Serotonin Increases Susceptibility to PulmonaryHypertension in BMPR2-Deficient Mice
Lu Long, Margaret R. MacLean, Trina K. Jeffery, Ian Morecroft, Xudong Yang,Nung Rudarakanchana, Mark Southwood, Victoria James, Richard C. Trembath, Nicholas W. Morrell
Abstract—Heterozygous germline mutations in the gene encoding the bone morphogenetic protein type II (BMPR-II)receptor underlie the majority (�70%) of cases of familial pulmonary arterial hypertension (FPAH), and dysfunctionof BMP signaling has been implicated in other forms of PAH. The reduced disease gene penetrance in FPAH indicatesthat other genetic and/or environmental factors may also be required for the clinical manifestation of disease. Of these,the serotonin pathway has been implicated as a major factor in PAH pathogenesis. We investigated the pulmonarycirculation of mice deficient in BMPR-II (BMPR2�/� mice) and show that pulmonary hemodynamics and vascularmorphometry of BMPR2�/� mice were similar to wild-type littermate controls under normoxic or chronic hypoxic (2-to 3-week) conditions. However, chronic infusion of serotonin caused increased pulmonary artery systolic pressure, rightventricular hypertrophy, and pulmonary artery remodeling in BMPR2�/� mice compared with wild-type littermates, aneffect that was exaggerated under hypoxic conditions. In addition, pulmonary, but not systemic, resistance arteries fromBMPR2�/� mice exhibited increased contractile responses to serotonin mediated by both 5-HT2 and 5-HT1 receptors.Furthermore, pulmonary artery smooth muscle cells from BMPR2�/� mice exhibited a heightened DNA synthesis andactivation of extracellular signal-regulated kinase 1/2 in response to serotonin compared with wild-type cells. In vitroand in vivo experiments suggested that serotonin inhibits BMP signaling via Smad proteins and the expression of BMPresponsive genes. These findings provide the first evidence for an interaction between BMPR-II–mediated signaling andthe serotonin pathway, perturbation of which may be critical to the pathogenesis of PAH. (Circ Res. 2006;98:0-0.)
Key Words: pulmonary arterial hypertension � bone morphogenetic protein � serotonin
Idiopathic pulmonary arterial hypertension (IPAH) is char-acterized by narrowing and obliteration of the small
arteries of the lung leading to increased pulmonary vascularresistance.1 Elevation of pulmonary arterial pressure leads toright ventricular failure. Patients present with dyspnea afterexertion, and many died of right heart failure within 3 yearsof diagnosis, before modern therapies.2 Treatment with pro-stanoids or endothelin receptor antagonists improves symp-toms and survival, although, for many, the long-term outlookremains poor.
Heterozygous germline mutations in the gene encoding thebone morphogenetic protein type II receptor (BMPR-II)occur in up to 70% of cases of familial PAH (FPAH).3,4
Similar mutations were found in up to 26% of cases ofidiopathic PAH.5 However, the low disease gene penetrancesuggests that other genetic or environmental factors arenecessary to manifest clinical disease.
A series of studies have implicated serotonin (or 5-HT) asa key mediator of PAH.6–8 For example, treatment of ratswith serotonin potentiates the effects of hypoxia on pulmo-
nary arterial pressure, right ventricular hypertrophy andpulmonary vascular remodeling.9 More recently, a poly-morphism in the 5-HT transporter (5-HTT) promoter, whichincreases expression of the 5-HTT, has been associated withIPAH,10 although this association has recently been chal-lenged.11 Hypoxia, a common cause of pulmonary hyperten-sion, increases the proliferative response of pulmonary arterysmooth muscle cells to serotonin.12 The compelling datasupporting the role of serotonin in PAH pathogenesis led usto question whether serotonin interacts with BMPR-II defi-ciency to increase susceptibility to pulmonary hypertension.
Here we report that right ventricular pressure, right ven-tricular weight, and pulmonary vascular morphometry aresimilar in the BMPR-II–deficient mouse (BMPR2�/�) andwild-type littermates under normoxic or chronic hypoxicconditions. However, chronic infusion of serotonin increasedpulmonary artery pressure in the BMPR2�/� mouse, espe-cially under conditions of chronic hypoxia. In addition, inisolated pulmonary arteries, deficiency of BMPR-II increasedthe contractile response to serotonin. We show that the
Original received September 22, 2005; revision received February 9, 2006; accepted February 15, 2006.From the Department of Medicine (L.L., T.K.J., X.Y., N.R., M.S., N.W.M.), University of Cambridge School of Clinical Medicine, Addenbrooke’s
and Papworth Hospitals, Cambridge; Institute of Biomedical and Life Sciences (M.R.M., I.M.), Glasgow University; and Department of Genetics andMolecular Medicine (V.J., R.C.T.), School of Medicine, Kings College London, United Kingdom.
Correspondence to Dr Nicholas W. Morrell, Division of Respiratory Medicine, Department of Medicine, Box 157, Addenbrooke’s Hospital, Hills Rd,Cambridge CB2 2QQ, United Kingdom. E-mail [email protected]
mechanism of this interaction includes increased serotonin-induced activation of extracellular signal-regulated kinase(ERK) 1/2 in BMPR-II–deficient cells and inhibition ofBMP/Smad signaling by serotonin.
Materials and MethodsExperimental DesignBMPR2 knockout mice were generated as previously described andkindly provided by Dr Beppu (Massachusetts General Hospital,Boston).13 Heterozygous null BMPR2�/� mice or wild-type littermatecontrols were used in these studies (strain C57BL/6J). Adult malemice (8 to 12 weeks of age; 25 to 30 g body weight [BW]) were usedthroughout. Animal experiments were reviewed and approved by thelocal animal care committee. Groups of wild-type and BMPR2�/�
mice were maintained in room air (21% O2) or in a normobarichypoxic chamber (10% O2) for 2 or 3 weeks, as described previous-ly.14 Additional groups of animals received serotonin via intraperi-toneal osmotic minipumps (Alzet, Palo Alto, Calif) at a rate of 5nmol/h over 2 weeks, or saline vehicle.9 After recovery, animalswere maintained in room air or in the hypoxic chamber.
