Haplotype association and synergistic effect of human aldosterone synthase (CYP11B2) gene polymorphisms causing susceptibility to essential hypertension in Indian patients

Background: Aldosterone synthase (CYP11B2) is a key enzyme involved in the terminal steps of aldosterone biosynthesis. Genetic variability in CYP11B2 gene has been associated with heterogeneous aldosterone production, which can affect sodium homeostasis and thereby regulation of blood pressure. Hence, the present study was aimed to explore the single-locus variations, haplotype and epistasis patterns of CYP11B2 (C-344T, intron-2 gene conversion and Lys173Arg) gene polymorphisms, and the risk contributed by them to the development of essential hypertension (EHT). Methods: A total of 279 hypertensive patients and 200 normotensive controls were enrolled in this study. C-344T and Lys173Arg polymorphisms of CYP11B2 gene were genotyped by PCR-RFLP method and intron-2 gene conversion (IC) polymorphism by allele-specific PCR analysis. Results: Single-locus analysis revealed significant association of CYP11B2 C-344T and Lys173Arg polymorphisms with EHT (p < 0.05). Considering the sexes, Lys173 allele was found to be at risk for hypertension in males (OR 1.40; 95% CI = 1.01–1.96). Unphased haplotype analysis revealed H1 (T-Conv-Lys; p = 0.0017) to have significant risk for EHT, while haplotype H4 (T-Wt-Arg) had a significant protective effect. Multifactor dimensionality reduction (MDR) interaction analysis found the overall best model with C-344T and IC polymorphisms exhibiting strong synergistic effect. Conclusion: The present study revealed a strong synergistic effect of CYP11B2 C-344T and IC polymorphisms causing susceptibility to EHT and haplotype H1 (-344T-Conv-Lys173) as the risk-conferring factor for hypertension predisposition.

Keywords: CYP11B2; essential hypertension; epistasis; haplotype; MDR; polymorphism; single locus

Essential hypertension (EHT) is a complex disorder resulting from interplay between an individual's genetic background and exposure to environmental factors ([1]). The identification of these factors has proved challenging since EHT acts as a major risk factor for cardio, cerebro and reno vascular diseases ([2]).

Renin–angiotensin–aldosterone system (RAAS) is a key pathway for studying cerebro vascular complications, because it influences blood pressure, vasoconstriction, thrombosis and vessel wall damage ([3]). Aldosterone, the main effectors of the RAAS pathway, affects sodium balance, intravascular volume and blood pressure ([4]). Aldosterone synthase (CYP11B2) is a key enzyme involved in the biosynthesis of aldosterone, in the zonaglomerulosa of the adrenal cortex. The aldosterone synthase gene, CYP11B2, encodes for a cytochrome P450 enzyme, involved in the terminal steps of aldosterone synthesis and its expression is regulated by angiotensin II and potassium ([5]). CYP11B2 gene is highly homologous to 11 β hydroxylase (the product of the CYP11B1 gene), involved in the terminal steps of cortisol biosynthesis. The CYP11B1 and CYP11B2 genes lie approximately 40 kb apart on human chromosome 8q21-22 ([6]) spanning approximately 7 kb and each comprising 9 exons and 8 introns (7; Figure 1). The sequence homology of the two genes is 95% identical within their coding regions, falling to 90% identity in the introns and the encoded proteins are 93% identical in their predicted amino acid sequences. Only two CYP11B genes are found in humans, whereas rats possess two more genes CYP11B3 and CYP11B4.

Graph: Figure 1. Location of six biallelic polymorphisms of CYP11B2 gene on chromosome 8. Exons 1–9 are indicated by vertical bars.

In humans, several genetic variants of CYP11B2 gene have been identified. According to National Center for Biotechnology Information (NCBI) database (http://www.ncbi.nlm.nih.gov), 227 SNPs have been identified in different populations, viz. European, Asian, African and North American. Among these, three polymorphic variants of CYP11B2 (rs1799998, C-344T promoter polymorphism; rs4546 the intron2 gene conversion (IC) polymorphism; and rs6414, 2713A/G, or Lys173Arg (K173R) polymorphism in exon 3) have been selected for the present study. These polymorphisms have been associated with hypertension and cardiovascular disease, exerting deleterious effects on normal regulation of blood pressure ([8],[9]).

