The m6A mark is deposited on RNA by a conserved writer complex. In plants, this complex comprises the methyltransferases MTA and MTB together with FIP37, VIRILIZER-like proteins, and HAKAI, counterparts of the METTL3-METTL14-WTAP machinery familiar from animal cells. The modification preferentially lands on adenosines within the DRACH sequence motif, clustering near stop codons and in 3-prime untranslated regions. Demethylase enzymes known as erasers, including ALKBH9B and ALKBH10B in plants, can strip the mark away, while YTH-domain reader proteins bind methylated transcripts and determine their fate, whether stabilization, degradation, translation, or sequestration. Because the mark is reversible and acts post-transcriptionally, it provides a rapid regulatory layer that operates without altering the underlying genetic sequence. Disrupting core writers such as MTA or FIP37 is lethal to plant embryos, underscoring how deeply this system is woven into normal development, from meristem maintenance to flowering time and cold-stress adaptation.

Whether plant viral RNAs actually carry m6A was unresolved for decades, but since 2017 a series of studies has delivered converging proof. The first biochemical evidence came from Alfalfa mosaic virus, whose RNAs were enriched by immunoprecipitation with anti-m6A antibodies. Subsequent work mapped high-confidence methylation peaks on Wheat yellow mosaic virus, Plum pox virus, Potato virus Y, and Pepino mosaic virus, the latter detected even inside purified viral particles. More recently, antibody-independent nanopore direct RNA sequencing has confirmed m6A on Cucumber mosaic virus genomes, with sites overlapping those found by antibody-based methods, and liquid chromatography coupled to tandem mass spectrometry has verified the modification chemically. In Sugarcane mosaic virus, a potyvirus of maize, researchers pinpointed a single modified adenosine at position 6556, deposited by the maize writer ZmMTA, and showed that a synonymous mutation at that site significantly enhanced infection. Bamboo mosaic virus yielded 122 candidate sites by direct RNA sequencing. Together these techniques leave little doubt that m6A is an authentic feature of diverse plant viral RNAs.

How cytoplasmic plant viruses gain access to a methyltransferase complex that normally resides in the nucleus has become one of the most intriguing mechanistic questions in the field. Unlike nuclear-replicating animal viruses such as influenza or HIV, most plant viruses replicate exclusively in the cytoplasm, seemingly far from the writer machinery. The answer, it turns out, is that viruses actively drag the machinery to themselves. In wheat, the WYMV NIb protein physically interacts with the writer TaMTB and triggers its translocation to cytoplasmic replication sites, where the enzyme methylates viral RNA at a defined position. In Arabidopsis infected with Cucumber mosaic virus, the viral coat protein binds the core writer MTB directly and promotes nuclear-to-cytoplasmic relocation of the entire complex. During Sugarcane mosaic virus infection, ZmMTA co-localizes with viral genomic RNA in cytoplasmic aggregates, though the exact relocation mechanism remains to be worked out. Notably, not every virus succeeds in this recruitment; recent antibody-independent analyses of Chikungunya and Dengue viruses found no detectable m6A, a reminder that epitranscriptomic engagement is selective rather than universal.

Once deposited, the mark can cut in either direction. In several systems m6A behaves as a classic antiviral flag. Hypermethylated Cucumber mosaic virus RNAs are recognized by the reader protein ECT8, which directs them into decay pathways, likely through processing bodies. In Pepino mosaic virus infection of Nicotiana benthamiana, overexpressing the writers MTA or HAKAI boosts viral RNA methylation and restricts infection, while the methylated transcripts are bound by the YTH-domain readers NbECT2A, 2B, and 2C, which recruit nonsense-mediated decay factors such as UPF3 and SMG7 to destroy them. Silencing NbECT2B or the NMD components increases viral RNA stability and susceptibility. In maize, the reader ZmECT23 recognizes methylated Sugarcane mosaic virus RNA and recruits the CCR4-NOT deadenylation complex to destabilize it, and in Nicotiana benthamiana the writer NbMTA targets Potato virus Y genomes for degradation.

