Intersegmental vessel formation in zebrafish: requirement for VEGF but not BMP signalling revealed by selective and non-selective BMP antagonists.

BACKGROUND AND PURPOSE: Bone morphogenetic proteins (BMPs) were first identified through their role in inducing bone and cartilage formation, but many other important functions have since been ascribed to BMPs, including dorsoventral patterning, angiogenesis and tissue homeostasis. Using dorsomorphin and LDN193189, selective small molecule inhibitors of BMP signalling, we investigated the role of BMP signalling in early vascular patterning in zebrafish. EXPERIMENTAL APPROACH: The effects of dorsomorphin and LDN193189 on vascular endothelial growth factor-a (VEGF) and BMP signalling in developing zebrafish and in human pulmonary artery endothelial cells were determined using confocal microscopy, Western blotting and quantitative PCR. KEY RESULTS: We showed that dorsomorphin, similar to the VEGF inhibitor SU5416, strongly inhibits intersegmental vessel formation in zebrafish and that this is due to inhibition of VEGF activation of VEGF receptor 2 (VEGFR2), leading to reduced VEGF-induced phospho-ERK (extracellular regulated kinase) 1/2 and VEGF target gene transcription. These effects occurred at concentrations of dorsomorphin that block BMP signalling. We also showed that LDN193189, an analogue of dorsomorphin, more potently blocks BMP signalling but has no effect on VEGF signalling in zebrafish and does not disrupt early vascular patterning. CONCLUSIONS AND IMPLICATIONS: Dorsomorphin inhibits both BMP and VEGF signalling, whereas LDN193189 is a more selective BMP antagonist. Results obtained in cardiovascular studies using dorsomorphin need to be interpreted with caution, and use of LDN193189 would be preferable due to its selectivity. Our data also suggest that BMP signalling is dispensable for early patterning of intersegmental vessels in zebrafish.

Regulation of bone morphogenetic protein signalling in human pulmonary vascular development.

The bone morphogenetic protein (BMP) type II receptor (BMPR-II) is predominantly expressed on the vascular endothelium in the adult lung. Although mutations in BMPR-II are known to underlie many cases of familial pulmonary arterial hypertension (FPAH), little is known regarding the expression of BMPs and their signalling pathways during normal lung development or the impact of BMPR-II mutations on endothelial cell function. We determined the cellular localization and expression levels of BMP4, BMP receptors, and activation of downstream signalling via phospho-Smad1 in a developmental series of human embryonic and fetal lungs by immunohistochemistry. The expression of BMP4 and BMP receptors was temporally and spatially regulated during lung development. BMPR-II expression correlated with phosphorylation of tissue Smad1 and was highest during the late pseudoglandular and early canalicular stage of lung development, when vasculogenesis is intense. Phospho-Smad1 expression was associated with markers of proliferation in endothelial cells. In vitro studies confirmed that BMPs 2 and 4 induced phosphorylation of Smad1/5 and pulmonary artery endothelial cell (PAEC) migration and proliferation. Adenoviral transfection of PAECs with mutant kinase-deficient BMPR-II, or siRNA knockdown of BMPR-II, inhibited Smad signalling and the proliferative response to BMP4. Our findings support a critical role for BMPs in lung vasculogenesis. Dysfunctional BMP signalling in PAECs during development may lead to abnormal pulmonary vascular development and contribute to the pathogenesis of FPAH.

Differential adrenomedullin release and endothelin receptor expression in distinct subpopulations of human airway smooth-muscle cells.

