Blood Pressure and Hypertension Are Associated With 7 Loci in the Japanese Population
Background— Two consortium-based genome-wide association studies have recently identified robust and significant associations of common variants with systolic and diastolic blood pressures in populations of European descent, warranting further investigation in populations of non-European descent.
Methods and Results— We examined the associations at 27 loci reported by the genome-wide association studies on Europeans in a screening panel of Japanese subjects (n=1526) and chose 11 loci showing association signals (1-tailed test in the screening, P<0.3) for an extensive replication study with a follow-up panel of 3 Japanese general-population cohorts (n ≤24 300). Significant associations were replicated for 7 loci—CASZ1, MTHFR, ITGA9, FGF5, CYP17A1-CNNM2, ATP2B1, and CSK-ULK3—with any or all of these 3 traits: systolic blood pressure (P=1.4×10−14 to 0.05), diastolic blood pressure (P=1.9×10−12 to 0.05), and hypertension (P=2.0×10−14 to 0.006; odds ratio, 1.10 to 1.29). The strongest association was observed for FGF5. In the whole study panel, the variance (R2) for blood pressure explained by the 7 single-nucleotide polymorphism loci was calculated to be R2=0.003 for male and 0.006 for female participants. Stratified analysis implied the potential presence of a gene-age-sex interaction, although it did not reach a conclusive level of statistical significance after adjustment for multiple testing.
Conclusions— We have confirmed 7 loci associated with blood pressure and/or hypertension in the Japanese. These loci can guide fine-mapping efforts to pinpoint causal variants and causal genes with the integration of multiethnic results.
Received August 25, 2009; accepted April 6, 2010.
Hypertension affects one third of the adult population worldwide and is a major risk factor of cardiovascular disease.1 Even within the normal range, increases in blood pressure are associated with the risk of death from vascular diseases such as stroke and ischemic heart disease.2 Although lifestyle influences (eg, excess salt and alcohol intake and lack of exercise) are known to increase blood pressure and the risk of developing hypertension, a substantial contribution of genetic factors to the overall disease pathogenesis has been documented by a number of epidemiological studies.3,4 Despite considerable efforts to study the molecular genetics of hypertension, the inherently complex nature has hampered progress in elucidating the involved genes.
Clinical Perspective on p 2309
The recent advent of genome-wide association (GWA) studies has enabled identification of common variants associated with common diseases and traits.5 Several GWA studies have thus far found loci associated with hypertension or blood pressure, few of which have attained genome-wide significance levels (eg, P<5×10−8).6 Therefore, blood pressure variation in the general population is assumably due to multiple variants with small effects; very large study samples are needed to provide definitive evidence of the principal hypertension-susceptibility gene(s).
Furthermore, meta-analysis of multiple GWA studies can facilitate detection of the variants with modest effects. With regard to hypertension, 2 consortia, the Cohorts for Heart and Aging Research in Genome Epidemiology Consortium6 and the Global Blood Pressure Genetics consortium,7 have been formed to conduct meta-analyses of GWA studies in populations of ≈30 000 individuals of European descent in each, with follow-up analysis using genotypes from large cohorts (>80 000 samples when combined). These large-scale analyses have identified 13 loci showing genome-wide significant association with systolic (SBP) and diastolic (DBP) blood pressures and/or hypertension, together with a number of other loci showing suggestive association.
