Identification of Hub Gene and Transcription Factor Related to Chronic Allograft Nephropathy Based on WGCNA Analysis

Kidney transplantation (KT) has surpassed dialysis as the optimal therapy for end-stage kidney disease. Yet, most patients could suffer from a slow but continuous deterioration of kidney function leading to graft loss mostly due to chronic allograft nephropathy (CAN) after KT. The dysregulated gene expression for CAN is still poorly understood.

To explore the pathogenesis of genomics in CAN, we analyzed the differentially expressed genes (DEGs) of kidney transcriptome between CAN and nonrejecting patients by downloading gene expression microarrays from the Gene Expression Omnibus database. Then, we used weighted gene coexpression network analysis (WGCNA) to analyze the coexpression of DEGs to explore key modules, hub genes, and transcription factors in CAN. Functional enrichment analysis of key modules was performed to explore pathogenesis. ROC curve analysis was used to validate hub genes.

As a result, 3 key modules and 15 hub genes were identified by WGCNA analysis. Three key modules had 21 mutual Gene Ontology term enrichment functions. Extracellular structure organization, extracellular matrix organization, and extracellular region were identified as significant functions in CAN. Furthermore, transcription factor 12 was identified as the key transcription factor regulating key modules. All 15 hub genes, Yip1 interacting factor homolog B, membrane trafficking protein, toll like receptor 8, neutrophil cytosolic factor 4, glutathione peroxidase 8, mesenteric estrogen dependent adipogenesis, decorin, serpin family F member 1, integrin subunit beta like 1, SRY-box transcription factor 15, trophinin associated protein, SRY-box transcription factor 1, metallothionein 3, lysosomal protein transmembrane, FERM domain containing kindlin 3, and cathepsin S, had a great diagnostic performance (AUC > 0.7).

This study updates information and provides a new perspective for understanding the pathogenesis of CAN by bioinformatics means. More research is needed to validate and explore the results we have found to reveal the mechanisms underlying CAN.