Postsynaptic Action Potentials Research Articles

Presynaptic NR2A-containing NMDA receptors implement a high-pass filter synaptic plasticity rule

The detailed characterization of synaptic plasticity has led to the replacement of simple Hebbian rules by more complex rules depending on the order of presynaptic and postsynaptic action potentials. Here, we describe a mechanism endowing a plasticity rule with additional computational complexity--a dependence on the pattern of presynaptic action potentials. The classical Hebbian rule is based on detection of conjunctive presynaptic and postsynaptic activity by postsynaptic NMDA receptors, but there is also accumulating evidence for the existence of presynaptic NMDA receptors in several brain structures. Here, we examine the role of presynaptic NMDA receptors in defining the temporal structure of the plasticity rule governing induction of long-term depression (LTD) at the cerebellar parallel fiber-Purkinje cell synapse. We show that multiple presynaptic action potentials at frequencies between 40 Hz and 1 kHz are necessary for LTD induction. We characterize the subtype, kinetics, and role of presynaptic NMDA receptors involved in the induction of LTD, showing how the kinetics of the NR2A subunits expressed by parallel fibers implement a high-pass filter plasticity rule that will selectively attenuate synapses undergoing high-frequency bursts of activity. Depending on the type of NMDA receptor subunit expressed, high-pass filters of different corner frequencies could be implemented at other synapses expressing NMDA autoreceptors.

Analysis of Caribbean ciguatoxin-1 effects on frog myelinated axons and the neuromuscular junction

Caribbean ciguatoxin-1 (C-CTX-1) induced, after about 1 h exposure, muscle membrane depolarisation and repetitive post-synaptic action potentials (APs) in frog neuromuscular preparations. This depolarising effect was also observed in a Ca 2+-free medium with a strong enhancement of spontaneous quantal transmitter release, compared with control conditions. The ciguatoxin-induced increase in release could be accelerated when Ca 2+ was present in the extracellular medium. C-CTX-1 also enhanced nerve-evoked quantal acetylcholine (ACh) release. At normal neuromuscular junctions loaded with the fluorescent dye FM1-43, C-CTX-1 induced swelling of nerve terminals, an effect that was reversed by hyperosmotic d-mannitol. In myelinated axons, C-CTX-1 increased nodal membrane excitability, inducing spontaneous and repetitive APs. Also, the toxin enlarged the repolarising phase of APs in control and tetraethylammonium-treated axons. Overall, our data suggest that C-CTX-1 affects nerve excitability and neurotransmitter release at nerve terminals. We conclude that C-CTX-1-induced up-regulation of Na + channels and the inhibition of K + channels, at low nanomolar concentrations, produce a variety of functional dysfunctions that are in part responsible for the human muscle skeletal symptoms observed in ciguatera. All these dysfunctions seem to result from the subtle balance between ionic currents, intracellular Na + and Ca 2+ concentrations, and engaged second messengers.

Formula omitted] regenerative-loop potentiation of glutamate synaptic transmission by microglia and astrocytes

