H2BK120ub (H2Bub1) — the ubiquitin tag on histone H2B that licenses H3K79 methylation, and the mark this graph was missing
also called H2Bub1 · uH2B · H2B-Ub · H2BK120ub1 · monoubiquitinated histone H2B · H2BK123ub (yeast)
A whole ubiquitin protein is clipped onto lysine 120 of histone H2B while a gene is being read, and that tag is what allows DOT1L to place the H3K79 mark that KMT2A-fusion leukaemia depends on. It is written by the RNF20/RNF40 enzyme as the gene is transcribed and stripped off again by deubiquitinases, so the level on any gene is a balance rather than a switch.
The same thing, for a scientist
H2BK120 monoubiquitination (H2Bub1; K123 in budding yeast) is a bulky, dynamic, co-transcriptional histone mark deposited on nucleosomes in the bodies of actively transcribed genes. In human cells the E3 is the RNF20/RNF40 heterodimer (BRE1A/BRE1B), stabilised and targeted by the WAC adaptor and recruited through the PAF1 complex riding elongating Pol II; the identity of the E2 is disputed between UbcH6/UBE2E1 and hRAD6/UBE2A-UBE2B. Unlike a methyl or acetyl group, the modification adds an entire 8.5 kDa protein to the nucleosome surface adjacent to the H3 core, which is why it acts as a docking and allosteric platform rather than only as a reader site: it directly stimulates DOT1L-mediated H3K79 methylation and SET1/COMPASS-mediated H3K4me2/3. **The reason this entry exists is arithmetic.** This graph carried 35 entries resting on H3K79me2 and 43 on H3K4me3, and not one entry on the modification that licenses both. Wang et al. showed by ChIP-seq that H2B ubiquitination is enriched across the bodies of MLL-fusion target genes, correlating with H3K79 methylation and elongation, and that Rnf20 is required to maintain Dot1l-dependent H3K79 methylation at *Hoxa9* and *Meis1* specifically [E3]. So the mark is not merely upstream in the abstract; it has been measured at the exact loci this disease runs on.
What it physically is, and why that matters mechanistically
H2Bub1 is not a small chemical group. It is a 76-residue ubiquitin moiety joined by an isopeptide bond to the epsilon-amine of H2B K120, sitting at the C-terminal end of the H2B alpha-C helix on the nucleosome disc face, near the H2A/H2B acidic patch and within a few nanometres of H3 K79 in the nucleosome core 12.
Three consequences follow, and they are why this mark behaves differently from every other mark in this graph:
- It is a direct allosteric effector, not just a reader docking site. Chemically ubiquitylated H2B reconstituted into defined nucleosomes stimulates hDOT1L methylation of H3K79 within the same nucleosome, through the DOT1L catalytic domain, with no reader protein required 4. Asymmetric dinucleosomes showed the stimulation is restricted to the nucleosome that carries the tag — an intranucleosomal effect, not a spreading one 4.
- It is deposited by the elongation machinery, so it is a record of transcription as much as an instruction to transcribe. In reconstituted human systems H2B ubiquitination per se does not change the level of hPAF-, SII- and p300-dependent transcription; it functions downstream of elongation and feeds forward into methylation 2.
- It is reversible and fast. The SAGA deubiquitinase module removes it, and the steady-state level at any gene is a writer/eraser balance. That is why knocking down the writer and knocking down the eraser can both be deleterious — see
mechanism-h2bub-removal-and-dub-redundancy.
The measurement that makes this a KMT2A entry rather than a chromatin-textbook entry
Wang et al. 2013 did the experiment that connects the mark to this disease 3:
| Experiment | Result |
|---|---|
| ChIP-seq for H2B ubiquitination in MLL-rearranged leukaemia | H2Bub enriched across the bodies of MLL-fusion target genes, correlating with H3K79 methylation and elongation |
| Rnf20 knockdown, H3K79me at Hoxa9 and Meis1 | local H3K79 methylation falls |
| Rnf20 knockdown, transcriptional output | reduced expression of MLL-fusion target genes, resembling Dot1l inhibition |
| Rnf20 knockdown, proliferation in vitro and in vivo | inhibited |
The authors' model — co-transcriptional recruitment of Rnf20 at fusion target genes amplifying Dot1l-mediated H3K79 methylation — places H2Bub1 upstream of the single most-cited mechanistic step in KMT2A-rearranged leukaemia.
