Abstracts

Dentate granule cells are thought to protect the hippocampus by filtering or blocking epileptiform activity from the entorhinal cortex. In previous studies, spontaneous seizures recorded in the dentate gyrus in models of temporal lobe epilepsy appear to include field excitatory post-synaptic potentials (fEPSPs), but it is not clear whether the firing rate of individual granule cells is altered. One possibility is that granule cells block epileptiform activity and maintain roughly the same firing rate for the duration of a seizure. Alternatively, the firing rate of granule cells could increase at seizure onset, and thus propagate or even amplify epileptiform input into the hippocampus. These possibilities can be distinguished by determining whether the firing rate of individual granule cells increases during seizures. Multiple, single units were recorded from the dentate gyrus in epileptic rats at least 4 months following pilocarpine-induced status epilepticus. Broadband, tetrode recordings were made before, during and after spontaneous seizures in awake, freely-moving rats. Seizures were confirmed by behavior, and seizure onset was determined electrographically by the paroxysmal onset of rhythmic fEPSPs in the dentate gyrus. Multiple, single units were isolated offline by digital filtering, threshold detection and cluster cutting. Individual units could be placed into one of two classes based on firing properties during rest that, when compared against unit recordings from the dentate gyrus reported in the literature, corresponded to putative interneurons (mean rate [gt] 5.0 Hz, unimodal distribution of inter-spike intervals (ISIs), symmetric spike waveform) and granule cells (mean rate [lt] 5.0 Hz, multimodal distribution of ISIs, asymmetric spike waveform). During seizures in two rats, granule cell firing rates increased dramatically (3.4[plusmn]1.4 Hz to 24.2[plusmn]14.9 Hz, N=5) and rapidly at the time of seizure onset, tens of seconds prior to the observation of population spikes in the field recordings. Following a seizure, granule cell activity was sharply attenuated for up to several minutes; i.e., during behavioral post-ictal depression. Within five minutes subsequent to the end of the seizure, granule cell activity returned to pre-seizure levels. Interestingly, in contrast to granule cells, firing rates for interneurons during seizures declined slightly, but not significantly (12.8[plusmn]3.3 Hz to 9.3[plusmn]2.1, N=5). These findings suggest that, in this model of temporal lobe epilepsy, dentate granule cells do not filter or block epileptiform activity from entering the hippocampus. Rather, a rapid increase in their firing suggests that granule cells might play an active role in amplifying or possibly generating seizure activity. (Supported by NIH NS07280.)