Host metabolic consequences of intracellular infection by members of the ESKAPE pathogens

Significance of the topic

Atypical and multidrug‑resistant bacterial pathogens of the ESKAPE group are major causes of hospital‑acquired infections. Several members, including Staphylococcus aureus (notably MRSA), can survive inside host cells, evading many antibiotic actions. Host‑directed therapies that modify host metabolism or nutrient availability are an attractive complement to classical antimicrobials because they may reduce selective pressure for resistance and target conserved host processes required for intracellular survival. This study applies high‑throughput host‑directed drug screening combined with a novel dual‑ionization GC‑time‑of‑flight metabolomics platform (GC‑ecTOF) to identify repurposable compounds that impair intracellular MRSA and to characterize host metabolic perturbations associated with infection and treatment.

Objectives and study overview

The primary goals were to:

• Screen a large library (5,599) of approved host‑targeted drugs for ability to reduce intracellular MRSA in A549 epithelial cells.
• Characterize host metabolic changes induced by MRSA invasion using GC‑ecTOF metabolomics.
• Identify candidate host‑directed drugs that modulate metabolic pathways and reduce intracellular bacterial burden in vitro and in vivo.

The screening pipeline combined high‑throughput fluorescence‑based quantification of GFP‑tagged MRSA and mCherry‑tagged host cells with secondary metabolomic and transcriptional profiling to propose mechanisms of action for prioritized hits. The nucleoside analogue 5‑fluoro‑2′‑deoxycytidine (5‑FdC) emerged as a lead repurposing candidate and was evaluated further in metabolic assays and murine infection models.

Methodology

Cellular infection model and screening:

• A549 human lung epithelial cells were infected with S. aureus USA300 LAC (GFP‑tagged) and host cells were mCherry‑tagged to enable ratiometric flow cytometric readouts of host viability and intracellular bacterial load in 96‑well format.
• Gentamicin protection assay: 1 h exposure to bacteria before gentamicin addition, MOI ~33, total invasion time 6 h.
• High‑throughput screen of 5,599 approved host‑directed compounds to identify those that decreased intracellular MRSA without overt host cytotoxicity.

Metabolomics sample preparation and analysis:

• Metabolite extraction from infected and treated A549 cells used acidified methanol:acetonitrile:water (2:2:1, v/v/v).
• Drying and two‑step derivatization: methoximation followed by tert‑butyl‑trimethylsilylation (TBDMS derivatization reported), enabling GC analysis of polar metabolites and fatty acids.
• GC‑MS analysis performed using both conventional single‑quadrupole GC‑MS and the novel GC‑ecTOF dual‑ionization time‑of‑flight platform for enhanced annotation confidence.

Data analysis and compound identification:

• Combined use of Electron Ionization (EI) and Chemical Ionization (CI) spectra obtained simultaneously from a single injection (GC‑ecTOF) provided both structural fragmentation patterns and molecular ion information for confident annotation.
• Retention index (RI), accurate mass (from CI), isotopic pattern fitting, and EI spectral matching (NIST) were integrated in a compound identification workflow to prioritize candidates.

Instrumentation used

Key instrumental parameters and components (summary of methods reported):

• GC‑ecTOF dual ionization GC‑TOF system (Bruker/Tofwerk ecosystem) enabling simultaneous EI (70 eV) and soft CI (NH4+ reactant) spectra from one injection.
• GC column: Rxi‑5MS, 30 m × 0.25 mm i.d., 0.25 µm film thickness.
• Injector: splitless, injection volume 1 µL; heated transfer lines ~180 °C.
• Ion source temperatures: EI 250 °C and CI 250 °C (StarBeam EI and HRP CI sources reported).
• Carrier gas: helium constant flow ~1.0 mL/min.
• GC oven program: 60 °C (1 min) then 10 °C/min to 325 °C, hold 10 min.
• Mass range acquisition: 1–1000 m/z; EI/CI switching speed ~10 Hz.

