Vasoconstriction of the pulmonary vasculature is one of the pathophysiologic features of pulmonary hypertension (1-3), and vasodilators are therapeutically effective in some patients with pulmonary hypertension. The vasodilators most commonly used are inhaled nitric oxide (NO), intravenous prostacyclin, and oral calcium entry blockers (4,5). However, none of these drugs is ideal, and alternative vasodilators are needed for treating those patients with pulmonary hypertension who are responsive to vasodilator therapy. Because NO and prostacyclin relax pulmonary vascular smooth muscle by increasing intracellular cyclic guanosine monophosphate (cGMP) and cyclic adenosine monophosphate (cAMP), respectively, drugs that increase these cyclic nucleotides by other mechanisms may be possible alternatives. Phosphodiesterase (PDE) inhibitors are in this category because they prevent the breakdown of cAMP or cGMP or both.
If PDE inhibitors are to be useful in the treatment of pulmonary hypertension, two important requirements must be met. First, the drugs must be effective relaxants of pulmonary blood vessels even when the endothelium is damaged or removed, because the pulmonary vascular endothelium is often dysfunctional in patients with pulmonary hypertension (6). Second, the drugs must be potent not only in healthy pulmonary arteries but also in those that have undergone the changes associated with the development of pulmonary hypertension. The latter cannot necessarily be assumed because, for other drug groups (e.g., K+ channel openers; NO donors), there are reports of changes in potency as pulmonary hypertension develops (7-11).
The aim of this study was to evaluate the vasorelaxant properties of some PDE inhibitors on rat pulmonary arteries and, in particular, to determine whether they meet the requirements referred to. Hence experiments have been carried out (a) on artery preparations with and without endothelium, and (b) in arteries from rats in which experimental pulmonary hypertension was induced by long-term exposure to hypoxia. A particular emphasis of this study was to compare early and established pulmonary hypertension (1 and 4 weeks of exposure to hypoxia, respectively).
The predominant PDE isoenzymes in human pulmonary arteries are PDE III (breaks down cAMP) and PDE V (breaks down cGMP) (12). Therefore two PDE III inhibitors, milrinone and the novel drug, SCA40 (6-bromo-8 (methylamino) imidazo [1,2a] pyrazine-2-carbonitrile; 13,14) and one PDE V inhibitor, zaprinast, were compared. The nonselective PDE inhibitor, 3-isobutyl-1-methyl xanthine (IBMX) also was examined.
Preliminary accounts of some of these data were presented to meetings of the Australasian Society of Clinical and Experimental Pharmacologists and Toxicologists (Melbourne, December 1996; 15) and the British Pharmacological Society (Edinburgh, September 1997; 16).
Specific-pathogen free, male Wistar rats (8-9 weeks old on the day of the experiment) were used in this study.
Treatment of the rats
Some of the rats were housed in a hypoxic chamber, as described previously (17) for a period of 1 or 4 weeks. The chamber was flushed continuously with a mixture of N2 and compressed air at a flow rate of 0.94 L/min. The ratio of the two gas mixtures was adjusted to maintain a level of 10% O2 in the chamber. Trays of soda lime and calcium sulfate were placed in the chamber to absorb excess CO2 and moisture, respectively. The levels of CO2 and O2 were monitored daily with a Datex Normocap Gas Monitor (Datex, Helsinki, Finland). Rats were exposed to room air for a maximum of 10 min each day when the chamber was cleaned and food, drinking water, soda lime, and calcium sulfate were replenished. Control rats (normoxic) were housed in room air containing 21% O2. The age of the rats at the commencement of the treatment was selected so that all rats were the same age (8-9 weeks) at the end of the treatment period. Rats were randomly assigned to hypoxic and control groups.
On the day of the experiment, the animals were anesthetized with sodium pentobarbitone (120 mg/kg, i.p.). The chest cavity was opened and 250 IU of heparin was injected into the right ventricle. A blood sample was collected to measure the hematocrit. The main pulmonary artery was then dissected out. The hearts were subsequently removed, divided into right ventricle (RV) and left ventricle plus septum (LV + S), blotted, and weighed.
