Four hours of MS67 in MV4;11, two KMT2A antibodies: the cheapest test of whether WRAD reaches the fusion or only wild-type KMT2A
also called the MS67 attribution experiment · N-versus-C terminal KMT2A blot under WDR5 degradation
A commercially characterised compound strips one protein out of leukemia cells within two hours. Running it in the index leukemia line and then blotting with two antibodies - one that sees the fusion protein and one that cannot - would finally say whether that protein machine touches the cancer-causing fusion or only the normal version of the gene. It takes an afternoon.
The same thing, for a scientist
`absence-no-proximity-degron-or-crispr-study-of-wrad-in-kmt2a-fusion-leukaemia` survives a 2026 re-test, and the reason it survives is a readout problem, not a reagent problem: MS67 already delivers rapid, reversible, near-complete WDR5 depletion in MV4;11 with a chromatin-association panel, but its MLL immunoblot cannot distinguish the KMT2A::AFF1 fusion from wild-type KMT2A. **The fix is an antibody choice.** KMT2A is cleaved by taspase1 into an N-terminal ~320 kDa species and a C-terminal ~180 kDa species; the direct fusion retains the N-terminal portion and loses everything C-terminal of the breakpoint. **An N-terminal antibody therefore detects the fusion and wild-type N320; a C-terminal antibody detects wild-type C180 only.** Running both on the same chromatin fraction under MS67 separates the two hypotheses this project has been unable to distinguish since the WRAD adjudication beat. Cost: one drug, 4 hours, one fractionation, two blots, three pre-validated controls.
Design
Cells. MV4;11 (KMT2A::AFF1, the index fusion). Add RS4;11 as a second KMT2A::AFF1 line if budget allows - the two sit at opposite ends of the MS67 sensitivity range for reasons the source paper could not explain, and that difference is itself informative.
Arms, all four validated in this line by the source paper so none needs new optimisation:
| arm | compound | role |
|---|---|---|
| 1 | MS67, 100 nM, 4 h | >25x DC50; 4 h is the paper's measured time to maximal degradation |
| 2 | MS67N, 100 nM, 4 h | VHL-non-binding diastereoisomer - binds WDR5, does not degrade. The control that matters |
| 3 | OICR-9429, 100 nM, 4 h | WIN-site occupancy without degradation |
| 4 | DMSO | vehicle |
| 5 | MS67, 100 nM, 4 h then 48 h washout | reversibility; the paper measures substantial rebound at 48 h |
Readout. Chromatin fractionation, then immunoblot the same fraction with (a) an N-terminal KMT2A antibody - detects the fusion and wild-type N320; (b) a C-terminal KMT2A antibody - detects wild-type C180 only; (c) WDR5, to confirm degradation; (d) RBBP5 and menin, to reproduce the source paper's panel; (e) histone H3 and a soluble marker, as fractionation controls. Quantify each band against the DMSO arm.
Antibody caveat that must be settled first: this project has already recorded that the immunogen range of one widely used KMT2A antibody (A300-086A) is not retrievable, and that every masking claim in one analysis rests on it. Whichever antibodies are chosen, their immunogen coordinates must be obtained from the vendor and recorded before the experiment, not after.
The scoreable predictions
Effect sizes are stated against the DMSO arm, measured on the chromatin fraction at 4 h, with MS67N and OICR-9429 required to show < 1.3-fold change on every band (if they do not, the experiment is uninterpretable and nothing below applies).
