eNOS (endothelial NO synthase) plays a critical role in the development of ventricular remodelling and cardiac hypertrophy. The purpose of the present study was to determine whether three common variants in NOS3 (the eNOS gene) are associated with the risk of LVH [LV (left ventricular) hypertrophy] in patients with essential hypertension. Three NOS3 genetic variants, −T786C (rs2070744), eNOS4a/b and +G894T (rs1799983), were genotyped in two independent case-control studies: the first study consisted of 1061 hypertensive patients with LVH and 1118 hypertensive patients without LVH, and the second sample consisted of 120 patients with LVH and 223 patients without LVH. Echocardiographic measurements were obtained in all of the hypertensive patients. Only the +G894T (E298D) variant of NOS3 was associated with a higher risk of LVH {OR (odds ratio), 1.67 [95% CI (confidence interval), 1.19–2.36]; P<0.01} in the first population, and replicated in the second population [OR, 1.41 (95% CI, 1.01–2.28); P<0.05] in a recessive model. Compared with carriers of the G allele (GT+GG), patients carrying the TT genotype had increased septal wall thickness (16.2%, P<0.01 and 11.7%, P<0.01 respectively), LV posterior wall thickness (8.3%, P<0.01 and 7.1%, P<0.01 respectively), LV mass index (14.0%, P<0.01 and 25.1%, P<0.01 respectively) and relative wall thickness (13.1%, P<0.01 and 16.2%, P<0.01 respectively) in the first and second populations. The results of the present study support that homozygosity for +G894T (E298D) in NOS3 is a genetic risk factor for the development of LVH in patients with hypertension.

INTRODUCTION

LVH [LV (left ventricular) hypertrophy] is a strong independent risk factor for cardiovascular morbidity and all-cause mortality [1,2]. The factors age, male gender, postmenopausal status in women, life style and diabetes, have been known to contribute to the development of LVH [3]. Although hypertension is a major cause of LVH, the degree of LVH does not parallel the level of BP (blood pressure), the duration of hypertension or reversal of hypertensive LVH by pharmacological treatments. Indeed, genetic factors account for 60% of BP-independent cardiac mass variances [4], suggesting the important role played by genetic factors in the development of LVH in essential hypertension.

NO is synthesized under the control of eNOS [endothelial NOS (NO synthase)] encoded by NOS3 located on chromosome 7q35 to 36 in the vascular endothelium and in cardiomyocytes [5]. In addition to acting as a potent vasodilator that can reduce cardiac load, NO can also stimulate angiogenesis, reduce cardiomyocyte hypertrophy, limit the production of extracellular matrix proteins by cardiac fibroblasts, and influence many processes involved in the pathophysiology of LV remodelling [68]. In animal models, long-term systemic inhibition of NO production with the NOS inhibitor L-NAME (NG-nitro-L-arginine methyl ester) or Nos3-deficient mice have been shown to induce greater LVH, fibrosis and dysfunction, especially concentric LVH [9,10]. Restoration of eNOS in the heart of Nos3-deficient mice has been shown to attenuate LVH and dysfunction in the TAC (transverse ascending aorta constriction) model [11], and also inhibits hypertrophy in the remote myocardium and preserves cardiac function after MI (myocardial infarction) [12]. Furthermore, the expression of NOS3 has been shown to be increased in the ventricular myocardium of failing human hearts [13], suggesting that the endothelial NO/NOS pathway has an important role in LV remodelling and LVH.

It has been reported that eNOS expression or function is affected by genetic variants [14]. Given the important role of eNOS in the regulation of cardiac remodelling in hypertension, we hypothesized that the genetic variants in NOS3 would be associated with susceptibility to LVH in human essential hypertension. In the present study, to test this hypothesis, we genotyped the common variants of NOS3 in two independent case-control studies.

MATERIALS AND METHODS

Study sample

The study was approved by the ethics committee of FuWai Hospital, Peking Union Medical College, and the participating hospitals. All subjects who participated in the study provided their written informed consent and were self-reported as Han nationality.

Two independent samples were included, and the subjects were recruited for investigating the prevalence and risk factors for LVH in the Chinese hypertensive population. The subjects of the first study, which consisted of 1061 hypertensive patients (mean age 59.1±8.7 years) with LVH and 1118 patients (mean age 57.6±9.1 years) without LVH, were recruited from 24 communities in Xinyang County, a region in central China. BPs were determined on the right arm with a mercury sphygmomanometer with standardized techniques after at least 5 min rest in the sitting position by the same investigator. Hypertension was defined as SBP (systolic BP)≥140 mmHg and/or DBP (diastolic BP)≥90 mmHg on an average of two measurements or by current anti-hypertensive treatment by WHO (World Health Organization) criteria [15]. Exclusion included secondary arterial hypertension, atrio-ventricular conduction block, chronic obstructive bronchitis, bronchial asthma, chronic myeloproliferative diseases, diabetes, hypertrophic cardiomyopathy, valvular heart diseases, pulmonary hypertension, coronary heart disease and heart failure. In the first population, 25.5% (271/1061) of patients with LVH and 25.2% (282/1118) of patients without LVH were on anti-hypertensive medication. The classes of drugs used were ACEIs (angiotensin-converting enzyme inhibitors), β-blockers, diuretics and calcium entry blockers.

The second study sample consisted of 120 hypertensive patients (mean age 59.2±7.4 years) with LVH and 223 patients (mean age 57.6±10.1 years) without LVH, recruited from Rizhao County, a region in Eastern China, following the same inclusion and exclusion criteria as the first study. In the second population, 16.8% (20/120) of patients with LVH and 17.6% (39/223) of patients without LVH were on anti-hypertensive medication.

