Abstract

Staphylococcus aureus is an opportunistic pathogen. In response to changing host environments, this bacterium has the capability to switch on selective sets of genes to enhance its chances for survival. This switching process is precisely controlled by global regulatory elements. There are two major groups of global regulatory elements in S. aureus, including two-component regulatory systems (TCRSs) and the SarA protein family. Presumably, the sensor proteins of the 16 TCRSs in S. aureus provide external sensing, while the response regulators, in conjunction with alternative transcription factors and the SarA protein family, function as effectors within the intricate regulatory network to respond to environmental stimuli. Sequence alignment and structural data indicate that the SarA protein family could be subdivided into three subfamilies: (1) single-domain proteins; (2) double-domain proteins; and (3) proteins homologous to the MarR protein family. Recent data using reporter gene fusions in animal models, have confirmed distinct expression profiles of selected regulatory and target genes in vitro vs. in vivo.

Introduction

Staphylococcus aureus continues to be a major pathogen in both community and nosocomial bacteremias [1]. In most cases, the bacteremia is a result of a local infection that has gained access to the bloodstream. Once the bacteria are in the bloodstream, patients are at risk for developing endocarditis and other metastatic complications [1]. Although the use of newer antimicrobial agents has controlled some of these infections [2], the recent report of methicillin-resistant S. aureus (MRSA) strains with high-level resistance to vancomycin (i.e. via the vanA gene), the drug of last resort for many MRSA strains [3–5], underscores the need to understand the pathogenesis of this organism. By understanding the pathogenic process, it is hoped that new therapeutic strategies against S. aureus will emerge.

The pathogenicity of S. aureus is a complex process involving a diverse array of extracellular and cell wall components that are coordinately expressed during different stages of infection (i.e. colonization→avoidance of host defense→growth and cell division→bacterial spread). The coordinated expression of diverse virulence factors in response to environmental cues during infections (e.g. expression of adhesins early during colonization vs. production of toxins late in infection to facilitate tissue spread) hints at the existence of global regulators in which a single regulatory determinant controls the expression of many unlinked target genes. Genomic scans have revealed two major families of global regulators in S. aureus: (1) two-component regulatory systems (TCRS), of which there are 16 in the genome [6–8]; and (2) the SarA homologs, a family of proteins homologous to SarA, a global regulator of virulence factors [6].

In this review, we will provide insights into the pathogenesis of S. aureus, focusing in particular on the regulation of virulence determinants by the SarA protein family and, to a lesser extent, by the TCRS. As many of these global regulatory elements activate multiple virulence genes simultaneously, we believe that regulatory genes implicated in virulence should be viewed as gene subsets rather than as individual virulence determinants. Moreover, given the complexity of global regulatory pathways, it is reasonable to consider staphylococcal virulence as the net result of multiple overlapping and interacting feedback networks, rather than as a series of linear pathways. For a detailed review of TCRS and the agr system in particular, readers are referred to three recent reviews [6–8]. Predicated upon recent work on gene expression with the green fluorescent protein (GFP) reporter system and in vivo transcript analyses, we will also provide, where available, data on the comparative activation of regulatory and target virulence genes in vivo vs. in vitro.

Regulation of virulence determinants in vitro

The coordinated regulation of extracellular and cell wall virulence determinants during growth has hinted at the contribution of global regulatory elements in S. aureus infections. During the exponential phase, cell-wall proteins with adhesive functions are actively synthesized, coinciding with the tissue-binding and colonization phases of infection. These cell-wall proteins, also called MSCRAMMs (microbial surface components recognizing adhesive matrix molecules), include protein A, as well as fibrinogen-binding, fibronectin-binding, and collagen-binding proteins (Table 1). In transition from exponential to postexponential phase, the expression of cell-wall proteins is repressed, while the synthesis of extracellular toxins and enzymes predominates. By virtue of their proteolytic activities (e.g. V8 protease) and toxin effects on host cells (e.g. α-toxin), the exotoxins synthesized during the postexponential phase likely facilitate the local invasion and hematogenous dissemination phases of infections.

Table 1

Virulence determinants regulated by sarA and agr

 sarA agr 
Aureolysin (Metalloprotease) − 
Enterotoxin B 
Enterotoxin C unknown 
α-hemolysin 
β-hemolysin 
δ-hemolysin 
γ-hemolysin 
exfoliatins A and B unknown 
Fatty acid modifying enzyme 
FemA unknown 
FemB unknown 
HysA (hyaluronate lyase) unknown 
Lipase − 
Phospholipase C unknown 
Set8 (exotoxin) unknown 
Set9 (exotoxin) unknown 
SplA, B, D and F (proteases) 
SspB (cysteine protease) − unknown 
Staphylokinase unknown 
TSST-1 
V8 protease (SspA) − 
Clumping factor A unknown no effect 
Clumping factor B no effect 
Coagulase − 
Collagen-binding protein − ± 
Capsular polysaccharide (type 5) 
Fibronectin-binding protein A − 
Fibronectin-binding protein B − 
GyrA unknown − 
LrgAB 
PBP3 − unknown 
Protein A − − 
SdrC unknown 
Vitronectin-binding protein unknown − 
 sarA agr 
Aureolysin (Metalloprotease) − 
Enterotoxin B 
Enterotoxin C unknown 
α-hemolysin 
β-hemolysin 
δ-hemolysin 
γ-hemolysin 
exfoliatins A and B unknown 
Fatty acid modifying enzyme 
FemA unknown 
FemB unknown 
HysA (hyaluronate lyase) unknown 
Lipase − 
Phospholipase C unknown 
Set8 (exotoxin) unknown 
Set9 (exotoxin) unknown 
SplA, B, D and F (proteases) 
SspB (cysteine protease) − unknown 
Staphylokinase unknown 
TSST-1 
V8 protease (SspA) − 
Clumping factor A unknown no effect 
Clumping factor B no effect 
Coagulase − 
Collagen-binding protein − ± 
Capsular polysaccharide (type 5) 
Fibronectin-binding protein A − 
Fibronectin-binding protein B − 
GyrA unknown − 
LrgAB 
PBP3 − unknown 
Protein A − − 
SdrC unknown 
Vitronectin-binding protein unknown − 

