Anda di halaman 1dari 239

Bacterial Sensing and Signaling

Contributions to Microbiology
Vol. 16

Series Editors

Axel Schmidt Witten


Heiko Herwald Lund

Bacterial Sensing
and Signaling
Volume Editors

Mattias Collin Lund


Raymond Schuch New York, N.Y.
42 figures, 2 in color, and 3 tables, 2009

Basel Freiburg Paris London New York Bangalore


Bangkok Shanghai Singapore Tokyo Sydney

Contributions to Microbiology
formerly Concepts in Immunopathology and Contributions to Microbiology and Immunology

Mattias Collin, PhD

Raymond Schuch, PhD

Department of Clinical Sciences


Division of Infection Medicine
Biomedical Center B14
Lund University
Klinikgatan 26
SE221 84 Lund (Sweden)

Laboratory of Bacterial
Pathogenesis and Immunology
The Rockefeller University
New York, NY 10021 (USA)

Library of Congress Cataloging-in-Publication Data


Bacterial sensing and signaling / volume editors, Mattias Collin, Raymond
Schuch.
p. ; cm. -- (Contributions to microbiology, ISSN 1420-9519 ; v. 16)
Includes bibliographical references and indexes.
ISBN 978-3-8055-9132-4 (hardcover : alk. paper)
1. Cell interaction. 2. Chemoreceptors. 3. Cellular signal transduction.
4. Bacteria--Physiology. I. Collin, Mattias. II. Schuch, Raymond.
[DNLM: 1. Bacteria--chemistry. 2. Bacterial Processes. 3. Quorum
Sensing--physiology. 4. Signal Transduction--physiology. QW 52 B1315 2009]
QR96.5.B338 2009
571.74--dc22
2009015269

Bibliographic Indices. This publication is listed in bibliographic services, including Current Contents and Index Medicus.
Disclaimer. The statements, opinions and data contained in this publication are solely those of the individual authors and
contributors and not of the publisher and the editor(s). The appearance of advertisements in the book is not a warranty,
endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the
editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products
referred to in the content or advertisements.
Drug Dosage. The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this
text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research,
changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader
is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and
precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.
All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by
any means electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and
retrieval system, without permission in writing from the publisher.
Copyright 2009 by S. Karger AG, P.O. Box, CH4009 Basel (Switzerland)
www.karger.com
Printed in Switzerland on acid-free and non-aging paper (ISO 9706) by Reinhardt Druck, Basel
ISSN 14209519
ISBN 9783805591324
e-ISBN 9783805591331

Contents

VII

18

33

65

88

103

120
136
150
161
182

Foreword
Collin, M. (Lund); Schuch, R. (New York, N.Y.)
Chemical Interactions between Organisms in Microbial Communities
Duan, K. (Calgary, Alta./Xian, Shaanxi); Sibley, C.D. (Calgary, Alta.); Davidson, C.J. (East
Lansing, Mich.); Surette, M.G. (Calgary, Alta.)
Autoinducer-2-Based Chemical Communication in Bacteria: Complexities of
Interspecies Signaling
Federle, M.J. (Chicago, Ill.)
The Molecular Basis of Excitation and Adaptation during Chemotactic Sensory
Transduction in Bacteria
Rao, C.V.; Ordal, G.W. (Urbana, Ill.)
Bacterial PEP-Dependent Carbohydrate: Phosphotransferase Systems Couple
Sensing and Global Control Mechanisms
Lengeler, J.W. (Magdeburg); Jahreis, K. (Osnabrck)
Correlations between Carbon Metabolism and Virulence in Bacteria
Poncet, S.; Milohanic, E.; Maz, A.; Nait Abdallah, J.; Ak, F. (Thiverval-Grignon);
Larribe, M.; Deghmane, A.-E.; Taha, M.-K. (Paris); Dozot, M.; De Bolle, X.;
Letesson, J.J. (Namur); Deutscher, J. (Thiverval-Grignon)
Stand-Alone Response Regulators Controlling Global Virulence Networks in
Streptococcus pyogenes
McIver, K.S. (College Park, Md.)
The Heme Sensor System of Staphylococcus aureus
Stauff, D.L.; Skaar, E.P. (Nashville, Tenn.)
Bacterial Sensing of Antimicrobial Peptides
Otto, M. (Bethesda, Md.)
RNA Thermosensors in Bacterial Pathogens
Johansson, J. (Ume)
Prevailing Concepts of c-di-GMP Signaling
Rmling, U.; Simm, R. (Stockholm)
Magnetosomes and Magneto-Aerotaxis
Frankel, R.B. (San Luis Obispo, Calif.); Bazylinski, D.A. (Las Vegas, Nev.)

194

Engineering Bacterial Signals and Sensors


Salis, H.; Tamsir, A.; Voigt, C. (San Francisco, Calif.)

226

Author Index
Subject Index

227

VI

Contents

Foreword

Over the last 1015 years, the study of how bacteria sense their environment and
respond accordingly has emerged as a focal point in the field of microbiology. Not
surprisingly, the bacterial adaptive response is now described by a panoply of interesting mechanisms, signals, behaviors, etc., involving everything from the movement
of flagella to the formation social groupings. Bacterial Sensing and Signaling, a volume
of the Karger book series Contributions to Microbiology, was initiated with the hope of
introducing the results of state-of-the-art research from internationally recognized
experts.
Chemical communication is undoubtedly the best-studied mechanism for passing
information between bacterial organisms and coordinating their behavior. As such,
Duan et al. begin the first section of this book with an introduction to the distinct
array of chemical signals that shape bacterial community relationships. Michael
Federle then follows with a description of his work on one such signal, termed autoinducer-2, with a focus on its role in cell-cell communication (or quorum sensing)
among different bacterial species and the mechanism by which such a signal is transduced across the bacterial membrane.
In keeping with the theme of signal transduction, Rao and Ordal next present a
comprehensive review of the molecular mechanisms by which individual bacteria
sense environmental attractants and repellents and transduce this information to the
flagellar motor to evoke a locomotive response (chemotaxis). Chemotaxis is the most
thoroughly understood bacterial adaptive behavior, and its description serves to
introduce the roles of bacterial two-component and phosphotransferase systems
(PTSs) systems in signal transmission and the processes of receptor multimerization
and methylation that allow response adaptation. The importance of PTSs in the coupling of sensory and regulatory mechanisms is further pursued by Lengeler and
Jahreis, who describe both the rapid chemotactic responses to carbohydrates and the
delayed responses associated with catabolite repression, as well as the concepts adaptation, memory, and learning associated with these signaling systems.

VII

The next several chapters are devoted to studies of bacterial virulence and in particular, the mechanisms of host detection, nutrient acquisition and host-defense
avoidance. Poncet et al. begin with a description of how the PTS-mediated carbohydrate signaling pathway responds to host-specific conditions and drives virulence
factor expression in both Gram-positive and Gram-negative pathogens. Kevin McIver
follows with a review of the so-called stand-alone response regulators of the human
pathogen Streptococcus pyogenes that serve to fine-tune virulence factor expression at
appropriate times during infection. Stauff and Skaar then detail a heme-sensing twocomponent system that communicates with transmembrane channel proteins (ABC
transporters) in pathogenic staphylococci and related Gram-positive bacteria. This
mechanism allows bacteria to acquire essential iron from host hemoglobin and avoid
the toxic effects of heme. In the subsequent chapter, Michael Otto covers how bacterial pathogens adapt to the human host by responding to human antimicrobial peptides (AMPs). Here, the author discusses how two-component systems induce
expression of resistance mechanisms against AMPs and how antimicrobial compounds are themselves signaling molecules.
The next three chapters present some of the more unusual and novel aspects of
bacterial signaling and sensing processes identified this far. In his chapter, Jrgen
Johansson introduces RNA thermosensors and the principles by which environmental temperature controls access to RNA expression signals and regulates virulence
gene expression in a variety of human pathogens. Rmling and Simm describe the
current knowledge about the signaling molecule cyclic-di-GMP (in fact a small cyclic
RNA) required for many bacterial responses including virulence regulation and the
transition between sessility and motility in bacteria. Next, Frankel and Bazylinski
review the intriguing field of magneto-aerotaxis which is based on the use of magnetic nanoparticles (magnetosomes) in the bacterial cytoplasm that serve to orient
bacteria and ensure migration within geomagnetic fields and maintenance of positions within favorable oxygen concentrations.
The last chapter, by Salis et al., describes how the individual abilities of bacteria to
sense and alter their environment can be harnessed for bioengineering applications.
By incorporating these capabilities into synthetic gene networks, bacteria can actually
be programmed for specific interactions and functions in the physical world.
We thank all of the authors who contributed to this volume. They have provided us
with comprehensive, interesting, and well-written chapters despite numerous other
duties and engagements. We further thank the series editor Dr. Heiko Herwald for
initiating this volume and Mr. Thomas Nold from Karger Publishers for helpful assistance and encouragement.
Mattias Collin, PhD, Lund, Sweden
Raymond Schuch, PhD, New York, USA

VIII

Foreword

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 117

Chemical Interactions between Organisms in


Microbial Communities
Kangmin Duana,c Christopher D. Sibleya Carla J. Davidsond
Michael G. Surettea,b
Departments of aMicrobiology and Infectious Diseases and bBiochemistry and Molecular Biology, University of
Calgary, Faculty of Medicine, Calgary, Alta., Canada; cMolecular Microbiology Laboratory, Faculty of Life Sciences,
Northwestern University, Xian, Shaanxi, China, and dDepartment of Microbiology and Molecular Genetics,
Michigan State University, East Lansing, Mich., USA

Abstract
Bacteria live almost exclusively in communities with other microorganisms, and often in association
with multicellular hosts. These communities are capable of maintaining complex structural and
functional stability over time, and exhibit fascinating properties of resiliency in response to environmental changes. This is a result of interactions between microbes and the environment and amongst
members of the community. A multitude of chemical interactions occur in microbial communities
where primary and secondary metabolites contribute to a wealth of interactions between organisms. The chemicals include a variety of nutrients, toxic or neutral metabolic byproducts, antibiotics,
and cell-cell signaling molecules. These chemical and physical signals facilitate microbial relationship that can be competitive, cooperative or neutral, and thus are responsible for determining community structure. In turn, the surrounding community changes the microenvironment of individual
cells who respond to chemical and environmental cues in a combinatorial manner. Current laboratory understanding of the genetics and mechanisms of interactions between microbes has the
power to help us understand how complex microbial communities behave in the natural environment. In this chapter we review the current understanding of microbial communication, from the
genetic and molecular aspects, to our current understanding of their ecological role.
Copyright 2009 S. Karger AG, Basel

Passive and Active Interactions between Bacteria

Microbial communities exhibit all the hallmarks of complex and social life [1].
Members in the community communicate, cooperate and compete to perform a wide
range of multicellular behaviors such as dispersal, foraging, biofilm formation, chemical warfare and quorum sensing [2]. These complex behaviors are mitigated by different kinds of interactions between individual bacteria. Interactions can be passive,

wherein one bacterium removes or alters a common resource, or active, wherein signals are produced which are specifically designed to alter the behavior of neighboring
bacteria.
In the simplest passive interaction, two individuals or populations compete for
a common resource such as nutrients or location. These competitions can be either
intra- or inter-strain/specific, but remain passive (fig. 1a). Competition experiments
between strains in defined media are often used to evaluate this type of interaction.
While the simplicity of this interaction suggests that the outcome can be easily predicted, in fact, demographic pressures and intake rates can lead to multiple outcomes,
such as coexistence or dominance of one bacterium over another [3]. Furthermore,
even in defined, homogeneous conditions evolution can lead to complex populations
[4]. In static (non-shaken) culture, the establishment of spatial heterogeneity can lead
to more complex competitive environments, such as adaptive radiation in standing
cultures of Pseudomonas fluorescens [57].
Competition can also be active, where one cell directly interferes with another (fig.
1b). For instance the production of bacteriocins, lantibiotics and other secondary
metabolites with antibiotic activity directly inhibit the growth of competitors. The
production of antibiotic-quenching enzymes such as lactonases that degrade N-acylhomoserine lactone (AHL) signals [8, 9] also represents active competition between
strains or species where one strain disrupts the normal regulation of another quorum-sensing organism.
In many communities, syntrophic interactions play important roles. These arise
when two or more strains cooperate in the metabolism of a resource (fig. 1c). A
simple example would be where the process involves multiple steps, with different
steps in the pathway carried out by different bacteria. This can lead to multicellullar
organization, as in the filamentous cyanobacteria where specialized cells, called heterocysts, develop in response to nitrogen limitation [1012]. These photosynthetic
bacteria separate the site of nitrogen fixation from photosynthesis to protect nitrogenase from the degradative effects of oxygen produced by photosynthesis. This results
in the development of a highly regulated spatial pattern mediated by peptide signaling [11], which is unusual among the Gram-negative bacteria.
Passive cooperative behavior can occur when cells/populations exploit the same
resource through a shared mechanism. An example would be the utilization of a complex resource with secreted enzymes, such as degradation of protein substrates by
exoproteases. Since both the enzyme and the products of digestion will diffuse into
the surrounding environment, this can be an inefficient mechanism for a single cell
and efficiency will asymptotically increase with cell density to a threshold concentration, after which competition dominates (fig. 1d). Within a population, coordinated
expression of exoenzymes when the cells reach a critical number is a common argument for quorum sensing and exoenzymes are frequently regulated in this way by bacteria. This would be an example of active cooperation (fig. 1e). It has also been argued
that the QS signals play a role in measuring the diffusivity of the environment [13].

Duan Sibley Davidson Surette

x
A

d Passive cooperation

a Passive competition
x
A

B
B

z
y

y
b Active competition

c Syntrophic interactions

e Active cooperation

Fig. 1. Summary of simple competitive and cooperative interactions between two cells or populations represented by A and B. Substrates and products are represented by x, y and z. These can be
nutrients but also can refer to binding sites in the case of physical interactions with the environment
or each other. They can be either passive or active interactions.

Examples of interspecies signaling in any of the above scenarios are rare and
often the pathways implicated in such examples are not well understood. Frequently,
interaction results in increased ability to grow together such as in the formation of
mixed-species biofilms. Furthermore, both passive and active cooperation are prone
to cheating where cells that do not contribute to the common good (such as secreting protease) exploit the resource. Thus these types of cooperative interactions might
be less common than the literature might suggest and additional factors might contribute to maintaining these interactions when they do exist. Indeed simple interactions between cells are probably rare; it is more likely individual cells must integrate a
number of inter- and intraspecific signals. The interactions will depend on numerous
factors that will vary over time and environmental conditions.
The importance of chemical interactions in natural microbial communities is an
area of intense research. Experiments addressing the problem of culturing microbes
in the laboratory hint at their potential importance. For example the majority of bacteria from any environment fail to grow on laboratory media [14] and it is unknown
whether the cause of this deficiency lies in the media composition or lack of community structure. These uncultivated organisms represent the great majority of microbes
and have the potential to contain many useful organisms and secondary metabolites.
Recent efforts to increase the recovery of uncultivated organisms have demonstrated the importance of small molecule interactions. Attempts to grow microbes
using diffusion chambers placed back in their natural environments results in

Chemical Interactions in Microbial Communities

a dramatic increase in the number and types of organisms recovered [15, 16]. For
obligate syntrophs this is expected, as growth would require the partner organisms
or the particular metabolites (or a means to remove toxic byproducts). Acquisition
of micronutrients using chemicals produced by other cells may also be a common
theme. Siderophores seem to play such an important role in some organisms ability
to grow [17]. These bacteria require iron but do not produce the chelating compounds
required to scavenge it [18, 19]. They appear to scavenge siderophores produced by
other bacteria (termed xenosiderophores) in order to grow.
Some organisms that are found initially to grow only in co-culture with other
organisms appear to rely on small molecule growth factors rather than nutritional
supplements produced by their partners. The helper bacteria provide some factor
necessary for the growth of the strain of interest. A recent study demonstrated that a
Psychrobacter isolate could grow when provided with a peptide produced by another
organism [20]. The authors demonstrated that a specific pentapeptide at relatively
low concentration (3.5 nm) was optimal for growth. This does not seem to be acting
as a specific nutrient, however the mechanism awaits further investigation.

Chemical Signaling: Communication, Cues and Chemical Manipulation

Interactions between bacteria in complex communities are mediated by chemical


messages that diffuse throughout the community. They are commonly referred to as
signal, cues, or communication. In particular, the broad use of the term signal in the
biological literature has confused the understanding of chemical interactions between
cells [21, 22]. A signal is any act, structure or chemical emission that alters the behavior of other organisms, in which both the chemical and the response it engenders has
evolved for this purpose [22, 23]. However, many cell-cell interactions may be inappropriately described as communication. For a signal to be classified as true communication it must be a compound that is created for the purposes of transmitting
information, can be perceived by other cells and it must engender a response in the
receiver. Furthermore, in order for this communication to be evolutionarily stable,
it is necessary for both the sender and the receiver to benefit from the exchange.
Therefore, the act of responding to a chemical that is produced by another organism
cannot be unequivocally considered a response due to communication. For example,
a bacterium that releases a metabolite for the purposes of removing toxins from the
cell, which can be sensed by a neighboring bacterium, is not engaging in communication. Rather, the neighbor is using this metabolite as a cue.
True interspecies communication requires cooperation. Cooperation is difficult
to explain in bacteria in that it requires an element of altruism, which, by definition,
increases another individuals fitness at a personal cost [24, 25]. For example, bacteria commonly produce extracellular scavenging compounds such as siderophores,
which are metabolically costly. In the environment, mutants can arise who do not

Duan Sibley Davidson Surette

produce siderophores, but still benefit from those produced by the rest of the population. Because they do not incur the cost of producing siderophores, these cheats can
actually have a competitive advantage over those cells that do produce them. This
has been elegantly illustrated using Pseudomonas aeruginosa mixed populations with
wild-type and siderophore mutants; without the metabolic burden of siderophore
production, cheaters will out compete the wild-type bacteria [26]. Clearly, the possibility of cheats arising in a population leads to an unstable situation that requires
either policing or population demographic pressures to maintain altruism within the
population. The selective advantage of altruism was nicely demonstrated in the case
of colicin production, in which the production of colicins results in the lysing of the
cell and is thus a true case of altruism. In this case, the behavior is selected in spatially
heterogeneous environments as a result of kin selection. In a heterogeneous, nonshaken environment, closely related individuals will be spatially associated with the
lysing cell, thus the destruction of non-related individuals in close proximity benefits
related kin by releasing some competitive pressure [27]. The topics of altruism, kin
selection and cheaters are beyond the scope of this chapter and we refer the reader
to several recent reviews that have thoroughly discussed the topic [23, 2831]. It is
important to emphasize, however, that true communication is subject to similar constraints as other altruistic behaviors.

Secondary Metabolites and Chemical Interactions on Bacteria

In order to understand the nature of competition and cooperation between microbes


(which ranges from marginal support to mutual dependence [32]) and between
microbes and higher organisms, it is important to identify and define the small bioactive metabolites that are the mediators of such relationships. They are often derived
from secondary metabolites; molecules that are biosynthesized via pathways that
branch off from essential metabolism. The compounds produced via such pathways
are low-molecular-weight molecules (<1,500 daltons) that are often produced during
a specific developmental stage of growth. They have varied chemical structures that
sometimes share a high degree of similarity to cofactors that are involved in primary
metabolism, but they have no essential function. The natural roles of these small molecules in biology have not been rigorously investigated and exploring their functions
in natural environment will revolutionize our understanding chemical interactions
between all forms of life [33].
The wide variety of secondary metabolites produced by bacteria and their seeming lack of utility begs the question of why bacteria invest in these energetically
expensive pathways. It is likely that in the natural environment, secondary metabolites serve unknown beneficial functions. However, when produced in pure culture
in the laboratory, they have a limited benefit to the producing organism. It has been
hypothesized that secondary metabolites once played a role in modulating ancient

Chemical Interactions in Microbial Communities

biochemical reactions such as primitive transcription and translation, but have been
subsequently replaced by polypeptides [34]. However, because they retained their
binding ability to DNA and proteins, secondary metabolites may have evolved from
molecules intimately associated with the primary synthesis of essential macromolecules, to be antagonists of such processes [34].
The importance of secondary metabolites cannot be overstated; it has been hypothesized that the doubling of the human lifespan in the 20th century is mainly due to
the use of plant and microbial secondary metabolites as antibiotics [35]. The demand
for these compounds has created a market that increases USD 30 billion per year [36].
More than 5,000 natural products have been discovered that have antimicrobial properties. However, most of these compounds are not clinically useful because they are
either too toxic or inactive in the human body. The actinomycetes can be credited as
being Mother Natures best chemists. Roughly 75% of all antibiotics are produced by
actinomycetes, of these, 75% are produced by a single genus, Streptomyces. Strains of
Streptomyces hydroscopicus can produce over 180 different secondary metabolites [2].
Many streptomycetes possess the genetic capacity to produce upwards of 25 different
secondary metabolites [37, 38]; most of these compounds are not produced under
laboratory conditions suggesting that they may only have a role in natural environments within microbial communities.

Bacterial Quorum Sensing

Bacterial quorum sensing has been extensively studied and is primarily a mechanism for cell-cell signaling within a population of cells [3941]. In its simplest form,
quorum sensing is mediated by the secretion of specific molecules and subsequent
response when they reach a threshold concentration. There are several classes of
compounds that have been described; they are often termed autoinducers since in
many systems the signal positively regulates its own synthesis. The five main classes
of compounds involved in bacterial quorum sensing are oligopeptides, N-AHLs,
quinolones, -butyrolactones and furanones. This is an active area of research and
in fact many new signals are still being identified, including a new class of AHLs
[42] and (S)-3-hydroxytridecan-4-one [43]. Most quorum-sensing signals are
secondary metabolites, a class of abundant, diverse, and largely underexplored molecules [33].
In Gram-positive bacteria the quorum-sensing molecules are often oligopeptides [4446]. They are found in the cells as a pre-peptide, which is inactive until
it is cleaved during secretion through the plasma membrane, and in some cases the
mature peptide is modified further. The signal freely diffuses in the extracellular environment and is typically received by neighboring cells by two component regulatory
systems. The peptide signal binds to the membrane-bound histidine kinase and modulates its kinase activity, resulting in the phosphorylation of the response regulator.

Duan Sibley Davidson Surette

The phosphorylated response regulator is an active transcriptional factor which then


regulates transcription of numerous target genes.
The second pathway of quorum sensing is mediated by N-AHL molecules which
are to date only known in Gram-negative bacteria [4749]. The system was first elucidated in Vibrio fischeri, which luminesce at high cell densities. In V. fischeri the
luxI gene encodes an enzyme that catalyses the synthesis of an N-3-(oxohexanoyl)
homoserine lactone autoinducer (3-OC6-AHL) [50]. At high concentrations the AHL
binds to the transcriptional regulator LuxR, which is then capable of binding to the
lux promoters, resulting in luminescence production. Subsequently, AHL mediated
quorum-sensing systems have been found in over 50 Gram-negative bacteria [51].
The synthase gene and response regulator gene are often homologous to V. harveyi
luxI and luxR, but other synthase and receptor system are known [52, 53]. Recently,
another type of signal that is related to the AHL signals is the aryl-homoserine lactones found in photosynthetic bacterium Rhodopseudomonas palustris [42]. Instead
of using the cellular fatty acyl group from fatty acid biosynthesis, the bacterium uses
p-coumaric acid from the environment to produce the aryl-homoserine lactone signal. Other bacteria found to produce this type of signals include Bradyrhizobium sp.
and Silicibacter pomeroyi [42].
The LuxS/autoinducer-2 (AI-2) system is mediated by a group of furanones [54,
55]. These furanones are derived from unstable 4,5-dihydroxy-2,3-petanedione
produced by LuxS [56, 57]. The LuxS/AI-2 system is found in both Gram-positive
and Gram-negative bacteria [51, 58], and because the same signal molecule is found
among bacteria, it is thought to be a universal signaling molecule interpretable by
many different species of bacteria. Further studies indicate the receptors for AI-2 in
Vibrio harveyi and Salmonella enterica serovar Typhimurium bind to different stereoisomers, the S- and R-forms of 2-methyl-2,3,3,4-tetrahydroxytetrahydrofuran,
respectively [54, 59]. Because the synthesis of AI-2 is linked to a central metabolic
pathway, the role of the AI-2/LuxS quorum-sensing signal has been in debate [58, 60,
61]. Instead, AI-2 may be a primary metabolite that has subsequently been adopted as
a cue to the density of cells in the immediate environment.
The A-factor in Streptomyces species is a representative type of -butyrolactones
that controls a variety of pathways including antibiotic production and aerial mycelium formation [6264]. This was one of the first cell-cell signaling systems described.
The -butyrolactone molecules were found to act by binding cytoplasmic receptors
within the cell. In general, these receptors are transcriptional repressors; once bound
with the -butyrolactone the transcriptional repressors release their binding sites
resulting in activation of numerous cellular responses [62].
These signaling systems vary in their specificity, cost and the response they engender, and in fact, many bacteria utilize multiple cell-cell signaling systems. P. aeruginosa possesses two AHL signaling pathways, the las and rhl systems [6568] and
an additional system mediated by 2-heptyl-3-hydroxy-4-quinolone (Pseudomonas
Quinolone Signal, PQS) [69]. This system acts to regulate the expression of a

Chemical Interactions in Microbial Communities

Quorum sensing

Cues and
chemical manipulators

Autoinducers
systems

QS targets

Other secondary
metabolites
(e.g. antibiotics)

Microorganism

Microorganism

Autoinducers

Environmental conditions
Temperatures, pH, nutrient levels, etc.

Fig. 2. Integration of quorum sensing with environmental signals.

number of target genes such as elastase (lasB) and phenazine synthesis (phzABCDE).
Interestingly, these genes are also targets of the AHL quorum-sensing signals, thus
multiple streams of information regarding the immediate environment can be integrated at the promoter level.
As well as being cell density-dependent, quorum-sensing systems are often modulated by environmental factors. In P. aeruginosa, there is a complex regulatory network involving at least 16 regulators that control quorum-sensing systems both at
the transcriptional and post-transcriptional levels. Information regarding temperature, pH, stress and nutrient depletion are integrated with quorum-sensing signals to
orchestrate a coordinated response (fig. 2).

Communication, Cues, Eavesdropping, and Interference

Quorum sensing as defined as cell-cell signaling within a population of clonal cells


and mediated by AHL or oligopeptides fits the stringent definition of communication. However, the role of AI-2 as a mediator of interspecies communication is not as
clear; AI-2 is derived from the toxic metabolic intermediate S-adenosyl-methionine
thus is equally important as a metabolite as a signaling molecule. Response to AI-2
concentrations may therefore not be the result of cell-cell communication but simply
the adoption of this metabolite as a cue.
Our laboratory has reported on the use of a cue by the opportunistic pathogen P.
aeruginosa from the oropharyngeal flora from cystic fibrosis (CF) patients. CF airways are complex polymicrobial environments in which the principal CF pathogen, P.
aeruginosa, is exposed to a number of bacterial species that produce the interspecies
signal AI-2. P. aeruginosa does not have the ability to produce AI-2 but responds to

Duan Sibley Davidson Surette

it by modulating the expression of a number of genes, many of which are important


virulence factors [70]. In this situation, there is likely no benefit gained by the Grampositive oropharyngeal flora in reporting the metabolic activity and spatial location
of the community to the pathogen. Therefore it is unlikely that AI-2 serves as a communication signal, but rather is a cue that is sensed by P. aeruginosa. It is intriguing to consider whether such interactions might be common mechanisms exploited
by pathogens during colonization of polymicrobial colonized host tissue. Another
example of small-molecule cueing comes from the CF lung whereby two common
CF pathogens chemically interact. Burkholderia cepacia responds to AHLs produced
by P. aeruginosa by increasing the production of virulence factors. P. aeruginosa does
not respond in the same way to B. cepacia and does not benefit from its presence,
therefore, communication between the two organisms is not occurring. This behavior
probably does not represent an evolved, coordinated behavior but rather a fortuitous
overlap in signal molecules that are produced and the range of signals to which each
bacteria can respond [23, 71, 72].
Finally, there is a type of chemical interaction referred to as chemical manipulation
or coercion. Chemical manipulation is a chemical emission that alters the behavior
and gene expression of other organisms, but contrary to communication, results in a
negative effect on the receiver. For example, Streptococcus gordonii can ferment sugars by a number of different mechanisms, one of which yields lactic acid. Veillonella
atypical cannot ferment sugars but utilizes lactic acid as a preferred carbon source.
When S. gordonii and V. atypica are grown in co-culture, on solid agar plates, S. gordonii amylase activity is induced by an as yet unknown chemical thereby increasing
the degradation of complex carbohydrates into lactic acid [73]. This may engender an
energetic cost to S. gordonii, therefore, the interaction is considered chemical manipulation by V. atypica.
With the exception of LuxS/AI-2, quorum-sensing systems are quite species/strain
specific. Interestingly, many bacteria have been shown to possess enzymes (e.g. lactonases and esterases) capable of degrading AHLs [8, 9]. It may be that AHL degradation
provides a competitive advantage to the producer by interfering with quorum sensing
in competing bacteria, although the role of these enzymes in natural environments
has yet to be established. Many bacteria have a wide variety of enzymes to catabolize
a broad spectrum of potential carbon sources and these lactonases and esterases may
function primarily in catabolism rather than in competition. However, as discussed
below, many compounds involved in interspecies competition or communication are
derived from chemicals involved in other pathways. Thus, roles in competition and
catabolism may not be mutually exclusive, and their effects may depend on local species density and environment.
Another intriguing example of interference comes from Staphylococcus aureus.
These Gram-positive bacteria signal using cyclized peptides, and the genes regulated
include virulence determinants [74]. Different strains of S. aureus produce different
signal peptides which can be grouped according to their cross-reactivity. Interestingly,

Chemical Interactions in Microbial Communities

peptides from one group are potent inhibitors of peptides from other groups [74, 75].
This is an example of chemical interference competition that impairs the pathogenesis of individuals of other strains, as well as a signaling molecule that has dual roles.

Antibiotics in Microbial Communities

Antibiotics are secondary metabolites with bacteriocidal or growth-inhibiting activity that are utilized by bacteria in microbial communities. In fact in a microbial community an antibiotic produced by one bacterial species may preclude the existence
of another species. For example, in dental plaque communities, bacteriocins play
such a competitive role. Bacteriocins are narrow spectrum proteinaceous antibiotics that have been found in many bacteria. Streptococcus mutans strains in plaque
produce a number of distinct bacteriocins, also termed mutacins [76]. Some of these
bacteriocins inhibit the growth of Streptococcus sanguinis strains. These bacteriocins
are probably responsible for the relatively low levels of S. sanguinis strains in plaque
containing high proportions of S. mutans [77]. In this way, some biofilm residents
may determine which other organisms coexist in these structures, and thus help the
establishment of a community with specific bacterial species, ensuring an ecological
balance of the oral microbial ecosystem.
Though bacteriocidal secondary metabolites are used in the environment, our use
of high concentrations of secondary metabolites as antibiotics may have clouded their
other roles in the environment [78]. When used in high concentrations, many secondary metabolites are bacteriocidal, but at lower concentrations these compounds
are potent modulators of gene expression. Thus it is probable that the main function
of the secondary metabolites in the environment is not antibiosis but rather as global
regulators of transcription [21, 33, 79, 80]. Up to 5% of the bacterial transcriptome
can be specifically modulated (activated or repressed) in the presence of subinhibitory concentrations of antibiotic, and this modulation of expression is independent of
stress responses [81, 82]. This suggests that secondary metabolites have the potential
to act as cues, chemical manipulators or signals.
In natural environments, antibiotics are likely present at subinhibitory concentrations and play a role other than growth inhibition [83, 84]. For example, it has been
shown that subinhibitory concentrations of exogenously added aminoglycosides induce
P. aeruginosa and Escherichia coli to grow as biofilms [85]. This implies that aminoglycosides are present in the environment and bacteria have adapted to respond to this cue.

Interrelationship of Quorum Sensing and Antibiotics in Microbial Communities

The fine line between signal and antibiotic also applies to quorum-sensing signals;
many of these have been shown to have antibiotic activity at high concentrations.

10

Duan Sibley Davidson Surette

For example, many oligopeptide autoinducers produced by Gram-positive bacteria,


and all of the quorum-sensing signals found in P. aeruginosa have antibiotic activities against other bacteria. In these cases, the inhibitory function of these autoinducers argues that they are not mediators of communication but rather cues sensed by
other species or molecules that manipulate or coerce specific responses from other
bacteria.
Pathways that respond to antibiotics and quorum-sensing signals may also be integrated. Genes regulated by subinhibitory concentrations of antibiotics such as those
encoding exoenzymes, lasB, rhlAB and phzA1B1C1D1E1 in P. aeruginosa are also
under the control of quorum-sensing systems [86]. It has been reported that subinhibitory concentrations of ceftazidime and tobramycin impair the production of quorum-sensing signals of P. aeruginosa [87]. Macrolides such as azithromycin have also
been reported to capable of suppressing quorum-sensing activity as well as the production of virulence factors [88, 89]. Combinatorial control of regulation is common
in many bacterial pathways and the co-regulation of many targets by antibiotics and
quorum-sensing signals suggests that they might have a common ecological role.
For example, phenazine production in P. aeruginosa is regulated by both subinhibitory concentrations of antibiotics and quorum-sensing signals. As stated
above, P. aeruginosa possesses three quorum-sensing systems: a quinolone signal
(PQS) system mediated by 2-heptyl-3-hydroxy-4-quinolone [69] and two N-AHLmediated quorum-sensing systems (the las and rhl systems) [6568]. Phenazine
compounds are broad-spectrum antibiotic metabolites produced by bacterial species
in Brevibacterium, Burkholderia, Pseudomonas and Streptomyces genera. They play
important roles in bacterial competitiveness in microbial communities [90] and function as important virulence factors [91, 92]. Two homologous operons are involved
in the synthesis of phenazine compounds in P. aeruginosa, phzA1B1C1D1G1 (phzA1)
and phzA2B2C2D2G2 (phzA2) [93]. Both the PQS and rhl system positively regulate phzA1 expression [67, 94], and the orphan LuxR-type quorum-sensing regulator
QscR negatively regulates the expression of phzA1, phzA2 as well as the elastase gene
lasB [95]. The results of our recent studies indicate that phzA1 operon is also activated by subinhibitory concentrations of tetracycline [86].
The regulation of phenazine compound synthesis, exoenzyme and protease expression by subinhibitory concentrations of tetracycline suggests that in a microbial community the antibiotics secreted by other microorganisms could serve as a signal to
alert P. aeruginosa of the existence of other bacteria. The subsequent increased pyocyanin production would help P. aeruginosa to eliminate competition. Genes such
as efflux pumps are also upregulated by subinhibitory concentrations of tetracycline
[unpubl. data], probably to expel antibiotic compounds. It has been estimated that
0.1% of soil actinomycetes produce tetracycline; it is therefore a prevalent antibiotic
in this community [96]. As a bacterium living in soil microbial community, P aeruginosa would have been exposed to these compounds throughout its evolution, and not
solely as a result of the clinical use of tetracyclines. The integration of these intrinsic

Chemical Interactions in Microbial Communities

11

regulatory systems suggests that the microbial community in which P. aeruginosa


resides is both complex and competitive.

Physical Interactions in Microbial Communities

It is useful to consider physical interactions in microbial communities because in many


cases they are a prerequisite for chemical interactions to occur. The best-characterized
networks of physical interactions between microbes come from the oral communities
[97, 98]. In the mouth there is a strong selective pressure to form mixed-species biofilms. Individual species of oral bacteria usually do not contain complete metabolic
pathways to degrade complex carbohydrates and, thus must form syntrophic communities. The consortia of microbes in a highly ordered spatial community use a syntrophic biochemical cascade to sequentially metabolize these compounds. Examples
of potential metabolic mutualisms also exist, such as between Streptococcus oralis and
Actinomyces naeslundii. Neither of these organisms is capable of growth on human
saliva as a sole carbon and nitrogen source in monoculture, but both are capable of
growing in saliva when co-cultured. Moreover, the growth of each species in co-culture appear to rely on contact-dependent, cooperative chemical interactions, because
isolated single-species colonies in the co-culture biofilm tended not to develop as well
as the coaggregated colonies [99].
Coaggregation implies the ability to interact physically with different species
and strains, and is a distinguishing feature of oral microbial communities. Using
a retrievable enamel chip system it was observed that after professional cleaning
few microbes are found associated with the enamel, and that colonization of the
tooth surface develops according to reproducible patterns in a process of succession
[100]. Initially, pioneering microbes attach to the enamel surface by interacting
with saliva-derived receptors and are dispersed in small clusters of very few cells
[100]. Streptococci represent the majority of the pioneering species; Actinomyces
spp., Veillonella spp., Neisseria spp., Prevotella spp., Gemella spp., Abiotrophia spp.,
and Rothia spp. also participate in primary colonization events [101103]. The subsequent attachment of secondary and tertiary colonizing species generates a mixed
species biofilm comprised of hundreds of different species. Fluorescence in situ
hybridization studies reveal that only 8 h after cleaning, the microbial communities
are already very diverse [99, 104]. More than 500 known species have been isolated
from the mouth [105], and 50% of these organisms cannot be cultured in the laboratory [106]. The evidence for the intimate cell-cell physical recognition between
oral bacteria has been demonstrated using an antibody approach. These experiments show that specific microbe-microbe interactions during secondary and tertiary colonization, typically involving one partner expressing a surface-associated
ligand for a cognate surface-exposed receptor (often a polysaccharide) by the other
partner [100].

12

Duan Sibley Davidson Surette

Perspective

Defining the role of competitive and cooperative interactions in microbial communities represents a challenge in microbial ecology. Chemical interactions will play an
important role in these communities but may not be as straightforward as monoculture experiments might indicate. The fact that quorum-sensing compounds have antibiotic activity at high concentration and antibiotics can have signaling-like properties
at low concentrations suggests that chemical interactions in bacterial communities
are potentially extremely complex. Because spatial gradients of signals, resources
and population density are the norm in most natural environments, these secondary
metabolites may play multiple roles within a community. Understanding of the contribution of different mechanisms of interactions and the role each organism plays in
overall community structure and dynamics will be a difficult but exciting frontier in
microbiology.

Acknowledgements
We thank members of the Surette laboratory for valuable discussion. M.G.S. is supported as an
Alberta Heritage Foundation for Medical Research (AHFMR) Scientist and Canada Research Chair
in Microbial Gene Expression. C.D.S. is supported by an AHFMR studentship and a Canadian
Graduate Scholarship from CIHR. C.J.D. is supported by the Defense Advanced Research Projects
Agency Fun Bio Program (HR0011-05-1-0057).

References
1 Crespi BJ: The evolution of social behavior in
microorganisms. Trends Ecol Evol 2001;16:178
183.
2 Demain AL, Fang A: The natural functions of secondary metabolites. Adv Biochem Eng Biotechnol
2000;69:139.
3 Hansen SR, Hubbell SP: Single-nutrient microbial
competition: qualitative agreement between experimental and theoretically forecast outcomes. Science
1980;207:14911493.
4 Rozen DE, Schneider D, Lenski RE: Long-term
experimental evolution in Escherichia coli. XIII.
Phylogenetic history of a balanced polymorphism. J
Mol Evol 2005;61:171180.
5 Bantinaki E, Kassen R, Knight CG, Robinson Z,
Spiers AJ, Rainey PB: Adaptive divergence in experimental populations of Pseudomonas fluorescens.
III. Mutational origins of wrinkly spreader diversity.
Genetics 2007;176:441453.
6 Rainey PB, Travisano M: Adaptive radiation in a
heterogeneous environment. Nature 1998;394:69
72.

Chemical Interactions in Microbial Communities

7 Spiers AJ, Kahn SG, Bohannon J, Travisano M,


Rainey PB: Adaptive divergence in experimental
populations of Pseudomonas fluorescens. I. Genetic
and phenotypic bases of wrinkly spreader fitness.
Genetics 2002;161:3346.
8 Dong YH, Gusti AR, Zhang Q, Xu JL, Zhang LH:
Identification of quorum-quenching N-acylhomoserine lactonases from Bacillus species. Appl
Environ Microbiol 2002;68:17541759.
9 Dong YH, Zhang XF, Xu JL, Zhang LH: Insecticidal
Bacillus thuringiensis silences Erwinia carotovora
virulence by a new form of microbial antagonism,
signal interference. Appl Environ Microbiol 2004;
70:954960.
10 Zhang CC, Laurent S, Sakr S, Peng L, Bedu S:
Heterocyst differentiation and pattern formation in
cyanobacteria: a chorus of signals. Mol Microbiol
2006;59:367375.
11 Golden JW, Yoon HS: Heterocyst formation in
Anabaena. Curr Opin Microbiol 1998;1:623629.
12 Golden JW, Yoon HS: Heterocyst development in
Anabaena. Curr Opin Microbiol 2003;6:557563.

13

13 Redfield RJ: Is quorum sensing a side effect of diffusion sensing? Trends Microbiol 2002;10:365370.
14 Rappe MS, Giovannoni SJ: The uncultured microbial majority. Annu Rev Microbiol 2003;57:369
394.
15 Kaeberlein T, Lewis K, Epstein SS: Isolatinguncultivable microorganisms in pure culture in a
simulated natural environment. Science 2002;296:
11271129.
16 Bollmann A, Lewis K, Epstein SS: Incubation of
environmental samples in a diffusion chamber
increases the diversity of recovered isolates. Appl
Environ Microbiol 2007;73:63866390.
17 Wandersman C, Delepelaire P: Bacterial iron
sources: from siderophores to hemophores. Annu
Rev Microbiol 2004;58:611647.
18 Matzanke BF, Bohnke R, Mollmann U, Reissbrodt
R, Schunemann V, Trautwein AX: Iron uptake and
intracellular metal transfer in mycobacteria mediated by xenosiderophores. Biometals 1997;10:193
203.
19 Ratledge C, Dover LG: Iron metabolism in pathogenic bacteria. Annu Rev Microbiol 2000;54:881
941.
20 Nichols D, Lewis K, Orjala J, Mo S, Ortenberg R,
OConnor P, Zhao C, Vouros P, Kaeberlein T, Epstein
SS: Short peptide induces an uncultivable microorganism to grow in vitro. Appl Environ Microbiol
2008;74:48894897.
21 Diggle SP, Gardner A, West SA, Griffin AS:
Evolutionary theory of bacterial quorum sensing:
when is a signal not a signal? Philos Trans R Soc
Lond B Biol Sci 2007;362:12411249.
22 Maynard Smith J, D. H: Animal Signals. New York,
NY, Oxford University Press, 2003.
23 Keller L, Surette MG: Communication in bacteria:
an ecological and evolutionary perspective. Nat Rev
Microbiol 2006;4:249258.
24 Hamilton WD: The genetical evolution of social
behaviour. I. J Theor Biol 1964;7:116.
25 Hamilton WD: The genetical evolution of social
behaviour. II. J Theor Biol 1964;7:1752.
26 Griffin AS, West SA, Buckling A: Cooperation and
competition in pathogenic bacteria. Nature 2004;
430:10241027.
27 Chao L, Levin BR: Structured habitats and the evolution of anticompetitor toxins in bacteria. Proc
Natl Acad Sci USA 1981;78:63246328.
28 Brown SP, Johnstone RA: Cooperation in the dark:
signalling and collective action in quorum-sensing
bacteria. Proc Biol Sci 2001;268:961965.
29 Velicer GJ: Social strife in the microbial world.
Trends Microbiol 2003;11:330337.
30 West SA, Griffin AS, Gardner A, Diggle SP: Social
evolution theory for microorganisms. Nat Rev
Microbiol 2006;4:597607.

14

31 Lenski RE, Velicer GJ: Games microbes play.


Selection 2000;1:8995.
32 Schink B: Synergistic interactions in the microbial
world. Antonie Van Leeuwenhoek 2002;81:257
261.
33 Surette MG, Davies J: A new look at secondary
metabolites; in Winans SC, Bassler BL (eds): Chemical
Communication among Bacteria. Washington, ASM
Press, 2008.
34 Davies J: What are antibiotics? Archaic functions
for modern activities. Mol Microbiol 1990;4:1227
1232.
35 Verdine GL: The combinatorial chemistry of nature.
Nature 1996;384:1113.
36 Demain AL: Small bugs, big business: the economic
power of the microbe. Biotechnol Adv 2000;18:499
514.
37 Bentley SD, Chater KF, Cerdeno-Tarraga AM,
Challis GL, Thomson NR, James KD, Harris DE,
Quail MA, Kieser H, Harper D, Bateman A, Brown
S, Chandra G, Chen CW, Collins M, Cronin A,
Fraser A, Goble A, Hidalgo J, Hornsby T, Howarth
S, Huang CH, Kieser T, Larke L, Murphy L, Oliver
K, ONeil S, Rabbinowitsch E, Rajandream MA,
Rutherford K, Rutter S, Seeger K, Saunders D, Sharp
S, Squares R, Squares S, Taylor K, Warren T,
Wietzorrek A, Woodward J, Barrell BG, Parkhill J,
Hopwood DA: Complete genome sequence of the
model actinomycete Streptomyces coelicolor A3(2).
Nature 2002;417:141147.
38 Ikeda H, Ishikawa J, Hanamoto A, Shinose M,
Kikuchi H, Shiba T, Sakaki Y, Hattori M, Omura S:
Complete genome sequence and comparative analysis of the industrial microorganism Streptomyces
avermitilis. Nat Biotechnol 2003;21:526531.
39 Konaklieva MI, Plotkin BJ: Chemical communication do we have a quorum? Mini Rev Med Chem
2006;6:817825.
40 Williams P, Winzer K, Chan WC, Camara M: Look
whos talking: communication and quorum sensing
in the bacterial world. Philos Trans R Soc Lond B
Biol Sci 2007;362:11191134.
41 Reading NC, Sperandio V: Quorum sensing: the
many languages of bacteria. FEMS Microbiol Lett
2006;254:111.
42 Schaefer AL, Greenberg EP, Oliver CM, Oda Y,
Huang JJ, Bittan-Banin G, Peres CM, Schmidt S,
Juhaszova K, Sufrin JR, Harwood CS: A new class of
homoserine lactone quorum-sensing signals. Nature
2008;454:595599.
43 Higgins DA, Pomianek ME, Kraml CM, Taylor RK,
Semmelhack MF, Bassler BL: The major Vibrio cholerae autoinducer and its role in virulence factor production. Nature 2007;450:883886.

Duan Sibley Davidson Surette

44 Podbielski A, Kreikemeyer B: Cell density-dependent regulation: basic principles and effects on the
virulence of Gram-positive cocci. Int J Infect Dis
2004;8:8195.
45 Lyon GJ, Novick RP: Peptide signaling in
Staphylococcus aureus and other Gram-positive bacteria. Peptides 2004;25:13891403.
46 Sturme MH, Kleerebezem M, Nakayama J,
Akkermans AD, Vaugha EE, de Vos WM: Cell to
cell communication by autoinducing peptides in
Gram-positive bacteria. Antonie Van Leeuwenhoek
2002;81:233243.
47 Smith D, Wang JH, Swatton JE, Davenport P, Price
B, Mikkelsen H, Stickland H, Nishikawa K, Gardiol
N, Spring DR, Welch M: Variations on a theme:
diverse N-acyl-homoserine lactone-mediated quorum sensing mechanisms in Gram-negative bacteria. Sci Prog 2006;89:167211.
48 Waters CM, Bassler BL: Quorum sensing: cell-tocell communication in bacteria. Annu Rev Cell Dev
Biol 2005;21:319346.
49 Whitehead NA, Barnard AM, Slater H, Simpson NJ,
Salmond GP: Quorum-sensing in Gram-negative
bacteria. FEMS Microbiol Rev 2001;25:365404.
50 Fuqua C, Winans SC, Greenberg EP: Census and
consensus in bacterial ecosystems: the LuxR-LuxI
family of quorum-sensing transcriptional regulators. Annu Rev Microbiol 1996;50:727751.
51 Bassler BL: Small talk. Cell-to-cell communication
in bacteria. Cell 2002;109:421424.
52 Miller MB, Bassler BL: Quorum sensing in bacteria.
Annu Rev Microbiol 2001;55:165199.
53 Camilli A, Bassler BL: Bacterial small-molecule signaling pathways. Science 2006;311:11131116.
54 Miller ST, Xavier KB, Campagna SR, Taga ME,
Semmelhack MF, Bassler BL, Hughson FM:
Salmonella typhimurium recognizes a chemically
distinct form of the bacterial quorum-sensing signal AI-2. Mol Cell 2004;15:677687.
55 Rickard AH, Palmer RJ Jr, Blehertl DS, Campagna
SR, Semmelhack MF, Egland PG, Bassler BL,
Kolenbrander PE: Autoinducer 2: a concentrationdependent signal for mutualistic bacterial biofilm
growth. Mol Microbiol 2006;60:14461456.
56 Surette MG, Miller MB, Bassler BL: Quorum sensing in Escherichia coli, Salmonella typhimurium, and
Vibrio harveyi: a new family of genes responsible for
autoinducer production. Proc Natl Acad Sci USA
1999;96:16391644.
57 Schauder S, Shokat K, Surette MG, Bassler BL: The
LuxS family of bacterial autoinducers: biosynthesis
of a novel quorum-sensing signal molecule. Mol
Microbiol 2001;41:463476.
58 Xavier KB, Bassler BL: LuxS quorum sensing: more
than just a numbers game. Curr Opin Microbiol
2003;6:191197.

Chemical Interactions in Microbial Communities

59 Chen X, Schauder S, Potier N, Van Dorsselaer A,


Pelczer I, Bassler BL, Hughson FM: Structural identification of a bacterial quorum-sensing signal containing boron. Nature 2002;415:545549.
60 Vendeville A, Winzer K, Heurlier K, Tang CM,
Hardie KR: Making sense of metabolism: autoinducer-2, LuxS and pathogenic bacteria. Nat Rev
Microbiol 2005;3:383396.
61 Winzer K, Hardie KR, Williams P: Bacterial cell-tocell communication: sorry, cant talk now gone to
lunch! Curr Opin Microbiol 2002;5:216222.
62 Horinouchi S: A microbial hormone, A-factor, as a
master switch for morphological differentiation and
secondary metabolism in Streptomyces griseus.
Front Biosci 2002;7:d20452057.
63 Ueda K, Miyake K, Horinouchi S, Beppu T: A gene
cluster involved in aerial mycelium formation in
Streptomyces griseus encodes proteins similar to the
response regulators of two-component regulatory
systems and membrane translocators. J Bacteriol
1993;175:20062016.
64 Horinouchi S, Beppu T: A-factor as a microbial hormone that controls cellular differentiation and secondary metabolism in Streptomyces griseus. Mol
Microbiol 1994;12:859864.
65 Passador L, Cook JM, Gambello MJ, Rust L, Iglewski
BH: Expression of Pseudomonas aeruginosa virulence genes requires cell-to-cell communication.
Science 1993;260:11271130.
66 Pearson JP, Passador L, Iglewski BH, Greenberg EP:
A second N-acyl-homoserine lactone signal produced by Pseudomonas aeruginosa. Proc Natl Acad
Sci USA 1995;92:14901494.
67 Latifi A, Winson MK, Foglino M, Bycroft BW,
Stewart GS, Lazdunski A, Williams P: Multiple
homologues of LuxR and LuxI control expression of
virulence determinants and secondary metabolites
through quorum sensing in Pseudomonas aeruginosa PAO1. Mol Microbiol 1995;17:333343.
68 Ochsner UA, Fiechter A, Reiser J: Isolation, characterization, and expression in Escherichia coli of the
Pseudomonas aeruginosa rhlAB genes encoding a
rhamnosyltransferase involved in rhamnolipid biosurfactant synthesis. J Biol Chem 1994;269:19787
19795.
69 Pesci EC, Milbank JB, Pearson JP, McKnight S,
Kende AS, Greenberg EP, Iglewski BH: Quinolone
signaling in the cell-to-cell communication system
of Pseudomonas aeruginosa. Proc Natl Acad Sci
USA 1999;96:1122911234.
70 Duan K, Dammel C, Stein J, Rabin H, Surette MG:
Modulation of Pseudomonas aeruginosa gene
expression by host microflora through interspecies
communication. Mol Microbiol 2003;50:1477
1491.

15

71 Lewenza S, Visser MB, Sokol PA: Interspecies communication between Burkholderia cepacia and
Pseudomonas aeruginosa. Can J Microbiol 2002;48:
707716.
72 Riedel K, Hentzer M, Geisenberger O, Huber B,
Steidle A, Wu H, Hoiby N, Givskov M, Molin S,
Eberl L: N-Acylhomoserine-lactone-mediated communication between Pseudomonas aeruginosa and
Burkholderia cepacia in mixed biofilms. Microbiology 2001;147:32493262.
73 Egland PG, Palmer RJ Jr, Kolenbrander PE:
Interspecies communication in Streptococcus gordonii-Veillonella atypica biofilms: signaling in flow
conditions requires juxtaposition. Proc Natl Acad
Sci USA 2004;101:1691716922.
74 Ji G, Beavis R, Novick RP: Bacterial interference
caused by autoinducing peptide variants. Science
1997;276:20272030.
75 Lyon GJ, Wright JS, Christopoulos A, Novick RP,
Muir TW: Reversible and specific extracellular
antagonism of receptor-histidine kinase signaling. J
Biol Chem 2002;277:62476253.
76 Hamada S, Ooshima T: Production and properties
of bacteriocins (mutacins) from Streptococcus
mutans. Arch Oral Biol 1975;20:641648.
77 Kreth J, Merritt J, Shi W, Qi F: Competition and
coexistence between Streptococcus mutans and
Streptococcus sanguinis in the dental biofilm. J
Bacteriol 2005;187:71937203.
78 Upton M, Tagg JR, Wescombe P, Jenkinson HF:
Intra- and interspecies signaling between Streptococcus salivarius and Streptococcus pyogenes mediated by SalA and SalA1 lantibiotic peptides. J
Bacteriol 2001;183:39313938.
79 Yim G, Wang HH, Davies J: The truth about antibiotics. Int J Med Microbiol 2006;296:163170.
80 Calabrese EJ, Baldwin LA: Defining hormesis. Hum
Exp Toxicol 2002;21:9197.
81 Goh EB, Yim G, Tsui W, McClure J, Surette MG,
Davies J: Transcriptional modulation of bacterial
gene expression by subinhibitory concentrations of
antibiotics. Proc Natl Acad Sci USA 2002;99:17025
17030.
82 Tsui WH, Yim G, Wang HH, McClure JE, Surette
MG, Davies J: Dual effects of MLS antibiotics: transcriptional modulation and interactions on the
ribosome. Chem Biol 2004;11:13071316.
83 Davies J: Are antibiotics naturally antibiotics? J Ind
Microbiol Biotechnol 2006;33:496499.
84 Yim G, Wang HH, Davies J: Antibiotics as signalling
molecules. Philos Trans R Soc Lond B Biol Sci 2007;
362:11951211.
85 Hoffman LR, DArgenio DA, MacCoss MJ, Zhang Z,
Jones RA, Miller SI: Aminoglycoside antibiotics
induce bacterial biofilm formation. Nature 2005;
436:11711175.

16

86 Liang H, Li L, Dong Z, Surette MG, Duan K: The


YebC family protein PA0964 negatively regulates
Pseudomonas aeruginosa quinolone signal (PQS)
system and pyocyanin production. J Bacteriol 2008.
87 Garske LA, Beatson SA, Leech AJ, Walsh SL, Bell
SC: Subinhibitory concentrations of ceftazidime
and tobramycin reduce the quorum sensing signals
of Pseudomonas aeruginosa. Pathology 2004;36:571
575.
88 Nalca Y, Jansch L, Bredenbruch F, Geffers R, Buer J,
Haussler S: Quorum-sensing antagonistic activities
of azithromycin in Pseudomonas aeruginosa PAO1:
a global approach. Antimicrob Agents Chemother
2006;50:16801688.
89 Tateda K, Standiford TJ, Pechere JC, Yamaguchi K:
Regulatory effects of macrolides on bacterial virulence: potential role as quorum-sensing inhibitors.
Curr Pharm Des 2004;10:30553065.
90 Mazzola M, Cook RJ, Thomashow LS, Weller DM,
Pierson LS 3rd: Contribution of phenazine antibiotic biosynthesis to the ecological competence of
fluorescent pseudomonads in soil habitats. Appl
Environ Microbiol 1992;58:26162624.
91 Laursen JB, Nielsen J: Phenazine natural products:
biosynthesis, synthetic analogues, and biological
activity. Chem Rev 2004;104:16631686.
92 Lau GW, Hassett DJ, Ran H, Kong F: The role of
pyocyanin in Pseudomonas aeruginosa infection.
Trends Mol Med 2004;10:599606.
93 Mavrodi DV, Bonsall RF, Delaney SM, Soule MJ,
Phillips G, Thomashow LS: Functional analysis of
genes for biosynthesis of pyocyanin and phenazine1-carboxamide from Pseudomonas aeruginosa PAO1.
J Bacteriol 2001;183:64546465.
94 Diggle SP, Winzer K, Chhabra SR, Worrall KE,
Camara M, Williams P: The Pseudomonas aeruginosa quinolone signal molecule overcomes the cell
density-dependency of the quorum sensing hierarchy, regulates rhl-dependent genes at the onset of
stationary phase and can be produced in the absence
of LasR. Mol Microbiol 2003;50:2943.
95 Lequette Y, Lee JH, Ledgham F, Lazdunski A,
Greenberg EP: A distinct QscR regulon in the
Pseudomonas aeruginosa quorum-sensing circuit. J
Bacteriol 2006;188:33653370.
96 Baltz RH: Antimicrobials from actinomycetes: back
to the future. Microbes 2007;2:125131.
97 Kolenbrander PE, Andersen RN, Blehert DS, Egland
PG, Foster JS, Palmer RJ Jr: Communication among
oral bacteria. Microbiol Mol Biol Rev 2002;66:486
505.
98 Kuramitsu HK, He X, Lux R, Anderson MH, Shi W:
Interspecies interactions within oral microbial communities. Microbiol Mol Biol Rev 2007;71:653670.

Duan Sibley Davidson Surette

99 Palmer RJ Jr, Kazmerzak K, Hansen MC,


Kolenbrander PE: Mutualism versus independence:
strategies of mixed-species oral biofilms in vitro
using saliva as the sole nutrient source. Infect
Immun 2001;69:57945804.
100 Palmer RJ Jr, Gordon SM, Cisar JO, Kolenbrander
PE: Coaggregation-mediated interactions of streptococci and actinomyces detected in initial human
dental plaque. J Bacteriol 2003;185:34003409.
101 Nyvad B, Kilian M: Comparison of the initial streptococcal microflora on dental enamel in cariesactive and in caries-inactive individuals. Caries Res
1990;24:267272.
102 Nyvad B, Kilian M: Microbiology of the early colonization of human enamel and root surfaces in vivo.
Scand J Dent Res 1987;95:369380.

103 Kolenbrander PE, Egland PG, Diaz PI, Palmer RJ Jr:


Genome-genome interactions: bacterial communities in initial dental plaque. Trends Microbiol 2005;
13:1115.
104 Wecke J, Kersten T, Madela K, Moter A, Gobel UB,
Friedmann A, Bernimoulin J: A novel technique for
monitoring the development of bacterial biofilms in
human periodontal pockets. FEMS Microbiol Lett
2000;191:95101.
105 Moore WE, Moore LV: The bacteria of periodontal
diseases. Periodontol 2000 1994;5:6677.
106 Kroes I, Lepp PW, Relman DA: Bacterial diversity
within the human subgingival crevice. Proc Natl
Acad Sci USA 1999;96:1454714552.

Michael G. Surette
Department of Microbiology and Infectious Diseases
University of Calgary, Faculty of Medicine
3330 Hospital Drive, N.W., Calgary T2N 4N1 (Canada)
Tel. +1 403 220 2744, Fax +1 403 270 2772, E-Mail surette@ucalgary.ca

Chemical Interactions in Microbial Communities

17

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 1832

Autoinducer-2-Based Chemical
Communication in Bacteria:
Complexities of Interspecies Signaling
Michael J. Federle
Department of Microbiology and Immunology, College of Medicine, and Center for Pharmaceutical
Biotechnology, College of Pharmacy, University of Illinois at Chicago, Chicago, Ill., USA

Abstract
Cell-cell communication in bacteria, called quorum sensing, relies on production, release, and detection of signaling molecules, termed autoinducers. Communication enables populations of cells to
synchronize gene expression and therefore behave as a group in a manner akin to cells in multicellular organisms. Most quorum-sensing systems allow communication within an individual species of
bacteria. However, one autoinducer, called AI-2, is produced and recognized by many different bacterial species, indicating that some bacteria communicate across species boundaries. Current studies are aimed at discovering the role that AI-2 plays in gene regulation. Differential gene expression
in response to AI-2 may cause bacterial behavioral changes, such as biofilm formation or transition
to a pathogenic state. Interestingly, multiple mechanisms to detect AI-2 exist. These differences likely
reflect variations in the role that AI-2 plays for different bacteria. Additionally, structural analyses of
the AI-2 receptor in V. harveyi have provided insight into bacterial trans-membrane signal transduction. A further understanding of bacterial quorum-sensing processes may facilitate development of
new technologies aimed at interfering with bacterial communication and virulence.
Copyright 2009 S. Karger AG, Basel

Bacteria, long thought to exist as asocial entities, are increasingly acknowledged to


use communication mechanisms to coordinate behaviors within a population. Cellto-cell communication in bacteria, also known as quorum sensing, relies on the ability
of bacteria to produce, secrete, and detect small molecules, termed autoinducers, in
their surrounding environment. Scores of examples have been characterized in which
a particular bacterial species produces a specific autoinducer molecule that is recognized only by that species. In addition, a newly discovered signaling molecule, called
AI-2, is produced by many species of bacteria and is capable of eliciting responses in
different species of bacteria.

Quorum sensing allows control of behaviors on a community-wide level. Bacterial


cells detect the buildup of secreted autoinducers and respond with programmed changes
in gene expression. Bacterial activities regulated by quorum sensing include biofilm
development, virulence factor regulation, bioluminescence induction, antibiotic production, sporulation, and competence initiation. The understanding that bacteria produce
and respond to extracellular signals makes it possible to develop technologies aimed at
obstructing bacterial communication systems. By interfering with these systems, it may
be possible to alter the course of a bacteriums ability to harm humans, animals, and industrial processes. The discovery that a single type of signaling molecule, AI-2, contributes to
the control of behaviors in diverse types of bacteria raises the possibility that a single drug
or class of inhibitors may be able to target communication in many bacterial species.

Intra- and Interspecies Quorum Sensing

The ability of bacteria to produce and release signals to coordinate gene expression in
a population was discovered in experimentally tractable bacteria whose quorum-sensing-dependent behaviors were easily quantifiable. The first bacterial cell-cell communication system studied at the molecular level was that of Vibrio fischeri that produces
density-dependent bioluminescence [1]. The V. fischeri quorum-sensing system has
become the paradigm for Gram-negative bacterial communication networks (fig. 1a).
Typically, V. fischeri and other Gram-negative quorum-sensing bacteria have a LuxI
enzyme that produces an acylated homoserine lactone (AHL) autoinducer by reacting
S-adenosyl methionine (SAM) with an acylated acyl-carrier protein [2]. Each species
uses a unique LuxI enzyme to generate an AHL autoinducer with a particular acyl
side-chain making these signals species-specific. The AHL molecules diffuse across
the membrane into the surrounding environment. Detection of AHL autoinducers
is accomplished by LuxR-type proteins which have the dual ability of binding AHLs
and DNA. At high cell densities, i.e., at high autoinducer concentrations, LuxR binds
to a cognate AHL and undergoes a conformational change allowing the LuxR-AHL
complex to bind DNA and control transcription of target genes. In V. fischeri, one
of the targets for LuxR regulation is the set of genes encoding luciferase, and thus,
at high autoinducer concentrations, LuxR promotes the expression of bioluminescence. Many other Gram-negative bacterial species use analogous LuxI/LuxR-type
circuits, but their target genes are variable. For example, Pseudomonas aeruginosa
controls biofilm development and secretion of enzymes contributing to its pathogenesis, Agrobacterium tumefaciens and Rhizobium leguminosarum regulate transfer of
plasmid DNA between cells, and Serratia marcescens controls sliding motility. Each of
these organisms uses a unique LuxI/R pathway or set of LuxI/R pathways, each utilizing different AHLs to regulate the above and other behaviors.
In Vibrio harveyi, a bioluminescent bacterial species related to V. fischeri, the
quorum-sensing circuitry is more complex (fig. 1b). V. harveyi relies on multiple

AI-2 Quorum Sensing

19

Low cell density

High cell density


HSL

Cytoplasmic
membrane

Cytoplasmic
membrane

LuxR

LuxI

LuxR

LuxI

a
AI-1

AI-2
LuxQ

LuxQ
LuxN

LuxN

LuxP

P H1 H1 P

P H1 H1 P

LuxP

H1 H1

H1 H1

Pi
P D1 D1 P

54

Hfq

D1 D1 P

P D1 D1 P

H2

LuxU
D2 P

H2 P

LuxO

Pi
P D1 D1

H2

D2

LuxU
D2

H2 P

LuxO

luciferase

LuxR

sRNA

D2 P

LuxR

LsrB
LsrB
LsrCD

LsrCD
LsrA

LsrA

LsrK
LuxS

LsrK

LuxS

P
P

LsrR
IsrA IsrC IsrD IsrB

LsrR IsrA IsrC IsrD IsrB


IsrK IsrR

IsrK IsrR

20

Federle

autoinducers to regulate light production. The predominant autoinducer contributing


to luciferase production is an AHL-denoted HAI-1 (V. harveyi autoinducer-1). HAI-1 is
N-(3- hydroxybutanoyl)homoserine lactone and it is synthesized by LuxM. The HAI-1
receptor, LuxN, is a membrane-spanning histidine sensor kinase protein belonging to
the well-conserved family of bacterial two-component signal transduction proteins.
Two-component histidine sensor kinases autophosphorylate by transferring phosphate from ATP to a conserved histidine residue located on the kinase. A phosphorylation cascade ensues, where phosphate passes from histidine (H1) to aspartic acid (D1),
and phosphate can continue to a second histidine (H2) and second aspartic acid (D2),
located on proteins connected to, or separate from, the kinase. Phosphorylation of the
terminal aspartic acid generally causes activation of a DNA-binding response regulator
protein. In the absence of HAI-1, LuxN autophosphorylates on the conserved H1 residue, which is then passed on to D1 located within the C-terminal domain of LuxN [3].
Phosphate is sequentially passed to H2 on LuxU, a phosphotransfer protein, and then
on to D2 located on LuxO, the DNA-binding response regulator [4]. Phosphorylation of
LuxO promotes the expression of genes encoding five small regulatory RNAs (sRNAs)
[5]. These five sRNAs, together with the sRNA chaperone Hfq, bind to and block translation of the luxR mRNA. (LuxR of V. harveyi is not a member of the canonical V. fischeri LuxI/LuxR family of proteins; it does not bind HSLs.) The V. harveyis LuxR protein
directly activates transcription of luciferase and regulates nearly 100 other genes [6].
When V. harveyi reaches high cell density, HAI-1 concentrations are also high. When
bound to autoinducer, the enzymatic activity of LuxN switches from kinase to phosphatase, and LuxN drains phosphate from LuxU and LuxO. Dephosphorylated LuxO
is unable to activate transcription of sRNA genes, and consequently, LuxR expression is
derepressed, and luciferase is expressed, and light is produced.
Interestingly, in the absence of the HAI-1 system, V. harveyi continues to control
luciferase production as a function of cell population density, albeit at 1,000 times
lower light production per cell. V. harveyi mutants were screened for those unable
to regulate light production in a background lacking the HAI-1 system, and a new

Fig. 1. Quorum-sensing signaling networks. Low cell density (left sides of panels), High cell density
(right sides of panels). a A LuxI/LuxR quorum-sensing circuit. The LuxI enzyme produces an AHL molecule that diffuses into the surroundings. At high cell densities, the LuxR protein binds the AHL and
subsequently binds to DNA promoting the transcription of genes X and Y. b V. harveyi AI-1- and AI-2dependent circuits. At low cell density, LuxN and LuxQ exist in kinase mode autophosphorylating at
histidine (H1) and aspartic acid (D1) residues. Phosphate is passed to LuxU at the histidine (H2) site,
and then to LuxO at an aspartic acid (D2). LuxO-P, with sigma factor 54, activates transcription of the
sRNA genes. The sRNAs, with the RNA chaperone Hfq repress luxR expression. At high cell densities,
AI-1 binds LuxN and AI-2 binds LuxP-LuxQ. Binding of autoinducers causes LuxN and LuxQ to switch
to phosphatase mode. Phosphate is drained from LuxU and LuxO, terminating expression of the sRNA
genes. LuxR is derepressed, and luciferase is expressed. c The E. coli and S. typhimurium Lsr system. At
low cell densities, LsrR binds to DNA and represses expression of the lsr operon. At high cell densities,
AI-2 is imported via the Lsr transporter and is phosphorylated by LsrK. AI-2-phosphate binds to LsrR
causing it to release DNA, thus derepressing lsr transcription [adapted from 8, 28].

AI-2 Quorum Sensing

21

autoinducer detection system was discovered [7]. Two genes identified in the screen
encode components of the autoinducer receptor. The first gene, luxP, encodes a
periplasmic binding protein LuxP, similar to the Escherichia coli ribose-binding
protein. The second gene, named luxQ, encodes a two-component histidine kinase
similar to LuxN. A third gene, luxS, encodes the autoinducer synthase, LuxS, which
produces a signal which was named, for lack of a specific chemical nature, AI-2.
V. harveyi integrates AI-2 information directly into the quorum-sensing pathway
described above. LuxP and LuxQ work together as the AI-2 receptor complex (see
structural details below). At low cell densities, when AI-2 concentrations are low,
LuxQ acts as a kinase in a manner analogous to LuxN. Phosphate follows the typical H1 to D1 transfer on LuxQ itself, and is then passed on to H2 of LuxU and D2 of
LuxO. When both HAI-1 and AI-2 concentrations are low, the kinase activities of both
LuxN and LuxQ maximize LuxO-phosphate levels in the cell, generating high levels
of the sRNAs. Consequently, little LuxR protein is made, and the cells do not produce
light. The phosphorylation pathway reverses under high autoinducer concentrations.
AI-2 binds to LuxP, switching LuxQ from kinase to phosphatase, lowering levels of
LuxU-P and LuxO-P. Ultimately, LuxR is derepressed and light is produced.
The luxS gene exists in over half of all sequenced bacterial genomes. Importantly,
species containing luxS genes generate and secrete AI-2 activity into their surroundings. One question that has emerged is whether AI-2 is an important communication
signal in other bacteria, and if so, what do these bacteria control with AI-2 information? For E. coli and Salmonella typhimurium, AI-2 plays a regulatory role. In these bacteria, a set of genes called lsr (luxS regulated) are induced that encode components of
the machinery used for the import and processing of AI-2 [8, 9]. This transporter is so
effective that levels of AI-2 are diminished to near background levels outside the cells.
The AI-2 receptor in the Lsr system, LsrB, is a periplasmic binding protein similar to
V. harveyis AI-2 receptor LuxP. While the overall fold of this family of proteins is highly
conserved (the prototype being the ribose binding protein, RbsB), primary sequence
similarity between LsrB, LuxP and RbsB is low. RbsBs similarity to LuxP and LsrB is 47
and 41%, respectively. However, LuxP is only 11% similar to LsrB. Low sequence similarity reflects the fact that each protein interacts with different types of downstream
components. Instead of interacting with a histidine sensor kinase protein, as is the
case for LuxP, the LsrB protein delivers AI-2 to an ABC transporter that imports AI-2
into the cytoplasm (fig. 1c). Transport of AI-2 is coupled to its phosphorylation and
sequestration, which is dependent on LsrK kinase activity [9]. Phosphorylated AI-2
subsequently binds and deactivates the transcriptional repressor LsrR. Transcription
increases of the genes encoding the Lsr transporter and modification enzymes used to
import, modify, and degrade AI-2. This positive feedback is required for rapid induction of the Lsr system and removal of AI-2 from the environment.
To date, several studies have identified additional genes differentially regulated in
response to AI-2 or LsrR/LsrK in both E. coli and S. typhimurium strains, including
pathogenic species. However, the ability of the Lsr transporter to rapidly reduce AI-2

22

Federle

levels in the surrounding environment has led to the hypothesis that E. coli and S.
typhimurium not only rely on AI-2 to identify their quorum, but also to diminish
AI-2 levels to confuse other species occupying the same environment. As proof of
this principle, E. coli and V. harveyi were grown together and the effects of this coculture on V. harveyi luciferase production was monitored. In the mixed species set
up, as V. harveyi population numbers increased, luciferase production was induced.
However, as AI-2 concentrations climbed, E. coli, in turn, induced its Lsr transporter,
and AI-2 levels in the co-culture dropped dramatically. The rate of light production
from V. harveyi declined, and the cells gene expression pattern was maintained at
the level appropriate for cell densities 10- to 100-fold below their actual population
number [9]. Few bacterial species exist in pure culture in nature, and it is likely that
communication networks are subject to quorum-sensing eavesdropping and interference, as the Lsr studies imply.

AI-2 Identification

The LuxS protein does not to share sequence homology to LuxI-type autoinducer
syntheses, and common extraction techniques used to purify AHLs from bacterial
cultures proved unsuccessful for isolating AI-2 activity, suggesting that it was a novel
chemical. Clues to LuxSs function and to AI-2s chemical nature came from the notion
that LuxS may be involved in a pathway with other genes lying within close proximity on the chromosome. luxS was often found near genes involved in methionine
metabolism and the reactive methyl cycle. In Borellia burgdorferi, for example, luxS is
preceded in an operon by the genes metK and pfs. MetK catalyzes the production of
SAM by combining adenosine with methionine (fig. 2). SAM is an essential coenzyme
whose function is to provide methyl groups to numerous substrates, for example, in
the biosynthesis of DNA, RNA, fatty acids, and proteins. SAM-dependent transmethylation reactions result in the release of S-adenosyl homocysteine (SAH) as a byproduct. SAH inhibits transmethylation reactions and is therefore toxic. The second gene
found near luxS in B. burgdorferi was pfs. The Pfs enzyme detoxifies SAH by removing
adenine and generating S-ribosyl homocysteine (SRH). Cells break down SRH into
homocysteine and 4,5-dihydroxy-2,3-pentanedione (DPD); however, no enzyme had
ever been identified to catalyze this reaction. Indeed, as predicted, LuxS catalyzes this
final reaction. Thus, bacteria that use Pfs and LuxS to recycle SAM pools generate
DPD as an additional byproduct, and DPD is responsible for AI-2 activity [10].

DPD and AI-2 Activity

Use of DPD as an intercellular signaling molecule is interesting from the standpoint


that this molecule is inherently unstable. DPD spontaneously cyclizes to form two

AI-2 Quorum Sensing

23

PPi, Pi

H2N

ATP

O
MetK

NH2
O

OH

THF

OH

Pfs

OH

N
N

OH
S-adenosyl
homocysteine
(SAH)

H2N
OH

OH

H2N

O
4,5-dihydroxy2,3-pentane dione
(DPD)

OH
Homocysteine

OH O

H2N

SH

LuxS

Methylated
product

OH

S-adenosyl
methionine
(SAM)

NH2

OH O

Methyltransferase

MetH

N
N

H2N

5-methyltetrahydrofolate

CH3

S
methionine

OH

S+

NH

O
HO

OH

OH
S-ribosyl
homocysteine
(SRH)

NH2

N
Adenine

Fig. 2. Reactive methyl cycle and production of DPD. SAM-dependent methyltransferases convert
SAM to SAH, accumulation of which confers product-feedback inhibition on methyltransferase reactions. SAH is detoxified to SRH by Pfs. SRH is converted to homocysteine and DPD by LuxS.
Homocysteine can be recycled to SAM via MetH, which generates methionine, and MetK.

enantiomers, 2S,4S-DHMF and 2R,4S-DHMF (fig. 3) [11]. Hydration of the ketone


in aqueous solutions leads to formation of S- and R-THMF. The multiple derivatives
of DPD raised the question: which of these isomers is recognized as AI-2 by bacteria? Because conversion between these molecules is spontaneous and fast, isolation
or synthesis of each compound for biological activity testing was impossible. Thus,
determining which forms of DPD are active came from biochemical and structural
studies of various DPD receptors. Crystals of V. harveyis LuxP protein containing the
AI-2 ligand within the binding pocket showed that a borated adduct of the S-THMF
molecule is the V. harveyi signal [12]. Consistent with this, Vibrio species are found in
marine environments where borate is plentiful. In contrast, levels of borate are low in
terrestrial environments, yet bacteria such as E. coli and S. typhimurium also produce
and respond to AI-2. These findings support the idea that borated DPD may not be
the only active form of the molecule. Indeed, crystal structures of LsrB, the AI-2 binding protein from S. typhimurium contained the non-borated R-THMF ligand [11].
Knowing the two AI-2 structures provided a mechanism by which interspecies
communication occurs. The LuxS enzyme, regardless of bacterial species origin, is

24

Federle

OH
CH3
O

HO
OH

OH

CH3
OH

HO

HO

(2R,4S)-2,4-dihydroxy-2
methylhydrofuran-3-one
(R-DHMF)

OH

OH

CH3

2H2O
+B(OH)4

HO

OH
B
O
O
CH3

HO

HO

(2S,4S)-2,4-dihydroxy-2
methylhydrofuran-3-one
(S-DHMF)

4,5-dihydroxy2,3-pentane dione
(DPD)

H2O

HO
(2S,4S)-2-methyl-2,3,3,4tetrahydroxytetrahydrofuran
(2S,4S)-2-methyl-2,3,3,4(S-THMF)
tetrahydroxytetrahydrofuran-borate
(S-THMF-borate)
H2O

HO

CH3

OH
HO

OH
O

(2R,4S)-2-methyl-2,3,3,4tetrahydroxytetrahydrofuran
(R-THMF)

Fig. 3. Interconverting forms of DPD. Cyclization of DPD generates two major stereoisomers S-DHMF
and R-DHMF. Hydration in aqueous solution forms S-THMF and R-THMF. S-THMF is able to complex
with borate, forming S-THMF-borate [adapted from 11].

responsible for the production of the linear DPD molecule, which through spontaneous rearrangements, produces a pool of interconverting compounds. A molecule
recognized by one species can rearrange to become the signal recognized by another
species. It remains to be seen whether forms of DPD other than the two now known
are recognized as AI-2 signals, or whether other types of adducts can be made from
DPD to generate new, unanticipated AI-2 molecules.
Environmental conditions affect the equilibrium of AI-2 molecules, and may contribute to a bacteriums ability to monitor its surrounding environment. For instance,
V. cholerae, a bacterium that contains luxS and responds to AI-2, has a life cycle that
alternates between marine environments and the human intestine. As discussed
above, the predominant form of AI-2 in the ocean is S-THMF-borate. However, in the
human gastrointestinal tract, where only trace amounts of borate are available, AI-2
likely exists only in non-borated forms. It is possible that V. cholerae distinguishes
these niches by measuring the availability of borated AI-2. Therefore, in addition to
population density, information encoded within AI-2 can also include the status of
the surrounding environment. It is also possible that bacteria detect multiple forms
of AI-2 using different protein receptors. If so, this would allow direct monitoring of
different AI-2 forms and the proportions of each.

Discriminating between AI-2 Signaling Effects and LuxS Metabolic Roles

A large effort has been underway to test the roles that LuxS and AI-2 play in controlling gene regulation. luxS null mutations have been generated in different bacterial

AI-2 Quorum Sensing

25

species, and subsequent studies have ranged in scope from measuring changes in
expression of a specific gene of interest in some species, to whole genome transcription profiles in others. For instance, specific virulence factors have been monitored in
luxS mutants of Clostridium perfringens, Shigella flexneri, Actinobacillus actinomycetemcomitans and Streptococcus pyogenes [1316]. Likewise, whole genome expression
profiles have been compared between wild-type strains and luxS mutants for many species, including E. coli W113 and E. coli O157:H7, Neisseria meningitidis, Streptococcus
pneumoniae, and Streptococcus mutans [1721]. Typically, large numbers of genes are
differentially regulated in luxS mutants (ranging from <1 to 10% of the genome) with
many of these genes being unresponsive to exogenously added AI-2. The inability
to complement luxS mutants with artificially added autoinducer raises the question
of whether luxS mutations affect gene regulation in an AI-2-dependent manner, or
whether the ramification of luxS mutations stem from alterations in SAM, SAH, or
SRH levels. Several possibilities in addition to disruption of the reactive methyl cycle
may also explain the lack of ability to complement luxS phenotypes, including the
appropriate timing of signal presentation, the percent of the population responding
to the signal, the concentration of the signal, if the signal requires chemical modification, or if feedback regulation affects the ability of cells to respond to AI-2.
In some quorum-sensing systems, a temporary window of time exists during
which cells are able to respond to extracellular signals. Studies of competence development, a quorum-sensing-controlled process in S. pneumoniae, have shown that
only 1530 min exist when cells develop and maintain maximum competence [22].
After this interval, the response to the competence-stimulating autoinducer rapidly
decreases. Shut-down of competence is now partially understood to be controlled by
degradation of transcription factors responsible for induction of the competent state
[23]. This stands as a clear example that programs triggered by autoinducers can be
heavily regulated processes, and precise temporal conditions must be met to generate
full responses. In the case of AI-2, cells experiencing only the correct concentrations
of autoinducer may also be critical for proper responses. Recent studies of biofilm
formation within mixed bacterial cultures of Streptococcus oralis and Actinomyces
naeslundii found that optimal AI-2 concentrations exist for biofilm formation, and
above and below this concentration, biofilm development is decreased [24].
Effects of added autoinducers on a bacterial culture also may be difficult to observe
when only a subpopulation of cells is responsive to the extracellular signal. In Bacillus
subtilis, competence develops in only 10% of the population in response to the competence-inducing autoinducer ComX [25]. Without an easily tractable phenotype,
like uptake of DNA, effects on gene expression via microarray or proteomic analysis,
or other methods that measure the mean change of expression of target genes in the
population, may be nearly impossible to observe.
When using AI-2 to artificially stimulate cells, another difficulty to be considered
is that some species of bacteria in which luxS has been deleted may be locked in a state
that is unresponsive to exogenously added autoinducer. In E. coli and S. typhimurium,

26

Federle

as mentioned, AI-2 is imported into the cell by the Lsr transporter. Transport occurs
quickly due to upregulation of the Lsr apparatus, and AI-2 is completely removed
from the surrounding environment in <2 h. In luxS mutants, exogenously supplied
AI-2 requires approximately 60 min longer than in the wild type for proper induction
of the Lsr system [8]. The response is possible only because a secondary, promiscuous
transporter is capable of importing some AI-2, which can then kick-start the system.
For species with other types of positive feedback regulation, it may be the case that
deletion of luxS abolishes the basal level of AI-2 required to ensure that detection
systems are primed for induction. Without this, cells are unable to respond to exogenously added AI-2.
Finally, phenotypes stemming from luxS mutations may be due to a combination
of effects on the active methyl cycle as well as a loss of the signaling molecule AI-2.
Genomic, proteomic, and metabalomic studies may provide initial clues for discriminating between gene and protein level changes in response to AI-2 concentrations versus metabolic disturbances. A recent study in S. mutans that measured changes in gene
expression using microarray analysis compared a wild-type strain to a luxS mutant
and to a luxS mutant supplemented with chemically synthesized DPD. Over 500 genes
(30% of genome) showed changes in gene expression in the luxS mutant, but only 59
of these genes were complemented by exogenously supplied DPD [21]. These results
indicate that the large majority of genes whose expression changed as a result of the
luxS mutation were not responsive to DPD under conditions tested. But the finding
that a substantial number of genes are controlled by AI-2 provides footing for future
studies to optimize conditions that may give insight to how S. mutans detects AI-2.
Further complications due to signal redundancy also complicate these analyses. In
fact, the species in which AI-2 was discovered, V. harveyi, does not have an obvious
luxS phenotype. A V. harveyi luxS mutant decreases light production about 100-fold
when grown to high cell density as compared to wild type [26]. However, a 100-fold
loss in light output accounts for less than 1% of the overall change in light production
observed when AI-1 and AI-2 autoinducers are removed simultaneously.

A Structural Study of the AI-2 Receptor of Vibrio harveyi

One long-term goal of quorum-sensing studies is to identify compounds that can


interfere with bacterial communication. To fully understand how small molecule
antagonists interfere with cell-cell communication systems, the endogenous signaling molecules and the mechanisms by which information is relayed into the cell must
be defined. The identification of the AI-2 signal, along with the identification of the
proteins used to convert AI-2 information into target gene expression, makes such
studies feasible. Of particular interest is to understand how AI-2 ligand binding in
the periplasm leads to changes in enzymatic activity of the LuxQ sensor kinase in the
cytoplasm.

