henhaobio
Sign in

bioRxivMolecular BiologyPreprint

Cellular context restricts a promiscuous m6A reader IDR to a single functional effector interface for mRNA decay

Wei Yee ChaniD, Enkai JiniD, Hector Cross, Laura MasinoiD, Fairouz IbrahimiD, Waleed S. AlbihlaliD, Duncan Berger, James R. BootiD, Aleksandra MichrowskaiD, Stéphane MouilleroniD, Mark J. SkeheliD, Kamil R. KranciD, Folkert Jacobus van WerveniD

PDFDOI

In-depth summary

Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (5 Aug 2026) · CC BY 4.0

Yeast m6A reader Pho92 uses one helical hydrophobic interface in its disordered region to engage Caf40; a single L35A substitution halts m6A-transcript decay in meiosis.

How m6A readers recruit decay effectors is poorly defined, partly because mammalian YTHDF paralogs are redundant. Yeast has one reader (Pho92) and a defined meiotic window, and this work shows that in vitro binding and tethered-reporter activity do not predict which interface is used on endogenous transcripts.

Key results

  1. ShownIn vitro, Pho92 IDR peptides bind human CNOT9 (Caf40 ortholog) through tryptophans and a helical segment. In cells, Caf40 binding depends on the helical region, not the tryptophans.

    BLI Kd: Pho92 1-25 15 ± 2 μM; 28-52 155 ± 21 μM; W5A 400 ± 50 μM; W18A 1220 ± 160 μM; W5A/W18A >3000 μM. Co-IP: W5A/W18A retained Caf40 binding; Δ1-70, Δ1-60 and Δα strongly reduced it. · In vitro, Pho92 directly interacts with Caf40…; In cells, Pho92 interaction with Caf40 requires…

  2. ShownCaf40 and Pop2 (Ccr4-NOT subunits) are needed for turnover of endogenous Pho92/m6A targets in meiosis. Depleting Not5 or other decay factors had little effect.

    Targets rose in CAF40-AID, POP2-AID and pho92Δ cells (mean log2 FC 0.71, 0.37, 0.26 in one listing; the text's other listing pairs Caf40 and pho92Δ as 0.26 vs 0.37). WT m6A signal fell about 50% after thiolutin; it stayed unchanged in caf40Δ, CAF40-AID and POP2-AID (n=2). caf40Δ meiosis: about 60% vs 90% WT at 24 hours. · Caf40 is specifically required for the turnover…; Figure legends

  3. ShownA single hydrophobic substitution, L35A, in the Pho92 helical region strongly weakens peptide binding to CNOT9 in vitro and stabilises m6A-modified transcripts in meiosis, with an RNA-seq effect comparable to the m6A-binding-dead W177A.

    BLI Kd 1410 ± 140 μM for 28-52 L35A vs 155 ± 21 μM wild type; the 4A peptide showed no binding. pho92-L35A m6A signal was unchanged at 0 and 15 min (n=2), and its RNA-seq target increase was comparable to W177A. · A single point mutation in the N-terminal IDR of…; Figure legends

  4. SuggestedIn a tethered reporter outside meiosis, the Pho92 IDR is promiscuous. Caf40 loss does not relieve repression, and the Not1 DUF3819 domain appears to be a redundant interface that Caf40 is proposed to occlude.

    Not1 enriched over 8-fold in caf40Δ proximity labelling. Reporter ratio: 0.2 WT, 0.4 not1-Δduf, 0.6 not1-Δduf + caf40Δ. Repression was retained in caf40Δ; ccr4Δ and pop2Δ gave small partial increases. · Caf40 is specifically required for the turnover…; Pho92 IDR is promiscuous in a heterologous…

  5. ShownHuman YTHDF2 can substitute for Pho92 in meiotic m6A-transcript decay in yeast. Its N-terminal IDR, tested in the tethered reporter, needs several hydrophobic patches rather than a single site.

    YTHDF2 N-terminus repressed the tethered reporter by more than 80%; YTHDF2 in pho92Δ cut m6A signal within 15 min of thiolutin, abolished by W491A. LIVFtoST fully derepressed; cLIVFtoST and 3pLIVFtoST ~0.6 vs ~0.2 WT. · The YTHDFs IDR are functionally active in yeast; The YTHDF N-terminus hydrophobicity is required…

Methods, data and reproducibility

TurboID proximity proteomics and AlphaFold modelling nominated Caf40; BLI, circular dichroism and co-IP mapped the interface. Tethered-reporter flow cytometry, auxin-degron RNA-seq, and m6A-ELISA transcription-block assays tested function.

