Oncofusion

start / graph / gene-ezh1

Genestrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q41-prc2-arm

EZH1 - the paralogue that makes the approved EZH2 drug the wrong drug for this disease

also called EZH1 · ENX-2 · KIAA0388 · Enhancer of zeste homolog 1

EZH1 is a near-twin of EZH2 that can put the same 'switch off' tag on the same spot of the same histone. Because leukaemia cells make both, a drug that hits only EZH2 leaves the job to EZH1 and the tag largely stays - which is why the approved EZH2-only drug is predicted not to work here and a drug hitting both does.

The same thing, for a scientist

EZH1 (UniProt Q92800, 747 aa) is the alternative catalytic subunit of PRC2: a PRC2 complex contains either EZH1 or EZH2, not both, and both methylate H3K27 [E1][E5]. In KMT2A-rearranged leukaemia EZH1 is the reason the catalytic subunit is not the target. *Ezh2*-null MLL-AF9 leukaemia retains locus-specific H3K27me3 and self-renews to a third transplant, attributed to partial compensation by Ezh1 [E2]; shRNA suppression of *Ezh1/Ezh2* **together** - like suppression of *Eed* or *Suz12* - arrests proliferation and forces differentiation, with minimal impact on non-transformed haematopoietic lines [E3]. The pharmacology follows: a dual EZH1/EZH2 inhibitor suppresses cellular H3K27me3 more than an EZH2-selective one and is more effective, including in leukaemia lines carrying MLL-AF9 and MLL-AF4 [E4], and in B-ALL lines DS-3201 (valemetostat) beat EPZ-6438 (tazemetostat) and worked in an MLL-AF4 PDX [E6]. This project's own DepMap panel shows why a screen never found it: EZH1's gene effect in 11 KMT2A-rearranged lines is **-0.100** against -0.102 in 43 comparator lines - no dependency and no selectivity - while the gene is transcribed in **7/7** panel lines at median 13.7 nTPM [E7]. **A single-gene CRISPR screen is structurally blind to a redundant pair, and EZH1/EZH2 is the clearest example of that blindness in this graph.** Structurally, PRC2:EZH1 is not merely a substitute: cryo-EM shows PRC2:EZH1 forms dimers that undergo a large conformational change on nucleosome binding and compact chromatin more effectively than monomers, through a loop that diverges between EZH1 and EZH2 [E5]. So EZH1 has an activity EZH2 does not, and 'redundant' describes the methyltransferase output, not the complex.

What it looks like
loading 7KSO…
PDB 7KSO ↗
shape

This is real experimental data — atoms placed by crystallography or cryo-EM, deposited by the people who did the experiment.

I have never read one of these pictures before

A protein is one long chain of amino acids that folds up. The ribbon traces that chain from one end to the other; it is not a wire, it is a simplification of where the backbone runs. Drag to rotate, scroll to zoom.

Domains are the stretches of chain that do a particular job, and they are what the coloured blocks above the picture mark out. Click one and the view flies to it. In this project the important question about any domain is usually whether the fusion still has it — the labels say lost where it does not.

Anything in yellow sticks is a bound small molecule, which is usually the drug.

Go deeper
Why this entry exists as a separate node from gene-ezh2

Because the therapeutic conclusion inverts depending on which one you mean, and the graph must be able to state that without ambiguity.

  • EZH2-selective (tazemetostat, EPZ-6438): approved, available, and predicted by three independent lines of evidence to be ineffective in KMT2A-rearranged leukaemia 246.
  • EZH1/EZH2 dual (valemetostat/DS-3201, UNC1999 as a probe): approved in Japan for a different disease, and the class with actual activity against MLL-AF9 and MLL-AF4 systems 46.

That is a different drug class, not a different potency, and a trial that tested the wrong one and reported a negative would be recorded by anyone reading only the result as 'PRC2 is not a target in KMT2A-rearranged leukaemia'. This graph should be able to refuse that reading, which is why this entry and drug-tazemetostat both state the prediction explicitly before anyone runs the experiment.

The counted absence is the point: tazemetostat AND (KMT2A OR MLL) returns exactly 1 PubMed record, and title-checking it finds a pharmacology review (PMID 35753676), not a study. See ana-prc2-arm-enumeration for the full query set.

