Journal of Human (2002) 16, 789–793 & 2002 Nature Publishing Group All rights reserved 0950-9240/02 $25.00 www.nature.com/jhh ORIGINAL ARTICLE Positive association of CYP11B2 with genetic predisposition to essential hypertension

K Tsukada1, T Ishimitsu1, M Teranishi1, M Saitoh1, M Yoshii1, H Inada1, S Ohta1, M Akashi1, J Minami1, H Ono1, M Ohrui2 and H Matsuoka1 1Department of Hypertension and Cardiorenal Medicine; 2Department of Health Care, Dokkyo University School of Medicine, Mibu, Tochigi, Japan

Predispositions to essential hypertension and cardio- polymorphism was determined in essential hyperten- vascular diseases are possibly associated with gene sion subjects (n ¼ 250) and normotensive subjects polymorphisms of the renin–angiotensin system. Gene (n ¼ 221). The distributions of three genotypes (TT, TC, polymorphisms of angiotensinogen and angiotensin- and CC) were significantly different between the hyper- converting enzyme have been suggested to be tensive and the normotensive groups (v2 ¼ 9.61, risk factors for hypertension and myocardial infarction. P ¼ 0.008). Namely, the frequency of C allele was higher Concerning the polymorphism of aldosterone synthase in the hypertensive patients than in the normotensive (CYP11B2) gene, earlier studies have shown inconsis- subjects (34.2 vs 26.5%, P ¼ 0.010). Our data suggest tent results in terms of its relation to hypertension. In that the À344C allele of CYP11B2 gene polymorphism is the present case–control study, we investigated the associated with the genetic predisposition to develop association of À344T/C polymorphism in the promoter essential hypertension. region of human CYP11B2 gene with genetic predis- Journal of Human Hypertension (2002) 16, 789–793. position to hypertension. The genotype of À344T/C doi:10.1038/sj.jhh.1001484

Keywords: essential hypertension; gene polymorphism; aldosterone

Introduction such as hypertension and myocardial infarction.6 However, the results of earlier studies are not always Aldosterone is the principal mineralocorticoid hor- consistent. Differences in background characteris- mone, which plays a major role in regulating sodium 1 tics of the study subjects such as ethnicity and balance and volume homeostasis. Human aldoster- selection criteria may be responsible for this incon- one synthase (CYP11B2) is a cytochrome P450 sistency. enzyme, which catalyses the terminal steps of In the present study, we investigated the À344T/C aldosterone synthesis in the zona glomerulosa cells 2,3 polymorphism of the CYP11B2 gene in carefully of the adrenal. Therefore, genetic variations of the selected normotensive subjects and patients with CYP11B2 gene that affect its expression may essential hypertension, and thereby evaluated the influence the development of salt-sensitive hyper- association of the genotypes with genetic predis- tension. Several common polymorphisms have been position to essential hypertension. described in the CYP11B2 gene.4,5 Among them, the À344T/C polymorphism, which is located at a putative binding site for the steroidogenic transcrip- tion factor (SF-1), has been recently attracting Subjects and methods interest with regard to its relation to cardiovascular Subjects diseases. Several studies have suggested that the C allele of this gene polymorphism is associated with A group of 250 patients with essential hypertension genetic predisposition to cardiovascular diseases (EH) (154 men and 96 women), aged 53 7 11 years, who had developed hypertension before the age of 50 years were selected from the outpatients clinic of Correspondence: Dr T Ishimitsu, Department of Hypertension and Cardiorenal Medicine, Dokkyo University School of Medicine, Dokkyo University School of Medicine Hospital. Mibu, Tochigi 321-0293, Japan. Hypertension was defined as systolic blood pres- E-mail: [email protected] sure exceeding 140 mmHg and/or diastolic blood CYP11B2 gene polymorphism in hypertension K Tsukada et al

