| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2007;49:113.)
© 2007 American Heart Association, Inc.
Original Articles |
From the British Heart Foundation Glasgow Cardiovascular Research Centre (M.B., S.M.M., E.C.F., C.D.H., D.M.W., N.J.R.B., M.C.I., R.F., A.F.D., J.M.C.C., E.D.), University of Glasgow, Glasgow; Clinical Pharmacology and the Cambridge Institute of Medical Research (M.B., D.C.), University of Cambridge, Addenbrookes Hospital, Cambridge; the Department of Cardiovascular Sciences (N.J.S.), University of Leicester, Glenfield Hospital, Leicester; Clinical Pharmacology and Barts and the London Genome Centre (M.C., P.B.M.), William Harvey Research Institute Barts and the London School of Medicine, Charterhouse Square, London; the Nuffield Department of Clinical Medicine and Department of Cardiovascular Medicine (M.F.), University of Oxford, Wellcome Trust Centre for Human Genetics, Oxford; and Medicine and Therapeutics (J.W.), Aberdeen Royal Infirmary, Aberdeen, United Kingdom.
Correspondence to John M.C. Connell, Medical Research Council Blood Pressure Group, Division of Cardiovascular and Medical Sciences, British Heart Foundation Glasgow Cardiovascular Research Centre, 126 University Place, Glasgow G12 8TA, United Kingdom. E-mail J.Connell{at}clinmed.gla.ac.uk
| Abstract |
|---|
|
|
|---|
Key Words: hypertension aldosterone synthase 11ß-hydroxylase polymorphisms transcriptional regulation
| Introduction |
|---|
|
|
|---|
15% of unselected patients with hypertension have a raised aldosterone/renin ratio, suggesting that altered regulation of aldosterone synthesis or release may be a common factor in essential hypertension.26
The late stages in aldosterone synthesis that occur in the zona glomerulosa of the adrenal cortex are key in the control of production of the hormone. The 3 terminal steps are a sequential 11ß-hydroxylation, 18-hydroxylation, and 18-oxidation of the precursor 11-deoxycorticosterone, all catalyzed by the enzyme aldosterone synthase. This enzyme is encoded by the CYP11B2 gene, the major regulator of which is angiotensin II. In the zona fasciculata, cortisol is synthesized by a similar pathway, but the final stage is an 11ß-hydroxylation of 11-deoxycortisol (S) by the highly homologous enzyme 11ß-hydroxylase. The CYP11B1 gene, which is controlled by adrenocorticotropic hormone (corticotropin [ACTH]), encodes 11ß-hydroxylase. CYP11B2 and CYP11B1 are situated
40 kb apart on chromosome 8 in humans. They share 95% sequence similarity, with the main differences being in the 5' regions that are responsible for the differences in their transcriptional regulation.7
Two common variants have been described previously in the CYP11B2 gene.8 The first is a single nucleotide substitution from cytosine to thymidine in the 5' promoter region (344 C/T), which disrupts a putative binding site for the steroidogenic factor 1. The second is an intron conversion in intron 2. This variant exists in 2 alternate forms, either the wild-type (Wt) or the conversion (Conv), where part of intron 2 is replaced by the corresponding intron of the adjacent CYP11B1 gene. These variants are in tight linkage disequilibrium (LD) so that the common haplotypes are T/Conv (38%), C/Wt (45%), and T/Wt (16%).9 The T allele at 344 and the Conv allele in intron 2 have been associated previously with hypertension, hypertension with a raised aldosterone/renin ratio, as well as increased levels of aldosterone.912 However, it should be mentioned that these are not consistent findings, with some studies not seeing any association and others observing an association with the C allele.1316
Interestingly, one consistent reported phenotypic association with the T/Conv haplotype is impaired 11ß-hydroxylase efficiency, indicated by raised ACTH-stimulated levels of the 11-deoxysteroids, 11-deoxycorticosterone and S (which are themselves substrates for the 11ß-hydroxylase enzyme).17,18 The molecular mechanism that results in this phenotype is unclear; in vitro reporter gene studies of the 344 C/T single nucleotide polymorphism (SNP) show no detectable effect on gene transcription, suggesting that this is not the causative variant.19 Ganapathipillai et al20 reported that the same biochemical phenotype was associated with a CYP11B1/B2 haplotype that included the 344 T variant, and it is likely that the 344 C/T SNP is in LD with causative variants in CYP11B1 that could account for this phenotype. We have reported previously that a SNP in exon 1 (Leu75Leu) of CYP11B1 was strongly associated with the phenotype,21 but this conservative polymorphism cannot, itself, account for the biochemical change. Accordingly, the aims of the present study were to investigate in detail the pattern of variation across the CYP11B1/CYP11B2 locus to define the precise LD with the 344 C/T and intron Conv variants. Investigation of the association of 11ß-hydroxylase efficiency with novel CYP11B1 SNPs and also analysis of their effects in vitro was performed to determine the genetic variation responsible for the observed impaired 11ß-hydroxylation.
