Cellular and molecular pathobiology of pulmonary arterial ... · pulmonary hypertension is the...

CoMKBNCV

Tsunpevpia

wv

MUMHHECLGVbVMC§C

Journal of the American College of Cardiology Vol. 43, No. 12 Suppl S© 2004 by the American College of Cardiology Foundation ISSN 0735-1097/04/$30.00Published by Elsevier Inc. doi:10.1016/j.jaac.2004.02.029

ellular and Molecular Pathobiologyf Pulmonary Arterial Hypertensionarc Humbert, MD, PHD,* Nicholas W. Morrell, MD,† Stephen L. Archer, MD,‡

urt R. Stenmark, MD,§ Margaret R. MacLean, BSC, PHD,� Irene M. Lang, MD,¶rian W. Christman, MD,# E. Kenneth Weir, MD,** Oliver Eickelberg, MD,††orbert F. Voelkel, MD,‡‡ Marlene Rabinovitch, MD§§lamart, France; Cambridge and Glasgow, United Kingdom; Edmonton, Canada; Denver, Colorado;ienna, Austria; Nashville, Tennessee; Minneapolis, Minnesota; Giessen, Germany; and Stanford, California

Pulmonary arterial hypertension (PAH) has a multifactorial pathobiology. Vasoconstriction,remodeling of the pulmonary vessel wall, and thrombosis contribute to increased pulmonaryvascular resistance in PAH. The process of pulmonary vascular remodeling involves all layersof the vessel wall and is complicated by cellular heterogeneity within each compartment of thepulmonary arterial wall. Indeed, each cell type (endothelial, smooth muscle, and fibroblast),as well as inflammatory cells and platelets, may play a significant role in PAH. Pulmonaryvasoconstriction is believed to be an early component of the pulmonary hypertensive process.Excessive vasoconstriction has been related to abnormal function or expression of potassiumchannels and to endothelial dysfunction. Endothelial dysfunction leads to chronicallyimpaired production of vasodilators such as nitric oxide and prostacyclin along withoverexpression of vasoconstrictors such as endothelin (ET)-1. Many of these abnormalitiesnot only elevate vascular tone and promote vascular remodeling but also represent logicalpharmacological targets. Recent genetic and pathophysiologic studies have emphasized therelevance of several mediators in this condition, including prostacyclin, nitric oxide, ET-1,angiopoietin-1, serotonin, cytokines, chemokines, and members of the transforming-growth-factor-beta superfamily. Disordered proteolysis of the extracellular matrix is also evident inPAH. Future studies are required to find which if any of these abnormalities initiates PAHand which ones are best targeted to cure the disease. (J Am Coll Cardiol 2004;43:13S–24S)© 2004 by the American College of Cardiology Foundation

nlgifip

C

Sdv(tnriipmttearmi

he pulmonary circulation is a low-pressure, high-flowystem with a great capacity for recruitment of normallynperfused vessels. As a consequence, the walls of pulmo-ary arteries are thin, in keeping with their low transmuralressure. Pulmonary arterial hypertension (PAH) is a dis-ase of the small pulmonary arteries, characterized byascular narrowing leading to a progressive increase inulmonary vascular resistance. The consequence of thisncreased right ventricle afterload is the failure of thefterload-intolerant right ventricle.

Vasoconstriction, remodeling of the pulmonary vesselall, and thrombosis contribute to the increased pulmonaryascular resistance in PAH (1). However, it is now recog-

From the *Service de Pneumologie et Reanimation Respiratoire, Centre desaladies Vasculaires Pulmonaires, UPRES EA2705, Hopital Antoine-Beclere,niversite Paris-Sud, Clamart, France; †Respiratory Medicine Unit, Department ofedicine, University of Cambridge School of Clinical Medicine, Addenbrooke’sospital, Cambridge, United Kingdom; ‡Vascular Biology Group and Pulmonaryypertension Program, Department of Medicine (Cardiology), University of Alberta,dmonton, Alberta, Canada; §Development Lung Biology Research, University ofolorado Health Sciences Center, Denver, Colorado; �Division of Biomedical andife Sciences, Institute of Biomedical and Life Sciences, Glasgow University,lasgow, United Kingdom; ¶Department of Cardiology, University of Vienna,ienna, Austria; #Center for Lung Research, Department of Pharmacology, Vander-ilt University School of Medicine, Nashville, Tennessee; **Department of Medicine,eterans Affairs Medical Center, Minneapolis, Minnesota; ††Department of Internaledicine, Justus-Liebig University, Giessen, Germany; ‡‡Pulmonary Hypertensionenter, University of Colorado Health Sciences Center, Denver, Colorado; and§Department of Pediatrics, Stanford University School of Medicine, Stanford,alifornia.

uManuscript received January 7, 2004; accepted February 3, 2004.

ized that pulmonary arterial obstruction by vascular pro-iferation and remodeling is the hallmark of PAH patho-enesis (2). The process of pulmonary vascular remodelingnvolves all layers of the vessel wall and is complicated by thending that cellular heterogeneity exists within the com-artment of the pulmonary arterial wall (2).

ELLULAR CHANGES

mooth muscle cells and fibroblasts. Each cell type (en-othelial, smooth muscle, and fibroblast) in the pulmonaryascular wall plays a specific role in the response to injury2). A feature common to all forms of PAH remodeling ishe distal extension of smooth muscle into small peripheral,ormally nonmuscular, pulmonary arteries within the respi-atory acinus. The cellular processes underlying muscular-zation of this distal part of the pulmonary arterial tree arencompletely understood. In addition, a hallmark of severeulmonary hypertension is the formation of a layer ofyofibroblasts and extracellular matrix between the endo-

helium and the internal elastic lamina, termed the neoin-ima. In some model systems, particularly in hypoxia mod-ls, the adventitial fibroblast appears to be the first cellctivated to proliferate and to synthesize matrix proteins inesponse to the pulmonary hypertensive stimulus (3). Theechanisms that enable the adventitial fibroblast to migrate

nto the media (and ultimately the intima) are currently
nclear, but there is good evidence to suggest that upregu-

lo

wlotttuttmebfrbaoEtmtkos

hpcppopcsmgalTh

ss

seadpiamstsTasgllieiiefIptPt(wiiisppcpalPRowpPd(oetP

14S Humbert et al. JACC Vol. 43, No. 12 Suppl SPathobiology of Pulmonary Arterial Hypertension June 16, 2004:13S–24S

ation of matrix metalloproteinases (MMP2 and MMP9)ccurs and that these molecules are involved in migration.In many forms of pulmonary hypertension, as the vessel

all thickens, a concomitant increase occurs in neovascu-arization of the vasa vasorum. This neovascularizationccurs primarily in the adventitia, although it extends intohe outer parts of the media. This adventitial vessel forma-ion could provide a conduit for circulating progenitor cellso access the vessel wall from the adventitial side. It isnknown whether circulating progenitor cells derived fromhe bone marrow contribute directly to the adventitialhickening and perhaps medial thickening, or whether bonearrow-derived progenitor cells simply enhance the prolif-

rative and migratory activity of the local adventitial fibro-lasts. Significant attention in the future will have to beocused on the role of circulating precursor cells to vascularemodeling (4). In addition, evidence shows that PAH maye associated with alterations of both rates of proliferationnd apoptosis, which, in balance, result in thickened,bstructive pulmonary arteries.ndothelial cells. Disorganized endothelial cell prolifera-

ion leading to formation of plexiform lesions is described inany cases of PAH (1,5). The initiating stimulus or injury

hat results in abnormal endothelial proliferation is un-nown, but may include hypoxia, shear stress, inflammation,r response to drugs or toxins on a background of geneticusceptibility.

