ENL's nuclear annotation has no citation at all, and every picture of ENL in a nucleus is of a mutant or a fusion, never the wild-type protein
also called MLLT1 localization · ENL condensates · ENL nuclear puncta · YEATS hotspot condensate · SEC phase separation
ENL is one of the commonest KMT2A fusion partners, and the standard protein database says it lives in the nucleus without citing a single experiment for it. Every published image of ENL forming structures inside a nucleus is of a mutated or fused version of the protein, deliberately over-produced - so what normal ENL looks like inside a normal nucleus is, as far as this search could establish, unrecorded.
The same thing, for a scientist
UniProt Q03111 (MLLT1/ENL, 559 aa) reads `SUBCELLULAR LOCATION: Nucleus.` with **no ECO evidence code and no PubMed reference of any kind** [E1] - the weakest annotation state encountered anywhere in this beat, on a gene with degree 38 in this graph. The imaging literature is entirely about **gain-of-function states**: (i) ENL, or its intrinsically disordered region alone, is sufficient to initiate liquid-liquid phase separation of the super elongation complex, and **fusion of ENL to MLL boosts SEC phase separation** [E2]; (ii) Wilms-tumour/AML **YEATS hotspot mutants** self-associate and form discrete dynamic nuclear puncta that wild-type ENL does not, when ectopically expressed [E3]; (iii) those mutant condensates form at leukaemogenic genes in a conditional knock-in mouse and are displaced by an acetyl-binding inhibitor [E4]; and (iv) mutant condensates are driven by multivalent interactions requiring both disordered regions and extrinsic elongation factors, with the explicit warning that **expression beyond the endogenous level produces non-functional condensates** [E5]. Nothing images endogenous wild-type ENL, and nothing images KMT2A::MLLT1 itself.
An annotation with no source
Fetched from UniProt REST on 2026-09-20, the complete subcellular-location comment for Q03111 is:
{"commentType": "SUBCELLULAR LOCATION",
"subcellularLocations": [{"location": {"value": "Nucleus", "id": "SL-0191"}}]}
No evidences array. No ECO code. No PubMed id. For comparison, in the same beat DOT1L carries ECO:0000269 and ECO:0000250, PSIP1 carries two ECO:0000269 references, and ELL carries five. ENL carries nothing.
This is not an accusation against UniProt - an uncited annotation is often an old, obviously-true curator statement. It is a warning about the import: a script that reads cc_subcellular_location and writes 'nucleus' into a gene entry produces, for ENL, a graph fact whose entire provenance is the string 'Nucleus'. The evidence code is the only thing that distinguishes that case from a measured one, and it is exactly what a bulk copy discards.
Everything imaged is a gain-of-function state
| what was imaged | state | system | source |
|---|---|---|---|
| SEC droplets nucleated by ENL or its IDR | overexpressed, plus in-vitro droplets | human cells, serum induction | 2 |
| discrete dynamic nuclear puncta | YEATS hotspot mutant, ectopically expressed | human and mouse cells | 3 |
| condensates at leukaemogenic genes | mutant knock-in, endogenous locus | mouse HSPC, in vivo AML | 4 |
| condensate requirements, mutagenesis series | mutant, titrated expression | human cells | 5 |
Read down the 'state' column: wild-type endogenous ENL appears nowhere. The nearest thing is 4, where the mutation is knocked in at the endogenous locus - which fixes the expression-level artefact but still images a mutant.
And 2 contains the only fusion-relevant sentence in the set: "the fusion of ENL with MLL can boost SEC phase separation." That is one line in one figure of a paper about serum response, and it is the whole published basis for saying KMT2A::MLLT1 has condensate behaviour.
The condensate caveats apply here in full, and one of them is the authors' own
This graph already holds localization-condensate-evidence-standards, and every item in it bites here: 1,6-hexanediol and classical FRAP cannot distinguish phase separation from binding to spatially clustered sites; hexanediol independently condenses chromatin in living cells; and low-complexity-domain interactions form dynamic hubs without detectable phase separation at physiological expression.