The transition from exponential to postexponential protein expression has been shown to be coordinately controlled by global regulators such as sarA and agr[9] (Table 1) [8,10]. The sarA locus encodes a 372-bp open reading frame with three upstream promoters driving three overlapping transcripts, each coding for the 14.5-kDa SarA protein [11]. Phenotypically, the sarA locus promotes the synthesis of fibronectin- and fibrinogen-binding proteins (for adhesion) as well as that of α-, β-, and δ-toxins (for tissue spread) (Table 1). Recent studies by Valle et al. suggested that SarA might activate development of biofilm, probably by increasing transcription of ica operon which encodes enzymes responsible for the production of extracellular polysaccharide adhesin called PIA [12]. Complementation analysis has indicated that SarA is the major regulatory molecule of this locus, and mediates its effect both directly by binding to target gene promoters (e.g. agr, hla and spa) and indirectly via the downstream effect on other regulons [10,13]. Indeed, studies by Dunman and colleagues demonstrated that the sarA locus controls the expression of over 100 target genes [14]. The expression of SarA peaks during late exponential phase [15] and coincides with activation of another global regulator called agr during the postexponential phase, presumably via SarA–agr promoter interaction [10,16]. The SarA-binding site on the agr promoter has been mapped by two groups [10,13,16], with our group identifying a 29-bp sequence without any direct repeats between the two agr promoters (P2 and P2) [10] and another group locating the binding site to three regions, each consisting of two footprints of ∼18 bp separated by 4–5 bp [16]. In a more recent study, this latter group identified a 7-bp consensus binding sequence (ATTTTAT) in a promoter selection assay with purified SarA [17]; however, these studies were not buttressed by DNase I footprinting assays in vitro or by deletion analysis in intact S. aureus cells. Thus, it is difficult to assess the validity of the 7-bp SarA-binding site, especially in light of the AT-rich nature of the promoter regions in an already AT-rich genome such as that of S. aureus.

The agr locus comprises two divergent transcripts, RNAII and RNAIII, encoding agrDBCA and hld, respectively [18]. agrDBCA corresponds to a quorum-sensing TCRS that leads to the generation of an autoinducing peptide (AIP) that activates AgrC and AgrA, corresponding to the sensor and the activator of a TCRS, respectively [19]. Cell density-dependent accumulation of AIP triggers AgrC activation via phosphorylation [20]. The sequence variability of AIPs has led to the identification of four AIP groups, recognizing four different AgrC receptors in S. aureus[21]. Remarkably, there is crosstalk at the level of ligand–receptor signaling as most AIPs activate their native receptor while inhibiting non-native receptor activation [21]. As with many TCRS, activation of the sensor protein such as AgrC would result in a second step phosphorylation of AgrA, the response regulator, which then binds to the agr promoter to initiate RNAII and RNAIII transcription. Thus, the expression of RNAIII, the agr effector molecule, also responds to induction by AIP. The agr locus plays an essential role in up-regulating exoprotein gene expression (e.g. hla, seb) while down-regulating the synthesis of cell-surface adhesins during the transition from exponential to postexponential phase. Additionally, environmental and host signals generated during the postexponential phase would also modulate regulatory and target genes involved in the SarA–agr transition (Fig. 1). This transition likely involves additional gene products, including SarR [15], a SarA homolog, SigB [22], a stress-induced transcription factor and ArlRS [23], a TCRS involved in autolysis and virulence, to modulate SarA and agr expression during the postexponential phase [22]. In addition, a membrane protein SvrA [24] may also contribute to agr expression. Activation of agr would also repress the transcription of sarS and sarT, both sarA homologs [25–27]. sarT is a repressor of hla, while sarS is an activator of protein A synthesis. More recently, we found that sarT activates the transcription of sarS[28]. Therefore, agr may repress protein A expression by down-regulating sarT and subsequently sarS, while SarA may activate α-toxin production by down-regulating sarT. Based on the transcriptional profiling of saeRS[29] and rot[30], wherein sae is an activator and rot is a repressor of hla, we speculate that these two loci may mediate the interaction between agr and sarT. Whether and to what extent rot, sae and sarT interact with each other to modulate hla transcription is not currently defined. More recently, we found that sarU, a homolog of sarA, is an activator of agr, and is repressible by SarT [31]. Thus, the agr locus, besides being activated by AIP and SarA, may also be activated by a positive feedback loop via the sarT–sarU pathway. In this pathway, activation of agr represses sarT. Repression of SarT leads to activation of sarU, an activator of agr. This will lead to amplification of the original agr signal. This pathway is consistent with our recent observation that the agr RNAIII promoter can be activated in vivo in agr deletion mutants [32].

Figure 1

Regulation of virulence determinants in S. aureus by global regulatory loci. Normally, the synthesis of cell surface adhesins such as fibronectin-binding protein A during the exponential phase coincides with the expression of SarA and Sae, suggesting regulation of fnbA by these two loci. And in transition from exponential to postexponential phase, the synthesis of cell wall proteins is disrupted and the production of extracellular toxins such as α-toxin would begin. This transition corresponds to the maximal expression of SarA and the ensuing activation of agr. SarA expression is controlled by SarR, a SarA protein homolog, and SigB, a stress-induced transcription factor. On the other hand, agr is controlled by SarA, a quorum sensing AIP, a TCRS called ArlRS, MgrA/Rat/NorR and also a novel membrane protein called SvrA. Activation of agr would lead to up-regulation of another TCRS system called Sae and down-regulation of a SarA protein homolog called Rot. This will eventually lead to repression of two gene products called SarT and SarS. SarT is a repressor of α-toxin and SarS is an activator of protein A synthesis, thus explaining the elevated production of α-toxin and repression of protein A upon agr activation. Activation of agr would also result in the amplification of the original signal by activating SarU, which is a positive regulator of agr.

