Quorum sensing: dynamic response of Pseudomonas aeruginosa to ...

Update

TRENDS in Microbiology

Vol.14 No.2 February 2006

Research Focus

Quorum sensing: dynamic response of Pseudomonas aeruginosa to external signals Victoria E. Wagner, John G. Frelinger, Richard K. Barth and Barbara H. Iglewski Department of Microbiology and Immunology, University of Rochester, Rochester, NY 14642, USA

A recent study suggests that the opportunistic pathogen Pseudomonas aeruginosa can actively monitor the host immune system. The P. aeruginosa outer membrane protein OprF was found to bind specifically to the cytokine interferon-g (IFN-g), and this interaction upregulated production of virulence factors through a cell–cell communication system known as quorum sensing (QS). Taken together with previous findings that P. aeruginosa QS can alter the host immune response (e.g. by activation of IFN-g), these data illustrate an exciting new element of bacteria–host interactions in which the P. aeruginosa quorum-sensing system both senses and modulates the host immune state.

Communication among bacteria and hosts Bacteria can communicate using diffusible molecules known as autoinducers to coordinate population activities. Many Gram-negative and Gram-positive microbes use this type of communication, termed quorum sensing (QS). Importantly, processes such as virulence factor production, biofilm development and antimicrobial resistance, which aid establishment and persistence of microorganisms in the environment and in the host, are regulated by QS. The synchronization of individual behavior by QS is thought to help bacteria evade host immune responses by upregulating virulence determinants only when a sufficient bacterial cell density has been achieved to overwhelm host defenses [1]. A major advance in immunology was the discovery that the immune system evolved ways of sensing viral or bacterial products. Perhaps the best studied of these sensing pathways are those that involve toll-like receptors that bind to viral or bacterial products such as lipopolysaccharide and are essential in activating antigenpresenting cells [2]. Such activation affects the innate immune response and is crucial in the subsequent adaptive immune response. A recent report by Wu et al. suggests that the opportunistic pathogen Pseudomonas aeruginosa actively monitors the host immune status and triggers virulence factor production in response to IFN-g [3]. This exciting report gives a new dimension to host– bacteria interactions. Although pathogens have long been known to develop strategies to evade the immune system, this might be the first example in which a regulatory Corresponding author: Iglewski, B.H. ([email protected]). Available online 6 January 2006 www.sciencedirect.com

system that probably evolved initially to sense environmental factors has since been harnessed to sense host immune system status. Here, we discuss this discovery in the context of the current understanding of QS in P. aeruginosa and highlight further avenues of investigation.

Quorum-sensing regulon in Pseudomonas aeruginosa P. aeruginosa can cause severe disease in immunocompromised hosts such as those with burns, AIDS or cystic fibrosis (CF). These infections are difficult to treat using conventional antibiotic therapies and often result in considerable morbidity and mortality. P. aeruginosa has two well-studied QS systems: the las and rhl systems, which are important in its pathogenesis, as shown using various animal models of infection [4]. Each system consists of a transcriptional regulatory protein (R-protein) – LasR or RhlR – and a cognate autoinducer signal molecule. N-(3-oxododecanoyl) homoserine lactone (3O–C12–HSL), which is produced by the synthase lasI, is the autoinducer molecule in the las system, whereas N-butyryl homoserine lactone (C4–HSL) is produced by rhlI in the rhl system. Once a threshold intracellular concentration of autoinducer is reached, the R-protein binds to its cognate autoinducer molecule to form a complex that modulates target gene expression [5]. Many of the virulence factors produced by P. aeruginosa, including secreted factors and cell-associated factors, are regulated by the las and rhl QS systems (Table 1). By monitoring expression of the rhl-regulated virulence factors, type-I P. aeruginosa lectin (lecA) and pyocyanin, Wu et al. made a startling observation: upregulation of lecA was substantially enhanced in the presence of exogenous interferon-g (IFN-g) [3]. This effect was abrogated in strains of P. aeruginosa PAO1 with mutations in rhlR or rhlI, suggesting that the rhl QS system is crucial in mediating the bacterial response to exogenous IFN-g [3]. Interestingly, the effect was specific for IFN-g because other cytokines did not promote transcription of lecA or rhlI or production of C4–HSL and pyocyanin [3]. Further investigation showed that the major outer-membrane protein OprF specifically binds to IFN-g and that this triggers the rhl QS system in an unknown manner [3]. An important unresolved question is the role of the las system in transducing the signal, or signals, responsible for mediating IFN-g enhancement of the rhl system. The las and rhl systems are not independent and are

Update

56



The discovery by Wu et al. [3] adds another level of complexity to the QS regulon. In light of previous research, which identified that QS is regulated by environmental stimuli and by proteins that monitor the metabolic status of the cell, the observation that exogenous cytokines can stimulate the rhl QS system is not entirely unprecedented [6,7] (Table 2). Interestingly, the IFN-g enhancement of the rhl system was observed only at high cell densities during stationary-phase growth, suggesting that the effect is also subject to growth-phase regulation. There is a complicated relationship between the stationary phase sigma factor rpoS and the QS regulons [8–10]. Elucidation of the mechanism through which the binding of IFN-g to OprF promotes transcription of rhlI and lecA and the production of pyocyanin, perhaps by activation of currently known regulators of the rhl system, will be of great interest.

