Pivotal Role for Endothelial Tetrahydrobiopterin in Pulmonary Hypertension
http://www.100md.com
循环学杂志 2005年第4期
the Department of Cardiovascular Medicine, University of Oxford, John Radcliffe Hospital, Oxford, UK (J.P.K., N.J.A., J.K.B., T.N., K.M.C.)
Department of Experimental Medicine and Toxicology, Imperial College London, Hammersmith Hospital, London, UK (L.Z., M.R.W.)
Childhood Infection Group, Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, UK (K.R.).
Abstract
Background— Pulmonary hypertension is a fatal disease characterized by vasoconstriction and vascular remodeling. Loss of endothelial nitric oxide bioavailability is implicated in pulmonary hypertension pathogenesis. Recent evidence suggests that the cofactor tetrahydrobiopterin (BH4) is an important regulator of nitric oxide synthase enzymatic function.
Methods and Results— In the hph-1 mouse with deficient BH4 biosynthesis, BH4 deficiency caused pulmonary hypertension, even in normoxic conditions, and greatly increased susceptibility to hypoxia-induced pulmonary hypertension. In contrast, augmented BH4 synthesis in the endothelium, by targeted transgenic overexpression of GTP-cyclohydrolase I (GCH), prevented hypoxia-induced pulmonary hypertension. Furthermore, specific augmentation of endothelial BH4 in hph-1 mice by crossing with GCH transgenic mice rescued pulmonary hypertension induced by systemic BH4 deficiency. Lung BH4 availability controlled pulmonary vascular tone, right ventricular hypertrophy, and vascular structural remodeling in a dose-dependent manner in both normoxia and hypoxia. Furthermore, BH4 availability had striking effects on the immediate vasoconstriction response to acute hypoxia. These effects of BH4 were mediated through the regulation of nitric oxide compared with superoxide synthesis by endothelial nitric oxide synthase.
Conclusions— Endothelial BH4 availability is essential for maintaining pulmonary vascular homeostasis, is a critical mediator in the pathogenesis of pulmonary hypertension, and is a novel therapeutic target.
Key Words: hypertension, pulmonary ; tetrahydrobiopterin ; superoxide ; endothelium ; nitric oxide synthase
Introduction
In a normal oxygen environment, the healthy adult pulmonary vascular bed is a low-pressure circuit. Pulmonary hypertension arises when vasoconstriction and structural remodeling of pulmonary arterioles lead to increased pulmonary vascular resistance. The consequent increase in pressure load causes right ventricular (RV) hypertrophy, which progresses to premature death from right-sided heart failure.1 Pulmonary hypertension can develop in isolation (idiopathic pulmonary arterial hypertension) or in association with conditions such as connective tissue diseases, congenital heart disease, pulmonary thromboembolism, or chronic hypoxia.2 Mutations in the BMPR2 gene, encoding a receptor for bone morphogenetic protein, appear to underlie some cases of familial idiopathic pulmonary arterial hypertension. However, BMPR2 mutations account for only 10% of sporadic idiopathic pulmonary arterial hypertension, and only 15% to 20% of individuals carrying a BMPR2 mutation develop pulmonary hypertension, suggesting that additional disease-causing or disease-modifying genes remain critical to understanding pulmonary hypertension pathogenesis.3
See p 2022
The endothelium plays a key role in pulmonary vascular homeostasis. Endothelial dysfunction is clearly implicated in the pathogenesis of pulmonary hypertension, but the exact mechanisms remain poorly understood.4 Endothelium-derived nitric oxide (NO), through potent vasodilator and antiproliferative effects on vascular smooth muscle cells, is critical in maintaining normal pulmonary vascular tone and structure.5 Mice deficient in endothelial NO synthase (eNOS) are more sensitive to hypoxia-induced pulmonary hypertension,6,7 whereas pulmonary gene transfer of eNOS is partially protective.8 Patients with pulmonary hypertension have low NO levels in their exhaled breath,9,10 and both inhaled NO and phosphodiesterase type 5 inhibitors, which act to increase NO-mediated cGMP signaling, have emerged as therapeutic strategies in pulmonary hypertension. Paradoxically, mice with hypoxia-induced pulmonary hypertension have increased eNOS protein levels without a concomitant increase in NO bioactivity.11,12 Scavenging by superoxide radicals may account in part for reduced NO bioactivity.13 Increased superoxide production has been observed in experimental models of pulmonary hypertension,14 and markers of oxidative stress are increased in patients with pulmonary hypertension.15 Superoxide mediates pulmonary vasoconstriction16 and stimulates pulmonary smooth muscle cell proliferation.17 These observations suggest that perturbation of NO synthesis, together with increased superoxide production, may play a central role in the pathogenesis of pulmonary hypertension.
The cofactor tetrahydrobiopterin (BH4) is an important regulator of NOS enzymatic activity.18 Without BH4, NO production is reduced and superoxide production is increased as a result of the loss of enzymatic coupling between the reduction in molecular oxygen and oxidation of L-arginine.19 In the systemic vasculature, BH4-mediated eNOS regulation appears to be an important determinant of NO bioavailability, independent of total eNOS protein levels,19,20 but the role of BH4 in the pulmonary circulation remains unexplored. We hypothesized that endothelial BH4 availability is a critical determinant of both the acute and chronic responses to hypoxia and the development of pulmonary hypertension, mediated through the effect of BH4 on eNOS activity and enzymatic coupling. To test this hypothesis, we compared pulmonary hemodynamics and the development of hypoxic pulmonary hypertension in genetic mouse models with graded reductions in BH4 levels; then, we tested the ability of transgenic endothelium-targeted augmentation of BH4 to rescue the effects of BH4 deficiency and protect against hypoxia-induced pulmonary hypertension. Furthermore, we investigated the effect of these changes in pulmonary BH4 availability on local NO compared with superoxide production by eNOS and the functional consequences of BH4-mediated eNOS regulation on acute pulmonary vasomotor function.
Methods
Mice
The hph-1 mouse, generated by ENU mutagenesis, was used as a model of BH4 deficiency. In these animals, tissue BH4 levels are low because of constitutively reduced expression of GTP cyclohydrolase I (GTPCH), the rate-limiting enzyme in BH4 biosynthesis.21–23 Wild-type (WT), hph-1 heterozygous (+/–), and hph-1 homozygous (hph) littermates on a C57BL/6 background were obtained by interbreeding hph-1 heterozygotes.23 In GCH-transgenic mice (GCH mice), overexpression of the human GTPCH under the control of the endothelial-specific Tie-2 promoter results in 2- to 3-fold elevation of endothelial BH4 levels.24 GCH mice used in these studies were bred for 8 generations onto a C57BL/6 background. To generate hph-1/GCH transgenic mice (hph/GCH), ie, hph-1 homozygotes carrying the Tie2-GCH transgene, hph-1 homozygotes were crossed with GCH mice, and offspring were backcrossed with hph-1. Mice were genotyped as previously described for hph-1 and GCH alleles.23,24 Mice were used at 15 weeks of age. Studies were conducted in accordance with the UK Home Office Animals (Scientific Procedures) Act 1986.
Normoxia and Chronic Hypoxia Studies
Animals were either housed in normal air or placed in a specially constructed normobaric hypoxic chamber (FIO2 10%) for 1 week.25 At 1 week, animals were weighed and anesthetized (Hypnorm 0.25 mL/kg; Midazolam 25 mg/kg IP), and RV systolic pressure (RVSP) was measured via direct cardiac puncture using a closed-chest technique in the spontaneously breathing, anesthetized animal.25 The animals were then euthanized, the hearts were removed, and the individual ventricular chambers were weighed (septum with left ventricle). The right lungs were snap-frozen in liquid nitrogen and stored at –80°C for biochemical measurements. The left lungs were fixed by inflation with 10% formalin, embedded in paraffin, and sectioned for histology.
BH4 Assay
Lung homogenates were oxidized with 1% iodine/2% KI in acid or base. Biopterins were determined by high-performance liquid chromatography (HPLC) with fluorescent detection, and BH4 was calculated by subtracting BH2+biopterin from total biopterins, as previously described.24
Lung Histology
Histological assessment of vascular remodeling was performed as previously described.25 Transverse lung sections were stained with van Gieson’s elastic method or smooth muscle -actin antibody (Clone 1A, Sigma) using an alkaline phosphate–conjugated avidin-biotin complex method and Vector Red substrate (Vector Laboratories), with methyl green counterstain (Vector Laboratories). "Distal muscularization" was defined as the proportion of vessels (<50 μm diameter) with immunoreactivity for smooth muscle -actin (as evidence for muscularization) over the total number of vessels stained with elastin. Three separate sections from each animal were quantified, and counting was performed by investigators blinded to genotype or treatment group.
eNOS Western Blotting
Lung homogenates were subjected to SDS-PAGE and transferred to nitrocellulose membranes. After transfer, membranes were incubated with mouse eNOS monoclonal antibodies (Transduction Labs), followed by horseradish peroxidase–conjugated secondary antibody. Proteins were visualized by chemiluminescence (SuperSignal, Pierce). Blot intensity was quantified with Scion Image software (release beta 4.0.2).
NOS Activity Assay
NOS activity was determined by 14C-arginine to 14C-citrulline conversion with HPLC quantification as described.24 Lung homogenates (400 μg protein) were incubated for 30 minutes at 37°C in 500 μL Krebs-HEPES buffer containing 14C-arginine (2 μL of 50 μCi/mL), FAD (10 μmol/L), FMN (10 μmol/L), and NADPH (1 mmol/L). Samples were run on a SCX 300 cation-exchange HPLC column (Sigma) with online scintillation detection. Background signals were corrected from samples with 14C-arginine alone without homogenates. The integrated 14C-citrulline elution peak was expressed as a percentage of total 14C counts.