What it does not establish, and what this beat found on re-examination, is that this upstream position translates into a therapeutic window. See ana-depmap-h2bub-machinery-kmt2a-selectivity: across 11 KMT2A-rearranged and 43 other leukaemia lines, RNF20 is less selectively required in the rearranged lines (difference +0.076, rank 16,229 of 17,728), while being broadly essential everywhere (pan-essentiality -0.734). Required is not the same as selective, and this beat's central correction is that the H2Bub machinery is the former and not the latter.
Where the mark sits in the graph's existing chain
The graph's mechanistic spine runs: fusion binds chromatin via menin/LEDGF (mechanism-menin-ledgf-chromatin-tethering) → recruits elongation machinery (mechanism-chromatin-elongation-hijack) → DOT1L deposits H3K79me2/3 (mechanism-dot1l-h3k79-methylation).
H2Bub1 inserts between the second and third steps and is a genuine requirement of the third, not a decoration on it. It also inserts into the second: H2BK34 ubiquitination stabilises MSL, PAF1 and RNF20/40 at active promoters and feeds back on P-TEFb chromatin association and Pol II processivity 5 — and P-TEFb is the component the KMT2A::AFF1 fusion recruits through AFF1/AFF4. So the ubiquitin layer touches both the mark and the elongation machinery.
| mark | H2BK120ub1 |
|---|---|
| histone | H2B |
| residue | K120 (human); K123 (S. cerevisiae) |
| chemistry | monoubiquitination - isopeptide bond to a full 76-residue ubiquitin, not a small chemical group |
| genomic distribution | gene bodies of actively transcribed genes; in MLL-rearranged leukaemia enriched across the bodies of MLL-fusion target genes |
| kmt2a relevance | required to maintain Dot1l-dependent H3K79 methylation at Hoxa9 and Meis1 |
| graph gap closed | Before this entry the graph held 35 entries mentioning H3K79me2 and 43 mentioning H3K4me3, with one passing mention of H2Bub and one of RNF20 and nothing else. |
writers
[ "gene-rnf20", "gene-rnf40", "gene-wac", "gene-ube2a", "gene-ube2b" ]
e2 disputed
[ "UBE2E1 (UbcH6; Zhu 2005)", "UBE2A/UBE2B (hRAD6; Kim 2009)" ]
erasers
[ "gene-usp22", "gene-atxn7l3", "USP27X", "USP51" ]
licenses
[ "H3K79me1/me2/me3 by DOT1L", "H3K4me2/me3 by SET1/COMPASS" ]
- The machinery that writes this mark is required by nearly every dividing cell (RNF20 pan-essentiality -0.734 across 1,208 DepMap lines), so an inhibitor of the writer is a general anti-proliferative with no KMT2A window, not a targeted agent.
- H2Bub1 also licenses H3K4me2/3 by SET1/COMPASS, so suppressing it hits the wild-type KMT2A/SET1-family programme as well as the fusion programme - the opposite of the selectivity that made DOT1L attractive.
- The mark is a writer/eraser balance, so incomplete inhibition of the writer will be partly buffered by reduced eraser flux, and the dose-response is likely narrower than a simple knockdown suggests.
- H2Bub1 has a substantial DNA-damage-response role independent of transcription; an agent that lowers it globally risks genotoxicity in normal tissue on top of proliferative toxicity.
- Has H2BK120ub ever been profiled in a human KMT2A::AFF1 infant B-ALL sample or line? Wang et al. worked in MLL-fusion models whose fusion identity the abstract does not state, and every H2Bub measurement in this disease that this beat could find is AML-weighted.
