Abstracts

Rationale: Intrahippocampal injection of tetanus neurotoxin (TeNT) induces a chronic non-lesional model of temporal lobe epilepsy characterized by spontaneous seizures. TeNT is a protease which cleaves vesicle associated membrane protein (VAMP). A key question for this model is the spatial and temporal distribution of the effects of the toxin and how it relates to the development of the epileptic syndrome. Methods: One microliter of TeNT or vehicle was stereotaxically injected into the right hippocampal CA3 region of Sprague-Dawley rats. Spontaneous recurrent seizures appeared after 1-2 weeks. At 8 days after injection, rats were anaesthetized, intracardially perfused with chilled protective saline and hippocampal slices prepared for in vitro recording. Whole-cell patch clamp recordings were made from pyramidal cells in CA1c (within the area exposed directly to TeNT) and CA1a (far from the injection) and from the same regions in control rats. Cells were labelled with neurobiotin and after the recording slices were fixed for subsequent immunohistochemistry (VAMP1 and VAMP2). VAMP positive terminals were counted, including around the labelled cells. Results: VAMP1 was preferentially expressed in synaptic terminals of interneurons, while VAMP2 was expressed mainly in excitatory terminals of principal cells. The density of VAMP1 positive terminals was significantly lower in CA1c of TeNT injected rats (20±3/100µm²) than in controls (36±2/100µm²). The density of VAMP2 terminals was also lower (18±5/100µm²) than in controls (43±3/100µm²). In CA3 VAMP1 and VAMP 2 were lower in TeNT rats both ipsilateral (VAMP1: 19.8±2.6 from 35.9±1.7 /100µm²; VAMP2: 17.6±5.1 from 43.3±2.6 /100µm²) and, to a lesser degree, contralateral to the injection (VAMP1: 30.0±4.0 from 37.4±1.9 /100µm²; VAMP2: 28.0±2.2 from 44.8±1.7 /100µm²). Functionally the cleavage of VAMPs in ipsilateral CA1c was associated with decreases in frequency of spontaneous excitatory postsynaptic currents (EPSCs; 0.8±0.3 Hz, controls 2.2±0.5 Hz) and reduced maximal amplitude of evoked EPSCs (129±23 pA, controls 594±171 pA). CA1a, far from the injection, and contralateral CA1a and c showed no significant changes in spontaneous or evoked EPSCs. Evoked and spontaneous inhibitory postsynaptic potentials were absent in CA1c of TeNT animals (control maximal evoked IPSCs 392.8±83.7 pA; control spontaneous IPSC frequency 5.4±1.4 Hz). Spontaneous IPSCs in CA1a were reduced to 3.4±0.9 Hz from 6.8±1.8 Hz. There were no changes in contralateral CA1. Conclusions: This study demonstrates that TeNT affects both excitatory and inhibitory transmission. The impact of TeNT is stronger on inhibition, shifting the excitation-inhibition balance towards the excitation and providing plausible explanation for development of epilepsy and seizures. The localization of the loss of VAMP, to ipsilateral CA1a and bilateral CA3 provides a plausible basis for the pathophysiology of the early stages of this model. Supported by Epilepsy Research UK and MRC