AI-2 Quorum Sensing

27

Genetic studies had indicated that LuxP controls LuxQs ability to switch from the
kinase state (at low AI-2 concentrations) to the phosphatase state (at high AI-2 concentrations), because deletions of luxP cause LuxQ to remain locked in the kinase
state [27]. To determine how conformational changes in LuxP control LuxQ enzymatic activity, combined structural and genetic studies were performed. V. harveyi
LuxP and the periplasmic domain of LuxQ (LuxQp) were expressed in E. coli in the
presence and absence of AI-2, and the protein complexes were purified and crystallized. LuxP and LuxQp have two sites of interaction, and each is required for proper
switching between kinase and phosphatase activities.
In protein complexes lacking AI-2, LuxP is in an open conformation and LuxP
interacts with LuxQp in a 1:1 stoichiometry forming a complex (the LuxPQp monomeric subunit) [27]. Contacts between LuxP and LuxQp occur at a location distal to the
membrane, and this interaction is necessary for LuxQ kinase activity because deleting this region causes LuxQ to favor the phosphatase state in an AI-2-independent
manner. The structure of the holo-LuxP-LuxQp protein complex displayed two major
differences from the apo-complex structure [28]. First, a second site of interaction
between LuxP and LuxQ was evident. While the first site of interaction is identical
to that observed in the apo-complex, the second site of interaction occurs because
AI-2 binding induces a major conformational change in LuxP, and allows new contacts to form between LuxP and a second LuxQp subunit (LuxQp). The second major
structural difference in the holo-complex is the generation of an asymmetric dimer,
consisting of two LuxPQp monomers (designated LuxPQp-LuxPQp). When LuxP
simultaneously contacts LuxQp and LuxQp, the complex is held together in an orientation that is asymmetric along the axis that is normal to the cytoplasmic membrane.
Interestingly, although a large conformational change occurs in each LuxP subunit
upon binding AI-2, virtually no conformational change is observed in either LuxQp
subunit. It appears that signal propagation does not occur through intramolecular
conformational changes that are passed across the membrane into the cytoplasmic
domains of the protein. Rather, in the case of LuxQ, the relative orientation in which
the periplasmic LuxQ subunits are arranged around the axis of symmetry determines
the enzymatic state of the receptor.
The current model suggests that when AI-2 concentrations are low, LuxP is in an
open conformation and only one contact is made with its paired LuxQ subunit. The
orientation of LuxQ-LuxQ dimers is determined by membrane-spanning and cytoplasmic domains of the protein and the dimer is situated in a symmetric orientation
(fig. 4, left). Upon binding AI-2, LuxP undergoes large movements to close around the
ligand. This movement reorients LuxP, making it possible to contact the LuxQp subunit. This contact drives rotation between the two LuxQ subunits into an asymmetric
position (fig. 4, right). This asymmetric orientation must signal to the cytoplasmic
domains to switch kinase activity off. The phosphatase activity of LuxQ is located in
the C-terminal aspartic acid (D1)-containing domain, and is not dependent on autoinducer concentration. Therefore, the activity of LuxQs kinase domain determines

28

Federle

90

90

P
P+H

D
P

P
D

Pi

Pi

Fig. 4. Model for AI-2 dependent LuxPQ receptor activity. The top panels display the LuxPQp receptor complex from a top-down view, and the lower panels are a 90 rotation, showing the complex
from the side. In the absence of AI-2 (left) LuxQ-LuxQ dimers exist in a symmetric orientation both
in the periplasm and the cytoplasm, and therefore are in kinase mode. AI-2 binding to LuxP (right)
causes a conformational change in LuxP, and induces interactions with LuxQ. Simultaneous contacts of LuxP with LuxQ and LuxQ rotates the complex into an asymmetric orientation. Rotation of
the periplasmic domains is conferred to the cytoplasmic domains, switching LuxQ from kinase to
phosphatase mode [figure 4 reproduced with permission, 28].

AI-2 Quorum Sensing

29

whether phosphate flows toward LuxU and LuxO or away from them. When the kinase
is active, it overrides the phosphatase activity; when the kinase is inactivated, the phosphatase activity prevails, and phosphate is drained away from LuxU and LuxO.
The AI-2-LuxP-LuxQp receptor complex is one of only a few bacterial signal
transduction receptor complexes with structures that have been determined with
and without bound ligand. Another such receptor system in which structural data
are available is the chemotaxis receptor. Chemotaxis systems are used by bacteria
to detect and subsequently move towards attractant and away from repellant molecules. Chemotaxis receptors measure changes in concentration of the attractants
or repellants and relay this information to a signaling network which controls the
bacteriums swimming direction. These receptors cluster to form arrays of protein
complexes in the membrane. Clustering of receptors is proposed to provide a mechanism to detect small fluctuations in attractant/repellant concentrations and then
respond with large changes in swimming behavior. Unlike the chemotaxis receptors, the LuxP-LuxQ quorum-sensing receptor does not appear to form higher-order
clusters [28]. If clustering of receptors provides added sensitivity to small changes in
chemotaxis attractants, perhaps arranging quorum-sensing receptors as independent
detectors protects the cell from responding to minor fluctuations in autoinducers.
This arrangement likely allows cells to measure the slow build-up of autoinducers
and at a critical concentration, commit to new behaviors that require a change in
global transcription.

Future Questions and Goals

As technologies continue to improve our ability to monitor gene, protein, and metabolite levels in living cells, and as we find new ways to identify small molecules produced by bacteria, discovery of new quorum-sensing systems will likely accelerate. To
fulfill the promise that an understanding of quorum sensing will allow us to manipulate bacterial behaviors by interfering with communication systems, autoinducer analogs and signaling inhibitors will need to be synthesized, or natural ones discovered
and tested in vivo. Large chemical libraries are becoming available, and experimental screening for quorum-sensing antagonists is a promising route. Likewise, rational design of compounds based on known autoinducers, and screens for compounds
occurring naturally in the environment, could be alternative resources of inhibitors.
If quorum-sensing systems are susceptible to chemical inhibition, development of
novel antimicrobial agents could ensue. With respect to AI-2-based quorum-sensing
systems, LuxS appears to be a promising target for directed inhibitors because the
LuxS crystal structure has been solved, all LuxS proteins synthesize the same moiety,
DPD, and its enzymatic mechanism is defined. By interfering with production of the
AI-2 signaling molecule, communication could be terminated in a diverse array of
species.

30

Federle

References
1 Nealson KH, Platt T, Hastings JW: Cellular control
of the synthesis and activity of the bacterial luminescent system. J Bacteriol 1970;104:313322.
2 Fuqua C, Parsek MR, Greenberg EP: Regulation of
gene expression by cell-to-cell communication:
acyl-homoserine lactone quorum sensing. Annu
Rev Genet 2001;35:439468.
3 Freeman JA, Lilley BN, Bassler BL: A genetic analysis of the functions of LuxN: a two-component
hybrid sensor kinase that regulates quorum sensing in Vibrio harveyi. Mol Microbiol 2000;35:139
149.
4 Freeman JA, Bassler BL: A genetic analysis of the
function of LuxO, a two-component response regulator involved in quorum sensing in Vibrio harveyi.
Mol Microbiol 1999;31:665677.
5 Lenz DH, Mok KC, Lilley BN, Kulkarni RV,
Wingreen NS, Bassler BL: The small RNA chaperone Hfq and multiple small RNAs control quorum
sensing in Vibrio harveyi and Vibrio cholerae. Cell
2004;118:6982.
6 Waters CM, Bassler BL: The Vibrio harveyi quorumsensing system uses shared regulatory components
to discriminate between multiple autoinducers.
Genes Dev 2006;20:27542767.
7 Bassler BL, Wright M, Silverman MR: Multiple signaling systems controlling expression of luminescence in Vibrio harveyi: sequence and function of
genes encoding a second sensory pathway. Mol
Microbiol 1994;13:273286.
8 Taga ME, Semmelhack JL, Bassler BL: The LuxSdependent autoinducer AI-2 controls the expression
of an ABC transporter that functions in AI-2 uptake
in Salmonella typhimurium. Mol Microbiol 2001;
42:777793.
9 Xavier KB, Bassler BL: Interference with AI-2mediated bacterial cell-cell communication. Nature
2005;437:750753.
10 Schauder S, Shokat K, Surette MG, Bassler BL: The
LuxS family of bacterial autoinducers: biosynthesis
of a novel quorum-sensing signal molecule. Mol
Microbiol 2001;41:463476.
11 Miller ST, Xavier KB, Campagna SR, Taga ME,
Semmelhack MF, Bassler BL, Hughson FM:
Salmonella typhimurium recognizes a chemically
distinct form of the bacterial quorum-sensing signal AI-2. Mol Cell 2004;15:677687.
12 Chen X, Schauder S, Potier N, Van Dorsselaer A,
Pelczer I, Bassler BL, Hughson FM: Structural identification of a bacterial quorum-sensing signal containing boron. Nature 2002;415:545549.

AI-2 Quorum Sensing

13 Day WA Jr, Maurelli AT: Shigella flexneri LuxS quorum-sensing system modulates virB expression but
is not essential for virulence. Infect Immun 2001;
69:1523.
14 Fong KP, Chung WO, Lamont RJ, Demuth DR:
Intra- and interspecies regulation of gene expression by Actinobacillus actinomycetemcomitans LuxS.
Infect Immun 2001;69:76257634.
15 Lyon WR, Madden JC, Levin JC, Stein JL, Caparon
MG: Mutation of luxS affects growth and virulence
factor expression in Streptococcus pyogenes. Mol
Microbiol 2001;42:145157.
16 Ohtani K, Hayashi H, Shimizu T: The luxS gene is
involved in cell-cell signalling for toxin production
in Clostridium perfringens. Mol Microbiol 2002;44:
171179.
17 DeLisa MP, Wu CF, Wang L, Valdes JJ, Bentley WE:
DNA microarray-based identification of genes
controlled by autoinducer 2-stimulated quorum
sensing in Escherichia coli. J Bacteriol 2001;183:
52395247.
18 Dove JE, Yasukawa K, Tinsley CR, Nassif X:
Production of the signaling molecule, autoinducer-2,
by Neisseria meningitidis: lack of evidence for a concerted transcriptional response. Microbiology 2003;
149:18591869.
19 Joyce EA, Kawale A, Censini S, Kim CC, Covacci A,
Falkow S: LuxS is required for persistent pneumococcal carriage and expression of virulence and biosynthesis genes. Infect Immun 2004;72:29642975.
20 Sperandio V, Torres AG, Giron JA, Kaper JB:
Quorum sensing is a global regulatory mechanism
in enterohemorrhagic Escherichia coli O157:H7. J
Bacteriol 2001;183:51875197.
21 Sztajer H, Lemme A, Vilchez R, Schulz S, Geffers R,
Yip CY, Levesque CM, Cvitkovitch DG, WagnerDobler I: Autoinducer-2-regulated genes in Streptococcus mutans UA159 and global metabolic effect of
the luxS mutation. J Bacteriol 2008;190:401415.
22 Hotchkiss RD: Cyclical behavior in pneumococcal
growth and transformability occasioned by environmental changes. Proc Natl Acad Sci USA 1954;
40:4955.
23 Luo P, Morrison DA: Transient association of an
alternative sigma factor, ComX, with RNA polymerase during the period of competence for genetic
transformation in Streptococcus pneumoniae. J
Bacteriol 2003;185:349358.
24 Rickard AH, Palmer RJ Jr, Blehert DS, Campagna
SR, Semmelhack MF, Egland PG, Bassler BL,
Kolenbrander PE: Autoinducer-2: a concentrationdependent signal for mutualistic bacterial biofilm
growth. Mol Microbiol 2006;60:14461456.

31

25 Magnuson R, Solomon J, Grossman AD:


Biochemical and genetic characterization of a competence pheromone from B. subtilis. Cell 1994;77:
207216.
26 Bassler BL, Wright M, Showalter RE, Silverman MR:
Intercellular signaling in Vibrio harveyi: sequence
and function of genes regulating expression of luminescence. Mol Microbiol 1993;9:773786.

27 Neiditch MB, Federle MJ, Miller ST, Bassler BL,


Hughson FM: Regulation of LuxPQ receptor activity by the quorum-sensing signal autoinducer-2.
Mol Cell 2005;18:507518.
28 Neiditch MB, Federle MJ, Pompeani AJ, Kelly RC,
Swem DL, Jeffrey PD, Bassler BL, Hughson FM:
Ligand-induced asymmetry in histidine sensor
kinase complex regulates quorum sensing. Cell
2006;126:10951108.

Michael J. Federle
Department of Microbiology and Immunology, College of Medicine
University of Illinois at Chicago, 900 S. Ashland (M/C 870)
MBRB Rm 3152, Chicago, IL 60607 (USA)
Tel. +1 312 413 0213, Fax +1 312 413 9303, E-Mail mfederle@uic.edu

32

Federle

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 3364

The Molecular Basis of Excitation


and Adaptation during Chemotactic
Sensory Transduction in Bacteria
Christopher V. Raoa George W. Ordalb
Departments of aChemical and Biomolecular Engineering and bBiochemistry, University of Illinois at
Urbana-Champaign, Urbana, Ill., USA

Abstract
Chemotaxis is the process by which cells sense chemical gradients in their environment and then
move towards more favorable conditions. In the case of Escherichia coli, the paradigm organism for
chemotaxis, the pathway is now arguably the best characterized in all of biology. If one broadens
their perspective to include other species of bacteria, then our knowledge of chemotaxis is far less
developed. In particular, the chemotaxis pathways in unrelated species are quite different despite
the conservation of many core signaling proteins. Here, we summarize the current state of knowledge regarding the chemotaxis pathways in E. coli and Bacillus subtilis, with a specific focus on the
mechanisms for excitation and adaptation. The mechanisms vary widely, and the B. subtilis process,
similar to those found in Thermotoga maritima and many archaea, may represent a new paradigm
for bacterial chemotaxis. For instance, B. subtilis has three interacting means for restoring prestimulus behavior after stimulation, including one involving CheYp feedback. The one shared with E. coli,
the receptor methylation system, is vastly different, as is the mechanism for conveying signals across
Copyright 2009 S. Karger AG, Basel
the membrane.

Chemotaxis is the process by which cells sense chemical gradients in their environment and then move towards more favorable conditions. In peritrichiously flagellated
bacteria such as Escherichia coli and Bacillus subtilis, directed migration involves transitions between smooth runs and reorientating tumbles. By altering the frequency of
runs and tumbles, these bacteria are able to migrate toward sources of attractants
such as amino acids and away from repellents such as metal ions in a manner akin
to a biased random walk [1]. Due to their small size, however, bacteria are unable to
spatially sense chemical gradients, unlike eukaryotic cells [2]. Rather, bacteria sense
chemical gradients temporally [3]. In other words, chemotaxis in bacteria involves
memory [4]; bacteria detect gradients by comparing their currently sensed chemical
environment with that sensed in the recent past. If the conditions are improving, then

1.0

0.8
Probability of CCW rotation

Fig. 1. Example of the excitation and


adaptation response in B. subtilis. The
cells were exposed to 0.5 mM asparagine beginning at the time denoted with
the downward arrow and ending with
the time denoted with the upward
arrow. The plot shows the average rotational bias of 15 cells. Counterclockwise
rotation correlates with smooth runs.
The measurements were made by first
tethering the cells to a glass coverslip
with an anti-flagellin antibody and
then measuring the direction of rotation of the cell body. Note the transient
response to the addition and removal
of asparagine, a characteristic of the
adaptational response. This figure was
reprinted from Kirby et al. [139].

0.6

0.4

0.2

0
0

6
8
Time (m)

10

12

the bacterium will continue running along its current path. However, if conditions
are deteriorating because, for example, the bacterium is moving away from a food
source, then it will tumble and try to find a new path that yields a more favorable trajectory. This memory is due to sensory adaptation as bacteria respond only to changes
in attractant and repellent concentrations rather than their absolute values (fig. 1).
Beginning with the pioneering work of Julius Alder in the late 1960s, bacterial chemotaxis has been studied extensively as a model system for bacterial signal
transduction [5, 6]. In the case of E. coli, where most of the investigations have been
directed, the pathway is now arguably the best characterized in all of biology. In particular, all of the participating proteins have been identified and characterized biochemically. Current research now is primarily directed towards understanding how these
proteins function as ensembles. Despite this progress, a number of open questions still
remain. Of notable significance, in the case of E. coli, is the structure and physiological significance of protein complexes. In particular, the chemotaxis proteins are not
homogeneously distributed in the cytosol but rather concentrated at distinct locations
within the cell [7, 8]. Furthermore, these complexes are structured over length scales
far greater than the size of individual proteins, akin to solid state signaling arrays [9].
If one broadens their perspective to include organisms outside the -proteobacteria,
then our knowledge of chemotaxis is far less developed. Despite the conservation of
many core signaling proteins, the chemotaxis pathways in unrelated species are quite
different. While some of the variation can be explained by the presence or absence of
different signaling proteins, others are due to how the specific proteins function within
the context of the integrated pathway. This latter feature suggests that simple sequencebased homology arguments cannot be used to infer the role of a given chemotaxis

34

Rao Ordal

14

protein in a given organism. These differences also afford a new mode for modeling
evolution by providing an avenue for studying this process at the pathway level.
In this chapter, we summarize the current state of knowledge regarding the chemotaxis pathways in E. coli and B. subtilis. Our review is heavily influenced by our own
research in B. subtilis chemotaxis. As a consequence, we focus primarily on chemotaxis in B. subtilis and highlight some of the differences and similarities with E. coli
chemotaxis. Furthermore, our review does not intend to be comprehensive, particularly with regard to E. coli chemotaxis as a number of excellent reviews have recently
been written [1016]. Finally, we note that significant progress towards elucidating
the chemotaxis pathways in other species of bacteria and archaea has been made,
particularly in the cases of Halobacterium salinarum [1721], Helicobacter pylori [16,
2227], Rhodobacter sphaeroides [2834], and Sinorhizobium meliltoi [3538].

Excitation

Paradigm
The primary signal transduction module involves a stable, ternary signaling complex comprising transmembrane receptors, CheW adaptor proteins, and CheA histidine kinases, where both the receptors and kinases form homodimers [39, 40] (see
table 1 for list of the signaling proteins). The basic model is that attractants bind to
the receptors outside the plasma membrane and cause a conformational change that
alters the rate of CheA autophosphorylation. The phosphoryl group on CheA is then
transferred to a soluble response regulator, CheY, that binds to the flagella motors and
alters their rotational bias [41, 42]. This basic paradigm appears to be conserved in all
species of bacteria.
In B. subtilis, attractant binding enhances CheA kinase activity, leading rapidly to
an increase in the concentration of phosphorylated CheY (CheYp) [43]. CheYp then
binds to the cytoplasmic face of the flagella and causes the motors to rotate counter
clockwise, yielding smooth swimming [44, 45]. In E. coli, the response is completely
the opposite, though the outcome the same. Attractant binding in this organism
inhibits kinase activity [39]. However, CheYp in E. coli causes the motors to rotate
clockwise and the cells to tumble. Thus, the net effect is the same: attractant binding leads to smooth swims due to the inhibition of the kinase and the subsequent
decrease in the concentration of CheYp [46, 47].

Receptor Structure
The chemotaxis receptors in E. coli, B. subtilis, and nearly all other organisms share
certain basic features (fig. 2). The first is that the receptors always form homodimers

Chemotactic Sensory Transduction

35

Table 1. Chemotaxis signaling proteins found in E. coli and B. subtilis


Protein

Description

E. coli

B. subtilis

CheA

Histidine kinase; phosphorylates CheY and other


response regulators (CheB and CheV) [40, 43, 54, 61, 117,
147160]

CheB

Receptor methylesterase; demethylates specific


methyl-glutamate residues on the receptors
[66, 94, 117, 129, 130, 140, 150, 154, 161173]

CheC

CheYp phosphatase; primary role is in adaptation


via binding to CheD [45, 121, 122, 134137, 139,
174180]

CheD

Receptor deamidase; primary role is in adaptation


via binding to CheC [121, 122, 134139, 178, 179]

CheR

Receptor methyltransferase; methylates specific


glutamate resides on the receptors [55, 125, 127,
140, 143, 164, 170, 181207]

CheV

Couples CheA to receptors; response regulator


domain can be phosphorylated; role in adaptation
[27, 5860, 94, 145, 178, 208]

CheW

Couples CheA to the receptors [22, 3942, 51, 54,


58, 59, 61, 67, 88, 209231]

CheY

Response regulator. When phosphorylated, CheYp


interacts with the flagellar motors and alters frequency
of runs and tumbles [28, 45, 106, 121, 122, 142, 169,
177, 224, 232272]

CheZ

CheYp phosphatase [174, 239, 259, 273287]

FliY

CheYp phosphatase; integral part of flagellar switch


[44, 45, 122, 174, 175, 179, 288, 289]

In the last two columns, x denotes the presence of that protein in the organism.

[48]. The second is that the N-terminal region usually, but not always, consists of an
exterior part that binds ligand, either directly [49] or via an intermediate binding protein [50]. The C-terminal region, on the other hand, consists of the cytoplasmic region
of the receptor. This region contains the so-called highly-conserved domain (HCD)
that binds to and interacts with the CheW adaptor protein and CheA kinase [51].
Flanking the HCD are the two methylation (adaptation) regions [52, 53]. Linking the
exterior N-terminal and interior C-terminal regions is the HAMP domain (see fig.
2) [54, 55]. Except for the HAMP domain, which is a parallel coiled coil, the interior

36

Rao Ordal

Sensing

TM1

TM2
N
HAMP
C

Methylation

Fig. 2. Domain structure of the chemotaxis receptor. Diagram


showing the domain organization of a typical chemotaxis receptor. Note the receptors form stable homodimers, whereas the figure only depicts a single monomer. N = Amino-terminus; TM1 =
transmembrane region 1; TM2 = transmembrane region 2; HAMP
= a common cytoplasmic linker domain found in histidine kinases,
adenlyl cyclases, methyl-accepting chemotaxis receptors, and
phosphatases; HCD = highly conserved (or signaling) domain; C =
carboxyl-terminus.

HCD

(cytosolic) region is an anti-parallel coiled coil having multiple heptad repeats [10,
56]. The number of repeats varies among species. Based on this number receptors can
be classified into seven groups [56].

Ternary Complex
The chemotaxis receptors, CheW adaptor proteins, and CheA kinases form a stable
ternary signaling complex (fig. 3). For the kinase to achieve full activity, it must associate with the receptors and CheW. In E. coli, the approximate stoichiometry of this
complex is 3.4 receptor dimers and 1.6 CheW adaptor proteins per CheA dimer [57].
These proteins form extended structures involving perhaps thousands of proteins (see
below: Structure and Higher-Order Organization). In B. subtilis, the stoichiometry of
the ternary signaling complex is unknown. Furthermore, for reasons discussed below,
it is unlikely to be the same as E. coli. We also note that there is an additional adaptor
protein, CheV, in B. subtilis. CheV consists of an N-terminal CheW-like domain and
a C-terminal CheY-like response regulator domain [58]. Our current understanding
suggests that CheV is redundant to CheW except that it also plays a role in adaptation
[59] (see below: CheV System). In particular, cheW and cheV null mutants are still
capable of excitation and adaptation, whereas cheWcheV null mutants are tumbly and
unresponsive to attractants due to the impaired coupling of the CheA kinase with the
receptors [60].

Chemotactic Sensory Transduction

37

Fig. 3. Ternary complex. In both E. coli and B. subtilis, the receptors form a stable ternary complex with the adaptor protein,
CheW, and histidine kinase, CheA. The complex enhances CheA
activity roughly 100-fold [39, 290]. In the case of B. subtilis, there is
also a second adaptor protein, CheV. CheW and CheV are redundant in the sense that only one is necessary to enable robust
kinase activity. Signaling from the receptors to the flagellar motors
is achieved through the phosphorylation of the CheY response
regulator. The basic mechanism is that CheA autophosphorylates
at a conserved histidine residue [291]. The phosphoryl group on
CheA is then transferred a conserved aspartate residue on CheY
(150). This His-Asp motif is common in many bacterial signal
transduction pathways [292].

W
A

Yp

Structure and Higher-Order Organization


Structures for both the cytoplasmic domain of the serine receptor, Tsr, from E. coli
[52] and an orphan chemotaxis receptor, Tm1143, from Thermotoga maritima [61]
have been determined. In the case of Tsr, the receptors form trimers-of-dimers within
the crystal [52]. Subsequent investigations by Parkinson and coworkers [62, 63] demonstrated that the receptors also form these trimers-of-dimers in vivo. Furthermore,
different receptors can assemble to form mixed trimers-of-dimers (e.g. Tsr and Tar,
the aspartate receptor) [62], though the individual monomers within a dimer are the
same receptor type [64]. It is now believed that the trimer-of-dimer assembly represents the basic building block within extended receptor assemblies [10]. In the case
of Tm1143 from T. maritima, the receptors form isolated dimers within the crystal.
Furthermore, Tm1143 cannot be modeled as a trimer-of-dimers, as it lacks the characteristic bend observed in Tsr [10, 61]. This would suggest that the basic building
block for the receptor assemblies in T. maritima are simply dimers. The chemoreceptors from B. subtilis closely align with those from T. maritima. In fact, of the seven
classes of chemoreceptors, as determined by the number of heptad repeats, the receptors from B. subtilis and T. maritima both fall in the same class [56]. Furthermore,
receptor chimeras involving the N-terminal ligand-binding domain of McpB and the
C-terminal cytoplasmic domain of Tm1143 are capable of mediating taxis towards
asparagine, the ligand for McpB, in B. subtilis [unpubl. results]. Collectively, these
results suggest that Tm1143 is a valid model for the chemotaxis receptors in B.
subtilis.
The chemotaxis proteins cluster at distinct locations in both E. coli and B. subtilis.
In the case of E. coli, these clusters form at the poles and future division sites within
the cell [65]. In the case of B. subtilis, clustering has also been observed [66], though
a comparably detailed study of localization has not yet been performed in this organism (fig. 4). In addition to clustering, cyro-electron tomographs of these clusters in E.
coli suggest that the receptors assemble in semi-ordered arrays [67, 68]. The structure

38

Rao Ordal

Fig. 4. Immunofluorescence
of McpB localization in B. subtilis. The image shows the localization of the McpB asparagine
receptor as determined by
antibody staining. The cells
were also labeled with propidium iodide. Similar localization
patterns are seen for the
receptors and signaling proteins. This figure was reprinted
from Kirby et al. [66].

of these arrays is still unknown although they have long been thought to resemble
triangular lattices involving trimer-of-dimer receptor subunits in E. coli [69]. As the
T. maritima receptors do not form trimers-of-dimers within crystals, Crane and colleagues [61] proposed that these receptors would instead form hedgerows within
the receptor lattice. These results would suggest that different lattice structures form
based on the chemoreceptor types, a result not entirely surprising given the variability observed in different species of bacteria.

Attractants
B. subtilis has ten chemotaxis receptors, eight transmembrane and two soluble [70].
Of these, the cognate ligands are known only for three [7073]. In the case of the
transmembrane receptors McpB and McpC, both can sense a number of amino
acids. However, only McpB can sense asparagine and only McpC, proline [72, 73].
Presumably, sensing occurs through the direct binding of the cognate amino acid to
the extracellular domain of the receptor. The wide range of ligands in the case of
McpC, though, suggests that other proteins may be involved (likely by binding the
ligand and then interacting with the receptor). The soluble receptor, HemAT, senses
oxygen through the direct binding to a heme group [70]. In E. coli, there are five receptors, four transmembrane and one membrane tethered. The four transmembrane

Chemotactic Sensory Transduction

39

receptors sense: serine and some repellents (Tsr); aspartate, maltose, and other repellents (Tar); ribose and galactose (Trg), and dipeptides (Tap). The membrane-tethered
receptor Aer indirectly senses oxygen [7477].
One known departure from this paradigm occurs in the case of sugar taxis. In B.
subtilis, the McpC receptor is absolutely required for taxis towards glucose. However,
chimeras involving the cytoplasmic C-terminal domain McpC and extracellular
N-terminal domain of McpB still facilitate taxis towards sugars, even though the
MpcB receptor alone does not. Both McpB and the chimera sense asparagine. These
results and others indicate that the cytoplasmic domain of the receptors is involved in
sugar taxis whereas the extracellular domain is involved in amino acid taxis. Likely,
sensing by the cytoplasmic domain is due to the operation of the phosphotransferase
system (PTS), where EnzI (but not EnzIp) appears to bind McpC and induce a stimulatory conformation change [71]. In E. coli, taxis towards sugars also involves the
PTS system. However the interaction is not through the receptors but rather though
the CheA kinase. In particular, unphosphorylated EI inhibits the kinase activity of
CheA [78]. We also note that the response towards sugars is also regulated indirectly
through the sensing of the redox state and proton motive force by Aer and Tsr, respectively [79]. See the chapter by Lengeler in this volume and Deutscher et al. [80] for
more complete discussions of E. coli taxis towards PTS sugars.

Repellents
A diverse catalog of substances includes repellents of B. subtilis taxis at, sometimes,
exceedingly low concentrations. For example, uncouplers of oxidative phosphorylation such as FCCP cause the repellent response (tumbling) at 10 nm. At first, the
repellents were thought to act by diminishing the protomotive force driving flagellar
rotation, an interpretation that accords with the Mitchell hypothesis concerning the
mode of action of uncouplers [81, 82]. However, the fact that the response was phasic (undergoes adaptation) as opposed to tonic (no adaptation) enabled a test of this
interpretation. Preincubation of one uncoupler greatly desensitized the bacterium to
further additions of the same uncoupler but usually had little effect on the sensitivity
to chemically dissimilar uncouplers [83]. At the time when these experiments were
performed, the chemotaxis receptors in B. subtilis had not been identified. Therefore,
to understand the mechanism of these uncouplers, one the authors decided to investigate amino acid transport, which was also inhibited by the uncouplers in the presence of glycerol (typically added as an energy source in chemotaxis experiments). At
the time, these transporters were the only known membrane proteins amendable to
analysis. Using these proteins as a surrogate, the mechanism for inhibition by different uncouplers was found to be varied. According to the most straightforward
interpretation of the results, three different mechanisms are involved depending on
the specific uncoupler. In the first case, the transport substrate could bind or the

40

Rao Ordal

uncoupler could bind but not both (competitive); in the second case, the transport
substrate had to bind for the uncoupler to bind (uncompetitive); in the last case, the
uncoupler inhibited whether the transport substrate was bound or not and the latter was unaltered in its KM (non-competitive). Thus, different uncouplers bound to
particular proteins in very specific ways [84]. At this point, which receptors mediate
taxis away from these repellents are not known. However, the fact that both excitation and adaptation occur indicates that an orthodox process is taking place. There
are also a number of repellents for E. coli, however these generally act at fairly high
concentrations and only in a few cases have the receptors have been identified [74,
85].

Transmission
In E. coli, the binding of attractant is known to induce a toward-the-cytosol (downward) movement of one of the four transmembrane (TM2) helices of the dimer. The
two first transmembrane (TM1 and TM1) helices and the other, second, transmembrane (TM2) helix are stationary [86, 87] (see fig. 2). Note that the prime is used to
denote the transmembrane region of the second receptor monomer within the dimer.
This downward piston-like motion is believed to alter the conformation of the HAMP
domain leading to weakened monomer/monomer interactions, increased flexibility,
and motions within the trimer-of-dimers complex [10]. These changes ultimately
lead to decreased CheA kinase activity [88]. The basic model for signal transmission
assumes that the ternary complex can exist in one of two allosteric states: active and
inactive. Attractant binding alters the equilibrium between the two states by reducing
the free energy of the inactive state, in other words making the inactive state energetically more favorable [8991]. Likewise, receptor methylation counteracts this
change by increasing the free energy of the inactive state (see below: Adaptation in E.
coli). We also note that there are many more elements and nuances to this simplified
description of the two-state model, including cooperative interactions among neighboring receptors within the signaling clusters and effects of heterogonous receptors.
By contrast, in B. subtilis, there is a rotation induced by attractant binding between the
helices TM1 and TM1 while the TM2 and TM2 helices remain stationary (fig. 5). In
particular, there is no vertical movement unlike the case in E. coli [92]. Interestingly,
only the TM2 and TM2 helices connect the extracellular ligand-binding domain to
the intracellular signaling domain. Therefore, these rotations in TM1 and TM1 likely
change the interactions of receptor dimers within the signaling lattice by altering
the packing (see below), though it is still not clear how these changes translate into
changes in kinase activity.
The transmission of the signal induced by attractant binding also involves a loosening in the packing of receptors within the lattice for both E. coli and B. subtilis
as determined by paraformaldehyde cross-linking experiments [93]. This loosening,

Chemotactic Sensory Transduction

41

281
37

30

44
40

34

278

277

33

41

274 285

282

284

II

275

273

36

38

286

280

279

43

31
32

42

39

35

35

39

42

32
272

283 276

31

43

279
280

286

39

276 283

38

36

273

275

II

282

284
277

278

34

40
41

33
44

37

30

285 274 281

Fig. 5. Rotation induced by the binding of asparagine to McpB in B. subtilis. The binding of attractant induces a rotation between the first two transmembrane helices, TM1 and TM1, as determined
by disulfide cross-linking experiments. In the figure, TM1 and TM2 denote the transmembrane helices for one receptor monomer and TM1 and TM2 denote the transmembrane helices for the other.
The numbers denote the specific amino acid residues. This figure was reprinted from Szurmant and
Ordal [146].

however, decreases as a function of time post attractant addition, indicating that the
adaptation process counteracts the loosening. Since the same response is observed in
both B. subtilis and E. coli, the loosening somehow activates the CheA kinase in the
former whereas it inhibits the kinase in the latter. Interestingly, a B. subtilis mutant
lacking both adaptor proteins CheW and CheV showed a looser structure than
wild-type and was not further affected by addition of attractant. However, a similar
mutant was still able to produce methanol upon the addition of attractant when the
CheB methylesterase was made constitutively active to account for the loss of kinase
activity. As methanol production is a result of the adaptational response [94] (see
below: The Methylation System), these results would indicate that, even in absence
of CheW and CheV, attractant increases the availability of methyl groups to the enzymatic action of CheB. We note that the degree of cross-linking decreased as function of time post addition in the cheWcheV null mutant, consistent with the role of

42

Rao Ordal

adaptation affecting receptor packing [93]. In addition, the rate of methanol production in this mutant was five times the corresponding rate in strains where CheB is
constitutively active in an otherwise wild-type background. These results suggest that
the presence of the adaptor proteins makes the receptors pack more densely and, as a
consequence, provide less access to CheR and CheB [94].
Also consistent with a model where attractant binding alters the packing of receptors within an ensemble is the fact that multivalent ligands are much more potent
agonists than monomeric ligands in both E. coli and B. subtilis. Furthermore, in the
case of E. coli, binding of multivalent galactose, whose taxis is mediated by chemotaxis receptor Trg, potentates the response to serine, whose taxis is mediated by the
chemotaxis receptor Tsr. Interestingly, in the same type of experiment, taxis to repellents no longer occurred. Indeed, leucine, a repellent in E. coli, became an attractant
when delivered as a large multivalent ligand [93, 9598].

Signaling Arrays and Receptor Crosstalk


As noted above, the chemotaxis receptors and associated proteins cluster at distinct
locations in the cell. These clusters likely result from the formation of large, extended
signaling arrays. Such arrays, as first postulated by Bray and colleagues [69, 99103],
would enable cells to sense small gradients over a wide range of concentrations. In
particular, a single ligand-bound receptor dimer (or trimer-of-receptor dimers in
the case of E. coli) could influence the activity of neighboring dimers (or trimers-ofdimers) within the array. In other words, the receptors form large cooperative structures. Such an amplification mechanism could potentially explain the long-standing
paradox of how the E. coli chemotaxis system is exquisitely sensitive to small changes
in receptor occupancy over a wide range of attractant concentrations. For example, a
1% change in receptor occupancy causes a 50% change in the motor response [104].
In addition to the amplification mechanism, the Bray model also predicted varying
levels of cooperativity depending on the concentration of attractant. For example, the
receptors within the lattice are highly cooperative at low ligand concentrations and
behave akin to isolated receptors at high concentrations.
Subsequent work by Victor Sourjik, Howard Berg, and others [105108] experimentally validated key aspects of the Bray model along with identifying a number of additional factors. In particular, they demonstrated that the sensitivity over
a broad range of concentrations is due, in part, to receptor methylation not only
compensating for the inactivating effects of attractant on the CheA kinase but also
reducing the affinity for attractant. Also, at high levels of methylation, the presence
of heterogeneous receptors (e.g., Tsr) can make the primary receptor (e.g., Tar for
aspartate) less sensitive to the attractant. Likewise, at low levels of methylation, the
primary receptor is able to suppress all kinase activity when bound with attractant
even though the heterogeneous receptors are unbound. Therefore, by varying the

Chemotactic Sensory Transduction

43

level of receptor methylation, cells are able to detect small gradient changes over a
wide range of concentrations for a particular attractant by tuning the affinity and
the threshold of the primary receptor. To understand the basic idea further, consider
the gedanken experiment involving a cell with just two types of receptors, one for
aspartate and one for serine, where the integrated activity of the kinase is proportional to the sum of the activity of the two receptors. In the absence of coupling, if
the aspartate receptor is bound, then the kinase activity will be reduced by one half
as the serine receptor is still unbound and capable of activating the kinase. However,
if the system is strongly coupled, the ligand-bound aspartate receptor can also deactivate the serine receptor and thus amplify the response, a behavior desirable at low
background attractant concentrations. Alternatively, the unbound serine receptor can keep the aspartate receptor in an active state even though it is bound with
attractant and thus attenuate the response, a behavior desirable at high background
attractant concentrations. Such coupling can tune the sensitivity of the receptor output to the concentration of attractant. In fact, Berg and Sourjik data suggest that
at low levels of receptor methylation (induced by a low background concentration
of attractant), a few ligand-bound receptors are sufficient to deactivate the entire
ensemble of receptors [106]. Likewise, at high levels of methylation (induced by
a high background of concentration of attractant), many ligand-bound receptors
are necessary to deactivate the ensemble. Thus, these results validate many aspects
of the Bray model, including varying degrees of coupling. They also suggest that
receptor methylation, in addition to its role in adaptation, tunes the sensitivity of
the receptor complex.
We note that a number of subsequent additions and reformulations to the original
Bray model have been proposed over the years, though the key ideas still remain [90,
91, 109113]. One open question is how receptors communicate with one another,
in particular between neighboring trimers-of-dimers. One set of experiments suggests that interactions between trimers may occur along the extracellular face of the
receptors [114]. Such a mechanism would explain the large cooperativities observed
experimentally (e.g. Hill coefficients >10) and also how allosteric signals are propagated over large distances [105, 108].
In B. subtilis, the role of cooperativity and heterogeneous receptor interactions has
not been explored to the same degree. Presumably, a similar mechanism exists as B.
subtilis is able to migrate up gradients over a similarly wide range of concentrations
for most amino acids [115, 116]. Furthermore, B. subtilis cells expressing the asparagine receptor McpB as the sole receptor have an initially tumbly bias though excite
normally in response to asparagine but do not completely adapt, perhaps due to an
insufficient number of receptors in the cell. Interestingly, we do note that these cells are
still capable of taxis. Likewise, in cells where the methylation sites on McpB have been
changed to unmethylatable aspartate residues and the other receptors are unchanged,
the cells are still capable of taxis. Furthermore, methanol is released in response to
asparagine, indicating the adaptation is occurring through the demethylation of other

44

Rao Ordal

receptors. This result indicates the other receptors are able to compensate for the loss
of the McpB methylation sites. Collectively, these results suggest that heterogonous
receptor interactions occur in B. subtilis as well [117, 118].

Phosphatases
E. coli has a single phosphatase, CheZ, that enhances the rate of CheY dephosphorylation. The phosphatase is thought to enable the pathway to respond to rapid changes
in attractant/repellent concentrations. In addition, CheZ localizes to the receptor
complexes, binding the short form of CheA (CheAs). In enteric bacteria, cheA has an
internal translation initiation site that yields the CheAs gene product. Localization
of CheZ maintains a uniform concentration of CheYp throughout the cell at steady
state, ensuring that the different motors receive the same signal [119, 120]. In B. subtilis, there are two phosphatases, FliY and CheC. FliY is localized at the flagellar motor
whereas CheC is thought to localize with the receptors. Of the two, FliY is the stronger phosphatase. A fliY615 mutant, in which CheYp has reduced binding affinity for
the phosphatase, is mostly smooth swimming. A cheC null mutant, on the other hand,
has a wild-type bias. In particular, we hypothesize that CheC functions primarily to
regulate CheD rather than dephosphorylate CheYp (see below: The CheC/CheD/
CheYp System) [45, 121123].

Adaptation

Adaptation in E. coli
The adaptation system in E. coli involves receptor methylation (fig. 6). The binding of attractant inhibits the ternary complex in E. coli, leading to significantly
reduced CheA kinase activity. This inhibition reduces the levels of CheYp and, as a
consequence, increases the likelihood of smooth swims. This excitatory response,
however, is only transitory. Activation of the ternary complex through attractant
binding also leads to increased methylesterification of specific glutamate side
chains on the receptors [124, 125]. This reaction is catalyzed by the CheR methyltransferase using S-adenosyl methionine as the methyl group donor [126].
Increased methylation of the receptor reduces the affinity of ligand and counteracts the inhibition due to ligand binding (i.e. reverses the downward piston shift
via restoring the original conformation of the HAMP domain [127]). Thus, CheA
kinase activity is eventually restored to prestimulus levels by increased methylation of the receptor, in other words adaptation. The reciprocal process occurs when
E. coli is exposed to repellents or the attractant is removed and is lost from the
receptors. This time CheA kinase activity is enhanced, leading to elevated CheYp

Chemotactic Sensory Transduction

45

Ligand

CH3

+ CH3
W

Fig. 6. Diagram of the chemotaxis pathway in E.


coli. Attractants bind to the receptor and inhibit
CheA activity. The result is that less CheY is phosphorylated, leading to an increase in smooth
runs. When the receptors become bound with
attractant, the rate of methylation by CheR
increases and the rate of demethylation by
phosphorylated CheB decreases. The net result
is increasing receptor methylation and increasing kinase activity, despite presence of the
kinase-inhibiting attractant.

Bp

B
Yp

Motor
(Tumble)

levels and an increased likelihood of reorientating tumbles [106]. Inhibition of the


ternary complex also causes the demethylation of the methylated glutamate side
chains [128]. The demethylation reaction is catalyzed by the CheB methylesterase
[129]. In order to be active, CheB must first be phosphorylated by CheA. However,
phosphorylation is not necessary for adaptation, as constitutively active variants
of CheB lacking the response regulator domain are still capable mediating adaptation [94, 130, 131]. Removing the methyl groups from the receptors increases their
affinity for attractant and also counteracts the addition of repellents or the loss of
attractant by decreasing CheA kinase activity, enabling adaptation. In summary,
adaptation in E. coli results from the reciprocal action of the CheR methyltransferase and CheB methylesterase.
The methylation and demethylation reactions are regulated in response to
changes in receptor conformation that occur due to positive (e.g. attractants) or
negative (e.g. repellents or attractant removal) stimuli. If one assumes that the receptors can exist either in an active or inactive conformation (i.e. a two-state model),
then the accumulated evidence indicates that active receptors are preferentially
demethylated by CheB whereas inactive receptors are preferentially methylated by
CheR [132]. This selective recognition forms two antagonizing negative feedback
loops where active receptors are deactivated due to demethylation by phosphorylated CheB and inactive receptors are activated due to methylation by CheR. These
antagonizing loops ensure that the fraction of active and inactive receptors always
returns to prestimulus levels following a perturbation due to either positive or negative stimuli.

46

Rao Ordal

Adaptation in B. subtilis
Unlike the single adaptation system in E. coli, at least three adaptation systems operate
in B. subtilis: the CheV system, the CheC/CheD/CheYp system, and the methylation
system. Each of these systems appears to partially enable adaptation, though completing the adaptation process over the full range of concentrations requires all three.

CheV System

CheV has two domains, an N-terminal CheW-like adaptor domain and a C-terminal
response regulator domain [58]. Because it has the adaptor domain, the CheV protein is redundant to CheW in the sense that both cheW and cheV null mutants are
still capable of adaptation and gradient sensing (though not over the same range of
concentrations as wild-type). However, a cheWcheV null mutant is tumbly, consistent with the inability of the receptors to interact with and activate CheA in such
strains [60]. While the mechanism underlying adaptation by CheV is still unknown,
we imagine that the phosphorylation of the response regulator domain by CheA
yields a conformational change that inhibits CheA kinase activity, likely by disrupting
the interaction between the receptor and the kinase [59] (fig. 7). Such a mechanism
would provide a negative feedback loop; active kinase phosphorylates CheV, which
in turn inhibits kinase activity. Interestingly, a CheV truncation mutant lacking the
response regulator domain fails to adapt to the addition of attractant, suggesting that
phosphorylation somehow tunes the mode of interaction. Remarkably, a similar phenotype is observed in a CheV mutant where the phosphorylation site is mutated [59].
We also note that the proposed mechanism for regulation by CheV is not capable of
completing the adaptation process, as CheVp levels will track CheAp levels due to the
phosphoryl group on CheV being readily hydrolyzed (fig. 8). Thus, the CheV system
is only capable of partially adapting the response. As a comparison, the methyl-glutamates on the receptors in E. coli are stable and thus capable of sustaining activation/inhibition in response to sustained negative/positive stimuli [128]. This integral
control is known to be necessary for complete adaptation when feedback is employed
[133]. Therefore, we hypothesize that CheV fine-tunes adaptation and only is significant in response to small gradients (see below: Integration of the Three Adaptation
Systems in B. subtilis).

The CheC/CheD/CheYp System

CheC is a CheYp phosphatase [122]. However, as noted above, its primary role does
not appear to involve dephosphorylating CheYp. Rather, CheC appears to regulate
CheD in response the CheYp levels [121]. CheD is a receptor deamidase that converts

Chemotactic Sensory Transduction

47

Ligand

Vp

W
A

Fig. 7. Diagram of the CheV adaptation system


in B. subtilis. Attractant binding activates the
CheA kinase, leading to increased phosphorylation of CheY and CheV. Phosphorylated CheV is
then thought to inhibit kinase activity by disrupting the coupling between the receptors and
CheA.

1.0
Wild-type
CheBCD

0.8

CCW bias

Fig. 8. Comparison of the full


adaptation system versus the
only CheV adaptation system
in B. subtilis. The cells were
exposed to 0.5 mM asparagine
beginning at the time denoted
with the upward arrow and
ending with the time denoted
with the downward arrow. The
cheBCD strain only possesses
the CheV adaptation system.
Note that this strain is only
capable of partial adaptation.
This figure was reprinted from
Saulmon et al. [123].

Yp

0.6
0.4
0.2

400

800
Time (s)

1,200

1,600

specific glutamines to glutamates [134]. In addition to its enzymatic function, CheD


regulates kinase activity. For example, cheD null mutants are tumbly, consistent with
CheD being necessary for kinase activation. The mode for this regulation is unknown,
though CheD has been shown to directly bind the receptors [135137]. This regulation by CheD, however, involves only a subset of receptors. For example, cells only
expressing the asparagine receptor, McpB, do not require CheD for taxis; however,
cells expressing only the proline receptor, McpC, do [138] (note that in an ensemble
of receptors such as in wild-type, all taxis is CheD-dependent [139]). Through the
use of receptor chimeras, the HAMP domain was determined to be the critical determinant. Chimeras of McpC, where the native HAMP was partially replaced by the
one from McpB, no longer require CheD for taxis. This dependency does not involve

48

Rao Ordal

Ligand

D
V

Yp

D
C

W
A

Yp

Fig. 9. Diagram of the CheC/CheD/CheYp adaptation system. Left part shows CheD bound to receptors. Right part shows what is believed to happen when attractant binds. The key element is that
higher levels of CheYp lead to more CheC-CheYp complexes, which then attracts CheD away from
the receptors. That loss inhibits the ability of the receptors to activate CheA kinase (adaptation). The
greater binding of CheD to CheC due to presence of CheYp also enhances CheYp dephosphorylation and thus sharpens the signal. For illustration purposes the CheD is shown away from the receptors and bound to CheC. Very likely, however, only some of the CheD leaves the receptors [121].

CheDs deamidase activity [138]. Finally, CheD can also bind to CheC and enhances
its phosphatase activity. Binding is CheYp-dependent, as CheC-CheYp complex is a
better binding partner than CheC alone. In addition, the interaction between CheD
and CheC/CheYp is competitive with the interaction between CheD and the receptors. This competitive interaction forms the basis for the CheC/CheD/CheYp adaptation system [121, 137].
The working model for the CheC/CheD/CheYp adaptation system is as follows
(fig. 9). Prior to stimulation, we imagine that a fraction of CheD is bound to the
receptors. When attractant binds the receptors, the kinase is activated and CheYp
levels are increased. CheYp then binds to CheC. This complex then provides an alternate binding target for CheD. This results in less CheD bound to the receptors. As a
consequence, the kinase is inhibited. This mechanism yields a negative feedback loop
involving CheYp as the regulatory signal. Likewise, when the attractant is removed
or, equivalently, a repellent added, the kinase is inhibited and the levels of CheYp
decrease. As a consequence, fewer CheC-CheYp complexes are formed and more
CheD is bound to the receptors. Because CheYp is rapidly dephosphorylated by the
CheC-CheD complex, this mechanism cannot completely adapt the response for the
reasons argued above (see above: CheV System), unless attractant-bound receptors
have a lower affinity for CheD as described below.
One alternate model is that receptor activation causes CheD to dissociate from
the receptors. Unbound CheD could then associate with CheC to increase CheYp

Chemotactic Sensory Transduction

49

dephosphorylation. Such a feedfoward mechanism would adapt the response at the


level of CheYp rather than CheA kinase activity as argued above. However, CheC
mutants with reduced phosphatase activity are still capable of reduced rates of taxis
whereas CheC mutants unable to bind CheD are incapable of taxis [121, 137]. These
results would argue for the former (depriving receptors of CheD) rather than the latter (enhanced phosphatase) mechanism. They also underscore that the primary role
of CheC is to regulate CheD during adaptation and not to dephosphorylate CheYp.

The Methylation System

The chemotaxis receptors in B. subtilis are also subject to reversible methylation by


the CheR methyltransferase and CheB methylesterase. Furthermore, the basic mechanism is the same in both E. coli and B. subtilis, as both enzymes from B. subtilis
can complement the corresponding E. coli null mutant [140, 141]. Despite employing
the same mechanism for adding and removing methyl groups at specific glutamate
residues, the role of methylation in adaptation in B. subtilis is quite different than in
E. coli. In particular, the receptors are subject to a cycle of rapid demethylation followed by slow methylation in response to both positive and negative stimuli (fig. 10).
Furthermore, the total number of methylated glutamates is roughly constant at steady
state [142]. By contrast, the receptors in E. coli are methylated in response to positive
stimuli and demethylated in response to negative stimuli. Here, the number of methylated glutamates roughly tracks the logarithm of the concentration of attractant at
steady state [128]. The differences between the two organisms are due to how receptor methylation affects kinase activity. In E. coli, methylation of the glutamate residues has an additive effect: the greater the number methylated, the more active the
kinase and the less able the receptors can bind attractant [143]. In B. subtilis, methylation has an antagonizing effect: methylation of certain glutamates activates the kinase
whereas methylation of others deactivates the kinase [118]. Thus, the cycle of rapid
demethylation followed by slow remethylation can be viewed as the movement of
methyl groups from one set of glutamates to another. While the details of receptor
methylation in B. subtilis are still unknown, data suggest that methylation of residue
630 on McpB activates the kinase whereas methylation of residue 637 deactivates it
[118]. Thus, the addition of attractant would first cause residue 630 to be demethylated and then residue 637 to be methylated. Likewise, upon the removal of attractant,
residue 637 is first demethylated and then residue 630 is methylated. Such a mechanism is consistent with the observed methylation cycle and could help explain how
selective methylation enables adaptation in B. subtilis. We also note that an additional
glutamate at position 371 on McpB is also subject to reversible methylation, though
its effect is still unknown [118].
Despite the appeal of the above model for adaptation through receptor methylation, a number of puzzles still remain. The first has to do with timing. When the

50

Rao Ordal

0.5

0.4

3 H-McpB (OD units)

Fig. 10. Methylation changes


on McpB in response to the
addition and removal of attractant. The downward arrow
denotes the time when 0.5 mM
asparagine was added to the
cells and the upward arrow the
time of removal. This assay was
performed by growing the cells
in radiolabeled [3H]methionine
to a steady state and then
measuring the degree of McpB
methylation using Western
blotting and fluorography.
Note the receptors are rapidly
demethylated in response
both to the addition and
removal of attractant. Likewise,
remethylation also occurs
though on a much slower time
scale. This figure was reprinted
from Kirby et al. [142].

0.3

0.2

0.1

0
20

30

40

50
Time (m)

60

70

80

cells are abruptly exposed to an attractant such as asparagine, the demethylation


event is rapid, occurring within the first few seconds. However, the remethylation is
much slower, taking almost 20 min to reach steady state [142]. Adaptation, as determined by the frequency of runs and tumbles, on the other hand, takes <1 min [144].
Based on the relative timing of different events, the remethylation step does not
appear to be involved in adaptation. Rather, the demethylation step in conjunction
with the other adaptation systems appears to be sufficient for the system to adapt.
The remethylation step must, therefore, have some other role, likely affecting the
sensitivity of the receptors to ligand. In other words, remethylation resets the other
adaptation systems so that they can responds to further additions of attractant or,
equivalently, sense gradients over wide ranges of concentrations. We also note that,
in nature, the apparently slow remethylation process might not be so slow given
that the cell will likely be confronted only with slowly changing concentrations of
attractant. By contrast, in the laboratory setting, we expose cells to very large, and
most likely unnatural, changes in attractant concentration in order to tease out various behaviors.
The second puzzle concerns selectivity. How do CheR and CheB determine which
residues need to methylated and demethylated? Some hints are available in various
deletion mutants. The receptors are weakly methylated in a cheD mutant whereas they
are strongly methylated at levels roughly twice wild-type in a cheC null mutant [135].
Furthermore, in a cheY null mutant, the receptors are rapidly demethylated upon the
addition of attractant. However, the receptors do not remethylate as is the case in
wild-type (fig. 11). Upon the removal of attractant, the receptors are remethylated

Chemotactic Sensory Transduction

51

0.6

Fig. 11. Methylation changes


on McpB in a cheY mutant in
response to the addition and
removal of attractant. The
assay and conditions are the
same as described in figure 10.
Note that the receptors do not
remethylate in response to
asparagine in the cheY mutant.
This figure was reprinted from
Kirby et al. [142].

3 H-McpB (OD units)

0.5

0.4

0.3

0.2

0.2

0
40

50

60

70
Time (m)

80

90

100

[142]. Collectively, these results suggest that the CheC/CheD/CheY adaptation system plays a role in selective methylation.
The third concerns reversibility. In a cheB null mutant, the cells are able to respond
to the addition of attractant but only weakly or not at all (at low concentrations) to
the removal. Furthermore, at low levels of attractants, the cells are also able to adapt
[66]. The fact that they adapt is not surprising as two other adaptation systems are
present. What is surprising is that the cells do not respond to negative stimuli. These
results would suggest that methylation plays a role in tuning the sensitivity of the
receptors, such that they are able to respond to positive and negative stimuli. But,
methylation does not seem important in regard to adaptation at low attractant concentrations. Perhaps different adaptation systems are active over different concentration ranges.
Integration of the Three Adaptation Systems in B. subtilis

A striking feature of the adaptation systems in B. subtilis is that they are in many
ways redundant yet do not function independently of one another. Deletion of
any one system leads to a moderate inhibition of chemotaxis. However, deletion
of any two significantly impairs chemotaxis at all concentrations of attractant. In
particular, there is no taxis for the cheBcheV and cheBcheC double mutants and
minimal taxis (7% relative to wild-type) for the cheCcheV mutant [unpubl. results].
Thus, all three systems contribute towards chemotaxis. The most important system appears to be the methylation system, followed by the CheC/CheD/CheYp

52

Rao Ordal

system. Also, as we have noted, these two systems may be coordinated with one
other. The working model for the time course of events following the addition
of the attractant asparagine is the following: (1) asparagine binds receptors; (2)
CheA activity increases; (3) CheYp levels increase; (4a) receptors are demethylated
at site 630; (4b) CheC binds CheYp and recruits CheD away from receptors; (4c)
CheVp levels increases; (5) CheA activity decreases; (6) at some point thereafter
CheAp, CheVp, and CheYp return to prestimulus levels (though not necessarily at
the same time); (7) receptors are slowly methylated at site 637. We also believe that
the CheV and CheC/CheD/CheYp are dominant at low concentrations of attractant whereas the methylation system is dominant at high concentrations [66, 141].
Clearly, our understanding of how these systems are integrated is still incomplete
as there are still many holes in our model. Filling them in, at present, is a current
research focus for both authors.

Concluding Remarks

Despite 40 years of progress, bacterial chemotaxis still presents us with a number of


unsolved puzzles. In the case of E. coli, the outstanding questions focus mainly on the
signaling clusters and arrays. In particular, what are the structures of these arrays and
how are signals communicated between adjacent receptors within the array? Part of
the justification for studying this system, beyond its inherent beauty, is that it be may
be the first pathway that we can understand and model from molecular detail all the
way to the integrated response. Furthermore, despite its apparent simplicity, the pathway yields a number of complex responses, making it a favorite among systems and
computational biologists.
In the case of B. subtilis, the outstanding question concerns how the three adaptation systems function both in isolation and as an integrated unit. While progress
has been made on the former, we still do not understand how the different adaptation systems work with one another to mediate taxis. We have succeeded in enumerating the pathway components and now need to understand how they all fit
together.
We finally note that B. subtilis is not unique in having multiple adaptation systems.
For example, Salmonella typhimurium appears to have a CheV adaptation system
[145] even though the pathway by all other measures is identical to E. coli. H. pylori,
on the other hand, lacks a methylation system and appears to gets by with just the
CheV adaptation system (albeit several of them) [27]. This diversity in the chemotaxis pathways among the different species of bacteria is still an unexplored arena.
Finally, we wish to point out that the B. subtilis chemotaxis pathway is very similar
to the chemotaxis pathways found in the archaea [146]. This similarity suggests that
the B. subtilis pathway may be close to the progenitor pathway from which all other
chemotaxis pathways are derived.

Chemotactic Sensory Transduction

53

Acknowledgements
We wish to thank George Glekas for helpful discussions. This work was supported by NIH grants
GM083601 to C.V.R. and GM054365 to G.W.O.

References
1 Berg HC, Brown DA: Chemotaxis in Escherichia
coli analysed by three-dimensional tracking. Nature
1972;239:500504.
2 Berg HC, Purcell EM: Physics of chemoreception.
Biophys J 1977;20:193219.
3 Brown DA, Berg HC: Temporal stimulation of
chemotaxis in Escherichia coli. Proc Natl Acad Sci
USA 1974;71:13881392.
4 Macnab RM, Koshland DE Jr: The gradient-sensing
mechanism in bacterial chemotaxis. Proc Natl Acad
Sci USA 1972;69:25092512.
5 Adler J: Chemoreceptors in bacteria. Science 1969;
166:15881597.
6 Adler J: Chemotaxis in bacteria. Science 1966;
153:708716.
7 Maddock JR, Shapiro L: Polar location of the
chemoreceptor complex in the Escherichia coli cell.
Science 1993;259:17171723.
8 Sourjik V, Berg HC: Localization of components of
the chemotaxis machinery of Escherichia coli using
fluorescent protein fusions. Mol Microbiol 2000;37:
740751.
9 Bray D: Signaling complexes: biophysical constraints on intracellular communication. Annu Rev
Biophys Biomol Struct 1998;27:5975.
10 Hazelbauer GL, Falke JJ, Parkinson JS: Bacterial
chemoreceptors: high-performance signaling in
networked arrays. Trends Biochem Sci 2008;33:9
19.
11 Eisenbach M: A hitchhikers guide through advances
and conceptual changes in chemotaxis. J Cell
Physiol 2007;213:574580.
12 Kollmann M, Sourjik V: In silico biology: from
simulation to understanding. Curr Biol 2007;17:
R132R134.
13 Kentner D, Sourjik V: Spatial organization of the
bacterial chemotaxis system. Curr Opin Microbiol
2006;9:619624.
14 Baker MD, Wolanin PM, Stock JB: Systems biology
of bacterial chemotaxis. Curr Opin Microbiol
2006;9:187192.
15 Baker MD, Wolanin PM, Stock JB: Signal transduction in bacterial chemotaxis. Bioessays 2006;28:9
22.

54

16 Rajagopala SV, Titz B, Goll J, Parrish JR, Wohlbold


K, McKevitt MT, Palzkill T, Mori H, Finley RL Jr,
Uetz P: The protein network of bacterial motility.
Mol Syst Biol 2007;3:128.
17 Falb M, Muller K, Konigsmaier L, Oberwinkler T,
Horn P, von Gronau S, Gonzalez O, Pfeiffer F,
Bornberg-Bauer E, Oesterhelt D: Metabolism of
halophilic archaea. Extremophiles 2008;12:177196.
18 Sasaki J, Spudich JL: Signal Transfer in Haloarchaeal
Sensory Rhodopsin- Transducer Complexes. Photochem Photobiol 2008;84:863868.
19 Sasaki J, Nara T, Spudich EN, Spudich JL:
Constitutive activity in chimeras and deletions
localize sensory rhodopsin II/HtrII signal relay to
the membrane-inserted domain. Mol Microbiol
2007;66:13211330.
20 Spudich JL: The multitalented microbial sensory
rhodopsins. Trends Microbiol 2006;14:480487.
21 Nutsch T, Oesterhelt D, Gilles ED, Marwan W: A
quantitative model of the switch cycle of an archaeal
flagellar motor and its sensory control. Biophys J
2005;89:23072323.
22 Terry K, Go AC, Ottemann KM: Proteomic mapping of a suppressor of non-chemotactic cheW
mutants reveals that Helicobacter pylori contains a
new chemotaxis protein. Mol Microbiol 2006;61:
871882.
23 Terry K, Williams SM, Connolly L, Ottemann KM:
Chemotaxis plays multiple roles during Helicobacter
pylori animal infection. Infect Immun 2005;73:803
811.
24 Josenhans C, Suerbaum S: The role of motility as a
virulence factor in bacteria. Int J Med Microbiol
2002;291:605614.
25 Andermann TM, Chen YT, Ottemann KM: Two
predicted chemoreceptors of Helicobacter pylori
promote stomach infection. Infect Immun 2002;70:
58775881.
26 Ottemann KM, Lowenthal AC: Helicobacter pylori
uses motility for initial colonization and to attain
robust infection. Infect Immun 2002;70:19841990.
27 Pittman MS, Goodwin M, Kelly DJ: Chemotaxis in
the human gastric pathogen Helicobacter pylori:
different roles for CheW and the three CheV paralogues, and evidence for CheV2 phosphorylation.
Microbiology 2001;147:24932504.

Rao Ordal

28 Del Campo AM, Ballado T, de la Mora J, Poggio S,


Camarena L, Dreyfus G: Chemotactic control of the
two flagellar systems of Rhodobacter sphaeroides is
mediated by different sets of CheY and FliM proteins. J Bacteriol 2007;189:83978401.
29 Porter SL, Wadhams GH, Armitage JP: In vivo and
in vitro analysis of the Rhodobacter sphaeroides
chemotaxis signaling complexes. Methods Enzymol
2007;423:392413.
30 Martin AC, Gould M, Byles E, Roberts MA, Armitage
JP: Two chemosensory operons of Rhodobacter
sphaeroides are regulated independently by sigma 28
and sigma 54. J Bacteriol 2006;188:79327940.
31 Porter SL, Wadhams GH, Martin AC, Byles ED,
Lancaster DE, Armitage JP: The CheYs of Rhodobacter
sphaeroides. J Biol Chem 2006;281:3269432704.
32 Thompson SR, Wadhams GH, Armitage JP: The
positioning of cytoplasmic protein clusters in bacteria. Proc Natl Acad Sci USA 2006;103:82098214.
33 Poggio S, Osorio A, Dreyfus G, Camarena L: The
flagellar hierarchy of Rhodobacter sphaeroides is
controlled by the concerted action of two enhancerbinding proteins. Mol Microbiol 2005;58:969983.
34 Wadhams GH, Martin AC, Warren AV, Armitage
JP: Requirements for chemotaxis protein localization in Rhodobacter sphaeroides. Mol Microbiol
2005;58:895902.
35 Hoang HH, Gurich N, Gonzalez JE: Regulation of
motility by the ExpR/Sin quorum-sensing system in
Sinorhizobium meliloti. J Bacteriol 2008;190:861
871.
36 Meier VM, Muschler P, Scharf BE: Functional analysis of nine putative chemoreceptor proteins in
Sinorhizobium meliloti. J Bacteriol 2007;189:1816
1826.
37 Rotter C, Muhlbacher S, Salamon D, Schmitt R,
Scharf B: Rem, a new transcriptional activator of
motility and chemotaxis in Sinorhizobium meliloti.
J Bacteriol 2006;188:69326942.
38 Attmannspacher U, Scharf B, Schmitt R: Control of
speed modulation (chemokinesis) in the unidirectional rotary motor of Sinorhizobium meliloti. Mol
Microbiol 2005;56:708718.
39 Borkovich KA, Kaplan N, Hess JF, Simon MI:
Transmembrane signal transduction in bacterial
chemotaxis involves ligand-dependent activation of
phosphate group transfer. Proc Natl Acad Sci USA
1989;86:12081212.
40 Gegner JA, Graham DR, Roth AF, Dahlquist FW:
Assembly of an MCP receptor, CheW, and kinase
CheA complex in the bacterial chemotaxis signal
transduction pathway. Cell 1992;70:975982.
41 Simon MI, Borkovich KA, Bourret RB, Hess JF:
Protein phosphorylation in the bacterial chemotaxis
system. Biochimie 1989;71:10131019.

Chemotactic Sensory Transduction

42 Wolfe AJ, Conley MP, Kramer TJ, Berg HC:


Reconstitution of signaling in bacterial chemotaxis.
J Bacteriol 1987;169:18781885.
43 Garrity LF, Ordal GW: Activation of the CheA
kinase by asparagine in Bacillus subtilis chemotaxis.
Microbiology 1997;143:29452951.
44 Bischoff DS, Ordal GW: Identification and characterization of FliY, a novel component of the Bacillus
subtilis flagellar switch complex. Mol Microbiol
1992;6:27152723.
45 Szurmant H, Bunn MW, Cannistraro VJ, Ordal GW:
Bacillus subtilis hydrolyzes CheY-P at the location
of its action, the flagellar switch. J Biol Chem 2003.
46 Barak R, Eisenbach M: Correlation between phosphorylation of the chemotaxis protein CheY and its
activity at the flagellar motor. Biochemistry 1992;
31:18211826.
47 Hess JF, Bourret RB, Oosawa K, Matsumura P,
Simon MI: Protein phosphorylation and bacterial
chemotaxis. Cold Spring Harb Symp Quant Biol
1988;53:4148.
48 Lee GF, Lebert MR, Lilly AA, Hazelbauer GL:
Transmembrane signaling characterized in bacterial
chemoreceptors by using sulfhydryl cross-linking in
vivo. Proc Natl Acad Sci USA 1995;92:33913395.
49 Yeh JI, Biemann HP, Prive GG, Pandit J, Koshland
DE Jr, Kim SH: High-resolution structures of the
ligand binding domain of the wild-type bacterial
aspartate receptor. J Mol Biol 1996;262:186201.
50 Kondoh H, Ball CB, Adler J: Identification of a
methyl-accepting chemotaxis protein for the ribose
and galactose chemoreceptors of Escherichia coli.
Proc Natl Acad Sci USA 1979;76:260264.
51 Ames P, Parkinson JS: Constitutively signaling fragments of Tsr, the Escherichia coli serine chemoreceptor. J Bacteriol 1994;176:63406348.
52 Kim KK, Yokota H, Kim SH: Four-helical-bundle
structure of the cytoplasmic domain of a serine
chemotaxis receptor. Nature 1999;400:787792.
53 Starrett DJ, Falke JJ: Adaptation mechanism of the
aspartate receptor: electrostatics of the adaptation
subdomain play a key role in modulating kinase
activity. Biochemistry 2005;44:15501560.
54 Mehan RS, White NC, Falke JJ: Mapping out regions
on the surface of the aspartate receptor that are
essential for kinase activation. Biochemistry 2003;
42:29522959.
55 Hulko M, Berndt F, Gruber M, Linder JU, Truffault
V, Schultz A, Martin J, Schultz JE, Lupas AN, Coles
M: The HAMP domain structure implies helix rotation in transmembrane signaling. Cell 2006;126:929
940.
56 Alexander RP, Zhulin IB: Evolutionary genomics
reveals conserved structural determinants of signaling and adaptation in microbial chemoreceptors.
Proc Natl Acad Sci USA 2007;104:28852890.

55

57 Li M, Hazelbauer GL: Cellular stoichiometry of the


components of the chemotaxis signaling complex. J
Bacteriol 2004;186:36873694.
58 Fredrick KL, Helmann JD: Dual chemotaxis signaling pathways in Bacillus subtilis: a sigma
D-dependent gene encodes a novel protein with
both CheW and CheY homologous domains. J
Bacteriol 1994;176:27272735.
59 Karatan E, Saulmon MM, Bunn MW, Ordal GW:
Phosphorylation of the response regulator CheV is
required for adaptation to attractants during Bacillus
subtilis chemotaxis. J Biol Chem 2001;276:43618
43626.
60 Rosario MM, Fredrick KL, Ordal GW, Helmann JD:
Chemotaxis in Bacillus subtilis requires either of
two functionally redundant CheW homologs. J
Bacteriol 1994;176:27362739.
61 Park SY, Borbat PP, Gonzalez-Bonet G, Bhatnagar J,
Pollard AM, Freed JH, Bilwes AM, Crane BR:
Reconstruction of the chemotaxis receptor-kinase
assembly. Nat Struct Mol Biol 2006;13:400407.
62 Ames P, Parkinson JS: Conformational suppression
of inter-receptor signaling defects. Proc Natl Acad
Sci USA 2006;103:92929297.
63 Ames P, Studdert CA, Reiser RH, Parkinson JS:
Collaborative signaling by mixed chemoreceptor
teams in Escherichia coli. Proc Natl Acad Sci USA
2002;99:70607065.
64 Chervitz SA, Lin CM, Falke JJ: Transmembrane signaling by the aspartate receptor: engineered disulfides reveal static regions of the subunit interface.
Biochemistry 1995;34:97229733.
65 Thiem S, Kentner D, Sourjik V: Positioning of
chemosensory clusters in E. coli and its relation to
cell division. EMBO J 2007;26:16151623.
66 Kirby JR, Niewold TB, Maloy S, Ordal GW: CheB is
required for behavioural responses to negative stimuli during chemotaxis in Bacillus subtilis. Mol
Microbiol 2000;35:4457.
67 Zhang P, Khursigara CM, Hartnell LM, Subramaniam
S: Direct visualization of Escherichia coli chemotaxis
receptor arrays using cryo-electron microscopy. Proc
Natl Acad Sci USA 2007;104:37773781.
68 Zhang P, Weis RM, Peters PJ, Subramaniam S:
Electron tomography of bacterial chemotaxis receptor assemblies. Methods Cell Biol 2007;79:373384.
69 Shimizu TS, Le Novere N, Levin MD, Beavil AJ,
Sutton BJ, Bray D: Molecular model of a lattice of
signalling proteins involved in bacterial chemotaxis.
Nat Cell Biol 2000;2:792796.
70 Hou S, Freitas T, Larsen RW, Piatibratov M,
Sivozhelezov V, Yamamoto A, Meleshkevitch EA,
Zimmer M, Ordal GW, Alam M: Globin-coupled sensors: a class of heme-containing sensors in Archaea
and Bacteria. Proc Natl Acad Sci USA 2001;98:9353
9358.

56

71 Kristich CJ, Glekas GD, Ordal GW: The conserved


cytoplasmic module of the transmembrane chemoreceptor McpC mediates carbohydrate chemotaxis in
Bacillus subtilis. Mol Microbiol 2003;47:13531366.
72 Muller J, Schiel S, Ordal GW, Saxild HH: Functional
and genetic characterization of mcpC, which encodes
a third methyl-accepting chemotaxis protein in
Bacillus subtilis. Microbiology 1997;143:32313240.
73 Hanlon DW, Ordal GW: Cloning and characterization of genes encoding methyl-accepting chemotaxis proteins in Bacillus subtilis. J Biol Chem 1994;
269:1403814046.
74 Reader RW, Tso WW, Springer MS, Goy MF, Adler
J: Pleiotropic aspartate taxis and serine taxis mutants
of Escherichia coli. J Gen Microbiol 1979;111:363
374.
75 Hazelbauer GL, Harayama S: Mutants in transmission of chemotactic signals from two independent
receptors of E. coli. Cell 1979;16:617625.
76 Manson MD, Blank V, Brade G, Higgins CF: Peptide
chemotaxis in E. coli involves the Tap signal transducer and the dipeptide permease. Nature 1986;
321:253256.
77 Rebbapragada A, Johnson MS, Harding GP,
Zuccarelli AJ, Fletcher HM, Zhulin IB, Taylor BL:
The Aer protein and the serine chemoreceptor Tsr
independently sense intracellular energy levels and
transduce oxygen, redox, and energy signals for
Escherichia coli behavior. Proc Natl Acad Sci USA
1997;94:1054110546.
78 Lux R, Jahreis K, Bettenbrock K, Parkinson JS,
Lengeler JW: Coupling the phosphotransferase system and the methyl-accepting chemotaxis proteindependent chemotaxis signaling pathways of
Escherichia coli. Proc Natl Acad Sci USA 1995;92:
1158311587.
79 Greer-Phillips SE, Alexandre G, Taylor BL, Zhulin
IB: Aer and Tsr guide Escherichia coli in spatial gradients of oxidizable substrates. Microbiology 2003;
149:26612667.
80 Deutscher J, Francke C, Postma PW: How phosphotransferase system-related protein phosphorylation
regulates carbohydrate metabolism in bacteria.
Microbiol Mol Biol Rev 2006;70:9391031.
81 Ordal GW, Goldman DJ: Chemotactic repellents of
Bacillus subtilis. J Mol Biol 1976;100:103108.
82 Ordal GW, Goldman DJ: Chemotaxis away from
uncouplers of oxidative phosphorylation in Bacillus
subtilis. Science 1975;189:802805.
83 Ordal GW: Recognition sites for chemotactic repellents of Bacillus subtilis. J Bacteriol 1976;126:72
79.
84 Brummett TB, Ordal GW: Inhibition of amino acid
transport in Bacillus subtilis by uncouplers of oxidative phosphorylation. Arch Biochem Biophys
1977;178:368372.

Rao Ordal

85 Tso WW, Adler J: Negative chemotaxis in Escherichia


coli. J Bacteriol 1974;118:560576.
86 Ottemann KM, Xiao W, Shin YK, Koshland DE Jr:
A piston model for transmembrane signaling of the
aspartate receptor. Science 1999;285:17511754.
87 Chervitz SA, Falke JJ: Molecular mechanism of
transmembrane signaling by the aspartate receptor:
a model. Proc Natl Acad Sci USA 1996;93:2545
2550.
88 Borkovich KA, Simon MI: The dynamics of protein
phosphorylation in bacterial chemotaxis. Cell 1990;
63:13391348.
89 Mello BA, Tu Y: Effects of adaptation in maintaining high sensitivity over a wide range of backgrounds for Escherichia coli chemotaxis. Biophys J
2007;92:23292337.
90 Mello BA, Tu Y: An allosteric model for heterogeneous receptor complexes: understanding bacterial
chemotaxis responses to multiple stimuli. Proc Natl
Acad Sci USA 2005;102:1735417359.
91 Rao CV, Frenklach M, Arkin AP: An allosteric
model for transmembrane signaling in bacterial
chemotaxis. J Mol Biol 2004;343:291303.
92 Szurmant H, Bunn MW, Cho SH, Ordal GW:
Ligand-induced conformational changes in the
Bacillus subtilis chemoreceptor McpB determined
by disulfide crosslinking in vivo. J Mol Biol 2004;
344:919928.
93 Lamanna AC, Ordal GW, Kiessling LL: Large
increases in attractant concentration disrupt the
polar localization of bacterial chemoreceptors. Mol
Microbiol 2005;57:774785.
94 Bunn MW, Ordal GW: Receptor conformational
changes enhance methylesterase activity during
chemotaxis by Bacillus subtilis. Mol Microbiol 2004;
51:721728.
95 Borrok MJ, Kolonko EM, Kiessling LL: Chemical
probes of bacterial signal transduction reveal that
repellents stabilize and attractants destabilize the
chemoreceptor array. ACS Chem Biol 2008;3:101
109.
96 Borrok MJ, Kiessling LL, Forest KT: Conformational
changes of glucose/galactose-binding protein illuminated by open, unliganded, and ultra-high-resolution ligand-bound structures. Protein Sci 2007;
16:10321041.
97 Gestwicki JE, Kiessling LL: Inter-receptor communication through arrays of bacterial chemoreceptors. Nature 2002;415:8184.
98 Gestwicki JE, Strong LE, Borchardt SL, Cairo CW,
Schnoes AM, Kiessling LL: Designed potent multivalent chemoattractants for Escherichia coli. Bioorg
Med Chem 2001;9:23872393.
99 Bray D, Levin MD, Lipkow K: The chemotactic
behavior of computer-based surrogate bacteria.
Curr Biol 2007;17:1219.

Chemotactic Sensory Transduction

100 Shimizu TS, Aksenov SV, Bray D: A spatially


extended stochastic model of the bacterial chemotaxis signalling pathway. J Mol Biol 2003;329:291
309.
101 Shimizu TS, Bray D: Modelling the bacterial chemotaxis receptor complex. Novartis Found Symp
2002;247:162181, 198206, 244152.
102 Bray D: Bacterial chemotaxis and the question of
gain. Proc Natl Acad Sci USA 2002;99:79.
103 Bray D, Levin MD, Morton-Firth CJ: Receptor clustering as a cellular mechanism to control sensitivity.
Nature 1998;393:8588.
104 Segall JE, Block SM, Berg HC: Temporal comparisons in bacterial chemotaxis. Proc Natl Acad Sci
USA 1986;83:89878991.
105 Sourjik V, Berg HC: Functional interactions between
receptors in bacterial chemotaxis. Nature 2004;428:
437441.
106 Sourjik V, Berg HC: Receptor sensitivity in bacterial
chemotaxis. Proc Natl Acad Sci USA 2002;99:123
127.
107 Asinas AE, Weis RM: Competitive and cooperative
interactions in receptor signaling complexes. J Biol
Chem 2006;281:3051230523.
108 Li G, Weis RM: Covalent modification regulates
ligand binding to receptor complexes in the chemosensory system of Escherichia coli. Cell 2000;100:357
365.
109 Mello BA, Shaw L, Tu Y: Effects of receptor interaction in bacterial chemotaxis. Biophys J 2004;87:1578
1595.
110 Mello BA, Tu Y: Quantitative modeling of sensitivity
in bacterial chemotaxis: the role of coupling among
different chemoreceptor species. Proc Natl Acad Sci
USA 2003;100:82238228.
111 Endres RG, Falke JJ, Wingreen NS: Chemotaxis
receptor complexes: from signaling to assembly.
PLoS Comput Biol 2007;3:e150.
112 Keymer JE, Endres RG, Skoge M, Meir Y, Wingreen
NS: Chemosensing in Escherichia coli: two regimes
of two-state receptors. Proc Natl Acad Sci USA
2006;103:17861791.
113 Albert R, Chiu YW, Othmer HG: Dynamic receptor
team formation can explain the high signal transduction gain in Escherichia coli. Biophys J 2004;86:
26502659.
114 Homma M, Shiomi D, Kawagishi I: Attractant binding alters arrangement of chemoreceptor dimers
within its cluster at a cell pole. Proc Natl Acad Sci
USA 2004;101:34623467.
115 Ordal GW, Gibson KJ: Chemotaxis toward amino
acids by Bacillus subtilis. J Bacteriol 1977;129:151
155.
116 Ordal GW, Villani DP, Gibson KJ: Amino acid
chemoreceptors of Bacillus subtilis. J Bacteriol
1977;129:156165.

57

117 Zimmer MA, Szurmant H, Saulmon MM, Collins


MA, Bant JS, Ordal GW: The role of heterologous
receptors in McpB-mediated signalling in Bacillus
subtilis chemotaxis. Mol Microbiol 2002;45:555
568.
118 Zimmer MA, Tiu J, Collins MA, Ordal GW:
Selective methylation changes on the Bacillus subtilis chemotaxis receptor McpB promote adaptation. J
Biol Chem 2000;275:2426424272.
119 McNamara BP, Wolfe AJ: Coexpression of the long
and short forms of CheA, the chemotaxis histidine
kinase, by members of the family Enterobacteriaceae.
J Bacteriol 1997;179:18131818.
120 Vaknin A, Berg HC: Single-cell FRET imaging of
phosphatase activity in the Escherichia coli chemotaxis system. Proc Natl Acad Sci USA 2004;101:17072
17077.
121 Muff TJ, Ordal GW: The CheC phosphatase regulates chemotactic adaptation through CheD. J Biol
Chem 2007;282:3412034128.
122 Szurmant H, Muff TJ, Ordal GW: Bacillus subtilis
CheC and FliY are members of a novel class of
CheY-P hydrolyzing proteins in the chemotactic
signal transduction cascade. J Biol Chem 2004;
279:2178721792.
123 Saulmon MM, Karatan E, Ordal, GW: Effect of loss
of CheC and other adaptational proteins on chemotactic behavior in Bacillus subtilis. Microbiology
2004;150:581589.
124 Kort EN, Goy MF, Larsen SH, Adler J: Methylation
of a membrane protein involved in bacterial
chemotaxis. Proc Natl Acad Sci USA 1975;72:3939
3943.
125 Kleene SJ, Toews ML, Adler J: Isolation of glutamic
acid methyl ester from an Escherichia coli membrane protein involved in chemotaxis. J Biol Chem
1977;252:32143218.
126 Springer WR, Koshland DE Jr: Identification of a
protein methyltransferase as the cheR gene product
in the bacterial sensing system. Proc Natl Acad Sci
USA 1977;74:533537.
127 Lai WC, Beel BD, Hazelbauer GL: Adaptational
modification and ligand occupancy have opposite
effects on positioning of the transmembrane signalling helix of a chemoreceptor. Mol Microbiol 2006;
61:10811090.
128 Toews ML, Goy MF, Springer MS, Adler J:
Attractants and repellents control demethylation of
methylated chemotaxis proteins in Escherichia coli.
Proc Natl Acad Sci USA 1979;76:55445548.
129 Stock JB, Koshland DE Jr: A protein methylesterase
involved in bacterial sensing. Proc Natl Acad Sci
USA 1978;75:36593663.

58

130 Lupas A, Stock J: Phosphorylation of an N-terminal


regulatory domain activates the CheB methylesterase in bacterial chemotaxis. J Biol Chem 1989;
264:1733717342.
131 Alon U, Surette MG, Barkai N, Leibler S: Robustness
in bacterial chemotaxis. Nature 1999;397:168171.
132 Morton-Firth CJ, Shimizu TS, Bray D: A freeenergy-based stochastic simulation of the Tar receptor complex. J Mol Biol 1999;286:10591074.
133 Yi TM, Huang Y, Simon MI, Doyle J: Robust perfect
adaptation in bacterial chemotaxis through integral
feedback control. Proc Natl Acad Sci USA 2000;97:
46494653.
134 Kristich CJ, Ordal GW: Bacillus subtilis CheD Is a
Chemoreceptor Modification Enzyme Required for
Chemotaxis. J Biol Chem 2002;277:2535625362.
135 Rosario MM, Ordal GW: CheC and CheD interact
to regulate methylation of Bacillus subtilis methylaccepting chemotaxis proteins. Mol Microbiol
1996;21:511518.
136 Rosario MM, Kirby JR, Bochar DA, Ordal GW:
Chemotactic methylation and behavior in Bacillus
subtilis: role of two unique proteins, CheC and
CheD. Biochemistry 1995;34:38233831.
137 Chao X, Muff TJ, Park SY, Zhang S, Pollard AM,
Ordal GW, Bilwes AM, Crane BR: A receptor-modifying deamidase in complex with a signaling
phosphatase reveals reciprocal regulation. Cell
2006;124:561571.
138 Kristich CJ, Ordal GW: Analysis of chimeric
chemoreceptors in Bacillus subtilis reveals a role for
CheD in the function of the McpC HAMP domain.
J Bacteriol 2004;186:59505955.
139 Kirby JR, Kristich CJ, Saulmon MM, Zimmer MA,
Garrity LF, Zhulin IB, Ordal GW: CheC is related to
the family of flagellar switch proteins and acts independently from CheD to control chemotaxis in
Bacillus subtilis. Mol Microbiol 2001;42:573585.
140 Kirsch ML, Zuberi AR, Henner D, Peters PD, Yazdi
MA, Ordal GW: Chemotactic methyltransferase
promotes adaptation to repellents in Bacillus subtilis. J Biol Chem 1993;268:2535025356.
141 Kirsch ML, Peters PD, Hanlon DW, Kirby JR, Ordal
GW: Chemotactic methylesterase promotes adaptation to high concentrations of attractant in Bacillus
subtilis. J Biol Chem 1993;268:1861018616.
142 Kirby JR, Saulmon MM, Kristich CJ, Ordal GW:
CheY-dependent methylation of the asparagine
receptor, McpB, during chemotaxis in Bacillus subtilis. J Biol Chem 1999;274:1109211100.
143 Terwilliger TC, Wang JY, Koshland DE Jr: Kinetics
of receptor modification. The multiply methylated
aspartate receptors involved in bacterial chemotaxis. J Biol Chem 1986;261:1081410820.

Rao Ordal

144 Ordal GW, Parker HM, Kirby JR: Complementation


and characterization of chemotaxis mutants of
Bacillus subtilis. J Bacteriol 1985;164:802810.
145 Frye J, Karlinsey JE, Felise HR, Marzolf B, Dowidar
N, McClelland M, Hughes KT: Identification of new
flagellar genes of Salmonella enterica serovar
Typhimurium. J Bacteriol 2006;188:22332243.
146 Szurmant H, Ordal GW: Diversity in chemotaxis
mechanisms among the bacteria and archaea.
Microbiol Mol Biol Rev 2004;68:301319.
147 Bilwes AM, Alex LA, Crane BR, Simon MI: Structure
of CheA, a signal-transducing histidine kinase. Cell
1999;96:131141.
148 Borkovich KA, Alex LA, Simon MI: Attenuation of
sensory receptor signaling by covalent modification. Proc Natl Acad Sci USA 1992;89:67566760.
149 Fuhrer DK, Ordal GW: Bacillus subtilis CheN, a
homolog of CheA, the central regulator of chemotaxis in Escherichia coli. J Bacteriol 1991;173:7443
7448.
150 Hess JF, Oosawa K, Kaplan N, Simon MI:
Phosphorylation of three proteins in the signaling
pathway of bacterial chemotaxis. Cell 1988;53:7987.
151 Lux R, Sim JH, Tsai JP, Shi W: Construction and
characterization of a cheA mutant of Treponema
denticola. J Bacteriol 2002;184:31303134.
152 Mourey L, Da Re S, Pedelacq JD, Tolstykh T, Faurie
C, Guillet V, Stock JB, Samama JP: Crystal structure
of the CheA histidine phosphotransfer domain that
mediates response regulator phosphorylation in
bacterial chemotaxis. J Biol Chem 2001;276:31074
31082.
153 Rudolph J, Tolliday N, Schmitt C, Schuster SC,
Oesterhelt D: Phosphorylation in halobacterial signal transduction. EMBO J 1995;14:42494257.
154 Stewart RC: Activating and inhibitory mutations in
the regulatory domain of CheB, the methylesterase
in bacterial chemotaxis. J Biol Chem 1993;268:1921
1930.
155 Swanson RV, Schuster SC, Simon MI: Expression of
CheA fragments which define domains encoding
kinase, phosphotransfer, and CheY binding activities. Biochemistry 1993;32:76237629.
156 Miller AS, Kohout SC, Gilman KA, Falke JJ: CheA
Kinase of bacterial chemotaxis: chemical mapping
of four essential docking sites. Biochemistry 2006;
45:86998711.
157 Jahreis K, Morrison TB, Garzon A, Parkinson JS:
Chemotactic signaling by an Escherichia coli CheA
mutant that lacks the binding domain for phosphoacceptor partners. J Bacteriol 2004;186:26642672.
158 Park SY, Quezada CM, Bilwes AM, Crane BR:
Subunit exchange by CheA histidine kinases from
the mesophile Escherichia coli and the thermophile
Thermotoga maritima. Biochemistry 2004;43:2228
2240.

