Identification of mitochondria-related genes as diagnostic biomarkers for diabetic nephropathy and their correlation with immune infiltration: New insights from bioinformatics analysis

Background: Diabetic nephropathy (DN) is a common and severe microvascular complication of diabetes. Mitochondrial dysfunction and immune inflammation are important factors in the pathogenesis of DN. However, the specific mechanisms and their intricate interactions in DN remain unclear. Besides, there are no effective specific predictive or diagnostic biomarkers for DN so far. Therefore, this study aims to elucidate the role of mitochondrial-related genes and their possibility as predictive or diagnostic biomarkers, as well as their crosstalk with immune infiltration in the progression of DN. Methods: Based on the GEO database and limma R package, the differentially expressed genes (DEGs) of DN were identified. Mitochondrial-related DEGs (MitoDEGs) were then obtained by intersecting these DEGs with mitochondria-related genes from the MitoCarta 3.0 database. Subsequently, the candidate hub genes were further screened by gene co-expression network analysis (WGCNA), and verified mRNA levels of these genes by real-time quantitative PCR (qRT-PCR) in high-glucose-treated human proximal tubular (HK-2) cells. The verified hub genes were utilized to construct a combined diagnostic model for DN, with its diagnostic efficacy assessed across the GSE30122 and GSE96804 datasets. Additionally, the immune infiltration pattern in DN was assessed with the CIBERSORT algorithm, and the Nephroseq v5 database was used to analyze the correlation between hub genes and clinical features of DN. Results: Seven mitochondria-related candidate hub genes were screened from 56 MitoDEGs. Subsequently, the expression levels of six of them, namely EFHD1, CASP3, AASS, MPC1, NT5DC2, and BCL2A1, exhibited significant inter-group differences in the HK-2 cell model. The diagnostic model based on the six genes demonstrated good diagnostic efficacy in both training and validation sets. Furthermore, correlation analysis indicated that EFHD1 and AASS, downregulated in DN, are positively correlated with eGFR and negatively with serum creatinine. Conversely, CASP3, NT5DC2, and BCL2A1, upregulated in DN, show opposite correlations. In addition, spearman analysis revealed that the six hub genes were significantly associated with the infiltration of immune cells, including M1 and M2 macrophages, mast cells, resting NK cells, gamma delta T cells, and follicular helper T cells. Conclusion: This study elucidated the characteristics of mitochondria-related genes and their correlation with immune cell infiltration in DN, providing new insights for exploring the pathogenesis of DN and facilitating the identification of new potential biomarkers and therapeutic targets.