Induction Of Long-term Facilitation Research Articles

Long-Lasting Adaptations of the NR2B-Containing NMDA Receptors in the Dorsomedial Striatum Play a Crucial Role in Alcohol Consumption and Relapse

A growing number of studies suggest that the development of compulsive drug seeking and taking depends on dorsostriatal mechanisms. We previously observed that ex vivo acute exposure of the dorsal striatum to, and withdrawal from, alcohol induces long-term facilitation (LTF) of the activity of NR2B-containing NMDA receptors (NR2B-NMDARs) in a mechanism that requires the Src family protein tyrosine kinase (PTK), Fyn (Wang et al., 2007). In the present study, we first compared alcohol's actions in rat dorsomedial (DMS) and the dorsolateral (DLS) subregions of the striatum, which differ in their anatomical connectivity and function. We found that alcohol-mediated induction of LTF of NR2B-NMDAR activity is centered in the DMS. Next, we tested whether in vivo exposure of rats to alcohol leads to long-term adaptations of the NMDAR system in the DMS. We observed that repeated daily administration of alcohol results in a long-lasting increase in the activity of the NR2B-NMDARs in the DMS. The same procedure leads to a prolonged activation of Fyn, increased NR2B phosphorylation, and membrane localization of the subunit. Importantly, similar electrophysiological and biochemical modifications were observed in the DMS of rats that consumed large quantities of alcohol. Finally, we show that inhibition of NR2B-NMDARs or Src family PTKs in the DMS, but not in the DLS, significantly decreases operant self-administration of alcohol and reduces alcohol-priming-induced reinstatement of alcohol seeking. Our results suggest that the upregulation of NR2B-NMDAR activity within the DMS by alcohol contributes to the maladaptive synaptic changes that lead to excessive alcohol intake and relapse.

A New Component in Synaptic Plasticity: Upregulation of Kinesin in the Neurons of the Gill-Withdrawal Reflex

To explore how gene products, required for the initiation of synaptic growth, move from the cell body of the sensory neuron to its presynaptic terminals, and from the cell body of the motor neuron to its postsynaptic dendritic spines, we have investigated the anterograde transport machinery in both the sensory and motor neurons of the gill-withdrawal reflex of Aplysia. We found that the induction of long-term facilitation (LTF) by repeated applications of serotonin, a modulatory transmitter released during learning in Aplysia, requires upregulation of kinesin heavy chain (KHC) in both pre- and postsynaptic neurons. Indeed, upregulation of KHC in the presynaptic neurons alone is sufficient for the induction of LTF. However, KHC is not required for the persistence of LTF. Thus, in addition to transcriptional activation in the nucleus and local protein synthesis at the synapse, our studies have identified a third component critical for long-term learning-related plasticity: the coordinated upregulation of kinesin-mediated transport.

Serotonin Induces Memory-Like, Rapamycin-Sensitive Hyperexcitability in Sensory Axons ofAplysiaThat Contributes to Injury Responses

The induction of long-term facilitation (LTF) of synapses of Aplysia sensory neurons (SNs) by serotonin (5-HT) has provided an important mechanistic model of memory, but little is known about other long-term effects of 5-HT on sensory properties. Here we show that crushing peripheral nerves results in long-term hyperexcitability (LTH) of the axons of these nociceptive SNs that requires 5-HT activity in the injured nerve. Serotonin application to a nerve segment induces local axonal (but not somal) LTH that is inhibited by 5-HT-receptor antagonists. Blockade of crush-induced axonal LTH by an antagonist, methiothepin, provides evidence for mediation of this injury response by 5-HT. This is the first demonstration in any axon of neuromodulator-induced LTH, a phenomenon potentially important for long-lasting pain. Methiothepin does not reduce axonal LTH induced by local depolarization, so 5-HT is not required for all forms of axonal LTH. Serotonin-induced axonal LTH is expressed as reduced spike threshold and increased repetitive firing, whereas depolarization-induced LTH involves only reduced threshold. Like crush- and depolarization-induced LTH, 5-HT-induced LTH is blocked by inhibiting protein synthesis. Blockade by rapamycin, which also blocks synaptic LTF, is interesting because the eukaryotic protein kinase that is the target of rapamycin (TOR) has a conserved role in promoting growth by stimulating translation of proteins required for translation. Rapamycin sensitivity suggests that localized increases in translation of proteins that promote axonal conduction and excitability at sites of nerve injury may be regulated by the same signals that increase translation of proteins that promote neuronal growth.

