The KCNH2 Genetic Polymorphism (1956, C>T) Is a Novel Biomarker That Is Associated with CCB and α,β-ADR Blocker Response in EH Patients in China

Background KCNH2 (hERG) potassium channels have an integral role in regulating the excitability of smooth muscle cells. Some pathways driven by angiotensin II, nitric oxide and adrenergic receptors blocker are involved in modulating the properties of KCNH2 potassium channels. And these pathways are closely related to blood pressure regulation. Therefore, we hypothesized that KCNH2 genetic polymorphisms may affect blood pressure response to the antihypertensive drug therapies. Materials and Methods To evaluate the interactions between KCNH2 genetic polymorphisms and individual blood pressure response to antihypertensive drugs, 370 subjects with essential hypertension (EH) were studied. In evaluating the interactions between KCNH2 genetic polymorphisms and drug response to blood pressure, multivariable ANOVA analysis followed by Bonferroni correction were carried out. Results There were statistically significant interactions between KCNH2 (1956, C>T) polymorphism and DBP change (P = 0.010), MAP change (P = 0.014) on azelnidipine or nitrendipine therapy patients at the end of 6 weeks. We found that the KCNH2 (1956,C>T) polymorphism was associated with the hypotensive effects of α,β-ADR blockers of DBP change at the end of 4 and 6 weeks' treatment in an age- and gender-dependent manner (P = 0.007 and 0.019, respectively). Similar results were also observed for changes in MAP at the end of 4 and 6 weeks (P-values were 0.035 and 0.078, respectively). While patients who received imidapril, candesartan and irbesartan therapy, no significant difference in drug response among KCNH2(1956,C>T) genotype was observed. Conclusion We have reported for the first time that KCNH2 (1956, C>T) polymorphism is associated with efficacy of antihypertensive drugs CCBs and ADR blockers, and may serve as a novel biomarker for individualized therapy for certain antihypertensive drugs.


Introduction
Essential hypertension is a heterogeneous disorder with differing causal factors in various patients. Essential hypertension comprises 95% of all causes of hypertension. The seventh Report of the Joint National Committee defined and classified hypertension in adults [1]. The diagnosis of hypertension is made when the average of 2 or more diastolic blood pressure (DBP) measurements on at least 2 subsequent visits is $90 mm Hg and (or) systolic blood pressure (SBP) $140 mm Hg. Isolated systolic hypertension is defined as SBP $140 mm Hg and DBP,90 mm Hg. Hypertension remains a major modifiable risk factor for cardiovascular disease despite important advances in our understanding of its pathophysiology and the availability of effective treatment strategies.
Studies have widely reported that genetic factors are important determinants of hypertension susceptibility and the drug response. Human ether-a-go-go-related gene (hERG or KCNH2) codes for the a subunit of the delayed-rectifier potassium channel (I kr ), and was first discovered as a cDNA homologous to Drosophila Eag gene screened from the human hippocampus cDNA library by Warmke and Ganetzky in 1994. KCNH2 potassium channels are widely expressed in human cardiac and smooth muscle cells. KCNH2 expression can also be detected in liver, pancreas, nervous and tumor tissues. The past researches have shown that functional abnormalities of KCNH2 potassium channels are related to increased risk for QT syndromes and tumors. While some studies have mentioned that the KCNH2 potassium channels also play a fundamental role in modulating the resting membrane potential in smooth muscles and neurons [2,3], suggesting that KCNH2 potassium channels may play a pivotal role in regulating the excitability of excitable cells.
Currently, many studies indicated that the nitric oxide (NO), ANG II, adrenergic receptor (ADR) blockers, L-type calcium channel blockers mediated pathways and some regulatory proteins are involved in modulating the properties of KCNH2 channels. Taglialatela et al. [4] have reported that the reactive oxygen species can evoke down-regulation of KCNH2 protein and decrease I kr but do not affect other potassium channels such as Beag, rDRK1, and mIRK1. In addition, NO release directly activates I kr in opossum esophagus circular muscle depending on Ca 2+ release from the sarcoplasmic reticulum stores [5]. G protein-coupled receptors such as a-ADR (via PLC, PKA, PKC pathway [6]) and b1-ADR (via activation of Gs-adenylate cyclase-cAMP-PKA-14-3-3 pathway [7]) can regulate the expression of KCNH2 potassium and (or) set its gating characteristics. Furthermore, cAMP/PKA/ PKC also can directly or indirectly interact with KCNH2 potassium channels [8]. Studies have observed that Ang II could increase I kr by about 30% with a time constant of approximately 30 s via AT1R in an ATP-dependent and PKC-mediated pathway; and the muscarinic agonist can occlude the effect of Ang II on I kr by enhancing I kr , suggesting that PLC-PIP2 pathway may be involved in regulating KCNH2 potassium channels [9,10,11]. It is worth noting that the interaction between L-type calcium and KCNH2 potassium channels is regulated by endothelin-1(ET-1) induced ANP secretion rather than PKC-mediated pathway [12],while small GTPase Rab11b disorders lead to an increase in L-type Ca 2+ current but a decrease in I kr [13]. Moreover, the KCNH2 potassium channels may directly regulate the characteristics of sodium and calcium channels [14]. Finally, some regulatory proteins such as Caveolin-1, dynamin-2 and monoubiquitination also play an important role in modulating the properties or endocytic degradation of KCNH2 potassium channels [15,16,17].
Interestingly, all the pathways outlined above are closely related to the pathological process of hypertension. Antihypertensive drugs like a-, or b-ADR blocker,ACE-inhibitor, and calcium channel blockers (CCB) are also able to regulate blood pressure via these pathways. The aromatic rings of KCNH2 Y652 and F656 located in S6 domain are the key determinants of a variety of xenobiotics (eg. doxazosin) or endogenous substances (eg. hormones) binding site [18], and the mutations Y652A (1956,C.T) and F656A (1966-1967insT) can attenuate the sensitivity of the targeting substances [19,20]. Guo et al. [21] found that the effect of drugs on KCNH2 channels is not necessarily to block KCNH2 channels, but through targeting the KCNH2interacting proteins such as Caveolin-1, and the mutations of KCNHE2 can influence such mutual effect [17]. KCNH2 (2690, A.C) mutation can create a phosphorylation site, and can also result in the increase in aldosterone synthesis, indicating that this mutation may change the channels activities [22.23]. Hence, we hypothesize that the mutations of KCNH2 (1956, C.T, 1966-1967insT and 2690, A.C) may influence the hypotensive effects of antihypertensive drugs. In this study, a total of 370 eligible patients was studied after a run-in period of 2 weeks and assigned to receive the antihypertensive drugs for 4 weeks or more.