Yet some viruses have turned the same mark to their advantage. When the WYMV NIb protein lures TaMTB to replication complexes, the resulting site-specific methylation stabilizes the viral RNA and prevents its degradation, and mutating the methylation site reduces viral stability and pathogenicity. A parallel strategy appears in the plant rhabdovirus Barley yellow striate mosaic virus, whose P6 messenger RNA is hypermethylated; loss of the mark reduces P6 stability and infectivity. Intriguingly, barley fights back with its own eraser, HvALKBH1B, which binds P6 mRNA into cytoplasmic condensates through liquid-liquid phase separation, a process requiring the protein’s intrinsically disordered region, thereby stripping the viral advantage. The lesson is that m6A is not inherently antiviral or proviral; it is a regulatory currency that whichever side controls it can spend.

The mark also shapes viral systemic movement, not merely local accumulation. Alfalfa mosaic virus depends on the host demethylase ALKBH9B removing m6A from its RNAs for efficient long-distance spread, an interaction mediated by the viral coat protein. Excessive methylation reduces phloem transport and restricts movement through the plant, so a balanced level of the mark is essential for the virus to travel. Consistent with this, infection by Plum pox virus and Potato virus Y lowers global m6A levels in Nicotiana benthamiana, and knocking down ALKBH9 homologs diminishes accumulation of both viruses. On the defensive side, the YTHDF-family readers ECT2, ECT3, and ECT5 bind Alfalfa mosaic virus RNAs in an m6A-dependent manner and act as restriction factors suppressing viral accumulation. Strikingly, this regulation is specific to ALKBH9B, since disrupting the related homologs ALKBH9A or ALKBH9C leaves infection untouched, revealing a fine-grained specificity in how plants deploy their epitranscriptomic arsenal.

Viruses, unsurprisingly, have evolved countermeasures. The 2b protein of Cucumber mosaic virus, already infamous as a suppressor of RNA silencing, competitively binds the writer components HAKAI and MTB, impairing the integrity of the writer complex in both nucleus and cytoplasm. This suppresses methylation not only on viral RNAs but across the host transcriptome; plants lacking m6A are more susceptible to a 2b-deficient mutant but not to wild-type virus, and simply expressing 2b in transgenic plants lowers global m6A levels. Pepino mosaic virus takes a different route: its RNA-dependent RNA polymerase interacts with the tomato writer SlHAKAI and the autophagy protein SlBeclin1, forming cytoplasmic granules that route SlHAKAI to autophagic destruction, thereby stripping methylation from viral RNA. In maize, the Sugarcane mosaic virus protease NIa-Pro hijacks the initiation factor ZmeIF4A3 into viral replication complexes, sterically blocking ZmMTA-mediated methylation and sparing viral RNA from ZmECT23-directed decay. Each case illustrates a mutually antagonistic arms race fought at the level of RNA chemistry.

Viral infection also rewrites the host’s own methylation landscape in ways that reshape immunity. Tobacco mosaic virus lowers m6A levels in tobacco, coinciding with upregulation of the eraser NbALKBH5, while Rice stripe virus and Rice black-streaked dwarf virus raise methylation levels in rice, enriching marks on transcripts involved in RNA silencing and hormone defense, including the antiviral genes OsAGO18 and OsSLRL1. In watermelon facing Cucumber green mottle mosaic virus, susceptible plants show early hypermethylation of stress-related transcripts, whereas resistant cultivars exhibit later hypomethylation that preserves defense gene expression. In Arabidopsis, CMV-induced hypomethylation upregulates the salicylic acid regulators NPR3 and CBP60a, and Potato virus Y triggers a transient rise in methylation at five to ten days post-inoculation that falls by day fourteen, accompanied by activation of a transcription factor that boosts NbMTA expression and accelerates viral RNA degradation. These virus-specific, time-dependent shifts integrate m6A into the broader networks of hormone signaling and stress adaptation.