However, although many animal models of pulmonaryypertension exist, none recapitulate the histology of PAH,articularly the presence of plexiform lesions. Endothelialells may respond to injury in various ways affecting therocess of vascular remodeling. Injury can alter not only cellroliferation and apoptosis but also homeostatic functionsf the endothelium (including coagulation pathways, androduction of growth factors and vasoactive agents). Theells comprising plexiform lesions are endothelial channelsupported by a stroma containing matrix proteins andyofibroblasts. Endothelial cells express markers of angio-

enesis, such as vascular endothelial growth factor (VEGF)nd its receptors (5). In addition, cells comprising plexiformesions of idiopathic PAH are monoclonal in origin (6).herefore, although the lesions themselves are probably

Abbreviations and AcronymsBMP � bone morphogenetic proteinsET � endothelin5-HT � 5-hydroxytryptamineNO � nitric oxidePAH � pulmonary arterial hypertensionTGF-� � transforming growth factor-betaTGF-�R2 � transforming growth factor-beta type-2

receptorVEGF � vascular endothelial growth factorVIP � vasoactive intestinal peptide

emodynamically irrelevant, they may represent more than fi

imply the result of severe elevation of intravascular pres-ures.

It has been suggested that the endothelial proliferationeen in these lesions may be a marker of a fundamentalndothelial abnormality in idiopathic PAH, possibly playingkey role in the pathogenesis of the condition. Intriguing

efects in growth suppressive genes have been reported inlexiform lesions of patients with idiopathic PAH, includ-ng transforming growth factor-beta (TGF-�) receptor-2nd the apoptosis-related gene, Bax (7). Thus, in approxi-ately 30% of plexiform lesions there is a somatic frame-

hift mutation in the transforming growth factor-betaype-2 receptor (TGF-�R2) gene encoding a prematuretop codon. Furthermore, in 90% of plexiform lesions theGF-�R2 protein is not expressed, in contrast to the

bundant expression in endothelial cells outside these le-ions. Thus, it has been proposed that somatic mutations inrowth regulatory genes allow clonal expansion of endothe-ial cells, that contribute to the formation of plexiformesions and vascular obliteration (7). Human herpesvirus-8nfection may also contribute to the growth of monoclonalndothelial cells in plexiform lesions from patients withdiopathic PAH (8). These findings suggest that triggers,ncluding vasculotropic viruses, can encourage the growth ofndothelial cells by dysregulating cell growth or growth-actor signaling.nflammatory cells. Inflammatory mechanisms appear tolay a significant role in some types of pulmonary hyper-ension including monocrotaline-induced cases in rats andAH of various origins in humans including connective

issue diseases and human immunodeficiency virus infection9). Interestingly, some patients with severe PAH associatedith systemic lupus erythematosus have improved with

mmunosuppressive therapy, emphasizing the relevance ofnflammation in this subset of patients (9). Patients withdiopathic PAH also have some immunological disturbancespeaking in favor of a possible role for inflammation in theathophysiology of this disease. Indeed, a subset of PAHatients have circulating autoantibodies including antinu-lear antibodies, as well as elevated circulating levels ofroinflammatory cytokines IL-1 and IL-6. Lung histologylso revealed inflammatory infiltrates (macrophages andymphocytes) in the range of plexiform lesions in severeAH as well as an increased expression of chemokinesANTES and fractalkine (10). Further analysis of the rolef inflammatory mechanisms is necessary to understandhether this component of the disease is relevant to itsathophysiology.latelets and thrombosis. Thrombotic lesions and plateletysfunction are potentially important processes in PAH11). In situ pulmonary artery thrombosis may be initiatedr aggravated by abnormalities in the clotting cascade, thendothelial cells, or the platelets. Biologic evidence showshat intravascular coagulation is a continuous process inAH patients, characterized by elevated plasma levels of
brinopeptide A- and D-dimers. In addition, procoagulant

altbtistp

tmsmpirtapVtq

M

Ppvedc(ooegeaPPpcwsPP

Fppg

15SJACC Vol. 43, No. 12 Suppl S Humbert et al.June 16, 2004:13S–24S Pathobiology of Pulmonary Arterial Hypertension

ctivity and fibrinolytic function of the pulmonary endothe-ium are altered in PAH. This dysfunction is reflected byhe demonstration of elevated plasma levels of von Wille-rand factor and plasminogen activator inhibitor type-1. Athe present time, it is widely accepted that the shear stresstself or injury of the lung vessels generates a thrombogenicurface with subsequent thrombotic lesions. Thus, it appearshis prothrombotic diathesis is shared by many forms ofulmonary hypertension, and is not unique to PAH.Moreover, an increasing body of evidence also suggests

hat enhanced interactions between platelets and the pul-onary artery wall may contribute to the functional and

tructural alterations of pulmonary vessels. Vascular abnor-alities in PAH may lead to release by platelets of various

rocoagulant, vasoactive, and mitogenic mediators. Indeed,n addition to its role in coagulation, the platelet stores andeleases important contributors to pulmonary vasoconstric-ion and remodeling such as thromboxane A2, platelet-ctivating factor, serotonin (5-hydroxytryptamine [5-HT]),latelet-derived growth factor (PDGF), TGF-�, andEGF. In most cases, however, it remains unclear whether

hrombosis and platelet dysfunction are causes or conse-

igure 1. Consequences of pulmonary artery endothelial cell dysfunctionulmonary artery endothelial cells (blue) have a decreased production ofromoting vasoconstriction and proliferation of pulmonary artery smoothuanosine monophosphate; ET � endothelin; ETA � endothelin recepto

uences of the disease (11). c

OLECULAR MECHANISMS

ulmonary vasoconstriction is believed to be an early com-onent of the pulmonary hypertensive process. Excessiveasoconstriction has been related to abnormal function orxpression of potassium channels, as well as to endothelialysfunction (1,2). Endothelial dysfunction leads to chroni-ally impaired production of vasodilators such as nitric oxideNO) and prostacyclin along with prolonged overexpressionf vasoconstrictors such as endothelin (ET)-1, which notnly affect vascular tone, but also promote vascular remod-ling and, therefore, represent logical pharmacological tar-ets (Fig. 1). It appears that most stimuli that acutelynhance vasoconstriction ultimately also cause cell prolifer-tion (e.g., K� channel inhibition, ET-1).rostacyclin, vasoactive intestinal peptide, and NO.rostacyclin (prostaglandin I2) is an important endogenousulmonary vasodilator acting through activation of theyclic adenosine monophosphate (cAMP)-dependent path-ays. Prostacyclin also inhibits the proliferation of vascular

mooth muscle cells and decreases platelet aggregation.rostacyclin synthesis is decreased in endothelial cells fromAH patients. Analysis of urinary metabolites of prostacy-

lmonary artery smooth muscle cell tone and proliferation. Dysfunctionalacyclin and nitric oxide, with an increased production of endothelin-1-e cells (red). cAMP � cyclic adenosine monophosphate; cGMP � cyclicTB � endothelin receptor B; PDE5 � phosphodiesterase type 5.

on puprost

musclr A; E

lin showed a decrease in the amount of excreted

6cpiads

psndmgss

mspptclctnwrpwcbpE(tapiapEtmaseleaocti

aPPlpcpiiplctsKh

hlpitcqdKnrmcmpccSv

FAcati


-ketoprostaglandin F1�, a stable metabolite of prostacy-lin, in patients with idiopathic PAH (12). Moreover,ulmonary endothelial cells of PAH patients are character-zed by reduced expression of prostacyclin synthase (13),nd prostacyclin therapy has been shown to improve hemo-ynamics, clinical status, and survival of patients displayingevere PAH.