| Methods |
|---|
|
|
|---|
A large hypertensive study population was derived from the Medical Research Council British Genetics of Hypertension (BRIGHT) Study. The BRIGHT Study is a large multicenter investigation into the genetic basis of hypertension, composed of >2500 affected sibling pairs. Hypertensive subjects were recruited from primary care with entry to the study being dependent on a pretreatment blood pressure of >145/95 mm Hg (mean of 3 readings). Details of the methods for blood pressure measurements and other demographic characteristics can be found at www.brightstudy.ac.uk. It was aimed to exclude subjects with a body mass index >30. Full details of the recruitment strategy have been published elsewhere.23 For the present investigation, we selected a random subgroup of 512 unrelated hypertensive subjects. Full demographic information (as medians and interquartile ranges) on the subjects included is shown in Table 1.
|
Ethical approval for the study was granted by the local ethics committees of the partner institutes, and fully informed written consent from all of the participants was obtained. All of the procedures were in accordance with institutional guidelines.
Polymorphism Detection
PCR
The nucleotide sequences for human CYP11B1 (accession numbers: D10169, M32863, X85218, M32878, and M32879) and CYP11B2 (accession numbers: D10170, D13752, and M32881) were obtained from the Entrez Nucleotide database. All of the PCRs were performed using Expand High Fidelity PCR system with proofreading capabilities, according to the manufacturers protocol (Roche Diagnostics). The amplification of CYP11B1 was performed as 2 reactions: from 2 kb of the 5' region to the end of exon 9 (
7.5 kb) and from the 3' region, downstream of the coding region (
750 bp). The amplification of CYP11B2 was split into 3 separate PCRs: (1) from the 5' region to intron 4 (
5 kb); (2) from exon 5 to exon 9 (
3 kb); and (3) the 3' region, downstream of the coding region (
1.5 kb). The sequences of the primers (MWG-Biotech) used for these reactions is available in an online supplement at http://hyper. ahajournals.org.
Sequencing
Automated sequencing of the entire region was performed using the ABI Big Dye Terminator v3.1 Cycle Sequencing kit (PE Applied Biosystems) and the ABI3730 Sequencer.
Haplotype Analysis
A Bayesian statistical method implemented in PHASE version 2.1 was used to construct phased haplotypes from phase-unknown genotype data.24 LD between all of the possible pairs of loci was calculated using EM estimation of haplotype frequencies and LD calculation, available online at http://request.mdacc.tmc.edu/
qhuang/Software/pub.htm.
Genotyping of BRIGHT Study for 1889 G/T and 1859 A/G
A subset of the BRIGHT Study population (n=512) was genotyped for the 1889 G/T and 1859 A/G SNPs using a nested PCR method to increase specificity, followed by sequencing. In both reactions, a heat-activated thermostart Taq polymerase (Abgene) was used. Please see the online supplement for the sequences of the primers used in both reactions.
Urinary Corticosteroid Metabolite Measurements
Urine samples (24-hour) were collected in plain plastic containers without preservative. Aliquots of urine were stored at 20°C. Excretion rates of steroid metabolites were measured by gas chromatography/mass spectrometry using the method of Shackleton25 with minor modifications. The ratio of tetrahydrodeoxycortisol (THS)/total cortisol (total F) was used as an index of 11ß-hydroxylase activity.