Given the appreciable ethnic differences in the clinical presentation of hypertension between populations of European and non-European ancestry,8 it is essential to test the genetic associations previously identified for Europeans in the other populations. We therefore conducted a cross-population comparison of susceptibility loci in a Japanese population. We also examined the influences of sex, age, and obesity on the strength of genetic association with blood pressure and/or hypertension to validate a complex interplay of genes and these nongenetic factors as previously reported in Europeans.4,9,10
Detailed characteristics of the subjects analyzed in each stage of the study are described in the online-only Data Supplement and Table 1. Our genetic studies for blood pressure and hypertension were originally organized as part of an ongoing GWA study for cardiometabolic disorders among Japanese subjects; the multistage design is summarized in Figure 1. In stage 1 of the GWA scan, 1526 Japanese samples were genotyped with 456 825 single-nucleotide polymorphism (SNP) markers. From the loci reported for Europeans,6,7 those showing a tendency for association (P<0.3 by 1-tailed test in stage 1) were chosen for genotyping in a replication (stage 2a) panel involving 5745 Japanese subjects (hereafter called the Amagasaki cohort) consecutively enrolled in a population-based setting as described elsewhere11 and 2425 subjects primarily recruited for the case-control study (1750 cases and 1776 controls) in stage 2a. Subsequently, the loci showing a significant association (P<0.05 in stage 2a) were tested in another replication (stage 2b) panel comprising 2 cohorts of 12 569 and 3975 subjects (hereafter called the Fukuoka cohort and the Kita-Nagoya Genomic Epidemiology [KING] Study cohort, respectively) randomly selected from the Japanese general population.12 From the stage 2b panel, 1544 cases and 5055 controls were further chosen for case-control comparison. All participants from these studies provided written informed consent, and the local ethics committees approved the protocols.
Blood pressure levels were classified according to the Japanese Society of Hypertension Guidelines for the Management of Hypertension (JSH2009) as described elsewhere.13 Cases were enrolled from clinical practices or annual medical checkups at medical institutions and university hospitals in accordance with the uniformly defined criteria. These criteria included (1) SBP ≥160 mm Hg and/or DBP ≥100 mm Hg for untreated subjects (grade II and III hypertension in JSH2009); (2) receiving long-term antihypertensive treatments; (3) no secondary form of hypertension as evaluated by an extensive workup that included serum creatinine and electrolytes, chest radiography, ECG, urinalysis, and other hematologic screening tests; and (4) age of onset ≤60 years. Normotensive controls were defined as follows: SBP <130 mm Hg and DBP <85 mm Hg without antihypertensive treatments (normal blood pressure in JSH2009) and age ≥50 years.
SNP Genotyping and Quality Control
In the stage 1 GWA scan, genotyping was performed with the Infinium HumanHap550 BeadArray (Illumina, San Diego, Calif), interrogating 555 352 SNPs (online-only Data Supplement). Data cleaning and analysis were performed with PLINK as described elsewhere.14 Population stratification was checked by multidimensional scaling analysis of the pairwise distance between samples measured over all SNPs (online-only Data Supplement). The λ value for the genomic control was 1.00 to 1.02, indicating the absence of systematic confounding such as population stratification in the GWA study samples (see text and Figure I in the online-only Data Supplement).
In the replication study, samples were genotyped with the TaqMan assay for 11 SNPs from 11 unique loci previously identified in Europeans.6,7 These included CASZ1 (rs880315), MTHFR (rs17367504), ITGA9 (rs155524), MDS1 (rs448378), FGF5 (rs16998073), EFCAB1 (rs1910252), CACNB2 (rs11014166), CYP17A1-CNNM2 (rs12413409), PLEKHA7 (rs406890), ATP2B1 (rs2681472), and CSK-ULK3 (rs1378942). The genotypic distribution of all tested SNPs was in Hardy-Weinberg equilibrium (P>10−3). We obtained successful genotyping call rates of >99% for the whole characterized sample.
SNP Association Analysis
The SNPs were tested for associations with blood pressure and hypertension with linear regression analysis and the Cochran-Armitage trend test, respectively. In the linear regression models, we adjusted continuous SBP and DBP for age, age2, and body mass index (BMI) separately by sex (all panels) and the sample enrollment site (GWA study panel only). For the individuals receiving antihypertensive therapies, blood pressure was imputed by adding 15 and 10 mm Hg for SBP and DBP, respectively.7,15 For the replication study of candidate SNP loci, we set in the stage 1 screening a significance level of P<0.3 (1-tailed) for the trait(s) that could be claimed by the original GWA studies at the individual loci based on the power calculation (Figure II and Table I in the online-only Data Supplement). In stages 2a and 2b, 1-tailed values of P<0.05 were considered statistically significant for the loci previously shown to have genome-wide significant (P<5×10−8) association in Europeans; for an association to be considered significant, it had to involve the same risk allele as that reported in Europeans and was accordingly assessed with the 1-tailed test. Otherwise, a significance level was set at 1-tailed value of P<0.05 after adjustment for multiple testing with the Bonferroni correction. For the quantitative trait analysis, the genetic effects estimated in each of the multistage Japanese panels were combined by using the inverse variance method, except for a few types of analyses mostly of gene-age-sex interaction, for which the subject data from 3 population cohorts were combined after calibrating intercohort differences in blood pressure. For the case-control study, cases and controls were pooled from multistage panels (online-only Data Supplement). We used PLINK, the R software, and the rmeta and meta packages to test for the associations.
