INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: hypertension; pulmonary - endothelins - receptors; endothelin - muscle; smooth; vascular - pulmonary artery

Pulmonary arterial hypertension is characterized by structural changes in the pulmonary vasculature, as well as vasoconstriction, and occurs either as primary pulmonary hypertension (PPH) or in association with diseases such as congenital systemic to pulmonary shunts (1). The arteriopathy found in these related diseases exhibits recognizable morphologic features, including medial hypertrophy, cellular intimal proliferation, intimal fibrosis, and obstructive lesions (2, 3). Endothelin-1 (ET-1) has been implicated in the development of pulmonary hypertension and associated vascular remodeling because of its dual vasoconstrictor (4, 5) and mitogenic actions (6). The expression of ET-1 is also increased in the lungs of patients with pulmonary arterial hypertension (7), and in experimental models this has been associated with an increased susceptibility to develop pulmonary vascular remodeling (8) and the augmented growth of pulmonary artery smooth muscle cells (PASMCs) in culture (9).

ET-1 is a 21 amino-acid peptide that is derived from big-ET-1 via the action of the phosphoramidon-sensitive metalloproteinase, endothelin-converting enzyme (ECE), which occurs in several isoforms and is present in both vascular smooth muscle cells and endothelial cells (10). Increasing evidence indicates that vascular smooth muscle cells, as well as endothelial cells, synthesize and release ET-1, particularly when stimulated by cytokines (11). Although ovine PASMCs express ET-1 (12), its role as an autocrine or paracrine regulator of human PASMCs has not been fully elucidated. Two distinct G-protein- coupled ET_A and ET_B receptors mediate the effects of ET-1 in the pulmonary vasculature and these receptors exhibit species, regional, and developmental differences in their distribution patterns (4, 13). Both types of receptor are involved in the ET-1-induced contraction of distal human pulmonary resistance arteries, whereas ET_A receptors mediate the response in proximal arteries (4, 5). The proliferative effects of ET-1 on PASMCs isolated from the main human pulmonary artery are also considered to be dependent on ET_A receptors (6), but the contribution of ET_A and ET_B receptors in regulating the growth of cells from distal resistance vessels is uncertain. An upregulation in vascular ET_A and ET_B expression has been demonstrated in experimental models of systemic hypertension and associated with augmented responses to ET-1 (14, 15). Some studies on hypoxia-induced pulmonary hypertension have also found an increase in the expression of ET-1 and ET receptors in the lung (16, 17), although the effects on pulmonary vascular ET-1 binding appear to vary (18). Little is known about the possible modulation of ET receptors in human pulmonary hypertension, but an increase in ET_A receptor density has been described in the pulmonary arteries and parenchyma of patients with congenital heart disease (21). Nonetheless, ET-1 receptor antagonists attenuate or reverse the development of experimental pulmonary hypertension and right ventricular hypertrophy (22, 23), and the endothelin system has become an attractive therapeutic target for the treatment of pulmonary hypertension (24).

Heterogeneity among PASMC phenotypes is now well recognized (12, 25) and it is therefore vital that cells for in vitro studies are derived from appropriate regions of the pulmonary vasculature. The importance of this is emphasized by our recent finding of regional variation in the serum-stimulated growth of human PASMCs in culture and the distinct antiproliferative effects of prostacyclin (PGI₂) analogues in cells derived from distal and proximal pulmonary arteries (26). The involvement of PGI₂ in modulating the release and action of ET-1 in human PASMCs is unclear, but it has been shown to inhibit both the production of and mitogenic response to ET-1 in bovine vascular cells (27), and in patients with severe pulmonary hypertension, continuous PGI₂ infusion improves the balance between pulmonary ET-1 release and clearance (28).

