LB.02.07] A GENOME-WIDE ASSOCIATION STUDY ON BLOOD PRESSURE DIPPING IN FINNISH HYPERTENSIVE PATIENTS

Abstract

Objective: Attenuated nocturnal fall (non-dipping) of blood pressure (BP) is a predictor of cardiovascular target organ damage. No genome-wide association studies (GWAS) on BP dipping have been previously reported. Design and method: We conducted a GWAS on BP dipping in 204 hypertensive Finnish men from the Genetics of Drug Responsiveness in Essential Hypertension (GENRES) study that included four four-week placebo periods before four antihypertensive drug treatment periods. Ambulatory BP was recorded at the end of each placebo and drug period. We analyzed dipping using ambulatory BP recordings after the placebo periods (173 subjects with four and 31 subjects with three recordings). The DNA samples were genotyped using Illumina HumanOmniExpress-12 Bead Chip. Residuals of mean systolic and diastolic BP dipping were calculated using linear regression with appropriate covariates. The residuals were tested for genetic associations under additive genetic model and signals with P-value < 1 × 10–5 were followed up in two independent Finnish cohorts; Haemodynamics in Primary and Secondary Hypertension (DYNAMIC) (N = 183) and Dietary, Lifestyle and Genetic determinants of Obesity and Metabolic Syndrome (DILGOM) (N = 180). Left ventricular mass index (LVMI) assessed by echocardiography was measured in GENRES subjects. Results: In GENRES, rs4905794 near BCL11B achieved genome-wide significance (β = −6.7 mmHg, P = 2.5 × 10–8 for systolic and β = −4.1 mmHg, P = 3.6 × 10–6 for diastolic dipping). In the replication analysis, the effect of rs4905794 was in the same direction but statistically insignificant in DYNAMIC (β = −1.1 mmHg, P = 0.5 for systolic, and β = −1.4 mmHg, P = 0.16 for diastolic dipping) and did not replicate in DILGOM. The G allele of rs4905794 that associated with larger BP dipping was also associated with smaller LVMI in GENRES subjects (β = −7.6 g/m2, P = 0.02). There were further 10 SNPs in 8 loci with P-values <1 × 10–5, but they did not replicate with P-values <0.05 in both DYNAMIC and DILGOM. Conclusions: We report here the first GWAS on BP dipping. rs4905794 near BCL11B had a genome-wide-significant signal for BP dipping and associated with LVMI in GENRES. Evidence from previous GWASs (lipoprotein(a) and aortic stiffness) and animal models (energy metabolism, cardiac hypertrophy) support the role of BCL11B in cardiovascular physiology.