Investigation of Second Genetic Hits at the BMPR2 Locus as a Modulator of Disease Progression in Familial Pulmonary Arterial Hypertension
Background— Primary pulmonary arterial hypertension (PAH) is a potentially devastating condition resulting from occlusion of the pulmonary arterioles by the formation of vascular lesions. Heterozygous mutations in the gene encoding the bone morphogenetic protein receptor type II (BMPR2) have been identified in both familial (FPAH) and idiopathic PAH. Mutant alleles are typically of low penetrance, indicating that other factors are required for the onset of PAH. Previous reports have suggested that the characteristic plexiform lesions in affected lungs are akin to neoplasia, showing monoclonal expansion and microsatellite instability. We hypothesized that in patients with germline mutations, BMPR2 might behave as a classic tumor suppressor gene, with somatic loss of the wild-type allele contributing to disease progression.
Methods and Results— To test this hypothesis, plexiform and concentric vascular lesions were serially microdissected from lung explant tissue derived from 7 FPAH cases. DNA was analyzed for loss of heterozygosity at BMPR2 and for microsatellite instability (MSI) at 5 loci. MSI was detected in 1 of 37 lesions at a single locus, BAT-26, whereas heterozygosity at BMPR2 was retained at all informative loci. We also describe a FPAH patient carrying biallelic constitutional missense mutations of BMPR2 who manifested disease at a stage and manner similar to heterozygous patients.
Conclusions— Taken together, these data demonstrate that MSI is uncommon in FPAH and suggest that somatic loss of the remaining wild-type BMPR2 allele in heterozygous mutation carriers likely does not play a significant role in modulating the onset or progression of FPAH.
Received August 25, 2004; revision received September 24, 2004; accepted October 5, 2004.
Primary pulmonary arterial hypertension (PAH) is an often fatal condition typified by the sustained elevation of mean pulmonary artery pressure, which leads to hypertrophy and subsequent failure of the right heart. Histologically, the disease is characterized by obstruction of the small pulmonary arteries resulting from the development of vascular plexiform and concentric lesions that in turn are defined by proliferating and migrating endothelial and smooth muscle cells.1,2 Although the majority of PAH patients spontaneously present with disease and are regarded as idiopathic cases (IPAH), up to 10% may have a family history of the condition (FPAH). Inheritance of PAH is autosomal-dominant with a sex bias to the reduced penetrance of mutant alleles.3 We and others have demonstrated that diverse germline heterozygous mutations in the gene encoding the bone morphogenetic protein receptor type II (BMPR2) underlie the majority of patients with FPAH and additionally, up to 40% of apparently sporadically affected subjects.4–7
See p 534
BMPR-II is a type 2 receptor of the transforming growth factor (TGF)-β family of cytokines, members of which are regarded as essential for the cellular processes of proliferation, differentiation, and apoptosis. The receptor comprises 4 functional domains (ligand binding, kinase, transmembrane, and cytoplasmic) and relays intracellular signals in complex with a type I receptor (BMPR-1A/1B) via the phosphorylation of a series of cytoplasmic mediators, of which the Smad proteins are the most extensively characterized. Activated Smads, in concert with other signaling partners, translocate to the nucleus, where they regulate the transcription of target genes.8 Although the Smad signaling pathway is regarded as a major target of TGF-β signal transduction, several Smad-independent substrates are now being elucidated, including the mitogen-activated protein kinases (MAPKs), extracellular signal–regulated kinase, c-Jun NH2-terminal kinase, and p38MAPK.9
Approximately 60% of the pathogenic mutations reported to date introduce premature truncation codons to the BMPR2 transcript, most of which are anticipated to undergo nonsense-mediated decay, supporting haploinsufficiency of the protein as the molecular mechanism of disease.10 By contrast, the remainder are point mutations that occur at functionally critical amino acid residues and have been independently demonstrated to have heterogeneous effects on important receptor functions, including ligand binding and phosphorylation of substrate.11,12 In PAH kindreds, mutant BMPR2 alleles display reduced penetrance, indicating that heterozygous mutation of the gene is required but not sufficient for the precipitation of disease.2 Other genetic or environmental factors are therefore also implicated.
