In vitro models for the blood-brain barrier

Area of Science:

• Pharmacology and Toxicology
• Neuroscience
• Biotechnology

Background:

• Accurate prediction of blood-brain barrier (BBB) transport is crucial for developing central nervous system drugs.
• Existing in vitro models often struggle to replicate the complex in vivo BBB environment.
• Computational models have shown limited success in predicting BBB permeability.

Purpose of the Study:

• To identify and evaluate continuous cell line models for predicting blood-brain barrier (BBB) transport.
• To investigate the specificity and correlation of various in vitro models with in vivo BBB data.
• To assess the ability of in vitro models to differentiate between various compound transport mechanisms.

Main Methods:

• Evaluated primary brain endothelial cells (bovine, human) and immortalized cell lines (rat, mouse) against in vivo BBB data.
• Utilized non-BBB cell lines (ECV/C6, MDCK, Caco-2) for comparative analysis.
• Developed a computational prediction model using Molsurf descriptors and partial least squares projection to latent structures (PLS).

Main Results:

• No overall linear correlation between in vivo and in vitro permeability was found across all models and compounds.
• MDCKwt and bovine brain endothelial cells showed moderate correlation for passively transported compounds (r2=0.65-0.74).
• The MDCKmdr-1 cell line demonstrated the best separation of passively distributed compounds from efflux substrates.

Conclusions:

• No single in vitro model accurately predicted all aspects of in vivo BBB transport.
• The MDCKmdr-1 cell line shows potential for identifying efflux substrates, improving in vitro-in vivo correlation.
• Development of combined in vitro testing strategies is likely necessary for reliable prediction of in vivo brain distribution.