SET / TAF-Ibeta domain map (Q01105, 290 aa) - dimerisation, earmuff, and the acidic tail that binds MLL

This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.

• Targeting the SET acidic region would hit wild-type SET as well as the fusion, and wild-type SET is a ubiquitous histone chaperone and the endogenous PP2A inhibitor - loss of PP2A inhibition is not a neutral background effect.
• If the MLL contact is what matters, an agent that blocks it may be phenocopied by menin inhibition and may share its resistance routes; conversely a fusion that keeps the CRM1 route can lose the MLL contact and survive.

1. Which residues of the SET acidic region contact MLL, and does that surface overlap the one wild-type SET uses for histones and for PP2A? If it does, the fusion's MLL binding and its chaperone activity are mutually exclusive and only one can be happening at a time.
2. Does the MLL that SET-NUP214 binds carry menin? Nobody has blotted menin in a SET-NUP214 immunoprecipitate. That single western blot is the cheapest decisive experiment in this beat.
3. Is the acidic region's MLL contact conserved in DEK-NUP214? DEK is a different chromatin protein with its own acidic domain, and if it also binds MLL the MLL recruitment is a property of the NUP214 nuclear body rather than of SET.

1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

   Human SET is UniProt Q01105, 290 aa, primary gene name SET, with annotated regions: disordered 1-45, dimerisation 31-78, earmuff domain 79-225, disordered 158-207 and disordered 236-290; its listed PDB entries are 2E50, 7MTO and 9I15.

   UniProtKB Q01105 (SET_HUMAN), Protein SET / Template-activating factor I, 290 aa · UniProt 2026
   ▸What this evidence cannot show

   Fetched live 2026-09-20 with gene_primary read back and length asserted. NOTE for the registry: gene_exact:SET AND organism_id:9606 AND reviewed:true returns EXACTLY ONE reviewed record, so the dispatch brief's warning that this query returns two reviewed human records did not reproduce. The SET hazard in this project is bibliographic, not a resolution trap.

   ▸Evidence that argues against this (1)

   • PUBLISHED CONFLICTING RESULT on what the 290-aa record can represent. SET is expressed as two isoforms with different first exons, TAF-Ialpha and TAF-Ibeta, and the leukaemia lines that carry the fusion express BOTH as fusion transcripts: RT-PCR specific for the alternative first exons shows LOUCY and MEGAL carry TAF-Ialpha-NUP214 mRNA alongside the TAF-Ibeta(SET)-NUP214 transcript. A single reviewed record with one length and one disordered N-terminal call cannot describe both, and every residue coordinate in this entry is implicitly in the beta frame. Any claim about the N-terminal 45 residues - the disordered region and the start of the dimerisation domain - is isoform-specific and must say so. Quentmeier H et al. 2009
2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.review

   SET/TAF-Ibeta is described as composed of an N-terminal dimerisation domain, a central earmuff domain named for its headphone-like structure, and a negatively charged C-terminal acidic domain; SET is expressed as two isoforms, TAF-Ialpha and TAF-Ibeta, generated by alternative splicing of the first two exons, and only TAF-Ibeta is present in SET-NUP214.

   NUP214 in Leukemia: It's More than Transport. · Mendes A, Fahrenkrog B · Cells 2019
   PMID 30669574DOI
   ▸What this evidence cannot show

   A review, cited for nomenclature and architecture rather than measurement. Full text read via Europe PMC (PMC6356203). The review names the gene 'TAF1' in this passage, which is a legacy usage - TAF1 is also the HGNC symbol of the TFIID subunit, an unrelated gene; this entry uses SET/TAF-I throughout to avoid that collision.

