Targeted Interneuron Ablation in the Mouse Hippocampus Can Cause Spontaneous Recurrent Seizures

AAV-DTR transfection and expression in the CA1 subregion of the hippocampus. A, Expression of DTR-EGFP can be seen by immunohistochemical labeling with anti-EGFP antibody at two weeks posttransfection. The transfected area extends throughout the medial-lateral expanse of the CA1 region of the hippocampus. Arrows indicate the pyramidal cell body layer of CA1. B, C, Higher power images demonstrate cell-specific expression in the oriens and pyramidal cell layers. D, E, Putative synaptic densities can be seen as puncta in the bright fluorescent band of the SLM region, as well as puncta in the pyramidal cell layer surrounding putative principal cell bodies (asterisks). F-H, Representative images from serial sections from one mouse demonstrate the area of transfection of the dorsal-medial region of CA1. Some expression can be seen in the deep cortical layers. Images were taken from sections corresponding to bregma -1.34 (F), -2.18 (G), and -2.80 (H). The injection was targeted to bregma -1.85.

Hippocampal interneurons are lost after targeted ablation by systemic DT administration. A-C, EGFP-expressing interneurons in the CA1 region coexpress GAD67 (arrows). D-F, At 6 d after DT treatment, a significant loss of EGFP-expressing (green, D) and GAD67-expressing (red, E) cells can be seen in the merged image (F), across all subregions of CA1. G. Quantification of the average number of virus-transfected cells (n = 3 mice/group) demonstrated a large population of EGFP-expressing cells in virus-transfected control mice that had not been given DT, and a significant reduction in EGFP-expressing cells in virus-transfected mice that had been given DT. H, Quantification of GAD67-expressing cells (in the same tissue as G for virus-transfected control and virus-transfected ablated) confirmed that the ablation resulted in a significant reduction of interneurons across all regions in the ablated (n = 3) compared to the control mice (grouped for comparison, n = 6: n = 3 virus-transfected control and n = 3 toxin control). Asterisks indicate significance at p < 0.01, two-way ANOVA, control versus ablated.

Interneuron ablation results in a specific reduction in the frequency of mIPSCs. Whole-cell recordings from CA1 pyramidal cells demonstrated that the mice that underwent the ablation procedure (DTR + DT) had a reduction in the frequency of mIPSCs. Data are arranged so that the start of each successive line is a continuation of the line above it, and the mIPSCs can be seen as sharp upward deflections from baseline.

Quantification of the properties of the mIPSCs and mEPSCs. A, The interevent interval, amplitude and decay time of both mIPSCs and mEPSCs recorded in each cell are plotted for saline-transfected controls treated with DT (saline + DT, blue), virus-transfected controls treated with saline (DTR + saline, green) and interneuron-ablated mice transfected with virus and treated with DT (DTR + DT, red). Symbols indicate data from individual cells, and bars indicate the group averages. Data are plotted for the time point of days 6-8 after DT treatment. At this time, CA1 pyramidal cells from the experimental mice had a specific deficit in the frequency of mIPSCs (asterisks, p < 0.01, ANOVA; Tables 1, 2) but not the amplitude or decay time, consistent with the loss of inhibitory synapses. B, The >10-fold shift in interevent interval of mIPSCs can also be seen as a large shift in the cumulative probability plot for the DTR + DT compared to controls.

Targeted interneuron ablation produces seizures. A, Example of a seizure recorded in the dorsal CA1 region of the hippocampus using a miniature wireless telemetry recording device mounted to the animal’s head with an electrode positioned near the CA1 pyramidal cell layer. B, Expansion of the boxed regions of the recording in A demonstrating the changes in the recorded waveforms during the seizure (1 and 2), the postictal depression (3), and return to normal baseline activity (4).

Interneuron ablation produces both convulsive and nonconvulsive seizures. Nonconvulsive and convulsive seizures were indistinguishable in the LFP recordings. The top and bottom seizures occurred ∼8 min apart in the same mouse. The first seizure (top) was associated with behavioral arrest and a gentle head nod side-to-side. The second seizure (bottom) was convulsive with forelimb clonus. Boxed regions are expanded below for both seizures to illustrate the similarities in waveforms during both high-frequency and large-amplitude periods of activity.

SRSs after interneuron ablation generally occur during a limited time period. A, Daily seizure activity is plotted for two weeks of baseline recording beginning with virus transfection and transmitter implantation. This was followed by continuous recording through the DT treatment period (shown to start at day 0, and lasting for 2 d, red bar) and beyond for several weeks for each mouse. SRSs were seen primarily in the first week after ablation. E20057 had nine seizures spanning from day 3 to day 8; four were confirmed to be convulsive (level 3 or higher). E20061 had three convulsive seizures (two level 3 and one level 5) on days 3-4. E20062 had a total of six seizures over a time period of ∼9 h. Seizures began just before the start of day 6 (11:33 P.M. on day 5); two were convulsive (level 4), four were nonconvulsive (level 2). E20064 had a total of 41 seizures that began on day 5; the first nine seizures were nonconvulsive (level 2), they then progressed to convulsive seizures (level 3-5). 42-5 had 10 seizures on day 4 to day 6; eight were confirmed nonconvulsive (level 2). 42-44 had two seizures on day 3; video from this day was either corrupt or missing. B, Interseizure interval is plotted for all seizure intervals ≤24 h for all mice. The histogram demonstrates that seizures were not confined to brief intervals, thus demonstrating that the interneuron-ablation treatment did not produce a discreet period of SE. C, The histogram of seizure durations illustrates that SRSs induced by interneuron ablation typically had a duration of 30-65 s.