Circulation Journal ORIGINAL ARTICLE Official Journal of the Japanese Circulation Society http://www.j-circ.or.jp Pulmonary Circulation Missense Mutations of the BMPR1B (ALK6) in Childhood Idiopathic Pulmonary Arterial Hypertension Ayako Chida, MD; Masaki Shintani, PhD; Tomotaka Nakayama, MD; Yoshiyuki Furutani, PhD; Emiko Hayama, PhD; Kei Inai, MD, PhD; Tsutomu Saji, MD, PhD; Shigeaki Nonoyama, MD, PhD; Toshio Nakanishi, MD, PhD

Background: Mutations in the bone morphogenetic receptor type 2 (BMPR2) gene, the -like 1 (ALK1) gene, and SMAD8 gene have been reported in heritable pulmonary arterial hypertension (HPAH) and in idiopathic pulmonary arterial hypertension (IPAH). However, almost 30% of HPAH cases and 60–90% of IPAH cases have no mutations in those . This suggests that there remain unidentified genes associated with HPAH and IPAH.

Methods and Results: This study screened for mutations in , SMAD1, SMAD2, SMAD3, SMAD4, SMAD5, SMAD6, SMAD7, bone morphogenetic protein receptor type 1A (BMPR1A) and bone morphogenetic protein recep- tor type 1B (BMPR1B) genes in 43 IPAH patients who had no mutations in BMPR2, ALK1 and SMAD8. Two mis- sense mutations (c.479 G>A S160N, c.1176 C>A F392L) in BMPR1B were each identified in 2 IPAH patients. Im- munoblot analysis revealed that the BMPR1B F392L protein promoted SMAD8 phosphorylation. The response to BMP was analyzed using promoter-reporter activities. The transcriptional activation of the BMPR1B F392L protein with SMAD8 increased above that of wild-type BMPR1B with SMAD8, and those of BMPR1B S160N and F392L with SMAD8 and SMAD4 were each increased above those of the wild-type BMPR1B with SMAD8 and SMAD4.

Conclusions: We identified 2 novel mutations in BMPR1B in 2 patients with IPAH. Our study suggests that BMPR1B mutations are associated with the pathogenesis of IPAH. (Circ J 2012; 76: 1501 – 1508)

Key Words: BMPR1B; Gain-of-function; Gene mutation; Pulmonary arterial hypertension

ulmonary arterial hypertension (PAH) is a progressive, sponds to sporadic disease in which there is neither a family severe, potentially fatal disease with an estimated in- history of PAH nor an identified risk factor.3 P cidence of approximately 1–2 patients per million per HPAH is inherited in an autosomal dominant fashion with year.1 In the absence of treatment, PAH leads to death with a 10–20% penetrance and affects females approximately twice median survival of 2.8 years for adults.2 as often as males.4 Bone morphogenetic protein (BMP) receptor 2 (BMPR2), a Editorial p 1329 member of the transforming growth factor (TGF)-β superfam- ily, was identified as a primary gene for HPAH on chromo- The latest classification of PAH, the “Dana Point classifica- some 2q33 in 2000.5,6 BMPR2 mutations have been identified tion”, proposed 5 subgroups of PAH: idiopathic PAH (IPAH); in more than 70% of subjects with one or more affected rela- heritable PAH (HPAH); drug- and toxin-induced PAH; PAH tives and 11–40% of IPAH.3,4 associated with other diseases such as collagen disease, HIV Other studies of the TGF-β superfamily revealed 2 further infection, portal hypertension, congenital heart disease, schis- genes responsible for PAH. Heterozygous mutations of activin tosomiasis and chronic hemolytic anemia; and persistent pul- receptor-like kinase 1 gene (ALK1), located on monary hypertension of the newborn (PPHN). IPAH corre- 12q13, were demonstrated in patients with hereditary hemor-

Received November 8, 2011; revised manuscript received January 17, 2012; accepted January 24, 2012; released online February 25, 2012 Time for primary review: 37 days Department of Pediatrics, National Defense Medical College, Tokorozawa (A.C., S.N.); Department of Pediatric Cardiology, Tokyo Women’s Medical University, Tokyo (A.C., M.S., Y.F., E.H., K.I., T. Nakanishi); and Department of Pediatrics, Toho University Medical Center, Omori Hospital, Tokyo (T. Nakayama, T.S.), Japan Grant: none. Mailing address: Toshio Nakanishi, MD, PhD, Department of Pediatric Cardiology, Tokyo Women’s Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan. E-mail: [email protected] ISSN-1346-9843 doi: 10.1253/circj.CJ-11-1281 All rights are reserved to the Japanese Circulation Society. For permissions, please e-mail: [email protected]

Circulation Journal Vol.76, June 2012 1502 CHIDA A et al.