EZH1 is not simply a weaker EZH2

Two differences that matter if anyone tries to design around the pair:

  1. Dimerisation and chromatin compaction. Cryo-EM of monomeric and dimeric PRC2:EZH1 shows a dramatic conformational change on nucleosome binding, and that dimers compact chromatin more effectively than monomers, dependent on a loop that diverges between EZH1 and EZH2 5. A compound that blocks the SET domain does not necessarily block compaction.
  2. Expression stoichiometry is line-specific. In the KMT2A-rearranged panel, EZH2 median is 47.2 nTPM against EZH1's 13.7 - but in NOMO-1, EZH2 is 13.4 and EZH1 is 13.1, i.e. approximately equal 7. The degree to which an EZH2-selective inhibitor is rescued by EZH1 is therefore predicted to vary between lines, and any head-to-head should report the ratio per line rather than pooling.

Both of these are reasons to measure H3K27me3 depletion directly rather than to assume class equivalence from target identity.

Accession resolution (Rules 62, 63, 65)

gene_exact:EZH1 AND organism_id:9606 AND reviewed:true returns exactly one reviewed human record: Q92800, primary gene name EZH1, 747 aa, protein name Histone-lysine N-methyltransferase EZH1 8. Resolved 2026-09-20 with gene_primary and length in fields.

The length check barely works for this pair. EZH2 (Q15910) is 746 aa and EZH1 is 747 - a one-residue difference. verify-accessions.py compares the stored length to UniProt's, so it will catch a swap only because the entry asserts the exact number. Do not rely on 'the length looked about right'.

