N6-methyladenosine (m6A) methylation in adipose tissue: role, regulation, and therapeutic potential

Abstract

Adipose tissue, is a core regulator of metabolic homeostasis, whose dysfunction plays a crucial role in metabolic diseases such as obesity and diabetes. Additionally, the browning of adipose tissue is important for thermogenesis and metabolic processes. In recent years, the RNA epigenetic modification N6-methyladenosine (m6A) has been shown to dynamically regulate adipocyte differentiation, lipid metabolism-related gene expression, and intercellular communication, playing a functional role in adipose tissue development and metabolic reprogramming. This review systematically summarizes the role and mechanism of m6A modification in lipolysis, adipocyte differentiation, and thermogenic capacity of adipose tissue. Furthermore, we discuss how m6A modification influences adipose tissue inflammation through regulating immune cells. Moreover, we also discuss small molecules targeting m6A modification in adipose tissue and m6A editing systems, which are closely related to the translational application of m6A modification in metabolic diseases. These findings not only refine the detailed molecular map of m6A and its role in the metabolic regulation of adipose tissue but also provide new insights and potential targets for precise interventions in the prevention and treatment of obesity and related metabolic diseases.

Introduction

Adipose tissue, as a multifunctional organ, plays multiple physiological roles including the regulation of energy storage, body thermogenesis, endocrine modulation, and immune responses [1–3]. The formation and differentiation of adipose tissue is a complex process involving the regulation of multiple transcription factors, such as PPAR-γ and the C/EBP family [4, 5]. Based on its regulatory role in energy metabolism homeostasis [6], adipose tissue can be classified into three types: white adipose tissue (WAT), brown adipose tissue (BAT) and beige adipose tissue. WAT is responsible for energy storage in the form of triglycerides; BAT generates heat through uncoupling protein (UCP1) [7]; beige adipose tissue, also known as brown-like adipose tissue, is a group of UCP1-positive adipocytes that also has high thermogenic capacity and glycolytic levels upon induction by stimuli, such as cold exposure and exercise [8–10]. Additionally, the imbalance in adipose tissue metabolic homeostasis can cause metabolic diseases such as obesity, diabetes, hyperlipidemia, and metabolic dysfunction-associated steatotic liver disease (MASLD) through the dysregulated secretion of adipose factors, involvement in immune regulation, and influence on inflammatory responses [11–14]. Currently, obesity and its related metabolic disorders are among the most challenging public health issues, with their incidence rates increasing globally. Therefore, understanding the molecular mechanisms underlying the different functions of adipose tissue is a crucial strategy for treating obesity and other metabolism-related diseases caused by adipose tissue dysfunction.

Chemical modification of RNA is an important post-transcriptional regulatory mechanism, and N6-methyladenosine (m6A) is a common modification in mRNA and non-coding RNA. By influencing the structure, splicing, stability, and translation of mRNA, m6A regulates functions such as gene expression and cell differentiation [15]. In mammals, the abundance and function of m6A in RNA are coordinated by the interactions among its “writers”(methyltransferases), “erasers” (demethylases), and “readers” (m6A-binding proteins). Through intricate interactions, "writers," "erasers," and "readers" ensure that m6A modification in RNA can be dynamically regulated and aligned with cellular demands. Recent studies have found that m6A methylation plays a significant role in adipogenesis, metabolism, immune response, and energy homeostasis of adipose tissue [16, 17].

With the continuous enrichment of research on m6A methylation modifications in adipose tissue, including the methylation installation, removal, and recognition by methyl-binding proteins in adipose tissue, a multi-angle regulation of adipose tissue function has been achieved. This review outlines the "writers," "erasers," and "readers" of m6A modification in adipose tissue, focusing on their roles and mechanisms in adipogenesis, lipolysis, and thermogenesis in adipose tissue, as well as the dynamic balance between these processes. It also discusses how m6A modification in adipose tissue influence chronic inflammation by regulating adipose tissue immune cells, further exploring the clinical application potential of targeting m6A modification, providing new strategies for the treatment of obesity and related metabolic diseases.

The dynamic regulation of m6A modification

N6-methyladenosine (m6A) modification was first discovered in 1974 in the mRNA of Novikoff hepatoma cells [18] and has since been identified across a wide range of organisms, including viruses, bacteria, fungi, plants, and animals [19]. m6A methylation refers to the methylation of the nitrogen at the 6th position of adenosine [20] and is characterized by its dynamic and reversible nature. It is predominantly enriched near stop codons and within the 3' untranslated regions (3'UTR) of mRNAs, particularly around the RRACH motif (R = G or A; H = A, C, or U) [21]. Further studies have revealed that m6A modification also occur in other RNA types, such as ribosomal RNA (rRNA), small nuclear RNA (snRNA), microRNA (miRNA), long non-coding RNA (lncRNA), and circular RNA (circRNA) [22–24]. These modifications play crucial roles in various biological processes, including immune metabolism, stem cell maintenance, and tumor development [25–27]. Dysregulation of m6A modification has been linked to human diseases such as obesity and cancer [28–30]. The function of m6A methylation is mediated by a complex enzymatic system, comprising methyltransferases ("writers"), demethylases ("erasers"), and recognition proteins ("readers"). These components work together to regulate RNA stability, splicing, translation efficiency, and subcellular localization.

Writers transfer methyl groups to RNA by forming methyltransferase complexes (MACs). METTL3, the first identified m6A methyltransferase, forms a stable heterodimer with METTL14, a core component of the methyltransferase complex, to cooperatively catalyze the covalent deposition of m6A modification groups in eukaryotic RNA. METTL3 serves as the catalytic subunit [31], while METTL14 enhances catalytic efficiency by promoting RNA binding [32, 33].The METTL3/METTL14 methyltransferase complex is responsible for methylating approximately 0.15% to 0.6% of adenosines in polyadenylated RNA [34]. The m6A MAC, composed of METTL3 and METTL14, catalyzes the m6A modification by specifically recognizing the DRACH motif (D = A/G/U, R = G/A, A* = methylatable adenosine, C = C, H = A/C/U) in RNA. This complex regulates the deposition of m6A modification across the transcriptome by targeting DRACH sequences, such as the classical motifs GAACU, thereby influencing mRNA stability and translation efficiency [35, 36]. Wilms tumor 1-associated protein (WTAP) binds to METTL3/14 and is essential for optimal substrate recruitment and the proper localization of METTL3/14 [37]. The loss of the WTAP-METTL3/14 complex reduces the RNA-binding ability of the MAC and impairs embryonic differentiation [37]. Recent study on liver transplant ischemia–reperfusion injury (IRI) found that WTAP is significantly upregulated in liver tissue after liver transplantation compared to non-liver transplant patients. Immune infiltration analysis revealed that WTAP is an immune-related m6A regulator closely associated with T and B cells activation level and infiltration degree. This suggests that WTAP may play a key role in the immune status of patients after liver transplantation and could serve as a biomarker for diagnosing the risk of IRI [38]. Additionally, METTL16 has been identified as a methyltransferase for U6 small nuclear RNA, which is crucial for the homeostasis of S-adenosylmethionine (SAM). METTL16 adds m6A modifications to the 3' UTR region of MAT2A mRNA, which encodes a common isoenzyme of mammalian methionine adenosyltransferase, thereby reducing its stability and inhibiting MAT2A expression, forming a negative feedback regulation of SAM synthesis. When intracellular SAM levels decrease in HeLa cervical cancer cell line, METTL16 activity is weakened, the m6A modification of MAT2A mRNA decreases, mRNA stability increases, thereby promoting MAT2A expression and SAM synthesis. If METTL16 or m6A-related regulatory factors (such as YTHDC1) are knocked down, the SAM-responsive regulation of MAT2A mRNA is eliminated, causing cells to be unable to dynamically regulate MAT2A expression according to SAM levels, thereby disrupting SAM homeostasis [39]. Other proteins have also been shown to contribute to the methyltransferase complex, including Vir-like m6A methyltransferase associated protein (VIRMA) [40], putative RNA-binding protein 15 (RBM15) [41], and Zinc Finger CCCH-Type Containing(ZC3H13) [42, 43].

"Eraser" refers to a class of enzymes responsible for removing methylation from RNA, primarily including fat mass and obesity-associated protein (FTO) [44] and ALKB homolog 5 (ALKBH5). Both enzymes contain a conserved double-stranded β-helix (DSBH) domain, which regulates their demethylase activity and indirectly influences other functional domains associated with their demethylation function [45]. FTO removes m6A modification by oxidizing m6A into N6-hydroxymethyladenosine (hm6A) and subsequently into N6-formyladenosine (f6A) [46]. Transcriptomic analysis of mouse preadipocytes combined with m6A-seq revealed that FTO targets thousands of m6A-modified genes involved in lipid metabolism and adipogenesis-related pathways [47]. Ubiquitous overexpression of FTO in mice increases food intake, leading to obesity, while its global knockout or loss of function results in reduced body weight and fat mass. Overexpression of FTO in mice increases food intake, leading to obesity, while its knockout or loss of function results in reduced body weight and fat mass [48]. Recent studies show that FTO regulates gut microbiota and m6A modification to protect the gut barrier and maintain neurological function. In arsenic exposure, FTO expression decreases, leading to gut mechanical barrier damage (reduced ZO-1/Occludin) and dysbiosis (e.g., reduced Desulfovibrio fairfieldensis). Overexpressing FTO restores tight junction protein levels, enhances the barrier, and upregulates mucin (Muc2) and digestive enzymes (e.g., maltase), thereby maintaining chemical barrier balance. FTO also restores Desulfovibrio fairfieldensis abundance, boosting hydrogen sulfide (H₂S) production. H₂S is transferred to the brain, alleviating arsenic-induced neurological issues like neurobehavioral impairments and anxiety-like behaviors [49]. Additionally, FTO inhibition increases m6A modification levels on ferritin heavy chain 1 (FTH1) mRNA, which is further recognized by YTHDF1, enhancing the translation efficiency of FTH1 mRNA. This leads to increased ferritin production, alleviates ferroptosis in spermatogenic cells, and improves sperm quality [50]. Unlike FTO, ALKBH5 directly removes the methyl group from N6-methyladenosine-modified RNA. Compared to FTO, ALKBH5 exhibits higher catalytic activity toward m6A through a Lys132/Tyr139-mediated proton shuttle network that accelerates hemiaminal intermediate decomposition, and preferentially binds to substrates with a consensus motif (A/G)m6AC via its unique 5'−3' single-stranded RNA orientation and specific interactions with key residues in the active site [51]. ALKBH5 has been shown to play a crucial role in mRNA export, fertility, and the prevention of abnormal RNA splicing. Knockout of ALKBH5 in male mice leads to elevated levels of m6A-modified mRNA, causing meiotic arrest and spermatocyte apoptosis, which impairs fertility [52]. Recent studies on ALKBH5 have found that liver-specific knockout of the ALKBH5 gene suppresses the glucagon receptor (GCGR) and mammalian target of rapamycin complex 1 (mTORC1) signaling pathways, reducing glucose and lipid levels. This beneficial effect may help improve type 2 diabetes and MASLD [53]. Additionally, reviews have discussed its role in gastrointestinal cancers and fibrosis-related diseases [54, 55], further highlighting the involvement of ALKBH5 dysregulation in various biological processes and diseases.

The reader proteins for m6A influence gene expression and various cellular events. The main m6A recognition proteins in mammals include three types: YT521-1B homolog (YTH) domain proteins, heterogeneous nuclear ribonucleoproteins (hnRNPs), and insulin-like growth factor 2 mRNA binding proteins (IGF2BPs) [56]. The YTH domain is highly conserved and dominant in eukaryotes, and it binds to the m6A sites on mRNA. Commonly known YTH family recognition proteins include YTHDF1-3, YTHDC1, and YTHDC2, all of which bind to single-stranded mRNA through the YTH domain to determine various downstream fates of RNA [57]. In terms of cellular localization, YTHDF1-3 proteins are located in the cytoplasm, YTHDC1 is located in the nucleus, and YTHDC2 is located in both the cytoplasm and the nucleus. Functionally, YTHDF1 promotes mRNA translation, while YTHDF2 induces mRNA instability by stimulating mRNA decay [58]. YTHDF3 can coordinate with YTHDF1 or YTHDF2 to control the metabolism of m6A-modified mRNA [59]. YTHDC1 can influence pre-mRNA splicing, promote XIST-mediated transcriptional silencing, accelerate the degradation of its target mRNA, and enhance mRNA export from the nucleus in an m6A-dependent manner [59]. YTHDC2 promotes mRNA translation and regulates mRNA decay [60]. Additionally, heterogeneous nuclear ribonucleoproteins, including hnRNPA2B1, hnRNPC and hnRNPG, can indirectly bind to m6A-modified mRNA, and they are primarily highly expressed in the nucleus to regulate the splicing and expression of m6A-modified mRNA [59]. Other reader proteins include insulin-like growth factor 2 binding proteins (IGF2BP) 1/2/3, which can stabilize mRNA and enhance translation [61].

The role of m6A in adipose function

M6A regulates lipolysis in adipose tissue

Adipose tissue, as a central regulator of energy homeostasis, mobilizes stored triglycerides through lipolysis —decomposing them into free fatty acids and glycerol—to provide energy and participate in metabolic signaling. This process is tightly regulated under both physiological (e.g., fasting, exercise) and pathological (e.g., obesity, insulin resistance) conditions [62, 63].Recent advancements in detecting methylated RNA modifications, coupled with deeper insights into adipose tissue function, have increasingly linked mRNA methylation modifications to adipose tissue activity. Among these, lipolysis in adipose tissue has emerged as a major research focus.

