PROJECT SUMMARY Alzheimer’s disease (AD) is a severe form of dementia that affects over 5 million Americans, making it one of the most common neurodegenerative disorders. Patients with AD exhibit a severe form of progressive memory loss that cannot be prevented with current pharmacological interventions, and are at increased risk of developing comorbid neurological disorders such as epilepsy. The lack of efficacious treatments and the intensive care required for late-stage AD patients also places a significant economic burden on families and the American healthcare system: an estimated ~$200 billion annually. Recent studies have demonstrated a link between epigenetic modifications and AD. One epigenetic modification of particular relevance is 5-hydroxymethylcytosine (5hmC) which is highly enriched in neurons and is dynamically regulated during neurodevelopment. Alterations in 5hmC have been observed in the brains of patients with AD and in mouse AD models, raising the possibility that aberrant DNA 5hmC might contribute to AD pathogenesis. GADD45A mediates the establishment of 5hmC in cells and also possesses RNA binding activity. Interestingly, Gadd45a-/- null mice exhibit decreased hippocampal long-term potentiation (LTP) and impaired long-term memory and reversal learning, whereas overexpression of Gadd45a in pyramidal excitatory neurons of wild-type mice enhances LTP and memory consolidation. In addition, our preliminary data and publicly available RNA-seq data sets indicate that GADD45A is downregulated in neurons isolated from AD patients. Gadd45a expression and 5hmC levels are also reduced in several mouse models of AD (including the 3xTG AD model to be used in this proposal). Based on these observations, we hypothesize that GADD45A plays an important role in neuronal function and cognition through epigenetic regulation of 5hmC levels and subsequent modulation of gene expression in neurons. We will test this hypothesis in Aim 1 by comparing 5hmC levels and distribution between Gadd45a-/- null mutants and WT littermates. We will also perform RNA-seq and ATAC-seq on these mice to functionally relate observed changes in 5hmC with specific alterations in gene expression and chromatin accessibility. Given that overexpression of Gadd45a enhances LTP and memory consolidation, we hypothesize that upregulation of GADD45A will be therapeutic in AD. Therefore, in Aim 2, we will examine whether overexpression of Gadd45a in the 3xTG mouse model of AD can ameliorate deficits in synaptic activity, improve learning and memory, and reduce spontaneous epileptiform discharges. These phenotypes were selected because they represent important pathological features of AD and are mechanistically related since cognitive decline in the AD brain is believed to be accelerated by neuronal hyperexcitability.