Lucigenin-Enhanced Chemiluminometry
Total lung superoxide production was measured by lucigenin-enhanced chemiluminescence as previously described.24 Lungs were homogenized in 400 μL Krebs-HEPES buffer. Samples (100 μL) of homogenate corrected for total protein content were added to 2 mL Krebs-HEPES buffer containing lucigenin (5 μmol/L) in a scintillation vial at 37°C, and chemiluminescence was quantified over 20 minutes.
Dihydroethidium Fluorescent Microtopography
Endothelial superoxide production in tissue sections of pulmonary arteries was detected by use of the fluorescent probe dihydroethidium, adapting a method previously described for mouse aortas.24 Fresh pulmonary artery cryosections (30 μm) were incubated in Krebs-HEPES buffer containing acetylcholine (10 μmol/L) for 30 minutes at 37°C with or without the NOS inhibitor NG-nitro-L-arginine methyl ester (L-NAME; 1 mmol/L; Calbiochem) and dihydroethidium (2 μmol/L; Molecular Probes) for 5 more minutes at 37°C in darkness. Images were obtained with a laser confocal microscope (Bio-Rad MRC 1024) at identical acquisition settings. Endothelial superoxide production was quantified with Image-Pro Plus software (MediaCybernetics) as the sum of mean red intensities of red objects on the luminal surface of the pulmonary artery. For each ring, mean fluorescence was calculated from 4 separate fields.
Acute Hypoxia Studies
Lungs of anesthetized mice were ventilated with air at a constant end-expiratory pressure (15 cm H20) and perfused in situ in the open chest at a flow rate of 2 mL/min as described previously.25 Lung preparations were preconstricted with angiotensin II and allowed to equilibrate for 15 minutes. The increase in pulmonary artery pressure (hypoxic pulmonary vasoconstriction, or HPV) was then recorded after hypoxic challenges (2%O2/5% CO2/93%N2) of 10 minutes in the absence or presence of L-NAME (1 mmol/L).
Statistical Analysis
Data are expressed as mean±SEM (n=number of animals). Data were analyzed by use of 1-way ANOVA with post-hoc Fisher test. Significance was set at P<0.05.
Results
BH4 Deficiency Causes Pulmonary Hypertension
We first assessed the possible importance of BH4 in pulmonary vascular homeostasis using the hph-1 mouse model of BH4 deficiency. In these animals, tissue BH4 levels in homozygotes are significantly reduced because of constitutively reduced expression of GTPCH, the rate-limiting enzyme in BH4 biosynthesis.21–23 Using breeding pairs of animals heterozygous for the hph-1 allele, we obtained WT, hph-1+/–, and hph littermates and observed a striking graded reduction in BH4 levels in lung tissue, with lung BH4 levels in hph animals reduced to only 15% of WT levels (Figure 1). RVSP in hph mice bred under normoxic conditions was significantly elevated; hph-1+/– mice showed an intermediate elevation of RVSP compared with WT mice (Figure 2a). In keeping with the elevated RVSP, we found significant RV hypertrophy in hph mice (Figure 2b), whereas there were no differences in left ventricular mass or total body weight between WT, hph-1+/–, and hph mice (Table).
Comparison of Body Weights, Left Ventricular Weights, and Hematocrits
We next evaluated the response to chronic hypoxia by exposure of mice to 10% oxygen for 1 week, which resulted in uniform elevation of the hematocrit in all animals (Table). Lung BH4 levels and the degree of BH4 deficiency in hph mice remained unchanged by hypoxia (Figure 1). As expected, WT mice developed pulmonary hypertension in response to hypoxia, evidenced by both elevated RVSP and RV hypertrophy (Figure 2). However, these responses to hypoxia were greatly exacerbated in hph mice (Figure 2), with an intermediate phenotype in hph-1+/– littermates.
Endothelial BH4 Restoration Prevents Pulmonary Hypertension in hph-1 Mice
To determine whether selective restoration of BH4 biosynthesis in the endothelium would be sufficient to rescue the adverse effects of systemic BH4 deficiency on the pulmonary vasculature, we crossed hph mice with GCH-transgenic mice (GCH mice) in which endothelium-targeted GTPCH overexpression selectively increases BH4 levels in the vascular endothelium 2- to 3-fold.24 Lung BH4 levels in hph-1/GCH-transgenic (hph/GCH) mice were similar to WT in both normoxia and hypoxia, reflecting the significant contribution of endothelial cells to total lung BH4 on a background of systemic BH4 deficiency (Figure 1). Remarkably, the pulmonary hypertension and RV hypertrophy observed in hph mice under normoxic conditions were completely abolished in hph/GCH mice (Figure 2). Moreover, the further increases in both RVSP and RV hypertrophy in hph mice exposed to chronic hypoxia were strikingly reduced in hph/GCH mice (Figure 2). Thus, targeted transgenic restoration of endothelial BH4 synthesis is sufficient to rescue the effects of systemic BH4 deficiency in the hph mouse.
Endothelial BH4 Augmentation Attenuates Hypoxia-Induced Pulmonary Hypertension
We next investigated the effects of increasing endothelial BH4 above WT levels on the development of hypoxia-induced pulmonary hypertension by subjecting GCH mice to chronic hypoxia. Lung BH4 levels in GCH mice were doubled compared with WT mice, and this level of BH4 augmentation was maintained in chronic hypoxia (Figure 1). In normoxia, RVSP and RV mass were no different in GCH mice compared with WT (Figure 2). However, endothelial BH4 augmentation resulted in striking protection against both hypoxia-induced pulmonary hypertension and RV hypertrophy in GCH mice (Figure 2).
Endothelial BH4 Controls Pulmonary Vascular Remodeling
Pulmonary vascular remodeling is a key pathological hallmark of pulmonary hypertension and an important pathogenic step in disease progression.1 Quantitative analysis of distal muscularization in histological lung sections revealed striking evidence of vascular remodeling in both hph-1+/– and hph lungs even in normoxic conditions, which was exacerbated by exposure to hypoxia. Endothelial BH4 restoration in hph/GCH lungs significantly reduced vascular remodeling in normoxia and abrogated the increased remodeling observed in hypoxic hph-1+/– and hph mice compared with WT mice (Figure 3). More strikingly, GCH mice were completely protected against the development of hypoxia-induced pulmonary vascular remodeling compared with WT mice (Figure 3).
BH4 Determines Severity of Pulmonary Hypertension in a Dose-Dependent Manner
The overall importance of BH4 availability in determining the development and severity of pulmonary hypertension was further supported by striking inverse correlations in individual animals from all 5 different genotypes (WT, hph-1+/–, hph, hph/GCH, and GCH) between lung BH4 levels and RVSP, RV hypertrophy, and pulmonary vascular remodeling (Figure 4). These relationships were highly significant in both normoxia and hypoxia. The divergence between normoxic and hypoxic phenotypes increased as pulmonary BH4 levels decreased, whereas the phenotypic responses to hypoxia diminished progressively with increasing BH4 levels.
NOS Function Is Regulated by BH4 Bioavailability
Because BH4 is a required cofactor for eNOS activity and is hypothesized to regulate eNOS activity and enzymatic coupling, we quantified eNOS protein levels and activity in lung tissue. Total eNOS protein levels in lung tissue were not significantly different between WT, hph-1+/–, hph, hph/GCH, and GCH mice (Figure 5a and 5b). As described previously,11 hypoxia resulted in a moderate increase in eNOS protein levels (1.4-fold) that was similar in all groups (data not shown). Thus, the differences in BH4 levels between the respective mouse lines were not accompanied by any evidence of changes in eNOS protein levels. In contrast, eNOS enzymatic activity, assessed by radiolabeled L-arginine to L-citrulline conversion in lung extracts, was significantly reduced in hph-1+/– and hph mice compared with WT mice but was restored to WT levels in hph/GCH mice and doubled in GCH mice (Figure 5c). Thus, changes in lung BH4 levels were quantitatively associated with a parallel reduction, restoration, or augmentation of lung eNOS activity, independently of eNOS protein levels.
The effect of BH4 deficiency on oxidative stress was evaluated by measurement of superoxide production in lung homogenates by lucigenin-enhanced chemiluminescence. Superoxide production in lungs from hph mice was significantly elevated but was restored to WT levels by targeted endothelial BH4 replacement in hph/GCH mice (Figure 5d). To specifically assess the effect of BH4 levels on eNOS coupling, endothelial superoxide production was measured by dihydroethidium fluorescent microtopography in pulmonary artery rings incubated in the absence or presence of the NOS inhibitor L-NAME. In BH4-deficient hph-1+/– and hph mice, endothelial superoxide production was increased compared with WT mice (Figure 6). Incubation with L-NAME increased superoxide generation in WT rings, suggesting a tonic scavenging effect of NO on superoxide. In contrast, L-NAME incubation decreased endothelial superoxide formation in hph-1+/– and hph mice, implying a significant contribution from eNOS-derived superoxide (Figure 6). Furthermore, this increased endothelial superoxide production and response to NOS inhibition were reversed by endothelial restoration of BH4 in hph/GCH mice and GCH mice, both of which showed responses similar to WT mice (Figure 6).