Impaired endothelium-derived vasodilation is further sup-orted by the demonstration of reduced NO synthase expres-ion in pulmonary endothelial cells from PAH patients (14). Aovel therapeutic strategy in PAH aims at increasing NO-ependent, cyclic guanosine monophosphate-mediated pul-onary vasodilation by inhibition of the breakdown of cyclic

uanosine monophosphate by phosphodiesterase type-5. In amall group of PAH patients, sildenafil has been shown to beafe and effective on a chronic basis (15).

Vasoactive intestinal peptide (VIP), a neuropeptide pri-arily functioning as a neurotransmitter, acts as a potent

ystemic and pulmonary vasodilator. It also inhibits theroliferation of vascular smooth muscle cells and decreaseslatelet aggregation; VIP acts through two receptor sub-ypes (VPAC-1 and -2), which are coupled to adenylateyclase and expressed in the lung vasculature (16). Stimu-ation of VPAC receptors leads to the activation of theAMP and cyclic guanosine monophosphate (cGMP) sys-ems. Low serum concentrations and decreased VIP immu-oreactivity were shown in pulmonary arteries from patientsith idiopathic PAH. In addition, higher expression of VIP

eceptors and elevated specific receptor binding activity inulmonary artery smooth muscle cells from idiopathic PAHas demonstrated, which presumably reflects VIP defi-

iency. Acute and chronic responses to inhaled VIP haveeen recently demonstrated in a small number of PAHatients (16).T-1. Through its action on the endothelin receptor A

ETA) in pulmonary artery smooth muscle cells, ET-1 leadso a rapid increase in intracellular calcium and sustainedctivation of protein kinase C. Early activation of the42/p44 isoforms of mitogen-activated protein kinase andnduction of the early growth response genes c-fos and c-junre also observed (2). The mitogenic action of ET-1 onulmonary artery smooth muscle cells occurs through theTA or endothelin receptor B (ETB) subtype, depending on

he anatomic location of cells. For instance, ETA mediateitogenesis in cells derived from the main pulmonary

rtery, whereas in cells from resistance arteries both receptorubtypes may contribute. There is strong evidence thatndothelium-derived ET-1 is a major player in the vasodi-ator/vasoconstrictor imbalance characteristic of PAH. Lev-ls of lung and circulating ET-1 are increased in animalsnd patients with pulmonary hypertension of various etiol-gies (17). These observations indicate that ET-1 is likely toontribute to the vasoactive component of PAH, as well aso the abnormal pulmonary vascular remodeling character-
stic of the condition. Results of chronic ET receptor s
ntagonist therapy support the relevance of this pathway inAH.otassium channels. Lessons relevant to PAH can be

earned from understanding the mechanism of hypoxiculmonary vasoconstriction, although PAH also involvesell proliferation and abnormalities of apoptosis (18). Hy-oxic pulmonary vasoconstriction is elicited when hypoxianhibits one or more voltage-gated potassium channels (Kv)n the pulmonary artery smooth muscle cells of resistanceulmonary arteries (Fig. 2). The resulting membrane depo-

arization increases the opening of voltage-gated calciumhannels, raising cytosolic calcium and initiating constric-ion. The Kv1.5 is downregulated in pulmonary arterymooth muscle cells in humans with PAH (19), and bothv1.5 and Kv2.1 are downregulated in rats with chronicypoxia-induced pulmonary hypertension (20).Furthermore, deoxyribonucleic acid microarray studies

ave shown downregulation Kv channel genes in PAHungs (21). The selective loss of these Kv channels leads toulmonary artery smooth muscle cell depolarization, anncrease in the intracellular calcium, and both vasoconstric-ion and cell proliferation. It is not clear whether these Kvhannel abnormalities are genetically determined or ac-uired. However, it is clear that the appetite suppressantsexfenfluramine and aminorex directly inhibit Kv1.5 andv2.1 (22). Augmenting Kv pathways should cause pulmo-ary vasodilation and promote regression of pulmonaryemodeling. Drugs including dichloroacetate and sildenafilay enhance the expression and function of these potassium

hannels. Most of the hemodynamic effects of NO areediated by cGMP, which causes vasodilation by activating

rotein kinase G, which phosphorylates and activates BKCAhannels, as one of several mechanisms by which it lowersytosolic calcium.erotonin. In PAH, circulating serotonin levels are ele-ated, whereas the level in platelets, the major repository of

igure 2. The balance of tone in pulmonary artery smooth muscle cells.ctivity of voltage-gated potassium channels (Kv) in the smooth muscle

ells of resistance pulmonary arteries affects vascular tone. In pulmonaryrterial hypertension, the selective loss of Kv channels such as Kv1.5 leadso pulmonary artery smooth muscle cell depolarization, an increase in thentracellular calcium, and both vasoconstriction and cell proliferation.

erotonin (5-hydroxytryptamine [5-HT]), is low (23).

5tbspwivPtreaiitctwrrwa5d

avmcmp5aTg2pibomaFrp

mi

Foa


-Hydroxytryptamine is produced by the gastrointestinalract enterochromaffin cells and pulmonary neuroepithelialodies and stored in platelets. A role for 5-HT has beenuggested in PAH (23,24). First, a correlation between highlasma 5-HT levels and PAH was observed in a patientith congenital thrombocytopathy characterized by a defect

n the platelet 5-HT storage capacities. Subsequently, ele-ated plasma 5-HT levels were demonstrated in a series ofAH patients (23). This could not be corrected by lung

ransplantation or epoprostenol therapy, indicating thataised plasma 5-HT cannot be the mere consequence oflevated pulmonary pressure (23). In the 1960s, an associ-tion between PAH and the anorexigen aminorex wasdentified. Aminorex induces platelet 5-HT release andnhibits monoamine oxidase, potentially inhibiting its me-abolism, thus increasing plasma 5-HT levels. More re-ently, it was shown that exposure to fenfluramine deriva-ives increased the risk of developing PAH. By interactingith the 5-HT transporter (5-HTT), these anorexigens

elease 5-HT from platelets and inhibit its reuptake andaise circulating free 5-HT. Additionally, treatment of ratsith 5-HT potentiates the effects of hypoxia on pulmonary

rterial pressure and remodeling. The mechanism by which-HT affects the pulmonary vasculature is still a matter of

