Oncofusion

start / graph / gene-pcgf1

Proteinmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q5-interactome4

PCGF1 — the PRC1.1 scaffold whose LOSS causes menin-inhibitor resistance, so it is a restraint on the leukaemia rather than a dependency of it

also called PCGF1 · NSPC1 · RNF68 · Polycomb group RING finger protein 1 · Nervous system Polycomb-1

PCGF1 is the central strut of a small protein machine (PRC1.1) that keeps certain genes switched off in leukaemia cells. Removing PCGF1 does not kill the leukaemia — it lets MYC come back on and makes the cells resistant to menin inhibitors, which is the opposite of what a drug target looks like.

The same thing, for a scientist

PCGF1 (Q9BSM1, 259 aa) is the PCGF-family RING-finger subunit that defines non-canonical PRC1.1 together with RING1A/B, KDM2B, BCOR or BCORL1, SKP1 and RYBP/YAF2 [E1]. In KMT2A-rearranged AML a group of **noncanonical menin targets** is co-occupied by menin and repressive H2AK119ub and is normally down-regulated by menin inhibition; loss of PCGF1 or BCOR epigenetically reactivates those targets, including **MYC**, and confers **resistance** to menin inhibitors [E2]. An independent bivalency screen put **PCGF1-PRC1.1** alongside MTF2-PRC2.1 and menin-KMT2A/B as the three complexes maintaining bivalent promoters, and found menin loss phenocopies polycomb disruption [E3]. A 2026 Perturb-seq map of the same disease reports that disrupting PCGF1 — named as an antagonistic regulator of the myeloid differentiation programme — confers **strong resistance to DOT1L inhibition** [E4]. DepMap agrees on the sign: across 17,728 genes PCGF1 ranks **17,405 (98.18th percentile)** for KMT2A-rearranged selectivity with a difference of **+0.145**, i.e. rearranged lines need PCGF1 *less* than other leukaemia lines [E5]. Structurally the PCGF1 RAWUL domain (aa 167-255) is the discriminating surface: it binds the BCOR/BCORL1 PUFD domain and not the BMI1/MEL18 partners, which is why PRC1.1 is a distinct complex rather than a variant of canonical PRC1 [E6].

What it looks like
Polycomb group RING finger protein 1259 aa
RING (RING-type zinc finger)RING (RING-type zinc finger) · residues 47–86RAWUL (RING finger- and WD40-associated ubiquitin-like domain)RAWUL (RING finger- and WD40-associated ubiquitin-like domain) · residues 167–2550100200
loading AF-Q9BSM1-F1…
AlphaFold AF-Q9BSM1-F1 ↗
colourshape

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.

very high (90+)confident (70–90)low (50–70)very low (<50)
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
What PCGF1 actually is

Mammals have six PCGF paralogues (PCGF1-6). Each nucleates a different PRC1 assembly around a RING1A/B catalytic subunit, and the six complexes differ in genomic localisation; only a small subset co-localises with H3K27me3 1. PCGF1 nucleates PRC1.1, the CpG-island-targeted variant, whose DNA-binding subunit is KDM2B and whose corepressor arm is BCOR or BCORL1 16.

The part that matters here is that PRC1.1 does not behave like textbook Polycomb. In primary human AML CD34+ cells, PRC1.1 binds a distinct set of genes that are devoid of H3K27me3 and transcriptionally active, including metabolic genes, and PRC1.1 components are significantly over-expressed relative to their normal counterparts 7. Whatever PRC1.1 is doing in leukaemia, it is not simply PRC2-dependent silencing.

UniProt annotates the PCGF1 architecture as a RING-type zinc finger at aa 47-86 and a RAWUL domain at aa 167-255 that is sufficient for the BCOR/BCORL1 interaction; aa 150-255 is annotated as required for interaction with the KDM2B-SKP1 heterodimer 10.

Why this entry says 'restraint', not 'dependency', and why that is the interesting claim

Three lines point the same way.

