Staphylococcus aureus Multiplexes Death-Effector Deoxyribonucleosides to Neutralize Phagocytes

Abstract

Adenosine synthase A (AdsA) is a key virulence factor of Staphylococcus aureus, a dangerous microbe that causes fatal diseases in humans. Together with staphylococcal nuclease, AdsA generates deoxyadenosine (dAdo) from neutrophil extracellular DNA traps thereby igniting caspase-3-dependent cell death in host immune cells that aim at penetrating infectious foci. Powered by a multi-technological approach, we here illustrate that the enzymatic activity of AdsA in abscess-mimicking microenvironments is not restricted to the biogenesis of dAdo but rather comprises excessive biosynthesis of deoxyguanosine (dGuo), a cytotoxic deoxyribonucleoside generated by S. aureus to eradicate macrophages of human and animal origin. Based on a genome-wide CRISPR-Cas9 knock-out screen, we further demonstrate that dGuo-induced cytotoxicity in phagocytes involves targeting of the mammalian purine salvage pathway-apoptosis axis, a signaling cascade that is concomitantly stimulated by staphylococcal dAdo. Strikingly, synchronous targeting of this route by AdsA-derived dGuo and dAdo boosts macrophage cell death, indicating that S. aureus multiplexes death-effector deoxyribonucleosides to maximize intra-host survival. Overall, these data provide unique insights into the cunning lifestyle of a deadly pathogen and may help to design therapeutic intervention strategies to combat multidrug-resistant staphylococci.

Keywords: Staphylococcus aureus; apoptosis; deoxyguanosine; deoxyribonucleosides; immune evasion; macrophage; phagocyte.

Figure 1

S. aureus deploys AdsA to synthesize deoxyguanosine. (A) SDS-PAGE analysis of purified rAdsA. Numbers to the left of the SDS-PAGE indicate the migration of molecular weight markers in kilodaltons. (B, C) Detection of dGuo formation by rAdsA. rAdsA was incubated with Nuc-digested DNA (B) or dGMP (C) at 37°C. Controls lacking Nuc-digested DNA, dGMP, or rAdsA are indicated (+ and – symbols). Reaction products and the formation of dGuo were analyzed via TLC. The migratory positions of single deoxyribonucleosides or dGMP were identified using pure standards. Representative images are shown. (D) Detection and quantification of rAdsA-derived dGuo by LC-MS/MS. rAdsA was incubated with dGMP at 37°C. Controls lacking dGMP or rAdsA are indicated (+ and – symbols). Reaction products and the formation of dGuo were analyzed and quantified via LC-MS/MS. (E) S. aureus-dependent hydrolysis of dGMP. The ability of S. aureus to hydrolyze dGMP in an AdsA-dependent manner was evaluated by assessing the release of inorganic phosphate (P_i) using a malachite green-based colorimetric assay. Wild-type S. aureus Newman (WT) or the mutant lacking adsA (ΔadsA) along with the complemented adsA variant (c-adsA) are indicated. (F) Analysis and quantification of S. aureus-dependent formation of dGuo. Lysostaphin-generated cell wall extracts of indicated strains were incubated with dGMP at 37°C and analyzed via TLC as described above. Individual dGuo spots of independent replicates were quantified by using ImageJ. Data are the mean (± standard deviation [SD]) values from at least three independent determinations. Statistically significant differences were analyzed with one-way analysis of variance (ANOVA) and Tukey’s multiple-comparison test (D–F); ns, not significant (P ≥ 0.05); *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 2

S. aureus generates deoxyguanosine to kill phagocytes. (A, B) Survival rates of human U937-derived macrophages (MФ) (A) or murine RAW264.7 MФ (B) exposed to dGuo (+) or left untreated (-). (C, D) Survival rates of U937-derived MΦ (C) or murine RAW264.7 MФ (D) exposed to rAdsA-derived dGuo. rAdsA was incubated with dGMP and reaction products containing dGuo were used to kill phagocytes. Controls lacked rAdsA or dGMP or included reaction buffer only as indicated with + and − symbols. (E, F) Survival of U937-derived MΦ (E) or murine RAW264.7 MФ (F) after treatment with culture medium (RPMI 1640) that had been conditioned by incubation with either wild-type (WT) S. aureus Newman, its adsA mutant (ΔadsA), or the complemented adsA variant (c-adsA) in the presence (black bars) or absence (white bars) of dGMP. Controls included culture medium that had been conditioned by incubation with dGMP only (gray bars). 160 µM (A) or 320 µM (B) of dGuo was used to treat the cells. Cell survival rates were analyzed 24 h (A, C–E) or 48 h (B, F) post-treatment. Data are the mean (± standard deviation [SD]) values from at least three independent determinations. Statistically significant differences were analyzed by a two-tailed Student’s t-test (A, B) or with one-way analysis of variance (ANOVA) and Tukey’s multiple-comparison test (C–F); ns, not significant (P ≥ 0.05); *P < 0.05; ***P < 0.001; ****P < 0.0001.