The aim of this study was to test the hypothesis that both ET_A and ET_B receptors affect the proliferation of distal human PASMCs and may therefore contribute to the vascular remodeling in pulmonary hypertension. Specifically, we sought to establish the distribution of ET_A and ET_B receptors in the normal and hypertensive human pulmonary circulation, determine the effects of ET receptor stimulation on the proliferation of isolated distal human PASMCs, investigate the autocrine role of ET-1 in these cells, and establish the influence of the PGI₂ analogue cicaprost on these processes.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Reagents

Dulbecco's modified Eagle's medium (DMEM) was from Life Technologies (Paisley, Scotland) and fetal bovine serum (FBS) from TCS CellWorks Ltd. (Botolph Claydon, Bucks, UK). PD145065, PD156707, and PD180988 were from Pfizer (Groton, CT), and cicaprost from Schering (Berlin, Germany). [¹²⁵I]-ET-1, [methyl-³H]-thymidine and Hyperfilm-³H were purchased from Amersham Pharmacia Biotech (Little Chalfont, Bucks, UK) and radioactive standards were from American Radiolabeled Chemicals Inc. (St. Louis, MO). Platelet-derived growth factor-BB (PDGF-BB) and transforming growth factor-1 (TGF-1) were from R & D Systems (Abingdon, Oxon, UK). Unless otherwise stated, other reagents were obtained from Sigma-Aldrich Co. Ltd (Poole, Dorset, UK).

Tissues and Cells

Lung tissues for receptor autoradiography and immunostaining were obtained at lung or heart-lung transplantation from patients with PPH (n = 12) and pulmonary hypertension associated with congenital heart disease (n = 10) (Table 1). Control tissues were obtained from unused donor lungs (n = 10) and at lobectomy or pneumonectomy for bronchial carcinoma (n = 3) and did not exhibit pulmonary vascular remodeling. Tissues were mounted on cork mats, surrounded with mounting medium, frozen in melting isopentane, and stored at 40° C (29). Cultures of human PASMCs (n = 7) were derived from different individuals, by microdissection and collagenase digestion of distal human pulmonary arteries (< 1 mm external diameter), and the smooth muscle phenotype was confirmed immunohistochemically, as previously described (26). Ethical approval was obtained from both Hammersmith and Harefield Hospital Ethics Committees.

Receptor Autoradiography and Ligand Binding

The distribution of ET_A and ET_B receptors was determined using [¹²⁵I]-ET-1 binding and in vitro autoradiography, as described (29). Briefly, serial cryostat sections (10 µm thick) were thaw-mounted on vectabond-treated slides and stored at 40° C. Sections were preincubated for 10 min in 50 mmol/L TRIS-HCl buffer (pH 7.2) containing 100 mmol/L NaCl and 5 mmol/L MgCl₂ and then incubated in fresh buffer containing 0.5% BSA and 0.1 nmol/L [¹²⁵I]-ET-1 for 120 min at 20 to 22° C. Sections were washed in three changes of ice-cold buffer, 5 min each, rinsed in cold distilled water, and dried under a stream of cold air. Nonspecific binding was determined in the presence of 10⁷ mol/L unlabeled ET-1 or 10⁵ mol/L PD145065. ET-1 is a dual ET_A/ ET_B nonselective agonist and the proportions of ET_A and ET_B subtypes were established using ET_A-selective antagonists (BQ-123, PD156707, PD180988), ET_B-selective antagonists (BQ-788), and the ET_B-selective agonist sarafotoxin 6c (S6c). Macroautoradiographic images were obtained by apposing labeled sections to Hyperfilm-³H, together with radioactive standards, for 3 d at 4° C, and standard curves were generated for each film, using the images of at least six radioactive standards and computer-assisted image analysis (KS 300; Imaging Associates, Thame, UK). Binding to specific tissue structures was delineated interactively, with at least six regions per vessel and up to seven separate vessels per sample being measured. The integrated grey values were converted to the amount of ligand bound using the standard calibration curves and binding expressed as attomoles of ligand bound per unit area (amol/mm²) (29). Equilibrium binding density was determined for the lung parenchyma, including alveolar walls, and five different levels of the pulmonary tree, corresponding to elastic (> 8 mm, 4 to 8 mm, and 1 to 4 mm external diameter), transitional (0.5 to 1.0 mm external diameter), and predominantly muscular arteries (0.1 to 0.5 mm external diameter).

