3.1 Molecular Targets Of GCK And Osteoporosis
A total of 206 and 6590 molecular targets were obtained for GCK and osteoporosis in this study, respectively. The overlapping targets of GCK and osteoporosis were considered as the potential targets of GCK-treated osteoporosis. Based on intersection analysis, a total of 138 molecular targets were identified as co-targets of GCK and osteoporosis, which were shown through a Venn diagram (Fig.1).
3.2 PPI Network Analysis Of Co-targets Of GCK And Osteoporosis
For investigating internal connection and important targets in these co-targets of GCK and osteoporosis, PPI network analysis was applied further here. After importing GCK-osteoporosis co-targets into STRING, we get GCK-osteoporosis target PPI network with the highest confidence (p≥ 0.900), which contains 95 nodes and 365 edges (Fig.2). In PPI network, the degrees of node PIK3R1, PIK3CA, STAT3, SRC, GRB2, PLCG1 and VEGFA were greater than 20, which was 35, 34, 30, 23, 22, 21 and 20, respectively. Targets information in PPI network queried from STRIING database was analyzed further by MCODE and cytoHubba tool of Cytoscape software for identifying the key targets. MCODE analysis returned five central gene clusters (Fig.3 and Table 1). In these clusters, just the top 1 module was with score > 6 in this PPI network (Fig.3a and Table 1), which contained SYK, STAT3, PLCG1, PIK3CD, PIK3CB, MAP2K1, JAK1, IL2, HSP90AA1, HCK, GRB2. Additionally, STAT3, PIK3R1, VEGFA, JAK2 and MAP2K1 were identified as hub genes in PPI network with cytoHubba tool based on six methods (Fig. 4 and Table 2), which involved in the three of five central gene clusters identified by MCODE analysis (Fig. 3a, 3b and 3c). This suggests that the related molecules have an important role in GCK-treated osteoporosis, especially for STAT3, PIK3R1, VEGFA, JAK2 and MAP2K1.
3.3 GO And KEGG Analysis Of Co-targets Of GCK And Osteoporosis
Go and KEGG analysis were further used to investigate the BP, CC, MF and signaling pathway of co-targets of GCK and osteoporosis. The results indicated that 210, 55 and 66 terms were enriched for BP, CC and MF, respectively. The top 20 BP, CC and MF terms (which were ranked based on p value) were shown in Fig.6a, 6b and 6c, respectively. These enriched BPs mainly included cell growth- and death-related processes (such as negative regulation of apoptotic process, positive regulation of MAP kinase activity, positive regulation of cell proliferation, regulation of phosphatidylinositol 3-kinase signaling and so on), protein synthetic and modification processes (such as phosphatidylinositol-3-phosphate biosynthetic process, phosphatidylinositol phosphorylation, peptidyl-tyrosine (auto)phosphorylation, protein (auto)phosphorylation, protein processing, and so on ), and some stress response processes (such as drug, hypoxia, inflammatory and innate immune response). Based on CC enrichment, these co-targets were found in plasma membrane, nuclear membrane, organelle, cytoplasm, extracellular matrix, extracellular secretion and exosomes, in which targets from membrane were the most, and the second is from cytoplasm and extracellular components. In MF enrichment analysis, these co-targets mostly involved in kinase activity, protein/receptor/enzyme/small molecular-binding function. It was worth noting that phosphatidylinositol-related biosynthetic process, protein modification, complex assembly, mediated-signaling and kinase activities were among the top in BP, CC and MF enrichment analysis. This suggested that phosphatidylinositol-related bioprocess and signaling pathway may be a potential target for GCK to treat osteoporosis.
KEGG analysis showed that co-targets of GCK and osteoporosis mainly involved in 88 pathways (the top 20 pathways were shown in Fig.6), most of which were cancer-related pathways (such as pathways in cancer, proteoglycans in cancer, prostate cancer, pancreatic cancer, non-small cell lung cancer and so on). However, there was only one pathway (osteoclast differentiation pathway, shown in Fig.7) which was significantly related to osteoporosis. There were 16 genes (c-Fms, MAP2K1, SYK, PIK3CD, PIK3CB, PIK3R1, MAPK14, PIK3CG, MAPK12, IKBKB, MAPK8, PIK3CA, CAMK4, GRB2, PPARG, JAK1) enriched in osteoclast differentiation pathway with a p value of 4.81×10-9. In these 16 genes, just c-Fms was located in the cell membrane, and the others were located in the cytoplasm and nucleus. This indicated that c-Fms may play a more key role than other targets through regulating downstream signal transduction in the osteoclast differentiation pathway for GCK-treated osteoporosis.
3.4 Molecular Docking Of GCK- c-Fms Interaction
Based on the results mentioned above, c-Fms-mediated signaling may be the most significant through interfering osteoclast differentiation for GCK-treated osteoporosis. To validate possible biological interaction between GCK and c-Fms, molecular docking was used here. The results showed that GCK could well insert into and bind to the active cavity on the surface of c-Fms with -7.21 Kcal/moL of binding energy (Fig.8 a and 8 b). The complex GCK-c-Fms was stabilized by three strong hydrogen bonds with residues including Glu 664 (3.19 Å), Glu 664 (2.62 Å) and Cys 666 (2.78 Å) (Fig.8c), respectively. It confirmed that GCK could interfere osteoclast differentiation through interacting with c-Fms, by which it exerts the efficacy in the treatment of osteoporosis.