Methicillin-resistant S. aureus USA300_FPR3757 was kindly provided by Hua Zhou, from Zhejiang University. MRSA252, Corynebacterium diphtheriae. S. aureus Newman, and Streptococcus agalactiae were from the collection of our Lab. Strains were identified by 16S rRNA sequencing. USA300, MRSA252, C. diphtheriae and Newman were grown at 37°C for 12–16 h in 30 mL Luria-Bertani (LB) broth (HuanKai Microbiology Technology Co., Ltd., Guangdong, China) with 220 rpm shaking. S. agalactiae was grown at 30°C for 12–16 h in 30 mL of LB broth with 220 rpm shaking. Uracil, thymine, cytosine, adenine, guanine, malonic acid, and antibiotics were purchased from Aladdin (Aladdin Biotech Co., Ltd., Shanghai, China). N-Acetyl-L-cysteine (NAC) was purchased from Solarbio (Beijing Solarbio Science and Technology Co., Ltd., Beijing, China). Citric acid, fumaric acid, succinic acid, and malic acid were purchased from Macklin (Macklin Biotech Co., Ltd., Shanghai, China). Carbonyl cyanide 3-chlorophenylhydrazone (CCCP) and 2′,7′-dichlorofluorescein diacetate (DCFH-DA) were obtained from Sigma-Aldrich (USA). 3,3′-diethyloxacarbocyanine iodide (DiOC23)) was purchased from Invitrogen (USA).

The minimum inhibitory concentrations of gentamicin against USA300 and E. coli-R_Gen were determined as previously described (Wiegand et al., 2008). Briefly, cells were cultured in 96-well plates containing LB medium with doubling serially diluted gentamicin ranging from 0.01 to 160 μg/mL. Overnight bacterial cultures were diluted 1:100 in fresh LB medium and cultured till OD₆₀₀ reached 0.5. The tray contained 90 µL of LB medium, with a series of two-fold dilutions of gentamicin and 10 µL of logarithmic-phase cells at 5 × 10⁶ CFU/mL. After overnight incubation, the minimal antibiotic concentrations exhibiting no visible growth was recorded as the MIC value. Three biological replicates were performed.

Antibacterial assays were performed as previously described (Peng et al., 2015b). Briefly, overnight bacteria were collected by centrifugation for 5 min at 8,000 rpm. Then washed with 0.85% sterile saline three times, and diluted to an OD₆₀₀ of 0.2 in 5 mL of M9 minimal medium (containing 10 mM acetate, 1 mM MgSO₄, and 100 µM CaCl₂). Relevant metabolites and/or antibiotics were added to the samples, which were then incubated at 37°C for 6 h. Finally, 100 µL of the culture was obtained and then serially diluted. An aliquot (10 µL) of each dilution was plated on LB agar to determine the bacterial count.

Sample preparation was performed as previously described (Cheng et al., 2014). In brief, 10 mL OD₆₀₀ of 1.0 cells were incubated in M9 minimal medium for 6 h, with or without uracil. Equivalent amounts of cells were collected by centrifugation at 8,000 rpm for 5 min, quenched using liquid nitrogen, and cold methanol was added prior to storage at −80°C. Cells were lysed by sonication for 5 min at a 200 W power setting, and metabolites were extracted with 1 mL of cold methanol (Sigma, United States) containing 10 μg of ribitol (Sigma, United States, used as the internal standard). Samples were centrifuged at 12,000 g at 4°C for 10 min and the supernatant was dried using a rotary vacuum centrifuge (LABCONCO, Germany). The dried extracts were then incubated with 50 µL of methoxyamine hydrochloride solution (20 mg/mL in pyridine) at 37°C for 1.5 h. Subsequently, samples were derivatized via a 37°C reaction for 1 h with the addition of 50 µL of N-methyl-N-(trimethylsilyl) trifluoroacetamide. Finally, the derivatized sample was centrifuged at 4°C and 12,000 g for 15 min, and the supernatant was removed for analysis. 1 µL of the derivatized sample was injected and analyzed using the gas chromatography–mass spectrometry (GC-MS-QP 2010 plus, Shimadzu, Japan). The inlet temperature, split ratio of the carrier gas (high-purity helium), and constant linear velocity were set at 300 °C, 5:1, and 40.0 cm/s respectively (Ye G. Z. et al., 2021). The metabolites were separated using the DB-5 MS capillary column (30 m × 250 μm × 0.25 μm, J&W Scientific Inc., Unite States). Mass signals of metabolites were acquired in the full scan mode. To obtain the retention index of the bacterial metabolites, which is equivalent to the retention time of n-alkanes, a light diesel sample was analyzed using the same instrumental parameters as those of the analytical samples.