790 pressure exceeding 90 mmHg on two or more con- amplified fragments were digested with HaeIII secutive visits.7,8 Individuals diagnosed as having restriction enzyme (New England BioLabs, MA, secondary causes of hypertension were excluded. USA) and were subjected to electrophoresis on The normotensive control group (NT) included 221 2.0% agarose gels. HaeIII digestion of the 538 bp healthy subjects (144 men and 77 women), aged PCK product yields 274, 138 and 126 bp fragments. 56 7 5 years, who were older than 50 years and free The existence of À344C creates an additional from a history of any cardiovascular diseases. They recognition site of HaeIII resulting in digestion of were recruited from participants of the health check the 274 bp fragment into 203 and 71 bp fragments. programme of Dokkyo University School of Medi- Fragments of 274 bp (T allele) and of 203 and 71 bp cine Hospital. The subjects were considered to be (C allele) were detected. normotensive when their blood pressure was lower than 140 mmHg in systole and 90 mmHg in diastole on multiple occasions. Blood pressure was mea- Statistical analysis sured by a sphygmomanometer, with the subjects in 7 a sitting position after a 10-min rest. All subjects Data are presented as means s.d. Clinical char- were native Japanese and unrelated to each other. acteristics between the two groups were compared Besides the routine laboratory tests, plasma renin by unpaired Student’s t-test for parametric data and by w2 test for categoric data. The allele and genotype activity, plasma angiotensin II, and aldosterone 2 concentrations were measured by radioimmunoas- frequencies in the two groups were compared by w says in 144 male subjects of the NT group. test. Parametric data of the three genotypes were The study protocol was approved by the institu- analysed by ANOVA followed by Dunnett’s t-test. A tional review board, and informed consent was P-value of less than 0.05 was considered to indicate obtained from each subject. statistical significance.

Results Genotype determination Clinical characteristics of study subjects Genomic DNA was isolated from peripheral leuko- cytes according to the standard method using Easy Clinical characteristics of the EH and NT subjects DNA kit (Invitrogen, Carlsbad, CA, USA), and the are listed in Table 1. Systolic and diastolic blood À344T/C polymorphism of the CYP11B2 gene was pressure was markedly higher in the EH group than determined by the analysis of restriction fragment in the NT group. Of the 250 EH patients, 229 length polymorphism (RFLP).9 The DNA fragment (91.6%) were taking antihypertensive medication. In containing À344T/C of the CYP11B2 gene was addition, as expected naturally, body mass index amplified by polymerase chain reaction (PCR). The (BMI), fasting blood glucose, and uric acid were used primers were 50-CAGGGAGGAGACCCCATGT- significantly higher in EH than in NT. A total of 68 GAC-30 (sense, from À528 to À548) and 50- patients with EH were taking lipid-lowering drugs, CCTCCACCCTGTTCAGCCC-30 (antisense, from and serum total cholesterol was lower in EH than in À11 to À29). The PCR profile included 35 cycles NT. However, serum triglycerides were higher and of denaturing at 941C for 60 s, annealing at 671C for serum HDL cholesterol was lower in EH than in NT. 60 s, and polymerization at 721C for 120 s. The The serum creatinine concentration was not sig-

Table 1 Clinical characteristics of the normotensive subjects and hypertensive patients

Parameter Normotensive (n=221) Hypertensive (n=250) P-value

Age (year) 56.1 7 5.4 52.7 7 11.0 o0.001 Sex, male/female 144/77 154/96 Body mass index (kg/m2) 23.9 7 2.4 25.2 7 3.3 o0.001 Habitual smoking, +/À 61/160 127/123 o0.001 Habitual alcohol intake, +/À 101/120 110/140 Systolic blood pressure (mmHg) 117 7 11 171 7 22 o0.001 Diastolic blood pressure (mmHg) 72 7 8 104 7 12 o0.001 Fasting blood glucose (mg/dl) 95 7 9 103 7 16 o0.001 Serum total cholesterol (mg/dl) 208 7 37 197 7 32 o0.001 Serum HDL cholesterol (mg/dl) 54 7 13 51 7 12 0.026 Serum triglycerides (mg/dl) 128 7 75 154 7 88 o0.001 Serum creatinine (mg/dl) 0.9 7 0.5 1.0 7 0.5 Serum uric acid (mg/dl) 5.3 7 1.3 5.8 7 1.2 o0.001

Data are means 7 s.d. HDL, high-density lipoprotein. In the hypertensive group, 229, 68, and 23 patients were on medications for hypertension, hyperlipidaemia and hyperuricaemia, respectively. Blood pressure of the hypertensive group is the value when antihypertensive drugs were not given.

