RNAIII-independent target gene control by the agr quorum-sensing system: insight into the evolution of virulence regulation in Staphylococcus aureus

Abstract

Cell-density-dependent gene regulation by quorum-sensing systems has a crucial function in bacterial physiology and pathogenesis. We demonstrate here that the Staphylococcus aureus agr quorum-sensing regulon is divided into (1) control of metabolism and PSM cytolysin genes, which occurs independently of the small regulatory RNA RNAIII, and (2) RNAIII-dependent control of additional virulence genes. Remarkably, PSM expression was regulated by direct binding of the AgrA response regulator. Our findings suggest that quorum-sensing regulation of PSMs was established before wide-ranging control of virulence was added to the agr regulon, which likely occurred by development of the RNAIII-encoding region around the gene encoding the PSM delta-toxin. Moreover, the agr regulon in the community-associated methicillin-resistant S. aureus MW2 considerably differed from that previously determined using laboratory strains. By establishing a two-level model of quorum-sensing target gene regulation in S. aureus, our study gives important insight into the evolution of virulence control in this leading human pathogen.

Figure 1. Model of target gene control by agr

The quorum-sensing circuit is shown at the top. The autoinducing peptide (AIP) with its characteristic thiolactone structure is produced from the AgrD precursor, modified and exported by AgrB. It activates the AgrC/AgrA two-component system. Target gene regulation, as determined by comparative qRT-PCR and genome-wide transcriptional profiling between S. aureus MW2 wild-type, agr deletion, and RNAIII deletion strains, is shown at the bottom. Targets regulated independently of RNAIII are shown at the left, which include several metabolic and psm genes (psmα and psmβ operons). The latter are regulated by AgrA binding to the respective psm promoter regions. RNAIII-dependent gene regulation is shown at the right. AgrA binds to the P2 and P3 promoters, activating the agr quorum-sensing feedback mechanism via RNAII expression, and RNAIII expression together with that of hld. Evolution of RNAIII around the hld mRNA likely has linked the RNAIII-dependent and independent parts of agr regulation. Many RNAIII-dependent agr targets contain a series of key virulence factors such as proteases and toxins. While regulation of many agr targets in an RNAIII-dependent fashion was shown in laboratory strains to occur via the SarA paralog Rot, the only SarA paralog regulated by agr in our study was SarH1.

Figure 2. Agr- and RNAIII-dependent gene regulation and PSM secretion

(A) Quantitative real-time PCR of selected agr-controlled target genes in S. aureus MW2 wild-type, and isogenic agr and RNAIII deletion strains. cDNA samples used were the same as for the microarray analyses, prepared from RNA isolated from cells grown to post-exponential growth phase (4 h). Probes were designed to align within the specific genes (MW0197) or within parts of the respective operons: MW0370 for the MW0370/MW0372 operon, psmβ1 for the psmβ operon, and parts of the psmα1 gene and the intergenic region between psmα1 and psmα2 for the psmα operon (See Supplemental Table 3 for probe sequences). Values are from 3 independent samples ± SEM. ***, p<0.001; **, p<0.01; *, p<0.05. (B) PSM concentration in culture filtrates of isogenic wild-type, agr, and RNAIII deletion strains. PSM amounts in 16-h culture filtrates were determined by RP-HPLC/ESI-MS of 3 independent samples. Values are ± SEM, relative to the parental MW2 (100%). *, no detectable amounts of any PSM. 0, below detection limit.

Figure 3. Binding of AgrA to psm promoter regions

(A) Promoter sequences of the psmα and psmβ loci. The ATG start codon of the first gene (psmα1, psmβ1) of the respective operon is marked in italic font and shaded in grey. Putative ribosomal binding sites (Shine-Dalgarno sequences, “SD”) are also shaded in grey. Putative AgrA binding sites in the psmα and psmβ promoter regions, according to comparison with the agr P2 and P3 promoters, are boxed and those found to be important for AgrA binding by EMSAs (panels B and C) are in blue. Fragments used for EMSAs are indicated by bars above the sequences (designated αS1, αS2, βS1, βS2, dark blue for fragments giving signals in the EMSAs, light blue for those without signal). Bases exchanged for mutated binding site analyses (panel C) are marked by a blue arrow and “G/C=>A/T”, describing the nature of the introduced mutation. The AgrA binding region, according to DNase footprinting analyses (Supplemental Figure 2), is shaded in light red. Start points of fragments used for cloning in psmα expression vectors (Figure 4A) are marked prom1, prom2, prom3 with an arrow (in purple). Transcription start sites were determined experimentally by 5′-RACE (in green, *). −10 and −35 regions, labeled in green, are deduced from the transcription start sites. (B) EMSAs of different fragments (S1, S2, see panel A) of psmα and psmβ promoters (25 pM) with AgrA (10 nM). The control fragment was amplified with the same oligonucleotides as used by Koenig et al. (Koenig et al., 2004). (C) EMSAs with natural and mutated binding sites of the psmα-S2 and the psmβ-S1 fragments, to which AgrA binding was demonstrated in (B), under corresponding experimental conditions. Shifts are marked. 50 mM acetyl phosphate was added to all EMSAs.

Figure 4. Control of psm expression by AgrA – translational analyses

(A) PSMα concentrations in culture filtrates of the MW2 psmα deletion strain complemented with the psmα locus under control of different lengths of the psmα promoter regions (see Figure 3A). PSM amounts in 16-h culture filtrates were determined by RP-HPLC/ESI-MS of 3 independent samples. Values are ± SEM. Maximal values obtained with the prom2 fragment were set to 100%. *, no detectable amounts of any PSMα. (B) Complementation of PSM expression in an agr mutant strain by agrA versus RNAIII. RNAIII or agrA were cloned in pTX15 (Peschel et al., 1996) for xylose-inducible control or pTX_Δ (Wang et al., 2007) for strong constitutive expression, and transformed in S. aureus 601055 mut6 (“agr”), which has a non-functional agr system. PSM production in the control strains S. aureus 601055 (pTX16) or S. aureus 601055 (pTX_Δ) was set to 100% (“wt”). Values for PSMα3 and δ-toxin are shown as examples. All pTX15 derivative strains were grown in TSB without glucose and with (“+”) or without (“−”) 0.5% xylose. All pTX_Δ derivative strains were grown in TSB. Expression of δ-toxin from the plasmid-encoded RNAIII is not shown. Plasmid-encoded is distinguishable from the chromosomally encoded δ-toxin in strain 601055 by its mass, because the plasmid-encoded MW2 RNAIII encodes a δ-toxin with a glycine to serine substitution at position 10 (Baba et al., 2002).