Assessment of the Cumulative Effect of Risk Variants
We assessed the cumulative effect of multiple SNPs by using a blood pressure risk score,6 which was a weighted sum across the SNPs (separately for SBP and DBP) combining the β coefficients and doses of the risk alleles. To illustrate a trend by grouping the individuals, the blood pressure risk score was rounded to 1 mm Hg for SBP (groups ≤−3 to ≥3) and 0.5 mm Hg for DBP (groups ≤−1.5 to ≥1.5), with each risk score representing deviations from the study mean. Within a risk score group, we calculated the mean and 95% confidence interval (CI) of empirical blood pressure.
The multilocus risk of hypertension for an individual was assessed as the sum of doses of the risk alleles weighted by the logarithm of the odds ratio (OR) at the SNPs. We simulated a population with 20% prevalence by using bootstrap sampling. The prevalence was drawn from the National Health and Nutrition Survey, Japan 2006 (survey data are available at http://www-bm.mhlw.go.jp/houdou/2008/04/dL/h0430–2c.pdf [in Japanese]). In this simulated population, we arranged the individuals in order of their multilocus risk, sorted them into 20 equal-sized groups (5% in each), and calculated the actual proportion of hypertensive subjects in each group. The mean and 95% CI of the group-wise prevalence were estimated on the basis of 1000 bootstrap sampling trials.
Test of Ethnic Diversity and Interaction With Sex, Age, and BMI
The per-allele effect size, β, of an SNP on blood pressure was compared between the ethnic groups and among the subgroups stratified by sex, age, and BMI. Blood pressure was standardized as a z score within each ethnic group before cross-population comparison. The interaction of effect estimates with ethnicity (Japanese versus European) and with sex, age, and BMI was analyzed by the Cochran Q test and ANOVA, respectively (online-only Data Supplement). To collectively assess the proportion of variance for blood pressure explained by an SNP, risk score, or BMI, we calculated the coefficient of determination (R2).
Screening of GWA Data
We found no significant association by genome-wide exploration in the Japanese (Figure III in the online-only Data Supplement). Using the stage 1 GWA data set, we then examined association signals at 27 unique loci for which significant or suggestive evidence of association was previously reported by 2 GWA meta-analyses of Europeans.6,7 Although no individual locus had strong statistical significance, the SNP markers from 11 (of 27) loci showed a tendency for association (1-tailed test, P<0.3) in the direction concordant with that previously reported for Europeans (Tables II and III in the online-only Data Supplement).
Replication of Selected SNPs in the Japanese
In the follow-up study (stages 2a and 2b), we tested blood pressure and hypertension associations with 11 SNP loci selected via the stage 1 screening. First, in the Amagasaki cohort panel (n=5745), we found significant (P<0.05) associations of 5 SNPs with DBP and/or SBP (Table IV in the online-only Data Supplement). In the stage 2a case-control panel, significant associations with hypertension were detected for 6 SNPs: 4 of the SNPs replicated for the blood pressure association plus rs155524 (ITGA9) and rs1378942 (CSK-ULK3). These were further genotyped in the Fukuoka Cohort Study panel (n=12 569) and the KING Study panel (n=3975), and the genetic associations were confirmed for 7 SNP loci (P<0.05): CASZ1 (rs880315), MTHFR (rs17367504), ITGA9 (rs155524), FGF5 (rs16998073), CYP17A1-CNNM2 (rs12413409), ATP2B1 (rs2681472), and CSK-ULK3 (rs1378942) (Tables 2 and 3⇓ and Table IV in the online-only Data Supplement). Here, 1-tailed P<0.0125 (0.05/4) was considered significant for 4 (of 11) typed SNP loci, CASZ1, ITGA9, MDS1, and EFCAB1, which showed suggestive association in Europeans. Despite modest intercohort fluctuations in the strength of association, the genetic associations were appreciably reproducible across all stages and different population cohorts in the Japanese (Table IV and Figure IV in the online-only Data Supplement).
