Introduction
Single cell proteomics (SCP) by mass spectrometry (MS) is an emerging analytical approach that continuously pushes the limits of technology towards improved reproducibility, throughput and sensitivity. Unlike conventional bulk proteomics, which measures the average protein composition of large populations of cells, single-cell proteomics provides information about cellular heterogeneity and distinct molecular states. This is particularly important because genetically similar cells can exhibit substantial differences in protein expression, signalling activity, and functional behaviour. Mass spectrometry (MS)-based single-cell proteomics combines highly sensitive mass spectrometric analysis with miniaturised sample preparation and liquid chromatography to identify and quantify proteins from the very small amounts of material present in individual cells. Recent advances in MS instrumentation, sample handling, and computational analysis have substantially increased the number of proteins that can be detected per cell. Consequently, single-cell proteomics is becoming an increasingly valuable tool for investigating complex biological processes, including cancer, immune-cell function, development, cellular responses to drugs, and anti-biotic resistance. [1-4]
As the protein content of a single eukaryotic cell is very low (in the 200pg range), SCP requires state-of-the-art equipment and workflows to reduce sample loss in an extremely low input setting. Previous successful approaches have included semi automated sample preparation within minimal volumes. [5]
While eukaryotic SCP is already a challenging task necessitating the most advanced hardware, downscaling this technique to bacteria pushes the boundaries of current technology. With an expected volume of ~2 fL, bacteria are approximately 1000 times smaller than eukaryotic cells (~2pL), and thus contain much less protein, with bacterial cell content estimated to be below 1pg protein even for actively growing cells. [1]
Major Challenges
The major challenges faced with bacteria includes efficient cell sorting due to the exceptionally small cell size, as well as sample loss during handling and sample contamination.
Gram-positive bacteria (e.g., Bacillus subtilis) are surrounded by a thick peptidoglycan cell wall, while Gram-negative bacteria (e.g., Escherichia coli) have a peptidoglycan wall, a periplasmic compartment, and an outer membrane. Therefore cell lysis needs to be compatible with down stream MS applications and provide effective sample preparation by breaking down the bacteria cell wall. [1]
The emergence of this analytical approach to proteomics is a unique one. Only one other publication attempts the first steps towards single cell bacteria proteomics. Using an alternative modified SCoPE MS approach, 12 confident proteins were confirmed from a single bacterium. However, such DDA-TMT based approaches suffer from a bias towards quantifying peptides present in the carrier proteome, which hampers potential detection of new or rare cellular subpopulations. [6]
Single Cell Proteomics (SCP)
Expanding on the results from Végvári, Á et al (2023), and establishing robust bacterial SCP will be critical in enabling direct proteomic studies of bacterial heterogeneity at the single-cell level. Individual bacteria have the ability to shift metabolic states, thereby allowing rapid adaptation to environmental changes, such as nutrient and temperature stresses. Most importantly, this plays a crucial role in persister cell formation during antibiotic stress and during infection. [7-9]
Analysis of bacterial heterogeneity at the single-cell level is already well established using fluorescence-based phenotypic observations of cellular processes, as well as through transcriptomics and metabolomics. Integrating SCP with these existing approaches would provide a critical layer of information for understanding bacterial heterogeneity.
Workflow Optimisation
Collaborators from the Research Institute of Molecular Pathology (IMP), Vienna, set out to investigate how to push the sensitivity of SCP to a level where single bacteria can yield meaningful results, establishing the first label-free proof-of-concept approach for bacterial single-cell proteomics (bacSCP). Researchers initially combined and adopted a robust and sensitive SCP sample preparation workflow with the capability to isolate individual bacterial cells.
Sample preparation parameters for bacSCP were tuned using E. coli cells. Cell lysis was enabled by the MS-compatible detergent N-Dodecyl-β-D-maltoside (DDM) and incubated at 50°C. Subsequently, bacteria were treated with Lysozyme to generate protoplasts or spheroplasts surrounded by a cell membrane only. The resulting cell-wall-deficient bacteria are spherical-shaped and of <2µm diameter. However, accurate isolation proved challenging using this method.
Alternative cell isolation and lysis procedures were explored. Notably, those previously demonstrated by Végvári et al. [6] Repeated freeze-thaw cycles followed by heating of intact bacteria lead to successful lysis. Furthermore, using cephalexin prior to cell wall removal, resulted in enlarged spheroplast cells and improved detectability. (See Fig. 1 below)
As a result, the above-mentioned cell-isolation and lysis conditions using E. coli cells led to the identification of 34 to 67 quantified E. coli protein groups per cell. The additional enlargement of spheroplasts using cephalexin, yielded 92 quantified E. coli protein groups per cell. Included within the results was the CRAPome [10] database. This was in order to account for contaminants, mainly originating from added trypsin as well as human skin keratins. It stands to reason therefore, that when dealing with samples of microscopic proportions, even a single dust particle could severely skew the final results. As expected, up to 98% of the total protein quantity detected in single cells originated from contaminants, reflecting both the low protein content of individual bacteria and the necessary large excess of trypsin added per sample. [1] By excluding the contaminants, an improved comparison can be ascertained and an estimate of 1–5pg of protein content can be deduced from a single bacteria.
Based on the success of isolation as confirmed by fluorescence and the aim to remain as close as possible to native conditions prior to lysis, intact stained cells were selected as the preferred condition. A subsequent bacSCP workflow using gram positive B. subtilis bacteria resulted in the quantification of ~36 bacterial proteins on average per cell.
It is suspected that the visible heterogeneity might at least partly result from technical variance and partly from biological heterogeneity. To confirm this hypothesis, future studies would require significantly enlarged sample sizes in order to improve statistical outcomes. [1]
Discussion
The study by Leodolter, J. et al (2026) is considered to be the first demonstration of bacterial single-cell proteomics capable of detecting stress-induced proteomic changes. The data provides evidence of biological regulation at the single bacterium level; and valuable first insights towards cell-to-cell heterogeneity present within cultured bacteria.
The team successfully demonstrated that bacSCP is applicable to both E.coli and B.subtilis, representing Gram-negative and Gram-positive model organisms. Non-contaminant protein detection in E. coli and B. subtilis demonstrated comparable depth with ~55 and ~65 protein groups respectively.
Being able to identify predominantly metabolic enzymes and ribosomes, and in the case of heat-stressed B. subtilis, stress response chaperones; highlights the sensitivity of the method implemented. The very low input sample and the high technical variability suggests some observed effects could be technical rather than biological. To more confidently distinguish between biological and technical variance, future investigations should consider a larger sample size. Applying bacSCP to a broader range of bacterial species would provide deeper insight into the robustness of the workflow.
Additional applications of this method could provide valuable insights into both virulence and antibiotic susceptibility of pathogenic bacterial strains. Pathologically relevant proteins are often not among the most abundant proteins in the cell; therefore to further investigate heterogeneity in broader antibiotic tolerance and resistance, bacSCP will require further improvements in sensitivity.