Pulmonary artery preparations
Ring preparations (3 mm in length) of main pulmonary artery were set up around two horizontal, stainless steel wires (4 mm in diameter) in a vertical organ bath containing physiologic salt solution (PSS; 37°C; 95% O2/5% CO2). The composition of the PSS was (in mM); NaCl, 118; KCl, 5.9; CaCl2, 1.5; MgSO4, 0.72; NaHCO3, 25; glucose, 11.7; and NaEDTA, 0.025. In some experiments, the endothelium was removed by gently rubbing the luminal surface of the vessel with a small pair of forceps. In the remainder of the experiments, care was taken not to damage the endothelium. Force in the circular muscle was recorded isometrically with a Statham Universal Transducer (UC3 + UL5) (Statham Instruments, Inc., Oxnard, CA, U.S.A.) or a Grass force displacement transducer (FT03) (Grass Instruments Co., Quincy, MA, U.S.A.) attached to a micrometer (Mitutoyo, Tokyo, Japan). The resting forces for preparations from control and hypoxic rats were 10 and 20 mN, respectively. These resting forces were based on previously published data (17,18) and take into account the different in vivo pulmonary artery pressures in control and hypoxic rats, respectively.
At the conclusion of the experiment, the following measurements were made: (a) the distance between the two stainless steel wires, and (b) the wet weight of the vessel, after blotting between filter paper for 30 s. The cross-sectional area in the plane perpendicular to the direction of the applied force [defined as (2 × vessel wall thickness × length of ring preparation)] was then calculated from the following formula: (Equation 1) where h is the distance between the wires plus the diameter of the two wires (mm), w is the wet weight (mg), and d is the density (1.06 mg/mm3).
Because the preparations were all the same length (3 mm), differences in cross-sectional area reflect differences in vessel wall thickness (17).
Blood-vessel preparations were allowed to equilibrate for 1 h (PSS changed every 15 min). They were then submaximally contracted with phenylephrine (0.1 μM), and when the contraction had reached equilibrium, acetylcholine (1 μM) was added to test for functional endothelium. The tissues were washed, and the PSS was then replaced with K+-depolarizing PSS (in which 80 mM NaCl was replaced with 80 mM KCl) to provide a reference contraction. The preparations were washed again to restore resting baseline force. Phenylephrine (0.1 μM) was then added to the baths and, once the contraction had stabilized, a concentration-response (relaxation) curve was obtained by adding cumulative concentrations (fivefold increments) of one of the following drugs: SCA40 (0.01-31 μM); milrinone (0.01-156 μM); zaprinast (0.2-625 μM); IBMX (0.2-125 μM), or forskolin (0.01-10 μM). In most experiments, two of these drugs were tested on any one preparation but in random order in different preparations. At the end of each experiment, pinacidil (0.3 mM) was added to the bath, fully to relax the tissue (19). This concentration of pinacidil maximally relaxes pulmonary arteries from both control and pulmonary hypertensive rats (17). Where maximum relaxation to pinacidil was below initial resting force, this indicated that preparations had inherent contractile tone. In one series of experiments with zaprinast, Nϑ-nitro-L-arginine methyl ester (L-NAME; 100 μM) was present in the bath 30 min before and also during the determination of the concentration-response curve.
Analysis of data
Contractions to K+ and phenylephrine were measured as force (mN) and expressed as stress (mN/mm2). Relaxant responses to vasodilator drugs were measured from the level of steady-state precontraction to phenylephrine. They were then expressed as a percentage of the full relaxation of the tissue, where full relaxation is defined as the difference between the force at steady-state precontraction and the force when the tissue was completely relaxed with 0.3 mM pinacidil. Inherent tone is defined as the difference between the initial resting force and the force when the tissue was completely relaxed with 0.3 mM pinacidil, and is expressed as stress (mN/mm2). The potencies of all drugs were calculated as the negative logarithm of the EC50 (negative log EC50) where EC50 is the concentration of drug producing 50% of the maximal response to the particular drug.
