Article Text


Pathways of vascular remodelling in PAH

Statistics from


1J. E. S. Park, 2P. D. Upton, 1P. M. de Souza, 2R. J. Davies, 2N. W. Morrell, 1M. J. D. Griffiths, 1S. J. Wort. 1Unit of Critical Care, NHLI, Imperial College, London, UK, 2University of Cambridge School of Clinical Medicine, Addenbrooke’s/CUHNHSFT and Papworth Hospitals, Cambridge, UK

Introduction Mutations in the bone morphogenetic protein (BMP) receptor II (BMPR-II), a member of the transforming growth factor β (TGFβ) receptor superfamily, are associated with familial pulmonary artery hypertension (PAH), whilst mutations of activin receptor-like kinase-1 (Alk-1) cause hereditary haemorrhagic telangiectasia (HHT). BMP9 binds BMPRII/Alk-1 complex in endothelial cells (ECs) and is reported to be a vascular quiescence factor.1 In addition, constitutively active Alk-1 in ECs induces endothelin-1 (ET-1) production, a contributor to vascular remodelling. Dysregulation of these pathways may therefore affect vascular integrity, as seen in both disease processes. We aimed to investigate the effect of BMP9 on ET-1 regulation.

Methods Human pulmonary artery endothelial cells (HPAECs) were treated with BMP9 (0–100 ng/ml) for up to 24 h in the presence/absence of mitogen-activated protein kinase (MAPK) pathway inhibitors (UO126, SB203580, SP600125; all 3 μM) and an inhibitor of IκB kinase 2 (IKK-2) phosphorylation (SC514, 3 μM), a component of the canonical nuclear factor-κB (NF-κB) pathway. MAPK phosphorylation was assessed by western analysis of whole-cell lysates. ET-1 gene expression and levels in supernatant were quantified.

Results BMP9 stimulated dose-dependent ET-1 release by HPAECs and increased ET-1 gene expression. A significant increase in ET-1 release was seen at physiological levels of BMP9 (1 ng/ml: p<0.001). ET-1 release was attenuated by inhibition of MAPK kinase 1 (MEK1)/extracellular signal-regulated kinase 1/2 (ERK1/2) and p38 phosphorylation (p<0.05 and p<0.01 respectively; see fig 1), but not c-Jun N-terminal kinase (JNK) and IKK-2 phosphorylation.

Conclusion BMP9 at physiological doses stimulates HPAECs to release ET-1 in an ERK/p38-dependent, SAMD-independent, manner. Hence BMP9, SMAD-independent, ET-1 production may have an important role in vascular dysregulation underlying PAH and HHT. Further work aims to investigate the effects of Alk-1/BMPRII deficiency on ET-1 regulation by BMP9 in HPAECs.

Abstract S111 Figure 1

The effect of inhibitors of mitogen-activated protein kinase kinase 1 (MEK1)/extracellular signal-regulated kinase 1/2 (ERK1/2)MEK1/ERK1/2 (UO126; 3 μM) and p38 (SB203580; 3 μM) phosphorylation on endothelin-1 (ET-1) release by bone morphogenetic protein 9 (BMP9; 1 ng/ml) stimulated human pulmonary artery endothelial cells (HPAECs). **p<0.01 (one-way analysis of variance (ANOVA)) compared with HPAECs incubated in 0.1% fetal calf serum (FCS) only; φp<0.05 and δp<0.01 compared with HPAECs treated with BMP9 only.



1A. Crosby, 1F. M. Jones, 1M. Southwood, 1S. Stewart, 2R. Schermuly, 3G. Butrous, 1D. W. Dunne, 1N. W. Morrell. 1University of Cambridge, Cambridge, UK, 2University of Giessen Lung Centre, Giessen, Germany, 3University of Kent, Kent, UK

Introduction Schistosomiasis is the most common worldwide cause of pulmonary hypertension. There is currently no established animal model to study the pathobiology of this important condition. We sought to develop and characterise a mouse model that recapitulates the pulmonary vascular pathology seen in patients with schistosomiasis.

