Ligand-based and e-pharmacophore modeling, 3D-QSAR and hierarchical virtual screening to identify dual inhibitors of spleen tyrosine kinase (Syk) and janus kinase 3 (JAK3)

The clinical efficacy of multiple kinase inhibitors has caught the interest of Pharmaceutical and Biotech researchers to develop potential drugs with multi-kinase inhibitory activity for complex diseases. In the present work, we attempted to identify dual inhibitors of spleen tyrosine kinase (Syk) and janus kinase 3 (JAK3), keys players in immune signaling, by developing ideal pharmacophores integrating Ligand-based pharmacophore models (LBPMs) and Structure-based pharmacophore models (SBPMs), thereby projecting the optimum pharmacophoric required for inhibition of both the kinases. The four point LBPM; ADPR.14 suggested the presence of one hydrogen bond acceptor, one hydrogen bond donor, one positive ionizable, and one ring aromatic feature for Syk inhibitory activity and AADH.54 proposed the necessity of two hydrogen bond acceptor, one hydrogen bond donor, and one hydrophobic feature for JAK3 inhibitory activity. To our interest, SBPMs identified additional ring aromatic features required for inhibition of both the kinases. For Syk inhibitory activity, the hydrogen bond acceptor feature indicated by LBPM was devoid of forming hydrogen bonding interaction with the hinge region amino acid residue (Ala451). Thus merging the information revealed by both LBPMs and SBPMs, ideal pharmacophore models i.e. ADPRR.14 (Syk) and AADHR.54 (JAK3) were generated. These models after rigorous statistical validation were used for screening of Asinex database. The systematic virtual screening protocol, including pharmacophore and docking-based screening, ADME property, and MM-GBSA energy calculations, retrieved final 10 hits as dual inhibitors of Syk and JAK3. Final 10 hits thus obtained can aid in the development of potential therapeutic agents for autoimmune disorders. Also the top two hits were evaluated against both the enzymes.

多种激酶抑制剂的临床疗效已引发制药与生物技术领域研究者的广泛关注，推动其针对复杂疾病开发具备多激酶抑制活性的潜在药物。本研究通过整合基于配体的药效团模型（Ligand-based pharmacophore models, LBPMs）与基于结构的药效团模型（Structure-based pharmacophore models, SBPMs）构建理想药效团，旨在筛选出免疫信号通路关键靶点——脾酪氨酸激酶（Syk）与Janus激酶3（JAK3）的双靶点抑制剂。针对Syk的抑制活性，四点式LBPM模型ADPR.14提示需具备1个氢键受体、1个氢键供体、1个正离子化特征与1个芳香环特征；针对JAK3的抑制活性，AADH.54模型则提出需具备2个氢键受体、1个氢键供体与1个疏水特征。值得注意的是，SBPMs额外鉴定出了同时抑制两类激酶所需的另一项芳香环特征。针对Syk的抑制活性，LBPMs所提示的氢键受体特征无法与铰链区氨基酸残基Ala451形成氢键相互作用。因此，整合LBPMs与SBPMs所揭示的信息，我们构建了最优药效团模型：用于Syk的ADPRR.14与用于JAK3的AADHR.54。上述模型经严格的统计验证后，被用于Asinex数据库的虚拟筛选。通过包含药效团筛选、基于分子对接的筛选、ADME性质评估以及MM-GBSA能量计算在内的系统性虚拟筛选流程，最终获得10个可同时抑制Syk与JAK3的双靶点抑制剂命中化合物。所得的10个最终命中化合物可为自身免疫性疾病潜在治疗药物的开发提供助力。此外，研究人员还对排名前两位的命中化合物开展了针对两种酶的活性评价。