Oncofusion

start / graph / gene-nutm1

Proteinstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q100-nutm1-comparator

NUTM1 (NUT) - a testis-restricted, fully disordered protein that writes no mark and instead switches ON a writer, and the 3' partner of the KMT2A-wild-type infant B-ALL with the opposite prognosis

also called NUTM1 · NUT · C15orf55 · NUT family member 1 · nuclear protein in testis

NUTM1 is a protein normally made only in maturing sperm cells, where it switches on a chromatin-modifying enzyme called p300 so that histones can be stripped out of the sperm genome. When a chromosome break fuses it to another gene in a baby's blood cell that switch gets turned on in the wrong place and causes leukaemia - but a leukaemia that, unlike the KMT2A-rearranged kind, responds very well to standard chemotherapy.

The same thing, for a scientist

NUTM1 (Q86Y26, 1132 aa) is the 3' partner of every NUTM1-rearranged malignancy, from NUT carcinoma to the KMT2A-wild-type infant B-ALL subtype that is this entry's reason for existing. Its own architecture is the point: live UniProt annotates **no structured domain at all** - eight predicted disordered regions and a single Pfam family, PF12881 (NUT), with no SET, JmjC, HAT, HDAC, bromodomain, chromodomain, Tudor, PHD or CXXC module [E1]. NUTM1 therefore **writes, erases and reads no histone or DNA modification itself**; it is an intrinsically disordered *allosteric activator* of a writer. Two transactivation domains, TAD1 (403-418) and TAD2 (419-470), each bind the TAZ2 domain of p300 as an amphipathic helix, with NMR structures (7XEZ, 7XFG) and a 1:2 stoichiometry that enhances NUT's own liquid-like condensation; the bipartite contact allosterically activates p300 [E2]. Physiologically that produces genome-wide H4K5ac/H4K8ac in post-meiotic spermatogenic cells, read by the first bromodomain of BRDT and required for histone-to-protamine exchange - Nut-null mice are sterile, arrested at histone removal [E3]. Fused to a chromatin-binding 5' partner the same loop runs in a somatic cell: p300 is recruited, activated and sequestered into hyperacetylated foci [E4], which at genome scale are megadomains up to ~2 Mb bounded by TAD boundaries [E5], and in leukaemia are a global rise in H3K27ac plus **global DNA hypomethylation that is independent of the 5' partner** [E6]. **Localization is the trap**: UniProt calls native NUT *Cytoplasm and Nucleus* with a note that it shuttles, and the founding paper shows native NUT carries both an NLS and a leptomycin-sensitive NES while BRD3-NUT and BRD4-NUT are **strictly nuclear** [E1][E7]. The nuclear-foci address that everyone quotes is a property of the **fusion**, not of the protein this entry is about.

What it looks like
NUT family member 11,132 aa
TAD1 (transactivation domain 1) · residues 403–418TAD2TAD2 (transactivation domain 2) · residues 419–47002004006008001,000
loading AF-Q86Y26-F1…
AlphaFold AF-Q86Y26-F1 ↗
colourshape

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.

very high (90+)confident (70–90)low (50–70)very low (<50)
I have never read one of these pictures before

A protein is one long chain of amino acids that folds up. The ribbon traces that chain from one end to the other; it is not a wire, it is a simplification of where the backbone runs. Drag to rotate, scroll to zoom.

Domains are the stretches of chain that do a particular job, and they are what the coloured blocks above the picture mark out. Click one and the view flies to it. In this project the important question about any domain is usually whether the fusion still has it — the labels say lost where it does not.

Anything in yellow sticks is a bound small molecule, which is usually the drug.

Go deeper
Why a protein with no domains is the interesting case

This graph's epigenetics dimension usually asks which mark does this protein write, erase or read. For NUTM1 the honest answer is none, and that is not an absence of data - it is the mechanism. Live UniProt (fetched 2026-09-22) returns for Q86Y26 exactly eight Region features, all Disordered and all ECO:0000256 (i.e. computational), and exactly one Pfam family, PF12881 NUT 1. There is no catalytic module and no acetyl-lysine, methyl-lysine or CpG reader module anywhere in 1,132 residues.

What NUTM1 has instead is two short transactivation helices that grip the TAZ2 four-helix bundle of p300 and allosterically switch the acetyltransferase on 2. The consequence is that NUTM1 is an epigenetic protein in the only sense that matters clinically - it changes the acetylation state of the genome - while being, structurally, a disordered adaptor. A domain-architecture screen looking for writers, erasers or readers would score it negative and be right about the architecture and wrong about the biology.

That also means the mark is set by p300, which this graph already holds as gene-ep300, and the drug-discovery consequence is the p300 HAT domain and the BET reader, not NUTM1 itself.

Subcellular and subnuclear localization, and why the database answer is about the fusion

Subcellular. UniProt lists Cytoplasm (SL-0086) and Nucleus (SL-0191), both ECO:0000269, with the note "Shuttles between nucleus and cytoplasm", and every one of those three statements points at the same citation, PMID 17934517 1.

What that paper actually shows (rule 91 - the cited paper is a pointer, not a source): it is a study of BRD-NUT oncoproteins. Native NUT "contains nuclear localization and export sequences that promote nuclear-cytoplasmic shuttling via a leptomycin-sensitive pathway", whereas "BRD3-NUT and BRD4-NUT are strictly nuclear, implying that the BRD moiety retains NUT in the nucleus via interactions with chromatin", confirmed by FRAP showing slower lateral nuclear diffusion for the fusions than for NUT 7. So UniProt's record for the native protein is correct and is sourced to a paper about the fusion - and the direction of the finding is that nuclear retention is contributed by the partner.

Subnuclear. The discrete nuclear foci that NUT immunohistochemistry is famous for, and the ~100 hyperacetylated megadomains they correspond to, are described for BRD4-NUT, not for native NUT 45. For native NUTM1 in its own cell type, no subnuclear address has been reported in anything read for this entry. Recorded as a bounded absence in data.localization, not as a nucleoplasm default.

Cell type and state. Normal expression is specifically testis (ECO:0000269, PMID 12543779) 1, and the functional window is narrower still: expression "is normally restricted to post-meiotic spermatogenic cells" 3. In disease the cell of origin is a B-cell progenitor - BRD9::NUTM1 expression alone produces a prepro-B-like leukaemia in mice, and all three pro-B-immunophenotype infants in the 85-case international cohort carried BRD9::NUTM1 68.

