Predicting a Drug's Membrane Permeability: A Computational Model Validated With in Vitro Permeability Assay Data.

Area of Science:

• Computational chemistry
• Pharmacology
• Biophysics

Background:

• Membrane permeability is crucial for drug efficacy, especially for small molecules targeting intracellular components.
• Accurate prediction of passive permeability is essential in drug design to ensure compounds reach their intracellular targets.

Purpose of the Study:

• To develop and validate a computational model for assessing passive permeability of compounds through lipid bilayers.
• To improve the quantitative prediction of permeability profiles compared to existing methods.

Main Methods:

• Utilized umbrella sampling molecular dynamics (MD) simulations to model compound passage through a lipid bilayer.
• Calibrated and validated the computational model using in vitro parallel artificial membrane permeability assay (PAMPA) data.
• Evaluated the model's predictive accuracy with custom-synthesized, closely related compounds.

Main Results:

• The developed MD model demonstrated substantially improved agreement with experimental PAMPA data.
• The computational approach showed superior quantitative prediction of permeability profiles.
• Progressive refinement through in vitro assays enhanced the model's predictive capabilities.

Conclusions:

• The study presents a validated computational model for predicting passive membrane permeability.
• This predictive tool can aid in the rational design of drug candidates with improved membrane transport properties.
• The synergistic approach combining MD simulations and PAMPA assays offers a robust method for permeability assessment.