Chemotactic Sensory Transduction

159 Kott L, Braswell EH, Shrout AL, Weis RM:


Distributed subunit interactions in CheA contribute
to dimer stability: a sedimentation equilibrium
study. Biochim Biophys Acta 2004;1696:131140.
160 Martin AC, Nair U, Armitage JP, Maddock JR: Polar
localization of CheA2 in Rhodobacter sphaeroides
requires specific Che homologs. J Bacteriol 2003;
185:46674671.
161 Anand GS, Goudreau PN, Stock AM: Activation of
methylesterase CheB: evidence of a dual role for the
regulatory domain. Biochemistry 1998;37:14038
14047.
162 Djordjevic S, Goudreau PN, Xu Q, Stock AM, West
AH: Structural basis for methylesterase CheB regulation by a phosphorylation-activated domain. Proc
Natl Acad Sci USA 1998;95:13811386.
163 Goldman DJ, Nettleton DO, Ordal GW: Purification
and characterization of chemotactic methylesterase
from Bacillus subtilis. Biochemistry 1984;23:675
680.
164 Goldman DJ, Ordal GW: In vitro methylation and
demethylation of methyl-accepting chemotaxis proteins in Bacillus subtilis. Biochemistry 1984;23:2600
2606.
165 Kehry MR, Bond MW, Hunkapiller MW, Dahlquist
FW: Enzymatic deamidation of methyl-accepting
chemotaxis proteins in Escherichia coli catalyzed by
the cheB gene product. Proc Natl Acad Sci USA
1983;80:35993603.
166 Kehry MR, Dahlquist FW: Adaptation in bacterial
chemotaxis: CheB-dependent modification permits
additional methylations of sensory transducer proteins. Cell 1982;29:761772.
167 Kehry MR, Doak TG, Dahlquist FW: Stimulusinduced changes in methylesterase activity during
chemotaxis in Escherichia coli. J Biol Chem 1984;
259:1182811835.
168 Kim C, Jackson M, Lux R, Khan S: Determinants of
chemotactic signal amplification in Escherichia coli.
J Mol Biol 2001;307:119135.
169 Li J, Swanson RV, Simon MI, Weis RM: The response
regulators CheB and CheY exhibit competitive
binding to the kinase CheA. Biochemistry 1995;34:
1462614636.
170 Martin AC, Wadhams GH, Shah DS, Porter SL,
Mantotta JC, Craig TJ, Verdult PH, Jones H,
Armitage JP: CheR- and CheB-dependent chemosensory adaptation system of Rhodobacter sphaeroides.
J Bacteriol 2001;183:71357144.
171 Nettleton DO, Ordal GW: Functional homology of
chemotactic methylesterases from Bacillus subtilis
and Escherichia coli. J Bacteriol 1989;171:120123.
172 Stewart RC, Dahlquist FW: N-terminal half of CheB
is involved in methylesterase response to negative
chemotactic stimuli in Escherichia coli. J Bacteriol
1988;170:57285738.

59

173 Stewart RC, Roth AF, Dahlquist FW: Mutations that


affect control of the methylesterase activity of CheB,
a component of the chemotaxis adaptation system
in Escherichia coli. J Bacteriol 1990;172:33883399.
174 Muff TJ, Foster RM, Liu PJ, Ordal GW: CheX in the
three-phosphatase system of bacterial chemotaxis. J
Bacteriol 2007;189:70077013.
175 Muff TJ, Ordal GW: Assays for CheC, FliY, and
CheX as representatives of response regulator phosphatases. Methods Enzymol 2007;423:336348.
176 Park SY, Lowder B, Bilwes AM, Blair DF, Crane BR:
Structure of FliM provides insight into assembly of
the switch complex in the bacterial flagella motor.
Proc Natl Acad Sci USA 2006;103:1188611891.
177 Motaleb MA, Miller MR, Li C, Bakker RG, Goldstein
SF, Silversmith RE, Bourret RB, Charon NW: CheX
is a phosphorylated CheY phosphatase essential for
Borrelia burgdorferi chemotaxis. J Bacteriol 2005;
187:79637969.
178 Saulmon MM, Karatan E, Ordal GW: Effect of loss
of CheC and other adaptational proteins on chemotactic behaviour in Bacillus subtilis. Microbiology
2004;150:581589.
179 Park SY, Chao X, Gonzalez-Bonet G, Beel BD,
Bilwes AM, Crane BR: Structure and function of an
unusual family of protein phosphatases: the bacterial chemotaxis proteins CheC and CheX. Mol Cell
2004;16:563574.
180 Kirby JR, Kristich CJ, Feinberg SL, Ordal GW:
Methanol production during chemotaxis to amino
acids in Bacillus subtilis. Mol Microbiol 1997;24:
869878.
181 Perez E, Stock AM: Characterization of the
Thermotoga maritima chemotaxis methylation system that lacks pentapeptide-dependent methyltransferase CheR:MCP tethering. Mol Microbiol
2007;63:363378.
182 Boldog T, Li M, Hazelbauer GL: Using Nanodiscs to
create water-soluble transmembrane chemoreceptors inserted in lipid bilayers. Methods Enzymol
2007;423:317335.
183 Mello BA, Tu Y: Effects of adaptation in maintaining high sensitivity over a wide range of backgrounds for Escherichia coli chemotaxis. Biophys J
2007;92:23292337.
184 Perez E, Zheng H, Stock AM: Identification of
methylation sites in Thermotoga maritima chemotaxis receptors. J Bacteriol 2006;188:40934100.
185 Li M, Hazelbauer GL: Adaptational assistance in
clusters of bacterial chemoreceptors. Mol Microbiol
2005;56:16171626.
186 Liberman L, Berg HC, Sourjik V: Effect of chemoreceptor modification on assembly and activity of the
receptor-kinase complex in Escherichia coli. J
Bacteriol 2004;186:66436646.

60

187 Levit MN, Stock JB: Receptor methylation controls


the magnitude of stimulus-response coupling in
bacterial chemotaxis. J Biol Chem 2002;277:36760
36765.
188 Levin MD, Shimizu TS, Bray D: Binding and diffusion of CheR molecules within a cluster of membrane receptors. Biophys J 2002;82:18091817.
189 Perazzona B, Spudich JL: Identification of methylation sites and effects of phototaxis stimuli on transducer methylation in Halobacterium salinarum. J
Bacteriol 1999;181:56765683.
190 Okumura H, Nishiyama S, Sasaki A, Homma M,
Kawagishi I: Chemotactic adaptation is altered by
changes in the carboxy-terminal sequence conserved among the major methyl-accepting chemoreceptors. J Bacteriol 1998;180:18621868.
191 Djordjevic S, Stock AM: Chemotaxis receptor recognition by protein methyltransferase CheR. Nat
Struct Biol 1998;5:446450.
192 Djordjevic S, Stock AM: Crystal structure of the
chemotaxis receptor methyltransferase CheR suggests a conserved structural motif for binding
S-adenosylmethionine. Structure 1997;5:545558.
193 Hanlon DW, Ying C, Ordal GW: Purification and
reconstitution of the methyl-accepting chemotaxis
proteins from Bacillus subtilis. Biochim Biophys
Acta 1993;1158:345351.
194 Weis RM, Koshland DE Jr: Reversible receptor
methylation is essential for normal chemotaxis of
Escherichia coli in gradients of aspartic acid. Proc
Natl Acad Sci USA 1988;85:8387.
195 Sanders DA, Koshland DE Jr: Receptor interactions
through phosphorylation and methylation pathways in bacterial chemotaxis. Proc Natl Acad Sci
USA 1988;85:84258429.
196 Nowlin DM, Bollinger J, Hazelbauer GL: Sitedirected mutations altering methyl-accepting residues of a sensory transducer protein. Proteins 1988;
3:102112.
197 Burgess-Cassler A, Ordal GW: Functional homology of Bacillus subtilis methyltransferase II and
Escherichia coli cheR protein. J Biol Chem 1982;257:
1283512838.
198 Springer MS, Zanolari B, Pierzchala PA: Ordered
methylation of the methyl-accepting chemotaxis
proteins of Escherichia coli. J Biol Chem 1982;257:
68616866.
199 Ullah AH, Ordal GW: In vivo and in vitro chemotactic methylation in Bacillus subtilis. J Bacteriol
1981;145:958965.
200 Stock JB, Maderis AM, Koshland DE Jr: Bacterial
chemotaxis in the absence of receptor carboxylmethylation. Cell 1981;27:3744.
201 Chelsky D, Dahlquist FW: Multiple sites of methylation in the methyl accepting chemotaxis proteins of
Escherichia coli. Prog Clin Biol Res 1981;63:371381.

Rao Ordal

202 Rollins CM, Dahlquist FW: Methylation of chemotaxis-specific proteins in Escherichia coli cells permeable to S-adenosylmethionine. Biochemistry
1980;19:46274632.
203 Kleene SJ, Hobson AC, Adler J: Attractants and
repellents influence methylation and demethylation
of methyl-accepting chemotaxis proteins in an
extract of Escherichia coli. Proc Natl Acad Sci USA
1979;76:63096313.
204 Springer MS, Goy MF, Adler J: Protein methylation
in behavioural control mechanisms and in signal
transduction. Nature 1979;280:279284.
205 Goy MF, Springer MS, Adler J: Failure of sensory
adaptation in bacterial mutants that are defective in
a protein methylation reaction. Cell 1978;15:1231
1240.
206 Van Der Werf P, Koshland DE Jr: Identification of a
gamma-glutamyl methyl ester in bacterial membrane protein involved in chemotaxis. J Biol Chem
1977;252:27932795.
207 Goy MF, Springer MS, Adler J: Sensory transduction in Escherichia coli: role of a protein methylation reaction in sensory adaptation. Proc Natl Acad
Sci USA 1977;74:49644968.
208 Rosu V, Chevance FF, Karlinsey JE, Hirano T,
Hughes KT: Translation inhibition of the Salmonella
fliC gene by the fliC 5 untranslated region, fliC
coding sequences, and FlgM. J Bacteriol 2006;188:
44974507.
209 Montefusco DJ, Shrout AL, Besschetnova TY, Weis
RM: Formation and Activity of Template-Assembled
Receptor Signaling Complexes. Langmuir 2007;23:
32803289.
210 Zhao J, Parkinson JS: Mutational analysis of the
chemoreceptor-coupling domain of the Escherichia
coli chemotaxis signaling kinase CheA. J Bacteriol
2006;188:32993307.
211 Wolanin PM, Baker MD, Francis NR, Thomas DR,
DeRosier DJ, Stock JB: Self-assembly of receptor/
signaling complexes in bacterial chemotaxis. Proc
Natl Acad Sci USA 2006;103:1431314318.
212 Hamel DJ, Dahlquist FW: The contact interface of a
120 kD CheA-CheW complex by methyl TROSY
interaction spectroscopy. J Am Chem Soc 2005;
127:96769677.
213 Bunn MW, Ordal GW: Transmembrane organization of the Bacillus subtilis chemoreceptor McpB
deduced by cysteine disulfide crosslinking. J Mol
Biol 2003;331:941949.
214 Shrout AL, Montefusco DJ, Weis RM: Templatedirected assembly of receptor signaling complexes.
Biochemistry 2003;42:1337913385.

Chemotactic Sensory Transduction

215 Alexandre G, Zhulin IB: Different evolutionary


constraints on chemotaxis proteins CheW and
CheY revealed by heterologous expression studies
and protein sequence analysis. J Bacteriol 2003;
185:544552.
216 Griswold IJ, Zhou H, Matison M, Swanson RV,
McIntosh LP, Simon MI, Dahlquist FW: The solution structure and interactions of CheW from
Thermotoga maritima. Nat Struct Biol 2002;9:121
125.
217 Francis NR, Levit MN, Shaikh TR, Melanson LA,
Stock JB, DeRosier DJ: Subunit organization in a
soluble complex of tar, CheW, and CheA by electron
microscopy. J Biol Chem 2002;277:3675536759.
218 Boukhvalova MS, Dahlquist FW, Stewart RC: CheW
binding interactions with CheA and Tar. Importance
for chemotaxis signaling in Escherichia coli. J Biol
Chem 2002;277:2225122259.
219 Griswold IJ, Dahlquist FW: The dynamic behavior
of CheW from Thermotoga maritima in solution, as
determined by nuclear magnetic resonance: implications for potential protein-protein interaction
sites. Biophys Chem 2002;101102:359373.
220 Francis NR, Levit MN, Shaikh TR, Melanson LA,
Stock JB, DeRosier DJ: Subunit organization in a
soluble complex of tar, CheW, and CheA by electron
microscopy. J Biol Chem 2002;277:3675536759.
221 Martin AC, Wadhams GH, Armitage JP: The roles
of the multiple CheW and CheA homologues in
chemotaxis and in chemoreceptor localization in
Rhodobacter sphaeroides. Mol Microbiol 2001;40:
12611272.
222 Skidmore JM, Ellefson DD, McNamara BP, Couto
MM, Wolfe AJ, Maddock JR: Polar clustering of the
chemoreceptor complex in Escherichia coli occurs
in the absence of complete CheA function. J
Bacteriol 2000;182:967973.
223 Hamblin PA, Bourne NA, Armitage JP:
Characterization of the chemotaxis protein CheW
from Rhodobacter sphaeroides and its effect on the
behaviour of Escherichia coli. Mol Microbiol
1997;24:4151.
224 Rowsell EH, Smith JM, Wolfe A, Taylor BL: CheA,
CheW, and CheY are required for chemotaxis to
oxygen and sugars of the phosphotransferase system in Escherichia coli. J Bacteriol 1995;177:6011
6014.
225 Hanlon DW, Marquez-Magana LM, Carpenter PB,
Chamberlin MJ, Ordal GW: Sequence and characterization of Bacillus subtilis CheW. J Biol Chem
1992;267:1205512060.
226 Gegner JA, Dahlquist FW: Signal transduction in
bacteria: CheW forms a reversible complex with the
protein kinase CheA. Proc Natl Acad Sci USA 1991;
88:750754.

61

227 McNally DF, Matsumura P: Bacterial chemotaxis


signaling complexes: formation of a CheA/CheW
complex enhances autophosphorylation and affinity for CheY. Proc Natl Acad Sci USA 1991;88:6269
6273.
228 Sanders DA, Mendez B, Koshland DE Jr: Role of the
CheW protein in bacterial chemotaxis: overexpression is equivalent to absence. J Bacteriol 1989;
171:62716278.
229 Conley MP, Wolfe AJ, Blair DF, Berg HC: Both
CheA and CheW are required for reconstitution of
chemotactic signaling in Escherichia coli. J Bacteriol
1989;171:51905193.
230 Liu JD, Parkinson JS: Role of CheW protein in coupling membrane receptors to the intracellular signaling system of bacterial chemotaxis. Proc Natl
Acad Sci USA 1989;86:87038707.
231 Chelsky D, Dahlquist FW: Chemotaxis in
Escherichia coli: associations of protein components. Biochemistry, 1980;19:46334639.
232 Yan J, Barak R, Liarzi O, Shainskaya A, Eisenbach
M: In vivo acetylation of CheY, a response regulator
in chemotaxis of Escherichia coli. J Mol Biol
2008;376:12601271.
233 Guhaniyogi J, Robinson VL, Stock AM: Crystal
structures of beryllium fluoride-free and beryllium
fluoride-bound CheY in complex with the conserved C-terminal peptide of CheZ reveal dual
binding modes specific to CheY conformation. J
Mol Biol 2006;359:624645.
234 Barak R, Yan J, Shainskaya A, Eisenbach M: The
chemotaxis response regulator CheY can catalyze its
own acetylation. J Mol Biol 2006;359:251265.
235 Barak R, Eisenbach M: Co-regulation of acetylation
and phosphorylation of CheY, a response regulator
in chemotaxis of Escherichia coli. J Mol Biol
2004;342:375381.
236 Barak R, Prasad K, Shainskaya A, Wolfe AJ,
Eisenbach M: Acetylation of the chemotaxis response
regulator CheY by acetyl-CoA synthetase purified
from Escherichia coli. J Mol Biol 2004;342:383401.
237 Ferre A, De La Mora J, Ballado T, Camarena L,
Dreyfus G: Biochemical study of multiple CheY
response regulators of the chemotactic pathway of
Rhodobacter sphaeroides. J Bacteriol 2004;186:5172
5177.
238 Sagi Y, Khan S, Eisenbach M: Binding of the chemotaxis response regulator CheY to the isolated, intact
switch complex of the bacterial flagellar motor: lack
of cooperativity. J Biol Chem 2003;278:25867
25871.
239 Wolanin PM, Webre DJ, Stock JB: Mechanism of
phosphatase activity in the chemotaxis response
regulator CheY. Biochemistry 2003;42:14075
14082.

62

240 Porter SL, Armitage JP: Phosphotransfer in


Rhodobacter sphaeroides chemotaxis. J Mol Biol
2002;324:3545.
241 Sourjik V, Berg HC: Binding of the Escherichia coli
response regulator CheY to its target measured in
vivo by fluorescence resonance energy transfer. Proc
Natl Acad Sci USA 2002;99:1266912674.
242 Schuster M, Silversmith RE, Bourret RB:
Conformational coupling in the chemotaxis response
regulator CheY. Proc Natl Acad Sci USA 2001;98:
60036008.
243 Deutschman WA, Dahlquist FW: Thermodynamic
basis for the increased thermostability of CheY
from the hyperthermophile Thermotoga maritima.
Biochemistry 2001;40:1310713113.
244 Lee SY, Cho HS, Pelton JG, Yan D, Berry EA,
Wemmer DE: Crystal structure of activated CheY.
Comparison with other activated receiver domains.
J Biol Chem 2001;276:1642516431.
245 Barak R, Eisenbach M: Acetylation of the response
regulator, CheY, is involved in bacterial chemotaxis.
Mol Microbiol 2001;40:731743.
246 Stewart RC, Jahreis K, Parkinson JS: Rapid phosphotransfer to CheY from a CheA protein lacking
the CheY-binding domain. Biochemistry 2000;39:
1315713165.
247 Cho HS, Lee SY, Yan D, Pan X, Parkinson JS, Kustu
S, Wemmer DE, Pelton JG: NMR structure of activated CheY. J Mol Biol 2000;297:543551.
248 Khan S, Pierce D, Vale RD: Interactions of the
chemotaxis signal protein CheY with bacterial flagellar motors visualized by evanescent wave microscopy. Curr Biol 2000;10:927930.
249 Da Re SS, Deville-Bonne D, Tolstykh T, V ron M,
Stock JB: Kinetics of CheY phosphorylation by
small molecule phosphodonors. FEBS Lett 1999;
457:323326.
250 Appleby JL, Bourret RB: Activation of CheY mutant
D57N by phosphorylation at an alternative site, Ser56. Mol Microbiol 1999;34:915925.
251 McEvoy MM, Bren A, Eisenbach M, Dahlquist FW:
Identification of the binding interfaces on CheY for
two of its targets, the phosphatase CheZ and the flagellar switch protein fliM. J Mol Biol 1999;289:1423
1433.
252 Usher KC, de la Cruz AF, Dahlquist FW, Swanson
RV, Simon MI, Remington SJ: Crystal structures of
CheY from Thermotoga maritima do not support
conventional explanations for the structural basis of
enhanced thermostability. Protein Sci 1998;7:403
412.
253 Appleby JL, Bourret RB: Proposed signal transduction role for conserved CheY residue Thr87, a member of the response regulator active-site quintet. J
Bacteriol 1998;180:35633569.

Rao Ordal

254 Schuster M, Abouhamad WN, Silversmith RE,


Bourret RB: Chemotactic response regulator mutant
CheY95IV exhibits enhanced binding to the flagellar
switch and phosphorylation-dependent constitutive
signalling. Mol Microbiol 1998;27:10651075.
255 Montrone M, Eisenbach M, Oesterhelt D, Marwan
W: Regulation of switching frequency and bias of
the bacterial flagellar motor by CheY and fumarate.
J Bacteriol 1998;180:33753380.
256 Shukla D, Zhu XY, Matsumura P: Flagellar motorswitch binding face of CheY and the biochemical
basis of suppression by CheY mutants that compensate for motor-switch defects in Escherichia coli. J
Biol Chem 1998;273:2399323999.
257 Sourjik V, Schmitt R: Phosphotransfer between
CheA, CheY1, and CheY2 in the chemotaxis signal
transduction chain of Rhizobium meliloti.
Biochemistry 1998;37:23272335.
258 Silversmith RE, Appleby JL, Bourret RB: Catalytic
mechanism of phosphorylation and dephosphorylation of CheY: kinetic characterization of imidazole
phosphates as phosphodonors and the role of acid
catalysis. Biochemistry 1997;36:1496514974.
259 Zhu X, Volz K, Matsumura P: The CheZ-binding
surface of CheY overlaps the CheA- and FliMbinding surfaces. J Biol Chem 1997;272:23758
23764.
260 Zhu X, Rebello J, Matsumura P, Volz K: Crystal
structures of CheY mutants Y106W and T87I/
Y106W. CheY activation correlates with movement
of residue 106. J Biol Chem 1997;272:50005006.
261 Zhou H, McEvoy MM, Lowry DF, Swanson RV,
Simon MI, Dahlquist FW: Phosphotransfer and
CheY-binding domains of the histidine autokinase
CheA are joined by a flexible linker. Biochemistry
1996;35:433443.
262 Sourjik V, Schmitt R: Different roles of CheY1 and
CheY2 in the chemotaxis of Rhizobium meliloti.
Mol Microbiol 1996;22:427436.
263 Sanna MG, Swanson RV, Bourret RB, Simon MI:
Mutations in the chemotactic response regulator,
CheY, that confer resistance to the phosphatase
activity of CheZ. Mol Microbiol 1995;15:1069
1079.
264 Shukla D, Matsumura P: Mutations leading to
altered CheA binding cluster on a face of CheY. J
Biol Chem 1995;270:2441424419.
265 Bischoff DS, Bourret RB, Kirsch ML, Ordal GW:
Purification and characterization of Bacillus subtilis
CheY. Biochemistry 1993;32:92569261.
266 Stock AM, Martinez-Hackert E, Rasmussen BF,
West AH, Stock JB, Ringe D, Petsko GA: Structure
of the Mg(2+)-bound form of CheY and mechanism of phosphoryl transfer in bacterial chemotaxis.
Biochemistry 1993;32:1337513380.

Chemotactic Sensory Transduction

267 Bourret RB, Drake SK, Chervitz SA, Simon MI, Falke
JJ: Activation of the phosphosignaling protein CheY.
II. Analysis of activated mutants by 19F NMR and
protein engineering. J Biol Chem 1993;268:13089
13096.
268 Drake SK, Bourret RB, Luck LA, Simon MI, Falke JJ:
Activation of the phosphosignaling protein CheY. I.
Analysis of the phosphorylated conformation by
19F NMR and protein engineering. J Biol Chem
1993;268:1308113088.
269 Kar L, Matsumura P, Johnson ME: Bivalent-metal
binding to CheY protein. Effect on protein conformation. Biochem J 1992;287:521531.
270 Roman SJ, Meyers M, Volz K, Matsumura P: A
chemotactic signaling surface on CheY defined by
suppressors of flagellar switch mutations. J Bacteriol
1992;174:62476255.
271 Volz K, Matsumura P: Crystal structure of
Escherichia coli CheY refined at 1.7-A resolution. J
Biol Chem 1991;266:1551115519.
272 Lukat GS, Stock AM, Stock JB: Divalent metal ion
binding to the CheY protein and its significance to
phosphotransfer in bacterial chemotaxis. Biochemistry 1990;29:54365442.
273 Lipkow K: Changing cellular location of CheZ predicted by molecular simulations. PLoS Comput Biol
2006;2:e39.
274 Cantwell BJ, Draheim RR, Weart RB, Nguyen C,
Stewart RC, Manson MD: CheZ phosphatase localizes to chemoreceptor patches via CheA-short. J
Bacteriol 2003;185:23542361.
275 Zhao R, Collins EJ, Bourret RB, Silversmith RE:
Structure and catalytic mechanism of the E. coli
chemotaxis phosphatase CheZ. Nat Struct Biol
2002;9:570575.
276 Boesch KC, Silversmith RE, Bourret RB: Isolation
and characterization of nonchemotactic CheZ
mutants of Escherichia coli. J Bacteriol 2000;182:
35443552.
277 Scharf BE, Fahrner KA, Berg HC: CheZ has no effect
on flagellar motors activated by CheY13DK106YW.
J Bacteriol 1998;180:51235128.
278 Sanna MG, Simon MI: Isolation and in vitro characterization of CheZ suppressors for the Escherichia
coli chemotactic response regulator mutant
CheYN23D. J Biol Chem 1996;271:73577361.
279 Bren A, Welch M, Blat Y, Eisenbach M: Signal termination in bacterial chemotaxis: CheZ mediates
dephosphorylation of free rather than switch-bound
CheY. Proc Natl Acad Sci USA 1996;93:10090
10093.
280 Blat Y, Eisenbach M: Conserved C-terminus of the
phosphatase CheZ is a binding domain for the
chemotactic response regulator CheY. Biochemistry
1996;35:56795683.

63

281 Blat Y, Eisenbach M: Oligomerization of the phosphatase CheZ upon interaction with the phosphorylated form of CheY. The signal protein of bacterial
chemotaxis. J Biol Chem 1996;271:12261231.
282 Wang H, Matsumura P: Characterization of the
CheAS/CheZ complex: a specific interaction resulting in enhanced dephosphorylating activity on
CheY-phosphate. Mol Microbiol 1996;19:695703.
283 Huang C, Stewart RC: CheZ mutants with enhanced
ability to dephosphorylate CheY, the response regulator in bacterial chemotaxis. Biochim Biophys Acta
1993;1202:297304.
284 Stock AM, Stock JB: Purification and characterization of the CheZ protein of bacterial chemotaxis. J
Bacteriol 1987;169:33013311.
285 Kuo SC, Koshland DE Jr: Roles of cheY and cheZ
gene products in controlling flagellar rotation in
bacterial chemotaxis of Escherichia coli. J Bacteriol
1987;169:13071314.
286 Snyder MA, Koshland DE Jr: Identification of the
esterase peptide and its interaction with the cheZ
peptide in bacterial sensing. Biochimie 1981;63:113
117.

287 Parkinson JS, Parker SR: Interaction of the cheC


and cheZ gene products is required for chemotactic
behavior in Escherichia coli. Proc Natl Acad Sci
USA 1979;76:23902394.
288 Senesi S, Celandroni F, Salvetti S, Beecher DJ, Wong
AC, Ghelardi E: Swarming motility in Bacillus
cereus and characterization of a fliY mutant
impaired in swarm cell differentiation. Microbiology
2002;148:17851794.
289 Celandroni F, Ghelardi E, Pastore M, Lupetti A,
Kolsto AB, Senesi S: Characterization of the chemotaxis fli Y and che A genes in Bacillus cereus. FEMS
Microbiol Lett 2000;190:247253.
290 Levit MN, Liu Y, Stock JB: Mechanism of CheA protein kinase activation in receptor signaling complexes. Biochemistry 1999;38:66516658.
291 Hess JF, Bourret RB, Simon MI: Histidine phosphorylation and phosphoryl group transfer in bacterial
chemotaxis. Nature 1988;336:139143.
292 Stock JB, Ninfa AJ, Stock AM: Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol Rev 1989;53:450490.

Dr. Christopher V. Rao


Department of Chemical and Biomolecular Engineering, University of Illinois at Urbana-Champaign
211 Roger Adams Laboratory, MC-712, Box C-3
600 S. Mathews Ave, Urbana, IL 61801 (USA)
Tel. +1 217 244 2247, Fax +1 217 333 5052, E-Mail chris@scs.uiuc.edu

64

Rao Ordal

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 6587

Bacterial PEP-Dependent Carbohydrate:


Phosphotransferase Systems Couple
Sensing and Global Control Mechanisms
Joseph W. Lengelera Knut Jahreisb
a
Max-Planck-Institut fr Dynamik komplexer technischer Systeme, Magdeburg, and bAG Genetik, Fachbereich
Biologie/Chemie, University of Osnabrck, Osnabrck, Germany

Abstract
The PEP-dependent carbohydrate:phosphotransferase systems (PTSs) of enteric bacteria constitute
a complex sensory system which involves as its central element a PEP-dependent His-protein kinase
(Enzyme I). As a unit, the PTS comprises up to 20 different transporters per cell which correspond to
its chemoreceptors for PTS carbohydrates, and several targeting subunits, which include in the low
[G + C] Gram-positive bacteria an ancillary Ser/Thr-protein kinase. The PTS senses the presence of
carbohydrates, in particular glucose, in the medium and the energy state of the cell, in the form of
either the intracellular PEP-to-pyruvate ratio or the D-fructose-bisphosphate levels. This information
is subsequently communicated to cellular targets, in particular those involved in the chemotactic
response of the cell towards PTS carbohydrates, and in sensing glucose in the medium, using cAMP
and several targeting subunits as intermediates. Peptide targeting subunits ensure the fast, transient, and yet accurate communication of the PTS with its more than hundred different targets,
avoiding at the same time unwanted cross-talk. Many elements of this sensory system are simultaneously elements of specific and global regulatory networks. Thus, the PTS controls, besides the immediate (in the ms to s range) chemotactic responses, the activity of the various carbohydrate
transporters and enzymes involved in carbon and energy metabolism through inducer exclusion,
and in a delayed response (in the min to h range) the synthesis of these transporters and catabolic
enzymes through catabolite repression. Indirect consequences of this program are phenomena
related to cell surface rearrangements, which include flagella synthesis, as well as memory, adaptation, and learning effects. The analogy between the PTS and other prokaryotic systems, and more
complex sensory systems from eukaryotic organisms which share elements with regulatory systems
Copyright 2009 S. Karger AG, Basel
is obvious.

Biological sensory or signal transduction systems comprise first a receptor, by definition, the stimulus-perceiving element. Receptors usually form a complex with a signal
transmitter or transducer. Together, these constitute the sensor which converts the
stimulus into a signal. In bacteria, an ATP-dependent His-protein kinase very often

constitutes the central element of a transducer, which during signal transduction, transfers its phosphoryl group to an aspartate residue of its receiver. The receiver is part of a
response regulator which modulates the activity of a target. Normally, a phosphatase, or
a protein which acts as a phosphate sink, completes this part of the sensory system.

Sensory (Signal Transduction) and Global Regulatory Systems Often


Share Elements in the Prokaryotes

At first, many bacterial sensory systems seemed to comprise only the two components of sensor and response regulator, hence their name two-component systems.
But after dozens of such systems have been analyzed in detail, it has been established
that they display an astonishing complexity which reaches far beyond a two-component structure in the simple sense [reviewed in 1]. Thus, a sensor may comprise many
subunits, and all sensors of a cell may cluster in supramolecular complexes containing
hundreds of proteins. Furthermore, there are membrane-bound sensors for extra- or
intercellular, and soluble sensors for intracellular stimuli. Like all biological sensory
and regulatory systems, the two-component systems act in a transient way, i.e., the
cell must be able to adapt to any longer lasting stimulus. This process involves further reversibly acting protein complexes, in particular when serine/threonine protein
kinases are involved (see Rao et al., pp. 3364, for details).
In the original definition, a group of elements whose common target affected the
behavior of a cell or an organism, e.g. its swimming (taxis) or the direction of its
growth (tropism), was called a sensory system. However, whenever they affected
metabolism or differentiation processes, e.g., because their target was a gene regulator,
they were considered to constitute a (global) regulatory system. This strict distinction
has become questionable because time and again systems are found which modulate
the sensory responses of a cell, and share elements with global regulatory systems
which control its physiological state. Furthermore, while immediate responses (in
the ms to s range) of a sensory system to a stimulus are mediated invariably at the
enzyme activity level, delayed or even late responses (in the min to h range), i.e., those
involving adaptation and learning, often also involve enzyme synthesis, and hence
gene regulatory processes. Thus, during the switch from feast to starvation and back,
bacteria often rearrange their cell surface drastically which may include the synthesis
and degradation of new chemoreceptors, and of the locomotory devices [2].
This article will concentrate on the mechanisms by which one of the best studied
bacterial systems, the PEP-dependent carbohydrate:phosphotransferase system (PTS)
of enteric bacteria, coordinates the response of a cell to chemical stimuli in relation
to its physiological state. The system involves up to 20 different transporters per cell,
a series of targeting subunits, and a PEP-dependent protein kinase [reviewed in 3, 4].
As a unit, the PTS senses the presence of carbohydrates in the medium as well as the
energy state of the cell, and it regulates directly carbohydrate transport, carbon and

66

Lengeler Jahreis

Regulatory part

Sensory part

Carbohydrate transport

PTS transport/phosphorylation

Inducer exclusion
Carbon Catabolism

Fructose

Catabolite repression

Mannose

PEP

Mannitol

PTS

Transcription
anti-termination

Pyruvate

etc.

Chemotaxis

Glycogen storage

Flagellar
Motility
motor

Glucose metabolism
Glucose

Mlc

Energy Metabolism
Catabolite repression
Cell surface rearrangements

Fig. 1. Schematic view of the PTS as a sensory and global regulatory system. See text for further
explanations.

energy metabolism [5, 6], a glucose-sensing system [7], and the chemotaxis towards
PTS carbohydrates [8] (fig. 1). Indirectly, it is also involved in cell surface rearrangements, which include flagella synthesis, and in prolonged epigenetic (adaptation and
learning) phenomena. Further details on bacterial chemotaxis, in particular within
Gram-positive bacteria, can be found in the article by Rao et al. (pp. 3364), while
details on the regulatory networks controlled by PTSs in Gram-positive and Gramnegative bacteria can be found in a recent review [9]. Wherever possible, we will also
indicate other bacterial sensory systems with such a dual role, in particular when they
also include PTSs.

Components and Structure of Various PTSs Involved in the Coupling of Sensory and
Regulatory Mechanisms

PTS Has a Dual Role in Carbohydrate Transport and in Sensing the Energy State of a Cell
More than 50 components of a cell constitute together the sensory part of the PTS.
Its general structure is similar in all species (fig. 2) [3, 4, 9]. It comprises, firstly, two
cytoplasmic proteins, the protein kinase Enzyme I (EI), and a phospho-carrier or histidine protein (HPr), and secondly, a substrate-specific Enzyme II (EII). In general,
EI and HPr are common to all EIIs which may number up to 20 different ones per

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

67

Enzyme II (Transporters and chemoreceptors)

Carbon catabolism
P~A P~FPr
Fructose

Mannose

C
D

Mannitol

P~B P~A

Mtl 1P

Glucose

P~B

Glc 6P

P~B

Fru 1P

P~B P~A

P~A

Man 6P

P~HPr

EI

PEP

HPr

P~EI

Pyruvate

Energy metabolism

Fig. 2. Elements of the PTS as carbohydrate transporters and as a sensory system. Up to 20 Enzymes
II with their membrane-bound domains EIICD, and their cytosolic domains EIIAB constitute so many
carbohydrate:phosphotransferase transporters and at the same time chemoreceptors. A protein
kinase Enzyme I (EI) and a phospho-carrier HPr are common to all EIIs, and constitute the signal
transducer. Domains are free or fused (overlapping in the figure). See text for further explanations.

cell. Each EII consists of one or two integral membrane-bound domains (EIIC/EIID),
which are necessary in substrate translocation through the membrane. They form a
complex with two hydrophilic domains which are central in substrate phosphorylation, namely EIIA, which is phosphorylated by HPr at a histidine (P~HEIIA), and
EIIB (P~CEIIB) which is phosphorylated by EIIA at a cysteine (rarely a histidine).
Often, individual EIIs together with EI and HPr are also designated as a PTS. Then,
they are named according to their major (inducing) substrate, e.g. the d-glucose (Glc),
N-acetyl-d-glucosamine (Nag), d-mannose (Man), or d-mannitol (Mtl) PTS.
The three (or four) domains EIIA to EIID of an EII are either three (or four)
distinct proteins, or the domains are fused in varying combinations and orders,
including fusions like HPr-EI, or EIIA-HPr. Specific HPr-like proteins (pseudoHPrs) rarely exist, e.g., (EIIAFru-FPr) or (EIIATag-TPr) for the d-fructose (Fru) or the
d-tagatose (Tag) PTS, respectively (fig. 2). These are also phosphorylated by EI, but
HPr can efficiently substitute for deleted pseudo-HPrs [3, 9]. In the various Gramnegative and Gram-positive bacteria, EIIs can be grouped, based on sequence similarity, into at least four genetic super-families, each with many sub-families [3, 10,
11]. The distinct domains remain active only when exchanged between PTSs of the
same genetic family.

68

Lengeler Jahreis

Next, we will discuss the structure of each of these components as far as this is
necessary to understand their role in sensing and signal transduction to their various
targets. The function itself will be described in a later section.

Enzymes II of the PTS Are Transport Systems and Chemoreceptors for


PTS Carbohydrates
Information on the structure and function of the integral membrane domains EIIC
(and EIID) is limited and based mostly on data from a few EIIs [reviewed in 3, 4].
With only few NMR and X-ray crystallography data available for the soluble domains,
analysis of the membrane domains relies on indirect data such as sequence alignments, domain complementation, and protein fusion studies [9, 10], cysteine crosslinking and accessibility data [1214], tryptophan fluorescence and phosphorescence
studies [15], and the analysis of mutants affected in transport, phosphorylation, and
in coupling both activities [1618]. According to such data, EIIC(+D) domains from
all PTS families comprise at least six typical trans-membrane helices, i.e., -helices
of 21 residues length with a hydrophobicity value 1.8. They also comprise additional, sometimes shorter, membrane-crossing structures with low hydrophobicity,
e.g., -sheet-rich segments, and extended hydrophilic structures, both with increased
sequence conservation compared to the canonical helices [19]. Thus, the membranebound EIICMtl domain consists of 42% -helical, and 26% -sheet structures, the
remainder forming random coils.
EIIs are active as homo-dimers, their interface situated between the membraneembedded IIC domains [12, 13]. The (2 6) trans-membrane helices of a dimer form
a channel-like structure, into which parts of their less hydrophobic structures fold
back to line up a more hydrophilic pathway for the translocation of the substrate [10].
EIIC dimers thus form a translocator, which alone is sufficient to catalyze slow substrate translocation. But for efficient transport and consequent substrate phosphorylation, a phosphorylated EIIB-EIIC interface is required [4, 10, 12, 13, 20]. PTSs were
classified originally as a distinct group of carbohydrate transporters because their
EIIs seemed to couple substrate translocation and phosphorylation in an obligatory
way, hence the name group translocation for this process. However, both functions
constitute distinct processes which can be uncoupled. Thus, low-affinity substrates
(analogues) can be transported in the non-phosphorylated form, and uncoupling
mutations have been isolated in EIIMtl and in EIIGlc, which allow efficient transport of
their high-affinity substrates in the absence of phosphorylation [21, 22].
A series of mutants affected in transport, phosphorylation and in coupling these
activities, have been analyzed, mostly in EIIMtl and EIIGlc [1618, 21, 22]. As expected
from our model, most residues essential for the three activities were located within
the postulated translocation pathway or at the interface between the EIIC and EIIB
domain, but not in the trans-membrane helices. Essential residues affected substrate

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

69

binding and translocation through the membrane, or the coupling of transport and
phosphorylation. Others were the preferred target for uncoupling mutations. Still
other residues were found to be critical for exposing substrate molecules bound in
the IIC domain to the phosphorylation site in the IIB domain [12, 13, 15]. These data
provide strong evidence that during this process, a conserved area of the EIIC domain
with the bound substrate molecule, and the area of the EIIB domain near the phosphorylated cysteine must be, at least temporarily, in close contact.
From such data and extensive work on the mechanism of transport and phosphorylation through EIIs [4, 20, 23, 24], the following sequence of steps emerges: substrate
molecules enter an EII dimer from the periplasmic site. Inside a channel-like structure, they are tightly bound within a translocation pathway, and lock the translocator
in a closed state. Next, the locked-in EIIC dimer with its trapped substrate molecules
is converted through structural rearrangements into an open form. These rearrangements, caused by the presence of a phosphorylated EIIB domain at the interface, concomitantly move the bound substrate molecules towards the inside. Still bound to the
EIIC-EIIB interface, the substrates are only now phosphorylated at the expense of the
P~Cys group, and released as substrate phosphates into the cytosol. The process is
repeated after the crucial cysteine has been rephosphorylated by a P~HEIIA domain.
In summary, not only transport activation and substrate phosphorylation by
means of the same high-energy phosphate is unique to PTSs, but also their use of this
constant de- and rephosphorylation during uptake as the first step in sensing these
substrates, i.e., in the chemoreception of PTS carbohydrates.

Enzyme I of the PTS, a PEP-Dependent His-Protein Kinase, and Its Substrate HPr
The first step in the PTS is catalyzed by EI, strictly speaking a PEP-dependent Hisprotein kinase (575 residues in E. coli) with multiple domains [reviewed in 3, 4, 9].
Similar to other protein kinases, EI autophosphorylates (P~HEI) in the dimeric form
in the presence of its substrate PEP and of Mg2+ at a conserved histidine which is
located in the N-terminal domain (EI-N) of the protein. Its C-terminal domain (EIC) contains the PEP-binding site and is necessary for dimerization. When separated,
both domains remain active and can complement each other both in vivo and in vitro
[25, 26]. The crystal structure of EI-N resembles the corresponding structure of the
enzyme pyruvate-orthophosphate-dikinase (PPDK) which, similar to PEP synthase
and EI, is also phosphorylated at a histidine [reviewed in 9]. In analogy to the mechanism as deduced from X-ray and NMR methods for PPDK, it has been postulated that
in EI, the PEP/pyruvate-binding site and the phosphorylated histidine are brought
sufficiently close together by a swiveling mechanism to allow the phosphoryl transfer
from the PEP-binding site to the crucial histidine [27]. Because the phosphorylation
cycle of EI plays a central role in chemotaxis of the cell to PTS carbohydrates, this
process will be discussed in more detail later.

70

Lengeler Jahreis

EI-N also contains the HPr-binding site. HPr is phosphorylated at the N-1 position of the imidazole ring of a histidine (P~HHPr) whose surroundings are conserved
in the various HPrs and even HPr-like proteins [3, 9]. The solute structures of the
complexes between EI-N and HPr on the one hand [28] and HPr and EIIAGlc from
Escherichia coli on the other [29, 30] have been determined. EI-N and EIIAGlc interact
with the same narrow region of HPr. Because the nature of these interactions and the
central role which HPr has in the Gram-positive bacteria are crucial for our further
understanding of the PTS and its interactions with its various targets, they will be
discussed in more details later.

The PEP-to-Pyruvate Ratio: A Key Parameter in Carbon and Energy Metabolism


When structuring complex metabolic and regulatory networks, areas are regularly
found in which a large number of enzymatic reactions converge to establish the equilibrium between two metabolic intermediates. Alternatively, information originating
from different sources is stored (integrated) in the configuration or concentration
of a signaling molecule. Such nodes constitute key parameters, and are the preferred
targets for sensing and global regulatory systems which control these networks [31].
The pyruvate node constitutes such a central element of carbon catabolism, which
is the target of the PTS as a sensory system. This node couples the upper part of
carbon metabolism with its lower part, and connects, furthermore, several catabolic
and anabolic pathways. The upper part, mostly responsible for the inter-conversion
of carbohydrates, comprises glycolysis and gluconeogenesis, from which emerge the
pentose-phosphate cycle and polysaccharide synthesis pathways. The lower part comprises the TCA cycle, as well as organic acid and C1-carbon metabolism. The upper
and lower parts together form the cellular functional unit (CFU) carbon catabolism/
quest for food. The enzymes clustered around the pyruvate node are regulated in a very
sophisticated way according to the needs of the organism, and the PEP-to-pyruvate
ratio therefore reflects faithfully the varying flux of their metabolites. Consequently,
a PEP-dependent protein kinase EI, rather than the usual ATP-dependent kinases, is
optimally suited to measure permanently the PEP-to-pyruvate ratio and to signal this
global information to behavioral and regulatory networks.

Components of the Glucose Pathway and of Its Control


For most organisms, d-glucose is a carbon and energy source of outstanding importance, whose metabolism requires a sophisticated control. In the enteric bacteria, only
the glucose PTS is induced by extracellular glucose, and hence the major transporter
among several others able to take up glucose. It consists of EIICBGlc (gene ptsG) and an
EIIA domain (fig. 2) whose gene crr (catabolite repression resistance) is encoded in an

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

71

operon together with the general components HPr and EI of the PTS (genes ptsHIcrr)
[reviewed in 3]. Such an arrangement is economic because this EIIA is not only involved
in the transport and phosphorylation of glucose (hence EIIAGlc) and of other PTSs (e.g.,
for sucrose), but it has a central role in the immediate and delayed control of carbohydrate transport and metabolism in general, hence its alternative name EIIACrr. Besides
their role in glucose transport and as the chemoreceptors for glucose in the medium, the
EIICB domains modulate the activity of the repressor Mlc (gene dgsA, also mlc) for the
ptsG (and other) operon(s) in relation to its transport activity (see below). These multiple roles of EIIABCGlc further emphasize the central role of glucose in bacterial life.

Diversification of Information by Targeting Subunits in Regulatory and


Sensing Networks
The PTS was discovered as a system which uses PEP to phosphorylate a series of carbohydrates, and was later recognized as a carbohydrate transport system which generates
carbohydrate phosphates concomitant with their transport at the expense of PEP. Such
a system could have been evolved starting from any ABC transporter, which couples an
ATP-binding domain directly to a translocator, provided the system substituted PEP for
ATP as the phosphoryl donor, and did not transfer the phosphate to water molecules,
but to the substrate molecules bound in the translocator. Thus, the intercalation of additional phospho-carriers, i.e., HPr, EIIA, and EIIB, between the PEP-dependent proteinkinase EI and the translocator EIIC appears to be superfluous. But, as stated before,
the PTS is also a complex sensory system which senses by means of EI the information
included in the intracellular PEP-to-pyruvate ratio, and the presence of (PTS) carbohydrates in the medium, and which communicates this information to several hundred
different targets. As for any sensory and regulatory system, the PTS must allow the fast,
transient, and yet accurate communication with its many targets, avoiding at the same
time unwanted cross-talk. Other biological systems have solved this major problem by
using targeting subunits [6, 31]. The usage of small molecules and targeting subunits
to carry information within and between cells is common among organisms. Because
such carriers often signal drastic metabolic or physiological changes, they are called
alarmones in the prokaryotic world, and, e.g., second messengers or neurotransmitters
among the eukaryotes. The corresponding molecules usually are rare, e.g., derivatives
of nucleotides like cAMP, and of carbohydrates and amino acids, or they carry unusual
modifications, as in phosphorylated or methylated peptides. Obviously, the additional
subunits HPr, EIIA, and EIIB are such targeting units which provide the PTS as a sensory system with several new capacities (fig. 3).
Firstly, due to their specific type of binding surfaces, HPr and EIIACrr can recognize many structurally diverse proteins, and therefore can act within a network as
universal joints [6, 32]. Thus, the HPr-binding surface on the N-terminal part of EI
(EI-N) and on EIIACrr is concave and located in a shallow depression [28], while that

72

Lengeler Jahreis

out

Membrane
TRANSPORT

in

Glycolysis
Inducer
exclusion

Chemotaxis
Gluconeogenesis

Glc 6P
P~IICBGlc

IIACrr

P~HPr

EI

P~EP

IICBGlc

P~IIACrr

HPr

P~EI

Pyruvate

Glc

Glycogen
DhaCatabolite
repression Phosphorylase metabolism
Transcription
anti-termination
Glc-Induction (MIc)

Fig. 3. Targeting subunits and the diversification of information within sensory networks as found
within enteric bacteria. The targeting subunits HPr, EIIA, and EIIB carry the information from the
cellular functional unit carbon metabolism, as mirrored in the flux of phosphoryl groups through
the PTS, to a series of targets in a rapid and transient, though precise way. See text for further
explanations.

on HPr is convex and located on a protrusion [29]. Furthermore, the central region of
both sites is hydrophobic, but surrounded by polar and negatively charged residues in
one partner, and by positively charged ones in the other partner. Although the binding surfaces for EI-N and EIIACrr on HPr have 17 residues in common (out of 18 for
EIIACrr and 23 for EI-N), their structural scaffolds differ drastically as also exemplified by the binding sites for HPr on a variety of EIIA domains from different PTSs
[29, 30, 3335], and on the glycogen phosphorylase [32], or by the differing binding
sites of EIIACrr on HPr and on the glycerol kinase [36].
Secondly, all phosphorylation reactions between PEP and the EIIB domains are
reversible, i.e., the PEP-to-pyruvate ratio is reflected at no additional energy costs in
the ratio of each phosphorylated to dephosphorylated PTS component (fig. 3). Only
the last step, i.e., transfer of the high-energy phosphate group to the substrate, is virtually irreversible. The reversibility of the phosphoryl group transfer thus allows the
metabolic network to control the phosphorylation state of PTS proteins in various
ways, which, although equivalent, are not identical, e.g., when the equilibrium constants and the velocities of the phosphoryl forward and backward transfer reactions,
or the various affinity constants differ among PTS components (see below).
Thirdly, targeting subunits allow sensory and regulatory systems to respond immediately (in ms to s) by controlling reactions at the level of enzyme/transporter activities, or in a delayed form (min) by controlling gene expression and enzyme synthesis
or degradation [31].

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

73

Last but not least, a PEP-dependent protein kinase is not only optimal as part of
a series of PTS carbohydrate transport systems, and as part of a system able to sense
the PEP-to-pyruvate ratio. But for each PEP molecule spent during uptake and synthesis of a substrate phosphate, and one ATP molecule spent during its metabolism,
two PEP molecules are generated, one of which can be used to rephosphorylate the
PTS. Consequently, the cell gains a functional sensory and regulatory network for its
multiple activities at virtually no net costs [reviewed in 3].
While in enteric bacteria EIIACrr is the central targeting subunit of the PTS, HPr is the
central unit in the low [G + C] Gram-positive bacteria (fig. 4) [reviewed in 6, 9]. Their
HPr becomes not only phosphorylated by PEP and EI at His15, but also by a protein
kinase at Ser46. In fact, the kinase is a bifunctional enzyme, hence its name HPrK/P.
In the presence of Fru-bisphosphate (FBP), i.e., in the presence of rapidly metabolizable carbohydrates, HPrK functions as an ATP-dependent kinase, thus producing
P-Ser46HPr. By a mechanism, called phospho-phosphorolysis, the enzyme catalyzes
in a reversible reaction the conversion of Pi and PPi. Consequently, HPrK/P dephosphorylates P-S46HPr in the presence of high concentrations of Pi, as found in starved
cells, and generates PPi. Alternatively, it phosphorylates S46HPr at the expense of PPi.
Because the latter is the energetically favored reaction, PPi needs to be hydrolyzed by a
pyro-phosphatase (YvoE) to allow efficient dephosphorylation of P-S46HPr.
The ratio of the various forms of HPr, in particular HPr, P~H15HPr, and P-S46HPr,
thus seems to be regulated mostly via the metabolites FBP, Pi, and PPi, and hence fast
and feast conditions of the cell. Because P-S46HPr is a poor substrate for the PEPdependent phosphorylation by EI, their ratio is shifted under feast conditions towards
P-Ser46HPr, the major targeting subunit for targets involved in inducer exclusion,
and in the CcpA-dependent catabolite repression. Here, as before for enteric bacteria,
the varying targeting subunits enable the PTS to fulfill its many roles in sensing the
energy state of the cell and to control, in a global way, its entire carbon metabolism.

Functions of the PTS Which Couple Sensing and Global Control

All functional subunits of the PTS can exist in the phosphorylated or unphosphorylated form. Enteric and low [G + C] Gram-positive bacteria use this information to
couple these targeting subunits, depending on their level of phosphorylation, to various regulatory cascades, and firstly to the chemotactic network.

Role of the PTS as a Sensory System in Chemotaxis


Increasing gradients of PTS carbohydrates elicit counterclockwise (ccw) flagella
rotation, smooth swimming, and thus, a positive chemotaxis. Mutants which lack a
specific EII do not exhibit chemotaxis towards their cognate substrates. In contrast,

74

Lengeler Jahreis

Transcription
anti-termination

Inducer exclusion

P~H15HPr
Chemotaxis

2 Pi
YvoE
Pi

PPi

P-S46HPr

P~EI
ATP

ADP

S46HPr
HPrK/P

P-S46HPr

Flagellar
motor

Motility

HPrK/P
Fru BP

CcpA.cre dependent Catabolite repression

Fig. 4. Targeting subunits and the diversification of information within sensory networks as found in
low [G + C] Gram-positive bacteria. The central targeting subunit HPr can be phosphorylated either
at a histidine (His15), or at a serine (Ser46). Each of the four possible phosphorylated and unphosphorylated forms communicate with many different targets, as will be discussed in the text.

mutants lacking EI or HPr do not chemotactically respond to any PTS substrate, even
if the corresponding EIIs are present [3739]. Mutated or truncated forms of EIIMtl
which cannot be phosphorylated, but still bind their substrates, as well as uncoupled
mutants which take up unphosphorylated mannitol, also do not elicit a chemotactic
response [39, 40]. Furthermore, extensive metabolism of the stimulating substrate is
not necessary, since non-metabolizable analogs, and normal substrates in mutants
with a defect in subsequent metabolism of the carbohydrate phosphates, are good
attractants. Finally, the apparent affinities for transport, phosphorylation and chemotactic response in wild-type cells are very similar. In summary, PTS carbohydrates are
sensed as chemo-effectors during uptake and phosphorylation [3739, 41].
For a long time it was not clear how the PTS signal is capable of influencing the rotation pattern of the flagellar motor. Motile bacteria, such as E. coli and B. subtilis, detect
various stimuli from the chemical or physical environment through different sensors,
the most important sensors being the membrane-bound methyl-accepting chemotaxis proteins (MCPs). But ultimately, all signals seem to affect the central chemotactic
complex which comprises as its central elements the proteins CheAWYZ (fig. 5). This
complex functions as an intrinsic oscillator (tumble generator) which eventually triggers the periodic change between the counter-clockwise (ccw) and the clockwise (cw)
flagellar rotation [reviewed in 8]. In brief, stimulus perception is signaled across the
inner membrane to a cytoplasmic domain of an MCP, which communicates through

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

75

out

Membrane

in

in

Membrane

out

MCP chemosensors
Ser
Asp
Mal
Rbs
Gal

P-CheY

CheA

ATP
+

CheY

P~CheA

Mtl
Man
Gut

EI

P~HPr
fast

Chemotaxis

CheZ

ADP

Enzymes II

cw

EI

slow

Pi

Flagellar
motor Motility

EI
EI

P~EP
slow

fast

P~B P~A
fast
Substrate-P

HPr

P~EI
P~EI

P~EI

Pyruvate

P~EI

Fig. 5. Principal components of the MCP and the PTS phospho-relays. The upper part shows the
protein phosphorylation reactions, which are modulated by MCP molecules in response to the binding or release of attractants such as serine or maltose. The lower part shows the reactions involved in
the uptake and phosphorylation of PTS substrates. Unphosphorylated EI molecules inhibit the autophosphorylation reaction of CheA, leading to a change in flagellar rotation. See text for further
explanations.

an ancillary protein, CheW, with the sensor kinase CheA. Using ATP as the phospho donor, CheA autophosphorylates at a His residue (P~HCheA) and subsequently
donates the phosphate group, first, to an Asp residue in the response regulator CheY
(P-DCheY). Phosphorylation of CheY induces a conformational change which enables
the protein to interact with the flagella switch and to trigger cw rotation, ccw being the
default state. P-DCheY is short-lived because of a self-catalyzed hydrolysis, a reaction
accelerated by the protein phosphatase CheZ. This signal transduction pathway is controlled by an adaptation mechanism which involves MCP methylation/demethylation.
This process employs a permanently active methyltransferase CheR, which increases
CheA activation, and a methylesterase CheB. Like CheY, CheB is phosphorylated by
P~HCheA and is thus activated. In its phosphorylated form, P-DCheB counteracts
CheA phosphorylation by a demethylation of the MCPs. This sophisticated interplay
between spontaneous CheA phosphorylation, its fast triggering of a tumble movement
through P-DCheY, and its slow counteraction by P-DCheB, acts as a spontaneous oscillator in flagellar rotation, which is endowed with a short-term memory, and which can
be modulated by the various sensors in different ways. Thus, an increase in attractant
or a decrease in a repellent concentration at an MCP causes inhibition of CheA autophosphorylation and smooth swimming, whereas a drop in attractant levels (or the
presence of a repellent) stimulates CheA and triggers cw rotation and tumbling.

76

Lengeler Jahreis

Except as part of this tumble generator, MCPs, whose general structure and function is similar in all bacteria, are not required as sensors for PTS chemotaxis [42, 43].
MCPs do not transport stimulating substrates into the cell, but rather measure the
external levels of attractants and repellents. As concluded before, this contrasts the
PTS as a sensor for which the constant phosphorylation and dephosphorylation of
its components during the uptake of PTS carbohydrates corresponds to the signal
[reviewed in 44]. In vitro studies with purified PTS and Che proteins [45] established
that the flagella rotation signal involves a change in phosphate flux through EI and
HPr. The in vitro results showed that unphosphorylated EI, but not P~HEI inhibits the autophosphorylation activity of CheA by up to 10-fold, and that this activity
decreases about 3-fold when the PEP concentration was lowered from 2 to 0.3 mM
in the test. No phosphoryl transfer between these two histidine-protein kinases could
be demonstrated, as confirmed in another study [46]. These results were corroborated by kinetic analyses of wild-type and mutant strains for a correlation between
the phosphorylation state of the PTS and CheA activity [41], and by the in vivo analysis of HPr mutants, which exhibited transport of PTS substrates, but were unable to
respond chemotactically to them. All these HPr derivatives showed a reduced phospho transfer from EI to HPr [47, 48].
The key to understand the role of the PTS in chemotaxis lies obviously in the complex phosphorylation cycle of its protein kinase EI, which requires a permanent monomer/dimer transition [49]. Autophosphorylation of EI occurs only in the dimeric
state (fig. 5). After phosphorylation, the dimers dissociate and subsequently transfer
the phosphate groups to HPr. This obligate dimerization prior to autophosphorylation appears to be the rate-limiting step in the EI phosphorylation cycle, since the rate
of association/dissociation is very slow in comparison to other protein kinases [49,
50]. Furthermore, dimerization of EI is promoted by PEP and Mg2+ and inhibited by
pyruvate [51]. Together, all these results suggest the following working model. During
fast uptake of a PTS carbohydrate through any EII, EI is dephosphorylated more
rapidly by HPr than rephosphorylated at the expense of PEP, in particular because
the PEP concentrations are lowered concomitantly. Consequently, the amount of
free EI increases transiently, thus causing inhibition of CheA autophosphorylation.
In the kinetic analyses, response was triggered by flash photolysis of caged glucose
and -methylglucoside in about 10 ms, and with a high sensitivity (threshold near 10
nm). Model calculations [52, 53] predict that at high intracellular PEP concentrations,
where mainly P~HEI should be present, even a slight decrease in PEP concentration
could cause the EI increase necessary to inhibit CheA, a decrease in P-DCheY levels,
and thus a positive chemotactic response.
Though deviating in details (see Rao et al., pp. 3364), PTS-dependent chemotaxis
in Bacillus subtilis is also modulated according to the phosphorylation levels of the
PTS proteins, which again reflects the universality of this sensing mechanism.
A similar logic can be found in the Aer-dependent energy taxis, which covers aerotaxis (to oxygen) and taxis to alternative electron acceptors [reviewed in 2, 8]. Aer

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

77

is homologous to the MCPs, but it is unique in that it does not comprise a periplasmic sensing domain. Instead Aer contains both PAS and HAMP domains including a
non-covalently bound flavin-adenine dinucleotide cofactor (FAD). Rather than sensing oxygen directly, Aer detects fast redox changes, most likely of the membranebound electron transport system. These redox changes modulate the activity of Aer
which in turn modulates the CheA autophosphorylation activity. This is reminiscent
of the PEP-to-pyruvate ratio sensed by the PTS, which mirrors carbon catabolism,
and concomitantly triggers an immediate behavioral response. Further common elements will be discussed later.

Role in Sensing Glucose and in Control of Glucose-Related Genes: The Mlc Regulator
E. coli K-12 utilizes different transporters for the uptake of d-glucose, in particular
the glucose PTS (EIICBGlc; gene ptsG), and the mannose-PTS or IIMan [reviewed in 3].
As expected for the transporter of a main substrate, the regulation of ptsG in E. coli is
very complex (fig. 6). It involves several transcription factors and auxiliary proteins
(Crp, ArcAB, Fis, SgrRT), a small regulatory RNA (SgrS), a small regulatory peptide
(MtfA), and three different sigma factors (RpoD, RpoS, RpoH), but above all a direct
linkage between the genetic regulatory mechanisms and the activity of the transporter
and sensor EIICBGlc [reviewed in 54].
The major specific regulator of ptsG expression is the repressor Mlc (mnemonic
for makes large colonies, previously DgsA), which is inactivated by glucose in the
medium. In contrast to other repressors, induction of Mlc is not catalyzed by direct
binding of glucose, or of any small molecular inducer. Instead, as part of a novel regulatory mechanism, the membrane-bound EIICBGlc binds Mlc, but only when it is in
its dephosphorylated form. Thus, in the absence of glucose, Mlc binds to its promoter
ptsGp, while in the presence of glucose, the dephosphorylated EIICB sequesters the
repressor away from its promoter, allowing induction and enhanced ptsG transcription [5558]. The complex between the membrane-bound EIICBGlc with a cytosolic
DNA-binding protein Mlc, thus forms an efficient glucose-sensing system. Perhaps
not surprisingly, DNA-microarray experiments revealed a pleiotropic role of Mlc in
the regulation of further carbohydrate transporters and catabolic enzymes, including the ptsHIcrr operon itself, in response to the presence of glucose (references in
Deutscher and Postma [9], and pers. unpubl. results).

Targeting Subunits Allow the PTS to React in a Global Way through Immediate and
Delayed Responses to Changes in the CFU Carbon Catabolism/Quest for Food
By using targeting subunits, the PTS becomes not only able to perform its multiple global regulatory functions, and to communicate in a reliable way with its many

78

Lengeler Jahreis

+ Glucose

Glucose
Glc

P~B
Glc 6P

SgrR

SgrT

Mlc

Mlc

Mlc
MtfA

MtfA

Crp
cAMP

+
Mlc

Mlc

ptsGOP

ptsG

ptsGOP

RNaseE

ptsG

SgrR

Fig. 6. Regulation of the major glucose transport system IICBGlc (encoded by ptsG) in E. coli. In the
absence of glucose, ptsG expression is repressed by Mlc. In the presence of glucose, Mlc is sequestered to the transporter IICBGlc, or to the auxiliary protein MtfA. Fine tuning of ptsG expression under
various growth conditions depends on the global-acting transcription factors cAMP-CrP, ArcA~P, Fis
and on several sigma factors. The stability of the ptsG-mRNA is triggered by the small regulatory RNA
SgrS. Moreover, a small polypeptide SgrT (also encoded by the sgrS gene), downregulates glucose
uptake by a direct inhibition of the EIICBGlc transport activity [76]. The expression of sgrS/T is regulated by the transcription activator SgrR, which in turn becomes activated during intracellular glucose-P or fructose-P stress.

targets, but also to react through immediate or delayed responses to changes in cellular carbon and energy metabolism (fig. 3, 4).
As exemplified by the fast (10 ms) chemotactic reaction of E. coli to the addition
of glucose, immediate control mechanisms can only act at the level of enzyme activity, in this particular case of EI, CheA, CheY, and the flagella motor (fig. 5). A second
immediate control mechanism, involving the PTS as sensing system, rapidly inhibits a
number of non-PTS transporters and some key enzymes, in particular those involved
in taking up inducing substrates, or in generating the real intracellular inducer, e.g.
glycerol kinase (GlpK) generating sn-glycerol-phosphate from glycerol. Hence, its
name inducer exclusion [reviewed in 3, 9]. In well-fed cells, e.g., during fast growth
on glucose, unphosphorylated EIIACrr accumulates in the cell, binds directly to the
mentioned transporters and enzymes, and immediately inhibits their activity. Efficient
binding of EIIACrr to its targets occurs only when the respective substrate is present.
This mechanism ensures that EIIACrr will not be wasted and titrated out by binding to
unoccupied transporters and enzymes. For obvious reasons, inducer exclusion is most

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

79

effective during the initial induction phase when the ratio of EIIACrr to its targets is
very high, because, in starving cells, containing the non-inhibitory P~HEIIACrr form,
and in fully induced cells, EIIACrr will be titrated out. Under these conditions, however, a delayed response, i.e., catabolite repression, will become active (see below).
As discussed before, targeting subunits often comprise a general interface, or various interfaces, which facilitates their role as a universal joint. Thus, the binding site
of EIIACrr for GlpK differs from its binding site for HPr. The former comprises a negatively charged Glu which comes close to the phosphorylated His on EIIACrr, explaining why only unphosphorylated EIIACrr inhibits GlpK during inducer exclusion. The
binding of EIIACrr leads to a locked-in conformational change of the enzyme which
thus becomes catalytically inactive and unable to synthesize the intracellular inducer
sn-glycerol-P [36]. Maltose is taken up by the maltose ATP-binding cassette transporter, which consists of two membrane-integral subunits (MalF and MalG), of two
copies of the ATPase subunit MalK, and of the periplasmic maltose-binding protein
MalE. Inhibition of maltose transport by EIIACrr in vivo is most likely caused by the
elimination of the substrate-stimulated ATPase activity of the transporter, which
indicates a direct interaction with the MalK subunits [59]. In this case, as for various membrane proteins such as the lactose-permease (LacY), the melibiose-permease
(MelB), or the raffinose-permease (RafB), knowledge concerning their interactions
with EIIACrr is rather limited, and no clear consensus sequence for a binding motif
could be proposed [60]. For LacY it was shown that during substrate transport, EIIACrr
seems to bind exclusively to the outward conformation, in which the carbohydratebinding site is faced to the periplasm, thus preventing an inward conformation,
which exposes the carbohydrate to the cytoplasm, and hence allows sugar uptake.
In the enteric bacteria, the PTS uses its targeting subunit EIIACrr, but now in the
phosphorylated form P~HEIIACrr, to control carbon catabolism and energy metabolism in a delayed response mechanism known as (EIIACrr-dependent) catabolite repression. Catabolite repression acts through global transcription control by means of the
global regulator Crp (mnemonic for catabolite receptor protein, also called CAP for
catabolite activator protein). Crp is activated by binding of the rare nucleotide cAMP,
which is synthesized from ATP by the enzyme adenylate cyclase CyaA (gene cyaA).
When complexed with cAMP, Crp binds to a consensus sequence found in the promoters of more than 100 genes and operons, which together form the crp-modulon.
This genetic unit which, compared to operons and regulons, constitutes a higher level
of complexity, includes basically all the genes of the CFU carbohydrate catabolism/
quest for food [reviewed in 3, 9, 61]. All the evidence indicates that P~HEIIACrr, when
bound to CyaA, stimulates cAMP synthesis by about 5-fold. Because P~HEIIACrr concentrations are high under the conditions of low PTS transport rates and high PEP
to pyruvate ratios, conditions normally found in starved cells, starvation triggers an
increased cAMP synthesis, and hence the pleiotropic activation of most members of
the crp-modulon. In contrast, feast conditions cause a dephosphorylation of EIIACrr,
and a decrease of the intracellular cAMP concentrations. Except for a few promoters

80

Lengeler Jahreis

whose activity is lowered by the cAMP-Crp complex, catabolite repression thus corresponds to the lack of activation of gene transcription.
The proposed model on CyaA activation is largely based on mutant analyses.
Although the results gave strong indications for a central role of P~HEIIACrr, in vitro
experiments with purified proteins, performed with the intention of displaying its
binding in the activation process, were not conclusive. Instead they hint at some, as
yet unknown, regulatory component also involved in the activity control of CyaA
[62]. Recent evidence demonstrated that a strict correlation between P~HEIIACrr
levels and intracellular cAMP concentrations only exist in cells growing at specific
growth rates between 0.3 and 0.7 h1, corresponding to generation times of about 140
to 60 min. Above and below this range the correlation became more and more diffuse, which again seems to argue for additional regulatory elements [63].
Directly related to the topic of delayed sensory responses are phenomena related
to long-term adaptation (maintenance) processes, and learning, as well as other epigenetic processes. Delayed answers are based on complex regulatory networks which
include slow feedback loops, e.g. the CheRB-dependent mechanism in chemotaxis.
Or they involve protein synthesis and degradation steps, which are also in themselves controlled by global regulatory mechanisms, as will be illustrated by three
examples:
(1) E. coli cells inoculated into soft-agar swarm plates containing 100 m of each
PTS substrate d-mannitol and d-glucitol (also called d-sorbitol) grow and swarm out
in two sharp concentric rings separated by about 5 mm [2, 64]. Cells from the outer
ring grow and follow mannitol exclusively, whereas cells from the inner ring consume
glucitol after a delay of about 45 min. Sequential metabolism is based on the fact that
synthesis of the sensor/transporter EIIGut and other glucitol-degrading enzymes is
strongly cAMP-Crp-dependent, whereas synthesis of EIIMtl and mannitol-degrading
enzymes depends only marginally on the presence of cAMP-Crp. Moreover, during
growth on mannitol, intracellular cAMP concentrations remain low. Hence, when
reinoculated into a similar plate, cells of the outer ring again form two sequential
rings, while cells from the inner ring, being preinduced for both sets of enzymes,
move for three to five generations in one concentric ring and consume both polyhydric alcohols simultaneously. This maintenance effect, which is based on the prehistory of the population, i.e., an epigenetic factor, can be perpetuated, when necessary,
for generations until the sensors and metabolic enzymes are repressed and lost from
the cells by degradation and dilution at each cell division.
(2) Expression of the 40-odd flagellar genes for flagellar structure, assembly and
function, is controlled by a three-level regulatory hierarchy. Genes flhDC at the highest level are organized in a master operon, whose products FlhDC are required for the
expression of all lower-level flagella genes (see Rao et al., pp. 3364, and Eisenbach [8]
for details). Because expression of this master operon, a member of the crp-modulon,
is strictly cAMP-Crp-dependent, only starved cells are fully flagellated and swim vigorously [64]. In addition, expression of the master operon is coupled to the cell cycle,

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

81

being affected by a number of global regulatory signals which together control a strict
coordination of cell surface rearrangements and cell division [65].
(3) In E. coli, the PTS is also involved in the regulation of the balance between
fermentation and respiration beyond its role in catabolite repression of the TCA
enzymes. Using a surface plasmon resonance spectroscopy (SPR)-based ligand fishing, a fermentation/respiration switch protein (FrsA) was found which interacts with
EIIACrr, and which seems to be involved in respiration control during growth on several sugars including glucose [66]. This resembles a link which seems to exist between
FlhDC, cell division, Aer, the MCP which senses the intracellular energy levels, and
transcription control for a number of enzymes involved in anaerobic respiration
[67].
As indicated before, inducer exclusion, catabolite repression, and related regulatory mechanisms exist also in the low [G + C] Gram-positive bacteria, but here the
targeting subunit HPr (fig. 4) functions as the equivalent of EIIACrr in enteric bacteria
(see above and Poncet et al., pp. 88102). But even in E. coli, HPr has a regulatory
function in glycogen metabolism. Basically, at the onset of the stationary phase when
cAMP levels increase, the expression of the glg genes involved in glycogen synthesis and degradation are activated by the cAMP-Crp complex, while the activity of
the degrading enzyme glycogen phosphorylase (GlgP), is inhibited by P~HHPr. In
starved cells of the late stationary phase, however, cAMP- and P~HHPr levels drop,
GlgP is activated by binding free HPr, and glycogen degradation begins [68, 69].
These examples demonstrate how the physiological state and, in particular, feast
and famine conditions are sensed by the PTS and used to control the CFU carbon
catabolism/quest for food which also includes flagella synthesis, cell motility, and the
overall chemosensory capacity of a cell.

Mathematical and Computer-Assisted Modeling of Complex Networks


Metabolism of a cell can be structured into meaningful CFUs, provided all relevant metabolic pathways, i.e., those with a common physiological role, are known.
Structuring requires further the knowledge of the true, i.e., measured in vivo, kinetics
of the various enzymes and the flux of key metabolites through the central nodes of
a CFU. This in turn requires detailed knowledge of the regulation of enzyme activity and synthesis, at both the local and the global level, as can be obtained through
genomics and quantitative proteomics. Finally, data on the sensory system(s), which
detect(s) the extra- and intracellular stimuli to which a given CFU responds, must
also be included [31].
The wealth of information on a CFU gathered by this approach can only be handled when assisted by mathematical modeling and computer simulation, a shown by
several groups for the PTS and its multiple roles as a global regulator [31, 53, 63,
7073]. Validation of such models through further experiments becomes imperative

82

Lengeler Jahreis

whenever the models point to additional parameters, e.g. diffusion limitations for
proteins within the cell [52], or when they indicate major deviations between in vitro
and in vivo data. Typical examples for this approach are the modeling of the control
of glucose metabolism by EIIGlc [73, 74], of phototaxis in Halobacterium salinarum
[75], and, hopefully, of the precise role the complex phosphorylation cycle of EI does
play in PTS-dependent chemosensing [53].

Conclusions

All complex sensory and regulatory systems must allow the fast, transient, and yet
accurate communication with their many targets, avoiding at the same time unwanted
cross-talk. Although all biological systems use targeting subunits to solve this problem, prokaryotic systems tend to communicate more directly between sensors and
their targets, while eukaryotic systems often intercalate sequential levels of targeting subunits, e.g., instead of one layer of targeting subunits, and one alarmone (second messenger) as in the PTS, cascades of protein kinases/phosphatases, and several
second messengers [31]. This trend is reinforced because eukaryotes have increased
redundancy among the components of complex networks and because their complexity increases even further from single cells and specific tissues to organs, organisms, and populations. The evolutionary pressure to encode, e.g., within mammals, a
maximum of peptides (up to 1 million) with a minimum of genes (2 30,000), not
only caused the multiple use of genes and gene products within the various stages of
embryogenesis and differentiation, but also within sensory and global control systems. The PTS with its comparatively primitive use of targeting subunits was obviously at the beginning of this evolutionary trend.

Acknowledgements
We gratefully acknowledge Lucille Schmieding for help with the manuscript. This work was financially supported by the Deutsche Forschungsgemeinschaft through the Sonderforschungsbereich
431 (Teilprojekt P14 to K. Jahreis).

References
1 Mascher T, Helmann JD, Unden G: Stimulus perception in bacterial signal-transducing histidine
kinases. Microbiol Mol Biol Rev 2006;70:910938.
2 Lengeler J: Chemotaxis a basic and universal phenomenon among microorganisms and eukaryotic
cells; in Eisenbach M (ed): Chemotaxis. London,
Imperial College Press, 2004, pp 752.

3 Postma PW, Lengeler JW, Jacobson GR:


Phosphoenolpyruvate: carbohydrate phosphotransferase systems; in Neidhardt FC, Curtiss III R,
Ingrahma JL, Lin ECC, Low KB, Magasanik B,
Reznikoff WS, Riley M, Schaechter M, Umbarger
HE (eds): Escherichia coli and Salmonella: Cellular
and Molecular Biology. Washington, ASM Press,
1996, pp 11471149.

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

83

4 Robillard GT, Broos J: Structure/function studies on


the bacterial carbohydrate transporters, enzymes II,
of the phosphoenolpyruvate-dependent phosphotransferase system. Biochim Biophys Acta 1999;
1422:73104.
5 Roseman S, Meadow ND: Signal transduction by
the bacterial phosphotransferase system. Diauxie
and the crr gene (J. Monod revisited). J Biol Chem
1990;265:29932996.
6 Lengeler J, Jahreis K: Phosphotransferase systems or
PTSs as carbohydrate transport and as signal transduction systems; in Konings WN, Kaback HR,
Lolkema JS (eds): Transport Processes in Eukaryotic
and Prokaryotic Organisms. Handbook of Biological
Physics. Vol. 2. Amsterdam, Elsevier Science, 1996,
pp 573598.
7 Erni B: Glucose transport by the bacterial phosphotransferase system: an interface between energy and
signal transduction; in Winkelmann G (ed):
Microbial Transport Systems. New York, WileyVCH, 2002, pp 115137.
8 Eisenbach M: Bacterial chemotaxis; in Eisenbach M
(ed): Chemotaxis. London, Imperial College Press,
2004, pp 53215.
9 Deutscher J FC, Postma PW: How phosphotransferase system-related protein phosphorylation regulates carbohydrate metabolism in bacteria. Microbiol
Mol Biol Rev 2004;70:9391031.
10 Lengeler JW, Jahreis K, Wehmeier UF: Enzymes II
of the phosphoenol pyruvate-dependent phosphotransferase systems: their structure and function in
carbohydrate transport. Biochim Biophys Acta
1994;1188:128.
11 Saier MH, Hvorup RN, Barabote RD: Evolution of
the bacterial phosphotransferase system: from carriers and enzymes to group translocators. Biochem
Soc Trans 2005;33:220224.
12 Van Montfort BA, Schuurman-Wolters GK,
Duurkens RH, Mensen R, Poolman B, Robillard
GT: Cysteine cross-linking defines part of the dimer
and B/C domain interface of the Escherichia coli
mannitol permease. J Biol Chem 2001;276:12756
12763.
13 Van Montfort BA, Schuurman-Wolters GK, Wind J,
Broos J, Robillard GT, Poolman B: Mapping of the
dimer interface of the Escherichia coli mannitol permease by cysteine cross-linking. J Biol Chem 2002;
277:1471714723.
14 Yagur-Kroll S, Amster-Choder O: Dynamic membrane topology of the Escherichia coli -glucoside
transporter BglF. J Biol Chem 2005;280:19306
19318.

84

15 Veldhuis G, Gabellieri E, Vos EP, Poolman B,


Strambini GB, Broos J: Substrate-induced conformational changes in the membrane-embedded
IIC(Mtl)-domain of the mannitol permease from
Escherichia coli, EnzymeII(Mtl), probed by tryptophan phosphorescence spectroscopy. J Biol Chem
2005;280:3514835156.
16 Lanz R, Erni B: The glucose transporter of the
Escherichia coli phosphotransferase system. Mutant
analysis of the invariant arginines, histidines, and
domain linker. J Biol Chem 1998;273:1223912243.
17 Oh H, Park Y, Park C: A mutated PtsG, the glucose
transporter, allows uptake of d-ribose. J Biol Chem
1999;274:1400614011.
18 Saraceni-Richards CA, Jacobson GR: A conserved
glutamate residue, Glu-257, is important for substrate binding and transport by the Escherichia coli
mannitol permease. J Bacteriol 1997;179:1135
1142.
19 Vervoort EB, Bultema JB, Schuurman-Wolters GK,
Geertsma ER, Broos J, Poolman B: The first cytoplasmic loop of the mannitol permease from
Escherichia coli is accessible for sulfhydryl reagents
from the periplasmic side of the membrane. J Mol
Biol 2005;346:733743.
20 Saraceni-Richards CA, Jacobson GR: Subunit and
amino acid interactions in the Escherichia coli mannitol permease: a functional complementation study
of coexpressed mutant permease proteins. J
Bacteriol 1997;179:51715177.
21 Otte S, Scholle A, Turgut S, Lengeler JW: Mutations
which uncouple transport and phosphorylation in
the d-mannitol phosphotransferase system of
Escherichia coli K-12 and Klebsiella pneumoniae
1033-5P14. J Bacteriol 2003;185:22672276.
22 Ruijter GJ, van Meurs G, Verwey MA, Postma PW,
van Dam K: Analysis of mutations that uncouple
transport from phosphorylation in enzyme IIGlc of
the Escherichia coli phosphoenolpyruvate-dependent phosphotransferase system. J Bacteriol 1992;
174:28432850.
23 Siebold C, Flukiger K, Beutler R, Erni B:
Carbohydrate transporters of the bacterial phosphoenolpyruvate:sugar phosphotransferase system.
FEBS Lett 2001;504:104111.
24 Veldhuis G, Broos J, Poolman B, Scheek RM:
Stoichiometry and substrate affinity of the mannitol
transporter, EnzymeIIMtl, from Escherichia coli.
Biophys J 2005;89:201210.
25 Fomenkov A, Valiakhmetov A, Brand L, Roseman
S: In vivo and in vitro complementation of the
N-terminal domain of enzyme I of the Escherichia
coli phosphotransferase system by the cloned
C-terminal domain. Proc Natl Acad Sci USA 1998;
95:84918495.

Lengeler Jahreis

26 Zhu PP, Szczepanowski RH, Nosworthy NJ,


Ginsburg A, Peterkofsky A: Reconstitution studies
using the helical and carboxy-terminal domains of
enzyme I of the phosphoenolpyruvate:sugar phosphotransferase system. Biochemistry 1999;38:15470
15479.
27 Oberholzer AE, Bumann M, Schneider P, Bachler C,
Siebold C, Baumann U, Erni B: Crystal structure of
the phosphoenolpyruvate-binding Enzyme I
domain from the Thermoanaerobacter tengcongensis PEP:sugar phosphotransferase system. J Mol Biol
2005;346:521532.
28 Garrett DS, Seok YJ, Peterkofsky A, Gronenborn
AM, Clore GM: Solution structure of the 40,000 Mr
phosphoryl transfer complex between the
N-terminal domain of enzyme I and HPr. Nat Struct
Biol 1999;6:166173.
29 Chen Y, Reizer J, Saier MH Jr, Fairbrother WJ,
Wright PE: Mapping of the binding interfaces of the
proteins of the bacterial phosphotransferase system,
HPr and IIAGlc. Biochemistry 1993;32:3237.
30 Wang G, Louis JM, Sondej M, Seok YJ, Peterkofsky
A, Clore GM: Solution structure of the phosphoryl
transfer complex between the signal transducing
proteins HPr and IIA(glucose) of the Escherichia
coli phosphoenolpyruvate:sugar phosphotransferase
system. EMBO J 2000;19:56355649.
31 Lengeler JW: Metabolic networks: a signal-oriented
approach to cellular models. Biol Chem 2000;381:
911920.
32 Wang G, Sondej M, Garrett DS, Peterkofsky A,
Clore GM: A common interface on histidine-containing phosphocarrier protein for interaction with
its partner proteins. J Biol Chem 2000;275:16401
16403.
33 Cornilescu G, Lee BR, Cornilescu CC, Wang G,
Peterkofsky A, Clore GM: Solution structure of the
phosphoryl transfer complex between the cytoplasmic A domain of the mannitol transporter IIMannitol
and HPr of the Escherichia coli phosphotransferase
system. J Biol Chem 2002;277:4228942298.
34 Van Nuland NA, Kroon GJ, Dijkstra K, Wolters GK,
Scheek RM, Robillard GT: The NMR determination
of the IIA(mtl) binding site on HPr of the Escherichia
coli phosphoenol pyruvate-dependent phosphotransferase system. FEBS Lett 1993;315:1115.
35 Williams DC Jr, Cai M, Suh JY, Peterkofsky A, Clore
GM: Solution NMR structure of the 48-kDa
IIAMannose-HPr complex of the Escherichia coli mannose phosphotransferase system. J Biol Chem
2005;280:2077520784.
36 Hurley JH, Faber HR, Worthylake D, Meadow ND,
Roseman S, Pettigrew DW, Remington SJ: Structure
of the regulatory complex of Escherichia coli IIIGlc
with glycerol kinase. Science 1993;259:673677.

37 Adler J, Epstein W: Phosphotransferase-system


enzymes as chemoreceptors for certain sugars in
Escherichia coli chemotaxis. Proc Natl Acad Sci USA
1974;71:28952899.
38 Lengeler J: Nature and properties of hexitol transport systems in Escherichia coli. J Bacteriol 1975;124:3947.
39 Lengeler J, Auburger AM, Mayer R, Pecher A:
The phosphoenolpyruvate-dependent carbohydrate:
phosphotransferase system enzymes II as chemoreceptors in chemotaxis of Escherichia coli K 12. Mol
Gen Genet 1981;183:163170.
40 Weng QP, Elder J, Jacobson GR: Site-specific mutagenesis of residues in the Escherichia coli mannitol
permease that have been suggested to be important
for its phosphorylation and chemoreception functions. J Biol Chem 1992;267:1952919535.
41 Lux R, Munasinghe VR, Castellano F, Lengeler JW,
Corrie JE, Khan S: Elucidation of a PTScarbohydrate chemotactic signal pathway in
Escherichia coli using a time-resolved behavioral
assay. Mol Biol Cell 1999;10:11331146.
42 Pecher A, Renner I, Lengeler JW: The phosphoenolpyruvate-dependent carbohydrate: phosphotransferase system enzyme II. A new class of
chemosensors in bacterial chemotaxis; in Sund H,
Veegher C (eds): Mobility and Recognition in Cell
Biology. Berlin, de Gruyter, 1983, pp 517531.
43 Rowsell EH, Smith JM, Wolfe A, Taylor BL: CheA,
CheW, and CheY are required for chemotaxis to
oxygen and sugars of the phosphotransferase system in Escherichia coli. J Bacteriol 1995;177:6011
6014.
44 Lengeler JW, Vogler AP: Molecular mechanisms of
bacterial chemotaxis towards PTS carbohydrates.
FEMS Microbiol Rev 1989;5:8192.
45 Lux R, Jahreis K, Bettenbrock K, Parkinson JS,
Lengeler JW: Coupling the phosphotransferase system and the methyl-accepting chemotaxis proteindependent chemotaxis signaling pathways of
Escherichia coli. Proc Natl Acad Sci USA 1995;
92:1158311587.
46 Johnson MS, Rowsell EH, Taylor BL: Investigation
of transphosphorylation between chemotaxis proteins and the phosphoenolpyruvate:sugar phosphotransferase system. FEBS Lett 1995;374:161164.
47 Grbl G, Vogler AP, Lengeler JW: Involvement of
the histidine protein of the phosphotransferase system in chemotactic signaling of Escherichia coli
K-12. J Bacteriol 1990;172:58715876.
48 Lengeler J, Bettenbrock K, Lux R: Signal transduction
through phosphotransferase systems; in TorrianiGorini M-A, Yagil E, Silver S (eds): Phosphate in
Microorganisms. Cellular and Molecular Biology.
Washington, ASM Press, 1994, pp 182188.

