|
|
||||||||
Endocrinology and Metabolism |
Department of Internal Medicine II, Hospital Clínico San Carlos, Madrid, Spain.
aAddress correspondence to this author at: Cea Bermúdez 66, 5° G. 28003 Madrid-Spain, Fax: +34-91-4429705; e-mail uinvest7.hcsc{at}salud.madrid.org.
| Abstract |
|---|
|
|
|---|
Methods: We recruited 738 unrelated persons from a cross-sectional population-based epidemiological survey in the province of Segovia in Central Spain (Castille). Metabolic syndrome was defined according to the recently modified National Cholesterol Education Program Adult Treatment Panel III guidelines.
Results: Haplotype analysis showed a statistically significant association between some NOS3 gene variants and features of metabolic syndrome. Relative to the most common haplotype, 121, the haplotype 212 was associated with an increased odds ratio (OR) for metabolic syndrome [OR = 1.81, 95% confidence interval (CI) 1.152.84], and for decreased HDL-cholesterol concentrations (OR 1.52, 95% CI 1.012.29), and with increased mean values for the homeostasis model assessment of insulin resistance (P = 0.043), and triglycerides (P = 0.026).
Conclusions: Our results suggest that genetic variation at the eNOS locus is associated with features of metabolic syndrome, and might represent a new genetic susceptibility component for insulin resistance, hypertriglyceridemia, and low HDL-cholesterol concentrations.
| Introduction |
|---|
|
|
|---|
Endothelial nitric oxide synthase (eNOS)1 is involved in production of nitric oxide (NO), a ubiquitous molecule responsible for the maintenance of normal endothelial function. Among its many roles, NO facilitates the uptake and metabolism of glucose in skeletal muscle (9)(10). By contrast, the superoxidative effects of NO, which may occur through functional variation in NO synthase genes, may play a role in insulin resistance and type 2 diabetes (11).
In humans, eNOS is encoded by the NOS32 gene, which is located at 7q3536. Several authors have reported an association between polymorphic alleles of the NOS3 gene and hypertension (12)(13), but the demonstration of such associations has not been replicated by others using the same genetic markers (14)(15). These discrepancies could be due to different ethnic populations, study design, and sample size. In addition, variants at the NOS3 gene have been related to type 2 diabetes mellitus (16), insulin resistance (17)(18)(19), plasma lipids, body mass index (20)(21), and metabolic syndrome (22).
Interestingly, it has recently been reported that genomic variation within the NOS3-proximal region of chromosome 7q influences both insulin resistance and blood pressure (23), suggesting that this locus may influence traits associated with metabolic syndrome. Furthermore, other studies of partial eNOS knockout mice demonstrate that these animals are hypertensive, hyperinsulinemic, dyslipidemic, and have decreased insulin-stimulated glucose uptake compared with wild-type mice (24). Because these pathophysiological disorders are central features of the metabolic syndrome, the purpose of this study was to investigate the relationship between NOS3 gene haplotypes and metabolic syndrome defined according to National Cholesterol Education Program Adult Treatment Panel III recently modified guidelines (25) in the general Spanish (Caucasian) population.
| Materials and Methods |
|---|
|
|
|---|
phenotype measurements
Anthropometric measurements included body mass index (kg/m2), waist circumference (cm), and waist-to-hip ratio. Waist-to-hip ratio was calculated by dividing waist circumference (cm) by hip circumference (cm). Waist measurements were made with a nonstretchable fiber measuring tape while study participants were standing erect in a relaxed position with both feet together on a flat surface. Waist girth was measured as the smallest horizontal girth between the costal margins and the iliac crests at minimal respiration. Hip measure was taken as the greatest circumference at the level of the greater trochanters (the widest portion of the hip) on both sides. Measurements were made to the nearest centimeter. Systolic and diastolic blood pressures were measured 3 times on persons in the seated position after 10 min of rest; measurements were read to the nearest even digit by use of a random-zero sphygmomanometer.
