Abstracts

The blood[ndash]brain barrier (BBB) is formed by the endothelial cells lining the brain microvessels. Complex tight junctions linking adjacent endothelial cells make brain capillaries around 100 times tighter than peripheral capillaries to small hydrophilic molecules. As a result, drugs required to act in the brain, including antiepileptic drugs (AEDs), have generally been made lipophilic, able to cross the brain endothelium via the lipid membranes. However, such lipophilic drugs are potential substrates for efflux carriers of the BBB which are known to be over-expressed in epileptic brain. We have tested the predictive ability of a new BBB model based on co-culture of C6 rat glioma and bovine aortic endothelial cells (BAEC) differentiated in a BBB phenotype under dynamic conditions (DIV-BBB). We also compared the permeation properties in this model against a widely used static transwell system. BAEC and C6 were grown in DMEM medium. In the DIV-BBB cells were co-cultured for 14 days using polypropylene hollow fibers inside a sealed chamber (the[italic][rdquo]extraluminal space[rdquo][/italic]) accessible by ports. A variable-speed pulsatile pump generates flow from the medium source through the capillary bundle allowing diffusion of nutrients out to the extra luminal space (ECS) through the 0.64 mM trans-capillary pores. Metabolic products are similarly removed. In parallel experiments, static BBB co-cultures were established in Transwell apparati. Permeability to a highly polar molecule (3H sucrose) and a widely used AED (14C Phenytoin) were measured. A similar experiment was performed after osmotic opening of the BBB with intraluminal mannitol. Transendothelial electrical resistance (TEER) was measured to assess the barrier integrity during the establishment on the BBB and during the course of the experiments. The TEER values in the DIV-BBB were on average [gt]800 Ohm cm², against a value of 40 in Transwells. The average sucrose permeability measured in 3 independent experiments was 5.5 x10^-6 cm/sec in the DIV-BBB against 8.2 x10^-5 cm/sec in the Transwell system (p[lt]0.05). Phenytoin permeability was on average 1.4 x10^-5 cm/sec in the DIV-BBB and 7.8 x10^-5 cm/sec in the Transwell model. Osmotic opening of the BBB by hyperosmolar injection of mannitol lead to a 1000x increase in both sucrose and Phenytoin permeability in the DIV-BBB. These results demonstrate that the DIV-BBB recapitulates some of the crucial characteristics of the in situ BBB, including tightness (high TEER), and selective permeability. Drug screening by traditional Transwell methods may give false permeability values, in particular for drugs that are poorly permeant across the BBB. Additional experiments with human cell isolated from drug resistant epileptics will further help characterize the utility of this model for CNS drug delivery studies. (Supported by NIH-2RO1 HL51614, NIH-RO1 NS 43284, NIH-RO1 NS38195 to Damir Janigro.)