The central role of writer proteins such as METTL3/METTL14, WTAP in the regulation of adipose tissue metabolism has been gradually revealed. Lipolysis of adipose tissue is mainly regulated by sympathetic nerves and insulin. Sympathetic nerves system (SNS) stimulates lipolysis through the neurotransmitter norepinephrine. The specific pathway is the binding of norepinephrine to β-adrenergic receptor 1–3(Adrb1-3) to activate the cAMP/protein kinase A (PKA) pathway in adipocytes, which further phosphorylates ATGL and hormone-sensitive lipase (HSL) to increase lipolysis [64, 65]. The methyltransferase complex formed by the heterodimerization of METTL3 with METTL14 regulates lipolysis indirectly by installing m6A in the transcripts of Adrb2, Adrb3, Atgl, and comparison gene recognition 58 (Cgi-58) to regulate lipolysis gene stability. It was found that the m6A content of Adrb2, Adrb3, Atgl and Cgi-58 transcripts was significantly increased in adipose tissue under high-fat diet (HFD)-induced obesity, and METTL3/14 expression was up-regulated, which increased the m6A level of these gene transcripts. The lipolytic signal is attenuated due to accelerated mRNA degradation of lipolytic genes through reader proteins such as YTHDF2 [66]. Functional knockdown of METTL14 in adipocytes decreases m6A modification levels in the transcripts of these lipolytic genes, thereby enhancing β-adrenergic receptor (Adrb) signaling and lipolysis. This conclusion was validated in further animal experiments in which adipocyte-specific deletion of METTL14 in mice was resistant to high-fat diet-induced obesity, insulin resistance, and MASLD [66]. Whether m6A modification regulates the signaling pathway of lipolysis by affecting HSL expression levels has not been reported. But in a correlation study, it was found that intermittent fasting increased HSL expression levels in the adipose tissue around the heart of obese mice, while m6A levels and METTL3 expression were down-regulated and FTO expression was up-regulated [67]. Further studies are needed to confirm the role of m6A modification in regulating HSL expression.

m6A modification not only directly regulates lipolysis in adipose tissue by regulating the expression of classical lipolysis genes, but also participates in the regulation of lipolysis through other pathways. The adipokine Angptl4, which has been reported to inhibit LPL and stimulate lipolysis in adipocytes [68], is regulated by demethylation modification of FTO, thereby leading to its elevated protein translation efficiency and thus enhanced lipolysis. In FTO knockout obese mice fed with HFD, the reduction in Angptl4 protein levels leads to inhibited lipolysis, thereby promoting obesity, which can be reversed by adenovirus-mediated expression of Angptl4 in the adipose tissue of HFD-FTO-KO mice [69]. Leptin, another adipokine, up-regulates FTO levels in adipose tissue and further removes the m6A methyl group of Perilipin5 (Plin5) gene, which encodes a scaffold protein for lipid droplets, thereby increasing its protein expression and promoting lipolysis [70]. Plin5, as a scaffold protein, plays an important role in controlling the process of triacylglycerol hydrolysis and lipolysis in adipose tissue [71]. Studies have found that Plin5 is involved in communication between lipid droplets and mitochondria, and it is hypothesized that this can promote the direct transfer of FFAs released during lipolysis. In fact, overexpression of Plin5 leads to increased mitochondrial biogenesis, enhanced transcription of electron transport chain complexes, and fatty acid oxidation genes, such as PPAR and PGC1α [72].Finally, the regulatory role of m6A modification in adipose lipolysis exhibits context-dependent specificity under pathological conditions. Intermittent hypoxia (IH) is the core pathological feature of obstructive sleep apnea syndrome (OSAS), which is accompanied by metabolic complications such as insulin resistance. It was found that METTL3 levels in adipose tissue decreased during IH, which reduced the level of m6A in monoacylglycerol lipase (MGLL) mRNA and further reduced YTHDF2-mediated recognition and degradation of Mgll mRNA to promote lipolysis. Interestingly, Atgl and Hsl were not affected by this modification [73]. This selective regulation suggests that m6A modification may have highly specific regulatory network under specific conditions.

In conclusion, m6A modification regulates lipolysis by modulating the mRNA stability or translation efficiency of key lipolytic genes(As shown in Fig. 1). In the obese state, the expression levels of key lipolytic genes such as Atgl and Hsl are reduced. Inhibition of m6A methylation writer enzymes' expression and activity may maintain the stability of Atgl and Hsl mRNA, increase their transcript abundance, and thereby enhance protein synthesis to promote lipolysis. In addition, non-classical lipolysis regulators such as Angptl4, Plin5 and Mgll may further strengthen lipolysis under the regulation of m6A modification, providing potential strategies for improving obesity. Therefore, exploring the differential regulation of m6A modification on lipolysis genes is crucial for improving adipose tissue lipolysis to combat obesity and related disorders.

Fig. 1.

m6A RNA methylation dynamically regulates adipose tissue lipolysis by controlling the stability of key lipolytic gene transcripts. Adipose tissue lipolysis is primarily regulated by the sympathetic nervous system and insulin. Sympathetic stimulation activates β-adrenergic receptors via norepinephrine (NE), triggering the cAMP/PKA pathway, which leads to the phosphorylation of Atgl and Hsl, thereby promoting lipolysis. Under obese conditions, elevated expression of the METTL3/METTL14 methyltransferase complex increases m6A deposition on mRNAs encoding critical lipolytic factors, including Atgl and adrenergic receptors (Adrb2/3). These m6A-modified transcripts are recognized and degraded by the reader protein YTHDF2, suppressing their expression and consequently inhibiting lipolytic signaling. Conversely, in pathological states such as intermittent hypoxia (IH), reduced METTL3 levels lead to decreased m6A methylation on the lipase gene Mgll. This diminishes YTHDF2-mediated mRNA decay, enhances MgllL expression, and promotes lipolysis. Additionally, the m6A demethylase FTO promotes lipolysis by erasing m6A marks from transcripts such as Angptl4 and Plin5, thereby increasing their mRNA stability and protein expression. Leptin has also been shown to upregulate FTO expression in adipose tissue. Figure created with BioRender.com

M6A modification is involved in adipogenesis

Adipogenesis refers to the process by which preadipocytes differentiate into adipocytes. This process includes the sequential stages from multifunctional fibroblasts (stem cells) to preadipocyte formation, initiation of adipocyte differentiation, terminal differentiation and lipid accumulation, and functional maintenance of mature adipocytes. At present, relevant reviews have sorted out the key regulatory genes in the process of adipogenesis, including peroxisome proliferator-activated receptor γ (PPARγ) and CCAAT/enhancer binding protein α (C/EBPα), as well as other important regulators, such as sterol regulatory element binding protein (SREBP), signal transducer and activator of transcription 5 (STAT5), Wingless-related integration site family proteins, chromatin remodeling complex (SWI/SNF), Krüppel family proteins, etc. [74]. Existing evidence suggests that m6A modification affects adipogenesis by regulating expression levels of some of the above adipogenic regulators.

PPARγ is an important transcriptional regulator of adipogenesis. Many in vitro and in vivo studies have shown that PPARγ is essential for adipocyte maturation and function [75, 76]. Ectopic expression of PPARγ in mouse fibroblasts initiates the entire adipogenic program to form functionally mature adipocytes [77]. Another important role of PPARγ in adipogenesis is the induction of the transcriptional activity of the adipogenic regulator CCAAT/enhancer binding protein-α (C/EBPα). Notably, PPARγ can drive adipogenesis independently of C/EBPα, as it retains the ability to induce adipogenic differentiation even in the absence of C/EBPα, thereby establishing itself as a critical master regulator of adipogenesis [78]. Research indicates that in the 3T3-L1 preadipocyte model, FTO suppresses miR-130 expression through m6A demethylation, while miR-130 serves as a direct negative regulator of PPARγ by targeting both the coding region (CR, nucleotide positions 1243–1350) and 3' untranslated region (3'UTR, nucleotide positions 1603–1730) of PPARγ mRNA. Consequently, FTO-mediated suppression of miR-130 indirectly upregulates PPARγ expression, thereby enhancing adipocyte differentiation and adipogenesis [79, 80].Besides, FTO knockdown resulted in the downregulation of several PPARγ target genes, such as ACSL1, ACOX1, and CD36, and impaired adipogenesis. However, supplementation with a PPARγ agonist (such as rosiglitazone) weakened the differentiation inhibition mediated by FTO knockdown, suggesting that the FTO/PPARγ axis plays an important role in regulating adipogenesis [81]. Another methyl-removing enzyme, ALKBH5, enhances the stability of the E3 ubiquitin ligase TNF receptor-associated factor 4 (TRAF4) by removing m6A levels of its mRNA, which further increases proteasomal ubiquitination of PPARγ and inhibits adipocyte differentiation [82]. Curcumin has been found to inhibit the expression of ALKBH5 to suppress the TRAF4-mediated ubiquitination and degradation pathway and promote adipocyte differentiation [82]. Meanwhile, METTL3 inhibits the transcription of PPARγ by increasing translation of the estrogen receptor ESR1, thereby inhibiting the differentiation of adipose progenitor cells (FAPs) into adipocytes, a mechanism that has been shown to be present in menopausal women [83].Next, METTL14 has been shown to enhance the stability of LINC00278 (a Y-linked lncRNA exclusively expressed in males) in an m6A-dependent manner. LINC00278 inhibits BRG1 degradation by binding to BRG1 (the core subunit of SWI/SNF complex), which promotes PPAR-γ2 transcription through chromatin remodeling, further inducing the differentiation of pre-adipocytes into mature adipocytes and leading to fat accumulation [84]. Besides, LINC00278 was specifically highly expressed in male adipose tissue, suggesting that this pathway may act synergistically through androgen receptor (AR) or other sex-related factors. These two studies suggest that the sex difference of PPARγ function in adipocyte differentiation can be regulated by m6A modification, and knocking down or inactivating the above m6A methyl-writing enzymes like METTL3/14 may be a potential therapeutic strategy for ameliorating sex-induced obesity and related metabolic diseases. In addition to PPARγ, studies on the role of m6A in regulating C/EBP proteins expression to influence adipocyte differentiation have not been reported. However, a study has found that m6A modification indirectly affects the level of C/EBPα by regulating the expression of family sequence similarity 134 member B (FAM134B) gene, thereby participating in the regulation of adipogenesis [85]. The specific mechanism is that there is a key m6A modification site in the CDS region of FAM134B mRNA. By recognizing this site, the m6A binding protein YTHDF2 leads to decreased stability (shortened lifespan) of FAM134B mRNA, thereby reducing its protein expression. After mutating this site (FAM134B-MUT), YTHDF2 was unable to mediate its degradation. Furthermore, FAM134B-MUT significantly enhanced adipogenesis by up-regulating the level of C/EBPα and PPARγ.

In addition, m6A modification can promote adipogenesis by indirectly affecting the phosphorylation level of SATA5. STAT5 is induced during adipogenesis and activated in adipocytes by growth hormone and prolactin [86], and activated STAT5 forms a dimer and translocalizes to the nucleus to regulate the expression of target genes involved in adipogenesis. Further studies indicate that deletion of the methyltransferase METTL3, by reducing the m6A level of JAK1 mRNA, enhances its stability mediated by the methylation-recognition protein YTHDF2, activates STAT5 phosphorylation, promotes its binding to the C/EBPβ promoter, and ultimately accelerates adipogenesis [87].Previous evidence showed that METTL3 inhibits the transcription of PPARγ by increasing the translation of estrogen receptor ESR1, thereby suppressing adipocyte differentiation. In these two studies, METTL3 seems to play the role of an "integrative inhibitory regulator" in adipocyte differentiation. Through m6A modification, it targets key genes like ESR1 and JAK1, suppressing core adipogenic transcription and early signaling to impede fat formation. This networked and dynamic regulation serves as a precise brake on differentiation, ensuring adipose tissue homeostasis.

Alternatively, WNT signaling has been shown to be potentially regulated by m6A modification. In a review comparing the effects of formula feeding (FF) versus breastfeeding (BF) on postnatal epigenetic programming (including DNA and RNA methylation) and subsequent obesity risk in infants, it was indicated that formula feeding leads to increased expression of the FTO gene. This upregulation may further activate the mTORC1-S6K1 signaling pathway and suppressing the WNT/β-catenin pathway, thereby relieving inhibition of adipogenesis and ultimately elevating the risk of obesity [88]. But, how FTO affects the activity of mTORC1/S6K1 signaling pathway through m6A modification, and whether this ultimately leads to the activation of WNT/β-catenin to inhibit adipogenesis, still requires further research for confirmation.

In addition to regulating key adipogenic factors, m6A modification also affects adipocyte generation by regulating other processes, such as autophagy and apoptosis. It has been found that inhibition of adipocyte autophagy and intracellular lysosomal function leads to a marked reduction in lipid accumulation in adipocytes, along with decreased protein levels of key differentiation-related factors such as C/EBPβ, ultimately inhibiting adipogenesis [89]. ATG5/7 has been shown to play an important role in facilitating autophagy in adipocytes [90, 91]. FTO promotes autophagosome formation by reducing the m6A levels of Atg5 and Atg7 mRNAs to maintain their stability [92]. FTO silencing leads to YTHDF2-mediated degradation of Atg5/7 mRNAs, inhibits autophagy, and impairs adipogenesis to reduce lipid storage in adipocytes. In the regulation of adipocyte apoptosis, FTO further inhibits the activation of the pro-apoptotic factor Caspase-3 and mitochondrial-dependent apoptosis in adipocytes by activating the JAK2/SATA3 signaling pathway and inhibiting mitochondrial unfolding protein response (UPRmt), ultimately reducing adipocyte apoptosis [93]. Some studies have found that some plant components can regulate the apoptosis process through m6A modification. Hinokiflavone (HF) specifically binds to IGF2BP2 to enhance its stability and functionality, which subsequently promotes the stability of m6A-modified BCL2 Interacting Mediator of cell death (Bim) mRNA and leads to the activation of mitochondrial apoptosis pathway. The upregulated Bim is involved in caspase9/3-mediated mitochondrial apoptosis in adipocytes, thereby ameliorates HFD-induced obesity [94]. However, inducing adipocyte apoptosis to treat obesity needs to be further explored. Moderate apoptosis in HFD-induced obese mice helps to control the number of adipocytes and maintain the stability of adipose tissue, but there are also certain risks in inducing adipocyte apoptosis. For example, excessive apoptosis may damage the endocrine function of adipose tissue, such as inhibiting the expression levels of leptin, adiponectin and resistin, which would impair the systemic homeostasis of metabolism. Therefore, adipogenesis is affected by autophagy and apoptosis, and m6A modification forms a dynamic balance network by targeting key genes involved in autophagy and apoptosis.

