Journal Article

Genetic underpinnings of the heterogeneous impact of obesity on lipid levels and cardiovascular disease

Smit, Roelof A. J., Berndt, Sonja I., Manson, JoAnn E., Young, Kristin L., Cai, Yanwei, Lundin, Jessica, Buyske, Steven, Spracklen, Cassandra N., Fornage, Myriam, Peters, Ulrike, Raffield, Laura M., Gordon-Larsen, Penny, Haiman, Christopher A., Darst, Burcu F., Wang, Zhe, Loos, Ruth J. F., Gignoux, Christopher R., Avery, Christy L., Graff, Mariaelisa, Guo, Boya, Kooperberg, Charles, North, Kari E., Lange, Leslie, Highland, Heather M., Rich, Stephen S., Kim, Daeeun, Buchanan, Victoria L., Kilpeläinen, Tuomas O., Lange, Ethan, Hysong, Micah R., Marouli, Eirini, Zhao, Chi

2025

Overview

Background Obesity is thought to increase cardiovascular disease (CVD) risk partly through dyslipidemia. Yet, obesity’s effects on dyslipidemia are not uniform. Understanding the shared genetic basis between obesity and lipid traits can provide insight into this heterogeneity and its implications for CVD risk. Methods We examined local genetic correlations between three lipid measures [high-density lipoprotein cholesterol (HDL), low-density lipoprotein cholesterol (LDL), and triglycerides (TG)] and body mass index (BMI) using genome-wide association study summary statistics from European ancestry UK Biobank participants. We identified genomic loci with opposing genetic effects on obesity and dyslipidemia risk (protective BMI-lipid loci) and those with concordant directions for both obesity and dyslipidemia risk (adverse BMI-lipid loci). Gene-based association analyses were used to prioritize potential causal genes. We then constructed polygenic risk scores for BMI (PRS BMI ) based on protective and adverse loci and assessed their associations with BMI, lipid levels, CVD, and related traits in the diverse Population Architecture using Genomics and Epidemiology (PAGE) study. PheWAS was performed in the All of Us cohort. Mendelian randomization (MR) was conducted to assess the causal impact of protective/adverse loci on cardiometabolic outcomes. Finally, we investigated the associations with fat distribution traits using MRI-based fat measures in the UK Biobank. Results Among 2495 regions, we identified 789 HDL, 26 LDL, and 494 TG loci with significant local genetic correlation with BMI (including overlapping loci). Of these, 3 HDL, 10 LDL, and 8 TG loci showed protective correlations. Gene-based analyses prioritized 18 candidate causal genes. The protective PRS BMI(+)HDL(+) was associated with higher BMI but favorable lipid profiles and reduced CVD risk in PAGE. PheWAS revealed protective associations with hyperlipidemia, atrial fibrillation, and Alzheimer’s disease. MR supported the favorable causal effects of these protective loci on several cardiometabolic outcomes. Notably, protective PRS BMI(+)TG(−) was uniquely associated with decreased visceral-to-abdominal subcutaneous adipose tissue ratio. Conclusions Identifying and validating genomic loci with shared genetic signals between BMI and lipid levels further supports the importance of genetics in defining the heterogeneous impact of obesity on dyslipidemia and CVD.

• Publisher: BioMed Central,BioMed Central Ltd,Springer Nature B.V,BMC
• Subject: Adipose tissue / Alzheimer's disease / Asian people / Atrial fibrillation / Biobanks / Bioinformatics / Biomedical and Life Sciences / Biomedicine / BMI-lipid pleiotropy / Body fat / Body Mass Index / Cancer Research / Cardiovascular diseases / Cardiovascular Diseases - blood / Cardiovascular Diseases - etiology / Cardiovascular Diseases - genetics / Cholesterol / Cohort analysis / Data processing / Diabetes / Disease / Dyslipidemia / Electronic health records / Epidemiology / Female / Genetic aspects / Genetic Predisposition to Disease / Genome-wide association studies / Genome-Wide Association Study / Genomes / Genomics / Glucose / Heterogeneities in obesity / High density lipoprotein / Hispanic Americans / Human Genetics / Humans / Hyperlipidemia / Lipids / Lipids - blood / Local genetic correlation / Low density lipoprotein / Low density lipoproteins / Magnetic resonance imaging / Male / Mediation / Medicine/Public Health / Mendelian randomization / Mendelian Randomization Analysis / Metabolic disorders / Metabolomics / Middle Aged / Multifactorial Inheritance / Neurodegenerative diseases / Obesity / Obesity - blood / Obesity - complications / Obesity - genetics / Polygenic inheritance / Polygenic risk scores / Polymorphism, Single Nucleotide / Systems Biology / Triglycerides / Triglycerides - blood / Type 2 diabetes / Whole genome sequencing