Glycosylation-Related Genes and Prognostic Signatures in Diabetic Nephropathy.

BACKGROUND: Diabetic nephropathy (DN) is the leading cause of end-stage renal disease worldwide, whose pathogenesis involves immune dysregulation and inflammatory response. Glycosylation plays key roles in numerous biological processes. This study aims to interrogate the role of glycosylation-related genes in tubulointerstitial immunoinflammatory injury in DN. METHODS: We utilized two tubulointerstitial transcriptome datasets from DN patients and normal individuals. Glycosylation-related hub genes were identified by integrating differential expression analysis, glycosylation-related gene sets, and machine learning. Immune cell infiltration was assessed using single-sample GSEA (ssGSEA), and functional enrichment analysis was performed via GO and KEGG. The expression levels of hub genes were validated in STZ-induced diabetic mouse model (n=5/group) followed by the evaluation of diagnostic efficiency and clinical significance. RESULTS: Six glycosylation-related hub genes (HEXB, B4GALT5, GALNT7, GCNT3, CGA, and VCAN) were identified, all closely associated with immune cell infiltration in DN. Enrichment analysis indicated their involvement in immune and inflammatory processes. CGA was significantly downregulated, while the other genes were upregulated in DN, which was experimentally validated in diabetic mice. ROC curve analysis revealed high diagnostic accuracy for all genes: HEXB (AUC = 0.892), B4GALT5 (AUC = 0.909), GALNT7 (AUC = 0.931), GCNT3 (AUC = 0.929), CGA (AUC = 0.898), and VCAN (AUC = 0.967). Elevated VCAN, GCNT3, and GALNT7 exhibited a positive association with renal function decline or proteinuria, providing valuable prognostic insights. CONCLUSION: This study highlights the significant role of glycosylation-related genes in DN pathogenesis, likely mediated through immune and inflammatory mechanisms. VCAN, GCNT3, and GALNT7 show particular promise as novel biomarkers for clinical diagnosis and immunotherapeutic targets, supporting their future clinical translation for DN management.