Although adrenomedullin (ADM) is implicated in the control of airway tone, regulation of ADM release from airway smooth-muscle cells (ASMCs) has not been explored. Preliminary experiments have indicated that human ASMC populations were heterogeneous in their rate of ADM release and expression of endothelin (ET)(A) and ET(B) receptors. We isolated these phenotypically distinct ASMCs from explants derived from the same airway segment. ASMCs possessing exclusively ET(A) receptors appeared smaller and proliferated faster than ET(A)/ET(B) isolates. Macroautoradiographic analysis confirmed the presence of both receptors in human bronchi. ADM release and messenger RNA expression was greater in ET(A)/ET(B) isolates compared with ET(A) isolates. No measurable ET release was detected from ASMCs. Exogenous ET-1 (1 to 100 nM) more potently stimulated the release of ADM from ET(A)/ET(B) compared with ET(A) isolates. In addition, ET-3 (1 to 100 nM) stimulated ADM release only from ET(A)/ET(B) isolates, implicating the ET(B) receptor in this response. Exogenous ET-1 potentiated platelet- derived growth factor-stimulated [3H]thymidine uptake in ET(A)/ ET(B) but not ET(A) isolates. ET-3 did not affect [3H]thymidine uptake in either cell type. Possession of ET(A)/ET(B) receptors is associated with higher rates of ADM release and slower proliferation, but a capacity for ET-1 stimulated DNA synthesis via ET(A) receptors. These results support a paracrine role for the regulation of ADM release predominantly via the ET(B) receptor in human ASMCs.

Inactivation of platelet-derived growth factor-BB following modification by ADP-ribosyltransferase.

1. Arginine-specific ADP-ribosyltransferase (ART1) is expressed on the surface of a number of cell types, and catalyses the transfer of ADP-ribose from NAD(+) to target proteins. We investigated whether extracellular proteins such as growth factors may serve as substrates for this enzyme, with subsequent alteration in their biological activity. Experiments were performed with rat skeletal muscle membranes and V79 Chinese hamster lung fibroblasts with doxycycline-inducible expression of human ART. 2. From a panel of growth factors, platelet-derived growth factor-BB (PDGF-BB) was found to be the best substrate for ART1, whereas the structural homologue PDGF-AA was not a substrate. Under conditions of maximum labelling 5 mol ADP-ribose was incorporated per mol of PDGF-BB. 3. Purified (ADP-ribosyl)-PDGF-BB did not stimulate a mitogenic or chemotactic response in human pulmonary smooth muscle cells, and showed a reduced capacity to bind to PDGF receptors in competition binding experiments, when compared to unmodified PDGF-BB. 4. PDGF-dependent [(3)H-methyl]-thymidine incorporation was measured in the ART1-transfected fibroblast cell line at physiological concentrations of PDGF-BB, and without addition of extracellular NAD(+). Fibroblasts expressing human ART1 at the cell surface showed reduced mitogenic responses to PDGF-BB, but not to PDGF-AA. This loss of mitogenic response in cells expressing ART1 activity was reversed by the addition of agmatine (an ART1 substrate). 5. In conclusion, we propose that PDGF-BB-dependent signalling may be regulated by post-translational modification of the growth factor by ART1 at the cell surface. This has been demonstrated in membranes of rat skeletal muscle, and the reaction confirmed in ART1-transfected fibroblasts.

Altered growth responses of pulmonary artery smooth muscle cells from patients with primary pulmonary hypertension to transforming growth factor-beta(1) and bone morphogenetic proteins.