Ethnic Heterogeneity in Effect Sizes
Of 27 candidate loci, the direction of association was concordant at 37% (10 of 27) and inverted at 19% (5 of 27) across the 3 traits between 2 ethnic groups (Table II in the online-only Data Supplement). Moreover, of 13 loci that attained genome-wide significance levels in Europeans, association was robustly replicated at 38% (5 of 13), was not reproducible at 54% (7 of 13) with an average power of >0.5, and was not assessable at 8% (1 of 13; no polymorphism detected for SNPs at SH2B3) in the Japanese. Thus, the real rate of positive (true positive plus false negative) seems to be between 5 of 13 and 9 of 13 for the reported genome-wide significant loci.
Among the 7 loci replicated in the whole study panel, there was significant (P<0.05) interethnic heterogeneity for rs880315 (CASZ1) and rs16998073 (FGF5). The effect sizes for these loci were significantly larger in the Japanese than in Europeans (Tables 2 and 3⇑ and Figure VA and VB in the online-only Data Supplement).
Cumulative Effect of 7 Associated Loci
Despite the small value of explained variance (R2) at each risk locus, knowledge about multiple-risk loci would allow identification of individuals with accumulated genetic risk.14 Therefore, we calculated empirical blood pressure levels with the study subjects categorized according to their blood pressure risk scores. Across the risk score groups, the mean level of SBP and DBP increased significantly in a stepwise fashion in the Japanese (Figure 2A; P<10−33 for each trend). We also investigated the combined risk of hypertension on the basis of 7 associated loci by analysis of the resampled data set. We thus found a 1.9-fold variation in hypertension prevalence from the lowest to the highest estimated risk groups for the combination of 7 associated loci in our study (Figure 2B).
Factors Modulating Genetic Associations
To estimate the impacts of risk factor variables on blood pressure variance in the Japanese, we first performed linear regression analysis with the stage 2a (Amagasaki Study) panel (Table V in the online-only Data Supplement). Sex was found to be a principal predictor of blood pressure (P<10−36); hence, the subjects were analyzed separately by sex. In both sexes, age and age2 were ranked in the first place (R2=0.108 to 0.162) and BMI was in the second place (R2=0.095 to 0.118) consistently for SBP and DBP. When the subjects were categorized into 5 age classes (30 to 39, 40 to 49, 50 to 59, 60 to 69, and 70 to 79 years), the mean blood pressure increased and the blood pressure distribution within an age class approached the normal distribution with advancing age in both sexes (Figure VI in the online-only Data Supplement). Besides sex, age, and BMI, although modest in the explained variance (R2), alcohol intake and smoking correlated with blood pressure increase (R2=0.002 to 0.015) and decrease (R2=0.003 to 0.012), respectively, in the studied population (Table V in the online-only Data Supplement).
We calculated R2 for blood pressure explained by BMI and that explained by the risk score on combining the 7 SNP data within each age class separately by sex on the basis of the results for the 3 population cohorts (online-only Data Supplement). In both sexes, BMI contributed to blood pressure variance to a larger extent at 30 to 49 years of age than at ≥50 and years of age. The cumulative genetic effects (represented by 7 SNPs) were almost unchanged throughout the age classes in women, whereas they were the highest at 40 to 49 years of age and tended to decrease with age (in particular, R2<0.0001 at 70 to 79 years of age) in men (Figure 3 and Figure VII in the online-only Data Supplement). There was a difference in the cumulative genetic effects on DBP at ≥70 years of age between the sexes (P=0.046, not adjusted for multiple comparisons; Table VI in the online-only Data Supplement). We then examined the modulation of genetic associations by sex, age, and BMI but obtained no conclusive evidence of interaction after the correction of multiple testing (see text and Tables VI through VIII in the online-only Data Supplement).
