SET::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
also called SET-NUP214 MLL1 · nucleoporin fusion condenses MLL1 · the wild-type MLL1 route
SET::NUP214 contains no piece of KMT2A, which is why it looked like a clean test of whether leukaemia can drive HOXA without the menin-KMT2A machinery. But the fusion grabs the ordinary, unbroken KMT2A protein and pulls it onto the same genes - and ordinary KMT2A is held by menin.
The same thing, for a scientist
Two independent reports place **wild-type MLL1 in the SET::NUP214 mechanism**: 1. **SET-NUP214 binds MLL through the SET acidic region**, and the two **cooperatively enhance the HOXA10 promoter**, with neither the SET region nor the NUP214 region sufficient alone [E1]. 2. The **phase-separated nuclear bodies of NUP98::HOXA9 and SET::NUP214 promote condensation of MLL1 together with CRM1**, are robustly associated with MLL1/CRM1, and co-localise with them on chromatin [E2]. And a third, older result runs in the same direction from the wild-type side: **the amino-terminal region of MLL - exactly the portion every KMT2A fusion retains - binds SET/TAF-Ibeta through TAF-Ibeta's acidic region, and the two synergistically upregulate Hoxa9** [E3]. **The consequence for this project's central claim.** The menin-interface hypothesis was being tested with SET::NUP214 as a KMT2A-free instance of HOXA activation, on the reasoning that a fusion with no menin-binding motif cannot be menin-dependent. **That reasoning is unsound here.** The fusion has no MBM of its own, but wild-type MLL1 does - MBM1 and MBM2 are in the N-terminus of the unbroken protein, and menin binds them. If the HOXA output depends on recruited wild-type MLL1, then a menin inhibitor could in principle displace it and lower HOXA in a leukaemia containing no KMT2A rearrangement at all. **Nobody has tested it.** There is not one published menin, MEN1 or menin-inhibitor experiment on this fusion or on either of its cell lines. **All of this is human T-ALL and AML cells or heterologous expression systems, not infant B-lineage KMT2A::AFF1.**
The correction this beat owes its own brief
The dispatch for this beat, and the Q55 comparator note it builds on, both describe SET::NUP214 as the second KMT2A-free HOXA-driving fusion and treat that as the thing that makes it a good test of menin-independence. The first half is true and the second does not follow.
- True: the fusion protein contains no KMT2A sequence, no CXXC domain and no menin-binding motif. Nothing in this entry changes that.
- Does not follow: therefore it does not need menin. A fusion can be menin-dependent by recruiting the wild-type protein that carries the menin contact. SET::NUP214 demonstrably recruits wild-type MLL1, by two independent methods, in two papers, thirteen years apart.
CALM-AF10 does not have this problem on the evidence this graph holds - no published result puts wild-type MLL1 in the CALM-AF10 complex. So the two comparator fusions are not equivalent tests, and treating them as two independent instances of the same separation overstates the case. CALM-AF10 remains the cleaner instance; SET::NUP214 is now an open question rather than a second confirmation.
How much weight this can carry, stated plainly
Not much yet, and the entry should not be read as asserting menin dependence. Three specific weaknesses:
- E1 is a promoter-reporter assay read at abstract level, in an unstated cell system, with the MLL construct unresolved to an accession. Reporter cooperativity is not chromatin occupancy.
- E2 was read at abstract level only - the paper is not in PMC. The abstract's MLL1 claim is made for 'the nuclear bodies of NUP98::HOXA9 and SET::NUP214' jointly, and the 3D-genome half of the paper is attributed specifically to NUP98::HOXA9. Whether the MLL1 condensation was demonstrated for SET::NUP214 in its own right is not established from the abstract.
- Condensation is not binding. This graph's standing condensate caveat applies: 1,6-hexanediol and classical FRAP cannot distinguish phase separation from binding to spatially clustered sites, and 'robustly associated' in an abstract does not say which assay produced it.
What it IS enough for is to block the inference the beat was commissioned to make, and to justify one cheap experiment.
The experiment that settles it, and why it is one western blot
LOUCY is a distributed, standard cell line. The founding paper already ran anti-NUP214 and anti-DOT1L immunoprecipitates in it and blotted them for SET, CRM1 and AF10 4. Adding an anti-menin lane to that same immunoprecipitate answers the question: if menin travels with SET-NUP214, the fusion's HOXA programme has a drugged target on it; if it does not, SET::NUP214 becomes the clean second instance the coordinator wanted and the menin-interface hypothesis is correspondingly sharpened.
