The dynamics of HDAC activity on memory formation

Abstract

Histone deacetylases (HDACs) have previously been shown to be critical for the formation of long-term memories. Recent findings now show that a specific HDAC isoform, HDAC2, negatively regulates formation of hippocampus-dependent memory. These recent findings published in Nature highlight potential new therapeutic interventions for the treatment of memory impairments associated with human neurological disorders.

Humans have a tremendous capacity for storing complex information for long periods of time. There are several molecular and cellular mechanisms that have been implicated in the process of storing stable long-term memories. One such mechanism involves the transcription of new genes, which has been shown to be essential for learning and memory processes. In addition, it is well-supported that new protein synthesis is required for the storage of long-term memory. Together, this suggests that new memories are formed by initial transcription of genes that lead to new protein synthesis. Since then, the molecular mechanisms responsible for the regulation of new gene transcription involved in the formation of stable long-term memories have caught the attention and imagination of a number of behavioral neuroscientists.

Epigenetics, a cellular mechanism thought to be static after development, has recently been shown to be a dynamic process occurring in non-dividing cells in the adult brain subject to environmental stimuli (Alarcon et al., 2004; Korzus et al., 2004; Chwang et al., 2006; Levenson et al., 2006; Wood et al., 2006; 2007; Lubin et al., 2008; Roth et al., 2009). Epigenetic mechanisms mediate gene transcription changes through the regulation of chromatin structure. For example, modifications to histone proteins produce lasting changes in chromatin structure, thereby producing lasting alterations in gene expression and patterns of protein synthesis for stable formation of long-term memory. Histone acetylation, or the addition of acetyl groups to histone proteins, especially histones H3 and H4, is an established method of such modifications (Verdone et al., 2005). The addition of these acetyl groups is facilitated by enzymes called histone acetyltransferases (HATs), and leads to a decrease of affinity between DNA and histone proteins, allowing the genes to be transcribed. The acetyl groups can be removed by histone deacetylases (HDACs), thus closing the genes to expression (Varga-Weisz and Becker, 1998). Histone deacetylation is not permissive for long-term memory storage, leading behavioral neuroscientists to focus their research efforts to investigating the permissive effects of HDAC inhibitors (HDACi) on gene expression and memory formation. Although several studies have shown that inhibition of HDACs by non-selective HDACi can enhance synaptic plasticity and memory formation (Levenson et al., 2004; Fischer et al., 2007), little is known about the actual HDAC isoform(s) mediating these effects. To date, there are eleven known HDAC isoforms in the rat (Broide et al., 2007). Thus, the identification of the specific HDAC isoforms responsible for memory enhancement by HDACi is critical for the use of more selective HDACi for the therapeutic treatment of memory impairments.

Now a recent publication in Nature from Guan et al. (2009) undertook an impressive series of molecular and behavioral experiments to demonstrate the importance of HDAC2 in the negative regulation of memory formation. First, Guan et al. (2009) established that suberoylanilide hydroxamic acid (SAHA), an HDACi specific to HDAC 1, 2, and 6, improved memory and synaptic plasticity. Guan et al. (2009) were able to eliminate a role of HDAC6 in SAHA effects by synthesizing a novel HDAC6 specific inhibitor, WT-161. Administration of this new compound had no effect on memory formation. Furthermore, they investigated the specific role of HDAC 1 and 2 via genetic knockout and overexpressor mice models. Interestingly, they found that HDAC2 overexpression (HDAC2OE) produced decreased dendritic spine density, synapse number, synaptic plasticity and attenuated memory formation compared to HDAC1 overexpression (HDAC1OE). Hdac2-deficiency (HDAC2KO) resulted in increased synapse number and enhanced memory formation that was similar to effects observed with the HDAC inhibitor SAHA. Crucially, HDAC2KO does not seem to be associated with gross changes in neuronal morphology or behavior, suggesting that HDAC2 can be safely targeted without broadly disrupting neuronal physiology. Taken together, the authors conclude that inhibition of the HDAC2 isoform is a viable therapeutic target for treatment of cognitive impairments associated with neurological disorders.