Preadipocyte proliferation plays an important role in adipogenesis. Moreover, cyclin and cyclin-dependent kinase (CDK) also affect the proliferation and differentiation of adipocyte precursors by participating in the process of cell cycle, which provides the necessary cellular basis for adipogenesis, and m6A modification is also involved in the above process. It has been found that the methyltransferase WTAP forms a methyltransferase complex with METTL3/14, which drives mitotic clonal expansion (MCE) progression and facilitates adipogenesis by elevating Cyclin A2 mRNA levels [95]. Knockdown of either WTAP or any component protein of the METTL3/METTL14 complex leads to cell cycle arrest in adipocyte precursors and inhibits adipogenesis. This molecular mechanism is functionally validated in the obese phenotype—WTAP knockdown significantly reduces the number and volume of adipocytes while effectively alleviating high-fat diet (HFD)-induced obesity in mice. Meanwhile, the expression level of Ccna2 can also be directly regulated by FTO. Studies have found that exogenous supplementation of branched-chain amino acids (BCAA) can inhibit the obesity phenotype of mice, and the underlying mechanism is that BCAA decreases the concentration of reduced nicotinamide adenine dinucleotide phosphate (NADPH) in adipose tissue and 3T3-L1 cells by reducing glucose-6-phosphate dehydrogenase (G6PD) expression, which is the rate-limiting enzyme in the pentose phosphate pathway.The reduction in NADPH further attenuates the demethylase activity of the fat mass and obesity-associated (FTO) protein, as NADPH directly binds to FTO and is essential for enhancing its activity in vivo. This leads to increased levels of m6A modification of Ccna2 and cyclin-dependent kinase 2 (CDK2) mRNAs, which are further recognized by YTH N6-methyladenosine RNA binding protein 2 (YTHDF2) for degradation, thereby inhibiting mitotic clonal expansion and adipogenesis [96]. In addition, the protein expression level of cyclin D1 (CCND1) was also regulated by m6A modification. Depletion of the Kruppel type zinc finger protein family protein ZFP217 in adipose tissue leads to increased expression of METTL3. This further enhances the m6A level of CCND1 mRNA, which is recognized and degraded by YTHDF2, leading to inhibition of adipogenesis [97]. These data suggest that m6A modification plays an important role in mitotic clonal expansion (MCE), a key step in adipogenesis, by regulating the expression or stability of cyclins (e.g., Cyclin A2, Cyclin D1) and related factors (e.g., CDK2).

Additionally, other evidence indirectly demonstrates the regulatory role of m6A modification in preadipocyte proliferation. For instance, studies have indicated that the RNA-binding proteins SRSF3 and hnRNP A1 synergize with METTL3 to form a complex that co-promotes the m6A modification of circCDK14, thereby driving its nuclear export to the cytoplasm. In the cytoplasm, circCDK14 acts as a sponge to sequester miR-4492-z, suppressing its functional activity and consequently relieving its repression of proliferation-related genes (cyclin B, cyclin D1, cyclin E1, and KI67) expression. This ultimately exerts a positive regulatory effect on the proliferation of yak intramuscular preadipocytes (YIMAs) [98].Moreover, it has been reported that the m6A demethylase FTO specifically binds to the mRNA of CTNNB1 (Catenin Beta 1, β-catenin) and mediates its m6A demethylation at site 420 in the coding sequence (CDS) and site 2816 in the 3’-UTR. This modification enhances the mRNA stability of CTNNB1 and upregulates its expression. As a positive regulator of chicken preadipocyte proliferation, the upregulated CTNNB1 subsequently promotes the expression of proliferation-related genes (CCND1, CCND2, CCNB2, and PCNA), ultimately facilitating chicken preadipocyte proliferation [99]. In addition to FTO, another m6A demethylase, ALKBH5, has been shown to bind to and mediate the m6A demethylation of lecithin-cholesterol acyltransferase (LCAT)mRNA, thereby reducing its stability and accelerating its degradation. Further investigations revealed that LCAT overexpression significantly impaired the proliferation capacity of chicken preadipocytes and induced G1-phase cell cycle arrest (with a blocked G1/S transition), whereas LCAT knockdown promoted preadipocytes proliferation [100].Therefore, ALKBH5 exerts a pro-proliferative effect on preadipocytes by suppressing LCAT mRNA expression.

In addition to the YTHDF family, a recent study reported that the m6A recognition protein IGF2BP3 participates in the regulation of obesity and insulin resistance by modulating the adipogenic differentiation process of mesenchymal stem cells (MSCs). IGF2BP3 is an important m6A reader that enhances mRNA stability [61]. During the natural adipogenic differentiation of MSCs, the expression level of IGF2BP3 gradually decreases. Further mechanistic exploration revealed that IGF2BP3 interacts with myosin light chain kinase (MYLK) mRNA in an m6A-dependent manner, extending its half-life and thereby increasing MYLK protein expression; while MYLK subsequently inhibits the phosphorylation of the ERK1/2 pathway, ultimately suppressing MSC adipogenic differentiation. Animal experiments demonstrated that specifically overexpressing IGF2BP3 in adipose tissue through adeno-associated virus serotype Rec2 (AAVRec2) significantly reduced body weight and body fat ratio in high-fat diet-induced obese mice, and improved insulin resistance and glucose tolerance [101].

Besides the above factors, m6A modification also regulates some other factors to affect adipogenesis. m6A modification was found to promote the mRNA stability of mitochondrial homology protein 2 (MTCH2) through the recognition of MTCH2 mRNA by YTHDF1, and MTCH2 acts as a key regulator to enhance the differentiation capacity of intramuscular preadipocytes, thereby driving adipogenesis. Reduction of m6A levels on MTCH2 mRNA results in impaired adipogenesis [102]. Heterogeneous ribonucleoprotein C (HNRNPC), as a novel m6A reader, also plays a role in adipogenesis. HNRNPC is widely distributed in a variety of tissues and gradually decreases with aging [103]. A PDGFRA -Cre-driven conditional knockout mouse model (Hnrnpcfl/fl; Pdgfra-Cre, abbreviated as APKO for adipocyte progenitor-specific knockout) on the C57BL/6 J background revealed that hnRNPC deficiency in adipocyte progenitors suppresses adipogenesis. Mechanistically, HNRNPC binds to the m6A modification of lymphocyte cytoplasmic protein (LCP1) mRNA to enhance its stability, and LCP1 re-expression partially reverses the inhibition of adipogenesis caused by HNRNPC depletion through modulation of cytoskeleton remodeling [103]. A key adipogenic differentiation factor, lecithin-cholesterol acyltransferase (LCAT), has recently been identified to be regulated by the m6A demethylase ALKBH5 in chicken adipose tissue. Overexpression of ALKBH5 in preadipocytes reduces global m6A RNA methylation levels and increases the demethylation of LCAT mRNA, leading to decreased LCAT mRNA stability. LCAT overexpression may accelerate preadipocyte differentiation by promoting activation of the PPARγ signaling pathway [100].

Comprehensive analysis of current research indicates that m6A modification regulates adipogenesis through multiple pathways, including the modulation of key transcription factors (e.g., PPARγ, C/EBPα), cell cycle regulators (e.g., Cyclin A2, CCND1, CDKs), autophagy (e.g., ATG5/7), and apoptosis (e.g., bim) and son on. (As shown in Fig. 2) Specifically, m6A modification directly or indirectly targets these critical mRNAs, influencing their stability and degradation efficiency to regulate biological processes in adipocytes such as mesenchymal stem cell (MSC) differentiation, mitotic clonal expansion (MCE), autophagy, and apoptosis. These interconnected processes form a dynamic equilibrium network that plays a pivotal role during adipogenic differentiation. Furthermore, studies have demonstrated that m6A modification plays a crucial role in adipogenesis under varying sexes, physiological states, and pathological conditions.

Fig. 2.

m6A modification regulates adipogenesis by influencing key transcription factors, autophagy, apoptosis, and the differentiation of mesenchymal stem cells (MSCs) and mitotic clonal expansion (MCE). This schematic summarizes the multi-layered control of adipocyte differentiation by m6A modification. m6A exerts bidirectional regulation over the master adipogenic transcription factors PPARγ and C/EBPα through the dynamic actions of methyltransferases (METTL3/METTL14), demethylases (FTO/ALKBH5), and reader proteins (YTHDF2/IGF2BP2). It promotes pro-adipogenic signaling by enhancing JAK-STAT activity and suppressing the inhibitory WNT/β-catenin pathway. Additionally, m6A modification regulates cellular homeostasis to promote differentiation primarily by suppressing autophagy and mitochondrial apoptosis, though certain reader proteins may induce apoptosis. Notably, the plant-derived Hinokiflavone (HF) binds specifically to IGF2BP2 and mediates m6A-dependent adipocyte apoptosis. During early commitment and mitotic clonal expansion, m6A facilitates the process by stabilizing mRNAs encoding cell cycle regulators and key differentiation factors. Figure created with BioRender.com

M6A is involved in the development of brown adipose tissue and the browning of white adipose tissue

Brown adipose tissue (BAT) refers to a kind of adipose tissue containing a large number of mitochondria. It mainly consumes energy to produce heat through uncoupling protein 1 (UCP1) and plays a role in regulating body temperature. m6A methylation regulates the development of BAT, the browning of white adipose tissue (WAT) and related metabolic homeostasis through the dynamic balance of "writer-eraser-reader".

BAT dissipates the mitochondrial electrochemical gradient through uncoupling protein 1 (UCP1) to generate heat [104–106]. Studies have found that m6A methyltransferases, such as METTL3 and WTAP, play important roles in the development and maturation of brown adipose tissue. METTL3 was significantly increased in the scapula after birth. Further studies showed that METTL3 was involved in regulating the expression of brown adipose related genes (including Prdm16, PPARγ, Pgc-1α, and Ucp1) through m6A modification to maintain the development and maturation of brown adipose tissue. Brown adipose tissue-specific METTL3 knockout in mice impairs BAT development and thermogenesis and promotes HFD-induced obesity and whole-body insulin resistance [107]. WTAP also affects the development and function of brown adipose tissue. Prostaglandin signaling indirectly promotes the development of brown adipose tissue by regulating the protein stability of WTAP. Prostaglandin receptor 3 (EP3) has been found to inhibit adenylyl cyclase by binding to Gai (a component of G protein-coupled receptor (GPCR) signaling), lowering cAMP levels, which further inhibits downstream PKA signaling. The inhibition of PKA signaling then leads to an increase in the phosphorylation level of ERK1/2, which blocks ubiquitin–proteasome-mediated degradation of WTAP. This helps maintain the stability of the WTAP protein. Next, m6A modification of Zfp410 mRNA was further enhanced by WTAP protein, which ultimately promoted the differentiation of brown adipocytes. However, blockade of Gai or ERK1/2, or restoration of PKA activity, abolished brown adipogenesis enhanced by EP3 agonists through inhibition of ERK1/2/WTAP/ZFP410 signaling [108]. A similar event was also reported in another study where BAT-specific knockdown of WTAP (WTAP-bKO) severely impaired BAT maturation in vivo by reducing BAT thermogenic related genes expression, resulting in interscapular BAT (iBAT) bleaching [109]. The specific mechanism is as follows: WTAP stabilizes the METTL3 protein by forming a stable complex with METTL3, maintaining the m6A modification level on the mRNA of the aforementioned genes, and promoting the mRNA stability of genes containing the PR domain 16 (Prdm16), PPARγ, and those related to mitochondrial function, thereby promoting the maturation and thermogenic function of brown adipose tissue. In addition, the stable expression of PPARγ is essential during brown adipose development, and m6A modification was found to maintain PPARγ levels by preventing its ubiquitination and proteasomal degradation [110]. The specific mechanism is that the methyl-recognition protein YTHDC1 directly interacts with PPARγ through its intrinsic disordered region (IDR), resulting in the A/B domain of PPARγ being covered from binding to E3 ubiquitin ligase ARIH2, thereby preventing its ubiquitin-mediated proteasomal degradation. However, loss of Ythdc1 in BAT increased PPARγ degradation and impaired interscapular BAT development, thermogenesis, and total energy expenditure. It is important to note that YTHDC1, a key member of the m6A reader family, has traditionally been recognized for its role in recognizing m6A modifications through its YTH domain, thereby participating in the regulation of various RNA processing events such as RNA splicing, nuclear RNA export, and RNA decay. However, this study reveals a non‑canonical functional mechanism of YTHDC1 in brown adipose tissue development. YTHDC1 interacts directly with the PPARγ protein via its intrinsically disordered regions (IDRs), forming nuclear puncta structures that physically block the binding of the E3 ubiquitin ligase ARIH2 to PPARγ. This protective mechanism effectively prevents the ubiquitination and subsequent proteasomal degradation of PPARγ. Unlike other m6A readers such as YTHDF1/2, YTHDC1 in this process acts directly at the protein level to maintain PPARγ stability, without affecting its mRNA level. This finding suggests that other key proteins involved in brown adipose tissue development (such as PRDM16, UCP1, etc.) might also be regulated by similar non‑canonical mechanisms. In addition to the aforementioned evidence, another study highlighted that the variant rs1421085 T > C in the FTO gene locus can enhance thermogenesis in brown adipose tissue (BAT), improve obesity, and this effect is temperature-dependent [111]. The study found that mice carrying the rs1421085 C allele in brown adipose tissue (knockin model) exhibited enhanced thermogenic capacity of BAT, characterized by smaller and denser brown adipocytes, and resistance to high-fat diet (HFD)-induced fat accumulation. Additionally, FTO, as an RNA demethylase, promotes thermogenesis in brown fat by regulating the mRNA stability of thermogenesis-related genes such as Ucp1. Moreover, the metabolic advantage of C allele mice (e.g., resistance to obesity) was observed at temperatures below thermoneutrality (e.g., 22 °C); however, this protective effect disappeared at thermoneutral temperatures (30 °C). The specific mechanism of this FTO gene locus mutation is that rs1421085 is located in the first intron of the FTO gene, which is an enhancer region. The C allele enhances the interaction between this enhancer and the FTO promoter, thereby increasing FTO transcription. Ultimately, this leads to increased stabilization of thermogenic genes, thus enhancing the thermogenic capacity of brown adipose tissue. Therefore, the m6A Methylation modification play a role in the development and maturation of brown adipose tissue by regulating the expression levels of thermogenic genes.