Figure 1

Regulation of virulence determinants in S. aureus by global regulatory loci. Normally, the synthesis of cell surface adhesins such as fibronectin-binding protein A during the exponential phase coincides with the expression of SarA and Sae, suggesting regulation of fnbA by these two loci. And in transition from exponential to postexponential phase, the synthesis of cell wall proteins is disrupted and the production of extracellular toxins such as α-toxin would begin. This transition corresponds to the maximal expression of SarA and the ensuing activation of agr. SarA expression is controlled by SarR, a SarA protein homolog, and SigB, a stress-induced transcription factor. On the other hand, agr is controlled by SarA, a quorum sensing AIP, a TCRS called ArlRS, MgrA/Rat/NorR and also a novel membrane protein called SvrA. Activation of agr would lead to up-regulation of another TCRS system called Sae and down-regulation of a SarA protein homolog called Rot. This will eventually lead to repression of two gene products called SarT and SarS. SarT is a repressor of α-toxin and SarS is an activator of protein A synthesis, thus explaining the elevated production of α-toxin and repression of protein A upon agr activation. Activation of agr would also result in the amplification of the original signal by activating SarU, which is a positive regulator of agr.

TCRS and the SarA protein family

There are two major families of regulatory elements that control the expression of virulence determinants in S. aureus. They include the TCRS and the SarA protein family. TCRS in prokaryotes generally entail a membrane sensor and a response regulator. Because the C-terminus of the sensor molecule and the N-terminal domain of the response regulator are usually conserved, Steve Gill at The Institute for Genomic Research (TIGR) has identified 16 TCRS in S. aureus by genome scanning, including six which have previously been studied (personal communication). The characterized systems include agrCA, saeRS, lytRS, arlRS, srrAB and yccFG[23,33–35]. Besides TCRS, a family of proteins homologous to SarA has also been found in the S. aureus genome (http://www.ncbi.nlm.nih.gov). We have termed this group of proteins ‘the SarA protein family’. Included in this family are five previously characterized SarA homologs (SarA, -R, -S, -T and -U), as well as four other as yet uncharacterized homologs [SarV (GI 13702067), SarX (GI 13700559), SarY (GI 13702097) and SarZ (GI 13702335)]. In addition, MgrA (also called Rat), a global regulator of autolysis and virulence (GI 13700577) [36] and two of its homologs (GI 13702065 and 13702470) also came up as hits in our search. Rot, a repressor of α-toxin synthesis [30], also shares homology with the smaller SarA homologs. In comparison to other smaller SarA homologs, Rot is larger in size (153 residues) and has an acidic pH. Alignment of SarS, SarU and SarY with other SarA homologs divulged that these three proteins are composed of two homologous domains, with each domain sharing sequence similarity to the smaller SarA homologs such as SarR and SarT (Fig. 2). We also aligned the smaller SarA homologs and found that SarZ, MgrA/Rat and its homologs are more homologous to MarR of Gram negative bacteria than to the SarA protein family (SarA, SarS and SarT). Based on sequence alignment and structures (see below), we propose to classify the SarA protein family into three subfamilies: (1) single-domain structures (SarA, SarR, SarT, SarV and SarX); (2) the two-domain structures (SarS, SarU and SarY); and (3) single-domain structures that are highly homologous to the MarR family (SarZ, MgrA/Rat and two of its homologs).

Figure 2

Alignment of SarA homologs of S. aureus. SarA, SarR, SarT and Rot belong to the single-domain proteins. Rot is unique for its larger size and an acidic pH. SarS and SarU represent two domain proteins, with two homologous halves connected by a linker region. SarS1 and SarS2 represent the N-terminal and C-terminal halves, respectively. Rat (also called MgrA/NorR) and SarZ are more homologous to the MarR protein family than to the SarA protein family and thus represent the third subfamilies within the SarA protein family. The HTH and the wing regions of the winged-helix structures are indicated.

Figure 2

Alignment of SarA homologs of S. aureus. SarA, SarR, SarT and Rot belong to the single-domain proteins. Rot is unique for its larger size and an acidic pH. SarS and SarU represent two domain proteins, with two homologous halves connected by a linker region. SarS1 and SarS2 represent the N-terminal and C-terminal halves, respectively. Rat (also called MgrA/NorR) and SarZ are more homologous to the MarR protein family than to the SarA protein family and thus represent the third subfamilies within the SarA protein family. The HTH and the wing regions of the winged-helix structures are indicated.

Sequence alignment has also been corroborated by crystal structure data. We have solved the crystal structures of SarR and SarS [37,38]. SarR is a dimeric structure, with each monomer comprising five α-helices, three β-strands and several loops (α1α2–β1α3α4–β2β3–α5). The SarR dimeric structure can also be visualized as a three-domain structure, with a central helical core and two winged-helix motifs. Within each winged-helix motif is a helix–turn–helix (HTH) motif (α3–α4) and a β-hairpin turn (β2–β3), both putative DNA-binding domains [37]. The overall structure is highly homologous to the winged-helix family of transcription factors as reported in eukaryotes [39]. In the middle of the SarR dimer is a canyon-like structure, outlined by a concave surface. Many basic residues (Lys and Arg), despite being randomly distributed throughout the SarR sequence, are located on the concave side, the putative DNA-binding side. An alignment of SarR with other SarA homologs revealed that many of these basic residues within the concave surface are conserved within the SarA protein family (Fig. 2). In contrast, residues implicated as part of the activation domain reside on the convex side and are less conserved.