Table 1. Examples of QS-regulated virulence factors in Pseudomonas aerguinosaa Gene Predominately las regulated lasB toxA aprA rhlR Predominantly rhl regulated rhlAB lecA lecb hcnABC phzA1–G1, phzM, phzS a

Protein Elastase Exotoxin A Alkaline protease RhlR Rhamnosyltransferase chain A, B Pseudomonas type-I lectin Pseudomonas type-II lectin Hydrogen-cyanide synthase Pyocyanin

See Ref. [1] for greater detail on QS-regulated virulence factors.

intertwined in a hierarchical manner, with the las system exerting both transcriptional and post-translational control of the rhl system [1]. In addition, the Pseudomonas quinolone signal (PQS) is known to regulate the rhl system positively [1]. However, production of the PQS is activated by the las system [1]. Whether or not the observed activation of the rhl system by exogenous cytokines is caused by stimulation of the las system or by another regulatory pathway (or pathways) remains to be determined (Figure 1).

Immunomodulation of the immune system by Pseudomonas aeruginosa QS Previous data have demonstrated that the autoinducer molecules themselves have immunomodulatory effects, which suggests that P. aeruginosa can not only actively sense, as suggested by Wu et al., but can also tailor

+

IFN-γ

T cell

OprF

Outer membrane

OprM MexB MexA

Inner membrane ???

3O–C12–HSL

(b) (a)

lasR

+

LasR +

LasR

LasR

+

pqsR

C4–HSL

+

RhlR

rhlR

rhlI

pqsH PqsR

PqsR

(c)

lasI

rsaL

+

Pyocyanin LecA

+

+ RhlR

RhlR

+

PQS

pqsA pqsB pqsC pqsD pqsE TRENDS in Microbiology

Figure 1. Current model of QS and proposed mechanisms of IFN-g-mediated QS activation. The autoinducer 3O–C12–HSL (green ovals) is synthesized by the las regulon and is actively transported outside of the cellular membrane by the MexAB–OprM efflux pump [1]. T cells are stimulated by 3O–C12–HSL to produce IFN-g (pink circles) [11]. Binding of IFN-g to OprF transduces a signal that might upregulate the rhl QS system by (a) activation of the PQS system; (b) upregulation of the las system through unknown regulator(s); or (c) activation through a currently unknown pathway. Two or more pathways might contribute to mediation of the observed effect on the rhl system. Positive regulation is shown by plus symbol (C) and red arrows. www.sciencedirect.com

Update


Table 2. Regulation of Pseudomonas aeruginosa QS-regulonsa Stimulus or gene Description Environmental stimuli Iron (fur) Ferric uptake regulation protein Oxygen (anr, dnr) Transcriptional regulators Anr and Dnr Two-component response systems gacA Response regulator GacA algR2 (algQ) Alginate regulatory protein AlgR2 algR Alginate biosynthesis regulatory protein AlgR pprB Two-component response regulator Transcriptional regulators vfr Transcriptional regulator Vfr mvaT Transcriptional regulator Sigma factors rpoN Sigma-54 factor rpoS Sigma factor RpoS Additional factors ppk Polyphosphate kinase dksA Suppressor protein DksA rsmA Regulator of secondary metabolites a

See Ref. [7] for greater detail.

