Interaction with specific HSP90 residues as a scoring function: validation in the D3R Grand Challenge 2015

Area of Science:

• Computational chemistry
• Drug discovery
• Structural biology

Background:

• Heat shock protein 90 (HSP90) is a key target in cancer therapy.
• Developing effective HSP90 inhibitors requires accurate prediction of ligand binding poses.
• Existing molecular docking methods face challenges in precisely identifying native binding conformations.

Purpose of the Study:

• To develop a novel scoring function, PocketScore, for enhanced identification of native binding poses.
• To evaluate PocketScore's performance in a drug discovery challenge.
• To understand the role of specific binding site residues in HSP90-inhibitor interactions.

Main Methods:

• Developed PocketScore based on interaction energy with key residues (Asp93, Thr184) and a water molecule.
• Integrated PocketScore into a molecular docking workflow.
• Applied the workflow to the Drug Design Data Resource (D3R) Grand Challenge 2015 (GC2015).

Main Results:

• PocketScore demonstrated strong performance in identifying native binding poses for HSP90 inhibitors.
• The scoring function successfully ranked 180 diverse molecules in the GC2015.
• Specific interactions with Asp93, Thr184, and a water molecule were found to be critical.

Conclusions:

• PocketScore effectively models interactions for aminopyrimidine and resorcinol-based HSP90 inhibitors.
• Accurate binding pose identification is crucial for successful virtual screening.
• PocketScore offers a valuable tool for improving docking power in drug discovery.