NOS Regulation by BH4 Modulates Acute Hypoxic Pulmonary Vasoconstriction
The acute constrictor response of the pulmonary circulation to hypoxia (HPV) serves to optimize pulmonary gas exchange under physiological conditions but contributes to elevated pulmonary pressures in pathological states characterized by chronic hypoxia. Both NO and superoxide can modulate HPV.27,28 To evaluate the functional importance of NO compared with superoxide production by eNOS in relation to endothelial BH4 availability, we quantified HPV in perfused lung preparations from WT, hph-1+/–, hph, hph/GCH, and GCH mice. In WT lungs, inhibition of NO production with L-NAME increased HPV, reflecting the vasodilatory effect of eNOS-derived NO (Figure 7).27 Baseline HPV in hph lungs was significantly greater than in WT lungs. Furthermore, the response to NOS inhibition by L-NAME in hph lungs was qualitatively different from WT lungs, causing a reduction in HPV to levels identical to those observed in WT mice. In hph-1+/– mice, the HPV responses were intermediate between WT and hph mice. Strikingly, the magnitude of HPV and the effect of L-NAME were both restored to WT levels in hph/GCH mice, whereas HPV in GCH mice was further attenuated to levels below those observed in WT animals (Figure 7). These observations suggest that BH4-dependent eNOS regulation has a direct functional effect on acute HPV that is mediated through NOS coupling.
Discussion
We have identified a novel and important role for BH4 in the pulmonary circulation. Our observations, based on a number of complementary murine genetic models, provide clear evidence that endothelial BH4 availability in the pulmonary circulation regulates the pathophysiological response to hypoxia. Specifically, endothelial BH4 levels mediate susceptibility to or protection from pulmonary hypertension. We first demonstrate that BH4 deficiency results in the development of pulmonary hypertension and vascular remodeling under normoxic conditions and greatly exacerbates the response to hypoxia. Selective restoration of endothelial BH4 levels rescues the effects of systemic BH4 deficiency, whereas augmentation of endothelial BH4 biosynthesis protects against the development of hypoxia-induced pulmonary hypertension and vascular remodeling. The striking quantitative correlations between lung BH4 levels and the development of pulmonary hypertension across the 5 genetic mouse models evaluated in our studies reveal that BH4 bioavailability controls both pulmonary vascular tone and structural remodeling in a dose-dependent manner. Finally, we show that the effects of genetic alterations in BH4 availability on the pulmonary vasculature are mediated through the regulation of NO compared with superoxide synthesis by eNOS in response to both acute and chronic hypoxia.
The elevated RVSP in the BH4-deficient hph-1 mice is at least comparable to that previously observed in mice deficient in eNOS (eNOS–/–), with both strains showing exacerbated responses to chronic hypoxia.6,7,29 However, vascular remodeling, a hallmark of pulmonary hypertension, was not seen in lungs of eNOS–/– mice in normoxia at sea-level altitudes,25,29 whereas vascular remodeling was clearly evident in both hph-1 heterozygotes and homozygotes compared with WT littermates, even in normoxia. This discordance in the severity of the phenotype between genetic eNOS deficiency compared with eNOS dysfunction resulting from genetic BH4 deficiency suggests that loss of NO production alone is not the sole mediator of disease pathogenesis. Rather, our findings highlight the importance of increased eNOS-dependent superoxide production in playing a further pathogenic role in both the vasoconstrictor and remodeling processes. Indeed, vascular superoxide production has a number of potentially important effects in the vascular wall, including effects on NO signaling through scavenging and peroxynitrite generation and through modulation of redox-sensitive signaling pathways.13 Superoxide controls proliferation and apoptosis of pulmonary smooth muscle cells,17 and pulmonary hypertension in fetal lambs subjected to ligation of ductus arteriosus is associated with increased superoxide generation.14 Our findings highlight the importance of endothelial BH4 availability as a reciprocal modulator of both NO and superoxide production by eNOS and suggest a critical role for BH4 availability in the pathogenesis of pulmonary hypertension. Thus, BH4 provides a mechanistic link between previous observations of increased NOS protein levels and reduced NO bioactivity.11,12 Targeting NOS regulation and enzymatic coupling, rather than eNOS protein levels or total enzymatic activity, may be a more promising therapeutic strategy in pulmonary hypertension.
The pulmonary phenotype in the eNOS–/– mice may be further complicated by the fact that these mice have left ventricular hypertrophy secondary to systemic hypertension,30 which is not seen in the BH4-deficient hph-1 mice. Thus, the effect of systemic BH4 deficiency, at least in hph-1 mice, appears to be more critical or specific to the pulmonary circulation as opposed to the systemic circulation. We note that this discordance was also seen in the transgenic mouse expressing a smooth muscle–specific dominant-negative BMPR-II, which developed pulmonary but not systemic hypertension, even though the SM22 smooth muscle–specific promoter directs systemic and pulmonary expression.31
The interplay between the various putative pathways implicated in pulmonary hypertension at both the cellular level (eg, endothelium, smooth muscle cells, platelets, macrophages, inflammatory cells, and progenitor cells) and the molecular level (eg, BMPR-II, endothelin, serine elastase, serotonin, potassium channels, angiopoietin, NO, and superoxide) is complex and poorly defined.1 It is unlikely that any single environmental factor or gene will explain all forms of pulmonary hypertension. Nevertheless, our findings underscore the central importance of endothelial dysfunction in pulmonary hypertension.4 The relative bioactivities of NO and reactive oxygen species are critical to maintaining pulmonary vascular homeostasis, and the balance may be shifted toward injury in pulmonary hypertension.5 Our studies show that BH4 has a pivotal effect on the pathogenesis of pulmonary hypertension mediated through eNOS enzymatic coupling that maintains the homeostatic balance between NO and superoxide bioactivity in the endothelium. The striking "concentration-response" relationship between lung BH4 levels and both functional and structural features of pulmonary hypertension suggests that alterations in BH4 availability modify susceptibility to or protection from pulmonary hypertension in both normoxia and hypoxia. Indeed, we observed a pulmonary hypertensive phenotype even in hph-1 heterozygotes in which the magnitude of BH4 deficiency is modest compared with hph-1 homozygotes yet is sufficient to alter both pulmonary tone and structure. These observations compare similarly with eNOS+/– (heterozygous) mice, which, like eNOS–/– mice, have raised RVSP and hypersensitivity to hypoxia.7 Like the eNOS gene, our data suggest that loss of 1 allele of the GCH1 gene encoding GTPCH, the rate-limiting enzyme in BH4 synthesis, is sufficient to produce a pulmonary vascular phenotype. Accordingly, we speculate that GCH1 is a candidate modifier gene in pulmonary hypertension. Investigating possible associations between genetic variation in the GCH1 gene and pulmonary hypertension may prove informative. Human GCH1 mutations are rare but cause severe BH4 deficiency syndromes such as hyperphenylalaninemia and dopa-responsive dystonia, because BH4 is also a cofactor for enzymes involved in hepatic phenylalanine metabolism and neuronal dopamine synthesis.26 The effect of BH4 deficiency on the pulmonary circulation in these patients remains to be studied.
We note that BH4 deficiency in hph-1 mice did not lead to severe pulmonary hypertension with plexiform lesions as seen in humans,1 even after exposure to hypoxia, and the mice do not suffer premature death. This may not be surprising, given that other recently published genetic mouse models of pulmonary hypertension do not recapitulate this extreme phenotype.31,32,33 Even in mice overexpressing S100A4/Mts1, only 5% of aging mice develop plexiform arteriopathy.34 These observations imply that additional pathways need to be invoked for full pathogenic expression of the human disease.
A further limitation of this study is that we had not determined late changes to longer durations of hypoxia. Endothelial dysfunction may play a larger role in the initiation and early pathogenesis of pulmonary hypertension,4 and it is possible that endothelial BH4 augmentation alone may not fully protect against greater structural changes accompanying prolonged hypoxia, when additional molecular pathways may predominate.
Given the complex pathophysiology of pulmonary hypertension, targeting of multiple pathways with combination drug therapy may be necessary. The ability of BH4 to both augment NO synthesis and decrease superoxide production addresses 2 pathogenic mechanisms simultaneously and may be a key therapeutic target, as illustrated by the striking salutary effects of endothelial BH4 restoration or augmentation in the hph/GCH and GCH mice. Pharmacological or molecular strategies that target endothelial BH4 availability in patients with pulmonary hypertension need to be evaluated. It is also interesting to note that HMG-CoA inhibitors (statins), currently generating significant interest after their reversal of experimental pulmonary hypertension in rats,35 upregulate GTPCH mRNA and BH4 levels in vascular endothelial cells.36
In conclusion, we have used a range of complementary gene-modified murine models to provide compelling evidence that demonstrates a novel and important role for endothelial BH4 in the pulmonary circulation. Regulation of eNOS function by BH4, a critical mechanism in the pathogenesis of pulmonary hypertension, may offer new therapeutic opportunities in a disease that remains inadequately understood and treated.
Acknowledgments
This work was supported by the British Heart Foundation (RG/02/006 to Dr Channon; PG/055/17036 to Dr Zhao/Dr Wilkins). Dr Khoo is a Wellcome Trust Cardiovascular Research Initiative Fellow.
Footnotes
Drs Khoo and Zhao contributed equally to this study.
References
Humbert M, Morrell NW, Archer SL, Stenmark KR, MacLean MR, Lang IM, Christman BW, Weir EK, Eickelberg O, Voelkel NF, Rabinovitch M. Cellular and molecular pathobiology of pulmonary arterial hypertension. J Am Coll Cardiol. 2004; 43: 13S–24S.
