Bronchial epithelium-associated pulmonary arterial muscle relaxation in the rat is absent in the fetus and suppressed by postnatal hypoxia

Transcription

1 Am J Physiol Lung Cell Mol Physiol 288: L384 L389, First published October 29, 2004; doi: /ajplung Bronchial epithelium-associated pulmonary arterial muscle relaxation in the rat is absent in the fetus and suppressed by postnatal hypoxia J. Belik, J. Pan, R. P. Jankov, and A. K. Tanswell Canadian Institutes of Health Research Group in Lung Development, Lung Biology, and Integrative Biology Programmes, Hospital for Sick Children Research Institute; Clinical Integrative Biology, Sunnybrook and Women s Research Institute; and Departments of Paediatrics and Physiology, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada Submitted 16 August 2004; accepted in final form 22 October 2004 Belik, J., J. Pan, R. P. Jankov, and A. K. Tanswell. Bronchial epithelium-associated pulmonary arterial muscle relaxation in the rat is absent in the fetus and suppressed by postnatal hypoxia. Am J Physiol Lung Cell Mol Physiol 288: L384 L389, First published October 29, 2004; doi: /ajplung We recently reported the existence of a bronchial epithelium-derived relaxing factor (BrEpRF) capable of reducing pulmonary arterial smooth muscle force generation in the newborn rat. We reasoned in this study that BrEpRF has physiological significance in the control of pulmonary vascular tone. We hypothesized that the release and/or activity of this factor can be stimulated and is suppressed prenatally or under hypoxic conditions postnatally. Therefore, we evaluated the pathways stimulated by the BrEpRF in fetal and newborn rat intrapulmonary arteries mounted with their adjacent bronchi in a wire myograph under both normoxic and hypoxic conditions. Under normoxic conditions, BrEpRF release/activation was observed in newborn vessels following methacholine stimulation of M 2 muscarinic receptors, which was mediated via a nitric oxide (NO)-dependent mechanism involving the phosphatidylinositol 3-kinase pathway. Hypoxia suppressed the BrEpRF-dependent modulation of basal and methacholine-induced pulmonary arterial muscle tone in newborn vessels without altering endothelium-dependent or -independent NO-mediated relaxation. In fetal pulmonary arteries studied under normoxic conditions, BrEpRF neither was active under basal conditions nor could it be stimulated with methacholine. We conclude that release/activation of the BrEpRF occurs by an oxygen-dependent mechanism in the newborn and is suppressed during late fetal life. These results suggest that the BrEpRF may be involved in postnatal adaptation of the pulmonary circulation and that its suppression may contribute to hypoxic pulmonary vasoconstriction. muscarinic receptors; pulmonary vascular resistance; nitric oxide; nitric oxide synthase; phosphatidylinositol 3-kinase DURING FETAL LIFE, pulmonary vascular resistance is very high, thereby allowing only a small percentage of the total cardiac output to flow through the lungs (14, 16). Pulmonary vascular adaptation following birth is associated with alveolar expansion, pulmonary vasodilatation, and flow of the total cardiac output through the lungs, thus allowing for adequate gas exchange (9). Matching of perfusion to ventilation in the lung is acutely controlled by the hypoxic pulmonary vasoconstriction response (7). The mechanisms responsible for the control and modulation of pulmonary vascular tone are highly complex and remain poorly understood. It is clear that locally produced humoral factors with either dilator (nitric oxide, prostacyclin, and adenosine) or constrictor (thromboxane, endothelin, isoprostanes) properties are involved in the process (8), although much remains to be understood about their cellular sources and precise roles during the fetal-newborn transition. The possibility of involvement of the bronchi in the control of pulmonary arterial tone has only received limited attention. Cross talk between the bronchi and arteries of the lung might be reasonably expected to occur, given their common developmental origin and close anatomical proximity. Flavahan et al. (5) were the first to report on the existence of a bronchial epithelium-derived factor capable of relaxing airway smooth