Echocardiographic measurement

Echocardiography was performed in all participants (HP 5500; Phillips Medical System, or an HDI 3000; ATL). M-mode and two-dimensional images were recorded at 30 frames/s on super TDK videotape (Aloka). Subjects were examined in the supine and left lateral position by three echocardiographers who were blinded to the genotype of the patient. Echocardiographic images were obtained in the parasternal long- and short-axis views, and apical two- and four-chamber views with standard transducer positions [16]. Three physician-echocardiographers supervised the echocardiographic examination. LV internal diameters, IVS (interventricular septum thickness) and PWT (posterior wall thickness) were measured on up to three cardiac cycles at end-diastole and endsystole according to the recommendations of the American Society of Echocardiography [17].

The LVM (LV mass) was calculated at end-diastole by use of the cube formula: 0.8×1.04 [(IVS+LVIDD+PWT)3−LVIDD3]+0.6 [18], where LVIDD is LV end-diastolic diameter. LVM was divided by height2.7 to obtain LVMIh2.7 (LVM index). LVH was defined as LVMI>49.2 g/m2.7 for men and 46.7 g/m2.7 for women [19]. RWT (relative wall thickness) was calculated by the formula: (IVS+PWT)/LVIDD [20].

Clinical data collection

A blood sample was taken from all participants after overnight fasting. Plasma and urine biochemical variables were determined by using standard methods with an automatic analyser (Hitachi 7060). All of the participants completed a questionnaire on current medication and life style. A complete medical history was obtained from all subjects, including family history of hypertension, diabetes mellitus and the following cardiovascular risk factors: alcohol intake, cigarette smoking, family history, weight, height, BMI (body mass index), SBP and DBP. BMI was calculated by the formula of weight (kg)/height (m2). Creatinine clearance rate was calculated by the formula of (140−age in years)×weight (kg)/[72×serum creatinine (mg/dl)] for men and (140−age in years)×weight (kg)/[72×serum creatinine (mg/dl)]×0.85 for women [21].

Genotyping

Genomic DNA was extracted from peripheral blood using a method described previously [22]. A total of 277 variants in NOS3 have been identified so far according to public databases (dbSNP; http://www.ncbi.nlm.nih.gov/SNP/). We undertook a search for variants within the promoter and coding regions of NOS3 and identified three important functional variants, −T786C (rs2070744), eNOS4a/b and +G894T (rs1799983), with the potential to alter the expression or function of eNOS according to reports in the literature [14]. At the 5′-flanking region, a variant within the promoter (−786T/C) and a variant of 27 bp VNTRs (variable number tandem repeats) within intron 4 have been associated with alterations in promoter activity. A variant located in exon 7 (+G894T) encodes an amino acid change from glutamate to aspartate. The amino acid change has been shown to result in the enzyme being more susceptible to proteolytic cleavage. Therefore three NOS3 genetic variants, −T786C (rs2070744), eNOS4a/b and +G894T (rs1799983), were genotyped in two independent case-control studies by PCR/RFLP (restriction fragment length polymorphism).

The primers for −T786C were 5′-GACACAGAACTACAAACCCC-3′ and 5′- GCAGGTCAGCAGAGAGACTA -3′. The 178-bp PCR products were digested with MspI (New England Biolabs) and yielded two DNA fragments of 41 bp and 137 bp for the C allele on a 4% (w/v) agarose gel and only one band for the T allele.

The eNOS4a/b genotype was determined by PCR using primers 5′-AGGCCCTATGGTAGTGCCTT-3′ and 5′-TCTCTTAGTGCTGTGGTCAC-3′. PCR products of 393 or 420 bp were yielded for the smaller allele (eNOS4a) and the large allele (eNOS4b) on a 3% (w/v) agarose gel respectively.

The primers for +G894T were 5′-GCTCAGCCCCAGAACCCCCT-3′ and 5′-GCTCCAGGGGCACCTCAAGG-3′. The 171-bp PCR products were digested with BanII (New England Biolabs) and yielded two DNA fragments of 53 bp and 118 bp for the G allele on a 4% (w/v) agarose gel and only one band for the T allele.

Reproducibility of genotyping was confirmed by bidirectional sequencing in 100 randomly selected samples, and the reproducibility was 100%.

Statistical analysis

Values are expressed as means±S.D. A χ2 test was used to test categorical variables, and the Hardy–Weinberg equilibrium was used to test the variant frequencies. One-way ANOVA and an unpaired Student's t test was used to compare quantitative variables in the clinical characteristics of the subjects in the two case-control studies. Association of the genetic variants with LVH was analysed by multivariate logistic regression. ORs (odds ratios) and 95% CIs (confidence intervals) were calculated by use of multivariate logistic regression analyses, and were adjusted by age, gender, BMI, SBP, DBP, smoking status, alcohol consumption and glucose. A general linear model-univariate analysis was performed where each cardiac echocardiographic parameter was used as a dependent variable, age, gender, BMI, SBP, DBP, glucose and creatinine clearance rate were used as co-variables, and the variants of −T786C, +G894T and eNOS4a/b were used as independent variables. A SHEsis software platform was used to analyse pair-wise LD (linkage disequilibrium). Statistical analysis was performed with the SPSS 13.0 package. A value of P<0.05 was considered significant. Bonferroni correction was used for correction of multiple comparisons.

RESULTS

The clinical characteristics of the subjects in the two case-control studies are shown in Table 1. The distribution of the genotypes of the three NOS3 variants is shown in Table 2. The genotype frequencies for all the three variants were in accordance with the Hardy–Weinberg equilibrium.

