Genetic polymorphisms that influence smooth muscle cell contraction and relaxation may affect the response to anti-hypertensive therapy. Our finding of lower blood pressure in the setting of treatment for hypertension, particularly with β-blockade, in Caucasian KCNMB1-E65K carriers in two community cohorts has implications for personalization of therapy in hypertension.
Genetic effects regulating arterial tone can have a direct consequence on blood pressure (BP). The large conductance calcium-activated potassium channel, also called the BK channel, regulates vascular smooth muscle cell membrane polarity and by extension cellular contraction and relaxation.1, 2, 3 The human β1-regulatory-subunit gene includes a non-synonymous polymorphism at codon 65 (KCNMB1-E65K, rs11739136) that increases the probability of BK-channel opening, following calcium stimulation. Fernandez-Fernandez et al.4, 5 first reported the protective effect of E65K against diastolic hypertension in a Spanish cohort (n=3876), yet this finding was not supported by Kokubo et al.,6 who after excluding subjects receiving anti-hypertensive medication found no association between E65K and BP in a cohort (n=3652) of Japanese subjects. We hypothesized that population and BP treatment differences contribute to the variable association of the E65K polymorphism with BP. To test this hypothesis, we studied E65K and BP phenotypes in two community-based cohorts (Heart Strategies Concentrating On Risk Evaluation, Heart SCORE and the Framingham Heart Offspring Study, FHS).
The KCNMB1 polymorphism rs11739136 was genotyped using a TaqMan assay, in DNA samples from Heart SCORE (n=1075) and FHS (n=1519) subjects without a history of cardiovascular disease. Baseline evaluation of Heart SCORE subjects, and examination cycle six of FHS subjects were used in these analyses, which were performed separately by cohort and race. We used analysis of covariance to assess whether E65K was associated with adjusted mean systolic and diastolic pressure. Similarly, logistic regression was used to estimate the adjusted odds of being in the upper quartile of systolic or diastolic pressure in relation to the E65K polymorphism. To allow for the assessment of an E65K interaction, we controlled anti-hypertensive therapy by adding a binary covariate to our regression models, rather than applying a constant to correct the pressures of treated individuals.7
Heart SCORE African-Americans had a lower K-allele frequency compared with Caucasians (6.2 versus 9.8%); genotype frequencies conformed to Hardy–Weinberg equilibrium. Hypertension was present in approximately 70% of African-American and 50% of Caucasian Heart SCORE subjects (Supplementary Table 1). The mean-adjusted systolic (133.7±7.3 versus 142.1±9.4 mm Hg, P=0.02) and diastolic (77.6±3.1 versus 79.7±2.6, P=0.02) pressures were lower in the presence of the K allele in Heart SCORE Caucasians. No association was seen in Heart SCORE African-Americans (P=0.39 and 0.78, respectively). Caucasian K-allele carriers were also significantly less likely to have systolic BP (SBP) in the upper quartile of Heart SCORE (SBP >144 mm Hg; odds ratio (OR) (95% confidence intervals (CI))=0.55 (0.33–0.91)). A protective effect was also found for elevated diastolic BP (DBP >86 mm Hg; OR (95% CI)=0.47 (0.28–0.79)).
We repeated our analyses in FHS participants. The K-allele frequency (10.0%) was similar to Caucasian Heart SCORE subjects; however, the E65K polymorphism was not associated with BP or hypertension. Considering the lack of association of E65K with BP in the Framingham Heart Study, we concluded that the KCNBM1 E65K polymorphism has no general effect on BP.
