Throughout our waking lives we are exposed to a continuous stream of experiences. Some of these experiences trigger changes in the strength of connections between neurons in the brain and begin the process of forming memories. However, these initial memory traces are fragile and only a small number will become long-term memories with the potential to last a lifetime. For this transition to occur, the brain must stabilize the memory traces through a process called consolidation.

During consolidation, the brain produces new proteins that strengthen the fragile memory traces. However, if a new experience occurs while an existing memory trace is being consolidated, the new experience could disrupt or even hijack the consolidation process. To avoid this problem, the brain performs most consolidation while we are asleep. But what happens if we wake up while consolidation is taking place? How does the brain prevent events that occur just after awakening from disrupting consolidation?

Levy et al. have now answered this question using a seemingly unlikely subject, the sea slug Aplysia. Sea slugs are capable of basic forms of learning, and their simple nervous systems and large neurons make them convenient to study. Blocking the production of new proteins in sleeping sea slugs prevents the animals from forming long-term memories, confirming that, like us, they do consolidate memories during sleep. Levy et al. now show that exposing sea slugs to new stimuli immediately after they wake up does not trigger the formation of new memories. However, if the slugs were treated with a drug that blocks protein production just beforehand, the new stimuli could trigger memory formation.

These findings show that proteins blocking the formation of new memories prevent an experience upon waking from being effective in producing memory. Removing this block – by inhibiting protein production – allows experiences just after waking to be encoded in memory. This even applies to experiences that are too brief to trigger memory formation in fully awake sea slugs. The next step following on from this work is to identify these memory blocking proteins and to work out how they prevent new memories from forming. A future challenge is to find out is whether the same proteins could ultimately be used to block unwanted memories.

New experiences that occur while a memory is being consolidated interact with the consolidation. The interactions can lead to problems in keeping apart related memories. This problem is partially solved by deferring some of the memory consolidation to sleep, a period in which new experiences are minimized. However, experiences when animals are wakened from sleep are likely to interact with the ongoing consolidation, thereby potentially causing maladaptive memories. We have found that during the sleep phase consolidation, new experiences produce a protein synthesis dependent process that blocks the formation of new memories, thereby preventing the formation of maladaptive memories.

Some of the potential problems that arise from successive experiences have been known for many years. An experience that occurs while a memory is being consolidated may interfere with the consolidation, and partially stops it. Indeed, the process of memory consolidation was discovered over 100 years ago by interference of memory formation as a result of a new learning trial (Lechner et al., 1999). Only much later was consolidation shown to depend on gene transcription and translation (Montarolo et al., 1986; Sweatt, 2010), and dependence on these molecular processes has become the defining feature of consolidation. A second interaction between an experience and an ongoing consolidation has been more recently discovered. An experience that is generally ineffective in producing long-term memory can become effective when it occurs during the consolidation of a memory that arises from another, effective experience. This phenomenon was originally described on a synaptic level, and was called synaptic tagging and capture, since training at synapses that have undergone short-term plasticity produces local tags (Frey and Morris, 1997; Martin et al., 1997). These tags can capture molecular products that are synthesized as a result of a consolidation process that arises from another training. Capture at the tags causes expression of long-term memory at synapses that would otherwise only display short-term memory. More recent studies have demonstrated behavioral tagging that is analogous to the synaptic tagging (Ballarini et al., 2009; de Carvalho Myskiw et al., 2013; Moncada and Viola, 2007; Moncado et al., 2015).

The problem of potentially non-adaptive interactions between memory consolidation and new experiences may be particularly acute when animals sleep. There is now abundant evidence that after completion of the early stages of consolidation which follow an experience, an additional stage of consolidation occurs while animals sleep (Rasch and Born, 2013; Walker and Stickgold, 2006), primarily during the night for diurnally active animals, or during the day for nocturnally active animals. Similar processes may govern consolidation that immediately follows training, and that during sleep. Memory consolidation and cortical plasticity during sleep are also accompanied by, or is dependent on protein synthesis (Grønli et al., 2013; Rasch and Born, 2013; Seibt et al., 2012; Tudor et al., 2016; Walker, 2005). There is also a general increase in cerebral protein synthesis during sleep phases associated with memory consolidation (Ramm and Smith, 1990). In addition, molecular events associated with memory consolidation during sleep are similar to those induced during consolidation following training (e.g., Luo et al., 2013). Consolidation-like processes occur during sleep even in the absence of a previous training trial. Animals and humans sleep every day, even when no overt training occurred in the previous active phase. Proteins supporting consolidation will be synthesized during sleep, even when sleep does not follow overt training (Ramm and Smith, 1990).

As in consolidation that follows training, new experiences during the sleep-phase may cause non-adaptive interaction with the consolidation. A transient experience that would not cause long-term memory could be amplified by the ongoing consolidation, and cause an unwanted long-term memory. The amplification of the experience could then in turn increase the possibility that the new experience will interfere with consolidation of the earlier memory during sleep. One way of preventing interference is to prevent new experiences by losing consciousness. Indeed, the earliest papers (Jenkins and Dallenbach, 1924) showing that sleep improves memory posited that it does so passively, by preventing interference from new experiences. In recent years it has become clear that active processes also strengthen memories while animals sleep (Rasch and Born, 2013; Walker and Stickgold, 2006). However, the presence of active processes promoting consolidation during sleep strengthens the need to limit experiences, or the potential memories elicited by such experiences, while memories are consolidated. Animals and humans often wake from sleep, and experiences during such times should interfere with consolidation. However, paradoxically, training during the inactive phase, when animals are awakened from sleep, is often relatively ineffective in producing memory (Lyons et al., 2005; Michel and Lyons, 2014; Page, 2015; Rawashdeh et al., 2007).

We explored the possibility that training during the sleep phase initiates an active block of learning, as a means of preventing such inappropriate learning from being consolidated, or of interfering with consolidation. If such a block is present, one would predict that if the block is removed, experiences adjacent to sleep should become more effective in initiating long-term memory. Under this condition, experiences that do not cause long-term memory when animals are active should cause long-term memory.

To explore mechanisms by which new experiences during the period when animals generally sleep are prevented from interfering with consolidation, we utilized an associative learning paradigm in Aplysia in which animals learn that a tasty food cannot be swallowed (Botzer et al., 1998; Katzoff et al., 2002; 2006, Lyons et al., 2005; Michel et al., 2013; Susswein et al., 1986). Training produces a gradual decrease in responsiveness to food, and animals stop responding after 10–25 min of training. Training during the active phase of the day produces short, intermediate, long-term (Botzer et al., 1998; Michel et al., 2013), and persistent (Levitan et al., 2012; Schwarz et al., 1991) memories, with the long-term memory dependent on protein synthesis adjacent to the training and persistent memory dependent on a somewhat later round of protein synthesis (Levitan et al., 2010). Training during the inactive, sleep phase of the day is ineffective in producing long-term memory (Lyons et al., 2005). We now report that training during the sleep phase can be made effective by treatment with a short-acting inhibitor of protein synthesis. This treatment permits even experiences that are too brief to cause long-term memory to become effective. Thus, protein synthesis following training has opposite effects on memory resulting from training during the active or the sleep phase, blocking the former while enabling the latter. Our findings indicate that training during the sleep phase initiates an active protein-synthesis dependent process that blocks the formation of new long-term memories. In addition, when the active block is removed, long-term memory formation is no longer dependent on protein synthesis at the time of the training. However, long-term memory formation is dependent on transcription and translation a number of hours after training.

