An elevation of oxidized forms of tetrahydrobiopterin (BH4), especially dihydrobiopterin (BH2), has been reported in the setting of oxidative stress, such as arteriosclerotic/atherosclerotic disorders, where endothelial nitric oxide synthase (eNOS) is dysfunctional, but the role of BH2 in the regulation of eNOS activity in vivo remains to be evaluated. This study was designed to clarify whether increasing BH2 concentration causes endothelial dysfunction in rats. To increase vascular BH2 levels, the BH2 precursor sepiapterin (SEP) was intravenously given after the administration of the specific dihydrofolate reductase inhibitor methotrexate (MTX) to block intracellular conversion of BH2 to BH4. MTX/SEP treatment did not significantly affect aortic BH4 levels compared with control treatment. However, MTX/SEP treatment markedly augmented aortic BH2 levels (291.1 ± 29.2 vs. 33.4 ± 6.4 pmol/g, P < 0.01) in association with moderate hypertension. Treatment with MTX alone did not significantly alter blood pressure or BH4 levels but decreased the BH4-to-BH2 ratio. Treatment with MTX/SEP, but not with MTX alone, impaired ACh-induced vasodilator and depressor responses compared with the control treatment (both P < 0.05) and also aggravated ACh-induced endothelium-dependent relaxations (P < 0.05) of isolated aortas without affecting sodium nitroprusside-induced endothelium-independent relaxations. Importantly, MTX/SEP treatment significantly enhanced aortic superoxide production, which was diminished by NOS inhibitor treatment, and the impaired ACh-induced relaxations were reversed with SOD (P < 0.05), suggesting the involvement of eNOS uncoupling. These results indicate, for the first time, that increasing BH2 causes eNOS dysfunction in vivo even in the absence of BH4 deficiency, demonstrating a novel insight into the regulation of endothelial function.
- endothelial function
(6r)-5,6,7,8-tetrahydrobiopterin (BH4) is an essential cofactor required for the biosynthesis of nitric oxide (NO) by all three NO synthase (NOS) isoforms: neuronal, inducible, and endothelial (eNOS) NOSs. Like cytochrome P-450, NOS is a monooxygenase that incorporates one atom of oxygen from O2 into the substrate l-arginine and reduces the other atom to water, and the monooxygenase reaction requires an electron donated by BH4 putatively (11, 35). When BH4 is deficient, electron transfer from the reductase domain of NOS to heme becomes uncoupled from l-arginine oxidation, and superoxide instead of NO is produced from the oxygenase domain (51). In the cardiovascular system, the gaseous signal molecule NO produced by eNOS plays crucial roles in maintaining normal vascular function as a major endothelium-derived relaxing factor and through a variety of antiatherogenic effects. The concept of endothelial dysfunction has evolved as a result of studies on diseased arteries both in experimental animals and in patients with vascular disease (16). A number of studies over the past decade have demonstrated that a limited availability of BH4, as found in various vascular diseases such as diabetes, hypertension, and hypercholesterolemia, causes endothelial dysfunction and uncoupling of eNOS (7, 18, 24, 35, 43), indicating that the regulation of endothelial BH4 homeostasis is critically important in determining eNOS function.
BH4 is formed by a de novo pathway (47) and a salvage pathway (36) that restores BH4 from its oxidized form, 7,8-dihydrobiopterin (BH2). Through the latter pathway, exogenously given BH2 as well as sepiapterin (SEP), a BH2 precursor, can be efficiently converted into BH4 in cells by dihydrofolate reductase (DHFR) (42). On the other hand, BH4 is readily oxidized to BH2 in physiological saline solution in vitro (12), and the occurrence of BH4 oxidation, resulting in the accumulation of BH2 in vascular cells, has been found in the setting of oxidative stress associated with diabetes (31, 44), hypertension (28), and atherosclerosis (29, 49). Several in vitro studies with purified eNOS (52) and cultured endothelial cells (9, 45) have recently demonstrated that decreased ratios of intracellular BH4 to BH2, rather than the simple depletion of BH4, may be relatively important as the trigger for NO insufficiency or superoxide production. However, it remains uncertain whether the increased concentration of oxidized biopterins is as valuable as the decreased BH4 concentration in determining eNOS activity. Moreover, there appears no study focusing on the effect of intrinsic BH2 per se on NO bioavailability in the vasculature in vivo.