After participants had fasted overnight, 20 mL of blood were obtained from an antecubital vein without compression. Plasma glucose was determined in duplicate by a glucose-oxidase method adapted to an automated analyzer (Hitachi 704, Boehringer Mannheim). Total cholesterol, triglycerides and HDL cholesterol were determined by enzymatic methods with commercial reagent sets (Boehringer Mannheim). LDL cholesterol was calculated by the Friedewald formula. Serum insulin, adiponectin, and leptin concentrations were determined by RIA (Human Insulin Specific, Human Adiponectin Specific, and Human Leptin Specific, respectively, Linco Research Inc.).
One and 2 h after glucose administration, we obtained blood samples and measured glucose and insulin concentrations. Insulin resistance was estimated by the homeostasis model assessment (HOMA-IR) method (27).
In the present study we used the National Cholesterol Education Program Adult Treatment Panel III definition of metabolic syndrome, recently modified (25) for analysis of the association between NOS3 haplotypes and metabolic syndrome in a population-based study.
genetic analyses
Twelve previously described NOS3 gene variants were genotyped (16). Five of these single nucleotide polymorphisms (SNPs) (IVS2 + 42, IVS626, IVS1130, E298D rs1799983, IVS25 + 15) were identified from de novo polymorphism screening in a human diversity panel with 47 samples from 4 distinct ethnic groups including our population of Segovia (16). The remaining 7 candidate SNPs (rs31800783, rs31800779, rs32070744, rs33800787, rs3918166, rs3918212, and rs3918220) were selected from dbSNP database to increase coverage of the gene. SNPs rs3918166, rs3918212, and rs3918220 were monomorphic in the population studied here. dbSNPs were genotyped by use of an adaptation of the homogenous MassExtend protocol supplied by Sequenom for the MassArray system. In summary, for each polymorphism we used an adaptation of the fluorescence polarization template-directed incorporation method to genotype DNA from leukocytes (16). Primer extension preamplificationamplified DNA samples were PCR amplified in 8 µL reactions with primers flanking the variant site, and unincorporated dNTP and remaining unused primer were degraded by exonuclease I and shrimp alkaline phosphatase at 37 °C for 45 min before the enzymes were heat inactivated at 95 °C for 15 min. At the end of the reaction, the samples were held at 4 °C. Single-base primer extension reactions were performed as previously described (28), and allele detection was performed by measuring fluorescence polarization on an LJL Analyst fluorescent reader (Molecular Devices).
statistical analysis
We compared genotype and allele frequency distributions with the
2 test and computed Hardy-Weinberg equilibrium to the expected genotype distribution. We used Student t-test and ANOVA (with Bonferroni test for post hoc comparisons) to compare continuous variables [expressed as means (SD)], and the
2 test to compare categorical variables. Continuous variables that did not have a gaussian distribution were log-transformed.
Logistic regression analysis was performed to evaluate associations of SNPs with metabolic syndrome. Adjusted odds ratios (OR) and their 95% confidence intervals (95% CI) were calculated. The null hypothesis was rejected in each statistical test when P <0.05. Analysis was performed with the Windows SPSS version 11.0 software (SPSS, Inc.).
Haplotype-tagging SNPs (htSNPs) were selected with the htSNP program (available online at http://www.gene.cimr.cam.ac.uk/clayton/software/stata/). This program operates on a Stata platform and enables the selection of htSNPs. htSNPs are the SNPs from a group of linked genetic variants that describe a large proportion of the total variance in all of the available SNPs. We selected 3 htSNPs (rs2070744, IVS1130, and rs3800787) that explained
54% of the variance incumbent in all SNPs. We then used the Thesias program to estimate the haplotype frequencies of the 3 htSNPs (available online from http://genecanvas.ecgene.net) (29). Because of the previously reported associations between these htSNPs and diabetes in the UK study (16), we chose to proceed by testing the association between the same htSNPs and features of the metabolic syndrome in the Spanish cohort.
