Stress-induced inactivation of the Staphylococcus aureus purine biosynthesis repressor leads to hypervirulence

Abstract

Staphylococcus aureus is a significant cause of human infection. Here, we demonstrate that mutations in the transcriptional repressor of purine biosynthesis, purR, enhance the pathogenic potential of S. aureus. Indeed, systemic infection with purR mutants causes accelerated mortality in mice, which is due to aberrant up-regulation of fibronectin binding proteins (FnBPs). Remarkably, purR mutations can arise upon exposure of S. aureus to stress, such as an intact immune system. In humans, naturally occurring anti-FnBP antibodies exist that, while not protective against recurrent S. aureus infection, ostensibly protect against hypervirulent S. aureus infections. Vaccination studies support this notion, where anti-Fnb antibodies in mice protect against purR hypervirulence. These findings provide a novel link between purine metabolism and virulence in S. aureus.

Fig. 1

Disruption of purR causes cell clumping of S. aureus USA300. In a, representative images of USA300, USA300 purR::ΦNΣ or the complemented purR::ΦNΣ mutant in culture tubes following growth in TSB with 10% (v/v) horse serum (TSB-S) for 3.5 h from a starting OD₆₀₀ equivalent of 0.03. In b, graphical representation of the relative sedimentation of bacterial aggregates in cultures as grown in a, reflected by the OD₆₀₀ values of the centre of liquid cultures after sitting without shaking for 5 min following shaking at 37 °C for 3.5 h. Data are mean ± SEM of four independent experiments. ***p-value < 0.001, based on a one-way analysis of variance (ANOVA) with a Bonferroni post-test. In c, the representative micrographs show bacterial cell clusters that arise during growth in TSB or TSB-S. White boxes define the region of interest that is depicted in the insets. Scale bars  = 40 µm. In d, transmission electron micrographs are shown for S. aureus USA300 and the USA300 purR::ΦNΣ strain grown in the presence (TSB-S) or absence (TSB) of horse serum. The representative images depict cells at ×11,000 magnification. Scale bars  = 1 µm. Source data are provided as a Source Data file

Fig. 2

The purR-dependent clumping phenotype requires fibronectin binding proteins and host fibronectin. In a, cultures were grown in TSB or TSB-S for 3.5 h and then imaged on a wide field microscope at ×40 magnification. White boxes define the region of interest that is depicted in the insets. Scale bars = 40 µm. Representative images are shown. In b, cultures were grown as in a and OD₆₀₀ was measured as described in the legend to Fig. 1 and in the Methods section. Data shown are mean ± SEM of four independent experiments. ***p-value < 0.001, based on a one-way ANOVA with a Bonferroni post-test. In c, WT and the purR::ΦNΣ mutant were grown in TSB, TSB-S, TSB containing 10% v/v of various levels of Fn-depleted horse serum or Fn-depleted horse serum with the addition of eluted fibronectin (Fn depletion 3+Fn). Measurement of OD₆₀₀ of cultures to evaluate clumping was performed as described above. Data shown are mean ± SEM of five independent experiments and two different Fn purifications. ***p-value < 0.001, based on a one-way ANOVA with a Bonferroni post-test. In d, biofilm forming ability of indicated strains was measured after growth in TSB in a standard 96-well plate biofilm assay (see the Methods section). Data shown are mean ± SEM of four experiments. **p-value < 0.01, ***p-value < 0.001, based on a one-way ANOVA with a Bonferroni post-test. Source data are provided as a Source Data file