Raw mass spectrometry data in NetCDF format exported by GC-MS solution 4.2 (Shimadzu, Japan) was used to perform peak processing using the XCMS method (Smith et al., 2006). Metabolites were firstly identified by searching the commercial mass spectra libraries, and then verified using available standards. Significance analysis was conducted using IBM SPSS Statistics 22.0 (SPSS Inc., Chicago, IL, United States), and statistically significant results were considered if the p-value was less than 0.05. R studio software version 4.0.3 was employed for hierarchical cluster analysis (HCA), and SIMCA-P + software 12.0 (Umetrics, Umeå, Sweden) was used for principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA). Z-score analysis was performed based on standard deviations from the mean using Microsoft Excel, and the enrichment pathways of significant metabolic was performed with MetaboAnalyst 5.0. Interactive pathways (iPath) analysis was conducted using iPath 3.0 (https://pathways.embl.de/). Data was processed using Microsoft Excel and figures were generated using GraphPad Prism 8.0 (San Diego, CA, United States).

Quantitative real-time PCR (qRT-PCR) was carried out as described previously, with modifications (Chen et al., 2022). Bacterial cells (2 mL) were harvested at OD₆₀₀ = 1.0 by centrifugation (12,000 g, 4°C, 3 min) and immediately quenched in liquid nitrogen. Then, the cells were lysed to extract total RNA using the bacterial RNA extraction kit (Vazyme, Jiangsu, China). Reverse transcription-PCR was conducted with 1 µg of total RNA, using the Evo M-mLV RT Mix Kit with gDNA Clean for qPCR (Accurate Biotechnology Co., Ltd., Guangdong, China), according to the instruction manual. The qRT-PCR was carried out in 96-well plates, each well containing a total volume of 20 µL liquid-composed of 10 µL 2× SYBR Green Pro Taq HS Premix, 9 µL PCR-grade water, 0.2 µL cDNA template, and 0.4 µL each of the primers (10 μM). Primers sequences are listed in Supplementary Table S1. As per the manufacturer’s instructions, three biological replicates were used, and all assays were performed on the CFX Connect Real-Time System (Bio-Rad, USA). Cycling parameters were as follows: an initial denaturation at 95°C for 30 s, 45 cycles at 95°C for 5 s, and 58°C for 30 s mRNA levels of target genes were normalized to that of gyrB, which is constitutively and steadily expressed under the conditions analyzed (Hartmann et al., 2014).