The strongest version of that last caveat is stated by the condensate paper itself: "expression of ENL mutants beyond the endogenous level leads to non-functional condensates" 5. So the field's own best work says the phenomenon is expression-level-dependent, and three of the four imaging datasets here are above endogenous level.
What survives all of it: 4's conditional knock-in forms condensates at endogenous expression, and an acetyl-binding inhibitor displaces them and delays AML in vivo. A drug that moves the structure and changes the outcome is the kind of evidence that a hexanediol caveat cannot dissolve.
What is missing, and why it is the cheapest experiment in this beat
- Endogenous wild-type ENL, imaged. Diffuse? Punctate? Chromatin-associated in mitosis? Unknown from this search.
- KMT2A::MLLT1, imaged. The fusion that this graph is partly about has never been photographed.
localization-endogenous-mll-af4-punctate-negativealready records the analogous negative for KMT2A::AFF1, so the field's one careful attempt at an endogenous fusion image gave a negative result - which makes the ENL version a real question rather than a formality. - Does the fusion make puncta at endogenous level? 2's boost is an overexpression result. A knock-in or a degron-tagged endogenous fusion would answer it.
All three are immunofluorescence experiments in cells this field already has.
| gene | gene-mllt1 |
|---|---|
| uniprot | Q03111 |
| length AA | 559 |
| compartment subcellular | nucleus - annotated in UniProt with NO evidence code and NO reference |
| compartment subnuclear | discrete dynamic nuclear puncta/condensates, but ONLY for YEATS hotspot mutants and for overexpressed SEC/ENL-IDR constructs; nothing published for endogenous wild-type ENL or for KMT2A::MLLT1 |
| expression level caveat | PMID 36272410 states that expression of ENL mutants beyond the endogenous level produces NON-FUNCTIONAL condensates; three of the four imaging datasets here are above endogenous level |
| shown for fusion | NO for imaging. The only fusion-relevant observation is that MLL-ENL boosts SEC phase separation in an overexpression droplet assay (PMID 32270036); KMT2A::MLLT1 has never been imaged in a cell. |
| patient material data | false |
| endogenous protein imaged | false |
uniprot evidence codes
{
"Nucleus": "NONE - the comment carries no evidences array at all"
}state imaged
{
"wild_type_endogenous": "not found in this search",
"wild_type_overexpressed": "SEC droplets (PMID 32270036)",
"YEATS_hotspot_mutant_ectopic": "discrete dynamic nuclear puncta (PMID 31853060, 36272410)",
"YEATS_hotspot_mutant_knock_in": "condensates at leukaemogenic genes in murine HSPC (PMID 38655899)",
"KMT2A_MLLT1_fusion": "never imaged; the only datum is that MLL-ENL boosts SEC phase separation in an overexpression assay (PMID 32270036)"
}methods
[ "live and fixed fluorescence imaging of tagged constructs", "in-vitro droplet reconstitution", "conditional knock-in mouse", "condensate-disrupting mutagenesis series", "pharmacological displacement with an acetyl-binding inhibitor" ]
species and systems
[ "human cell lines (non-leukaemic)", "mouse nephrogenesis assay", "mouse haematopoietic stem/progenitor cells", "conditional knock-in mouse AML" ]
- An acetyl-binding ENL inhibitor displaces mutant condensates, but the same YEATS pocket reads acetylation on normal target genes in normal haematopoiesis, so the therapeutic window is a dose window rather than a target window.
- If the condensate is expression-level-dependent, then any resistance mechanism that lowers fusion expression could paradoxically improve condensate function rather than abolish it, per the authors' own supra-endogenous observation.
- ENL and AF9/MLLT3 are paralogues sharing YEATS and AHD; a compound targeting the ENL YEATS acetyl pocket is a plausible AF9 binder too, and this graph should not assume selectivity between them without a measured pair.
- Because wild-type endogenous ENL has never been imaged, there is no baseline for a pharmacodynamic imaging assay - a trial could not currently score 'ENL relocalised' in a patient sample.
- What does endogenous wild-type ENL look like in a nucleus? No image was found in this search, and it is the baseline against which every mutant and fusion result is being read.
