Integrated Genome-Wide Analysis of DNA Methylation and Gene Expression in the Hippocampi of 5xFAD Alzheimer’s Disease Mouse Model

Background : DNA methylation forms 5-methylcytosine and its regulation in the hippocampus is critical for learning and memory. Indeed, dysregulation of DNA methylation is associated with neurological diseases. Alzheimer’s disease (AD) is the predominant of dementia and a neurodegenerative disorder. Methods : We examined the learning and memory function in 3-and 9-month-old wild-type and 5xfamiliar Alzheimer’s disease (5xFAD) transgenic mice by performing the object recognition memory and Y-maze tests, and identified the hippocampal amyloid beta burden. To investigate the epigenetically regulated genes involved in the development or neuropathology of AD, we performed genome-wide DNA methylation sequencing and RNA sequencing analyses in the hippocampus of 9-month-old wild-type and 5xFAD tg mice. To validate the genes inversely regulated by epigenetics, we confirmed their methylation status and mRNA levels. Results : At 9 months of age, 5xFAD tg mice showed significant cognitive impairment and amyloid-beta plaques in the hippocampus. DNA methylation sequencing identified a total of 13,777 differentially methylated regions, including 4484 of hyper-and 9293 of hypomethylated regions, that are associated with several gene ontology (GO) terms including ‘nervous system development’ and ‘axon guidance’. In RNA sequencing analysis, we confirmed a total of 101 differentially expressed genes, including 52 up-and 49 downregulated genes, associated with GO functions such as ‘positive regulation of synaptic transmission, glutamatergic’ and ‘actin filament organization’. Through further integrated analysis of DNA methylation and RNA sequencing, three epigenetically regulated genes were selected: thymus cell antigen 1, theta ( Thy1 ), myosin VI ( Myo6 ), and filamin A-interacting protein 1-like ( Filip1l ). The methylation level of Thy1 decreased and its mRNA levels increased, whereas that of Myo6 and Filip1l increased and their mRNA levels decreased. The common functions of these three genes may be associated with the neural cytoskeleton and synaptic plasticity. Conclusions : We suggest that the candidate genes epigenetically play a role in AD-associated neuropathology (i.e., amyloid-beta plaques) and memory deficit by influencing neural structure and synaptic plasticity. Furthermore, counteracting dysregulated epigenetic changes may delay or ameliorate AD onset or symptoms.


Introduction
DNA methylation, a process in which a methyl group is added to the C5 position of cytosine, thereby forming 5-methylcytosine [1], controls gene expression by attracting proteins that suppress gene transcription or preventing the attachment of transcription factors [1].During development, the DNA methylation pattern in the genome undergoes changes via dynamic processes, including de novo DNA methylation and demethylation.After development, differentiated cells acquire persistent and distinct DNA methylation patterns that regulate the transcription of specific genes in each tissue [2].However, the patterns of DNA methylation can change following development, a process possibly influenced by cellular activity or the microenvironment [3].In the central nervous system, particularly in the hippocampus, precise control of DNA methylation is crucial for normal learning and memory function.DNA methylation changes due to congenital mutations or acquired environmental factors can lead to various psychiatric and neurological disorders, often causing mental impairment [4].
Alzheimer's disease (AD) is the most prevalent neurodegenerative disorder with symptoms such as deterioration in memory, cognitive, and executive abilities, as well as alterations in personality [5].The primary pathological features of AD are the accumulation of amyloid beta (Aβ) plaques and formation of neurofibrillary tangles (NFTs) that occur throughout the brain regions, particularly in the cerebral cortex and hippocampus [6,7].The accumulation of Aβ and tau proteins disrupts the transmission of signals and substances between neurons, ultimately resulting in neuronal death [8].Recent studies have demonstrated that not only genetic factors but also epigenetic imbalance may attribute to the abnormal expression of genes associated with synaptic plasticity and cognition in AD [9][10][11].However, the relationship between DNA methylation in genetic control studies of AD and the gross pathology of AD remains unclear [12][13][14].
In this study, we investigated the integrated analysis of global DNA methylation and gene expression in the hippocampi of 5xfamiliar Alzheimer's disease (5xFAD) transgenic mice and explored the role of epigenetically regulated genes in the development and neuropathology of AD.