Essential Role of Somatic and Synaptic Protein Synthesis and Axonal Transport in Long-Term Synapse-Specific Facilitation at Distal Sensorimotor Connections inAplysia

To investigate further the cellular mechanisms underlying long-term facilitation (LTF) and long-term synapse-specific facilitation (LTSSF), we studied the role of axonal transport and somatic and synaptic protein synthesis at proximal and distal synapses of Aplysia siphon sensory neurons (SNs). The long soma-synapse distances (2.5 to 3 cm) of the SN distal synapses impose important temporal and mechanistic constraints on long-term facilitation and on intracellular signaling. Excitatory postsynaptic potentials (EPSPs) evoked by SNs in central and peripheral siphon motor neurons were used to assay LTF 24-30 h after various pharmacological treatments. Inhibition of protein synthesis via anisomycin application at either the SN soma or distal synapses blocked the induction of LTF and LTSSF normally produced by synaptic application of the facilitating transmitter serotonin (5-hydroxytryptamine). Further, disruption of axonal transport by application of nocodazole to the isolated siphon nerve completely blocked LTF at distal synapses. These results indicate an essential role for somatic and synaptic protein synthesis and active axonal transport in LTSSF at distal synapses, and raise intriguing questions for current synaptic marking/capture models of synapse specificity and LTF.

CAMP Acts on Exchange Protein Activated by cAMP/cAMP-Regulated Guanine Nucleotide Exchange Protein to Regulate Transmitter Release at the Crayfish Neuromuscular Junction

Glutamatergic synapses are highly modifiable, suiting them for key roles in processes such as learning and memory. At crayfish glutamatergic neuromuscular junctions, hyperpolarization and cyclic nucleotide-activated (HCN) ion channels mediate hormonal modulation of glutamatergic synapses and a form activity-dependent long-term facilitation (LTF) of synaptic transmission. Here, we show that a new target for cAMP, exchange protein activated by cAMP (Epac) or cAMP-regulated guanine nucleotide exchange protein, is involved in the hormonal enhancement of synaptic transmission by serotonin. Induction of LTF "tags" synapses, rendering them responsive to cAMP in an HCN-independent manner. Epac also mediates the enhancement of tagged synapses. Thus, the cAMP-dependent enhancement of transmission is mediated by two separate pathways, neither of which involves protein kinase A.

Coregulation of glutamate uptake and long-term sensitization in Aplysia.

In Aplysia, long-term facilitation (LTF) at sensorimotor synapses of the pleural-pedal ganglia is mediated by an increase in the release of a neurotransmitter, which appears to be glutamate. Glutamate uptake also is increased in sensory neurons 24 hr after the induction of long-term sensitization (Levenson et al., 2000b). The present study investigated whether the same signaling pathways were involved in the long-term increase in glutamate uptake as in the induction of LTF. Thus, roles for cAMP, PKA (cAMP-dependent protein kinase), MAPK (mitogen-activated protein kinase), and tyrosine kinase in the regulation of glutamate uptake were tested. We found that 5-HT increased cAMP and activated PKA in sensory neurons. Exposure of pleural-pedal ganglia to analogs of cAMP or forskolin increased glutamate uptake 24 hr after treatments. Inhibitors of PKA (KT5720), MAPK (U0126 and PD98059), and tyrosine kinase (genistein) blocked the long-term increase in glutamate uptake produced by 5-HT. In addition, bpV, a tyrosine phosphatase inhibitor, facilitated the ability of subthreshold levels of 5-HT to increase glutamate uptake. Inhibition of PKC, which is not involved in LTF, had no effect on the long-term increase in glutamate uptake produced by 5-HT. Furthermore, activation of PKC by phorbol-12,13-dibutyrate did not produce long-term changes in glutamate uptake. The results demonstrate that the same constellation of second messengers and kinases is involved in the long-term regulation of both glutamate release and glutamate uptake. These similarities in signaling pathways suggest that regulation of glutamate release and uptake during formation of long-term memory are coordinated through coregulation of these two processes.

Roles of Ca2+, hyperpolarization and cyclic nucleotide-activated channel activation, and actin in temporal synaptic tagging.

