Menin-FANCD2: one 2003 co-immunoprecipitation, a curated sufficient region of menin 214-390, and nothing since
also called menin-FANCD2 · MEN1-FANCD2 · menin Fanconi interaction
In 2003 one laboratory showed that menin and the Fanconi repair protein FANCD2 can be pulled out of a cell together, and that the two stick together more after radiation. Nobody has repeated that experiment in a human blood cell, nobody has solved a structure of the pair, and the only later study designed to test whether the two proteins actually work together concluded that they do not.
The same thing, for a scientist
The entire protein-level evidence for a menin-FANCD2 interaction is PMID 12874027 (Jin, Mao, Schnepp, Sykes, Silva, D'Andrea, Hua; Cancer Res 2003) [E1]. The abstract states that menin specifically interacts with FANCD2, that the interaction is enhanced by gamma-irradiation, that menin-null mouse embryonic fibroblasts are more sensitive to DNA damage, and that menin is found on chromatin and in the nuclear matrix with the nuclear-matrix fraction increasing after irradiation [E1]. IntAct curates four experiments from that one paper: anti-tag co-immunoprecipitation on a FLAG-tagged construct read out by western blot, the reciprocal anti-tag pull with FANCD2 identified by mass spectrometry (peptide 10-22), a second anti-tag experiment, and one anti-bait co-immunoprecipitation [E2]. From that curation the single most useful structural fact is a **sufficient binding region on menin of residues 214-390** [E2]. No domain was mapped on FANCD2; no structure exists; no affinity was measured; no stoichiometry was established. The 2008 follow-up that tested the functional prediction is `mech-men1-and-fancd2-act-in-distinct-crosslink-repair-processes`, and it is negative. This entry exists to fix the resolution of the claim, because the gap between *an association was detected in mouse fibroblasts in 2003* and *menin acts through the Fanconi pathway in KMT2A-rearranged leukaemia* is the whole width of the hypothesis it is being used to support.
What the 2003 paper actually establishes, and at what resolution
Four propositions, each graded separately, because they are not equally strong.
| proposition | grade | why |
|---|---|---|
| menin and FANCD2 co-immunoprecipitate in that system | established | it is a direct observation in the reporting system, which is what the grading rule makes established |
| the association increases after gamma-irradiation | established | same; a within-experiment comparison |
| menin-null MEFs are more sensitive to DNA damage | established | a phenotype, and it is independently corroborated by the DEB hypersensitivity of MEN1 patient lymphocytes 3 |
| menin functions in DNA repair through FANCD2 | weak | the paper's own word is "suggest"; the association is correlative, and the designed test of it later failed 4 |
The fourth row is the one that gets cited as if it were the first three.
Cells. The abstract names mouse embryonic fibroblasts for the sensitivity experiment. The 2008 paper, discussing this work, states that "these immunoprecipitation assays were performed in MEFs" 4. So the co-IP system is mouse fibroblast, not human, and not haematopoietic.
Tagged or endogenous. IntAct's feature annotation records a FLAG tag on the bait in three of the four curated experiments, and one experiment curated as anti-bait co-immunoprecipitation with no tag annotation 2. So at least one pull appears to have used an antibody against the protein itself, but the bulk of the evidence is tagged and therefore over-expressed.
Menin 214-390, and what can and cannot be read off it
IntAct's curation of PMID 12874027 annotates, on the menin side of one co-IP, sufficient binding region: 214-390 2. Menin is 610 aa, so this is a 177-residue central fragment — 29% of the protein.
What that supports: the contact does not require menin's N-terminal 213 residues or its C-terminal 220, in an over-expression co-IP.
What it does not support, and where a careless reader would go wrong: it is not an interface, not a set of contact residues, and not a claim about the MBM pocket. Menin's peptide-binding cavity — the site that accepts KMT2A MBM1, JunD, and every clinical menin inhibitor (interface-menin-kmt2a-mbm1, interface-menin-jund-pocket) — is formed by residues contributed from across the folded protein, and a large central fragment being sufficient tells you almost nothing about whether the pocket is involved.
The reason to record the number anyway is that it is testable and cheap: the region 214-390 overlaps the region 300-376 that the same era's work mapped as necessary for menin's binding to RelA (mechanism-menin-represses-nfkb-transactivation), which raises a concrete question — do FANCD2 and NF-kB compete for the same surface of menin, and does an MBM-pocket inhibitor touch either? None of that has been asked.