Resistance genetics. In Zhou et al. 2024, chromatin-focused CRISPR screens in human and mouse KMT2A-rearranged AML identified loss of PRC1.1 subunits — PCGF1 or BCOR — as a menin-inhibitor resistance mechanism, acting by de-repressing a class of noncanonical menin targets that carry both active menin and repressive H2AK119ub. MYC is among them, and MYC inhibition resensitises 2.

A second drug class, same direction. A 2026 Perturb-seq map of KMT2A-rearranged AML reports PCGF1 disruption confers strong resistance to DOT1L inhibition 4. Two different epigenetic drugs, one direction: less PRC1.1 means more resistance. (These two papers share senior authors, so this is one group's programme replicated across perturbation types, not two labs agreeing — see the caveat on 4.)

Unbiased dependency data. PCGF1 sits at the 98th percentile of a genome-wide KMT2A-selectivity scan, on the anti-dependency side, and so do the other PRC1.1 members the graph carries — BCOR +0.102 (95.5th), RNF2 +0.096 (94.9th), BCORL1 +0.061 (87.5th) 5. The whole module points one way, which is harder to dismiss as noise than any single gene would be.

The practical consequence: PCGF1 is not a target, it is a biomarker and a resistance gene. A KMT2A-rearranged patient whose leukaemia has lost PRC1.1 function is predicted to respond poorly to a menin inhibitor and to a DOT1L inhibitor, and to be susceptible to BCL2 inhibition instead 2.

Cell type, cell state, localisation and epigenetic readout
AxisWhat is documentedWhere
Cell typeHuman and mouse KMT2A-rearranged AML models and lines; primary AML CD34+ cells27
Cell stateLoss of PRC1.1 reduces monocytic gene signatures; PRC1.1 antagonises the myeloid differentiation programme24
SubcellularNuclear, chromatin-associated; PRC1.1 is recruited to DNA by the KDM2B ZF-CxxC domain17
SubnuclearNot established for PCGF1 in any KMT2A-rearranged cell. No speckle, Cajal-body or condensate assignment is published. Recorded here as a gap, not as diffuse nucleoplasmic by default.—
Epigenetic readoutH2AK119ub, deposited by the RING1A/B catalytic subunit; PRC1.1 targets carry high H2AK119ub and H3K4me3 but low H3K27me318
Chromatin targetingNon-methylated CpG islands, via the KDM2B ZF-CxxC domain78

No published work assigns PCGF1 a B-lineage phenotype in KMT2A::AFF1 infant ALL. Every functional result above is AML or T-ALL. That is the single largest caveat on this entry for this project's disease.

Is PCGF1 an interactor of KMT2A::AFF1?

No published experiment shows PCGF1 binding KMT2A::AFF1 or any KMT2A fusion protein. The documented relationship is genetic and chromatin-level: menin/KMT2A and PRC1.1 co-occupy or compete at a shared set of promoters 23. Classified here as chromatin antagonist, not interactome member, on the discipline this beat applies elsewhere — co-occupancy on chromatin is not a protein-protein interaction.

The nearest physical link runs through a fusion partner rather than through PCGF1: ATM-phosphorylated ENL (MLLT1) binds the PRC1 E3-ligase module via BMI1 9. BMI1 is PCGF4, not PCGF1, and BCOR PUFD explicitly does not bind BMI1 6, so that link belongs to canonical PRC1 and does not transfer to PRC1.1.