Two studies13,14 observed that a proportion of abnormally proliferating endothelial cells from plexiform lesions in a series of patients with IPAH and others who had developed the disease secondary to the ingestion of appetite suppressant drugs displayed monoclonal expansion. The monoclonal cell growth within such lesions was interpreted as being akin to neoplasia, suggesting that PAH onset necessitated somatic mutations in genes that conferred a selective growth advantage to cells.13,15 In support of this hypothesis, microsatellite instability (MSI) and/or mutation of TGFβRII or the proapoptotic gene BAX was subsequently demonstrated in a proportion of plexiform lesions from PAH lungs.15
Because several other genes in the TGF-β/BMP superfamily have known roles in neoplasia,16 we hypothesized that FPAH and BMPR2 mutation–positive IPAH might follow the classic “2-hit” model of tumorigenesis and that inactivation of the remaining wild-type BMPR2 allele might be one of the somatic mutations necessary to precipitate disease. To test this theory, we examined microdissected pulmonary vascular concentric and plexiform lesions from 7 FPAH cases for evidence of somatic loss of the wild-type BMPR2 allele. We also describe a PAH patient who showed a typical disease course, despite harboring constitutional missense mutations of both BMPR2 alleles.
Patient ascertainment was performed as previously described.10 In brief, subjects were defined as affected if pulmonary artery mean pressure exceeded 25 mm Hg and pulmonary artery wedge pressure was normal or if postmortem analysis identified plexogenic pulmonary arteriopathy.2 Subjects with FPAH were defined by having at least 1 affected relative displaying evidence of linkage to the BMPR2 locus. All patients were of western European descent. Patient 1 (GER02) was of German origin with 1 affected sibling. Patient 2 (FRA01), of French origin, had an affected parent and 3 affected siblings. Patients 3 (UK01), 4 (UK07), 5 (UK11), 6 (UK13), and 7 (UK14) were from the United Kingdom and had 8, 1, 2, 1, and 1 affected relatives, respectively. Germline mutations were detected among patients and carriers of the BMPR2 disease haplotype in families of UK01, UK11, and UK13, as previously described (Table 1).
Patient 5559, who died at the age of 20 years, was diagnosed with PAH after cardiac catheterization revealed pulmonary artery pressure of 154/69 mm Hg. This individual had a recorded paternal family history of PAH but no known maternal history. Samples were not available from either parent.
Mutation Analysis of the BMPR2 Gene in Case 5559
DNA was isolated from peripheral blood and fresh-frozen lung samples by standard methods. The 13 coding exons of the gene and flanking intronic sequences were amplified by polymerase chain reaction (PCR) with primer pairs described elsewhere.10 Both strands of the purified PCR product were sequenced with the same primers on an ABI377 DNA analyzer with the BigDye terminator kit. The presence of identified mutations was excluded from a panel of 150 normal chromosomes, also by direct sequencing. To determine the relative proportions of the 2 mutations in the compound heterozygote patient, PCR products generated from blood and lung sources were cloned into the PCR2.1 vector according to the manufacturer’s instructions (Promega). Transformation into the Escherichia coli strain DH5α was performed by the heat-shock procedure and overnight cultures miniprepped with the Qiagen kit. A total of 70 colonies were prepared from each source, and sequencing was carried out as described earlier.
Laser Capture Microdissection
Serial paraffin sections (10 μm) were mounted on membrane-based slides for microdissection in conjunction with 1M1000 digital imaging software (Leica Microsystems). Immediately before microdissection, sections were stained with hematoxylin and eosin to permit identification of tissue components. At least 6 serial sections were obtained from each patient sample, and all identified vascular lesions were collected. Concentric and plexiform lesions were sampled, together with random fields of normal lung parenchyma, as detailed in Table 1. Lesions were highlighted and dissected from tissue sections with a laser beam (0.6 μm, at 32 mW, 30 Hz) and an 0.8-second pulse, as illustrated in Figure 1.
