Methylome-Wide Association Study of Central Adiposity Implicates Genes Involved in Immune and Endocrine Systems

Anne E. Justice, Geetha Chittoor, Rahul Gondalia, Phillip E. Melton, Elise Lim, Megan L. Grove, Eric A. Whitsel, Ching‐Ti Liu, L. Adrienne Cupples, Lindsay Fernández‐Rhodes, Weihua Guan, Jan Bressler, Myriam Fornage, Eric Boerwinkle, Yun Li, Ellen W. Demerath, Nancy L. Heard‐Costa, Daniel Levy, James D. Stewart, Andrea Baccarelli, Lifang Hou, Karen N. Conneely, Trevor A. Mori, Lawrence J. Beilin, Rae‐Chi Huang, Penny Gordon‐Larsen, Annie Green Howard, Kari E. NorthEpigenomics 2020 · DOI

Abstract

We conducted a methylome-wide association study to examine associations between DNA methylation in whole blood and central adiposity and body fat distribution, measured as waist circumference, waist-to-hip ratio, and waist-to-height ratio adjusted for body mass index, in 2,684 African American adults in the Atherosclerosis Risk in Communities (ARIC) study. We validated significantly associated Cytosine-phosphate-Guanine methylation sites (CpGs) among adults using the Women’s Health Initiative and Framingham Heart Study participants (combined N=5,743) and generalized associations in adolescents from The Raine Study (N=820). We identified 11 CpGs that were robustly associated with one or more central adiposity trait in adults and 2 in adolescents, including CpG site associations near TXNIP, ADCY7, SREBF1, and RAP1GAP2 that had not previously been associated with obesity-related traits.

Introduction

While Body Mass Index (BMI) remains the standard clinical metric for obesity, it is an imperfect tool. BMI fails to distinguish between lean muscle and adipose tissue, and more importantly, it does not account for the anatomical distribution of fat. Central adiposity—the accumulation of visceral fat around abdominal organs—is a significantly stronger predictor of cardiometabolic conditions, including Type 2 Diabetes and hypertension, than total body mass alone.

Most existing epigenetic research has focused on BMI within relatively homogeneous populations of European descent. This study addresses two critical gaps in the literature: the need for specificity in isolating the "belly fat" signal from overall body size, and the need for diversity by focusing on African American adults—a group disproportionately affected by obesity-related morbidity but historically underrepresented in genomic research.

Methods

The research team performed a Methylome-Wide Association Study (MWAS) to identify chemical modifications to DNA that regulate gene expression without altering the underlying sequence. Specifically, they examined DNA methylation at Cytosine-phosphate-Guanine (CpG) sites.

Study Design and Cohorts

  • Discovery Phase: The study analyzed approximately 390,000 CpG sites in the blood of 2,684 African American adults from the Atherosclerosis Risk in Communities (ARIC) study.
  • Statistical Framework: Linear Mixed Models (LMM) were employed to control for technical noise, while Bayesian Model Averaging was used to select optimal covariates, including age, smoking status, and educational attainment.
  • Validation and Generalization: Significant findings were meta-analyzed across 5,743 adults from the Women’s Health Initiative (WHI) and the Framingham Heart Study (FHS). To determine if these markers emerge early in life, the team tested for generalization in 820 adolescents from The Raine Study.
  • Biological Interrogation: Data from the GTEx project and the GWAS Catalog were used to map these methylation sites to gene expression and existing genetic risk factors.

Key Results

Identification of a Distinct Epigenetic Signature

By adjusting for BMI, the researchers isolated 18 CpG sites uniquely associated with central adiposity. Following validation across multi-ethnic adult cohorts, 11 CpG sites remained robustly associated with abdominal fat distribution. Notably, 72% of these associations showed consistent directional effects across different ancestries, suggesting a conserved biological mechanism for fat distribution.

Novel Genetic Associations

The study implicated several genes previously unlinked to obesity traits:

  • TXNIP (Thioredoxin Interacting Protein): Increased methylation at cg19693031 was strongly associated with lower central adiposity. This association was significant in both adults and 17-year-olds, suggesting that metabolic risk markers are established early in the life course.
  • ADCY7 and SREBF1:** These genes are involved in immune response and lipid metabolism, respectively. Their identification supports the hypothesis that visceral fat acts as an active endocrine organ.
  • RAP1GAP2:** Implicated in platelet function, suggesting a potential link between abdominal fat and cardiovascular risk through non-traditional pathways.

Immune-Metabolic Cross-talk

A significant finding was the overlap between central adiposity markers and genes active in macrophages and T-cells. This reinforces the theory that central fat promotes systemic inflammation, bridging the gap between metabolic storage and immune system activation.

Discussion

One of the most striking observations was the minimal overlap between these methylation sites and known GWAS "hits" for obesity. This suggests that epigenetics captures environmental influences—such as diet, stress, and socioeconomic factors—that traditional genetic studies may overlook.

Limitations

Despite the robust findings, the study has limitations. The use of whole blood as a proxy for adipose tissue may not capture tissue-specific methylation patterns. Furthermore, the cross-sectional nature of the data prevents definitive conclusions regarding causality; it remains unclear whether methylation changes precede or follow the accumulation of central fat. Finally, the lack of comprehensive data on physical activity and alcohol consumption across all cohorts may leave some confounding factors unaddressed.

Conclusion

This MWAS identifies a unique epigenetic fingerprint for central adiposity that is independent of overall BMI. The discovery of 11 robust markers, particularly those present in both adults and adolescents, provides new insights into the early biological pathways of metabolic disease. Future longitudinal research is required to determine the causal direction of these associations and their potential as biomarkers for clinical intervention.

Summary updated

Abstract

<b>Aim:</b> We conducted a methylome-wide association study to examine associations between DNA methylation in whole blood and central adiposity and body fat distribution, measured as waist circumference, waist-to-hip ratio and waist-to-height ratio adjusted for body mass index, in 2684 African-American adults in the Atherosclerosis Risk in Communities study. <b>Materials & methods:</b> We validated significantly associated cytosine-phosphate-guanine methylation sites (CpGs) among adults using the Women's Health Initiative and Framingham Heart Study participants (combined n = 5743) and generalized associations in adolescents from The Raine Study (n = 820). <b>Results & conclusion:</b> We identified 11 CpGs that were robustly associated with one or more central adiposity trait in adults and two in adolescents, including CpG site associations near <i>TXNIP</i>, <i>ADCY7</i>, <i>SREBF1</i> and <i>RAP1GAP2</i> that had not previously been associated with obesity-related traits.
BibTeX
@inproceedings{justice2020methylome,
  author = {Anne E. Justice and Geetha Chittoor and Rahul Gondalia and Phillip E. Melton and Elise Lim and Megan L. Grove and Eric A. Whitsel and Ching‐Ti Liu and L. Adrienne Cupples and Lindsay Fernández‐Rhodes and Weihua Guan and Jan Bressler and Myriam Fornage and Eric Boerwinkle and Yun Li and Ellen W. Demerath and Nancy L. Heard‐Costa and Daniel Levy and James D. Stewart and Andrea Baccarelli and Lifang Hou and Karen N. Conneely and Trevor A. Mori and Lawrence J. Beilin and Rae‐Chi Huang and Penny Gordon‐Larsen and Annie Green Howard and Kari E. North},
  title = {{Methylome-Wide Association Study of Central Adiposity Implicates Genes Involved in Immune and Endocrine Systems}},
  booktitle = {Epigenomics},
  year = {2020},
  doi = {10.2217/epi-2019-0276},
}

Loading related content...

Figure