Facts and figures
uniprotQ9BSM1
gene symbolPCGF1
protein namePolycomb group RING finger protein 1
length AA259
alphafold IDAF-Q9BSM1-F1
accession resolutionUniProt REST query gene_exact:PCGF1 AND organism_id:9606 AND reviewed:true, run 2026-09-19; one reviewed hit; primary gene name read back as PCGF1 and length 259 confirmed from the same fetched record, not from recall.
complexPRC1.1 (non-canonical / variant PRC1)
classificationchromatin antagonist and resistance gene; NOT a demonstrated interactor of KMT2A::AFF1
classification reasonno published co-IP, proximity-labelling or structural evidence of PCGF1 contacting any KMT2A fusion protein; the documented relationship is promoter-level antagonism plus a genetic resistance phenotype
interaction evidence classCRISPR screens; ChIP-seq co-occupancy; Perturb-seq; crystallography of PCGF1-BCOR
shown for fusionKMT2A-rearranged AML, human and mouse; the specific fusion is not always stated and no result is in KMT2A::AFF1 B-ALL
epigenetic readoutH2AK119ub deposited by RING1A/RING1B; PRC1.1 target promoters are H3K4me3-high and H3K27me3-low
clinical relevancepredicted biomarker of menin-inhibitor and DOT1L-inhibitor resistance; PRC1.1-deficient cells reported susceptible to BCL2 inhibition
aliases retired or alternate
[
  "NSPC1",
  "RNF68"
]
complex partners
[
  "gene-rnf2",
  "gene-kdm2b",
  "gene-bcor",
  "gene-bcorl1",
  "gene-skp1",
  "gene-rybp"
]
domains
[
  {
    "name": "RING (RING-type zinc finger)",
    "start": 47,
    "end": 86,
    "source": "UniProt Q9BSM1, fetched 2026-09-19"
  },
  {
    "name": "RAWUL (RING finger- and WD40-associated ubiquitin-like domain)",
    "start": 167,
    "end": 255,
    "source": "UniProt Q9BSM1, fetched 2026-09-19; the BCOR/BCORL1 PUFD-binding surface of PMID 23523425"
  }
]
localization
{
  "subcellular": "nucleus, chromatin-associated",
  "subnuclear": "not established in any KMT2A-rearranged system",
  "chromatin_targeting": "non-methylated CpG islands via the KDM2B ZF-CxxC domain"
}
essentiality
{
  "depmap_scan": "analyses/depmap-kmt2a-selective-scan-full",
  "pan_essentiality_all_lines": -0.0384,
  "n_all_lines": 1208,
  "mean_kmt2a_rearranged": 0.0593,
  "mean_other_leukemia": -0.0855,
  "kmt2a_selectivity_difference": 0.1448,
  "rank": 17405,
  "n_scanned": 17728,
  "percentile": 98.178,
  "interpretation": "anti-dependency direction; not a proliferation target"
}
depmap kmt2a selectivity
{
  "chronos_mean_kmt2a_rearranged": 0.0593,
  "chronos_mean_other_leukaemia": -0.0855,
  "difference": 0.1448,
  "pan_essentiality_all_lines": -0.0384,
  "selectivity_rank": 17405,
  "rank_denominator": 17728,
  "provenance": "Joined 2026-09-21 from analyses/depmap-kmt2a-selective-scan-full (generated 2026-09-19T10:12:06Z), the project's own genome-wide scan over 11 KMT2A-rearranged and 43 other leukaemia DepMap lines. NOT re-fetched and NOT independent of that scan: every caveat it carries applies here, including that the KMT2A-rearranged arm is a small number of lines drawn from nine individuals, that the rank denominator is 17,728 genes, and that a difference is not a dependency. Added because only 26 of 308 gene entries carried this number, which made any KMT2A-selectivity question answerable for 8% of the corpus.",
  "how_to_read": "A negative difference means MORE required in KMT2A-rearranged lines than in other leukaemia lines. A rank near 17,728 is ANTI-selective. Neither is evidence of a fusion-driven dependency on its own: check pan_essentiality_all_lines before calling anything selective, and note that a paralogue-redundant gene can score near zero for reasons that have nothing to do with this disease - see ana-the-paralogue-redundancy-excuse-is-not-tested-and-my-proxy-decided-the-answer."
}
How this could bite us back
  • Anything that disrupts PRC1.1 would, on this evidence, be expected to REDUCE menin-inhibitor and DOT1L-inhibitor efficacy rather than increase it. A combination partner chosen because 'it is a polycomb complex in the same genes' would be actively harmful.
  • The reverse manoeuvre - stabilising or increasing PRC1.1 activity to resensitise - has no chemical starting point at all, and raising H2AK119ub globally would deepen the differentiation block the disease already has.
  • MYC is the reactivated node, and MYC is not a tractable direct target. The proposed workaround, BCL2 inhibition in PRC1.1-deficient cells, inherits venetoclax's own resistance biology.
  • If PRC1.1 status becomes a response biomarker it will need an assay that works on an infant ALL aspirate. PCGF1 protein loss, BCOR truncation and KDM2B CxxC disruption are three different measurements and no single test covers them.
Still open
  1. Does PRC1.1 loss occur in patients relapsing on revumenib or ziftomenib? Every result here is engineered loss in a model; sequencing PCGF1, BCOR, BCORL1 and KDM2B in paired pre- and post-menin-inhibitor samples would settle whether this is a real clinical mechanism.
  2. Does any of this hold in KMT2A::AFF1 infant B-ALL? Every functional result in this entry is AML or T-ALL.
  3. Is PRC1.1 required for KMT2A-rearranged leukaemic stem cell function, as van den Boom's KDM2B data suggest, even though DepMap sees no bulk requirement? A xenograft LSC assay using PCGF1 rather than KDM2B depletion would separate these.
  4. Does the PCGF1 RAWUL-BCOR PUFD interface have enough of a pocket for a chemical probe, and would a PRC1.1 disruptor be pro- or anti-leukaemic given that loss confers drug resistance?
  5. Where in the nucleus is PRC1.1 in a KMT2A-rearranged blast? No subnuclear localisation is published, and the fusion's own condensate behaviour makes that a non-trivial question.
Every source behind this entry (10)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Six major PRC1 groups exist, each containing a distinct PCGF subunit plus a RING1A/B ubiquitin ligase and a unique set of associated polypeptides; they differ in genomic localisation and only a small subset co-localises with H3K27me3, and RYBP or YAF2 association excludes canonical CBX/PHC/SCM subunits while stimulating RING1B activity toward H2AK119ub1.

