In inv(16), disrupting the fusion interface swaps BRG1 for RING1B at three MYC enhancers - the cleanest version anywhere of the experiment nobody has run on menin
also called BRG1-RING1B swap · AI-10-49 MYC enhancer exchange · the inv(16) chromatin-replacement experiment
In a different leukaemia, a drug that breaks apart a fusion protein frees a normal transcription factor to sit on three switches that control MYC, and when it sits there it kicks off the machine that turns genes on and replaces it with a machine that turns them off - and the cells die. That is exactly the kind of measurement nobody has made for menin inhibitors in the leukaemia this project studies.
The same thing, for a scientist
Pulikkan et al. give the field's most complete account of what a fusion-interface inhibitor does at chromatin [E1]. In inv(16) AML, the small molecule AI-10-49 disrupts the CBFbeta-SMMHC/RUNX1 interaction; RUNX1 then binds three MYC distal enhancers (ME1 at +0.18 Mb, ME2 at +0.5 Mb, and element E3 of the five-element BDME/BENC cluster at +1.7 Mb) and represses MYC by mediating **replacement of the SWI/SNF component BRG1 with the PRC1 component RING1B**, causing apoptosis. Four features make it the reference experiment rather than one more chromatin paper. **(1) It is ordered in time**: RUNX1 and RING1B occupancy rise together from ~2.5 h, reaching 90% by 5 h, while BRG1 falls only between 4 and 6 h - the repressor arrives before the activator leaves, which is a replacement model and not a vacancy model. **(2) It is a physical interaction, not a coincidence**: ChIP-re-ChIP shows RUNX1 co-occupies E3 with RING1B but not with BRG1, and the association is drug-induced. **(3) It is genetically necessary**: RNF2 knockdown, or the RING1B-activity inhibitor PRT4165, partially rescues MYC expression under drug. **(4) It is locus-specific**: RING1B is recruited to ME1, ME2 and E3 but not to N-ME, the NOTCH-dependent T-cell MYC enhancer used as a non-myeloid control. The obvious question - **does menin inhibition do the same thing at the same element in KMT2A-rearranged cells?** - has never been asked. A PubMed query for RING1B or RNF2 together with MYC and enhancer returns exactly two records on 2026-09-20, one of which is this paper [E5].
The mapping, stated so it can be falsified
| inv(16) | KMT2A-rearranged |
|---|---|
| CBFbeta-SMMHC sequesters RUNX1 away from DNA | KMT2A fusion is held on chromatin by menin (and LEDGF) |
| AI-10-49 disrupts the fusion/partner interface | revumenib, ziftomenib disrupt the menin/KMT2A interface |
| a normal transcription factor (RUNX1) is released and becomes a repressor | the fusion is evicted; no released repressor is named |
| RUNX1 recruits RING1B to three MYC enhancers, evicting BRG1 | unmeasured |
| MYC falls; apoptosis | MYC falls in some models; differentiation is the dominant phenotype |
| RNF2 knockdown partially rescues MYC | unmeasured |
The mapping is not clean, and the place where it breaks is the most interesting part. In inv(16) the drug releases a repressor; in KMT2A-rearranged disease the drug removes an activator. Those are different causal shapes, and the inv(16) result therefore does not predict the KMT2A result - it supplies a measurement template, not a conclusion. If the same BRG1-to-RING1B exchange were found after menin inhibition, the mechanism reaching it would have to be different, and the obvious candidate is the one Neville et al. describe: PRC1.1 arriving at freed fusion-target chromatin and depositing H2AK119ub 2. hyp-menin-inhibition-swaps-brg1-for-ring1b-at-benc states that as a falsifiable hypothesis and exp-brg1-ring1b-timecourse-at-benc-under-menin-inhibition states the experiment.
The caveat that stops this being a PRC1.1 result
Pulikkan et al. measure H3K27me3 as their repressive readout alongside RING1B occupancy 1. H3K27me3 is a PRC2 mark. The direct catalytic product of RING1B is H2AK119ub, and it is not measured. Nor is any PCGF paralogue, BCOR, BCORL1 or KDM2B assayed, so the PRC1 flavour is not identified and nothing in the paper establishes that PRC1.1 was involved rather than canonical PRC1 or another variant.
