Journal of Human Hypertension (2010) 24, 345–350 & 2010 Macmillan Publishers Limited All rights reserved 0950-9240/10 $32.00 www.nature.com/jhh ORIGINAL ARTICLE CYP3A5 polymorphism and sensitivity of blood pressure to dietary salt in Japanese men

L Zhang1,2, K Miyaki3, W Wang1 and M Muramatsu1,2 1Department of Epidemiology and Biostatistics, School of Public Health and Family Medicine, Capital Medical University, Youanmenwai, Beijing, PR China; 2Department of Molecular Epidemiology, Medical Research Institute, Tokyo Medical and Dental University, Chiyoda-ku, Tokyo, Japan and 3Department of Preventive Medicine of Cerebrovascular Disease, Keio University School of Medicine, Shinjuku-ku, Tokyo, Japan

The cytochrome P-450 3A5 (CYP3A5) has recently a weak association was detected for diastolic blood been implicated in renal sodium reabsorption and blood pressure (DBP) (*1/*1 vs *3/*3, P ¼ 0.038), which was pressure regulation. The genetic effect of CYP3A5*1 strengthened after adjustment with age and body mass (expressor) and *3 (reduced-expressor) variants on index (*1/*1 vs*3/*3, P ¼ 0.007; *1/*1 vs*1/*3, P ¼ 0.045). blood pressure has been studied in African Americans Significant interactions between the genotype and salt and Caucasians, but not yet in the Asian population. In intake were observed in both SBP (P ¼ 0.046) and DBP this cross-sectional study, 238 Japanese male workers (P ¼ 0.003). SBP and DBP were significantly associated were examined to determine whether CYP3A5 *1*3 with the level of salt intake in*3/*3 (Po0.001) but not affects the blood pressure level, taking daily salt intake in *1 carriers. *1carrier had higher blood pressure into account as a potential gene–environment interac- than *3/*3, but only in those with low salt intake. tion. The A6986G polymorphism was determined by These results suggest that CYP3A5 variants may be a melting curve analysis, and the salt intake level was determinant of salt sensitivity of blood pressure in inferred from spot urine specimen by calculating 24-h Japanese men. urinary sodium excretion. CYP3A5 *1*3 per se had no Journal of Human Hypertension (2010) 24, 345–350; association with systolic blood pressure (SBP) and only doi:10.1038/jhh.2009.74; published online 8 October 2009

