AT-9283

Janus kinase 2 (JAK2) plays a critical role in orchestrating hematopoiesis, and its deregulation leads to various blood disorders, most importantly myeloproliferative neoplasms (MPNs). Ruxolitinib, fedratinib, momelotinib, and pacritinib are FDA-/EMA-approved JAK inhibitors effective in relieving symptoms in MPN patients but show variable clinical profiles due to poor JAK selectivity. The development of next-generation JAK2 inhibitors is hampered by the lack of comparative functional analysis and knowledge of the molecular basis of their selectivity. Here, we provide mechanistic profiling of the four approved and six clinical-stage JAK2 inhibitors and connect selectivity data with high-resolution structural and thermodynamic analyses. All of the JAK inhibitors potently inhibited JAK2 activity. Inhibitors differed in their JAK isoform selectivity and potency for erythropoietin signaling, but their general cytokine inhibition signatures in blood cells were comparable. Structural data indicate that high potency and moderate JAK2 selectivity can be obtained by targeting the front pocket of the adenosine 5'-triphosphate-binding site.

Drug-target interactions (DTIs) prediction is an important step in drug discovery. As traditional biological experiments or high-throughput screening are high cost and time-consuming, many deep learning models have been developed. Overfitting must be avoided when training deep learning models. We propose a simple framework, called OverfitDTI, for DTI prediction. In OverfitDTI, a deep neural network (DNN) model is overfit to sufficiently learn the features of the chemical space of drugs and the biological space of targets. The weights of trained DNN model form an implicit representation of the nonlinear relationship between drugs and targets. Performance of OverfitDTI on three public datasets showed that the overfit DNN models fit the nonlinear relationship with high accuracy. We identified fifteen compounds that interacted with TEK, a receptor tyrosine kinase contributing to vascular homeostasis, and the predicted AT9283 and dorsomorphin were experimentally demonstrated as inhibitors of TEK in human umbilical vein endothelial cells (HUVECs).