Dissected material was contained in a sterile PCR plastic collecting tube cap containing 40 μL of lysis buffer (TE [10 mmol/L Tris-HCl, 1 mmol/L EDTA], pH 8.0); 1 mg/mL proteinase K; 1% Tween-20). For DNA extraction, samples were incubated overnight at 37°C. An additional 4 μL of proteinase K (20 mg/mL) was added to each sample before each of 2 additional overnight incubations at 65°C. The samples were finally incubated at 95°C for 10 minutes to inactivate the enzyme.
MSI and Loss of Heterozygosity Analysis
MSI analysis was performed with 5 microsatellite markers: BAT-25, a thymine repeat in intron 16 of the c-kit oncogene; BAT-26, an adenine repeat located within the fifth intron of the DNA mismatch repair gene hMSH2; a guanine repeat in exon 3 of BAX; a 10-base adenine repeat in exon 3 of the TGFβRII gene17–20; and a 7-base adenine repeat in BMPR2. All familial and sporadic cases were analyzed. Primer details for these microsatellites are available on request.
Three polymorphic markers were used to detect loss of heterozygosity (LOH) at the BMPR2 locus: a previously undescribed marker within the first intron of BMPR2; D2S3009, located in intron 12; and D2S2289, which maps 420 kb distal of the gene. Markers used in the MSI study were amplified with fluorescently labeled primers as either a duplex (TGFβRII and BMPR2) or a multiplex (BAT-25, BAT-26, and BAX) PCR. Markers used in the analysis of LOH were amplified individually. Each reaction comprised 1× Qiagen multiplex Mastermix, 0.2 μmol/L of each primer, and 2 μL of extracted genomic DNA to a final volume of 10 μL. Markers were visualized on 6% polyacrylamide gels on an ABI 377 DNA analyzer and with the use of GeneScan 3.0 software. PCR cycling conditions were as follows: preincubation at 95°C for 15 minutes to activate the polymerase; then 95°C for 30 seconds, 57°C for 90 seconds, and 72°C for 60 seconds for 40 cycles; followed by a final step of 60°C for 30 minutes.
Molecular Genetic Analysis of a Patient With Biallelic BMPR2 Mutations
Both R491Q and K512T mutations were observed as heterozygous mutations in exon 11 of BMPR2 on sequencing of DNA isolated from 2 independent sources of tissue, namely, blood and lung. The PCR products of the exon 11 generated from these tissues were next cloned to determine (1) whether the mutations were in the cis or trans and (2) whether there was evidence of a wild-type allele, which would indicate mosaicism. Among the 140 clones analyzed, 70 from each tissue, the 2 mutations were always detected on different chromosomes, thus confirming the patient to be a compound heterozygote for BMPR2 (Figure 2). Additionally, the mutations were present in approximately equal proportions in both tissues (Figure 2A) and importantly, there was no evidence of a wild-type allele. Neither mutation was present in the control panel of 150 normal chromosomes. Furthermore, both wild-type amino acids at positions 491 and 512 in the BMPR-II cDNA demonstrated evolutionary conservation across species, indicating their relevance to receptor function (Figure 2B).
Investigation of LOH at the BMPR2 Locus in Microdissected Human Lung
A total of 37 vascular lesions (21 plexiform, 16 concentric) were analyzed from 7 FPAH cases, and the results were compared with normal lung parenchyma in each case. D2S3009 was informative in all 7 cases, whereas D2S2289 and the intron 1 polymorphisms were each informative in 6 cases (Table 2). Successful PCR amplification was achieved in 29 of 32 (91%) of informative lesions for intron 1, 35 of 37 (95%) for D2S3009, and 30 of 32 (94%) for D2S2289. In all cases wherein genotyping was successful, heterozygosity was retained at all informative loci (Table 2 and Figure 3).
No Evidence of MSI of 5 Genes in PAH Human Lung
MSI analysis was conducted on the same 37 lesions, together with normal parenchyma. Success rates for PCR amplification ranged from 31 of 37 lesions (84%) for TGFBR2 to 36 of 37 (97%) for BAT-25 and BAX (Table 3). Aberrantly sized fragments, indicative of microsatellite mutation, were identified in a single concentric lesion from a familial case for marker BAT-26 and subsequently confirmed in a second independent PCR (Figure 4). All other results were normal (Table 3 and Figure 4).