In addition, m6A modification is also involved in the regulation of white adipose tissue browning. Induction of beige adipocytes in white adipose tissue, also known as WAT Browning, improves glucose and lipid metabolism. METTL3 prevents the degradation of kruppel-like factor 9 (Klf9) by increasing the m6A levels on Klf9 mRNA. It was found that METTL3 deletion significantly reduced the mRNA level of Klf9 in BAT, and this effect also existed in white adipose tissue (WAT), suggesting that Klf9 level is regulated by m6A modification. The expression of Klf9 and UCP1 increased synchronously during WAT browning induced by cold exposure, indicating that KLf9 is involved in WAT browning [112]. Specific knockdown of METTL3 in BAT results in enlargement of lipid droplets and suppression of UCP1 expression, phenotypes that are markedly reversed by overexpression of Klf9.These findings illuminate how the METTL3-Klf9 module likely operates during WAT browning to ensure effective induction of thermogenic genes like UCP1.Thus, METTL3 plays a key role in promoting BAT function and adipose tissue browning by stabilizing Klf9 mRNA, especially under cold exposure. White adipose tissue-specific depletion of METTL3 disrupted WAT browning in HFD-fed mice. This result was supported by the study that METTL3 enhanced the mRNA m6A modification level of the genes encoding hexokinase type 2 (Hk2), phosphofructokinase-liver (Pfk-l) and pyruvate kinase muscle isoform (Pk-m), which further led to their increased recognition by IGFBP2 to increase the stability of their mRNAs and promote glycolysis. In contrast, ablation of METTL3 in mature beige adipose tissue results in inhibition of glycolytic capacity and thermogenesis, along with a reduction in preadipocyte proliferation [113]. In addition to methyltransferases, methyl-removal enzymes have also been shown to play important roles in the browning process. Knockdown of FTO in adipose tissue has been shown to promote thermogenesis and beige adipocyte transformation. Deletion of FTO increases the m6A modification level of hypoxia-inducible factor 1α (Hif1a) mRNA, which is further recognized by YTHDC2 to promote the translation of Hif1a mRNA, leading to the increase of HIF1A protein level. HIF1A promotes Ucp1 expression and the WAT browning process by activating transcription of thermogenic genes such as Ppaggc1a, Prdm16, and Pparg [114]. In the previous section on the functional regulation of m6A in brown adipose tissue, we mentioned that the mutation at the FTO gene locus (rs1421085 T > C) enhances FTO expression to promote thermogenic function in brown adipose tissue (BAT), with the primary mechanism being that FTO, as an m6A demethylase, stabilizes Ucp1 mRNA, and this effect is only evident under low-temperature conditions (below 22 °C). Here, we emphasize that adipose tissue-specific knockout of FTO promotes the browning of WAT, with the mechanism involving elevated m6A levels of Hif1a mRNA due to FTO deficiency, which enhances HIF1A protein translation through YTHDC2 and subsequently activates thermogenic genes. This apparent contradiction may stem from the dual roles of FTO in different types of adipose tissue—acting as a thermogenesis promoter in established brown adipose tissue while functioning as an inhibitor of browning in white adipose tissue. The key lies in the precise regulation of FTO expression levels and its tissue-specific effects: in brown adipose tissue, moderate physiological upregulation (such as the 1.5–twofold increase in expression caused by the rs1421085 C allele) optimizes existing thermogenic mechanisms; whereas in white adipose tissue, complete FTO ablation releases inhibition of the HIF1A pathway, activating alternative thermogenic pathways. Together, these findings reveal the complex role of FTO in the energy balance regulatory network and provide a theoretical foundation for developing precision obesity treatment strategies tailored to individuals with different genetic backgrounds.

In addition to thermogenic and glycolytic genes, m6A modification has also been shown to affect white adipose tissue browning by regulating the levels of a number of other factors. For example, YTHDF1 promotes the mRNA translation of bone morphogenetic protein 8b (Bmp8b) to induce WAT browning in an m6A-dependent manner, and knockdown of Bmp8b blocks the promotion of browning by YTHDF1 overexpression [115]. In addition, the m6A modification-mediated regulation of WAT browning was also found to be affected by exercise. β-hydroxybutyric acid (BHBA), a metabolite secreted after acute exercise, has been shown to upregulate m6A modification in WAT, a process that promotes transcription of key browning genes by enhancing m6A-dependent translation of cAMP Response Element-Binding Protein Binding Protein (CREBBP) histone acetyltransferase to increase chromatin accessibility [116].

Finally, recent studies have revealed distinct regulatory roles of m6A modification in WAT and BAT. In BAT, the lack of METTL14-mediated m6A modification enhances systemic insulin sensitivity by promoting the secretion of prostaglandins PGE₂ and PGF₂α. Mechanistically, METTL14 induces m6A deposition on mRNAs encoding prostaglandin-synthetic enzymes, such as PTGES₂ (PGE₂ synthase) and CBR₁ (PGF₂α reductase), which are subsequently degraded via YTHDF2/3 recognition [117]. This pathway operates independently of the classical thermogenic protein UCP1, unveiling a novel mechanism by which BAT-derived secretory factors regulate metabolic homeostasis. But, knockdown of METTL14 in WAT yields different outcomes. Integrated m6A-seq and RNA-seq analyses demonstrate activation of apoptotic pathways, accompanied by elevated expression of necrosis factor-related apoptosis-inducing ligand (TRAIL) and TNF receptor superfamily member 1a (TNFR1) mRNAs [118]. This leads to adipocyte apoptosis and systemic insulin resistance. It is noteworthy that METTL14 exhibits non-canonical functions independent of m6A catalysis. In this study, apoptosis-related genes upregulated in METTL14-deficient white adipose tissue were divided into two groups. The first group (e.g., TRAIL, TNFR1, RIPK1) follows the canonical m6A pattern, showing reduced methylation and increased mRNA stability/expression. The second group (e.g., TNFα, DR5, CASP3) showed significant expression changes without altered m6A levels. Because knocking down core m6A targets only partially rescued the apoptosis and insulin resistance phenotypes, these results provide direct evidence for METTL14’s non-canonical roles. Thus, targeting these m6A-independent pathways in WAT represents a promising strategy for treating insulin resistance. Anyway, all these findings highlight a bidirectional m6A regulatory network in adipose tissues—whereas in BAT, the loss of METTL14 enhances metabolic health by promoting prostaglandin production, in WAT its deficiency disrupts metabolic balance through apoptosis induction—providing tissue-specific insights for targeting m6A modifications in obesity and metabolic disorders.

In summary, m6A modification plays important roles in the development and maturation of BAT, the browning of WAT, and the regulation of BAT and WAT functions, which contributes to the maintenance of metabolic homeostasis. (As shown in Fig. 3).

Fig. 3.

M6A is involved in the development of brown adipose tissue and the browning process of white adipose tissue. m6A RNA methylation orchestrates key processes in brown adipose tissue (BAT) development and white adipose tissue (WAT) browning. During BAT development, METTL3-mediated m6A modification regulates the expression of core thermogenic genes (e.g., Prdm16, Ppargc1a, Ucp1). PPARγ protein stability is maintained by YTHDC1, which directly binds to PPARγ and protects it from ubiquitin-mediated degradation, and YTHDC1 exerts a non-canonical function in this process. Prostaglandin receptor 3 (EP3) signaling enhances brown adipogenesis by stabilizing WTAP to promote Zfp410 mRNA methylation. In the context of WAT browning, METTL3 promotes beige adipocyte formation by stabilizing Klf9 mRNA to induce UCP1 expression and by enhancing the stability of glycolytic gene transcripts (e.g., Hk2) via IGFBP2 recognition. Conversely, FTO deficiency elevates Hif1a m6A, enhancing its YTHDC2-dependent translation and subsequent activation of thermogenic gene programs. Besides, the rs1421085 (T-to-C) variant at the FTO gene locus enhances the thermogenic capacity of BAT. Additional regulatory layers include YTHDF1-mediated translational promotion of Bmp8b mRNA and exercise-induced β-hydroxybutyrate, which enhances m6A-dependent CREBBP translation via upregulating m6A modification in white adipose tissue, promoting transcription of key browning genes and browning. Notably, m6A exerts tissue-specific effects: in BAT, METTL14 deficiency stabilizes prostaglandin synthase mRNA, improving systemic insulin sensitivity; in WAT, however, METTL14 knockdown activates apoptotic pathways, leading to adipocyte loss and insulin resistance. Figure created with BioRender.com

Potential role of M6A modification-regulated immune cells function in adipose tissue

Adipose tissue is mainly used to store excess energy in the body, but under the condition of pathological obesity, the volume of adipocytes increases. At the same time, the impaired angiogenesis in adipose tissue leads to the decrease of oxygen supply to adipocytes, which leads to the infiltration of a large number of immune cells in adipose tissue, such as T cells and macrophages. The massive infiltration of immune cells in adipose tissue is closely related to the long-term chronic inflammation of adipose tissue [119–121]. There is evidence that m6A modification is involved in the activation, infiltration, and phenotypic differentiation of immune cells [122]. The role of m6A modification in regulating the function of T cells, macrophages, dendritic cells and other immune cells in adipose tissue, as well as their potential effects on adipose tissue inflammation, will be reviewed and discussed here. (As shown in Fig. 4).

Fig. 4.

The potential regulation of adipose tissue immune cell function by m6A RNA methylation. This schematic summarizes the potential role of m6A modification in orchestrating immune responses within adipose tissue. In T cells, m6A regulates proliferation, apoptosis, and regulatory T cell (Treg) homeostasis. METTL3 promotes STAT5 activation and T cell expansion, while FTO protects CD8⁺ T cells from apoptosis. Additionally, USP47 fine-tunes YTHDF1 activity to restrain c-Myc translation and maintain Treg metabolic homeostasis. m6A also governs T cell subset differentiation, infiltration, and cytokine production (see main text). In macrophages, WTAP deficiency drives pro-inflammatory M1 differentiation and suppresses M2 phenotypes by downregulating IDH1 and αα-ketoglutarate. Furthermore, Chronic unpredictable mild stress (CUMS) amplifies inflammation through two pathways: (1) accumulating serum tryptophan to enhance M2 polarization and crown-like structure formation; and (2) upregulating adipocyte FTO to weaken antioxidant defenses. Notably, adipocytes, rather than macrophages, are the primary drivers of inflammatory factor release in this pathway. In dendritic cells (DCs), YTHDF1 impairs antigen cross-presentation, and METTL3 enhances the production of Th17-polarizing cytokines via miR-338-3p. m6A further modulates DC cytokine release and tissue recruitment (see main text). The diagram also outlines m6A-mediated regulation in other immune populations: B cell development, NK cell cytotoxicity and maturation, and neutrophil activation and mobilization, as detailed in the main text. Figure created with BioRender.com

T cells

As one of the main immune cells in adipose tissue, T cells play an important role in regulating the chronic inflammation of adipose tissue in the obese state. On the one hand, studies have found that CD8⁺T cells in adipose tissue promote the inflammatory response of adipose tissue by releasing proinflammatory cytokines such as INF-γ and IL-6 [123]. On the other hand, CD8⁺T cells, by working with macrophages, also participate in the formation of coronal structures (CLS) in adipose tissue, which are peripheral coronal structures surrounding dead/necrotic adipocytes [124]. These changes further induce the persistence of low-level chronic inflammation in adipose tissue in the obese state. Depletion of CD4⁺ and CD8⁺T cells significantly reduced inflammation in perigonadal adipose tissue in obese mice induced by a HFD feeding [125]. Further studies found that m6A methylation modification was found to directly regulate T cell development and T cell immune infiltration. By targeting the suppressor of cytokine signaling (SOCS) gene, METTL3 causes its mRNA to be rapidly degraded under IL-7 stimulation, which activates the downstream target gene STAT5 and initiates the reprogramming of T cells, ultimately promoting the differentiation and proliferation of T cells. On the contrary, knockout of METTL3 in T cells inhibits m6A methylation, leading to increased stability and slowed degradation (prolonged half-life) of mRNAs from SOCS family genes such as Socs1, Socs3, and Cish due to the loss of the m6A mark, thereby upregulating their mRNA and protein expression levels. The elevated SOCS proteins subsequently suppress the IL-7/STAT5 signaling pathway, ultimately impairing T cell proliferation and differentiation [126]. Deletion of ALKBH5 in CD4⁺T cells results in increased m6A modification of INF-γ mRNA and decreases the expression of the proinflammatory INF-γ [127].At the same time, m6A modification is also involved in the apoptosis process of T cells. FTO protects CD8⁺T cells from excessive apoptosis by down-regulating the m6A modification level of Fas mRNA, reducing its stability, thereby limiting the expression of Fas [128]. Although these results indicate a strong association between m6A modification and T cells functions, whether m6A is involved in the chronic inflammation of adipose tissue in the obese state by regulating the proliferation, infiltration, apoptosis of T cells and the release of inflammatory factors needs to be further verified.

