Date Published: August 1, 2019
Publisher: Public Library of Science
Author(s): Dharambir K. Sanghera, Ruth Hopkins, Megan W. Malone-Perez, Cynthia Bejar, Chengcheng Tan, Huda Mussa, Paul Whitby, Ben Fowler, Chinthapally V. Rao, KarMing A. Fung, Stan Lightfoot, J. Kimble Frazer, Ouliana Ziouzenkova.
Dyslipidemia is a well-established risk factor for cardiovascular diseases. Although, advances in genome-wide technologies have enabled the discovery of hundreds of genes associated with blood lipid phenotypes, most of the heritability remains unexplained. Here we performed targeted resequencing of 13 bona fide candidate genes of dyslipidemia to identify the underlying biological functions. We sequenced 940 Sikh subjects with extreme serum levels of hypertriglyceridemia (HTG) and 2,355 subjects were used for replication studies; all 3,295 participants were part of the Asian Indians Diabetic Heart Study. Gene-centric analysis revealed burden of variants for increasing HTG risk in GCKR (p = 2.1×10-5), LPL (p = 1.6×10-3) and MLXIPL (p = 1.6×10-2) genes. Of these, three missense and damaging variants within GCKR were further examined for functional consequences in vivo using a transgenic zebrafish model. All three mutations were South Asian population-specific and were largely absent in other multiethnic populations of Exome Aggregation Consortium. We built different transgenic models of human GCKR with and without mutations and analyzed the effects of dietary changes in vivo. Despite the short-term of feeding, profound phenotypic changes were apparent in hepatocyte histology and fat deposition associated with increased expression of GCKR in response to a high fat diet (HFD). Liver histology of the GCKRmut showed severe fatty metamorphosis which correlated with ~7 fold increase in the mRNA expression in the GCKRmut fish even in the absence of a high fat diet. These findings suggest that functionally disruptive GCKR variants not only increase the risk of HTG but may enhance ectopic lipid/fat storage defects in absence of obesity and HFD. To our knowledge, this is the first transgenic zebrafish model of a putative human disease gene built to accurately assess the influence of genetic changes and their phenotypic consequences in vivo.
Dyslipidemia is a well-established risk factor for cardiovascular disease and a principal cause of mortality in individuals with type 2 diabetes (T2D). Circulating blood lipid phenotypes are heritable risk factors for the development of atherosclerosis and their measurements are used clinically to predict future coronary artery disease (CAD) risk and therapy for primary prevention [1,2]. Epidemiological studies suggest that elevated serum triglyceride (TG) concentration is a strong independent risk factor for CAD [1,3]. There is an inverse correlation between serum TG and serum high-density cholesterol (HDL-C) that is associated with increased risk of cardiovascular dysfunction, despite the level of low density cholesterol (LDL-C) being normal. This combination of lipid alterations is defined as atherogenic dyslipidemia, which is a significant risk factor for the development of CAD . Lowering of LDL-C has been the major focus in CAD prevention following treatment with HMG-CoA reductase inhibitors (statins). However, the mortality rate of CAD remains elevated particularly in the patients with T2D and insulin resistance, and reasons for their discordant effects in diabetics remain unknown .
Of a total of 2,709 individuals studied, targeted sequencing was performed on 940 subjects and 1,769 subjects were used for the replication studies. All these participants were part of the AIDHS/SDS [15–17]. Of the 940 sequenced samples, 820 passed the stringent QC based on multiple parameters and were used for further analysis. S1 Table describes details of the lipid candidate gene regions selected for targeted resequencing. A summary of high-quality variants analyzed for their distribution and association with lipid-related traits, diabetes and other cardiometabolic risk factors is provided in S2 Table.
In this investigation, we have attempted to identify functional variants by resequencing 13 known candidate genes of dyslipidemia using an endogamous population of Punjabi Sikhs known to have high risk for cardiovascular diseases [11,12,38–41]. Despite considerable success of GWAS, whole-genome, and exome sequencing, including studies from our group [15,42–44], the genetic mechanisms that predispose people to metabolic and cardiovascular disease risk factors remain poorly understood. Of these 13 selected loci with prior evidence of association with three major lipids (HDL cholesterol, LDL cholesterol, and TG) in European populations [10,12,45,46], variants in ANGPTL3, GCKR, MLXIPL, LPL, TRIB1 and APOE genes have been shown to be associated with lipid phenotypes in South Asians . Fine mapping of ~195 kb region encompassing Chr11q23.3 [APO-A1-C3-A4-A5, ZNF259, and BUD13] by targeted genotyping revealed a strong association of this region with HTG (rs964184; p = 1.6×10-39) in Punjabi Sikhs and South Asians) . Here we have intended to capture putatively functional rare and less common variants from coding, non-coding, and intergenic regions including variants influencing gene regulation and expression within and around these known candidate genes. The degree of clinical heterogeneity existing in the CAD or cardiometabolic phenotypes imposes serious limitations in our ability to effectively measure genetic risk, environmental exposure, and their interactions. Additionally, most post-GWA studies on candidate gene sequencing have predominantly been focused on European populations which provide limited information on the usefulness of variants in populations of non-European ancestry. Moreover, the post-GWAS exome arrays could capture the majority of low-frequency variants in European populations only when the sample size exceeded >300,000 . However, such studies in other disparate populations are sparse. The current investigation in family and population based sample from the AIDHS/SDS is an effort to identify missing heritability associated with GWAS-driven loci of dyslipidemia, specifically the HTG by using candidate gene resequencing.