Noxious stimuli, such as electrical shocks to the animal's tail, enhance Aplysia's gill- and siphon-withdrawal reflex. Previous experimental work has indicated that this behavioral enhancement, known as dishabituation (if the reflex has been habituated) or sensitization (if it has not been habituated), might be mediated, at least in part, by the endogenous monoaminergic transmitter serotonin (5-HT). To assess 5-HT's role in dishabituation and sensitization of Aplysia withdrawal reflex, we treated Aplysia with the serotonergic neurotoxin 5,7-dihydroxytryptamine (5,7-DHT). We found that 5,7-DHT treatment significantly reduced the dishabituation of the withdrawal reflex produced by tail shock. Treatment with the neurotoxin also blocked the heterosynaptic facilitation of monosynaptic connections between siphon sensory neurons and their follower cells, which contributes to the behavioral enhancement. Analysis by high-performance liquid chromatography indicated that 5,7-DHT treatment significantly reduced 5-HT levels in the Aplysia CNS. Moreover, the neurotoxic effects of 5,7-DHT appeared to be relatively specific for serotonergic pathways. Thus, 5,7-DHT treatment did not disrupt the ability of nonserotonergic facilitatory interneurons, the L29 cells, to facilitate the connections of siphon sensory neurons. Also, 5,7-DHT reduced 5-HT-dependent, but not dopamine-dependent, histofluorescence in Aplysia central ganglia. Finally, 5,7-DHT does not reduce the levels of the facilitatory peptides SCPA and SCPB within the Aplysia CNS. Our results, together with those of Mackey et al. (1989), indicate that 5-HT plays a major role in mediating dishabituation and sensitization of Aplysia's withdrawal reflex.

Recent studies have shown that, in addition to being modulated by presynaptic facilitation, the sensory neurons of the gill-and siphon-withdrawal reflex ofAplysia are also capable of being modulated by transient presynaptic inhibition produced by the peptide Phe-Met-Arg-Phe-NH2. These two modulatory effects involve different second-messenger systems: the facilitation is mediated through cAMP-dependent protein phosphorylation, and the inhibition is mediated through the lipoxygenase pathway of arachidonic acid. To explore the behavioral function of this inhibition, we have carried out a parametric analysis of the effect of tail shock on the siphon-withdrawal reflex. In addition to producing sensitization of the withdrawal reflex, tail shock also transiently inhibits the reflex. The inhibition is produced by relatively weak shock, whereas sensitization is more prominent and may mask the inhibition with stronger shock. Furthermore, inhibition is not observed after habituation training. Cellular studies suggest that the behavioral inhibition is mediated, at least in part, by presynaptic inhibition of transmitter release from the siphon sensory neurons. Moreover, we have identified an interneuron within the left pleural ganglion (LPL16) that shows Phe-Met-Arg-Phe-NH2 immunoreactivity, is activated by tail shock, and simulates the presynaptic inhibitory actions produced by tail shock. Therefore, our results suggest that presynaptic inhibition mediated by Phe-Met-Arg-Phe-NH2 and its lipoxygenase second messenger contributes to behavioral inhibition of the siphon-withdrawal reflex.Modern studies of the properties of stimuli that serve as reinforcers for sensitization and classical conditioning in vertebrate learning reveal that these stimuli usually have two components, a prominent facilitatory component and a lessobvious inhibitory component. The facilitatory component is important for sensitization and conventional classical conditioning, whereas the inhibitory component is important for conditioned inhibition (1, 2). These dual properties of unconditioned stimuli have also been studied recently in invertebrates (6,14,(20)(21)(22)(23)(24)(25)(26)35). We describe here the existence of an inhibitory component of the unconditioned stimulus for learning in Aplysia and show that this component appears to use Phe-Met-Arg-Phe-NH2 as one of its transmitters.

Several lines of evidence suggest that 5-HT plays a significant role in presynaptic facilitation of the siphon sensory cells contributing to dishabituation and sensitization of the gill- and siphon-withdrawal reflex in Aplysia. Most recently, Glanzman et al. (1989) found that treatment with the 5-HT neurotoxin, 5,7-DHT markedly reduced both synaptic facilitation and behavioral dishabituation. To provide more direct evidence for a role of 5-HT, we have attempted to identify individual serotonergic facilitator neurons. Hawkins (1989) used histological techniques to locate several serotonergic neurons in the ring ganglia that send axons to the abdominal ganglion and are therefore possible serotonergic facilitators. These include one neuron in the B cluster of each cerebral ganglion, which we have identified electrophysiologically and named the CB1 cells. Both glyoxylic acid histofluorescence and 5-HT immunofluorescence indicate that the CB1 neurons are serotonergic. In a semiintact preparation, the CB1 neurons respond to cutaneous stimulation which produces dishabituation and sensitization (such as tail shock) with an increase in firing, which may outlast the stimulation by 15 min. Intracellular stimulation of a CB1 neuron in a manner similar to its response to tail shock produces facilitation of the EPSPs from siphon sensory neurons to motor neurons, as well as broadening of the action potential in the sensory neurons in tetraethylammonium solution. These results strongly suggest that the identified serotonergic CB1 neurons participate in mediating presynaptic facilitation contributing to dishabituation and sensitization of the gill- and siphon-withdrawal reflex in Aplysia.

Locomotion in Aplysia is mediated by a central program. We have found that this program is triggered by the biogenic amine, serotonin (5HT), and modulated by an endogenous source of neuropeptides, bag cell extract. In examining the triggering of locomotion, we found that injecting 5-HT into the bloodstream of quiescent animals elicited locomotion in both intact, freely moving animals and split-foot preparations. Increasing the concentration of 5-HT from lop7 to lop3 M produced both a monotonic increase in the number of steps elicited over a 5-min period after injection and a monotonic decrease in the first interstep interval. The threshold for triggering locomotion was lop7 M in split-foot preparations. Application of a saturated salt solution to the tail-a natural trigger for escape locomotionfacilitated 5-HT locomotion. The effects of 5-HT on locomotion were relatively specific since they were not mimicked by three other biogenic amines known to exist in Aplysia (dopamine, octopamine, and histamine). 5-HT could still trigger locomotion after isolating the pedal and pleural ganglia from the rest of the central nervous system, suggesting that a primary site for the action of 5-HT on locomotion lies in the pedal and pleural ganglia. In examining the modulation of locomotion, we found that injecting bag cell extract, an endogenous source of neuropeptides in Aplysia which elicits egg-laying behavior, reversibly suppressed 5-HTtriggered locomotion. As in the case with serotonin, bag cell extract suppression of locomotion was still observed in animals in which the pleural and pedal ganglia were isolated from the rest of the central nervous system. The suppression of locomotion was specific to bag cell extract since no suppression was produced by extracts of other central nervous tissue (cerebral ganglia). Moreover, the suppressive effect of bag cell extract was abolished by prior treatment of the extract with pronase, suggesting that the active suppressive agent is one or more of the peptides specific to the bag cells. Thus, locomotion in Aplysia provides a useful model system for examining both the triggering and the modulation of a fixed action pattern.