Hypoxic Vasoconstriction of Rat Main Pulmonary Artery: Role ... : Journal of Cardiovascular Pharmacology (original) (raw)

Pulmonary hypertension may be secondary to hypoxic vasoconstriction, resulting in complex changes in the pulmonary vascular bed. Whether occurring as a primary illness or secondary to a chronic obstructive pulmonary disease, pulmonary hypertension is increasing in prevalence (1). The mechanisms for the development and maintenance of hypoxic pulmonary vasoconstriction are still poorly understood. Perturbation of vasodilatory signal transduction in hypoxia-induced pulmonary hypertension has been attributed to the suppression of endogenous vasodilator substances (2-4), decreased expression of receptors for vasodilators (5), and impaired guanylyl cyclase activity (6). In recent years, a body of evidence has emerged indicating that O2 level per se can regulate ion channel activity (7,8). Results from patch-clamp studies on isolated pulmonary vascular myocytes from rats and dogs have shown that acute hypoxia inhibits outward potassium (K+) current, causing depolarization (9). Most subsequent studies have indicated that the K+ current inhibited by hypoxia is voltage-gated and sensitive to 4-aminopyridine (10,11).

Pulmonary hypertension presents a poor prognosis regardless of its cause and the medical treatment for the disease remains unsatisfactory. The recent development of therapy, such as nitric oxide, prostacyclin, and analogues, represents a significant advance in the management of pulmonary hypertension (12,13). However, the therapeutic effects of these agents are limited by their short-half lives. A new pharmacologic treatment approach consists in prolonging the action of nitric oxide or prostacyclin by combining them with dilator agents and mediating their effects through inhibition of cyclic adenosine monophosphate (AMP) or guanosine monophosphate (GMP) hydrolysis (14,15). If numerous agents increase cyclic AMP or cyclic GMP levels through adenylyl or guanylyl cyclase activity, the hydrolysis of cyclic nucleotides is catalyzed solely by phosphodiesterase (PDE).

There are currently eleven partially characterized PDEs that are derived from at least 15 genes in the mammalian genome (16). Four of these families (1, 3, 4, and 5) are known to play a significant role in the regulation of vascular tone (17) and have been identified in human pulmonary circulation (18). The activity of PDEs 1-5 has been described in all pulmonary arteries of the rat (19). Isoenzyme-selective inhibitors are available for most PDE families (20). The inhibition of cyclic AMP and cyclic GMP hydrolysis by PDE inhibitors (PDEIs) has been shown to promote pulmonary vasodilatation (14), and the development of selective PDEIs may be of interest, alone or in combination with other relaxing agents, because PDE activity might be increased in the pulmonary vasculature in cases of chronic hypoxia (1,19).

The purpose of this study was first to investigate the relaxant effect of a nonselective PDEI (theophylline), a selective cyclic GMP-insensitive cyclic AMP PDEI (rolipram, PDE4I), a selective cyclic GMP-sensitive cyclic AMP PDEI (siguazodan, PDE3I), and a cyclic GMP PDEI (zaprinast, PDE5I) on hypoxic vasoconstriction of isolated rat main pulmonary arteries.

Recent evidence suggests involvement of potassium channels in the relaxant effects of PDEIs in coronary arteries (21) and small mesenteric arteries (22). The second aim of this study was therefore to assess the role of adenosine triphosphate-sensitive K+ channels (KATP), small-conductance Ca2+-activated K+ channels, large-conductance Ca2+-activated K+ channels, and 4-aminopyridine-sensitive voltage-gated K+ channels in these effects. In addition, because nitric oxide has been implicated in the regulation of pulmonary artery tone, we also examined the modulator role of endogenous nitric oxide by inhibition of nitric oxide synthase with l-_N_G-nitro-arginine methyl ester (l-NAME).

METHODS

Pulmonary artery preparation

Male Wistar rats (Dépré, St. Doulchard, France), weighing 270-370 g, were anesthetized by intraperitoneal injection of sodium pentobarbitone (100 mg/kg) and exsanguinated by transection of the carotid arteries. The heart and lungs were removed en bloc from the thoracic cavity and placed in ice-cold oxygenated Krebs-Henseleit solution (in m_M_: NaCl, 119; KCl, 5.4; CaCl2, 2.5; KH2PO4, 0.6; MgSO4, 1.2; NaHCO3, 25; glucose 11.7). The proximal right and left branches of the main pulmonary arteries were isolated and cleaned of all visible fat and connective tissue.

