H2BK34ub — a second ubiquitination site on H2B that also stimulates DOT1L, written by MSL1/MSL2, and important at HOXA9 and MEIS1
also called H2B K34ub · H2BK34ub1 · MSL2-dependent H2B ubiquitination
There is a second place on histone H2B that gets a ubiquitin tag, put there by a different enzyme, and it turns on DOT1L just as effectively. It was shown to matter for switching on HOXA9 and MEIS1 — the two genes KMT2A-fusion leukaemia is defined by.
The same thing, for a scientist
H2BK34 ubiquitination is the less-known second H2B ubiquitination mark, deposited not by RNF20/RNF40 but by **MSL2 within the MOF-MSL complex**, with MSL1 [E1] - an activity that is real on free histones but **contested on nucleosomes**, where an independent reinvestigation found MSL1/2 modify them very poorly (PMID 30930284, counter-evidence on [E1]). Three things make it more than a footnote for this graph. (1) **It regulates the same downstream marks**: H2BK34ub by MSL1/2 directly regulates H3K4 and H3K79 methylation through trans-tail crosstalk in vitro and in cells [E1]. (2) **It was shown at the right loci**: MSL1/2 activity is important for transcription activation at the **HOXA9 and MEIS1** loci [E1] — the two genes the KMT2A-fusion programme is defined by, and the same two genes where Rnf20 loss lowers H3K79me. (3) **It works by a different structural mechanism**: cryo-EM of a Dot1L–H2BK34ub nucleosome complex shows H2BK34ub induces an almost identical Dot1L orientation and binding pattern to H2BK120ub, but the K34-anchored ubiquitin does **not** contact Dot1L directly; instead it distorts DNA and histone structure around the modified site [E2]. Functionally this is the strongest argument in this beat against a site-directed strategy: two different ubiquitination sites, written by two different ligases, converge on the same DOT1L activation. The MOF-MSL complex is already in this graph as `gene-kat8` (MOF/KAT8, the H4K16 acetyltransferase) and as `subtraction-kmt2a-mof-contact`; what the graph did not record is that the same complex carries an E3 ligase activity aimed at H2B.
One complex, two chromatin activities, and a retained KMT2A contact
The MOF-MSL complex possesses two distinct chromatin-modifying activities: H4K16 acetylation through MOF (KAT8) and H2BK34 ubiquitination through MSL2 1. The graph already carries the KMT2A–MOF contact in subtraction-kmt2a-mof-contact and KAT8 as a protein entry, and it carries H4K16ac in eight entries. It did not carry the second activity at all.
That is a specific, checkable gap with a specific consequence: if the KMT2A fragment retained in the fusion contacts MOF, then it is in contact with a complex that can ubiquitinate H2B, and the H2Bub layer may reach fusion target genes by a route that has nothing to do with RNF20/RNF40 or PAF1C. Nobody has tested that. It is the strongest new hypothesis this beat generated and is written up as hyp-msl-route-to-h2bub-at-fusion-targets.
The caution is that subtraction-kmt2a-mof-contact is, as its id says, a subtraction entry — it concerns a contact examined in the context of what the fusion loses. Any reader building on this should read that entry before assuming the contact is retained in KMT2A::AFF1.
Contested, and this entry now says so in the body. The only independent reinvestigation of the activity found that purified MSL1/MSL2 ubiquitylate free histone substrates efficiently but modify nucleosomes very poorly, implying that efficient H2BK34 ubiquitylation requires a prior structural alteration of the nucleosome (PMID 30930284, Krajewski & Vassiliev 2019; counter-evidence on 1). Two further complications belong beside it: H2BK34 also carries a conserved methyl mark in mouse meiosis (PMID 32933557), so the residue is not dedicated to ubiquitin; and neither MSL2 nor MSL1 is selectively required in KMT2A-rearranged lines in this project's DepMap scan.
The feedback loop onto P-TEFb, which is the fusion's own machinery
Wu et al. went on to show that H2BK34 ubiquitylation by MOF-MSL is part of the network operating in early transcription elongation: MSL, PAF1 and RNF20/40 complexes are recruited and stabilised at active promoters by direct binary interactions, and the stabilised assembly regulates chromatin association of P-TEFb through a positive feedback loop, facilitating Pol II transition during early elongation 3.