muscle. Subsequently, Fernandes et al. (4) demonstrated that a tracheal epithelium-derived factor was capable of relaxing aortic smooth muscle when the tissues were coaxially mounted in vitro. More recently, Prazma et al. (12) reported that this airway epithelium-derived relaxant factor was capable of modulating the tracheal vascular tone in vivo in the anesthetized rat. However, the potential role of this airway epithelium-derived relaxant factor on the control of pulmonary vascular tone has received little attention. In studies on pulmonary arterial reactivity in newborn rats, we have recently reported the presence of a bronchial epithelium-derived relaxing factor or factors (BrEpRF) that reduced pulmonary artery force generation, independently of the endothelium (2). The converse was not true, in that bronchial force generation was not affected when studied in the presence of the adjacent pulmonary artery. Given the above observations in the newborn, we reasoned that the BrEpRF might have physiological significance in the control of pulmonary vascular tone during the transition from fetal to postnatal life. We therefore hypothesized that the release and/or activity of this factor would be suppressed in the fetus and would be induced by increased oxygen tension. We also studied the methacholine response of fetal and newborn rat pulmonary arteries with their bronchi attached to evaluate whether BrEpRF release and/or activation could be stimulated pharmacologically. As described herein, we found that the BrEpRF was not active in fetal vessels and was suppressed by hypoxia in newborn vessels. METHODS Materials. U was obtained from Cayman Chemical (Ann Arbor, MI). All other chemicals were obtained from Sigma Chemical (Oakville, ON, Canada) and dissolved in Krebs-Henseleit buffer except for wortmannin, which was first dissolved in DMSO (10 2 M) and subsequently in Krebs-Henseleit buffer. Address for reprint requests and other correspondence: J. Belik, Univ. of Toronto, Division of Neonatology, Hospital for Sick Children, 555 Univ. Ave., Toronto, Ontario, Canada M5G 1X8 ( jaques.belik@sickkids.ca). The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. L /05 $8.00 Copyright 2005 the American Physiological Society

2 Institutional review. All procedures involving animals were conducted according to criteria established by the Canadian Council for Animal Care. Approval for the study was obtained from the Animal Care Review Committee of the Hospital for Sick Children Research Institute. Animal preparation. Four- to eight-day-old newborn Sprague- Dawley rats (Charles River, ON, Canada) were killed with an overdose of intraperitoneal pentobarbital sodium (50 mg/kg), and their lungs were removed immediately after death. Fetal rats of days of gestation (term 21 days) were delivered by hysterotomy (intramuscular pentobarbital sodium anesthesia, 20 mg/kg), and their lungs were removed immediately after delivery. Organ bath studies. Fourth-generation left lung intralobar pulmonary artery ring segments (average diameter 100 m, length 2 mm) were dissected free and mounted in a wire myograph in isolation or with the bronchi attached. Tissues were bathed in 5 ml of Krebs-Henseleit buffer (115 mm NaCl, 25 mm NaHCO 3, 1.38 mm NaHPO 4, 2.51 mm KCl, 2.46 mm MgSO 4, 1.91 mm CaCl 2, and 5.56 mm dextrose) bubbled with air-6% CO 2 (normoxia, organ bath PO Torr) or nitrogen-6% CO 2 (hypoxia, organ bath PO 2 40 Torr) and were maintained at 37 C. After1hofequilibration, the optimal resting tension of the tissue was determined by repeated stimulation with 128 mm KCl until maximum active tension was reached. All subsequent force measurements were obtained at optimal resting tension. The optimal resting tension was directly proportional to age, not affected by the presence of the attached bronchi, and varied between 1 and 2 mn. Isometric changes were digitized and recorded online (Myodaq; Danish Myo Technology, Aarhus, Denmark). Pulmonary arterial muscle force generation was evaluated by stimulating with KCl ( mm) and the thromboxane mimetic U ( M). Contractile responses were normalized to the tissue cross-sectional area as (width diameter) 2 and expressed as mn/mm 2. The relaxant responses to sodium nitroprusside (SNP) and acetylcholine (ACh) were determined by precontracting