More recently, SarS, a 250-residue DNA-binding protein that is an activator of protein A synthesis, was also found to be a winged-helix protein [38]. Contrary to SarA and SarR which form a homodimeric structure, SarS contains two homologous but non-identical halves, forming a winged-helix structure that is similar to the dimeric SarR structure from a single SarS molecule. Each of the two domains, comprising five α-helices and three β-strands, is connected by a well-ordered loop (Fig. 3). The degree of interaction between the homologous halves is more extensive than the homodimeric SarR structure. Similar to SarR, the putative DNA-binding surface, represented by a tract of positively charged residues, is located on the concave surface. However, in distinction to SarR, the acidic residues, noted on the opposite side of the putative DNA-binding surface, form a continuous, negatively charged tract that likely represents the activation motif.

Figure 3

Comparison of the SarR and SarS structures. The SarR structure is composed of two identical monomers, one in green and the other in yellow. In contrast, SarS is composed of two homologous but non-identical halves connected by a linker region. Both are winged-helix structures similar to the winged-helix proteins of eukaryotes.

Figure 3

Comparison of the SarR and SarS structures. The SarR structure is composed of two identical monomers, one in green and the other in yellow. In contrast, SarS is composed of two homologous but non-identical halves connected by a linker region. Both are winged-helix structures similar to the winged-helix proteins of eukaryotes.

We also obtained the preliminary structure of MgrA/Rat at 3.5× resolution (unpublished data). Although limited in fine details at this resolution, we were able to ascribe six α-helices and three β-strands to each of the monomers within the Rat dimeric structure. The overall structure of MgrA/Rat is topologically more similar to MarR than to SarR or SarS of the SarA protein family. Analogous to MarR [40], MgrA/Rat is a winged-helix structure with extensive coiled-coil interactions between two unique α-6 helices from each monomer. Thus, structural data have also confirmed the uniqueness of each of the three subfamilies within the SarA protein family as predicted by sequence alignment.

Based on the above data, it would appear that members of the SarA protein family have similar DNA-binding domains as evidenced by a positively charged tract on the concave surface of the dimeric structures. Modeling of SarR to target DNA suggests that the HTH binds to the major groove of DNA, while the β-hairpin turn interacts with the minor groove. The latter interaction is unique for the SarA protein family when compared with the mammalian winged-helix counterparts. In contrast, the activation domains among family members differ, with patches of acidic residues on SarR, and a continuous tract of negatively charged residues on SarS. Differences in the activation domain may partially explain the divergent functions among the SarA family members.

Predicated upon the structures, we speculated three possible mechanisms of regulation. First, the concave DNA-binding surface suggests that the target DNA may need to be foreshortened to fit into the aperture (71 Å for SarR) of the canyon-like structure to make contact with the DNA-binding surface (92 Å for SarR) [37]. Second, modeling of the SarA structure with a 135-bp DNA promoter fragment divulged the plausible binding of three SarA dimers to the agr promoter, with the three SarA dimers in the middle encircled by the target DNA, resulting in a closed configuration. This is reminiscent of the eukaryotic chromatin structure, with histone proteins encased by promoter DNA. We hypothesize that the closed configuration is not amenable to promoter activation while the open configuration, resulting from lesser interaction among the SarA dimers, is. Third, because a dimeric-like structure is necessary for the proper function of winged-helix proteins, the formation of a winged-helix structure by two heterologous halves of SarS hints at the possibility that heterodimer formation between compatible SarA family members (e.g. SarA and SarR) may occur to interfere with the function of the homodimer.

Regulation of virulence determinants in vivo

As noted above, the capacity of S. aureus to cause a multitude of human diseases (e.g. abscesses, endocarditis, and osteomyelitis) suggests that the pathogenesis of S. aureus infections is highly complex. It is very likely that distinct networks of multiple virulence genes are expressed in response to distinct host signals, including those found in blood and specific target tissues, and those related to innate host defense factors that emerge during the infectious process (e.g. PMNs, platelet antimicrobial peptides and chemokines, and macrophages). Recognizing that host factors likely contribute to, and modulate the expression of virulence determinants in vivo, recent studies have focused on defining the in vivo role of putative virulence factors such as adhesins, toxins and global regulons in relevant S. aureus infection models. Selected, salient and representative investigations are summarized below.

Many of the initial, seminal studies on the in vivo role of putative virulence loci in S. aureus utilized gene knockout strategies. For example, inactivation of either the sarA or agr loci has been shown to result in reduced virulence in several experimental staphylococcal infection models [41–45]. Of note, Cheung et al. showed that inactivation of the sarA locus, with or without concomitant inactivation of agr, led to a significant reduction in virulence of S. aureus in an experimental rabbit endocarditis model (i.e. decrease in both infectivity rates and intravegetation bacterial densities) [42,43]. Nilsson et al. [44] and Abdelnour et al. [41] have used a murine model of arthritis to study the role of sarA or agr in the induction and progression of septic arthritis. Similarly, Blevins et al. [45] have recently examined the impact of sarA on the capacity of S. aureus to cause septic arthritis and osteomyelitis in a murine model of musculoskeletal infection. Collectively, these latter investigations have shown that the agr, and especially the sarA loci are both likely activated in vivo, and contribute to the induction and progression of the experimental staphylococcal musculoskeletal infections. Similarly, important roles for clfA, fnbA, and hla in in vivo virulence of S. aureus have been confirmed in experimental endocarditis by either strategic gene knockout or adoptive gene transfer studies with Lactococcus lactis[46]. A detailed description of these investigations is beyond the scope of this review.

