Pulmonary arterial hypertension (PAH) is a severe disease affecting small pulmonary arteries, with a progressive remodeling leading to elevated pulmonary vascular resistance and right ventricular failure 123. PAH can be idiopathic, heritable or associated with drug or toxin exposure or other conditions such as connective tissue diseases, human immunodeficiency virus infection, congenital heart diseases, and portal hypertension 12. Germline mutations in the bone morphogenetic protein receptor type 2 (BMPR2) gene are detected in 10 to 40% of idiopathic PAH and in 58% to 74% of patients with a family history of PAH 456. Accumulated evidence indicates that PAH patients carrying a BMPR2 mutation develop PAH approximately 10 years earlier than non-carriers, with a more severe hemodynamic compromise at diagnosis, and are less likely to respond to acute vasodilator testing 456. In addition, few cases of PAH in ACVRL1 and Endoglin mutants have been reported 6789. PAH patients with an ACVRL1 mutation are characterized by a younger age at diagnosis and death, as compared to PAH patients without BMPR2 and ACVRL1 mutation 6.

A recent report in Respiratory Research by Austin and colleagues 10 showed that PAH patients with missense mutations in BMPR2 gene had more severe disease than patients with truncating mutation with a significant younger age at diagnosis, a younger age at death and a shorter survival from diagnosis to death or lung transplantation. Moreover, they demonstrated a statistically significant difference in age at diagnosis between carriers and non-carriers, and subgroup analysis revealed this to be the case only for females 10.

To test the influence of gender in patients from the French PAH network, we decided to compare clinical, functional, and hemodynamic characteristics between PAH patients carriers of a BMPR2 mutation and patients without identified mutation, according to gender. To analyse the influence of the type of BMPR2 mutation in clinical outcomes of PAH, clinical phenotypes were compared between BMPR2 missense mutation carriers, BMPR2 truncating mutation carriers and patients carriers of a large rearrangement or a splice defect in the BMPR2 gene.

In this study, we show that BMPR2 mutation carriers are younger at diagnosis with a more severe hemodynamic compromise, but this is not influenced by gender. However, we observed a trend for more severe prognosis of the disease in males, and particularly in male patients carrying a BMPR2 mutation. This observation is consistent with our recent observation that PAH mortality is most closely associated with male gender 16. Moreover, we compared BMPR2 missense mutation carriers, BMPR2 truncating mutation carriers and patients carrying a splice defect or a large rearrangement in the BMPR2 gene. We found no influence of mutation type on clinical phenotypes in BMPR2 mutation carriers with PAH.

Although no significant impact of gender was observed on age at diagnosis and outcomes in our patient population, it should be emphasized that PAH mostly occur in females, irrespective of BMPR2 mutation status (sex ratio females/males: 2.4 in BMPR2 mutation non-carriers, and 2.4 in BMPR2 mutation carriers). To explain overrepresentation of female patients it has been suggested that estrogens and estrogen metabolism might participate in the pathogenesis of PAH 1718. Estrogen is a potent mitogen of pulmonary vascular smooth muscle cells, and cytochrome P450 1B1 (CYP1B1), highly expressed in lung and particularly in endothelial cells, catalyzes the oxidation of estrogens to 2-hydroxy (2-OHE) and 4-hydroxy (4-OHE) estrogens. West et al. 17 have showed significantly decreased transcript levels of CYP1B1 in affected female compared to unaffected females with a BMPR2 mutation. They hypothesized that low level of CYP1B1 might result in increased local concentrations of estrogens, increasing the risk of PAH. Moreover, in extrahepatic tissue, oxidation of estrogens also occurs by hydroxylation at the C-16 position by other P450 enzymes, predominantly resulting in 16α-hydroxyestrone (16α-OHE1) which stimulates cellular proliferation by constitutively activating the estrogen receptor. Austin et al. 18 demonstrated a decrease of the 2-OHE/16α-OHE1 ratio in affected female compared to non-affected females with a BMPR2 mutation. Thus, altered estrogen metabolism could contribute to the penetrance of PAH in women and CYP1B1 could be a sex specific modifier gene.

