Crosstalk Between N6-Methyladenosine and Other Epigenetic Mechanisms in Central Nervous System Development and Disorders

Abstract

A variety of epigenetic mechanisms—such as DNA methylation, histone alterations, RNA chemical modifications, and regulatory non-coding RNAs—collectively influence gene regulation and cellular processes. Among these, N6-methyladenosine (m⁶A) represents the most widespread internal modification in eukaryotic mRNA, exerting significant influence on RNA metabolic pathways and modulating mRNA function at multiple levels. Studies have shown that m⁶A modification is highly enriched in the brain and regulates central nervous system development and various physiological functions. Recent studies have demonstrated that m⁶A interacts with other epigenetic regulators and triggers epigenetic remodeling, which further affects the development and occurrence of central nervous system diseases. In this review, we provide an up-to-date overview of this emerging research hotspot in biology, with a focus on the interplay between m⁶A and other epigenetic regulators. We highlight their potential roles and regulatory mechanisms in epigenetic reprogramming during central nervous system development and disease, offering insights into potential novel targets and therapeutic strategies for CNS disorders.

1. Introduction

Epigenetic regulation involves transmissible changes in gene activity that occur without altering the underlying DNA sequence. These modifications typically arise from chemical alterations to DNA or chromatin-associated proteins. Key epigenetic mechanisms include DNA methylation, histone tail modifications, regulatory non-coding RNAs (ncRNAs), and various RNA chemical marks. Among these, RNA modifications are particularly widespread in eukaryotic cells, with more than 170 types identified to date [1]. Notably, RNA methylation represents the majority, comprising approximately two-thirds of all RNA modifications. Common types include N6-methyladenosine (m⁶A), 5-methylcytosine (m⁵C), 1-methyladenosine (m¹A), 5-hydroxymethylcytosine (hm⁵C), pseudouridine (ψ), and 2′-O-methylation (Nm) (Figure 1A) [2,3]. Among them, m⁶A, first reported in 1974, is the most prevalent methylation mark in eukaryotic RNA, constituting around 80% of total RNA methylation events [4,5]. It participates in regulating multiple biological functions of RNA, but was only recently studied due to the limitations of early technology [6,7,8]. The regulation of m⁶A mainly depends on the methyltransferase complex, demethylases, and methyl-binding proteins, which dynamically adjust the modification through demethylation [9,10]. The interaction between m⁶A and other epigenetic regulators has been increasingly recognized as a key contributor to epigenetic remodeling, further adding to the complexity of epigenetic mechanisms. This is currently a hot research topic.

Central nervous system (CNS) development is a highly coordinated process that requires regulation at multiple levels, including epigenetic modifications, to ensure the formation of normal functions. Sun et al. and Huang et al. demonstrated that m⁶A is highly enriched in the mammalian brain and plays essential roles in regulating early central nervous system (CNS) development as well as maintaining physiological functions in adulthood [11,12]. For example, m⁶A-mediated gene regulation has been demonstrated to be a crucial step in neural cell differentiation, and m⁶A controls the self-renewal and differentiation of embryonic stem cells in mammals [13]. However, the role of m⁶A in adverse events in the CNS is still poorly understood. The crosstalk between m⁶A and other epigenetic regulators increases the complexity of epigenetic reprogramming and involves multiple aspects, including CNS development and neurological and psychiatric disorders [9].

It is important to critically acknowledge that this research field still faces substantial limitations and knowledge gaps. For instance, the functional consequences of specific m⁶A modification sites remain poorly understood, and there is currently a lack of standardized tools that enable the precise manipulation of m⁶A levels at defined loci [14]. Although several small-molecule inhibitors, such as FTO inhibitors, have demonstrated therapeutic potential in preliminary studies, their target specificity, bioavailability, and long-term safety have not yet been fully validated. Moreover, existing studies are predominantly focused on developmental stages or cancer biology, with relatively limited systematic investigation into the role of m⁶A in specific neuropsychiatric or neurodegenerative diseases. Finally, the in vivo bidirectional regulatory mechanisms between m⁶A and other epigenetic layers—such as histone modifications and non-coding RNAs—remain largely unexplored, which constrains the broader application of m⁶A research in the context of epigenetic reprogramming.

This review systematically summarizes recent progress in this emerging field, with a particular focus on the interplay between m⁶A and other epigenetic mechanisms. It further highlights the potential roles and regulatory pathways of m⁶A and its modulators in mediating epigenetic reprogramming during neurodevelopment and disease pathogenesis.

2. Molecular Regulatory Mechanisms and Biological Functions of m⁶A

2.1. Molecular Regulatory Mechanisms of m⁶A

The regulation of m⁶A modification is mediated through the interplay of three classes of proteins, methyltransferases (“writers”), demethylases (“erasers”), and m⁶A-binding proteins (“readers”) [15], which, together, determine the methylation status and functional outcomes of target transcripts (Figure 1B).

The installation of m⁶A marks on RNA is predominantly carried out by a methyltransferase complex, in which methyltransferase-like 3 (METTL3) functions as the enzymatic core. This writer complex also comprises essential cofactors, including methyltransferase-like 14 (METTL14), which facilitates RNA substrate recognition, and Wilms tumor 1-associating protein (WTAP), which contributes to the localization and stability of the complex [16]. METTL3 and METTL14 form a stable heterodimer, which is central to both substrate recognition and methyl group transfer [13]. WTAP, in turn, interacts with this core dimer to ensure proper subnuclear localization within nuclear speckles and to facilitate full methylation activity [17]. Notably, WTAP depletion impairs the RNA-binding ability of METTL3 [18]. Beyond these core elements, several accessory proteins—such as VIRMA (also known as KIAA1429), ZC3H13, RBM15, and CBLL1—associate with the complex to refine methylation site selection and modulate transcription-linked signaling pathways [19,20].