Figure 1. Patient disposition.

rhagic telangiectasia (HHT) in association with PAH in 2001.7 Tohoku University, Kagoshima University, Sakakibara Me- Furthermore, Harrison et al demonstrated a ALK1 mutation morial Hospital, National Hospital Organization Nagasaki in 1 IPAH patient without a family history of HHT in 2005.8 Medical Center, Social Insurance Chukyo Hospital, Toyama In addition, we reported 5 ALK1 mutations in children with University, Keio University, Nagano Children’s Hospital, IPAH/HPAH.9 More rarely, mutations in endoglin (ENG) Kyoto Second Red Cross Hospital, Tsukuba University, Hok- have been identified in patients with PAH, predominantly with kaido Children’s Hospital and Medical Center, Gunma Chil- coexistent HHT.10–13 dren’s Medical Center, Okinawa Prefectural Nambu Medical In 2009, we reported the first nonsense mutation ofSMAD8 Center and Children’s Medical Center, Okayama University, in an IPAH patient who had no mutations in BMPR2 or ALK1.14 Oita University, Shizuoka Children’s Hospital, Kitano Hospi- In the same year, another group reported that pulmonary hyper- tal and Beijing Anzhen Hospital (Figure 1). These subjects in- tension and lung tumorigenesis are promoted in Smad8 mutant clude 21 IPAH/HPAH patients from our previous study 9 and mice, which reinforced our hypothesis that SMAD8 is involved 21 patients from the second cohort of another study.14 The di- in the pathogenesis of IPAH.15 agnosis of IPAH/HPAH was made through clinical evaluation, These genetic studies have considerably increased our un- chest radiography, electrocardiography, echocardiography and derstanding of the molecular basis of PAH. However, almost cardiac catheterization on the basis of current international 30% of HPAH cases and 60–90% of IPAH cases have no mu- consensus criteria; mean pulmonary artery pressure >25 mmHg tations in BMPR2, ALK1, ENG and SMAD8. at rest or >30 mmHg during exercise.18 Patients with PAH We hypothesized that other genes that belong to the TGF-β associated with another disease such as portal hypertension, superfamily or other signal pathways (BMP/MAP kinase congenital heart disease including small ventricular septal de- p38 pathway, Toll-like pathway, Rho-kinase pathway, and so fect and atrial septal defect and PPHN were excluded from this on),16,17 might be associated with the onset of IPAH/HPAH. At study by trained cardiologists. This study was approved by an the beginning, we attempted to screen for 10 genes: ENG, institutional review committee of Tokyo Women’s Medical SMAD1, SMAD2, SMAD3, SMAD4, SMAD5, SMAD6, SMAD7, University. Written informed consent was obtained from all BMP receptor type 1A (BMPR1A) and BMP receptor type 1B patients or their guardians in accordance with the Declaration (BMPR1B), involved in the TGF-β/BMP signaling pathway in of Helsinki. IPAH/HPAH patients who had no mutations in BMPR2, ALK1 and SMAD8. Molecular Analysis Genomic DNA was prepared from peripheral blood lympho- cytes or lymphoblastoid cell lines transformed by the Epstein- Methods Barr virus, as described previously.19 The BMPR2 and ALK1 Subjects coding regions and exon-intron boundaries were amplified Seventy-four unrelated IPAH/HPAH patients were recruited from genomic DNA using primers, as described in previous from Tokyo Women’s Medical University, Toho University, reports (PRIMER information was obtained from Deng et

