Carborane clusters in computational drug design: a comparative docking evaluation using AutoDock, FlexX, Glide, and...

Area of Science:

• Medicinal Chemistry
• Computational Drug Design
• Boron Neutron Capture Therapy (BNCT)

Background:

• Boron-containing compounds are vital for cancer therapy and diagnostics.
• Existing computational drug design software often lacks parameters for boron atoms, hindering development.
• This limitation impedes the design of novel boron-based therapeutics and diagnostics.

Purpose of the Study:

• To develop efficient computational strategies for designing boron-containing drugs.
• To address the lack of boron atom parameters in molecular modeling software.
• To validate a novel approach using carbon atom substitution for boron atoms.

Main Methods:

• Developed strategies involve replacing boron atom types with carbon atom types.
• Validated methods by docking closo- and nido-carboranyl antifolates into human dihydrofolate reductase (hDHFR).
• Utilized docking software: AutoDock, Glide, FlexX, and Surflex for validation.

Main Results:

• AutoDock and Glide showed comparable efficiency in docking closo-carboranyl antifolates.
• AutoDock, Glide, and Surflex demonstrated similar performance for nido-carboranyl antifolates.
• Software performance was not significantly impacted by variations in carboranyl antifolate structures.
• Predicted binding energies aligned with experimental data across all tested programs.

Conclusions:

• The proposed carbon-for-boron substitution strategy effectively overcomes parameter limitations in computational drug design.
• This method enhances the accuracy and applicability of molecular docking for boron-containing compounds.
• The validated approach facilitates the development of new boron-based therapeutics and diagnostics, particularly for cancer treatment.