Does the fusion's programme fall apart in the same order in KMT2A::AFF1 as in MLL::AF9? The chain's only ordering measurement, repeated on the index fusion
also called index-fusion degron ordering · AFF1 degron timecourse
There is exactly one experiment that shows which step of this leukaemia's machinery fails first when the cancer protein is removed, and it was done on a different fusion protein in a different kind of leukaemia. This is that experiment, done on the one that causes infant leukaemia.
The same thing, for a scientist
`mechanism-kmt2a-fusion-causal-chain` contains one measured arrow out of five: degron-tagged MLL::AF9 removal changes elongation within 15 minutes, with loss of the active chromatin landscape following [E1]. **No degron model of KMT2A::AFF1 exists**, so the ordering has never been tested on the index fusion - and there is a specific reason to expect it may differ. MLL::AF9 recruits DOT1L directly through the retained AHD, where DOT1L, AF4, BCOR and CBX8 compete for one mutually exclusive site [E4]; KMT2A::AFF1 retains an SEC component and recruits DOT1L indirectly. **If the elongation and methylation arms are wired differently, their failure order should differ** - and if it does, the chain's internal direction is partner-dependent, which is a different object from the single figure the field draws.
Design
Line. SEM and RS4;11 (human KMT2A::AFF1, B-lineage), with MOLM-13 or a published MLL::AF9 degron line as the reference arm so the comparison is run in one hand rather than against a literature value. An infant-derived KMT2A::AFF1 PDX if one can be tagged.
Construct. Endogenous knock-in of FKBP12(F36V) at the KMT2A N-terminus so the tag is carried by both the fusion and, unavoidably, wild-type KMT2A from that allele - which must be stated, because the wild-type protein is itself part of what a menin inhibitor perturbs 5. A VHL-recruiting degrader arm is included because CRBN-recruiting dTAG molecules fail on some recalcitrant oncoproteins 2.
Timepoints. 0, 5, 10, 15, 30, 60, 120, 240 min and 8, 24, 48 h. The early grid is set by the 15-minute result being replicated; the late grid by H3K79me2 being removed only by dilution.
Readouts, all spike-in calibrated. TT-seq or PRO-seq for nascent transcription and pausing index; CUT&RUN for the fusion, AFF1/SEC, DOT1L, H3K79me2 and H3K4me3; western for fusion protein to confirm degradation kinetics. Scoring on nascent transcription at fusion targets, not steady-state mRNA.
Controls. Untagged parental plus degrader; tagged line plus vehicle; a non-target locus panel; and a division tracker, because at the late timepoints mark loss and cell division are confounded.
What each outcome decides
| result | reading |
|---|---|
| Elongation change precedes chromatin loss, same order, similar timing | the chain's one measured arrow generalises across partners; the MLL::AF9 import is safe for this claim |
| Elongation change precedes chromatin loss but much more slowly | ordering is conserved, kinetics are not - and every schedule argument built on the MLL::AF9 number is wrong |
| H3K79me2 falls first or simultaneously | the arrow is partner-dependent; the chain must be drawn per fusion and the AFF1 disease has no measured ordering of its own |
| Neither moves within 4 h despite complete degradation | the programme is maintained by something the degron does not remove, and the whole chain needs a step it does not have |
The comparison is the experiment. A KMT2A::AFF1 timecourse without a same-hand MLL::AF9 arm would be uninterpretable against a published number generated on different instruments.
Why this has not been done
The graph already carries exp-kmt2a-aff1-degron-occupancy-titration and exp-endogenous-tag-and-degron-knock-in-kmt2a-aff1, which propose the reagent. This entry is the ordering question specifically, and it is the one that repairs a named defect in the spine rather than adding a capability. The obstacle is the reagent, not the assay: dTAG chemistry and spike-in CUT&RUN are both standard 23, and the missing piece is an endogenously tagged index-fusion line that nobody has published.
| design | Endogenous FKBP12(F36V) knock-in at the KMT2A N-terminus in SEM and RS4;11 (KMT2A::AFF1), with a same-hand MLL::AF9 degron reference arm; CRBN- and VHL-recruiting degrader arms; timecourse 0, 5, 10, 15, 30, 60, 120, 240 min and 8, 24, 48 h. |
|---|---|
| readout | Spike-in-calibrated TT-seq or PRO-seq (nascent transcription, pausing index) and CUT&RUN for fusion, AFF1/SEC, DOT1L, H3K79me2 and H3K4me3; western for degradation kinetics; division tracker at late timepoints. |
| controls | untagged parental plus degrader; tagged line plus vehicle; non-target locus panel; same-hand MLL::AF9 arm; division tracking to separate mark dilution from active removal |
| cost tier | high |
| timeline | 9-12 months including line generation |
| why not done | no endogenously degron-tagged KMT2A::AFF1 line has been published; the assay components are standard and the reagent is not |
| chain | mechanism-kmt2a-fusion-causal-chain |
tests
[ "mechanism-kmt2a-fusion-causal-chain" ]
decision rule
{
"ordering_conserved": "the chain's one measured arrow generalises across partners",
"ordering_conserved_kinetics_differ": "schedule arguments built on the 15-minute number do not transfer",
"ordering_reversed_or_simultaneous": "the arrow is partner-dependent and the chain must be drawn per fusion",
"no_change_despite_degradation": "the programme is maintained by a step the chain does not contain"
}- A degron result licenses a schedule argument that a drug cannot deliver: complete removal in minutes has no pharmacological counterpart, and the most likely misuse of this experiment is to quote its kinetics in a dosing rationale.
