Long-term synaptic facilitation in the absence of short-term facilitation in Aplysia neurons.

Serotonin (5-HT) induces both short-term and long-term facilitation of the identified synaptic connections between sensory and motor neurons of Aplysia. Three independent experimental approaches showed that long-term facilitation can normally be expressed in the absence of short-term facilitation: (i) The 5-HT antagonist cyproheptadine blocked the induction of short-term but not long-term facilitation; (ii) concentrations of 5-HT below threshold for the induction of short-term facilitation nonetheless induced long-term facilitation; and (iii) localized application of 5-HT to the sensory neuron cell body and proximal synapses induced long-term facilitation in distal synapses that were not exposed to 5-HT and had not expressed short-term facilitation. These results suggest that short-term and long-term synaptic facilitation are induced in parallel in the sensory neurons and that the short-term process, because it is induced and expressed at the synapse, can occur locally, but the long-term process, because of its dependence on a nuclear signal, is expressed throughout the neuron.

Coincident induction of long-term facilitation in Aplysia: cooperativity between cell bodies and remote synapses.

Induction of long-term synaptic changes at one synapse can facilitate the induction of long-term plasticity at another synapse. Evidence is presented here that if Aplysia sensory neuron somata and their remote motor neuron synapses are simultaneously exposed to serotonin pulses insufficient to induce long-term facilitation (LTF) at either site alone, processes activated at these sites interact to induce LTF. This coincident induction of LTF requires that (i) the synaptic pulse occur within a brief temporal window of the somatic pulse, and (ii) local protein synthesis occur immediately at the synapse, followed by delayed protein synthesis at the soma.

Acquisition and retention of long-term habituation in Aplysia: correlation of behavioral and cellular processes.

To examine the cellular mechanisms responsible for transition from a short-term to a long-term behavioral modification, a rapid training procedure was developed for producing long-term habituation of the defensive withdrawal of gill and siphon in Aplysia. Four ten-trial training sessions, with 1(1/2)-hour intersession intervals, produced habituation that was retained for more than 1 week. This 5-hour procedure could be applied to a test system in the isolated abdominal ganglion where the cellular changes accompanying the acquisition of long-term habituation can be examined. During acquisition, intracellular recordings were obtained from L7, a major gill and siphon motor neuron, and the pattern of stimulation used in the behavioral experiments was applied to an afferent nerve. Acquisition was associated with a progressive decrease in the complex excitatory synaptic potential produced in L7 by afferent nerve stimulation. When retention was tested 24 hours later, the synaptic decrement was still evident. Thus, a behaviorally meaningful stimulus sequence, consisting of only 40 patterned stimuli, leads to changes in synaptic effectiveness lasting one or more days in a neural pathway involved in short-term habituation of this reflex.

Two functional effects of decreased conductance EPSP's: synaptic augmentation and increased electrotonic coupling.

Three electronically coupled motor neurons, which mediate inking behavior in Aplysia californica, receive both increased and decreased conductance excitatory postsynaptic potentials (EPSP's). The increased conductance EPSP's reduce electrical coupling among the cells, whereas the decreased conductance EPSP's increase electrical coupling. The decreased conductance EPSP's also augment the action of a previously ineffective sensory input and this augmentation is enhanced by the increase in electrical coupling. Both effects combine to trigger a stereotypic behavioral response.

Associative Learning in Aplysia: evidence for conditioned fear in an invertebrate.

Aversive classical conditioning of Aplysia californica, a gastropod mollusk suited for neurobiological study, produces a learned reaction to the chemosensory conditioned stimulus that is expressed as a marked facilitation of four defensive responses: two graded reflexes (head and siphon withdrawal), an all-or-none fixed act (inking), and a complex fixed action pattern (escape locomotion). In addition, the conditioned stimulus produces a concomitant depression of at least one appetitive response, feeding. These extensive and selective actions of the conditioned stimulus in Aplysia resemble the actions of conditioned fear stimuli in higher mammals and suggest that the functional equivalent of fear occurs in invertebrates and thus may be an adaptive mechanism that is widespread in the animal kingdom.

Pharmacological dissociation of modulatory effects of serotonin in Aplysia sensory neurons.

