Abstract: BACKGROUND: The etiology of knee osteoarthritis is complex and its mechanisms are not fully understood. Research on candidate target genes for knee osteoarthritis will help further clarify the pathogenesis of the disease and provide a basis for precision treatment.
OBJECTIVE: To identify therapeutic targets for knee osteoarthritis based on summary data using Mendelian randomization combined with bioinformatics methods, followed by cellular validation.
METHODS: Gene expression profiles GSE46750, GSE55235, GSE82107, and GSE206848 were downloaded from the Gene Expression Omnibus database. Differentially expressed genes were obtained using R software with screening criteria of |log2FC| > 0.585 and adjusted P < 0.05. Module genes with the highest correlation were acquired using the Weighted Gene Co-expression Network Analysis algorithm and intersected with differentially expressed genes. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analyses were performed on the intersecting genes. Genetic genes significantly associated with knee osteoarthritis were obtained from the eQTLGen database using summary-data-based Mendelian randomization analysis. Genes jointly identified by both bioinformatics and summary-data-based Mendelian randomization analysis were defined as core genes. The binding of celecoxib to core genes was evaluated via molecular docking and dynamics simulations. Immune infiltration analysis of core genes was performed using the CIBERSORT algorithm. Human chondrocytes were divided into a normal group and an experimental group (interleukin-1β-induced osteoarthritis cell models). The mRNA expression of adrenomedullin, human osteopontin, and lysosomal membrane protein 5 was detected by qPCR.
RESULTS AND CONCLUSION: Bioinformatics analysis identified 229 differentially expressed genes. Gene Ontology enrichment analysis showed that these genes were mainly associated with biological functions such as inflammatory response, positive regulation of response to external stimulus, regulation of cell activation, and chemotaxis. Kyoto Encyclopedia of Genes and Genomes enrichment analysis revealed significant enrichment in pathways including phagosome, osteoclast differentiation, complement and coagulation cascades, and interleukin-17 signaling pathway. Summary-data-based Mendelian randomization analysis identified 76 significantly associated genetic genes (P < 0.05, FDR < 0.05, HEIDI test P > 0.05). Adrenomedullin, human osteopontin, and lysosomal membrane protein 5 were identified as core genes. Human osteopontin and lysosomal membrane protein 5 were negatively correlated with the progression of knee osteoarthritis, while adrenomedullin was positively correlated with the progression of knee osteoarthritis. Molecular docking and molecular dynamics simulations confirmed favorable structure-activity relationships between the core genes and celecoxib. Immune infiltration analysis suggested that adrenomedullin, human osteopontin, and lysosomal membrane protein 5 were correlated with multiple immune cell types. qPCR detection showed that the mRNA expression of human osteopontin and lysosomal membrane protein 5 in the experimental group was lower than that in the normal group (P < 0.001), while the mRNA expression of adrenomedullin was higher than that in the normal group (P < 0.001). These findings indicate that adrenomedullin, human osteopontin, and lysosomal membrane protein 5 are key genes in the progression of knee osteoarthritis and may serve as promising novel targets for its prevention and treatment.

Key words: gene expression profile, differential genes, key genes, knee osteoarthritis, summary-data-based Mendelian randomization analysis, molecular docking, molecular dynamics, experimental validation