Different subsets of T cells may have different functions. For example, Th1 and Th17 subsets mainly play a pro-inflammatory role, while Th2 and Treg subsets mainly play an anti-inflammatory role. Therefore, exploring the regulation of m6A modification on the function of different T cell subsets may be an effective strategy to alleviate chronic inflammation in adipose tissue. In obese mice, pro-inflammatory Th1, Th17, and CD8 + T cells increase in adipose tissue while Th2 cells decrease [129],whereas lean mice primarily harbor anti-inflammatory Th2 and Treg cells [130]. In Th17 cells, METTL3 knockdown stabilizes SOCS3 mRNA and reduces IL-17A and CCR5 expression, thereby inhibiting Th17 differentiation and infiltration [131]. Additionally, FTO deficiency in intestinal epithelial cells causes S1P accumulation, which activates macrophages to secrete SAA1/3 and promote Th17 differentiation [132]. Given that macrophage infiltration is a hallmark of adipose inflammation [133],modulating macrophage-mediated Th17 differentiation via these pathways could improve inflammation. Furthermore, reducing Th1 proportions in adipose tissue enhances insulin sensitivity and alleviates inflammation [134, 135]. obesity elevates both Th1 cell counts and IFN-γ levels in adipose tissue [136]. m6A also regulates Th1 cells: METTL3 knockdown inhibits Th1 differentiation, while METTL14 loss increases IFNγ and TNFα secretion [137]. These findings suggest that m6A influences disease progression by regulating key factors in pro-inflammatory T cell differentiation, though its specific role in chronic adipose inflammation warrants further investigation.

Among anti-inflammatory T cells, Treg, as one of the subsets of CD4⁺ T cells, accounts for only a small proportion of peripheral T cells. The number of Treg cells in VAT progressively decreases with age in HFD-induced or genetically obese mice [138]. In obese individuals, the number of Treg cells is significantly reduced, whereas the proportion of proinflammatory monocytes is elevated [139]. Depletion of Treg cells in adipose tissue results in a stronger proinflammatory environment that promotes the development of type 2 diabetes [140]. Related studies have reported that a significant reduction in the number of Treg cells in VAT is strongly associated with an increase in inflammatory mediators and a decrease in insulin sensitivity [141]. These studies suggest that the down-regulation of Treg also plays an important role in obesity-induced adipose tissue inflammation. This imbalance suggests that the adipose tissue microenvironment in obesity may affect Treg homeostasis at multiple levels—including epitranscriptomic regulation. Therefore, investigating the specific molecular mechanisms underlying Treg number and function, particularly RNA modifications such as m6A that have garnered significant attention in recent years, is crucial for understanding the etiology of adipose tissue inflammation. To increase the proportion of Treg cells in the obesity-associated chronic inflammatory state would provide new strategies for the treatment of obesity related disorders.

m6A modification holds therapeutic potential by regulating Regulatory T cell (Treg) function. The deubiquitinase USP47 removes K63-linked ubiquitination from YTHDF1, impairing its interaction with eIF3A and reducing the promotion of m6A-modified c-Myc mRNA translation; this limits c-Myc levels to prevent excessive glycolysis, thereby maintaining Treg metabolic and functional homeostasis [142]. While YTHDF2 deficiency does not affect peripheral homeostasis, it increases Treg apoptosis and impairs immunosuppressive function within the tumor microenvironment (TME) [143].. In the TME, enhanced TNF signaling upregulates YTHDF2, which accelerates the degradation of transcripts encoding NF-κB negative regulators, thus sustaining NF-κB activation and protecting tumor-infiltrating Tregs from apoptosis. Given m6A’s critical role in Treg metabolic reprogramming (e.g., the c-Myc pathway) and survival (e.g., the NF-κB pathway), metabolic stressors in chronic adipose inflammation—such as free fatty acids and TNF—likely disrupt the m6A network. The adipose microenvironment may resemble the TME in altering m6A readers like YTHDF1 and YTHDF2, leading to Treg dysfunction, a mechanism that warrants further investigation.

In addition, Th2 has been found to have decreased expression levels in adipose tissue in the obese state, and Th2 promotes the expression of IL-10 in obesity by secreting anti-inflammatory factors such as IL-4 and IL-5 to inhibit inflammatory responses [144]. Therefore, in the obese state, a decline in Th2 cells within adipose tissue may directly impair their capacity to sustain an anti-inflammatory milieu. However, the upstream molecular mechanisms driving this Th2 cell imbalance remain unclear. Given that cell differentiation and functional states are subject to precise post-transcriptional regulation, m6A modification as a critical layer of this regulatory network is likely to play a pivotal role in this process.m6A modification plays an important role in the differentiation of Th2 cells.m6A modification regulates Th2 cell differentiation and the pathogenesis of asthma through multiple mechanisms. Overexpression of METTL3 in CD4 + T cells alleviates pulmonary inflammation in asthmatic mice by promoting Th1 differentiation and reducing Th2/Th17 differentiation [145]. PM2.5 inhibits the expression of ALKBH5, thereby increasing the m6A methylation of SRSF1 mRNA and enhancing its stability, which promotes Th2 polarization and exacerbates allergic asthma in children [146]. In human bronchial epithelial cells, knockdown of METTL3 reduces m6A methylation of SOX5 mRNA, leading to increased SOX5 expression. This subsequently upregulates GATA3, promoting Th2 differentiation and IL-4 production, ultimately exacerbating T2 asthma. Conversely, overexpression of METTL3 reverses this process [147]. While the aforementioned studies primarily elucidated the regulatory role of m6A in Th2 differentiation within the context of allergic asthma, the core principles they revealed—that environmental factors (e.g., PM2.5) or key enzymes (e.g., METTL3, ALKBH5) alter m6A modification levels to subsequently impact the expression of Th2-defining transcription factors (such as GATA3) or splicing factors (such as SRSF1) which provide a novel perspective for understanding Th2 cell depletion in obese adipose tissue. In the setting of chronic adipose inflammation, it remains an open question whether similar metabolic stressors (e.g., elevated free fatty acids, hyperglycemia) or local cytokine milieus might likewise disrupt the m6A "code" within adipose-resident T cells or antigen-presenting cells, thereby suppressing Th2 differentiation or function through analogous molecular pathways (such as the SOX5-GATA3 axis). This represents a highly promising avenue for future investigation.

In summary, it is known that different types of T cells promote or inhibit the progression of chronic adipose tissue inflammation by secreting pro-inflammatory or anti-inflammatory factors. The proportion of pro-inflammatory T cells is increased and the proportion of anti-inflammatory T cells is decreased in obesity. m6A modification plays a role in the differentiation, survival and cytokine release of different types of T cells. The potential role of m6A in adipose T cells functions and its involvement in obesity-associated chronic adipose tissue inflammation is worthy of further exploration, which may provide new strategies to improve adipose metabolism and combat obesity.

Macrophages and dendritic cells

Macrophages, as one of the main immune cells in adipose tissue, have been proven to be closely related to adipose tissue inflammation. Studies have found that F4/80⁺ macrophages account for approximately 12% of total cells in adipose tissue from lean mice, while in adipose tissue from obese mice, the proportion of F4/80⁺ macrophages increases to approximately 41%. This indicates a significant increase in macrophage expression levels in adipose tissue under obese conditions. Macrophages can be classified into two major types based on their functions and different cell surface markers: M1 type pro-inflammatory macrophages and M2 type anti-inflammatory macrophages [148]. In obese mice, the number of adipose M1-type macrophages is increased, which contributes to inflammation and metabolic dysfunction in adipose tissue through the production of proinflammatory cytokines [149].Further investigations indicate that m6A modification in adipose tissue macrophages (ATMs) is closely associated with adipose tissue inflammation. Specifically, WTAP knockout directly downregulates intracellular m6A levels in macrophages and suppresses the expression of M2 macrophage marker genes (Arg1, Ym1, and Retnla) as well as the surface marker CD206, thereby preventing their differentiation into anti-inflammatory M2 macrophages. Conversely, WTAP deficiency enhances M1 polarization. Under palmitic acid (PA) stimulation, WTAP-deficient macrophages secrete higher levels of pro-inflammatory cytokines, such as IL-1β and TNF-α, compared to wild-type macrophages, driving these WTAP-deficient cells to skew toward a pro-inflammatory M1 phenotype [150].Mechanistically, WTAP deficiency in ATMs suppresses the mRNA stability and translation efficiency of Idh1 by downregulating its m6A modification. Subsequently, the downregulated IDH1 protein expression impedes the conversion of isocitrate to α-ketoglutarate (α-KG), resulting in a significant decrease in intracellular α-KG production. As a core metabolite orchestrating macrophage metabolism and polarization, the depletion of α-KG leads to a marked reduction in oxidative phosphorylation (OXPHOS) levels within macrophages [150].The integrity of oxidative metabolism is an essential prerequisite for macrophage polarization toward the anti-inflammatory M2 phenotype. Impaired metabolic reprogramming renders macrophages incapable of maintaining this anti-inflammatory phenotype, keeping them persistently skewed toward the pro-inflammatory M1 state, which ultimately drives the initiation and progression of chronic inflammation in adipose tissue. Furthermore, macrophage dysfunction triggered by WTAP deficiency inhibits the beiging and energy metabolism of adipose tissue, further exacerbating adipose tissue metabolic disorders [150].This, in turn, provides a pathological microenvironment for the persistence of inflammation, thereby creating a vicious cycle between inflammation and metabolic dysfunction. Overall, this study primarily highlights how m6A modification in ATMs orchestrates phenotypic transitions in macrophages to modulate the release of pro-inflammatory cytokines, thereby contributing to adipose tissue inflammation. Furthermore, similar studies have indicated that macrophages and adipose tissue inflammation are interconnected via m6A mechanisms. Chronic unpredictable mild stress (CUMS) leads to the accumulation of tryptophan (TRP) in the serum; the elevated TRP activates the aryl hydrocarbon receptor (AhR) in macrophages to induce their M2 polarization, resulting in the robust secretion of the chemokine CXCL1. This leads to a significant increase in the number of macrophage-formed “crown-like structures” (CLS) in mammary adipose tissue, which are typical hallmarks of adipose tissue inflammation [151]. Concurrently, macrophage-induced upregulation of CXCL1 further increases FTO expression in adipocytes, which reduces the m6A modification of KEAP1 mRNA and thereby enhances its stability. The resulting elevation of KEAP1 strengthens its interaction with the core oxidative stress regulator NRF2, subsequently accelerating the proteasomal degradation of NRF2. This impairs the antioxidant capacity of adipocytes and triggers the production of excessive reactive oxygen species (ROS) accompanied by mitochondrial fission. [151]. As a crucial inflammatory mediator, ROS further exacerbates the inflammatory state of the adipose tissue. Lastly, evidence suggests that FTO may exacerbate adipose tissue inflammation by indirectly modulating inflammatory factor expression in adipose tissue macrophages (ATMs). Within the FTO intron, an obesity-associated non-coding region acts as a long-range enhancer that cis-regulates IRX3 expression via chromatin looping—primarily in the brain—thereby affecting adiposity and body weight [152].Furthermore, it was found that in ATMs, phosphorylated IRX3 dimerizes and translocates to the nucleus, where it directly binds to the promoters of pro-inflammatory genes (e.g., IL-1α, IL-1β, IL-6, TNF) to enhance transcription. This process escalates the release of pro-inflammatory cytokines, contributing to adipose tissue inflammation [153].While the specific regulation of IRX3 by FTO within adipose tissue remains to be fully clarified, these findings offer a hypothesis: FTO-mediated elevation of IRX3 in adipose tissue may contribute to the aggravation of inflammation during obesity. Given that research linking macrophages to adipose tissue inflammation specifically via m6A mechanisms is still in its infancy, the direct evidence available for review remains limited. Nevertheless, the existing evidence provides indirect links demonstrating the association between macrophages and adipose tissue inflammation.

