基于生物信息学鉴定铁死亡相关基因作为糖尿病肾病潜在的诊断生物标志物

Identification of ferroptosis-related genes as potential diagnostic biomarkers for diabetic nephropathy based on bioinformatics.

作者信息

Guo Binbin, Li Minhui, Wu Peipei, Chen Yan

机构信息

International Special Medical Department, Shengli Oilfield Central Hospital, Dongying, Shandong, China.

Department of Pediatrics Internal Medicine, Dongying Municipal Children's Hospital, Dongying, Shandong, China.

出版信息

Front Mol Biosci. 2023 Aug 1;10:1183530. doi: 10.3389/fmolb.2023.1183530. eCollection 2023.

This study investigated to probe ferroptosis-related diagnostic biomarkers and underlying molecular mechanisms in Diabetic nephropathy (DN). GSE30122 and GSE1009 from GEO database were used as training and verification sets, respectively, to screen differentially expressed ferroptosis-related genes (FRGs). These genes were further analyzed using GO, KEGG, and GSEA methods, and screened with PPI, LASSO, and SVM-RFE to identify ferroptosis-related diagnostic biomarkers for DN. A diagnostic model was established using the Glm function and verified with ROC curve. The relationship between these biomarkers and immune cell was analyzed, and qRT-PCR and Western blot were used to detect the expression of these biomarkers in kidney tissues and identify the effect of TP53 on DN development. Fifty one differentially expressed FRGs were enriched in bioprocesses such as signaling pathway, oxidative stress and chemical stress response, and mTOR signaling pathway. , , , , , and were identified as ferroptosis-related diagnostic biomarkers for DN. showed the most differential expression. ROC analysis showed that AUC values of , , , , , and were 0.751, 0.705, 0.725, 0.882, 0.691, and 0.675, respectively. The AUC value of DN diagnosis model was 0.939 in training set and 1.000 in verification set. qRT-PCR results confirmed significant differences in these six biomarkers between DN and normal kidney tissue ( < 0.05), and correlation analysis showed that five biomarkers were significantly correlated with infiltrating immune cells ( < 0.05). Furthermore, western blots showed that promotes apoptosis through PI3K-AKT signaling in DN. , , , , , and have potential as diagnostic biomarkers for DN. The diagnostic model containing the above six biomarkers performs well in the diagnosis of DN. Five of the six biomarkers are strongly associated with several infiltrating immune cells. may play an essential role in the development of DN.

本研究旨在探究糖尿病肾病（DN）中铁死亡相关的诊断生物标志物及其潜在分子机制。分别将来自基因表达综合数据库（GEO数据库）的GSE30122和GSE1009用作训练集和验证集，以筛选差异表达的铁死亡相关基因（FRGs）。使用基因本体论（GO）、京都基因与基因组百科全书（KEGG）和基因集富集分析（GSEA）方法对这些基因进行进一步分析，并通过蛋白质-蛋白质相互作用（PPI）、套索回归（LASSO）和支持向量机递归特征消除（SVM-RFE）进行筛选，以鉴定DN的铁死亡相关诊断生物标志物。使用广义线性模型（Glm）函数建立诊断模型，并通过ROC曲线进行验证。分析了这些生物标志物与免疫细胞之间的关系，并使用实时定量聚合酶链反应（qRT-PCR）和蛋白质免疫印迹法（Western blot）检测这些生物标志物在肾组织中的表达，并确定TP53对DN发展的影响。51个差异表达的FRGs富集于信号通路、氧化应激和化学应激反应以及哺乳动物雷帕霉素靶蛋白（mTOR）信号通路等生物过程中。、、、、和被鉴定为DN的铁死亡相关诊断生物标志物。显示出最大差异表达。ROC分析表明，、、、、和的曲线下面积（AUC）值分别为0.751、0.705、0.725、0.882、0.691和0.675。DN诊断模型在训练集中的AUC值为0.939，在验证集中为1.000。qRT-PCR结果证实，这六种生物标志物在DN和正常肾组织之间存在显著差异（ < 0.05），相关性分析表明，五种生物标志物与浸润性免疫细胞显著相关（ < 0.05）。此外，蛋白质免疫印迹法表明，通过PI3K-AKT信号通路促进DN中的细胞凋亡。、、、、和有潜力作为DN的诊断生物标志物。包含上述六种生物标志物的诊断模型在DN诊断中表现良好。六种生物标志物中的五种与几种浸润性免疫细胞密切相关。可能在DN的发展中起重要作用。