A host-directed adjuvant sensitizes intracellular bacterial persisters to antibiotics

Ethics statement

All animal experiments were approved by the UNC Institutional Animal Care and Use Committee (IACUC protocol ID: 24-029.0). Animal housing and feeding were centrally managed by the UNC Division of Comparative Medicine. Human blood was drawn by certified personnel at the UNC Center for AIDS Research and performed in compliance with ethics guidelines set forth by the institution’s Institutional Review Board (IRB). All participants provided informed consent before donating blood.

General materials

Antibiotics including rifampicin, moxifloxacin hydrochloride, gentamicin sulfate, chloramphenicol (Cam) (>95% purity) were purchased from Fisher Scientific. Vancomycin hydrochloride (>99%) was from Alfa Aesar. Compound KL1 (>90%) was obtained from ChemBridge, Enamine and also synthesized in-house (Supplementary Notes) to validate compound fidelity. Analogues KL2–6 (>90%) were purchased from ChemDiv, ChemBridge and Vitas M Chemical. KL7 was synthesized in-house (>90%) (Supplementary Notes). BIX-01294, an EHMT2/G9a inhibitor (>98%), was from MedChemExpress. RAW 264.7 macrophages (TIB-71) and THP-1 (TIB-202) monocytes were from ATCC and distributed by the UNC Tissue Culture Facility. RAW 264.7 cells were cultured in complete Dulbecco’s modified Eagle medium (DMEM, Gibco) supplemented with 10% (v/v) heat-inactivated fetal bovine serum (FBS) (Avantor Seradigm), 2 mM L-glutamine (Gibco), 1× non-essential amino acids solution (Gibco) and 1 mM sodium pyruvate (Gibco) below 18 passages without exceeding 90% confluency. THP-1 monocytes were maintained in RPMI 1640 medium (Gibco) with 10% FBS, 2 mM L-glutamine and 50 µM 2-mercaptoethanol (Gibco). BMDMs were isolated from C57BL/6J mice (JAX 000664) and cultivated in complete DMEM⁶³. Immortalized BMDMs were generated using a CRE-J2 retroviral infection method⁶⁴. Briefly, BMDMs were incubated in 50% (v/v) L929-conditioned medium and infected with CRE-J2 retrovirus on days 5 and 7. Transduced cells were cultured in conditioned medium, gradually reduced to 20%. Immortalized BMDMs were maintained in RPMI 1640 with 2 mM L-glutamine, 1× penicillin–streptomycin (GenClone), 10% (v/v) heat-inactivated FBS (R&D Systems) and 20% conditioned medium.

Bacterial strains and growth conditions

S. aureus strains LAC, HG003, MW2, JE2, JE2-lux (JE2luxABDCE) (kindly provided by Roger Plaut, Center for Biologics Evaluation and Research, Silver Spring)²⁸, ΔG4 (kindly provided by Anthony Richardson, University of Pittsburgh, Pittsburgh)³⁵ and bacteraemia isolates⁸ were routinely cultured in tryptic soy broth (TSB, Fisher Scientific) or Mueller–Hinton broth (MHB, Oxoid) at 37 °C with shaking (225 r.p.m.) unless otherwise stated. Clinical isolates were obtained under an IRB exemption from a pre-existing collection. S. enterica serovar Typhimurium 14028S was cultured in LB broth (Lennox) at 37 °C with shaking. M. tuberculosis strains N0155 (lineage 2) and N1283 (lineage 4)⁶⁵ were grown in Middlebrook 7H9 broth (BD Difco) with 10% (v/v) oleic albumin dextrose catalase, 0.2% (v/v) glycerol, 0.05% tyloxapol (w/v) and 0.1 mM sodium propionate (ThermoFisher) to optical density at 600 nm (OD₆₀₀) = 0.6 (ref. ⁶⁶).

MIC assay

Stationary-phase bacterial cultures were diluted 1:1,000 in MHB containing serially diluted Rif (0–200 µg ml⁻¹), Mox (0–20 µg ml⁻¹) or Van (0–16 µg ml⁻¹) in 96-well assay plates (Corning) in triplicates (200 µl per well). Plates were sealed with Breathe-EASIER membranes (Diversified Biotech) and incubated statically at 37 °C for 24 h. The MICs were determined by the absence of bacterial growth. Three independent assays were performed to ensure reproducibility.

