Selective Inhibition of ADAMMetalloproteases as a Novel Approach for Modulating ErbBPathways in Cancer
J. Fridman,E. Caulder,M. Hansbury,Xiangdong Liu,Genjie Yang,Qian Wang,Y. Lo,Bin-Bing S. Zhou,Max Pan,S. Thomas,J. Grandis,J. Zhuo,W. Yao,R. Newton,S. Friedman,P. Scherle
2007-01-01
Abstract:Purpose: ErbB receptor signaling pathways are important regulators of cell fate, and their dysregulation, through (epi)genetic alterations, plays an etiologic role in multiple cancers. ErbB ligands are synthesized as membrane-bound precursors that are cleaved by members of the ADAM family of zinc-dependent metalloproteases. This processing, termed ectodomain shedding, is essential for the functional activation of ErbB ligands. Recent studies suggest that elevated levels of ErbB ligands may circumvent the effectiveness of ErbB-targeted therapeutics. Here, we describe the discovery and preclinical development of potent, selective inhibitors of ErbB ligand shedding. Experimental Design: A series of biochemical and cell-based assays were established to identify selective inhibitors of ErbB ligand shedding. The therapeutic potential of these compounds was assessed in multiple in vivo models of cancer and matrix metalloprotease ^ related toxicity. Results: INCB3619 was identified as a representative selective, potent, orally bioavailable small-molecule inhibitor of a subset of ADAM proteases that block shedding of ErbB ligands. Administration of INCB3619 to tumor-bearing mice reduced ErbB ligand shedding in vivo and inhibited ErbB pathway signaling (e.g., phosphorylation of Akt), tumor cell proliferation, and survival. Further, INCB3619 synergized with clinically relevant cancer therapeutics and showed no overt or compounding toxicities, including fibroplasia, the dose-limiting toxicity associated with broad-spectrummatrix metalloprotease inhibitors. Conclusions: Inhibition of ErbB ligand shedding offers a potentially novel and well-tolerated therapeutic strategy for the treatment of human cancers and is currently being evaluated in the clinic. The ErbB family is composed of four members, epidermal growth factor (EGF) receptor (EGFR; HER1 or ErbB1), HER2 (neu or ErbB2), HER3 (ErbB3), and HER4 (ErbB4). Eleven ligands have been reported to bind to EGFR, HER3, and HER4 and show some level of receptor selectivity (1). HER2 has no confirmed extracellular ligand and serves as the preferred dimerization partner for other ErbB receptors (2). Ligand binding results in receptor dimerization, kinase domain activation (except for HER3 which lacks a functional kinase active site), receptor phosphorylation, and the activation of multiple downstream signaling cascades with pleiotropic oncogenic effects when dysregulated—a common event in cancer (3–5). Accordingly, the ErbB family represents a validated therapeutic target for the treatment of cancer and has resulted in the development of multiple therapeutics, including the antibodies trastuzumab (directed against HER2) and cetuximab (directed against EGFR) as well as low molecular weight, ATP-competitive tyrosine kinase inhibitors (TKI) targeting EGFR—gefitinib and erlotinib (reviewed in ref. 5). Although such targeted therapeutics have showndramatic antitumor effects in a subset of patients, many do not respond and most eventually become resistant. Despite the fact that certain tumor traits have been identified that may be associated with clinical responses to EGFR TKIs (e.g., EGFR gene amplification or a subset of somatic mutations in exons 18 to 21, or mesenchymal characteristics), they can account for only a fraction of those patients achieving clinical benefit from treatment. Substantial preclinical and clinical evidence exist, suggesting that the different ErbB pathways may compensate for one another, thus providing a mechanism of resistance against an agent targeting an individual pathway. For Cancer Therapy: Preclinical Authors’Affiliations: Drug Discovery, Incyte Corporation,Wilmington, Delaware and Department of Otolaryngology, University of Pittsburgh Medical Center and University of Pittsburgh Cancer Institute, Pittsburgh, Pennsylvania Received 8/24/06; revised11/17/06; accepted11/30/06. The costs of publication of this article were defrayed in part by the payment of page charges.This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section1734 solely to indicate this fact. Conflict of interest statement. The authors (excluding S.M. Thomas and J.R. Grandis) declare they all currently are, or have been, employed by Incyte Corporation. Requests for reprints: Jordan S. Fridman, Drug Discovery, Experimental Station, Incyte Corporation, E400, Route 141and Henry Clay Road,Wilmington, DE19880. Phone: 302-498-6930; Fax: 302-425-2760; E-mail: jfridman@incyte.com. F2007 American Association for Cancer Research. doi:10.1158/1078-0432.CCR-06-2116 www.aacrjournals.org Clin Cancer Res 2007;13(6) March15, 2007 1892 Research. on May 1, 2017. © 2007 American Association for Cancer clincancerres.aacrjournals.org Downloaded from instance, increased levels of the EGFR ligand transforming growth factor-a (TGF-a) have been