with U (10 6 M). To stimulate BrEpRF release and/or activity, pulmonary arteries with bronchi attached were precontracted with U (10 7 M, EC 50) in the presence of the endothelial nitric oxide synthase (enos) inhibitor L-N 5 -(1-iminomethyl)ornithine hydrochloride (L-NIO; 10 4 M) to suppress methacholine-induced endothelium-dependent nitric oxide (NO) formation, and the percentage relaxation induced by methacholine (10 5 M) was measured. To assess whether the relaxing factor acted via the NO-cGMP phosphatidylinositol 3-kinase (PI3- kinase) pathway, the methacholine response was measured in the presence of the soluble guanylate cyclase inhibitor 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ; 10 5 M) or wortmannin (10 7 M), which have been shown by others to block the NO-cGMP and PI3-kinase pathways, respectively (18, 20). To evaluate the involvement of muscarinic receptors in methacholine-induced relaxation of pulmonary arteries with attached bronchi, the response was evaluated in the presence of selective inhibitors to the three muscarinic receptors: M 1 (pirenzepine, 10 7 M), M 2 (methoctramine tetrahydrochloride, 10 6 M), and M 3 (p-fluorohexahydro-sila-difenidol hydrochloride, 10 6 M). The inhibitors concentrations were based on reported evidence of effective blockade (3, 11, 15, 21, 23). All inhibitors were added to the muscle bath 20 min before the experiments. Data analysis. Data were evaluated by Student s t-test or two-way ANOVA, as indicated. Statistical significance was accepted if P All statistical analyses were performed with the Number Cruncher Statistical System (Kaysville, UT). Data are presented as means (SD). RESULTS In an attempt to induce the release of BrEpRF, we evaluated the response to methacholine in U stimulated pulmonary arteries alone and with the bronchi attached. The rationale L385 for this approach was based on evidence that methacholine induces release of tracheal epithelium-derived relaxant factor in adult rats (4). Figure 1A illustrates a representative muscle force generation tracing obtained from a pulmonary artery alone or with the bronchus attached. In the presence of the enos-specific blocker L-NIO (10 4 M), methacholine induced relaxation solely in the pulmonary artery with the attached bronchus. Figure 1B illustrates the percentage relaxation measured in air-bubbled, U prestimulated newborn pulmonary artery either alone (n 4) or with bronchi attached (n 16). As shown, the methacholine-relaxant response of the newborn pulmonary artery is completely suppressed in the presence of the specific enos inhibitor L-NIO. In contrast, pulmonary arteries with the bronchi attached and in the presence of L-NIO exhibited significant muscle relaxant response to methacholine. The methacholine-induced relaxation response of pulmonary arteries with attached bronchi was reduced, but not statistically significantly lower, compared with the pulmonary artery alone (Fig. 1B). Fig. 1. A: representative force generation tracing of a newborn pulmonary artery both with and without the bronchus attached. B: U precontracted (10 7 M) and methacholine stimulated (MCh; 10 5 M) newborn pulmonary artery relaxation with (n 4) and without (n 4) an endothelial nitric oxide synthase (enos) inhibitor [L-N 5 -(1-iminomethyl)ornithine hydrochloride (L-NIO); 10 4 M] or with the bronchus attached [pulmonary (Pulm.) artery bronchi] in the absence (n 4) and presence (n 32) of L-NIO. **P 0.01 compared with MCh by t-test.

3 L386 Methacholine-induced relaxation of pulmonary arteries with attached bronchi was almost completely suppressed (Fig. 2) in the presence of the soluble guanylate cyclase inhibitor ODQ (10 5 M). The methacholine response was also significantly (P 0.01) reduced in the presence of the PI3-kinase inhibitor wortmannin (10 7 M) and the nonspecific muscarinic receptor inhibitor atropine (10 5 M). All these measurements were obtained in the presence of L-NIO. To identify the muscarinic receptor mediating methacholineinduced relaxation, we measured the response in the presence of increasing concentrations of M 1 -, M 2 -, and M 3 -selective muscarinic receptor blockers. Comparing the three inhibitors, we found the relaxation response to be significantly reduced at lower concentrations of the M 2 blocker (Fig. 3). The