| model | fusion band (N-terminal minus C-terminal) | wild-type C180 band | verdict |
|---|---|---|---|
| WRAD reaches the fusion | falls >= 2.0-fold | falls >= 2.0-fold, within 2-fold of the fusion's magnitude | the WRAD-fusion contact is real and load-bearing for chromatin residence |
| WRAD is a wild-type-only passenger | changes < 1.5-fold | falls >= 2.0-fold | the second-shell model; WRAD is not a fusion target |
| second shell with partial transfer | falls 1.5-2.0-fold, and less than C180 | falls >= 2.0-fold | the outcome the current evidence most predicts |
| neither | < 1.5-fold | < 1.5-fold | WDR5 is not required for KMT2A-complex chromatin residence in this line at all, which would contradict the source paper's own MLL panel and should be read as a fractionation failure first |
Falsifier for the whole design: if the N-terminal and C-terminal antibodies do not resolve distinct bands of the expected sizes in the DMSO arm, the experiment cannot be scored and no result from it should be reported. Check this on a whole-cell lysate before committing the fractionation.
Why this and not the experiment the absence entry proposes
absence-no-proximity-degron-or-crispr-study-of-wrad-in-kmt2a-fusion-leukaemia lists as feasible 'TurboID from a tagged KMT2A::AFF1 with a wild-type-allele degron' and 'tiling CRISPR of ASH2L and DPY30 in MV4;11 or SEM'. Both are better experiments and both are months of work: a knock-in tag on the fusion allele, a degron on the wild-type allele, streptavidin proteomics.
This design answers a strictly narrower question - does removing WDR5 remove the fusion from chromatin - in an afternoon, using a compound whose DC50, kinetics, washout profile, selectivity (4000+ protein proteomics, 22 methyltransferases, 45 kinases, 44 GPCRs/channels/transporters) and negative controls are already published for this exact cell line. It should be run first, because a negative result would tell the lab not to build the knock-in, and a positive result would be the first direct evidence of a WRAD-fusion relationship in any system.
The one thing it cannot do is direction: chromatin co-loss is not contact. A TurboID or a stringent co-IP is still required to say WRAD binds the fusion. This design says whether the fusion's chromatin residence depends on WDR5, which is the question the therapeutic argument actually turns on.
| design | MV4;11 (KMT2A::AFF1) treated 4 h with MS67 100 nM against MS67N, OICR-9429 and DMSO, plus a 48 h washout arm; chromatin fractionation; parallel immunoblot with an N-terminal KMT2A antibody (detects fusion + wild-type N320) and a C-terminal KMT2A antibody (wild-type C180 only), with WDR5, RBBP5, menin, histone H3 and a soluble marker on the same fraction. |
|---|---|
| readout | fold-change in chromatin-bound fusion band and wild-type C180 band versus DMSO at 4 h |
| controls | MS67N (binds WDR5, cannot recruit VHL) and OICR-9429 (occupancy without degradation) must both show <1.3-fold change on every band; N- and C-terminal antibodies must resolve distinct bands of expected size in a whole-cell lysate before the fractionation is attempted |
| cost tier | low |
| timeline | one week including antibody validation; the treatment itself is four hours |
| prerequisite | obtain and record the immunogen coordinates of both KMT2A antibodies from the vendor BEFORE the experiment, and confirm the N-terminal epitope lies 5' of the MV4;11 breakpoint |
| answers absence | absence-no-proximity-degron-or-crispr-study-of-wrad-in-kmt2a-fusion-leukaemia |
falsifiers
[ "The N-terminal and C-terminal antibodies do not resolve distinct bands of expected size in the DMSO whole-cell lysate: the experiment is unscoreable and must not be reported.", "MS67N or OICR-9429 changes any band by more than 1.3-fold: the effect is not degradation-dependent and the design is uninterpretable." ]
measurements
[
{
"quantity": "effect_size_threshold",
"subject": "chromatin-bound fusion band fold-change | MS67 4 h vs DMSO, MV4;11",
"value": 2,
"units": "fold decrease",
"n": null,
"universe": "pre-registered threshold for 'WRAD reaches the fusion'; <1.5-fold scores as 'wild-type-only passenger'",
"null": 1,
"note": "pre-registered before the experiment, per rule 96 half one: interval, cell type and effect size all named"
}
]- A positive result would be read as 'WDR5 is a target in KMT2A::AFF1', and this corpus's DepMap reading says the opposite: WDR5 is depleted LESS in KMT2A-rearranged lines than in leukaemia lines generally, a lineage-controlled selectivity of +0.191, the worst of 41 candidates scored here. Chromatin dependence and therapeutic selectivity are different claims and this experiment measures only the first.