    PCGF homologs, CBX proteins, and RYBP define functionally distinct PRC1 family complexes · Gao Z, Zhang J, Bonasio R, Strino F, Sawai A, Parisi F, Kluger Y, Reinberg D · Molecular Cell 45(3):344-56 2012
    PMID 22325352DOI
    What this evidence cannot show

    Abstract fetched and read on 2026-09-19; full text not read. Proteomics in HEK293 cells and mouse embryonic stem cells, not in leukaemia. The composition statement is biochemistry and is not a claim that any one complex is present in a KMT2A-rearranged cell.

    Evidence that argues against this (1)
    • The subunit roster is not fixed across systems. Isshiki et al. define PRC1.1 as KDM2B + RING1B + PCGF1 + BCOR-or-BCORL1, without SKP1 or RYBP in the headline definition, so which polypeptides count as core PRC1.1 depends on the purification. An entry that treats any one roster as canonical is overstating. Isshiki Y et al. 2019
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In KMT2A-rearranged AML a group of noncanonical menin targets bivalently co-occupied by active menin and repressive H2AK119ub is normally downregulated after menin inhibition; loss of PRC1.1 subunits such as PCGF1 or BCOR causes menin-inhibitor resistance by epigenetic reactivation of these targets including MYC; genetic or pharmacological MYC inhibition resensitises; and PRC1.1-deficient cells show reduced monocytic gene signatures and susceptibility to BCL2 inhibition, with venetoclax overcoming menin-inhibitor resistance.

    Epigenetic regulation of noncanonical menin targets modulates menin inhibitor response in acute myeloid leukemia · Zhou X, Zhang L, Aryal S, Veasey V, Tajik A, Restelli C, Moreira S, Zhang P, Zhang Y, Hope KJ, Zhou Y, Cheng C, Bhatia R, Lu R · Blood 144(19):2018-2032 2024
    PMID 39158067DOI
    What this evidence cannot show

    Abstract fetched and read on 2026-09-19; full text not read. All models are **AML**, human and mouse KMT2A-rearranged; nothing here is infant B-ALL or KMT2A::AFF1 specifically, and this project's disease is infant B-ALL. Resistance is demonstrated by engineered loss of PRC1.1 subunits, not by observing PRC1.1 loss arise in patients on a menin inhibitor, so the clinical frequency of this mechanism is not established by this paper.

