IntroductionMetabolism is an essential component of biology, enabling rapid and extensive reprogramming through the transcription and translation processes of RNA.1,2,3 Metabolism also controls the activity of enzymes responsible for reversible RNA modifications such as N6-methyladenosine (m6A).3,4,5 m6A is the most prevalent internal modification of messenger RNA (mRNA) in higher eukaryotes. m6A has widespread effects, having been shown to participate in the control of RNA stability, splicing, and translation, and the dysregulation of m6A alterations is closely tied to various developmental diseases, as well as cancer development.6,7,8,9,10 This chemical modification is catalyzed by the METTL3–METTL14 complex, an S-adenosylmethionine (SAM)-dependent RNA methyltransferase (MTase).9,10 The discovery of fat mass and obesity-associated protein (FTO) as the first RNA demethylase highlights the reversible and dynamic nature of m6A modification, which is similar to that of extensively studied reversible DNA and histone modifications.11 Metabolism has an important role in influencing epigenetics and cell fate decisions.3,5,12,13,14,15 However, certain aspects of metabolic enzymes and metabolites, as well as their regulatory mechanisms in RNA epigenetics, remain unclear.Methionine metabolism, together with the availability of related environmental nutrients, plays a crucial role in RNA epigenetics, thereby regulating a variety of cellular processes.5,16,17 Methionine is converted into SAM in an MTase-catalyzed reaction that yields methylated substrates with roles in histone, DNA, and RNA methylation.5,18,19 We previously found that SAM, a product of methionine metabolism, facilitates m6A methylation in tumor cells.13 Although SAM supplementation can rescue histone methylation caused by methionine deficiency, it is unable to fully restore the downregulation of mRNA m6A (Supplementary information, Fig. S1a–c). Methionine metabolism and its key metabolic genes have been implicated in cancer development and metastasis through epigenetic changes related to methylation,5,13,16 highlighting the potential of targeting methionine metabolism for cancer therapy. However, it remains to be determined whether SAM-independent mechanisms play a role in the regulation of mRNA m6A modification by methionine metabolic genes and thus affect tumor biology.In addition to resulting from concomitant aberrations in signaling pathways, alterations in metabolites can also be key factors influencing physiological and pathological processes.1,12,20 Adenosine (ADO) is a metabolite of the methionine cycle and an endogenous nucleoside that functions as both an intermediate metabolite and an extracellular signaling molecule.21 It is primarily produced intracellularly through the degradation of AMP or the reversible hydrolysis of S-adenosylhomocysteine (SAH) catalyzed by the methionine metabolism enzyme adenosylhomocysteinase (AHCY). ADO is then transported extracellularly by bidirectional nucleoside transporters or released through nucleotide metabolic pathways into the extracellular environment. ADO signaling controls several physiological and pathological processes,21,22,23 such as sleep–wake regulation, vasodilation, epilepsy, chronic inflammation, fibrosis, diabetes, and cancer development, via the ADO receptors. However, the importance of intracellular ADO metabolism and receptor-independent ADO signaling mechanisms remains unclear.21Here, we demonstrate that the methionine metabolism enzyme AHCY specifically regulates mRNA methylation through a SAM-independent pathway by complexing with ADO. Mechanistically, the AHCY–ADO complex significantly enhances the direct physical interaction of AHCY with FTO by increasing AHCY dimerization. This interaction inhibits the binding of FTO to specific RNA molecules containing m6A motifs, which preferentially bind to the FTO Q86 site, thus increasing m6A levels and upregulating lipogenesis genes such as ACACA and SCD1, ultimately promoting fatty acid biosynthesis, tumor cell proliferation and tumorigenesis.ResultsAHCY increases mRNA m6A methylationTo thoroughly evaluate methionine metabolism and its associated metabolic pathways, we used an established metabolic library to perform CRISPR screening,24 focusing on the effects of m6A modification. Given the complexity of RNA structures, we selected two fluorescent reporter plasmids containing a GFP circular RNA with a GGACU consensus sequence: one with the m6A motif in a linear context (SSm6A)25,26 and the other with the motif in a hairpin structure (RSVm6A).25,26 The GFP pre-mRNA transcript in both reporter systems is assembled by backsplicing, generating a circular RNA that connects two GFP exon fragments (Fig. 1a). m6A methylation of the GGACU motif in the circular RNA can drive translation initiation of the GFP transcript, resulting in a GFP fluorescence signal. The RNA m6A modification is dynamically and reversibly regulated by the MTases METTL3 and METTL14 and the demethylases FTO and ALKBH5 (Supplementary information, Fig. S1d–i). Consequently, the GFP signals from both the SSm6A and RSVm6A reporters can be used as a readout of m6A methylation.Fig. 1: AHCY specifically increases mRNA m6A methylation.a Schematic diagram of two circular RNA (circRNA) translational reporters containing a GGACU motif that can be backspliced to generate circRNAs that drive GFP translation. One reporter had a linear structure (SSm6A), and the other had a hairpin structure (RSVm6A). b Overview of the CRISPR screen performed under conditions in which the medium was supplemented with 50 μM S-adenosylmethionine (SAM). c Positive regulators of m6A identified in the screen using the SSm6A and RSVm6A reporters. The displayed genes are associated with significantly enriched pathways, and AHCY is highlighted in red. d Schematic diagram of the methionine cycle. e Liquid chromatography-mass spectrometry (LC-MS/MS) quantification of the mRNA m6A/A ratio in WT and AHCY KO clones of HEK293T and SW480 cells (left). Immunoblot analysis of AHCY and β-actin (Actin) in HEK293T and SW480 cells upon KO of AHCY (right). f LC-MS/MS quantification of the mRNA m6A/A ratio in the indicated cells expressing vector or AHCY shRNAs. g Immunoblot analysis of AHCY, MAT2A, and Actin in HEK293T and HCT116 cells with AHCY or MAT2A knockdown. h–k SAM content (h) and LC-MS/MS quantification of the mRNA m6A/A ratio (i) in the indicated cells expressing vector, AHCY shRNAs, or MAT2A shRNAs. Scatterplots showing positive correlations between RNA methylation and AHCY mRNA expression (j) and between RNA methylation and the expression of genes related to one-carbon metabolism (k) in tissues from colorectal cancer (GSE190388) and ovarian cancer patients (GSE119168). Pearson’s correlation test. Data are presented as mean ± S.D. (n = 3, unless otherwise specified). Two-tailed unpaired Student’s t-test (h, i). One-way ANOVA with least significant difference t-test (LSD-t) (e, f). **P