Pharmacological dissociation of memory: anisomycin, a protein synthesis inhibitor, and leupeptin, a protease inhibitor, block different learning tasks

Temporal phases of long-term potentiation (LTP): myth or fact?

Long-term potentiation (LTP) remains the most widely accepted model for learning and memory. In accordance with this belief, the temporal differentiation of LTP into early and late phases is accepted as reflecting the differentiation of short-term and long-term memory. Moreover, during the past 30 years, protein synthesis inhibitors have been used to separate the early, protein synthesis-independent (E-LTP) phase and the late, protein synthesis-dependent (L-LTP) phase. However, the role of these proteins has not been formally identified. Additionally, several reports failed to show an effect of protein synthesis inhibitors on LTP. In this review, a detailed analysis of extensive behavioral and electrophysiological data reveals that the presumed correspondence of LTP temporal phases to memory phases is neither experimentally nor theoretically consistent. Moreover, an overview of the time courses of E-LTP in hippocampal slices reveals a wide variability ranging from <1 h to more than 5 h. The existence of all these conflictual findings should lead to a new vision of LTP. We believe that the E-LTP vs. L-LTP distinction, established with protein synthesis inhibitor studies, reflects a false dichotomy. We suggest that the duration of LTP and its dependency on protein synthesis are related to the availability of a set of proteins at synapses and not to the de novo synthesis of plasticity-related proteins. This availability is determined by protein turnover kinetics, which is regulated by previous and ongoing electrical activities and by energy store availability.

Cognitive neuroepigenetics: the next evolution in our understanding of the molecular mechanisms underlying learning and memory?

A complete understanding of the fundamental mechanisms of learning and memory continues to elude neuroscientists. Although many important discoveries have been made, the question of how memories are encoded and maintained at the molecular level remains. To date, this issue has been framed within the context of one of the most dominant concepts in molecular biology, the central dogma, and the result has been a protein-centric view of memory. Here we discuss the evidence supporting a role for neuroepigenetic mechanisms, which constitute dynamic and reversible, state-dependent modifications at all levels of control over cellular function, and their role in learning and memory. This neuroepigenetic view suggests that DNA, RNA and protein each influence one another to produce a holistic cellular state that contributes to the formation and maintenance of memory, and predicts a parallel and distributed system for the consolidation, storage and retrieval of the engram.

Protein synthesis and consolidation of memory-related synaptic changes

Although sometimes disputed, it has been assumed for several decades that new proteins synthesized following a learning event are required for consolidation of subsequent memory. Published findings and new results described here challenge this idea. Protein synthesis inhibitors did not prevent Theta Bust Stimulation (TBS) from producing extremely stable long-term potentiation (LTP) in experiments using standard hippocampal slice protocols. However, the inhibitors were effective under conditions that likely depleted protein levels prior to attempts to induce the potentiation effect. Experiments showed that induction of LTP at one input, and thus a prior episode of protein synthesis, eliminated the effects of inhibitors on potentiation of a second input even in depleted slices. These observations suggest that a primary role of translation and transcription processes initiated by learning events is to prepare neurons to support future learning. Other work has provided support for an alternative theory of consolidation. Specifically, if the synaptic changes that support memory are to endure, learning events/TBS must engage a complex set of signaling processes that reorganize and re-stabilize the spine actin cytoskeleton. This is accomplished in fast (10 min) and slow (50 min) stages with the first requiring integrin activation and the second a recovery of integrin functioning. These results align with, and provide mechanisms for, the long-held view that memories are established and consolidated over a set of temporally distinct phases. This article is part of a Special Issue entitled SI: Brain and Memory.

