"In vitro" models of the blood-brain barrier : applications and evaluation of a new human immortalized brain capillary endothelial cell line

An optimal in vitro permeability screening model for the evaluation of BBB drug
permeability accounts for active and passive transport processes, as well as for nondefined
drug-cell interactions. In addition, it should be as little laborious as possible,
and preferably from humans. Such an in vitro model would be an important tool for
the evaluation of BBB permeability in drug development. However, most of the
established in vitro models use cells of non-human origin, which is not optimal for the
prediction of brain permeability in humans. Therefore, we evaluated the immortalized
human brain capillary endothelial cell line BB19 as an in vitro model of the BBB.
BB19 cells are derived from human brain endothelium and are immortalized with
E6E7 genes of human papilloma virus. They express factor VIII-related antigen and
von Willebrand’s factor. Cells exhibit cobblestone-like morphology. However,
tightness of the cell monolayer had not been investigated so far. Restrictive tight
junctions are a prerequisite for drug transport studies. Therefore, the sucrose
permeability of BB19 cells on different filters was tested and compared to porcine
brain capillary endothelial cells. However, the tightness of BB19 cell monolayers was
insufficient and still needs further optimization. Furthermore, the ability of BB19 cells
to discriminate between the paracellular marker sucrose and the transcelluar marker
propranolol was tested in a transport assay. However, hardly any discrimination
between sucrose and propranolol (Papp = 1.30 x 10-5 vs. 2.18 x 10-5) was seen. The
effects of different supplements such as chlorophenylthio-cAMP with the
phosphodiesterase inhibitor RO-20-1724, dexamethasone, 1,25 dihydroxyvitamin D3,
and C6-conditioned medium on cell morphology, ZO-1 expression, and tightness of
the BB19 cell monolayers were investigated. Cells showed an improvement towards
a more primary BCEC morphology with C6-conditioned medium, dexamethasone,
and 1,25-dihydroxyvitamin D3. In a next step, we studied the expression of important
BBB transporters, such as P-gp, MRPs, SLCs, and BCRP. The presence of P-gp,
MRP4, and BCRP has been shown on mRNA level, by immunostaining, and Western
blot. MRP1, MRP2, MRP5, OAT3, and OAT4 were also detected by RT-PCR.
Functional properties of the BBB were shown with uptake of propranolol, morphine,
and sucrose. The uptake data was similar to the results gained from the established
porcine brain capillary endothelial cell model. Uptake studies with daunomycin and
the P-gp inhibitor verapamil showed functional activity of P-gp. We conclude that
BB19 cells might be feasible as a human in vitro model of the BBB for drug uptake
studies. However, for the assessment of transport studies, further improvements of
this model are necessary.
Furthermore, we assessed the expression and inducibility of CYPs, that may play a
role in the metabolism of CNS-active drugs, in BB19 cells. So far, only little is known
about the expression and functional role of CYPs at the BBB. The presence of
CYP1A1, and to a lower extent, of CYP3A4 could be shown on RNA level. Treatment
with benzo[a]pyrene induced the CYP1A1 transcript level approximately 11-fold,
whereas the treatment with rifampicin did not significantly change the expression
level of CYP3A4. CYP1A1 was also detected by immunostaining and Western blot.
However, no inducibility of CYP1A1 could be observed on protein level. No functional
activity could be shown in the P450-GLOTM assay.
Only little is known about the impact of P-gp for the distribution of neuroleptic drugs
into the brain. In collaboration with the Psychiatric Clinic of the University of Mainz,
Germany, the potential of neuroleptic drugs and drug metabolites to modulate P-gp
was studied in uptake and efflux assays in the P-gp overexpressing cell line
P388/mdr1. Aripiprazole and quetiapine, and to a lower extend risperidone, the
metabolite 9-OH risperidone, the metabolite norclozapine, and olanzapine could be
identified as P-gp inhibitors. Clozapine was a weak inhibitor of P-gp, while
haloperidol did not show any modulation of P-gp function.
In an industrial collaboration project, we evaluated the BBB permeability of different
preclinical CNS active compounds in porcine BCECs, and aimed to develop a
combined in vitro model, simultaneously studying BBB penetration and
pharmacolological effect.
The investigated preclinical compounds could be ranked according to their BBB
permeability in porcine BCEC.
Until now, drug candidates pass through a consecutive series of screening assays,
where in a first screening procedure the antagonism or inhibition of a receptor or
enzyme is tested in vitro. In a second step, the membrane permeability properties are
assessed in cell-culture models (White, 2000). However, CNS-active compounds first
have to pass the BBB (which we assessed with the transport through porcine BCEC);
only this fraction will contribute to the effects on the receptor (which we obtained in a
target receptor screening assay). This situation was incorporated in our combined
approach, improving the screening–criteria for the selection of potential drug
candidates, by reason that a compound with a moderate pharmacological efficacy
(which usually is excluded from further screens), but with high BBB permeability,
might be a better drug candidate, than a compound with a high pharmacological
efficacy but poor BBB permeability properties.
In addition, knowing the dose-response relationship of a test compound in the target
receptor screening assay from a standard curve, we evaluated, whether it was
possible to directly estimate the extent of BBB permeability of a test compound in this
combined in vitro model, which could save elaborate analytical measurements.
In a first approach, we performed the transport of CNS active compounds from AP to
BL in porcine BCEC. The collected samples from the BL compartment were applied
in the target receptor screening assay. Our preliminary experimental results revealed
that this combined in vitro assay might be a promising new approach for the
identification of drug candidates. However, with the future goal of application in
industrial drug screening, extensive further opimization is necessary.