Computational prediction of solubilizers' effect on partitioning

Area of Science:

• Computational chemistry
• Pharmacokinetics
• Analytical chemistry

Background:

• Drug partitioning influences absorption and distribution.
• Solubilizers can alter drug membrane interactions.
• Predictive models are crucial for efficient drug development.

Purpose of the Study:

• To develop a computational model predicting solubilizer effects on drug partitioning.
• To evaluate drug partitioning using immobilized artificial membrane (IAM) chromatography.
• To establish structure-property relationships using molecular descriptors.

Main Methods:

• Utilized immobilized artificial membrane (IAM) chromatography to assess membrane/water partitioning.
• Calculated 2D (MOE) and 3D (Volsurf) molecular descriptors.
• Applied partial least squares (PLS) statistics to establish quantitative structure-property relationships (QSPR).

Main Results:

• Identified statistically significant relationships between molecular descriptors and IAM partitioning data.
• Achieved high predictive accuracy with 2D descriptors (R(2)Y=0.99, Q(2)=0.82-0.83), with logP being a key descriptor.
• Developed 3D descriptor models (R(2)Y=0.53-0.64, Q(2)=0.40-0.54) using five descriptors.

Conclusions:

• It is feasible to predict drug partitioning in artificial phospholipid membranes.
• The developed computational model effectively accounts for the influence of solubilizers.
• This approach can guide the selection of appropriate solubilizers in drug formulation.