    Evidence that argues against this (1)
    • The opposite sign is on record in a different lineage: in T-ALL, KDM2B/PRC1.1 acts as a **tumour suppressor** whose loss initiates disease, restraining NOTCH1-driven activation. Read together with this paper, PRC1.1's direction of effect is lineage- and context-dependent, and 'PRC1.1 loss helps the leukaemia' is a statement about one AML context rather than a general rule. Isshiki Y et al. 2019
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    A cell-surface bivalency readout coupled to whole-genome CRISPR-Cas9 screens identified specific roles for MTF2-PRC2.1, PCGF1-PRC1.1 and Menin-KMT2A/B complexes in maintaining bivalency; genetic loss or pharmacological inhibition of menin phenocopies polycomb disruption and derepresses bivalent genes in both cancer cells and pluripotent stem cells.

    Targeting Menin disrupts the KMT2A/B and polycomb balance to paradoxically activate bivalent genes · Sparbier CE, Gillespie A, Gomez J, Kumari N, Motazedian A, Chan KL, Bell CC, Gilan O, Chan YC, Popp S, Gough DJ, Eckersley-Maslin MA, Dawson SJ, Lehner PJ, Sutherland KD, Ernst P, McGeehan GM, Lam EYN, Burr ML, Dawson MA · Nature Cell Biology 25(2):258-272 2023
    PMID 36635503DOI
    What this evidence cannot show

    Abstract fetched and read on 2026-09-19; full text not read. The reporter is a bivalent MHC class I gene in cancer cell lines and pluripotent stem cells, not a leukaemia differentiation phenotype, and the menin arm is a pharmacological phenocopy in those systems. Generalising 'menin inhibition activates bivalent genes' to KMT2A-rearranged patient blasts is an extrapolation this paper does not make.

    Evidence that argues against this (1)
    • Zhou et al. report the same axis with the opposite emphasis: after menin inhibition the bivalent noncanonical targets are typically **downregulated**, and it takes PRC1.1 loss to reactivate them. The two results reconcile only if direction depends on which polycomb arm survives, which means neither paper alone licenses a prediction about what menin inhibition does to a given bivalent gene. Zhou X et al. 2024
  4. E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    A Perturb-seq map of KMT2A-rearranged AML found a synergistic hub of KAT6A, menin and DOT1L silencing a core myeloid differentiation programme, and reported that disruption of antagonistic regulators of that programme, such as the PRC1.1 component PCGF1, confers strong resistance to DOT1L inhibition.

    PMID 41882099DOI
    What this evidence cannot show

    Abstract fetched and read on 2026-09-19; full text not read. Same senior-author group as [E2] (Lu R, Cheng C), so this is not an independent laboratory even though it is an independent drug class and assay. AML only; no B-lineage model.

    Evidence that argues against this (1)
    • The shared authorship with [E2] is the main reason to discount this as confirmation. A genuinely independent replication of 'PRC1.1 loss confers epigenetic-drug resistance in KMT2A-rearranged leukaemia' has not been published as of this cycle, and the central claim of this entry would be materially strengthened or weakened by one. Zhou X et al. 2024
  5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    In this project's genome-wide KMT2A-rearranged selectivity scan of 17,728 genes, PCGF1 ranks 17,405 (98.178th percentile) with a difference of +0.1448 (KMT2A-rearranged mean +0.0593 against other-leukaemia mean -0.0855) and a pan-essentiality of -0.0384 across 1,208 lines. The other PRC1.1 members score in the same direction: BCOR +0.1019 (rank 16,929), RNF2 +0.0964 (rank 16,819), BCORL1 +0.0609 (rank 15,515). For contrast MEN1 ranks 1 (-0.8656) and KMT2A ranks 11 (-0.5411).