Facts and figures
symbolEZH1
uniprotQ92800
length AA747
uniprot primary gene nameEZH1
uniprot resolution notegene_exact:EZH1 AND organism_id:9606 AND reviewed:true returned exactly one row; primary gene name EZH1 matched; length 747 asserted. Paralogue EZH2 is Q15910 at 746 aa - a one-residue difference, so the accession rather than the length is the discriminator. Resolved 2026-09-20.
paraloguegene-ezh2
complexprotein-prc2-complex
mark writtenmark-h3k27me3
catalytictrue
PDB noteEntry titles fetched from the RCSB entry API on 2026-09-20 and read back: 7KSO = 'Cryo-EM structure of PRC2:EZH1-AEBP2-JARID2' (a PRC2.2 assembly), 7KSR = 'PRC2:EZH1_A from a dimeric PRC2 bound to a nucleosome'. Polymer entities were NOT fetched and no chain-to-UniProt mapping or residue span is asserted here - per the QUEEN_BRIEF structural rules, a span must be derived by aligning the fetched entity sequence to a same-day UniProt sequence, which this beat did not do. RCSB returned no primary-citation PubMed id for either entry, so the link to PMID 33514705 is by subject and deposition, not by a stored identifier.
expression hpa nTPM median kmt2a r panel13.7
therapeutic consequenceEZH2-selective inhibition is predicted to be ineffective; EZH1/EZH2 dual inhibition is the class with reported activity in MLL-AF9 and MLL-AF4 systems
subnuclear localizationNOT PUBLISHED. No subnuclear compartment has ever been assigned to EZH1 in any cell type. The related question that HAS been answered is chromatin engagement - PRC2-EZH1 compacts chromatin without SAM (PMID 19026781), binds nucleosomes as a monomer and a dimer with flexible compaction modules (this entry's E5), and stabilises Pol II preinitiation complexes in postmitotic muscle (PMID 39433902) - and none of it locates EZH1 within the nucleus. Recorded deliberately as an unpublished gap, distinct from the not-applicable case of a membrane protein such as CD34 in this same tranche.
shown for fusionNO. Measured 2026-09-21 with NCBI esearch: EZH1[tiab] AND (MLL[tiab] OR KMT2A[tiab] OR MLL-AF4[tiab] OR MLL-AF9[tiab] OR mixed lineage leukemia[tiab]) AND (localiz*[tiab] OR localis*[tiab] OR subcellular[tiab] OR nuclear body[tiab] OR nuclear bodies[tiab] OR speckle*[tiab] OR condensat*[tiab] OR nucleolar[tiab] OR immunofluorescence[tiab] OR compartment*[tiab]) returns 0. Operands: gene clause 354, gene AND fusion 11 - so an EZH1-and-KMT2A literature exists, at eleven papers, and not one of them makes any spatial claim. Positive control through the identical shape with (MLLT1[tiab] OR ENL[tiab]) returns 23. Zero in titles and abstracts; supporting figures not searched (brief rule 80).
expression per line nTPM
THP-18.600THP-1: 8.6MV4-1111.6MV4-11: 11.6NOMO-113.1NOMO-1: 13.1MOLM-1313.7MOLM-13: 13.7OCI-AML-214.8OCI-AML-2: 14.8SEM18.6SEM: 18.6RS4;1119.4RS4;11: 19.4
PDB IDS
[
  "7KSO",
  "7KSR"
]
essentiality
{
  "depmap_kmt2a_r_mean": -0.0998,
  "depmap_other_leukaemia_mean": -0.1021,
  "depmap_difference": 0.0022,
  "depmap_rank_of_17728": 9427,
  "depmap_pan_essentiality_all_lines": -0.1343,
  "depmap_index_myeloid_mean": -0.1247,
  "depmap_index_lymphoid_mean": -0.0563,
  "source": "analyses/prc2-dependency-lineage-split/results.json",
  "interpretation": "dispensable alone; the double EZH1/EZH2 perturbation is the informative one and DepMap does not contain it"
}
expression per line nTPM — the numbers behind the chart
{
  "SEM": 18.6,
  "RS4;11": 19.4,
  "MV4-11": 11.6,
  "THP-1": 8.6,
  "NOMO-1": 13.1,
  "MOLM-13": 13.7,
  "OCI-AML-2": 14.8
}
drugs
[
  "drug-valemetostat"
]
localization
{
  "subcellular": "Nucleus, and nothing else. UniProt Q92800 annotates a single location, Nucleus, at ECO:0000269 - no cytoplasmic pool, no organellar pool. Unlike its paralogue's literature, there is no imaging of EZH1 in any haematopoietic cell.",
  "subnuclear": "UNPUBLISHED - and this is the unpublished case, not the not-applicable one. EZH1 is a nuclear protein and the question is perfectly answerable; nobody has asked it. What IS established is chromatin ENGAGEMENT rather than a compartment: PRC2-Ezh1 compacts chromatin on reconstituted templates even without SAM, cryo-EM shows monomeric and dimeric PRC2:EZH1 on a nucleosome with flexible modules involved in chromatin compaction, and in postmitotic muscle PRC2-EZH1 stabilises Pol II preinitiation complexes. None of that is a subnuclear address. No polycomb-body, chromocentre, speckle or condensate assignment exists for EZH1 anywhere.",
  "how_shown": "electron microscopy of reconstituted chromatinised templates with and without SAM, plus mouse tissue expression comparison (PMID 19026781); cryo-EM of PRC2:EZH1 monomer and dimer on nucleosome (this entry's E5); nascent transcription and PIC-stability assays in postmitotic mouse skeletal muscle (PMID 39433902); single-molecule and epigenomic analysis of the EZH1 Q571R compaction mutant (PMID 42398503); UniProt Q92800 SUBCELLULAR LOCATION comment.",
  "systems": "in vitro reconstitution, mouse kidney and skeletal muscle, thyroid cancer models; no leukaemia and no KMT2A-rearranged cell has ever been used to localise EZH1",
  "node": "localization-polycomb-bodies-kmt2a-r"
}
cell type
{
  "expressed_in": "NON-PROLIFERATIVE adult tissue preferentially - Ezh1 is more abundant in non-proliferative adult organs while Ezh2 expression is tightly associated with proliferation [E_q66_loc1]. In the leukaemia panel this entry already measures EZH1 by HPA across KMT2A-rearranged lines, so expression is present but not restricted. The expression pattern is the single most important thing about EZH1 for this disease, and it points AWAY from the cycling blast.",
  "required_in": "Not established anywhere in leukaemia as an EZH1-specific requirement. The entry's own lineage split gives EZH1 a gene effect of -0.100 in 11 KMT2A-rearranged lines against -0.102 in 43 other leukaemia lines [E7] - no selectivity and essentially no effect. Every functional result attributed to EZH1 in this disease comes from a DUAL EZH1/2 inhibitor: UNC1999 with 5-azacytidine in KMT2A::MLLT3 iPSC-derived AML-HSPCs [E_q66_cell1], DS-3201/valemetostat in B-ALL lines [E6], and the dual inhibitors of [E1] and [E4].",
  "required_in_context": "Postmitotic skeletal muscle, for circadian gene expression [E_q66_loc2] - the only cell type with a demonstrated EZH1-specific requirement, and it is terminally differentiated.",
  "how_shown": "mouse tissue expression comparison in ageing kidney (PMID 19026781); iPSC-derived HSPC differentiation from a KMT2A::MLLT3 patient with ChIP-seq and dual EZH1/2 inhibition (PMID 41438598); Ezh1 perturbation in postmitotic mouse skeletal muscle (PMID 39433902); genome-scale CRISPR fitness with a lineage split (this entry's E7)",
