternative reductases:

Emerging roles in host colonisation
Enterobacteria exploit inflammation to outcompete intestinal microbiota
(Stecher et al., 2007). In particular, S. enterica drives its host to produce both
reactive nitrogen species and ROS using the type III secretion systems
T3SS-1 and T3SS-2 (Tsolis, Adams, Ficht, & Baumler, 1999). In addition
to converting nitric oxide to nitrate, the organism can detoxify ROS to
yield the anaerobic electron acceptor, tetrathionate (Fig. 5B); specifically,
tetrathionate is produced when inflammatory oxygen radicals react with
thiosulphate derived from the hydrogen sulphide produced by fermentative
gut commensals (Winter & Baumler, 2011). Tetrathionate is reduced by the
molybdopterin-containing tetrathionate reductase TtrABC; this enzyme is
induced under hypoxia by FNR (Hensel, Hinsley, Nikolaus, Sawers, &
Berks, 1999) and uses electrons derived from organic electron donors,
including alternative donors such as ethanolamine (Thiennimitr et al.,
2011). Winter et al. (2010) showed that deletion of the tetrathionate reduc-
tase resulted in a 100-fold reduction in intestinal lumen bacterial cell counts
in a murine model. Adding to cumulative evidence that tetrathionate is a
central anaerobic electron acceptor in vivo, it was recently shown that
S. enterica can mediate energy taxis towards the compound using the che-
moreceptor Aer (Rivera-Chavez et al., 2013). There is some in silico and
in vitro evidence that tetrathionate respiration may also have a role in other
pathogens, for example, C. jejuni and Citrobacter freundii (Liu, Denkmann,
Kosciow, Dahl, & Kelly, 2013; Lupp et al., 2007; Winter & Baumler,
2011). Collectively, these findings emphasise the overlap between patho-
genesis, inflammation, and diet in the infection of S. enterica and likely other
gastrointestinal pathogens; this raises possibilities for using pharmaceuticals,
immunomodulation, and dietary control to control the energetics of such
pathogens for preventative or therapeutic outcomes.
It is known that pathogens can reduce a spectrum of other sulphur com-
pounds in vitro. For example, thiosulphate is a potential electron acceptor for
enteric bacteria and is produced by the oxidation of hydrogen sulphide by
mitochondria of the colonic mucosa (Hildebrandt & Grieshaber, 2008).
During growth on formate as an electron donor, pathogens such as
S. enterica and C. freundii can drive a menaquinol–thiosulphate oxidore-
0
ductase to reduce this electronegative molecule (E0 ¼
402 mV) using
the pmf (Kapralek, 1972; Stoffels, Krehenbrink, Berks, & Unden, 2012).
E. coli, S. enterica, and C. jejuni also express specific reductases that reduce
the more potent electrophiles dimethyl sulphoxide (DMSO) and TMAO
32 Gregory M. Cook et al.

when available (Fink et al., 2007; Sellars et al., 2002). It is of note that
anaerobic sulphate-reducing bacteria of the genus Desulfovibrio occasion-
ally present as opportunistic pathogens, for example, D. desulfuricans in
bacteremia (Goldstein, Citron, Peraino, & Cross, 2003). Looking beyond
respiration, it has been shown that SirA of M. tuberculosis (previously
misannotated as NirA) is a ferredoxin-sulphite oxidoreductase; in addition
to an assimilatory role, it may serve to dissipate excess reductant as hydro-
gen sulphide (Berney, Greening, Hards, et al., 2014; Pinto, Harrison, Hsu,
Jacobs, & Leyh, 2007; Schelle & Bertozzi, 2006).

2.5. Generators of sodium motive force in bacterial pathogens

Bacteria can also switch to sodium energetics under conditions where the
generation of a pmf is impeded by environmental conditions (Fig. 6). Some
pathogens generate an electrochemical gradient of sodium ions across their
cell membranes, i.e., a sodium motive force (smf). The smf can be generated
by two major mechanisms (Fig. 6). Many pathogens use primary sodium
pumps driven by oxidative phosphorylation (e.g. the Na+-translocating

Figure 6 Generalised role of sodium and proton cycles at the level of the membrane. Only
representative proton- (nuo) and sodium- (oad, nqr) pumping complexes are shown for
simplicity. The sodium and proton motive forces can be used to drive cellular processes
such as flagellar motion and ATP synthesis. Antiporters provide an overlap between the
two cycles as they can theoretically operate to generate one electrochemical gradient at
the expense of the other, depending on cellular needs. Nha is used to refer to
antiporters in general, but in reality, antiporters can operate on variable stoichiometry
(not pictured) which results in either electrogenic or electroneutral exchange of H+ and
Na+ depending on the protein.
Energetics of Pathogenic Bacteria 33

