Investigate biomarkers and regulatory mechanisms related to RNA modification in diabetic nephropathy through bioinformatics and clinical experiments

In recent years, RNA modifications have demonstrated significant potential in diagnosing and treating diabetic nephropathy (DN). This research utilized bioinformatics techniques to investigate the biomarkers associated between DN and RNA modifications, offering fresh perspectives on the diagnosis and treatment of DN. Candidate genes were acquired via differential expression analysis and weighted gene co-expression network analysis (WGCNA). Biomarkers were selected through machine learning algorithms and expression level validation. Subsequently, a nomogram was constructed for the biomarkers, followed by gene set enrichment analysis (GSEA), gene multiple association network integration algorithm (GeneMANIA) analysis, immune infiltration analysis, molecular regulatory network construction, and drug prediction analysis. Lastly, reverse transcription quantitative polymerase chain reaction (RT-qPCR) was employed to validate the expression of biomarkers in clinical samples in relation to bioinformatics analysis results (P < 0.05). This study identified 39 candidate genes, among which PFKFB3 and SLC7A5 were confirmed as biomarkers. The nomogram achieved a high AUC of 0.900, indicating robust predictive power. GSEA revealed that the two biomarkers were co-enriched in multiple pathways including peroxisome and primary immunodeficiency. GeneMANIA analysis showed that PFKFB3 and SLC7A5 interacted with 20 other genes in various ways. They also played a role in the regulation of immune cell activities (for instance, there was a positive correlation (cor) between PFKFB3 and activated dendritic cells (cor = 0.61, P = 1.520×10⁻⁴)). Regulatory network analysis indicated that the 2 biomarkers were regulated by 28 microRNAs (miRNAs) and 86 transcription factors (TFs), and three potential drugs such as pfk-158 were predicted. The results of RT-qPCR were consistent with the results of bioinformatics analysis. This study identified PFKFB3 and SLC7A5 as biomarkers related to DN and RNA modifications through bioinformatics analysis, and unveils potential therapeutic targets in DN.