We examined a number of molecular correlates of long-term memory formation during the sleep phase. Changes were examined in the transcription of Aplysia CREB1, CREB2 and C/EBP, molecules that are associated with memory formation (Alberini, 2009). We found no changes in CREB2 expression. Treatments that did and that did not cause memory formation led to increased C/EBP expression, indicating that C/EBP expression is not sufficient for memory formation. By contrast, only a treatment causing long-term memory caused a significant increase in the expression of CREB1, indicating that this increase is a correlate of memory formation.

A previous report (Lyons et al., 2005) showed that training Aplysia during their sleep phase is ineffective in producing long-term memory. Our aim was to test the hypothesis that this training is ineffective, because it is adaptive to prevent new learning during an ongoing process of memory consolidation that occurs during sleep. If the training during sleep phase were effective two problems would arise: (1) the molecular events underlying consolidation could enhance the efficacy of the training at this time, leading to dysfunctional long-term memory of events that would otherwise not be remembered; (2) new learning during sleep phase consolidation could interfere with the consolidation. We hypothesize: (a) training during the sleep phase induces the synthesis of an active blocker of memory formation. If the blocker is removed, we predict: (b) the ongoing consolidation process will enhance the ability of training to produce long-term memory. In addition, if the ongoing consolidation is itself blocked, we predict: (c) training will no longer produce long-term memory. These predictions were tested.

Memory is consolidated during the inactive phase

It was first important to show that Aplysia consolidate memory during sleep. Similar to almost all animals, Aplysia display a circadian rhythm of activity. They spend most of their inactive phase immobile, and relatively unresponsive to external stimuli (Kupfermann, 1968; Strumwasser, 1971). Recent work has shown that immobility during the inactive phase has characteristics of sleep (Vorster et al., 2014). In mammals, memory consolidation occurs in part during the inactive phase, while animals sleep (Rasch and Born, 2013; Walker and Stickgold, 2006). Since consolidation generally depends on protein synthesis (Montarolo et al., 1986; Sweatt, 2010), an effective way to examine whether consolidation occurs during sleep is to block protein synthesis at this time, and then determine whether this treatment blocks memory formation. To this end, Aplysia were trained with inedible food during the active period, and were treated with 10 µM of the protein synthesis inhibitor anisomycin 12 hr later, during the sleep phase. A previous study had already shown that injecting 10 µM anisomycin 10 min before training during the active phase blocked the expression of memory 24 hr later (Levitan et al., 2010). Thus, consolidation occurs during the hours following training. However, 10 µM anisomycin applied several hours after training did not block 24 hr memory (Levitan et al., 2010). Before treatment with anisomycin, we confirmed that all of the animals were immobile, and presumably were asleep. Treatment with 10 µM anisomycin, but not with the vehicle (artificial seawater - ASW) 12 hr after training blocked 24 hr memory (Figure 1A). Thus, after a period in which protein synthesis is not required, 24 hr memory again depends on protein synthesis, during the inactive phase, when Aplysia sleep. This experiment confirms that active memory consolidation occurs during sleep in Aplysia, and that consolidation can be interrupted by blocking protein synthesis.

Figure 1

Download asset Open asset

Figure 1—source data 1

Memory is consolidated, but training is ineffective, during the sleep phase.

https://doi.org/10.7554/eLife.17769.004

Download elife-17769-fig1-data1-v1.xlsx

Training during the inactive phase does not produce long-term memory

Previous studies examined the efficacy of training Aplysia during the inactive, sleep phase (Fernandez et al., 2003; Lyons et al., 2006). Using a number of learning paradigms, including learning with inedible food (Lyons et al., 2005), training during the inactive sleep phase was ineffective in producing memory tested 24 hr later. We confirmed that training with inedible food during the sleep phase (3–6 hr after the change in lighting) does not produce 24 hr memory (Figure 1B1). In this experiment, all animals were inactive and presumably asleep before being awakened and trained. The inability to detect 24 hr memory after training during the sleep period reflects an inability to form a new memory, rather than an inability to retrieve the memory. As shown previously (Lyons et al., 2005), memory is readily retrieved and expressed during the inactive period, as demonstrated by the presence of memory at night 36 hr after training during the active period (Figure 2A2). This experiment indicates that a blocker of memory formation is active during the sleep phase. The blocker prevents experiences that would have given rise to memory during the active phase from doing so during the sleep phase (Figure 1B2).

Figure 2

Download asset Open asset

Figure 2—source data 1

3 min training with anisomycin during the inactive phase produces memory.

https://doi.org/10.7554/eLife.17769.006

Download elife-17769-fig2-data1-v1.xlsx

Brief training during the sleep phase causes long-term memory

Consolidation-like activity occurs during sleep, even in the absence of an overt training during the previous active phase. Is the previous active phase training required for the brief training during the sleep phase with anisomycin to be effective in producing long-term memory? We tested the possibility that a 3 min training during the sleep phase after treatment with 10 µM anisomycin produces long-term memory, even without a previous active phase training. Naïve animals were trained for 3 min. They were injected with 10 µM anisomycin 10 min before the training. Memory was tested 24 hr later. For comparison, memory after training was compared to that before and after a non-abbreviated training session during the active period. Animals tested 24 hr after the brief training with anisomycin during the sleep phase responded like animals that had been previously trained during the active phase (Figure 2A,B). Thus, the combination of 3 factors which each alone is either ineffective in producing long-term memory (training during the sleep phase, training with a protein synthesis inhibitor, training stopped after 3 min), caused memory when combined.

Would the 3 min training during the sleep phase produce long-term memory even without the treatment with 10 µM anisomycin, or is the long-term memory dependent on the anisomycin treatment? Training animals for 3 min during the sleep phase after ASW treatment, rather than after 10 µM anisomycin, produced no memory (Figure 2C). We also tested the possibility that the brief training with anisomycin caused memory that is only expressed during the sleep phase. Animals were tested 12 hr after the brief training with 10 µM anisomycin, or with ASW. The 3 min training produced memory 12 hr later only when training was with 10 µM anisomycin (Figure 2D), but not with ASW (Figure 2E).

The findings suggest that training during the sleep phase initiates a protein synthesis dependent block of long-term memory formation. When the block is removed by the anisomycin treatment, learning is more effective during the sleep phase than during the active phase. Even an experience that is too brief to cause long-term memory during the active phase, a 3 min training, is effective during the sleep phase (Figure 2F). In addition, memory is expressed earlier. It was previously shown (Botzer et al., 1998) that after effective training during the day, long-term memory is expressed only 24 hr after training, and no memory is expressed at 12 hr after the training. However, 12 hr after a 3 min training with 10 µM anisomycin during the inactive phase, memory is expressed.

We have suggested that the 3 min training during sleep is effective in producing memory because of the ongoing consolidation during sleep, and the lack of a protein synthesis dependent blocker of memory formation. However, it is conceivable that the 10 µM anisomycin treatment per se makes the 3 min training effective, even without the consolidation process. Although the 3 min training alone during the active phase is ineffective in producing memory (Levitan et al., 2010), the combination of a 3 min training and treatment with 10 µM anisomycin might produce memory, even during the active phase. We tested this possibility. However, we found that the 3 min training 10 min after 10 µM anisomycin treatment during the active phase was ineffective in producing long-term memory (data not shown). Thus, this combination is effective in producing memory only during the sleep phase, presumably because of the ongoing consolidation process.

