Comprehensive analysis identified and validated BRD4, CHD1, and KDM7A as potential biomarkers for osteoarthritis

Abstract

Abstract Background Osteoarthritis (OA) is a common degenerative joint disease and costly public health problem. Current treatments for OA provide only limited symptomatic benefits. The onset of OA can be a long-term silent process and the articular cartilage will likely have been damaged before the onset of typical symptoms. Therefore, better diagnostic and treatment methods are needed. Objective Chromatin regulators (CRs) are a class of regulators of epigenetics and play an important role in OA. However, the mechanisms of CRs in OA are unclear. Identifying and validating signature CRs in OA will assist in the diagnosis and treatment of OA. And it is essential to further explore the potential mechanisms by which CRs intervene in OA. Methods Firstly, the publicly available Gene Expression Omnibus database was used to download the OA-related chip data sets GSE55235 and GSE55457. Afterward, we extracted the expression matrix of chromatin regulator-related genes (CRRGs) in the OA-related data sets and screened for differentially expressed CRRGs (DECRRGs). GO and KEGG enrichment analyses were performed on the DECRRGs. Subsequently, we constructed the PPI network and screened for the key cluster network. We used three machine-learning methods to screen for potential biomarkers. Two additional OA datasets (GSE12021 and GSE1919) were used to validate the diagnostic value of these biomarkers. Then we performed an immune cell infiltration analysis. Finally, we explore the potential mechanisms through which these biomarkers intervene in OA in the context of the relevant literature published in PubMed over the last five years. Results We obtained the expression matrix for 717 CRRGs. Through differentially expressed analysis, we obtained 85 DECRRGs. By three machine-learning methods, we screened four potential biomarkers. The diagnostic value of these potential biomarkers was evaluated by two additional OA datasets (GSE12021 and GSE1919). Finally, we obtained three biomarkers (BRD4, CHD1, and KDM7A). Through immune cell infiltration analysis, we found that all these biomarkers were significantly associated with mast cells. Through reviewing the relevant literature published in PubMed in the last five years, we have initially revealed the potential mechanisms by which these biomarkers intervene in OA. Conclusion BRD4, CHD1, and KMD7A are potential biomarkers of OA. BRD4 may play an important role in the inflammatory response to OA. CHD1 and KDM7A may contribute to altered epigenetic markers in OA by affecting histone modifications. Our study laid the foundation for further studies to follow.