Enzymatic activity was measured as previously described (Chen et al., 2022). In brief, cells were collected and then resuspended in sterile saline to OD₆₀₀ = 1.0 in M9 medium after washing three times. Samples (30 mL) were collected by centrifugation at 8,000 rpm for 5 min. Cells were resuspended following lysis using lysostaphin and sonication (200 W total power with 60% output, 2 s pulse, and 3 s pause) over ice for 20 min. Enzymatic activity was determined spectrophotometrically by monitoring the reduction of 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) at 566 nm using the following reaction mixtures. The samples were centrifuged at 12,000 g for 10 min and concentration of protein concentration was determined using a BCA assay (Beyotime, Shanghai, China). Supernatants containing 200 µg of total proteins were added to an α-ketoglutaric dehydrogenase (OGDH) reaction mixture containing 0.15 mM MTT, 2.5 mM MgCl₂, 5 mM phenazine methosulfate (PMS), 0.2 mM thiamine pyrophosphate (TPP), 5 mM α-ketoglutaric acid potassium salt, and 100 mM PBS [pH 7.2]. An isocitrate dehydrogenase (ICDH) reaction mixture (0.15 mM MTT, 1 mM PMS, 2.5 mM MgCl₂, 100 mM Tris–HCl [pH 8.8], 3.5 mM isocitrate) with 10 µg of total proteins from supernatants was added to a succinate dehydrogenase (SDH) reaction mix (0.15 mM MTT, 1 mM PMS, 2 mM sodium succinate and 100 mM PBS), making up a final volume of 200 µL in a 96-well plate. Subsequently, the plate was incubated at 37°C for 20 min in the dark and absorbance value by colorimetric readings was measured at 566 nm. The activity of citrate synthase (CS) and complex I of electron transport chain was measured using the citrate synthase assay kit (Suzhou Keming Biotechnology Co., Ltd., Suzhou, China) and complex I activity assay kit (Beijing Solarbio Science and Technology Co., Ltd., Beijing, China).

ATP levels were determined via the BacTiter-GloTM Microbial Cell Viability Assay (Promega, Madison, WI, United States), as previously described (Ye J. Z. et al., 2021). Overnight bacterial cultures were collected by centrifugation at 8,000 rpm for 5 min. Samples were washed three times with 0.85% sterile saline, suspended in M9 minimal medium, and diluted to the concentration with an OD₆₀₀ of 0.2. Uracil (0, 1.25, 2.5, 5, or 10 mM) was added to the medium, and incubated at 37°C and 220 rpm for 6 h. After 10-fold dilution, 50 µL of the sample and 50 µL of the kit solution were added to a 96-well plate together. Finally, the absorbance was measured in the microplate reader (Biotek, Synergy HT, Vermont, USA) according to the manufacturer’s instructions.

The NAD⁺/NADH assay kit with WST-8 (Beyotime, Shanghai, China) was used to measure NAD⁺/NADH. First, cells were collected and diluted to an OD₆₀₀ value of 1.0 with or without uracil (10 mM), in M9 minimal media. After 6 h of incubation, 5 mL of cell pellets were washed with PBS (pH = 7.2) and resuspended in 600 µL of liquid mixture (precooled extraction buffer: PBS, 1:1). Subsequently, pellets were lysed with lyases by sonication for 20 min, and then centrifuged at 12,000 g for 10 min at 4°C. Absorbance of the supernatants was measured at 450 nm using a plate reader (Biotek, Synergy HT, Vermont, USA) with a reference wavelength of 655 nm.

Overnight bacterial cultures were pelleted by centrifugation at 8,000 rpm for 5 min. Samples were washed three times with 0.85% sterile saline, suspended in M9 minimal medium, and diluted to an OD₆₀₀ of 0.6. Uracil (10 mM) or/and gentamicin (400 μg/mL) were added to the medium and incubated at 37 °C, 220 rpm for 6 h. Then, 194 µL of bacterial cells and 4 µL of 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA, Sigma, United States), which of the final concentration was 20 µM, and added to a 96-well plate and incubated at 37°C for 1 h in the dark. Fluorescence units were immediately measured at an excitation wavelength of 485 nm and emission wavelength of 515 nm using a plate reader (Biotek, Synergy HT, Vermont, Unites States).