Profound MSI is a hallmark of hereditary nonpolyposis colon carcinoma (HNPCC) and is also found in a proportion of sporadic HNPCC-spectrum tumors, such as endometrial carcinoma.21 The underlying cause of MSI is a defect in mismatch repair, which results in tumorigenesis through an accumulation of somatic mutations in genes important for regulating cell cycle, growth, or apoptosis. A lower level of MSI occurs in tumors that are outside the HNPCC spectrum. Previous studies of endothelial cells microdissected from plexiform lesions of PAH lungs have shown monoclonal expansion in 17 of 22 lesions (77%) from 4 patients and microsatellite mutation rates ranging from 21% for BAX to 50% for BAT26.13,15 This suggested that endothelial cell expansion in plexiform lesions is akin to neoplasia and might result from an accumulation of somatic mutations, either through MSI or other mutational mechanisms. We have now conducted similar analyses in a series of FPAH cases in whom BMPR2 has been fully characterized. We hypothesized that at least in those cases with a germline BMPR2 mutation, BMPR2 might behave as a classic tumor suppressor gene and undergo somatic loss of the wild-type allele in pulmonary vascular lesions, as documented for the type I receptor BMPR1A in juvenile polyposis and Cowden syndromes.22
We performed serial microdissection of plexiform and concentric vascular lesions and normal lung parenchyma on paraffin-embedded lung from 7 FPAH patients. All had previously been analyzed for a germline BMPR2 mutation by a combination of direct sequencing, Southern blotting, and dosage PCR, and mutations had been detected in 3 of these cases.
Mononucleotide-repeat tracts in 5 genes were examined for MSI, including BAT-25 and BAT-26, which are the recommended diagnostic markers for HNPCC.23 Only 1 mutation was found, in a single gene (BAT-26) in 1 concentric lesion; no mutations were identified among 19 plexiform lesions, in contrast with the results of Yeager et al.15 There are 2 significant differences between the 2 studies that might account for this. First, Yeager et al15 used whole-genome amplification (WGA) of microdissected DNA before analysis of individual genes. In the present study, an optimized DNA extraction protocol obviated the need for WGA. WGA can be prone to allele dropout or other artifacts, but these were controlled for in the previous PAH study. Thus, an alternative explanation is that our results might represent a difference between BMPR2 mutation-positive and mutation-negative cases. Yeager et al studied IPAH cases, and although their BMPR2 status was not determined, mutations have been detected in fewer than half of IPAH patients.6,7 In the present study, we focused on familial cases that were linked to BMPR2, including 3 with characterized mutations. MSI is apparently an uncommon phenomenon among this group of patients. Similarly, we examined whether somatic LOH at the BMPR2 locus occurred in pulmonary vascular lesions from these patients. Heterozygosity was retained in all informative cases for 3 markers that encompass the BMPR2 gene. These results suggest that somatic loss of the wild-type BMPR2 allele in affected lung is not a function of disease progression in FPAH. However, we have previously noted an almost-complete absence of BMPR-II protein in FPAH lung on immunohistochemistry.24 Thus other “second hits,” such as somatic mutation, hypermethylation of the BMPR2 promoter, or other factors secondary to perturbation of upstream regulatory proteins, remain possible mechanisms.
The BMPR2 gene has been knocked out in the mouse by 2 independent groups in an effort to assess its importance in early development. As reported by Beppu et al,25 although heterozygous mice harboring an ablated BMPR2 allele appear phenotypically normal, mutant homozygotes die in utero at gastrulation. A second homozygous mutant of BMPR2, generated by the inframe deletion of exon 2, survived longer than the knockout model but was nevertheless lethal during embryogenesis.26 These findings strongly argue that at least 1 functional BMPR2 allele is necessary for survival, therefore raising the question of how the biallelic germline pathogenic mutation in the patient we describe herein was compatible with normal development and survival until adulthood.