- Does KMT2A::MLLT1 form puncta at endogenous expression? The one fusion observation is an overexpression droplet assay; a degron- or knock-in-tagged endogenous fusion would settle it, and the analogous careful experiment on KMT2A::AFF1 gave a NEGATIVE result.
- Is the acetyl-binding inhibitor that displaces mutant-ENL condensates active against KMT2A::MLLT1 leukaemia, where the YEATS domain is retained in the fusion? The in-vivo pharmacology exists; the indication has not been tested.
- Can condensation be separated from chromatin binding for ENL? Every current perturbation removes both, so 'the condensate is the oncogenic unit' remains under-determined.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
UniProt Q03111 (human MLLT1/ENL, 559 aa) annotates subcellular location as 'Nucleus' with no ECO evidence code and no literature reference attached to the annotation.
UniProtKB Q03111 (ENL_HUMAN) - subcellular location comment with no evidence array, fetched 2026-09-20 · UniProt Consortium · UniProtKB 2026What this evidence cannot show
A database record. Resolution: gene_exact:MLLT1 AND organism_id:9606 AND reviewed:true returned one reviewed record, gene_primary MLLT1, accession Q03111, length 559. An absent evidence array means the annotation is not attributed in the record; it does NOT mean no experiment exists, and it is common for old, uncontroversial curator statements. The finding reported here is about provenance, not about whether ENL is nuclear.
Evidence that argues against this (1)
- It would be wrong to treat this as evidence that ENL is not nuclear. ENL is a YEATS-domain histone-acetylation reader in the super elongation complex and is recovered in chromatin fractions in essentially every biochemical study of it, so the nuclear assignment is almost certainly correct - the point is that a graph asserting it should cite the biochemistry rather than the uncited database line.
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
SEC components compartmentalise and concentrate P-TEFb by liquid-liquid phase separation, with ENL or its intrinsically disordered region alone sufficient to initiate SEC droplet formation and AFF4 acting to fluidise the droplets; SEC droplets form rapidly on serum exposure and are required for rapid transcriptional induction, and fusion of ENL with MLL can boost SEC phase separation.
ENL initiates multivalent phase separation of the super elongation complex (SEC) in controlling rapid transcriptional activation. · Guo C, Che Z, Yue J, Xie P, Hao S, Xie W, Luo Z, Lin C · Sci Adv 2020What this evidence cannot show
Overexpression and in-vitro droplet work in non-leukaemic human cells with a serum-response readout; the MLL-ENL statement is a single result inside a paper about SEC dynamics, not a leukaemia experiment, and no KMT2A-rearranged cell line is involved. The standing condensate caveat applies: droplet formation by an overexpressed IDR does not establish phase separation of the endogenous complex, and the assays used cannot separate LLPS from binding to clustered sites. Read: abstract only (PubMed efetch, 2026-09-20).
Evidence that argues against this (1)
- At physiological expression, low-complexity-domain interactions that control transcription form dynamic hubs WITHOUT detectable phase separation (the standing finding recorded in localization-condensate-evidence-standards). An IDR that is 'sufficient to initiate droplet formation' when overexpressed is therefore weak evidence about what the endogenous SEC does, and the MLL-ENL 'boost' inherits that weakness.
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In human and mouse cells, recurrent hotspot mutations in the ENL YEATS domain confer gain-of-function in chromatin recruitment and transcriptional control; ectopically expressed ENL mutants exhibit greater self-association and form discrete and dynamic nuclear puncta characteristic of local high-concentration regulatory hubs, and this self-association is functionally linked to enhanced chromatin occupancy and gene activation.
Impaired cell fate through gain-of-function mutations in a chromatin reader. · Wan L, Chong S, Xuan F, Liang A, Cui X, Gates L, Carroll TS, Li Y, Feng L, Chen G, Wang SP, Ortiz MV · Nature 2020What this evidence cannot show
RULE 73: the disease context is Wilms tumour (paediatric kidney cancer) and the nephrogenesis assay is murine - not leukaemia, not KMT2A-rearranged, not B-lineage. The puncta are formed by ECTOPICALLY EXPRESSED mutants; the paper's own framing is that mutants do this and wild-type does not, so the entry cannot be read as describing normal ENL. Read: abstract only (PubMed efetch, 2026-09-20).