Simonneau G, Galie N, Rubin LJ, Langleben D, Seeger W, Domenighetti G, Gibbs S, Lebrec D, Speich R, Beghetti M, Rich S, Fishman A. Clinical classification of pulmonary hypertension. J Am Coll Cardiol. 2004; 43: 5S–12S.
Newman JH, Trembath RC, Morse JA, Grunig E, Loyd JE, Adnot S, Coccolo F, Ventura C, Phillips JA 3rd, Knowles JA, Janssen B, Eickelberg O, Eddahibi S, Herve P, Nichols WC, Elliott G. Genetic basis of pulmonary arterial hypertension: current understanding and future directions. J Am Coll Cardiol. 2004; 43: 33S–39S.
Budhiraja R, Tuder RM, Hassoun PM. Endothelial dysfunction in pulmonary hypertension. Circulation. 2004; 109: 159–165.
Hampl V, Herget J. Role of nitric oxide in the pathogenesis of chronic pulmonary hypertension. Physiol Rev. 2000; 80: 1337–1372.
Steudel W, Scherrer-Crosbie M, Bloch KD, Weimann J, Huang PL, Jones RC, Picard MH, Zapol WM. Sustained pulmonary hypertension and right ventricular hypertrophy after chronic hypoxia in mice with congenital deficiency of nitric oxide synthase 3. J Clin Invest. 1998; 101: 2468–2477.
Fagan KA, Fouty BW, Tyler RC, Morris KG Jr, Hepler LK, Sato K, LeCras TD, Abman SH, Weinberger HD, Huang PL, McMurtry IF, Rodman DM. The pulmonary circulation of homozygous or heterozygous eNOS-null mice is hyperresponsive to mild hypoxia. J Clin Invest. 1999; 103: 291–299.
Champion HC, Bivalacqua TJ, Greenberg SS, Giles TD, Hyman AL, Kadowitz PJ. Adenoviral gene transfer of endothelial nitric-oxide synthase (eNOS) partially restores normal pulmonary arterial pressure in eNOS-deficient mice. Proc Natl Acad Sci U S A. 2002; 99: 13248–13253.
Kaneko FT, Arroliga AC, Dweik RA, Comhair SA, Laskowski D, Oppedisano R, Thomassen MJ, Erzurum SC. Biochemical reaction products of nitric oxide as quantitative markers of primary pulmonary hypertension. Am J Respir Crit Care Med. 1998; 158: 917–923.
Machado RF, Londhe Nerkar MV, Dweik RA, Hammel J, Janocha A, Pyle J, Laskowski D, Jennings C, Arroliga AC, Erzurum SC. Nitric oxide and pulmonary arterial pressures in pulmonary hypertension. Free Radic Biol Med. 2004; 37: 1010–1017.
Fagan KA, Morrissey B, Fouty BW, Sato K, Harral JW, Morris KG Jr, Hoedt-Miller M, Vidmar S, McMurtry IF, Rodman DM. Upregulation of nitric oxide synthase in mice with severe hypoxia-induced pulmonary hypertension. Respir Res. 2001; 2: 306–313.
Le Cras TD, McMurtry IF. Nitric oxide production in the hypoxic lung. Am J Physiol Lung Cell Mol Physiol. 2001; 280: L575–L582.
Cai H, Harrison DG. Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ Res. 2000; 87: 840–844.
Brennan LA, Steinhorn RH, Wedgwood S, Mata-Greenwood E, Roark EA, Russell JA, Black SM. Increased superoxide generation is associated with pulmonary hypertension in fetal lambs: a role for NADPH oxidase. Circ Res. 2003; 92: 683–691.
Bowers R, Cool C, Murphy RC, Tuder RM, Hopken MW, Flores SC, Voelkel NF. Oxidative stress in severe pulmonary hypertension. Am J Respir Crit Care Med. 2004; 169: 764–769.
Liu JQ, Folz RJ. Extracellular superoxide enhances 5-HT–induced murine pulmonary artery vasoconstriction. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L111–L118.
Wedgwood S, Black SM. Role of reactive oxygen species in vascular remodeling associated with pulmonary hypertension. Antioxid Redox Signal. 2003; 5: 759–769.
Katusic ZS. Vascular endothelial dysfunction: does tetrahydrobiopterin play a role; Am J Physiol Heart Circ Physiol. 2001; 281: H981–H986.
Alp NJ, Channon KM. Regulation of endothelial nitric oxide synthase by tetrahydrobiopterin in vascular disease. Arterioscler Thromb Vasc Biol. 2004; 24: 413–420.
Landmesser U, Dikalov S, Price SR, McCann L, Fukai T, Holland SM, Mitch WE, Harrison DG. Oxidation of tetrahydrobiopterin leads to uncoupling of endothelial cell nitric oxide synthase in hypertension. J Clin Invest. 2003; 111: 1201–1209.
McDonald JD, Cotton RG, Jennings I, Ledley FD, Woo SL, Bode VC. Biochemical defect of the hph-1 mouse mutant is a deficiency in GTP-cyclohydrolase activity. J Neurochem. 1988; 50: 655–657.
Cosentino F, Barker JE, Brand MP, Heales SJ, Werner ER, Tippins JR, West N, Channon KM, Volpe M, Luscher TF. Reactive oxygen species mediate endothelium-dependent relaxations in tetrahydrobiopterin-deficient mice. Arterioscler Thromb Vasc Biol. 2001; 21: 496–502.
Khoo JP, Nicoli T, Alp NJ, Fullerton J, Flint J, Channon KM. Congenic mapping and genotyping of the tetrahydrobiopterin-deficient hph-1 mouse. Mol Genet Metab. 2004; 82: 251–254.
Alp NJ, Mussa S, Khoo J, Guzik TJ, Cai S, Jefferson A, Rockett KA, Channon KM. Tetrahydrobiopterin-dependent preservation of nitric oxide–mediated endothelial function in diabetes by targeted transgenic GTP-cyclohydrolase I overexpression. J Clin Invest. 2003; 112: 725–735.
Zhao L, Mason NA, Morrell NW, Kojonazarov B, Sadykov A, Maripov A, Mirrakhimov MM, Aldashev A, Wilkins MR. Sildenafil inhibits hypoxia-induced pulmonary hypertension. Circulation. 2001; 104: 424–428.
Thony B, Auerbach G, Blau N. Tetrahydrobiopterin biosynthesis, regeneration and functions. Biochem J. 2000; 347 (pt 1): 1–16.
Fagan KA, Tyler RC, Sato K, Fouty BW, Morris KG Jr, Huang PL, McMurtry IF, Rodman DM. Relative contributions of endothelial, inducible, and neuronal NOS to tone in the murine pulmonary circulation. Am J Physiol. 1999; 277: L472–L478.
Weissmann N, Winterhalder S, Nollen M, Voswinckel R, Quanz K, Ghofrani HA, Schermuly RT, Seeger W, Grimminger F. NO and reactive oxygen species are involved in biphasic hypoxic vasoconstriction of isolated rabbit lungs. Am J Physiol Lung Cell Mol Physiol. 2001; 280: L638–L645.
Steudel W, Ichinose F, Huang PL, Hurford WE, Jones RC, Bevan JA, Fishman MC, Zapol WM. Pulmonary vasoconstriction and hypertension in mice with targeted disruption of the endothelial nitric oxide synthase (NOS 3) gene. Circ Res. 1997; 81: 34–41.
Huang PL, Huang ZH, Mashimo H, Bloch KD, Moskowitz MA, Bevan JA, Fishman MC. Hypertension in mice lacking the gene for endothelial nitric oxide synthase. Nature. 1995; 377: 239–242.
West J, Fagan K, Steudel W, Fouty B, Lane K, Harral J, Hoedt-Miller M, Tada Y, Ozimek J, Tuder R, Rodman DM. Pulmonary hypertension in transgenic mice expressing a dominant-negative BMPRII gene in smooth muscle. Circ Res. 2004; 94: 1109–1114.
Beppu H, Ichinose F, Kawai N, Jones RC, Yu PB, Zapol WM, Miyazono K, Li E, Bloch KD. BMPR-II heterozygous mice have mild pulmonary hypertension and an impaired pulmonary vascular remodeling response to prolonged hypoxia. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L1241–1247.
MacLean MR, Deuchar GA, Hicks MN, Morecroft I, Shen S, Sheward J, Colston J, Loughlin L, Nilsen M, Dempsie Y, Harmar A. Overexpression of the 5-hydroxytryptamine transporter gene: effect on pulmonary hemodynamics and hypoxia-induced pulmonary hypertension. Circulation. 2004; 109: 2150–2155.
Greenway S, van Suylen RJ, Du Marchie Sarvaas G, Kwan E, Ambartsumian N, Lukanidin E, Rabinovitch M. S100A4/Mts1 produces murine pulmonary artery changes resembling plexogenic arteriopathy and is increased in human plexogenic arteriopathy. Am J Pathol. 2004; 164: 253–262.
Girgis RE, Li D, Zhan X, Garcia JG, Tuder RM, Hassoun PM, Johns RA. Attenuation of chronic hypoxic pulmonary hypertension by simvastatin. Am J Physiol Heart Circ Physiol. 2003; 285: H938–H945.
