Oncofusion

start / graph / localization-partner-baseline-compartment-is-often-miscalled

Explainermoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q28-subnuclear

A partner's 'normal' compartment is usually read off a different tissue, and for at least two KMT2A partners that reading is wrong in blood cells

also called partner baseline localisation · EEN endophilin A2 is nuclear · AF3p21 bipartite NLS

When people say a fusion partner is a cytoplasmic protein, they usually mean somebody saw it in the cytoplasm of a nerve cell or a bone cell. Two MLL partners turn out to sit in the nucleus of blood cells instead, which means the story that MLL drags cytoplasmic proteins inward is partly an artefact of looking in the wrong tissue.

The same thing, for a scientist

The standard narrative for the non-elongation KMT2A partners is that MLL drags a cytoplasmic protein into the nucleus, where a retained oligomerisation or scaffolding module then does the damage. Two documented cases break the premise rather than the conclusion. **EEN/endophilin A2**, an MLL partner and a member of an endocytic-vesicle protein family, is localised **predominantly in the nuclei** of haematopoietic, fibroblast and epithelial lines - 'in contrast with its reported cytoplasmic localization in neurons and osteoclasts' - and shuttles between nucleus and cytoplasm, with a cell-cycle-dependent distribution that is perichromosomal in prometaphase, on the bipolar spindle in metaphase and anaphase, and at the midzone and midbody in telophase, tracking a protein level that peaks in G2/M [E1]. **AF3p21**, an MLL partner in therapy-related leukaemia, carries a bipartite nuclear localisation signal and is 'localized exclusively in the cell nucleus' by Western blot and immunocytochemistry [E2]. Both facts were published as partner-biology papers and neither appears to have been carried back into the fusion literature. The practical rule for this beat: **before recording that a fusion relocalises its partner, check where the partner is in a blood cell**, because the compartment label usually comes from whichever tissue first characterised the protein.

Go deeper
Why this matters more than a bookkeeping correction

Three downstream claims in this graph depend on the partner's baseline compartment.

  1. 'The fusion relocalises the partner.' If the partner is already nuclear in the relevant cell type, there is nothing to relocalise, and the observation that the chimera is nuclear stops being evidence for anything.
  2. 'A cytoplasmic partner contributes twice - as a module and by depleting the cytoplasm of its wild-type self.' This is documented for AFDN and is a real mechanism, but it only exists where a cytoplasmic pool exists.
  3. 'Forced oligomerisation is the conserved core.' This model predicts the partner's compartment is irrelevant, and partner-is-already-nuclear cases are mildly confirmatory for it, since the transforming module works whichever side of the envelope its wild-type counterpart sits on.

The EEN case also carries an independent warning about method: EEN's distribution changes through the cell cycle, so a single fixed-cell image of an asynchronous population averages four different localisations 1. That is a general hazard for every punctate-nuclear claim in this beat and it is rarely controlled.

What would have to be checked for each partner

A defensible partner-baseline statement needs three things, and most of this graph's partner entries have one:

  • the cell type in which the localisation was determined, and whether it is haematopoietic;
  • whether the reagent could see endogenous protein or only a tag;
  • whether the protein shuttles, in which case the steady-state picture is a ratio rather than a compartment.

The septin case is the one that passes: SEPT6 alone, overexpressed in the same cells as the fusion, stays cytoplasmic in the same figure 3. That is a proper baseline. The AFDN case passes too, because the endogenous partner's move is the measurement 4.