Figure 3

Genome-wide CRISPR-Cas9 screen uncovers host factors conferring susceptibility to cytotoxic deoxyguanosine. (A) Chemical structure of deoxyguanosine (dGuo). (B) Schematic diagram illustrating the CRISPR-Cas9 screening approach used to identify host determinants that mediate susceptibility of the U937 macrophage cell line to dGuo. (C) Discovery of the top candidate genes following dGuo treatment of U937 cells by next generation sequencing (NGS). Data were analyzed using the MaGeCK-based robust rank aggregation (RRA) score analysis and are representative of two independent replicates (see also Supplementary Figure 3 ). A smaller RRA score indicates more essentiality. The twelve top-ranked genes are highlighted.

Figure 4

Staphylococcal deoxyguanosine kills macrophages by targeting the purine salvage pathway-apoptosis axis. (A, B) Survival of wild-type (WT) U937-derived macrophages (MФ) (A) or murine RAW264.7 MФ (B) exposed to dGuo in the presence (+) or absence (-) of 10 µM dipyridamole (Dipy.) or nitrobenzylthioinosine (NBTI), both inhibitors of hENT1. Cells were also exposed to the inhibitors only or left untreated. (C, D) Survival of WT U937-derived MΦ and their SLC29A1 ^−/−, DCK ^−/−, or CASP3 ^−/− variants after treatment with dGuo (C) or after treatment culture medium (RPMI 1640) that had been conditioned by incubation with either wild-type (WT) S. aureus Newman, its adsA mutant (ΔadsA), or the complemented adsA variant (c-adsA) in the presence (black bars) or absence (white bars) of dGMP. Controls included culture medium that had been conditioned by incubation with dGMP only (gray bars) (D). (E) Immunoblotting of lysates obtained from dGuo-exposed (+) or untreated (-) WT U937-derived MΦ with caspase-3 and GAPDH-specific antibodies (α-CASP3 and α-GAPDH, respectively). GAPDH was used as a loading control. Numbers to the left of blots indicate the migration of molecular weight markers in kilodaltons. (F) Lysates of dGuo-exposed (+) or untreated (-) WT U937-derived MΦ were analyzed for caspase-3 activity using a colorimetric assay. (G, H) Analysis of dGuo-dependent induction of apoptosis in WT U937-derived MΦ via FACS (G) or immunofluorescence microscopy (H). dGuo-exposed (+) or untreated (-) Mϕ were stained using FITC-annexin-V/PI and analyzed as indicated (see also Supplementary Figure 5 ). White bars depict a length of 100 μm. Representative images are shown. 80 µM (E), 160 µM (A, C, F–H), or 320 µM (B) of dGuo was used to treat the cells. Cell survival rates were analyzed 24 h (A, C, D) or 48 h (B) post-treatment. Induction of apoptosis was analyzed 18 h (F) or 24 h (E, G, H) post-treatment. Data are the mean (± standard deviation [SD]) values from three independent determinations. Statistically significant differences were analyzed with one-way analysis of variance (ANOVA) and Tukey’s multiple-comparison test (A–D) or by a two-tailed Student’s t-test (F, G) or; ns, not significant (P ≥ 0.05); ***P < 0.001; ****P < 0.0001.

Figure 5

S. aureus multiplexes death-effector deoxyribonucleosides to maximize macrophage cell death. (A) Survival rates of U937-derived macrophages (MФ) exposed to purine deoxyribonucleosides (dGuo and dAdo) as indicated with + and - symbols. (B) Lysates of purine deoxyribonucleoside-exposed (+) or untreated (-) WT U937-derived MΦ were analyzed for caspase-3 activity using a colorimetric assay. Cells were exposed to 160 µM of dGuo or dAdo alone, or received a combination of both (A, B). (C) Survival of U937-derived Mϕ after treatment with culture medium (RPMI 1640) that had been conditioned by incubation with wild-type S. aureus Newman in the presence (black bars) or absence (white bars) of dGMP and/or dAMP as indicated with + and - symbols. Controls included culture medium that had been conditioned by incubation with dGMP and/or dAMP only (gray bars). (D) Survival of U937-derived MΦ after treatment with RPMI 1640 that had been conditioned by incubation with either wild-type (WT) S. aureus Newman, its adsA mutant (ΔadsA), or the complemented adsA variant (c-adsA) in the presence (black bars) or absence (white bars) of a dGMP/dAMP cocktail. Controls included culture medium that had been conditioned by incubation with a dGMP/dAMP mixture only (gray bars). Cell survival rates or induction of apoptosis were analyzed 18 h (A, B) or 24 h (C, D) post-treatment. Data are the mean (± standard deviation [SD]) values from three independent determinations. Statistically significant differences were analyzed with one-way analysis of variance (ANOVA) and Tukey’s multiple-comparison test; ns, not significant (P ≥ 0.05); *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. (E) Scheme illustrating S. aureus-mediated killing of macrophages via synchronous targeting of the purine salvage pathway-apoptosis axis. hENT1 promotes uptake of S. aureus Nuc/AdsA-derived purine deoxyribonucleosides into phagocytes. dGuo (red spots) is converted via DCK to dGMP, while dAdo-based (blue spots) formation of dAMP is promoted by ADK and DCK. Generation of purine deoxyribonucleoside monophosphates triggers the accumulation of dGTP and dATP, ultimately culminating in the activation of caspase-3 and apoptotic cell death.