In tomorrow’s Science, researchers led by André Fischer at the Laboratory for Aging and Cognitive Diseases in the European Neuroscience Institute in Göttingen, Germany, demonstrate that alterations in learning-dependent acetylation at a specific histone site may be an early biomarker of memory impairment during aging. The group also shows that restoring histone acetylation can restore memory consolidation and learning-induced changes in gene expression that are ordinarily altered in aging brains.

“The world's scientific community may be one step closer to understanding age-related memory loss and to developing a drug that might help boost memory,” remarked David Sweatt from the University of Birmingham at Alabama in an e-mail comment to ARF about this study. Sweatt published an accompanying Perspectives article in the same issue of Science.

In their experiments, first authors Shahaf Peleg, Farahnaz Sananbenesi, and Athanasios Zovoilis used a contextual fear conditioning task to measure associative hippocampal learning in young (three-month-old), adult (eight-month-old), and middle-aged (16-month-old) mice. Mice in all age groups learned to associate a new environment with a mild foot shock. However, middle-aged mice showed less freezing behavior during a memory test, indicating that their associative memory was slightly impaired. Researchers also saw similar impairments in a Morris water maze test for spatial memory.

When Fischer and colleagues analyzed hippocampal gene expression of young mice after the fear conditioning task, they saw altered expression of more than 1,500 genes linked to associative learning. In contrast, they saw almost no change in the gene expression of middle-aged mice after learning. Middle-aged mice also failed to increase acetylation of histone H4 lysine 12 (H4K12) for those genes that were upregulated after learning in young mice. Increased acetylation is generally associated with both a more open chromatin structure and with increased gene expression.

Changes in acetylation at other histone sites and for genes that were downregulated or not regulated during learning were similar among age groups. Histone acetylation, hippocampal gene expression, and the activity of histone acetyltransferase and histone deacetylase (HDAC) enzymes were also similar between young and middle-aged mice that had not gone through the learning task. “Under basal conditions, there is no difference,” summarized Fischer in an interview with ARF. “But when [the mice] have to learn something and form a memory…the young mice are regulating 1,500 genes to go up transiently and then to go down. In the old mice, which are actually just a little bit impaired, none of this happens.”

Unlike most histone modifications, which tend to cluster in gene promoter regions, the scientists observed H4K12 acetylation mainly in gene coding regions. Fischer suggests that H4K12 acetylation is therefore less important to initiate transcription of these upregulated genes than it is to elongate transcription.

Fischer likens the potential role of H4K12 in learning-associated gene expression to the process of starting and driving a car. All of the other machinery and components for gene expression are present, and transcription initiation (starting the car) can still occur properly. “But if you only have three wheels,” says Fischer, “you still cannot drive anywhere. The fourth wheel that is missing is this H4K12 [acetylation]. Everything else is fine, but the genome is not responding.”

Finally, the team showed that restoration of H4K12 acetylation using pharmacologic HDAC inhibitors can restore expression of learning-regulated genes and improve associative learning during fear conditioning. To test this, the authors injected suberoylanilide hydroxamic acid (SAHA) into the hippocampi of middle-aged mice one hour before they attempted the associative learning task. Mice treated with SAHA showed increased H4K12 acetylation in the coding regions of learning-associated genes, as well as higher expression of these genes. The authors saw similar results when they treated mice with the pan-HDAC inhibitor sodium butyrate, but not when they treated them with a different HDAC inhibitor that did not increase H4K12 acetylation.

These findings are consistent with previous studies that also showed better learning following treatment with HDAC inhibitors in mouse models of neurodegenerative disease, including two recent studies from the Sweatt and Arancio labs conducted in AD mouse models (see Kilgore et al., 2010; Francis et al., 2009; Sananbenesi and Fischer, 2009; and ARF related news story). Together, these results suggest that histone deacetylase inhibitors may improve cognitive function in both normal and disease-related memory decline.

Additional research is needed before these results can be extended to humans. Even so, Fischer is optimistic about the therapeutic potential of these findings. “I think there is now a clear focus on what needs to be targeted, and researchers just need to take it to the next step,” he says. “I think we aren’t far from a clinical application anymore.”—Elizabeth Eyler.

Elizabeth Eyler is a freelance writer in Baltimore, Maryland.

References:
Peleg S, Sananbenesi F, Zovoilis A, Burkhardt S, Bahari-Javan S, Agis-Balboa RC, Cota P, Wittnam JL, Gogol-Doering A, Opitz L, Salinas-Reister G, Dettenhofer M, Kang H, Farinelli L, Chen W, Fischer A. Altered histone acetylation is associated with age-dependent memory impairment in mice. Science. 2010 May 7. Abstract

Sweatt JD. Epigenetics and Cognitive Aging. Science 2010 May 7. Abstract