Does the ongoing consolidation in the presence of 10 µM anisomycin facilitate only a brief training, or will it also facilitate a longer training, which is effective during the active phase, but not during the sleep phase? We examined this possibility by training animals during the sleep phase until they stopped responding (full training), and testing memory 24 hr later. They were treated 10 min before training with either 10 µM anisomycin Figure 3A), or with ASW Figure 3B). A third group was treated with 10 µM anisomycin, but was not trained (Figure 3C). Only the animals that were trained after treatment with 10 µM anisomycin displayed 24 hr memory. This experiment shows that both an abbreviated and a longer training session are effective in producing memory after treatment with 10 µM anisomycin. In addition it shows that anisomycin alone during the sleep phase is ineffective in producing memory without training.

Figure 3

Download asset Open asset

Figure 3—source data 1

Full training during the sleep phase with anisomycin causes long-term memory.

https://doi.org/10.7554/eLife.17769.008

Download elife-17769-fig3-data1-v1.xlsx

Dual function of protein synthesis during sleep

Injecting 10 µM anisomycin during the sleep phase allowed training procedures that would have been ineffective in producing long-term memory to become effective. Our interpretation is that a protein-synthesis dependent blocker of memory is inhibited by the anisomycin, and therefore is not present to suppress memory formation. However, consolidation leading to long-term memory generally requires protein synthesis. Even if it inhibited synthesis of the blocker, why was long-term memory formed when protein synthesis was blocked? We suggest that the proteins having a role in memory consolidation are either specifically expressed during sleep, or are expressed more strongly during sleep. These proteins are expressed even in the absence of an explicit training during the previous active phase, and their presence substitutes for the proteins synthesized just after training during the active phase. These proteins are synthesized throughout the sleep phase, and in our experiments they were already expressed when animals were trained later in the sleep phase. Their expression explains why training can produce long-term memory with a protein synthesis inhibitor.

This hypothesis suggests that temporally separated processes of protein synthesis explain our findings. Protein synthesis functioning in consolidation begins within the first hours of the sleep phase, and these proteins can enhance memory formation from new experiences during the sleep phase, explaining why even a brief training can produce long-term memory. However, the new experiences themselves initiate an additional process of protein synthesis that blocks the formation of new memories, explaining why training during the sleep phase is generally ineffective in causing long-term memory. Treatment with 10 µM anisomycin just before training blocks synthesis of the memory blocker, leaving intact the previously synthesized proteins associated with consolidation that enhance memory formation (see Figure 5E).

We tested this hypothesis by separately blocking the protein synthesis associated with consolidation, and the protein synthesis that blocks memory formation. We predict that blocking the protein synthesis associated with consolidation will block long-term memory formation during the sleep phase, even if the memory blocker induced by training is also not present. Since consolidation is likely to occur throughout the sleep phase, a short-acting blocker of protein synthesis that is applied early in the sleep phase may block the proteins that support consolidation, but will no longer be present at a later time, when animals are trained. In intact animals, the length of time that anisomycin effectively blocks protein synthesis depends on the concentration used. Smaller doses of anisomycin transiently block protein synthesis, whereas larger doses have longer-lasting effects (Flood et al., 1973). To find a transiently active dosage of anisomycin, we examined the effects of different concentrations at different times before a full training session during the active phase. We found that 10 µM anisomycin was effective in blocking memory formation 1 hr before training (Figure 4A), but was not effective 2 hr before training (Figure 4B). By contrast, a three times larger dose (30 µM anisomycin) was still effective in blocking memory formation when injected 2.5 hr before training (Figure 4C,D). We also tested the effect of a dosage of anisomycin that is a third of that used previously (3.3 µM in place of 10 µM anisomycin), and found that this concentration injected 10 min before training is too low to inhibit long-term memory formation (Figure 4E). Thus, 10 µM anisomycin is close to threshold in blocking protein synthesis, and its effect is relatively short-lasting. Injection of this concentration early in the sleep phase, 2 hr or more before training, will not affect synthesis of the proposed memory blocker, but will affect synthesis of proteins that support consolidation during sleep.

Figure 4

Download asset Open asset

Figure 4—source data 1

Length of protein synthesis inhibition is concentration dependent.

https://doi.org/10.7554/eLife.17769.010

Download elife-17769-fig4-data1-v1.xlsx

Aplysia were treated with 10 µM anisomycin 1 to 3 hr after the change in lighting that signaled the start of the sleep phase, but 2 hr before training began (Figure 5A). A second group of animals was treated with 10 µM anisomycin 30 min before the training (Figure 5B). Animals in both groups received a second injection of 10 µM anisomycin 10 min before a 3 min training. Our expectation is that the treatment 2 hr before training will block the synthesis of proteins associated with consolidation during sleep, without affecting the synthesis of the memory blocker. Thus, if sleep-phase proteins functioning in consolidation are required for long-term memory formation during the sleep phase, this treatment should block memory formation. By contrast, the injection 30 min before training will leave intact proteins that have already been synthesized in the early hours of the sleep phase, before the anisomycin was injected.

Figure 5

Download asset Open asset

Figure 5—source data 1

Block of sleep-time expression of proteins blocks memory formation.

https://doi.org/10.7554/eLife.17769.012

Download elife-17769-fig5-data1-v1.xlsx

Treatment with 10 µM anisomycin 2 hr before training blocked the formation of long-term memory (Figure 5A), as shown by a significant increase in the response time 24 hr later, with respect to animals that were treated with anisomycin only once, 10 min before training (Figure 5C). By contrast, treatment with 10 µM anisomycin 30 min before training did not block the formation of long-term memory (Figure 5B), as shown by a significant decrease in response, with respect to the response time of naïve animals trained during the active phase (Figure 5D). These findings support the hypothesis that proteins synthesized during sleep substitute for the proteins that are generally synthesized as a result of long term training, and that presence of these proteins is required for long-term memory formation during the sleep phase (Figure 5E). This experiment also indicates that the double injection of anisomycin per se did not block memory formation, since animals injected 30 min before training, and again 10 min before training, displayed long-term memory.

These findings show that 10 µM anisomycin 2 hr before training has opposite effects during the active and inactive phases of activity. During the active phase, this treatment does not block long-term memory formation, presumably because its concentration within the animal decreases with time, and becomes too low to block the protein synthesis initiated by the training. By contrast, during the sleep phase it blocks long-term memory formation, presumably because it blocks the background expression of molecules that support memory consolidation during sleep.

Memory formation during the sleep phase is dependent on transcription and later translation

A hallmark of long-term memory is dependence on gene transcription and translation (Montarolo et al., 1986). However, we have shown that memory formation at night only occurs when protein synthesis is blocked just preceding the training, although protein synthesis in the hours preceding training is necessary for long-term memory formation. Consolidation is a slow process, and is dependent on a number of waves of mRNA and protein synthesis (Barzilai et al., 1989). The effects of 10 µM anisomycin are relatively short-lasting (see Figure 3). After training, one might still expect that long-term memory formation will depend on molecular signals initiating transcription of mRNAs, and their subsequent translation, after the effects of anisomycin present during or just after training have worn off. To determine whether memory formation is dependent on transcription, we examined the effects of the reversible transcription blocker 5,6-Dichlorobenzimidazole 1-β-D-ribofuranoside (DRB) on memory formation. After treating animals with anisomycin, they were trained 10 min later for 3 min, and then treated with DRB. Memory was then tested 24 hr later.

