Transcriptome and single-cell RNA sequencing data identify and validate candidate biomarkers associated with lysosomal and ferroptosis in osteoarthritis

BACKGROUND: Osteoarthritis (OA) represents the most common degenerative joint disease, with emerging evidence linking it to lysosomal dysfunction and ferroptotic cell death. This study aimed to identify candidate biomarkers associated with lysosomal function and ferroptosis in OA, thereby providing a theoretical basis for subsequent experimental research. METHODS: We obtained OA datasets and ferroptosis-related genes (FRGs) from publicly accessible repositories, while lysosome-related genes (LRGs) were compiled from published literature. Potential candidate genes emerged through the intersection of differentially expressed genes (DEGs), LRGs, and FRGs. Machine learning approaches, along with Receiver Operating Characteristic (ROC) analysis and gene expression profiling, enabled candidate biomarker identification. Nomogram analysis was utilized to assess the diagnostic performance of identified candidate biomarkers. We conducted functional pathway enrichment and immune cell infiltration analyses. Cell type determination was achieved through single-cell sequencing analysis, with key cellular populations characterized using the candidate biomarkers. In vitro validation of candidate biomarker expression was performed using Reverse Transcription quantitative Polymerase Chain Reaction (RT-qPCR). RESULTS: The intersection of 2621 DEGs, 875 LRGs, and 565 FRGs yielded 12 candidate genes. Through subsequent analysis, SLC7A5 (8140) and SLC3A2 (6520) emerged as candidate biomarkers, both showing reduced expression in OA samples with excellent discriminatory ability between sample groups. Pathway enrichment revealed involvement in lysosomal and spliceosome signaling. Strong positive correlations were observed between both SLC7A5 and SLC3A2 with eosinophil infiltration (r > 0.40, P < 0.05). Fibroblasts were identified as a major cell population in OA synovium, with both biomarkers showing markedly decreased expression in OA model cells compared to normal chondrocytes. CONCLUSION: This study identified SLC7A5 and SLC3A2 as OA candidate biomarkers associated with both lysosome and ferroptosis. Additionally, fibroblasts were identified as important contributors to OA pathology at the cellular level, providing a foundation for further investigation into their cellular mechanisms. THE TRANSLATIONAL POTENTIAL OF THIS ARTICLE: This study identified candidate biomarkers-SLC7A5 and SLC3A2-associated with lysosomal dysfunction and ferroptosis, and highlighted fibroblasts as a potentially important cell type in OA. Collectively, these findings offer a theoretical foundation and generate testable hypotheses regarding potential molecular and cellular targets. This work may guide future research aimed at developing early diagnostic approaches and therapeutic strategies, although further experimental and clinical validation is required.