Measurement of relaxation

Each arterial segment was suspended in a 10-ml tissue bath by gently threading the rings onto a fixed, horizontal surgical-steel wire (300 μm in diameter, 5 mm in length). Once the wire was anchored, a second wire of the same dimensions was connected to a force transducer (UF1-Pioden strain gauge transducers). The tissue baths were filled with Krebs-Henseleit solution (composition as above, pH 7.4, 37°C) and gassed continuously with 95% O2 and 5% CO2 during equilibration. Isometric tension was amplified (Transbridge TBM4, WPI, Aston, U.K.) and recorded as a function of time on a chart recorder (Linseis 2000, Linseis, Munich, Germany). The arterial rings were subjected to an initial load of 1 g and were washed every 15 min with Krebs-Henseleit solution for 1 h. The tissues were allowed to equilibrate for another hour.

Functional procedures

After equilibration, vessels were exposed to 80 m_M_ potassium chloride until a maximal contraction was obtained. The Krebs solution was then changed four times at 5-min intervals, allowing vessels to relax back to baseline. Vessels where then re-exposed to 80 m_M_ potassium chloride and the average of these two contractions was recorded. After they were washed by changing Krebs solution four times at 5-min intervals, the vessels were then allowed to equilibrate. To assess endothelial integrity, the rings were contracted with phenylephrine (1 μ_M_) and then challenged with acetylcholine (0.1 μ_M_). Rings that did not relax to acetylcholine were withdrawn from experiments.

After endothelial function was assessed, vessels were washed again by changing Krebs solution four times at 5-min intervals; then vessels were allowed to equilibrate. When stable resting tension was recovered, the rings were precontracted with phenylephrine (0.1 μ_M_) and then placed under hypoxic conditions (95% N2 and 5% CO2). Under hypoxia, the vessels first relaxed and then developed a sustained contraction (Fig. 1). Once a stable contraction had been reached, the potassium channel blockers (glibenclamide 1 μ_M_, KATP channel; charybdotoxin 0.1 μ_M_, BKCa, large-conductance Ca2+-sensitive K+ channel; apamin 0.3 μ_M_, SKCa, small conductance Ca2+-sensitive K+ channel; and 4-aminopyridine 1 m_M_, Kv, voltage dependent) or the nitric oxide synthase inhibitor l-NAME (0.1 m_M_) was added to the bath of one of the branches of the pulmonary arteries. The other branch from the same animal was used as control.

F1-18

FIG. 1.:

Sample of trace of hypoxic vasoconstriction obtained in first branches of rat pulmonary arteries submitted to a mixture of 95% N2 and 5% CO2. In this case relaxation was achieved with theophylline by cumulative concentration.

Approximately 30 min after addition of K+-channel blockers or l-NAME—once a stable contraction had been reached again—cumulative concentration responsecurves were constructed with theophylline (nonselective PDEI, 0.01 μ_M_-3 m_M_), siguazodan (PDE3I), rolipram (PDE4I), or zaprinast (PDE5I), the range of concentrations used with the three selective PDEIs being 0.01 μ_M_-30 μ_M_. Each additional concentration was added to the bath every 5-15 min until a plateau was reached. The effects of agonists were expressed as a percentage of maximal relaxation induced with papaverine (0.1 m_M_) added to the bath at the end of the experiment as previously described (23,24). The formula for calculation is (A - x) ÷ (A - B), where A is the absolute tension before beginning the cumulative concentration response curve, B is the absolute tension after adding papaverine, and x is the tension level obtained after each addition of agonist. The agonist concentration producing 50% of it maximal response was derived from log-logit transformation of individual concentration-response curves. The effects of l-NAME or potassium channels on hypoxic vasoconstriction were expressed as percent change of the level of contraction induced by hypoxia

Statistical analysis

Data are expressed as mean ± SEM. Differences among groups were analyzed by ANOVA for repeated measures followed by the Bonferroni corrected t test or by Student's t test for paired or unpaired data as appropriate.