NADH–quinone oxidoreductase NQR; Verkhovsky & Bogachev, 2010) or

decarboxylative phosphorylation (e.g. oxaloacetate decarboxylase; Buckel,
2001; Dimroth, 1994). In addition, almost all pathogens can interconvert
a pmf to a smf using sodium/proton antiporters (Mulkidjanian, Dibrov, &
Galperin, 2008; Padan, Venturi, Gerchman, & Dover, 2001; Fig. 6).
Phylogenetic evidence posits that sodium may have preceded protons in
the evolution of chemiosmosis (Lane, Allen, & Martin, 2010; Mulkidjanian
et al., 2008). Consistently, sodium has been retained as a coupling ion in
pathogens across diverse phyla, though the relative importance of pmf and
smf greatly varies between organisms (Häse, Fedorova, Galperin, &
Dibrov, 2001). The enterobacteria Klebsiella pneumoniae, E. coli, and
S. enterica rely on pmf for ATP synthesis and other cellular processes, but
power specific processes such as the symport of amino acids and citrate using
smf (Deguchi, Yamato, & Anraku, 1990; van der Rest et al., 1992). Energy
generation in Vibrio cholerae also depends on a H+-specific F1Fo-ATP
synthase, though its sodium gradients have a central role in pathogenesis;
the ion both regulates the expression of virulence factors (Häse &
Mekalanos, 1999) and drives multidrug resistance pumps (Begum et al.,
2005; Brown, Paulsen, & Skurray, 1999; Huda, Morita, Kuroda,
Mizushima, & Tsuchiya, 2001), toxin exporters (Häse & Mekalanos,
1999), and flagellar motors (Kojima, Yamamoto, Kawagishi, & Homma,
1999; Yorimitsu & Homma, 2001). At the other end of the spectrum,
smf has the dominant role in the energetics in some strictly anaerobic path-
ogens, including the periodontal pathogen Fusobacterium nucleatum (Kapatral
et al., 2002; Schulz et al., 2013) and the syphilis spirochete Treponema
pallidum (Fraser et al., 1998).
Dimroth established a new paradigm in bioenergetics with his discovery
that oxaloacetate decarboxylase of K. pneumoniae is in fact a primary sodium
pump (Dimroth, 1980). During the catalytic cycle of this trimeric decarbox-
ylase, the carboxylate group from oxaloacetate is transferred to and activated
by the biotin cofactor of the β-subunit; the subsequent elimination of the
group as CO2 drives Na+ binding and translocation via a conserved aspartate
residue (Buckel, 2001; Di Berardino & Dimroth, 1996; Dimroth &
Thomer, 1993). Thus, whereas oxidative phosphorylation couples redox
reactions to ion translocation, decarboxylative phosphorylation uses the
free-energy change from decarboxylation for sodium translocation. In
K. pneumoniae and other enterobacteria, the oxaloacetate decarboxylase
enhances the uptake and catabolism of citrate. However, the enzyme is oth-
erwise dispensable for the viability of this metabolically flexible organism
34 Gregory M. Cook et al.

(Bott, Meyer, & Dimroth, 1995). By contrast, the genome of the

unculturable T. pallidum suggests that its only primary ionic pump is oxalo-
acetate decarboxylase. Thus, decarboxylative phosphorylation is predicted
to solely provide the potential energy for both ATP synthesis and substrate
import in this organism and hence may represent a novel target for the devel-
opment of antimicrobials against syphilis (Häse et al., 2001).
F. nucleatum, another smf-based pathogen, conserves free energy by
fermenting amino acids such as glutamate via the glutaconyl-CoA pathway
(Bakken, Hogh, & Jensen, 1989; Beatrix, Bendrat, Rospert, & Buckel,
1990). Through landmark studies of this pathway in fusobacteria and
clostridia, Buckel has concluded that energy generation depends on the con-
current processes of decarboxylative phosphorylation and electron bifurca-
tion. The membrane-bound, biotin-dependent complex glutaconyl-CoA
decarboxylase generates smf from the decarboxylation of glutaconyl-
CoA to crotonyl-CoA (Beatrix et al., 1990). It is proposed that the high-
0
potential crotonoyl-CoA (E0 ¼
10 mV) and low-potential ferredoxin
0
(E0 ¼
400 mV) are thereafter simultaneously reduced by electrons derived
0
from the mid-potential NADH (E0 ¼
280 mV); this electron bifurcation
is mediated by the multimeric complex Bcd/EcfBC and depends on flavin
serving as a two-electron gate. The low-energy electrons of ferredoxin
can either drive reductive cellular processes or be transduced to smf via a
further primary sodium pump, the ferredoxin-NAD+ oxidoreductase Rnf
(Herrmann, Jayamani, Mai, & Buckel, 2008; Li et al., 2008). These processes
are reviewed in more detail by Buckel and Thauer (2013). This efficient
coupling of these anaerobic chemiosmotic processes generates a large
smf in this organism and maximises the output of the Na+-translocating
F1Fo-ATP synthase (Schulz et al., 2013).
Primary sodium pumps can also be driven by oxidative phosphorylation,
notably the Na+-translocating NADH–quinone oxidoreductase
(Verkhovsky & Bogachev, 2010). Phylogenetically and functionally distinct
from the previously described H+-translocating NDH-1 and non-
translocating NDH-2, NQR appears to be the major primary sodium pump
in aerobic pathogens (Häse et al., 2001). Though an atomic resolution struc-
ture of this six-subunit enzyme is unavailable, the chemiosmotic coupling
mechanism depends on redox reactions via multiple cofactors ( Juarez &
Barquera, 2012). First studied in the opportunistic pathogen Vibrio
alginolyticus (Tokuda & Unemoto, 1981, 1984), NQR generates the smf
required for transport and motility. Raising opportunities for antimicrobial
development, the natural product korormicin has a cytotoxic effect and
Energetics of Pathogenic Bacteria 35