DRB is not water soluble, and DMSO injection inhibits feeding in intact, behaving Aplysia (unpublished observation). We therefore used ethanol to dissolve DRB. To rule out the possibility that ethanol, or ethanol with DRB, damages animals and affects the ability to respond to food, we first examined the effects of ethanol alone and of DRB dissolved in ethanol on learning 24 hr after the injections, when their effects had worn off, and on memory 48 hr after the injection. Neither ethanol (Figure 6A1) nor ethanol plus DRB (Figure 6A2) affected the ability to train animals 24 hr after the treatment, or to obtain memory 24 hr after the training.

Figure 6

Download asset Open asset

Figure 6—source data 1

Learning during the inactive phase is transcription dependent.

https://doi.org/10.7554/eLife.17769.014

Download elife-17769-fig6-data1-v1.xlsx

After a 3 min training at night with anisomycin, animals were injected with either ethanol (Figure 6B2), or DRB dissolved in ethanol (Figure 6B3). A control group also examined memory 24 hr after 10 µM anisomycin treatment at night, without subsequent training, followed by ethanol plus DRB (Figure 6B1). Significant memory was shown in animals treated with anisomycin alone, whereas no memory was shown in animals treated with anisomycin plus DRB. Thus, memory 24 hr after a 3 min training at night with anisomycin is dependent on transcription, and presumably also on later rounds of translation that occur after the effects of anisomycin have worn off (Figure 6C).

As an indicator of whether a later round of translation is required after training, we used a higher concentration of anisomycin (30 µM in place of 10 µM), and then trained animals for 3 min. The higher concentration presumably has a longer-lasting effect on protein synthesis (Flood et al., 1973), accounting for a longer-lasting inhibition of memory formation when Aplysia were trained during the active phase (see Figure 5C). Training with 30 µM anisomycin was ineffective in producing memory 24 hr later (Figure 7A2), presumably because it not only blocks protein synthesis transiently at the time of the training, but also blocks later rounds of synthesis that follow transcription e.g., Barzilai et al., 1989) (Figure 6C).

Figure 7

Download asset Open asset

Figure 7—source data 1

Dissociation between memory and C/EBP expression.

https://doi.org/10.7554/eLife.17769.016

Download elife-17769-fig7-data1-v1.xlsx

Other effects of anisomycin are not relevant to memory formation

In addition to blocking protein synthesis, anisomycin also increases MAP-kinase activity, and thereby causes a large increase in CCAAT enhancer-binding protein (C/EBP) mRNA expression (Alberini et al., 1994; Cano et al., 1994). Increased C/EBP is an early step in the formation of memory in Aplysia (Alberini et al., 1994), and in other animals (Alberini, 2009). In addition, training with inedible food during the active phase caused a significant (5–10 fold) increase in C/EBP mRNA expression in the buccal ganglia (Levitan et al., 2008), with respect to C/EBP mRNA expression in controls whose lips had been stimulated with food, a procedure that does not lead to memory formation (Schwarz et al., 1988). In another Aplysia learning paradigm, a tyrosine phosphatase inhibitor was used to increase MAP-kinase activity, and thereby also presumably increase C/EBP mRNA expression. This treatment alone was not effective in allowing long-term memory to be formed after sleep phase training (Lyons et al., 2006). However, when this treatment was combined with a histone deacetylation inhibitor, the combination of both was effective in allowing long-term memory to be formed by sleep phase training (Lyons et al., 2006). These finding raise the possibility that anisomycin during the sleep phase permits memory formation because it increases C/EBP expression, rather than because it inhibits the synthesis of a protein that actively blocks memory formation. We tested this possibility.

Using quantitative real time polymerase chain reaction (qRT-PCR), we found that injection of 10 µM anisomycin produced a large increase in the expression of C/EBP mRNA (see Figure 7B). As shown above (Figure 3C), 10 µM anisomycin treatment alone, with no training, has no effect on subsequent memory, indicating that increased C/EBP mRNA expression per se does not produce long-term memory. However, it is possible that increased C/EBP mRNA expression underlies the ability of a brief training during the sleep phase to produce long-term memory, rather than the block of protein synthesis produced by 10 µM anisomycin. We examined this possibility.

Injection of 10 µM anisomycin 10 min before training during the active phase blocks memory, but during the sleep phase it facilitates memory. If increased C/EBP is responsible for the facilitation of long-term memory during sleep, there should be a difference in the C/EBP expression in the buccal ganglia after 10 µM anisomycin treatment during the active and sleep phases. Using qRT-PCR, we measured C/EBP mRNA expression in the buccal ganglia 2 hr after anisomycin treatment during both the active and the sleep phases. Expression was compared to that in response to animals injected with ASW, during the active and the sleep phases respectively. During both the active and sleep phases, there were large (over 100-fold) increases in C/EBP mRNA expression in response to 10 µM anisomycin (Figure 7B), with no significant difference in expression between the two phases. Thus, C/EBP is increased when anisomycin blocks memory, as well as when it is required to form long-term memory.

We also tested the effects of a 3.3 µM dose of anisomycin. We found that 3.3 µM anisomycin injected 10 min before training during the active phase did not block memory formation (see Figure 4E). In addition, it did not facilitate memory formation during the inactive phase (Figure 7A3), presumably because the dose is too low to block protein synthesis sufficiently. However, 3.3 µM anisomycin still caused a large (over 80-fold) increase in C/EBP expression (Figure 7B). The difference between C/EBP production during the sleep phase in response to 10 µM or 3.3 µM anisomycin was not significant. Thus, there is a dissociation between the effect of anisomycin on protein synthesis, which is likely to be critical for its effects on memory, and the effect of anisomycin on C/EBP, whose increase is not systematically associated with the ability to form long-term memory.

Although increases in C/EBP caused by anisomycin treatment could not explain the ability to produce long-term memory, it is possible that increases in C/EBP are nonetheless correlated with training procedures that are effective in producing long-term memory. If this is so, training with 10 µM anisomycin at night should produce an increase in C/EBP over that produced by anisomycin alone, whereas training with ASW at night should produce no increase in C/EBP expression over that produced by ASW alone. We tested this possibility by measuring C/EBP expression 2 hr after a 3 min training with ASW or with 10 µM anisomycin. A 3 min training with ASW causes no long-term memory (Figure 2C), but nonetheless caused a 7-fold increase in C/EBP expression (Figure 8A1), which is similar to the increase in C/EBP expression caused by training during the day (Levitan et al., 2008). A 3 min training after treatment with 10 µM anisomycin also caused a significant increase in C/EBP expression (Figure 8A2). Note that the increased C/EBP expression after anisomycin treatment is 2 orders of magnitude larger than that after ASW treatment, because of the effect of the anisomycin. However, in both conditions training caused an increase in C/EBP expression. These data indicate that increased C/EBP mRNA expression is not a necessary correlate of memory formation.

Figure 8

Download asset Open asset

Figure 8—source data 1

Increased C/EBP expression is a correlate of training, not of memory formation.

https://doi.org/10.7554/eLife.17769.018

Download elife-17769-fig8-data1-v1.xlsx

The data above suggest that the increase in C/EBP expression seen after training, either during the active phase (Levitan et al., 2008), when training causes memory, or during the inactive phase in which training does not cause memory, is a correlate of the training procedure per se, rather than being a correlate of memory formation. Additional evidence also supported this suggestion. When Aplysia fasciata are trained during their active phase, but in isolation from conspecifics, no long-term memory is produced (Schwarz et al., 1998). We found that both training in isolation, and training in the presence of conspecifics produced significant increases in C/EBP expression (Figure 8B). Thus, increased C/EBP expression is a correlate of the training procedure, rather than being a correlate of memory formation.