- Which E2 actually charges RNF20/RNF40 on chromatin in a leukaemia cell - UBE2E1 or UBE2A/UBE2B? The two primary papers disagree and DepMap gives neither a selective phenotype, so the question is open and cheap to answer with a degron pair.
- Does the H2Bub level at fusion target genes change under menin inhibition? If the fusion's chromatin residence drives co-transcriptional RNF20 recruitment, revumenib should collapse H2Bub at HOXA9/MEIS1 before H3K79me2 falls, and nobody has looked.
- Is there a subset of fusion target genes whose H3K79me2 is H2Bub-independent? Those genes would be the ones that survive an RNF20 loss and would predict the escape route from any writer-directed strategy.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
In human cells the 600 kDa RNF20/RNF40 complex is the E3 ligase and UbcH6 the E2-conjugating enzyme for H2B-Lys120 monoubiquitination; UbcH6 interacts with RNF20/40 and with the hPAF complex, the trimeric complex stimulates activity in vitro, all three are recruited to transcriptionally active genes, RNF20 overexpression raises H2Bub1 and subsequently H3K4 and H3K79 methylation and HOX gene expression, and RNAi against RNF20/40 or hPAF lowers all three.
Monoubiquitination of human histone H2B: the factors involved and their roles in HOX gene regulation. · Zhu B, Zheng Y, Pham AD, Mandal SS, Erdjument-Bromage H, Tempst P, Reinberg D · Mol Cell 2005What this evidence cannot show
Abstract read in full from the eutils record; full text not read. The HOX-gene readout is in a non-leukaemic human cell context and concerns HOX genes generally, not a KMT2A fusion. This paper assigns the E2 to UbcH6 (UBE2E1); a later paper assigns it to hRAD6 - see counter-evidence and `gene-ube2a`.
Evidence that argues against this (1)
- The E2 identity is genuinely contested in the primary literature. Zhu et al. 2005 name UbcH6 (UBE2E1); Kim et al. 2009 in Cell state that hRAD6 serves as the cognate E2 and is recruited through direct interaction with hPAF-bound hBRE1. This graph records both and asserts neither as settled. Kim J et al. 2009
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
hRAD6 serves as the cognate E2-conjugating enzyme for human H2B ubiquitylation, is recruited to transcribed genes through direct interaction with hPAF-bound hBRE1 and ubiquitylates chromatinised H2B at lysine 120; hPAF-mediated transcription is required for efficient H2B ubiquitylation; H2B ubiquitylation per se does not affect the level of hPAF-, SII- and p300-dependent transcription and functions downstream; and H2B ubiquitylation directly stimulates hSET1-dependent H3K4 di- and trimethylation.
RAD6-Mediated transcription-coupled H2B ubiquitylation directly stimulates H3K4 methylation in human cells. · Kim J, Guermah M, McGinty RK, Lee JS, Tang Z, Milne TA, Shilatifard A, Muir TW, Roeder RG · Cell 2009What this evidence cannot show
Abstract read in full; full text not read. Reconstituted chromatin transcription systems plus cell work; not leukaemia. The paper's direct-stimulation result is for hSET1/H3K4, not for DOT1L/H3K79 - the H3K79 arm rests on separate papers cited in `mechanism-h2bub-dot1l-trans-histone-crosstalk`.
Evidence that argues against this (1)
- hRAD6 recruitment through hPAF-bound hBRE1 was established in human cells with the pathway reconstituted, and RAD6 has two human paralogues (UBE2A and UBE2B) that are not distinguished by the antibodies and reagents typical of this era. Which paralogue operates at a given locus is not settled by this work. Cell 2009
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In MLL-rearranged leukaemia, ChIP-seq showed H2B ubiquitination is enriched in the body of MLL-fusion target genes, correlating with sites of H3K79 methylation and transcription elongation; Rnf20 is required to maintain local H3K79 methylation by Dot1l at Hoxa9 and Meis1; Rnf20 suppression reduced expression of MLL-fusion target genes resembling Dot1l inhibition, and inhibited proliferation in tissue culture and in vivo.
