Cytoplasmic RNA-binding protein of the YTHDF family and one of the three paralogous readers of N6-methyladenosine (m6A) in messenger RNA. Its C-terminal YTH domain engages m6A through an aromatic cage, while its long N-terminal low-complexity region drives assembly of multivalent ribonucleoprotein condensates. YTHDF3 binds essentially the same pool of m6A-modified transcripts as YTHDF1 and YTHDF2, in proportion to the number of methylated sites, and the three paralogs act largely redundantly to destabilise those transcripts by recruiting deadenylation and decay machinery, so that the effect on mRNA stability and on cell differentiation becomes fully apparent only when all three are depleted together. Together with the bound mRNA, YTHDF3 partitions into phase-separated compartments including P-bodies, cytoplasmic stress granules and neuronal RNA granules, and it is required for efficient stress granule formation. Whether YTHDF3 also actively promotes translation is disputed: it was originally reported to bind 40S and 60S ribosomal proteins and to enhance translation initiation in synergy with YTHDF1, but a later study using simultaneous depletion of all three paralogs found no translation-promoting effect and attributed the shared function entirely to mRNA degradation. Additional proposed activities - reading N1-methyladenosine in mRNA, and reading N6-methyladenine in genomic DNA - rest on antibody-based mapping methods whose specificity has been independently challenged.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0003729 mRNA binding | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic assertion of mRNA binding across the YTH family, seeded from a broad set of plant, fly, fission yeast, budding yeast and zebrafish members. Reason: Correct and consistent with the direct human data; mRNA is the substrate class on which the m6A reader activity is exercised. |
| GO:0005737 cytoplasm | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic assertion that this YTHDF clade acts in the cytoplasm, in contrast to the nuclear YTHDC readers. Reason: Correct. The cytoplasmic restriction is the defining compartmental feature separating YTHDF1/2/3 from YTHDC1, and it is confirmed by direct human evidence. |
| GO:1990247 N6-methyladenosine-containing RNA reader activity | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic assertion of m6A reader activity, the defining function of the YTH domain. Reason: Core molecular function, and the phylogenetic call is unusually secure because experimental m6A-binding data exist for members across fly, mouse, budding yeast and zebrafish as well as human. The appearance of Q7Z739 itself in the WITH/FROM list is the expected marker that this gene's own experimental annotation helped seed the ancestral node, not circularity. Supporting Evidence: PMID:28106072 Here, we report that YTHDF3 promotes protein synthesis in synergy with YTHDF1, and affects methylated mRNA decay mediated through YTHDF2. |
| GO:0061157 mRNA destabilization | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic assertion that YTHDF-clade readers destabilise their bound transcripts. Reason: Core biological process, and the one outcome of m6A reading on which both sides of the translation dispute agree. Supporting Evidence: PMID:32492408 Instead, the DF paralogs act redundantly to mediate mRNA degradation and cellular differentiation. |
| GO:0000932 P-body | IEA GO_REF:0000120 | ACCEPT | Summary: Automated subcellular-location assignment to processing bodies. Reason: Correct; directly observed, and mechanistically expected because m6A-mRNA-YTHDF condensates partition into P-bodies. |
| GO:0003723 RNA binding | IEA GO_REF:0000002 | KEEP AS NON CORE | Summary: InterPro YTH-domain signature (IPR007275) mapped to generic RNA binding. Reason: True but uninformative relative to what this gene already carries. GO:1990247 N6-methyladenosine-containing RNA reader activity specifies both the ligand and the modification and is the term that should represent this protein's molecular function. Retained as a correct parent rather than modified, because one of the other annotations to this same term (PMID:32194978) concerns m1A rather than m6A binding and so is not simply subsumed by the m6A reader term. |
