Author Manuscript Published OnlineFirst on March 13, 2018; DOI: 10.1158/0008-5472.CAN-17-2316 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Inhibin is a novel paracrine factor for tumor angiogenesis and metastasis Priyanka Singh1#, Laura M. Jenkins1#, Ben Horst1, Victoria Alers1, Shrikant Pradhan1, Prabhjot Kaur2, Tapasya Srivastava2, Nadine Hempel3, Balázs Győrffy4, Eugenia V. Broude5, Nam Y. Lee6, and Karthikeyan Mythreye5,1,* 1Department of Chemistry and Biochemistry, University of South Carolina, Columbia, South Carolina, USA 2Department of Genetics, University of Delhi, South Campus, India. 3Department of Pharmacology, Penn State University College of Medicine, Hershey PA 4MTA TTK Lendület Cancer Biomarker Research Group, Institute of Enzymology; Semmelweis University 2nd Department of Pediatrics, Budapest, Hungary. 5Department of Drug Discovery and Biomedical Sciences, School of Pharmacy, 6 Division of Pharmacology, College of Pharmacy, Ohio State University, Columbus, Ohio, USA # Equal contribution Running title: Inhibin promotes angiogenesis and metastasis * Correspondence to K. Mythreye: [email protected] Mailing address: Department of Chemistry and Biochemistry University of South Carolina, 631 Sumter Street Columbia SC 29208 Tel: (803) 576-5806, Fax: (803) 777-9521 Authors declare no conflict of interest. Keywords: Inhibin, Endoglin, ALK1, Angiogenesis, Ovarian cancer, Paracrine 1 Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 13, 2018; DOI: 10.1158/0008-5472.CAN-17-2316 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Abstract Inhibin is a heterodimeric TGF-β family ligand that is expressed in many cancers and is a selective biomarker for ovarian cancers, however its tumor-specific functions remain unknown. Here we demonstrate that the α subunit of Inhibin (INHA), which is critical for the functionality of dimeric Inhibin A/B, correlates with microvessel density (MVD) in human ovarian tissues and xenografts and is predictive of poor clinical outcomes in multiple cancers. We demonstrate that Inhibin regulated angiogenesis is necessary for metastasis. While Inhibin had no direct impact on tumor cell signaling, both tumor cell-derived and recombinant Inhibin elicit a strong paracrine response from endothelial cells by triggering SMAD1/5 activation and angiogenesis in vitro and in vivo. Inhibin-induced angiogenesis was abrogated via anti-Inhibin antibodies. The endothelial-specific TGF-β receptor complex comprising ALK1 and endoglin were crucial mediators of Inhibin signaling, offering a molecular mechanism for Inhibin-mediated angiogenesis. These results are the first to define a role for Inhibin in tumor metastasis and vascularization and offer an antibody-based approach for targeting Inhibin therapeutically Significance: Inhibin is a predictor of poor patient survival in multiple cancers and is a potential target for anti-angiogenic therapies. 2 Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 13, 2018; DOI: 10.1158/0008-5472.CAN-17-2316 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Introduction Inhibition of angiogenesis - the growth of new blood vessels from pre-existing vasculature is a clinically validated anti-cancer strategy for numerous tumor types. However, while the VEGF/VEGF receptor (VEGFR) signaling axis is widely recognized as the principal target of this therapeutic approach, current FDA-approved anti-VEGF drugs have demonstrated sub-optimal responses in the clinic. In many cases, including in ovarian cancers (OVCA), patient relapse, acquired resistance and cytotoxicity is commonly observed following anti-VEGF therapy. The identification of new vascular targets with potentially fewer adverse effects is therefore critical for improving therapeutic outcomes. TGF-β family members, particularly TGF-β1 and BMP9, are essential regulators of angiogenesis (1). Targeting these for anti-angiogenic therapy remains a formidable challenge due to their non-endothelial pleiotropic functions. Here, we focus on the unique TGF-β family member Inhibin, an endocrine hormone that sharply declines at the onset of menopause in healthy normal women and remains low (2) unlike other TGF-β family members and prototypical angiogenic factors like VEGF. Importantly, when Inhibin becomes elevated in postmenopausal women, this elevation becomes a diagnostic and prognostic marker for OVCA where along with CA125 detects 95% of ovarian tumors with 95% specificity (3,4). Inhibin is a heterodimeric member