At crayfish neuromuscular junctions, cAMP increases transmitter released by action potentials by activating two effectors, hyperpolarization and cyclic nucleotide-activated channels (HCNCs) and a separate target that has been tentatively identified as exchange protein activated by cAMP (Epac). Intense electrical activity in the motor neuron induces a long-term facilitation (LTF) of transmitter release in which hyperpolarization from an electrogenic Na+-K+ exchanger activates HCNCs. The coupling of HCNCs to transmission involves actin. After LTF induction, cAMP further increases transmission in an HCNC-independent manner, activating the second target. This relaxation of the requirement for HCNC activation to enhance release is called temporal synaptic tagging. Tagging lasts at least 1 d but develops only in the 10 min period after electrical activity. The HCNCs are activated by the post-tetanic hyperpolarization occurring during this time. Both synaptic tagging and LTF induction depend on presynaptic Ca2+ accumulation during activity; both are blocked by EGTA-AM, and LTF is also prevented by stimulation in a low-[Ca2+] medium. Actin depolymerizers prevent induction of LTF and tagging, with little effect on HCNCs, whose sensitivity to cAMP and HCNC blockers is unaffected by tagging. Enhancement of actin polymerization can rescue tagging from HCNC block, suggesting that actin acts at a step after HCNC activation. These and other recent results suggest a model in which HCNC activation, followed by a process involving actin polymerization, acts cooperatively with [Ca2+] to induce tagging, after which only Epac activation is required for cAMP to further enhance transmission.

Coincident Induction of Long-Term Facilitation at Sensory-Motor Synapses in Aplysia: Presynaptic and Postsynaptic Factors

Induction of long-term synaptic changes at one synapse can facilitate the induction of long-term plasticity at another synapse. Here we show that if Aplysia sensory neuron (SN) somata and their remote motor neuron (MN) synapses are simultaneously exposed to serotonin (5HT) pulses, which at either site alone are insufficient to induce long-term facilitation (LTF), processes activated at these sites interact to induce LTF. Coincident induction of LTF requires: (1) that the synaptic pulse occurs within a brief temporal window of the somatic pulse and (2) that local protein synthesis occurs immediately at the synapse, followed by delayed protein synthesis at the soma. LTF at the SN-MN synapses can also be induced with cell-wide application of repeated pulses of 5HT. However, these two forms of LTF differ mechanistically: (1) coincident LTF requires protein synthesis in the postsynaptic motor neuron, whereas repeated 5HT LTF does not, and (2) repeated 5HT LTF is accompanied by intermediate-term (3 h) facilitation, whereas coincident LTF is not. Thus LTF expressed in the same temporal domain can result from different underlying mechanisms.

Changes in cat medullary neurone firing rates and synchrony following induction of respiratory long-term facilitation.

1. Long-term facilitation is a respiratory memory expressed as an increase in motor output lasting more than an hour. This change is induced by repeated hypoxia, stimulation of carotid chemoreceptors, or electrical stimulation of the carotid sinus nerve or brainstem mid-line. The present work addressed the hypothesis that persistent changes in medullary respiratory neural networks contribute to long-term facilitation. 2. Carotid chemoreceptors were stimulated by close arterial injection of CO(2)-saturated saline solution. Phrenic nerve efferent activity and up to 30 single medullary neurones were recorded simultaneously in nucleus tractus solitarii (NTS) including the dorsal respiratory group (DRG), Botzinger-ventral respiratory group (Böt-VRG), and nucleus raphe obscurus of nine adult cats, anaesthetized, injected with a neuromuscular blocking agent, vagotomized and artificially ventilated. 3. The firing rates of 87 of 105 neurones (83 %) changed following induction of long-term facilitation. Nine of eleven DRG and Böt-VRG putative premotor inspiratory neurones had increased firing rates with long-term facilitation. Fourteen of twenty-one raphe obscurus neurones with control firing rates less than 4 Hz had significant long-term increases in activity. 4. Cross-correlogram analysis suggested that there were changes in effective connectivity of neuron pairs with long-term facilitation. Joint peristimulus time histograms and pattern detection methods used with 'gravity' analysis also detected changes in short time scale correlations associated with long-term facilitation. 5. The results suggest that changes in firing rates and synchrony of VRG and DRG premotor neurones and altered effective connectivity among other functionally antecedent elements of the medullary respiratory network contribute to the expression of long-term facilitation.

Activation and degradation of the transcription factor C/EBP during long-term facilitation in Aplysia.