    Genome-wide KMT2A-rearranged selectivity scan, 17,728 genes (DepMap Chronos via Open Targets) · Broad Institute DepMap / Open Targets Platform; analyses/depmap-kmt2a-selective-scan-full 2026
    What this evidence cannot show

    Numbers read directly out of analyses/depmap-kmt2a-selective-scan-full/results.json on 2026-09-19 (file generated 2026-09-19T10:12:06Z). Eleven KMT2A-rearranged against 43 other leukaemia lines; none of the eleven is infant-derived and the group is AML-heavy. PCGF1 is near-neutral in absolute terms (pan-essentiality -0.038), so a positive difference here is a difference between two small numbers and should not be read as proof that PRC1.1 protects KMT2A-rearranged cells.

    Evidence that argues against this (1)
    • DepMap measures bulk proliferation of established lines in culture with no drug present, so it cannot see a resistance phenotype at all. A gene whose loss confers drug resistance has no reason to score as a dependency, which makes the agreement between this scan and Zhou 2024 weaker than it looks: the scan rules out PCGF1 as a proliferation dependency, it does not confirm the resistance mechanism. Zhou X et al. 2024
  6. E6established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    A protein interaction domain in BCOR, named PUFD (PCGF Ub-like fold discriminator), binds the RING finger- and WD40-associated ubiquitin-like (RAWUL) domain of PCGF1 and PCGF3 but not of PCGF2 (MEL18) or PCGF4 (BMI1); the crystal structure of BCOR PUFD bound to PCGF1 shows the discrimination arises from the identity of specific residues within the interaction surfaces.

    Structure of the polycomb group protein PCGF1 in complex with BCOR reveals basis for binding selectivity of PCGF homologs · Junco SE, Wang R, Gaipa JC, Taylor AB, Schirf V, Gearhart MD, Bardwell VJ, Demeler B, Hart PJ, Kim CA · Structure 21(4):665-71 2013
    PMID 23523425DOI
    What this evidence cannot show

    Abstract fetched and read on 2026-09-19. Crystallography of isolated domains, not of an assembled PRC1.1 on chromatin; binding preference in vitro does not by itself establish which complexes exist in a leukaemic nucleus.

    Evidence that argues against this (1)
    • The isolated-domain picture was later corrected by the same group: BCOR PUFD alone does NOT stably assemble with KDM2B, residues N-terminal to PUFD are required, and NMR shows the PUFD termini are disordered until PCGF1 binds. Fragment affinities therefore understate how conditional PRC1.1 assembly is, and reading 'PCGF1 RAWUL is a defined druggable pocket' off this structure would be too simple. Wong SJ et al. 2020
  7. E7strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Mass-spectrometry-defined non-canonical PRC1.1 is critically important for human leukaemic stem cells: downmodulation of PRC1.1 members such as the DNA-binding subunit KDM2B strongly reduces proliferation in vitro and delays or abrogates leukemogenesis in humanised xenografts; PRC1.1 components are significantly over-expressed in primary AML CD34+ cells; and ChIP-seq shows PRC1.1 binds a distinct set of transcriptionally active genes devoid of H3K27me3, enriched for metabolic functions.

    Non-canonical PRC1.1 Targets Active Genes Independent of H3K27me3 and Is Essential for Leukemogenesis · van den Boom V, Maat H, Geugien M, Rodriguez Lopez A, Sotoca AM, Jaques J, Brouwers-Vos AZ, Fusetti F, Groen RWJ, Yuan H, Martens ACM, Stunnenberg HG, Vellenga E, Martens JHA, Schuringa JJ · Cell Reports 14(2):332-46 2016
    PMID 26748712DOI
    What this evidence cannot show

    Abstract fetched and read on 2026-09-19. The functional depletion is of **KDM2B**, not of PCGF1, so PCGF1's own requirement in leukaemic stem cells is inferred from complex membership rather than demonstrated. Primary AML CD34+ cells were not selected for KMT2A rearrangement.