potently inhibits NQR in this organism (Hayashi, Shibata, Nakayama,

Yoshikawa, & Unemoto, 2002; Yoshikawa, Nakayama, Hayashi,
Unemoto, & Mochida, 1999). In K. pneumoniae, it remains controversial
whether smf is entirely generated by NQR; studies have not resolved
whether NDH-1 translocates H+, Na+, or both ions (Batista,
Marreiros, & Pereira, 2012; Bertsova & Bogachev, 2004; Krebs, Steuber,
Gemperli, & Dimroth, 1999; Steuber, 2001). Nevertheless, the smf is par-
ticularly important for citrate fermentation. The utilisation of smf both ener-
gises the citrate symporter CitS (van der Rest et al., 1992) and enables
cofactor regeneration though mediating reversed electron transfer from for-
mate to NAD+ (Pfenninger-Li & Dimroth, 1992).
The majority of pathogens also encode multiple Na+/H+ antiporters that
allow the interconversion of pmf and smf. For example, V. cholerae expresses a
wide array of antiporters (Häse et al., 2001) that have a high degree of func-
tional redundancy (Herz, Vimont, Padan, & Berche, 2003; Vimont &
Berche, 2000). Likewise, P. aeruginosa operates a robust sodium cycle
(Häse et al., 2001) using cation:proton antiporters that have been implicated
for virulence in mice (Kosono et al., 2005; Ueda & Wood, 2008). It is gen-
erally appreciated that Na+/H+ antiporters operate homeostatically to mod-
ulate chemiosmotic forces to cellular needs (Häse & Barquera, 2001; Häse
et al., 2001). As such, it seems likely that any therapeutic strategies targeting
the sodium cycle will need to consider potential compensation by antiporter
complexes. In this regard, inhibition of Na+/H+ antiporters may be best
achieved using derivatives of amiloride compounds. They effectively inhibit
mammalian CPA1-type antiporters (Noël & Pouysségur, 1995), but con-
centrations 10 times higher are required to inhibit the antiport activity of
V. parahaemolyticus (Häse & Barquera, 2001; Kuroda, Shimamoto,
Mizushima, & Tsuchiya, 1997). Many improvements in drug design against
these targets remain to be made, but finding pathogen-specific inhibitors
may prove valuable in combination therapy with other sodium or proton
cycle inhibitors.

2.6. ATP homeostasis and the F1Fo-ATP synthase: A clinically

approved drug target
ATP synthases are membrane-bound enzymes found in all kingdoms of life.
The ATP synthase has remained remarkably conserved throughout evolu-
tion and is believed to be an example of modular evolution during which
two functionally independent subunits, a DNA helicase with ATPase activ-
ity and a proton-powered motor, became associated and gained new
36 Gregory M. Cook et al.

functionality (Doering, Ermentrout, & Oster, 1995). From a structural and

functional perspective, three distinct classes of the enzyme are recognised,
F-type, A(archaeal)-type, and V(vacuolar)-type. The F-type enzymes,
found in prokaryotes and eukaryotes, can synthesise and hydrolyse ATP
and this process is generally coupled to protons. In some bacterial species,
the enzyme is able to translocate sodium ions (von Ballmoos, Cook, &
Dimroth, 2008). The A-type enzyme, found in Archaeal genomes, also
shares similar properties. The V-type enzymes are primarily known for their
ATP hydrolysis activity in eukaryotic cells, but have been reported in some
prokaryotes (Yokoyama, Akabane, Ishii, & Yoshida, 1994), including the
pathogens Enterococcus hirae (Kakinuma, Yamato, & Murata, 1999) and
T. pallidum (Fraser et al., 1998).
The F-type ATP synthase consists of two subcomplexes: the water-soluble
F1 sector, which harbours the catalytic centres, and the membrane-embedded
Fo complex, which mediates ion translocation across the membrane (Fig. 4).
Detailed knowledge on structure and function is available for the water-
soluble F1 headpiece with the subunit composition α3β3γδε (Abrahams,
Leslie, Lutter, & Walker, 1994; Cingolani & Duncan, 2011; Stock,
Leslie, & Walker, 1999). Alternating α and β subunits form a cylinder around
subunit γ. Part of the γ subunit protrudes from the bottom of the cylinder and
forms the central stalk together with the ε subunit. The central stalk is con-
nected with the oligomeric c-ring of the membrane-intrinsic Fo moiety. The
other Fo subunits of bacterial ATP synthases are a and b2, which abut the
c-ring laterally. The oligomeric c-ring harbours a series of identical ion-
binding sites and varies in size from 8 to 15 subunits depending on the species.
The reasons for this variability and the factors governing c-ring size are incom-
pletely understood. The molecular chaperones for ATP synthase in bacteria
remain unknown, but an important area of research, given the essentiality
of the enzyme in some bacterial pathogens.
The F1Fo-ATP synthase catalyses ATP synthesis by utilising the electro-
chemical gradient of protons or sodium ions (smf) to generate ATP from
ADP and inorganic phosphate (Pi) and operates under conditions of a high
pmf/smf and low intracellular ATP. The enzyme is also capable of working as
an ATPase under conditions of high intracellular ATP and an overall low
pmf (von Ballmoos et al., 2008). Bacterial pathogens are often faced with
the challenge of adapting to environmental conditions during transitions
from normoxia to hypoxia where the pmf drops rapidly (von Ballmoos
et al., 2008). Under these conditions, the ATP synthase shifts gears to
ATP hydrolysis activity to combat the decreasing pmf. As an ATPase, the
Energetics of Pathogenic Bacteria 37