igure 3. Serotonin receptors and transporter in pulmonary artery smooth mr both in pulmonary artery smooth muscle cells contribute to pulmonaryrtery smooth muscle cells. A role for other 5-hydroxytryptamine (5-HT)

ebate (Fig. 3). h

The 5-HTT expression, activity, or both in pulmonaryrtery smooth muscle cells contribute to the pulmonaryascular remodeling occurring in both clinical and experi-ental PAH (24); 5-HTT is encoded by a single gene on

hromosome 17q11.2, and a variant in the upstream pro-oter region of the 5-HTT gene has been described. This

olymorphism with long (L) and short (S) forms affects-HTT expression and function with the L-allele inducinggreater rate of 5-HTT gene transcription than the S-allele.he L-allelic variant was found to be present in homozy-ous forms in 65% of idiopathic PAH patients but only in7% of controls (25). Moreover the 5-HTT gene polymor-hism contributes to interindividual differences in hypoxia-nduced 5-HTT expression and potentially affects suscepti-ility to hypoxic pulmonary hypertension (26). Miceverexpressing the 5-HTT gene exhibit spontaneous pul-onary hypertension in absence of hypoxia (and exagger-

ted pulmonary hypertension after hypoxic exposure) (27).inally, recent studies have shown that selective serotonin

euptake inhibitors protect against hypoxic pulmonary hy-ertension in mice (28).In human large pulmonary arteries, the 5-HT1 receptorediates 5-HT–induced contraction. Further investigation

dentified the 5-HT1B as that mediating contraction in

cells. The 5-hydroxytryptamine transporter (5-HTT) expression, activity,lar remodeling. The 5-HT1B mediates contraction in human pulmonarytors such as 5-HT2A and 5-HT2B has also been suggested.

usclevascurecep

uman small muscular pulmonary arteries (29). In addition,

tincr5si5h5khrnti

sflis5p

hccgrhpvTpi((vt(1BsfwrP

Fva


here is an increase in the expression of the 5-HT1B receptorn PAH. Contractile responses to 5-HT in the rat pulmo-ary circulation are mediated by the 5-HT2A receptor inontrol rats, but in chronic hypoxic pulmonary hypertensiveats the response is increased, and this is mediated by the-HT1B receptor. Molecular studies confirmed that mes-enger ribonucleic acid (mRNA) for the 5-HT1B receptor isncreased in these vessels. Converging evidence that the-HT1B receptor may be involved in the development ofypoxia-induced PAH comes from studies using the-HT1B/1D antagonist and studies in the 5-HT1B receptornockout mouse (30). Development of right ventricularypertrophy, and enhanced vasoconstriction to 5-HT1-eceptor stimulation, is absent in chronic hypoxic pulmo-ary hypertensive 5-HT1B knockout mice compared withheir wild-type controls, and pulmonary vascular remodel-ng is markedly reduced.

A role for other 5-HT receptors such as 5-HT2B has beenuggested. The 5-HT2B receptor is activated by nordexfen-uramine, the active dexfenfluramine metabolite. Interest-

ngly, development of chronic hypoxic pulmonary hyperten-ion is ablated in 5-HT2B receptor knockout mice, and the-HT2B receptor transcript is increased in idiopathic PAH

�

igure 4. Potential roles of transforming growth factor-beta (TGF-�) superfariety of physiological processes, including cell proliferation, differentiation, imnd differentiation factor.

atients (31). An interesting link between the K channel h

ypothesis and the role of serotonin is the finding that K�

hannel inhibitors cause serotonin release and inhibit K�

urrents in megakaryocytes (32). Moreover, the anorexi-ens, which inhibit serotonin reuptake and cause serotoninelease, are K� channel blockers (22). This led to theypothesis that chronic depolarization of platelets andulmonary artery smooth muscle cells could lead to aasconstricted, pro-proliferative, serotoninemic phenotype.GF-� superfamily. The TGF-� superfamily is com-osed of multifunctional mediators, including the TGF-�soforms (TGF-�1–3), the bone morphogenetic proteinsBMPs), activins, and growth and differentiation factors33,34). The TGF-� superfamily has diverse roles in a wideariety of physiological processes (Fig. 4). Germline muta-ions in the gene coding for BMP type-II receptorBMPR2) have been identified in 60% of familial PAH and0% to 30% of idiopathic PAH (35–37). The absence ofMPR2 mutations in some families and in the majority of

poradic and associated cases suggests that there may beurther genes, possibly related to the BMP/TGF-� path-ay, to be identified. Indeed, mutations in the TGF-�

eceptors, ALK-1 and endoglin, have been identified inAH patients with a personal or family history of hereditary

in vascular remodeling. The TGF-� superfamily has diverse roles in a widety, and inflammation. BMP � bone morphogenetic protein; GDF � growth

amilymuni

emorrhagic telangiectasia (38,39).

kwAmBaptcBSRoctac

roBhSomsoer1

tScnaabpRne

ogeBseoTh

ha

tsslprioRbosmBrutvB

tmslnrada

oadphBttontfwbfi

wliTtahb


The BMPR-II is a constitutively active serine/threonineinase receptor, signaling via formation of heterocomplexesith one of three type-I receptors (ALK-3/BMPR-IA,LK-6/BMPR-IB or ALK-2) in response to ligand. Theain ligands identified for BMPR-II are BMP2, BMP4,MP7, GDF5, and GDF6. The BMPR-II phosphorylatesglycine-serine–rich domain on the proximal intracellular

ortion of the associated type-I receptor. Activation of theype-I receptor kinase domain initiates phosphorylation ofytoplasmic signaling via the Smad family of proteins. TheMPs signal via a specific set of Smad proteins (Smad1,mad5, and Smad8) termed “Receptor activated” or-Smads, which complex with the common partner Smadr Co-Smad, Smad4, to allow translocation of this signalingomplex to the nucleus where they can regulate generanscription. However, Smads bind only weakly to DNAnd require the presence of transcriptional co-activators oro-repressors.

There are multiple levels at which BMP signaling isegulated, including the presence of endogenous inhibitorsf BMP-receptor interactions (chordin, noggin, andAMBI), the formation of specific type-II/type-I receptoreterocomplexes, the activation of inhibitory Smads (I-mads, Smad6, and Smad7), and the cell-specific expressionf transcription factors. Such diverse levels of regulationay be responsible for the tissue specificity of BMP

ignaling and may, for example, underlie the lung specificityf PAH. Human pulmonary artery smooth muscle cells andndothelial cells express a wide range of TGF-� superfamilyeceptors, including BMPR-II and BMPR-IB, and bind25I-TGF-� and 125I-BMP4.

Furthermore, activation of these receptors by BMPs leadso phosphorylation of Smad1 and induction of mRNAs formad6 and Smad7. Although signaling via Smads is wellharacterized, there is increasing evidence that MAP ki-ases, including ERK, p38MAPK, and JNK kinases, arectivated in specific cell types by TGF-� and BMPs (33). Itppears that in some cases the p38 kinase pathway canypass the Smad pathway and mediate some of the BMPR2athway effects on nuclear transcription and apoptosis.ecent evidence suggests that abnormal activation of alter-ative signaling pathways may be critical to the pathogen-sis of PAH (40).