logic 50 (M) for the M 2 -selective muscarinic receptor blocker was and significantly lower (P 0.01) than that of the M 1 ( )- and M 3 ( )-selective muscarinic receptor blockers. These data suggest that M 2 is the muscarinic receptor primarily responsible for methacholine-induced relaxation. As shown in Fig. 4A, and in keeping with our previously reported data (2), newborn pulmonary arteries with bronchi attached generated significantly (P 0.01) less force than arteries alone when stimulated with U under normoxic conditions. Under hypoxic conditions, the U dose response for arteries without and with the attached bronchus was no longer different (Fig. 4A). The hypoxia effect on force generation was only observed in the pulmonary arteries with attached bronchi. As evident from Fig. 4A, in response to U (10 6 M), there was no significant effect of hypoxia on pulmonary arteries alone [force 2.4 (0.6) on air and 2.3 (0.7) mn/mm 2 under hypoxia]. In contrast, in response to the same molar concentration of U-46619, the force of pulmonary artery with attached bronchi significantly increased from 1.8 (0.6) on air to 2.7 (0.4) under hypoxic conditions (P 0.01). Methacholine-induced relaxation of U prestimulated newborn pulmonary arteries with attached bronchi was significantly reduced (P 0.01) under hypoxic conditions (Fig. 4B). Fig. 3. MCh (10 5 M)-induced relaxation of U prestimulated (10 7 M) newborn pulmonary arteries with attached bronchi preincubated with incremental molar concentrations of M 1- (pirenzepine, 10 7 M; n 4), M 2- (methoctramine tetrahydrochloride, 10 6 M; n 4), or M 3-(p-fluorohexahydro-sila-difenidol hydrochloride, 10 6 M; n 4) selective muscarinic receptor inhibitors. **P 0.01 compared with noninhibitor-exposed tissue values by 2-way ANOVA for repeated measurements. To exclude a direct effect of hypoxia on the NO-mediated smooth muscle relaxant response as an explanation for the above findings, we also evaluated the pulmonary artery response to endothelium-dependent (ACh) and -independent (SNP) agonists. We found that the presence of an attached bronchus did not alter the pulmonary artery muscle response to either relaxant agonist (Fig. 5), suggesting that the pulmonary arterial muscle NO-mediated relaxation is not acutely dependent on the O 2 tension, at least within the range tested. Finally, we evaluated fetal rat pulmonary arteries alone and with attached bronchi under both normoxic and hypoxic conditions. The BrEpRF was not active in fetal tissue under normoxic conditions as evidenced by the similar dose response to KCl for the pulmonary arteries alone and with bronchi attached (Fig. 6, top). Under hypoxic conditions, however, the pulmonary artery with attached bronchi generated a significantly greater (P 0.01) force compared with the pulmonary artery alone (Fig. 6, bottom). In addition, the methacholineinduced relaxant response of air-bubbled U prestimulated fetal pulmonary arteries with attached bronchi was negligible [0.3 (0.7) % relaxation]. In these arteries with attached bronchi, no significant difference in the U induced (10 6 M) force was observed when studied under air [1.7 (0.5)] and hypoxic [1.5 (0.4)] conditions. DISCUSSION Fig. 2. MCh-induced (10 5 M) relaxation of U prestimulated (10 7 M) newborn pulmonary arteries with attached bronchi in the absence (n 34) or presence of the soluble guanylate cyclase inhibitor 1H-[1,2,4]oxadiazolo[4,3- a]quinoxalin-1-one (ODQ, 10 5 M; n 4), the phosphatidylinositol 3-kinase inhibitor wortmannin (10 7 M; n 4), and the nonselective muscarinic inhibitor atropine (10 5 M; n 4). All these measurements were obtained in the presence of the enos inhibitor (L-NIO, 10 4 M). **P 0.01 compared with control values by 1-way ANOVA. We have previously reported that in the newborn rat, the presence of an attached bronchus significantly reduces the thromboxane analog and KCl-stimulated pulmonary artery force generation when studied in air-bubbled Krebs-Henseleit solution (2). This effect is dependent on the presence of an intact bronchial epithelium, but not pulmonary artery endothelium, and appears to be mediated by neuronal NOS and PI3-kinase (2). These data suggest that under basal (nonstimulated) normoxic conditions, the newborn rat s bronchi are continuously producing a BrEpRF.