| GO:0005829 cytosol | IEA GO_REF:0000044 | ACCEPT | Summary: Automated transfer from the UniProt subcellular-location vocabulary. Reason: Correct primary location. |
| GO:0010494 cytoplasmic stress granule | IEA GO_REF:0000120 | ACCEPT | Summary: Automated subcellular-location assignment to stress granules. Reason: Correct; directly observed by super-resolution imaging under stress. |
| GO:0034063 cytoplasmic stress granule assembly | IEA GO_REF:0000117 | KEEP AS NON CORE | Summary: ARBA machine-learning assignment of a role in stress granule assembly. Reason: Genuine and directly demonstrated, but stress-conditional and downstream of the m6A reader activity - the granules form because polymethylated mRNAs act as multivalent scaffolds for YTHDF low-complexity domains. A consequence of the core function rather than a separate one. |
| GO:1990247 N6-methyladenosine-containing RNA reader activity | IEA GO_REF:0000117 | ACCEPT | Summary: ARBA machine-learning assignment of the m6A reader activity. Reason: Agrees with abundant direct experimental evidence; core molecular function. |
| GO:0005515 protein binding | IPI PMID:22190034 Global landscape of HIV-human protein complexes. | REMOVE | Summary: IntAct-curated interaction with the HIV-1 capsid protein (UniProtKB:P04591-PRO_0000038596) from a global HIV-human protein complex survey. Reason: No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Bare protein binding carries no functional information. Any antiviral role of YTHDF3 is better expressed through its m6A reader activity and the type I interferon signalling annotation this gene already carries. Removal does not imply the reported interaction is false. |
| GO:0005515 protein binding | IPI PMID:28514442 Architecture of the human interactome defines protein commun... | REMOVE | Summary: High-throughput interactome detection of the YTHDF3-YTHDF1 (UniProtKB:Q9BYJ9) interaction. Reason: Bare protein binding carries no functional information. The YTHDF1-YTHDF3 association is real and recurrent across interactome datasets, but its functional meaning is captured by the shared m6A reader activity and by the granule and translation annotations, not by this term. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | REMOVE | Summary: Binary interactome detection of an interaction with FAF1 (UniProtKB:Q9UNN5). Reason: Bare protein binding carries no functional information and no downstream function has been attached to this particular interaction. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005515 protein binding | IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling... | REMOVE | Summary: Proteome-scale interactome detection of the YTHDF3-YTHDF1 (UniProtKB:Q9BYJ9) interaction. Reason: Bare protein binding carries no functional information; the same paralog interaction is already recorded from several other datasets. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005515 protein binding | IPI PMID:40205054 Multimodal cell maps as a foundation for structural and func... | REMOVE | Summary: Multimodal cell-map interactome detection of the YTHDF3-YTHDF1 (UniProtKB:Q9BYJ9) interaction. Reason: Bare protein binding carries no functional information; redundant with the other paralog-interaction rows. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005737 cytoplasm | IEA GO_REF:0000107 | ACCEPT | Summary: Ensembl Compara transfer of the cytoplasmic localisation from mouse Ythdf3 (UniProtKB:Q8BYK6). Reason: Correct and independently established for the human protein. |
| GO:0043488 regulation of mRNA stability | IEA GO_REF:0000107 | ACCEPT | Summary: Ensembl Compara transfer from mouse Ythdf3. Reason: Core biological process; matches direct human evidence from both sides of the translation dispute. |
| GO:0060339 negative regulation of type I interferon-mediated signaling pathway | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Ensembl Compara transfer of an innate-immune regulatory role from mouse Ythdf3. Reason: Plausible and consistent with m6A-dependent turnover of interferon-pathway transcripts, but a context-specific physiological output several steps downstream of the reader activity, and supported for human only by orthology and sequence similarity. Retained as non-core. |
