Oncofusion

start / graph / import-class-murine-mll-af9-myeloid

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

Import class 1: murine MLL-AF9 myeloid transformation, the field's default assay and this graph's largest single import

also called mouse MLL-AF9 import · myeloid progenitor immortalisation import

Most of what is known about how these fusion proteins work was learned by putting MLL-AF9 into mouse blood cells and seeing whether they turn into leukaemia. That assay is fast, reliable and answers a lot - but it is a different fusion, a different species and a different blood lineage from infant MLL-AF4 leukaemia.

The same thing, for a scientist

The dominant experimental system in KMT2A biology is retroviral MLL-AF9 (or MLL-ENL) transduction of murine haematopoietic progenitors, read out as serial replating, immortalisation, or transplantation. It underlies the founding DOT1L dependency result [E1], the wild-type-allele requirement [E2], the SET-domain dispensability result [E3], and the PAF1c synergy result [E4]. Forty-six entries in this graph cite at least one of those five keystone papers [E5]. **The import is threefold and the three parts have different strengths.** *Species*: usually the weakest concern for direct protein contacts, strongest for expression and dosage. *Lineage*: the assay scores myeloid progenitor immortalisation, and the index disease is B-lymphoid - a difference that is not cosmetic, because the same Mll-AF4 allele in mice produces either B-precursor ALL or AML [E6]. *Fusion*: MLL-AF9 carries the AF9 AHD, a DOT1L-binding module KMT2A::AFF1 does not have, which is why H3K79 mechanism specifically does not transfer (`mechanism-kmt2a-aff1-dot1l-recruitment-route`). **What this class gets right.** Requirement claims - *this gene is needed for transformation* - transfer reasonably well, because they identify bottlenecks rather than contacts. **What it gets wrong.** Contact claims and dose-response claims: *DOT1L recruitment scales with AHD occupancy* is true and is a statement about AF9.

Go deeper
Why the field chose this assay, which is not an accident to be corrected

Murine MLL-AF9 myeloid transformation is fast (replating in days), penetrant, transplantable, and genetically tractable with conditional alleles. The index fusion has no comparably tractable model: chimeric Mll-Af4 knock-ins are slow and variable, human t(4;11) lines are hard to engineer, and infant primary material is scarce. A demand that every mechanism be re-derived in KMT2A::AFF1 would stop the work rather than correct it. The useful response is to know which claims the assay can carry across the boundary, not to discount it.

A worked contrast inside a single paper

Krivtsov et al. 2008 is the clearest illustration that lineage is a real variable rather than a caveat-phrase: a single conditional Mll-AF4 allele produces B-precursor ALL in some mice and AML in others, and the H3K79 profiles of the resulting ALLs match human MLL-rearranged ALL 6. The fusion is held constant and the disease changes. Any result read out only in myeloid progenitors has, by construction, sampled one of the two outcomes that fusion can produce.

Facts and figures
classspecies + lineage + fusion import
source systemmurine haematopoietic progenitors transduced with MLL-AF9
entries citing keystone papers46
keystone pmids
[
  "21741597",
  "20159607",
  "24813891",
  "20541477",
  "19956800"
]
transfers well
[
  "requirement/bottleneck claims",
  "contacts mapped to the KMT2A moiety"
]
transfers badly
[
  "contacts mapped to the 3' partner",
  "dose-response between recruitment and phenotype",
  "lineage-specific downstream programmes"
]
How this could bite us back
  • Rejecting this class wholesale would remove the founding evidence for DOT1L as a target in this disease, which no index-fusion experiment replaces at the same depth.
  • Accepting it wholesale means the graph's mechanistic spine is myeloid, and an infant B-ALL programme built on it inherits target genes and differentiation endpoints that may not be the ones enforced in a pro-B cell.
Still open
  1. How many of the 46 entries make a contact claim rather than a requirement claim? That subset is the one whose transfer to KMT2A::AFF1 is actually at risk.
  2. Does any published side-by-side compare a murine MLL-AF9 result with the same experiment in a murine Mll-Af4 B-ALL model? Krivtsov 2008 built the model; the comparison does not appear to have been run systematically.
Every source behind this entry (6)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Abnormal chromatin at MLL-AF9 target loci in leukaemia stem cells is specific to H3K79me2; Dot1l inactivation downregulates direct MLL-AF9 targets and the MLL-translocation signature, and MLL-AF9 leukaemia is Dot1l-dependent in vivo.

    MLL-rearranged leukemia is dependent on aberrant H3K79 methylation by DOT1L. · Bernt KM, Zhu N, Sinha AU, Vempati S, Faber J, Krivtsov AV, Feng Z, Punt N, Daigle A, Bullinger L, Pollock RM, Richon VM, Kung AL, Armstrong SA · Cancer Cell 2011
    PMID 21741597DOI
    What this evidence cannot show

    Abstract fetched from PubMed and read on 2026-09-19; full text not read. Conditional Dot1l deletion in a murine MLL-AF9 leukaemia - a triple import (mouse, myeloid, AF9). This is the founding dependency result for the entire DOT1L programme in this disease and it was never run in KMT2A::AFF1.