Drugs

The drugs and chemicals used and their sources were: theophylline sodium anisate (Delalande, Quétigny, France), siguazodan (SK&F 94836, (R,S)-2 cyano-1-methyl-3-[4-(methyl-6-oxo-1,4,5,6 tetrahydrophyridazine-3-yl)]-guanidine; SmithKline Beecham, Sussex, U.K.), rolipram (ZK 62771, 4-(3-cyclopentyloxy-4-methoxyphenyl)2-pyrolidone; SmithKline Beecham), zaprinast (M&B, 22,948, 2-_o_-propoxyphenyl-8-azapurin-6-one, Dagenham, Essex, U.K.); glibenclamide (Laboratoire Hoechst, Paris la Défense, France); 4-aminopyridine and l-NAME (Sigma Chimie, La Verpillère, France); and charybdotoxin and apamin (Latoxan, Rosans, France). Apamin, 4-aminopyridine, l-NAME, and theophylline were dissolved in distilled water, charybdotoxin in saline, glibenclamide in a mixture of dimethyl sulfoxide and distilled water (1:1), and siguazodan, rolipram, and zaprinast in a mixture of dimethyl sulfoxide, ethanol, and distilled water (0.15:0.5:0.35); the stock solutions were then diluted in distilled water. All the stock solutions were kept frozen until use and all the diluted solutions were prepared just before administration. The maximal concentrations of dimethyl sulfoxide (0.003%) or ethanol (0.42%) added to the bath did not by themselves exert any effect on hypoxic vasoconstriction and did not modify the reactivity of the preparation. All concentrations are expressed as final bath concentrations.

RESULTS

Influence of potassium channel blockers and nitric oxide synthase inhibition on hypoxic vasoconstriction

Charybdotoxin (0.1 μ_M_, BKCa, channel blocker) and apamin (0.3 μ_M_; SKCa, channel blocker) had no effect on hypoxic vasoconstriction, whereas glibenclamide (1 μ_M_, KATP-sensitive channel blocker) induced a 15% ± 2% increase in the hypoxic vasoconstriction (the level of tension raised from 1.24 ± 0.05 g to 1.31 ± 0.05 for a mean baseline tone of 0.83 ± 0.02, p < 0.05 vs. timematched controls, n = 33) and 4-aminopyridine (Kv, channel blocker) induced a significant 28% ± 2% decrease on hypoxic vasoconstriction (the level of tension decreased from 1.43 ± 0.05 to 1.31 ± 0.04, for a mean baseline tone of 0.85 ± 0.02, p < 0.01 vs. time-matched controls, n = 32). The blockade of nitric oxide synthase by l-NAME (0.1 m_M_) induced a more pronounced increase in hypoxic vasoconstriction (increase of 27% ± 3%) than that observed with glibenclamide (the level of tension raised from 1.35 ± 0.04 g to 1.53 ± 0.05 for a mean baseline tone of 0.88 ± 0.03, p < 0.01 vs. control experiments and vs. experiments performed in the presence of glibenclamide, n = 28). The combination of nitric oxide synthase and KATP-sensitive channel inhibition by the combined addition of l-NAME and glibenclamide did not further increase the hypoxic vasoconstriction, compared with that obtained with l-NAME alone (the level of tension raised from 1.36 ± 0.04 g to 1.57 ± 0.05, for a mean baseline tone of 0.82 ± 0.03, corresponding to 28% ± 3%, n = 12).

Relaxant effects of phosphodiesterase inhibitors

All the PDEIs tested induced strong relaxation of the rat main pulmonary arteries contracted under hypoxic conditions, with siguazodan being the most potent and theophylline the most efficient. The pD2 and Emax values for these relaxant agents are given in Table 1.

T1-18

TABLE 1:

Maximal effects (Emax) and potencies (pD2) values

Influence of potassium channel blockade on relaxation induced by phosphodiesterase inhibitors

Glibenclamide induced a significant shift to the right of the concentration-response curve for the PDEI5 zaprinast (n=8, ANOVA, p < 0.01), with a reduction of the pD2 value (pD2 = 6.28 ± 0.15 and 5.76 ± 0.11 in control and glibenclamide experiments, respectively) (Fig. 2). The maximal effect of zaprinast (30 μ_M_) was also significantly reduced by glibenclamide (Emax = 68% ± 4% and 54% ± 4% in control and glibenclamide experiments, respectively, p < 0.05). Glibenclamide had no significant effect on the concentration-response curves for theophylline (n = 10) and siguazodan (n = 10). For the concentration-response curve for rolipram there was a trend toward a decrease of the maximal effect that did not reach statistical significance (p = 0.061, n = 8).

