Introduction

Hypertension is the most common risk factor for atherosclerotic cardiovascular diseases, and its morbidity is greater than 50% in subjects aged 65 years and older in Japan and most Western countries. In Japan, the number of patients with hypertension is estimated to be greater than 40 million.1 Hypertension is a multifactorial disease in which genetic and environmental factors are closely related. Specifically, genetic factors can influence blood pressure elevation by 30–50%.2 Thus, over the last decade, many genetic studies have aimed to clarify the causal genes of hypertension, as reviewed previously.3 Genetic research on hypertension started using the candidate gene approach by investigating renin–angiotensin system (RAS)-related genes.4, 5, 6 Recently, several large-scale genome-wide association studies (GWASs) for hypertension7, 8, 9 were performed, and 50 single nucleotide polymorphisms (SNPs) were identified as possible causal genes.

The goal of genetic studies on hypertension is mainly to clarify the causal genes of hypertension and the mechanisms of blood pressure elevation. Consequently, these approaches may lead to the production of novel antihypertensive drugs. Another important aim is to establish personalized pharmacotherapy based on genetic information. These were the two main aims of the Japanese Millennium Project for Hypertension.10 As mentioned, the first aim may be reasonably addressed by large-scale GWAS, but the second aim has not yet been examined in the field of hypertension research.

Recent studies indicate that the heterogeneity of patients’ responses to antihypertensive treatment is, at least in part, genetically determined.11 This finding underscores the role of pharmacogenetic/pharmacogenomic research in identifying either functional genetic variations or those variations inherited by linkage disequilibrium. These variations may serve as markers to provide a more individualized evaluation and selection of agents from different drug classes to treat hypertension.12 In this review article, we discuss current trends and future directions of pharmacogenomic studies on antihypertensive drugs.

Current trends in pharmacogenetic/pharmacogenomic studies of hypertension

Polymorphisms of genes involved in the RAS, such as ACE I/D6 polymorphisms and AGT M235T5 polymorphisms, were discovered to be associated with high blood pressure nearly 20 years ago. These studies investigated the association of RAS gene polymorphisms and responses to antihypertensive drugs, including angiotensin I-converting enzyme inhibitors (ACEIs),13 angiotensin II receptor blockers (ARBs),14 beta blockers15 and diuretics.16 Our group also endeavored to clarify the susceptible gene polymorphisms to thiazide diuretics17 and calcium channel blockers (CCBs)18 by candidate genes approaches, although there are very limited reports about gene polymorphisms associated with the effects of CCBs. We clarified the associations between CACNA1D rs312481G>A and rs3774426C>T, and CACNA1C 527974G>A and treatment with a more significant reduction in BP for the combined presence of the L type of CCB’s antihypertensive effect in this paper.18

In the Japanese Millennium Genome Project, we identified a strong candidate gene, regulator of G-protein-coupled receptor (GPCR) signaling 2 (RGS2), associated with human hypertension19, 20 and the efficacies of antihypertensive drugs.21 RGS2 is a molecule that has a key role in GPCR signaling, including activation of the angiotensin II type 1 receptor. The RGS2-knockout mouse is characterized by a high blood pressure phenotype that is hyper-responsive to ARB therapy.22 We investigated the correlation between genetic mutations and polymorphisms in RGS2 and human hypertension. We found that rare mutations and common polymorphisms in RGS2 were associated with hypertension in Japanese19 and African American patients.23 Furthermore, we indicated that a SNP in RGS2 could increase responsiveness to the antihypertensive effects of ACEIs and CCBs.21 Taken together, we consider that genetic variations in RGS2 could be new pharmacogenomic targets for some antihypertensive drugs.

However, results from these early pharmacogenetic studies using candidate gene approaches that focused on RAS-related genes or other genes related to blood pressure (BP) regulation are inconsistent. Harrap et al.24 indicated that ACE I/D polymorphisms were not associated with stroke onset or ACEI efficacy in a subanalysis of the randomized control trial, The Perindopril pROtection aGainst REcurrent Stroke Study (PROGRESS) In 2005, Arnett et al.25 reported the first results from the GenHAT study, which was a large-scale genetic analysis of participants in the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) study, a randomized clinical trial that examined the protective effects of four antihypertensive drugs against cardiovascular events. From this report,25 ACE I/D polymorphisms genotyped in 38 000 subjects were found to be unassociated with any cardiovascular event and increased responsiveness to ACEIs, thereby confirming Harrap’s report.24 These results greatly influenced genetic analysis using candidate gene approaches of RAS-related genes. There are several reports on candidate gene polymorphism analyses using the GenHAT data, including vasoactive substance-like natriuretic peptide,26 nitric oxide,27 matrix metalloproteinase28 and coagulation factor polymorphisms.29, 30 These studies indicate that gene polymorphisms targeted by candidate gene approaches may have a strong influence on the efficacy of antihypertensive and other cardiovascular drugs; however, most of these results have not been replicated. Thus, candidate gene approaches to pharmacogenetics/pharmacogenomics studies for antihypertensive drugs may not be sufficient to identify all relevant gene variations that affect the responsiveness to antihypertensive drugs. Therefore, GWAS for pharmacogenomic studies of antihypertensive drugs is needed.31