P 2 X 7 purinergic receptors have been implicated in chronic neuropathic and neuroinflammatory pain as well as in depression. These receptors are predominantly found in the central nervous system on microglial cells and on glutamatergic nerve terminals. Here, we develop hypotheses concerning mechanisms by which transient high-frequency impulse firing in glutamatergic terminals, such as occurs in nociceptor terminals accompanying neuropathic/neuroinflammatory pain, can lead to long-lasting changes in neural network function that is mediated by surrounding glial cells. The hypothesis consists of two parts. In the first, glutamate released by low-frequency (2 Hz) terminal action potentials is insufficient to generate postsynaptic action potentials, but these are generated by brief high-frequency input bursts. Glutamate released by these bursts is partly removed by transporters on the enveloping astrocyte processes and also excites AMPA receptors on these processes, which then release ATP. This ATP is partly metabolised to adenosine, which acts on presynaptic A1 receptors to inhibit glutamate release. The remaining ATP acts on the presynaptic P 2 X 7 receptors to facilitate glutamate release by both the high-frequency burst of action potentials as well as by a continuous low-frequency (2Hz) action potential firing that occurs in the absence of a neuropathic/neuroinflammatory insult. The positive feedback of terminal glutamate release, triggering astrocyte ATP release and leading to further glutamate release through activation of P 2 X 7 receptors, is then sufficient to allow the normal low-frequency (2 Hz) action potentials to now elicit postsynaptic action potentials after the insult is removed. In the second part of this model, the high concentration of ATP derived from astrocytes at the terminal attracts microglia by chemotaxis. The P 2 X 7 receptors on these microglia are then engaged, resulting in microglia secreting the cytokine TNF α . This acts on postsynaptic TNF-R1 receptors to increase the number of AMPA receptors there, thus enhancing the efficacy of synaptic transmission. The TNF α also acts on presynaptic TNF-R1 to increase the amount of glutamate released by each nerve terminal impulse. Experimental tests can be made of this hypothesis that P 2 X 7 receptors on the presynaptic terminal and those on the microglia synergistically act to ensure feedback pathways that reset to a high level the efficacy of synaptic transmission, thus ensuring chronic neuropathic/neuroinflammatory pain even when the initial insult has subsided.

On the mechanism of action of muscle fibre activity in synapse competition and elimination at the mammalian neuromuscular junction

Activity-dependent competition plays a crucial role in the refinement of synaptic connections in the peripheral and central nervous system. The reduction in number of axons innervating each neuromuscular junction during development, i.e. synapse elimination, appears to be one such competitive activity-driven event. Recently, we showed that asynchronous firing of competing presynaptic terminals is a key player in synapse elimination. Although some previous studies suggested that activity of the postsynaptic cell may be an intermediary in the disposal of redundant presynaptic inputs, the mechanism involved remains unknown. In the present study, in order to assess the role of evoked muscle activity in this process, we inhibited the generation of postsynaptic action potentials in muscle fibers in vivo, through the overexpression of inwardly rectifying Kir2.1 and Kir2.2 channels, via electroporation of the soleus muscle in the mouse hindlimb. Electrophysiological and morphological data show that overexpression of potassium channels in the endplate region of neonatal muscle fibres induces membrane hyperpolarization and an increase in conductance, inhibition of the action potential mechanism and prolonged persistence of polyneuronal innervation. These changes are not seen in muscle fibres with overexpression of a non-conducting Kir2.1 mutant. Our results are compatible with the interpretation that the block of action potential generation, even in single endplates, can inhibit synapse elimination through local signalling.

Learning to Encode Timing: Mechanisms of Plasticity in the Auditory Brainstem

Mechanisms of plasticity have traditionally been ascribed to higher-order sensory processing areas such as the cortex, whereas early sensory processing centers have been considered largely hard-wired. In agreement with this view, the auditory brainstem has been viewed as a nonplastic site, important for preserving temporal information and minimizing transmission delays. However, recent groundbreaking results from animal models and human studies have revealed remarkable evidence for cellular and behavioral mechanisms for learning and memory in the auditory brainstem.

Phase Encoding in the Mauthner System: Implications in Left–Right Sound Source Discrimination