Animals
The animals in this study were cared for and maintained according to protocols authorized by the Institutional Animal Care and Use Committee of Chonnam National University (CNU IACUC-YB-2023-49).Research involving animals was conducted in accordance with the guidelines outlined in the Guide for the Care and Use of Laboratory Animals issued by the US National Institutes of Health.
We utilized 3-and 9-month-old male 5xFAD tg mice models with AD-like manifestations through the production of transgenes containing mutations in amyloid precursor protein (APP, the Swedish, Florida, and London AD mutations) and presenilin 1 (PS-1, M146L and L286V mutations) under the control of the thymus cell antigen 1 (Thy1) promoter.The model mice exhibited elevated Aβ42 production in young mice and formation of amyloid plaques within 2 months [15].Male 3-and 9-month-old wild-type mice were used as controls.

Novel Object Recognition Test (NORT)
The novel object recognition test (NORT) was conducted to evaluate the cognitive abilities of the mice in terms of learning and memory.The test protocol was conducted as previously described [16].Briefly, mice (n = 7 per group except for the 3-month-old wild-type mice of n = 9) were acclimated to an open acrylic chamber measuring 45 × 45 × 30 cm.To mitigate the impact of odor, the acrylic box and each item were thoroughly cleaned using a 70% (v/v) ethanol solution after every use.During the training session, we introduced two identical objects (referred to as object A) into the chamber and the mice explored the arena for 10 min.After a 24-h training period, one object A was substituted with an unfamiliar object B, and testing was conducted for 10 min.The recording system was positioned overhead of the chamber, and the mo-tions of the mice were documented with video tracking using Viewer3 (BIOSERVE GmbH, Mainz, Germany).The percentage preference for each object was calculated by dividing the number of visits to a particular object by the total number of visits to both objects.

Y-maze Test
The Y-maze test was employed to quantify working and reference memories by observing spontaneous alternation behaviors.The test was conducted as previously described [17].Briefly, the activity of mice (n = 7 per group except for 3-month-old wild-type mice of n = 9) was recorded for 8 min and subsequently evaluated using a computer tool (Viewer3; BIOSERVE GmbH).Alternation behavior was defined as the successive entry of three separate arms into overlapping triplet sets.The spontaneous alternation percentage (SAP) was determined by dividing the number of alternations by the total number of arm entries minus two, and multiplying the result by 100.

Immunohistochemistry
Immunohistochemical staining was performed as previously described [18].Briefly, the brain tissues (n = 6 per group) were preserved in a solution containing 4% paraformaldehyde, embedded in paraffin, and cut into 5 µm thick sections.Sections were immunostained using the Vectastain Elite ABC kit (Vector Laboratories Inc., Burlingame, CA, USA), according to the manufacturer's instructions.For antigen retrieval, sections were immersed in citrate buffer solution (pH 6.0) and boiled for 30 min.To perform immunoperoxidase labeling, the activity of the naturally occurring peroxidase enzyme was inhibited by treating the sample with a solution containing 0.3% hydrogen peroxide in absolute methanol for 15 min.
To perform immunohistochemistry for Aβ, brain sections were incubated with normal horse serum (Vector Laboratories Inc.) to prevent non-specific binding.Slides were then incubated overnight with mouse anti-6E10 antibody (cat.no.SIG-39320; Covance, Emeryville, CA, USA) at a 1:1000 dilution.After washing, the sections were incubated with the corresponding secondary antibodies.Before mounting, the sections were counterstained with Harris hematoxylin.ImageJ (version 1.8.0.;National Institutes of Health, Bethesda, MD, USA) was utilized to measure the quantity of Aβ plaque in the immunostained hippocampus.

