Identification of Genetic Loci Simultaneously Associated With Multiple Cardiometabolic Traits

Authors

Alexis C Wood
Amit Arora
Michelle Newell
Victoria L Bland
Jin Zhou
Nicola Pirastu
Jose M Ordovas
Yann C Klimentidis

Publication Date

4-1-2022

Journal

Nutrition, Metabolism and Cardiovascular Diseases

DOI

10.1016/j.numecd.2022.01.002

PMID

35168826

PMCID

PMC9275655

PubMedCentral® Posted Date

4-1-2023

PubMedCentral® Full Text Version

Author MSS

Published Open-Access

yes

Keywords

Adiposity, Cholesterol, HDL, Genetic Loci, Genome-Wide Association Study, Humans, Hypertension, Polymorphism, Single Nucleotide, Waist-Hip Ratio, GWAS, cardiometabolic disease, pleiotropy, risk alleles, fat distribution, Epidemiology, Risk Factors

Abstract

Background and aims: Cardiometabolic disorders (CMD) arise from a constellation of features such as increased adiposity, hyperlipidemia, hypertension and compromised glucose control. Many genetic loci have shown associations with individual CMD-related traits, but no investigations have focused on simultaneously identifying loci showing associations across all domains. We therefore sought to identify loci associated with risk across seven continuous CMD-related traits.

Methods and results: We conducted separate genome-wide association studies (GWAS) for systolic and diastolic blood pressure (SBP/DBP), hemoglobin A1c (HbA1c), low- and high- density lipoprotein cholesterol (LDL-C/HDL-C), waist-to-hip-ratio (WHR), and triglycerides (TGs) in the UK Biobank (N = 356,574-456,823). Multiple loci reached genome-wide levels of significance (N = 145-333) for each trait, but only four loci (in/near VEGFA, GRB14-COBLL1, KLF14, and RGS19-OPRL1) were associated with risk across all seven traits (P < 5 × 10-8). We sought replication of these four loci in an independent set of seven trait-specific GWAS meta-analyses. GRB14-COBLL1 showed the most consistent replication, revealing nominally significant associations (P < 0.05) with all traits except DBP.

Conclusions: Our analyses suggest that very few loci are associated in the same direction of risk with traits representing the full spectrum of CMD features. We identified four such loci, and an understanding of the pathways between these loci and CMD risk may eventually identify factors that can be used to identify pathologic disturbances that represent broadly beneficial therapeutic targets.