The contrast with KMT2A, stated as a contrast

Same disease, same age band, same cohorts, opposite outcome - and the mechanisms are not analogous, which is the part worth being careful about.

KMT2A-rearranged infant B-ALLNUTM1-rearranged infant B-ALL
what the fusion bringsa CXXC CpG-binding module and a menin-binding motif on the 5' side, an elongation-machinery recruiter on the 3' sidea disordered p300 activator on the 3' side; the chromatin address comes from the 5' partner
the markH3K79me2/3 written by DOT1L, recruited through the 3' partnerH3K27ac (and H3K18ac in the carcinoma condensate) written by p300, plus global DNA hypomethylation
meninthe interface the whole therapeutic programme targetsno reported role; NUTM1 has no MBM-like motif and no menin contact in anything read here
HOXA9upalso up, in the ACIN1/CUX1/BRD9/AFF1 partner group 8
4-year EFS, uniformly treated Interfant infants74.0% for the KMT2A-wild-type NUTM1-negative comparator arm; ~36-37% for KMT2A-rearranged in an independent infant cohort100% (n=35) 8

The convergence in that table is more interesting than the divergence. Both diseases run a posterior-HOXA programme - the Blood 2026 functional work names "posterior HoxA genes" among the co-opted programmes 6, and the international cohort's BLUEPRINT case shows HOXA9/HOXA10 expression in an ACIN1-NUTM1 patient "comparable with levels observed in two cases with KMT2A rearrangement", with reduced CpG methylation and high H3K4me3/H3K27ac over HOXA9-HOXA10 8. HOXA9 activation therefore does not require the KMT2A-menin interface. That is a real constraint on any claim that the menin interface is the necessary route to the HOXA programme, and it is stated as a constraint, not as a refutation: nothing read here tests menin dependence in a NUTM1-rearranged cell either way.

The interactome record for this protein is an instrument failure in both directions

Measured live, 2026-09-22 9:

  • IntAct REST findInteractions/Q86Y26 returns 84 interaction records / 24 named human partners, and every single one is a two-hybrid variant or a CLASH record from an ORFeome-scale screen (PMID 25416956, 31515488, 22810586, 36217029, 23622248). ALOX5, CCHCR1, FAM161A, HSPBAP1, IFIT2, KANSL1, KPNA2, LHX8, LMO1/2/3, MORN4, NEBL, P4HA3, PRKAA2, PSME4, RIC8A, SAMD4A, SH2D4A, TP53BP2, ZMYND10, ZNF277.
  • EP300, CREBBP, BRD4 and BRDT are absent from all 84. The interaction this protein's entire literature is about - the one with two NMR structures 2, an XL-AP-MS identification 10 and a founding recruitment paper 4 - is not filed under NUTM1's accession.
  • It is in IntAct, filed under a fusion-construct interactor: EBI-8101903, name brd4-nut, Homo sapiens, 3 records, all fluorescence imaging / colocalization, all PMID 20676058: EP300 (host Homo sapiens HCC2429), H4C16 (histone H4), and one Cercopithecus aethiops record.

This is cycle 54's finding reproduced in a second fusion family. Cycle 54 established that IntAct curates records in which the participant is a KMT2A fusion construct (mll_af9_human_protein and three siblings, 47 records) and that this project's gene_of() normaliser drops them all. The same curation pattern exists for NUT, under brd4-nut, and it holds the only NUT-p300 record in the database.

And it runs the other way too. A coverage metric reading NUTM1's interactome off IntAct would see 84 records and 24 partners and score this protein well sampled. It is not sampled at all: 24 Y2H hits and zero of the four interactions that matter. Record count is not evidence class. Rule 93's second clause applies exactly - absence of a famous interaction is an instrument alarm, not a fact about biology.

Two more database defects in this record, both rule-91 shaped
  1. UniProt's FUNCTION for Q86Y26 cites one paper, and it is not the function paper. The only ECO:0000269 function citation is PMID 30447097, a genome-wide shRNA screen in HepG2 hepatocellular carcinoma cells identifying C15orf55 as a positive regulator of TERT transcription through two GC motifs in the SP1 site of the TERT promoter 11. That is a real result, it is about this gene, and it is filed under the retired symbol. Meanwhile PMID 30257209 - the paper that established what NUTM1 does in the cell that normally expresses it, p300-dependent genome-wide H4 hyperacetylation in spermatogenesis, with a knockout phenotype 3 - is absent from UniProt's nine-reference list entirely. Checked programmatically against the live record: 30257209 in refs? False, 20676058 in refs? False, 17934517 in refs? True 1.
  2. The localization annotation is sourced to a fusion paper, as set out above. Both defects are flagged for the PI, not edited; neither is an error in this graph.
Identifier traps measured for this gene

All counts from PubMed E-utilities esearch, one tag per term, querytranslation read back from each response and stored in data.enumeration_record_counts (cycle-54 registry row 123: a count of 0 with an EMPTY translation is an instrument failure, not a result; none of these came back empty).