Keywords: CYP3A5; polymorphism; salt intake; blood pressure; interaction

Introduction wild-type A allele and the variant G allele express CYP3A5*1 and CYP3A5*3, respectively. Whereas The 3A (CYP3A) enzyme is the *1 allele expresses the wild-type CYP3A5 involved in the metabolism of both endogenous , the *3 allele leads to splicing of an substances and xenobiotics, and accounts for the 1 alternative exon that results in a premature stop metabolism of over 50% of clinically used drugs. codon and reduces CYP3A5 protein expression.7 CYP3A consists of four , CYP3A4, CYP3A43, 2 Thus, CYP3A5*1 and *3 represent the expressor and CYP3A5 and CYP3A7, located on 7q21-7q22.1. The reduced expressor, which are the key determinants CYP3A5 gene is the major class representing about of the activity and expression of CYP3A5. 50% of the hepatic CYP3A content, and is also 3,4 CYP3A5 has recently been shown to regulate highly expressed in the kidney. An extreme blood pressure. CYP3A5 is present in the kidney interindividual and interethnic variability for the 5,6 and converts cortisol and corticosterone to 6b- expression of CYP3A5 has been reported. This hydroxycortisol and 6b-corticosterone, respectively, difference is largely because of the A6986G poly- 5 which leads to sodium absorption and water reten- morphism (rs776746) in intron 3, in which the tion at the renal tubules.8 Thompson et al.8 geno- typed the CYP3A5*1/*3 polymorphism, which is Correspondence: Professor M Muramatsu, Department of Mole- likely to influence salt and water retention and risk cular Epidemiology, Medical Research Institute, Tokyo Medical for salt-sensitive hypertension. Previous publica- and Dental University, 2-3-10 Kandasurugadai, Chiyoda-ku, tions have examined CYP3A5 polymorphisms in Tokyo 101-0062, Japan. E-mail: [email protected] African Americans and Caucasians, and the *1 allele Received 10 March 2009; revised 10 August 2009; accepted 14 was repeatedly shown to be the risk allele for August 2009; published online 8 October 2009 hypertension.9–11 CYP3A5, salt intake and blood pressure L Zhang et al 346 There are differences in the frequencies of this and Cr was assessed by Jaffe’s method. Urinary allele between racial groups.5 Non-Caucasians have sodium excretion (24-h) was calculated by the a higher prevalence of CYP3A5 expression and have formula, 24HUNaClV (mEq dayÀ1) ¼ 21.98 Â X- different risks of diseases that are associated with Na0.392, which was then converted into international the CYP3A5 expressor phenotype. However, there is system of units mmol dayÀ1 and g dayÀ1. A variable no report to our knowledge that determined the of 24HUNaClV after converting units to grams was effect of this variant on blood pressure in the Asian applied to the subsequent statistical analysis. The population. Replicating the result in different ethnic method has been validated to be robust and effective groups is important. Thus, in this study, we in monitoring trends in dietary sodium intake.14 addressed the effect of the CYP3A5 A6986G poly- morphism on blood pressure in Japanese. As the Japanese diet is characterized by a high consump- Genotype determinations tion of salt,12 we considered dietary salt intake as a DNA was extracted from whole blood using a potential modification factor when investigating the classical phenol/chloroform method. CYP3A5 geno- effect of genetic factor on blood pressure. mic sequence (NT_007933) was obtained from the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/). The A6986G poly- Materials and methods morphism (dBSNP: rs776746) in intron 3 was searched and PCR primers and hybridization probes Subjects were designed as follows: forward primer: A cross-sectional study was conducted among 329 50-ACTGCCCTTGCAGCATTTAG-30, reverse primer: unrelated workers in a company in Kanagawa 50-CATACCCCTAGTTGTACGAC-30, anchor probe: Prefecture of Japan, who underwent a health exam- 50 LCRed640-GTTTGGACCACATTACCCTTCATC-30 ination in 2003. Details of the study population were phosphate, sensor probe: 50-CTTTTGTCTTTCAA described elsewhere.13 Owing to the small number of TATCTCTTCCC-30 fluorescein. The A6986G poly- women (n ¼ 37), they were excluded from the morphism was genotyped using PCR, followed by analysis. We also excluded those who are currently melting curve analysis using LightCycler480 (Roche under treatment for hypertension (n ¼ 38), diabetes Diagnostics, Penzberg, Germany). Briefly, PCR was (n ¼ 8) and hyperlipidemia (n ¼ 8). Accordingly, a carried out with 0.05 mmol lÀ1 upstream primer total of 238 healthy Japanese men were enrolled in and 0.5 mmol lÀ1 downstream primer in a reaction the genotype and phenotype association study and mixture containing 0.2 mmol lÀ1 anchor probe, in the following interactive studies. The study was 0.2 mmol lÀ1 sensor probe, 17.5 ng of dried down approved by the Ethical Committees of the Tokyo DNA, 3.0 mmol lÀ1 MgCl2, 0.5 ml10Â PCR buffer, Medical and Dental University, and all participants 0.2 mmol lÀ1 each dNTP and 0.5 U Faststart DNA provided written consent. Polymerase (Roche Diagnostics) in a total volume of 5 ml. The cycling programme consisted of 5 min of initial denaturation at 95 1C and 55 cycles of Methods denaturation at 95 1C for 10 s, annealing at 60 1C for Details of medical history were obtained by inter- 10 s and extension at 72 1C for 10 s. After completion view. Height and weight were measured, and body of PCR, melting curve analysis was carried out by mass index (BMI) was calculated as weight in heating the mixture at 95 1C for 60 s and then kilograms divided by height in metres squared. maintaining it at 40 1C for 60 s. The plate was then Brachial blood pressure was measured using a heated from 40 1Cto801C by the gradient of 0.1 1C posterior wall velocity/ankle-brachial index device per second. Melting curve data were collected and (Nippon Colin, Aichi, Japan), after subjects rested classified using LightCycler genotype software; quietly for at least 10 min in a supine position. subjects with one lower Tm curve was identified Blood pressure was automatically measured twice as the AA (*1/*1) genotype, those with one higher and the average mean value was recorded. This Tm curve was identified as the GG (*3/*3) genotype device was used as it is approved by the US Food and those with two curves were identified as AG and Drug Administration as VP-2000/1000. Only (*1/*3) genotypes. data for blood pressure were used for the study. The daily sodium intake of individuals was assessed by using a method to estimate 24-h urinary Statistical analyses sodium excretion from a spot urine sample. Partici- Hardy–Weinberg equilibrium for the distribution of pants were asked to urinate and empty their bladder genotype was determined by the w2-test. Analysis of completely and a second morning ‘spot’ urine variance was used to test differences of systolic specimen was collected while they underwent the blood pressure (SBP), diastolic blood pressure (DBP) physical examination. All specimens were refriger- and other continuous variables according to geno- ated at 4 1C within 24 h and frozen at À20 1C within types. Analysis of covariance was used to analyse 7 days. The sodium concentration of the spot urine the interactive effects of genotype, with estimated sample was assessed by emission flame photometry 24HUNaClV on blood pressure adjusted for age

