Association between endothelial nitric oxide synthase gene polymorphisms and risk of ischemic stroke: A meta-analysis
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.198170
Source of Support: None, Conflict of Interest: None
Previously published studies that have examined whether the three polymorphisms, G894T, T786C, and 4b/a in the endothelial nitric oxide synthase (eNOS) gene, are associated with ischemic stroke (IS) have reported conflicting results. Thus, we performed a meta-analysis to examine the potential association between these three single nucleotide polymorphisms (SNPs) of the eNOS gene and IS risk. A literature search was carried out for eligible candidate gene studies published before August 05, 2015 in the PubMed, Embase, and Google Scholar databases. The following combinations of main keywords were used in our study: ('endothelial nitric oxide synthase') or ('eNOS') and ('G894T, 4b/a, and T786C') and ('polymorphism') or ('polymorphisms') and ('Ischemic Stroke' or 'IS') and ('Cerebral Infarction' or 'CI') and ('genetic polymorphism' or 'single nucleotide polymorphisms' or 'SNP'). Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were estimated by using fixed or random effects model. Meta-regression analysis was used to investigate the potential sources of heterogeneity. Begg's funnel plots were used to explore the publication bias, and heterogeneity was assessed by I2 test. Twenty seven case-control studies involving 6733 cases and 7305 controls were analyzed in our meta-analysis. Significant association was observed for G894T (OR = 1.17; 95% CI: 1.08 to 1.28; P< 0.001) and 4b/a (OR = 1.25; 95% CI: 1.13 to 1.39; P < 0.001) whereas a non-significant association was observed for T786C (OR = 1.11; 95% CI: 0.98 to 1.26; P =0.109) eNOS gene polymorphisms and IS. Our meta-analysis establishes that the G894T and 4b/a polymorphisms of eNOS gene are significantly associated with the risk of IS. However, a non-significant association was found between T786C polymorphism of the eNOS gene and IS risk. Further prospective large epidemiological studies need to be done to confirm these findings.
Keywords: Endothelial nitric oxide synthase, gene polymorphism, ischemic stroke, meta-analysis
Stroke is the second major leading cause of death and adult disability after ischemic heart disease (IHD). Stroke accounts for nearly 5.7 million deaths globally, and 87% of these deaths occurr in low and middle income nations. The incidence of stroke in South Asian countries has increased by more than 100%, while this has decreased by 42% in developed European countries in the last four decades., Stroke is a multi-factorial disease; epidemiological and animal studies have strongly recommended the role of genetic influence in the pathogenesis of ischemic stroke (IS). The genetic influences are probably polygenic, and IS itself has a number of different subtypes, which may each have different pathogenesis. This increase in the incidence of stroke in developing countries could have been influenced by both environmental and genetic factors.
In addition to genetic variants, non-genetic variants including environmental and lifestyle risk factors may also contribute to the risk of developing complex diseases such as stroke. Lifestyle risk factors such as diet, exercise, smoking, and alcohol consumption can play a large role in determining whether or not someone develops a disease. Family history is also an important tool for determining the risk of developing stroke. The genetic and non-genetic factors help us to identify the risk factors that can take control over the risk of stroke.
The eNOS gene is located on chromosome 7 (7q35-q36) and consists of 26 exons. It codes for an enzyme that generates nitric oxide (NO) in the vascular endothelium. NO mediates vasodilation in the endothelium, inhibits adhesion of platelets and leukocytes and limits oxidation of atherogenic low-density lipoproteins in the vascular endothelium. Impaired endothelium-dependent vasodilation is a common feature of atherosclerotic vessels, which appears to be partly due to the reduction in the activity of vascular endothelial nitric oxide synthase. Impaired nitric oxide-dependent vasomotor reactivity has been implicated in the pathophysiology of stroke. Because it has an important role in the physiology of the vasculature, genetic variations could alter the expression and activity of eNOS, thereby contributing to the development of stroke. A meta-analysis  published in 2009 suggested that TT genotype of G894T polymorphism is not associated with the risk of IS. Studies published in the literature examining the association between eNOS polymorphism and IS have shown conflicting results. Discrepancy in the methodological quality of studies may explain the presence of the conflicting results. No meta-analysis till date has explored the association of methodological quality and outcome. Large epidemiological studies have been performed explaining the relationship of the eNOS gene polymorphisms (G894T, T786C, 4b/a) and the risk of IS; however, the results are still controversial and not conclusive. Because a single study neither has a large number of sample size nor has the power to draw profound conclusions, carrying out a meta-analysis seemed to be an appropriate approach in order to get solid conclusive results. Therefore, the aim of this study was to perform an updated meta-analysis of the association between eNOS gene polymorphisms and the risk of IS, as well as to investigate whether or not the methodological quality of study affects the results.