Circulation Journal Vol.76, June 2012 BMPR1B Mutations in Childhood IPAH 1503 al6).20 Amplified products were purified using the QIAquick were as follows: anti-HA rat antibody (Roche, Mannheim, polymerase chain reaction (PCR) purification method (QIA- Baden-Württemberg, Germany), anti-Myc rabbit antibody (Cell GEN, Hilden, Germany) and screened with bi-directional direct Signaling Technology, MA, USA) and anti-phospho-Smad1/ sequencing with a ABI 3130xl DNA Analyzer (Applied Bio- Smad5/Smad8 rabbit antibody ( Technology). A systems, Foster City, CA, USA). Eighteen BMPR2 mutations human BMP4 -linked immunosorbent assay was from and 7 ALK1 mutations were detected by direct sequencing. R&D Systems (Abingdon, Oxon, UK). Some of these results were described in previous reports.9,14 After direct sequencing of BMPR2 and ALK1, multiplex Cells, Transfection and Western Blotting ligation-dependent probe amplification (MLPA) was used COS1 cells were grown in DMEM/F-12 (Sigma, St. Louis, to detect exonic deletions/duplications of BMPR2, ALK1 and MO, USA) supplemented with 10% fetal bovine serum (Gibco, ENG in 49 patients who had no mutations in BMPR2 and New York, NY, USA) and 100 units/ml penicillin-streptomy- ALK1. MLPA was performed with 100 ng of genomic DNA cin (Gibco). Transfection was performed with a Lipofectamine according to the manufacturer’s instructions using a SALSA 2000 reagent (Invitrogen, Carlsbad, CA, USA) according to MLPA HHT/PPH1 probe set (MRC-Holland, Amsterdam, The the manufacturer’s instructions. For experiments on SMAD8 Netherlands). Probe amplification products were run on an ABI phosphorylation and gene expression, some of the cells were 3130xl DNA Analyzer using a GS500 size standard (Applied cultured in DMEM/F12 containing 0.1% FBS for 4 h, before Biosystems). MLPA peak plots were visualized using Gene- the addition of human BMP4 in DMEM/F-12 containing 0.1% Mapper software v4.0 (Applied Biosystems). For each sample, FBS for 1 h. Twenty-four h after transfection, the cells were peak heights of BMPR2 probes were normalized against the lysed in lysis buffer (1 mol/L Tris-HCl [pH 8.0] 50 mmol/L, sum of all control peaks. Patients’ samples were then normal- 0.5 mol/L EDTA 1 mmol/L [pH8.0], 5 mol/L NaCl 120 mmol/L, ized to the mean of 3 normal control samples. MLPA analysis NP-40 0.25%). For Western blotting, the lysates were sepa- revealed that 5 of 49 patients had exonic deletions in BMPR2. rated on 10% resolving SDS-polyacrylamide gels, and Some of these results were described in previous studies.14 The were transferred to a polyvinylidene fluoride membrane by 30 patients who had BMPR2 or ALK1 mutations were excluded semidry blotting. For phospho-SMAD8 and SMAD8 de- from this study. tection, the membranes were blocked in TBS-T (50 mmol/L Among the 44 patients with no mutations in BMPR2 or Tris-HCl [pH 7.6], 137 mmol/L NaCl, 0.1% [w/v] Tween 20) ALK1, all coding exons and adjacent intronic regions for ENG, containing 1% bovine serum albumin (BSA) for 1 h at room SMAD1, SMAD2, SMAD3, SMAD4, SMAD5, SMAD6 and temperature. The membranes were rinsed with TBS-T and SMAD8 were amplified using polymerase chain reaction. PCR incubated with primary antibody against phospho-SMAD1/ amplified products were purified and directly sequenced like SMAD5/SMAD8 (1:5,000), Myc-SMAD8 (1:30,000) and HA- BMPR2 and ALK1. SMAD8 nonsense mutation was detected in BMPR1B (1:2,000) for 1 h at room temperature. The mem- 1 patient, as