The five routes, and an honest unreadable
PMID 12874027 could not be read. Recorded here in full so nobody repeats the attempt:
| route | result |
|---|---|
1. Europe PMC core EXT_ID:12874027 | pmcid: null |
2. {PMCID}/fullTextXML | unavailable — no PMCID |
3. efetch db=pmc | unavailable — efetch db=pubmed returns no <ArticleId IdType="pmc"> either |
| 4. PMC HTML page | unavailable — no PMCID |
| 5. DOI resolution | no DOI exists: absent from PubMed's ArticleIdList, absent from Crossref for the exact title, and OpenAlex returns doi: null, oa_status: "closed", any_repository_has_fulltext: false |
| publisher site directly | aacrjournals.org HTTP 403 via both curl and the fetch tool |
This is a genuinely closed 2003 paper with no digital object identifier. Per concept-known-but-unreadable-literature, nothing in this entry states what the paper "probably" shows beyond its abstract and the IntAct curation of it.
| partner a | gene-men1 |
|---|---|
| partner b | gene-fancd2 |
| motif a | sufficient binding region on menin, residues 214-390 (IntAct curation of PMID 12874027); 177 of menin's 610 residues |
| motif b | none mapped; IntAct records only an MS-identified FANCD2 peptide at residues 10-22, which is an identification artefact of the experiment and not an interface |
| system | mouse embryonic fibroblasts; no human and no haematopoietic system reported |
| n source publications in intact | 1 |
| druggability | not assessable. No structure, no affinity, no mapped interface, and no evidence that the contact uses the MBM pocket that every clinical menin inhibitor occupies. Whether revumenib-class compounds touch this interaction is unknown and untested. |
interface residues a
[]
interface residues b
[]
structures
[]
structures absent
[ "no PDB deposit contains menin with FANCD2", "no PDB deposit contains FANCD2 with any non-FA protein" ]
methods
[ "anti-tag co-immunoprecipitation, FLAG tag on bait, western blot readout", "anti-tag co-immunoprecipitation, FLAG tag, FANCD2 identified by mass spectrometry (peptide 10-22)", "anti-bait co-immunoprecipitation, western blot readout" ]
ligands
[]
readability
{
"pmid": "12874027",
"routes_tried": 5,
"outcome": "unreadable",
"detail": "EPMC core pmcid null; efetch db=pubmed has no pmc ArticleId; no DOI in PubMed, Crossref or OpenAlex; OpenAlex oa_status closed; aacrjournals.org HTTP 403",
"title": "Menin associates with FANCD2, a protein involved in repair of DNA damage.",
"authors": "Jin S, Mao H, Schnepp RW, et al.",
"venue": "Cancer Res",
"year": 2003,
"title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
}- If a menin inhibitor does displace FANCD2 from menin, the class acquires a DNA-repair liability that no trial has looked for and no label mentions.
- If it does not displace FANCD2, then this interaction is not a route to menin-independent effects of the drug, and the fifth explanation for the Fanconi signal loses its only pharmacological handle.
- Any interpretation resting on this interface inherits a single unreplicated 2003 experiment in mouse fibroblasts. Building a combination rationale on it would be building on one co-IP.
- Does menin co-immunoprecipitate with FANCD2 in a human KMT2A-rearranged cell line at endogenous levels? Nobody has tried.
- Does the menin 214-390 region overlap the MBM/JunD pocket in three dimensions, and would revumenib compete with FANCD2? Both are answerable from existing structures plus one competition experiment.
- Do FANCD2 and RelA compete for menin? The curated regions, 214-390 and 300-376, overlap.
- Is the gamma-irradiation enhancement of the association reproducible outside mouse fibroblasts, and does it require ATM-dependent FANCD2 S222 phosphorylation?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Menin specifically interacts with FANCD2; the interaction is enhanced by gamma-irradiation; loss of menin in mouse embryonic fibroblasts leads to increased sensitivity to DNA damage; menin localises to chromatin and nuclear matrix and the nuclear-matrix association is enhanced by gamma-irradiation.
Menin associates with FANCD2, a protein involved in repair of DNA damage. · Jin S, Mao H, Schnepp RW, Sykes SM, Silva AC, D'Andrea AD, Hua X · Cancer Res 2003What this evidence cannot show
`established` for the detection of an association in the reported system only. Mouse embryonic fibroblasts; largely tagged constructs; 2003, before the ID2 complex, the K561 ubiquitin switch and the structure of menin were understood. Full text unreadable through all five routes plus the publisher site. The paper's own conclusion is hedged ('these results suggest'), and nothing in it is a leukaemia experiment.
