Analysis and identification of ferroptosis-related diagnostic markers in rheumatoid arthritis

Rheumatoid arthritis (RA) is an autoimmune, inflammatory joint disease. There is growing evidence that ferroptosis is involved in the pathogenesis of RA. This study aimed to search for diagnostic markers of ferroptosis in RA and to analyse the potential mechanisms and clinical value. RA-associated datasets were used from the publicly available GEO database. Three methods of machine learning were applied to screen biomarkers. The diagnostic efficacy of the results was also verified by receiver operating characteristic (ROC) curve, external dataset, qRT-PCR and Western blot. Enrichment analysis was performed in this process, while protein–protein interaction (PPI) analysis and immune infiltration correlation analysis were performed using biomarkers, and competing endogenous RNA (ceRNA) networks were constructed to search for prospective therapeutic targets. MMP13 and GABARAPL1 can be used as ferroptosis diagnostic genes in RA. The ROC curve and PPI result demonstrated that MMP13 and GABARAPL1 had an excellent diagnostic value. The results of signature genes in the external dataset, qRT-PCR and Western blot further confirm our findings. The enrichment analysis showed that p53, MAPK and NOD-like receptor signalling pathways may be involved in the process of ferroptosis in RA. In addition, two ferroptosis diagnostic genes in RA participate in the occurrence of ferroptosis in RA via oxidative stress, metabolism and immune response. Immune infiltration analysis showed that RA extensively infiltrated B cells, T cells, macrophages and other immune cells. Persistent immune activation may be an essential reason for the progression of ferroptosis in RA. We also obtained five potential therapeutic agents for RA and some long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) regulating ferroptosis diagnostic genes. Our study suggests that MMP13 and GABARAPL1, which are closely linked with oxidative stress and immunological modulation, can be used as ferroptosis-related potential diagnostic markers in RA and provide new clues regarding the diagnostic and therapeutic targets of ferroptosis in RA.

类风湿关节炎（Rheumatoid arthritis, RA）是一种自身免疫性炎症性关节疾病。越来越多的证据表明铁死亡（ferroptosis）参与了RA的发病机制。本研究旨在筛选RA中铁死亡的诊断标志物，并分析其潜在机制与临床价值。研究使用了公开可得的GEO（Gene Expression Omnibus）数据库中与RA相关的数据集。采用三种机器学习方法筛选生物标志物。通过受试者工作特征（Receiver Operating Characteristic, ROC）曲线、外部数据集、定量实时聚合酶链反应（qRT-PCR）及蛋白质印迹（Western blot）验证了所得结果的诊断效能。本研究过程中开展了富集分析，并利用生物标志物进行蛋白-蛋白相互作用（protein–protein interaction, PPI）分析与免疫浸润相关性分析，同时构建竞争性内源RNA（competing endogenous RNA, ceRNA）网络以筛选潜在治疗靶点。研究发现基质金属蛋白酶13（MMP13）和GABARAPL1可作为RA中铁死亡的诊断基因。ROC曲线与PPI分析结果显示，MMP13与GABARAPL1具有优异的诊断价值。外部数据集、qRT-PCR及蛋白质印迹的验证结果进一步证实了本研究的结论。富集分析表明，p53、丝裂原活化蛋白激酶（mitogen-activated protein kinase, MAPK）及NOD样受体信号通路可能参与了RA中铁死亡的发生过程。此外，RA中的两种铁死亡诊断基因通过氧化应激、代谢及免疫应答参与RA中铁死亡的进程。免疫浸润分析显示，RA患者体内广泛存在B细胞、T细胞、巨噬细胞及其他免疫细胞的浸润。持续的免疫激活可能是RA中铁死亡进展的关键诱因。本研究还筛选出5种RA潜在治疗药物，以及若干调控铁死亡诊断基因的长链非编码RNA（long non-coding RNAs, lncRNAs）和微小RNA（microRNAs, miRNAs）。本研究表明，与氧化应激及免疫调控密切相关的MMP13与GABARAPL1可作为RA中铁死亡相关的潜在诊断标志物，为RA中铁死亡的诊疗靶点提供了新的研究思路。