Construction of the inducible reporter strains

The plasmid pALC2084, which contains a tetracycline (Tet)-inducible GFP-expressing cassette⁶⁷ (kindly provided by Ambrose Cheung, Dartmouth College, Hanover), was first transformed into S. aureus strain RN4220 via electroporation using a Gene Pulser Xcell system (Bio-Rad)⁶⁸. The plasmid was then purified and transformed into the recipient S. aureus strain HG003. Single colonies were isolated and cultured in TSB containing 10 µg ml⁻¹ Cam and 0–2 µM aTc (Sigma) for 2–3 h. GFP induction was verified at excitation and emission wavelengths of 475 nm and 508 nm, respectively, using a Synergy H1 microplate reader (BioTek). For mKate reporter construction, the mKate2 gene was amplified from pRN10 (Addgene, plasmid 84454)⁶⁹ using a specific primer pair (Supplementary Table 3) and Q5 High-Fidelity DNA Polymerase (New England Biolabs). The PCR product was purified using a MinElute kit (Qiagen). Both the Tet-inducible vector pRMC2 (Addgene, plasmid 68940)⁷⁰ and the mKate insert were digested with KpnI-HF and EcoRI-HF (New England Biolabs), gel purified using a QIAquick PCR purification kit (Qiagen), and ligated using T4 DNA ligase (New England Biolabs) at 16 °C for 16 h. The construct was transformed into MAX Efficiency DH5α competent cells (ThermoFisher). Site-directed mutagenesis was performed using a QuikChange kit (Agilent) and a primer pair (Supplementary Table 3) to incorporate an upstream ribosomal binding site. The final pRMC2–mKate plasmid was verified by restriction digest and Sanger sequencing, and then transformed into RN4220 and subsequently HG003. aTc-induced mKate expression was confirmed at 633 nm (excitation 588 nm) using the Synergy H1 reader.

Live-cell microscopy

RAW 264.7 macrophages were seeded in Nunc 8-well Lab-Tek chambered coverglass (ThermoFisher) at 2–3 × 10⁵ cells per well in 500 µl complete minimum essential medium (MEM, Gibco) supplemented with 10% (v/v) heat-inactivated FBS and 2 mM L-glutamine at 37 °C with 5% CO₂ for 16 h. Cells were infected with Tet-inducible GFP- or mKate-expressing S. aureus at a multiplicity of infection (MOI) of 20 (no Rif) or 100 (with Rif) by centrifugation at 1,000 g for 2 min, followed by incubation at 37 °C for 35 min. After washing once in 500 µl complete MEM, cells were treated with or without 10 µg ml⁻¹ Rif in the presence of 50 µg ml⁻¹ Gen at 37 °C for 1.5 h (no Rif) or 4 h (with Rif). Rif-treated cells were washed three times and incubated in Gen-containing medium (50 µg ml⁻¹) at 37 °C for 2–3 h to allow resuscitation of intracellular persisters. Cells were then treated with 2 µM aTc in the presence of 50 µg ml⁻¹ Gen at 37 °C for 3 h to induce reporter expression. GFP-expressing S. aureus-infected macrophages were imaged using a Leica SP8X Falcon microscope with excitation at 484 nm (WLL2), emission collection at 495–570 nm and a detector gain of 100 V. Imaging parameters included: 1,024 × 1,024-pixel scan, 16-bit depth, speed of 600, pixel averaging of 2, pinhole of 1 AU and acquisition using a ×63 oil-immersion lens (NA 1.4, 1–4× zoom).

