C. elegans is capable of showing long-term memory for habituation training of the tap withdrawal response. As in other organisms this training is best produced by distributed or spaced training and lasts at least 24 hours. The neural circuit underlying the tap withdrawal response has been described and there is increasing evidence that the sensory neurons use glutamate as their neurotransmitter (Rankin and Wicks, 2000). Long-term habituation training for the current experiments consisted of groups of approximately 20 worms that received 4 blocks of 20 taps at a 60s interstimulus interval (ISI) with blocks separated by one-hour rest periods. Testing was conducted 24 hours after training with individual worms given one block of 10 taps also at a 60s ISI. Trained groups were compared to a control group that received a single tap on training day and the same individual testing on day two. The results showed that wild-type worms that received distributed training show significantly smaller responses compared to single-tap controls (p<0.01). This retention has been observed at 24, 48 and 72 hours following training.
eat-4 (
ky5) mutants also received distributed habituation training and although previous studies have shown that
eat-4 (
ky5) mutants display enhanced short-term habituation the
eat-4 (
ky5) mutant did not show long-term memory for the habituation training. In a follow-up study,
eat-4 (
ky5) mutants were given the same distributed habituation training protocol but using a stronger stimulus (trains of 7 taps/sec) rather than the single tap. This resulted in trained
eat-4 (
ky5) mutant worms showing decreased reversal magnitude on Day 2 compared to untrained
eat-4 (
ky5) controls. This suggests that stronger stimulation likely results in increased glutamate release from the sensory neuron in these mutants. The alpha-amino-3-hydroxy-5-methyl-4-isoxazole-proprionate (AMPA)-receptor subtype mutant
glr-1 (
n2461) is capable of short-term habituation, however, when long-term habituation training was administered to the
glr-1 (
n2461) mutant this resulted in no long-term retention of the training on Day 2. When
glr-1 (
n2461) worms were given the stronger stimulation (trains of taps) they still did not show long-term habituation. Thus the AMPA receptor does appear to be necessary for long-term memory regardless of level of stimulation. To further elucidate the role of glutamate in long-term memory wild-type worms were placed on plates streaked with either 10 mM 6,7-dinitrquinoxaline-2,3-dione (DNQX; a competitive AMPA-receptor blocker) or vehicle (35 mM sodium hydroxide; NaOH) during training on Day 1. Day 2 testing was conducted drug free. The vehicle-administered worms displayed long-term retention while the worm exposed to DNQX did not. This finding, coupled with the long-term memory deficits observed in the
glr-1 mutant strain, lead to the conclusion that activation of the glutamate receptor subtype AMPA is required for the formation of long-term retention. The role of the N-methyl-D-aspartate (NMDA) glutamate receptor in long-term habituation training was also investigated. Wild-type worms were trained with the non-competitive NMDA antagonists (5R,10S)-(+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]-cyclohepten-5,10-imine hydrogen maleate (MK801). When worms were trained in the presence of MK801 there was no evidence for long-term memory of habituation training. This supports what has been theorized in mammalian models of long-term synaptic plasticity in that activation of AMPA-type receptors may be required for NMDA-type receptor activation and therefore when either receptor is blocked, it effectively attenuates long-term memory. Research funded by the Natural Sciences and Engineering Research Council of Canada.