Endothelin receptors on PASMCs (passage 4-10) were characterized using cells seeded in 24- or 48-well plates (0.5-1.0 × 10⁴ cells/well) and grown to 80 to 90% confluent. The cells were washed twice in serum-free DMEM medium and incubated in fresh DMEM containing 0.3% (wt/vol) BSA and [¹²⁵I]-ET-1 (50 pmol/L) for 120 min at 37° C, equilibrium being reached by 90 min. Nonspecific binding was determined in the presence of excess unlabeled ET-1 (100 nmol/L). Binding was characterized by saturation analysis with 5 to 500 pmol/L [¹²⁵I]-ET-1 and competitive inhibition experiments, cells being incubated with 50 pmol/L [¹²⁵I]-ET-1 and increasing concentrations (10¹³ to 10⁵ mol/L) of ET-3, S6c, BQ-123, PD145065, PD156707, or PD180988. After incubation, cells were washed three times with 0.5 ml ice-cold DMEM, lyzed with 0.2 mol/L NaOH and 0.1% (wt/vol) SDS for 30 min and bound ligand measured using a gamma-counter. All experiments were performed in triplicate, repeated at least twice, and were standardized for cell number.

Immunohistochemistry and Morphometry

The localization of [¹²⁵I]-ET-1 binding sites was further assessed by comparison with adjacent sections, immunostained using the avidin-biotin complex (ABC) and indirect immunofluorescence methods (29). Primary antisera were raised to -smooth muscle actin (IA4), smooth muscle myosin (hSM-V), cytokeratin-18 (KS-B17.2; Sigma-Aldrich), CD31 (JC/70A; Dako Ltd., Ely, UK) and smoothelin (R4A; Monosan, Uden, The Netherlands).

The dimensions of pulmonary arteries were determined using hematoxylin-eosin-stained sections of lung and computer-assisted image analysis. The arterial diameter was determined by measuring the distance between external laminae measured at two perpendicular axes. The medial thickness in distal muscular arteries (0.1 to 0.5 mm external diameter) was derived from measurements of the distance between internal and external laminae, values being obtained for up to 12 vessels per sample and expressed as a percentage of the external artery diameter, as previously described (30).

DNA Synthesis and Proliferation of PASMCs

Cells were suspended in DMEM containing 10% FBS, seeded (10⁴ cells/well) in 24-well plates, grown until 80 to 90% confluent, and quiesced by incubation in serum-free DMEM for 2 h followed by serum deprivation in DMEM containing 0.1% FBS for 72 h. The incorporation of [methyl-³H]-thymidine was then measured over 24 h (26) and the effects of ET-1 and sarafotoxin 6c (S6c) determined in the absence and presence of PDGF-BB (0.1 to 10 ng/ml), cicaprost (10⁷ mol/L), or forskolin (10⁵ mol/L), and selective (BQ-123, BQ-788) or nonselective ET antagonists (PD145065). To determine the effects of endogenous ET release on cell proliferation, PASMCs were plated and quiesced as described above and then incubated for up to 11 d in DMEM containing 10% FBS, in the absence and presence of either PD145065 (10⁵ mol/L) or the endothelin-converting enzyme-1 inhibitor phosphoramidon (10⁵ mol/L). The medium was replaced every 2 to 3 d and cells counted using a hemocytometer and trypan blue exclusion.

Endothelin Release and cAMP Production

Release of ET immunoreactivity was determined by radioimmunoassay and fast protein liquid chromatography (FPLC) (31) after stimulation with 10% FBS and TGF-1 (10 ng/ml) for up to 48 h. Intracellular cAMP accumulation was assessed after 15-min stimulation at room temperature in the presence of 5 × 10⁵ mol/L 3-isobutyl-1-methylxanthine (IBMX), using a ¹²⁵I-labeled cAMP assay system (NEN^TM; Life Science Products Inc., Zaventen, Belgium) as described (26).