The review’s authors argue that these findings position m6A as a promising lever for crop protection, whether through overexpressing writer components, modulating reader activity, or breeding viral methylation sites out of susceptible hosts. Significant questions remain, however. Recent work warns that false-positive signals can arise in both antibody-based and some antibody-independent mapping methods, making rigorous negative controls essential, and emerging single-molecule technologies are expected to sharpen the resolution of modification detection. Researchers still do not fully understand what sequence or structural features direct methylation to particular viral transcripts, how the temporal choreography of methylation unfolds during infection, or how reliably these mechanisms can be engineered into durable field resistance. What is already clear is that the epitranscriptome has moved from the margins of plant virology to its center, and that the contest over a single methyl group on adenosine may help decide the outcome of infections that devastate staple crops worldwide.

Subject of Research: The role of m6A RNA modification in plant-virus interactions

Article Title: The expanding role of m6A RNA modification in plant-virus dynamics: friend, foe, or both?

Article References: Liu, J.-H., Yu, H., & Duan, C.-G. (2026). The expanding role of m6A RNA modification in plant-virus dynamics: friend, foe, or both?. Advanced Biotechnology, 4(1), Article 7. https://doi.org/10.1007/s44307-026-00100-3

Image Credits: AI Generated

DOI: 10.1007/s44307-026-00100-3

Keywords: m6A, RNA modification, epitranscriptome, plant viruses, plant immunity, methyltransferase, YTH-domain proteins, RNA decay, viral countermeasures, crop resistance, Cucumber mosaic virus, RNA methylation

The study, conducted by Hidenori Tani of Yokohama University of Pharmacy, takes an unusually careful comparative approach. Rather than comparing mRNA and lncRNA stability data from different experiments, different laboratories or different measurement platforms, Tani assembled half-life measurements derived from the same metabolic labeling and sequencing assays, allowing messenger RNAs and long non-coding RNAs to be evaluated on a genuinely level playing field. This design matters, because the field has been plagued by comparisons in which differences in assay chemistry, cellular context or transcript abundance could masquerade as biological differences in decay regulation. By holding the measurement method constant, the analysis isolates the one variable of interest: whether the sequence of a transcript encodes its own stability.

The findings for messenger RNAs confirm the power of codon-mediated regulation. When Tani re-estimated codon stabilization coefficients, quantitative measures of how strongly each codon is associated with transcript longevity, inside every cross-validation fold to eliminate information leakage between training and test data, the resulting model predicted mRNA half-life with a cross-validated coefficient of determination of 0.174 and a Spearman rank correlation of 0.44. Critically, the codon stabilization coefficient on its own, yielding an R-squared of 0.084, outperformed a six-member set of translation-independent features, which managed only 0.008. That ordering, with codon optimality beating a battery of sequence characteristics such as GC content, transcript length and known destabilizing elements, held consistently across three additional datasets, two distinct measurement technologies and two different species.

The picture for long non-coding RNAs could hardly be more different. In a carefully matched set of 364 lncRNAs measured in the same HeLa cell assay as the messenger RNAs, sequence-based models of half-life performed no better than chance, yielding an R-squared of minus 0.050, a value meaning the predictions were actually worse than simply guessing the average half-life for every transcript. The result was not a quirk of a small or unrepresentative sample. When Tani turned to the largest published lncRNA stability dataset, encompassing 33,285 transcripts, k-mer composition features achieved an R-squared of essentially zero, at 0.0007. Even the cross-validated rank correlation reported in that dataset’s original study, 0.091, is vanishingly small in absolute terms.

Here the study makes a subtle but important statistical argument. A Spearman correlation of 0.091 in a dataset of more than thirty thousand transcripts does clear the permutation null, meaning it is statistically detectable rather than pure noise. But detectability is not the same as explanatory power, and the contrast between the two RNA classes is one of magnitude, not of presence or absence. Messenger RNA half-life is meaningfully sequence-encoded; long non-coding RNA half-life, by every measure Tani applied, is not. The distinction has practical consequences for anyone attempting to model RNA decay: a signal that is real but negligible cannot support the kind of predictive machinery that works for coding transcripts.