Hattori Y, Nakanishi N, Akimoto K, Yoshida M, Kasai K. HMG-CoA reductase inhibitor increases GTP cyclohydrolase I mRNA and tetrahydrobiopterin in vascular endothelial cells. Arterioscler Thromb Vasc Biol. 2003; 23: 176–182.(Jeffrey P. Khoo, MD; Lan )
Department of Experimental Medicine and Toxicology, Imperial College London, Hammersmith Hospital, London, UK (L.Z., M.R.W.)
Childhood Infection Group, Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, UK (K.R.).
Abstract
Background— Pulmonary hypertension is a fatal disease characterized by vasoconstriction and vascular remodeling. Loss of endothelial nitric oxide bioavailability is implicated in pulmonary hypertension pathogenesis. Recent evidence suggests that the cofactor tetrahydrobiopterin (BH4) is an important regulator of nitric oxide synthase enzymatic function.
Methods and Results— In the hph-1 mouse with deficient BH4 biosynthesis, BH4 deficiency caused pulmonary hypertension, even in normoxic conditions, and greatly increased susceptibility to hypoxia-induced pulmonary hypertension. In contrast, augmented BH4 synthesis in the endothelium, by targeted transgenic overexpression of GTP-cyclohydrolase I (GCH), prevented hypoxia-induced pulmonary hypertension. Furthermore, specific augmentation of endothelial BH4 in hph-1 mice by crossing with GCH transgenic mice rescued pulmonary hypertension induced by systemic BH4 deficiency. Lung BH4 availability controlled pulmonary vascular tone, right ventricular hypertrophy, and vascular structural remodeling in a dose-dependent manner in both normoxia and hypoxia. Furthermore, BH4 availability had striking effects on the immediate vasoconstriction response to acute hypoxia. These effects of BH4 were mediated through the regulation of nitric oxide compared with superoxide synthesis by endothelial nitric oxide synthase.
Conclusions— Endothelial BH4 availability is essential for maintaining pulmonary vascular homeostasis, is a critical mediator in the pathogenesis of pulmonary hypertension, and is a novel therapeutic target.
Key Words: hypertension, pulmonary ; tetrahydrobiopterin ; superoxide ; endothelium ; nitric oxide synthase
Introduction
In a normal oxygen environment, the healthy adult pulmonary vascular bed is a low-pressure circuit. Pulmonary hypertension arises when vasoconstriction and structural remodeling of pulmonary arterioles lead to increased pulmonary vascular resistance. The consequent increase in pressure load causes right ventricular (RV) hypertrophy, which progresses to premature death from right-sided heart failure.1 Pulmonary hypertension can develop in isolation (idiopathic pulmonary arterial hypertension) or in association with conditions such as connective tissue diseases, congenital heart disease, pulmonary thromboembolism, or chronic hypoxia.2 Mutations in the BMPR2 gene, encoding a receptor for bone morphogenetic protein, appear to underlie some cases of familial idiopathic pulmonary arterial hypertension. However, BMPR2 mutations account for only 10% of sporadic idiopathic pulmonary arterial hypertension, and only 15% to 20% of individuals carrying a BMPR2 mutation develop pulmonary hypertension, suggesting that additional disease-causing or disease-modifying genes remain critical to understanding pulmonary hypertension pathogenesis.3
See p 2022
The endothelium plays a key role in pulmonary vascular homeostasis. Endothelial dysfunction is clearly implicated in the pathogenesis of pulmonary hypertension, but the exact mechanisms remain poorly understood.4 Endothelium-derived nitric oxide (NO), through potent vasodilator and antiproliferative effects on vascular smooth muscle cells, is critical in maintaining normal pulmonary vascular tone and structure.5 Mice deficient in endothelial NO synthase (eNOS) are more sensitive to hypoxia-induced pulmonary hypertension,6,7 whereas pulmonary gene transfer of eNOS is partially protective.8 Patients with pulmonary hypertension have low NO levels in their exhaled breath,9,10 and both inhaled NO and phosphodiesterase type 5 inhibitors, which act to increase NO-mediated cGMP signaling, have emerged as therapeutic strategies in pulmonary hypertension. Paradoxically, mice with hypoxia-induced pulmonary hypertension have increased eNOS protein levels without a concomitant increase in NO bioactivity.11,12 Scavenging by superoxide radicals may account in part for reduced NO bioactivity.13 Increased superoxide production has been observed in experimental models of pulmonary hypertension,14 and markers of oxidative stress are increased in patients with pulmonary hypertension.15 Superoxide mediates pulmonary vasoconstriction16 and stimulates pulmonary smooth muscle cell proliferation.17 These observations suggest that perturbation of NO synthesis, together with increased superoxide production, may play a central role in the pathogenesis of pulmonary hypertension.
The cofactor tetrahydrobiopterin (BH4) is an important regulator of NOS enzymatic activity.18 Without BH4, NO production is reduced and superoxide production is increased as a result of the loss of enzymatic coupling between the reduction in molecular oxygen and oxidation of L-arginine.19 In the systemic vasculature, BH4-mediated eNOS regulation appears to be an important determinant of NO bioavailability, independent of total eNOS protein levels,19,20 but the role of BH4 in the pulmonary circulation remains unexplored. We hypothesized that endothelial BH4 availability is a critical determinant of both the acute and chronic responses to hypoxia and the development of pulmonary hypertension, mediated through the effect of BH4 on eNOS activity and enzymatic coupling. To test this hypothesis, we compared pulmonary hemodynamics and the development of hypoxic pulmonary hypertension in genetic mouse models with graded reductions in BH4 levels; then, we tested the ability of transgenic endothelium-targeted augmentation of BH4 to rescue the effects of BH4 deficiency and protect against hypoxia-induced pulmonary hypertension. Furthermore, we investigated the effect of these changes in pulmonary BH4 availability on local NO compared with superoxide production by eNOS and the functional consequences of BH4-mediated eNOS regulation on acute pulmonary vasomotor function.
Methods
Mice
The hph-1 mouse, generated by ENU mutagenesis, was used as a model of BH4 deficiency. In these animals, tissue BH4 levels are low because of constitutively reduced expression of GTP cyclohydrolase I (GTPCH), the rate-limiting enzyme in BH4 biosynthesis.21–23 Wild-type (WT), hph-1 heterozygous (+/–), and hph-1 homozygous (hph) littermates on a C57BL/6 background were obtained by interbreeding hph-1 heterozygotes.23 In GCH-transgenic mice (GCH mice), overexpression of the human GTPCH under the control of the endothelial-specific Tie-2 promoter results in 2- to 3-fold elevation of endothelial BH4 levels.24 GCH mice used in these studies were bred for 8 generations onto a C57BL/6 background. To generate hph-1/GCH transgenic mice (hph/GCH), ie, hph-1 homozygotes carrying the Tie2-GCH transgene, hph-1 homozygotes were crossed with GCH mice, and offspring were backcrossed with hph-1. Mice were genotyped as previously described for hph-1 and GCH alleles.23,24 Mice were used at 15 weeks of age. Studies were conducted in accordance with the UK Home Office Animals (Scientific Procedures) Act 1986.
Normoxia and Chronic Hypoxia Studies
Animals were either housed in normal air or placed in a specially constructed normobaric hypoxic chamber (FIO2 10%) for 1 week.25 At 1 week, animals were weighed and anesthetized (Hypnorm 0.25 mL/kg; Midazolam 25 mg/kg IP), and RV systolic pressure (RVSP) was measured via direct cardiac puncture using a closed-chest technique in the spontaneously breathing, anesthetized animal.25 The animals were then euthanized, the hearts were removed, and the individual ventricular chambers were weighed (septum with left ventricle). The right lungs were snap-frozen in liquid nitrogen and stored at –80°C for biochemical measurements. The left lungs were fixed by inflation with 10% formalin, embedded in paraffin, and sectioned for histology.
BH4 Assay
Lung homogenates were oxidized with 1% iodine/2% KI in acid or base. Biopterins were determined by high-performance liquid chromatography (HPLC) with fluorescent detection, and BH4 was calculated by subtracting BH2+biopterin from total biopterins, as previously described.24
Lung Histology
Histological assessment of vascular remodeling was performed as previously described.25 Transverse lung sections were stained with van Gieson’s elastic method or smooth muscle -actin antibody (Clone 1A, Sigma) using an alkaline phosphate–conjugated avidin-biotin complex method and Vector Red substrate (Vector Laboratories), with methyl green counterstain (Vector Laboratories). "Distal muscularization" was defined as the proportion of vessels (<50 μm diameter) with immunoreactivity for smooth muscle -actin (as evidence for muscularization) over the total number of vessels stained with elastin. Three separate sections from each animal were quantified, and counting was performed by investigators blinded to genotype or treatment group.
eNOS Western Blotting
Lung homogenates were subjected to SDS-PAGE and transferred to nitrocellulose membranes. After transfer, membranes were incubated with mouse eNOS monoclonal antibodies (Transduction Labs), followed by horseradish peroxidase–conjugated secondary antibody. Proteins were visualized by chemiluminescence (SuperSignal, Pierce). Blot intensity was quantified with Scion Image software (release beta 4.0.2).
NOS Activity Assay
NOS activity was determined by 14C-arginine to 14C-citrulline conversion with HPLC quantification as described.24 Lung homogenates (400 μg protein) were incubated for 30 minutes at 37°C in 500 μL Krebs-HEPES buffer containing 14C-arginine (2 μL of 50 μCi/mL), FAD (10 μmol/L), FMN (10 μmol/L), and NADPH (1 mmol/L). Samples were run on a SCX 300 cation-exchange HPLC column (Sigma) with online scintillation detection. Background signals were corrected from samples with 14C-arginine alone without homogenates. The integrated 14C-citrulline elution peak was expressed as a percentage of total 14C counts.
