RNF40 (BRE1B) — the obligate partner of RNF20, and the one of the pair with a documented link to differentiation and bivalency resolution

This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.

• If the H2Bub layer is what lets a poised gene become active, suppressing it could **deepen** the differentiation block rather than relieve it - the opposite of the intended effect of any writer-directed agent in this disease.
• RNF40 is broadly required across cancer lines with no KMT2A selectivity, so the toxicity argument that applies to RNF20 applies here too.
• Panobinostat, the one clinically available agent shown to hit this axis, is a pan-HDAC inhibitor whose paediatric toxicity profile is driven by effects that have nothing to do with H2Bub.
• The H2B ligase pair is a putative tumour suppressor in the literature that characterised it (PMID 18832071), so an agent that suppresses this axis carries a transformation risk that no leukaemia paper in this beat has excluded.

1. Would raising H2Bub1 - rather than lowering it - relieve the differentiation block in KMT2A-rearranged cells? [E2] shows the mark rises during differentiation and that RNF40 is needed to resolve bivalency; every intervention this beat considered goes the other way, and the forward direction is untested in leukaemia.
2. Is RNF40's bivalency-resolution function preserved in haematopoietic progenitors? The whole inference in this entry rests on an hMSC result.
3. Why do RNF20 and RNF40 have different DepMap severities (-0.734 vs -0.490) if they are an obligate heterodimer? A partner-independent RNF20 function, or unequal guide efficiency, are the two obvious explanations and they are distinguishable.
4. RNF40 deposits H2Bub120 across transcribed gene bodies and has never been imaged at subnuclear resolution in any cell. Does it concentrate anywhere - transcription factories, speckle margins - in a KMT2A-rearranged blast, where the fusion's own output is elongation-coupled? This is a genuinely unasked question on a protein whose whole literature is 119 abstracts.

1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

   RNF20 and RNF40, both homologues of yeast Bre1, form the 600 kDa complex that is the E3 ligase for human H2B-Lys120 monoubiquitination; the complex is recruited to transcriptionally active genes and RNAi against it reduces H2Bub1 and downstream H3K4 and H3K79 methylation.

   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 2005
   PMID 16307923DOI
   ▸What this evidence cannot show

   Abstract read in full; full text not read. RNAi against 'the RNF20/40 complex' does not separate the two subunits.

   ▸Evidence that argues against this (1)

   • Depleting this ligase is reported to be transforming, not therapeutic. Shema et al. reduced RNF20, the major H2B-specific E3, in human cells: only a subset of genes responded, RNF20 SUPPRESSES several proto-oncogenes, and its depletion augmented EGF-driven transcription, increased cell migration and elicited transformation and tumorigenesis, with frequent RNF20 promoter hypermethylation observed in tumours. The pair this entry describes is therefore a putative tumour suppressor in the literature that characterised it, which is direct opposition to treating either subunit as a target. Not leukaemia and not KMT2A-rearranged, and RNF20 rather than RNF40 - but the two are an obligate heterodimer and this entry's own E1 is the paper that established that. Shema E et al. 2008
2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   Histone H2B monoubiquitination increases significantly during differentiation of human mesenchymal stem cells and various lineage-committed precursor cells across diverse organisms; the H2B ubiquitin ligase RNF40 is required for induction of differentiation markers and transcriptional reprogramming of hMSCs; this function depends on CDK9 and the WAC adaptor protein; and RNF40 is required for resolution of the H3K4me3/H3K27me3 bivalent poised state on lineage-specific genes during the transition from inactive to active chromatin.

   The histone H2B monoubiquitination regulatory pathway is required for differentiation of multipotent stem cells. · Karpiuk O, Najafova Z, Kramer F, Hennion M, Galonska C, Konig A, Snaidero N, Vogel T, Shchebet A, Begus-Nahrmann Y, Kassem M, Simons M, Shcherbata H, Beissbarth T, Johnsen SA · Mol Cell 2012
   PMID 22681891DOI
   ▸What this evidence cannot show

   Abstract read in full; full text not read. Human mesenchymal stem cells and assorted precursors - **not haematopoietic, not leukaemic, and not KMT2A-rearranged**. The transfer of a bivalency-resolution requirement from hMSC differentiation to the KMT2A-fusion differentiation block is this project's inference and is explicitly flagged as untested. One author surname is rendered without its diacritic here.

   ▸Evidence that argues against this (1)

   • The transcriptional consequence of moving H2Bub may be much smaller than a reprogramming model needs. Wang et al. found that the strong global H2Bub increase in Atxn7l3-null mouse ES cells and MEFs produced no more than modest changes in Pol II occupancy and NO change in Pol II elongation, and concluded that H2Bub deubiquitylation does not directly regulate global Pol II transcription. A differentiation-reprogramming role for the mark has to survive that null. The authors themselves note the activity regulates only a subset of genes, so this opposes a global model rather than a locus-level one. Wang F et al. 2021
3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   Panobinostat's anti-leukaemic activity in MLL-rearranged ALL involves depletion of H2B ubiquitination via suppression of the RNF20/RNF40/WAC E3 ligase complex, and knockdown of WAC phenocopies loss of H2B ubiquitination with concomitant cell death.