The present study was conducted to clarify whether an endogenous BH2 abundance impairs endothelial function in vivo. To this end, we evaluated the effects of a combined treatment with SEP and methotrexate (MTX), a DHFR inhibitor that blocks the intracellular conversion of BH2 to BH4, on pterin contents in the aorta, eNOS function assessed by the extent of endothelium-dependent vasodilatation, and vascular superoxide generation in rats.
MATERIALS AND METHODS
Male Wistar rats (age: 10–19 wk old, weight: 350–490 g) supplied by Kyudo (Kumamoto, Japan) were used.
This study was approved by the Animal Care and Use Committee of the University of the Ryukyus and was performed in accordance with the Guidelines for Animal Experimentation of the University of the Ryukyus.
Surgical preparation of rats.
Rats were anesthetized with an intraperitoneal injection of pentobarbital sodium (60 mg/kg). Subsequently, pentobarbital sodium at 40 mg/kg was subcutaneously injected to maintain a constant level of anesthesia throughout the experimental period. The trachea was cannulated to ensure patency of the airway. A heparinized catheter was inserted into the left femoral vein for the intravenous administration of SEP or vasoacting drugs, respectively. Aortic pressure (AoP) was measured through a catheter that was inserted into the aorta via the right carotid artery. The catheter was connected to a pressure transducer (TP-400T, Nihon Kohden, Tokyo, Japan). Heart rate was continuously monitored with a cardiotachometer (AT-601G, Nihon Kohden) triggered by the electrocardiogram. A flow probe (1RB, Transonic Systems, Ithaca, NY), which was connected to an ultrasonic flowmeter (T106, Transonic Systems), was positioned around the root of the right femoral artery for the measurement of mean femoral arterial blood flow. Femoral vascular resistance was calculated from the following equation: femoral vascular resistance (in mmHg·min·ml−1) = mean AoP/femoral arterial blood flow. Data were continuously recorded on a chart recorder (8K-23, NEC San-ei Instrument, Tokyo, Japan).
MTX (5 mg/kg ip), which was dissolved in PBS (pH 7.4), or the vehicle (PBS) was given 4 h before an intravenous injection of SEP or saline (0.9% NaCl). The doses of MTX and SEP and the timing of administration were determined by referring to previous reports (36, 42). In the previous studies, MTX exerted nearly complete inhibition of BH4 formation from its precursors in vivo. Thus, it is likely that MTX, at the dose used in this study, blocked DHFR almost completely. Rats were allowed to stabilize for at least 30 min after completion of the operation before SEP or saline injection, and the baseline values of cardiovascular parameters were measured.
Depressor responses, estimated by the peak reduction in diastolic AoP, to intravenous bolus injections of ACh (0.05, 0.1, 0.2, and 0.5 μg/kg) and sodium nitroprusside (SNP; 1, 2, 5, and 10 μg/kg) were consecutively tested 10 min after a bolus injection of SEP (0.3 mg/kg) or saline (0.5 ml/kg). The results were compared between the data obtained in MTX- or MTX/SEP-treated rats and in control rats. Additionally, pressor responses to norepinephrine (1.0 μg/kg iv) and depressor responses to l-isoproterenol (0.1 μg/kg iv) were tested.
Measurement of biopterins.
BH2 and BH4 levels in the thoracic aorta were analyzed at the end of the experiments in rats treated with PBS/saline (n = 10), MTX/saline (n = 7), or MTX/SEP (n = 8). The thoracic aorta was freshly excised after bleeding and homogenized in 10% (wt/vol) extraction buffer [50 mM Tris·HCl (pH 7.4), 1 mM EDTA, and 10 mM DTT] using a glass mortar and pestle. The homogenate was centrifuged at 4°C and 20,630 g for 10 min. The supernatant was stored at −30°C.