We reconstructed haplotypes from the 3 htSNPs (rs2070744, IVS1130, and rs3800787) and retained 7 common haplotypes (prevalence at >5%) for further analysis. The coding of the haplotypes refers to the allele at each genetic locus (i.e., 1 or 2). Each number within the haplotype is ordered according to its genomic location. Thus, the 1st number refers to allele 1 or 2 at the rs2070744 SNP, the 2nd number refers allele 1 or 2 at the IVS1130 SNP, and the 3rd number refers to allele 1 or 2 at the rs3800787 SNP. Comparison of haplotype frequencies was done with
2 tests. For quantitative analysis, haplotype effects are expressed as increases/decreases of the phenotypic mean with respect to the most frequent (the intercept). For qualitative analysis, results are expressed as ORs.
| Results |
|---|
|
|
|---|
|
We observed significant differences in genotypic distributions and haplotype frequencies for the prevalence of metabolic syndrome and its related phenotype traits. For the NOS3 IVS1130 SNP, the prevalence of the AA genotype was greater in individuals with insulin resistance determined by HOMA-IR compared with those without insulin resistance (36.9% vs 26.5%, P = 0.020). In addition to increased insulin resistance, AA homozygous individuals had higher total cholesterol concentrations and a trend toward higher LDL-cholesterol and fasting glucose concentrations than individuals with AG heterozygous and GG homozygous genotypes (Table 2
).
|
For the rs3800787 SNP, the CC genotype was significantly more frequent in individuals with metabolic syndrome than in those without metabolic syndrome (16.4% vs 12.5%, P = 0.010) and in those with decreased HDL-cholesterol concentrations (16.1% vs 12.7%, P = 0.044). Moreover, the CC genotype was also associated with a higher HOMA-IR value (Table 3
). Thus, the C (2) allele carriers at the rs3800787 locus had higher mean (SD) values for fasting glucose [5.2 (1.5) vs 4.8 (1.3) mmol/L, P = 0.015], and HOMA-IR [3.3 (2.9) vs 2.8 (1.9), P = 0.013] than those with the GG genotype. Moreover, this C allele was also associated with a greater OR for metabolic syndrome (OR 1.79, 95% CI 1.222.64) and for decreased HDL-cholesterol concentrations (OR 1.71, 95% CI 1.112.63).
|
In haplotype analyses (Table 4
), we observed that the haplotype 212 was associated with an increased OR for metabolic syndrome (OR 1.81, 95% CI 1.152.84, P = 0.01), even after adjustment for variables that may affect NO metabolism such as smoking, alcohol intake, type 2 diabetes, and antihypertensive medication (OR 1.87, 95% CI 1.143.08, P = 0.01). Moreover, this haplotype was also associated with a greater OR for decreased HDL-cholesterol concentrations (OR 1.52, 95% CI 1.012.29, P = 0.04) and with increased mean values of HOMA-IR (+35.7%, P = 0.043), and triglycerides (+23.7%, P = 0.026) compared with the most frequent haplotype 121.
|
| Discussion |
|---|
|
|
|---|
eNOS activation increases muscle blood flow, with increased delivery of insulins major substrate, glucose, to the muscle cell (36)(37)(38). These findings are in agreement with our data suggesting that genetic defects in the NOS3 gene may be useful markers of risk for features of the metabolic syndrome, such as dyslipidemia and insulin resistance.
To our knowledge, this is the first study investigating the relationships between eNOS haplotypes and metabolic syndrome in a large population-based study. Other studies utilized different study design and smaller sample size (19)(22). In concordance with our results, Fernández ML et al. (22) also reported an association between NOS3 SNPs and metabolic syndrome, but in hypertensive patients. Other eNOS haplotype studies have been focused on the association with hypertension in hypertensive patients (39) and in Brazilian patients without and with type 2 diabetes (40). In contrast to these studies, our study did not find any association between eNOS haplotypes and hypertension. Some other studies have found evidence of association between NOS3 variants and hypertension (12)(13), but 2 other studies (14)(15) did not.
