Alzheimer’s disease (AD) is a degenerative disease of the central nervous system characterized by progressive cognitive and behavioral impairment, which seriously affects the quality of life of the elderly (Shah et al., 2016; Chhatwal et al., 2022). AD is caused by complex interactions between genetic and environmental factors and lacks effective treatment measures. Neuropathological features of AD, including synaptic loss, synaptic disorder, and impaired plasticity, may form the pathological bases behind the development of cognitive impairment (Zhu et al., 2018; Qu et al., 2021). As the structural basis of brain function, synaptic connections continue to undergo dynamic changes. During development, learning, stress, and other processes, synapses exhibit remarkable plasticity and are easily disturbed by many factors. Mounting evidence has shown that the abnormal aggregation of amyloid-beta (Aβ), tau, and other proteins can cause synaptic disorder in fragile brain regions by disturbing and prolonging long-term potentiation (LTP). As a result, there are changes in postsynaptic membrane receptors, altering synaptic plasticity, and changing the morphology and density of dendritic spines (Jeong et al., 2021). These lead to decreased synapses and eventually cause various disorders, primarily cognitive and memory failure (Zhang et al., 2008; Bufill et al., 2013). In this review, we postulate that the mechanisms underlying the abnormal aggregation of Aβ, tau, and other proteins which abrogate synaptic function may be epigenetic in nature. Epigenetics focuses on the study of changes in gene expression without any changes in DNA sequence. A study found that the PP2A gene encoding the protein phosphatase 2A was in a demethylated state in the hippocampus of AD patients and mice (Zhou et al., 2008). Demethylation at L309 of PP2A leads to an increase in tau protein, while reduction of methylation at L309 reduces tau hyperphosphorylation, thus reducing intersynaptic protein function. Scientists have found that micro RNA (miRNA) imbalance in AD patients and models is related to amyloid precursor protein (APP) expression, production of Aβ, deposition of pathological tau protein, inflammatory response, and damage to the cytoskeleton, leading to synaptic disorder (Wang et al., 2019; Jash et al., 2020). In addition, several studies have revealed that histone acetylation is related to synaptic disorders and learning and memory disorders (Graff et al., 2014; Yamakawa et al., 2017). Therefore, epigenetic studies have opened a new chapter for exploring the pathogenesis and treatment of these disorders. This article reviews the epigenetic mechanisms and treatments regarding DNA methylation, histone modifications, and RNA modifications in synaptic dysregulation in AD.

In a healthy adult brain, the hippocampus and cerebral cortex functions are crucial for learning and memory. LTP plays an essential role in their maintenance. LTP is a persistent enhancement phenomenon that occurs in the signal transmission of two neurons, which can stimulate two neurons synchronously, which is one of several phenomena related to synaptic plasticity, which is the ability of synapses to change strength (Jeong et al., 2021). Sustained LTP is closely associated with increased synaptic calcium ion (Ca²⁺) levels and the enhancement of kinase activity, which maintains healthy neural activity and connections (Dudai and Morris, 2013). In pathological tissue sections of AD patients, it was found that Aβ oligomers can activate calcineurin in cell signaling pathways, leading to dendritic spine loss and resulting in LTP disorders. This triggers synaptic damage and impairs learning and memory function (Alberdi et al., 2010). It was also found that the lack of PP2A led to axon growth disorders in AD mouse models (Kuchibhotla et al., 2014). In the transgenic rTg4510 mice expressing tau protein with P301L mutation, mutant tau can cause dendritic spine loss and synapse collapse, causing severe cognitive impairment (Kuchibhotla et al., 2014).

Synaptic failure is a critical component of the neurodegenerative process of AD, especially in early AD (Spires-Jones and Hyman, 2014; Wu et al., 2021). Synaptic plasticity denotes changes in the efficacy of synaptic transmission in response to neuronal activity; it also promotes related structural plasticity responses exemplified by dendritic spine remodeling (Weng et al., 2011). In AD mouse models, highly expressed delta-secretase induced the formation of senile plaques and neurofibrillary tangles in the brain, causing neuroinflammation and neuron loss, leading to an apparent cognitive disorder. Overexpression of delta-secretase significantly reduced the formation of dendritic spines in wild-type hAPP/hMAPT double transgenic mice. These findings support that delta-secretase plays an essential role in AD development by influencing synaptic disorders (Latina et al., 2018).