Long-term facilitation (LTF) of the sensory-to-motor synapses that mediate defensive reflexes in Aplysia requires induction of the transcription factor Aplysia CCAAT/enhancer binding protein (ApC/EBP) as an early response gene. We examined the time course of ApC/ EBP DNA binding during the induction of LTF: Binding activity was detected within 1 h of the sensitization treatment with serotonin, reached a maximum at 2 h, and decreased after 6 h. How are DNA binding and the turnover of ApC/EBP regulated? We find that phosphorylation of ApC/EBP by mitogen-activated protein (MAP) kinase is essential for binding. MAP kinase appears to be activated through protein kinase C. We also showed that ApC/EBP is degraded through the ubiquitin-proteasome pathway but that phosphorylation by MAP kinase renders it resistant to proteolysis. Thus, phosphorylation by MAP kinase is required for ApC/EBP to act as a transcription activator as well as to assure its stability early in the consolidation phase, when genes essential for the development of LTF begin to be expressed.

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.

Long-term facilitation of phrenic nerve activity in cats: responses and short time scale correlations of medullary neurones.

1. Stimulation of either peripheral chemoreceptors or nucleus raphe obscurus results in long-term facilitation of phrenic motoneurone activity. The first objective of this work was to measure the concurrent responses of neurones in the nucleus raphe obscurus, the nucleus tractus solitarii, and the regions of the retrofacial nucleus, nucleus ambiguus and nucleus retroambigualis during induction of long-term facilitation. A second goal was to assess functional relationships of the chemoresponsive raphe neurones with neurones in the other monitored locations and with phrenic motoneurones. 2. Up to thirty single medullary neurones and phrenic nerve efferent activity were recorded simultaneously in fifteen anaesthetized, paralysed, vagotomized, artificially ventilated adult cats. Carotid chemoreceptors were stimulated by close arterial injection of 200 microliters of CO2-saturated saline solution. Spike trains were analysed with cycle-triggered histograms and two statistical tests for respiratory modulation. Peristimulus-time histograms and cumulative sum histograms were used to assess responses to stimulation. Cross-correlation was used to test for non-random temporal relationships between spike trains. Spike-triggered average histograms provided evidence for functional associations with phrenic motoneurones. 3. One hundred and thirteen of 348 neurones were monitored in the nucleus raphe obscurus. The firing rates of twenty-nine raphe neurones increased during stimulation; eighteen decreased. In twenty-one pairs of concurrently monitored raphe neurones, the firing rate of one increased its activity during stimulation then decreased, while the other showed an increase that began as the rate of the former declined. Eighteen chemoresponsive raphe neurones had short time scale features in their phrenic spike-triggered averages. Short time scale features were found in cross-correlograms from 184 of 1407 neurone pairs. 4. The data suggest parallel routes by which carotid chemoreceptors influence medullary raphe neurones and support the hypotheses that mid-line respiratory-related neuronal assemblies transform information from those receptors and regulate the gain of respiratory motor output.

Dynamics of protein phosphorylation in sensory neurons of Aplysia

Protein phosphorylation plays important roles in the mechanisms underlying serotonin (5-HT)-induced presynaptic facilitation of Aplysia sensory neurons. To study mechanisms involved in facilitation, we investigated the pattern of protein phosphorylation in sensory neurons as a function of different durations of 5-HT. Two minutes and 1.5 hr treatments with 5-HT altered the phosphorylation of 5 and 10 proteins, respectively. These different duration treatments with 5-HT produced unique effects on the phosphorylation of different sets of proteins. This result suggests that cells may encode and measure the duration of a stimulus by the pattern of specific proteins that are phosphorylated or dephosphorylated. In addition, because the changes in phosphorylation produced by 2 min treatments with 5-HT were not observed after 25 min treatments with 5-HT, mechanisms must exist for the transient phosphorylation of some proteins even when the 5-HT treatment persists. Anisomycin, an inhibitor of protein synthesis, blocked the effect of 1.5 hr treatments with 5-HT on the phosphorylation of six proteins but had no effect on the phosphorylation change of four other proteins. Both CPT-cAMP (an activator of protein kinase A) and PDAc (an activator of protein kinase C) mimicked the effects of 5-HT on four proteins. Interestingly, the effect of 5-HT on these four proteins did not require protein synthesis. CPT-cAMP, but not PDAc, mimicked the effect of 5-HT on one protein (L55) and, the effect of 5-HT on this protein appeared to require protein synthesis. Because both activation of PKA and protein synthesis are involved in the induction of long-term facilitation, protein L55 is a good candidate for a protein that might play a key role in long-term facilitation. Finally, the effects of 5-HT on four proteins were not mimicked by either CPT-cAMP or PDAc. This finding raises the interesting possibility that some effects of 5-HT are mediated by second-messenger systems other than PKA or PKC.

