Protein-peptide association kinetics beyond the seconds timescale from atomistic simulations

Area of Science:

• Computational chemistry
• Biophysics
• Drug discovery

Background:

• Protein-ligand interactions are crucial for drug design.
• Atomistic molecular dynamics (MD) simulations struggle to capture long binding timescales.

Purpose of the Study:

• To compute full protein-peptide kinetics for the Mdm2-PMI complex.
• To enable accurate kinetic estimations beyond the seconds timescale using MD simulations.

Main Methods:

• Utilized a multi-ensemble Markov model framework.
• Employed all-atom molecular dynamics (MD) simulations.
• Validated results against mutagenesis data and experimental measurements.

Main Results:

• Achieved direct kinetic estimates beyond the seconds timescale with high accuracy.
• Revealed a rugged binding funnel with multiple binding pathways.
• Identified interconverting conformations on the milliseconds timescale contributing to strong binding.

Conclusions:

• The multi-ensemble Markov model framework effectively predicts protein-ligand binding kinetics.
• This approach overcomes limitations of traditional MD simulations for long timescales.
• Provides insights into the complex binding mechanisms of Mdm2 and PMI.