ObjectiveThe mechanisms of chondrocytes ferroptosis in osteoarthritis (OA) have not yet been fully elucidated. This study aimed to identify key ferroptosis related genes (FRGs) involved in chondrocytes ferroptosis. MethodsLASSO, SVM-RFE, and receiver operating characteristic curve (ROC) were performed to screen key differentially expressed FRGs (DEFRGs). Functional analyses were conducted using GO, and KEGG analyses. Unsupervised clustering analysis was used to identify ferroptosis related patterns. The CeRNA network was constructed to predict the upstream miRNAs and lncRNAs. Finally, we validated the role of EGFR in chondrocytes ferroptosis using in vivo and in vitro experiments. ResultsA total of 42 DEFRGs were identified between OA and normal cartilages. GO and KEGG analyses indicated that these DEFRGs were significantly engaged in ferroptosis related biological processes and pathways, such as cellular response to oxidative stress, positive regulation of programmed cell death, MAPK and PI3K-Akt signaling pathways. Moreover, four key DEFRGs, including ACSF2, AURKA, EGFR, and KLHL24, were considered as potential biomarkers of OA. Moreover, two distinct ferroptosis related patterns were determined, and a total of 882 differentially expressed genes were identified which might participate in extracellular matrix degradation and inflammatory response. In addition, the CeRNA network showed that EGFR could be competitively regulated by 3 lncRNAs and 4 miRNAs. Significantly, the expression of EGFR was downregulated in human OA cartilages, OA mouse model, and erastin induced chondrocytes. EGFR inhibition could induce the occurrence of chondrocytes ferroptosis and ECM degradation which could be reversed by the addition of Ferrostatin-1. ConclusionOur study has identified ACSF2, AURKA, EGFR, and KLHL24 as ferroptosis-related biomarkers in OA. Furthermore, we have conducted a preliminary investigation into the role of EGFR in regulating chondrocytes ferroptosis. These findings offer novel insights into the molecular mechanisms underlying OA.
Read full abstract