Oncofusion

start / graph / exp-psip1-inducible-knockdown-established-kmt2a-aff1-xenograft

Experimentspeculative confidence 1 of 5: Our own inference or analogy. Not shown to be true.Q33-ledgf-axis

The experiment that closes the LEDGF hole: doxycycline-inducible PSIP1 knockdown switched on AFTER a human KMT2A::AFF1 xenograft is established

also called inducible LEDGF knockdown xenograft · PSIP1 maintenance experiment

Put a switch on the LEDGF gene in human leukaemia cells, let the leukaemia take hold in a mouse, then flip the switch off and see whether the leukaemia dies. That one experiment would tell us whether a LEDGF drug could treat a patient, rather than only prevent a disease that has already happened.

The same thing, for a scientist

Every published LEDGF-removal experiment in a KMT2A-rearranged animal removes LEDGF before transformation (`absence-ledgf-removal-from-established-kmt2a-leukaemia`). The decisive missing experiment is a maintenance experiment: transduce a human KMT2A::AFF1 line and an infant-derived KMT2A::AFF1 PDX with a doxycycline-inducible shRNA or a dTAG/AID degron against PSIP1, engraft into NSG or NSG-SGM3 mice, wait for confirmed disease burden by bioluminescence or peripheral blast percentage, then induce. The primary readout is survival; the mechanistic readouts are fusion and LEDGF occupancy at HOXA9/MEIS1 by CUT&RUN and HOXA9/MEIS1 transcript levels. The design must carry a degron arm, because the single largest interpretive risk is that an shRNA leaves enough residual protein to make a null result uninterpretable, and it must carry a lymphoid model, because all existing in vivo LEDGF data are myeloid (MLL-ENL, MLL-AF9) while the patients this project cares about have infant B-ALL.

Go deeper
Arms
ArmConstructInductionPurpose
1dox-inducible shPSIP1 (two independent hairpins)at confirmed burdenthe maintenance question
2dox-inducible shPSIP1at day 0, before engraftmentinternal prevention control, reproducing the published result in a human lymphoid system
3PSIP1-dTAG or AID degrondTAG-13 / auxin at confirmed burdenremoves the residual-protein escape from a negative result
4dox-inducible shSCRat confirmed burdendox and shRNA-machinery control
5dox-inducible shMEN1at confirmed burdenpositive control: menin is established as a maintenance dependency
6shPSIP1 + rescue with shRNA-resistant PSIP1 IBD point mutant (F406/D366-class IBD mutant)at confirmed burdenseparates the IBD arm from the PWWP arm
7shPSIP1 + rescue with PWWP-dead PSIP1at confirmed burdenthe other half of arm 6

Arms 6 and 7 are what turn a dependency result into a drug-discovery result, because a whole-protein knockdown cannot say which module to drug and the graph already disagrees with itself on that point (interface-psip1-pwwp-h3k36me-nucleosome versus target-phase2-psip1-ibd-dossier).

Models, and why the lymphoid one is not optional

SEM (line-sem-kmt2a-aff1) and RS4;11 (line-rs4-11-kmt2a-aff1-adult) are the workhorse human t(4;11) lines; MV4-11 (line-mv4-11-kmt2a-aff1-myeloid) carries the same fusion in a myeloid context and is therefore the internal lineage control rather than the index model. At least one infant-derived KMT2A::AFF1 PDX should be included, because the cell-state beat established this cycle that KMT2A-rearranged disease splits by lineage rather than by fusion across five independent axes, and because menin inhibition kills in the lymphoid disease while it differentiates in the myeloid one. An axis validated only in MOLM-13 and MLL-AF9 mice is validated in the wrong lineage for infant ALL.

Engraftment feasibility is not hypothetical: LEDGF-depleted human CD34+ cells engraft NSG mice in the HIV gene-therapy setting 4.

What each outcome would mean
  • Leukaemia regresses. LEDGF is a maintenance dependency and the IBD is a genuine second-line target, most interestingly in menin-inhibitor-resistant disease where the MEN1 gatekeeper mutations preserve the KMT2A contact 3.
  • Leukaemia is unaffected. The axis is an initiation-stage dependency only. That does not make it useless - it would make it a candidate for MRD or prevention in the infants with a detectable preleukaemic clone - but it removes the relapse indication that is the entire commercial and clinical case.
  • Leukaemia slows but does not regress. The most likely result, and the one that needs arm 5 to interpret: if shMEN1 in the same experiment produces regression and shPSIP1 produces delay, the two arms of the clamp are not equivalent and menin primacy survives.
  • Occupancy of the fusion at HOXA9/MEIS1 does not fall, but output does. This is the outcome the Zhu 2016 result predicts 2 and it would be the most informative of all: it would mean LEDGF is not a tether for the fusion in established disease but a functional cofactor, and that a PWWP-directed chromatin-anchoring inhibitor is targeting the wrong mechanism.
Cost, and why this has not been done

Nothing here is technically hard. Inducible knockdown in SEM, NSG engraftment and CUT&RUN are all routine, and the reagents exist. The likeliest explanation for a decade-old gap is division of labour: the LEDGF-leukaemia literature is dominated by one virology group (Debyser/De Rijck/Christ, Leuven) whose animal work is built around transduce-then-transplant designs inherited from HIV gene therapy, and by Cleary's group, whose LEDGF work is mechanistic and in vitro. A maintenance xenograft is a leukaemia-laboratory experiment that the LEDGF field has not had reason to run.