In Vitro Expression Studies
Site-Directed Mutagenesis
The 5' regulatory region (1937bp) of CYP11B1 was cloned into the pGL3 basic vector (Promega Corp). The CYP11B1 1889 G/T and 1859 A/G SNPs were constructed in vitro using this pGL3/B1 5'regulatory region construct. Site-directed mutagenesis was performed using the Quick-Change site-directed mutagenesis kit (Stratagene Ltd) following the protocol described previously.26 Please see the online supplement for the sequences of the primers used. The entire insert in each case was sequenced to confirm the presence of the desired mutations and that no unwanted base changes had been incorporated.
Cell Culture and Transfection
Y-1 mouse adrenal cells (American Type Culture Collection) were cultured at 37°C in 5% CO2 in F-12K Nutrient Mixture, Kaighns Modification (GIBCO) supplemented with 15% horse serum (GIBCO), 2.5% fetal bovine serum (Sigma-Aldrich), and 1% penicillin/streptomycin (GIBCO). Cells were plated at 3.5x105 cells per well in 6-well plates and incubated for 24 hours before transfection. Transfection was carried out using 6 µL of Fugene (Roche Diagnostics) and 1 µg of Wt or mutant plasmid in MEM/Eagles medium(2 ml) for 42 hours. The pSV-ß-galactosidase control vector (0.04 µg; Promega Corp) was added as a marker of transfection efficiency. After transfection, the cells were treated with agonists (forskolin 10 µmol/L or ACTH 1 µmol/L) for 6 hours. The cells were then lysed and luciferase activity measured according to the manufacturers protocols (Promega Corp). Protein (BCA assay, Pierce Biotechnology) and ß-galactosidase activity (Promega Corp) measurements were used to determine cell number and transformation efficiency, and the luciferase results were corrected appropriately.
Statistics
A
2 test was used to determine whether the allele frequencies of 1889 G/T and 1859 A/G within the BRIGHT subgroup were in Hardy-Weinberg equilibrium. Steroid excretion data did not follow a normal distribution and so were compared by nonparametric methods (Mann-Whitney test). The in vitro expression results for each construct were compared with the appropriate control using a 2-sample t test. In all instances, a P value of <0.05 was accepted as significant.
| Results |
|---|
|
|
|---|
|
CYP11B2 Polymorphisms
Sixty-four variants were identified in CYP11B2 that are in LD with the 344 C/T and intron Conv variants (Figure 1). Most of these changes were part of the intron Conv variant in intron 2. Of the remainder, 3 single base substitutions were found in exons 3 and 6. Two of these are synonymous changes with no effect on amino acid sequence: Phe168Phe and Arg374Arg. These mutations have been described previously by the National Center for Biotechnology Information, accessible on the dbSNP website (http://ncbi.nlm.nih.gov/SNP), with the identification numbers rs4546 and rs4538, respectively.
The third, at codon 173 (rs4539), does cause an amino acid change from Lys to Arg but has been shown to have no effect on aldosterone synthase activity in vitro.2728 Seven single base substitutions were also identified within the 5' regions immediately upstream of the CYP11B2 gene.
CYP11B1 Polymorphisms
Nineteen variants were identified in CYP11B1 that show association with the 344 C/T and intron Conv variants in CYP11B2 (Figure 1). As with CYP11B2, most of these were within intronic regions (introns 2 through 6 and 8). Two SNPs that were identified in exons 1 (Leu75Leu) and 2 (Asp82Asp) have been described previously (National Center for Biotechnology Information rs6410 and rs5283, respectively). Two novel SNPs were identified within the 5'regulatory region at positions 1889 (G/T) and 1859 (A/G) upstream of the ATG start codon. These SNPs are in LD with the T/Conv haplotype in CYP11B2 (Figure 2).
|
Because of their potential to alter CYP11B1 gene transcription, the 1889 G/T and 1859 A/G SNPs were identified as key functional candidates to account for the impaired 11ß-hydroxylase efficiency associated previously with the 344 T and the intron 2 Conv alleles.