the host immune response [11]. Purified 3O–C12–HSL stimulates various host signaling pathways to inhibit or activate immune cell responses, whereas C4–HSL does not [11–13] (Box 1). Observation of the immunomodulatory effect of 3O–C12–HSL has led to the hypothesis that QS modulates T-cell responses away from a protective T-helper-1 (TH1) host response, thus promoting P. aeruginosa survival in vivo [14]. Importantly, the autoinducer 3O–C12–HSL activates IFN-g production in T cells [11–13]. With the data from Wu et al. [3], a cyclic network can be proposed to exist between the P. aeruginosa QS system and T cells, whereby the las QS system promotes production of IFN-g, which is then sensed by the P. aeruginosa cell and results in upregulation of the rhl system (Figure 1). Interestingly, a recent study determined that both 3O–C12–HSL and C4–HSL can penetrate and activate their respective LasR and RhlR chimeric transcription factors in monkey kidney COS-1 cells [15]. The resemblance of the autoinducer binding domain of R-proteins such as LasR to domains present in several mammalian signaling proteins and transcription factors suggests the existence of eukaryotic proteins that could interact with autoinducers but these have yet to be identified. These interactions could provide a means by which P. aeruginosa customizes host-cell activities through production and secretion of autoinducers. Evolutionary considerations and unanswered questions The findings that P. aeruginosa can both stimulate and sense IFN-g are intriguing ones. IFN-g is a cytokine with pleiotropic effects that is produced in large amounts by Box 1. Examples of effects of 3O–C12–HSL on host cells † Stimulates IFN-g production in T cells [11]. † Stimulates interleukin-8 production in human lung fibroblasts [11]. † Inhibits interleukin-12 and tumor-necrosis factor a in lipopolysaccharide-stimulated macrophages [13]. † Promotes immunoglobulin-E production in interleukin-4 stimulated monocytes [11]. † Causes apoptosis in macrophages and neutrophils [12]. www.sciencedirect.com


57

natural killer cells, CD8 T cells and TH1 cells [16]. It has several important functional properties including enhancement of major histocompatibility complex expression, activation of macrophages to kill intracellular bacteria and promotion of switching to particular isotypes in B cell antibody responses. IFN-g is an important cytokine in cell-mediated TH1 inflammatory responses and, thus, is an essential effector molecule for many pathogen responses [16]. If P. aeruginosa can sense IFN-g, it might indeed have crucial clinical consequences, as suggested by Wu et al. [3]. Intriguingly, Pseudomonas infections are more common in transplant and immunocompromised individuals. The sensing of IFN-g raises several interesting issues. First, why sense IFN-g and not other cytokines that are perhaps more involved in the innate immune response? Although IFN-g is an essential cytokine for many cellmediated responses, classically, Pseudomonas has been associated with deficiency of granulocytes (cells in an early innate inflammatory response), although this does not eliminate a potential role for T cells. Also, IFN-g might have a role in activating neutrophils and, in some cases, is produced by neutrophils. Therefore, IFN-g could be relevant in chronic inflammation such as that seen in CF patients [11,17]. Nevertheless, IFN-g might be expected to have a more important role in viral and intracellular bacterial infections such as mycobacteria rather than in infections of extracellular bacteria such as Pseudomonas [16]. Second, from an evolutionary perspective, it is interesting to consider how such recognition might evolve. It is appealing to propose that IFN-g sensing by Pseudomonas is a result of strong selection during infections of individuals. However, alternative hypotheses must be considered because Pseudomonas is an opportunistic pathogen and, as such, might be expected to be most influenced by evolutionary selective pressures found in its native environment (e.g. in soil and water) rather than those within an opportunistic environment (e.g. in immunocompromised individuals). Does this suggest that this recognition might be a fortuitous (or nonfortuitous) consequence of a system that has evolved for entirely different purposes? If so, this does not eliminate the potential clinical importance of the observation. However, it raises the possibility that a key to understanding how direct interaction with IFN-g arose in Pseudomonas might lie in the study of natural predators of this bacterium. Third, do other bacteria sense IFN-g or are other cytokines detected? OprF, which was demonstrated to bind specifically to IFN-g [3], is conserved across many bacterial species. The discovery of lectin-like activity of some cytokines that is mediated by portions of the cytokine molecule other than the classic receptor-binding domain is an intriguing area of further investigation [18]. Indeed, as with many important results, recent data on the interaction of quorum sensing and the immune system raise as many questions as they provide answers. Pursuing answers to these questions will help to unravel these complex bacteria–host interactions.

Update

58


Acknowledgements We thank John Looney, Melanie J. Filiatrault Muhlebach for helpful discussions and critical manuscript. This work was supported by an NIH (NIHR37AI37713). V.E.W. is supported by an NIH (5T32AI07285).

and Marianne review of the grant to B.H.I training grant

References 1 de Kievit, T.R. and Iglewski, B.H. (2000) Bacterial quorum sensing in pathogenic relationships. Infect. Immun. 68, 4839–4849 2 Janeway, C.A., Jr. and Medzhitov, R. (2002) Innate immune recognition. Annu. Rev. Immunol. 20, 197–216 3 Wu, L. et al. (2005) Recognition of host immune activation by Pseudomonas aeruginosa. Science 309, 774–777 4 Fuqua, W. et al. (1994) Quorum sensing in bacteria: the LuxR–LuxI family of cell density-responsive transcriptional regulators. J. Bacteriol. 176, 269–275 5 Whiteley, M. and Greenberg, E.P. (2001) Promoter specificity elements in Pseudomonas aeruginosa quorum-sensing-controlled genes. J. Bacteriol. 183, 5529–5534 6 Albus, A. et al. (1997) Vfr controls quorum sensing in Pseudomonas aeruginosa. J. Bacteriol. 179, 3928–3935 7 Lazdunski, A.M. et al. (2004) Regulatory circuits and communication in Gram-negative bacteria. Nat. Rev. Microbiol. 2, 581–592 8 Latifi, A. et al. (1996) A hierarchal quorum-sensing cascade in Pseudomonas aeruginosa links the transcriptional activators LasR and RhlR (VsmR) to expression of the stationary-phase sigma factor RpoS. Mol. Microbiol. 21, 1137–1146


