Mechanisms of Staphylococcus aureus survival of trimethoprim-sulfamethoxazole-induced thymineless death

Abstract

Trimethoprim-sulfamethoxazole (SXT) is commonly used to treat diverse Staphylococcus aureus infections, including those associated with cystic fibrosis (CF) pulmonary disease. Studies with Escherichia coli found that SXT impairs tetrahydrofolate production, leading to DNA damage, stress response induction, and accumulation of reactive oxygen species (ROS) in a process known as thymineless death (TLD). TLD survival can occur through the uptake of exogenous thymidine, countering the effects of SXT; however, a growing body of research has implicated central metabolism as another potentially important determinant of bacterial survival of SXT and other antibiotics. Here, we conducted studies to better understand the mechanisms of TLD survival in S. aureus. We found that thymidine abundances in CF sputum were insufficient to prevent TLD of S. aureus, highlighting the importance of alternative survival mechanisms in vivo. In S. aureus cultured in vitro with SXT and low thymidine, we frequently identified adaptive mutations in genes encoding carbohydrate, nucleotide, and amino acid metabolism, supporting reduced metabolism as a common survival mechanism. Although intracellular ROS levels rose with SXT treatment in vitro, survival was not improved in the presence of ROS scavengers, unlike in E. coli. SXT challenge induced the SOS response, which was alleviated by added thymidine. Finally, an inactivating mutation in the phosphotransferase gene ptsI conferred both limitation in cellular ATP and improved survival against TLD. Collectively, these results suggest that alterations in core metabolic functions, particularly those that reduce ATP levels, predominantly confer S. aureus survival and persistence during SXT treatment, potentially identifying novel targets for co-treatment.IMPORTANCEStaphylococcus aureus is a ubiquitous organism and one of the leading causes of human infections, many of which are difficult to treat due to persistence, antibiotic resistance, or antibiotic tolerance. As our arsenal of effective antibiotics dwindles, the need for improved treatments becomes increasingly urgent, necessitating a better understanding of the precise mechanisms by which pathogens evade our most critical antimicrobial agents. Here, we report a systematic characterization of the mechanisms of S. aureus survival to treatment with the first-line antistaphylococcal antibiotic trimethoprim-sulfamethoxazole, identifying pathways and candidate targets for enhancing the efficacy of available antimicrobial agents.

Keywords: Staphylococcus aureus; antibiotic resistance; antifolate drugs; persistence; thymineless death.

Fig 1

Kinetics of survival of two laboratory S. aureus strains and a TD-SCV mutant in the presence and absence both of trimethoprim-sulfamethoxazole and thymidine supplementation. S. aureus strains (A) Newman, (B) JE2, and (C) Newman-derived ∆thyA cultured in LB over 24 h. Each strain was supplemented with the concentrations of thymidine (THY) indicated in the legend (0–16 µg/mL); strains Newman and JE2 were treated with 8 µg/mL TMP and 152 µg/mL SMX; ∆thyA was not treated with SXT (indicated in panel C) due to its natural ability to undergo TLD in the absence of thymidine. A control experiment including supplementation with 128 µg/mL THY to simulate wild-type-like growth was included for ∆thyA only (C). Data represent mean values ± SD (n ≥ 2).

Fig 2

Concentrations of free thymidine and its analogs in sputum samples from PwCF. (A) LC-MS/MS analysis of free deoxythymidine triphosphate, deoxythymidine diphosphate, deoxythymidine monophosphate, and thymidine in 53 CF sputum samples. Significance was determined using Kruskal-Wallis and Dunn’s multiple comparison tests; significant P-values are reported. (B) All analyzed samples were categorized as S. aureus-negative, S. aureus-positive (non-SCV), or S. aureus SCV-positive and the corresponding thymidine concentration as determined by LC-MS/MS; all data are reported as median with interquartile range. For panels A and B, the dotted line indicates the 0.041 µg/mL limit of detection. Significance was determined using either the (A) Wilcoxon signed rank test (theoretical mean of 0.25 µg/mL) or the (B) Kruskal-Wallis test for multiple comparisons (right); only significant P-values are reported.