described previously.14 After the above mutation branes were rinsed with TBS-T and incubated with HRP-goat screening, we also screened SMAD7, BMPR1A and BMPR1B anti-rabbit IgG (Invitrogen) for phosphor-Smad8 and Smad8 mutations for the remaining 43 patients by direct sequencing detection, or with anti-rat IgG (Rockland, Pennsylvania, USA) (Figure 1). Available data on the characteristics and hemody- for BMPR1B detection. Blots were then washed with TBS-T namic parameters of the 43 patients with IPAH/HPAH are and bound complexes were detected using enhanced chemilu- provided in Table S1 in the online supplement. minescence (ImageQuant LAS 4000 mini, GE Healthcare). All sequences generated were compared with wild-type BMPR2 (GenBank NM_001204), ALK1 (GenBank NM_ Luciferase Assay 000020), ENG (GenBank NM_000118), SMAD1 (GenBank COS1 was transfected using the Lipofectamine 2000 re- NM_005900), SMAD2 (GenBank NM_005901), SMAD3 (Gen- agent (Invitrogen) with 3GC2-Lux and wild-type or mutant Bank NM_005902), SMAD4 (GenBank NM_005359), SMAD5 pcDNA3.0-BMPR1B and/or pcDNA3.0-SMAD8 and/or pc­ (GenBank NM_005903), SMAD6 (GenBank NM_005585), DNA3.0-SMAD4. Some of the cells were treated with human SMAD8 (GenBank NM_005905), SMAD7 (GenBank NM_ BMP4 in DMEM/F-12 containing 0.1% FBS for 12 h. Twenty- 005904), BMPR1A (GenBank NM_009009) and BMPR1B four h after transfection, the cells were harvested. Firefly and (GenBank NM_001203). renilla luciferase activities were measured with the Dual lucif- When a mutation was detected, we confirmed that it was not erase reporter assay (Promega, Madison, WI, USA) following present in 450 healthy controls by direct sequencing. the manufacturer’s instructions. Results are expressed as the ratio of firefly luciferase activity to renilla luciferase activity. Preparation of Plasmid All assays were performed in triplicate. Mouse pcDNA3.0-hemagglutinin (HA)-Bmpr1b, human pc­ DNA3.0-6xMyc-SMAD8, pcDNA3.0-SMAD4 and BMP-re- Immunocytochemistry sponsive promoter reporter construct 3GC2-Lux were kindly COS1 cells grown on glass coverslips were transfected with provided by Dr K. Miyazono (Tokyo, Japan). 3GC2-Lux con- wild-type or mutant pcDNA3.0-HA-BMPR1B using the Li- tains 3 repeats of a GC-rich sequence derived from the proxi- pofectamine 2000 reagent (Invitrogen). Cells were washed mal BMP response element in the Smad6 promoter.21 We pre- twice with phosphate-buffered saline (PBS) 48 h after trans- viously utilized the 3GC-Lux reporter gene for functional fection. Cells were then fixed for 15 min at room tempera- analysis of the SMAD8 mutant in PAH patients.14 In other ture in 4% paraformaldehyde in PBS and washed 2 times with reports, 3GC2-Lux has also been used to assess the interaction PBS. Cells were incubated in 3% BSA in TBS-T for 30 min of genes belonging to the BMP signal pathway.22–24 Mouse at room temperature for blocking nonspecific binding. After constitutively active (ca) Bmpr1b was generated by mutation blocking, cells were incubated with anti-HA rat antibody of Gln-203 into aspartic acid. (Roche, Mannheim, Baden-Württemberg, Germany; 1:3,000), Site-directed mutagenesis was carried out using a site-di- at room temperature for 1 h. After washing, detection was rected mutagenesis (Stratagene, CA, USA). The construct- achieved by incubation with anti-rat IgG antibody CY2 (Rock- ed plasmids were verified by sequencing. The antibodies used land; 1:3,000). Cells were mounted with Prolong Gold anti-

