Abstract A113: A fragment-based approach towards the identification of small molecule inhibitors of fascin 1

Fascin 1 is an actin-bundling protein that is dramatically overexpressed in a variety of invasive tumors and thought to have a critical role in cancer cell metastasis. However, as a drug target it is highly challenging due to its mechanism of protein-protein interaction and the lack of knowledge around the critical actin-binding sites. Using a fragment-based approach, biophysical assay screening and X-ray crystallography, we have been able to identify and optimize novel fascin 1 inhibitors. Furthermore we have developed robust and reproducible biochemical binding and bundling assays which have allowed us to develop ligands with submicromolar affinity. Fascin 1 cross-links filamentous actin (F-actin) into parallel bundles that are involved in the formation of dynamic cellular protrusions (such as lamellipodia and filopodia) used during cell migration. It also contributes to the formation of actin-rich finger-like protrusions, termed invadopodia, that tumor cell lines use to degrade the tumor extracellular matrix (ECM). Fascin 1 expression is low or absent in normal epithelia but is dramatically increased in a variety of tumor types including bladder, colon, lung, ovarian and pancreas. Its increased expression is associated with the most aggressive and metastatic tumor types and has been shown to be an independent prognostic indicator of poor clinical outcome. Fascin knockdown has been shown to reduce tumor cell invasion both in vitro and in vivo , suggesting fascin as a valid target for small molecule inhibitors. We have utilized a fragment-based screening approach incorporating surface plasmon resonance (SPR) to identify small molecules that bind to fascin 1. Our fragment library (∼1000 compounds) yielded 53 hit compounds that were confirmed by dose-response. Using these hits, X-ray co-crystallography identified four independent ligand binding sites within fascin. We have concentrated our efforts on one site, a deeply enclosed pocket between fascin domains 1 and 2. A number of fragment hits and commercial analogues were found to bind in this pocket and open a channel towards the surface of the protein. Subsequent virtual screening studies identified BDP-00010834, a compound which binds in this pocket with an SPR K d of 29.7μM. Moreover, BDP-00010834 shows activity (IC 50 = 50μM) in a functional biochemical screening assay that measures fascin-mediated F-actin bundling. Structure-based optimization together with X-ray co-crystallography has generated compounds with a 25- to 50-fold increase in both binding affinity and functional activity over this screening hit. Best-in-series compounds include BDP-00011163 (K d = 1.2μM, IC 50 = 0.8μM) and BDP-00011165 (K d = 0.6μM, IC 50 = 1.1μM). Citation Format: Daniel Croft, Stuart Francis, Alexander Schuettelkopf, Charles Parry, Gillian Goodwin, Angelo Pugliese, Laura McDonald, Maeve Clarke, Andrew Pannifer, Jen Lonsdale, Christopher Gray, Justin Bower, Heather McKinnon, Laura Machesky, Martin Drysdale. A fragment-based approach towards the identification of small molecule inhibitors of fascin 1. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr A113.