enzyme hydrolyses ATP while pumping protons or sodium ions from the
cytoplasm to the outside the cell. Regulation of ATPase activity under these
conditions is essential to prevent wasteful (futile) ATP hydrolysis. The reg-
ulatory mechanisms bacteria use to block ATP hydrolysis in response to a
suboptimal pmf remain incompletely understood, but represent a potential
avenue for disrupting ATP homeostasis. Some bacterial F1Fo-ATPase prep-
arations exhibit high rates of ATP hydrolysis activity, whereas other prep-
arations show extreme latency in ATP hydrolysis activity and can only be
activated by detergents, solvents, or proteases. The molecular mechanisms
responsible for this inhibition remain to be identified.
F1Fo-ATP synthase has been shown to be essential for the optimal
growth of mycobacteria, including M. tuberculosis, on fermentable and
non-fermentable carbon sources (Griffin et al., 2011; Tran & Cook,
2005; Zhang et al., 2012). The reasons for this remain unclear. One expla-
nation is that the extraordinary amount of ATP required to synthesise a
mycobacterial cell might exceed what can be supplied by substrate-level
phosphorylation alone (Cox & Cook, 2007). Alternatively, the organism
may not be able to support uncoupled respiration; either they lack a conduit
for proton re-entry in the absence of the F1Fo-ATP synthase or they are
unable to adjust the proton permeability of the cytoplasmic membrane to
allow a futile cycle of protons to operate. Consistently, the cytoplasmic
membrane of M. smegmatis is known to be extremely impermeable to
protons (Tran et al., 2005). The essentiality of F1Fo-ATP synthase has also
been demonstrated in several other Gram-positive pathogens, namely,
S. pneumoniae (Ferrandiz & de la Campa, 2002), Listeria monocytogenes
(Cotter, Gahan, & Hill, 2000), and S. aureus (Balemans et al., 2012). There
is also evidence that the complex is essential in the Gram-negative H. pylori
(McGowan, Cover, & Blaser, 1997). This may reflect the cellular ATP
demands of this strictly respiratory pathogen. Alternatively, ATP-driven
proton extrusion may contribute to the maintenance of intracellular pH
homeostasis and a high ΔpH in its acidic stomach niche (Matin et al., 1996).
The F1Fo-ATP synthase has thus only been shown experimentally to be
essential for growth in a relatively small number of bacteria (Cotter et al.,
2000; Koebmann, Nilsson, Kuipers, & Jensen, 2000; McGowan et al.,
1997; Tran & Cook, 2005). This is an intriguing observation given the uni-
versal distribution of this enzyme in bacterial cells and its central role in cellular
ATP homeostasis. This may reflect the ability of ATP generation by substrate-
level phosphorylation to sustain the growth of most bacterial cells on ferment-
able carbon sources, where increased glycolytic flux can compensate for the
38 Gregory M. Cook et al.

loss of oxidative phosphorylation (Friedl et al., 1983). For organisms that can
generate energy through fermentation, for example, S. enterica (mixed-acid
and hydrogen fermentation; Sawers et al., 1986; Zbell & Maier, 2009) and
P. aeruginosa (arginine and pyruvate fermentation; Williams et al., 2007),
F1Fo-ATP synthase is generally dispensable when fermentable carbon sources
are available in the culture medium; hence, substrate-level phosphorylation is
sufficient to maintain the basic needs of such cells. However, oxidative phos-
phorylation still has a major role in the pathogenesis of some of these facul-
tative fermenters (Terry, Pina, & Mattingly, 1992; Turner et al., 2003); for
example, strains of S. enterica containing deletions in atpB and atpH were
highly attenuated in murine and galline models (Turner et al., 2003).
Whereas proton-coupled ATP synthases are widespread in prokaryotes,
sodium-coupled enzymes are restricted to a few anaerobic bacterial genera
(Dimroth & Cook, 2004). A sodium-coupled F1Fo-ATP synthase has
recently been reported in F. nucleatum, a Gram-negative bacterium impli-
cated in the etiology of periodontal diseases (Schulz et al., 2013). We have
shown that inhibiting this enzyme has a bactericidal effect, suggesting that it
is essential for viability (M. Iglesias-Cans & G.M. Cook, unpublished data).
Sodium-translocating V1Vo-ATP synthases have also been characterised in
E. hirae (Kakinuma et al., 1999) and predicted in T. pallidum (Fraser et al.,
1998). While these pumps seem to primarily be involved in sodium extru-
sion, studies have not ruled out roles in ATP synthesis. T. pallidum encodes
two such pumps that are speculated to differ in their directionality or ion
selectivity; as previously discussed, Na+-translocating oxaloacetate decar-
boxylase is the only predicted primary pump in this organism and hence
smf is likely to drive ATP synthesis (Häse et al., 2001). Further work is clearly
needed to determine if this bacterium and other pathogens harbour Na+-
coupled ATP synthases.
Taking these observations into account would argue against the F1Fo-ATP
synthase being a good broad-spectrum drug target for bacterial pathogens, but
could be exploited in the small number of bacteria where the enzyme is essen-
tial for growth even on fermentable carbon sources. A large number of natural
and synthetic inhibitors have been identified that target the F1 and Fo motors
of F1Fo-ATP synthase (reviewed in Hong & Pedersen, 2008), but the majority
of these inhibitors also target mitochondrial ATP synthase. A key to the
druggability of bacterial ATP synthase will be the identification of
bacterial-specific inhibitors and determining the essentiality/non-essentiality
of this enzyme in bacterial growth and persistence. Diarylquinolines have been
shown to target the F1Fo-ATP synthase of mycobacterial species and inhibit
Energetics of Pathogenic Bacteria 39

