*3.4. Proteomic Analysis of EVs from hHSC*

We next used MS to investigate the protein components in EVs from LX-2 hHSC (Supplemental Table S4). Analysis of three separate LX-2 hHSC EV samples resulted in the identification of between 567 and 762 proteins, 524 of which were common to all three samples (Figure 7A). In light of the profibrogenic actions of EVs from activated mHSC or hHSC (Figure 2E), it was interesting that of the 524 LX-2 hHSC EV proteins, 206 were shared with the 337 proteins in EVs from P1 mHSC (Figure 7B). Only 26 LX-2 hHSC proteins were shared with EVs from D4 mHSC, (Figure 7B). All three types of EVs had 21 proteins in common with one another (Figure 7B). Quantitative analysis (Figure 7C–F) showed that when proteins in each group were ranked based on expression levels in LX-2 EVs, those shared with EVs from D4 mHSC but not P1 mHSC had the lowest expression (five proteins; quantitative value range = 8–80), while those shared with EVs from P1 mHSC but not with EVs from D4 mHSC had the highest expression (185 proteins; quantitative range = 100–1000 for the top 20) (Figure 7C,F). The most abundant protein was FN1 (Figure 7F), the presence of which was confirmed in LX-2 hHSC EVs by Western blot analysis (Figure 2F). Top-ranked proteins specific to LX-2 hHSC EVs had an intermediate level of expression (313 proteins; quantitative value 50–200 for the top 20) while the proteins in LX-2 hHSC EVs that were shared with EVs from both D4 and P1 mHSC were very variably expressed (21 proteins; quantitative range 5–500) (Figure 7D,E).

**Figure 7.** Proteomic composition of EVs from LX-2 hHSC and/or D4 or P1 mHSC. (**A**) Summary of quantitative features of EV proteins analyzed from LX-2 hHSC EV samples. (**B**) Venn diagram showing distribution of proteins between EVs from LX-2 hHSC, D4 mHSC and P1 mHSC. Also shown are the identities and quantifications of (**C**) the five proteins shared between LX-2 hHSC EVs and D4 mHSC EVs; (**D**) the 20 most abundant proteins specific to EVs from LX-2 hHSC; (**E**) the 21 proteins shared by EVs from all three cell types; and (**F**) the 20 most abundant proteins shared by EVs from LX-2 hHSC and P1 mHSC.

GO analysis was performed on the total proteins identified in EVs from activated human or mouse HSC, as well as those that were shared or not shared between them (Figure 8). For the entire EV proteome from LX-2 cells, the principal components were related to exosome, cytoplasm, cytosol, membrane, nucleus, ECM, focal adhesion, extracellular space, cell surface, cell-cell adherins, perinuclear region of cytoplasm, mitochondrion, nucleoplasm and myelin sheath (Figure 8A). Many of these components were shared with P1 mHSC EVs but additional components in this shared group included melansome, actin cytoskeleton and proteosome complex (Figure 8B). The same or highly similar components were represented by the proteins that were unique to EVs from either LX-2 cells (Figure 8C) or P1 mHSC (Figure 8D). The outcome of this component analysis was largely reflected in the KEGG analysis of the respective EV protein groups (Figure 9). For example, the top 20 KEGG pathways for the complete LX-2 hHSC EV proteome (Figure 9A) showed substantial overlap with the pathways for the P1 mHSC EV proteome (Figure 6A) and included focal adhesion, PI3K-Akt signaling, regulation of actin cytoskeleton, pathways in cancer, proteoglycans in cancer, ribosome, phagosome, leukocyte transendothelial migration, protein processing in the ER and carbon metabolism. These pathways were also predominant for the 206 proteins shared between LX-2 hHSC EVs and P1 mHSC EVs (Figure 9B). Some of the same or related components were evident for LX-2 hHSC EV-specific proteins (regulation of cytoskeleton, pathways in cancer, ribosome, PI3k-AKT signaling, proteoglycans in cancer, phagosome, focal adhesion, cell adhesion molecules, alcoholism) (Figure 9C) or for P1 mHSC EV-specific proteins (proteasome, ECM-receptor interactions) (Figure 9D) but the remaining components in these groups were quite dissimilar and diverged from those typically noted above. Finally, STRING analysis of the entire LX-2 hHSC EV proteome (Supplemental Figure S4) or of the EV proteins shared between LX-2 cells and P1 mHSC (Supplemental Figure S5) revealed major interactions between nodes associated with collagens, ECM, metabolic enzymes, vesicular transport, chaperones or ribosomes and these were very similar to those for the entire P1 mHSC EV proteome (Supplemental Figure S3).

**Figure 8.** Cellular component analysis of proteins in EVs from LX-2 hHSC and/or P1 mHSC. GO analysis was performed on proteins in EVs from LX-2 hHSC as in Figure 5. The figure shows the 20 most highly ranked cellular components for proteins that were (**A**) in the entire LX-2 HSC EV proteome; (**B**) shared between EVs from LX-2 hHSC and P1 mHSC; (**C**) specific to LX-2 hHSC EVs; or (**D**) specific to P1 mHSC EVs. Only cellular components with significant enrichment (*p* < 0.05) are shown.

**Figure 9.** KEGG pathway analysis of proteins in EVs from LX-2 hHSC and/or P1 mHSC. EV proteins were analyzed as in Figure 6. The figure shows the top 20 pathways identified for (**A**) the entire LX-2 hHSC EV proteome; (**B**) proteins shared between LX-2 hHSC EVs and P1 mHSC EVs; or (**C**) proteins unique to LX-2 hHSC EVs. (**D**) Pathways identified for EV proteins that were specific to P1 mHSC. Only pathways with significant enrichment (*p* < 0.05) are shown.