The paired teleost Mauthner (M)-cells and their associated network serve as an excellent system to study the biophysical basis of decision making. In teleosts, an abrupt sound evokes an M-spike, triggering a C-start escape that is usually directed away from a sound source. The response latency is minimized by electrical synapses between auditory afferents and the M-cell lateral dendrite. Here, we demonstrate that the electrical synapses also mediate phase encoding. Ramped sound pressure waves (150-250 Hz) evoked electrotonic postsynaptic potentials in the M-cell locked to two diametrically opposed phase angles that were frequency dependent but intensity independent. Phase encoding was also evident at the behavioral level underwater, because the stimuli evoked directional C-starts with an onset that was phase locked to the sound wave. In interneurons inhibitory to the M-cell, these same stimuli also evoked phase-locked electrotonic postsynaptic potentials and action potentials. The resulting chemical and electrical (i.e., field effect) inhibitions functioned tonically and phasically, respectively. Phase encoding could be important in underwater sound source localization, which is thought to require a neural computation involving a phase comparison between the pressure and the directional particle motion components of sound. This computation may be implemented by an interplay between phase-dependent afferent excitation and feedforward inhibition that activates the appropriate M-cell and directs the C-start away from the sound source.

Synaptic decision making: flipping switch-like synapses with cubic autocatalysis.

Event Abstract Back to Event Synaptic decision making: flipping switch-like synapses with cubic autocatalysis. In the brain, the process of learning is a noisy decision making process. At the level of a single synapse, when two connected neurons are active, each generating a different noisy, but perhaps correlated, sequence of action potentials, the synapse connecting them faces a decision ? how or whether to modify the strength of that connection. Recent work suggests that this change in synapse strength might come in the form of one of two binary decisions (O'Connor et al., PNAS102:9679-9684, 2005). For a synapse in a low-state, the decision is whether to increase in strength (long-term potentiation; LTP), or remain the same. For a synapse in a high-state, the decision is whether to decrease in strength (long-term depression; LTD) or remain the same. When plasticity is measured in a population of synapses, graded learning rules can be measured which reflect the statistics of the single synapse decisions. One form of learning rule which has generated a great deal of interest over the last decade is spike-timing dependent plasticity (STDP), in which the amount and direction of plasticity is a function of the timing difference between presynaptic and postsynaptic action potentials. In hippocampal synapses however, the story is more complicated: both the magnitude and direction of plasticity are influenced by other activity parameters such as spiking frequency, the number of pairings delivered (Wittenberg and Wang, J Neurosci 26:6610-6617, 2006). The biophysical mechanism by which neural activity changes the strength and state of a synapse is well-studied. Both increases in synapse strength and decreases in synapse strength are initiated by increases in postsynaptic [Ca2+] at the synapse above baseline [Ca2+] levels. Large calcium transients drive synapses to potentiate, whereas moderate and prolonged calcium transient drive synapses to depress. The goal of this work is to synthesize and summarize these experimental findings with a simple biophysical model for a switch-like synapse. We construct a simple model based on the bistability arising from cubic autocatalysis, which is reminiscent of the Activator/Inhibitor models for spatial pattern formation in animal coats. With this approach we are able to naturally capture the features of the learning rule determined at the CA3-CA1 synapse described above. Conference: Computational and systems neuroscience 2009, Salt Lake City, UT, United States, 26 Feb - 3 Mar, 2009. Presentation Type: Poster Presentation Topic: Poster Presentations Citation: (2009). Synaptic decision making: flipping switch-like synapses with cubic autocatalysis.. Front. Syst. Neurosci. Conference Abstract: Computational and systems neuroscience 2009. doi: 10.3389/conf.neuro.06.2009.03.273 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 04 Feb 2009; Published Online: 04 Feb 2009. Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract Supplemental Data The Authors in Frontiers Google Google Scholar PubMed Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.

Postsynaptic action potentials are required for nitric-oxide-dependent long-term potentiation in CA1 neurons of adult GluR1 knock-out and wild-type mice.

