Gene environment interactions and postural blood pressure in community exposed elderly men : the VA normative aging study

Background: Evidence has shown a positive association between environmental air pollution and blood pressure (BP) change, a marker of autonomic function and a predictor of cardiovascular outcomes. We investigated the associations between postural changes in systolic and diastolic BP and fine particulate matter (PM2.5), single nucleotide polymorphisms (SNPs), and SNP-by-PM2.5 interactions. Methods: Systolic and diastolic BP measurements were obtained from men in the Normative Aging Study who lived in the Boston area and visited the study center every 3-5 years. Postural BP change was defined as the difference between standing and supine measurements for systolic and diastolic BP. PM2.5 was measured continuously with exposures averaged over 48 hours prior to study visit for analyses. Blood samples were analyzed for 975 SNPs in 62 candidate genes and included haplotype tag SNPs and known or suspected functional SNPs. Data were stratified into a screening cohort and testing cohort. In the screening cohort, the relationship between postural change in BP and SNP, PM2.5, and SNP-by-PM2.5 interactions was analyzed using linear mixed effect models with random intercepts for each subject to account for repeated BP measurements within the study period. Models were adjusted for age and body mass index (BMI). A positive outcome in the screening cohort was defined as a P-value < 0.1 for the interaction term. SNPs meeting these criteria were subsequently analyzed in the testing cohort. SNPs were confirmed as significant predictors if they also met criteria of P < 0.05 in the testing cohort with an association in the same direction as in the screening cohort. All confirmed associations were analyzed within the full cohort data in models adjusted for age, BMI, smoking status and use of hypertension medications. Results: 930 participants with a mean (±SD) age of 72 (±7) provided information on exposure, outcomes and covariates of interest. Average systolic BP was 132.97 (±17.9) mmHg with a postural change in pressure of 2.4 (±12.7) mmHg. Average diastolic BP was 78.11 (±10.4) mmHg with -0.5 (±6.5) mmHg change. SNP-by-PM2.5 interactions in ITPR2 and MMP1 decreased the magnitude of change in systolic BP and a polymorphism in the PHF11 gene suppressed postural change in diastolic pressure. For example, the main effect of the minor allele of rs9568232 in PHF11 was associated with greater postural change in diastolic BP (β = 2.5, P = .001) and the interaction was associated with a suppression in diastolic BP change (β = -2.4, P = .001) for an increase of 10 μg/m3 PM2.5. The effect of PM2.5 in subjects with the major variant was non-significant (β = 0.3, P = .23). Conclusions: We observed SNPs associated with decreases in magnitude of postural change in BP and found different candidate genes to be associated with differences in postural change in diastolic and systolic blood pressure. Our findings also suggest that considering gene main effects alone may overlook relationships modified by environmental exposures such as PM2.5.