Prediction of biomarkers associated with membranous nephropathy: Bioinformatic analysis and experimental validation
Miaoru Han,Yi Wang,Xiaoyan Huang,Ping Li,Wenjun Shan,Haowen Gu,Houchun Wang,Qinghua Zhang,Kun Bao
DOI: https://doi.org/10.1016/j.intimp.2023.111266
IF: 5.714
2023-11-30
International Immunopharmacology
Abstract:Membranous nephropathy (MN), the most prevalent form of nephrotic syndrome in non-diabetic adults globally, is currently the second most prevalent and fastest-increasing primary glomerular disease in China. Numerous renal disorders are developed partly due to ferroptosis. However, its relationship to the pathogenesis of MN has rarely been investigated in previous studies; actually, ferroptosis is closely linked to the immune microenvironment and inflammatory response, which might affect the entire process of MN development. In this study, we aimed to identify ferroptosis-related genes that are potentially related to immune cell infiltration, which can further contribute to MN pathogenesis. The microarray datasets were downloaded from the Gene Expression Omnibus (GEO) database. Ferroptosis-related differentially expressed genes (FDEGs) were identified, which were further used for functional enrichment analysis. The common genes identified using the Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression algorithm and the support vector machine recursive feature elimination (SVM-RFE) algorithm were used to identify the characteristic genes related to ferroptosis. The feasibility of the 7 genes as a distinguishing factor was assessed using the receiver operating characteristic (ROC) curve, with the area under the curve (AUC) score serving as the evaluation metric. Gene set enrichment analysis (GSEA) and correlation analysis of these genes were further performed. The correlation between the expression of these genes and immune cell infiltration inferred by single sample gene set enrichment analysis (ssGSEA) algorithm was explored. As a result, 7 genes, including NR1D1 , YTHDC2 , EGR1 , ZFP36 , RRM2 , RELA and PDK4 , which were most relevant to immune cell infiltration, were identified to be potential diagnostic genes in MN patients. Next, the signature genes were validated with other GEO datasets. In the subsequent steps, we conducted quantitative real-time fluorescence PCR (qRT-PCR) analysis and immunohistochemistry (IHC) method on the cationic bovine serum albumin (C-BSA) induced membranous nephropathy (MN) rat model and the passive Heymann nephritis (pHN) rat model to examine characteristic genes. Finally, we analysed the mRNA expression patterns of hub genes in MN patients and normal controls using the Nephroseq V5 online platform. In concise terms, our study successfully identified biomarkers specific to MN patients and delved into the potential interplay between these markers and immune cell infiltration. This knowledge bears significance for the diagnosis and prospective treatment strategies for individuals affected by MN.
pharmacology & pharmacy,immunology