Seizure-induced pathologic plasticity and cell death in epileptogenic focal cortical dysplasia: converging evidences from human patients and an experimental model

RATIONALE: We previously demonstrated in a rat model of chronic epilepsy associated with cortical malformations, the methylazoxymethanol (MAM)-pilocarpine -MP- rats, that status epilepticus (SE) and subsequent seizures set in motion a maladaptive plasticity altering the morphology of cortical pyramidal neurons and NMDA receptor expression and localization. We have here exploited the same model to verify i) the activation of cell death pathways in both neurons and glia during the course of epilepsy and ii) whether these changes could be related to epilepsy course. METHODS: MAM rats treated with pilocarpine and diazepam (MDP, n=5), not experiencing SE, were used as controls. MP rats were analyzed at 18h after SE (acute phase, MP-acute, n=3), 3-5 days after epilepsy onset (early chronic phase, MP-EC, n=3) and 3 and 6 months after epilepsy onset (chronic phase, MP-3m, n=8, MP-6m, n=5). Morphometric analysis of neocortex and hippocampus was performed and cell degeneration assessed with Fluoro-Jade (FJ) staining. We also used single- and double-labeling immunocytochemistry to study the activation of different pathways possibly leading to cell death. RESULTS: Our morphometric analysis revealed progressively reduced cortical thickness and hippocampal and thalamic volumes during chronic epilepsy course. FJ analysis revealed the activation of cell death processes not temporally restricted to the primary insult of pilo-induced SE, but extending to the different epileptic stages considered up to 6 months (Fig 1). We also found a region-specific pattern of FJ activation, degenerating neurons being more evident at early chronic epileptic stages in cortex and thalamus and at late chronic stages in hippocampal CA layers. Granule cells were consistently FJ- throughout all stages. Reactive astrocytes with larger cell bodies, thicker processes, and more eccentric nuclei became increasingly evident. Caspase3 and c-Jun activation became also progressive in chronic MP rats. Double IF experiments with glial and neuronal markers revealed a clear activation of c-Jun in cortical and hippocampal neurons, but not in glia, suggesting the activation of excitotoxic cell death pathways leading to neuronal death. Caspase3 was active in hippocampal CA pyramidal neurons, likely indicating apoptosis, and particularly in cortical and hippocampal glial cells (Fig 2). Triple-labeling data with FJ, caspase3 and GFAP revealed that caspase3+ glial cells were consistently FJ negative. Thus, caspase3 activation in glia may be related to astroglial plasticity and not simply to cell death. CONCLUSIONS: Our data support the content that neuronal death processes play a relevant role in the pathologic plasticity set in motion by SE and subsequent seizures in the malformed brain of MP rats. They also demonstrated that cell death was related also to the occurrence of spontaneous seizures and not the long-term consequence of SE. Finally, they indicated that different intracellular pathways leading to cell death or remodeling were activated by seizures in neurons and glia.