Post-GWAS methodologies for localisation of functional non-coding variants: ANGPTL3

Federico Oldoni,Jutta Palmen,Claudia Giambartolomei,Philip Howard,Fotios Drenos,Vincent Plagnol,Steve E Humphries,Philippa J Talmud,Andrew J.P Smith

doi:10.1016/j.atherosclerosis.2015.12.009

Federico Oldoni, Jutta Palmen + Show 7 more

Open Access

https://doi.org/10.1016/j.atherosclerosis.2015.12.009

Copy DOI

Abstract

Genome-wide association studies have confirmed the involvement of non-coding angiopoietin-like 3 (ANGPTL3) gene variants with coronary artery disease, levels of low-density lipoprotein cholesterol (LDL-C), triglycerides and ANGPTL3 mRNA transcript. Extensive linkage disequilibrium at the locus, however, has hindered efforts to identify the potential functional variants. Using regulatory annotations from ENCODE, combined with functional in vivo assays such as allele-specific formaldehyde-assisted isolation of regulatory elements, statistical approaches including eQTL/lipid colocalisation, and traditional in vitro methodologies including electrophoretic mobility shift assay and luciferase reporter assays, variants affecting the ANGPTL3 regulome were examined. From 253 variants associated with ANGPTL3 mRNA expression, and/or lipid traits, 46 were located within liver regulatory elements and potentially functional. One variant, rs10889352, demonstrated allele-specific effects on DNA-protein interactions, reporter gene expression and chromatin accessibility, in line with effects on LDL-C levels and expression of ANGPTL3 mRNA. The ANGPTL3 gene lies within DOCK7, although the variant is within non-coding regions outside of ANGPTL3, within DOCK7, suggesting complex long-range regulatory effects on gene expression. This study illustrates the power of combining multiple genome-wide datasets with laboratory data to localise functional non-coding variation and provides a model for analysis of regulatory variants from GWAS.

Highlights

Genome-wide association studies (GWAS) have revealed novel gene loci and confirmed many candidate genes for numerous traits including lipid levels and coronary artery disease (CAD) risk [1,2]
Identification of functional variants from GWAS can be a laborious process with large blocks of linkage disequilibrium (LD) surrounding the lead single nucleotide polymorphism (SNP) often resulting in unfeasible numbers of variants to examine in the laboratory
Three complementary laboratory methods were employed to examine the effect of this reduced set of variants on regulatory potential: electrophoretic mobility shift assay (EMSA), a relatively inexpensive and quick method to identify the effect of variants on DNA-protein interactions; luciferase reporter assay to examine the role of these variants in gene expression; and FAIRE

Summary

Introduction

Genome-wide association studies (GWAS) have revealed novel gene loci and confirmed many candidate genes for numerous traits including lipid levels and coronary artery disease (CAD) risk [1,2]. Despite the increasing number of loci associated with such traits, identification of functional or causal variants at each locus has not been met with similar advances. This lack of progress is largely due Abbreviations: GWAS, genome-wide association study; LDL-C, low density lipoprotein cholesterol; HDL-C, high density lipoprotein cholesterol; TG, triglycerides; EMSA, electrophoretic mobility shift assay; SNP, single nucleotide polymorphism. A typical example of this occurs at the angiopoietin-like 3 (ANGPTL-3) gene locus, associated with triglyceride (TG) levels, low-density lipoprotein cholesterol (LDL-C) levels, and CAD risk in several independent GWAS [1,3,4], where the functional variant(s) associated with these traits have not been identified. ANGPTL-3 can inhibit lipoprotein lipase [5] and endothelial lipase [6], and through the fibrinogen-like domain, induce endothelial cell adhesion and migration [7]

Methods

Results

Conclusion