BBB Seminar: Boris Lenhard
Disease susceptibility variants mapping within long-range regulatory domains of transcription factor genes
Boris Lenhard, Computational Biology Unit, Bergen Center for Computational Science and Sars Centre for Marine Molecular Biology, University of Bergen
Many genes with multiple developmental regulatory roles are spanned by megabase arrays of highly conserved non-coding elements (HCNEs) that form extended chromosomal domains termed genomic regulatory blocks (GRBs). Besides those (target) genes, GRBs often contain other, functionally unrelated genes kept in cis with a target gene because enhancers regulating target genes are located in their introns or beyond. We identified GRBs in regions of recently published single nucleotide polymorphisms (SNPs) associated with risk for type 2 diabetes and obesity, using minimal conserved synteny and density plots of highly conserved non-coding elements. We find that the recently identified diabetes risk genes ALX4 and TCF7L2 and HHEX, which are transcriptional regulators, are targets of GRB regulation. Strikingly, the recently identified SNPs near FLJ393370, CDKAL1 and PKN2 fall into the regulatory blocks of PITX2, SOX4, and LMO4, respectively, all developmental transcription factors expressed in the pancreas. FTO, which contains a haplotype conferring risk for obesity, lies within the GRB of IRX3, a transcription factor expressed in the hypothalamus and endoderm. The identified regulatory genes are functionally more plausible as disease risk factors than the genes physically closest to the SNPs. The testing of conserved non-coding elements near or coinciding with the reported T2D-associated SNPs in reporter constructs in zebrafish shows that they drive expression in the manner compatible with the GRB target genes and in patterns indicative of their involvement in T2D pathogenesis. This shows that human enhancers driving disease genes are often located inside the introns of non-related neighboring genes, and they can be tested in zebrafish to evaluate genomic regions for their involvement in T2D. The results call for a re-interpretation of common variants conferring disease risk as mutations in non-coding elements regulating distant developmental control genes. I will discuss the general approach and the present tools used for the computational analysis of many other similar cases in the genome.
Host: Marit Bakke, Department of Biomedicine