Hemodynamic EvaluationFollowing treatments, mice were anesthetized using Hypnorm (fen-tanyl and fluanisone 0.25 mL/kg and midazolam 25 mg/kg byintraperitoneal injection). BW was recorded, and right ventricularsystolic pressure (RVSP) was measured via direct cardiac puncture,as previously described.14 To assess right ventricular hypertrophy,the heart was removed and the right ventricle (RV) free wall wasdissected from the left ventricle plus septum (LV�S) and weighedseparately. In groups of mice, an isolated perfused lung preparationwas used to determine the pressure-flow relationship in the pulmo-nary circulation, as previously described.14
MorphometryMouse lungs were removed and fixed by inflation with 10%formalin. Sections were stained with a monoclonal anti–�-smoothmuscle actin antibody (1:100 dilution, DAKO) and lightly counter-stained with hematoxylin. The wall thickness of muscular arteriesaccompanying terminal bronchioles (40- to 50-�m diameter) wasdetermined as previously described.15 In addition, the proportion ofarteries accompanying alveolar ducts and with immunoreactivity for�-smooth muscle actin (as an index of muscularization) was ex-pressed as a percentage of total vessels counted for each animal bya blinded observer. At least 30 vessels were analyzed per animal.
Cell CulturePulmonary arteries (�500-�m internal diameter) were microdis-sected from wild-type and BMPR2�/� mice. Pulmonary arterysmooth muscle cells (PASMCs) were explant derived and grown in20% FBS before passage. The smooth muscle phenotype wasconfirmed by positive immunofluorescent staining using an antibodyto smooth muscle specific �-actin (Sigma). Early passage cells(passage 2 to 3) were plated in 24-well plates grown to subconflu-ence, then quiesced in 0.1% FBS for 24 hours before serotonintreatment (0.1 to 10 �m) in 0.1% FBS. [3H]-Thymidine (0.5�Ci) wasadded for the final 6 hours. For cell-counting studies, cells wereplated at 15 000 per well in 24-well plates, adhered overnight in 10%serum, then serum deprived for 24 hours before serotonin treatment(1 �mol/L). Cell counts were performed using a hemocytometerbefore addition of serotonin (day 0) and after exposure to serotoninfor 3 days. In some experiments, selective 5-HT antagonists wereincluded, including the 5-HT2B antagonist SB215505 (1 �mol/L), the5-HT2A antagonist ketanserin (0.1 to 1 �mol/L), or the 5-HTTinhibitor fluoxetine (1 �mol/L).
MyographyThe contractile response of small pulmonary arteries (�350- to450-�m internal diameter) from male wild-type and BMPR2�/� mice
was examined using a wire myograph, as described previously.16
Responses to serotonin (0.1 nmol/L to 100 �mol/L), the 5-HT2A/2B
agonist �-methyl-5-HT (1 nmol/L to 100 mmol/L), the 5-HT1B/1D
agonist 5-carboxamidotryptamine (5-CT, 1 nmol/L to 300 mmol/L)or phenylephrine (0.1 nmol/L to 100 �mol/L) were determined. Theeffect of selective 5-HT receptor antagonists ketanserin (5-HT2A) (30nmol/L) and SB224289 (5-HT1B) (200 nmol/L) were also studied.First order mesenteric arteries were also studied. The vessels wereset at a resting tension of 2.5 mN (predetermined in preliminaryexperiments as optimal resting tension) before construction ofcumulative concentration response curves. Responses to serotoninand its agonists or phenylephrine were expressed as a percentage ofthe response to 50 mmol/L KCl to calculate the maximum contrac-tion (Emax). The sensitivity to the various agonists (pEC50 values) wascalculated from individual cumulative concentration response curvesby graphical interpolation (Graphpad Prism).
Quantitative RT-PCRTotal RNA was extracted from BMPR2�/� and wild-type mouselungs (n�3 of each) then reverse transcribed. Quantitative PCR wasperformed using Eurogen SYBR1 green core reagents and sampleswere run on a Stratagene MX4000 multiplex QPCR system. Furtherquantitative PCR analysis was performed for expression of theinhibitor of DNA binding 3 (Id3) gene in mouse PASMCs treatedwith BMP2 and/or serotonin and normalized to �-actin. All sampleswere analyzed in the same sample run for Id3 and �-actin.
Western ImmunoblottingProtein was extracted from wild-type and BMPR2�/� lung andsubjected to immunoblotting. Blots were probed with a polyclonalantibody to BMPR-II as previously described,17 then stripped andreprobed with a monoclonal antibody to �-actin (AC-15, Sigma,UK). In protein derived from lungs and PASMCs, we determined thephosphorylation of Smad1/5 (951b, Cell Signaling, UK) or ERK1/2(4377, Cell Signaling) using rabbit monoclonal antibodies as de-scribed previously.18 An antibody to total ERK1/2 was used as aloading control (Cell Signaling). The activation of Smad1/5 andERK1/2 was studied in cell monolayers following stimulation byBMPs or serotonin, as previously described.18 Some studies wereperformed in the presence of diphenyleneiodonium (DPI), an inhib-itor of the NAD(P)H oxidase.
StatisticsData are presented as means�SE. Data between groups werecompared using a 2-tailed t test or a 1-way analysis of variancefollowed by Tukey’s HSD test, whichever was appropriate. P�0.05indicated a statistically significant result.
ResultsPhenotyping of BMPR2�/� and Wild-Type MiceUnder Normoxia and Chronic HypoxiaThe mean RVSP in BMPR2�/� mice maintained in room airwas indistinguishable from wild-type littermate controls (Fig-ure 1a). In addition, the ratio RV/LV�S was similar betweengenotypes (Figure 1b). We questioned whether BMPR2�/�
mice would develop more severe PAH than littermate con-trols in hypoxia. Exposure of animals to 2 or 3 weeks ofcontinuous hypoxia increased RVSP and right ventricularhypertrophy (RVH) in all animals, with no significant differ-ences between wild-type and BMPR2�/� mice (Figure 1a and1b). To ensure that measurements were not missing subtlechanges in pulmonary vascular resistance between genotypesat baseline, we examined the pressure-flow characteristics ofwild-type and BMPR2�/� mouse lungs. Experiments wereperformed under both normoxic and acute hypoxic conditions(2% oxygen) to assess pulmonary vascular resistance under
2 Circulation Research March 31, 2006
by guest on June 28, 2018http://circres.ahajournals.org/
resting and elevated tone, respectively. Pressure-flow curvesfor wild-type and BMPR2�/� mice were similar, indicating nodifference in pulmonary vascular resistance between geno-types (Figure 1c and 1d).