The critical role of the BMP pathway in vascular devel-pment is evident from studies in knockout mice. Homozy-osity for a null mutation in BMPR2 is lethal during earlymbryogenesis (41), and mice deficient in Smad5, one of theMP-restricted Smads, die owing to defects in angiogene-

is, with failure to recruit vascular smooth muscle tondothelial structures. The net result of TGF-� signalingn vascular growth and structure is complex. Whether theGF-� superfamily inhibits or promotes cell proliferation isighly context-specific.In situ hybridization and immunohistochemical studies

ave demonstrated that both BMPR-II mRNA and protein
re present predominantly on the pulmonary vascular endo- i
helium, macrophages, and to a lesser extent on medialmooth muscle cells (42). Lung BMPR-II protein expres-ion is dramatically reduced in patients harboring an under-ying BMPR2 mutation predicted to cause truncation of therotein (42). In addition, BMPR-II expression is markedlyeduced in PAH cases in which no BMPR2 mutation wasdentified (41). A small but significant reduction was alsobserved in cases of secondary pulmonary hypertension.educed BMPR-II expression was specific for this receptorecause no change was observed in the level of expression ofther endothelial markers, including CD31. These findingstress the importance of understanding how other environ-ental and genetic factors regulate the expression ofMPR-II in lung cells. Thus, further characterization of the

egulation of BMPR-II expression is likely to add to ournderstanding of exogenous factors influencing BMPR-IIranscription and may provide important clues as to why theascular abnormality is restricted to the lung, particularly asMPR-II is widely expressed in normal adult tissues.A recently published study has provided further support for

he hypothesis that intact BMP signaling is important for theaintenance of the normal pulmonary vasculature (43). In this

tudy the type-I receptor BMPR-IA was downregulated in theung tissue of a heterogeneous group of patients with pulmo-ary hypertension. Furthermore, investigators showed a recip-ocal relationship between BMPR-IA expression and that ofngiopoietin-1, and they demonstrated that angiopoietin-1ownregulates BMPR-IA expression in human pulmonaryrtery endothelial cells (43).

Additionally, BMP2, -4, and -7 inhibit the proliferationf smooth muscle cells derived from normal pulmonaryrteries and from PAH patients with congenital heartiseases, but they fail to suppress proliferation of cells fromatients with idiopathic or familial PAH (43). An attractiveypothesis is that a failure of the growth inhibitory effects ofMPs in idiopathic or familial PAH cells could contribute

o the vascular obliteration and remodeling that characterizehe condition. The failure to suppress growth of idiopathicr familial PAH cells was observed in all cases, whether orot specific BMPR2 mutations were identified, suggestinghat defective BMP-mediated signaling may be a commonactor in idiopathic or familial PAH. The mechanism byhich BMPR2 mutations disrupt BMPR-II signaling hasegun to be elucidated (Fig. 5) (40,44,45). Interestingly, aeature common to all mutants is a gain of functionnvolving p38MAPK activation.

In pulmonary artery smooth muscle cells from patientsith idiopathic PAH, TGF-�1 causes enhanced cell pro-

iferation in contrast to the growth inhibitory effect observedn normal cells (43). This is not due to alterations inGF-�1 receptor ratios or downregulation of TGF-�1

ype-II receptor (44). Transforming growth factor-beta islso known to increase production of extracellular matrix. Inuman lung fibroblasts, TGF-� increases elastin expressiony stabilization of elastin mRNA, and thus it is possible that
ncreased elastin expression observed in PAH may be due to

atcTTfiagilwrAfa(r

umtvtsbi

bevBt(es

FBemTwwlcwrs


lterations in this pathway. Although studies from other cellypes have found TGF-� to induce collagen production, noorrelation has been found between procollagen andGF-� staining in lungs of PAH patients (46). TheGF-� superfamily may regulate the activity of other

actors implicated in vascular remodeling. The TGF-�1nduces ET-1 in human pulmonary artery cells probably viactivation of protein kinase A (47). Connective tissuerowth factor production can also be stimulated by TGF-�n pulmonary fibroblasts (48). Clearly much remains to beearned of the interaction of the TGF-�/BMP pathwayith other factors already demonstrated to play important

oles in the control of vascular tone and growth.ngiogenesis and apoptosis. Vascular endothelial growth

actor is an endothelial-cell-specific-angiogenic mitogencting via two high-affinity tyrosine kinase receptorsVEGFR-1 and VEGFR-2). Although the physiological

igure 5. Consequences of bone morphogenetic protein type-II receptorMPR2 mutations occur within exon 1 of the gene and would be predictxpress the mutant protein, resulting in haploinsufficiency. Although this fiutations involving the ligand binding or kinase domain of BMPR-II couldhe mechanism by which BMPR-II mutants disrupt BMP/Smad signalingithin the ligand binding or kinase domain of BMPR-II leads to failure oild-type receptor trafficking. In contrast, noncysteine mutations within t

uciferase reporter gene. Interestingly, BMPR-II mutants with missense mapable of activating the Smad-responsive luciferase reporter gene. Howeveas ligand-independent activation of p38MAPK and enhanced serum-indu

educed cell-surface expression of BMPR-II favors activation of p38MA

ignaling in a mutation-specific manner. Thus, a feature common to all m

ole of the abundantly expressed VEGF in the lung is (

nknown, it has been proposed that VEGF supports pul-onary endothelial cell maintenance and survival. In PAH,

he VEGF expression is increased within the pulmonaryasculature, including the plexiform lesions (5,49). Al-hough the isoform VEGF-A has been most extensivelytudied in the context of pulmonary hypertension and haseen proposed to play a protective role, a recent studydentified a pathogenic role for VEGF-B.

In contrast to VEGF-A, the VEGF-B appears to exacer-ate remodeling as VEGF-B knockout mice (VEGF-B�/�)xposed to chronic hypoxia exhibit significantly less pulmonaryascular remodeling compared with wild-type mice (VEGF-�/�) (50). Recent animal studies have emphasized the posi-

ive effects of VEGF in models of pulmonary hypertension51). Indeed, cell-based VEGF gene transfer has proved anffective method of preventing the development and progres-ion of pulmonary hypertension in the monocrotaline model

R2) mutations on signaling. Mutation analysis demonstrated that somecause nonsense-mediated messenger ribonucleic acid decay and failure tomay be true for some mutations, it was also found in transfected cells that

t a dominant negative effect on BMPR-II signaling via the Smad pathway.terogeneous, and mutation specific. Thus, substitution of cysteine residuescking of the mutant protein to the cell surface, which may interfere with

nase domain reach the cell surface but fail to activate a Smad-responsiveions involving the cytoplasmic tail reached the cell surface but were stillature common to all mutants transfected into normal mouse epithelial cellsoliferation. Based on the results of these studies it was hypothesized thatependent pro-proliferative pathways, while inhibiting Smad-dependents is a gain of function involving p38MAPK activation.

(BMPed tonding

exeris he

f traffihe ki

utatr, a feced pr

PK-dutant

51). Vascular endothelial growth factor would minimize

pot

wenhars(tbtpmav

fiedin

psics

vammapicme

pRaeDpo

Fa


rogression of the disease by preventing loss of existing vesselsr by inducing the development of new blood vessels withinhe lung (51).