Statistical Analysis

Data, expressed as mean ± SEM or mean and 95% confidence interval, were analyzed by Student's t test or one-way ANOVA, with post-hoc Tukey's test. Dissociation constant (K_d) and maximum binding capacity (B_max) were derived using iterative nonlinear regression (GraphPAD Prism^TM, version 3; GraphPad Software, San Diego, CA). p < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Distribution of ET_A and ET_B Receptors

Specific [¹²⁵I]-ET-1 binding sites were localized to pulmonary arteries, veins, airways, and parenchyma and regional differences demonstrated in binding density and the proportion of ET_A and ET_B receptors in the media of pulmonary arteries (Figures 1 and 2). Equilibrium binding density was greater (p < 0.001) in distal than in trunk or lobar arteries (Figure 2A), there being an inverse correlation between vessel size and [¹²⁵I]-ET-1 binding (r² = 0.851, p < 0.001). Regional variation in binding was associated with the differential distribution of ET_A and ET_B receptors (Figure 2B). The ET_A subtype predominated in proximal arteries (97 ± 1% ET_A; > 8.0 mm external diameter, n = 13), whereas the proportion of ET_B receptors increased significantly in distal arteries (36 ± 3% ET_B; p < 0.001; 0.5 to 1.0 mm external diameter; n = 17). ET_B receptors also represented the main receptor subtype in the lung parenchyma (59 ± 2%; p < 0.001; n = 18) and bronchial smooth muscle (82 ± 3%; p < 0.001; n = 8).

View larger version (139K): [in this window] [in a new window]   Figure 1.   Autoradiographic images showing [125I]-ET-1 binding in the media of pulmonary arteries (arrows), parenchyma, and alveolar walls (arrowhead). Sections incubated with 107 mol/L BQ-123 (ETB sites), S6c (ETA sites), or ET-1 (NS, nonspecific binding). Proportion of ETB receptors increased from proximal (A) to intermediate (B) and distal regions (C ). Nonspecific binding represented as much as 33% and < 10% of total binding in proximal and distal arteries, respectively. Asterisk indicates terminal bronchiole.

View larger version (20K): [in this window] [in a new window]   Figure 2.   Regional heterogeneity in ETA and ETB receptor distribution. (A) Equilibrium binding density (amol/mm2) varied between distal and proximal arteries (ap < 0.001 versus > 8.0 mm; bp < 0.05 versus 1.0 to 4.0 mm external diameter). (B) The proportion (%) of ETA and ETB receptors differed in distal and proximal arteries (ap < 0.05; bp < 0.001 versus > 8.0 mm external diameter). Data are mean ± SEM, for 10 to 20 vessels/region, in four different arterial regions and a total of 18 subjects.

In the most distal arterial segments (0.1 to 0.5 mm external diameter), medial thickness was 2.23- and 2.57-fold greater (p < 0.05) in patients with PPH and hypertension associated with congenital heart disease than in control subjects. The number of [¹²⁵I]-ET-1 binding sites in the distal arterial region was also greater in the hypertensive groups than in the control group (Table 2), although the proportions of ET_A and ET_B receptors were equal. Equilibrium binding density in the parenchyma, including the alveolar walls, was greater in the PPH (160.0 ± 10.1 amol/mm²; p < 0.001; n = 6) and patients with congenital heart disease (114.7 ± 10.1 amol/mm²; p < 0.05; n = 6) than in the control subjects (78.40 ± 2.6 amol/mm²; n = 6). There was, however, no change in the relative proportion of ET_A and ET_B subtypes. No significant differences were found between the control and the PPH groups in the specific binding exhibited either by proximal arterial regions (80.1 ± 17.5 amol/ mm²; n = 5 versus 57.7 ± 6.9 amol/mm²; n = 6; 1 to 4 mm external diameter; p = 0.166) or by bronchial smooth muscle (220.0 ± 30.5 amol/mm²; n = 4 versus 263.5 ± 69.2 amol/mm²; n = 5; p = 0.614).

