Oxidation of tetrahydrobiopterin leads to uncoupling of endothelial cell nitric oxide synthase in hypertension (original) (raw)
The endothelial NO synthase produces ROSs in hypertensive vessels. In rats with DOCA-salt hypertension (blood pressure, 189 ± 4 mmHg vs. 126 ± 2 mmHg; P < 0.01), vascular superoxide production was increased fourfold as compared to sham-operated rats (Figure 1a). Removal of the endothelium significantly reduced superoxide levels in hypertensive rats but had no effect in normotensive rats (Figure 1a), suggesting that a significant portion of the increase in O2•– formation in hypertensive vessels is derived from the endothelium. Because NO synthase is abundant in the endothelium, we determined if this enzyme contributed to O2•– production in hypertension. The NO synthase inhibitor L-nitroarginine methyl ester (L-NAME) reduced O2•– production in hypertensive vessels but not in control vessels, suggesting that the NO synthase is a significant source of this radical in hypertension (Figure 1a).
Role of the endothelial NO synthase as a source of ROSs in hypertensive vessels. (a) Effect of endothelial removal and NO synthase inhibition (L-NAME, 1 mM) on vascular superoxide (O2•–) production estimated by lucigenin chemiluminescence in sham-operated and DOCA-salt–hypertensive rats (n = 6–10 per group). (b) Vascular O2•– production estimated by lucigenin chemiluminescence in sham-operated and DOCA-salt–hypertensive wild-type (C57BL/6), eNOS–/–, and nNOS–/– knockout mice (n = 6–10 per group). (c) Vascular O2•– production measured by the SOD-inhibitable cytochrome c reduction assay in sham-operated and DOCA-salt–hypertensive wild-type, eNOS–/–, and nNOS–/– knockout mice (n = 5–13 per group).
To confirm this finding and to determine the NOS isoform responsible for O2•– production, we performed additional studies in genetically altered mice. In wild-type (C57BL/6) mice, we found that DOCA-salt hypertension produced an increase in vascular O2•– production very similar to that observed in rats (Figure 1, b and c). This increase in O2•– production was largely blocked when NO synthase was inhibited by L-NAME (Figure 1, b and c). Neuronal NOS expression has been reported to be increased in aortas of spontaneously hypertensive rats (33) and could serve as a source of O2•– inhibitable by L-NAME. However, in _nNOS_–/– mice with DOCA-salt hypertension (blood pressure, 134 ± 4 mmHg), vascular O2•– production was increased to a similar extent as that observed in wild-type mice (Figure 1, b and c). As in wild-type mice, L-NAME reduced O2•– production in nNOS–/– mice. In contrast, in eNOS–/– mice with DOCA-salt hypertension (blood pressure, 144 ± 5 mmHg), there was only a modest increase in O2•– production, and this was not altered by L-NAME (Figure 1, b and c). These findings strongly support the concept that DOCA-salt hypertension is associated with an increase in O2•– production from eNOS and that nNOS does not significantly contribute to this phenomenon.
One explanation for an increase in O2•– production in hypertensive vessels might relate to a decrease in the activity of SOD in these vessels. Total SOD activity, however, was identical in vessels of sham-operated and hypertensive mice, averaging 7.6 ± 1.0 U/mg of protein and 7.4 ± 1.1 U/mg of protein, respectively (n = 3 for each; difference not significant).
Oxidation of tetrahydrobiopterin in hypertension. Tetrahydrobiopterin plays a crucial role as a cofactor for all NO synthases (22, 34). In its absence, purified eNOS produces ROSs rather than NO (15–17). Thus, one mechanism whereby eNOS could become uncoupled in vivo may be due to diminished levels of tetrahydrobiopterin. In preliminary experiments, we found that endothelial removal before homogenization of vessels reduced total biopterin levels by 60–70%, suggesting that the major site of tetrahydrobiopterin synthesis in the vessel wall is the endothelium. Total biopterin content in aortas from control and DOCA-salt–hypertensive mice was similar (Figure 2). In vessels from mice with DOCA-salt hypertension, however, the tetrahydrobiopterin content was substantially reduced, and the content of oxidized forms of tetrahydrobiopterin (7,8-dihydrobiopterin and biopterin) was correspondingly increased.
Tetrahydrobiopterin and oxidized biopterin content in aortas from sham-operated and DOCA-salt–hypertensive mice: effect of oral treatment with tetrahydrobiopterin, p47phox, or eNOS deficiency. Tetrahydrobiopterin (H4B) and oxidized biopterin (7,8-H2B plus biopterin) were determined using HPLC analysis after differential oxidation (n = 3–4 per group). Three to six aortas were pooled for each measurement. *P < 0.05 versus sham, **P < 0.05 versus DOCA (C57BL/6).