Microarray Data
Initially, three samples per group were processed for the microarray analysis, but a decrease in the total number of detected transcripts and an abnormal distribution in the Principal Component Analysis were observed for one of mice included in the wild-type group (data not shown).Thus, we decided to exclude the data derived from this specific subject in order to improve the quality and reliability of the downstream analysis.Consequently, the hippocampal tissues of 9-month-old wild-type (n = 2) and 5xFAD tg (n = 3) mice were subjected to microarray analysis.A flowchart of methylation sequencing (methyl-seq) and RNA sequencing (RNA-seq) for identifying the target genes involved in AD development and pathology is shown in Fig. 1.
A window size of 1000 bp and a step size of 500 bp were employed.Significant regions were defined as those containing ≥2 cytosine-phosphorothioate-guanines (CpGs) with a mean methylation difference of at least 10% and a q-value of <0.01.MethylKit software (v.1.9.0) [20] was used to annotate peaks and evaluate the distribution of methylation peaks across genomic features.The University of California Santa Cruz genome browser [21] was used to classify genomic characteristics into four distinct types of regions: promoter, exon, intron, and intergenic.To understand the molecular mechanisms associated with DMR, gene ontology (GO) enrichment analysis was performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) tool (National Institutes of Health, Bethesda, MD, USA) [22].

Combined Analysis of RNA Expression and DNA Methylation
To identify genes that underwent changes in RNA expression levels due to DNA methylation (i.e., a negative relationship between RNA expression and DNA methylation), the expression levels were measured using the Pearson correlation coefficient.A correlation was considered significant at p-value of <0.05.

Bisulfite Sequencing Polymerase Chain Reaction (PCR)
To determine methylation levels from the result of methylation mapping, BS-seeker2 (v.2.1.2,https://gu oweilong.github.io/BS_Seeker2/index.html) was used; the methylation levels were calculated at single-base resolution.To compare the methylation profiles on cytosinephosphorothioate-guanine (CpG) sites among all groups, the values of only CpG sites were selected (n = 2 per group).The bisulfite primers used in this study are listed in Table 1.

Quantitative Real-time Reverse Transcription PCR (qRT-PCR)
Real-time reverse transcription polymerase chain reaction (RT-PCR) was performed using a StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) and PowerUpTM 2X SYBRTM Green Master Mix (Applied Biosystems).The Ct values of genes were matched to those of the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and the relative gene expression was determined using the 2 −∆∆Ct technique (n = 2 per group) [28].The primers used in this study are listed in Table 2.

Statistical Analysis
Data were analyzed using a two-tailed t-test and are reported as the mean value ± the standard error.The level of significance was set at p < 0.05.

Confirmation of Behavioral Phenotype and Hippocampal Aβ Aggregates in Old 5xFAD tg Mice
We employed 3-and 9-month-old wild-type and 5xFAD tg mice to investigate AD-like cognitive impairment and histopathological changes in the hippocampus (Fig. 2).Fig. 2a shows illustrative images of animal movements in the NORT behavioral task used to examine the learning and memory functions.There was no significant difference in the percentage of visits to the new object between 3-month-old wild-type and 5xFAD tg mice, whereas 9-month-old 5xFAD tg mice showed a significantly de-creased visits to the new object compared with 9-month-old wild-type mice in the NORT test session (Fig. 2b).Another behavioral task, the Y-maze test, was used to assess working and reference memories.Similar to the results of NORT, no significant change was observed in spontaneous alternation behavior in the Y-maze test at 3 months, except for a slight decrease in SAP in 3-month-old 5xFAD tg mice (Fig. 2c).However, SAP levels were lower at 9-months in 5xFAD tg mice than in that of wild-type mice.
To confirm Aβ aggregates in the hippocampi of 5xFAD tg mice, immunohistochemical analysis was performed.Fig. 2d presents representative microphotographs of the Aβ-immunostained hemispheres in wild-type and 5xFAD tg mice.There were almost no 6E10-positive areas in the hippocampus of both 3-and 9-month-old wild-type mice (Fig. 2e).The 6E10-positive area comprised approximately 1.07% of the entire hippocampus in 3-month-old 5xFAD tg mic, which was significantly increased by approximately 8.32% in 9-month-old 5xFAD tg mice.We observed that 5xFAD tg mice developed age-dependent cognitive deficits and histopathology similar to those in patients with AD; therefore, we employed the hippocampal tissues of 9-month-old animals for further microarray analysis.

