The combination of single-cell and Seq-RNA sequences revealed homeostatic chondrocyte osteoarthritic immune infiltrate

Abstract

Abstract Osteoarthritis (OA), a degenerative joint disease, is identified as the primary contributor to disability and pain. The significance of OA biomarkers lies in their role in comprehending the underlying mechanisms of the disease, facilitating the development of innovative targeted treatments, and potentially enhancing patient outcomes. Single cell sequencing was employed to identify genes expressed in a minimum of 5% of the homeostatic chondrocytes cluster. Subsequently, the hdWGCNA method was utilized to construct co-expression networks and facilitate differential expression analysis, enabling the identification of hub genes. Additionally, a machine learning approach was employed to screen 319 OA and normal samples from the GEO dataset, aiming to identify biomarkers capable of distinguishing between OA and normal samples. Furthermore, deep learning techniques were applied to validate the prognosis of OA and normal samples. The IOBR method was utilized to evaluate the infiltration of immune cells in OA tissues and explore the correlation between biomarkers and infiltrating immune cells. We identified 15 potential crucial genes (NCL, HNRNPA2B1, XBP1, IGFBP3, EIF2S2, CCT3, SERPINE2, ICAM1, TXNRD1, FHL2, PTPRE, GTPBP4, NRP2, LAMB3, LCN2) that could impact the prognosis of osteoarthritis and potentially contribute to the development of osteoarthritic joints. Furthermore, the involvement of immune cell infiltration in the initiation and progression of OA suggests that enhancing its management could be beneficial. In summary, our study integrates the findings derived from the analysis of a substantial volume of RNA-seq and scRNA-seq data obtained from stable-state chondrocytes of osteoarthritis (OA) chondrocytes, thereby contributing to a deeper understanding of the molecular mechanisms underlying the pathogenesis of OA.