Abstract
BACKGROUND: Diabetic kidney disease (DKD) is a common diabetes complication that increases global morbidity and mortality. To identify DKD biomarkers and explore autophagy-related mechanisms to find potential therapeutic targets for DKD treatment. METHODS: After standardizing and analyzing the GSE142025 and Merged_GSE47183_GSE32591 datasets, DKD-related transcriptional changes (DEGs) were identified using computational methods, volcano plots, and heatmaps. Functional enrichment analysis of GO and KEGG pathways explored the role of autophagy-related genes. Key genes were identified through a PPI network, and potential DKD biomarkers were selected using machine learning. ROC curve analysis evaluated the biomarkers' diagnostic effectiveness. qPCR and immunohistochemical staining confirmed biomarker expression in DKD mouse kidneys. snRNA-seq identified cell-specific transcriptional profiles and signaling networks. Pseudotemporal analysis highlighted DKD-related PCT cell dynamics. RESULTS: This study showed autophagy-related gene enrichment in pathways like inflammatory response and TNF/NF-κB signaling. Four biomarkers (COL1A2, CSF1R, PTPRC, and TYROBP) showed diagnostic potential for DKD. Immune cell infiltration analysis revealed differences between DKD and control groups. qPCR and staining indicated significant upregulation of these biomarkers. snRNA-seq identified cell clusters and DEGs, with altered signaling pathways and PCT cell communication. Network centrality analysis highlighted the differing roles of PCT in cell communication between control and DKD groups. Pseudotime trajectory analysis and BEAM revealed potential regulatory mechanisms in PCT cell differentiation and development in DKD. CONCLUSIONS: This study offers new insights into DKD biomarkers and autophagy-related mechanisms, suggesting potential therapeutic targets and laying the groundwork for future research and treatment strategies.