Resolve acetyl-K1130/K1133 occupancy on the fusion versus wild-type KMT2A in the same cell - the one experiment that makes a shared-sequence site interpretable
also called fusion-resolved acetyl PRM
The chemical tag site sits on a stretch of protein that both the normal MLL1 and the cancer fusion carry, so an ordinary measurement cannot tell which protein was tagged. This experiment separates the two proteins first, then measures the tag on each, which is the only way the question means anything.
The same thing, for a scientist
The tryptic peptide APPIKPIKPVTR (Q03164 1126-1137) is identical in wild-type KMT2A and in every KMT2A fusion, so no bottom-up proteomics on a fusion-bearing cell can attribute acetylation to one species. This experiment separates the species before measuring: epitope-tagged fusion knock-in (or partner-directed IP with the stated controls), followed by targeted parallel-reaction-monitoring quantitation of the mono- and di-acetylated forms of APPIKPIKPVTR against synthetic heavy standards. It answers the beat's central unanswerable question and also supplies, for the first time, an absolute stoichiometry for a site that has only ever been measured as a western-blot ratio.
Why the obvious version does not work
Do not run an acetylome on MV4-11 or SEM and report acetyl-K1130. This has effectively already happened - the site's founding detection was in a panel including MV4-11 - and it cannot answer the question, because the peptide is shared. Two true facts, the deposit used a KMT2A::AFF1 line and the deposit found acetyl-K1130, do not join into the fusion is acetylated at K1130. The species must be separated first.
Also do not rely on the 2015 paper's site-specific antibody without re-validating it. It is a custom polyclonal raised at Millipore in 2015 against NH2-APPIK(ac)PIK(ac)PVTR; whether it still exists as a reagent is unknown, and it was raised against the doubly acetylated peptide so it may not detect mono-acetylated species at all.
Design
Separation (choose one, the first is strongly preferred):
- Endogenous epitope tag on the fusion allele. CRISPR knock-in of a 3xFLAG or HiBiT tag at the 3' end of the fusion ORF in SEM or RS4;11 (lymphoid, the index disease) and MV4-11 (myeloid, for the lineage comparison). Anti-FLAG IP recovers fusion only; anti-KMT2A-N IP on the same lysate recovers both, and fusion-depleted flow-through approximates wild-type.
- Partner-directed IP. Anti-AFF1 C-terminal IP. ⚠️ This also recovers wild-type AFF1 and the endogenous SEC, so it needs a KMT2A-negative control line and cannot give clean wild-type comparator material. Use only if the knock-in fails.
Measurement: targeted PRM on the tryptic peptide APPIKPIKPVTR, quantifying four species - unmodified, K1130-ac only, K1133-ac only, and doubly acetylated - against synthetic heavy-labelled standards of each. The standards are essential and are also the deliverable that settles a separate open question: whether K1130 and K1133 can be distinguished by MS at all, given they share one peptide.
Arms: fusion-IP and wild-type-enriched material, 3 biological replicates, ± the SIRT1 inhibitor EX527 as a positive control that the mark is measurable and movable in these cells (EX527 raised it markedly in MEFs in the source paper).
Controls that the design lives or dies by:
- a synthetic heavy standard for each acetylated form - without it there is no stoichiometry, only a ratio;
- EX527 arm - if the mark does not move under SIRT1 inhibition in a leukaemia cell, the whole axis may simply not operate there, which is itself a publishable answer;
- a KMT2A-rearrangement-negative leukaemia line (e.g. REH, NALM-6) for the wild-type-only baseline;
- time of collection recorded, because the mark is circadian in every system where it has been measured and an unrecorded harvest time makes a negative uninterpretable.
What each outcome means
| result | interpretation |
|---|---|
| fusion:wild-type occupancy ratio > 1.5 | the fusion is preferentially acetylated at a shared site - the hypothesis survives and a fusion-selective property of shared sequence exists |
| ratio < 1/1.5 | the fusion is preferentially deacetylated - equally interesting, and points at SIRT1 recruitment by the partner |
| ratio within 1.5-fold | the hypothesis is dead and the retained acetylation site should be closed as a subject |
| the mark is undetectable in these cells | the axis does not operate in leukaemia at all - the most likely single outcome, and it closes the question cleanly |
🟠 The last row is the honest modal expectation. Every measurement of this mark has been in HEK293, MEFs or mouse liver; no haematopoietic cell has ever been tested. A well-controlled undetectable is a good result and the experiment should be sized and written so that it counts as one.