Baseline characteristics of patients
Genotyping by sanger-sequencing failed to find the KCNH2 (1966-1967insT) polymorphism in 85 randomly selected patients. The KCNH2 (2690, A.C) mutation (C allele frequency was 2.9%) was not associated with the hypotensive effects of the antihypertensive drugs, so the data were not listed in article. Baseline characteristics of the patients among gender, age, and KCNH2 (1956,C.T) genotypes groups are shown in Table 1.

Genotypes and allele frequency
In our study population, the prevalence of KCNH2 (1956, C.T) polymorphism CC genotype, CT genotype, and TT genotype were 80.5%, 18.6% and 0.9%, respectively. The T allele frequency was 10.1%, and the C allele frequency was 89.9%. This population had no significant deviations in genotype distributions from expected Hardy-Weinberg equilibrium (see table 2).

Calcium channel blockers (azelnidipine & Nitrendipine):
The descent of blood pressure after azelnidipine or nitrendipine administration was significantly between KCNH2 (1956,C.T) CC genotype and CT/TT genotype groups (as shown in Figure 1). After adjustment for gender, BMI and age, Pvalues for DBP change and MAP (mean arterial pressure) change at the end of 6 weeks were 0.010 and 0.014, respectively. While Pvalues for SBP change at the end of 4 and 6 weeks were 0.193 and 0.059 respectively. Interestingly, our study also found that the relationship between KCNH2 (1956,C.T) mutation and the effects of CCB is gender-dependent. Significant difference in SBP change followed by age, BMI and Bonferroni correction was observed at the end of 2 weeks (genotype*gender, P = 0.028), and a marginally significant difference in MAP change at the end of 6 weeks was also observed (genotype*gender, P = 0.060). The results of subgroups stratified by genotype-gender-specific analyses showed that at the end of 2 weeks, significant difference in SBP change between KCNH2 (1956, C.T) wild type homozygotes and the T allele carriers with Azelnidipine or Nitrendipine therapy was observed only in male patients (Table 3).
3.2 a,b-ADR blockers (doxazosin, celiprolol, atenolol & bisoprolol): As shown in Table 4, patients treated with a,b-ADR blockers(doxazosin, celiprolol, atenolol or bisoprolol monotherapy), a significant KCNH2 (1956,C.T) genotype specific interaction with age and (or) gender in DBP change and MAP change at the end of 4 weeks and 6 weeks was observed (Table 4). Subgroups stratified by genotype-gender, genotype-age, genotype-genderage-specific analysis found that the significant differences in DBP change and MAP change in carriers of the KCNH2 (1956, C.T) CC genotype between male (M-CC) and female (W-CC) patients were observed at the end of 6 weeks. Further study found that the significant differences in DBP and MAP changes between M-CC and W-CC were limited to patients .55 years of age. In addition, when combining doxazosin, atenolol and bisoprolol treatment groups, significant difference in changes in heart rate (HR), DBP and MAP between age#55y-CC and age.55y-CC group was observed at the end of 4 weeks. Similarly, when combining doxazosin, celiprolol, atenolol and bisoprolol treatment groups, significant difference in changes in HR, DBP and MAP in carriers of the CC genotype between age#55 y and age.55 y was also observed at the end 4 weeks (Table 5).