Neocortical long-term potentiation (LTP) consists of both presynaptic and postsynaptic components that rely on nitric oxide (NO) and the GluR1 subunit of the AMPA receptor, respectively. In this study, we found that hippocampal LTP, induced by theta-burst stimulation in mature (>8-week-old) GluR1 knock-out mice was almost entirely NO dependent and involved both the alpha splice variant of NO synthase-1 and the NO synthase-3 isoforms of NO synthase. Theta-burst induced LTP was also partly NO-dependent in wild-type mice and made up approximately 50% of the potentiation 2 h after tetanus. Theta-burst stimulation reliably produced postsynaptic spikes, including a high probability of complex spikes. Inhibition of postsynaptic somatic spikes with intracellular QX314 or local TTX application prevented LTP in the GluR1 knock-out mice and also blocked the NO component of LTP in wild types. We conclude that theta-burst stimulation is particularly well suited to producing the postsynaptic somatic spikes required for NO-dependent LTP.

The ups and downs of synaptic plasticity: influences on this particular ‘market’

The ups and downs of synaptic plasticity: influences on this particular ‘market’

How Secure IsIn VivoSynaptic Transmission at the Calyx of Held?

The medial nucleus of the trapezoid body (MNTB) receives excitatory input from giant presynaptic terminals, the calyces of Held. The MNTB functions as a sign inverter giving inhibitory input to the lateral and medial superior olive, where its input is important in the generation of binaural sensitivity to cues for sound localization. Extracellular recordings from MNTB neurons show complex spikes consisting of a prepotential, thought to reflect synaptic activation, followed by a postsynaptic action potential. This makes the synapse ideal to study synaptic transmission in vivo because presynaptic and postsynaptic activity can be monitored with a single electrode. Recent in vivo and in vitro studies have observed isolated prepotentials in the MNTB suggesting that, under certain stimulus conditions, synaptic transmission fails. We investigated synaptic transmission at the calyx of Held in the MNTB of the adult cat and concluded that synaptic transmission was completely secure in terms of rate of transmitted spikes. However, synaptic transmission was found to be less secure in terms of timing. With increasing spike rate, the synaptic delay showed an increase of up to 100 micros, as well as a decrease in amplitude of the action potential. This variability in delay is of a surprisingly high magnitude given the hypothesized role of these binaural circuits in sound localization and given the fact that this is one of the largest synapses in the mammalian brain.

Driver or Coincidence Detector: Modal Switch of a Corticothalamic Giant Synapse Controlled by Spontaneous Activity and Short-Term Depression

Giant synapses between layer 5B (L5B) neurons of somatosensory (barrel) cortex and neurons of the posteromedial nucleus (POm) of thalamus reside in a key position of the cortico-thalamo-cortical (CTC) loop, yet their synaptic properties and contribution to CTC information processing remain poorly understood. Fluorescence-guided local stimulation of terminals were combined with postsynaptic whole-cell recordings in thalamus to study synaptic transmission at an identified giant synapse. We found large EPSCs mediated by Ca(2+)-permeable AMPA and NMDA receptors. A single presynaptic electrical stimulus evoked a train of postsynaptic action potentials, indicating that a single L5B input can effectively drive the thalamic neuron. Repetitive stimulation caused strong short-term depression (STD) with fast recovery. To examine how these synaptic properties affect information transfer, spontaneous and evoked activity of L5B neurons was recorded in vivo and played back to giant terminals in vitro. We found that suprathreshold synaptic transmission was suppressed because of spontaneous activity causing strong STD of the L5B-POm giant synapse. Thalamic neurons only spiked after intervals of presynaptic silence or when costimulating two giant terminals. Therefore, STD caused by spontaneous activity of L5B neurons can switch the synapse from a "driver mode" to a "coincidence mode." Mechanisms decreasing spontaneous activity in L5B neurons and inputs synchronized by a sensory stimulus may thus gate the cortico-thalamo-cortical loop.