In addition to increasing C/EBP expression, anisomycin could increase the expression of additional genes associated with memory formation, and thereby induce memory during training in the sleep phase. We examined the possibility that the expression of two additional genes associated with memory formation, cAMP- response element-binding proteins 1 and 2 (CREB1 and CREB2) (Lee et al., 2008), are changed by doses of anisomycin that facilitate memory formation. Since CREB2 is a repressor of memory formation (Bartsch et al., 1995), regulation of CREB2 might be associated with the block of memory during the sleep phase. As noted above, 3.3 µM anisomycin does not block memory formation during the active phase, or facilitate memory formation during the inactive phase, whereas 10 µM anisomycin is effective in blocking memory during the active phase, and enhancing memory formation during the inactive phase. If 10 µM anisomycin were to significantly change the expression of one of these genes, over that expressed after treatment with 3.3 µM anisomycin, this change could underlie the ability of training during the inactive phase to cause long-term memory. Unlike the effect of anisomycin on C/EBP, neither 3.3 µM nor 10 µM anisomycin during the inactive phase produced significant increases in the expression of CREB1 or of CREB2, with respect to expression after treatment with ASW, although both concentrations produced moderate increases that approached significance (Figure 9A, and legend – note that the bars are normalized to expression after ASW, which is 100%). There were no significant differences in expression between treatment with 3.3 µM and 10 µM anisomycin (Figure 9A), effectively ruling out changed expression of either of these two genes as a result of anisomycin treatment alone from being the underlying cause of memory formation with anisomycin treatment during the inactive phase.

Figure 9

Download asset Open asset

Figure 9—source data 1

Increased CREB1 expression after training is a correlate of memory.

https://doi.org/10.7554/eLife.17769.020

Download elife-17769-fig9-data1-v1.xlsx

Molecular correlates of sleep phase memory formation

We examined whether changes in gene expression after training are correlates of training, or of memory formation, after a 3 min training with anisomycin during the sleep phase. As shown above, C/EBP expression is significantly increased by a 3 min training at night, whether or not the training produces memory. Thus, its increase is a correlate of training, but not a correlate of memory formation. We also examined changes in the expression of CREB1 or CREB2 in the buccal ganglia 2 hr after a 3 min training during the inactive phase after treatment with either ASW or anisomycin. CREB1, but not CREB2 expression was significantly increased only after treatment with anisomycin (Figure 9B1 and B2), indicating that increased CREB1 expression is a molecular correlate of long-term memory formation in our learning paradigm. Since C/EBP expression is downstream from CREB1 expression (Alberini et al., 1994; Shifrin and Anderson, 1999), and there is no increase in CREB1 expression in ineffective training, transcription factors other than CREB1, or the CREB1 protein that is already present and phosphorylated, must be driving expression of C/EBP in the buccal ganglia in ineffective training. Thus, the combination of anisomycin and training during the inactive phase has an effect on a specific mRNA whose product supports memory formation, CREB1. By contrast, CREB2 expression was not changed by training after treatment with either ASW or with anisomycin, indicating that its expression is not altered by the training procedure, when the training does or does not produce long-term memory.

Experiments in which related training trials follow one another have provided major insights into the nature of memory consolidation. Indeed, consolidation was first identified by the finding that a training trial occurring during consolidation may interfere with the consolidation (Lechner et al., 1999). More recently, it has been found that training protocols leading to short-term memory tag synapses that are to be consolidated (Frey and Morris, 1997; Martin et al., 1997). These tags can cause a training that would only cause short-term memory to capture a consolidation process that was initiated by another training session which causes long-term memory (Ballarini et al., 2009; de Carvalho Myskiw et al., 2013; Moncada and Viola, 2007).

The complex interactions that may occur between an ongoing process of memory consolidation and a second, related training potentially causes problems. There are presumably good adaptive reasons why some experiences are consolidated, and cause long-term memories, whereas others are not, and cause only short-term memory, or even no memory at all. The ability of an experience to disrupt or to capture an ongoing consolidation may therefore be maladaptive.

One potential solution to the problem of maladaptive interactions between different experiences is to defer a major part of the consolidation to the sleep phase. New experiences will rarely occur during sleep, and therefore will not interfere with the consolidation. However, consolidation is unlikely to be instantly turned off when animals are awakened during the sleep phase, and non-adaptive interactions between new experiences and consolidation may then occur. In this paper, we have shown that new experiences when animals are awakened from sleep can indeed capture the ongoing consolidation process, and that an active protein synthesis dependent process blocks such capture.

When animals are trained during the active phase, protein synthesis-dependent consolidation processes occur in the hours after training, and again during the inactive phase (Figure 1A), when Aplysia sleep (e.g., Vorster et al., 2014). As in many other animals (Rasch and Born, 2013; Walker and Stickgold, 2006), memories are consolidated during sleep (Figure 1A). Blocking protein synthesis in either of these two periods blocks long-term memory formation. Even a relatively brief-acting blocker, such as 10 µM anisomycin, effectively blocks consolidation when applied either during the hour preceding training (Figure 4A), or during the sleep phase, 12 hr after training (Figure 1A).

When animals are trained during the sleep phase, no long-term memory is seen (Figure 1B), despite the presence of sleep-phase consolidation products, because memory formation is actively blocked. Treatment with 10 µM anisomycin removes the block, and allows training during the sleep phase to form long-term memory (Figures 1C, 2 and 3). Although translation immediately following training is blocked by the anisomycin, proteins supporting molecular processes underlying memory during sleep are present, because they have been synthesized in the hours preceding the training. These substitute for the translation generally elicited by training. Blocking the expression of these proteins by treatment with 10 µM anisomycin 2 hr before training blocks long-term memory (Figure 5). Memory after training during the sleep phase is dependent on transcription and translation following training, since the transcription inhibitor DRB (Figure 6B), or a longer-lasting block of protein synthesis (30 µM anisomycin), block memory formation (Figure 7A2).

The ability of a training session to induce memory during the sleep phase only in the presence of anisomycin raises two questions, which will be discussed separately below: (1) why is memory formation during the sleep phase not blocked by anisomycin, as it is during the active phase? (2) Why does long-term memory after training during the sleep phase occur ONLY after treatment with anisomycin?