Although gene inactivation and adoptive gene transfer studies have identified a virulent role for global regulons and structural genes in vivo, until recently understanding the interplay and regulation of such putative virulence genes has not been possible. To understand the precise interaction among S. aureus virulence factors and host stimuli, a sensitive method to identify the temporal expression of staphylococcal genes in vivo was required. The recent development of differential fluorescent induction in which promoter fragments-of-interest are fused to a downstream promoterless GFP reporter gene has greatly facilitated detection of the activation of multiple S. aureus genes in vivo. Using this technology, a growing body of evidence underscores the notion that a number of gene-expression and gene-regulatory paradigms established in vitro are not precisely paralleled in vivo, using experimental endocarditis as a model system of multiorgan infection [32,47,48] (Table 2). For example, Cheung et al. [47] have shown that the sarA P2 promoter (which is weakly transcribed in vitro) and the sarA P1 promoter, are both well expressed in vivo at the surface of experimental cardiac vegetations; in contrast, the sarA P3 promoter remained relatively silent both in vitro and in vivo. As noted above, it is well documented that the full in vitro expression of agr RNAIII is highly dependent on the prior activation of agr RNAII. However, we recently showed that within infective vegetations in the experimental endocarditis model, agr RNAIII expression can be prompted within target tissues in animals infected with agr knockout mutants [32]. Further, van Wamel et al. [48] and others [49] have demonstrated that the optimal expression of capsular polysaccharide genes (cap) in vitro is co-regulated by both agr and sarA. In contrast, in the experimental infective endocarditis model, the agr locus is clearly the most critical for the in vivo expression of cap5[48]. Moreover, Goerke et al. [50] have suggested that the expression of the hla gene was inducible in vivo in the absence of a functional agr locus. Thus, two agr-deficient isolates were shown to express hla in vivo to near parental levels in a guinea pig model of subcutaneous device infection [50]. Lastly, Xiong et al. (American Society of Microbiology Annual Meeting May 2003) have recently studied the in vivo regulation of fnbA expression in experimental endocarditis. As expected, sarA enhanced, and agr repressed fnbA expression in vivo, paralleling the known regulatory paradigm in vitro. Interestingly, fnbA expression was enhanced during stationary growth phase in vitro, as well as in well-established infection in vivo, in the absence of both sarA and agr signals.

Table 2

Comparison of S. aureus virulence factor expression in vitro and in vivo in animal models

S. aureus promoter Class In vitro expression In vivo expression Animal model Reference 
sarA P1 global regulator + (log-phase) endocarditis [47] 
sarA P2 global regulator − endocarditis [47] 
sarA P3 global regulator − − endocarditis [47] 
Agr RNAIII global regulator + (stationary-phase) endocarditis; skin infection [32,51] 
agr RNAIII (agr mutants) global regulator − ± endocarditis [32] 
hla extracellular toxin + (stationary-phase) subcut. device infection; skin infection [50,51] 
hla (agr mutant) extracellular toxin − subcut. device infection; skin infection [50] 
cap5 surface exopolysaccharide + (stationary-phase); regulated by sarA and agr + regulated mainly by agr endocarditis, skin infection [48,51] 
fnbA surface adhesin + (exponential phase) regulated by sarA and agr + expressed in the presence or absence of sarA and agr endocarditis ASM 2003 
S. aureus promoter Class In vitro expression In vivo expression Animal model Reference 
sarA P1 global regulator + (log-phase) endocarditis [47] 
sarA P2 global regulator − endocarditis [47] 
sarA P3 global regulator − − endocarditis [47] 
Agr RNAIII global regulator + (stationary-phase) endocarditis; skin infection [32,51] 
agr RNAIII (agr mutants) global regulator − ± endocarditis [32] 
hla extracellular toxin + (stationary-phase) subcut. device infection; skin infection [50,51] 
hla (agr mutant) extracellular toxin − subcut. device infection; skin infection [50] 
cap5 surface exopolysaccharide + (stationary-phase); regulated by sarA and agr + regulated mainly by agr endocarditis, skin infection [48,51] 
fnbA surface adhesin + (exponential phase) regulated by sarA and agr + expressed in the presence or absence of sarA and agr endocarditis ASM 2003 

+ indicates detectable promoter expression with a good signal; − represents weak to non-detectable promoter expression; ± represents detectable promoter expression, but relatively weaker than in vitro signal.

Taken together, the above cumulative data on in vivo gene expression strongly implicate: (i) a major role of specific host micro-environmental signals in the in vivo activation of key S. aureus virulence genes; and/or (ii) the up-regulation of other regulatory genes in addition to sarA and agr that impact virulence gene networks (e.g. sae, or sarA homologs, as discussed above). Potentially of equal importance, the gene regulatory pathways impacted by sarA and agr in the above model systems, were frequently target-tissue specific. Thus, regulatory pathway impacts that were overtly seen within cardiac vegetations were not always operative in other target tissues in the same animal (e.g. kidney or spleen). These findings speak to potential impacts of micro-environmental factors (e.g. pH; osmolarity) and/or host niche-specific factors (e.g. presence of PMNs and/or platelets) on S. aureus gene expression in vivo. Recently, fibrinogen was identified as one of these host niche-specific factors, and shown to promote in vivo expression of RNAIII and two downstream targets of RNAIII, hla and cap5[51]. In this study, fibrinogen depletion prevented the in vivo activation of RNAIII; this resulted in significant reductions in tissue bacterial burden, morbidity and mortality during skin infection with wild-type, but not with agr-deficient bacteria. While the role of fibrinogen in promoting adherence of S. aureus to injured tissues and foreign bodies is well recognized, this was the first study to indicate that it provided an important in vivo signal for up-regulation of quorum sensing-dependent virulence gene expression. As in the endocarditis model [32], not all of the bacteria isolated from the various infected sites demonstrated equivalent virulence gene expression. Understanding how virulence genes are expressed in vivo may provide the basis for designing unique therapeutic strategies, such as new antimicrobial agents, targeted immunotherapies and preventive vaccines.