The role of uracil in bacterial respiration was determined based on the changes in iodonitrotetrazolium chloride (INT) (Sigma, USA) (Liu Y. et al., 2021). Briefly, overnight bacterial cultures were washed, and diluted to an OD₆₀₀ value of 1.0 in M9 minimal media, with or without uracil (10 mM). Subsequently, 1 mM INT and 0.6 mM NADH (Macklin Biotech Co., Ltd., Shanghai, China) were added as substrates: the solution was then incubated for 45 min in the dark at 37°C. The reaction was terminated by addition of 5% trichloroacetic acid. Insoluble formazan was collected by centrifugation at 13,000 g for 5 min and then extracted with ethanol. Absorbance of the supernatant was measured at 485 nm every 5 min for 45 min.

DiOC₂3) was used to measure membrane potential (PMF) (Novo et al., 1999). In brief, overnight bacteria were collected at an OD₆₀₀ value of 0.2, and incubated in 5 mL of M9 minimal medium for 6 h at 37°C and 220 rpm. Cells (495 μL) containing 1 × 10⁷ CFU were stained with 5 μL of 3 mM DiOC₂3) for 30 min in the dark. Red (PE) and green (FTIC) fluorescence intensity was determined using a flow cytometer (Beckman Coulter, Brea, CA, United States). The PMF was calculated with formula 1.5 + log10 (red/green).

A gentamicin ELISA rapid test kit (CUSABIO, Hubei, China) was used to detect intracellular gentamicin content. In brief, overnight bacteria were collected at a concentration with an OD₆₀₀ value of 1.0. Subsequently, gentamicin was added and incubated for 6 h at 37°C and 220 rpm, with or without uracil. 30 mL of cells were harvested, lysated with lysostaphin (Macklin Biotech Co., Ltd., Shanghai, China) and sonication (200 W total power with 60% output, 2 s pulse, and 3 s pause) over ice for 20 min. The supernatant was collected after centrifugation at 12,000 g for 10 min at 4°C. Finally, gentamicin concentration was assessed based as per the rapid diagnostic kit instructions.

Bacterial cultures from −80°C were diluted 1: 100 in fresh 50 mL LB medium, and cultured to an OD₆₀₀ of 1.0 at 37°C, and oscillated at 220 rpm. Following collection by the centrifugation at 8,000 rpm for 5 min, cells were diluted with PBS to the concentration with 5 × 10⁹ CFU/mL cells. BALB/c mice (8–10 weeks) were injected with 100 μL of the bacterial suspensions with the concentration of 5 × 10⁹ CFU/mL cells. Bacterial infection was performed by tail vein injection. After 6 h, mice were randomly divided into four groups (n = 8 per group), and separately treated with PBS, gentamicin (5 mg kg⁻¹), uracil (50 mg kg⁻¹), and gentamicin (5 mg kg⁻¹) + uracil (50 mg kg⁻¹) twice a day. The survival of mice was recorded daily for 7 days.

To observe the effect of nucleotides on antibiotic activity against methicillin-resistant S. aureus, USA300, was incubated with gentamicin or gentamicin plus five single exogenous nucleotides. More than 95% of the cells survived when exposed to 400 μg/mL gentamicin alone. However, the bactericidal effect of gentamicin was significantly increased by combined treatment with cytosine, thymine, adenine, uracil or guanine (Figure 1A). In particular, the efficacy of gentamicin was potentiated approximately 400-fold by uracil or guanine. Due to the strong synergistic bactericidal effect of uracil and its wide application in therapy, we further explored the role of uracil in synergistic sterilization with aminoglycosides (Ramesh et al., 2020). We found that exogenous uracil increased gentamicin-mediated bacterial killing in a dose-dependent manner (Figure 1B). In addition, the synergistic efficacy of gentamicin with uracil was enhanced with an increase in the gentamicin dose and time (Figures 1C, D). Similar synergistic bactericidal effects of uracil in combination with other aminoglycosides (amikacin, kanamycin, and tobramycin) were also observed (Supplementary Figure S1A). However, uracil did not augment the bactericidal effect of other types of antibiotics (ampicillin, tetracycline, erythromycin, and ciprofloxacin) against USA300 (Supplementary Figure S1B). To determine whether the enhanced-bactericidal effects of uracil and gentamicin in combination persisted until extended periods of time, viable bacteria were killed repeatedly for 20 days (1 cycle/day). The synergistic bactericidal effect on the 10th day was 10 folds weaker than that on the first day, but after 20 days, the efficacy was restored (Supplementary Figure S1C). These results indicate that uracil can effectively and consistently potentiate aminoglycosides activity to eliminate MRSA.