  • Three symbols, wildly unequal recall. NUTM1[tiab] = 304. The retired symbol C15orf55[tiab] = 3 - and one of those three is the paper UniProt cites for the gene's entire FUNCTION field 11. The other synonym, NUT[tiab], returns 9,391, because NUT is a three-letter token in food science and because nuclear protein in testis is this gene's own former name. Compare id sets, never counts.
  • A localization query on this gene is contaminated by the gene's own name. NUTM1[tiab] AND nuclear[tiab] = 102, and "nuclear protein in testis"[tiab] alone = 200. A large share of the 102 are papers spelling out the acronym in the abstract, not reporting a localization. This is a new collision class for the registry: the gene's approved symbol expands to a phrase containing a localization term, so the localization query cannot be read at face value.
  • The p300 literature is written under the OLD symbol. NUTM1[tiab] AND p300[tiab] = 6; NUT[tiab] AND p300[tiab] = 25, and every founding paper in this entry is in the 25 and not in the 6. A beat that queried only the approved symbol would conclude the p300 link is thin.
  • The 5' partner contaminates in the opposite direction, knowingly. Registry row 120 records that BRD9 AND leukemia returns BRD9::NUTM1 fusion papers rather than ncBAF papers and prescribes avoiding the subunit symbol. For this beat that contamination is the signal, and the row's advice is inverted on purpose - BRD9[tiab] AND NUTM1[tiab] = 2, and both are papers this beat is built on.
  • Controls. Genuine zero: TMEM103[tiab] = 0 with the term echoed back unparsed (TMEM103[tiab]). Genuine on-topic zero with a parsed translation: SLC12A6[tiab] AND NUTM1[tiab] = 0, translation "SLC12A6"[Title/Abstract] AND "NUTM1"[Title/Abstract] - both operands resolve, the conjunction is empty. Neither shape is the cycle-54 failure mode, which returns an EMPTY translation.
Facts and figures
uniprotQ86Y26
gene symbolNUTM1
protein nameNUT family member 1
length AA1132
alphafold IDAF-Q86Y26-F1
hgncHGNC:29919
cytoband15q14
accession resolutionUniProt REST query gene_exact:NUTM1 AND organism_id:9606 AND reviewed:true, run 2026-09-22; exactly one reviewed hit; primary gene name read back as NUTM1 with synonyms C15orf55 and NUT, uniProtkbId NUTM1_HUMAN, sequence length 1132 read from the same fetched record.
PDB noteBoth are NMR solution structures of the p300 TAZ2 domain in complex with a BRD4-NUT F1c binding motif (7XFG motif #1, 7XEZ motif #2); titles verified live against the RCSB entry API 2026-09-22. There is no structure of full-length NUTM1 and, with eight predicted disordered regions, there is unlikely to be one.
domains noteThese two TADs are the ONLY structured elements this protein is known to have, and they are structured only when bound - both adopt amphipathic helices on the TAZ2 surface and are disordered free in solution. UniProt annotates neither of them and annotates no domain at all, so they are recorded here from the primary structural paper. Pfam for Q86Y26 is exactly one family, PF12881 (NUT), matching twice.
disordered regions noteEight Region features, all described 'Disordered' and all ECO:0000256 (automatic sequence analysis). Kept out of data.domains because they are predictions rather than measured boundaries, and because a predicted disorder region is not a figure lane.
cell typenormal expression is testis-specific and, functionally, restricted to POST-MEIOTIC spermatogenic cells (PMID 30257209). In disease the cell of origin is a B-cell progenitor: BRD9-NUTM1 alone produces a prepro-B-like leukaemia in mice (PMID 41460961), and all three pro-B-immunophenotype infants in the 85-case international cohort carried BRD9-NUTM1 (PMID 34211097).
cell statein spermatogenesis, the histone-to-protamine exchange window; in leukaemia, a B-lymphoid-committed state that simultaneously carries leukaemic stem cell properties - the Blood 2026 'dual oncogenic role' (PMID 41460961).
epigenetic readoutNOT APPLICABLE AS AN ACTIVITY - AND EMPHATICALLY APPLICABLE AS A MECHANISM. NUTM1 writes, erases and reads no histone or DNA modification, and the reason is domain architecture fetched live rather than a failure to find one: UniProt Q86Y26, fetched 2026-09-22, is 1132 aa carrying EIGHT Region features, every one of them 'Disordered' and ECO:0000256, NO domain, motif, zinc-finger or coiled-coil feature of any kind, and a Pfam complement of exactly PF12881 (NUT). There is no SET, JmjC, HAT, HDAC, bromo, chromo, Tudor, PHD, CXXC, YEATS, WD40, MBD or DNMT module anywhere in the protein. WHAT IT DOES INSTEAD is the whole point of this entry: two transactivation helices, TAD1 403-418 and TAD2 419-470, bind the TAZ2 domain of p300 and ALLOSTERICALLY ACTIVATE the acetyltransferase (NMR 7XEZ/7XFG, PMID 36690674), so NUTM1 is a disordered ACTIVATOR of a writer. The marks are therefore real and they are p300's: physiologically genome-wide H4K5ac and H4K8ac in post-meiotic spermatogenic cells, read by the first bromodomain of BRDT and required for histone-to-protamine exchange, with Nut-null mice sterile and arrested at histone removal (PMID 30257209); in the fusion, hyperacetylated chromatin comprising H3K27ac and H3K18ac within a p300-dependent condensate (PMID 36690674) and, at genome scale, ~100 megadomains up to 2 Mb (PMID 26220994). IN THE LEUKAEMIA THE READOUT IS TWO MARKS AND ONE OF THEM IS A DNA MARK: global enhancement of H3K27 acetylation, and GLOBAL DNA HYPOMETHYLATION THAT IS INDEPENDENT OF THE 5-PRIME FUSION PARTNER (PMID 41460961) - unusual, because partner-independence means the methylome phenotype is attributable to the NUTM1 moiety rather than to the DNA-binding module the partner contributes, which is the opposite of the HOXA phenotype, where partner identity decides whether HOXA9 goes up at all (PMID 34211097). NO MECHANISM FOR THE HYPOMETHYLATION IS ESTABLISHED in anything read here: NUTM1 has no DNMT or TET module and no reported contact with one, so whether this is a consequence of the proliferative or stem state, of a p300-dependent change in DNMT recruitment, or of something the fusion does directly, is open, and the discriminating experiment has not been done. REOPENING CONDITIONS for the NOT-APPLICABLE half: (a) any report of a catalytic or reader module in NUTM1 that UniProt does not yet annotate; (b) any experiment showing NUTM1 binding a modified histone peptide directly rather than through p300 or a BET partner; (c) any demonstration that the NUTM1 moiety contacts a DNMT, a TET or a DNA-methylation reader, which would make the hypomethylation a direct activity rather than a downstream state.
epigenetic roleactivator of a writer (p300/CBP); not a writer, eraser or reader itself
interaction evidence classNMR structures of two peptide-TAZ2 complexes plus GST pulldowns (PMID 36690674); XL-AP-MS of the fusion against wild-type BRD4 (PMID 28484033); fluorescence-imaging colocalization curated in IntAct under a fusion construct (PMID 20676058). NO proximity-labelling map exists, and no interactome of any NUTM1 fusion in a lymphoid cell exists.
classificationthe 3' partner shared by every NUTM1-rearranged malignancy; in this graph it is the KMT2A-WILD-TYPE COMPARATOR, not a KMT2A interactor. NUTM1 has no reported contact with KMT2A, menin, the MBM, the CXXC domain or the WIN site in anything read for this entry.