observed in trastuzumabtreated breast cancer patients with progressive disease (6). Similar results were obtained in a preclinical model of trastuzumab resistance that showed a concordant increased sensitivity to an EGFR-selective TKI (7). Moreover, a nonbiased analysis of patients receiving gefitinib described significantly elevated expression of the ErbB ligands amphiregulin and TGFa in nonresponders, agreeing with in vitro data suggesting that high ErbB ligand concentrations can circumvent the effectiveness of an EGFR-targeted agent (8, 9) or the less selective ErbB kinase inhibitor GW572016 (10). These data are consistent with the preclinical observation that targeting the EGFR pathway at multiple points (ligand binding and ATP pocket) may be more efficacious than either method in isolation (11, 12)—a hypothesis currently being evaluated clinically. Finally, even in patients who do respond to EGFR-selective TKIs, the emergence of secondary mutations in the EGFR kinase domain affords resistance to the ATP-competitive TKIs (13–16). Although this is an evolving field, these and other data support the following concepts: (a) although ErbBtargeted agents have shown clinical promise, there is significant room for improvement; (b) a multifaceted approach to inhibiting multiple ErbB pathways would be expected to improve therapeutic response; and (c) inhibiting the activation of multiple ErbB ligands represents a logical and perhaps clinically relevant point of intervention. Although certain broad spectrum matrix metalloprotease (MMP) inhibitors (e.g., marimastat) can inhibit EGFR ligand shedding in vitro (among other activities), this has not been investigated in vivo nor has the therapeutic hypothesis been tested clinically due to the lesser potency of these drugs against key ADAM protease—relative to MMPs (see below)—and their inability to be tolerated at higher doses (17). Here, we describe the discovery and preclinical development of selective inhibitors of ErbB ligand shedding and show that these compounds inhibit ErbB ligand shedding and tumor growth in vivo without inducing joint fibroplasias, the dose-limiting toxicity of many MMP inhibitors. These compounds possess the pharmacologic properties making them ideal candidates to test the therapeutic potential of sheddase inhibition for the treatment of cancer in a clinical setting. Materials andMethods In vitro metalloprotease and pharmacology assays. Sheddase inhibitors were synthesized at Incyte Corp as described (18). Human MMPs, ADAM9, and ADAM10 were purchased from R&D Systems (Minneapolis, MN) and assayed as recommended by the manufacturer. The peptide substrate (7-methoxycoumarin-4-yl) acetyl-Pro-Leu-Gly-Leu-(3[2,4-dinitrophenyl]-L-2, 3-diaminopropionyl)-Ala-Arg-NH2 was used for the MMP assays as recommended by the manufacturer (R&D Systems). Partially purified porcine ADAM17 was obtained from spleen as described (19). The fluorogenic peptide substrate, (7-methoxycourmarin-4-yl)-acetyl-Pro-Leu-Ala-Gln-Ala-Val-(3-[2,4-dinitrophenyl]-L-2, 3-diaminopropionyl)-Arg-Ser-Ser-Ser-Arg-NH2, was used for the ADAM9, ADAM10, and ADAM17 assays following the manufacturer’s recommendations (R&D Systems). INCB3619 was also screened against a panel of G-protein coupled receptors and ion channels for inhibitory activity (ExpresSProfile, Cerep, Inc., Paris, France). Cell culture and generation of stable cell lines. Cell lines were obtained from American Type Culture Collection (Manassas, VA) and were cultured as recommended by the supplier. Retroviral transduction was used to generate derivatives of the NMuMG and Rat-1 cells as described previously (20). The coding region of the human cDNA for TGF-a was cloned into pMSCVpuro (BD Biosciences, Mountain View, CA) following amplification from a pool of human cDNA (BD QUICKClone). Human umbilical vascular endothelial cells were obtained from Cambrex Corp (East Rutherford, NJ) and treated following the manufacturer’s protocols. Analysis of ErbB ligand shedding and its effect. The TGF-a cleavage assay used SCC9 cells stimulated with 10 Amol/L phorbol 12-myristate 13-acetate (Sigma-Aldrich, St. Louis, MO) for 30 min in serumfree medium following a 30-min pretreatment with DMSO or metalloprotease inhibitor. Supernatants were collected and assayed for soluble TGF-a by a commercially available ELISA kit (R&D Systems). Alternatively, NMuMG or Rat-1 cells were grown for 24 h in the presence of DMSO or sheddase inhibitor, and supernatants were analyzed in a similar manner. Amphiregulin release was measured in supernatants of HCT116 stimulated with phorbol 12-myristate 13-acetate (1 Amol/L, 30 min) using an ELISA kit (R&D Systems). The heparin-binding EGF-like growth factor (HB-EGF) cleavage assay used Chinese hamster ovary cells stably transfected with human placental alkaline phosphatase– tagged HB-EGF (In Vitro Biology Department of Incyte Corporation, Wilmington, DE). The transfected Chinese hamster ovary cells were grown in 96-well plates and treated with DMSO or compound for 10 min followed by phorbol 12-myristate 13-acetate stimulation (2 Amol/L, 45 min). Al