Journal of Human Hypertension CYP3A5, salt intake and blood pressure L Zhang et al 347 and BMI as continuous variables, in which an inter- (Po0.001) and DBP (Po0.001). 24HUNaClV did action term, 24HUNaClV  CYP3A5*1/*3 genotype not associate with either SBP or DBP. (*3/*3 ¼ 1; *1 carriers ¼ 0), was included. As the Figure 1 shows the correlations of 24HUNaClV expectation that CYP3A5*1 allele carriers have a with SBP and DBP according to the CYP3A5 higher CYP3A5 enzyme activity, we combined genotype using the general linear model. The slopes CYP3A5*1/*1 and CYP3A5*1/*3 genotypes as one of the relationship between 24HUNaClV and blood group for the analyses. A general linear model was pressure significantly differed between the two applied to compare blood pressure levels in differ- genotype groups (Po0.05). *3/*3 subjects had ent genotypes according to the 24HUNaClV level. significantly increasing slopes of blood pressure All probability values presented were for two-tailed with increasing 24HUNaClV (Figure 1a, SBP: test and the values of Po0.05 were considered to b ¼ 1.265; Po0.001; Figure 1b, DBP: b ¼ 0.912; indicate statistical significance. Statistical analyses P ¼ 0.001), whereas the slopes were not significant were carried out using the Statistical Package of in *1 carriers (Figure 1a, SBP: b ¼À0.178; P ¼ 0.463; Social Science for Windows version 13.0 (SPSS Inc., Figure 1b, DBP: b ¼À0.466; P ¼ 0.340). Chicago, IL, USA). Next, the subjects were divided into high and low salt intake groups according to the median of 24HUNaClV. The ranges of 24HUNaClV in low and Results high groups were 3.12–9.91 g and 9.98–20.71 g, respectively. The differences in blood pressure The frequencies of CYP3A5 *3/*3,*1/*3 and *1/*1 between two 24HUNaClV levels according to the genotypes were 43.3, 45.8 and 10.9%, respectively. CYP3A5 genotype are shown in Figure 2. In the low The frequency of the *1 allele was 33.8%. The 24HUNaClV level, DBP was higher in the *1 carrier genotyping results were consistent with the Hardy– than in the *3/*3 genotype (P ¼ 0.006), the same Weinberg equilibrium (P ¼ 0.722). trend of SBP was detected (P ¼ 0.095), whereas the The characteristics of subjects in different geno- difference was not detected in the high 24HUNaClV type groups are shown in Table 1. There were no level. There was difference between two 24HU- significant differences in mean age, BMI, spot urine NaClV levels in subjects with the *3/*3 genotype Na excretion, spot urine creatinine excretion, esti- (DBP: P ¼ 0.024, SBP: P ¼ 0.064). This result was mated 24HUNaV and 24HUNaClV among genotype consistent with the former results analysed using groups. The CYP3A5 genotype associated with DBP continuous data. (Po0.05). The *1/*1 group has the highest DBP and exceeded that of the *3/*3 group by 7 mm Hg (P ¼ 0.038). The difference was strengthened after Discussion adjustment with age and BMI (*1/*1 vs*3/*3, P ¼ 0.007; *1/*1 vs*1/*3, P ¼ 0.045). SBP was not CYP3A5*1 and *3 have a significant difference in different among genotypes. their allele frequencies among different ethnic We then sought a potential interaction between groups. The CYP3A5*1 allele varies from over 45% the CYP3A5 genotype and 24HUNaClV on blood in those of African descent to 8–15% in whites, and pressure levels using analysis of covariance 23–40% in Asians.15 This difference is thought to be (Table 2). Significant interactive effects between an evolutional consequence for the need to retain the CYP3A5*1/*3 genotype and 24HUNaClV on water under different climates.8 both SBP (P ¼ 0.046) and DBP (P ¼ 0.003) were The association of CYP3A5 alleles with blood detected. There were associations of CYP3A5*1/*3 pressure has been initially studied in Blacks, in genotypes with SBP (P ¼ 0.027) and DBP (P ¼ 0.001). which the *1 allele is abundant. Givens et al.9 found Age and BMI were also associated with SBP a significant association between CYP3A5*1/*3 and

