PRC1.1 is not both an effector and a resistance gene by coincidence: the two published readings are one activity measured from opposite ends
also called PRC1.1 effector versus gatekeeper · H2AK119ub dose hypothesis
Two recent papers say opposite-sounding things about the same Polycomb machine: one says a menin drug switches it on to silence the cancer's genes, the other says losing it makes the drug stop working. We propose these are the same fact seen from either side - the drug needs that machine to do its job, so cells without it are resistant.
The same thing, for a scientist
`mechanism-menin-inhibitor-silencing-is-prc1-1-deposition-not-only-eviction` records that menin inhibition induces **PRC1.1-dependent H2AK119ub** to silence a subset of fusion targets, arising from progressive loss of DOT1L-mediated H3K79 methylation [E1]. `resistance-prc1-1-loss-menin-inhibitor-nup98` records that **loss of PCGF1 or BCOR confers menin-inhibitor resistance** by reactivating non-canonical menin targets including MYC [E2]. The graph carries both and reconciles neither. **Hypothesis: there is one activity - H2AK119ub deposition by PRC1.1 at menin-inhibitor-responsive loci - and the two papers measure its presence and its absence.** If so the relationship is a **dose relationship**, and three things follow that neither paper states: baseline PRC1.1 capacity should predict depth of response; H2AK119ub should be a pharmacodynamic biomarker of a menin inhibitor rather than of polycomb; and RYBP or YAF2, which set the complex's catalytic rate [E3], should modulate response without being resistance genes in a knockout screen.
Why the two readings are not in fact contradictory
| paper | perturbation | what it measures | what it concludes |
|---|---|---|---|
| 1 | menin inhibitor, PRC1.1 intact | H2AK119ub appears at fusion targets | PRC1.1 executes the silencing |
| 2 | PRC1.1 subunit knockout, then menin inhibitor | response is lost, MYC reactivates | PRC1.1 loss causes resistance |
These are the presence and absence arms of one experiment that nobody ran together. A necessary effector is automatically a resistance gene when it is lost - that is not a coincidence, it is the definition. The apparent tension comes from the vocabulary: 'effector' and 'resistance gene' belong to different literatures and are rarely applied to one complex in one sentence.
What makes the hypothesis non-trivial is its quantitative form. If PRC1.1 is the executing arm, then response depth should be a graded function of PRC1.1 capacity, not a binary of loss versus intact - and nothing published tests a partial reduction.
The alternative this hypothesis has to beat
Two-locus-class model. PRC1.1 could act as the effector at one set of loci and as a repressor of non-canonical menin targets such as MYC at a different set, with no shared activity. On that model, PRC1.1 loss would both abolish drug-induced silencing at class A and de-repress class B, and the resistance would be driven by class B alone.
The two models make different predictions, which is what makes this worth stating:
| one-activity model | two-locus-class model | |
|---|---|---|
| partial PRC1.1 reduction | graded loss of response | threshold, with MYC de-repression dominating |
| H2AK119ub at MYC in drug-treated PRC1.1-intact cells | increases with the rest | already present at baseline, does not change |
| RYBP/YAF2 double depletion | phenocopies PCGF1 loss | may not, if locus selection rather than catalysis is what matters |
hyp-prc2-is-the-fusions-repressive-arm is a third and separate proposal about a different complex; this one is about PRC1.1 only.
| statement | The PRC1.1-dependent H2AK119ub deposited after menin inhibition and the menin-inhibitor resistance caused by PRC1.1 subunit loss are the presence and absence arms of one activity, so response depth is a graded function of PRC1.1 catalytic capacity rather than a binary of intact versus lost. |
|---|---|
| origin | literature-gap |
| prior probability | 0.5 |
| alternative model | two locus classes - PRC1.1 as effector at drug-silenced fusion targets and as an independent repressor of non-canonical menin targets including MYC, with no shared activity |
| chain | mechanism-kmt2a-fusion-causal-chain |
predictions
[ "Partial PRC1.1 reduction (hypomorph, degron titration, or RYBP/YAF2 depletion) gives graded loss of menin-inhibitor response rather than a threshold.", "Spike-in-calibrated H2AK119ub rises at menin-inhibitor-responsive loci with a timecourse that follows H3K79me2 decay rather than menin displacement.", "Baseline PRC1.1 subunit expression predicts depth of response across KMT2A-rearranged models.", "BCOR- or PCGF1-mutant patient samples show reduced menin-inhibitor sensitivity with preserved target engagement." ]
experiments
[ "exp-prc1-1-capacity-titration-under-menin-inhibition" ]
analyses
[ "ana-h3k27me3-under-menin-inhibition-the-absence-is-narrower-than-stated" ]
falsifiers
[ "Partial PRC1.1 reduction gives a threshold rather than a graded response.", "H2AK119ub at MYC is unchanged by menin inhibition in PRC1.1-intact cells, which would put the resistance route at loci the effector route never touches.", "Menin-inhibitor-induced silencing survives complete loss of PRC1.1 catalytic activity, which would make PRC1.1 a modifier rather than the effector.", "Baseline PRC1.1 expression does not predict response across models." ]
- If the drug's effect requires host PRC1.1, then BCOR and PCGF1 mutations - recurrent in myeloid disease - are a pre-existing resistance population, and no menin-inhibitor trial has stratified on them.