of the TGF-β family composed of an alpha (α) (coded by INHA) and beta (β) subunit (INHBA or INHBB). Combinations of these subunits give rise to either Inhibin A (αβA) or Inhibin B (αβB) (5). While Inhibin null mice (INHA-/-) are viable, they present with alterations to the ovarian vasculature and result in spontaneous gonadal tumors (6). In humans however, Inhibin levels are elevated in multiple cancer types including ovarian, prostate, adrenal, stomach and pancreatic cancers with indications for a role for Inhibin in prostate cancer metastasis (7-11). Despite these findings, the functional consequences of elevated tumor-derived Inhibin have yet to be determined. Several Inhibin binding proteins/receptors were previously reported (12). However unlike other TGF-β members, whose signal transduction mechanisms have been well studied, the mechanisms of Inhibin signaling remain largely unclear. The best-characterized Inhibin binding protein is the epithelial cell surface TGF-β co-receptor TβRIII/betaglycan (13). Inhibin binding to betaglycan fails to activate any discernable downstream pathways in epithelial cells. Others 3 Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 13, 2018; DOI: 10.1158/0008-5472.CAN-17-2316 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. and we previously demonstrated, several tumor suppressor functions for betaglycan, which is lost in the majority of human cancers (14), but little is known about the impact of elevated Inhibin on non-epithelial cells that do not express significant betaglycan. Given the urgent need to identify new anti-angiogenic pathways and targets to complement and improve existing therapies, we examined the potential role of Inhibin as a novel regulator of angiogenesis and metastasis. Importantly, we demonstrate Inhibin as an unexpected, clinically relevant, paracrine factor of tumor-induced angiogenesis and define the underlying mechanism of Inhibin action and therapeutic potential. Materials and Methods Cell Lines and Reagents: Ovarian epithelial carcinoma cell lines were obtained either from Duke Gynecology/Oncology Bank (Durham, NC) and ATCC. Authentication was carried out at the University of Colorado (Denver, CO) sequencing facility. HMEC-1 (human dermal microvascular endothelial cells) from ATCC CRL-3243 and MEECs (murine embryonic endothelial cells) ENG+/+ and ENG-/- were as described previously (15). HUVEC (human umbilical vein endothelial cells) was purchased from Lonza, USA. HMEC-1s were grown as per ATCC instructions. Epithelial carcinoma cell lines A2780, HEY, IGROV, OVCA247, M41, OVCA3, OVCA4, OVCA420, OVCA429, OVCA448, SKOV3 and PA1 were cultured in RPMI- 1640 (ATCC® 30-2001™) containing L-glutamine, 10% FBS and 100 U of penicillin- streptomycin. All cells lines were maintained at 37°C in a humidified incubator at 5% CO2, routinely checked for mycoplasma 3 times a year and experiments conducted within 3-6 passages depending on the cell line. Antibodies phospho-SMAD1/5 (#9516), phospho-SMAD2/3 (#8828S) and SMAD2/3 (#5678S) were from Cell Signaling Technology (Danvers, CA), SMAD1/5 (#ab75273) from Abcam, Cambridge, MA, USA. Mouse anti-HA antibody, Rabbit anti-HA antibody and mouse anti-Myc antibody were from invitrogen. Monoclonal antibodies to CD31 (#F8402) and Inhibin α (#ab47720) for IHC were purchased from Sigma-Aldrich and Abcam, respectively. Anti-INHA antibody (polyclonal #sc22048, Santa Cruz) and (monoclonal #sc365439, Santa Cruz) were used as indicated. ML347, Dorsomorphin and SB351432 were from Sigma-Aldrich, TRCN105 was a gift from TRACON pharmaceuticals (http://www.traconpharma.com/trc105.php). Inhibin A was from Sigma-Aldrich (# I9149) and R&D Systems (# 8506-AB). Lentiviral particles were generated at the COBRE Center for 4 Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 2018 American Association for Cancer Research. Author Manuscript Published OnlineFirst on March 13, 2018; DOI: 10.1158/0008-5472.CAN-17-2316 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Targeted Therapeutics Core Facility at SC. For INHA knockdown, SKOV3 cells were infected shRNA lentivirus, selected in 2 μg/ml Puromycin and stable cell lines maintained in 1 μg/ml Puromycin. Transient DNA transfections of HMEC-1 and COS7 were performed using either Targetfect (#HUVEC-01) from Targeting systems (El Cajon, CA) or Lipofectamine 2000 (#11668019) from Life Technologies (Carlsbad, CA). RNA isolation and Quantitative Polymerase Chain Reaction (qRT-PCR) analysis Total RNAs