In idiopathic PAH, VEGFR-1 expression is increased,hereas within the plexiform lesions it is VEGFR-2 that is

xpressed (52). In rats, it has been shown that the combi-ation of chronic blockade of VEGFR-2 and chronicypoxia could cause pulmonary endothelial cell dysfunctionnd cell death, allowing the selection of an apoptosis-esistant proliferating endothelial cell phenotype and theubsequent development of severe pulmonary hypertension53). Because endothelial cell death, cell proliferation, andhe development of severe pulmonary hypertension could belocked by a broad-spectrum caspase inhibitor, it appearedhat the selection of an apoptosis-resistant endothelial cellhenotype might be the critical event responsible for pul-onary artery endothelial cell proliferation (53). Therefore,

poptosis of endothelial cells may underlie the propensity toascular disease (53).

Various other growth factors including PDGF, basicbroblast growth factor, insulin-like growth factor-1, andpidermal growth factor have also been implicated in theevelopment of remodeling and all have been reported to be

ncreased in the pulmonary hypertensive lung. The mecha-

igure 6. Schematic view of pulmonary arterial hypertension (PAH) pathond it is unlikely that one factor or gene mutation will explain all forms a

ism that leads to induction of these growth factors in the d

ulmonary vasculature is unclear, though reactive oxygenpecies have been implicated because hydrogen peroxidenduces PDGF expression in human pulmonary endothelialells, as does hypoxia and mechanical stretch and sheartress.

Angiopoietin-1 is an angiogenic factor essential for lungascular development (43). Produced by smooth-muscle cellsnd precursor pericytes, angiopoietin-1 stabilizes the develop-ent of blood vessels by recruiting muscle cells, throughigration and division, to endothelial tubes, creating mature

rterial structures. The receptor for angiopoietin-1, TIE2, isresent only on vascular endothelium. The ligand-receptornteraction between angiopoietin-1 secreted by smooth-muscleells and endothelium-specific TIE2 during organ develop-ent induces the proliferation of muscle cells around the

ndothelium vascular network.After development is completed, angiopoietin-1 is ex-

ressed at a minimally detectable level in the human lung.ecent studies attempted to analyze the putative role of

ngiopoietin-1 in pulmonary hypertension, but they reachedntirely antithetical conclusions (43,54,55). The findings byu and colleagues (43) suggest that all forms of nonfamilial

ulmonary hypertension are characterized by upregulationf angiopoietin-1 and phosphorylated TIE2, correlating

iology. Pulmonary arterial hypertension has a multifactorial pathobiology,ses of PAH.

physnd ca

irectly with the severity of the disease. A mechanistic link

bsoreepts

eshTm

aothiprmiutipowmpPmvehsphdsewpmMsfes

wtt

5crtBpaiBi

eMsiewMmcMgsciwaCpmHudndpsmp

RvPi

R


etween familial PAH and acquired pulmonary hyperten-ion was supported by the finding that angiopoietin-1 shutsff the expression of BMPR1A, a transmembrane proteinequired for BMPR2 signaling, in pulmonary arteriolarndothelial cells (43). Interestingly, rodents engineered toxpress angiopoietin-1 in the lung develop pulmonary hy-ertension (54). These animals manifest diffuse medialhickening in small pulmonary vessels, resulting frommooth muscle cell hyperplasia.

In addition, angiopoietin-1 stimulates pulmonary arteriolarndothelial cells through a TIE2 pathway to produce andecrete 5-HT (54). These revelations suggest that pulmonaryypertensive vasculopathy occurs through an angiopoietin-1/IE2/5-HT paracrine pathway and imply that these signalingolecules may be targets for strategies to treat this disease.By contrast, Zhao et al. (55) have demonstrated that

ngiopoetin-1 may have a protective role in at least some formsf pulmonary hypertension. In their study, cell-based generansfer with angiopoietin-1 improved survival and pulmonaryemodynamics in monocrotaline-exposed rats by a mechanism

nvolving the inhibition of apoptosis and protection of theulmonary microvasculature (55). These two approaches of theole of angiopoietin-1 in PAH reflect distinct views on theechanisms leading to the disease. On the one hand, the

nvestigators assume that the primary cellular defect contrib-ting to the disease is smooth muscle cell hyperplasia, and thathis is mediated by excess angiopoietin-1. On the other hand,t is hypothesized that endothelial apoptosis underlies diseaserogression, and that this can be prevented by administrationf angiopoietin-1. At the present time one cannot concludehether angiopoietin-1 is the cause or the cure of PAH, butore data are required to evaluate the angiopoietin-1/TIE2

athway in this condition (56).roteolysis. Evidence that proteolysis of the extracellularatrix may be important in the pathobiology of pulmonary

ascular disease came from observations of degradation oflastin in pulmonary arteries from patients with a congenitaleart defect and pulmonary vascular disease (57). Thesetudies were supported by work in a variety of rat models ofulmonary hypertension (hypoxia, monocrotaline) in whicheightened activity of elastase in the pulmonary arteries wasocumented as a very early feature after the injurioustimulus (57). Subsequent studies showed that infusion oflastase inhibitors suppressed the disease process (58,59). Itas then shown that serum factors could induce theroduction of an endogenous vascular elastase from smoothuscle cells and that the mechanism appeared to involveAP kinase activity and nuclear partitioning of the tran-

cription factor AML1 (60). This is a transcription factoror neutrophil elastase and a putative transcription factor forndogenous vascular elastase, based on studies using anti-ense blockade of AML1 to repress elastase activity.

Moreover, repression of nuclear partitioning of AML1 asell as repression of phosphorylation of Erk (a member of

he MAP kinase family) was achieved with NO donors, and
his also repressed elastase activity (61). Suppression of
-HT receptors repressed elastase activity and TGF-� inhronic hypoxia. Also, to be further investigated is theelationship between BMPR-II and the induction of elas-ase activity. Hypothetically, it has been proposed thatMPR-II induces Smadl/5 interaction with AML1, thusreventing its interaction with Smad4 and repressing itsbility to partner with other transcription factors and tonduce elastase activity. Conversely, a mutation inMPR-II would result in derepression of AML1 and

nduction of elastase and other AML1-dependent genes.Evidence from the published data indicates that serine

lastases activate MMPs and also repress tissue inhibitors ofMPs. In other injury models, elastase activity has been

hown to precede MMP activity and is responsible for itsnduction. In the monocrotaline model, an elevation inlastase activity is seen on day 2 after injection of the toxin,hereas MMP activity is not increased until day 21. BothMPs and elastases can degrade most components of theatrix in addition to elastin and collagen. Degradation of

ollagen leads to ligation of �3 integrins activation of theAP kinase pathway and transcription of tenascin C. This

lycoprotein cooperatively interacts with growth factorsuch as epidermal growth factor in inducing smooth muscleell proliferation. Repression of this pathway by elastasenhibitors has been shown in cell and organ culture and inhole animals to induce apoptosis of smooth muscle cells,

nd regression of severe vascular disease (58,59).onclusions. It is clear that PAH has a multifactorialathobiology, and it is unlikely that one factor or geneutation will explain all forms and cases of PAH (Fig. 6).owever, the current understanding of the mechanisms

nderlying PAH has allowed the rapid development ofrugs including prostacyclin, endothelin receptor antago-ists, and phosphodiesterase inhibitors. Our improved un-erstanding of additional pathways in this condition willresumably lead to the development of novel therapeutictrategies in the near future, such as ion channel replace-ent therapy or cell-based therapies, using bone marrow

recursor cells.

eprint requests and correspondence: Dr. Marc Humbert, Ser-ice de Pneumologie, Hopital Antoine-Beclere, 157, Rue de laorte de Trivaux, 92140 Clamart, France. E-mail: humbert@

psc.u-psud.fr.