Methods Mice were infected percutaneously with 30 Schistosoma mansoni cercariae and the development of lung and liver pathology was studied at 7, 12, 17 and 20 weeks postinfection. We determined the degree of right ventricular (RV) hypertrophy. Liver and lung egg counts were measured at specified time points. Pulmonary vascular remodelling was assessed by morphometry, following immunohistochemistry with antibodies for smooth muscle α-actin, von Willebrand factor, CD68 and CD45. A cytokine array was also performed.

Results Chronically infected animals had a relatively large lung egg burden with marked pulmonary vascular remodelling and perivascular inflammation. Morphometery revealed a significant increase in medial thickness in small and medium sized peripheral vessels and a significant increase in muscularisation in small vessels (20–70 mm). In addition, we observed the presence of plexiform-like lesions in these mice. Lung egg burden positively correlated with liver egg burden and the degree of RV hypertrophy. Plasma cytokines increased with time and correlated with the degree of pulmonary vascular remodelling. Mouse pulmonary artery smooth muscle cells showed an increase in migration to interleukin-13.

Conclusions This study provides evidence for extensive pulmonary vascular remodelling in a mouse model of pulmonary schistosomiasis, including the formation of plexiform-like lesions. The degree of remodelling and RV hypertrophy was related to the lung egg burden. This experimental model will be of use to study the pathobiology and treatment of pulmonary hypertension associated with schistosomiasis and will be valuable in studying the role of inflammation in pulmonary hypertension.


1P. D. Upton, 2J. E. S. Park, 2R. J. Davies, 2M. Griffiths, 2S. J. Wort, 2P. M. De Souza, 1N. W. Morrell. 1University of Cambridge School of Clinical Medicine, Addenbrooke’s/CUHNHSFT and Papworth Hospitals, Cambridge, UK, 2Unit of Critical Care, NHLI, Imperial College, London, UK

Introduction Bone morphogenetic protein 9 (BMP9) has recently been identified as the ligand for the activin receptor-like kinase-1 (ALK1)/BMP receptor II (BMPR-II) complex in endothelial cells. Mutations in BMPR-II underlie pulmonary arterial hypertension and mutations in ALK1 cause hereditary haemorrhagic telangiectasia or a syndrome comprising both disease phenotypes. Both diseases are associated with an unstable pulmonary vasculature susceptible to remodelling events. We recently reported that BMP9 induces interleukin-8 (IL-8) and E-selectin expression in human pulmonary artery endothelial cells (HPAECs), factors that can mediate inflammation but may also regulate vessel stability. Here, we examine the regulation of a similarly pleiotropic chemokine, monocyte chemoattractant protein-1 (MCP-1), by BMP9 in HPAECs.

Methods HPAECs were treated with M199/0.1% fetal bovine serum (FBS) with or without BMP9 (1 ng/ml). Lysates from cells treated for 1 h were analysed for Smad phosphorylation with western blotting. RNA was extracted from cells treated for 8 h and quantitative PCR used to determine mRNA expression. Supernatants were collected from cells treated for 24 h and analysed for MCP-1 and IL-8 using ELISA. To assess the contribution of Smads, cells were transfected with small interfering RNA (siRNA) for Smad2 or Smad4, and RNA extracted from cells treated for 8 h for expression analysis.

Results BMP9 induced phosphorylation of Smad1/5 and Smad2 in HPAECs. BMP9 induced the expression and secretion of IL-8 by HPAECs, whereas BMP9 potently repressed MCP-1 expression and release. The IL-8 induction by BMP9 was reversed by siRNA for Smad4 and Smad2, whereas the repression of MCP-1 was only reversed by Smad4 siRNA.

Conclusions BMP9 induces IL-8 and represses MCP-1 in HPAECs, suggesting that BMP9 is not eliciting an inflammatory response but is discretely repressing specific inflammatory responses or signals that alter vascular structure. Loss of ALK1/BMPR-II functions may lead to dysregulated BMP9-mediated chemokine responses in HPAECs and promote a more inflammatory state and altered vascular integrity.


