Identifying potent Janus kinase 3 inhibitors using structure-guided virtual screening for inflammatory and neoplastic disease therapy

Janus kinases (JAKs) are potential therapeutic targets for anti-inflammatory and anti-cancer agents due to their involvement in cytokine signaling and cell proliferation. One of the major issues in the development of JAK inhibitors is the problem of selectivity for certain isoforms; since the isoforms are highly homologous, selective targeting is difficult. Of the JAKs, Janus kinase 3 (JAK3) which is mainly found in immune cells, is the most suitable isoform to target selectively to enhance the efficacy of treatment. In this study, we used a structure-based virtual screening method to screen PubChem for high-affinity JAK3 inhibitors using known JAK3-inhibitor complex structures. Through stringent filtering criteria, including structural similarity, physicochemical properties, and molecular interactions, we identified two promising compounds, CID:68715657 and CID:68585456, which showed potential JAK3 inhibition activity. These compounds showed better binding affinity than the parent molecules, and the structural modifications also improved the interaction with JAK3, indicating better potency and selectivity. Molecular dynamics (MD) simulations and MM-PBSA confirmed the stability of JAK3 complexes with CID:68715657 and CID:68585456, which further support their prospect as therapeutic targets of JAK3-related diseases. However, this study is limited by its reliance on computational predictions without experimental validation and the constraints of the PubChem database in capturing novel chemical scaffolds. Taken together, the results offer a sound basis for the further optimization of these compounds as highly effective and selective JAK3 inhibitors.