Circulation Journal Vol.76, June 2012 1504 CHIDA A et al.

Figure 2. BMPR1B mutations in idiopathic pulmonary arterial hypertension patients. (A) Two mutations, c.479 G>A p.S160N and c.1176 C>A p.F392L, were identified in 2 probands. (B) Schematic representation of BMPR1B wild type and the locations of the 2 mutations. (C) Alignment of BMPR1B proteins among humans, rhesus, mouse, dog, elephant, opossum, chicken, X_tropicalis and zebrafish showing conservation of serine 160 and phenylalanine 372 in these species. (D) Pedigrees of the patients’ families.

fade reagent with 46’-diamidino-2-phenylindole-2 HCl (DAPI) (Invitrogen). Fluorescence digital images were recorded with Results an LSM 5 PASCAL Laser Scanning Microscope (Carl Zeiss, Sequence Analysis New York, NY, USA). We screened mutations in SMAD7, BMPR1A and BMPR1B genes in 43 patients with IPAH/HPAH who had no mutations Statistical Analysis in BMPR2, ALK1 and SMAD8 (Figure 1). In this study, no All results are expressed as mean ± SD. For the statistical com- mutations were identified in SMAD7 and BMPR1A. parison of 2 samples, a 2-tailed Student’s t-test was used where We identified, however, 2BMPR1B missense mutations in applicable. Values of P<0.05 were considered significant. Sta- 2 independent probands with IPAH. In proband A, c.479 G>A tistical analyses were performed using JMP for Windows (ver- p.S160N was identified Figure( 2A). In proband B, c.1176 sion 8; SAS Institute, NC, USA). C>A p.F392L was identified (Figure 2A). As depicted in Figure 2B, BMPR1B consists of an extracellular ligand-bind- ing domain, a transmembrane domain, an amino acid stretch involved in phosphorylation called GS domain and a serine- threonine kinase domain. Mutation S160N is located outside

Circulation Journal Vol.76, June 2012 BMPR1B Mutations in Childhood IPAH 1505

Figure 3. SMAD8 is strongly phosphorylated by BMPR1B F392L with BMP4 in Western blotting. (A) The addition of BMP4 induced SMAD8 phosphorylation in the presence of wild-type or mutant BMPR1B equally. Confluent cells were stimulated with 100 ng/ml BMP4 in DMEM/0.1% FBS for 60 min, followed by lysis for total protein. (B) The ratio of phospholylated-Smad8 densitometry to myc-Smad8 densitometry expressed in-fold increases in the presence of BMPR1B F392L with BMP4. Values represent mean ± SD of 3 independent experiments. Statistical differences between groups were assessed by using a Student’s t-test. *P<0.05.

Figure 4. SMAD8 is strongly phosphorylated by ca-BMPR1B F392L. (A) The ca-BMPR1B F392L mutant strongly induced SMAD8 phosphorylation more than ca-BMPR1B and ca-BMPR1B S160N. (B) The ratio of phospholylated-Smad8 densitometry to myc- Smad8 densitometry expressed in-fold increases in the presence of ca-BMPR1B F392L. Values represent mean ± SD of 3 inde- pendent experiments. Statistical differences between groups were assessed by using a Student’s t-test. **P<0.005.

