YTHDF2 regulates self non-coding RNA metabolism to control inflammation and tumorigenesis

Seungwon Yang, Yan‐Hong Cui, Haixia Li, Jiangbo Wei, Gayoung Park, Ming Sun, Michelle Verghese, Emma Wilkinson, Tae Gyu Nam, Linnea L. Lungstrom, Xiao‐Long Cui, Tae Young Ryu, Jing Chen, Marc Bissonnette, Chuan He, Yu‐Ying HeNature Communications 2025 · DOI

This research paper, published in Nature Communications, reveals a novel molecular "brake" that prevents our own RNA from triggering chronic inflammation and skin cancer. The study demonstrates that the protein YTHDF2 acts as a guardian by recognizing and degrading a specific non-coding RNA (U6 snRNA), preventing it from accidentally activating the immune system's alarm bells.

TL;DR

Researchers have discovered that the $m^6A$ reader protein YTHDF2 prevents "self-inflicted" inflammation by degrading U6 snRNA before it can bind to the immune receptor TLR3. When UVB light (sunburn) or genetic mutations deplete YTHDF2, U6 accumulates in endosomes, triggering an inflammatory cascade that promotes skin tumorigenesis.

What Problem Does This Paper Address?

Our cells are filled with "self" RNA, which looks dangerously similar to the "non-self" RNA found in viruses. To avoid constant, damaging inflammation, the body must distinguish between the two. While we know a lot about how $m^6A$ ($N^6$-methyladenosine) methylation regulates messenger RNA (mRNA), its role in "self" non-coding RNAs (ncRNAs) is less clear.

The authors investigate how the cell prevents U6 snRNA—a ubiquitous molecule usually involved in gene splicing—from escaping its normal duties and acting as a "damage-associated molecular pattern" (DAMP) that triggers the innate immune receptor TLR3. Understanding this mechanism is critical for explaining why environmental stressors like UVB light lead to skin inflammation and cancer.

Methods

The authors employed a multi-disciplinary approach combining molecular biology, biochemistry, and in vivo mouse models:

  • $m^6A$ Mapping: Used UHPLC-MS/MS and $m^6A$-seq to quantify and locate methylation on U6 snRNA.
  • Endosomal Proteomics/Transcriptomics: Developed "Endo-IP" (endosome immunoprecipitation) to isolate and analyze the contents of endosomes, specifically looking for RNA-protein interactions.
  • Genetic Models: Created skin-specific conditional knockout (cKO) mice lacking Ythdf2 to observe the effects of its absence on UVB-induced inflammation and tumor formation.
  • Protein Biochemistry: Used mass spectrometry to identify YTHDF2-interacting proteins and mapped the specific binding domains between U6 and TLR3 using deletion mutants.

Datasets

  • New Data: The study generated RNA-seq and $m^6A$-seq datasets from HaCaT (human keratinocyte) cells, deposited under GEO accession GSE145925. Mass spectrometry data is available via PRIDE (PXD059422).
  • Public Data: The authors analyzed existing datasets for autoimmune diseases, including Systemic Lupus Erythematosus (SLE) and Type I Diabetes (e.g., GSE110169, GSE50772), to correlate YTHDF2 levels with human disease states.

Key Insights and Ideas

  • U6 as an Immune Trigger: U6 snRNA, typically a nuclear splicing factor, can act as a potent ligand for TLR3 when it accumulates in endosomes.
  • The $m^6A$ "Stability Tag": Approximately 16.5% of U6 snRNA is $m^6A$-methylated. This modification isn't for function, but for disposal. YTHDF2 recognizes this tag and recruits the CNOT1 decay machinery to keep U6 levels low.
  • Distinct Binding Mechanism: Unlike the synthetic TLR3 agonist poly I:C, which binds to the LRR20 domain of TLR3, U6 snRNA binds to the LRR21 domain. This suggests a unique structural interface for "self" RNA recognition.
  • UVB's Double Whammy: UVB exposure doesn't just damage DNA; it actively inhibits the "policeman" YTHDF2. It does this by inducing YTHDF2 dephosphorylation (at Serine 39) and triggering its degradation via p62-mediated autophagy. This allows U6 to flood the system and drive inflammation.

Key Results

  • Tumorigenesis: Mice lacking skin-specific Ythdf2 developed significantly more and larger skin tumors following chronic UVB exposure compared to wild-type mice.
  • Inflammatory Signaling: Loss of YTHDF2 led to a massive increase in pro-inflammatory cytokines (TNF-$\alpha$, IL-6, COX-2) in both human cells and mouse skin.
  • Molecular Half-life: In control cells, U6 snRNA has a half-life of ~4 hours. When YTHDF2 is knocked down, U6 becomes significantly more stable, leading to its accumulation.
  • Clinical Correlation: YTHDF2 expression was found to be significantly lower in human skin squamous cell carcinoma (SCC) samples and in the blood of patients with SLE and Type I diabetes, suggesting this pathway is relevant to human pathology.

Limitations and Future Work

  • Specificity of Inhibitors: The authors noted that Actinomycin D, used for stability assays, might affect multiple RNA polymerases, potentially complicating half-life measurements.
  • Other Readers: While YTHDF2 is the primary focus, the role of other $m^6A$ readers (like YTHDF1 or YTHDF3) in this specific endosomal pathway remains to be fully explored.
  • Therapeutic Potential: The study identifies the YTHDF2-U6-TLR3 axis as a target. Future work could investigate whether stabilizing YTHDF2 or preventing U6-TLR3 binding could serve as a treatment for chronic inflammatory skin conditions or as a preventative measure against skin cancer.

Summary updated

Abstract

The role of m<sup>6</sup>A RNA methylation of self non-coding RNA remains poorly understood. Here we show that m<sup>6</sup>A-methylated self U6 snRNA is recognized by YTHDF2 to reduce its stability and prevent its binding to Toll-like receptor 3 (TLR3), leading to decreased inflammatory responses in human and mouse cells and mouse models. At the molecular level, endosomal U6 snRNA binds to the LRR21 domain in TLR3, independent of m<sup>6</sup>A methylation, to activate inflammatory gene expression, a mechanism that is distinct from that of the best known synthetic TLR3 agonist poly I:C. Both U6 snRNA and YTHDF2 are localized to endosomes via the transmembrane protein SIDT2, where YTHDF2 functions to prevent the U6-TLR3 interaction. We further show that UVB exposure inhibits YTHDF2 by inducing its dephosphorylation and autophagic protein degradation in human keratinocytes and mouse skin. Skin-specific deletion of Ythdf2 in mice enhanced the UVB-induced skin inflammatory response and promoted tumor initiation. Taken together, our findings demonstrate that YTHDF2 plays a crucial role in controlling inflammation by inhibiting m<sup>6</sup>A U6-mediated TLR3 activation, suggesting that YTHDF2 and m<sup>6</sup>A U6 are potential therapeutic targets for preventing and treating inflammation and tumorigenesis.
BibTeX
@inproceedings{yang2025ythdf,
  author = {Seungwon Yang and Yan‐Hong Cui and Haixia Li and Jiangbo Wei and Gayoung Park and Ming Sun and Michelle Verghese and Emma Wilkinson and Tae Gyu Nam and Linnea L. Lungstrom and Xiao‐Long Cui and Tae Young Ryu and Jing Chen and Marc Bissonnette and Chuan He and Yu‐Ying He},
  title = {{YTHDF2 regulates self non-coding RNA metabolism to control inflammation and tumorigenesis}},
  booktitle = {Nature Communications},
  year = {2025},
  doi = {10.1038/s41467-025-64898-7},
}
Figure