Importantly we considered whether the increased rates of diastolic hypertension in the Spanish (27.0%)5 compared with Heart SCORE (14.8%) and Framingham Heart Study subjects (6.8%) may explain the variable association of E65K with BP. To examine the role of hypertension in the association of E65K with BP, we tested our a priori hypothesis that E65K interacts with anti-hypertensive therapy, age and gender.5, 8 There was no interaction between E65K and age in Heart SCORE, and tests for interaction with gender were non-significant in both cohorts. By comparison, in Heart SCORE, the test for interaction between genotype status and anti-hypertensive medication use was significant (P=0.003, DBP model). K carriers receiving anti-hypertensive therapy had lower systolic and diastolic pressures (Table 1); BP was not different in the absence of anti-hypertensive medication use. We confirmed this finding in FHS, with older (based on median age of 58 years) K carriers exhibiting lower systolic and diastolic pressure in the presence of treatment (three-way test for interaction P=0.0009, SBP model; 0.002, DBP model).
Next, we determined the role of specific medication classes in the relationship of E65K with BP. Given the limited number of subjects within each subgroup, we pooled data from the Heart SCORE Caucasian and FHS cohorts in order to maximize our power to study a medication class effect.9 The frequency of anti-hypertensive therapy use did not differ by genotype (P>0.05). Among the treated subjects (n=671), 50.4% reported receiving diuretics (n=338), 37.1% renin-angiotensin system blockers (n=249), 34.8% β-blockers (n=234) and 23.5% calcium-channel blockers (n=158). A significant proportion (46.6%, n=313) reported receiving multiple classes of agents. Despite a limited sample size, we found significantly lower diastolic pressure among K carriers receiving β-blockade (Table 1). No significant difference in BP was noted by genotype status among the other medication classes. When the analysis was restricted to subjects receiving β-blockers alone (n=35), the results were similar. Formal tests for interaction between β-blockade and E65K genotype were not significant (P=0.24, 0.08 for SBP and DBP models, respectively). There was no difference in the frequency of β-blockade use by E65K status (P=0.28).
Our results showing a selective association of E65K with BP in the setting of anti-hypertensive medication use are supported by the INVEST trial in which K carriers receiving verapamil achieved BP control more rapidly and effectively compared to non-carriers.8 We identified a trend towards lower pressures among all classes of medication, but statistical significance was reached solely in the β-blockade subgroup. We consider two molecular mechanisms by which E65K would have a selective effect on BP in the setting of hypertension and β-blockade therapy. First, the increased calcium-triggered BK activity inherent to the mutant β1-subunit protein4 may have a general effect of preserving BK channel-dependent vasodilation in the setting of reduced overall β1-subunit expression caused by hypertension.10, 11 Alternatively, the E65K mutant β1-subunit may directly influence the BP response to β-blocker drugs through the physical interaction of the BK channel with the β-adrenergic receptor.12
Because our results are consistent with earlier studies,4, 5, 8 we consider a false positive association to be less likely. Limited effect and sample sizes may have restricted our power to detect a genotype–phenotype association, especially in African-American Heart SCORE subjects whose K-allele frequency was lower and sample size smaller. Our assessment of the effect of anti-hypertensive medication was challenging in this cohort study design. Conservative power calculations suggested a sample size of >11 000 would be needed to adequately assess the impact of β-blockade on E65K-carrier status in a population-based cohort. Our finding of a significant interaction with anti-hypertensive medication use, despite the constraint of sample size, suggests a true association.
Essential hypertension remains a complex disease, which is influenced by both genetic and environmental factors, and gene–environmental interaction contributes to individual disease burden and treatment response. The finding of a selective effect of E65K on BP in the setting of hypertension and β-blockade, supports future in vivo molecular studies to more clearly define the role of the E65K polymorphism in BP regulation in response to therapy.
This project is funded, in part, under a grant with the Pennsylvania Department of Health (SER) (contract no. ME-02-384). The Department specifically disclaims responsibility for any analyses, interpretations or conclusions. This project was supported by the National Institutes of Health program project Grant no. HL077378 and training grant no. HL069770 (MEM). This work was supported by the National Heart, Lung and Blood Institute's Framingham Heart Study (contract no. N01-HC-25195) and Boston University School of Medicine.
About this article
Disclosure/conflict of interest
Expert Review of Clinical Pharmacology (2012)