Abstract P154: Associations between DNA Methylation of Four Genes and Risk of Cardiovascular Disease in the Cardiovascular Health Study
DNA methylation of gene regulatory regions plays an important role in modulating expression. We hypothesized that methylation of four genes selected a priori due to their relationship with cardiovascular disease (CVD) would be associated with CVD risk factors, incident CVD, and mortality. The genes were 1) Tumor Necrosis Factor (TNF) which encodes for the cytokine TNF involved in systemic inflammation; 2) Superoxide Dismutase 3 (SOD3) associated with antioxidant function; 3) Synuclein Gamma (SNCG) believed to be related to neurodegenerative diseases; and 4) Serpin A5 (SERPINA5), a serine protease inhibitor related to coagulation and inflammation. We used quantitative methylation-specific PCR to measure the methylation status of each gene at baseline and 8 years later in 492 older adults enrolled in the Cardiovascular Health Study (CHS). The CHS is a multi-site observational cohort investigating CVD in older adults initiated in 1989/90 with follow-up through the present. We evaluated associations between percent methylation and CVD risk factors using linear regression adjusted for age and gender. Incident CVD and mortality were evaluated using Cox proportional hazards regression over 16 years for morbid events and 8 years for mortality. A Principal Components Analysis constructed four orthogonal factors comprised of the 8 gene methylation measures accounting for 85% of the variance. Mean age of the sample was 71.3 years and all were non-Hispanic Caucasians. Each gene was associated with risk factors at baseline for CVD such as body-mass index (TNF, SNCG, SOD3), blood pressure (TNF, SNCG), hypertension (SERPINA5), diabetes (SNCG, SERPINA5), elevated lipids (TNF, SOD3), and individual coagulation/inflammatory markers including C-reactive protein (SOD3), Interleukin 6 (SERPINA5), fibrinogen (TNF) or Factor 7 (TNF). Differences in associations with CVD risk factors by gender were found for SNCG and SERPINA5. Borderline associations were found between risk of mortality (p=0.07) and incident stroke (p=0.06) for Factors 1 and 4, respectively, each loading primarily with methylation variability of the TNF and SNCG genes. No associations were found between the PCA factors and incident coronary heart disease or heart failure. This may be the first study to show relationships between DNA methylation and CVD risk factors and subsequent disease suggesting a role for epigenetics in gene expression of these phenotypes. Additional studies with larger sample sizes and additional genes are needed to expand these findings and to investigate the impact of DNA methylation modification on subclinical CVD.
- © 2013 by American Heart Association, Inc.