Oncofusion

start / graph / interface-menin-pias4-declined-the-cited-paper-is-about-parp1-and-heat-shock

Interfaceweak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.q160-the-prey-only-interactome

MEN1-PIAS4 - DECLINED. The single supporting publication is a PARP-1 heat-shock sumoylation study in which menin is never named, and the row is a HeLa co-IP passenger

also called menin-PIASy · MEN1-PIAS4 · menin sumoylation

A database lists the SUMO-attaching enzyme PIASy as a menin partner. The only paper behind it is about how heat shock changes PARP-1, and menin is not mentioned in it at all.

The same thing, for a scientist

**Verdict: DECLINED, on the same shape rule 91 names for UniProt localization citations - the pointer is real and the paper is about something else.** IntAct holds one MEN1-PIAS4 record: anti-tag co-IP, *physical association*, HeLa, MEN1 as prey, intact-miscore 0.40, from **PMID 19779455** [E1]. That paper is Martin, Schwamborn, Schreiber, Werner, Guillier, Zhang, Bischof, Seeler and Dejean, *EMBO J* 2009: *PARP-1 transcriptional activity is regulated by sumoylation upon heat shock*. Its subject is the PIASy-PARP-1 axis, heat-shock-induced poly-sumoylation of PARP-1, recruitment of PIASy, SUMO and Ubc9 to the **HSP70.1** promoter, and RNF4-mediated ubiquitination of PARP-1 [E2]. **Menin appears nowhere in the abstract, the title, the MeSH terms or the keywords.** The row is a menin peptide recovered in a PIASy pull-down from HeLa lysate. **This is declined and it is the most interesting decline in the beat**, because PIAS4 is the one prey-only partner whose *localization* is a named subnuclear body - UniProt gives **Nucleus, PML body** with three experimental evidence codes [E3] - and a menin-SUMO link would connect this disease to PML::RARA's biology by more than analogy. **The corpus already carries a SUMO thread on the LEDGF side (`gene-sumo1`) and the PML::RARA compartment case in full (`localization-pml-nuclear-bodies-pmlrara`); what it does not carry is any claim that menin or the fusion itself is SUMOylated, or any PIAS or SENP entry** [E4].

Go deeper
Why it is declined
criterionverdict
independent replicationno - one record, one publication, one method
the source study is about this pairno - PIASy-PARP-1 under heat shock; menin is not named
endogenous on the menin sideunknown; the row is anti tag coip with PIASy tagged
expression / compartment compatibleyes - both nuclear, HeLa expresses both
a reciprocal menin-bait experiment recovers PIAS4no - none exists in the record set

The compartment and expression checks that killed interface-menin-cd226-declined-proteome-scale-ap-ms both pass here. The decline rests entirely on the last three rows: one record, from a paper about a different protein pair, with no reciprocal.

What would reverse it in one experiment: a menin immunoprecipitation from any cell blotted for SUMO2/3, or a menin-bait AP-MS recovering PIAS4. Neither exists.

The reason to keep this entry, and a false absence I wrote and had to withdraw

PML nuclear bodies are a subnuclear compartment this project already works on, and SUMO is a thread it already carries.

WITHDRAWN 2026-09-24, before shipping. This section first asserted that nothing in the corpus connects SUMO to this disease, and that the question of whether SUMO matters here had never been asked. Both were false, and one grep found it. gene-sumo1 records that LEDGF/PSIP1 is constitutively SUMOylated at K75, K250, K254 and K364, that deSUMOylation by SENP1 or the K364R mutation raises LEDGF's abundance and transactivation capacity, and that the tag is therefore a negative regulator of LEDGF dose - which is what the KMT2A fusion depends on. localization-pml-nuclear-bodies-pmlrara works the PML::RARA compartment case in full, including the finding that reassembling the bodies was not the cure 4.

What IS absent is narrower and is still worth stating:

  1. No entry asserts that menin, or a KMT2A fusion protein itself, is SUMOylated. The corpus's SUMO thread runs entirely through LEDGF and through the PML::RARA analogy; neither reaches menin or the oncoprotein.
  2. No PIAS-family SUMO E3 ligase and no SENP isopeptidase has an entry. gene-pias1 through gene-pias4 and every gene-senp* are absent from disk, checked this cycle, while gene-sumo1 exists - the graph carries the tag and not the machinery that puts it on or takes it off 4.

So the question this decline leaves is not 'does SUMO matter here'. The corpus already argues that it does, on the LEDGF side. It is: does the SUMO machinery touch menin or the oncoprotein, and would a menin IP blotted for SUMO2/3 show anything? That experiment does not exist, and this PIAS4 row is the only pointer towards it anywhere in the interactome.

