Peptide-level reanalysis finds a small cleavage shift but no major accessibility bias under osmotic stress

Abstract

Environmental stress can alter protein folding, molecular interactions, aggregation state, and protease accessibility before proteomic sample preparation. I reanalyzed the publicly available *Yersinia pseudotuberculosis* TMT11plex proteomics data associated with Süer et al. (PNAS, 2026) to test whether the 0.5 M NaCl osmotic-stress condition produced a peptide-level proteolytic-accessibility effect large enough to materially bias protein quantification. The biological context includes the stress-specific stimulon framework previously developed in Co-PATHOgenex (Fernandez et al., 2024). In the source PNAS study, osmotic-stress stimulon genes showed marked mRNA-protein decoupling across the three pathogens examined. The present secondary analysis focuses specifically on whether altered peptide accessibility could account for a substantial part of the proteomic signal in Y. pseudotuberculosis. I evaluated four complementary signatures using the public protein-, peptide-, and peptide-spectrum-match-level outputs: reporter dropout, missed-cleavage behavior, sensitivity of protein-level summaries to removal of missed-cleavage peptides, and condition-specific within-protein peptide heterogeneity. Missed-cleavage reporter signal increased reproducibly across the three biological replicates by approximately 0.36-0.51 percentage points, and missed-cleavage peptides showed a small positive shift relative to fully cleaved peptides. However, excluding all missed-cleavage peptides changed protein-level peptide summaries only minimally (median absolute change 0.0185 log2; approximately 1.3% on a linear scale), while all-peptide and fully-cleaved-only summaries remained highly concordant (Pearson r = 0.985). Osmotic stress also did not show unusually high within-protein peptide dispersion relative to the other stress conditions and ranked near the middle of the ten tested conditions. These results indicate that a small cleavage-state difference is detectable, but the available peptide-level evidence does not support a major proteolytic-accessibility bias as an explanation for the protein-level quantitative signal. In this specific respect, the reanalysis independently supports the robustness of the source study's findings. A key limitation is that treatment and TMT reporter channel were fixed together across replicates, so the small cleavage shift cannot be uniquely separated from a reporter-channel effect. This work is an independent secondary analysis of publicly available data. It has not been peer reviewed, and the source study authors were not involved in the reanalysis.