menin relationshipNONE REPORTED, and now BOUNDED rather than merely unobserved. NUTM1 carries no MBM-like RWRFP motif, and no menin contact appears in the 84 IntAct records, in UniProt's interaction list, or in any of the eleven papers cited here. Bounded by query, run twice each with parsed querytranslations recorded in data.enumeration_record_counts: NUTM1[tiab] AND menin[tiab] = 0, NUTM1[tiab] AND MEN1[tiab] = 0 - both names tried, per rule 72 - and NUTM1[tiab] AND DOT1L[tiab] = 0 with NUTM1[tiab] AND H3K79[tiab] = 0. The index disease's interface and its mark share NO title-or-abstract literature with this gene. Rule 80: titles and abstracts only; supporting figures not searched.
subnuclear localizationNOT ESTABLISHED for native NUTM1. The discrete nuclear foci, and the ~100 hyperacetylated megadomains up to 2 Mb that they correspond to, are reported for BRD4-NUT and BRD3-NUT, which are strictly nuclear because the BRD moiety retains them on chromatin (PMID 17934517, 26220994). BOUNDED: nothing read for this entry reports a subnuclear compartment for endogenous NUTM1 in a spermatogenic cell or in any lymphoid cell. Note the query hazard: NUTM1[tiab] AND localization[tiab] returns 2 records and NUTM1[tiab] AND nuclear[tiab] returns 102, the latter heavily contaminated by abstracts spelling out the gene's own former name 'nuclear protein in testis' (200 records for that phrase alone). Searched in titles and abstracts only; supporting figures not searched (rule 80).
localization shown for fusionBRD4::NUTM1, BRD3::NUTM1 (carcinoma); nobody has imaged BRD9::NUTM1 or ACIN1::NUTM1 in a B-cell progenitor
previous symbols
[
  "NUT",
  "C15orf55"
]
PDB IDS
[
  "7XEZ",
  "7XFG"
]
domains
[
  {
    "name": "TAD1 (transactivation domain 1)",
    "start": 403,
    "end": 418,
    "source": "PMID 36690674 main text and Fig. 2a; the validated TAZ2-binding motif, with L410G/L411G/Y413G/I414G abrogating binding. Figure 1d's legend gives 377-418, which are GST construct boundaries, not the domain."
  },
  {
    "name": "TAD2 (transactivation domain 2)",
    "start": 419,
    "end": 470,
    "source": "PMID 36690674 main text and Fig. 2b; resolved helices within it are 433-440 and 448-460, and Fig. 2e's alignment window is 430-458."
  }
]
disordered regions uniprot
[
  [
    1,
    56
  ],
  [
    337,
    365
  ],
  [
    383,
    405
  ],
  [
    476,
    584
  ],
  [
    693,
    714
  ],
  [
    873,
    892
  ],
  [
    922,
    1017
  ],
  [
    1031,
    1132
  ]
]
localization
{
  "subcellular": "nucleus AND cytoplasm - nucleocytoplasmic shuttling. UniProt Q86Y26 lists Cytoplasm (SL-0086) and Nucleus (SL-0191), both ECO:0000269, with the note 'Shuttles between nucleus and cytoplasm'; the experiment behind all three is PMID 17934517, which shows native NUT carries both an NLS and a leptomycin-sensitive NES.",
  "subnuclear": "NOT ESTABLISHED for native NUTM1. The discrete nuclear foci, and the ~100 hyperacetylated megadomains up to 2 Mb that they correspond to, are reported for BRD4-NUT and BRD3-NUT, which are strictly nuclear because the BRD moiety retains them on chromatin (PMID 17934517, 26220994). BOUNDED: nothing read for this entry reports a subnuclear compartment for endogenous NUTM1 in a spermatogenic cell or in any lymphoid cell. Note the query hazard: NUTM1[tiab] AND localization[tiab] returns 2 records and NUTM1[tiab] AND nuclear[tiab] returns 102, the latter heavily contaminated by abstracts spelling out the gene's own former name 'nuclear protein in testis' (200 records for that phrase alone). Searched in titles and abstracts only; supporting figures not searched (rule 80).",
  "shown_for_fusion": "BRD4::NUTM1 and BRD3::NUTM1 - strictly nuclear, with discrete foci and megadomains; NOT shown for any NUTM1 fusion of B-ALL (BRD9, ACIN1, CUX1, ZNF618), and not for KMT2A::AFF1 or any KMT2A fusion, where NUTM1 plays no part.",
  "localization_shown_for_fusion": "BRD4::NUTM1, BRD3::NUTM1 (carcinoma); nobody has imaged BRD9::NUTM1 or ACIN1::NUTM1 in a B-cell progenitor"
}
epigenetic marks
[
  "H4K5ac",
  "H4K8ac",
  "H3K27ac",
  "H3K18ac",
  "DNA methylation (global, direction: DOWN / hypomethylation)"
]
intact record
{
  "endpoint": "https://www.ebi.ac.uk/intact/ws/interaction/findInteractions/Q86Y26",
  "fetched": "2026-09-22",
  "records_under_accession": 84,
  "named_human_partners_under_accession": 24,
  "evidence_classes_under_accession": [
    "two hybrid array",
    "two hybrid prey pooling approach",
    "validated two hybrid",
    "clash"
  ],
  "source_publications_under_accession": [
    "25416956",
    "31515488",
    "22810586",
    "36217029",
    "23622248"
  ],
  "ep300_crebbp_brd4_brdt_present_under_accession": false,
  "fusion_construct_interactor": "EBI-8101903 (name 'brd4-nut', Homo sapiens, 3 interactions)",
  "fusion_construct_records": [
    {
      "partner": "EP300",
      "accession": "Q09472",
      "method": "fluorescence imaging",
      "type": "colocalization",
      "host": "Homo sapiens HCC2429 lung non-small cell carcinoma cell",
      "pmid": "20676058"
    },
    {
      "partner": "H4C16",
      "accession": "P62805",
      "method": "fluorescence imaging",
      "type": "colocalization",
      "host": "Homo sapiens HCC2429 lung non-small cell carcinoma cell",
      "pmid": "20676058"
    },
    {
      "partner": "q6b828_chlae",
      "accession": "Q6B828",
      "method": "fluorescence imaging",
      "type": "colocalization",
      "host": "Cercopithecus aethiops simian cells transformed with SV40",
      "pmid": "20676058"
    }
  ],
  "reachability_note": "findInteractor?query=nut returns 26 interactors and finds EBI-8101903; findInteractor?query=brd4_nut returns 0. The construct is reachable under one spelling and not another, so this enumeration is a floor."
}
uniprot defects flagged not edited
[
  "FUNCTION rests on one ECO:0000269 citation, PMID 30447097, a HepG2 TERT-promoter shRNA screen filed under the retired symbol C15orf55. The founding physiological-function paper with a knockout phenotype, PMID 30257209, is absent from the entry's nine-reference list (checked programmatically against the live record).",
  "All three localization statements (Cytoplasm, Nucleus and the shuttling note) cite PMID 17934517, a study of BRD-NUT FUSION oncoproteins in carcinoma cells. The native-NUT shuttling result inside it is real, so the annotation is defensible, but the citation is a fusion paper standing in for the native protein's address - rule 91's shape.",
  "No domain or motif feature is annotated, so the two experimentally mapped, structurally solved p300-binding TADs (403-418 and 419-470; PDB 7XEZ/7XFG) are invisible to anyone reading the feature table."
]
enumeration record counts
{
  "tool": "PubMed E-utilities esearch, one tag per term (rule 94), querytranslation read back from every response (registry row 123)",
  "date": "2026-09-22",
  "NUTM1[tiab]": {
    "count": 304,
    "querytranslation": "\"NUTM1\"[Title/Abstract]"
  },
  "C15orf55[tiab]": {
    "count": 3,
    "querytranslation": "\"C15orf55\"[Title/Abstract]"
  },
  "NUT[tiab]": {
    "count": 9391,
    "querytranslation": "\"NUT\"[Title/Abstract]"
  },
  "\"nuclear protein in testis\"[tiab]": {
    "count": 200,
    "querytranslation": "\"nuclear protein in testis\"[Title/Abstract]"
  },
  "NUTM1[tiab] AND nuclear[tiab]": {
    "count": 102,
    "querytranslation": "\"NUTM1\"[Title/Abstract] AND \"nuclear\"[Title/Abstract]",
    "read_as": "CONTAMINATED - the gene's former name contains the query term"
  },
  "NUTM1[tiab] AND localization[tiab]": {