Facts and figures
designDoxycycline-inducible shRNA and orthogonal degron knockdown of PSIP1 in human KMT2A::AFF1 cell lines and an infant-derived PDX, induced only after leukaemic burden is confirmed in NSG/NSG-SGM3 recipients.
readoutPrimary: overall survival and leukaemic burden (bioluminescence, human CD45 chimerism). Secondary: LEDGF, KMT2A-N, menin and H3K36me2/3 CUT&RUN at HOXA9/MEIS1/PROM1/CDK6; HOXA9 and MEIS1 transcript levels; immunophenotype for differentiation.
controlsshSCR + dox; shMEN1 + dox as a positive maintenance control; the same constructs induced at day 0 as an internal prevention control; shRNA-resistant wild-type, IBD-mutant and PWWP-mutant PSIP1 rescue arms.
cost tiermed
timeline9-14 months including PDX expansion
falsifiable predictionIf LEDGF is a maintenance dependency, induction at established burden extends survival relative to shSCR by a margin comparable to shMEN1; if it is an initiation-only dependency, the day-0 arm separates and the established-burden arm does not.
How this could bite us back
  • A positive result licenses a LEDGF-directed programme into a protein that also performs DNA end resection, R-loop suppression and co-transcriptional splicing, all of which would be removed alongside it.
  • shRNA removes the whole protein; a drug would occupy one module. A positive knockdown result therefore over-predicts what an IBD ligand or a PWWP ligand could achieve, and the rescue arms exist to bound that over-prediction.
  • In T-ALL, loss of Psip1 ACCELERATES leukaemia initiation, so a systemic LEDGF-lowering agent used in a lymphoid disease carries a documented tumour-promoting risk in a neighbouring lineage.
  • If leukaemia regresses only in the myeloid model, the result would be read as validation of the axis while being irrelevant to the infant B-ALL population.
Still open
  1. Should the degron arm use dTAG or AID? dTAG needs a knock-in and is cleaner; AID is faster but leaks. In a PDX neither is trivial, and the choice may decide whether the experiment is done at all.
  2. What knockdown depth is needed before a negative result is interpretable? The menin-inhibitor literature suggests occupancy-based targets tolerate substantial residual protein, and no dose-response for LEDGF exists in any system.
  3. Should the readout include the reciprocal AFF1::KMT2A product, which a subset of t(4;11) patients express? No LEDGF experiment has ever considered it.
Every source behind this entry (4)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The published in vivo LEDGF requirement is for initiation: Psip1 was deleted from early embryonic HSC by Vav-iCre before MLL-ENL transduction and transplantation, and no arm deletes Psip1 in an animal that already has leukaemia.

    PMID 29084774DOI
    What this evidence cannot show

    Read in full through the PMC HTML route after fullTextXML returned HTTP 500 and efetch db=pmc returned no <body>. Murine, myeloid, MLL-ENL.

    Evidence that argues against this (1)
    • Dominant-negative LEDGF fragments did impair clonogenic growth of already-transformed human and mouse MLL-fusion cells, so an in-vitro maintenance effect exists and the proposed xenograft is not being run against a blank expectation. Méreau H et al. 2013
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    In already-transformed MLL-AF9 HSPCs, LEDGF knockdown decreased wild-type MLL occupancy but INCREASED MLL-fusion occupancy at target loci, while reducing chromatin association of AF4, CDK9, BRD4, P-TEFb and Ser2-phosphorylated Pol II - so the occupancy and output readouts of this experiment can dissociate.

    ASH1L Links Histone H3 Lysine 36 Dimethylation to MLL Leukemia. · Zhu L, Li Q, et al. · Cancer Discov 2016
    PMID 27154821DOI
    What this evidence cannot show

    Murine, in vitro, ChIP-qPCR at Hoxa9 and Meis1 rather than genome-wide. Read via efetch db=pmc; 36,284 characters with a body element.

    Evidence that argues against this (1)
    • A second group achieved only 30% Ledgf knockdown in a mouse ESC TetR reporter system and saw only a slight increase in MLL(N) ChIP, so the magnitude of the effect is not well established. Kerry J et al. 2017
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Somatic MEN1 mutations at the drug-menin interface cause clinical resistance to menin inhibition while leaving the menin-KMT2A interaction intact, which is the specific clinical setting in which a LEDGF-directed agent would be tested.

    MEN1 mutations mediate clinical resistance to menin inhibition. · Perner F, Stein EM, et al. · Nature 2023
    PMID 36922589DOI
    What this evidence cannot show

    The prediction that the LEDGF arm survives in these cells is an inference from the structural argument, not a measurement; nobody has profiled LEDGF occupancy in a MEN1-mutant resistant sample.

    Evidence that argues against this (1)
    • If the clamp is cooperative, an intact menin-KMT2A interaction in a drug-resistant cell keeps the LEDGF affinity at its tenfold-enhanced value, which makes the LEDGF grip harder to break in exactly the population where it would be used. Murai MJ et al. 2014
  4. E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    LEDGF/p75-depleted human haematopoietic cells engraft the NSG mouse, which establishes the feasibility of the human arm of this design.

    PMID 22334021DOI
    What this evidence cannot show

    An HIV gene-therapy study, not a leukaemia study; engraftment is a weaker endpoint than long-term competitive repopulation, and it says nothing about whether LEDGF loss is tolerated in a leukaemic graft.

    Evidence that argues against this (1)
    • Conditional Psip1 knockout mice show an approximately twofold reduction in peripheral white cells and impaired serial colony formation, so 'tolerated' has a measured cost. El Ashkar S et al. 2018
Connected to