   The HDAC inhibitor panobinostat (LBH589) exerts in vivo anti-leukaemic activity against MLL-rearranged acute lymphoblastic leukaemia and involves the RNF20/RNF40/WAC-H2B ubiquitination axis. · Garrido Castro P, van Roon EHJ, Pinhancos SS, Trentin L, Schneider P, Kerstjens M, Te Kronnie G, Heidenreich O, Pieters R, Stam RW · Leukemia 2018
   PMID 28690313DOI
   ▸What this evidence cannot show

   Abstract read in full; full text not read. Panobinostat is a pan-HDAC inhibitor with multiple epigenetic effects, so the RNF40 arm is one contributor of unquantified weight.

   ▸Evidence that argues against this (1)

   • The selectivity this entry's clinical citation implies is absent when a comparator group is run. In this project's genome-wide DepMap scan RNF40's KMT2A-selectivity difference is +0.006, rank 9,953 of 17,728 - dead centre of the genome - on a pan-essentiality of -0.490; its partner RNF20 is +0.076 at rank 16,229 on -0.734, i.e. marginally LESS required in KMT2A-rearranged lines while being broadly essential. Panobinostat depleting H2Bub in MLL-rearranged ALL is a necessity result in one arm; it is not evidence that the axis is selectively required in this disease. Eleven KMT2A-rearranged lines, AML-weighted, none infant-derived, against a positive literature that is infant B-ALL. Broad Institute DepMap; Cellosaurus; Open Targets Platform; scan run by this project et al. 2026
4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset

   In the genome-wide DepMap KMT2A-selectivity scan of 17,728 genes, RNF40 has a KMT2A-rearranged mean gene effect of -0.520 against -0.527 in other leukaemia lines, a difference of +0.006 and a rank of 9,953 of 17,728, on a pan-essentiality of -0.490 across 1,208 lines - no selectivity in either direction.

   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 2026
   ▸What this evidence cannot show

   Eleven KMT2A-rearranged lines, AML-weighted, none infant-derived. A difference of +0.006 is indistinguishable from zero and should be read as 'no signal', not as evidence of equality.

   ▸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`. RNF40's -0.520 against the comparator is also close to the pan-essential range, so the more likely reading is a broadly required gene with little KMT2A-specific margin rather than a selective dependency the scan failed to resolve. Oncofusion knowledge graph (internal analysis) 2026
5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

   UniProt O75150 records a single subcellular location for RNF40, Nucleus, resting on one coded experimental citation (PubMed:10944455), and HPA immunofluorescence independently gives a single main location, Nucleoplasm, at reliability Supported.

   UniProtKB O75150 (BRE1B_HUMAN) SUBCELLULAR LOCATION comment, fetched 2026-09-21 · UniProt Consortium 2026
   ▸What this evidence cannot show

   One coded citation is thin for a protein with a thousand residues and a defined chromatin function. Brief rule 91: PMID 10944455 is a retinoblastoma-binding paper about 'RBP95', not a localisation study, so the annotation is a pointer rather than a source. No subnuclear resolution in either record.

   ▸Evidence that argues against this (1)

   • A bare 'Nucleus' record does not anticipate what RNF40 is actually reported to do. Loss of RNF40 decreases NF-kB activity in colorectal cancer cells and reduces colitis burden in mice - a signalling-pathway output, not a chromatin-mark output - which means the protein's functional footprint extends beyond the compartment its one-line record describes. Either RNF40 acts on NF-kB from chromatin, in which case the interesting fact is which loci, or it has a role the record does not capture at all; the record cannot distinguish them. J Crohns Colitis 2019
6. E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   RNF20 and RNF40 are required in mouse B cells for homologous recombination and for class switch recombination, and DNA double-strand breaks raise global H2Bub without raising H3K4me3 or H3K79me2.

   E3 Ubiquitin Ligases RNF20 and RNF40 Are Required for Double-Stranded Break (DSB) Repair: Evidence for Monoubiquitination of Histone H2B Lysine 120 as a Novel Axis of DSB Signaling and Repair. · Mol Cell Biol 2019
   PMID 30692271DOI
   ▸What this evidence cannot show

   Mouse B cells, no human data, no KMT2A fusion. Read at abstract level. The experiments treat RNF20 and RNF40 largely as a unit, which is biologically defensible for an obligate heterodimer but means the RNF40-specific contribution is not separated. The H3K79me2 non-response is reported for the DSB context only and should not be generalised.

