Pulmonary arterial hypertension (PAH) is a devastating cardiovascular disease without a clear mechanism or drugs for treatment. Therefore, it is crucial to reveal the underlying molecular mechanism and identify potential drugs for PAH. In this study, we first integrated three human lung tissue datasets (GSE113439, GSE53408, GSE117261) from GEO. A total of 151 differentially expressed genes (DEGs) were screened, followed by KEGG and GO enrichment analyses and PPI network construction. Five hub genes (CSF3R, NT5E, ANGPT2, FGF7, and CXCL9) were identified by Cytoscape (Cytohubba). GSEA and GSVA were performed for each hub gene to uncover the potential mechanism. Moreover, to repurpose known and therapeutic drugs, the CMap database was retrieved, and nine candidate compounds (lypressin, ruxolitinib, triclabendazole, L-BSO, tiaprofenic acid, AT-9283, QL-X-138, huperzine-a, and L-741742) with a high level of confidence were obtained. Then ruxolitinib was selected to perform molecular docking simulations with ANGPT2, FGF7, NT5E, CSF3R, JAK1, JAK2, JAK3, TYK2. A certain concentration of ruxolitinib could inhibit the proliferation and migration of rat pulmonary artery smooth muscle cells (rPASMCs) in vitro. Together, these analyses principally identified CSF3R, NT5E, ANGPT2, FGF7 and CXCL9 as candidate biomarkers of PAH, and ruxolitinib might exert promising therapeutic action for PAH.
Keywords: DEGs; PAH; bioinformatics; hub gene; molecular docking; potential drugs.