Identification of new inhibitors of protein kinase R guided by statistical modeling

We report the identification of new, structurally diverse inhibitors of interferon-induced, double-stranded RNA-activated protein kinase (PKR) using a combined experimental and computational approach. A training set with which to build a predictive statistical model was generated by screening a set of 80 known Ser/Thr kinase inhibitors against recombinant human PKR, resulting in the identification of 28 compounds from 18 chemical classes with <0.1 μM ⩽ IC50 ⩽ 20 μM. The model built with this data was used to screen a database of 5 million commercially available compounds in silico to identify candidate inhibitors. Testing of 128 structurally diverse candidates resulted in the confirmation of 20 new inhibitors from 11 chemical classes with 2 μM ⩽ IC50 ⩽ 20 μM. Testing of 34 analogs in the newly identified pyrimidin-2-amine active series provided initial SAR. One newly identified inhibitor, N-[2-(1H-indol-3-yl)ethyl]-4-(2-methyl-1H-indol-3-yl)pyrimidin-2-amine (compound 51), inhibited intracellular PKR activation in a dose-dependent manner in primary mouse macrophages without evident toxicity at effective concentrations.

Figure optionsDownload as PowerPoint slide