1J. E. S. Park, 2A. R. Lyon, 1L. R. Hector, 1M. J. D. Griffiths. 1Unit of Critical Care, NHLI, Imperial College, London, UK, 2Department of Cardiac Medicine, NHLI, Imperial College, London, UK

Introduction Irreversible alveolar capillary membrane (ACM) remodelling accompanies chronic heart failure (CHF), contributing to dyspnoea, the predominant symptom that limits quality of life in CHF. Monocyte chemoattractant protein-1 (MCP-1) is a monocyte chemokine which also modulates cell fate and stimulates collagen expression by fibroblasts. Plasma levels are increased in animal models of CHF and other pulmonary pathologies involving ACM remodelling, and correlate with mortality in patients with CHF.1 We hypothesised that mechanical strain at the pulmonary microvasculature associated with raised left atrial pressure upregulates MCP-1 activity leading to pulmonary inflammation and ACM remodelling.

Methods and Results Cyclic mechanical strain (CMS: 30%; 0.3 Hz; 4 h, Flexercell 4000X) applied to EA.hy 926 and human lung microvascular endothelial cells (HLMVECs: Promocell) grown on collagen-coated flexible membranes (Bioflex) upregulated MCP-1 gene expression and protein release via extracellular signal-regulated kinase (ERK) 1/2- and c-Jun N-terminal kinase (JNK) 1/2-dependent pathways with associated phosphorylation of these mitogen-activated protein kinases (MAPKs) (fig 1). Supernatants from stretched but not unstretched cells increased markers of myofibroblast (human fetal lung fibroblast-1 cells) and pulmonary vascular smooth muscle differentiation (α-smooth muscle actin and vimentin expression), and monocyte (THP-1) migration in vitro. Lungs from adult male Sprague–Dawley rats at 16 weeks after left coronary artery ligation with haemodynamic evidence of left ventricular dysfunction2 were less compliant, showed evidence of airspace leucocytosis on bronchoalveolar lavage (BAL) and histology, but not pulmonary oedema (wet:dry ratio: p = 0.74, n = 10). MCP-1 gene expression was increased in whole lung (p<0.05, n = 5); MCP-1 protein levels were increased in BAL (16.9 (4.7) vs 7.51 (2.1), mean (SEM), n = 4) and serum (233.6 (8.99) vs 150.4 (19.99), n = 3) in the CHF model animals compared with sham-operated controls.

Conclusion Mechanical strain at the pulmonary microvasculature increased expression of MCP-1 via ERK- and JNK-dependent pathways which may decrease lung compliance and impair gas exchange by inducing inflammation and remodelling of the ACM. MCP-1, or the induction of MCP-1 by CMS, may provide an attractive therapeutic target for sufferers of dyspnoea in CHF.

Abstract S114 Figure 1

Effect of mitogen-activated protein kinase (MAPK) pathway inhibitors on stretch-induced monocyte chemoattractant protein-1 (MCP-1) release by EA.hy 926 cells. **p<0.001 compared with non-stretched cells (one-way analysis of variance (ANOVA)); ϕp<0.01 (one-way ANOVA) compared with cells stretched in the absence of MAPK inhibitor.



1R. Condliffe, 2A. G. Hameed, 2C. Paiva, 3J. Suntharalingam, 2S. E. Francis, 2C. M. H. Newman, 2D. C. Crossman, 3N. W. Morrell, 2A. Lawrie. 1Royal Hallamshire Hospital, Sheffield, UK, 2University of Sheffield, Sheffield, UK, 3University of Cambridge, Cambridge, UK

Background and Objectives Despite improvements in the overall management of pulmonary arterial hypertension (PAH) the disorder still causes significant morbidity and mortality. Current treatments fail to reverse the disease, and clinical assessment does not always differentiate between, or reflect the local pathogenesis within, the heart or pulmonary circulation. Current proposed biomarkers—for example, brain natriuretic peptide (BNP and NT-proBNP)—mostly measure cardiac effects rather than the pulmonary circulation. Subsequently, there has been increasing interest in identifying a biomarker for PAH that tracks with lung pathology and treatment. Through our desire to understand disease pathogenesis, our studies in vitro and in animal models have identified two candidate molecules osteoprotegerin (OPG) and tumour necrosis factor-related apoptosis-inducing ligand (TRAIL). The aim of this study was to investigate the potential of OPG and TRAIL as biomarkers in idiopathic pulmonary arterial hypertension (IPAH) and chronic thromboembolic pulmonary hypertension (CTEPH).