Serotonin Increases Susceptibility to PAH inBMPR2�/� MiceSerotonin had no significant effect on RVSP or RVH inwild-type mice (Figure 2a and 2b). In contrast, BMPR2�/�
mice exhibited a significant increase in RVSP when exposedto serotonin (22.4�0.9 mm Hg versus 17.7�0.8 mm Hg)
(Figure 2a). In addition, serotonin infusion significantlyincreased the proportion of small arteries that became mus-cularized, although with no significant change in wall thick-ness of already muscularized arteries (Figure 3a and 3b).Infusion of serotonin for 2 weeks under hypoxic conditionsfurther exaggerated the differences between genotypes. Thus,under hypoxic conditions, BMPR2�/� mice treated with sero-tonin exhibited increased RVSP (37.4�1.0 mm Hg versus30.0�1.6 mm Hg), RVH, and pulmonary artery musculariza-tion compared with wild-type mice or BMPR2�/� mice treatedwith saline (Figures 2c and 2d and 3c and 3d). Serotonin did
Figure 1. Pulmonary hemodynamics in wild-type and BMPR2�/� mice. a, Right ventricular systolic pressure was measured under nor-moxic conditions (n�9) and after 2 (n�5) and 3 (n�10) weeks of chronic hypoxia. b, The ratio RV/LV�S was used as an index of rightventricular hypertrophy. c and d, Pressure-flow curves in the isolated perfused mouse lung (n�6) under normoxic conditions (c) andunder acute hypoxic (2% oxygen) conditions (d). *P�0.05 compared with normoxic mice; †P�0.01 compared with normoxic mice.
Long et al Serotonin and BMPR2 in Pulmonary Hypertension 3
by guest on June 28, 2018http://circres.ahajournals.org/
not affect left ventricular pressures or heart rate and did notincrease left ventricular weights in any experimental group(data not shown).
Expression of BMPR-II and Smad Signaling inMouse LungWestern blotting for BMPR-II protein confirmed the reduc-tion in receptor expression in BMPR2�/� mice compared with
controls (supplemental Figure I, available online at http://circres.ahajournals.org). Quantitative RT-PCR also con-firmed that BMPR2�/� mice express BMPR-II mRNA atapproximately 50% of levels found in wild-type littermates(supplemental Figure I). No change in BMPR-II mRNAtranscript levels was observed in animals treated with sero-tonin. Therefore, we determined the phosphorylation ofSmad1/5 proteins, the main signaling intermediaries down-
Figure 2. Serotonin potentiates suscepti-bility to pulmonary hypertension inBMPR2�/� mice. a, Right ventricular sys-tolic pressure in normoxic mice treatedwith saline (n�9) or serotonin (n�12) for 2weeks. b, RV/LV�S ratios as an index ofright ventricular hypertrophy. c and d, Thesame indices studied in mice treated withsaline (n�8) or serotonin (n�12) for 2weeks under hypoxic conditions. *P�0.05compared with corresponding salinetreated mice.
Figure 3. Serotonin exaggerates vascularremodeling in BMPR2�/� mice. a, Per-centage wall thickness in small musculararteries accompanying terminal bronchi-oles in normoxic mice treated with saline(n�9) or serotonin (n�12) for 2 weeks. b,Percentage of muscularized arteries atthe level of alveolar ducts. c and d, Thesame indices studied in mice treatedwith saline (n�8) or serotonin (n�12) for2 weeks under hypoxic conditions.*P�0.05 compared with correspondingsaline treated mice.
4 Circulation Research March 31, 2006
by guest on June 28, 2018http://circres.ahajournals.org/
stream of BMP receptors, in mouse lungs exposed to 2 weeksof normoxia or hypoxia with or without serotonin infusion.No difference in basal phosphorylation of Smad1/5 wasobserved in wild-type or BMPR2�/� mice under normoxicconditions. Two weeks of hypoxia consistently increased thephosphorylation of Smad1/5 in wild-type and heterozygousmouse lung (Figure 4a and 4b), with no difference betweengenotypes. Infusion of serotonin in wild-type mice had noconsistent effect on phosphorylation of Smad1/5 under nor-moxic conditions (data not shown), although the phosphory-lation signal was low in normoxia. Thus, we examined theeffect of serotonin infusion on Id3 mRNA expression innormoxic mouse lungs. Basal Id3 expression was similar inwild-type and BMPR2�/� lungs, but serotonin depressed Id3expression in both (see supplemental Figure II). Underhypoxic conditions serotonin infusion consistently reduced
the elevated level of phosphorylated Smad1/5 (Figure 4c and4d) in both wild-type and BMPR2�/� mice. Taken together,these data suggest that serotonin inhibits BMP signalingpathways in the normoxic and hypoxic lung.
Heightened Proliferation to Serotonin inBMPR2�/� PASMCsTo further explore the interaction between serotonin andBMPR-II deficiency, we performed experiments in isolatedPASMCs. PASMC monolayers were �95% positive forsmooth muscle �-actin, whereas mouse fibroblasts did notstain when cultured under the same conditions (not shown).Serotonin increased [3H]-thymidine incorporation inPASMCs derived from BMPR2�/� mice, compared withwild-type littermates (Figure 5a). The observed increase inDNA synthesis corresponded to increased cell number (Fig-
Figure 4. Phospho-Smad1/5 expression in mouse lung. Immunoblots for phospho-Smad1/5 in wild-type (a) and BMPR2�/� (b) lungsunder normoxic (n�3) and hypoxic (n�4) conditions. Two-week serotonin infusion reduced lung phospho-Smad1 levels under hypoxicconditions in wild-type (c) and BMPR2�/� (d) lungs (n�4 for serotonin treated and saline treated groups). Quantification of the phospho-Smad1/5 signal, compared with each �-actin loading control, is shown next to each blot. *P�0.05.
