BIDIRECTIONAL LONG-TERM MODIFICATION OF SYNAPTIC EFFECTIVENESS IN THEADULT AND IMMATURE HIPPOCAMPUS

Previously we showed that delivering 900 pulses to the Schaffer collat eral-CA1 pathway at 1-3 Hz causes a lasting depression of synaptic eff ectiveness that is input specific and dependent on NMDA receptor activ ation (Dudek and Bear, 1992a). Here we describe experiments aimed at f urther characterizing this homosynaptic long-term depression (LTD) and comparing it with long-term potentiation (LTP). To address the questi on of whether depressed synapses can still be potentiated and vice ver sa, LTP was saturated with repeated high-frequency tetani, and then LT D was induced with low-frequency stimulation (LFS). A second strong te tanus then restored the potentiation, indicating that the same synapse s whose transmission had been depressed by LFS were capable of subsequ ently supporting potentiation. In a complementary experiment, LTD was induced first and then a strong high-frequency tetanus was delivered. We found that the resulting LTP achieved the same absolute magnitude a s that observed in control slices that had received the high-frequency stimulation alone. Next, the postnatal development of LTD was investi gated in slices prepared from rats at 6-35 d of age. The consequences of LFS were far more pronounced in slices from young rats. LTD followi ng 900 pulses at 1 Hz measured -45 +/- 4% in CA1 of rats less than 2 w eeks old as compared with -20 +/- 4 in animals at 5 weeks postnatal. I t was also found that LTD precedes the developmental onset of LTP in C A1. Finally, we addressed the question of whether LTD could be saturat ed by repeated episodes of LFS in slices prepared from 3-week-old rats . It was observed that a floor to the LTD effect is reached after -180 0 pulses and measures approximately -50%. From these data, we estimate that the dynamic range over which the population of Schaffer collater al synapses can be modified from a naive state under our experimental conditions is roughly +/- 50%.