The identification of demethylases has revealed that m⁶A modification is a reversible and dynamically regulated process. Two key enzymes responsible for m⁶A demethylation are fat-mass- and obesity-associated protein (FTO) and AlkB homolog 5 (ALKBH5). FTO has been shown to remove both m⁶A and N6,2′-O-dimethyladenosine (m⁶Am) modifications from mRNA, with a markedly higher catalytic preference—approximately 100-fold—for m⁶Am over m⁶A [21,22,23]. However, recent data do not support FTO as an eraser of m⁶A, and indicate that FTO does not remove m⁶A under normal physiological conditions, but only acts on m⁶Am [22,24,25]. Therefore, the removal of m⁶A by FTO may be the result of environmental and cell-type-specific regulation. ALKBH5 reverses m⁶A by oxidation, affecting mRNA output, metabolism, and the assembly of mRNA processing factors in nuclear speckles [26,27]. In summary, knocking down FTO and ALKBH5 in cell lines significantly increases the level of m⁶A modifications on mRNA.

A specific RNA-binding protein (RBP) is required for mRNA with m⁶A modification to perform specific biological functions, commonly found in the conserved YT521-B homologous domain (YTH) family of proteins, including YTH domain family proteins (YTHDF) 1, 2, and 3 and YTH-domain-containing proteins (YTHDC) 1 and 2, which bind to m⁶A-modified RNA through the m⁶A binding domain [28,29]. YTH-domain-containing proteins are key mediators of m⁶A function, influencing various aspects of mRNA metabolism, including splicing, nuclear export, translation, and degradation [30,31]. In addition to YTH proteins, other RNA-binding proteins (RBPs) can recognize m⁶A-modified transcripts through conventional RNA-binding domains. These include heterogeneous nuclear ribonucleoproteins (hnRNPs) [32], fragile X mental retardation protein (FMRP) [33], insulin-like growth factor 2 mRNA-binding proteins (IGF2BPs) [34], and eukaryotic initiation factor 3 (eIF3) [33], among others. Interestingly, recent findings suggest that certain RBPs, such as G3BP1 and G3BP2, may be repelled by m⁶A marks, thereby adding further complexity to m⁶A-mediated post-transcriptional gene regulation [35].

2.2. Biological Functions of m⁶A

m⁶A can affect a wide range of downstream functions, including RNA splicing and output from the nucleus, as well as the translation, degradation, and stability regulation of mRNA in the cytoplasm (Figure 1B).

m⁶A can regulate RNA splicing and output through various mechanisms. In the nucleus, m⁶A splicing is mediated by several binding proteins, which regulate the selective splicing of target transcripts by directly binding to YTH-domain-containing proteins such as YTHDC1 [36], or by relying on m⁶A structural switches such as hnRNPC and hnRNPG [37]. In terms of nuclear output, YTHDC1 promotes m⁶A-modified transcript output mediated by nuclear RNA export factor 1 (NXF1) through interaction with SRSF3 [38], while FMRP binds to m⁶A-modified RNA and promotes its nuclear output through exportin 1 (CRM1) [39]. Although m⁶A modification has an unstable effect on target RNA, a positive correlation has been observed between m⁶A and mRNA translation efficiency. In the cytoplasm, YTHDF1 and YTHDF3 bind near the stop codon of mRNA, promoting translation [40,41]. YTHDF1 has been shown to enhance mRNA translation by directly associating with the eIF3 complex [42,43]. Additionally, m⁶A modifications themselves can facilitate cap-independent translation through a direct interaction with eIF3, independent of YTH proteins [44]. However, emerging evidence has questioned the traditional classification of YTHDF1 and YTHDF3 as translation enhancers. Instead, it has been proposed that all three YTHDF family members may primarily contribute to mRNA degradation driven by m⁶A modification [6]. Among the YTHDF proteins, YTHDF2 appears to play the most prominent role in promoting mRNA decay. It can mediate degradation through multiple mechanisms: by directing m⁶A-tagged transcripts to cytoplasmic processing bodies (P-bodies) [30], by recruiting the CCR4-NOT deadenylase complex to initiate degradation independently of P-bodies [45], or by cooperating with adaptor protein HRSP12 and ribonuclease RNase P/MRP to cleave target RNAs [46]. Conversely, some proteins have been identified that stabilize m⁶A-modified transcripts. For instance, FMRP has been reported to bind competitively with YTHDF2 to shared m⁶A-marked targets, thereby preventing their degradation and supporting transcript stability [47]. In addition, IGF2BP1-3 have also been shown to stabilize mRNA expression by recognizing m⁶A-marked mRNA [48,49].

3. The Role of m⁶A in the Development of the Central Nervous System

m⁶A is highly abundant in the mammalian brain and serves as a critical regulator of neural development, contributing to the regulation of various processes, such as neurogenesis, glial cell formation, and axonal development (Figure 2A).

