Bone morphogenetic protein 10 is increased in pre-capillary pulmonary hypertension patients

Abstract

Aims: Pre-capillary pulmonary hypertension (precPH) results in increased right atrial (RA) stretch and pressure. The right atrium is the major source of bone morphogenetic protein 10 (BMP10) in adults, primarily produced by RA cardiomyocytes. The aim of this study was to investigate BMP10 expression in the right heart and systemic circulation and to identify potential triggers for increased BMP10 secretion associated with precPH.

Methods and results: We examined BMP10 mRNA and protein expressions in RA tissue. Circulating BMP10 plasma levels were determined using enzyme-linked immunosorbent assay. BMP10 transcriptional activity was studied using a BMP-responsive element luciferase assay. Correlation analyses were performed between circulating BMP10 and RA dilatation as well as right ventricular (RV) function. Finally, we determined the impact of pressure unloading on BMP10 activity in chronic thromboembolic pulmonary hypertension (CTEPH) patients before and after pulmonary endarterectomy (PEA). BMP10 mRNA's protein and activity were significantly increased in the precPH right atrium. While circulating BMP10 protein levels were elevated, no significant changes were observed in BMP10 transcriptional activity between precPH and controls. Interestingly, RA dilatation, increased RA pressure, high N-terminal pro b-type natriuretic peptide levels, and reduced RV ejection fraction were associated with high BMP10 activity. Finally, pressure unloading after PEA in a cohort of CTEPH patients resulted in reduced BMP10 activity.

Conclusion: RA BMP10 expression and plasma levels are increased in precPH, likely triggered by excessive RA dilatation and pressure overload. Future studies are needed to determine whether increased BMP10 release is an adaptive mechanism or a potential therapeutic target.

Keywords: BMP; BMP10; Pulmonary hypertension; Right atrium; TGF-beta.

Graphical Abstract

Increased BMP10 in precPH patients compared with controls. Higher BMP10 expression and activity in the right atrium and greater BMP10 expression, but preserved activity, in the circulation. Increased BMP10 activity associates with disease severity. Created with BioRender.com.

Figure 1

Study design. Schematic overview of study populations used for each analysis. We studied BMP10 gene expression in fresh RA tissue of five CTEPH from the main cohort undergoing PEA and nine control subjects undergoing coronary bypass or aortic valve replacement. Due to limited fresh RA tissue availability, we also collected post-mortem or transplanted paraffin-embedded RA tissue from PAH patients (n = 2 iPAH and 2 hPAH) and from six controls to study BMP10 protein expression and activity using histological analysis. Furthermore, we included 48 pre-capillary PH patients [n = 22 iPAH, n = 14 hPAH (n = 12 BMPR2 mutation carrier and n = 2 GDF2 mutation carrier), and n = 12 CTEPH] and 16 controls, from which we measured circulating BMP10 protein expression and transcriptional activity in peripheral blood samples. In a subset of patients, we also collected RA blood samples (n = 37 PH patients, n = 11 controls) to check BMP10 levels directly in the right atrium.

Figure 2

Up-regulated local BMP10 expression in the right atrium of precPH patients. (A) Quantification of the relative BMP10 mRNA expression in control (n = 9) and precPH (n = 5 CTEPH) RA tissues. (B and C) Quantification of total RA BMP10 fluorescent area and RA cardiomyocytes BMP10 intensity levels in control (n = 6) and precPH (n = 4 PAH) paraffin-embedded RA tissue sections stained against BMP10, respectively. Representative immunofluorescent stainings of BMP10, Ulex-rhodamine (Ulex, endothelium), and cardiac TroponinT (cTnT, myocardium) in the negative control sample for anti-rabbit Alexa488 and anti-mouse Alexa647 (D), in the control (E), and in the precPH (F) RA tissues at 60×-oil magnification. (D’–F’) Alexa488 single-channel images from the stainings in (D–F). (E′′ and F′′) Zoom-in images from (E′ and F′) to appreciate the sarcomeric pattern of the BMP10 staining in the cardiomyocytes and the homogeneous staining in the vessels. Scale bars = 50 μm. Brightness and contrast for the Alexa488 channel have not been modified. Normality of data was checked and transformed if needed, and statistical differences between precPH patients and controls were tested using an independent sample t-test.