9 Schuster, M. et al. (2004) The Pseudomonas aeruginosa RpoS regulon and its relationship to quorum sensing. Mol. Microbiol. 51, 973–985 10 Wagner, V.E. et al. (2003) Microarray analysis of Pseudomonas aeruginosa quorum-sensing regulons: effects of growth phase and environment. J. Bacteriol. 185, 2080–2095 11 Smith, R.S. and Iglewski, B.H. (2003) P. aeruginosa quorum-sensing systems and virulence. Curr. Opin. Microbiol. 6, 56–60 12 Tateda, K. et al. (2003) The Pseudomonas aeruginosa autoinducer N-3-oxododecanoyl homoserine lactone accelerates apoptosis in macrophages and neutrophils. Infect. Immun. 71, 5785–5793 13 Chhabra, S.R. et al. (2003) Synthetic analogues of the bacterial signal (quorum sensing) molecule N-(3-oxododecanoyl)-L-homoserine lactone as immune modulators. J. Med. Chem. 46, 97–104 14 Telford, G. et al. (1998) The Pseudomonas aeruginosa quorum-sensing signal molecule N-(3-oxododecanoyl)-L-homoserine lactone has immunomodulatory activity. Infect. Immun. 66, 36–42 15 Williams, S.C. et al. (2004) Pseudomonas aeruginosa autoinducer enters and functions in mammalian cells. J. Bacteriol. 186, 2281–2287 16 Schroder, K. et al. (2004) Interferon-g: an overview of signals, mechanisms and functions. J. Leukoc. Biol. 75, 163–189 17 Ethuin, F. et al. (2004) Human neutrophils produce interferon-g upon stimulation by interleukin-12. Lab. Invest. 84, 1363–1371 18 Beschin, A. et al. (2004) Functional convergence of invertebrate and vertebrate cytokine-like molecules based on a similar lectin-like activity. Prog. Mol. Subcell. Biol. 34, 145–163

0966-842X/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.tim.2005.12.002

Contact-dependent inhibition: bacterial brakes and secret handshakes E. Susan Slechta and Matthew A. Mulvey Pathology Department, Division of Cell Biology and Immunology, University of Utah School of Medicine, Salt Lake City, UT 84132-2501, USA

Interbacterial communication can be mediated by soluble secreted factors and direct cell–cell contact. Recently, Aoki et al. identified a new contact-dependent communication pathway by which strains of uropathogenic Escherichia coli can inhibit the growth of other microbes within a mixed population. Two novel gene products – CdiA and CdiB, which seem to be members of a two-partner secretion family with homologs in many pathogens – mediate this contact-dependent inhibition (CDI). A third gene product, CdiI, provides immunity to CDI, as does expression of either P or S pili. The interplay between CDI and immunity factors could directly affect the course of an infection and modulate both the dispersion and the chronic persistence of bacterial pathogens within the host.

Interbacterial communication Many bacteria have evolved sophisticated cell–cell communication systems that enable microbes to assess Corresponding author: Mulvey, M.A. ([email protected]). Available online 4 January 2006 www.sciencedirect.com

their own population density and the presence of competitors or potential mating partners [1–4]. These systems often rely on the accumulation of secreted molecules called autoinducers. Upon reaching a certain threshold concentration within the local environment, autoinducers alter the expression of quorum-sensing genes, which, in turn, can affect a myriad of bacterial functions, including spore formation, bioluminescence, conjugation, antibiotic production and virulence [4–6]. In addition, variants of autoinducing peptides secreted by some microbes can interfere with the growth, toxicity and overall virulence of other bacteria within a particular niche. An example of this has been described in Staphylococcus aureus; autoinducers from one variant of S. aureus can abrogate expression of virulence genes and extracellular proteins in a second variant [7], which might attenuate the ability of the latter to infect host tissues. This seemingly hostile mode of communication is not limited to autoinducers. For example, many Grampositive and Gram-negative bacteria can secrete colicins, which are toxins that can kill a narrow spectrum of nearby target cells expressing specific cell-surface receptors [8].