3.1. Neurogenesis

Neurogenesis is a multifaceted biological process wherein neural stem cells proliferate, differentiate into lineage-specific progenitors, migrate to target brain regions, undergo structural and functional remodeling, and, ultimately, form synaptic connections essential for neural circuit integration. Yoon et al. revealed that m⁶A plays an important role in regulating cortical neurogenesis [50]. Specifically, the loss of METTL14 in the E13.5 mouse cortex led to a downregulation of m⁶A modification on transcripts related to the cell cycle, stem cell maintenance, and neuronal differentiation, resulting in the prolonged cell cycle progression of radial glia cells (RGCs) (Figure 2B) [50]. This suggests that m⁶A can mediate cell cycle exit, which is necessary for maintaining normal cortical neurogenesis. Neuroglial cells (NSCs) lacking YTHDF2 exhibit decreased proliferation and differentiation, which is associated with transcripts regulating the JAK-STAT pathway. Neurons originating from these NSCs exhibit reduced neurite outgrowth and heightened vulnerability to oxidative stress (Figure 2C) [51]. Moreover, the loss of YTHDF2 markedly disrupts the self-renewal capacity of embryonic cortical neural progenitor cells and impairs neurogenesis. This disruption correlates with the upregulation of m⁶A-modified transcripts involved in neurodevelopment and cortical neuronal differentiation, underscoring the critical role of YTHDF2-dependent mRNA decay in orchestrating cortical neurogenesis during embryogenesis.

3.2. Gliogenesis

NSCs are the stromal cells of the nervous system, characterized by processes but without dendrites and axons, accounting for about 90% of the total cells in the CNS, and mainly play roles in repair and regeneration, support, the maintenance of substance metabolism, and nutrition. In YTHDF2- and METTL14-deficient NSCs, it has been shown that NSCs cannot differentiate into astrocytes, affecting the generation of astrocytes (Figure 2C) [50,51]. The novel m⁶A-binding protein proline-rich coiled-coil 2A (PRRC2A) regulates the proliferation of oligodendrocyte progenitor cells and the generation of oligodendrocytes. PRRC2A binds to the CDS of Olig2 mRNA with the GGACU motif of m⁶A modification, resulting in a reduction in oligodendrocyte myelin formation and inducing motor and cognitive defects (Figure 2D) [52]. In addition, the loss of the METTL14 gene in oligodendrocytes can lead to a reduction in specific neurofascin (NfASC) splice variants, causing abnormal Ranvier node morphology and the abnormal splicing of the paranodal antibody–neurofascin 155 (NF155) transcript, ultimately resulting in a reduction in the number of oligodendrocytes [53]. Therefore, m⁶A modification plays a crucial role in neuroglial cell generation.

3.3. Axonal Growth

Axons are the primary channels for neuronal signal transmission in the nervous system, which extend along specific pathways over long distances to reach their targets and form precise neural connections. Axonal growth is influenced by various factors, among which guidance molecules with attractive or repulsive effects play a major role. m⁶A can affect the mRNA translation involved in axonal guidance and extension. Qi et al. reported that FTO, a demethylase, and its binding protein YTHDF1, respectively, bind to axonal RNAs, suggesting their potential involvement in the regulation of axonal mRNA metabolism [54]. FTO promotes Gap43 translation via demethylation in the axons of dorsal root ganglion (DRG) neurons, thereby facilitating axonal elongation. In the dorsal spinal cord (DSC), YTHDF1 recognizes methylated axon-guidance-related mRNA Rono3.1 and promotes its translation, thereby regulating axonal guidance [55]. Recent studies have demonstrated that both YTHDF1 and YTHDF2 are present in cerebellar granule cells and their axonal projections [26]. Silencing either of these m⁶A reader proteins in granule cell axons leads to enhanced parallel fiber elongation and increased cerebellar synaptogenesis. This effect appears to be linked to the local translation of WNT signaling components—Dvl1 and Wnt5a—regulated by YTHDF1 and YTHDF2, respectively. These findings highlight the pivotal role of m⁶A modification in modulating axonal development.

4. m⁶A Regulates Physiological Functions of the Central Nervous System

4.1. Behavioral Adaptation

Experimental evidence in adult mice indicates that fear-based behavioral training can dynamically modulate m⁶A in the brain. Exposure to both contextual and cued fear conditioning transiently suppresses FTO expression in dorsal hippocampal CA1 neurons and the medial prefrontal cortex, which, in turn, elevates m⁶A levels on a subset of mRNA transcripts [56,57]. The depletion of FTO before training further enhances contextual fear memory (Figure 3B). Under restraint stress conditions, mice with hippocampal METTL3 and FTO deficiencies show lower resilience to stress stimuli, suggesting that m⁶A-mediated mRNA modification may improve stress adaptation [58,59]. m⁶A is also considered crucial for reward mechanisms, as impaired reward stimulation and altered behavioral adaptability are observed in mice lacking FTO [60]. FTO deficiency is a protective factor against chronic stress and relieves depressive and anxiety-like behaviors in FTO knockout mice [61]. However, the absence of FTO, METTL14, or YTHDF1 shows impairments in the cognitive behavior of mice. Overall, abnormal changes in m⁶A can lead to various abnormal neurobehavioral changes, inhibit behavioral adaptability, and induce the occurrence of neuropsychiatric disorders.

4.2. Learning and Memory

Several studies on the role of m⁶A in the central nervous system emphasize its importance in learning and memory processes (Figure 3B). For example, METTL3 has a direct role in regulating hippocampus-dependent memory consolidation by promoting the translation of early response genes in neurons [62]. This is related to the increased translation efficiency of early genes mediated by METTL3-mediated m⁶A modification, which is crucial for long-term memory consolidation. Emerging evidence links m⁶A reader proteins to the regulation of learning and memory. Upon neuronal activation, YTHDF1 enhances the translation of m⁶A-modified mRNAs in neurons, facilitating synaptic plasticity and cognitive function. YTHDF1 knockout mice exhibit deficits in learning and memory, accompanied by impaired synaptic transmission and reduced long-term potentiation in the hippocampus [7]. Remarkably, the targeted re-expression of YTHDF1 in the adult hippocampus can rescue these impairments. These findings suggest that the YTHDF1-mediated translational control of m⁶A-tagged transcripts plays a pivotal role in cognitive processes. Thus, m⁶A modification may represent a key epigenetic mechanism underlying memory decline in various neurological disorders.