Outlook

The regulation of virulence determinants in S. aureus is complex. Contributing to this complexity are the growth phase dependency of virulence factors, the host and environmental signals, and the redundancy in adhesins, exotoxins and proteases. With the advance of genomic data, it is now clear that many of the virulence determinants in S. aureus are controlled by TCRS and/or the SarA protein family. Although the structure and function relationship of TCRS has been worked out in some details in prokaryotes, we still do not understand the precise regulatory mechanism(s) of proteins within the SarA protein family. While some of the pathways within the regulatory network have been elucidated, significant gaps remain in our knowledge in regard to the direct and indirect interactions among these regulatory elements. More importantly, we currently have very incomplete data-sets on the precise host and environmental signals that trigger the activation or repression of these regulatory genes, eventuating in alterations in target gene expression. With the availability of GFP variant [47] and Lux reporters [52], it is now feasible to monitor specific gene expression in vivo with fluorescence microscopy and real-time luminescence imaging [53]. These studies of gene expression in vivo should expedite our detailed understanding of how staphylococcal virulence genes are controlled by regulatory factors within the host. One is led to believe that not all regulators are created equal. Our future role will be to identify those regulators that serve as important ‘hubs’ in the central control of virulence gene expression. Animal studies support the notion that agr and SarA represent such network hubs. However, it is essential that we fully understand the intricacy of these regulators before we can target these regulatory elements for therapeutic interventions. Our overall goal is to disrupt the relevant regulatory networks in the organism, without incurring reciprocal genetic impacts on other regulatory pathways that might be eventually detrimental to the host.

References

[1]
Boyce
J.M.
(
1997
)
Epidemiology and prevention of nosocomial infections
. In:
The Staphylococci in Human Disease
  (
Crossley
K.B.
Archer
G.L.
, Eds.), pp.
309
329
.
Churchill Livingstone
,
New York
.
[2]
Lowy
F.
(
1998
)
Staphylococcus aureus infections
.
N. Engl. J. Med.
 
339
,
520
532
.
[3]
Chang
S.
Sievert
D.M.
Hageman
J.C.
Boulton
M.L.
Tenover
F.C.
Downes
F.P.
Shah
S.
Rudrik
J.T.
Pupp
G.R.
Brown
W.J.
Cardo
D.
Fridkin
S.K.
Vancomycin-resistant Staphylococcus aureus Investigative Team
, (
2003
)
Infection with vancomycin-resistance S. aureus containing vanA resistance gene
.
N. Engl. J. Med.
 
348
,
1342
1347
.
[4]
Sievert
D.M.
Boulton
M.L.
Stoltman
G.
Johnson
D.
Stobierski
B.G.
Downes
F.P.
Somsel
P.A.
Rudrik
J.T.
Brown
W.
Hafeez
W.
Lundstrom
T.
Flanagan
E.
Johnson
R.
Mitchell
J.
Chang
S.
(
2002
)
Staphylococcus aureus resistance to vancomycin — United States 2002
.
Morbidity and Mortality Weekly Reports
 
51
,
565
566
.
[5]
Hiramatsu
K.
Hanaki
H.
Ino
T.
Yabuta
K.
Oguri
T.
Tenover
F.C.
(
1997
)
Methicillin resistant Staphylococcus aureus clinical strain with reduced vancomycin susceptibility
.
J. Antimicrob. Chemother.
 
40
,
135
136
.
[6]
Cheung
A.L.
Zhang
G.
(
2002
)
Regulations of virulence determinants in Staphylococcus aureus by the SarA protein family
.
Front. Biosci.
 
7
,
D1825
D1842
.
[7]
Arvidson
S.
Tegmark
K.
(
2001
)
Regulation of virulence determinants in Staphylococcus aureus
.
Int. J. Med. Microbiol.
 
291
,
159
170
.
[8]
Novick
R.P.
(
2003
)
Autoinduction and signal transduction in the regulation of staphylococcal virulence
.
Mol. Microbiol.
 
48
,
1429
1449
.
[9]
Chan
P.F.
Foster
S.J.
(
1998
)
Role of SarA in virulence determinant production and environmental signal transduction in Staphylococcus aureus
.
J. Bacteriol.
 
180
,
6232
6241
.
[10]
Chien
Y.
Cheung
A.L.
(
1998
)
Molecular interactions between two global regulators, sar and agr, in Staphylococcus aureus
.
J. Biol. Chem.
 
237
,
2645
2652
.
[11]
Bayer
M.G.
Heinrichs
J.H.
Cheung
A.L.
(
1996
)
The molecular architecture of the sar locus in Staphylococcus aureus
.
J. Bacteriol.
 
178
,
4563
4570
.
[12]
Valle
J.
Toledo-Arna
A.
Berasain
C.
Ghigo
J.
Amorena
B.
Penades
J.
Lasa
I.
(
2003
)
SarA and not sigmaB is essential for biofilm development by Staphylococcus aureus
.
Mol. Microbiol.
 
48
,
1075
1087
.
[13]
Chien
C.-T.
Manna
A.C.
Projan
S.J.
Cheung
A.L.
(
1999
)
SarA, a global regulator of virulence determinants in Staphylococcus aureus, binds to a conserved motif essential for sar dependent gene regulation
.
J. Biol. Chem.
 
274
,
37169
37176
.
[14]
Dunman
P.M.
Murphy
E.
Haney
S.
Palacios
D.
Tucker-Kellogg
Q.
Wu
S.
Brown
E.L.
Zagursky
R.J.
Shlaes
D.
Projan
S.J.
(
2001
)
Transcriptional profiling based identification of S. aureus genes regulated by the agr and/or sarA loci
.
J. Bacteriol.
 
183
,
7341
7353
.
[15]
Manna
A.C.
Cheung
A.L.
(
2001
)
Characterization of sarR, a modulator of sar expression in Staphylococcus aureus
.
Infect. Immun.
 