Aortic biopterin contents were determined by a HPLC system (LV-10AD, Shimadzu, Kyoto, Japan) according to the method of Fukushima and Nixon (19) with some modifications (21). The amount of BH4 was estimated from the difference between the total biopterin (acid oxidized biopterin level = BH4 + BH2 + biopterin) and alkaline-stable biopterin (alkaline oxidized biopterin level = BH2 + biopterin). Since the amount of fully oxidized biopterin detected in fresh tissue extracts is reportedly insignificant (42), the biopterin measured as alkaline-stable biopterin is termed “BH2” or referred to as “oxidized biopterin” throughout this report.
Measurement of NO metabolites.
A blood sample obtained from the same rat used for the BH4 assay was deproteinized by the addition of an equal volume of methanol. The plasma levels of nitrite and nitrate, NO metabolites, were measured using the Griess method with an automated NO detector HPLC system (ENO-20, Eicom, Kyoto, Japan), as described previously (54).
Organ chamber experiments.
After bleeding at 60 min after the administration of SEP or saline, the abdominal aorta was excised in separate rats (n = 5–6) and rapidly placed in cold Krebs-Henseleit solution (KHS) to remove fat and connective tissues from the artery. The composition of KHS was (in mM) 120 NaCl, 4.8 KCl, 1.25 CaCl2, 1.2 MgSO4, 1.2 KH2PO4, 25.0 NaHCO3, and 11.0 glucose. The aorta was cut into 2-mm ring segments. One end of the artery ring was fixed in an organ bath, which was filled with KHS maintained at 37°C and aerated with 95% O2-5% CO2. The other end was connected to an isometric force-displacement transducer (TB-611T, Nihon Kohden) for the measurement of tension. After 90 min of stabilization under an optimal resting tension of 1 g, ring segments were contracted with KCl (30 mM) to test their viability and thereafter rinsed three times with KHS. The vasorelaxing responses of phenylephrine (10−6 M)-precontracted rings to ACh (10−9–10−4 M) or SNP (10−10–10−5 M) were evaluated. These vasoactive agents were cumulatively added to the organ bath. At the end of the experiments, each segment was maximally relaxed by 10−4 M papaverine hydrochloride to confirm the sufficient relaxant ability of the preparation. Cumulative relaxation data were expressed as a percentage of phenylephrine-induced contraction.
Assessment of vascular superoxide production.
Superoxide anion production in aortic segments was determined using lucigenin-enhanced chemiluminescence according to previously published methods (14, 39). Thoracic aortic rings (3 mm long) were prepared as described above and then placed in ice-cold Krebs-HEPES buffer of the following composition (in mM): 99.0 NaCl, 4.7 KCl, 2.5 CaCl2, 1.2 MgSO4, 1.0 KH2PO4, 25.0 NaHCO3, 20.0 Na-HEPES, and 5.6 glucose (pH 7.35). Aortic rings were incubated at 37°C in Krebs-HEPES buffer bubbled with 95% O2-5% CO2 for 30 min in the absence and presence of SOD (150 U/ml). In some of the segments, NG-nitro-l-arginine methyl ester (l-NAME; 1 mM), a NOS inhibitor (3), or NADH (0.1 mM), an electron donor used to increase vascular NAD(P)H oxidase activity (13), was added to the incubation buffer for 30 min. The specimen was transferred to a vial containing 5 μM lucigenin (N,N′-dimethyl-9,9′-biacridinium dinitrate) and placed in a luminometer (MiniLumat LB9506, Berthold Technologies, Bad Wildbad, Germany). After 15 min of equilibration, counts over 30-s periods were obtained every 30 s for the next 5 min and averaged. Background counts obtained in the absence of tissue were subtracted from the average of each sample. The wet weight of each specimen was measured for normalization of the count. Data were expressed as relative light units (RLU) per milligram of wet weight.