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

85

49 Chauvin F, Brand L, Roseman S: Enzyme I: the first


protein and potential regulator of the bacterial
phosphoenolpyruvate:glycose phosphotransferase
system. Res Microbiol 1996;147:471479.
50 Meadow ND, Mattoo RL, Savtchenko RS, Roseman
S: Transient state kinetics of Enzyme I of the
phosphoenolpyruvate:glycose phosphotransferase
system of Escherichia coli: equilibrium and secondorder rate constants for the phosphotransfer
reactions with phosphoenolpyruvate and HPr. Biochemistry 2005;44:1279012796.
51 Dimitrova MN, Peterkofsky A, Ginsburg A:
Opposing effects of phosphoenolpyruvate and pyruvate with Mg2+ on the conformational stability and
dimerization of phosphotransferase enzyme I from
Escherichia coli. Protein Sci 2003;12:20472056.
52 Francke C PP, Westerhoff HV, Blom JG, Peletier
MA: Why the phosphotransferase system of
Escherichia coli escapes diffusion limitation. Biophys
J 2002;85:612622.
53 Sauter T, Gilles ED: Modeling and experimental
validation of the signal transduction via the
Escherichia coli sucrose phosphotransferase system.
J Biotechnol 2004;110:181199.
54 Bhm A, Boos W: Gene regulation in prokaryotes
by subcellular relocalization of transcription factors.
Curr Opin Microbiol 2004;7:151156.
55 Lee SJ, Boos W, Bouche JP, Plumbridge J: Signal
transduction between a membrane-bound transporter, PtsG, and a soluble transcription factor, Mlc,
of Escherichia coli. EMBO J 2000;19:53535361.
56 Nam TW, Cho SH, Shin D, Kim JH, Jeong JY, Lee
JH, Roe JH, Peterkofsky A, Kang SO, Ryu S, Seok
YJ: The Escherichia coli glucose transporter enzyme
IICB(Glc) recruits the global repressor Mlc. EMBO
J 2001;20:491498.
57 Tanaka Y, Kimata K, Aiba H: A novel regulatory
role of glucose transporter of Escherichia coli: membrane sequestration of a global repressor Mlc.
EMBO J 2000;19:53445352.
58 Zeppenfeld T, Larisch C, Lengeler JW, Jahreis K:
Glucose transporter mutants of Escherichia coli
K-12 with changes in substrate recognition of
IICB(Glc) and induction behavior of the ptsG gene.
J Bacteriol 2000;182:44434452.
59 Bhm A, Diez J, Diederichs K, Welte W, Boos W:
Structural model of MalK, the ABC subunit of the
maltose transporter of Escherichia coli: implications
for mal gene regulation, inducer exclusion, and subunit assembly. J Biol Chem 2002;277:37083717.
60 Sondej M, Sun J, Seok YJ, Kaback HR, Peterkofsky A:
Deduction of consensus binding sequences on proteinsthatbindIIAGlcofthephosphoenolpyruvate:sugar
phosphotransferase system by cysteine scanning
mutagenesis of Escherichia coli lactose permease. Proc
Natl Acad Sci USA 1999;96:35253530.

86

61 Gosset G, Zhang Z, Nayyar S, Cuevas WA, Saier


MH Jr: Transcriptome analysis of Crp-dependent
catabolite control of gene expression in Escherichia
coli. J Bacteriol 2004;186:35163524.
62 Park YH, Lee BR, Seok YJ, Peterkofsky A: In vitro
reconstitution of catabolite repression in Escherichia
coli. J Biol Chem 2006;281:64486454.
63 Bettenbrock K, Sauter T, Jahreis K, Kremling A,
Lengeler JW, Gilles ED: Correlation between growth
rates, EIIACrr phosphorylation, and intracellular
cyclic AMP levels in Escherichia coli K-12. J Bacteriol
2007;189:68916900.
64 Vogler AP, Lengeler JW: Indirect role of adenylate
cyclase and cyclic AMP in chemotaxis to phosphotransferase system carbohydrates in Escherichia coli
K-12. J Bacteriol 1987;169:593599.
65 Soutourina O, Kolb A, Krin E, Laurent-Winter C,
Rimsky S, Danchin A, Bertin P: Multiple control of
flagellum biosynthesis in Escherichia coli: role of
H-NS protein and the cyclic AMP-catabolite activator protein complex in transcription of the flhDC
master operon. J Bacteriol 1999;181:75007508.
66 Koo BM, Yoon MJ, Lee CR, Nam TW, Choe YJ, Jaffe
H, Peterkofsky A, Seok YJ: A novel fermentation/
respiration switch protein regulated by enzyme
IIAGlc in Escherichia coli. J Biol Chem 2004;
279:3161331621.
67 Prss BM, Campbell JW, Van Dyk TK, Zhu C,
Kogan Y, Matsumura P: FlhD/FlhC is a regulator of
anaerobic respiration and the Entner-Doudoroff
pathway through induction of the methyl-accepting
chemotaxis protein Aer. J Bacteriol 2003;185:534
543.
68 Koo BM SY: Regulation of glycogen concentration
by the histidine-containing phosphocarrier protein
HPr in Escherichia coli. J Microbiol 2001;39:2430.
69 Seok YJ, Koo BM, Sondej M, Peterkofsky A:
Regulation of E. coli glycogen phosphorylase activity by HPr. J Mol Microbiol Biotechnol 2001;3:385
393.
70 Bettenbrock K, Fischer S, Kremling A, Jahreis K,
Sauter T, Gilles ED: A quantitative approach to
catabolite repression in Escherichia coli. J Biol Chem
2006;281:25782584.
71 Rohwer JM, Meadow ND, Roseman S, Westerhoff
HV, Postma PW: Understanding glucose transport
by the bacterial phosphoenolpyruvate:glycose phosphotransferase system on the basis of kinetic measurements in vitro. J Biol Chem 2000;275:
3490934921.
72 Wang J, Gilles ED, Lengeler JW, Jahreis K: Modeling
of inducer exclusion and catabolite repression based
on a PTS-dependent sucrose and non-PTS-dependent glycerol transport systems in Escherichia coli
K-12 and its experimental verification. J Biotechnol
2001;92:133158.

Lengeler Jahreis

73 Becker AK, Zeppenfeld T, Staabl A, Seitz S, Boos W,


Morita T, Aiba H, Mahr K, Titgemeyer F, Jahreis K:
YeeI, a novel protein involved in modulation of the
activity of the glucose-phosphotransferase system
in Escherichia coli K-12. J Bacteriol 2006;188:5439
5449.
74 Van der Vlag J, vant Hof R, van Dam K, Postma PW:
Control of glucose metabolism by the enzymes of
the glucose phosphotransferase system in Salmonella
typhimurium. Eur J Biochem 1995;230:170182.

75 Nutsch T, Marwan W, Oesterhelt D, Gilles ED:


Signal processing and flagellar motor switching
during phototaxis of Halobacterium salinarum.
Genome Res 2003;13:24062412.
76 Wadler C, Vanderpool CK: A dual function for a
bacterial small RNA: SgrS performs base pairingdependent regulation and encodes a functional
polypeptide. Proc Natl Acad Sci USA 2007;104:
2045420459.

Joseph W. Lengeler
AG Genetik, Fachbereich Biologie/Chemie der Universitt Osnabrck
Barbarastrasse 11, DE49069 Osnabrck (Germany)
Tel. +49 541 969 2288, Fax +49 541 969 2293
E-Mail Lengeler@Biologie.Uni-Osnabrueck.de

Bacterial PEP-Dependent Carbohydrate: PTSs Couple Sensing and Global Control Mechanisms

87

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 88102

Correlations between Carbon Metabolism


and Virulence in Bacteria
Sandrine Ponceta Eliane Milohanica Alain Maza
Jamila Nait Abdallaha Francine Aka Mireille Larribeb
Ala-Eddine Deghmaneb Muhamed-Kheir Tahab Marie Dozotc
Xavier De Bollec Jean Jacques Letessonc Josef Deutschera
a

Laboratoire de Microbiologie et Gntique Molculaire, CNRS-AgroParisTech-INRA, Thiverval-Grignon;


Unit des Neisseria, Institut Pasteur, Paris, France, and cResearch Unit in Molecular Microbiology,
University of Namur, Namur, Belgium
b

Abstract
Bacteria have developed several mechanisms which allow the preferred utilization of the most efficiently metabolizable carbohydrates when these organisms are exposed to a mixture of carbon
sources. Interestingly, the same or similar mechanisms are used by some pathogens to control various steps of their infection process. The efficient metabolism of a carbon source might serve as signal for proper fitness. Alternatively, the presence of a specific carbon source might indicate to
bacterial cells that they thrive in infection-related organs, tissues or cells and that specific virulence
genes should be turned on or switched off. Frequently, virulence gene regulators are affected by
changes in carbon source availability. For example, expression of the gene encoding the Streptococcus
pyogenes virulence regulator Mga is controlled by the classical carbon catabolite repression (CCR)
mechanism operative in Firmicutes. The activity of PrfA, the major virulence regulator in Listeria
monocytogenes, seems to be controlled by the phosphorylation state of phosphotransferase system
(PTS) components. In Vibrio cholerae synthesis of HapR, which regulates the expression of genes
required for motility, is controlled via the Crp/cAMP CCR mechanism, whereas synthesis of Salmonella
enterica HilE, which represses genes in a pathogenicity island, is regulated by the carbohydrateCopyright 2009 S. Karger AG, Basel
responsive, PTS-controlled Mlc.

In their natural environment, bacteria are frequently exposed to major changes in


growth conditions. These changes can include temperature, salt concentrations,
supply of essential nutrients, oxygen supply, etc. In order to sustain these changes,
bacteria have developed mechanisms which allow them to adapt to drastic environmental alterations. The response to carbon source availability seems to be of special
importance, because many bacteria dispose of more than one mechanism to adapt

to changes in carbohydrate composition. These mechanisms include induction of a


specific carbohydrate transport and utilization system by the presence of the corresponding carbon source and their repression when a more efficiently utilizable carbohydrate is present in addition. The latter phenomenon is called carbon catabolite
repression (CCR) and is often mediated by more than one signal transduction pathway, which can include regulation of transcriptional activators or repressors, antiterminators, carbohydrate transporters or metabolic enzymes. In Enterobacteria
and Firmicutes, CCR is mediated via components of the phosphoenolpyruvate
(PEP):carbohydrate phosphotransferase system (PTS) which catalyses the uptake and
phosphorylation of numerous sugars and sugar derivatives [1]. The EIICs of the PTS
are integral membrane proteins catalyzing the transport step of their substrate. In
contrast, the two general PTS components EI and HPr and the carbohydrate-specific
proteins (or domains) EIIA and EIIB are in the cytoplasm or associated to the cytoplasmic side of the inner membrane and form a phosphorylation cascade by using
PEP as phosphoryl donor. Phosphorylation occurs at histidyl (EI, HPr and EIIAs as
well as EIIBs of the mannose family) or cysteyl residues (other EIIBs). Depending
on their phosphorylation state, several PTS proteins also carry out regulatory functions by either interacting with or phosphorylating their target proteins. In addition,
Firmicutes and several Proteobacteria possess a bifunctional enzyme that catalyses
the ATP-dependent phosphorylation of HPr at Ser-46 as well as the dephosphorylation of P-Ser-HPr. P-Ser-HPr exclusively serves regulatory functions. Interestingly,
the availability of carbon sources also affects the virulence of certain human, animal
and plant pathogens and it seems that mechanisms similar to those that control CCR
are used for the regulation of pathogenicity. As we will see, several steps of the complex infection process can respond to carbon source availability, including toxin production, cell adhesion, immune evasion, intestinal colonization and motility. Some
of these effects are mediated by changes in the activity of transcription regulators
controlling the expression of virulence genes.
In this review, we will briefly summarize the known mechanisms used for CCR in
bacteria and subsequently describe some cases where identical or related mechanisms
are used for the regulation of virulence.

Mechanisms Controlling CCR in Enterobacteria

CCR in Enterobacteria is mainly mediated via inducer exclusion. During the uptake
of an efficiently utilized carbon source (PTS or non-PTS), EIIAGlc has been shown
to be present mainly in the dephosphorylated form [2]. Dephospho EIIAGlc interacts with several non-PTS permeases, such as the lactose and melibiose permeases,
the MalK protein of the maltose ABC transporter, or the catabolic enzyme glycerol
kinase. In each case, only EIIAGlc, but not P~EIIAGlc interacts with the target proteins
and inhibits their activity [1]. As a consequence, the presence of an efficiently utilized

Carbon Metabolism and Virulence

89

carbon source, which lowers the PEP/pyruvate ratio and therefore the degree of phosphorylation of the PTS proteins, inhibits the uptake of lactose, melibiose, etc., and
this regulatory mechanism was therefore called inducer exclusion.
Genetic and biochemical data suggested that the phosphorylated form of EIIAGlc
plays a role in the regulation of adenylate cyclase (Cya) activity. This enzyme catalyses
the conversion of ATP into cAMP, which in Enterobacteria forms a complex with
the cAMP receptor protein Crp and allows it to bind to operator sites in front of
numerous catabolic operons thereby stimulating (or in a few instances inhibiting)
their expression [1]. However, the cAMP/Crp mechanism does not seem to be the
major mechanism of CCR, but is rather responsible for the phenomenon of diauxie.
For example, under conditions leading to the expression of the lac operon in the presence of glucose (induction with IPTG or use of a lacI mutant), glucose was still preferentially used over lactose. However, under these conditions the lag phase, which in
experiments with wild-type cells in the absence of IPTG covers the period between
glucose exhaustion and resumption of growth on lactose, had disappeared and as a
consequence also the phenomenon of diauxie [3].

Other Interaction Partners of PTS Proteins in Enterobacteria

EIIAGlc is not the only PTS component known to interact with non-PTS proteins. In
addition to EIIAGlc, several other PTS proteins of Enterobacteria have been shown to
carry out regulatory functions. For example, Escherichia coli HPr as well as P~HisHPr interact with the enzyme glycogen phosphorylase. However, although P~HisHPr has a higher affinity, only binding of HPr stimulates the activity of the enzyme
[4]. The dephosphorylated EIIBGlc domain of the glucose PTS permease interacts
with the transcription regulator Mlc and thus prevents it from binding to its DNA
targets, which includes also an operator site in front of ptsG, the gene encoding the
EIIBCGlc [1]. Mlc is also regulated by the Mlc titration factor MtfA, which controls
Mlc-mediated repression in response to yet unknown signals [5]. Finally, EI plays
a role in the chemotactic response and dephospho EI seems to negatively affect the
autokinase function of the two-component system protein CheA [6]. The P~His-HPrand P~EIIB-mediated phosphorylations of certain antiterminators and transcription
activators will be discussed in the following chapter.

Mechanisms Controlling CCR in Firmicutes

As already pointed out, CCR in Firmicutes is also mediated via PTS proteins, but the
central regulator is HPr and not EIIAGlc. In fact, HPr exists in four different forms
in Firmicutes. Like in Enterobacteria, phosphorylation at His-15 requires PEP and
is catalyzed by EI. In Firmicutes HPr becomes also phosphorylated at Ser-46 in a

90

Poncet et al.

reaction catalyzed by the bifunctional enzyme HPr kinase/P-Ser-HPr phosphorylase


(HprK/P). In addition, this enzyme catalyses the dephosphorylation of P-Ser-HP via
an unusual mechanism requiring inorganic phosphate and leading to the formation
of pyrophosphate (phosphorolysis) [7]. Firmicutes therefore usually contain HPr,
P~His-HPr, P-Ser-HPr and also doubly phosphorylated (P~His,P-Ser)-HPr. The
amount of the various forms of HPr is controlled by the presence of PTS sugars in the
medium and by intracellular metabolites. High concentrations of ATP and fructose1,6-bisphosphate (FBP), which result from the rapid metabolism of a carbon source
such as glucose, stimulate the kinase function of HprK/P, whereas high concentrations of inorganic phosphate, which occur in cells growing on less efficient carbon
sources, such as TCA cycle intermediates, favor its phosphorylase activity [7]. To
mediate CCR, P-Ser-HPr interacts with the catabolite control protein A (CcpA) and
allows this member of the LacI/GalR repressor family to bind to its DNA targets,
the so-called catabolite response elements (cre) [8], imperfect palindromes which in
Firmicutes precede most catabolic operons. In Bacillus subtilis about 10% of the genes
are controlled by this mechanism.
Several CcpA-independent CCR mechanisms exist in Firmicutes. For example,
P-Ser-HPr also mediates inducer exclusion. In lactobacilli and lactococci, maltose
uptake is inhibited by the presence of glucose and other efficiently metabolized carbohydrates. However, this inhibition disappeared when mutations were introduced in
the two lactic acid bacteria, which prevented the formation of P-Ser-HPr (Ser-46-Ala
replacement in HPr or hprK inactivation). In contrast, overproduction of P-Ser-HPr in
Lactobacillus casei owing to the Val267Phe replacement in HprK/P led to permanent
maltose transport inhibition, i.e. even in the absence of an efficient carbon source [9].
P~His-HPr mediates glycerol exclusion in Firmicutes. This PTS protein phosphorylates glycerol kinase at a conserved histidyl residue and thereby stimulates its activity.
Unphosphorylated glycerol kinase is not sufficiently active to support growth of for
example B. subtilis on media containing glycerol as the sole carbon source. When a
carbohydrate is efficiently transported and metabolized, this leads to the dephosphorylation of HPr at His-15 and the formation of P-Ser-HPr (a poor substrate for PEP
and EI). Under these conditions, glycerol kinase is barely phosphorylated and glycerol, which usually enters bacterial cells via facilitated diffusion, is therefore not efficiently converted to the inducer glycerol-3-P [10]. The uptake of lactose and raffinose
in certain lactic acid bacteria is also regulated by P~His-HPr-catalyzed phosphorylation. However, in contrast to glycerol kinase the corresponding non-PTS transporters
(LacS, RafP) carry an EIIAGlc-like domain fused to their C-terminus, which is phosphorylated by P~His-HPr [11].
P~His-HPr phosphorylates and controls also several antiterminators and transcription activators, which contain a specific domain recognized by certain PTS
proteins as phosphorylation target (PTS regulation domain, PRD). The antiterminators belong to the BglG/SacY family, whereas PRD-containing transcription activators can be either of the NifA/NtrC or DeoR family [1]. Again, P~His-HPr-mediated

Carbon Metabolism and Virulence

91

phosphorylation stimulates the activity of these transcription regulators and as a consequence the efficient utilization of a carbon source prevents the expression of the
corresponding genes, which usually encode PTS proteins. This CCR mechanism is
CcpA-independent [12], but nevertheless requires the formation of P-Ser-HPr. This is
due to the fact that P-Ser-HPr is a poor substrate for the phosphorylation by PEP and
EI and only when sufficient HPr is converted into P-Ser-HPr, P~His-HPr-mediated
phosphorylation of these transcription regulators is significantly slowed [13]. The
PRDs can also be the targets of phosphorylation by P~EIIB domains of the PTS. This
phosphorylation has a negative effect on the activity of the transcription regulators
and serves for induction of the corresponding PTS operon. When the PTS substrate is
absent, the EIIB will be mainly present in the phospho-form and phosphorylate and
inactivate the corresponding transcription regulator. When the PTS substrate is present, the P~EIIB will transfer its phosphoryl group mainly to the carbohydrate and not
to the transcription regulator, which therefore remains active.
In B. subtilis, HPr and P-Ser-HPr have been shown to interact with the enzyme
glyceraldehyde-3-P dehydrogenase (GapA). However, only P-Ser-HPr was found to
inhibit the activity of the enzyme [14]. Similarly, HPr and P~His-HPr interact with
and regulate the transcription activator YesS of the AraC/XylS family [S. Poncet, S.
Brinster, M. Soret, P. Mervelet, J. Deutscher, P. Noirot, unpubl results], which controls
the expression of an operon encoding the enzymes for rhamnogalacturonan degradation [15]. In this case only P~His-HPr seems to stimulate YesS activity.

Carbohydrate Availability and Virulence in Bacteria

Early reports about the effects of carbon source availability on the virulence of bacterial pathogens date back to the beginning of the second half of the last century. For
example, a few years after the discovery of the PTS [16], enterotoxin B production by
Staphylococcus aureus was reported to be submitted to CCR [17] and many similar
reports followed [1822]. The molecular details of the mechanisms mediating these
effects were of course not understood at that time and it was only much later that for
example carbohydrate-mediated activation [23] or inhibition [24, 25] of specific transcription regulators controlling the expression of usually numerous virulence genes
was discovered. In the following chapters we will discuss some of the well-studied
examples of carbon source-mediated virulence control, first in Firmicutes and subsequently in Proteobacteria.

CCR and Virulence in Firmicutes

About three decades ago it was reported that addition of glucose, sucrose or trehalose to a semisynthetic growth medium stimulates the synthesis of the Streptococcus

92

Poncet et al.

pyogenes M protein [22]. The M protein, which is encoded by the emm gene, is a
cell-wall-associated essential virulence determinant that provides resistance to
phagocytosis and is required for adherence to various host tissues and extracellular
matrix components [26]. Several avirulent S. pyogenes mutants no longer producing the M protein were isolated and found to contain deletions just upstream from
emm [27] in a gene encoding a response regulator of two-component systems. This
gene was called mga for multiple gene regulator of group A Streptococcus. Mga is
required not only for its own synthesis and that of the M protein, but also for other
proteins mediating cell adhesion, cell invasion and immune evasion [28]. Because
expression of mga, which occurs from the two promoters P1 and P2, responds to
environmental signals, it was hypothesized that these promoters might also be the
target of P-Ser-HPr/CcpA-mediated regulation. In fact, P-Ser-HPr/CcpA can exert
carbon catabolite activation (CCA) when the cre site is located about 10 bp upstream
from the -35 promoter sequence (fig. 1) [1]. Indeed, an almost perfect cre site precedes the P1 promoter and CcpA has recently been shown to bind to this operator
[23]. The requirement for P-Ser-HPr as co-repressor was not tested, but it is likely
that regulation of mga expression follows the classical CCA mechanism outlined in
figure 1. Deletion of the cre site or inactivation of ccpA specifically diminished the
synthesis of the mga transcript from the P1 promoter [23]. In another recent paper, a
comparison of the transcriptome of a S. pyogenes wild-type and a ccpA mutant strain
grown in nutrient-rich medium to early or late exponential phase has been carried
out. In agreement with the above results it revealed that CcpA of S. pyogenes controls not only the expression of numerous genes related to carbon metabolism, but
also many genes encoding well-characterized virulence factors such as the cytolysin
streptolysin S encoded by the sag operon [29] and CcpA was shown to bind to sagAp
[30]. Interestingly, neither mga nor emm expression was altered in the ccpA mutant.
This might be due to mga expression from the second promoter and the effect of
ccpA inactivation on P1 activity might only become detectable under specific growth
conditions and/or during specific steps of the infection process. Nevertheless, this
suggests that the effects of the ccpA mutation on virulence genes observed in the transcriptome analysis are Mga-independent and that the CcpA effects are either direct or
mediated by other virulence regulators. Interestingly, the S. pyogenes speB gene, which
is strongly expressed in tissue during infection, is repressed by a CcpA-independent
mechanism mediated by an aldolase encoded by the lacD.1 gene, which is associated
with an operon encoding components of a lactose class PTS [31]. Substrate binding
was proposed to control the repressor activity of the aldolase. A PTS specific for maltose and maltotriose was reported to be important for growth on saliva and for colonization of the oropharynx. Mutants affected in maltose/maltotriose utilization were
attenuated in both activities [32]. Finally, many S. pyogenes strains contain a protein
resembling multiple sugar metabolism regulators (MsmR) from other bacteria. In S.
pyogenes, the gene encoding MsmR is located in a region containing genes for virulence factors, mostly surface proteins, and some of these genes are indeed controlled

Carbon Metabolism and Virulence

93

hly

10

35

+
?

PrfA box
+

PrfA
PEP

OUT

Sug
EIIASug

P~HPr

EI

IN

P~EIIBSug
EIICGlc

Pyruvate

P~EIIASug

HPr

P~EI

EIIBSug

HprK/P
Sug6-P

P-Ser-HPr

CcpA
?
+
cre

35 10
P1

mga
P2

membrane

Fig. 1. CCR-related regulation of virulence genes in Firmicutes. Shown are the PTS phosphorylation
cascade and PTS-mediated sugar uptake (the P-proteins are written in grey letters to indicate that
during sugar uptake they are present in low amounts) and the regulation of S. pyogenes mga expression (stimulated by rapidly metabolizable sugars) and of L. monocytogenes PrfA activity (inhibited by
rapidly metabolizable sugars). Expression of mga is controlled by the classical CCA mechanism operative in Firmicutes. The uptake of sugars stimulates the formation of P-Ser-HPr, which interacts with
CcpA and allows this member of the LacI/GalR family to bind to an operator site located upstream
from the mga promoter P1 (the involvement of P-Ser-HPr has not yet been experimentally proven).
In contrast, the activity of PrfA seems to be regulated by an EII component of yet unknown sugar
specificity. According to the present model, PrfA activity is either inhibited by one or more dephospho EIIAs or EIIBs () or stimulated by P~EIIA or P~EIIB (+), where the P-protein could either interact
with or phosphorylate PrfA.

by MsmR [33]. However, a correlation between sugar metabolism and MsmR activity
and virulence has so far not been established.
In contrast to S. pyogenes, the Listeria monocytogenes transcription activator PrfA,
which regulates the expression of numerous virulence genes, was inhibited when this
organism was exposed to efficiently utilized carbohydrates such as glucose, fructose,
maltose or cellobiose, but not by less favorable sugars such as galactose [24, 25]. It
was therefore assumed that prfA expression might be controlled via P-Ser-HPr/CcpAmediated CCR. However, inactivation of ccpA did not relieve the repressive effect of the
above carbon sources on PrfA activity [34]. In a heterologous B. subtilis system (lacZ
expressed from a PrfA-regulated promoter), overproduction of P-Ser-HPr (more than
95% of the HPr converted to P-Ser-HPr owing to the V267F hprK mutation) was found
to strongly inhibit PrfA activity [35]. However, P-Ser-HPr does not seem to directly
affect PrfA activity. P-Ser-HPr is a poor substrate for PEP-dependent phosphorylation

94

Poncet et al.

and the phosphoryl group transfer via the PTS phosphorylation cascade is drastically slowed when most of the HPr is converted into P-Ser-HPr [1]. However, to be
active PrfA probably needs a fully functional PTS, because inactivation of EI or HPr
in the heterologous B. subtilis system also significantly lowered PrfA activity [36].
Preliminary results with a L. monocytogenes ptsI mutant support the concept that PrfA
activity requires a functional PTS [E. Milohanic and J. Deutscher, unpubl. results]. In
contradiction with the above results, expression of PrfA-controlled genes was reported
to be slightly elevated in a L. monocytogenes ptsH mutant [37]. However, these experiments were carried out with cells grown in glucose-containing BHI medium, which
leads to PrfA inhibition. The ptsH mutant cannot take up glucose and this might be
responsible for the slightly elevated PrfA activity observed for the mutant strain. The
commonly accepted concept of PrfA regulation predicts that one or several sugar-specific EIIA or EIIB components of the PTS interact with PrfA and stimulate (P~EIIs)
or inhibit (EIIs) its activity (fig. 1). In addition to PTS components, the L. monocytogenes response regulator ResD also controls virulence gene expression in response to
carbohydrate availability. In resD mutants, efficiently metabolized carbohydrates had
lost their repressive effect on the expression of several PrfA-controlled virulence genes
[38]. This effect was suggested to be related to the observed low expression of the mpt
operon in the resD mutant, which probably leads to low levels of EIIABMan/Glc (MptA).
In agreement with this concept, preliminary experiments in our laboratory revealed
that disruption of the gene encoding EIIABMan/Glc (mptA) led to 3- to 5-fold elevated
PrfA activity (expression of the gus reporter gene fused to the hly promoter) in cells
grown to stationary phase in LB medium in the absence of a repressing sugar [F. Ak,
E. Milohanic, J. Deutscher, unpubl. results] (fig. 1). Finally, the utilization of glycerol as
carbon source was reported to increase the expression of PrfA-controlled genes [39].
As already mentioned, although glycerol is not transported via the PTS, its utilization
depends on functional EI and HPr. In firmicutes, P-His-HPr phosphorylates glycerol
kinase at a conserved histidyl residue and thereby stimulates its activity about 10-fold
[1]. Indeed, in glycerol-grown firmicutes the major fraction of glycerol kinase is present in the phosphorylated form [40]. Accordingly, L. monocytogenes cells grown on
glycerol contain mainly P~His-HPr [39] and consequently the EIIAs and EIIBs are
probably also primarily present in the phospho form. These results suggest again a correlation between the phosphorylation state of PTS proteins and PrfA activity.
In several pathogenic clostridial species, CcpA was also reported to control the
expression of virulence genes. While it is not clear whether glucose repression of
toxin A and B synthesis in Clostridium difficile is mediated via P-Ser-HPr/CcpA or via
the alternative factor TexR [41], CcpA of Clostridium perfringens was reported to be
necessary for regulation of -toxin (collagenase) production and efficient sporulation
[42] and to cause repression of type IV pilus-dependent gliding motility [43].
In Streptococcus pneumoniae, inactivation of ccpA affects the synthesis of capsular polysaccharides and the ccpA mutant exhibits diminished virulence [44]. CcpA
appears to be necessary for efficient colonization of the nasopharynx and survival

Carbon Metabolism and Virulence

95

and multiplication in the lung [45]. One of the bacterial components important for
the colonization of the nasopharynx is the surface-attached -galactosidase BgaA, the
synthesis of which is controlled by CcpA [46]. Finally, a lactose/cellobiose and a mannose type PTS of S. pneumoniae were found to be specifically expressed during infection in mouse lung tissues [47].

CCR and Virulence in Proteobacteria without HprK/P

A correlation between carbon metabolism and virulence was shown for Vibrio cholerae
in an early publication reporting that the presence of glucose affects toxin production
by this organism [20]. In more recent studies the Crp/cAMP regulation mechanism
was confirmed to control toxin production [48] and was additionally shown to affect
quorum sensing, motility and intestinal colonization [49], characteristics which all contribute to virulence. Inactivation of crp affects several virulence-related genes including those required for motility, the ompT, ompU and ompW genes, which code for
outer membrane proteins, rpoE, which encodes the alternative E factor necessary for
intestinal colonization and the vpsR gene, which controls genes required for exopolysaccharide production. Part of these effects is probably mediated via HapR, a pivotal
regulator of genes involved in virulence and biofilm formation. Expression of hapR is
controlled by the Crp/cAMP CCR mechanism (fig. 2). This mechanism was also proposed to be responsible for the regulation of leukotoxin production in the pathogen
Aggregatibacter actinomycetemcomitans, which causes aggressive periodontitis. A ptsH
mutant was reported to produce much less leukotoxin and to contain significantly
lower cAMP levels. Adding exogenous cAMP could restore normal leukotoxin production [50]. Interestingly, in V. cholerae the PTS also plays a role in biofilm-associated
growth. Biofilm-associated cells with a functional PTS exhibited reduced growth and
this effect disappeared when the EI-encoding ptsI gene was disrupted [51].
In Salmonella enterica serovar Typhimurium the global PTS-controlled transcription regulator Mlc (see above: Other Interaction Partners of PTS Proteins in
Enterobacteria) was found to be necessary for full expression of several genes (hilA,
hilD and invF) located in the pathogenicity island I. In fact, Mlc represses expression
of the hilE gene from its promoter P3 (fig. 2). HilE in turn represses the expression
of genes in the pathogenicity island I. An mlc mutant therefore exhibited elevated
hilE expression and consequently reduced levels of invasion and cytotoxicity [52].
In a murine model, a 100- to 1,000-fold attenuation of virulence was observed for
a Salmonella typhimurium ptsI mutant [53]. A similar attenuation was reported for
Salmonella choleraesuis mutants defective in adenylate cyclase and Crp after oral
administration to mice and pigs [54, 55]. Because the ptsI mutation prevents the formation of P~EIIAGlc, the activator of adenylate cyclase (fig. 2), it is tempting to assume
that the effect of the ptsI mutation on virulence might also be mediated via the Crp/
cAMP regulation mechanism.

96

Poncet et al.

IN

OUT
Glc

PEP

EI

EIIASug

P~HPr

P~EIIBGlc
EIICGlc

P~EI

Pyruvate

P~EIIAGlc

HPr

EIIBGlc

+
Crp
+

35

cAMP

10

Cya

hapR

Sug6-P

Mlc
ATP

35

10

hilE
membrane

Fig. 2. CCR-related regulation of virulence genes in Proteobacteria. Synthesis of the V. cholerae virulence regulator HapR is controlled by one of the CCR mechanisms operative in Proteobacteria.
P~EIIAGlc, which is mainly formed when no efficiently metabolizable carbon source is present, stimulates the Cya-catalyzed formation of cAMP and the Crp/cAMP complex probably binds to an operator site upstream from the hapR gene to stimulate its expression. In contrast, expression of the S.
enterica hilE gene, which codes for a repressor of genes located in a pathogenicity island, is repressed
by Mlc. Mlc of Enterobacteria has been shown to be sequestered to the membrane (and thereby
inactivated) by the dephospho EIIBGlc domain (which is fused to EIICGlc), which prevails during transport of glucose or other PTS sugars. Under these conditions, HilE is synthesized and represses the
genes in the pathogenicity island. This model is in agreement with the observation that the metabolism of glucose or mannose represses expression of the genes in the pathogenicity island [52].

Erwinia amylovora is the causative agent of fire blight in rosaceous plants and the
utilization of sucrose seems to be a prerequisite for successful infection of the plant
cells. Sucrose is transported via a PTS in this organism and mutants unable to transport
sucrose had lost the capacity to produce fire blight symptoms on apple seedlings [56].
A homologue of the carbon storage regulation system was recently reported to
indirectly control expression of the Yersinia pseudotuberculosis virulence regulator
RovA. While overexpression of the small RNAs encoded by the csrC- and csrB-like
genes activated rovA expression, a CsrA-like protein repressed RovA synthesis [57].

CCR and Virulence in Proteobacteria Containing HprK/P

Several bacteria (mainly Proteobacteria) possess paralogues of the common PTS


proteins EI and HPr and usually one or two sugar-specific EIIAs. Some of them are
involved in the transport of specific carbohydrates (FPr of E. coli specifically serves
for fructose transport), but for most of them their detailed function is not known. It

Carbon Metabolism and Virulence

97

seems that many of them carry out exclusively regulatory functions as they are lacking EIIB and EIIC components for sugar transport. For example, the ptsO, ptsP and
ptsN genes of Enterobacteria encode HPr, EI and EIIAFru paralogues, which form a
separate PTS phosphorylation cascade (although cross-reactions with the main PTS
proteins exist), which plays a role in the regulation of nitrogen metabolism in E. coli.
In Pseudomonas putida, similar PTS proteins control the degradation of toluene and
other aromatic compounds [1] and the utilization of dimethyl sulfone [58]. In addition, E. coli PtsN was reported to interact with and to control TrkA, the ATP-binding
domain of a K+ ABC transporter [59]. In Enterobacteria, ptsN is part of the RpoE
(E) regulon and overproduction of PtsN was recently reported to suppress the lethal
effect of rpoE inactivation [60]. Interestingly, several pathogenic Proteobacteria possess not only a truncated PTS composed of EI and HPr paralogues and either an
EIIAFru- or EIIAMan-like component, but also HprK/P. It seems that HprK/P and the
PTS components of these pathogens control certain steps of the infection process.
In Proteobacteria possessing an HprK/P, two principal gene organizations can be
observed [61]. In -Proteobacteria, the hprK, ptsM (encoding an EIIA of the mannose
PTS family) and ptsO genes are organized together with genes encoding a two-component system, which has been shown to be essential for the virulence of for example
Agrobacterium tumefaciens [62] or Brucella melitensis [63]. For B. melitensis we could
show that the genes for the two-component system BvrR/BvrS and the pts genes are
organized in an operon [M. Dozot, X. de Bolle, J.J. Letesson, S. Poncet, J. Deutscher,
unpubl. results]. In - and - Proteobacteria possessing an HprK/P, the hprK gene and
ptsN or ptsM and sometimes also ptsO and ptsP are located next to rpoN or rpoNrelated genes, such as yhbJ. Interestingly, pathogenic Neisseria (Neisseria meningitidis
and Neisseria gonorrhoeae) were reported to possess an inactivated, truncated RpoN
[64]. In the case of N. meningitidis, we could establish that ptsN, hprK and yhbJ are
organized in an operon [S. Poncet, J. Nait Abdallah, M.-K. Taha, M. Larribe, A.-E.
Deghmane, J. Deutscher, unpubl. results]. In B. melitensis as well as in N. meningitidis, the incomplete PTS phosphorylation cascade is functional and HprK/P is also
capable of phosphorylating HPr at Ser-46.
Especially the organization of hprK and some pts genes in an operon together
with the genes for a two-component system essential for the virulence of pathogenic
-Proteobacteria suggested that the PTS might also be involved in virulence regulation, possibly via a cross-talk between PTS proteins and the two-component system.
For example, Brucella abortus BvrR/BvrS was proposed to interact with PTS components in order to directly or indirectly adjust the metabolism of Brucella during transit
into the intracellular niche [65]. Moreover, in a large-scale screen for Brucella virulence factors, ptsN [66] and ptsH [67] were among the genes that led to an attenuated
phenotype when inactivated. In addition, some of the B. melitensis pts mutants we
constructed were found to exhibit strongly reduced synthesis of VirB proteins, which
form a type IV secretion system also essential for virulence [M. Dozot, X. de Bolle,
J.J. Letesson, unpubl. results]. In B. melitensis, synthesis of VirB proteins is controlled

98

Poncet et al.

by environmental signals [68] and it is therefore not surprising that their synthesis is
also influenced by carbon source availability. Similarly, a N. meningitidis hprK mutant
exhibited significantly reduced cell adhesion [61], which seems to be related to the
enhanced synthesis of polysialic acid-containing capsule, which was also observed
for this mutant [S. Poncet, M.-K. Taha, M. Larribe, A.-E. Deghmane, J. Deutscher,
unpubl. results].
Finally, ptsP inactivation in Legionella pneumophila [69] or Pseudomonas aeruginosa [70] also led to an almost complete loss of their virulence. The mutant strains grew
very slowly in animal models or on human epithelial cells. Trans-complementation
restored the virulence, thus establishing that the loss of virulence is related to the
absence of a functional PTS phosphorylation cascade.

Conclusions

Seen the importance of carbon metabolism in bacteria for the gain of energy and
the synthesis of building blocks for all different kinds of cellular processes, it is not
surprising that virulence of bacterial pathogens also responds to carbon source availability. However, it is astonishing that in most of the above-presented examples the
correlation between carbon source availability and bacterial virulence is based on
mechanisms identical or similar to those controlling carbon catabolite repression.

References
1 Deutscher J, Francke C, Postma PW: How phosphotransferase system-related protein posphorylation
regulates carbohydrate metabolism in bacteria.
Microbiol Mol Biol Rev 2006;70:9391031.
2 Hogema BM, Arents JC, Bader R, Eijkemans K,
Yoshida H, Takahashi H, Aiba H, Postma PW:
Inducer exclusion in Escherichia coli by non-PTS
substrates: the role of the PEP to pyruvate ratio in
determining the phosphorylation state of enzyme
IIAGlc. Mol Microbiol 1998;30:487498.
3 Kimata K, Takahashi H, Inada T, Postma P, Aiba H:
cAMP receptor protein-cAMP plays a crucial role
in glucose-lactose diauxie by activating the glucose
transporter gene in Escherichia coli. Proc Natl Acad
Sci USA 1997;94:1291412919.
4 Koo BM, Seok Y-J: Regulation of glycogen concentration by the histidine-containing phosphocarrier
protein HPr in Escherichia coli. J Microbiol
2001;39:2430.

Carbon Metabolism and Virulence

5 Becker AK, Zeppenfeld T, Staab A, Seitz S, Boos W,


Morita T, Aiba H, Mahr K, Titgemeyer F, Jahreis K:
YeeI, a novel protein involved in modulation of the
activity of the glucose-phosphotransferase system
in Escherichia coli K-12. J Bacteriol 2006;188:5439
5449.
6 Lux R, Jahreis K, Bettenbrock K, Parkinson JS,
Lengeler JW: Coupling of the phosphotransferase
system and the methyl-accepting chemotaxis protein-dependent chemotaxis signaling pathways of
Escherichia coli. Proc Natl Acad Sci USA 1995;92:
1158311587.
7 Mijakovic I, Poncet S, Galinier A, Monedero V,
Fieulaine S, Janin J, Nessler S, Marquez JA, Scheffzek
K, Hasenbein S, Hengstenberg W, Deutscher J:
Pyrophosphate-producing protein dephosphorylation by HPr kinase/phosphorylase: a relic of early
life? Proc Natl Acad Sci USA 2002;99:1344213447.
8 Fujita Y, Miwa Y, Galinier A, Deutscher J: Specific
recognition of the Bacillus subtilis gnt cis-acting
catabolite-responsive element by a protein complex
formed between CcpA and seryl-phosphorylated
HPr. Mol Microbiol 1995;17:953960.

99

9 Monedero V, Yebra MJ, Poncet S, Deutscher J:


Maltose transport in Lactobacillus casei and its regulation by inducer exclusion. Res Microbiol 2008;
159:94102.
10 Darbon E, Servant P, Poncet S, Deutscher J:
Antitermination by GlpP, catabolite repression via
CcpA, and inducer exclusion elicited by P~GlpK
dephosphorylation control Bacillus subtilis glpFK
expression. Mol Microbiol 2002;43:10391052.
11 Geertsma ER, Duurkens RH, Poolman B: The activity of the lactose transporter from Streptococcus
thermophilus is increased by phosphorylated IIA
and the action of -galactosidase. Biochemistry
2005; 44:1588915897.
12 Lindner C, Hecker M, Le Coq D, Deutscher J:
Bacillus subtilis mutant LicT antiterminators exhibiting enzyme I- and HPr-independent antitermination affect catabolite repression of the bglPH operon.
J Bacteriol 2002;184:48194828.
13 Krger S, Gertz S, Hecker M: Transcriptional analysis of bglPH expression in Bacillus subtilis: evidence
for two distinct pathways mediating carbon catabolite repression. J Bacteriol 1996;178:26372644.
14 Pompeo F, Luciano J, Galinier A: Interaction of
GapA with HPr and its homologue, Crh: novel levels of regulation of a key step of glycolysis in Bacillus
subtilis? J Bacteriol 2007;189:11541157.
15 Ochiai A, Itoh T, Kawamata A, Hashimoto W,
Murata K: Plant cell wall degradation by saprophytic
Bacillus subtilis strains: gene clusters responsible
for rhamnogalacturonan depolymerization. Appl
Environ Microbiol 2007;73:38033813.
16 Kundig W, Ghosh S, Roseman S: Phosphate bound
to histidine in a protein as an intermediate in a
novel phospho-transferase system. Proc Natl Acad
Sci USA 1964;52:10671074.
17 Morse SA, Mah RA, Dobrogosz WJ: Regulation of
staphylococcal enterotoxin B. J Bacteriol 1969;98:4
9.
18 Forget A, Paquette G, Roy A, Fredette V: Correlations
between virulence and other characters of Clostridium
perfringens type A. Appl Microbiol 1969;18:668676.
19 De Repentigny J, Mathieu LG: A comparison of use
of carbohydrate, cell structure, and toxin formation
in naturally occurring and thymine-requiring
Staphylococcus aureus strains. Can J Microbiol
1971;17:13431350.
20 Callahan LT III, Richardson SH: Biochemistry of
Vibrio cholerae virulence. III. Nutritional requirements for toxin production and the effects of pH on
toxin elaboration in chemically defined media.
Infect Immun 1973;7:567572.

100

21 Stinson RS, Talburt DE: Antigenic variation and


increase in pathogenicity in Pseudomonas aeruginosa as a result of growth on glucose or N-hexadecane
as sole carbon source. Can J Microbiol 1978;24:675
679.
22 Pine L, Reeves MW: Regulation of the synthesis of
M protein by sugars, Todd Hewitt broth, and horse
serum, in growing cells of Streptococcus pyogenes.
Microbios 1978;21:185212.
23 Almengor AC, Kinkel TL, Day SJ, McIver KS: The
catabolite control protein CcpA binds to Pmga and
influences expression of the virulence regulator
Mga in the group A Streptococcus. J Bacteriol
2007;189:84058416.
24 Park SF, Kroll RG: Expression of listeriolysin and
phosphatidylinositol-specific phospholipase C is
repressed by the plant-derived molecule cellobiose
in Listeria monocytogenes. Mol Microbiol 1993;
8:653661.
25 Milenbachs AA, Brown DP, Moors M, Youngman P:
Carbon-source regulation of virulence gene expression in Listeria monocytogenes. Mol Microbiol
1997;23:10751085.
26 Navarre WW, Schneewind O: Surface proteins of
gram-positive bacteria and mechanisms of their targeting to the cell wall envelope. Microbiol Mol Biol
Rev 1999;63:174229.
27 Spanier JG, Jones SJ, Cleary PP: Small DNA deletions creating avirulence in Streptococcus pyogenes.
Science 1984;225:935938.
28 Hondorp ER, McIver KS: The Mga virulence regulon: infection where the grass is greener. Mol
Microbiol 2007;66:10561065.
29 Shelburne SA III, Keith D, Horstmann N, Sumby P,
Davenport MT, Graviss EA, Brennan RG, Musser
JM: A direct link between carbohydrate utilization
and virulence in the major human pathogen group
A Streptococcus. Proc Natl Acad Sci USA 2008;
105:16981703.
30 Kinkel TL, McIver KS: CcpA-mediated repression
of streptolysin S expression and virulence in the
group A Streptococcus. Infect Immun 2008;76:3451
3463.
31 Loughman JA, Caparon MG: A novel adaptation of
aldolase regulates virulence in Streptococcus pyogenes. EMBO J 2006;25:54145422.
32 Shelburne SA III, Keith DB, Davenport MT,
Horstmann N, Brennan RG, Musser JM: Molecular
characterization of group A Streptococcus maltodextrin catabolism and its role in pharyngitis. Mol
Microbiol 2008;69:436452.
33 Nakata M, Podbielski A, Kreikemeyer B: MsmR, a
specific positive regulator of the Streptococcus pyogenes FCT pathogenicity region and cytolysinmediated translocation system genes. Mol Microbiol
2005;57:786803.

Poncet et al.

34 Behari J, Youngman P: A homolog of CcpA mediates catabolite control in Listeria monocytogenes but
not carbon source regulation of virulence genes. J
Bacteriol 1998;180:63166324.
35 Deutscher J, Herro R, Bourand A, Mijakovic I,
Poncet S: P-Ser-HPr a link between carbon metabolism and the virulence of some pathogenic bacteria. Biochim Biophys Acta 2005;1754:118125.
36 Herro R, Poncet S, Cossart P, Buchrieser C, Gouin
E, Glaser P, Deutscher J: How seryl-phosphorylated
HPr inhibits PrfA, a transcription activator of
Listeria monocytogenes virulence genes. J Mol
Microbiol Biotechnol 2005;9:224234.
37 Mertins S, Joseph B, Goetz M, Ecke R, Seidel G,
Sprehe M, Hillen W, Goebel W, Mller-Altrock S:
Interference of components of the phosphoenolpyruvate phosphotransferase system with the
central virulence gene regulator PrfA of Listeria
monocytogenes. J Bacteriol 2007;189:473490.
38 Larsen MH, Kallipolitis BH, Christiansen JK, Olsen
JE, Ingmer H: The response regulator ResD modulates virulence gene expression in response to carbohydrates in Listeria monocytogenes. Mol Microbiol
2006;61:16221635.
39 Joseph B, Mertins S, Stoll R, Schr J, Umesha KR,
Luo Q, Mller-Altrock S, Goebel W: Glycerol
metabolism and PrfA activity in Listeria monocytogenes. J Bacteriol 2008;190:54125430.
40 Deutscher J, Bauer B, Sauerwald H: Regulation of
glycerol metabolism in Enterococcus faecalis by
phosphoenolpyruvate-dependent phosphorylation
of glycerol kinase catalyzed by enzyme I and HPr of
the phosphotransferase system. J Bacteriol 1993;
175:37303733.
41 Dupuy B, Sonenshein AL: Regulated transcription
of Clostridium difficile toxin genes. Mol Microbiol
1998;27:107120.
42 Varga J, Stirewalt VL, Melville SB: The CcpA protein is necessary for efficient sporulation and
enterotoxin gene (cpe) regulation in Clostridium
perfringens. J Bacteriol 2004;186:52215229.
43 Mendez M, Huang IH, Ohtani K, Grau R, Shimizu
T, Sarker MR: Carbon catabolite repression of type
IV pilus-dependent gliding motility in the anaerobic pathogen Clostridium perfringens. J Bacteriol
2008;190:4860.
44 Giammarinaro P, Paton JC: Role of RegM, a homologue of the catabolite repressor protein CcpA, in
the virulence of Streptococcus pneumoniae. Infect
Immun 2002;70:54545461.
45 Iyer R, Baliga NS, Camilli A: Catabolite control
protein A (CcpA) contributes to virulence and regulation of sugar metabolism in Streptococcus pneumoniae. J Bacteriol 2005;187:83408349.

Carbon Metabolism and Virulence

46 Kaufman GE, Yother J: CcpA-dependent and -independent control of -galactosidase expression in


Streptococcus pneumoniae occurs via regulation of
an upstream phosphotransferase system-encoding
operon. J Bacteriol 2007;189:51835192.
47 Meng J-P, Yin Y-B, Zhang X-M, Huang Y-S, Lan K,
Cui F, Xu S-X: Identification of Streptococcus pneumoniae genes specifically induced in mouse lung
tissues. Can J Microbiol 2008;54:5865.
48 Liang W, Pascual-Montano A, Silva AJ, Benitez JA:
The cyclic AMP receptor protein modulates quorum sensing, motility and multiple genes that affect
intestinal colonization in Vibrio cholerae. Microbiology 2007;153:29642975.
49 Liang W, Silva AJ, Benitez JA: The cyclic AMP
receptor protein modulates colonial morphology in
Vibrio cholerae. Appl Environ Microbiol 2007;73:
74827487.
50 Isaza MP, Duncan MS, Kaplan JB, Kachlany SC:
Screen for leukotoxin mutants in Aggregatibacter
actinomycetemcomitans: genes of the phosphotransferase system are required for leukotoxin biosynthesis. Infect Immun 2008;76:35613568.
51 Houot L, Watnick PI: A novel role for enzyme I of
the Vibrio cholerae phosphoenolpyruvate phosphotransferase system in regulation of growth in a biofilm. J Bacteriol 2008;190:311320.
52 Lim S, Yun J, Yoon H, Park C, Kim B, Jeon B, Kim D,
Ryu S: Mlc regulation of Salmonella pathogenicity
island I gene expression via hilE repression. Nucleic
Acids Res 2007;35:18221832.
53 Kok M, Bron G, Erni B, Mukhija S: Effect of enzyme
I of the bacterial phosphoenolpyruvate: sugar phosphotransferase system on virulence in a murine
model. Microbiology 2003;149:26452652.
54 Stabel TJ, Mayfield JE, Morfitt DC, Wannemuehler
MJ: Oral immunization of mice and swine with an
attenuated Salmonella choleraesuis [cya-12 (crpcdt)19] mutant containing a recombinant plasmid.
Infect Immun 1993;61:610618.
55 Kennedy MJ, Yancey RJ Jr, Sanchez MS, Rzepkowski
RA, Kelly SM, Curtiss III R: Attenuation and immunogenicity of cya crp derivatives of Salmonella
choleraesuis in pigs. Infect Immun 1999;67:4628
4636.
56 Bogs J, Geider K: Molecular analysis of sucrose
metabolism of Erwinia amylovora and influence on
bacterial virulence. J Bacteriol 2000;182:53515358.
57 Heroven AK, Bhme K, Rohde M, Dersch P: A Csrtype regulatory system, including small non-coding
RNAs, regulates the global virulence regulator RovA
of Yersinia pseudotuberculosis through RovM. Mol
Microbiol 2008;68:11791195.

101

58 Kouzuma A, Endoh T, Omori T, Nojiri H, Yamane


H, Habe H: The ptsP gene encoding the PTS family
protein EINtr is essential for dimethyl sulfone utilization by Pseudomonas putida. FEMS Microbiol Lett
2007;275:175181.
59 Lee CR, Cho SH, Yoon MJ, Peterkofsky A, Seok YJ:
Escherichia coli enzyme IIANtr regulates the K+
transporter TrkA. Proc Natl Acad Sci USA 2007;
104:41244129.
60 Hayden JD, Ades SE: The extracytoplasmic stress
factor, E, is required to maintain cell envelope
integrity in Escherichia coli. PLoS ONE 2008;3:
e1573.
61 Bol G, Mijakovic I, Maz A, Poncet S, Taha M-K,
Larribe M, Darbon E, Khemiri A, Galinier A,
Deutscher J: Transcription regulators potentially
controlled by HPr kinase/phosphorylase in Gramnegative bacteria. J Mol Microbiol Biotechnol 2003;
5:206215.
62 Charles TC, Nester EW: A chromosomally encoded
two-component sensory transduction system is
required for virulence of Agrobacterium tumefaciens. J Bacteriol 1993;175:66146625.
63 Sola-Landa A, Pizarro-Cerda J, Grillo MJ, Moreno
E, Moriyon I, Blasco JM, Gorvel JP, Lopez-Goni I: A
two-component regulatory system playing a critical
role in plant pathogens and endosymbionts is present in Brucella abortus and controls cell invasion
and virulence. Mol Microbiol 1998;29:125138.
64 Laskos L, Dillard JP, Seifert HS, Fyfe JA, Davies JK:
The pathogenic neisseriae contain an inactive rpoN
gene and do not utilize the pilE 54 promoter. Gene
1998;208:95102.

65 Lamontagne J, Butler H, Chaves-Olarte E, Hunter J,


Schirm M, Paquet C, Tian M, Kearney P, Hamaidi L,
Chelsky D, Moriyn I, Moreno E, Paramithiotis E:
Extensive cell envelope modulation is associated
with virulence in Brucella abortus. J Proteome Res
2007;6:15191529.
66 Delrue RM, Lestrate P, Tibor A, Letesson JJ, De
Bolle X: Brucella pathogenesis, genes identified
from random large-scale screens. FEMS Microbiol
Lett 2004;231:112.
67 Wu Q, Pei J, Turse C, Ficht TA: Mariner mutagenesis
of Brucella melitensis reveals genes with previously
uncharacterized roles in virulence and survival.
BMC Microbiol 2006;6:102.
68 Dozot M, Boigegrain RA, Delrue RM, Hallez R,
Ouahrani-Bettache S, Danese I, Letesson JJ, De
Bolle X, Khler S: The stringent response mediator
Rsh is required for Brucella melitensis and Brucella
suis virulence, and for expression of the type IV
secretion system virB. Cell Microbiol 2006;8:1791
1802.
69 Higa F, Edelstein PH: Potential virulence role of the
Legionella pneumophila ptsP ortholog. Infect Immun
2001;69:47824789.
70 Tan M-W, Rahme LG, Sternberg JA, Tompkins RG,
Ausubel FM: Pseudomonas aeruginosa killing of
Caenorhabditis elegans used to identify P. aeruginosa
virulence factors. Proc Natl Acad Sci USA 1999;96:
24082413.

Josef Deutscher
Microbiologie et Gntique Molculaire
Route de Thiverval, FR78850 Thiverval-Grignon (France)
Tel. +33 1 30 81 54 47, Fax +33 1 30 81 54 57
E-Mail Josef.Deutscher@grignon.inra.fr

102

Poncet et al.

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 103119

Stand-Alone Response Regulators


Controlling Global Virulence Networks
in Streptococcus pyogenes
Kevin S. McIver
Department of Cell Biology & Molecular Genetics and Maryland Pathogen Research Institute, University of
Maryland, College Park, Md., USA

Abstract
Global regulation of virulence gene expression via transcriptional regulators plays a central role in
the ability of the bacterial pathogen Streptococcus pyogenes (the group A Streptococcus, GAS) to rapidly adapt during infection. The stand-alone regulators Mga, RofA-like proteins (RALPs), and RopB/
Rgg control important and diverse virulence regulons in response to growth-related signals and
other environmental conditions in GAS. Stand-alone regulated genes encode factors important for
colonization of tissues, immune evasion, persistence, dissemination, metabolism, and the response
to stressors. Although conserved core regulons have been established for each, recent studies have
revealed significant inter-serotype and even intra-serotype variation in the regulatory patterns presented by the stand-alone regulators. This chapter will look at each stand-alone regulatory pathway
in depth and discuss how these important global networks influence virulence as well as interact
with each other to produce an integrated response during GAS infection.
Copyright 2009 S. Karger AG, Basel

Rapid adaptation to changing environments encountered during infection is essential


for bacterial pathogens to persist in various tissues, evade the host immune response,
and disseminate to other niches. The ability to respond to these challenges typically
involves sensing relevant signals and linking them to a coordinate change in gene
expression, resulting in expression of a virulence gene regulon encoding factors appropriate for the given situation. Thus, pathogens will often exhibit a global alteration in
their transcriptome upon encountering a specific host environment. Furthermore,
different response regulatory pathways will often influence each other in order to
amplify, modulate, or repress the others response. Such interaction between regulatory networks allows the pathogen to fine-tune its response to complex environments
as well as remain sensitive to new signals that might be encountered.

The group A Streptococcus (GAS, Streptococcus pyogenes) is a strict human pathogen, causing a wide array of diseases ranging from self-limiting infections (pharyngitis or strep throat) to life-threatening invasive disorders (necrotizing fasciitis,
streptococcal toxic shock syndrome) to immune sequelae (acute rheumatic fever) [1,
2]. Infection occurs in the throat or on the surface of the skin, where GAS grows
extracellularly or sometimes intracellularly [1, 2]. Severe invasive disease involves the
spread of GAS from normal sites of colonization to deeper tissues, the bloodstream,
lymphatics, and various organs. Thus, a hallmark of GAS disease is their ability to
successfully adapt and persist at many different tissue sites in the human host.
GAS exhibits coordinate regulation of virulence gene expression in response to
changing host environments. Both growth ex vivo in relevant host fluids such as blood
and saliva or growth in vivo during infection of animal models results in a dramatic
remodeling of the GAS transcriptome, which points to a major role for global regulatory networks during infection [3, 4]. Although such pathways have been well studied
in other bacterial pathogens, we have only recently begun to unravel the mysteries of
this process in GAS. Unlike Gram-positive pathogens such as Staphylococcus aureus
or Bacillus anthracis, GAS does not appear to use alternative sigma factors to regulate virulence gene expression in response to stress or growth phase [5]. Instead, they
depend on transcriptional regulators, both activators and repressors, whose activities
are responsive to environmental conditions [6].
One approach utilizes classical two-component signal transduction systems (TCS)
and GAS possesses, on average, 13 such regulatory networks based on the available
genome sequences. In TCS, a sensor kinase detects the specific signal and phosphorylates its cognate DNA-binding response regulator leading to alterations in gene expression. Several TCS have now been characterized and found to play a significant role in
the pathogenesis of GAS. These include the CovRS system, which is important for the
response to stress and the switch to an invasive phenotype [7, 8]; the Ihk/Irr system,
which controls the ability of GAS to resist neutrophil killing [9], and SptRS, which is
critical for adaptation and growth of GAS in human saliva [10]. However, this chapter
will focus on a second major group of regulatory networks found in GAS that also allow
the pathogen to adapt to host environments called stand-alone response regulators.

Stand-Alone Response Regulators: Virulence and Growth Phase

The term stand-alone response regulator was originally coined to describe transcriptional regulators and their paralogs that control global virulence regulons in response
to changing environmental conditions in GAS, yet where the exact signals and cognate sensory elements involved remain undefined [6]. To date, three stand-alone
regulatory pathways have been studied extensively in GAS: the multiple gene regulator Mga, the RofA-like protein family or RALPs, and Rgg/RopB (see sections below).
In general, GAS exhibit a growth phase-dependent profile of virulence expression

104

McIver

density

Mga

RALPs

RopB/Rgg

(RofA, Nra, Ralp3, RivR)


Activation/Repression

Activation/Repression

Activation/Repression

Immune Evasion+
Adherence+
Metabolism+/

Adherence+/
Internalization+/
Cell Toxicity+/

Exoproteins+
Nutrient Stress
Peroxide stress+/

Adhesins
M, M-like proteins
Internalization Fn-binding proteins
Immune
Modulation

ScpA, Sic
M, M-like proteins

Sugar & Iron


Utilization

PTS operons
Iron uptake

Secreted & SpeB, SLO, MF


Surface Factors SpeG, Grab

Adhesins

Pilus FCT operon


Fn-binding proteins

Cell
damage

Streptolysin S
SpeB, SpeA

Amino Acid
Metabolism

Arginine, Histidine
& Serine Catabolism

Mga, RopB/Rgg
MsmR, RALPs

Other
Regulators

CovRS, Ihk/Irr
FasBCA, Mga

Other
Regulators

Exponential

Transition

Stationary

Colonization/acute

Persistence

Dissemination

time

Fig. 1. Growth, metabolism, and stand-alone response regulatory systems of GAS. An in vitro growth
curve is shown in background with phase of growth indicated. Boxes representing each stand-alone
network are placed at the time during growth where they exhibit maximal expression and the relevant point during infection is given at bottom. Prevalence of activation versus repression is indicated
by size and whether functions are activated (+) or repressed () are shown as superscript. Overall
processes and specific molecules regulated by each are shown.

controlled in large part by stand-alone response regulators (fig. 1) [6, 11]. Expression
of the Mga regulon is maximal during exponential growth in vitro likely representing a favorable environment in vivo [12, 13]. Upon encountering conditions that lead
to stationary phase growth, first RALPs then Rgg/RopB can negatively influence the
Mga regulon, presumably allowing persistence and spread to other tissue sites [14
16]. A common theme with the stand-alone regulators is that they appear intimately
linked to GAS metabolism in order to monitor growth, either by directly responding to metabolic signaling and/or via regulating metabolic operons. This chapter will
look at each stand-alone regulatory pathway in depth and then discuss how these
important global virulence networks interact with each other to produce an integrated response.

The Mga Regulon: Responding to the Good Life

The surface molecule M protein (emm) is a major virulence determinant of GAS for
resistance to phagocytosis, adherence to host tissues and triggering internalization
into non-phagocytic cells [2, 17]. Classification of GAS into over 100 serotypes is
based on the hypervariable N-terminal region of M protein, whereas a monoclonal
antibody to the C-terminus reveals two classes of M protein. Strains expressing the

Response Regulators in S. pyogenes

105

class II M protein also produce the serum opacity factor (SOF+), whereas class I GAS
strains do not (SOF). Mga was identified based on spontaneous M protein negative
class I M12 isolates resulting from small deletions located just upstream of the emm
gene at a locus called virR for virulence regulator [18]. A comparable locus mry (M
protein RNA yield) was found in another class I (M6) GAS strain that also positively
regulated emm transcription and M protein production [19]. Since both virR and mry
were found to encode the same predicted DNA-binding transcriptional regulator,
they were renamed mga for multiple gene regulator of GAS [2022].
Mga has been found in all sequenced GAS genomes and strains tested [16, 23].
Two divergent alleles of mga exist, mga-1 is linked to class I strains associated with
throat infections whereas mga-2 is found almost exclusively in class II strains that
cause disease at both throat and skin sites [23, 24]. Thus, divergent mga alleles, and
the genes that they regulate in these strains, have likely evolved to take advantage of
distinct environments in the host and may be directly involved in tissue specialization
in GAS [23]. All of the stimuli that lead to increased expression of the Mga regulon
in vitro would also be expected to support strong growth for GAS, including elevated
CO2, normal body temperature, increased iron levels, and metabolizable sugars [25
28]. Therefore, expression of the Mga regulon responds to environments conducive
for rapid growth.

The Core Mga Virulence Regulon


Mga strongly activates the transcription of a number of established GAS virulence
genes, as well as mga, during the exponential phase of growth (fig. 1). These core
Mga-regulated genes encode mostly cell wall-associated surface molecules important
for adherence to host tissues, internalization into non-phagocytic cells, and evasion
of the host immune responses. It has been hypothesized that Mga activates factors
important for early stages of GAS infection, enabling the pathogen to colonize a
promising niche and fight off the host immune response triggered by its subsequent
replication.
Mga transcriptionally activates genes that are closely linked to mga and emm in a
region of the GAS chromosome called the mga locus, including the downstream scpA
encoding a cell wall-associated C5a peptidase (ScpA) involved in cleavage of the C5a
chemotaxin of complement; thereby preventing the recruitment of PMNs [29]. Mga
can also autoactivate mga transcription, providing a mechanism to amplify the Mga
regulon [3032]. The genomic organization of the mga locus varies between strains in
the number of Mga-regulated emm and emm-like (mrp, enn) genes present [3335].
Mrp (FcrA) binds IgG and fibrinogen and is antiphagocytic in the absence of M protein, while Enn generally binds IgA or IgG3, but has not shown antiphagocytic properties [3638]. Typically, mga-1 strains possess a single emm, while the vast majority
of mga-2 strains have three M-like genes (mrp/emm/enn) [24].

106

McIver

The secreted inhibitor of complement (sic) found only in M1 strains is the only
secreted Mga-regulated virulence factor and has the ability to block the membrane
attack complex of complement and inhibit innate immune effectors such as host antimicrobial peptides [3941]. Similarly, the fibronectin-binding surface adhesin Fba
is only found in a few serotypes and also binds the complement regulators factor H
and factor H-like protein 1 to block opsonophagocytosis [42, 43]. The co-transcription of fba with scpA leads to its high-level Mga regulation [44, 45]. In class II SOF+
GAS strains, the Mga-activated sof and co-transcribed sfbX are located 12 kb away
from mga [46]. SfbX and SOF (SfbII) are fibronectin-binding surface proteins that
allow GAS to bind the host ECM while SOF also exhibits apoproteinase activity leading to serum opacity [47, 48]. The gene encoding streptococcal collagen-like protein
A (sclA/scl1) is also Mga regulated [49, 50]. The SclA/Scl1 protein absorbs human
plasma LDL and binds to 21 integrins on host cells, allowing for internalization and
re-emergence for immune evasion [51]. In some GAS serotypes, Mga can also regulate speB encoding the secreted cysteine protease SpeB [32, 52], which can degrade
both host proteins and Mga-regulated adhesins on the surface of GAS [53].

Mga Regulates Carbohydrate Utilization and Metabolism


Microarray analysis of the Mga regulon from three different GAS serotypes found
that Mga both activated and repressed expression of greater than 10% of the GAS
genome during exponential growth [52], indicating that Mga can also act as a repressor of gene expression. Only a single new gene was discovered (grm) that was highly
regulated (>20-fold) similar to the other core virulence genes, suggesting that most
core Mga regulon genes have been identified [52]. Instead, most of the novel Mgaregulated genes showed low-level regulation, possibly indirect, and represented operons important for the transport and utilization of carbohydrates as well as iron and
amino acids [52]. Loss of Mga resulted in an altered ability of GAS to grow in defined
media containing only sugar sources corresponding to the repressed operons [52].
Thus, Mga is able to regulate genes important not only for virulence, but also for the
metabolic homeostasis of GAS.

DNA Binding and Transcriptional Activation


Mga is an approximately 62-kDa protein that must bind to promoters for high level
transcriptional activation of core regulon genes [54, 55]. Three categories of Mgaactivated promoters have been established based on in vitro studies: Class A binding
sites represent the majority of Mga-regulated promoters and have a single Mga binding site just overlapping the -35 region [54], likely allowing Mga to interact with the
subunit of RNA polymerase. The binding of Mga distal to the start of transcription

Response Regulators in S. pyogenes

107

in class B promoters suggests that DNA bending might be required to allow interaction with RNA polymerase [56]. The dual Mga-binding sites found within the class
C autoregulated Pmga are unique amongst Mga-regulated promoters. Pmga contains two transcriptional start sites, P1 (distal) and P2 (proximal), with the two Mgabinding sites located between them [30, 31, 57]. Binding of Mga to the distal site I is
required for expression from the major P2 start site, although Mga appears to bind to
both sites independently [31, 57]. Transcription from P1 can occur in the absence of
Mga and is activated by the catabolite repression regulator CcpA early in growth to
initiate the Mga response [31, 57, 58].
DNase I footprinting experiments indicate that Mga protects a large 4559 base
pair non-palindromic region of DNA [31, 59]. Although a consensus site was proposed early on [59], the inclusion of newly identified sites has only led to a less
defined sequence. In vitro transcription assays have established that Mga is sufficient
to activate expression from class A and B promoters [56], although additional factor
can enhance activity [60].
Two conserved helix-turn-helix (HTH) domains, HTH-3 (residues 5372) and
HTH-4 (residues 107126), have been mapped near the N-terminus of Mga [20, 32,
55] that are required for DNA binding and transcriptional activation [55, 61]. While
HTH-4 is absolutely essential, HTH-3 may be primarily important for activation of
Pmga [55]. Recently, a small conserved domain CMD-1 at the N-terminus (residues
1015) may also contribute to DNA binding [61].

The Mga Regulon and GAS Pathogenesis


The core Mga virulence regulon plays a major role in GAS pathogenesis in vivo [2] as
seen in models of GAS systemic, skin, and throat infections [43, 62]. Recent genomewide transcriptional analyses of GAS during growth ex vivo and in vivo have further
established its importance in an infection [4]. During growth in whole human blood,
the Mga regulon rapidly reached maximum expression during exponential phase
similar to in vitro studies [63]. In a longitudinal study of pharyngitis in cynomolgus macaques, expression of the Mga regulon peaked during the early acute phase
of infection when the bacteria were actively growing and clinical symptoms were the
most severe [64]. Of interest, Mga expression in vivo correlated with the expression
carbohydrate utilization genes during this same period.

The Mga Family of Virulence Regulators


Mga orthologs can be been found in many pathogenic streptococci, including S. dysgalactiae, S. pneumoniae, S. equi, S. gordonii, S. mitis, S. sanguinis, and S. uberis [61,
6567]. The group C Streptococcus (GCS, S. dysgalactiae) ortholog DmgB/MgrC

108

McIver

positively regulates a GCS emm homologue demB [65, 66] and possesses comparable DNA-binding domains. The S. pneumoniae ortholog MgrA is important for both
nasopharyngeal colonization and pneumonia [67]. MgrA represses gene expression
of the rlrA pathogenicity islet, including the RALP regulator RlrA and a pilin biosynthesis operon [68]. The B. anthracis virulence tripartite toxin regulator AtxA and
capsule synthesis regulator AcpA exhibit regions of homology to Mga [69, 70].
Several domains characterized in Mga can be found in other known or predicted
Gram-positive transcription regulators. The N-terminal HTH domains of Mga
appear to be present in many regulators, including other stand-alone virulence regulators such as RALPs (Mga_HTH Pfam) [71]. Although motifs with similarity to the
receiver phosphorylation domains of the TCS response regulator CheY have been
predicted in the C-terminus of Mga [22], there has been no validation of this function. Two regions within Mga with strong similarity to the dual phosphotransferase
system (PTS) regulation domains (PRD) in the antiterminator LicT and other proteins involved in the regulation of sugar metabolism have recently been identified
[72, 73]. Activity of these PRD regulators is modulated by phosphorylation via the
PTS phosphorelay in response to the utilization of different carbohydrate sources.
Pfam now designates these regions as a member of the PRD superfamily (PRD_Mga)
[71] and this domain can be found in the RALPs as well. In fact, the B. anthracis AtxA
ortholog contains functional PRD domains that modulate its activity via phosphorylation by the PTS pathway [70]. Thus, the Mga family of virulence regulators and
possibly the RALPs as well may have evolved from metabolic sugar regulators and
could still retain the ability to sense the carbon flow of the bacteria as a means to follow growth.

RALPs: Regulating Life at the Transition

The RofA-like protein (RALP) family of transcriptional regulators was initially identified in GAS and now has orthologous members in several pathogenic streptococci,
including S. dysgalactiae (GCS, GGS), S. agalactiae (GBS), and S. pneumoniae, as well
as in Listeria. In GAS, the four members share on average 52% similarity and 29%
identity to each other and are approximately 58 kDa in size [74]. Established RALPs
are primarily involved in regulating expression of genes important for GAS-host cell
interactions and avoidance of host cell damage during the transition to stationary
phase growth (fig. 1) [14, 75, 76]. RALP-regulated genes include those encoding various MSCRAMMs (microbial surface components recognizing adhesive matrix molecules) such as those for pilus biosynthesis (Cpa, FctAB), fibronectin-binding proteins
(SfbI and Sfb2); hemolysins (Streptolysin S); proteases (SpeB); superantigens (SpeA),
and other stand-alone virulence regulators such as Mga [14, 7678]. Although much
of our knowledge has been derived from the study of two RALPs, RofA and Nra,
recent work has shed light on the function of the other GAS family members (RALP3

Response Regulators in S. pyogenes

109

& RALP4) as well as RALP members in other pathogens such as RlrA (RALP5) in S.
pneumoniae and RogB in S. agalactiae (GBS) [67, 7981].
GAS strains possess at least RofA or Nra and some strains can possess several
RALP family members in their genome, resulting in a widely variable combination
of RALP members, RALP-regulated genes, and patterns of gene expression between
different GAS serotypes [23, 78, 82]. As seen with mga, molecular epidemiology has
suggested that the RALPs rofA and nra have also undergone balancing selection with
strong association of the two alleles to different tissue niches (throat or skin) during
infection [23]. However, others have found no association between RALPs and tissue
tropism [78]. Despite this, RALPs as a whole influence host cell attachment, internalization, and increased intracellular persistence at the transition into stationary phase
(fig. 1).

RofA (RALP1)
RofA (regulator of F) represents the paradigm RALP and was first identified as a positive transcriptional activator of prtF encoding fibronectin-binding protein F (SfbI)
from a serotype M6 GAS under anaerobic growth conditions [83]. The autoregulated
rofA gene is found immediately adjacent to and divergently transcribed from prtF in
the M6 genome [84]. Subsequent studies in other GAS strains (M6, M2) have shown
that RofA also negatively influences the expression of key virulence genes (sagA,
speB), as well as mga [14, 78]. A rofA mutant exhibited reduced attachment and internalization to HEp-2 epithelial cells along with decreased host cell viability [14]. Since
there has not been a transcriptome analysis of a rofA mutant in GAS, the full extent of
a RofA regulon has yet to be determined.
RofA has predicted helix-turn-helix motifs homologous to Mga and is capable of interacting with specific binding sites within the intergenic region shared by
rofA and prtF [74, 84]. From these studies a consensus RofA-binding site was proposed (5-TTTTCACCAAAAANCAT-3) and was used to identify a potential site
in the promoter of the pilus-associated cpa in the SF370 M1 genome sequence [74].
Interestingly, Cpa is an Nra-regulated gene in other serotypes. No potential RofAbinding sites have been identified upstream of genes repressed by RofA (mga, speA,
sagA), suggesting that this control might be indirectly mediated by other regulatory
mechanisms.

Nra (RALP2)
Nra was first identified in an M49 GAS strain as a negative regulator for the adjacent cpa, encoding a collagen-binding pilus-associated surface protein, as well as
the unlinked prtF2, encoding a second fibronectin-binding protein (SfbII) [76, 85].

110

McIver

Similar to RofA, Nra repressed the expression of several virulence genes, including speB, speA, sagA, and the stand-alone regulator mga. The down modulation of
secreted superantigens and cytolysins helps to explain the reduction in host cell damage observed with nra mutants. Maximum expression of nra occurs immediately
following the transition to stationary phase growth, yet unlike RofA, Nra does not
appear to respond to changing atmospheric conditions [75, 76]. Although Nra is
homologous to RofA within its putative Mga-related DNA-binding domains, there
are no published studies demonstrating the expected interaction between Nra and
target promoters.
A transcriptome and proteome analysis of Nra in the M49 background found that
it regulated approximately 10% of the GAS genome across all phases of growth, both
activated and repressed, with the maximum occurring at the transition to stationary phase [86]. Nra primarily downregulated expression of virulence genes, including
those encoding pilus biogenesis in the FCT pathogenicity island (Fn binding, collagen
binding, and T antigen), cytolysin-mediated translocation (CMT; slo, nga), and capsule synthesis (hasABC). Nra repressed the expression of other stand-alone regulators
such as Rgg, RALP3, and RivR (RALP4), providing a transcriptional link between the
different global virulence pathways [86]. Nra positively influenced the expression of
metabolic genes such as those for the transport and utilization of carbohydrates [86],
demonstrating a connection between virulence and metabolism for RALPs as well.
The AraC-like transcriptional regulator MsmR, which can be found with rofA or
nra in certain FCT regions, was able to activate expression of nra and the CMT genes
(slo, nga) in M49. Thus, MsmR and Nra appear to act as opposing regulators in a
circuit to control adhesin expression in GAS. Interestingly, recent studies have found
that the polarity of Nra-MsmR-mediated regulation can be different depending on the
serotype examined. Luo et al. [87] found that MsmR actually represses nra expression
and Nra activated pilus-associated FCT genes in an M53 GAS strain representative of
skin infection isolates, while Nra had no effect on the Mga virulence regulon. These
results highlight an emerging theme of strain-specific regulatory patterns involving
stand-alone regulators.

Ralp3 (RALP3)
Analysis of the available GAS genome sequences established the serotype-specific distribution (M1, M4, M12, M28, and M49) of the ralp3 gene, which is located near the
streptolysin S operon (sag), and immediately upstream and divergently transcribed
from the largest predicted ORF in GAS encoding the fibrinogen-binding surface protein Lsp/Epf [86, 88]. This genomic region has been termed ERES for the eno ralp3
epf/lsp sagA pathogenicity region [86]. In an invasive M1T1 strain of GAS, mutations
in ralp3 influenced host-cell interactions and antimicrobial peptide sensitivity leading to the inability to survive in whole blood and an attenuation of systemic infection

Response Regulators in S. pyogenes

111

in mice [88]. The M1 Ralp3 was found to strongly repress capsule synthesis (has) and
cysteine protease (speB) expression. Ralp3 in an M1 strain repressed expression of
lsp/epf, while in an M49 strain Ralp3 had the opposite effect, demonstrating strainspecific differences in this pathway as well. The M49 Ralp3 was able to activate nra,
which in turn downregulates ralp3 forming a closed regulatory circuit [86]. It also
repressed expression of the AraC regulator MsmR and the fibronectin-binding protein SFbII. As with all RALPs, expression of ralp3 was maximal during transition to
stationary phase. Ralp3 has a predicted DNA-binding motif homologous to RofA and
Nra, but has not been shown to bind DNA.

RivR (RALP4)
The expression of rivR, which stands for Ralp-IV (RALP4), is directly repressed by
the major virulence-associated TCS, CovRS [89]. DNA microarray analysis in an M1
strain determined that RivR activates the stand-alone regulator Mga and its virulence
regulon [60]. RivR is associated with a small regulatory RNA, RivX and a rivRX
mutant was attenuated for invasive skin infection in mice. Interestingly, either rivR
or rivX gene was able to separately complement the Mga activation defect [60]. Since
RivR is able to enhance Mga transcriptional activation in vitro, it may be occurring
through direct interaction of the proteins or through RivR binding DNA and supporting Mga-dependent activation. The influence of a CovR-repressed RALP on the
Mga virulence regulon provides an intriguing linkage between TCS and stand-alone
regulatory networks in GAS.

Rgg/RopB: Sensing the Stresses of Infection

Members of the Rgg family of transcriptional regulators can be found in the genomes
of many different low G+C Gram-positive bacteria, including S. pyogenes [90, 91], S.
gordonii [92], S. oralis [93], S. sanguinis [94], S. mutans [95], Lactococcus lactis [96],
and Listeria monocytogenes. A common theme for all of these orthologs is that they
are likely DNA-binding proteins that transcriptionally regulate an adjacent gene or
genes that do not appear to share any functional similarity between the different systems [9799].
The Rgg ortholog in GAS (RopB/Rgg) has been extensively studied over the last 10
years, with all of the work performed in two serotypes, M5 and mostly M49. As with
other family members, Rgg/RopB is a predicted DNA-binding protein of approximately 33 kDa. Interestingly, rgg/ropB exhibits a G+C content (32%) that is significantly lower than the 39% G+C found in the rest of the GAS genome, suggesting that
this regulator may have been obtained through some form of horizontal gene transfer
[90, 91].

112

McIver

Expression of the extracellular SpeB cysteine protease, a major virulence factor


of GAS, is dependent upon the Rgg-like transcriptional regulator called RopB [91]
or Rgg [90]. Similar to other Rgg-like regulators, ropB/rgg is found adjacent to and
divergently transcribed from speB in the GAS genome. Although SpeB production
is strongly growth phase regulated, RopB/Rgg does not appear to be mediator of this
control [97]. Two independent transcriptome analyses as well as proteomic studies of
both secreted and cytoplasmic proteins have now been performed on an Rgg mutant
in the M49 background [15, 100103], providing the field with a high-resolution picture of the Rgg regulon in GAS. It is evident from these data that RopB/Rgg is a global
transcriptional regulator, influencing >30% of genome in all phases of growth, with an
overall trend of repression during exponential growth and activation during stationary phase [103]. However, a recent transcriptome analysis of RopB/Rgg in two M1
strains and a second M49 strain indicate a much smaller influence on gene expression in other serotypes [104]. In fact, speB and the adjacent Spy2040 represented the
only core regulon genes regulated in all the strains studies. Thus, RopB/Rgg plays
an important role in modulating virulence and metabolism during times of nutrient
limitation and stress associated with post-exponential growth (fig. 1), but can exhibit
inter- and intra-serotypic variation.

The Rgg/RopB Regulon and GAS Pathogenesis


In addition to activating expression of the cysteine protease SpeB, RopB/Rgg regulates a large number of other virulence-associated genes and regulatory pathways in
GAS [15, 100103]. RopB/Rgg represses the genes encoding the cytolysin-mediated
translocation (CMT) system (slo, nga/spn), secreted DNases such as MF-1 and MF-3,
streptokinase (ska), and Mga-regulated proteins (sic, sclA/scl1, scpA, and emm). At the
same time, RopB/Rgg activates not only speB, but also the secreted superantigen gene
speG and the gene for the 2-macroglobulin-binding protein (grab) found on the surface of GAS. RopB/Rgg mutants are also more resistant to inducible H2O2 stress and
exhibit increased virulence during systemic infection of mice via the intraperitoneal
route [105], leading to the suggestion that RopB/Rgg may be a novel sensoregulator
of GAS peroxide resistance.
Many of the effects on virulence gene expression may be mediated by the
influence of RopB/Rgg on TCS as well as other stand-alone regulators. Similar
to RALPs, RopB/Rgg represses mga transcription and expression of Mga regulon genes, which may contribute to shutting down this regulon during stationary
phase. RopB/Rgg also positively influences several TCS networks linked to virulence, including CovRS, FasBCA, LytRS, and Ihk/Irr [15, 103]. Whether RopB/Rgg
directly affects expression of these regulatory pathways or whether it occurs indirectly through changes to the metabolic state of the cell (see below) is currently not
known.

Response Regulators in S. pyogenes

113

The RopB/Rgg Regulon and Metabolism


RopB/Rgg also coordinates amino acid catabolism and is essential for the growth of
GAS in non-glucose sugars by repressing these functions in the high glucose environment found during exponential growth [101, 106]. This includes exponential phase
repression of arginine (arcABC), histidine (hutI), and serine (sdhA) catabolism genes
as well as operons important for utilization of the alternative sugar sources fructose,
mannose, and sucrose [103, 106]. Thus, RopB/Rgg mutants do not preferentially
use glucose, but will ferment arginine; an environment that may resemble growth in
necrotic lesions [101]. In addition, expression of several genes important for thermal
(clpE, clpL) and oxidative stress (ahpCF) is repressed by RopB/Rgg early in growth;
therefore, ropB/rgg mutants are more tolerant to high temperature, peroxide, and
some antibiotics [105, 106]. Thus, RopB/Rgg appears to keep those functions needed
in nutrient limiting environments downregulated when glucose is available.

DNA Binding and Transcriptional Activation


Direct interaction between RopB/Rgg and promoter-binding sites has been demonstrated for the intergenic PspeB and PropB promoters in the M5 background [97].
Specific binding of purified RopB to sites adjacent to ropB was critical for transcription of both genes, indicating that the promoters were overlapping. At this time, a
consensus-binding sequence for RopB/Rgg in GAS has not been proposed. Several
mutations in RopB/Rgg based on conserved amino acids identified amongst multiple
Rgg family members resulted in an inactive regulator, including those in the predicted
helix-turn-helix DNA-binding domain [107]. RopB/Rgg can form homodimers in
vitro and dominant negative mutants in conserved residues provide in vivo support
for this hypothesis [107]. Finally, recent studies have revealed the requirement for a
lactose aldolase LacD.1 that has been adapted to repress expression of speB independent of its metabolic enzymatic function [108, 109]. Co-immunoprecipitation experiments suggest that LacD.1 forms a complex with RopB/Rgg to promote repression,
perhaps by sequestering it and preventing binding at PspeB [108]. Thus, further studies will be necessary to unravel the complexities of RopB/Rgg regulation in GAS.

Interactions between Stand-Alone Regulators

Based on the studies of individual GAS stand-alone response regulatory systems in GAS,
it is clear that they are intimately linked to each other as well as to classical TCS systems.
This provides GAS with a mechanism to fine-tune responses to different signals and integrate them into a broader response. Some RALPs (RofA, Nra) and Rgg/RopB can repress
mga expression post-exponentially in certain serotypes, leading to downregulation of

114

McIver

the Mga response [14, 15, 76, 88]. RivR (RALP4), which is directly repressed by the TCS
CovRS, is able to enhance activation of mga and Mga-regulated genes [89]. Nra can
downregulate the expression of Rgg and other RALPs such as Ralp3 and RivR [86]. Rgg/
RopB influences several virulence-associated TCS, including CovRS. Finally, the modulation of metabolic gene expression by both Mga and RopB/Rgg, combined with their
abilities to sense metabolic fluxes, suggests a possible feedback regulatory circuit. These
last findings raise the intriguing suggestion that growth phase regulation of virulence in
GAS involves monitoring metabolic processes by stand-alone regulators (fig. 1). Given
that GAS possesses a relatively small genome (ca. 1.9 Mbp) and must obtain most of its
nutrients from the human host, sensing the metabolic environment would be an excellent indicator of the well-being of the pathogen during infection.

Acknowledgements
The author would like to thank all members of the McIver laboratory, past and present. The information described here was the result of contributions from numerous laboratories in the field and
the author apologizes for any work that was excluded due to space limitations. K.S.M. was supported by a grant from the National Institutes of Health (NIH/NIAID AI47928).

