A joint effort involving research teams from Vanderbilt University Medical Center (VUMC) and the University of Michigan has linked the activities of a regulatory gene network and functional defects in insulin-producing pancreatic beta cells to type 2 diabetes. Details of the study are published in a Nature paper titled, “Genetic risk converges on regulatory networks mediating early type 2 diabetes.”
As a disease with serious health implications that affects nearly 35 million people in the United States, type 2 diabetes has in some ways been well studied and characterized. Large genome-wide association studies have connected more than 400 genomic signals to a greater risk of developing the disease. But these are mostly sites in noncoding enhancer regions of the genome. Exactly how “this genetic variation at the population level relates to molecular changes in gene expression, tissue architecture and cellular physiology in type 2 diabetes is not well understood,” said Marcela Brissova, PhD, a research professor of medicine at VUMC and co-lead on the new paper.
Furthermore, scientists have used rodent models of islet cells to study the processes that contribute to type 2 diabetes development. But as with most animal models of human diseases, there are key differences between rodent and human islets. As such, “there is a need for studies to define the mechanisms that initiate and sustain islet dysfunction in primary human islets, which is where we focused our efforts,” according to Alvin Powers, MD, a professor of biomedical science, director of the Vanderbilt Diabetes Research and Training Center, and one of the study leads.
For the study, the researchers used a combination of functional, transcriptomic, and spatially resolved multiplex imaging datasets to link gene regulatory networks and unique islet cell signatures to early-stage type 2 diabetes cases. Specifically, they integrated data on islet function in mouse models with gene expression information from whole islets and purified beta and alpha cells, and the islet microenvironment architecture assessed using multiplexed imaging.
“[W]e used integrative data analytic methods to build gene networks and connect them to type 2 diabetes GWAS signals and islet physiology,” Stephen Parker, PhD, an associate professor of computational medicine & bioinformatics, human genetics, and biostatistics at the University of Michigan and a co-lead author, explained. Building on findings from single cells, the researchers looked for population-level differences in gene expression between people with early-onset type 2 diabetes and normal controls.
Their analysis revealed that the RFX6 gene regulatory network, which is influential in pancreatic beta cells, was downgraded in type 2 diabetes cases suggesting a likely role for the network in increasing disease risk. “Our single-cell to population-scale results suggest that RFX6 exerts a disproportionate transcriptional influence on β cell state and that its dysregulation is a key molecular event in early T2D pathogenesis,” the research team wrote.
Additional analysis using data from human pseudoislet cells showed that disrupting that RFx6 activity in beta cells reduced insulin secretion and altered chromatin architecture in genomic regions enriched for type 2 diabetes GWAS signals. The team also analyzed phenotype and genotype data from the UK Biobank and found evidence that decreased expression of RFX6 in the pancreatic islet cell was causally associated with type 2 diabetes cases.
This study serves a two-fold purpose in the researchers’ minds. It provides a starting point from which scientists can try to identify additional disease-driving events in type 2 diabetes cases. To help with studies along these lines, the team has made their data publicly available through various user-friendly and interactive web portals. They also plan to explore the initial cause or causes of RFX6 dysregulation and whether it can be targeted to prevent the beta cell defects from arising in the first place.
The team also believes that the work done here provides a potentially useful template for identifying the regulatory networks that drive other diseases for which there is GWAS data available.
