Abstract
Knee osteoarthritis (KOA) is a degenerative joint disease with a genetic component. Nonetheless, it remains largely unknown how risk variants influence the OA risk through their effects on proteins. This study aimed to identify new and effective drug targets for the diagnosis of KOA. A proteome-wide association study was conducted using summary datasets from genome-wide association studies of OA and protein quantitative trait locus data. The Gene Expression Omnibus dataset GSE63359 was used to identify differentially expressed genes between controls and KOA. Subsequently, Mendelian randomization and colocalization analyses were performed on the intersecting proteins identified using the aforementioned methods to assess the association between protein levels and KOA risk. Clinical samples were included, and RT-qPCR and enzyme-linked immunosorbent assay were used to validate the expression of risk proteins between the KOA and control groups. Extracellular matrix protein 1 (ECM1) expression was causally related to KOA. Further support from colocalization analysis suggested that ECM1 might serve as a potential drug target for KOA. Validation of clinical samples indicated that ECM1 levels were higher in the KOA group than in the control group and that ECM1 demonstrated significant diagnostic efficacy for KOA (area under the curve = 0.85). This study identified ECM1 as a risk factor for the pathogenesis of KOA and showed promising therapeutic targets for KOA treatment.