Lucigenin-Enhanced Chemiluminometry
Total lung superoxide production was measured by lucigenin-enhanced chemiluminescence as previously described.24 Lungs were homogenized in 400 μL Krebs-HEPES buffer. Samples (100 μL) of homogenate corrected for total protein content were added to 2 mL Krebs-HEPES buffer containing lucigenin (5 μmol/L) in a scintillation vial at 37°C, and chemiluminescence was quantified over 20 minutes.
Dihydroethidium Fluorescent Microtopography
Endothelial superoxide production in tissue sections of pulmonary arteries was detected by use of the fluorescent probe dihydroethidium, adapting a method previously described for mouse aortas.24 Fresh pulmonary artery cryosections (30 μm) were incubated in Krebs-HEPES buffer containing acetylcholine (10 μmol/L) for 30 minutes at 37°C with or without the NOS inhibitor NG-nitro-L-arginine methyl ester (L-NAME; 1 mmol/L; Calbiochem) and dihydroethidium (2 μmol/L; Molecular Probes) for 5 more minutes at 37°C in darkness. Images were obtained with a laser confocal microscope (Bio-Rad MRC 1024) at identical acquisition settings. Endothelial superoxide production was quantified with Image-Pro Plus software (MediaCybernetics) as the sum of mean red intensities of red objects on the luminal surface of the pulmonary artery. For each ring, mean fluorescence was calculated from 4 separate fields.
Acute Hypoxia Studies
Lungs of anesthetized mice were ventilated with air at a constant end-expiratory pressure (15 cm H20) and perfused in situ in the open chest at a flow rate of 2 mL/min as described previously.25 Lung preparations were preconstricted with angiotensin II and allowed to equilibrate for 15 minutes. The increase in pulmonary artery pressure (hypoxic pulmonary vasoconstriction, or HPV) was then recorded after hypoxic challenges (2%O2/5% CO2/93%N2) of 10 minutes in the absence or presence of L-NAME (1 mmol/L).
Statistical Analysis
Data are expressed as mean±SEM (n=number of animals). Data were analyzed by use of 1-way ANOVA with post-hoc Fisher test. Significance was set at P<0.05.
Results
BH4 Deficiency Causes Pulmonary Hypertension
We first assessed the possible importance of BH4 in pulmonary vascular homeostasis using the hph-1 mouse model of BH4 deficiency. In these animals, tissue BH4 levels in homozygotes are significantly reduced because of constitutively reduced expression of GTPCH, the rate-limiting enzyme in BH4 biosynthesis.21–23 Using breeding pairs of animals heterozygous for the hph-1 allele, we obtained WT, hph-1+/–, and hph littermates and observed a striking graded reduction in BH4 levels in lung tissue, with lung BH4 levels in hph animals reduced to only 15% of WT levels (Figure 1). RVSP in hph mice bred under normoxic conditions was significantly elevated; hph-1+/– mice showed an intermediate elevation of RVSP compared with WT mice (Figure 2a). In keeping with the elevated RVSP, we found significant RV hypertrophy in hph mice (Figure 2b), whereas there were no differences in left ventricular mass or total body weight between WT, hph-1+/–, and hph mice (Table).
Comparison of Body Weights, Left Ventricular Weights, and Hematocrits
We next evaluated the response to chronic hypoxia by exposure of mice to 10% oxygen for 1 week, which resulted in uniform elevation of the hematocrit in all animals (Table). Lung BH4 levels and the degree of BH4 deficiency in hph mice remained unchanged by hypoxia (Figure 1). As expected, WT mice developed pulmonary hypertension in response to hypoxia, evidenced by both elevated RVSP and RV hypertrophy (Figure 2). However, these responses to hypoxia were greatly exacerbated in hph mice (Figure 2), with an intermediate phenotype in hph-1+/– littermates.
Endothelial BH4 Restoration Prevents Pulmonary Hypertension in hph-1 Mice
To determine whether selective restoration of BH4 biosynthesis in the endothelium would be sufficient to rescue the adverse effects of systemic BH4 deficiency on the pulmonary vasculature, we crossed hph mice with GCH-transgenic mice (GCH mice) in which endothelium-targeted GTPCH overexpression selectively increases BH4 levels in the vascular endothelium 2- to 3-fold.24 Lung BH4 levels in hph-1/GCH-transgenic (hph/GCH) mice were similar to WT in both normoxia and hypoxia, reflecting the significant contribution of endothelial cells to total lung BH4 on a background of systemic BH4 deficiency (Figure 1). Remarkably, the pulmonary hypertension and RV hypertrophy observed in hph mice under normoxic conditions were completely abolished in hph/GCH mice (Figure 2). Moreover, the further increases in both RVSP and RV hypertrophy in hph mice exposed to chronic hypoxia were strikingly reduced in hph/GCH mice (Figure 2). Thus, targeted transgenic restoration of endothelial BH4 synthesis is sufficient to rescue the effects of systemic BH4 deficiency in the hph mouse.
Endothelial BH4 Augmentation Attenuates Hypoxia-Induced Pulmonary Hypertension
We next investigated the effects of increasing endothelial BH4 above WT levels on the development of hypoxia-induced pulmonary hypertension by subjecting GCH mice to chronic hypoxia. Lung BH4 levels in GCH mice were doubled compared with WT mice, and this level of BH4 augmentation was maintained in chronic hypoxia (Figure 1). In normoxia, RVSP and RV mass were no different in GCH mice compared with WT (Figure 2). However, endothelial BH4 augmentation resulted in striking protection against both hypoxia-induced pulmonary hypertension and RV hypertrophy in GCH mice (Figure 2).
Endothelial BH4 Controls Pulmonary Vascular Remodeling
Pulmonary vascular remodeling is a key pathological hallmark of pulmonary hypertension and an important pathogenic step in disease progression.1 Quantitative analysis of distal muscularization in histological lung sections revealed striking evidence of vascular remodeling in both hph-1+/– and hph lungs even in normoxic conditions, which was exacerbated by exposure to hypoxia. Endothelial BH4 restoration in hph/GCH lungs significantly reduced vascular remodeling in normoxia and abrogated the increased remodeling observed in hypoxic hph-1+/– and hph mice compared with WT mice (Figure 3). More strikingly, GCH mice were completely protected against the development of hypoxia-induced pulmonary vascular remodeling compared with WT mice (Figure 3).
BH4 Determines Severity of Pulmonary Hypertension in a Dose-Dependent Manner
The overall importance of BH4 availability in determining the development and severity of pulmonary hypertension was further supported by striking inverse correlations in individual animals from all 5 different genotypes (WT, hph-1+/–, hph, hph/GCH, and GCH) between lung BH4 levels and RVSP, RV hypertrophy, and pulmonary vascular remodeling (Figure 4). These relationships were highly significant in both normoxia and hypoxia. The divergence between normoxic and hypoxic phenotypes increased as pulmonary BH4 levels decreased, whereas the phenotypic responses to hypoxia diminished progressively with increasing BH4 levels.
NOS Function Is Regulated by BH4 Bioavailability
Because BH4 is a required cofactor for eNOS activity and is hypothesized to regulate eNOS activity and enzymatic coupling, we quantified eNOS protein levels and activity in lung tissue. Total eNOS protein levels in lung tissue were not significantly different between WT, hph-1+/–, hph, hph/GCH, and GCH mice (Figure 5a and 5b). As described previously,11 hypoxia resulted in a moderate increase in eNOS protein levels (1.4-fold) that was similar in all groups (data not shown). Thus, the differences in BH4 levels between the respective mouse lines were not accompanied by any evidence of changes in eNOS protein levels. In contrast, eNOS enzymatic activity, assessed by radiolabeled L-arginine to L-citrulline conversion in lung extracts, was significantly reduced in hph-1+/– and hph mice compared with WT mice but was restored to WT levels in hph/GCH mice and doubled in GCH mice (Figure 5c). Thus, changes in lung BH4 levels were quantitatively associated with a parallel reduction, restoration, or augmentation of lung eNOS activity, independently of eNOS protein levels.
The effect of BH4 deficiency on oxidative stress was evaluated by measurement of superoxide production in lung homogenates by lucigenin-enhanced chemiluminescence. Superoxide production in lungs from hph mice was significantly elevated but was restored to WT levels by targeted endothelial BH4 replacement in hph/GCH mice (Figure 5d). To specifically assess the effect of BH4 levels on eNOS coupling, endothelial superoxide production was measured by dihydroethidium fluorescent microtopography in pulmonary artery rings incubated in the absence or presence of the NOS inhibitor L-NAME. In BH4-deficient hph-1+/– and hph mice, endothelial superoxide production was increased compared with WT mice (Figure 6). Incubation with L-NAME increased superoxide generation in WT rings, suggesting a tonic scavenging effect of NO on superoxide. In contrast, L-NAME incubation decreased endothelial superoxide formation in hph-1+/– and hph mice, implying a significant contribution from eNOS-derived superoxide (Figure 6). Furthermore, this increased endothelial superoxide production and response to NOS inhibition were reversed by endothelial restoration of BH4 in hph/GCH mice and GCH mice, both of which showed responses similar to WT mice (Figure 6).