And gene-sumo1 supplies an argument against this row that this entry had not used: it reasons that a SUMO interaction record is an enzyme-substrate relationship with a transient intermediate, not a complex. That applies to PIAS4-menin directly and is an independent third reason to decline.

Facts and figures
partner agene-men1
partner bPIAS4
partner b uniprotQ8N2W9
partner b length AA510
partner a uniprotO00255
partner a length AA610
verdictDECLINED
adjudicated byq160, cycle 146, 2026-09-24
druggabilityNot assessable as an interface. The adjacent druggable idea is SUMO-targeted degradation, the mechanism of arsenic trioxide in PML::RARA - which has never been posed for a KMT2A fusion because nobody has shown one is sumoylated.
n methods1
n platforms1
n publications1
endogenous sideno - PIASy is the tagged side
intact miscore0.4
cell statethe source study's manipulated state is HEAT SHOCK; the IntAct row does not record which condition it came from
epigeneticsSUMO E3 ligase; in the source study PIASy, SUMO and Ubc9 are recruited to and released from the HSP70.1 promoter with PARP-1 kinetics. No chromatin readout involving menin exists.
absence classnever-measured
absence class noteNo menin immunoprecipitation has been blotted for SUMO2/3 in any published work found here, and no PML-body colocalisation with menin or with a KMT2A fusion has been reported. The LEDGF side is already covered by gene-sumo1; the menin and fusion sides are not.
interface residues a
[]
structures
[]
ligands
[
  "drug-revumenib",
  "drug-ziftomenib"
]
records
[
  "MEN1-PIAS4, anti tag coip, physical association, HeLa, MEN1 prey, miscore 0.40, PMID 19779455"
]
cell systems
[
  "HeLa"
]
localization
{
  "compartment": "nucleus",
  "subnuclear": [
    "PML body (ECO:0000269, three codes)"
  ],
  "shown_for_fusion": "NO - nobody has looked in any fusion. The corpus DOES carry SUMO biology via gene-sumo1 (LEDGF SUMOylation at K75/K250/K254/K364) and localization-pml-nuclear-bodies-pmlrara, but nothing asks whether menin or a KMT2A fusion is SUMOylated or visits a PML body."
}
measurements
[
  {
    "quantity": "count",
    "subject": "entries in this corpus asserting SUMOylation of menin or of a KMT2A fusion protein",
    "value": 0,
    "units": "entries",
    "n": 2931,
    "universe": "sentence-level scan of all node files outside q160 for a SUMOylation term co-occurring with menin/MEN1 or with a KMT2A fusion, 2026-09-24; 27 files mention SUMO at all, and gene-sumo1 carries a full LEDGF SUMOylation account",
    "null": null,
    "note": "CORRECTED before shipping. This measurement first read 'entries connecting SUMO to menin or a KMT2A fusion = 0', which was FALSE: gene-sumo1 and localization-pml-nuclear-bodies-pmlrara both connect SUMO to this disease. The surviving zero is the narrow one - no entry says MENIN or the FUSION ITSELF is SUMOylated. A bounded absence within this corpus, not within the literature."
  },
  {
    "quantity": "count",
    "subject": "PIAS-family SUMO E3 ligase or SENP isopeptidase entries in this graph",
    "value": 0,
    "units": "entries",
    "n": 2957,
    "universe": "id listing of data/graph/nodes/*/ for gene-pias* and gene-senp*, 2026-09-24",
    "null": null,
    "note": "gene-sumo1 exists; the machinery that conjugates and removes the tag does not."
  }
]
How this could bite us back
  • If menin is a SUMO substrate, SUMO-pathway inhibitors now in trials would be modifying the target of revumenib and ziftomenib in an uncontrolled way, in patients who might receive both.
  • Declining this row must not be read as evidence that SUMO is irrelevant here. The correct statement is that the question has never been asked, which is a much weaker and much more actionable claim.
Still open
  1. Is menin SUMOylated? A menin IP blotted for SUMO2/3 in one KMT2A-rearranged line answers it. gene-sumo1 shows the corpus already knows how to reason about a SUMO row; it has never had a menin one to reason about.
  2. Does menin or a KMT2A fusion visit PML nuclear bodies? localization-pml-nuclear-bodies-pmlrara works the PML::RARA compartment case in full and concludes that reassembling the bodies was NOT the cure - so the transfer is not automatic, and the prior question of whether a KMT2A fusion is anywhere near a PML body has never been asked.
  3. Does the supplementary material of PMID 19779455 list menin? This decline rests on the abstract and MeSH only, and the tables were not opened.
  4. Why does this graph carry gene-sumo1 but no PIAS ligase and no SENP isopeptidase? The tag is written up and the enzymes that add and remove it are absent, which leaves any SUMO argument here unanchored on the machinery side.
Every source behind this entry (4)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    IntAct REST returns one MEN1-PIAS4 (O00255-Q8N2W9) record: anti-tag co-immunoprecipitation, interaction type 'physical association', host Homo sapiens HeLa epitheloid cervical carcinoma cell, MEN1 as PREY, intact-miscore 0.40, from PMID 19779455. There is no reciprocal MEN1-bait record for this pair.

