Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Pulmonary hypertension, whether occurring as a primary illness or secondary to a COPD, is increasing in prevalence (American Thoracic Society Board of Directors, 1995). The mechanisms responsible for the development and maintenance of hypoxia-induced pulmonary hypertension are still poorly understood. A defect in pulmonary vasodilatation may account, in part, for the maintenance of abnormal vascular tone seen with hypoxia-induced pulmonary hypertension. Perturbations of vasodilatory signal transduction in hypoxia-induced pulmonary hypertension have been attributed to: a) altered production of vasodilatory mediators (Badesch et al, 1989; Shaul et al, 1991, 1995; Su and Block, 1995); b) decreased expression of receptors for vasodilators (Shaul et al, 1990); c) decreased responsiveness to exogenous vasodilators which stimulate cyclic nucleotide production (Adnot et al, 1991; Wanstall and O'Donnell 1992); and d) impaired guanylyl cyclase activity (Crawley et al, 1992). Vasodilation may also be modulated by the rate of hydrolysis of vasodilatory second messengers (cAMP and cGMP). Cyclic nucleotide PDEs are responsible for the inactivation of these second messengers and play a significant role in modulating the amplitude and duration of vasodilatory stimuli (Kauffman et al, 1987). PDEs are also major mediators of "cross-talk" between vasodilatory and vasoconstrictive second messenger signaling pathways (Beavo, 1995; Conti et al, 1995). It has not been determined whether increased PDE expression and activity contribute to attenuated vasodilatory responses in HPH by accelerating hydrolysis of cAMP and cGMP.

There are currently seven partially characterized PDE families which are derived from at least 15 genes in the mammalian genome (Conti et al, 1995). Four of these families (1, 3, 4 and 5) are known to play a significant role in the regulation of vascular tone and may be important in the regulation of vascular responses to injury (Beavo 1995; Conti et al, 1995; Polson and Strada, 1996). Recent reports suggest that PDEs may be up-regulated after vascular graft preparation or vascular injury and that PDEs may modulate proliferation of vascular smooth muscle. In cultured vascular smooth muscle cells, exposure to hypoxia resulted in a time-dependent decrease in cAMP levels and a concomitant increase in PDE activity of the soluble fractions of PDE types 3 and 4 (Pinsky et al, 1993). In the rat aorta, balloon catheter angioplasty stimulated a biphasic increase in expression of the cAMP-specific PDE4B isoform (Smith et al, 1995). Furthermore, selective inhibitors of PDE3 and PDE4 significantly inhibited fetal calf serum-stimulated [³H]thymidine incorporation and proliferation of rat vascular smooth muscle cells (Pan et al, 1994; Polson and Strada, 1996). PDEs 1, 3, 4 and 5 have been identified in the pulmonary vasculature (Rabe et al, 1994; Polson and Strada, 1996; Dent et al, 1994). However, little is known about the role that PDEs play in regulating the magnitude and duration of the vasodilatory responses in hypertensive pulmonary vasculature. Accordingly, the purpose of this investigation was to determine whether PDEs play a significant functional role in regulating pulmonary vasodilation in HPH. The aims of the study were to determine the effects of PDE inhibition on PA relaxation and contraction in rats with HPH and the effect of HPH on PDE3 expression.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Induction of pulmonary hypertension. Pulmonary hypertension was established by exposing rats to a normobaric, hypoxic environment under the supervision of the laboratory animal veterinary staff as described previously (Griffith et al, 1994). Male Sprague-Dawley rats (225250 g) were randomly divided into control and HPH groups. Rats were made hypoxic gradually to minimize stress and loss of animals. Rats were placed in clear Lucite hypoxia chambers and exposed to 15% O₂ (FIO₂ 0.15) for 24 h. After 24 h of adaptation at 15% O₂, the rats were exposed to 10% O₂ (Fio₂ 0.1) for an additional 13 days. O₂ tension was measured daily by a Beckman O₂ analyzer (model C-2). Animals were maintained on a 12:12-h light-dark cycle and were given food and water ad libitum. Chambers were opened daily for 10 to 15 min. for cleaning and feeding. Control rats were maintained under similar conditions in the same room but were allowed to breath room air (21% O₂; FIO₂ 0.21).

Vessel preparation. Normal rats and rats with HPH were anesthetized with an intraperitoneal injection of pentobarbital sodium (50 mg/kg) (Sigma Chemical Co., St. Louis, MO) and exsanguinated by transection of the left renal artery. The heart and lungs were removed en bloc from the thoracic cavity and placed in ice-cold, oxygenated KHSS (115 mM NaCl, 25 mM NaHCO₃, 1.38 mM NaH₂PO₄, 2.51 mM KCl, 2.46 mM MgSO₄, 1.91 mM CaCl₂, 5.56 mM dextrose). (Individual components were obtained from Sigma.) The proximal right and left branches of the main PA were isolated, cleaned of all visible fat and connective tissue and cut into segments (2.5-3.5 mm in length) for use in tissue bath studies or were snap frozen in liquid nitrogen and stored at 70°C.

Measurement of contractile responses. Each arterial segment was suspended in a tissue bath by gently threading the ring onto a fixed, horizontal surgical steel wire (300 µm in diameter; 5 mm in length). Once anchored, a second wire of the same dimensions connected to a force transducer (Grass model FT 03C) was also threaded into the lumen of the ring. The tissue bath was filled with KHSS (37°C, pH 7.4) and gassed continuously with 95%O₂/5%CO₂. Isometric tension was recorded as a function of time on a chart recorder (Gould, model 2400). The arterial rings were stretched to optimal resting tension (PA-0.7g) for maximum active tension development and equilibrated for 1 h (Griffith et al, 1994). The rings were then contracted with KCl (80 mM) to establish the maximum active tension developed in response to membrane depolarization (P_o). The high potassium solution was washed out with KHSS, and the vessels were allowed to relax to base line for a 15-min equilibration period. This procedure was repeated twice before inclusion of a segment in an experiment. Vessel segments that failed to relax to base line were excluded from study. The maximum active tension developed by PA rings from normal rats and rats with HPH was not significantly different (Wagner, R. S., data not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Representative tracing of experimental protocol. PA rings were contracted with KCl (80 mM) to establish the maximum active tension developed in response to membrane depolarization (Po). The rings were then contracted with U46619 in the presence or absence of a PDE inhibitor. The contractile response to U46619 is expressed as a percent of Po. In subsequent experiments, relaxation responses are expressed as a percent of the U46619-induced contraction.

Northern blot analysis. Frozen PA tissue was ground to fine powder in a mortar and pestle pretreated with a 0.1% DEPC and prechilled with liquid N₂. Total RNA was isolated with TRI Reagent (Sigma). The concentration of total RNA recovered was determined with a Beckman, DU Series 640 spectrophotometer at 260 nm and 280 nm. After size separation of 15 to 20 µg of total RNA from each sample by gel electrophoresis on a 1% agarose-formaldehyde gel, total RNA was transferred to a nylon membrane (Fisher, Pittsburgh, PA) by capillary action and cross-linked to the membrane by UV irradiation (Stratagene, La Jolla, CA). The membranes were prehybridized at 42°C for 2 h in a hybridization solution containing deionized formamide (50%), 25 mM NaH₂PO₄ (pH 7.4), 20 mM EDTA, 375 mM NaCl, 5X Denhardt's, N-lauroyl sarcosine (3.2%), heparin (0.5 mg/ml) and 100 µg/ml salmon sperm DNA. Rat PDE3A (RcGIP2) cDNA and PDE3B cDNA (generously provided by Dr. V. Manganiello, National Institutes of Health) were labeled with [-³²P]dCTP by random priming (Ambion, Austin, TX) (Taira et al, 1993). Rat PDE4 A, B, C and D cDNAs (generously provided by Dr. Graeme B. Bolger, University of Utah) were labeled in a similar fashion. The labeled probe was added directly to the prehybridization solution and hybridization was continued for 16 to 18 h at 42°C. Membranes probed with PDE3 were washed once each in 2× SSC/1% SDS, 1× SSC/0.5% SDS and 0.5× SSC/0.25% SDS for 30 min at 60°C before autoradiography with X-OMAT film (Kodak) for 5 days at 70°C. Membranes were reprobed with ³²P-labeled cDNA for GAPDH (Gibco, Grand Island, NY) and washed as above at 55°C. Membranes probed with PDE4 were washed twice each in 2× SSC/0.1% SDS and then 0.1× SSC/0.1% SDS for 30 min at 65°C before autoradiography with X-OMAT film (Kodak) for 5 days at 70°C. Membranes were reprobed with ³²P-labeled cDNA for pT7 RNA 18S (Antisense Control Template; Ambion, Austin, TX) and washed as above at 55°C. Results were quantified by scanning densitometry, and target mRNA/GAPDH ratios were calculated.

Data analysis. Unless otherwise indicated, the data reported in the text are means ± S.E.M. Differences between control and experimental groups were analyzed by use of the unpaired, two-tailed Student's t test. The null hypothesis was rejected if P < .05.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of PDE inhibition on PA vasoconstriction. The effect of PDE inhibition on PA vasoconstriction was determined by contracting PA ring segments with U46619 (100 nM) in the presence or absence of PDE inhibitors. The contractile response to U46619 was expressed as a percent of P_o (fig. 1A). Milrinone (1 µM), rolipram (10 and 50 µM) and SB207499 (3 µM) had no effect on basal tension in PA rings from normal rats and rats with HPH (data not shown). Active tension development in response to U46619 (100 nM) did not differ between PA rings from normal rats (n = 18) and rats with HPH (n = 30) in the absence of PDE inhibitors (104 ± 7% and 98 ± 4% of P_o, respectively) (fig. 2). Treatment of PA rings from normal rats with milrinone (1 µM; n = 12) did not significantly affect vasoconstriction. However, treatment with rolipram [(10 µM; n = 7) or (50 µM; n = 3)] or SB207499 (3 µM; n = 3) significantly reduced (*P < .05) active tension development in response to U46619 (100 nM) to 75 ± 7%, 36.2 ± 11.1% and 40.7 ± 18.6% of P_o, respectively (fig. 2A). Treatment of PA rings from normal rats with combinations of mirinone and rolipram also significantly reduced (*P < .05) active tension development (n = 3) (fig. 2A). Treatment of PA rings from rats with HPH with milrinone (1 µM; n = 12) or rolipram (10 µM; n = 3) did not affect PA vasoconstriction. However, treatment of PA rings from rats with HPH with a higher concentration of rolipram (50 µM; n = 8) or SB207499 (3 µM; n = 3) did significantly reduce (*P < .05) active tension development to 72 ± 8% and 55.1 ± 13.4% of P_o (fig. 2B). Treatment of PA rings from rats with HPH with a combination of mirinone and rolipram also significantly reduced (*P < .05) active tension development (n = 3) (fig.2B).

View larger version (41K): [in this window] [in a new window]   Fig. 2.   Effect of PDE inhibitors on PA vasoconstriction. (A) Treatment of PA rings from normal rats with milrinone (1 µM; n = 12) did not significantly affect vasoconstriction. However, treatment with rolipram [(10 µM; n = 7) or (50 µM; n = 3)] or SB207499 (3 µM; n = 3) significantly reduced (*P < .05 compared with vehicle control; P < .05 compared with 10 µM rolipram) active tension development in response to U46619 (100 nM) (75 ± 7%, 36.2 ± 11.1% and 40.7 ± 18.6% of Po, respectively). Treatment of PA rings from normal rats with combinations of mirinone and rolipram also significantly reduced (*P < .05) active tension development (n = 3). (B) Treatment of PA rings from rats with HPH with milrinone (1 µM; n = 6) or rolipram (10 µM; n = 4) had no effect on active tension development in response to U46619 (100 nM). However, a higher concentration of rolipram (50 µM; n = 8) or SB207499 (3 µM; n = 3) significantly reduced (*P < .05) active tension development to U46619 (72 ± 7.5 and 55.1 ± 13.4 of Po, respectively). Treatment of PA rings from rats with HPH with combinations of mirinone and rolipram also significantly reduced (*P < .05 compared with vehicle control; *P < .05 compared with 1 µM milrinone; or 50 µM rolipram alone) active tension development (n = 3).

Effect of PDE inhibition on relaxation of PA rings from normal rats. PA ring segments from normal rats were contracted with U46619 in the presence or absence of PDE inhibitors, and cumulative concentration-response curves were generated for isoproterenol and forskolin. Isoproterenol induced concentration-dependent relaxation of precontracted PA rings from normal rats with an EC₅₀ of 259 ± 73 nM (n = 4). Treatment with either milrinone (1 µM) or rolipram (10 µM) significantly increased (*P < .05) isoproterenol-induced relaxation, which reduced the EC₅₀ for isoproterenol from 259 ± 73 nM to 26 ± 6 nM (n = 3) and 110 ± 14 nM (n = 4), respectively (fig. 3A). Forskolin also induced concentration-dependent relaxation of PA rings with an EC₅₀ of 234 ± 40 nM (n = 4). Treatment with either milrinone (1 µM) or rolipram (10 µM) significantly increased (*P < .05) forskolin-induced relaxation, which reduced the EC₅₀ for forskolin from 234 ± 40 nM to 40 ± 7 nM (n = 4) and 85 ± 8 nM (n = 4), respectively (fig. 3A).

View larger version (25K): [in this window] [in a new window]   Fig. 3.   PDE inhibition improves relaxation of PA rings from normal rats. (A) Treatment with either milrinone (1 µM) or rolipram (10 µM) significantly increased (*P < .05) isoproterenol-induced relaxation of PA rings from normal rats reducing the EC50 for isoproterenol from 259 ± 73 nM (n = 4) to 26 ± 6 nM (n = 3) and 110 ± 14 nM (n = 4), respectively. (B) Treatment with either PDE inhibitor also significantly increased (*P < .05) forskolin-induced relaxation reducing the EC50 for forskolin from 234 ± 40 nM (n = 4) to 40 ± 7 nM (n = 4) and 85 ± 8 nM (n = 4), respectively (fig. 2A).

Effect of PDE inhibition on relaxation of PA rings from rats with HPH. PA ring segments from rats with HPH were contracted with U46619 in the presence or absence of PDE inhibitors and cumulative concentration-response curves were generated for isoproterenol and forskolin. Isoproterenol induced concentration-dependent relaxation of precontracted PA rings from rats with HPH (fig. 4A). However, maximum relaxation was significantly greater (*P < .05; n = 4) in PA rings from normal rats than rats with HPH (43 ± 9% vs. 81 ± 5% of U46619-stimulated active tension development, respectively). Treatment of PA rings from rats with HPH with milrinone (1 µM) significantly improved (*P < .05; n = 4) maximum isoproterenol-induced relaxation (81 ± 5% of U46619-stimulated active tension development without milrinone versus 49 ± 4% with milrinone) (fig. 4A). Treatment of PA rings from rats with HPH with 10 µM rolipram had no effect on isoproterenol-induced relaxation at any concentration of isoproterenol tested (data not shown). However, treatment of PA rings from HPH rats with 50 µM rolipram significantly improved (*P < .05; n = 4) maximum isoproterenol-induced (500 nM) relaxation (81 ± 5% of U46619-stimulated active tension development without rolipram versus 57 ± 5% with rolipram) (fig. 4B).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   PDE inhibition improves beta adrenergic receptor-mediated relaxation of PA rings from rats with HPH. (A) Treatment of PA rings from rats with HPH with milrinone (1 µM) significantly improved (*P < .05; n = 4) maximum isoproterenol-induced relaxation from 81 ± 5% U46619 without milrinone to 49 ± 4% U46619 with milrinone. (B) Treatment of PA rings from HPH rats with 50 µM rolipram significantly improved (*P < .05; n = 4) maximum isoproterenol-induced (500 nM) relaxation from 81 ± 5% U46619 without rolipram to 57 ± 5% U46619 with rolipram.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   PDE inhibition improves forskolin-mediated relaxation of PA rings from rats with HPH. (A) Treatment of PA rings from rats with HPH with milrinone (1 µM) significantly (*P < .05; n = 4) lowered the EC50 for forskolin-induced relaxation from 548 ± 112 nM in rings without milrinone to 58 ± 14 nM. (B) Treatment of PA rings from rats with HPH with rolipram (50 µM) significantly (*P < .05; n = 4) lowered the EC50 for forskolin-induced relaxation from 548 ± 112 nM in rings without rolipram to 96 ± 25 nM.

Effect of dbu-cAMP on relaxation of PA rings from normal rats and rats with HPH. The cell-permeable cAMP analog, dbu-cAMP (100 µM), induced relaxation of precontracted PA rings from normal rats and rats with HPH (fig. 6). At 10 min, the degree of relaxation stimulated by dbu-cAMP was significantly (*P < .05; n = 4) greater in PA rings from normal rats than that in PA rings from rats with HPH. Although the T_1/2 was significantly shorter (*P < .05; n = 4) in PA rings of normal rats than that in PA rings from rats with HPH (8.1 ± 2.1 min and 15.9 ± 2.2 min, respectively), the maximum relaxation of the precontracted rings was identical in PA rings from normal rats and rats with HPH. A second cAMP analog, 8-Br-cAMP, did not induce relaxation of precontracted PA rings from rats with HPH (n = 4).

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Time course of dbu-cAMP-induced relaxation. PA rings from normal rats and rats with HPH were contracted with U46619 (100 nM) before stimulation with dbu-cAMP (100 µM) or 8-Br-cAMP (100 µM). At 10 min, the degree of relaxation stimulated by dbu-cAMP was significantly (*P < .05; n = 4) greater in PA rings from normal rats than that in PA rings from rats with HPH. However, the degree of relaxation was not significantly different in normal and HPH PA rings at 30 and 60 min post-stimulation (n = 4). 8-Br-cAMP did not stimulate relaxation in PA rings from rats with HPH (n = 4).

PDE expression is altered in PA from rats with HPH. Total RNA isolated from PA of normal and HPH rats was probed with ³²P-labeled cDNA for PDE3A (RcGIP2), PDE3B, PDE4A, PDE4B, PDE4C and PDE4D. Transcripts for PDE3B, PDE4A, PDE4C and PDE4D were not detected. In three separate experiments, PDE3A 8-kb and 10-kb mRNA transcripts were significantly (*P < .05; n = 3) increased in PAs from rats with HPH (3.8 ± 1.6-fold and 3.9 ± 1.2-fold, respectively) (figs. 7and 9). Northern analysis of PDE4B expression suggested that there is a 30% reduction in expression of the 4-kb mRNA transcript, but the difference was not statistically significant (P = .06; n = 3) (figs. 8 and 9).

View larger version (39K): [in this window] [in a new window]   Fig. 7.   Expression of PDE3A is increased in PA from rats with HPH. (A) Lane 1, control PA-RNA; lane 2, PA-RNA from rats with HPH. Intervening lanes were removed for clarity. Northern analysis of PDE3A expression demonstrated a 3.8 ± 1.6-fold and 3.9 ± 1.2-fold increase in 8-kb and 10-kb mRNA transcripts, respectively. (Representative of three separate Northern blots. RNA from 2 PAs was pooled in each lane.) (B) After probing the blot for PDE3A, the blot was reprobed with GAPDH to evaluate lane loading. To quantitate expression of PDE3A mRNA transcripts, densitometry was used to determine the ratio of PDE mRNA transcripts to GAPDH mRNA transcripts. (C) Photograph of RNA gel showing quality of RNA and equality of RNA loading.

View larger version (34K): [in this window] [in a new window]   Fig. 9.   Analysis of fold change in PDE expression. PDE expression in PAs from rats with HPH was normalized to PDE expression in PAs from normal rats. Northern analysis of PDE3A expression demonstrated a 3.8 ± 1.6-fold and 3.9 ± 1.2-fold increase in 8-kb and 10-kb mRNA transcripts, respectively. Although expression of PDE4B appears to be decreased, the difference from normal did not reach statistical significance.

View larger version (48K): [in this window] [in a new window]   Fig. 8.   Expression of PDE4B may be decreased in PA from rats with HPH. (A) Lane 1, control PA-RNA; lane 2, PA-RNA from rats with HPH; lane 3, RNA from the right ventricles of control rats. Northern analysis of PDE4B expression suggests that there is a 30% reduction in expression of the 4-kb mRNA transcript but the difference was not statistically significant (P = .06; n = 3; RNA from 3 PAs was pooled in each lane.) RNA from the right ventricles of control rats was included as a positive control. (B) After probing the blot for PDE4B, the blot was reprobed with ribosomal 18S to evaluate lane loading. To quantitate expression of PDE4B mRNA transcripts, densitometry was used to determine the ratio of PDE mRNA transcripts to ribosomal mRNA 18S transcripts. (C) Photograph of RNA gel showing quality of RNA and equality of RNA loading.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The purpose of this investigation was to determine whether PDEs play a significant functional role in regulating pulmonary artery contraction and relaxation in rats with HPH. We found that inhibition of PDE4 activity with rolipram (10 and 50 µM) or SB207499 (3 µM) in PAs from normal rats significantly reduced the magnitude of the contractile response to U46619, but that inhibition of PDE3 with milrinone (1 µM) did not. Combinations of milrinone (1 µM) and rolipram (10 or 50 µM) significantly reduced the magnitude of the contractile response to U46619 in PAs from normal rats, but the degree of the reduction was not significantly different from that achieved with rolipram (10 or 50 µM) alone. Neither milrinone (1 µM) nor rolipram (10 µM) had an effect on the contractile response to U46619 in PAs from rats with HPH. However, inhibition of PDE4 activity with 50 µM rolipram significantly reduced the magnitude of the contractile response to U46619 in PAs from rats with HPH but to a lesser degree than in PAs from normal rats. Inhibition of PDE4 activity with SB207499 (3 µM) also significantly reduced the magnitude of the contractile response to U46619 in PAs from rats with HPH. A combination of milrinone (1 µM) and rolipram (50 µM) significantly reduced the magnitude of the contractile response to U46619 in PAs from rats with HPH to a greater degree than that achieved with rolipram (50 µM) alone. This suggests that in PAs from rats with HPH, this combination of PDE3 and PDE4 inhibitors may synergistically inhibit PDE activity in a manner similar to that reviewed by Polson and Strada (1996). Cumulatively, these data suggest that both PDE3 and PDE4 modulate contractile responses of PAs from rats with HPH.

In addition to investigating the effect of PDE inhibition on PA contraction, we also investigated the effect of PDE inhibition on PA relaxation. Isoproterenol and forskolin stimulated concentration-dependent relaxation of precontracted PAs from normal rats and rats with HPH. The data confirm that beta adrenergic receptor-mediated relaxation is significantly attenuated in PA rings from rats with HPH and also show that forskolin-mediated relaxation is significantly attenuated. This finding was somewhat surprising because forskolin is a direct activator of adenylyl cyclase and Shaul et al. (1990) have shown that GTP- and sodium fluoride-stimulated adenylyl cyclase activity is not altered by HPH. In light of normal adenylyl cyclase activity with HPH, the most probable explanations for attenuated relaxation to forskolin are: 1) increased hydrolysis of cAMP or 2) impaired signal transduction distal to cAMP formation and hydrolysis. Indeed, we found that inhibition of either PDE3 or PDE4 activity significantly improved both isoproterenol- and forskolin-induced relaxation in PA rings from rats with HPH. These data suggest: 1) that increased hydrolysis of cAMP may be responsible, in part, for impaired PA vasodilation in rats with HPH and 2) that both PDE3 and PDE4 modulate relaxation of PAs from normal rats and rats with HPH.

The degree of relaxation achieved by PA rings from rats with HPH in the presence of a PDE inhibitor generally did not equal that obtained by PA rings from normal rats under similar conditions. Maximum relaxation stimulated by isoproterenol in PA rings from rats with HPH did not equal that achieved by PA rings from normal rats in the presence of PDE3 or PDE4 inhibitors even when a higher concentration of rolipram was used to treat PA rings from rats with HPH. This is likely caused partly by decreased expression of beta adrenergic receptors in PAs of rats with HPH (Shaul et al, 1990) but may be complicated by increased PDE activity. Even with beta adrenergic receptor down-regulation, the degree of maximum relaxation stimulated by isoproterenol in the presence of PDE3 and PDE4 inhibitors in PAs from rats with HPH approached that of isoproterenol-stimulated PAs from normal rats in the absence of PDE inhibitors. These data suggest that PDE inhibition may enhance the vasodilatory activity of beta adrenergic receptor agonists.

When PAs were stimulated with forskolin in the presence of 10 µM rolipram, relaxation was significantly improved in PA rings from normal rats but was not affected in PA rings from rats with HPH. However, increasing the concentration of rolipram to 50 µM did significantly improve forskolin-stimulated relaxation of PA rings from rats with HPH. It is possible that at this concentration, rolipram may have also inhibited PDE3 activity. We were unable to find an IC₅₀ for the inhibition of PDE4 activity by rolipram in rat pulmonary arteries. Komas et al. (1991) did publish an IC₅₀ of 1.0 ± 0.2 µM for rolipram inhibition of PDE4 purified from rat aorta. In the same study, rolipram at concentrations greater than 200 µM did not inhibit purified PDE3, which suggests that rolipram is selective for PDE4 at a concentration of 50 µM. Because a higher concentration of rolipram was required to inhibit PDE4 activity in PA rings from rats with HPH, it is highly probable that PDE4 activity is increased in these arteries.

Although the data from this study strongly suggest that PDE activity is increased in PA rings from rats with HPH, the possibility still exists that impaired signal transduction distal to cAMP formation and hydrolysis could, in part, account for attenuated relaxation of PA rings from rats with HPH. Therefore, we investigated the effect of cAMP analogs on relaxation of PA rings from normal rats and rats with HPH. The stable cAMP analog, dbu-cAMP, relaxed precontracted PA rings from normal rats and rats with HPH. The magnitude of relaxation achieved at 10 min was significantly greater in PAs from normal rats than PA rings from rats with HPH, but at 30 and 60 min poststimulation, the magnitude of relaxation did not differ. Additionally, the rate of initial relaxation induced by dbu-cAMP was faster in PA rings from normal rats. The difference in the degree of relaxation seen at 10 min and the longer T_1/2 for relaxation of PAs from rats with HPH may reflect slower dbu-cAMP penetration into thicker, hypertrophied PA rings from rats with HPH or impaired activity in signal transduction pathways activated by cAMP or its analog. In contrast to the effect of dbu-cAMP, 8-Br-cAMP, a relatively selective activator of PKA, did not relax precontracted PA rings from rats with HPH. Several potential explanations exist for this discrepancy. It is possible that cAMP-mediated vasodilation is not mediated entirely through activation of PKA. The results of our relaxation studies are very similar to the findings of MacDonald and Diamond (1994) in the rat aorta. Their data demonstrated that although 8-Br-cAMP activated PKA, it did not stimulate relaxation. In addition, they found that dbu-cAMP did not activate PKA but did stimulate relaxation. Their data suggest that the relaxation stimulated by dbu-cAMP resulted not from stimulation of PKA, but from stimulation of other signal transduction pathways. Indeed, it has been shown that cAMP may also stimulate relaxation by activation of cGMP-dependent kinase or indirect gating of non-ATP-sensitive K⁺ channels (Lincoln et al, 1990; Haynes et al, 1992). These findings suggest that elevating intracellular levels of cAMP in PAs from both normal rats and rats with HPH via inhibition of PDE3 or PDE4 activity may improve relaxation through cAMP-dependent activation of multiple signal transduction pathways.

The findings from our studies suggest that cAMP-PDEs (PDE3 and PDE4) play a significant role in regulating pulmonary artery contraction and relaxation and are linked to the pathophysiology of altered tone in HPH. However, little is known about how chronic hypoxia and/or pulmonary hypertension affect PDE expression and activity. Therefore, we evaluated PDE3 and PDE4 expression in PAs from normal and HPH rats. We found that PDE3A expression is actively up-regulated during HPH whereas PDE4B expression may be down-regulated during HPH. A complimentary study found that cAMP-PDE activity is significantly increased in the proximal and intrapulmonary branches of PAs from rats with HPH (Johnston et al, 1996). This suggests that increased expression of PDE3A may correlate with increased PDE activity. However, because it appears that expression of PDE4B may be decreased whereas our pharmacological data suggest that its activity is increased, in future studies it will be necessary to measure PDE activity.

In summary we have shown that: 1) inhibition of PDE3 and PDE4 significantly improves beta adrenergic receptor-mediated and forskolin-mediated relaxation of PA rings normal rats and from rats with HPH; 2) the cell-permeable cAMP analog, dbu-cAMP, induces a similar degree of relaxation in PA rings from normal and HPH rats; and 3) expression of PDE3A mRNA is increased during HPH. These findings strongly suggest that PDEs play an important role in the development and maintenance of HPH. Many diseases such as neonatal persistent pulmonary hypertension, congenital heart disease, adult respiratory distress syndrome, neuromuscular disorders and COPD are complicated by pulmonary hypertension (Golan et al, 1995; Higenbottam and Cremona, 1993; Kinsella and Abman, 1995; Vender, 1994). Vasodilators which stimulate increases in intracellular levels of cAMP are currently used to treat patients with pulmonary hypertension. Unfortunately, the effects of this type of therapy are often short-lived and responses are highly variable (Lunn, 1995; Cremona and Higenbottam, 1995; Alpert et al, 1994). The data from our study suggest that vasodilator therapy may be unsuccessful, in part, because of increased PDE activity. Therefore, it may be beneficial to consider developing therapeutic strategies which incorporate the use of PDE inhibitors into current treatment regimes to potentiate the beneficial effects of vasodilators and prolong their duration of action.