start / graph / mechanism-q120-dot1l-eviction-and-h3k79me2-persistence-are-separable-at-three-days

MechanismmoderateQ120-h3k79me2-second-line-and-the-mll1-epitope

In MOLM13 a menin inhibitor removes three quarters of DOT1L from the fusion's own sites and leaves H3K79me2 there completely intact for three days - the writer and its mark are separable on this timescale, and in RS4;11 they are not

also called the enzyme leaves and the mark stays · H3K79me2 persistence

The drug strips away the enzyme that puts a chemical tag on histones, but in one of the two cell lines the tag itself is still fully there three days later. Removing the machine does not immediately remove what the machine made.

The same thing, for a scientist

At the 467 MOLM13 bins where an MLL1 N-terminus and menin co-occupy, three days of 330 nM VTP-50469 leaves [E1]: | | DOT1L (the writer) | H3K79me2 (its mark) | |---|---:|---:| | MOLM13 share ratio | **0.2219** | **0.8667** | | floor-normalised | 0.243 | 0.876 | | at the MLL1-N held set | 0.2773 | **0.9980** (1.0082 over floor) | | fraction of held bins losing rank | 0.5652 | **0.2498** | | per-bin median share ratio | 0.3045 | **1.0845** | | the same in RS4;11 | 0.2455 | 0.7008 | **DOT1L occupancy falls to a quarter and its mark does not move at all.** At the MLL1-N held set three quarters of the bins actually *gain* H3K79me2 rank under drug and the per-bin median ratio is above 1. **Three readings, and this measurement does not choose between them.** 1. **Kinetic.** H3K79 methylation has no known demethylase; the mark is thought to be lost by replicative dilution. Three days of a drug that does not arrest MOLM13 immediately is perhaps two to three divisions, and a mark diluted 1/2 per division should be visibly down. It is not - so either the dilution model is wrong here or the residual DOT1L is enough. 2. **Residual-enzyme sufficiency.** 22% of DOT1L occupancy may be ample to maintain a steady-state mark, in which case occupancy is a poor proxy for catalysis and the ChIP is measuring the wrong variable. 3. **Line difference.** RS4;11 loses a third of the mark at the same construction, so whatever maintains it in MOLM13 does not operate, or does not suffice, in the index fusion. MOLM13's partner AF9 binds DOT1L directly and AFF1 does not, which predicts the opposite of what is seen. **What is not in doubt is the dissociation itself**, because it is measured with two antibodies in the same nuclei against the same bins with a matched input, and the input share ratio at those bins is 0.9272.

Go deeper

Why this bears on a number this graph has recorded as never measured

absence-h3k79me2-decay-half-life-never-measured records, with a dated and controlled query, that nobody has published a decay constant for H3K79me2 after its writer is blocked, and that in this literature the word half-life almost always means the drug's.

This measurement does not supply the half-life and it does bound it. In MOLM13, three days after DOT1L occupancy at the fusion's own sites falls to 0.22, the mark there is unchanged within 1%. If the mark were decaying with a half-life comparable to the observation window it would be visibly down. So either the half-life at these sites exceeds three days in this line, or the mark is being maintained by the residual enzyme.

A menin inhibitor is not the right instrument for that number - it removes the writer indirectly and incompletely. A DOT1L inhibitor removes the catalysis directly, and GSE260742 contains an SGC0946 arm for H3K79me2 in murine MLLAF9 that this beat did not touch. That is the cheaper route and it is one download away.

Facts and figures

script	analyses/q120-h3k79me2-second-line-and-the-mll1-epitope/arm1-molm13-ordering.py
status	run
mechanism summary	Writer eviction and mark persistence are separable on a three-day timescale in MOLM13 and only partly separable in RS4;11.
cell line	MOLM13 and RS4;11
drug	VTP-50469 330 nM, 3 days
epigenetic readout	H3K79me2 share ratio 0.9980 at the MLL1-N held set in MOLM13 while DOT1L occupancy at the same bins is 0.2773
mark written	H3K79me2
writer of that mark	gene-dot1l
eraser of that mark	none known - no H3K79 demethylase has been identified, which is why replicative dilution is the default model for mark loss and why a three-day persistence is informative
shown for fusion	KMT2A::MLLT3 in MOLM13 (mark fully persists) and KMT2A::AFF1 in RS4;11 (mark falls to 0.70); not shown for KMT2A::AFDN, which has no H3K79me2 sample in this series
absence class	measured-not-analysed
absence class note	The decay-rate question itself remains never-measured, but the data that bounds it is deposited: GSE260742 contains an SGC0946 (DOT1L inhibitor) H3K79me2 arm in murine MLLAF9 that would remove catalysis directly rather than indirectly, and it has not been analysed. That is compute, not wet work.
absence class assigned	2026-09-23 (Q120)

How this could bite us back

If H3K79me2 persists for three days after a menin inhibitor in AML, then combining with a DOT1L inhibitor has a real substrate to act on rather than a redundancy - the opposite of what the DOT1L occupancy number alone implies - and a combination trial designed off occupancy data would be designed off the wrong variable.
It also means H3K79me2 cannot serve as an early pharmacodynamic marker of menin inhibition in MLL-AF9 disease. A three-day on-treatment biopsy would read fully methylated chromatin in a patient whose menin occupancy had fallen to 0.16.

Still open

Does 22% residual DOT1L occupancy support a steady-state H3K79me2 level? A dose-response of a DOT1L inhibitor with paired DOT1L and H3K79me2 ChIP would give the occupancy-to-mark transfer function, and nobody has published one.
How many divisions did MOLM13 complete in the three days of this experiment? Without a growth curve for the treated culture the dilution model cannot be tested, and GSE127507 deposits none.
Is the persistence specific to the fusion's own sites, or genome-wide? The H3K79me2 genome-wide mid-occupancy floor in MOLM13 is 0.9899, so the mark is flat everywhere - which makes the RS4;11 result, where the floor is 0.9851 but the fusion's sites fall to 0.70, the site-specific one.

