A new high-resolution metabolomics platform uncovers hidden metabolic diversity in individual bacteria, revealing rare resistance-like states that could reshape understanding of persistent infections.
Study: Single-bacterium metabolome revealing heterogeneous cellular states in bacterial populations. Image credit: Tatevosian Yana/Shutterstock.com
In a recent study published in Nature Communications, researchers developed a highly sensitive metabolomics platform, SinBactM, enabling detailed metabolic profiling at the level of individual bacterial cells. The system combines induced nanoelectrospray ionization mass spectrometry (InESI-MS) with micro-extraction to map bacterial metabolites from individual cells while preserving overall bacterial morphology after extraction.
Applied across diverse species, SinBactM distinguishes distinct metabolic signatures. This morphology-preserving approach could facilitate future integration with complementary techniques such as single-bacterium RNA sequencing. It could also advance multi-omics studies and offer new insights into heteroresistance in specific contexts such as Klebsiella pneumoniae. This level of resolution may help in understanding why some bacterial infections persist despite seemingly effective treatment.
Limits Of Population-Level Metabolomics In Bacteria
Single-cell analysis in bacteria is crucial for uncovering metabolic diversity driving variation in growth, stress responses, and antibiotic resistance. Although bacterial metabolism underpins survival and adaptation, its organization at the single-cell level remains difficult to resolve. Metabolite distributions vary widely between cells and are closely linked to physiological states, including heteroresistance under metabolic stress.
Conventional mass spectrometry (MS) captures only population-level averages, masking this heterogeneity. While single-bacterium RNA sequencing provides indirect insights, it is limited by low transcript abundance and structural barriers such as cell walls. Optical methods, including Raman spectroscopy and fluorescence imaging, offer limited metabolic coverage and suffer from low specificity, leaving a key gap in single-cell metabolic profiling and limiting understanding of real-world treatment failure.
SinBactM Combines Micro-Extraction With InESI Mass Spectrometry
In this study, researchers developed SinBactM in-house to achieve highly sensitive individual-bacterium level metabolic profiling.
The system integrates pipette-based micro-extraction procedures with InESI-MS, enabling analysis of picoliter-scale samples without conventional liquid-phase separation. After the team immobilized single bacterial cells on coated glass slides, a fine glass capillary delivered extraction solvent to generate a tiny droplet over each cell. After extraction, researchers analyzed the droplet using InESI-MS. This approach concentrated metabolites into ~4.8–5.8 pL volumes, enhancing detection sensitivity.
The team applied SinBactM to multiple bacterial species, including Gram-negative and Gram-positive strains, to assess its broad applicability. They used dimensionality reduction and clustering methods to analyze metabolic fingerprints, allowing clear separation of species and subpopulations. To test clinical relevance, the researchers profiled colistin-heteroresistant Klebsiella pneumoniae (HR-S) and its resistant derivatives, identifying distinct metabolic states within and between strains.
The team further examined metabolic responses under antibiotic stress by treating bacteria with colistin and tracking time-dependent metabolic shifts. Unsupervised clustering revealed heterogeneous subpopulations, while pseudotemporal analysis mapped dynamic metabolic trajectories across cells.
To validate robustness, they applied supervised classification models and tested mixed bacterial populations, demonstrating approximate discrimination of subtypes based on single-cell metabolomic signatures. Additional experiments, including antibiotic accumulation assays (performed using ciprofloxacin in E. coli) and isotopic labeling, supported the reliability of the platform.
Single-Cell Metabolomics Reveals Species-Specific Signatures
The SinBactM platform successfully enabled sensitive, untargeted metabolomic profiling of bacteria at a single-cell level and demonstrated clear discrimination across multiple bacterial species. SinBactM resolved over 500 metabolic features in Bacillus subtilis. Of these, 141 ions were putatively annotated, including amino acids and polyamines.
Metabolic signatures were consistent within species but clearly separated between different bacterial species, reflecting robust biological identity. The team observed distinct clustering among Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, and Bacillus subtilis. They also observed species-specific enrichment of key metabolites such as spermidine in B. subtilis and phenylalanine in P. aeruginosa.
Importantly, SinBactM revealed pronounced metabolic heterogeneity within populations rather than uniform responses. Unsupervised analysis identified multiple subpopulations. Pseudotemporal analysis revealed progressive metabolic transitions, including a rare subpopulation with a resistant-like metabolic state.
In clinical K. pneumoniae isolates, the platform distinguished heteroresistant (HR-S) and resistant strains, identifying 82 significantly altered metabolites under antibiotic stress. HR-S subpopulations were distributed across eight distinct clusters. In contrast, resistant cells concentrated in a single cluster closely resembling a resistant-like HR-S subset. Pseudotime analysis mapped a trajectory of increasing resistance-associated metabolism. This was accompanied by shifts in amino acid and polyamine pathways.
Machine learning models predicted strain composition in mixed samples, where results were reasonably accurate but not exact. Overall, the results demonstrate that SinBactM can robustly resolve species identity, uncover hidden subpopulations, and track dynamic metabolic reprogramming under antibiotic pressure at individual-bacterium resolution.
SinBactM Enables High-Resolution Mapping Of Bacterial Metabolism
The study findings position the SinBactM platform as a powerful approach for single-bacterium metabolomic profiling and reveal marked metabolic heterogeneity within bacterial populations. The detection of rare, resistance-like subpopulations and continuous metabolic trajectories highlights a dynamic landscape of bacterial responses to antibiotic stress. While these findings provide important clues to heteroresistance, the functional links between specific metabolic states and resistance mechanisms remain to be validated.
Looking ahead, improving ionization efficiency, extraction strategies, and metabolite coverage will expand analytical depth. Current limitations include incomplete metabolite coverage, lack of chromatographic separation, and MS/MS validation affecting data certainty, potential salt interference, low throughput, and the need for improved metabolic quenching methods.
Enhancing sensitivity, throughput, and metabolic quenching methods will further strengthen the application in complex microbial systems.
The morphology-preserving design of SinBactM could facilitate integration with single-cell transcriptomics, enabling multi-omics analyses. These advances could transform the study of microbial ecology, host–pathogen interactions, and antibiotic resistance at single-cell resolution, with potential relevance for improving infection outcomes.
Download your PDF copy by clicking here.
Journal Reference
Zhan, L. et al. (2026). Single-bacterium metabolome revealing heterogeneous cellular states in bacterial populations. Nature Communications. DOI: 10.1038/s41467-026-72373-0. https://www.nature.com/articles/s41467-026-72373-0
New Methods Increase Understanding of How Bacteria Manage DNA