4 L387 relaxant response similar in magnitude to the one observed in these arteries when exposed to the NO donor SNP. Methacholine is a muscarinic agonist similar to ACh. It induces contraction of airway smooth muscle, likely via stimulation of M 1 receptors (21). Similar to ACh, methacholine can also relax the adult rat pulmonary arteries via an endotheliumdependent muscarinic receptor of either the M 1 (23) or M 3 type (10). In this study, we showed that the methacholine-relaxant response of the newborn pulmonary artery is mediated via enos and is completely suppressed in the presence of the specific enos inhibitor L-NIO. In contrast, pulmonary arteries with the bronchi attached exhibited a significant muscle-relaxant response to methacholine in the presence of L-NIO. This suggests that in preparations with an attached bronchus, methacholine is able to induce pulmonary artery muscle relaxation via an enos-independent mechanism, likely modulated by BrEpRF release and/or activation. The methacholine-induced relaxation of the newborn pulmonary artery with attached bronchi is mediated via NO, given its dependence on the Fig. 4. A: thromboxane mimetic U dose-response curves for pulmonary arteries alone (Pa) and pulmonary arteries with bronchi attached (Pa Bronchi) in air- (Pa, n 17; Pa Bronchi, n 32) and nitrogen-bubbled (Pa, n 4; Pa Bronchi, n 7) Krebs-Henseleit solution. **P 0.01 compared with Pa values by 2-way ANOVA for repeated measurements. B: U precontracted (10 7 M) and MCh-stimulated (10 5 M) newborn pulmonary artery with the bronchus attached relaxation in air- (n 32) and nitrogen-bubbled (n 4) Krebs-Henseleit solution in the presence of L-NIO. **P 0.01 compared with air values. An airway epithelium-derived factor capable of relaxing canine bronchial smooth muscle was first reported by Flavahan et al. (5). It was subsequently shown to be present in the tracheal epithelium on the basis of the observation that methacholine- or histamine-stimulated tracheal tissue was able to relax endothelium-denuded adult rat aorta smooth muscle (4). In this study, we sought to determine whether BrEpRF release and/or activity could also be stimulated pharmacologically, in a similar fashion. We found that methacholine stimulation of newborn pulmonary arteries with bronchi attached induced a Fig. 5. Acetylcholine (ACh; top) and sodium nitroprusside (SNP; bottom) dose-response curves for thromboxane mimetic U prestimulated (10 6 M) pulmonary artery with bronchi attached in air- (ACh, n 11; SNP, n 4) and nitrogen-bubbled (ACh, n 13; SNP, n 8) Krebs-Henseleit solution.

5 L388 Fig. 6. KCl dose-response curves for pulmonary artery alone (Pa) and pulmonary arteries with bronchi attached (Pa Bronchi) in air- (top: Pa, n 7; Pa Bronchi, n 8) and nitrogen-bubbled (bottom: Pa, n 4; Pa Bronchi, n 8) Krebs-Henseleit solution. **P 0.01 compared with Pa values by 2-way ANOVA for repeated measurements. soluble guanylate cyclase, and involves the PI3-kinase pathway. Last, in contrast to its dual effect of epithelium-dependent relaxation and smooth muscle force generation on the airway, methacholine only induces endothelium-dependent relaxation in pulmonary arteries. Thus the possibility that bronchial smooth muscle contraction influences the attached pulmonary arterial relaxation potential cannot be excluded in the present study and deserves further investigation. Our finding that the same downstream pathways were stimulated in pulmonary arteries by methacholine as by bronchial epithelium alone suggests that muscarinic receptors are involved in the release and/or activation of BrEpRF. Muscarinic cholinergic receptors are divided into four major subtypes (M 1 to M 4 ) according to their pharmacological properties. In the rat lung, all receptor subtypes appear to be prevalent except for M 4 (21). In an attempt to identify the muscarinic receptor involved in methacholine-induced relaxation, we constructed dose-response curves for inhibitors with moderate selectivity for M 1, M 2, and M 3 receptors. Given the nonavailability of highly specific muscarinic receptor inhibitors (at high concentrations all receptors are inhibited with any of the blockers), we reasoned that this approach would enable us to identify the receptor most likely to mediate