Modeling of Biomolecular Systems Interactions, Dynamics, and Allostery: Bridging Experiments and Computations - September 10-14, 2014, Istanbul, Turkey

Modeling of Biomolecular Systems Interactions, Dynamics, and Allostery Poster Session I

27-POS Board 27 Discovery of a Cryptic Druggable Pocket in Human p53 Ozlem Demir 1 , Linda V. Hall 3,4 , Faezeh Salehi 2,3 , Scott Rychnovsky 7 , Peter Kaiser 3,4,5 , Richard H. Lathrop 2,3,5 , Vicki Feher 1,10 , Rommie E. Amaro 1,8,9 . 1 University of California, San Diego, La Jolla, CA, USA, 2 University of California, Irvine, Irvine, CA, USA, 3 University of California, Irvine, Irvine, CA, USA, 4 University of California, Irvine, Irvine, CA, USA, 5 University of California, Irvine, Irvine, CA, USA, 6 University of California, Irvine, Irvine, CA, USA, 7 University of California, Irvine, Irvine, CA, USA, 8 University of California, San Diego, La Jolla, CA, USA, 9 University of California, San Diego, La Jolla, CA, USA, 10 University of California, San Diego, La Jolla, CA, USA. The tumor suppressor p53 has a major role in the defense of a cell against cancer. Cancer can only proceed after p53 or its pathways are inactivated by mutations. Thus, reactivation of mutant p53 with small-molecules have been a long-standing idea for potential cancer treatment. We explored the dynamic ensemble of many p53 mutants as well as the wild-type protein using molecular dynamics (MD) simulations using NAMD suite. In the MD-generated ensembles of p53 mutants, we have identified a transiently open binding pocket occluded in the available crystal structures. Virtual screening against different conformations of this cryptic pocket identified 45 promising compounds among which stictic acid emerged as a potential p53 reactivation compound. Stictic acid demonstrated dose-dependent p21 activation in human osteosarcoma cells with R175H mutant of p53. Encouraged by this result, we have performed virtual screening of 1.7 million compounds against a single pocket-open conformation of p53. Among the top 1% the compounds with the highest binding scores, we used machine learning to select a set of 138 diverse compounds out of which 15 compounds (~11% hit rate) were found to be potential p53 reactivation compounds by biological assays. Our findings highlight this cryptic druggable pocket as a promising pharmaceutical target for p53 reactivation.

80