NOS Regulation by BH4 Modulates Acute Hypoxic Pulmonary Vasoconstriction
The acute constrictor response of the pulmonary circulation to hypoxia (HPV) serves to optimize pulmonary gas exchange under physiological conditions but contributes to elevated pulmonary pressures in pathological states characterized by chronic hypoxia. Both NO and superoxide can modulate HPV.27,28 To evaluate the functional importance of NO compared with superoxide production by eNOS in relation to endothelial BH4 availability, we quantified HPV in perfused lung preparations from WT, hph-1+/–, hph, hph/GCH, and GCH mice. In WT lungs, inhibition of NO production with L-NAME increased HPV, reflecting the vasodilatory effect of eNOS-derived NO (Figure 7).27 Baseline HPV in hph lungs was significantly greater than in WT lungs. Furthermore, the response to NOS inhibition by L-NAME in hph lungs was qualitatively different from WT lungs, causing a reduction in HPV to levels identical to those observed in WT mice. In hph-1+/– mice, the HPV responses were intermediate between WT and hph mice. Strikingly, the magnitude of HPV and the effect of L-NAME were both restored to WT levels in hph/GCH mice, whereas HPV in GCH mice was further attenuated to levels below those observed in WT animals (Figure 7). These observations suggest that BH4-dependent eNOS regulation has a direct functional effect on acute HPV that is mediated through NOS coupling.
Discussion
We have identified a novel and important role for BH4 in the pulmonary circulation. Our observations, based on a number of complementary murine genetic models, provide clear evidence that endothelial BH4 availability in the pulmonary circulation regulates the pathophysiological response to hypoxia. Specifically, endothelial BH4 levels mediate susceptibility to or protection from pulmonary hypertension. We first demonstrate that BH4 deficiency results in the development of pulmonary hypertension and vascular remodeling under normoxic conditions and greatly exacerbates the response to hypoxia. Selective restoration of endothelial BH4 levels rescues the effects of systemic BH4 deficiency, whereas augmentation of endothelial BH4 biosynthesis protects against the development of hypoxia-induced pulmonary hypertension and vascular remodeling. The striking quantitative correlations between lung BH4 levels and the development of pulmonary hypertension across the 5 genetic mouse models evaluated in our studies reveal that BH4 bioavailability controls both pulmonary vascular tone and structural remodeling in a dose-dependent manner. Finally, we show that the effects of genetic alterations in BH4 availability on the pulmonary vasculature are mediated through the regulation of NO compared with superoxide synthesis by eNOS in response to both acute and chronic hypoxia.
The elevated RVSP in the BH4-deficient hph-1 mice is at least comparable to that previously observed in mice deficient in eNOS (eNOS–/–), with both strains showing exacerbated responses to chronic hypoxia.6,7,29 However, vascular remodeling, a hallmark of pulmonary hypertension, was not seen in lungs of eNOS–/– mice in normoxia at sea-level altitudes,25,29 whereas vascular remodeling was clearly evident in both hph-1 heterozygotes and homozygotes compared with WT littermates, even in normoxia. This discordance in the severity of the phenotype between genetic eNOS deficiency compared with eNOS dysfunction resulting from genetic BH4 deficiency suggests that loss of NO production alone is not the sole mediator of disease pathogenesis. Rather, our findings highlight the importance of increased eNOS-dependent superoxide production in playing a further pathogenic role in both the vasoconstrictor and remodeling processes. Indeed, vascular superoxide production has a number of potentially important effects in the vascular wall, including effects on NO signaling through scavenging and peroxynitrite generation and through modulation of redox-sensitive signaling pathways.13 Superoxide controls proliferation and apoptosis of pulmonary smooth muscle cells,17 and pulmonary hypertension in fetal lambs subjected to ligation of ductus arteriosus is associated with increased superoxide generation.14 Our findings highlight the importance of endothelial BH4 availability as a reciprocal modulator of both NO and superoxide production by eNOS and suggest a critical role for BH4 availability in the pathogenesis of pulmonary hypertension. Thus, BH4 provides a mechanistic link between previous observations of increased NOS protein levels and reduced NO bioactivity.11,12 Targeting NOS regulation and enzymatic coupling, rather than eNOS protein levels or total enzymatic activity, may be a more promising therapeutic strategy in pulmonary hypertension.
The pulmonary phenotype in the eNOS–/– mice may be further complicated by the fact that these mice have left ventricular hypertrophy secondary to systemic hypertension,30 which is not seen in the BH4-deficient hph-1 mice. Thus, the effect of systemic BH4 deficiency, at least in hph-1 mice, appears to be more critical or specific to the pulmonary circulation as opposed to the systemic circulation. We note that this discordance was also seen in the transgenic mouse expressing a smooth muscle–specific dominant-negative BMPR-II, which developed pulmonary but not systemic hypertension, even though the SM22 smooth muscle–specific promoter directs systemic and pulmonary expression.31
The interplay between the various putative pathways implicated in pulmonary hypertension at both the cellular level (eg, endothelium, smooth muscle cells, platelets, macrophages, inflammatory cells, and progenitor cells) and the molecular level (eg, BMPR-II, endothelin, serine elastase, serotonin, potassium channels, angiopoietin, NO, and superoxide) is complex and poorly defined.1 It is unlikely that any single environmental factor or gene will explain all forms of pulmonary hypertension. Nevertheless, our findings underscore the central importance of endothelial dysfunction in pulmonary hypertension.4 The relative bioactivities of NO and reactive oxygen species are critical to maintaining pulmonary vascular homeostasis, and the balance may be shifted toward injury in pulmonary hypertension.5 Our studies show that BH4 has a pivotal effect on the pathogenesis of pulmonary hypertension mediated through eNOS enzymatic coupling that maintains the homeostatic balance between NO and superoxide bioactivity in the endothelium. The striking "concentration-response" relationship between lung BH4 levels and both functional and structural features of pulmonary hypertension suggests that alterations in BH4 availability modify susceptibility to or protection from pulmonary hypertension in both normoxia and hypoxia. Indeed, we observed a pulmonary hypertensive phenotype even in hph-1 heterozygotes in which the magnitude of BH4 deficiency is modest compared with hph-1 homozygotes yet is sufficient to alter both pulmonary tone and structure. These observations compare similarly with eNOS+/– (heterozygous) mice, which, like eNOS–/– mice, have raised RVSP and hypersensitivity to hypoxia.7 Like the eNOS gene, our data suggest that loss of 1 allele of the GCH1 gene encoding GTPCH, the rate-limiting enzyme in BH4 synthesis, is sufficient to produce a pulmonary vascular phenotype. Accordingly, we speculate that GCH1 is a candidate modifier gene in pulmonary hypertension. Investigating possible associations between genetic variation in the GCH1 gene and pulmonary hypertension may prove informative. Human GCH1 mutations are rare but cause severe BH4 deficiency syndromes such as hyperphenylalaninemia and dopa-responsive dystonia, because BH4 is also a cofactor for enzymes involved in hepatic phenylalanine metabolism and neuronal dopamine synthesis.26 The effect of BH4 deficiency on the pulmonary circulation in these patients remains to be studied.
We note that BH4 deficiency in hph-1 mice did not lead to severe pulmonary hypertension with plexiform lesions as seen in humans,1 even after exposure to hypoxia, and the mice do not suffer premature death. This may not be surprising, given that other recently published genetic mouse models of pulmonary hypertension do not recapitulate this extreme phenotype.31,32,33 Even in mice overexpressing S100A4/Mts1, only 5% of aging mice develop plexiform arteriopathy.34 These observations imply that additional pathways need to be invoked for full pathogenic expression of the human disease.
A further limitation of this study is that we had not determined late changes to longer durations of hypoxia. Endothelial dysfunction may play a larger role in the initiation and early pathogenesis of pulmonary hypertension,4 and it is possible that endothelial BH4 augmentation alone may not fully protect against greater structural changes accompanying prolonged hypoxia, when additional molecular pathways may predominate.
Given the complex pathophysiology of pulmonary hypertension, targeting of multiple pathways with combination drug therapy may be necessary. The ability of BH4 to both augment NO synthesis and decrease superoxide production addresses 2 pathogenic mechanisms simultaneously and may be a key therapeutic target, as illustrated by the striking salutary effects of endothelial BH4 restoration or augmentation in the hph/GCH and GCH mice. Pharmacological or molecular strategies that target endothelial BH4 availability in patients with pulmonary hypertension need to be evaluated. It is also interesting to note that HMG-CoA inhibitors (statins), currently generating significant interest after their reversal of experimental pulmonary hypertension in rats,35 upregulate GTPCH mRNA and BH4 levels in vascular endothelial cells.36
In conclusion, we have used a range of complementary gene-modified murine models to provide compelling evidence that demonstrates a novel and important role for endothelial BH4 in the pulmonary circulation. Regulation of eNOS function by BH4, a critical mechanism in the pathogenesis of pulmonary hypertension, may offer new therapeutic opportunities in a disease that remains inadequately understood and treated.
Acknowledgments
This work was supported by the British Heart Foundation (RG/02/006 to Dr Channon; PG/055/17036 to Dr Zhao/Dr Wilkins). Dr Khoo is a Wellcome Trust Cardiovascular Research Initiative Fellow.
Footnotes
Drs Khoo and Zhao contributed equally to this study.
References
Humbert M, Morrell NW, Archer SL, Stenmark KR, MacLean MR, Lang IM, Christman BW, Weir EK, Eickelberg O, Voelkel NF, Rabinovitch M. Cellular and molecular pathobiology of pulmonary arterial hypertension. J Am Coll Cardiol. 2004; 43: 13S–24S.