Microscopic Localization of ET Receptors

Intra-acinar arteries and arterioles exhibited both ET_A and ET_B receptors, colocalized with smooth muscle markers in the media (Figure 3A). Pulmonary arteries from hypertensive patients exhibited both medial hypertrophy and neointimal proliferation and a differential distribution of immunohistochemical markers (Figures 3B-3D). Smooth muscle actin and myosin immunoreactivity was localized to both the media and neointima throughout the arterial tree. Smoothelin, a marker for differentiated smooth muscle cells (32), was restricted to the medial layer in proximal arteries, whereas cytokeratin 18 immunoreactivity, a marker of synthetic dedifferentiated smooth muscle cells (33), was localized to the neointima in preacinar arteries. [¹²⁵I]-ET-1 binding was prominent throughout the media, representing both ET_A and ET_B receptors, whereas binding to the neointima was generally less dense and comprised mainly ET_A receptors (Figures 3C and 3D). ET_A and ET_B receptors were also localized to the thin-walled distal branches of arteries containing plexiform lesions, pulmonary veins, the parenchyma, and alveolar walls, the latter displaying immunoreactivity for markers of endothelium and epithelium but not smooth muscle cells.

View larger version (98K): [in this window] [in a new window]   Figure 3.   Microscopic distribution of [125I]-ET-1 binding. Sections of pulmonary artery, from controls (A) and hypertensive subjects (B-D), showing the localization of smooth muscle myosin (SM-myosin), -smooth muscle actin (-SMA), smoothelin, cytokeratin-18, and CD31 immunoreactivity and the distribution of ETA and ETB receptors in the media (arrowheads) and neointima (asterisks). Arteries from hypertensive patients exhibited characteristic medial hypertrophy (B) and neointimal proliferation (C and D). Arrow indicates internal elastic lamina.

Analysis of ET Receptors on PASMCs

Distal PASMC isolates displayed high affinity [¹²⁵I]-ET-1 binding, comprising both ET_A and ET_B receptors (Figures 4A and 4B). Selective ET agonists and antagonists distinguished two receptor populations (Figure 4A), the rank order of affinity being PD156707 = BQ-123 = PD180988 >> ET-3 > S6c for ET_A receptors and ET-3 = S6c >> PD156707 = BQ-123 = PD180988 for ET_B receptors (Table 3). The overall proportions of ET_A and ET_B receptors in cell isolates (67:33%; n = 7) corresponded with those of distal arteries in tissue sections (64:36%; 0.1 to 1.0 mm external diameter; n = 19). Saturation analysis indicated that although there was no significant difference in binding affinity, the number of binding sites varied with the expression of ET_B receptors (Figure 4B). Binding characteristics were unchanged during continuous culture (passages 4-10) and after the recovery of frozen cells.

View larger version (18K): [in this window] [in a new window]   Figure 4.   Characterization of [125I]-ET-1 binding to distal PASMCs. (A) Inhibition with nonselective (ET-1, PD145065), ETA-selective (BQ-123), and ETB- selective competitors (ET-3, S6c). (B) Saturation binding showing differences in receptor density (p = 0.0127) between PASMC isolates expressing similar proportions of ETA and ETB receptors (closed squares) (Bmax 83.6 ± 6.7 fmol/106 cells; Kd, 57 ± 6 pmol/L) and those possessing mainly ETA receptors (open squares) (Bmax 48.7 ± 4.7 fmol/106 cells; Kd 64 ± 4 pmol/L). Data are mean ± SEM, for three PASMC isolates measured in triplicate.

Effects of ET-1 and S6c on DNA Synthesis and Proliferation of PASMCs

ET-1 (10⁹ to 10⁷ mol/L) stimulated [methyl-³H]-thymidine uptake in serum-deprived cells (Figure 5A), but did not affect the fivefold (p < 0.001) increase induced by PDGF-BB or serum. ET-1 attenuated the inhibitory effects of cicaprost and forskolin on DNA synthesis (Figure 5A). PASMC isolates with equivalent proportions of ET_A and ET_B receptors (49:51%; n = 3) exhibited concentration-dependent responses to both ET-1 and S6c (Figure 5B) that were mediated by ET_A and ET_B receptors (Figures 5C and 5D), whereas isolates with predominantly ET_A receptors (81:19%; n = 4) responded only to ET-1 (data not shown).