Drugs and solutions
The drugs used were acetylcholine (Sigma, St. Louis, MO, U.S.A.); forskolin (Sigma); heparin sodium (Commonwealth Serum Laboratories, Melbourne, Victoria, Australia); 3-isobutyl-1-methyl xanthine (IBMX; Sigma), milrinone (Sigma); L-NAME (Sigma); pentobarbitone sodium (Nembutal; Boehringer Ingelheim, Artarmon, NSW, Australia); phenylephrine hydrochloride (Sigma); pinacidil (Leo Pharmaceuticals, Ballerup, Denmark); SCA40 (6-bromo-8-methylaminoimidazo[1,2α]pyrazine-2-carbonitrile; gift from Professor P.A. Bonnet, University of Montpellier, France); U46619 (9.11-dideoxy-11α,9α-epoxymethano-prostaglandin F2α; Upjohn, Kalamazoo, MI, U.S.A.); zaprinast (Sigma).
Solutions of drugs were prepared as follows: acetylcholine (10 mM) and L-NAME (10 mM) in deionized water; phenylephrine (10 mM) and milrinone (10 mM) in 0.01 M HCl; and IBMX (10 mM) and zaprinast (10 mM) in 0.05 M NaOH. All dilutions were made in PSS. Pinacidil was dissolved in 0.1 M HCl, and dilutions were prepared in deionized water. Forskolin (10 mM) and SCA40 (10 mM) were dissolved in absolute ethanol. Dilutions of forskolin were in absolute ethanol. For SCA40, a 1 mM dilution was prepared in 50% ethanol, and subsequent dilutions were in deionized water.
Mean values were calculated and are quoted with their standard errors (SEM). All mean values, except for those expressed as a percentage value, were analyzed by parametric one-way analysis of variance (ANOVA) followed by a Tukey-Kramer or Dunnett's post hoc test, whichever was appropriate. Responses expressed as percentage values were analyzed by nonparametric one-way ANOVA followed by Dunn's post hoc test (Graphpad Instat 2.0).
Responses to PDE inhibitors on pulmonary arteries from control rats: preparations with and without endothelium
On pulmonary artery preparations, precontracted with phenylephrine (0.1 μM), each of the PDE inhibitors produced concentration-dependent relaxation and, at the highest concentrations used, relaxation was complete (i.e., maximal responses corresponded to 100% of the full relaxation elicited with pinacidil; see Methods; Fig. 1). When the endothelium was removed, the vasorelaxant responses to SCA40, milrinone, and IBMX were not altered (Fig. 1). In contrast, for zaprinast, removal of the endothelium caused a shift of the concentration-response curve to a higher concentration range (Fig. 1). In the absence of endothelium, the highest concentration of zaprinast used did not produce complete relaxation (86 ± 2.8%; n = 4). In endothelium-intact preparations, the NO synthase inhibitor, L-NAME (100 μM), caused a significant shift in the zaprinast concentration-response curve comparable to that seen when the endothelium was removed (Fig. 1).
Development of pulmonary hypertension in chronically hypoxic rats
On the day of the experiment, rats that had been exposed to hypoxia weighed significantly less than control rats (Fig. 2). Values for hematocrit, ratio of RV/(LV + S), and pulmonary artery cross-sectional area were all significantly increased when compared with control values (Fig. 2). These findings indicate the development of polycythemia, right ventricular hypertrophy, and pulmonary vascular hypertrophy, all of which are characteristic of chronic hypoxic pulmonary hypertension (17).
The polycythemia and right ventricular hypertrophy were significantly greater after 4 weeks of hypoxia than after 1 week of hypoxia (Fig. 2). The pulmonary vascular hypertrophy was not different in the two groups of hypoxic rats.
Functional characteristics of pulmonary arteries from pulmonary hypertensive rats
Pulmonary arteries from control rats had minimal inherent tone, but arteries from rats exposed to hypoxia for 1 week had a significant amount of tone (Fig. 3) and also showed spontaneous contractile activity. The development of inherent tone in these arteries was associated with a significant reduction in the size of the 80 mM K+ contraction (Fig. 3). However, the sum of the K+ contraction plus the inherent tone (a measure of overall contractile ability) was the same in control and 1-week hypoxic rats (1 week hypoxia, 29 ± 2.1 mN/mm2; n = 16; control, 34 ± 2.4 mN/mm2; n = 19). After 4 weeks of hypoxia, inherent tone was again minimal, as in arteries from control rats (Fig. 3), but the K+ contraction (Fig. 3) and the overall contractile ability (sum of K+ plus inherent tone; 44 ± 2.9 mN/mm2; n = 14) were significantly greater than control values (p < 0.05). The phenylephrine contraction varied in size in the three groups of rats but was always less than the K+ contraction (40-65% of K+; see Fig. 3).