Discussion
Our study indicates that the hypotensive effects of azelnidipine and nitrendipine are more sensitive in T allele carriers than wildtype carriers of KCNH2 (1956, C.T) in EH patients. We also observed that the association of which the hypotensive effect of azelnidipine or nitrendipine is genotype-gender dependent. Literature reported that the sensitivity of doxazosin [19] and W-7 (an inhibition of calmodulin) [24] to KCNH2 potassium channels are decreased due to the mutations like Y652A, F656A in the KCNH2 pore-S6 region. Zhang et al. [25] reported for verapamil caused high-affinity block of KCNH2 channel current is close to its block of L-type Ca2+ channels', whereas diltiazem only weakly suppresses KCNH2 current, and nifedipine has no effect. These studies suggest that the direct actions of dihydropyridine CCBs on KCNH2 potassium channels are less likely. Meanwhile, antioxidant stress approach [26] of CCB may play a pivotal role in increasing the activity of KCNH2 potassium channels [5]. So, according to our results, we can assume that the ability of azelnidipine/ nitrendipine and its downstream semiochemical may interact with variant KCNH2 channels more weakly than the wild-type channels. Then the excitability and contractibility inhibited by CCB on variant allele carriers of KCNH2 potassium channels are more sensitive in smooth muscles [2], and eventually, the hypotensive effects of azelnidipine or nitrendipine on EH patient's therapy is more apparent in carriers of the KCNH2(1956,C.T) T allele.
Unexpectedly, our study shows that the SBP changes between KCNH2 (1956,C.T) CC and CT+TT genotypes are significantly different at the end of 2 and 6 weeks but not at the end of 4 weeks after azelnidipine or nitrendipine therapy (see Fig 2 and table 2). Studies have shown that the modest fluctuations of Ca 2+ concentration may lead to marked changes on KCNH2 K + current: a reduction of I kr in result of external Ca 2+ elevation was not due to Ca 2+ acting as a direct blocker of the open pore, but as an allosteric modulator of the KCNH2 potassium channels [27,28]. However, Koyama et al. [29] found that CCBs have an excellent antioxidant stress effect, which subsequently increase the activity of KCNH2 potassium channels [5]. From the studies above, the phenomenon may be intrigued by the dynamic regulation of calcium channels blocking effect and the antioxidant stress effect of azelnidipine or nitrendipine on KCNH2 potassium channels: the effects of azelnidipine or nitrendipine antioxidant stress pathway can account for T allele carriers, while the calcium channels blocking effect may be a better explanation for CC genotype carriers, both of which should be studied intensively.
As summarized in table 4 and Table 5, the KCNH2 (1956,C.T) polymorphism is associated with the effects of a,b-ADR blockers in a genotype-age-gender dependent manner. Interestingly, stratified analyses according to the interactions between genotype and age, gender were observed, the significantly different effects of a,b-ADR blockers on EH patients only exist in KCNH2 (1956,C.T) CC*gender-, CC*age-or CC*gender*age-specific groups. The hypotensive effects of b-ADR blockers are more obvious in age#55y-CC than age.55-CC genotype carriers, and a significant difference in PP change between CC and CT+TT genotypes at the end of 4 weeks was also observed. However, a-ADR blocker in the men CC genotype carriers group is more sensitive than women group (see Figure 2). When analyzing combined a,b-ADR blockers, we found that the effects in men CC genotype carriers group is better than women, so does it between age#55y-CC and age.55y-CC genotype carriers, but there is no significant difference in T allele carriers between the group.
Studies have shown that activating PKA and cAMP in endoplasmic reticulum surface can up-regulate the expression of KCNH2 [30,31]. Progesterone and b-estradiol which can induce the trafficking defect of KCNH2 [32,33,34]. While, dihydrotestosterone can repair damaged KCNH2 and improve its expression [35]. Moreover, the expression of KCNH2 mRNA in females is lower than that in males, and significant difference in KCNH2 expression between age,55 y and age.55 y group in LQTS patients is also reported [36]. Additionally, Thomas et al. [19] reported that doxazosin can directly block KCNH2, and the sensitivity is influenced by the mutation of KCNH2 (1956,C.T), while other research found that celiprolol involved in modulating the function of endothelial cells and had certain antioxidant stress effects [37,38]. Furthermore, b-ADR blockers possess weak affinity with KCNH2 potassium channels, while atenolol did not block I kr current [30,31], and b1-ADR blockers (a1-ADR blockers) can be involved in regulating the expression of a1-ADR (b1-ADR) directly or indirectly [39]. Together, the aforementioned studies may provide us some clues to the reason why the KCNH2 (1956,C.T) polymorphism is related to the genotype-age-gender dependent effects of a,b-ADR blockers, and the high sensitivity of a,b-ADR blockers in men or age,55 y group.
We also found that the KCNH2 (1956,C.T) was related to HR change (P = 0.035) rather than BP change at the end of 6 weeks between age#55 yCC(n = 36) and age.55yCC(n = 62) medicated with imidapril, candesartan or irbesartan. Irbesartan and candesartan could decrease the QT interval in hypertension patients due to the increased I kr currents [40,41]. The inhibition of I kr was realized via its phosphorylation by ANGII-AT1R-PKC pathway  Table 3. Stratified analyses of the difference hypotensive effects of CCB between KCNH2 (1956,C.T) genotype and gender interaction in EH patients.   or Ang II directly interact with KCNH2, and that was not influenced by the change of Ca 2+ concentration or PKA pathway. This indicates that it may activate a special subtype of PKC [42], which may be a leading cause to interpret why the KCNH2(1956,C.T) polymorphism is not associated with the response to imidapril, candesartan or irbesartan in EH patents. At present, only a few literatures have reported the mechanisms of these pathways, but its exact mechanism is still unclear. Our study indicated that the function of KCNH2 potassium channels,which regulated by a,b-ADR, oxidative stress, regulator protein (Caveolin-1), effectors (PIP2,PKA/PKC) mediated pathways may through interacting with the pore-S6 region. Furthermore, the KCNH2 (1956C.T) polymorphism dependent on age, as well as gender difference is important to determinate the sensibility of drugs, which are involved in modulating the expression or the properties of KCNH2 potassium channels. Interestingly, our results are in accordance with the in vitro functional studies for KCNH2 (1956C.T) mutation. Hence, we conclude that the KCNH2 (1956,C.T) polymorphism may serve as a novel biomarker for prediction of the response to CCBs and a,b-ADR blockers in EH patients.