Evidence that argues against this (1)
- This is the cleanest demonstration in the set that PUNCTA ARE A MUTANT PROPERTY. It therefore argues against, not for, the common shorthand that 'ENL forms condensates'. Wild-type ENL in this work is the negative control, and any entry citing ENL condensate biology for the wild-type protein or for a KMT2A::MLLT1 fusion is citing a control as a result.
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In a conditional knock-in mouse, mutant ENL perturbs haematopoiesis and triggers aggressive AML; mutant ENL forms condensates in haematopoietic stem/progenitor cells at key leukaemogenic genes, mutagenesis that disrupts condensate formation impairs its chromatin and oncogenic function, and an acetyl-binding inhibitor of mutant ENL displaces the condensates from target loci and delays AML initiation and progression in vivo.
Condensate-Promoting ENL Mutation Drives Tumorigenesis In Vivo Through Dynamic Regulation of Histone Modifications and Gene Expression. · Liu Y, Li Q, Song L, Gong C, Tang S, Budinich KA, Vanderbeck A, Mathias KM, Wertheim GB, Nguyen SC, Outen R, Joyce EF · Cancer Discov 2024What this evidence cannot show
RULE 73: murine conditional knock-in of a POINT MUTANT, producing myeloid leukaemia. This is not a KMT2A fusion and not a B-lineage disease; ENL-mutant AML and KMT2A::MLLT1 leukaemia are different entities that happen to share a protein. The strength of the design - endogenous locus, genetic condensate-disrupting mutants, and pharmacological displacement with an outcome readout - is the reason this item is graded strong despite the system mismatch. Read: abstract only (PubMed efetch, 2026-09-20).
Evidence that argues against this (1)
- Displacement by an acetyl-binding inhibitor shows the condensates depend on YEATS-domain acetyl-lysine binding; it does not show that the condensate, rather than the chromatin binding it requires, is the oncogenic unit. The two are experimentally entangled here because the same interaction is needed for both, and a compound that blocked condensation without blocking binding would be required to separate them.
- E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
ENL YEATS hotspot mutations trigger aberrant condensates at native genomic targets through multivalent homotypic and heterotypic interactions requiring mutation-induced structural change in the YEATS domain, two disordered regions of opposing charge, and incorporation of extrinsic elongation factors; extensive mutagenesis establishes condensate formation as a driver of oncogenic gene activation, and expression of ENL mutants beyond the endogenous level leads to non-functional condensates.
Hotspot mutations in the structured ENL YEATS domain link aberrant transcriptional condensates and cancer. · Song L, Yao X, Li H, Peng B, Boka AP, Liu Y, Chen G, Liu Z, Mathias KM, Xia L, Li Q, Mir M · Mol Cell 2022What this evidence cannot show
Human cell systems with mutant constructs; again a mutant, not the wild-type protein and not a fusion. The final clause is the load-bearing caveat for this whole literature and comes from the authors: above-endogenous expression produces condensates that do not work, so condensate presence is not by itself informative unless expression level is controlled. Read: abstract only (PubMed efetch, 2026-09-20).
Evidence that argues against this (1)
- The authors' own expression-level result is the strongest internal counter-argument available: it means a positive condensate image at unknown expression level carries no information about function. Applied backwards to PMID 32270036's 'MLL-ENL boosts SEC phase separation', which is an overexpression experiment, it says that observation cannot currently be interpreted functionally.
localisation record for ENL: an uncited nuclear annotation and a condensate literature entirely about mutants
the ENL condensate papers themselves report that supra-endogenous expression yields non-functional condensates
the equivalent endogenous-fusion imaging experiment has been done for KMT2A::AFF1 and was negative; it has never been done for KMT2A::MLLT1
- related to →KMT2A::MLLT1 (MLL-ENL), t(11;19)(q23;p13.3)
KMT2A::MLLT1 has never been imaged; the sole fusion observation is an overexpression droplet assay
worked example of a location value with no evidence array at all