  "nodes": [
    "cell-type-hsc-bone-marrow",
    "cell-type-pro-b"
  ]
}
cell state
{
  "maintains_state": "REPRESSION AT PRC2 TARGETS, in a KMT2A::MLLT3 model - and only as half of a pair. A significant subset of genes downregulated in KMT2A::MLLT3 iPSC-derived AML-HSPCs are direct PRC2 targets, and dual EZH1/2 inhibition with 5-azacytidine reactivates them specifically in the AML-HSPCs toward normal expression [E_q66_cell1]. The state is PRC2-maintained; the EZH1 share of it is unmeasured.",
  "expressed_in_state": "QUIESCENCE and terminal differentiation, by paralogue division of labour: Ezh1 is the non-proliferative-tissue paralogue and Ezh2 the proliferative one [E_q66_loc1]. That is a direct prediction for this disease - EZH1 should matter most in the quiescent, therapy-surviving fraction and least in the cycling bulk - and it has never been tested in a leukaemia.",
  "regulates_state": "Bidirectional transcriptional control in postmitotic cells: PRC2-EZH1 both represses through chromatin state and ACTIVATES by stabilising Pol II preinitiation complexes [E_q66_loc2]. An EZH1 inhibitor is therefore not unambiguously a de-repressor.",
  "how_shown": "dual EZH1/2 inhibition plus 5-azacytidine with ChIP-seq and expression profiling in isogenic normal versus AML iPSC-derived HSPCs (PMID 41438598); nascent transcription and PIC stability in postmitotic muscle (PMID 39433902); tissue expression comparison (PMID 19026781)",
  "nodes": [
    "cell-state-primed-bivalent-chromatin",
    "cell-state-quiescence",
    "cell-state-differentiation-block"
  ]
}
How this could bite us back
  • The redundancy cuts both ways. A dual EZH1/EZH2 inhibitor removes H3K27 methylation in normal haematopoietic and B-lineage compartments as completely as in the leukaemia, and EZH2 loss is the documented tumour-suppressor lesion of myeloid neoplasia - so the more complete the inhibition, the closer the pharmacology sits to a state that predisposes to a second malignancy in a survivor with decades of life ahead.
  • If a trial of an EZH2-selective agent is run in this disease and reports a negative, the negative will be attributed to the biology rather than to EZH1. Any such trial must include a pharmacodynamic H3K27me3 measurement, or its result will be uninterpretable in exactly the way the pinometostat trial's was.
  • PRC2:EZH1 has a catalysis-independent chromatin-compaction activity, so a SET-domain inhibitor is not a clean loss-of-function reagent for EZH1 and a negative inhibitor result does not exclude an EZH1 dependency.
  • An EZH1 inhibitor is not unambiguously a de-repressor: in postmitotic cells PRC2-EZH1 regulates transcription both negatively AND positively, activating by stabilising Pol II preinitiation complexes (PMID 39433902). Inhibiting it could therefore lower expression of genes a de-repression rationale predicts it would raise, and no leukaemia study measures nascent transcription under a dual EZH1/2 inhibitor.
  • The catalytic site may be the wrong handle. PRC2-Ezh1 methylates weakly and its distinctive activity - direct repression and chromatin COMPACTION - works in the ABSENCE of the SAM cofactor (PMID 19026781). A SAM-competitive inhibitor such as valemetostat or UNC1999 would not touch that activity, so 'dual EZH1/2 inhibition' may be dual only for the catalytic arm.
  • Expression works against the indication: Ezh1 is abundant in non-proliferative adult tissue and Ezh2 tracks proliferation (PMID 19026781). In a rapidly cycling blast EZH1 is the wrong paralogue to be limiting, and this entry's DepMap row agrees - -0.100 in KMT2A-rearranged lines against -0.102 in other leukaemia lines, no difference. The cost of the dual agent is therefore paid in non-proliferative normal tissue, where EZH1 is most abundant.
  • EZH1 has an EZH1-SPECIFIC oncogenic mechanism that a class-wide inhibitor could not be assumed to reverse: the thyroid EZH1 Q571R mutation enhances chromatin compaction and pushes PRC2 into H3K36me2-marked chromatin, and the equivalent EZH2 Q570R does NOT compact (PMID 42398503). Reasoning about EZH1 from EZH2 is unsafe in both directions.
Still open
  1. Has anyone measured the EZH1:EZH2 protein ratio, or the fraction of PRC2 complexes containing each, in primary infant KMT2A::AFF1 blasts? Every statement about redundancy here is from mouse myeloid genetics or from cell-line RNA; the ratio is what determines how much residual H3K27me3 an EZH2-selective inhibitor would leave in the index disease.
  2. Is the compensation locus-specific in the same way in B-lineage cells? PMID 22396593 reports locus-specific persistence of H3K27me3 in Ezh2-null MLL-AF9 cells; if the compensated loci are the fusion's targets, a dual inhibitor is required, and if they are not, an EZH2-selective agent might still reach the relevant loci.
  3. Does PRC2:EZH1 dimerisation and chromatin compaction (PMID 33514705) contribute anything in leukaemia, and would a SET-domain inhibitor leave it intact? If compaction is catalysis-independent, the degrader-versus-inhibitor question that this graph already asks for DOT1L applies to EZH1 as well.
  4. Is EZH1 the paralogue that matters in the QUIESCENT fraction of KMT2A-rearranged blasts? The tissue-expression division of labour predicts exactly that, this graph invokes quiescence to explain measurable residual disease, and no one has measured EZH1 in a sorted quiescent leukaemic population.
  5. Can any experiment separate the EZH1 from the EZH2 contribution in this disease? Every functional result in this entry uses a dual inhibitor, and the DepMap signal for EZH1 alone is flat - so the entry's central claim, that EZH1 redundancy makes the EZH2-selective drug the wrong drug, currently rests on the absence of an EZH1 effect rather than on its presence.
  6. Where is EZH1 in a nucleus? No subnuclear assignment exists for EZH1 in any cell type. Given that BMI1 recruitment to polycomb bodies requires PRC2, the reciprocal question - whether EZH1-containing PRC2 is in those bodies at all - is unasked.
  7. Does the SAM-independent compaction activity operate in leukaemia? If it does, a catalytic inhibitor leaves the mechanism intact and the correct assay is chromatin accessibility, not H3K27me3.
Every source behind this entry (11)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    EZH2 or its close homolog EZH1 functions as the catalytic subunit of PRC2, so there are two PRC2 complexes containing either EZH2 or EZH1, and little was known about the function of EZH1 in tumorigenesis before dual inhibitors were developed.