Fig. 3

purR mutations lead to transcriptional upregulation of the purine biosynthesis operon and fnbAB. a Consensus PurBox sequence for B. subtilis and L. lactis (data adapted from refs. ^,). b Promoter sequences of purE, purA and fnbA, with putative PurBox sequences underlined. WT (c) or purR::ΦNΣ mutant (d) containing a luciferase construct with the promoter sequence of fnbA or fnbB (see the Methods section) were grown in TSB and OD₆₀₀ and luminescence monitored. Data shown are mean ± SEM of three experiments. In e, f and g, indicated strains were grown to OD₆₀₀ of 0.2, 0.6 or 1.0, total RNA was extracted and RT-PCR analysis performed for relative abundance of fnbA (e), fnbB (f) and purE (g) transcripts. All data were normalized to levels of rpoB and expressed as fold change using WT pALC (empty plasmid) as comparator at each OD₆₀₀ value. Data shown are mean ± SEM of four independent experiments. *p-value < 0.05, **p-value < 0.01, ***p-value < 0.001, based on a one-way ANOVA with a Bonferroni post-test. Source data are provided as a Source Data file

Fig. 4

A S. aureus purR mutant is hypervirulent via FnbAB. In a, mice (9–12 per group) were infected with ~1 × 10⁷ CFU of WT USA300, USA300 purR::ΦNΣ or complemented purR::ΦNΣ mutant and survival monitored over 72 h. ***p-value < 0.001, based on a Mantel-Cox test. In b, animals were infected as in a, but with 2–2.5 × 10⁶ CFU, and c weight loss monitored daily for 48 h. **p-value < 0.01, ***p-value < 0.001, based on a one-way ANOVA with a Bonferroni post-test. In d, animals from b were sacrificed at 48 h post infection (hpi), and heart, kidney and liver were harvested and bacterial burdens determined. Data shown are mean ± SEM, *p-value < 0.05, **p-value < 0.01, ***p-value < 0.001, based on a Student’s unpaired t-test. In e, two animals per bacterial strain were infected as in a, with ~1 × 10⁷ CFU, sacrificed at 24 hpi and organs harvested. Organs were paraffin embedded, sectioned and stained with H&E and a Gram stain. Representative images are shown. In f, animals were infected as in a, with ~1 × 10⁷ CFU, with the inclusion of WTΔfnbAB and purR::ΦNΣΔfnbAB strains, and monitored for 72 h. ***p-value < 0.001, based on a Mantel-Cox test. In g, the heart, kidney and liver from the animals infected in e were harvested at the point of sacrifice and bacterial burden determined. Data shown are mean ± SEM, *p-value < 0.05, **p-value < 0.01, ***p-value < 0.001, based on a Student’s unpaired t-test. Source data are provided as a Source Data file

Fig. 5

Anti-staphylococcal antibodies ameliorate purR hyper-clumping. a WT or the purR::ΦNΣ mutant were grown in TSB, TSB-S or TSB with 10% v/v fresh human serum (TSB-HuS) for 3 h and relative clumping ability was measured using OD₆₀₀ as described above. Data shown are mean ± SEM of four independent experiments. *p-value < 0.05, **p-value < 0.01 and ***p-value < 0.001, based on a one-way ANOVA with a Bonferroni post-test. WT (b) or the purR::ΦNΣ mutant (c) were grown in TSB (white bars), TSB-S (white bars), TSB-HuS (grey bars) or TSB with IgG-depleted human serum (HuS) (black bars) for 3 h and relative clumping ability measured as above. Data shown are mean ± SEM of four experiments, with four donors. **p-value < 0.01, ***p-value < 0.001, based on a one-way ANOVA with a Bonferroni post-test. In d, whole cell lysates of WT, WT pfnbA or WTΔfnbAB were used for western blots, with human serum (from donors in panels b and c) or a rabbit anti-Fnb serum (far right blot) used as a source of primary antibody. Source data are provided as a Source Data file

Fig. 6

Vaccination with S. aureus expressing FnbAB is protective against a challenge with a purR mutant. a Vaccination scheme, with six animals per group. b Survival of animals challenged with 1 × 10⁷ CFU of WT or purR::ΦNΣ S. aureus following vaccination, as outlined in a. *p-value < 0.05, **p-value < 0.01, based on a Mantel-Cox test, as compared to WT vaccinated, purR::ΦNΣ challenged animals. c Whole cell lysate of WT, WT pfnbA or WTΔfnbAB were used for a western blot, with serum from vaccinated animals or a rabbit anti-Fnb serum (far right) used as a primary antibody