These data suggest that oxidation of tetrahydrobiopterin leads to tetrahydrobiopterin deficiency in hypertension (Figure 2) and raise the question of what may be the initial source of ROSs in hypertension leading to tetrahydrobiopterin oxidation. Previous studies have shown that peroxynitrite, the reaction product of NO and O2•–, and a variety of other oxidants that could potentially be derived from O2•– can oxidize tetrahydrobiopterin (26, 35, 36). We hypothesized that the vascular NADPH oxidase might be an initial source of O2•– that leads to formation of peroxynitrite and other oxidants, since this enzyme system represents an important source of O2•– in endothelial cells (37, 38). We therefore studied mice lacking p47_phox_, a critical component of the vascular NADPH oxidase (39, 40). In contrast to the findings in wild-type mice with DOCA-salt hypertension, levels of tetrahydrobiopterin and oxidized biopterin were similar to those observed in sham-operated mice (Figure 2). In addition, there was no increase in vascular O2•– production in DOCA-salt–hypertensive p47_phox_-deficient mice (Figure 3). In contrast to DOCA-salt–treated wild-type mice, exposure of aortas from p47_phox_-deficient mice with DOCA-salt hypertension to L-NAME did not reduce O2•– production (Figure 3). These findings indicate that in the absence of a functioning NADPH oxidase, eNOS is not uncoupled by DOCA-salt hypertension. A significant role of reactive nitrogen species derived from eNOS in tetrahydrobiopterin oxidation is suggested by the fact that in DOCA-salt–treated eNOS-deficient mice, there was less tetrahydrobiopterin oxidation than in hypertensive wild-type mice (Figure 2).
Effect of NADPH oxidase deficiency (p47phox–/–) and treatment with tetrahydrobiopterin (H4B) on vascular O2•– production in DOCA-salt hypertension. (a) Vascular O2•– production estimated by lucigenin chemiluminescence in sham-operated and DOCA-salt–hypertensive mice (n = 6–10 per group). (b) Vascular O2•– production measured by the SOD-inhibitable cytochrome c reduction assay (n = 5–13 per group).
Effect of oral treatment with tetrahydrobiopterin in hypertension. The above studies demonstrated that eNOS is uncoupled in hypertension and suggest that tetrahydrobiopterin oxidation may underlie this phenomenon. To further evaluate this concept, we treated mice with oral tetrahydrobiopterin and measured vascular O2•– production. Treatment with oral tetrahydrobiopterin (5 mg per day), beginning one day after surgery, increased tetrahydrobiopterin by about twofold in hypertensive mice and to a lesser extent in control mice but did not change the levels of oxidized biopterin in either of these groups (Figure 2). Tetrahydrobiopterin treatment resulted in a substantial reduction of O2•– production in mice with DOCA-salt hypertension (Figure 3). Of note, L-NAME no longer caused a decrease in O2•– production in aortas of hypertensive mice treated with tetrahydrobiopterin (Figure 3, a and b). In vessels of sham-operated mice there was no effect of treatment with tetrahydrobiopterin on O2•– production (Figure 3).
Reduced pteridines such as tetrahydrobiopterin may exert antioxidant effects in vitro that could contribute to the effect on O2•– production we observed in hypertensive mice fed H4B (41). Treatment with oral tetrahydroneopterin (H4N, 5 mg per day), which has antioxidant properties similar to those of H4B (42) but poorly sustains eNOS catalysis, failed to reduce vascular O2•– production in mice with DOCA-salt hypertension (O2•– production, 5.87 ± 2.1 × 103 counts/mg per minute; n = 4). This indicates that tetrahydrobiopterin is not simply acting as an antioxidant in hypertension.
Because uncoupling of eNOS not only causes the enzyme to produce O2•– but also decreases NO formation, we studied the effect of oral tetrahydrobiopterin treatment on NO formation, reflected by nitrosyl hemoglobin levels in the blood. In hypertensive mice, nitrosyl hemoglobin levels, as determined by ESR measurements, were significantly reduced as compared with those in sham-operated mice (Figure 4). This abnormality was corrected by oral tetrahydrobiopterin treatment (Figure 4). In keeping with the concept that the NADPH oxidase is at least in part responsible for oxidation of tetrahydrobiopterin and reducing NO production in hypertension, we found that nitrosyl hemoglobin levels in hypertensive p47_phox_-deficient mice were similar to those observed in controls (Figure 4).
