How Robust Is the Ligand Binding Transition State?

Area of Science:

• Computational chemistry
• Biophysics
• Pharmacology

Background:

• Drug efficacy is often better predicted by drug binding kinetics (on-rate, off-rate) than thermodynamics alone.
• Optimizing drug compounds requires predictive computational models based on kinetics.
• Understanding ligand binding transition states is crucial for these models, despite their short lifetimes.

Purpose of the Study:

• To computationally model ligand unbinding events for soluble epoxide hydrolase (sEH) inhibitors.
• To analyze ligand binding transition state ensembles (TSEs) for inhibitors with varying residence times.
• To identify challenges and opportunities for kinetics-based drug design.

Main Methods:

• Utilized the weighted ensemble method REVO (resampling of ensembles by variation optimization).
• Simulated unbinding paths for five sEH inhibitors with residence times from 14.25 to 31.75 minutes.
• Analyzed unbinding ensembles, focusing on transition state ensemble features and protein-ligand interactions.

Main Results:

• Achieved average prediction accuracy within an order of magnitude for residence times.
• Observed significant differences in TSEs (spatial distribution, protein-ligand interactions) for ligands with similar bound poses.
• Identified commonalities in TSEs when considering general features like ligand degrees of freedom.

Conclusions:

• Ligand binding transition state ensembles exhibit complex behavior, varying even for similar bound poses.
• General features like ligand degrees of freedom show similarities across different TSEs.
• Significant challenges remain for rational, kinetics-based drug design due to the complexity of transition states.