Methods Serum samples for OPG and TRAIL were obtained from 35 patients with IPAH, 37 patients with CTEPH (before and after pulmonary artery endarterectomy (PEA)) and 39 controls. Serum concentrations were compared and correlations with pulmonary haemodynamics and prognostic significance were assessed.

Results OPG concentrations were significantly elevated in IPAH (median 4505 pg/ml) when compared with CTEPH (1528 pg/ml) and controls (1241 pg/ml). TRAIL concentrations were significantly decreased in IPAH (median 39.4 pg/ml) when compared with CTEPH (107 pg/ml) and controls (146.8 pg/ml). Concentrations of OPG and TRAIL and the OPG:TRAIL ratio correlated with pulmonary haemodynamics in IPAH (OPG:TRAIL vs cardiac index r = −0.59); however, no correlations existed in CTEPH. OPG concentration above a threshold of 5000 pg/ml predicted poorer survival. OPG was significantly lower in patients with CTEPH at 3 months following PEA but there was no significant difference in the levels of TRAIL.

Conclusion PAH is characterised by elevated OPG and decreased TRAIL serum concentrations. Both correlate with pulmonary haemodynamics while OPG concentrations may have prognostic significance. The difference in OPG and TRAIL concentrations between IPAH and CTEPH may reflect a much greater degree of distal pulmonary vascular remodelling in IPAH. The OPG–TRAIL pathway has potential as a biomarker in PAH and merits further investigation; a prospective study is currently underway.


1H. J. Durrington, 1P. Upton, 1T. Criley, 1B. Dunmore, 2S. Hoer, 2P. J. Lehner, 2J. Boname, 3L. Butler, 3D. Blackbourne, 1N. W. Morrell. 1Department of Medicine, University of Cambridge School of Clinical Medicine, Box 157, Addenbrooke’s Hospital, Cambridge, UK, 2Cambridge Institute for Medical Research, University of Cambridge, Cambridge, UK, 3Cancer Research UK Institute for Cancer Studies, University of Birmingham, Birmingham, UK

Bone morphogenetic proteins (BMPs) are critically involved in cell differentiation, oncogenesis and early development. In man, dysfunction of the BMP type II receptor (BMPR-II) is associated with pulmonary arterial hypertension (PAH). Since Kaposi sarcoma-associated herpesvirus (KSHV) may be an aetiological factor in the pathogenesis of PAH, we sought to determine whether KSHV influences BMPR-II function. Here, we show that KSHV infection reduces the cell surface localisation of BMPR-II and induces expression of the viral lytic gene, K5, a ubiquitin E3 ligase. Heterologous expression of K5 leads to ubiquitination of BMPR-II and its lysosomal degradation (fig 1), an effect dependent on both the RING domain of K5 and a membrane proximal lysine, in the cytoplasmic domain of BMPR-II. These findings suggest that disruption of BMP signalling may play a role in the pathobiology of diseases caused by KSHV and provide an important association between KSHV and PAH.

Moreover, we demonstrate the presence of an endogenous lysosomal pathway important in the regulation of mammalian BMPR-II. We show that members of the Nedd4 family of E3 ligases may regulate endogenous BMPR-II through this lysosomal pathway.

Abstract S116 Figure 1

BMPR-II is located at the cell surface in HeLa cells and is internalised into discrete, cytoplasmic compartments in HeLa-K5 cells. Using a 5′myc-tagged bone morphogenetic protein type II receptor (BMPR-II) construct, HeLa and HeLa-K5 cells were transfected, then stained with a myc antibody and a secondary antibody conjugated to fluorescein isothiocyanate (FITC; green). In HeLa cells BMPR-II is localised at the cell membrane (A), but in HeLa-K5 cells BMPR-II is internalised into discrete collections within the cytoplasm (B). Co-staining with the lysosomal marker, lysosomal-associated membrane protein 1 (LAMP-1; red) produces a yellow colour in HeLa-K5 cells, confirming that BMPR-II is internalised from the cell surface to a lysosomal compartment in the presence of K5. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI; blue). GFP, green fluorescent protein.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.