In the mollusk Aplysia the neurotransmitter serotonin (5HT) has a fundamental modulatory role in several forms of learning and memory that involve an increase in the efficacy of synaptic transmission between tail sensory neurons (SNs) and motor neurons. The classical 5HT antagonist cyproheptadine (CYP) permits dissociation of three forms of serotonergic modulation in these SNs: (i) CYP reversibly blocks spike-broadening induced either by exogenous application of 5HT or by sensitizing stimulation of a tail nerve. (ii) CYP does not block 5HT-induced or tail input-induced increases in SN somatic excitability. (iii) Concomitant with its block of spike-broadening, CYP reversibly blocks 5HT-induced facilitation of synaptic transmission from SNs. These results suggest that endogenously released 5HT may act at different receptor subtypes that are coupled to different forms of neuromodulation in tail SNs of Aplysia.

Differential classical conditioning of a defensive withdrawal reflex in Aplysia californica.

The defensive siphon and gill withdrawal reflex of Aplysia is a simple reflex mediated by a well-defined neural circuit. This reflex exhibits classical conditioning when a weak tactile stimulus to the siphon is used as a conditioned stimulus and a strong shock to the tail is used as an unconditioned stimulus. The siphon withdrawal component of this reflex can be differentially conditioned when stimuli applied to two different sites on the mantle skin (the mantle shelf and the siphon) are used as discriminative stimuli. The differential conditioning can be acquired in a single trial, is retained for more than 24 hours, and increases in strength with increased trials. Differential conditioning can also be produced within the field of innervation of a single cluster of sensory neurons (the LE cluster) since two separate sites on the siphon skin can serve as discriminative stimuli. The finding that two independent afferent inputs that activate a common set of interneurons and motor neurons can be differentially conditioned restricts the possible cellular loci involved in the associative learning.

Associative learning in Aplysia: cellular correlates supporting a conditioned fear hypothesis.

Aversive associative learning in Aplysia california survives restraint of the animal and surgical exposure of the central nervous system. The learning is expressed in the intracellularly recorded activity of identified motor neurons mediating three different defensive behaviors: escape locomotion, inking, and siphon withdrawal. In each case, animals that had previously received paired training showed significant facilitation of synaptic input to motor neurons during test stimulation in the presence of the conditioned stimulus. Animals without such training showed no facilitation of input to the motor neurons. Resting potential and input resistance appeared unaffected by conditioning and were not altered by application of the conditioned stimulus. These results show that the conditioned facilitation of defensive responses cannot be explained by subthreshold actions of the conditioned stimulus on the motor neurons and support the hypothesis that Aplysia learn to associate the conditioned stimulus with a fearlike central state.

Cellular analysis of long-term habituation of the gill-withdrawal reflex of Aplysia californica.

Long-term habituation training in Aplysia californica produces a profound depression in the efficacy of synaptic transmission between mechanoreceptor neurons and gill motor neurons. This depression persists for more than 3 weeks. Thus a critical synaptic site for plasticity underlying long-term habituation is the same as that for short-term habituation. For this simple form of learning, short- and long-term memory share a common locus and aspects of a common mechanism: synaptic depression.

Long-term habituation of a defensive withdrawal reflex in aplysia.

A tactile stimulus to the siphon of Aplysia produces a defensive withdrawal reflex consisting of contraction of the siphon, the gill, and the mantle shelf. We studied long-term habituation of this reflex using two types of preparations, one focusing on the siphon component and the other on the gill component of the reflex. Siphon withdrawal, studied in unrestrained animals, showed marked habituation within a single ten-trial training session. Five daily training sessions produced habituaton that built up across days and lasted for at least 3 weeks. Furthermore, spaced training produced significantly longer lasting habituation than massed training. Gill withdrawal, studied in a restrained animal, also showed long-term retention of habituation. Since the neural circuitry of gill withdrawal is relatively well understood, it may be possible to study the cellular mechanisms underlying a long-term behavioral modification.

Lack of coincidence between neural and behavioral manifestations of cortical spreading depression.

The presence of cortical spreading depression is typically inferred from the presence of hypesthesia. The electrocorticogram and slow-potential change were recorded during cortical spreading depression and it was found that hypesthesia remained long after the cortex recovered from neural depression. Hypesthesia, therefore, is an unreliable indicant of cortical spreading depression; if cortical spreading depression is used as a research tool, neural activity must be monitored. These data offer a special problem for memory transfer studies.

