Identification of Novel Anaplastic Lymphoma Kinase (ALK) Inhibitors Using a Common Feature Pharmacophore Model Derived from Known Ligands Crystallized with ALK

• Published in: Chemical biology & drug design 2013-02, Vol.81 (2), p.175-184
• Main Authors: Xie, Huan-Zhang, Lan, Hai, Pan, You-Li, Zou, Jun, Wang, Ze-Rong, Li, Lin-Li, Huang, Qi, Zhang, Hui, Yang, Sheng-Yong
• Format: Article
• Language: English
• Subjects: anaplastic lymphoma kinase; Antineoplastic Agents - chemistry; Antineoplastic Agents - pharmacology; Cell Line, Tumor; Cell Survival - drug effects; Drug Screening Assays, Antitumor; Humans; kinase inhibitor; Ligands; Molecular Docking Simulation; Mutation; pharmacophore model; Protein Kinase Inhibitors - chemistry; Protein Kinase Inhibitors - pharmacology; Receptor Protein-Tyrosine Kinases - antagonists & inhibitors; Receptor Protein-Tyrosine Kinases - chemistry; Receptor Protein-Tyrosine Kinases - genetics; Receptor Protein-Tyrosine Kinases - metabolism; Structure-Activity Relationship; virtual screening
• ISSN: 1747-0277; 1747-0285
• DOI: 10.1111/cbdd.12084

In this investigation, a common feature pharmacophore model of anaplastic lymphoma kinase inhibitors was developed based on several known anaplastic lymphoma kinase inhibitors that were co‐crystallized with anaplastic lymphoma kinase. The established pharmacophore model Hypo1 was carefully validated and then adopted to screen two in silico chemical databases, Specs (202 408 compounds) and Enamine (1 105 894 compounds), for retrieving novel anaplastic lymphoma kinase inhibitors. The hit compounds were further filtered using a fast bumping‐check tool and molecular docking. Finally, 25 compounds were selected and purchased from market. The bioactivity of these compounds was firstly measured at the cellular level against a typical anaplastic lymphoma kinase mutant‐driven cancer cell line, Karpas299. And six of them showed a good anti‐viability activity. The kinase inhibitory potency against the recombinant human anaplastic lymphoma kinase kinase was tested to the most active compound at the cellular level, T0508‐5181 (from Specs), which gave a half maximal inhibitory concentration (IC50) of 5.3 μm. In this study, the common features of three known ALK inhibitors that were crystallized with ALK have been identified for constructing a novel pharmacophore model. The established pharmacophore model Hypo1 had been carefully validated, and then compared with Hypo1′ that was built using a traditional ligand‐based pharmacophore modeling method, which results further demonstrated the correctness of Hypo1. This pharmacophore model was adopted to screen in silico chemical databases for novel ALK inhibitors.