Identifying Novel Drug Targets by iDTPnd: A Case Study of Kinase Inhibitors

Hammad Naveed,Corinna Reglin,Thomas Schubert,Xin Gao,Stefan T Arold,Michael L Maitland

doi:10.1016/j.gpb.2020.05.006

Abstract

Current FDA-approved kinase inhibitors cause diverse adverse effects, some of which are due to the mechanism-independent effects of these drugs. Identifying these mechanism-independent interactions could improve drug safety and support drug repurposing. Here, we develop iDTPnd (integrated Drug Target Predictor with negative dataset), a computational approach for large-scale discovery of novel targets for known drugs. For a given drug, we construct a positive structural signature as well as a negative structural signature that captures the weakly conserved structural features of drug-binding sites. To facilitate assessment of unintended targets, iDTPnd also provides a docking-based interaction score and its statistical significance. We confirm the interactions of sorafenib, imatinib, dasatinib, sunitinib, and pazopanib with their known targets at a sensitivity of 52% and a specificity of 55%. We also validate 10 predicted novel targets by using in vitro experiments. Our results suggest that proteins other than kinases, such as nuclear receptors, cytochrome P450, and MHC class I molecules, can also be physiologically relevant targets of kinase inhibitors. Our method is general and broadly applicable for the identification of protein–small molecule interactions, when sufficient drug–target 3D data are available. The code for constructing the structural signatures is available at https://sfb.kaust.edu.sa/Documents/iDTP.zip.

Highlights

Proteins that contain kinase domains are involved in numerous cellular processes including signaling, proliferation, apoptosis, and survival [1,2]
Type I kinase inhibitors bind to the active forms of kinase domains in an ATP-competitive manner, with the aspartate amino acid facing into the active site
For these drugs, we obtained an average sensitivity and specificity of 31(±34)% and 78(±25)%, respectively, using the cut-off of 0.85 specified in our previous study [38] (Table 1, Table S2). This large standard deviation suggests that the algorithm’s performance based on a positive signature is not very reliable when the targets and non-targets share significant similarity. This might be due to a combination of reasons, i) the ATP binding pocket is structurally conserved across the kinase domains, ii) the orientation of the DFG motifs differs across the kinase domains, and iii) there are subtle changes in the binding interaction of the kinase inhibitors with kinase domains [42]

Summary

Introduction

Proteins that contain kinase domains are involved in numerous cellular processes including signaling, proliferation, apoptosis, and survival [1,2]. Given that the ATP binding site has necessarily conserved features across most kinase domains, several kinase inhibitors interact with the human kinome broadly and are not very selective (on average, 135 or 26% of all human kinases interact with one or more kinase inhibitors included in this study) [10,11]. This broad reactivity affects the inhibitor’s efficacy and toxicity [12,13,14]. Predicting kinase inhibitor targets or off-targets is central for the rapid and cost-efficient development of inhibitors, as it allows a better understanding of a drug’s adverse effects and exploration of the drug repositioning opportunities [15]

Methods

Results

Conclusion