Discovery of novel genes and genetic regulatory mechanisms through integrative genomic, epigenetic and transcriptional analysis in human pancreatic islet cells

• Main Author: Moran Castany, Ignasi
• Other Authors: Ferrer, Jorge
• Published: Imperial College London 2015
• Subjects: 610
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.733073

Diabetes mellitus is a metabolic disorder characterized by a rise in blood sugar levels, estimated to currently affect more than 300 million people worldwide. Even though pancreatic islet β-cells play a central role in all major forms of the disease, their transcriptome is still not fully characterised. Furthermore, type 2 diabetes susceptibility is known to be affected by non-coding genomic variation, but the impact of these variants on gene regulation and their disease implications remain to be elucidated. In the first part of this project, I integrated the analysis of RNA sequencing datasets with maps of epigenomic features, resulting in the identification of 1,128 long non-coding RNA (lncRNA) genes in human pancreatic islets. Islet lncRNAs were enriched for tissue-specific expression, and a subset were shown to be regulated during endocrine differentiation or in response to changes in glucose. Furthermore, lncRNA orthologs in mouse islets were identified, which displayed similar regulatory properties to their human counterparts. Taken together, these analyses led to the identification of a novel class of genes active in human and mouse pancreatic islets, opening new avenues for the study of diabetes development and therapeutic strategies. In the second part of this project, I analysed the RNA-seq datasets and genotypes of 27 human islet samples, and developed a methodology to study the effect of non-coding genomic variation on gene regulation. This revealed hundreds of genes under allele-specific expression, consistently across many independent individuals. Some of these genes performed known roles in human islets, or resided within GWAS loci for type 2 diabetes and related traits. These findings could be used to identify functional cis-regulatory genetic variants impacting the human islet transcriptome, some of which could lead to the development of diabetes. Collectively, these projects uncovered novel components of the human islet transcriptome, and provided new methodologies to study the effects of genomic regulatory variation on gene transcription.