Figure 3

Up-regulated local BMP10 activity in the right atrium of precPH patients. (A) Quantification of positive pSMAD1/5/8 nuclei in vascular and non-vascular cells within the RA tissues from precPH (n = 4 PAH) and controls (n = 5). (B and C) Representative immunofluorescent staining of positive pSMAD1/5/8 nuclei in vascular and non-vascular cells from control and precPH with rhodamine and Alexa488 single-channel images on the sides. (D) Quantification of positive ID3 nuclei in vascular and non-vascular cells within the RA tissues from precPH (n = 4 PAH) and controls (n = 5). (E and F) Representative immunofluorescent staining of positive ID3 nuclei in vascular and non-vascular cells from control and precPH with rhodamine and Alexa488 single-channel images on the sides. Arrowheads indicate positive pSMAD1/5/8 and ID3 nuclei. Zoom-in images are included within (B, C, E, and F). Nuclei were counterstained with Hoechst 33342 and vessels with Ulex-rhodamine (B, C, E, and F). Negative control images are shown in Supplementary material online, Figure S2A. Scale bars = 50 μm. Vascular and non-vascular measurements are plotted with their own Y-axis on the left or right side, respectively. Normality of data was checked and transformed if needed, and statistical differences between precPH patients and controls were tested using a Wilcoxon rank-sum test (in A and D).

Figure 4

Higher BMP10 plasma levels in precPH patients compared with controls. (A and B) BMP10 protein circulating plasma levels in precPH patients (n = 48) and subgroups (n = 48: 22 iPAH, 14 hPAH, and 12 CTEPH), respectively, vs. controls (n = 16). (C and D) BMP9 protein circulating plasma levels in precPH patients (n = 45) and subgroups (n = 45: 20 iPAH, 14 hPAH, and 11 CTEPH), respectively, vs. controls (n = 16). (E and F) Correlation between BMP10 and BMP9 plasma levels in precPH patients (n = 45) or subgroups (n = 45: 20 iPAH, 14 hPAH, and 11 CTEPH), respectively, vs. controls (n = 16). Logarithmic Y-axis is used in graphs (A–D). Data in (A and B) are y + 1 for logarithmic scale transformation. Normality of data was checked and transformed if needed. Statistical differences between precPH patients or precPH subgroups and controls were tested with an independent sample t-test or a one-way ANOVA, respectively. Associations were tested with univariate linear regression analysis.

Figure 5

BMP10 transcriptional activity in precPH patients and controls. (A) Schematic explanation of the BRE-LUC reporter assay to determine BMP transcriptional activity in venous serum. Specific trap antibodies targeting BMP9 or BMP9 and BMP10 are used to assess BMP10 activity. Created with BioRender.com. B) Relative BMP transcriptional activity as a luciferase read-out from the HMEC-BRE-LUC, endothelial cells expressing a BMP-specific luciferase reporter, in control (n = 15) and precPH subgroups (n = 21 iPAH, n = 13 hPAH, and n = 11 CTEPH) after incubation with phosphate-buffered saline (PBS) (baseline), anti-BMP9, or ALK1-Fc (inhibition of BMP9 and BMP10). (C) BMP10 activity in controls and precPH subgroups has been calculated from the subtraction of anti-BMP9 and ALK1-Fc to total BMP activity. Normality of data was checked and transformed if needed. Statistical differences between precPH patients and controls, and between baseline conditions and trap antibodies, were tested with an independent sample t-test or a one-way ANOVA, after which pairwise t-testing with Bonferroni correction was applied, respectively.

Figure 6

Patients with more right atrial dilatation, reduced RV ejection fraction, and higher NT-proBNP have higher levels of circulation BMP10 activity. (A and B) BMP10 transcriptional activity in precPH patients with RA or RV dilation, respectively. (C–E) BMP10 transcriptional activity in precPH patients with high RAP, reduced RVEF, or high NT-proBNP, respectively. PrecPH patients were stratified according to RA volume (>79 mL/mm² for male patients or >69 mL/mm² for female patients), RV end-diastolic volume index (≥109 mL/mm² for males, and ≥97 mL/mm² for females), RAP (>14 mmHg), RVEF (<35%), and NT-proBNP levels (>1100 ng/L). Normality of data was checked and transformed if needed. Statistical differences between both groups were tested with an independent samples t-test.

Figure 7

Effect of pressure unloading on BMP10 activity in precPH patients. (A) Serum relative BMP activity at baseline and post-PEA in CTEPH patients. Incubation with anti-BMP9 only blocked BMP9 activity, while ALK1-Fc blocked both BMP9 and BMP10 activities. (B) Calculated BMP10 transcriptional activity at baseline and post-PEA (n = 13), respectively. BMP10 transcriptional activity is calculated by subtracting BMP activity values after incubation with the trap antibodies. Normality of data was checked and transformed if needed. Statistical differences between baseline conditions and trap antibodies, and between baseline and post-PEA, were tested with an independent sample t-test.