69
,
885
896
.
[16]
Rechtin
T.M.
Gillaspy
A.F.
Schumacher
M.A.
Brennan
R.G.
Smeltzer
M.S.
Hurlburt
B.K.
(
1999
)
Characterization of the SarA virulence gene regulator of Staphylococcus aureus
.
Mol. Microbiol.
 
33
,
307
316
.
[17]
Sterba
K.M.
Mackintosh
S.G.
Blevins
J.S.
Hurlburt
B.K.
Smeltzer
M.S.
(
2003
)
Characterization of Staphylococcus aureus SarA binding sites
.
J. Bacteriol.
 
185
,
4410
4417
.
[18]
Kornblum
J.
Kreiswirth
B.
Projan
S.J.
Ross
H.
Novick
R.P.
(
1990
)
Agr: A polycistronic locus regulating exoprotein synthesis in Staphylococcus aureus
. In:
Molecular Biology of the Staphylococci
  (
Novick
R.P.
, Ed.), pp.
373
402
.
VCH Publishers
,
New York
.
[19]
Lyon
G.J.
Mayville
P.
Muir
T.W.
Novick
R.P.
(
2000
)
Rational design of a global inhibitor of the virulence response in Staphylococcus aureus, based in part on localization of the site of inhibition to the receptor-histidine kinase, AgrC
.
Proc. Natl. Acad. Sci. USA
 
97
,
13330
13335
.
[20]
Lina
G.
Jarraud
S.
Ji
G.
Greenland
T.
Pedraza
A.
Etienne
J.
Novick
R.P.
Vandenesch
F.
(
1998
)
Transmembrane topology and histidine protein kinase activity of AgrC, the agr signal receptor in Staphylococcus aureus
.
Mol. Microbiol.
 
28
,
655
662
.
[21]
Lyon
G.J.
Wright
J.S.
Christopoulos
A.
Novick
R.P.
Muir
T.W.
(
2002
)
Reversible and specific extracellular antagonism of receptor-histidine kinase signaling
.
J. Biol. Chem.
 
277
,
6247
6253
.
[22]
Cheung
A.L.
Chien
Y.T.
Bayer
A.S.
(
1999
)
Hyperproduction of alpha hemolysin in a sigB mutant is associated with elevated SarA expression in Staphylococcus aureus
.
Infect. Immun.
 
67
,
1331
1337
.
[23]
Fournier
B.
Klier
A.
Rapoport
G.
(
2001
)
The two-component system ArlS-ArlR is a regulator of virulence gene expression in Staphylococcus aureus
.
Mol. Microbiol.
 
41
,
247
261
.
[24]
Garvis
S.
Mei
J.M.
Ruiz-Albert
J.
Holden
D.W.
(
2002
)
Staphylococcus aureus svrA: a gene required for virulence and expression of the agr locus
.
Microbiology
 
148
,
3243
3253
.
[25]
Tegmark
K.
Karlsson
A.
Arvidson
S.
(
2000
)
Identification and characterization of SarH1, a new global regulator of virulence gene expression in Staphylococcus aureus
.
Mol. Microbiol.
 
37
,
398
409
.
[26]
Cheung
A.L.
Schmidt
K.A.
Bateman
B.
Manna
A.C.
(
2001
)
SarS, a SarA homolog repressible by agr, is an activator of protein A synthesis in S. aureus
.
Infect. Immun.
 
69
,
2448
2455
.
[27]
Schmidt
K.A.
Manna
A.C.
Gill
S.
Cheung
A.L.
(
2001
)
SarT: a repressor of alpha-hemolysin synthesis in Staphylococcus aureus
.
Infect. Immun.
 
69
,
4748
4758
.
[28]
Schmidt
K.A.
Manna
A.C.
Cheung
A.L.
(
2003
)
sarT influences sarS expression in Staphylococcus aureus
.
Infect. Immun.
 
71
,
5139
5148
.
[29]
Giraudo
A.T.
Cheung
A.L.
Nagel
R.
(
1997
)
The sae locus of Staphylococcus aureus controls exoprotein synthesis at the transcriptional level
.
Arch. Microbiol.
 
168
,
53
58
.
[30]
McNamara
P.J.
Milligan-Monroe
K.C.
Khalili
S.
Proctor
R.A.
(
2000
)
Identification, cloning, and initial characterization of rot, a locus encoding a regulator of virulence factor expression in Staphylococcus aureus
.
J. Bacteriol.
 
182
,
3197
3203
.
[31]
Manna
A.C.
Cheung
A.L.
(
2003
)
sarU, a sarA homolog, is repressed by SarT and regulates virulence genes in Staphylococcus aureus
.
Infect. Immun.
 
71
,
343
353
.
[32]
Xiong
Y.-Q.
van Wamel
W.
Nast
C.C.
Yeaman
M.R.
Cheung
A.L.
Bayer
A.S.
(
2002
)
Activation and transcriptional interaction between agr RNAII and RNAIII in Staphylococcus aureus in vitro and in an experimental endocarditis model
.
J. Infect. Dis.
 
186
,
668
677
.
[33]
Brunskill
E.W.
Bayles
K.W.
(
1996
)
Identification and molecular characterization of a putative regulatory locus that affects autolysis in Staphylococcus aureus
.
J. Bacteriol.
 
178
,
611
618
.
[34]
Yarwood
J.M.
McCormick
J.K.
Schlievert
P.M.
(
2001
)
Identification of a novel two-component regulatory system that acts in global regulation of virulence factors of Staphylococcus aureus
.
J. Bacteriol.
 
