Hippocampal memory formation requires neuroplasticity, which is produced by the orchestrated expression of neuronal genes through chromatin modification1,2. Histone acetylation is a post-translational modification of histone proteins that helps to restructure chromatin and regulate the ability to store and recall previously acquired information3. Histone acetyltransferases (HATs) transfer acetyl groups (-COCH3) to the lysine residues of histone tails, thus neutralizing the positive charges and decreasing their affinity for negatively charged DNA that is wound around the nucleosome. This process allows the condensed chromatin structure to relax and local gene expression to increase. For this process to occur, metabolic production of the acetyl-group donor, acetyl coenzyme A (acetyl-CoA), is indispensable, but the precise mechanisms remain poorly understood.
Previous investigations have shown that manipulating the concentration of intracellular acetyl-CoA can alter histone acetylation and gene expression4,5. With this knowledge, GGR alumnus Dr. Philipp Mews, who completed this thesis work in Dr. Shelley Berger’s lab, set out to determine whether the metabolic enzymes that generate acetyl-CoA from intermediary metabolites might directly control the epigenetic modifications necessary for neuronal development and memory storage. In a recent Nature publication, he demonstrates that acetyl-CoA synthetase 2 (ACSS2) directly regulates histone acetylation in post-mitotic neuronal cultures and gene expression changes that affect spatial memory in mammals. Philipp recalls that at the time he conceived the project, metabolic enzymes participating in precise gene regulation of the nucleus was a highly controversial notion.
“The idea that key metabolic enzymes could bind directly to chromatin to fuel the local chromatin modification machinery was bold and novel, and to see it borne out so clearly in my data was really exciting,” said Philipp.
Philipp used the Cath.-a-differentiated (CAD) cell line to investigate the function of ACSS2. CAD cells are derived from mouse catecholaminergic cells and differentiate to express neuronal properties upon serum deprivation. He observed that ACSS2 shifted from cytoplasmic localization in undifferentiated CAD cells to nuclear localization upon differentiation, suggesting ACSS2 may have a role in the neuronal development of these cells. Transcriptome analysis of differentiated CAD cells revealed upregulation of neuron-specific genes as well as increased acetylation of these genes compared to the surrounding genome. Additionally, knockdown of ACSS2 abolished this induction of neuronal gene expression upon differentiation. Philipp concluded that ACSS2 is critical for the upregulation of genes implicated in neuronal differentiations.
To determine whether ACSS2 enrichment correlated with histone acetylation, Philipp performed ChIP-seq analyses in undifferentiated and differentiated CAD cells. Areas bound by ACSS2 were proximal to genes linked to neuronal differentiation and 80% of these ACSS2 peaks overlapped an acetylation peak or had an acetylation peak near the targeted transcription start site. The co-occupancy of ACSS2 and acetylation marks suggests ACSS2 may have a role in providing acetyl-CoA for HAT enzymes. To investigate this, Philipp measured the levels of acetyl-CoA in ACSS2 knockdown cells and cells treated with ACSS2 siRNA and found significant reduction of the metabolite in both cases. Global transcription-linked acetylation marks were similarly reduced upon ACSS2 knockdown. Experiments with primary mouse hippocampal neurons corroborated these findings, supporting the hypothesis that ACSS2 functions in neuronal histone acetylation.
Previous studies have shown that there is a strong link between histone acetylation in neurons and memory formation3. ACSS2 co-immunoprecipitated with H3K9ac and H3K27ac, which are key substrates of transcriptional coactivators CBP and p300 that have roles in long-term memory6. ACSS2 also co-immunoprecipitated with CBP in differentiated CAD cells, correlating with data showing colocalization of the two proteins in the mouse hippocampus. To examine the function of ACSS2 in hippocampus-dependent spatial memory, ACSS2 expression was attenuated in the dorsal hippocampus of adult mice by shRNA. An object-location memory paradigm revealed impaired long-term spatial memory upon ACSS2 knockdown. mRNA-seq of the dorsal hippocampus of untreated control mice exposed a small number of upregulated genes which encode proteins that mediate the strength of neural connections immediately after learned behavior. Upregulation was reduced by ACSS2 knockdown, as was induction of memory retrieval-associated genes. Together, the in vitro and in vivo findings are the first evidence linking metabolic signaling to chromatin restructuring in the brain and memory consolidation.
ACSS2 is a chromatin-bound source of acetyl Co-A for histone acetylation in neurons and is essential for memory formation in mammals. (A) ACSS2 binding cor relates with histone aceylation and gene expression of differentiation-linked genes in neuronal cultures. The co-occupancy of ACSS2 and acetylation marks suggests ACS22 may have a role in providing acetyl Co-A for histone acetyltransferase (HAT) enzymes (B) In an object-location memory paradigm, mice explore three different objects during a training period. After 24 hours, the mice are re-exposed to the objects with one object changed to a new location. The amount of time spent exploring the novel object-location cor responds with spatial object memory; the ACSS2 knockdown mice had reduced total object exploration, suggesting ACSS2 is critical for long-termmemory formation.
Philipp’s work establishes an association between cellular metabolism, gene regulation, and neuroplasticity by elucidating the neuronal function of ACSS2 as a chromatin-bound source of acetyl-CoA for histone acetylation. Localization of ACSS2 is linked to the increased histone acetylation and transcription of neuronal genes required for spatial memory formation and retrieval. Understanding this metabolic pathway and other epigenetic mechanisms brings us one step closer to unraveling the complex functions and behaviors of the brain and how these may become dysregulated in neuropsychiatric disorders, such as anxiety or depression. ACSS2 is now a novel enzymatic target for developing therapies to restore histone acetylation due to its critical role at the interface of neural metabolism and chromatin restructuring.
Philipp is currently a postdoctoral fellow in Dr. Eric Nestler’s lab at the Friedman Brain Institute at the Icahn School of Medicine at Mount Sinai in NYC. His long-term goal is to understand the complex interplay between epigenetic gene regulation and neuronal circuit connectivity.
Kandel, E.R., Dudai, Y. & Mayford, M.R. The molecular and systems biology of memory. , 163-186 (2014).
Zovkic, I.B., Guzman-Karlsson, M.C. & Sweatt, J.D. Epigenetic regulation of memory formation and maintenance. , 61-74 (2013).
Gräff, J & Tsai, L.-H. Histone acetylation: molecular mnemonics on the chromatin. , 97-111 (2013).
Cai, L., Sutter, B.M., Li, B. & Tu, B.P. Acetyl-CoA induces cell growth and proliferation by promoting acetylation of histones at growth genes. , 426-437 (2011).
Wellen, K.E. . ATP-citrate lyase links cellular metabolism to histone acetylation. , 1076-1080 (2009).
Vecsey, C.G. Histone deacetylase inhibitors enhance memory and synaptic plasticity via CREB:CBP-dependent transcriptional activation. , 6128-6140 (2007).