| GO:0061157 mRNA destabilization | IEA GO_REF:0000120 | ACCEPT | Summary: Automated/orthology assignment of the mRNA destabilisation role. Reason: Core biological process, agreeing with the direct human data. |
| GO:0000932 P-body | IDA PMID:32492408 A Unified Model for the Function of YTHDF Proteins in Regula... | ACCEPT | Summary: Direct observation of YTHDF3 in processing bodies. Reason: Correct, and mechanistically tied to the degradation arm of the core function. |
| GO:0003723 RNA binding | IDA PMID:32194978 Cytoplasmic m(1)A reader YTHDF3 inhibits trophoblast invasio... | KEEP AS NON CORE | Summary: Direct demonstration that YTHDF3 binds m1A-carrying RNA, identified by mass spectrometry among nine candidate m1A readers. Reason: The binding experiment stands, but the claim that this constitutes a physiologically important reader function is weakened by the substrate's abundance. The year before, a systematic reinvestigation showed that mRNA m1A had been massively over-called because the commonly used m1A antibody cross-reacts with the 5' cap (PMID:31719534), and m1A-seq is the mapping method this study relies on. Retained as non-core generic RNA binding rather than promoted to a specific m1A reader function; the informative molecular function for this protein remains GO:1990247. Supporting Evidence: PMID:32194978 among them, YTH domain-containing protein 3 (YTHDF3), could bind directly to m1A-carrying RNA PMID:31719534 These results demonstrate that high-stoichiometry m1A sites are exceedingly rare in mRNAs and that previous mappings of m1A to 5'UTRs were the result of antibody cross-reactivity to the 5' cap. |
| GO:0005737 cytoplasm | IDA PMID:32492408 A Unified Model for the Function of YTHDF Proteins in Regula... | ACCEPT | Summary: Direct observation of cytoplasmic localisation. Reason: Correct; the cytosol (GO:0005829) annotation is the more specific form of the same observation. |
| GO:0010494 cytoplasmic stress granule | IDA PMID:32451507 m(6)A-binding YTHDF proteins promote stress granule formatio... | ACCEPT | Summary: Super-resolution imaging placing YTHDF proteins in clusters at the periphery of and between stress granule cores. Reason: Directly observed and structurally characterised. Supporting Evidence: PMID:32451507 Here, we show that m6A-modified mRNAs are enriched in SGs, and that m6A-binding YTHDF proteins are critical for SG formation. |
| GO:0034063 cytoplasmic stress granule assembly | IDA PMID:32451507 m(6)A-binding YTHDF proteins promote stress granule formatio... | KEEP AS NON CORE | Summary: Depletion of YTHDF1/3 inhibits stress granule formation and the recruitment of mRNAs into granules. Reason: Genuine and directly demonstrated, but stress-conditional and downstream of the m6A reader activity - the granules form because polymethylated mRNAs act as multivalent scaffolds for YTHDF low-complexity domains. A consequence of the core function rather than a separate one. Supporting Evidence: PMID:32451507 Depletion of YTHDF1/3 inhibits SG formation and recruitment of mRNAs to SGs. |
| GO:0043488 regulation of mRNA stability | IMP PMID:32194978 Cytoplasmic m(1)A reader YTHDF3 inhibits trophoblast invasio... | ACCEPT | Summary: YTHDF3 knockdown/overexpression changes IGF1R mRNA levels in trophoblast cells. Reason: Core biological process. The direction of effect (YTHDF3 promotes target mRNA degradation) agrees with the consensus degradation model, independently of whether the mark being read in this particular study is m1A or m6A. Supporting Evidence: PMID:32194978 YTHDF3 could promote IGF1R mRNA degradation and thus inhibit IGF1R protein expression along with its downstream matrix metallopeptidase 9 signaling pathway, consequently decreasing migration and invasion of trophoblast. |
| GO:0043488 regulation of mRNA stability | IDA PMID:32492408 A Unified Model for the Function of YTHDF Proteins in Regula... | ACCEPT | Summary: Direct demonstration that the DF paralogs act redundantly on the stability of m6A-modified mRNAs. Reason: Core biological process; this is the function both camps in the translation dispute agree on. Supporting Evidence: PMID:32492408 The ability of DF proteins to regulate stability and differentiation becomes evident only when all three DF paralogs are depleted simultaneously. |