Patients and study design
The study protocol was approved by the Ethical Committee of the Institute of Clinical Pharmacology, Central South University, China. The registration number (ChiCTR-RO-12002612) was validated by the Chinese Clinical Trial Registry, and written informed consent was obtained from all patients prior to study entry. The clinical data and DNA samples were graciously provided by Shanghai Institute of Hypertension, Ruijin Hospital affiliated with Shanghai Jiaotong University. This study was an open label clinical trial, in which a total of 453 eligible patients was enrolled after a run-in period of 2 weeks and assigned to receive the drugs for 4 weeks or more, and details of the protocol were shown in Table 6. Blood pressure determinations were performed in the morning after a light breakfast with subjects in the seated position, and following a 30 min quiet resting period. Blood pressure and heart rate were measured by trained nurses, with an automatic blood pressure monitor with intellisense, which allows the detection of alteration of the heart rate by greater than or equal to beat/min and of the blood pressure by greater than or equal to 1 mmHg. Monitors were validated against a mercury sphygmomanometer. The blood pressure values were determined as the average of three measurements taken 10 min apart. Values for SBP and DBP were defined by Korotkoff phase I and V, respectively. Pulse pressure was calculated as PP = (SBP-DBP); mean arterial pressure(MAP) was calculated as MBP = DBP+(PP/ 3).
According to our purpose, all the patients fulfilled the following inclusion and exclusion criteria: male or female patients, heart rates within 55,90 beats/min, SBP$140 mmHg and/or DBP$90 mmHg were included. Patients with secondary hypertension, coronary heart disease, diabetes, obesity (BMI.30 kg/ m 2 ), stroke, renal or liver dysfunction, malignant tumor or  pregnancy women and those whose blood pressure measurements above 180/110 mmHg or remaining lower than 140/90 mmHg during the wash-out period, were withdrawn from the study. Finally, 83 samples were excluded from 453 eligible patients due to the small sample size, loss to follow-up and unqualified g-DNA.