Western blot analysis.
Thoracic aortic tissue was freshly isolated and snap frozen in liquid nitrogen. The artery was homogenized in 0.6 ml of ice-cold homogenization buffer containing 50 mM Tris (pH 8.0), 150 mM NaCl, 0.1% SDS, 20 mM CHAPS, and 0.5% Complete protease inhibitor cocktail (Roche, Mannheim, Germany). After homogenization, the tissue homogenate was centrifuged at 15,000 g for 10 min at 4°C, and the supernatant was used for Western blot analysis. To detect total eNOS, phosphorylated eNOS, and actin, 50 μg protein was solubilized in Laemmli sample buffer containing 2.5% 2-mercaptoethanol at 60°C for 3 min and subjected to 6% SDS-PAGE at room temperature. To evaluate the eNOS dimer (active form)-to-monomer (inactive form) ratio, low-temperature SDS-PAGE was run at 70 V and 4°C using nonheated tissue homogenates and reducing sample buffer, as previously reported (22, 27). Gels were transferred to nitrocellulose membranes, and membranes were blocked with 5% nonfat milk. They were then incubated with specific primary antibodies against eNOS, phosphorylated Ser1177 eNOS (BD Transduction Laboratories, Franklin Lakes, NJ), or β-actin (Amersham-GE Healthcare, Buckinghamshire, England) in 5% nonfat milk and Tris-buffered saline-Tween. After being washed and incubated with horseradish peroxidase-conjugated secondary antibodies (Amersham-GE Healthcare), membranes were developed with enhanced chemiluminescence (Amersham-GE Healthcare). Quantitative densitometry was performed by a lumino-image analyzer (LAS-4000 mini EPUV, Fuji Film, Tokyo, Japan) and Multi Gauge (version 3.0) software.
The drugs used in this study were MTX (Wako, Osaka, Japan), acetylcholine chloride (Dai-ichi, Tokyo, Japan), norepinephrine (Sankyo, Tokyo, Japan), papaverine hydrochloride (Takeda, Osaka, Japan), phenylephrine hydrochloride, l-isoproterenol hydrochloride (Kowa, Nagoya, Japan), SEP, SNP, SOD, NADH, and l-NAME (Sigma, St. Louis, MO). Drugs were dissolved in or diluted with physiological saline solution.
Dose-response data were analyzed by repeated-measures ANOVA followed by a Bonferoni/Dunn test. For multiple comparisons between groups, one-way ANOVA was used with a Dunnett post hoc test. Unpaired data were analyzed by an unpaired Student's t-test. All statistical tests were performed using the computer program StatView-J 5.0 (SAS Institute Japan, Tokyo, Japan). The level for statistical significance was P < 0.05. All results are expressed as means ± SE.
Effect of combined treatment with MTX and SEP on aortic BH4 and BH2 levels.
Treatment with MTX, a DHFR inhibitor blocking the intracellular conversion of BH2 to BH4, did not significantly affect aortic BH4 (Fig. 1A) or BH2 (Fig. 1B) levels compared with control treatment. On the other hand, vascular BH2 contents in rats that received MTX/SEP increased markedly compared with those in vehicle-receiving control rats (291.1 ± 29.2 vs. 33.4 ± 6.4 pmol/g wet wt, P < 0.01; Fig. 1B), although the vascular contents of BH4 were comparable with the control group (Fig. 1A). The ratio of aortic BH4 to BH2 levels significantly decreased not only in MTX/SEP-treated rats but also in rats treated with MTX alone (Fig. 1C).
Effect of combined treatment with MTX and SEP on hemodynamic responses to ACh and SNP in vivo.