4.3. Circadian Rhythm

The regulation of the 24 h rhythm of biochemistry, physiology, and behavior by the circadian clock is crucial for maintaining normal health status [63]. The circadian clock has been implicated in the pathophysiology of sleep disturbances, neurodegenerative conditions such as Alzheimer’s disease, and various psychiatric disorders including depression. Evidence from animal studies indicates that diminished m⁶A modification on key circadian transcripts—such as Per1, Per3, Tef, Dbp, Nfil3, Bhlhe41, and Nr1d—can extend circadian cycles in mice, implying that m⁶A is essential for sustaining clock gene rhythmicity [64]. Specifically, METTL3 deficiency leads to the nuclear retention of unmethylated Per2 and Bmal1 transcripts, disrupting the temporal expression of core clock genes (Figure 3B). Moreover, the elimination of the m⁶A-modified region in the CK1δ gene, which encodes a kinase involved in clock protein turnover, enhances its translation in the brain and results in prolonged behavioral rhythms in mice [64]. These findings underscore the pivotal role of m⁶A methylation in the regulation of circadian timing mechanisms.

5. The Regulation of Central Nervous System Development and Diseases by the Crosstalk Between m⁶A and Other Epigenetic Factors

5.1. Inter-Crosstalk Between m⁶A and Other Epigenetic Factors

5.1.1. Inter-Crosstalk Between m⁶A and DNA Methylation

Among various epigenetic mechanisms, DNA methylation is one of the most prevalent in mammals, playing a key role in the regulation of gene expression [65,66]. This modification predominantly occurs at CpG islands within gene promoter regions. In both eukaryotic and prokaryotic genomes, the major forms of DNA methylation are 5-methylcytosine (5mC) and N6-methyladenine (6mA) [67,68,69]. However, the crosstalk between DNA methylation and m⁶A leads to diverse changes in both modifications. m⁶A modification on DNA methyltransferase genes promotes the stability of DNA methyltransferase [70]. Conversely, DNA methyltransferase interacts with m⁶A methyltransferase or demethylase to regulate m⁶A levels. For example, in pancreatic cancer, exposure to cigarette smoke condensate (CSC) has been shown to interfere with the recruitment of DNA methyltransferases DNMT1 and DNMT3A to the METTL3 promoter. This disruption reduces DNA methylation at the METTL3 locus, leading to elevated METTL3 expression and a subsequent increase in global m⁶A levels [71]. In esophageal squamous cell carcinoma, CSC can also cause the low methylation of DNMT1 and DNMT3a in the ALKBH5 CpG island, leading to the low expression of ALKBH5 and the downregulation of m⁶A levels (Figure 4A) [72]. Therefore, there is a wide range of co-transcriptional interferences between m⁶A and DNA methylation modifications.

5.1.2. m⁶A and Histone Modification

Histone modifications encompass several chemical alterations, including acetylation, methylation, phosphorylation, ubiquitination, SUMOylation, and ADP-ribosylation, with methylation and acetylation being the most extensively studied [73]. Methylation typically targets arginine (R) and lysine (K) residues, whereas acetylation predominantly occurs on lysine residues. Both modifications influence gene transcription by either promoting or repressing expression, depending on the specific residues and the context of modification. Studies have found that there is also co-transcriptional interaction between RNA modification and histone modification. For example, the m⁶A methyltransferase complex METTL3-METTL14 interacts with acetyltransferase and methyltransferase to downregulate H3K27ac and upregulate H3K27me3 levels, respectively (Figure 4B). YTHDC1, an m⁶A reader protein, has been shown to associate with the histone demethylase KDM3B, guiding its recruitment to m⁶A-enriched chromatin loci. This interaction facilitates the removal of H3K9me2 marks and enhances the transcription of associated genes (Figure 4D) [74]. Histone modification enzymes can also affect the transport of m⁶A-binding proteins. For instance, after infection with herpes simplex virus-1 (HSV-1), the m⁶A-binding protein Recombinant Mouse Heterogeneous nuclear ribonucleoproteins A2/B1 (hnRNPA2B1) recognizes viral DNA and undergoes homodimerization. Histone demethylase jumanji domain-containing 6 (JMJD6) induces hnRNPA2B1 demethylation, promoting hnRNPA2B1 translocation from the nucleus to the cytoplasm, inducing IFN-α/β production through activating the TBK1-IRF3 pathway and mediating immune response (Figure 4C) [75].