Long et al Serotonin and BMPR2 in Pulmonary Hypertension 5
by guest on June 28, 2018http://circres.ahajournals.org/
ure 5b). Increased [3H]-thymidine incorporation in responseto serotonin in BMPR2�/� cells was blocked by the 5-HT2A
antagonist, ketanserin, at both 0.1 �mol/L and 1 �mol/Lconcentrations, but not by a 5-HT2B antagonist or an inhibitorof the serotonin transporter (Figure 5c). Additional studiesfailed to show an exaggerated response to the 5-HT1 agonist,5-carboxamidotryptamine, in BMPR2�/� PASMCs (notshown).
Potentiation of Serotonin Signaling inBMPR2-Deficient CellsTo further investigate the basis of the exaggerated growthresponse to serotonin in BMPR2-deficient PASMCs, wedetermined the activation of ERK1/2, reported to be involvedin the proliferative response to serotonin.19,20 Immunoblottingdemonstrated that serotonin exposure caused a greater acti-
vation of ERK1/2 in BMPR2�/� PASMCs compared withwild-type cells (Figure 6a). In addition, activation of ERK1/2by serotonin was inhibited by preincubation with ketanserin,consistent with the results of the [3H]-thymidine studies. Theactivation of ERK1/2 by serotonin is thought to be mediatedvia an increase in intracellular superoxide (O��
2 ).20 Inhibitionof intracellular O��
2 generation by the inhibitor of NAD(P)Hoxidase, DPI, significantly reduced the activation of ERK1/2by serotonin in wild-type and BMPR2�/� cells (Figure 6b).
Serotonin Interferes With BMP Signaling In VitroNext we examined whether serotonin inhibited BMP signal-ing in PASMCs. We stimulated PASMCs with BMP2 (30ng/mL) for 1 hour and, as expected, found that phosphoryla-tion of Smad1/5 was reduced in BMPR2�/� cells (Figure 7a).More importantly, preincubation of PASMCs with serotoninfor 30 minutes before stimulation with BMP2 showed inhi-bition of Smad1/5 phosphorylation in wild-type andBMPR2�/� cells (Figure 7a). To investigate further the func-tional impact of serotonin on BMP signaling, we chose a
Figure 5. Proliferation to serotonin in PASMCs. a, Cells derivedfrom wild-type and BMPR2�/� mice showing increased[3H]-thymidine incorporation in cells derived from BMPR2�/�
mice in 0.1% FBS. b, Cell counts in wild-type and BMPR2�/�
cells exposed to serotonin for 3 days showing increased prolif-eration in BMPR2�/� cells. c, [3H]-Thymidine incorporation inresponse to serotonin (1 �mol/L) in BMPR2�/� PASMCs and theeffect of coincubation with selective antagonists of the 5-HT2B
receptor (SB215505, 1 �mol/L), 5-HT2A receptor (ketanserin,1 �mol/L), and the 5-HTT (fluoxetine, 1 �mol/L). *P�0.05, datarepresentative of experiments performed with cells from 5 wild-type and 5 BMPR2�/� mice.
Figure 6. Enhanced activation of ERK1/2 in BMPR2�/�
PASMCs. a, PASMCs from wild-type and BMPR2�/� mice wereexposed to serotonin (1 �mol/L) and protein immunoblotted forphosphorylated ERK1/2 and total ERK1/2. Experiments wereperformed in cells from 3 different mice of each genotype. Rep-resentative blots are shown. Bar graphs represent the quantifi-cation of all blots by image analysis and normalizing for totalERK1/2. b, Representative immunoblot and quantification (n�3)showing the effect of inhibition of reactive oxygen species gen-eration by DPI on ERK1/2 phosphorylation in wild-type andBMPR2�/� cells. *P�0.05 compared with corresponding controlconditions; †P�0.05 compared with serotonin treatment alone.
6 Circulation Research March 31, 2006
by guest on June 28, 2018http://circres.ahajournals.org/
well-documented transcriptional target of BMP/Smad signal-ing, Id3. Real-time quantitative RT-PCR studies revealed thatserotonin powerfully inhibited the transcription of Id3 mRNAinduced by BMP2 in wild-type and BMPR2�/� cells (Figure7b and 7c), although the basal expression of Id3 was notsignificantly different in wild-type and BMPR2�/� cells.Interestingly, the transcription of Id3 mRNA in response toBMP2 was markedly reduced in BMPR2�/� cells (�3.5-fold)compared with wild type (�40-fold), although the inhibitoryeffect of serotonin was still evident (Figure 7c).
Pulmonary Arteries From BMPR2�/� MiceDemonstrate Increased Contractile Responsesto SerotoninIncreased vascular tone, as well as vascular remodeling,contributes to pulmonary hypertension in animal models.Therefore, we investigated the contractile response of in-trapulmonary arteries from wild-type and BMPR2�/� mice toserotonin. In BMPR2�/� mouse vessels, serotonin induced amore profound contraction, with the maximum responsebeing increased nearly 2-fold to that observed in wild-typevessels (Figure 8a and supplemental Table I). The affinity ofserotonin (pEC50) was not altered. In contrast, phenylephrine,an a1 receptor agonist, produced similar contractile responsesin vessels from wild-type and BMPR2�/� mice, demonstratingpharmacological restriction of the exaggerated response toserotonin (Figure 8b).
To determine whether the increased contractile response toserotonin in BMPR2�/� arteries was specific to pulmonaryarteries, we studied mesenteric arteries. These experimentsrevealed no difference in the contractile responses to seroto-nin and phenylephrine in wild-type and BMPR2�/� mesen-teric arteries (see supplemental Figure III).
We investigated the serotonin receptor subtype involved inthe enhanced responses to serotonin in BMPR2�/� mice.Responses to the 5-HT2A/2B agonist, �-methyl-5-HT, and the5-HT1 agonist, 5-CT, were determined. The efficacy (notaffinity) of �-methyl-5-HT was increased in BMPR2�/�
vessels (Figure 8d and supplemental Table I). There was aleftward shift of the response curve to 5-CT in the BMPR2�/�
mice (Figure 8c). However, a maximum response to 5-CTwas not achieved in the BMPR2�/� vessels, at the concentra-tion achievable, and so neither pEC50 nor Emax could becalculated for BMPR2�/� data. The response to 0.1 mmol/Lwas, however, doubled in the BMPR2�/� vessels (P�0.05,Figure 8c). The exaggerated contractile response to serotoninin BMPR2�/� mice was inhibited in the presence of ketanserinor SB224289 (Figure 8e). These data support the involvementof 5-HT2A and 5-HT1B in the exaggerated contractile responseto serotonin in BMPR2�/� mice.