Table 1
Clinical characteristics

Values are means±S.D. or percentages. Ccr, creatinine clearance rate. †P<0.01 compared with hypertensive patients without LVH.

The first study populationThe second study population
CharacteristicWithout LVHWith LVHWithout LVHWith LVH
Patients (n1118 1061 223 120 
Age (years) 57.6±9.1 59.1±8.7 57.6±10.1 59.2±7.4 
Men (%) 37.9 32.4 37 34.2 
BMI (kg/m225.5±3.5 27.1±6.5† 25.7±3.3 26.0±3.1† 
SBP (mmHg) 160.0±22.6 164.1±28.6† 157.4±18.1 160.7±18.1† 
DBP (mmHg) 96.6±11.4 96.5±14.4† 91.3±10.3 93.6±11.9† 
Glucose (mmol/l) 5.58±1.77 5.53±1.64 5.56±1.40 5.65±1.74 
Cigarette smoking (%) 19.4 16.1 17.9 19.2 
Alcohol intake (%) 22.3 19.0 17.9 26.4 
Ccr (ml/min) 65.6±7.6 66.2±7.8 53.9±7.4 57.6±8.4 
IVS (mm) 9.3±1.4 10.8±1.5† 9.3±1.4 11.7±1.7† 
PWT (mm) 9.1±1.1 11.5±1.4† 9.2±1.2 11.3±1.3† 
LVIDD (mm) 43.6±4.2 48.0±4.8† 43.3±4.2 47.0±5.0† 
LVISD (mm) 28.1±4.7 30.6±7.4† 28.3±4.4 30.3±6.4† 
RWT (%) 43.0±7.6 47.1±8.4† 42.8±7.3 48.9±8.3† 
LVM (g) 132.1±28.8 199.0±27.6† 129.6±31.0 200.3±33.2† 
LVMI (g/m2.737.5±5.4 59.5±9.6† 36.5±6.9 58.5±10.1† 
Hypertension treatment (%) 25.2 25.5 17.6 16.8 
The first study populationThe second study population
CharacteristicWithout LVHWith LVHWithout LVHWith LVH
Patients (n1118 1061 223 120 
Age (years) 57.6±9.1 59.1±8.7 57.6±10.1 59.2±7.4 
Men (%) 37.9 32.4 37 34.2 
BMI (kg/m225.5±3.5 27.1±6.5† 25.7±3.3 26.0±3.1† 
SBP (mmHg) 160.0±22.6 164.1±28.6† 157.4±18.1 160.7±18.1† 
DBP (mmHg) 96.6±11.4 96.5±14.4† 91.3±10.3 93.6±11.9† 
Glucose (mmol/l) 5.58±1.77 5.53±1.64 5.56±1.40 5.65±1.74 
Cigarette smoking (%) 19.4 16.1 17.9 19.2 
Alcohol intake (%) 22.3 19.0 17.9 26.4 
Ccr (ml/min) 65.6±7.6 66.2±7.8 53.9±7.4 57.6±8.4 
IVS (mm) 9.3±1.4 10.8±1.5† 9.3±1.4 11.7±1.7† 
PWT (mm) 9.1±1.1 11.5±1.4† 9.2±1.2 11.3±1.3† 
LVIDD (mm) 43.6±4.2 48.0±4.8† 43.3±4.2 47.0±5.0† 
LVISD (mm) 28.1±4.7 30.6±7.4† 28.3±4.4 30.3±6.4† 
RWT (%) 43.0±7.6 47.1±8.4† 42.8±7.3 48.9±8.3† 
LVM (g) 132.1±28.8 199.0±27.6† 129.6±31.0 200.3±33.2† 
LVMI (g/m2.737.5±5.4 59.5±9.6† 36.5±6.9 58.5±10.1† 
Hypertension treatment (%) 25.2 25.5 17.6 16.8 
Table 2
The distribution of the genotypes −T786C, +G894T and eNOS4a/b in the hypertensive patients

P<0.01 compared with hypertensive patients without LVH, as measured using the χ2 test.

−T786C [n (%)]+G894T [n (%)]eNOS4a/b [n (%)]
PatientsTTTCCCGGGTTT4b4a/b4a
Study 1          
 Patients without LVH (n=1118) 1019 (91.1) 96 (8.6) 3 (0.3) 634 (56.7) 411 (36.8) 73 (6.5) 990 (88.6) 122 (10.9) 6 (0.5) 
 Patients with LVH (n=1061) 982 (92.6) 76 (7.2) 3 (0.2) 506 (54.4) 434 (37.3) 121 (11.4)† 921 (86.8) 133 (12.5) 7 (0.7) 
Study 2          
 Patients without LVH (n=223) 189 (84.8) 31 (13.9) 3 (1.3) 123 (55.2) 79 (35.4) 21 (9.4) 193 (86.5) 30 (13.5) 0 (0) 
 Patients with LVH (n=120) 105 (87.5) 13 (10.8) 2 (1.7) 61 (50.8) 43 (35.8) 16 (13.3)† 101 (84.2) 19 (15.8) 0 (0) 
−T786C [n (%)]+G894T [n (%)]eNOS4a/b [n (%)]
PatientsTTTCCCGGGTTT4b4a/b4a
Study 1          
 Patients without LVH (n=1118) 1019 (91.1) 96 (8.6) 3 (0.3) 634 (56.7) 411 (36.8) 73 (6.5) 990 (88.6) 122 (10.9) 6 (0.5) 
 Patients with LVH (n=1061) 982 (92.6) 76 (7.2) 3 (0.2) 506 (54.4) 434 (37.3) 121 (11.4)† 921 (86.8) 133 (12.5) 7 (0.7) 
Study 2          
 Patients without LVH (n=223) 189 (84.8) 31 (13.9) 3 (1.3) 123 (55.2) 79 (35.4) 21 (9.4) 193 (86.5) 30 (13.5) 0 (0) 
 Patients with LVH (n=120) 105 (87.5) 13 (10.8) 2 (1.7) 61 (50.8) 43 (35.8) 16 (13.3)† 101 (84.2) 19 (15.8) 0 (0) 