Synaptic disorders can directly lead to memory and behavioral dysfunction, and understanding its changes and regulatory mechanisms is essential. Therefore, there are significant synaptic disorders in AD, including synapse loss, changes in the morphology and number of dendritic spines, and synaptic transmission disorders (Figure 1).

There are many causes of synaptic dysregulation in AD, including cytoskeleton structure destruction caused by excessive deposition of Aβ and tau proteins, neurotransmitter system disorder (Bukke et al., 2020). Other factors include an inflammatory immune response (Uddin et al., 2021), lipid and issues of biological energy metabolism (Xiang et al., 2015). Oxidative stress, and epigenetic alterations also play a role in synaptic dysregulation (Mastroeni et al., 2015). Epigenetic regulation disorders play a crucial role in the pathogenesis of AD, including abnormal DNA methylation, histone modification, and RNA interference, and have essential value in the pathogenesis of cellular metabolic and synaptic disorders (Van den Hove et al., 2014; Nikolac et al., 2021).

Excessive Aβ binding to N-methyl-D-aspartate receptors depolarizes postsynaptic neurons, increases Ca²⁺ influx, activates multiple Ca²⁺-dependent signaling pathways, and changes downstream-related proteins, including calpain and calcineurin enzymes (De Felice et al., 2009). Inactivation of PP2A and increased Fyn levels lead to in mitochondrial permeability and functional breakdown changes, which ultimately leads to synaptic disorder (De Felice et al., 2009; Kamat et al., 2016). It is also involved in the inflammatory immune response, oxidative stress, and other causes of nerve cell metabolism disorders. It subsequently leads to synaptic disorder resulting in learning and memory disorders. However, the initiating factors that induce synaptic lesions include genetic and environmental factors. Individuals with abundant Aβ and tau protein expression have not yet developed AD, and approximately 1/2 of identical twins have AD (Gatz et al., 2006).

Furthermore, APP, beta-site amyloid precursor protein cleaving enzyme (BACE), and presenilin-1 (PSEN1) gene mutations are common in AD patients, and similar promoter mutations have different clinical manifestations in some cases. As a result, APP, BACE, and PSEN1 genes contain controllable methylated CpG sites (Ko et al., 2015; Tecalco-Cruz et al., 2020). Autopsy of AD patients revealed that the APP gene in the cortex was completely demethylated. However, no similar findings were noted in the healthy controls or patients with Pick disease. In vitro studies have found that hypomethylation caused by folic acid deprivation enhances the expression of BACE and PSEN1 genes, which returns to normal after folic acid supplementation (Fuso et al., 2008).

Therefore, we believe that epigenetic regulation is involved in AD neuropathological processes under the influence of the environment and other factors. Epigenetic regulation of synaptic function in AD is also involved in regulating synaptic disorders and plasticity and has an essential impact on learning and cognitive functions.

Synaptic disorders in AD can be improved by regulating epigenetic mechanisms. In 2019, Roman and others reported that reduced levels of SAM induce DNA demethylation, leading to overexpression of pathologically related genes in AD (Roman et al., 2019). Methionine reacts with ATP to form SAM, SAH, and homocysteine. According to a recent study, moderately elevated homocysteine increased the relative risk of dementia in older adults from 1.15 to 2.5 times, with population-attributable risk rising from 4.3 to 31% (Raszewski et al., 2016). Therefore, homocysteine reduction therapy with vitamin B may significantly slow the rate of brain shrinkage and cognitive decline in older adults by upregulating DNA methylation (Smith et al., 2018). Li et al. (2016) showed that folic acid upregulated DNA methylation in N2a-APP cells and AD transgenic mice and analyzed the role of folic acid in JAK-STAT and LTD signaling pathways and its value in AD treatment. Therefore, increasing methylation levels and activities may represent a potential therapeutic strategy for ameliorating AD (Wang et al., 2013; Li et al., 2016).