5.1.3. m⁶A and ncRNAs (miRNAs, lncRNAs, and circRNAs)

Non-coding RNAs (ncRNAs), a diverse class of RNA molecules that do not encode proteins, play essential regulatory roles in epigenetic processes. Among them, microRNAs (miRNAs), long non-coding RNAs (lncRNAs), and circular RNAs (circRNAs) have attracted the most attention in epigenetic research [76].

miRNAs are a type of small endogenous RNA, typically 20–24 nucleotides in length. In mammalian systems, the METTL3-mediated methylation of primary miRNAs (pri-miRNAs) enhances their recognition by the RNA-binding protein DGCR8, thereby promoting the efficient processing and maturation of miRNAs (Figure 4E) [77]. Additionally, heterogeneous nuclear ribonucleoprotein A2/B1 (hnRNPA2B1) supports this process by interacting with DGCR8 and facilitating its association with pri-miRNAs, thus contributing to sustained miRNA biogenesis [71,77,78,79]. Since both m⁶A on mRNA and miRNA binding sites are located on the 3′UTR, it is speculated that m⁶A and miRNA may have co-binding effects. Zhang et al. reported that m⁶A modification exists in the 3′UTR of Yes-associated protein 1 (YAP), and this modification promotes the binding of miR-582-3p to YAP [80]. Therefore, m⁶A modification may trigger the binding of miRNA and target genes (Figure 4F). Conversely, miRNA can also control m⁶A levels by inhibiting the expression of m⁶A demethylases [81]. Therefore, there is a bidirectional relationship between miRNA and m⁶A, and the positive and negative feedback loops illustrate the complex relationship between miRNA and m⁶A.

LncRNAs are RNA molecules with lengths generally ranging from 200 to 100,000 nucleotides, some of which fine-tune gene expression through diverse mechanisms in a tissue-specific manner [82]. Recent studies have revealed that the lncRNA regulation of genes is influenced by m⁶A modification, which affects lncRNA degradation and stability, thereby regulating gene expression (Figure 4G) [83,84,85,86]. On the other hand, m⁶A increases lncRNA accessibility to RNA-binding proteins (RBPs) by exposing its purine-rich sequence. Therefore, m⁶A-dependent structural changes in RNA can facilitate the direct binding of m⁶A-modified lncRNA to specific regions of RBP, thereby regulating gene expression [87]. Additionally, m⁶A-binding protein YTHDF3 can interact with lncRNA (Figure 4H) [88]. For example, METTL3-mediated m⁶A modification in support of YTHDF3 contributes to the upregulation of LncRNA MALAT1 expression, thereby enhancing the metastatic potential of non-small cell lung cancer [89].

Unlike conventional linear RNAs, circRNAs possess a covalently closed loop structure that confers resistance to exonuclease-mediated degradation, resulting in enhanced transcript stability [90]. Similar to microRNAs, circRNAs can undergo m⁶A mediated by METTL3 and are recognized by m⁶A reader proteins such as YTHDF1 and YTHDF2, suggesting their involvement in the epitranscriptomic regulation of gene expression [91]. Timoteo et al. revealed that METTL3 regulates m⁶A levels on circRNA, while YTHDC1 affects circRNA splicing. The collaborative regulation of METTL3 and YTHDC1 is involved in various biological processes of circRNA, including translation [92] and splicing [93]. Subsequently, the crosstalk between circRNA and m⁶A modification has been confirmed in numerous diseases. For instance, in patients with colorectal cancer liver metastasis, circNSUN2 is recognized by the m⁶A-binding protein, facilitating its transport from the nucleus to the cytoplasm, increasing the stability of high mobility group A2 (HMGA2), and promoting colon cancer metastasis (Figure 4J) [94]. The m⁶A modification of circular RNAs also plays a role in modulating physiological responses, including immune regulation. Exogenously introduced circRNAs have been shown to function as potent adjuvants, promoting antigen-specific T cell activation, antibody production, and antitumor immunity. However, the recognition of m⁶A-modified circRNAs by the m⁶A reader protein YTHDF2 can suppress these immune responses, indicating a regulatory mechanism by which m⁶A influences circRNA-mediated immunomodulation. Therefore, m⁶A modification on circRNA inhibits innate immunity (Figure 4I) [95]. Numerous studies suggest that the crosstalk between circRNA and m⁶A modification should be considered simultaneously for their roles in disease regulation.

5.2. The Regulatory Role of Crosstalk Between m⁶A and Other Epigenetic Factors in Central Nervous System Development and Diseases

In recent years, extensive interactions between m⁶A RNA modification and other epigenetic mechanisms have been identified, which collaboratively regulate gene expression during central nervous system development and contribute to the pathogenesis of various neurological disorders (Table 1).

5.2.1. Neurogenesis Disorder

During embryonic neural development, the distortion of cell-type diversity and the abnormal assembly of circuits and higher-level structures can lead to the occurrence of many mental disorders [96,97]. NSCs are a type of cell with division potential and self-renewal ability, which can produce various types of nerve tissue cells through asymmetric division. When METTL14 is deficient in embryonic NSCs, the downregulation of m⁶A increases the stability of histone acetyltransferase Crebbp and Ep300 mRNA, promotes H3K27ac, inhibits proliferation-related genes, and activates the expression of differentiation-related genes, leading to the loss of NSCs’ ground state and inducing embryonic neurodevelopmental disorders [98]. As the basic structural and functional unit of the nervous system, the normal development of neurons is closely related to the construction of the central nervous system. It has been shown that mRNA in the neuronal cell body can be transported to dendrites and axons, and new proteins can be synthesized through mRNA local dynamic translation to regulate neuronal development and the correct establishment of neural networks. Related studies have found that lncRNA Dubr interacts with m⁶A-binding proteins YTHDF1 and YTHDF3 through m⁶A modification sites. The knockdown of Dubr or mutation of m⁶A sites in Dubr can accelerate the degradation of YTHDF1 and YTHDF3 proteins, affecting the mRNA translation of Calmodulin and Tau genes related to neuronal development, thereby inhibiting the axon growth of sensory neurons and the correct migration of cortical neurons [12]. In addition, m⁶A-mediated histone regulation also plays an important role in adult neurogenesis and neuronal development. The silencing of METTL3 in adult neural stem cells (ANSC) leads to proliferation inhibition and promotes ANSC differentiation towards the glial cell lineage, affecting the morphological maturation of newborn neurons. The downregulation of m⁶A levels induced by METTL3 deficiency is associated with the inhibition of histone methyltransferase EZH2 protein levels and H3K27me3, ultimately affecting the expression of genes related to proliferation, cell cycle progression, and neuronal development in ANSCs. The overexpression of EZH2 can repair the proliferation and neuronal development defects induced by METTL3 knockout in ANSCs [13]. Therefore, the crosstalk between m⁶A and epigenetic regulatory factors such as histone modification and lncRNA may be one of the mechanisms leading to neurodevelopmental disorders.