   ▸Evidence that argues against this (1)

   • The attribution to RNF40 specifically is not established by this experiment, which treats RNF20 and RNF40 largely as one unit - defensible for an obligate heterodimer and inadequate for a gene-level claim. That the two are separable is shown elsewhere: RNF40 loss has its own phenotype in colorectal epithelium and colitis, where the readout is NF-kB activity rather than H2Bub-dependent repair. Until an RNF40-only allele is tested in B cells, the B-lineage record here is a record about the heterodimer. J Crohns Colitis 2019
7. E7moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

   Human Protein Atlas v25.1 assigns RNF40 the single main subcellular location Nucleoplasm at reliability Supported and records it as detected in all immune cell types with low immune cell specificity.

   Human Protein Atlas v25.1, subcellular (immunofluorescence) and Blood Atlas records for RNF40, fetched 2026-09-21 · Human Protein Atlas 2026
   ▸What this evidence cannot show

   Antibody immunofluorescence in non-haematopoietic lines. 'Nucleoplasm' in HPA's vocabulary is the absence of a resolved substructure, not a positive statement that RNF40 avoids nucleoli, speckles or bodies. Blood Atlas values are RNA.

   ▸Evidence that argues against this (1)

   • 'Nucleoplasm only' is the default output when an antibody resolves no substructure, so it must not be read as positive evidence that RNF40 avoids nucleoli, speckles or bodies - which is how a bare subcellular field will be read. The only spatial information anyone has for this protein is the genomic distribution of the mark it deposits, H2Bub120 across transcribed gene bodies, and a mark distribution is not an image. The absence recorded in this entry is therefore an absence of resolution, not a demonstration of homogeneity. Mol Cell Biol 2019

• ← related toRNF20 (BRE1A) — the H2B ubiquitin ligase that MLL-rearranged leukaemia needs, and that DepMap says every other cancer line needs just as much

  Counter-evidence (Shema 2008, PMID 18832071): reducing RNF20, the obligate partner of RNF40, augmented EGF-driven transcription, increased migration and elicited transformation and tumorigenesis in human cells, with frequent RNF20 promoter hypermethylation in tumours - the pair is a putative tumour suppressor, which opposes treating either subunit as a target

• ← related toThe differentiation block — what it actually is, operationally, in KMT2A::AFF1 leukaemia

  Counter-evidence (Wang 2021, PMID 33731875): a strong global H2Bub increase in Atxn7l3-null mouse cells produced only modest Pol II occupancy changes and no change in elongation, so the differentiation-reprogramming role this entry builds on has to survive a global transcriptional null

• ← refutesEvery gene in the H2B ubiquitination axis, ranked against the whole genome for KMT2A selectivity — and the writers are not selective

  Counter-evidence (DepMap scan, this project): RNF40's KMT2A-selectivity difference is +0.006 at rank 9,953 of 17,728 on a pan-essentiality of -0.490, and RNF20 is +0.076 at rank 16,229 on -0.734 - marginally LESS required in KMT2A-rearranged lines while broadly essential, so the panobinostat necessity result is not evidence of selective requirement

• causes →H2BK120ub (H2Bub1) — the ubiquitin tag on histone H2B that licenses H3K79 methylation, and the mark this graph was missing

  obligate heterodimer partner of RNF20 in the human BRE1 ligase

• ← bindsRNF20 (BRE1A) — the H2B ubiquitin ligase that MLL-rearranged leukaemia needs, and that DepMap says every other cancer line needs just as much

  obligate heterodimer forming the ~600 kDa human BRE1 E3 ligase

• regulates →The primed chromatin state — what is actually established, and why the word 'bivalent' should be used carefully here

  required to resolve H3K4me3/H3K27me3 bivalency during differentiation (shown in hMSCs)

• related to →Bivalent and primed chromatin at fusion target loci — and why the H3K4/H3K27 story is weaker than usually told

  the H2Bub layer is part of the machinery for leaving the poised state

• expressed in →Pre-B cell — the stage KMT2A::AFF1 leukaemia never reaches, and why that boundary is informative

  required in mouse B cells for HR and class switch recombination (PMID 30692271); the only dedicated haematopoietic RNF40 experiment

• related to →The soluble-nucleoplasm versus chromatin-bound axis - the one localisation readout that keeps a size dimension, and nobody has run it on a KMT2A fusion

  chromatin association is inferred from H2Bub120 deposition, never imaged; recorded as an absence rather than upgraded

• ← related toCDK9 - the kinase the whole SEC story runs through, which this graph held only as half of a two-protein summary entry

  RNF40-dependent H2Bub1 induction during human mesenchymal stem cell differentiation is dependent upon CDK9 and the WAC adaptor, which are required for H2B monoubiquitination - independent support for the CDK9 requirement from a different system.