    Novel orally bioavailable EZH1/2 dual inhibitors with greater antitumor efficacy than an EZH2 selective inhibitor. · Honma D, Kanno O, Watanabe J, Kinoshita J, Hirasawa M, Nosaka E, et al. · Cancer Sci 2017
    PMID 28741798DOI
    What this evidence cannot show

    Abstract read from the eutils record; full text not read. This is the sponsor's own paper for the dual-inhibitor programme, so it is cited here for the structural fact about complex composition rather than for efficacy comparisons.

    Evidence that argues against this (1)
    • 'Either EZH1 or EZH2' is a statement about complex composition, not about functional equivalence. Cryo-EM shows PRC2:EZH1 dimerises and compacts chromatin through a loop that diverges between the paralogues, so the two complexes are not interchangeable in every activity. Grau D et al. 2021
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Genome-wide analysis of PRC2-mediated H3K27 trimethylation in Ezh2-null MLL-AF9 leukaemia demonstrated locus-specific persistence of H3K27me3 despite inactivation of Ezh2, suggesting partial compensation by Ezh1, and Ezh2-null leukaemias maintained self-renewal up to the third round of transplantation.

    Polycomb repressive complex 2 is required for MLL-AF9 leukemia. · Neff T, Sinha AU, Kluk MJ, Zhu N, Khattab MH, Stein L, et al. · Proc Natl Acad Sci U S A 2012
    PMID 22396593DOI
    What this evidence cannot show

    Abstract read from the eutils record; full text not read. Compensation by Ezh1 is an inference from residual locus-specific H3K27me3, not from an Ezh1 deletion or an Ezh1 ChIP - the paper's own wording is 'suggesting'. Mouse MLL-AF9, myeloid, no B-lineage or infant arm. This entry should not upgrade 'suggesting partial compensation' to 'EZH1 deposits the residual mark' without a direct measurement.

    Evidence that argues against this (1)
    • Residual locus-specific H3K27me3 after Ezh2 loss has an alternative explanation that this experiment does not exclude: the mark is chemically stable and is diluted by replication rather than actively removed, so a pre-existing H3K27me3 pattern can persist for several divisions without any methyltransferase depositing it. Distinguishing persistence from re-deposition needs a pulse-chase or a calibrated timecourse, not an endpoint ChIP. Danis E et al. 2015
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In an shRNA screen in MLL-AF9;Nras-G12D acute myeloid leukaemia, suppression of the PRC2 components Eed, Suz12 or Ezh1/Ezh2 caused proliferation arrest and differentiation of leukaemia cells with minimal impact on growth of several non-transformed haematopoietic cell lines, partly through direct PRC2 repression of self-renewal-limiting genes including Cdkn2a.