5.2.2. Depression

Depression is one of the most common mental disorders worldwide and is characterized by emotional dysfunction [99,100,101]. Impaired synaptic plasticity is an important neuropathological basis for depression, which is usually associated with the dysregulation of relevant functional genes [102,103,104]. Studies on a chronic unpredictable mild stress (CUMS) rat model found that depression-like behavior induced by impaired synaptic plasticity may be due to the hypomethylation of the FTO promoter caused by the low expression of DNA methyltransferases Dnmt1 and Dnmt3a, affecting the expression of plasticity-related genes, and, thus, regulating depressive behavior [58]. It is known that patients with depression exhibit symptoms such as cognitive, emotional, memory, and neurogenic impairments [105], for which improving cognitive ability is a major goal of depression management [106]. Based on this, Niu et al. revealed that METTL3 is highly expressed in CUMS rats, and its silencing alleviates cognitive deficits by promoting the m⁶A-mediated processing and maturation of miR-221-3p through DGCR8, which, in turn, suppresses the expression of GRB2-associated binding protein 1 (Gab1)—a key regulator of synaptic plasticity, inflammation, oxidative stress, and apoptosis in the brain—thereby exacerbating cognitive impairment in CUMS models [107]. Furthermore, a survey of the Chinese Han population showed an association between ALKBH5 and severe depression, increasing the possibility of a high-risk correlation between ALKBH5 and this emotional disorder. Animal studies found an increased expression of ALKBH5 and depression-like behavior in mice subjected to chronic unpredictable stress. Mechanistic studies showed that CUMS caused the abnormal expression of circRNA STAG1, leading to the increased transport of ALKBH5 into the nucleus and the downregulation of the m⁶A modification of fatty acid amide hydrolase (FAAH) in astrocytes, promoting FAAH stability and inducing depression-like behavior [108]. These results suggest that there is extensive crosstalk between m⁶A, DNA methylation, miRNA, circRNA, and other epigenetic mechanisms that regulate the pathogenesis of depression, which may become a new target for depression treatment.

5.2.3. Glioblastoma

Glioblastoma (GBM) is the most common and malignant primary intracranial tumor, characterized by high heterogeneity and poor prognosis [109,110]. However, epigenetic analysis has shown that dysregulated epigenetic mechanisms are involved in the formation of GBM [111], including crosstalk between various epigenetic factors. For example, crosstalk between m⁶A and histone modification can affect GBM resistance to therapy. The treatment of GBM cells with temozolomide (TMZ) has been reported to induce increased m⁶A modification in specific RNA transcripts, which indirectly affects the expression of histone modification enzymes (e.g., those regulating H3K27ac), leading to changes in chromatin accessibility, transcriptional plasticity, and therapy resistance [112]. Conversely, the crosstalk between m⁶A and histone modification is also a molecular mechanism for GBM treatment. H3K9 demethylase JMJD1C promotes the demethylation of the miR-302a promoter region to increase the expression of miR-302a, which negatively regulates METTL3 and inhibits GBM growth [113]. Furthermore, crosstalk between m⁶A and lncRNA promotes the proliferation of GBM. The m⁶A binding protein IGF2BP2 can recognize modification sites on lncRNA CASC9 (cancer susceptibility 9) and enhance its stability, thereby increasing the stability of hexokinase 2 mRNA, which is beneficial for aerobic glycolysis in GBM and promotes GBM proliferation [114]. In addition, studies have found that m⁶A-modified lncRNAs can predict the prognosis of GBM, such as lncRNA, MIR9-3HG, and LINC00515 as protective genes and LINC00900 and MIR155HG as risk genes [115]. Therefore, based on the interaction between m⁶A and lncRNA, m⁶A-modified lncRNA GBM risk prediction models have been established, providing new ideas for the treatment of GBM. These include the risk model of m⁶A-modified lncRNA RP11-552D4.1, which is associated with the immune status, immune suppression biomarkers, and chemosensitivity of GBM patients [116]. The GBM prognosis model constructed by m⁶A-modified lncRNAs AC005229.3, SOX21-AS1, AL133523.1, and AC004847.1 is used to predict and evaluate the survival rate of GBM patients [117].

