INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

VASCULAR COMPLICATIONS are the main causes of morbidity and mortality in patients with diabetes. Several lines of evidence suggest that endothelial damage could play a key role in the development of both micro- and macroangiopathy in diabetes. Recently, the role of the L-arginine/nitric oxide (NO) pathway in the regulation of vascular smooth muscle tone has attracted increasing interest. Endothelium-derived NO is a potent endogenous vasodilator (17). In addition to regulating vascular tone, endothelium-derived NO suppresses vascular smooth muscle proliferation (9), inhibits platelet adhesion and aggregation (20), and interferes with leukocyte-endothelial cell interaction (15), which lead to the development of atherosclerosis.

Despite the overwhelming evidence of impaired endothelium-dependent vasodilation in diabetes mellitus (21, 24), there are sporadic reports of preserved endothelium-dependent vasodilation (2). Possible explanations for discrepancies between studies include differences in several risk factors for endothelium-dependent vasodilation: hyperlipidemia (4), smoking (6), diabetic patients with and without complications (16), hypertension with diabetes with coronary heart disease (19), and a family history of coronary artery disease (7). In addition, several lines of evidence suggest that genetic factors contribute to the impairment of vascular endothelial function (3). Among previously reported polymorphisms in the endothelial constitutive NO synthase (ecNOS) gene, a 27-base pair repeat polymorphism in intron 4 of the ecNOS gene (4a/a genotype) was found to be associated with a smoking-dependent risk of coronary artery disease (27). This genotype was also associated with a history of myocardial infarction. These observations raise the possibility that this polymorphism is associated with endothelial dysfunction in patients with Type 2 diabetes, which is a potent risk factor for the development of atherosclerosis and coronary artery disease.

The aim of this study was to evaluate the possible relationships between ecNOS gene polymorphism, endothelium-dependent vasodilation, and endothelium-independent vasodilation in patients with Type 2 diabetes. In addition, because strong evidence for gene-environment interaction between smoking and the ecNOS gene has been obtained (27), we also investigated how cigarette smoking affects the relationship between ecNOS gene polymorphism and endothelial dysfunction.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. A total of 95 patients with Type 2 diabetes were studied. The patients were selected from 120 consecutive patients admitted to Osaka City University Hospital for educational programs on diabetes. The diagnosis of diabetes was based on a previous history of diabetes or on the criteria given in the Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus (23). Patients with a history of cardiovascular diseases, those with advanced diabetic complications (macroalbuminuria or proliferative retinopathy), and those treated with antioxidants vitamins, angiotensin-converting enzyme inhibitors, or insulin were excluded from the present study. Each subject gave informed consent to participate in this study. The Ethics Committee on Clinical Investigation of the Osaka City University Medical School approved the study methodology.

Endothelium-dependent and -independent vasodilations of the right brachial artery. Endothelium-dependent (flow-mediated, FMD) and -independent (nitroglycerin-induced, NTG) vasodilations of the right brachial artery were measured by the same examiner. The measurements were performed according to the method described by Celermajer et al. (5, 6) in a temperature-controlled (22°C) room. The arterial diameter was measured using an ultrasonic phase-locked echo-tracking system, which was equipped with a high-resolution, real-time 7.5-MHz linear scanner in B-mode (SSD 610; Aloka, Tokyo). The first ultrasound examination was performed with subjects in the supine position after they had rested for at least 15 min. A longitudinal section of the right brachial artery 2-12 cm above the elbow was scanned. The same observer, who was unaware of clinical details and the stage of the experiment, measured vessel diameter. The arterial diameter was measured from the anterior to the posterior interface between the media and adventitia ("m" line) (14) at a fixed distance from an anatomic marker (6). The mean diameter was calculated from four cardiac cycles synchronized with the R wave peaks on the electrocardiogram (ECG). All measurements were made at end diastole. Blood flow was then increased by inflating a pneumatic tourniquet to a pressure of 250 mmHg for 5 min. The second scan was taken for 30 s before and 90 s after cuff deflation; additional scans were recorded 3, 5, and 10 min later. Another resting scan was obtained 15 min later to confirm the vessel recovery. Scans were then obtained 3, 5, and 10 min after administration of a sublingual spray of nitroglycerin (300 µg/spray). The ECG was monitored continuously, and blood pressure was measured in the left arm during the ultrasound study. The maximal diameter changes caused by percent FMD (%FMD) and percent NTG (%NTG) were expressed as the percent change relative to that at the initial resting scan. In addition, the intima-media thickness of the right brachial arterial posterior wall was measured 2 cm proximal to the elbow joint as previously reported (12).

Reproducibility of the ultrasound study. Ten diabetic patients were examined on two different occasions 7 days apart to estimate the intraobserver variability of the values of %FMD and %NTG of the brachial artery diameter by the same examiner, who was unaware of the values from the first examination. The coefficient of variation was 4.3% for %FMD and 3.8% for %NTG.