Frequency-Dependent Glycinergic Inhibition Modulates Plasticity in Hippocampus

Previous studies have demonstrated the presence of functional glycine receptors (GlyRs) in hippocampus. In this work, we examine the baseline activity and activity-dependent modulation of GlyRs in region CA1. We find that strychnine-sensitive GlyRs are open in the resting CA1 pyramidal cell, creating a state of tonic inhibition that "shunts" the magnitude of EPSPs evoked by electrical stimulation of the Schaffer collateral inputs. This GlyR-mediated shunting conductance is independent of the presynaptic stimulation rate; however, pairs of presynaptic and postsynaptic action potentials, repeated at frequencies above 5 Hz, reduce the GlyR-mediated conductance and increase peak EPSP magnitudes to levels at least 20% larger than those seen with presynaptic stimulation alone. We refer to this phenomenon as rate-dependent efficacy (RDE). Exogenous GlyR agonists (glycine, taurine) block RDE by preventing the closure of postsynaptic GlyRs. The GlyR antagonist strychnine blocks postsynaptic GlyRs under all conditions, occluding RDE. During RDE, GlyRs are less responsive to local glycine application, suggesting that a reduction in the number or sensitivity of membrane-inserted GlyRs underlies RDE. By extending the RDE induction protocol to include 500 paired presynaptic and postsynaptic spikes, we can induce long-term synaptic depression (LTD). Manipulations that lead to reduced functionality of GlyRs, either pharmacologically or through RDE, also lead to increased LTD. This result suggests that RDE contributes to long-term synaptic plasticity in the hippocampus.

A Precisely Timed Asynchronous Pattern of ON and OFF Retinal Ganglion Cell Activity during Propagation of Retinal Waves

Patterns of coordinated spontaneous activity have been proposed to guide circuit refinement in many parts of the developing nervous system. It is unclear, however, how such patterns, which are thought to indiscriminately synchronize nearby cells, could provide the cues necessary to segregate functionally distinct circuits within overlapping cell populations. Here, we report that glutamatergic retinal waves possess a substructure in the bursting of neighboring retinal ganglion cells with opposite light responses (ON or OFF). Within a wave, cells fire repetitive nonoverlapping bursts in a fixed order: ON before OFF. This pattern is absent from cholinergic waves, which precede glutamate-dependent activity, providing a developmental sequence of distinct activity-encoded cues. Asynchronous bursting of ON and OFF retinal ganglion cells depends on inhibition between these parallel pathways. Similar asynchronous activity patterns could arise throughout the nervous system, as inhibition matures and might help to separate connections of functionally distinct subnetworks.

Different requirements for action potentials in the induction of different forms of long‐term potentiation

The role of postsynaptic action potentials (APs) in the induction of long-term potentiation (LTP) remains unclear, but has important implications for theories of associative learning in the brain. In area CA1 of hippocampus, at least three discrete forms of LTP coexist, each displaying unique decay kinetics and involving different signalling and effector systems. The present work investigates whether these forms of LTP also differ in their requirement for postsynaptic APs. Inhibition of APs during theta-burst stimulation (TBS) had no effect on the persistence of short-lasting LTP (LTP 1), but reduced the persistence of more durable forms (LTP 2 and 3). Calcium imaging revealed different requirements for APs in generating calcium signals in spines, dendrites, and somata, consistent with their known roles in the induction of each form of LTP. Finally, short-lasting LTP was endowed with dramatically enhanced persistence by the presentation of TBS-patterned APs alone. These data reveal that the requirement for APs in LTP induction is dependent on the form of LTP under investigation, supporting the contention that different neuronal learning mechanisms coexist in hippocampal area CA1.

Dopamine Receptor Activation Is Required for Corticostriatal Spike-Timing-Dependent Plasticity

