Identifying Cryptic Binding Sites with Mixed Solvent MD Simulation and SiteMap

Area of Science:

• Computational chemistry
• Structural biology
• Drug discovery

Background:

• Identifying cryptic binding sites in proteins is crucial for structure-based drug discovery but challenging due to their absence in apo structures.
• Existing methods often fail to reveal these transient or hidden pockets, limiting the exploration of potential drug targets.

Purpose of the Study:

• To develop and validate a novel computational workflow for the effective identification of cryptic protein binding sites.
• To enhance early-phase drug discovery by enabling the exploration of previously inaccessible druggable targets.

Main Methods:

• Integration of mixed solvent molecular dynamics (MxMD) simulations with the SiteMap computational tool.
• MxMD was used to sample diverse protein conformations, inducing the exposure of hidden pockets.
• SiteMap was employed to identify and rank these exposed cryptic binding sites.

Main Results:

• The developed "induce-and-identify" workflow successfully identified cryptic binding sites in a challenging dataset of 65 examples.
• The workflow achieved a high success rate, identifying the correct cryptic binding site within the top 5 predictions in 78.5% of cases.
• Demonstrated robustness and effectiveness in revealing previously hidden protein pockets.

Conclusions:

• The combined MxMD + SiteMap workflow is a powerful and reliable tool for structure-based drug discovery.
• This method significantly expands the scope of druggable targets by effectively inducing and identifying cryptic binding sites.
• Facilitates early-phase drug discovery efforts by providing a systematic approach to uncover challenging protein targets.