| design | CRISPR epitope-tag knock-in on the fusion allele in SEM or RS4;11 (lymphoid) and MV4-11 (myeloid); anti-tag IP to isolate fusion, anti-KMT2A-N IP and tag-depleted flow-through for wild-type; targeted PRM quantitation of unmodified, K1130-ac, K1133-ac and doubly acetylated APPIKPIKPVTR against synthetic heavy standards; 3 biological replicates; +/- EX527; harvest time recorded. |
|---|---|
| readout | absolute acetylation stoichiometry at K1130 and K1133 on fusion versus wild-type KMT2A from the same cell |
| controls | synthetic heavy-labelled standard for each acetyl form; EX527 arm as a movability control; KMT2A-rearrangement-negative line (REH or NALM-6) as a wild-type-only baseline; recorded circadian harvest time; a non-tryptic protease pilot because acetylation blocks tryptic cleavage at the very residues being measured |
| cost tier | high |
| timeline | 9-15 months (knock-in generation dominates; the PRM assay is 6-8 weeks once the standards exist) |
| decision rule | fusion:wild-type occupancy ratio outside 1.5-fold in either direction supports hyp-the-partner-not-the-site-...; inside 1.5-fold kills it; undetectable in leukaemia cells closes the subject |
| cheaper prerequisite | Before any of this: look up Q03164 K1133 in the Weinert 2018 CBP/p300 acetylome supplementary tables (PXD005252) to see whether the site is CBP/p300-dependent, and look up Q03164 in the Choudhary 2009 Science supplementary tables to see which cell line contributed the KMT2A peptides. Two paywalled lookups that could reshape or cancel this experiment. |
tests
[ "hyp-the-partner-not-the-site-is-what-would-make-a-retained-kmt2a-modification-fusion-specific" ]
measurements
[
{
"quantity": "decision threshold on the fusion:wild-type occupancy ratio",
"subject": "acetyl-K1130/K1133 PRM",
"value": 1.5,
"units": "fold",
"n": 3,
"universe": "3 biological replicates per arm, one cell line at a time",
"null": 1,
"note": "a ratio of 1.0 is the null - the same sequence in the same cell with the same enzyme environment"
}
]- This is a 9-15 month, high-cost experiment whose most likely single outcome is 'undetectable in leukaemia cells'. That outcome is worth having and it is worth saying out loud at the point of proposing, not after.
- A fusion-specific difference in occupancy at a shared site is a biomarker, not a target. Even a clean positive leaves the intervention problem exactly where it was: the only handles are CBP/p300 and SIRT1, both global.
- Two paywalled supplementary-table lookups could make this experiment unnecessary or reshape it entirely. Running the expensive thing before the cheap thing is the failure mode to avoid here.
- Is APPIKPIKPVTR a tractable PRM peptide? Acetylation blocks tryptic cleavage at K1130 and K1133, so the modified and unmodified forms are not the same peptide and a stoichiometry measurement needs either a different protease or chemical acetylation of the unmodified pool. This must be piloted before the knock-in is started.
- Does the site-specific antibody from the 2015 paper still exist? If it does, a western-blot version of this experiment on fusion-IP material is a fraction of the cost and could be run as a go/no-go before committing to the PRM route.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
The tryptic peptide APPIKPIKPVTR at Q03164 1126-1137 is shared between wild-type KMT2A and KMT2A fusions, so bottom-up proteomics cannot attribute acetylation at K1130/K1133 to either species without prior separation.
UniProtKB Q03164 canonical sequence and ana-kmt2a-exon-to-domain-map breakpoint coordinates · UniProt; this project 2026What this evidence cannot show
The peptide identity was verified in this beat. Whether it is a good PRM target has NOT been assessed - it is short (12 residues), proline-rich and doubly missed-cleavage-prone at the acetylated lysines, all of which make it a difficult peptide. An alternative protease (Glu-C, chymotrypsin) may be needed and that should be piloted before committing.
Evidence that argues against this (1)
- Acetylation blocks tryptic cleavage at the modified lysine, so the acetylated and unacetylated forms of this region are NOT the same peptide - the unmodified form will be cleaved at K1130 and K1133 into fragments too short to detect. Measuring true stoichiometry therefore requires a non-tryptic protease or a chemical-acetylation reference, and this is a real design problem, not a detail.
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Acetyl-K1130/K1133 rises markedly under the SIRT1 inhibitor EX527 and under the NAMPT inhibitor FK866, providing a validated positive control for the measurement.
NAD(+)-SIRT1 control of H3K4 trimethylation through circadian deacetylation of MLL1 · Aguilar-Arnal L, Katada S, Orozco-Solis R, Sassone-Corsi P · Nature Structural & Molecular Biology 22(4):312-8 2015What this evidence cannot show
Demonstrated in MEFs and HEK293 only. Whether EX527 moves this mark in a leukaemia cell is exactly what is unknown, so the 'positive control' is itself a hypothesis until it fires once.
Evidence that argues against this (1)
- EX527 at the 50 microM used in the source paper is well above its SIRT1 IC50 and will hit other targets. A concentration-response and a genetic SIRT1 arm would be needed before attributing any change to SIRT1.
fusion:wild-type occupancy ratio outside 1.5-fold supports it; inside kills it
the shared-peptide problem this experiment exists to solve
EX527 arm as the movability control