Oncofusion

start / graph / concept-dpy30-what-is-established-for-wild-type-kmt2a-and-what-for-the-fusion

Explainerstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q23-wrad-adjudication

DPY30: the founding negative does not cover it, and it has no KMT2A contact to lose

also called DPY30 versus the fusion · DPY-30 evidence split · HDCMD6P

DPY30 is the odd one out. It was not even known to be part of this machine when the famous 2004 paper was written, and it does not touch KMT2A at all - it holds onto ASH2L. So the usual argument, that the fusion deletes the surface the protein grips, does not apply to it in the first place.

The same thing, for a scientist

**(a) Wild-type.** DPY30 (Q9C005, 99 aa) homodimerises and docks the ASH2L SDI motif at 2:1 stoichiometry; it makes **no direct contact with KMT2A**. Functionally it contributes only modestly to allosteric stimulation of KMT2 activity in vitro while its ASH2L interface is critical for correct genome-wide placement of H3K4me2/me3 [E1]. It is required in haematopoiesis: conditional Dpy30 deletion in adult mouse causes severe pancytopenia [E2], and knockdown impairs human CD34+ progenitor proliferation [E3]. **(b) Fusion.** Nothing, and the founding negative does not reach it. The string 'DPY' appears zero times in Yokoyama 2004 [E4]; DPY30 was first placed in the human SET1-like complexes in 2007 [E5]. The alias-expanded fusion query returns 3 records, none an association experiment [E6]. **(c) Domain.** DPY30's binding partner is ASH2L, not KMT2A. The retained/lost framing therefore applies to it only at one remove: DPY30 reaches KMT2A through ASH2L, which reaches the SET domain, which the direct fusion lacks. That two-step dependence is why DPY30 is the subunit for which 'does it associate with the fusion?' has the least determinate answer of the four.

Go deeper
Why DPY30 is categorically different from the other three

WDR5, RBBP5 and ASH2L each have a mapped physical contact with KMT2A, and each contact is C-terminal to the breakpoint. DPY30 has no mapped KMT2A contact at all. Its published interface is with the ASH2L SDI motif 1, and structurally its position in the MLL1-WRAD complex on a nucleosome has been described as ambiguous even in recent cryo-EM 7.

This has a consequence the graph has not stated. The clean structural argument - 'the fusion deletes the surface, therefore no association' - is an argument about WDR5, RBBP5 and ASH2L. For DPY30 the argument has to be inherited through ASH2L, and inherited arguments are where errors hide. If ASH2L were ever shown to reach a fusion by a non-SET route, DPY30 would come with it, and no experiment would have had to change.

The scope error this corrects

Four entries in this graph carry the evidence block E_yokoyama2004 with the claim that 'oncogenic mutant (fusion) forms of MLL retain the ability to interact with menin but not the other identified complex components', and one of the four is gene-dpy30. The paper does not mention DPY30, and DPY30 was not an identified complex component in 2004. The claim text is a faithful paraphrase of the abstract; it is the placement in gene-dpy30 that is wrong, because it reads as a negative about DPY30 and the source cannot supply one. The fix is to restrict that block's scope where it appears in gene-dpy30, not to remove it - the paper is still the right citation for the complex's discovery.

The co-dependency story, and where it ended up

DPY30 entered this graph as the one WRAD subunit whose leukaemia dependency tracks MEN1 (r = 0.446, Holm p = 0.015 across 54 lines). That result has since failed four independent designs in this project - a variance-matched null over a 2,743-gene pool, a within-subtype test, an essentiality-adjusted test and a KMT2A-rearrangement-specificity scan - and the entries recording the failures are already in the graph. The current position is that DPY30's correlation with MEN1 is reliably non-zero and not distinctive, which is not a reason to pick anything. This entry does not reopen it; it records that the co-dependency argument was never an argument about physical association with the fusion in the first place, and cannot substitute for one.

Aliases, and a clean zero

DPY30 publishes as DPY30, DPY-30, DPY30L and HDCMD6P. Bare-symbol counts: DPY30 119 records, DPY30L 1, HDCMD6P 0. The retired symbol carries no hidden literature at all, which is unusual and worth recording so nobody re-runs it. The fusion-paired query over all four spellings returns 3 records: a 2026 review of the ASH2L-DPY30 interface, a peptide-inhibitor paper, and a WDR5 review.