View larger version (32K): [in this window] [in a new window]   Figure 5.   Effects of ET-1 and S6c on DNA synthesis. (A) ET-1 stimulated DNA synthesis in serum-deprived PASMCs (ap < 0.05 versus control). Inhibitory effects of cicaprost and forskolin on DNA synthesis (bp < 0.001) were attenuated by ET-1 (cp < 0.01 versus cicaprost; versus forskolin). Data are mean ± SEM, for seven PASMC isolates. (B) Concentration-dependent effect of ET-1 and S6c on cicaprost-induced inhibition of DNA synthesis (bp < 0.001). ET-1 (C ) and S6c (D) stimulated DNA synthesis (ap < 0.001) in cicaprost-treated PASMCs expressing 45% ETA and 55% ETB receptors. BQ-123 (bp < 0.001) and BQ-788 (cp < 0.01) inhibited the response to ET-1, whereas BQ-788 selectively inhibited the effect of S6c (ep < 0.001) and PD145065 abolished the effects of both peptides (dp < 0.001; fp < 0.001). Data are mean ± SEM, for four to six replicates.

ET-1 Release by PASMCs and Effects on Cell Proliferation

ET-like immunoreactivity was detected in PASMC-conditioned medium after stimulation with either 10% FBS or TGF-1 (10 ng/ml) and 0.1% FBS for 6 to 48 h. On reversed-phase FPLC, the immunoreactivity coeluted with synthetic human ET-1 (Figure 6A). Incubation of PASMCs with cicaprost (10¹⁰-10⁷ mol/L) for 24 h induced a concentration-dependent reduction in ET-1 release compared with control values of 5.2 ± 1.1 fmol/10⁵ cells (n = 5), and this effect was mimicked by IBMX and forskolin (Figure 6B).

View larger version (22K): [in this window] [in a new window]   Figure 6.   Release of ET-1 from distal PASMC isolates. (A) Chromatographic analysis of ET-like immunoreactivity in PASMC-conditioned medium after 10 h, coeluting with synthetic ET-1. (B) Concentration-dependent inhibition of ET-1 release by cicaprost, IBMX, and forskolin after 24 h (ap < 0.05; bp < 0.001 versus control). Data are mean ± SEM, for five PASMC isolates measured in triplicate.

Serum-stimulated PASMC proliferation was significantly inhibited by both phosphoramidon and PD145065 (Figures 7A- 7C). In cells expressing similar proportions of ET_A and ET_B receptors, phosphoramidon-induced inhibition of proliferation was reversed by exogenous ET-1 and, in turn, inhibited by both ET_A and ET_B antagonists (Figure 7B). In contrast, PASMCs with predominantly ET_A receptors displayed no significant response to S6c and only ET_A antagonists blocked the effect of ET-1 on cell proliferation (Figure 7C). In addition to stimulating ET-1 release, TGF-1 (10 ng/ml) increased PASMC proliferation (163 ± 6% of control at 6 d; p < 0.01; n = 4), the response being attenuated by both ET_A and ET_B selective antagonists (data not shown).

View larger version (17K): [in this window] [in a new window]   Figure 7.   Effect of endogenous ET-1 production on PASMC proliferation. (A) Inhibition of serum-stimulated proliferation by PD145065 (ap < 0.05; bp < 0.01; cp < 0.001). Data are mean ± SEM for three PASMC isolates measured in quadruplicate. (B and C ) Inhibition of serum-stimulated proliferation by phosphoramidon (ap < 0.05) and effect of ET-1 (bp < 0.001). In isolates expressing similar proportions of ETA and ETB receptors (B), both ETA (BQ-123; cp < 0.01) and ETB receptors (BQ-788; dp < 0.001) mediated the response to ET-1 and it was abolished by dual receptor blockade (BQ-123 + BQ-788; ep < 0.001). In PASMCs expressing predominantly ETA receptors (C ), ET-1, but not S6c, stimulated proliferation and this was abolished by BQ-123 (cp < 0.05) or PD145065 (dp < 0.01). Data are mean ± SEM of quadruplicate measurements at 7 d, in the absence (open squares) and presence of phosphoramidon (closed squares).

