Abstracts

Rationale: An impairment of the blood-brain barrier (BBB) is frequently described in human temporal lobe epilepsy (TLE) and in animal models of chronic limbic epilepsy, whereas a growing body of evidences suggests that BBB failure contributes directly to the chronic epileptogenicity. Here we attempted to correlate the BBB permeability with the chronicity of epilepsy, by comparing a model of limbic epilepsy secondary to status epilepticus (SE) and the kindling model of progressive hyperexcitability. Methods: 80 rats were implanted with deep electrodes and cortical screws. 30 rats received lithium (3mEq.kg-1) 18h before pilocarpine injection (30mg.kg-1), whereas 10 rats received lithium only. EEG recording and behavioural observation were performed during SE and at various time-points of silent and chronic periods. 30 rats were given 30mg.kg-1 of pentylenetetrazol (PTZ) and 10 rats received the vehicle every second day for 28 days. EEG recording and behavioural scoring (1 to 6) were performed after each injection. For both models, treated rats and age-matched controls were sacrificed at different stages of epileptogenesis or kindlingv developement. The integrity/disruption of the BBB was evaluated by immunohistochemistry and western blot. We checked the expression and localization of the main proteins that constitute the tight junctions (ZO-1, Claudin-5, Occludin) and the leakage of IgGs which accumulate in parenchyma and in neurons. Results: 1- In the pilocarpine model, tight junction degradation and IgG leakage were obvious in limbic structures and persisted from SE to the chronic phase. 2- In PTZ kindling, a decreased expression of tight junction proteins and a leakage of IgGs were observed in hippocampus, thalamus, cingular and entorhinal cortices only during stages 1,2,3. Then, the expression of tight junction proteins rebounded as soon as stage 3 and the BBB was progressively rebuilt. Conclusions: This comparative study shows that BBB permeability is transient in chemical kindling and chronic in the lithium-pilocarpine model. Bearing in mind that BBB failure is epileptogenic, the differences between both models are consistent with their ability to generate spontaneous seizures. This is related neither to the intensity of seizures nor to excitotoxic processes, since BBB disruption occurs only after short seizures in the early phases of kindling. It seems likely that kindling induces a kind of preconditioning that protects or consolidates the BBB. Large scale approaches are needed to identify the mechanisms which confer this “vascular tolerance” against seizures.