Baseline hemodynamic variables just before the administration of SEP or saline in the three groups of anesthetized rats are shown in Table 1. These values were not significantly different between the control group and treated groups (Table 1). Systolic, diastolic, and mean blood pressures 30 min after SEP injection in MTX/SEP-treated rats, but not in MTX-treated rats, were significantly elevated by 12%, 13%, and 12% above those in control rats, respectively (Fig. 2, A–C), and mean blood pressure was significantly correlated positively with aortic BH2 content (Fig. 2D) and inversely with the ratio of aortic BH4 to BH2 contents (Fig. 2E).
Depressor and vasodilator responses to graded doses of ACh, an endothelium-dependent vasodilator, in rats treated with MTX alone were not significantly altered compared with those in the control group (Fig. 3A), whereas those in rats treated with MTX/SEP were significantly attenuated (P < 0.05; Fig. 3B). On the other hand, depressor and vasodilator responses to graded doses of SNP, an endothelium-independent NO donor, in MTX-pretreated groups irrespective of SEP or saline administration were not significantly different from those in the control group (Fig. 4). In addition, depressor responses to isoproterenol (n = 4 and 3), a non-NO-mediated vasodilator, and pressor responses to norepinephrine (n = 8 and 7) were both unaffected by the treatment with MTX/SEP or MTX alone compared with the corresponding responses of the control group (n = 10).
Effect of combined treatment with MTX and SEP on vascular reactivity ex vivo.
In harmony with the above results obtained in situ experiments, ACh-induced relaxations of isolated aortas from rats receiving MTX/SEP but not MTX alone were significantly diminished without affecting SNP-induced relaxations of the aortic segments (Fig. 5). Thus, these results consistently indicate that combined treatment with MTX/SEP selectively impairs endothelium-dependent vasodilator function, whereas treatment with MTX alone does not cause endothelial dysfunction in spite of significantly decreasing the BH4-to-BH2 ratio.
Effect of combined treatment with MTX and SEP on plasma concentrations of nitrite plus nitrate.
Plasma concentrations of nitrite plus nitrate, a marker of NO production, in rats with combined treatment with MTX/SEP (8.4 ± 0.4 μmol/l, n = 8, P = 0.068) and with MTX alone (9.1 ± 0.4 μmol/l, n = 7, P = 0.357) tended to decline compared with control rats (10.1 ± 0.6 μmol/l, n = 10), although they did not reach a statistically significant level.
Effect of combined treatment with MTX and SEP on vascular superoxide production.
Basal superoxide production, as assessed by lucigenin-enhanced chemiluminescence, which was sensitive to SOD (150 U/ml), of thoracic aortas from rats treated with MTX/SEP (14.7 ± 1.3 RLU/mg, n = 7) was significantly greater by 43% (P < 0.05) than that in the control group (10.3 ± 0.6 RLU/mg, n = 6), as shown in Fig. 6A. Importantly, this statistically significant difference in superoxide production disappeared when assayed in the presence of l-NAME, a NOS inhibitor (Fig. 6B), suggesting that eNOS-derived superoxide production is involved in the higher basal superoxide production in the MTX/SEP-treated group. The addition of NADH (0.1 mM) to the aortic segment produced remarkable increases in superoxide, but NADH-stimulated superoxide levels in the MTX/SEP-treated group and the control group were comparable (77.8 ± 6.7 RLU/mg, n = 7, and 73.1 ± 6.9 RLU/mg, n = 5, respectively), implying that there was no difference in vascular NAD(P)H activity in the two groups.
Effect of SOD on the vascular reactivity of rats treated with MTX and SEP.
The application of SOD into the organ chamber did not affect ACh-induced relaxations of phenylephrine-precontracted aortic rings from control rats (Fig. 7A) but significantly ameliorated impaired ACh-induced relaxations in MTX/SEP-treated rats (Fig. 7B). In contrast, SNP-induced relaxations of aortic segments in both control and MTX/SEP-treated groups were unaltered by the treatment with SOD (both n = 6; data not shown).
Effect of combined treatment with MTX and SEP on eNOS dimerization and phosphorylation.