   ▸Evidence that argues against this (1)

   • DIRECT PUBLISHED CONTRADICTION of this review's isoform statement. 'Only TAF-Ibeta is present in SET-NUP214' is contradicted by the primary cell-line data: an RT-PCR designed to discriminate the alternative first exons found LOUCY and MEGAL express TAF-Ialpha-NUP214 mRNA in addition to the TAF-Ibeta-NUP214 transcript. The review is being cited here for architecture and nomenclature, which is the right use of it, but the isoform sentence in the same passage is wrong at the mRNA level and this entry should not carry it as settled. Note that the western in the primary paper detects one 140 kDa band with an anti-SET antibody and cannot separate the two, so the protein-level question is genuinely open in both directions. Quentmeier H et al. 2009
3. E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

   The amino-terminal region of MLL (MLLN) interacts with TAF-Ibeta/SET through TAF-Ibeta's acidic region; GST-MLLN pulls down both TAF-Ibeta and histones; and MLLN and TAF-Ibeta synergistically upregulate Hoxa9 transcription and co-immunoprecipitate on chromatin containing the Hoxa9 promoter.

   Synergistic action of MLL, a TRX protein with template activating factor-I, a histone chaperone. · Shimoyama T, Kato K, Miyaji-Yamaguchi M, Nagata K · FEBS Lett 2005
   PMID 15670842DOI
   ▸What this evidence cannot show

   A 2005 biochemical study on a wild-type MLL N-terminal fragment, not a fusion protein and not in a leukaemia cell. This evidence item is carried in this graph's `gene-set` entry and its identifiers were re-fetched from PubMed for this entry rather than copied; the DOI and title were taken from the fetched esummary record on 2026-09-20.

   ▸Evidence that argues against this (1)

   • PUBLISHED CONFLICTING RESULT on the SELECTIVITY the entry's title asserts. Calling this 'the acidic tail that binds MLL' names one partner for a surface that has several documented ones: Set/TAF-Ibeta and pp32, as INHAT subunits, bind unacetylated, hypoacetylated and repressively marked histones but not hyperacetylated histones, associate with histone deacetylases in vitro and in vivo, and repress transcription from a chromatin-integrated template. Histones and MLLN are both basic ligands for the same acidic region, and a GST pulldown that recovers TAF-Ibeta AND histones together - which is what this study reports - does not establish that MLL is the physiological occupant. The title overstates what the mapping shows. Kutney SN et al. 2004
4. E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

   SET-NUP214 interacts with MLL via the SET acidic region of SET-NUP214, and the two cooperatively enhance the promoter activity of the HOXA10 gene, with neither the SET region alone nor the NUP214 region alone sufficient.

   SET-NUP214 and MLL cooperatively regulate the promoter activity of the HoxA10 gene. · Cigdem S, et al. · Genes Cells 2021
   PMID 34320268DOI
   ▸What this evidence cannot show

   ABSTRACT-LEVEL ONLY for this beat: the paper is not in PMC. A promoter-reporter assay plus deletion mapping, with the cell system unstated in the abstract. 'MLL' is not resolved to an accession in the abstract and the MLL2/MLL4 alias hazard (brief rule 66) means the bare word is not self-resolving; read here as MLL1/KMT2A on the HOXA10 context and recorded as unresolved.

   ▸Evidence that argues against this (1)

   • AUTHORS' OWN NEGATIVE IN THE FOUNDING PAPER, at the endogenous locus this promoter assay stands in for. A reporter construct plus deletion mapping measures cooperation on a transfected template; the corresponding endogenous readout in the fusion-positive line was run in 2008 and was flat - siRNA knockdown of SET-NUP214 in LOUCY left histone H3 acetylation at the HOXA locus UNALTERED within the experiment's time frame, and immunoprecipitation failed to demonstrate an interaction between SET-NUP214 and the INHAT components SET and PP32. A SET-acidic-region mechanism at a HOXA promoter predicts an acetylation change; the one measurement of it did not find one. The authors bound their negative by the siRNA time frame, which is the correct caveat and not a dismissal. Van Vlierberghe P et al. 2008
5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.review

   Only the TAF-Ibeta isoform of SET is present in the SET-NUP214 fusion protein.

   NUP214 in Leukemia: It's More than Transport. · Mendes A, Fahrenkrog B · Cells 2019
   PMID 30669574DOI
   ▸What this evidence cannot show

   Review statement, attributed there to a primary report. It sits in tension with E6, which finds TAF-Ialpha-NUP214 mRNA in both fusion-positive cell lines; the two are reconcilable if the alpha transcript is not translated or not detected at protein level, but no source read in this beat resolves that.