Induction of long-term facilitation in Aplysia sensory neurons by local application of serotonin to remote synapses.

Long-term synaptic facilitation at the connections of Aplysia sensory neurons onto their target cells involves alterations in gene expression. How then are the relevant cellular signals for the induction and expression of long-term synaptic changes conveyed between the nucleus and remote synaptic terminals? We have explored this question using a set of remote, peripheral terminals of siphon sensory cells, which are approximately 3 cm from the sensory cell body in the abdominal ganglion. We found that these remote synapses, like the proximal synapses previously studied in dissociated cell culture, can exhibit long-term facilitation 24 hr after cell-wide serotonin application. Furthermore, serotonin applications restricted to the remote synaptic terminals nevertheless produced long-term facilitation, indicating that signals generated in synaptic regions can trigger the long-term process, perhaps via retrograde signals to the nucleus to modify gene expression, followed by anterograde signals back to the terminal. Serotonin applications restricted to the cell body and proximal synapses of the sensory neuron also produced long-term facilitation at remote synapses, although to a lesser extent, suggesting that long-term facilitation is expressed cell-wide, but that superimposed on this cell-wide facilitation there appears to be a component that is synapse-specific.

Presynaptic long-term facilitation at the crayfish neuromuscular junction: voltage-dependent and ion-dependent phases

Long-term facilitation (LTF) of synaptic transmission was investigated in the crayfish opener muscle to determine the factors necessary for its induction and expression. LTF was induced without action potentials by intracellular depolarization of presynaptic nerve terminals. Following induction, the synaptic transmission was enhanced by about 80% for a period of several hours. Intracellular recordings from pre- and postsynaptic cells, combined with ionic and pharmacological tests, permitted dissection of LTF into 2 phases: an initial tetanic phase that depended on the presence of both sodium and calcium ions and a subsequent long-lasting phase. This latter long-lasting enhancement of synaptic transmission was induced by repeated depolarizations of synaptic terminals but did not depend on the influx of sodium or calcium ions or on intracellular release of calcium ions. Both tetanic and long-lasting phases of LTF are attributable to activity of a single neuron, i.e., they are homosynaptic phenomena. Furthermore, LTF is associated with an increase of quantal release, whereas the size of quanta remains unchanged. During the long-lasting phase of LTF, the nerve terminal releases more transmitter for a given depolarization than before induction of LTF. Thus, the locus of LTF is presynaptic. Our findings suggest the presence of a voltage-dependent mechanism in the presynaptic membrane different from voltage-gating of Na or Ca channels. Such a mechanism may be important in the establishment of long-lasting synaptic changes at the crayfish neuromuscular junction and perhaps in other neural systems.

Long-term facilitation of synaptic transmission demonstrated with macro-patch recording at the crayfish neuromuscular junction

Recordings of synaptic currents from the crayfish opener muscle were made with a macro-patch recording technique, permitting clear detection of neurotransmitter quanta at individual nerve terminals before and after induction of long-term facilitation (LTF). Depolarization of the terminal by propagated action potentials or by local intracellular pulses induced LTF. The quantal content was increased on average by 93%. Binomial analysis indicated increased probability of release and also increased number of available quantal units. The increase occurred regardless of a blockade of sodium, calcium or potassium channels by appropriate pharmacological agents. Presynaptic recording with an intracellular microelectrode showed no change in presynaptic electrical properties. Also, there were no changes in the synaptic delay. It is concluded that LTF results from a depolarization-dependent alteration of synaptic release sites.

Long-term facilitation and low-frequency depression in a crayfish phasic motor axon