DiscussionThis study has demonstrated that deficiency of BMPR-IIincreases susceptibility to PAH induced by serotonin in mice.
Figure 7. Serotonin antagonizes BMPsignaling in PASMCs. a, PASMCs fromwild-type and BMPR2�/� mice wereexposed to BMP2 and/or serotonin andprotein immunoblotted for phosphorylat-ed Smad1/5. �-Actin was used as load-ing control. Results are representative of3 separate experiments, and bar graphshows quantification of mean data from3 experiments. *P�0.05 compared withcontrol values, †P�0.05 compared withcorresponding BMP-treated sample. band c show the change in Id3 mRNAexpression normalized to �-actin by real-time PCR in response to BMP2 (30ng/mL) with and without serotonin(1 �mol/L) in wild-type and BMPR2�/�
PASMCs. Note the different axis scalesfor wild-type and heterozygous samples.Results represent the means of 3 sepa-rate experiments. *P�0.01 comparedwith control, **P�0.01 compared withBMP2 treatment, †P�0.05 comparedwith BMP2 treatment.
Long et al Serotonin and BMPR2 in Pulmonary Hypertension 7
by guest on June 28, 2018http://circres.ahajournals.org/
Although the pulmonary vascular phenotype of BMPR2�/�
and wild-type littermates was similar under normoxic orchronic hypoxic conditions, infusion of serotonin increasedRVSP, RVH, and pulmonary vascular remodeling inBMPR2�/� mice compared with controls. These data providethe first evidence for cross-talk between the BMP andserotonin pathways, both key systems implicated in thepathogenesis of pulmonary hypertension. Serotonin infusionwas associated with a reduction in lung Smad1/5 activation inhypoxic mice. Serotonin inhibited BMP signaling inPASMCs, as evidenced by inhibition of Smad1/5 phosphor-ylation and inhibition of Id3 transcription. Furthermore,PASMCs isolated from BMPR2�/� mice exhibited a height-ened DNA synthesis to serotonin and increased activation ofERK1/2 via O��
2 . Moreover, we found that pulmonary, but notsystemic, arteries from BMPR2�/� mice demonstrated anincreased contractile response to serotonin, compared witharteries from wild-type littermates. Taken together, these datasupport the hypothesis that a deficiency of BMPR-II in-creases susceptibility to PAH and that serotonin may be apermissive factor in the manifestation of disease.
We reasoned that BMPR2�/� mice may be a useful geneticmodel of FPAH in humans, because many disease causingmutations in human BMPR2 likely result in haploinsuffi-
ciency.4 However, we observed no difference betweenBMPR2�/� mice and wild-type littermates with regard toRVSP, RVH, or muscularization of peripheral pulmonaryarteries. In addition, pressure-flow curves in the isolatedperfused lung suggested no difference in pulmonary vascularresistance. These results are consistent with a recent report ofthe pulmonary vascular phenotype in the same BMPR2�/�
mice used in the present study,21 although they differ fromearlier results in the same animal.22 Beppu et al22 reportedthat BMPR2�/� mice exhibit mild elevation of pulmonaryvascular resistance and increased wall thickness of muscularpulmonary arteries compared with wild-type littermates un-der normoxic conditions. Taking these reports together thepulmonary vascular phenotype in unchallenged BMPR2�/�
mice is, at best, only subtly different from wild-type controls.Song et al21 recently demonstrated that BMPR2�/� micedevelop more severe pulmonary hypertension than wild-typecontrols in the setting of pulmonary inflammation induced byadenovirus-mediated overexpression of 5-lipoxygenase.However, overexpression of a dominant-negative BMPR-II invascular smooth muscle appears to be sufficient for thedevelopment of pulmonary hypertension in transgenic mice.23
Taken together, the in vivo data suggest that BMPR2 haplo-insufficiency is not in itself sufficient to generate significant
Figure 8. Increased contractile responses to sero-tonin in BMPR2�/� pulmonary arteries. a and b,Cumulative concentration-response curves to sero-tonin (a) and phenylephrine (b) in small musculararteries from wild-type and BMPR2�/� mice. c andd, Cumulative concentration-response curves to5-CT (c) and �-methyl-5-HT (d) in small musculararteries from wild-type and BMPR2�/� mice. e,Cumulative concentration-response curves to sero-tonin in BMPR2�/� pulmonary arteries in the pres-ence and absence of ketanserin (30 nmol/L) andSB224289 (200 nmol/L).
8 Circulation Research March 31, 2006
by guest on June 28, 2018http://circres.ahajournals.org/
pulmonary hypertension in mice but increases susceptibilityto specific pulmonary hypertensive stimuli including seroto-nin and inflammation induced by 5-lipoxygenase. In contrastto these studies, mice deficient in the BMPR-II ligandBMP424 are protected from hypoxia-induced pulmonary hy-pertension. These studies suggest that deficiency of a singleligand is not the functional equivalent of receptor deficiencyand may indicate a specific function of BMPs during hypoxia.
Because serotonin has been widely implicated in thepathogenesis of IPAH,8 we investigated the possibility thatserotonin could influence the pulmonary vascular phenotypeof BMPR2�/� mice. Serotonin was infused over a period of 2weeks at a dose (5 nmol/h) previously shown to elevate bloodserotonin levels in rats.9 In the present study, serotoninincreased susceptibility to PAH in BMPR2�/�, but not wild-type, mice. This potentiation became more evident underchronic hypoxic conditions. Although numerous mechanismsmight contribute to the effects of serotonin in BMPR2�/�
mice, we hypothesized that serotonin may exacerbate adeficiency in BMP signaling. Interestingly, chronic hypoxiawas associated with an increase in Smad1/5 phosphorylationin mouse lung, consistent with recent reports of increasedlung BMP expression under similar hypoxic conditions.24
Hypoxia-induced activation of Smad1/5 occurred to a similarextent in wild-type and BMPR2�/� lungs. However, the invivo environment is complex with multiple ligands andreceptors capable of Smad1/5 activation. Nevertheless, im-munoblotting of mouse lung protein after 2 weeks of seroto-nin infusion showed a reduction in hypoxia-induced Smad1/5activation. Furthermore, serotonin infusion reduced Id3mRNA expression in wild-type and BMPR2�/� mouse lungs.We have previously demonstrated that Smad1/5 activationexerts antiproliferative effects on human PASMCs.18 Thus, asuppression of Smad1/5 phosphorylation during serotonininfusion might exert a permissive effect on vascular remod-eling via Smad responsive genes.
