Closing the gap between the in-vivo and in-vitro blood–brain barrier tightness

Numerous in-vitro models of the blood–brain barrier (BBB) have been developed in the hope to mimic as closely as possible the in-vivo BBB characteristics. Most models however display BBB tightness properties still very remote from those found in-vivo. We describe here the properties of an in-vitro BBB model in three configurations: primary porcine brain endothelial cells (PBEC) grown in a monoculture, or as a co-culture in close proximity to rat glial cells (contact), or with the latter at distance (non-contact). The BBB tightness as reflected by measurements of the permeability (Pe) to sucrose and of the transendothelial electrical resistance (TEER) showed that only the contact co-culture closely mimic the in-vivo BBB (Pe = 0.19 ⁎ 10− 6 ± 0.01 cm/s and TEER up to 1650 Ω ⁎ cm2). While no changes in the expression pattern of three of the major tight junction proteins, claudin-5, occludin and ZO-1, were observed using immunohistochemistry, western blot analysis showed that the expression levels of claudin-5 and occludin increase when PBEC are cultured in contact with glial cells. In addition, we found, in the contact co-culture model, a reduced sensitivity of the TEER to vinblastine, a P-glycoprotein (Pgp) substrate that disrupts the cell cytoskeleton, indicating an improved functionality of the Pgp transporters in this configuration. We conclude that the close proximity of astrocytes is crucial to the development of a tight BBB.