the methacholine-induced relaxant response. Methoctramine tetrahydrochloride, an M 2 receptor-selective inhibitor, significantly reduced the methacholine-induced relaxation at a lower molar concentration and exhibited a significantly lower logic 50 compared with the M 1 - and M 3 -selective inhibitors, suggesting that M 2 receptors modulate the methacholine-induced muscle relaxation. In this study, we demonstrated that under hypoxic conditions (buffer PO 2 40 Torr), the presence of attached bronchi altered neither the newborn pulmonary arterial muscle force generation nor the methacholine-induced relaxation, suggesting that the BrEpRF release and/or activity was suppressed by hypoxia. In addition, we showed that BrEpRF is not active in fetal vessels, even under normoxic conditions (buffer PO Torr). The mechanisms accounting for the dependence of BrEpRF release and/or activity on O 2 tension in the rat lung are not clear from the present data. In sheep, the pulmonary arterial response to histamine is O 2 and age dependent (6). Chronic (19), but not acute, hypoxia has been shown to alter the hypoxic pulmonary vasoconstriction response. Hypoxia has also been shown to induce, in fetal pulmonary vascular smooth muscle, the production of reactive oxygen species and NO, the combination of which may result in the formation of peroxynitrite anion (14). We have recently shown that in contrast to the adult, peroxynitrite anion induces significant pulmonary arterial contraction of newborn vessels and the formation of 8-isoprostane, another potent pulmonary vasoconstrictor (1). The possibility that peroxynitrite anion and/or 8-isoprostane formation, together with suppression of BrEpRF, are major contributors to hypoxic vasoconstriction in the newborn is a focus of our ongoing studies. The O 2 dependence of the methacholine-induced relaxant response is likely to be related to its effect on BrEpRF release and/or the signal transduction pathway upstream from the NO release. This conclusion is based on the fact that the endothelium-derived (ACh) and donor-dependent, NO-mediated pulmonary artery muscle relaxation was not affected under the hypoxic conditions utilized in our experiments. This is in contrast to the observed abrogation of the endothelium-dependent relaxation in the newborn pulmonary artery of the pig (22) and newborn rat (unpublished observations) subjected to chronic hypoxia. Last, the methodology employed in this study to induce in vitro hypoxia deserves further comment. The buffer PO 2 under the experimental condition employed in this study is significantly higher than the pulmonary arterial blood PO 2 in vivo (40 Torr), yet the vessel s tissue PO 2 levels are likely similar. This conclusion is based on the fact that in vitro tissue oxygenation depends on O 2 diffusion across the vessel wall as opposed to the vasa vasorum blood perfusion-dependent O 2 delivery in vivo. It is of interest that the BrEpRF appears to be either not present or not active during fetal life. As shown by our data in this study, this likely relates to the relatively hypoxic environment of the fetus (Pa O2 22 Torr in late gestation). Similar to our observation of the BrEpRF, expression of the airway epithelial sodium channel is PO 2 sensitive and upregulated immediately after birth (13). There is also evidence that the cgmp-pkg pathway in fetal pulmonary vessels is upregulated following O 2 exposure, suggesting that it is PO 2 regulated (14). In contrast to the newborn, hypoxia increased the KCl-induced, but not the U induced, force generation of pulmonary arteries with attached bronchi. Such distinct effect may be related to the mechanisms by which U causes contraction. Agonist-dependent differences in intracellular Ca 2 release, or hypoxia-mediated Ca 2 sensitization of the contractile apparatus (17), may be developmentally regulated through pathways on which BrEpRF exerts its effects. Whether the BrEpRF is acutely regulated by changes in gene expression, release of vasoactive molecules, or as is likely, both, remains to be elucidated. In summary, we have shown that the bronchial modulation of basal pulmonary arterial muscle tone under normoxic con-