By this multistaged replication study involving a total of 25826 Japanese subjects, we have confirmed genetic associations at 7 loci with blood pressure and/or hypertension in a population of non-European ancestry. In addition to 5 (of 7) loci—FGF5, MTHFR, CYP17A1-CNNM2, ATP2B1, and CSK-ULK3—that have genome-wide significance in Europeans, our study replicated 2 loci, CASZ1 and ITGA9, for which suggestive evidence of association was previously reported.6,7 Most of the replicated variants were associated concordantly with both blood pressure traits (SBP and DBP) and dichotomous hypertension, except for MTHFR and ITGA9, for which either blood pressure or hypertension association was rather pronounced in the Japanese (Tables 2 and 3⇑ and Figure VA and VB in the online-only Data Supplement). Furthermore, despite the small number of SNP loci described here (7 SNPs) and the small proportion of blood pressure variation explained by these SNPs (R2=0.005), the conjoint effect of multiple risk alleles on blood pressure levels and hypertension is sufficient to increase the cardiovascular risk (Figure 2A and 2B). Moreover, while examining the impacts of confounding factors on the genetic associations with blood pressure and/or hypertension, we detected a trend for male-specific interactions between age and the genetic associations; ie, the SNP–blood pressure association is influenced by age in a sex-specific manner, at least for the loci tested in the present study (Figure 3). Although the interaction has not been proven to be statistically significant, this seems to be in good agreement with the previous studies,4,9 which documented the modulation of genetic effects on blood pressure by sex and age.
There is substantial overlap of common genetic variants influencing blood pressure and the risk of hypertension between the Japanese (or East Asians) and Europeans, whereas some ethnic heterogeneity is likely to exist in the effect sizes for each risk allele. In the Japanese, the strongest association was observed for FGF5. Although this locus shows a robust and significant association in Europeans,7 we detected significant cross-population heterogeneity (P=0.01 for SBP) in the strength of association with blood pressure; the SBP and DBP effects (on the SD scale) were >1.4 times larger in the Japanese than in Europeans (Figure VA in the online-only Data Supplement). Furthermore, the OR for hypertension significantly differed at FGF5 between the populations (OR, 1.29 and 1.10 in the Japanese and Europeans, respectively; Figure VB in the online-only Data Supplement). Despite a less conserved linkage disequilibrium block in the Japanese than in Europeans near rs16998073 (3.4 kb upstream of FGF5 on 4q21; Figure VIII in the online-only Data Supplement), we could still detect strong association signals at this locus across the 3 Japanese cohorts (Figure IV in the online-only Data Supplement). The FGF5 gene itself is a promising candidate because it encodes a member of the fibroblast growth factor family, and the protein (fibroblast growth factor 5) is noted for its effects in promoting angiogenesis in the heart.16
Similarly, significant cross-population heterogeneity was observed at CASZ1. Although suggestive evidence of association with SBP has been identified at this locus in Europeans (β=0.53 and P=4.8×10−6),6 the genetic association was stronger for all 3 traits in the Japanese (β=1.08 and P=2.2×10−8 for SBP, β=0.79 and P=4.9×10−12 for DBP, OR=1.18 and P=3.0×10−7 for hypertension; Tables 2 and 3⇑). The associated SNP, rs880315, is located in the intron of the CASZ1 gene. Although little is known about its molecular function, CASZ1 (encoding the castor homolog 1, zinc finger) is reportedly a cell survival gene that controls apoptosis and tumor formation.17 One mechanism linking CASZ1 to vascular or heart dysfunction would be inflammatory responses involving cell adhesion, permeability, and apoptosis.
Genetic associations were confirmed at 5 other loci—MTHFR, CYP17A1-CNNM2, ATP2B1, CSK-ULK3, and ITGA9—in the Japanese, with some interethnic fluctuation in the effect sizes. On the other hand, several candidate loci previously reported by GWA studies of Europeans6,7 did not pass our stage 1 screening. This could reflect the limited power to detect the modest effect size (average power=0.62 in the stage 1 panel; Table I in the online-only Data Supplement) and/or the lack of (or weaker) association among the Japanese subjects due partly to differences in linkage disequilibrium patterns between the ethnic groups.