References
1 Bisno AL, Brito MO, Collins CM: Molecular basis of
group A streptococcal virulence. Lancet Infect Dis
2003;3:191200.
2 Cunningham MW: Pathogenesis of group A streptococcal infections. Clin Microbiol Rev 2000;13:470
511.
3 Shelburne SA, Musser JM: Virulence gene expression in vivo. Curr Opin Microbiol 2004;7:283289.
4 Musser JM, DeLeo FR: Toward a genome-wide systems biology analysis of host-pathogen interactions
in group A Streptococcus. Am J Pathol 2005;167:
14611472.
5 Opdyke JA, Scott JR, Moran CP: A secondary RNA
polymerase sigma factor from Streptococcus pyogenes. Mol Microbiol 2001;42:495502.
6 Kreikemeyer B, McIver KS, Podbielski A: Virulence
factor regulation and regulatory networks in Streptococcus pyogenes and their impact on pathogen-host
interactions. Trends Microbiol 2003;11:224232.
7 Churchward G: The two faces of Janus: virulence
gene regulation by CovR/S in group A streptococci.
Mol Microbiol 2007;64:3441.
8 Sumby P, Whitney AR, Graviss EA, DeLeo FR, Musser
JM: Genome-wide analysis of group a streptococci
reveals a mutation that modulates global phenotype
and disease specificity. PLoS Pathog 2006;2:e5.

Response Regulators in S. pyogenes

9 Voyich JM, Braughton KR, Sturdevant DE, Vuong


C, Kobayashi SD, Porcella SF, Otto M, Musser JM,
DeLeo FR: Engagement of the pathogen survival
response used by group A Streptococcus to avert
destruction by innate host defense. J Immunol 2004;
173:11941201.
10 Shelburne SA 3rd, Sumby P, Sitkiewicz I, Granville
C, DeLeo FR, Musser JM: Central role of a bacterial
two-component gene regulatory system of previously unknown function in pathogen persistence in
human saliva. Proc Natl Acad Sci USA 2005;102:
1603716042.
11 Chaussee MA, Dmitriev AV, Callegari EA, Chaussee
MS: Growth phase-associated changes in the transcriptome and proteome of Streptococcus pyogenes.
Arch Microbiol 2008;189:2741.
12 McIver KS, Scott JR: Role of mga in growth phase
regulation of virulence genes of the group A
Streptococcus. J Bacteriol 1997;179:51785187.
13 Unnikrishnan M, Cohen J, Sriskandan S: Growthphase-dependent expression of virulence factors in
an M1T1 clinical isolate of Streptococcus pyogenes.
Infect Immun 1999;67:54955499.
14 Beckert S, Kreikemeyer B, Podbielski A: Group A
streptococcal rofA gene is involved in the control of
several virulence genes and eukaryotic cell attachment
and internalization. Infect Immun 2001;69:534537.

115

15 Chaussee MS, Sylva GL, Sturdevant DE, Smoot LM,


Graham MR, Watson RO, Musser JM: Rgg influences the expression of multiple regulatory loci to
coregulate virulence factor expression in Streptococcus
pyogenes. Infect Immun 2002;70:762770.
16 Podbielski A: Ubiquitous occurrence of virR and
scpA genes in group A streptococci. Med Microbiol
Immunol (Berl) 1992;181:227240.
17 Navarre WW, Schneewind O: Surface proteins of
Gram-positive bacteria and mechanisms of their
targeting to the cell wall envelope. Microbiol Mol
Biol Rev 1999;63:174229.
18 Spanier JG, Jones SJ, Cleary P: Small DNA deletions
creating avirulence in Streptococcus pyogenes.
Science 1984;225:935938.
19 Caparon MG, Scott JR: Identification of a gene that
regulates expression of M protein, the major virulence determinant of group A streptococci. Proc
Natl Acad Sci USA 1987;84:86778681.
20 Chen C, Bormann N, Cleary PP: VirR and Mry are
homologous trans-acting regulators of M protein
and C5a peptidase expression in group A streptococci. Mol Gen Genet 1993;241:685693.
21 Scott JR, Cleary P, Caparon MG, Kehoe M, Heden
L, Musser JM, Hollingshead S, Podbielski A: New
name for the positive regulator of the M protein of
group A Streptococcus. Mol Microbiol 1995;17:799.
22 Perez-Casal J, Caparon MG, Scott JR: Mry, a transacting positive regulator of the M protein gene of
Streptococcus pyogenes with similarity to the receptor proteins of two-component regulatory systems. J
Bacteriol 1991;173:26172624.
23 Bessen DE, Manoharan A, Luo F, Wertz JE,
Robinson DA: Evolution of transcription regulatory
genes is linked to niche specialization in the bacterial pathogen Streptococcus pyogenes. J Bacteriol
2005;187:41634172.
24 Hollingshead SK, Readdy TL, Yung DL, Bessen DE:
Structural heterogeneity of the emm gene cluster in
group A streptococci. Mol Microbiol 1993;8:707
717.
25 Caparon MG, Geist RT, Perez-Casal J, Scott JR:
Environmental regulation of virulence in group A
streptococci: transcription of the gene encoding M
protein is stimulated by carbon dioxide. J Bacteriol
1992;174:56935701.
26 McIver KS, Heath AS, Scott JR: Regulation of virulence by environmental signals in group A streptococci: influence of osmolarity, temperature, gas
exchange, and iron limitation on emm transcription. Infect Immun 1995;63:45404542.
27 Pine L, Reeves MW: Regulation of the synthesis of
M protein by sugars, Todd Hewitt broth, and horse
serum, in growing cells of Streptococcus pyogenes.
Microbios 1978;21:185212.

116

28 Podbielski A, Peterson JA, Cleary P: Surface protein-CAT reporter fusions demonstrate differential
gene expression in the vir regulon of Streptococcus
pyogenes. Mol Microbiol 1992;6:22532265.
29 Simpson WJ, LaPenta D, Chen C, Cleary PP:
Coregulation of type 12 M protein and streptococcal C5a peptidase genes in group A streptococci:
evidence for a virulence regulon controlled by the
virR locus. J Bacteriol 1990;172:696700.
30 Geist RT, Okada N, Caparon MG: Analysis of
Streptococcus pyogenes promoters by using novel
Tn916-based shuttle vectors for the construction of
transcriptional fusions to chloramphenicol acetyltransferase. J Bacteriol 1993;175:75617570.
31 McIver KS, Thurman AS, Scott JR: Regulation of
mga transcription in the group A Streptococcus: specific binding of Mga within its own promoter and
evidence for a negative regulator. J Bacteriol 1999;
181:53735383.
32 Podbielski A, Flosdorff A, Weber-Heynemann J:
The group A streptococcal virR49 gene controls
expression of four structural vir regulon genes.
Infect Immun 1995;63:920.
33 Haanes EJ, Heath DG, Cleary PP: Architecture of
the vir regulons of group A streptococci parallels
opacity factor phenotype and M protein class. J
Bacteriol 1992;174:49674976.
34 Hollingshead SK, Arnold J, Readdy TL, Bessen DE:
Molecular evolution of a multigene family in group
A streptococci. Mol Biol Evol 1994;11:208219.
35 Podbielski A: Three different types of organization
of the vir regulon in group A streptococci. Mol Gen
Genet 1993;237:287300.
36 Bessen DE, Fischetti VA: Nucleotide sequences of
two adjacent M or M-like protein genes of group A
streptococci: different RNA transcript levels and
identification of a unique immunoglobulin A-binding
protein. Infect Immun 1992;60:124135.
37 Courtney HS, Hasty DL, Dale JB: Anti-phagocytic
mechanisms of Streptococcus pyogenes: binding of
fibrinogen to M-related protein. Mol Microbiol 2006;
59:936947.
38 Podbielski A, Hawlitzky J, Pack TD, Flosdorff A,
Boyle MD: A group A streptococcal Enn protein
potentially resulting from intergenomic recombination exhibits atypical immunoglobulin-binding
characteristics. Mol Microbiol 1994;12:725736.
39 kesson P, Sjholm AG, Bjrck L: Protein SIC, a
novel extracellular protein of Streptococcus pyogenes
interfering with complement function. J Biol Chem
1996;271:10811088.
40 Fernie-King BA, Seilly DJ, Willers C, Wurzner R,
Davies A, Lachmann PJ: Streptococcal inhibitor of
complement (SIC) inhibits the membrane attack
complex by preventing uptake of C567 onto cell
membranes. Immunol 2001;103:390398.

McIver

41 Frick IM, kesson P, Rasmussen M, Schmidtchen


A, Bjrck L: SIC, a secreted protein of Streptococcus
pyogenes that inactivates antibacterial peptides. J
Biol Chem 2003;278:1656116566.
42 Pandiripally V, Gregory E, Cue D: Acquisition of
regulators of complement activation by Streptococcus
pyogenes serotype M1. Infect Immun 2002;70:6206
6214.
43 Terao Y, Kawabata S, Kunitomo E, Murakami J,
Nakagawa I, Hamada S: Fba, a novel fibronectinbinding protein from Streptococcus pyogenes, promotes bacterial entry into epithelial cells, and the
fba gene is positively transcribed under the Mga
regulator. Mol Microbiol 2001;42:7586.
44 Podbielski A, Woischnik M, Pohl B, Schmidt KH:
What is the size of the group A streptococcal vir
regulon? The Mga regulator affects expression of
secreted and surface virulence factors. Med Microbiol Immunol 1996;185:171181.
45 Pritchard KH, Cleary PP: Differential expression of
genes in the vir regulon of Streptococcus pyogenes is
controlled by transcription termination. Mol Gen
Genet 1996;250:207213.
46 Green NM, Zhang S, Porcella SF, Nagiec MJ, Barbian
KD, Beres SB, Lefebvre RB, Musser JM: Genome
sequence of a serotype M28 strain of group A
Streptococcus: potential new insights into puerperal
sepsis and bacterial disease specificity. J Infect Dis
2005;192:760770.
47 Courtney HS, Hasty DL, Li Y, Chiang HC, Thacker
JL, Dale JB: Serum opacity factor is a major
fibronectin-binding protein and a virulence determinant of M type 2 Streptococcus pyogenes. Mol
Microbiol 1999;32:8998.
48 Jeng A, Sakota V, Li Z, Datta V, Beall B, Nizet V:
Molecular genetic analysis of a group A Streptococcus
operon encoding serum opacity factor and a novel
fibronectin-binding protein, SfbX. J Bacteriol 2003;
185:12081217.
49 Lukomski S, Nakashima K, Abdi I, Cipriano VJ,
Shelvin BJ, Graviss EA, Musser JM: Identification
and characterization of a second extracellular collagen-like protein made by group A Streptococcus:
control of production at the level of translation.
Infect Immun 2001;69:17291738.
50 Rasmussen R, Eden A, Bjrck L: SclA, a novel collagen-like surface protein of Streptococcus pyogenes.
Infect Immun 2000;68:63706377.
51 Caswell CC, Lukomska E, Seo NS, Hk M,
Lukomski S: Scl1-dependent internalization of
group A Streptococcus via direct interactions with
the 21 integrin enhances pathogen survival and
re-emergence. Mol Microbiol 2007;64:13191331.

Response Regulators in S. pyogenes

52 Ribardo DA, McIver KS: Defining the Mga regulon:


comparative transcriptome analysis reveals both
direct and indirect regulation by Mga in the group
A Streptococcus. Mol Microbiol 2006;62:491508.
53 Rasmussen M, Bjrck L: Proteolysis and its regulation at the surface of Streptococcus pyogenes. Mol
Microbiol 2002;43:537544.
54 Almengor AC, McIver KS: Transcriptional activation of sclA by Mga requires a distal binding site in
Streptococcus pyogenes. J Bacteriol 2004;186:7847
7857.
55 McIver KS, Myles RL: Two DNA-binding domains
of Mga are required for virulence gene activation in
the group A Streptococcus. Mol Microbiol 2002;43:
15911602.
56 Almengor AC, Walters MS, McIver KS: Mga is sufficient to activate transcription in vitro of sof/sfbX and
other Mga-regulated virulence genes in the group A
Streptococcus. J Bacteriol 2006;188:20382047.
57 Okada N, Geist RT, Caparon MG: Positive transcriptional control of mry regulates virulence in the group
A Streptococcus. Mol Microbiol 1993;7:893903.
58 Almengor AC, Kinkel TL, Day SJ, McIver KS: The
catabolite control protein CcpA binds to Pmga and
influences expression of the virulence regulator
Mga in the group A Streptococcus. J Bacteriol 2007;
189:84058416.
59 McIver KS, Heath AS, Green BD, Scott JR: Specific
binding of the activator Mga to promoter sequences
of the emm and scpA genes in the group A
Streptococcus. J Bacteriol 1995;177:66196624.
60 Roberts SA, Scott JR: RivR and the small RNA RivX:
the missing links between the CovR regulatory cascade and the Mga regulon. Mol Microbiol 2007;66:
15061522.
61 Vahling CM, McIver KS: Domains required for
transcriptional activation show conservation in the
Mga family of virulence gene regulators. J Bacteriol
2006;188:863873.
62 Kihlberg BM, Cooney J, Caparon MG, Olsn A,
Bjrck L: Biological properties of a Streptococcus
pyogenes mutant generated by Tn916 insertion in
mga. Microb Pathog 1995;19:299315.
63 Graham MR, Virtaneva K, Porcella SF, Barry WT,
Gowen BB, Johnson CR, Wright FA, Musser JM:
Group A Streptococcus transcriptome dynamics
during growth in human blood reveals bacterial
adaptive and survival strategies. Am J Pathol 2005;
166:455465.
64 Virtaneva K, Porcella SF, Graham MR, Ireland RM,
Johnson CA, Ricklefs SM, Babar I, Parkins LD,
Romero RA, Corn GJ, Gardner DJ, Bailey JR, Parnell
MJ, Musser JM: Longitudinal analysis of the group A
Streptococcus transcriptome in experimental pharyngitis in cynomolgus macaques. Proc Natl Acad Sci
USA 2005;102:90149019.

117

65 Geyer A, Schmidt KH: Genetic organisation of the


M protein region in human isolates of group C and
G streptococci: two types of multigene regulator-like
(mgrC) regions. Mol Gen Genet 2000;262:965976.
66 Vasi J, Frykberg L, Carlsson LE, Lindberg M, Guss
B: M-like proteins of Streptococcus dysgalactiae.
Infect Immun 2000;68:294302.
67 Hava DL, Camilli A: Large-scale identification of
serotype 4 Streptococcus pneumoniae virulence factors. Mol Microbiol 2002;45:13891406.
68 Hemsley C, Joyce E, Hava DL, Kawale A, Camilli A:
MgrA, an orthologue of Mga, Acts as a transcriptional repressor of the genes within the rlrA pathogenicity islet in Streptococcus pneumoniae. J Bacteriol
2003;185:66406647.
69 Drysdale M, Bourgogne A, Koehler TM:
Transcriptional analysis of the Bacillus anthracis
capsule regulators. J Bacteriol 2005;187:51085114.
70 Tsvetanova B, Wilson AC, Bongiorni C, Chiang C,
Hoch JA, Perego M: Opposing effects of histidine
phosphorylation regulate the AtxA virulence transcription factor in Bacillus anthracis. Mol Microbiol
2007;63:644655.
71 Finn RD, Mistry J, Schuster-Bockler B, GriffithsJones S, Hollich V, Lassmann T, Moxon S, Marshall
M, Khanna A, Durbin R, Eddy SR, Sonnhammer
EL, Bateman A: Pfam: clans, web tools and services.
Nucleic Acids Res 2006;34:D247D251.
72 Hondorp ER, McIver KS: The Mga virulence regulon: infection where the grass is greener. Mol
Microbiol 2007;66:10561065.
73 Deutscher J, Francke C, Postma PW: How phosphotransferase system-related protein phosphorylation
regulates carbohydrate metabolism in bacteria.
Microbiol Mol Biol Rev 2006;70:9391031.
74 Granok AB, Parsonage D, Ross RP, Caparon MG:
The RofA binding site in Streptococcus pyogenes is
utilized in multiple transcriptional pathways. J
Bacteriol 2000;182:15291540.
75 Molinari G, Rohde M, Talay SR, Chhatwal GS,
Beckert S, Podbielski A: The role played by the
group A streptococcal negative regulator Nra on
bacterial interactions with epithelial cells. Mol
Microbiol 2001;40:99114.
76 Podbielski A, Woischnik M, Leonard BA, Schmidt
KH: Characterization of nra, a global negative regulator gene in group A streptococci. Mol Microbiol
1999;31:10511064.
77 Patti JM, Hk M: Microbial adhesins recognizing
extracellular matrix macromolecules. Curr Opin
Cell Biol 1994;6:752758.
78 Kreikemeyer B, Beckert S, Braun-Kiewnick A,
Podbielski A: Group A streptococcal RofA-type
global regulators exhibit a strain-specific genomic
presence and regulation pattern. Microbiology 2002;
148:15011511.

118

79 Gutekunst H, Eikmanns BJ, Reinscheid DJ: Analysis


of RogB-controlled virulence mechanisms and gene
repression in Streptococcus agalactiae. Infect Immun
2003;71:50565064.
80 Hava DL, Hemsley CJ, Camilli A: Transcriptional
regulation in the Streptococcus pneumoniae rlrA
pathogenicity islet by RlrA. J Bacteriol 2003;185:413
421.
81 Tettelin H, Nelson KE, Paulsen IT, Eisen JA, Read TD,
Peterson S, Heidelberg J, DeBoy RT, Haft DH, Dodson
RJ, Durkin AS, Gwinn M, Kolonay JF, Nelson WC,
Peterson JD, Umayam LA, White O, Salzberg SL,
Lewis MR, Radune D, Holtzapple E, Khouri H, Wolf
AM, Utterback TR, Hansen CL, McDonald LA,
Feldblyum TV, Angiuoli S, Dickinson T, Hickey EK,
Holt IE, Loftus BJ, Yang F, Smith HO, Venter JC,
Dougherty BA, Morrison DA, Hollingshead SK,
Fraser CM: Complete genome sequence of a virulent
isolate of Streptococcus pneumoniae. Science 2001;
293:498506.
82 Bessen DE, Kalia A: Genomic localization of a T
serotype locus to a recombinatorial zone encoding
extracellular matrix-binding proteins in Streptococcus
pyogenes. Infect Immun 2002;70:11591167.
83 Fogg GC, Gibson CM, Caparon MG: The identification of rofA, a positive-acting regulatory component
of prtF expression: use of an m---based shuttle
mutagenesis strategy in Streptococcus pyogenes. Mol
Microbiol 1994;11:671684.
84 Fogg GC, Caparon MG: Constitutive expression of
fibronectin binding in Streptococcus pyogenes as a
result of anaerobic activation of rofA. J Bacteriol
1997;179:61726180.
85 Kreikemeyer B, Nakata M, Oehmcke S, Gschwendtner
C, Normann J, Podbielski A: Streptococcus pyogenes
collagen type I-binding Cpa surface protein. Expression profile, binding characteristics, biological functions, and potential clinical impact. J Biol Chem 2005;
280:3322833239.
86 Kreikemeyer B, Nakata M, Koller T, Hildisch H,
Kourakos V, Standar K, Kawabata S, Glocker MO,
Podbielski A: The Streptococcus pyogenes serotype
M49 Nra-Ralp3 transcriptional regulatory network
and its control of virulence factor expression from
the novel eno ralp3 epf sagA pathogenicity region.
Infect Immun 2007;75:56985710.
87 Luo F, Lizano S, Bessen DE: Heterogeneity in the
polarity of Nra regulatory effects on streptococcal
pilus gene transcription and virulence. Infect
Immun 2008;76:24902497.
88 Kwinn LA, Khosravi A, Aziz RK, Timmer AM,
Doran KS, Kotb M, Nizet V: Genetic characterization and virulence role of the RALP3/LSA locus
upstream of the streptolysin S operon in invasive
M1T1 group A Streptococcus. J Bacteriol 2007;189:
13221329.

McIver

89 Roberts SA, Churchward GG, Scott JR: Unraveling


the regulatory network in Streptococcus pyogenes:
the global response regulator CovR represses rivR
directly. J Bacteriol 2007;189:14591463.
90 Chaussee MS, Ajdic D, Ferretti JJ: The rgg gene of
Streptococcus pyogenes NZ131 positively influences
extracellular SPE B production. Infect Immun 1999;
67:17151722.
91 Lyon WR, Gibson CM, Caparon MG: A role for
trigger factor and an rgg-like regulator in the transcription, secretion and processing of the cysteine
proteinase of Streptococcus pyogenes. EMBO J 1998;
17:62636275.
92 Sulavik MC, Tardif G, Clewell DB: Identification of
a gene, rgg, which regulates expression of glucosyltransferase and influences the Spp phenotype of
Streptococcus gordonii Challis. J Bacteriol 1992;174:
35773586.
93 Fujiwara T, Hoshino T, Ooshima T, Sobue S, Hamada
S: Purification, characterization, and molecular analysis of the gene encoding glucosyltransferase from
Streptococcus oralis. Infect Immun 2000;68:2475
2483.
94 Vickerman MM, Sulavik MC, Clewell DB: Oral
streptococci with genetic determinants similar to
the glucosyltransferase regulatory gene, rgg. Infect
Immun 1995;63:45244527.
95 Qi F, Chen P, Caufield PW: Purification and biochemical characterization of mutacin I from the
group I strain of Streptococcus mutans, CH43, and
genetic analysis of mutacin I biosynthesis genes.
Appl Environ Microbiol 2000;66:32213229.
96 Sanders JW, Leenhouts K, Burghoorn J, Brands JR,
Venema G, Kok J: A chloride-inducible acid resistance mechanism in Lactococcus lactis and its regulation. Mol Microbiol 1998;27:299310.
97 Neely MN, Lyon WR, Runft DL, Caparon M: Role
of RopB in growth phase expression of the SpeB
cysteine protease of Streptococcus pyogenes. J
Bacteriol 2003;185:51665174.
98 Rawlinson EL, Nes IF, Skaugen M: Identification of
the DNA-binding site of the Rgg-like regulator LasX
within the lactocin S promoter region. Microbiol
2005;151:813823.

99 Vickerman MM, Minick PE, Mather NM:


Characterization of the Streptococcus gordonii chromosomal region immediately downstream of the
glucosyltransferase gene. Microbiol 2001;147:3061
3070.
100 Chaussee MS, Phillips ER, Ferretti JJ: Temporal production of streptococcal erythrogenic toxin B (streptococcal cysteine proteinase) in response to nutrient
depletion. Infect Immun 1997;65:19561959.
101 Chaussee MS, Somerville GA, Reitzer L, Musser JM:
Rgg coordinates virulence factor synthesis and
metabolism in Streptococcus pyogenes. J Bacteriol
2003;185:60166024.
102 Chaussee MS, Watson RO, Smoot JC, Musser JM:
Identification of Rgg-regulated exoproteins of
Streptococcus pyogenes. Infect Immun 2001;69:822
831.
103 Dmitriev AV, McDowell EJ, Kappeler KV, Chaussee
MA, Rieck LD, Chaussee MS: The Rgg regulator of
Streptococcus pyogenes influences utilization of nonglucose carbohydrates, prophage induction, and
expression of the NAD-glycohydrolase virulence
operon. J Bacteriol 2006;188:72307241.
104 Dmitriev AV, McDowell EJ, Chaussee MS: Interand intraserotypic variation in the Streptococcus
pyogenes Rgg regulon. FEMS Microbiol Lett
2008;284:4351.
105 Pulliainen AT, Hytonen J, Haataja S, Finne J:
Deficiency of the Rgg regulator promotes H2O2
resistance, AhpCF-mediated H2O2 decomposition
and virulence in Streptococcus pyogenes. J Bacteriol
2008;190:32253235.
106 Chaussee MA, Callegari EA, Chaussee MS: Rgg
regulates growth phase-dependent expression of
proteins associated with secondary metabolism and
stress in Streptococcus pyogenes. J Bacteriol 2004;
186:70917099.
107 Loughman JA, Caparon MG: Contribution of invariant residues to the function of Rgg family transcription regulators. J Bacteriol 2007;189:650655.
108 Loughman JA, Caparon MG: A novel adaptation of
aldolase regulates virulence in Streptococcus pyogenes. EMBO J 2006;25:54145422.
109 Loughman JA, Caparon MG: Comparative functional analysis of the lac operons in Streptococcus
pyogenes. Mol Microbiol 2007;64:269280.

Dr. Kevin S. McIver


Department of Cell Biology & Molecular Genetics and
Maryland Pathogen Research Institute, University of Maryland
3124 Biosciences Research Building, College Park, MD 20742 (USA)
Tel. +1 301 405 4136, Fax +1 301 314 1248, E-Mail Kmciver@umd.du

Response Regulators in S. pyogenes

119

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 120135

The Heme Sensor System of


Staphylococcus aureus
Devin L. Stauff Eric P. Skaar
Department of Microbiology and Immunology, Vanderbilt University Medical Center, Nashville, Tenn., USA

Abstract
The important human pathogen Staphylococcus aureus is able to satisfy its nutrient iron requirement by acquiring heme from host hemoglobin in the context of infection. However, heme acquisition exposes S. aureus to heme toxicity. In order to detect the presence of toxic levels of exogenous
heme, S. aureus is able to sense heme through the heme sensing system (HssRS) two-component
system. Upon sensing heme, HssRS directly regulates the expression of the heme-regulated ABC
transporter HrtAB, which alleviates heme toxicity. Importantly, the inability to sense or respond to
heme alters the virulence of S. aureus, highlighting the importance of heme sensing and detoxification to staphylococcal pathogenesis. Furthermore, potential orthologues of the Hss and Hrt systems
are found in many species of Gram-positive bacteria, a possible indication that heme stress is a challenge faced by bacteria whose habitats include host tissues rich in heme.
Copyright 2009 S. Karger AG, Basel

Staphylococcus aureus Pathogenesis

Staphylococcus aureus is a Gram-positive bacterial pathogen that is responsible for a


tremendous amount of morbidity and mortality [1]. S. aureus is the most frequent
causative agent of hospital-acquired infections and community-acquired infections
that necessitate admission to emergency rooms in the USA [2]. Antibiotic-resistant
strains of S. aureus such as community-acquired methicillin-resistant S. aureus
(CA-MRSA) are increasingly prevalent, even among otherwise healthy individuals [3,
4]. This point is underscored by the fact that MRSA infections are now responsible
for more US deaths annually than the human immunodeficiency virus [3, 5]. Taken
together, these facts have galvanized efforts aimed at identifying novel therapeutic
targets against S. aureus. One area that has received significant attention in this regard
is the identification of factors that allow S. aureus to adapt to and survive within the
hostile host environment.

A commensal of the skin and anterior nares, S. aureus enters tissues of the host
when epithelial barriers become compromised [68]. Following tissue invasion, S.
aureus is able to cause infection of virtually any anatomical site leading to diseases
ranging from skin and soft tissue infections to endocarditis, osteomyelitis, pneumonia, meningitis, and septicemia. Regardless of the site of infection, S. aureus invariably
encounters host blood during infection. Blood is the bearer of cells of the immune
system that are important for combating staphylococcal infection, as well as the carrier of nutrients that invasive bacteria must acquire in order to colonize and proliferate within host tissues [911].

Staphylococcal Heme Metabolism

One nutrient that S. aureus must acquire in the context of infection is iron. Iron is
required for a number of cellular processes including nucleotide biosynthesis, aerobic
respiration, and resistance to reactive oxygen species [12, 13]. Within host tissues, free
iron is actively maintained at concentrations that are too low to support the growth of
S. aureus and most other bacterial pathogens [12]. Nonetheless, host tissues are rich
in other forms of iron, such as iron in the context of the porphyrin heme. Heme is
the cofactor for hemoglobin within erythrocytes and myoglobin within myocytes and
therefore is highly abundant within the host [12, 14]. In order to satisfy its nutrient iron
requirement during infection, S. aureus expresses a set of cell wall-anchored, membrane-associated, and cytoplasmic proteins that are devoted to the acquisition of iron
from heme associated with hemoglobin [9, 10]. Designated the iron-regulated surface
determinants (Isd) and the heme transport system (Hts), these factors bind hemoglobin, remove the cofactor heme, traffic heme across the cell wall, import heme into the
cytoplasm, and degrade heme to release free iron to satisfy nutrient needs [9, 10, 1517].
Mutation of Isd or Hts family members attenuates the virulence of S. aureus, highlighting the importance of heme acquisition to staphylococcal pathogenesis [17, 18].

The Adaptive Response of S. aureus to Heme

Conceptually, heme acquisition on the part of a bacterial pathogen raises an interesting paradox. Although a number of bacterial pathogens express systems dedicated to
the acquisition of heme during infection, heme is a toxic molecule that is capable of
generating reactive oxygen species and damaging cells. Therefore, biological systems
that acquire, synthesize, or associate with heme must also possess systems devoted
to the avoidance of heme toxicity [10, 1922]. Given the significant association of
S. aureus with blood during infection and the efficiency of the staphylococcal heme
uptake machinery, it is likely that S. aureus expresses systems that protect this bacterium from the toxic side effects of heme acquisition during infection.

S. aureus Heme Sensing

121

An initial indication that S. aureus may express a heme detoxification system(s)


came from experiments designed to test whether S. aureus is sensitive to the toxic
effects of heme and whether the bacterium is able to respond accordingly [23]. These
experiments revealed two important observations. First, S. aureus is indeed sensitive
to heme toxicity. When S. aureus is subcultured into medium containing a high heme
concentration, no bacterial proliferation occurs as measured by growth curve analysis
due to a bactericidal activity that heme exerts on S. aureus [23]. Second, S. aureus is
able to sense the presence of exogenous heme and adapt to growth in its presence.
This can be demonstrated by pre-exposing S. aureus to a sub-toxic concentration
of heme, then subculturing the bacteria into medium containing an otherwise toxic
heme concentration and monitoring bacterial proliferation. When staphylococci are
pre-exposed to heme, appreciable growth occurs and cells are more resistant to the
bactericidal activity of heme [23]. These results suggest that S. aureus is capable of
sensing and adapting to the presence of exogenous heme, presumably through a coordinated change in the expression of a heme detoxification system.

The Heme Regulated Transporter (HrtAB)

To test whether S. aureus expresses a heme detoxification system in response to


growth in heme, a proteomics-based technique known as two-dimensional difference
gel electrophoresis (2D-DIGE) [24, 25] was utilized to search for proteins that change
in abundance when S. aureus is exposed to heme. 2D-DIGE allows a quantitative
comparison to be made of the abundance of individual proteins in a complex sample
across multiple conditions. Following quantitative comparison of proteins between
samples, individual proteins that change in abundance in a given condition can be
identified by mass spectrometry.
Using 2D-DIGE, 21 proteins that display alterations in abundance when S. aureus
is grown in high heme were identified [26]. Most of these proteins exhibit modest 2to 5-fold changes in expression, however a single protein is increased in abundance
an impressive 45-fold when S. aureus is grown in heme. This protein was identified by
mass spectrometry as the ATPase component of an uncharacterized staphylococcal
ABC transport system [23, 26]. Ubiquitous amongst bacteria and eukaryotic organisms, ABC transporters consist of an integral membrane permease and a peripherally
associated ATPase that together couple the cleavage of ATP to the translocation of
small molecules, ions, and proteins across the cell membrane [2729]. In order to
reflect the heme-responsive expression pattern of the ATPase component, this particular ABC transporter was named the heme-regulated ABC transporter (hrtAB; hrtA,
ATPase; hrtB, permease) [23, 26]. HrtA and HrtB are predicted to be co-transcribed,
an indication that they work together to form a functional transport system.
To test whether HrtAB plays a role in the previously described adaptive response
of S. aureus to heme toxicity, mutants that lack the members of the hrtAB operon

122

Stauff Skaar

were generated [23]. These mutants were tested for the ability to adapt to growth in
an otherwise toxic heme concentration by pre-exposure to heme and subculture into
medium containing a high level of heme. These experiments revealed that mutants
that lack either hrtA or hrtB fail to adapt to heme toxicity [23]. Furthermore, these
mutants are highly sensitive to the bactericidal effect of heme on staphylococci. These
findings indicate that HrtAB is critical for the ability of S. aureus to adapt to and proliferate in toxic concentrations of heme. Together with the heme-regulated expression
of HrtAB, these data point to a model in which S. aureus senses heme and upregulates
the expression of HrtAB, which results in relief from heme toxicity [23]. Given the
predicted function of HrtAB as an ABC transporter, it was initially hypothesized that
HrtAB exports excess heme that may accumulate in the staphylococcal cytoplasm
due to the efficiency of heme uptake by the Isd and Hts systems. However, it is equally
likely that HrtAB does not export heme itself from the cytoplasm, but exports byproducts of heme toxicity or heme metabolites that accumulate in bacterial cells upon
heme acquisition, and thus alleviates heme toxicity through this mechanism. In support of this later contention, mass spectrometry-based tracking experiments using
isotopically labeled heme have not revealed a role for HrtAB in staphylococcal heme
efflux [unpubl. data].
During systemic infections, S. aureus is intimately associated with vertebrate blood,
the most abundant source of heme in the host. Furthermore, the staphylococcal Isd
and Hts systems have evolved to rapidly and efficiently acquire heme as a nutrient
iron source. To cope with the toxicity inherent to the rapid intracellular amassing of
heme, S. aureus relies on the heme-dependent upregulation of HrtAB to avoid heme
toxicity [9, 10, 23, 30]. In keeping with this, it would appear likely that a mutant strain
of S. aureus with impaired resistance to heme toxicity would have a diminished ability to cause disease as such a strain might succumb to heme toxicity during infection.
However, S. aureus hrtA displays a remarkable phenotype in a murine abscess model
of systemic infection. Rather than being attenuated in its ability to infect mice and
cause disease, S. aureus hrtA exhibits increased virulence [23]. Mice infected with
S. aureus hrtA develop significant signs of disease early after infection and survive
for a reduced length of time compared to mice infected with wild-type S. aureus. This
increased virulence was further manifested upon autopsy, which revealed that infection with S. aureus hrtA leads to considerable liver abscess formation and increased
bacterial burden in this organ. Surprisingly, this hypervirulence appears to be liverspecific as examination of other organs in the same infection model does not reveal
differences in virulence between S. aureus hrtA and wild type.
A potential mechanistic explanation for the liver-specific hypervirulence of S.
aureus hrtA surfaced during comparison of the immune response to S. aureus wild
type and hrtA. Cell populations in the livers of mice infected with wild type and
hrtA were subjected to FACS-based analysis. These experiments revealed no appreciable strain-specific differences in the immune response to staphylococci colonizing
the kidneys and spleens of infected mice. Furthermore, examination of infected livers

S. aureus Heme Sensing

123

did not reveal any differences in the levels of a number of cells of the innate immune
system including natural killer cells, dendritic cells, and invariant natural killer T cells
[23]. However, significant differences between livers from mice infected with wild
type and hrtA were observed in terms of the number of granulocytes and phagocytes present at the site of infection [23]. Specifically, livers of mice infected with
hrtA harbor half the number of CD11b+/CD11c phagocytes and fewer CD11b+/
Ly6G+ granulocytes than those from mice infected with wild type [23]. This is consistent with histological analysis of both sets of livers that revealed fewer polymorphonuclear cells present in tissue infected with hrtA. These results suggest that S. aureus
hrtA either interferes with the recruitment of phagocytes and granulocytes to the
site of infection or that hrtA exhibits an increased ability to kill these cell types as
they migrate toward invading staphylococci.
S. aureus is notorious for its ability to combat cells of the immune system through
the secretion of effector molecules that interfere with leukocyte or lymphocyte function [31]. One possible explanation for the increased ability of S. aureus hrtA to
inhibit neutrophil recruitment to the site of infection is that when this heme-sensitive
strain is under heme stress, it alters the quantity or the types of effector molecules that
it secretes, thereby altering host cell function or survival. In order to test whether S.
aureus hrtA exhibits a heme-dependent alteration in the amount or types of secreted
effector molecules, microarray and secreted protein profile experiments were performed on S. aureus hrtA grown in the presence and absence of heme. These experiments revealed that upon exposure to heme, S. aureus hrtA dramatically increases
the expression and secretion of a number of immunomodulatory proteins that interfere with the recruitment of phagocytes to the site of infection [23, 30]. The secreted
effector response of S. aureus hrtA to heme stress seems to specifically inhibit the
recruitment of host immune cells to the site of infection rather than increase the lysis
of these cells, as the expression of a number of cytolytic exotoxins in this strain is
repressed [30]. This secreted protein response correlates with the observation that
fewer neutrophils and granulocytes are present in abscesses formed by S. aureus
hrtA, providing a possible explanation for the increased virulence of this strain [23].
These data are all consistent with a model in which S. aureus hrtA encounters host
heme during infection, which induces heme stress and causes this strain to secrete
effector molecules that interfere with host phagocyte recruitment and function [23,
30]. This enables S. aureus hrtA to form larger abscesses and reach higher levels that
wild-type S. aureus in the liver.
It is unclear why the virulence of hrtA is perturbed only in the livers of infected
mice and not in other organs; however, it is possible that S. aureus encounters greater
heme stress when infecting the murine liver than the kidney or spleen. In this regard,
it can be envisioned that hemoglobin catalysis in the liver following normal erythrocyte turnover as a result of intravascular hemolysis increases the amount of heme
available for interaction with invading staphylococci. It should be pointed out that
the heme stress experienced by S. aureus within the host must be at a level significant

124

Stauff Skaar

enough to alter virulence gene expression but not so dramatic as to negatively impact
growth of S. aureus hrtA during infection. The fact that S. aureus hrtA specifically
increases the expression of immunomodulatory virulence factors in response to heme
stress raises the tantalizing possibility that regulatory circuits exist within staphylococci that respond to stress and are capable of activating a secreted protein response
that specifically targets immune cell recruitment [30].
The ABC transporter HrtAB has been demonstrated to be critical for the ability
of S. aureus to proliferate in otherwise toxic heme levels [23, 26, 30]. Perturbation
of this system also increases the virulence of this important human pathogen [23].
However, the original observation of a heme-dependent increase in HrtAB expression that led to the discovery of this system is a particularly important finding,
suggesting the presence of a regulatory system that responds to heme and alters
gene expression accordingly [26]. In order to identify potential candidates for an
HrtAB-regulating system, the genomic context surrounding hrtAB in the S. aureus
chromosome was investigated for genes encoding potential regulatory systems.
This simple genomics-based approach was conceptually feasible because bacterial regulatory systems are often genomically co-localized with their cognate target
genes [32].
Interrogation of the S. aureus chromosome for potential HrtAB-regulating systems revealed the presence of two open reading frames immediately adjacent to
hrtAB that are predicted to be transcribed in the opposite direction from hrtAB [23].
Importantly, these genes are annotated as members of a bacterial two-component
system (TCS). This TCS is predicted to be composed of a membrane-localized histidine kinase and a cytoplasmic response regulator with a DNA-binding domain. In
general, membrane-localized histidine kinases recognize a particular environmental signal, which is thought to induce a conformational change in the protein, leading to autophosphorylation on a conserved histidine residue within the cytoplasmic
domain of the histidine kinase [33]. Response regulators catalyze transphosphorylation from the histidine kinase they recognize to a response regulator aspartic acid
residue, an event that increases the ability of these proteins to activate the transcription of target genes [34]. Interestingly, ABC transporter-TCS clusters have only been
found in the Gram-positive Firmicutes and are generally thought to be involved in
resistance to bacteriocins, which are antibacterial compounds made by one bacterial
strain to inhibit the growth of other surrounding bacteria [32]. It is thought that bacteria that synthesize a particular bacteriocin encode a TCS that recognizes this compound, controlling a resistance response. However, another ABC transporter-TCS
cluster in S. aureus (the GraRS/ApsRS TCS and the VraFG ABC transporter) has
recently been found to be involved in the resistance of this organism to the toxicity
of certain cell wall-active antibiotics as well as antimicrobial peptides [3537]. These
studies suggest that ABC transporter-TCS clusters may not be exclusively devoted
to bacteriocin resistance, but may be involved in the resistance to toxic host compounds or antimicrobial agents.

S. aureus Heme Sensing

125

The Heme Sensor System (HssRS)

Given the heme-regulated expression pattern of HrtAB and the fact that a TCS is
found co-localized with hrtAB on the S. aureus chromosome, a model was envisioned
in which this TCS responds to heme, thereby inducing the expression of HrtAB and
initiating the S. aureus heme detoxification response. To test whether the hrtABproximal TCS is involved in the response of S. aureus to heme toxicity, mutant strains
lacking either the histidine kinase or the response regulator gene were generated. As
with S. aureus hrtA, inactivation of either the response regulator or histidine kinase
leads to an inability to adapt to heme toxicity or proliferate in high heme concentrations [23]. Furthermore, a strain of S. aureus lacking the hrtA-proximal response
regulator displays an increased bacterial burden in the murine liver upon infection,
a similar phenotype observed for S. aureus hrtA [23]. This suggests that this TCS
functions along with HrtAB to control the heme toxicity response of S. aureus during
infection. This TCS was therefore designated the heme sensor system (hssRS; hssR,
response regulator; hssS, histidine kinase) [23].
The fact that HssRS is required for the adaptation of S. aureus to heme toxicity as
well as the predicted function of HssRS as a sensor system that regulates gene expression together imply that HssRS directly controls transcription of hrtAB. Three experiments were carried out to test this hypothesis. First, the level of HrtAB transcript was
measured by RT-PCR in wild-type and S. aureus hssR grown in the presence or
absence of heme. In wild-type S. aureus, heme induces an increase in the level of HrtAB
transcript, suggesting that the heme-regulated expression of HrtAB occurs at the level
of gene expression [23]. This heme-dependent increase in the HrtAB transcript is not
observed in a S. aureus strain lacking the hssR gene, implying that the HssRS TCS is
critical for the ability of S. aureus to increase the expression of HrtAB in response to
heme. Second, a plasmid-based reporter construct was generated to indirectly monitor HrtAB expression by fusing the hrtAB promoter to the coding sequence for the
XylE reporter enzyme. Wild-type S. aureus harboring this hrtAB promoter-xylE fusion
exhibits a significant heme-dependent increase in reporter activity, further indicating that the heme-dependent expression of HrtAB occurs through an increase in synthesis of the HrtAB transcript driven by the hrtAB promoter [23, 38]. Furthermore,
no heme-dependent increase in reporter activity occurs in S. aureus hssR or hssS,
establishing the importance of the HssRS TCS in the regulation of HrtAB expression
in response to heme [23, 38]. Third, 2D-DIGE was utilized to determine the abundance of all of the S. aureus cytoplasmic proteins in wild type and hssR grown in the
presence and absence of heme [38]. This experiment revealed two important findings. First, HrtA is increased in abundance in response to heme in wild type but not
in hssR, further establishing the importance of HssRS in the regulation of HrtAB
expression [38]. Second, HrtA is the only cytoplasmic protein that depends on hssR
for a heme-dependent increase in abundance [38]. This suggests that the HssRS TCS
may specifically regulate the expression of HrtAB rather than a set of genes [38].

126

Stauff Skaar

These experiments demonstrate a dependence on HssRS for the response of S.


aureus to heme toxicity through the regulation of HrtAB expression. However, they
do not delineate the signaling events in S. aureus that connect HssRS to HrtAB
expression. A series of experiments was therefore carried out to elucidate the mechanism by which HssRS regulates HrtAB expression in response to heme. Potential
residues required for HssS-HssR signaling were identified by aligning the amino acid
sequence of the predicted HssS cytoplasmic domain with the cytoplasmic domains
from a number of known and characterized TCS histidine kinases. These experiments revealed a highly conserved histidine residue in HssS (His249) that aligns
with the histidine residue known to be phosphorylated in response to signal recognition in other histidine kinases [38]. Similar alignments with HssR revealed a
conserved aspartic acid residue (Asp52) that aligns with those residues known to
undergo phosphorylation in other response regulators [38]. Based on the identification of these residues, it was postulated that upon signal recognition, HssS catalyzes
autophosphorylation at His249 and that HssR subsequently catalyzes transphosphorylation to Asp52, activating the latter protein. To test this hypothetical signal transduction pathway, recombinant HssS and HssR as well as their predicted
phosphotransfer site mutants were purified and tested for in vitro signaling. While
HssS undergoes time-dependent autophosphorylation in these experiments, HssS
mutated at its conserved histidine residue fails to undergo phosphorylation [38].
When HssR is added to phosphorylated HssS, HssR is able to catalyze transphosphorylation, indicating that these proteins are capable of phosphotransfer in vitro [38].
Furthermore, HssR mutated at its predicted phosphorylation site fails to catalyze
transphosphorylation from HssS [38]. These experiments suggest that HssS undergoes autophosphorylation at His249 and that HssR catalyzes transphosphorylation
to Asp52. However, these experiments were all carried out in vitro and therefore
do not fully establish the importance of these signaling events in the functioning of
HssRS in S. aureus.
To this end, the HssS and HssR mutants that fail to undergo signaling in vitro were
tested for functionality in vivo. This was accomplished by creating plasmids containing a full-length copy of either wild-type HssS or HssS mutated at His249, which
is required for in vitro autophosphorylation. Strains harboring these plasmids were
tested for the ability to rescue the heme-sensitive phenotype of hssS by growth curve
analyses. Importantly, the plasmid bearing wild-type HssS rescues the growth of S.
aureus lacking the hssS gene while the plasmid containing the mutant form of HssS
that fails to undergo in vitro autophosphorylation is unable to rescue hssS [38]. These
differences occur in spite of significant expression of both proteins. Furthermore, a
similar result is seen with HssR: while wild-type HssR rescues S. aureus lacking the
hssR gene, HssR mutated at the aspartic acid residue required for in vitro phosphotransfer is unable to complement hssR even though both proteins are expressed by S.
aureus [38]. Cumulatively, these results suggest that HssS-HssR signaling is essential
for the ability of S. aureus to respond to the toxicity of heme.

S. aureus Heme Sensing

127

Given the observation that HssRS is required for an increase in transcription of


hrtAB driven by the hrtAB promoter and the fact that HssR is modeled to contain a
DNA-binding domain, it was predicted that HssR binds directly to the HrtAB promoter upon phosphorylation. To test this prediction, recombinant HssR was phosphorylated in vitro and added to magnetic beads coated with DNA corresponding to
the hrtAB promoter. Proteins that associate with the hrtAB promoter are precipitated
along with the magnetic beads used in this assay and can then be eluted from the
beads and analyzed by immunoblot analysis. It was found that phosphorylated HssR
is able to bind to such beads coated with hrtAB promoter DNA [38]. Importantly,
phosphorylated HssR does not associate with beads coated with hrtAB intragenic
DNA and non-phosphorylated HssR does not bind to the hrtAB promoter in vitro
[38]. Furthermore, HssR mutated at its phosphorylation residue is not able to bind
to the hrtAB promoter [38]. These experiments suggest that upon phosphorylation,
HssR binds directly to the hrtAB promoter. To test the in vivo relevance of these
observations, pull-down experiments using DNA-coated magnetic beads similar to
those described above were performed using cytoplasmic extracts from S. aureus
grown in the presence and absence of heme. These experiments revealed that HssR
binds to the hrtAB promoter when S. aureus encounters heme, and that no binding of
HssR to the hrtAB promoter occurs when S. aureus is grown in medium lacking heme
[38]. Together, these results suggest that HssS senses heme and initiates a signaling
pathway that leads to HssR phosphorylation and binding of phosphorylated HssR to
the hrtAB promoter.
Response regulators such as HssR that contain DNA-binding domains usually
exhibit specificity for certain sites within the promoters of the genes they regulate.
Such sites often consist of short direct repeat sequences to which the effector domain
of a response regulator is thought to associate through a head-to-tail dimerization
mechanism [34]. Accordingly, the hrtAB promoter was interrogated for direct repeat
sequences as a means of identifying potential HssR-binding sites. These analyses
revealed a perfect direct repeat sequence upstream of the hrtAB coding sequence
(GTTCATATT(N2)GTTCATATT) [38]. To test whether this direct repeat is involved
in the response of the hrtAB promoter to heme-dependent activation by HssRS, the
hrtAB promoter-xylE fusion reporter construct was subjected to mutational analyses.
Truncation of the hrtAB promoter up to this direct repeat fails to alter activity of
the hrtAB promoter, but truncation of half or all of this repeat completely eliminates
hrtAB promoter heme responsiveness [38]. Furthermore, mutation of four residues
within the hrtAB promoter that are conserved in the hrtAB promoters of other species of staphylococci also eliminates the response of this promoter to heme [38]. This
suggests that the hrtAB promoter direct repeat is required for the ability of this promoter to respond to heme through HssRS. In light of the fact that HssR binds directly
to the hrtAB promoter, in vitro promoter-binding experiments were carried out to
test whether HssR binds to this direct repeat. To this end, HssR was eluted from the
hrtAB promoter with increasing concentrations of double-stranded oligonucleotides

128

Stauff Skaar

corresponding to either the wild-type hrtAB promoter direct repeat or a pseudodirect repeat mutated at the four residues required for the functioning of the hrtAB
promoter in vivo. While the intact direct repeat elutes HssR, the mutated direct repeat
fails to elute HssR from the hrtAB promoter [38]. These results suggest that upon
phosphorylation, HssR binds to the direct repeat within the hrtAB promoter.
Together, the known features of HssS-HssR signaling and the mechanism by which
these events impinge on the hrtAB promoter to regulate expression of HrtAB are
shown in schematic form in figure 1. Nonetheless, a number of aspects of the HssRS
and HrtAB systems remain to be elucidated. First, the transport substrate for HrtAB
is unknown. Originally, it was proposed that HrtAB removes excess heme from the
staphylococcal cytoplasm in order to alleviate heme toxicity [23, 30]. However, it is
equally possible that a toxic molecule accumulates in S. aureus upon intracellular
amassing of exogenously acquired heme. It can also be envisioned that breakdown
of heme by S. aureus results in the buildup of toxic heme metabolites that must be
exported from the staphylococcal cytoplasm. More studies focused on the function of
HrtAB need to be carried in order to understand the mechanism of heme toxicity in
S. aureus and the means by which HrtAB alleviates heme toxicity.
Another aspect of HssRS/HrtAB function that is not fully understood is the
nature of the signal that is directly sensed by the histidine kinase HssS. Although
HssS may sense heme by direct binding, HssS also could sense one of the toxic effects
that heme has on the cell. Genomics-based approaches have not revealed a potential
heme-binding domain in HssS, and the predicted sensing domain of this kinase has
little sequence similarity to any protein with a known function [23]. Furthermore,
no amino acids that typically engage in axial coordination of the iron atom of heme
are conserved in the putative sensing domain of all of the potential HssS orthologues
[unpubl. observations]. Identification of the ligand for HssS will lay the groundwork
for a mechanistic understanding of how this histidine kinase is activated upon heme
exposure. This would represent a significant advance for understanding HssS function and the mechanism employed by S. aureus to regulate the tight balance between
heme acquisition and heme toxicity.
HssS represents one of the few histidine kinases that respond to a molecular
marker of vertebrate tissue and one of two histidine kinases known to respond to
heme. Importantly, other TCS that recognize heme are found in the Gram-positive
pathogen Corynebacterium diphtheriae, a bacterium that is not closely related to S.
aureus [3942]. In C. diphtheriae, the TCS ChrAS and HrrAS are capable of responding to heme or hemoproteins [41]. However, in addition to the absence of any significant sequence identity between the staphylococcal and corynebacterial systems,
two major aspects of ChrAS and HrrAS distinguish them from the S. aureus HssRS
system. First, ChrAS and HrrAS both regulate expression of the C. diphtheriae heme
oxygenase HmuO, while HssRS does not appear to regulate expression of the S.
aureus heme oxygenases [38, 41, 42]. HmuO is capable of catabolizing heme and is
therefore a critical factor required for heme metabolism in C. diphtheriae [43]. In

S. aureus Heme Sensing

129

Fe

Cell wall

HssS

HrtA

HrtB

HrtB

Isd/
Hts

Membrane

Cytoplasm

HrtA

HssR
P

Fe
HrtAB

HssR
P
5

HssR

Fig. 1. Signaling events that connect heme sensing by the HssRS TCS to upregulation of HrtAB and
heme detoxification: (1) Heme is imported into the staphylococcal cytoplasm through the actions of
the Isd and Hts systems. (2) Heme activates the histidine kinase HssS directly through a receptorligand interaction, or indirectly through the heme-mediated accumulation of toxic metabolites. (3)
Activation of HssS results in autophosphorylation and (4) transphosphorylation of HssR. (5)
Phosphorylated HssR binds to a direct repeat sequence within the hrtAB promoter. (6) This increases
the transcription of the hrtAB genes by recruitment of RNA polymerase. (7) Newly synthesized HrtAB
alleviates heme toxicity (8) through the export of heme or an unknown molecule that accumulates
in the S. aureus cytoplasm upon heme exposure.

S. aureus, IsdG and IsdI are iron-regulated heme oxygenases that also break down
heme [10, 16, 44]. However, 2D-DIGE experiments have failed to reveal a role for
HssRS in the regulation of IsdG/I expression, indicating that HssRS does not regulate
these heme-degrading enzymes in S. aureus [38]. Second, despite the fact that HssRS
from S. aureus and ChrAS/HrrAS from C. diphtheriae both respond to heme, the
histidine kinase HssS and the kinases ChrS/HrrS differ dramatically in terms of the
predicted structure of their signal-sensing domains. While HssS is predicted to have
an N-terminal periplasmic sensing domain flanked by two transmembrane helices,
ChrS/HrrS are predicted to have sensing domains composed of five transmembrane
helices flanked by short loops [23, 42]. This difference in predicted structure suggests
that HssS may sense heme through a mechanism that is different from that of ChrS/
HrrS or that these TCS might sense different effects that heme has on the Grampositive bacterial cell. Interestingly, C. diphtheriae strains lacking the chrAS genes
exhibit elevated heme sensitivity that is not dependent on reduced HmuO expression
[39]. This suggests the presence of a ChrAS-regulated heme-detoxification system

130

Stauff Skaar

in C. diphtheriae [39]. It is tempting to speculate that C. diphtheriae encodes a system functionally analogous to the S. aureus HrtAB system that protects this organism
from heme toxicity.
The ability of bacterial pathogens to sense and respond to heme is not limited to
Gram-positive organisms. The Gram-negative mammalian pathogen Bordetella pertussis also encodes a heme sensing system [4551]. In B. pertussis, the heme receptor
BhuR binds to heme, transducing signals into the periplasm that activate the cytoplasmic membrane protein RhuR [46, 48, 50]. RhuR alters the activity of the cytoplasmic
protein RhuI, which regulates the expression of the B. pertussis heme uptake system
BhuRSTUV [46, 48, 50]. Through this mechanism, B. pertussis is able to sense exogenous heme and increase the expression of systems required for heme uptake from
the host. Thus the B. pertussis RhuIR/BhuR system differs from the HssRS system of
S. aureus in that the former is devoted to the sensing of host heme for the purposes of
heme acquisition, whereas the latter is required for sensing host heme to avoid heme
toxicity [23, 38, 4551]. Furthermore, HssRS is a TCS, while RhuI is an extracytoplasmic sigma factor and RhuR is a regulator of RhuI activity. Thus, the mechanism
by which heme is sensed by these systems as well as the details of signaling are likely
distinct. It appears that the pathogens S. aureus, C. diphtheriae and B. pertussis are
all able to sense host heme and transduce signals that alter gene expression, but that
these events occur through distinct mechanisms and for different reasons.

Evolutionary Implications of Heme Sensing in Gram-Positive Bacteria

A role for the TCS HssRS and the heme-regulated ABC transporter HrtAB has been
established in the ability of S. aureus to avoid heme toxicity as well as in the pathogenesis of this organism [23, 38]. Interestingly, a number of distinct Gram-positive
bacteria also encode potential orthologues of these systems (fig. 2) [23]. Most of
these bacteria are either pathogens or saprotrophs (microorganisms that break down
tissues of dead animals) and thus are likely to encounter host heme. For example,
Bacillus anthracis, the cause of anthrax infection, is a bacterium that has a significant bloodstream component throughout the late stages of systemic infection [52].
B. anthracis also is capable of inducing hemolysis and has been shown to acquire
iron from heme, potentially exposing this organism to heme toxicity during infection [5356]. Accordingly, B. anthracis encodes potential HssRS/HrtAB orthologues
(fig. 2) and has been shown to adapt to heme toxicity [23]. Although the putative
B. anthracis Hss and Hrt systems have not been experimentally proven to be functional in this organism or orthologous to the corresponding systems in S. aureus,
these initial observations suggest that B. anthracis must respond to heme toxicity at
some point throughout its life cycle. Potential Hss and Hrt genes are also found in
the pathogenic or saprotrophic Listeriae such as L. monocytogenes and L. innocua as
well as other Bacilli including B. cereus and B. thuringiensis (fig. 2). Exiguobacterium

S. aureus Heme Sensing

131

hrtA

hrtB

hssR

hssS

other
S. aureus COL
S. epidermidis ATCC12228

69.7

53.1

70.5

64.5

68.0

56.7

68.3

63.4

72.9

63.4

67.9

60.4

47.9

29.7

36.5

52.2

48.0

29.9

51.8

35.9

49.3

30.9

49.3

30.9

34.4

48.4

48.9

31.4

34.4

48.0

49.8

30.6

43.2

29.7

44.9

25.8

S. haemolyticus JCSC1435
S. saprophyticus ATCC15305
L. innocua Clip11262
L. monocytogenes EGD-e

35.3

35.1

**
34.4
30.2

42.5

B. anthracis Ames

48.4

48.4

B. cereus ATCC14579
B. weihenstephanensis
KBAB4

B. thuringiensis AI Hakam

B. thuringiensis israelensis
E. sibiricum 255-15

Fig. 2. Conservation of HssRS and HrtAB across Gram-positive bacteria. Sequenced bacterial
genomes were interrogated for potential orthologues of S. aureus hssRS and hrtAB genes by BLAST
analyses. The indicated bacterial strains were found to encode potential Hss/Hrt systems. Boxes indicate open reading frames and point in the direction of transcription. Dark grey boxes: HrtA; light
grey boxes: HrtB, empty boxes: HssR, boxes with diagonal lines: HssS; black boxes: non-Hss/Hrt open
reading frames (other genes). Numbers within boxes indicate the annotated gene number. Numbers
below boxes are percent similarities with respect to the corresponding S. aureus COL Hss/Hrt gene.
Percent similarities were calculated at the amino acid level using a Lipman-Pearson Protein Alignment
(gap penalty = 4, gap length penalty = 12). Single asterisk denotes large separation between genes;
double asterisk denotes that this potential HssR orthologue is split into two open reading frames in
the genome of this species.

sibiricum, an extremophile isolated from Siberian permafrost, also encodes potential


Hss/Hrt systems (fig. 2). The details of the life cycle of this organism are currently
unknown, but the fact that it encodes potential Hss/Hrt systems suggests that it may
experience exogenous heme exposure in its natural habitat. Importantly, Hss and
Hrt genes appear to be absent from the genomes of a number of non-pathogenic and
non-saprotrophic Gram-positive bacteria including B. subtilis and B. licheniformis,
a possible indication that Hss and Hrt are only required for the fitness of bacteria that encounter exogenous heme throughout their life cycles. The fact that Hss

132

Stauff Skaar

and Hrt genes are widespread throughout Gram-positive bacteria that associate with
heme-containing tissues suggests that the ability of these organisms to overcome
heme toxicity has been critical for their capacity to occupy different niches, whether
pathogenic or saprotrophic. Furthermore, the fact that genes encoding Hss and Hrt
are present in a number of human pathogens establishes the members of these systems as potential targets for therapeutic agents that could interfere with bacterial
pathogenesis through the perturbation of heme metabolism. This idea is supported
by the alteration in virulence observed upon inactivation of the staphylococcal Hrt
systems.

Acknowledgments
We would like to thank members of the Skaar laboratory for reading of the manuscript. This work
was supported by the Searle Scholars Program, United States Public Health Service Grant AI69233
from the National Institute of Allergy and Infections Diseases, and the Southeast Regional Centers
of Excellence in Emerging Infections and Biodefense (SERCEB). E.P.S. holds an Investigator in
Pathogenesis of Infectious Disease Award from the Burroughs Wellcome Fund. D.L.S. was supported by T32 HL069765 from the National Institute of Allergy and Infectious Diseases.

References
1 Fridkin SK, Hageman JC, Morrison M, Sanza LT,
Como-Sabetti K, Jernigan JA, Harriman K, Harrison
LH, Lynfield R, Farley MM: Methicillin-resistant
Staphylococcus aureus disease in three communities.
N Engl J Med 2005;352:14361444.
2 Moran GJ, Krishnadasan A, Gorwitz RJ, Fosheim
GE, McDougal LK, Carey RB, Talan DA: Methicillinresistant S. aureus infections among patients in the
emergency department. N Engl J Med 2006;355:666
674.
3 Klevens RM, Morrison MA, Nadle J, Petit S,
Gershman K, Ray S, Harrison LH, Lynfield R,
Dumyati G, Townes JM, Craig AS, Zell ER, Fosheim
GE, McDougal LK, Carey RB, Fridkin SK: Invasive
methicillin-resistant Staphylococcus aureus infections in the United States. JAMA 2007;298:1763
1771.
4 Creech CB 2nd, Kernodle DS, Alsentzer A, Wilson
C, Edwards KM: Increasing rates of nasal carriage of
methicillin-resistant Staphylococcus aureus in
healthy children. Pediatr Infect Dis J 2005;24:617
621.
5 CDC: HIV/AIDS Surveillance Report. 2007;17.

S. aureus Heme Sensing

6 Kuehnert MJ, Kruszon-Moran D, Hill HA,


McQuillan G, McAllister SK, Fosheim G, McDougal
LK, Chaitram J, Jensen B, Fridkin SK, Killgore G,
Tenover FC: Prevalence of Staphylococcus aureus
nasal colonization in the United States, 20012002. J
Infect Dis 2006;193:172179.
7 Wertheim HF, Melles DC, Vos MC, van Leeuwen W,
van Belkum A, Verbrugh HA, Nouwen JL: The role
of nasal carriage in Staphylococcus aureus infections.
Lancet Infect Dis 2005;5:751762.
8 Creech CB 2nd, Talbot TR, Schaffner W: Community-associated methicillin-resistant Staphylococcus aureus: the way to the wound is through the
nose. J Infect Dis 2006;193:169171.
9 Skaar EP, Schneewind O: Iron-regulated surface
determinants (Isd) of Staphylococcus aureus: stealing iron from heme. Microbes Infect 2004;6:390
397.
10 Reniere ML, Torres VJ, Skaar EP: Intracellular metalloporphyrin metabolism in Staphylococcus aureus.
Biometals 2007;20:333345.
11 Corbin BD, Seeley EH, Raab A, Feldmann J, Miller
MR, Torres VJ, Anderson KL, Dattilo BM, Dunman
PM, Gerads R, Caprioli RM, Nacken W, Chazin WJ,
Skaar EP: Metal chelation and inhibition of bacterial growth in tissue abscesses. Science 2008;319:
962965.

133

12 Bullen JJ, Griffiths, E.: Iron and Infection: Molecular,


Physiological and Clinical Aspects. New York,
Wiley, 1999.
13 Jordan A, Reichard P: Ribonucleotide reductases.
Annu Rev Biochem 1998;67:7198.
14 Crichton R: Inorganic Biochemistry of Iron
Metabolism: From Molecular Mechanisms to
Clinical Consequences. Chichester, Wiley, 2001.
15 Mazmanian SK, Skaar EP, Gaspar AH, Humayun
M, Gornicki P, Jelenska J, Joachmiak A, Missiakas
DM, Schneewind O: Passage of heme-iron across
the envelope of Staphylococcus aureus. Science 2003;
299:906909.
16 Skaar EP, Gaspar AH, Schneewind O: IsdG and IsdI,
heme-degrading enzymes in the cytoplasm of
Staphylococcus aureus. J Biol Chem 2004;279:436
443.
17 Torres VJ, Pishchany G, Humayun M, Schneewind
O, Skaar EP: Staphylococcus aureus IsdB is a hemoglobin receptor required for heme iron utilization. J
Bacteriol 2006;188:84218429.
18 Skaar EP, Humayun M, Bae T, DeBord KL,
Schneewind O: Iron-source preference of Staphylococcus aureus infections. Science 2004;305:1626
1628.
19 Graca-Souza AV, Maya-Monteiro C, Paiva-Silva
GO, Braz GR, Paes MC, Sorgine MH, Oliveira MF,
Oliveira PL: Adaptations against heme toxicity in
blood-feeding arthropods. Insect Biochem Mol Biol
2006;36:322335.
20 Tolosano E, Fagoonee S, Hirsch E, Berger FG,
Baumann H, Silengo L, Altruda F: Enhanced splenomegaly and severe liver inflammation in haptoglobin/hemopexin double-null mice after acute
hemolysis. Blood 2002;100:42014208.
21 Stojiljkovic I, Evavold BD, Kumar V: Antimicrobial
properties of porphyrins. Expert Opin Investig
Drugs 2001;10:309320.
22 Francis SE, Sullivan DJ Jr, Goldberg DE: Hemoglobin
metabolism in the malaria parasite Plasmodium falciparum. Annu Rev Microbiol 1997;51:97123.
23 Torres VJ, Stauff DL, Pishchany G, Bezbradica JS,
Gordy LE, Iturregui J, Anderson KL, Dunman P,
Joyce S, Skaar EP: A Staphylococcus aureus regulatory system that responds to host heme and modulates virulence. Cell Host Microbe 2007;1:109119.
24 Friedman DB: Quantitative proteomics for twodimensional gels using difference gel electrophoresis; in Matthiesen R (ed): Mass Spectrometry Data
Analysis in Proteomics. Totowa/NJ, Humana Press,
2006.
25 Lilley KS, Friedman DB: All about DIGE: quantification technology for differential-display 2D gel
proteomics. Expert Rev Proteomics 2004;1:401
409.

134

26 Friedman DB, Stauff DL, Pishchany G, Whitwell


CW, Torres VJ, Skaar EP: Staphylococcus aureus
redirects central metabolism to increase iron availability. PLoS Pathog 2006;2.
27 Borst P, Elferink RO: Mammalian ABC transporters
in health and disease. Annu Rev Biochem 2002;
71:537592.
28 Pohl A, Devaux PF, Herrmann A: Function of
prokaryotic and eukaryotic ABC proteins in lipid
transport. Biochim Biophys Acta 2005;1733:2952.
29 Young J, Holland IB: ABC transporters: bacterial
exporters-revisited five years on. Biochim Biophys
Acta 1999;1461:177200.
30 Stauff DL, Bagaley D, Torres VJ, Joyce R, Anderson
KL, Kuechenmeister L, Dunman PM, Skaar EP:
Staphylococcus aureus HrtA is an ATPase required
for protection against heme toxicity and prevention
of a transcriptional heme stress response. J Bacteriol
2008;190:35883596.
31 Foster TJ: Immune evasion by staphylococci. Nat
Rev Microbiol 2005;3:948958.
32 Joseph P, Fichant G, Quentin Y, Denizot F: Regulatory
relationship of two-component and ABC transport
systems and clustering of their genes in the Bacillus/
Clostridium group, suggest a functional link between
them. J Mol Microbiol Biotechnol 2002;4:503513.
33 Mascher T, Helmann JD, Unden G: Stimulus perception in bacterial signal-transducing histidine
kinases. Microbiol Mol Biol Rev 2006;70:910938.
34 Gao R, Mack TR, Stock AM: Bacterial response
regulators: versatile regulatory strategies from common domains. Trends Biochem Sci 2007;32:225
234.
35 Li M, Lai Y, Villaruz AE, Cha DJ, Sturdevant DE,
Otto M: Gram-positive three-component antimicrobial peptide-sensing system. Proc Natl Acad Sci
USA 2007;104:94699474.
36 Li M, Cha DJ, Lai Y, Villaruz AE, Sturdevant DE,
Otto M: The antimicrobial peptide-sensing system
aps of Staphylococcus aureus. Mol Microbiol
2007;66:11361147.
37 Meehl M, Herbert S, Gotz F, Cheung A: Interaction
of the GraRS two-component system with the
VraFG ABC transporter to support vancomycinintermediate resistance in Staphylococcus aureus.
Antimicrob Agents Chemother 2007;51:26792689.
38 Stauff DL, Torres VJ, Skaar EP: Signaling and DNAbinding activities of the Staphylococcus aureus
HssR-HssS two-component system required for
heme sensing. J Biol Chem 2007;282:2611126121.
39 Bibb LA, King ND, Kunkle CA, Schmitt MP:
Analysis of a heme-dependent signal transduction
system in Corynebacterium diphtheriae: deletion of
the chrAS genes results in heme sensitivity and
diminished heme-dependent activation of the hmuO
promoter. Infect Immun 2005;73:74067412.

Stauff Skaar

40 Schmitt MP: Transcription of the Corynebacterium


diphtheriae hmuO gene is regulated by iron and
heme. Infect Immun 1997;65:46344641.
41 Schmitt MP: Identification of a two-component signal transduction system from Corynebacterium
diphtheriae that activates gene expression in response
to the presence of heme and hemoglobin. J Bacteriol
1999;181:53305340.
42 Bibb LA, Kunkle CA, Schmitt MP: The ChrA-ChrS
and HrrA-HrrS signal transduction systems are
required for activation of the hmuO promoter and
repression of the hemA promoter in Corynebacterium
diphtheriae. Infect Immun 2007;75:24212431.
43 Schmitt MP: Utilization of host iron sources by
Corynebacterium diphtheriae: identification of a
gene whose product is homologous to eukaryotic
heme oxygenases and is required for acquisition of
iron from heme and hemoglobin. J Bacteriol 1997;
179:838845.
44 Wu R, Skaar EP, Zhang R, Joachimiak G, Gornicki P,
Schneewind O, Joachimiak A: Staphylococcus aureus
IsdG and IsdI, heme-degrading enzymes with structural similarity to monooxygenases. J Biol Chem
2005;280:28402846.
45 Kirby AE, Metzger DJ, Murphy ER, Connell TD:
Heme utilization in Bordetella avium is regulated by
RhuI, a heme-responsive extracytoplasmic function
sigma factor. Infect Immun 2001;69:69516961.
46 King ND, Kirby AE, Connell TD: Transcriptional
control of the rhuIR-bhuRSTUV heme acquisition
locus in Bordetella avium. Infect Immun 2005;73:
16131624.
47 Murphy ER, Sacco RE, Dickenson A, Metzger DJ,
Hu Y, Orndorff PE, Connell TD: BhuR, a virulenceassociated outer membrane protein of Bordetella
avium, is required for the acquisition of iron from
heme and hemoproteins. Infect Immun 2002;70:
53905403.

48 Kirby AE, King ND, Connell TD: RhuR, an extracytoplasmic function sigma factor activator, is essential for heme-dependent expression of the outer
membrane heme and hemoprotein receptor of
Bordetella avium. Infect Immun 2004;72:896907.
49 Vanderpool CK, Armstrong SK: The Bordetella bhu
locus is required for heme iron utilization. J Bacteriol
2001;183:42784287.
50 Vanderpool CK, Armstrong SK: Heme-responsive
transcriptional activation of Bordetella bhu genes. J
Bacteriol 2003;185:909917.
51 Brickman TJ, Vanderpool CK, Armstrong SK: Heme
transport contributes to in vivo fitness of Bordetella
pertussis during primary infection in mice. Infect
Immun 2006;74:17411744.
52 Mock M, Fouet A: Anthrax. Annu Rev Microbiol
2001;55:647671.
53 Shannon JG, Ross CL, Koehler TM, Rest RF:
Characterization of anthrolysin O, the Bacillus
anthracis cholesterol-dependent cytolysin. Infect
Immun 2003;71:31833189.
54 Skaar EP, Gaspar AH, Schneewind O: Bacillus
anthracis IsdG, a heme-degrading monooxygenase.
J Bacteriol 2006;188:10711080.
55 Gaspar AH, Marraffini LA, Glass EM, Debord KL,
Ton-That H, Schneewind O: Bacillus anthracis
sortase A (SrtA) anchors LPXTG motif-containing
surface proteins to the cell wall envelope. J Bacteriol
2005;187:46464655.
56 Maresso AW, Chapa TJ, Schneewind O: Surface protein IsdC and Sortase B are required for heme-iron
scavenging of Bacillus anthracis. J Bacteriol 2006;188:81458152.

Eric P. Skaar
Department of Microbiology and Immunology, Vanderbilt University Medical Center
21st Avenue South, Medical Center North, Room A5102
Nashville, TN 37232 (USA)
Tel. +1 615 343 0002, Fax +1 615 343 7392, E-Mail eric.skaar@vanderbilt.edu

S. aureus Heme Sensing

135

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 136149

Bacterial Sensing of Antimicrobial Peptides


Michael Otto
Laboratory of Human Bacterial Pathogenesis, National Institute of Allergy and Infectious Diseases,
The National Institutes of Health, Bethesda, Md., USA

Abstract
Antimicrobial peptides (AMPs) form a crucial part of human innate host defense, especially in neutrophil phagosomes and on epithelial surfaces. Bacteria have a variety of efficient resistance mechanisms to human AMPs, such as efflux pumps, secreted proteases, and alterations of the bacterial cell
surface that are aimed to minimize attraction of the typically cationic AMPs. In addition, bacteria
have specific sensors that activate AMP resistance mechanisms when AMPs are present. The prototypical Gram-negative PhoP/PhoQ and the Gram-positive Aps AMP-sensing systems were first
described and investigated in Salmonella typhimurium and Staphylococcus epidermidis, respectively.
Both include a classical bacterial two-component sensor/regulator system, but show many structural, mechanistic, and functional differences. The PhoP/PhoQ regulon controls a variety of genes
not necessarily limited to AMP resistance mechanisms, but apparently aimed to combat innate host
defense on a broad scale. In contrast, the staphylococcal Aps system predominantly upregulates
AMP resistance mechanisms, namely the D-alanylation of teichoic acids, inclusion of lysyl-phosphatidylglycerol in the cytoplasmic membrane, and expression of the putative VraFG AMP efflux pump.
Notably, both systems are crucial for virulence and represent possible targets for antimicrobial
Copyright 2009 S. Karger AG, Basel
therapy.

The first line of defense against invading microorganisms is formed by the human
innate immune system. This system reacts swiftly long before the acquired host
defense, which is dependent on the production of antibodies, becomes effective. It
comprises phagocytes such as neutrophils that recognize invariant structures on the
microorganisms surface (pathogen-associated molecular patterns, PAMPs) and kill
microorganisms by the combined activities of reactive oxygen species and antimicrobial proteins and peptides [1, 2]. Additionally, antimicrobial peptides (AMPs) are
secreted by epithelial cells and other cell types [3]. They form an especially important
part of host defense on epithelial and mucosal surfaces such as in the gut and on
the skin, and contribute significantly to balancing the composition of the colonizing microflora. While AMPs are evolutionarily ancient and in lower organisms often
represent the only available mechanism of immune defense, their important role in

higher organisms, including humans, has been recognized more recently. In addition
to their microbicidal activities, they may also function as signaling molecules, connecting the innate and acquired immune systems by the activation of immune cell
types such as T cells [4].
There are three major AMP classes in humans: the defensins, cathelicidins, and
thrombocidins (table 1). Defensins have a -sheet structure and 3 disulfide bridges.
They are further classified by their disulfide bridging pattern, distinguishing and
defensins. The only cathelicidin produced in humans is the -helical LL-37. Finally,
while defensins and cathelicidins are produced by a variety of cell types that include
neutrophils, the thrombocidins are released from platelets. All these AMPs are cationic (CAMPs), which is a typical feature of most AMPs and believed to have evolved
to interact with the negatively charged bacterial surface [5]. However, there are also
anionic AMPs. For example, the anionic peptide DCD-1L, a proteolytic product of
dermcidin, is found in human sweat [6].

Mechanisms of Bacterial Resistance to Antimicrobial Peptides

As a consequence of the long interplay between bacteria and other microorganisms


with host AMPs during evolution, bacteria have invented a series of mechanisms to
combat AMPs [5]. These mechanisms can be classified in mechanisms aimed at (1)
destruction of AMPs, (2) change of the AMP target to make it less susceptible to
AMPs, and (3) removal of AMPs from their site of action (table 2).
Many AMPs can be efficiently inactivated via proteolytic digestion by secreted
bacterial proteases, such as Staphylococcus epidermidis and Staphylococcus aureus
aureolysins [7, 8], PgtE from Salmonella enterica serovar Typhimurium [9], and several others from diverse microbial species [10]. However, during evolution hosts have
learned to produce AMPs that are less susceptible to this relatively non-specific bacterial mechanism of evading AMP activity. The frequent presence of multiple disulfide
bridges in AMPs is believed to have arisen to provide resistance to proteolytic digestion by bacterial proteases [5].
Most AMPs function by forming pores in the cytoplasmic membrane of bacteria.
As membrane integration of the AMP is dependent on the physicochemical properties of the membrane, resistance can be achieved by altering those properties, for
example by altering fluidity of the phospholipid bilayer via introduction of fatty acids
with different chain lengths [11].
Mechanisms of AMP removal from the site of action are the most frequently
found mechanisms of AMP resistance. They comprise two major mechanisms, efflux
pumps and alteration of the bacterial cell surface composition. Furthermore, the
streptococcal SIC proteins (streptococcal inhibitors of complement) and the S. aureus
staphylokinase interact specifically with AMPs to prevent them from reaching the
cytoplasmic membrane [12, 13].

Bacterial Sensing of Antimicrobial Peptides

137

Table 1. Sequences of major human antimicrobial peptides


Class

Peptide

Sequence

-Defensins

HNP1
HNP2
HNP3
HNP4
HD5
HD6

ACYCRIPACIAGERRYGTCIYQGRLWAFCC
CYCRIPACIAGERRYGTCIYQGRLWAFCC
DCYCRIPACIAGERRYGTCIYQGRLWAFCC
VCSCRLVFCRRTELRVGNCLIGGVSFTYCCTRV
ATCYCRTGRCATRESLSGVCEISGRLYRLCCR
CHCRRSCYSTEYSYGTCTVMGINHRFCC

-Defensins

HBD1
HBD2
HBD3
HBD4
HBD5

DHYNCVSSGGQCLYSACPIFTKIQGTCYRGKAKCCK
TCLKSGAICHPVFCPRRYKQIGTCGLPGTKCCKKP
GIINTLQKYYCRVRGGRCAVLSCLPKEEQIGKCSTRGRKCCRRKK
EFELDRICGYGTARCRKKCRSQEYRIGRCPNTYACCLRKWDESLLNRTKP
GLDFSQPFPSGEFAVCESCKLGRGKCRKECLENEKPDGNCRLNFLCCRQRI

Cathelicidin

LL-37

LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES

Dermcidin

DCD-1,
DCD-1L

SSLLEKGLDGAKKAVGGLGKLGKDAVED LESVGKGAVHDVKDVLDSV(L)

Efflux pumps such as the MtrCE system of Neisseria gonorrhoeae and QacA of S.
aureus work by energy-dependent constant export of toxic substances out of the cell
and away from the membrane [14, 15]. Although they primarily export hydrophobic
drugs, they have relatively low specificity and may also accept a subset of AMPs as
substrates [16, 17]. However, in the case of S. aureus QacA, it has been shown that
resistance to the platelet AMP, tPMP-1 is not mediated by efflux pump activity. This
exemplifies that real transport phenomena should always be shown by biochemical studies to rule out that low-level resistance is mediated by mere binding effects
[18].
Mechanisms based on alteration of the bacterial surface often differ between
Gram-positive and Gram-negative bacteria owing to the different composition of
the Gram-positive and Gram-negative cell surface. In the Gram-negative Salmonella
typhimurium, AMP resistance is achieved by modification of lipid A either by acylation (PagP) [19] or addition of an aminoarabinose moiety (Pmr system) [20]. In
the Gram-positive S. aureus, substitution of teichoic acids with d-alanine confers
AMP resistance by decreasing the negative net charge of teichoic acids, which leads
to diminished attraction of CAMPs [21]. This mechanism is catalyzed by enzymes
encoded in the dlt locus and contributes to CAMP resistance in many Gram-positive
bacteria, for example Streptococcus pneumonia and Listeria monocytogenes [22, 23].
Furthermore, formation and integration into the membrane of lysyl-phosphatidylglycerol (L-PG), a phospholipid that is distinguished by a positive net charge not
found in other common bacterial membrane phospholipids, equally leads to reduced

138

Otto

Table 2. Mechanisms of resistance to antimicrobial peptides


Mechanism
Proteolytic cleavage
PgtE
OmpT
Aureolysin, SepA, V8 protease
Unidentified proteases

External AMP-binding proteins


SIC protein
Staphylokinase
AMP exporters
MtrCDE
EpiFEG
RosA/RosB
Alteration of cell surface charge
Modification of lipid A
Teichoic acid alanylation
Phospholipid lysinylation
Other mechanisms
Mycolic acid modification
Reduction of cytoplasmic membrane
potential
Production of protective exopolymers
Inhibition of AMP production

Species

Ref.

Salmonella enterica
Escherichia coli
Staphylococcus aureus,
Staphylococcus epidermidis
Pseudomonas aeruginosa,
Enterococcus faecalis, Proteus mirabilis,
Porphyromonas gingivalis, Prevotella ssp.

9
65
7, 8

Streptococcus pyogenes
Staphylococcus aureus

66
13

Neisseria gonorrhoeae
Staphylococcus epidermidis,
many bacteriocin producers
Yersinia ssp.

67
68, 69

Salmonella enterica,
many Gram-negative bacteria
Staphylococcus aureus,
many Gram-positive bacteria
Staphylococcus aureus, many bacteria

71

Mycobacterium tuberculosis
Staphylococcus aureus

73
74

Klebsiella pneumonia, Staphylococcus


epidermidis, many other bacteria
Shigella spp.

26, 28, 75

10

70

21, 72
24

30

attraction of CAMPs to the cytoplasmic membrane [24] and inhibits killing by neutrophils [25]. The transmembrane enzyme that catalyzes formation and likely integration
of L-PG into the membrane, MprF, is present in many bacteria and thus represents a
widespread mechanism of AMP resistance. It is important to stress that these latter
mechanisms are specific for CAMPs. The presence of anionic AMPs in human sweat
likely represents a response of the human innate immune system to circumvent those
mechanisms.
Resistance mechanisms against anionic AMPs have only been investigated more
recently. Interestingly, in S. epidermidis the cationic exopolymer polysaccharide

Bacterial Sensing of Antimicrobial Peptides

139

intercellular adhesin (PIA) and the anionic poly--glutamic acid (PGA) have been
shown to protect from cationic and anionic AMPs [2628]. These results indicate
that charged bacterial exoploymers do not work exclusively by electrostatic repulsion such as in the cases of teichoic acid d-alanylation. Alternative mechanisms that
may explain the findings with PIA and PGA include charge-independent mechanical
exclusion and charge-dependent sequestration of AMPs, a mechanism also proposed
for example for the cationic antibiotic tobramycin that is sequestered by the anionic
Pseudomonas aeruginosa exopolymer alginate [29]. Alginate and PIA are typical components of the extracellular matrix in bacterial biofilms, and possibly main contributors to the observed high resistance of biofilms to AMPs.
Finally, some yet poorly understood mechanisms of AMP resistance may be unique
to some bacteria. For example, Shigella species appear to inhibit production of LL-37
and human -defensin 1 in human rectal epithelial cells [30].