F2-18

FIG. 2.:

Concentration-response curves for theophylline (nonselective phosphodiesterase inhibitor [PDEI]), siguazodan (PDE3I), rolipram (PDE4I), and zaprinast (PDE5I) without (filled circles) or with (open circles) the KATP-sensitive potassium channel antagonist glibenclamide (0.1 μ_M_). Results are expressed as mean ± SEM.

Charybdotoxin, apamin, and 4-aminopyridine had no effect on the concentration-response curves of any of the PDEIs used in this study in terms of either efficacy or potency (at least six experiments in each group; data not shown).

The inhibition of nitric oxide synthase by l-NAME (100 μ_M_, Fig. 3) significantly displaced to the right the concentration-response curve for theophylline and zaprinast (control experiments vs. l-NAME experiments, ANOVA, p < 0.05 for both theophylline [n = 7] and zaprinast [n = 7]. For the concentration-response curve for theophylline neither the Emax nor the pD2 were significantly modified by l-NAME, whereas for zaprinast the potency was significantly reduced by l-NAME (pD2 = 6.27 ± 0.08 for zaprinast alone and 5.68 ± 0.12 for zaprinast and l-NAME, p < 0.02). The difference in Emax value for zaprinast without or with l-NAME (68% ± 6% and 56% ± 6%, respectively) did not reach statistical significance.

F3-18

FIG. 3.:

Concentration-response curves for theophylline (nonselective phosphodiesterase inhibitor [PDEI]), siguazodan (PDE3I), rolipram (PDE4I), and zaprinast (PDE5I) without (filled triangles) or with (open triangles) blockade of nitric oxide synthase with l-N G-nitro-arginine methyl ester (0.1 m_M_). Results are expressed as mean ± SEM.

For rolipram the concentration-response curves obtained in the absence or in the presence of l-NAME were significantly distinct (ANOVA, p < 0.05), whereas neither the maximal effect (Emax = 61% ± 9% and 47% ± 7% in control and l-NAME experiments, respectively, not statistically significant) nor the potency (pD2 = 6.44 ± 0.27 for rolipram alone and 6.26 ± 0.13 for rolipram and l-NAME) was significantly affected by l-NAME.

The relaxation obtained in the presence of siguazodan was not significantly affected by l-NAME.

To investigate the link between KATP-sensitive channels and nitric oxide synthase in the relaxant effect of siguazodan and zaprinast, we performed the experiment with a combination of glibenclamide and l-NAME (concentrations as described above). Experiments with l-NAME alone were used as control experiments. As shown in Figure 4, in the presence of l-NAME, glibenclamide did not significantly displace the concentration-response curves for siguazodan or zaprinast.

F4-18

FIG. 4.:

Concentration-response curves for siguazodan and zaprinast obtained in the presence of l-N G-nitro-arginine methyl ester (l-NAME, 0.1 m_M_) alone (filled squares) or in the presence of l-NAME (0.1 m_M_) and glibenclamide (1 μ_M_) (open squares). Results are expressed as mean ± SEM.

DISCUSSION

Influence of potassium channels and endogenous nitric oxide on hypoxic vasoconstriction