Table 1 Clinical characteristics according to CYP3A5*1/*3 genotype groups

Characteristics GG (*3/*3) GA (*3/*1) AA (*1/*1) P-value

Genotype frequency % (n) 43.3 (103) 45.8 (109) 10.9 (26) 0.722 Age (years) 45.5±12.1 45.6±11.7 49.2±9.9 0.324 BMI (kg mÀ2) 23.5±3.5 23.2±3.1 22.8±3.9 0.625 SBP (mm Hg) 133.5±16.4 134.8±18.0 138.8±20.6 0.388 DBP (mm Hg) 80.1±11.6* 81.7±12.3 87.0±15.2 0.038 Spot urine Na excretion (mmol dayÀ1) 167.2±45.8 158.2±61.0 140.8±63.0 0.085 Spot urine creatinine excretion (mg dayÀ1) 1388.4±235.9 1336.6±219.0 1353.7±190.8 0.241 Estimated 24HUNaV (mmol dayÀ1) 177.3±46.2 172.2±49.1 173.5±61.3 0.750 Estimated 24HUNaClV (g) 10.3±2.7 10.0±2.9 10.1±3.6 0.750

Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; SBP, systolic blood pressure; 24HUNaV, 24-h urinary sodium excretion value; 24HUNaCl, calculated 24-h urinary NaCl excretion value. *Post-test used Bonferroni, Po0.05. Values are mean±s.d. or percentage. P values are calculated from analysis of variance or Pearson’s w2-test.

Journal of Human Hypertension CYP3A5, salt intake and blood pressure L Zhang et al 348 Table 2 Multiple linear regression analysis of the interactive effect of CYP3A5*1/*3 genotype and 24HUNaClV on blood pressure adjusted by age and BMI

Parameters SBP P-value DBP P-value b (95% CI) b (95% CI)

*3/*3 vs *1 carrier À17.1 (À32.2 to À2.0) 0.027 À16.7 (À26.3 to À7.0) 0.001 *3/*3  24HUNaClV vs 1.5 (0.02À2.9) 0.046 1.4 (0.5À2.3) 0.003 *1 carrier  24HUNaClV 24HUNaClV 0.05 (À0.8 to 0.9) 0.914 À0.3 (À0.8 to 0.3) 0.385 Age 0.6 (0.4–0.7) o0.001 0.5 (0.4–0.6) o0.001 BMI 1.3(0.6–1.9) o0.001 1.1(0.7–1.5) o0.001

Abbreviations: BMI, body mass index; CI, confidence interval; DBP, diastolic blood pressure; SBP, systolic blood pressure; 24HUNaCl, calculated 24-h urinary NaCl excretion value. The CYP3A5*1 carriers is the reference group. The interaction between the 24HUNaClV and CYP3A5*1 carriers is the reference group. b values and 95% CI were derived from multiple regression analysis models adjusted for age and BMI. CYP3A5*1/*3 polymorphism was assigned as: CYP3A5*3/*3 ¼ 1; CYP3A5*1 carriers ¼ 0.