The Gram-Negative PhoP/PhoQ Antimicrobial Peptide Sensor

The PhoP/PhoQ system is a virulence regulator originally described in S. typhimurium [31, 32]. It regulates a series of genes and cellular activities [33], such as Mg2+
transport (mgtA and mgtCB genes) [34], survival in macrophages (mgtCB genes)
[35], modification of lipopolysaccharide (LPS, pagP gene) [19, 36], and also activates
a further two-component regulatory system, pmrA/pmrB, by activation of pmrD transcription [37]. The pmrA/pmrB system also upregulates genes involved in LPS modification and resistance to polymyxin B (pbgP, pbgE, ugd) [37].
PhoP/PhoQ is a classical and one of the most intensively studied bacterial twocomponent signal transduction systems [31, 32]. PhoQ is the histidine kinase sensor part that is present as a dimer in the bacterial cytoplasmic membrane. It has
two transmembrane helices, one extracellular (periplasmic) loop of 145 amino acid
length, and a cytoplasmic domain that contains a conserved autophosphorylation
and an ATP-binding site. Upon stimulation, PhoQ trans-autophosphorylates within
the dimer, and a phosphate is transferred to a conserved aspartate residue on PhoP.
PhoP is the response regulator DNA-binding protein of the system that interacts with
target gene promoters (fig. 1).
Three stimulants of the PhoQ sensor have been described: low concentration of
divalent cations [38], low pH [39], and CAMPs [40]. Recent studies have shown that
PhoQ is repressed in the presence of divalent cations such as Mg2+ or Ca2+ [40]. By
solving the crystal structure of PhoQ in the presence of Ca2+, it was found that these
divalent cations bind to a set of acidic residues termed the acidic patch that is adjacent
to two -helices [41]. Binding of Ca2+ ions alleviates some of the charge repulsion
between the acidic patch and the membrane that can be predicted from the crystal
structure and which is believed to be important during the activation process. NMR
studies have shown that apparently, there is no difference in the structure of PhoQ

140

Otto

Teichoic acids

AMP

Outer
membrane
Ca2+

at
ch

Peptidoglycan

ac
id
ic
p

AMP

Ca2+

Ca2+

ApsX

ApsS

PhoQ

ApsR

P
Phop

dlt (D-alanylation of teichoic acids)


mprF (Lysyl-phosphatidylglycerol
integration into the membrane)
vraFG (putative AMP efflux pump)

pagP (LPS modification)


mgtA, mgtCB (Mg2+ transport)
pmrD (activation of pmrAB system)

Fig. 1. The Gram-negative and Gram-positive AMP sensors. a The Staphylococcus aps system. The
Gram-positive cell surface, containing a thick layer of peptidoglycan and anionic teichoic acids, is
shown schematically. Binding of a cationic AMP leads to activation of the system in a mechanistically
yet poorly understood fashion. The extracellular loop of the ApsS sensor is extremely short, carrying
three negative charges that presumably are crucial for the interaction with the cationic AMP. A phosphorylation cascade as typical for two-component systems is believed to activate target gene transcription, with the ApsX protein participating in signal transduction in an unknown way. It is not
known whether the ApsS sensor forms a dimer, but dimer formation of sensor proteins is common.
b The Salmonella PhoP/PhoQ system. Activation by CAMPs, which penetrate the Gram-negative outer
membrane in a yet undefined manner, occurs by deplacement of divalent cations in an acidic patch
of the PhoQ sensor. The PhoQ acidic region is much larger than the acidic extracellular loop of ApsS.
This leads to structural rearrangements including notably two -helices that move away from the
membrane, and ultimately autophosphorylation of ApsS and subsequent phosphorylation of PhoP.
Main regulatory targets of the systems are shown at the bottom. While both systems control many
targets, those involved in CAMP resistance are the dlt, mprF, and vraFG loci in Staphylococcus, and
the pagP locus in Salmonella. Furthermore, the pbgPE operon, which is also involved in CAMP resistance by lipid A modification, is regulated by pmrAB, which is under control of phoPQ.

with Mg2+ versus Ca2+, indicating that the type of divalent ion is not critical for keeping the PhoQ dimer in a repressed state [41].
Removing repression of PhoQ by lowering the concentration of divalent cations
in principle would lead to activation of the system. However, this is unlikely to represent a signal that occurs in vivo, because metal ions are abundant in host tissues and
fluids. Rather, activation of the PhoP/PhoQ system is believed to be triggered in vivo
by low pH and CAMPs. Bader et al. [40] reconstituted PhoQ in membrane vesicles
and showed that CAMPs lead to increased levels of phosphorylated PhoP and target

Bacterial Sensing of Antimicrobial Peptides

141

gene activation in that system. Activation by CAMPs can be competed out by increasing concentrations of Mg2+, indicating that CAMPs and divalent cations bind to the
same site in PhoQ. Based on these studies, a model was developed in which CAMPs
substitute divalent cations at the acidic patch and owing to their larger size, thereby
force two -helices of the PhoQ protein away from the membrane, ultimately leading
to PhoQ activation [40, 42].
Similarly to CAMPs, PhoQ can be activated by low pH. This was also shown
using reconstituted PhoQ in membrane vesicles [40]. It is important to stress that
the effects of pH and CAMPs on PhoQ phosphorylation are additive, suggesting a
different mechanism for the interaction of these two signals with the sensor protein.
Interestingly, activation by low pH does not involve loss of capacity to bind divalent
cations, which would represent a hypothetical mechanism for pH-dependent activation of PhoQ, but leads to increased flexibility of the PhoQ protein. This increased
flexibility would allow the two -helices to move in a model similar to the one proposed for PhoQ activation by CAMPs.
Activation by CAMPs and pH occurs in the presence of ~1 mm Mg2+, which is
about the divalent cation concentration in a neutrophil phagosome [40]. Thus, these
two signals are good candidates for stimuli that activate PhoQ in vivo. Furthermore,
the additive effects of pH and CAMPs on PhoQ indicate that for maximal activation,
both signals need to be present. This limits the expression of PhoQ-regulated CAMP
resistance mechanisms to situations that are relevant for pathogen survival in vivo,
notably in the phagosome and the acidic intestinal microenvironment around Paneth
cells, which secrete the CAMP HD-5 (human -defensin 5).
While only investigated in detail in Salmonella, PhoQ homologues have been found
in a variety of Gram-negative pathogens. In some of those such as Shigella flexneri,
Yersinia pestis, the insect pathogen Photorhabdus luminescens, and the plant pathogen
Erwinia chrysanthemi, there are results that suggest that PhoQ is also involved in sensing CAMPs and low pH and affects virulence [4347]. In contrast to these eukaryoteassociated bacteria, the acidic patch crucial for CAMP recognition is missing in P.
aeruginosa, which may cause infection, but is primarily a bacterium found in soil and
water. Sensing CAMPs with strongly varying structure is much more essential for
eukaryote-associated pathogens, indicating that either the acidic patch in PhoQ has
evolved during the interaction with the host, or has been lost in organisms in which
CAMP sensing is not as crucial for survival [42].
Despite the evidence outlined so far, the role of the PhoP/PhoQ system in CAMP
and pH sensing is debatable. It has been stressed that the PhoP regulon comprises
many more genes than those activated by the CAMP polymyxin B [48]. Furthermore,
PmrA/PmrB appears to interact with PhoP/PhoQ post-translationally via PmrD and is
primarily involved in low pH sensing. Finally, it has been argued that most CAMPs are
larger than porins in the outer membrane of Gram-negative bacteria and thus, would
cause perturbation that may be sensed by other systems before PhoQ can be reached.
For example, polymyxin B promoted the expression of genes regulated by RcsB and

142

Otto

the alternative sigma factor RpoS. Taken together, it appears that while the interaction
of CAMPs with PhoQ has been studied on a very detailed molecular level, there may
be more systems involved in CAMP sensing in a complicated regulatory network.

The Gram-Positive Aps Antimicrobial Peptide Sensor

While the Gram-negative PhoP/PhoQ system as a regulator of virulence and CAMP


resistance has been known for a while and its role in CAMP sensing has been
described about 2 years ago [40], a Gram-positive sensor with an equivalent function
in CAMP sensing and regulation of Gram-positive CAMP resistance mechanisms has
remained elusive. The PhoR/PhoP system described in some Gram-positive bacteria
to respond to inorganic phosphate and control virulence [49] must not be confused
with the Gram-negative PhoP/PhoQ sensor. Recently, during studies aimed to determine genome-wide responses to the human AMP human -defensin 3 (hBD3) in the
nosocomial pathogen and skin colonizer S. epidermidis, using transcriptional profiling with microarrays, my group identified a Gram-positive CAMP sensor/gene regulator [50]. The AMP human -defensin 3 (hBD3), the only -defensin that maintains
anti-staphylococcal activity at physiological salt concentration as found in the natural
habitat of skin-colonizing staphylococci [51], induced expression of the major staphylococcal mechanisms involved in CAMP resistance: the dlt operon for d-alanylation
of teichoic acids and the mprF gene for formation of L-PG. In addition, several transporters were upregulated that may function to remove CAMP from the cytoplasmic
membrane. Exceptionally strong upregulation was detected for the VraFG transporter,
previously described to confer resistance to the glycopeptide antibiotic vancomycin
[52]. The vraFG genes are located in the S. epidermidis and S. aureus genome adjacent to an apparent operon consisting of genes coding for a two-component system
and a third gene without striking similarity to genes with known function. Based
on the fact that in bacteria, regulatory systems are often involved in the regulation
of adjacent genes, we constructed deletion mutants in each of the three genes and
characterized genome-wide changes in gene expression compared to the wild-type
strain. The results showed that this system, which we termed aps for antimicrobial
peptide sensor, regulates the major CAMP resistance mechanisms, dlt and mprF, in
addition to vraFG. Notably, all three aps components were essential for gene regulation, indicating that this system represents an unusual three-component regulatory
system consisting of a classical two-component system with a sensor histidine kinase
and a DNA-binding response regulator, whereas the third component has a yet undefined role in signal transduction. The apsS and apsR genes have also been called graS
and graR due to their involvement in resistance to glycopeptide antibiotics [53, 54].
We suggest using the aps name, because it is based on the presumably natural role of
this system and publications on graRS have failed to recognize the three-component
nature of the regulatory system.

Bacterial Sensing of Antimicrobial Peptides

143

Similar to results achieved in Salmonella with the PhoP/PhoQ system [33], analysis of the aps regulon using deletion mutants revealed genes that were not found to
be under regulation of CAMPs [50]. In addition, differential expression of autolysins
was observed in all three aps gene deletion mutants. However, it needs to be stressed
that the deletion of a gene represents an unnatural situation that is not likely to be
achieved in this extreme fashion by differential phosphorylation of the ApsS/ApsR/
(ApsX), or the PhoP/PhoQ regulatory cascades. Thus, these results do not necessarily contradict the notion that the primary tasks of these systems are the sensing of
CAMPs and regulation of CAMP resistance mechanisms.
The ApsS sensor is a transmembrane protein with two predicted transmembrane
helices and an extracellular acidic loop, features that are reminiscent of the PhoQ
architecture. However, there is no significant sequence similarity and most strikingly,
the ApsS extracellular loop is only 9 amino acids long and thus much smaller than the
acidic patch of PhoQ (fig. 1). Interaction of this 9 amino acid loop with CAMPs has
been shown using specific antibodies developed against the loop epitope that blocked
upregulation of aps target genes [50].
In contrast to the S. epidermidis aps system, which was activated by any CAMP
tested, we found activation of the S. aureus aps system only with specific CAMPs,
including indolicidin and LL-37, for example, but not using hBD3 and several other
peptides [55]. Using complementation vectors that expressed heterologous S. epidermidis or hybrid proteins consisting of the S. aureus apsS gene with an S. epidermidis loop inserted in an apsS deletion mutant of S. aureus, we could demonstrate
that substrate specificity of the system is due to amino acid sequence differences in
the loop between S. aureus and S. epidermidis. The biological significance of the fact
that S. aureus is more discriminatory in accepting only some CAMPs as aps stimuli
is not known. Further mechanistic details on AMP sensing by ApsS are not known
yet. However, divalent cations do not appear to have a significant AMP-dependent
impact on aps-dependent gene regulation [Li and Otto, unpubl. data], suggesting
that the regulators involved in divalent cation-dependent gene expression are different from aps and the mechanism of ApsS activation differs from that used by
PhoQ.
In addition to regulating the dlt, mprF, and vraFG loci (with a somewhat differing relative emphasis compared to S. epidermidis), the aps regulon in S. aureus comprises genes involved in the biosynthesis of lysine [55]. This is in accordance with
the increased need for lysine when L-PG production via MprF is stimulated. Thus,
basic metabolic adaptation supports the specific upregulation of CAMP resistance
mechanisms during stimulation of the aps regulon by CAMPs. It needs to be stressed
that these results achieved in the hypervirulent community-associated MRSA strain
MW2 [56, 57] are contradictory to those achieved by other authors, who did not find
the mprF and lysine biosynthesis genes in the graRS regulon of S. aureus strain SA113
[58], which is a mutant in the global regulatory system agr [59]. Thus, there might be
strain-specific differences in the aps regulon in S. aureus.

144

Otto

Sensing CAMPs and responding with dedicated resistance mechanisms is supposed


to be important for staphylococcal virulence. In a peritoneal mouse infection model,
increased CFU counts of the wild-type strain in comparison to the apsS deletion mutant
strain in the kidneys indicated a role for the aps system in S. aureus infection [55].
Homologues of the aps system are found in a multitude of Gram-positive pathogens,
including Clostridium difficile, L. monocytogenes, Bacillus anthracis, Staphylococcus
haemolyticus, and S. pneumonia [50]. Interestingly, homologues of the ApsX third
component are only found in staphylococcal species. Thus, determining the role of
ApsX in the staphylococcal Aps system and those of the ApsS homologues in other
bacteria will be an important task for future research.

Anionic Antimicrobial Peptides and Non-Specific Sensing

The recently discovered human anionic AMP dermcidin does not activate the aps
system due to its negative net charge and thus represents a valuable means to investigate gene regulatory responses to an AMP not mediated via that sensor [50]. In S.
epidermidis, dermcidin (in its proteolytically processed, active form called DCD-1) led
to an upregulation of the gene coding for the secreted protease SepA, and increased
extracellular proteolytic activity [7]. SepA had a major role in degrading DCD-1, while
it only had marginal effects on hBD3 and LL-37, which may be explained by the secondary and tertiary structures of these AMPs. Global regulatory systems, namely the
agr, saeRS, and sarA regulators were changed in a fashion to promote increased production of secreted proteases including SepA. Notably, these regulatory changes were
also achieved by incubation with hBD3, indicating that they are independent of AMP
charge. Interestingly, it has been found recently that the saeRS system responds to a
series of phagocyte-derived signals [60] in addition to AMPs, suggesting a key role in
the regulation of immune evasion mechanisms in S. aureus. These results indicate that
while anionic AMPs may be interpreted as an adaptation of human innate host defense
to bacterial AMP resistance mechanisms specific for CAMPs, bacteria still have less
specific ways to combat those peptides. These comprise sensing mechanisms that do
not discriminate between positively and negatively charged AMPs. The detailed mechanism of the staphylococcal adaptive response to dermcidin and other AMPs in an
aps-independent manner needs to be further analyzed. It appears to involve a broad
gene-regulatory response most likely caused by membrane perturbations and is thus
reminiscent of the Gram-negative rpoS-dependent response to polymyxin B.

A Potential Target for Antimicrobial Therapeutics?

In a time of desperate search for new antimicrobial agents and targets, it has been frequently proposed to develop antimicrobial therapeutics based on AMPs [61]. While

Bacterial Sensing of Antimicrobial Peptides

145

the commonly bactericidal mode of action of AMPs is a clear advantage, preexisting


bacterial resistance mechanisms and AMP sensors such as those discussed herein,
in addition to the sensitivity of AMPs to proteolytic digestion, make this approach
appear problematic. Nevertheless, due to the empty pipeline in antibiotic development and strongly increasing antibiotic resistance in premier pathogens such as S.
aureus, a closer look at AMPs as potential therapeutics is certainly warranted. In that
context, targeting AMP sensor mechanisms involved in regulating bacterial resistance to AMPs may form a valuable part [62]. The therapeutic benefit from targeting a regulatory system is debatable for many reasons. For example, regulators are
commonly not essential to the microbe and some, such as staphylococcal agr, may
regulate virulence factors in opposing fashion [63]. Thus, blocking regulators may
sometimes lead to increased expression of virulence determinants [64]. However, the
fact that AMP sensors are fairly conserved among Gram-negative or Gram-positive
bacteria is encouraging in that regard.

Acknowledgement
This work was supported by the Intramural Program of the National Institute of Allergy and
Infectious Diseases, NIH.

References
1 Faurschou M, Borregaard N: Neutrophil granules
and secretory vesicles in inflammation. Microbes
Infect 2003;5:13171327.
2 Heine H, Ulmer AJ: Recognition of bacterial products by toll-like receptors. Chem Immunol Allergy
2005;86:99119.
3 Hancock RE, Diamond G: The role of cationic antimicrobial peptides in innate host defences. Trends
Microbiol 2000;8:402410.
4 Bowdish DM, Davidson DJ, Scott MG, Hancock RE:
Immunomodulatory activities of small host defense
peptides. Antimicrob Agents Chemother 2005;49:
17271732.
5 Peschel A, Sahl HG: The co-evolution of host cationic antimicrobial peptides and microbial resistance. Nat Rev Microbiol 2006;4:529536.
6 Schittek B, Hipfel R, Sauer B, Bauer J, Kalbacher H,
Stevanovic S, Schirle M, Schroeder K, Blin N, Meier
F, Rassner G, Garbe C: Dermcidin: a novel human
antibiotic peptide secreted by sweat glands. Nat
Immunol 2001;2:11331137.
7 Lai Y, Villaruz AE, Li M, Cha DJ, Sturdevant DE,
Otto M: The human anionic antimicrobial peptide
dermcidin induces proteolytic defence mechanisms
in staphylococci. Mol Microbiol 2007;63:497506.

146

8 Sieprawska-Lupa M, Mydel P, Krawczyk K, Wojcik K,


Puklo M, Lupa B, Suder P, Silberring J, Reed M, Pohl
J, Shafer W, McAleese F, Foster T, Travis J, Potempa J:
Degradation of human antimicrobial peptide ll-37 by
Staphylococcus aureus-derived proteinases. Antimicrob Agents Chemother 2004;48:46734679.
9 Guina T, Yi EC, Wang H, Hackett M, Miller SI: A
PhoP-regulated outer membrane protease of
Salmonella enterica serovar typhimurium promotes
resistance to -helical antimicrobial peptides. J
Bacteriol 2000;182:40774086.
10 Schmidtchen A, Frick IM, Andersson E, Tapper H,
Bjorck L: Proteinases of common pathogenic bacteria degrade and inactivate the antibacterial peptide
ll-37. Mol Microbiol 2002;46:157168.
11 Jing W, Prenner EJ, Vogel HJ, Waring AJ, Lehrer RI,
Lohner K: Headgroup structure and fatty acid chain
length of the acidic phospholipids modulate the
interaction of membrane mimetic vesicles with the
antimicrobial peptide protegrin-1. J Pept Sci 2005;
11:735743.
12 Akesson P, Sjoholm AG, Bjorck L: Protein sic, a
novel extracellular protein of Streptococcus pyogenes
interfering with complement function. J Biol Chem
1996;271:10811088.

Otto

13 Jin T, Bokarewa M, Foster T, Mitchell J, Higgins J,


Tarkowski A: Staphylococcus aureus resists human
defensins by production of staphylokinase, a novel
bacterial evasion mechanism. J Immunol 2004;172:
11691176.
14 Hagman KE, Pan W, Spratt BG, Balthazar JT, Judd
RC, Shafer WM: Resistance of Neisseria gonorrhoeae
to antimicrobial hydrophobic agents is modulated
by the mtrRCDE efflux system. Microbiology 1995;
141:611622.
15 Rouch DA, Cram DS, DiBerardino D, Littlejohn
TG, Skurray RA: Efflux-mediated antiseptic resistance gene qacA from Staphylococcus aureus: common ancestry with tetracycline- and sugar-transport
proteins. Mol Microbiol 1990;4:20512062.
16 Kupferwasser LI, Skurray RA, Brown MH, Firth N,
Yeaman MR, Bayer AS: Plasmid-mediated resistance to thrombin-induced platelet microbicidal
protein in staphylococci: role of the qacA locus.
Antimicrob Agents Chemother 1999;43:23952399.
17 Tzeng YL, Ambrose KD, Zughaier S, Zhou X, Miller
YK, Shafer WM, Stephens DS: Cationic antimicrobial peptide resistance in Neisseria meningitidis. J
Bacteriol 2005;187:53875396.
18 Bayer AS, Kupferwasser LI, Brown MH, Skurray
RA, Grkovic S, Jones T, Mukhopadhay K, Yeaman
MR: Low-level resistance of Staphylococcus aureus
to thrombin-induced platelet microbicidal protein-1
in vitro associated with qacA gene carriage is independent of multidrug efflux pump activity.
Antimicrob Agents Chemother 2006;50:24482454.
19 Guo L, Lim KB, Poduje CM, Daniel M, Gunn JS,
Hackett M, Miller SI: Lipid A acylation and bacterial resistance against vertebrate antimicrobial peptides. Cell 1998;95:189198.
20 Gunn JS, Lim KB, Krueger J, Kim K, Guo L, Hackett
M, Miller SI: PmrA-PmrB-regulated genes necessary for 4-aminoarabinose lipid A modification and
polymyxin resistance. Mol Microbiol 1998;27:1171
1182.
21 Peschel A, Otto M, Jack RW, Kalbacher H, Jung G,
Gotz F: Inactivation of the dlt operon in Staphylococcus
aureus confers sensitivity to defensins, protegrins,
and other antimicrobial peptides. J Biol Chem 1999;
274:84058410.
22 Abachin E, Poyart C, Pellegrini E, Milohanic E,
Fiedler F, Berche P, Trieu-Cuot P: Formation of
d-alanyl-lipoteichoic acid is required for adhesion
and virulence of Listeria monocytogenes. Mol
Microbiol 2002;43:114.

Bacterial Sensing of Antimicrobial Peptides

23 Kovacs M, Halfmann A, Fedtke I, Heintz M, Peschel


A, Vollmer W, Hakenbeck R, Bruckner R: A functional dlt operon, encoding proteins required for
incorporation of d-alanine in teichoic acids in
Gram-positive bacteria, confers resistance to cationic antimicrobial peptides in Streptococcus pneumoniae. J Bacteriol 2006;188:57975805.
24 Peschel A, Jack RW, Otto M, Collins LV, Staubitz P,
Nicholson G, Kalbacher H, Nieuwenhuizen WF,
Jung G, Tarkowski A, van Kessel KP, van Strijp JA:
Staphylococcus aureus resistance to human defensins
and evasion of neutrophil killing via the novel virulence factor MprF is based on modification of membrane lipids with l-lysine. J Exp Med 2001;193:
10671076.
25 Kristian SA, Durr M, Van Strijp JA, Neumeister B,
Peschel A: MprF-mediated lysinylation of phospholipids in Staphylococcus aureus leads to protection
against oxygen-independent neutrophil killing.
Infect Immun 2003;71:546549.
26 Kocianova S, Vuong C, Yao Y, Voyich JM, Fischer
ER, DeLeo FR, Otto M: Key role of poly--dlglutamic acid in immune evasion and virulence of
Staphylococcus epidermidis. J Clin Invest 2005;115:
688694.
27 Vuong C, Kocianova S, Voyich JM, Yao Y, Fischer
ER, DeLeo FR, Otto M: A crucial role for exopolysaccharide modification in bacterial biofilm formation, immune evasion, and virulence. J Biol Chem
2004;279:5488154886.
28 Vuong C, Voyich JM, Fischer ER, Braughton KR,
Whitney AR, DeLeo FR, Otto M: Polysaccharide
intercellular adhesin protects Staphylococcus epidermidis against major components of the human
innate immune system. Cell Microbiol 2004;6:269
275.
29 Mah TF, Pitts B, Pellock B, Walker GC, Stewart PS,
OToole GA: A genetic basis for Pseudomonas aeruginosa biofilm antibiotic resistance. Nature 2003;426:
306310.
30 Islam D, Bandholtz L, Nilsson J, Wigzell H,
Christensson B, Agerberth B, Gudmundsson G:
Downregulation of bactericidal peptides in enteric
infections: a novel immune escape mechanism with
bacterial DNA as a potential regulator. Nat Med
2001;7:180185.
31 Groisman EA, Chiao E, Lipps CJ, Heffron F:
Salmonella typhimurium PhoP virulence gene is a
transcriptional regulator. Proc Natl Acad Sci USA
1989;86:70777081.
32 Miller SI, Kukral AM, Mekalanos JJ: A two-component regulatory system (phoP phoQ) controls
Salmonella typhimurium virulence. Proc Natl Acad
Sci USA 1989;86:50545058.

147

33 Groisman EA: The pleiotropic two-component


regulatory system PhoP-PhoQ. J Bacteriol 2001;
183:18351842.
34 Soncini FC, Garcia Vescovi E, Solomon F, Groisman
EA: Molecular basis of the magnesium deprivation
response in Salmonella typhimurium: identification
of PhoP-regulated genes. J Bacteriol 1996;178:5092
5099.
35 Smith RL, Kaczmarek MT, Kucharski LM, Maguire
ME: Magnesium transport in Salmonella typhimurium: regulation of mgtA and mgtCB during invasion of epithelial and macrophage cells. Microbiology
1998;144:18351843.
36 Belden WJ, Miller SI: Further characterization of
the PhoP regulon: identification of new PhoPactivated virulence loci. Infect Immun 1994;62:5095
5101.
37 Gunn JS, Miller SI: PhoP-PhoQ activates transcription of pmrAB, encoding a two-component regulatory system involved in Salmonella typhimurium
antimicrobial peptide resistance. J Bacteriol 1996;
178:68576864.
38 Garcia Vescovi E, Soncini FC, Groisman EA: Mg2+
as an extracellular signal: environmental regulation
of salmonella virulence. Cell 1996;84:165174.
39 Alpuche Aranda CM, Swanson JA, Loomis WP,
Miller SI: Salmonella typhimurium activates virulence gene transcription within acidified macrophage phagosomes. Proc Natl Acad Sci USA
1992;89:1007910083.
40 Bader MW, Sanowar S, Daley ME, Schneider AR,
Cho U, Xu W, Klevit RE, Le Moual H, Miller SI:
Recognition of antimicrobial peptides by a bacterial
sensor kinase. Cell 2005;122:461472.
41 Cho US, Bader MW, Amaya MF, Daley ME, Klevit
RE, Miller SI, Xu W: Metal bridges between the
PhoQ sensor domain and the membrane regulate
transmembrane signaling. J Mol Biol 2006;356:1193
1206.
42 Prost LR, Miller SI: The salmonellae PhoQ sensor:
mechanisms of detection of phagosome signals. Cell
Microbiol 2008;10:576582.
43 Derzelle S, Turlin E, Duchaud E, Pages S, Kunst F,
Givaudan A, Danchin A: The PhoP-PhoQ twocomponent regulatory system of Photorhabdus
luminescens is essential for virulence in insects. J
Bacteriol 2004;186:12701279.
44 Moss JE, Fisher PE, Vick B, Groisman EA,
Zychlinsky A: The regulatory protein PhoP controls
susceptibility to the host inflammatory response in
Shigella flexneri. Cell Microbiol 2000;2:443452.

148

45 Llama-Palacios A, Lopez-Solanilla E, RodriguezPalenzuela P: Role of the PhoP-PhoQ system in the


virulence of Erwinia chrysanthemi strain 3937:
Involvement in sensitivity to plant antimicrobial
peptides, survival at acid Hh, and regulation of pectolytic enzymes. J Bacteriol 2005;187:21572162.
46 Rebeil R, Ernst RK, Gowen BB, Miller SI,
Hinnebusch BJ: Variation in lipid A structure in the
pathogenic yersiniae. Mol Microbiol 2004;52:1363
1373.
47 Oyston PC, Dorrell N, Williams K, Li SR, Green M,
Titball RW, Wren BW: The response regulator PhoP
is important for survival under conditions of macrophage-induced stress and virulence in Yersinia
pestis. Infect Immun 2000;68:34193425.
48 Groisman EA, Mouslim C: Sensing by bacterial
regulatory systems in host and non-host environments. Nat Rev Microbiol 2006;4:705709.
49 Hulett FM: The signal-transduction network for
Pho regulation in Bacillus subtilis. Mol Microbiol
1996;19:933939.
50 Li M, Lai Y, Villaruz AE, Cha DJ, Sturdevant DE,
Otto M: Gram-positive three-component antimicrobial peptide-sensing system. Proc Natl Acad Sci
USA 2007;104:94699474.
51 Harder J, Bartels J, Christophers E, Schroder JM:
Isolation and characterization of human -defensin3, a novel human inducible peptide antibiotic. J Biol
Chem 2001;276:57075713.
52 Kuroda M, Kuwahara-Arai K, Hiramatsu K:
Identification of the up- and down-regulated genes
in vancomycin-resistant Staphylococcus aureus
strains Mu3 and Mu50 by cDNA differential hybridization method. Biochem Biophys Res Commun
2000;269:485490.
53 Neoh HM, Cui L, Yuzawa H, Takeuchi F, Matsuo M,
Hiramatsu K: Mutated response regulator graR is
responsible for phenotypic conversion of Staphylococcus aureus from heterogeneous vancomycinintermediate resistance to vancomycin-intermediate
resistance. Antimicrob Agents Chemother 2008;52:
4553.
54 Meehl M, Herbert S, Gotz F, Cheung A: Interaction
of the GraRS two-component system with the
VraFG ABC transporter to support vancomycinintermediate resistance in Staphylococcus aureus.
Antimicrob Agents Chemother 2007;51:26792689.
55 Li M, Cha DJ, Lai Y, Villaruz AE, Sturdevant DE,
Otto M: The antimicrobial peptide-sensing system
aps of Staphylococcus aureus. Mol Microbiol 2007;
66:11361147.

Otto

56 Baba T, Takeuchi F, Kuroda M, Yuzawa H, Aoki K,


Oguchi A, Nagai Y, Iwama N, Asano K, Naimi T,
Kuroda H, Cui L, Yamamoto K, Hiramatsu K:
Genome and virulence determinants of high virulence community-acquired MRSA. Lancet 2002;
359:18191827.
57 Voyich JM, Braughton KR, Sturdevant DE, Whitney
AR, Said-Salim B, Porcella SF, Long RD, Dorward
DW, Gardner DJ, Kreiswirth BN, Musser JM, DeLeo
FR: Insights into mechanisms used by Staphylococcus
aureus to avoid destruction by human neutrophils. J
Immunol 2005;175:39073919.
58 Herbert S, Bera A, Nerz C, Kraus D, Peschel A,
Goerke C, Meehl M, Cheung A, Gotz F: Molecular
basis of resistance to muramidase and cationic antimicrobial peptide activity of lysozyme in staphylococci. PLoS Pathog 2007;3:e102.
59 Vuong C, Saenz HL, Gotz F, Otto M: Impact of the
agr quorum-sensing system on adherence to polystyrene in Staphylococcus aureus. J Infect Dis 2000;
182:16881693.
60 Geiger T, Goerke C, Mainiero M, Kraus D, Wolz C:
The virulence regulator Sae of Staphylococcus aureus:
promoter activities and response to phagocytosisrelated signals. J Bacteriol 2008;190:34193428.
61 Hancock RE, Sahl HG: Antimicrobial and hostdefense peptides as new anti-infective therapeutic
strategies. Nat Biotechnol 2006;24:15511557.
62 Brodsky IE, Gunn JS: A bacterial sensory system
that activates resistance to innate immune defenses:
potential targets for antimicrobial therapeutics. Mol
Interv 2005;5:335337.
63 Otto M: Staphylococcus aureus and Staphylococcus
epidermidis peptide pheromones produced by the
accessory gene regulator agr system. Peptides 2001;
22:16031608.
64 Vuong C, Kocianova S, Yao Y, Carmody AB, Otto M:
Increased colonization of indwelling medical devices
by quorum-sensing mutants of Staphylococcus epidermidis in vivo. J Infect Dis 2004;190:14981505.
65 Stumpe S, Schmid R, Stephens DL, Georgiou G,
Bakker EP: Identification of OmpT as the protease
that hydrolyzes the antimicrobial peptide protamine
before it enters growing cells of Escherichia coli. J
Bacteriol 1998;180:40024006.

66 Frick IM, Akesson P, Rasmussen M, Schmidtchen


A, Bjorck L: Sic, a secreted protein of Streptococcus
pyogenes that inactivates antibacterial peptides. J
Biol Chem 2003;278:1656116566.
67 Shafer WM, Qu X, Waring AJ, Lehrer RI: Modulation
of Neisseria gonorrhoeae susceptibility to vertebrate
antibacterial peptides due to a member of the resistance/nodulation/division efflux pump family. Proc
Natl Acad Sci USA 1998;95:18291833.
68 Otto M, Peschel A, Gotz F: Producer self-protection
against the lantibiotic epidermin by the ABC transporter EpiFEG of Staphylococcus epidermidis
T3298. FEMS Microbiol Lett 1998;166:203211.
69 Peschel A, Gotz F: Analysis of the Staphylococcus
epidermidis genes epiF, -E, and -G involved in epidermin immunity. J Bacteriol 1996;178:531536.
70 Bengoechea JA, Skurnik M: Temperature-regulated
efflux pump/potassium antiporter system mediates
resistance to cationic antimicrobial peptides in
Yersinia. Mol Microbiol 2000;37:6780.
71 Miller SI, Ernst RK, Bader MW: LPS, TLR4 and
infectious disease diversity. Nat Rev Microbiol 2005;
3:3646.
72 Perego M, Glaser P, Minutello A, Strauch MA,
Leopold K, Fischer W: Incorporation of d-alanine
into lipoteichoic acid and wall teichoic acid in
Bacillus subtilis. Identification of genes and regulation. J Biol Chem 1995;270:1559815606.
73 Gao LY, Laval F, Lawson EH, Groger RK, Woodruff
A, Morisaki JH, Cox JS, Daffe M, Brown EJ:
Requirement for kasB in mycobacterium mycolic
acid biosynthesis, cell wall impermeability and
intracellular survival: Implications for therapy. Mol
Microbiol 2003;49:15471563.
74 Yeaman MR, Bayer AS, Koo SP, Foss W, Sullam PM:
Platelet microbicidal proteins and neutrophil
defensin disrupt the Staphylococcus aureus cytoplasmic membrane by distinct mechanisms of action. J
Clin Invest 1998;101:178187.
75 Campos MA, Vargas MA, Regueiro V, Llompart
CM, Alberti S, Bengoechea JA: Capsule polysaccharide mediates bacterial resistance to antimicrobial
peptides. Infect Immun 2004;72:71077114.

Michael Otto
9000 Rockville Pike
Bethesda, MD 20892 (USA)
Tel. +1 301 443 5209, Fax +1 301 480 3633
E-Mail motto@niaid.nih.gov

Bacterial Sensing of Antimicrobial Peptides

149

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 150160

RNA Thermosensors in Bacterial Pathogens


Jrgen Johansson
Department of Molecular Biology, Ume University, Ume, Sweden

Abstract
During the course of an infection, a pathogenic bacterium has to sense the environment and adjust
its gene expression appropriately. One such environmental cue is the difference in temperature
inside and outside the host. RNA thermosensors are structures that can respond to differences in
temperature by altering their conformation and thereby allowing/preventing binding of the ribosome to the translational start site. This chapter discusses different types of RNA thermosensors in
general and RNA thermosensors known to control virulence gene expression in particular.
Copyright 2009 S. Karger AG, Basel

In order to survive in an often harsh environment, a bacterium has to constantly survey its surroundings and adjust its physiology. Such regulation is essential, especially
for pathogenic bacteria, which employ many different protective measures during an
infection. For many decades, this regulation was believed to be exerted almost exclusively at the transcriptional level, but new data have highlighted the importance of
post-transcriptional regulation. The mRNA molecule is not passively awaiting decoding, instead it can form specific structures or bind certain regulatory elements and
thereby control its stability or ability to initiate translation. The last few years have
revealed the importance of RNA as a mediator of gene expression. RNA-mediated
regulation can roughly be divided into two classes: trans-acting RNAs are generally small non-coding RNAs (ncRNAs) that by sequence complementarity can bind
target mRNA encoded elsewhere on the chromosome, and thereby control its fate.
In contrast, if the transcript harbors a regulatory region that affects the expression
of the downstream coding sequence, this regulatory region acts in cis. During the
last few years, small ncRNAs have attracted much attention, especially in eukaryotes
where very short (2127 nucleotides) ncRNAs bind target mRNAs and either block
translation or induce their degradation. In bacteria, the ncRNAs discovered thus far
are larger, with sizes generally being around 100 nucleotides, although ncRNAs over
500 nucleotides are known [1]. Through different regions of complementarity, one

bacterial ncRNA can often control several mRNA targets in a synchronized manner.
This is the case for iron-storage genes in Escherichia coli, whose expression is repressed
in the absence of iron. Such regulation is exerted by an ncRNA, RyhB, which by binding many of the iron-storage encoding mRNAs initiates their decay. The absence of
RyhB at high iron concentrations allows expression of iron-storage proteins [2].
New forms of cis-acting RNA elements, different from classic attenuators [3], were
discovered a few years ago. These elements, riboswitches, are able to directly bind
certain metabolites with high specificity. Riboswitches generally lie upstream of genes
involved in various metabolic pathways [3]. By binding a metabolite that is connected
to such a pathway, the riboswitch can monitor the concentration of the metabolite and either turn synthesis of the downstream genes on or off . Riboswitches are
almost exclusively built up by an aptamer domain that binds the metabolite and an
expression platform that controls expression of the downstream genes in response to
metabolite binding. Binding of the metabolite to the aptamer domain induces a conformational change at the expression platform that results in either of three outcomes:
an altered ability of the RNA polymerase to proceed through an intrinsic transcriptional terminator [4, 5], an altered translation initiation [6] or transcriptional decay
[7]. The transcriptional termination mechanism is most common in Gram-positive
bacteria, where it is believed to control expression of more than 2% of the genes [3].
The recent finding of riboswitches in eukaryotes and archaeas indicates extensive distribution of such regulatory entities in all kingdoms [3].

RNA Thermosensors

During the last 20 years, it has become evident that both pathogenic and non-pathogenic bacteria are able to control gene expression by RNA thermosensors. The underlying idea is that the RNA structure is altered with an increased temperature, such
that certain occluded regions, i.e. Shine-Dalgarno (SD) sites, are liberated and translation can initiate.
In 1989, Altuvia et al. [8] beautifully showed that an RNA segment in the bacteriophage was present in two different RNA secondary structures, A and B, at equilibrium (fig. 1a). At low temperatures (<37C), structure B was more dominant, therefore
the SD site was accessible due to an interaction between the anti-SD site and the antianti-SD site. The free SD site allowed translation of the cIII mRNA which in turn
positively regulated the lysogenic pathway. If the temperature was increased (45C),
the structure was rearranged and structure A became more dominant. The SD site
in structure A was masked by an interaction with the anti-SD site, which prevented
translation of cIII and thereby drove the bacteriophage into the lytic cycle. In contrast
to other thermosensors, translation of cIII is reduced with increasing temperature.
The concerted expression of many genes essential for growth at higher temperatures requires the expression of a heat-shock sigma factor H, encoded by rpoH.

Temperature Sensing by RNA

151

Translation of rpoH in E. coli is controlled by a thermosensor mechanism. Through


various experiments, it was shown that two different regions in the rpoH structural
mRNA could interact and form a secondary structure that prevented translation initiation at low temperatures (fig. 1b) [9, 10]. When the temperature was increased, the
structure was weakened and a sequence downstream of the AUG start codon that
showed complementarity to a part of the 16S rRNA was opened up and translation
could initiate. A short form of the rpoH thermosensor was constructed where two
additional hairpins were deleted. For the minimal thermosensor to respond to an
increase in temperature, some base-pairing mismatches had to be introduced, indicating a role for the additional hairpins whereby they could somewhat destabilize the
main hairpin. Both the cIII and the rpoH thermosensors are unique in the sense that
most of their functional domains lie downstream of the AUG start codon.
The most widespread class of thermosensors by far is the ROSE elements (repression of heat-shock gene expression). Originally, it was described in the plant symbiont
Bradyrhizobium japonicum [11], but has now been identified in many Gram-negative
bacteria including E. coli and Salmonella [12]. The ROSE thermosensors are comprised of 34 different RNA hairpins where the SD and the AUG start codon are
sequestered in the last hairpin (fig. 1c). The roles of the other hairpins are probably
to ensure proper conformation during transcription or alternatively, to form a tertiary structure supporting translation initiation at higher temperatures [13, 14]. A
mini-ROSE element only retaining the hairpin with the SD site and the AUG start
codon showed reduced heat inducibility [14]. Destabilization of the top of the hairpin
upstream of the hairpin containing the SD increased expression of the downstream
gene at low temperatures. Common for all ROSE thermosensors is the presence of a
G residue that first was believed to be unable to base-pair and therefore cause a bulge
in the anti-SD side of the hairpin [14]. However, recent NMR studies have shown that
the G residues form an energetically unstable G-G base-pairing with a residue of the
SD region [15]. The NMR study suggests that by increasing the temperature to 37C,
an internal loop harboring a U-U base-pair is disrupted. Increasing the temperature
further to 42C dislocates the G-G base-pair and an A-U base-pair, resulting in an SD
site accessible for the ribosome. Deletion of the G-residue stabilized the hairpin and
completely abolished translation induction at higher temperatures.
A fundamentally different type of RNA thermosensor is represented by the ncRNA,
DsrA. By directly binding to the SD region of hns, DsrA could repress translation initiation of hns. In contrast, DsrA was able to directly bind an anti-SD site upstream of
rpoS, and thereby liberate its SD site, leading to translation initiation [16]. When synthesized at low temperatures (25C), DsrA was 85 nucleotides long and could activate
rpoS translation (fig. 1d). With increasing temperature, more DsrA was processed by
RNaseE to a 60- to 61-nucleotide-long ncRNA that lacked the bases that could bind
and sequester the anti-SD of rpoS. At lower temperatures, full-length DsrA was 6-fold
more stable than at higher temperatures. Both temperature-mediated mechanisms
(processing and stability) resulted in a much higher level of full-length (active) DsrA

152

Johansson

Low T
High Mg2+
5

High T
Low Mg2+

HighT

5
3

5
3
LowT

HighT
G

3
5

3
5

LowT

c
+DsrA
63nt

5
63nt

3
5

?DsrA
3

d
Fig. 1. Schematic models of RNA thermosensors in different organisms. See text for further details. a Structure A: Binding of
the ribosome (upward diagonal spheres) to the SD site (light grey box) of the bacteriophage cIII mRNA is impaired by a
secondary structure at high (45C) temperature and low Mg2+ concentration. Structure B: At lower temperatures (<37C) and
higher Mg2+ concentrations, an alteration of the RNA structure enables the binding of the ribosome to the SD site and translation initiation at the AUG-start codon (black box). b Translation initiation of the E. coli rpoH mRNA requires an interaction
between 16S rRNA and a downstream box (dark grey box) within the rpoH structural gene. At low temperatures (<30C) the
downstream box is occluded, preventing binding of the ribosome. At higher temperatures (42C) the mRNA structure is
altered, allowing binding of the ribosome to the downstream box and interaction of the ribosome to the SD site. Translation
can initiate. c Translation of heat-shock protein encoding mRNAs in Rhizobium species is prohibited at low temperatures
(30C) due to an interaction between the SD site and an anti-SD site. This interaction prevents binding of the ribosome. The
SD site is released at higher temperatures, allowing binding of the ribosome and translation initiation. d Translation of the E.
coli rpoS mRNA is inhibited by an upstream sequence that by masking the rpoS SD site prevents binding of the ribosome. The
small regulatory RNA, DsrA (long dashed line), can interact with the occluding upstream sequence, thereby liberating the SD
site and allowing translation initiation. DsrA is expressed at a higher level and is more active at low (25C) temperatures.

Temperature Sensing by RNA

153

at low temperatures and therefore also an increased level of S in the cell [17, 18].
It is however not known if the signal for DsrA processing at varying temperatures
is mediated by a structural alteration in DsrA or if it is due to increased activity of
RNaseE or another yet unknown factor. We can therefore not conclude that DsrA is a
bona fide RNA thermosensor.
In the soil bacterium Streptomyces albus, the expression of a heat-shock protein,
Hsp18, was shown to be both transcriptionally and post-transcriptionally regulated
[19]. It was shown that translation was initiated only at high (41C) temperatures and
not at low (30C). After transition from high to low temperatures, translation was
rapidly inhibited although the transcript was still present. Although proteinacious
components cannot be excluded, the results suggest a mechanism whereby an antiSD site sequesters the SD site and prevents translation at low temperatures.

Cold Sensors

Can an RNA structure prevent translation at higher temperatures and only allow translation at lower temperatures (<25C)? Many of the genes encoding cold-shock proteins (Csps) in E. coli lie downstream of long 5-untranslated RNA regions. Deletion
analysis in the 5-UTR preceding cspA identified a region that repressed translation
at higher temperatures. Since the stability of the different deletion mutant transcripts
did not vary, the authors speculate that a conformational change in the structure of
the 5-UTR between different temperatures could account for the thermoregulated
expression of cspA [20]. This is an attractive idea, although it would require a larger
structural rearrangement (as for the cIII thermosensor) than for RNA thermosensors
activated by an increased temperature. Further experiments will hopefully shed light
on this question.

Eukaryotic RNA Thermosensors

RNA thermosensors have been found in many different bacteria and in bacteriophages. Are they also present in eukaryotes? Some circumstances suggest that the
induced translation of Drosophila melanogaster Hsp90 mRNA and human Hsp70
mRNA during heat-shock is through RNA thermosensor mechanisms [21, 22]. In
a suggested model, translation at low temperatures is believed to occur through a
normal, cap-dependent scanning mechanism. In contrast, translation of the Hsp90
mRNA during heat-shock is suggested to occur through a cap-independent bacterialike mechanism where the 18S rRNA is recruited by base-pairing to a region close to
the AUG start codon [21]. This putative region is thought to respond to differences
in temperature in a manner similar to the bacterial RNA thermosensors. However,
although attractive, no evidence has yet been shown ruling out other mechanisms.

154

Johansson

RNA Thermosensors in Bacterial Pathogens

To save energy, pathogenic bacteria often express virulence genes only when entering the host. Different pathogens can use various methods of detecting the temperature shift between outside and inside the host. For example, when Salmonella enterica
serovar Typhimurium is transferred from lower temperatures to 37C, a transcriptional repressor (TlpA) is not able to form multimers and can therefore only bind
DNA poorly, leading to increased virulence gene expression [23]. Also, a similar shift
in temperature causes a conformational alteration of the promoter region of the virulence regulator virF in Shigella flexneri. This DNA alteration prevents binding of the
nucleoid-associated protein H-NS and virulence gene expression can initiate [24].
However, most examples of direct responses to alterations in temperature come from
RNA thermosensors.
Yersinia pestis is one of the most efficient human pathogens and was the causative
agent of the plague during the 14th century in Europe. Expression of the Y. pestis virulence factors requires a transcriptional activator, LcrF, which is present only at 37C
and not at 26C. In 1993, Hoe and Goguen [25] showed that the absence of LcrF at
lower temperatures was not due to proteolytic degradation. They also showed that
thermoregulated expression of LcrF expression was mediated both in Y. pestis and in
E. coli. Measurement of the protein/mRNA ratio suggested that although the lcrF messenger was present at low temperatures, its translation was approximately 5-fold lower.
Conclusively, the authors suggest that an interaction between an anti-SD site and the
SD site could prevent translation initiation at low temperatures. A structural alteration
of this hairpin at higher temperatures would free the SD site and translation would
initiate. Unfortunately, no further experiments were performed to verify this model.
S. enterica serovar Typhimurium causes salmonellosis, an infection where the
patient normally develops diarrhea, fever and abdominal cramps. Most virulence factors in Salmonella are encoded on various pathogenicity islands, which carry genes
required for different steps of the infection process. The expression, proper folding
and secretion of virulence factors require the presence of certain chaperones, with
many of them being upregulated during different stress conditions. Recently, the
Narberhaus laboratory presented evidence that one such chaperone, AgsA, was controlled both transcriptionally and post-transcriptionally [26] (fig. 2a). Using translational fusions, it was observed that agsA expression was induced approximately
3-fold when transcription was directed from a pBAD promoter. Using RNA prediction programs and RNA-probing experiments, it was found that the 5-untranslated
RNA upstream of the agsA structural mRNA formed two distinct hairpins, with the
second hairpin having the SD site trapped by binding to an anti-SD site. Certain hairpin-stabilizing base substitutions located at or very close to the anti-SD site that by
RNA-probing experiments were shown to close the hairpin, completely disrupted
the function of the thermosensor. This again highlights the importance of a nonperfect interaction between the SD site and the anti-SD site as was observed for the

Temperature Sensing by RNA

155

HighT

5
3

LowT

5
3

HighT
LowT
5

3
NoPrfA

3
PrfA

Virulencegene
expression

Fig. 2. Schematic functional models of 5-UTR located RNA thermosensors in pathogenic bacteria.
See text for further details. a Translation of the S. typhimurium agsA mRNA at low temperatures (30C)
is prevented by a SD site and anti-SD site interaction, that hampers the binding of the ribosome. At
higher temperatures (42C), the SD site is unchained, allowing binding of the ribosome and translation initiation. b At low temperatures (<30C), the 5-UTR of the L. monocytogenes transcriptional
activator prfA forms a secondary structure that masks the SD site and prevents translation. At higher
temperatures (37C), the secondary hairpin structure is partially disrupted, enabling binding of the
ribosome and translation initiation. Translation of prfA activates virulence genes.

ROSE thermosensors. Base substitutions that further destabilized the region around
the anti-SD site increased expression both at low and high temperatures. A mini-agsA
construct, harboring only the second hairpin, showed lower expression compared to
the wild-type, but with an increased induction level, thus indicating that the role of
the first hairpin could be to stabilize the structures. Toe-printing experiments showed
that the ribosome was unable to bind at low temperatures, but could form an initiation complex at higher temperatures. The authors claim that the agsA and the lcrF
thermosensors are part of a four U class of thermosensors, referring to the four or
more U residues that comprise the anti-SD site. Other members of this class would be
5-UTRs preceding Staphylococus aureus groES and Brucella melitensis dnaJ, although
these putative thermosensors have yet to be verified experimentally.
Listeria monocytogenes is a food-borne bacterial pathogen that causes severe infections, primarily in immunocompromised individuals and pregnant women, due to
its ability to pass the intestinal, placental and blood-brain barrier. The bacterium can
induce its own phagocytosis, escape the vacuole that is formed, proliferate within the
host cell and use actin polymerization to disseminate, by a wide range of virulence

156

Johansson

genes positively regulated by the transcriptional regulator PrfA. Early, it was shown that
the PrfA protein was only expressed at body temperature, 37C, although the mRNA
encoding the protein was present in high amounts at temperatures <30C [27, 28].
Upstream of the prfA encoding mRNA lies a 115-nucleotide-long 5-UTR. Computer
prediction and chemical probing experiments showed that the prfA-UTR formed one
long hairpin where the SD site was partially masked by an anti-SD site [28] (fig. 2b).
When the prfA-UTR was removed and the promoter replaced by a T7 promoter, the
PrfA protein was expressed at both low and high temperatures, indicating a role for
the prfA-UTR in temperature regulation. Base substitution mutations, that by computer prediction analysis and RNA-probing experiments were shown to destabilize the
prfA-UTR structure, displayed a derepressed expression at low temperatures, in both
E. coli and L. monocytogenes. When such derepressed mutants were expressed at low
temperatures in L. monocytogenes, the expression of PrfA led to an increased expression of various virulence genes, and also enabled the bacterium to invade and proliferate in cell culture [28]. These results indicate that expression of PrfA is sufficient for
activation of the virulence machinery at low temperatures. A mutation that restored
disrupted base-pairing in the hairpin partially restored thermoregulation and base substitutions in the anti-SD, predicted to close the SD, resulted in reduced expression of
PrfA at 37C. In vitro translation of the prfA mRNA carrying the wild-type prfA-UTR
showed a clear thermoregulated expression, with PrfA mostly expressed at higher temperatures. In contrast, when in vitro translation of prfA mRNAs carrying mutations
that destabilized the prfA-UTR structure was performed, much less thermoregulation
could be detected [28]. The above experiments show that thermoregulated expression
of PrfA is due to structural alterations in the prfA-UTR and not to any other, unknown
factor. At low temperatures, the SD site is closed due to an interaction with an antiSD site and binding of the ribosome is prevented. With increasing temperature, the
region surrounding the prfA SD site is destabilized, allowing binding of the ribosome
so translation can initiate (fig. 2b). Basal expression of PrfA at low temperatures is
governed by a promoter located in the middle of the prfA thermosensor, resulting in a
32-nucleotide-long UTR not being thermoregulated. By placing the long prfA-UTR in
front of gfp, GFP expression became thermoregulated, with non-fluorescent bacteria
at low temperatures and green bacteria at 37C [28]. This raises the possibility that any
gene of interest can be thermoregulated, which could be of interest for biotechnological applications.

Structural Requirements

Although the basic idea is similar, it appears there is a very large structural difference between riboswitches acting at the translational level and RNA thermosensors.
Except for the thermoswitch [8], the structural modification of a thermosensor
upon an increase in temperature is modest compared to the structural alterations

Temperature Sensing by RNA

157

occurring in a riboswitch when binding a metabolite [6, 15]. This is exemplified by


the anti-SD site in the thiamine pyrophosphate riboswitch that sequesters the SD site
in the presence of the metabolite TPP. In the absence of TPP, a structural rearrangement of the riboswitch forces the anti-SD site to interact with an anti-anti-SD instead
of the SD site, thus allowing translation initiation [6]. Except for the thermosensor,
no thermosensor has been described having such an anti-anti-SD site. Instead, the
NMR data from the ROSE thermosensor [15] suggest the structural rearrangements
are very modest when the temperature increases.
Although a thermosensor can consist of several different hairpins that contribute to
the thermosensing capacity, the thermosensor activity per se resides in the hairpin with
the SD and anti-SD regions. In contrast, riboswitches have an aptamer (metabolite binding) domain and a regulatory domain that almost exclusively lies in disparate hairpins.
It is very striking that most, if not all, suggested thermosensors harbor non-perfect
base-pairing between the SD and the anti-SD. The weak interaction is most probably very important for a proper temperature-mediated opening of the structure.
Mutations stabilizing an SD and anti-SD interaction have impaired the thermoregulation with no increased expression at the higher temperature [13, 26, 28]. Moreover,
many SD hairpins contain a bulge region close to the SD region, which appears to
function as the primary sensing site. It opens even at lower temperatures, with the
surrounding SD site being released at higher temperatures [15].

Questions and Future Perspectives

How do we find new thermosensors? Recently, the Narberhaus laboratory used a


bioinformatics approach to identify new thermosensors. This searchable database,
RNA-SURIBA (Structures of Untranslated Regions in Bacteria), predicts putative
thermosensor structures in 5-UTRs preceding annotated genes [29]. Such programs
can prove especially useful for finding orphan thermosensors controlling virulence
genes in pathogenic bacteria.
Can thermosensors respond to stimuli other than temperature? Interestingly, the
thermosensor controlling the cIII translation in bacteriophage also responds to
variations in the level of Mg2+. Low concentrations of Mg2+ (0.5 mm Mg2+) favored
the formation of structure A, in a manner similar to an increase in temperature,
thus resulting in inhibited translation of cIII [8]. It could therefore be hypothesized
that some thermosensors could respond to stimuli other than temperature (mono- or
divalent cations, pH, etc). Such double regulation would be an intelligent way for a
pathogen to ensure translation of an important protein under two different conditions. Analogously, a tandem riboswitch responding to both S-adenosylmethionine
and coenzyme B12 was recently identified upstream of metE in Bacillus clausii [30].
Although the thermosensor itself resides within the RNA structure, it cannot be
completely ruled out that the thermosensor would require certain protein co-factors

158

Johansson

in vivo to enhance their function or stability. Such protein co-factors would probably
be RNA-binding proteins like Hfq, RNA helicases and/or ribosomal proteins (i.e. S1).

Conclusions

This chapter clearly shows that RNA thermosensors can be important regulatory
devices for the control of gene expression. In a domino-effect manner, a regulatory
protein that is responding to a signal and controls certain genes also has to be regulated by something. This is avoided by the use of RNA thermosensors that instead
can directly be the sensors of a temperature difference. When more knowledge is
gathered about RNA thermosensors and their function, this will hopefully give us
a larger probability of identifying new thermosensors from the folding of an RNA
structure. This could especially be of interest in pathogenic bacteria which could use
an increase in temperature as an indicator of their spatial orientation (inside or outside the host).

Acknowledgements
The author thanks Dr. Guo for critically reading the manuscript. I apologize to those whose work
was not cited due to space constraints. Supported by the Wenner-Gren Foundations, Ume
University, the Swedish Research Council Grant 621-2006-4450 and EU (BacRNA 2005 Contract
No. 018618).

References
1 Toledo-Arana A, Repoila F, Cossart P: Small noncoding RNAs controlling pathogenesis. Curr Opin
Microbiol 2007;10:182188.
2 Mass E, Gottesman S: A small RNA regulates the
expression of genes involved in iron metabolism in
Escherichia coli. Proc Natl Acad Sci USA 2002;99:
46204625
3 Winkler WC, Breaker RR: Regulation of bacterial
gene expression by riboswitches. Annu Rev Microbiol
2005;59:487517.
4 Mironov AS, Gusarov I, Rafikov R, Lopez LE,
Shatalin K, Kreneva RA, Perumov DA, Nudler E:
Sensing small molecules by nascent RNA: a mechanism to control transcription in bacteria. Cell 2002;
111:747756.
5 Winkler WC, Nahvi A, Sudarsan N, Barrick JE,
Breaker RR: An mRNA structure that controls gene
expression by binding S-adenosylmethionine. Nat
Struct Biol 2003;10:710707.

Temperature Sensing by RNA

6 Winkler WC, Nahvi A, Breaker RR: Thiamine


derivatives bind messenger RNAs directly to regulate bacterial gene expression. Nature 2002;419:952
956.
7 Winkler WC, Nahvi A, Sudarsan N, Barrick JE,
Breaker RR: Control of gene expression by a natural
metabolite-responsive ribozyme. 2004;428:281286.
8 Altuvia S, Kornitzer D, Teff D, Oppenheim AB:
Alternative mRNA structures of the cIII gene of
bacteriophage determine the rate of its translation
initiation. J Mol Biol 1989;210:265280.
9 Morita M, Kanemori M, Yanagi H, Yura T: Heatinduced synthesis of 32 in Escherichia coli: structural and functional dissection of rpoH mRNA
secondary structure. J Bacteriol 1999;181:401410.
10 Morita MT, Tanaka Y, Kodama TS, Kyogoku Y,
Yanagi H, Yura T: Translation induction of heatshock transcription factor 32: evidence for a built-in
RNA thermosensor. Genes Dev 1999;13:655665.

159

11 Narberhaus F, Weiglhofer W, Fischer HM, Hennecke


H: The Bradyrhizobium japonicum rpoH1 gene
encoding a 32-like protein is part of a unique heatshock gene cluster together with GroESL1 and three
small heat-shock genes. J Bacteriol 1996;178:5337
5346.
12 Waldminghaus T, Fippinger A, Alfsmann J,
Narberhaus F: RNA thermometers are common in
- and -proteobacteria. Biol Chem 2005;386:1279
1286.
13 Nocker A, Hausherr T, Balsiger S, Krstulovic NP,
Hennecke H, Narberhaus F: mRNA-based thermosensor controls expression of rhizobial heatshock genes. Nucleic Acids Res 2001;29:48004807.
14 Chowdhury S, Ragaz C, Kreuger E, Narberhaus F:
Temperature-controlled structural alterations of an
RNA thermometer. J Biol Chem 2003;4791547921
15 Chowdhury S, Maris C, Allain FH-T, Narberhaus F:
Molecular basis for temperature sensing by an RNA
thermometer EMBO J 2006;25:24872497.
16 Sledjeski DD, Gupta A, Gottesman S: The small
RNA, DsrA, is essential for the low temperature
expression of RpoS during exponential growth in
Escherichia coli. EMBO J 1996;15:39934000.
17 Repoila F, Gottesman S: Signal transduction cascade
for regulation of RpoS: temperature regulation of
DsrA. J Bacteriol 2001;183:40124023.
18 Repoila F, Majdalani N, Gottesman S: Small noncoding RNAs, coordinators of adaptation processes
in Escherichia coli: the RpoS paradigm. Mol Microbiol
2003;48:855861.
19 Servant P, Mazodier P: Heat Induction of hsp18
gene expression in Streptomyces albus G: transcriptional and posttranscriptional regulation. J Bacteriol
1996;178:70317036.
20 Yamanaka K, Mitta M, Inouye M: Mutation analysis
of the 5-untranslated region of the cold-shock cspA
mRNA of Escherichia coli. J Bacteriol 1999;181:6284
6291.
21 Ahmed R, Duncan RF: Translational regulation of
Hsp90 mRNA. AUG-proximal 5-untranslated
region elements essential for preferential heat-shock
translation. J Biol Chem 2004;279:4991949930.

22 Yueh A, Schneider RJ: Translation by ribosome


shunting on adenovirus and hsp70 mRNAs facilitated by complementarity to 18S rRNA. Genes Dev
2000;14:414421.
23 Hurme R, Berndt KD, Normark SJ, Rhen M: A proteinaceous gene regulatory thermometer in
Salmonella. Cell 1997;90:5564.
24 Falconi M, Colonna B, Prosseda G, Micheli G,
Gualerzi CO: Thermoregulation of Shigella and
Escherichia coli EIEC pathogenicity. A temperaturedependent structural transition of DNA modulates
accessibility of virF promoter to transcriptional
repressor H-NS. EMBO J 1998;17:70337043.
25 Hoe NP, Goguen JD: Temperature sensing in Yersinia
pestis: translation of the LcrF activator protein is
thermally regulated. J Bacteriol 1993;175:79017909.
26 Waldminghaus T, Heidrich N, Brantl S, Narberhaus
F: FourU: a novel type of RNA thermometer in
Salmonella. Mol Microbiol 2007;65:413424.
27 Renzoni A, Klarsfeld A, Dramsi S, Cossart P:
Evidence that PrfA, the pleiotropic activator of virulence genes in Listeria monocytogenes, can be present but inactive. Infect Immun 1997;65:15151518.
28 Johansson J, Mandin P, Renzoni A, Chiaruttini C,
Springer M, Cossart P: An RNA thermosensor control expression of virulence genes in Listeria monocytogenes. Cell 2002;110:551561.
29 Waldminghaus T, Gaubig LC, Narberhaus F:
Genome-wide bioinformatic prediction and experimental evaluation of potential RNA thermometers.
Mol Gen Genom 2007;278:555564.
30 Sudarsan N, Hammond MC, Block KF, Welz R,
Barrick JE, Roth A, Breaker RR: Tandem riboswitch
architectures exhibit complex gene control functions. Science 2006;314:300304.

Jrgen Johansson
Department of Molecular Biology, Ume University
SE90187 Ume (Sweden)
Tel. +46 90 785 6720, Fax +46 90 772 630
E-Mail Jorgen.johansson@molbiol.umu.se

160

Johansson

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 161181

Prevailing Concepts of c-di-GMP Signaling


Ute Rmling Roger Simm
Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden

Abstract
Recently, the list of ubiquitous bacterial secondary messengers which include cAMP and ppGpp has
been extended by 3,5-cyclic diguanylic acid (c-di-GMP). C-di-GMP metabolism is tuned by the
tightly controlled activity of diguanylate cyclases and c-di-GMP-specific phosphodiesterases. As
c-di-GMP-metabolizing enzymes are not only found frequently in bacterial genomes, but also are
often numerous in individual genomes, the c-di-GMP metabolic network is highly complex whereby
signaling specificity is adjusted on the level of expression, enzymatic activity, protein localization
and, most likely, receptor affinity. The targets of c-di-GMP, which include protein and RNA receptors,
are subsequently being unraveled. Besides the transition between sessility and motility, probably
the most ancient regulatory control of bacterial behavior by c-di-GMP, many more phenotypes such
as virulence are affected by c-di-GMP. However, the exact molecular mechanisms of c-di-GMP action
Copyright 2009 S. Karger AG, Basel
remain to be discovered.

Signal transduction via second messengers is a common mechanism not only to regulate basic cellular functions, but also complex behavior in eukaryotes [1]. Thereby,
purine nucleotides are frequently used. Compared to eukaryotes, an apparent reduction of the signaling pathways seemed to exist in bacteria with only the use of the
purine nucleotides cAMP and ppGpp, also called an alarmone. The paradigm of second messenger signaling was for a long time cAMP signaling studied in the laboratory
work horse Escherichia coli [2]. Although synthesis of cAMP is subject to a reasonably complex regulation, the overall signaling pathway is linear, with one adenylate
cyclase, one cAMP-specific phosphodiesterase and one cAMP receptor protein (fig.
1a), whereby the cAMP degradation has mostly been neglected. Similarly, the ppGpp
metabolism is not much more complex (fig. 1b). Nevertheless, numerous genes and
phenotypes are regulated by cAMP as well as ppGpp signaling.
In recent years, novel secondary signaling molecules have been discovered in bacteria [36]. In particular, the c-di-GMP signaling system is not only highly abundant
in many bacterial species, but already the metabolism of this secondary messenger in
an individual species is more complex than observed before for a secondary signaling

molecule in bacteria [7]. As a comparison to the metabolism of cAMP and ppGpp,


E. coli has 31 proteins presumably involved in c-di-GMP metabolism [8]. Although
the diversity of genes and phenotypes regulated by c-di-GMP signaling is just beginning to be unraveled, one principal role of c-di-GMP signaling is the regulation of the
transition between motility and sessility in bacteria. This review describes the current
view on the basic principles of c-di-GMP signaling.

Biosynthesis of C-di-GMP by GGDEF Domain Proteins

In a condensation reaction, two molecules of GTP are ligated to c-di-GMP by the


activity of the GGDEF domain which therefore is a diguanylate cyclase. The GG(D/E)
EF domain has received its name from the characteristic highly conserved motif
within the domain whereby all amino acids besides the tryptophan contribute to substrate binding and catalysis [9]. Consequently, any amino acid exchanges in this motif
highly reduced or abolished the catalytic activity of the respective GGDEF domain [5,
6, 1012]. In homology to the well-studied adenylate cyclases [13], diguanylate cyclases exert their function as dimers [9] with an analogous two-metal catalytic mechanism using Mg2+ or Mn2+ as metal ions [14]. However, other oligomeric stages such
as trimers and tetramers exist. Dimerization, an intrinsic property of the GGDEF
domain [15] or induced upon activation of the protein [6], is required, yet not sufficient for the enzymatic activity. Consequently, GGDEF domains alone have low
enzymatic activity and require activation provided by N-terminal signaling or sensor
domains [16].

Degradation of C-di-GMP by EAL and HD-GYP Domain Proteins

The EAL domain hydrolyzes c-di-GMP to the linear dinucleotide 5-pGpG with
high specificity, therefore it is a c-di-GMP-specific phosphodiesterase [1618]. PDE
activity requires Mg2+ or Mn2+ and is strongly inhibited by Ca2+ and Zn2+. Although
name-giving, the EAL motif is not the most conserved motif in the domain [16],
even if the highly conserved glutamate is required for enzymatic activity [16, 19].
Fig. 1. Metabolism of secondary messenger molecules in E. coli and S. typhimurium. a The second
messenger cAMP is synthesized by the adenylate cyclase (AC) encoded by the cya gene [2] after
activation by phosphorylation [114]. Degradation of cAMP is performed by the cpdA gene product
[115]. Through binding to the cAMP receptor protein CRP, an active cAMP-CRP complex is formed,
which positively or negatively influences the expression of numerous genes. b Metabolism of the
alarmone ppGpp in E. coli and S. typhimurium. Synthesis of ppGpp is performed by RelA and SpoT,
while SpoT has also ppGpp hydrolyzing activity. In response to nutrient stress, ppGpp influences
various physiological processes in the cell such as translation, replication and transcription. c
Metabolism of c-di-GMP in E. coli. There exist 19 potential diguanylate cyclases (GGDEF domain proteins) and 17 potential phosphodiesterases (EAL domain proteins) in E. coli.

162

Rmling Simm

cpdA

AC

ATP

cAMP

AMP

CRP

GTP + ATP

SpoT

RelA

50S

GDP

30S

ppGpp
transcription
translation

replication

MASE1

MASE1
PAS

HRSDF

PAS

PGSEL

PAS

ELM

GGDEF

EAL

pGpG cdiGMP

ELI

PAS

GTP cdiGMP

EAL
cdiGMP pGpG

GAF

EAL

GTP cdiGMP pGpG

GGDEF

cdiGMP pGpG

PAS

PAS

pGpG cdiGMP

GGDEF

PAS

EAL

EAL

SGYDF

GTP cdiGMP

GGDEF

EAL

cdiGMP

cdiGMP pGpG
BLUF

GGDEF

pGpG cdiGMP

pGpG

cdiGMP

GGDEF
HTH
DPL

EAL

pGpG cdiGMP

pGpG cdiGMP

GTP
GGDEF

cdiGMP GTP
GGDEF

HA
MP

cdiGMP GTP
GGDEF

cdiGMP GTP

cdiGMP GTP

GGDEF

GTP
GGDEF

EAL

HTH

GGDEF
cdiGMP GTP cdiGMP

SGYDF

HA
MP

GGDEF

GTP cdiGMP

GTP cdiGMP

cdiGMP pGpG
EAL

ELL
cdiGMP pGpG
EAL

EAL

cdiGMP

GTP
PAS

FGNDL

cdiGMP pGpG

LIS

pGpG cdiGMP GTP


EAL

EAL

GGDEF

EAL

pGpG
cdiGMP
GTP
GGDEF
MASE2

Prevailing Concepts of c-di-GMP Signaling

163

Alignment of active EAL domain proteins identified other more conserved extended
motifs such as ELTE, (L/I)DDFGTG and EGVE, apparently required for enzymatic
activity. Recently, the impact of highly conserved individual residues in these motifs
and at other sites in the domain on the catalytic activity has been investigated [19].
Although EAL domains are enzymatically active as monomers [16], the full-length
proteins can have different oligomeric stages [16, 19].
As preticted by bioinformatic analysis [20, 21], a subgroup of the HD superfamily of metal-dependent phosphohydrolases, the HD-GYP domain, was experimentally confirmed to possess c-di-GMP-specific phosphodiesterase activity [22]. The
HD-GYP domain has no sequence similarity to the EAL domain. In contrast to EAL
domains, HD-GYP domains hydrolyze c-di-GMP readily to GMP. At present, the
impact of the different end products (pGpG vs. GMP) derived through the enzymatic
activity of EAL domain and HD-GYP phosphodiesterases is not clear.

Abundance of GGDEF, EAL and HD-GYP Domain Proteins in Bacterial Genomes

Proteins involved in c-di-GMP metabolism are highly abundant. Many species


throughout the bacterial kingdom harbor the c-di-GMP signaling pathway [7]. When
present in one species, usually more than one gene encoding a GGDEF, EAL and/
or HD-GYP domain protein is found. -Proteobacteria living in aquatic environments such as Vibrio and Shewanella spp. have top scores with respect to abundance
of these proteins, while the c-di-GMP-metabolizing enzymes are usually more scarce
in individual genomes of Gram-positive bacteria. However, there is substantial variation even within bacterial genera. In general, c-di-GMP signaling seems to be more
variable than e.g. signaling by phosphor transfer. While commensal E. coli as represented by the laboratory strain E. coli K-12 have 12 GGDEF, 12 EAL and 7 GGDEFEAL domain proteins summing up to 31 proteins involved in c-di-GMP metabolism,
Shigella flexneri, an E. coli pathovar, has 3 GGDEF, 6 EAL and 4 GGDEF-EAL domain
proteins (in total 13 proteins). On the other hand, the number of sensor kinases
diminishes only from 30 in E. coli to 25 in S. flexneri. Salmonella and E. coli genes
encoding proteins involved in c-di-GMP metabolism are usually stand-alone genes
not located in operons or clustered together with other GGDEF/EAL domain proteins which might contribute to the high variability (and flexibility) observed in c-diGMP signaling [23, 24].
Although the genomes of Salmonella and E. coli do not encode HD-GYP domain
proteins, the HD-GYP domain is not significantly less abundant than the EAL domain
(not counting the EAL domains coupled to a GGDEF domain). Usually these two
phosphodiesterase types co-exist, however there is a significant proportion of bacterial genomes not encoding HD-GYP domain proteins. The presence of HD-GYP
domain proteins as sole c-di-GMP-dependent phosphodiesterases is infrequent,
observed in species isolated from extreme environments, in Treponema pallidum, the

164

Rmling Simm

causative agent of syphilis and the obligate intracellular pathogen Lawsonia intracellularis, which causes disease in domestic and laboratory animals. The reason(s) for the
preference of one type of phosphodiesterase, if any, are not known.

Modulation of the Enzymatic Activity of C-di-GMP-Metabolizing Enzymes

As required for a second messenger signal, the c-di-GMP concentration is considered


to dynamically respond to changing internal and external conditions. Consequently
c-di-GMP-metabolizing domains are rarely found as stand-alone proteins [25, 26].
Usually the enzymatic domains are part of multi-domain proteins with one or more
N-terminal signaling domains (fig. 2). There exist membrane-bound and cytoplasmic
c-di-GMP-metabolizing proteins indicating that c-di-GMP metabolism is regulated
by external and internal signals.
A significant fraction of GGDEF, EAL and HD-GYP proteins with predicted cytoplasmic or membrane location contain N-terminal sequences of unknown function
indicating that uncharacterized signaling and sensing mechanisms are involved in the
control of c-di-GMP metabolism.
Although some of the membrane domains adjacent to GGDEF, EAL and HD-GYP
domains can be classified, e.g. MASE1, MASE2 or MHYT [27, 28], functional characterization remains mainly elusive. However, two examples indicate that membranebound proteins with periplasmic domains involved in c-di-GMP signaling sense
periplasmic substrate-binding proteins and thereby modulate their readout [12, 29]
(fig. 2). Another example indicates sensing of the antibiotic tobramycin by a membrane-bound EAL domain protein [30].
Among the characterized cytoplasmic sensor and signaling domains coupled to
c-di-GMP-metabolizing proteins, PAS, GAF and Rec are predominant. The GAF and
PAS clans contain domain families whose members are sequentially diverse, smallmolecule binding units, which sense for example oxygen and the redox status or bind
small nucleotides. Indeed, it has been demonstrated that heme-binding PAS domains
of GGDEF-EAL domain proteins from E. coli and Gluconacetobacter xylinus [31, 32]
sense oxygen directly. Thereby, the phosphodiesterase activity of AxPDE1 in G. xylinus, a PAS-GAF-GGDEF-EAL protein, is inhibited when oxygen binds to the hemecontaining PAS domain [32] (fig. 2).
Rec represents the receiver domain in two-component systems whereby phosphor
transfer from a membrane-bound sensor kinase to the conserved aspartate at position 59 in the receiver domain is the activation mechanism for the response regulator.
The output of a conventional response regulator upon activation by phosphorylation is enhanced sequence-specific DNA binding via the C-terminal helix-turn-helix
domain, but a significant fraction of response regulators have a C-terminal GGDEF or
EAL domain and therefore c-di-GMP synthesis or degradation as response [15, 33, 34]
(fig. 2). Frequently, the response regulators which have c-di-GMP signal modulation

Prevailing Concepts of c-di-GMP Signaling

165

a Phosphorylation of recevier domain


P

cdiGMP
Rec

Rec

GGDEF

PleD
Caulobacter crescentus CB15

GTP

b Binding of c-di-GMP to GGDEF domain


Rec

cdiGMP
GGDEF

Rec

cdiGMP
GTP

c Binding of GTP to GGDEF domain


GTP
GGDEF

EAL

GAF

cdiGMP pGpG

CC_3396
Caulobacter crescentus CB15
cdiGMP pGpG

d O2-binding to PAS domain


O2
PAS

PleD

GGDEF

EAL

PDEA1
Gluconacetobacter xylinus

e Protein sensing in the periplasm


cdiGMP pGpG

SBP_Bac_1
HA
MP

GGDEF

VC0703
Vibrio cholerae

EAL

f Sensing of antibiotics

cdiGMP pGpG
EAL

PA2818
Pseudomonas aeruginosa

Fig. 2. Signals regulating the activity of diguanylate cyclases and phosphodiesterases. a


Phosphorylation of receiver domains. Phosphorylation of the conserved aspartate in the receiver
domain (Rec) of response regulators is a common mechanism to activate the diguanylate cyclase
function of GGDEF domains as exemplified with the response regulator PleD [6]. b Binding of c-diGMP to GGDEF domain. C-di-GMP binding with subsequent allosteric product inhibition of diguanylate cyclase activity is predicted to occur in approximately 60% of GGDEF domains that contain an
I-site [43]. c Binding of GTP to GGDEF domain. Binding of GTP to GGDEF domains with degenerated
GGDEF motif activates the phosphodiesterase function of N-terminal EAL domains [18]. d Oxygen
binding to PAS domain. The binding of oxygen to the PAS domain of G. xylinus inhibits the phosphodiesterase A1 in G. xylinus [32]. e Protein sensing in the periplasm. The GGDEF-EAL domain protein
MbaA in V. cholerae senses a periplasmic putative norspermidine-binding protein loaded with substrate, which diminishes the ability of MbaA to inhibit biofilm formation [29]. f The EAL domain protein Arr (PA2818) senses the antibiotic tobramycin [30]. Domains: GAF, nucleotide/nucleoside-binding
domain; SBP_Bac_1, solute-binding protein family 1.

as an output are part of non-classical two-component systems, e.g. an integrative part


of three-component systems which also contain a conventional response regulator
[3538].
The domain structure GGDEF-EAL is especially frequent, with more than onethird of the GGDEF and more than half of the EAL domains found in this combination. The HD-GYP domain is also found, although rarely, in combination
with the GGDEF domain. GGDEF domains adjacent to EAL domains represent a
distinct subgroup of GGDEF domains [unpubl. data]. The GG(D/E)EF and other
motifs characteristic for diguanylate cyclase activity are frequently subject to alteration indicating non-enzymatic roles for the GGDEF domains in the GGDEF-EAL

166

Rmling Simm

constellation [39]. Some GGDEF domains with a non-canonical GGDEF motif have
retained the capability to bind the substrate GTP. However, upon binding, GTP is not
cleaved, but the GGDEF domain allosterically enhances the phosphodiesterase activity of the C-terminal EAL domain [18, 40, 41] (fig. 2). This control mechanism might
have evolved to avoid drainage of the GTP pools by uncontrolled phosphodiesterase
activity. Control of PDE activity by small effector molecules has also been shown in
eukaryotes [42].
A c-di-GMP dimer bound at the interface between the GGDEF and N-terminal
Rec2 domain responsible for non-competitive product inhibition of the diguanylate
cyclase activity of PleD was identified by resolving the crystal structure of PleD [9,
43]. An arginine and aspartate residue with direct contacts to the c-di-GMP ligand
constitutes the RxxD core motif required for c-di-GMP binding to the inhibitory site
(I-site) in the GGDEF domain. C-di-GMP binding and allosteric product inhibition
is modulated by the sequence context of RxxD and actually inhibited by the charged
residues in the Rec2 domain that interact with the c-di-GMP dimer. Allosteric control of diguanylate cyclase activity by c-di-GMP seems to be common, since the RxxD
motif is present in more than 60% of the proteins with consensus GGDEF motif.

C-di-GMP Binding Sites

Secondary signaling requires a target molecule (receptor) binding the messenger


to confer the downstream physiological events. Thereby, the target molecule can be
directly the effector, but also an adaptor protein which links the signal to a physiological response. Until now, a variety of c-di-GMP receptors have already been identified,
although many more are expected to be discovered. The nature of the detected receptors also starts to shed some light on the regulatory mechanisms of c-di-GMP signaling
which can be on the transcriptional, post-transcriptional and post-translational level.
Bioinformatic analysis predicted that the PilZ domain, named after PilZ (PA2960),
a 118-amino-acid protein required for twitching motility in Pseudomonas aeruginosa,
is a candidate protein to bind c-di-GMP [44]. Subsequently, the binding of c-di-GMP
to PilZ domains was experimentally confirmed [4548] (fig. 3). A few highly conserved residues clustered in two separate motifs, the RxxxR and D/NxSxxG sequence
motifs located within the first 30 amino acids of the PilZ domain, are required for
ligand contact and binding. Upon binding of c-di-GMP, an N-terminal seven-residue
loop containing the highly conserved RxxxR motif undergoes a dramatic conformational change [45, 47, 49]. This conformational change, named the c-di-GMP switch,
makes the PilZ domain more compact and significantly alters the relative position of
N-terminal domains which provides, in combination with the bound c-di-GMP, novel
surfaces for allosteric interactions [49]. The high sequence diversity of PilZ domains
and variable domain composition of the proteins indicates an early divergence into
different functional protein families with a diversity of downstream physiological

Prevailing Concepts of c-di-GMP Signaling

167

100

200

Glycos
transf_2

BcsA

YcgR

YcgR

PleD

Rec

PilZ

Rec

GGDEF

GAF

PelD

FleQ

PilZ

FleQ

AAA+

HTH_8

Fig. 3. C-di-GMP receptors. The


PilZ domain is a family of c-diGMP-binding proteins present
in combination with various
N-terminal domains [44, 49].
Examples of PilZ domains are
the bacterial cellulose synthase
BcsA and motility-controlling
YcgR in E. coli and S. typhimurium [45]. Presumably 60% of
the GGDEF domains bind c-diGMP through the I-site as in
PleD [43]. Individual proteins
with c-di-GMP-binding capacity are the transcriptional regulator FleQ and PelD encoded by
the pel operon in P. aeruginosa.
GEMM riboswitches as Vc2 also
bind c-di-GMP. Dark grey dots
indicate amino acids involved
in c-di-GMP binding. Light grey
dots indicate less than 50%
diminished c-di-GMP-binding
capacity.

Vc2 RNA

events affected by PilZ domain proteins. Thereby, PilZ domain proteins can be an
integral part of effector proteins as it is exemplified for the cellulose synthase BcsA
with a C-terminal PilZ domain. More frequently, however, PilZ domains are standalone proteins, fused to another PilZ domain or C-terminal of a YcgR-like domain.
In those instances, PilZ domain proteins likely functions as adaptors. PilZ domain
proteins are involved in the regulation of physiological processes such as alginate biosynthesis, biofilm formation, cellulose biosynthesis, motility, extracellular enzyme
production, virulence and toxin production [45, 47, 48, 50, 51] whereby c-di-GMP
binding to the PilZ domain is required for repression of motility and virulence and
activation of exopolysaccharide synthesis.
Besides the PilZ domain family, there are a few other individual protein receptors
identified for c-di-GMP (fig. 3). These c-di-GMP-binding proteins have been detected
experimentally after analysis of c-di-GMP-dependent processes. Indeed, besides conserved binding motifs such as the RxxD motif of the c-di-GMP-binding I-site, there
is marginal or no sequence similarity between different types of c-di-GMP-binding
proteins. This apparent lack of sequence homology indicates that c-di-GMP-binding
sites cannot be predicted from sequence comparison but extensive experimentation is
still a prerequisite for the identification of novel c-di-GMP-binding proteins.
Pronounced phenotypes mediated by c-di-GMP signaling in P. aeruginosa are pellicle formation (a bacterial network at the air-liquid interface) and the appearance of
wrinkled colonies of agar-grown bacteria which requires the biosynthesis of pel and

168

Rmling Simm

psl exopolysaccharides [52]. Polysaccharide biosynthesis is activated by elevated c-diGMP levels transcriptionally and post-transcriptionally [34].
On the other hand, the transcription of genes involved in the biosynthesis of pel
and psl exopolysaccharides [53] is negatively regulated by the transcriptional regulator
FleQ, the master regulator of flagella gene expression in P. aeruginosa [54]. Thereby,
FleQ binds to the promoter region of the pel biosynthesis operon and represses transcription. Upon c-di-GMP binding, however, FleQ is relieved from promoter binding
to cause the derepression of the expression of pel and other EPS genes [55]. Since
c-di-GMP has only a marginal effect on transcription of motility genes, the interaction of FleQ with the promoters for motility genes is different from the interaction
with the promoters of pel and other EPS genes.
Posttranscriptional regulation of the Pel exopolysaccharide requires PelD, the c-diGMP-binding protein encoded by the operon for biosynthesis of Pel polysaccharide
[56]. PelD was identified by the systematic search for the c-di-GMP-binding protein
among the proteins encoded by genes of the biosynthesis operon for the Pel polysaccharide that lacked a PilZ domain protein as a c-di-GMP receptor. PelD does not have
any significant homology to other common c-di-GMP-binding sites, however, PelD
contains an RxxD motif similar to the I-site in GGDEF domain proteins [43].
Riboswitches are mRNA domains located in the 5-untranslated region (UTR) that
control gene expression in response to changing concentrations of their target ligand.
C-di-GMP, which is actually a small cyclic RNA, binds to a highly conserved riboswitch called GEMM (Genes for the Environment, for Membranes and for Motility)
[57, 58]. The GEMM motif was identified by bioinformatic analysis of 5-UTR
regions for conserved structured RNA motifs. The GEMM motif is widespread in
Gram-positive and Gram-negative bacteria, common in - and -proteobacteria. The
GEMM motif is present in the 5-UTR regions of genes encoding a diversity of gene
products. Frequently, genes or operons that mediate phenotypes known to be modulated by c-di-GMP signaling, such as whole operons encoding flagella biosynthesis,
motility genes and GGDEF and EAL domain proteins contain a GEMM motif.
Depending on the position in the 5-UTR, the riboswitch can control transcriptional termination or translation initiation. It can function as an on or off switch as
exemplified by the effect of c-di-GMP on expression of a competence gene in Vibrio
cholerae, which is switched on, and on expression of the flagella biosynthesis operon in
Clostridium difficile, which is switched off upon high c-di-GMP levels [58]. Although
the GEMM riboswitch is widespread, its relative abundance is by far not sufficient to
explain the diverse effects of c-di-GMP on transcription and mRNA stability.

Transition between Sessility and Motility by C-di-GMP Signaling

There is a diversity of phenotypes modulated positively or negatively by c-di-GMP


signaling. However, common principles of regulation by c-di-GMP have become

Prevailing Concepts of c-di-GMP Signaling

169

apparent as demonstrated in a variety of bacteria. Numerous studies have shown


that biofilm formation is positively regulated by c-di-GMP signaling [5, 10, 5963]
(fig. 4). On the other hand, motility is a phenotype inhibited by elevated c-di-GMP
concentrations (fig. 4). Consequently, the transition between motility and sessility
is a behavioral pattern regulated by c-di-GMP signaling in many bacteria such as P.
aeruginosa, E. coli and Salmonella typhimurium, V. cholerae and Shewanella [5, 26].
Early experiments have been conducted with randomly chosen diguanylate cyclases
and phosphodiesterases overexpressed to globally manipulate the c-di-GMP levels [5,
61, 6466]. Thereby, the general effect of alterations in c-di-GMP concentrations on
motility and biofilm formation indicates that c-di-GMP signaling acts downstream of
other regulatory mechanisms and/or can override existing regulatory circuits. On the
level of chromosomal expression, however, specific diguanylate cyclases and phosphodiesterases regulate mobility and biofilm formation.
Although the above-described transition pattern holds true for all bacteria tested,
there is substantial variation in the details how and on which level c-di-GMP promotes sessility or inhibits the motile state of the bacterium. In S. typhimurium, the
central regulator of biofilm formation CsgD is a major target for c-di-GMP signaling
whereby the expression is regulated on the transcriptional and post-transcriptional
level [59]. Consequently, at least two diguanylate cyclases are required to upregulate
the expression of CsgD [59], and four phosphodiesterases downregulate the expression of CsgD [67]. Downstream of CsgD expression the diguanylate cyclase AdrA
produces c-di-GMP dedicated mainly to the biosynthesis of cellulose and partially to
the biosynthesis of curli fimbriae on a post-transcriptional level [59, 68, 69]. However,
transcription of the genes constituting the cellulose biosynthesis operon is not altered
by c-di-GMP signaling [69]. This is in contrast to V. cholerae and P. aeruginosa, where
c-di-GMP signaling alters transcription of operons encoding proteins for exopolysaccharide biosynthesis [34].
Equally, c-di-GMP regulates motility on different levels in individual bacterial
species and among species from the transcription of flagellar genes to the interference
with flagellar function. Transcriptional control of flagella genes by c-di-GMP signaling has been observed in Pseudomonas putida, V. cholerae and Vibrio parahaemolyticus [64, 70, 71]. Post-transcriptional control of flagella biosynthesis occurs in C.
crescentus and V. fisheri [72, 73]. In addition, upon completion of the flagellar assembly, c-di-GMP can interfere with flagella function by inhibiting the rotation of flagella
or regulation of the flagella reversal rate. Functional interference has been observed
in S. typhimurium, E. coli, C. crescentus and P. aeruginosa [47, 60, 74, 75]. In addition,
c-di-GMP signaling controls flagella positioning and ejection [73, 76]. The regulation
of flagella-mediated swimming and swarming motility by c-di-GMP has been extensively covered in a recent review [77].
In addition, type IV pili-mediated twitching motility in P. aeruginosa is regulated
by c-di-GMP signaling [40, 78]. However, the c-di-GMP-binding protein required
for repression of type IV pili-mediated twitching motility is unknown as the PA2960

170

Rmling Simm

Motility
swimming
swarming
twitching
Virulence
toxin production
cytotoxicity
invasion
IL-8 production

2 GTP

pGpG
2 GMP

GGDEF

EAL
HD-GYP

phage resistance
heavy metal resistance
chemotaxis

c-di-GMP

c-di-GMP

Sessility
biofilm formation
rugose colony
morphology
adhesive matrix
components fimbriae
exopolysaccharides
Virulence
toxin production
cytotoxicity
intracellular replication
Cell morphology
cell elongation
long term survival
heterocyst development
photosynthesis
cell-cell communication

Fig. 4. Phenotypes regulated by c-di-GMP signaling. On each side, phenotypes stimulated by the
respective c-di-GMP concentration (low [left side] or high [right side]) are indicated.

protein name-giving for the family of PilZ domain c-di-GMP receptors actually does
not bind c-di-GMP [48].

