Computational insights into the interaction of small molecule inhibitors with HRI kinase domain

Abstract

The Heme-Regulated Inhibitor (HRI) kinase regulates globin synthesis in a heme-dependent manner in reticulocytes and erythroid cells in bone marrow. Inhibitors of HRI have been proposed to lead to an increased amount of haemoglobin, benefitting anaemia patients. A series of indeno[1,2-c]pyrazoles were discovered to be the first known in vitro inhibitors of HRI. However, the structural mechanism of inhibition is yet to be understood. The aim of this study was to unravel the binding mechanism of these inhibitors using molecular dynamic simulations and docking. The docking scores were observed to correlate well with experimentally determined pIC₅₀ values. The inhibitors were observed to bind in the ATP-binding site forming hydrogen bonds with the hinge region and van der Waals interactions with non-polar residues in the binding site. Further, quantitative structure–activity relationship (QSAR) studies were performed to correlate the structural features of the inhibitors with their biological activity. The developed QSAR models were found to be statistically significant in terms of internal and external predictabilities. The presence of chlorine atoms and the hydroxymethyl groups were found to correlate with higher activity. The identified binding modes and the descriptors can support future rational identification of more potent and selective small molecule inhibitors for this kinase which are of therapeutic importance in the context of various human pathological disorders.

Keywords:

• HRI
• EIF2AK1
• inhibitors
• docking
• indeno[1
• 2-c]pyrazoles

Abbreviations:

• EIF2AK1 – Eukaryotic Initiation factor alpha kinase I
• HRI – Heme Regulated Inhibitor Kinase
• PKR – RNA-dependent Protein Kinase
• PERK – PKR-like Endoplasmic Reticulum Kinase
• GCN2 – General Control Non-derepressible 2 Kinase
• PDB – Protein Databank
• BLAST – Basic Local Alignment Search Tool
• QSAR – Quantitative Structure–Activity Relationship
• ATP – Adenosine triphosphate
• MLR – Multiple Linear Regression
• kNN-MFA – k-Nearest Neighbour Molecular Field Analysis

Acknowledgements

MJ thanks DST-PURSE for HPC facility and Bioinformatics Centre for infrastructure support. SP, MJ, APK and JKP gratefully acknowledge DBT grant BT/PR8749/BID/7/473/2013. The authors thank Abhishek Subramanian and Ganesh Shahane for initiating various aspects of this work as a part of their MSc projects.