Simplified in vitro 3D co-culture-based blood-brain barrier model using transwell.

The blood-brain barrier (BBB) is a major hurdle for treatment of brain diseases. To overcome this, precise and reproducible BBB model is one of the key factors for successful evaluation of BBB-penetrating efficacy of developmental drugs. Thus, in vitro BBB model recapitulating the physiological structure of the BBB is a valuable tool for drug discovery and development for brain diseases. Here, we develop a simplified 3D co-culture-based BBB model using immortalized human brain endothelial cells and immortalized human astrocytes mixed with Matrigel allowing model preparation within 30 min. We directly compare our 3D BBB model to a 2D BBB model comprised solely of immortalized brain endothelial cells, to demonstrate that our 3D BBB model blocks penetration of Dextran molecules with various molecular weights, remain durable and impermeable even in a BBB-degrading condition, and rapidly form tight junctions while the 2D BBB model do not. In conclusion, this establishes our simplified 3D BBB model as a valuable tool for high throughput screening of drug candidates for brain diseases.

Functional validation of the simplified in vitro 3D Co-culture based BBB model.

Various methods of generating 2D and 3D in vitro blood-brain barrier (BBB) models have previously been published with the objective of developing therapeutics for brain diseases. In general, published methods including our published method demonstrate that in vivo-like semi-permeable barrier can be generated. To further verify that an in vitro BBB model closely represents BBB, functional validation is required. Here, we functionally validate our in vitro 3D BBB model using rituximab as a representative therapeutic antibody and previously published anti-TfR (transferrin receptor) antibodies as representative BBB-penetrating antibodies. We demonstrate that our BBB model can efficiently block rituximab while allowing receptor-mediated transcytosis (RMT) of anti-TfR antibodies. In addition, we showed that RMT efficacy of anti-TfR antibodies with different binding affinity can be displayed using our BBB model. In conclusion, this demonstrates that our BBB model functionally mimics the BBB as well as having BBB-like physical properties, further establishing our BBB model as a screening tool for discovery and development of therapeutics for brain diseases.