    "count": 2,
    "querytranslation": "\"NUTM1\"[Title/Abstract] AND \"localization\"[Title/Abstract]"
  },
  "NUTM1[tiab] AND p300[tiab]": {
    "count": 6,
    "querytranslation": "\"NUTM1\"[Title/Abstract] AND \"p300\"[Title/Abstract]"
  },
  "NUT[tiab] AND p300[tiab]": {
    "count": 25,
    "querytranslation": "\"NUT\"[Title/Abstract] AND \"p300\"[Title/Abstract]",
    "read_as": "the p300 literature is written under NUT, not NUTM1 - 25 against 6, and every founding paper is in the 25"
  },
  "NUTM1[tiab] AND KMT2A[tiab]": {
    "count": 15,
    "querytranslation": "\"NUTM1\"[Title/Abstract] AND \"KMT2A\"[Title/Abstract]"
  },
  "NUTM1[tiab] AND MLL[tiab]": {
    "count": 2,
    "querytranslation": "\"NUTM1\"[Title/Abstract] AND \"MLL\"[Title/Abstract]",
    "read_as": "registry row 124 satisfied - MLL was run as well as KMT2A; here the union adds little, because the NUTM1 literature postdates the rename"
  },
  "BRD9[tiab] AND NUTM1[tiab]": {
    "count": 2,
    "querytranslation": "\"BRD9\"[Title/Abstract] AND \"NUTM1\"[Title/Abstract]"
  },
  "control_genuine_zero TMEM103[tiab]": {
    "count": 0,
    "querytranslation": "TMEM103[tiab]",
    "read_as": "term echoed back UNPARSED - a real zero, not an instrument failure"
  },
  "control_parsed_zero SLC12A6[tiab] AND NUTM1[tiab]": {
    "count": 0,
    "querytranslation": "\"SLC12A6\"[Title/Abstract] AND \"NUTM1\"[Title/Abstract]",
    "read_as": "both operands translate; the conjunction is genuinely empty"
  },
  "NUTM1[tiab] AND menin[tiab]": {
    "count": 0,
    "querytranslation": "\"NUTM1\"[Title/Abstract] AND \"menin\"[Title/Abstract]",
    "read_as": "GENUINE ZERO - parsed translation, both operands resolve, run twice with the same result"
  },
  "NUTM1[tiab] AND MEN1[tiab]": {
    "count": 0,
    "querytranslation": "\"NUTM1\"[Title/Abstract] AND \"MEN1\"[Title/Abstract]",
    "read_as": "GENUINE ZERO, run twice. With the menin query above, the two names together bound the absence of any menin-NUTM1 literature in titles and abstracts"
  },
  "NUTM1[tiab] AND DOT1L[tiab]": {
    "count": 0,
    "querytranslation": "\"NUTM1\"[Title/Abstract] AND \"DOT1L\"[Title/Abstract]",
    "read_as": "GENUINE ZERO, run twice"
  },
  "NUTM1[tiab] AND H3K79[tiab]": {
    "count": 0,
    "querytranslation": "\"NUTM1\"[Title/Abstract] AND \"H3K79\"[Title/Abstract]",
    "read_as": "GENUINE ZERO, run twice - the index disease's mark and this gene share no title-or-abstract literature at all"
  },
  "double_run_note": "Every zero above was run twice through the identical code path, minutes apart, and returned the same count with the same non-empty parsed querytranslation (cycle-54 registry row 123). The positive counts were re-run too and were identical. Rule 80 applies to all of them: these bound TITLES AND ABSTRACTS, not papers - supporting figures and tables were not searched, and one measured instance of that limit is recorded in disease-nutm1-r-b-all (SLC12A6 is a genuinely reported NUTM1 partner and SLC12A6[tiab] AND NUTM1[tiab] is a true zero)."
}
How this could bite us back
  • The obvious drug is not against NUTM1. NUTM1 is 1,132 disordered residues with no pocket; the tractable nodes are p300's HAT domain, its TAZ2 surface, and the BET reader. All three are required in normal cells - p300/CBP loss is developmentally lethal and BET inhibition has dose-limiting thrombocytopenia - so any therapeutic window comes from the fusion's addiction, not from target selectivity.
  • Targeting the 5' partner is worse than it looks in this family. BRD9 has clinical-stage degraders, but BRD9::NUTM1 retains only the 5' portion of BRD9, and a degrader recruits an E3 to a ligand-binding surface the fusion may not retain. The BRD9 breakpoints in infants sit at exons 3, 8 and 14, so which BRD9 elements survive differs between patients: a degrader could hit wild-type BRD9 in every cell and miss the fusion in some.
  • This disease is already cured by chemotherapy in the great majority of cases - 4-year EFS 100% in 35 uniformly treated infants, all reaching CR, 91% with MRD under 0.05%, none transplanted in first remission. The clinically useful direction is therefore treatment REDUCTION, and any targeted agent has to beat a near-perfect standard arm. An intervention that improves a mechanism and worsens toxicity is a net harm here, which is the reverse of the KMT2A-rearranged calculation.
  • The cases that would make a targeted agent necessary are the exceptions, and those are the ones with the least mechanism behind them: a ZNF618::NUTM1 with a poor outcome, a CUX1::NUTM1 adult who relapsed early, two paediatric cases with high-risk features. The HOXA9-negative partner group is where the failures cluster and is also where the biology is least studied.
  • If the chemosensitivity really is a dependence on active transcription, then any agent that slows transcription - a BET inhibitor, a CDK9 inhibitor, a p300 inhibitor, all of which are the mechanistically obvious choices here - could ANTAGONISE the chemotherapy that currently cures these children. That is a specific, testable and unflattering prediction of the stated mechanism, and it should be tested in combination before any of these agents is given alongside induction.
Still open
  1. What makes the NUTM1-r methylome globally hypomethylated, and why is that independent of the 5' partner when the HOXA phenotype is not? NUTM1 has no DNMT or TET module and no reported contact with one. The discriminating experiment - DNMT/TET occupancy or activity in a NUTM1-fusion-expressing B progenitor against an isogenic control - has not been done.
  2. Is a NUTM1-rearranged blast sensitive to menin inhibition? Boer 2021 raises exactly this question and nobody has answered it. The HOXA9-positive partner group (ACIN1, CUX1, BRD9, AFF1) runs a posterior-HOXA programme without a KMT2A fusion, so the answer decides whether HOXA dependence or the KMT2A-menin interface is what menin inhibitors actually exploit.
  3. Does native NUTM1's nucleocytoplasmic shuttling matter in the fusion? The NES sits in the NUTM1 moiety and is retained in essentially every fusion, yet BRD-NUT fusions are strictly nuclear. Whether a non-BET 5' partner such as ACIN1 or CUX1 overrides that NES as effectively as a double bromodomain does is unknown, and it is a testable determinant of how much fusion protein ever reaches chromatin.
  4. Is the p300 TAZ2 interface druggable where the NUTM1 moiety is concerned? Two short helices binding a four-helix bundle through a consensus motif shared with p53 TAD2 and p63 TAD is a shallow, promiscuous interface - and its promiscuity is itself the pharmacological problem.
  5. Why is NUTM1-r leukaemia so chemosensitive? The Blood 2026 answer, a dependence on active transcription, predicts that transcription-directed agents should be the effective ones and that a transcription-sparing regimen should be worse - an actionable and falsifiable prediction no trial has tested.
Every source behind this entry (11)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProtKB Q86Y26 (NUTM1_HUMAN, NUT family member 1) is 1132 aa, annotates eight disordered regions and no structured domain, carries exactly one Pfam family PF12881 (NUT), is localized to Cytoplasm and Nucleus with a shuttling note (all ECO:0000269 on PMID 17934517), is specifically expressed in testis (ECO:0000269, PMID 12543779), and rests its FUNCTION field on one citation only (PMID 30447097).