1. 1. Alberini CM
   2. Ghirardi M
   3. Metz R
   4. Kandel ER

   (1994) C/EBP is an immediate-early gene required for the consolidation of long-term facilitation in Aplysia

   Cell 76:1099–1114.

   https://doi.org/10.1016/0092-8674(94)90386-7

   • PubMed
   • Google Scholar
2. 1. Alberini CM

   (2009) Transcription factors in long-term memory and synaptic plasticity

   Physiological Reviews 89:121–145.

   https://doi.org/10.1152/physrev.00017.2008

   • PubMed
   • Google Scholar
3. 1. Arzi A
   2. Shedlesky L
   3. Ben-Shaul M
   4. Nasser K
   5. Oksenberg A
   6. Hairston IS
   7. Sobel N

   (2012) Humans can learn new information during sleep

   Nature Neuroscience 15:1460–1465.

   https://doi.org/10.1038/nn.3193

   • PubMed
   • Google Scholar
4. 1. Ballarini F
   2. Moncada D
   3. Martinez MC
   4. Alen N
   5. Viola H

   (2009) Behavioral tagging is a general mechanism of long-term memory formation

   PNAS 106:14599–14604.

   https://doi.org/10.1073/pnas.0907078106

   • PubMed
   • Google Scholar
5. 1. Barnes DC
   2. Wilson DA

   (2014) Slow-wave sleep-imposed replay modulates both strength and precision of memory

   Journal of Neuroscience 34:5134–5142.

   https://doi.org/10.1523/JNEUROSCI.5274-13.2014

   • PubMed
   • Google Scholar
6. 1. Bartsch D
   2. Casadio A
   3. Karl KA
   4. Serodio P
   5. Kandel ER

   (1998) CREB1 encodes a nuclear activator, a repressor, and a cytoplasmic modulator that form a regulatory unit critical for long-term facilitation

   Cell 95:211–223.

   https://doi.org/10.1016/S0092-8674(00)81752-3

   • PubMed
   • Google Scholar
7. 1. Bartsch D
   2. Ghirardi M
   3. Skehel PA
   4. Karl KA
   5. Herder SP
   6. Chen M
   7. Bailey CH
   8. Kandel ER

   (1995) Aplysia CREB2 represses long-term facilitation: relief of repression converts transient facilitation into long-term functional and structural change

   Cell 83:979–992.

   https://doi.org/10.1016/0092-8674(95)90213-9

   • PubMed
   • Google Scholar
8. 1. Barzilai A
   2. Kennedy TE
   3. Sweatt JD
   4. Kandel ER

   (1989) 5-HT modulates protein synthesis and the expression of specific proteins during long-term facilitation in Aplysia sensory neurons

   Neuron 2:1577–1586.

   https://doi.org/10.1016/0896-6273(89)90046-9

   • PubMed
   • Google Scholar
9. 1. Botzer D
   2. Markovich S
   3. Susswein AJ

   (1998)

   Multiple memory processes following training that a food is inedible in Aplysia

   Learning & Memory 5:204–219.

   • PubMed
   • Google Scholar
10. 1. Cano E
    2. Hazzalin CA
    3. Mahadevan LC

    (1994) Anisomycin-activated protein kinases p45 and p55 but not mitogen-activated protein kinases ERK-1 and -2 are implicated in the induction of c-fos and c-jun

    Molecular and Cellular Biology 14:7352–7362.

    https://doi.org/10.1128/MCB.14.11.7352

    • PubMed
    • Google Scholar
11. 1. de Carvalho Myskiw J
    2. Benetti F
    3. Izquierdo I

    (2013) Behavioral tagging of extinction learning

    Proceedings of the National Academy of Sciences 110:1071–1076.

    https://doi.org/10.1073/pnas.1220875110

    • PubMed
    • Google Scholar
12. 1. Dhawan P
    2. Bell A
    3. Kumar A
    4. Golden C
    5. Mehta KD

    (1999)

    Critical role of p42/44(MAPK) activation in anisomycin and hepatocyte growth factor-induced LDL receptor expression: activation of Raf-1/Mek-1/p42/44(MAPK) cascade alone is sufficient to induce LDL receptor expression

    Journal of Lipid Research 40:1911–1919.

    • PubMed
    • Google Scholar
13. 1. Diekelmann S
    2. Büchel C
    3. Born J
    4. Rasch B

    (2011) Labile or stable: opposing consequences for memory when reactivated during waking and sleep

    Nature Neuroscience 14:381–386.

    https://doi.org/10.1038/nn.2744

    • PubMed
    • Google Scholar
14. 1. Dudai Y

    (2004) The neurobiology of consolidations, or, how stable is the engram?

    Annual Review of Psychology 55:51–86.

    https://doi.org/10.1146/annurev.psych.55.090902.142050

    • PubMed
    • Google Scholar
15. 1. Fernandez RI
    2. Lyons LC
    3. Levenson J
    4. Khabour O
    5. Eskin A

    (2003) Circadian modulation of long-term sensitization in Aplysia

    PNAS 100:14415–14420.

    https://doi.org/10.1073/pnas.2336172100

    • PubMed
    • Google Scholar
16. 1. Flood JF
    2. Rosenzweig MR
    3. Bennett EL
    4. Orme AE

    (1973) The influence of duration of protein synthesis inhibition on memory

    Physiology & Behavior 10:555–562.

    https://doi.org/10.1016/0031-9384(73)90221-7

    • PubMed
    • Google Scholar
17. 1. Frey U
    2. Morris RG

    (1997) Synaptic tagging and long-term potentiation

    Nature 385:533–536.

    https://doi.org/10.1038/385533a0

    • PubMed
    • Google Scholar
18. 1. Gev S
    2. Achituv Y
    3. Susswein AJ

    (1984) Seasonal determinants of the life cycle in two species of Aplysia found in shallow waters along the Mediterranean coast of Israel

    Journal of Experimental Marine Biology and Ecology 74:67–83.

    https://doi.org/10.1016/0022-0981(84)90038-8

    • Google Scholar
19. 1. Ghirardi M
    2. Montarolo PG
    3. Kandel ER

    (1995) A novel intermediate stage in the transition between short- and long-term facilitation in the sensory to motor neuron synapse of aplysia

    Neuron 14:413–420.

    https://doi.org/10.1016/0896-6273(95)90297-X

    • PubMed
    • Google Scholar
20. 1. Grønli J
    2. Soulé J
    3. Bramham CR

    (2013) 2014. Sleep and protein synthesis-dependent synaptic plasticity: impacts of sleep loss and stress

    Frontiers in Behavioral Neuroscience 7:224.

    https://doi.org/10.3389/fnbeh.2013.00224

    • PubMed
    • Google Scholar
21. 1. Hu JY
    2. Levine A
    3. Sung YJ
    4. Schacher S

    (2015) cJun and CREB2 in the postsynaptic neuron contribute to persistent long-term facilitation at a behaviorally relevant synapse

    Journal of Neuroscience 35:386–395.

    https://doi.org/10.1523/JNEUROSCI.3284-14.2015

    • PubMed
    • Google Scholar
22. 1. Jenkins JG
    2. Dallenbach KM

    (1924) Obliviscence during Sleep and Waking

    The American Journal of Psychology 35:605–612.

    https://doi.org/10.2307/1414040

    • Google Scholar
23. 1. Kandel ER

    (2012) The molecular biology of memory: cAMP, PKA, CRE, CREB-1, CREB-2, and CPEB

    Molecular Brain 5:14.

    https://doi.org/10.1186/1756-6606-5-14

    • PubMed
    • Google Scholar
24. 1. Katzoff A
    2. Ben-Gedalya T
    3. Hurwitz I
    4. Miller N
    5. Susswein YZ
    6. Susswein AJ

    (2006) Nitric oxide signals that aplysia have attempted to eat, a necessary component of memory formation after learning that food is inedible

    Journal of Neurophysiology 96:1247–1257.

    https://doi.org/10.1152/jn.00056.2006

    • PubMed
    • Google Scholar
25. 1. Katzoff A
    2. Ben-Gedalya T
    3. Susswein AJ

    (2002)

    Nitric oxide is necessary for multiple memory processes after learning that a food is inedible in aplysia

    Journal of Neuroscience  22:9581–9594.

    • PubMed
    • Google Scholar
26. 1. Krishnan HC
    2. Gandour CE
    3. Ramos J
    4. Wrinkle MC
    5. Sanchez-Pacheco JJ
    6. Lyons LC

    (2016)

    Acute sleep deprivation blocks short- and long-term operant memory in Aplysia

    Sleep 39:sp-00313-16.

