INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MANY EFFORTS ARE CURRENTLY DIRECTED at identifying gene alterations that increase the susceptibility for coronary artery disease and myocardial infarction (13). Although in recent years the search for such alterations was limited to a few "candidate genes," the decoding of the entire human genome sequence and novel technologies for assessing sequence variations on a genome scale will prompt comprehensive studies of comparative genomic diversity in human populations (4). Even without these novel techniques, the identification of the number of gene alterations associated with cardiovascular disease is rapidly increasing; however, the underlying mechanisms or even the functional implications of such gene alterations are often poorly understood. The understanding of the influence of a variant on the level of protein expression or on structure and function of the encoded proteins may not always be helpful in highlighting possible mechanisms leading to disease (17). Thus it is important to link observed associations with dynamic and functional phenotypes influenced by the same genetic alterations with the aim to build up a basis for cardiovascular pharmacogenetics (21).

A single base exchange polymorphism (G894T) in the endothelial nitric oxide (NO) synthase (eNOS) gene results in a glutamate or aspartate, respectively, at position 298 in the eNOS protein. This substitution resembles a conservative or silent substitution that appears to be not located in any functional consensus sequence.

Nevertheless, the 894T allele has been recently associated with hypertension (14), acetylcholine-induced coronary spasm (25), myocardial infarction (9, 18), and increased blood pressure responses to systemic ₁-adrenoceptor activation (16). In a study by Wang et al. (23), there was a trend for a decreased eNOS enzyme activity in eNOS 894T allele carriers by up to 80% of that in GG homozygotes, which was, however, not significant (23). Together, the presence of an endothelial dysfunction secondary to the eNOS 894T allele is strongly suggested.

Given the obviously silent character of the substitution, the 894T allele was repeatedly suspected to be merely in a linkage disequilibrium with another functional marker, which was, however, ruled out, e.g., for the functional T-786C exchange in the eNOS promoter region (22).

A recent study (15) demonstrated in human endothelial cells and human hearts that eNOS with aspartate, but not glutamate, at position 298 is cleaved, resulting in the generation of 100- and 35-kDa products (15). In contrast, immunoblotting of recombinant or native eNOS yielded a single major protein band in the predicted position for eNOS. Thus the eNOS 894T allele leads to the generation of protein products with differing susceptibility to cleavage, suggesting that, in contrast to the prior predictions, this polymorphism has a functional effect on the eNOS protein.

However, the distinct relevance of the 894T allele for the regulation of vasomotor tone, which is characterized by complex interactions of numerous endogenous mechanisms (1, 8), remains to be further defined. Because recent evidence suggests that adenosine (ADO)-induced vasodilation is partly NO dependent (10, 20), this study was conducted to assess the impact of the eNOS 894T allele on coronary vasomotor tone at rest and during ADO-induced hyperemia.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study was conducted in accordance with the local ethics committee and the Revised Declaration of Helsinki.

Study population. Ninety-seven consecutive patients undergoing routine diagnostic cardiac catheterization for unclear chest pain were recruited when coronary angiography and intravascular ultrasound examination yielded either no plaques or plaques with <10% area of stenosis in the entire coronary system. All medication was stopped 24 h before the study. Demographic data are given in Table 1. Measurements were performed in the left anterior descending (LAD) coronary artery. All participants were Caucasians of German descent from the area of Essen, Germany. All gave written informed consent.

Catheterization procedure. The catheterization procedure was performed in accordance with institutional guidelines. Intracoronary Doppler flow velocity measurements were performed as described previously (2) using a 0.014-in. Doppler guide wire with a 12-MHz transducer (FloWire; Cardiometrics, Mountain View, CA) connected to a console (FloMod, Cardiometrics). In brief, the Doppler guide wire was positioned under fluoroscopic guidance into the midsegment of the respective coronary artery. Great care was taken to ensure a stable position of the Doppler guide wire to obtain an optimal signal for the entire duration of the protocol. All data were stored continuously on a videotape system (S-VHS, Sony; Tokio, Japan) for playback and off-line analysis.