Target genes related to pharmacokinetic function

According to the review article11 by Turner et al., most current studies in pharmacogenetics/pharmacogenomics of antihypertensive drugs are using candidate gene approaches to investigate receptors, enzymes and peptides related to physiologic blood pressure regulation, including those involved in the renin–angiotensin and sympathetic nervous systems, as well as those genes that regulate production of vasoactive substances and water–electrolyte metabolism. Other approaches to clarify the gene polymorphisms that increase responsiveness to antihypertensive drugs are to target mutations and polymorphisms in pharmacokinetic enzyme-associated genes. Certainly, the drug metabolism of most current antihypertensive drugs is not via cytochrome P450 (CYP).12 However, some β-blockers, CCBs and ARBs are metabolized by CYP. We focused on polymorphisms in the gene that codes for CYP2C9, which is a partial metabolite of the ARB, losartan.32 We resequenced CYP2C9 in 724 Japanese individuals, including 39 hypertensive patients receiving losartan treatment. Of the two novel missense mutations identified, the Arg132Gln variant showed a fivefold lower intrinsic clearance of diclofenac metabolized by CYP2C9 when expressed in a baculovirus-insect cell system, whereas the Arg335Gln variant had no substantial effect on clearance. Several known missense variations were also found, and7% of Japanese participants (53 out of 724) carried one of the deleterious alleles (CYP2C9*3, *13, *14, *30 and Arg132Gln) as heterozygotes. After 3 months of losartan treatment, systolic blood pressure was unchanged in two patients with the CYP2C9*1/*30 variation, suggesting that they exhibited impaired in vivo CYP2C9 activity. CYP2C9*30 might be associated with a diminished response to the antihypertensive effects of losartan.32, 33 These rare mutations related to the effects of antihypertensive drugs should be clarified for all antihypertensive drugs.

Genome-wide approaches to gene polymorphisms that increase responsiveness to antihypertensive drugs

As mentioned above, the results of pharmacogenetic/pharmacogenomic studies using candidate genes to identify gene polymorphisms that increase responsiveness to antihypertensive drugs are largely controversial. Thus, GWAS also had a desirable application in this research field. To the best of our knowledge, there is only a report by Turner et al.34 that associated SNPs with the effects of thiazide diuretics using GWAS and the 100K Affymetrix GeneChip microarray system (Affymetrix, Santa Clara, CA, USA). In this report, haplotype trend regression identified a region of chromosome 12q15 in which haplotypes constructed from three successive single-nucleotide polymorphisms (rs317689, rs315135 and rs7297610), in close proximity to lysozyme (LYZ), YEATS domain-containing 4 (YEATS4) and fibroblast growth receptor substrate 2 (FRS2), were significantly associated with the diastolic BP response to thiazide diuretics. These results were replicated in other ethnic sample sets in this study.34 Recently, this group reported that expression analysis in thiazide-treated subjects showed significantly different pre-treatment leukocyte YEATS4 expression in the rs7297610 genotype group.35 These results support the use of GWAS analyses to identify novel genes influencing antihypertensive drug responses. The same group proposed a new investigation, called the Pharmacogenomic Evaluation of Antihypertensive Responses (PEAR) study, and recruited 800 hypertensive individuals and randomized them to receive atenolol or hydrochlorothiazide.36 The results remain unreported.

In Japan, two major pharmacogenomic studies to identify gene polymorphisms related to the effects of antihypertensive drugs were performed. One is the Hypertension Objective Treatment Based on Measurement by Electrical Devices of Blood Pressure (HOMED-BP) Study, which is a large-scale intervention trial using a PROBE design to determine the optimal target BP level on the basis of self-measured BP with an optimal initial dose of antihypertensive medication.37 We are also carrying out a substudy of HOMED-BP, called the HOMED-BP-GENE study, which is a GWAS to clarify the genetic background of those patients that respond to antihypertensive medications, including ARBs, ACEIs and CCBs. Another pharmacogenomic GWAS in Japan is the GEANE (Gene Evaluation for Antihypertensive Effect of Drugs) study, which is a multicenter trial to examine the response to 3 months of treatment with ARBs, CCBs and thiazide diuretics using a crossover design. These two studies are focusing on SNPs that increase responsiveness to ARBs, CCBs, ACEIs and thiazide diuretics, which are commonly used as first-line drugs in many published hypertension guidelines.1 From these two studies, useful genetic information to establish a tailored therapeutic strategy for hypertension based on a patient’s genetic information could be obtained.

Possibilities of personalized medicine using pharmacogenomic information

Anticancer drug treatment relying on personalized medicine, which considers a patient’s genetic background, is currently under heavy investigation.38 Recently, there was a recommendation considering two patient polymorphisms, CYP2C9 and VKORC1, and their response to warfarin.39 However, tailored medicine for hypertension treatment based on genetic information has not yet been established, although several pharmacogenetic/pharmacogenomic studies have been completed as mentioned in this review. There are several challenges that should be resolved before such tailored medicine can be employed:

  1. 1

    Gene polymorphisms and mutations strongly associated with the effects of antihypertensive drugs have not determined. Most obtained results from pharmacogenetic/pharmacogenomic studies have not been validated or replicated in other studies.

  2. 2

    Clinical parameters are not clear in this research field. The measured BP is quite variable due to the conditions of its measurements. Therefore, it is very difficult to judge responders from non-responders for given antihypertensive drugs.

  3. 3

    Finally, antihypertensive drugs do not have serious or lethal side effects such as those of anticancer drugs or warfarin. Clinicians may not strongly desire personalized medicine for hypertension therapy. Therefore, hypertensive patients are often taking an antihypertensive drug that is less effective than necessary.

The cost of antihypertensive therapies in aging populations has increased steeply, especially in the United States, Japan and other East Asian countries. If personalized medicine for antihypertensive drugs could be established, the cost of antihypertensive drugs may be reduced due to the discontinuation of the use of less effective antihypertensive drugs. Although there are several obstacles in this research field, the effort must be made to establish personalized medicine for antihypertensive treatments to prescribe the most effective drugs to each hypertensive patient.