A cellular mechanism of classical conditioning in Aplysia: activity-dependent amplification of presynaptic facilitation.

A training procedure analogous to differential classical conditioning produces differential facilitation of excitatory postsynaptic potentials (EPSP's) in the neuronal circuit for the siphon withdrawal reflex in Aplysia. Thus, tail shock (the unconditioned stimulus) produces greater facilitation of the monosynaptic EPSP from a siphon sensory neuron to a siphon motor neuron if the shock is preceded by spike activity in the sensory neuron than if the shock and spike activity occur in a specifically unpaired pattern or if the shock occurs alone. Further experiments indicate that this activity-dependent amplification of facilitation is presynaptic in origin and involves a differential increase in spike duration and thus in Ca2+ influx in paired versus unpaired sensory neurons. The results of these cellular experiments are quantitatively similar to the results of behavioral experiments with the same protocol and parameters, suggesting that activity-dependent amplification of presynaptic facilitation may make a significant contribution to classical conditioning of the withdrawal reflex.

Behavioral dissociation of dishabituation, sensitization, and inhibition in Aplysia.

Three forms of nonassociative learning (habituation, dishabituation, and sensitization) have commonly been explained by a dual-process view in which a single decrementing process produces habituation and a single facilitatory process produces both dishabituation and sensitization. A key prediction of this view is that dishabituation and sensitization should always occur together. However, we show that dishabituation and sensitization, as well as an additional process, inhibition, can be behaviorally dissociated in Aplysia by (i) their differential time of onset, (ii) their differential sensitivity to stimulus intensity, and (iii) their differential emergence during development. A simple dual-process view cannot explain these results; rather, a multiprocess view appears necessary to account for nonassociative learning in Aplysia.

Long-term sensitization of a defensive withdrawal reflex in Aplysia.

When a weak tactile stimulus is applied to the siphon of Aplysia californica, the animal withdraws the siphon between the parapodia. This defensive withdrawal reflex can be facilitated (sensitized) if the animal is previously given 4 days of training, consisting of four brief noxious stimuli each day. The sensitization of this reflex can last for up to 3 weeks after training and is mediated by the abdominal ganglion which also mediates long-term habituation. This preparation may provide a system for analyzing the neural mechanism of long-term behavioral modifications of complexity which is intermediate between habituation and associative learning.

Serotonin induces temporally and mechanistically distinct phases of persistent PKA activity in Aplysia sensory neurons.

The cAMP signaling cascade has been implicated in several stages of memory formation. We have examined activation of this cascade by serotonin (5-HT) in the sensory neurons of Aplysia. We find that different patterns of 5-HT exposure induce three distinct modes of PKA activation. First, a single 5 min pulse induces transient (5 min) PKA activation that requires neither transcription nor translation. Second, 4-5 pulses induce intermediate-term persistent activation (3 hr duration) that requires translation but not transcription. Third, 5 pulses of 5-HT, as well as continuous (90 min) exposure, induce long-term persistent activation 20 hr later, which requires both transcription and translation. Thus, in the sensory neurons, different patterns of 5-HT give rise to three independent phases of PKA activation that differ in their induction requirements, their temporal profiles, and their molecular mechanisms.

Parallel molecular pathways mediate expression of distinct forms of intermediate-term facilitation at tail sensory-motor synapses in Aplysia.

Three distinct temporal phases of synaptic facilitation (short-, intermediate-, and long-term) are induced by serotonin (5-HT) at sensory (SN) to motor (MN) synapses in Aplysia. Here, we characterize two mechanistically distinct forms of intermediate-term facilitation (ITF) at tail SN-MN synapses. One form, activity-independent ITF, is produced by five spaced pulses of 5-HT in the absence of SN activity. Its induction requires protein synthesis, and its expression requires persistent activation of PKA but not PKC. The other form, activity-dependent ITF, is produced by a single pulse of 5-HT coincident with SN activation. Its induction does not require protein synthesis, and its expression requires persistent activation of PKC but not PKA. These results demonstrate that SN-MN synapses can exhibit two distinct forms of ITF that are mediated by parallel molecular pathways.

Molecular mechanisms underlying a unique intermediate phase of memory in aplysia.

