Smoking Impairs the Activity of Endothelial Nitric Oxide Synthase in Saphenous Vein
Abstract Smoking impairs the endothelium-dependent relaxation of arteries and veins, with the maximum relaxation in response to the calcium ionophore A23187 of saphenous vein rings being reduced from 53±4% in nonsmokers to 27±5% in smokers. We have investigated whether this endothelial dysfunction was attributable to altered activity or concentration of nitric oxide synthase (NOS). The concentration of NOS in saphenous vein endothelium, determined by Western blotting and immunohistochemistry, was not different in nonsmokers and smokers. Nitrite production from vein strips stimulated with A23187 was higher in nonsmokers (median 23.6 nmol·cm−2·h−1) than smokers (median 3.3 nmol·cm−2·h−1), P=.001, this difference being abolished when vein strips were preincubated in the presence of NG-monomethyl-l-arginine. Organ chamber studies to monitor the endothelium-dependent relaxation of vein rings in response to A23187 showed that preincubation of rings from smokers with either l-arginine (3 mmol/L) or superoxide dismutase (250 U/mL) did not improve the maximum relaxation. In contrast, preincubation of vein rings from smokers with 20 μmol/L tetrahydrobiopterin increased the maximum relaxation from 27±5% to 51±6%, P=.01. Preincubation of vein from smokers with tetrahydrobiopterin also significantly increased nitrite and cGMP production in response to stimulation with A23187. The impaired endothelium-dependent relaxation of saphenous vein rings from smokers appears to be caused by a reduction in the activity of endothelial NOS that is attributable to an inadequate supply of the coenzyme tetrahydrobiopterin.
- Received June 30, 1995.
- Accepted December 20, 1995.
Saphenous vein is used widely for surgical bypass of critical atherosclerotic lesions in the coronary and leg arteries. Smoking is both a cause of the underlying disease and a major risk factor for bypass graft failure.1 2 3 Therefore, recently there has been a controversy as to whether smokers should be eligible for bypass surgery.4 5 Whereas other cardiovascular risk factors, such as hypertension and hyperlipidemia, are measured objectively, the clinician usually relies on the patients’ history to diagnose smoking. However, some patients deceive the surgeon about their smoking habits, and an objective blood or urine marker of smoking will identify covert smokers.1 6
Smoking adversely influences several atherosclerotic and thrombotic factors relevant to bypass patency, including platelet function, eicosanoid metabolism, and fibrinogen concentrations.7 8 9 It is also widely recognized that smoking damages the endothelium, but how smoking damages the endothelium is poorly understood. Recently, it has been reported that smoking impairs the endothelium-dependent relaxation of both arteries in vivo and vein rings in vitro.10 11 The obligatory role of the endothelium in modulating vasomotor tone was described first by Furchgott and Zawadski in 1980.12 Endothelium-derived relaxing factor is NO, which is synthesized constitutively in arterial endothelium from l-arginine and molecular oxygen by NOS.13 NO contributes to resting arterial tone, but it also has important effects on both platelet function and smooth muscle cell proliferation.13 14 15 16 In veins, there is little basal release of NO, but NO is released in response to a variety of agonists.17 Veins used as an arterial bypass conduit adapt to the arterial circulation. Changes in endothelium-dependent relaxation may be one adaptation.18 Saphenous vein remains in use for coronary artery bypass and is the optimum conduit for femorodistal bypass grafts, especially when the distal anastomosis is below the knee. Studies on the endothelium-dependent relaxation of such grafts have been reported only in vitro or in animal studies.18 19 Nevertheless, since smoking is associated with impaired endothelium-dependent relaxation in both the superficial femoral artery and the saphenous vein used to bypass disease in this artery, it is important to understand how smoking causes this manifestation of endothelial dysfunction. Here we describe in vitro studies with saphenous vein to investigate the molecular basis of the impaired endothelium-dependent relaxation in smokers.