6 ditions in the newborn is suppressed by hypoxia and is not functional during late fetal life in the rat. These results suggest that the BrEpRF is involved in postnatal adaptation of the pulmonary circulation and that its suppression contributes to hypoxic pulmonary vasoconstriction in the newborn period. GRANTS This work was supported by grants from the Canadian Institutes for Health Research. REFERENCES 1. Belik J, Pan J, Jankov R, and Tanswell A. Peroxynitrite inhibits relaxation and induces pulmonary artery muscle contraction in the newborn rat. Free Radic Biol Med 37: , Belik J, Pan J, Jankov RP, and Tanswell AK. A bronchial epitheliumderived factor reduces pulmonary vascular tone in the newborn rat. J Appl Physiol 96: , Chiba Y, Sakai H, and Misawa M. Characterization of muscarinic receptors in rat bronchial smooth muscle in vitro. Res Commun Mol Pathol Pharmacol 102: , Fernandes LB, Paterson JW, and Goldie RG. Co-axial bioassay of a smooth muscle relaxant factor released from guinea-pig tracheal epithelium. Br J Pharmacol 96: , Flavahan NA, Aarhus LL, Rimele TJ, and Vanhoutte PM. Respiratory epithelium inhibits bronchial smooth muscle tone. J Appl Physiol 58: , Gordon JB, Clement DC, and Chu K. Developmental changes in vascular responses to histamine in normoxic and hypoxic lamb lungs. J Appl Physiol 70: , Gurney AM. Multiple sites of oxygen sensing and their contributions to hypoxic pulmonary vasoconstriction. Respir Physiol Neurobiol 132: 43 53, Heymann MA. Control of the pulmonary circulation in the fetus and during the transitional period to air breathing. Eur J Obstet Gynecol Reprod Biol 84: , Iwamoto HS, Teitel DF, and Rudolph AM. Effects of lung distension and spontaneous fetal breathing on hemodynamics in sheep. Pediatr Res 33: , McCormack DG, Mak JC, Minette P, and Barnes PJ. Muscarinic receptor subtypes mediating vasodilation in the pulmonary artery. Eur J Pharmacol 158: , L McGowan SE, Smith J, Holmes AJ, Smith LA, Businga TR, Madsen MT, Kopp UC, and Kline JN. Vitamin A deficiency promotes bronchial hyperreactivity in rats by altering muscarinic M 2 receptor function. Am J Physiol Lung Cell Mol Physiol 282: L1031 L1039, Prazma J, Coleman CC, Shockley WW, and Boucher RC. Tracheal vascular response to hypertonic and hypotonic solutions. J Appl Physiol 76: , Rafii B, Tanswell AK, Otulakowski G, Pitkanen O, Belcastro-Taylor R, and O Brodovich H. O 2-induced ENaC expression is associated with NF- B activation and blocked by superoxide scavenger. Am J Physiol Lung Cell Mol Physiol 275: L764 L770, Raj U and Shimoda L. Oxygen-dependent signaling in pulmonary vascular smooth muscle. Am J Physiol Lung Cell Mol Physiol 283: L671 L677, Reinheimer T, Mohlig T, Zimmermann S, Hohle KD, and Wessler I. Muscarinic control of histamine release from airways. Inhibitory M 1- receptors in human bronchi but absence in rat trachea. Am J Respir Crit Care Med 162: , Rudolph AM and Heymann MA. The circulation of the fetus in utero. Methods for studying distribution of blood flow, cardiac output and organ blood flow. Circ Res 21: , Sauzeau V, Rolli-Derkinderen M, Lehoux S, Loirand G, and Pacaud P. Sildenafil prevents change in RhoA expression induced by chronic hypoxia in rat pulmonary artery. Circ Res 93: , Secondo A, Sirabella R, Formisano L, D Alessio A, Castaldo P, Amoroso S, Ingleton P, Di Renzo G, and Annunziato L. Involvement of PI3-K, mitogen-activated protein kinase and protein kinase B in the up-regulation of the expression of nnos and nnos splicing variants induced by PRL-receptor activation in GH3 cells. J Neurochem 84: , Shimoda LA, Sham JS, and Sylvester JT. Altered pulmonary vasoreactivity in the chronically hypoxic lung. Physiol Res 49: , Tsui DY, Gambino A, and Wanstall JC. S-nitrosocaptopril: in vitro characterization of pulmonary vascular effects in rats. Br J Pharmacol 138: , Tulic MK, Wale JL, Petak F, and Sly PD. Muscarinic blockade of methacholine induced airway and parenchymal lung responses in anaesthetised rats. Thorax 54: , Tulloh RM, Hislop AA, Boels PJ, Deutsch J, and Haworth SG. Chronic hypoxia inhibits postnatal maturation of porcine intrapulmonary artery relaxation. Am J Physiol Heart Circ Physiol 272: H2436 H2445, Wilson PS, Khimenko PL, Barnard JW, Moore TM, and Taylor AE. Muscarinic agonists and antagonists cause vasodilation in isolated rat lung. J Appl Physiol 78: , 1995.