The C-344T (rs1799998) variant is a common polymorphism in the promoter region of CYP11B2, which involves a C to T substitution in the steroidogenic transcription factor 1 (SF-1) binding site ([10]), which occurs in approximately 30% of African Americans and 46% of Europeans (http://www.ncbi.nlm.nih.gov/SNP/). The promoter C-344T polymorphism has been reported to be associated with serum aldosterone level and its production ([11]), blood pressure ([12],[13]), urinary aldosterone excretion ([14]), aldosterone:renin ratio (ARR) ([15]) and left ventricular size and mass ([16]).

Second variant, an Intron 2 gene conversion polymorphism (IC or Iw/Ic), wherein a part of intron 2 of the CYP11B2 gene is replaced by the corresponding intron of CYP11B1 gene, has also been studied in detail ([10]). There are reports of Intron 2 polymorphism being associated with hypertension. However, there is considerable heterogeneity among these studies ([8],[17]). Lys173Arg polymorphism, resulting in the substitution of arginine (Arg) by lysine (Lys) at position 173 (K173R) in the protein sequence, has been described in the Chilean and Japanese patients with hypertension ([18],[19]). However, the mechanism by which these polymorphisms might result in increased aldosterone production is not clear.

Multiple genes/loci contribute to the etiology of EHT either independently or by synergistic action with other genes/loci which are related to the regulation of blood pressure. To unravel fully the genetic architecture of EHT, focus must be laid on the transmission of multilocus haplotypes, the synergistic effects between candidate genes and interaction of genetic variants with each other ([20],[21]). Single-locus association analyses are ineffective in explaining the genetic contribution in case of complex traits ([22]), such as blood pressure. So, the present study was aimed to explore the contribution of single-locus, haplotype and epistasis patterns of CYP11B2 polymorphisms in the development of EHT.

The present case–control study included a total of 279 hypertensive patients and 200 normotensive controls. Hypertensive patients with confirmed diagnosis (with systolic blood pressure (SBP) ≥ 140 mmHg and diastolic blood pressure (DBP) ≥ 90 mmHg) and those who were already on antihypertensive drugs at the time of investigations were recruited from the outpatient clinics of Osmania General Hospital, Hyderabad and Durgabai Deshmukh Hospital and Research Centre, Hyderabad, India. Cases associated with other common conditions like diabetes, coronary artery disease, ischemia and secondary hypertensive cases (based on blood urea, serum creatinine, lipid profiles, etc.) were excluded from the study. Normal healthy controls (with SBP/DBP <130/90 mmHg) who were on routine health check-up were selected at random from the same population for comparison with the patient group. The study was approved by the institutional ethical committee and informed consent was taken from patients and controls for participating in the study.

About 5 ml of venous blood samples were obtained from patients and controls in EDTA Vaccutainer tubes. Genomic DNA was isolated from these samples by using rapid non-enzymatic method ([23]) and used for the analysis of CYP11B2 polymorphisms.

C-344T polymorphism was determined by PCR-RFLP wherein a fragment of 537 bp was generated by primer pair: forward 5′-CAG GAG GAG ACC CCA TGT GAC-3′ and reverse 5′-CCT CCA CCC TGT TCA GCC C-3′ ([14]). The PCR reaction was performed in a 10 µl volume of 20 pmoles of each primer, 0.2 mM each dNTP, 1 µl buffer and 0.25 U Taq DNA polymerase, with an initial denaturation of 94°C for 5 min, followed by 35 cycles of 1 min at 94°C, 1 min at 59°C, 2 min at 72°C and finally 5 min at 72°C. Following amplification, the PCR products were digested with 10 U of Hae III (New England Biolabs, UK) restriction enzyme and incubated at 37°C overnight and subjected to electrophoresis on 2% agarose gels. The genotyping was done based on the loss of restriction site due to the substitution of C by T. As a result -344T allele generates a single fragment of 273 bp and -344C allele generates two fragments of 202 and 71 bp.