Our data from whole mouse lungs exposed to 2 weeks ofserotonin infusion suggested that serotonin may inhibit BMPsignaling in vivo. The effects in vivo are likely to be complex,with numerous inputs from diverse lung cellular compart-ments, BMPs and, BMP receptors. We, therefore, investi-gated this phenomenon further in isolated PASMCs fromwild-type and BMPR2�/� mice. BMP4-induced Smad1/5activation was reduced in BMPR2�/� cells. This is consistentwith a recent report in BMPR2�/� PASMCs in which multipleBMP ligands showed reduced activation of Smad1/5.25 In thatstudy, complete ablation or knockdown of BMPR-II expres-sion led to a gain of function in terms of increased activationof Smad 1/5 in response to BMP6 and -7. In the present study,exposure of PASMCs to serotonin inhibited the BMP2-induced phosphorylation of Smad1/5 in wild-type andBMPR2�/� cells. In addition, we determined the effect ofserotonin on the transcription of a known target gene of BMPsignaling, the helix-loop-helix factor inhibitor of DNA bind-ing 3 (Id3), implicated in vascular smooth muscle growth anddifferentiation.26 The transcription of Id3 in response toBMP2 was reduced in BMPR2�/� PASMCs compared withcontrols, and serotonin exerted a marked inhibitory effect onId3 transcription in wild-type and BMPR2�/� cells. Taken
together, these findings suggest that serotonin inhibits theactivation of BMP-dependant antiproliferative pathways inPASMCs.
A further key finding in this study was that PASMCs fromBMPR2�/� mice exhibited an increased proliferative responseto serotonin compared with wild-type cells. This proliferativeresponse was inhibited by the 5-HT2A receptor antagonist,ketanserin, but not by inhibitors of the 5-HT1, 5-HT2B, or the5-HTT. Serotonin caused an exaggerated activation of theproproliferative ERK1/2 pathway in BMPR2�/� PASMCs.This effect was also mediated via the 5-HT2A receptor andwas at least partly dependant on O��
2 generation. Takentogether, our in vitro results suggest that serotonin inhibitsantiproliferative Smad1/5 signaling in wild-type andBMPR2�/� PASMCs and selectively enhances ERK1/2 pro-proliferative pathways in BMPR2�/� cells. We have previ-ously suggested that an imbalance between antiproliferativeSmad1/5 signals and proproliferative ERK1/2 pathways con-tribute to proliferation in BMPR2 mutant human PASMCs,18
a concept supported by the present study. These resultsprovide a potential mechanism for the increased susceptibilityto pulmonary hypertension induced by serotonin on a back-ground of BMPR2 deficiency. Although our data confirm arole for O��
2 in the activation of ERK1/2 by serotonin, itremains to be determined whether deficiency of BMPR2 leadsto intrinsic alterations in intracellular O��
2 generation.Recent work has implicated the serotonin transporter,
5-HTT, in the proliferative response of human PASMCs toserotonin10 and mice overexpressing the 5-HTT developspontaneous PAH in normoxic conditions.27 The 5-HT2B
receptor has previously been shown to exert effects onvascular remodeling, matrix deposition, and transforminggrowth factor-� expression in hypoxic rats and mice.28
However, our data suggest that the functional interactionbetween serotonin and BMPR-II deficiency occurs mainly viathe 5-HT2A receptor in the mouse.
In addition to vascular remodeling, vasoconstriction playsa role in the pathogenesis of PAH. Therefore, we studied thecontractile response to serotonin in pulmonary arteries usingwire myography. BMPR2�/� pulmonary arteries demonstratedan increased contractile response to serotonin. The efficacy,but not affinity, of serotonin was increased in the BMPR2�/�
mice, suggesting that there may be an increase in serotoninreceptor number. To investigate this further, we studied theeffects of the 5-HT2A/2B agonist �-methyl-5-HT and the 5-HT1
agonist 5-CT and observed an increase in the Emax of�-methyl-5-HT, suggesting an increase in 5-HT2A/2B receptorsin these vessels. The lack of contraction of mouse pulmonaryartery to 5-HT2B agonists (unpublished observation) makes itlikely that it is 5-HT2A mediating contraction in these exper-iments. The response to 5-CT was also increased. We knowthat contractile responses to 5-CT are inhibited by both acuteand chronic treatment with 5-HT1B/1D antagonists.16 The5-HT1B receptor subtype mediates contraction in human smallmuscular pulmonary arteries.29 It is also the 5-HT1B receptorthat mediates hypoxic pulmonary hypertension in rat andmouse models,16 and, hence, it is likely that there is anincreased response to 5-HT1B receptors in the BMPR2�/�
vessels studied here. The involvement of these specific
Long et al Serotonin and BMPR2 in Pulmonary Hypertension 9
by guest on June 28, 2018http://circres.ahajournals.org/
receptor subtypes was confirmed with the use of selectiveinhibitors of 5-HT2A and 5-HT1B, both of which were able toinhibit the heightened contractile response to serotonin inBMPR2�/� arteries.
Our results indicate the presence of a relatively specificfunctional antagonism between the BMPR-II and serotoninsignaling pathways, with deficiency of BMPR-II potentiatingthe contractile and growth response to serotonin in vitro andincreasing susceptibility to PAH and pulmonary vascularremodeling in vivo. The molecular mechanism of this inter-action includes inhibition of Smad signaling and BMP targetgene transcription by serotonin and enhanced activation ofproproliferative ERK1/2 pathways by serotonin in the settingof BMPR2 deficiency. These findings provide a link between2 key systems widely implicated in the pathogenesis ofpulmonary arterial hypertension.
AcknowledgmentsThis work was funded by The British Heart Foundation. We thankDr Hideyuki Beppu (Harvard Medical School) for providing us withthe BMPR2�/� mouse.