Single action potentials (APs) backpropagate into the higher-order dendrites of striatal spiny projection neurons during cortically driven "up" states. The timing of these backpropagating APs relative to the arriving corticostriatal excitatory inputs determines changes in dendritic calcium concentration. The question arises to whether this spike-timing relative to cortical excitatory inputs can also induce synaptic plasticity at corticostriatal synapses. Here we show that timing of single postsynaptic APs relative to the cortically evoked EPSP determines both the direction and the strength of synaptic plasticity in spiny projection neurons. Single APs occurring 30 ms before the cortically evoked EPSP induced long-term depression (LTD), whereas APs occurring 10 ms after the EPSP induced long-term potentiation (LTP). The amount of plasticity decreased as the time between the APs and EPSPs was increased, with the resulting spike-timing window being broader for LTD than for LTP. In addition, we show that dopamine receptor activation is required for this spike-timing-dependent plasticity (STDP). Blocking dopamine D(1)/D(5) receptors prevented both LTD and LTP induction. In contrast, blocking dopamine D(2) receptors delayed, but did not prevent, LTD and sped induction of LTP. We conclude (1) that, in combination with cortical inputs, single APs evoked in spiny projection neurons can induce both LTP and LTD of the corticostriatal pathway; (2) that the strength and direction of these synaptic changes depend deterministically on the AP timing relative to the arriving cortical inputs; (3) that, whereas dopamine D(2) receptor activation modulates the initial phase of striatal STDP, dopamine D(1)/D(5) receptor activation is critically required for striatal STDP. Thus, the timing of APs relative to cortical inputs alone is not enough to induce corticostriatal plasticity, implying that ongoing activity does not affect synaptic strength unless dopamine receptors are activated.

STDP Provides the Substrate for Igniting Synfire Chains by Spatiotemporal Input Patterns

Spike-timing-dependent synaptic plasticity (STDP), which depends on the temporal difference between pre- and postsynaptic action potentials, is observed in the cortices and hippocampus. Although several theoretical and experimental studies have revealed its fundamental aspects, its functional role remains unclear. To examine how an input spatiotemporal spike pattern is altered by STDP, we observed the output spike patterns of a spiking neural network model with an asymmetrical STDP rule when the input spatiotemporal pattern is repeatedly applied. The spiking neural network comprises excitatory and inhibitory neurons that exhibit local interactions. Numerical experiments show that the spiking neural network generates a single global synchrony whose relative timing depends on the input spatiotemporal pattern and the neural network structure. This result implies that the spiking neural network learns the transformation from spatiotemporal to temporal information. In the literature, the origin of the synfire chain has not been sufficiently focused on. Our results indicate that spiking neural networks with STDP can ignite synfire chains in the cortices.

Structural plasticity controlled by calcium based correlation detection

Hebbian learning in cortical networks during development and adulthood relies on the presence of a mechanism to detect correlation between the presynaptic and the postsynaptic spiking activity. Recently, the calcium concentration in spines was experimentally shown to be a correlation sensitive signal with the necessary properties: it is confined to the spine volume, it depends on the relative timing of pre- and postsynaptic action potentials, and it is independent of the spine's location along the dendrite. NMDA receptors are a candidate mediator for the correlation dependent calcium signal. Here, we present a quantitative model of correlation detection in synapses based on the calcium influx through NMDA receptors under realistic conditions of irregular pre- and postsynaptic spiking activity with pairwise correlation. Our analytical framework captures the interaction of the learning rule and the correlation dynamics of the neurons. We find that a simple thresholding mechanism can act as a sensitive and reliable correlation detector at physiological firing rates. Furthermore, the mechanism is sensitive to correlation among afferent synapses by cooperation and competition. In our model this mechanism controls synapse formation and elimination. We explain how synapse elimination leads to firing rate homeostasis and show that the connectivity structure is shaped by the correlations between neighboring inputs.

