Abstracts

Rationale: Broadly speaking, anti-epileptic drugs work to either enhance inhibition (benzodiazapenes, barbiturates, etc.) or reduce excitation/excitability (phenytoin, topiramate, etc.). Studying the interactions between excitation and inhibition in neural networks has been technically challenging, limited primarily to the scale of either broad pharmacological manipulations or paired intracellular recordings. Here, we have attempted to bridge this gap, using a combination of recently developed imaging and genetic techniques, to study the pathological neuronal network dynamics that lead to seizures in acute slices of hippocampus and entorhinal cortex. Specifically, we looked at cell-type specific (interneuron vs. excitatory cell) firing patterns at seizure onset to gain insight into the interplay between inhibition and excitation. Methods: Using a recently developed laser-scanning strategy, Targeted Path Scanning (TPS), in conjunction with two-photon excitation of bath-applied, calcium-sensitive dye, Indo-1 AM, we imaged epileptiform activity in slices of hippocampal formation from GAD67-GFP (GIN) mice. In this way, we were able to record simultaneously activity in populations of GABAergic interneurons (I-cells) and putative excitatory neurons (E-cells). We compared observed I-cell and E-cell calcium dynamics to their electrical activity by targeting simultaneous patch clamp recording to GFP and GFP- cells. Finally, we tested our hypothesis that GABAA synapses were driven into a depolarizing regime and contributing to seizure onset using pharmacological manipulations and a transgenic mouse, Clomeleon-1, which expresses a genetically encoded chloride sensor. Results: At 4-AP-induced SLE onset, we observed a high amplitude pre-ictal calcium spike that is significantly larger in I-cells than it is in E-cells (n=25, p<0.05). We hypothesize that I-cells fire hard enough to become depolarizing, resulting in a positive feedback loop capable of generating seizures. Simultaneous dual patch recordings of GFP-expressing interneurons and putative E-cells confirm the elevated I-cell firing rates and reveal seemingly random firing before the I-cell-dominated pre-ictal spike. After the pre-ictal spike, however, E-cell spikes follow I-cell spikes by approximately 3ms, suggesting a monosynaptic, excitatory GABAergic connection. Furthermore, acetazolamide, a drug that hyperpolarizes the GABA reversal potential by blocking production of the depolarizing, GABA-synapse permeant ion, bicarbonate, dramatically reduces ictal-like activity. Finally, using Clomeleon mice, which express a genetically-encoded chloride sensor, we directly imaged a substantial (10-20mM) increase in intracellular chloride at SLE onset. Conclusions: We have provided evidence that, at seizure onset, interneurons produce a large calcium transient, corresponding to elevated firing rates. This leads to postsynaptic chloride accumulation, which results in paired I-before-E action potentials. Together, these results suggest that, in the acute slice preparation, interneurons become transiently excitatory, producing a positive feedback network that contributes to SLE generation.