A literature search was carried out for eligible candidate gene studies published till August 05, 2015 using the PubMed, Embase, and Google Scholar databases. The following combinations of main keywords were used in our study: ('endothelial nitric oxide synthase') or ('eNOS') and ('G894T, 4b/a, and T786C') and ('polymorphism') or ('polymorphisms') and ('ischemic stroke' or 'IS') and ('cerebral infarction' or 'CI') and ('genetic polymorphism' or 'single nucleotide polymorphisms' or 'SNP'). The search was performed without any boundary on language; however, we selected only those studies that were conducted on human adult subjects. In order to identify additionally published articles, all references were reviewed carefully. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were estimated by using fixed or random effects model. Meta-analysis was carried out using STATA, version 13.0 (StataCorp. 2013. Stata Statistical Software: Release 13. College Station, TX: StataCorp LP).
Inclusion and exclusion criteria
For inclusion in the meta-analysis, the following criteria were set: (1) A case-control study investigating the association between G894T in exon 7 or 4b/a in intron 4 or T786C (in the promoter region) genetic polymorphisms of the eNOS gene and susceptibility to IS; (2) cases meeting the diagnostic criteria for IS; (3) studies reporting a sufficient genotypic frequency for data required for meta-analysis. The major reasons for excluding studies were: (1) Not a case-control study; (2) duplicate publications were present with overlapping subjects from the same study; and, (3) no available data was reported. This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guideline.
As per the PRISMA guidelines, each full-text article was checked for eligibility by two authors (AK and SM) independently and the following data was extracted from the eligible studies:First author surname, publication year, country, number of cases and controls, population, genotyping method, age, sex, genotype frequency, etc., Disagreements were resolved by discussion among all the authors until a consensus was obtained.
We also examined the methodological quality of every study which was included in our meta-analysis using a quality assessment scale developed for genetic association studies, which was modified by us to increase the relevance of our study. Both the traditional epidemiological considerations and genetic issues were taken into consideration by using this scale. The scores ranged from 0 (worst) to 17 (best). The contents of the quality scale are illustrated in [Table 1]. The quality of this study was independently assessed by two authors (AK and SM). Inconsistency over the quality scores were resolved by discussion among all the authors and a subsequent agreement was reached.
Hardy–Weinberg equilibrium (HWE) was tested for genotype distributions in the controls by using the chi-square test. The association between G894T or T786C or 4b/ a genetic polymorphisms of eNOS gene and susceptibility to IS was assessed by the pooled ORs with their corresponding 95% confidence intervals (95% CIs) under three genetic models, consisting of dominant, recessive, and allelic models. After considering the possible heterogeneity between studies, a statistical test was conducted for heterogeneity by using Cochran's Q statistic and I 2 metric. In our study, the I 2 values exceeding 50% and heterogeneity at the 10% level of significance were considered to be indicators of significant heterogeneity. Begg's funnel plot was used to assess the potential for a publication bias. Stratified analyses were done according to two major ethnic groups/populations. Meta-regression analysis was done to check if the methodological quality of studies was significantly associated with the size of the observed effect.
A total of 59 published articles were identified by using the pre-specified search strategy. [Figure 1] displays the study flow chart. Out of the 59 articles which were retrieved, 19 studies were excluded due to their irrelevance to our interest, and 5 studies were excluded because they were in duplicate records. As per the inclusion criteria, 27 case-control studies were included in our meta-analysis. Sixteen studies were in the Asian population, whereas 11 studies were in the Caucasian population. The publication years of the studies included in this meta-analysis ranged from 1998 to 2014. Twenty four studies in this meta-analysis had controls in accordance with Hardy-Weinberg Equilibrium (HWE). The quality score of all the included studies was observed to be moderately high. Out of the 27 studies, 14 had hospital-based source of control, 9 studies had population-based source of control, and 4 studies did not report the source of control. [Table 2] gives a summary of the characteristics and methodological quality of all the included studies.
