Fig. 1. Retrograde amnesia resulting from FPI in mice was assessed using contextual fear response trained 24 h prior to FPI and tested 1 week after brain injury. (A) The average percent fear-associated freezing in the training context over the 5-min test. (B) The percent freezing by minute for the 5-min test in the training context. * P<0.05 compared with sham. Data are group means±standard error of the mean.

Fig. 2. FPI induces opposing shifts in hippocampal excitability. (A) Extracellular I/O curves for the DG from PP stimulation. The slope of the fEPSP normalized to the maximal sham value is plotted versus stimulation intensity. Insets are fEPSP waveforms from sham, FPI and FPI superfused with zero calcium aCSF all at 60 mV stimulation intensity. (B) Extracellular I/O curves for area CA1 from Schaffer collateral (SC) stimulation. The responses from FPI mice are smaller compared with those from sham animals. Insets are fEPSP waveforms from sham, FPI and sham superfused with zero calcium aCSF all at 60 mV stimulation intensity. (C) Effect of injury on the fEPSP-population E-S relation. Extracellular E-S curves were constructed from the ratio of the amplitude of the PS and the slope of the fEPSP amplitude (n=5, 4 for sham and FPI respectively) over a range of nine stimulation strengths (see Experimental Procedures).

Fig. 3. Extracellular I/O curves with modulators of inhibition in DG and area CA1. (A) DHK (300 μM) perfusion in slices from FPI animals returns injury-induced DG hyperexcitability to that measured in sham slices. (B) BMI (30 μM) offsets injury-induced area CA1 decrease in net synaptic efficacy. * P<0.05 compared with sham. † P<0.05 compared with drug treatment. Data are group means±standard error of the mean.

Fig. 4. FPI induces opposing shifts in inhibitory synaptic transmission. (A) IPSC I/O curves for the DG from local stimulation. Net IPSC charge transfer normalized to the maximal sham value is plotted versus stimulation intensity. Insets are IPSC waveforms from sham, FPI at 150 μA stimulation intensity. (B) IPSC I/O curves for area CA1 from local stimulation. Net IPSC charge transfer normalized to the maximal sham value is plotted versus stimulation intensity. Insets are IPSC waveforms from sham, FPI at 150 μA stimulation intensity. * P<0.05 compared with control. Insets are averaged (n=3 for all records) eIPSCs at 70 μA stimulation. Data are group means±standard error.

Fig. 5. Slice recordings from dentate granule neurons from FPI animals demonstrate reductions in spontaneous inhibitory activity 1 week after FPI. The mIPSCs are smaller and less frequent in dentate granule neurons from FPI animals. (A) Continuous sweeps of mIPSCs from sham (left panel) and FPI (right panel) dentate granule cells at −60 mV. (B) Cumulative frequency amplitude histograms for neurons in (A) with representative mIPSCs (C; average of 50) from sham and injured dentate granule neurons in (A). Histograms of mean (D) median mIPSC amplitude, (E) net charge transfer, and (F) mIPSC frequency demonstrate significant reductions in inhibitory activity of neurons from FPI mice compared with ‘control’ animals. * P<0.05 compared with control. Data are group means±standard deviation.

Fig. 6. Slice recordings from area CA1 pyramidal neurons from FPI animals demonstrate increases in spontaneous inhibitory activity 1 week after FPI. The mIPSCs are larger in CA1 pyramidal neurons of FPI mice. (A) Cumulative frequency amplitude histograms for neurons from sham and FPI animals with representative mIPSCs (B; average of 50). Histograms of mean (C) median mIPSC amplitude and (D) net charge transfer demonstrate significant increases in inhibitory activity of neurons from FPI mice compared with control animals. (E) Histogram demonstrating that mIPSC frequency is not significantly different in sham compared with FPI slices. * P<0.05 compared with control. Data are group means±standard deviation.

Total number of neurons in the ipsilateral hippocampal subregions of sham and FPI mice^a