BACKGROUND: Mutations in the type II receptor for bone morphogenetic protein (BMPR-II), a receptor member of the transforming growth factor-beta (TGF-beta) superfamily, underlie many cases of familial and sporadic primary pulmonary hypertension (PPH). We postulated that pulmonary artery smooth muscle cells (PASMCs) from patients with PPH might demonstrate abnormal growth responses to TGF-beta superfamily members. METHODS AND RESULTS: For studies of (3)H-thymidine incorporation or cell proliferation, PASMCs (passages 4 to 8) were derived from main pulmonary arteries. In control cells, 24-hour incubation with TGF-beta(1) (10 ng/mL) or bone morphogenetic protein (BMP)-2, -4, and -7 (100 ng/mL) inhibited basal and serum-stimulated (3)H-thymidine incorporation, and TGF-beta(1) and BMPs inhibited the proliferation of serum-stimulated PASMCs. In contrast, TGF-beta(1) stimulated (3)H-thymidine incorporation (200%; P<0.001) and cell proliferation in PASMCs from PPH but not from patients with secondary pulmonary hypertension. In addition, BMPs failed to suppress DNA synthesis and proliferation in PASMCs from PPH patients. Reverse transcription-polymerase chain reaction of PASMC mRNA detected transcripts for type I (TGF-betaRI, Alk-1, ActRI, and BMPRIB) and type II (TGF-betaRII, BMPR-II, ActRII, ActRIIB) receptors. Receptor binding and cross-linking studies with (125)I-TGF-beta(1) confirmed that the abnormal responses in PPH cells were not due to differences in TGF-beta receptor binding. Mutation analysis of PASMC DNA failed to detect mutations in TGF-betaRII and Alk-1 but confirmed the presence of a mutation in BMPR-II in 1 of 5 PPH isolates. CONCLUSIONS: We conclude that PASMCs from patients with PPH exhibit abnormal growth responses to TGF-beta(1) and BMPs and that altered integration of TGF-beta superfamily growth signals may contribute to the pathogenesis of PPH.

Adrenomedullin expression and growth inhibitory effects in distinct pulmonary artery smooth muscle cell subpopulations.

The vasodilator peptide adrenomedullin is elevated in patients with pulmonary hypertension and has been implicated in the inhibition of vascular remodeling. We questioned whether adrenomedullin is released by human pulmonary artery smooth muscle cells (PASMCs) and inhibits PASMC growth and release of endothelin, a known smooth muscle cell mitogen. The majority of PASMCs isolated from proximal pulmonary arteries and all PASMCs from distal pulmonary arteries released adrenomedullin, although at differing rates (mean, 177 +/- 28 and 62 +/- 11 fmol/10(5) cells/24 h, respectively). These cells were designated ADM+. However, some proximal PASMC isolates did not release adrenomedullin, designated ADM-. Northern blot analysis confirmed adrenomedullin expression in proximal ADM+ but not ADM- isolates. ADM- and distal ADM+ PASMCs proliferated faster in serum than did proximal ADM+ cells. Adrenomedullin potently and dose-dependently (mean EC(50) = 2.2 +/- 0.5 nM) increased intracellular cyclic adenosine monophosphate (cAMP) in ADM- isolates via specific adrenomedullin receptors. In contrast, both adrenomedullin and calcitonin gene-related peptide modestly elevated cAMP in 50% of ADM+ isolates. Adrenomedullin dose-dependently inhibited platelet-derived growth factor-stimulated [3H]thymidine incorporation and endothelin release in ADM- cells but did not affect [3H]thymidine uptake in ADM+ isolates. We conclude that distinct subpopulations of human PASMCs release and respond to adrenomedullin. The heterogeneity of adrenomedullin release and the inhibition of PASMC DNA synthesis and endothelin release suggest that adrenomedullin may function as a paracrine mediator in the inhibition of pulmonary vascular remodeling.

Vascular remodeling and ET-1 expression in rat strains with different responses to chronic hypoxia.

Chronic hypoxia leads to a greater degree of pulmonary hypertension in the Wistar-Kyoto (WKY) rat than in the Fischer 344 (F-344) rat. We questioned whether this difference is associated with baseline differences in pulmonary artery anatomy, a greater degree of hypoxia-induced pulmonary vascular remodeling in the WKY rat, and/or differences in expression of endothelin (ET)-1. Male F-344 and WKY rats were maintained in normoxia or normobaric hypoxia for 21 days. Morphometry revealed that baseline pulmonary artery anatomy was similar in the two strains. However, during chronic hypoxia, the WKY rats developed a greater degree of muscularization of small pulmonary arteries. Baseline plasma and lung immunoreactive ET-1 levels were similar in the WKY and F-344 rats and increased significantly during hypoxia in the WKY rats. Northern analysis demonstrated increased lung preproET-1 mRNA during hypoxia in both strains, with a greater increase in WKY rats. Immunostaining demonstrated increased ET-1 in bronchial epithelium and peripheral pulmonary arteries during hypoxia, although to a greater degree in the WKY rats. We conclude that the WKY strain demonstrates increased susceptibility to hypoxia-induced pulmonary vascular remodeling compared with the F-344 strain and that increased lung and circulating ET-1 levels during hypoxia may partly explain this difference.

