Integrative Profiling for BBB Permeability Using Capillary Electrochromatography, Experimental Physicochemical...

Area of Science:

• Pharmacology
• Analytical Chemistry
• Biophysics

Background:

• Accurate prediction of blood-brain barrier (BBB) permeability is crucial for optimizing drug pharmacokinetics in early development.
• Traditional in vivo methods for determining BBB penetration (log BB) are resource-intensive and time-consuming.
• There is a need for high-throughput, in vitro methods to efficiently assess drug candidates' BBB permeability.

Purpose of the Study:

• To develop and validate a novel in vitro method for high-throughput prediction of blood-brain barrier (BBB) log BB values.
• To integrate experimental data from capillary electrochromatography (CEC) and potentiometric titrations for enhanced predictive accuracy.
• To establish a machine learning model for rapid classification of compound BBB permeability.

Main Methods:

• Utilized open-tubular capillary electrochromatography (CEC) with liposome-coated capillaries to mimic biological membranes.
• Combined CEC retention factor (k'), pKa, and log D7.4 to develop a predictive regression model for log BB.
• Employed machine learning algorithms (Dynamic Time Warping, k-NN, Bag-of-SFA-Symbols) for classification based on CEC electropherograms.

Main Results:

• A regression model (log BB = -2.45 + 0.1k' + 0.3logD7.4 + 0.27pKa) achieved an R² of 0.64.
• Preliminary CEC analyses showed promising correlation between log k' and log BB for neutral drugs.
• Machine learning classification of CEC electropherograms yielded 0.81 accuracy and 0.81 F1weighted score.

Conclusions:

• The developed in vitro CEC-based method offers a viable, high-throughput alternative for predicting blood-brain barrier (BBB) permeability.
• Integration of physicochemical parameters and chromatographic data enhances the accuracy of log BB prediction.
• Machine learning analysis of CEC data enables rapid classification of compound BBB penetration potential.