There is an obvious interaction between macrophages and T cells in the process of adipose tissue inflammation. Obesity-induced activation of CD4⁺ T cells requires antigen presentation in the context of MHC II molecules, the expression of T-cell costimulatory molecules, and cytokines that drive differentiation of CD4⁺ T cell subsets [154, 155]. Studies have shown that adipose tissue macrophages in visceral adipose tissue can present antigens on MHC II molecules, express T cell costimulatory markers and induce CD4⁺ T cell proliferation in an antigen-dependent manner [156]. Specific knockdown of MHC II on adipose tissue macrophages in a mouse model of HFD-induced obesity significantly reduces the accumulation of inflammatory macrophages and CD4 + T cells in adipose tissue, as well as decreased INF-γ expression in adipose T cells [157].Depletion of adipose tissue macrophages reduces the frequency of Tregs in adipose tissue, whereas injection of isolated adipose tissue macrophages into epididymal white adipose tissue significantly enhances Treg expression levels. This regulation is macrophage phenotype-dependent. Anti-inflammatory macrophages enhance Treg function and survival by secreting cytokines such as Semaphorin-4A, which interacts with neuropilin-1 (nrp1) on Tregs. In contrast, during obesity, adipose tissue macrophages shift toward a pro-inflammatory phenotype, leading to increased secretion of proinflammatory factors and decreased expression of Semaphorin-4A in adipose tissue, ultimately suppressing Tregs function [158]. m6A plays an important role in the crosstalk between T cells and macrophages. Myeloid knockdown of METTL3 impairs YTHDF3-dependent degradation of Pentraxin 3 (Ptx3) mRNA, thereby enhancing M2 macrophage activation to promote a Th2 cell response, which significantly exacerbates allergic airway inflammation [159]. Knockdown of METTL14 in macrophages reduced the m6A modification level of Epstein-Barr virus-induced protein 3 (Ebi3) transcript and increased the protein expression level of Ebi3, leading to the increase of IL-27 and IL-35, which directly inhibited the activation and effector function of CD8⁺ T cells. This is accompanied by a decrease in the capacity of CD8⁺ T cells to secrete INF-γ, which eventually leads to the accumulation of functionally impaired CD8⁺ T cells [160]. These results suggest that m6A modification affects the polarization state and cellular function of macrophages by dynamically regulating inflammation-related gene expression in macrophages, which may further affect T cell function. The further polarized M1-type macrophages, together with regulated T cells, release inflammation-related factors that contribute to adipose tissue inflammation in the context of obesity. This provides a new perspective for understanding the molecular mechanisms of adipose tissue inflammation and suggests that targeting m6A modification in macrophages may represent a potential strategy for intervening in obesity-related metabolic inflammation.

Dendritic cells (DCs), derived from hematopoietic stem cells, function as professional antigen-presenting cells in the innate immune system. They recognize and uptake pathogen antigens, process them into antigen peptide-MHC complexes, and present these complexes to naive T cells. Concurrently, they release co-stimulatory signals and cytokines, thereby initiating adaptive immune responses. Dendritic cells also play a critical role in chronic inflammation of adipose tissue and insulin resistance [161]. In a study using a conditional cell ablation system in HFD-fed mice, where transgenic expression of the diphtheria toxin receptor under the control of the CD11c promoter was employed to deplete CD11c + dendritic cell subsets in obese mice, ablation of CD11c + cells led to significant reduction in local and systemic inflammatory markers, accompanied by marked improvement in insulin sensitivity. In obese mice and patients with insulin resistance, specific subtypes of DCs, CD11c(high)F4/80(low) in mice and CD11c (+) CD1c (+) in humans, accumulate in adipose tissue [162]. CD11c(high)F4/80(low) DCs from obese mice are able to induce naive T cells to differentiate into Th17 cells [162, 163]. This suggests that crosstalk between DCs and T cells function affects adipose tissue inflammation in the obese state. Therefore, investigating how the abundance, subset differentiation, and functional states—particularly their capacity for cross-talk with T cells—of dendritic cells (DCs) in adipose tissue are regulated under obese conditions is crucial for elucidating the mechanisms underlying chronic inflammation. Within this process, post-transcriptional regulation of gene expression, especially the dynamically reversible m6A modification, likely constitutes a central regulatory layer.

Studies have reported that m6A modification plays a role in functional crosstalk between DCs and T cells. Knocking down YTHDF1 in gastric cancer tumor cells promotes the recruitment of mature DCs (CD11c + MHCII +) in the tumor microenvironment, enhances the expression of MHCII on the surface of these dendritic cells, and increases the secretion of interleukin-12 (IL-12). This, in turn, promotes the infiltration of CD4⁺ and CD8⁺ T cells and the secretion of interferon-γ (IFN-γ) by T cells [164]. Besides, knockdown of YTHDF1 in classical DCs enhances cross-presentation of tumor antigens and cross-priming of CD8⁺ T cells [165]. Mechanistically, YTHDF1 promotes the translation of mRNAs encoding lysosomal proteases including cathepsins B, D, and L in dendritic cells (DCs), increasing the levels of these lysosomal proteases, which leads to excessive degradation of tumor antigens in DCs, thereby reducing the number of antigens available for cross-presentation. When YTHDF1 is absent or these cathepsins are directly inhibited, antigen degradation is reduced, cross-presentation is enhanced, ultimately resulting in stronger CD8⁺ T cell-mediated antitumor immune responses. The infiltration of T cells into adipose tissue contributes to adipose tissue inflammation by promoting lipolysis. In this process, Th17 cells play a direct role through the release of proinflammatory factors. Additionally, evidence suggests that the crosstalk between Th17 cells and DCs is modulated by m6A modification. METTL3 was found to promote the cleavage and processing maturation of miR-338-3p in activated DCs by increasing the m6A modification level of miR-338-3p. MiR-338-3p enhanced the activity of MAPK p38 signaling pathway by inhibiting dual-specificity phosphatase 16 (Dusp16), thereby increasing the production of Th17-polarizing cytokines such as IL-6, IL-1β, and IL-23 [166]. At the same time, evidence that m6A is involved in inflammation and metabolism-related diseases by regulating the generation, activation and migration of DCs have gradually enriched [167]. This further points to the diversity and potential of m6A modification in the regulation of DCs function. Extending these findings to the context of chronic fat tissue inflammation, some insights can be drawn. The metabolic stress factors related to obesity include lipotoxicity and endoplasmic reticulum stress, which are likely to cause abnormal activation of dendritic cells in fat tissue by disrupting the m6A modification homeostasis of these cells. This may lead to their excessive secretion of pro-inflammatory cytokines such as IL-6 and IL-23, promoting the expansion of Th17 cells. This polarization tendency of Th17 cells will further exacerbate the inflammatory response in fat tissue through a positive feedback mechanism. However, the current mechanism of action remains an unexplored research field. Whether m6A modification regulates chronic inflammation in fat tissue by influencing the function of dendritic cells and the underlying molecular mechanisms are all issues that deserve further in-depth exploration.

Other immune cells

In addition to the above immune cells, B cells, NK cells, and neutrophils, as important components of adipose tissue immune cells, have also been shown in relevant studies to have their functions influenced by m6A modifications.

B lymphocytes play an important role in maintaining adipose tissue homeostasis, metabolism, and immune response [168]. In the obese state, the ratio of B cells with a pro-inflammatory phenotype is significantly increased, which further aggravates metabolic disorders and insulin resistance. Thus, elucidating the upstream molecular regulators driving the pro-inflammatory skewing of adipose tissue-resident B cells holds significant implications for intervening in obesity-associated metabolic inflammation. The development, activation, and effector functions of B cells are highly dependent on tightly regulated gene expression programs, which may be broadly modulated by m6A modifications. Knockdown of METTL14 impairs interleukin-7 (IL-7)-induced proliferation of pre-B cells and the transition of large pre-B cells to small pre-B cells in the early stage of B-cell development, ultimately leading to abnormal B-cell development [169]. In germinal center B-cell (GCB) cells, loss of METTL3 slows the cell cycle and decreases the expression of genes involved in proliferation and oxidative phosphorylation, which indirectly leads to B-cell immune dysfunction [170]. A recent report also revealed a novel role of m6A in regulating B cell function. In response to activation signals, ALKBH5 and translational regulatory long non-coding RNA1 (treRNA1) enter the nucleus to form a complex with DDX46 (RNA helicase) to coordinate the removal of m6A modification on BCR (B cell receptor)-related transcripts. This demethylation increases the transcripts stability and facilitates their efficient translation by facilitating their transport to the cytoplasm through interaction with the RNA-binding protein HuR [171]. Deletion of any link in this axis leads to impaired transcripts processing and decreased expression of BCR-related genes, leading to B-cell dysfunction. These lines of evidence all suggest that m6A modification plays an important role in B cell maturation and cellular function. The aforementioned studies have revealed the central role of m6A in regulating fundamental biological processes in B cells, including early development, proliferative metabolism, and BCR signal transduction. When contextualized within the metabolic inflammatory milieu of obesity, a plausible hypothesis emerges: obesity-associated aberrant metabolites (e.g., oxidized lipids, hyperglycemia) or inflammatory signals in the adipose microenvironment may disrupt the activity of m6A-modifying enzymes (such as METTL3, METTL14, or ALKBH5) within B cells, thereby reprogramming the methylome landscape of their transcriptome. This alteration could specifically impact polarization-determining gene networks (e.g., those governing metabolic reprogramming or BCR signal intensity) that dictate B cell fate (pro-inflammatory vs. regulatory), ultimately culminating in the expansion and hyperactivation of pro-inflammatory B cell subsets within adipose tissue. However, how m6A modification dynamically regulates the phenotypic switching of adipose tissue B cells (such as pro-inflammatory/anti-inflammatory balance), and whether targeting this pathway can improve obesity-related metabolic disorders remain to be further studied.

NK cells are enriched and chronically activated in adipose tissue during obesity, promoting inflammation by releasing pro-inflammatory factors like TNFα and participating in insulin resistance via macrophage activation; indeed, HFD feeding leads to an increase in NK cell numbers in murine adipose tissue [172]. Therefore, suppressing the aberrant activation of adipose-resident NK cells represents a novel strategy for alleviating metabolic inflammation. Given the core role of m6A in gene expression, studies have found that YTHDF2 is significantly upregulated during NK cell maturation, whereas YTHDF2 knockout significantly reduces the expression of Eomes, a key regulator of terminal maturation [173]. Furthermore, YTHDF2 deficiency leads to decreased expression of the activating receptors CD226 and NKG2D, and reduces IFN-γ secretion upon IL-12 and IL-18 stimulation, thereby suppressing NK cell effector functions [173]. On the other hand, METTL3 directly regulates the Ptpn11 gene (encoding SHP-2) via m6A methylation, affecting the IL-15-mediated AKT-mTOR/MAPK-ERK signaling pathway to maintain terminal NK cell maturation [174]. Mechanistically, METTL3 maintains SHP-2 protein levels primarily by enhancing translation efficiency rather than affecting mRNA stability; thus, METTL3 loss reduces SHP-2 expression, impairing NK cell metabolic activity and proliferative capacity. In summary, m6A precisely regulates NK cell fate and function. It is hypothesized that in obesity-associated chronic inflammation, metabolic stressors (e.g., free fatty acids, hypoxia) act as “aberrant stimuli” that disrupt m6A homeostasis in adipose NK cells. This dysregulation may abnormally enhance METTL3/YTHDF2-mediated translation of pro-maturation molecules, driving excessive activation and the release of IFN-γ and TNFα, while facilitating adaptive survival in the inflammatory environment via molecules like SHP-2. Consequently, regulating m6A levels to inhibit excessive NK cell activation may be an effective strategy to improve chronic adipose inflammation, a mechanism that warrants further investigation.

Under physiological conditions, neutrophils are present at low levels in adipose tissue and are primarily localized within blood vessels. However, in obesity-induced chronic inflammation, neutrophils are recruited by endogenous danger signals and become among the first immune cells to arrive at inflammatory sites. They subsequently perform immune clearance functions through phagocytosis, degranulation, and the release of neutrophil extracellular traps (NETs). Although this is a sterile process, proteases and antimicrobial peptides carried by NETs can act as potent damage-associated molecular patterns, which may further activate macrophages and other immune cells to release large quantities of inflammatory cytokines, thereby driving a vicious cycle of inflammatory network amplification [175]. Therefore, the aberrant recruitment and hyperactivation of neutrophils represent a critical early event that initiates and amplifies the inflammatory cascade in adipose tissue. This underscores the importance of deciphering the upstream molecular switches that regulate neutrophil infiltration into adipose tissue and their functional states under obese conditions—key to breaking the vicious cycle of inflammation. These switches likely involve the precise spatiotemporal control of cellular programs governing chemotaxis, survival, and effector functions. Recent evidence further reveals that neutrophils participate in high-fat diet (HFD)-induced chronic inflammation in mouse adipose tissue in which circadian rhythm gene regulation is involved. Specifically, deletion of the Bmal1 gene in neutrophils alters the expression of inflammation-related chemokines and receptors at the gene level, including reduced expression of CXCL2 and CXCR2, thereby decreasing neutrophil recruitment to adipose tissue. Simultaneously, Bmal1 deficiency in neutrophils increases CXCR4 gene expression, which promotes a neutrophil senescence phenotype that may protect against excessive inflammatory activity and limits inflammatory cell infiltration. Through these two mechanisms, the elevation of inflammation levels in adipose tissue is ultimately suppressed [176]. Although a direct link between m6A and Bmal1 in neutrophils remains unclear, it is noteworthy that studies in hepatocytes have shown that Bmal1 deletion can increase reactive oxygen species levels, leading to elevated expression of METTL3 and YTHDF2 as well as increased m6A modification [177]. We therefore speculate that a similar "Bmal1–ROS–m6A" regulatory axis may exist in neutrophils and could potentially modulate chronic adipose tissue inflammation by influencing Bmal1 itself or its downstream inflammatory target genes. This hypothesis integrates metabolic dysregulation (potentially via ROS), core clock genes, and epitranscriptomic regulation into a unified framework, offering an innovative perspective on how obesity systematically reprograms immune cell function. If such an axis exists, it implies that inflammation in adipose tissue may be intrinsically linked to circadian rhythm disruption and aberrant m6A modification.