Preparations from all groups of rats (except those from which the endothelium was deliberately removed) relaxed in response to acetylcholine (% of full relaxation; control, 38 ± 4.4; n = 19; 1 week hypoxia, 21 ± 1.4; n = 16; 4 weeks hypoxia, 44 ± 6.1; n = 14), indicating a functional endothelium. The relaxant response to acetylcholine after 1 week of hypoxia, but not after 4 weeks, was significantly less than that in arteries from control rats (p < 0.05).
Responses to PDE inhibitors and forskolin in pulmonary arteries from pulmonary hypertensive rats
Each of the PDE inhibitors completely relaxed pulmonary arteries from both groups of pulmonary hypertensive rats (i.e., they fully reversed the phenylephrine-induced tone plus any inherent tone). In pulmonary arteries from rats exposed to hypoxia for 1 week, the potencies of zaprinast, milrinone, and SCA40 were all significantly less than corresponding values in control rats (10-fold, sixfold, and fourfold, respectively; Table 1). However, in preparations from rats exposed to hypoxia for 4 weeks, the potencies of each of these PDE inhibitors were not significantly different from control values (Table 1). The potency of IBMX was not significantly different from control values in either group of hypoxic rats (Table 1).
The adenylate cyclase activator, forskolin, also was examined in arteries from control and hypoxic rats. The potency of forskolin was significantly reduced in arteries from both 1- and 4-week hypoxic rats when compared with controls (negative log EC50 values: control, 7.80 ± 0.07; n = 5; 1 week hypoxia, 6.73 ± 0.06; n = 3; p < 0.01; 4 weeks hypoxia, 7.22 ± 0.17; n = 6; p < 0.05). The reduction in potency was greater after 1 week of hypoxia (11-fold) than after 4 weeks of hypoxia (fourfold).
This study provided comparative data on the pulmonary vasorelaxant effects of four PDE inhibitors in a rat model of pulmonary hypertension. The PDE III inhibitors, milrinone and SCA40, the PDE V inhibitor, zaprinast and the nonselective PDE inhibitor, IBMX, were effective in relaxing precontracted main pulmonary artery preparations. Moreover, all of the drugs remained potent in pulmonary arteries taken from rats with established hypoxic pulmonary hypertension.
Of the drugs examined, zaprinast was the only one to be affected by removal of the endothelium, with responses being significantly reduced when compared with data in endothelium-intact preparations. The NO synthase inhibitor, L-NAME, had the same effect as removal of the endothelium. These data indicate that the PDE V inhibitor, zaprinast, produces part of its pulmonary vasorelaxant effect by inhibiting the breakdown of cGMP produced specifically by the action of endothelial NO. The source of the cGMP responsible for the residual effect of zaprinast, seen in the absence of endothelium or presence of L-NAME, is yet to be determined. The effect of endothelium removal on responses to zaprinast agrees with previous findings in various nonpulmonary vessels (20-22), and the data with L-NAME reflect previous results with NO synthase inhibitors in rat aorta (22), cat pulmonary vascular bed (23), and in vivo in lambs (24) and rats (25).
Although responses to zaprinast were attenuated by removal of the endothelium or inhibition of NO synthase, neither of these procedures eliminated responses to this PDE V inhibitor altogether. Thus it can be concluded that zaprinast, and presumably other PDE V inhibitors, will produce pulmonary vasodilation even when endothelial function is severely impaired. However, for optimal effect, a functional endothelium would be required.