Main results and discussion

Host metabolic perturbations induced by intracellular MRSA:

• MRSA infection produced marked rewiring of host central carbon metabolism, including altered pool sizes of amino acids and changes in tricarboxylic acid (TCA) cycle flux indicators.
• Elevations in branched‑chain fatty acids (BCFAs) were observed in infected A549 cells, consistent with bacterial biosynthesis of membrane lipids derived from branched‑chain amino acids (BCAAs). BCFAs (notably anteiso‑C15:0 and anteiso‑C17:0) were detected inside host cell extracts and correlated with intracellular bacterial presence.
• Certain metabolites (for example purines and cystathionine) were only detected or confidently annotated using the ecTOF platform, highlighting the value of combined EI/CI data for expanded metabolome coverage.

Effect of 5‑FdC and mechanism‑related observations:

• 5‑FdC treatment reduced intracellular MRSA burden in A549 cells and modulated host central carbon metabolism, particularly amino acid pools and TCA‑related metabolites, suggesting interference with bacterial nutrient acquisition or host metabolic support pathways.
• Reduction of intracellular BCFAs in treated cells served as a metabolomic marker of decreased intracellular bacterial content after 5‑FdC exposure.
• In murine infection models, 5‑FdC co‑administration with antibiotics decreased bacterial loads in lung and spleen and improved health readouts compared with controls, supporting translational potential.

Analytical innovation and advantages:

• GC‑ecTOF’s simultaneous acquisition of CI (molecular ion/accurate mass) and EI (fragmentation) spectra from a single injection improved annotation confidence versus conventional single‑mode GC‑MS, enabling detection of metabolites otherwise missed.
• Integrated EI/CI/RI workflows yielded more robust identification of host‑pathogen metabolic alterations, facilitating mechanistic hypotheses for drug action.

Benefits and practical applications of the method

• High‑throughput host‑directed screening coupled with metabolomics provides a route to identify repurposable drugs that limit intracellular pathogens by targeting host pathways rather than bacterial targets, potentially lowering resistance selection pressure.
• GC‑ecTOF extends GC‑MS capabilities by capturing both structural and molecular ion data in a single run, improving metabolite coverage and identification reliability—valuable for complex host‑pathogen matrices.
• Detection of BCFAs in host cell extracts offers a practical biomarker for intracellular S. aureus burden and treatment response in cell models and possibly ex vivo samples.

Future trends and potential uses

• Further optimization and standardization of dual‑ionization GC‑TOF methods could broaden routine metabolomics applications in infection biology and drug screening workflows.
• Follow‑up mechanistic studies are needed to define the precise host targets and metabolic nodes modulated by 5‑FdC, and to evaluate safety, dosing, and combination regimens with conventional antibiotics.
• Translational development may explore BCFA quantification in tissue or biofluid samples as biomarkers for intracellular S. aureus infection and therapeutic monitoring.
• Applying similar integrated screening‑metabolomics pipelines to other intracellular ESKAPE pathogens could reveal shared host dependencies amenable to broad host‑directed interventions.

Conclusion

The integrated screening and GC‑ecTOF metabolomics pipeline identified 5‑FdC as a promising host‑directed agent that reduces intracellular MRSA burden and perturbs host central carbon metabolism linked to bacterial adaptation. The GC‑ecTOF dual‑ionization approach increased metabolite coverage and confidence in compound annotation, and branched‑chain fatty acids were highlighted as a metabolomic marker of intracellular S. aureus. These results support further mechanistic and preclinical development of host‑targeting strategies against intracellular Gram‑positive pathogens.

References

1. Sci Rep. 2019. 9(1):4876.
2. Arch Toxicol. 2019. 93(2):341–353.
3. mSphere. 2018. 3(4):e00374‑18.

Conflict of interest note: Several authors are employees of instrument vendors that provided access to the GC‑ecTOF system; this was disclosed by the authors and is relevant when interpreting performance claims.