References1. Jeffery TK, Morrell NW. Molecular and cellular basis of pulmonary
vascular remodeling in pulmonary hypertension. Prog CardiovascDis. 2002;45:173–202.
3. The International PPH Consortium, Lane KB, Machado RD, PauciuloMW, Thomson JR, Philips JA, Loyd JE, Nichols WC, Trembath RC.Heterozygous germ-line mutations in BMPR2, encoding a TGF-�receptor, cause familial primary pulmonary hypertension. Nat Genetics.2000;26:81–84.
4. Machado RD, Pauciulo MW, Thomson JR, Lane KB, Morgan NV,Wheeler L, Phillips JA III, Newman J, Williams D, Galie N, Manes A,McNeil K, Yacoub M, Mikhail G, Rogers P, Corris P, Humbert M,Donnai D, Martensson G, Tranebjaerg L, Loyd JE, Trembath RC, NicholsWC. BMPR2 haploinsufficiency as the inherited molecular mechanismfor primary pulmonary hypertension. Am J Hum Genet. 2001;68:92–102.
5. Thomson JR, Machado RD, Pauciulo MW, Morgan NV, Humbert M,Elliot GC, Ward K, Yacoub M, Mikhail G, Rogers P, Newman J, WheelerL, Higenbottam T, Gibbs JSR, Egan J, Crozier A, Peacock A, Allcock R,Corris P, Loyd JE, Trembath RC, Nichols WC. Sporadic primary pulmo-nary hypertension is associated with germline mutations of the geneencoding BMPR-II, a receptor member of the TGF-� family. J MedGenet. 2000;37:741–745.
6. Eddahibi S, Raffestin B, Pham I, Launay JM, Aegerter P, Sitbon M,Adnot S. Treatment with 5-HT potentiates development of pulmonaryhypertension in chronically hypoxic rats. Am J Physiol. 1997;272:H1173–H1181.
7. Eddahibi S, Fabre V, Boni C, Martres MP, Raffestin B, Hamon M, AdnotS. Induction of serotonin transporter by hypoxia in pulmonary vascularsmooth muscle cells: relationship with the mitogenic action of serotonin.Circ Res. 1999;84:329–336.
8. MacLean MR, Herve P, Eddahibi S, Adnot S. 5-Hydroxytryptamine andthe pulmonary circulation: receptors, transporters and relevance to pul-monary arterial hypertension. Br J Pharmacol. 2000;131:161–168.
9. Eddahibi S, Raffestin B, Pham I, Launay J-M, Aegerter P, Sitbon M,Adnot S. Treatment with 5-HT potentiates development of pulmonaryhypertension in chronically hypoxic rats. Am J Physiol. 1997;272:H1173–H1181.
10. Eddahibi S, Humbert M, Fadel E, Raffestin B, Darmon M, Capron F,Simonneau G, Dartevelle P, Hamon M, Adnot S. Serotonin transporteroverexpression is responsible for pulmonary artery smooth muscle hyper-plasia in primary pulmonary hypertension. J Clin Invest. 2001;108:1141–1150.
11. Machado RD, Koehler R, Glissmeyer E, Veal C, Suntharalingam J, KimM, Carlquist J, Town M, Elliott CG, Hoeper M, Fijalkowska A, KurzynaM, Thomson JR, Gibbs SR, Wilkins MR, Seeger W, Morrell NW,
Gruenig E, Trembath RC, Janssen BR. Genetic association of the sero-tonin transporter in pulmonary arterial hypertension. Am J Respir CritCare Med. In press.
12. Eddahibi S, Chaouat A, Morrell N, Fadel E, Fuhrman C, Bugnet AS,Dartevelle P, Housset B, Hamon M, Weitzenblum E, Adnot S. Poly-morphism of the serotonin transporter gene and pulmonary hypertensionin chronic obstructive pulmonary disease. Circulation. 2003;108:1839–1844.
13. Beppu H, Kawabata M, Hamamoto T, Chtil A, Minowa O, Noda T,Miyazono K. BMP type II receptor is required for gastrulation and earlydevelopment of mouse embryos. Dev Biol. 2000;221:249–258.
14. Zhao L, Long L, Morrell NW, Wilkins MR. NPR-A-deficient mice showincreased susceptibility to hypoxia-induced pulmonary hypertension. Cir-culation. 1999;99:605–607.
15. Morrell NW, Morris KG, Stenmark KR. Role of angiotensin convertingenzyme and angiotensin II in development of hypoxic pulmonary hyper-tension. Am J Physiol. 1995;269:H1186–H1194.
16. Keegan A, Morecroft I, Smillie D, Hicks MN, MacLean MR. Contri-bution of the 5-HT1B receptor to chronic hypoxia-induced pulmonaryhypertension: converging evidence using 5-HT1B receptor knockout miceand the 5-HT1B/1D receptor antagonist GR127935. Circ Res. 2001;89:1231–1239.
17. Rosenzweig BL, Imamura T, Okadome T, Cox GN, Yamashita H, tenDijke P, Heldin C-H, Miyazano K. Cloning and characterization of ahuman type II receptor for bone morphogenetic proteins. Proc Natl AcadSci U S A. 1995;92:7632–7636.
18. Yang X, Long L, Southwood M, Rudarakanchana N, Upton PD, JefferyTK, Atkinson C, Chen H, Trembath RC, Morrell NW. DysfunctionalSmad signaling contributes to abnormal smooth muscle cell proliferationin familial pulmonary arterial hypertension. Circ Res. 2005;96:1053–1063.
19. Lee SL, Simon AR, Wang WW, Fanburg BL. H2O2 signals 5-HT-induced ERK MAP kinase activation and mitogenesis of smooth musclecells. Am J Physiol Lung Cell Mol Physiol. 2001;281:L646–L652.
20. Lee SL, Wang WW, Finlay GA, Fanburg BL. Serotonin stimulatesmitogen-activated protein kinase activity through the formation ofsuperoxide anion. Am J Physiol Lung Cell Mol Physiol. 1999;277:L282–L291.
21. Song Y, Jones JE, Beppu H, Keaney JF Jr, Loscalzo J, Zhang YY.Increased susceptibility to pulmonary hypertension in heterozygousBMPR2-mutant mice. Circulation. 2005;112:553–562.