Short- and long-term synaptic facilitation induced by serotonin at Aplysia sensory-motor (SN-MN) synapses has been widely used as a cellular model of short- and long-term memory for sensitization. In recent years, a distinct intermediate phase of synaptic facilitation (ITF) has been described at SN-MN synapses. Here, we identify a novel intermediate phase of behavioral memory (ITM) for sensitization in Aplysia and demonstrate that it shares the temporal and mechanistic features of ITF in the intact CNS: (1) it declines completely prior to the onset of LTM, (2) its induction requires protein but not RNA synthesis, and (3) its expression requires the persistent activation of protein kinase A. Thus, in Aplysia, the same temporal and molecular characteristics that distinguish ITF from other phases of synaptic plasticity distinguish ITM from other phases of behavioral memory.

Development assembly of learning in Aplysia.

Development can provide a powerful analytic approach for distinguishing and analysing specific behavioral, cellular and molecular processes as they emerge during ontogeny. Recently, such a developmental strategy has been used to investigate the functional assembly of different forms of non-associative learning (habituation, dishabituation and sensitization) in the marine mollusc Aplysia. This analysis has shown that different forms of learning, as well as their cellular analogs at central synapses, emerge according to very different developmental timetables. Subsequent behavioral studies in adult Aplysia showed that these same forms of learning were also clearly dissociable in the mature animal. These results, taken with earlier studies, suggest that a commonly held 'dual-process' view of non-associative learning, which attempts to account for all forms of non-associative learning as the interaction of only two processes (one decremental and one incremental) requires revision, and that a multi-process view, which includes the possibility of inhibitory as well as facilitatory interactions, is required to account adequately for all of the behavioral features of nonassociative learning.

Multiple overlapping processes underlying short-term synaptic enhancement.

Recently there have been exciting advances in understanding the mechanisms and functional roles of a form of short-term synaptic enhancement (STE) that results from an activity-dependent accumulation of Ca2+ in the presynaptic terminal. This form of STE is composed of at least four processes: fast-decaying facilitation (FI), slow-decaying facilitation (F2), augmentation (AUG) and post-tetanic potentiation (PTP). Recent results suggest that these processes can now be distinguished mechanistically by the site of their induction within the presynaptic terminal: FI and F2 appear to be induced by a rapid, high concentration of Ca2+ at or near the site of exocytosis, whereas AUG and PTP seem to be induced by lower levels of Ca2+ with slower kinetics, possibly within the core of the terminal. STE is highly conserved across diverse species, and appears to serve as a flexible mechanism for temporal information processing in systems ranging from peripheral motor control to higher cortical integration.

Contribution of postsynaptic Ca2+ to the induction of post-tetanic potentiation in the neural circuit for siphon withdrawal in Aplysia.

Recent studies in Aplysia have revealed a novel postsynaptic Ca(2+) component to posttetanic potentiation (PTP) at the siphon sensory to motor neuron (SN-MN) synapse. Here we asked whether the postsynaptic Ca(2+) component of PTP was a special feature of the SN-MN synapse, and if so, whether it reflected a unique property of the SN or the MN. We examined whether postsynaptic injection of BAPTA reduced PTP at SN synapses onto different postsynaptic targets by comparing PTP at SN-MN and SN-interneuron (L29) synapses. We also examined PTP at L29-MN synapses. Postsynaptic BAPTA reduced PTP only at the SN-MN synapse; it did not affect PTP at either the SN-L29 or the L29-MN synapse, indicating that the SN and the MN do not require postsynaptic Ca(2+) for PTP with all other synaptic partners. The postsynaptic Ca(2+) component of PTP is present at other Aplysia SN-MN synapses; tail SN-MN synapses also showed reduced PTP when the MN was injected with BAPTA. Surprisingly, in both tail and siphon SN-MN synapses, there was an inverse relationship between the initial size of the EPSP and the postsynaptic component to PTP; only the initially weak SN-MN synapses showed a BAPTA-sensitive component. Homosynaptic depression of initially strong SN-MN synapses into the size range of initially weak synapses did not confer postsynaptic Ca(2+) sensitivity to PTP. Finally, the postsynaptic Ca(2+) component of PTP could be induced in the presence of APV, indicating that it is not mediated by NMDA receptors. These results suggest a dual model for PTP at the SN-MN synapse, in which a postsynaptic Ca(2+) contribution summates with the conventional presynaptic mechanisms to yield an enhanced form of PTP.