Considering the Lys173Arg (K173R) polymorphism, a 1286 bp PCR product was amplified using specific primers: forward 5′-AGG CAG CTT CTA CCA GGG CCC CAG TCA CTC-3′ and reverse 5′-CCC CTC CCC TGC AAA TCT CAT CCC TTA-3′ and PCR conditions as described by Fardella et al. ([18]). The amplified product was digested with 5 U of Bsu361 (New England Biolabs) and incubated overnight at 37°C. The 173 Arg genotype produced two bands (1037 and 249 bp) 173 Lys genotype produced a single band (1286 bp) and the heterozygotes displayed all three bands (1286, 1037, and 249 bp).

Intron 2 gene conversion (IC) polymorphism was determined by amplification refractory mutation system (ARMS) wherein a 418 bp fragment was generated using reverse primer 5′-AGG AAC CTC TGC ACG GCC-3′; intron conversion allele (Ic) forward allele specific primer 5′-CAG AAA ATC CCT CCC CCC TA-3′; and no-conversion allele (Iw) forward allele specific primer 5′-TGG AGA AAA GCC CTA CCC TGT-3′. The genotypes were detected on 1.5% agarose gel. A 10% random sample of the study population was double genotyped in a blinded fashion with 100% concordant results.

SPSS (version 16.0) package was used to analyze the data for descriptive statistics and computation of means and standard deviations. χ2 statistics was computed for significance of differences in the distribution of genotypes between patients and controls. The frequencies of the marker alleles were estimated by allele counting method and tested for Hardy–Weinberg equilibrium (HWE). Odds ratios (OR) and 95% confidence interval (CI) were computed for different combinations of genotypes to estimate the risk contributing to the onset of hypertension. Each genotype was assessed by logistic regression analysis using dominant (major homozygote vs. heterozygotes plus minor homozygotes) and recessive (major homozygote plus heterozygotes vs. minor homozygotes) genetic models. A p-value of ≤0.05 was considered statistically significant.

Haplotype frequencies were estimated using the EH/EH+ program ([24]) and compared by χ2 test from a series of 2 × 2 contingency tables by combining other haplotypes. EH is a program to test and estimate linkage disequilibrium between different markers or between a disease locus and markers. In this program, the data consist of a number of individuals collected at random from a population. Based on these sample data, the EH program estimates allele frequencies for each marker. Haplotype frequencies are estimated with allelic association (H1) and without (H0). The EH program also provides log likelihood, χ2 and the number of degrees of freedom under hypotheses H0 and H1 ([25]). A probability value less than 0.00625 (0.05/8) was considered statistically significant after Bonferroni correction applied for haplotype-based multiple testing. The omnibus-likelihood ratio test was performed to examine the differences in haplotype frequency profiles between the cases and controls. The statistical significance of comparisons was assessed by comparing twice the difference in logarithm of the likelihood (−2 Ln L) with the χ2 distribution with appropriate degrees of freedom. The pairwise linkage disequilibrium coefficient was calculated with estimated haplotype frequencies using the 2LD program ([26]). The extent of disequilibrium was expressed in terms of D′ and r2.

Multifactor dimensionality reduction (MDR) analysis was performed using MDR software to determine the presence of epistatic interaction and the genotypic combination of the two genes that may confer high or low risk for development of hypertension. Evaluation of gene–gene and gene–environment interaction was performed using a four-step process outlined by Moore and Williams ([20]). The analysis was implemented using open-source MDR software package (version 1.1.0). Best models with possible combinations of the polymorphisms were considered based on 10-fold cross-validation and maximum testing accuracy.

The demographic data of the patients and controls are summarized in Table 1. The mean age recorded for patients was 55.57 ± 9.78 and for controls was 47.63 ± 9.65 years. Onset of hypertension in the present study ranged between 35 and 65 years with high frequency of cases ranging between 45 and 60 years. The difference in the mean ages of patients (55.57 ± 9.78 years) and controls (47.63 ± 9.65 years) was mainly due to high frequency of patients above the age of 60 years as compared with controls, while the frequency in the age group between 45 and 60 years was same in patients (52.6%) and controls (52.5%). Controls above the age of 60 years were reluctant to participate in the study. There was significant elevation in the mean levels of age, BMI, SBP and DBP in hypertensive patients as compared with controls (p < 0.05).