Simonneau G, Galie N, Rubin LJ, Langleben D, Seeger W, Domenighetti G, Gibbs S, Lebrec D, Speich R, Beghetti M, Rich S, Fishman A. Clinical classification of pulmonary hypertension. J Am Coll Cardiol. 2004; 43: 5S–12S.
Newman JH, Trembath RC, Morse JA, Grunig E, Loyd JE, Adnot S, Coccolo F, Ventura C, Phillips JA 3rd, Knowles JA, Janssen B, Eickelberg O, Eddahibi S, Herve P, Nichols WC, Elliott G. Genetic basis of pulmonary arterial hypertension: current understanding and future directions. J Am Coll Cardiol. 2004; 43: 33S–39S.
Budhiraja R, Tuder RM, Hassoun PM. Endothelial dysfunction in pulmonary hypertension. Circulation. 2004; 109: 159–165.
Hampl V, Herget J. Role of nitric oxide in the pathogenesis of chronic pulmonary hypertension. Physiol Rev. 2000; 80: 1337–1372.
Steudel W, Scherrer-Crosbie M, Bloch KD, Weimann J, Huang PL, Jones RC, Picard MH, Zapol WM. Sustained pulmonary hypertension and right ventricular hypertrophy after chronic hypoxia in mice with congenital deficiency of nitric oxide synthase 3. J Clin Invest. 1998; 101: 2468–2477.
Fagan KA, Fouty BW, Tyler RC, Morris KG Jr, Hepler LK, Sato K, LeCras TD, Abman SH, Weinberger HD, Huang PL, McMurtry IF, Rodman DM. The pulmonary circulation of homozygous or heterozygous eNOS-null mice is hyperresponsive to mild hypoxia. J Clin Invest. 1999; 103: 291–299.
Champion HC, Bivalacqua TJ, Greenberg SS, Giles TD, Hyman AL, Kadowitz PJ. Adenoviral gene transfer of endothelial nitric-oxide synthase (eNOS) partially restores normal pulmonary arterial pressure in eNOS-deficient mice. Proc Natl Acad Sci U S A. 2002; 99: 13248–13253.
Kaneko FT, Arroliga AC, Dweik RA, Comhair SA, Laskowski D, Oppedisano R, Thomassen MJ, Erzurum SC. Biochemical reaction products of nitric oxide as quantitative markers of primary pulmonary hypertension. Am J Respir Crit Care Med. 1998; 158: 917–923.
Machado RF, Londhe Nerkar MV, Dweik RA, Hammel J, Janocha A, Pyle J, Laskowski D, Jennings C, Arroliga AC, Erzurum SC. Nitric oxide and pulmonary arterial pressures in pulmonary hypertension. Free Radic Biol Med. 2004; 37: 1010–1017.
Fagan KA, Morrissey B, Fouty BW, Sato K, Harral JW, Morris KG Jr, Hoedt-Miller M, Vidmar S, McMurtry IF, Rodman DM. Upregulation of nitric oxide synthase in mice with severe hypoxia-induced pulmonary hypertension. Respir Res. 2001; 2: 306–313.
Le Cras TD, McMurtry IF. Nitric oxide production in the hypoxic lung. Am J Physiol Lung Cell Mol Physiol. 2001; 280: L575–L582.
Cai H, Harrison DG. Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ Res. 2000; 87: 840–844.
Brennan LA, Steinhorn RH, Wedgwood S, Mata-Greenwood E, Roark EA, Russell JA, Black SM. Increased superoxide generation is associated with pulmonary hypertension in fetal lambs: a role for NADPH oxidase. Circ Res. 2003; 92: 683–691.
Bowers R, Cool C, Murphy RC, Tuder RM, Hopken MW, Flores SC, Voelkel NF. Oxidative stress in severe pulmonary hypertension. Am J Respir Crit Care Med. 2004; 169: 764–769.
Liu JQ, Folz RJ. Extracellular superoxide enhances 5-HT–induced murine pulmonary artery vasoconstriction. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L111–L118.
Wedgwood S, Black SM. Role of reactive oxygen species in vascular remodeling associated with pulmonary hypertension. Antioxid Redox Signal. 2003; 5: 759–769.
Katusic ZS. Vascular endothelial dysfunction: does tetrahydrobiopterin play a role; Am J Physiol Heart Circ Physiol. 2001; 281: H981–H986.
Alp NJ, Channon KM. Regulation of endothelial nitric oxide synthase by tetrahydrobiopterin in vascular disease. Arterioscler Thromb Vasc Biol. 2004; 24: 413–420.
Landmesser U, Dikalov S, Price SR, McCann L, Fukai T, Holland SM, Mitch WE, Harrison DG. Oxidation of tetrahydrobiopterin leads to uncoupling of endothelial cell nitric oxide synthase in hypertension. J Clin Invest. 2003; 111: 1201–1209.
McDonald JD, Cotton RG, Jennings I, Ledley FD, Woo SL, Bode VC. Biochemical defect of the hph-1 mouse mutant is a deficiency in GTP-cyclohydrolase activity. J Neurochem. 1988; 50: 655–657.
Cosentino F, Barker JE, Brand MP, Heales SJ, Werner ER, Tippins JR, West N, Channon KM, Volpe M, Luscher TF. Reactive oxygen species mediate endothelium-dependent relaxations in tetrahydrobiopterin-deficient mice. Arterioscler Thromb Vasc Biol. 2001; 21: 496–502.
Khoo JP, Nicoli T, Alp NJ, Fullerton J, Flint J, Channon KM. Congenic mapping and genotyping of the tetrahydrobiopterin-deficient hph-1 mouse. Mol Genet Metab. 2004; 82: 251–254.
Alp NJ, Mussa S, Khoo J, Guzik TJ, Cai S, Jefferson A, Rockett KA, Channon KM. Tetrahydrobiopterin-dependent preservation of nitric oxide–mediated endothelial function in diabetes by targeted transgenic GTP-cyclohydrolase I overexpression. J Clin Invest. 2003; 112: 725–735.
Zhao L, Mason NA, Morrell NW, Kojonazarov B, Sadykov A, Maripov A, Mirrakhimov MM, Aldashev A, Wilkins MR. Sildenafil inhibits hypoxia-induced pulmonary hypertension. Circulation. 2001; 104: 424–428.
Thony B, Auerbach G, Blau N. Tetrahydrobiopterin biosynthesis, regeneration and functions. Biochem J. 2000; 347 (pt 1): 1–16.
Fagan KA, Tyler RC, Sato K, Fouty BW, Morris KG Jr, Huang PL, McMurtry IF, Rodman DM. Relative contributions of endothelial, inducible, and neuronal NOS to tone in the murine pulmonary circulation. Am J Physiol. 1999; 277: L472–L478.
Weissmann N, Winterhalder S, Nollen M, Voswinckel R, Quanz K, Ghofrani HA, Schermuly RT, Seeger W, Grimminger F. NO and reactive oxygen species are involved in biphasic hypoxic vasoconstriction of isolated rabbit lungs. Am J Physiol Lung Cell Mol Physiol. 2001; 280: L638–L645.
Steudel W, Ichinose F, Huang PL, Hurford WE, Jones RC, Bevan JA, Fishman MC, Zapol WM. Pulmonary vasoconstriction and hypertension in mice with targeted disruption of the endothelial nitric oxide synthase (NOS 3) gene. Circ Res. 1997; 81: 34–41.
Huang PL, Huang ZH, Mashimo H, Bloch KD, Moskowitz MA, Bevan JA, Fishman MC. Hypertension in mice lacking the gene for endothelial nitric oxide synthase. Nature. 1995; 377: 239–242.
West J, Fagan K, Steudel W, Fouty B, Lane K, Harral J, Hoedt-Miller M, Tada Y, Ozimek J, Tuder R, Rodman DM. Pulmonary hypertension in transgenic mice expressing a dominant-negative BMPRII gene in smooth muscle. Circ Res. 2004; 94: 1109–1114.
Beppu H, Ichinose F, Kawai N, Jones RC, Yu PB, Zapol WM, Miyazono K, Li E, Bloch KD. BMPR-II heterozygous mice have mild pulmonary hypertension and an impaired pulmonary vascular remodeling response to prolonged hypoxia. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L1241–1247.
MacLean MR, Deuchar GA, Hicks MN, Morecroft I, Shen S, Sheward J, Colston J, Loughlin L, Nilsen M, Dempsie Y, Harmar A. Overexpression of the 5-hydroxytryptamine transporter gene: effect on pulmonary hemodynamics and hypoxia-induced pulmonary hypertension. Circulation. 2004; 109: 2150–2155.
Greenway S, van Suylen RJ, Du Marchie Sarvaas G, Kwan E, Ambartsumian N, Lukanidin E, Rabinovitch M. S100A4/Mts1 produces murine pulmonary artery changes resembling plexogenic arteriopathy and is increased in human plexogenic arteriopathy. Am J Pathol. 2004; 164: 253–262.
Girgis RE, Li D, Zhan X, Garcia JG, Tuder RM, Hassoun PM, Johns RA. Attenuation of chronic hypoxic pulmonary hypertension by simvastatin. Am J Physiol Heart Circ Physiol. 2003; 285: H938–H945.
Hattori Y, Nakanishi N, Akimoto K, Yoshida M, Kasai K. HMG-CoA reductase inhibitor increases GTP cyclohydrolase I mRNA and tetrahydrobiopterin in vascular endothelial cells. Arterioscler Thromb Vasc Biol. 2003; 23: 176–182.(Jeffrey P. Khoo, MD; Lan )