H3K79me1 - the first methyl group, the one H2B ubiquitination is not needed for, and the state HOX loci fall back to as blood cells mature
also called H3K79 monomethylation · H3K79me · histone H3 lysine 79 monomethylation · H3K76me1 (trypanosome equivalent)
H3K79me1 is a single methyl group on lysine 79 of histone H3, the first of up to three that DOT1L can add. It is the state that normal blood cells settle into at their HOX genes as they mature, and unlike the higher states it does not need the ubiquitin tag on histone H2B to be made.
The same thing, for a scientist
H3K79me1 is the mono-methylated state of the DOT1L substrate lysine. Three properties separate it from H3K79me2/3 and none of them are visible if the mark is treated as one thing. **(1) It is H2Bub-independent.** Ubiquitylation of H2B-K123 is dispensable for monomethylation of H3-K79 but required for the transition from mono- to the subsequent states, and Dot1 still binds chromatin normally without it, so H2Bub regulates processivity rather than recruitment [E1]. **(2) It is the resting state of HOX chromatin in normal differentiation.** H3K79me2 is converted to H3K79me1 at HOX loci as haematopoietic cells mature, coinciding with the fall in HOX expression; the me1-to-me2 conversion is the step the DOT1L cofactor AF10 controls [E2]. **(3) It has its own reader and its own genomic logic.** The SMN TUDOR domain binds H3K79me1 through an aromatic cage - the first protein shown to read this modification, and the first reader described that recognises both methyl-arginine and methyl-lysine [E3] - and in the human genome H3K79me1 tracks gene activation while H3K79me3 tracks repression [E4]. In yeast, Dot1-*repressed* genes are the ones enriched for H3K79me1 [E5]. For a KMT2A-fusion graph the operational point is that an anti-H3K79me2 ChIP that loses signal has not necessarily lost methylation; it may have lost a methyl group.
Why a separate entry, when the project's own citation records 'H3K79 methylation'
Because the states are not interchangeable in the one direction that matters clinically. A DOT1L inhibitor collapses the higher states first: H2Bub is required for the mono-to-di transition but not for mono itself 1, so partial SAM-competitive inhibition should produce a chromatin state that is mono-methylated and not unmethylated, and me1 is the state associated with the mature, low-HOX configuration 2 and with activation rather than repression genome-wide 4.
That makes the therapeutically relevant question "did you reach me1, or did you reach zero?" - and an antibody raised against H3K79me2 cannot answer it. The pinometostat trial reported H3K79 methylation reduction as its pharmacodynamic endpoint 6; which state was measured, and what the residual state was, is the fact that would distinguish incomplete target engagement from complete engagement with a surviving mark.
This graph holds 35-odd entries resting on H3K79me2 and, before this cycle, none on the individual states. The PI's standing granularity instruction is one entry per thing, and Q3b already applies it to H2B ubiquitination in mark-h2bk120ub and mark-h2bk34ub.
The reader, and how much weight it can carry
SMN - the spinal muscular atrophy protein - associates with H3K79me1 through the aromatic cage of its TUDOR domain, and most patient-derived SMN TUDOR mutants fail to bind it 3. This is biochemistry in a non-haematopoietic context and no function in leukaemia has been shown, so it belongs in this entry as the answer to "is there a reader at all" and not as a therapeutic hypothesis.
It matters for one structural reason. If mono-, di- and trimethyl states are generated non-processively from a common pool, as the kinetics say they are (mechanism-h3k79-nonprocessive-deposition), then a reader that discriminates me1 from me2 is the mechanism by which state-specific function could exist at all. One such reader is now known. Whether the leukaemic H3K79me2 domain has a reader remains open and is one of this beat's live gaps.
| mark | H3K79me1 |
|---|---|
| histone | H3 |
| residue | K79 (human and mouse); K79 in S. cerevisiae; K76 in trypanosomes |
| chemistry | mono-methylation of a lysine epsilon-amine on the nucleosome lateral surface, not the histone tail |
| writer | gene-dot1l |
| writer is sole | yes - DOT1L is the only known H3K79 methyltransferase in mammals |
| h2bub dependence | independent: H2B-K123ub is dispensable for H3K79 monomethylation and required only for the transition to higher states |
| erasers known | none established; see mechanism-h3k79-demethylation-contested |
| genomic association human | gene activation (Barski 2007, CD4+ T cells) |
| genomic association yeast | Dot1-repressed genes (Park 2026) |
| haematopoietic behaviour | HOX loci shift from H3K79me2 to H3K79me1 as cells mature, coinciding with falling HOX expression |
| cofactor controlling transition to me2 | gene-mllt10 (AF10) |
readers known
[ "SMN TUDOR domain (in vitro, non-haematopoietic)" ]
- A residual H3K79me1 state that a me2-directed pharmacodynamic assay cannot see would make a trial read as target-engaged while the chromatin is still partly marked - the failure mode in which a drug is abandoned for lack of efficacy at a dose that never fully removed the mark.