Prostacyclin analogues differentially inhibit growth of distal and proximal human pulmonary artery smooth muscle cells.

BACKGROUND: Prostacyclin has proved to be a beneficial treatment for patients with severe pulmonary hypertension. We postulated that the response may reflect, at least in part, inhibition of pulmonary artery smooth muscle cell (PASMC) growth. METHODS AND RESULTS: Human PASMCs were derived from distal (<1-mm external diameter, n=8) and proximal (>8-mm external diameter, n=12) pulmonary arteries obtained at transplant surgery and pneumonectomy. The effects of the stable prostacyclin analogues on [methyl-(3)H]thymidine incorporation and cell proliferation were investigated by using immunohistochemically characterized cells. Distal cells proliferated faster than did proximal PASMCs and displayed a distinct sensitivity to cicaprost and iloprost. Both analogues inhibited thymidine uptake over 24 hours (20% to 60%, P<0.001; n=8) and abolished stimulation of DNA synthesis by platelet-derived growth factor-BB (10 ng/mL) in distal but not proximal cells. The inhibitory effect of cicaprost was mimicked by isoproterenol (10(-5) mol/L), forskolin (10(-5) mol/L), and dibutyryl cAMP (5x10(-4) mol/L) and was potentiated by the phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (5x10(-5) mol/L). Cicaprost (10(-10) to 10(-6) mol/L) inhibited the proliferation of PASMCs, which had been stimulated with either platelet-derived growth factor-BB or serum, and increased cAMP production. These effects were potentiated by 3-isobutyl-1-methylxanthine and attenuated by the adenylyl cyclase inhibitor 2',5'-dideoxyadenosine (10(-5) to 10(-4) mol/L). CONCLUSIONS: ++Cicaprost and iloprost inhibit DNA synthesis and proliferation to a greater extent in distal compared with proximal human PASMCs, acting at least in part via a cAMP-dependent mechanism. The results are consistent with the hypothesis that prostacyclin analogues inhibit vascular remodeling in pulmonary hypertension and demonstrate heterogeneity among human PASMCs.

Angiotensin II activates MAPK and stimulates growth of human pulmonary artery smooth muscle via AT1 receptors.

To determine a potential role for the renin-angiotensin system in the growth of human pulmonary artery (PA) smooth muscle, we studied the localization of angiotensin (ANG) II-receptor subtypes by autoradiography in sections of human PA and in cultured PA smooth muscle cells (PASMCs) and examined the growth responses to ANG II in vitro. Specific 125I-labeled [Sar1,Ile8]ANG II binding was demonstrated within the pulmonary arterial media, but binding to cultured cells varied between isolates. Binding in tissues and cells was inhibited by the ANG II type 1 (AT1) receptor antagonist losartan but not by the type 2 (AT2) receptor antagonist PD-123319. Microautoradiographic studies indicated that cultured PASMCs exhibit heterogeneity with regard to ANG II binding sites. Addition of ANG II to serum-deprived PASMCs, exhibiting a relatively high level of 125I-[Sar1,Ile8]ANG II binding, led to a dose-dependent stimulation of DNA synthesis at 24 h and protein synthesis at 48 h. ANG II led to an increase in cell size without an increase in cell number. These effects were inhibited by losartan but not by PD-123319. In addition, ANG II led to rapid activation of mitogen-activated protein kinase (MAPK), and ANG II-stimulated DNA synthesis was inhibited by the specific inhibitor of MAPK PD-98059. We conclude that the AT1 receptor is expressed by human PASMCs in vivo and in vitro and is coupled to activation of MAPK and increased DNA and protein synthesis in vitro. These results are consistent with the hypothesis that ANG II may be involved in human pulmonary vascular remodeling.