Evidence that argues against this (1)
- The designed follow-up concluded the opposite about function: MEN1 and FANCD2 mutants have strikingly different mutation spectra and there is no genetic interaction between them for crosslink sensitivity, so MEN1's role in ICL repair is distinct from the Fanconi genes'. Marek LR et al. 2008
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
IntAct curates four experiments for the MEN1-FANCD2 pair, all from PMID 12874027: two anti-tag co-immunoprecipitations with a FLAG tag on the bait (one read by western blot, one by mass spectrometry identifying FANCD2 peptide 10-22), a third anti-tag experiment, and one anti-bait co-immunoprecipitation. The curation records a sufficient binding region on MEN1 of residues 214-390. No other publication supports this pair in IntAct.
IntAct molecular interaction database, PSICQUIC MITAB 2.7 query on interactor O00255 (MEN1), retrieved 2026-09-20 · IntAct (EMBL-EBI) 2026What this evidence cannot show
A curation of one paper is not independent evidence about that paper; it is a structured transcription of it, and the curator's 'sufficient binding region' reflects whatever deletion constructs the authors happened to test. Because the source paper is unreadable here, the deletion series behind 214-390 could not be inspected.
Evidence that argues against this (1)
- That IntAct holds only one source publication for this pair is itself the counter-point: two decades of menin affinity-purification mass spectrometry, including the BioPlex dual-proteome network and the Polycomb complexome map, have never recovered FANCD2 as a menin interactor. Huttlin EL et al. 2021
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Peripheral mononuclear cells from patients with familial MEN1 show an increased frequency of chromosome aberrations and reduced survival after exposure to the crosslinking agent diepoxybutane, relative to controls; hypersensitivity to alkylating agents exists in MEN1 patients and does not correlate with any particular MEN1 mutation.
Enhanced sensitivity to alkylating agent in lymphocytes from patients with multiple endocrine neoplasia type 1. · Itakura Y, Sakurai A, Katai M, Ikeo Y, Hashizume K · Biomed Pharmacother 2000What this evidence cannot show
Small clinical cytogenetics study from 2000; n is not stated in the abstract and the full text was not consulted. Diepoxybutane hypersensitivity is the diagnostic test for Fanconi anaemia, which is exactly why this observation motivated the search for a menin-FA link - and also why it is easy to over-read: a shared cellular phenotype is compatible with two independent pathways converging on the same lesion.
Evidence that argues against this (1)
- Sharing crosslinker hypersensitivity with FA cells is precisely the premise the 2008 Drosophila study set out to test, and it found the underlying mutation spectra to be different: single-base deletions in homopolymeric tracts for MEN1, large deletions for FANCD2. Marek LR et al. 2008
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The 2008 Drosophila study describes the 2003 experiments as immunoprecipitation assays performed in MEFs, and offers two readings of why a physical association could coexist with no genetic interaction: a mammalian-versus-Drosophila difference, or the two proteins acting together in the S-phase checkpoint but not in crosslink repair.
MEN1 and FANCD2 mediate distinct mechanisms of DNA crosslink repair. · Marek LR, Kottemann MC, Glazer PM, Bale AE · DNA Repair (Amst) 2008What this evidence cannot show
This is a second-hand description of the 2003 methods, taken from the discussion of a later paper by a different group. It is used here only because the primary text is unreadable, and it should be replaced with the primary description if anyone gains access.
Evidence that argues against this (1)
- Two readings offered by authors are speculation, not data, and the entry should not inherit them as though they were findings. A physical association with no genetic interaction is also explained by the association being real but non-functional, which the authors' two readings do not cover. Nakanishi K et al. 2005
menin side of the interface
- part of →FANCD2 — the FA effector menin binds, and the weakest FA gene in this graph's own selectivity scan
FANCD2 side of the interface
the functional inference from the 2003 co-IP failed a designed genetic test
a second non-KMT2A menin partner interface; unlike JunD, no evidence it uses the drug pocket
the curated 214-390 region is not a surface: it spans three spatial lobes and 29.9% of menin's total solvent-accessible area
overlaps the curated FANCD2 region by 15 of 228 residues
menin chromatin and nuclear-matrix localisation comes from the same 2003 paper as the FANCD2 interaction
PMID 12874027 is cited in seven entries for the FANCD2 axis and in none for its one-sentence chromatin/nuclear-matrix result, which this entry recovers
arm 4 tests whether Val371-Gln400 overlaps the curated FANCD2 region 214-390