- If H3K79me1 is itself activating in human cells, then driving a KMT2A-fusion leukaemia from me2 to me1 removes less of the oncogenic signal than the pharmacodynamic number suggests.
- Normal maturing progenitors depend on this same mono-methylated state at HOX loci, so a state-selective intervention has no obvious tumour-versus-host margin built into it.
- In KMT2A-rearranged blasts under a DOT1L inhibitor, does H3K79me2 loss leave H3K79me1 behind or go to zero, and does the answer differ between fusion-target gene bodies and the rest of the genome? No published dataset in this disease reports the three states separately with calibrated quantification.
- Does H3K79me1 have any reader in haematopoietic cells? SMN is the only reader described for any H3K79 state and was found outside blood.
- If maturing normal progenitors park HOX loci at H3K79me1 rather than at zero, is the leukaemic defect an excess of the me1-to-me2 step specifically, and would a partial inhibitor that restored the normal me1 set-point be better tolerated than one aiming for full ablation?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Ubiquitylation of histone H2B-K123 is dispensable for monomethylation of H3-K79 (and H3-K4) but is required for the transition from monomethylation to subsequent methylation states; Dot1 binds chromatin normally in the absence of H2B-K123 ubiquitylation, so the ubiquitin regulates processive activity rather than enzyme recruitment.
Histone H2B ubiquitylation controls processive methylation but not monomethylation by Dot1 and Set1. · Shahbazian MD, Zhang K, Grunstein M · Mol Cell 2005What this evidence cannot show
Abstract read from the eutils record; full text not read. Saccharomyces cerevisiae, and the residue is H2B-K123 not the human K120. Chromatin immunoprecipitation plus mass spectrometry in bulk chromatin, so this is a genome-average statement and not a locus-level one. Nothing here is measured in a KMT2A-fusion context.
Evidence that argues against this (1)
- The human structural work shows H2Bub does more than license processivity: cryo-EM of DOT1L on ubiquitinated nucleosomes shows DOT1L reading the ubiquitin through a C-terminal hydrophobic helix and the acidic patch through an arginine anchor, and single-molecule FRET shows H2Bub enhancing a NON-catalytic nucleosome-destabilising function of DOT1L. A model in which H2Bub only tunes the mono-to-di step understates it. Jang S et al. 2019
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
H3K79me2 is converted to H3K79me1 at HOX loci as haematopoietic cells mature, coinciding with a decrease in HOX gene expression, and the H3K79me1-to-H3K79me2 conversion is regulated by the DOT1L cofactor AF10.
AF10 regulates progressive H3K79 methylation and HOX gene expression in diverse AML subtypes. · Deshpande AJ, Deshpande A, Sinha AU, et al. · Cancer Cell 2014What this evidence cannot show
Abstract read from the eutils record; full text not read. Mouse haematopoietic maturation and mouse leukaemia models (MLL-AF9, MLL-AF6, NUP98-NSD1); no KMT2A::AFF1 model is named in the abstract, and KMT2A::AFF1 contains no AF10. 'Conversion' is inferred from the anticorrelated distribution of the two states across maturation stages rather than from a pulse-chase on individual nucleosomes.
Evidence that argues against this (1)
- The kinetic mechanism argues against reading 'conversion' as a regulated switch between discrete states. Dot1 methylates non-processively, which means the states cannot be generated independently of one another and a shift in the state distribution is what a change in total enzymatic flux looks like. On that model AF10 changes how much methylation happens, not which state is chosen. Frederiks F et al. 2008
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The TUDOR domain of SMN associates with histone H3 monomethylated on lysine 79 via an aromatic cage; SMN is the first protein shown to associate with H3K79me1 and the first histone-mark reader shown to recognise both methylated arginine and methylated lysine, and most SMN TUDOR mutants found in spinal muscular atrophy patients fail to associate with H3K79me1.
The TUDOR domain of SMN is an H3K79(me1) histone mark reader. · Binda O, Kimenyi Ishimwe AB, Galloy M, et al. · Life Sci Alliance 2023What this evidence cannot show
Abstract read from the eutils record; full text not read. In vitro biochemistry with peptides and mutational analysis, in a spinal muscular atrophy context. No leukaemia, no haematopoietic cell, and no demonstration that the interaction occurs on nucleosomal chromatin in cells or has a transcriptional consequence. Binding affinity is not stated in the abstract.
