Does the sarcoma fusion bind menin? One co-immunoprecipitation with the right control settles six years of the question
also called menin co-IP on YAP1::KMT2A::YAP1 · menin pulldown YKY versus KMT2A::AFF1
Express the sarcoma fusion protein and the leukaemia fusion protein in the same cells, then pull on menin and see which one comes with it. The prediction is that the leukaemia one does and the sarcoma one does not, and nobody has ever checked.
The same thing, for a scientist
A side-by-side co-immunoprecipitation in one cell background, designed to test the single load-bearing prediction of the exon map. Arms: (1) YAP1::KMT2A::YAP1, (2) VIM::KMT2A, (3) KMT2A::AFF1 as the positive control, (4) a KMT2A N-terminal fragment (aa 1-1400) as a second positive control for the interface alone, (5) a YKY construct with the CXXC zinc finger deleted, (6) empty vector. Readouts: reciprocal co-IP of endogenous menin and endogenous LEDGF/PSIP1 with each tagged construct, plus a revumenib competition arm to confirm that any recovered menin signal is through the drug-addressable pocket. Secondary readout: RT-qPCR of HOXA9, HOXA10 and MEIS1 in each arm, which tests the transcriptional half of the argument in the same experiment. The system is immortalised human fibroblasts - not because it is ideal but because it is the only system this entity has, and the constructs already exist in one laboratory.
Why it is worth doing and why it is cheap
The whole comparative argument of this beat rests on a coordinate derivation: the menin-binding motif is at aa 6-25, the sarcoma fusions start at KMT2A exon 2 or later, therefore no menin. That is a prediction, not a measurement, and this project has been repeatedly bitten by treating a well-reasoned inference as a finding.
The constructs needed already exist: YKY, YAP1::KMT2A and KMT2A::YAP1 were built and profiled in immortalised fibroblasts 1. The additional reagents are standard and the menin co-IP from a KMT2A-fusion-expressing cell is an established assay in this field.
The negative result is the informative one, which is unusual and worth saying: if YKY does not bind menin, the coordinate argument is confirmed and the drug class is formally excluded from this entity. If it does bind menin, an uncharacterised second menin surface exists on KMT2A and a large amount of this project's interface reasoning needs revisiting.
Controls this experiment must not skip
- A KMT2A::AFF1 arm in the same cells on the same day. A menin co-IP that fails for technical reasons looks exactly like the predicted negative. This project's standing rule applies: never trust a zero from an assay you have not shown can return a one.
- Revumenib competition on the positive control. Confirms the recovered menin is bound through the drug-addressable pocket rather than through a bridging complex.
- LEDGF/PSIP1 in parallel. IBM1 is in exon 1 and lost in all arms; IBM2 is in exon 2 and retained in the VIM arm by three residues. So the VIM arm is predicted menin-negative but possibly LEDGF-positive, which would be a genuinely novel intermediate state and is invisible unless LEDGF is blotted.
- A fibroblast is not a sarcoma cell. A positive result in fibroblasts would need repeating in a mesenchymal stem cell or in the tumour itself; a negative result is more transferable because menin expression is ubiquitous.
| design | Side-by-side co-immunoprecipitation of endogenous menin and endogenous LEDGF/PSIP1 with epitope-tagged fusion constructs expressed in one immortalised human fibroblast background, with reciprocal IP in both directions and a revumenib competition arm on the positive control. |
|---|---|
| readout | menin and LEDGF/PSIP1 recovery per arm by western blot, quantified against input; secondary RT-qPCR of HOXA9, HOXA10 and MEIS1 |
| controls | KMT2A::AFF1 arm run in the same cells on the same day (the assay must be shown able to return a positive); revumenib competition on that arm; IgG and untagged controls; LEDGF blotted in parallel because IBM2 status differs between the two sarcoma arms |
| cost tier | low |
| timeline | 6-10 weeks if the published constructs are obtainable; 4-5 months if they must be rebuilt |
| prediction | menin recovered from the KMT2A::AFF1 and KMT2A aa 1-1400 arms and not from YKY or VIM::KMT2A; LEDGF possibly recovered from VIM::KMT2A alone; no HOXA9/MEIS1 induction in either sarcoma arm |
| what a surprise would mean | menin recovery from YKY would imply a second, uncharacterised menin-binding surface on KMT2A outside aa 6-25, which would require re-examination of every coordinate-based interface claim in this graph |
arms
[ "YAP1::KMT2A::YAP1 (YKY)", "VIM::KMT2A", "KMT2A::AFF1 (positive control)", "KMT2A aa 1-1400 N-terminal fragment (interface-only positive control)", "YKY with the CXXC zinc finger deleted", "empty vector" ]
tests
[ "hyp-the-menin-interface-requirement-is-polarity-dependent-not-kmt2a-intrinsic" ]
- A negative co-IP in fibroblasts would be used to exclude menin inhibitors from this entity, and that inference is only as good as the positive control run beside it. Without the KMT2A::AFF1 arm the experiment is worthless in the direction it is most likely to point.