To explore whether uracil played a role through metabolic regulation, which potentiated gentamicin-mediated killing. GC-MS was used to analyze metabolomes of USA300 cells in the presence or absence of uracil. Unsupervised hierarchical clustering identified that 65 metabolites in the bacteria (Supplementary Figure S2A). Among these metabolites, 32.31%, 20%, 29.23%, 13.85%, and 4.61% of metabolites were carbohydrates, amino acids, fatty acids and lipids, nucleotides, and others (Supplementary Figure S2B). In total, 39 metabolites showed differential abundances (p < 0.05) with uracil (Figure 2A). The Z-score plot demonstrated that upon uracil treatment, 2′-deoxyridine, uracil, 5-methyluridine, thymine and uridine were the top five most upregulated metabolites, while alanine, glutamic acid, pyroglutamic acid, docosanoic acid and lysine were the top five most downregulated metabolites (Figure 2B). These differential metabolites were mainly divided into five categories: carbohydrates (38.46%), amino acids (17.95%), fatty acids (23.08%), nucleotides (17.95%), and others (2.56%) (Figure 2C). These results indicate that exogenous uracil induces significant reprogramming of the MRSA metabolome.

Principal component analysis was used to identify two principal components, where component t [1] distinguished USA300 from USA300 + uracil (Figure 3A). The discriminating variables are presented in an S-plot (Figure 3B). Cutoff values for the absolute value of the covariance p and correlation p (corr) were ≥0.05. The axes that are plotted in the S-plot from the predictive component are p vs. p (corr), representing the magnitude (modeled covariation) and reliability (modeled correlation), respectively. The selection of potentially biochemically interesting compounds needs a combination of covariance and correlation information, which is the purpose of the S-plot (Wiklund et al., 2008). Ten crucial biomarkers were identified, of which alanine, lysine, aspartic acid, glutamic acid, and pyroglutamic acid had decreased. In contrast, succinic acid, ethylene glycol, uracil, 2′-deoxyuridine, and 2-hydroxy pyridine increased in response to uracil treatment (Figure 3C). Increase in succinic acid, an intermediate of the TCA cycle, as well as the elevation of aspartic acid and glutamic acid, metabolites capable of directly entering the TCA cycle–suggest a potential role for the TCA pathway. Consistently, results of the pathway enrichment analysis showed that six metabolic pathways were significantly enriched, of which alanine, aspartate, and glutamate metabolism, TCA cycle, and aminoacyl-tRNA biosynthesis exhibited highest impact (Figure 3D). Bacterial metabolites enriched in the TCA cycle and butanoate metabolism had all increased in response to uracil treatment, including citric acid, succinic acid, fumaric acid, malic acid, and 4-aminobutyric acid (Figure 3E). Further analysis using iPath 3.0 provided a global overview of the metabolome, revealing increased TCA cycle and nucleotide metabolism, and decreased carbohydrate metabolism in USA300 upon treatment with uracil (Figure 3F). These results demonstrate that uracil promotes the TCA cycle in MRSA.

To further validate that the TCA cycle was enhanced upon uracil treatment, expression levels of genes and the activity of enzymes involved in the TCA cycle were measured. Among the 14 genes detected, mRNA expression levels of seven genes were elevated, including glta, acnA, sucA/B/C, sdhA, and mqo (2,541), while the remainder were unchanged in the presence of uracil (Figure 4A). Genes exhibiting elevated expression encoded for enzymes including one subunit of citrate synthase (CS), aconitate hydratase (ACO), succinyl-CoA synthetase (SCS), succinate dehydrogenase (SDH), malate dehydrogenase (MDH), and two subunits of α-ketoglutarate dehydrogenase (OGDH) in the TCA cycle. To further confirm the activation of the TCA cycle, the activity of three key enzymes in the TCA cycle, CS, isocitrate dehydrogenase (ICDH), and OGDH were measured. Consistent with qRT-PCR results, the activity of CS and OGDH were increased in USA300 cells following uracil treatment (Figure 4B).

