Staphylococcus aureus regulates the expression and production of the staphylococcal superantigen-like secreted proteins in a Rot-dependent manner

Abstract

Staphylococcus aureus overproduces a subset of immunomodulatory proteins known as the staphylococcal superantigen-like proteins (Ssls) under conditions of pore-mediated membrane stress. In this study we demonstrate that overproduction of Ssls during membrane stress is due to the impaired activation of the two-component module of the quorum-sensing accessory gene regulator (Agr) system. Agr-dependent repression of ssl expression is indirect and mediated by the transcription factor repressor of toxins (Rot). Surprisingly, we observed that Rot directly interacts with and activates the ssl promoters. The role of Agr and Rot as regulators of ssl expression was observed across several clinically relevant strains, suggesting that overproduction of immunomodulatory proteins benefits agr-defective strains. In support of this notion, we demonstrate that Ssls contribute to the residual virulence of S. aureus lacking agr in a murine model of systemic infection. Altogether, these results suggest that S. aureus compensates for the inactivation of Agr by producing immunomodulatory exoproteins that could protect the bacterium from host-mediated clearance.

Fig. 1. agr inactivation results in the increased production of Ssls

A) S. aureus strain Newman was grown to stationary phase in medium supplemented with indicated concentrations of gramicidin. Exoproteins were collected, precipitated, separated using SDS-PAGE, transferred to nitrocellulose, and indicated Ssls were detected by immunoblot. B) S. aureus strain Newman wildtype (WT) or the strain lacking hrtA (ΔhrtA) were grown to stationary phase in medium supplemented with hemin and the exoproteins were analyzed as in Panel A. C) Indicated strains were grown to stationary phase in medium with or without hemin. Exoproteins were collected, precipitated, separated using SDS PAGE, and stained with Coomassie blue. Arrowhead indicates Ssls 1–11. D) S. aureus strain Newman WT and an isogenic agr mutant strain were grown to stationary phase and the exoproteins were collected and processed as in Panel A.

Fig. 2. Membrane stress affects the activation of Agr

A) S. aureus strain Newman containing the P3_agr-yfp reporter plasmid was grown to stationary phase in medium supplemented with indicated concentrations of gramicidin and the YFP fluorescence of normalized cultures (OD₆₀₀) was measured. Values represent average of three independent experiments +/− standard deviation (S.D.). Asterisk (*) indicates statistically significant difference compared to no gramicidin as determined by Student's t test (P≤0.05). B) S. aureus strain Newman WT, ΔhrtA, and the Δagr mutant strains containing the P3_agr-yfp reporter were grown in media with or without 0.5 µM hemin, and fluorescence of normalized cultures (OD₆₀₀) was measured at stationary phase. Values represent average of three independent experiments +/− standard deviation (S.D.). Asterisk (*) indicates statistically significant differences as determined by Student's t test (P≤0.05). C) S. aureus strain Newman WT containing the P3_agr-yfp reporter was grown in media to mid-log phase, washed extensively with PBS, and resuspended in media containing increasing concentrations of culture supernatants collected from stationary phase cultures of WT, Δagr, or ΔhrtA+2 µM hemin as the AIP source. Values represent average of three independent samples +/− standard deviation (S.D.). Asterisk (*) indicates statistically significant differences as determined by Student's t test (P≤0.05).

Fig. 3. Rot is involved in ssl regulation

A) S. aureus strain Newman lacking agrA was transformed with an empty plasmid (p) or an agrA complementation plasmid (pagrA). The strains were grown to stationary phase and the exoproteins were collected, precipitated, separated using SDS-PAGE, and stained with Coomassie blue. Arrowhead indicates Ssls. B) S. aureus strain Newman lacking agr was transformed with a tetracycline-inducible RNAIII plasmid. The strain was grown in 0, 100, or 250 ng/mL tetracycline until it reached stationary phase. Exoproteins were collected, precipitated, separated using SDS-PAGE, and stained with Coomassie blue. Arrowhead indicates Ssls. C) Exoproteins from Panel B were also transferred to nitrocellulose and analyzed by immunoblot with anti-Ssl7 and Ssl9 antibodies. Proteins from whole cell lysates were immunoblotted with an anti-Rot antibody. D) Indicated S. aureus strains (Newman background) were grown to stationary phase and exoproteins processed and analyzed as in Panel B. E) Exoproteins from Panel D were also analyzed by immunoblot with anti-Ssl7 and Ssl9 antibodies. Proteins from whole cell lysates were immunoblotted with an anti-Rot antibody. F) Indicated S. aureus strains (Newman background) were grown to stationary phase and exoproteins were processed as in Panel A and immunoblotted with an anti-Ssl7 antibody, which detected Ssl7 and Protein A (SpA). Corresponding whole cell lysates were immunoblotted against Rot.