Inhibition of mTOR by rapamycin in the amygdala or hippocampus impairs formation and reconsolidation of inhibitory avoidance memory

Mammalian target of rapamycin (mTOR), a central regulator of protein synthesis in neurons, has been implicated in synaptic plasticity and memory. Here we show that mTOR inhibition by rapamycin in the basolateral amygdala (BLA) or dorsal hippocampus (DH) impairs both formation and reconsolidation of memory for inhibitory avoidance (IA) in rats. Male Wistar rats received bilateral infusions of vehicle or rapamycin into the BLA or DH before or after IA training or retrieval. Memory retention was tested at different time points after drug infusion. Rapamycin impaired long-term IA retention when given before or immediately after training or retrieval into the BLA. When infused into the DH, rapamycin produced memory impairment when given before training or immediately after retrieval. The impairing effects of post-retrieval rapamycin required memory retrieval and were not reversed by a reminder shock. The results provide the first evidence that mTOR in the BLA and DH might play a role in IA memory reconsolidation.

The role of protein synthesis in memory consolidation: progress amid decades of debate

A major component of consolidation theory holds that protein synthesis is required to produce the synaptic modification needed for long-term memory storage. Protein synthesis inhibitors have played a pivotal role in the development of this theory. However, these commonly used drugs have unintended effects that have prompted some to reevaluate the role of protein synthesis in memory consolidation. Here we review the role of protein synthesis in memory formation as proposed by consolidation theory calling special attention to the controversy involving the non-specific effects of a group of protein synthesis inhibitors commonly used to study memory formation in vivo. We argue that molecular and genetic approaches that were subsequently applied to the problem of memory formation confirm the results of less selective pharmacological studies. Thus, to a certain extent, the debate over the role of protein synthesis in memory based on interpretational difficulties inherent to the use of protein synthesis inhibitors may be somewhat moot. We conclude by presenting avenues of research we believe will best provide answers to both long-standing and more recent questions facing field of learning and memory.

On the role of hippocampal protein synthesis in the consolidation and reconsolidation of object recognition memory

Upon retrieval, consolidated memories are again rendered vulnerable to the action of metabolic blockers, notably protein synthesis inhibitors. This has led to the hypothesis that memories are reconsolidated at the time of retrieval, and that this depends on protein synthesis. Ample evidence indicates that the hippocampus plays a key role both in the consolidation and reconsolidation of different memories. Despite this fact, at present there are no studies about the consequences of hippocampal protein synthesis inhibition in the storage and post-retrieval persistence of object recognition memory. Here we report that infusion of the protein synthesis inhibitor anisomycin in the dorsal CA1 region immediately or 180 min but not 360 min after training impairs consolidation of long-term object recognition memory without affecting short-term memory, exploratory behavior, anxiety state, or hippocampal functionality. When given into CA1 after memory reactivation in the presence of familiar objects, ANI did not affect further retention. However, when administered into CA1 immediately after exposing animals to a novel and a familiar object, ANI impaired memory of both of them. The amnesic effect of ANI was long-lasting, did not happen after exposure to two novel objects, following exploration of the context alone, or in the absence of specific stimuli, suggesting that it was not reversible but was contingent on the reactivation of the consolidated trace in the presence of a salient, behaviorally relevant novel cue. Our results indicate that hippocampal protein synthesis is required during a limited post-training time window for consolidation of object recognition memory and show that the hippocampus is engaged during reconsolidation of this type of memory, maybe accruing new information into the original trace.

Retrieval induces hippocampal-dependent reconsolidation of spatial memory

Nonreinforced retrieval can cause extinction and/or reconsolidation, two processes that affect subsequent retrieval in opposite ways. Using the Morris water maze task we show that, in the rat, repeated nonreinforced expression of spatial memory causes extinction, which is unaffected by inhibition of protein synthesis within the CA1 region of the dorsal hippocampus. However, if the number of nonreinforced retrieval trials is insufficient to induce long-lasting extinction, then a hippocampal protein synthesis-dependent reconsolidation process recovers the original memory. Inhibition of hippocampal protein synthesis after reversal learning sessions impairs retention of the reversed preference and blocks persistence of the original one, suggesting that reversal learning involves reconsolidation rather than extinction of the original memory. Our results suggest the existence of a hippocampal protein synthesis-dependent reconsolidation process that operates to recover or update retrieval-weakened memories from incomplete extinction.