Histone H2B ubiquitin ligase RNF20 is required for MLL-rearranged leukemia. · Wang E, Kawaoka S, Yu M, Shi J, Ni T, Yang W, Zhu J, Roeder RG, Vakoc CR · Proc Natl Acad Sci U S A 2013What this evidence cannot show
Abstract read in full; full text not read. shRNA knockdown rather than genetic deletion or degron, so hypomorphic and subject to off-target and partial-knockdown effects. Mouse MLL-fusion models; the abstract does not state which fusions or whether any B-ALL/KMT2A::AFF1 model was included, and this project's disease is infant B-ALL. 'Required for MLL-rearranged leukaemia' is a statement about necessity in those models, not a demonstration of selectivity against non-MLL leukaemia. **Resolved:** GSE43725 is three samples of one murine RN2 (MLL-AF9/NrasG12D) line, n=1 per hairpin, with an unflagged cooperating NRAS G12D. 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 (2)
- DepMap CRISPR across 11 KMT2A-rearranged and 43 other leukaemia lines finds RNF20 is not selectively required in the rearranged group: difference +0.076 (rank 16,229 of 17,728 scanned genes, i.e. in the tail pointing the opposite way), on a pan-essentiality of -0.734 across 1,208 lines. RNF20 is required by nearly every cancer line, which is compatible with the knockdown phenotype but removes the selectivity that would make it a target. Broad Institute DepMap; Cellosaurus; Open Targets Platform; analysis run by this project 2026
- Two entries in this graph that read as independent support for H2B-ubiquitination biology in KMT2A-rearranged leukaemia rest on the same three mouse samples. Source independence at the dataset level is lower here than the citation count suggests. NCBI Gene Expression Omnibus 2026
- E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Chemically ubiquitylated H2B reconstituted into chemically defined nucleosomes directly stimulates hDot1L-mediated intranucleosomal methylation of H3 K79; the effect is mediated through the catalytic domain of hDot1L, most likely allosterically, and asymmetric incorporation into dinucleosomes showed the enhancement is limited to the nucleosome bearing ubiquitylated H2B.
Chemically ubiquitylated histone H2B stimulates hDot1L-mediated intranucleosomal methylation. · McGinty RK, Kim J, Chatterjee C, Roeder RG, Muir TW · Nature 2008What this evidence cannot show
Abstract read in full; full text not read. Fully reconstituted in vitro system with semisynthetic histones; the isopeptide linkage is a chemical mimic produced by expressed protein ligation and desulphurisation, not the native enzymatic product. The quantitative stimulation fold is not given in the abstract.
Evidence that argues against this (1)
- Chemically ubiquitylated H2B in chemically defined nucleosomes is the cleanest possible demonstration of direct stimulation and the furthest from chromatin. Intranucleosomal methylation in a defined system does not establish the magnitude of the effect on a real gene, where nucleosome density, competing marks and Pol II passage all intervene. Nature 2008
- E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
H2B K34 ubiquitylation by the MOF-MSL complex is part of the network operating in early transcription elongation; 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 P-TEFb link is inferred here to be relevant to KMT2A::AFF1 because the fusion recruits P-TEFb through AFF1/AFF4, which is this project's inference and not the paper's claim.