Journal of Human Hypertension CYP11B2 gene polymorphism in hypertension K Tsukada et al

791 Table 2 Genotype and allele frequencies of the À344T/C polymorphism of the CYP11B2 gene in Normotensive subjects and hypertensive patients

Normotensive Hypertensive (n=221) (n=250)

Genotype frequency T/T, n (%) 124 (56.1) 105 (42.0) T/C, n (%) 77 (34.8) 119 (47.6) C/C, n (%) 20 (9.1) 26 (10.4) Figure 1 Plasma renin activity (left panel) and plasma concentra- Allele frequency tions of angiotensin II (middle panel) and aldosterone (right T (%) 73.5 65.8 panel) in 144 male subjects with TT (n ¼ 81), TC (n ¼ 50), and CC C (%) 26.5 34.2 (n ¼ 13) genotypes. *Po0.05, **Po0.01.

Genotype distribution in hypertensive vs normotensive: w2=9.61, P=0.008. Figure 1 depicts the indices of the circulating Allele distribution in hypertensive vs normotensive: w2=6.60, P=0.010. renin–angiotensin–aldosterone system in male NT subjects with TT (n ¼ 81), TC (n ¼ 50), and CC Table 3 Numbers of smokers and nonsmokers in normotensive (n ¼ 13) genotypes. The CC group had lower plasma subjects and hypertensive patients grouped by the CYP11B2 renin activity and lower plasma concentrations of genotype angiotensin II and aldosterone than the TT and TC groups. Genotype

TT TC CC Discussion Normotensive (n=221) Nonsmoker 92 52 16 In the present study, we investigated the association Smoker 32 25 4 between the À344T/C polymorphism in the promo- Hypertensive (n=250) ter region of the human CYP11B2 gene and the Nonsmoker 55 61 13 development of hypertension. Our results indicate Smoker 50 58 13 that the genotype distribution differs between hypertensive and normotensive groups, and that the À344C allele is associated with genetic predis- nificantly different between the two groups. position to develop EH. Although the ratios of habitual drinkers were not A number of studies have suggested the implica- different between EH and NT, smokers were more tion of CYP11B2 gene polymorphism in the patho- frequent in EH than in NT. In the EH group, 41 physiology of cardiovascular diseases. It was first subjects had complications because of other cardio- reported that the À344C allele of CYP11B2 gene vascular diseases. Namely, 37 subjects had present polymorphism was associated with left ventricular or past history of coronary artery disease, five mass in Finnish young adults without clinical heart subjects had past history of cerebral stroke, and diseases.10 The same authors have also reported that one subject had both. the À344C allele was associated with higher systolic blood pressure levels in Finnish men.11 With regard to the mechanism of developing hypertension, it has Genotype and allele frequencies been reported that baroreflex sensitivity is decreased in subjects carrying the C allele.12 However, in the Table 2 shows the frequencies of each genotype and French population, the frequency of the C allele was allele. The distributions of three genotypes were in rather lower in hypertensive patients than in accordance with the Hardy–Weinberg equilibrium normotensive subjects.13 It was also reported that both in the NT and EH groups. However, the À344T/C polymorphism is associated with the risk genotype distribution was significantly different of myocardial infarction in Finland.14 However, between the two groups (w2 ¼ 9.61, P ¼ 0.008). such association was not observed in later studies Namely, the frequency of the C allele was higher undertaken in England and Germany.15,16 In the and the frequency of the T allele was lower in EH Finnish study, the risk of myocardial infarction was than in NT (w2 ¼ 6.60, P ¼ 0.010). In the EH group, increased especially when the C allele coexisted there was no significant difference in the C allele with habitual smoking;14 however, such interaction frequency between the 41 subjects with cardiovas- between CYP1B2 gene polymorphism and smoking cular diseases and the 209 subjects without them was not observed in our study as to the development (0.34 vs 0.34). As shown in Table 3, the ratios of of essential hypertension. habitual smokers were not different among the three The relationship between gene polymorphism genotypes either in the NT or the EH group. and predisposition to cardiovascular diseases may

Journal of Human Hypertension CYP11B2 gene polymorphism in hypertension K Tsukada et al

