Methylation Landscapes of Cartilage in Hip Osteoarthritis.

OBJECTIVE: To elucidate different methylation landscapes between cartilage of femoral neck fracture and preserved and damaged cartilages in hip osteoarthritis (OA). METHODS: Genome-wide DNA methylation data were acquired from two data sets in GEO database (GSE63106 and GSE63695), which were based on Illumina HumanMethylation450 BeadChip arrays. A total of 63 hip samples were selected for further analysis, including 19 cartilages obtained from patients with femoral neck fracture, 14 preserved cartilages, and 30 damaged cartilages obtained from patients with OA. We identified the differential methylated positions (DMPs) and genes between different cartilage groups. RESULTS: There were 116,750 DMPs and 51,200 DMPs identified in preserved and damaged cartilages compared to cartilage in femoral neck fracture, respectively, while there were no signals found between preserved and damaged cartilages. Gene ontology analysis showed that most of differential methylated genes were enriched in extracellular matrix and structure organization, collagen-containing extracellular matrix, and KEGG enrichment highlighted PI3K-AKT and AMPK signaling pathways, which were known to be crucial for the progression of OA. Further construction of protein-protein interaction networks with differential methylated genes elucidated molecular basis of the disease. Three hypermethylated genes (NOTCH1, GREM1, and DYSF) and three hypomethylated genes (HDAC4, S100A10, and RUNX1) were selected to detect the relative expression in different cartilages, and their expression was correlated with the methylation status within the genes. CONCLUSION: We demonstrated the differential methylated genes across the whole genome not only on preserved cartilage but also on damaged cartilage during OA. The molecular network highlighted the potential therapy targets which may be involved in the initiation or progression of the disease.