Abstract IA29: Targeting undruggable proteins for degradation with small molecules

Abstract

Abstract Many well-studied oncoproteins are currently viewed to be “undruggable”. The infamous teratogen thalidomide was serendipitously discovered to be a potent antimyeloma drug in the 1990s but its mechanism of action was unknown. In 2010 thalidomide was shown by Handa and coworkers to bind to cereblon, which is the substrate-recognition subunit of a cullin-dependent ubiquitin ligase. Moreover, thalidomide was found to inhibit this ligase in vitro, suggesting that thalidomide is a cereblon antagonist. However, it soon became apparent that loss of cereblon rendered myeloma cells resistant to thalidomide (and related drugs such as lenalidomide, collectively referred to as IMiDs). This suggested to us that cereblon, once bound to an IMiD, acquires a new (neomorphic) ability to target a myeloma-relevant protein for destruction. To look for such a protein we built a bicistronic expression vector that encodes both a protein open reading frame (ORF)-firefly luciferase fusion protein and, from the same mRNA, renilla luciferase for normalization purposes, with both proteins under the control of an internal ribosomal entry site (IRES). We then made a library of ~12,000 such ORF-luciferase fusions and introduced them into 293T cells in quadruplicate in multiwell plate format (1 ORF/well), treated the cells with lenalidomide or vehicle, and monitored changes in firefly luciferase/renilla luciferase as a surrogate for changes in ORF stability. In this way we discovered that cereblon, once bound to lenalidomide, destabilizes IKZF1 and IKZF3. We went on to show that downregulation of IKZF1 and IKZF3 is intimately linked to the antimyeloma activity of the IMiDs. Ben Ebert and his group working in parallel and using complimentary approaches reached the same conclusion. Published and unpublished work from investigators such as Jay Bradner and Craig Crews suggests IMiDs can be reengineered to serve as molecular matchmakers between cereblon and novel targets. We are currently developing new screening methodologies to identify chemicals and genetic perturbants that directly or indirectly destabilize proteins of interest. Citation Format: William G. Kaelin, Jr. Targeting undruggable proteins for degradation with small molecules. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Targeting the Vulnerabilities of Cancer; May 16-19, 2016; Miami, FL. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(1_Suppl):Abstract nr IA29.