792 not be always universal among different ethnic cant difference in 24-h ambulatory blood levels groups. In the Japanese, unlike the results of our among the genotypes; however, the nocturnal de- study, two earlier studies have failed to find a cline in blood pressure was greater in CC homo- significant relationship between CYP11B2 zygotes. Moreover, the prevalence of cardiovascular gene polymorphism and hypertension.17,18 Tsujita disease was less in CC than in other genotypes. In et al17 have enrolled a large number of subjects the EH group of our study, the presence or absence randomly selected from the municipal population of cardiovascular complications did not affect the and could not find significant differences in genotype distribution of CYP11B2 gene polymorph- blood pressure between TT, TC, and CC genotypes. ism, although the number of patients with cardio- This result is not necessarily in conflict with vascular diseases is limited. Further large-scale our findings. Young subjects of the study by study is needed to elucidate the relation of CYP11B2 Tsujita et al may possibly develop hypertension gene polymorphism to the incidence of cardiovas- later in their life and systolic hypertension caused cular diseases. by arteriosclerosis in the elderly may not reflect With regard to the plasma aldosterone levels, genetic predisposition to EH. In the present study, Pojoga et al21 have shown that the CC genotype was the subject selection was performed so that such associated with increased plasma aldosterone in ambiguity was excluded, because all subjects in the French patients with EH, while Brand et al22 have EH group had developed hypertension before they reported that plasma aldosterone was not affected by were 50 years old and all subjects in the NT group the genotypes of À344T/C polymorphism in German were older than 50 years. normotensive subjects. In the present study, most On the other hand, Kato et al18 have performed a patients with EH who were already treated pharma- case–control study in which they failed to find any cologically and plasma aldosterone levels were not relation between the gene polymorphisms of the evaluated. In male NT subjects of this study, the renin–angiotensin–aldosterone system and EH. In circulating renin–angiotensin system was sup- the subanalysis of their study, similar criteria to our pressed and the plasma aldosterone level was lower study were applied to select the normotensive and in the CC group than in the TC or TT groups. In hypertensive subjects. Namely, even when the accordance with our results, Davies et al23 have hypertensive patients who had developed hyperten- reported that urinary excretion of aldosterone is less sion before 45 years of age and the normotensive in the CC genotype than in the TC and TT subjects more than 60 years of age were compared, a genotypes. Therefore, it is speculated that the CC significant difference in genotype frequencies was genotype may be related to the development of low- not detected. Discordance of the results of the study renin hypertension; however, the mechanism is not by Kato et al with our results may be attributed to supposed to be brought about by the increased the influence of environmental factors. Most of our transcription of the aldosterone synthetase gene. study subjects lived in Tochigi, a rural to suburban The results of this study revealed the association district, while the participants of the study by Kato of CYP11B2 gene polymorphism with genetic pre- et al had been recruited in Tokyo, a big city. The disposition to hypertension. The frequency of the quality and quantity of stress, which contribute to À344C allele was higher in patients with EH than in the development of hypertension, are supposed to normotensive subjects. However, the cross-sectional be different between urban and rural lives. In studies, including our current one, cannot be addition, people in Tochigi are known to ingest completely free from some bias in selecting the much salt as compared to urban people. According study subjects. The NT subjects of this study were to the national nutritional survey performed in 1999, recruited from the participants of a health check the average salt intake was 13.9 g/day in Tochigi and programme and the EH patients were the outpatients 12.5 g/day in Tokyo. Therefore, salt-induced hyper- of our university hospital. Besides the genetic tension is more likely to occur in our study subjects predisposition to hypertension, it may be possible than in the subjects living in Tokyo. In this aspect, that unexpected difference in genetic background the observed relationship between the C allele of the had existed between the case and control groups of CYP11B2 gene and the development of EH may only this study. Therefore, the results of this study should be applicable to rural Japanese subjects consuming be confirmed by further prospective studies and more salt. Indices of salt intake such as urinary Na linkage analyses in sib-pairs. excretion were not evaluated in our study, and the salt-loading study in Spanish subjects did not demonstrate any relationship between salt sensitiv- ity and CYP11B2 gene polymorphism.19 Therefore, Acknowledgements further salt balance studies are needed to elucidate the implication of the CYP11B2 genotype in salt We thank Ms Yasuko Kawamura, Ms Mika Nomura, sensitivity in Japanese subjects. Ms Jun Chou, Ms. Masako Minato, Ms Machiko Matsubara et al20 have examined the association Sakata and Ms Yasuko Mamada for technological of CYP11B2 gene polymorphism with circadian assistance in executing the study. This study was blood pressure rhythm. They observed no signifi- supported in part by a grant-in-aid for scientific

Journal of Human Hypertension CYP11B2 gene polymorphism in hypertension K Tsukada et al

