Background

m⁶A is the most abundant internal modification in mRNA and long non-coding RNA, among more than 150 known types of post-transcriptional RNA modifications [1–3]. m⁶A was first reported in 1974 [4,5], but the field remained relatively inactive until two key discoveries. First, in 2011, FTO (fat mass and obesity-associated protein) was identified as the first m⁶A demethylase, demonstrating the reversibility of this modification [6]. Interestingly, the m⁶A methylase METTL3 had been identified earlier, in 1994 [7]. Second, in 2012, two groups independently reported genome-wide mapping approaches, MeRIP-seq and m⁶A-seq, using an m⁶A-specific antibody in combination with next-generation sequencing [8,9]. Identification of additional relevant proteins soon established the principles governing m⁶A regulation and function. The core m⁶A methylase complex contains at least seven subunits, including METTL3 (catalytic subunit) and METTL14 (structural support essential for the METTL3 activity). m⁶A is typically found at the consensus RRACH sequence, often near stop codons and within 3′ UTR of mRNAs. The m⁶A level is dynamically regulated by the METTL3 complex (“writer”) and two demethylases (“erasers”), FTO and ALKBH5, during development and in response to environmental changes. m⁶A is recognized by “reader” proteins, such as the YTH domain family (YTHDC1–2, YTHDF1–3) and insulin-like growth factor 2-binding proteins (IGF2BPs), which influence RNA stability, splicing, nuclear transport, localization, translation, and RNA–protein interactions. Reflecting this versatility, m⁶A participates in diverse cellular processes, developmental events, and disease pathogenesis.

The following workflow represents a standard approach for studying the role of m⁶A in a specific mRNA within a given biological context. First, Mettl3 or Mettl14 is depleted by knockout (KO) or knockdown (KD) in cells or animals to remove m⁶A genome-wide, and the resulting phenotypes are described. Second, the genomic distribution of m⁶A is mapped with m⁶A-seq, which includes RNA-seq as the input. Third, mRNAs dysregulated at both the RNA and m⁶A levels following KO or KD are identified. Fourth, putative causal genes are determined by combining this list with literature evidence. The final and most critical step is proving a causal link between loss of m⁶A and the observed phenotype; however, this has been challenging. Since KO and KD effects may reflect the combined disruption of many genes, identifying a small subset sufficient to explain the phenotype requires site-specific demethylation of m⁶A. This, in turn, demands prior mapping at the single-nucleotide resolution, which m⁶A-seq cannot provide due to its use of RNA fragments up to ~200 bases. Although single-nucleotide mapping and site-specific demethylation techniques have been reported, the field is still developing (see references in [10]). A more widely used alternative, described here, is to transfect a luciferase reporter plasmid containing the putative m⁶A target sequence, with and without RRACH motif mutations, and measure luciferase activity as a surrogate for protein levels. The same reporter can also be used to examine how m⁶A affects RNA stability. While this ectopic expression approach is less direct than endogenous site-specific demethylation, it remains a practical and validated method, as demonstrated by several published examples listed at the end of this article. This protocol focuses on the final step, starting from the mutations of the adenine in the third nucleotide of each RRACH motif, assuming that m⁶A mapping data are already available. The RRACH motifs in the Runx2 gene are used as examples. The mRNA and protein of Runx2 are downregulated in the mouse limb bud cells of Mettl14 conditional KO (cKO) mice at embryonic day 12.5, accompanied by reduced m⁶A in the 3′ UTR [11]. For details on m⁶A-seq, readers are referred to comprehensive published protocols [12–14].