Predicting MDCK cell permeation coefficients of organic molecules using membrane-interaction QSAR analysis

Area of Science:

• Computational chemistry
• Pharmacokinetics
• Drug discovery

Background:

• Predicting drug absorption in the gastrointestinal tract is crucial for drug development.
• Understanding molecule-membrane interactions is key to accurate permeability predictions.

Purpose of the Study:

• To develop predictive models for organic compound partitioning in gastrointestinal cells using membrane-interaction quantitative structure-activity relationship (MI-QSAR).
• To elucidate the key molecular descriptors influencing gastrointestinal absorption.

Main Methods:

• Constructed MI-QSAR models using a training set of 22 diverse compounds with measured Madin- Darby canine kidney (MDCK) cell permeability.
• Employed molecular dynamic simulations to analyze solute-membrane interactions with a lipid monolayer model.
• Utilized multidimensional linear regression and stepwise methods for model optimization.

Main Results:

• Developed MI-QSAR models for gastrointestinal absorption.
• Identified key descriptors: ClogP, highest occupied molecular orbital energy (E(HOMO)), stretch energy (E(s)), principal moment of inertia Y (PM(Y)), total connectivity (C(t)), and non-bonded interaction energy (E(nb)).
• ClogP emerged as the most significant descriptor for predicting permeability.

Conclusions:

• Gastrointestinal permeability is influenced by both drug molecule properties and the molecule-membrane interaction process.
• MI-QSAR provides a valuable framework for predicting drug absorption and optimizing drug design.