In vitro and in vivo characterization of wild type BMP9 and a non-osteogenic variant in models of pulmonary arterial hypertension

Abstract

Endothelial dysfunction and the resulting vascular remodeling are hallmarks of pulmonary hypertension, a debilitating disease of high arterial pressure in the lungs and the right side of the heart. Mutations in the BMPR2 signaling pathway are associated with the development of pulmonary arterial hypertension. Previous pre-clinical studies demonstrated that exogenous administration of recombinant human wild type BMP9 (WT BMP9) enhances BMPR2/ALK1 mediated signaling and reverses experimental pulmonary hypertension in rat models. However, BMP9 induces osteogenic activity in progenitor cells through activation of ActR2A and ActR2B receptor complexes potentially leading to unwanted bone formation in non-osteogenic tissues. The cellular activity of human WT BMP9 and amino acid substitution variants was characterized in vitro in terms of BMPR2 and ActR2 signaling. We identified a mutant variant of human BMP9 that maintains its activity in endothelial cells, specifically preserving BMPR2 signaling while eliminating osteogenic signaling associated with ActR2A/B activation in mesenchymal precursor cells. Rat models of pulmonary hypertension served as in vivo models to characterize efficacy and safety of BMP9 supplementation therapy. While WT BMP9 effectively activates BMPR2 signaling across species in rat, cynomolgus monkey and human systems, our human BMP9 mutant variant is inactive on rat BMPR2/ALK1 receptor complexes. Therefore, WT BMP9 was used to examine disease reversal in the preclinical monocrotaline model rat of pulmonary hypertension. WT BMP9 failed to improve right ventricular systolic pressure or right ventricular hypertrophy, despite clear target engagement shown by upregulation of SMAD7. Telemetry studies of WT BMP9 in the Sugen 5416 and hypoxia rat model of pulmonary hypertension indicated no significant change in pulmonary pressure but led to increased systemic blood pressure and decreased heart rate. Additionally, escalating doses in naive rats caused severe dose-limiting effects and morbidity at 500 µg/kg/day or higher. Given these findings including the absence of therapeutic efficacy in a relevant PAH animal model and dose limiting toxicity in rats, a therapeutic window for BMP9 treatment could not be established.

Fig 1. Pharmacological characterization of WT BMP9 and A347E variant in vitro and in vivo.

(A) WT BMP9 and A347E variant in cell-based human MSC assay measuring SP7 expression as a marker for osteogenic differentiation in a dose-response experiment (0.0195nM-10nM). (B) pSMAD1 activation in human endothelial cells in a dose-response experiment (0.000169nM-10nM). (C-E) Human primary pulmonary artery endothelial cells were treated with WT BMP9 and A347E variant at 10nM for 24h and gene expression was measured using Taqman PCR (C) ID2, (D) TGFBI, (E) PAI1. (F) pSMAD1 activation in primary rat endothelial cells in a dose-response experiment (0.000169nM-10nM). (G) WT BMP9 and A347E variant (30 µg/kg) were dosed SC to naïve rats (n = 4-5). Lung tissue was harvested 6h post dose and SMAD7 expression was evaluated in lung tissue as a PD marker by Taqman PCR. (H) WT BMP9 (30 µg/kg; IV) and A347E variant (1, 10, 30, 100 µg/kg; SC) was dosed to cynos (n = 3). Lung tissue was harvested 6h post dose and SMAD7 expression was evaluated in lung tissue as a PD marker by Taqman PCR.

Fig 2. In vivo PK evaluation of WT BMP9 in naïve SD rats.

(A) WT BMP9 (100 µg/kg; IV) plasma PK evaluation in naïve rats (n = 4). (B) Rat lung and kidney tissue PK (n = 4) at the indicated dose levels via IV administration 1h after dose. (C) WT BMP9 (100 µg/kg) rat plasma PK after SC administration (20 min: n = 1; 60 min: n = 4). (D) Rat lung tissue PK time course of WT BMP9 (100 µg/kg) at the indicated timepoints after a single SC administration (n = 4). (E) Rat lung tissue PK (n = 4) at the indicated dose levels after a single SC administration or daily injections over a 5-day period (5x30µg/kg) 3h after the last dose. (F) Gene expression analysis by Taqman PCR of lung tissue from PK time course study (100 µg/kg; n = 4). (G) Gene expression analysis by Taqman PCR of lung tissue from PK dose response study with SC administration 3h after dose (n = 4).

Fig 3. Evaluation of chronic in vivo efficacy in 21 days rat MCT model in prevention mode.

Rats were treated with MCT (60 mg/kg; n = 15; SC), or vehicle (control; n = 8; SC) followed by daily dosing of vehicle, WT BMP9 (3-100 µg/kg; n = 12; SC) or Imatinib (100 mg/kg; n = 12; PO) for 21 days. WT BMP9 was purchased from R&D systems or manufactured in-house in E. coli. (A) Fulton index (RV/LV + S) a measure of RVH 21 days post MCT injection. Comparison vs. control: ### p < 0.0001; vs MCT + Vehicle: * p = 0.0315; *** p < 0.0001. (B) Invasive measurements of RVSP 21 days post MCT injection. Comparison vs. control: ### p < 0.0001; vs MCT + Vehicle: ** p = 0.0017. (C) Fold change vs MCT vehicle in expression of various BMPR2 target genes by Taqman PCR from lung tissue of vehicle and WT BMP9 treated groups (Error represented as standard error of the mean).

Fig 4. Hemodynamic effects of WT BMP9 in rats.

(A) Continuous hemodynamic measurement of SBP over 48h after a single SC dose of WT BMP9 (3-100 µg/kg) in Su/Hx rats (n = 6). (B) HR changes quantified at 6h timepoint post dose in Su/Hx rats (n = 6). (C) SBP changes after repeat dose experiments with 5 daily doses of WT BMP9 (100 µg/kg) in Su/Hx rats (n = 6). (D) HR measurements over 48h after a single SC dose of WT BMP9 (3-100 µg/kg) in Su/Hx rats (n = 6). (E) SBP changes quantified at 6h timepoint post-dose in Su/Hx rats (n = 6). (F) HR changes after repeat dose experiments with 5 daily doses of WT BMP9 (100 µg/kg) in Su/Hx rats (n = 6). (G) Daily body weight changes during 5-day repeat dose study with a single daily dose of WT BMP9 (30-500 µg/kg) via SC or IV routes in naïve rats (n = 5). Two-way ANOVA test with multiple comparison. Comparison vs. day 1: * p = 0.0363. (H) Gene expression analysis by Taqman PCR from lung tissue at day 5 post dose in naïve rats (n = 5).