of these functional domains, and mutation F392L is located in of age revealed a mean pulmonary arterial pressure (mPAP) of the serine-threonine kinase domain. The alignment of the 66 mmHg, right atrial pressure (RAP) of 12 mmHg and cardiac BMPR1B protein between 9 distantly related species showed index (CI) of 2.0 L · min–1 · m–2. His condition progressed to that these amino acids are highly conserved (Figure 2C). World Health Organization (WHO) functional class III at 7 years of age. He has been receiving epoprostenol, home oxy- Clinical Characteristics gen therapy, a cardiotonic drug, vasodilator, anticoagulant and Proband A (Patient No. 11) When the patient was 6 years diuretics since the age of 7 years. His current condition is WHO old, right ventricular hypertrophy was identified by electro- functional class III at 17 years old. There is no family history cardiography during a health examination and he was diag- of PAH (Figure 2D). His younger brother died of viral en- nosed with IPAH. The first symptom was mild dyspnea on cephalitis at 2 years of age. The other family members have exercise at 7 years of age. His hemodynamic data at 7 years not been screened for BMPR1B mutations because their blood

Circulation Journal Vol.76, June 2012 1506 CHIDA A et al.

Figure 5. Luciferase activity induced by BMPR1B mutants. (A) Both ca-BMPR1B S160N with SMAD8 and ca-BMPR1B F392L with SMAD8 increased BMP responsive activity, and both ca-BMPR1B mutant with SMAD8 and SMAD4 induced higher activity than ca-BMPR1B with SMAD8 and SMAD4. (B) After stimulation with human BMP4 (100 pg/ml) for 12 h, BMPR1B F392L and SMAD8 with or without SMAD4 induced higher activity than wild-type BMPR1B and SMAD8 with or without SMAD4. Values represent mean ± SD of 3 independent experiments. Statistical differences between groups were assessed by using a Student’s t-test. ***P<0.0001.

samples were not obtainable. He has no malformation of the Luciferase Assay digits, limbs and genitalia. We investigated the transcriptional activity mediated by wild- Proband B (Patient No. 18) This patient’s first symptom type or mutant BMPR1B and SMAD8 with or without SMAD4 was syncope at 12 years of age. Her hemodynamic data at 7 to determine whether mutant BMPR1B could increase BMP- years of age revealed a mPAP of 111 mmHg, RAP of 11 mmHg, responsive promoter-reporter activity. CI of 3.5 L · min–1 · m–2 and a pulmonary artery wedge pressure The first luciferase assay showed that, after stimulation with of 9 mmHg. Her WHO functional class was III at 13 years of human BMP4, BMPR1B F392L and SMAD8 with or without age. She has been receiving epoprostenol, home oxygen ther- SMAD4 induced higher activity than wild-type BMPR1B and apy, vasodilator, anticoagulant and diuretics since the age of SMAD8 with or without SMAD4 (Figure 5A). 13 years. Her current condition is WHO functional class II at The second luciferase assay showed that both ca-BMPR1B 22 years old. There is no family history of PAH, but the pa- S160N with SMAD8 and ca-BMPR1B F392L with SMAD8 tient’s 39-year-old father was identified as having the same increased BMP responsive activity, and both ca-BMPR1B mutation (Figure 2D,S1). The patient’s mother did not have mutant with SMAD8 and SMAD4 induced higher activity the same mutation. The other family members have not been than ca-BMPR1B with SMAD8 and SMAD4. In particular, screened for BMPR1B mutations because their blood samples ca-BMPR1B-F392L with SMAD8 and SMAD4 induced ap- were not obtainable. She has no malformation of the digits, proximately 2-fold higher activity than ca-BMPR1B with limbs and genitalia. SMAD8 and SMAD4 (Figure 5B).

Western Blotting Analysis Immunocytochemistry The addition of BMP4 induced Myc-SMAD8 phosphorylation COS1 cells transfected with the wild-type or mutant BMPR1B in the presence of wild-type or mutant BMPR1B. In addition, were subjected to immunofluorescence staining. Wild-type and BMPR1B F392L with BMP4 increased Myc-SMAD8 phos- 2 mutants exhibited intense and equal staining of plasma mem- phorylation more than wild-type BMPR1B and BMPR1B brane and cytoplasm (Figure S2), suggesting that intracellular S160N (Figures 3A,B). production and transportation of BMPR1B were not affected As shown in Figures 4A and B, additional Western blot- by these mutations. ting analysis revealed that the ca-BMPR1B F392L mutant induced Myc-SMAD8 phosphorylation more strongly than ca-BMPR1B and ca-BMPR1B S160N. Discussion In this study, we first describe 2 missense mutations inBMPR1B