| GO:0045948 positive regulation of translational initiation | IDA NOT PMID:32492408 A Unified Model for the Function of YTHDF Proteins in Regula... | UNDECIDED | Summary: Explicit negative assertion - with all three DF paralogs analysed together, no translation-promoting effect was detectable in HeLa cells. Reason: GOA carries both sides of this dispute on the same term: this NOT|involved_in IDA from PMID:32492408 and a positive involved_in IDA from PMID:28106076. Both are direct experimental annotations from competent laboratories, and the disagreement is substantive rather than a curation slip - it turns on whether single-paralog depletion or simultaneous triple depletion is the informative experiment, and on cell type (HeLa). This cannot be adjudicated from the cached record, and neither annotation may be removed. UNDECIDED is recorded for both rows so that the contradiction is visible rather than silently resolved. Supporting Evidence: PMID:32492408 Furthermore, we find that DF proteins do not induce translation in HeLa cells. PMID:32492408 In contrast to the prevailing model, we show that DF proteins bind the same m6A-modified mRNAs rather than different mRNAs. |
| GO:0061157 mRNA destabilization | IDA PMID:32492408 A Unified Model for the Function of YTHDF Proteins in Regula... | ACCEPT | Summary: The DF paralogs, acting redundantly, degrade m6A-modified mRNAs in proportion to the number of m6A sites. Reason: Core biological process and the central positive claim of the unified model. Supporting Evidence: PMID:32492408 Instead, the DF paralogs act redundantly to mediate mRNA degradation and cellular differentiation. |
| GO:0070925 organelle assembly | IDA PMID:31292544 m(6)A enhances the phase separation potential of mRNA. | MODIFY | Summary: YTHDF proteins undergo liquid-liquid phase separation with polymethylated mRNAs and partition into P-bodies, stress granules and neuronal RNA granules. Reason: Correct in kind but far too general - "organelle assembly" spans mitochondria, cilia and every other organelle, whereas what was shown is the assembly of specific membraneless ribonucleoprotein compartments. The two specific child terms below capture the observation precisely; GO:0034063 is already annotated to this gene from an independent study. Proposed replacements: cytoplasmic stress granule assembly P-body assembly Supporting Evidence: PMID:31292544 The resulting mRNA-YTHDF complexes then partition into different endogenous phase-separated compartments, such as P-bodies, stress granules or neuronal RNA granules. |
| GO:1901163 regulation of trophoblast cell migration | IMP PMID:32194978 Cytoplasmic m(1)A reader YTHDF3 inhibits trophoblast invasio... | KEEP AS NON CORE | Summary: YTHDF3 depletion increases, and overexpression decreases, trophoblast invasion and migration via IGF1R. Reason: A real cell-type-specific physiological output, but two steps downstream of the molecular activity (reader binding, then target mRNA turnover, then IGF1R-dependent motility) and confined to one cell type. Retained as non-core. |
| GO:1990247 N6-methyladenosine-containing RNA reader activity | IDA PMID:31292544 m(6)A enhances the phase separation potential of mRNA. | ACCEPT | Summary: m6A-dependent binding demonstrated in the context of phase separation, with condensation driven specifically by mRNAs containing multiple m6A residues. Reason: Core molecular function, here shown with an unusually direct readout of multivalent m6A recognition. |
| GO:1990247 N6-methyladenosine-containing RNA reader activity | IDA PMID:32451507 m(6)A-binding YTHDF proteins promote stress granule formatio... | ACCEPT | Summary: m6A binding demonstrated as the basis for recruitment of methylated mRNAs into stress granules; both the YTH domain and the N-terminal disordered region are required. Reason: Core molecular function. Supporting Evidence: PMID:32451507 Both the N-terminal intrinsically disordered region and the C-terminal m6A-binding YTH domain of YTHDF proteins are important for SG formation. |
