Successive Statistical and Structure-Based Modeling to Identify Chemically Novel Kinase Inhibitors

Abstract

Kinases are frequently studied in the context of anticancer drugs. Their involvement in cell responses, such as proliferation, differentiation, and apoptosis, makes them interesting subjects in multitarget drug design. In this study, a workflow is presented that models the bioactivity spectra for two panels of kinases: (1) inhibition of RET, BRAF, SRC, and S6K, while avoiding inhibition of MKNK1, TTK, ERK8, PDK1, and PAK3, and (2) inhibition of AURKA, PAK1, FGFR1, and LKB1, while avoiding inhibition of PAK3, TAK1, and PIK3CA. Both statistical and structure-based models were included, which were thoroughly benchmarked and optimized. A virtual screening was performed to test the workflow for one of the main targets, RET kinase. This resulted in 5 novel and chemically dissimilar RET inhibitors with remaining RET activity of <60% (at a concentration of 10 μM) and similarities with known RET inhibitors from 0.18 to 0.29 (Tanimoto, ECFP6). The four more potent inhibitors were assessed in a concentration range and proved to be modestly active with a pIC 50 value of 5.1 for the most active compound. The experimental validation of inhibitors for RET strongly indicates that the multitarget workflow is able to detect novel inhibitors for kinases, and hence, this workflow can potentially be applied in polypharmacology modeling. We conclude that this approach can identify new chemical matter for existing targets. Moreover, this workflow can easily be applied to other targets as well.

Fundamental considerations in drug design

“…Burggraaff et al have successfully carried out a statistical and structure-based virtual screening for the discovery of several RET kinase inhibitors. 128 …”

Large-scale comparison of machine learning methods for profiling prediction of kinase inhibitors

“…Generally, these computational methods can be classified into two major categories: structure- and ligand-based kinase inhibition and/or profiling prediction approaches (called virtual assay). Molecular docking, commonly used in structure-based prediction methods for kinase inhibition, has good generalizability, but its accuracy depends on the crystal structure of the kinase and the accuracy of the scoring function [ 13 , 14 ]. Ligand-based methods include pharmacophore modelling, and quantitative structure–activity relationship (QSAR) [ 15 – 21 ].…”

QSPRpred: a Flexible Open-Source Quantitative Structure-Property Relationship Modelling Tool

“…In the traditional sense, QSPR modelling focuses mainly on describing the relationship between the compound structure and a property of interest, but proteochemometric modelling (PCM) has emerged as an extension that also introduces the protein target information into the equation [13,14]. A PCM approach can extrapolate similarities and differences across (super)families and is therefore promising in poly-pharmacology and off-target prediction [15], as well as a strategy for data augmentation and relevant binding residue identification [16,17]. Although in the traditional sense, the architecture is identical to that of a single-task model, it includes bio-activity endpoints for multiple proteins, by featurizing each compound-protein combination separately [18].…”