Design and synthesis of a potential selective JAK-3 inhibitor for the treatment of rheumatoid arthritis using predictive QSAR models

Abstract

Rheumatoid arthritis is a chronic autoimmune illness characterized by joint dysfunction. New drugs are available, among them Janus Kinase enzyme inhibitor drugs, whose significant side effects are associated with neutropenia, thrombocytopenia, and dyslipidemia due to nonselective inhibition of this family of enzymes. This research aim is to create a selective inhibitor for the Janus Kinase 3 enzyme (JAK3) using structure-based virtual screening (SBVS) and ligand-based virtual screening (LBVS) methodologies. As a result, we developed 127 predictive models for JAK1, JAK2, JAK3, and TYK2, yielding four predictive models with R2 better than 0.750 for each counterpart. Following the models development, employed 120 peficitinib analogues were rationally created. Next were predicted the IC50 values were predicted in the four models. The affinity was estimated using autodock, and the percentage of protein plasma binding as the partition coefficient was calculated employing ADMELAB. Finally, synthetic viability of the candidates was examined using Swissadme. Identifying ILER1 as the top candidate out of the 120 studied, it demonstrated high selectivity towards JAK3 over enzymes JAK1, JAK2, and TYK2 based on predicted IC50 values calculated by the neural network with an IC50 of 4,1 nM, which is 8,8 times smaller than the nearest value (JAK1 = 36,1 nM), as well as a safe, toxicity-free profile and good pharmacokinetic characteristics in comparison to commercial drugs. ILER1 (Ethyl 2-((7H-pyrrolo [2,3-d] pyrimidine-4-yl) amino)-3-methylpentanoate) was subsequently synthesized and characterized for future in vitro evaluation.

Keywords

• QSAR
• Virtual screening
• Organic synthesis
• Rational design