A new protocol for predicting novel GSK-3β ATP competitive inhibitors

Area of Science:

• Medicinal Chemistry
• Computational Drug Discovery
• Biochemistry

Background:

• Glycogen synthase kinase 3β (GSK-3β) is implicated in cancer, type-2 diabetes, and Alzheimer's disease.
• Targeting GSK-3β offers therapeutic potential for these conditions.
• Identifying novel inhibitors with diverse chemical structures is crucial.

Purpose of the Study:

• To develop and validate a novel computational protocol for identifying new GSK-3β ATP competitive inhibitors.
• To discover compounds with topologically diverse scaffolds.
• To combine ligand-based and structure-based drug design approaches.

Main Methods:

• Building and validating 3D QSAR models (CoMFA, CoMSIA) using known GSK-3β inhibitors.
• Virtual screening of a large maleimide derivative library (PubChem) against GSK-3β using FlexX-dock.
• Filtering compounds using Lipinski's rules.
• Predicting inhibitory activity using the developed 3D QSAR models.

Main Results:

• A library of 10,429 maleimide derivatives was filtered and screened.
• 617 virtual hits were identified.
• 93 compounds were predicted to have GSK-3β inhibition < 15 nM.
• 23 compounds were confirmed as GSK-3β inhibitors from literature, with inhibition ranging from 1.3 to 480 nM.
• The virtual screening protocol achieved a hit rate > 25%.

Conclusions:

• The developed protocol effectively identifies novel GSK-3β inhibitors.
• The combination of ligand- and structure-based methods validates both approaches.
• The protocol is capable of discovering inhibitors with diverse chemical scaffolds, suitable for therapeutic development.