DataSheet1_Identification of Potent and Selective JAK1 Lead Compounds Through Ligand-Based Drug Design Approaches.pdf

JAK1 plays a significant role in the intracellular signaling by interacting with cytokine receptors in different types of cells and is linked to the pathogenesis of various cancers and in the pathology of the immune system. In this study, ligand-based pharmacophore modeling combined with virtual screening and molecular docking methods was incorporated to identify the potent and selective lead compounds for JAK1. Initially, the ligand-based pharmacophore models were generated using a set of 52 JAK1 inhibitors named C-2 methyl/hydroxyethyl imidazopyrrolopyridines derivatives. Twenty-seven pharmacophore models with five and six pharmacophore features were generated and validated using potency and selectivity validation methods. During potency validation, the Guner-Henry score was calculated to check the accuracy of the generated models, whereas in selectivity validation, the pharmacophore models that are capable of identifying selective JAK1 inhibitors were evaluated. Based on the validation results, the best pharmacophore models ADHRRR, DDHRRR, DDRRR, DPRRR, DHRRR, ADRRR, DDHRR, and ADPRR were selected and taken for virtual screening against the Maybridge, Asinex, Chemdiv, Enamine, Lifechemicals, and Zinc database to identify the new molecules with novel scaffold that can bind to JAK1. A total of 4,265 hits were identified from screening and checked for acceptable drug-like properties. A total of 2,856 hits were selected after ADME predictions and taken for Glide molecular docking to assess the accurate binding modes of the lead candidates. Ninety molecules were shortlisted based on binding energy and H-bond interactions with the important residues of JAK1. The docking results were authenticated by calculating binding free energy for protein–ligand complexes using the MM-GBSA calculation and induced fit docking methods. Subsequently, the cross-docking approach was carried out to recognize the selective JAK1 lead compounds. Finally, top five lead compounds that were potent and selective against JAK1 were selected and validated using molecular dynamics simulation. Besides, the density functional theory study was also carried out for the selected leads. Through various computational studies, we observed good potency and selectivity of these lead compounds when compared with the drug ruxolitinib. Compounds such as T5923555 and T5923531 were found to be the best and can be further validated using in vitro and in vivo methods.

JAK1（Janus激酶1）可通过与不同类型细胞内的细胞因子受体相互作用，在细胞内信号转导中发挥重要作用，且与多种癌症的发病机制及免疫系统病理过程密切相关。本研究采用基于配体的药效团建模结合虚拟筛选与分子对接方法，旨在筛选出针对JAK1的强效且选择性优异的先导化合物。首先，以52种名为C-2甲基/羟乙基咪唑并吡咯并吡啶衍生物的JAK1抑制剂为数据集，构建基于配体的药效团模型。共生成27个包含5个和6个药效团特征的药效团模型，并通过活性验证与选择性验证方法对其进行评估。在活性验证环节，通过计算古纳-亨利（Guner-Henry）得分以评估所构建模型的准确性；而在选择性验证环节，则对能够识别选择性JAK1抑制剂的药效团模型进行评估。基于验证结果，筛选出最优的8个药效团模型：ADHRRR、DDHRRR、DDRRR、DPRRR、DHRRR、ADRRR、DDHRR及ADPRR，并将其用于针对Maybridge、Asinex、Chemdiv、Enamine、Lifechemicals及Zinc数据库的虚拟筛选，以挖掘能够与JAK1结合的新型骨架分子。本次筛选共得到4265个命中化合物，并对其类药性质进行了合规性校验。经ADME（吸收-分布-代谢-排泄）性质预测后，筛选出2856个合格命中化合物，并采用Glide分子对接对候选先导化合物的精准结合模式进行评估。基于结合能及与JAK1关键残基的氢键相互作用，最终筛选出90个候选化合物。通过MM-GBSA（分子力学-广义玻恩表面积）计算及诱导契合对接方法，对蛋白-配体复合物的结合自由能进行测算，以验证对接结果的可靠性。随后采用交叉对接方法，进一步筛选出针对JAK1的选择性先导化合物。最终筛选出5个对JAK1兼具强效与选择性的最优先导化合物，并通过分子动力学模拟对其进行验证。此外，还针对筛选出的先导化合物开展了密度泛函理论研究。通过一系列计算研究发现，相较于上市药物鲁索替尼（ruxolitinib），这些先导化合物展现出更优异的强效性与选择性。其中T5923555与T5923531表现最为突出，可通过体外及体内实验开展进一步验证。