Circulation Journal Vol.76, June 2012 BMPR1B Mutations in Childhood IPAH 1507 in IPAH patients. BMPR1B is a member of the BMP family than with SMAD8 only in a living body, so these results might that belongs to the TGF-β superfamily. The TGF-β/BMP sig- be appropriate. Several studies have revealed that BMPR2 nal pathway has 2 types of receptors. There are 7 type 1 recep- mutations that are located outside of the functional domains tors (ALK1, ALK2, BMPR1A known as ALK3, ALK4, ALK5, were identified in IPAH/HPAH patients, PVOD patients and BMPR1B known as ALK6 and ALK7) and 5 type 2 recep- associated-PAH patients.4,35–37 There might be an unknown tors (ActR2A, ActR2B, TGF-β R2, AMHR2 and BMPR2).25 special function outside of the already-known functional do- BMPs bind independently to both type 1 and type 2 receptors. mains in both BMPR2 and BMPR1B. Further functional For example, BMP4 can bind to one of the type 1 receptors, analysis on mutant S160N is needed. BMPR1B, and one of the type 2 receptors, BMPR2. Upon li- The age at onset of our patients with a BMPR1B mutation gand binding, the type 2 receptors phosphorylate and activate was younger than that of other IPAH patients. It was difficult the type 1 receptors. The activated type 1 receptors propagate to identify other differences in their phenotypes. The father of the signal by phosphorylating a family of transcription fac- Proband B has the same BMPR1B mutation, but he has no tors, called Smads. BMPR1B activates SMAD1, SMAD5 and clinical signs of PAH to date. This is not surprising because SMAD8 by phosphorylation. These activated Smads complex BMPR2 mutations have very low penetrance in familial PAH. with a common partner Smad, SMAD4, and accumulate in the It is necessary to study an increased number of subjects to nucleus where they interact with transcriptional regulators for investigate the phenotype of PAH with BMPR1B mutation in target genes.25,26 detail. In addition, to undertake further analysis of the function BMPR1B mutations are known to be associated with brachy- of BMPR1B in the pathogenesis of PAH, further investigations dactyly type A2 and type C/symphalangism-like phenotype.27,28 using human PASMCs and/or animal models with BMPR1B Acromesomelic chondrodysplasia with genital anomalies is mutation will be necessary. also associated with BMPR1B mutations.29 Four missense mutations (p.I200K, p.Q249R, p.R486Q and p.R486 W) and 1 Acknowledgments 27–29 small deletion (c.del 359-366) have been reported to date. We are grateful to the patients and their family members. We thank Dr In addition, functional analysis of 3 of 4 missense mutations Kohei Miyazono for providing the plasmids. We thank Dr Hisato Yagi, revealed a loss of function in all.27,28 However, both mutations Dr Yoshihisa Matsushita and Dr Maya Fujiwara for their excellent techni- that we identified in IPAH revealed a gain-of-function in our cal assistance. We also thank Dr Rumiko Matsuoka for her supervision. experimental condition. It has been hypothesized that an imbalance of increased Funding Sources TGF-β levels and decreased BMP signals induced by BMPR2 None. mutation leads to PAH.30 In addition, ALK1 mutations in HHT 31 revealed the down-regulation of SMAD1/5/8 signaling, and Disclosures SMAD8 nonsense mutation that we identified revealed loss None. of BMP signals.14 Our findings conflict with the above hypothesis because the BMPR1B mutations caused promotion of BMP signals. References BMPR1B is expressed in human pulmonary arterial smooth 1. Gaine SP, Rubin LJ. Primary . Lancet 1998; muscle cells (PASMCs) at higher levels than in human micro- 352: 719 – 725. 2. D’Alonzo GG, Barst RJ, Ayres SM, Bergofsky EH, Brundage BH, vascular endothelial cells and human pulmonary artery en- Detre KM, et al. Survival in patients with primary pulmonary hyper- dothelial cells.32 A previous report suggested that BMPR1B tension: Results from a natural prospective registry. Ann Intern Med plays an important role in PASMC mitosis of PAH patients.33 1991; 115: 343 – 349. Quantification using real-time PCR revealed that the BMPR1B 3. Simonneau G, Robbins IM, Beghetti M, Channick RN, Delcroix M, Denton CP, et al. Updated clinical classification of pulmonary hyper- expression in PASMCs of PAH patients was more than 10- tension. J Am Coll Cardiol 2009; 54(1 Suppl): S43 – S54. fold higher than in PASMCs of controls.33 This report sug- 4. Machado RD, Eickelberg O, Elliott CG, Geraci MW, Hanaoka M, gests that BMP signals through BMPR1B are promoted in Loyd JE, et al. Genetics and genomics of pulmonary arterial hyper- PASMCs of IPAH patients. Because BMPR1B S160N and tension. J Am Coll Cardiol 2009; 54(1 Suppl): S32 – S42. 5. Lane KB, Machado RD, Pauciulo MW, Thomson JR, Phillips JA F392L promoted BMP signals in the present study, they might 3rd, Loyd JE, et al. Heterozygous germline mutations in BMPR2, be associated with the pathogenesis of PAH. Furthermore, encoding a TGF-beta receptor, cause familial primary pulmonary another study reported that disruption of BMPR2 led to dimin- hypertension: The International PPH Consortium. Nat Genet 2000; ished signaling by BMP2 and BMP4, and augmented signal- 26: 81 – 84. ing by BMP6 and BMP7 in PASMCs.34 Thus, not only inhibi- 6. Deng Z, Morse JH, Slager SL, Cuervo N, Moore KJ, Venetos G, et al. Familial primary pulmonary hypertension (gene PPH1) is caused tion but also promotion of BMP signals might be associated by mutations in the bone morphogenetic protein receptor-II gene. Am with the onset of PAH. J Hum Genet 2000; 67: 737 – 744. In this study, 1 BMPR1B mutant, F392L, strongly induced 7. Trembath RC, Thomson JR, Machado RD, Morgan NV, Atkinson C, SMAD8 phosphorylation and increasing transcriptional acti- Winship I, et al. Clinical and molecular genetic features of pulmo- nary hypertension in patients with hereditary hemorrhagic telangiec- vation in the presence of SMAD8 or SMAD8/SMAD4. In tasia. N Engl J Med 2001; 345: 325 – 334. contrast, another mutant, S160N, did not induce them so mark- 8. Harrison RE, Berger R, Haworth SG, Tulloh R, Mache CJ, Morrell edly. This difference of effect might depend on the position of NW, et al. Transforming growth factor-beta, receptor mutations and mutation in the BMPR1B gene. Mutation F392L is located in pulmonary arterial hypertension in childhood. Circulation 2005; 111: 435 – 441. one of the functional domains, the serine-threonine kinase 9. Fujiwara M, Yagi H, Matsuoka R, Akimoto K, Furutani M, Imamura domain, but mutation S160N is located outside of the func- S, et al. Implications of mutations of activin receptor-like kinase 1 tional domains. However, mutation S160N might also cause gene (ALK1) in addition to bone morphogenetic protein receptor gain-of-function because ca-BMPR1B S160N with SMAD8 II gene (BMPR2) in children with pulmonary arterial hypertension. Circ J 2008; 72: 127 – 133. and SMAD4 induced higher activity than ca-BMPR1B with 10. Harrison RE, Flanagan JA, Sankelo M, Abdalla SA, Rowell J, SMAD8 and SMAD4 in the luciferase assay. We guess that Machado RD, et al. Molecular and functional analysis identifies coexistence with SMAD8 and SMAD4 is more physiological ALK-1 as the predominant cause of pulmonary hypertension related