    Evidence that argues against this (1)
    • The index-fusion version of this dependency does exist, earlier and in a different form: H3K79 profiles define murine and human MLL-AF4 leukaemias and DOT1L suppression silences MLL-AF4 targets - so the DOT1L dependency is not purely an AF9 import even though its mechanistic account is. Krivtsov AV et al. 2008
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    The wild-type Mll allele is required for MLL-AF9-induced leukemogenesis and for maintenance of MLL-AF9-transformed cells, with menin recruiting both wild-type MLL and the fusion to HOX loci.

    MLL-AF9-induced leukemogenesis requires coexpression of the wild-type Mll allele. · Thiel AT, Blessington P, Zou T, Feather D, Wu X, Yan J, Zhang H, Liu Z, Ernst P, Koretzky GA, Hua X · Cancer Cell 2010
    PMID 20159607DOI
    What this evidence cannot show

    Abstract fetched from PubMed and read on 2026-09-19; full text not read. Mouse genetics in one fusion and a transplantation model; no human allele-dosage data exist. 'Required' is binary in that design, so the quantitative dependence on wild-type dosage is unmeasured.

    Evidence that argues against this (1)
    • If the wild-type allele requirement is partly about competition for menin and PAF1c rather than about the fusion itself, the requirement could differ in a fusion whose transcriptional output depends less on the same cofactors - and no one has repeated the experiment in a KMT2A::AFF1 background. Muntean AG et al. 2010
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    The MLL1 SET domain and its methyltransferase activity are dispensable for haematopoietic stem-cell maintenance and for MLL-AF9-driven leukemogenesis.

    The histone methyltransferase activity of MLL1 is dispensable for hematopoiesis and leukemogenesis. · Mishra BP, Zaffuto KM, Artinger EL, Org T, Mikkola HK, Cheng C, Djabali M, Ernst P · Cell Rep 2014
    PMID 24813891DOI
    What this evidence cannot show

    Abstract fetched from PubMed and read on 2026-09-19; full text not read. Mouse SET-domain deletion and point-mutant models with MLL-AF9 in a myeloid context; a catalytically dead SET domain still occupies chromatin, so this tests catalysis rather than scaffolding. Does not address KMT2A::AFF1.

    Evidence that argues against this (1)
    • The SET domain is lost from the fusion in every KMT2A translocation, so its dispensability for the FUSION is trivially expected; the load-bearing part of this result is about the wild-type allele, where the import question is about species and lineage rather than about the partner. Thiel AT et al. 2010
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    MLL interacts directly with the PAF complex through sequences flanking the CxxC domain; PAFc augments MLL- and MLL-AF9-mediated activation of Hoxa9, and deletions abolishing the interaction eliminate MLL-AF9-mediated immortalisation.

    The PAF complex synergizes with MLL fusion proteins at HOX loci to promote leukemogenesis. · Muntean AG, Tan J, Sitwala K, Huang Y, Bronstein J, Connelly JA, Basrur V, Elenitoba-Johnson KS, Hess JL · Cancer Cell 2010
    PMID 20541477DOI
    What this evidence cannot show

    Abstract fetched from PubMed and read on 2026-09-19; full text not read. The fusion tested is MLL-AF9 and the immortalisation assay is murine; the interaction is established for PAFc as a complex and the abstract does not say which subunit contacts MLL. This paper is the load-bearing citation for several PAF1c subunit entries in this graph, one of which (gene-leo1) does not record its provenance.

    Evidence that argues against this (1)
    • The PAFc contact maps to the KMT2A moiety, not the partner, so this is one of the claims most likely to survive the partner boundary - which makes the unflagged import in gene-leo1 a provenance defect rather than a substantive error. Yokoyama A et al. 2010
  5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    Forty-six of the 848 entries in this graph cite at least one of PMIDs 21741597, 20159607, 24813891, 20541477 or 19956800, the five murine MLL-AF9 keystone papers.

    What this evidence cannot show

    Our own count over the node JSON files on 2026-09-19, by literal PMID string match. It counts an entry once whether the paper is its load-bearing citation or a passing counter-evidence note, so it over-states load-bearing use; it also misses entries that describe the same results without citing these five papers, so it under-states the class's true reach.

    Evidence that argues against this (1)
    • A citation count is a measure of how much a paper is used, not of how much a claim depends on it. Some of these 46 cite the keystone papers precisely in order to disclaim them, which the count cannot see. Oncofusion knowledge graph et al. 2026
  6. E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    A conditional murine Mll-AF4 allele induces either B-precursor acute lymphoblastic leukaemia or acute myeloid leukaemia, and the ALL expression profile overlaps human MLL-rearranged ALL.

    H3K79 methylation profiles define murine and human MLL-AF4 leukemias. · Krivtsov AV, Feng Z, Lemieux ME, Faber J, Vempati S, Sinha AU, Xia X, Jesneck J, Bracken AP, Silverman LB, Kutok JL, Kung AL, Armstrong SA · Cancer Cell 2008
    PMID 18977325DOI
    What this evidence cannot show

    Abstract fetched from PubMed and read on 2026-09-19; full text not read. The abstract does not give the proportion of mice developing each disease, so 'either' is qualitative. A chimeric knock-in Mll-AF4, not the human t(4;11) product.

    Evidence that argues against this (1)
    • Lineage plasticity from one allele could mean lineage is downstream of a shared programme, in which case myeloid readouts are more transferable than this entry suggests, not less. Vu LP et al. 2017
Connected to