Oncofusion

start / graph / import-class-dot1l-af10-om-lz

Explainerstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q-partner-provenance

Import class 3: the DOT1L-AF10 coiled-coil axis, an MLLT10 mechanism with no counterpart in the index fusion

also called OM-LZ import · AF10 octapeptide leucine zipper · DOT1L-AF10 higher-order complex

AF10 grabs the H3K79-writing enzyme through a coiled stretch of protein, and that grip has to form a larger assembly to drive leukaemia. The infant fusion MLL-AF4 contains no such stretch, so mechanism described through this route is about AF10-containing leukaemias.

The same thing, for a scientist

The second documented route to DOT1L is the AF10 octapeptide-motif/leucine-zipper region. UniProt P55197 annotates residues 703-784 as *'Transactivation domain; required for DOT1L-binding'* with the leucine zipper at 750-778 [E1]. Structural analysis of the DOT1L-AF10 complex gives mechanistic insight framed explicitly around MLL-AF10 leukaemia [E2], and a **higher-order configuration of the heterodimeric coiled coils** - not simple heterodimerisation - potentiates leukemogenic activity [E3]. Endogenous AF10 also directs progressive H3K79 methylation and is required for transformation by MLL-AF9, MLL-AF6 and NUP98-NSD1 [E4]. In the CRISPR-ChIP partition, MLLT10 directs the **endogenous** DOT1L complex [E5]. **Fourteen entries in this graph cite at least one of these papers** [E6]. AFF1 carries no leucine-zipper or coiled-coil family in its UniProt Pfam complement, so this axis reaches KMT2A::AFF1 only through *endogenous* AF10 acting on genes the fusion has activated - which is a different claim from the one an imported entry usually makes.

Go deeper
The partner-specificity trap this class sets

AF10 is unique among the players here in being both the endogenous director of H3K79 methylation and a recurrent fusion partner, in KMT2A::MLLT10 and PICALM::MLLT10. So the two-unit model of H3K79 regulation - oncogenic unit directed by the fusion, endogenous unit directed by MLLT10 - collapses into a single unit in exactly those genotypes, and the sign of every prediction about menin inhibition changes there.

For KMT2A::AFF1 the trap is different and quieter. The index fusion has no AF10 and no AHD, so if the CRISPR-ChIP partition is correct and complete, all H3K79 methylation in t(4;11) cells should belong to the endogenous MLLT10-directed unit. That is a strong, testable, and slightly uncomfortable prediction: it implies menin inhibition should barely move H3K79me2 in infant disease.

Higher-order assembly, not binding, is what this axis contributes

The 2019 PNAS result is the one most often compressed into 'AF10 binds DOT1L'. What it shows is that a higher-order configuration of the two coiled-coil domains potentiates leukemogenic activity 3. That is an oligomeric-state claim, and oligomeric-state claims transfer across partners worse than binding claims, because they depend on local concentration and valency rather than on a conserved contact.

This also links the class to the condensate literature: assemblies whose activity depends on multivalency are the ones for which 1,6-hexanediol and classical FRAP cannot distinguish phase separation from clustered binding, so a mechanistic reading of this axis should say which assay produced it.

Facts and figures
classinterface import - structurally inapplicable to the index fusion
interfaceOM-LZ (P55197 703-784; leucine zipper 750-778)
entries citing class14
residual route for index fusionendogenous AF10 acting on genes the fusion has activated
uncomfortable predictionif the partition is complete, all H3K79me2 in t(4;11) cells belongs to the endogenous MLLT10-directed unit, so menin inhibition should barely move it
carried by
[
  "MLLT10"
]
absent from
[
  "AFF1"
]
How this could bite us back
  • If endogenous AF10 supplies the H3K79 methylation that t(4;11) cells need, then DOT1L inhibition in infant disease is removing an endogenous activity that normal haematopoiesis also runs on, and the therapeutic index is narrower than a fusion-directed model implies.
  • Targeting the DOT1L-AF10 coiled coil would be a fusion-agnostic intervention marketed as a fusion-specific one.
Still open
  1. Does MLLT10 knockout reduce H3K79me2 at KMT2A::AFF1 target genes in a human t(4;11) line, and by how much relative to a MLL-AF9 line?
  2. Is the higher-order coiled-coil assembly detectable at endogenous expression, or only in overexpression systems where multivalent assemblies form readily?
Every source behind this entry (6)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProtKB P55197 (primary gene name MLLT10, 1,068 aa) annotates residues 703-784 as 'Transactivation domain; required for DOT1L-binding' and 750-778 as a leucine zipper.

    What this evidence cannot show

    Accession fetched directly on 2026-09-19 with primary gene name and length read back. UniProt regions are curatorial calls drawn from primary literature, not crystallographic footprints; the boundaries should be treated as annotation rather than measurement.

    Evidence that argues against this (1)
    • A region annotated as both a transactivation domain and the DOT1L-binding element conflates two functions; a construct deleting 703-784 removes transactivation as well, so separation of function is not established by the annotation itself. Zhang H et al. 2018
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Structural and functional analysis of the DOT1L-AF10 complex gives mechanistic insight into MLL-AF10-mediated leukemogenesis.

