Control of Staphylococcus aureus Quorum Sensing by a Membrane-Embedded Peptidase

Abstract

Gram-positive bacteria process and release small peptides, or pheromones, that act as signals for the induction of adaptive traits, including those involved in pathogenesis. One class of small signaling pheromones is the cyclic autoinducing peptides (AIPs), which regulate expression of genes that orchestrate virulence and persistence in a range of microbes, including staphylococci, listeriae, clostridia, and enterococci. In a genetic screen for Staphylococcus aureus secreted virulence factors, we identified an S. aureus mutant containing an insertion in the gene SAUSA300_1984 (mroQ), which encodes a putative membrane-embedded metalloprotease. A ΔmroQ mutant exhibited impaired induction of Toll-like receptor 2-dependent inflammatory responses from macrophages but elicited greater production of the inflammatory cytokine interleukin-1β and was attenuated in a murine skin and soft tissue infection model. The ΔmroQ mutant phenocopies an S. aureus mutant containing a deletion of the accessory gene regulatory system (Agr), wherein both strains have significantly reduced production of secreted toxins and virulence factors but increased surface protein A abundance. The Agr system controls virulence factor gene expression in S. aureus by sensing the accumulation of AIP via the histidine kinase AgrC and the response regulator AgrA. We provide evidence to suggest that MroQ acts within the Agr pathway to facilitate the optimal processing or export of AIP for signal amplification through AgrC/A and induction of virulence factor gene expression. Mutation of MroQ active-site residues significantly reduces AIP signaling and attenuates virulence. Altogether, this work identifies a new component of the Agr quorum-sensing circuit that is critical for the production of S. aureus virulence factors.

Keywords: Agr; Staphylococcus aureus; accessory gene regulator; peptidase; peptide signaling; quorum sensing; virulence.

FIG 1

MroQ is a putative membrane-embedded protease that affects macrophage activation by S. aureus secreted factors. (A) Illustration of the gene arrangement surrounding mroQ and its proximity to the agr operon. nrd, nitroreductase family protein; sdrH, serine aspartate repeat family protein; hydrolase, carbon-nitrogen family hydrolase. (B) Phyre2 topology prediction analysis of MroQ within the membrane. Green boxes, predicted α-helices. (C) Amino acid sequence alignment of the EEXXXR and FXXXH motifs that comprise the type II CAAX protease active site among four predicted family members in S. aureus. (D) IL-1β, IL-6, and KC production by BMM after addition of supernatant from the WT, ΔmroQ, or ΔmroQ + mroQ strain grown in TSB. (Top) WT BMM; (bottom) MyD88^−/− BMM. The data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3). P values were determined by one-way ANOVA with Tukey’s posttest. ***, P < 0.001; ****, P < 0.0001.

FIG 2

MroQ is important for S. aureus skin and soft tissue infection. (A) Bacterial burden in skin abscesses of mice at 120 h postinfection with the WT (n = 8), ΔmroQ mutant (n = 8), or ΔmroQ + mroQ mutant (n = 8). P values were determined by a nonparametric one-way ANOVA (Kruskal-Wallis test) with Dunn’s posttest. *, P < 0.05; **, P < 0.01. (B) Representative images of skin abscesses at 120 h postinfection with the WT, ΔmroQ, and ΔmroQ + mroQ strains. (C) IL-6, KC, MCP-1, IL-1β, CCL3, and CCL4 levels in abscess homogenates of mice infected with the WT (n = 8), ΔmroQ mutant (n = 8), or ΔmroQ + mroQ mutant (n = 6). P values were determined by a nonparametric one-way ANOVA (Kruskal-Wallis test) with Dunn’s posttest. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; NS, not significant.

FIG 3

A ΔmroQ mutant resembles a Δagr mutant for protein secretion and perturbations in macrophage activation. (A) TCA-precipitated exoproteins from the WT, ΔmroQ, ΔmroQ + mroQ, Δagr, and Δagr ΔmroQ strains collected after growth in TSB. (B) IL-1β, IL-6, and KC production by BMM after addition of supernatant from the WT, ΔmroQ, ΔmroQ + mroQ, Δagr, and Δagr ΔmroQ strains grown in TSB. The data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3). ****, P < 0.0001 by one-way ANOVA with Tukey’s posttest.

FIG 4

MroQ contributes to Agr function. (A) LukA, LukS-PV, and Hla immunoblots of TCA-precipitated exoproteins from WT, ΔmroQ, ΔmroQ + mroQ, and Δagr strains. Protein A levels were detected based upon binding of anti-LukA antibody to protein A. MWM, molecular weight marker; NS, nonspecific band. (B) Rabbit red blood cell lysis of cell-free culture filtrates derived from WT, ΔmroQ, ΔmroQ + mroQ, and Δagr strains. (C) (Left) Bacterial burden in skin abscesses of mice at 120 h postinfection with the WT (n = 8), ΔmroQ (n = 8), ΔmroQ + mroQ (n = 8), and Δagr (n = 8) strains. P values were determined by a nonparametric one-way ANOVA (Kruskal-Wallis test) with Dunn’s posttest. *, P < 0.05; **, P < 0.01; ***, P < 0.001. (Right) Representative images of skin abscesses at 120 h postinfection with the WT, ΔmroQ, ΔmroQ + mroQ, and Δagr strains are shown. Hemolysis assay data are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3).

