Coagulase-Negative Staphylococcal Strain Prevents Staphylococcus aureus Colonization and Skin Infection by Blocking Quorum Sensing

Abstract

Coagulase-negative staphylococci (CoNS) and Staphylococcus aureus are part of the natural flora of humans and other mammals. We found that spent media from the CoNS species Staphylococcus caprae can inhibit agr-mediated quorum sensing by all classes of S. aureus. A biochemical assessment of the inhibitory activity suggested that the S. caprae autoinducing peptide (AIP) was responsible, and mass spectrometric analysis identified the S. caprae AIP as an eight-residue peptide (YSTCSYYF). Using a murine model of intradermal MRSA infection, the therapeutic efficacy of synthetic S. caprae AIP was evident by a dramatic reduction in both dermonecrotic injury and cutaneous bacterial burden relative to controls. Competition experiments between S. caprae and MRSA demonstrated a significant reduction in MRSA burden using murine models of both skin colonization and intradermal infection. Our findings indicate that important interactions occur between commensals that can impact disease outcomes and potentially shape the composition of the natural flora.

Keywords: AIP; CoNS; MRSA; Staphylococcus aureus; agr; coagulase-negative staphylococci; commensal; microbiome; quorum sensing.

Figure 1. S. caprae inhibits all classes of S. aureus quorum-sensing

A. Schematic of agr system operon. Staphylococcal autoinducing peptides (AIP) are processed and secreted in an AgrB-dependent process. Binding of the cognate AIP to AgrC results in phosphotransfer to AgrA, which induces transcription at the P2 and P3 promoters of the agr operon and RNAIII, respectively. Non-cognate staphylococcal AIPs can bind to AgrC and prevent signal transduction. B. Screen of clinical coagulase-negative staphylococcal (CoNS) isolates for inhibition of S. aureus agr type I–IV P3-GFP reporter. CoNS spent media was added to 10% final, and the 24 hr fluorescence time point is shown. Results are pooled from three experiments, each with N=4 replicates per condition. Two-way ANOVA with multiple comparisons (Dunnett’s correction) was performed. **** indicates p < 0.0001; *, p < 0.05.

Figure 2. Initial characterization and identification of the S. caprae quorum-sensing inhibitor

A. The S. caprae spent media was treated with 65 °C heat, proteinase K (1 mg/mL final), EGTA (5 mM final), or NaOH (to a pH of 11, then neutralized). Each sample was added to MRSA agr type I reporter to 10% final and compared to no treatment control. Florescence was measure at 24 hr, and the results are pooled from two experiments, each with N=3 replicates per condition and normalized to the media control. One-way ANOVA with multiple comparisons (Dunnett’s correction) was performed. **** indicates p < 0.0001. B. Hemolysis activity assay of MRSA wildtype and constitutive AgrC (R238H) treated with S. caprae spent media. Activity is expressed as the percent of MRSA spent media needed to achieve 50% rabbit blood cell lysis as calculated by a four-parameter logistic curve. Results are pooled from three experiments, each with N=3 replicates per condition. The pooled EC₅₀s were then analyzed by two-way ANOVA with each untreated vs. treated pair compared (Sidak multiple comparison test). ** indicates p < 0.01. C. The S. caprae agrBD genes were cloned into pEPSA5, and transformed into an S. aureus agr deletion mutant. Spent media from this strain, and additionally an empty vector control, were tested against a MRSA agr type I P3-YFP reporter. The six-hour time point is shown, and the results are pooled from two experiments, each with N=4 replicates per condition. One-way ANOVA with multiple comparisons (Dunnett’s correction) was performed. **** indicates p < 0.0001. D. Structure of S. caprae AIP with predicted y₇, y₆ and y₅ fragments. The AIP is an eight-residue peptide with the last five residues constrained as a thiolactone ring. E. Selected ion chromatogram within ± 5 ppm of m/z 1015.3871 for synthetic AIP standard (retention time 3.91, measured m/z 1015.3843). F. Selected ion chromatogram within ± 5 ppm of m/z 1015. 3871 for 30% acetonitrile/70% water fraction of S. caprae spent media (retention time 3.92, measured m/z 1015.3836).