Post-translational protein modification as the substrate for long-lasting memory

Prevailing models of memory identify mRNA translation as necessary for long-lasting information storage. However, there are enough instances of memory storage in the virtual absence of protein synthesis to prompt consideration of alternative models. A comprehensive review of the protein synthesis literature leads us to conclude that the translational mechanism is exclusively a permissive, replenishment step. Therefore, we propose that post-translational modification (PTM) of proteins already at the synapse is the crucial instructive mechanism underlying long-lasting memory. A novel feature of this model is that non-random spontaneous (or endogenous) brain activity operates as a regulated positive-feedback rehearsal mechanism, updating network configurations by fine-tuning the PTM state of previously modified proteins. Synapses participating in memory storage are therefore supple, a feature required for networks to alter complexity and update continuously. In analogy with codons for amino acids, a long-lasting memory is represented by a 'degenerate code' - a set of pseudo-redundant networks that can ensure its longevity.

A genetic deficiency in calpastatin and isovalerylcarnitine treatment is associated with enhanced hippocampal long-term potentiation

The Milan hypertensive strain (MHS) of rats, in addition to having hypertension, is also characterized by a genetic deficiency in calpastatin, the endogenous inhibitor of calpain. Since this protease has been implicated in long-term potentiation (LTP), we have investigated whether induction of this form of plasticity was altered in this strain of rats as compared to control animals (Milan normotensive strain, MNS). Progressive induction of LTP by increasing numbers of high frequency trains resulted in a greater degree of potentiation measured with all inducing protocols in MHS as compared with MNS animals. This difference was not related to the hypertension, since another hypertensive strain (the SHR strain) and a segregated Milan hypertensive strain, expressing only the hypertension but not the calpastatin deficiency (the MHNE strain), exhibited an LTP indistinguishable from control rats. Treatment of MHNE rats for 2 months with isovalerylcarnitine, a compound that increases calpain activity, also resulted in a greater amount of LTP induced by high frequency trains. These effects were not related to an enhancement of the NMDA receptor dependent component of responses to burst stimulation. These results are consistent with the idea that conditions under which calpain activation is facilitated are associated with a greater degree of synaptic potentiation.

Genetic dissection of consolidated memory in Drosophila

Behavioral and pharmacological experiments in many animal species have suggested that memory is consolidated from an initial, disruptable form into a long-lasting, stable form within a few hours after training. We combined these traditional approaches with genetic analyses in Drosophila to show that consolidated memory of conditioned (learned) odor avoidance 1 day after extended training consisted of two genetically distinct, functionally independent memory components: anesthesia-resistant memory (ARM) and long-term memory (LTM). ARM decayed away within 4 days, was resistant to hypothermic disruption, was insensitive to the protein synthesis inhibitor cycloheximide (CXM), and was disrupted by the radish single-gene mutation. LTM showed no appreciable decay over 7 days, was sensitive to CXM, and was not disrupted by the radish mutation.

Altered synaptic plasticity and memory formation in nitric oxide synthase inhibitor-treated rats

Nitric oxide (NO) is a messenger molecule that is produced in the brain from the metabolism of L-arginine to L-citrulline. Growing evidence suggests a physiological role for NO in long-term potentiation (LTP). Since LTP is a form of synaptic plasticity thought to be involved in learning and memory, we have tested whether inhibition of endogenous NO production affects memory capacities of rats. We found that the NO synthase [L-arginine, NADPH:oxygen oxidoreductase (nitric oxide-forming), EC 1.14.13.39] inhibitor N omega-nitro-L-arginine, at doses blocking LTP in hippocampal slices, impairs spatial learning in a radial arm maze and olfactory memory in a social recognition test. In contrast, N omega-nitro-L-arginine left shock-avoidance learning unaffected. These results indicate that NO is involved in some but not all forms of memory and further support the existence of a causal link between LTP and spatial learning.