The R491Q mutation has been described in other PAH families5 and is thus, clearly pathogenic. K512T has not previously been reported in PAH cases. However, this mutation was not detected among 150 control chromosomes. Furthermore, this mutation has previously been characterized functionally and was shown to perturb BMPR-II signaling.11 It is therefore unlikely to be an innocuous polymorphism. A second possible explanation is that 1 of the 2 mutations arose postzygotically, and the patient might be a mosaic, with a sufficient proportion of wild-type alleles to rescue development. The patient’s father was also affected with PAH, and thus, 1 mutation is presumed to be paternally inherited. Parental DNA samples were not available, and thus, it is not possible to determine whether the second mutation was inherited from the unaffected mother. However, no wild-type alleles were detected among 70 clones analyzed from each of blood and lung-derived DNA. This argues strongly against mosaicism resulting from a late-postzygotic mutation and suggests that patient 5559 is constitutionally compound heterozygous for 2 different BMPR2 mutations.
Both of these missense mutations have been extensively characterized, and although both clearly perturb the BMPR-II signaling pathway, their effects are subtly distinct biochemically. The kinase p38MAPK is normally a downstream target of BMPR-II after activation of the receptor by ligand. However, both mutant receptors elevate the phosphorylation of this substrate constitutively, with the K512T mutation stimulating this effect to a significantly greater degree than R491Q. Perhaps the most critical distinction between these 2 mutations lies in their respective effects on activation of the Smad pathway. In luciferase reporter assays, the kinase domain mutation R491Q leads to a reduction of activation of the reporter to below endogenous levels, consistent with a dominant-negative effect on Smad signaling. In contrast, K512T reduced reporter gene activity by comparison with cells transfected with the wild-type receptor but did not compromise endogenous signaling activity.11 We therefore propose that K512T is a pathogenic mutation that impairs BMPR-II signaling activity but retains sufficient residual Smad activity to permit normal embryonic development in the compound-heterozygote patient. The observation that despite apparently minimal BMPR-II activity, her age at PAH diagnosis and subsequent disease progression were comparable to those in individuals with heterozygous BMPR2 mutations, and this suggests that further perturbation of BMPR-II signaling in heterozygous carriers is not a significant factor in the precipitation or progression of FPAH.
In conclusion, we have examined microdissected pulmonary vascular lesions from a series of FPAH patients. MSI was uncommon, being found in only 1 of 37 lesions for a single mononucleotide repeat. Furthermore, no evidence for LOH was found at the BMPR2 locus. We have also investigated a patient with compound heterozygosity for 2 distinct BMPR2 mutations. Surprisingly, the course of the disease in this patient was not dissimilar to the typical presentation and progression of PAH. Taken together, these 2 seemingly disparate sets of data suggest that somatic loss of wild-type BMPR2 alleles or mutation of proapoptotic genes is unlikely to play a significant role in the pathogenesis of FPAH. The identification of other genetic or environmental factors that modulate the onset and progression of this disease remains an important priority.
This study was supported by a program grant from the British Heart Foundation (BHF) to Drs Trembath and Morrell (Ref RG/2000012). Victoria James is supported by a BHF PhD studentship (Ref FS/02/082/14741), Dr Harrison by a BHF clinical PhD studentship (Ref FS/2001069), and Dr Aldred by a Wellcome Trust Advanced Training fellowship (Ref 064271/Z/01/Z). We are particularly grateful to Phil Grocott, Leica Microsystems Ltd, for providing access to the laser microdissection instrument. We also thank Nita Williams and Charis Eng, Ohio State University, for communicating their laser capture microdissection DNA extraction protocol, and Jenny Edwards and Peter Greaves, MRC Toxicology Unit, Leicester, for assistance in preliminary laser capture microdissection optimizations.
↵*The first 2 authors contributed equally to this work.
Deng Z, Morse JH, Slager SL, Cuervo N, Moore KJ, Venetos G, Kalachikov S, Cayanis E, Fischer SG, Barst RJ, Hodge SE, Knowles JA. Familial primary pulmonary hypertension (gene PPH1) is caused by mutations in the bone morphogenetic protein receptor-II gene. Am J Hum Genet. 2000; 67: 737–744.
Thomson JR, Machado RD, Pauciulo MW, Morgan NV, Humbert M, Elliot GC, Ward K, Yacoub M, Mikhail G, Rogers P, Newman J, Wheeler L, Higenbottam T, Gibbs JS, Egan J, Crozier A, Peacock A, Allcock R, Corris P, Loyd JE, Trembath RC, Nichols WC. Sporadic primary pulmonary hypertension is associated with germline mutations of the gene encoding BMPR-II, a receptor member of the TGF-β family. J Med Genet. 2000; 37: 741–745.