Discrepancies between these studies may be attributable to sample size, study design, and population stratification secondary to ethnic diversity. In this regard, marked interethnic differences in the distribution of eNOS genetic variants and haplotype frequencies have been reported (41)(42). In addition, discrepancies in association studies may also result from consideration limited to only one polymorphism rather than combinations of polymorphisms (haplotypes) (43)(44). Therefore, it is generally believed that analyzing combinations of genetic markers in a region of interest (haplotypes) can be much more informative than testing the effects of genetic markers one by one (44).
At variance with our study, the majority of published studies reporting association between NOS3 gene variants and features of the metabolic syndrome have used several single markers (SNPs) to characterize the genetic architecture. Although genetic association studies using single markers can be valuable in cases in which SNPs are likely to be functional, the probability of detecting a false-positive association increases proportionally with the number of statistical comparisons made. Thus, if the available SNPs are not believed to be functionally related to the phenotype (i.e., they are thought to flag only the true functional variant through high-linkage disequilibrium), association models that make use of haplotypes, which require fewer statistical comparisons to test the same underlying hypotheses, may be more appropriate. Thus, because haplotype approaches are less likely to lead to false discovery, the genetic associations identified in this way may be of greater value.
Recent studies suggest that the extensive variation in human beings is best described by haplotypes (45)(46). Haplotypes are more likely to identify disease associations than single polymorphisms because they reflect global gene structure and encompass the majority of common variations in a gene. Identification of a haplotype associated with increased or decreased disease risk should facilitate identification of the actual functional variant that affects disease risk, because this variant should lie on chromosome regions identified by that haplotype.
The present study does not address the molecular basis for the association of metabolic syndrome with specific eNOS haplotypes. The effect of eNOS variants on enzyme activity and on the production of NO is still a controversial issue. Some polymorphisms in the promoter region (T-786C) have been shown to decrease promoter activity (47). Recent findings, however, suggest that these polymorphisms do not have a significant effect on NO bioavailability (48). Additional but controversial data on the effect on impaired enzyme function of polymorphisms in exon 7 have also been reported (49)(50). Furthermore, it has been suggested that eNOS haplotypes, instead of eNOS genotypes alone, could have a major effect on NO bioavailability (51). Therefore, the molecular mechanisms underlying the association between specific eNOS haplotypes and abnormalities in eNOS activity deserve further investigation.
In summary, our results suggest that genetic variation at the NOS3 locus may increase an individuals susceptibility to certain components of the metabolic syndrome, specifically insulin resistance, hypertriglyceridemia, and low HDL-cholesterol concentrations.
| Acknowledgments |
|---|
This work was supported by grants FEDER 2FD1997-2309 from Fondo Europeo de Desarrollo Regional; FISS 03/1618 from Fondo de Investigaciones Sanitarias and a grant from Red de Centros RCMN (C03/08), Madrid, Spain.
Partial support also came from educational grants from Eli Lilly Lab, Spain and Bayer Pharmaceutical Co., Spain. We also acknowledge financial support by an educational grant from Santander Central Hispano Bank (GSCH 5292564).
| Footnotes |
|---|
2 Human gene: NOS3, nitric oxide synthase 3 (endothelial cell). ![]()
| References |
|---|
|
|
|---|
C Mutation in the 5'-Flanking Region of the Endothelial Nitric Oxide Synthase Gene Is Associated With Coronary Spasm. Circulation 1999;99:2864-2870.The following articles in journals at HighWire Press have cited this article:
![]() |
M.-A. Cornier, D. Dabelea, T. L. Hernandez, R. C. Lindstrom, A. J. Steig, N. R. Stob, R. E. Van Pelt, H. Wang, and R. H. Eckel The Metabolic Syndrome Endocr. Rev., December 1, 2008; 29(7): 777 - 822. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Imamura, R. Takahashi, R. Murakami, H. Kataoka, X. W. Cheng, Y. Numaguchi, T. Murohara, and K. Okumura The effects of endothelial nitric oxide synthase gene polymorphisms on endothelial function and metabolic risk factors in healthy subjects: the significance of plasma adiponectin levels Eur. J. Endocrinol., February 1, 2008; 158(2): 189 - 195. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |