American Journal of Respiratory Cell and Molecular Biology

Primary pulmonary hypertension is characterized by increased pulmonary vascular resistance and smooth muscle proliferation. Stable analogs are increasingly being used to treat this disease, although no data exists comparing their effects on proliferation. We therefore investigated the antiproliferative activity of several prostacyclin (PGI2) analogs on human pulmonary arterial smooth muscle cells, including UT-15 and iloprost, analogs that have recently completed successful clinical trials. Serum-induced proliferation, as assessed by [3H]thymidine incorporation (30 h) or cell number (48 h), was significantly inhibited with a 10-fold difference in potency, ranking in effectiveness UT-15 > iloprost > cicaprost > beraprost. Effects were reversed by the adenylyl cyclase inhibitor, 2,5 ′ dideoxyadenosine (DDA) but not SQ22536. Intracellular cyclic AMP (cAMP) was elevated by all analogs and inhibited by DDA, although SQ22536 was a highly variable inhibitor, suggesting that different pathways might mediate cAMP generation. UT-15 produced a significantly larger and more sustained increase in cAMP compared with other analogs, with iloprost being the weakest elevator. Thus, PGI2 analogs potently inhibit proliferation of human pulmonary artery, probably via a cAMP-dependent pathway, although cAMP elevation in itself is not a good predictor of antiproliferative potency.

Primary pulmonary hypertension (PPH) is a progressive and fatal cardiovascular disorder and is characterized by a sustained elevation of pulmonary artery pressure that cannot be attributed to any demonstrable cause (1). Increased pulmonary vascular resistance probably results from a combination of pulmonary vasoconstriction, vascular-wall remodelling (intimal and smooth muscle proliferation), and thrombosis (1). Although the underlying trigger for these changes remains unknown, patients with PPH have enhanced levels of the potent vasoconstrictor and growth promoting agents, thromboxane and endothelin, together with impaired synthesis of the vasodilatory and antiproliferative substances, prostacyclin (PGI2) and nitric oxide (2, 3). Thus an imbalance in production of these factors may contribute to the pathogenesis of PPH.

Much of the rationale for the treatment of PPH has been based around the premise that vasoconstriction is a major underlying feature of the disease. Despite this, less than 30% of patients respond to vasodilator oral therapy with improved lung function (1). On the other hand, long-term continuous infusion of PGI2 has been shown to significantly improve hemodynamics and long-term survival in a much wider population of patients, even those demonstrating no vasodilatory response to acute infusion (4, 5). Thus other factors, such as effects of PGI2 on smooth muscle cell proliferation and platelet aggregation, are likely to mediate clinical improvement (4, 5). Because the half-life of prostacyclin is short (∼ 2–3 min), a number of stable analogs have been developed with extended half-lives and increased bioavailabilities, including beraprost, cicaprost, and iloprost (6, 7). Both iloprost and beraprost appear beneficial in the treatment of PPH and data suggest that they may offer some advantages over PGI2 infusion (8-10). Recently, preliminary results have been reported suggesting that UT-15, a novel PGI2 analog, produces favorable effects in pulmonary artery hypertension unresponsive to conventional medical therapy (11).

Like PGI2, stable analogs are potent inhibitors of platelet aggregation and produce profound vasodilation in vivo and relaxation of pulmonary arteries ex vivo (6, 7, 12). In a limited number of studies, PGI2 analogs have been shown to inhibit coronary and aortic smooth muscle proliferation in culture and to inhibit neointimal formation in rat balloon-injured arteries (13-16). In addition, both iloprost and cicaprost have been shown to inhibit cell growth stimulated by platelet-derived growth factor in distal pulmonary artery, though cicaprost was reported to have little effect on growth in proximal cells (17). Although the mechanism of action on cell proliferation is not well understood, several lines of evidence suggest that the major biologic activities of PGI2 and its analogs are mediated by cell surface prostanoid (IP) receptors (6). These receptors are known to couple to adenylyl cyclase via the G-protein, Gs (7). This is consistent with the observation that IP receptor agonists increase cAMP in many cell types, including smooth muscle (6, 18), and that the effects of PGI2 analogs can be potentiated by agents that prevent the breakdown of cAMP (19, 20). Thus, it has been proposed that cAMP is the main mediator of PGI2-induced effects in vascular smooth muscle (6). However, recent data from our laboratory challenge this view, because cAMP generation, but not relaxation, elicited by iloprost was reversed by adenylyl cyclase inhibition (18). Moreover, IP receptors can couple to multiple G-protein pathways and a number of PGI2 analogs have been reported to potently activate prostanoid EP receptor subtypes (7).

In view of the obvious clinical importance of these analogs, we sought to compare the effect of these analogs on proliferation in cultured human pulmonary artery smooth muscle cells (HPASMC), because differences in the ability of agents to inhibit proliferation may have a bearing on clinical efficacy. We investigated the effects of the novel PGI2 analog, UT-15 (Remodulin) for which no in vitro data exists, and compared it with iloprost, cicaprost, and beraprost. Experiments were also designed to assess the role of adenylate cyclase in mediating the antiproliferative effects of these agents.

Cell Culture

Primary cultures of human pulmonary artery (1st and 2nd order) smooth muscle cells (HPASMC) were obtained from Clonetics (San Diego, CA) at the 4th passage. Cells were grown in smooth muscle cell growth medium (SmGM-2; Clonetics) supplemented with fetal bovine serum (FBS) (5%), gentamycin (50 μg/ml), human fibroblast growth factor (2 ng/ml), human EGF (0.5 ng/ml), and insulin (5 μg/ml) at 37°C in a humidified atmosphere of 5% CO2. After reaching confluence, cells were washed with PBS (Gibco, Paisley, UK) and treated with trypsin-EDTA (Gibco) for further passage. Cells between passages 8 and 10 were used for all experiments. The smooth muscle phenotype was confirmed immunohistochemically using a LSAB+ kit (DAKO, Ely, UK) and a mouse anti-human α-actin monoclonal antibody (Roche, Lewes, UK). The presence of endothelial cells was excluded using a mouse anti-human antibody to CD31.

Cell Counting

HPASMC were seeded onto six-well plates at a density of ∼ 2 × 104 cells/ml and grown in supplemented SmGM-2 for 24 h. After 24 h, the media was replaced with SmGM-2, and cells starved in growth factor-free medium for 48 h. Following this, the medium was replaced with fresh SmGM-2 containing 10% FBS or platelet derived growth factor (PDGF) and the relevant test agent. After 48 h, cells were washed and incubated with trypsin-EDTA for 2.5 min before being neutralized with SmGM-2 containing 10% FBS. Cells were centrifuged and resuspended in 0.5–1 ml of media to which trypan blue (0.4%) had been added. Viable (nonstained) cells were counted blindly using a hemocytometer by a person not knowing the treatment protocol. Results are expressed either as actual cell number or the percentage inhibition of growth at 48 h in the presence of serum.

[3H]thymidine Uptake

Relative rates of DNA synthesis were assessed by determination of [3H]thymidine incorporation into trichloroacetic acid (TCA) precipitable material. As above, cells were grown in supplemented SmGM-2 for 24 h and starved for a further 48 h, after which the medium was replaced with SmGM-2 containing 1 μCi/ml methyl- [3H]thymidine (Amersham Life Science, Amersham, UK) and 1 μM unlabeled thymidine, either alone or supplemented with 10% FBS and the appropriate test drugs. Cultures were incubated for 30 h in the thymidine solution, after which the medium was removed, the cells washed with ice-cold PBS and then incubated with 5% TCA for 1 h at 4°C to precipitate the DNA. After 1 h, cells were washed with 5% TCA followed by 75% ethanol and samples left to dry at room temperature. The precipitated material was dissolved in a buffer containing 1% sodium dodecyl sulfate, 0.1 M NaOH, 2% Na2CO3, and samples diluted in scintillation fluid for counting. Results were expressed either as actual [3H]thymidine incorporation (counts per minute) or percentage inhibition of incorporation stimulated by 10% FBS.

Concentration and Time-Response for cAMP

Cells were plated in six-well plates and grown to 80% confluence. For concentration–response curves, washed cells were stimulated for 15 min with SmGM-2 containing a specific concentration of the PGI2 analog. In some experiments, the medium also contained the adenylyl cyclase inhibitors, 2,5′ dideoxyadenosine (DDA, 100 μM) or SQ22536 (100 μM), in which case cells were pretreated for 15 min with these inhibitors. For the time-response, cells were stimulated with 30 nM of the analog from 5 min to 72 h. At the end of each experiment, cells were washed and lysed with ice-cold ethanol (pH 3.4) containing 0.5 mM of 3-isobutyl-1-methylxanthine (IBMX). The lysate was centrifuged (10 min at 13,000 rpm) to pellet tissue fragments and the supernatant collected. The pellet was dissolved in 0.05 M Tris buffer (pH 7.5) containing 4 mM EDTA and proteins determined using a Biorad (Munich, Germany) assay kit, as per the manufacturer's instructions. The ethanol in the supernatant was evaporated at 55°C under a flow of N2 and the residue resuspended in Tris/EDTA buffer. Samples were stored at −20°C for up to 2 wk. Levels of cAMP were measured using a competitive radioreceptor binding assay kit (Biotrak) obtained from Amersham.

Transient Transfection of the Human IP Receptor

Human embryonic kidney (HEK-293) cells were cultured in MEM, supplemented with l-glutamine (2 mM), 10% FBS, penicillin (100 U/ ml), and streptomycin (10 μg/ml) at 37°C in a humidified atmosphere of 5% CO2. Cells were transfected with the human IP receptor using lipofectamine (Life Technologies, Inc.) according to the manufacturer's instructions. The human IP receptor clone was excised from pBluescript with a KpnI and XbaI restriction digest and subcloned into pcDNA3.1Zeo(x) using standard molecular cloning techniques. Transfected and nontransfected cells were washed with PBS and exposed to fresh medium for 15 min in the absence and presence of 30 nM UT-15. The cAMP was extracted and assayed as above.


All drugs were made up in the appropriate culture medium with the exception of DDA, which was made as a 100 mM stock in dimethlylsulphoxide and then diluted in medium. DDA was obtained from Affiniti Research Products Ltd (Exeter, Devon, UK), IBMX and PDGF from Sigma Chemical Co. (Poole, Dorset, UK). SQ22536 was synthesized by the Department of Medicinal Chemistry, the Wolfson Institute, (UCL, London, UK). Iloprost and cicaprost were kindly donated by Schering AG (Berlin, Germany), beraprost sodium (sodium [±]-[1R*,2R*,3aS*,8bS*]2,3,3a,8b-tetrahydro-2-hydroxy-1 ([E]-[3S*]-3hyroxy-4-methyl-1-octen-6-yny1)-1H-cyclopenta[b] benzofuran-5-butyrate) by Toray Industries, Inc. (Tokyo, Japan) and UT-15 ([1R,2R,3aS,9aS]-[[2,3, 3a,4,9,9a-hexahydro-2-hydroxy-1- [(3)-3-hydroxyoctyl]-1H-benz [f] inden-5-yl]oxy]acetic acid) by United Therapeutics (Washington, MD).


Each experiment was repeated at least five times and results expressed as mean ± SEM of n observations. The concentration of agent causing 50% of the maximal cAMP elevation or inhibition of cell proliferation (EC50) was calculated using the sigmoidal-curve fitting routine in Origin (MicroCal, Northampton, MA). Statistical analysis was performed using the program SigmaStat (Jandal, San Rafael, CA). Significance between a single treatment was assessed using a t test. For comparisons between groups, one- or two-way analysis of variance (ANOVA) was used and where appropriate, correction for multiple comparisons was performed. A P value < 0.05 was considered significant.

Effect of UT-15 on Cell Growth: Comparison with Other Prostacyclin Analogs

The effects of UT-15 on proliferation stimulated by FBS (10%) was investigated in HPASMC. Under control conditions, FBS increased cell number (P < 0.001) by 70% from 3.66 ± 0.15 × 104 to 6.2 ± 0.2 × 104 cells/ml (n = 6), measured at 48 h in the absence or presence of FBS, respectively (Figure 1A). Incubation with UT-15 (30 nM) significantly inhibited HPASMC proliferation stimulated by FBS, reducing cell numbers close to unstimulated levels (93% reduction in cell number increase, P < 0.001 one-way ANOVA, n = 6). Similarly, UT-15 inhibited DNA synthesis (another index of proliferative activity), reducing [3H]thymidine incorporation by 53% from 5628 ± 924 to 2613 ± 122 cpm (n = 6, P < 0.01; Figure 1B). When data from several different experiments were pooled, UT-15 consistently caused a profound reduction in FBS-stimulated HPASMC proliferation as measured by cell counting (91.6 ± 2.1%, n = 50; P < 0.001) and [3H]thymidine incorporation (60.5 ± 3.9%, n = 28; P < 0.001).

In initial experiments, the antiproliferative activity of UT-15 was compared with other PGI2 analogs under the same experimental conditions using the same batch of cells. Like UT-15, iloprost and cicaprost significantly reduced proliferation of HPASMC stimulated by FBS. All three agonists displayed marked antiproliferative activity at 30 nM as assessed by blinded cell counting. Growth was 16.1 ± 2.4%, 27.6 ± 2.6%, and 43.9 ± 5.7% (n = 12) of that observed with FBS alone for UT-15, iloprost, and cicaprost, respectively (Figure 2B). Similar results were obtained with [3H]thymidine incorporation, where all three agonists reduced [3H]thymidine uptake compared with FBS-stimulated control (n = 12, P < 0.01). In both assays, cicaprost was significantly (P < 0.01) less effective than either iloprost or UT-15 at inhibiting cell growth. Furthermore, iloprost was less effective than UT-15 at inhibiting cell numbers (P < 0.05), though no significant difference was observed when comparing results obtained in the [3H]thymidine assay (Figure 2C). In another series of experiments, the concentration– responses of the various PGI2 analogs on cell proliferation of HPASMC were assessed. UT-15, iloprost, cicaprost, and beraprost all dose-dependently inhibited proliferation with an EC50 of 4.2 nM, 21.0 nM, 24.1 nM, and 40.0 nM, with UT-15 being significantly more potent (P < 0.01, n = 5) at inhibiting cell proliferation than either of the other three analogs, and produced maximal inhibition at a 100-fold lower (0.1 μM) concentration. Cell growth was essentially abolished by UT-15 and iloprost, albeit at different concentrations, whereas maximal inhibition (84%) elicited by cicaprost and beraprost was significantly (P < 0.05) less than that observed with either iloprost or UT-15 (Figure 2D).

Profile of Intracellular cAMP Changes Induced by Prostacyclin Analogs

To confirm that UT-15 acts as an agonist at the IP receptor, cAMP levels were measured in wild-type HEK-293 cells and HEK-293 cells transiently transfected with the human IP receptor. In the presence of 30 nM UT-15, there was a significant elevation of cAMP over basal values (∼ 112-fold, n = 6; P < 0.0001) in transiently transfected cells but no increase was observed in wild-type cells (Figure 3). These results therefore demonstrate that UT-15 is a potent stimulator of the IP receptor.

To assess whether differences in the antiproliferative activity of these agonists related to their ability to elevate intracellular cAMP, the concentration–response and time course to cAMP induced by the PGI2 analogs was examined in HPASMC. When concentration–response curves to intracellular cAMP were constructed in HPASMC, there were noticeable differences in the efficacy and potency of these agents (Figure 4A). UT-15 had a significantly higher efficacy (P < 0.001) than the other three analogs, maximally increasing cAMP by 200-fold from a basal level of 1.7 ± 0.12 to 365.2 ± 29.7 pmol/mg protein at 1 μM (n = 12), whereas iloprost had the lowest efficacy, only increasing cAMP to 58.7 ± 4.7 pmol/mg protein (n = 6) at 1 μM. However, the EC50 for UT-15, cicaprost, and iloprost were not significantly different, being 8.2 nM (95% confidence limits 4.2 to 16.3 nM), 7.1 nM (95% confidence limits 5.6 to 8.9 nM), and 4.8 nM (95% confidence limits 2.7 to 8.7 nM), respectively. In contrast, the EC50 for beraprost was 100-fold higher, being 98.2 nM (95% confidence limits 89.5 to 107.7 nM).

The time course of these rises in cAMP was assessed for UT-15, iloprost, and cicaprost. UT-15 produced an elevation in intracellular cAMP which peaked at ∼ 15 min. (Figure 4B). Intracellular cAMP levels remained significantly elevated compared with basal at 6 h after drug treatment (27-fold, n = 12; P < 0.05) and between 48 and 72 h cAMP was still elevated by 4- to 5-fold. Both cicaprost and iloprost elevated intracellular cAMP concentration, though levels only remained significantly elevated over basal for up to 2 h after treatment with either agent (19-fold and 9-fold, respectively; n = 6, P < 0.05). Again, the magnitude of the increase induced by either iloprost or cicaprost was less than that observed with UT-15 over all time points up to and including 6 h (P < 0.02).

Effect of Adenylate Cyclase Inhibitors on cAMP Accumulation and Cell Proliferation

While nonhydrolysable analogs of cAMP have been shown to inhibit smooth muscle cell growth, a direct role for the cAMP pathway in mediating the antiproliferative effects of PGI2 analogs has not been examined to date. We therefore investigated the effects of two P-site adenylate cyclase inhibitors, namely, DDA and 9-(tetrahydro-2-furanyl)- 9H-purin-6-amine (SQ22536) on growth in HPASMC. Pretreatment of cells with DDA (100 μM) substantially inhibited the effects of UT-15 on cell number by 52% (P < 0.01, n = 14; Figure 5D) and [3H]thymidine incorporation by 42% (P < 0.001, n = 5; Figure 5B). In contrast, SQ22536 (100 μM) did not significantly reverse the antiproliferative effects of UT-15 in either cell growth assay. Furthermore, a combination of both adenylate cyclase inhibitors did not produce a greater inhibition of the response to UT-15 than that observed by DDA alone. In other experiments, cAMP elevation induced by UT-15 was substantially inhibited at all concentrations by 100 μM DDA, the maximal response being reduced by 94% from 448.0 ± 21.9 to 26.0 ± 3.8 pmol/mg protein (P < 0.001, n = 6; Figure 5D). SQ22536 (100 μM) also significantly inhibited the rise in cAMP elicited by UT-15, reducing the maximal response by 77% from 315.0 ± 8.7 to 72.4 ± 3.6 pmol/mg protein. However, the maximal levels of cAMP induced by UT-15 in the presence of DDA (26.0 ± 3.8 pmol/mg protein) were significantly less (P < 0.001, n = 6) than that observed with SQ22536.

To assess whether the mode of action UT-15 was similar against a single growth factor, experiments were performed in which HPASMC were stimulated with PDGF. UT-15 essentially abolished proliferation stimulated by PDGF (Figure 6), as measured by cell counting (94% reduction in cell number increase, P < 0.001, n = 5), and [3H]thymidine incorporation (66% reduction of incorporation, P < 0.001, n = 5). Furthermore, DDA substantially reversed the effects of UT-15 on cell number (79%, P < 0.001, n = 5), and [3H]thymidine incorporation (75%, P < 0.001, n = 5), whereas SQ22536 had no observable effect on UT-15–mediated antiproliferative activity.

The effects of the adenylate cyclase inhibitors were also studied against iloprost. SQ22536 produced a small (though not significant) reversal of the iloprost-induced antiproliferative activity, eliciting a 10% increase in cell number and a 15% increase in [3H]thymidine incorporation (Figures 7A and 7B). Much larger effects were observed with DDA, which significantly (P < 0.001, n = 5) inhibited the antiproliferative effects of iloprost to a similar degree as that observed with UT-15. However, unlike experiments with UT-15, a greater inhibition (P < 0.05) of the iloprost response on cell numbers was achieved with a combination of DDA and SQ22536. Furthermore, DDA significantly inhibited the elevation in cAMP mediated by both iloprost (69%, P < 0.001, n = 6) (Figure 7D) and cicaprost (75%, P < 0.001, n = 6) (Figure 7F). In contrast to its marked inhibitory effects on UT-15–induced cAMP elevation, SQ22536 had only a modest inhibitory action on cAMP levels increased by iloprost, though it should be noted that the scale of cAMP generation is vastly different. SQ22536 reduced the maximal response by only 27% (P < 0.001, n = 6) to 42.8 ± 3.4 pmol/mg protein (Figure 7C). However, this inhibitor had little or no effect on cAMP elevation induced by cicaprost (Figure 7E).

In the present study, we investigated the antiproliferative effects of iloprost, beraprost, and UT-15 in human pulmonary artery, three stable PGI2 analogs currently being used clinically to treat PPH, and compared effects with cicaprost, the most selective IP agonist available (7). These analogs produced substantial inhibition of FBS-stimulated cell proliferation and DNA synthesis in HPASMC in the nanomolar concentration range, and dose-dependently increased intracellular cAMP, though the correlation between cAMP generation and inhibition of cell growth was not clear. However, the adenylyl cyclase inhibitor DDA did blunt effects on proliferation, DNA synthesis, and intracellular cAMP generation by ∼ 75%, with a trend toward a greater inhibition in the presence of both adenylate cyclase inhibitors, which was significant for iloprost (Figure 7). Thus, our results strongly suggest that cAMP is the main mediator of the antiproliferative effects of these stable analogs in human pulmonary artery. Similar conclusions were made recently, where the antiproliferative effects of cicaprost in distal HPASMC were partially reversed by DDA (∼ 40%) (17). Our results are also consistent with the well-documented growth inhibitory effects of cAMP observed in many cell types, including vascular smooth muscle (17, 21, 22) and also with the observation that antiproliferative activity of the stable PGI2 analog U-61,431F can be potentiated by a specific inhibitor of the cAMP-regulated phosphodiesterase in bovine aorta (20). Moreover, the endogenous prostaglandins PGI2, PGE1, and PGE2, which all activate Gs-coupled receptors and increase cAMP, are good inhibitors of animal and human smooth muscle cell proliferation (7, 22-24).

Our results presented here clearly show that PGI2 analogs are potent inhibitors of growth from cells derived from proximal human pulmonary artery. This is in contrast to previous findings in which it was reported that cicaprost had no significant effect on proliferation of cells derived from a similar region of the lung (17). There are a number of factors that could explain these discrepancies. First, PDGF was used as the growth stimulus, whereas we primarily used serum, so that it is possible the growth stimulus may influence the ability of an agent to inhibit proliferation. Second, DNA synthesis was used as the marker for cell growth, which in our hands appears to be a less sensitive marker than cell counting (compare Figures 2B and 2C). Lastly, the effects on proliferation were only reported for cicaprost and not other PGI2 analogs, and the concentrations of this agent used did not exceed 100 nM.

It is not possible at this stage to determine whether the effect of these PGI2 analogs are related solely to activation of the IP receptor, because there are currently no IP receptor antagonists available. The EC50 we obtained for cAMP generation for cicaprost and iloprost does agree well with data from displacement radioligand binding studies, which gives an inhibitory constant (Ki) of 10 and 11 nM, respectively at the IP receptor, although this does not hold true for beraprost, where a Ki of 16 nM has been reported (7). No data exists for UT-15, although our results in HEK-293 cells showed that it does indeed activate the IP receptor, only elevating cAMP in cells transfected with the receptor. Recently, mice lacking the IP receptor have been developed (25), and in these animals, the ability of cicaprost to produce hypotension and inhibit platelet aggregation was lost. This strongly suggests that the IP receptor is probably the major receptor pathway mediating the effects of cicaprost, although the role of this receptor in modulating proliferation or mediating the responses to other analogs has so far not been tested. It is known that some PGI2 analogs, including iloprost and beraprost, potently activate the EP1 and EP3 receptor subtype in the nanomolar range (7). Of these two receptors, only the EP3 receptor subtype has been identified functionally in human pulmonary artery (26). Although it would appear that the major signaling pathway of this receptor subtype is inhibition of adenylate cyclase via Gi, a number of splice variants of the EP3 receptor have recently been identified, two of which positively couple to adenylate cyclase (7). Activation of additional receptors coupled to different isoforms of adenylyl cyclase isoforms may well explain our observations with SQ22536, which significantly inhibited cAMP generated by UT-15 and to a much lesser extent iloprost, but did not affect cAMP elevated by cicaprost. Previous studies have reported differential sensitivity for P-site inhibitors depending on the adenylyl cyclase isoform present in the tissue (27). Consistent with this notion, we found in earlier studies in guinea pig aorta, that SQ22536 at the same concentration completely abolished the rise in cAMP induced either by iloprost or cicaprost, whereas DDA only inhibited cAMP elevation by ∼ 50% (18). Thus the adenylyl cyclase isoforms activated by PGI2 analogs in aorta and pulmonary artery appear different. However, which ones are being activated in human pulmonary artery remains to be elucidated, although molecular studies in rodents have identified 5 adenylyl cyclase isoforms in the lung, namely type 2, 3, 4, 7, and 9 (28). Our data is consistent with at least two isoforms being activated, one that is sensitive to inhibition by both SQ22536 and DDA, and one that is sensitive just to DDA.

Although SQ22536 substantially inhibited cAMP generation, it surprisingly had no significant effect on either UT-15 or iloprost-mediated inhibition of FBS- or PDGF-stimulated cell growth. One possible interpretation is that cAMP levels were not sufficiently reduced to affect cell growth. Certainly, cAMP levels were significantly lower in the presence of DDA compared with SQ22536, and further inhibition of the iloprost response was observed with a combination of the two inhibitors. This is in contrast to other studies, where SQ22536 does effectively oppose the action of IP agonists in other cell types. For example, it was reported to completely reverse the protective effects of iloprost against neutrophil-induced lung injury (19) and also reduced the inhibitory action of IP agonists on platelet aggregation (29). At higher concentrations (1 mM), it does also inhibit the antiproliferative effects of endothelin in human hepatic stellate which was reported to be mediated via release of PGI2 and PGE2. We did not try higher concentrations of SQ22536 in these studies, but some isoforms of adenylate cyclase require > 500 μM to inhibit the enzyme (27).

We observed that PGI2 analogs had markedly differing effects on the levels and profile of cAMP generated in HPASMC, with UT-15 being the most efficacious agent. To what extent these differences relate to the chemical and pharmacologic properties of these analogs remains unclear. Certainly, iloprost and beraprost are racemic mixtures of two and four sterioisomers, respectively, which appear to have different potencies, whereas cicaprost is a single isomeric form (30). However, effects on cAMP generation in HPASMC did not necessarily correlate with ability to inhibit proliferation. Although UT-15 was consistently the most effective antiproliferative agent, cicaprost and beraprost were significantly weaker analogs at inhibiting proliferation. It is unlikely that these effects are related to differences in chemical stability, because PGI2 analogs are stable in aqueous solution (31). Furthermore, cicaprost is metabolically stable in vivo, undergoing no metabolism in a variety of species, including humans (31). The simplest interpretation is that both UT-15 and iloprost activate additional mechanisms, which we hypothesize to be other receptors coupled to adenylate cyclase, that enhances the ability of these agents to inhibit proliferation. Indeed, clinical data indicates iloprost to be more efficacious than cicaprost in the treatment of peripheral vascular disease (12, 32), although there is no comparative data for the outcome of PPH. Moreover, evidence from earlier studies indicate that the inhibitory effects of PGI2 and PGE1 on vascular smooth muscle proliferation were additive, suggesting different mechanisms (15). Interestingly, the ability of PGE2 (and a range of other EP2 agonists) and cicaprost to inhibit human neutrophil activation did not correlate well with their ability to increase cAMP levels (33). However, each isoform of adenylate cyclase differs markedly in its response to other regulatory signals such as G, protein kinase A, protein kinase C, and calcium/calmodulin, adding a further complexity to the regulation of this system (28).

There is growing evidence that deficiencies in PGI2 synthesis may play a role in the genesis and progression of vascular remodeling in PPH. In patients with a severe form of the disease, decreased expression of PGI2 synthase, the enzyme responsible for the production of PGI2, was observed (34). Moreover, recent studies in mice showed that overexpression of PGI2 synthase prevented the development of pulmonary hypertension and right ventricular hypertrophy in response to hypoxia (35). Therefore, it may not be surprising that long-term therapy with PGI2 has resulted in increased life expectancy and exercise tolerance in patients with PPH (5). However, the complexity and complications associated with the intravenous application of an agent with a short biologic half-life have made this approach of limited utility, prompting the therapeutic evaluation of stable PGI2 analogs. We observed potent effects on proliferation in HPASMC up to 48 h after the initial application with several analogs, suggesting little desensitization of the response under our experimental conditions. However, this is in direct contrast to studies in bovine coronary artery where the antiproliferative response to iloprost completely desensitized at 24 h, although there was no desensitization to either PGE1 or the adenylyl cyclase activator, forskolin (15). This could reflect real differences in the control of IP receptor expression in coronary and pulmonary artery. Our results also indicate that the antiproliferative activity of PGI2 analogs may not solely relate to activation of IP receptors.

It is not clear at this stage whether inhibition of smooth muscle proliferation significantly contributes to the clinical benefit of PGI2 analogs that is observed in patients with PPH. Although the earliest pathologic feature of PPH is medial hypertrophy, neointimal proliferation occurs later on in the disease process, suggesting excess growth of other vascular wall elements, including endothelial cells (36). Thus, other mechanisms of action of PGI2 analogs are likely to be operative in the clinical setting. Clearly, future studies are required not only to assess the role of both smooth muscle and endothelial cells in relation to the progression of pulmonary vascular remodeling, but also to understand the cellular mechanisms associated with the therapeutic benefit of stable PGI2 analogs in PPH and other vascular diseases.

L.H.C. is an MRC Senior Fellow in Basic Biomedical Science and A.T. is a Wellcome Trust Senior Fellow in Clinical Science. The authors thank Professor Narumiya (Kyoto University, Japan) for the gift of the human IP receptor cDNA. This work was supported by United Therapeutics (Washington, DC) and the Medical Research Council, UK (G117/180). The authors would like to state that they have no financial interest (stocks and shares) in United Therapeutics. Professor Rubin does, however, serve as a consultant and investigator for this company in the clinical trials/development of UT-15 (Remodulin) for pulmonary hypertension.

1. Gaine S. P., Rubin L. J.Primary pulmonary hypertension. Lancet3521998719725
2. Christman B. W., McPherson C. D., Newman J. H., King G. A., Bernard G. R., Groves B. M., Loyd J. E.An imbalance between the excretion of thromboxane and prostacyclin metabolites in pulmonary hypertension. N. Engl. J. Med.32719927075
3. Giaid A.Nitric oxide and endothelin-1 in pulmonary hypertension. Chest1141998208S212S
4. Higenbottam T., Wheeldon D., Wells F., Wallwork J.Long-term treatment of primary pulmonary hypertension with continuous intravenous epoprostenol (prostacyclin). Lancet1198410461047
5. Barst R. J., Rubin L. J., Long W. A., McGoon M. D., Rich S., Badesch D. B., Groves B. M., Tapson V. F., Bourge R. C., Brundage B. H.A comparison of continuous intravenous epoprostenol (prostacyclin) with conventional therapy for primary pulmonary hypertension: The Primary Pulmonary Hypertension Study Group. N. Engl. J. Med.3341996296302
6. Coleman R. A., Smith W. L., Narumiya S.International union of pharmacology classification of prostanoid receptors: properties, distribution, and structure of the receptors and their subtypes. Pharmacol. Rev.461994205229
7. Narumiya S., Sugimoto Y., Ushikubi F.Prostanoid receptors: structure, properties, and functions. Physiol. Rev.79199911931226
8. Olschewski H., Ghofrani H. A., Walmrath D., Schermuly R., Temmesfeld-Wollbruck B., Grimminger F., Seeger W.Inhaled prostacyclin and iloprost in severe pulmonary hypertension secondary to lung fibrosis. Am. J. Respir. Crit. Care Med.1601999600607
9. Nagaya N., Uematsu M., Okano Y., Satoh T., Kyotani S., Sakamaki F., Nakanishi N., Miyatake K., Kunieda T.Effect of orally active prostacyclin analogue on survival of outpatients with primary pulmonary hypertension. J. Am. Coll. Cardiol.34199911881192
10. Hoeper M. M., Schwarze M., Ehlerding S., Adler-Schuermeyer A., Spiekerkoetter E., Niedermeyer J., Hamm M., Fabel H.Long-term treatment of primary pulmonary hypertension with aerosolized iloprost, a prostacyclin analogue. N. Engl. J. Med.342200018661870
11. Barst, R. J., G. Simonneau, R. Stuart, S. D. Blackburn, R. Naeije, and L. J. Rubin. 2000. Efficacy and safety of chronic subcutaneous infusion UT-15 in pulmonary arterial hypertension (PAH). Circulation 102(Suppl. 2):100. (Abstr.)
12. Grant S. M., Goa K. L.Iloprost: a review of its pharmacodynamic and pharmacokinetic properties, and therapeutic potential in peripheral vascular disease, myocardial ischaemia and extracorporeal circulation procedures. Drugs431992889924
13. Fredrich, M., and B. Muller. 1992. Prostacyclin and atherosclerosis. In G. M. Rubanyi and J. R. Vane, editors. Prostacyclin: new prospectives for basic research and novel therapeutic indications. Elsevier Science Publishers, Amsterdam. 169–180.
14. Koh E., Morimoto S., Jiang B., Inoue T., Nabata T., Kitano S., Yasuda O., Fukuo K., Ogihara T.Effects of beraprost sodium, a stable analogue of prostacyclin, on hyperplasia, hypertrophy and glycosaminoglycan synthesis of rat aortic smooth muscle cells. Artery201993242252
15. Grosser T., Bonisch D., Zucker T. P., Schror K.Iloprost-induced inhibition of proliferation of coronary artery smooth muscle cells is abolished by homologous desensitization. Agents Actions Suppl4519958591
16. Harada M., Toki Y., Numaguchi Y., Osanai H., Ito T., Okumura K., Hayakawa T.Prostacyclin synthase gene transfer inhibits neointimal formation in rat balloon-injured arteries without bleeding complications. Cardiovasc. Res.431999481491
17. Wharton J., Davie N., Upton P. D., Yacoub M. H., Polak J. M., Morrell N. W.Prostacyclin analogues differentially inhibit growth of distal and proximal human pulmonary artery smooth muscle cells. Circulation102200031303136