Proximal saphenous vein was obtained from patients undergoing high ligation and excision for correction of varicose veins; this procedure was approved by the local ethics committee. Smokers were defined as current smokers of at least 10 manufactured cigarettes per day with at least a 10–pack year history of smoking. Nonsmokers had never smoked. Smoking status was confirmed by measurement of whole blood carboxyhemoglobin and/or plasma cotinine from a sample obtained on the morning of planned surgery. Cotinine was measured by gas liquid chromatography, and a concentration of >200 nmol/L was indicative of current smoking. Carboxyhemoglobin was measured with an IL282 carboximeter, and a concentration of >2% was indicative of current smoking. Patients with diabetes or hyperlipidemia, patients on antihypertensive therapy or nitrate therapy, and patients with previous history of venous thrombosis or venous surgery were excluded.
Organ Chamber Studies of Endothelium-Dependent Relaxation
Saphenous vein samples were transported to the laboratory in ice-cold, oxygenated solution containing (mmol/L) 118.4 NaCl, 4.7 KCl, 2.5 CaCl2, 1.2 MgSO4 · 7H2O, 24.9 NaHCO3, 1.2 KH2PO4, 11.1 glucose, and 10 μmol/L indomethacin, which had been passed through a 0.22 μmol/L filter. Excess fat and connective tissue were dissected from the vein, and 5-mm rings were cut and suspended between metal stirrups in the chamber of an organ bath that contained 10 mL Krebs-Ringer buffer at 37°C, oxygenated with 95% O2 admixed with 5% CO2. The upper stirrup was connected by surgical silk to an isometric type D transducer (Bioscience) and the lower stirrup to a glass tissue holder. The vein ring was allowed to equilibrate for 60 minutes, during which time the chamber solution was changed every 15 minutes. Vein rings were stretched at 0.5-g intervals and exposed to phenylephrine (0.3 μmol/L) at each level of stretch until the optimum-length tension relationship was obtained, usually between 2 and 3 g. Vein rings that did not contract at phenylephrine concentrations of 2 μmol/L were considered to be diseased and were discarded.20 After a further 30 minutes of equilibration, a dose-response curve to phenylephrine (10−9 to 10−4 mol/L) was obtained. The vein ring was then thoroughly washed with Krebs-Ringer solution and allowed to equilibrate for a further 30 minutes before agonist-induced relaxation. For this procedure, vein rings were exposed to a submaximal dose of phenylephrine and accumulative doses of bradykinin or calcium ionophore A23187 (10−9 to 10−5 mol/L), which were added at 2-minute intervals or until a stable response was observed. After each relaxation concentration-response curve, the vein rings were again washed thoroughly with Krebs-Ringer solution and a further 30 minutes of equilibration was allowed before a new concentration-response curve was performed. For some experiments, vein rings were preincubated with L-NAME (0.1 mmol/L) for 30 minutes before a concentration-response curve for A23187 was determined. In other circumstances, vein rings from nonsmokers and smokers were preincubated with 3 mmol/L l-arginine, 250 U/mL superoxide dismutase, and 50 μmol/L sepiapterin or 20 μmol/L tetrahydrobiopterin [(6RS)-5,6,7,8-tetrahydro-l-biopterin] (both from Dr B. Schirck’s Laboratory, Jona, Switzerland) for 30 minutes before the concentration-response curve for A23187 was effected; when arginine was included, the pH of the Krebs solution was adjusted to 7.3 with 2 mol/L HCl. Some nonsmokers’ vein rings were preincubated with 30 ng/mL nicotine for 30 minutes before the dose-response curve for A23187 was effected.
Radioimmunoassay of cGMP
NO released from endothelium is assumed to stimulate guanylate cyclase in the underlying smooth muscle, thereby increasing the concentration of cGMP. The endothelium-dependent relaxation of paired vein rings (±20 μmol/L tetrahydrobiopterin) in response to A23187 (0.5×10−6 mol/L) was observed. After 3 minutes, the rings were detached, frozen in liquid nitrogen, and stored at −70°C. Before assay, rings were homogenized in 6% trichloroacetic acid, and the supernatant (centrifuged at 10 000g for 10 minutes) was extracted with ether. The extract, containing cGMP, was dried under N2 at 60°C and reconstituted to 0.5 mL in sodium acetate buffer, and cGMP was determined, after acetylation, by radioimmunoassay (Biotrak cGMP, Amersham Plc).
