CALM-AF10 reaches DOT1L through exactly the same OM-LZ interface as KMT2A::MLLT10 - the recruitment transfers, the addressing does not

Structural and functional work framed around **MLL1-AF10**, not CALM-AF10. The retained AF10 module is the same, so the interface transfers by construction; the octamerisation REQUIREMENT has not been separately tested in CALM-AF10.

• Per the brief's condensate and oligomerisation caveats, a higher-order assembly requirement demonstrated with engineered constructs is not the same as showing the assembly forms at endogenous stoichiometry in a patient cell. Song X et al. 2019

Murine retroviral minimal construct; transforms more efficiently than the full-length fusion, so it is a tool rather than a model of the patient protein.

• PUBLISHED CONFLICTING RESULT on the attribution, from a mouse model of the same region. This item's title attributes the minimal construct's activity to the CLATHRIN-BINDING domain. The model built to test that attribution found against it: excluding the C-terminal portion of CALM required for optimal clathrin binding produced myeloproliferative disease rather than AML - so the region matters - but the development of leukaemia COULD NOT BE ATTRIBUTED DIRECTLY to interference with endocytosis, full-length fusion showed no consistent effect on growth-factor endocytosis, and the authors propose HOMO-OLIGOMERISATION, measured by FRET, as what endocytic 5' partners actually supply. Sufficiency of the fragment is not in dispute; what the fragment does is. Stoddart A et al. 2012

Murine bone-marrow serial colony-plating. **Verified in the paper's full text on 2026-09-20 via the PMC HTML route (PMC4425349), not from the abstract**: the paper states that the capacity to generate a significant number of third-round colonies is dependent on the ability of CALM-AF10 to interact with hDOT1L, with the OM-LZ deletion mutant as the comparator. The same paper's headline mechanism - that hDOT1L contributes by PREVENTING nuclear export of CALM-AF10 - is in tension with later work showing that nuclear export is REQUIRED for transformation; see mechanism-calm-af10-crm1-nuclear-export-requirement.

• Okada et al. report that hDOT1L contributes to transformation partly by preventing nuclear export of CALM-AF10, i.e. by making the fusion MORE nuclear. Conway et al. later showed that a nuclear-restricted CALM-AF10 does not transform and that export is necessary. Both cannot be straightforwardly true and the graph should not smooth them. Conway AE et al. 2013

Murine bone marrow and HEK293 imaging.

• PUBLISHED CONFLICTING RESULT: three claims are being made about one piece of protein and none of the experiments separates them. CALM 544-553 is an NES; the same C-terminal span is the clathrin-binding region; and the same span confers a propensity to homo-oligomerise, shown by fluorescence resonance energy transfer. Excluding the clathrin-binding C-terminus changes the mouse disease from AML to a myeloproliferative disease, while the leukaemia itself could not be attributed to interference with endocytosis. So deleting or restoring this region moves at least three variables at once, and 'a bare NES fused to AF10 immortalises bone marrow' is also 'a bare oligomerisation-competent peptide fused to AF10 immortalises bone marrow'. Stoddart A et al. 2012

The system is AF10-translocation leukaemia including CALM-AF10; the mouse/human split is not resolved from the abstract alone. This is the best available answer to 'what addresses CALM-AF10 to chromatin' and it is an acetylation-reader answer, not a menin answer.

• PUBLISHED CONFLICTING RESULT about what the combination arm is worth. The cooperativity here is between two tool compounds in mice, and the clinical record for the DOT1L half of it, in the population where the H3K79 rationale is strongest, is thin: pinometostat in 51 adults with advanced acute leukaemia enriched for MLL rearrangement produced complete remission in 2 patients, both with t(11;19), with no maximum tolerated dose established and the authors describing the activity as modest. A MOZ/MORF-plus-DOT1L combination proposal inherits that, and the entry should not carry 'cooperatively induces differentiation' as though the DOT1L component had a clinical foundation. Stein EM et al. 2018

Murine CALM-AF10 AML with confirmation of the TGM2 arm in human U-937. A requirement for CTCF is not specific to this fusion - CTCF is required by many things - and the entry treats it as an architectural hint, not as a fusion-specific mechanism.

• PUBLISHED CONFLICTING RESULT that offers a less fusion-specific reading of the CTCF phenotype. The chromatin signature reported at TGM2 on CTCF loss - gain of H3K27me3, loss of H3K4me3 and H3K27ac - is a polycomb switch, and polycomb is independently required on BOTH sides of this graph's comparison: Bmi1 is indispensable for the development of MLL-AF9 leukaemia in vivo, with Bmi1-null cells showing enhanced differentiation and retaining fewer leukaemia stem cells. A CTCF knockout that induces macrophage-like differentiation through a promoter gaining H3K27me3 may therefore be reporting a general PRC1-mediated differentiation switch that any of these leukaemias would show, rather than an architectural dependency specific to CALM-AF10. Yuan J et al. 2011

Structural and cellular work on endogenous AF10. It establishes what the fusion LOSES; whether that loss matters is addressed in mechanism-calm-af10-pzp-loss-is-a-removed-brake.

• PUBLISHED CONFLICTING RESULT: the PZP requirement has the opposite sign in the fusion. In endogenous AF10, PZP recognition of unmodified H3K27 is REQUIRED for H3K79 dimethylation and for proliferation of DOT1L-addicted leukaemic cells. In CALM-AF10, RESTORING a functional AF10 PZP domain promotes nuclear localisation, is required for chromatin association - and ABOLISHES transformation and leukaemogenesis. The same module is necessary in one context and suppressive in the other. A single statement that 'PZP recognition of H3 is required for H3K79 dimethylation' cannot be carried into the fusion, and the cellular arm of the 2015 work was in KMT2A-rearranged lines rather than in any AF10 fusion. Klein BJ et al. 2021

Coordinates are in UniProt P55197-1 (1,068 aa) numbering, which is NOT the frame used by the CALM-AF10 clinical literature - the 2026 review works in a 1,027-residue AF10 frame and Conway et al. in a third (NCBI NP_001182555). Never carry an AF10 residue number between a CALM-AF10 paper and a structure without converting the frame.

• METHODOLOGICAL DEFEATER: a crystallised span is not a functional boundary, and adding it as a fourth frame makes the numbering problem worse rather than better. The constructs behind 6CKN, 6CKO and 6JN2 were chosen to diffract; the functional limits of the AF10 OM-LZ were set by deletion in a transformation assay, where a defined AF10 fragment fused to a CALM C-terminus suffices to induce AML in mice. There is no reason for a crystallisable span and a minimal transforming span to coincide, and quoting 704-779 or 699-782 as 'the OM-LZ' imports a structural-biology convenience into a claim about what the fusion needs. Deshpande AJ et al. 2011