DNA study. Polymorphism of the ecNOS gene in intron 4 was assessed by the polymerase chain reaction (PCR) method from peripheral leukocytes as previously reported (27). The PCR products were separated by electrophoresis in 2% agarose gel. We identified two alleles as ecNOS4a for four tandem 27-bp repeats (393 bp) and ecNOS4b for tandem repeats (420 bp).

Biochemical analysis. For each diabetic patient, the 24-h urinary albumin excretion was the mean value obtained from 3 consecutive days. Normoalbuminuria was defined as a urinary albumin excretion <20 µg/min and microalbuminuria as a urinary albumin excretion 20 µg/min and <200 µg/min. The plasma glucose and HbA1c levels were measured as previously described (13). Serum total cholesterol, triglyceride, HDL cholesterol, and creatinine levels were measured by an autoanalyzer. Urinary albumin was measured by immunoturbidimetry (TIA MicroAlb Kit; Nittobo, Tokyo).

Statistical analysis. Values are expressed as means ± SE unless otherwise indicated. Differences in variables among the groups were analyzed by unpaired t-test or ²-test. Because of the low prevalence of the ecNOS4a/a genotype, multiple regression analyses were performed to assess the magnitude of individual effects on vascular endothelial function. The following factors were considered as independent variables: age, gender (male = 1, female = 0), body mass index (BMI), cigarette years, duration of diabetes, mean blood pressure, fasting plasma glucose, HbA1c, low-density lipoprotein (LDL) cholesterol, triglyceride, HDL cholesterol, and presence of the ecNOS4a allele (4b/b = 0, 4b/a or 4a/a = 1) (model 1); and age, gender (male = 1, female = 0), BMI, duration of diabetes, mean blood pressure, fasting plasma glucose, HbA1c, LDL cholesterol, triglyceride, HDL cholesterol, and presence of the ecNOS4a allele (4b/b = 0, 4b/a or 4a/a = 1) (model 2). Because the smoking-related risk for coronary disease has been demonstrated for subjects with ecNOS4a/a genotype, the association between gene polymorphism and endothelial function was separately analyzed in smokers and nonsmokers. The %FMD and %NTG were compared among the four groups (smokers with and without ecNOS4a allele and nonsmokers with and without ecNOS4a allele). These procedures were performed on a Macintosh computer using the StatView V Statistical System. Values of P < 0.05 were considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The frequencies of ecNOS genotypes in our diabetic patients (4a/a, n = 3, 3%; 4b/a, n = 30, 32%; and 4b/b, n = 62, 65%) were consistent with those reported by Wang et al. (27) (1%, 32%, and 67%, respectively). Clinical characteristics of the patients are shown in Table 1. There was no significant difference in the percentage of patients treated with statins between the patients with and without ecNOS4a allele (4a/a or 4b/a, n = 9, 27%; and 4b/b, n = 19, 31%).

ecNOS gene polymorphism and endothelial function, %FMD and %NTG. The %FMD was significantly lower in the patients with ecNOS4a allele (5.10 ± 0.67%) than in those without ecNOS4a allele (7.07 ± 0.40%) (P < 0.05) (Fig. 1A). On the other hand, there were no significant differences in %NTG between the patients with and without ecNOS4a allele (4a/a or 4b/a, 10.90 ± 0.84%; and 4b/b, 12.58 ± 0.61%, respectively) (Fig. 1B). Because of the small number of patients with ecNOS4a/a genotype, multiple regression analysis was performed to determine the impact of ecNOS4a allele on %FMD and %NTG in all patients. The presence of ecNOS4a allele and mean blood pressure were independent determinants for reduced %FMD (R² = 0.122, P = 0.0025), and the duration of diabetes was an independent determinant for reduced %NTG (R² = 0.076, P = 0.0069) in all patients (model 1) (Table 2).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   A: percent flow-mediated (endothelium-dependent) vasodilation (%FMD) between patients with and without endothelium constitutive nitric oxide (ecNOS) 4a allele. Data are expressed as medians, interquartiles (open boxes), and ranges (10th and 90th percentile levels). %FMD was significantly lower in patients with ecNOS4a allele than in those without ecNOS4a allele. *P < 0.05 vs. patients without ecNOS4a allele. B: percent nitroglycerin-mediated (endothelium-independent) vasodilation (%NTG) between patients with and without ecNOS4a allele. There was no significant difference in %NTG between patients with and without ecNOS4a allele.