Facts and figures
subunitDPY30
uniprotQ9C005
length AA99
wild type contactthe ASH2L SDI motif at 2:1 stoichiometry; NO direct KMT2A contact is mapped
contact retained in direct fusionnot applicable - DPY30 does not contact KMT2A directly; it inherits ASH2L's SET-domain dependence
founding negative covers dpy30false
dpy30 mentions in yokoyama 20040
year placed in human set1 complexes2007
fusion association evidencenone published
pubmed records subunit x fusion3
pubmed query(DPY30 OR "DPY-30" OR DPY30L OR HDCMD6P) AND (MLL-AF4 OR MLL-AF9 OR MLL-ENL OR MLL-AF10 OR MLL-AF6 OR MLL-ELL OR "MLL fusion" OR "MLL fusion protein" OR "KMT2A fusion" OR "MLL-rearranged" OR "KMT2A-rearranged")
verdictno direct KMT2A contact at all; the 2004 negative predates its discovery and does not cover it; no measured association with any KMT2A fusion protein
alias record counts
HDCMD6P0.000HDCMD6P: 0DPY30L1.000DPY30L: 1DPY30119.0DPY30: 119
aliases searched
[
  "DPY30",
  "DPY-30",
  "DPY30L",
  "HDCMD6P"
]
alias record counts — the numbers behind the chart
{
  "DPY30": 119,
  "DPY30L": 1,
  "HDCMD6P": 0
}
How this could bite us back
  • Conditional Dpy30 deletion in adult mice causes severe pancytopenia, so the on-target marrow toxicity of a DPY30 agent is documented before any efficacy has been.
  • The ASH2L-derived peptides that define this target's tractability also inhibit MYC-dependent cancers without KMT2A rearrangement, so the modality is not fusion-selective and an efficacy signal could not be attributed to the fusion.
  • DPY30 is 99 residues; gnomAD LOEUF cannot be powered at that length, so the human-constraint reassurance previously recorded for this target is not evidence of tolerability.
  • DPY30 sits in all six SET1/MLL complexes; any agent removes H3K4-methylation positioning globally, in a patient population of infants whose normal haematopoiesis has no reserve.
Still open
  1. Does DPY30 reach a KMT2A fusion protein by any route at all? Because it has no direct KMT2A contact, the only available routes are through ASH2L or through a bridge nobody has proposed, and neither has been tested.
  2. Is the anti-leukaemia activity of ASH2L-derived DPY30-binding peptides a KMT2A-axis effect or a MYC effect? The same peptides work in MYC-dependent cancers without KMT2A rearrangement, and the two explanations have not been separated.
  3. Does DPY30 loss reduce H3K4me2/me3 preferentially at KMT2A-fusion target loci rather than globally? That is the direct test of the positioning-not-catalysis model and it is one CUT&RUN experiment in SEM cells.
Every source behind this entry (8)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    DPY30 homodimerises and docks the ASH2L SDI motif at 2:1 stoichiometry, contributes only modestly to positive heterotypic allosteric regulation of KMT2 methyltransferase activity in vitro, and its extensive DPY30-ASH2L interface is critical for correct genome-wide placement of H3K4me2 and H3K4me3.

    PMID 30270175DOI
    What this evidence cannot show

    Structural and biochemical work in a non-leukaemia system; it does not address MEN1, KMT2A fusions, or whether the placement function matters in a transformed cell.

    Evidence that argues against this (1)
    • The catalytic output this placement function serves is genetically dispensable for MLL-AF9 leukaemogenesis in mouse, so correct placement of H3K4 methylation is not established as load-bearing in this disease. Mishra BP et al. 2014
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Conditional Dpy30 deletion in the adult mouse haematopoietic system causes severe pancytopenia with accumulation of HSCs and early progenitors defective in multilineage reconstitution, and Dpy30-deficient HSCs fail to sustain long term in mixed chimeras.

    Dpy30 is critical for maintaining the identity and function of adult hematopoietic stem cells. · Yang Z, Shah K, Khodadadi-Jamayran A, et al. · J Exp Med 2016
    PMID 27647347DOI
    What this evidence cannot show

    Complete genetic deletion in mouse, which is a harsher perturbation than partial pharmacological inhibition in an adult human; it bounds tolerability rather than predicting it.

    Evidence that argues against this (1)
    • DPY30-binding peptides derived from ASH2L inhibited the growth of MLL-rearranged and MYC-dependent blood cancer cells, which argues a partial pharmacological block can be tolerated where complete genetic deletion is not. Shah KK et al. 2019
  3. E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    DPY30 is required for the growth of several MLL1-fusion-driven leukaemia cell lines, and DPY30 knockdown in human CD34+ progenitors impairs proliferation and myelomonocytic differentiation while promoting haemoglobin production.