Every source behind this entry (1)

E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
In MOLM13 under 330 nM VTP-50469 for 3 days, at the 467 co-occupied 2 kb bins DOT1L has a within-sample share ratio of 0.2219 and H3K79me2 of 0.8667; at the 1,389-bin MLL1-N held set DOT1L is 0.2773 and H3K79me2 is 0.9980 with only 24.98% of bins losing H3K79me2 rank and a per-bin median ratio of 1.0845; the matched input share ratio at the co-occupied bins is 0.9272. In RS4;11 the same constructions give DOT1L 0.2455 and H3K79me2 0.7008.
Q120 re-analysis of GSE127507, MOLM13 arm; analyses/q120-h3k79me2-second-line-and-the-mll1-epitope/arm1-molm13-ordering.py and results-arm1.json, run 2026-09-23 · Iqbal Lab oncofusion knowledge graph 2026
What this evidence cannot show
CST #5427 anti-H3K79me2 reads a histone MARK, i.e. a consequence. It cannot distinguish halted deposition from active demethylation from replicative dilution, and it cannot attribute any of them to the fusion. CST #77087S clone 1W4Z anti-DOT1L cannot distinguish DOT1L held by the fusion's partner moiety from DOT1L in wild-type complexes - and the two lines compared here differ in exactly that moiety, MLLT3/AF9 (direct DOT1L contact) versus AFF1 (indirect, through the AF4-family/SEC axis). A ChIP measures occupancy, not catalysis: 22% residual DOT1L occupancy is not 22% residual methyltransferase activity, and this series has one timepoint so no rate can be fitted. NO SPIKE-IN NORMALISATION. All 26 GSE127507 records state a Drosophila spike-in was run, measured, and deliberately not used, and no scale factor is deposited; no GSE260742 record contains spike-in vocabulary at all. Every cross-condition number here is therefore a WITHIN-SAMPLE share or rank, which cancels the unknown per-sample scale factor exactly. No raw fold change is reported as a biological quantity anywhere in this beat. n = 1 per antibody per condition per line in both series. No number in this beat carries a spread and none is precise.
Evidence that argues against this (1)
- Nature Structural & Molecular Biology 30:1592-1606 2023

Connected to

related to →Bounding an absence the graph already asserts: the H3K79me2 decay half-life, counted, dated and controlled - and why three of the six hits are about the drug's half-life
bounds the unmeasured number from one side: three days after DOT1L occupancy falls to 0.22 at the fusion's own MOLM13 sites, H3K79me2 there is unchanged within 1%
related to →H3K79me2 - the di-methyl state the leukaemia actually runs on, and the one menin turns out to govern only half of
the mark persists intact for three days after its writer is largely evicted, in MOLM13
depends on →DOT1L — the target where excellent chemistry met an indirect interaction
the dissociation is defined by DOT1L occupancy at the same bins falling to 0.2219 while the mark it writes does not move
supports →The four-way ordering given a second cell line: menin, DOT1L and the MLL1-N channel reproduce between KMT2A::AFF1 and KMT2A::MLLT3 to within 0.04 of a share ratio, and H3K79me2 - the one channel the ordering was built around - diverges by 0.21 to 0.30 and changes places with MLL1-N
the mechanistic reading of the one channel that does not replicate between the two lines
← depends onHypothesis: H3K79me2 at a KMT2A fusion's own sites survives menin inhibition when the C-terminal partner contacts DOT1L directly (MLLT3/AF9) and decays when it does not (AFF1) - so the mark's response is a property of the partner and not of menin, which predicts the opposite of the direct-contact intuition and is testable in one line
the mechanism the hypothesis needs: DOT1L occupancy falls to 0.22 in MOLM13 and the mark it writes does not move
← testsThe cheapest unrun test in this beat's queue: GSE260742 deposits an H3K79me2 ChIP under the DOT1L inhibitor SGC0946 in the same murine MLLAF9 cells as its FLAG and MLL1 tracks, which removes the catalysis directly instead of indirectly and would separate 'the mark is stable' from 'residual DOT1L is sufficient' - one download
the experiment that would choose between that mechanism entry's three readings
← supportsA DOT1L inhibitor DOUBLES H3K79me2's mid-occupancy share (1.89-2.01) and a menin inhibitor CUTS it (0.69-0.78), in THREE murine samples one of which is a replicate (corrected cycle 97; the id still says four) with no overlap between the ranges, while the FLAG and MLL1-N channels sit at 0.905-0.929 under both drugs - so the two drugs do not act on one axis and 'a fraction of the achievable effect' presumes they do
a second axis on which the DOT1L channel and the mark separate: the enzyme's inhibition and the interface inhibitor's move the mark's compartmental distribution in opposite directions
← supportsThe active chromatin landscape does not lose a single peak call within 24 hours of the fusion being destroyed: H3K79me2 holds 99.7% at both 3 h and 24 h, H3K9ac 100.0% at both, H3K27ac 83.3%/67.2%, H3K4me3 99.6%/78.9% - while the authors, measuring signal at fusion target genes, report a marked H3K79me2 reduction by 24 h
SAME DISSOCIATION UNDER A CLEANER PERTURBATION: that entry separates DOT1L eviction from H3K79me2 persistence under a drug at three days; here removing the fusion entirely gives DOT1L as the most-displaced protein (68.7%) and H3K79me2 as the least-moved mark (99.7% at both 3 h and 24 h).