In the first case-control study, only +G894T was found to be significantly associated with LVH (P<0.01). The TT genotype of the +894 locus conferred a 1.67-fold risk for LVH (95% CI, 1.19–2.36; P=0.002) after adjustment for age, gender, BMI, SBP, DBP, smoking status, alcohol consumption and glucose by multivariate logistic regression analysis (Table 3). The influence of +894T on LVH is in a recessive model. A general linear model-univariate test on +G894T showed that the subgroup of patients with LVH carrying the TT genotype had increased IVS (12.2±1.8 mm compared with 10.5±1.5 mm, P<0.01), LV PWT (13.0±3.4 mm compared with 12.0±5.2 mm, P<0.01), LVMI (69.3±14.5 g/m2.7 compared with 60.8±7.2 g/m2.7, P<0.01) and RWT (53.6±10.1% compared with 47.4±9.0%, P<0.01) compared with carriers of the G allele (GT+GG). No significant difference was found in LVIDD (P>0.05), age (P>0.05), SBP (P>0.05), DBP (P>0.05) and LVISD (LV end-systolic diameter) (P>0.05) among the patients with different genotypes (Table 4). These associations were not found in hypertensive patients without LVH (Table 5).

Table 3
The association of the +G894T genotype with LVH in the patients with hypertension

ORs and 95% CIs were calculated by use of multivariate logistic regression analyses. Adjusted ORs were stratified by age, gender, BMI, SBP, DBP, smoking status, alcohol consumption and glucose. *P<0.05 compared with patients without LVH in Study 2; and †P<0.01 compared with patients without LVH in Study 1.

Genotype [n (%)]TT compared with GT+GG
SubjectsGT+GGTTCrude OR (95% CI)Adjusted OR (95% CI)
Study 1     
 Patients without LVH (n=1118) 1045 (93.5) 73 (6.5) 
 Patients with LVH (n=1061) 940 (88.6) 121 (11.4) 1.77 (1.31–2.40)† 1.67 (1.19–2.36)† 
Study 2     
 Patients without LVH (n=223) 202 (90.6) 21 (9.4) 
 Patients with LVH (n=120) 104 (86.7) 16 (13.3) 1.43 (1.05–2.37)* 1.41 (1.01–2.28)* 
Genotype [n (%)]TT compared with GT+GG
SubjectsGT+GGTTCrude OR (95% CI)Adjusted OR (95% CI)
Study 1     
 Patients without LVH (n=1118) 1045 (93.5) 73 (6.5) 
 Patients with LVH (n=1061) 940 (88.6) 121 (11.4) 1.77 (1.31–2.40)† 1.67 (1.19–2.36)† 
Study 2     
 Patients without LVH (n=223) 202 (90.6) 21 (9.4) 
 Patients with LVH (n=120) 104 (86.7) 16 (13.3) 1.43 (1.05–2.37)* 1.41 (1.01–2.28)* 
Table 4
Echocardiographic variables according to genotype +G894T in the hypertensive patients with LVH

Values are means±S.D. General linear model-univariate analysis was performed, with each cardiac echocardiographic parameter used as a dependent variable, the independent variable was the variant of +G894T, and co-variables were age, gender, BMI, SBP, DBP, glucose and creatinine clearance rate. Bonferroni correction was used for correction of multiple comparisons. †P<0.01 compared with GT+GG.

The first populationThe second population
VariablesGT+GG (n=940)TT (n=121)GT+GG (n=104)TT (n=16)
IVS (mm) 10.5±1.5 12.2±1.8† 11.1±1.7 12.4±1.5† 
PWT (mm) 12.0±5.2 13.0±3.4† 11.3±1.3 12.1±1.0† 
LVIDD (mm) 48.2±2.3 47.1±4.4 6.6±3.0 48.3±4.84 
LVISD (mm) 30.3±3.7 30.2±7.5 30.2±4.1 29.0±3.1 
RWT (%) 47.4±9.0 53.6±10.1† 43.8±8.4 50.9±8.1† 
LVMI (g/m2.760.8±7.2 69.3±14.5† 56.6±8.4 70.8±10.9† 
Age (years) 60.1±8.7 58.1±8.5 58.8±7.2 59.6±7.5 
DBP (mmHg) 95.2±13.2 97.8±15.8 95.4±10.2 91.8±13.5 
SBP (mmHg) 162.8±29.7 165.4±27.5 158.7±18.2 162.7±18.1 
The first populationThe second population
VariablesGT+GG (n=940)TT (n=121)GT+GG (n=104)TT (n=16)
IVS (mm) 10.5±1.5 12.2±1.8† 11.1±1.7 12.4±1.5† 
PWT (mm) 12.0±5.2 13.0±3.4† 11.3±1.3 12.1±1.0† 
LVIDD (mm) 48.2±2.3 47.1±4.4 6.6±3.0 48.3±4.84 
LVISD (mm) 30.3±3.7 30.2±7.5 30.2±4.1 29.0±3.1 
RWT (%) 47.4±9.0 53.6±10.1† 43.8±8.4 50.9±8.1† 
LVMI (g/m2.760.8±7.2 69.3±14.5† 56.6±8.4 70.8±10.9† 
Age (years) 60.1±8.7 58.1±8.5 58.8±7.2 59.6±7.5 
DBP (mmHg) 95.2±13.2 97.8±15.8 95.4±10.2 91.8±13.5 
SBP (mmHg) 162.8±29.7 165.4±27.5 158.7±18.2 162.7±18.1 
Table 5
Echocardiographic variables according to genotype +G894T in the hypertensive patients without LVH