    What this evidence cannot show

    Fetched live 2026-09-22 by gene_exact:NUTM1 AND organism_id:9606 AND reviewed:true, one reviewed hit, accession and length 1132 read back from the returned record. The eight disordered regions are ECO:0000256, i.e. automatic sequence analysis, not experiment. The absence of a domain annotation is an absence in the database and not proof that no structured module exists; it is corroborated here by the PONDR disorder prediction and by the helix-on-binding behaviour of the two TADs in PMID 36690674 [E2], which is the signature of an intrinsically disordered activator rather than a folded domain.

    Evidence that argues against this (1)
    • UniProt's nine-reference list does not include PMID 30257209 or PMID 20676058, the two papers that establish this protein's function and its p300 recruitment. An architecture read only off this record would be complete and an activity read only off it would be badly wrong. Verified programmatically against the live entry. Shiota H et al. 2018
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    NUT contains two transactivation domains, TAD1 (residues 403-418) and TAD2 (419-470), each of which binds the p300 TAZ2 domain as an amphipathic helix; the bipartite TAD/TAZ2 contact allosterically activates p300 and drives acetylation-dependent liquid-like condensation on chromatin comprising H3K27 and H3K18 acetylation together with BRD4L/S, CDK9, MED1 and RNA polymerase II.

    PMID 36690674DOI
    What this evidence cannot show

    NMR structures 7XEZ and 7XFG (both verified live against the RCSB entry API 2026-09-22: 'NMR solution structures of p300 TAZ2 domain in complex with BRD4-NUT F1c domain binding motif #2' and '#1'), BMRB 36480/36481. The residue numbering is internally inconsistent in the paper and this entry uses the main-text definition: Figure 1d's legend gives TAD1 as 377-418, which are GST construct boundaries, while the main text defines TAD1 as 403-418 and validates that motif by mutation (L410G/L411G/Y413G/I414G abrogate binding); Figure 2e's alignment window narrows to 405-414 and 430-458, and the TAD2 helices resolved in the structure are 433-440 and 448-460. Peptide-scale NMR plus in-vitro droplets, in a carcinoma context, not a leukaemia one; the inference that the same TADs operate in a B-ALL fusion rests on the fact that nearly the entire NUTM1 open reading frame is retained in those fusions, and that inference is ours.

    Evidence that argues against this (1)
    • A separate 2023 study concludes that the molecular features driving condensate formation and those driving gene expression by BRD4-NUT are 'overlapping but distinct', so condensation and transcriptional output must not be treated as one readout. Scientific Reports 2023
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Nut expression is normally restricted to post-meiotic spermatogenic cells, where it uses p300 and/or CBP to raise H4 acetylation at K5 and K8 genome-wide, creating binding sites for the first bromodomain of the testis BET protein BRDT that then mediates histone removal; inactivation of Nut causes male sterility with spermatogenesis arrested at the histone-removal stage.