Macrophages infected with mKate-expressing S. aureus were stained with 100–300 nM LysoTracker Green DND-26 (ThermoFisher) at 37 °C for 30 min, then washed and maintained in 200 µl complete MEM containing 1 µg ml⁻¹ Hoechst 33342, 50 µg ml⁻¹ Gen and 2 µM aTc. Live-cell z-stack and time-lapse imaging were performed using an Olympus FV3000RS confocal microscope (Olympus) equipped with a stage-top incubator at 37 °C, 5% CO₂ and 100% humidity. The settings were as follows: (1) 405-nm (Hoechst 33342), 488-nm (LysoTracker Green) and 594-nm (mKate) diode lasers were used for excitation with GaAsP detectors; (2) a 516 × 516-pixel scan was applied; (3) pixel averaging was set to 10; (4) pinhole size was set to 1 AU; and (5) images were obtained using a ×60 oil-immersion objective lens (NA 1.4). Images were exported using ImageJ/FIJI (v.1.54)⁷¹ without gamma correction.

ImageStream analysis

Cells were loaded into an Amnis ImageStreamX Mark II system (EMD Millipore) for imaging and fluorescence detection. For capturing live cells harbouring viable GFP-expressing S. aureus, 405-nm and 488-nm colinear lasers were set to 40 mW and 70–90 mW for live/dead cell discrimination and GFP expression, respectively. In-focus single-cell images were acquired with a ×60 magnification and a low-speed flow rate using the INSPIRE software (EMD Millipore). The exported data were analysed using IDEAS 6.2 software (EMD Millipore).

High-throughput screen for intracellular S. aureus energy modulators

RAW 264.7 macrophages were seeded in 4 ml complete MEM in Costar 6-well tissue culture-treated plates (Corning) at 8 × 10⁵ cells per well and incubated at 37 °C and 5% CO₂ for 16 h. Cells were infected with S. aureus strain JE2-lux at an MOI of 100 for 25 min at 37 °C. Infected cells were washed once with 2 ml PBS and incubated with 10 mM EDTA in PBS (2 ml per well) at 37 °C for 5–10 min to detach. Cells were pelleted at 300 g for 6 min, resuspended in complete MEM with 50 µg ml⁻¹ Gen at 2.5 × 10⁵ cells per ml, and dispensed (80 µl per well) into 384-well white and black plates (ThermoFisher) for luminescence and fluorescence measurements, respectively. Uninfected macrophages were added to column 1 and 24 as background and viability controls. Plates were preloaded with 80 nl of 10 mM compounds using a Mosquito LV pipetting system (SPT Labtech). Vehicle (0.1% dimethylsulfoxide (DMSO)) and 10 µg ml⁻¹ Rif were included for quality control. After cell dispensing, plates were centrifuged at 500 g for 2 min and incubated at 37 °C and 5% CO₂ for 4 h. Luminescence was recorded at 1.25-mm depth with a detector gain of 250 using a Synergy H1 reader. To assess cell viability, 10 µl CellTiter-Fluor (Promega) was added to black plates and incubated for 30 min in the dark before fluorescence was measured at 7-mm depth (gain 100). The kinase-targeted compound library (>4,700 rule-of-five-compliant compounds) was from the UNC Center for Integrative Chemical Biology and Drug Discovery (CICBDD) (Supplementary Table 4). Luminescence and fluorescence signals were normalized to 0.1% DMSO controls. EC₅₀ values for Van and Rif against intracellular MRSA strain JE2-lux were determined using GraphPad Prism. The screen was performed once for hit identification.

Antibiotic survival assays

For S. aureus planktonic cultures, 16- to 18-h stationary-phase cultures were diluted 1:100 in MHB and incubated at 37 °C with shaking at 225 r.p.m. for 3 h to reach the mid-exponential phase. Cultures were then treated with 10 µg ml⁻¹ Rif and/or 0–40 µM KL1 and incubated for 24 h under the same conditions. Bacteria were washed in one volume of 1% NaCl three times at 20,000 g for 5 min, followed by resuspension in one volume of 1% NaCl. Suspensions were serially diluted 10-fold and plated on tryptic soy agar (TSA) to quantify surviving S. aureus. Tolerant bacterial frequencies were calculated by normalizing c.f.u.s to the initial input.

