Large, drug-like lead compounds can require more time in the optimization phase to address problems inherent in the lead. Smaller leads have a higher chance of binding efficiently but can be harder to detect owing to lower affinity for the target. In recent years, several groups have demonstrated that evolution of low molecular weight, weakly-binding fragment hits is possible, representing a complementary approach to drug discovery that is especially attractive when co-crystal structure determination is used.An effort at Merck recently identified a series of potent and selective phosphodiesterase 10 (PDE10) inhibitors employing the fragment approach. The Merck Fragment Library (MFL) is a collection of 1600 soluble, low molecular weight compounds designed to be screened at high concentration for weak but efficient binders. Screening the MFL for PDE10 inhibitors identified a low molecular weight pyrimidine (1 ) with a Ki? of 8.7 �M. Initial optimization with similarity screening followed by parallel synthesis resulted in rapid potency improvements with minimal loss of binding efficiency.An early lead identified in this process is 2 , a 4.8 nM inhibitor of the PDE10 enzyme with ~100x selectivity over other phosphodiesterases and excellent pharmacokinetics. Extensive use of inhibitor-bound X-ray crystal structures of PDE10 and in silico modeling suggested further modifications to the 2-position of the pyrimidine, resulting in the preparation of 3 , which has a Ki? of 8.2 pM and selectivity of >5,000x over other PDEs.

Discovery and optimization of a series of pyrimidine PDE10 inhibitors through fragment screening, structure-based design, and parallel synthesis

247th National Meeting (2014)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.