Discovery And Validation Of Unique Associations Of DNA Methylation Regions With 24-hour Blood Pressure Phenotypes In African Americans
Michelle L. Roberts,Theodore A. Kotchen,Pan Xiaoqing,Li Yingchuan,Chun Yang,Liu Pengyuan,Tao Wang,Purushottam Laud,Thomas Chelius,Yannick Munyura,David L. Mattson,Yong Liu,Allen W. Cowley,Srividya Kidambi,Liang Mingyu
DOI: https://doi.org/10.1161/hyp.78.suppl_1.P175
IF: 9.8968
2021-01-01
Hypertension
Abstract:DNA methylation, an epigenetic mark, may reflect the interactions between DNA, environment, and lifestyle. It has been implicated in the development and progression of hypertension, a risk factor for cardiovascular disease. We hypothesize that regions of DNA methylation in blood cells can explain 24h BP phenotypes in African Americans. We performed Reduced Representation Bisulfite Sequencing (RRBS) in a discovery cohort of 281 African Americans. Several DNA methylation regions (MRs) were significantly associated with continuously monitored 24-h, daytime, or nighttime SBP, DBP, PP, and MAP after adjustments for covariates age, sex, and body mass index (False Discovery Rate (FDR) = 0.013 - 0.050). Each of these MRs explained a substantial portion of 24h BP variance, ranging from 6.5% - 9.4%. After FDR adjustment, there were no MRs significantly associated with clinic BPs (FDR > 0.1374), calculated by the average of 4 resting measurements (2 per arm) by sphygmomanometer. To interrogate specific regions of DNA methylation, our lab developed a potentially clinically applicable, deep, and targeted methylation sequencing method called Bisulfite-Specific PCR ULtrapLEx Targeted Sequencing (BULLET-Seq), and tested it in two reference samples for three MRs of interest. BULLET-Seq is able to accurately quantify the 10% changes in the dilution series when methylation rates ranged from ~40% to 90% (a chr19 methylation region; R 2 = 0.95 - 0.97) and can modestly measure these changes when rates range from ~2% to 4% (a chr5 region, R 2 = 0.82), and is questionable when methylation rates are below 2% (a chr13 region, R 2 = 0.03 - 0.27). Validation of the chr19 MR in an independent cohort (n=117) was performed in a single BULLET-Seq run. After covariate adjustment, the chr19 region was significantly associated with 24h BPs (SBP, DBP, and MAP; FDR < 0.05), confirming the findings from the discovery cohort. The MR accounted for up to 1.75% of BP variance in the 24h phenotypes. In conclusion, the reported DNA MRs have potential to be excellent markers for the cumulative effect of factors that influence 24h BPs and the BULLET-Seq workflow can be applied in clinical and population settings to screen up to thousands of patients.