To evaluate intracellular S. aureus persister frequencies, RAW 264.7 macrophages or BMDMs were seeded into 24-well tissue culture-treated plates (Corning) at 4 × 10⁵ cells per well in 500 µl complete MEM, and incubated at 37 °C and 5% CO₂ for 16 h. THP-1 cells were seeded at 2 × 10⁵ cells per well in complete RPMI 1640 medium with 100 nM phorbol 12-myristate 13-acetate (PMA) (Cayman) and differentiated into macrophages for 3 days, then rested in PMA-free medium for 16 h before infection. Macrophages were infected with S. aureus strains at MOI 10–20 via centrifugation at 1,000 g for 2 min, followed by 35 min incubation at 37 °C. After removing the spent medium, cells were incubated in 500 µl complete MEM containing antibiotics (10 µg ml⁻¹ Rif, 50 µg ml⁻¹ Mox or 20 µg ml⁻¹ Van) with or without 0–100 µM KL1 or its analogues (KL2–7), 20 µM BHA (Sigma) or 0–10 µM BIX-01294, in the presence of 50–100 µg ml⁻¹ Gen at 37 °C for 6–24 h. Cells were washed three times with 1 ml PBS and lysed with 200 µl 0.5% (v/v) Triton X-100 (Fisher Scientific) at 37 °C for 5 min, followed by the addition of 800 µl PBS. Released intracellular bacteria were serially diluted and plated on TSA. Persister frequencies were normalized to the intracellular c.f.u.s at the time of antibiotic exposure. For comparisons, final bacterial loads were normalized to untreated controls.

For Salmonella infections, immortalized BMDMs were seeded in 12-well plates (4 × 10⁵ cells per well) in RPMI 1640 supplemented with 2 mM L-glutamine and 10% FBS at 37 °C and 5% CO₂ for 24 h. Stationary-phase Salmonella Typhimurium were opsonized with 8% (v/v) mouse serum in RPMI 1640 for 20 min and used to infect immortalized BMDMs at MOI 15 (ref. ²⁶). Plates were centrifuged at 110 g for 5 min, then incubated at 37 °C for 30 min. Cells were washed once with PBS and treated with 5 µg ml⁻¹ ciprofloxacin (MP Biomedicals) in the presence of 10–80 µM KL1 or 0.2% DMSO for 24 h. Cells were washed three times in 500 µl PBS, lysed with 0.2% Triton X-100 at 37 °C for 10 min and plated on LB agar. Persister frequencies were calculated as c.f.u.s at 24 h (T₂₄) over input c.f.u.s (T₀).

For M. tuberculosis, bacterial cultures were washed twice with PBS, passed through a 5-µm filter and resuspended in DMEM⁷². BMDMs were infected at MOI 1 for 4 h, washed three times with PBS and treated with 200 µg ml⁻¹ Gen for 2 h. Cells were then incubated in DMEM containing 0.1% DMSO, 1 µg ml⁻¹ Rif, 100 µM KL1, or a combination of Rif and KL1 in the presence of 20 µg ml⁻¹ Gen in triplicate at 37 °C for 3 days. Cells were washed three times with PBS, then lysed with 200 µl of 0.1% Triton X-100 for 5 min. The extracted intracellular bacteria were plated on Middlebrook 7H10 agar plates (BD Difco). The number of viable bacteria was enumerated after 21 days of incubation at 37 °C and 5% CO₂.