    What this evidence cannot show

    One record. IntAct does not label an expansion method on this row, but a co-IP followed by MS in a paper reporting a multi-protein sumoylation pathway may well have been spoke-expanded by the depositor without the flag being set, in which case the binary pair was never measured.

    Evidence that argues against this (1)
    • A miscore of 0.40 is above several rows this beat treats more generously, so declining on the record alone is not consistent scoring. The decline rests on the source study's subject, not on the score. IntAct (EMBL-EBI) 2026
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    PMID 19779455 (Martin et al., EMBO J 2009) reports that the SUMO E3 ligase PIASy interacts with PARP-1 and mediates heat-shock-induced poly-sumoylation of PARP-1; that PIASy is indispensable for full activation of the inducible HSP70.1 gene; that PIASy, SUMO and Ubc9 are rapidly recruited to the HSP70.1 promoter on heat shock and released with kinetics similar to PARP-1; and that RNF4 mediates heat-shock-inducible ubiquitination of PARP-1 and regulates its stability. Neither menin nor MEN1 appears in the title, abstract, keywords or MeSH terms of the record.

    PARP-1 transcriptional activity is regulated by sumoylation upon heat shock. · Martin N, Schwamborn K, Schreiber V, Werner A, Guillier C, Zhang XD, Bischof O, Seeler JS, Dejean A · EMBO J 2009
    PMID 19779455DOI
    What this evidence cannot show

    Abstract, title, keywords and MeSH read; the body and supplementary tables were NOT opened, so this establishes that menin is absent from the paper's stated subject, not that menin is absent from every table in it. `expected_absent_terms` carries the polarity: the absence IS the support for the decline.

    Evidence that argues against this (1)
    • A protein can be a genuine partner and still not reach the abstract. The MeSH list does include 'Transcriptional Activation' and 'Protein Processing, Post-Translational', and a sumoylation survey in that paper could legitimately have identified menin as a PIASy substrate without it being the headline. Only the supplementary tables would settle it, and they were not read. Martin N et al. 2009
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProtKB Q8N2W9 (PIAS4_HUMAN) is 510 aa. Its recorded subcellular location is Nucleus, PML body, with three experimental evidence codes (ECO:0000269). Tissue specificity: highly expressed in testis and, at lower levels, in spleen, prostate, ovary, colon and peripheral blood leukocytes.

    What this evidence cannot show

    Rule 91: the three ECO:0000269 codes mean curators read experiments, and none of those papers was opened here. PML-body residence is the only subnuclear annotation among this beat's declined partners and is recorded because it is the interesting fact, not because it supports the interaction.

    Evidence that argues against this (1)
    • PML-body residence is an argument for a compartment overlap only if menin visits PML bodies, and menin's own UniProt record carries a single undifferentiated 'Nucleus'. No colocalisation has ever been reported. The compartment argument here is one-sided. UniProt Consortium 2026
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset

    A scan of all node files outside this beat on 2026-09-24 finds 27 mentioning SUMO. gene-sumo1 records that LEDGF/PSIP1 is constitutively SUMOylated at K75, K250, K254 and K364, that deSUMOylation by SENP1 or the K364R mutation enhances LEDGF transactivation capacity and cellular abundance, and that the tag is a negative regulator of LEDGF dose; it also argues that a SUMO interaction record is an enzyme-substrate relationship rather than a complex. localization-pml-nuclear-bodies-pmlrara works the PML::RARA nuclear-body case in full. No entry asserts that menin or a KMT2A fusion protein is SUMOylated, and no gene-pias* or gene-senp* entry exists on disk.

    What this evidence cannot show

    A sentence-level regex over serialised JSON splits badly on quoted strings, so the co-occurrence count is approximate. What was verified by hand is that gene-sumo1 and localization-pml-nuclear-bodies-pmlrara exist and say what is quoted here, and that no gene-pias* or gene-senp* id is on disk. The claim about menin and the fusion is an absence WITHIN THIS CORPUS and was NOT tested against PubMed.

    Evidence that argues against this (1)
    • This whole evidence item exists because the first version of this entry asserted the opposite - that nothing in the corpus connects SUMO to this disease. A reader should treat the surviving narrow zero with the same suspicion: it rests on a regex over JSON, and the broad version of the same claim was wrong. this project 2026
Connected to