SBP among healthy African-American adults. The elevating effect was more prominent in *3*3 sub- SBP of the *1/*1 group exceeded that of the jects than in *1 carriers. In two previous reports, *1 homozygous *3/*3 group by 19.3 mm Hg. They also carriers have been shown to express more enzyme and found that *1/*3 individuals had an eightfold higher to be salt sensitive than *3*3 subjects.11,16 Our results kidney CYP3A5 content and an 18-fold higher point to an opposite direction. Potential explanations CYP3A5 catalytic activity than *3/*3 individuals. for this discrepancy are as follows: two former studies Bochud et al.16 also studied the population in were conducted in Blacks, not in Asians, and one Seychelles and found that age and urinary sodium possibility is the difference in diet. The salt intake in excretion were significant effect modifiers of the the Black population may be low because of more association between the CYP3A5*1 allele and vegetables and less processed food.17 In such a case, the ambulatory blood pressure. As with Caucasians, a CYP3A5 expressor may become more sensitive to salt. study conducted in Finland showed that the *1/*3 In contrast, Japanese diet is abundant in salt and the genotype had a significantly higher prevalence of dietary salt intake usually exceeds 10g dayÀ1 (the hypertension than did the *3/*3 genotype.10 median of our study group was 9.98g dayÀ1). Under To our knowledge, this genetic effect has not yet such a habitual high-salt diet, the blood pressure- been addressed in Asians. Our study using appar- elevating effect of CYP3A5 in *1 carrier might have ently healthy Japanese men showed that the already saturated. Thus, in only the CYP3A5 low CYP3A5*1/*3 polymorphism associated with blood expressor (*3*3) might the salt sensitivity become pressure, and that there was an interaction with salt overt. To support this notion, it was reported that, intake. The *1/*1 genotype had a higher DBP than although *1 carriers express large amounts of CYP3A5 did the *3/*3 genotype, indicating that the genetic mRNA and protein, *3 carriers also produce small effect is more prominent on DBP in this population. amounts of normally spliced CYP3A5 mRNA and When salt intake was taken into account, the resulting protein.18 On the basis of studies in a interactive effect of genotype and salt intake was spontaneous hypertensive rat model, excessive intrar- significant on both SBP and DBP. (SBP: Po0.046, enal conversion of cortisol to 6b-hydroxycorticosterone DBP: Po0.003). As these effects remained significant by CYP3A isoenzymes is thought to mediate tubular after being adjusted for age and BMI, CYP3A5*1/*3 sodium reabsorption, which leads to blood pressure seems to be an independent modifier of individual increase.19 On the other hand, functional studies in salt sensitivity. These results are generally in the epithelial kidney cells have indicated a role of CYP3A5 same line with previous reports that the *1 allele is a in 6b-hydroxysteroid formation, which may, in con- risk factor for hypertension. trast, counteract with the mineralocorticoid receptor In our study, DBP rather than SBP was strongly from being activated. The presence of a hypothetical associated with the *1 allele. SBP is predominately second enzyme might provide an additional pathway determined by stroke volume, and thus, in theory, to the mineralocorticoid receptor from being occupied should be more influenced by CYP3A enzyme by cortisol.20 Such a hypothesis could potentially activity. However, in our study population, DBP explain the diminished salt sensitivity of blood gave a positive association, but SBP only gave a pressure in *1 carriers compared with *3*3 subjects weak trend with the increased amount of *1 allele as shown in our results. (0 allele: 133.5±16.4, 1 allele: 134.8±18.0 and 2 Hypertensive subjects were excluded from this alleles: 138.8±20.6 mm Hg). study, as dietary salt restriction and using antihy- Further investigation taking the environmental pertensive drugs might affect the results. It is factor of salt intake into account led us to find possible that the AA genotype might be a stronger different reactions of the CYP3A5 genotypes on predictor of higher SBP and DBP in hypertensive blood pressure. To our surprise, the blood pressure- subjects. Salt sensitivity of blood pressure is more

Journal of Human Hypertension CYP3A5, salt intake and blood pressure L Zhang et al 349 Association between 24HuNaClV and adjusted SBP Comparison of SBP and DBP between low and high 24HUNaClV levels according to CYP3A5 genotype according to CYP3A5*1/*3 genotype 200.0 *3/*3 *3/*3 82.5 High 24HUNaClV *1 carrier *1 carriers 82.5 *3/*3 DBP 180.0 P for interaction=0.046 77.5 *1 carriers Low 24HUNaClV 83.0

136.3 160.0 High 24HUNaClV 135.9 SBP 130.4 Low 24HUNaClV 140.0 135.4

60 90 120 150 Blood Pressure 120.0 Adjusted SBP (mmHg) Figure 2 Comparison of systolic blood pressure (SBP) and y *3/*3 = 120.4 + 1.265x diastolic blood pressure (DBP) between low and high 24HUNaClV w 100.0 P for slope < 0.001 levels according to the CYP3A5*1/*3 genotype. The difference in y *1 carriers =137.4 – 0.178x blood pressure between the CYP3A5*3/*3 group and the *1 P for slope = 0.463 carriers group in the low 24HUNaClV level (P ¼ 0.095 in SBP; P ¼ 0.006 in DBP). wwThe difference in blood pressure between two 80.0 24HUNaClV levels in the *3/*3 group (P ¼ 0.064 in SBP; P ¼ 0.024 3.00 6.00 9.00 12.00 15.00 18.00 21.00 in DBP). 24HuNaClV(g)

