Mechanisms of hyperexcitability and efficacy of antiepileptic drugs in hippocampal-entorhinal networks in the Reduced Intensity Status Epilepticus (RISE) model of chronic epilepsy

Abstract

A third of epilepsy patients are resistant to anti-epileptic drug (AED) treatment leading to reduced quality of life, increased treatment costs and complexities surrounding polytherapy. The overall aim of this project was to explore dynamic network changes in the excitability and efficacy of AEDs in: acute models of epileptiform activity, chronic models of epileptogenesis and in resected human tissue, in vitro. Initial studies investigated the differences in the neuronal network excitability induced by 0[Mg]2+ in rat brain slices prepared using either a standard NaCl-based aCSF or a sucrose-based aCSF. Standard prepared slices were more excitable in comparison to sucrose-based aCSF prepared slices. Immunohistochemical investigations for parvalbumin demonstrated a reduction of interneurons in slices prepared in the standard way. There was little difference in response to combination AEDs, but this could be due to increased latency to first seizure in sucrose prepared slices. LTP was suggested to play a role in the resistance to AEDs. These results suggest sucrose prepared slices better preserve the neuronal network in vitro, and serve as a better acute model for assessing AEDs and mechanisms of resistance. Sucrose perfused slices were prepared from rodents that had undergone a refined chronic Li-pilocarpine-based model of epileptogenesis (RISE) to investigate the effects of six AED combinations on network excitability (24 hrs and 1, 5 and 12 weeks post status). Ictal-like discharges (IDs) were seen in significantly greater numbers in slices from RISE animals compared to age-matched controls. Additionally, RISE slices showed a consistently shorter latency to first seizure across all time points. Investigations exploring the efficacy of different AED combinations during epileptogenesis showed that the tiagabine and carbamazepine combination was most effective in reducing measures of ictal activity whilst the combination of lamotrigine and gabapentin was least effective. The resistance of different drug combinations was also variable depending on the stage of epileptogenesis. These findings suggest that vulnerable networks show underlying hyperexcitability even at stages when chronic behavioural seizures are not yet developed, and that the RISE model may provide insights into the variable efficacy of AEDs. In comparison to chronically epileptic rodent tissue, epileptic human tissue from the temporal lobe was not as excitable, and often required stronger ID inducing manipulations. Once IDs were initiated in vitro, inter-event intervals between seizures were longer in comparison rodent epileptic tissue. Discrepancies in excitability could be attributed to the likelihood that damage within human tissue is likely to be subtle, hence require more stimulation to induce ictal-like activity (Gabriel et al., 2004). There was a developmental trend for excitability, in response to low concentrations of the NMDA antagonist MK801 (100-300 nM), to decrease in controls and remain elevated in epileptic animals. The NOS inhibitor, 7-nitraindazole, failed to stop the induction of IDs by low concentrations of MK801. Additionally, low concentrations of MK801 had no significant effects on the frequency and amplitude of field IPSPs in control and SE latent period slices. Further investigations are required to elucidate the mechanisms of how altered excitatory drive of inhibition may promote network excitability in epilepsy Overall, my findings suggest network changes in excitability occur at stages when chronic behavioural seizures are not yet developed, and that the RISE model may provide insights into the variable efficacy of AEDs and underlying mechanisms of epileptogenesis