KMT2A::AFF1 recruits P-TEFb through AFF1/AFF4 in the super elongation complex; that is the graph's mechanism-chromatin-elongation-hijack. So the H2B ubiquitination layer is not only upstream of the H3K79 mark — it is also, through this loop, upstream of the elongation factor the fusion exists to recruit. Whether the loop runs in a KMT2A-fusion cell is untested; that the two systems touch the same node is established.
| mark | H2BK34ub1 |
|---|---|
| histone | H2B |
| residue | K34, within a conserved basic patch on the H2B tail |
| writer complex | MOF-MSL (MSL2 RING E3 with MSL1) |
| mechanism vs k120 | same Dot1L orientation and productive conformation, but by DNA/histone distortion around K34 rather than direct ubiquitin-enzyme contact |
| sister activity same complex | H4K16 acetylation by MOF/KAT8 (gene-kat8) |
| therapeutic implication | a second, independently written ubiquitination site that activates DOT1L equally well is a built-in bypass of any strategy aimed at H2BK120 or at RNF20/RNF40 |
| activity caveat | MSL1/MSL2 ubiquitylate free histones efficiently but nucleosomes very poorly in vitro (PMID 30930284); efficient H2BK34ub may require prior nucleosome alteration |
| competing modification at same residue | H2BK34 methylation, conserved meiotic mark in mouse testes (PMID 32933557; yeast counterpart H2BK37me) |
writers
[ "gene-msl2" ]
licenses
[ "H3K79 methylation by DOT1L", "H3K4 methylation" ]
loci implicated
[ "HOXA9", "MEIS1" ]
- Two independent ubiquitination sites converging on the same DOT1L activation is a compensation mechanism that exists before any drug is given; a strategy against RNF20/RNF40 has a documented escape route on day one.
- The writer complex also carries MOF, whose H4K16 acetylation is essential across cancer lines (KAT8 pan-essentiality -1.452); targeting the complex rather than the subunit is not viable.
- MSL2 itself is essentially dispensable in DepMap (mean -0.083), so removing it is unlikely to remove enough of the signal to matter even if the route is real.
- The bypass argument in this entry assumes H2BK34ub is deposited readily on chromatin; the one independent in vitro reinvestigation says MSL1/2 modify nucleosomes very poorly (PMID 30930284), which would make the bypass narrow rather than automatic.
- Can MSL1/MSL2 ubiquitinate an ordinary nucleosome at all? PMID 30930284 says barely, and if the enzyme needs the nucleosome structurally altered first, then H2BK34ub is downstream of some other remodelling event rather than an independent entry point to DOT1L.
- How much of the total H2B ubiquitin pool is K34 rather than K120, in any cell, and specifically in a KMT2A-rearranged cell? If K34ub is a very minor species the bypass argument weakens considerably, and the mass-spectrometry to answer it is routine.
- Does the retained KMT2A fragment's MOF contact deliver MSL2 - and therefore H2B ubiquitination - to fusion target genes? Untested, and it would be a route to the mark entirely independent of RNF20/RNF40 and PAF1C.
- Does H2BK34ub rise when H2BK120ub is removed? That is the compensation experiment that decides whether RNF20 inhibition would be bypassed, and it is a single western blot in an RNF20-degron line.
- Is the MSL-PAF1-RNF20/40-P-TEFb feedback loop intact in KMT2A-rearranged cells, where P-TEFb is already being recruited by the fusion?
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
MSL2 in the MOF-MSL complex is a histone ubiquitin E3 ligase; MSL2 together with MSL1 has robust histone ubiquitylation activity targeting nucleosomal H2B on lysine 34, a site within a conserved basic patch on the H2B tail; H2BK34ub by MSL1/2 directly regulates H3K4 and H3K79 methylation through trans-tail crosstalk both in vitro and in cells; MSL1/2 activity is important for transcription activation at the HOXA9 and MEIS1 loci; and the MOF-MSL complex therefore possesses two distinct chromatin-modifying activities, H4K16 acetylation and H2BK34 ubiquitylation.
The RING finger protein MSL2 in the MOF complex is an E3 ubiquitin ligase for H2B K34 and is involved in crosstalk with H3 K4 and K79 methylation. · Wu L, Zee BM, Wang Y, Garcia BA, Dou Y · Mol Cell 2011What this evidence cannot show
Abstract read in full; full text not read. The HOXA9/MEIS1 result is in a non-leukaemic context - the abstract does not name a KMT2A-rearranged cell line and this beat does not assume one. 'Important for transcription activation' is not quantified in the abstract. Whether H2BK34ub contributes materially to total H2Bub in any cell is not addressed; it may be a minor species. **Resolved: the 'non-leukaemic context' is HeLa.** The full text (PMC4119175, recovered through `efetch db=pmc` after Europe PMC returned HTTP 500) contains **zero occurrences of the stem 'leukem'** in 55,724 characters. A cervical adenocarcinoma line is not merely non-leukaemic; it shares no lineage, no developmental origin and no chromatin context with the disease this graph is about. Resolved 2026-09-19 by the Q18b unstated-systems sweep from the study's deposition, not from its text; bounded as a statement about the deposited data rather than about the whole study.