Assessment of adjuvant activity in primary human neutrophils

Blood samples from healthy donors were processed immediately following a standard protocol⁷³. Whole blood was mixed 1:1 with 3% dextran T-500 (Pharmacosmos)/0.9% NaCl and incubated at room temperature for 20 min to sediment erythrocytes. Leucocyte-rich supernatants were transferred to 50-ml tubes and centrifuged at 250 g for 10 min at 4 °C. Pellets were resuspended in 0.9% NaCl equal to the starting blood volume. Ficoll-Paque Plus (Cytiva, Fisher Scientific) was underlaid (10 ml per tube) beneath the cell suspension, followed by centrifugation at 400 g for 40 min at 20 °C (no brake). Pellets were resuspended in 20 ml cold 0.2% NaCl for 30 s, followed by addition of 20 ml cold 1.6% NaCl to lyse remaining erythrocytes. Neutrophils were pelleted at 250 g for 6 min, washed in RPMI 1640 with 2% FBS, 2 mM L-glutamine and 10 mM HEPES, and centrifuged at 250 g for 4 min. Cells were resuspended in the same medium, rested at 37 °C for 0.5–1 h, filtered through 100-μm strainers (VWR) and dispensed at 1 × 10⁵ neutrophils per well (500 µl) in 24-well ultra-low attachment plates. Cells were pretreated with 100 µM KL1 or 0.25% DMSO at 37 °C for 1–1.5 h in 6 replicates, infected with S. aureus JE2-lux (MOI 10) at 37 °C for 1 h, followed by treatment with 50 µg ml⁻¹ Mox or 10 µg ml⁻¹ Rif and 25 µg ml⁻¹ Gen for 4 h. Cell suspensions were serially diluted and plated on TSA with 0.4% activated charcoal (Sigma) to neutralize residual antibiotics, allowing enumeration of surviving bacteria. Neutrophils from ≥4 donors were tested.

Antibiotic survival assays in human PBMC-derived macrophages

Human PBMCs were isolated from fresh blood samples using Ficoll-Paque density gradient centrifugation, as described for neutrophil isolation. The interphase was collected into five volumes of complete RPMI 1640 and centrifuged at 400 g for 10 min. Cells were washed twice with 20 ml PBS at 300 g for 5 min and rested in complete medium at 37 °C for 30 min. Debris was removed using 100-μm cell strainers. For differentiation, 1–2 × 10⁵ PBMCs were seeded in 1 ml complete medium containing 1× penicillin–streptomycin and 20–50 ng ml⁻¹ GM-CSF (PeproTech) per well in 24-well plates for 4–6 days, with medium changed every other day. On day 4 or 6, macrophages were polarized with or without 100 ng ml⁻¹ lipopolysaccharide (LPS) (Sigma) and 20–50 ng ml⁻¹ IFNγ (PeproTech) for 1–2 days. Polarized macrophages were stimulated with 500 nM PMA at 37 °C for 2–3 h and washed three times in RPMI 1640. Macrophages (polarized and non-polarized, stimulated and unstimulated) were pretreated with 40–100 µM KL1, 20 µM BHA or 0.1–0.25% DMSO in 500 µl complete medium at 37 °C for 1 h in triplicates. Macrophages were infected with S. aureus at MOI 10 by centrifugation at 1,000 g for 2 min and incubated at 37 °C for 35 min. Cells were subsequently treated with 10 µg ml⁻¹ Rif and 50–100 µg ml⁻¹ Gen in the presence of KL1, BHA or DMSO, and incubated at 37 °C with 5% CO₂ for 6–18 h. Cells were washed three times with PBS, lysed in 200 µl 0.5% Triton X-100 for 5 min, diluted with 800 µl PBS and plated on TSA.

S. aureus murine infection and isolation of kidney cells

To visualize viable intracellular S. aureus, C57BL/6J female mice (6–9 weeks old) were systemically infected with Tet-inducible GFP-expressing S. aureus strain HG003 (5 × 10⁶ c.f.u.) via tail vein injection⁷. At 1 dpi, mice received 10 mg kg⁻¹ Rif in 2.5% DMSO, 12.5% PEG300 (Sigma) and 85% sterile H₂O via intraperitoneal injection. Control mice received vehicle (2.5% DMSO/12.5% PEG300) only. At 2 dpi (24-h Rif treatment), kidneys were homogenized in 5 ml cold PBS with DNase I (50 U ml⁻¹, bovine pancreas; ThermoFisher) using a Stomacher 80 Biomaster (Seward) at fast speed for 2 min twice⁷⁴. Homogenates were filtered through 70-µm strainers and pelleted at 300 g for 5 min at 4 °C. After three PBS washes, cells were resuspended in cold PBS with 1% FBS and 50 µg ml⁻¹ Gen, with or without 2 µM aTc, for 3 h. Samples were stained with LIVE/DEAD fixable violet dead cell stain (1 µl ml⁻¹; ThermoFisher) at 4 °C for 30 min before ImageStream analysis.

