Mutual potentiation drives synergy between trimethoprim and sulfamethoxazole

Abstract

Trimethoprim (TMP)-sulfamethoxazole (SMX) is a widely used synergistic antimicrobial combination to treat a variety of bacterial and certain fungal infections. These drugs act by targeting sequential steps in the biosynthetic pathway for tetrahydrofolate (THF), where SMX inhibits production of the THF precursor dihydropteroate, and TMP inhibits conversion of dihydrofolate (DHF) to THF. Consequently, SMX potentiates TMP by limiting de novo DHF production and this mono-potentiation mechanism is the current explanation for their synergistic action. Here, we demonstrate that this model is insufficient to explain the potent synergy of TMP-SMX. Using genetic and biochemical approaches, we characterize a metabolic feedback loop in which THF is critical for production of the folate precursor dihydropterin pyrophosphate (DHPPP). We reveal that TMP potentiates SMX activity through inhibition of DHPPP synthesis. Our study demonstrates that the TMP-SMX synergy is driven by mutual potentiation of the action of each drug on the other.

Fig. 1

Synergistic activity of anti-folate combinations against E. coli and S. aureus. a Targets of anti-folate compounds. b–d E. coli BW25113 strain was grown overnight in LB medium. Cultures were washed twice and resuspended in M9-glucose, then inoculated into a 96-well round-bottom plate (Corning) containing the same medium with a range of concentrations of SMX, TMP, MAC173979, or combination of two compounds. Concentration ranges were as follows: SMX (0.024–25 µg ml⁻¹), TMP (0.0078–1 µg ml⁻¹), and MAC173979 (0.05–25 µg ml⁻¹). MICs were determined by visible growth after 24 h incubation at 37 °C. Synergy was assessed by calculating FICI. FICI_m, minimum value of FICI in the tested combinations is shown. Synergy (FICI_m ≤ 0.5). No interaction (FICI_m > 0.5). b–d Graphical representations of E. coli BW25113 checkerboard assays are shown. Representative data from at least three independent experiments are shown. b SMX and MAC173979. c TMP and MAC173979. d SMX and TMP

Fig. 2

Folate deficiency potentiates SMX activity. a Schematic of E. coli folate metabolism. Green characters indicate metabolites and enzymes that affect SMX susceptibility. Blue characters indicate enzymes that affect both SMX and TMP susceptibility. b MICs of SMX and TMP for E. coli folate pathway mutants were determined after 24 h of incubation at 37 °C in M9-glucose medium. The selected E. coli single-gene deletion mutants and their parental strain BW25113 (WT) from the Keio collection were used. c Effects of inosine (Ino) and methionine (Met) on SMX MIC against E. coli BW25113. Ino and Met were added to the medium at 10 µg ml⁻¹. MICs of SMX were determined after 24 h incubation at 37 °C in M9-glucose medium. Representative data from at least three independent experiments are shown

Fig. 3

TMP inhibits DHPPP biosynthesis pathway. The impact of TMP on intracellular levels of a 7,8-dihydroneopterin (DHN) and b, e, f 6-hydroxymethyl-7,8-dihydropterin (DHPt), c PABA were determined by using LC-MS/MS. The results represent the mean and standard deviation of three biological replicates. **p < 0.05. NS indicates no significant difference (p > 0.05). p-values of pairwise comparisons were calculated by using the Student’s t-test. d Effects of TMP on DHPPP biosynthesis pathway. Red characters are metabolites that decrease intracellular amount in response to TMP treatment (THF, GTP, and ATP were reported previously)

Fig. 4

A metabolic feedback loop amplifies SMX-TMP activity. a, b SMX-TMP synergy was assessed by growth kinetics in the presence of a sub-inhibitory concentration of each drug and in combination. The results represent the mean and SD of three biological replicates. a Wild type (E. coli BW25113 strain) was grown in the presence of SMX (0.06 µg ml⁻¹), TMP (0.125 µg ml⁻¹), or a combination of SMX (0.06 µg ml⁻¹) and TMP (0.125 µg ml⁻¹). b ΔnudB (E. coli BW25113 ΔnudB) was grown in the presence of SMX (0.003 µg ml⁻¹), TMP (0.0125 µg ml⁻¹), or a combination of SMX (0.003 µg ml⁻¹) and TMP (0.0125 µg ml⁻¹). c SMX-TMP synergy against ΔnudB (E. coli BW25113 ΔnudB) was assessed by FICI. FICI_m, minimum value of FICI for the tested combinations is shown. Representative data from at least three independent experiments are shown. d Relative metabolite abundance and metabolic flux are shown. No drug treatment (i), sub-inhibitory concentrations of each drug (ii), and combination of sub-inhibitory concentrations of each drug (iii). SMX treatment inhibits the accumulation of DHF by TMP that results in potentiation of TMP activity, and TMP treatment decreases DHPPP that potentiates SMX activity. Cyclic mutual potentiation results in amplified depletion of the essential cofactor THF