- Overexpressed tagged constructs can force interactions that do not occur at endogenous stoichiometry, so a POSITIVE result here would be the weaker one and would need endogenous confirmation in tumour material that barely exists.
- If this experiment is run and reported as 'menin-independent sarcoma fusion', a clinician could read that as licence to try something else entirely; the experiment tests a binding event, not a therapeutic hypothesis.
- Are the published fibroblast lines available from the group that made them? That single email determines whether this is a two-month or a five-month experiment.
- Should the panel include a KMT2A::MAML2 construct? If that thymoma fusion has 5' polarity it would be a second positive control from a solid tumour, which would separate lineage from architecture in one experiment.
- Is a revumenib-resistant MEN1 mutant worth including as a specificity control, given this project's resistance-allele work?
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Immortalised human fibroblasts expressing YAP1::KMT2A::YAP1, YAP1::KMT2A and KMT2A::YAP1 constructs have been made and profiled by expression analysis, with YKY-driven upregulated genes overlapping those of YKY-expressing tumours and of other YAP1 fusions - so the reagents this experiment needs already exist.
Insertion of the CXXC domain of KMT2A into YAP1: An unusual mechanism behind the formation of a chimeric oncogenic protein. · Teku G, Nilsson J, Magnusson L, et al. · Genes Chromosomes Cancer 2023What this evidence cannot show
Read from the abstract only; the full text is unreachable through all five of this project's routes (no PMCID; Wiley DOI returns HTTP 403). Whether the lines were banked, are available on request, or were assessed for transformation is therefore unknown, and the experiment may require rebuilding the constructs from the published exon structure.
Evidence that argues against this (1)
- If the lines are unavailable and the constructs must be rebuilt, the exon structure must be read from a full text nobody in this project can open. The breakpoints are however independently published as exon indices in two other series, so the constructs are reconstructible from those. Warmke LM et al. 2025
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
The prediction under test is derived here (2026-09-21) from coordinates: the KMT2A menin-binding motif at aa 6-25 lies in exon 1, and the YAP1 arm contributes only KMT2A exons 4/5-6 while the VIM arm contributes KMT2A exon 2 onward, so neither can present the motif.
Ensembl REST (ENST00000534358) and UniProtKB (Q03164) records, KMT2A exon boundaries intersected with the feature table · Ensembl (EMBL-EBI); UniProt Consortium · Ensembl REST / UniProtKB 2026What this evidence cannot show
A coordinate prediction, which is the reason the experiment is needed. It assumes the reported transcript is the expressed species and that no cryptic upstream initiation restores KMT2A N-terminal sequence - neither of which has been checked in any tumour.
Evidence that argues against this (1)
- Menin also contacts KMT2A through a second motif region (MBM2) and the full interface characterised structurally spans more than aa 6-10, so an experiment that finds weak residual binding would be hard to interpret as either a clean positive or a clean negative. The revumenib competition arm exists to disambiguate that case. Kao YC et al. 2020
the low-cost experiment that would close this absence
co-IP of endogenous menin and LEDGF against YKY, VIM::KMT2A and a KMT2A::AFF1 positive control
does the YKY protein bind menin despite lacking aa 6-25
LEDGF blotted in parallel because IBM2 is retained in the VIM arm by a three-residue margin
the measurement that would replace this coordinate argument
RECOMMENDS ADDING AN ARM: YKY inserts only KMT2A exons 4-5/6 and carries no SET, so it supplies essentially no MLL1 moiety and the MBM hypothesis is INDETERMINATE for it -- the one arm in an 11-lesion table where the framework makes no prediction. A VIM::KMT2A construct is the determinate arm. The counter-argument is that a positive pulldown on YKY would be the more surprising and more damaging result, so the recommendation is to add the VIM arm rather than to redirect the experiment.
the mirror of the YKY problem: there KMT2A supplies almost no MLL1 moiety, here the partner supplies almost no partner