Abstracts

Rationale: Temporal lobe epilepsy is the most common type of epilepsy in adults. Seizures initiate in mesial temporal lobe structures, which include dentate gyrus. Somatostatin-immunoreactive (SS+) interneurons are a predominant GABAergic interneuron class in the hilus. They form inhibitory synapses on granule cell dendrites close to excitatory synapses from entorhinal cortex. Loss of many hilar SS+ interneurons is commonly found in patients and models. It is unclear whether surviving SS+ interneurons are damaged, unchanged, or adapted to compensate. Methods: Mice were used that selectively express GFP in a subset of SS+ interneurons. Pilocarpine (380 mg/kg) was administered at 39 ± 1 (sem) days old to induce status epilepticus. Beginning at least 10 days later, mice were video-monitored to verify development of spontaneous seizures. Controls included naïve and pilocarpine-treated mice that did not experience status epilepticus or develop epilepsy. At 74 ± 3 days old, horizontal slices (300 μm thick) were prepared. Visually identified SS+ interneurons were recorded and labeled at 32°C using patch pipettes filled with cesium-based internal solution containing 20 mM biocytin. Spontaneous and miniature IPSCs were recorded at 0 mV holding potential. Slices were fixed and processed for biocytin-labeling using a whole-mount protocol. Labeled cells were scanned with a confocal microscope. Cells were 3-D reconstructed using Neurolucida. Results: Miniature IPSC amplitudes were similar in control and epileptic mice (13 ± 1 vs 15 ± 1 pA). There was a trend toward more frequent mIPSCs in epileptic mice, which were 138% of control values (4.6 ± 1.4 vs 3.3 ± 0.5 Hz, P = 0.44, n = 10, 11). Spontaneous IPSCs were similar in amplitude (21 ± 3 vs 19 ± 1 pA), while frequencies in epileptic mice were 134% of control values (7.6 ± 1.5 vs 5.7 ± 1.7 Hz, P = 0.40, n = 14, 23). Both control and epileptic biocytin-labeled interneurons (n = 20, 20) demonstrated common features of HIPP cells, including axon projections to the molecular layer and hilar somata with variable morphology: fusiform, oval, multipolar, or pyramidal. Dendrites had long, thin spines and usually remained in the hilus, except ~30% cells in both groups extended 1-2 dendrites into the molecular layer. SS+ interneurons in epileptic mice also exhibited morphological changes. Soma area was larger, 147% of control values (227 ± 12 vs 154 ± 10 μm2, P < 0.001). There was a trend toward more dendritic ends in epileptic mice, 145% of control values (119 ± 22 vs 82 ± 21, P = 0.25). Numbers of primary dendrites were similar (5.0 ± 0.3 vs 4.8 ± 0.3), but total dendritic length was 1.2 times longer in epileptics (3234 ± 211 vs 2669 ± 159 μm, P < 0.05). Conclusions: In epileptic mice SS+ interneurons have longer dendrites and larger somata. These findings, and a trend toward increased mIPSC frequency, suggest surviving SS+ interneurons enlarge and provide more target area for synaptic input.