FIG 5

Interrogating the effects of MroQ on Agr peptide processing or signaling. (A) Model depicting the potential locations where MroQ might facilitate Agr function. We propose that MroQ functions at the level of the peptide-processing module (AgrBD) or at the level of the membrane-embedded histidine kinase (AgrC). (B) P3-gfp (pDB59) promoter activation (relative fluorescence units [RFU]/OD₆₀₀) in the ΔmroQ and Δagr strains upon addition of conditioned medium from the WT, ΔmroQ, ΔmroQ + mroQ, and Δagr strains. (C) P3-gfp/pDB59 promoter activation (RFU/OD₆₀₀) in the ΔmroQ, ΔagrB, and ΔmroQ ΔagrB strains upon addition of conditioned medium (CM) from the WT and Δagr strains. (D) Whole-cell lysates of the WT, ΔmroQ, ΔmroQ + mroQ, and Δagr strains constitutively expressing AgrD-6×His (pOS1-P_sarA-agrD-6×His) followed by immunoblotting with anti-His monoclonal antibody to detect full-length (unprocessed) AgrD. (E) P3-gfp/pDB59 promoter activation (RFU/OD₆₀₀) in ΔagrB and Δagr strains upon addition of conditioned medium from the Δagr and Δagr ΔmroQ strains constitutively expressing agrBD/pOS1-P_HELP-agrBD. ****, P < 0.0001 by one-way ANOVA with Tukey’s posttest (C) and a two-tailed t test (E). The data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3).

FIG 6

MroQ active-site residues are required for regulation of Agr by MroQ. (A) Amino acid sequences of the EEXXXR and FXXXH motifs that comprise the type II CAAX protease active site. The locations of site-directed amino acid substitutions are underlined. (B) Hla immunoblots of TCA-precipitated exoproteins from the WT, ΔmroQ, ΔmroQ + mroQ, ΔmroQ + mroQ(E141A), ΔmroQ + mroQ(E142A), ΔmroQ + mroQ(H180A), and Δagr strains. (C) Rabbit red blood cell lysis of cell-free culture filtrates derived from the WT, ΔmroQ, ΔmroQ + mroQ, ΔmroQ + mroQ(E141A), ΔmroQ + mroQ(E142A), ΔmroQ + mroQ(H180A), and Δagr strains. (D) LukA and LukS-PV immunoblots of TCA-precipitated exoproteins from the WT, ΔmroQ, ΔmroQ + mroQ, ΔmroQ + mroQ(E141A), ΔmroQ + mroQ(E142A), ΔmroQ + mroQ(H180A), and Δagr strains. Protein A levels were detected based upon the binding of anti-LukA antibody to protein A. MWM, molecular weight marker; NS, nonspecific band. (E) IL-1β, IL-6, and KC production by BMM after addition of supernatant from the WT, ΔmroQ, ΔmroQ + mroQ, ΔmroQ + mroQ(E141A), ΔmroQ + mroQ(E142A), ΔmroQ + mroQ(H180A), and Δagr strains grown in TSB. ****, P < 0.0001 by one-way ANOVA with Tukey’s posttest. The data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3).

FIG 7

MroQ amino acid substitution E141A is required for Agr activation and full virulence. (A) P3-gfp (pDB59) promoter activation (relative fluorescence units [RFU]/OD₆₀₀) in ΔmroQ and Δagr strains upon addition of conditioned medium from the WT, ΔmroQ, ΔmroQ + mroQ, ΔmroQ + mroQ(E141A), ΔmroQ + mroQ(E142A), ΔmroQ + mroQ(H180A), and Δagr strains. The data shown are from one of at least three experiments conducted in triplicate. Means ± SD are shown (n = 3). ****, P < 0.0001 by one-way ANOVA with Tukey’s posttest. (B) Bacterial burden in skin abscesses of mice at 5 days postinfection with the WT (n = 16), ΔmroQ (n = 16), ΔmroQ + mroQ (n = 14), ΔmroQ + mroQ(E141A) (n = 17) ΔmroQ + mroQ(E142A) (n = 17), ΔmroQ + mroQ(H180A) (n = 17), and Δagr (n = 16) strains. P values were determined by a nonparametric one-way ANOVA (Kruskal-Wallis Test) with Dunn’s posttest. *, P < 0.05; ***, P < 0.001.