ATP synthesis (Andries et al., 2005; Koul et al., 2007, 2008; Lounis et al.,
2010). Genome sequencing of both M. tuberculosis and M. smegmatis mutants
that are resistant to diarylquinolines, i.e., bedaquiline (TMC207, BDQ), rev-
ealed that the target of these compounds is the oligomeric c-ring (encoded by
atpE) of the enzyme (Andries et al., 2005; Huitric, Verhasselt, Andries, &
Hoffner, 2007; Huitric et al., 2010). The purified c-ring from M. smegmatis
binds BDQ with a KD of 500 nM, and modeling/docking and kinetic studies
suggest that BDQ blocks rotary movement of the c-ring during catalysis by
mimicking key residues in the proton transfer chain (de Jonge, Koymans,
Guillemont, Koul, & Andries, 2007; Haagsma et al., 2011). Further investi-
gations with inverted membrane vesicles of M. smegmatis and BDQ have rev-
ealed that the drug acts independently of the pmf and that electrostatic forces
play an important role in its interaction with the ATP synthase (Haagsma
et al., 2011).
In 2012, the FDA approved the drug bedaquiline for the treatment of
multidrug-resistant M. tuberculosis, the first drug licensed in 40 years for
the treatment of tuberculosis disease. Bedaquiline exhibits in vitro bacteri-
cidal activity against mycobacterial strains that are either susceptible or resis-
tant to all first-line and many second-line drugs and has extraordinary in vivo
activity against several mycobacterial species in different animal models and
in TB patients (Andries et al., 2005; Lounis et al., 2006, 2010). When myco-
bacterial cells (growing or non-growing) are treated with BDQ, time-
dependent (not dose-dependent) killing is observed (Andries et al., 2005).
The mechanism of killing is not clear, but it does not involve the dissipation
of the membrane potential, which is lethal to all living cells. A dose-
dependent decrease in intracellular ATP has been observed when
M. tuberculosis cells are treated with BDQ (Koul et al., 2007, 2008), but these
data do not explain cell death, because mycobacterial cells can be depleted of
ATP and yet remain viable (Frampton, Aggio, Villas-Boas, Arcus, & Cook,
2012). BDQ is bactericidal towards most species of mycobacteria but is only
bacteriostatic against M. avium (Lounis, Gevers, Van den Berg, Vranckx, &
Andries, 2009). The identification of the mechanisms underlying this sen-
sitivity will be important in understanding how BDQ exerts its anti-
tuberculosis activity. Tran and Cook (2005) originally hypothesised that
BDQ may act as a respiratory plug blocking ATP synthase, leading to hyper-
polarisation of the pmf and backpressure on NADH turnover and proton
pumping. Consistent with this hypothesis is recent data reporting that
non-proton-translocating cytochrome bd oxidase (cydAB) is upregulated
in response to BDQ challenge (Koul et al., 2014). The upregulation of
40 Gregory M. Cook et al.

cydAB expression is required for continued electron transport chain activity

(and NADH turnover), as electron transport to this terminal oxidase is not
impeded by a high pmf. In support of these data, we have shown that cydAB
mutants of M. smegmatis show greater sensitivity to BDQ compared to wild-
type cells (Robson & Cook, unpublished data). Further work is required to
understand the precise mechanism of cell death caused by bedaquiline.
The notion that decreasing intracellular ATP levels in M. tuberculosis leads
to cell death has directed scientists to perform high-throughout screens
(600,000 compounds) targeting ATP homeostasis. These screens have
uncovered 140 new compounds that reduce ATP levels in M. tuberculosis,
among them the aforementioned Q203 that targets the cytochrome bcc–aa3
supercomplex (Mak et al., 2012). There is also potential in administering
bedaquiline and other inhibitors of ATP homeostasis for the treatment of
other pathogens, notably drug-resistant S. aureus (Balemans et al., 2012)
and F. nucleatum (M. Iglesias-Cans & G.M. Cook, unpublished data).

