Abstract

The Janus kinases (JAKs) play a pivotal role in cytokine receptor signaling pathways via activation of downstream signal transducers and activators of transcription (STAT) pathway. Intracellular pathways that include JAKs are critical to immune cell activation and pro-inflammatory cytokine production. Selective inhibitors of JAKs are potentially disease-modifying anti-inflammatory drugs for the treatment of rheumatoid arthritis (RA). Each of the four members of the JAK family plays an individual role in the oncogenesis of the immune system, and therefore, the development of potent and specific inhibitors for each member is needed. Although there is a high sequence homology and structural identity of JAK1 and JAK2, such as a very similar binding mode of inhibitors at the ATPbinding site of enzymes, obvious differences surrounding the JAK1 and JAK2 ATP-binding sites provide a platform for the rational design of JAK2- and JAK1-specific inhibitors. In the present study, a dataset of 33 compounds characterized by a common scaffold of 2-amino-[1,2,4]triazolo[1,5-α]pyridine with well-defined in vitro activity values was computationally explored. Most of the compounds included in the dataset had higher ligand efficiency against JAK2 than JAK1. To improve further the selectivity of these triazolopyridines, Common Pharmacophore Hypotheses (CPHs) were generated and 3D-QSAR studies were carried out on them, in order to comprehend on the molecular features responsible for their selectivity. The proposed computational approach was applied in order to perform an in silico database virtual screening study with the aim to discover novel potent and selective JAK2 inhibitors.

Keywords: 3D-QSAR model, JAK1 kinase, JAK2 kinase, PHASE, Rheumatoid arthritis, triazolopyridines.