The strengths of the present study are that we used 3 independent cohorts of the Japanese population; that the study subjects were enrolled from regions of Japan with no strong population stratification18; and that the subjects totaled 25 826 (49.7% female), belonging to a wide range of age groups (18 to 97 years). In the population at large, blood pressure (in particular, SBP) tends to increase and the prevalence of hypertension becomes higher with advancing age, when prominent sex differences are widely recognizable.19 When we examined modulation of SNP–blood pressure association by sex and age (Figure 3 and Table VI in the online-only Data Supplement), we unexpectedly found that R2 for blood pressure, explained by a combination of SNPs or by an individual SNP, fluctuates with age in a sex-specific manner. The possibility of this gene-age-sex interaction is worthy of note but should be carefully interpreted because it has not attained a conclusive level of statistical significance after adjustment for multiple testing. For both SBP and DBP, R2 explained by the 7 SNPs was highest at 40 to 49 years of age, when R2 explained by BMI was also the highest, and tended to decrease at older ages in men but was almost stable in women. Older subjects are more likely to have hypertension and to be taking antihypertensive drugs than younger adults. While analyzing blood pressure, we adjusted for the use of antihypertensive therapy by adding 15 and 10 mm Hg to SBP and DBP, respectively, as has been carried out by previous studies to reduce bias and to improve statistical power.7,15 It is possible that SBP and DBP measures thus imputed as quantitative traits incorporate some bias in the older age groups, thereby leading to underestimation of R2. This speculation, however, cannot explain the sexual dimorphism detected in the possible interactions of the genetic associations with age. That is, considering that age-related changes in R2 explained by BMI were similarly observable in both sexes, the male-specific fluctuations in R2 explained by the 7 SNPs cannot be attributed to simple bias in blood pressure and warrant further investigation.
We acknowledge that there are several limitations in the present study. Although we recruited a major part of the study subjects (stages 2a and 2b) from 3 cohorts of the Japanese population, there is some intercohort diversity in the range of age and treatment profiles (Table 1). Given the appreciable influence of age on the genetic association and the potential age-related bias in adjustment for the use of antihypertensive therapy, these factors could reduce the power of detecting the association of genetic variants with modest effects. In this respect, larger sample sizes (particularly in the stage 1 screening) are required to thoroughly examine modest associations at the loci previously identified by the GWA studies of Europeans6,7 in the Japanese; the power calculation is given in Figure II and Table I and explained in the text of the online-only Data Supplement. Moreover, because cause-and-effect relationships cannot be inferred by cross-sectional studies, longitudinal studies need to be performed to validate sex- and age-specific modulation of blood pressure associations and to clarify the underlying mechanisms.
We have confirmed that 7 loci are associated with blood pressure and/or hypertension in the Japanese. These loci can guide our fine-mapping efforts to pinpoint causal variants and causal genes with the integration of multiethnic results.
We thank all the people who have continuously supported the hospital-based cohort study at the International Medical Center of Japan, the Amagasaki Study and the Kyushu University Fukuoka Cohort Study, and the KING Study. We also thank Dr Suminori Kono, Masahiro Ogasawara, Kazutoyo Ienaga, Dr Chikanori Makibayashi, and the many physicians of the Amagasaki Medical Association for their assistance in collecting the DNA samples and accompanying clinical information.
Sources of Funding
This work was supported by grants for the Core Research for Evolutional Science and Technology from the Japan Science Technology Agency; the Program for Promotion of Fundamental Studies in Health Sciences, National Institute of Biomedical Innovation Organization; and the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
Pilia G, Chen WM, Scuteri A, Orrú M, Albai G, Dei M, Lai S, Usala G, Lai M, Loi P, Mameli C, Vacca L, Deiana M, Olla N, Masala M, Cao A, Najjar SS, Terracciano A, Nedorezov T, Sharov A, Zonderman AB, Abecasis GR, Costa P, Lakatta E, Schlessinger D. Heritability of cardiovascular and personality traits in 6,148 Sardinians. PLoS Genet. 2006; 2: e132.
Seda O, Tremblay J, Gaudet D, Brunelle PL, Gurau A, Merlo E, Pilote L, Orlov SN, Boulva F, Petrovich M, Kotchen TA, Cowley AW Jr, Hamet P. Systematic, genome-wide, sex-specific linkage of cardiovascular traits in French Canadians. Hypertension. 2008; 51: 1156–1162.