Genome-Wide DNA Methylation Analysis in the Hippocampus of 5xFAD tg Mouse
We performed DNA methyl-seq analysis in the hippocampi of 9-month-old wild-type and 5xFAD tg mice.As shown in the heatmap, a total of 13,777 DMRs were identified, including 4484 hypermethylated and 9293 hypomethylated regions (Fig. 3a).The number of hypomethylated regions was more than twice that of the hypermethylated regions, consistent with the fact that global DNA methylation is reduced in the brains of patients with AD [29].Nevertheless, the proportion of functional gene regions, including promoters and exons, was higher in the hypermethylated regions of 5xFAD tg mice (48% of hypermethylated regions and 23% of hypomethylated regions) (Fig. 3b).The distribution of DMR in genes was also analyzed, namely CpG island (CpGi), shore (short distance [0-2 kb] from CpGi), and other.A total of 18% and 16% of the hypermethylated regions and 1% and 12% of the hypomethylated regions were located in the CpGi and shore, respectively.

Transcriptomic Alternations in the Hippocampus of 5xFAD tg Mouse
To examine changes in gene expression in the hippocampal tissue of 5xFAD tg mice, we performed RNAseq analysis.We identified 101 DEGs between 9-monthold wild-type and 5xFAD tg mice (Fig. 4).DEGs, including 52 upregulated and 49 downregulated genes, are shown as a dot distribution map with log (FC) and log (p-value) (Fig. 4a), and a heatmap (Fig. 4b).

DNA Methylation and mRNA Levels Validation of Candidate Genes Deduced from Combined Methyl-seq and RNA-seq Analyses
Based on our criteria, we identified five candidate genes related to AD development and pathology from a combined analysis of methyl-and RNA-seq data: thymus cell antigen 1, theta (Thy1), retinoblastoma binding protein 7 (Rbbp7), A-kinase anchoring protein 2 (Akap2), myosin VI (Myo6), and filamin A-interacting protein 1like (Filip1l).In the combined analysis of microarray data, DNAs encoding Thy1 and Rbbp7 were hypomethylated, and RNA levels were increased in the hippocampus of 5xFAD tg mice compared with those of wild-type mice.Additionally, DNAs encoding Akap2, Myo6, and Filip1l were hypermethylated, and RNA levels were decreased in 5xFAD tg mice.The Pearson correlation coefficient was only detected for Thy1, Myo6, and Filip1l (Table 3).The DMRs of Myo6 and Filip1l are located in intron 1, and those of Rbbp7 and Akap2 are positioned in intron 2 of each respective gene, while that of Thy1 is confirmed to exon 4.
To validate the microarray data, we confirmed the DNA methylation status and mRNA levels of candidate genes in the hippocampi of 9-month-old wild-type and 5xFAD tg mice (Fig. 5).In terms of DNA methylation status, the methylation of genes, including Rbbp7, Myo6, and Filip1l, was significantly increased in 5xFAD tg mice, while no significant change in the methylation level of Akap2 was observed, although the tendency was slightly increased in 5xFAD tg mice (Fig. 5a).The DNA methylation level of Thy1 significantly decreased in 5xFAD tg mice.Except for the genes showing no significant methylation change (Akap2) and those that did not coincide with the methyl-seq data (Rbbp7), we performed qRT-PCR to confirm the mRNA levels of the other three genes (Thy1, Myo6, and Filip1l) in the hippocampi of wild-type and 5xFAD tg mice.The mRNA level of Thy1 was increased, while Myo6 and Filip1l were decreased in inverse accordance with their DNA methylation status, although the differences were not significant (Fig. 5b).Lastly, we summarized the gold standard dataset in Table 4.