| GO:1990247 N6-methyladenosine-containing RNA reader activity | IDA PMID:32492408 A Unified Model for the Function of YTHDF Proteins in Regula... | ACCEPT | Summary: YTHDF3 binds the same m6A-modified transcripts as YTHDF1 and YTHDF2, with occupancy scaling with the number of m6A sites. Reason: Core molecular function. Notably, the study that most sharply disputes YTHDF3's translation role is also among the strongest evidence for its m6A reader activity. Supporting Evidence: PMID:32492408 In contrast to the prevailing model, we show that DF proteins bind the same m6A-modified mRNAs rather than different mRNAs. |
| GO:0060339 negative regulation of type I interferon-mediated signaling pathway | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Curator-judged sequence-similarity transfer from mouse Ythdf3 (UniProtKB:Q8BYK6). Reason: Plausible and consistent with m6A-dependent turnover of interferon-pathway transcripts, but a context-specific physiological output several steps downstream of the reader activity, and supported for human only by orthology and sequence similarity. Retained as non-core. |
| GO:0005515 protein binding | IPI PMID:28106072 YTHDF3 facilitates translation and decay of N(6)-methyladeno... | REMOVE | Summary: Tandem-affinity purification of YTHDF3 complexes recovering YTHDF1 (UniProtKB:Q9BYJ9) and YTHDF2 (UniProtKB:Q9Y5A9). Reason: Bare protein binding carries no functional information. The functional content of these interactions - cooperative action of the three paralogs on shared m6A targets - is already represented by the reader activity and the mRNA-turnover annotations. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005829 cytosol | TAS PMID:28106072 YTHDF3 facilitates translation and decay of N(6)-methyladeno... | ACCEPT | Summary: Cytosolic localisation stated by the authors. Reason: Correct and the most specific of the cytoplasmic location terms carried by this gene. |
| GO:0043022 ribosome binding | IDA PMID:28106076 Cytoplasmic m(6)A reader YTHDF3 promotes mRNA translation. | KEEP AS NON CORE | Summary: Co-immunoprecipitation of YTHDF3 with 40S and 60S ribosomal proteins, used as the mechanistic basis for a translation-promoting role. Reason: The interaction data are direct, but the functional interpretation they were used to support - that YTHDF3 promotes translation initiation - is contested by PMID:32492408, which found no translation-promoting effect when all three DF paralogs were depleted together. Since an abundant cytoplasmic mRNA-binding protein can co-purify with ribosomes without regulating them, this is retained as a genuine observation but demoted from core pending resolution of the translation question. Supporting Evidence: PMID:28106076 We found that YTHDF3 interacts with the ribosomal proteins |
| GO:0045727 positive regulation of translation | IMP PMID:28106072 YTHDF3 facilitates translation and decay of N(6)-methyladeno... | KEEP AS NON CORE | Summary: YTHDF3 knockdown reduces protein synthesis from m6A-modified transcripts, reported as synergistic with YTHDF1. Reason: Contested. PMID:32492408 reports that the DF paralogs do not induce translation in HeLa cells and that their shared, redundant output is mRNA degradation; GOA carries an explicit NOT annotation for the related term GO:0045948 from that study. The single-paralog knockdown used here is precisely the experimental design the later work argues is confounded by paralog redundancy. Not removed - this is an experimental annotation and the effect may be real in the cell types assayed - but demoted from core, with the dispute recorded. Supporting Evidence: PMID:28106072 Here, we report that YTHDF3 promotes protein synthesis in synergy with YTHDF1, and affects methylated mRNA decay mediated through YTHDF2. PMID:32492408 Furthermore, we find that DF proteins do not induce translation in HeLa cells. |
| GO:0045948 positive regulation of translational initiation | IDA PMID:28106076 Cytoplasmic m(6)A reader YTHDF3 promotes mRNA translation. | UNDECIDED | Summary: YTHDF3 reported to act at the initiation stage of translation, in cooperation with YTHDF1 and via 40S/60S ribosomal subunit contacts. Reason: GOA carries both sides of this dispute on the same term: this positive involved_in IDA and a NOT|involved_in IDA from PMID:32492408. Both are direct experimental annotations from competent laboratories, and the disagreement is substantive rather than a curation slip - it turns on whether single-paralog depletion or simultaneous triple depletion is the informative experiment, and on cell type (HeLa). This cannot be adjudicated from the cached record, and neither annotation may be removed. UNDECIDED is recorded for both rows so that the contradiction is visible rather than silently resolved. Supporting Evidence: PMID:28106076 In this study, we show that YTHDF3 promotes translation, thus playing an important role in the initial stages of translation. PMID:28106076 YTHDF3, in cooperation with YTHDF1, facilitates the translation of targeted mRNAs through binding to m 6 A-modified mRNAs and interacting with 40S and 60S ribosome subunits. |
