BMP9 activates the Smad1,5,8 pathway in IOSE397 cells. A-B, IOSE397 cells were incubated for 1 hour with increasing concentrations of BMP9 (A) or for the indicated periods of time −/+ BMP9 (5ng/ml) in 0.1% FBS media (B). Western blots were performed with the indicated antibodies. C, IOSE397 cells were incubated −/+ BMP9 (5ng/ml) for different periods of time in 0.1% FBS media and Id1 levels were analysed by qRT-PCR and normalized to β-actin. Fold changes relative to untreated samples were determined (mean ± S.E.M, n=3). D, Cells were treated as in C and analysed by western blotting with indicated antibodies. Statistical analysis compared treated to untreated samples. * = P< 0.05.

BMP9 binds and signals via ALK2 in IOSE397 cells. A, IOSE397 cells were transiently transfected with siRNA oligonucleotides as indicated and with pGL3(BRE)-luciferase reporter gene and EF-LacZ and treated −/+ BMP9 (5ng/ml) for 15 hours. Normalised luciferase activity is shown as fold induction relative to untreated samples (mean ± S.E.M, n=3). B, IOSE397 cells were transiently transfected with siRNA oligonucleotides as indicated and treated for 1 hour −/+ BMP9 (5ng/ml) and western blots were performed with indicated antibodies. C, IOSE397 and Huvec cells were affinity labelled with ¹²⁵[I]BMP9 and crosslinked ligand receptor complexes were immunoprecipitated (I.P.) with specific antisera as indicated and subjected to SDS-PAGE and autoradiography. D, IOSE397 cells were transiently transfected with siRNA oligonucleotides as indicated and treated as in A. Normalised luciferase activity is shown as fold activations relative to untreated samples (mean ± S.E.M., n=4). Statistical analysis compared siRNA treated to non silencing (N.S.) samples throughout. * = P< 0.05.

BMP9 promotes proliferation of IOSE397 cells and ovarian cancer cell lines via an ALK2/ActRIIA/BMPRII/Smad1/Smad4 pathway. A, Proliferation curves of IOSE397, TR175 and OVCA433 cells incubated for different periods of time −/+ BMP9 (5ng/ml) in 0.1% FBS media (mean ± S.E.M., n=3). B left panel, OVCA433 cells were transiently transfected with siRNA oligonucleotides as indicated, treated −/+ BMP9 (5ng/ml) in 0.1% FBS for 4 days and counted (mean ± S.D, n=3). B right panel, OVCA433 cells were treated −/+ BMP9 (5ng/ml) and −/+ dorsomorphin (Dm, 1μM) in 0.1% FBS media for 4 days and counted. (mean ± S.E.M, n=5). C, OVCA433 cells were transiently transfected with siRNA oligonucleotides as indicated, treated −/+ BMP9 (5ng/ml) in 0.1% FBS for 6 days and counted (mean ± S.D, n=2). D, OVCA433 cells were transiently transfected with siRNA oligonucleotides as indicated and treated as in B. (mean ± S.D, n=3). B ,C, D, Data are displayed as fold increase relative to untreated samples. * = P< 0.05, ** = P < 0.005.

Serum derived BMP9 promotes proliferation of IOSE397 cells and ovarian cancer cell lines. A, IOSE397 cells were incubated for 1 hour with increasing concentrations of ALK1ecd −/+ BMP9 (5ng/ml) in 5% FBS media. Western blots were performed with indicated antibodies. B, cells were incubated −/+ ALK1ecd (16 F.M.E.) in 5% FBS (IOSE397) or 10% FBS (TR175 and OVCA433) and counted at day 6. Data are displayed as percent of untreated cells (mean ± S.D, n≥3). C, IOSE397 cells were incubated for 1 hour with increasing concentrations dorsomorphin (Dm) −/+ BMP9 (5ng/ml) in 5% FBS media. Western blots were performed with indicated antibodies. D, Cells were incubated −/+ dorsomorphin (Dm, 1μM) in the same conditions as in C and counted at day 4. Data are displayed as percentage of untreated samples (mean ± S.E.M, n=3). * = P< 0.05.

Autocrine BMP9 promotes ovarian cancer cell lines proliferation. A left panel, BMP9 RNA levels of IOSE and EOC cell lines were analysed by qRT-PCR and normalized to β-actin. IOSE397 BMP9 RNA content was assigned an arbitrary value of 1 (mean ± S.E.M, n=3). A right panel, Bioassay for the analysis of BMP9 production in EOC and IOSE397 cell lines. Cells were serum starved (0.1% FBS media) and after 15 hours, counted and media collected and assayed for BMP9 content. Results are expressed as pg of BMP9 /100,000 cells/ hour (mean ± S.D., n=6). B left panel IOSE397, TR175 and OVCA433 cells were incubated with different concentrations of ALK1ecd (F.M.E.) in 0.1% FBS and counted at day 6. Data are displayed as percent of untreated cells (mean ± S.E.M, n=3). B right panel, IOSE397, TR175 and OVCA433 cells were transiently transfected with BMP9 siRNA oligonucleotides, serum starved and counted 6 days later. Data are displayed as percent of non silencing siRNA control transfected (N.S) cells, (mean ± S.D, n≥3). C, Independent stable cell lines expressing non-silencing (N.S) and two different shRNAs targeted against BMP9 were generated by retroviral infection of TR175 and OVCA433 cells. Cells were serum starved in 0.1% FBS, treated −/+ BMP9 (5ng/ml) and counted at day 6 (mean ± S.E.M, n=3). D, TR175 and OVCA433 cells were transiently transfected with ALK2 siRNA oligonucleotides as indicated, serum starved and counted 4 days later. Data are displayed as percent of non silencing control siRNA transfected (N.S) cells (mean ± S.D, n≥3)* = P< 0.05, ** = P < 0.005.

BMP9 is expressed in human epithelial ovarian cancer tissue and not in normal OSE cells. Sections of formalin fixed paraffin embedded human ovarian tissues were stained with BMP9 antibody and counterstained with haematoxylin. A, Representative images of normal human OSE cells and EOCs. Scale bars represent 500 μm. B, Summary of BMP9 IHC of a human ovarian cancer TMA. Staining was scored as positive or negative for BMP9 staining and percentage positive samples for each ovarian cancer subtype are shown.