    Evidence that argues against this (1)
    • This is the entry's strongest internal tension and it is unresolved. van den Boom reports PRC1.1 as ESSENTIAL for leukaemic stem cells, while Zhou 2024 reports PRC1.1 loss as a resistance state the leukaemia survives, and the genome-wide scan puts every PRC1.1 member on the anti-dependency side. The likeliest reconciliation is that the KDM2B requirement is a leukaemic-stem-cell/xenograft phenotype invisible to both bulk culture and a drug-resistance screen, but that is a hypothesis and this entry does not assert it as fact. Broad Institute DepMap / Open Targets Platform; analyses/depmap-kmt2a-selective-scan-full 2026
  8. E8strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    KDM2B-PRC1.1 target promoters in mouse thymocytes carry high H2AK119ub1 and high H3K4me3 but low H3K27me3, i.e. they are transcriptionally active or primed for activation rather than classically Polycomb-silenced, and PRC1.1 restrains excessive transcriptional activation by active NOTCH1.

    KDM2B in polycomb repressive complex 1.1 functions as a tumor suppressor in the initiation of T-cell leukemogenesis · Isshiki Y, Nakajima-Takagi Y, Oshima M, Aoyama K, Rizk M, Kurosawa S, Saraya A, Kondo T, Sakaida E, Nakaseko C, Yokote K, Koseki H, Iwama A · Blood Advances 3(17):2537-2549 2019
    PMID 31471323DOI
    What this evidence cannot show

    Abstract fetched and read on 2026-09-19. Mouse thymocytes and human T-ALL: a different lineage and a different oncogenic driver from KMT2A::AFF1. The chromatin signature is carried into this entry as a description of PRC1.1 biology, not as a measurement made in a KMT2A-rearranged cell.

    Evidence that argues against this (1)
    • The lineage caveat cuts hard. Here PRC1.1 is a tumour suppressor whose loss initiates leukaemia; in the KMT2A-rearranged AML work PRC1.1 loss is an acquired drug-resistance state in an already-established leukaemia. The chromatin signature may be conserved while the phenotype inverts, and this entry makes no claim that it does not. Zhou X et al. 2024
  9. E9moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    ATM phosphorylates the MLL fusion partner ENL (MLLT1) at conserved SQ sites in response to DNA double-strand breaks; this increases ENL's interaction with the PRC1 E3-ubiquitin-ligase complex via BMI1, enriching PRC1 at transcription elongation sites near breaks and ubiquitinating H2A to repress transcription.

    PMID 25921070DOI
    What this evidence cannot show

    Abstract fetched and read on 2026-09-19. The PRC1 involved is the **BMI1 (PCGF4)** canonical complex, not PCGF1-PRC1.1, so this citation is carried here to mark what the physical link is NOT. It is a DNA-damage-response phenomenon in non-leukaemic cells and no KMT2A fusion protein was tested.

    Evidence that argues against this (1)
    • PCGF1 and PCGF4/BMI1 bind mutually exclusive corepressor partners: BCOR PUFD binds PCGF1 and PCGF3 but explicitly not BMI1 or MEL18. An ENL-BMI1 link therefore gives PRC1.1 no physical foothold on the fusion at all, and citing it as evidence for PCGF1 would be exactly the complex-level conflation this beat exists to avoid. Junco SE et al. 2013
  10. E10established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProtKB Q9BSM1 (PCGF1_HUMAN, Polycomb group RING finger protein 1) is 259 aa and annotates a RING-type zinc finger at aa 47-86, a region required for repressor activity at aa 86-247, a region required for interaction with the KDM2B-SKP1 heterodimeric complex at aa 150-255, and a RAWUL domain at aa 167-255 that is sufficient for interaction with BCOR and BCORL1.

    UniProtKB Q9BSM1 (PCGF1_HUMAN), Polycomb group RING finger protein 1 · UniProt Knowledgebase, reviewed (Swiss-Prot) 2026
    What this evidence cannot show

    Record fetched 2026-09-19 and the primary gene name (PCGF1) and length (259 aa) read back from the same fetched record rather than from recall. Domain boundaries are UniProt curator annotations aggregating several primary papers; they are not remeasured here and can shift between releases.

    Evidence that argues against this (1)
    • The RAWUL boundary given by UniProt (167-255) is not the construct boundary used in the crystallographic work, and domain limits defined by a database curator and by a crystallographer routinely differ by tens of residues. Any figure or docking exercise should use the modelled range from the PDB entry rather than these annotations. Junco SE et al. 2013
Connected to