Evidence that argues against this (1)
- The kinetic argument predicts that state-specific readers should be rare and weakly discriminating: because the methylation states are produced non-processively they co-occur, which the authors of that work state 'is expected to limit the possibilities for regulation by putative demethylases or binding proteins'. A single peptide-binding TUDOR domain is not yet evidence against that expectation. Frederiks F et al. 2008
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In genome-wide maps of 20 histone methylations in human CD4+ T cells, monomethylation of H3K79 (together with H3K27me1, H3K9me1, H4K20me1 and H2BK5me1) is linked to gene activation, whereas trimethylation of H3K79 (with H3K27me3 and H3K9me3) is linked to repression.
High-resolution profiling of histone methylations in the human genome. · Barski A, Cuddapah S, Cui K, et al. · Cell 2007What this evidence cannot show
Abstract read from the eutils record; full text not read. Resting human CD4+ T cells only - one cell type, not haematopoietic stem or progenitor cells and not leukaemic blasts. The activation/repression assignment is a correlation between ChIP-seq occupancy and expression, not a perturbation. Antibody specificity across adjacent methylation states is the well-known weakness of every study of this class and is not independently established here.
Evidence that argues against this (1)
- The opposite polarity is reported in yeast when the same question is asked functionally rather than by correlation: Dot1-ACTIVATED genes are enriched for H3K79me3 while a majority of Dot1-REPRESSED genes carry H3K79me1. Whether the sign flips between species or between correlative and perturbative readouts is unresolved, and this entry should not be cited for a fixed activation polarity of either state. Park NH et al. 2026
- E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
ChIP-seq mapping of H3K79me1, me2 and me3 across the yeast genome identifies three distinct gene groups each predominantly marked by one state; the H3K79me1 zone is diminished by loss of Rad6 and weakly disrupted by loss of H4K16 acetylation, and a majority of Dot1-repressed genes are associated with H3K79me1.
Trans-histone crosstalk establishes distinct H3K79 methylation zones with differential transcriptional functions. · Park NH, Kim HR, Kim HY, et al. · Nucleic Acids Res 2026What this evidence cannot show
Abstract read from the eutils record; full text not read. Saccharomyces cerevisiae. Published 2026 and not yet independently replicated. The zone assignment rests on state-specific antibodies, which is precisely the reagent class whose cross-reactivity between adjacent methylation states is hardest to exclude, and the abstract does not report a peptide-array specificity control. The authors state the mechanism only as 'may be established and/or maintained via' H2B ubiquitination.
Evidence that argues against this (1)
- This is the direct modern challenge to the functional-redundancy model, and the redundancy model is the older and better-established one: Dot1 introduces methyl groups non-processively, the states therefore cannot be generated independently, and yeast Dot1-dependent silencing depended on GLOBAL H3K79 methylation levels rather than on any one state. Both cannot be fully right and the entry records the conflict rather than resolving it. Frederiks F et al. 2008
- E6established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
In the adult phase 1 of the DOT1L inhibitor pinometostat, H3K79 methylation was reduced in patients as intended, with 2 complete remissions among 51 treated adults.
The DOT1L inhibitor pinometostat reduces H3K79 methylation and has modest clinical activity in adult acute leukemia. · Stein EM, Garcia-Manero G, Rizzieri DA, et al. · Blood 2018What this evidence cannot show
Abstract read from the eutils record; full text not read. Cited here only for the pharmacodynamic endpoint and the remission count. The trial's maximum tolerated dose was never reached, so 'reduced as intended' is a statement about a dose range that was not taken to its limit. Which methylation state the pharmacodynamic assay measured, and what residual state remained, is not resolvable from the abstract - which is the specific gap this entry exists to name.
Evidence that argues against this (1)
- Some KMT2A-rearranged leukaemias are susceptible to pinometostat concentrations far below those that deplete the canonical HOXA9/MEIS1 axis, through non-canonical H3K79me2-dependent pathways running mainly through FLT3. If low-level H3K79me2 loss is already cytotoxic by a separate route, the relationship between residual methylation state and clinical effect is not monotonic and the 'did you reach me1 or zero' framing is necessary but not sufficient. Richter WF et al. 2021
- ← co-occurs withHistone H3.3 - one protein, two genes, and the reason this graph should not file it as gene-h3f3a
H3.3K79me1 is higher than H3.2/H3.1 at equal methylation rate
sole mammalian H3K79 methyltransferase; deposits the mono-methyl state
H2Bub is dispensable for mono-methylation and required only for the transition beyond it
AF10 controls the H3K79me1-to-me2 conversion at HOX loci
state-resolved granularity for the mark that entry treats as one thing
me1 to me2 conversion is the AF10-controlled step
distributive kinetics set the me1 share of the distribution
does partial inhibition leave me1 behind or reach zero
loss of H4K16ac weakly disrupts the H3K79me1 zone in yeast