    PMID 25139354DOI
    What this evidence cannot show

    shRNA knockdown in cell lines and primary progenitors. 'Required for growth of MLL1-fusion-driven lines' is a dependency observation, not evidence of association with the fusion protein, and no KMT2A-wild-type leukaemia comparator establishes selectivity.

    Evidence that argues against this (1)
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The string 'DPY' does not appear anywhere in the full text of Yokoyama 2004, so the paper's statement that fusion MLL retains menin but not other identified complex components cannot be a statement about DPY30.

    PMID 15199122DOI
    What this evidence cannot show

    Checked against the PMC HTML record flattened to 68,159 characters of text; figures were not OCR'd, so a label appearing only inside an image would be missed. The publisher does not release full-text XML for this article, so the HTML route is the only one available.

    Evidence that argues against this (1)
    • DPY30 nevertheless reaches KMT2A through ASH2L, which the same paper did map to the SET domain, so the 2004 result bears on DPY30 indirectly even though it never names it. Haddad JF et al. 2018
  5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    hDPY-30 was identified as a component of the human Set1-like H3K4 methyltransferase complexes alongside ASH2L, RBBP5 and WDR5 in 2007, three years after the founding paper was published.

    PMID 17500065DOI
    What this evidence cannot show

    The paper's subject is the MLL3/MLL4 complex and PTIP; hDPY-30 is named in a composition list rather than characterised. An earlier incidental mention cannot be excluded but would still postdate 2004.

    Evidence that argues against this (1)
    • DPY30's assignment to these complexes has been revisited on stoichiometric rather than co-purification grounds, and a systematic reconstitution shows the MLL1 core complex assembles hierarchically with occupancy dependent on temperature and concentration, so 'component' is a conditional status. Namitz KEW et al. 2023
  6. E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    An alias-expanded PubMed query pairing DPY30, DPY-30, DPY30L or HDCMD6P with the named common KMT2A fusions returns 3 records, none of which tests association between DPY30 and a fusion protein; bare-symbol counts are DPY30 119, DPY30L 1 and HDCMD6P 0.

    What this evidence cannot show

    Counted absence bounded by four spellings. The yeast/worm orthologue names Sdc1 and dpy-30 were not searched against KMT2A fusions because no KMT2A fusion exists in those organisms. Complex-level framing and supplementary-only data remain invisible.

    Evidence that argues against this (1)
    • A paper framing its result as 'the MLL1 core complex' or 'COMPASS' rather than naming DPY30 would not be retrieved, and this project has previously produced a wrong absence claim exactly that way. Schmidt L et al. 2019
  7. E7strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In multistate cryo-EM of the MLL1-WRAD complex bound to an H2B-ubiquitinated nucleosome, portions of the ASH2L structure and the position of DPY30 remain ambiguous, and MLL1-WRAD docking on nucleosomes is variable.

    Multistate structures of the MLL1-WRAD complex bound to H2B-ubiquitinated nucleosome. · Rahman S, Hoffmann NA, Worden EJ, et al. · Proc Natl Acad Sci U S A 2022
    PMID 36095189DOI
    What this evidence cannot show

    Reconstituted complexes on designer nucleosomes at limited resolution; ambiguity in a density map is a statement about the experiment as well as about the molecule.

    Evidence that argues against this (1)
    • A crystal structure of the DPY-30 dimerisation/docking domain bound to the ASH2L SDI motif resolves that specific interface precisely, so the ambiguity concerns DPY30's position relative to the nucleosome rather than its contact with ASH2L. Haddad JF et al. 2018
  8. E8moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    Cell-penetrating peptides derived from ASH2L that bind the DPY30 ASH2L-binding groove inhibit DPY30's interaction with ASH2L, reduce H3K4 methylation, and significantly inhibit the growth of MLL-rearranged leukaemia and other MYC-dependent haematologic cancer cells.

    PMID 31251903DOI
    What this evidence cannot show

    Cell-penetrating peptides in culture; no in vivo data, no pharmacokinetics, and 'MLL-rearranged' here does not specify a t(4;11) or infant system. Growth inhibition in MYC-dependent cancers alongside MLL-rearranged cells argues the mechanism is not KMT2A-specific.

    Evidence that argues against this (1)
    • DPY30 was independently characterised as the chromatin modulator that MYC hijacks, with DPY30 promoting endogenous MYC expression - so a DPY30 agent's activity in MLL-rearranged cells may run through MYC rather than through anything KMT2A-related. Yang Z et al. 2018
Connected to