Ligand-based computer aided drug design reveals new tropomycin receptor kinase a (TrkA) inhibitors

[Display omitted] •161 TrkA inhibitors were modeled by a combination pharmacophore and Genetic algorithm regression QSAR workflow.•The pharmacophore generation step was performed via sequential analysis of the 161 TrkA inhibitors.•The resulted pharmacophores were converted into numerical descriptors and been engaged in genetic algorithm and multiple linear regression QSAR workflow.•The optimal QSAR model was investigated and used to screen the NCI database for TrkA inhibitors.•21 hits out of 41 acquired, exhibited anti-TrkA activity with the most potent hit showing an IC50 value of 2.4μM against TrkA. Targeting tropomycin kinase A (TrkA) by small molecule inhibitors is considered as a promising strategy for treating several human cancers. To achieve this goal, a ligand based QSAR model was applied using the Discovery studio 4.5 (DS 4.5). Hence, a total list of 161 TrkA inhibitors was investigated. The TrkA inhibitors were extensively explored to detect their optimal physicochemical properties and pharmacophoric binding modes, which were converted into numeric descriptors and allowed to compete within the context of the Genetic Function Algorithm (GFA) approximations to find the subset of terms that correlates best with the activity. The resulted successful QSAR equation had statistical criteria of (r2129=0.67, r2LOO=0.61 r2PRESS against 32 external test inhibitors=0.50). Afterwards, the most successful pharmacophore: HypoB-T5-3, was used to screen compounds within the National Cancer institute (NCI) database. Only 41 compounds were retrieved and 21 of them exhibited anti-TrkA activity. The most potent hit had an IC50 value of 2.4μM. Later, upon docking the active hits into the TrkA binding pocket, important interactions were revealed including hydrogen bonding with the amino acids Asp668 and Lys544 in addition to the cation-π interactions with the sidechain of Arg559.