Discussion
In this study, we performed an in-depth combined analysis of DNA methylation and gene expression in the hippocampal tissues of 5xFAD tg mice, a mouse model of AD.Cognitive impairment, including working and reference memory, was detected in 9-month-old, but not in 3month-old, 5xFAD tg mice, and the age-dependent memory deficits are similar to the behavioral phenotypes reported in other studies using the Morris water maze and contextual fear conditioning paradigms [30,31].Furthermore, Aβ production was significantly increased in the hippocampus of 9-month-old 5xFAD tg mice.Therefore, we selected the hippocampi of 9-month-old wild-type and 5xFAD mice as samples to confirm DNA methylation-regulated gene expression.A total of 13,777 DMRs, including 4484 hyperand 9293 hypomethylated regions, were identified using DNA methyl-seq, and 101 DEGs, including 52 up-and 49 downregulated genes, were identified using RNA-seq analysis.Although hypomethylated regions were much more abundant than the hypermethylated regions, the distribution of hypomethylation in genes was largely in non-functional regions, which might have affected similar numbers of upand downregulated genes.Finally, in the combined analysis, three genes, namely Thy1, Myo6, and Filip1l, were selected as candidates that potentially play epigenetic roles in the progression or aggravation of AD symptoms and histopathological lesions.Given that acquired environmental factor or aging influences the development of neurological illnesses, previous studies have employed epigenome-wide association studies to identify changes in DNA methylation linked to the characteristics of diseases [32,33].Over the past few decades, the significance of epigenetic pathways in the development of neurological disorders has been increasingly recognized [32].AD, marked by a gradual deterioration in cognitive function and neuronal death, is the most prevalent of the late-onset neurodegenerative diseases [34].Increasing evidence emphasizes the involvement of epigenetic variation in AD development since gene mutations account for only approximately 5% of all patients with AD [35].Aging is the primary risk factor for the development of AD; recent studies have indicated that both aging and AD are linked to significant alterations in the epigenetic control of gene expression, particularly DNA methylation [36,37].DNA methylation analysis has been conducted in several AD mouse models.Cong et al. [38] investigated genome-wide CpG methylation in the cortex of 11-monthold APP/PS1 mice, showing that 2346 CpG sites involving 485 genes could be critically associated with AD.The study suggested that epigenetic alternation of the gene for transforming growth factor beta 1 (TGF-β1) may have a pathological role in the early stages of AD [38].Moreover, Sanchez-Mut et al. [14] analyzed DNA methyla-tion changes in the frontal cortex of 12-month-old APP/PS1 and 18-month-old 3xTg mice through a genome-wide promoter DNA methylation array.Three hypermethylated candidate genes (thromboxane A2 receptor (TBXA2R), sorbin and SH3 domain-containing 3 (SORBS3) and spectrin beta 4 (SPTBN4)) were identified via an integrative analysis of DNA methylation changes and confirmed in the two mouse models and in AD patients, suggesting that cAMP response element-binding protein (CREB) signaling and the axon initial segment, which are associated to the three silenced genes, could contribute to the pathobiology of AD [14].Herein, we identified three candidate genes related to the symptoms and pathological progression of AD through a combined analysis of DNA methyl-and RNA-seq in the hippocampus of 9-month-old 5xFAD tg mice.First, Thy1 encoding DNA was hypomethylated, and mRNA expression was upregulated in the hippocampus of 5xFAD tg mice.Thy1 is a highly conserved glycoprotein connected to glycophosphatidylinositol and found on the surface of several cell types such as thymocytes, fibroblasts, endothelial cells, mesangial cells, and neurons [39].As regulators of cell-cell or cell-matrix interactions, the physiological effects of Thy1 vary based on cell and tissue type [39].Thy1, mainly expressed by neurons in the brain, makes up 2-7% of the total surface protein, making it one of the most common surface glycoproteins in the nervous system [40].Thy1 is specifi- cally found at the ends of axon terminals in neurons and induces actin cytoskeleton contraction, resulting in neuronal retraction [41,42].Neurite outgrowth is important for the treatment of brain injury and neurodegenerative diseases, including AD [43].In addition to its effects on neurite outgrowth, Thy1 is also directly associated with synaptic physiology.The absence of Thy1 in mice disrupts long-term potentiation in the dentate gyrus under specific conditions [44], suggesting that Thy1 plays a role in the regulation of synaptic neurotransmitter release [45,46].Hence, we propose that epigenetically increased expression of Thy1 in the hippocampus of 5xFAD tg mice may have detrimental effects on axonal regeneration and synaptic transmission associated with the development and progression of AD.