MiRNA-206 is highly expressed in APP/PSEN1 transgenic mice, mainly in the plasma, cerebrospinal fluid, and hippocampus. It is associated with the downregulation of BDNF, which has also been observed in AD patients (Tian et al., 2014; Silvestro et al., 2019). Intranasal AM206, an antagonist of miRNA-206, was found to block the Aβ pathogenicity (1-42) and increase BDNF levels, synaptic density, and neurogenesis (Satoh, 2012). Similarly, Wang and others found that melatonin supplementation could reverse EPACs/miR-124/Egr1 signal conduction and restore the amplitude and slope of field excitatory postsynaptic potential (Wang et al., 2013). They also found it can improve the density of spines and percentage of mushroom spines in CA1 neurons and reduce synaptic disorder and memory deficits in an anisodamine-induced AD model (Wang et al., 2013).

Cao et al. (2020) found that the H3K4me3 histone marker and catalytic enzyme were significantly increased in the prefrontal cortex in AD patients and mice models. Subsequently, introducing a specific Sgk1 inhibitor into mice reduced excessive phosphorylated tau proteins, restored the prefrontal cortex and glutamate synaptic function, and improved the memory defects in AD mice (Cao et al., 2020). Additionally, several AD mouse models have been treated with HDAC inhibitors, such as sodium butyrate, trichostatin A, and valproic acid, showing improvements in learning and memory, some resulting from reduced Aβ levels (Sung et al., 2013). A phase 3 clinical trial of valproate-regulated histone acetylation-modifying therapy in dementia did not improve outcomes in dementia patients (Tariot et al., 2011). Insufficient evidence supports valproic acid in treating dementia for cognitive, psychiatric symptoms or disease-modifying (Liu and Wang, 2020). The failure of these clinical trials may be due to insufficient plasma concentrations of sodium valproate to engage relevant molecular targets, particularly glycogen synthase −3β. In practice, patients cannot use higher doses of valproate.

Chatterjee et al. (2018) identified interaction partners of target proteins and epigenetic targets through the human protein interaction network. These networks include DNA methyltransferase inhibitors, histone deacetylase inhibitors, histone acetyltransferase modulators, histone methyltransferase inhibitors, and histone demethylase inhibitors. These can be used as epigenetic drugs to reverse the epigenetic changes and correct memory deficits and behavioral abnormalities after improved synaptic disorder in AD, which can help develop new drugs for AD treatment (Ganai et al., 2016; Chatterjee et al., 2018). Some researchers have also launched a phase I clinical trial of Vorinostat (an HDAC inhibitor) in the treatment of mild Alzheimer’s disease. But so far there is no clear conclusion of safety and efficacy. Therefore, we look forward to further large-scale clinical studies.

In summary, epigenetic modifications play an essential role in the occurrence and development of synaptic dysregulation in AD. Modification abnormalities, such as DNA methylation, histone modification, and RNA interference, lead to abnormal transcription of AD-related genes, resulting in dysfunction of related proteins and enzymes, synaptic disorder, and memory behavior disorders. In theory, diseases caused by epigenetic modifications can be reversed with drugs. For example, folic acid supplementation can increase DNA methylation and inhibit Aβ deposition. Melatonin can reverse EPACs/miR-124/Egr1 signaling, improving the density of spines and the percentage of mushroom spines in CA1 neurons. Sgk1 inhibitors can reduce hyperphosphorylated tau proteins by regulating H3K4me3 and restoring PFC glutamate synaptic function, providing a new approach for AD treatment. However, the clinical application of epigenetic drugs may be a long way off due to the unknown side effects and complex co-regulatory mechanisms that limit their development. Therefore, further research on the epigenetic mechanisms in the pathogenesis and development of synaptic dysregulation in AD is essential to discover safe and specific drugs that regulate epigenetic mechanisms.

ZC and MW: conception and design and administrative support. ZC, MW, QL, WZ, XW, and XY: provision of study materials. ZC, MW, and QL: collection and assembly of data. ZC, XW, and XY: data analysis and interpretation. XY: revision of the final version. All authors: manuscript writing and final approval of the manuscript.