Facts and figures
rulecheck the partner's compartment in a haematopoietic cell before asserting relocalisation
partners checked
AF3p212,001AF3p21: 2001EEN/endophilin A2 (SH3GL1)2,004EEN/endophilin A2 (SH3GL1): 2004SEPT62,005SEPT6: 2005AFDN/AF62,014AFDN/AF6: 2014
partners checked — the numbers behind the chart
[
  {
    "partner": "EEN/endophilin A2 (SH3GL1)",
    "reported_compartment": "cytoplasmic (neurons, osteoclasts)",
    "measured_in_blood_lineage": "predominantly nuclear, shuttling",
    "pmid": "15214844",
    "title": "Subcellular localization of EEN/endophilin A2, a fusion partner gene in leukaemia.",
    "authors": "Cheung N, So CW, Yam JW, et al.",
    "venue": "Biochem J",
    "year": 2004,
    "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."
  },
  {
    "partner": "AF3p21",
    "reported_compartment": "-",
    "measured_in_blood_lineage": "exclusively nuclear, bipartite NLS, 80 kDa",
    "pmid": "11241789",
    "title": "Genomic organization, tissue expression, and cellular localization of AF3p21, a fusion partner of MLL in therapy-related leukemia.",
    "authors": "Hayakawa A, Matsuda Y, Daibata M, et al.",
    "venue": "Genes Chromosomes Cancer",
    "year": 2001,
    "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."
  },
  {
    "partner": "SEPT6",
    "reported_compartment": "cytoplasmic",
    "measured_in_blood_lineage": "cytoplasmic (293T control, overexpressed)",
    "pmid": "15761502",
    "title": "Dimerization of MLL fusion proteins and FLT3 activation synergize to induce multiple-lineage leukemogenesis.",
    "authors": "Ono R, Nakajima H, Ozaki K, et al.",
    "venue": "J Clin Invest",
    "year": 2005,
    "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."
  },
  {
    "partner": "AFDN/AF6",
    "reported_compartment": "cytoplasmic",
    "measured_in_blood_lineage": "cytoplasmic, moved to nucleus by the fusion",
    "pmid": "24695851",
    "title": "MLL-AF6 fusion oncogene sequesters AF6 into the nucleus to trigger RAS activation in myeloid leukemia.",
    "authors": "Manara E, Baron E, Tregnago C, et al.",
    "venue": "Blood",
    "year": 2014,
    "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."
  }
]
How this could bite us back
  • A therapeutic rationale built on 'restoring the partner to its correct compartment' would be aimed at a compartment assignment that, for at least two partners, came from the wrong tissue.
  • Cell-cycle-dependent partner localisation means any imaging-based pharmacodynamic readout would move with a drug's effect on cell-cycle distribution, independent of any effect on the fusion.
Still open
  1. Is endogenous EEN/endophilin A2 nuclear in primary human haematopoietic progenitors, or only in established lines?
  2. How many of this graph's partner entries record a compartment taken from a non-haematopoietic tissue without saying so?
  3. Does any partner's cell-cycle-dependent redistribution confound a published fixed-cell localisation claim about its fusion?
Every source behind this entry (4)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    EEN/endophilin A2, an MLL fusion partner, is localised predominantly in the nuclei of haemopoietic, fibroblast and epithelial cell lines, in contrast with its reported cytoplasmic localisation in neurons and osteoclasts; it shuttles between nucleus and cytoplasm and shows cell-cycle-dependent distribution - perichromosomal in prometaphase, on the bipolar spindle in metaphase and anaphase, and at the midzone and midbody in telophase - with protein levels highest in G2/M.

    PMID 15214844DOI
    What this evidence cannot show

    Read from the PMC HTML page (PMC1134040) after fullTextXML failed and efetch db=pmc returned no <body>; the stripped text measured 49,296 characters, so route 4 of the unreadable test succeeded. The work is on wild-type EEN, not on an MLL-EEN fusion, and the cell lines are established lines rather than primary haematopoietic cells.

    Evidence that argues against this (1)
    • Nucleocytoplasmic shuttling means the steady-state picture depends on the export inhibitor, fixation and cell-cycle phase, so 'predominantly nuclear' is a condition-dependent statement rather than a fixed property. Hayakawa A et al. 2001
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    AF3p21, an MLL fusion partner in therapy-related leukaemia with t(3;11)(p21;q23), encodes a 722-residue protein with an SH3 domain, a proline-rich domain and a bipartite nuclear localisation signal, and Western blot and immunocytochemical analyses showed the 80 kDa protein is localised exclusively in the cell nucleus.

    PMID 11241789DOI
    What this evidence cannot show

    Abstract-level; the Europe PMC core record returns no pmcid. The cell type used for immunocytochemistry is not stated in the abstract, and the localisation is of wild-type AF3p21, not of the fusion. In the same patient the der(3) chromosome formed no fusion transcript, so this is a reciprocal-negative case.

    Evidence that argues against this (1)
    • A bipartite NLS predicted from sequence and an exclusively nuclear image together still do not establish that the endogenous protein is nuclear in a haematopoietic cell, which is the cell type the fusion arises in. Cheung N et al. 2004
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Overexpression of SEPT6 alone did not shift its physiological cytoplasmic localisation, in the same confocal experiment in which MLL-SEPT6 gave small nuclear speckles - the design that makes a partner-baseline claim defensible.

    PMID 15761502DOI
    What this evidence cannot show

    293T cells, transient transfection. The baseline control is for an overexpressed partner in a non-haematopoietic line, not for endogenous SEPT6 in a blood cell.

    Evidence that argues against this (1)
    • Sept6-null mice show no gross abnormality and Sept6 loss does not alter MLL-SEPT6-induced disease, so the cytoplasmic baseline, though real, may carry no functional weight. Eguchi M et al. 2004
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In MLL-AF6-rearranged cells endogenous AF6 moves from cytoplasm to nucleus and silencing the fusion restores the cytoplasmic pool - a design in which the endogenous partner's compartment change is itself the measurement.

    PMID 24695851DOI
    What this evidence cannot show

    One partner, one fusion; the RAS-GTP readout is a functional consequence rather than an independent localisation measurement.

    Evidence that argues against this (1)
    • If a partner is already predominantly nuclear in haematopoietic cells, as EEN is, the sequestration design has no signal to detect and a negative result would be uninformative rather than reassuring. Cheung N et al. 2004
Connected to