  • TurboID proximity proteomics of Pho92 in meiotic wild-type, slz1Δ and caf40Δ cells (LC-MS/MS, MaxQuant LFQ) · Saccharomyces cerevisiae (SK1) · n = triplicates per group
  • Poly(A) RNA-seq of meiotic cells: pho92 IDR mutants, pho92Δ, and nine auxin-degron depletion strains · Saccharomyces cerevisiae (SK1) · n = minimum 20 million reads per sample · GSE339410, GSE339409, GSE339501
  • BLI of synthetic Pho92 and YTHDF2 peptides against recombinant human CNOT9 (residues 16-284) · human protein expressed in E. coli · n = n ≥ 3 independent replicates
  • Dual YFP/RFP λN-boxB tethered reporter flow cytometry of Pho92 and YTHDF1-3 fragments and mutants · Saccharomyces cerevisiae (BY4741) · n = 3 biological replicates; minimum 10,000 cells per sample
  • m6A-ELISA transcription-block (thiolutin) stability assays in meiotic cells · Saccharomyces cerevisiae (SK1) · n = n=2 biological repeats
  • DAPI counting of meiotic divisions in wild-type, caf40Δ, pho92Δ · Saccharomyces cerevisiae (SK1) · n = n = 200 cells; 3 biological repeats
Methods & tools
TurboID, MaxQuant 2.0.3.1, Perseus 1.4.0.2, AlphaFold2-Multimer v2.3.2, AlphaFold3 server, ChimeraX 1.6.1, Biolayer interferometry (Octet R8), Circular dichroism spectroscopy, nf-core/rnaseq, STAR, DESeq2 with IHW, clusterProfiler
Reagents & environment
nf-core/rnaseq 3.10.1 with Nextflow v23.10.0, STAR 2.7.10a, DESeq2 1.48.1; and nf-core/rnaseq 3.25.0 with Nextflow v25.10.4, STAR 2.7.11b, DESeq2 1.50.2; RSEM v1.3.1; AlphaFold v2.3.2-multimer; ChimeraX 1.6.1.
Assay conditions & compute
AlphaFold2-Multimer run on the Francis Crick Institute HPC cluster using one GPU; AlphaFold3 run on the hosted server.
Reproducibility gaps
The only code link is the AlphaFold3 output-processing tool. No repository for the custom analysis scripts is given, and no proteomics deposition is stated.
Limitations (6)
  • Yeast Caf40 could not be produced recombinantly, so all biophysical binding assays used the human ortholog CNOT9. · stated by the authors
  • The AlphaFold screening cut-offs for candidate interactors were described by the authors as relatively lenient, so the candidate shortlist is prediction-based. · stated by the authors
  • How active translation couples to Pho92-Caf40-dependent decay is unresolved. Whether translation restricts YTHDF IDR promiscuity in mammals is untested; YTHDF function was assessed only in yeast. · stated by the authors
  • The tethered reporter bypasses the translation requirement and runs under vegetative growth, so the authors caution that it can give outcomes that differ from physiological decay of endogenous transcripts. · stated by the authors
  • The transcript stability assays (m6A-ELISA transcription block) rest on n=2 biological repeats and a bulk m6A signal rather than per-transcript decay rates, which limits statistical power. · noted in this summary
  • Full-length Pho92 L35A and Caf40-binding-deficient alleles were not shown to retain equivalent translation engagement or localisation, so effects on other factors cannot be excluded. · noted in this summary

Read this if you study m6A reader mechanisms, IDR-mediated effector recruitment, or Ccr4-NOT, or want an example of tethered-reporter results diverging from endogenous function.

Citations
1

Abstract

N6-methyladenosine (m6A) is a conserved mRNA modification that regulates transcript stability, yet how m6A readers engage effectors remains unknown. We show that during yeast meiosis, the YTH-domain protein Pho92 promotes the decay of m6A-modified transcripts through an uncharacterised, promiscuous intrinsically disordered region (IDR). In vitro, the Pho92 IDR makes multiple contacts with the Ccr4-NOT deadenylase complex and drives tethered reporter decay in the absence of any single subunit. Yet in meiosis, turnover of endogenous m6A-modified transcripts strictly depends on a single interface: direct binding of the Pho92 IDR to Caf40. A single hydrophobic residue substitution is sufficient to abolish Caf40 binding and halt transcript decay. Human YTHDF proteins can functionally substitute for Pho92, but require multiple hydrophobic patches within the IDR for decay activity. Thus, on endogenous m6A-modified transcripts during meiosis, the promiscuous Pho92 IDR is restricted to a single functional interface, revealing unappreciated specificity in m6A-directed mRNA decay.

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISVarier '22Sene '26Ricci '26Scutenaire '22Lasman '20Lee '24
80% semantic matchBiochemistry
R. A. Varier, T. Sideri, C. Capitanchik+15 more·bioRxiv·Jan 20, 2022