In psychological terms memory describes the processes utilized by the brain for long-term storage of information. Early studies implicated both protein translation and gene transcription as vital to the formation of long-term memories (30
). Subsequent studies have shown that formation of long-term memory is a complex process that requires the engagement of many distinct signaling pathways and the regulation of numerous genes (32
A recent study by Levenson et al. (35
) has suggested that the same processes that lead to formation of long-term memory also lead to epigenetic marking of the genome. Contextual fear conditioning is a hippocampus-dependent learning paradigm whereby an animal learns to associate a novel context with an aversive stimulus (36
). Acetylation of hippocampal histone H3, but not H4, is significantly increased after an animal is trained using a contextual fear conditioning paradigm (35
). Formation of long-term contextual fear memory requires NMDA-receptor-dependent synaptic transmission and the ERK MAPK signaling cascade in the hippocampus (39
), and inhibition of either of these critical cellular processes blocks the memory-associated increase in acetylation of H3 (35
). These observations were the first to demonstrate that epigenetic tagging of the genome occurred during consolidation of hippocampus-dependent long-term memory.
Interestingly, in these same studies Levenson et al found that a different form of long-term memory, latent inhibition, was associated with altered acetylation of histone H4, while H3 acetylation was unaltered by this paradigm. This finding along with many others suggests the intriguing possibility that a type of epigenetic code might exist for memory formation, whereby specific types of memories are associated with specific patterns of histone modifications (11
Addition of acetyl groups to lysine residues within histone proteins is accomplished via the action of histone acetyltransferases (HATs). If acetylation of histones is functionally significant for consolidation of long-term memory, then disruption of HAT activity would be predicted to interfere with long-term memory formation. CREB Binding Protein (CBP) is a transcriptional coactivator; an enzyme that contains endogenous HAT activity (45
). Several studies have investigated long-term memory formation in genetically manipulated mice with impaired CBP function. One class of mice examined contain an allele of CBP that codes for a truncated form of the protein, which acts in a dominant negative manner (CBPDN+/−
mice exhibit significant deficits in various forms of long-term memory including step-through passive avoidance, novel object recognition and cued fear conditioning (46
). While these studies provided the first evidence that CBP might play a role in long-term memory formation, the widespread developmental derangements exhibited by CBPDN+/−
animals prevented straightforward interpretation of the performance of these animals in various memory tasks(46
To further elucidate the role of CBP in long-term memory formation, three very sophisticated recent studies have made CBP-deficient animals that lack the severe developmental problems present in the CBPDN+/−
animals. The first study improved upon the CBPDN+/−
mice by linking the dominant negative allele of CBP to an inducible promoter (CBPI-DN+/−
). Activation of the dominant negative allele after animals had developed normally, led to impaired acquisition of the spatial watermaze task and novel object recognition, two forms of hippocampus-dependent memory (48
). In another series of experiments, animals that lacked one allele of CBP (CBP+/−
) exhibited impairments in contextual and cued fear memory, and novel object recognition (49
). In both studies, administration of an HDAC inhibitor restored normal long-term memory formation, suggesting that the balance of HAT/HDAC activity was altered in these mice (48
) and causative of the memory deficits. Finally, mice that carry inactivating mutations in the CREB-binding (KIX) domain of CBP exhibit deficits in long-term memory for contextual fear conditioning and novel object recognition (50
). These results support a role for CBP-mediated histone acetylation in memory formation and suggest that KIX-interacting transcription factors like CREB recruit CBP histone acetyltransferase activity during long-term memory storage.
The above studies demonstrate that histone acetylation is regulated by, and disruption of HAT activity impairs, long-term memory formation. Together, these observations suggest that any perturbations in the processes that regulate chromatin structure influence long-term memory formation in the behaving animal in vivo. However, can augmentation of histone acetylation enhance memory formation? To directly test this, several studies have investigated the effect of HDAC inhibitors on long-term memory formation. In a study by Yeh et al.(24
), direct infusion of the HDAC inhibitor trichostatin A into the amygdala significantly enhanced formation of fear potentiated startle memory. Additionally, Levenson et al. (35
) demonstrated that systemic administration of the HDAC inhibitor sodium butyrate enhanced formation of contextual fear memory, and Wood et al demonstrated that direct intrahippocampal infusion of an HDAC inhibitor enhanced hippocampus-dependent contextual fear conditioning (50
). Finally, a fascinating recent study by Abel, Wood, and coworkers demonstrated that these memory-enhancing effects of HDAC inhibitors depend on the CREB/CBP pathway (51
), nicely unifying the HDAC inhibitor studies with the CBP-HAT oriented studies described in the preceding paragraphs.
In the clinical setting, extinction of aversive and maladaptive memories is often a significant challenge. Might regulation of chromatin structure also be involved in extinction of memories? Bredy et al (52
) have shown that extinction of fear conditioning in laboratory animals is associated with alterations in chromatin structure, implicating regulation of these processes in fear extinction. Moreover, using a conditioned fear paradigm, two different groups have shown that extinction of conditioned fear is accelerated when animals are administered HDAC inhibitors in vivo (53
). Thus, extinction of learned fear, like fear learning itself, appears to involve chromatin modifications and be subject to enhancement with HDAC inhibitors. These studies both provide a new insight into molecular mechanisms of memory extinction, they suggest a potential new route for pharmacotherapy of maladaptive fear behaviors.
In a recent series of studies my laboratory has investigated the capacity of DNA methylation, the other major epigenetic molecular mechanism besides histone modification, to regulate synaptic plasticity and memory in adult animals (55
). In our first series of studies in this area we found that inhibitors of DNMTs that likely block the net effects of both maintenance and de novo DNMTs could alter DNA methylation in adult CNS tissue and block hippocampal Long-term Potentiation (LTP) in physiologic studies vitro (55
). In additional more recent studies we found that de novo DNMT gene expression (DNMT3a and DNMT3b) is upregulated in the adult rat hippocampus following contextual fear conditioning, and that generalized DNMT inhibition blocks memory formation in this same paradigm (56
). In addition, fear conditioning was associated with rapid methylation and transcriptional silencing of the memory suppressor gene Protein Phosphatase 1 (PP1) and demethylation and transcriptional activation of the synaptic plasticity gene reelin. These findings have the surprising implication that both DNA methylation and demethylation might be involved in long-term memory consolidation. Overall these results suggest that DNA methylation is dynamically regulated in the adult nervous system and that this cellular mechanism is a crucial step in memory formation.
These studies of DNA methylation are at a very early stage and do not address several open questions. First, the biochemical mechanism for active DNA demethylation is unknown and indeed the general idea of active DNA demethylation has a contentious history (reviewed in 57
). In addition, the mechanistic interplay between histone acetylation and DNA methylation/demethylation has not been worked out, and the possibility exists of reciprocal regulation between the two mechanisms (56
). Finally, the signal transduction processes that might be controlling DNA methylation and demthylation in the adult CNS are completely unexplored at present.
Despite these numerous unanswered questions, taken together all these studies of histone acetylation and DNA methylation indicate that long-term behavioral memory processes regulate, and are regulated by, the epigenome.