Since metabolomics showed that citric acid, succinic acid, fumaric acid, and malic acid were increased in USA300 exposed to exogenous uracil, these four metabolites were separately added to the medium to investigate their roles in enhancing aminoglycoside activity. We found that the aforementioned metabolites could independently promote bactericidal effects of gentamicin on USA300 (Figure 4C). Blocking the TCA cycle with malonic acid (a competitive inhibitor of SDH) rescued the viability of the bacteria exposed to uracil, demonstrating the crucial role of the TCA cycle in uracil-mediated synergistic sterilization with aminoglycosides (Figure 4D). Hence, we speculated that uracil might affect NADH and PMF. As expected, the intracellular NAD⁺/NADH ratio decreased and PMF increased after exogenous uracil treatment (Figures 4E, F). Subsequently, PMF inhibitor carbonyl cyanide-chlorophenyl hydrazone (CCCP) was added to the medium to examine the role of PMF in uracil-mediated synergistic sterilization with aminoglycosides. The results showed that uracil-enabled killing of USA300 by gentamicin was decreased following CCCP treatment (Figure 4G). Uniformly, exogenous uracil increased intracellular gentamicin content in USA300 cells (Figure 4H). Taken together, uracil treatment elevates TCA cycle flux and promotes the conversion of NAD⁺ to NADH and increases PMF, thereby enhancing gentamicin uptake and potentiating its bactericidal effects.

As the TCA cycle plays a key role in aerobic respiration, uracil may affect the bacterial electron transport chain. Thus far, it has been shown that two types of respiratory oxygen reductases are present; however, bc1 complex and cytochrome c oxidase do not exist in S. aureus (Schurig-Briccio et al., 2014). Thus, the activity and gene expression of complex I in the respiration chain were determined. As expected, the activity and gene expression of complex I were increased in uracil-treated USA300 (Figures 5A, B). Meanwhile, uracil increased bacterial respiration in a time-dependent manner (Figure 5C). In addition, uracil promoted the production of the intracellular ATP level in USA300 (Figure 5D). Increased bacterial respiration is typically accompanied by the production of reactive oxygen species (ROS). We observed that uracil alone increased the production of ROS, and this increase was higher following combined treatment with gentamicin and uracil (Figure 5E). Consistently, adding the ROS scavenger N-acetyl-L-cysteine (NAC) blocked the synergistic bactericidal activity of gentamicin plus uracil, demonstrating the involvement of ROS in uracil-mediated synergistic sterilization with aminoglycosides (Figure 5F). These results indicate that exogenous uracil promotes bacterial respiration and elevates the ROS levels, thereby increasing the efficacy of aminoglycosides.

The synergistic bactericidal effects of uracil and gentamicin against other pathogenic bacteria were further examined in vitro. The results showed that uracil increased the efficacy of gentamicin (Figure 6A). A mouse model of MRSA infection was used to assess the synergistic effect in vivo. BALB/c mice were infected with USA300 and treated with a single dose of gentamicin and/or uracil. Gentamicin or uracil treatment in isolation were ineffective in improving the survival of infected mice (only 10% for uracil and 20% for gentamicin) (Figure 6B). Interestingly, the combined administration of gentamicin and uracil greatly increased the survival of infected mice (>80%) (Figure 6B). These data suggest that uracil increases the killing efficacy of gentamicin against other Gram-positive bacteria in vitro. Moreover, combined treatment significantly improve survival rate in mice infected with MRSA.