Long-term potentiation is impaired in rat hippocampal slices that produce spontaneous sharp waves

Abstract

Sharp waves (SPWs) occur in the hippocampal EEG during behaviours such as alert immobility and slow-wave sleep. Despite their widespread occurrence across brain regions and mammalian species, the functional importance of SPWs remains unknown. Experiments in the present study indicate that long-term potentiation (LTP) is significantly impaired in slices, prepared from the temporal aspect of rat hippocampus, that spontaneously generate SPW activity. This was probably not due to anatomical and/or biochemical abnormalities in temporal slices because stable LTP was uncovered in field CA1 when SPWs were eliminated by severing the projection from CA3. The same procedure did not alter LTP in slices lacking SPWs. Robust and stable LTP was obtained in the presence of SPWs in slices treated with an adenosine A1 receptor antagonist, a finding that links the present results to mechanisms related to the LTP reversal effect. In accord with this, single stimulation pulses delivered intermittently in a manner similar to the SPW pattern interfered with LTP to a similar degree as spontaneous SPWs. Taken together, these results suggest the possibility that SPWs in the hippocampus constitute a neural mechanism for forgetting.

Figure 1. Properties of spontaneously occurring SPWs in temporal slices from rat hippocampus

A, SPWs are eliminated by infusion of the AMPA receptor antagonist CNQX. Representative traces from CA3 stratum pyramidale prior to (above) and at the end of a 30 min CNQX wash-in (below) are shown. Calibration: 250 ms, 200 μV. B, simultaneous extracellular and intracellular recordings from CA3 stratum pyramidale during SPW activity. Over 800 events were averaged to produce these records (standard error indicated by dashed line), and the traces were normalized to peak amplitude for illustration purposes. Individual records are also shown (inset: scale bars for extracellular trace (above) represent 100 ms and 200 μV; scale bars for intracellular record (below) represent 100 ms and 10 pA). C, representative example of a single SPW recorded in CA3b stratum pyramidale. Small amplitude, high frequency activity is apparent, particularly on the ascending phase. Calibration: 30 ms, 50 μV. D, SPWs were accompanied by high frequency (150–200 Hz) ripple oscillations. 235 SPW/ripple complexes recorded from CA3b stratum pyramidale were averaged and bandpass filtered from 0.3 to 100 Hz to isolate the SPW (grey) or bandpass filtered from 150 to 300 Hz to isolate the ripple (black). Errors are shown as dashed lines above and below the traces. Calibration: 20 ms, 10 μV. A segment of the ripple record (indicated by bar) is enlarged and shown (inset). Scale bars (inset) represent 5 ms and 2.5 μV.

Figure 2. LTP was significantly decreased in slices from temporal hippocampus with spontaneously occurring SPWs

A, spontaneous field recording in CA3 stratum radiatum of slices prepared from temporal hippocampus (above) or septal hippocampus (below). SPW activity was only observed in the former case. Note that the phase of in vitro SPWs was negative-going in stratum radiatum, in contrast to the positive-going waves recorded in stratum pyramidale (see Fig. 1). Scale bars represent 500 ms and 0.2 mV. B, control and potentiated responses in CA3 stratum radiatum of a SPW-producing temporal slice (above) and a septal slice exhibiting no SPW activity (below). LTP decayed to near baseline in the presence of SPWs. Calibration: 5 ms, 0.4 mV. C and D, summary of all data for slope (C) and amplitude (D) measures across time. Slope and amplitude were normalized to 10 min baseline values. Theta bursts were delivered at the time indicated with an arrow.

Figure 3. Elimination of SPWs in CA1 of temporal hippocampal slices by cutting the projection from CA3 allowed for induction of lasting LTP

A, representative field recordings from CA1 stratum radiatum of temporal hippocampal slices. The top trace depicts an example of SPW activity in a slice with an intact connection between CA3 and CA1, while the bottom trace shows a typical case with no SPW activity following surgical isolation of CA1 from CA3. Scale bars represent 500 ms and 0.2 mV. B, representative EPSPs in CA1 stratum radiatum during baseline and after the delivery of theta burst triplets in a slice with SPWs (above) and another slice in which SPWs were eliminated by removing afferent inputs from CA3 (below). Calibration: 5 ms, 0.5 mV, C and D, group data for normalized slope (C) and amplitude (D) measures in CA1. E, summary of data for SPW-deficient septal slices in which a cut was placed between CA3 and CA1 prior to induction of LTP. Note that LTP does not appear to be affected by the procedure.

Figure 4. DPCPX, an antagonist of the A1-type adenosine receptor, blocked the impairment of LTP in SPW-producing slices

A, summary of LTP data for slices from temporal hippocampus pre-treated with DPCPX. B, example control and potentiated EPSPs recorded from CA3 stratum radiatum of slices infused with the adenosine antagonist DPCPX. Calibration bars represent 5 ms and 0.5 mV. C, LTP was not impaired in DPCPX-treated slices when stimulation current was lowered to counteract DPCPX-induced increases in baseline response amplitudes. D, spontaneous field recordings from CA3 stratum radiatum show that DPCPX had little effect on ongoing SPW activity. Calibration: 500 ms, 0.2 mV. E, peak frequency and power within the 0.1–7 Hz frequency band may have been slightly increased by DPCPX infusion, but the effects were highly variable and not significant.

Figure 5. Stimulation pulses modelled after spontaneous SPW activity had a detrimental effect on LTP

In these experiments, the same theta burst triplet protocol was employed to induce LTP except that irregularly timed stimulation pulses (i.e. SPW pulse stimulation; SPWPS) with a mean interpulse interval of 385 ms were delivered throughout the 60 second intervals that separated the four theta burst triplets. Up to 3 additional SPWPS episodes were delivered in the 5 min following theta burst stimulation. LTP was completely reversed in 3 out of 7 cases; the data shown here represent the average values across all 7 slices.