Sections (from segments adjacent to the vein rings used for organ chamber studies) were taken for light and scanning electron microscopy. For immunocytochemistry, vein samples were fixed by immersion in a solution of 1% paraformaldehyde in PBS (0.01 mol/L phosphate buffer, pH 7.2 to 7.4; 0.15 mol/L sodium chloride) for 4 to 6 hours. After washing at 4°C in PBS containing 15% sucrose, tissues were frozen and cryostat sections (8 μm) taken for immunostaining.
A specific rabbit polyclonal antiserum raised to a synthetic 15-residue peptide, based on the deduced amino acid sequences of cDNA encoding the bovine and human eNOSs, was used to investigate the localization of eNOS. This antiserum has previously been characterized elsewhere.21 22 A rabbit polyclonal antiserum and a mouse monoclonal antibody to the endothelial markers vWf (Reference A 082, Dakopatts) and platelet–endothelial cell adhesion molecule-1 (CD31; gift from Dr Masurov, Wellcome Research Laboratories, Kent, UK), respectively, were used to assess endothelial integrity and confirm the localization of the eNOS immunostaining.
Tissues were stained by the ABC peroxidase method. Endogenous peroxidase was blocked by immersion in 0.03% hydrogen peroxidase in methanol for 30 minutes followed by washing in PBS (three washes, 10 minutes each). After blocking nonspecific binding by incubating in 3% normal serum from the same species in which the biotinylated secondary antiserum was raised (either goat or horse) for 20 minutes, sections were blotted to remove excess serum and incubated overnight with primary antiserum diluted (eNOS and CD31 1:1000; eNOS/vWf 1:8000) in PBS containing 0.05% BSA and 0.01% sodium azide. Sections were then washed in PBS and successively incubated with either biotinylated goat or horse antiserum to rabbit or mouse IgG (Vector Laboratories) diluted 1:100 in PBS/BSA and freshly prepared ABC reagent (Vectastain, Vector Laboratories) for 30 and 60 minutes, respectively. Peroxidase activity was revealed using glucose oxidase–diaminobenzidine with nickel enhancement. Slides were dehydrated, cleared and mounted in Pertex Mounting Media (CellPath), then screened and photographed using an Olympus BX 60 microscope.
As control, serial sections were stained as above but either omitting primary antiserum or replacing it with the respective preimmune serum.
Quantification of Immunostaining
To assess the relative intensity of immunostaining and therefore derive information on how much of the enzyme of interest was present in the endothelium of these vessels, computer-assisted image analysis was performed using a Symphony image analysis system (Seescan). Digitized images representing the whole transverse section of the vessel were segmented by interactive thresholding to separate immunostaining (endothelium) from background. Intensity of staining was determined from an assessment of the proportion of light absorbed by the sample. This intensity was presented as gray shades, representing the fraction of incident light transmitted through the sample, scaled by 256. An averaged optical density measurement was achieved when the negative of the logarithm (to base 10) of the fraction was calculated. To eliminate inconsistencies, all samples to be quantified were prepared and stained at the same time. Measurements were taken in a darkened room to minimize background illumination, and the stability of the light source was regularly assessed. At least three sections were analyzed from each of 10 vein samples, representing nonsmokers (n=5) and smokers (n=5). All data were presented as mean±SEM.
Isolation of Endothelial Cells and Western Blotting
Segments of saphenous vein (2 to 4 cm in length) were opened longitudinally, washed with PBS, and the lumenal surface was floated on collagenase (1 mg/mL) for 30 minutes. Loosened endothelial cells were removed with single passage of a cell scraper and harvested by centrifugation. This procedure yielded 4000 to 6000 endothelial cells, which were stored as a pellet at −70°C until Western blotting was performed. Immediately before Western blotting, the pellet was thawed in water (200 μL), and membranous and particulate fractions were collected by centrifugation (100 000g, 60 minutes). The pellet was dissociated by boiling in gel-loading buffer containing 4% SDS and 0.5% dithiothreitol. Polyacrylamide gel electrophoresis was performed on 8% to 25% PHAST gels (Pharmacia). Western blotting on Immobilon-P membranes (Amersham Plc) used the following antibodies: mouse monoclonal antibodies to intercellular adhesion molecule-1 at 1:500 dilution (Serotec) and rabbit polyclonal antibodies to eNOS at 1:500 dilution (Santa Cruz Biotechnology) and vWf at 1:1000 dilution (Dako). The Western blots were developed using enhanced chemiluminescence (ECL kits, Amersham Plc) and bands quantitated by densitometric scanning.