EFERENCES

1. Voelkel NF, Tuder RM, Weir EK. Pathophysiology of primarypulmonary hypertension. In: Rubin L, Rich S, editors. PrimaryPulmonary Hypertension. New York, NY: Marcel Dekker, 1997:83–129.

2. Jeffery TK, Morrell NW. Molecular and cellular basis of pulmonaryvascular remodeling in pulmonary hypertension. Prog Cardiovasc Dis2002;45:173–202.

3. Stenmark KR, Gerasimovskaya E, Nemenoff RA, Das M. Hypoxicactivation of adventitial fibroblasts: role in vascular remodeling. Chest2002;122:326S–34S.

4. Davie NJ, Crossno JT, Frid MG, et al. Hypoxia-induced pulmo-
nary artery advential remodeling and neovascularization: contribu-

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

4

4

4

4

4

4

4


tion of progenitor cells. Am J Physiol Lung Cell Mol Physiol2004;286:668 –78.

5. Cool CD, Stewart JS, Wehareha P, et al. Three-dimensional recon-struction of pulmonary arteries in plexiform pulmonary hypertensionusing cell-specific markers. Evidence for a dynamic and heterogeneousprocess of pulmonary endothelial cell growth. Am J Pathol 1999;155:411–9.

6. Lee SD, Shroyer KR, Markham NE, Cool CD, Voelkel NF, TuderRM. Monoclonal endothelial cell proliferation is present in primarybut not secondary pulmonary hypertension. J Clin Invest 1998;101:927–34.

7. Yeager ME, Halley GR, Golpon HA, Voelkel NF, Tuder RM.Microsatellite instability of endothelial cell growth and apoptosis geneswithin plexiform lesions in primary pulmonary hypertension. Circ Res2001;88:e2–e11.

8. Cool CD, Rai PR, Yeager ME, et al. Expression of humanherpesvirus-8 in primary pulmonary hypertension. N Engl J Med2003;349:1113–22.

9. Dorfmuller P, Perros F, Balabanian K, Humbert M. Inflammation inpulmonary arterial hypertension. Eur Respir J 2003;22:358–63.

0. Balabanian K, Foussat A, Dorfmuller P, et al. CX (3) C chemokinefractalkine in pulmonary arterial hypertension. Am J Respir Crit CareMed 2002;165:1419–25.

1. Herve P, Humbert M, Sitbon O, et al. Pathobiology of pulmonaryhypertension: the role of platelets and thrombosis. Clin Chest Med2001;22:451–8.

2. Christman BW, McPherson CD, Newman JH, et al. An imbalancebetween the excretion of thromboxane and prostacyclin metabolites inpulmonary hypertension. N Engl J Med 1992;327:70–5.

3. Tuder RM, Cool CD, Geraci MW, et al. Prostacyclin synthaseexpression is decreased in lungs from patients with severe pulmonaryhypertension. Am J Respir Crit Care Med 1999;159:1925–32.

4. Giaid A, Saleh D. Reduced expression of endothelial nitric oxidesynthase in the lungs of patients with pulmonary hypertension. N EnglJ Med 1995;333:214–21.

5. Michelakis ED, Tymchak W, Noga M, et al. Long-term treatmentwith oral sildenafil is safe and improves functional capacity andhemodynamics in patients with pulmonary arterial hypertension.Circulation 2003;108:2066–9.

6. Petkov V, Mosgoeller W, Ziesche R, et al. Vasoactive intestinalpeptide as a new drug for treatment of primary pulmonary hyperten-sion. J Clin Invest 2003;111:1339–46.

7. Giaid A, Yanagisawa M, Langleben D, et al. Expression ofendothelin-1 in the lungs of patients with pulmonary hypertension.N Engl J Med 1993;328:1732–9.

8. Archer S, Rich S. Primary pulmonary hypertension: a vascular biologyand translational research “work in progress.” Circulation 2000;102:2781–91.

9. Yuan JX, Wang J, Juhaszova M, Gaine SP, Rubin LJ. Attenuated K�

channel gene transcription in primary pulmonary hypertension. Lancet1998;351:726–7.

0. Michelakis ED, McMurtry MS, Wu XC, et al. Dichloroacetate, ametabolic modulator, prevents and reverses chronic hypoxic pulmonaryhypertension in rats: role of increased expression and activity ofvoltage-gated potassium channels. Circulation 2002;105:244–50.

1. Geraci MW, Moore M, Gesell T, et al. Gene expression patterns inthe lungs of patients with primary pulmonary hypertension: a genemicroarray analysis. Circ Res 2001;88:555–62.

2. Weir EK, Reeve HL, Huang JMC, et al. Anorexic agents aminorex,fenfluramine, and dexfenfluramine inhibit potassium current in ratpulmonary vascular smooth muscle and cause pulmonary vasoconstric-tion. Circulation 1996;94:2216–20.

3. Herve P, Launay JM, Scrobohaci ML, et al. Increased plasmaserotonin in primary pulmonary hypertension. Am J Med 1995;99:249–54.

4. MacLean MR, Herve P, Eddahibi S, Adnot S. 5-Hydroxytryptamineand the pulmonary circulation: receptors, transporters and relevance topulmonary arterial hypertension. Br J Pharmacol 2000;131:161–8.

5. Eddahibi S, Humbert M, Fadel E, et al. Serotonin transporteroverexpression is responsible for pulmonary artery smooth musclehyperplasia in primary pulmonary hypertension. J Clin Invest 2001;
108:1141–50.
6. Eddahibi S, Chaouat A, Morrell N, et al. Polymorphism of theserotonin transporter gene and pulmonary hypertension in chronicobstructive pulmonary disease. Circulation 2003;108:1839–44.

7. MacLean MR, Deuchar GA, Hicks MN, et al. Over-expression of the5-hydroxytryptamine transporter gene: effect on pulmonary hemody-namics and hypoxia-induced pulmonary hypertension. Circulation2004. In press.

8. Marcos E, Adnot S, Pham MH, et al. Serotonin transporter inhibitorsprotect against hypoxic pulmonary hypertension. Am J Respir CritCare Med 2003;168:487–93.

9. Morecroft I, Heeley RP, Prentice HM, Kirk A, MacLean MR.5-Hydroxytryptamine receptors mediating contraction in human smallmuscular pulmonary arteries: importance of the 5-HT1B receptor. Br JPharmacol 1999;128:730–4.

0. Keegan A, Morecroft I, Smillie D, Hicks MN, MacLean MR.Contribution of the 5-HT (1B) receptor to hypoxia-induced pulmo-nary hypertension: converging evidence using 5-HT (1B)-receptorknockout mice and the 5-HT (1B/1D)-receptor antagonistGR127935. Circ Res 2001;89:1231–9.