Genotyping. DNA was extracted from whole EDTA blood using a standard procedure (QIAmp Blood Kit, Qiagen) according to the manufacturer's recommendations. Genotyping was performed using 5'-AAGGCAGGAGACAGTGGATGGA-3' as the sense primer and 5'-CCCAGTCAATCCCTTTGGTGCTCA-3' as the antisense primer. Cycling conditions (Biometra Uno II, Biometra Analytical; Göttingen, Germany) were as follows: initial denaturation at 94°C for 5 min, 35 cycles with 94°C for 1 min, 64°C for 1 min, and 72°C for 1 min, and a final extension step of 72°C for 7 min. Restriction analysis was performed using BanII (GIBCO Life Technologies), yielding an unrestricted PCR product of 248 bp for the 894T allele and two completely restricted products of 163 and 85 bp for the 894G allele. Genotyping was conducted in a blinded fashion, i.e., investigators were unaware of patient data.

Statistics. The results are expressed as means ± SE. Data were analyzed using the SPSS software package 8.0.0G by SPSS. Intragroup variations as well as comparisons between groups were performed for continuous variables by Student's t-test; categorial data were compared by ²-test or Fisher's exact test where appropriate. To test the influence of the 894T allele on CBF parameters, a general linear model was used, including MAP and heart rate, which were correlated with CBF. The P values presented for comparisons between genotypes with respect to CBF parameters were taken from the general linear model. A P value of 0.05 was considered to indicate statistical significance.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Genotyping analysis yielded 14 individuals homozygous for the 894T allele (TT), 39 homozygous for the G894 allele (GG), and 44 heterozygotes (TG). The resulting allele frequency for the 894T allele (32%) was much higher than that observed in Japanese individuals (9, 14, 18, 25); however, it was comparable with that previously observed in Caucasian subjects (16). Different genotypes were comparable with respect to their demographic and clinical data (Table 1).

At rest, CSA (TT: 6.3 ± 1.0 mm²; TG: 5.2 ± 0.3 mm²; GG: 6.2 ± 0.5 mm²) and MAP (TT: 99.6 ± 3.7 mmHg; TG: 104.8 ± 2.5 mmHg; GG: 99.2 ± 2.8 mmHg) were similar between the genotypes. Homozygous carriers of the 894T allele displayed a significantly reduced APV (11.6 ± 1.4 cm/s) compared with heterozygotes (15.6 ± 1.0 cm/s) and individuals without the 894T allele (16.5 ± 1.0 cm/s, P < 0.05; Fig. 1). CVR was significantly higher in eNOS 894T homozygotes than in individuals without the 894T allele (6.5 ± 1.0 vs. 4.4 ± 0.4 mmHg · ml¹ · min¹, P < 0.05; Fig. 1). As expected, CVR values for the heterozygotes were intermediate between the other genotypes (5.4 ± 0.5 mmHg · ml¹ · min¹), although not significantly different. CBF showed a similar trend between the genotypes, being lowest in the 894T homozygotes [TT: 19.8 ± 3.4 ml/min; TG: 24.4 ± 2.3 ml/min; GG: 27.0 ± 2.3 ml/min, not significant (NS); Fig. 1].

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Coronary flow parameters under resting conditions. Values are means ± SE. eNOS, endothelial nitric oxide synthase gene; TT and GG, homozygosity for the 894T and G894 alleles, respectively; TG, heterozygosity; CSA, cross-sectional area; MAP, mean arterial pressure at the coronary ostium; APV, average peak velocity; CVR, coronary vascular resistance; CBF, coronary blood flow. P values were taken from the general linear model, as described in METHODS.

