Pulmonary Hypertension in a GTP-Cyclohydrolase 1–Deficient Mouse
Background— GTP-cyclohydrolase 1 (GTP-CH1) catalyzes the first step for the de novo production of tetrahydrobiopterin (BH4), a cofactor for nitric oxide synthase (NOS). The hyperphenylalaninemic mutant mouse (hph-1) displays a 90% reduction in GTP-CH1 activity. Reduced BH4 decreases NOS activity and may lead to endothelial dysfunction, and there is increasing evidence that a dysfunction of the NOS pathway may be implicated in pulmonary hypertension. The aim of the study was to investigate whether reduced BH4 in the hph-1 mouse results in a pulmonary hypertensive phenotype.
Methods and Results— Morphological characterization of the heart, lung, and kidney and measurements of systemic and right ventricular blood pressures were performed in both hph-1 and wild-type mice. BH4 and NOx levels were also measured. Hph-1 mice had significantly lower NOx and BH4 levels, consistent with previous findings. Both morphological and in vivo data were indicative of a pulmonary but not systemic hypertensive phenotype. We observed increased right ventricle–left ventricle plus septum ratios attributable only to an increase in right ventricular mass, increased smooth muscle medial area in pulmonary resistance vessels, and significantly higher right ventricular pressures in vivo. There were no significant differences between left ventricular masses and systemic pressures, and there was no observed evidence of systemic hypertension in kidney sections between wild-type and hph-1.
Conclusions— This study demonstrates that mice deficient in GTP-CH1/BH4 display a pulmonary hypertensive but not systemic hypertensive phenotype.
Received June 15, 2004; revision received January 14, 2005; accepted January 20, 2005.
The enzyme GTP-cyclohydrolase 1 (GTP-CH1) catalyzes the first step for the de novo production of tetrahydrobiopterin (BH4), a cofactor for nitric oxide synthase (NOS).1,2 Supplementation of BH4 restores endothelial function in a number of human and animal disease models,3–5 but less is known about the functional consequences of BH4 deficiency in blood vessels in vivo.
See p 2022
Recently, we found that GTP-CH1 is developmentally regulated in the pulmonary vasculature of the pig and that in the first few weeks of life, endothelium-dependent relaxation of pulmonary artery is enhanced by addition of BH4 (M.N., unpublished data, 2001). These observations suggested that BH4 might affect pulmonary vascular function and that loss of BH4 would impair NO responses. Because endothelial NOS (eNOS) protein is decreased in certain patients with primary pulmonary hypertension,6 and eNOS-knockout mice display significantly elevated right ventricular (RV)/pulmonary pressures under normoxia,7,8 increased RV pressures and remodeling after hypoxia,8 and a reduction in pulmonary pressures after adenoviral transfection of eNOS,7 we hypothesized that BH4 deficiency would result in a pulmonary hypertensive phenotype. To test this, we have studied the hyperphenylalaninemic mutant mouse (hph-1), a model that has a 90% reduction in GTP-CH1 activity.9
Tissue Collection for Morphological Characterization
Experiments were performed under a Home Office License and conducted according to the Animals Scientific Procedures Act 1986, United Kingdom. Twelve- to 17-week-old (≈25 g), sex-matched wild-type mice (C57BL/6 × CBA [WT]) and homozygous hph-1 mice were studied.
Perfusion and Fixing of Tissues for Paraffin Sectioning
Airway perfusion with 10% buffered formol saline via a cannula at a perfusion pressure of 20 cm H2O was performed, and fixed lungs were cut into transverse sections and dehydrated in 70% ethanol before being processed in a Hypercenter XP enclosed tissue processor. Kidneys were immersion fixed in formol saline and cut into 3 to 4 transverse and longitudinal sections before dehydration and processing. All subsequent histological analysis was performed by an individual blind to the mouse type.
Measurement of Pulmonary Arterial Medial Area
Standard immunohistochemical staining of 4-μm paraffin wax sections was performed with a mouse α-smooth muscle actin primary antibody 1:3000 dilution (DAKO). Light microscopy (Zeiss Axio-skop2) was used to anatomically identify ≈10 to 15 arteries per animal at the level of the alveolar ducts, respiratory bronchioles, and terminal bronchioles, and the arterial smooth muscle medial area was measured with Open Laboratory Software version 3.04 (Improvision). The medial smooth muscle area was expressed as a percentage of the total external area of the vessel. Extension of muscle into normally nonmuscularized arteries (<25 μm) was assessed as described previously into nonmuscular, partially muscular, and fully muscular arteries.8
Histological Examination of Pulmonary Sections
Standard hematoxylin-and-eosin staining and elastic van Gieson staining were performed on 4-μm lung sections and analyzed for evidence of any structural abnormalities, such as fibrosis and thickening of the alveolar septa, or for any evidence of edema that may have occurred secondary to left ventricular (LV) dysfunction.
