Genetic ablation of the BMPR2 gene in pulmonary endothelium is sufficient to predispose to pulmonary arterial hypertension

Abstract

Background: Pulmonary arterial hypertension (PAH) is a rare but fatal lung disease of diverse origins. PAH is now further subclassified as idiopathic PAH, familial PAH, and associated PAH varieties. Heterozygous mutations in BMPR2 can be detected in 50% to 70% of patients with familial PAH and 10% to 40% of patients with idiopathic PAH. Although endothelial cells have been suspected as the cellular origin of PAH pathogenesis, no direct in vivo evidence has been clearly presented. The present study was designed to investigate whether endothelial Bmpr2 deletion can predispose to PAH.

Methods and results: The Bmpr2 gene was deleted in pulmonary endothelial cells using Bmpr2 conditional knockout mice and a novel endothelial Cre transgenic mouse line. Wide ranges of right ventricular systolic pressure were observed in mice with heterozygous (21.7 to 44.1 mm Hg; median, 23.7 mm Hg) and homozygous (20.7 to 56.3 mm Hg; median, 27 mm Hg) conditional deletion of Bmpr2 in pulmonary endothelial cells compared with control mice (19.9 to 26.7 mm Hg; median, 23 mm Hg) at 2 to 7 months of age. A subset of mice with right ventricular systolic pressure >30 mm Hg exhibited right ventricular hypertrophy and an increase in the number and wall thickness of muscularized distal pulmonary arteries. In the lungs of these mice with high right ventricular systolic pressure, the expression of proteins involved in the pathogenesis of PAH such as serotonin transporter and tenascin-C was elevated in distal arteries and had a high incidence of perivascular leukocyte infiltration and in situ thrombosis.

Conclusions: Conditional heterozygous or homozygous Bmpr2 deletion in pulmonary endothelial cells predisposes mice to develop PAH.

Figure 1

Tg(Alk1-cre)-L1 (L1cre) mice express the Cre recombinase in the pulmonary vascular endothelial cells. a and b, Lungs of 5-month-old L1cre;R26R bigenic mice were stained with X-gal for the β-galactosidase expression to visualize the cells in which Cre-mediated recombination has occurred. Sections of whole mount X-gal stained lungs were counter stained with nuclear fast red (NFR; a), or with anti-αSMA antibodies (b). Insets are magnified views of the area indicated by arrows. Note that X-gal positive cells reside in vascular ECs, but not in airway epithelial cells or in smooth muscle cells. c, PCR detection of the Bmpr2 null allele (i.e., Bmpr2^1loxP: deletion of exons 4 and 5 of Bmpr2 by Cre) from genomic DNA isolated from multiple organs/tissues of 2-month-old L1cre(-);Bmpr2^f/f (left) and L1cre(+);Bmpr2^f/f (right), demonstrating a lung-specific Cre activity. A primer set amplifying an Alk1 locus (~190 bps) was used as a control for PCR reaction. d, Representative Western blotting analysis with anti-BMPR2 antibody on whole lung lysate shows reduced levels of BMPR2 protein (~120 kd) in L1cre(+);Bmpr2^f/f lungs compared to L1cre(-);Bmpr2^f/f lungs: about 30% reduction (P<0.09, n=3). GAPDH (~36 kd) was used as a loading control.

Figure 2

Endothelial-specific Bmpr2 deletion resulted in elevation of RVSP and RV hypertrophy. a, RV systolic pressure of control, L1cre(+);Bmpr2^+/f and L1cre(+);Bmpr2^f/f mice was measured at 2-6 months of age. Each circle represents the RVSP of an individual mouse. RVSPs greater than 30 mmHg (the PH group) were shown as open circles. Proportions of PH mice were indicated above the circles. RVSP of L1cre(+);Bmpr2^f/f was significantly higher than that of controls (p = 0.001). Median value of RVSPs was used for statistical evaluation due to divergent distribution in L1cre(+);Bmpr2^+/f and L1cre(+);Bmpr2^f/f groups. Mean values of RVSPs were 23.3±0.5, 26.7±1.5, and 31.7±1.9 for control, +/f and f/f groups, respectively. b, Systemic systolic arterial pressure was not different by genotypes of mice. Mean and SEM (in parenthesis) was shown in each bar. c, The RV/(LV+S) ratio was greater in homozygous group than in control or heterozygous group (p<0.05). d, Non-PH (N-PH; open bars) and PH (solid bars) groups were evaluated separately. Statistical differences (p<0.05) were indicated by letters above each bar (e.g. `A' is different from `B' or `C'; `B,C' is different from `A', but not from `B' nor `C').