Table 1. Demographic and clinical characteristics of the study population.

6 Mean ± SD for age, BMI, SBP, DBP. *Significant difference at p < 0.05 from control.

The genotype and allele distributions of three CYP11B2 polymorphisms are shown in Table 2. The observed genotype frequencies for all three CYP11B2 polymorphisms were consistent with HWE in hypertensive group, whereas distribution of C-344T and Lys173Arg polymorphisms deviated from HWE in the control group. Considering single-locus analysis, there was a significant association of CYP11B2 C-344T (χ2 = 6.074; p = 0.048) and Lys173Arg polymorphisms (χ2 = 6.152; p = 0.046) with hypertension. In case of Intron-2 gene conversion polymorphism (IC), statistically insignificant difference was found for the genotype and allele distributions between the patients and controls.

Table 2. Genotype and allele distributions of the three CYP11B2 polymorphisms in patients and controls.

7 Genotype frequencies are expressed as number (%) and allele frequencies are indicated as fractions. IC, Intron 2 gene conversion polymorphism; Conv, conversion; Wt, wild-type; OR, odds ratio; CI, confidence interval. *Significant p-value for the comparisons made between patients and controls.

Under the assumption of dominant or recessive models, we observed that under dominant model CYP11B2 Lys173Arg polymorphism was associated with an elevated risk of EHT (OR 1.65; 95%CI = 1.11–2.45; p = 0.013), while IC polymorphism conferred a protective effect (OR 0.67; 95%CI = 0.47–0.98; p = 0.036; Table 2). Univariate analysis also indicated that the frequency of the Lys173 allele was significantly higher in hypertensive males than in controls (62.1% vs. 53.8%). Other polymorphisms did not show any significant difference with respect to gender (Table 3).

Table 3. Distribution of the CYP11B2 gene polymorphisms between males and females among hypertensive and control subjects.

8 HT, hypertensive patients; NT, normotensive controls; Conv, conversion; Wt, wild-type.*p-Value significant at 0.05 level.

To understand the relationship among the three polymorphisms genotyped in CYP11B2, pairwise linkage disequilibrium was measured by D′. Using the 2LD program, it was observed that the C-344T promoter polymorphism was in strong LD with Lys173Arg polymorphism in hypertensives than in normotensives (D′: 0.638 vs. 0.551). Whereas, moderate LD was shown in between C-344T and IC polymorphisms (D′: 0.521 vs. 0.103) (Table 4).

Table 4. The pairwise linkage disequilibrium analysis.

9 The upper triangular data denoted the pairwise linkage disequilibrium coefficients in hypertensives, and the lower triangular data in normotensives.

Haplotype analysis using the EH/EH+ program was conducted to study the overall unphased haplotype frequency profiles of CYP11B2 gene which showed significant difference between the cases and controls (χ2 = 14.48; df 7; p = 0.0433; Table 5), indicating presence of disease-predisposing and disease-protective haplotypes in patients with EHT. An additional individual haplotype analysis for each haplotype using χ2 test indicated that among all eight haplotypes, the frequency of haplotype H1 (T-Conv-Lys; p = 0.0017) was significantly higher (24.3% vs. 15.9 in patients and controls), while that of haplotype H4 (T-Wt-Arg; p = 0.0004) was significantly lower (4.5% vs. 9.9%) in cases than in controls at the significant level of p < 0.00625 even after Bonferroni's correction (p < 0.05/8 = 0.00625 for eight individual haplotype analysis). In case of haplotype H6 (C-Conv-Arg; p = 0.0353), the frequency was lower in cases than in controls (3.8% vs. 7.2%) with significance of p < 0.05, but turned insignificant after Bonferroni's correction was applied.

Table 5. Haplotype structure and frequencies of the CYP11B2 C-344T, IC, and K173R polymorphisms in hypertensive and normotensive subjects.