Regulation of cAMP Production

In PASMCs expressing similar proportions of ET_A and ET_B receptor subtypes, intracellular cAMP production was stimulated by cicaprost and attenuated in a concentration-dependent manner by both ET-1 and S6c (Figures 8A and 8B). The responses were inhibited by BQ-788 and PD145065, but not by BQ-123, suggesting that ET_B receptors mediated the reduction in cAMP production (Figures 8C and 8D).

View larger version (20K): [in this window] [in a new window]   Figure 8.   Effects of ETA and ETB receptor stimulation on intracellular cAMP production in PASMCs. (A and B) Cicaprost-stimulated cAMP production (ap < 0.001) was inhibited by both ET-1 (bp < 0.05; cp < 0.001) and S6c (dp < 0.01). (C and D) The inhibitory effect of ET-1 (ap < 0.01) and S6c (bp < 0.001) was attenuated by BQ-788 (cp < 0.001) and PD145065 (dp < 0.001; ep < 0.05), but not by Q-123. Data are mean ± SEM of quadruplicate measurements.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We determined the differential distribution of ET_A and ET_B receptors in the hypertensive and normal human pulmonary artery tree and found that both receptor subtypes influence the proliferation of distal human PASMCs. ET-1 exerted a significant positive effect on DNA synthesis and attenuated the antiproliferative effects of the PGI₂ analogue cicaprost and cAMP signaling. Furthermore, the release of endogenous ET-1 contributes, at least in part, to the serum-stimulated proliferation of human PASMCs in vitro.

ET_A and ET_B receptor subtypes exhibited differential distributions in the proximal and distal segments of human pulmonary arteries, underlining our previous findings of heterogeneity among human PASMCs in these regions (26). The differences in ET receptor localization appear to be functionally significant as they parallel regional variations in ET_A and ET_B receptor-mediated contraction of human proximal and distal pulmonary arteries (4, 5). Studies with experimental models of hypoxia-induced pulmonary hypertension have described differing effects on ET_A and ET_B receptor expression (16, 17) and [¹²⁵I]-ET-1 binding in pulmonary arteries (18, 19). However, the desensitization or downregulation of pulmonary ET_B receptors has been postulated to contribute to the development of human pulmonary hypertension (34). In the present study, we found an apparent increase in ligand binding, but no difference in the relative proportions of ET_A and ET_B receptors, in distal pulmonary arteries and parenchyma in tissue sections of lung from hypertensive patients compared with control subjects. This may have important consequences, as changes in the density of ET_A and ET_B receptors on vascular smooth muscle cells can influence both the mitogenic (35) and vasoconstrictor effects of ET-1 (14, 15). In a preliminary report on ET receptors in lung biopsies from patients with congenital heart disease, Lutz and colleagues (21) also described an increase in ligand binding to distal pulmonary arteries and parenchyma, although, surprisingly, ET_B receptor expression was found to be low. This contrasts both with the present findings and the results of an independent autoradiographic study of patients with scleroderma-associated fibrotic lung disease that demonstrated increased binding in the parenchyma of the fibrotic lung, reflecting a selective increase in the proportion of ET_B receptors (38). We considered that [¹²⁵I]-ET-1 binding in the lung parenchyma represented, in part, ET_A and ET_B receptors localized to the pulmonary microvasculature, but these receptors were not characterized further because of the limited resolution of the autoradiographic technique and the close proximity of endothelial, epithelial, and interstitial cells.