Rat-2 fibroblasts express specific adrenomedullin receptors, but not calcitonin-gene-related-peptide receptors, which mediate increased intracellular cAMP and inhibit mitogen-activated protein kinase activity.

Rat-2 fibroblasts demonstrate specific binding of 125I-labelled rat adrenomedullin (KD=0.43 nM; Bmax=50 fmol/mg of protein) in the absence of 125I-labelled calcitonin-gene-related peptide (CGRP) binding. Therefore Rat-2 cells were used to examine the pharmacology and signal transduction pathways of adrenomedullin receptors. We examined the effects of adrenomedullin, the CGRP receptor antagonist CGRP-(8-37) and the amylin antagonists AC187 and AC253 on receptor binding and cAMP production. AC253, AC187 and CGRP-(8-37) inhibited 125I-adrenomedullin binding, with respective IC50 values of 25+/-8, 129+/-39 and 214+/-56 nM. Adrenomedullin dose-dependently increased intracellular cAMP (approximate EC50=1.0 nM). CGRP-(8-37), AC253 and AC187 antagonized adrenomedullin-stimulated cAMP production at micromolar concentrations. Using kinase-substrate assays, Mono Q FPLC and 'phospho-specific' Western blotting, we found that adrenomedullin alone abolished basal mitogen-activated protein kinase (MAPK) activity and dose-dependently inhibited platelet-derived-growth-factor-stimulated MAPK activity. Radioimmunoassay for adrenomedullin of media from Rat-2 cells showed a linear release of adrenomedullin-like immunoreactivity of 3.1 fmol/h per 2x10(6) cells. Gel-filtration chromatography showed that this adrenomedullin-like immunoreactivity co-eluted with synthetic rat adrenomedullin. Northern blotting with a rat adrenomedullin cDNA probe was used to confirm the presence of adrenomedullin mRNA. However, neither Northern blotting nor reverse transcriptase-PCR showed the presence of the cloned adrenomedullin receptor (L1). We conclude that the Rat-2 cell line expresses a specific adrenomedullin receptor (coupled to cAMP production and regulation of MAPK) and secretes adrenomedullin, which may participate in a regulatory control loop.

Expression of adrenomedullin (ADM) and its binding sites in the rat uterus: increased number of binding sites and ADM messenger ribonucleic acid in 20-day pregnant rats compared with nonpregnant rats.

RIA of nonpregnant rat uterus extracts showed 0.68 +/- 0.08 pmol/g adrenomedullin (ADM) and 3.23 +/- 0.08 pmol/g calcitonin gene-related peptide (CGRP). In the pregnant (20 days gestation) uterus, the ADM content was 0.90 +/- 0.17 pmol/g, and CGRP could not be detected. ADM messenger RNA was detected at high levels in the uterus, with a 1.8-fold increase in expression in pregnancy. Pharmacologically distinct binding sites for ADM (Bmax = 21 +/- 2 fmol/mg protein, dissociation constant = 80 +/- 6 pM), and CGRP (Bmax = 101 +/- 18 fmol/mg protein, dissociation constant = 140 +/- 20 pM) were identified in nonpregnant uterus. Competition for 125I[Tyr0]alphaCGRP binding was shown by both ADM and CGRP (8-37), whereas CGRP and CGRP (8-37) did not compete for 125I-ADM-binding sites. The density of the ADM-binding sites was 10 times greater in pregnant uterus (Bmax = 211 +/- 39 fmol/mg protein, P < 0.01) than nonpregnant uterus. CGRP receptor messenger RNA was identified in both nonpregnant and pregnant uteri. In isolated nonpregnant rat uteri, CGRP and ADM attenuated the contractile response to galanin by 77 +/- 10% and 57 +/- 10%, respectively. The responses to both CGRP and ADM were abolished by CGRP (8-37). These results demonstrate, for the first time, the presence of ADM and specific binding sites for both ADM and CGRP in the rat uterus.