10 Conv, conversion; Wt, wild-type. *Calculated using χ2 test from a series of 2 × 2 contingency tables by combining other haplotypes. †Calculated using χ2 test from 8 × 2 contingency table with degree of freedom (df) of 7. ‡Likelihood ratio statistic for omnibus test in the EH/EH+ program.

MDR analysis revealed a two-way interaction model including C-344T and IC polymorphisms as the overall best model with maximum testing accuracy of 0.5729 and CV consistency of 9/10. The interaction information analysis revealed a strong synergism between the markers C-344T and IC contributing to the development of EHT.

The distribution of cases and controls for the two interacting polymorphisms (-344T and IC) are represented in Figure 2. High-risk (dark grey) and low-risk (light grey) genotypic combinations were determined based on the threshold value for the present data which was 1.39 (279/200). It was observed that TT genotype of C-344T when present in combination with IcIc (Conv) genotype of Ic polymorphism conferred a twofold risk (17/8), while when present in combination with IwIc genotype conferred a threefold risk (53/17) for developing EHT. Further, IwIw (Wild) and IwIc genotypes of IC when present in combination with CC genotype of C-344T (34/17 and 13/7, respectively) conferred a twofold risk for developing hypertension.

Graph: Figure 2. Distribution of high-risk (dark shaded) and low-risk (light shaded) genotypes among the markers studied. The summary of the distribution illustrates the hypertensives (left bars) and normotensives (right bars) for each genotype combination.

EHT is a multifactorial disease with major clinical manifestations and multiple etiologies, and a significant cause of disability and death in developed countries ([27]). The host genetic susceptibility, combined with environmental and clinical risk factors, may play a crucial role in the development and progression of cardiovascular diseases ([28]).

The aldosterone-synthase gene (CYP11B2) plays a pivotal role in catalyzing the biosynthesis of aldosterone ([29]), which is involved in re-absorption of sodium ions and water in the kidney, leading to elevated blood pressure. It has been reported that variations at CYP11B2 gene are associated with EHT and influence aldosterone secretion ([30]). In the present study, association of three polymorphic variants (C-344T, IC and Lys173Arg) of CYP11B2 with EHT provided interesting insights.

Several reports including the present study revealed a positive association of C-344T promoter polymorphism with EHT. The results were conflicting, with some studies showing a positive association of T allele with hypertension ([31],[32],[33]), some with the C allele ([12],[34],[35]), and others reporting a lack of association with hypertension ([36]). This inconsistency in the finding of the results can be attributed to the differences in background characteristics of the study subjects such as ethnicity, selection criteria, age, environmental factors and techniques employed. Despite these contradictory findings, a recent meta-analysis designed to address this controversy did conclude that the -344T allele is indeed associated with a higher, though modest risk of hypertension, in African, Caucasian, Japanese and Chinese cohorts but that it has minimal influence on aldosterone excretion ([33]). In addition, in a multi-ethnic cross-sectional study of 1313 middle-aged men and women (456 white, 441 of African origin and 416 South Asian), the TT genotype was associated with 14% higher plasma aldosterone levels, 3.7 mmHg higher systolic and 2.1 mmHg higher diastolic blood pressure than CC (p < 0.05) ([11]).

The exact mechanism whereby the C-344T polymorphism of CYP11B2 gene variant may lead to higher blood pressure remains unknown. It has been suggested that the C-344T variant by itself does not directly influence promoter activity, rather binding of steroidogenic factor-1 (SF-1) to this site downregulates activity of CYP11B2 promoter by making SF-1 less available to functionally affect other CYP11B2 promoter sites, which could alter expression of the gene ([37]). Moreover, in vitro studies showed that C allele binds SF-1 four times more than it does the T allele ([10]), suggesting a modulating effect of this variation in the transcription of the enzyme and, in turn on the aldosterone production. A study by Davies et al. ([31]) proposed that -344T allele has physiological advantage for absorbing or maintaining sodium and high salt intake, which in turn, induces extra body sodium and fluid accumulation leading to disturbances in blood pressure.