To evaluate KL1 adjuvant activity in vivo, mice (male:female = 1:1) were infected with 2 × 10⁷ c.f.u.s of strain HG003 via intravenous injection. At 6 hpi, mice were intraperitoneally treated with 10 mg kg⁻¹ Rif, alone or with 100 mg kg⁻¹ KL1 in 7% DMSO, 40% PEG300 and 53% sterile H₂O. Rif and KL1 were given once daily and every 12 h, respectively, for 2 days. Organs were homogenized either by repetitively rolling a serological pipette over the samples in sample bags (liver) or by bead beating using a Precellys 24 Touch homogenizer (Bertin Technologies) at 5,000 r.p.m. for 25 s twice with a 5-s interval (spleen and kidney). Homogenates were serially diluted in 1% NaCl and plated on TSA to enumerate surviving bacteria. Bacterial burden (c.f.u.s g⁻¹) was normalized to tissue weight. Two independent experiments using 6–8 mice each (14 mice per group in total) from different litters were conducted.

For Kaplan–Meier survival analysis, mice (female) were infected with 2 × 10⁸ c.f.u.s of S. aureus JE2-lux. At 6 hpi, mice received either 100 mg kg⁻¹ KL1 or the vehicle (7% DMSO, 40% PEG300 and 53% sterile H₂O). At 12 hpi, a single dose of 1 mg kg⁻¹ Rif, alone or with 100 mg kg⁻¹ KL1, was administered. Mice were monitored twice daily for signs of disease and euthanized when they reached humane endpoints, defined as change in mobility, severe lethargy, dehydration and a loss of >20% body weight. Survival was recorded until natural death or humane euthanasia. Data from 2 independent experiments (12–13 mice per group) were pooled. Significance was determined using Mantel–Cox test.

Salmonella Typhimurium murine infection

S. Typhimurium cultures were back diluted to OD₆₀₀ = 0.1 in LB broth and grown to exponential phase at 37 °C with shaking. Bacterial cells were washed three times with PBS before infection. Female C57BL/6J mice (10 weeks old) were fasted for ≥4 h and infected orally with 1 × 10¹⁰ c.f.u.s in 200 µl PBS. At 2 dpi, mice received 150 mg kg⁻¹ cefotaxime (CTX), alone or with 100 mg kg⁻¹ KL1 in 7% DMSO, 40% PEG300 and 53% sterile H₂O via intraperitoneal injection, given twice daily for 2 and 6 days. Organs were homogenized in PBS using a Mixer Mill MM400 (Retsch) with 3.2-mm stainless steel beads (BioSpec) at 30 Hz for 2 min. Homogenates were serially diluted and plated on LB agar to enumerate surviving bacteria. Five mice per group were used.

Reactive species assays

Reactive species were measured using the luminescent probe L-012 (Wako Chemical) and the fluorescent dye fluorescein-boronate (Fl-B)⁶³. For L-012 probing, RAW 264.7 macrophages were seeded in 100 µl complete MEM in Falcon 96-well white plates (Corning) at 3.84 × 10⁴ cells per well and incubated at 37 °C for 16 h. Cells were infected with S. aureus JE2 at MOI 10 by centrifugation at 1,000 g for 2 min, followed by a 40-min incubation at 37 °C. Cells were treated with 40–100 µM KL1 or 1–10 µM BIX-01294 in the presence of 50 µg ml⁻¹ Gen in 5 replicates for 4–8 h. After three washes with 200 µl prewarmed PBS, 100 µl of 300 µM prewarmed Hanks’ balanced salt solution (Gibco) was added, and luminescence was immediately measured at 37 °C using a Synergy H1 reader. Two consecutive reads were averaged.

For Fl-B staining, macrophages were similarly seeded in 96-well black clear-bottom plates. After treatment with 40–100 µM KL1, 100 µM KL2, 40–100 µM inactive analogue KL7, 1–10 µM BIX-01294 or 20 µM BHA, infected cells were washed twice with 200 µl PBS and incubated with 100 µl of 25–50 µM Fl-B in PBS at 37 °C for 30 min. Following three washes, fluorescence was measured at 535 nm emission with 485 nm excitation (gain 100, area scan) using the Synergy H1 reader.