| GO:0061157 mRNA destabilization | IMP PMID:28106072 YTHDF3 facilitates translation and decay of N(6)-methyladeno... | ACCEPT | Summary: YTHDF3 depletion, alone and together with YTHDF1/2, causes accumulation of m6A-modified transcripts. Reason: Core biological process. This part of the 2017 study is fully consistent with the later unified model, which agrees that the paralogs act redundantly on mRNA decay. Supporting Evidence: PMID:28106072 Cells deficient in all three YTHDF proteins experience the most dramatic accumulation of m6A-modified transcripts. |
| GO:1990247 N6-methyladenosine-containing RNA reader activity | IDA PMID:28106072 YTHDF3 facilitates translation and decay of N(6)-methyladeno... | ACCEPT | Summary: Direct demonstration of m6A-dependent transcript binding by YTHDF3. Reason: Core molecular function. Supporting Evidence: PMID:28106072 Here, we report that YTHDF3 promotes protein synthesis in synergy with YTHDF1, and affects methylated mRNA decay mediated through YTHDF2. |
| GO:1990247 N6-methyladenosine-containing RNA reader activity | IDA PMID:28106076 Cytoplasmic m(6)A reader YTHDF3 promotes mRNA translation. | ACCEPT | Summary: m6A-dependent mRNA binding by the cytoplasmic reader YTHDF3. Reason: Core molecular function. Note that although the translation conclusions of this paper are disputed, the underlying m6A-binding observation is not. Supporting Evidence: PMID:28106076 YTHDF3, in cooperation with YTHDF1, facilitates the translation of targeted mRNAs through binding to m 6 A-modified mRNAs and interacting with 40S and 60S ribosome subunits. |
| GO:0003723 RNA binding | HDA PMID:22681889 The mRNA-bound proteome and its global occupancy profile on ... | KEEP AS NON CORE | Summary: YTHDF3 identified as an mRNA-bound protein in a proteome-wide interactome capture experiment. Reason: True but uninformative relative to the m6A reader term this gene already carries. The high-throughput capture confirms mRNA association without specifying the recognition determinant. |
| GO:1990247 N6-methyladenosine-containing RNA reader activity | IDA PMID:22575960 Topology of the human and mouse m6A RNA methylomes revealed ... | ACCEPT | Summary: YTHDF3 recovered as an m6A-binding protein during the original transcriptome-wide mapping of the human and mouse m6A methylomes. Reason: Core molecular function. Caveat recorded for transparency: the cached record for this reference is abstract-only and does not name YTHDF3, so the supporting data could not be inspected here. Per project policy the curator, who read the full text, is deferred to; the conclusion is in any case supported by six independent IDA annotations to this term. |
| GO:1990247 N6-methyladenosine-containing RNA reader activity | IDA PMID:24284625 N6-methyladenosine-dependent regulation of messenger RNA sta... | ACCEPT | Summary: m6A binding by YTHDF3 assayed alongside YTHDF2 in the study that established m6A-dependent mRNA degradation. Reason: Core molecular function. The title and abstract foreground YTHDF2, but the cached full text shows a GST-YTHDF3 expression construct and an anti-YTHDF3 antibody in the methods, so YTHDF3 was directly assayed - a textbook case of a paper whose abstract names a paralog while the full text also covers the annotated gene. |
Loading supporting content…
Download this section (compressed HTML)Q: Does YTHDF3 promote translation initiation, or not? GOA currently carries an IDA asserting it (PMID:28106076) and a NOT|involved_in IDA denying it (PMID:32492408) for the same term. Is the discrepancy explained by paralog redundancy masking the effect in single knockdowns, by cell type, or by the reporter system used?