Smoking, %FMD, and %NTG in each genotype. Because the smoking-related risk for coronary disease has been demonstrated for subjects with the ecNOS4a/a genotype, we explored the possible interactions among cigarette smoking, ecNOS polymorphism, and endothelial function. There were significant differences in %FMD (P = 0.0159) but not in %NTG among the four groups. The %FMD was significantly lower in smokers with the ecNOS4a allele (4.94 ± 1.06%) than in smokers without this allele (7.05 ± 0.52%) and nonsmokers without it (7.10 ± 0.63%) (P < 0.05 and P < 0.05, respectively) (Fig. 2A). In addition, multiple regression analysis revealed that presence of the ecNOS4a allele was an independent determinant for reduced %FMD in smokers (R² = 0.217, P = 0.0020) but not in nonsmokers (model 2) (Table 3).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   A: %FMD in four groups of patients: smokers with ecNOS4a allele (4a/a + 4b/a), smokers without ecNOS4a allele (4b/b), nonsmokers with ecNOS4a allele, and nonsmokers without ecNOS4a allele. Parentheses indicate numbers of patients in each group. %FMD was significantly lower in smokers with ecNOS4a allele than in both the smokers and nonsmokers without this allele. *P < 0.05 vs. smokers without ecNOS4a allele. P < 0.05 vs. nonsmokers without ecNOS4a allele. B: %NTG in patients divided into four groups as described above. There were no significant differences in %NTG among the four groups.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This is the first study demonstrating that 4a/b polymorphism of the ecNOS gene is associated with endothelium-dependent vasodilation but not with endothelium-independent vasodilation in patients with Type 2 diabetes. Our findings that the ecNOS4a allele has a significant effect on endothelium-dependent vasodilation but not on endothelium-independent vasodilation suggest that this is a specific effect limited to NO production and/or release. We previously reported (12) a similar finding for patients with Type 2 diabetes, in a study in which endothelium-dependent vasodilation was significantly lower in Type 2 diabetic patients than in control subjects. We also found a significant association of the ecNOS4a allele with impairment of endothelium-dependent vasodilation in smokers but not in nonsmokers. These results indicate that the ecNOS4a allele is an independent risk factor for the impairment of endothelium-dependent vasodilation in diabetic patients, especially among smokers. A similar effect in smokers has been described previously (27), in a study in which a significant association of 4a/b polymorphism of the ecNOS gene with coronary artery disease was observed. Continuous release of NO by endothelium is an important endothelial function relevant to the prevention of atherogenesis, and there is clear evidence that NO-dependent basal vasodilation is impaired in long-term smokers (5). Our findings demonstrate associations among documented endothelial dysfunction, a variant of the ecNOS gene and smoking, which is an established risk factor for impaired endothelial function. Although our study does not identify any mechanism involved in the contribution of the ecNOS gene to endothelial dysfunction, we hypothesize that polymorphism of the ecNOS gene could affect NO production-induced reactive hyperemia. NO contributes to all phases of reactive hyperemia in the human peripheral vasculature (8). Evidence for a relationship between ecNOS gene polymorphism and impaired function of ecNOS was obtained in a previous study of NO metabolite (NOx) levels in patients with different ecNOS genotypes. Homozygotes for the ecNOS4a allele had a nearly 20% lower mean plasma NOx level, and carriers of the ecNOS4a allele had a significantly lower NOx level than did noncarriers (25). Thus, because in vivo and in vitro studies have demonstrated that the endothelium, by reduced NO synthesis and/or release, blunts endothelium-dependent vasodilation (1, 26), there may be less production of NO in patients with the ecNOS4a allele. Change in endothelial function may have important clinical implications for the pathogenesis of cardiovascular diseases. As suggested by the present study, functional alterations of the endothelium-derived NO pathway, especially those involved in the pathogenesis of diabetic vascular complications, may be due to decreased endothelial release of NO related to the presence of the ecNOS4a allele. Among few reported polymorphic markers in ecNOS gene, associations between these polymorphisms and endothelial function are controversial. The G894T polymorphism of the ecNOS gene was shown to be associated with an enhanced vascular responsiviness to phenylephrine (18). On the other hand, the Glu298Asp polymorphism of the ecNOS gene was reported to have no association with endothelial function (22). The present study showed the first evidence for an association between the 4a/b polymorphism of the ecNOS gene and endothelium-dependent vasodilation in Type 2 diabetes.

Blunting of endothelium-dependent vasodilation by smoking was observed in patients with the ecNOS4a allele but not in those without it in the present study. Furthermore, multiple regression analysis demonstrated that this allele was an independent determinant for reduced endothelium-dependent vasodilation in smokers but not in nonsmokers. Previously, Wang et al. (27) presented strong evidence for a gene-environment interaction between smoking and the ecNOS gene. The association between ecNOS gene polymorphism and ecNOS activity was reported to be modifiable by smoking (28). Smoking-induced vascular damage is considered a consequence of enhanced degradation of NO secondary to formation of oxygen-derived free radicals (11) as well as thrombogenic effects (10). The antioxidant vitamin C has been reported to improve endothelial dysfunction in smokers (11). Taken together, these findings suggest that smoking could be risk factor for cardiovascular disease by enhancing degradation of NO by free radicals. In the subjects with the 4a allele of the ecNOS gene, NO might be susceptible to degradation by the free radicals derived from smoking.

In conclusion, we have found a modest association between ecNOS polymorphism of the NO synthase gene and endothelium-dependent vasodilation and have identified increased risk for impaired endothelium-dependent vasodilation in diabetic patients with the ecNOS4a allele associated with cigarette smoking.