1. Stimulation of the fast axon to the closer muscle of the isolated crayfish claw preparation at 0.1 Hz leads to low-frequency depression (LFD) of the excitatory postsynaptic potentials (EPSPs) recorded from single muscle fibers. EPSP amplitude is reduced by 50% after only 30 stimuli at 0.1 Hz, and declines during maintained 0.1 Hz stimulation with a time constant of 10–15 min. 2. LFD is not affected by temperature changes ranging from 10 °C to 25 °C. However, ouabain (10−4M-10−3M) in the extracellular solution slightly reduces the rate of LFD. 3. When 0.1 Hz stimulation is delivered following a period of conditioning stimulation (5–10 Hz) which causes depression, a facilitated EPSP appears. Facilitation persists for 1/2 h or more after induction. This long-term facilitation (LTF) has some features in common with that seen after stimulation of a tonic motoneuron in crayfish and crabs. 4. The amplitude of LTF is directly related to the number of conditioning impulses. However, the effect is pattern-sensitive: the same number of impulses delivered in a bursting pattern enhances the expression of LTF. 5. LTF is enhanced by lowering the temperature or applying ouabain (10−3M) during conditioning. 6. The results implicate sodium loading of phasic axon nerve terminals during stimulation as a factor in enhancing EPSP amplitude. This suggests that the mechanism for induction of LTF at phasic neuromuscular synapses is similar to that in crustacean tonic neuromuscular synapses.

Long-term facilitation alters transmitter releasing properties at the crayfish neuromuscular junction.

Synaptic transmission at the neuromuscular junction of the excitatory axon supplying the crayfish opener muscle was examined before and after induction of long-term facilitation (LTF) by a 10-min period of stimulation at 20 Hz. Induction of LTF led to a period of enhanced synaptic transmission, which often persisted for many hours. The enhancement was entirely presynaptic in origin, since quantal unit size and time course were not altered, and quantal content of transmission (m) was increased. LTF was not associated with any persistent changes in action potential or presynaptic membrane potential recorded in the terminal region of the excitatory axon. The small muscle fibers of the walking-leg opener muscle were almost isopotential, and all quantal events could be recorded with an intracellular microelectrode. In addition, at low frequencies of stimulation, m was small. Thus it was possible to apply a binomial model of transmitter release to events recorded from individual muscle fibers and to calculate values for n (number of responding units involved in transmission) and p (probability of transmission for the population of responding units) before and after LTF. In the majority of preparations analyzed (6/10), amplitude histograms of evoked synaptic potentials could be described by a binomial distribution with a small n and moderately high p. LTF produced a significant increase in n, while p was slightly reduced. The results can be explained by a model in which the binomial parameter n represents the number of active synapses and parameter p the mean probability of release at a synapse. Provided that a pool of initially inactive synapses exists, one can postulate that LTF involves recruitment of synapses to the active state.

Correlation of presynaptic and postsynaptic events during establishment of long-term facilitation at crayfish neuromuscular junction

Repetitive stimulation (10-20 Hz) of the motor axon supplying the opener muscle in the crayfish leg produces long-lasting enhancement of excitatory postsynaptic potentials. This long-term facilitation (LTF) was investigated by recording simultaneously from the presynaptic nerve terminal and from the innervated muscle fiber with intracellular microelectrodes. On cessation of stimulation, the facilitated postsynaptic potential declines in amplitude when monitored with low-frequency test stimuli. A rapid decline (phase I) occurs over the first 30 s and is succeeded by a more gradual decline lasting several minutes (phase II). Finally, a residual potentiation with a very slow decay (phase III) persists for several hours. Simultaneous pre- and postsynaptic recordings were made during induction of LTF with stimuli delivered at 20 Hz for 10 min. During the tetanus, excitatory postsynaptic potentials were enhanced 20-fold, while action potentials in the presynaptic terminal declined in amplitude from 108.6 to 97.2 mV, and the presynaptic membrane became hyperpolarized by 6.4 mV. The Na+ pump inhibitor ouabain (0.5-1.0 mM) abolished the hyperpolarization, indicating that the latter resulted from activation of an electrogenic Na+ pump. The reduction in amplitude of the presynaptic action potential was consistent with a reduced transmembrane concentration gradient for Na+. Thus, it is suggested that a significant accumulation of Na+ occurs during repetitive stimulation of crayfish motor axons. Decay of phase II of LTF, but not of phases I or III, had approximately the same time course as the decay of Na+ accumulation in the terminals, monitored by changes in the presynaptic action potential. Thus it is probable that in crayfish this phase of LTF is linked to an increased intraterminal Na+ concentration. Injection of Na+ from a microelectrode into the presynaptic terminal produced enhancement of the excitatory postsynaptic potential lasting for many minutes, as well as changes in presynaptic membrane potential and action potential similar to those seen during repetitive stimulation. The results provide the first direct measurements of electrical and ionic changes in axonal terminals during prolonged periods of activity leading to LTF, and support the hypothesis that accumulation of intraterminal Na+ is associated with one phase of LTF.