Unbiased transcriptome mapping and modeling identify candidate genes and compounds of osteoarthritis

Abstract

Osteoarthritis (OA) is a chronic degenerative joint disease characterized by progressive cartilage loss, subchondral bone remodeling, and synovial inflammation. Given that the current therapies for advanced OA patients are limited, the understanding of mechanisms and novel therapies are urgently needed. In this study, we employed the weighted gene co-expression network (WGCNA) method and the connectivity map (CMap) database to identify the candidate target genes and potential compounds. Four groups of co-expressing genes were identified as the OA-related modules. The biological annotations of these modules indicated some critical hallmarks of OA and aging, such as mitochondrial dysfunctions and abnormal energy metabolism, and the signaling pathways, such as MAPK, TNF, and PI3K/Akt signaling pathways. Some genes, such as RELA and GADD45B, were predicted to extensively involve these critical pathways, indicating their potential functions in OA mechanisms. Moreover, we constructed the co-expressing networks of modules and identified the hub genes based on network topology. GADD45B, MAFF, and MYC were identified and validated as the hub genes. Finally, anisomycin and MG-262 were predicted to target these OA-related modules, which may be the potential drugs for OA therapy. In conclusion, this study identified the significant modules, signaling pathways, and hub genes relevant to OA and highlighted the potential clinical value of anisomycin and MG-262 as novel therapies in OA management.