Association between G894T polymorphism of eNOS gene and ischemic stroke
The eNOS G894T gene polymorphism was assessed in 19 case-control studies with a total of 4505 IS cases and 5038 controls. Borderline significant association was observed for G894T polymorphism of eNOS gene and IS under dominant [GT + TT vs GG: OR = 1.22; 95% CI: 0.99 to 1.50; P = 0.06] and allelic [T vs G: OR = 1.23; 95% CI: 0.95 to 1.59; P < 0.12] models. Significant association was observed for G894T polymorphism of eNOS gene and IS under recessive model [TT vs GT + GG: OR = 1.21; 95% CI: 1.03 to 1.41; P = 0.019]. In the subgroup analysis based on ethnicity, significant association between G894T polymorphism of eNOS gene and risk of IS was observed in 9 Asian studies under dominant (OR = 1.57, 95% CI: 1.13 to 2.19, P = 0.008), recessive (OR = 1.67, 95% CI: 1.22 to 2.28, P = 0.001) and allelic (OR = 1.62, 95% CI: 1.04 to 2.54, P = 0.03) models. However, no significant association was observed in 10 Caucasian studies under any of the three models [Figure 2].
Association between T786C polymorphism of eNOS gene and ischemic stroke
The eNOS T786C gene polymorphism was assessed in 10 case-control studies with a total of 2348 IS cases and 2879 controls. Borderline significant association was observed for T786C polymorphism of eNOS gene and IS under dominant [CT + CC vs TT: OR = 1.11; 95% CI; 0.98 to 1.26, P = 0.10] and recessive [CC vs CT + TT: OR = 1.58, 95% CI: 0.93 to 2.66, P = 0.08] models, whereas significant association was observed under allelic model [C vs T: OR = 1.18, 95% CI: 1.04 to 1.34, P = 0.011].
In the subgroup analysis based on ethnicity, significant association between T786C polymorphism of eNOS gene and risk of IS was observed in 7 Asian studies under dominant (OR = 1.17, 95% CI: 1.01 to 1.35, P = 0.03) and allelic (OR = 1.25, 95% CI: 1.09 to 1.45, P = 0.002) models but not under the recessive model. No significant association was observed in the 3 Caucasian studies under any of the three models [Figure 3].
Association between 4b/a polymorphism of eNOS gene and ischemic stroke
The eNOS 4b/a gene polymorphism was assessed in 16 case-control studies with a total of 3730 IS cases and 4220 controls. Significant association was observed for 4b/a polymorphism of eNOS gene and susceptibility to IS under recessive model [aa vs ab + bb: OR = 1.65, 95% CI: 1.01 to 2.72, P = 0.04]. Borderline significant association was observed for 4b/a polymorphism of eNOS gene and susceptibility to IS under dominant [aa + ab vs bb: OR = 1.16; 95% CI; 0.95 to 1.43, P = 0.15] and allelic [a vs b: OR = 1.18, 95% CI: 0.91–1.53, P = 0.21] models.
In the subgroup analysis based on ethnicity, significant association between 4b/a polymorphism of eNOS gene and risk of IS was observed in the 12 Asian studies under dominant (OR = 1.31, 95% CI: 1.07 to 1.61, P = 0.009), recessive (OR = 2.45, 95% CI: 1.72 to 3.49, P = <0.001) and allelic (OR = 1.42, 95% CI: 1.11 to 1.81, P = 0.006) models. No significant association was observed in 4 Caucasian studies under any of the three models [Figure 4]. Begg's funnel plot suggested the presence of moderate publication bias in the studies [Figure 5].
Meta-regression analysis suggested that the quality of studies did not significantly affect the strength of association between eNOS gene polymorphisms and risk of IS [Figure 6].
The present meta-analysis suggests that the eNOS gene is an important candidate gene because of its association with IS in the Asian population but not in the Caucasian population. NO is known to be a potent endogenous vasodilator which inhibits the platelet aggregation and reduces the adherence of leukocytes to vascular endothelium. In addition, it suppresses the proliferation of vascular smooth muscle cells. In this regard, reduced NO production has been found to be linked with vascular dysfunction. The elevation of blood pressure in healthy individuals has been reported due to the inhibition of eNOS, whereas the production of NO is found to be diminished in patients with essential hypertension under basal conditions. The eNOS gene catalyzes the synthesis of NO, which is a regulator of cerebral blood flow as well as of smooth muscle cell proliferators in the arterial endothelium. Among the eNOS variants, G894T is the most studied variant and is found to be associated with hypertension in Caucasian as well as Asian populations., It is widely accepted that the polymorphism in eNOS gene modifies the NO production which may predispose the patient to hypertension, which is a prominent risk factor for IS.
Biochemical evidence exists for the decreased eNOS gene promoter activation, reduced protein expression, and reduced enzyme activation because the precise molecular effects of polymorphisms in the eNOS gene might be linked to the pathophysiology of stroke. It is a well-established fact that endothelial dysfunction occurs in response to cardiovascular risk factors and precedes the development of atherosclerosis. The T786C polymorphism in the 5-flanking region of the eNOS gene probably influences the promoter activity and thereby the endothelial synthesis of NO. A study  has demonstrated that T786C polymorphism in the 5-flanking region of the eNOS gene is in linkage disequilibrium with 4b/a polymorphism.