Little evidence supports the role of IC polymorphism in pathophysiology of hypertension. Previous studies found an association of gene conversion with hypertension ([15],[17],[32]). The present study conferred protective effect for individuals with lw/lw genotype under dominant model (OR 0.67; 95%CI = 0.47–0.98; p = 0.036; Table 2). Brand et al. ([38]) did not reveal any association of IC polymorphism with hypertension.

Lys173Arg polymorphism was putatively involved in low-renin hypertension in Chilean and Japanese patients ([18],[19]). Fardella et al. ([18]) found that the lysine coding allele was highly distributed among a large number of Chilean patients with hypertension. Similarly, the present study confirms significant association of Lys173Arg polymorphism with hypertension in general (p = 0.046; Table 2) as well as when males were considered (p = 0.043; Table 3). In contrast to the present study, Gu et al. ([32]) reported that female hypertensives showed higher frequency of lysine isoform when compared with controls. A study by Davies et al. ([39]) on K173R and IC polymorphisms hypothesized that they are less likely to be of physiological significance and independently effect CYP11B2 transcription because they do not alter enzyme activity in vitro and are in linkage disequilibrium. However, our findings indicate that the Lys173Arg and IC polymorphisms confer more effects on the status of hypertension.

Although methods based on single-nucleotide polymorphisms (SNPs) yield important insights, haplotype-based methods can provide additional statistical power to detect genes involved in complex trait diseases and information on factors that influence the dependency among genetic markers ([40]). Haplotype analysis of CYP11B2 polymorphisms (C-344T, IC, and Lys173Arg) revealed that haplotypes H4 (T-Wt-Arg) and H6 (C-Wt-Arg) had protective effect, while haplotype H1 (-344T-Conv-Lys173) was disease causing. This finding is consistent with results of Gu et al. ([32]), who reported that haplotype H4 (-344T-173Lys-IC (conversion)) was significantly higher in hypertensives. A previous study by Nejatizadeh et al. ([17]) reported that prevalence of T-Ic haplotype in hypertensives serves as risk predisposing haplotype, and T-Iw haplotype prevalent in controls serves as protective.

MDR analysis in the present study revealed a two-locus model including CYP11B2 C-344T and IC polymorphisms as the overall best interaction model. It was observed that TT genotype of C-344T when present in combination with IcIc (Conv) genotype of Ic polymorphism conferred a twofold risk (17/8), while when present in combination with IwIc genotype conferred a threefold risk (53/17) for developing EHT. Further, IwIw (Wild) and IwIc genotypes of IC when present in combination with CC genotype of C-344T (34/17 and 13/7, respectively) conferred a twofold risk for developing hypertension.

The present study has certain limitations. The present data are part of the major project conducted among patients using several gene polymorphisms as markers that are likely to cause risk for hypertension. CYP11B2 is one of such markers. The recruitment of the subjects in the present study covered a period of 7 years (2005–2012). Secondary hypertensive cases and those associated with conditions like diabetes and thyroid disorders, were eliminated limiting the sample size to 279. The power of the sample size calculated by R software was 70% which is closer to approved level (80%). Although we could conclude the significant association of unphased haplotype groups with EHT, we could not correlate it with the functional role of CYP11B2 gene and its variants since this objective was not planned originally and the required facilities for functional studies were not available. We recommend further studies on patients from India with global comparison to establish significant role of CYP11B2 in causing EHT.

In conclusion, analysis of the present data reveals that interactions between polymorphisms of the CYP11B2 (C-344T, IC and Lys173Arg) may have an independent as well as synergistic effect on the pathogenesis of hypertension, and therefore may explain differences in an individual's genetic susceptibility. The haplotype architecture that has been established in this study and the potentially informative variants implicated within these haplotypes, would be of use in other genetic association studies of the CYP11B2 locus in Indian population. It also can be added to the panel of markers that enables screening of EHT patients prior to expression of the condition.

We thank the patients and their families for their invaluable contribution.

The authors declare no conflicts of interest.

This research was supported by Indian Council of Medical Research [number 45/15/2006//BMS] and University Grants Commission [number F.3-80/2003/(SR)], New Delhi, India.