Values are means±S.D. General linear model-univariate analysis was performed, with each cardiac echocardiographic parameter used as a dependent variable, the independent variable was the variant of +G894T, and co-variables were age, gender, BMI, SBP, DBP, glucose and creatinine clearance rate. Bonferroni correction was used for correction of multiple comparisons.

The first populationThe second population
VariablesGT+GG (n=1045)TT (n=73)GT+GG (n=202)TT (n=21)
IVS (mm) 9.3±1.3 9.5±1.3 9.3±1.4 9.4±1.5 
PWT (mm) 9.1±1.0 9.3±1.2 9.2±1.3 9.1±1.0 
LVIDD (mm) 44.0±4.5 43.6±5.0 42.8±3.5 42.1±4.4 
LVISD (mm) 28.7±4.6 28.1±5.3 28.8±2.8 29.3±2.0 
RWT (%) 42.3±7.2 43.7±7.6 42.9±7.8 42.1±6.4 
LVMI (g/m2.737.9±6.1 37.2±6.2 36.6±6.9 36.6±6.8 
Age (years) 57.8±9.3 57.4±8.9 58.3±7.2 56.9±7.5 
DBP (mmHg) 95.2±11.4 97.9±11.3 95.4±10.2 91.8±10.0 
SBP (mmHg) 158.8±22.3 161.9±22.8 156.2±17.2 158.6±19.1 
The first populationThe second population
VariablesGT+GG (n=1045)TT (n=73)GT+GG (n=202)TT (n=21)
IVS (mm) 9.3±1.3 9.5±1.3 9.3±1.4 9.4±1.5 
PWT (mm) 9.1±1.0 9.3±1.2 9.2±1.3 9.1±1.0 
LVIDD (mm) 44.0±4.5 43.6±5.0 42.8±3.5 42.1±4.4 
LVISD (mm) 28.7±4.6 28.1±5.3 28.8±2.8 29.3±2.0 
RWT (%) 42.3±7.2 43.7±7.6 42.9±7.8 42.1±6.4 
LVMI (g/m2.737.9±6.1 37.2±6.2 36.6±6.9 36.6±6.8 
Age (years) 57.8±9.3 57.4±8.9 58.3±7.2 56.9±7.5 
DBP (mmHg) 95.2±11.4 97.9±11.3 95.4±10.2 91.8±10.0 
SBP (mmHg) 158.8±22.3 161.9±22.8 156.2±17.2 158.6±19.1 

The second case-control study confirmed the association of the TT genotype of the +894 locus with LVH [OR, 1.41 (95% CI, 1.01–2.28); P=0.01] after adjustment for conventional risk factors (Table 3). The influence was also in a recessive model. A general linear model-univariate analysis of +G894T showed that the subgroup of patients with LVH carrying the TT genotype had increased IVS (12.4±1.5 mm compared with 11.1±1.5 mm, P<0.01), LV PWT (12.1±1.0 mm compared with 11.3±1.3 mm, P<0.01), LVMI (70.8±10.9 g/m2.7 compared with 56.6±8.4g/m2.7, P<0.01) and RWT (50.9±8.1% compared with 43.8±8.4%, P<0.01) compared with the G allele carriers (GT+GG). No significant difference was found in LVIDD (P>0.05), age (P>0.05), SBP (P>0.05), DBP (P>0.05) and LVISD (P>0.05) among the patients with different genotypes (Tables 4 and 5).

A SHEsis software platform was used to analyse pair-wise LD. We found that the three NOS3 genetic variants were not in LD between −T786C (rs2070744) and eNOS4a/b (D′=0.659, r2=0.272), or between −T786C (rs2070744) and +G894T (rs1799983) (D′=0.675, r2=0.007), or between eNOS4a/b and +G894T (rs1799983) (D′=0.719, r2=0.013).

DISCUSSION

The present study provides evidence that +G894T, a common variant of NOS3, is associated with an increased risk of LVH in hypertensive patients in a recessive model. Multivariate adjustments support that the effects are independent of common traditional cardiovascular risk factors, including BP levels, age, gender and other clinical factors that influence cardiac size. To the best of our knowledge, the present study is the first investigation to test the association of the variation +G894T with LVH in large-scale community-based cohorts, and the results were replicated in a second independent population.

In recent years the regulation of the expression of eNOS in endothelial cells has become clearer. This regulatory cycle involves a series of post-translational modifications and intracellular compartmentalization to allow for co-ordinated substrate availability, calcium-dependent activation, enzyme catalysis and NO release [23,24]. Of these steps, specific localization of eNOS with caveolae is critical for enzyme regulation and activity. In endothelial cells, Cav-1 (caveolin-1) is the major protein constituent of caveolae; this protein has been thought to hold or ‘store’ eNOS in an inactive form wherein phosphorylation causes dissociation and eNOS activation. Basal eNOS bound to Cav-1 is therefore essential for its activation, leading to the regulation of endothelial cell NO production [25].