Fig 3

Characteristics of adaptive mutants cultured from LB following SXT challenge. (A) Proportion of colonies exhibiting the TD-SCV morphology following 120 h of SXT exposure of Newman under high thymidine conditions (1, 4, and 16 µg/mL). Phenotypes were determined by plating on blood agar, and total bacterial abundances were measured on chocolate agar plates (n = 4). (B) Survival of a subset of adaptive mutants isolated after 24 h from low thymidine concentrations relative to parental counterpart with re-exposure to SXT. Shown is change in log CFU/mL between 0 and 6 h in LB with the addition of SXT and in the absence of supplemental thymidine. Data represent mean ± SD (n ≥ 2). The isolates listed above contain nonsynonymous mutations in the following genes (Table S1): memE (NWM-TS01), menF (NWM-TS02), hemE (NWM-TS04), aroB (NWM-TS05), aroA2 (NWM-TS06), hemH (NWM-TS07), ptsI (NWM-TS08), thiN (NWM-TS10), ndk and hepT (NWM-TS17), tkt (NWM-TS56), pyk (JE2-TS02), pta/eutD (JE2-TS09), nrdE (JE2-TS10), and deoB (JE2-TS20); no mutations were detected in isolates NWM-TS26 and NWM-TS30.

Fig 4

Effect of oxygen and reactive oxygen species on S. aureus survival of SXT. (A) Survival kinetics of S. aureus Newman cultured under aerobic versus anaerobic conditions in LB over 10 h in the absence of additional thymidine and treated with SXT; data are mean ± SD (n = 3). (B) Intracellular ROS, measured by flow cytometry and fluorescent indicator CMH₂DCFDA, in S. aureus Newman, JE2, and ptsI mutant NWM-TS08 with and without SXT at 0 and 6 h. In total, 100,000 events were collected for each condition, where possible. Representative data of at least three replicate experiments. (C) Survival kinetics of S. aureus Newman with and without ROS scavenger Trolox (Tx) and SXT in LB over 6 h in the absence of added thymidine; SXT-treated conditions are indicated by a dotted line and data are mean ± SD (n = 2). (D) Survival kinetics of S. aureus JE2 carrying transposons in ROS detoxification genes (sodM::Tn, katA::Tn, and crtN::Tn) with and without SXT in LB over 6 h in the absence of additional thymidine; SXT-treated conditions are indicated by a dotted line and data are mean ± SD (n = 3).

Fig 5

SOS induction, ATP levels, and survival during SXT treatment after ATP depletion. (A) recA-GFP expression in response to sub-inhibitory levels of SXT (1.25–0.5 µg/mL SXT) or mitomycin C (0.005 µg/mL MMC; control) in Newman (with and without 128 µg/mL thymidine) and NWM-TS08 at 6 h post-SXT exposure; RFU was normalized to OD₆₀₀ and significance was reported as **** when P ≤ 0.05 (two-way ANOVA). Data are mean ± SD (n ≥ 3). (B) Relative whole-culture ATP levels at 2 h post-SXT exposure (RLU normalized to CFU/mL) presented as percentage of wild type for treated and untreated Newman, NWM-TS08, and NWM-TS08 complemented with wild-type ptsI (NWM-TS08 + pCN24-ptsI); data are mean ± SD (n ≥ 3), and significance is reported as *, ***, or **** where P ≤ 0.05 (two-way ANOVA). (C) Survival kinetics of Newman in LB with and without SXT and added CCCP or arsenate (Ars) as indicated; SXT-treated conditions are indicated by a dotted line and data are mean ± SD (n = 3).

Fig 6

Model of the proposed mechanism of SXT action in S. aureus. Upon SXT exposure, S. aureus undergoes thymidine starvation and DNA damage, inducing the SOS response and, ultimately, cell death. SOS induction can be mitigated, and survival conferred, by environmental thymidine levels above a threshold of 0.25 µg/mL. In contrast, with continued thymidine depletion, both ROS and ATP levels increase through undefined mechanisms. Of these, only ATP is strongly associated with lethality. Multiple conditions that decrease ATP levels, including decreased electron transport, decreased substrate phosphorylation, or decreased ATP production (either through slow growth or by impairing metabolism or ATP production) improve survival with SXT. Several details of this model remain unknown (dotted lines), including the points at which ROS accumulation and increased ATP production occur, whether thymidine supplementation prevents DNA damage itself, and the mechanism(s) through which ATP levels correlate with lethality. Created in BioRender (L. Gonsalves, 2024, BioRender.com/f49f324).