- Any WDR5-directed intervention removes WDR5 from NSL, KMT2B, SETD1A and SETD1B as well, and WDR5 is DepMap common-essential. Essential-function toxicity is the dominant risk for this whole target class regardless of how this experiment comes out.
- Are MS67 and MS67N obtainable? The paper names no supplier and the design's entire cost advantage rests on not having to synthesise them.
- Would the same design in RS4;11 - the least MS67-sensitive KMT2A::AFF1 line - give the opposite answer, and if so is the difference in WRAD dependence or in drug uptake?
- Chromatin co-loss is not contact. What is the cheapest follow-up that establishes direction - a stringent co-IP of the fusion under MS67N, or a TurboID?
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
MS67 achieves DC50 3.7 nM and Dmax 94% against WDR5 in MV4;11 with apparent degradation at 2 h and maximal at ~4 h, has MS67N and OICR-9429 as validated non-degrading controls in that line, and rebounds substantially by 48 h after washout - so every parameter this design needs is already published for this exact system.
A selective WDR5 degrader inhibits acute myeloid leukemia in patient-derived mouse models. · Yu X, Li D, Kottur J, Shen Y, Kim HS, Park KS, Tsai YH, Gong W, Wang J, Suzuki K, Parker J, Herring L, Kaniskan HU, Cai L, Jain R, Liu J, Aggarwal AK, Wang GG, Jin J · Sci Transl Med 2021What this evidence cannot show
Full text read via PMC8500670 on 2026-09-24. MS67 and MS67N are academic compounds from the Jin laboratory; the paper does not state a commercial supplier and availability is an assumption of this design, not a fact established by the citation. The 4 h timepoint is the paper's time to maximal degradation in MV4;11 specifically and is longer in other lines.
Evidence that argues against this (1)
- Degradation removes WDR5 from NSL and from KMT2B, SETD1A and SETD1B simultaneously, so a fall in chromatin-bound fusion under MS67 shows dependence on WDR5, not dependence on a WDR5-KMT2A-fusion contact. The design's positive result is therefore weaker than 'WRAD reaches the fusion' and the table above should be read with that discount. Oncofusion Knowledge Graph 2026
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
UniProt models KMT2A as three separately annotated species - full length, the MLL cleavage product N320 and the MLL cleavage product C180 - so an N-terminal and a C-terminal antibody address different molecules, which is the basis of this design's discrimination.
UniProt Q03164 (KMT2A_HUMAN), subcellular location comments for the full-length protein and for the MLL cleavage products N320 and C180 · UniProt 2026What this evidence cannot show
Quoted from this corpus's rule-91 record of the Q03164 location comments rather than from a fresh UniProt fetch this cycle. The N320/C180 split is taspase1 cleavage of the WILD-TYPE protein; the fusion is a separate species that shares the N-terminal region, and no antibody epitope map for the specific reagents a lab would buy is supplied here.
Evidence that argues against this (1)
- If the chosen N-terminal antibody's epitope lies 3' of the KMT2A breakpoint used in MV4;11, it will not detect the fusion at all and the experiment silently measures wild-type twice. The breakpoint is intron-dependent and this is why the immunogen coordinates must be obtained before the experiment. Oncofusion Knowledge Graph 2026
- tests →No proximity, degron or tiling-CRISPR study has ever been run on WRAD in KMT2A-fusion leukaemia
four hours of MS67 and two KMT2A antibodies would close this absence's degron arm
every parameter of this design comes from the MS67 characterisation recorded there
asks whether the index fusion's chromatin residence depends on WDR5