Levy D, Ehret GB, Rice K, Verwoert GC, Launer LJ, Dehghan A, Glazer NL, Morrison AC, Johnson AD, Aspelund T, Aulchenko Y, Lumley T, Köttgen A, Vasan RS, Rivadeneira F, Eiriksdottir G, Guo X, Arking DE, Mitchell GF, Mattace-Raso FU, Smith AV, Taylor K, Scharpf RB, Hwang SJ, Sijbrands EJ, Bis J, Harris TB, Ganesh SK, O'Donnell CJ, Hofman A, Rotter JI, Coresh J, Benjamin EJ, Uitterlinden AG, Heiss G, Fox CS, Witteman JC, Boerwinkle E, Wang TJ, Gudnason V, Larson MG, Chakravarti A, Psaty BM, van Duijn CM. Genome-wide association study of blood pressure and hypertension. Nat Genet. 2009; 41: 677–687.
Newton-Cheh C, Johnson T, Gateva V, Tobin MD, Bochud M, Coin L, Najjar SS, Zhao JH, Heath SC, Eyheramendy S, Papadakis K, Voight BF, Scott LJ, Zhang F, Farrall M, Tanaka T, Wallace C, Chambers JC, Khaw KT, Nilsson P, van der Harst P, Polidoro S, Grobbee DE, Onland-Moret NC, Bots ML, Wain LV, Elliott KS, Teumer A, Luan J, Lucas G, Kuusisto J, Burton PR, Hadley D, McArdle WL; Wellcome Trust Case Control Consortium, Brown M, Dominiczak A, Newhouse SJ, Samani NJ, Webster J, Zeggini E, Beckmann JS, Bergmann S, Lim N, Song K, Vollenweider P, Waeber G, Waterworth DM, Yuan X, Groop L, Orho-Melander M, Allione A, Di Gregorio A, Guarrera S, Panico S, Ricceri F, Romanazzi V, Sacerdote C, Vineis P, Barroso I, Sandhu MS, Luben RN, Crawford GJ, Jousilahti P, Perola M, Boehnke M, Bonnycastle LL, Collins FS, Jackson AU, Mohlke KL, Stringham HM, Valle TT, Willer CJ, Bergman RN, Morken MA, Döring A, Gieger C, Illig T, Meitinger T, Org E, Pfeufer A, Wichmann HE, Kathiresan S, Marrugat J, O'Donnell CJ, Schwartz SM, Siscovick DS, Subirana I, Freimer NB, Hartikainen AL, McCarthy MI, O'Reilly PF, Peltonen L, Pouta A, de Jong PE, Snieder H, van Gilst WH, Clarke R, Goel A, Hamsten A, Peden JF, Seedorf U, Syvänen AC, Tognoni G, Lakatta EG, Sanna S, Scheet P, Schlessinger D, Scuteri A, Dörr M, Ernst F, Felix SB, Homuth G, Lorbeer R, Reffelmann T, Rettig R, Völker U, Galan P, Gut IG, Hercberg S, Lathrop GM, Zelenika D, Deloukas P, Soranzo N, Williams FM, Zhai G, Salomaa V, Laakso M, Elosua R, Forouhi NG, Völzke H, Uiterwaal CS, van der Schouw YT, Numans ME, Matullo G, Navis G, Berglund G, Bingham SA, Kooner JS, Connell JM, Bandinelli S, Ferrucci L, Watkins H, Spector TD, Tuomilehto J, Altshuler D, Strachan DP, Laan M, Meneton P, Wareham NJ, Uda M, Jarvelin MR, Mooser V, Melander O, Loos RJ, Elliott P, Abecasis GR, Caulfield M, Munroe PB. Genome-wide association study identifies eight loci associated with blood pressure. Nat Genet. 2009; 41: 666–676.
Wang X, Poole JC, Treiber FA, Harshfield GA, Hanevold CD, Snieder H. Ethnic and gender differences in ambulatory blood pressure trajectories: results from a 15-year longitudinal study in youth and young adults. Circulation. 2006; 114: 2780–2787.