In a previous study, it has been reported that −T786C in the promoter region of NOS3 affects the promoter activity of NOS3 in vitro [26]. eNOS4a/b influences NOS3 mRNA and protein levels. All of these NOS3 variants have been shown to be significantly associated with plasma NO concentration in vivo [27]. The +G894T variant of NOS3 results in a substitution of an aspartate for glutamate residue at amino acid position 298 of the eNOS protein. Some studies have shown that the eNOS protein containing an aspartate residue at position 298 is more susceptible to cleavage by proteases, which results in eNOS dysfunction [28,29]. Other studies questioned the results as an artefact caused by Western blotting preparation [30]. A recent study, however, has found evidence that eNOS production is dysregulated in subjects carrying the TT compared with those carrying the GG genotype [31]. The primary difference is an altered localization of the protein at caveolae, leading to diminished shear-dependent responses and impaired co-ordination of the eNOS regulatory cycle [31]. The results of the present study do not support associations of −T786C and eNOSa/b of the eNOS gene with increased susceptibility for the development of LVH in hypertensive patients. We found that only the +G894T variant of the three variants tested was associated with a greater risk of LVH in hypertensive patients.

Our observations are consistent with the results of Lapu-Bula et al. [32] in that the 894T variant of the eNOS gene contributes to an increased susceptibility to LVH and a high to normal BP in African-Americans. Our results did not support the observation by Minushkina et al. [33], which shows that the G allele of NOS3 is a risk allele for LVH in patients with essential hypertension. In addition, eNOS4a/b has been reported to be associated with higher IVS in Turkish hypertensive patients [34]. The difference between the studies described above and the present study could be due to (i) the different genetic background of the two populations, (ii) the relatively small sample size confounds the result in the Minushkina et al. [33] study (the total number of patients with essential hypertension were 109, of them, 76 had LVH), and (iii) the subjects in the Minushkina et al. [33] study were recruited from clinics, whereas the patients in the present study were recruited from communities.

We also examined whether the disease-associated alleles were related to specific vascular risk factors, including hypertension, diabetes, age and gender. After adjusting for those conventional cardiovascular risk factors, the association remained significant, suggesting that the contribution of this variant to the risk of LVH is independent of conventional cardiovascular risk factors.

The strength of the present study is the large sample size, and community-based sample. Nevertheless, there are some limitations in the present study, including that the size of the second case-control study was relatively small and all study subjects were hypertensive patients. However, because the second study replicated the first study results, it strengthens the likelihood that the association is a true association. A longitudinal follow-up study is warranted to elucidate whether there are protective effects of NOS3 variants on cardiac hypertrophy in the future, considering the powerful ability of LVH to predict morbidity and mortality.

In conclusion, a genetic variant of NOS3 could contribute to LVH in patients with hypertension. The results need to be replicated in other nationalities and confirmed with a prospective trial. The mechanisms need to be further explored.

FUNDING

This work was supported by the Ministry of Science and Technology of China [grant numbers 2006CB503805, 2006DFA31500 (to R. H.)].

Abbreviations

     
  • BMI

    body mass index

  •  
  • BP

    blood pressure

  •  
  • Cav-1

    caveolin-1

  •  
  • CI

    confidence interval

  •  
  • DBP

    diastolic BP

  •  
  • IVS

    interventricular septum thickness

  •  
  • LD

    linkage disequilibrium

  •  
  • LV

    left ventricular

  •  
  • LVH

    LV hypertrophy

  •  
  • LVIDD

    LV end-diastolic diameter

  •  
  • LVISD

    LV end-systolic diameter

  •  
  • LVM

    LV mass

  •  
  • LVMI

    LVM index

  •  
  • NOS

    NO synthase

  •  
  • eNOS

    endothelial NOS

  •  
  • OR

    odds ratio

  •  
  • PWT

    posterior wall thickness

  •  
  • RWT

    relative wall thickness

  •  
  • SBP

    systolic BP

References

References
1
Schillaci
 
G.
Verdecchia
 
P.
Porcellati
 
G.
Cuccurullo
 
O.
Cosco
 
C.
Perticone
 
F.
 
Continuous relation between left ventricular mass and cardiovascular risk in essential hypertension
Hypertension
2000
, vol. 
35
 (pg. 
580
-
586
)
2
Levy
 
D.
Garrison
 
R. J.
Savage
 
D. D.
Kannel
 
W. B.
Castelli
 
W. P.
 
Prognostic implications of echocardiographically determined left ventricular mass in the Framingham Heart Study
N. Engl. J. Med.
1990
, vol. 
322
 (pg. 
1561
-
1566
)
3
Galderisi
 
M.
Anderson
 
K. M.
Wilson
 
P. W.
Levy
 
D.
 
Echocardiographic evidence for the existence of a distinct diabetic cardiomyopathy (the Framingham Heart Study)
Am. J. Cardiol.
1991
, vol. 
68
 (pg. 
85
-
89
)
4
Deschepper
 
C. F.
Boutin-Ganache
 
I.
Zahabi
 
A.
Jiang
 
Z.
 
