Ligand-based cheminformatics and free energy-inspired molecular simulations for prioritizing and optimizing G-protein coupled receptor kinase-6 (GRK6) inhibitors in multiple myeloma treatment

Arnab Bhattacharjee; Supratik Kar; Probir Kumar Ojha

doi:10.1016/j.compbiolchem.2025.108347

Ligand-based cheminformatics and free energy-inspired molecular simulations for prioritizing and optimizing G-protein coupled receptor kinase-6 (GRK6) inhibitors in multiple myeloma treatment

Comput Biol Chem. 2025 Jan 13:115:108347. doi: 10.1016/j.compbiolchem.2025.108347. Online ahead of print.

Authors

Arnab Bhattacharjee¹, Supratik Kar², Probir Kumar Ojha³

Affiliations

¹ Drug Discovery and Development Laboratory (DDD Lab), Department of Pharmaceutical Technology, Jadavpur University, Kolkata 700032, India.
² Chemometrics and Molecular Modeling Laboratory, Department of Chemistry and Physics, Kean University, 1000 Morris Avenue, Union, NJ 07083, USA.
³ Drug Discovery and Development Laboratory (DDD Lab), Department of Pharmaceutical Technology, Jadavpur University, Kolkata 700032, India. Electronic address: probirojha@yahoo.co.in.

PMID: 39824142
DOI: 10.1016/j.compbiolchem.2025.108347

Abstract

Multiple myeloma (MM) is the second most frequently diagnosed hematological malignancy, presenting limited treatment options with no curative potential and significant drug resistance. Recent studies involving genetic knockdown established the crucial role of GRK6 in upholding the viability of MM cells, emphasizing the need to identify potential inhibitors. Computational exploration of GRK6 inhibitors has not been attempted previously. Herein, the present study reports a multilayered lead prioritization and optimization framework using chemometrics and molecular simulations. 2D QSAR studies revealed that hydrogen bonding and polar interactions enhanced GRK6 inhibitory activity, while increased electron accessibility posed a risk of off-target effects. The pharmacophore hypothesis (DDHRRR_1) featured two hydrogen bond donors, one hydrophobic region, and three aromatic rings, laying the foundation for the 3D QSAR models. Hydrophobic groups, such as pyridine and pyrazole, were shown to enhance inhibition, while smaller groups, like ethyl and hydroxyl, reduced activity. 12,557 DrugBank compounds were screened using the developed chemometric models and molecular docking in tandem, which led to the identification of 7 potential parent leads for subsequent QSAR-guided structural optimizations. 350 lead analogs were generated and the top 4 were further analyzed using molecular docking, ADMET, molecular dynamics, and metadynamics analysis based on Principal Component Analysis (PCA), Probability Density Function (PDF), and Free Energy Landscapes (FEL). Upon cumulative retrospection, we propose a novel analog of DB07168 (DB07168-A13) (docking score: -11.2 kcal/mol, MM-GBSA binding energy: -55.2 kcal/mol) as the most promising GRK6 inhibitor, warranting further in vitro validation, for addressing prospective therapeutic intervention in MM.

Keywords: Analog generation; Free energy landscapes; GRK6; Pharmacophore; Probability density function; QSAR.