An important part of this study was the examination of PDE inhibitors in pulmonary arteries from rats with hypoxic pulmonary hypertension. Rats exposed to hypoxia have been widely used to study both the structural and functional properties of pulmonary blood vessels in pulmonary hypertension, but the period of hypoxic exposure varies from 2 days to 4 weeks or more (8-10,26-28). Significant increases in pulmonary artery pressure are first observed after as little as 2 or 3 days of hypoxia (8,26). Changes in the functional properties of pulmonary blood vessels also can be detected after a few days of hypoxia (8,28) but can take ∼4 weeks of hypoxia to become stabilized (28). Hence it has been suggested that studies undertaken in rats exposed to hypoxia for <4 weeks reflect a transition period in the development of these changes (28). In view of this, we compared rats exposed to hypoxia for 1 week (early pulmonary hypertension, reflecting the transition period) and 4 weeks (established pulmonary hypertension, reflecting a time of stabilized changes). Both groups of hypoxic rats had significant right ventricular hypertrophy, indicative of increased pulmonary artery pressure. This was significantly greater after the longer period of hypoxic exposure, presumably reflecting the ongoing increase in workload to the right ventricle (29). Likewise arteries from both groups of hypoxic rats showed vascular hypertrophy. Interestingly this was of comparable magnitude in the two groups of rats, suggesting that the majority of the change in structure of main pulmonary artery occurs in the early stages of pulmonary hypertension. However, several differences were noted in the functional behavior of pulmonary arteries from rats with early pulmonary hypertension compared with established pulmonary hypertension.
After 1 week of hypoxia, the arteries showed spontaneous contractile activity and had inherent tone (i.e., were not fully relaxed under resting conditions). This confirms previous findings in 1-week hypoxic rats (9,19) and also reflects data obtained in rats exposed to hypoxia for 4 days (28) or 2 weeks (18,27). The development of inherent tone was accompanied by a corresponding decrease in the maximal contraction to K+. However, the overall contractile ability, defined as the sum of the inherent tone and the contraction to K+, was not different from the value in control arteries, as previously noted in 2-week hypoxic rats (18). After 1 week of hypoxia, there was also some diminution in endothelial function, in that the relaxant response to a standard concentration of acetylcholine was reduced. In arteries from 4-week hypoxic rats, the inherent tone, spontaneous contractions, and diminished endothelial function were no longer apparent. The only difference between these arteries and arteries from control rats was a significant increase in overall contractile ability. Note that the parameter used to quantify contractile ability took into account the increased vessel wall thickness in the pulmonary hypertensive arteries. Thus the increased contractile ability did not simply reflect the increased amount of smooth muscle; it is likely that the actual contractile properties of the smooth muscle were altered in established pulmonary hypertension. This could possibly be due to a change in vascular smooth-muscle cell phenotype, as suggested by Twarog et al. (28).
A significant finding from our study was that the data for the PDE inhibitors also differed in arteries from 1-week and 4-week hypoxic rats. Zaprinast, milrinone, and SCA40, but not IBMX, were each less potent in arteries from rats with early pulmonary hypertension (1 week of hypoxia) but after 4 weeks of hypoxia, when pulmonary hypertension was established, the potency of these drugs was restored. The potency change after 1 week of hypoxia did not simply reflect the need for the drugs to reverse the inherent tone, as well as the phenylephrine-induced contraction; if this were the explanation, the potency of all of the drugs would have been reduced.
There are at least three theoretic explanations for the reduction in potency of the PDE inhibitors seen in early pulmonary hypertension. First there could be increased levels of the relevant PDE enzymes such that higher concentrations of the PDE inhibitors would be needed to produce sufficient inhibition of the enzymes for vasorelaxation to occur. In support of this, cGMP and cAMP PDE enzymes have been reported to be increased in pulmonary arteries from rats exposed to hypoxia for 2 weeks (27,30). A second possibility is that basal levels of cGMP and cAMP are reduced in early pulmonary hypertension, as reported previously in main pulmonary arteries from hypoxic rats (1 or 2 weeks of hypoxia; 10,27) and in rat lungs (1 week of hypoxia; 31). The reduced pulmonary vascular endothelial function in main pulmonary artery from 1-week hypoxic rats, seen in this and other studies (19), is likely to contribute to any reduction in basal levels of cGMP. The third possibility is that the effectiveness of the cyclic nucleotides as vasorelaxants is reduced, but as yet there is no direct experimental evidence to support this concept. On the contrary, it has been shown that the activation of cGMP-dependent protein kinase is not altered in hypoxic rats (10).