Vein segments (3 to 5 cm) were flushed with warm Krebs solution (10 mL) and opened longitudinally. The vein strips were stretched flat and placed, endothelium side uppermost, in 10-cm2 dishes and repeatedly washed and agitated in warm Krebs solution until all the red cells and hemoglobin had been removed and the absorbance of the washings at 575 nm was zero. The vein strips then were divided into equal pieces of at least 1.5 cm2 in area. For unilateral ligation of saphenous vein, the vein was divided into two or three pieces, but for bilateral ligation, the vein could be divided into three to five pieces. Vein strips were preincubated for 30 minutes in Krebs solution (2 mL) before the addition of three doses of A23187 (1 nmol) at 5- to 10-minute intervals. Some preincubations included 0.25 mmol/L L-NMMA, 10−6 mol/L cycloheximide, or 50 μmol/L sepiapterin. The concentration of A23187 used was that which induced maximum endothelium-dependent relaxation, and an NOS inhibitor (L-NMMA) from which nitrite could not be produced was used. After 1 hour, the medium was removed, centrifuged for 2 minutes at 10 000g, and the absorbance at 575 nm was measured. If the absorbance at 575 nm was <0.03, the medium was divided into aliquots and stored at −20°C. Nitrite was measured using the Griess reaction, monitoring absorbance at 527 nm, detection limit in the range of 1 to 2 μmol/L, and by high-pressure anion-exchange chromatography (Dionex 2000i) with conductivity detection of anions (including nitrate), detection limit in the range of 5 to 10 nmol/L. Endothelin concentrations were measured using the high-sensitivity endothelin-1,2 [125I] assay system from Amersham. Sepiapterin and tetrahydrobiopterin were obtained from Dr Schirck’s Laboratory, and all other specialist chemicals were obtained from Sigma.
ANOVA was performed using Statview 4.0 for Macintosh computers.
The saphenous veins were harvested from 22 nonsmokers and 24 current smokers. There was no difference in age or sex between the nonsmokers and smokers: nonsmokers, 7 men and 15 women with median age 42 years (range, 25 to 61 years); smokers, 10 men and 14 women with median age 46 years (range, 25 to 69 years). The smokers had a higher carboxyhemoglobin concentration (median, 3.0%; range, 2.4% to 8.0%) than the nonsmokers (median, 1.6%; range, 1.1% to 1.9%), P<.00001. There was no difference in plasma cholesterol concentration between nonsmokers (median, 4.96; range, 3.26 to 5.78 mmol/L) and smokers (median, 4.20; range, 2.93 to 6.22 mmol/L). Light microscopic examination of segments of proximal saphenous vein did not reveal any intimal thickening or damage. Scanning electron microscopy demonstrated that the endothelium was well preserved, with no significant areas of endothelial denudation and no differences between nonsmokers and smokers.23
Immunostaining for eNOS was evident in the endothelium of all samples investigated and was present in all luminal endothelial cells. This localization was supported by the similar pattern of staining seen for vWf and CD31. Visual comparison of staining intensity for eNOS suggested no marked difference between veins from nonsmokers and smokers (Fig 1⇓). This observation was supported by quantifying intensity of staining, revealing that there was no significant difference in the relative amounts of antigen (eNOS) present in the endothelium of veins from nonsmokers and smokers.