    The Polycomb complex PRC2 supports aberrant self-renewal in a mouse model of MLL-AF9;Nras(G12D) acute myeloid leukemia. · Shi J, Wang E, Zuber J, Rappaport A, Taylor M, Johns C, et al. · Oncogene 2013
    PMID 22469984DOI
    What this evidence cannot show

    Abstract read from the eutils record; full text not read. shRNA knockdown, so residual protein is present and the comparison between single and combined Ezh1/Ezh2 suppression is not resolved in the abstract - 'Ezh1/Ezh2' is written as a pair and the reader cannot tell from the abstract whether single-gene arms were run. The model carries a co-operating Nras-G12D mutation, which is not a feature of most infant KMT2A::AFF1 disease. Mouse, myeloid.

    Evidence that argues against this (1)
    • The therapeutic window claimed here - leukaemia arrests, non-transformed haematopoietic lines are spared - is measured in immortalised non-transformed LINES, not in primary HSPCs and not in vivo. EZH2 is required for normal early B-cell development and Igh rearrangement in mice, so in a B-lineage disease the spared compartment in this assay is not the compartment at risk. Su IH et al. 2003
  4. E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    EZH1/2 dual inhibitors suppressed trimethylation of H3K27 in cells more than EZH2-selective inhibitors and showed greater antitumour efficacy in vitro and in vivo, and a haematological cancer panel assay showed efficacy against leukaemia with fusion genes including MLL-AF9, MLL-AF4 and AML1-ETO.

    Novel orally bioavailable EZH1/2 dual inhibitors with greater antitumor efficacy than an EZH2 selective inhibitor. · Honma D, Kanno O, Watanabe J, Kinoshita J, Hirasawa M, Nosaka E, et al. · Cancer Sci 2017
    PMID 28741798DOI
    What this evidence cannot show

    Abstract read from the eutils record; full text not read. Sponsor-authored; a panel assay with no n, no potency values and no fusion-versus-non-fusion comparison in the abstract. The headline efficacy demonstration is EZH2-mutant DLBCL, not leukaemia. 'More than EZH2-selective inhibitors' is a comparison of cellular H3K27me3 depletion whose dose-matching is not stated in the abstract.

    Evidence that argues against this (1)
    • Greater H3K27me3 depletion is not automatically a better therapeutic outcome, because the same depletion in normal B cells raises H3K27ac and increases chromosomal translocation frequency where AID activity is high or DNA repair is deficient. A more complete inhibitor is also a more complete exposure of that hazard. Tao J et al. 2025
  5. E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Cryo-EM structures of PRC2:EZH1 as a monomer and as a dimer bound to a nucleosome show a dramatic conformational change on nucleosome binding; mutation of a divergent EZH1/EZH2 loop abrogates nucleosome binding and methyltransferase activity of PRC2:EZH1, and dimers are more effective than monomers at promoting chromatin compaction, a function that also requires the divergent loop.

    Structures of monomeric and dimeric PRC2:EZH1 reveal flexible modules involved in chromatin compaction. · Grau D, Zhang Y, Lee CH, Valencia-Sánchez M, Zhang J, Wang M, et al. · Nat Commun 2021
    PMID 33514705DOI
    What this evidence cannot show

    Abstract read from the eutils record; full text not read. Reconstituted complexes on defined nucleosome substrates in vitro - the dimer's existence and stoichiometry in a leukaemia cell are not established by this work, and the authors themselves state a speculation about why flexibility and dimerisation matter in cells. Residue numbering for the divergent loop is not taken from this abstract and is deliberately not asserted here (see the QUEEN_BRIEF warning that residue indices read straight off a deposit are often wrong).

    Evidence that argues against this (1)
    • Chromatin compaction measured on reconstituted arrays is exactly the kind of observation the condensate-assay caution applies to: compaction in vitro and clustered binding in cells are different claims, and neither 1,6-hexanediol sensitivity nor classical FRAP can tell a phase-separated compartment from binding to spatially clustered sites - and 1,6-hexanediol itself rapidly condenses chromatin in living cells. Itoh Y et al. 2021
  6. E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    The EZH1/EZH2 dual inhibitor DS-3201 (valemetostat) inhibited growth of B-cell ALL cell lines more significantly and strongly than the EZH2-specific inhibitor EPZ-6438 (tazemetostat), and significantly inhibited growth of a B-ALL patient-derived xenograft harbouring MLL-AF4.