Whatever the reason for the reduction in potency of the PDE inhibitors in rats with early pulmonary hypertension, it was encouraging that in established pulmonary hypertension (i.e., after exposure of the rats to hypoxia for 4 weeks), the potency of these drugs was no longer reduced. This was in contrast to data for the adenylate cyclase activator, forskolin (our study), and the NO donor, FK409 (which acts via activation of guanylate cyclase; 9). The potencies of both of these drugs were significantly reduced after 4 weeks, as well as 1 week, of hypoxia, although the reduction in potency was most pronounced in preparations from the 1-week hypoxic group.
In summary, this study showed that inhibitors of PDE III and V are good pulmonary vasorelaxants. The data suggest that inhibitors of both of these PDE isoenzymes would be effective even if the endothelium is damaged or functionally impaired. However, PDE V inhibitors require a fully functional endothelium for optimal effect. The drugs remained potent in rats with established pulmonary hypertension. This may be important in relation to the possible use of PDE inhibitors in the clinical treatment of patients with this disorder, although it should be remembered that data in rats cannot necessarily be extrapolated to human pulmonary hypertension. Finally, the study has highlighted a number of differences in the functional properties of main pulmonary artery from rats exposed to hypoxia for only 1 week compared with those exposed to hypoxia for longer periods (4 weeks). Therefore in studies on pulmonary vascular function and responsiveness to drugs in pulmonary hypertension, caution must be exercised in interpreting and comparing data from different studies in which rats have been exposed to hypoxia for different periods.
Acknowledgment: This study was supported by the National Health and Medical Research Council of Australia and this financial support is gratefully acknowledged. J.C.W. is a NH&MRC Senior Research Fellow. We thank Professor P-A. Bonnet, University of Montpellier, France, for his generous gift of SCA40, and Dr. Ian Ahnfelt-Ronne, Leo Pharmaceuticals, for provision of pinacidil.
1. Reid LM. Hypoxia and hyperoxia: patterns in vascular remodelling and cell adaptation. Prog Respir Res
2. Vender RL. Chronic hypoxic pulmonary hypertension; cell biology to pathophysiology. Chest
3. Voelkel NF, Tuder RM. Cellular and molecular mechanisms in the pathogenesis of severe pulmonary hypertension. Eur Respir J
4. Salvaterra CG, Rubin LJ. Investigation and management of pulmonary hypertension in chronic obstructive pulmonary disease. Am Rev Respir Dis
5. Rubin LJ. The medical management of pulmonary hypertension. In: Morice AH, ed. Clinical pulmonary hypertension.
London: Portland Press, 1995:235-45.