See exp-menin-in-the-set-nup214-complex-and-menin-inhibitor-in-loucy.
| mechanism class | co-factor recruitment |
|---|---|
| fusion | fusion-set-nup214 |
| recruited protein | wild-type MLL1 / KMT2A |
| contact region on fusion | the SET acidic region (SET aa 236-290) |
| menin binding motif on recruited protein | MBM1 and MBM2, in the wild-type KMT2A N-terminus |
| why this matters | The inference 'no menin-binding motif in the fusion, therefore menin-independent' fails whenever the fusion recruits the wild-type protein that carries the motif. This is the first documented instance of that failure mode in this graph's comparator set. |
| calm af10 comparison | No published result places wild-type MLL1 in the CALM-AF10 complex, so CALM-AF10 remains the cleaner test of menin-independence and SET::NUP214 does not currently provide a second independent instance. |
| menin tested in this system | false |
experiments
[ "exp-menin-in-the-set-nup214-complex-and-menin-inhibitor-in-loucy" ]
hypotheses
[ "hyp-set-nup214-hoxa-runs-through-recruited-wild-type-mll1" ]
- If this route is real, a menin inhibitor in SET::NUP214 disease inherits every resistance mechanism the graph already records for menin inhibitors - MEN1 pocket mutation, adaptive non-genetic states, RAS/FLT3 co-mutation - without any of the clinical evidence base, because no trial has enrolled this genotype.
- The CRM1 route and the MLL1 route are not mutually exclusive, so blocking menin may lower HOXA only partially, and a partial HOXA reduction is exactly what the founding knockdown data show is compatible with retained programme.
- Conversely, if menin is NOT in the complex, that is a real negative for menin-inhibitor repurposing and should stop anyone proposing it on the strength of HOXA-high alone - HOXA-high is not a menin-dependence biomarker.
- Is menin in the SET-NUP214 complex? One anti-menin lane on an immunoprecipitate that has already been run answers it.
- Does a menin inhibitor lower HOXA in LOUCY or MEGAL? If it does, menin-inhibitor eligibility is currently drawn too narrowly, because eligibility is written around KMT2A rearrangement and NPM1 mutation.
- Does NUP98::HOXA9, which this graph already records as menin/MLL1-dependent, share the mechanism? If the MLL1 recruitment is a general property of nucleoporin-fusion nuclear bodies, SET::NUP214's menin exposure should look like NUP98-fusion menin exposure, which is already documented.
- Is the recruited MLL1 catalytically required, or is it a scaffold? A SET domain-dead MLL1 would separate the two.
- E1moderate 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 SET-NUP214 and MLL 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 2021What this evidence cannot show
ABSTRACT-LEVEL ONLY: not in PMC, not read in full. A promoter-reporter assay plus deletion mapping; cell system unstated in the abstract. 'MLL' is not resolved to an accession and brief rule 66 makes the bare word ambiguous between KMT2B and KMT2D in principle; read here as MLL1/KMT2A on the HOXA10 context, which is an inference. Reporter cooperativity is not evidence of chromatin co-occupancy.
Evidence that argues against this (1)
- A BioID proximity-labelling study of SET-NUP214 did NOT recover MLL1 among its results - MLL1 and DOT1L appear in that paper only in the introduction, citing prior work, while the significantly overrepresented categories are mitochondrial proteins and metabolic regulators. That is a weak negative because the system is HCT-116 colorectal carcinoma cells transfected with an exogenous tagged fusion, but it is the only unbiased survey of this fusion's environment and it did not find MLL1. Mendes A et al. 2020
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The phase-separated nuclear bodies of NUP98::HOXA9 and SET::NUP214 promote the condensation of MLL1, are robustly associated with MLL1/CRM1, and co-localise with them on chromatin.
Phase-separated nuclear bodies of nucleoporin fusions promote condensation of MLL1/CRM1 and rearrangement of 3D genome structure. · Oka M, et al. · Cell Rep 2023What this evidence cannot show
ABSTRACT-LEVEL ONLY: Europe PMC returns pmcid null for this record, so PMC routes are unavailable, and the Cell Reports DOI was not pursued within this beat. Cell system, MLL1 antibody and the assay behind 'robustly associated' are all unknown here. The claim is made jointly for two fusions and the 3D-genome result is attributed in the abstract to NUP98::HOXA9 alone, so how much of the MLL1 result is SET::NUP214-specific is unresolved. This graph's standing condensate caveat applies: hexanediol sensitivity and classical FRAP cannot distinguish LLPS from binding to spatially clustered sites, and the abstract does not say which assay was used.
