Abstracts

Rationale: In most cases, temporal lobe epilepsies (TLE) are drug-resistant. This disorder often starts with an initial insult followed by a progressive alteration of underlying neuronal networks which leads the brain to a pathological state characterized by recurrent seizures. The epileptogenesis process often involves an initial seizure-free period. Nevertheless, during this latent period, some electrophysiological markers do appear in mesial structures. Typically, hippocampal spikes are observed in local field potentials few days after the initial lesion in various animal models of TLE. Although some recent studies have suggested that changes occur in the frequency of these epileptic spikes, it is still unclear whether the shape of these events evolves over the latent period and how this evolution, if any, can be related to underlying pathophysiological mechanisms.Methods: Our study aims to characterize and to understand, in terms of possible alteration of neuronal networks, the evolution of epileptic spikes during epileptogenesis. This study was performed using an in vivo mouse model of mesial TLE (Neuroscience 2002; 112(1):101-111). In this model, after an intra-hippocampal micro-injection of ka nic acid (KA), both structural (sclerosis) and functional (abnormal intracerebral electroencephalographic signals or iEEG) changes are observed. We have developed an automatic method to detect spikes in iEEG signals. To prevent any bias in the statistical analysis, the method has been optimized in order to keep the false detection rate as low as possible. The extracted spikes were characterized according to two features, namely their amplitude and duration. In addition, a computational model of the hippocampus - CA1 sub-field network with detailed level - (IEEE Trans Biomed Eng 2009; 56(12):2782-95) was used to physiologically interpret any observed modifications through excitability-related parameters (connectivity patterns among principal neurons and interneurons, conductances associated with AMPAergic, NMDAergic and GABAergic currents).Results: Over the latent period, results showed a high increase of the number of the hippocampal spikes. Moreover, we also found significant increases of both amplitude and duration of the spikes. Using the computational model, these changes can be explained as i) an increase of AMPAergic currents and/or ii) a decrease of GABAergic currents. The GABAA reversal potential was also found to play a major role on both amplitude and duration.Conclusions: A recent study (Epilepsia 2010; 51(3):371-383) showed that the frequency of interictal spikes in an in vivo model of TLE in rat (repeated injections of KA) increase during epileptogenesis. The authors suggest that such increase is a predictive bio-marker of epileptogenicity. In this respect, our results go beyond their quantification of the spike frequency and show that the amplitude and duration of these typical events may reflect some plasticity-related effects taking place during epileptogenesis. Financial supports: INSERM, Universit de Rennes 1.