6. Dinh-Xuan AT, Higenbottam TW, Clelland TW, et al. Impairment of endothelium-dependent pulmonary artery relaxation in chronic obstructive lung disease. N Engl J Med
7. Wanstall JC, Kay CS, O'Donnell SR. Pinacidil-induced relaxation in pulmonary arteries isolated from pulmonary hypertensive and normotensive rats and precontracted with different spasmogens. Pulm Pharmacol
8. Wanstall JC, Hughes IE, O'Donnell SR. Reduced relaxant potency of nitroprusside on pulmonary artery preparations taken from rats during the development of hypoxic pulmonary hypertension. Br J Pharmacol
9. Wanstall JC, Kaye JA, Gambino A. The in vitro pulmonary vascular effects of FK409 (nitric oxide donor): a study in normotensive and pulmonary hypertensive rats. Br J Pharmacol
10. Crawley DE, Zhao L, Giembycz MA, et al. Chronic hypoxia impairs soluble guanylyl cyclase-mediated pulmonary arterial relaxation in the rat. Am J Physiol
11. Rodman DM. Chronic hypoxia selectively augments rat pulmonary artery Ca++
channel-mediated relaxation. Am J Physiol
12. Rabe KF, Tenor H, Dent G, Schudt C, Nakashima M, Magnussen H. Identification of PDE isozymes in human pulmonary artery and effect of selective PDE inhibitors. Am J Physiol
13. Cook SJ, Martin A, Buchheit K, et al. Further analysis of the mechanisms underlying the tracheal relaxant action of SCA40. Br J Pharmacol
14. Cortijo J, Villagrassa, V, Navarrete C, et al. Effects of SCA40 on human isolated bronchi and human polymorphonuclear leukocytes: comparison with rolipram and levcromakalim. Br J Pharmacol
15. Crilley TK, Wanstall JC. A comparative study of inhibitors of phosphodiesterases III and V in pulmonary arteries from pulmonary hypertensive and normotensive rats. Proc Aust Soc Clin Exp Pharmacol Toxicol
16. Wanstall JC, Crilley TK. Phosphodiesterase inhibitors on pulmonary arteries from rats with hypoxic pulmonary hypertension: comparison of 1 and 4 weeks hypoxia. Br J Pharmacol Suppl
17. Wanstall JC, O'Donnell SR. Responses to vasodilator drugs on pulmonary artery preparations from pulmonary hypertensive rats. Br J Pharmacol
18. Wanstall JC, Hughes IE, O'Donnell SR. Evidence that nitric oxide from the endothelium attenuates inherent tone in isolated pulmonary arteries from rats with hypoxic pulmonary hypertension. Br J Pharmacol
19. Wanstall JC, Crilley TK. Relaxation of rat pulmonary artery by adrenomedullin involves endothelium-derived nitric oxide and is attenuated in hypoxic pulmonary hypertension. Pharmacol Commun
20. Martin W, Furchgott RF, Villani GM, Jothianandan D. Phosphodiesterase inhibitors induce endothelium-dependent relaxation of rat and rabbit aorta by potentiating the effects of spontaneously released endothelium-derived relaxing factor. J Pharmacol Exp Ther
21. Harris A, Lemp B, Bentley R, Perrone M, Hamel L, Silver P. Phosphodiesterase isozyme inhibition and the potentiation by zaprinast of endothelium-derived relaxing factor and guanylate cyclase stimulating agents in vascular smooth muscle. J Pharmacol Exp Ther
22. Komas N, Lugnier C, Stoclet JC. Endothelium-dependent and independent relaxation of the rat aorta by cyclic nucleotide phosphodiesterase inhibitors. Br J Pharmacol
23. McMahon TJ, Ignarro LJ, Kadowitz PJ. Influence of zaprinast on vascular tone and vasodilator responses in the cat pulmonary vascular bed. J Appl Physiol
24. Braner DAV, Fineman JR, Chang R, Soifer SJ. M & B 2948, a cGMP phosphodiesterase inhibitor, is a pulmonary vasodilator in lambs. Am J Physiol
25. Dundore RL, Pratt PF, O'Connor B, Buchholz A, Pagani ED. Nϑ
-nitro-L-arginine attenuates the accumulation of aortic cyclic GMP and the hypotension produced by zaprinast. Eur J Pharmacol
26. Reid L. The pulmonary circulation: remodelling in growth and disease. Am Rev Respir Dis
27. MacLean MR, Sweeney G, Baird K, McCulloch KM, Houslay M, Morecroft I. 5-Hydroxytryptamine receptors mediating vasoconstriction in pulmonary arteries from control and pulmonary hypertensive rats. Br J Pharmacol
28. Twarog BM, Takuno H, Petrou S, Wahlqvist I, Marcus R, Campbell GL. Pathogenesis of pulmonary hypertension in the rat model. Chest
29. Wagenvoort CA, Wagenvoort N. Pathology of pulmonary hypertension.
New York: Wiley, 1977.
30. MacClean MR, Johnston ED, McCulloch KM, Pooley L, Houslay MD, Sweeney G. Phosphodiesterase isoforms in the pulmonary arterial circulation of the rat: changes in pulmonary hypertension. J Pharmacol Exp Ther
31. Rhoades R, Whittle E. Selective inhibition of hypoxia on rat lung cyclic AMP. Respir Physiol