Massague J, Chen Y-G. Controlling TGF-β signalling. Genes Dev. 2000; 14: 627–644.
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, Nichols WC. BMPR2 haploinsufficiency as the inherited molecular mechanism for primary pulmonary hypertension. Am J Hum Genet. 2001; 68: 92–102.
Rudarakanchana R, Flanagan JA, Chen H, Upton PD, Machado R, Patel D, Trembath RC, Morrell NW. Functional analysis of bone morphogenetic protein type II receptor mutations underlying primary pulmonary hypertension. Hum Mol Genet. 2002; 11: 1517–1525.
Nishihara A, Watabe T, Imamura T, Miyazono K. Functional heterogeneity of bone morphogenetic protein receptor-II mutants found in patients with primary pulmonary hypertension. Mol Biol Cell. 2002; 13: 3055–3063.
Rampino N, Yamamoto H, Ionov Y, Li Y, Sawai H, Reed JC, Perucho M. Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype. Science. 1997; 275: 967–969.
Bapat BV, Madlensky L, Temple LKF, Hiruki T, Redston M, Baron DL, Xia L, Marcus VA, Soravia C, Mitri A, Shen W, Gryfe R, Berk T, Chodirker BN, Cohen Z, Gallinger S. Family history characteristics, tumor microsatellite instability and germline MSH2 and MLH1 mutations in hereditary colorectal cancer. Hum Genet. 1999; 104: 167–176.
Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshleman JR, Burt RW, Meltzer SJ, Rodriguez-Bigas MA, Fodde R, Ranzani GN, Srivastava S. A National Cancer Institute Workshop on Microsatellite Instability for Cancer Detection and Familial Predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res. 1998; 58: 5248–5257.
Markowitz SD, Wang J, Myeroff LL, Parsons R, Sun L, Lutterbaugh J, Fan RS, Zborowska E, Kinzler KW, Vogelstein B, Brattain MG, Willson JKV. Inactivation of the type II TGF-β receptor in colon cancer cells with microsatellite instability. Science. 1995; 268: 1336–1338.
Peltomäki P. Deficient DNA mismatch repair: a common etiologic factor for colon cancer. Hum Mol Genet. 2001; 10: 735–740.
Zhou XP, Woodford-Richens K, Lehtonen R, Kurose K, Aldred M, Hampel H, Launonen V, Virta S, Pilarski R, Salovaara R, Bodmer WF, Conrad BA, Dunlop M, Hodgson SV, Iwama T, Jarvinen H, Kellokumpu I, Kim JC, Leggett B, Markie D, Mecklin JP, Neale K, Phillips R, Piris J, Rozen P, Houlston RS, Aaltonen LA, Tomlinson IP, Eng C. Germline mutations in BMPR1A/ALK3 cause a subset of cases of juvenile polyposis syndrome and of Cowden and Bannayan-Riley-Ruvalcaba syndromes. Am J Hum Genet. 2001; 69: 704–711.
Umar A, Boland CR, Terdiman JP, Syngal S, de la Chapelle A, Ruschoff J, Fishel R, Lindor NM, Burgart LJ, Hamelin R, Hamilton SR, Hiatt RA, Jass J, Lindblom A, Lynch HT, Peltomäki P, Ramsey SD, Rodriguez-Bigas MA, Vasen HF, Hawk ET, Barrett JC, Freedman AN, Srivastava S. Revised Bethesda guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability. J Natl Cancer Inst. 2004; 96: 261–268.
Atkinson C, Stewart S, Upton PD, Machado R, Thomson JR, Trembath RC, Morrell NW. Primary pulmonary hypertension is associated with reduced pulmonary vascular expression of type II bone morphogenetic protein receptor. Circulation. 2002; 105: 1672–1678.
Delot EC, Bahamonde ME, Zhao M, Lyons KM. BMP signaling is required for septation of the outflow tract of the mammalian heart. Development. 2003; 130: 209–220.