22. Beppu H, Ichinose F, Kawai N, Jones RC, Yu PB, Zapol WM, MiyazonoK, Li E, Bloch KD. BMPR-II heterozygous mice have mild pulmonaryhypertension and an impaired pulmonary vascular remodeling response toprolonged hypoxia. Am J Physiol Lung Cell Mol Physiol. 2004;287:L1241–L1247.
23. West J, Fagan K, Steudel W, Fouty B, Lane K, Harral J, Hoedt-Miller M,Tada Y, Ozimek J, Tuder R, Rodman DM. Pulmonary hypertension intransgenic mice expressing a dominant-negative BMPRII gene in smoothmuscle. Circ Res. 2004;94:1109–1114.
24. Frank DB, Abtahi A, Yamaguchi DJ, Manning S, Shyr Y, Pozzi A,Baldwin HS, Johnson JE, de Caestecker MP. Bone morphogenetic protein4 promotes pulmonary vascular remodeling in hypoxic pulmonary hyper-tension. Circ Res. 2005;97:496–504.
25. Yu PB, Beppu H, Kawai N, Li E, Bloch KD. Bone morphogenetic protein(BMP) type II receptor deletion reveals BMP ligand-specific gain ofsignaling in pulmonary artery smooth muscle cells. J Biol Chem. 2005;280:24443–24450.
26. Kumar MS, Hendrix JA, Johnson AD, Owens GK. Smooth muscle{alpha}-actin gene requires two E-boxes for proper expression in vivoand is a target of class I basic helix-loop-helix proteins. Circ Res.2003;92:840–847.
27. MacLean MR, Deuchar GA. Hicks MN, Morecroft I, Shen S, Sheward J,Colston J, Loughlin L, Nilsen M, Dempsie Y, Harmar AJ. Over-expression of the 5-hydroxytryptamine transporter gene: effect on pul-monary haemodynamics and hypoxia-induced pulmonary hypertension.Circulation. 2004;109:2150–2155.
28. Launay JM, Herve P, Peoc’h K, Tournois C, Callebert J, Nebigil CG,Etienne N, Drouet L, Humbert M, Simonneau G, Maroteaux L. Functionof the serotonin 5-hydroxytryptamine 2B receptor in pulmonary hyper-tension. Nat Med. 2002;8:1129–1135.
29. Morecroft I, Heeley RP, Prentice HM, Kirk A, MacLean MR. 5-Hydroxy-tryptamine receptors mediating contraction in human small muscularpulmonary arteries: importance of the 5-HT1B receptor. Br J Pharmacol.1999;128:730–734.
10 Circulation Research March 31, 2006
by guest on June 28, 2018http://circres.ahajournals.org/
MorrellRudarakanchana, Mark Southwood, Victoria James, Richard C. Trembath and Nicholas W.
Lu Long, Margaret R. MacLean, Trina K. Jeffery, Ian Morecroft, Xudong Yang, Nung-Deficient MiceBMPR2Serotonin Increases Susceptibility to Pulmonary Hypertension in
is online at: Circulation Research Information about subscribing to Subscriptions:
http://www.lww.com/reprints Information about reprints can be found online at: Reprints:
document. Permissions and Rights Question and Answer about this process is available in the
located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is
can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:
by guest on June 28, 2018http://circres.ahajournals.org/
Online Figure 1. Expression of BMPR-II mRNA and protein. a, Western blot of BMPR-II protein expression in wild type (+/+) and BMPR2+/- mice (n=3 of each) showing reduced expression of BMPR-II in heterozygous mice. Beta-actin is shown as a loading control; b, levels of BMPR-II transcripts in control mice and BMPR2+/- mice and after exposure to serotonin for 2 weeks. *P<0.05 compared with wild type control mice, samples repeated in triplicate from n=3 mice; data represent means and 95% confidence intervals.
Online Figure 2
+/+ +/+ 5HT +/- +/- 5HT0
1
2
3
4
5
**
Id3/β
-act
in e
xpre
ssio
n
Online Figure 2. Real time RT-PCR expression of Id3 mRNA normalised to β-actin in wild type (+/+) and BMPR2 heterozygous (+/-) mouse lungs under control normoxic conditions or after 2 weeks of serotonin ((5HT) exposure. Results represent the means of 3 separate experiments. *P<0.05 compared with control.
Online Figure 3
a
-11 -10 -9 -8 -7 -6 -5 -4 -30
100
200
300
400+/+ (n=4)
+/- (n=4)
Phenylephrine (log mol/L)C
ontr
actil
e re
spon
se(%
of r
espo
nse
to50
mm
ol/L
KC
l)
b
-11 -10 -9 -8 -7 -6 -5 -4 -30
100
200
300
400 +/+ (n=4)
+/- (n=4)
5-HT (log mol/L)
Con
trac
tile
resp
onse
(% o
f res
pons
e to
50m
mol
/L K
Cl)
Online Figure 3. a, cumulative concentration response curves to serotonin and b, phenylephrine in small mesenteric arteries from wild type and BMPR2+/- mice.
Online Table 1
Table 1. Affinity (pEC50) and efficacy (Emax) of serotonin (5-HT), a-methyl-5-HT and 5-carboxamidotryptamine (5-CT) in small muscular arteries from wild type (+/+) and BMPR2+/-(+/-) mice
Data shown as mean ± SEM, n= number of mice. Statistical analysis was by ANOVA with Neuman-Keuls post test ***P<0.001, *P<0.05 cf +/+ ; ¶¶¶P<0.001 cf +/- 5-HT nc. Not calculated
![Selective serotonin reuptake inhibitors [SSRIs] for stroke recoveryclok.uclan.ac.uk/6814/19/17551 - Selective serotonin reuptake... · Hackett, Maree (2012) Selective serotonin reuptake](https://static.documents.pub/doc/80x56/5f9c1bce9667ca02083a93ee/selective-serotonin-reuptake-inhibitors-ssris-for-stroke-selective-serotonin.jpg)
![Selective serotonin reuptake inhibitors [SSRIs] and ... SSRIs SNRIs prevention... · Selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine ... and tension-type](https://static.documents.pub/doc/80x56/5ce01be988c99399558de41a/selective-serotonin-reuptake-inhibitors-ssris-and-ssris-snris-prevention.jpg)