Association between 24HuNaClV and adjusted DBP involved. Recent studies have identified that there according to CYP3A5 genotype is gene–gene–environment interaction between 140.0 23 *3/*3 CYP3A5, ABCA1 and salt intake. More biophysio- *1 carriers logical pathways may be involved in the homo- 120.0 P for interaction=0.003 *3/*3 eostasis of sodium retention and blood pressure *1 carriers regulation, which acts along with the increase in salt intake. It is conceivable that there might be genetic 100.0 or environmental factors still undiscovered, which could change the direction of the effect of a certain variation in different ethnic groups. Further studies 80.0 are definitely warranted to provide a more compre- hensive explanation for the basis of our observation. In conclusion, our study using Japanese men 60.0

Adjusted DBP (mmHg) showed that CYP3A5*1*3 is a modifier of blood y *3/*3 = 70.6 + 0.912x pressure by changing salt sensitivity. Although the P for slope = 0.001 CYP3A5*1 carrier tends to have higher blood 40.0 y *1 carriers = 87.4 – 0.466x P for slope = 0.340 pressure, the blood pressure of CYP3A5*3 homo- zygous is more vulnerable to salt intake. The 20.0 physiological role of these variants needs to be 3.00 6.00 9.00 12.00 15.00 18.00 21.00 further studied under the context of wider genetic 24HuNaClV(g) and environmental influences. Figure 1 Interaction between the CYP3A5*3/*3 group and the *1 carriers group and the effect of 24HUNaClV level on systolic blood pressure (SBP) (a) and diastolic blood pressure (DBP) (b). Regression line and probability values for slopes in different What is known about this topic genotype groups are indicated. K CYP3A5*1/*3 polymorphism is the key determinant of the expression of CYP3A5, which has an extreme interindividual and interethnic variability. frequent in hypertensive than in normotensive K 21 CYP3A5*1/*3 polymorphism influence salt and water subjects. It is estimated that 26% of Americans retention and is one of the candidate genes for salt-sensitive with normal blood pressure and about 58% of those hypertension. with high blood pressure may be salt sensitive.22 This could also be true for salt sensitivity of blood What this study adds K CYP3A5*1 allele is strongly associated with diastolic blood pressure in hypertensive subjects. Further study pressure rather than systolic blood pressure in Japanese is needed to clarify the effect of the CYP3A5 males. polymorphism on salt sensitivity in primary K CYP3A5*1*3 gene polymorphism is a modifier of blood hypertensives. pressure by changing salt sensitivity. K Although CYP3A5*1 carrier tends to have higher blood It is obvious that salt sensitivity of blood pressure pressure, the blood pressure of CYP3A5*3 homozygous is not accounted for by a single gene, but rather are more vulnerable to salt intake. various genetic and environmental factors are