Transition between Sessility and Motility Occurs on the Single Cell Level

The status sessility is strongly associated with the production of adhesive extracellular
matrix components by the bacterial cell used for attachment to surfaces and/or other
bacterial cells to build up three-dimensional biofilm consortia. Some examples of
adhesive components where the production is stimulated by c-di-GMP are the holdfast of Caulobacter cresentus, the exopolysaccharide cellulose produced by enterobacteria and the fruit-degrading bacterium G. xylinus and the cell surface protein LapA
of P. fluorescens [33, 79, 80]. Sessility of cells enhances biofilm formation, although the
term sessility is not synonymous to biofilm formation, as the term sessility refers to the
status of an individual cell, while the term biofilm is referring to cell populations.
Actually, intriguing studies by Klausen et al. [81] using the continuous-flow model
which allows the observation of biofilm formation in real-time on the single cell level
have shown that the biofilm community consists of populations of sessile cells as
well as motile cells. Both cell populations are required to build up a mature biofilm.
Extrapolating the behavior of cells in the population to c-di-GMP concentrations,
one can assume that the c-di-GMP levels are substantially different in individual cells.
How such a different c-di-GMP concentration can arise is not clear, but individual
cells in the biofilm sense different microenvironments.

Prevailing Concepts of c-di-GMP Signaling

171

OD600

2
1
0

UMR1

1703

4264

3611

1827

1344

3375

adrA

2123

3388

1987

Fig. 5. Specificity and redundancy of c-di-GMP-metabolizing enzymes in biofilm formation in S.


typhimurium. a Knockout in genes encoding GGDEF and EAL domain proteins were screened for
biofilm formation in liquid culture. Shown are 10 knockouts that display a consistently altered phenotype compared to wild-type S. typhimurium UMR1 (left) [116] in this assay and/or other biofilm
assays [59, 67]. The numbers shown are the annotated gene numbers in S. typhimurium (LT2). b
Quantification of planktonic cells. The OD as determined from the top layer of the cultures after
sedimentation of the cell clumps for 30 min reflects the number of planktonic cells in the culture
which is inversely proportional to biofilm formation. The STM1703 and STM1827 mutants form more
and larger cell clumps compared to S. typhimurium UMR1 resulting in a clear medium and low OD600
values. The STM3375 and STM1987 mutants are clearly downregulated in adherence and clumping
ability resulting in more planktonic cells and high OD600 values.

Role of C-di-GMP Signaling in Virulence

Besides sessility and motility, a variety of different phenotypes are regulated by c-diGMP signaling. Virulence is a phenotype frequently found to be affected by c-diGMP signaling in animal as well as plant pathogens [35, 37, 8286]. Thereby, based
on early studies on the role of c-di-GMP in virulence [87, 88], there exists the common paradigm that high c-di-GMP levels suppress the virulence phenotypes of acute
infections, while low c-di-GMP concentrations stimulate virulence properties. On the
other hand, undoubtedly, biofilm formation, which requires elevated c-di-GMP levels
for expression, is a virulence factor in the establishment of chronic infections [89].
These findings have lead to the view that c-di-GMP signaling mediates the transition
between acute and chronic infections.
Up to now, there is only a glimpse on the mechanisms how c-di-GMP suppresses virulence phenotypes. In V. cholerae, low c-di-GMP levels created by the
expression of the phosphodiesterase VieA are required for virulence in a mouse
model [88]. Specifically, absence of VieA suppress the production of cholera toxin

172

Rmling Simm

a major virulence factor in cholera. C-di-GMP affects the transcription of cholera


toxin indirectly by suppressing the expression of the transcriptional regulator ToxT
[88].
In S. typhimurium, another enteric pathogen, c-di-GMP has an inhibitory effect
on virulence properties. C-di-GMP signaling inhibits invasion of a gastrointestinal
epithelial cell line and the gut epithelium of mice [90]. In addition to this the bacterial
induction of the proinflammatory cytokine IL-8 in the gastrointestinal epithelial cell
line is also abolished by high c-di-GMP levels.
As described above, screens for reduction of virulence detected clear-cut phenotypes that implicate an inhibitory role for c-di-GMP in acute infections. In contrast, systematic studies of genes involved in c-di-GMP metabolism for virulence
phenotypes do not give such a clear-cut picture, although a similar trend can be
observed. A systematic knockout and overexpression study of GGDEF and EAL
domain proteins in P. aeruginosa clearly showed that biofilm formation requires signaling through c-di-GMP [35]. However, the role of c-di-GMP in virulence was less
clear. Type III secretion-mediated cytotoxicity in a cell culture model, an indicator
of virulence, and virulence in a burn wound mouse model were positively affected
by either GGDEF or EAL domain proteins. In addition, comparison of the genes
demonstrated to have a role in virulence in mice showed only a partial overlap with
those with a role in cytotoxicity.
A similar ambivalent role of c-di-GMP signaling in virulence has been observed in
the plant pathogen Xanthomonas campestris. Virulence to Chinese radish and secretion of virulence factors is positively affected by either GGDEF, EAL or HD-GYP
domain proteins [82]. As in P. aeruginosa, in vivo and in vitro phenotypes of proteins
were only partially congruent.
The requirement of the EAL-like protein STM1344 for virulence of S. typhimurium in a mouse model of systemic infection indicated an inhibitory role of c-diGMP in virulence [87]. However, detailed analysis showed that STM1344 does
not exert phosphodiesterase activity, but rather enhances c-di-GMP levels in the
cell [91]. In extrapolation, STM1344-mediated phenotypes, resistance to hydrogen peroxide stress and reduced macrophage killing most likely require c-di-GMP
signaling.
These studies together with a recent study on the role of c-di-GMP signaling in the
intracellular pathogen Anaplasma phagocytophilum [85], which only possesses one
GGDEF domain protein, suggest that tight regulation of c-di-GMP signaling during
infection is required for virulence.

Additional Phenotypes Affected by C-di-GMP Signaling

The diversity of phenotypes demonstrated or indicated to be regulated by c-diGMP signaling is broad. Overexpression or knockout studies have discovered that

Prevailing Concepts of c-di-GMP Signaling

173

phenotypes like cell-to-cell communication, growth competition, chemotaxis, longterm survival, heavy metal resistance, phage resistance, transposition frequency,
antibiotic resistance, cell morphology, heterocyst development and photosynthesis are regulated by c-di-GMP signaling [63, 65, 92102]. Of particular note is the
recent first experimental description of a c-di-GMP signaling pathway in a Grampositive bacterium [101]. In Mycobacterium smegmatis, which has only one functional GGDEF-EAL domain protein, c-di-GMP signaling is involved in long-term
survival.
A convenient way to elucidate the range of physiological pathways affected by
c-di-GMP signaling in one organism is overexpression of GGDEF and EAL domain
proteins. Ideally, by overexpression of a highly potent diguanylate cyclase, c-di-GMP
concentrations in the cytoplasm reach sufficient concentration to saturate receptors
for all physiological pathways stimulated or inhibited by c-di-GMP signaling. On the
other hand, depletion of the cell from c-di-GMP by overexpression of a c-di-GMPspecific phosphodiesterase has the opposite effect on cell physiology. This strategy
in combination with microarray analysis was used to elucidate c-di-GMP-dependent pathways in V. cholerae and E. coli [64, 65]. In both species, a distinct subset of
genes was transcriptionally altered in response to high c-di-GMP levels. Remarkably,
a major fraction of genes regulated by altered c-di-GMP concentrations was of
unknown function (35% in E. coli; 51% in V. cholerae), indicating that phenotypes
or regulatory mechanisms related to c-di-GMP metabolism are mainly unexplored.
Besides the somewhat expected effect of c-di-GMP on cell envelope-related proteins,
the microarray studies showed that energy metabolism and transport and binding of
small molecules is subject to c-di-GMP-mediated regulation on both organisms. The
microarray studies also showed that, as the expression of a variety of transposases and
integrases are affected by c-di-GMP signaling, transposition events might be subject
to c-di-GMP regulation in V. cholerae [64], while iron uptake might be subject to
c-di-GMP regulation in E. coli [65].

Redundancy and Specificity of C-di-GMP Signaling

As there is commonly more than one c-di-GMP synthesizing and one c-di-GMPdegrading enzyme encoded by the genome of a species, the question about the
mechanisms of specificity of the c-di-GMP-metabolizing enzymes arises. Obviously,
there is specificity of the individual GGDEF and EAL domain proteins as the knockouts of individual GGDEF and EAL domain proteins provide clear-cut phenotypes
despite the presence of numerous GGDEF, EAL and/or HD-GYP domain proteins
on the chromosome. For example, the biofilm phenotype in Yersinia pestis exclusively requires the diguanylate cyclase HmsT for expression at ambient temperature
on agar plates [10, 103], while the diguanylate cyclase AdrA is required for the biosynthesis of the exopolysaccharide cellulose in S. typhimurium on agar plates [68]. In

174

Rmling Simm

Caulobacter cresentus, the diguanylate cyclase PleD is required for flagella ejection,
stalk formation and biosynthesis of the hold-fast [33, 76]. Specificity is also observed
for phosphodiesterases as the phosphodiesterase STM1703 (YciR) exclusively affects
CsgD expression, but not swimming and swarming motility in S. typhimurium and
E. coli. Consequently, specificity can extend to the metabolic circuit of c-di-GMP as
corresponding diguanylate cyclases and phosphodiesterases, which inversely affect
the same phenotype, probably metabolize the same c-di-GMP pool [41, 60, 62, 104].
However, specificity is not absolute, but depends on environmental and genetic factors as the strain background, the expression level of the c-di-GMP-metabolizing
protein and changes in the amino acid sequence of c-di-GMP-metabolizing enzymes
which lead to higher enzymatic activity [105109].
The mechanisms of specificity are just at the beginning to be dissected, however,
there is rising evidence that more than one mechanism assures the specificity of the
signaling system.
Temporal control of cyclic di-GMP-metabolizing enzymes is a mechanism of
specificity which involves regulated expression on a transcriptional and post-transcriptional level in response to extra- and/or intracellular signals. For example, the
diguanylate cyclase AdrA is exclusively activated by the biofilm regulator CsgD under
all environmental conditions tested. On the other hand, global regulatory proteins
such as the stress sigma factor RpoS or the carbon storage regulator CsrA can activate
and repress GGDEF and EAL domain proteins in a concerted action [11, 62, 108,
110]. However, expression control can only partially explain the specificity of c-diGMP-metabolizing enzymes, as even overexpression of GGDEF, EAL or HD-GYP
proteins does not necessarily affect a certain phenotype [35].
A second mechanism of specificity relies on the regulation of enzymatic activities
of the c-di-GMP-metabolizing enzymes on environmental or intracellular stimuli. As
discussed before, tight regulation of activities of c-di-GMP-metabolizing enzymes is
achieved by allosteric product inhibition and the allosteric regulation by N-terminal
domains with sensing or signaling capability (fig. 2). Protein-protein interactions
might also affect the enzymatic activities of c-di-GMP-metabolizing enzymes, as the
HD-GYP domain protein RpfG has been shown to interact with various GGDEF
domain proteins [111].
Besides the highly regulated production of c-di-GMP, the regulated degradation of
c-di-GMP is equally important to generate localized c-di-GMP pools. Substantially
higher c-di-GMP concentrations than in the wild-type were obtained when the phosphodiesterase STM4264 was deleted in S. typhimurium [67] indicating that STM4264
degrades the major part of the c-di-GMP produced in the cell.
A third mechanism involves co-localization of enzymes involved in c-di-GMP
turnover with their targets. Stalk cell development in C. crescentus requires c-di-GMP
and the response regulator diguanylate cyclase PleD. Upon activation of PleD by
phosphorylation, PleD is simultaneously localized to the differentiating pole [6] suggesting that proximity to the target is required for PleD functionality.

Prevailing Concepts of c-di-GMP Signaling

175

Specificity of C-di-GMP Signaling A Phenomenon Based on Receptor Affinity?

The specificity of c-di-GMP signaling raised the question about the presence and
mechanisms of localized c-di-GMP pools as it seemed to be in contrast to the assumption that c-di-GMP is a freely diffusible molecule. The experimentally found 80%
membrane association for the c-di-GMP molecule in G. xylinus points also to a localization of the molecule [112].
First attempts to estimate the c-di-GMP concentration calculated 8 molecules per
bacterial cell [87]. A more realistic number is probably the recently estimated 250
molecules per cell for S. typhimurium and 2,500 molecules per cell for V. cholerae
[113]. The c-di-GMP signaling network is highly modular with many receptors for
c-di-GMP such as adaptor proteins as PilZ domain proteins, but also GGDEF domain
proteins with I-sites. The ratio of potential c-di-GMP-binding sites to free c-di-GMP
molecules in the cell is unknown, however one can assume a limited number of c-diGMP molecules in the cell and the presence of several c-di-GMP receptors, which
modulate physiological events, expressed at the same time. The specificity of the c-diGMP signaling would therefore be determined by the expression levels of c-di-GMP
receptors and the binding affinity for c-di-GMP factors which determines the sequestration of c-di-GMP by a certain protein [48].

Conclusions

Many basic principles of c-di-GMP signaling have been unraveled. Enzymes required
for the biosynthesis and hydrolysis of c-di-GMP have been identified and their
enzymatic activities have been characterized subsequently with several mechanisms
of regulation of c-di-GMP metabolism. Widespread effector mechanisms for c-diGMP are the binding of c-di-GMP to the PilZ domain and sensing of c-di-GMP by
riboswitches. Common phenotypes regulated by c-di-GMP signaling are the transition between sessility and motility and regulation of virulence properties. However,
although many common principles and phenotypes exist in c-di-GMP signaling,
there is also a huge diversity of metabolism, effector mechanisms and phenotypes
present and yet to be discovered. Still, we have only scratched the surface in this new
area of research.

Acknowledgements
Work in the laboratory of U.R. is supported by Vetenskapsrdet, the Carl Trygger Foundation and
the European Commission under contract No. MEST-2004-0875.

176

Rmling Simm

References
1 Conti M, Beavo J: Biochemistry and physiology of
cyclic nucleotide phosphodiesterases: essential
components in cyclic nucleotide signaling. Annu
Rev Biochem 2007;76:481511.
2 Kolb A, Busby S, Buc H, Garges S, Adhya S:
Transcriptional regulation by cAMP and its receptor protein. Annu Rev Biochem 1993;62:749795.
3 Tischler AD, Camilli A: Cyclic diguanylate (c-diGMP) regulates Vibrio cholerae biofilm formation.
Mol Microbiol 2004;53:857869.
4 Witte G, Hartung S, Buttner K, Hopfner KP:
Structural biochemistry of a bacterial checkpoint
protein reveals diadenylate cyclase activity regulated
by DNA recombination intermediates. Mol Cell
2008;30:167178.
5 Simm R, Morr M, Kader A, Nimtz M, Rmling U:
GGDEF and EAL domains inversely regulate cyclic
di-GMP levels and transition from sessility to motility. Mol Microbiol 2004;53:11231134.
6 Paul R, Weiser S, Amiot NC, Chan C, Schirmer T,
Giese B, Jenal U: Cell cycle-dependent dynamic
localization of a bacterial response regulator with a
novel di-guanylate cyclase output domain. Genes
Dev 2004;18:715727.
7 Galperin MY: A census of membrane-bound and
intracellular signal transduction proteins in bacteria: bacterial IQ, extroverts and introverts. BMC
Microbiol 2005;5:35.
8 Rmling U: Characterization of the rdar morphotype, a multicellular behaviour in Enterobacteriaceae.
Cell Mol Life Sci 2005;62:12341246.
9 Chan C, Paul R, Samoray D, Amiot NC, Giese B,
Jenal U, Schirmer T: Structural basis of activity and
allosteric control of diguanylate cyclase. Proc Natl
Acad Sci USA 2004;101:1708417089.
10 Kirillina O, Fetherston JD, Bobrov AG, Abney J,
Perry RD: HmsP, a putative phosphodiesterase, and
HmsT, a putative diguanylate cyclase, control Hmsdependent biofilm formation in Yersinia pestis. Mol
Microbiol 2004;54:7588.
11 Jonas K, Edwards AN, Simm R, Romeo T, Rmling
U, Melefors : The RNA binding protein CsrA controls cyclic di-GMP metabolism by directly regulating the expression of GGDEF proteins. Mol
Microbiol 2008;70:236257.
12 Ferreira RB, Antunes LC, Greenberg EP, McCarter
LL: Vibrio parahaemolyticus ScrC modulates cyclic
dimeric GMP regulation of gene expression relevant
to growth on surfaces. J Bacteriol 2008;190:851
860.
13 Linder JU: Class III adenylyl cyclases: molecular
mechanisms of catalysis and regulation. Cell Mol
Life Sci 2006;63:17361751.

Prevailing Concepts of c-di-GMP Signaling

14 Wassmann P, Chan C, Paul R, Beck A, Heerklotz H,


Jenal U, Schirmer T: Structure of BeF3-modified
response regulator PleD: implications for diguanylate cyclase activation, catalysis, and feedback inhibition. Structure 2007;15:915927.
15 Ryjenkov DA, Tarutina M, Moskvin OV, Gomelsky
M: Cyclic diguanylate is a ubiquitous signaling molecule in bacteria: insights into biochemistry of the
GGDEF protein domain. J Bacteriol 2005;187:1792
1798.
16 Schmidt AJ, Ryjenkov DA, Gomelsky M: The ubiquitous protein domain EAL is a cyclic diguanylatespecific phosphodiesterase: enzymatically active
and inactive EAL domains. J Bacteriol 2005;187:
47744781.
17 Tamayo R, Tischler AD, Camilli A: The EAL domain
protein VieA is a cyclic diguanylate phosphodiesterase. J Biol Chem 2005;280:3332433330.
18 Christen M, Christen B, Folcher M, Schauerte A,
Jenal U: Identification and characterization of a
cyclic di-GMP-specific phosphodiesterase and its
allosteric control by GTP. J Biol Chem 2005;280:
3082930837.
19 Rao F, Yang Y, Qi Y, Liang ZX: Catalytic mechanism
of cyclic di-GMP-specific phosphodiesterase: a
study of the EAL domain-containing RocR from
Pseudomonas aeruginosa. J Bacteriol 2008;190:3622
3631.
20 Galperin MY, Natale DA, Aravind L, Koonin EV: A
specialized version of the HD hydrolase domain
implicated in signal transduction. J Mol Microbiol
Biotechnol 1999;1:303305.
21 Galperin MY, Nikolskaya AN, Koonin EV: Novel
domains of the prokaryotic two-component signal
transduction systems. FEMS Microbiol Lett 2001;
203:1121.
22 Ryan RP, Fouhy Y, Lucey JF, Crossman LC, Spiro S,
He YW, Zhang LH, Heeb S, Camara M, Williams P,
Dow JM: Cell-cell signaling in Xanthomonas campestris involves an HD-GYP domain protein that
functions in cyclic di-GMP turnover. Proc Natl
Acad Sci USA 2006;103:67126717.
23 Rashid MH, Rajanna C, Ali A, Karaolis DK:
Identification of genes involved in the switch between
the smooth and rugose phenotypes of Vibrio cholerae. FEMS Microbiol Lett 2003;227:113119.
24 Drenkard E, Ausubel FM: Pseudomonas biofilm formation and antibiotic resistance are linked to phenotypic variation. Nature 2002;416:740743.
25 Rmling U, Gomelsky M, Galperin MY: c-di-GMP:
The dawning of a novel bacterial signaling system.
Mol Microbiol 2005;57:629639.
26 Rmling U, Amikam D: Cyclic di-GMP as a second
messenger. Curr Opin Microbiol 2006;9:218228.

177

27 Galperin MY, Gaidenko TA, Mulkidjanian AY,


Nakano M, Price CW: MHYT, a new integral membrane sensor domain. FEMS Microbiol Lett 2001;
205:1723.
28 Nikolskaya AN, Mulkidjanian AY, Beech IB,
Galperin MY: MASE1 and MASE2: two novel integral membrane sensory domains. J Mol Microbiol
Biotechnol 2003;5:1116.
29 Karatan E, Duncan TR, Watnick PI: NspS, a predicted polyamine sensor, mediates activation of
Vibrio cholerae biofilm formation by norspermidine. J Bacteriol 2005;187:74347443.
30 Hoffman LR, DArgenio DA, MacCoss MJ, Zhang Z,
Jones RA, Miller SI: Aminoglycoside antibiotics
induce bacterial biofilm formation. Nature 2005;
436:11711175.
31 Delgado-Nixon VM, Gonzalez G, Gilles-Gonzalez
MA: Dos, a heme-binding PAS protein from
Escherichia coli, is a direct oxygen sensor.
Biochemistry 2000;39:26852691.
32 Chang AL, Tuckerman JR, Gonzalez G, Mayer R,
Weinhouse H, Volman G, Amikam D, Benziman M,
Gilles-Gonzalez MA: Phosphodiesterase A1, a regulator of cellulose synthesis in Acetobacter xylinum, is
a heme-based sensor. Biochemistry 2001;40:3420
3426.
33 Aldridge P, Paul R, Goymer P, Rainey P, Jenal U:
Role of the GGDEF regulator PleD in polar development of Caulobacter crescentus. Mol Microbiol 2003;
47:16951708.
34 Hickman JW, Tifrea DF, Harwood CS: A chemosensory system that regulates biofilm formation
through modulation of cyclic diguanylate levels.
Proc Natl Acad Sci USA 2005;102:1442214427.
35 Kulesekara H, Lee V, Brencic A, Liberati N, Urbach
J, Miyata S, Lee DG, Neely AN, Hyodo M, Hayakawa
Y, Ausubel FM, Lory S: Analysis of Pseudomonas
aeruginosa diguanylate cyclases and phosphodiesterases reveals a role for bis-(3-5)-cyclic-GMP
in virulence. Proc Natl Acad Sci USA 2006;103:
28392844.
36 Martinez-Wilson HF, Tamayo R, Tischler AD,
Lazinski DW, Camilli A: The Vibrio cholerae hybrid
sensor kinase VieS contributes to motility and biofilm regulation by altering the cyclic diguanylate
level. J Bacteriol 2008;190:64396447.
37 Merkel TJ, Stibitz S, Keith JM, Leef M, Shahin R:
Contribution of regulation by the bvg locus to respiratory infection of mice by Bordetella pertussis.
Infect Immun 1998;66:43674373.
38 Paul R, Jaeger T, Abel S, Wiederkehr I, Folcher M,
Biondi EG, Laub MT, Jenal U: Allosteric regulation
of histidine kinases by their cognate response regulator determines cell fate. Cell 2008;133:452461.

178

39 Tamayo R, Schild S, Pratt JT, Camilli A: Role of


cyclic di-GMP during Eltor biotype Vibrio cholerae
infection: characterization of the in vivo-induced
cyclic di-GMP phosphodiesterase CdpA. Infect
Immun 2008;76:16171627.
40 Kazmierczak BI, Lebron MB, Murray TS: Analysis
of FimX, a phosphodiesterase that governs twitching motility in Pseudomonas aeruginosa. Mol
Microbiol 2006;60:10261043.
41 Kuchma SL, Brothers KM, Merritt JH, Liberati NT,
Ausubel FM, OToole GA: BifA, a c-di-GMP phosphodiesterase, inversely regulates biofilm formation
and swarming motility by Pseudomonas aeruginosa
PA14. J Bacteriol 2007;189:81658178.
42 Conti M, Jin SL: The molecular biology of cyclic
nucleotide phosphodiesterases. Prog Nucleic Acid
Res Mol Biol 1999;63:138.
43 Christen B, Christen M, Paul R, Schmid F, Folcher
M, Jenoe P, Meuwly M, Jenal U: Allosteric Control
of Cyclic di-GMP Signaling. J Biol Chem 2006;281:
3201532024.
44 Amikam D, Galperin MY: PilZ domain is part of the
bacterial c-di-GMP binding protein. Bioinformatics
2006;22:36.
45 Ryjenkov DA, Simm R, Rmling U, Gomelsky M:
The PilZ domain is a receptor for the second messenger c-di-GMP. The PilZ domain protein YcgR
controls motility in enterobacteria. J Biol Chem
2006;281:3031030314.
46 Ramelot TA, Yee A, Cort JR, Semesi A, Arrowsmith
CH, Kennedy MA: NMR structure and binding
studies confirm that PA4608 from Pseudomonas
aeruginosa is a PilZ domain and a c-di-GMP binding protein. Proteins 2007;66:266271.
47 Christen M, Christen B, Allan MG, Folcher M, Jeno
P, Grzesiek S, Jenal U: DgrA is a member of a new
family of cyclic diguanosine monophosphate receptors and controls flagellar motor function in
Caulobacter crescentus. Proc Natl Acad Sci USA
2007;104:41124117.
48 Merighi M, Lee VT, Hyodo M, Hayakawa Y, Lory S:
The second messenger bis-(3-5)-cyclic-GMP and
its PilZ domain-containing receptor Alg44 are
required for alginate biosynthesis in Pseudomonas
aeruginosa. Mol Microbiol 2007;65:876895.
49 Benach J, Swaminathan SS, Tamayo R, Handelman
SK, Folta-Stogniew E, Ramos JE, Forouhar F, Neely
H, Seetharaman J, Camilli A, Hunt JF: The structural basis of cyclic diguanylate signal transduction
by PilZ domains. EMBO J 2007;26:51535166.
50 Pratt JT, Tamayo R, Tischler AD, Camilli A: PilZ
domain proteins bind cyclic diguanylate and regulate diverse processes in Vibrio cholerae. J Biol Chem
2007;282:1286012870.

Rmling Simm

51 McCarthy Y, Ryan RP, ODonovan K, He YQ, Jiang


BL, Feng JX, Tang JL, Dow JM: The role of PilZ
domain proteins in the virulence of Xanthomonas
campestris pv. campestris. Mol Plant Pathol 2008;
9:819824.
52 Friedman L, Kolter R: Two genetic loci produce distinct carbohydrate-rich structural components of
the Pseudomonas aeruginosa biofilm matrix. J
Bacteriol 2004;186:44574465.
53 Dasgupta N, Wolfgang MC, Goodman AL, Arora
SK, Jyot J, Lory S, Ramphal R: A four-tiered transcriptional regulatory circuit controls flagellar biogenesis in Pseudomonas aeruginosa. Mol Microbiol
2003;50:809824.
54 Arora SK, Ritchings BW, Almira EC, Lory S,
Ramphal R: A transcriptional activator, FleQ, regulates mucin adhesion and flagellar gene expression
in Pseudomonas aeruginosa in a cascade manner. J
Bacteriol 1997;179:55745581.
55 Hickman JW, Harwood CS: Identification of FleQ
from Pseudomonas aeruginosa as a c-di-GMPresponsive transcription factor. Mol Microbiol
2008;69:376389.
56 Lee VT, Matewish JM, Kessler JL, Hyodo M,
Hayakawa Y, Lory S: A cyclic-di-GMP receptor
required for bacterial exopolysaccharide production. Mol Microbiol 2007;65:14741484.
57 Weinberg Z, Barrick JE, Yao Z, Roth A, Kim JN,
Gore J, Wang JX, Lee ER, Block KF, Sudarsan N,
Neph S, Tompa M, Ruzzo WL, Breaker RR:
Identification of 22 candidate structured RNAs in
bacteria using the CMfinder comparative genomics
pipeline. Nucleic Acids Res 2007;35:48094819.
58 Sudarsan N, Lee ER, Weinberg Z, Moy RH, Kim JN,
Link KH, Breaker RR: Riboswitches in eubacteria
sense the second messenger cyclic di-GMP. Science
2008;321:411413.
59 Kader A, Simm R, Gerstel U, Morr M, Rmling U:
Hierarchical involvement of various GGDEF
domain proteins in rdar morphotype development
of Salmonella enterica serovar Typhimurium. Mol
Microbiol 2006;60:602616.
60 Merritt JH, Brothers KM, Kuchma SL, OToole GA:
SadC reciprocally influences biofilm formation and
swarming motility via modulation of exopolysaccharide production and flagellar function. J
Bacteriol 2007;189:81548164.
61 Thormann KM, Duttler S, Saville RM, Hyodo M,
Shukla S, Hayakawa Y, Spormann AM: Control of
formation and cellular detachment from Shewanella
oneidensis MR-1 biofilms by cyclic di-GMP. J
Bacteriol 2006;188:26812691.
62 Weber H, Pesavento C, Possling A, Tischendorf G,
Hengge R: Cyclic-di-GMP-mediated signaling
within the sigma network of Escherichia coli. Mol
Microbiol 2006;62:10141034.

Prevailing Concepts of c-di-GMP Signaling

63 Rahman M, Simm R, Kader A, Basseres E, Rmling


U, Mollby R: The role of c-di-GMP signaling in an
Aeromonas veronii biovar sobria strain. FEMS
Microbiol Lett 2007;273:172179.
64 Beyhan S, Tischler AD, Camilli A, Yildiz FH:
Transcriptome and phenotypic responses of Vibrio
cholerae to increased cyclic di-GMP level. J Bacteriol
2006;188:36003613.
65 Mendez-Ortiz MM, Hyodo M, Hayakawa Y,
Membrillo-Hernandez J: Genome wide transcriptional profile of Escherichia coli in response to high
levels of the second messenger c-di-GMP. J Biol
Chem 2006;281:80908099.
66 Caiazza NC, Merritt JH, Brothers KM, OToole GA:
Inverse regulation of biofilm formation and swarming motility by Pseudomonas aeruginosa PA14. J
Bacteriol 2007;189:36033612.
67 Simm R, Lusch A, Kader A, Andersson M, Rmling
U: Role of EAL-containing proteins in multicellular
behavior of Salmonella enterica serovar Typhimurium.
J Bacteriol 2007;189:36133623.
68 Rmling U, Rohde M, Olsen A, Normark S,
Reinkster J: AgfD, the checkpoint of multicellular
and aggregative behaviour in Salmonella typhimurium regulates at least two independent pathways.
Mol Microbiol 2000;36:1023.
69 Zogaj X, Nimtz M, Rohde M, Bokranz W, Rmling
U: The multicellular morphotypes of Salmonella
typhimurium and Escherichia coli produce cellulose
as the second component of the extracellular matrix.
Mol Microbiol 2001;39:14521463.
70 Choy WK, Zhou L, Syn CK, Zhang LH, Swarup S:
MorA defines a new class of regulators affecting flagellar development and biofilm formation in diverse
Pseudomonas species. J Bacteriol 2004;186:7221
7228.
71 Boles BR, McCarter LL: Vibrio parahaemolyticus
scrABC, a novel operon affecting swarming and
capsular polysaccharide regulation. J Bacteriol 2002;
184:59465954.
72 OShea TM, Klein AH, Geszvain K, Wolfe AJ, Visick
KL: Diguanylate cyclases control magnesium-dependent motility of Vibrio fischeri. J Bacteriol 2006;188:
81968205.
73 Huitema E, Pritchard S, Matteson D, Radhakrishnan
SK, Viollier PH: Bacterial birth scar proteins mark
future flagellum assembly site. Cell 2006;124:1025
1037.
74 Ko M, Park C: Two novel flagellar components and
H-NS are involved in the motor function of
Escherichia coli. J Mol Biol 2000;303:371382.
75 Toutain CM, Caizza NC, Zegans ME, OToole GA:
Roles for flagellar stators in biofilm formation by
Pseudomonas aeruginosa. Res Microbiol 2007;158:
471477.

179

76 Hecht GB, Newton A: Identification of a novel


response regulator required for the swarmer-tostalked-cell transition in Caulobacter crescentus. J
Bacteriol 1995;177:62236229.
77 Wolfe AJ, Visick KL: Get the message out: cyclic-DiGMP regulates multiple levels of flagellum-based
motility. J Bacteriol 2008;190:463475.
78 Huang B, Whitchurch CB, Mattick JS: FimX, a multidomain protein connecting environmental signals
to twitching motility in Pseudomonas aeruginosa. J
Bacteriol 2003;185:70687076.
79 Tal R, Wong HC, Calhoon R, Gelfand D, Fear AL,
Volman G, Mayer R, Ross P, Amikam D, Weinhouse
H, Cohen A, Sapir S, Ohana P, Benziman M: Three
cdg operons control cellular turnover of cyclic diGMP in Acetobacter xylinum: genetic organization
and occurrence of conserved domains in isoenzymes. J Bacteriol 1998;180:44164425.
80 Monds RD, Newell PD, Gross RH, OToole GA:
Phosphate-dependent modulation of c-di-GMP
levels regulates Pseudomonas fluorescens Pf0-1 biofilm formation by controlling secretion of the
adhesin LapA. Mol Microbiol 2007;63:656679.
81 Klausen M, Aaes-Jorgensen A, Molin S, TolkerNielsen T: Involvement of bacterial migration in the
development of complex multicellular structures in
Pseudomonas aeruginosa biofilms. Mol Microbiol
2003;50:6168.
82 Ryan RP, Fouhy Y, Lucey JF, Jiang BL, He YQ, Feng
JX, Tang JL, Dow JM: Cyclic di-GMP signaling in
the virulence and environmental adaptation of
Xanthomonas campestris. Mol Microbiol 2007;63:
429442.
83 Lee SH, Butler SM, Camilli A: Selection for in vivo
regulators of bacterial virulence. Proc Natl Acad Sci
USA 2001;98:68896894.
84 He J, Baldini RL, Deziel E, Saucier M, Zhang Q,
Liberati NT, Lee D, Urbach J, Goodman HM, Rahme
LG: The broad host range pathogen Pseudomonas
aeruginosa strain PA14 carries two pathogenicity
islands harboring plant and animal virulence genes.
Proc Natl Acad Sci USA 2004;101:25302535.
85 Lai TH, Kumagai Y, Hyodo M, Hayakawa Y, Rikihisa
Y: Anaplasma phagocytophilum PleC histidine
kinase and PleD diguanylate cyclase two-component system and role of cyclic di-GMP in host-cell
infection. J Bacteriol 2009;191:693700.
86 Claret L, Miquel S, Vieille N, Ryjenkov DA,
Gomelsky M, Darfeuille Michaud A: The flagellar
sigma factor FliA regulates adhesion and invasion
of Crohns disease-associated Escherichia coli via a
c-di-GMP-dependent pathway. J Biol Chem 2007;
282:3327533283.

180

87 Hisert KB, MacCoss M, Shiloh MU, Darwin KH,


Singh S, Jones RA, Ehrt S, Zhang Z, Gaffney BL,
Gandotra S, Holden DW, Murray D, Nathan C: A
glutamate-alanine-leucine (EAL) domain protein of
Salmonella controls bacterial survival in mice, antioxidant defence and killing of macrophages: role of
cyclic diGMP. Mol Microbiol 2005;56:12341245.
88 Tischler AD, Camilli A: Cyclic diguanylate regulates
Vibrio cholerae virulence gene expression. Infect
Immun 2005;73:58735882.
89 Goodman AL, Kulasekara B, Rietsch A, Boyd D,
Smith RS, Lory S: A signaling network reciprocally
regulates genes associated with acute infection and
chronic persistence in Pseudomonas aeruginosa. Dev
Cell 2004;7:745754.
90 Lamprokostopoulou A, Monteiro C, Rmling U:
Cyclic di-GMP signaling controls virulence properties of Salmonella enterica serovar Typhimurium
through CsgD, a master regulator of biofilm formation. Submitted, 2009.
91 Simm R, Remminghorst U, Ahmad I, Zakikhany K,
Rmling U: Role of STM1344 in CsgD expression. J
Bacteriol 2009, in press.
92 Thomas C, Andersson CR, Canales SR, Golden SS:
PsfR, a factor that stimulates psbAI expression in
the cyanobacterium Synechococcus elongatus PCC
7942. Microbiology 2004;150:10311040.
93 Neunuebel MR, Golden JW: The Anabaena sp.
strain PCC 7120 gene all2874 encodes a diguanylate
cyclase and is required for normal heterocyst development under high-light growth conditions. J
Bacteriol 2008;190:68296836.
94 Hyde DR, Tu CP: tnpM: a novel regulatory gene
that enhances Tn21 transposition and suppresses
cointegrate resolution. Cell 1985;42:629638.
95 Camilli A, Bassler BL: Bacterial small-molecule signaling pathways. Science 2006;311:11131136.
96 Hall BG: The rtn gene of Proteus vulgaris is actually
from Escherichia coli. J Bacteriol 1997;179:2433
2434.
97 Alksne LE, Rasmussen BA: Expression of the AsbA1,
OXA-12, and AsbM1 -lactamases in Aeromonas
jandaei AER 14 is coordinated by a two-component
regulon. J Bacteriol 1997;179:20062013.
98 Rychlik I, Martin G, Methner U, Lovell M, Cardova
L, Sebkova A, Sevcik M, Damborsky J, Barrow PA:
Identification of Salmonella enterica serovar
Typhimurium genes associated with growth suppression in stationary-phase nutrient broth cultures
and in the chicken intestine. Arch Microbiol 2002;
178:411420.
99 Huber B, Riedel K, Hentzer M, Heydorn A,
Gotschlich A, Givskov M, Molin S, Eberl L: The cep
quorum-sensing system of Burkholderia cepacia
H111 controls biofilm formation and swarming
motility. Microbiology 2001;147:25172528.

Rmling Simm

100 Gronewold TM, Kaiser D: The act operon controls


the level and time of C-signal production for
Myxococcus xanthus development. Mol Microbiol
2001;40:744756.
101 Kumar M, Chatterji D: Cyclic di-GMP: a second
messenger required for long-term survival, but not
for biofilm formation, in Mycobacterium smegmatis.
Microbiology 2008;154:29422955.
102 Sarand I, Osterberg S, Holmqvist S, Holmfeldt P,
Skarfstad E, Parales RE, Shingler V: Metabolismdependent taxis towards (methyl)phenols is coupled
through the most abundant of three polar localized
Aer-like proteins of Pseudomonas putida. Environ
Microbiol 2008;10:13201334.
103 Jones HA, Lillard JW Jr, Perry RD: HmsT, a protein
essential for expression of the haemin storage
(Hms+) phenotype of Yersinia pestis. Microbiology
1999;145:21172128.
104 Girgis HS, Liu Y, Ryu WS, Tavazoie S: A comprehensive genetic characterization of bacterial motility. PLoS Genet 2007;3:16441660.
105 Beyhan S, Yildiz FH: Smooth to rugose phase variation in Vibrio cholerae can be mediated by a single
nucleotide change that targets c-di-GMP signaling
pathway. Mol Microbiol 2007;63:9951007.
106 Beyhan S, Tischler AD, Camilli A, Yildiz FH:
Differences in gene expression between the classical
and El Tor biotypes of Vibrio cholerae O1. Infect
Immun 2006;74:36333642.
107 Tamayo R, Pratt JT, Camilli A: Roles of cyclic diguanylate in the regulation of bacterial pathogenesis.
Annu Rev Microbiol 2007;61:131148.
108 Hammer BK, Bassler BL: Distinct sensory pathways
in Vibrio cholerae El Tor and classical biotypes modulate c-di-GMP levels to control biofilm formation.
J Bacteriol 2009;191:169177.
109 Solano C, Garcia B, Valle J, Berasain C, Ghigo JM,
Gamazo C, Lasa I: Genetic analysis of Salmonella
enteritidis biofilm formation: critical role of cellulose. Mol Microbiol 2002;43:793808.

110 Waters CM, Lu W, Rabinowitz JD, Bassler BL:


Quorum sensing controls biofilm formation in
Vibrio cholerae through modulation of cyclic diGMP levels and repression of vpsT. J Bacteriol
2008;190:25272536.
111 Andrade MO, Alegria MC, Guzzo CR, Docena C,
Rosa MC, Ramos CH, Farah CS: The HD-GYP
domain of RpfG mediates a direct linkage between
the Rpf quorum-sensing pathway and a subset of
diguanylate cyclase proteins in the phytopathogen
Xanthomonas axonopodis pv citri. Mol Microbiol
2006;62:537551.
112 Weinhouse H, Sapir S, Amikam D, Shilo Y, Volman
G, Ohana P, Benziman M: c-di-GMP-binding protein, a new factor regulating cellulose synthesis in
Acetobacter xylinum. FEBS Lett 1997;416:207211.
113 Simm R, Morr M, Remminghorst U, Andersson M,
Rmling U: Quantitative determination of cyclicdi-GMP concentrations in nucleotide extracts of
bacteria by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Anal
Biochem 2009;386:5358.
114 Stulke J, Hillen W: Coupling physiology and gene
regulation in bacteria: the phosphotransferase sugar
uptake system delivers the signals. Naturwissenschaften 1998;85:583592.
115 Imamura R, Yamanaka K, Ogura T, Hiraga S, Fujita
N, Ishihama A, Niki H: Identification of the cpdA
gene encoding cyclic 3,5-adenosine monophosphate phosphodiesterase in Escherichia coli. J Biol
Chem 1996;271:2542325429.
116 Rmling U, Bian Z, Hammar M, Sierralta WD,
Normark S: Curli fibers are highly conserved
between Salmonella typhimurium and Escherichia
coli with respect to operon structure and regulation.
J Bacteriol 1998;180:722731.

Ute Rmling
Department of Microbiology, Tumor and Cell Biology
Karolinska Institutet, FE280, SE171 77 Stockholm (Sweden)
Tel. +46 8 524 87319, Fax +46 8 330744
E-Mail ute.romling@ki.se

Prevailing Concepts of c-di-GMP Signaling

181

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 182193

Magnetosomes and Magneto-Aerotaxis


Richard B. Frankela Dennis A. Bazylinskib
a

Department of Physics, California Polytechnic State University, San Luis Obispo, Calif. and bSchool of Life
Sciences, University of Nevada, Las Vegas, Nev., USA

Abstract
Magnetotactic bacteria orient and migrate along geomagnetic field lines. Magneto-aerotaxis
increases the efficiency of respiring microaerophilic cells to efficiently find and maintain a position at
a preferred microaerobic oxygen concentration. Magneto-aerotaxis could also facilitate access to
regions of higher nutrient and electron acceptor concentration via periodic excursions above and
Copyright 2009 S. Karger AG, Basel
below the preferred oxygen concentration level.

General Features of Magnetotactic Bacteria

The magnetotactic bacteria are a morphologically and metabolically diverse group


of freshwater and marine prokaryotes that align and migrate along geomagnetic field
lines, a phenomenon called magnetotaxis [1]. They generally inhabit water columns
or sediments with vertical chemical concentration stratification, where they occur
predominantly at the oxic-anoxic interface (OAI) and the anoxic regions of the habitat
or both [2]. Phylogenetically, all known magnetotactic bacteria belong to the domain
Bacteria and are associated with different subgroups of the Proteobacteria and the
Nitrospira phyla [2, 3].
There are relatively few axenic cultures of magnetotactic bacteria [4]. The cultured
strains include Magnetospirillum magnetotacticum strain MS-1, M. gryphyswaldense
strain MSR-1, M. magneticum strain AMB-1, a marine vibrio strain MV-1 (the name
Magnetovibrio blakemorei will be proposed), a marine coccus strain MC-1 (the name
Magnetococcus marinus will be proposed), Desulfovibrio magneticus strain RS-1, and
a few uncharacterized marine strains. All the cultured organisms, except D. magneticus, are facultatively anaerobic or obligate microaerophiles. All are chemoorganoheterotrophic although the marine strains can also grow chemolithoautotrophically [5].
The genomes of several strains, including M. magnetotacticum strain MS-1, M. magneticum strain AMB-1 [6] and strain MC-1 have been partially or completely sequenced.

All studied magnetotactic bacteria are motile by means of flagella and have a cell
wall structure characteristic of Gram-negative bacteria [4]. The arrangement of flagella varies between species/strains and can be either polarly monotrichous, bipolar, or in tufts (lophotrichous). Like other flagellated bacteria, magnetotactic bacteria
propel themselves through the water by rotating their helical flagella. Reported swimming speeds vary between species/strains, from ca. 40 to 1,000 m/s. In general, the
magnetotactic spirilla are at the slower end (<100 m/s) [7] and the magnetotactic
cocci are at the faster end of the range (>100 m/s) [8].
Magnetotaxis involves passive orientation and active migration along the ambient magnetic field [4]. Killed cells in suspension also orient along the field but do
not migrate. While many magnetotactic bacteria migrate persistently in one direction
relative to the field under oxic conditions, they are able to reverse direction without
turning around under anoxic conditions [9]. Other bacteria, particularly magnetotactic spirilla, migrate in both directions along the field with occasional spontaneous
reversals of the swimming direction without turning around under both oxic and
anoxic conditions [9, 10].

Magnetosomes

Orientation of magnetotactic bacteria in a magnetic field is due to the presence of


magnetosomes, intracellular structures comprising magnetic iron mineral crystals
enveloped by a phospholipid bilayer membrane [11]. The magnetosome membrane
originates as an invagination of the cytoplasmic membrane and appears to control
the nucleation and growth of the mineral crystal at particular locations in the cell [12,
13]. Whether the invagination pinches off to become a true membrane vesicle within
the cell is under debate [13]. Regardless, magnetosomes are almost always adjacent to
the cell membrane where they appeared to be anchored.
The magnetosome mineral crystals consist either of magnetite, Fe3O4, or greigite,
Fe3S4 [4]. These crystals are typically of order 35120 nm in length, which is within
the permanent single-magnetic-domain (SD) size range for both minerals, although
magnetite crystals with lengths up to 250 nm are known [14]. In the majority of magnetotactic bacteria, the magnetosomes are organized in one or more straight chains of
various lengths parallel to the axis of motility of the cell (fig. 1). Non-chain configurations of magnetosomes occur in some species, usually at the side of the cell where
the flagella are inserted. Recent progress in elucidating the biomineralization process
and the construction of the magnetosome chain in magnetotactic bacteria has been
reviewed [12, 13].
All known freshwater magnetotactic bacteria and some marine, estuarine and salt
marsh strains have magnetite magnetosomes. Other strains in the latter habitats contain greigite magnetosomes. These include an unusual multicellular bacterium and a
variety of relatively large, rod-shaped bacteria [3]. The magnetosome greigite crystals

Magnetosomes and Magneto-Aerotaxis

183

2 m

Fig. 1. Transmission electron micrographs: (a) the marine coccus strain MC-1 and (b) M. magnetotacticum strain MS-1. Both organisms have a single chain of magnetite (Fe3O4) magnetosomes (white
arrows). Strain MC-1 has two tufts of flagella, whereas strain MS-1 has two polar flagella. The size bar
at the bottom is for both a and b.

are thought to form from non-magnetic iron-sulfide precursors [15]. Some magnetotactic bacteria, including some forms of the multicellular organism described above,
contain magnetite and greigite magnetosomes, co-organized within the same magnetosome chains but with distinct morphologies for each mineral [4, 16]. Interestingly,
all the magnetite magnetosomes are tooth- or bullet-shaped.
In Magnetospirillum species and strain MC-1, the genes for the proteins implicated in magnetite magnetosome biosynthesis are located within a genomic island.
In Magnetospirillum gryphiswaldense the magnetosome genes are located within a
hypervariable 130-kb stretch of the genome within the magnetosome island [17,
18]. The genes encoding for the proteins MamA, MamB, MamJ and MamK are
located on the mamAB operon, while the genes for MamC and MamD are located
on the mamDC operon. Functions for magnetosome-membrane-associated proteins have been determined for MamJ and MamK. MamJ was demonstrated to be
essential for the assembly of magnetosome chains in M. gryphiswaldense, probably
through interaction with MamK [12], and MamK is involved in the formation of a
network of actin-like filaments that comprise the magnetosomal cytoskeleton and
is responsible for the linear chain-like alignment of magnetosomes within the cell
[13].

184

Frankel Bazylinski

Cellular Magnetic Dipole

Magnetosomes within the SD size range are uniformly magnetized with the maximum magnetic dipole moment per unit volume. Magnetite crystals larger than SD
size are non-uniformly magnetized because of the formation of domain walls or
other spin structures that significantly reduce the magnetic moments of the crystals.
Crystals with lengths below ca. 35 nm are superparamagnetic. Although superparamagnetic particles are SD, thermally-induced reversals of their magnetic moments
result in a time-averaged magnetic moment of zero. Thus by controlling particle size,
magnetotactic bacteria optimize the magnetic dipole moment per magnetosome.
For magnetosomes arranged in a chain, as in M. magnetotacticum, magnetostatic
interactions between the SD crystals cause the magnetic moments to spontaneously
orient parallel to each other along the chain direction [4]. This results in a permanent magnetic dipole for the entire chain with a magnetization approaching its saturation value (0.6 T). Since the chain of magnetosomes is fixed within the cell, the
entire cell is oriented in the magnetic field by the torque exerted on the magnetic
dipole, causing the cell to migrate along the magnetic field as it swims. The permanent magnetic structure of magnetosome chains has been demonstrated by electron
holography [19].
Reported and estimated permanent magnetic moments of several organisms are
ca. 1.0E15 Am2 [19]. The corresponding magnetic energy in the geomagnetic field
of 50 T is 5.0E20 J. This value is greater than thermal energy at room temperature,
4.1E21 J. The average orientation of a cell along the magnetic field as it swims is
determined by the ratio of magnetic to thermal energy; for a ratio of 10, the average
projection of the magnetic dipole on the magnetic field, <cos > = 0.9. This means
the cell can migrate along the field at 90% of its forward speed. Thus a magnetotactic
bacterium is, in effect, a self-propelled magnetic compass needle.

Magnetotaxis

In the original model of magnetotaxis, all magnetotactic bacteria were assumed


to have a polar preference in their swimming direction and were microaerophiles
indigenous to sediments [1]. In this model, the geomagnetic field directs swimming cells downward towards the sediments along the downward-inclined field
lines. Once cells reach their preferred microhabitat, they presumably stop swimming and adhere to sediment particles until conditions changed, e.g., disturbance
of the sediments caused cells to be displaced into the oxic water column. This
explained why magnetotactic bacteria from northern hemisphere sites appeared to
predominantly migrate northward along the geomagnetic field under oxic conditions whereas as those from southern hemisphere sites migrated predominantly
southward [1, 9].

Magnetosomes and Magneto-Aerotaxis

185

Magneto-Aerotaxis

The discovery of large populations of magnetotactic bacteria at the OAI in the water
columns of certain chemically-stratified aquatic habitats, and the isolation of obligately microaerophilic, coccoid, magnetotactic bacteria strains in pure culture, has
led to a revised view of magnetotaxis [9]. The original model did not completely
explain how bacteria in the anoxic zone of a water column benefit from magnetotaxis,
nor did it explain how the magnetotactic cocci such as strain MC-1 form horizontal
microaerophilic bands in semi-solid oxygen gradient media in shake tubes instead
of accumulating and growing at the bottom of the tubes. Bands of strain MC-1 and
M. magnetotacticum in oxygen concentration gradients in thin, flattened capillaries
moved up the capillary, eventually to the meniscus, when the head space gas was
switched from air to pure N2. When the N2 was replaced with air, the bands moved
back to their original position in the capillary. Pure O2 caused the bands to move
further down the capillary. This showed that magnetotactic bacteria have both aerophilic and aerophobic responses and that magnetotaxis and aerotaxis work together
to guide cells to their optimal [O2] (fig. 2). The behavior observed in strain MC-1
and M. magnetotacticum has been denoted magneto-aerotaxis (M-A) [9]. Based on
observation of individual cells in microaerobic bands, two different M-A mechanisms, polar and axial, have been proposed for strain MC-1 and M. magnetotacticum,
respectively [9].

Polar Magneto-Aerotaxis

Polar M-A is a two-state model in which the sense of flagellar rotation in each state
is determined by the [O2] [9]. When [O2] is greater than optimal, ccw rotation
causes cells to migrate along field lines through the optimal [O2] toward lower [O2],
at which point the cell switches to State 2 with cw flagellar rotation, causing the cells
to back up along the field lines toward higher [O2]. It is possible that the cells might
detect redox potential, or internal energy level, rather than molecular oxygen [20].
In experiments with strain MC-1 in an oxygen gradient in a thin capillary, a number
of cells formed an immobilized biofilm at the optimal [O2] while other cells made
straight line passes through the biofilm in both directions [9]. The polar M-A model
accounts for the fact that cells swam away from the biofilm following a reversal of
the magnetic field. In this situation, cells in either state do not encounter the redox
condition that would switch them into the other state and hence do not reverse their
swimming direction. This shows that the orientation of the magnetic field is important in polar M-A.
In some polar strains, short wavelength light (<500 nm) can apparently switch the
state of the cell, causing ccw rotation even under reducing conditions for which the
oxygen concentration is suboptimal [9].

186

Frankel Bazylinski

SH

NH
Bgeo
ccw

ccw

[O2]
O A T Z
[S=]
cw

cw

Fig. 2. Diagram of magnetoaerotaxis. Magnetotactic bacteria in the northern hemisphere (NH)


are oriented in the downward inclined geomagnetic field (Bgeo). Counterclockwise (ccw) flagellar rotation causes the cells to migrate downward, whereas clockwise (cw) rotation causes the
cells to migrate upward. Migration along the magnetic field causes cells to efficiently move
between the oxic water layer above the oxic-anoxic transition zone (OATZ = oxic-anoxic interface (OAI)) and the anoxic water layer below. In the southern hemisphere (SH) the geomagnetic field is inclined upward and the cellular magnetic dipole has reversed orientation in the
cells.

Axial Magneto-Aerotaxis

In axial M-A, cells migrate along magnetic field line by means of a temporal sensory mechanism that occurs in many non-magnetotactic, chemotactic bacteria
[21]. Cells sample the oxygen concentration as they swim and compare the present
concentration with that in the recent past. The change in oxygen concentration
with time is connected to the probability of switching the sense of flagellar rotation
(cw or ccw) and hence the direction of migration along the field [9]. Axial magneto-aerotactic cells moving away from the optimal oxygen concentration toward
higher or lower oxygen concentration have an increased probability of reversing
the sense of flagellar rotation, and hence the direction of migration, causing them
to return to the band. Cells moving toward the optimum oxygen concentration
have a decreased probability of reversing the sense of flagellar rotation. M. magnetotacticum forms microaerobic bands in which the cells are in constant motion in
the band. Since the cells use the magnetic field to provide an axis but not a direction of motility, the orientation of the field is not important, as shown by the fact
that while the cells rotate following a magnetic field reversal, the band remains
intact.

Magnetosomes and Magneto-Aerotaxis

187

Redoxtaxis

The polar M-A model can be extended to a more complex redoxtaxis in habitats in
which rapid chemical oxidation of reduced chemical species such as sulfur near the
OAI results in separated pools of reductants and oxidants [3]. For some magnetotactic
bacteria, it might be necessary to perform excursions to anoxic zones of their habitat
in order to accumulate reduced sulfur compounds as a source of electrons to support
growth. In this situation, polar magnetotaxis could efficiently guide bacteria, either
downward to accumulate reduced sulfur species or upward to oxidize stored sulfur
with oxygen. The oxidized state would result from the almost complete consumption
of stored sulfur or another electron donor, and the cells would swim down along the
field toward the anoxic zone where they could replenish the depleted stock of electron
donor using nitrate or other compounds as alternative electron acceptor. Eventually,
they would reach a reduced state in which the electron acceptor is depleted. In this
state the cells would swim back up along the field and return to the microoxic zone
where oxygen is available to them as an electron acceptor. The advantage of polar
M-A is that an oxygen concentration gradient is not necessary for efficient orientation in the anoxic zone, thereby enabling a rapid return of the cell along relatively
large distances to the preferred microoxic conditions. A further benefit would be that
cells avoid the waste of energy by constant movement along gradients, but instead can
attach to particles in preferred microniches until they reach an unfavorable internal
redox state that triggers a magnetotactic response either parallel or antiparallel to the
geomagnetic field lines. In any case, greater than optimal concentrations of oxygen
would switch cells immediately to the oxidized state provoking the typical downseeking response of magnetotactic bacteria under oxic conditions.
Cells of strain MC-1, like other, uncultivated, magnetotactic cocci, are small (ca. 1
m diameter) with twin, multiflagellar bundles on one side of the cell. Magnetotactic
cocci have been reported to swim at speeds in excess of 100 m/s (about 100 body
lengths per second) [8]. In [O2] gradients in flat, thin capillaries, cells of strain MC-1
form microaerophilic bands of cells [9]. Some cells within the band make long,
straight traverses through the band whereas others stop swimming and attach to the
walls of the capillary or to each other at the OAI. Cells of magnetotactic cocci thus
appear to alternate between active swimming and sessile behavior.
Cells of strain MC-1 grow chemolithoautotrophically with sulfide and other
reduced sulfur sources as electron donors and molecular oxygen as the terminal electron acceptor [5]. In addition, these cells also fix atmospheric dinitrogen [Bazylinski,
unpubl. data]. This is presumably true for other magnetotactic cocci that inhabit the
OAI in many marine and brackish habitats. However, oxidation of S2 by O2 is autocatalytic, so an inverse [O2]/[S2] double gradient (from the downward diffusion of
O2 from air at the surface and the upward diffusion of S2 from the anaerobic zone
through the action of sulfate-reducing bacteria) will form even without the presence
of bacteria. Therefore, bacteria actually have to compete with molecular O2 for S2.

188

Frankel Bazylinski

Consumption of S2 and O2 by bacteria at the OAI makes the gradients steeper. The
coexistence or overlap region (both O2 and S2 present together) is only a few hundred m deep [21] and has very low (<1 m) concentrations of both O2 and S2. Thus
although cells have to contend with relatively low nutrient concentrations, as well as
diffusion-limited flux of S2 from below and O2 from above into the overlap region,
the OAI in these [O2]/[S2] inverse gradients is on optimal positions for these organisms particularly if the habitat is nitrogen-limited and cells are fixing N2.
Nutrient limitation is a fact of life in many marine habitats, and results in predominantly small, fast swimming cells [22, 23]. Smaller cells require lower levels of
nutrients to grow and they have a higher surface-to-volume ratio (S/V~1/R), which
increases their rate of nutrient uptake relative to their volume. This is especially
advantageous in low nutrient conditions. However, consumption of nutrients results
in a greater local depletion because of diffusion limitation. Cells can solve this problem by swimming and relying on chemotaxis to find areas of locally higher nutrient
concentration. At minimum, cells have to swim fast and straight enough to outrun
nutrient diffusion (about 30 m/s for 1 s). However, small cells lose their heading in
times of the order of milliseconds from buffeting by Brownian motion. One solution
is swimming faster so as to get farther before going off course, which is presumably
the reason why small cells that swim fast are the rule in marine environments [23].
However, faster swimming also burns more cellular energy because the viscous drag
on cells depends on their velocity.
Cells of strain MC-1 and similar marine magnetotactic cocci with bilophotrichous
flagellation are fast swimmers yet have their magnetic dipole to keep their heading.
As noted above, fast swimming perhaps allows them to make traverses from one side
of the overlap region to the other to sequentially access higher concentrations of S2
and O2. However, small cells such as the cocci have low carrying capacity so they have
to make shorter, more frequent, traversals than larger cells. In this case, the horizontal chemical stratification could guarantee a payoff that would cover the cost of fast
swimming. Then why do cells of strain MC-1 sometimes stop swimming, as seen
in the bands in the flat capillaries? The answer might involve nitrogen fixation, an
energy-demanding process that only occurs at O2 concentrations less than about 5 m
[24]. Since the O2 concentration in the OAI is even less, cells can fix N2 there. If a cell
is fixing N2, its energy balance might improve if it stops swimming altogether.
Cells of M. magnetotacticum, like all other magnetospirilla, have a single flagellum
at both poles of the cell and swim at about 40 m/s, forwards and backwards with
equal facility. Cultivated cells grow heterotrophically on certain organic acids (e.g.,
succinic acid) as an electron source with O2 or nitrate as the terminal electron acceptor [4]. When O2 is the only electron acceptor available in [O2] gradients, cells form
microaerophilic bands, seeking a preferred O2 concentration that presumably maximizes the proton motive force generated by transfer of electrons [20, 24]. Cells can
be seen to be in constant motion making straight-line excursions above and below
the band. However, because there is no autocatalytic oxidation of electron donor by

Magnetosomes and Magneto-Aerotaxis

189

acceptor, access to nutrients is mostly limited by the diffusion of O2 and electron


source and consumption by the cells. In this situation, cells need only outrun diffusion in order to access increased concentrations of electron donor and acceptor below
and above the preferred O2 concentration, respectively. There is no need to incur the
cost of faster swimming because the cellular magnetic dipole allows cells to maintain
their heading, minimizing the straight run time for temporal chemotaxis.
Cells of the magnetospirilla, like those of strain MC-1, also fix N2, but since they
do not expend as much energy swimming as does strain MC-1, they likely do not
need to stop swimming to conserve energy for N2 fixation. It should be noted that the
situation for magnetospirilla in natural environments might be more complex than
that for the magnetotactic cocci. The fact that cells of magnetotactic spirilla freshly
collected from natural environments or newly isolated in pure culture often display
polar magnetotaxis in the hanging drop assay might indicate this. Many of the magnetotactic cocci collected from natural environments contain sulfur-rich globules
indicating that they are actively oxidizing S2 at the OAI. Many of the cultivated magnetospirilla possess genes encoding for CbbM, a type II ribulose-1,5-bisphosphate
carboxylase/oxygenase, a key enzyme of the Calvin-Benson-Bassham cycle for autotrophy. Thus the magnetospirilla might be able to grow chemolithoautotrophically
like strain MC-1 and may also use inorganic electron donors as well as organic ones.

Deviations from the Magneto-Aerotaxis Models

Polar M-A has been observed in some of the freshwater spirilla [D. Schler, pers.
commun.], bacteria that are nominally axial magneto-aerotactic. Magnetic polarity
was most pronounced in strains that were freshly isolated but gradually converted to
axial M-A upon repeated subcultivation. Polar M-A has also been observed in cells
of M. gryphiswaldense [D. Schler, pers. commun.] and M. magnetotacticum [D.A.
Bazylinski, unpubl. data] grown in semi-solid [O2]-gradient medium and placed in
highly reduced medium under the microscope. However, these experiments were not
entirely reproducible and thus the trigger that causes cells to switch between axial and
polar M-A is not known. Since the difference between axial and M-A at the molecular level is not known, it is possible that the two models represent the endpoints of a
continuum of responses.
The predominance of freshwater, southward swimming (SS), magnetotactic cocci
in a pond in the northern hemisphere was reported by Cox et al. [8] without discussion. Simmons et al. [25] recently observed a population of uncultured, marine magnetotactic bacteria, collected from the anoxic zone of a coastal pond in the northern
hemisphere, that were primarily SS under oxic conditions. Other, polar magnetotactic,
bacteria in the sample were generally northward swimming (NS) as expected although
on occasion the ratio of SS to NS cells was >0.1. Since the SS cells were not identified,
it is not clear whether they are microaerophiles, leaving open the possibility that they

190

Frankel Bazylinski

use the magnetic field to find a preferred position in a vertical concentration gradient
of a molecule or ion other than O2 or at a specific oxidation-reduction potential. If the
organism turns out to be microaerophilic, then the SS response is difficult to understand on the basis of the M-A models. However, since the cells do not migrate up to
the surface of the pond, something must cause them to reverse direction and swim
downward in the water column. Alternatively, they may not be actively swimming but
attached to suspended particulates. The solution to this intriguing mystery will require
examination of the motility of the unusual cells in an oxygen concentration gradient.

Bacterial Hemerythrins and Magneto-Aerotaxis

The genomes of M. magnetotacticum, M. magneticum and strain MC-1 each contain


approximately 30 or more open reading frames (ORFs-possible genes) that encode for
putative proteins with hemerythrin-like domains. None of these proteins have been
characterized, however. Given that magnetotactic bacteria occur predominantly at the
OAI and/or anoxic regions of the water column, O2-binding proteins such as hemerythrins may serve as a sensory mechanism for O2, and thus play a key role in M-A.
Hemerythrin domains contain a sequence motif that includes five histidine residues
and two carboxylate ligands that coordinate two iron atoms; reversible O2-binding
occurs at the di-iron site. For magnetotactic bacteria, some ORFs that encode for putative hemerythrin-like proteins are located within the magnetosome membrane protein
gene islands in strain MC-1 (Mmc1DRAFT_1515 from draft genome) and M. gryphiswaldense (ORF12, ORF13) [15, 17]. Other putative multi-domain proteins from
other magnetotactic bacteria also include hemerythrin domains associated with signal
transduction domains (e.g., histidine kinases, methyl-accepting chemotaxis proteins).
For magnetotactic bacteria migrating within and through the OAI, hemerythrins
may serve to bind O2 when the cell is exposed to elevated O2 concentrations, and then
release the O2 when the cell descends into anoxic conditions. Multi-domain proteins
with both signal transduction and hemerythrin domains suggests a role for these proteins in O2 sensing. Even single-domain hemerythrins may serve a sensory function,
if they are co-transcribed and/or acting with signal transduction proteins. Given the
prevalence of hemerythrin-like ORFs in the known genomes of magnetotactic bacteria, including those within the magnetosome protein gene island [18], hemerythrins
may play a role in M-A (including directing flagellar rotation).
The genomes of M. magnetotacticum, M. magneticum strain AMB-1 and strain
MC-1 also show numerous ORFs that encode for putative proteins with PAS domains,
providing many potential candidate genes for aero-, redox-, and (perhaps) phototaxis
in these bacteria. In bacteria, PAS domains are responsible for sensing stimuli such
as [O2], redox potential, and light. For example, the aerotaxis receptor (Aer) responds
to oxygen concentration in the environment, and is the first step in the intracellular pathway that governs the sense of flagellar rotation in Escherichia coli [26]. As

Magnetosomes and Magneto-Aerotaxis

191

mentioned above, the polar M-A coccus, strain MC-1, displays a negative phototaxis
in response to short-wavelength light, but the mechanism is unknown. It is difficult
to infer the precise identity of the stimulus that the PAS-containing protein is sensitive to based on amino acid sequence alone. This is also the case for numerous ORFs
that encode putative methyl-accepting chemotaxis proteins in M. magnetotacticum,
M. magneticum, and strain MC-1, including putative hemerythrins.

Conclusion

Magnetotactic bacteria have solved the problem of constructing an internal, permanent, magnetic dipole that is sufficiently robust so that a cell will be oriented along
the geomagnetic field as it swims, yet be no longer than the length of the cell itself (ca.
12 m). The solution, the magnetosome chain, is very elegant and efficient in that
it makes maximum use of a small amount of magnetite, assuming that cells want to
maximize the ratio of magnetic moment to volume of magnetite to keep metabolic
cost to a minimum. However, there are many microaerophilic organisms that form
aerotactic bands without the aid of magnetism, including, for example, non-magnetic
mutants of magnetotactic bacteria. Simulations of axial magnetotactic bacteria confirm the fact that M-A is more efficient than aerotaxis alone for finding the optimal
[O2], meaning magnetotactic bacteria would find the optimal concentration before
non-magnetic aerotactic bacteria with the same swimming speed, but only at high
inclinations of the geomagnetic field. Many polar magnetotactic bacteria are fast
swimmers, ca. 100 body lengths per second or more, so the efficiency argument may
hold over a greater range of geomagnetic inclination for these organisms. Nevertheless,
the question of whether aerotactic efficiency alone is sufficient to account for the persistence of magnetotaxis in bacteria over geologic time scales is still open.

Acknowledgements
We thank S. Schbbe and T. J. Williams for discussions. D.A.B. was supported by US National
Science Foundation Grant EAR-0311950. We dedicate this paper to Dott. Salvatore Bellini, a pioneer in the study of magnetically-sensitive bacteria and protists.

References
1 Blakemore RP: Magnetotactic bacteria. Annu Rev
Microbiol 1982;36:217238.
2 Simmons SL, Sievert SM, Frankel RB, Bazylinski
DA, Edwards KJ: Spatiotemporal distribution of
marine magnetotactic bacteria in a seasonally stratified coastal pond. Appl Environ Microbiol 2004;
70:62306239.

192

3 Spring S, Bazylinski DA: Magnetotactic bacteria; in


The Prokaryotes. Published on the web at http://
www.springer-ny.com/. New York, Springer, 2000.
4 Bazylinski, DA, Frankel RB: Magnetosome formation in prokaryotes. Nat Rev Microbiol 2004;2:217
230.

Frankel Bazylinski

5 Williams TJ, Zhang CL, Scott JH, Bazylinski DA:


Evidence for autotrophy via the reverse tricarboxylic acid cycle in the marine magnetotactic coccus
strain MC-1. Appl Environ Microbiol 2006;72:1322
1329.
6 Matsunaga T, Okamura Y, Fukuda Y, Wahyudi AT,
Murase Y, Takeyama H: Complete genome sequence
of the facultative anaerobic magnetotactic bacterium Magnetospirillum sp. strain AMB-1. DNA Res
2005;12:157166.
7 Maratea D, Blakemore RP: Aquaspirillum magnetotacticum sp. nov., a magnetic spirillum. Int J Syst
Bacteriol 1981;31:452455.
8 Cox BL, Popa R, Bazylinski DA, Lanoil B, Douglas
S, Belz A, Engler DL, Nealson, KH: Organization
and elemental analysis of P-, S-, and Fe-rich inclusions in a population of freshwater magnetococci.
Geomicrobiol J 2002;19:387406.
9 Frankel RB, Bazylinski DA, Johnson, M, Taylor BL:
Magneto-aerotaxis in marine, coccoid bacteria.
Biophys J 1997;73:9941000.
10 Spormann AM, Wolfe RS: Chemotactic, magnetotactic, and tactile behavior in a magnetic spirillum.
FEMS Microbiol Lett 1984;22:171177.
11 Gorby YA, Beveridge TJ, Blakemore RP:
Characterization of the bacterial magnetosome
membrane. J Bacteriol 1988;170:834841.
12 Schler D: Magnetoreception and Magnetosomes
in Bacteria. Berlin, Springer, 2007, pp 1319.
13 Komeili A: Molecular mechanisms of magnetosome
formation. Annu Rev Biochem 2007;76:351366.
14 Lins U, McCartney MR, Farina M, Buseck PR, Frankel
RB: Crystal habits and magnetic microstructures of
magnetosomes in coccoid magnetotactic bacteria.
Appl Environ Microbiol 2005;71:49024905.
15 Psfai M, Buseck PR, Bazylinski DA, Frankel RB:
Reaction sequence of iron sulfides in bacteria and
their use as biomarkers. Science 1998;280:880883.
16 Lins U, Keim CN, Evans FF, Farina M, Buseck PR:
Magnetite (Fe3O4) and greigite (Fe3S4) crystals in multicellular magnetotactic prokaryotes. Geomicrobiol J
2007;24:4350.

17 Schbbe S, Kube M, Scheffel A, Wawer C, Heyen U,


Meyerdierks A, Madkour MH, Mayer F, Reinhardt
R, Schler D: Characterization of a spontaneous
nonmagnetic mutant of Magnetospirillum gryphiswaldense reveals a large deletion comprising a
putative magnetosome island. J Bacteriol 2003;185:
57795790.
18 Ullrich S, Kube M, Schbbe S, Reinhardt R, Schler
D: A hypervariable 130-kilobase genomic region of
Magnetospirillum gryphiswaldense comprises a magnetosome island which undergoes frequent rearrangements during stationary growth. J Bacteriol
2005;187:71767184.
19 Dunin-Borkowski RE, McCartney MR, Frankel RB,
Bazylinski DA, Posfai M, Buseck PR: Magnetic
microstructure of magnetotactic bacteria by electron holography. Science 1998;282:18681870.
20 Alexandre G, Greer-Phillips S, Zhulin IB: Ecological
role of energy taxis in microorganisms. FEMS
Microbiol Rev 2004;28:113126.
21 Taylor BL, Zhulin IB, Johnson MS: Aerotaxis and
other energy-sensing behavior in bacteria. Annu
Rev Microbiol 1999;53:103128.
22 Schultz HN, Jorgensen BB: Big bacteria. Annu Rev
Microbiol 2001;55:105137.
23 Mitchell JG: The influence of cell size on marine
bacterial motility and energetics. Microb Ecol 1991;
22:227238.
24 Zhulin IB, Bespalov VA, Johnson MS, Taylor BL:
Oxygen taxis and proton motive force in Azospirillum
brasilense. J Bacteriol 1996;178:51995204.
25 Simmons SL, Bazylinski DA, Edwards KJ: Southseeking magnetotactic bacteria in the northern
hemisphere. Science 2006;311:371374.
26 Watts KJ, Johnson MS, Taylor BL: Minimal requirements for oxygen sensing by the aerotaxis receptor
Aer. Mol Microbiol 2006;59:13171326.

Richard B. Frankel
Department of Physics, California Polytechnic State University
San Luis Obispo, CA 93407 (USA)
Tel. +1 805 756 2795, Fax +1 805 756 2435
E-Mail rfrankel@calpoly.edu

Magnetosomes and Magneto-Aerotaxis

193

Collin M, Schuch R (eds): Bacterial Sensing and Signaling.


Contrib Microbiol. Basel, Karger, 2009, vol 16, pp 194225

Engineering Bacterial Signals and Sensors


Howard Salisa Alvin Tamsirb Christopher Voigta
a
Department of Pharmaceutical Chemistry and bBiochemistry, Cell & Developmental Biology, and
Genetics (Tetrad) Graduate Program, University of California, San Francisco, Calif., USA

Abstract
In the emerging field of synthetic biology, a central goal is to reliably engineer bacteria to respond to
environmental signals according to a pre-determined genetic program. The sensor systems and
genetic circuitry inside bacteria are the eyes and brain of a new class of biotechnological applications in which bacteria are used as living, self-replicating computers that can beneficially interact
with the physical world. These engineered gene networks are constructed by extracting natural sensor systems and other genetic parts from multiple organisms and recombining them into novel configurations. This chapter is a how-to guide. It describes several strategies for engineering new
bacterial sensor systems and synthetic gene networks that are capable of sensing a desired stimulus
and generating interesting dynamical or pattern-forming responses. We also provide specification
sheets describing many two-component and quorum-sensing systems, focusing on the information
that one needs to know in order to use them for engineering applications.
Copyright 2009 S. Karger AG, Basel

Introduction

Bacteria have evolved a multitude of chemical mechanisms to detect changes in their


natural environment [1], adapt their metabolism in response to scarcity or excess of
resources [2], and alter their possibly pathogenic lifestyle according to the presence of a
host [3]. Sensors and signals exist because they provide each bacterium, or a population
of bacteria, with a competitive advantage over its opponents in their ecosystem. Because
bacteria are found in almost every environment on Earth, the diversity of sensors and
signals that enable their survival across such different conditions is truly astounding.
Modern biotechnology has co-opted bacteria and other microbes to perform useful
work for human society [4]. Many every-day chemicals are produced by genetically
engineered microbial sources, including nutritional supplements [5] and the plastics
found in consumer products [6]. However, modern biotechnology has not taken full
advantage of the ability of bacteria to sense and respond to their environment. Such
sensors may improve product yields by regulating metabolism according to changing
oxygen and substrate levels.

In the emerging field of synthetic biology, a central goal is to reliably engineer


bacteria to respond to environmental signals according to a genetic program [7]. The
development of new technologies including large-scale DNA synthesis and sequencing allow us to decode the genomes of intractable microbes, extract useful sensor
systems, and redesign them for use in other hosts, such as Escherichia coli. The existing diversity of sensor systems, generated by the history of evolution, is now accessible and provides us with a wealth of genetic components. By rewiring these genetic
components together into new combinations, we create novel sensor systems that
respond to environmental signals that are important to industrial or medical applications. These new sensor systems are the eyes in a new class of biotechnological
applications in which bacteria are used as living, self-replicating computers that can
beneficially interact with the physical world [8]. As we engineer new types of bacteria,
we are continuing to contribute towards the diversity of life, except that we are limited
only by our imaginations whereas evolution is restricted to conferring fitness.
In this chapter, we discuss several successful examples of engineering bacterial
sensors and signals and formulate strategies for engineering future systems. We start
with a brief overview of bacterial sensor systems. In the next section, Specifications
of Well-Characterized Bacterial Sensors and Signals, we focus on the characteristics of two-component and quorum-sensing systems. The important characteristics
of these systems are compiled into a condensed specification sheet, a style commonly found in engineering handbooks. In the following section, Engineering New
Bacterial Sensors, we present both directed evolution and semi-rational strategies for
creating new bacterial sensors and illustrate with successful examples. In the final
section, Engineering Synthetic Gene Networks with Rewired Sensors, we describe
how the components of existing sensor systems can be rewired to generate new
responses to existing signals. We conclude with a look towards the future.

An Overview of Bacterial Sensors


Bacteria sense and respond to a broad range of stimuli. These include temperature
or pH changes, nutritional starvation, stresses to the outer or inner membranes, new
food sources, toxins, and quorum signals. The mechanisms for sensing these environmental changes are as varied as the stimuli, but there are some archetypal systems that
may be summarized.

Allosteric Transcription Factors


Allosteric transcription factors represent the broadest and most commonly studied
type of bacterial sensor. When bound by cytoplasmic small molecules, these proteins will undergo an induced conformational change that alters their DNA-binding

Engineering Bacterial Sensors

195

specificity, enabling the targeted regulation of gene transcription. Allosteric transcription factors often regulate the expression of transporters and enzymes that import
and break down specific nutritional sources. For example, the LacI, MalT, and AraC
transcription factors respectively regulate the lactose, maltose, and arabinose disaccharide utilization operons [9]. Similarly, inducible proteins can also respond to toxins, such as antibiotics, by activating the expression of efflux pumps or other proteins
that nullify the toxins effects.
Further, some allosteric transcription factors are global regulators of gene expression that enact large shifts in cell metabolism. The transcription factors Crp and Fnr
[10] sense ATP availability and respiratory potential by respectively binding to cyclic
AMP and molecular oxygen. When cyclic AMP accumulates, Crp upregulates over
100 genes involved in carbon utilization and energy production. Likewise, the key
genes responsible for aerobic versus anaerobic metabolism are regulated by Fnr in
response to the presence or absence of oxygen.
Multiple allosteric transcription factors can bind to the same promoter to combinatorially regulate the rate of transcription. The cooperative binding of both types
of allosteric transcription factors global and specific can fine tune the expression
level from these promoters and create logical-like decisions in response to new nutritional sources. For example, the arabinose utilization operon is regulated by both the
AraC and Crp transcription factors, ensuring that its activation only occurs when
both extra energy is needed (at low ATP concentrations) and when arabinose is sufficiently present [9].

Riboswitches
Riboswitches are RNA-based sensors that form three-dimensional structures capable
of switching their conformational shapes from one state to another in response to
intracellular environmental changes [11]. These conformational changes are induced
by the presence of certain small molecules or through variations in temperature and
lead to the modulation of key steps in gene expression, including transcriptional
termination, translational initiation, and mRNA stability. Because riboswitches can
directly couple the post-transcriptional regulation of a gene to the concentration of
a key metabolite, they offer a rapid mechanism for adapting cellular metabolism to
changing intracellular conditions.
For example, a riboswitch located in the ydhL mRNA of Bacillus subtilis responds
to the absence of adenine by causing the formation of an early transcriptional terminator, shutting down the production of an adenine efflux pump [12, 13]. A short
region of the mRNA called an aptamer binds to adenine molecules and forms a
tertiary structure. This structure is mutually exclusive with the formation of a long
RNA hairpin that is responsible for terminating transcription. When adenine is no
longer present, newly transcribed mRNA transcripts no longer form the tertiary

196

Salis Tamsir Voigt

structure, allowing early transcriptional termination to occur. Similarly, a riboswitch


in the thiM mRNA of E. coli responds to thiamine pyrophosphate by repressing
the translation initiation of a kinase involved in the thiamine biosynthetic pathway
[14].
Riboswitches also play an important role as temperature sensors in many physiological processes, including virulence induction as well as heat and cold shock [15].
The mRNAs of the heat-shock 32 factor [16] and the CspA cold-shock regulator [17]
both contain an RNA thermosensor that regulates translation initiation according to
temperature changes.

Two-Component Systems
Bacteria need a rapid mechanism for detecting and adapting to environmental
changes that occur outside the cell. Lacking a transmembrane component, both allosteric transcription factors and riboswitches are limited to sensing only intracellular
conditions. A two-component system (TCS) is an evolutionary conserved solution
to this problem. It is a transmembrane sensor kinase that biochemically transduces
a signal to a cytoplasmic response regulator, which binds to DNA to regulate transcriptional initiation. The sensor kinase is responsible for detecting changes in one
or more environmental stimuli located in the periplasm and responding by phosphorylating its cognate response regulator. The response regulator behaves similarly to
an allosteric transcription factor; once phosphorylated, it increases its DNA-binding
specificity and regulates the rate of transcriptional initiation.
In a TCS, the link between the stimulus and response is determined by specific
protein-protein interactions. These interactions permanently wire together the inputoutput response of the cell to changing environmental cues, activating or repressing
gene expression according to a single stimulus. Multiple response regulators can bind
to a single promoter, creating complex genetic programs that use multiple stimuli as
inputs. A good example is the FocA formate transporter. It is activated by the ArcAB
TCS under anaerobic conditions and repressed by the NarXL TCS in response to
high nitrate availability [18, 19]. The focA promoter also binds to the Crp and Fnr
transcription factors to integrate a total of four oxygen and energy-related signals as
inputs into its genetic program.
Even though the input-output relationship is hard-wired together, the modular architecture of TCSs makes them particularly prone to evolutionary adaptation.
Relatively few mutations to the sensor domain of the kinase can result in the sensing
of a new environmental stimulus. Likewise, mutations to the response regulator or the
sensor kinases phosphorylation activity can alter the genetic response to an existing
stimulus. Specific examples of such mutations will be described below: Engineering
New Bacterial Sensors. This facile adaptation enables bacteria to quickly alter their
sensing capabilities when entering new environments.

Engineering Bacterial Sensors

197

Quorum Sensors
A quorum sensor is a sensor system that detects and responds to a diffusible molecule
that is produced by a population of organisms, frequently from multiple species. The
concentration of the diffusible molecule often called an autoinducer dynamically
changes according to multiple factors, including the production rate of the autoinducer, the number of organisms producing the autoinducer, and the volume of the
enclosed space [20]. As the number of autoinducer-producing organisms increases,
the concomitant increase in autoinducer concentration can activate or repress gene
expression. Many physiological behaviors, such as virulence and biofilm formation,
only become effective at high or low population densities and are regulated accordingly by quorum-sensing systems.
While there is no single architecture for a quorum-sensing system, there are conserved families of sensor systems that detect quorum signals [21]. In Gram-negative
bacteria, there is a prototypical architecture that consists of an autoinducer synthetase
that produces a diffusible, membrane-permeable homoserine lactone (HSL) and an
allosteric transcription factor that binds to it. This architecture is named the LuxIRtype after the first such quorum sensor system discovered in the marine bacterium
Vibrio fischeri. In Gram-positive bacteria, the autoinducer is typically a secreted short
peptide that binds to membrane-bound receptors. The sensor kinases of some TCSs
can also detect diffusible small molecules. For example, the QseB sensor kinase in E.
coli binds to epinephrine and regulates FlhCD production [22], which is a transcription factor that controls flagellar biosynthesis.
Interestingly, quorum signals are not directly coupled to a single environmental
stimulus. In an ecosystem of microbes, each species can regulate the production of
a quorum signal in response to different stimuli, leading to a mixing of signals. As
a result, the concentration of autoinducer and its effect on gene regulation depends
on the summation of all signals and in proportion to the organisms sending them. In
addition, bacteria are not required to express both components of a quorum-sensing
system, leading to the ability to selectively sense or manipulate signals in the environment, but not necessarily the obligation to act on them. For example, the E. coli MsqR
transcription factor binds to the universal quorum signal AI-2 (a furanosyl borate
diester) [23]. Finally, autoinducers can also act as cross-kingdom signaling conduits,
enabling bacteria to respond to changes in animal host physiology [24].

Starvation and Stress Sensors


Bacteria quickly adapt to dangerous environments by sensing a variety of stressful
physical and chemical stimuli and regulating their gene expression to compensate.
The sensor architectures described above are often utilized to sense the presence
of toxic stimuli. For example, the TCSs Rst and Cpx detect instabilities in the inner

198

Salis Tamsir Voigt

membrane caused by excess unfolded protein and accordingly trigger the E-mediated
envelope stress response, which activates the expression of a variety of periplasmic
chaperones and proteases [25]. However, specialized sensors are sometimes required
to detect certain stress stimuli, including amino acid starvation, the presence of reactive oxygen species, and DNA damage.
The ribosome-bound RelA kinase is responsible for detecting amino acid starvation [26]. Each time an uncharged tRNA binds to the ribosomal A-site, the RelA
kinase synthesizes one molecule of ppGpp or pppGpp. When a free amino acid
becomes depleted, its corresponding tRNA remains uncharged, causing an accumulation of ppGpp and triggering a genome-wide shift in transcriptional regulation
called the stringent response. By binding to DksA, ppGpp can destabilize the open
complex formation of RNA polymerase at a promoter. Promoters that feature weak
open complex formation are further destabilized, causing transcriptional repression.
Conversely, promoters with strong open complex formation are either unaffected or
activated by the increasing amount of free RNA polymerase. As a result, the entire
cellular metabolism is shifted to maintain a set biomass to protein ratio. The amino
acid biosynthetic pathways are activated while the ribosome, nucleotide, and fatty
acid synthesis pathways are repressed. The SpoT hydrolase also contributes to ppGpp
regulation and responds to carbon, iron, and fatty acid starvation.
Reactive oxygen species, such as singlet oxygen (O2 1g), superoxide (O2), and
hydrogen peroxide (H2O2), initiate massive cellular damage by violently oxidizing
chemical groups or disrupting the iron-sulfur clusters inside enzymes. Bacterial
sensors must rapidly detect reactive oxygen species and trigger the production
of antioxidants and enzymes that convert them to non-toxic forms. In E. coli, the
transcription factor SoxR [27] is induced by superoxide and activates the SoxS
regulon, which includes superoxide dismutase (SodA). Similarly, hydrogen peroxide induces the OxyR factor, activating the expression of the catalase KatG [28]. In
Rhodobacter sphaeroides, a model photosynthetic bacteria, an anti- factor ChrR
senses the presence of singlet oxygen and responds by triggering a general stress
response, which includes the increased production of singlet-quenching carotenoids [29].
Bacteria are constantly exposed to radiation and carcinogens that result in DNA
damage. DNA damage is both detected and repaired by a variety of repair enzymes,
including lyases, exonucleases, polymerases, helicases, and ligases. In E. coli, these
enzymes include the general nucleotide excision repair machinery (UvrABC and others), the photolyase Phr that detects and repairs pyrimidine dimers, the base excision
repair enzyme (Ung) that detects and reverses the deamination of cytosine to uracil, and the methyl-directed mismatch repair pathway (MutHLS). Because errors in
the replication process can expose single-stranded DNA (ssDNA), the last chance to
repair such errors is the SOS response [30, 31], mediated by the ssDNA-binding RecA
protein. In response to ssDNA damage, the SOS regulon is activated with intermittent, timed pulses until the damage is repaired [32].