18. Turcato S., Clapp L. H.The effects of the adenylyl cyclase inhibitor SQ22536 on iloprost-induced vasorelaxation and cAMP elevation in isolated guinea-pig aorta. Br. J. Pharmacol.1261999845847
19. Riva C. M., Morganroth M. L., Ljungman A. G., Schoeneich S. O., Marks R. M., Todd R. F., Ward P. A., Boxer L. A.Iloprost inhibits neutrophil-induced lung injury and neutrophil adherence to endothelial monolayers. Am. J. Respir. Cell Mol. Biol.31990301309
20. Shirotani M., Yui Y., Hattori R., Kawai C.U-61,431F, a stable prostacyclin analogue, inhibits the proliferation of bovine vascular smooth muscle cells with little antiproliferative effect on endothelial cells. Prostaglandins41199197110
21. Orekhov A. N., Tertov V. V., Kudryashov S. A., Khashimov K., Smirnov V. N.Primary culture of human aortic intima cells as a model for testing antiatherosclerotic drugs: effects of cyclic AMP, prostaglandins, calcium antagonists, antioxidants, and lipid-lowering agents. Atherosclerosis601986101110
22. Owen N. E.Effect of prostaglandin E1 on DNA synthesis in vascular smooth muscle cells. Am. J. Physiol.2501986C584C588
23. Nilsson J., Olsson A. G.Prostaglandin E1 inhibits DNA synthesis in arterial smooth muscle cells stimulated with platelet-derived growth factor. Atherosclerosis5319847782
24. Jourdan K. B., Evans T. W., Lamb N. J., Goldstraw P., Mitchell J. A.Autocrine function of inducible nitric oxide synthase and cyclooxygenase-2 in proliferation of human and rat pulmonary artery smooth-muscle cells: species variation. Am. J. Respir. Cell Mol. Biol.211999105110
25. Murata T., Ushikubi F., Matsuoka T., Hirata M., Yamasaki A., Sugimoto Y., Ichikawa A., Aze Y., Tanaka T., Ueno A., Oh S.-ishi, and S. NarumiyaAltered pain perception and inflammatory response in mice lacking prostacyclin receptor. Nature3881997678682
26. Qian Y. M., Jones R. L., Chan K. M., Stock A. I., Ho J. K.Potent contractile actions of prostanoid EP3 receptor agonists on human isolated pulmonary artery. Br. J. Pharmacol.1131994369374
27. Johnson R. A., Desaubry L., Bianchi G., Shoshani I., Lyons E., Taussig R., Watson P. A., Cali J. J., Krupinski J., Pieroni J. P., Iyengar R.Isozyme-dependent sensitivity of adenylyl cyclases to P-site-mediated inhibition by adenine nucleosides and nucleoside 3′-polyphosphates. J. Biol. Chem.272199789628966
28. Simonds W. F.G protein regulation of adenylate cyclase. Trends Pharmacol. Sci.2019996673
29. Armstrong R. A., Jones R. L., MacDermot J., Wilson N. H.Prostaglandin endoperoxide analogues which are both thromboxane receptor antagonists and prostacyclin mimetics. Br. J. Pharmacol.871986543551
30. Ashby B.Comparison of iloprost, cicaprost and prostacyclin effects on cyclic AMP metabolism in intact platelets. Prostaglandins431992255261
31. Hildebrand M.Inter-species extrapolation of pharmacokinetic data of three prostacyclin-mimetics. Prostaglandins481994297312
32. Pope J.Treatment of systemic sclerosis. Curr. Opin. Rheumatol.51993792801
33. Armstrong R. A.Investigation of the inhibitory effects of PGE2 and selective EP agonists on chemotaxis of human neutrophils. Br. J. Pharmacol.116199529032908
34. Tuder R. M., Cool C. D., Geraci M. W., Wang J., Abman S. H., Wright L., Badesch D., Voelkel N. F.Prostacyclin synthase expression is decreased in lungs from patients with severe pulmonary hypertension. Am. J. Respir. Crit. Care Med.159199919251932
35. Geraci M. W., Gao B., Shepherd D. C., Moore M. D., Westcott J. Y., Fagan K. A., Alger L. A., Tuder R. M., Voelkel N. F.Pulmonary prostacyclin synthase overexpression in transgenic mice protects against development of hypoxic pulmonary hypertension. J. Clin. Invest.103199915091515
36. Heath D., Edwards J. E.The pathology of hypertensive pulmonary vascular disease. Circulation181958533541
Address correspondence to: Dr. Lucie Clapp, Centre for Clinical Pharmacology, UCL, Rayne Institute (4th Floor), 5 University Street, London W1C 6JJ, UK. E-mail:

Abbreviations: cyclic AMP, cAMP; 2'5′dideoxyadenosine, DDA; fetal bovine serum, FBS; human pulmonary artery smooth muscle cells, HPASMC; 3-isobutyl-1-methylxanthine, IBMX; primary pulmonary hypertension, PPH; platelet-derived growth factor, PDGF; prostacyclin, PGI2; prostacyclin receptor, IP receptor; smooth muscle cell growth medium, SmGM; trichloroacetic acid, TCA.


No related items
American Journal of Respiratory Cell and Molecular Biology

Click to see any corrections or updates and to confirm this is the authentic version of record