183
,
1113
1123
.
[35]
Fukuchi
K.
Kasahara
Y.
Asai
K.
Kobayashi
K.
Moriya
S.
Ogasawara
N.
(
2000
)
The essential two-component regulatory system encoded by yccF and yccG modulates ftsAZ operon in Bacillus subtilis
.
Microbiology
 
146
,
1573
1583
.
[36]
Ingavale
S.
van Wamel
W.
Cheung
A.L.
(
2003
)
Characterization of RAT, an autolysis regulator in Staphylococcus aureus
.
Mol. Microbiol.
 
48
,
1451
1466
.
[37]
Liu
Y.
Manna
A.C.
Li
R.
Martin
W.E.
Cheung
A.L.
Zhang
G.
(
2001
)
Crystal structure of the SarR protein from Staphylococcus aureus
.
Proc. Natl. Acad. Sci. USA
 
98
,
6877
6882
.
[38]
Li
R.
Manna
A.C.
Dai
S.
Cheung
A.L.
Zhang
G.
(
2003
)
Crystal structure of the SarS protein from Staphylococcus aureus
.
J. Bacteriol.
 
185
,
4219
4225
.
[39]
Gajiwala
K.S.
Chen
H.
Cornille
F.
Roques
B.P.
Reith
W.
Mach
B.
Burley
S.K.
(
2000
)
Structure of the winged-helix protein hRFX1 reveals a new mode of DNA binding
.
Nature
 
403
,
916
921
.
[40]
Alekshun
M.N.
Levy
S.B.
Mealy
T.R.
Seaton
B.A.
Head
J.F.
(
2001
)
The crystal structure of MarR, a regulator of multiple antibiotic resistance at 2.3 angstrom resolution
.
Nat. Struct. Biol.
 
8
,
710
714
.
[41]
Abdelnour
A.
Arvidson
S.
Bremell
T.
Ryden
C.
Tarkowski
A.
(
1993
)
The accessory gene regulator (agr) controls Staphylococcus aureus virulence in a murine arthritis model
.
Infect. Immun.
 
61
,
3879
3885
.
[42]
Cheung
A.L.
Eberhardt
K.J.
Chung
E.
Yeaman
M.R.
Sullam
P.M.
Ramos
M.
Bayer
A.S.
(
1994
)
Diminished virulence of a saragr mutant of Staphylococcus aureus in the rabbit model of endocarditis
.
J. Clin. Invest.
 
94
,
1815
1822
.
[43]
Cheung
A.L.
Yeaman
M.
Sullam
P.M.
Witt
M.D.
Bayer
A.S.
(
1994
)
The role of the sar locus of Staphylococcus aureus in the induction of endocarditis in rabbits
.
Infect. Immun.
 
62
,
1719
1725
.
[44]
Nilsson
I.
Bremell
T.
Rydén
C.
Cheung
A.L.
Tarkowski
A.
(
1996
)
The role of staphylococcal accessory regulator (sar) in septic arthritis
.
Infect. Immun.
 
64
,
4438
4443
.
[45]
Blevins
J.S.
Elasri
M.O.
Allmendinger
S.D.
Beenken
K.E.
Skinner
R.A.
Thomas
J.R.
Smeltzer
M.S.
(
2003
)
Role of sarA in the pathogenesis of Staphylococcus aureus musculoskeletal infection
.
Infect. Immun.
 
71
,
516
523
.
[46]
Que
Y.A.
Francois
P.
Haefliger
J.A.
Entenza
J.M.
Vaudaux
P.
Moreillon
P.
(
2003
)
Reassessing the role of Staphylococcus aureus clumping factor and fibronectin-binding protein by expression in Lactococcus lactis
.
Infect. Immun.
 
69
,
6296
6302
.
[47]
Cheung
A.L.
Nast
C.C.
Bayer
A.S.
(
1998
)
Selective activation of sar promoters with the use of green fluorescent protein (gfpuv) transcriptional fusions as the detection system in the rabbit endocarditis model
.
Infect. Immun.
 
66
,
5988
5993
.
[48]
van Wamel
W.
Xiong
Y.-Q.
Bayer
A.S.
Yeaman
M.R.
Nast
C.C.
Cheung
A.L.
(
2002
)
Regulation of Staphylococcus aureus type 5 capsular polysaccharides by agr and sarA in vitro and in an experimental endocarditis model
.
Microb. Pathog.
 
33
,
73
79
.
[49]
Luong
T.
Sau
S.
Gomez
M.
Lee
J.C.
Lee
C.Y.
(
2002
)
Regulation of Staphylococcus aureus capsular polysaccharide expression by agr and sarA
.
Infect. Immun.
 
70
,
444
450
.
[50]
Goerke
C.
Fluckiger
U.
Steinhuber
A.
Zimmerli
W.
Wolz
C.
(
2001
)
Impact of the regulatory loci agr, sarA and sae of Staphylococcus aureus on the induction of alpha-toxin during device-related infection resolved by direct quantitative transcript analysis
.
Mol. Microbiol.
 
40
,
1439
1447
.
[51]
Rothfork
J.
Dessus-Babus
S.
van Wamel
W.
Cheung
A.L.
Gresham
H.
(
2003
)
Fibrinogen depletion attenuates Staphylococcus aureus infection by preventing density-dependent virulence gene up-regulation
.
J. Immunol.
 
171
,
5388
5395
.
[52]
Qazi
S.N.
Counil
E.
Morrissey
J.
Rees
C.E.
Cockayne
A.
Winzer
K.
Chan
W.C.
Williams
P.
Hill
P.J.
(
2001
)
agr expression precedes escape of internalized Staphylococcus aureus from the host endosome
.
Infect. Immun.
 
69
,
7074
7082
.
[53]
Francis
K.P.
Joh
D.
Bellinger-Kawahara
C.
Hawkinson
M.J.
Purchio
T.F.
Contag
P.R.
(
2000
)
Monitoring bioluminescent Staphylococcus aureus infections in living mice using a novel luxABCDE construct
.
Infect. Immun.
 
68
,
3594
3600
.