Evidence that argues against this (3)
- The enzyme's activity on the substrate that matters is weak. Krajewski & Vassiliev assayed purified MSL1/MSL2 against free histones and against nucleosomes and found that MSL1/2 ubiquitylate histone substrates efficiently but modify NUCLEOSOMES very poorly, concluding that efficient H2BK34 ubiquitylation requires a prior structural alteration of the nucleosome. 'Robust nucleosomal H2B ubiquitylation activity' is therefore contested at the level of the in vitro assay that established it. In vitro, purified components, no cellular or leukaemia data - but it is the only independent reinvestigation of the activity this beat found. Krajewski WA et al. 2019
- Neither writer is required in the disease. In this project's genome-wide DepMap scan MSL2 has a KMT2A-selectivity difference of -0.015 (rank 7,112 of 17,728) on a pan-essentiality of -0.083, and MSL1 -0.073 (rank 2,078) on -0.365; KAT8 in the same complex is +0.004. A route said to be important at HOXA9 and MEIS1 leaves no trace on a knockout screen in 11 KMT2A-rearranged lines. The honest read is that a near-dispensable gene has little dynamic range in this assay, so this is weak evidence rather than none. Broad Institute DepMap; Cellosaurus; Open Targets Platform; scan run by this project et al. 2026
- Transferring an H2BK34ub result from HeLa to KMT2A-rearranged infant leukaemia crosses lineage, developmental stage and transformation mechanism in one step. The mark's deposition machinery may well be general, but this entry's claim is about its behaviour in a fusion-driven context, and nothing in the source addresses that. Oncofusion knowledge graph (internal analysis) 2026
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Cryo-EM structures of a Dot1L-H2BK34ub nucleosome complex and of the H2BK34ub nucleosome alone show H2BK34ub induces an almost identical orientation and binding pattern of Dot1L on the nucleosome as H2BK120ub, positioning Dot1L for the productive conformation; but the K34-anchored ubiquitin does not directly interact with Dot1L as H2BK120ub does, instead inducing DNA and histone distortion around the modified site.
H2B Lys34 Ubiquitination Induces Nucleosome Distortion to Stimulate Dot1L Activity. · Ai H, Sun M, Liu A, Sun Z, Liu T, Cao L, Liang L, Qu Q, Li Z, Deng Z, Tong Z, Chu G, Tian X, Deng H, Zhao S, et al. · Nat Chem Biol 2022What this evidence cannot show
Abstract read in full; full text not read. Reconstituted nucleosomes with chemically installed ubiquitin; resolutions not stated in the abstract. A structure showing a productive conformation does not establish the physiological abundance or importance of the mark.
Evidence that argues against this (1)
- K34 is not a residue dedicated to ubiquitin, which complicates every read-out at that position. Crespo et al., combining a systematic yeast histone mutational screen with mass spectrometry across sporulation, identified methylation of H2BK37 as a meiotic mark and showed the mouse paralogue **H2BK34** is likewise enriched for methylation during meiosis in the testes. A structural sufficiency result for K34-ubiquitin says nothing about which modification occupies K34 in a given cell, and an antibody or MS read-out has to distinguish them. Yeast and mouse germline, not human leukaemia. Crespo M et al. 2020
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
H2B K34 ubiquitylation by the MOF-MSL complex operates in early transcription elongation; MSL2 knockdown affects global H2BK34ub and multiple cotranscriptionally regulated histone modifications; the MSL, PAF1 and RNF20/40 complexes are recruited and stabilised at active gene promoters by direct binary interactions, and the stabilised complexes regulate chromatin association of P-TEFb through a positive feedback loop facilitating Pol II transition during early elongation.
H2B ubiquitylation promotes RNA Pol II processivity via PAF1 and pTEFb. · Wu L, Li L, Zhou B, Qin Z, Dou Y · Mol Cell 2014What this evidence cannot show
Abstract read in full; full text not read. Not a leukaemia system and not a KMT2A-fusion system. The link to the fusion's P-TEFb recruitment is this project's inference and is stated as untested.
Evidence that argues against this (1)
- The P-TEFb feedback loop is reported with the opposite sign elsewhere. Chen et al. found that reducing PAF1 INCREASES release of paused Pol II at thousands of genes by recruiting the super elongation complex, raising Ser2P. A positive PAF1/H2Bub-to-P-TEFb feedback loop and PAF1 acting as a brake on pause release cannot both be the general rule, and the second result is the one obtained genome-wide in human cells. Chen FX et al. 2015
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
In the project's genome-wide DepMap KMT2A-selectivity scan, neither MSL2 nor MSL1 is selectively required in KMT2A-rearranged lines: MSL2 difference -0.015 (rank 7,112/17,728) on a pan-essentiality of -0.083, and MSL1 difference -0.073 (rank 2,078) on -0.365; KAT8/MOF is strongly pan-essential (-1.452) with no selectivity (difference +0.004).