Combined treatment with MTX/SEP did not significantly affect the eNOS dimer-to-monomer ratio in aortas compared with the control treatment (n = 5; Fig. 8). MTX/SEP treatment also had no effect on the levels of phosphorylated Ser1177 eNOS or total eNOS protein in aortas compared with the control treatment (n = 5; Fig. 9).
This is the first study to examine the effect of increasing the vascular BH2 concentration on endothelial function in vivo. To increase endogenous vascular BH2 levels in vivo, we used pharmacological treatment with the BH2 precursor SEP and the DHFR inhibitor MTX together instead of exogenously given BH2 itself. In consequence, the treatment yielded a ninefold increase in aortic BH2 levels in rats without significant changes in BH4 levels. Thus, the experimental conditions seem quite suitable for estimating the effect derived exclusively from a generous amount of BH2 on endothelial function.
The present study demonstrates that the abundance of vascular BH2 produced by the combined treatment with MTX/SEP is associated with moderate hypertension and a selective impairment of endothelium-dependent vascular dilatation elicited by ACh. Elevation of blood pressure is primarily caused by a rise in vascular tone in systemic resistance vessels. In this study, an increase in vascular BH2 levels may cause endothelial dysfunction under both agonist-stimulated and basal conditions, leading to a slight elevation of systemic blood pressure. Despite significantly decreased ratios of BH4 to BH2 contents being seen in rats treated with MTX alone, the treatment did not affect vasodilator responses to ACh or SNP. These findings implicate that increasing endogenous BH2 levels, rather than BH4 deficiency, may contribute substantially to endothelial dysfunction in the intact animal at least under a condition in which a BH2 precursor was given after the inhibition of DHFR. These data also indicate that a reduction in the ratio of BH4 to BH2 does not always relate to an impairment of endothelial function. Indeed, we (38) have recently shown that endothelial dysfunction seen in a model of genetic hypertension is associated with even increased vascular contents of BH4 but without alteration in the BH4-to-oxidized biopterin ratio.
We observed that the profound increases in BH2 levels seen in aortas of rats with the combined treatment were also accompanied by elevated superoxide generation and that the impairment of ACh-induced relaxation was recovered by SOD. These results indicate that the endothelial dysfunction observed in the MTX/SEP-treated group may have resulted from superoxide production. Moreover, the significant rise in superoxide was diminished in the presence of l-NAME, suggesting that the source of the enhanced superoxide production in the MTX/SEP group is, at least partly, eNOS. It is generally recognized that NADPH oxidases are the predominant sources of ROS in the vasculature (4). In this case, however, considering our data that the NADH-stimulated production of superoxide in aortas of rats that had received MTX/SEP was comparable with the control group, the participation of NADPH oxidase-derived superoxide may have been limited in the present experiments. Accordingly, it is most likely that the endothelial dysfunction produced by the abundance of endogenous BH2 in the present study is attributable in part to eNOS uncoupling. Although the eNOS dimer-to-monomer ratio, a marker of eNOS uncoupling (55), was unaltered by MTX/SEP treatment, this negative result might not entirely exclude the possibility of involvement of eNOS uncoupling in our study because it has been reported that eNOS uncoupling is not always associated with eNOS monomerization (2, 18). In harmony with our findings, a previous in vitro study (53) demonstrated that the eNOS homodimer-to-monomer ratio is not dependent on changes in levels of BH2. In line with our findings, previous in vivo studies using various disease models suffering from oxidative stress, such as insulin resistance in high-fructose diet-fed rats (44), diabetes in db/db mice (40), systemic hypertension in Dahl-salt sensitive rats (46), pulmonary hypertension in lambs (20), and hyperglycemia in Zucker diabetic fatty rats (9), have shown that there is an increase in BH2 in the vasculature, which was associated with diminished NO availability and increased superoxide production by eNOS. BH2 lacks the cofactor activity for NOS but competes with BH4 at the binding site of NOS with equal