Finally, while the primary focus of this review is on the regulatory mechanisms of m6A-mediated adipose tissue inflammation involving distinct immune cell subsets in the adipose tissue immune microenvironment, the potential contribution of intrinsic m6A regulatory mechanisms within adipocytes cannot be overlooked. Alterations in m6A modifications within adipocytes may govern the stability and translational efficiency of key adipokines and chemokines, thereby initiating the downstream recruitment of distinct immune cell subsets. Future studies using adipocyte-specific knockout models are warranted to further dissect this complex crosstalk and provide a more comprehensive perspective on adipose tissue inflammation.

Therapeutic strategies

Given the multifaceted regulatory potential of m6A modification in adipose tissue lipolysis, adipogenesis, thermogenesis, and immune cell function, the development of targeted m6A intervention strategies is of great significance. specifically, screening for distinct m6A regulatory targets (writers, erasers, and readers), developing diverse intervention modalities (such as small molecule inhibitors/activators, antisense oligonucleotides, siRNA, and CRISPR-mediated gene editing), and exploring efficient delivery systems targeting adipose tissue and adipocytes will provide novel insights and feasible pathways for the prevention and treatment of obesity and metabolic syndrome-related diseases.

In recent years, evidence supporting the targeting of m6A modifications to ameliorate obesity has been accumulating. Recent studies have proposed an innovative intervention strategy based on Proteolysis Targeting Chimera (PROTAC) technology, aiming to treat obesity and related metabolic disorders by targeting the degradation of the FTO protein to regulate m6A modification. Specifically, this research demonstrated that FTO degradation-targeting chimera (FTO-DT) can directly target FTO within adipose tissue [178].The research team selected meclofenamic acid (MA), which exhibits high selectivity for FTO, as the targeting ligand. Based on structural biological analysis, they determined the binding mode between MA and FTO (at Ser229) and utilized the C-3 position of MA as the attachment point. By coupling this with pomalidomide, a ligand for the E3 ubiquitin ligase CRBN, via a linker, they successfully constructed the PROTAC molecule FTO-DT. Mechanistic studies revealed that the action of FTO-DT is highly dependent on the ubiquitin–proteasome pathway, as proteasome inhibitors completely abrogated FTO-DT-mediated FTO degradation. The molecule achieved efficient and long-lasting degradation of FTO protein in 3T3-L1 adipocytes at nanomolar levels (0.33 nM) and showed no cytotoxicity at effective concentrations. By reversing FTO-mediated m6A demethylation, FTO-DT remodels adipose tissue function primarily through two pathways: first, by significantly enhancing mitochondrial quantity and function to promote ATP generation and inducing browning gene expression in subcutaneous white adipose tissue, thereby driving the shift from energy storage to energy expenditure; and second, by alleviating oxidative stress via the reduction of ROS and lipid peroxidation products while upregulating glutathione levels, thus inhibiting excessive lipid deposition and abnormal differentiation. In vivo experiments further confirmed that FTO-DT precisely targets subcutaneous and epididymal adipose tissues. It dose-dependently reduced body weight and body fat rates in high-fat diet (HFD) mice by increasing systemic energy expenditure without affecting food intake. Furthermore, this strategy significantly ameliorated glucose metabolism disorders, enhanced insulin sensitivity, and improved accompanying hepatic steatosis.In summary, FTO-DT achieves effective regulation of lipid metabolism and systemic energy homeostasis through the precise targeted degradation of FTO protein in adipose tissue, providing a new theoretical basis and direction for drug development in the treatment of obesity and its complications. However, research into this strategy is still in its infancy, and several issues remain to be addressed. The study only verified changes in overall m6A levels in adipocytes following FTO-DT treatment but failed to identify the specific m6A-modified target genes directly regulated within the adipose tissue. Simultaneously, critical data regarding druggability is severely lacking. The study did not conduct complete pharmacokinetic studies; thus, the in vivo metabolic characteristics, exposure levels, and the dose–response relationship remain unclear. Moreover, the FTO-DT molecule exhibits poor water solubility and was administered solely via intraperitoneal injection, without an evaluation of oral bioavailability. This limitation significantly restricts its potential for clinical translation in the treatment of chronic metabolic diseases such as obesity.

Beyond PROTAC Technology, specific compounds have been developed to inhibit the catalytic activity of m6A effector molecules. Through virtual screening, Entacapone was identified as a compound capable of directly binding to FTO and inhibiting its activity in vitro [179]. Furthermore, Entacapone was shown to reduce body weight and lower fasting blood glucose concentrations in diet-induced obese mice. The underlying molecular mechanism likely relies on the Entacapone-FTO-FOXO1 regulatory axis, wherein the mRNA of the transcription factor Forkhead box protein O1 (FOXO1) serves as a direct substrate for FTO. Additionally, a recent review proposed an emerging therapeutic strategy targeting FTO: Cold Atmospheric Plasma (CAP). CAP represents the fourth state of matter and is composed of various reactive oxygen and nitrogen species (RONS). It can be administered via flexible clinical forms such as plasma-activated liquid (PAL), sprays, and hydrogels. The study highlights CAP’s characteristics of being anti-inflammatory, exhibiting low toxicity, and possessing pleiotropic effects, with proven efficacy in diabetes and cancer [180]. Concurrently, research indicates that CAP can inhibit FTO expression [181], positioning it as a promising novel modality for targeting FTO to treat inflammation-driven metabolic syndrome. Current research on m6A target inhibition predominantly focuses on FTO. Beyond the aforementioned inhibitors, other agents include Meclofenamic acid (MA) [182], Hydroxyquinoline-based inhibitors [183] Pyridine dicarboxylic acid (PDCA) [183], and IOX1 (5-carboxy-8-hydroxyquinoline) [183]. However, the development of inhibitors for key m6A molecules remains heavily skewed towards FTO, with a notable lack of targeted inhibitors for writers like METTL3/METTL14 and major readers such as the YTHDF family. Furthermore, the specific inhibitory effects associated with FTO in adipose tissue have not yet been fully validated. This lack of confirmation extends to the efficacy of relevant inhibitors in suppressing FTO activity within adipose tissue, as well as the question of whether such inhibitor-induced FTO suppression can effectively ameliorate obesity. Therefore, further investigation into whether known FTO inhibitors can effectively ameliorate obesity and metabolic syndrome is essential.

Beyond small molecule inhibitors, gene editing targeting m6A represents a promising future strategy for ameliorating obesity and metabolic syndrome. Current research has developed programmable Targeted RNA Methylation (TRM) systems. By fusing catalytically inactive Cas13 (dCas13) with modified m6A methyltransferase domains, two core editors—dCas13–M3 and dCas13–M3M14—were constructed. Furthermore, the addition of nuclear localization signals (NLS) or nuclear export signals (NES) to these editors enables site-specific m6A modification of RNA in both the nucleus and cytoplasm [184]. For instance, employing TRM (such as dCas13–M3nls) to target functional m6A sites on lipolysis-related genes in adipose tissue during obesity can increase m6A levels. Depending on the site characteristics, this can regulate mRNA stability or translation efficiency (e.g., promoting the translation of lipolytic genes or accelerating the degradation of pro-obesity genes), thereby enhancing lipolytic signaling to alleviate obesity. While this system is primarily designed for the targeted delivery of methyltransferases and cannot directly enzymatically remove elevated m6A modifications on pathogenic mRNAs associated with obesity and metabolic diseases, it can leverage its ‘precise RNA targeting’ capability to function as a competitive inhibitor. By binding to target m6A sites, it blocks the continuous methylation induced by endogenous methyltransferases. Specifically, dCas13 can precisely bind to mRNA sequences via guide RNAs; if the guide RNA is designed to target sequences surrounding high-m6A sites, dCas13 occupies the region, sterically hindering the binding of the endogenous m6A writer complex (METTL3-METTL14). This prevents further methylation, indirectly inhibiting the rise of m6A levels and reducing the methylation of pro-obesity mRNAs, ultimately suppressing their expression to improve obesity outcomes. However, it is important to note that this technology still carries off-target risks. Although studies confirm that off-target methylation does not significantly alter intracellular transcript abundance, the accumulation of long-term, low-abundance off-target modifications may interfere with physiological processes, limiting direct application in in vivo models or therapeutic scenarios. Simultaneously, there is a lack of delivery vectors for the TRM system that specifically target adipose tissue. Existing AAV vectors (e.g., AAV8, AAV9) exhibit a tropism for liver and muscle, with low efficiency for adipose tissue targeting. Moreover, adipose tissue comprises distinct subtypes—including subcutaneous, visceral, and brown adipose tissue—which possess significant metabolic differences; for instance, dysfunction of visceral fat is closely linked to metabolic syndrome. Therefore, developing delivery systems specific to adipose subtypes is crucial to avoid detrimental effects on other functional adipose tissues (e.g., brown or beige adipose tissue) when using TRM to target m6A. Future research should focus on developing adipose-specific TRM delivery systems to facilitate the use of site-specific m6A editing for obesity improvement. Additionally, it is necessary to develop delivery systems for TRM variants that utilize demethylases and readers as effectors, thereby achieving comprehensive and bidirectional regulation of m6A modifications in adipose tissue.

In conclusion, targeting m6A modifications within adipose tissue offers a promising multi-dimensional approach for the treatment of obesity and metabolic syndrome. The strategies discussed herein, including PROTAC-mediated protein degradation, pharmacological inhibition of FTO, and programmable RNA methylation editing, demonstrate significant potential in remodeling lipid metabolism and thermogenesis. However, the translation of these innovations into clinical reality faces critical bottlenecks. These challenges include the limited druggability and pharmacokinetic validation of emerging molecules like FTO-DT, the scarcity of inhibitors targeting m6A writers and readers beyond FTO, and the off-target risks associated with gene editing. Most notably, the lack of delivery systems capable of distinguishing between distinct adipose depots hinders precise intervention. Consequently, future research must prioritize the structural optimization of therapeutic agents to improve bioavailability and the engineering of next-generation delivery vectors with high affinity for specific adipose subtypes. Overcoming these barriers will be pivotal in unlocking the full therapeutic value of m6A modulation for metabolic health.

Discussion

The role and mechanism of m6A methylation modifications in adipose tissue have been increasingly elucidated, particularly in lipolysis and adipogenesis. Impaired lipolysis and excessive adipocyte proliferation are critical drivers of obesity progression. m6A modification play pivotal roles in these processes by regulating the expression of lipolytic genes, modulating cyclin levels, controlling key adipogenic factors, and influencing adipocyte autophagy and apoptosis. Moreover, under obese conditions, brown adipose tissue exhibits reduced thermogenic capacity, enlarged lipid droplets, decreased mitochondrial numbers, and impaired mitochondrial function. Recent studies indicate that m6A modification regulate BAT development and the browning of white adipose tissue, offering a new perspective for improving BAT dysfunction in obesity. Targeting this complex m6A-mediated regulatory network in adipose tissue offers novel therapeutic opportunities for obesity. Additionally, m6A modification alter the expression of biosynthetic genes in BAT, affecting its secretome and thereby regulating systemic metabolic homeostasis, which is a newly identified regulatory paradigm. These findings underscore the tissue-specific and diversified functions of m6A regulatory networks. In the future research, whether m6A modification regulate adipose-derived secretory factors (e.g., leptin, adiponectin, and resistin) to further modulate systemic metabolic homeostasis could be studied. However, studies on m6A modification in regulating the function of BAT remain in their infancy, particularly regarding their impact on BAT thermogenic function and BAT-mediated systemic metabolic regulation. The roles of m6A in regulating the cellular senescence, autophagy, apoptosis, and proliferation of BAT also remain poorly understood.

In addition to impaired lipolysis, dysregulated adipogenesis, and brown adipose tissue dysfunction, chronic inflammation in adipose tissue is also a key hallmark of obesity. Chronic inflammation in adipose tissue leads to the overactivation of immune cells within it, which is a dysfunctional and detrimental activation. This ultimately depletes immune resources, induces immunosuppression, thereby weakening the immune system's overall surveillance and defense capabilities against other diseases and resulting in a state of "immunosuppression" in the body. m6A modification plays an important role in regulating the function of immune cells, which may be involved in regulating adipose inflammation. Studies have shown that m6A modification influence the differentiation of T cells into anti-inflammatory or pro-inflammatory phenotypes, indirectly modulating inflammation levels in adipose tissue. Furthermore, m6A regulates the interactions between macrophages, dendritic cells, and T cells, further impacting the immune status of adipose tissue. Although the roles of m6A modification in B lymphocytes, NK cells, and neutrophils within adipose tissue remain to be fully elucidated, substantial evidence highlights the importance of m6A in the development and function of these immune cells. Therefore, uncovering the m6A modification-related regulatory pathways in adipose immune cells would provide new insights for addressing chronic inflammation in adipose tissue.