Circulating adrenomedullin does not regulate systemic blood pressure but increases plasma prolactin after intravenous infusion in humans: a pharmacokinetic study.

Adrenomedullin has been proposed to be a circulating hormone regulating systemic and pulmonary blood pressure. A potential therapeutic role in the management of pulmonary hypertension has been suggested based on animal studies, but the pharmacokinetics and pharmacodynamics in human subjects have not been studied. We have infused adrenomedullin into volunteers at 3.2 pmol/kg.min, which more than quadrupled (52 pmol/L) normal circulating concentrations. At this dose no change in heart rate or blood pressure was noted. When infused at 13.4 pmol/kg.min to achieve a concentration over 40 times normal circulating levels (448 pmol/L), there was a significant fall in diastolic blood pressure from 69 +/- 2 to 53 +/- 2 mm Hg and a significant increase in pulse rate from 57 +/- 3 to 95 +/- 4 beats/min. Circulating PRL concentrations rose from 197 +/- 46 to 372 +/- 64 IU/L (mean +/- SEM; P < 0.01). No effect was seen on ACTH, TSH, FSH, LH, or cortisol. When the infusion was discontinued, baseline pulse and blood pressure were reestablished after 20 min. Adrenomedullin has a MCR of 27.4 +/- 3.6 mL/kg.min, with a circulating half life of 22 +/- 1.6 min and an apparent volume of distribution of 880 +/- 150 mL/kg. Column chromatography of plasma taken during infusion and decay of adrenomedullin showed no evidence of the production of additional molecular forms. These results are consistent with a peptide that is markedly tissue bound. Plasma adrenomedullin concentrations were increased in patients with renal impairment (14.1 +/- 0.9 pmol/L) compared to those in healthy volunteers (8.1 +/- 0.7 pmol/L), with a good correlation (r = 0.86) between circulating adrenomedullin and plasma creatinine. The circulating concentration of adrenomedullin necessary to affect blood pressure greatly exceeds that observed in healthy volunteers and in patients with a range of pathological conditions. Thus, adrenomedullin may be a paracrine regulator of vascular smooth muscle in humans.

Characterisation and molecular identification of adrenomedullin binding sites in the rat spinal cord: a comparison with calcitonin gene-related peptide receptors.

Calcitonin gene-related peptide (CGRP) and its receptors are found in mammalian spinal cord. We show, for the first time, binding sites for the novel related peptide adrenomedullin in rat spinal cord microsomes. 125I-Adrenomedullin binding showed high affinity (KD = 0.45 +/- 0.06 nM) and sites were abundant (Bmax = 723 +/- 71 fmol/mg of protein). CGRP, amylin, and calcitonin did not compete at these sites (Ki > 10 microM). High-affinity CGRP binding sites (KD = 0.18 +/- 0.01 nM) were much less numerous (Bmax = 17.7 +/- 2.4 fmol/mg of protein) and showed competition by unlabeled adrenomedullin (Ki = 34.6 +/- 2.4 nM). Chemical cross-linking revealed a major band for 125I-adrenomedullin of M(r) = 84,400 +/- 1,200 and a minor band of M(r) = 122,000 +/- 8,700. 125I-CGRP cross-linking showed bands of lower molecular weight (M(r) = 74,500 +/- 5,000 and 61,000 +/- 2,200). Enzymic deglycosylation of the adrenomedullin binding site showed a considerable carbohydrate content. Neither adrenomedullin nor CGRP was able to increase cyclic AMP in spinal cord. Adrenomedullin mRNA was present in spinal cord, at one-third of its level in lung, and adrenomedullin immunoreactivity was present, at a low concentration (40 fmol/g of tissue). Thus, the presence of abundant binding sites and adrenomedullin mRNA and immunoreactivity anticipate an as yet undefined function for this peptide in spinal cord.