Block of theta-burst-induced long-term potentiation by (1S,3S)-1-aminocyclopentane-1,3-dicarboxylic acid: further evidence against long-term potentiation as a model for learning.

It has been previously reported that block of high-frequency stimulation-induced long-term potentiation of synaptic transmission in the hippocampus does not necessarily lead to impairment of spatial learning. Here we show that (1S,3S)-1-aminocyclopentane-1,3-dicarboxylic acid, an agonist at group II metabotropic glutamate receptors, completely blocked long-term potentiation induced by a theta-burst type of stimulation protocol (five pulses at 75 Hz per train, 200 ms inter-train interval) in the CA1 region in vivo. The drug did not significantly affect synaptic responses during each train whereas inter-train facilitation of excitatory postsynaptic potentials was slightly reduced. It also produced a large reduction in paired-pulse facilitation (50 ms inter-stimulus interval), possibly indicating that an increase in inhibition might be involved in the block of long-term potentiation. The drug dose used (5 microliters of a 10 mM solution i.c.v.) was half the dose which inhibited high-frequency stimulation-induced long-term potentiation in earlier experiments but which did not prevent learning of spatial tasks. We conclude that long-term potentiation induced by a more physiological stimulation protocol which uses theta-like inter-train intervals does not appear to accurately model the synaptic changes which are believed to occur during learning either.