Figure 3

Mice in the PH group exhibited increased number of muscularized distal arteries and thickening of medial layers. a and b, Representative histological sections of control (a) and PH (b) lungs stained with anti-αSMA antibody. Insets in (a) and (b) are magnified views of the areas indicated by arrows. Note, readily identifiable strong αSMA-positive small arteries in the PH group lungs (b) and thickening of arterial walls (inset in b). Scale bars in insets represent 25 μm. c and d, Percentages of small arteries positive for αSMA staining (c) and medial thickness of distal arteries (d) were compared among five groups, i.e. control, PH and N-PH groups in each mutant genotype. Statistical differences were indicated by letters above each bar.

Figure 4

Vascular lesions of the PH lungs mimic some pathological features of human PAH. a and b, H&E staining of control and PH lungs, showing focal leukocyte infiltration around pulmonary arteries of PH lungs of L1cre(+);Bmpr2^f/+ or L1cre(+);Bmpr2^f/f mice. Immunohistochemical studies revealed that the infiltrated cells were mostly CD68-positive mononuclear cells (inset in b). Br, Bronchus; PA, pulmonary artery. c and d, Immunostaining with anti-αSMA (c, d) and elastic van Giesen staining (d) showed excessive thickening of aSMA-positive layers of small arteries. EL, external elastic lamina. e, Anti-von Willebrand factor (vWF; d) antibodies showed occluded arteries resembling concentric vascular lesions. f, H&E staining showing a representative in situ thrombotic lesion in a PH lung. Lumen of the affected vessel was occluded by formation of fibrin(ogen)-positive thrombus (inset). Scale bars represent 50 μm.

Figure 5

L1cre(+);Bmpr2^f/f mice showed a higher proliferation index in ECs and SMCs of 4week-old lungs. a and b, Lung samples from four week old L1cre(-);Bmpr2^f/f (a) and L1cre(+);Bmpr2^f/f (b) mice were immunostained with PCNA antibodies followed by light hemotoxylin counter staining. Distal small arteries of about 50 μm diameter are shown as representative images. ECs and SMCs are marked with arrows and arrows with asterisks, respectively. The purplish staining represents PCNA-negative nuclei (indicated by blue arrows), while brown staining represents PCNA-positive nuclei (indicated by red arrows). Scale bars represent 30 μm. c and d, The ratios of PCNA-positive nuclei to total nuclei of ECs (c) and of SMCs (d) at distal arteries were compared between two groups. Significantly higher ratios of PCNA-positive ECs and SMCs were detected (p<0.001, n=7). Pulmonary hypertension (PH) is a lung disease of diverse etiologies. PH is classified into arterial, venous, hypoxic, thromboembolic, and miscellaneous varieties. Of these varieties, pulmonary arterial hypertension (PAH) typically carries the worst prognosis. PAH is promoted by the imbalance of hormones, growth factors, neurotransmitters, or environmental stresses that leads to pulmonary vascular constriction, cell proliferation, or remodeling. Bone morphogenetic protein receptor type II (BMPR2) signaling plays a critical role in PAH pathogenesis, as germline mutations of BMPR2 are associated with about 25-30% of all PAH cases. PAH pathogenesis involves multiple vascular and non-vascular cell types. We show that mice with the genetic ablation of Bmpr2 in pulmonary endothelial cells (pECs) exhibited an elevation of RVSP, RV hypertrophy, and histopathological features reminiscent of human PAH lungs, demonstrating for the first time in vivo that Bmpr2 mutation in endothelium is sufficient to predispose to PAH. Our data suggest that impaired BMPR2 signaling in pECs may increase the risk of pECs to damages that render the pulmonary vessels more susceptible to dysregulated remodeling. One of the major impediments for PAH studies is limited access to biological samples because pathological samples are available only from lung explants and autopsy specimens, at the very late stage of the disease. Animal models that reproduce key features of PAH provide relevant pathological samples from early to late stages of PAH. Development of non-invasive monitoring systems to detect PH from these mutant mice at early phase will facilitate the usefulness of this animal model for various PAH studies.