- Any combination that inhibits polycomb to relieve repression would, on this hypothesis, antagonise the menin inhibitor's own effector arm, and EZH1/EZH2 inhibitor combinations are actively proposed in this disease.
- A pharmacodynamic biomarker that depends on the host's polycomb capacity will read as non-response in patients whose drug is working on the fusion but whose effector arm is impaired.
- Do the two published systems overlap at all at the locus level? Neither paper reports the other's locus set, so the simplest test - are the drug-silenced loci and the MYC-bypass loci the same genes - has not been done and needs no new experiment.
- Is H2AK119ub measurable in patient material on a menin inhibitor? If it is, the chain gains its first pharmacodynamic readout downstream of target engagement.
- Does the bivalency-maintenance role reported in normal cells make PRC1.1 a shared toxicity surface, so that a strategy exploiting it in leukaemia also perturbs normal progenitors?
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Menin inhibition induces PRC1.1-dependent deposition of H2AK119ub to silence a subset of MLL-fusion-protein targets, and the enhanced PRC1.1 activity arises specifically from progressive loss of DOT1L-mediated H3K79 methylation, independently of other routes.
DOT1L provides transcriptional memory through PRC1.1 antagonism. · Neville D, Ferguson DT, Heikamp EB, Lai Z, et al. · Nat Cell Biol 2026What this evidence cannot show
Abstract read 2026-09-20. Cell-line and mouse work; not in infant KMT2A::AFF1 material. The causal route from H3K79 loss to PRC1.1 gain is the paper's own claim and is the part that makes the dose form of this hypothesis testable.
Evidence that argues against this (1)
- The same paper reports that DOT1L inhibition produces a genome-wide H2AK119ub increase while menin inhibition produces a targeted one, so the two drugs engage PRC1.1 differently and a single dose relationship may not describe both. Zhou X et al. 2024
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Chromatin-focused CRISPR screens in human and mouse KMT2A-rearranged AML show that loss of PRC1.1 subunits such as PCGF1 or BCOR leads to menin-inhibitor resistance by epigenetic reactivation of non-canonical menin targets including MYC.
Epigenetic regulation of noncanonical menin targets modulates menin inhibitor response in acute myeloid leukemia. · Zhou X, Zhang L, Aryal S, Veasey V, et al. · Blood 2024What this evidence cannot show
Abstract read 2026-09-20. Knockout-level loss, AML models. A knockout screen cannot distinguish a necessary effector from an independent repressor of a bypass gene, which is exactly the ambiguity this hypothesis exists to resolve.
Evidence that argues against this (1)
- MYC reactivation is a recognised general bypass in this disease and is reported for BET-inhibitor resistance through PRC2 suppression as well, so a resistance phenotype routed through MYC is weak evidence about which complex was the effector. Rathert P et al. 2015
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The H2AK119ub ligase activity of PRC1 is set by its catalytic subunit composition and is highly regulated by RYBP/YAF2-dependent stimulation, with RYBP shaping H2AK119ub at Polycomb targets.
RYBP stimulates PRC1 to shape chromatin-based communication between Polycomb repressive complexes. · Rose NR, King HW, Blackledge NP, Fursova NA, et al. · Elife 2016What this evidence cannot show
Abstract read 2026-09-20. Mouse embryonic stem cells, not leukaemia. Supplies the rate-setting subunit that makes a dose experiment possible; the transfer to a leukaemic blast is an assumption.
Evidence that argues against this (1)
- RYBP and YAF2 appear interchangeable in the published biochemistry, so depleting one may change nothing, and a negative RYBP result would not discriminate between the two models. Gao Z et al. 2012
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Menin loss or pharmacological menin inhibition phenocopies polycomb disruption and derepresses bivalent genes, with PCGF1-PRC1.1 identified alongside MTF2-PRC2.1 and Menin-KMT2A/B as a maintainer of bivalency.
Targeting Menin disrupts the KMT2A/B and polycomb balance to paradoxically activate bivalent genes. · Sparbier CE, Gillespie A, Gomez J, Kumari N, et al. · Nat Cell Biol 2023What this evidence cannot show
Abstract read 2026-09-20; full text read for the H3K27me3 arm. K-562 and pluripotent stem cells. This is the third published position on PRC1.1 near menin - as a bivalency maintainer in normal cells - and it is not obviously the same activity as either of the other two.
Evidence that argues against this (1)
- In this system menin inhibition ACTIVATES bivalent genes while the effector model has menin inhibition driving PRC1.1 to SILENCE fusion targets; if both hold, PRC1.1's direction of effect is locus-class dependent, which favours the two-locus-class alternative. Neville D et al. 2026
COSTS THE TEST THAT HYPOTHESIS PROPOSED: its free first step requires peak calling, because neither study deposited locus sets - only coverage tracks.
the complement of its second prediction: that hypothesis proposes H2AK119ub as the menin inhibitor's pharmacodynamic biomarker, and this one measures that H3K79me2 cannot be - 0.0166 and 0.0101 against a 0.0584 two-culture spread in a PCGF1 knockout
the effector arm of the proposed single activity
the absence arm of the proposed single activity
resolves break 3 of the chain - the drug's output executed by the host cell
RYBP/YAF2 set the catalytic rate the dose form of this hypothesis requires
replaces the knockout with a dose axis
MAKES HALF ITS TEST FREE: one side of the proposed intersection is now a published gene-level list; only Neville's side still needs peak calling.