In search of cardiovascular candidate genes: interactions between phenotypes and genotypes
Hypertension
2002
, vol. 
39
 (pg. 
332
-
336
)
5
Barouch
 
L. A.
Harrison
 
R. W.
Skaf
 
M. W.
Rosas
 
G. O.
Cappola
 
T. P.
Kobeissi
 
Z. A.
Hobai
 
I. A.
Lemmon
 
C. A.
Burnett
 
A. L.
O'Rourke
 
B.
, et al 
Nitric oxide regulates the heart by spatial confinement of nitric oxide synthase isoforms
Nature
2002
, vol. 
416
 (pg. 
337
-
339
)
6
Kim
 
N. N.
Villegas
 
S.
Summerour
 
S. R.
Villarreal
 
F. J.
 
Regulation of cardiac fibroblast extracellular matrix production by bradykinin and nitric oxide
J. Mol. Cell Cardiol.
1999
, vol. 
31
 (pg. 
457
-
466
)
7
Murohara
 
T.
Asahara
 
T.
Silver
 
M.
Bauters
 
C.
Masuda
 
H.
Kalka
 
C.
Kearney
 
M.
Chen
 
D.
Symes
 
J. F.
Fishman
 
M. C.
, et al 
Nitric oxide synthase modulates angiogenesis in response to tissue ischemia
J. Clin. Invest.
1998
, vol. 
101
 (pg. 
2567
-
2578
)
8
Ritchie
 
R. H.
Schiebinger
 
R. J.
LaPointe
 
M. C.
Marsh
 
J. D.
 
Angiotensin II-induced hypertrophy of adult rat cardiomyocytes is blocked by nitric oxide
Am. J. Physiol.
1998
, vol. 
275
 (pg. 
H1370
-
H1374
)
9
Sládek
 
T.
Gerová
 
M.
Znojil
 
V.
Devát
 
L.
 
Morphometric characteristics of cardiac hypertrophy induced by long-term inhibition of NO synthase
Physiol. Res.
1996
, vol. 
45
 (pg. 
335
-
338
)
10
Ichinose
 
F.
Bloch
 
K. D.
Wu
 
J. C.
Hataishi
 
R.
Aretz
 
H. T.
Picard
 
M. H.
Scherrer-Crosbie
 
M.
 
Pressure overload induced LV hypertrophy and dysfunction in mice are exacerbated by congenital NOS3 deficiency
Am. J. Physiol. Heart Circ. Physiol.
2004
, vol. 
286
 (pg. 
H1070
-
H1075
)
11
Buys
 
E. S.
Raher
 
M. J.
Blake
 
S. L.
Neilan
 
T. G.
Graveline
 
A. R.
Passeri
 
J. J.
Llano
 
M.
Perez-Sanz
 
T. M.
Ichinose
 
F.
Janssens
 
S.
, et al 
Cardiomyocyte-restricted restoration of nitric oxide synthase 3 attenuates left ventricular remodeling after chronic pressure overload
Am. J. Physiol. Heart Circ. Physiol.
2007
, vol. 
293
 (pg. 
H620
-
H627
)
12
Janssens
 
S.
Pokreisz
 
P.
Schoonjans
 
L.
Pellens
 
M.
Vermeersch
 
P.
Tjwa
 
M.
Jans
 
P.
Scherrer-Crosbie
 
M.
Picard
 
M. H.
Szelid
 
Z.
, et al 
Cardiomyocyte-specific overexpression of nitric oxide synthase 3 improves left ventricular performance and reduces compensatory hypertrophy after myocardial infarction
Circ. Res.
2004
, vol. 
94
 (pg. 
1256
-
1262
)
13
Stein
 
B.
Eschenhagen
 
T.
Rüdiger
 
J.
Scholz
 
H.
Förstermann
 
U.
Gath
 
I.
 
Increased expression of constitutive nitric oxide synthase III, but not inducible nitric oxide synthase II, in human heart failure
J. Am. Coll. Cardiol.
1998
, vol. 
32
 (pg. 
1179
-
1186
)
14
Hingorani
 
A. D.
Liang
 
C. F.
Fatibene
 
J.
Parsons
 
A.
Hopper
 
R. V.
Trutwein
 
D.
Stephens
 
N. G.
O'shaughnessy
 
K. M.
Brown
 
M. J.
 
A common variant of the endothelial nitric oxide synthase (Glu298>Asp) is a major risk factor for coronary artery disease in the UK
Circulation
1999
, vol. 
100
 (pg. 
1515
-
1520
)
15
Whitworth
 
J. A.
 
2003 World Health Organization (WHO)/International Society of Hypertension (ISH) statement on management of hypertension
J. Hyperten.
2003
, vol. 
21
 (pg. 
1983
-
1992
)
16
Tajik
 
A. J.
Seward
 
J. B.
Hagler
 
D. J.
Mair
 
D. D.
Lie
 
J. T.
 
Two-dimensional real-time ultrasonic imaging of the heart and great vessels
Mayo Clin. Proc.
1978
, vol. 
53
 (pg. 
271
-
303
)
17
Sahn
 
D. J.
DeMaria
 
A.
Kisslo
 
J.
Weyman
 
A.
 
Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements
Circulation
1978
, vol. 
58
 (pg. 
1072
-
1083
)
18
Devereux
 
R. B.
Alonso
 
D. R.
Lutas
 
E. M.
Gottlieb
 
G. J.
Campo
 
E.
Sachs
 
I.
Reichek
 
N.
 
Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy finding
Am. J. Cardiol.
1986
, vol. 
57
 (pg. 
450
-
458
)
19
De Simone
 
G.
Devereux
 
R. B.
Daniels
 
S. R.
Koren
 
M. J.
Meyer
 
R. A.
Laragh
 
J. H.
 
Effect of growth on variability of left ventricular mass: assessment of allometric signals in adults and children and their capacity to predict cardiovascular risk
J. Am. Coll. Cardiol.
1995
, vol. 
25
 (pg. 
1056
-
1062
)
20
Ganau
 
A.
Devereux
 
R. B.
Roman
 
M. J.
de Simone
 
G.
Pickering
 
T. G.
Saba
 
P. S.
Vargiu
 
P.
Simongini
 
I.
Laragh
 
J. H.
 
Patterns of left ventricular hypertrophy and geometric remodeling in essential hypertension
J. Am. Coll. Cardiol.
1992
, vol. 
19
 (pg. 
1550
-
1558
)
21
Cockcroft
 
D. W.
Gauh
 
M. H.
 
Prediction of creatinine clearance from serum creatinine
Nephron
1976
, vol. 
16
 (pg. 
31
-
41
)
22
Leadon
 
S. A.
Cerutti
 
P. A.
 
A rapid and mild procedure of the isolation of DNA from mammalian cells
Anal. Biochem.
1982
, vol. 
120
 (pg. 
282
-
288
)
23
Garcia-Cardena
 
G.
Martasek
 
P.
Masters
 
B. S.
Skidd
 
P. M.
Couet
 
J.
Li
 
S.
Lisanti
 
M. P.
Sessa
 
W. C.
 
Dissecting the interaction between nitric oxide synthase (NOS) and caveolin. Functional significance of the NOS caveolin binding domain in vivo
J. Biol. Chem.
1997
, vol. 
272
 (pg. 
25437
-
25440
)
24
Shaul
 
P. W.
Anderson
 
R. G.
 
Role of plasmalemmal caveolae in signal transduction
Am. J. Physiol.
1998
, vol. 
275
 (pg. 
L843
-
L851
)
25
Govers
 
R.
Rabelink
 
T. J.
 
Cellular regulation of endothelial nitric oxide synthase
Am. J. Physiol. Renal Physiol.
2001
, vol. 
280
 (pg. 
F193
-
F206
)
26
Nakayama
 
M.
Yasue
 
H.
Yoshimura
 
M.
Shimasaki
 
Y.
Kugiyama
 
K.
Ogawa
 
H.
Motoyama
 
T.
Saito
 
Y.
Ogawa
 
Y.
Miyamoto
 
Y.
 
T−786>C mutation in the 5′-flanking region of endothelial nitric oxide synthase gene is associated with coronary spasm
Circulation
1999
, vol. 
99
 (pg. 
2864
-
2870
)
27
Wang
 
X. L.
Sim
 
A. S.
Wang
 
M. X.
Murrell
 
G. A. C.
Trudinger
 
B.
Wang
 
J.
 
Genotype dependent and cigarette specific effects on endothelial nitric oxide synthase gene expression and enzyme activity
FEBS Lett.
2000
, vol. 
471
 (pg. 
45
-
50
)
28
Tesauro
 
M.
Thompson
 
W. C.
Rogliani
 
P.
Qi
 
L.
Chaudhary
 
P. P.
Moss
 
J.
 
Intracellular processing of endothelial nitric oxide synthase isoforms associated with differences in severity of cardiopulmonary diseases: cleavage of proteins with aspartate versus glutamate at position 298
Proc. Natl. Acad. Sci. U.S.A.
2000
, vol. 
97
 (pg. 
2832
-
2835
)
29
Ahmadi
 
K. R.
Weale
 
M. E.
Xue
 
Z. Y.
Soranzo
 
N.
Yarnall
 
D. P.
Briley
 
J. D.
Maruyama
 
Y.
Kobayashi
 
M.
Wood
 
N. W.
Spurr
 
N. K.
, et al 
A single-nucleotide polymorphism tagging set for human drug metabolism and transport
Nat. Genet.
2005
, vol. 
37
 (pg. 
84
-
89
)
30
Fairchild
 
T. A.
Fulton
 
D.
Fontana
 
J. T.
Gratton
 
J. P.
McCabe
 
T. J.
Sessa
 
W. C.
 
Acidic hydrolysis as a mechanism for the cleavage of the Glu298-Asp variant of human endothelial nitric-oxide synthase
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
26674
-
26679
)
31
Joshi
 
M. S.
Mineo
 
C.
Shaul
 
P. W.
Bauer
 
J. A.
 
Biochemical consequences of the NOS3 Glu298Asp variation in human endothelium: altered calveolar localization and impaired response to shear
FASEB J.
2007
, vol. 
21
 (pg. 
2655
-
2663
)
32
Lapu-Bula
 
R.
Quarshie
 
A.
Lyn
 
D.
Oduwole
 
A.
Pack
 
C.
Morgan
 
J.
Nkemdiche
 
S.
Igho-Pemu
 
P.
Onwuanyi
 
A.
Li
 
R.
Ofili
 
E.
 
The 894T allele of endothelial nitric oxide synthase gene is related to left ventricular mass in African Americans with high-normal blood pressure
J. Natl. Med. Assoc.
2005
, vol. 
97
 (pg. 
197
-
205
)
33
Minushkina
 
L. O.
Zateĭshchikov
 
D. A.
Zateĭshchikova
 
A. A.
Zotova
 
I. V.
Kudriashova
 
O. Y.
Nosikov
 
V. V.
Sidorenko
 
B. A.
 
NOS3 gene polymorphism and left ventricular hypertrophy in patients with essential hypertension
Kardiologiia
2002
, vol. 
42
 (pg. 
30
-
34
)
34
Olcay
 
A.
Ekmekci
 
C. G.
Ozbek
 
U.
Sezer
 
M.
Barcin
 
C.
Arslan
 
E.
Boztosun
 
B.
Nisanci
 
Y.
 
Negative association of endothelial nitric oxide gene polymorphism with hypertension in Turkish patients: effect of eNOS polymorphism on left ventricular hypertrophy
Cardiovasc. Ultrasound
2006
, vol. 
4
 pg. 
33