- Tagging the endogenous locus perturbs wild-type KMT2A as well as the fusion, so an apparently fusion-specific ordering may partly report the loss of the normal protein.
- Can the KMT2A N-terminus be tagged without disturbing the menin/LEDGF contacts that sit in the first 160 residues? If not, the tag must go at the partner C-terminus and the design loses its wild-type-KMT2A control value.
- Is a PDX-derived infant KMT2A::AFF1 line taggable at all, or does the experiment have to be run in adult-derived lines that carry a different cell of origin?
- Would a reversible chemical approach - washing out a menin inhibitor rather than degrading the fusion - answer the ordering question more cheaply, at the cost of never removing the fusion itself?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
FKBP12(F36V)-degron-tagged MLL::AF9 degradation changes transcriptional elongation at a core subset of target genes within 15 minutes, with loss of the active chromatin landscape occurring subsequently.
MLL::AF9 degradation induces rapid changes in transcriptional elongation and subsequent loss of an active chromatin landscape. · Olsen SN, Godfrey L, Healy JP, Choi YA, et al. · Mol Cell 2022What this evidence cannot show
Abstract read 2026-09-20. One fusion, one cell system. This is the result being replicated, and the entire justification for the experiment is that it exists exactly once.
Evidence that argues against this (1)
- Complete degradation is not partial pharmacological occupancy, and clinical KMT2A::AFF1 material shows resistance downstream of sustained target engagement - so an ordering established by degron does not by itself predict what a drug does in a patient. Mahdavi L et al. 2026
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
VHL-recruiting dTAG molecules give rapid, direct control of tagged protein levels and overcome a limitation of CRBN-recruiting dTAG molecules, which fail to degrade certain recalcitrant oncoproteins.
Rapid and direct control of target protein levels with VHL-recruiting dTAG molecules. · Nabet B, Ferguson FM, Seong BKA, Kuljanin M, et al. · Nat Commun 2020What this evidence cannot show
Abstract read 2026-09-20. Methodological citation; the failure mode it describes is why this design carries two degrader chemistries rather than one.
Evidence that argues against this (1)
- Tagging the endogenous KMT2A N-terminus degrades wild-type KMT2A from the same allele as well as the fusion, so a degron result in this design is not purely a fusion result - which is the same confound that makes menin-inhibitor experiments hard to interpret. Sparbier CE et al. 2023
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Acute ENL degradation in an MLL-AF4-positive line suppresses both initiation and elongation of RNA polymerase II genome-wide, with pronounced effects at genes carrying a disproportionate share of the fusion's programme.
Transcription control by the ENL YEATS domain in acute leukaemia. · Erb MA, Scott TG, Li BE, Xie H, et al. · Nature 2017What this evidence cannot show
Abstract read 2026-09-20. Establishes that chemical-genetic degradation plus nascent-transcription readout is tractable in an AFF1-context human line - i.e. the assay half of this design is already demonstrated in the right disease; only the fusion tag is missing.
Evidence that argues against this (1)
- ENL degradation affects transcription genome-wide, not only at fusion targets, so a degron-plus-nascent-transcription design must be scored on fusion-target selectivity rather than on magnitude, or it will report a global effect as a fusion effect. Brzezinka K et al. 2019
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The retained AF9 AHD binds AF4, DOT1L, BCOR and CBX8 at one mutually exclusive site, with the four complexes superposing at 1.4-1.9 A over the structured AF9 residues.
BCOR Binding to MLL-AF9 Is Essential for Leukemia via Altered EYA1, SIX, and MYC Activity. · Schmidt CR, Achille NJ, Kuntimaddi A, Boulton AM, et al. · Blood Cancer Discov 2020What this evidence cannot show
Abstract read 2026-09-20. This is the structural reason to expect the ordering to be partner-dependent: in MLL::AF9 the elongation and methylation arms compete for one surface, and in KMT2A::AFF1 they do not sit on the same surface at all.
Evidence that argues against this (1)
- AFF1 and AFF4 also form distinct SECs with non-overlapping target specificity, so 'the AFF1 wiring' is itself more than one thing, and a single AFF1 degron line may not represent the fusion class either. Lu H et al. 2015
- E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Menin loss or pharmacological menin inhibition phenocopies polycomb disruption through the Menin-KMT2A/B axis and paradoxically activates bivalent genes.
Targeting Menin disrupts the KMT2A/B and polycomb balance to paradoxically activate bivalent genes. · Sparbier CE, Gillespie A, Gomez J, Kumari N, et al. · Nat Cell Biol 2023What this evidence cannot show
Abstract read 2026-09-20. Cited here for the design constraint only: any perturbation that also touches wild-type KMT2A/KMT2B carries a non-fusion component, and this design must state that it does rather than assume it does not.
Evidence that argues against this (1)
- Men1 deletion produces consequences distinct from Mll1;Mll2 co-deletion, so the wild-type component is real but bounded, and a degron design should not be abandoned because of it. Chen Y et al. 2019
The cause-versus-consequence question this amendment leaves open at the HOXA9 locus is exactly what this proposed experiment would order.
repeats the chain's only ordering measurement on the index fusion
the index fusion, for which no degron model exists
- tests →Reversibility and kinetics — does the program collapse when the fusion is removed, and how fast?
fills the degron_models_missing entry for KMT2A::AFF1
the mutually exclusive AHD is the structural reason to expect partner-dependent ordering
the decisive experiment for this hypothesis