    Dual inhibition of EZH1/2 induces cell cycle arrest of B cell acute lymphoblastic leukemia cells through upregulation of CDKN1C and TP53INP1. · Ito J, Yamagata K, Shinohara H, Shima Y, Katsumoto T, Aikawa Y, et al. · Int J Hematol 2023
    PMID 36280659DOI
    What this evidence cannot show

    Abstract read from the eutils record; full text not read. The head-to-head is a cell-line growth comparison; without matched H3K27me3 depletion between arms it cannot separate 'dual target coverage' from 'more potent compound'. One MLL-AF4 PDX, with no n, dose or patient age in the abstract, and no infant arm.

    Evidence that argues against this (1)
    • The mechanism the paper establishes - CDKN1C and TP53INP1 de-repression - was confirmed in B-ALL lines generally rather than in the MLL-AF4 PDX, so the PDX result and the mechanism are two separate claims from the same paper and the second does not automatically apply to the first. Ito J et al. 2023
  7. E7moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    EZH1 gene effect in 11 KMT2A-rearranged DepMap leukaemia lines is -0.100 against -0.102 in 43 other leukaemia lines (difference +0.002, rank 9,427 of 17,728, pan-essentiality -0.134), while EZH1 is transcribed in 7 of 7 Human Protein Atlas panel lines at median 13.7 nTPM - so the absence of dependency is redundancy with EZH2 rather than absence of the transcript.

    What this evidence cannot show

    A single-gene knockout screen cannot detect a redundant pair by construction, so this number is evidence that EZH1 alone is dispensable and is not evidence about PRC2 catalysis. 11 index lines, none infant. HPA covers 7 of them. nTPM is bulk RNA, not protein, and does not report the EZH1:EZH2 protein ratio or which complex predominates. Raw effects, no z, no permutation null.

    Evidence that argues against this (1)
    • The expression control is a rebuttal to one artefact and not a proof of function. A transcribed methyltransferase can still be excluded from the relevant loci - the compensation reported in Ezh2-null leukaemia was LOCUS-SPECIFIC, not global, so EZH1 may cover only a subset of the PRC2 landscape and the residual pattern after an EZH2-selective inhibitor need not resemble the untreated one. Neff T et al. 2012
  8. E8established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    EZH1 resolves to a single reviewed human UniProt record, Q92800 (EZH1_HUMAN), primary gene name EZH1, 747 amino acids, protein name Histone-lysine N-methyltransferase EZH1; EZH2 is Q15910 at 746 amino acids.

    What this evidence cannot show

    Resolved 2026-09-20. The one-residue length difference from EZH2 means the length check prescribed by Rule 65 is technically valid but has no margin; the accession is the load-bearing identifier.

    Evidence that argues against this (1)
    • This graph has already logged two cases where a paralogue was returned in place of the gene searched (CSNK2A1/CSNK2A2, HDAC7/HDAC9), and in both the wrong answer looked right. EZH1 and EZH2 do not currently alias each other in UniProt, but both carry the ambiguous historical name 'enhancer of zeste homolog' in free text, so literature-derived attributions are less safe than the database lookup. UniProt 2026
  9. E_q66_loc1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    EZH1's relationship to chromatin is PHYSICAL rather than catalytic, which makes its localisation the functional variable: PRC2-Ezh1 methylates H3K27 only weakly and Ezh1 knockdown does not change global H3K27me2/3, yet PRC2-Ezh1 directly and robustly represses transcription from chromatinised templates and COMPACTS CHROMATIN in the ABSENCE of the methyltransferase cofactor SAM, shown by electron microscopy; Ezh1 targets a subset of Ezh2 genes and is more abundant in NON-PROLIFERATIVE adult organs while Ezh2 expression tracks proliferation.

    Ezh1 and Ezh2 maintain repressive chromatin through different mechanisms. · Margueron R; Li G; Sarma K; Blais A; Zavadil J; Woodcock CL; Dynlacht BD; Reinberg D · Mol Cell 2008
    PMID 19026781DOI
    What this evidence cannot show

    In vitro chromatinised templates with electron microscopy, plus mouse tissue expression comparison in ageing kidney - not haematopoietic, not leukaemic, and not human. The compaction result is a reconstitution, so it shows PRC2-Ezh1 CAN compact chromatin, not that it does so at any particular locus in a cell. The proliferation-versus-quiescence expression contrast is the part most relevant to this graph and is the part measured least directly. Read: abstract only. Identifiers copied field by field from the efetch record fetched 2026-09-21, ArticleId anchored per brief rule 81.