Secondly, we compared clinical phenotypes of BMPR2 missense mutation carriers, BMPR2 truncating mutation carriers and patients carrying a splice defect or a large rearrangement in BMPR2 gene. Whereas Austin and colleagues 10 showed that PAH patients with missense mutations in BMPR2 gene had more severe disease than patients with truncating mutation with a significant younger age at diagnosis and a shorter survival from diagnosis to death or lung transplantation, we did not find any influence of mutation type on disease pattern or natural history among patients with BMPR2 mutation in the French PAH network. Austin and colleagues 10 hypothesized that this difference may be due to the fact that truncating mutations are degraded by nonsense-mediated decay (NMD), a mRNA surveillance mechanism that detects and degrades mRNA transcripts containing premature termination codons (PTCs), leaving only the wild-type mRNA detectable. By contrast, missense mutations escape NMD resulting in stable transcript that may produce dominant negative proteins 1920. Because some truncating mutations escape NMD, Austin et al. confirmed the activity of NMD pathway in BMPR2 truncating mutation carriers. BMPR2 truncating mutations escaping NMD pathway may represent a minority of cases, with only 7 out of the 62 BMPR2 truncating mutation carriers escaping NMD in this study 10. Indeed, PTCs located more than 55 nucleotides upstream the last exon escape NMD. Analysis of all the truncating mutations in the French PAH network showed that 9 out of the 51 BMPR2 truncating mutation carriers had mutations predicted to escape NMD (Additional file 1). Nevertheless, if we excluded these 9 patients from the analysis of impact of BMPR2 mutation type on clinical outcomes, no significant differences were seen between BMPR2 missense mutation carriers, BMPR2 truncating mutation carriers and patients carrying a splice defect or a large rearrangement in BMPR2 gene. However, this analysis cannot include all BMPR2 mutations that could potentially escape NMD, because it has been demonstrated that truncating mutation may also escape NMD pathway by reinitiation of translation by use of an alternative translation start site downstream the PTC, producing a BMPRII protein of a lower molecular weight 1920. Indeed, other factors may influence the relation between BMPR2 mutation type and phenotype of PAH. It has been demonstrated that the level of expression of the wild-type BMPR2 allele may influence the clinical development of PAH in carriers of truncating BMPR2 mutations 21. Moreover, different missense BMPR2 mutations could lead to different level of BMPRII protein expression on cell surface, depending on the level of cytosolic retention of BMPR-II transcript 2223. All these factors may explain the difficulty to correlate severity and penetrance of PAH to the BMPR2 mutation type.

The differences observed between our data and those reported by Austin et al. related to gender, may be explained by a selection bias with a low number of PAH patients without a BMPR2 mutation in the cohort published by Austin et al. (only 10 male BMPR2 mutation non-carriers as compared to 79 in the French PAH Network). This low number of male patients not carrying a BMPR2 mutation may be due to a preferential recruitment of patients with a family history of PAH in the study by Austin. Indeed, 74% of their patients had such a family history, as compared to 22% in the cohort of New York Presbyterian Pulmonary Hypertension Center 5 and 20% in the French PAH network 6. This resulted to 72% of PAH patients carrying a BMPR2 mutation in the cohort reported by Austin and colleagues, as compared to 17% in the cohort from the New York Presbyterian Pulmonary Hypertension Center 5, and 25% in the French PAH network 6. The lower number of BMPR2 mutation carriers in the cohort from the New York Presbyterian Pulmonary Hypertension Center 5 could be explained by the fact that patients with a considered idiopathic PAH were not screened for large rearrangement of BMPR2. In the French PAH network, all patients with idiopathic or familial PAH were screened for point mutations and large rearrangements of BMPR2 gene, to avoid selection bias.

The familial clustering in the French PAH network may explain the differences observed between our data and those reported by Austin et al. Indeed, familial clustering could impact data in terms of genes and environmental interactions that may alter disease expression beyond that of a single gene mutation. In contrast, in the French PAH network, 81 out of the 115 BMPR2 mutation carriers (70.5%) were the only cases reported in their families, while 14 families had 2 cases reported and 2 families had 3 cases reported. Thus the majority of PAH cases we studied were the only reported cases from their families and familial clustering represented only a minority of reported cases.

Finally, PAH patients with a family history of PAH without identification of either BMPR2 or ACVRL1 mutation (n = 13) were excluded from the analysis in order to limit the risk of misclassification, although their clinical and hemodynamic characteristics were broadly similar to those of BMPR2 mutation carriers (age at PAH diagnosis 38.9 ± 12.5 yrs, mPAP 64 ± 13 mmHg, PVRi 25.9 ± 9.5 mmHg/L/min/m²). Moreover, ACVRL1 and Endoglin are two other genes well known to predispose to PAH. However, similar approach was not possible in ACVRL1 and Endoglin mutation carriers because the number of patients was too low (1 male and 8 female ACVRL1 mutation carriers, and only one patient with an Endoglin mutation).

In conclusion, our data indicate that mutation type has no influence on disease pattern or natural history among patients with BMPR2 mutations. However BMPR2 mutation carriers (males or females) are younger at diagnosis with a similar evolution of the disease than non-carriers, leading to a younger age at death of BMPR2 mutation carriers compared to non-carriers.

2-OHE: 2-hydroxyestrogens; 4-OHE: 4-hydroxyestrogens; 16α-OHE1: 16α-hydroxyestrone; BMPR2: Bone morphogenetic protein receptor type 2; CYP1B1: Cytochrome P450 1B1; NMD: nonsense-mediated decay; PAH: Pulmonary arterial hypertension; PTC: Premature termination codons.

The authors declare that they have no competing interests.

BG and DM performed the statistical analysis and wrote the manuscript. ME, AY, BS, FC, GS, FS and MH helped to conception, design and wrote the manuscript. All authors read and approved the final manuscript.