The proportions of ET_A or ET_B receptors identified in cultured PASMCs corresponded with those determined for distal pulmonary arteries in tissue sections, and the binding capacity of isolated cells also varied in parallel with the expression of ET_B receptors. Thus, PASMC isolates exhibiting equivalent proportions of ET_A and ET_B receptors had a significantly greater binding capacity than did those displaying a majority of ET_A receptors. These differences were not associated with a variation in receptor affinity and did not change during subculture up to passage 10. This contrasts with earlier studies on isolated rat and rabbit vascular smooth muscle cells where a rise (35) or decline in ET_B receptor expression was observed during culture (36).

Stimulation of both ET_A and ET_B receptors promoted the proliferation of distal human PASMCs treated with cicaprost or cultured in the presence of phosphoramidon. Stimulation of ET_B receptors also reduced agonist-stimulated intracellular cAMP levels selectively, this being consistent with the negative coupling of ET_B receptors to adenylyl cyclase (39). However, it is likely that additional mechanisms are involved in mediating the effects of ET-1 on distal PASMC proliferation, (39) and further investigations will be required to determine the extent of possible crosstalk between ET-1 and cAMP signaling in modulating the proliferation of these cells. We demonstrated further that cicaprost inhibited the release of ET-1 from serum-stimulated human PASMCs. The effect was mimicked by IBMX and forskolin, indicating the involvement of a phosphodiesterase-sensitive cAMP-dependent mechanism, and suggests that the inhibition of ET-1 release may contribute, in part, to the antiproliferative effects of PGI₂ analogues on distal human PASMCs in vitro (26). Indeed, the inhibitory response may underlie the beneficial effects of PGI₂ therapy, reducing the net production of ET-1 by the lung in pulmonary hypertensive patients (28), and could contribute in the long-term response to PGI₂ treatment, which appears to be independent of its vasodilator properties (40).

Previous studies have indicated that ET_A and ET_B antagonists act synergistically to inhibit ET-1-induced contractions of pulmonary arteries (41) and postulated that the combined blockade of both receptor subtypes would be more effective than selective ET_A receptor antagonism in the treatment of severe pulmonary hypertension (5). Few studies have compared directly the effects of these two strategies on cardiovascular structure, but dual ET antagonists are reported to prevent or reverse vascular remodeling and right ventricular hypertrophy in experimental pulmonary hypertension (22, 23). Our results suggest that both ET_A and ET_B receptors modulate the proliferation of distal human PASMCs in culture. Potential targets in vivo include cells in intimal lesions, which comprise mainly smooth muscle cells or myofibroblasts, as well as the media. Neointimal cells in middle-proximal regions of the arterial tree also exhibited cytokeratin-18, but not smoothelin immunoreactivity, and may represent an undifferentiated smooth muscle cell phenotype (32, 33). Specific [¹²⁵I]-ET-1 binding sites, comprising ET_A rather than ET_B receptors, were localized throughout the neointima, and although less numerous than in the adjacent media, could be another site of action for ET-1.

There are some limitations to this study. Firstly, despite demonstrating that the binding capacity of isolated PASMCs varied with the relative expression of ET_A and ET_B receptors, changes in cell type and number or variation in receptor occupancy could have influenced ET receptor density measured in tissue sections. Secondly, the apparent increase in ET receptor expression in pulmonary hypertension should be investigated at the molecular as well as the protein level, and the distribution of ET receptor subtypes in the pulmonary microvasculature needs to be defined in greater detail. The impact of ET receptor antagonists and PGI₂ analogues on vascular remodeling in the hypertensive human pulmonary circulation also requires further investigation.

In summary, our findings provide further evidence of regional heterogeneity among smooth muscle cell phenotypes in human pulmonary arteries and support the hypothesis that both ET_A and ET_B receptors affect the proliferation of distal human PASMCs. The counterregulatory effects of ET-1 and cicaprost or cAMP stimulation on cultured human PASMCs suggests that combined therapy, using an ET antagonist and a PGI₂ analogue, could be used to attenuate or reverse pulmonary vascular remodeling in patients with pulmonary hypertension.