Kidney Specific Regulatory Network in Mouse Uncovers Functional, Evolutionary and Disease Dynamics

Abstract

Transcription factors (TFs) operate in a combinatorial fashion to regulate the expression of a gene or a group of genes; however, their tissue-specific regulatory interactions are not fully characterized. In this study, we construct and investigate kindey-specific regulatory (KSR) network for mouse. We obtained upstream regions of genes in the mouse genome from ENSEMBL and extracted DNase 1 Hypersensitive sites (DHS) for 8-week mouse kidney from ENCODE project. Similarly, the position weight matrices (PWMs) for TF binding motifs (BMo) were extracted from JASPAR. Jolma, TRANSFAC and mapped in the mouse genome using FIMO. These BMo were integrated with obtained DHS signals (narrow peak) in 5 KBs upstream regions. The resulting TFs and their targeted genes were modeled as directed interaction network comprising of 619 TFs and their corresponding 13500 target genes. We trimmed the resulting network by only keeping the genes that function as TFs. Resulting TF-TF network (of 619 nodes) was analyzed to provide a holistic picture of TF-TF interactions in mouse kidney tissue while the global network was studied for conservation across 61 species and relevance in kidney associated diseases. We observed that genes related to diseases were significantly enriched in second and third layers in network hierarchy. Conservation analysis of Mouse KSR revealed >50% conservation in close relatives such as rat, human, dog, squirrel and less conserved in invertebrates and yeast, thus elucidating network complexity increases with increase in kidney functionality from lower to higher species. In addition, mouse KSR was examined in its closest relative, rat for segments of nephron - TAL (Thick ascending limb), PT (Proximal tubules), IMCD (Inner medullary collecting duct), which revealed a significant enrichment of TFs for their corresponding original group in mouse KSR. Further, this network was investigated in diverse model kidney diseases such as hypertension, diabetes and kidney renal clear cell carcinoma (KIRC). The compendium of the network reported in this study can form a roadmap for increasing our understanding of the variations in regulatory wiring in kidney diseases.