5.2.4. Other Diseases

In other CNS diseases, there is also crosstalk between m⁶A and other epigenetic regulatory factors. In autism, METTL3 stabilizes the expression of lncRNA MALAT1 by upregulating its m⁶A modification, which facilitates MALAT1 to recruit DNA methyltransferases DNMT1, DNMT3A, and DNMT3B to the promoter region of SFRP2, promoting SFRP2 methylation and reducing its expression, ultimately leading to Wnt/β-catenin signaling pathway inhibition, relieving hippocampal neuron apoptosis and autism-like symptoms [118]. In neurodegenerative diseases, the investigation of why paraquat (PQ) increases the risk of Parkinson’s disease (PD) found that PQ induces the m⁶A modification of lncRNA CDC5L and lncRNA STAT3, affecting their expression, increasing intracellular ROS production, affecting autophagy-related biological functions, and, ultimately, causing damage to and the death of dopaminergic neurons, leading to the onset of PD [119,120]. In stroke research, the currently successful in vitro stroke model is the oxygen and glucose deprivation/reoxygenation (OGD/R) protocol for neurons, which causes oxidative damage to surrounding neurons through ischemia–reperfusion [121,122]. Studies have found that OGD/R may induce neuronal cell apoptosis by increasing the m⁶A of lncRNA D63785 mediated by METTL3, downregulating the expression of LncD63585 and regulating the accumulation of miR-422a [123]. In conclusion, the occurrence of central nervous system diseases is not only regulated by a single epigenetic modification, but may involve the reprogramming of multiple epigenetic factors, so it is necessary to broaden the research field, explore new mechanisms, and discover new targets.

Table 1. Crosstalk mechanisms between m⁶A and other epigenetic regulatory factors in regulating central nervous system development and disease occurrence.

Categories	m6A Related Enzymes	Categories of Epigenetics	Related Components	Mechanisms	Biological Influence	References
Neurodevelopmental disorders	METTL14	Histone modification	Crebbp and Ep300	Upregulated H3K27ac level, inhibited proliferation genes, and activated differentiation genes	Disordered NSC ground state	[98]
	METTL3		EZH2	Inhibited H3K27me3 level; regulated P53 signaling pathway	Promoted ANSC to glial cell line; affected the development of newborn neurons	[13]
	YTHDF1YTHDF3	ncRNA	LncRNA Dubr	Accelerated degradation of YTHDF1 and YTHDF3; affected translation of Calmodulin and Tau	Inhibited axon growth; affected cortical neuron migration	[12]
Depression	FTO	DNA methylation	Dnmt1and Dnmt3a	Downregulated FTO expression	Influenced plasticity-related gene expression	[58]
	METTL3	ncRNA	pri-miR-221	Upregulated miR-221-3p; inhibited Gab1 expression	Induced cognitive impairment	[107]
	ALKBH5		circRNA STAG1	Reduced ALKBH5 level; promoted FAAH degradation	Induced depression-like behavior	[108]
Glioblastoma	METTL3	Histone modification	EZH2	Increased H3K27ac level; degraded nonsense-mediated mRNA	Increased drug resistance	[112]
			JMJD1C	Inhibited SOCS2 expression	Inhibited GBM growth	[113]
	IGF2BP2	ncRNA	lncRNA CASC9	Increased the stability of HK2; promoted aerobic sugar degradation	Promoted GBM proliferation	[114]
	—		lncRNA RP11-552D4.1	—	Reflected immune infiltration disorders; predicted GBM risk	[116]
	—		lncRNA AC005229.3/SOX21- AS1/AL133523.1/ AC004847.1	—	Reflected immune response function; predicted GBM prognosis	[117]
Autism	METTL3	ncRNA	lncRNA MALAT1	Downregulated SFRP2 expression; inhibited Wnt/β-catenin signaling pathway	Reduced autism-like symptoms and hippocampal neuronal apoptosis	[118]
Parkinson’s disease	—	ncRNA	lncRNA CDC5L and lncRNA STAT3	Increased ROS production; enhanced autophagy	Induced dopaminergic neuron damage and death	[119, 120]
Stroke	METTL3	ncRNA	LncRNA D63785	Increased the accumulation of miR-422a	Induced neuronal apoptosis	[123]

Table 1. Crosstalk mechanisms between m⁶A and other epigenetic regulatory factors in regulating central nervous system development and disease occurrence.

Categories	m6A Related Enzymes	Categories of Epigenetics	Related Components	Mechanisms	Biological Influence	References
Neurodevelopmental disorders	METTL14	Histone modification	Crebbp and Ep300	Upregulated H3K27ac level, inhibited proliferation genes, and activated differentiation genes	Disordered NSC ground state	[98]
	METTL3		EZH2	Inhibited H3K27me3 level; regulated P53 signaling pathway	Promoted ANSC to glial cell line; affected the development of newborn neurons	[13]
	YTHDF1YTHDF3	ncRNA	LncRNA Dubr	Accelerated degradation of YTHDF1 and YTHDF3; affected translation of Calmodulin and Tau	Inhibited axon growth; affected cortical neuron migration	[12]
Depression	FTO	DNA methylation	Dnmt1and Dnmt3a	Downregulated FTO expression	Influenced plasticity-related gene expression	[58]
	METTL3	ncRNA	pri-miR-221	Upregulated miR-221-3p; inhibited Gab1 expression	Induced cognitive impairment	[107]
	ALKBH5		circRNA STAG1	Reduced ALKBH5 level; promoted FAAH degradation	Induced depression-like behavior	[108]
Glioblastoma	METTL3	Histone modification	EZH2	Increased H3K27ac level; degraded nonsense-mediated mRNA	Increased drug resistance	[112]
			JMJD1C	Inhibited SOCS2 expression	Inhibited GBM growth	[113]
	IGF2BP2	ncRNA	lncRNA CASC9	Increased the stability of HK2; promoted aerobic sugar degradation	Promoted GBM proliferation	[114]
	—		lncRNA RP11-552D4.1	—	Reflected immune infiltration disorders; predicted GBM risk	[116]
	—		lncRNA AC005229.3/SOX21- AS1/AL133523.1/ AC004847.1	—	Reflected immune response function; predicted GBM prognosis	[117]
Autism	METTL3	ncRNA	lncRNA MALAT1	Downregulated SFRP2 expression; inhibited Wnt/β-catenin signaling pathway	Reduced autism-like symptoms and hippocampal neuronal apoptosis	[118]
Parkinson’s disease	—	ncRNA	lncRNA CDC5L and lncRNA STAT3	Increased ROS production; enhanced autophagy	Induced dopaminergic neuron damage and death	[119, 120]
Stroke	METTL3	ncRNA	LncRNA D63785	Increased the accumulation of miR-422a	Induced neuronal apoptosis	[123]

