Integrative analyses of biomarkers and pathways for diabetic nephropathy

Background: Diabetic nephropathy (DN) is a widespread diabetic complication and a major cause of terminal kidney disease. There is no doubt that DN is a chronic disease that imposes substantial health and economic burdens on the world’s populations. By now, several important and exciting advances have been made in research on etiopathogenesis. Therefore, the genetic mechanisms underlying these effects remain unknown. Methods: The GSE30122, GSE30528, and GSE30529 microarray datasets were downloaded from the Gene Expression Omnibus database (GEO). Analyses of differentially expressed genes (DEGs), enrichment of gene ontology (GO), the Kyoto Encyclopedia of Genes and Genomes (KEGG), and gene set enrichment analysis (GSEA) were performed. Protein-protein interaction (PPI) network construction was completed by the STRING database. Hub genes were identified by Cytoscape software, and common hub genes were identified by taking intersection sets. The diagnostic value of common hub genes was then predicted in the GSE30529 and GSE30528 datasets. Further analysis was carried out on the modules to identify transcription factors and miRNA networks. As well, a comparative toxicogenomics database was used to assess interactions between potential key genes and diseases associated upstream of DN. Results: Samples from 19 DNs and 50 normal controls were identified in the GSE30122 dataset. 86 upregulated genes and 34 downregulated genes (a total of 120 DEGs). GO analysis showed significant enrichment in humoral immune response, protein activation cascade, complement activation, extracellular matrix, glycosaminoglycan binding, and antigen binding. KEGG analysis showed significant enrichment in complement and coagulation cascades, phagosomes, the Rap1 signaling pathway, the PI3K-Akt signaling pathway, and infection. GSEA was mainly enriched in the TYROBP causal network, the inflammatory response pathway, chemokine receptor binding, the interferon signaling pathway, ECM receptor interaction, and the integrin 1 pathway. Meanwhile, mRNA-miRNA and mRNA-TF networks were constructed for common hub genes. Nine pivotal genes were identified by taking the intersection. After validating the expression differences and diagnostic values of the GSE30528 and GSE30529 datasets, eight pivotal genes (TYROBP, ITGB2, CD53, IL10RA, LAPTM5, CD48, C1QA, and IRF8) were finally identified as having diagnostic values. Conclusion: Pathway enrichment analysis scores provide insight into the genetic phenotype and may propose molecular mechanisms of DN. The target genes TYROBP, ITGB2, CD53, IL10RA, LAPTM5, CD48, C1QA, and IRF8 are promising new targets for DN. SPI1, HIF1A, STAT1, KLF5, RUNX1, MBD1, SP1, and WT1 may be involved in the regulatory mechanisms of DN development. Our study may provide a potential biomarker or therapeutic locus for the study of DN.