DepMap CRISPR (Chronos) genome-wide KMT2A-selectivity scan of 17,728 protein-coding genes, 11 Cellosaurus-annotated KMT2A-rearranged leukaemia lines vs 43 other leukaemia lines · Broad Institute DepMap; Cellosaurus; Open Targets Platform; analysis run by this project (analyses/depmap-kmt2a-selective-scan-full) · Database and derived analysis 2026What this evidence cannot show
Eleven KMT2A-rearranged lines, AML-weighted. MSL2's near-zero pan-essentiality means the assay has little dynamic range for it; a non-essential gene cannot show a large selectivity difference in either direction.
Evidence that argues against this (1)
- SCAN CONFOUND: this number comes from the project's genome-wide KMT2A-selectivity scan, whose index arm is MYELOID-ENRICHED (8 AML lines against 5 B-ALL). A gene whose requirement differs by lineage is mis-scored in BOTH directions by that design — CEBPA ranks 25 genome-wide with no dependency at all in KMT2A-rearranged B-ALL, and conversely a gene required in one lineage and not the other can average to apparent non-selectivity. See `ana-lineage-confound-myeloid-tf-dependencies`. For MSL1 and MSL2 the more informative limitation is different: the K34 mark's writer complex overlaps the NSL complex in composition, so single-gene knockouts distribute the phenotype across a shared membership the scan scores independently. Oncofusion knowledge graph (internal analysis) 2026
- ← related toMSL2 — the second H2B ubiquitin ligase, sitting inside the MOF complex the KMT2A fragment contacts
Counter-evidence (Krajewski and Vassiliev 2019, PMID 30930284): purified MSL1/MSL2 ubiquitylate free histone substrates efficiently but modify nucleosomes very poorly, implying efficient H2BK34 ubiquitylation needs the nucleosome structurally altered first - the founding claim of robust nucleosomal activity is contested at the level of the assay that established it
Counter-evidence (DepMap scan, this project): MSL2 -0.015 (rank 7,112) on a pan-essentiality of -0.083, MSL1 -0.073 (rank 2,078), KAT8 +0.004 - a route said to matter at HOXA9 and MEIS1 leaves no trace on a knockout screen across 11 KMT2A-rearranged lines
Counter-evidence (Chen 2015, PMID 26279188): a positive MSL/PAF1/RNF20-to-P-TEFb feedback loop cannot be the general rule alongside the genome-wide human finding that PAF1 depletion increases pause release and Ser2P by recruiting SEC
- related to →MSL2 — the second H2B ubiquitin ligase, sitting inside the MOF complex the KMT2A fragment contacts
Counter-evidence (Crespo 2020, PMID 32933557): H2BK34 is not a residue dedicated to ubiquitin - the mouse paralogue of yeast H2BK37 is enriched for METHYLATION during meiosis in the testes - so any K34 read-out must identify the modification rather than assume it
- related to →Does the fusion carry an H2B ubiquitin ligase? A co-IP and an H2BK34ub CUT&RUN in KMT2A::AFF1 cells
Design objection (Egelhofer 2011, PMID 21131980; Crespo 2020, PMID 32933557): more than 25% of 200+ histone-modification antibodies failed specificity and more than 20% of the survivors failed in ChIP, and H2BK34 also carries a conserved meiotic methyl mark - so the validation panel needs a K34-methylated control, and the most likely single outcome of Part B is a confident map of the wrong thing
- ← causesMSL2 — the second H2B ubiquitin ligase, sitting inside the MOF complex the KMT2A fragment contacts
RING E3 that ubiquitinates nucleosomal H2B at K34 with MSL1
positions Dot1L for the productive conformation by nucleosome distortion
- analogous to →H2BK120ub (H2Bub1) — the ubiquitin tag on histone H2B that licenses H3K79 methylation, and the mark this graph was missing
a second H2B ubiquitination site converging on the same DOT1L activation
MSL1/2 activity important for transcription activation at HOXA9
MSL1/2 activity important for transcription activation at MEIS1
the hypothesised fusion-proximal route runs through the K34 mark
- ← related toThe H2BK34ub HOXA9/MEIS1 result is HeLa: the full text contains the word leukaemia zero times
resolves the non-leukaemic context to HeLa; the full text contains no occurrence of the stem leukem