Engineering Bacterial Sensors

199

Light Sensors
Bacteria detect visible light to regulate photosynthesis, virulence, and metabolism. The
presence of light can signify both a potential energy source and the bacteriums location, by indicating its depth in a marine environment or whether it has invaded a host
body. Light sensors are composed of rhodopsin, BLUF, phytochrome, LOV, or cryptochrome protein domains and are bound either covalently or reversibly to an excitable
chromophore, including retinal, flavins, and tetrapyrryoles (e.g. bacteriochlorophyll)
[33]. The chromophore transitions from the ground to an excited state by absorbing
photons of a specific wavelength and subsequently induces a protein conformational
change, such as a flavin-cysteinyl adduct. The conformational change modulates the
activity of the protein, linking the magnitude of light absorption to an output response.
Light sensors can exhibit a variety of light-dependent output responses, serving as
DNA-binding transcription factors or antagonists to other proteins.
For example, the YtvA light sensor in B. subtilis uses a blue-light sensing LOV
domain to reversibly sequester an anti- factor, upregulating the B stress response
[34]. Similarly, the AppA transmembrane receptor in R. sphaeroides sequesters the
photosynthesis-regulating PpsR transcription factor in the absence of blue light [35].
Light sensors can also play the role of the histidine kinase in a TCS [36] and are ubiquitous in free-living soil or marine bacteria.

Specifications of Well-Characterized Bacterial Sensors and Signals

In this section, we discuss the specifications of TCSs and the LuxIR-type of quorumsensing systems, focusing on ones that regulate gene expression inside E. coli. We
direct our attention to the characteristics that are most relevant to using them in engineering applications. While our focus is on two-component and quorum-sensing systems, both inducible transcription factors and riboswitches [37] are also active targets
for engineering new sensors and signals.

Two-Component Systems
The prototypical TCS is composed of a sensor kinase and a DNA-binding response
regulator. In response to a stimulus, the kinase autophosphorylates itself at a conserved histidine residue and transfers the phosphate group to an aspartic acid residue on the response regulator. Once phosphorylated, the response regulator regulates
transcriptional initiation by binding to one or more DNA operators and interacting
with RNA polymerase. Sensor kinases often bind to the periplasmic membrane as
dimers or larger multimers. The mechanistic path from stimulus to gene regulation
depends on a number of traits of the TCS, including the domain structure of the

200

Salis Tamsir Voigt

sensor kinase, the DNA-binding properties of the response regulator, and the number
and positioning of the DNA operators inside the promoter. By manipulating these
traits, one can create sensor kinases that respond to new signals or promoters that
respond to different response regulators.
In figure 1, we summarize the relevant details of 21 well-characterized TCSs in E.
coli, including the stimulus of each sensor kinase and the effect on its cognate response
regulator, the presence of a HAMP linker domain in the sensor kinase, the activity of
the sensor kinase on its cognate and non-cognate response regulators, a brief list of
E. coli promoters regulated by the response regulator, the number of operator sites
located in these promoters, and the consensus DNA sequence of the response regulators operator site. Next, we briefly discuss the relevance of this information when
engineering TCSs.

The Stimulus
The N-terminal domain of most sensor kinases is responsible for sensing its one or
more stimuli. While ill-conserved and thus unable to be grouped into specific domain
families, recent crystal structures [3840] suggest that the sensing domains of many
sensor kinases are PAS-like (PER, ARNT, SIM) domains [41] with one or more transmembrane segments that loop through the plasma membrane. These sensor domains
possess alternative conformational shapes which are stabilized or destabilized by the
presence or absence of their stimulus in a dose-dependent manner.
For most sensor kinases, the stimulus is a well-defined small molecule, ion, or protein that binds to the sensor domain. For example, the sensor domains of AtoS and
CusS bind directly to acetoacetate and Cu2+ ions, respectively [42, 43]. In other cases,
the stimulus remains unidentified. For example, the stimulus of CreC has been narrowed down to an intermediate in the fermentation or respiration pathways that consume gluconeogenic carbon sources [44]. Other stimuli are an indirect indicator of a
change in the environmental state. The CpxP chaperone mediates the unfolded perisplasmic protein response by binding to and inhibiting the sensor domain of CpxA in
the absence of unfolded protein [45]. Likewise, the cytoplasmic sensor domain of ArcB
indirectly senses the oxygen availability by binding to the reduced forms of quinones
[46]. In other TCSs, the stimuli may be a poorly defined environmental characteristic,
such as the osmolarity stimulus of EnvZ [47] or the stress response of RstB [48], that
may alter the sensing domain via subtle changes in amino acid protonation or electrostatic interactions.
Understanding the sensor kinases mechanism of sensing is important when engineering them for new applications. The stimulus must be well characterized and
externally alterable. Depending on the application, small molecules are ideal stimuli
because their concentration in the growth environment can be carefully controlled
and they are sometimes specific to a single sensor kinase. Conversely, temperature,

Engineering Bacterial Sensors

201

pH, and other environmental variables can simultaneously affect multiple sensor
kinases and other homeostatic regulatory mechanisms.

The HAMP Linker


Once the sensor domain initiates an allosteric change in conformational shape, that
motion must be physically transmitted to the histidine kinase domain to alter its
kinase activity. In about 30% of all sensor kinases, that motion is transmitted by
a conserved HAMP linker domain [49], which consists of two parallel helices and
is typically located C-terminal to the last periplasmic domain. The presence of a
HAMP linker implies that the conformational motion is a rotation along an axis
perpendicular to the membrane [50]. The Pfam [51] and SMART [52] databases use
hidden Markov models and multiple sequence alignments to predict the presence of
the HAMP linker and other protein domains based on evolutionary conservation.
In one example, the HAMP domain of the NarX sensor kinase was completely
replaced by the HAMP domain from CpxA with no significant alteration in its
response to stimulus [53]. However, replacing only a portion of the NarX HAMP
yielded a non-functional sensor kinase, demonstrating that the HAMP domain, as
a whole, is somewhat modular. In other examples, replacing the HAMP linker created a functional sensor kinase only when the HAMP domain adjacent to the sensor
domain was utilized, suggesting that there may be specific, non-modular interactions
between the sensor and HAMP domains that mediate the rotational motion [54]. The
presence or absence of the HAMP linker therefore becomes important when engineering new hybrid sensor kinases.

Kinase Activity and Cross-Talk


The C-terminal end of the prototypical sensor kinase contains a dimerization and
histidine kinase (HK) domain followed by an ATPase domain. In response to stimulus and the transmission of conformational motion from the sensor domain, the histidine kinase will phosphorylate its conserved histidine residue using the activity of
the ATPase domain. Independent of its phosphorylation, the histidine kinase domain
can specifically bind to its cognate response regulator. If the conserved histidine residue is phosphorylated, the phosphate group is transferred to a conserved aspartate
residue on the response regulator and the response regulator unbinds from the histidine kinase domain.
In some sensor kinases, the HisP AspP phosphorylation path can also proceed
in reverse. A non-phosphorylated histidine kinase domain can bind to a phosphorylated response regulator and transfer its aspartate phosphate group back to the
histidine residue on the kinase domain. Sensor kinases that possess both the kinase

202

Salis Tamsir Voigt

Fig. 1. The relevant characteristics of 21 two-component systems (TCS) from E. coli [22, 23, 39, 4244,
47, 48, 6792]. The activity of each sensor kinase (SK) is either bifunctional (B) or kinase only (K), predicted by homology modeling in Alves and Savageau [55]. The presence (Y) or absence (N) of a
HAMP linker in the sensor kinase is noted according to the PFAM database [51]. The stimulus and its
effect on the phosphorylated response regulator (RR) is shown for each SK and RR cognate pair. A
brief list of E. coli promoters containing RR-binding DNA sites is included alongside their DNA
sequence [93]. These sequences are either the consensus or highest affinity RR-binding sites.
Nucleotide abbreviations used: R (A/G), Y (C/T), N (any), W (A/T), M (A/C), K (T/G), D (A/G/T).

Engineering Bacterial Sensors

203

and phosphatase activities are called bifunctional. An interesting study examined the
structural characteristics of known bifunctional sensor kinases and found that they
typically possess an ATP lid similar to the one in EnvZ, a known bifunctional sensor
kinase [55]. Using this property, they used homology modeling to predict whether
a particular sensor kinase is likely to be bifunctional. Bifunctional sensor kinases
behave similarly to a futile enzyme cycle, with the total rate of phosphorylation equal
to the difference between the kinase and phosphatase activities. Under certain conditions, futile enzyme cycles can generate large changes in the phosphorylation level
of substrates in response to small changes in the kinase or phosphatase activity [56],
leading to an ultrasensitive behavior.
In certain special cases, a sensor kinase will possess a longer HisP AspP HisP
AspP phosphorelay pathway, mediated by additional C-terminal domains. The intermediate aspartate residue is located in a response regulator-like domain in the sensor
kinase that is C-terminal to the ATPase domain. It is followed by a histidine phosphotransferase (Hpt) domain that phosphorylates the response regulator. For example, the BarA, TorS, ArcB, and EvgS sensor kinases possess this longer phosphorelay
pathway. The longer phosphorelay pathway may be advantageous when integrating
multiple complex stimuli by increasing the number of phosphotransfer events whose
rates may be individually regulated. These additional phosphotransfer events are ideal
targets for creating sensor kinases with multiple independent stimuli.
The histidine kinase domain can also less specifically bind to non-cognate response
regulators and phosphorylate them, called cross-talk. In vitro studies demonstrate
that many sensor kinases will phosphorylate a non-cognate response regulator [57,
58], leading to the perplexing question of how signals are correctly transduced from
stimulus to response regulator without inadvertent gene regulation by non-cognate
response regulators. One hypothesis is that each sensor kinases activity is only one
component in a much larger network of equally important kinase and phosphatase
activities along with strong substrate competition [59]. These other components
include the phosphatase activity of bifunctional sensor kinases, spontaneous phosphorylation of the response regulator by acetylphosphate, and auto-dephosphorylation of the response regulator. In addition, the in vivo concentrations of sensor kinases
are relatively low, compared to the response regulators, leading to strong binding
competition between cognate and non-cognate response regulators. When combined
together, the network of interactions can maintain a phosphorylated response regulator at a steady concentration in response to a cross-talk stimulus even while retaining
sensitivity to the correct stimulus.

The Response Regulator and Gene Regulation


Response regulators are DNA-binding transcription factors that preferentially bind
to their cognate DNA operators once their aspartate residues are phosphorylated. In

204

Salis Tamsir Voigt

general, there are two mechanisms for increasing the DNA-binding affinity. In the
first case, phosphorylation causes the response regulator to form homodimers or
tetramers, which increases the surface area for potential protein-DNA interactions
between the multimer and DNA operator. Because multimerization of proteins often
occurs with a head-to-head symmetry, these DNA operators will typically be palindromic sequences. In the second case, a phosphorylated response regulator first
binds to its cognate DNA operator and then becomes stabilized by forming attractive
protein-protein interactions with additional adjacently bound copies of the response
regulator. These interactions are typically made in a head-to-tail manner with the
DNA operator sequence consisting of multiple repeats.
Once bound to its DNA operator, the response regulator can regulate transcriptional initiation by either increasing (activation) or decreasing (repression) the rate
of recruitment or conformational change of the Holoenzyme (RNA polymerase and
factor) at the promoter. The extent of activation or repression depends on how
often the response regulator is bound to its operator, the position of the operator
with respect to the RNA polymerase-binding site, and the strength of the attractive or
repulsive protein-protein interactions between the response regulator and RNA polymerase. The binding affinity of the response regulator to its DNA operator can also be
altered by modifying the sequence of the operator.
In the simplest case of a single DNA operator, the location of the operator with
respect to the transcriptional start site (+1) determines how a bound response regulator regulates transcription. In general, when an operator is located between the 35
and +20 nucleotides, bound response regulator will compete with RNA polymerase
binding and repress transcription. Alternatively, operators located between the 65
and 30 nucleotides offer bound regulator the opportunity to recruit Holoenzyme by
forming attractive protein-protein interactions with factor or the -subunits of RNA
polymerase. Therefore, it is the position of the DNA operator and not necessarily
the response regulators characteristics that determine its regulation of transcription.
In fact, many response regulators can either activate or repress transcription based
on the position of their operator closest to the promoter, including ArcA, NarL, and
PhoP in E. coli. When engineering synthetic promoters, one can reposition the operators of these dual-type response regulators to achieve either increased or decreased
transcription in response to stimuli.
When multiple DNA operators surround the promoter region, response regulators
and other transcription factors can interact with one another to generate a more complicated response to stimuli. Overlapping DNA operators cause multiple response
regulators to compete for binding, creating an OR regulatory logic that activates
or represses transcription in the presence of either response regulator. Overlapping
operators may also cause a subtractive logic if one response regulator activates transcription and the other represses it. When two different adjacently bound response
regulators form attractive protein-protein interactions, the regulatory logic is additive; the transcriptional activation or repression by both stimuli is far more than

Engineering Bacterial Sensors

205

when either stimulus is present. Finally, the presence of distant DNA operators can
allow multiple copies of a response regulator to mediate the formation of DNA loops,
enabling transcriptional repression by preventing Holoenzyme binding. The formation of these DNA loops is often assisted by transcription factors that bend or kink
DNA, such as H-NS, IHF, or Lrp. By combining all of these interactions together,
one can engineer a synthetic promoter to increase or decrease its transcription rate
according to the presence or absence of multiple stimuli and in a dose-dependent
manner.

Quorum-Sensing Systems
Many cellular behaviors, including bioluminescence, virulence, and biofilm formation, require a high cell density in the microbial community before becoming effective. A quorum-sensing system can activate or repress gene expression once the cell
density has reached a threshold or when the system volume is limiting. Quorumsensing systems also engage in inter-kingdom signaling between bacteria and their
natural hosts [24].
Quorum-sensing systems differ greatly between Gram-negative and -positive bacteria. In Gram-negative bacteria, the LuxIR-type of quorum-sensing system consists
of two proteins, an enzyme that synthesizes a diffusible small molecule and a cytoplasmic response regulator that regulates gene expression [20]. The synthetase catalyzes the production of a HSL, which is a small amphipathic molecule that readily
crosses the lipid bilayer of a membrane. Once inside the intracellular environment,
HSL molecules bind to the response regulator, causing the binding specificity towards
their cognate DNA operators to increase. In natural promoters, the DNA operator is
typically located upstream of the promoter, leading to the activation of transcription
once the response regulator has bound to it.
In contrast, Gram-positive bacteria use short peptides as signaling molecules. The
peptides are ribosomally produced, post-translationally modified (e.g. by cleavage or
covalent addition), and then secreted across the membrane through an active transport system, such as an ABC transporter or the Sec secretion pathway [60]. After
diffusing through the extracellular medium, the peptides indirectly regulate gene
expression by stimulating the histidine kinase of a TCS.
We will focus our attention on the LuxIR-type system because it has been extensively used in developing engineering applications. In figure 2, we list LuxIR-type
quorum-sensing systems from a variety of Gram-negative bacteria, including their
synthetases and response regulators, the diffusible chemical signal produced by the
synthetase, the DNA operator that binds to the response regulator, and a brief list of
example promoters in their respective natural hosts.

206

Salis Tamsir Voigt

Synthesis of Signal
The synthesis of an HSL is catalyzed by the LuxI family of enzymes. The precursors
are S-adenosyl methionine and an acyl group, carried by an acyl carrier protein [61,
62]. The acyl group varies according to the synthetase; it is typically a fatty hydrocarbon chain with 320 carbons, possibly modified by carbonyl or hydroxyl groups
at the carbon. The acyl group is conjugated to S-adenosyl methionine, followed by
lactonization of the ring structure. For example, 3-oxo-hexanoyl homoserine lactone
(3O6-HSL) contains a six carbon acyl chain with a carbonyl group at the carbon
position. Likewise, tetradecenoyl homoserine lactone (C14-HSL) contains an acyl
group with a 14 straight chain hydrocarbon group. More recently, several bacteria
were found to acylate their HSL molecules with scavenged p-coumaric acid rather
than a fatty hydrocarbon chain [63]. Most natural enzyme synthetases produce significant amounts of only one type of HSL molecule, but there are exceptions. Likewise,
the response regulators typically bind to only type of HSL molecule, although there is
evidence of cross-talk in binding specificities [64].

Response to Signal
Gene regulation by HSL is concentration-dependent. At low concentrations, only
a tiny fraction of the response regulator is bound by AHL and little transcriptional
activation is observed. As the HSL concentration increases, more response regulator is bound by HSL and transcription activates, up to a maximum rate. In a natural microbial community of low density, the expression of the enzyme synthetase
is low, causing the HSL concentration to remain below the threshold of transcriptional activation. Once the population density inside the community increases,
there are more bacteria expressing the enzyme synthetase and the increased concentration of HSL can trigger transcriptional activation. In addition, the natural
promoters that drive the enzyme synthetase expression are often activated by the
HSL-binding response regulator, forming a positive feedback loop. Once the HSL
concentration crosses the threshold in a subset of the microbial population, the
positive feedback loop causes the expression of the enzyme synthetase and the HSL
concentration to rapidly increase, leading to gene activation in the entire microbial
population.
Once bound by HSL, the response regulators typically dimerize in the cytoplasm
and then bind to their cognate DNA operator. For the LuxR transcription factor, the
operator is a 20 base pair palindromic sequence called the lux box. The operator is
found 42.5 bp upstream of the transcriptional start site, enabling a bound LuxR to
form attractive interactions with RNA polymerase. Other response regulators in the
LuxR family of transcription factors bind with the same mechanism and often have
DNA operators with similar consensus sequences. In addition, by mutating the DNA

Engineering Bacterial Sensors

207

Synthetase

Response
regulator

Natural host

Signal molecule

DNA operator

Example
promoters

CerI

CerR

Rhodobacter
sphaeroides

C14-HSL

ND

ND

94

EsaI

EsaR

Pantoea
stewartii

3OC6-HSL

GCCTGTACTATAGTGCAGGT

esaR

95

LasI

LasR

Pseudomonas
aeruginosa

3OC12-HSL

ACCTGCCAGTTCTGGCAGGT

lasB, rsaL

96

LuxI

LuxR

Vibrio fischeri

3OC6-HSL

ACCTGTAGGATCGTACAGGT

luxI

61,97

RhlI

RhlR

Pseudomonas
aeruginosa

C4-HSL

TCCTGTGAAATCTGGCAGTT

rhlI, rhlA

96

TraI

TraR

Agrobacterium
tumefaciens

3OC8-HSL

ATGTGCAGATCTGCACAT

traA, traC

98

VanI

VanR

AACTGTTCGATCGAACAGGT

VanI

99

YenI

YenR

ND

100

Vibrio anguillarum 3OC10-HSL


Yersinia
enterocolitica

3OC6-HSL, C6-HSL
3OC10-HSL, 3OC12-HSL,
3OC14-HSL

ND

Ref.

Fig. 2. A list of well-characterized quorum-sensing systems and their attributes [61, 94100]. Each
quorum-sensing system is composed of a synthetase that produces a diffusible signal molecule,
which binds to a response regulator. The response regulator binds to a DNA operator to regulate
transcription. Example promoters from the natural hosts are shown. ND = Not determined.

operator, it is possible to alter the binding specificity of the LuxR-DNA interaction


[65]. Other mutations resulted in HSL-independent binding of LuxR to its DNA
operator.
While the LuxR family of response regulators typically functions as a transcriptional activator, it is possible to create a synthetic promoter that instead uses the LuxR
transcription factor as a repressor [66]. By moving the lux operator in between the
35 and 10 positions of the promoter, bound LuxR will form steric clashes with
RNA polymerase, preventing its recruitment to the promoter. When developing engineering applications, it is highly useful to use LuxR and other LuxR family response
regulators as either activators or repressors as the need requires.

Engineering New Bacterial Sensors

The diversity of the millions of bacterial species that inhabit the planet and their
wealth of sensor systems almost guarantees that, somewhere out there, there is a sensor system that has evolved to respond to your chemical or physical stimulus of interest. This is especially true for nutritional sources or toxins. However, an individual
bacterial strain will only possess the sensor systems that are important for detecting
changes in its natural environment. Consequently, when developing biotechnological

208

Salis Tamsir Voigt

applications, our favorite genetically tractable host may lack the ability to sense a
stimulus of interest and we will need to add to or modify its existing sensor systems.
This section describes two general strategies for engineering new bacterial sensors
and demonstrates their use with successful examples.
The first strategy takes advantage of the wealth of sensor systems in nature and
chooses one that responds to the stimulus of interest. The challenge is to correctly
interface the capabilities of the recombinant sensor system into the new bacterial host.
The first strategy is demonstrated by reviewing the construction of a recombinant
light-responsive TCS in E. coli [101], which uses a semi-rational method of creating
hybrid sensor kinases. The second strategy is named directed evolution and simulates
the process of evolution itself, sped up by many orders of magnitude, to modify an
existing sensor system to respond to a new stimulus. This strategy is illustrated with
recent examples of using directed evolution to generate new autoinducer synthetases
and response regulators.

A Semi-Rational Strategy for Creating Hybrid Sensor Kinases


A hybrid sensor kinase is a chimera or fusion protein between the sensor kinases
from two different TCSs. The hybrid sensor kinase has the ability to detect the same
stimuli as the first sensor kinase and accordingly modulate kinase activity. The target
of the hybrids kinase activity is the same response regulator as the second sensor
kinase. Thus, the hybrid has the input stimulus of the first sensor kinase and the output response of the second.
In the past two decades, at least 12 hybrid sensor kinases have been constructed
to help elucidate the mechanisms of signal transduction across the membrane. These
hybrid sensor kinases support the hypothesis that there is a conserved signal transduction mechanism mediated by a HAMP or a HAMP-like domain. For example, the
TarEnvZ hybrid phosphorylates the OmpR response regulator in response to high
concentrations of l-aspartate in E. coli [102]. It was constructed by fusing together
the first 255 amino acids of the Tar receptor and the last 229 amino acids of the EnvZ
kinase, containing the Tar aspartate-binding transmembrane domain, the Tar HAMP
linker, and the EnvZ HK and ATPase domains. Other examples include the TarTsr
[103], TrgEnvZ [104], TrgTsr [105], TapTar [106], NarXTar [107], AerTsr [108],
NarXNarQ [54], NarQNarX [54], and NarXCpxA [54] hybrids in E. coli and the
McpBMcpC hybrid in B. subtilis [109].
Many of the hybrids replace an ill-defined stimulus, such as osmolarity or excess
unfolded protein, with an externally controllable one, such as serine or nitrate, allowing a higher throughput mutational analysis of the system. In addition, by combinatorially swapping the domains inside the sensor kinase, one can demonstrate the
extent of their modularity. However, all of the components of the above hybrid sensor
kinases originated from a single bacterial species. While one might expect that the

Engineering Bacterial Sensors

209

TCSs inside the same bacterial strain are partly modular, the same expectation may
not remain true for bacteria which are more evolutionary distant. The next step is to
test such a hypothesis by creating a functional hybrid sensor kinase whose components were extracted from bacteria in different families. At the same time, we would
like to augment our favorite genetically tractable host E. coli with an amazingly
useful sensing capability, the ability to see light.
Most photosynthetic bacterial species are capable of detecting changes in light
to synchronize their metabolism. The Synechocystis sp. PC6803 strain from the
cyanobacteria family possesses the two-component sensor kinase Cph1, containing
a perisplasmic and transmembrane structure composed of the PAS, GAF, and phytochrome-binding domains connected to the cytoplasmic histidine kinase and ATPase
domains. In the presence of red light (650 nm), the phycocyanobilin (PCB) phytochrome causes the formation of a covalent adduct, inducing the dimerization of a pair
of Cph1 proteins, and halting the histidine kinase activity [110112]. The histidine
kinase activity is restored when the covalent adduct and the dimerization process are
reversed. The reverse process occurs quickly when Cph1 is exposed to far red light
(720 nm) or more slowly when Cph1 is in the dark.
The goal was to develop a light-sensing TCS in E. coli that can regulate gene transcription in response to varying intensities of red light (650 nm). To do this, the histidine kinase portion of the hybrid sensor kinase must be able to phosphorylate a
DNA-binding response regulator. The TarEnvZ hybrid was an inspiration since it
exhibits the desired behavior; it can directly alter gene expression using the OmpR
response regulator and the ompC promoter. Consequently, the light sensor was chosen to be a hybrid Cph1EnvZ sensor kinase. Before beginning the experimentation,
it was prudent to develop a strategy to identify the best way to fuse together domains
from the Cph1 and EnvZ proteins. A screen was also developed to verify whether
such a fusion was functional.
A rational strategy for creating new hybrid sensor kinases was formulated by
examining the similarities of the previously constructed ones. The key similarity is
the location of the fusion point that demarcates the border between the two sensor
kinases, illustrated in figure 3. The fusion points are found within the HAMP linker,
which consists of two helical regions separated by a connector. The fusion points lie
1018 amino acids either upstream (N-terminal) or downstream (C-terminal) of
these connecting regions. In addition, when the amino acid sequences of the two sensor kinases are correctly aligned, the ending point of the first sensor kinase is always
within two amino acids of the beginning point of the second one. This strong pattern suggests a systematic method of identifying where to best combine two sensor
kinases to create a hybrid. The pattern arises when both sensor kinases use the same
conserved signal transduction mechanism to modulate kinase activity.
Importantly, in order to more accurately determine the location of the fusion point,
we must correctly align the amino acid sequences of the two sensor kinases and verify
that the two sensor kinases do indeed use the same signal transduction mechanism.

210

Salis Tamsir Voigt

Hybrid sensor kinases

HAMP alignments

Fig. 3. A semi-rational strategy for creating hybrid sensor kinases. Top: The domain alignments
(boxes) and amino acid compositions (black lines) for six different functional hybrid sensor kinases
are diagrammed (not to scale), each with a common fusion point located between the HAMP and
histidine kinase (HisKA) domains. At the top and bottom are the sensor kinase names and the amino
acids that constitute each hybrid. Bottom: The amino acid sequence alignments of the HAMP domain
are shown for each pair of sensor kinases. Aligned amino acids are bolded. At the top and bottom of
each alignment, the corresponding predicted secondary structures are indicated as either (H) helical
or (C) connecting along with the fusion point (*). Secondary structure predictions were made using
SCRATCH [122]. Numbers in parentheses refer to the amino acids included in the alignment.

Engineering Bacterial Sensors

211

To do this, we can not rely on a multiple sequence alignment algorithm alone. Because
the algorithm seeks to minimize its alignment score, the results will sometimes favor
alignments that have no basis in evolutionary conservation. Instead, we must combine the multiple sequence alignment algorithm with a comparison of the secondary
structures and conserved protein domains between the two sensor kinases. We start
by examining the protein domains inside each sensor kinase. If the HAMP linker is
present in both sensor kinases, we can focus our comparison to the region surrounding the HAMP linker. If the HAMP linker is absent, there may be a HAMP-like region
with the same function. Consequently, we instead focus on the region consisting of
the ~100 amino acids upstream (N-terminal) of the histidine kinase domain. This
region will be responsible for transmitting the signal transduction motion from the
sensor domain to the histidine kinase.
Using one of many algorithms [reviewed in 113], we then calculate the secondary
structures that form in this region, such as helical, -sheet, coiled-coil, or disordered.
The signature for the HAMP linker is a pair of helical regions linked by a disordered
connector. Importantly, other protein domains may appear HAMP-like with similar secondary structures, even though their stated function is quite different. These
secondary structure prediction algorithms are a more course-grained version of the
PFAM method and can sometimes discern similarities between regions of evolutionary distant protein domains.
Finally, we use ClustalW [114] to perform the amino acid alignment of the two
sensor kinases, using only the region surrounding the HAMP linker. It is important
to limit the scope of the alignment because our goal is to identify the common fusion
point and the secondary structures inside the region responsible for signal transduction. If we were to include the entire amino acid sequence, then the sensor domains
and other non-conserved regions may cause the alignment to become skewed, yielding an incorrect fusion point. We can verify that the amino acid alignment is correct
by overlaying the secondary structure predictions. A correct alignment should also
yield a close alignment in the secondary structures between the two sensor kinases.
In figure 3, we demonstrate this strategy when applied to the Cph1EnvZ hybrid
[101]. The Cph1 light sensor does not contain a HAMP linker. However, when we
focus on the region upstream of its histidine kinase domain, calculate its secondary
structures, and align the regions amino acid sequence with EnvZ, we find that there is
a significant similarity between the two regions. Both contain a pair of helical regions
separated by a disordered connector (not shown due to space constraints); the Cph1
helical region is much longer, though, which may be required for proper signal transduction. When the H222 fusion point of the TarEnvZ hybrid is located on the alignment, we note that the corresponding fusion point in Cph1 is Q517. Consequently,
the strategy suggests that the first 517 amino acids from Cph1 fused to the last 229
amino acids from EnvZ would produce a functional hybrid sensor kinase.
Of course, in applying the strategy here, we are relying upon the successful construction of the TarEnvZ and other EnvZ hybrids. Previous work demonstrates that

212

Salis Tamsir Voigt

the H222 fusion point in EnvZ can result in a functional hybrid sensor kinase. In general, however, we would like to apply the same strategy to other sensor kinases besides
EnvZ. To do this, we may perform the same alignment procedure, except that we align
the HAMP or HAMP-like regions from three sensor kinases: the two sensor kinases
that we wish to hybridize and EnvZ, as a reference sequence. The fusion points of the
two sensor kinases are determined by locating the positions aligned with the reference
H222 fusion point in the EnvZ sequence.
This strategy can fail in many circumstances. When the two sensor kinases do not
use the HAMP-mediated signal transduction mechanism to regulate histidine kinase
activity there is then no evolutionary basis for correctly choosing the fusion point.
For example, a hybrid sensor kinase was attempted by fusing together the first 212
215 amino acids of Tar and the last 697700 amino acids of ArcB, but the product
did not differentially regulate the ArcA-binding lldP promoter in response to aspartate [115]. In this case, it would be more difficult to locate the correct fusion point
using evolutionary conservation because the ArcB sensor kinase lacks a HAMP-like
domain and does not contain a significant periplasmic portion. As always, there is no
guarantee that secondary structure predictions or sequence alignments will return
the correct fusion points. Consequently, we need to expand our search, making this a
semi-rational strategy whose purpose is to narrow down the number of constructed
hybrid sensor kinases that need to be tested for correct function. When developing
the Cph1EnvZ hybrid, a small library was generated that varied the Cph1 fusion
point up to 9 amino acids around the predicted Q517 fusion point [101].
Besides the correct fusion points, a functioning Cph1EnvZ hybrid also requires
a source of the phytochrome PCB. The two-gene metabolic pathway pcyA and ho1
from Synechocystis synthesizes PCB from heme and was placed under control of the
Para/lac synthetic promoter, which is activated by arabinose or IPTG when the AraC
or LacI transcription factors are present. The pathway was inserted into the pPLPCB plasmid, containing a p15A origin of replication and an ampicillin resistance
marker. In addition, each Cph1EnvZ hybrid was placed under control of the Ptet
promoter in a pTet-CphEnvZ plasmid containing a ColE1 origin of replication and
the chloramphenicol resistance marker. The plasmids pPL-PCB and pTet-CphEnvZ
were co-transformed into the CP919 E. coli strain, a rbsB- and rbsK- derivative of
RU1012 E. coli. RU1012 contains a chromosomal copy of lacZ under control of the
ompC promoter and lacks the tetR and lacI repressors, but contains a copy of the araC
transcription factor. Synthesis of PCB was induced with addition of arabinose and
light-dependent regulation was reported with a -galactosidase assay.
Light induction of the ompC promoter by each library member of the Cph1EnvZ
hybrid was tested using a high throughput plate assay. After an initial overnight culture
in LB with appropriate antibiotics, two 2-ml cultures in 24-well plates were each inoculated with a 1:1,000 dilution. The first 24-well plate was exposed to a standard 100
W broad-spectrum glowing lamp while the second 24-well plate was wrapped in aluminum foil and kept in darkness. Both plates were grown at 37C for 4 h. Afterwards,

Engineering Bacterial Sensors

213

each culture was then diluted 1:100 and transferred to an identical 24-well plate under
the same light condition and grown for another 4 h. The intermediate dilution step
maintains the cultures in the exponential phase of growth while allowing additional
time for the Cph1-EnvZ hybrid to regulate gene expression. A standard Miller assay
using o-nitrophenyl--galactosidase was then performed on each culture in the light
and dark, looking for the Cph1EnvZ hybrid that yielded the maximum differential
in ompC expression between the light and dark conditions. Perhaps surprisingly, the
most active Cph1EnvZ hybrid utilized the predicted Q517 fusion point in Cph1,
demonstrating the usefulness of the proposed semi-rational strategy.
Finally, a different strategy for creating hybrid sensor kinases was recently explored
[116]. The amino acids responsible for determining the binding specificity of a histidine kinase to its cognate response regulator were determined by analyzing the frequency of their co-evolution. By transplanting these amino acids from one histidine
kinase to another, it was shown that the binding specificities towards the response regulator can also be swapped. Specifically, the response regulator-binding amino acids
in the EnvZ histidine kinase were replaced by their corresponding sequences from the
RstB, CpxA, PhoR, AtoS, and PhoQ histidine kinases. The mutant EnvZ kinase lost
its ability to phosphorylate OmpR, but gained kinase activity towards the RstA, CpxR,
PhoB, AtoC, and PhoP response regulators. Using this strategy, the stimulus of EnvZ
is rewired to regulate a promoter that binds to an alternate response regulator.

A Directed Evolution Strategy for Creating Quorum-Sensing Systems


Directed evolution is an iterative procedure that incrementally optimizes a molecular
target for a selectable function or behavior. It has been extensively used to improve
the kinetics and thermostability of enzymes, modify substrate specificity, and increase
expression [117, 118]. The procedure begins with a protein that exhibits a minimal
amount of the desired activity. Using random mutagenesis or gene recombination, a
pool of protein variants is generated. The mutations may also be selectively targeted
to a region of the protein most likely to be responsible for the activity. A well-designed
screen then identifies which mutations resulted in improved protein function. The
protein variants carrying these mutations are selected and carried into the next iteration of the directed evolution procedure. By repeating the procedure, mutations that
yield the desired improved activity will become enriched. Importantly, no structural
or biochemical knowledge is required to perform the procedure and the generated
mutations often provide key insights to the structure-function relationship.
The directed evolution procedure has been applied to both the LuxI enzyme
synthetase and the LuxR response regulator to alter their signal-generating or signal-receiving specificities. Collins et al. [119] developed a dual positive and negative selection scheme that successfully altered the HSL-binding specificity of LuxR
from 3OC6-HSL to a straight chain acyl HSL, such as C12-HSL. The altered binding

214

Salis Tamsir Voigt

specificity required only four mutations (T33A, S116A, M135I, and R67M), demonstrating the evolutionary plasticity of LuxR. Importantly, the first three mutations
were determined by using a positive-only screen that searched for broadened HSLbinding specificities [120]. A LuxR with these three mutations bound to both 3OC6HSL and straight chain acyl groups of varying lengths. Afterwards, the HSL-binding
specificity was narrowed to only straight chain acyl groups by applying both the positive and negative selection schemes. The positive scheme enriches for mutations that
resulted in LuxR binding to straight chain acyl groups while the negative scheme
selected against mutations in LuxR that allowed 3OC6-HSL binding.
The catalytic rate of AHL synthesis by LuxI was also increased by the directed evolution procedure [121]. Using a similar positive selection scheme, four mutations in
LuxI (K19R, E34G, V39V, and E63G) resulted in an 80-fold increase in the production
rate of 3OC6-HSL by LuxI. As a side effect, the production rate of the straight chain
C6-HSL also increased, suggesting that a dual positive-negative screen may be required
to maintain product specificity. The expression level of this LuxI variant was increased
2-fold, indicating that the per-protein catalytic rate increased by about 40-fold.
These two examples illustrate how directed evolution can alter the HSL-binding
specificity of the response regulator as well as production rate and specificity of the
enzyme synthetase. When developing biotechnological applications, it is highly useful
to have multiple available channels of cell-cell communication. Each of these channels requires an orthogonal quorum-sensing system that produces and responds to
different HSL-signaling molecules with no cross-talk. By using directed evolution,
additional orthogonal quorum-sensing systems may be created.

Engineering Synthetic Gene Networks with Rewired Sensors

Natural sensor systems are the eyes and ears of their hosts. Their signals are fed into
the downstream genetic circuitry responsible for coordinating and carrying out cellular functions. Using the newest techniques in genetic engineering, we can completely
redesign the regulatory interactions of these genetic networks, creating synthetic versions that exhibit radically different behaviors. By extracting the genetic components
of natural sensor systems the sensor kinases, synthetases, response regulators, and
promoters and recombining them with other genetic parts [8], such as transcription
factors, epigenetic switches, and fluorescent reporters, these synthetic genetic networks can exhibit a variety of interesting dynamical behaviors, including bistability,
oscillations, and pattern formation [7].
In the emerging field of synthetic biology, we use creativity and engineering principles, such as quantitative modeling and abstraction, to identify optimal ways to connect these genetic parts together in order to exhibit a certain interesting and desirable
behavior. The goal is to reliably engineer microbes to become living, self-replicating
computers that use sensors to beneficially interact with their environment [123].

Engineering Bacterial Sensors

215

The variety of constructed sensor-driven synthetic gene networks is best illustrated


with examples. A synthetic gene network enabled E. coli cells to invade cancer-derived
HeLa, HepG2, or U2OS cells only when tumor-indicative conditions were present by
connecting the output of a sensor system to the expression of the invasin (inv) gene
from Yersinia pseudotuburculosis [124]. The inv expression was activated only at high
cell density or during hypoxic growth conditions, which are the environmental conditions associated with the accumulation of bacteria at a tumor site and the high respiratory rate of cancerous cells. Cell density was sensed by using the quorum-sensing
LuxIR system coupled to a LuxR-activated promoter (Plux) while the hypoxic response
was mediated by fdhF promoter. The fdhF promoter integrates sensor signals from
the nitrate-sensing NarX/NarL and NarQ/NarP TCSs and the formate-sensing Fnr
and oxygen-sensing Fhl transcription factors to detect anaerobic growth conditions.
In two additional examples, the NtrB/NtrC TCS in E. coli has been modified to
either generate spontaneous oscillations [125] or to create artificial cell-cell communication [126]. In the first example, a combination of a positive and negative feedback loop on the expression of the NtrC response regulator and the LacI repressor
generates oscillations in -galactosidase activity. Interestingly, the NtrB sensor kinase
is overexpressed so that phosphorylation of NtrC is constitutive and independent
of nitrogen availability, making NtrC behave like a transcriptional activator. In the
second example, the NtrB/NtrC system is converted into a cell-cell communication
system that uses acetate as the diffusible signaling molecule. Knocking out both the
pta and ntrB genes causes the rate of phosphorylation of NtrC by acetyl phosphate to
be proportional to the amount of acetate in the media. As the cell density increases,
higher concentrations of secreted acetate are interconverted into acetyl phosphate,
leading to more NtrC-P. The expression of a GFP reporter is activated by NtrC-P
by the natural glnAp2 promoter, allowing observations of the quorum-sensing signal.
These examples illustrate how one can modify natural TCSs to create new types of
behaviors.
In the remainder of this section, the focus is on two synthetic gene networks that
use quorum sensing to either create spatial patterns on an agar plate or a synthetic
predator-prey system that oscillates. The regulatory interactions of these synthetic
gene networks are shown in figure 4.

A Pattern-Forming Synthetic Gene Network


Pattern formation is an important cellular process that drives the differentiation of
cells by regulating their gene expression according to the concentration of a diffusible
signaling molecule, called a morphogen. By using a quorum-sensing system to create
and sense a morphogen, one can study how bacterial gene networks respond to morphogen gradients to create different spatial patterns. The insights gained by manipulating and observing these simple genetic networks can elucidate the governing

216

Salis Tamsir Voigt

PLtetO-1

Plux

Sender

lacl

luxl
Plac

PPR

cl

lacl

A Petri dish
with sender S
and receiver

Output
GFP
(au)
gfp

S
OFF
ON
OFF

AHL (M)

a
PLtetO-1

Predator strain

luxR

Prey strain

Plac/ara

lasI
3OC12-HSL
Plac/ara

Plux

Band-pass
filter

lasR

luxI

Plux
3OC6-HSL

Cell density (103 cells/mL)

ccdA

25

0 M IPTG

20
15
10
5
0
0

ccdB

50

80
70
60
50
Predator
Prey
40
30
20
10
0
100
150
0

Time (h)

ccdB
5 M IPTG

50 100 150 200 250 300 350 400

Time (h)

80
70
60
50
40
30
20
10
00

1,000 M IPTG

50 100 150 200 250 300 350 400

Time (h)

Fig. 4. Rewiring synthetic genetic networks using quorum-sensing systems. a A sender E. coli strain
transmits a quorum-sensing signal through the band-pass filter and output modules, located in a
receiver E. coli strain, to activate gfp expression. The band-pass filter causes gfp expression to turn on
at an intermediate AHL concentration and to remain off at either high or low AHL concentrations. A
ring pattern of gfp expression spontaneously forms when a Petri dish containing LB agar and receiver
strain is seeded, at the center, with sender strain. b Orthogonal quorum-sensing signals sent and
received by predator and prey strains spontaneously generate oscillations in cell density. The predator strain produces the 3OC12-HSL signal, which binds to the LasR response regulator in the prey
strain and activates the ccdB toxin. Likewise, the 3OC6-HSL signal, produced by the prey strain, binds
to the predators LuxR response regulator to activate the ccdA anti-toxin. High predator cell densities
increase ccdB expression in the prey strain, leading to prey strain death. Low prey cell densities
decrease ccdA anti-toxin in the predator strain, leading to predator strain death. A quantitative model
predicts that varying the IPTG concentration will modulate the period of the resulting oscillations by
altering the rates of predator ccdB and prey luxI expression, which is confirmed by experimental
characterization.

Engineering Bacterial Sensors

217

principles of these systems [127], which may apply to the same types of networks
present in higher organisms.
Basu et al. [128] constructed a synthetic multicellular system that generates a complex two-dimensional ring pattern. Shown in figure 4a, the system consists of two E.
coli strains, a sender and a receiver, both growing on an agar plate. The sender strain
expresses LuxI to create the HSL diffusible signal. The receiver strain uses the LuxR
response regulator to activate a gene network that behaves like a genetic version of
a band-pass filter. An electronic band-pass filter is an analog circuit that converts
an input signal into an output signal only when the input signal has an intermediate
value. If the input is too low or too high, then the output is zero. In this case, a genetic
band-pass filter uses HSL concentration as the input signal and the GFP reporter as
the output. Because HSL diffuses away from the sender cells, the HSL concentration
decreases as the distance from the source increases. Receiver cells that sense either a
high or low concentration of HSL will repress GFP expression while, at an intermediate HSL concentration, the receiver cells will activate GFP expression. Consequently,
if the sender cells are placed in the center of the agar plate with the receiver cells surrounding it, a ring pattern will spontaneously form. Other patterns are also possible
by seeding the sender cells at different locations on the agar plate.
A genetic implementation of a band-pass filter is implemented by using Plux to drive
the expression of both the LacI and CI repressors. A CI-repressed promoter (PPR) then
drives the expression of another copy of the LacI repressor. Finally, a LacI-repressed
promoter (Plac) drives the expression of the GFP output. At low HSL concentrations,
the Plux promoter produces only a basal rate of expression of the CI and LacI repressors.
With a low concentration of CI repressor, the PPR promoter expresses the second copy
of the LacI repressor, leading to repressed GFP expression. At high HSL concentrations,
LacI is actively produced by the Plux promoter, which also represses the GFP output.
At intermediate HSL concentrations, the first copy of LacI is expressed at a minimal
level while the second copy is repressed by CI. Consequently, there is not enough LacI
repressor to sufficiently repress GFP expression and the output is turned on.
Importantly, any genetic network that features a pair of similarly acting repressors in this feedforward loop configuration will behave like a band-pass filter; it will
express a gene when the input signal is within an intermediate value. We call this a
band-pass filter design. By identifying natural genetic networks that are configured
like a band-pass filter, we can infer their behavior and better understand why genetic
networks have evolved such designs.

An Oscillatory Predator-Prey System in Escherichia coli


Predator-prey systems are classic examples of cross-species interactions that result in
interesting dynamical behaviors. In ecosystems spanning from forests and oceans to
microscopic communities of bacteria, organisms compete for nutritional resources to

218

Salis Tamsir Voigt

survive. Predatory species feed on prey species for nutrition while prey species extract
chemical nutrients from the environment. Because the growth and death rates of both
predator and prey species depend on one another, these systems can exhibit oscillations in the number of predator and prey organisms. Balagadde et al. [129] recreated a
synthetic version of a predator-prey system by designing two strains of E. coli that use
a pair of orthogonal quorum-sensing systems to alter each others growth and death
rates. The authors were successful in creating a synthetic system that can generate
oscillations in the cell densities of predator and prey E. coli strains with long periods
of ~180 h. Interestingly, by manipulating the strengths of cell-cell communication, the
period of oscillations was also controllable. In figure 4b, the synthetic gene networks
inside the predator and prey E. coli strains are shown. The dynamics of a quantitative
model, described in the Balagadde et al. paper, demonstrate how tuning the strength
of cell-cell communication with IPTG can control the period of oscillations.
The predator-prey system consists of two synthetic gene networks. E. coli cells were
labeled as the predator or prey strain according to which synthetic gene network they
possessed. The predator gene network constitutively expresses the LasI synthetase and
the LuxR response regulator. LasI produces the 3OC12-HSL-signaling molecule while
LuxR binds to the 3OC6-HSL autoinducer. Conversely, the prey gene network uses a
LacI-repressed promoter (Plac/ara) to express the LuxI synthetase and the LasR response
regulator, which produce 3OC6-HSL and bind to 3OC12-HSL, respectively. The rate
of expression from this promoter may be increased by adding the IPTG inducer. These
two quorum-sensing systems do not significantly interact with one another, creating
two independent cell-cell communication channels. The rates of production of the
3OC6-HSL- and 3OC12-HSL-signaling molecules are proportional to the cell densities of the prey and predator strains, respectively. By altering the growth or death rates
according to the concentrations of these diffusible signals, one can create the crossspecies interactions that are required to generate predator-prey oscillations.
The death rates of the predator and prey strains are controlled by regulating the
expression of the ccdB toxin and the ccdA anti-toxin. Both networks use the Plux promoter, which can bind to both LuxR and LasR. In the prey gene network, the Plux
promoter drives the expression of the toxin. When the predator cell density is high,
the 3OC12-HSL concentration increases and activates toxin production, leading to
prey cell death. Conversely, in the predator gene network, the Plux promoter drives
the expression of the anti-toxin. When the prey cell density is high, the 3OC6-HSL
concentration increases, resulting in increased anti-toxin production and higher cell
viability in the predator strain. Consequently, oscillations will ensue with alternating
high and low cell densities of both predator and prey strains.
A quantitative model was developed that describes how the molecular interactions
in the system give rise to oscillations. Its simplified version consists of four ordinary
differential equations describing the rates of change of the concentrations of the two
signaling molecules as well as the predator and prey cell densities. A derivation of
the model is available in the supplementary information of the original article [129].

Engineering Bacterial Sensors

219

An important prediction of the model is that the addition of IPTG will increase the
period of oscillations. The IPTG inducer increases the constitutive rate of transcription of the Plac/ara promoter, which regulates toxin production in the predator strain
and production of LasR and LuxI in the prey strain. In figure 4b, the dynamics of the
quantitative model are shown with varying concentrations of IPTG. With no IPTG
inducer, the predator strain does not produce the toxin, resulting in predator overgrowth. Once IPTG is added, oscillations are predicted to arise with a period of ~150
h at 5 m IPTG and ~250 h at 1 mm IPTG.
The dynamics of the predator-prey system were characterized by co-culturing predator and prey E. coli strains in an advanced 9 nl microchemostat. Either the predator
or prey strain constitutively expresses GFP, allowing one to measure their cell densities. Their respective densities were measured for over 500 h by using an automated
bright-field and fluorescence microscopy platform. Oscillations in predator and prey
cell densities were observed with periods ranging from 50 to 180 h, depending on the
microchemostat dilution rate and IPTG concentration. The characterization experiments verified the models prediction that increasing the IPTG concentration results
in longer periods of oscillation.
The synthetic predator-prey system in E. coli demonstrates how one can use a pair
of orthogonal quorum-sensing systems to engineer a two-way communication channel across variants of the same organism. This cell-cell communication method can be
used in a variety of synthetic gene networks to exhibit some very interesting dynamical or pattern-forming behaviors. In this example, the signaling molecules regulated
the growth and death rates of the two strains, giving rise to oscillatory behavior that
was measured over extremely long time periods in an advanced microchemostat system. Future synthetic gene networks can improve on this design by using multiple
quorum-sensing systems as inputs to a circuit module, such as a band-pass filter, to
generate even more complex types of dynamical or spatial behaviors.

Concluding Remarks

Bacteria are living, self-replicating chemical computers capable of both sensing and
altering their environment. In the quest for new biotechnologies, their genetic code is
redesigned to beneficially interact with the physical world, carrying out tasks that are
important to humankind. In this chapter, both the resources and strategies for augmenting bacteria with new sensor systems are provided, including several successful
examples. By constructing synthetic gene networks that use new or existing sensor
systems, downstream processes can be controlled according to environmental conditions, connecting the eyes and brains to the hands that carry out a task. Sensordriven synthetic gene networks can optimize metabolic pathways, regulate bacterial
therapeutic systems, and counter bacterial pathogens, by adjusting the particular process according to changing nutrient, oxygen, or autoinducer availabilities.

220

Salis Tamsir Voigt

References
1 Galperin MY: Bacterial signal transduction network
in a genomic perspective. Environ Microbiol 2004;6:
552567.
2 Pal C, Papp B, Lercher MJ: Adaptive evolution of
bacterial metabolic networks by horizontal gene
transfer. Nat Genet 2005;37:13721375.
3 Beier D, Gross R: Regulation of bacterial virulence
by two-component systems. Curr Opin Microbiol
2006;9:143152.
4 Withers ST, Keasling JD: Biosynthesis and engineering of isoprenoid small molecules. Appl Microbiol
Biotechnol 2007;73:980990.
5 Wendisch VF, Bott M, Eikmanns BJ: Metabolic
engineering of Escherichia coli and Corynebacterium
glutamicum for biotechnological production of
organic acids and amino acids. Curr Opin Microbiol
2006;9:268274.
6 Verlinden RA, Hill DJ, Kenward MA, Williams CD,
Radecka I: Bacterial synthesis of biodegradable
polyhydroxyalkanoates. J Appl Microbiol 2007;102:
14371449.
7 Andrianantoandro E, Basu S, Karig DK, Weiss R:
Synthetic biology: new engineering rules for an
emerging discipline. Mol Syst Biol 2006;2:2006 0028.
8 Voigt CA: Genetic parts to program bacteria. Curr
Opin Biotechnol 2006;17:548557.
9 Kaplan S, Bren A, Zaslaver A, Dekel E, Alon U:
Diverse two-dimensional input functions control
bacterial sugar genes. Mol Cell 2008;29:786792.
10 Korner H, Sofia HJ, Zumft WG: Phylogeny of the
bacterial superfamily of Crp-Fnr transcription regulators: exploiting the metabolic spectrum by controlling alternative gene programs. FEMS Microbiol
Rev 2003;27:559592.
11 Serganov A, Patel DJ: Ribozymes, riboswitches and
beyond: regulation of gene expression without proteins. Nat Rev Genet 2007;8:776790.
12 Mandal M, Breaker RR: Gene regulation by riboswitches. Nat Rev Mol Cell Biol 2004;5:451463.
13 Zakataeva NP, Gronskiy SV, Sheremet AS, Kutukova
EA, Novikova AE, Livshits VA: A new function for
the Bacillus PbuE purine base efflux pump: efflux of
purine nucleosides. Res Microbiol 2007;158:659
665.
14 Serganov A, Polonskaia A, Phan AT, Breaker RR,
Patel DJ: Structural basis for gene regulation by a
thiamine pyrophosphate-sensing riboswitch. Nature
2006;441:11671171.
15 Waldminghaus T, Fippinger A, Alfsmann J,
Narberhaus F: RNA thermometers are common in
- and -proteobacteria. Biol Chem 2005;386:1279
1286.

Engineering Bacterial Sensors

16 El-Samad H, Kurata H, Doyle JC, Gross CA,


Khammash M: Surviving heat shock: control strategies for robustness and performance. Proc Natl
Acad Sci USA 2005;102:27362741.
17 Phadtare S: Recent developments in bacterial coldshock response. Curr Issues Mol Biol 2004;6:125
136.
18 Kaiser M, Sawers G: Nitrate repression of the
Escherichia coli pfl operon is mediated by the dual
sensors NarQ and NarX and the dual regulators
NarL and NarP. J Bacteriol 1995;177:36473655.
19 Kaiser M, Sawers G: Overlapping promoters modulate Fnr- and ArcA-dependent anaerobic transcriptional activation of the focApfl operon in Escherichia
coli. Microbiology 1997;143:775783.
20 Waters CM, Bassler BL: Quorum sensing: cell-tocell communication in bacteria. Annu Rev Cell Dev
Biol 2005;21:319346.
21 Miller MB, Bassler BL: Quorum sensing in bacteria.
Annu Rev Microbiol 2001;55:165199.
22 Clarke MB, Hughes DT, Zhu C, Boedeker EC,
Sperandio V: The QseC sensor kinase: a bacterial
adrenergic receptor. Proc Natl Acad Sci USA 2006;
103:1042010425.
23 Gonzalez Barrios AF, Zuo R, Hashimoto Y, Yang L,
Bentley WE, Wood TK: Autoinducer 2 controls biofilm formation in Escherichia coli through a novel
motility quorum-sensing regulator (MqsR, B3022).
J Bacteriol 2006;188:305316.
24 Hughes DT, Sperandio V: Inter-kingdom signalling:
communication between bacteria and their hosts.
Nat Rev Microbiol 2008;6:111120.
25 Rhodius VA, Suh WC, Nonaka G, West J, Gross CA:
Conserved and variable functions of the E stress
response in related genomes. PLoS Biol 2006;4:e2.
26 Traxler MF, Summers SM, Nguyen HT, Zacharia
VM, Hightower GA, Smith JT, Conway T: The
global, ppGpp-mediated stringent response to
amino acid starvation in Escherichia coli. Mol
Microbiol 2008;68:11281148.
27 Gaudu P, Weiss B: SoxR, a [2Fe-2S] transcription
factor, is active only in its oxidized form. Proc Natl
Acad Sci USA 1996;93:1009410098.
28 Blanchard JL, Wholey WY, Conlon EM, Pomposiello
PJ: Rapid changes in gene expression dynamics in
response to superoxide reveal SoxRS-dependent
and independent transcriptional networks. PLoS
ONE 2007;2:e1186.
29 Campbell EA, Greenwell R, Anthony JR, Wang S,
Lim L, Das K, Sofia HJ, Donohue TJ, Darst SA: A
conserved structural module regulates transcriptional responses to diverse stress signals in bacteria.
Mol Cell 2007;27:793805.

221

30 Nohmi T: Environmental stress and lesion-bypass


DNA polymerases. Annu Rev Microbiol 2006;60:
231253.
31 Da Rocha RP, Paquola AC, Marques Mdo V, Menck
CF, Galhardo RS: Characterization of the SOS regulon of Caulobacter crescentus. J Bacteriol 2008;190:
12091218.
32 Friedman N, Vardi S, Ronen M, Alon U, Stavans J:
Precise temporal modulation in the response of the
SOS DNA repair network in individual bacteria.
PLoS Biol 2005;3:e238.
33 Purcell EB, Crosson S: Photoregulation in prokaryotes. Curr Opin Microbiol 2008;11:168178.
34 Gaidenko TA, Kim TJ, Weigel AL, Brody MS, Price
CW: The blue-light receptor YtvA acts in the environmental stress signaling pathway of Bacillus subtilis. J Bacteriol 2006;188:63876395.
35 Moskvin OV, Kaplan S, Gilles-Gonzalez MA,
Gomelsky M: Novel heme-based oxygen sensor
with a revealing evolutionary history. J Biol Chem
2007;282:2874028748.
36 Swartz TE, Tseng TS, Frederickson MA, Paris G,
Comerci DJ, Rajashekara G, Kim JG, Mudgett MB,
Splitter GA, Ugalde RA, Goldbaum FA, Briggs WR,
Bogomolni RA: Blue-light-activated histidine kinases: two-component sensors in bacteria. Science
2007;317:10901093.
37 Isaacs FJ, Dwyer DJ, Collins JJ: RNA synthetic biology. Nat Biotechnol 2006;24:545554.
38 Reinelt S, Hofmann E, Gerharz T, Bott M, Madden
DR: The structure of the periplasmic ligand-binding
domain of the sensor kinase CitA reveals the first
extracellular PAS domain. J Biol Chem 2003;278:
3918939196.
39 Kneuper H, Janausch IG, Vijayan V, Zweckstetter
M, Bock V, Griesinger C, Unden G: The nature of
the stimulus and of the fumarate binding site of the
fumarate sensor DcuS of Escherichia coli. J Biol
Chem 2005;280:2059620603.
40 Pappalardo L, Janausch IG, Vijayan V, Zientz E,
Junker J, Peti W, Zweckstetter M, Unden G, Griesinger
C: The NMR structure of the sensory domain of the
membranous two-component fumarate sensor (histidine protein kinase) DcuS of Escherichia coli. J Biol
Chem 2003;278:3918539188.
41 Taylor BL, Zhulin IB: PAS domains: internal sensors
of oxygen, redox potential, and light. Microbiol Mol
Biol Rev 1999;63:479506.
42 Yamamoto K, Ishihama A: Transcriptional response
of Escherichia coli to external copper. Mol Microbiol
2005;56:215227.
43 Grigoroudis AI, Panagiotidis CA, Lioliou EE, Vlassi
M, Kyriakidis DA: Molecular modeling and functional analysis of the AtoS-AtoC two-component
signal transduction system of Escherichia coli.
Biochim Biophys Acta 2007;1770:12481258.

222

44 Cariss SJ, Tayler AE, Avison MB: Defining the


growth conditions and promoter-proximal DNA
sequences required for activation of gene expression
by CreBC in Escherichia coli. J Bacteriol 2008;190:
39303939.
45 Raivio TL, Laird MW, Joly JC, Silhavy TJ: Tethering
of CpxP to the inner membrane prevents spheroplast induction of the cpx envelope stress response.
Mol Microbiol 2000;37:11861197.
46 Malpica R, Sandoval GR, Rodriguez C, Franco B,
Georgellis D: Signaling by the arc two-component
system provides a link between the redox state of
the quinone pool and gene expression. Antioxid
Redox Signal 2006;8:781795.
47 Maris AE, Walthers D, Mattison K, Byers N, Kenney
LJ: The response regulator OmpR oligomerizes via
-sheets to form head-to-head dimers. J Mol Biol
2005;350:843856.
48 Ogasawara H, Hasegawa A, Kanda E, Miki T,
Yamamoto K, Ishihama A: Genomic SELEX search
for target promoters under the control of the
PhoQP-RstBA signal relay cascade. J Bacteriol 2007;
189:47914799.
49 Galperin MY, Nikolskaya AN, Koonin EV: Novel
domains of the prokaryotic two-component signal
transduction systems. FEMS Microbiol Lett 2001;
203:1121.
50 Hulko M, Berndt F, Gruber M, Linder JU, Truffault
V, Schultz A, Martin J, Schultz JE, Lupas AN, Coles
M: The HAMP domain structure implies helix rotation in transmembrane signaling. Cell 2006;126:929
940.
51 Finn RD, Tate J, Mistry J, Coggill PC, Sammut SJ,
Hotz HR, Ceric G, Forslund K, Eddy SR,
Sonnhammer EL, Bateman A: The Pfam protein
families database. Nucleic Acids Res 2008;36:D281
D288.
52 Letunic I, Copley RR, Pils B, Pinkert S, Schultz J,
Bork P: SMART 5: domains in the context of
genomes and networks. Nucleic Acids Res 2006;34:
D257D260.
53 Appleman JA, Stewart V: Mutational analysis of a
conserved signal-transducing element: the HAMP
linker of the Escherichia coli nitrate sensor NarX. J
Bacteriol 2003;185:8997.
54 Appleman JA, Chen LL, Stewart V: Probing conservation of HAMP linker structure and signal transduction mechanism through analysis of hybrid
sensor kinases. J Bacteriol 2003;185:48724882.
55 Alves R, Savageau MA: Comparative analysis of
prototype two-component systems with either
bifunctional or monofunctional sensors: differences
in molecular structure and physiological function.
Mol Microbiol 2003;48:2551.

Salis Tamsir Voigt

56 Goldbeter A, Koshland DE Jr: Ultrasensitivity in


biochemical systems controlled by covalent modification. Interplay between zero-order and multistep
effects. J Biol Chem 1984;259:1444114447.
57 Yamamoto K, Hirao K, Oshima T, Aiba H, Utsumi
R, Ishihama A: Functional characterization in vitro
of all two-component signal transduction systems
from Escherichia coli. J Biol Chem 2005;280:1448
1456.
58 Skerker JM, Prasol MS, Perchuk BS, Biondi EG,
Laub MT: Two-component signal transduction
pathways regulating growth and cell cycle progression in a bacterium: a system-level analysis. PLoS
Biol 2005;3:e334.
59 Laub MT, Goulian M: Specificity in two-component
signal transduction pathways. Annu Rev Genet
2007;41:121145.
60 Sturme MH, Kleerebezem M, Nakayama J,
Akkermans AD, Vaugha EE, de Vos WM: Cell to
cell communication by autoinducing peptides in
Gram-positive bacteria. Antonie Van Leeuwenhoek
2002;81:233243.
61 Schaefer AL, Val DL, Hanzelka BL, Cronan JE Jr,
Greenberg EP: Generation of cell-to-cell signals in
quorum sensing: acyl homoserine lactone synthase
activity of a purified Vibrio fischeri LuxI protein.
Proc Natl Acad Sci USA 1996;93:95059509.
62 Byers DM, Gong H: Acyl carrier protein: structurefunction relationships in a conserved multifunctional protein family. Biochem Cell Biol 2007;85:
649662.
63 Schaefer AL, Greenberg EP, Oliver CM, Oda Y,
Huang JJ, Bittan-Banin G, Peres CM, Schmidt S,
Juhaszova K, Sufrin JR, Harwood CS: A new class of
homoserine lactone quorum-sensing signals. Nature
2008;54:595599.
64 Schaefer AL, Hanzelka BL, Eberhard A, Greenberg
EP: Quorum sensing in Vibrio fischeri: probing
autoinducer-LuxR interactions with autoinducer
analogs. J Bacteriol 1996;178:28972901.
65 Antunes LC, Ferreira RB, Lostroh CP, Greenberg
EP: A Mutational analysis defines Vibrio fischeri
LuxR binding sites. J Bacteriol 2008;190:4392
4397.
66 Egland KA, Greenberg EP: Conversion of the Vibrio
fischeri transcriptional activator, LuxR, to a repressor. J Bacteriol 2000;182:805811.
67 Liu X, De Wulf P: Probing the ArcA-P modulon of
Escherichia coli by whole genome transcriptional
analysis and sequence recognition profiling. J Biol
Chem 2004;279:1258812597.
68 Cho BK, Knight EM, Palsson BO: Transcriptional
regulation of the fad regulon genes of Escherichia
coli by ArcA. Microbiology 2006;152:22072219.

Engineering Bacterial Sensors

69 Matta MK, Lioliou EE, Panagiotidis CH, Kyriakidis


DA, Panagiotidis CA: Interactions of the antizyme
AtoC with regulatory elements of the Escherichia
coli atoDAEB operon. J Bacteriol 2007;189:6324
6332.
70 Avison MB, Horton RE, Walsh TR, Bennett PM:
Escherichia coli CreBC is a global regulator of gene
expression that responds to growth in minimal
media. J Biol Chem 2001;276:2695526961.
71 Janausch IG, Garcia-Moreno I, Lehnen D, Zeuner Y,
Unden G: Phosphorylation and DNA binding of the
regulator DcuR of the fumarate-responsive twocomponent system DcuSR of Escherichia coli.
Microbiology 2004;150:877883.
72 Wang H, Gunsalus RP: Coordinate regulation of the
Escherichia coli formate dehydrogenase fdnGHI and
fdhF genes in response to nitrate, nitrite, and formate: roles for NarL and NarP. J Bacteriol 2003;185:
50765085.
73 Partridge JD, Browning DF, Xu M, Newnham LJ,
Scott C, Roberts RE, Poole RK, Green J:
Characterization of the Escherichia coli K-12
ydhYVWXUT operon: regulation by FNR, NarL
and NarP. Microbiology 2008;154:608618.
74 Blauwkamp TA, Ninfa AJ: Physiological role of the
GlnK signal transduction protein of Escherichia coli:
survival of nitrogen starvation. Mol Microbiol 2002;
46:203214.
75 Claverie-Martin F, Magasanik B: Role of integration
host factor in the regulation of the glnHp2 promoter
of Escherichia coli. Proc Natl Acad Sci USA 1991;
88:16311635.
76 Simon G, Jourlin C, Ansaldi M, Pascal MC,
Chippaux M, Mejean V: Binding of the TorR regulator to cis-acting direct repeats activates tor operon
expression. Mol Microbiol 1995;17:971980.
77 Merkel TJ, Nelson DM, Brauer CL, Kadner RJ:
Promoter elements required for positive control of
transcription of the Escherichia coli uhpT gene. J
Bacteriol 1992;174:27632770.
78 Wright JS, Olekhnovich IN, Touchie G, Kadner RJ:
The histidine kinase domain of UhpB inhibits UhpA
action at the Escherichia coli uhpT promoter. J
Bacteriol 2000;182:62796286.
79 Rothenbucher MC, Facey SJ, Kiefer D, Kossmann
M, Kuhn A: The cytoplasmic C-terminal domain of
the Escherichia coli KdpD protein functions as a K+
sensor. J Bacteriol 2006;188:19501958.
80 Miyashiro T, Goulian M: Stimulus-dependent differential regulation in the Escherichia coli PhoQ
PhoP system. Proc Natl Acad Sci USA 2007;104:
1630516310.
81 Prost LR, Daley ME, Le Sage V, Bader MW, Le
Moual H, Klevit RE, Miller SI: Activation of the
bacterial sensor kinase PhoQ by acidic pH. Mol Cell
2007;26:165174.

223

82 Blanco AG, Sola M, Gomis-Ruth FX, Coll M:


Tandem DNA recognition by PhoB, a two-component signal transduction transcriptional activator.
Structure 2002;10:701713.
83 Leonhartsberger S, Huber A, Lottspeich F, Bock A:
The hydH/G Genes from Escherichia coli code for a
zinc and lead responsive two-component regulatory
system. J Mol Biol 2001;307:93105.
84 Lee LJ, Barrett JA, Poole RK: Genome-wide transcriptional response of chemostat-cultured Escherichia coli
to zinc. J Bacteriol 2005;187:11241134.
85 Nishino K, Honda T, Yamaguchi A: Genome-wide
analyses of Escherichia coli gene expression responsive to the BaeSR two-component regulatory system. J Bacteriol 2005;187:17631772.
86 De Wulf P, McGuire AM, Liu X, Lin EC: Genomewide profiling of promoter recognition by the twocomponent response regulator CpxR-P in Escherichia
coli. J Biol Chem 2002;277:2665226661.
87 Batchelor E, Walthers D, Kenney LJ, Goulian M:
The Escherichia coli CpxA-CpxR envelope stress
response system regulates expression of the porins
ompF and ompC. J Bacteriol 2005;187:57235731.
88 Yoshida T, Qin L, Egger LA, Inouye M: Transcription
regulation of ompF and ompC by a single transcription factor, OmpR. J Biol Chem 2006;281:17114
17123.
89 Nishino K, Inazumi Y, Yamaguchi A: Global analysis of genes regulated by EvgA of the two-component regulatory system in Escherichia coli. J Bacteriol
2003;185:26672672.
90 Masuda N, Church GM: Regulatory network of acid
resistance genes in Escherichia coli. Mol Microbiol
2003;48:699712.
91 Hagiwara D, Sugiura M, Oshima T, Mori H, Aiba H,
Yamashino T, Mizuno T: Genome-wide analyses
revealing a signaling network of the RcsC-YojNRcsB phosphorelay system in Escherichia coli. J
Bacteriol 2003;185:57355746.
92 Vianney A, Jubelin G, Renault S, Dorel C, Lejeune P,
Lazzaroni JC: Escherichia coli tol and rcs genes participate in the complex network affecting curli synthesis. Microbiology 2005;151:24872497.
93 Keseler IM, Collado-Vides J, Gama-Castro S,
Ingraham J, Paley S, Paulsen IT, Peralta-Gil M, Karp
PD: EcoCyc: a comprehensive database resource for
Escherichia coli. Nucleic Acids Res 2005;33:D334
D337.
94 Puskas A, Greenberg EP, Kaplan S, Schaefer AL: A
quorum-sensing system in the free-living photosynthetic bacterium Rhodobacter sphaeroides. J Bacteriol
1997;179:75307537.

224

95 Von Bodman SB, Ball JK, Faini MA, Herrera CM,


Minogue TD, Urbanowski ML, Stevens AM: The
quorum-sensing negative regulators EsaR and
ExpR(Ecc), homologues within the LuxR family,
retain the ability to function as activators of transcription. J Bacteriol 2003;185:70017007.
96 Pesci EC, Pearson JP, Seed PC, Iglewski BH:
Regulation of las and rhl quorum sensing in Pseudomonas aeruginosa. J Bacteriol 1997;179:31273132.
97 Urbanowski ML, Lostroh CP, Greenberg EP:
Reversible acyl-homoserine lactone binding to purified Vibrio fischeri LuxR protein. J Bacteriol 2004;
186:631637.
98 White CE, Winans SC: Cell-cell communication in
the plant pathogen Agrobacterium tumefaciens.
Philos Trans R Soc Lond B Biol Sci 2007;362:1135
1148.
99 Milton DL, Hardman A, Camara M, Chhabra SR,
Bycroft BW, Stewart GS, Williams P: Quorum sensing in Vibrio anguillarum: characterization of the
vanI/vanR locus and identification of the autoinducer N-(3-oxodecanoyl)-l-homoserine lactone. J
Bacteriol 1997;179:30043012.
100 Atkinson S, Chang CY, Sockett RE, Camara M,
Williams P: Quorum sensing in Yersinia enterocolitica controls swimming and swarming motility. J
Bacteriol 2006;188:14511461.
101 Levskaya A, Chevalier AA, Tabor JJ, Simpson ZB,
Lavery LA, Levy M, Davidson EA, Scouras A,
Ellington AD, Marcotte EM, Voigt CA: Synthetic
biology: engineering Escherichia coli to see light.
Nature 2005;438:441442.
102 Utsumi R, Brissette RE, Rampersaud A, Forst SA,
Oosawa K, Inouye M: Activation of bacterial porin
gene expression by a chimeric signal transducer in
response to aspartate. Science 1989;245:12461249.
103 Krikos A, Conley MP, Boyd A, Berg HC, Simon MI:
Chimeric chemosensory transducers of Escherichia
coli. Proc Natl Acad Sci USA 1985;82:13261330.
104 Baumgartner JW, Kim C, Brissette RE, Inouye M,
Park C, Hazelbauer GL: Transmembrane signalling
by a hybrid protein: communication from the
domain of chemoreceptor Trg that recognizes sugarbinding proteins to the kinase/phosphatase domain
of osmosensor EnvZ. J Bacteriol 1994;176:1157
1163.
105 Feng X, Baumgartner JW, Hazelbauer GL: Highand low-abundance chemoreceptors in Escherichia
coli: differential activities associated with closely
related cytoplasmic domains. J Bacteriol 1997;179:
67146720.
106 Weerasuriya S, Schneider BM, Manson MD:
Chimeric chemoreceptors in Escherichia coli: signaling properties of Tar-Tap and Tap-Tar hybrids. J
Bacteriol 1998;180:914920.

Salis Tamsir Voigt

107 Ward SM, Delgado A, Gunsalus RP, Manson MD: A


NarX-Tar chimera mediates repellent chemotaxis to
nitrate and nitrite. Mol Microbiol 2002;44:709719.
108 Repik A, Rebbapragada A, Johnson MS, Haznedar
JO, Zhulin IB, Taylor BL: PAS domain residues
involved in signal transduction by the Aer redox
sensor of Escherichia coli. Mol Microbiol 2000;36:
806816.
109 Kristich CJ, Glekas GD, Ordal GW: The conserved
cytoplasmic module of the transmembrane chemoreceptor McpC mediates carbohydrate chemotaxis in
Bacillus subtilis. Mol Microbiol 2003;47:13531366.
110 Yeh KC, Wu SH, Murphy JT, Lagarias JC: A
cyanobacterial phytochrome two-component light
sensory system. Science 1997;277:15051508.
111 Hubschmann T, Borner T, Hartmann E, Lamparter
T: Characterization of the Cph1 holo-phytochrome
from Synechocystis sp. PCC 6803. Eur J Biochem
2001;268:20552063.
112 Strauss HM, Schmieder P, Hughes J: Lightdependent dimerisation in the N-terminal sensory
module of cyanobacterial phytochrome 1. FEBS
Lett 2005;579:39703974.
113 Bystroff C, Krogh A: Hidden Markov models for
prediction of protein features. Methods Mol Biol
2008;413:173198.
114 Chenna R, Sugawara H, Koike T, Lopez R, Gibson
TJ, Higgins DG, Thompson JD: Multiple sequence
alignment with the Clustal series of programs.
Nucleic Acids Res 2003;31:34973500.
115 Kwon O, Georgellis D, Lin EC: Rotational on-off
switching of a hybrid membrane sensor kinase TarArcB in Escherichia coli. J Biol Chem 2003;278:
1319213195.
116 Skerker JM, Perchuk BS, Siryaporn A, Lubin EA,
Ashenberg O, Goulian M, Laub MT: Rewiring the
specificity of two-component signal transduction
systems. Cell 2008;133:10431054.
117 Arnold FH, Volkov AA: Directed evolution of biocatalysts. Curr Opin Chem Biol 1999;3:5459.
118 Zhao H: Directed evolution of novel protein functions. Biotechnol Bioeng 2007;98:313317.

119 Collins CH, Leadbetter JR, Arnold FH: Dual selection enhances the signaling specificity of a variant
of the quorum-sensing transcriptional activator
LuxR. Nat Biotechnol 2006;24:708712.
120 Collins CH, Arnold FH, Leadbetter JR: Directed
evolution of Vibrio fischeri LuxR for increased sensitivity to a broad spectrum of acyl-homoserine lactones. Mol Microbiol 2005;55:712723.
121 Kambam PK, Sayut DJ, Niu Y, Eriksen DT, Sun L:
Directed evolution of LuxI for enhanced OHHL
production. Biotechnol Bioeng 2008;101:263272.
122 Cheng J, Randall AZ, Sweredoski MJ, Baldi P:
SCRATCH: a protein structure and structural feature prediction server. Nucleic Acids Res 2005;33:
W72W76.
123 Endy D: Foundations for engineering biology.
Nature 2005;438:449453.
124 Anderson JC, Clarke EJ, Arkin AP, Voigt CA:
Environmentally controlled invasion of cancer cells
by engineered bacteria. J Mol Biol 2006;355:619
627.
125 Ninfa AJ, Selinsky S, Perry N, Atkins S, Xiu Song Q,
Mayo A, Arps D, Woolf P, Atkinson MR: Using twocomponent systems and other bacterial regulatory
factors for the fabrication of synthetic genetic
devices. Methods Enzymol 2007;422:488512.
126 Bulter T, Lee SG, Wong WW, Fung E, Connor MR,
Liao JC: Design of artificial cell-cell communication
using gene and metabolic networks. Proc Natl Acad
Sci USA 2004;101:22992304.
127 Haseltine EL, Arnold FH: Implications of rewiring
bacterial quorum sensing. Appl Environ Microbiol
2008;74:437445.
128 Basu S, Gerchman Y, Collins CH, Arnold FH, Weiss
R: A synthetic multicellular system for programmed
pattern formation. Nature 2005;434:11301134.
129 Balagadde FK, Song H, Ozaki J, Collins CH, Barnet
M, Arnold FH, Quake SR, You L: A synthetic
Escherichia coli predator-prey ecosystem. Mol Syst
Biol 2008;4:187.

Howard Salis
University of California, San Francisco
1700 4th St., Byers Hall MC 2540, Room 408C
San Francisco, CA 94158 (USA)
Tel. +1 415 514 9745, Fax +1 415 502 4690, E-Mail salish@picasso.ucsf.edu

Engineering Bacterial Sensors

225

Author Index

Ak, F. 88

Nait Abdallah, J. 88

Bazylinski, D.A. 182

Ordal, G.W. 33
Otto, M. 136

Davidson, C.J. 1
De Bolle, X. 88
Deghmane, A.-E. 88
Deutscher, J. 88
Dozot, M. 88
Duan, K. 1
Federle, M.J. 18
Frankel, R.B. 182
Jahreis, K. 65
Johansson, J. 150
Larribe, M. 88
Lengeler, J.W. 65
Letesson, J.J. 88

Poncet, S. 88
Rao, C.V. 33
Rmling, U. 161
Salis, H. 194
Sibley, C.D. 1
Simm, R. 161
Skaar, E.P. 120
Stauff, D.L. 120
Surette, M.G. 1
Taha, M.-K. 88
Tamsir, A. 194
Voigt, C. 194

Maz, A. 88
McIver, K.S. 103
Milohanic, E. 88

226

Subject Index

Active interactions, bacteria 2


Acylated homoserine lactone, Vibrio quorum
sensing 1921
A-factor 7
Allosteric transcription factors, sensors 195,
196
Antibiotics
environmental function 10
market 6
quorum sensing relationship 1012
sources 6
Antimicrobial peptides
resistance mechanisms 137140
sensing
anionic peptides and nonspecific sensing
145
Aps sensor 143145
PhoP/PhoQ sensor 140143
therapeutic targeting 145, 146
types and sequences 137, 138
Aps 143145
AraC 196
Autoinducer-2
4,5-dihydroxy-2,3-pentanedione in
signaling 2325
Escherichia coli 22, 23
identification 23
LsrB receptor 22
LuxP interactions 2730
LuxS metabolism discrimination
2527
prospects for study 30
Pseudomonas aeruginosa response
8, 9
Salmonella typhimurium 22, 23
transport 22, 23
Vibrio harveyi quorum sensing 22

Bacillus subtilis
carbon catabolite repression control 9092
chemotaxis
adaptation
CheC/CheD/CheTp system 4750
CheV system 47
integration of systems 52, 53
methylation system 5052
excitation
attractants 39, 40
paradigm 35
phosphatases 45
protein types 36
receptor structure 3538
receptor-adaptor protein-kinase
complex 37
repellents 40, 41
signaling array and receptor crosstalk
4345
transmission of attractant signals
4143
Bordetella pertussis, heme sensing 131
Carbon catabolite repression
carbohydrate availability and virulence 92
control
enterobacteria 89, 90
Firmicutes 9092
overview 89
virulence effects
Firmicutes 9296
proteobacteria 9699
Cathelicidins, see Antimicrobial peptides
Caulobacter cresentus, cyclic diguanylic acid
signaling 175
CcpA 9396
Che proteins 3538, 42, 4553, 7577

227

Chemical manipulation 9
Chemotaxis
adaptation
Bacillus subtilis
CheC/CheD/CheTp system 4750
CheV system 47
integration of systems 52, 53
methylation system 5052
Escherichia coli 45, 46
excitation
attractants 39, 40
paradigm 35
phosphatases 45
protein types 36
receptor structure 3538
receptor-adaptor protein-kinase complex
37
repellents 40, 41
signaling array and receptor crosstalk
4345
transmission of attractant signals 4143
overview 33, 34
phosphoenolpyruvate-dependent
carbohydrate:phosphotransferase system
role 7478
prospects for study 53
ChrAS 129, 130
Cooperation, definition 4, 5
Corynebacterium diphtheriae, heme sensing
129131
CovRS 104, 105, 112, 113, 115
Cph1 212214
Crp 196
CsgD 170
CspA 154
Cyclic diguanylic acid
binding sites 167169
metabolizing enzymes
abundance in bacterial genomes 164,
165
biosynthesis 162, 163
degradation 162, 164
overexpression studies 174
regulation 165167
overview of signaling 161, 162, 173, 174
redundancy and specificity of signaling
174176
sessility-motility transition signaling
overview 169171
single cell effects in biofilms 171
virulence signaling 172, 173

228

Defensins, see Antimicrobial peptides


4,5-Dihydroxy-2,3-pentanedione, AI-2
signaling role 2325
EAL domain proteins 162, 164166, 174
EnvZ 212214
Enzyme I, see Phosphoenolpyruvatedependent carbohydrate:
phosphotransferase system
Enzyme II, see Phosphoenolpyruvatedependent carbohydrate:
phosphotransferase system
Escherichia coli
AI-2 22, 23
chemotaxis
adaptation 45, 46
excitation
attractants 39, 40
paradigm 35
phosphatases 45
protein types 36
receptor structure 3538
receptor-adaptor protein-kinase
complex 37
repellents 40, 41
signaling array and receptor crosstalk
4345
transmission of attractant signals
4143
phosphoenolpyruvate-dependent
carbohydrate:phosphotransferase system
81, 82
predator-prey system synthetic gene
network 218220
Fnr 196, 197
GGDEF domain proteins 162167, 174
Group A Streptococcus, see Streptococcus
pyogenes
HAI-1 21, 22
HAMP domain
chemotaxis 36, 37
hybrid sensor kinase engineering 209213
specifications 202
HD-GYP domain proteins 162, 164166
Hemerythrin, magneto-aerotaxis role 191, 192
Heme sensing, see Staphylococcus aureus
Histidine kinase
Enzyme I 70, 71, 9799

Subject Index

specifications 202, 204


Homoserine lactone, quorum sensing
206, 207, 218
HrtAB 122125
HssRs 126131
Light, sensors 200
Listeria monocytogenes
carbon catabolite repression and virulence
94, 95
RNA thermosensors 156, 157
LsrB 22
LuxP, see Autoinducer-2
LuxS, see Autoinducer-2
Magnetotaxis
bacteria features 182, 183
cellular magnetic dipole 185
magneto-aerotaxis
axial magneto-aerotaxis 187
deviations from models 190, 191
hemerythrin function 191, 192
overview 186
polar magneto-aerotaxis 186
magnetosomes 183, 184
models 185
McpB 39, 42, 44
McpC 39, 40
Mga regulon, see Streptococcus pyogenes
Mlc 78
Nra 110, 111
NtrC 216
Oral bacteria, physical interactions 12
OxyR 199
Passive interactions, bacteria 1, 2
Phenazine 11, 12
PhoP 140143
PhoQ 140143
Phosphoenolpyruvate-dependent
carbohydrate:phosphotransferase
system
carbohydrate transport and energy state
sensing roles 6769
carbon catabolite repression control
enterobacteria 89, 90
Firmicutes 9092
proteobacteria and virulence 9799
chemotaxis role 7478

Subject Index

Enzyme I
histidine protein kinase activity 70, 71,
9799
overview 67, 68
Enzyme II
carbohydrate transport and
chemoreception 69, 70
overview 67, 68, 89
functional overview 66, 67
glucose pathway components and control
71, 72
glucose sensing role 78
modeling of complex networks 82, 83
phosphoenolpyruvate-to-pyruvate ratio in
carbon and energy metabolism 71
prospects for study 83
targeting subunits
regulatory and sensory networks 7274
responses to changes in carbon
metabolism/quest for food 7882
PrfA 95
Pseudomonas aeruginosa
AI-2 response 8, 9
quorum sensing 7, 8, 11, 12
Pseudomonas fluorescens, adaptive radiation 2
QacA 138
Quorum sensing
antibiotic relationship 1012
directed evolution for system engineering
214, 215
molecules and pathways 6, 7
sensors 198
species specificity 9, 19
specifications
overview 206, 208
signal response 207, 208
signal synthesis 207
RALPs, see Streptococcus pyogenes
Redoxtaxis, mechanisms 188190
Regulatory system, definition 66
RelA 199
Response regulator, specifications 204206
Rgg 104, 105, 112114
RhuI 131
RhuR 131
Riboswitch, sensors 196
RivR 112
RNA thermosensors
bacterial pathogens 155157

229

RNA thermosensors (continued)


cis- versus trans-acting RNA 150, 151
cold sensors 154
eukaryotic thermosensors 154
prospects for study 158, 159
ROSE thermosensors 152
structural requirements 157, 158
types 151154
RofA 110
RopB 104, 105, 112114
Salmonella typhimurium
AI-2 22, 23
cyclic diguanylic acid and virulence
signaling 173
RNA thermosensors 155
Sensor kinase
hybrid sensor kinase engineering 209214
stimulus 201, 202
Sensory system, definition 66
Siderophore, scavenging 4, 5
Signal, definition 4
Staphylococcus aureus
heme
adaptive responses 121, 122
evolutionary implications of heme
sensing 131133
HrtAB transporter 122125
HssRs sensor system 126131
metabolism 121
pathogenesis 120, 121
quorum sensing interference 9, 10
Starvation, sensors 198, 199
Streptococcus pneumoniae, carbon catabolite
repression and virulence 95, 96
Streptococcus pyogenes
carbon catabolite repression and virulence
92, 93
clinical features of infection 104
virulence gene expression
Mga regulon
carbohydrate utilization and
metabolism regulation 107
core virulence regulon 106, 107
DNA binding and transcriptional
regulation 107, 108
orthologs 108, 109
overview 105, 106
pathogenesis role 108
RALPs
Nra 110, 111

230

overview 109, 110


Ralp3 111, 112
RivR 112
RofA 110
Rgg/RopB regulon
DNA binding and transcriptional
activation 114
metabolism regulation 114
overview 112, 113
pathogenesis role 113
stand-alone response regulators
interactions 114, 115
overview 104, 105
two-component signal transduction
systems 104
Stress, sensors 198, 199
Synthetic gene networks, engineering 215220
Thermosensing, see RNA thermosensors
Thermotoga maritima, Tm1143 38
Thrombocidins, see Antimicrobial peptides
Tsr 38
Two-component system
heme sensing 125127
hybrid sensor kinase engineering 210
sensors 197
specifications 200, 201, 203
Streptococcus pyogenes virulence gene
expression 104
Vibrio fischeri, quorum sensing 19, 198
Vibrio harveyi
LuxP-AI-2 interactions 2730
quorum sensing 1923
Yersinia pestis, RNA thermosensors 155

Subject Index