Histological Examination of Kidneys
Kidney sections (4 μm) were stained with hematoxylin and eosin and elastic van Gieson and analyzed for evidence of renal damage caused by sustained systemic hypertension.
Measurement of Ventricular Weight
The RV was dissected from the LV plus septum (LV+S) and each component weighed separately. The RV:LV+S ratio was calculated. Additionally, RV and LV+S were indexed to body weight.
Hph-1 and WT mice were anesthetized with 5% halothane (Abbot Laboratories) and placed supine on a thermostatically controlled heating blanket (37°C). Anesthesia was maintained with 2% to 3% halothane inhalation through a small mask. To measure systemic blood pressure, the right carotid artery was isolated and a fluid-filled (heparin; 100 U/mL diluted in 0.9% saline), 0.28-mm internal-diameter cannula (Critchley Electrical Products Pty Ltd) introduced into the artery. To measure RV pressure, the right jugular vein was isolated and a fluid-filled cannula introduced into the heart and advanced into the RV. After a 5-minute stabilization period, mean systemic and RV pressures were measured and analyzed with MacLab version 3.4/e and Chart 4 Windows software.
In a separate set of mice from the same generation as those used in the present study, BH4 levels were measured in the brain, liver, and kidney with high-performance liquid chromatography and electrochemical detection, as described previously.9
Serum samples were analyzed for NOx levels. Briefly, whole blood was obtained by cardiac puncture and centrifuged at 12 000g at 4°C and the serum (supernatant) removed. Serum samples were centrifuged in 0.5-μm centrifuge filters (Millipore) at 12 000g for 1 hour. Plasma NOx was measured by a modified Griess reaction wherein plasma NO3− is first converted to NO2− by nitrate reductase (1 hour; 37°C), and the NO2− generated was quantified in terms of the diazonium product formed after reaction with sulfanilamide and N-(1-naphthyl)ethylenediamine.10
Data are expressed as mean±SEM, and statistical significance was tested with either ANOVA with Tukey-Kramer post hoc test or Student t test, where appropriate.
Histological Examination of Pulmonary Vasculature
The mean percentage of smooth muscle medial area in all arteries was greater in hph-1 mice (Figure 1a), with a significant difference seen in the arteries associated with the alveolar ducts (Figures 1a and 1b). The proportions of precapillary arteries at the level of the alveoli that were either nonmuscular (no actin staining), partially muscular (<75% circumferential actin staining), or fully muscular (>75% actin circumferential staining) differed between hph-1 and WT mice (Figure 1c), with fewer nonmuscular vessels and more partially or fully muscular vessels in hph-1 mice.
The mean RV:LV+S ratio was higher in hph-1 mice than in WT mice, and this was entirely due to the RV component, as shown when indexed to body weight (Figures 2A, 2B, and 2C).
Detailed examination of hematoxylin-and-eosin–stained lung sections revealed no abnormalities such as edema, fibrosis, hypercellularity, or thickening of the alveolar septa.
There was a significant increase in mean RV pressure in hph-1 mice (Figure 2D), with no significant difference in mean systemic blood pressure (hph-1 86.7±2.56 mm Hg, WT 93.6±3.0 mm Hg).
No features of hypertension were identified in the kidneys. Specifically, features indicative of hypertensive nephropathy (eg, intimal fibroelastosis, hyaline arteriosclerosis, and fibrinoid necrosis) were not seen (Figure 3).
BH4 levels were significantly lower in hph-1 mice than in WT in all tissues studied (Figure 4). BH4 measurements were technically difficult to obtain in lung tissue, but given that there was at least a 50% reduction in the other tissues studied, it is reasonable to conclude that this represents a global effect in this animal model.
Serum NOx levels were significantly lower in hph-1 mice than in WT mice (Figure 5).
The results from this study demonstrate that mice deficient in GTP-CH1/BH4 display the structural and hemodynamic features of pulmonary hypertension. All 3 structural characteristics of pulmonary hypertension (RV hypertrophy, increased smooth muscle wall area of resistance arteries, and extension of muscle into normally nonmuscular arteries) were present in hph-1 mice, and RV pressures were elevated. This identifies the critical importance of GTPCH-1 in the lung.
GTP-CH1 is a key enzyme controlling the synthesis of BH4 and is expressed in pulmonary vascular endothelium (M.N., unpublished observations, 2001). Although there are no genetic knockouts of GTP-CH1, possibly because of embryonic lethality, the hph-1 mouse is a chemical mutant that shows a 90% reduction in GTP-CH1 activity and a reduction of BH4, and is a useful model to study BH4 deficiency. The neurological phenotype of hph-1 mice has been studied extensively,9,11,12 and in addition to BH4 deficiency, these mice have reduced NOS activity that can be restored ex vivo by supplementation with BH4.9,13,14 Morphological examination of the pulmonary vasculature of hph-1 mice showed increased smooth muscle medial area in the smallest resistance arteries, with no change in lumen area. There was also extension of muscle into normally nonmuscularized precapillary arteries. These structural changes are characteristic of pulmonary hypertension, and the increase in wall-to-lumen ratio in the small pulmonary resistance vessels would be expected to enhance responses to contractile stimuli15 and increase resting pulmonary vascular resistance. Consistent with these findings, mean RV pressures were elevated in vivo, which confirms the presence of elevated pulmonary pressures. These findings were seen in both male and female mice in studies undertaken by individuals blind to the mouse type. Although a pulmonary hypertensive phenotype was evident in both sexes, it appeared more profound in the female mice.