We found that both glibenclamide (KATP-sensitive potassium channel blocker) and l-NAME (nitric oxide synthase inhibitor) induced a significant increase of hypoxic vasoconstriction of the main branches of rat pulmonary arteries. The effect of combining these two drugs was not superior to the effect of l-NAME alone. These results concur with those of Armstead (25), who found that hypoxic pial artery dilatation resulted from the sequential release of nitric oxide, cyclic GMP, and opioids, which in turn activate the KATP-sensitive potassium channels. Armstead has shown that the nitric oxide synthase inhibitor _N_-nitro-L-arginine (l-NNA) and glibenclamide both attenuated the hypoxic pial artery dilatation. The co-administration of these two drugs did not further decrease the already attenuated hypoxic pial vessel dilatation but blocked the release of cyclic GMP in cerebrospinal fluid, suggesting that nitric oxide and KATP channel contributions to hypoxic dilatation must occur at a common site. In the current study, the increase in hypoxic vasoconstriction of rat pulmonary arteries was more pronounced with l-NAME (+27%) than with glibenclamide (+15%), suggesting that endogenous nitric oxide exerts an antagonistic effect against hypoxic vasoconstriction that is only partly mediated through an activation of KATP-sensitive potassium channels. In our laboratory, we have already suggested that in the isolated perfused rat lung, exogenous nitric oxide (nitroglycerine) may partly exert its relaxant effect through KATP-sensitive potassium channels (26). The other part of the relaxant effect is likely to be attributed to the activation of protein kinase G by cyclic GMP (27).

It has recently been suggested that l-NAME had no effect on hypoxic vasoconstriction of isolated rat pulmonary arteries (28) after 30 min of hypoxia. It is likely that the influence of l-NAME is more pronounced when added after a longer hypoxic period (˜1.5 h in our study), which might be associated with an increase in endothelial production of nitric oxide.

The Kv antagonist, 4-aminopyridine, induced a 28% decrease in hypoxic vasoconstriction. This effect is unlikely to be attributable to a nonspecific toxic effect on the vascular smooth muscle cells because this loss of tone was stable in time and because the concentration of this antagonist is the most commonly used (29,30). The first hypothesis is that the prolonged (˜2 h) hypoxia before administration of 4-aminopyridine is responsible for an acidosis that induces changes in the potassium current through Kv channels. It has been shown that acidosis stimulates vasodilatation in the coronary microcirculation (31) and, in contrast, constricts the pulmonary vasculature (26,32). This result concurs with that of Ahn and Hume (33), who have demonstrated that superfusing canine pulmonary vascular smooth muscle cells with 20 m_M_ sodium butyrate to lower the intracellular pH (pHi) enhanced a 4-aminopyridine-sensitive Kv current by ˜20% over control levels. On the other hand, Berger et al. (30) have observed that Kv current is diminished by lowering pHi in rat pulmonary vascular cells and that 4-aminopyridine (3 m_M_) abolished this change in K+ currents. It is widely accepted that voltage-gated K+ channels may be important in determining the membrane potential and play a role in hypoxic pulmonary vasoconstriction (34). It has also been hypothesized that another potassium current (a voltage-gated, low-threshold, non-inactivating K+ current, _I_K(N)) might be a critical factor in the phenomenon of hypoxic vasoconstriction (11,35). Indeed, in the study by Walker et al. (35) hypoxia and 4-aminopyridine both inhibited _I_K(N) and induced contraction, whereas when 4-aminopyridine was administered under hypoxic conditions, as it is in our study, the amplitude of _I_K(N) returned to control levels and the hypoxic contraction was decreased.

Relaxant effects of phosphodiesterase inhibitors

We have observed that PDE3I (siguazodan), PDE4I (rolipram), and PDE5I (zaprinast) were able to produce strong relaxation of rat main pulmonary arteries constricted under hypoxic conditions. The relaxant efficacy of rolipram was less pronounced than that of siguazodan and zaprinast, but rolipram and siguazodan were more potent than zaprinast. The level of relaxation induced by PDE5 inhibition was the same in our study (˜70%) as in isolated perfused rat lungs (˜68%) described by Cohen et al. (36). This is in agreement with previous studies that have shown that PDE3, PDE4, and PDE5 (37,38) are abundantly distributed in lung tissue. Maclean et al. (19) have observed that in chronic hypoxic rats the total cyclic AMP and cyclic GMP PDE activity was increased compared with control experiments. For cyclic GMP, differences were significant only in first-branch pulmonary arteries and in intrapulmonary arteries but not in main pulmonary arteries or in pulmonary resistance arteries, whereas for cyclic AMP the difference was also significant in main pulmonary arteries. This finding might be of further clinical interest in the management of chronic pulmonary hypertension, since Hanasato et al. (39) have demonstrated that E-4010, an orally effective selective PDE5I, attenuates pulmonary hypertension, right ventricular hypertrophy, and pulmonary arterial remodeling induced by exposure to chronic hypoxia in rats without affecting systemic arterial blood pressure, heart rate, or cardiac output. Furthermore, Kodama and Adachi (40) have observed that long-term treatment with E-4010 improves survival in monocrotaline-induced pulmonary hypertensive rats.

Influence of potassium channels and endogenous nitric oxide on response to selective phosphodiesterase inhibitors

In our study the concentration-response curve for zaprinast was displaced to the right by the blockade of either endogenous nitric oxide by l-NAME or KATP-sensitive potassium channels by glibenclamide. This finding allows us to hypothesize that PDE5I relaxes best when the endogenous production of cyclic GMP is elevated, as suggested by Jeffery et al. (37). The effects of zaprinast are at least partly mediated through activation of KATP-sensitive potassium channels that seem to be activated by cyclic GMP (25). This idea was supported by the fact that glibenclamide did not further displace the concentration-response curve for zaprinast obtained in the presence of l-NAME.

The potentiation of hypoxic vasoconstriction and the antagonism on zaprinast-induced relaxation induced by nitric oxide synthase inhibition indicate that nitric oxide and cyclic GMP have a role in modulating pulmonary artery tone in healthy humans (41).

The lack of effect of any of the potassium channel blockers on the concentration-response curve for the PDE3I used in this study (siguazodan) is in agreement with the results of Taylor and Benoit (22), who found that the PDE3I milrinone was able to reduce the vasoconstriction of rat small mesenteric arteries induced by norepinephrine without the contribution of K+ channel regulation. Since PDE3 is a cyclic GMP-sensitive cyclic AMP PDE, we would have expected l-NAME to increase the relaxant effect of siguazodan by inhibiting nitric oxide production and thus cyclic GMP production and we would have expected the concentration-response curve for rolipram to be unaffected by l-NAME. Conversely, there was a slight inhibitory effect of l-NAME on the concentration response-curve for rolipram and, although not significant, for siguazodan. It has recently been reported that L-arginine (the substrate for nitric oxide production) and amrinone or milrinone (both PDE3Is) were additive in the vasodilatation of porcine internal mammary arteries and that l-NAME inhibited the effects of milrinone (42). It is likely that the rightward displacement of the concentration-response curves for rolipram and siguazodan observed in the current study in the presence of l-NAME is linked to the suppression of the additive effect of endogenous nitric oxide produced under our hypoxic conditions. The rightward displacement of the concentration-response curve of theophylline in the presence of l-NAME is not surprising. This antagonism is likely to be attributable to the lack of selectivity of theophylline. Among all the mechanisms involved in the relaxing effects of theophylline, inhibition of both PDE3 and PDE5 might explain the results of the current study.

In conclusion, (a) nitric oxide exerts a permanent antagonistic effect against hypoxic vasoconstriction that is partly mediated through activation of KATP, (b) neither small-conductance nor large-conductance Ca2+-sensitive potassium channels are involved in the response of pulmonary arteries to hypoxia and the relaxing effects of PDEI, (c) under our hypoxic conditions there is a 4-AP-sensitive potassium current (maybe an _I_K(N)) that increases the constriction of first branches of rat pulmonary arteries and that is likely to be partly activated through a pH decrease, (d) PDE3I and PDE5I are powerful relaxing agents against hypoxic vasoconstriction that might be of clinical interest in the management of pulmonary hypertension, and (e) PDE5Is require cyclic GMP produced by guanylate cyclase after activation of the endothelial nitric oxide synthase-nitric oxide pathway for optimal effect and PDE5I effects are partly mediated by KATP-sensitive potassium channels.

Acknowledgment:

Special thanks to Pr. Advenier for careful reading of the manuscript. This work was funded by grants from the University of Burgundy and the Conseil Régional of Burgundy.

REFERENCES

1. Wagner RS, Smith CJ, Taylor AM, et al. Phosphodiesterase inhibition improves agonist-induced relaxation of hypertensive pulmonary arteries. J Pharmacol Exp Ther 1997;282:1650-7.

2. Liu SF, Crawley DE, Barnes PJ, et al. Endothelium-derived relaxing factor inhibits hypoxic pulmonary vasoconstriction in rats. Am Rev Respir Dis 1991;143:32-7.

3. Shaul PW, Muntz KH, DeBeltz D, et al. Ontogeny of the adenylate cyclase system in the pulmonary vascular smooth muscle of the fetal lamb. J Cardiovasc Pharmacol 1991;17:125-33.

4. Shaul PW, North AJ, Brannon TS, et al. Prolonged in vivo hypoxia enhances nitric oxide synthase type I and type III gene expression in adult rat lung. Am J Respir Cell Mol Biol 1995;13:167-74.

5. Shaul PW, Muntz KH, DeBeltz D, et al. Effects of prolonged hypoxia on adenylate cyclase activity and beta-adrenergic receptors in pulmonary and systemic arteries of the rat. Circ Res 1990;66:1526-34.

6. Crawley DE, Zhao L, Giembycz MA, et al. Chronic hypoxia impairs soluble guanylyl cyclase-mediated pulmonary arterial relaxation in the rat. Am J Physiol 1992;263:L325-32.

7. Weir EK, Archer SL. The mechanism of acute hypoxic pulmonary vasoconstriction: the tale of two channels. FASEB J 1995;9:183-9.

8. Peers C. Oxygen-sensitive ion channels. Trends Pharmacol Sci 1997;18:405-8.

9. Post JM, Gelband CH, Hume JR. [Ca2+]i inhibition of K+ channels in canine pulmonary artery. Novel mechanism for hypoxia-induced membrane depolarization. Circ Res 1995;77:131-9.

10. Yuan XJ, Goldman WF, Tod ML, et al. Hypoxia reduces potassium currents in cultured rat pulmonary but not mesenteric arterial myocytes. Am J Physiol 1993;264:L116-23.

11. Osipenko ON, Evans AM, Gurney AM. Regulation of the resting potential of rabbit pulmonary artery myocytes by a low threshold, O2-sensing potassium current. Br J Pharmacol 1997;120:1461-70.

12. Olschewski H, Walmrath D, Schermuly R, et al. Aerosolized prostacyclin and iloprost in severe pulmonary hypertension. Ann Intern Med 1996;124:820-4.

13. Manktelow C, Bigatello LM, Hess D, et al. Physiologic determinants of the response to inhaled nitric oxide in patients with acute respiratory distress syndrome. Anesthesiology 1997;87:297-307.

14. Ichinose F, Adrie C, Hurford WE, et al. Selective pulmonary vasodilation induced by aerosolized zaprinast. Anesthesiology 1998;88:410-6.

15. Schermuly RT, Roehl A, Weissmann N, et al. Subthreshold doses of specific phosphodiesterase type 3 and 4 inhibitors enhance the pulmonary vasodilatory response to nebulized prostacyclin with improvement in gas exchange. J Pharmacol Exp Ther 2000;292:512-20.

16. Torphy TJ, Page C. Phosphodiesterases: the journey towards therapeutics. Trends Pharmacol Sci 2000;21:157-9.

17. Polson JB, Strada SJ. Cyclic nucleotide phosphodiesterases and vascular smooth muscle. Annu Rev Pharmacol Toxicol 1996;36:403-27.

18. Rabe KF, Tenor H, Dent G, et al. Identification of PDE isozymes in human pulmonary artery and effect of selective PDE inhibitors. Am J Physiol 1994;266:L536-43.

19. Maclean MR, Johnston ED, McCulloch KM, et al. Phosphodiesterase isoforms in the pulmonary arterial circulation of the rat: changes in pulmonary hypertension. J Pharmacol Exp Ther 1997;283:619-24.

20. Torphy TJ. Phosphodiesterase isozymes: molecular targets for novel antiasthma agents. Am J Respir Crit Care Med 1998;157:351-70.

21. Chen CH, Nakaya Y, Minami K, et al. Effects of pimobendan and its active metabolite, UD-CG 212 Cl, on Ca2+-activated K+ channels in vascular smooth-muscle cells. J Cardiovasc Pharmacol 1997;30:739-43.

22. Taylor MS, Benoit JN. Effect of milrinone on small mesenteric artery vasoconstriction: role of K(+) channels. Am J Physiol 1999;277:G69-78.