793 research (10218209) from the Ministry of Education, 12 Ylitalo A et al. Baroreflex sensitivity and variants of Science and Culture of Japan. the renin angiotensin system genes. J Am Col Cardiol 2000; 35: 194–200. 13 Brand E et al. Structural analysis and evaluation of the aldosterone synthase gene in hypertension. Hyperten- sion 1998; 32: 198–204. References 14 Hautanen A et al. Joint effects of an aldosterone synthase (CYP11B2) gene polymorphism and classic 1 White PC. Disorders of aldosterone biosynthesis and risk factors on risk of myocardial infarction. Circula- action. N Engl J Med 1994; 331: 250–258. tion 1999; 100: 2213–2218. 2 Kawamoto T et al. Role of steroid 11 beta-hydroxylase 15 Patel S, Steeds R, Channer K, Samani NJ. Analysis of and steroid 18-hydroxylase in the biosynthesis of promoter region polymorphism in the aldosterone glucocorticoids and mineralocorticoids in humans. synthase gene (CYP11B2) as a risk factor for myocar- Proc Natl Acad Sci USA 1992; 89: 1458–1462. dial infarction. Am J Hypertens 2000; 13: 134–139. 3 Curnow KM et al. The product of the CYP11B2 gene is 16 Hengstenberg C et al. Evaluation of the aldosterone required for aldosterone biosynthesis in the human synthase (CYP11B2) gene polymorphism in patients adrenal cortex. Mol Endocrinol 1991; 5: 1513–1522. with myocardial infarction. Hypertension 2000; 35: 4 White PC, Slutsker L. Haplotype analysis of CYP11B2. 704–709. Endocr Res 1995; 21: 437–442. 17 Tsujita Y et al. Lack of association between genetic 5 Fardella CE et al. in P450c11AS in polymorphism of CYP11B2 and hypertension in Chilean patients with low renin hypertension. J Clin Japanese: the Suita Study. Hypertens Res 2001; 24: Endocrinol Metab 1996; 81: 4347–4351. 105–109. 6 White PC, Hautanen A, Kupari M. Aldosterone 18 Kato N et al. Comprehensive analysis of the renin– synthase (CYP11B2) polymorphisms and cardiovascu- angiotensin gene polymorphisms with relation to lar function. J Steroid Biochem Mol Biol 1999; 69: hypertension in the Japanese. J Hypertens 2000; 18: 409–412. 1025–1032. 7 Joint National Committee on prevention, detection, 19 Poch E et al. Molecular basis of salt sensitivity in evaluation, and treatment of high blood pressure. The human hypertension. Evaluation of renin–angioten- sixth report of the Joint National Committee on sin–aldosterone system gene polymorphisms. Hyper- prevention, detection, evaluation, and treatment of tension 2001; 38: 1204–1209. high blood pressure. Arch Intern Med 1997; 157: 20 Matsubara M et al. Aldosterone synthase gene 2413–2446. (CYP11B2) C-334T polymorphism, ambulatory blood 8 Guidelines Subcommittee: 1999 World Health Organi- pressure and nocturnal decline in blood pressure in zation–International Society of Hypertension Guide- the general Japanese population: the Ohasama Study. lines for the Management of Hypertension. Guidelines J Hypertens 2001; 19: 2179–2184. Subcommittee. J Hypertens 1999; 17: 151–183. 21 Pojoga L et al. Genetic determination of plasma 9 Komiya I et al. Lys(173)Arg and À344T/C variants of aldosterone levels in essential hypertension. Am J CYP11B2 in Japanese patients with low-renin hyper- Hypertens 1998; 11: 856–860. tension. Hypertension 2000; 35: 699–703. 22 Brand E et al. Aldosterone synthase gene (CYP11B2) 10 Kupari M et al. Associations between human aldoster- C-344T polymorphism in Caucasians from the Berlin one synthase (CYP11B2) gene polymorphisms and left Salt-Sensitivity Trial (BeSST). J Hypertens 1999; 17: ventricular size, mass, and function. Circulation 1998; 1563–1567. 97: 569–575. 23 Davies E et al. Aldosterone excretion rate and blood 11 Hautanena A et al. Associations between aldosterone pressure in essential hypertension are related to synthase gene polymorphism and the adrenocortical polymorphic differences in the aldosterone synthase function in males. J Intern Med 1998; 244: 11–18. gene CYP11B2. Hypertension 1999; 33: 703–707.

Journal of Human Hypertension