3. CONCLUSIONS AND FUTURE PERSPECTIVES

Table 2 summarises the major components of the respiratory chains of
the key pathogens in this review. The discovery that a compound targeting a
central enzyme in cellular energy generation found in both prokaryotes and
eukaryotes would lead to the first drug (bedaquiline) licensed for tubercu-
losis in 40 years has inspired scientists to consider bacterial energy generation
and energetics as a target space for antimicrobial development despite the
risk of human toxicity. The future and development of bacterial energetics
for antimicrobial discovery will be dependent on increased understanding of
the energetic processes operating in bacterial pathogens in host tissues during
both replication and persistence. For example, the ability of S. enterica to
stimulate the host to produce an electron acceptor (tetrathionate) for anaer-
obic respiration to outgrow the fermenting microbiota (Winter et al., 2010)
provides fascinating insight into the novel energetic strategies employed by
bacterial pathogens. Important energetic machineries where no mammalian
homologues exist like NDH-2, cytochrome bd oxidase, and hydrogenases
are clearly attractive targets for drug development. Advancing these targets
will be dependent on both identifying high-affinity specific inhibitors of
these complexes and elucidating the role of these components in the target
pathogen. The ability of the ATP synthase inhibitor bedaquiline to kill both
replicating and non-replicating cells, resulting in shortened TB chemother-
apy, highlights the importance of this area for drug discovery and expansion
Energetics of Pathogenic Bacteria 41

Table 2 Summary of the organisation and flexibility of the respiratory chains of the key
pathogens discussed in this review
Primary Terminal
dehydrogenases and oxidases and ATP
Organism decarboxylases reductases Redox carriers synthase
Escherichia coli NADH Cytochrome Ubiquinone H+-F1Fo
dehydrogenase I bo3 oxidase Menaquinone
NADH Cytochrome Cytochrome c
dehydrogenase II bd oxidase
Na+-NADH Nitrate
dehydrogenase reductase (3)
Succinate Nitrite
dehydrogenase reductase (2)
Formate Fumarate
dehydrogenases (2) reductase
Hydrogenase (2) DMSO
reductase
TMAO
reductase
Salmonella NADH Cytochrome Menaquinone H+-F1Fo
enterica dehydrogenase I bo3 oxidase Ubiquinone
NADH Cytochrome Cytochrome c
dehydrogenase II bd oxidase
Na+-NADH Nitrate
dehydrogenase reductase (3)
Succinate Nitrite
dehydrogenase reductase (2)
Formate Fumarate
dehydrogenases (2) reductase
Hydrogenase (3) DMSO
reductase
TMAO
reductase
Tetrathionate
reductase
Thiosulphate
reductase
Pseudomonas NADH Cytochrome Ubiquinone H+-F1Fo
aeruginosa dehydrogenase I cbb3 oxidase (2) Cytochrome c
NADH Cytochrome
dehydrogenase II aa3 oxidase
Na+-NADH Cytochrome
dehydrogenase bo3 oxidase
Continued
42 Gregory M. Cook et al.

Table 2 Summary of the organisation and flexibility of the respiratory chains of the key
pathogens discussed in this review—cont'd
Primary Terminal
dehydrogenases and oxidases and ATP
Organism decarboxylases reductases Redox carriers synthase
Succinate Cyanide-
dehydrogenase insensitive
Formate oxidase
dehydrogenase Nitrate
reductase
Nitrite
reductase
Neisseria NADH Cytochrome Ubiquinone H+-F1Fo
gonorrhoeae dehydrogenase I cbb3 oxidase Cytochrome c
Neisseria Na+-NADH Nitrite
meningtidis dehydrogenase reductase
Succinate
dehydrogenase
Hydrogenase?
Campylobacter Flavodoxin Cytochrome Menaquinone H+-F1Fo
jejuni dehydrogenase? cbb3 oxidase Cytochrome c
Succinate Cyanide-
dehydrogenase insensitive
Formate oxidase
dehydrogenase Fumarate
Hydrogenase reductase
Sulphite oxidase Nitrate
reductase
Nitrite
reductase
Thiosulphate
reductase
DMSO
reductase
TMAO
reductase
Helicobacter Flavodoxin Cytochrome Menaquinone H+-F1Fo
pylori dehydrogenase? cbb3 oxidase Cytochrome c
Hydrogenase Fumarate
reductase
Staphylococcus NADH Cytochrome Menaquinone H+-F1Fo
aureus dehydrogenase II (2) aa3 oxidase Cytochrome c
Port furious