Western Blotting on Isolated Endothelial Cells
The minimum number of saphenous vein endothelial cells necessary to provide clear bands for eNOS (140 kD) and ICAM-1 (95 kD) was in the range 1500 to 2000 cells. Blotting for vWf required only ≈500 cells. The ratio of NOS to ICAM-1 signal was 0.7±0.2 in cells harvested from nonsmokers (n=5) and 0.7±0.1 in endothelial cells harvested from smokers (n=3). The ratio of NOS to vWf signal was 0.12±0.04 in cells from nonsmokers and 0.11±0.03 in cells from smokers. These results confirm the findings on immunohistochemistry that the concentration of eNOS is similar in endothelial cells isolated from saphenous veins of nonsmokers and smokers.
Organ Chamber Studies of Endothelium-Dependent Relaxation in Nonsmokers and Smokers
The concentration of phenylephrine required to give half-maximal contraction, EC50, was similar for vein rings from smokers (1.6 μmol/L) and nonsmokers (1.7 μmol/L). Similarly, the EC50 for relaxation of precontracted vein rings in response to sodium nitroprusside was similar in smokers (1.23×10−7 mol/L) and nonsmokers (1.03×10−7 mol/L). The difference in endothelium-dependent relaxation of vein rings from smokers and nonsmokers in response to increasing concentrations of the calcium ionophore A23187 is shown in Fig 2⇓; the mean maximum relaxation of nonsmoker vein rings (53±4%) was double that in vein rings from smokers (27±5%), P=.0008. These relaxations could be blocked by 10−4 mol/L L-NAME (Fig 3⇓). The EC50 for relaxation in response to A23187 was similar in nonsmokers and smokers. Preincubation of vein rings from nonsmokers (n=4) with nicotine (30 ng/mL, the approximate plasma concentration observed in smokers) did not reduce the maximum relaxation in response to A23187 (Fig 3⇓). Preincubation of the smoker vein rings (n=4) with 250 U/mL superoxide dismutase did not influence either the mean maximum relaxation or the EC50 for A23187 (Fig 3⇓). In contrast, preincubation of vein rings from smokers with 3 mmol/L l-arginine decreased the EC50 for A23187 from 84 nmol/L to 30 nmol/L (P=.047) but did not influence the mean maximum relaxation (Fig 3⇓). Preincubation with 20 μmol/L tetrahydrobiopterin significantly increased the mean maximum relaxation of vein rings from smokers (n=5) in response to A23187, from 27±4% to 51±6%, P=.01 (Fig 4⇓). There was a similar increase in maximum relaxation, to 49±5% when vein rings were preincubated with 50 μmol/L sepiapterin (Fig 4⇓). In contrast, preincubation with tetrahydrobiopterin did not influence the relaxation of vein rings from nonsmokers (n=5) in response to A23187 (Fig 4⇓).
Nitrite Production From Saphenous Vein Strips
Hemoglobin binds NO and can inhibit the endothelium-dependent relaxation of saphenous vein rings. Therefore, the release of NO from saphenous vein endothelium, monitored by the accumulation of nitrite ions, was investigated after careful dissection and thorough washing of vein strips to provide a medium devoid of hemoglobin. Basal accumulation of nitrite was close to zero but increased in response to stimulation of vein strips with A23187: the A23187-stimulated accumulation of nitrite ions from nonsmoker (n=13) vein (median 23.7 nmol·cm–2·h−1) was significantly higher than that from smoker (n=13) vein (median 3.3 nmol·cm−2·h−1), ANOVA P<.001 (Table 1⇓). The ratio of nitrate/nitrite assessed by ion-exchange chromatography before reduction of nitrate was similar in nonsmokers (1.7:1) and smokers (1.9:1). The accumulation of nitrite in response to A23187 stimulation was inhibited almost completely by 2.5×10−4 mol/L L-NMMA (Table 1⇓). There was no measurable accumulation of nitrite in vein strips from which endothelium had been removed or damaged with a scalpel blade. The wide range of A23187-stimulated nitrite production may be attributable in part to the inaccuracy of measuring vein area. However, the ranges for smokers and nonsmokers were distinct. This might not have been the case had we relied on the patients’ truthfulness about smoking. Three patients who claimed to be ex-smokers but had carboxyhemoglobin levels of 2.4% to 4.4% on the morning of surgery are included as smokers. Preincubation of vein strips with 20 μmol/L tetrahydrobiopterin abolished the difference in nitrite production between smokers and nonsmokers by increasing the production in smokers to a median of 19.6 nmol·cm–2·h−1 (Table 1⇓).