The second DNA methylation-regulated gene confirmed in this study was Myo6, which was hypermethylated; its mRNA levels decreased in the hippocampus of 5xFAD tg mice.Myo6, unlike other myosins, exhibits atypical retrograde movement toward the minus point of actin filaments [47] and is abundantly expressed in the brain, particularly at synapses [48].Osterweil et al. [48] discovered that the hippocampi of mice lacking Myo6 showed notable impairments in stimulation-induced endocytosis of the alphaamino-3-hydroxy-5-methyl-4-isoxazole propionic acid receptor (AMPAR, a glutamate receptor).Furthermore, the phosphorylation of Myo6, which is associated with mitochondrial recruitment and presynaptic filament anchoring, is involved in energy supply and calcium removal in the presynaptic region during intense synaptic activity [49].Deficits in this process induce local energy depletion and intracellular calcium overloading.Consequently, deficiency in Myo6 expression levels or function leads to unusual dendritic spines, synapse reduction, impairment of synaptic efficacy, long-term depression, and astrogliosis [48], all of which are characteristics of hippocampal lesions in patients with AD [50].Therefore, we propose that epigenetic downregulation of Myo6 expression can affect synaptic plasticity, which is associated with pathological lesions and impaired learning and memory in AD.
The last methylation-affected gene investigated in this study was Filip1l, whose DNA was hypermethylated, and mRNA level was decreased in 5xFAD tg mice.To date, Filip1l has been studied in the field of cancer and reported to be associated with epithelial-mesenchymal transition, chemoresistance, and extracellular matrix migration in cancer cells [51].Filip1l is highly similar to Filip1, with ~50% amino acid similarity [52,53].The intracellular location and functions of Filip1l are similar to those of Filip1, suggesting that Filip1l can be a Filip1 family protein that binds to filamin A and induces the degradation of filamin A, an actin-binding protein [53].Therefore, Filip1l may exert its actions via the inhibition of filamin A. Filamin A modulates the rearrangement of the actin cytoskeleton by interacting with integrins, transmembrane receptor complexes, and secondary messengers [54].Mutations in filamin A generally lead to aberrant neuronal migration and connective and vascular tissue abnormalities [55].Followed by senile Aβ plaques and neurofibrillary tangles (NFTs), a recent study suggested that a third proteopathy, an altered conformation of the scaffolding protein filamin A, could be intimately connected to the amyloid and tau pathologies in AD [56].Changes in filamin A function facilitate the continu-ous activation of toll-like receptor 4 by Aβ plaques, thereby inducing neuroinflammation [56].Therefore, the epigenetically decreased Filip1l expression in the hippocampus of 5xFAD tg mice might be involved in the development and pathogenesis of AD, possibly connected to the function of filamin A. Future investigations remain warranted to determine the physiological binding partner of Filip1l and analyze the potential link between Filip1l and filamin A.
The DMRs of the candidate genes are located inside the gene body.Methylation of promoter or control regions located outside of the transcribed sequences (e.g., enhancer and insulator) is increasingly being recognized as functionally significant in inverse gene expression [57].The roles of DNA methylation in gene body have been disregarded, but some recent studies have emphasized its importance.A whole-genome study has shown that exons are more highly methylated than introns, and changes in the degree of methylation occur at the exon-intron border, possibly suggesting a role for methylation in the regulation of splicing [58].Shukla et al. [59] has suggested that binding of the transcription repressor CCCTC-binding factor (CTCF), which can be regulated by DNA methylation, could stop the progression of RNA polymerase II and, since this would affect splicing, this might provide evidence regarding the connection between DNA methylation and splicing.Furthermore, Anastasiadi et al. [60] clearly demonstrated the important, preserved role of methylation at the level of the first intron, and its inverse correlation with gene expression irrespectively of tissue and species.These observations suggest a previously uncharacterized role for DNA methylation of gene body at the transcriptional level, resulting in changes in mRNA and protein expression via alternative splicing.However, further studies are needed to collect additional evidence for a direct correlation between DNA methylation and the expression of the candidate genes presented in this study.
Although we described three new candidate genes unknown to date through a combined analysis of DNA methylation and RNA-seq, the sequence analyses were performed on a small-scale, which is a limitation of the present study.Moreover, it remains unclear whether the confirmed epigenetic alterations are causal factors or are consequences of AD-like symptoms in 5xFAD mice.Several epigenetically regulated genes, including ABCA7, ANK1, APP, BACE1, BIN1, CDH23, DIP2A, DUSP22, homeobox, MAPT, PSEN2, RHBDF2, RPL13, SERPINF1, SERPINF2, SLC24A4, and SORL1, have been verified so far to be linked to AD neuropathology, including Aβ plaques and tau aggregates [61][62][63][64][65].Although extensive research has been conducted to understand the significance of DNA methylation in the progression of AD, the application of this information in clinical practice is still in its early stages.This could be attributed to the challenge of reaching a specific brain region of interest and to the fact that biomarkers can in general only be assessed after death.To overcome these obstacles, it is necessary to conduct prospective or retrospective studies on the progression of AD on a large scale, using blood as a substitute for brain tissue.Nevertheless, we believe that the current preclinical investigation contributes to expanding the knowledge and the pool of genes potentially associated with AD epigenetics, and provides the possibility of developing novel drugs based on DNA methylation.

