Figure 3

Proteomics Analysis Indicated that Disrupting the TLR4 Gene Reduced Secretion of Pro-Inflammatory Proteins. To determine differences in the proteome of free proteins in growth media and purified EVs released from edited and unedited hMSCs, we carried out a comparative MS proteomic analysis. Data are available via ProteomeXchange with identifier PXD033253. (A) Heat map showing the levels of proteins secreted by hMSCs with and without TLR4 gene editing from four patients shown as Z-scores (abundance between -2 and 2). (B) Heat map of EV-encapsulated proteins in EVs secreted by hMSCs with and without gene editing of cells from four patients shown as Z-scores (abundance between -2 and 2). (C) To graphically present the quantitative data, we constructed a volcano plot (log₂ fold-change edited vs. unedited cells, plotted against log₁₀ of statistical difference). For free-protein secretion, q-value was used to determine the statistical strength of protein identification, with ±1 as the cutoff region for significant changes in secretion after editing. (D) For EV protein content, the p value was used for statistical strength of protein identification, with ±0.8 as the cutoff region of significant changes. Points above the non-axial vertical line at 0.05 represent proteins with significantly different abundances (p < 0.05). Results show a significant reduction in the release of proteins involved in immune regulation (red) and extracellular organization pathways (orange). n = 4. (E) The number of validated protein signatures found in each experimental group, separately and combined. The complete list of protein names is specified in Supplementary Table 1. Abbreviations TLR4, toll-like receptor 4; hMSCs, human mesenchymal stromal cells; EVs, extracellular vesicles; FBS, fetal bovine serum; MS proteomics, mass spectrometry proteomics.