3.1 Identification of FLS and celastrol treatment
After confirming that the FLS expressed the marker vimentin (Figure 1A), we treated them with 5 μM celastrol for 24 h. This treatment reduced cell viability, reflected in a greater number of dead cells than in the control cultures (Figure 1B).
3.2Gene and protein expression in FLS in response to celastrol
Our analysis detected 26,565 expressed genes and 3,372 expressed proteins (in the form of 18,843 unique peptides) (Figure 2A-B). The Mascot search algorithm was used to identify proteins.
Celastrol down-regulated 3,324 genes and up-regulated 4,479 genes, corresponding to a total of 7,803 differentially expressed genes (DEGs) (Figure 2C). The drug up-regulated 106 proteins and down-regulated 188, corresponding to a total of 294 differentially expressed proteins (DEPs) (Figure 2D).
3.3 Gene ontology of DEGs and DEPs
The distribution of DEPs and DEGs was analyzed across the three gene ontology categories of biological processes, cellular components, and molecular functions (http://geneontology.org). Among biological processes, DEGs and DEPs were enriched in processes related to response to oxidative stress, response to oxygen levels, extracellular structure and regulation of small-molecule metabolic processes. Among cellular components, DEGs and DEPs were enriched in collagen-containing extracellular matrix, focal adhesion, cell-substrate adherens junction and cell-substrate junction (Figure 3A). Among molecular functions, DEGs and DEPs were enriched in binding of cell adhesion molecules, ubiquitin-like protein transferase activity and ubiquitin-like protein transferase activity followed by structural components of extracellular matrix.
3.4 Correlation between proteomic and transcriptomic analyses
We observed a weak correlation between levels of transcripts and levels of their cognate proteins when we examined the entire set of genes and proteins (r = 0.2067), as well as the subset of DEGs and DEPs (r =0.2507). Altogether, we identified 36 genes up-regulated and 15 genes down-regulated at both the RNA and protein levels in response to celastrol. We identified another 27 genes that were transcriptionally down-regulated but translationally up-regulated, and two genes that were transcriptionally up-regulated but translationally down-regulated (Figure 4).
3.5 Docking of celastrol in target proteins
Our transcriptomic and proteomic analyses suggested that celastrol acts against several proteins involved in the extracellular matrix of FLS. Therefore, we explored the potential binding of celastrol to several such proteins whose crystal structures are known (Figure 5). In all cases, we managed to obtain reasonable complexes in which the primary forms of interaction are hydrogen bonding and hydrophobic interactions. For instance, celastrol is predicted to interact with CHI3L1 via hydrogen-bonds with Ser103, Gly143, Ser187, and Arg144 (Figure 5A). The drug is predicted to interact with ANXA6 via hydrogen-bonds with Lys224 and hydrophobic interaction with Glu643 (Figure 5B).
(A) Chitinase 3 like 1 (PDB ID: 1HJW).
(B) Cartilage oligomeric matrix protein (3FBY).
(C) Matrix metalloprotease 1 (1SU3)
(D) Laminin subunit gamma 1 (5XAU)
(E) Transforming growth factor beta-induced (5NV6)
(F) Annexin A6 (1M9I)
(G) Thrombospondin 2 (2RHP)
(H) Thrombospondin 1 (2OUH).