    Evidence that argues against this (1)
  10. E_q66_loc2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In postmitotic cells, PRC2-EZH1 is a chromatin-state and Pol II-complex regulator rather than only a silencer: in postmitotic skeletal muscle cells the circadian master regulator BMAL1 directly controls oscillatory assembly of core PRC2-EZH1 components, and EZH1 is essential for circadian gene expression at alternate Zeitgeber times through STABILISATION OF RNA POLYMERASE II PREINITIATION COMPLEXES, controlling nascent transcription - so PRC2-EZH1 regulates transcription both negatively and positively by modulating chromatin states and basal transcription complex stability.

    PRC2-EZH1 contributes to circadian gene expression by orchestrating chromatin states and RNA polymerase II complex stability. · Liu P; Nadeef S; Serag MF; Paytuví-Gallart A; Abadi M; Della Valle F; Radió S; Roda X; Dilmé Capó J; Adroub S; Hosny El Said N; Fallatah B; Celii M; Messa GM; Wang M; Li M; Tognini P; Aguilar-Arnal L; Habuchi S; Masri S; Sassone-Corsi P; Orlando V · EMBO J 2024
    PMID 39433902DOI
    What this evidence cannot show

    Mouse postmitotic skeletal muscle in a circadian-biology setting: not haematopoietic, not leukaemic, and the cell is terminally differentiated, which is the opposite of a blast. The positive-regulation claim - PIC stabilisation - is the novel part and rests on nascent-transcription and complex-stability assays rather than on a direct structural readout. Nothing here licenses a subnuclear COMPARTMENT assignment; it is a chromatin-engagement result. Read: abstract only.

    Evidence that argues against this (1)
    • The property that makes EZH1 interesting in postmitotic muscle is exactly the one that would make it irrelevant in a blast: Ezh1 is more abundant in NON-PROLIFERATIVE adult organs while Ezh2 tracks proliferation. A rapidly cycling leukaemic cell is the tissue context in which EZH1 should matter least, and this entry's own DepMap row is consistent with that - EZH1 gene effect is -0.100 in the 11 KMT2A-rearranged lines against -0.102 in 43 other leukaemia lines, no difference at all. Margueron R; Li G; Sarma K; Blais A; Zavadil J; Woodcock CL; Dynlacht BD; Reinberg D 2008
  11. E_q66_cell1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In a patient-derived KMT2A::MLLT3 iPSC model of AML, PRC2 targets are the repressed axis and a DUAL EZH1/2 inhibitor reopens them cell-type-specifically: analysis of iPSCs derived from a patient with KMT2A::MLLT3 AML identified key transcriptional activators and repressors of the altered regulatory landscape, ChIP-seq indicated that a significant subset of genes downregulated in AML iPSC-derived haematopoietic stem and progenitor cells are direct PRC2 targets, and treatment with the dual EZH1/2 inhibitor UNC1999 plus 5-azacytidine reactivated these PRC2 target genes SPECIFICALLY IN AML-HSPCs, toward normal expression patterns.

    Targeting dysregulated epigenetic and transcription factor networks in KMT2A-rearranged AML using iPSC models. · Palau A; Thier J; Naughton A; Tae-Jun Kwon A; Cabrerizo Granados D; Hofmann S; Kaczkowski B; Zhong X; Lehmann S; Arner E; Lundin V; Lennartsson A · Blood Neoplasia 2026
    PMID 41438598DOI
    What this evidence cannot show

    An iPSC-derived model of KMT2A::MLLT3 AML - the right fusion family but the myeloid partner and a reprogrammed system, not primary infant KMT2A::AFF1 B-lineage material; claim written with the system first per brief rule 73. UNC1999 is a DUAL EZH1/EZH2 inhibitor used with 5-azacytidine, so neither the EZH1 contribution nor the single-agent effect is resolved - this is the recurring problem with every EZH1 claim in this entry. 'Specifically in AML-HSPCs' means relative to the isogenic normal comparator in the same experiment, which is the right control. Read: abstract only.

    Evidence that argues against this (1)
    • No experiment anywhere separates the EZH1 contribution from the EZH2 contribution in this disease, and the dependency data say there is nothing to separate: this entry's own DepMap analysis gives EZH1 a gene effect of -0.100 in 11 KMT2A-rearranged lines against -0.102 in 43 other leukaemia lines. A dual inhibitor plus a hypomethylating agent reactivating PRC2 targets is therefore evidence about PRC2, not about EZH1, and the entry's title - that EZH1 makes the approved EZH2 drug the wrong drug - rests on the paralogue's REDUNDANCY rather than on any EZH1-specific requirement. Oncofusion knowledge graph; DepMap Chronos via Open Targets 2026
Connected to