Conclusions
Our study identified three epigenetically regulated genes associated with the neural cytoskeleton and synaptic plasticity in AD neuropathology through a combined analysis of DNA methyl-and RNA-seq in the hippocampus of wild-type and 5xFAD tg mice.DNA methylation may have a significant impact on the preservation of normal hippocampal functions, such as learning and memory [66], hence alterations in DNA methylation could be linked to the disease activity, progression, and clinical prognosis of AD [32].Therefore, although no treatment based on DNA methylation has been developed yet, counteracting epigenetic changes may facilitate the development of new therapeutic interventions to delay or ameliorate the symptoms of AD.
gional Innovation Strategy through the National Research Foundation of Korea funded by the Ministry of Education (2021RIS-002).

Fig. 2 .
Fig. 2. Cognitive deficits and Aβ burden in 9-month-old 5xFAD tg mice.(a) Representative illustrations of mouse movements in the NORT.(b) Percentage of visits to objects A or B in the NORT test session.(c) SAP in Y-maze test.(d) Representative microphotographs of the 6E10-immunostained hemisphere.Boxes with dotted lines represent hippocampal areas.(e) Percentage of the 6E10-positive area in the hippocampus.*p < 0.05, **p < 0.01, and ***p < 0.001 indicate significance between the selected groups.5xFAD, 5xfamiliar Alzheimer's disease; SAP, spontaneous alternation percentage; mon, months; ns, not significant; NORT, novel object recognition test.

Fig. 3 .
Fig. 3. Global DNA methyl-seq analysis in the hippocampi of wild-type and 5xFAD tg mice.(a) Heatmap showing DNA hypermethylated regions of 4484 and hypomethylated regions of 9293 in the hippocampi of 5xFAD tg mouse.(b) Venn diagrams showing the distribution of hyper-or hypomethylation in genes.(c) Bar graphs showing the top 10 GO biological processes associated hyperand hypo-methylated genes.The number aside the GO term represents the number of DMGs related with the GO.5xFAD, 5xfamiliar Alzheimer's disease; DMR, differentially methylated region; DMG, differentially methylated gene; hyper, hypermethylation; hypo, hypomethylation; CpGi, CpG island; GO, gene ontology.

Fig. 4 .
Fig. 4. Gene expression profiles in the hippocampus of wild-type and 5xFAD tg mice.(a) Scatter plot showing the correlation between the fold change in RNA expression and p-value.(b) Heatmap showing 52 upregulated and 49 downregulated genes in the hippocampi of 5xFAD tg mouse.(c) Bar graphs showing GO biological processes associated up-and down-regulated genes.The number aside the GO term represents the number of DEGs related with the GO.5xFAD, 5xfamiliar Alzheimer's disease; FC, fold change; GO, gene ontology; DEGs, differentially expressed genes.