Haplotype Analysis
Haplotype analysis revealed 9 haplotypes that have frequencies >1% and describe 75% of chromosomes in the data set. Four of these common haplotypes each have a frequency of >5% and, in combination, account for 68% of chromosomes in the data set (Figure 2). These haplotypes are conserved across the entire region, confirming that there is strong LD across the locus, with a D' value of
0.75 to 1.0 and an r2 value of
0.5 to 1.0.
Genotyping of BRIGHT Study for 1889 G/T and 1859 A/G
To determine whether the 1889 G/T and 1859 A/G SNPs associate with 11ß-hydroxylase efficiency, their frequency and phenotypic associations were investigated in the hypertensive BRIGHT population.
Demographic Data for BRIGHT Subgroup
Blood pressure and other relevant demographic data are shown in Table 1; the population was significantly hypertensive but not significantly overweight. The only significant difference in characteristics of subject groups separated according to the 1889 (GG versus TT P=0.015) or 1859 (AA versus GG P=0.013) SNPs was with waist/hip ratio.
Genetic Analysis of CYP11B1
The frequencies of the 1889 and 1859 SNPs are illustrated in Table 2 (1889 G: 0.53, T: 0.47; 1859 A: 0.5, G: 0.5). The alleles of both 1889 G/T and 1859 A/G were in Hardy-Weinberg equilibrium with
2 values of 0.91 and 0.82, respectively.
|
Association With THS/Total F Metabolites
The major urinary corticosteroid metabolites examined (tetrahydrocortisol, (allo)tetrahydrocortisol, tetrahydrocortisone, dehydroepiandrosterone, aetiocholanolone, tetrahydroaldosterone, and tetrahydrodeoxycortisol) are listed in Table 3. There was no significant difference in total F metabolite excretion rate (micrograms per 24 hours) between the 1889 G/T (P=0.4513) or 1859 A/G (P=0.2789) genotypes. The THS/total F ratio (index of 11ß-hydroxylase activity) was significantly higher in 1889 TT homozygotes than in GG homozygotes (P=0.024). A similar pattern was seen with the 1859 SNP, with GG homozygotes tending to have a higher ratio than AA homozygotes (P=0.056; Table 3). There was no statistically significant difference between the 1889 TT and GT groups, but the THS/F ratio remained significantly higher in the heterozygous group than the GG subjects (P=0.019). A similar pattern is seen for 1859, with no statistical difference between the GG and AG groups but a trend to a higher ratio in the heterozygous group than the AA homozygotes (P=0.066).
|
In Vitro Expression Studies
To investigate whether the altered 11ß-hydroxylase efficiency associated with the 1889 G/T and 1859 A/G SNPs in the BRIGHT population is because of an altered transcriptional response, the effects of these variants on reporter gene expression were analyzed in vitro. The impact of the 1889 G/T and 1859 A/G single base changes on gene expression using luciferase reporter gene constructs revealed no significant effect on basal expression levels. The construct containing the 1889 T and the 1859 G alleles has a significantly reduced response to stimulation with either 1 µmol/L of ACTH (P<0.001) or 10 µmol/L of forskolin (P=0.036) when compared with the contrasting allele combination (Figure 3).
|
| Discussion |
|---|
|
|
|---|
We now demonstrate that the overall sequence variation across the CYP11B locus is high, consistent with previous reports.32,33 Recent work has demonstrated that the 344C/T and intron Conv variants in CYP11B2 are in LD with SNPs in CYP11B1, specifically in exon 1 (225 G/A and Leu75Leu), intron 3 (2803 A/G), and intron 6 (4265 G/A).20,21 We have demonstrated previously that an increased THS excretion was correlated with haplotypes carrying the 344 T allele and that the strongest association was with the Leu75Leu SNP in CYP11B1. However, this is a conservative change and is unlikely to be functional. In the present study we have demonstrated that there is tight LD across the entire CYP11B locus. This is confirmed by the presence of relatively few common haplotypes, with 4 (each with a frequency >5%) accounting for 68% of chromosomes in the data set.
Our detailed sequencing identified several candidate variants that could account for the observed phenotypes associated previously with the 344C/T and intron Conv variants in CYP11B2. Several of these were identified in CYP11B2, and it is possible that one (or more) may contribute to the increased aldosterone levels that have been associated previously with the T/Conv haplotype.912 This is currently being investigated.