Effect of NADPH oxidase deficiency (p47_phox–/–_) and tetrahydrobiopterin (H4B) treatment on NO production. NO production was analyzed by ESR measurements of nitrosyl hemoglobin levels (n = 5).
Endothelium-dependent vascular relaxation in DOCA-salt hypertension: role of hydrogen peroxide. It has recently been demonstrated that depletion of tetrahydrobiopterin in intact vessels has minimal effect on endothelium-dependent vasodilation, but that these responses are markedly inhibited by the addition of catalase, suggesting that they are mediated by H2O2 (43, 44). We therefore examined the effect of catalase on endothelium-dependent vasodilation in aortas from sham-operated and DOCA-salt–hypertensive mice and in tetrahydrobiopterin-treated mice with DOCA-salt hypertension. Relaxations evoked by the calcium ionophore A23187 were only minimally reduced in vessels from hypertensive mice as compared with sham-operated mice (Figure 5a). Catalase had no effect on relaxations to A23187 in sham-operated vessels; however, in vessels from hypertensive mice, catalase inhibited these responses by approximately 50% (Figure 5a). This effect of catalase was no longer observed in hypertensive mice treated with tetrahydrobiopterin. Endothelium-independent relaxation to nitroglycerin was similar between hypertensive and sham-operated mice and was not affected by catalase (Figure 5b). The vasorelaxation studies were not performed in eNOS-deficient mice, since we have previously shown that aortas from these mice lack relaxation in response to A23187 (32).
Endothelium-dependent vascular relaxation in DOCA-salt hypertension: role of hydrogen peroxide. Effect of catalase (1200 U/ml) on endothelium-dependent vasorelaxations to calcium ionophore A23187 (a) and endothelium-independent vasorelaxations to nitroglycerin (b) in aortas from sham-operated, DOCA-salt–hypertensive mice and hypertensive mice treated with H4B (n = 6–10, *P < 0.05). (c) H2O2 production in aortas from sham-operated, DOCA-salt–hypertensive mice (with and without oral H4B treatment) and DOCA-salt–treated eNOS–/– and p47_phox–/–_ mice. H2O2 production was determined by the fluorometric Amplex red assay in the presence of A23187 (10–6 M). The effect of NO synthase inhibition (1 mM L-NAME) and PEG catalase (1000 U) was studied (n = 3–6).
In additional experiments, we examined the ability of A23187 to stimulate release of H2O2 from vascular segments using the Amplex red assay. As shown in Figure 5c, the production of H2O2 in response to A23187 was increased by approximately twofold in vessels from mice with DOCA-salt hypertension. This increased response was not observed if vessels were treated with L-NAME, if the animals were treated with oral tetrahydrobiopterin, or when either eNOS- or p47_phox_-deficient mice were studied (Figure 5c). Endothelial removal decreased vascular H2O2 production in DOCA-salt–hypertensive mice (from 283 ± 21 pmol H2O2 per milligram to 223 ± 22 pmol H2O2 per milligram, P < 0.05) but had no significant effect in sham mice (133 ± 16 pmol H2O2 per milligram vs. 167 ± 18 pmol H2O2 per milligram). The specificity of the Amplex red signal for H2O2 was confirmed using PEG catalase (1000 U) (Figure 5c).
Taken together, these data suggest that endothelium-dependent vasodilation is in part mediated by H2O2 in the setting of DOCA-salt hypertension. Our studies of H2O2 production strongly suggest that a major source of H2O2 in this situation is uncoupled eNOS.
Effect of tetrahydrobiopterin treatment and NADPH oxidase deficiency on the progression of hypertension. The effect of tetrahydrobiopterin treatment and NADPH oxidase deficiency on blood pressure was evaluated at 10, 20, and 30 days after DOCA-salt treatment. At the early time point (10 days), there was a trend toward a lower blood pressure in tetrahydrobiopterin- and NADPH oxidase–deficient DOCA-salt mice (Table 1). In animals with untreated DOCA-salt hypertension, blood pressure continued to increase for the ensuing 20 days. In mice treated with tetrahydrobiopterin and mice lacking p47phox, blood pressures remained unchanged after the initial increase at 10 days and were statistically lower than those in untreated DOCA-salt mice (Table 1).
Effect of oral H4B treatment or NADPH oxidase deficiency on systolic blood pressure (mmHg)