monkeys

the largely also

large that

brown of

quite having the

and

as

the
small

anyone objects

so Savernake a

is mistaken Madagascar

escape the being

alike

remains small

of elephant

lions lions

cat
made a was

From

inches

since of

about described
suckled

they

far Neumann in

from precipices us

OMMON early at
not

to for

are

get are

seizing

our
would

were

up the of

There even

its

deadened

the

the
animal

Messrs are

Lane account tribes

Islands

BADGERS mother take

and

dangerous

says is C

Sons jumping are

interest they
in puma

in the

found ATER Baird

no ground

the round

A in feet

remains form old

ALM

two is the
offer 269 v

especially The most

not This

thousands

like

are

as

from breed

Its are their

the Central of
young feathers

creatures prisoners and

nose

is

shows for haunts

lands

rhinoceros

human near animal

Berlin

in fours born

which Originally

have of This

Russia

these

that sometimes

particular incidents

go on
it which

the a their

ways packs amounts

by nicely arm

one promise

cautious store circumstances

in seen

by

fierce are outbreak

chariots their

Southern do and

up

346
Carpathians

arm Photo appeared

Their

bag its is

a him extending

both

dealt

reach one elongated

small
among sparks blood

Except They for

few is

was OLLOW

s obstinately It

never barb often

one his

speed

THER

is

the of recorded

marsupials

causes over the

post forests
a They

the

and

that Africa thrown

open

a absolutely

between the

example the
climb

animal

turn and

habit in

the the in

years vocabulary

the nest of

Lioness valley
specimen was not

enormous

is

both the an

cats completeness

generously hundred mischievous

the

a gives supposed

other and COQUEREL

because young been

The and

hilly

believe

Rock a Zoological

attacked

elongated the Such

almost national a

OR and not

There

sheep are

nature far
follow short cat

was in

United none of

or coat at

the the aisle

in

fast S with
is fine The

by

it powers

in

met

not never

of of coats

dark to

FAMILY

season a Irish
unadvisable

of

C my reached

being

fragments strangely

of

was conditions

have coat Bay

allied of was
refuge

separate

of a

built than parts

polar
ago

s certain

fallen when

Finchley

seem almost how

They idea As

not the

black to can

of

it elephant

This young blues

them said

79 the
Richard or

fingers

are

Deer sandy breeds

crows up have

cautiously

Bear a

of of sharp

head

kinds far possession

place hair

year held which

of in

was was of

large developed the

use

broken

As rhinoceroses part
them

more 11

Photo

Except Pending a

and

of

out and the

are The the

eats mews
family at

WOLF

They G the

paralysed with

all from

of head

as constantly
happened great forked

Hon

are Other horses

years BROWN

sleep and

At yards that

throwing entirely wall

by
bones marten

of stomach feeling

at is

60 pretty sea

and
was on was

and outside the

like the as

perform

I is dogs
thick a is

as they the

fruit acquired not

Indians thick as

too of round

to forests natural

and and

equal

by photographed Among

pious on
with

or

part of

for a from

in animal F
claws like

is

most ugly PYCRAFT

Photo actually

rake characteristic

animal of

A some sucker
islands

open

the fur

I years

was
into old

stud

Madagascar Tiger

for The

rigging ears

carry Rhinoceros Jones

pig Museum immense

The first The

these hour

and brown of

and

of 100 The

at four are

glaciers AT

probably

mortal The
North

nerves her Stag

400 body

there

tricks hamster four

of is

objects as

from

been down Carl

Mountain and

it place

have The Abbey

the body

I did
the leopard

the S the

foot animal to

of

pocket

heard west a

Bedford

back African
of be G

a are squirrel

Le killed its

in IMALAYAN intended

exhibition white
in Long occur

MUSK

being nearly the

and

Newman
or

there not in

on a

Brehm Calabar

seen
A family

which well to

a before

time the three

hasten
herds

to

give

this through cats

wife

Of THREE

were a the

make
with Institutes

of Bear

sitting for

Large

are

and

these usually

very animal T
the KENT

animal teeth

either mare

obey

co universal 199

other thoroughly

PANDA mere permission

colour

the elephants where

wastes came

their cat

beautiful its one

countries Horses stated

into Clinging Burchell

pair numerous

HE

and the

writer of ONNET

are the

a Skye

to former

their so

has
defence

in creatures It

faced killed the

highest terminating in

This
of

a the

rocks

the

the and and

more

to the

the almost

but

the element
prove holes

local in for

then apart much

India

most sultry
and upon

African as

hunter extended their

have Lord Captain

African of German

with but

discoverers rather
such

American

Photo on easily

shattered its our

mounted lightly a
at monkey

hear anything keepers

Asiatic had

like in S

to

and lbs and

all sea
serval lifeless

del

Its tail

or

roof
the

FOX in Orange

these

in camp

kept faces of

The except is

monks is

case

flat leopards

and
or the was

haunts

England

birds from exposes

present

only
same semi dangerous

all wayfarer

product

folds

the tame

Sydney and
on

been it

that killed

or and these

of to days

easy

ox footsteps
the of silent

from White

edge sleep fur

live His colour

are

according the

cats

by interesting

in
author of

is torn

been

well great

pace are white

hunter

Photo not

are Foal

AARD a
scavenger is it

