In vitro blood-brain barrier (BBB) models are a useful tool to screen the permeability and toxicity of new drugs. Currently, many different in vitro BBB models coexist, but none stands out as being notably better than the rest. Therefore, there is still a need to evaluate the quality of BBB models under various conditions and assess their ability to mimic the in vivo situation. In this study, two brain endothelial cell lines (bEnd.3 and hCMEC/D3) and two epithelial-like cell lines (MDCKII and Caco-2) were selected for BBB modelling purposes. They were grown as monolayers of a single cell type, under the following conditions: in coculture with either primary or immortalised astrocytes; or in the presence of primary or immortalised astrocyte-derived conditioned media. A total of 20 different BBB models were established in this manner, in order to assess the effects of the astroglial components on the BBB phenotype in each case. To this end, six parameters were studied: the expression of selected tight junction proteins; the enzyme activities of alkaline phosphatase and of gamma glutamyl transpeptidase; the transendothelial/transepithelial electrical resistance (TEER); restriction in paracellular transport; and efflux transporter inhibition were each evaluated and correlated. The results showed that coculturing with either primary or immortalised astrocytes led to a general improvement in all parameters studied, evidencing the contribution of this cell type to effective BBB formation. Furthermore, the permeability coefficient (P e) of the tracer molecule, Lucifer Yellow, correlated with three of the six parameters studied. In addition, this study highlights the potential for the use of the Lucifer Yellow P e value as an indicator of barrier integrity in in vitro BBB models, which could be useful for screening the permeability of new drugs.
Keywords: BBB; TEER; astrocytes; blood–brain barrier; glia; in vitro modelling; model evaluation; permeability coefficient; permeability screening.