    Nut Directs p300-Dependent, Genome-Wide H4 Hyperacetylation in Male Germ Cells · Shiota H, Barral S, Buchou T, et al. · Cell Reports 2018
    PMID 30257209DOI
    What this evidence cannot show

    Mouse germ cells. This is the physiological function and the only knockout phenotype, and it is in a cell type with no relationship to B-lymphopoiesis; the marks it establishes (H4K5ac/H4K8ac) are NOT the marks reported in the leukaemia (H3K27ac plus global DNA hypomethylation) [E6], so the germ-cell mechanism should be read as the template for the loop and not as the leukaemic readout.

    Evidence that argues against this (1)
    • UniProt's FUNCTION field for Q86Y26 records none of this and instead cites a hepatocellular-carcinoma shRNA screen placing the gene on the TERT promoter via SP1. Two non-overlapping accounts of what this protein does, one of them the database's. Amisaki M et al. 2019
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In the fusion, after the BRD4 bromodomains bind acetylated chromatin, the NUT moiety strongly interacts with and recruits p300, stimulates its catalytic activity and initiates feed-forward cycles of recruitment that create hyperacetylated chromatin domains; sequestration of p300 into BRD4-NUT foci is the principal oncogenic mechanism and inactivates p53, with knockdown releasing p300 and restoring differentiation and apoptosis.

    Oncogenesis by sequestration of CBP/p300 in transcriptionally inactive hyperacetylated chromatin domains · Reynoird N, Schwartz BE, Delvecchio M, et al. · EMBO Journal 2010
    PMID 20676058DOI
    What this evidence cannot show

    A patient-derived NUT carcinoma line, not leukaemia, and the 5' partner is BRD4 rather than BRD9. The paper's 'transcriptionally inactive' framing was later refined - the megadomain literature reports active transcription of the underlying DNA within the hyperacetylated expanses - so the two descriptions of the same structures are not identical and this entry does not adopt the inactive framing as settled.

    Evidence that argues against this (1)
    • The 2015 megadomain study describes the same hyperacetylated expanses as driving transcription of the underlying DNA, i.e. transcriptionally ACTIVE, which is the opposite characterisation of the same structures. Alekseyenko AA et al. 2015
  5. E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    BRD4-NUT nuclear foci correspond to roughly 100 hyperacetylated chromatin expanses of up to ~2 Mb ('megadomains') that originate from selected pre-existing enhancers, broaden progressively and are delimited by topologically associating domain boundaries; megadomain locations are largely cell-lineage-specific, with the MYC and TP63 regions targeted in all NUT carcinomas tested.

    The oncogenic BRD4-NUT chromatin regulator drives aberrant transcription within large topological domains · Alekseyenko AA, Walsh EM, Wang X, et al. · Genes & Development 2015
    PMID 26220994DOI
    What this evidence cannot show

    NUT carcinoma patient cells and naive cells induced to express BRD4-NUT. Whether megadomains of this scale form in a NUTM1-rearranged B-ALL blast, with a 5' partner that is not a BET protein, is NOT established by this paper and is not established anywhere read for this entry - the leukaemia evidence is a global H3K27ac rise and new open chromatin regions [E6], which is compatible with megadomains but does not demonstrate them.

    Evidence that argues against this (1)
    • Megadomain targeting is reported to be cell-state-dependent, so a lineage-specific target list obtained in carcinoma cells does not transfer to a B-cell progenitor. Genetics 2023
  6. E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    NUTM1-rearranged leukaemia is a discrete entity with a unique transcriptional and epigenetic landscape featuring global DNA hypomethylation irrespective of the 5' fusion partner; NUTM1 fusions drive B-lymphoid lineage commitment while conferring leukaemic stem cell properties, establish an aberrant chromatin state marked by global enhancement of H3K27 acetylation and distinctive open chromatin regions co-opting B-lineage and stemness programmes including NF-kB and posterior HoxA genes, and BRD9-NUTM1 alone induces serially transplantable prepro-B-like leukaemia in vivo; the cells are profoundly chemosensitive, mechanistically linked to a dependence on active transcription.

    Comprehensive molecular and functional analysis of NUTM1-rearranged leukemia · Nishimura K, Isobe T, Shigehiro T, et al. · Blood 2026
    PMID 41460961DOI
    What this evidence cannot show

    Read from the published abstract, not the full text; the per-partner n behind the hypomethylation claim, the assay used for 'global' DNA methylation, and the definition of 'profound sensitivity to chemotherapy' are all unread here and should be checked before any of them is quoted as a number. 'Irrespective of the 5-prime fusion partner' is the paper's own wording and is a strong claim resting on however many partners were profiled. The mouse model uses a single representative fusion, BRD9-NUTM1.

    Evidence that argues against this (1)
    • The same journal carries an editorial comment on this paper which this entry has not read and which may qualify its conclusions; recorded so a reader knows the commentary exists rather than being told the primary paper stands unchallenged. Shiloh R et al. 2026
  7. E7strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Native NUT contains both nuclear localization and nuclear export sequences and shuttles between nucleus and cytoplasm by a leptomycin-sensitive pathway, whereas BRD3-NUT and BRD4-NUT are strictly nuclear, implying that the BRD moiety retains NUT in the nucleus through chromatin interaction; FRAP shows the fusions diffuse laterally in the nucleus significantly more slowly than NUT, and siRNA withdrawal of either fusion produces squamous differentiation and cell-cycle arrest.

    PMID 17934517DOI
    What this evidence cannot show

    This is the paper UniProt cites for all three of its localization statements about NUTM1, and it is a study of the fusion oncoproteins in carcinoma cells - the native-NUT shuttling result is one experiment inside it. No localization experiment on endogenous NUT in a spermatogenic cell, and none in any lymphoid cell, is reported here or anywhere else read for this entry. The paper predates the NUTM1 symbol and uses NUT throughout, which is why it is reachable under the old symbol and not the new one.