Figure 3. Inhibition of S. aureus agr types I–IV with synthetic S. caprae AIPs

The agr P3-YFP reporters of S. aureus types I–IV (shown in panels A, B, C, and D) were treated with a dose-response of synthetic S. caprae AIP, and versions with one additional residue (AIP+1) and one less residue (AIP-1) than the native structure. Fluorescence was monitored throughout growth and fitted to a four-parameter logistic curve (GraphPad PRISM 7). One representative result is shown for each reporter (N=4 replicates per condition), and the entire experiment was repeated with similar results. Specific IC₅₀ values are reported in Table S2.

Figure 4. S. caprae AIP attenuates MRSA pathogenesis

A. Dermonecrotic lesion size following infection with MRSA WT +/− S. caprae AIP (n=4–8). Error bars represent SEM. Post test p value for every indicated time point =<0.0005 for comparisons of lesion size reduction resulting from both 5 μg and 10 μg AIP treatments. B. Weight loss and measurements following infection for the indicated groups (n=4–8). Error bars represent SEM. Post test p value (*)=<0.05, (**)=<0.01, (**)=<0.005, (***)=<0.001. C. Representative images of tissue injury following infection with MRSA WT +/− the indicated amounts of S. caprae AIP or agr null strain five days after infection. D. Representative images of in vivo bioluminescence induction one day after challenge with MRSA constitutive Lux +/− 10 μg of S. caprae-AIP or an Δagr MRSA Lux. E. Time course comparison of in vivo bioluminescence after intradermal challenge with MRSA Lux +/− 10 μg of S. caprae-AIP or an Δagr MRSA Lux (n=8–12). Error bars represent SEM. Post test p value (*)=<0.05, (**)=<0.01, (***)=<0.001.

Figure 5. S. caprae AIP-mediated quorum quenching in vivo corresponds with protection against MRSA-induced dermatopathology

A. Representative images of in vivo bioluminescence induction 2.5 hours after challenge with MRSA agr P3-Lux +/− 10 μg of S. caprae AIP. B. Time course comparison of in vivo bioluminescence after intradermal challenge with agr P3-Lux +/− 10 μg of S. caprae AIP (n=5). Error bars represent SEM. Post test p value (***)=<0.005. C. Dermonecrotic lesion size following infection with agr P3-Lux +/− 10 μg of S. caprae AIP (n=5). Error bars represent SEM. Post test p value (*)=<0.05, (**)=<0.01, (***)=<0.001. D. Left: representative images showing gross appearance of gross tissue injury five days after infection with agr P3-Lux +/− 10 μg of S. caprae AIP. Right: Representative H&E stained tissue sections prepared 5 days after infection agr P3-Lux +/− 10 μg of S. caprae AIP. Arrows denote distinct areas of tissue necrosis.

Figure 6. S. caprae reduces early MRSA burden following epicutaneous inoculation

A. Representative images of in vivo bioluminescence induction one day after epicutaneous administration of MRSA constitutive Lux alone or together with S. caprae. B. Representative images of tissue injury following epicutaneous application of the indicated groups 1 and 3 days after challenge. C. Time course comparison of in vivo bioluminescence after MRSA mono (MRSA alone) or co (MRSA/S. caprae) application (n=6). Error bars represent SEM. Post test p value (*)=<0.05.

Figure 7. S. caprae reduces early MRSA burden following intradermal inoculation

A. Representative images of in vivo bioluminescence induction 3.5 hours after challenge with MRSA agr P3-Lux +/− 10 μg of S. caprae AIP, or equal CFU of the CoNS S. caprae or S. capitis. B. Representative images show dermonecrosis 5 days after the bacterial challenge. C. Time course comparison of in vivo bioluminescence after intradermal challenge with within the indicated conditions (n=5–10). Error bars represent SEM. Post test p value (*)=<0.05, (**)=<0.01, (***)=<0.005. D. Time course of dermonecrotic lesion size in the indicated challenge conditions (n=5–10). Error bars represent SEM. Post test p value (*)=<0.05, (**)=<0.01.