Effects of cycloheximide on the retention of olfactory learning and maternal experience effects in postpartum rats

We have previously shown that cycloheximide, a protein synthesis inhibitor, blocks retention of experimentally based material responding in postpartum rats. In the following study, we investigated further the effect of this drug on maternal experience effects and, in particular, on the retention of olfactory learning. Dams were injected SC with cycloheximide or saline following a one-hour maternal experience with pups scented with one of two artificial odorants. When tested eight days later, saline but not cycloheximide animals demonstrated a preference for the odor which they had experience on pups. Animals were then tested for maternal responsiveness with pups that were scented with an odor that was either the same or different from the one used in the maternal experience. All animals receiving cycloheximide, regardless of the pup odor at testing, exhibited long latencies to become maternal, replicating our earlier finding that the retention of learning in maternal behavior is susceptible to disruption by protein synthesis inhibitors. A similar disruption of the experience effect was found when the pup odor at testing was different from the exposure odor. This suggests that olfactory learning may normally play a role in the mediation of postpartum experience effects.

Stable depression of potentiated synaptic responses in the hippocampus with 1-5 Hz stimulation

Adult rats with two chronic stimulating electrodes in the Schaffer collateral/commissural system of the hippocampus and one recording electrode in the stratum radiatum (apical dendrites) of field CA1 were administered high-frequency stimulation (10 brief bursts at theta frequency) to produce long-term potentiation (LTP). 'Low frequency' stimulation (100 pulses at 1 Hz alone or followed by 250 pulses at 5 Hz) delivered 5-15 min later had no effect on LTP in 18% of the rats, caused a transient reversal in 18% of the group, but produced an apparent reversal of LTP for the remainder of a 1 h test session in 64% of the animals. LTP did not recover in animals tested 24 h later, at which point a second episode of high-frequency stimulation but without subsequent low-frequency stimulation was administered. This produced an LTP effect that persisted for a 1 h test session in 94% of the cases and that was still present in 86% of the animals tested 24 h later. Low-frequency stimulation applied prior to induction of LTP had no lasting effects on evoked responses not did it affect responses to a control stimulating electrode in those cases in which it reversed LTP. Possible implications of these results for hypotheses concerning the substrates of LTP and mechanisms of forgetting are discussed.

Anoxia reveals a vulnerable period in the development of long-term potentiation

Transient anoxia occurring 1-2 min after high-frequency stimulation selectively prevented the stable expression of long-term potentiation (LTP). Anoxia occurring after this brief vulnerable period did not reverse LTP. Experiments on the duration of anoxia necessary to block LTP expression indicated that simply reducing synaptic transmission was insufficient but that membrane depolarization was not required. The effects of anoxia on LTP were blocked by antagonists of A1 adenosine receptors. It is concluded that LTP develops in about one minute and that the chemistries operating in this period are easily disrupted by an event triggered by adenosine receptors.

The nature and causes of hippocampal long-term potentiation

One of the most fascinating features of the hippocampus is its capacity for plasticity. Long-term potentiation (LTP), a stable facilitation of synaptic potentials after high-frequency synaptic activity, is very prominent in hippocampus and is a leading candidate memory storage mechanism. Here, we discuss the nature and causes of LTP and relate them to endogenous rhythmic neuronal activity patterns and their potential roles in memory. Anatomical studies indicate that LTP is accompanied by postsynaptic structural modifications while pharmacological studies strongly suggest that LTP is not due to an increase in presynaptic transmitter release. In field CA1, LTP induction appears to be triggered by a postsynaptic influx of calcium through NMDA receptor-linked channels. Possible roles of several calcium-sensitive enzyme systems in LTP are discussed and it is argued that activation of a calcium-dependent protease (calpain) could produce the structural changes linked to LTP. Rhythmic bursting activity is highly effective in inducing LTP and it is argued that the endogenous hippocampal theta rhythm plays a role in LTP induction in vivo. Finally, studies indicate that LTP and certain types of memory share a common pharmacology and the use of electrical brain stimulation as a sensory cue suggests that LTP develops when the significance of that cue is learned.