    • Google Scholar
27. 1. Kupfermann I

    (1968) A circadian locomotor rhythm in Aplysia Californica

    Physiology & Behavior 3:179–181.

    https://doi.org/10.1016/0031-9384(68)90051-6

    • Google Scholar
28. 1. Lechner HA
    2. Squire LR
    3. Byrne JH

    (1999)

    100 years of consolidation--remembering Müller and Pilzecker

    Learning & Memory 6:77–87.

    • PubMed
    • Google Scholar
29. 1. Lee JA
    2. Kim HK
    3. Kim KH
    4. Han JH
    5. Lee YS
    6. Lim CS
    7. Chang DJ
    8. Kubo T
    9. Kaang BK

    (2001) Overexpression of and RNA interference with the CCAAT enhancer-binding protein on long-term facilitation of Aplysia sensory to motor synapses

    Learning & Memory 8:220–226.

    https://doi.org/10.1101/lm.40201

    • PubMed
    • Google Scholar
30. 1. Lee YS
    2. Bailey CH
    3. Kandel ER
    4. Kaang BK

    (2008) Transcriptional regulation of long-term memory in the marine snail Aplysia

    Molecular Brain 1:3.

    https://doi.org/10.1186/1756-6606-1-3

    • PubMed
    • Google Scholar
31. 1. Levitan D
    2. Lyons LC
    3. Perelman A
    4. Green CL
    5. Motro B
    6. Eskin A
    7. Susswein AJ

    (2008) Training with inedible food in Aplysia causes expression of C/EBP in the buccal but not cerebral ganglion

    Learning & Memory 15:412–416.

    https://doi.org/10.1101/lm.970408

    • PubMed
    • Google Scholar
32. 1. Levitan D
    2. Saada-Madar R
    3. Teplinsky A
    4. Susswein AJ

    (2012) Localization of molecular correlates of memory consolidation to buccal ganglia mechanoafferent neurons after learning that food is inedible in Aplysia

    Learning & Memory 19:503–512.

    https://doi.org/10.1101/lm.026393.112

    • PubMed
    • Google Scholar
33. 1. Levitan D
    2. Twitto R
    3. Levy R
    4. Lyons LC
    5. Susswein AJ

    (2010) A brief retraining regulates the persistence and lability of a long-term memory

    Learning & Memory 17:402–406.

    https://doi.org/10.1101/lm.1820010

    • PubMed
    • Google Scholar
34. 1. Liu RY
    2. Cleary LJ
    3. Byrne JH

    (2011) The requirement for enhanced CREB1 expression in consolidation of long-term synaptic facilitation and long-term excitability in sensory neurons of Aplysia

    Journal of Neuroscience 31:6871–6879.

    https://doi.org/10.1523/JNEUROSCI.5071-10.2011

    • PubMed
    • Google Scholar
35. 1. Livak KJ
    2. Schmittgen TD

    (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method

    Methods 25:402–408.

    https://doi.org/10.1006/meth.2001.1262

    • PubMed
    • Google Scholar
36. 1. Luo J
    2. Phan TX
    3. Yang Y
    4. Garelick MG
    5. Storm DR

    (2013) Increases in cAMP, MAPK activity, and CREB phosphorylation during REM sleep: implications for REM sleep and memory consolidation

    Journal of Neuroscience 33:6460–6468.

    https://doi.org/10.1523/JNEUROSCI.5018-12.2013

    • PubMed
    • Google Scholar
37. 1. Lyons LC
    2. Collado MS
    3. Khabour O
    4. Green CL
    5. Eskin A

    (2006) The circadian clock modulates core steps in long-term memory formation in Aplysia

    Journal of Neuroscience 26:8662–8671.

    https://doi.org/10.1523/JNEUROSCI.2307-06.2006

    • PubMed
    • Google Scholar
38. 1. Lyons LC
    2. Rawashdeh O
    3. Katzoff A
    4. Susswein AJ
    5. Eskin A

    (2005) Circadian modulation of complex learning in diurnal and nocturnal Aplysia

    PNAS 102:12589–12594.

    https://doi.org/10.1073/pnas.0503847102

    • PubMed
    • Google Scholar
39. 1. Martin KC
    2. Casadio A
    3. Zhu H
    4. Yaping E
    5. Rose JC
    6. Chen M
    7. Bailey CH
    8. Kandel ER

    (1997) Synapse-specific, long-term facilitation of aplysia sensory to motor synapses: a function for local protein synthesis in memory storage

    Cell 91:927–938.

    https://doi.org/10.1016/S0092-8674(00)80484-5

    • PubMed
    • Google Scholar
40. 1. Michel M
    2. Gardner JS
    3. Green CL
    4. Organ CL
    5. Lyons LC

    (2013) Protein phosphatase-dependent circadian regulation of intermediate-term associative memory

    Journal of Neuroscience 33:4605–4613.

    https://doi.org/10.1523/JNEUROSCI.4534-12.2013

    • PubMed
    • Google Scholar
41. 1. Michel M
    2. Lyons LC

    (2014) Unraveling the complexities of circadian and sleep interactions with memory formation through invertebrate research

    Frontiers in Systems Neuroscience 8:133.

    https://doi.org/10.3389/fnsys.2014.00133

    • PubMed
    • Google Scholar
42. 1. Moncada D
    2. Viola H

    (2007) Induction of long-term memory by exposure to novelty requires protein synthesis: evidence for a behavioral tagging

    Journal of Neuroscience 27:7476–7481.

    https://doi.org/10.1523/JNEUROSCI.1083-07.2007

    • PubMed
    • Google Scholar
43. 1. Moncado D
    2. Ballarini F
    3. Martinez C
    4. Viola H

    (2015)

    Synaptic Tagging and Capture: From Synapses to Behavior Springer

    231–259, The behavioral tagging hypothesis and its implications for long-term memory formation, Synaptic Tagging and Capture: From Synapses to Behavior Springer, 10.1007/978-1-4939-1761-7_14.

    • Google Scholar
44. 1. Montarolo PG
    2. Goelet P
    3. Castellucci VF
    4. Morgan J
    5. Kandel ER
    6. Schacher S

    (1986) A critical period for macromolecular synthesis in long-term heterosynaptic facilitation in Aplysia

    Science 234:1249–1254.

    https://doi.org/10.1126/science.3775383

    • PubMed
    • Google Scholar
45. 1. Nader K
    2. Schafe GE
    3. Le Doux JE

    (2000) Fear memories require protein synthesis in the amygdala for reconsolidation after retrieval

    Nature 406:722–726.

    https://doi.org/10.1038/35021052

    • PubMed
    • Google Scholar
46. 1. Nader K

    (2003) Memory traces unbound

    Trends in Neurosciences 26:65–72.

    https://doi.org/10.1016/S0166-2236(02)00042-5

    • PubMed
    • Google Scholar
47. 1. Page TL

    (2015) Circadian regulation of learning and memory

    Current Opinion in Insect Science 7:87–91.

    https://doi.org/10.1016/j.cois.2014.12.001

    • Google Scholar
48. 1. Parsons RG
    2. Gafford GM
    3. Baruch DE
    4. Riedner BA
    5. Helmstetter FJ

    (2006) Long-term stability of fear memory depends on the synthesis of protein but not mRNA in the amygdala

    European Journal of Neuroscience 23:1853–1859.