Evidence that argues against this (1)
- H2B K34 ubiquitylation by MOF-MSL is a SECOND H2B ubiquitination site, and the entry is about K120. Placing them together invites the reader to treat H2Bub as one quantity when the two marks have different writers, different positions and different reported consequences for elongation. Molecular Cell 2014
H2Bub is dispensable for mono-methylation and required only for the transition beyond it
H2Bub licenses the transition beyond mono-methylation
strongest H2Bub dependence of the three states; Rad6 loss abolishes the me3 zone
H2Bub enhances the non-catalytic nucleosome-destabilising activity
H2BK120ub licenses DOT1L-mediated H3K79 methylation; Rnf20 loss lowers H3K79me at Hoxa9 and Meis1
direct intranucleosomal stimulation of DOT1L catalysis
- part of →Plain-English: chromatin and epigenetics — the sticky notes on the DNA, and which ones matter here
a histone post-translational modification
deposited co-transcriptionally by PAF1C-associated machinery on elongating Pol II
H2Bub directly stimulates hSET1-dependent H3K4 di- and trimethylation
- ← depends onThe trans-histone crosstalk H2Bub → DOT1L → H3K79me: modulatory for me1, effectively required for me2/me3
the crosstalk requires the H2B ubiquitin mark
catalytic RING subunit of the E3 that monoubiquitinates H2B K120
obligate heterodimer partner of RNF20 in the human BRE1 ligase
adaptor required for H2B ubiquitination; depletion abolishes the mark
E2 conjugating enzyme for H2B K120 ubiquitination in one primary account; contested
PAF1C/RTF1 is required for H2B ubiquitination and imposes its transcription-coupled specificity
Rtf1 is required for H2B ubiquitination; its effects on H3K4 and H3K79 methylation are an indirect consequence
hPAF-dependent recruitment of hBRE1-hRAD6 to the Pol II machinery is required for efficient H2B ubiquitination
the H3K4 arm of the same crosstalk
a second H2B ubiquitination site converging on the same DOT1L activation
erases the mark; steady-state level is a writer/eraser balance
catalytic subunit of the SAGA DUB module; ablation lowers rather than raises the mark
limiting adaptor for three H2Bub deubiquitinases; its loss strongly raises the mark
DOT1L binds the ubiquitin conjugated to H2B K120
ranks every writer, carrier and eraser of the mark against the whole genome
the mark is necessary without being selectively necessary
- ← related toNobody knows which way to push H2Bub in this disease, and the only available compound pushes it upward
which way to move the mark is undetermined in this disease
depletes H2B ubiquitination indirectly, by transcriptional suppression of the ligase complex
the entire KMT2A-specific H2Bub literature is six records, two of them primary
- regulates →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
enriched across the bodies of MLL-fusion target genes and required for H3K79me there
- ← related toHOXA9 — required for survival of established MLL-rearranged leukemia, and dispensable for starting it
BULK ONLY: MOLM-13 carries more global H2BK120ub than K562 and bortezomib reduces it alongside MLL-programme transcripts. No H2BK120ub ChIP at HOXA9 exists; recorded so the claim is not later upgraded by accident.
INVERTED RELATIONSHIP: MEIS1 is a NON-HISTONE substrate of USP22, the catalytic subunit of the SAGA H2A/H2B deubiquitination module, which protects MEIS1 from proteasomal degradation in KMT2A-rearranged B-ALL. Whether the H2Bub activity is involved in the MEIS1 effect is untested.
REQUIRED FOR, NOT A WRITER OF: CDK9 activity is essential for maintaining global and gene-associated H2Bub1 levels, acting through non-histone substrates (Pol II CTD, SPT5/SUPT5H, NELF-E, probably UBE2A) and the RNF20/RNF40-WAC ligase. Human somatic cells, not a fusion system.
the Rtf1 histone modification domain stimulates Rad6/Bre1 H2B monoubiquitination; HMD point mutants lose the mark (yeast H2BK123)
- ← regulatesCTR9 — a PAF1C subunit the field assumes is a fusion partner, with no paper that tested it
CTR9 is the Rad6 (E2) dock required for PAF1C-induced stimulation of H2B monoubiquitination; confounded in cells by CTR9's scaffold role
catalytic centre of the SAGA/TFTC-STAGA DUB module, which deubiquitinates histone H2A and H2B; direction of the cellular effect differs between genetic and chemical loss
the H2BK120ub requirement cited here is a single murine MLL-AF9 line
- ← related toThe unnamed KMT2A-rearranged B-ALL line is SEM, and the deposition says KMT2A::AFF1 in a genotype field
USP22 is the SAGA H2B deubiquitinase opposing the RNF20 ligase activity resolved in the sibling entry