During ADO-induced hyperemia, CSA (TT: 6.3 ± 1.0 mm²; TG: 5.2 ± 0.3 mm²; GG: 6.3 ± 0.6 mm²) and MAP (TT: 89.7 ± 5.6 mmHg; TG: 91.8 ± 3.1 mmHg; GG: 86.7 ± 3.4 mmHg) again displayed no significant differences among genotypes. APV (TT: 38.0 ± 3.8 cm/s; TG: 42.3 ± 1.6 cm/s; GG: 46.1 ± 2.2 cm/s, NS; Fig. 2), CVR (TT: 2.0 ± 0.6 mmHg · ml¹ · min¹; TG: 1.8 ± 0.3 mmHg · ml¹ · min¹; GG: 1.4 ± 0.1 mmHg · ml¹ · min¹, NS; Fig. 2), and CBF (TT: 64.2 ± 9.6 ml/min; TG: 67.9 ± 5.0 ml/min; GG: 78.7 ± 6.2 ml/min, NS; Fig. 2) showed the same tendencies as at rest for the comparison of 894T homozygotes with heterozygotes and individuals without the 894T allele. However, the differences reached no statistical significance. CFVR was significantly increased in homozygous carriers of the 894T allele when compared with G894 homozygotes (TT: 3.5 ± 0.2; GG: 2.9 ± 0.1, P < 0.05; Fig. 3), whereas heterozygotes showed again an intermediate value without statistical significance (3.0 ± 0.1). CFR (TT: 3.3 ± 0.3; TG: 3.1 ± 0.2; GG: 3.0 ± 0.2, NS; Fig. 3) and CFVR_corr (TT: 2.9 ± 0.2; TG: 92.9 ± 0.1; GG: 2.8 ± 0.1, NS; Fig. 3) were not different between the genotypes.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Coronary flow parameters under adenosine (18 µg ic)-induced hyperemia. Values are means ± SE. P values were taken from the general linear model, as described in METHODS.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Relative coronary flow parameters. Values are means ± SE. CFVR, coronary flow velocity reserve; CFR, coronary blood flow reserve; CFVRcorr, coronary flow velocity reserve corrected for the underlying baseline APV as previously reported (18). P values were taken from the general linear model, as described in METHODS.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CBF at rest. The 894T allele of a single base exchange polymorphism (G894T) in the eNOS gene appears to be associated with impaired endothelial function, because it is associated with hypertension (14), acetylcholine-induced coronary spasm (25), myocardial infarction (18), and increased blood pressure responses to systemic ₁-adrenoceptor activation (16). This hypothesis is further supported by our present findings on coronary vasomotor dysfunction associated with the eNOS 894T allele, appearently in an allele dose-dependent manner. Because the epicardial vasomotor tone, which is represented by CSA, was not different among genotypes, the coronary vasomotor dysfunction reflects primarily microvascular alterations, reflected by an increased CVR and subsequently decreased APV. Our data further support the concept of a decreased eNOS enzyme activity, which was shown to be reduced by up to 80% in TT homozygotes (23) and could be a result of an increased susceptibility to cleavage of the eNOS protein in 894T allele carriers (22).

ADO-induced hyperemia. Recent evidence suggests that ADO-induced vasodilation is at least partly NO dependent (10, 20), although this remains controversial (5). ADO, when administered into the coronary lumen, is expected to enhance the release of NO from endothelial cells via the adenosine 2A receptor (11, 12), which has been shown to mediate coronary vasodilation (3). In chronically instrumented dogs, endothelial NO release together with the activation of ATP-dependent K⁺ channels contributes to ADO-mediated vasodilation (6). In contrast, another study by Duffy et al. (7) did not detect any attenuation of coronary flow reserve with N^G-monomethyl-L-arginine. Because high amounts of exogenous ADO are capable of overcoming NO-related deficiencies by direct receptor-mediated mechanisms (19), the findings by Duffy et al. (7) cannot exclude a significant contribution of NO to ADO-mediated vasodilation.

Study limitations. Although our results indicate a vasomotor dysfunction secondary to the eNOS 894T allele and, therefore, keep in line with existing reports (9, 14, 16, 18, 22, 23, 25), several limitations of the present study should be considered. 1) A number of 97 individuals with angiographically excluded coronary artery disease represents a quite large population of such cases. However, the number of included individuals who are homozygous for the 894T allele is still rather small (n = 14). 2) Although all concomitant medication was stopped 24 h before the angiographic procedure and there were no differences regarding concomitant medication between the genotypes, a potential effect cannot be fully ruled out.