    https://doi.org/10.1111/j.1460-9568.2006.04723.x

    • PubMed
    • Google Scholar
49. 1. Purcell AL
    2. Sharma SK
    3. Bagnall MW
    4. Sutton MA
    5. Carew TJ

    (2003) Activation of a tyrosine kinase-MAPK cascade enhances the induction of long-term synaptic facilitation and long-term memory in Aplysia

    Neuron 37:473–484.

    https://doi.org/10.1016/S0896-6273(03)00030-8

    • PubMed
    • Google Scholar
50. 1. Raju U
    2. Koumenis C
    3. Nunez-Regueiro M
    4. Eskin A

    (1991) Alteration of the phase and period of a circadian oscillator by a reversible transcription inhibitor

    Science 253:673–675.

    https://doi.org/10.1126/science.1871602

    • PubMed
    • Google Scholar
51. 1. Ramm P
    2. Smith CT

    (1990) Rates of cerebral protein synthesis are linked to slow wave sleep in the rat

    Physiology & Behavior 48:749–753.

    https://doi.org/10.1016/0031-9384(90)90220-X

    • PubMed
    • Google Scholar
52. 1. Rasch B
    2. Born J

    (2013) About sleep's role in memory

    Physiological Reviews 93:681–766.

    https://doi.org/10.1152/physrev.00032.2012

    • PubMed
    • Google Scholar
53. 1. Rawashdeh O
    2. de Borsetti NH
    3. Roman G
    4. Cahill GM

    (2007) Melatonin suppresses nighttime memory formation in zebrafish

    Science 318:1144–1146.

    https://doi.org/10.1126/science.1148564

    • PubMed
    • Google Scholar
54. 1. Schwartz JH
    2. Castellucci VF
    3. Kandel ER

    (1971)

    Functioning of identified neurons and synapses in abdominal ganglion of Aplysia in absence of protein synthesis

    Journal of Neurophysiology 34:939–953.

    • PubMed
    • Google Scholar
55. 1. Schwarz M
    2. Blumberg S
    3. Susswein AJ

    (1998) Social isolation blocks the expression of memory after training that a food is inedible in Aplysia fasciata

    Behavioral Neuroscience 112:942–951.

    https://doi.org/10.1037/0735-7044.112.4.942

    • PubMed
    • Google Scholar
56. 1. Schwarz M
    2. Feldman E
    3. Susswein AJ

    (1991) Variables affecting long-term memory of learning that a food is inedible in Aplysia

    Behavioral Neuroscience 105:193–201.

    https://doi.org/10.1037/0735-7044.105.1.193

    • PubMed
    • Google Scholar
57. 1. Seibt J
    2. Dumoulin MC
    3. Aton SJ
    4. Coleman T
    5. Watson A
    6. Naidoo N
    7. Frank MG

    (2012) Protein synthesis during sleep consolidates cortical plasticity in vivo

    Current Biology 22:676–682.

    https://doi.org/10.1016/j.cub.2012.02.016

    • PubMed
    • Google Scholar
58. 1. Shifrin VI
    2. Anderson P

    (1999) Trichothecene mycotoxins trigger a ribotoxic stress response that activates c-Jun N-terminal kinase and p38 mitogen-activated protein kinase and induces apoptosis

    Journal of Biological Chemistry 274:13985–13992.

    https://doi.org/10.1074/jbc.274.20.13985

    • PubMed
    • Google Scholar
59. 1. Strumwasser F

    (1971) The cellular basis of behavior in Aplysia

    Journal of Psychiatric Research 8:237–257.

    https://doi.org/10.1016/0022-3956(71)90022-7

    • PubMed
    • Google Scholar
60. 1. Susswein AJ
    2. Schwarz M
    3. Feldman E

    (1986)

    Learned changes of feeding behavior in Aplysia in response to edible and inedible foods

    Journal of Neuroscience 6:1513–1527.

    • PubMed
    • Google Scholar
61. Book

    1. Sweatt JD

    (2010)

    Mechanisms of Memory (Second Edition)

    London, UK, Burlington: MA, San Diego, CA:Academic Press.

    • Google Scholar
62. 1. Tudor JC
    2. Davis EJ
    3. Peixoto L
    4. Wimmer ME
    5. van Tilborg E
    6. Park AJ
    7. Poplawski SG
    8. Chung CW
    9. Havekes R
    10. Huang J
    11. Gatti E
    12. Pierre P
    13. Abel T

    (2016) Sleep deprivation impairs memory by attenuating mTORC1-dependent protein synthesis

    Science Signaling 9:ra41.

    https://doi.org/10.1126/scisignal.aad4949

    • PubMed
    • Google Scholar
63. 1. Tully T
    2. Preat T
    3. Boynton SC
    4. Del Vecchio M

    (1994) Genetic dissection of consolidated memory in Drosophila

    Cell 79:35–47.

    https://doi.org/10.1016/0092-8674(94)90398-0

    • PubMed
    • Google Scholar
64. 1. Vorster AP
    2. Krishnan HC
    3. Cirelli C
    4. Lyons LC

    (2014) Characterization of sleep in Aplysia californica

    Sleep 37:1453–1463.

    https://doi.org/10.5665/sleep.3992

    • PubMed
    • Google Scholar
65. 1. Walker MP
    2. Stickgold R

    (2006) Sleep, memory, and plasticity

    Annual Review of Psychology 57:139–166.

    https://doi.org/10.1146/annurev.psych.56.091103.070307

    • PubMed
    • Google Scholar
66. 1. Walker MP

    (2005) A refined model of sleep and the time course of memory formation

    Behavioral and Brain Sciences 28:51–64.

    https://doi.org/10.1017/S0140525X05000026

    • PubMed
    • Google Scholar
67. 1. Waser PM

    (1968) The spectral sensitivity of the eye of Aplysia californica

    Comparative Biochemistry and Physiology 27:339–347.

    https://doi.org/10.1016/0010-406X(68)90776-7

    • Google Scholar
68. 1. Zhang Y
    2. Liu RY
    3. Heberton GA
    4. Smolen P
    5. Baxter DA
    6. Cleary LJ
    7. Byrne JH

    (2011) Computational design of enhanced learning protocols

    Nature Neuroscience 15:294–297.

    https://doi.org/10.1038/nn.2990

    • PubMed
    • Google Scholar

Author details

1. Roi Levy

   The Leslie and Susan Gonda (Goldschmied) Multidisciplinary Brain Research Center, The Mina and Everard Goodman Faculty of Life Sciences, Bar Ilan University, Ramat Gan, Israel

   Contribution

   RL, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   For correspondence

   roiv.levy@gmail.com

   Competing interests

   The author declares that no competing interests exist.

2. David Levitan

   The Leslie and Susan Gonda (Goldschmied) Multidisciplinary Brain Research Center, The Mina and Everard Goodman Faculty of Life Sciences, Bar Ilan University, Ramat Gan, Israel

   Contribution

   DL, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The author declares that no competing interests exist.

3. Abraham J Susswein

   The Leslie and Susan Gonda (Goldschmied) Multidisciplinary Brain Research Center, The Mina and Everard Goodman Faculty of Life Sciences, Bar Ilan University, Ramat Gan, Israel

   Contribution

   AJS, Conception and design, Analysis and interpretation of data, Drafting or revising the article

   For correspondence

   avy@biu.ac.il

   Competing interests

   The author declares that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0001-6107-5901

• 3,774

  views

• 469

  downloads

• 9

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.