Suggested experts: RNA modification biology, Translational control
Q: If all three YTHDF paralogs bind the same transcripts and act redundantly, is there any YTHDF3-specific function at all, or is the correct unit of annotation the DF1/DF2/DF3 group?
Suggested experts: Gene Ontology curation, RNA modification biology
Q: Does the YTH domain of YTHDF3 bind N6-methyladenine in DNA with an affinity and specificity that could matter in cells? A co-crystal or cryo-EM structure of the YTH domain with 6mA-containing duplex DNA would settle whether the aromatic cage can accommodate a deoxyribose backbone.
Suggested experts: Structural biology of nucleic acid recognition
Q: Is there any regulated N6-methyladenine in mammalian genomic DNA for YTHDF3 to read? Several independent audits attribute reported mammalian 6mA to contamination, antibody cross-reactivity and nucleotide-salvage misincorporation, and animals appear to have lost the AMT1 writer lineage.
Suggested experts: Chromatin and DNA modification, Comparative genomics
Q: How much mRNA m1A actually exists, and is YTHDF3's reported m1A binding physiologically meaningful given that mRNA m1A mapping was shown to be dominated by antibody cross-reactivity with the 5' cap?
Suggested experts: Epitranscriptomics, Analytical nucleic acid chemistry
Q: Is the interaction of YTHDF3 with 40S and 60S ribosomal proteins a regulatory contact or incidental co-purification of an abundant cytoplasmic mRNA-binding protein with polysomes?
Suggested experts: Translational control
Experiment: Perform matched ribosome profiling and mRNA-seq in a panel of cell lines carrying degron-tagged YTHDF1, YTHDF2 and YTHDF3, comparing acute single, double and triple depletion on the same day, with m6A-site-resolved analysis of translation efficiency. Acute degradation avoids the compensatory upregulation that confounds stable knockdown, and the panel tests whether HeLa is atypical.
Hypothesis: The translation-promoting role of YTHDF3 is detectable only in specific cell types or under specific stress conditions, which explains the contradiction between PMID:28106076 and PMID:32492408.
Type: acute protein degradation with ribosome profiling
Experiment: Measure binding of purified recombinant YTHDF3 YTH domain to matched m6A-RNA, 6mA-ssDNA and 6mA-dsDNA oligonucleotides by isothermal titration calorimetry and fluorescence polarisation, reporting dissociation constants side by side, and attempt a co-crystal or cryo-EM structure with the 6mA-DNA ligand. An aromatic-cage mutant provides the specificity control.
Hypothesis: The YTH domain of YTHDF3 cannot bind 6mA in duplex DNA with physiologically relevant affinity.
Type: quantitative binding biophysics and structural biology
Experiment: Perform CUT&RUN and ChIP-seq for endogenously tagged YTHDF3 in unstressed human cells, with a nuclear-localised positive control reader and an IgG negative control, and quantify the nuclear versus cytoplasmic pool by fractionation and quantitative western blot. Absence of reproducible chromatin occupancy would argue against a genomic reader role independently of whether 6mA exists.
Hypothesis: YTHDF3 is not associated with chromatin in cells.
Type: chromatin occupancy profiling
Experiment: Re-map the YTHDF3 binding sites on IGF1R mRNA in trophoblast cells using an m1A antibody that lacks 5'-cap cross-reactivity alongside an m6A antibody and a misincorporation-based, antibody-free method, and test whether an aromatic-cage YTH mutant that abolishes m6A binding also abolishes the IGF1R and migration effects.
Hypothesis: The trophoblast phenotype attributed to m1A reading is in fact mediated by m6A.
Type: antibody-independent modification mapping with structure-guided mutants
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)