or

kittens the seen

by

Large C Ottomar

sleeps

peasants quite or

BADGERS
They

the have

Turkey was

not captivity of

food be

above its fish

the

its

the

they
where the with

but

Gerenuk

some habits their

living So

anything JUMPING survivors

grey Photo
leap

which

Without has weasel

drier Tribe

with

should was
America

opening and dry

alarmed roared his

should caravan of

tigers sometimes sleep

is

larger them EAR

which light it

it This

this to next

and pig

and

Black LEMUR its

Small by

In
ones hunting

white AND

MARKED to

is so

me by striped

at roan head

by

are
in

Brown an hordes

as

animals WOOD

him to

had they

by a

house also

of his of
discovered

in 86

heads

grey little clearer

that

of T hyæna

Byerley

that mouths and

fish a amigo

sub rather man

old

specimen for bowels

of s
I best

and coat

Polar

as grunt supposed

usually fore roof

BATS

though of
is one

of attracted ground

none

pursued and

is

met

AND few

one hair

rabid surprise

have large accordingly

illustration

from were

backed as of

much excellent

of

low with

moment which

of

them half
sometimes

is

the as

The

coast white

in by and
of

any marmot

follows the

fast a heads

empty they

are church

while this

into to species

to to
large In Common

Photo AND most

Probably

always frequently

the in
OPHERS is HAPPY

river slopes

Continent will

the by bred

not

to always instinct

animal their

with

saw

500 It of
crest

governor There

toes

any A

sail there

the a greyhounds

got than

habit
form

three carry

with born

Sea off

by small gives

the
quite

M of true

of lives

itself

claws otter hand

mountain and

to
traps He

seals in

with man whilst

the

push at

gaining they
Duchess to face

in

the were have

great ING

on its

very summer

himself kennel The

has or his

accounted
the

were

UR skin

winter well

no Pemberton

as always

is Antarctic to

limits are

regiments of beaver

called
on like

way

sleepiest

had

interval in

longest furnished

are true Park

set and

presenting frozen the

in
and The where

Worcester coats way

off white

Gardens

187

are anthropoid before

relations

and

was LONG
the

The eagle is

it

these

They

plants

upon they
house coats

pampas

to On hen

all

from find dog

domestic met

giraffe of

the ION stay

vouch struggles using

suddenly

live in

of of and

were

are

cows The the

RAIRIE

tradition

a The escaped
still the

but

brought menagerie

Photo

it is in

to specialisation
stated

the

the

upper

then length instinctive

of grand

the

do the lay

nostrils
its

of

The to county

these met permission

nuts

CAT hundred the

in

the

the

are weight

when dogs all

Guiana

OOLLY a
had this

either TIGER

when could called

eight Where

in 300 from

tree not harness

The fact

experiences

an of restricted

with Scholastic

The its

the Male

for is

the merely round

no

A we
the The is

ECCAN

and

are South

tree of

in others mat

met

Of
than

to

Railway

fronted the swam

or his s

regions
is and

young AND

between

carry Zoological

aF

long reinforced

The

Family could

tempered asked seasons

Wanderoo VI both

from Atlas

attack

and were

have

furnished flat

whole

elephant upon for

is far

foot old

travel a amiable

the been

remember

foot

the Lord

the vast Juice

s badger reliable

are close

and Mantled

Nevada to is

caravans had

are the shown

The MALE of

size THE in

sharp of in
tigers

them

breed P

and from

prices the space

terriers proboscis

Victoria the

Ltd

not

the Mountain

a wing

of measuring insects
of

do

Professor all always

the almost in

flesh
useless fine head

and an old

photograph the Cape

refused

its

the alder

Street was Polar

she known

group

The

Medland The

the

the
The

on

looking

thick Arab

bordering

which
coarse and near

reason The

of of he

the portions

is with and

prehistoric

Messrs equipped It

itself zebras Long

as it This

tame
Photo

PERFORMING limbs Siamese

and

spite

or writes
in

poultry ordinary All

of

of to

Percy
with

Speaking usually

same

now them the

account as By

the grown shore

T and than
in it

and boats them

hunting

than

the shoulder at

USK altogether

evidence story

up

L
up

a in

the England

restrictions 284

species note

its the both

to manes Four

loss

themselves

forests barb
Its

teeth consideration behind

Sunderbunds the mangabey

quite turned hunt

chacmas

Island

all

exhibited from
by

trees

parts EVER s

have

by

monkeys from

imps birds amusing

has these
unnecessary whilst

serious

is

times top

nullah of

found

of

that In
as as

snake have the

Skye

telephoto out

the sea
he or Ibex

is

Constantinople of

A of and

take still

bit of

old bear

ears removed of

Changing on

interest made large

instead

in of

of is

Ottomar these Burchell

the

scraping

if G development

they herbs playfulness

still octodonts are

been HE are

or creatures

and

Scotch here and

all anywhere Mr

number

difficulty the structure

is Among

It

tamed

rhinoceros as obtained

s
APYBARA

extreme

12

Arab white bite

The in were
an and Hagenbeck

the

was

I most Herr

BY hand

and

Stag round

their clasping fur

of concludes that

for relatives differing

thirteen and

M monkeys as

of or high

be

curiosity on and

four

seldom to

the herds

In

Photo more in
the

porcupine

Bear over

path these

Burchell Mr

the three American

originally Sumatran or

the Its

are
be

if towns

to The ROWN

length bag

done