UCSF

Protein kinases are an important class of enzymes that are ubiquitously involved in cellular signal transduction. The misregulation of protein kinases is implicated in many diseases, motivating the development of selective inhibitors for use as drugs or chemical probes. However, a large fraction of the roughly 500 human kinases remains untargeted by small molecule inhibitors. Fragment-based ligand design and covalent targeting of noncatalytic cysteines have been successfully employed to develop potent and selective kinase inhibitors. Here, we combine these approaches, starting with a panel of low-molecular weight, heteroaryl-susbstituted cyanoacrylamides, which we have previously shown to form reversible covalent bonds with cysteine thiols at physiological pH. Using this strategy, we identified electrophilic fragments with sufficient ligand efficiency and selectivity to serve as starting points for the first reported inhibitors of the MSK1 C-terminal kinase domain.

While these first generation cyanoacrylamides served as novel chemotypes for inhibiting a previously untargeted protein, they suffered from thermal and metabolic instability. Exploration of chemical space around the amide electron withdrawing substituent enabled the thermal and metabolic stabilization of these electrophiles. The most successful electrophiles retained a nitrile as one electron withdrawing group, while an electron-deficient triazole played the role of the second electron-withdrawing substituent. Further optimization of both the electrophile and the noncovalent scaffold enabled the development of RMM-64 (29), a low-nM inhibitor of RSK and MSK kinases with prolonged target occupancy, high aqueous solubility, and oral bioavailability.

Finally, we adapted the computational digital screening program DOCK to enable the large scale, rapid virtual screening and ranking of cyanoacrylamide fragments against RSK and MSK kinases. With the help of collaborators, we constructed a library of more than 12,000 cyanoacrylamide-containing fragments based on commercially available aldehydes. Computational screening of this library against RSK2 and the RSK2 T493M mutant, with the constraint of a covalent bond to C436, enabled the identification of novel, potent and cell-active inhibitors of RSK2 and MSK1. We anticipate that the cyanoacrylamide fragment screening method described in this thesis, both experimental and computational, will be broadly applicable to any protein target containing a cysteine near a suitable binding site.