ATP measurement

S. aureus strains LAC, HG003 and JE2 were grown in TSB for 16–18 h and subcultured at 1:1,000 to 1:40,000 dilution in fresh TSB containing 40 µM KL1 or 0.1–0.5% DMSO in 6 replicates at 37 °C for 4 h. Cultures (100 µl per well) were aliquoted into Falcon 96-well white plates and mixed 1:1 with BacTiter-Glo reagent (Promega), then incubated in the dark on a rocker at room temperature for 5–20 min. Luminescence was measured at 1.25-mm depth with gain 250 using a Synergy H1 reader and normalized to c.f.u.s for relative ATP levels.

To correlate lux-based bioluminescence and bacterial metabolic activity, stationary-phase JE2 and JE2-lux cultures were subcultured at 1:200 in MHB, incubated at 37 °C for 3 h with shaking, then treated with 0–0.5 mM sodium arsenate (Sigma) for 30 min. Parallel cultures diluted 1:6 in PBS were incubated at 37 °C with shaking for 3 h, then treated with 1% glucose, 5 mM sodium pyruvate and 0.5% casamino acids, or vehicle for 30 min. Bioluminescence (JE2-lux) and ATP levels (JE2) were measured immediately. Bacterial suspensions were plated for c.f.u. enumeration.

Seahorse analysis

This analysis measures real-time changes in cellular metabolism and has been applied to study S. aureus respiration⁷⁵. Stationary-phase cultures were diluted 1:100 in fresh TSB in triplicate and incubated for 2–3 h until OD₆₀₀ reached 0.3. Cultures were then diluted 1:500 to 1:1,000 in TSB or Seahorse XF DMEM (Agilent) supplemented with 100 mM glucose, 2 mM L-glutamine and 1 mM sodium pyruvate, and dispensed into poly-D-lysine (PDL)-coated Seahorse XF HS miniplates (100 µl per well). PDL coating involved adding 100 µg ml⁻¹ PDL in sterile H₂O for 30 min, followed by two washes with H₂O and air-drying. Bacteria were adhered by centrifugation at 1,400 g for 10 min. An additional 80 µl of medium was added (180 µl total), and miniplates were loaded into a calibrated Seahorse XF HS Mini Analyzer (Agilent). KL1 (40 µM) or 0.1–0.5% DMSO was injected, and oxygen consumption rates (OCRs) and extracellular acidification rates (ECARs) were monitored at 37 °C for 4–6 h. Data were analysed using Seahorse Analytics and plotted with GraphPad Prism.

Transcriptomic analysis

RAW 264.7 macrophages were seeded in 6-well plates (8 × 10⁵ cells per well) in 4 ml complete MEM and incubated at 37 °C and 5% CO₂ for 16 h. Cells were infected with S. aureus JE2-lux at MOI 20 for 35 min, then incubated in fresh media with 100 µg ml⁻¹ Gen and 40 µM KL1 or 0.1% DMSO for 24 h. Cells were washed with PBS, dissociated in 10 mM EDTA (2 ml per well) at 37 °C for 5 min, and washed three times in cold PBS. Viable cells (2.4–4.2 × 10⁶) were pelleted and submitted to Azenta for RNA sequencing (paired-end, 30 million reads per sample). Reads were aligned to the mouse GRCm38 genome using STAR v.2.5.2b.

Databases

Biological test results of the identified compound KL1 (PubChem CID: 2881454) were retrieved from PubChem (https://pubchem.ncbi.nlm.nih.gov/)^39,76. Gene expression profiles were examined using Expression Atlas (https://www.ebi.ac.uk/gxa/home)⁷⁷. Biological functions and subcellular localization data were sourced from the UniProt Knowledgebase (https://www.uniprot.org/)³⁶. Protein association networks among the differentially expressed genes were analysed using STRING (https://string-db.org/)⁷⁸. Chemical structures were generated using ChemDraw software 21.0.0 (PerkinElmer).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.