High-throughput virtual screening of phenylpyrimidine derivatives as selective JAK3 antagonists using computational methods

• Published in: Journal of biomolecular structure & dynamics Vol. 42; no. 14; pp. 7574 - 7599
• Main Authors: Faris, Abdelmoujoud, Ibrahim, Ibrahim M., Hadni, Hanine, Elhallaoui, Menana
• Format: Journal Article
• Language: English
• Published: England Taylor & Francis 21.09.2024
• Subjects: 3D-QSAR; ADME-Tox; atom-based model; Binding Sites; covalent docking; field-based model; High-Throughput Screening Assays - methods; Humans; Hydrogen Bonding; Hydrophobic and Hydrophilic Interactions; JAK3; Janus Kinase 3 - antagonists & inhibitors; Janus Kinase 3 - chemistry; Janus Kinase 3 - metabolism; Molecular Docking Simulation; Molecular Dynamics Simulation; pharmacophore modeling; phenyl pyrimidine derivatives; Protein Binding; Protein Kinase Inhibitors - chemistry; Protein Kinase Inhibitors - pharmacology; Pyrimidines - chemistry; Pyrimidines - pharmacology; Quantitative Structure-Activity Relationship; Rheumatoid arthritis; tyrosine kinase enzyme; ADME-Tox; phenyl pyrimidine derivatives; atom-based model; tyrosine kinase enzyme; Rheumatoid arthritis; molecular dynamics simulation; 3D-QSAR; covalent docking; pharmacophore modeling; field-based model; JAK3
• ISSN: 0739-1102; 1538-0254; 1538-0254
• DOI: 10.1080/07391102.2023.2240413

In this study, we used phenylpyrimidine derivatives with known biological activity against JAK3, a critical tyrosine kinase enzyme involved in signaling pathways, to find similar compounds as potential treatments for rheumatoid arthritis. These inhibitors inhibited JAK3 activity by forming a covalent bond with the Cys909 residue, which resulted in a strong inhibitory effect. Phenylpyrimidine is considered a promising therapeutic target. For pharmacophore modeling, 39 phenylpyrimidine derivatives with high pIC50 (Exp) values were chosen. The best pharmacophore model produced 28 molecules, and the five-point common pharmacophore hypothesis from P HASE (DHRRR_1) revealed the requirement for a hydrogen bond donor feature, a hydrophobic group feature, and three aromatic ring features for further design. The validation of the pharmacophore model phase was performed through 3D-QSAR using partial least squares (P LS). The 3D-QSAR study produced two successful models, an atom-based model (R2 = 0.95; Q2 = 0.67) and a field-based model (R2 = 0.93; Q2 = 0.76), which were used to predict the biological activity of new compounds. The pharmacophore model successfully distinguished between active and inactive medications, discovered potential JAK3 inhibitors, and demonstrated validity with a ROC of 0. 77. ADME-Tox was used to eliminate compounds that might have adverse effects. The best pharmacokinetics and affinity derivatives were selected for covalent docking. A molecular dynamics simulation of the selected molecules and the protein complex was performed to confirm the stability of the interaction with JAK3, whereas MM/GBSA simulations further confirmed their binding affinity. By using the principle of retrosynthesis, we were able to map out a pathway for synthesizing these potential drug candidates. This study has the potential to offer valuable and practical insights for optimizing novel derivatives of phenylpyrimidine. Communicated by Ramaswamy H. Sarma