    Structural and functional analysis of the DOT1L-AF10 complex reveals mechanistic insights into MLL-AF10-associated leukemogenesis. · Zhang H, Zhou B, Qin S, Xu J, Harding R, Tempel W, Nayak V, Li Y, Loppnau P, Dou Y, Min J · Genes Dev 2018
    PMID 29563185DOI
    What this evidence cannot show

    Identifier, title, journal, year and DOI fetched from PubMed on 2026-09-19 and copied from that record; the abstract was not re-read in this cycle and the claim is stated at the level of the title, so no structural detail is asserted here. The system is MLL-AF10 throughout.

    Evidence that argues against this (1)
    • A structure of an isolated complex does not establish that the same geometry forms on chromatin in a fusion context, and this graph holds the general caution that a PDB entry existing is not a structure suitable for design or for mechanism. Song X et al. 2019
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    A higher-order configuration of the heterodimeric DOT1L-AF10 coiled-coil domains potentiates their leukemogenic activity.

    A higher-order configuration of the heterodimeric DOT1L-AF10 coiled-coil domains potentiates their leukemogenenic activity. · Song X, Yang L, Wang M, Gu Y, Ye B, Fan Z, Xu RM, Yang N · Proc Natl Acad Sci U S A 2019
    PMID 31527241DOI
    What this evidence cannot show

    Identifier, title, journal, year and DOI fetched from PubMed on 2026-09-19 and copied from that record; abstract not re-read this cycle, so the claim is asserted only at the level of the title's own statement. An oligomeric-state claim depends on local concentration and valency, which are properties of the assay as much as of the protein.

    Evidence that argues against this (1)
    • Multivalent assembly claims are exactly the class for which the two commonest assays - 1,6-hexanediol sensitivity and classical FRAP - cannot distinguish phase separation from binding to clustered sites, so 'higher-order configuration' may describe the readout rather than the mechanism. Zhang H et al. 2018
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Endogenous AF10 regulates progressive H3K79 methylation and HOX expression, and AF10 inactivation impairs transformation by MLL-AF9, MLL-AF6 and NUP98-NSD1.

    AF10 regulates progressive H3K79 methylation and HOX gene expression in diverse AML subtypes. · Deshpande AJ, Deshpande A, Sinha AU, Chen L, Chang J, Cihan A, Fazio M, Chen CW, Zhu N, Koche R, Dzhekieva L, Ibáñez G, Dias S, Banka D, Krivtsov A, Luo M, Roeder RG, Bradner JE, Bernt KM, Armstrong SA · Cancer Cell 2014
    PMID 25464900DOI
    What this evidence cannot show

    Abstract fetched from PubMed and read on 2026-09-19; full text not read. Mouse models; all three fusions tested are myeloid and none is KMT2A::AFF1, so the transferability argument this evidence supports is itself untested in the index fusion.

    Evidence that argues against this (1)
    • If endogenous AF10 is required for transformation by fusions that do not contain it, the 'endogenous' unit is part of the oncogenic requirement rather than a separable housekeeping activity, which blurs the two-unit partition at its most important boundary. Gilan O et al. 2023
  5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    CRISPR-ChIP assigns the endogenous DOT1L complex's catalytic direction to MLLT10, separately from a menin-dependent fusion-directed unit.

    CRISPR-ChIP reveals selective regulation of H3K79me2 by Menin in MLL leukemia. · Gilan O, Talarmain L, Bell CC, Neville D, Knezevic K, Ferguson DT, Boudes M, Chan YC, Davidovich C, Lam EYN, Dawson MA · Nat Struct Mol Biol 2023
    PMID 37679565DOI
    What this evidence cannot show

    Abstract read from the efetch record on 2026-09-19; full text unreachable (Europe PMC not open access, Unpaywall closed, bioRxiv HTTP 429 on four attempts). The deposited data show every H3K79 experiment is murine MLL-AF9, so the partition itself is an import - see ana-crispr-chip-fusion-identity-resolved. Confidence reduced accordingly. **Regraded moderate -> established on 2026-09-20** under `concept-the-confidence-grading-rule`: the assignment of catalytic direction to MLLT10 is the paper's own result.

    Evidence that argues against this (1)
    • MLLT10 knockout collapsing part of the H3K79 signal in a reporter-based screen is equally consistent with AF10 being required for DOT1L stability or nuclear localisation generally, rather than directing a distinct complex. Deshpande AJ et al. 2014
  6. E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    Fourteen of the 848 entries in this graph cite at least one of PMIDs 25464900, 29563185, 31527241 or 33993518.

    What this evidence cannot show

    Our own count by literal PMID string match over node JSON on 2026-09-19; counts an entry once whether the citation is load-bearing or a counter-evidence note, and misses entries describing AF10-DOT1L mechanism without citing these four papers.

    Evidence that argues against this (1)
    • Many of the fourteen are AF10/MLLT10 entries where this mechanism is the subject rather than an import, so the number of entries actually importing the axis into the index fusion is smaller. Oncofusion knowledge graph et al. 2026
Connected to