Perturbation Approaches for Exploring Protein Binding Site Flexibility to Predict Transient Binding Pockets.

Area of Science:

• Computational chemistry
• Structural biology
• Drug discovery

Background:

• Standard molecular dynamics (MD) simulations struggle to capture slow protein motions crucial for drug design.
• Long-timescale protein dynamics and transient pockets are often neglected in computational approaches.
• Understanding binding site flexibility is key to designing novel compounds.

Purpose of the Study:

• To introduce two novel nonequilibrium MD methods for identifying protein binding site conformational changes and transient pockets.
• To provide computationally efficient tools for exploring protein pocket flexibility.
• To aid in the functional characterization of protein pockets and ligand design.

Main Methods:

• Development of Langevin-RIP (L-RIP) using short Langevin MD simulations with side-chain perturbations.
• Introduction of RIPlig applying perturbations to pseudoligands within binding pockets.
• Validation of methods on four proteins with varying binding site flexibility.

Main Results:

• L-RIP and RIPlig identified transient pocket regions in under 10 ns total simulation time.
• Observed α-helix distortions in HSP90 and DFG loop flipping in Src kinase using L-RIP.
• Detected transient pockets missed by 100 ns of conventional MD simulations.

Conclusions:

• L-RIP and RIPlig are efficient computational tools for exploring protein binding site flexibility.
• These methods enhance the identification of transient pockets for drug design.
• The approaches facilitate functional characterization of protein pockets.