In Figure 1, the major findings of Guan et al. (2009) are outlined and the potential directions for future research are highlighted. The effects of HDAC1OE, HDAC2OE, or HDAC2KO genetic manipulation on memory formation and synaptic plasticity are summarized in Panel A (Fig. 1A). One of the most striking findings of this study is that HDAC2KO induced a large increase in acetylated histone H3 (AcH3) and acetylated histone H4 (AcH4) levels at several gene promoter regions associated with learning and memory, including Egr1, Fos, Glur1, and Creb. Interestingly, they observed that for some of these genes the effect of HDAC1OE was similar to the effect of HDAC2KO. Specifically, AcH4 levels at the Fos promoter and AcH3 levels at the Creb promoter were elevated in both HDAC2KO and HDAC1OE mice. Could it be that HDAC1 and HDAC2 function in a compensatory manner? Indeed, Guan et al. (2009) note in their supplemental information that HDAC2KO mice exhibited large increases in HDAC1 expression. Perhaps HDAC1OE induced reductions in HDAC2 expression. Additionally, the promoter region of the Arc gene, which has been shown to be involved in learning and memory, predominantly associates with HDAC1, a notable exception to HDAC2 association with the vast majority of genes important to learning and memory. Thus, although HDAC2 is demonstrated to be a negative regulator of memory formation and synaptic plasticity, HDAC1 may make some important contributions as well.

Figure 1. A summary of the findings in Guan et al., 2009.

(A) The three genetic models used by Guan and colleagues include the knockout of Hdac2 (HDAC2KO) and overexpression of either HDAC 1 (HDAC1OE) or HDAC 2 (HDAC2OE) compared to wildtype (WT). Interestingly, the authors report that HDAC2KO leads to an increase in HDAC1 in the brain. Double arrows indicate the primary effects of the genetic manipulation, while the single arrow indicates the known secondary effects. The question marks indicate unknown secondary effects of the genetic manipulations. (B) Guan et al., 2009 demonstrate that treatment with the HDACi SAHA or WT-161 strongly implicates HDAC2 in learning and memory. SAHA inhibits HDAC 1, 2, and 6. Unlike HDAC 1 and 2, HDAC6 is primarily localized in the cytoplasm and is involved in the acetylation of tubulin. Furthermore, the specific inhibitor of HDAC6, WT-161, had no effects on learning and memory. It is important to note that HDACs can alter acetylation of not only histones and tubulin, but transcription factor subunits as well, such as Rel A.

Another highlight of the genetic models used by Guan et al. (2009) are the widespread learning deficits induced by overexpression of HDAC2, but not HDAC1. As we note in Figure 1A, HDAC2OE mice exhibited significant impairments in memory in three different paradigms: contextual fear conditioning, Pavlovian (tone) fear conditioning, and the Morris water maze, highlighting the importance of HDAC2 in several types of learning and memory. Intriguingly, in their experiments the impairment in Pavlovian (tone) fear conditioning, which is amygdale-based, was even greater in magnitude than the impairment in contextual fear conditioning, which is hippocampal-based. This suggests that hippocampal-dependent learning and memory may be differentially regulated by other HDACs compared to amygdala-based learning and memory, making amygdala-dependent learning even more sensitive to HDAC2 overexpression, thus leaving this interesting concept for future research.

While evidence presented by Guan et al. (2009) suggests that tubulin acetylation is not relevant to alterations in memory formation and synaptic plasticity, they do suggest that other molecular targets may be relevant as well. Interestingly, there are other molecular targets for HDACs besides histones and tubulin. In Figure 1B we present another possible target for HDAC2 function within the cell. For example, the p65 subunit (RelA) of the transcription factor nuclear actor-kappa B (NF-kB) complex interacts with HDAC1 and HDAC2 to negatively regulate gene expression (Ashburner et al., 2001). In addition, RelA is acetylated and deacetylated a process shown to be crucial for long-term memory formation (Yeh et al., 2004). Therefore, a plausible effect of Hdac2 deficiency is not only an enhancement of histone acetylation but perhaps also an enhancement of RelA acetylation and enhanced NF-kB DNA binding activity, thereby leading to enhanced memory (Lubin and Sweatt, 2007). Hence, while the studies by Guan et al. (2009) strongly support HDAC2 activity as a negative regulator of hippocampal-dependent memory formation and is consistent with the hypothesis that enhanced histone acetylation is involved in enhanced memory formation, it cannot be ruled out that these results may also be indicative of a more general role for enhanced protein acetylation by HDACi in enhanced memory formation as well (Swank and Sweatt, 2001; Yeh et al., 2004).

In conclusion, the studies performed by Guan et al. (2009) strongly implicates HDAC2 activity as a negative regulator of memory formation. This major finding illustrates that the continued investigation of the role of HDAC isoforms in the process of memory storage is necessary. Future experiments could address a number of interesting questions. For example, while histones are an important molecular target for HDACs, they are not the only targets, and subsequent research should investigate the multiple actions of HDACs. Similarly, the distinct actions of HDACs at particular gene promoter sites, and in particular brain areas, may be important mechanisms of differentially regulating memory in response to various environmental stimuli. Finally, other questions may be directed at the effect of HDAC2 overexpression on other epigenetic mechanisms, such as other histone modifications or DNA methylation. Thus, future studies should include these questions in order to provide a more complete understanding of the role of these epigenetics mechanisms in the regulation of gene transcription during long-term memory storage.