Fig. 4. Rot influences the activation of the ssl promoters

A) Schematic of the ssl locus in the S. aureus strain Newman chromosome. B) Northern blot analysis of ssl1 and 7 transcripts in 5 hour cultures of Newman WT, Δagr, and Δagr/ssl strains. C) Newman WT and the Δagr and Δagr/rot mutant strains were transformed with plasmids containing the indicated promoters driving gfp expression, and GFP fluorescence was monitored at the indicated time points. Values represent average of three independent experiments +/− S.D. D) 5’-RACE products were sequenced to define the putative transcript start sites and upstream alignment revealed nucleic acids shared by all sequences (gray) and highly conserved between sequences (light gray). Putative −35 and −10 promoter regions are indicated. Underlined region indicates the SaeR consensus sequence. E) Newman lacking agr, saeS, and agr/saeS were transformed with plasmids containing the indicated promoters driving gfp expression, and GFP fluorescence was monitored at the indicated time points. Values represent average of three independent experiments +/− S.D.

Fig. 5. Rot binds to the ssl promoters

A) Whole cell lysates from S. aureus strain Newman WT, Δagr, and the Δagr/rot double mutant grown to stationary phase were mixed with ssl9 promoter, ssl11 promoter, control DNA (DNA) conjugated to magnetic beads or beads alone. Captured proteins were separated using SDS-PAGE, transferred to nitrocellulose, and immunoblotted against Rot. B) Purified Rot was mixed with the ssl11 promoter conjugated to magnetic beads in the presence of increasing concentration of either unconjugated ssl11 promoter or control DNA (DNA) and Rot levels bound to the ssl11 promoter were detected by immunoblot as described in Panel A.

Fig. 6. Ssls are over-produced by agr defective clinical isolates

A) S. aureus USA300 containing the P3_agr-yfp reporter plasmid was grown to stationary phase in medium supplemented with indicated concentrations of gramicidin and YFP fluorescence was measured. Values represent average of three independent experiments +/− standard deviation (S.D.). Asterisk (*) indicates statistically significant difference compared to no gramicidin as determined by Student's t test (P≤0.05). B) USA300 WT was grown to stationary phase in media supplemented with indicated concentrations of gramicidin. Exoproteins were collected, precipitated, separated using SDS-PAGE, transferred to nitrocellulose, and analyzed by immunoblot with anti-Ssl1 and Ssl11 antibodies. C) USA300 and the indicated isogenic mutants were grown to stationary phase and exoproteins were processed as in Panel B, and analyzed by immunoblot with anti-Ssl 1, 9, and 11 antibodies. Corresponding whole cell lysates were immunoblotted against Rot. D) Strains described in Panel C were transformed with a plasmid containing the ssl11 promoter driving gfp expression, and GFP fluorescence was monitored at the indicated time points. Values represent average of three independent experiments +/− standard deviation (S.D.). Asterisk (*) indicates statistically significant difference compared to no gramicidin as determined by Student's t test (P≤0.05). E) Parental clinical isolate strains (WT), the naturally occurring isogenic agr defective strain (agr*), and the corresponding complemented strain (pagrC) were grown to stationary phase. Exoproteins were prepared as in panel B and immunoblotted using an anti-Ssl11 antibody. Corresponding whole cell lysates were immunoblotted against Rot.

Fig. 7. Disruption of ssls 1–11 reduced the bacterial burden of an agr mutant in vivo

A) S. aureus strain Newman WT, Δagr mutant, and the Δagr/ssl 1–11 double mutant were grown to stationary phase and the exoproteins were collected, precipitated, separated using SDS-PAGE, and stained with Coomassie blue. B) 10 mice each were systemically infected with 10⁷ CFU/mL of indicated strains. After four days, mice were sacrificed and bacterial burden determined by enumeration of CFUs in the hearts and kidneys. Asterisk (*) denotes statistical significance determined by Student’s t test (P≤0.05).