The association between eNOS gene polymorphism and risk of IS has not yet been well-established because conflicting evidence is present in the literature. Six published meta-analyses that examined the association between eNOS gene polymorphisms and risk of stroke also showed conflicting evidence in the results [Table 3]. The possible reason for inconsistent results include (i) variation in the methodology, (ii) lack of proper selection of studies, (iii) inappropriate or lack of exploration of possible cause of heterogeneity in inter-study, and (iv) publication bias. In the present meta-analysis, we have taken appropriate measures to overcome the above shortcomings.
G894T polymorphism of eNOS gene
Five previously published meta-analyses suggest a significant association between G894T polymorphism and risk of IS with an OR ranging from 1.3 to 2.01 under the dominant model of inheritance in the Asian population but not in the Caucasian population. The present meta-analysis also observed a significant association in the Asian population under the dominant model of inheritance even after assessing the heterogeneity using meta-regression analysis. The present meta-analysis also found an association with the risk of IS under the recessive model in the Asian population, which is consistent with the findings of the earlier published meta-analysis. The present meta-analysis also confirms the significant association of G894T polymorphism with risk of stroke in the Asian population under the allelic model.
T786C polymorphism of eNOS gene
Significant association was observed between T786C polymorphism and risk of IS in a few earlier published meta-analyses; however, other studies failed to observe a significant association [Table 3]. Our meta-analysis suggests that T786C polymorphism significantly contributes to the risk of IS in Asians but not in Caucasians.
Most of the meta-analyses observed an association between 4b/a polymorphism and risk of IS in the Asian population. The present meta-analysis also confirms a significant association between 4b/a polymorphism and the risk of IS in the Asian population but not in the Caucasian population.
A meta-analysis that included 33 studies demonstrated that eNOS (G894T, 4b/a, and T786C) gene polymorphisms are associated with the risk of IS in Asians but not in Caucasians. Another meta-analysis provided strong evidence that the eNOS T786C polymorphism is not associated with the risk of IS whereas the 4b/a and G894T polymorphisms might be associated with IS, at least in the Asian population. A meta-analysis including 12 association studies with 2836 ischemic stroke and 3354 controls demonstrated that T786C polymorphism is associated with the risk of stroke mainly in Asians among the population aged 60–65 years in particular, but not in the Caucasian population. Another meta-analysis demonstrated that T786C polymorphism may not be associated with the risk of stroke while G894T and 4b/a polymorphisms might be associated with the risk of stroke at least in Asians.
This comprehensive meta-analysis of 27 case-control studies was conducted on three well characterized genetic variants of the eNOS gene (G894T, T786C, 4b/a) and the studies were found to be significantly associated with all the three polymorphisms in the Asian population and the risk of IS. In this study, the effects of dominant, recessive, and allelic models were estimated. However, non-significant association was observed in all the three polymorphisms in the Caucasian population of eNOS gene. Our findings are found to be in concordance with the previously published meta-analyses by Niu et al., and Wang et al. In the case of overall analysis between the T786C polymorphism and risk of IS, no significant association was found. A previous meta-analysis published by Liu et al. also showed a non-significant association between T786C polymorphism and IS risk.
Heterogeneous evidence is observed even in the different meta-analyses published in the literature, and differences in methodology or study population used by published meta-analyses, inappropriate statistical test, and selection bias may explain the variation in the study results. The present meta-analysis examined heterogeneity by using meta-regression for the precise estimation of association between eNOS polymorphisms and the risk of IS.
The present meta-analysis must be carefully interpreted because of certain limitations. First, the studies included were found to be varied in environmental factors, ethnicity, and age. Second, the use of different methodologies for genotyping method, matching criteria, and selection of controls may have led to heterogeneity. Therefore, more credible evidences are required to illustrate solid conclusions regarding the association between eNOS gene polymorphisms and risk of IS.
In summary, our comprehensive meta-analysis shows a significant association between the three eNOS gene polymorphisms (G894T, T786C, 4b/a) and risk of IS in the Asian population whereas no significant association was observed in the Caucasian population. Further, no overall significant association was observed between the T786C polymorphism and IS risk. Further large prospective epidemiological studies are required to confirm these findings and draw accurate conclusions.
We acknowledge Translational Research in Cerebral Stroke- biology (TRICS) study for providing us the resources to conduct this meta-analysis.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]
[Table 1], [Table 2], [Table 3]