   ▸Evidence that argues against this (1)

   • RT-PCR distinguishing the alternative first exons of the two TAF-I isoforms found that LOUCY and MEGAL express TAF-Ialpha-NUP214 mRNA as well as TAF-Ibeta-NUP214, so the transcript is not exclusively beta even if the protein is. Quentmeier H et al. 2009
6. E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   RT-PCR specifically recognising the alternative first exons of the two TAF-I isoforms showed that LOUCY and MEGAL express TAF-Ialpha-NUP214 mRNA in addition to the TAF-Ibeta(SET)-NUP214 transcript; both lines express a 140 kDa fusion protein.

   SET-NUP214 fusion in acute myeloid leukemia- and T-cell acute lymphoblastic leukemia-derived cell lines. · Quentmeier H, et al. · J Hematol Oncol 2009
   PMID 19166587DOI
   ▸What this evidence cannot show

   Cell lines only. The western blot detects a single 140 kDa band with an anti-SET antibody, which cannot distinguish an alpha-derived from a beta-derived fusion protein because the isoforms differ only at the extreme N-terminus.

   ▸Evidence that argues against this (1)

   • AUTHORS' OWN STATED LIMITATION, unclosed for seventeen years, and the field's review disagrees with the result. The mRNA finding does not reach the protein: the anti-SET western detects a single 140 kDa band and cannot distinguish an alpha-derived from a beta-derived fusion protein, because the isoforms differ only at the extreme N-terminus. No isoform-specific antibody, no N-terminal peptide by mass spectrometry, and no ribosome-profiling evidence has been published since. Meanwhile the standing review of NUP214 in leukaemia states flatly that only TAF-Ibeta is present in SET-NUP214. One of the two is wrong and the experiment that would settle it - an N-terminal peptide from the endogenous fusion in LOUCY - has not been done. Mendes A et al. 2019
7. E7strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   In LOUCY, immunoprecipitation failed to demonstrate a direct interaction between SET-NUP214 and the INHAT components SET and PP32, and histone H3 acetylation at the HOXA locus was unaltered by SET-NUP214 knockdown within the experiment's time frame.

   The recurrent SET-NUP214 fusion as a new HOXA activation mechanism in pediatric T-cell acute lymphoblastic leukemia. · Van Vlierberghe P, et al. · Blood 2008
   PMID 18299449DOI
   ▸What this evidence cannot show

   A negative co-IP in one cell line: absence of a band is weaker than presence of one, and the authors say only that the interaction could not be demonstrated. The acetylation result is explicitly bounded by the limited time frame of the siRNA experiment. Full text read via the PMC HTML route (PMC2343598).

   ▸Evidence that argues against this (1)

   • PUBLISHED CONFLICTING RESULT against over-reading this double negative. A failed co-IP in one cell line does not license 'the INHAT route is not operating': Set/TAF-Ibeta and pp32 do form an INHAT complex, do bind hypoacetylated and repressively marked histones, do associate with HDACs and do repress transcription from a chromatin template, all demonstrated biochemically and in cells. What the founding paper shows is that it could not detect the complex on the fusion in LOUCY, with the antibodies and conditions used, and that acetylation at HOXA did not move within its siRNA time frame - a limit the authors state themselves. Absence of a band is the weakest negative available, and this one is being asked to carry a mechanistic exclusion. Kutney SN et al. 2004

• part of →SET / TAF-I beta - a histone chaperone that binds the retained N-terminus of MLL, co-activates HOXA with it, and is itself a HOXA-activating fusion partner that recruits DOT1L

  isoform caveat: LOUCY and MEGAL express TAF-Ialpha-NUP214 as well as TAF-Ibeta-NUP214, so a single 290 aa record cannot frame every coordinate here

• binds →KMT2A / MLL1 — the gene that breaks

  the SET acidic region binds the MLL N-terminus in both the wild-type and the SET-NUP214 direction

• ← depends onSET::NUP214 is KMT2A-free in its sequence but not in its mechanism: it binds and condenses wild-type MLL1, which reopens the menin question this beat was sent to close

  the contact is the SET acidic region (aa 236-290)