1. Launay JM, Herve P, Peoc’h K, et al. Function of the serotonin5-hydroxytryptamine 2B receptor in pulmonary hypertension. NatMed 2002;8:1129–35.

2. Weir EK, Reeve HL, Peterson DA, Michelakis ED, Nelson DP,Archer SL. Pulmonary vasoconstriction, oxygen sensing, and the roleof ion channels: Thomas A. Neff lecture. Chest 1998;114:17S–22S.

3. Massague J, Chen YG. Controlling TGF-beta signaling. Genes Dev2000;14:627–44.

4. Shi Y, Massague J. Mechanisms of TGF-beta signaling from cellmembrane to nucleus. Cell 2003;113:685–700.

5. Lane KB, Machado RD, Pauciulo MW, et al. Heterozygous germlinemutations in a TGF-� receptor, BMPR2, are the cause of familialprimary pulmonary hypertension. Nat Genet 2000;26:81–4.

6. Deng Z, Morse JH, Slager SL, et al. Familial primary pulmonaryhypertension (gene PPH1) is caused by mutations in the bonemorphogenetic protein receptor-II gene. Am J Hum Genet 2000;67:737–44.

7. Thomson JR, Machado RD, Pauciulo MW, et al. Sporadic primarypulmonary hypertension is associated with germline mutations of thegene encoding BMPR-II, a receptor member of the TGF-� family.J Med Genet 2000;37:741–5.

8. Trembath R, Thomson JR, Machado RD, et al. Clinical and moleculargenetic features of pulmonary hypertension in patients with hereditaryhemorrhagic telangiectasia. N Engl J Med 2001;345:325–34.

9. Chaouat A, Coulet F, Favre C, et al. Endoglin germline mutation in apatient with hereditary hemorrhagic telangiectasia and dexfenfluramine-associated pulmonary arterial hypertension. Thorax 2004;59:446–8.

0. Rudarakanchana N, Flanagan JA, Chen H, et al. Functional analysis ofbone morphogenetic protein type II receptor mutations underlyingprimary pulmonary hypertension. Hum Mol Genet 2002;11:1517–25.

1. Beppu H, Kawabata M, Hamamoto T, et al. BMP type II receptor isrequired for gastrulation and early development of mouse embryos.Dev Biol 2000;221:249–58.

2. Atkinson C, Stewart S, Upton PD, et al. Primary pulmonary hyper-tension is associated with reduced pulmonary vascular expression oftype II bone morphogenetic protein receptor. Circulation 2002;105:1672–8.

3. Du L, Sullivan CC, Chu D, et al. Signaling molecules in nonfamilialpulmonary hypertension. N Engl J Med 2003;348:500–9.

4. Morrell NW, Yang X, Upton P, et al. Altered growth responses ofpulmonary artery smooth muscle cells from patients with primarypulmonary hypertension to transforming growth factor-�1 and bonemorphogenetic proteins. Circulation 2001;104:790–5.

5. Machado RD, Pauciulo MW, Thomson JR, et al. BMPR2 haploin-sufficiency as the inherited molecular mechanism for primary pulmo-nary hypertension. Am J Hum Genet 2001;68:92–102.

6. Botney MD, Bahadori L, Gold LI. Vascular remodeling in primarypulmonary hypertension: potential role for transforming growthfactor-�. Am J Pathol 1994;144:286–95.

7. Markewitz BA, Farrukh IS, Chen Y, Li Y, Michael JR. Regulation ofendothelin-1 synthesis in human pulmonary arterial smooth musclecells: effects of transforming growth factor-� and hypoxia. CardiovascRes 2001;49:200–6.

8. Kucich U, Rosenbloom JC, Herrick DJ, et al. Signaling events
required for transforming growth factor-� stimulation of connective

4

5

5

5

5

5

5

5

5

5

5

6

6


tissue growth factor expression by cultured human lung fibroblasts.Arch Biochem Biophys 2001;395:103–12.

9. Tuder RM, Chacon M, Alger L, et al. Expression of angiogenesis-relatedmolecules in plexiform lesions in severe pulmonary hypertension: evidencefor a process of disordered angiogenesis. J Pathol 2001;195:367–74.

0. Wanstall JC, Gambino A, Jeffery TK, et al. Vascular endothelialgrowth factor-B–deficient mice show impaired development of hy-poxic pulmonary hypertension. Cardiovasc Res 2002;55:361–8.

1. Campbell AI, Zhao Y, Sandhu R, Stewart DJ. Cell-based gene transfer ofvascular endothelial growth factor attenuates monocrotaline-induced pul-monary hypertension. Circulation 2001;104:2242–8.

2. Hirose S, Hosoda Y, Furuya S, Otsuki T, Ikeda E. Expression ofvascular endothelial growth factor and its receptors correlates closelywith formation of the plexiform lesion in human pulmonary hyper-tension. Pathol Int 2000;50:472–9.

3. Taraseviciene-Stewart T, Kasahara Y, Alger L, et al. Inhibition of theVEGF receptor-2 combined with chronic hypoxia causes cell death-dependent pulmonary endothelial cell proliferation and severe pulmo-nary hypertension. FASEB J 2001;15:427–38.

4. Sullivan CC, Du L, Chu D, et al. Induction of pulmonary hyperten-sion by an angiopoietin 1/TIE2/serotonin pathway. Proc Natl AcadSci U S A 2003;100:12331–6.

5. Zhao YD, Campbell AIM, Robb M, Ng D, Stewart DJ. Protectiverole of angiopoietin-1 in experimental pulmonary hypertension. CircRes 2003;92:984–91.

6. Rudge JS, Thurston G, Yancopoulos GD. Angiopoietin-1 and pul-monary hypertension: cause or cure? Circ Res 2003;92:947–9.

7. Rabinovitch M. Elastase and the pathobiology of unexplained pulmo-nary hypertension. Chest 1998;114:213S–24S.

8. Cowan KN, Jones PL, Rabinovitch M. Elastase and matrixmetalloproteinase inhibitors induce regression, and tenascin-Cantisense prevents progression, of vascular disease. J Clin Invest2000;105:21–34.

9. Cowan KN, Heilbut A, Humpl T, Lam C, Ito S, Rabinovitch M.Complete reversal of fatal pulmonary hypertension in rats by a serineelastase inhibitor. Nat Med 2000;6:698–702.

0. Wigle DA, Thompson KE, Yablonsky S, et al. AML1-like transcrip-tion factor induces serine elastase activity in ovine pulmonary arterysmooth muscle cells. Circ Res 1998;10:252–63.

1. Mitani Y, Zaidi SH, Dufourcq P, Thompson K, Rabinovitch M.Nitric oxide reduces vascular smooth muscle cell elastase activitythrough cGMP-mediated suppression of ERK phosphorylation andAML1B nuclear partitioning. FASEB J 2000;14:805–14.

Cellular and molecular pathobiology of pulmonary arterial ... · pulmonary hypertension is the...

Documents

Transcript of Cellular and molecular pathobiology of pulmonary arterial ... · pulmonary hypertension is the...