Shi G, Gu CC, Kraja AT, Arnett DK, Myers RH, Pankow JS, Hunt SC, Rao DC. Genetic effect on blood pressure is modulated by age: the Hypertension Genetic Epidemiology Network Study. Hypertension. 2009; 53: 35–41.
Pausova Z, Gaudet D, Gossard F, Bernard M, Kaldunski ML, Jomphe M, Tremblay J, Hudson TJ, Bouchard G, Kotchen TA, Cowley AW, Hamet P. Genome-wide scan for linkage to obesity-associated hypertension in French Canadians. Hypertension. 2005; 46: 1280–1285.
Nanri A, Yoshida D, Yamaji T, Mizoue T, Takayanagi R, Kono S. Dietary patterns and C-reactive protein in Japanese men and women. Am J Clin Nutr. 2008; 87: 1488–1496.
Ogihara T, Kikuchi K, Matsuoka H, Fujita T, Higaki J, Horiuchi M, Imai Y, Imaizumi T, Ito S, Iwao H, Kario K, Kawano Y, Kim-Mitsuyama S, Kimura G, Matsubara H, Matsuura H, Naruse M, Saito I, Shimada K, Shimamoto K, Suzuki H, Takishita S, Tanahashi N, Tsuchihashi T, Uchiyama M, Ueda S, Ueshima H, Umemura S, Ishimitsu T, Rakugi H. The Japanese Society of Hypertension guidelines for the management of hypertension (JSH 2009). Hypertens Res. 2009; 32: 3–107.
Takeuchi F, Serizawa M, Yamamoto K, Fujisawa T, Nakashima E, Ohnaka K, Ikegami H, Sugiyama T, Katsuya T, Miyagishi M, Nakashima N, Nawata H, Nakamura J, Kono S, Takayanagi R, Kato N. Confirmation of multiple risk Loci and genetic impacts by a genome-wide association study of type 2 diabetes in the Japanese population. Diabetes. 2009; 58: 1690–1699.
Vatner SF. FGF induces hypertrophy and angiogenesis in hibernating myocardium. Circ Res. 2005; 96: 705–707.
Yuan ZR, Wang R, Solomon J, Luo X, Sun H, Zhang L, Shi Y. Identification and characterization of survival-related gene, a novel cell survival gene controlling apoptosis and tumorigenesis. Cancer Res. 2005; 65: 10716–10724.
Yamaguchi-Kabata Y, Nakazono K, Takahashi A, Saito S, Hosono N, Kubo M, Nakamura Y, and Kamatani N. Japanese population structure, based on SNP genotypes from 7003 individuals compared to other ethnic groups: effects on population-based association studies. Am J Hum Gene. 2008; 83: 445–456.
Safar ME, Smulyan H. Hypertension in women. Am J Hypertens. 2004; 17: 82–87.
A substantial contribution of genetic factors to hypertension or high blood pressure has been documented by a number of epidemiological studies. However, identifying common variants affecting blood pressure using genome-wide association studies has proved challenging compared with the success of genome-wide association studies of other common diseases. Recently, 2 consortium-based genome-wide association studies have successfully identified robust and significant associations of common variants with systolic and diastolic blood pressures in populations of European descent, warranting further investigation in populations of non-European descent. In the present study, we examined the associations at 27 loci reported by the genome-wide association studies on Europeans in a screening panel of Japanese subjects (n=1526) and chose 11 loci showing association signals (1-tailed test in the screening, P<0.3) for an extensive replication study with a follow-up panel of 3 Japanese general population cohorts (n ≤24 300). Significant associations were replicated for 7 loci—CASZ1, MTHFR, ITGA9, FGF5, CYP17A1-CNNM2, ATP2B1, and CSK-ULK3—with any or all of the following 3 traits: systolic blood pressure (P=1.4×10−14 to 0.05), diastolic blood pressure (P=1.9×10−12 to 0.05), and hypertension (P=2.0×10−14 to 0.006; odds ratio, 1.10 to 1.29). Besides confirming the association at 7 loci in the Japanese, we have detected the potential presence of gene-age-sex interactions, emphasizing the importance of such interactions in the molecular approach to hypertension genetics.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.109.904664/DC1.