6. Therapeutic Potential of Targeting Crosstalk Between m⁶A and Other Epigenetic Modulators

In recent years, drug development targeting m⁶A regulators has made initial progress, providing both theoretical and technical foundations for intervening in their crosstalk with other epigenetic mechanisms. Emerging evidence suggests that m⁶A modifications form a coordinated regulatory network with DNA methylation, histone modifications, and non-coding RNAs during various physiological and pathological processes. Therefore, the combinatorial targeting of these pathways may enhance therapeutic efficacy and overcome the limitations of single-target interventions.

Several small-molecule compounds have been developed to target m⁶A regulators. Rhein was the first identified natural inhibitor of FTO, followed by the development of more selective and potent inhibitors such as FB23, FB23-2, CS1, CS2, and Bisantrene, which have shown the ability to inhibit leukemia cell proliferation or reduce neuronal apoptosis. The METTL3 inhibitor STM2457 has demonstrated anti-tumor activity in multiple cancer types and is currently in early-phase clinical trials [124]. The inhibitors of ALKBH5 are still in the early stages of investigation, but preliminary evidence indicates their potential role in diseases such as glioblastoma [125,126].

Although most of these compounds are currently explored within the oncology field, the growing literature suggests that targeting m⁶A regulators may also indirectly influence DNA methylation (e.g., DNMT1 expression), histone deacetylation (e.g., HDACs), or the stability and expression of miRNAs/lncRNAs, thereby facilitating the integrated modulation of multiple epigenetic mechanisms. For instance, the FTO inhibitor FB23-2 can modulate the expression of neuroinflammation-associated miRNAs (e.g., miR-146a) through m⁶A regulation, thereby affecting downstream TLR signaling and histone acetylation [127]. METTL3-mediated m⁶A modification has been shown to promote the stability of EZH2, a histone methyltransferase, influencing H3K27me3 levels and suggesting direct crosstalk between m⁶A and histone modifications with potential therapeutic relevance [128].

Furthermore, the development of multi-target epigenetic inhibitors is emerging as a promising strategy. Some novel compounds can simultaneously interfere with m⁶A regulatory proteins and other epigenetic factors (such as DNMTs and HDACs), enabling combinatorial intervention approaches. These agents have demonstrated neuroprotective effects in models of neurodegenerative diseases [129].

In summary, although no clinical-grade drugs currently exist that specifically target the crosstalk between m⁶A and other epigenetic regulators, several candidate molecules have demonstrated synergistic effects in preclinical studies. However, most of these compounds remain at the in vitro or early in vivo research stages, with limited clinical validation. Existing clinical trials—such as those involving the METTL3 inhibitor STM2457—are primarily focused on oncology, and systematic in vivo evaluation in neurological disease models is still lacking. These limitations highlight the need for further pharmacokinetic optimization, toxicity profiling, and long-term safety assessment to support future translational applications. Collectively, these findings lay the groundwork for the development of multi-faceted network-based epigenetic therapies for neurological disorders.

7. Summary and Outlook

Advancements in sequencing technologies have progressively uncovered the physiological roles of m⁶A modification, including its regulation of RNA splicing, nuclear export, translation, and stability. Concurrently, growing evidence has illuminated the interplay between m⁶A and other epigenetic regulators—such as DNA methylation, histone modifications, and long non-coding RNAs—shedding light on the mechanisms of epigenetic reprogramming. Given the remarkable complexity, heterogeneity, and plasticity of the nervous system, elucidating the crosstalk between m⁶A and other epigenetic pathways may offer novel insights for enhancing therapeutic strategies in central nervous system disorders.

From a translational perspective, therapeutic strategies that target m⁶A regulatory proteins hold promise for the development of novel pharmacological interventions against a range of CNS disorders, including Alzheimer’s disease, major depressive disorder, and post-stroke cognitive impairment. On the one hand, the pivotal role of m⁶A in neurogenesis, synaptic plasticity, and neuroinflammatory modulation makes it an attractive target for the precision treatment of neuropsychiatric diseases. On the other hand, the extensive crosstalk between m⁶A, DNA methylation, and histone modifications provides a theoretical rationale for combined epigenetic targeting. Future investigations should integrate cutting-edge technologies such as single-cell sequencing, spatial transcriptomics, and AI-driven drug discovery platforms to identify cell-type-specific m⁶A nodes and construct highly precise epigenetic intervention models. At the drug development level, high-throughput screening coupled with structure-guided optimization will be essential to improve the brain penetrance and target selectivity of enzyme inhibitors—for example, those directed against FTO or METTL3—thereby enhancing their therapeutic potential in CNS diseases.

In summary, an in-depth understanding of m⁶A regulation and its epigenetic crosstalk not only promises breakthroughs in the mechanistic elucidation of neurological disorders, but also furnishes a theoretical and practical foundation for developing clinically translatable precision epigenetic interventions.