Pulmonary hypertension can occur secondary to LV dysfunction, but the absence of edema in lung sections, the preservation of normal systemic blood pressure, and the lack of hypertrophy of the LV suggest that it is more likely that the pulmonary hypertension observed in the hph-1 mice resulted from reduced NO availability and subsequent pulmonary vascular changes. Mean RV pressures were significantly elevated in the hph-1 mice, and it would now be interesting to undertake a more detailed evaluation of the pulmonary hemodynamics, because this would facilitate a direct comparison with other models.
We were not able to measure BH4 accurately in the lungs and therefore could not determine whether BH4 levels were reduced in the relevant cells in the lungs; however, overall, the hph-1 mice displayed a 90% reduction in GTP-CH1 activity, and there was a >50% reduction in BH4 in all tissues in which it was measured. Because the defect in GTP-CH1 is thought to be global in hph-1 mice, it appears likely that BH4 was indeed reduced in the lung, as in other tissues. Therefore, the simplest explanation of the present data are that GTP-CH1/BH4 deficiency in hph-1 mice reduces NO generation, and this is supported by the observation of a reduction in the serum NOx levels in the hph-1 mice. An alternative explanation is that the deficiency of BH4 alters levels of aromatic amino acid derivatives (ie, biogenic amines), because BH4 is a cofactor for the aromatic amino acid hydroxylases. Indeed, the concentration of BH4 required to activate the aromatic amino acid hydroxylases is 20 orders of magnitude greater than that required to activate NOS.16 Therefore, it may be predicted that deficiency of BH4 would preferentially affect the levels of biogenic amines, including 5-hydroxytryptamine (5HT) and noradrenaline, an effect seen in the brains of hph-1 mice.17 However, in contrast to the effects we observed, deficiency of 5HT would be expected to render the animals resistant to pulmonary hypertension.18 It would now be interesting to determine to what extent biogenic amines may have been reduced in the lung in this model and why pulmonary hypertension predominates. One possibility is that deficiency of BH4 affects all isoforms of NOS; alternatively, an increase in superoxide production from BH4-deficient NOS may be important in mediating the effect.14 It is also possible that reduced β-noradrenergic receptor stimulation or noradrenaline-stimulated NO release may contribute to the phenotype.19
Hph-1 mice showed pulmonary but not systemic hypertension. This contrasts with a previous study in which an increase in systemic blood pressure was measured by the tail-cuff method in hph-1 mice.14 The reasons for this difference are not known but may relate to the different methods of assessing blood pressure or the effects of anesthetic on systemic blood pressure; however, the pulmonary vasculature appears to be particularly sensitive to changes in NOS activity,20,21 and inhibition of NOS results in attenuation of endothelium-dependent vasodilation in the pulmonary vasculature.22 The present study identifies pulmonary hypertension in the hph-1 mouse and suggests a critical role of GTP-CH1 in the regulation of pulmonary vascular tone.
This study was sponsored by the Medical Research Council and The British Heart Foundation.
Tayeh MA, Marletta MA. Macrophage oxidation of l-arginine to nitric oxide, nitrite, and nitrate: tetrahydrobiopterin is required as a cofactor. J Biol Chem. 1989; 264: 19654–19658.
Tiefenbacher CP, Chilian WM, Mitchell M, DeFily DV. Restoration of endothelium-dependent vasodilation after reperfusion injury by tetrahydrobiopterin. Circulation. 1996; 94: 1423–1429.
Tiefenbacher CP, Bleeke T, Vahl C, Amann K, Vogt A, Kubler W. Endothelial dysfunction of coronary resistance arteries is improved by tetrahydrobiopterin in atherosclerosis. Circulation. 2000; 102: 2172–2179.
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.
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.
Folkow B. The haemodynamic consequences of adaptive structural changes of the resistance vessels in hypertension. Clin Sci. 1971; 41: 1–12.
Katusic ZS. Vascular endothelial dysfunction: does tetrahydrobiopterin play a role? Am J Physiol Heart Circ Physiol. 2001; 281: H981–H986.
Cases A, Haas J, Burnett JC, Romero JC Hemodynamic and renal effects of acute and progressive nitric oxide synthesis inhibition in anesthetized dogs. Am J Physiol Regul Integr Comp Physiol. 2001; 280: R143–R148.