    Evidence that argues against this (1)
    • The gene's own former name is 'nuclear protein in testis', and the 2003 founding paper says NUT 'is predicted to encode a nuclear protein' - a prediction, taken from the name, five years before the shuttling experiment. The naming has very likely propagated a nuclear-only reading of a protein the experiment calls nucleocytoplasmic. French CA et al. 2003
  8. E8strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In 85 NUTM1-rearranged B-ALL cases collected by the Ponte di Legno Childhood ALL Working Group and Interfant-99/-06, NUTM1 rearrangement accounted for 35 of 161 evaluated KMT2A-wildtype Interfant infants (21.7%); 4-year event-free survival was 100% in the 35 uniformly treated NUTM1-rearranged infants against 74.0% (95% CI 65.1-81.0) in 126 NUTM1-wildtype/KMT2A-wildtype infants (p=0.001), all cases reached complete remission, 41 of 45 (91.1%) had MRD below 0.05% against 48% in KMT2A-rearranged infant ALL, and none received a transplant in first remission; ten partners were identified among 71 cases, with ACIN1 (44%), BRD9 (26%) and CUX1 (15%) commonest in infants, all three pro-B-immunophenotype infants carried BRD9-NUTM1, and the ACIN1-NUTM1 BLUEPRINT case showed HOXA9/HOXA10 expression comparable to two KMT2A-rearranged cases with reduced CpG methylation and high H3K4me3/H3K27ac over that locus.

    PMID 34211097DOI
    What this evidence cannot show

    Read in full from the Europe PMC open-access full text (PMC8478641); PubMed carries NO abstract for this letter, so an abstract-only workflow would have missed every number in it. Retrospective, with cases assembled across many protocols over 1986-2019 for the Ponte di Legno arm; only the 35 Interfant infants are uniformly treated, and the 100% EFS in that arm is an n=35 Kaplan-Meier estimate with no events, so its lower confidence bound is not 100%. The comparator for the 74.0% figure is NUTM1-wildtype/KMT2A-wildtype infants, NOT KMT2A-rearranged infants - the KMT2A-rearranged contrast in this entry comes from the 48% MRD figure and from an independent cohort, and conflating the two comparators would overstate the contrast.

    Evidence that argues against this (1)
    • NUTM1-rearranged B-ALL is not uniformly favourable at the case level: a ZNF618::NUTM1 case is reported explicitly as challenging the prognostic expectation with a poor outcome, a CUX1::NUTM1 adult case relapsed early and is described as part of a recurring pattern of poor prognosis in adults, and two paediatric cases presented with high-risk features. Boer 2021 also states NUTM1 fusions had not been reported in adults, which two subsequent adult case reports contradict. Chen X et al. 2025
  9. E9established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    IntAct REST returns 84 interaction records for Q86Y26, all of them two-hybrid variants or CLASH from ORFeome-scale screens and none of them EP300, CREBBP, BRD4 or BRDT; the NUT-EP300 record exists in IntAct only under the fusion-construct interactor EBI-8101903 'brd4-nut', which carries three fluorescence-imaging colocalization records from PMID 20676058 (EP300 in HCC2429 cells, histone H4C16, and one Cercopithecus aethiops record).

    What this evidence cannot show

    Queried live 2026-09-22 through two endpoints: findInteractions on the accession (84 records, totalElements read back from the response) and findInteractor on the free-text token 'nut' (26 interactors, from which EBI-8101903 was identified by name), then findInteractions on that interactor id (3 records). The construct search was run as 'nut' and as 'brd4_nut', the latter returning 0 - so the construct is reachable under one spelling and not another, and this enumeration is a FLOOR on how many NUT fusion constructs IntAct holds, exactly as cycle 54's 47/18 figure was a floor. The three brd4-nut records are typed 'colocalization', not MI:0407 direct interaction, so even the fusion-side record is an imaging co-localisation rather than a binding assay; the binding evidence is [E2] and [E10].

    Evidence that argues against this (1)
    • The absence could in principle be correct rather than an instrument artefact, if no curator had ever deposited the NUT-p300 contact. It is not: the record exists in the same database under the fusion-construct interactor, which settles the question and makes this a filing problem rather than a curation gap. Reynoird N et al. 2010
  10. E10strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Cross-linking, affinity purification and mass spectrometry identified EP300 as uniquely associated with BRD4 through the NUT fusion in both NUT carcinoma and non-NUT-carcinoma cell types, and ZNF532 as associated with BRD4-NUT in patient cells; ChIP sequencing of NUT, ZNF532, BRD4, EP300 and H3K27ac showed hyperacetylated megadomains, and a ZNF532-NUT translocation was independently found in a newly diagnosed patient.

    Ectopic protein interactions within BRD4-chromatin complexes drive oncogenic megadomain formation in NUT midline carcinoma · Alekseyenko AA, Walsh EM, Zee BM, et al. · Proceedings of the National Academy of Sciences 2017
    PMID 28484033DOI
    What this evidence cannot show

    NUT carcinoma and 293T cells; the comparison is BRD4-NUT against wild-type BRD4, so it establishes that the NUT moiety brings EP300 into a BRD4 complex, not that native NUT binds EP300 in a cell that normally expresses it. None of these records is in IntAct under NUTM1's accession [E9].

    Evidence that argues against this (1)
    • ZNF532 was detectable in patient cells but not in 293T, so the composition of the NUT-nucleated complex is cell-context-dependent and a partner list from one system does not transfer - including to a B-cell progenitor, where no such purification has been done. Alekseyenko AA et al. 2017
  11. E11moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    A genome-wide shRNA screen in HepG2 cells identified C15orf55 (the retired symbol for NUTM1) as a positive regulator of TERT expression, acting by binding two GC motifs in the SP1 binding site of the TERT promoter between base pairs -58 and +36; this is the only citation behind UniProt's FUNCTION annotation for Q86Y26.

    Identification of genes involved in the regulation of TERT in hepatocellular carcinoma · Amisaki M, Tsuchiya H, Sakabe T, et al. · Cancer Science 2019
    PMID 30447097DOI
    What this evidence cannot show

    Hepatocellular carcinoma cell line plus a TCGA survival correlation, published under the retired symbol C15orf55, which is why it is reachable only by a query this beat would otherwise never run (C15orf55[tiab] returns 3 records in total). Nothing connects this activity to leukaemia or to the p300 loop, and it is recorded here because it is what the database says the gene does, not because this entry endorses it as the function.

    Evidence that argues against this (1)
    • The physiological function established by knockout is p300-dependent H4 hyperacetylation in spermatogenesis, which this paper does not mention and which UniProt does not cite. Rousseaux S et al. 2022
Connected to