Furthermore, in this review, we have separately discussed the distinct regulatory roles of m6A in adipogenesis, thermogenesis, and immune cells within adipose tissue. However, emerging evidence is gradually revealing potential interactions among these m6A-mediated regulatory networks in adipose tissue. This represents a highly promising research direction, and recent studies have already provided supportive evidence. Specifically, research indicates that CXCL1 secreted by tumor-associated macrophages (TAMs) acts on adipocytes to upregulate FTO. This increases KEAP1 mRNA demethylation and protein expression [185],which accelerates the degradation of NRF2 and suppresses its antioxidant target genes, such as HO-1 and SOD [186]. Consequently, elevated reactive oxygen species (ROS) levels trigger mitochondrial fission and activate the PKA pathway, leading to HSL phosphorylation and lipolysis [151] In turn, excessive oxidative stress activates the β-adrenergic receptor/adenylyl cyclase/protein kinase A (PKA) pathway, which induces the phosphorylation of HSL at Ser660 and ultimately promotes lipolysis [151]. This study clearly illustrates the crosstalk between immune cells and adipocyte, highlighting the m6A eraser FTO as a critical molecule that mediates this interaction and regulates the lipolytic function of adipocytes. Furthermore, m6A epitranscriptomic regulation in macrophages serves as a crucial upstream factor dictating adipocyte function and adipose tissue homeostasis. In adipose tissue macrophages (ATMs), WTAP stabilizes Idh1 mRNA expression via m6A modification to promote IDH1 protein translation, which subsequently catalyzes the generation of α-ketoglutarate (α-KG) [150].Conversely, WTAP deficiency reduces αα-KG levels, downregulating thermogenic genes (Ucp1, Cox5a, Cox7a) in scWAT and BAT, thereby impairing thermogenesis. Moreover, Wtap-knockout ATMs polarize towards a pro-inflammatory M1 phenotype, increasing cytokines (IL-1β, TNF-α) and crown-like structures. Ultimately, this inflammatory cascade promotes adipocyte hypertrophy, hepatic lipid accumulation, and exacerbates obesity and insulin resistance under high-fat diet conditions [150].Collectively, these findings suggest that m6A regulators function as pivotal nodes in the intercellular crosstalk between immune cells and adipocytes, orchestrating a complex network that integrates oxidative stress, inflammation, and metabolic homeostasis within adipose tissue.

Finally, non-coding variants of the m6A eraser FTO show a strong genetic association with human obesity [187, 188] primarily, by regulating IRX3 expression [152]. Specifically, an FTO intronic enhancer distally upregulates brain IRX3, which impairs sympathetic innervation and reduces energy expenditure. In adipose tissue macrophages (ATMs), IRX3 promotes pro-inflammatory cytokine release (e.g., IL-1αα, TNF), which suppresses adrenergic signaling and lipolysis in adipocytes [153].Concurrently, these cytokines downregulate thermogenic genes (e.g., Ucp1) in brown and beige fat, thereby promoting energy storage and exacerbating obesity and hepatic steatosis [153].Notably, while FTO influences this pathway via genetic variants rather than direct m6A demethylation, the potential m6A-dependent regulation of downstream IRX3 warrants further study. Collectively, these findings underscore the pivotal roles of m6A regulators (FTO, WTAP) and their targets in orchestrating immune-adipocyte crosstalk, establishing a promising framework for investigating m6A-mediated cross-regulation in adipose tissue homeostasis.

Besides, current research on m6A in adipose tissue has primarily focused on methyltransferases and demethylases, while the functions of m6A reader proteins remain underexplored. As the final effector in the m6A regulatory axis, reader proteins determine whether target mRNA translation is enhanced or suppressed and influence mRNA stability. In the context of lipolysis and adipogenesis, YTHDF1 and YTHDF2 are the most extensively studied readers, whereas the role of adipose YTHDF3 remains less characterized. Unlike YTHDF1, which promotes translation of m6A-marked mRNAs, and YTHDF2, which destabilizes target mRNAs, YTHDF3 exhibits dual functionality—it can cooperatively bind m6A sites shared with YTHDF1/YTHDF2, amplifying their regulatory effects [189]. Notably, YTHDF3 expression is significantly downregulated in obese individuals compared to lean controls and shows a negative linear correlation with BMI [190]. Thus, investigating the role of YTHDF3 in adipose m6A networks represents a promising direction. In addition, research on the role of YTHDC subfamily proteins in m6A-mediated regulation of adipose function is relatively scarce. Unlike YTHDF1/2/3, YTHDC1 is primarily located in the nucleus, while YTHDC2 is present both in the nucleus and the cytoplasm. Existing studies have pointed out that YTHDC1 inhibits the ubiquitination and degradation of PPARγ by directly interacting with it in the cell which is regarded as a non-classical function of YTHDC1, while YTHDC2 promotes the development of brown adipose tissue by enhancing the expression of thermogenesis-related genes. However, overall, research on YTHDC in adipose tissue is still in its early stages. Therefore, a promising direction for future research will be to focus on how the YTHDC protein affects mRNA splicing, transport, stability, and translation, thereby participating in the development and functional processes of adipocytes and adipose tissues.

In addition, available evidence indicates that METTL3 mainly relies on canonical m6A catalytic action to affect target mRNAs in adipose tissue. We have previously summarized how METTL3 regulates key processes, including lipolysis, adipogenesis, and adipose immune cell function, via its m6A catalytic activity. Nevertheless, we also noted earlier that two studies have demonstrated the existence of non-canonical, m6A-catalysis-independent mechanisms for METTL14 in adipose tissue. Similarly, relevant studies have indicated that METTL3 can also exert regulatory functions via m6A-independent pathways. For instance, research has demonstrated that METTL3 drives the senescence-associated secretory phenotype (SASP) at the transcriptional level through chromatin relocalization [191].In normal cells, METTL3 broadly colocalizes with METTL14; however, under senescent conditions, METTL3 exhibits significantly increased binding to the upstream regions of transcription start sites (TSS), whereas METTL14 preferentially binds to distal regions located over 10 kb away from the gene body (i.e., the distal enhancers of SASP genes). Further mechanistic investigations reveal that METTL3 binds to NF-κB p65 sites in the promoter regions of SASP genes and directly interacts with NF-κB p65. This interaction stabilizes the enrichment of p65 at the promoters and enhances both its phosphorylation and nuclear chromatin-binding capacity [191]. Additionally, METTL3 interacts with RNA polymerase II (Pol II), recruiting it to the TSS regions of SASP genes to lay the foundation for transcription initiation. Concurrently, METTL3 and the enhancer-relocated METTL14 mediate the formation of chromatin loops between the promoters and enhancers of SASP genes via direct protein–protein interactions. This significantly shortens the spatial distance between these two regions and further amplifies the transcriptional efficiency of Pol II [191]. Importantly, this entire cascade of events is independent of the m6A methyltransferase activity of METTL3. Catalytically inactive METTL3 mutants can still normally bind to SASP gene promoters and execute the aforementioned regulatory functions, ultimately driving the robust expression of SASP genes—such as IL6, CXCL3, and IL1β—and the subsequent secretion of corresponding cytokines to establish the SASP. The non-canonical functions of METTL3 elucidated in this study have been demonstrated across various cell types, though adipocytes have yet to be investigated. During aging and the progression of obesity-related phenotypes, the levels of SASP-associated transcription factors are elevated within adipose tissue. It raises the intriguing question of whether, under conditions of adipose tissue aging or obesity, METTL3 might similarly undergo intranuclear spatial redistribution. Upon dissociation from METTL14, METTL3 could potentially bind to the promoter regions of SASP-related genes (such as IL6 and IL1β) in adipose tissue, synergizing with transcription-related proteins like NF-κB p65 and Pol II to activate the SASP pathway via chromatin remodeling and transcriptional regulation. This cascade would subsequently promote adipose tissue inflammation and exacerbate adipocyte apoptosis, dysregulated lipolysis, or impaired adipogenesis, ultimately inducing adipose tissue dysfunction. Therefore, further exploration of the potential m6A-independent functional pathway of METTL3 in adipose tissue is highly imperative for ameliorating adipose tissue dysfunction mediated by aging and inflammation. In addition, another study has revealed that METTL3 in hepatocytes reduces chromatin accessibility by recruiting HDAC1/2 to mediate histone deacetylation, thereby repressing the transcription of Cd36 and Ccl2 [192].This results in decreased free fatty acid (FFA) uptake by hepatocytes and a reduction in hepatic macrophage infiltration, which suppresses hepatic lipid accumulation and inflammatory responses, ultimately delaying the transition from non-alcoholic fatty liver (NAFL) to non-alcoholic steatohepatitis (NASH). Conversely, in the NASH state, the nucleocytoplasmic translocation of METTL3 leads to its functional inactivation, which exacerbates the pathological progression. Given that both hepatocytes and adipocytes are key metabolically active cells, they share certain commonalities in the regulatory mechanisms of lipid metabolism. Therefore, it is plausible to hypothesize that METTL3 in adipocytes might recruit histone modification-associated proteins, such as HDAC1/2, to mediate histone deacetylation at the promoter regions of key lipid metabolism genes (e.g., Cd36, FASN ATGL). This would regulate chromatin accessibility, thereby repressing or promoting the transcription of these specific genes, and ultimately impacting processes such as fatty acid uptake, adipogenesis, or lipolysis in adipocytes. However, these hypotheses warrant further experimental investigation for validation.

To rigorously distinguish between the m6A-dependent and m6A-independent functions of METTL3 in adipose biology, future studies could employ catalytically inactive mutants of METTL3 and other m6A modifiers. These mutants eliminate the catalytic activity of METTL3 while preserving its other functions. Specifically, CRISPR/Cas9 technology can be utilized to knock out endogenous METTL3 in in vitro cultured preadipocytes or mature adipocytes, followed by the reconstitution (rescue) with either wild-type (WT) or mutant METTL3. By evaluating changes in adipocyte-related phenotypes—such as SASP gene expression, lipolysis, thermogenesis, and adipogenesis—researchers can determine whether these phenotypic alterations rely on the canonical or non-canonical pathways of METTL3. Furthermore, more direct evidence could be acquired through targeted RNA m6A editing technologies. For instance, utilizing the CRISPR-dCas13 system fused with m6A effector proteins allows for the precise installation or erasure of m6A modifications at specific sites on the mRNAs of particular adipose-related genes (e.g., Cd36 and IL6), without altering the global expression levels of the METTL3 protein. By observing whether such targeted manipulations can phenocopy or abrogate the cellular phenotypes induced by METTL3 overexpression or knockdown, it can be ultimately validated whether a given phenotype is directly mediated by changes in the m6A modification levels of specific genes.

Future research should focus on further perfecting the m6A modification map in different adipose tissues and exploring the differential role of m6A modification in adipose tissue function. Special attention should be paid to the differential expression of m6A levels and the construction of modification maps of its target genes in pathological states such as obesity. This will help reveal the unique mechanism of action of m6A modification in the microenvironment of chronic inflammation in adipose tissue. At the same time, how m6A regulates the function of immune cells and the interaction and crosstalk between immune cells in chronic inflammation is also an important direction for future research. In-depth study of the role of m6A in immune cell function will not only help to improve the immune deficiency caused by obesity, but also provide new targets and strategies for the treatment of obesity and its related metabolic diseases. Furthermore, research focusing on different readers of m6A modification in adipose tissue is also necessary. Currently, research on m6A-regulated intercellular crosstalk in adipose tissue is in its infancy. Future exploration of m6A-mediated interactions between adipocytes and immune cells, as well as among different immune cells, may yield new strategies for treating obesity and related metabolic diseases and inflammation. Besides, whether the YTHDC family regulates adipose tissue function under obesity conditions remains to be thoroughly explored. Concurrently, further exploring and integrating the non-canonical functions of m6A writers, erasers, and readers may elucidate the pleiotropic regulatory landscape of m6A modifications within adipose tissue. Lastly, adipose tissue angiogenesis plays a pivotal role in preserving metabolic health, insulin sensitivity, and the homeostasis of the adipose tissue microenvironment. Notably, the obese state is frequently accompanied by impaired angiogenesis in adipose tissue [193].To date, accumulating evidence has demonstrated the regulatory roles of m6A modifications in angiogenesis, involving key m6A regulators such as YTHDF2 [194], IGF2BP2 [195] and METTL3 [196]. However, direct evidence linking m6A modifications to the regulation of angiogenesis specifically within adipose tissue remains elusive. Therefore, this represents a highly compelling avenue for in-depth future research.

Conclusion

Increasing evidence suggests that m6A modification plays an important role in the regulation of adipose tissue function. In addition to directly regulating lipolysis and differentiation in adipocytes, attention should also be focused on the impact of m6A modification on the secretory factors of adipose tissue, as well as its role in regulating the thermogenic function of adipose tissue. This may provide new perspectives for improving obesity and metabolic disorders. Besides, further exploration and verification of the relationship between m6A modification and immune cells functions in context of chronic inflammation of adipose tissue is also an important research direction. At the same time, the research on the function of m6A in adipose tissue still needs to be further expanded, especially in the aspects of adipose aging, adipose tissue angiogenesis, insulin sensitivity, and the interaction between adipose tissue and other organs, such as skeletal muscle and liver. It is necessary to construct a panoramic map of m6A modification in different adipose tissues and different functions of adipose tissues. In addition, exploring the interaction between m6A modification and immune cells in adipose tissue may provide an effective bridge to elucidate the function and mechanism of immune metabolism. In advancing m6A-targeted therapeutic strategies for adipose tissue, it is also essential to further explore the mechanisms of action of known small-molecule inhibitors of FTO in adipose tissue and pathological states such as obesity, alongside the development of delivery systems for "TRM" (Targeted RNA Modification) that target demethylases and methyl recognition proteins beyond methyltransferases.

Authors contribution

The search and collection of literatures was performed by Q.L., W.-Q. P. and L.G. The first draft of the manuscript was written by Q.L. and L.G. co-wrote the manuscript. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This work was funded by the National Natural Science Foundation of China (Grant Nos. 82571021, 32070751, 31871435), the Shanghai Natural Science Foundation (Grant No. 25ZR1401328), the Shanghai Oriental Talents Program (Top Project, Grant No. BJJY2025018), and the Program for Overseas High-level Talents at Shanghai Institutions of Higher Learning (Grant No. TP2022100). Additionally, this work was supported by the Key Laboratory of Exercise and Health Sciences (Shanghai University of Sport), Ministry of Education (Project No. 2025KF001). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

No new data were generated or analyzed in support of this research.

Declarations

Conflict of interests

Authors declare no conflict of interests.