Endothelin Release From Saphenous Vein Strips
The potent vasoconstrictor endothelin-1 is also released from venous endothelium. The basal release of endothelin from saphenous vein strips from nonsmokers and smokers was similar (Table 2⇓). Stimulation of vein strips with A23187 resulted in a significant reduction in endothelin release, ANOVA P=.023 (Table 2⇓).
cGMP Concentration in Saphenous Vein
The concentration of cGMP in veins after stimulation with 10−6 mol/L A23187 was 116.5±31.6 fmol/mg protein (mean±SD) in nonsmokers (n=5) and 60.5±10.2 fmol/mg protein in smokers (n=4), P=.031. The concentration of cGMP in unstimulated vein was low, 23.4±7.0 fmol/mg protein in vein from nonsmokers (n=5) and 27.0±8.5 fmol/mg protein in vein from smokers (n=4). When smoker veins were preincubated with 20 μmol/L tetrahydrobiopterin before stimulation with A23187, the tissue concentration of cGMP increased to 127±24.1 fmol/mg protein.
Smoking has adverse effects on the endothelium-dependent relaxation of both arteries in vivo and veins in vitro, which may be reversible on cessation of smoking.10 11 24 25 In contrast, neither endothelium-independent relaxation of vein rings in vitro, eg, in response to sodium nitroprusside, nor the response of vein rings to phenylephrine is different in smokers. Endothelium-dependent relaxing factor, or NO, is one regulator of resting arterial tone, but the endothelium also synthesizes other vasoactive molecules, including endothelium-derived hyperpolarizing factor, prostacyclin, and the potent vasoconstrictor endothelin-1. Smoking is recognized to impair the synthesis of prostacyclin by the endothelium.26 The purpose of this study was to identify whether smoking also impaired the synthesis of NO by the endothelium. Since conditions such as hypertension, hypercholesterolemia, and diabetes impair endothelium-dependent relaxation, we were particularly careful in the selection of patients from whom saphenous vein was harvested, excluding patients with the listed conditions. We also confirmed the smoking status of patients with the use of objective smoking markers, the smokers all being heavy current smokers.
Several different mechanisms could underlie the impaired endothelium-dependent relaxation in smokers, including changes in the synthesis and release of vasoactive molecules other than NO. For this reason, all experiments were performed in the presence of indomethacin to inhibit prostanoid synthesis. The measurement of nitrite release from saphenous vein in short-term culture provides convincing evidence that smoking influences the synthesis and release of NO. Nitrite and nitrate are the aqueous oxidation products of NO, which were analyzed by a sensitive anion-exchange/conductivity method designed for the analysis of water. The release of nitrite from nonsmoker vein was significantly higher and in a separate range from nitrite release from the vein of heavy smokers (Table 1⇑). The ratio of nitrite/nitrate was similar in nonsmokers and smokers, eliminating the possibility of differential oxidation of NO at the vessel wall. In nonsmokers, nitrite release was reduced considerably when vein was preincubated with the NOS inhibitor L-NMMA, and the minimal effect of cycloheximide argues for the possibility that this nitrite arises from constitutively expressed NOS. Low levels of endothelin-1 were released in short-term culture, but there was no difference between nonsmokers and smokers, indicating that smoking did not augment the production of this vasoconstrictor.
Impaired release of endothelial NO may have many causes, including abnormal supply and uptake of l-arginine by the endothelium, altered activity of cell surface receptors and intracellular signaling molecules lined to the l-arginine/NO pathway, reduced synthesis or activity of NOS, and inactivation of NO by free radicals or other molecules. Since both receptor-mediated (bradykinin) and receptor-independent (A23187) endothelium-dependent relaxation of vein rings is impaired in smokers,10 we had already excluded the consideration that smoking influenced signal transduction events at the endothelial plasma membrane. If the impaired release of NO was attributable to either reduced synthesis or faster turnover of NOS, we would have hoped to observe a marked reduction in the immunostaining for NOS of saphenous vein endothelium from smokers, but there was no difference in the immunostaining of saphenous vein from nonsmokers and smokers (Fig 1⇑). Moreover, the ratio of eNOS to ICAM-1 or eNOS to vWf in human saphenous vein endothelial cells, assessed by Western blotting, was similar in cells isolated from nonsmokers and heavy smokers. Since preincubation of smoker vein rings with either l-arginine or superoxide dismutase did not effect a significant improvement of endothelium-dependent relaxation, we also would discount the hypothesis that the impairment in endothelium-dependent relaxation in smokers arises from limitations of substrate supply or rapid inactivation of NO by free radicals. From these observations, we conclude that the concentration of venous eNOS is not reduced in smokers but rather that the activity of eNOS is impaired by smoking.
Smoking allows the absorption of hundreds of noxious chemicals into the circulation. Perhaps it would not be surprising if one or more of these chemicals contributed to the inhibition of NOS, an enzyme with a complex active site that requires several cofactors to contribute to the oxidative synthesis of NO and citrulline from l-arginine. Nicotine is responsible for many of the cardiovascular effects of smoking.27 However, preincubation of vein rings with nicotine did not suggest that nicotine contributed directly to the inhibition of NOS. Tetrahydrobiopterin, one of the essential coenzymes for NOS activity, is synthesized in endothelial cells from GTP in a multistep pathway in which the final step is the enzymatic conversion of sepiapterin to tetrahydrobiopterin.28 Exogenous tetrahydrobiopterin stimulates the activity of NOS derived from cultured human umbilical vein endothelial cells, and an inhibitor of tetrahydrobiopterin synthesis attenuates the endothelium-dependent relaxation of canine coronary arteries in response to A23187.28 29 30 Intra-arterial infusion of tetrahydrobiopterin also induces strong vasodilatation in humans.31 Preincubation of saphenous vein from smokers with both tetrahydrobiopterin and its biosynthetic precursor sepiapterin improved the endothelium-dependent relaxation in response to A23187. In contrast, the endothelium-dependent relaxation of vein rings from nonsmokers did not improve after preincubation with tetrahydrobiopterin. Further support of the observation that preincubation with tetrahydrobiopterin improves the endothelial function in saphenous vein from smokers arises from the influence of tetrahydrobiopterin on nitrite release and the tissue concentrations of cGMP. NO binds to guanylate cyclase in smooth muscle to stimulate the synthesis of cGMP and hence effect a signaling cascade leading to vasorelaxation. Preincubation of vein rings from smokers with tetrahydrobiopterin was associated with a significant increase in the concentration of cGMP in the vessel wall after stimulation with A23187. Our evidence indicates that smoking impairs the synthesis of tetrahydrobiopterin in saphenous vein endothelium. The aromatic amines absorbed into the circulation from the combustion of tobacco could provide effective inhibitors of the enzymes in tetrahydrobiopterin synthesis. This, in turn, could limit the activity of eNOS and impair endothelium-dependent relaxation.
In conclusion, it would appear that the impaired endothelium relaxation of saphenous vein rings from smokers is attributable to a limitation in the supply or synthesis of tetrahydrobiopterin, which reduces the activity of NOS. Smoking is a potent risk factor for both peripheral atherosclerosis and infrainguinal vein graft occlusion.1 32 In the maturing vein graft, endothelial synthesis of NO is likely to limit intimal hyperplasia by preventing platelet aggregation and smooth muscle cell proliferation, in addition to modulating vasomotor tone.13 14 15 This highlights the possibility that an improved supply of tetrahydrobiopterin could improve the patency of vein grafts in those who are unable to stop smoking.
Selected Abbreviations and Acronyms
|L-NAME||=||NG-nitro-l-arginine methyl ester|
|vWf||=||von Willebrand factor|
This study was supported in part by the Tobacco Products Research Trust. We thank Rachid Mir-Hasseine for superb technical assistance.
Wiseman S, Kenchington G, Dain R, Marshall CE, McCollum CN, Greenhalgh RM, Powell JT. Influence of smoking and plasma factors on patency of femoropopliteal vein grafts. Br Med J. 1989;299:643-646.
FitzGibbon GM, Leach AJ, Kafka HP. Atherosclerosis of coronary artery bypass grafts and smoking. Can Med Assoc J. 1987;136:45-47.
Bosch X, Perez-Villa F, Sanz G. Effect of smoking habits on the preventive effect of aspirin and dipyridamole on early aortocoronary bypass occlusion. Eur Heart J. 1992;12:170.
Underwood MJ, Bailey JS, Shiu M, Higgs R. Should smokers be offered coronary bypass surgery? Br Med J. 1993;306:1047-1050.
Powell JT, Greenhalgh RM. Arterial bypass surgery and smokers. Br Med J. 1994;308:607-608.
Sillet R, Wilson MB, Malcolm RE, Ball KP. Deception among smokers. Br Med J. 1978;65:197-200.
Nowak J, Murray JJ, Oates JA, FitzGerald GA. Biochemical evidence of a chronic abnormality in platelet and vascular function in healthy individuals who smoke cigarettes. Circulation. 1987;76:6-14.
Lassila R, Seyberth HW, Schweer H, Koskenvuo M, Laustiola KE. Vasoactive and atherogenic effects of cigarette smoking: a study of monozygotic twins discordant for smoking. Br Med J. 1988;297:955-957.
Scott-Burden T, Schini VB, Elizondo E, Junquero DC, Vanhoutte PM. Platelet-derived growth factor suppresses and fibroblast growth factor enhances cytosine-induced production of nitric oxide by cultured smooth muscle cells. Circ Res. 1992;71:1088-1100.
Scott-Burden T, Vanhoutte PM. The endothelium as a regulator of vascular smooth muscle proliferation. Circulation. 1993;87(suppl V):V-51-V-55.
Ku DD, Caulfield B, Kirklin JK. Endothelium-dependent responses in long-term human coronary artery bypass grafts. Circulation. 1991;83:402-411.
Miller VM, Reigel MN, Hollier LH, Vanhoutte PM. Endothelium-dependent responses in autogenous femoral veins grafted into the arterial circulation of the dog. J Clin Invest. 1987;80:1350-1357.
Ishii T, Okadome K, Komori K, Odashiro T, Sugimachi K. Natural course of endothelial-dependent and -independent responses in autogenous femoral veins grafted into the arterial circulation of the dog. Circ Res. 1993;72:1004-1010.
Michel T, Li KG, Busconi L. Phosporylation and subcellular translocation of endothelial nitric oxide synthase. Proc Natl Acad Sci U S A. 1993;90:6252-6256.
Higman DJ, Powell JT, Greenhalgh RM, Coady A, Moss J. Is thickening of the basal lamina in the saphenous vein a hallmark of smoking? Br Heart J. 1994;71:45-50.
Celermajer DS, Sorensen KE, Georgakopoulos D, Bull C, Thomas O, Robinson J, Deanfield JE. Cigarette smoking is associated with dose-related and potentially reversible impairment of endothelium-dependent dilation in healthy young adults. Circulation. 1993;88:2149-2155.
Reinders JH, Brinkman HJM, van Mourik JA, de Groot PG. Cigarette smoke impairs endothelial cell prostacyclin production. Arteriosclerosis. 1986;6:15-23.
Werner-Felmayer G, Werner ER, Fuchs D, Hausen A, Reibnegger G, Schmidt K, Weiss G, Wachter H. Pteridine biosynthesis in human endothelial cells. J Biol Chem. 1993;268:1842-1846.
Rosenkranz-Weiss P, Sessa WP, Milstein S, Kaufman S, Watson CA, Pober JS. Regulation of nitric oxide synthesis by proinflammatory cytokines in human umbilical vein endothelial cells. J Clin Invest. 1994;93:2236-2244.
Cosentine F, Katusic ZS. Tetrahydrobiopterin and dysfunction of endothelial nitric oxide synthase in coronary arteries. Circulation. 1995;91:139-144.
Fowkes FGR. Aetiology of peripheral atherosclerosis. Br Med J. 1989;298:405-406.