Journal of Human Hypertension CYP3A5, salt intake and blood pressure L Zhang et al 350 Acknowledgements 12 Shimazu T, Kuriyama S, Hozawa A, Ohmori K, Sato Y, Nakaya N et al. Dietary patterns and cardiovascular This work was supported by the National Natural disease mortality in Japan: a prospective cohort study. Science Foundation of China 30800949. Int J Epidemiol 2007; 36: 600–609. 13 Zhang L, Miyaki K, Araki J, Song Y, Kimura T, Omae K et al. Interaction of angiotensin I-converting enzyme References insertion-deletion polymorphism and daily salt intake influences hypertension in Japanese men. Hypertens 1 Paulussen A, Lavrijsen K, Bohets H, Hendrickx J, Res 2006; 29: 751–758. Verhasselt P, Luyten W et al. Two linked mutations in 14 Kawasaki T, Itoh K, Uezono K, Sasaki H. A simple transcriptional regulatory elements of the CYP3A5 method for estimating 24 h urinary sodium and gene constitute the major genetic determinant of potassium excretion from second morning voiding polymorphic activity in humans. Pharmacogenetics urine specimen in adults. Clin Exp Pharmacol Physiol 2000; 10: 415–424. 1993; 20: 7–14. 2 Gellner K, Eiselt R, Hustert E, Arnold H, Koch I, Haberl 15 Chowbay B, Zhou S, Lee EJ. An interethnic compar- M et al. Genomic organization of the human CYP3A ison of polymorphisms of the genes encoding locus: identification of a new, inducible CYP3A gene. drug-metabolizing enzymes and drug transporters: Pharmacogenetics 2001; 11: 111–121. experience in Singapore. Drug Metab Rev 2005; 37: 3 Haehner BD, Gorski JC, Vandenbranden M, Wrighton 327–378. SA, Janardan SK, Watkins PB et al. Bimodal distribu- 16 Bochud M, Eap CB, Elston RC, Bovet P, Maillard M, tion of renal cytochrome P450 3A activity in humans. Schild L et al. Association of CYP3A5 genotypes with Mol Pharmacol 1996; 50: 52–59. blood pressure and renal function in African families. 4 Koch I, Weil R, Wolbold R, Brockmoller J, Hustert E, J Hypertens 2006; 24: 923–929. Burk O et al. Interindividual variability and tissue- 17 Charlton KE, Steyn K, Levitt NS, Zulu JV, Jonathan D, specificity in the expression of cytochrome P450 3A Veldman FJ et al. Diet and blood pressure in South mRNA. Drug Metab Dispos 2002; 30: 1108–1114. Africa: intake of foods containing sodium, potassium, 5 Hustert E, Haberl M, Burk O, Wolbold R, He YQ, calcium, and magnesium in three ethnic groups. Klein K et al. The genetic determinants of the Nutrition 2005; 21: 39–50. CYP3A5 polymorphism. Pharmacogenetics 2001; 11: 18 Kuehl P, Zhang J, Lin Y, Lamba J, Assem M, Schuetz J 773–779. et al. Sequence diversity in CYP3A promoters and 6 Thompson EE, Kuttab-Boulos H, Yang L, Roe BA, characterization of the genetic basis of polymorphic Di Rienzo A. Sequence diversity and haplotype CYP3A5 expression. Nat Genet 2001; 27: 383–391. structure at the human CYP3A cluster. Pharmaco- 19 Watlington CO, Kramer LB, Schuetz EG, Zilai J, Grogan genomics J 2006; 6: 105–114. WM, Guzelian P et al. Corticosterone 6 beta-hydro- 7 Beeks E, Kessels AG, Kroon AA, van der Klauw MM, xylation correlates with blood pressure in sponta- de Leeuw PW. Genetic predisposition to salt- neously hypertensive rats. Am J Physiol 1992; 262: sensitivity: a systematic review. J Hypertens 2004; 22: F927–F931. 1243–1249. 20 Morris DJ, Latif SA, Rokaw MD, Watlington CO, 8 Thompson EE, Kuttab-Boulos H, Witonsky D, Yang L, Johnson JP. A second enzyme protecting mineralocor- Roe BA, Di Rienzo A. CYP3A variation and the ticoid receptors from glucocorticoid occupancy. Am J evolution of salt-sensitivity variants. Am J Hum Genet Physiol 1998; 274: C1245–C1252. 2004; 75: 1059–1069. 21 Weinberger MH, Fineberg NS, Fineberg SE, Weinberger 9 Givens RC, Lin YS, Dowling AL, Thummel KE, M. Salt sensitivity, pulse pressure, and death in normal Lamba JK, Schuetz EG et al. CYP3A5 genotype predicts and hypertensive humans. Hypertension 2001; 37: renal CYP3A activity and blood pressure in healthy 429–432. adults. J Appl Physiol 2003; 95: 1297–1300. 22 Weinberger MH, Miller JZ, Luft FC, Grim CE, Fineberg 10 Kivisto KT, Niemi M, Schaeffeler E, Pitkala K, Tilvis R, NS. Definitions and characteristics of sodium sensi- Fromm MF et al. CYP3A5 genotype is associated with tivity and blood pressure resistance. Hypertension diagnosis of hypertension in elderly patients: data 1986; 8: II127–II134. from the DEBATE Study. Am J Pharmacogenomics 23 Eap CB, Bochud M, Elston RC, Bovet P, Maillard MP, 2005; 5: 191–195. Nussberger J et al. CYP3A5 and ABCB1 genes 11 Ho H, Pinto A, Hall SD, Flockhart DA, Li L, Skaar TC influence blood pressure and response to treatment, et al. Association between the CYP3A5 genotype and and their effect is modified by salt. Hypertension 2007; blood pressure. Hypertension 2005; 45: 294–298. 49: 1007–1014.

Journal of Human Hypertension