affinity (9, 26). Thus, it is conceivable that an increase in the endothelial BH2 level results in impairment of endothelial function by compromising the normal BH4 bioavailability. Therefore, in these previous studies and our study, the mechanism involved in eNOS dysfunction can be explained mostly by a relative deficiency of BH4 based on a hypothesis that BH2 accumulated in endothelial cells may bind eNOS with significant affinity rather than being an inert product of BH4 oxidation (9). On the other hand, phosphorylation or dephosphorylation of eNOS at specific residues has been shown to modulate superoxide generation from uncoupled eNOS independently of BH4 (8, 30). However, phosphorylation of eNOS at Ser1177 was not significantly changed by MTX/SEP treatment. It is still unknown whether or not the treatment might have an impact on the phosphorylation state of eNOS at other functional residues (17). A recent cellular study (45) demonstrated that an increase in intracellular BH2 levels provokes inhibition of dephosphorylation of eNOS at Ser116 and was associated with a decrease in eNOS activation and an increase in ROS production. These findings convey the latest view that some of the effect of oxidized BH4 on NOS-dependent ROS synthesis may reflect changes in the NOS phosphorylation state (32), although a mechanistic explanation concerning the molecular linkage between BH2 and eNOS phosphorylation remains unknown.
Our findings obtained with intact animals suggest that DHFR plays a crucial role in determining endothelial function, particularly in the presence of an intracellular BH2 burden. The importance of endothelial DHFR as a key determinant of eNOS dysfunction was first reported by Chalupsky and Cai (6) and further confirmed by different laboratories (10, 45) using DHFR-targeted RNA interference in cultured cells. Nevertheless, in contrast to these studies with genetic knockdown of DHFR, our data with MTX-alone treatment showed insignificant changes in aortic BH4 levels, indicating that BH4 formation via the salvage pathway might be trivial under the physiologically normal state in the intact animal, as it is usually considered that intracellular BH4 levels are primarily regulated by the activity of the rate-limiting enzyme GTP cyclohydrolase 1 in the de novo pathway (34, 37). However, the distinct role of the salvage pathway in determining BH4 levels under in vivo normal conditions remains obscure. Further in vivo studies may be needed to clarify whether DHFR in the endothelium is of value as a therapeutic target for various cardiovascular disorders. In this connection, several recent articles have shown that decreased activity or expression of DHFR is involved in eNOS dysfunction in angiotensin-treated endothelial cells (6), ischemia-induced impairment of coronary vasodilatation in isolated rat hearts (15), and streptozotocin-induced diabetes in mice (41).
A number of investigations have demonstrated that supplementation with BH4 improves impaired endothelial function in experimental animals or humans with major cardiovascular risk factors or with common cardiovascular disorders (18, 24), forming the basis for several clinical trials. In contrast, some reports have shown that acute incubation with BH4 or the precursor SEP further impairs endothelium-dependent relaxation in isolated arteries (23, 33, 48, 50). These unfavorable effects of exogenous BH4 or SEP on endothelial function have so far been explained primarily by the autooxidation of BH4 (12, 25). Alternatively, these disappointing results might be ascribed, in part, to the intracellular accumulation of BH2, similar to the present study in that conversion of BH2 to BH4 by DHFR is significantly attenuated. In clinical circumstances, MTX has been a widely used agent for cancer (5) and is the preferred therapeutic for rheumatoid arthritis (1). Thus, caution might be taken regarding a potential untoward effect when administration of SEP is considered in patients, particularly in those with a diminished DHFR activity due to oxidative stress-associated cardiovascular disorders or an antifolate drug like MTX.
In conclusion, the present study shows that the abundance of vascular BH2, even in the absence of a deficiency of absolute amount of BH4, causes endothelial dysfunction, at least partly, through eNOS uncoupling in rats in vivo.
No conflicts of interest, financial or otherwise, are declared by the author(s).
- Copyright © 2011 the American Physiological Society