However, of particular interest to this study is that the 344C/T and intron Conv variants in CYP11B2 were shown to be in strong LD with several genetic variants in CYP11B1. Of the 19 variants identified in CYP11B1 that are in LD with 344 C/T and intron Conv, only 2 were located within the coding sequence, and both of these were synonymous changes (Leu75Leu and Asp82Asp). As in CYP11B2, the majority of variants identified in CYP11B1 were in intronic regions. However, 2 novel SNPs were identified in the 5' regulatory region upstream of the CYP11B1 gene at positions 1889 G/T and 1859 A/G. These may alter transcriptional regulation of the CYP11B1 gene and, therefore, contribute the observed biochemical phenotype of decreased 11ß-hydroxylation.
To investigate whether the 1889 G/T and 1859 A/G SNPs do indeed associate with altered 11ß-hydroxylation, we genotyped a subgroup from the BRIGHT hypertension study for both of these SNPs. Our data demonstrate a significant association between these SNPs and the ratio of THS/total F, the index of impaired 11ß-hydroxylase efficiency. Those subjects homozygous for the 1889 T allele showed lower 11ß-hydroxylation efficiency (increased urinary THS/total F ratio) than those homozygous for the G allele. Similarly, subjects homozygous for the 1859 G allele showed a trend toward a reduced 11ß-hydroxylation efficiency than those homozygous for the A allele, although this did not quite reach statistical significance. As mentioned previously, it is these alleles (1889 T and 1859 G) that are in LD with the CYP11B2 T/Conv haplotype. Importantly, there was no difference in urinary excretion of total cortisol metabolites between the 2 genotype groups for each SNP.
The effects of the 1889 G/T and 1859 A/G SNPs on gene transcription were tested in vitro using a luciferase reporter gene system. Compared with Wt (ie, 1889 G and 1859 A), constructs containing the 1889 T and 1859 G alleles reduced reporter gene activity by
50% in response to the agonists ACTH and the adenylate cyclase agonist forskolin, indicating decreased transcription. These data strongly support the theory that the reduced 11ß-hydroxylation associated previously with the T/Conv haplotype in CYP11B2 is accounted for by variants in CYP11B1, such as 1889 T and 1859 G. The is one of the first studies in humans to identify novel genetic variants in a candidate gene for hypertension, which not only associate with an intermediate phenotype but have also been shown to have functional effects on gene transcription in vitro.
Perspectives
Evidence of a slightly reduced 11ß-hydroxylase efficiency is a consistent feature of subjects with essential hypertension compared with normotensive control subjects.3436 This study demonstrates that the 1889 G/T and 1859 A/G SNPs in CYP11B1, which associate with reduced 11ß-hydroxylase efficiency, are in LD with those variants in CYP11B2 that associate with higher aldosterone levels. How the 2 effects are related is currently being investigated.
| Acknowledgments |
|---|
This study was supported by funding from the Medical Research Council (N.J.S., J.M.C.C., and E.D.) and the British Heart Foundation (S.M.M., J.M.C.C., and E.D.).
Disclosures
None.
Received June 13, 2006; first decision July 11, 2006; accepted October 3, 2006.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S Sookoian, T F Gianotti, and C J Pirola Role of the C-344T aldosterone synthase gene variant in left ventricular mass and left ventricular structure-related phenotypes Heart, July 1, 2008; 94(7): 903 - 910. [Abstract] [Full Text] [PDF] |
||||
![]() |
P.-a. B. Shih and D. T. O'Connor Hereditary Determinants of Human Hypertension: Strategies in the Setting of Genetic Complexity Hypertension, June 1, 2008; 51(6): 1456 - 1464. [Full Text] [PDF] |
||||
![]() |
J. M. C. Connell, S. M. MacKenzie, E. M. Freel, R. Fraser, and E. Davies A Lifetime of Aldosterone Excess: Long-Term Consequences of Altered Regulation of Aldosterone Production for Cardiovascular Function Endocr. Rev., April 1, 2008; 29(2): 133 - 154. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Y. Deng Genetic basis of polygenic hypertension Hum. Mol. Genet., October 15, 2007; 16(R2): R195 - R202. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |