Figure 1.
Savirin inhibits RNA III levels in S. aureus, but not S. epidermidis, without affecting agr-independent growth.
(A) Chemical structure of savirin (3-(4-propan-2-ylphenyl) sulfonyl-1H-triazolo [1,5-a] quinazolin-5-one). Effect of savirin (5 µg ml−1) vs vehicle control on (B) RNAIII levels induced by 50 nM AIP1 at 1 hr in MRSA strain USA300 LAC; (C) RNAIII levels in LAC without exogenous AIP1 at 5 hrs; (D) growth of LAC compared to growth of LAC Δagr; (E) RNAIII levels in S. epidermidis induced by overnight culture supernatant containing S. epidermidis AIP at 1 hr; and (F) growth of S. epidermidis. Data are represented as mean ± SEM, n = 3 experiments (B, C, D, & F) or n = 6 (E) performed in triplicate. ***p<0.001 **p<0.01, *p<0.05 by two-tailed Student's t-test.
Figure 2.
Savirin inhibits AgrA function in S. aureus both in vitro and within the organism.
(A) In silico docking of savirin to AgrAC from S. aureus. Space-filled representation of the C-terminal agrA DNA binding domain (gray) bound to target DNA. Surface residues that differ in S. epidermidis are blue, a naturally occurring dysfunctional mutation is shown in orange, and an essential Cys is shown in yellow. Savirin is shown in stick representation. An enlarged view of the boxed area shows the savirin docking site and surrounding residues. (B) Effect of increasing concentrations of savirin (5–160 µg ml−1 or 13.5–432 µM) vs vehicle on AgrAC-FAM labeled oligonucleotide complex formation by EMSA. The IC50 for savirin inhibition was quantified by densitometry of the bands. Data are representative of 3 independent experiments. (C) Effect of increasing concentrations of savirin (0.4–6.3 µM or 0.29–2.33 µg ml−1) vs vehicle on agrA reporter activation in an agr null strain expressing a plasmid for agrA where agr::P3 drives luminescence, AH3048, after 6 hr of growth. AIP2 as an inhibitor of non-agrII AgrC signaling was used as a specificity control. Viability is represented as OD600. Data are represented as the mean ± SEM of quadruplicates of a representative experiment of 3 independent experiments. ***p<0.001 **p<0.01, *p<0.05 by two-tailed Student's t-test.
Figure 3.
Savirin inhibits agr-regulated gene transcription and secreted virulence factor production.
(A) Graphic representation by category of microarray results in LAC after 5 hrs of culture with 50 nM AIP1 and 5 µg ml−1 of savirin. Of 205 agr-regulated transcripts, 122 were affected by savirin by 2 fold or greater and p<0.05, n = 3. All major virulence factors are represented; others shown are representative of the category. (B) Effect of savirin (5 µg ml−1) on transcription induced by 50 nM AIP1 in LAC for hla, psm alpha, and pvl relative to 16S at 1 hr and for agrA and agrC at 5 hr. (C) Effect of savirin (5 µg ml−1) on overnight alpha-hemolysin production by LAC. (D) Effect of savirin (5 µg ml−1) vs. vehicle on the capacity of culture supernatant from a clinical agr I blood stream isolate to lyse human neutrophils as measured by LDH release after 2 hr. Data are represented as the mean ± SEM, n = 3 independent experiments. ***p<0.001 **p<0.01, *p<0.05 by two-tailed Student's t-test.
Figure 4.
Savirin inhibits AgrA-dependent transcription in clinical isolates.
Effect of savirin (5 µg ml−1) vs. vehicle on psm alpha transcripts determined by qRT-PCR in clinical S. aureus isolates of each agr allele after 5 hr of culture. Data are represented as the mean of 5 replicates. Significance determined by two-way repeated measures ANOVA.
Figure 5.
Savirin promotes agr-dependent host defense in vivo and in vitro.