23. Tkachenko MN, Sagach VF, Bazilyuk OV, et al. Involvement of endothelium in vasodilating effects of vintoperol. J Cardiovasc Pharmacol 1992;20:S90-3.

24. Walch L, Labat C, Gascard JP, et al. Prostanoid receptors involved in the relaxation of human pulmonary vessels. Br J Pharmacol 1999;126:859-66.

25. Armstead WM. Relationship among NO, the KATP channel, and opioids in hypoxic pial artery dilation. Am J Physiol 1998;275:H988-94.

26. Dumas M, Dumas JP, Rochette L, et al. Comparison of the effects of nicorandil, pinacidil and nitroglycerin on hypoxic and hypercapnic pulmonary vasoconstriction in the isolated perfused lung of rat. Br J Pharmacol 1996;117:633-8.

27. Moncada S, Palmer RM, Higgs EA. Nitric oxide: physiology, pathophysiology, and pharmacology. Pharmacol Rev 1991;43:109-42.

28. Jones RD, Morice AH. The effect of the nitric oxide synthase inhibitor _N_-gamma-nitro-L-arginine methyl ester on hypoxic pulmonary vasoconstriction. Eur J Pharmacol 2000;402:111-7.

29. Onoue H, Katusic ZS. Role of potassium channels in relaxations of canine middle cerebral arteries induced by nitric oxide donors. Stroke 1997;28:1264-70.

30. Berger MG, Vandier C, Bonnet P, et al. Intracellular acidosis differentially regulates Kv channels in coronary and pulmonary vascular muscle. Am J Physiol 1998;275:H1351-9.

31. Ishizaka H, Kuo L. Acidosis-induced coronary arteriolar dilation is mediated by ATP-sensitive potassium channels in vascular smooth muscle. Circ Res 1996;78:50-7.

32. Wray S. Smooth muscle intracellular pH: measurement, regulation, and function. Am J Physiol 1988;254:C213-25.

33. Ahn DS, Hume JR. pH regulation of voltage-dependent K+ channels in canine pulmonary arterial smooth muscle cells. Pflugers Arch 1997;433:758-65.

34. Archer SL, Weir EK, Reeve HL, et al. Molecular identification of O2 sensors and O2-sensitive potassium channels in the pulmonary circulation. Adv Exp Med Biol 2000;475:219-40.

35. Walker SD, Edwards G, Weston AH. Functional significance of IK(N) in rat pulmonary artery. Br J Pharmacol 1998;98:88P.

36. Cohen AH, Hanson K, Morris K, et al. Inhibition of cyclic 3′-5′-guanosine monophosphate-specific phosphodiesterase selectively vasodilates the pulmonary circulation in chronically hypoxic rats. J Clin Invest 1996;97:172-9.

37. Jeffery TK, Wanstall JC. Phosphodiesterase III and V inhibitors on pulmonary artery from pulmonary hypertensive rats: differences between early and established pulmonary hypertension. J Cardiovasc Pharmacol 1998;32:213-9.

38. De Witt BJ, Marrone JR, Kadowitz PJ. Comparison of responses to siguazodan, rolipram, and zaprinast in the feline pulmonary vascular bed. Eur J Pharmacol 2000;406:233-8.

39. Hanasato N, Oka M, Muramatsu M, et al. E-4010, a selective phosphodiesterase 5 inhibitor, attenuates hypoxic pulmonary hypertension in rats. Am J Physiol 1999;277:L225-32.

40. Kodama K, Adachi H. Improvement of mortality by long-term E4010 treatment in monocrotaline-induced pulmonary hypertensive rats. J Pharmacol Exp Ther 1999;290:748-52.

41. Stamler JS, Loh E, Roddy MA, et al. Nitric oxide regulates basal systemic and pulmonary vascular resistance in healthy humans. Circulation 1994;89:2035-40.

42. Wallace AW, Tom WL. Interaction of L-arginine and phosphodiesterase inhibitors in vasodilation of the porcine internal mammary artery. Anesth Analg 2000;90:840-6.

Keywords:

Hypoxic pulmonary vasoconstriction; Phosphodiesterases; Siguazodan; Rolipram; Zaprinast; Nitric oxide; ATP-sensitive K+ channels; Voltage-gated K+ channels

© 2001 Lippincott Williams & Wilkins, Inc.