The analysis also confronts one of the most fashionable tools in modern genomics: pretrained genomic language models. These deep neural networks, trained on enormous corpora of genomic sequence, have been touted as universal feature extractors capable of discovering biological signals without explicit programming. Tani tested models including HyenaDNA, which can process nearly complete transcripts at single-nucleotide resolution, and DNABERT-2. Even with one model covering 99.6 percent of transcripts in their entirety, the language models left messenger RNA prediction at an R-squared of just 0.042, and lncRNA prediction at values between 0.0002 and 0.0036. The implication is sobering: whatever the language models learned about genomic sequence, they did not uncover a hidden stability code in non-coding transcripts that simpler approaches had missed.

To rule out confounding, Tani stratified transcripts by predicted coding potential, testing whether a subset of lncRNAs with translated open reading frames might behave like messenger RNAs after all. They did not. Nor did differences in transcript length, GC content or sample size explain the gap: when lncRNAs and mRNAs were matched on all three characteristics and analyzed with an identical feature set, the messenger RNAs yielded an R-squared of 0.033 while the lncRNAs yielded 0.0002. Perhaps most convincingly, a signal-injection experiment established that the analytical pipeline could reliably detect effects explaining as little as 0.5 percent of the variance in half-life, a sensitivity threshold comfortably above every single lncRNA result reported in the study. If a sequence-based stability signal existed in these transcripts at even a modest level, the method would have found it.

The biological interpretation is as interesting as the statistical one. The mechanistic chain linking codon optimality to decay runs through translation itself: slow-moving ribosomes on unoptimized codons recruit decay factors, and the DEAD-box helicase Dhh1p in yeast monitors codon optimality directly to couple translation to destruction. Long non-coding RNAs, which are largely untranslated, simply cannot participate in this feedback loop. Their lifespans are instead governed by other forces, including RNA-binding proteins, nuclear retention mechanisms, structural elements and, in many organisms, the nonsense-mediated decay pathway, which can eliminate lncRNAs carrying premature stop codons. A growing literature documents hundreds of short-lived non-coding transcripts in mammalian cells and shows that nonsense-mediated decay restricts lncRNA expression even in RNAi-capable budding yeasts, underscoring that non-coding transcript turnover has its own logic, written in a different biochemical language.

The study’s broader lesson extends beyond RNA biology into the methodology of machine-learning-driven science. By re-estimating features within each cross-validation fold, Tani guarded against data leakage, the insidious practice pattern that has inflated reported performance across computational biology, and the work aligns with recent guidance on reproducibility in machine-learning applications. The bottom-line recommendation is blunt: mRNA-derived decay models should not be transferred to long non-coding RNAs without re-validation. For researchers designing RNA therapeutics, engineering synthetic transcripts or interpreting lncRNA dysregulation in disease, the message is clear. The stability of the non-coding transcriptome cannot be read from codon usage tables, and understanding how these molecules are scheduled for destruction will require looking elsewhere, at the proteins and structures that shepherd them through the cell.

Subject of Research: Codon optimality as a predictor of mRNA half-life and its failure to predict long non-coding RNA stability

Article Title: Codon optimality predicts mRNA half-life but does not transfer to lncRNAs

Article References: Tani, H. (2026). Codon optimality predicts mRNA half-life but does not transfer to lncRNAs. Molecular Genetics and Genomics, 301(1), Article 196. https://doi.org/10.1007/s00438-026-02520-1

Image Credits: AI Generated

DOI: 10.1007/s00438-026-02520-1

Keywords: codon optimality, mRNA stability, long non-coding RNA, RNA decay, RNA half-life, machine learning, cross-validation, genomic language models, nonsense-mediated decay, transcriptome, RNA-binding proteins, gene regulation