(A) Effect of 10 µg of savirin vs. vehicle on infection with 2×107 LAC AH 1677 (agr::P3 driving yfp) of Nox2−/− mice (n = 8 per group) in an air-pouch model 24 hrs after infection. Parameters shown include the fluorescence of bacteria in a lavage of the pouch, weight loss, and bacterial burden in the pouch and in the kidney. (B) Effect of 10 µg savirin vs vehicle on infection with 5×107 LAC Δagr of wild-type C57BL/6 mice (n = 4 per group) in an air-pouch model 24 hr after infection. Parameters shown include weight loss and bacterial burden in a lavage of the pouch and kidney. (C) Effect of savirin (5 µg) vs vehicle injected at the time of infection with 4×107 LAC agr+ or Δagr subcutaneously in the flank (n = 10–15 mice per group) of immunocompetent hairless SKH1 mice. Parameters shown include images of the infection sites at day 3; abscess area for agr+ infected mice; ulcer area for agr+ infected mice; weight loss over 3 days for agr+ infected mice; bacterial burden measured as CFU from the skin of agr+ infected mice at days 3 and 7; and bacterial burden measured as CFU from the spleen of agr+ infected mice at days 3 and 7. (D) Effect of delayed addition of savirin in vitro and in vivo. Parameters shown include the in vitro effect of savirin (5 µg ml−1) added at 3 hrs on RNAIII levels at 5 hr. **p<0.01 by two tailed Student's t test and savirin (5 µg) vs. vehicle injected 24 and 48 hrs after infection (n = 12–13 mice per group), depicting abscess area (arrows indicate timing of savirin/vehicle injection), ulcer size, bacterial burden in the skin at day 7, and bacterial burden in the spleen at day 7. All data from mouse infection represented as mean ±SEM ***p<0.001 **p<0.01, *p<0.05 by two-tailed Mann-Whitney U test.
Figure 6.
Effect of savirin treatment on in vitro host-dependent killing of LAC agr+ and Δagr.
(A) Percent intracellular survival of LAC agr+ (plus AIP1) or Δagr treated with savirin (5 µg ml−1) vs. vehicle for 5 hr prior to opsonization and phagocytosis by mouse macrophages (MOI 1∶1). Viable intracellular CFU set at 100% after internalization for 1 hr. Mean ± s.e.m., n = 3 independent experiments performed in triplicate. (B) Log CFU remaining of 1.0×108 LAC agr+ (plus AIP1) or Δagr treated with savirin (5 µg ml−1) vs. vehicle for 5 hr prior to incubation at pH 2.5 for 2 hr. Mean ± SEM, n = 6. ***p<0.001 **p<0.01, *p<0.05 by two-tailed Student's t-test.
Figure 7.
Passage of agr+ LAC with savirin in vivo or in vitro does not induce resistance or tolerance to savirin inhibition of agr signaling.
(A & B) In vivo passage of LAC sequentially through the skin of 10 individual mice for 24 hr in the presence of either 16 µg erythromycin and 0.12 µg clindamycin (A) or 5 µg savirin (B). (A) Percent survival after incubation overnight with 16 µg ml−1 erythromycin and increasing concentrations of clindamycin of non-passaged and passaged LAC, mean ± SEM, n = 3. (B) Percent AIP1 induced fold change in RNAIII after incubation for 1 hr with increasing concentrations of savirin in non-passaged or passaged LAC, mean ± SEM, n = 4–5. (C & D) In vitro passage serially for 10 days of LAC with either 16 µg ml−1 of erythromycin and 0.12 µg ml−1 clindamycin (C) or 5 µg ml−1 savirin (D). (C) Percent survival after incubation overnight with 16 µg ml−1 erythromycin and increasing concentrations of clindamycin of non-passaged and passaged LAC, mean ± SEM, n = 3. (D) Percent AIP1 induced fold change in RNAIII after 1 hr incubation with increasing concentrations of savirin in non-passaged and passaged LAC, mean ± SEM, n = 3. ***p<0.001 by two-tailed Student's t-test. (E & F) Assessment of savirin resistance at the colony level of LAC passaged in vivo with either antibiotics (E/C) or savirin. (E) In vivo passaged bacteria were plated on skim milk agar plates (diluted to give ∼15–20 colonies/plate) containing either vehicle or 10 µg ml−1 savirin for 72 hr. Non-passaged LAC Δagr is shown for comparison. Arrows are pointing to zones of proteolysis. The black bar is 5 mm. (F) Quantification of proteolytic and non-proteolytic colonies after 72 hr on milk agar plates containing either vehicle or savirin. Data are represented as mean ± SEM, n = 8 replicates of a representative experiment of two independent experiments.
Figure 8.
Model depicting the mechanism of action of savirin on S. aureus virulence.
Savirin blocks AgrA binding to its promoter sites subsequent to AIP secretion, binding, and signaling through AgrC. This blocking prevents the positive stimulation of P2 preventing increased AgrA and AgrC expression, the production of RNAIII which alters expression of multiple secreted virulence factors, and the direct promoter function of AgrA on PSM production.