Genotyping
Sequencing in 85 samples in the frequency of KCNH2 rs8179011 (1966-1967insA/T) was not reported in Chinese population. Genotyping of KCNH2 rs1137617 (1956 C.T) polymorphisms were verified by sanger sequencing and polymerase chain reaction and restriction fragment length polymorphism assays (PCR-RFLP).The PCR primers used in the amplification were: 59TACAAGGG CTCTGTGG39 (forward) and 59TGGTGGAAGC GGATGAAC39 (reverse). The oligonucleotide primers were synthesized by Invitrogen Trading (shanghai) Figure 2. KCNH2 (1956,C.T) genetic polymorphism in the response to doxazosin is gender-specific. A,B,C depicts that the changes in systolic blood pressure, diastolic blood pressure and mean arterial pressure in men-CC genotype carriers v.s women-CC genotype carriers with essential hypertension after 2 weekend,4 weekend and 6 weekend follow-up to those who were medications with doxazosin. D showed that all follow-up records have no missing at the end of 2,4 and 6 weeks and the polymorphism of KCNH2(C195T) related to the effects of doxazosin is gender-specific,*P,0.05 as compared with corresponding men-CC. doi:10.1371/journal.pone.0061317.g002 Co., Ltd. The optimized PCR system were carried out with a total volume of 25 ul composed of 10x PCR buffer (2.5 ul), 10x dNTP (2.5 ul), 10 uM for each of the forward and reverse primers (0.5 ul), H 2 O (16.8 ul), g -DNA (2 ul), Taq-ase (0.2 ul), and it was performed for thirty-six cycle of amplification: denaturation at 94uC for 30 s, annealing at 52.8uC for 30 s, and elongation at 72uC for 30 s. While an initial denaturation step was implemented at 94uC for 5 min, a final elongation at 72uC for 5 min. All the processes forenamed were carried out with silver tank PCR instrument (eppendorf AG, Germany). Digestion of the PCR products with Csp6 I (Fermentas) was carried out according to the criteria. The digestion products were analyzed by gel electrophoresis using 2.5% agarose (gene tech. company, Shanghai, China), sequencing were assisted by Shanghai Majorbio Bio-pharm Technology Company. KCNH2 rs1805123(2690,A.C) genotyping was carried out as described by Bezzina et al. [21]. A modified amplification system composed of 2xGC buffer(12.5 ul),10x dNTP (2.5 ul),10 uM sense and anti-sense primer (1 ul), H 2 O (5.8 ul), g-DNA (2 ul), Taq-ase (0.2 ul) was used. The amplification conditions were as follows: initial denaturation at 94uC for 5 min, followed by 36 cycles of 94uC for 30 s, 61.8uC for 30 s, 72uC for 30 s, and then a final elongation at 72uC for 5 min.

Statistical Analysis
All data were presented as mean values 6S.D unless otherwise specified. Difference in the baseline characteristics between genders (men v.s women), KCNH2(1956,C.T) genotypes(CC v.s CT+TT), and ages(age#55 y v.s age.55) were compared by independent-samples T-test or Wilcoxon rank sum test, appropriately. Hardy-Weinberg equilibrium for genotypic distribution of KCNH2 (1956, C.T) polymorphism was analyzed by using x2 test. Allele frequencies were determined by direct gene counting. The differences between groups of the changes (after treatmentbefore treatment) in HR, SBP DBP, MAP, and PP were calculated using multivariable ANOVA analysis followed by Bonferroni correction for multiple comparisons. P-values were adjusted by BMI, gender and age when needed. To address whether significant interaction between age-genotype, gender-genotype, age-gender-genotype specific and the response on antihypertensive therapy, multivariate ANOVA and Stratified Analysis were used. A two-tailed P-value,0.05 was considered significant. Statistical analyses were performed using SPSS 19.0 for Windows software (SPSS, Chicago, IL).