A Shout Out to Immature Synapses. Focus on “Different Roles for AMPA and NMDA Receptors in Transmission at the Immature Retinogeniculate Synapse”

A major topic of investigation in the field of developmental neuroscience is to understand how newly formed excitatory synapses receive and transmit information. It is now widely accepted that neural activity plays an important role in a number of developmental processes, including the refinement and establishment of orderly connections (Grubb and Thompson 2004). However, it remains unclear how many immature excitatory synapses (i.e., ones that utilize glutamate as a neurotransmitter) participate in the relay of activity. Contributing to this problem is the finding that immature glutamatergic synapses are comprised largely of N-methyl-D-aspartate (NMDA) receptors and lack AMPA receptors that mediate fast excitatory transmission. Such synapses are said to be “silent,” incapable of causing postsynaptic action potentials from resting levels because in addition to glutamate, the NMDA receptor requires membrane depolarization to relieve a voltage-dependent Mg 2 block. How then do silent synapses “speak” loudly enough to cause postsynaptic firing and thereby successfully relay information? In a recent article, Liu and Chen (2008) provide some answers to this question. They examine the synaptic responses and intrinsic properties of developing neurons by making use of an in vitro thalamic slice preparation that maintains the excitatory synaptic connections between retinal ganglion cells and relay cells of the dorsal lateral geniculate nucleus (LGN). In an elegant series of whole cell recording experiments, they show how a constellation of properties, including ligand-gated channel kinetics, receptor subunit composition, extended presence of neurotransmitter

The development of synaptic plasticity induction rules and the requirement for postsynaptic spikes in rat hippocampal CA1 pyramidal neurones

Coincident pre- and postsynaptic activity induces synaptic plasticity at the Schaffer collateral synapse onto CA1 pyramidal neurones. The precise timing, frequency and number of coincident action potentials required to induce synaptic plasticity is currently unknown. In this study we show that the postsynaptic activity required for the induction of long-term potentiation (LTP) changes with development. In acute slices from adult rats, coincident pre- and postsynaptic theta burst stimulation (TBS) induced LTP and we show that multiple high-frequency postsynaptic spikes are required. In contrast, in acute slices from juvenile (P14) rats, TBS failed to induce LTP unless the excitatory postsynaptic potentials (EPSPs) were of sufficient magnitude to initiate action potentials. We also show that coincident individual pre- and postsynaptic action potentials are only capable of inducing LTP in the juvenile when given at a frequency greater than 5 Hz and that the timing of individual pre- and postsynaptic action potentials relative to one another is not important. Finally, we show that local tetrodotoxin (TTX) application to the soma blocked LTP in adults, but not juveniles. These data demonstrate that somatic spiking is more important for LTP induction in the adult as opposed to juvenile rats and we hypothesize that the basis for this is the ability of action potentials in the postsynaptic CA1 pyramidal neurone to back-propagate into the dendrites. Therefore, the pre- and postsynaptic activity patterns required to induce LTP mature as the hippocampus develops.

Learning Real-World Stimuli in a Neural Network with Spike-Driven Synaptic Dynamics

We present a model of spike-driven synaptic plasticity inspired by experimental observations and motivated by the desire to build an electronic hardware device that can learn to classify complex stimuli in a semisupervised fashion. During training, patterns of activity are sequentially imposed on the input neurons, and an additional instructor signal drives the output neurons toward the desired activity. The network is made of integrate-and-fire neurons with constant leak and a floor. The synapses are bistable, and they are modified by the arrival of presynaptic spikes. The sign of the change is determined by both the depolarization and the state of a variable that integrates the postsynaptic action potentials. Following the training phase, the instructor signal is removed, and the output neurons are driven purely by the activity of the input neurons weighted by the plastic synapses. In the absence of stimulation, the synapses preserve their internal state indefinitely. Memories are also very robust to the disruptive action of spontaneous activity. A network of 2000 input neurons is shown to be able to classify correctly a large number (thousands) of highly overlapping patterns (300 classes of preprocessed Latex characters, 30 patterns per class, and a subset of the NIST characters data set) and to generalize with performances that are better than or comparable to those of artificial neural networks. Finally we show that the synaptic dynamics is compatible with many of the experimental observations on the induction of long-term modifications (spike-timing-dependent plasticity and its dependence on both the postsynaptic depolarization and the frequency of pre- and postsynaptic neurons).