Testing physical models of passive membrane permeation.

Area of Science:

• Computational chemistry and molecular modeling
• Pharmacokinetics and drug design
• Biophysics and membrane transport

Background:

• Predicting passive membrane permeability is crucial for drug design.
• Current methods often rely on statistical models, indirectly capturing physical principles.
• A need exists for physics-based approaches that directly model molecular interactions.

Purpose of the Study:

• To investigate and evaluate molecular mechanics-based models for passive membrane permeability prediction.
• To assess model performance against diverse experimental data and in silico predictions.
• To systematically analyze the contribution of individual physical factors to permeability.

Main Methods:

• Utilized molecular mechanics force fields (e.g., partial charges) and implicit solvent models.
• Developed physics-based models for passive membrane permeation.
• Evaluated models against parallel artificial membrane permeability assays (PAMPA), cell-based assays, in vivo measurements, and other computational predictions.

Main Results:

• Conformation-dependent free energy of desolvation was identified as a primary predictor of passive membrane permeation.
• This desolvation energy alone showed good agreement with experimental permeability data.
• Incorporating deionization and entropy loss factors further improved model accuracy and captured size-dependence.

Conclusions:

• Physics-based molecular mechanics models offer a robust alternative to statistical methods for predicting membrane permeability.
• Desolvation energy is a key determinant of passive permeation rates.
• These models provide valuable insights for optimizing drug candidates' pharmacokinetic properties.