Docking-based identification of small-molecule binding sites at protein-protein interfaces

Area of Science:

• Computational Biology
• Structural Biology
• Drug Discovery

Background:

• Protein-protein interactions are crucial in biology but difficult to target with small molecules due to a lack of clear binding pockets.
• Identifying protein interface hot-spots computationally often requires the complex structure, which is frequently unavailable.

Purpose of the Study:

• To develop and validate an integrative computational strategy for detecting protein-protein inhibitor binding sites on unbound proteins.
• To enable the identification of potential drug targets when complex structures are unknown.

Main Methods:

• Integration of molecular dynamics (MD) simulations to generate transient cavities in unbound proteins.
• Application of docking-based interface hot-spot prediction to identify suitable cavities.
• Validation using protein-protein complexes with known inhibitors and available unbound structures.

Main Results:

• Short MD simulations successfully generated transient cavities resembling known inhibitor binding sites.
• Docking simulations accurately identified the best cavities, comparable to using the actual protein interface.
• The predicted pockets were suitable for protein-ligand docking in several instances.

Conclusions:

• The proposed integrative approach effectively detects protein-protein inhibitor binding sites on unbound proteins.
• This method is valuable for targeting protein-protein interactions when complex structures are unavailable.
• The strategy shows promise for identifying novel drug targets in various protein complexes.