Evidence that argues against this (1)
- METHODOLOGICAL DEFEATER plus a published alternative explanation for the same observation. Co-condensation and co-localisation on chromatin are not binding. The same laboratory's earlier work shows CRM1 is ALREADY bound at the HOX clusters in LOUCY without the fusion, and that knocking the fusion down does not significantly reduce it - so a picture in which MLL1, CRM1 and SET::NUP214 all sit on the same chromatin is equally well explained by three proteins independently addressed to the same locus as by the fusion recruiting MLL1. The condensate assays that would be used to argue otherwise cannot distinguish phase separation from binding to spatially clustered sites, and the abstract does not say which was run. Oka M et al. 2019
- 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 2005What this evidence cannot show
A 2005 biochemical study on a wild-type MLL N-terminal FRAGMENT and wild-type TAF-Ibeta, not on the fusion, not in a leukaemia cell. It supports the contact and says nothing about whether it occurs in SET-NUP214-positive cells. Identifiers re-fetched from PubMed esummary on 2026-09-20 rather than copied from the sibling entry that also cites this paper.
Evidence that argues against this (1)
- PUBLISHED CONFLICTING RESULT that supplies a confound this item cannot control. The MLLN/TAF-Ibeta contact is a property of WILD-TYPE SET, and wild-type SET is not a bystander in this disease: SET is overexpressed in 60 of 214 acute myeloid leukaemia patients at diagnosis, a prevalence of 28%, associated with worse overall and event-free survival and with protein phosphatase 2A inhibition. So an MLL1-SET contact will occur in a large minority of leukaemias that carry no fusion at all. Demonstrating that SET binds MLL1 therefore does not demonstrate that SET::NUP214 recruits MLL1 to anywhere in particular - the fusion's contribution would be the ADDRESS, and this item measures the contact. Cristóbal I et al. 2012
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In LOUCY, the founding study ran anti-NUP214, anti-PP32, anti-SET, anti-CRM1 and anti-hDOT1L immunoprecipitates and blotted them for SET, CRM1, DOT1L and AF10; no menin or MEN1 blot appears anywhere in the paper.
The recurrent SET-NUP214 fusion as a new HOXA activation mechanism in pediatric T-cell acute lymphoblastic leukemia. · Van Vlierberghe P, et al. · Blood 2008What this evidence cannot show
This is a statement about what the paper contains, made from a full-text read via the PMC HTML route (PMC2343598) and a string search of the flattened text for 'menin' and 'MEN1'. It bounds the paper, not the literature; the literature-wide bound is in `ana-set-nup214-menin-bounded-absence`. An absence of a blot is not evidence of an absence of the protein.
Evidence that argues against this (1)
- AN ARGUMENT FROM A 2008 FIGURE LIST, against which a directly relevant positive experiment now exists. The founding authors worked under a CRM1-and-DOT1L model in which menin had no place, so the missing blot reflects their model rather than a considered negative. The informative test was run in the nearest lesion class: menin-MLL1 inhibition is active in NUP98-fusion leukaemia at an IC50 similar to KMT2A-rearranged and NPM1c cells, and evicts MLL1 and the fusion from chromatin together in a nucleoporin fusion carrying no menin motif of its own. That is the experiment this item is standing in for, and it has been done next door with a positive result. Heikamp EB et al. 2022
the condensate reading of the same assembly
wild-type SET is overexpressed in 28% of 214 AML patients, so an MLL1-SET contact occurs in many fusion-negative leukaemias and does not establish that the fusion supplies the address
wild-type MLL1, via the SET acidic region and the fusion nuclear body
the recruited protein is wild-type MLL1/KMT2A, which carries MBM1 and MBM2
if the recruited MLL1 is load-bearing, menin is on the SET::NUP214 HOXA programme - untested
- analogous to →Why a nucleoporin fusion needs menin-MLL1: the second menin-dependent fusion family, and what it borrows rather than carries
the same nucleoporin-fusion-recruits-MLL1 shape, already documented for NUP98 fusions and already menin-dependent there
- depends on →SET / TAF-Ibeta domain map (Q01105, 290 aa) - dimerisation, earmuff, and the acidic tail that binds MLL
the contact is the SET acidic region (aa 236-290)
the hypothesis generalises this mechanism to a menin dependency
the recruited-MLL1 finding is what demotes SET::NUP214 from a confirmation to an open question