Circulation Journal Vol.76, June 2012 1508 CHIDA A et al.

to hereditary haemorrhagic telangiectasia. J Med Genet 2003; 40: 27. Lehmann K, Seeman P, Stricker S, Sammar M, Meyer B, Süring K, 865 – 871. et al. Mutations in bone morphogenetic protein receptor 1B cause 11. Chaouat A, Coulet F, Favre C, Simonneau G, Weitzenblum E, Soubrier brachydactyly type A2. Proc Natl Acad Sci USA 2003; 100: 12277 – F, et al. Endoglin germline mutation in a patient with hereditary 12282. haemorrhagic telangiectasia and dexfenfluramine associated pulmo- 28. Lehmann K, Seeman P, Boergermann J, Morin G, Reif S, Knaus P, nary arterial hypertension. Thorax 2004; 59: 446 – 448. et al. A novel R486Q mutation in BMPR1B resulting in either a 12. Harrison RE, Berger R, Haworth SG, Tulloh R, Mache CJ, Morrell brachydactyly type C/symphalangism-like phenotype or brachydac- NW, et al. Transforming growth factor-beta receptor mutations and tyly type A2. Eur J Hum Genet 2006; 14: 1278 – 1254. pulmonary arterial hypertension in childhood. Circulation 2005; 111: 29. Demirhan O, Türkmen S, Schwabe GC, Soyupak S, Akgül E, Tastemir 435 – 441. D, et al. A homozygous BMPR1B mutation causes a new subtype of 13. Mache CJ, Gamillscheg A, Popper HH, Haworth SG. Early-life pul- acromesomelic chondrodysplasia with genital anomalies. J Med Genet monary arterial hypertension with subsequent development of diffuse 2005; 42: 314 – 317. pulmonary arteriovenous malformations in hereditary haemorrhagic 30. Newman JH, Phillips JA 3rd, Loyd JE. Narrative review: The enigma telangiectasia type 1. Thorax 2008; 63: 85 – 86. of pulmonary arterial hypertension: New insights from genetic stud- 14. Shintani M, Yagi H, Nakayama T, Saji T, Matsuoka R. A new non- ies. Ann Intern Med 2008; 148: 278 – 283. sense mutation of SMAD8 associated with pulmonary arterial hyper- 31. Gu Y, Jin P, Zhang L, Zhao X, Gao X, Ning Y, et al. Functional tension. J Med Genet 2009; 46: 331 – 337. analysis of mutations in the kinase domain of the TGF-beta receptor 15. Huang Z, Wang D, Stansbury KI, Jones PL, Martin JF. Defective ALK1 reveals different mechanisms for induction of hereditary hem- pulmonary vascular remodeling in Smad8 mutant mice. Hum Mol orrhagic telangiectasia. Blood 2006; 107: 1951 – 1954. Genet 2009; 18: 2791 – 2801. 32. Upton PD, Long L, Trembath RC, Morrell NW. Functional charac- 16. Herpin A, Cunningham C. Cross-talk between the bone morphoge- terization of bone morphogenetic protein binding sites and Smad1/5 netic protein pathway and other major signaling pathways results in activation in human vascular cells. Mol Pharmacol 2008; 73: 539 – tightly regulated cell-specific outcomes. FEBS J 2007; 274: 2977 – 552. 2985. 33. Takeda M, Otsuka F, Nakamura K, Inagaki K, Suzuki J, Miura D, et 17. Fukumoto Y, Shimokawa H. Recent progress in the management of al. Characterization of the bone morphogenetic protein (BMP) sys- pulmonary hypertension. Circ J 2011; 75: 1801 – 1810. tem in human pulmonary arterial smooth muscle cells isolated from 18. Rich S, Dantzker DR, Ayres SM, Bergofsky EH, Brundage BH, a sporadic case of primary pulmonary hypertension: Roles of BMP Detre KM, et al. Primary pulmonary hypertension: A national pro- type 1B receptor (activin receptor-like kinase 6) in the mitotic action. spective study. Ann Intern Med 1987; 107: 216 – 223. Endocrinology 2004; 145: 4344 – 4354. 19. Yoshida MC, Satoh H, Sasaki M, Semba K, Yamamoto T, Toyoshima 34. Yu PB, Beppu H, Kawai N, Li E, Bloch KD. Bone morphogenetic K. Regional location of novel yes-related proto-oncogene, syn, on protein (BMP) type II receptor deletion reveals BMP ligand-specific human chromosome 6 at band q21. Jpn J Cancer Res 1986; 77: gain of signaling in pulmonary artery smooth muscle cells. J Biol 1059 – 1061. Chem 2005; 280: 24443 – 24450. 20. Berg JN, Gallione CJ, Stenzel TT, Johnson DW, Allen WP, Schwartz 35. Sztrymf B, Coulet F, Girerd B, Yaici A, Jais X, Sitbon O, et al. CE, et al. The activin receptor-like kinase 1 gene: Genomic structure Clinical outcomes of pulmonary arterial hypertension in carriers of and mutations in hereditary hemorrhagic telangiectasia type 2. Am J BMPR2 mutation. Am J Respir Crit Care Med 2008; 177: 1377 – Hum Genet 1997; 61: 60 – 67. 1383. 21. Ishida W, Hamamoto T, Kusanagi K, Yagi K, Kawabata M, Takehara 36. Humbert M, Deng Z, Simonneau G, Barst RJ, Sitbon O, Wolf M, et K, et al. Smad6 is a Smad1/5-induced inhibitor: Characteriza- al. BMPR2 germline mutations in pulmonary hypertension associ- tion of bone morphogenetic protein-responsive element in the mouse ated with fenfluramine derivatives.Eur Respir J 2002; 20: 518 – 523. Smad6 promoter. J Biol Chem 2000; 275: 6075 – 6079. 37. Roberts KE, McElroy JJ, Wong WP, Yen E, Widlitz A, Barst RJ, et 22. Murakami G, Watabe T, Takaoka K, Miyazono K, Imamura T. Co- al. BMPR2 mutations in pulmonary arterial hypertension with con- operative inhibition of bone morphogenetic protein signaling by genital heart disease. Eur Respir J 2004; 24: 371 – 374. Smurf1 and inhibitory Smads. Mol Biol Cell 2003; 14: 2809 – 2817. 23. Aoki H, Fujii M, Imamura T, Yagi K, Takehara K, Kato M, et al. Synergistic effects of different bone morphogenetic protein type I Supplementary Files receptors on alkaline phosphatase induction. J Cell Sci 2001; 114: Supplementary File1 1483 – 1489. Table S1. Baseline Characteristics and Hemodynamic Parameters of 24. Mochizuki T, Miyazaki H, Hara T, Furuya T, Imamura T, Watabe T, 43 IPAH/HPAH Patients et al. Roles for the MH2 Domain of Smad7 in the specific inhibition of transforming growth factor-β superfamily signaling. J Bio Chem Figure S1. Sequence analysis of BMPR1B mutation in the family of 2004; 279: 31568 – 31754. proband B. 25. Laurent D, Jean-Jacques F, Sabine B. Emerging role of bone mor- Figure S2. Subcellular distribution of wild-type BMPR1B, BMPR1B phogenetic proteins in angiogenesis. Cytokine Growth Factor Rev S160N and BMPR1B F392L in COS1 cells. 2009; 20: 203 – 212. 26. Murakami M, Kawachi H, Ogawa K, Nishino Y, Funaba M. Recep- Please find supplemental file(s); tor expression modulates the specificity of transforming growth http://dx.doi.org/10.1253/circj.CJ-11-1281 factor-beta signaling pathways. Genes Cells 2009; 14: 469 – 482.

Circulation Journal Vol.76, June 2012