The protease inhibitor leupeptin interferes with the development of LTP in hippocampal slices

The effect of leupeptin, an inhibitor of thiol-proteases, was tested on the induction of long-term potentiation (LTP) in field CA1 of hippocampal slices. Two h of drug application did not produce substantial changes while a greater than 3-h application caused a sizeable reduction in the degree of LTP induced. Leupeptin had no obvious effects on the facilitation of postsynaptic responses occurring within or between the short high frequency bursts used to induce LTP, suggesting that the drug acted on cellular chemistries occurring after the initial physiological events that normally trigger LTP. These results are consistent with the hypothesis that a calcium-activated thiol protease (calpain) is involved in the induction of LTP.

Effects of two ACTH analogs on successive odor discrimination learning in rats

Olfactory discrimination learning has been shown to provide a powerful tool to investigate the mechanisms involved in the formation, storage and retrieval of information in rodent CNS. In the present study we tested the effects of two ACTH analogs, which were previously reported to influence the processes of learning and memory, on various olfactory learning tasks. The ACTH(4-9) analog HOE 427 produced an apparent increase in storage of olfactory information as shown by the difficulty experienced by the animals to rapidly reverse their behavioral responses to previously learned odors. Similarly, the ACTH (4-9) analog ORG 2766 appears to enhance the storage of olfactory information when administered either before or after the learning trials. These data are consistent with the notion that ACTH and related analogs facilitate performance in a variety of learning tasks. In addition, our results suggest possible mechanisms by which some neuroactive peptides might modulate learning and memory processes in the CNS.

Long-term potentiation: persisting problems and recent results

In this paper we discuss recent experimental results pertinent to three unresolved issues regarding the long-term potentiation (LTP) effect: the nature of its enduring substrates, the biochemical mechanisms that produce it, and its potential role in memory. LTP appears to be triggered by a postsynaptic influx of calcium and is associated with alterations in the shape of dendritic spines and probably the formation of new synapses. We discuss the possibility that morphological reorganization also modifies membrane surface chemistry of synaptic elements. Evidence is presented that LTP is not associated with changes in presynaptic calcium currents. Activation of protein kinase C is shown to be insufficient for the induction of LTP, although it may play a modulatory role. The hypothesis that activation of a calcium-sensitive protease (calpain) is pivotal to the establishment of LTP is supported by experiments showing that a calpain inhibitor, leupeptin, blocks LTP. Furthermore, activation of NMDA receptors, an event implicated in LTP induction, is accompanied by calcium-sensitive proteolysis of spectrin, a major dendritic cytoskeletal protein. The finding that stimulation patterns designed to mimic naturally-occurring cell discharge patterns are highly effective for LTP induction greatly strengthens the hypothesis that LTP actually occurs during the encoding of information in cortical systems. Potential contributions of LTP to learning are explored using computer simulations of a simple cortical network.

Anisomycin, an inhibitor of protein synthesis, blocks late phases of LTP phenomena in the hippocampal CA1 region in vitro

Long-term potentiation (LTP) with its extremely long duration has been frequently regarded as an elementary mechanism of information storage in the nervous system or at least as a suitable model for the study of mechanisms underlying functional plasticity and processes of learning and memory formation. Considering the necessity of an increased protein synthesis for memory consolidation and for the maintenance of LTP in granular synapses in vivo it was of interest to determine whether the LTP of the CA1 region of the hippocampus depends on protein synthesis as well. For the solution of this question anisomycin (ANI), a reversible blocker of protein synthesis, was used at a concentration of 20 microM, which blocked the [3H]leucine incorporation in hippocampal slices by at least 85%. It has been shown that in the CA1 region in vitro the maintenance of LTP (i.e. a late phase greater than 5 h) depends on an ongoing protein synthesis. A 3-h treatment with ANI immediately following multiple tetanization resulted in gradually developing loss of field excitatory postsynaptic potential (EPSP) and population spike (PS) potentiation (15 +/- 19% increase of the PS instead of the 96 +/- 14% increase in non-treated control experiments at the 8th h after tetanization). Furthermore, a late PS potentiation (greater than 6 h) of a second non-tetanized pathway to CA1 pyramidal cells has been observed (increase by 64 +/- 18% at the 8th h) for the first time. This potentiation was ANI-sensitive as well and suggests that the maintenance of LTP is dependent on a postsynaptic mechanism.