That matters here more than it usually would, because Sijm et al. show H2AK119ub1 dosage is largely uncoupled from H3K27me3 3 - so the H3K27me3 gain cannot be read backwards as a measure of the PRC1 catalytic event. An entry that presented this as 'PRC1.1 replaces BRG1' would be over-reading a real result by exactly one complex.
The graph's existing treatment of this paper, in fusion-cbfb-myh11, cites it correctly as a counter-evidence note about single-agent interface disruption being insufficient without JQ1. Nothing there is wrong; what was missing is the element-level mechanism, which is what this entry adds.
The second lesson, which is about combinations and not about Polycomb
The mechanistic account is complete and the single agent still did not clear the disease: eliminating inv(16) leukaemia in human cells and in a mouse model required combining the interface inhibitor with the BET inhibitor JQ1 1. fusion-cbfb-myh11 already records this as a direct warning for the lab's menin-interface programme.
The enhancer mechanism explains why, and the explanation is transferable. BRD4 occupies the same cluster 4; RING1B replaces BRG1 there but BRD4 is not reported to leave. A BET inhibitor removes the remaining activator from the same element. So the successful combination is not an empirical pairing - it is two drugs acting on one enhancer, and the reason to expect synergy is legible before the experiment. If menin inhibition turns out to recruit PRC1 at BENC in KMT2A-rearranged cells, the same argument predicts menin-plus-BET synergy at that element for the same reason, which is a specific, testable prediction rather than a general hope.
| domain | engineering |
|---|---|
| element | noncoding-benc-myc-enhancer-cluster |
| functional test in source | CRISPR deletion of the RUNX1 site within ME1 (41 bp), ME2 (67 bp) or E3 (275 bp of ~2,800 bp): 40-50% reduction in MYC, 60-70% reduction in viability, each enhancer singly |
| prc1 flavour identified | false |
| repressive mark measured | H3K27me3 (a PRC2 mark), not H2AK119ub |
| what would make this a prc1 1 result | assay H2AK119ub at the same enhancers, and blot or ChIP a PCGF1/BCOR/KDM2B subunit; neither was done |
| asked in kmt2a rearranged cells | false |
mapping
[
{
"bio": "AI-10-49 disrupts the CBFbeta-SMMHC/RUNX1 interface in inv(16) AML",
"analog": "revumenib/ziftomenib disrupt the menin/KMT2A interface in KMT2A-rearranged leukaemia"
},
{
"bio": "released RUNX1 recruits RING1B to three MYC enhancers and evicts BRG1",
"analog": "UNMEASURED - no released repressor is named for the menin axis, and no enhancer-level occupancy has been reported"
},
{
"bio": "RNF2 knockdown or PRT4165 partially rescues MYC under drug",
"analog": "UNMEASURED - PRC1.1 subunit loss is known to confer menin-inhibitor resistance, which is the corresponding genetic epistasis at whole-cell rather than locus level"
},
{
"bio": "repressor arrives at 2.5-5 h, activator leaves at 4-6 h (replacement, not vacancy)",
"analog": "the time-resolution this project's menin-inhibitor chromatin experiments have never had"
},
{
"bio": "single-agent interface disruption mechanistically correct but therapeutically insufficient; JQ1 required",
"analog": "already recorded in fusion-cbfb-myh11 as a direct warning for the menin-interface programme"
}
]generated hypotheses
[ "hyp-menin-inhibition-swaps-brg1-for-ring1b-at-benc" ]
enhancers assayed in source
[ "ME1 (+0.18 Mb)", "ME2 (+0.5 Mb)", "BDME E1-E5 (+1.7 Mb), with E3 the drug-induced RUNX1 peak", "N-ME / NDME (+1.4 Mb, NOTCH-dependent, T-cell) as non-myeloid negative control" ]
- The inv(16) precedent's strongest single lesson is negative for this project: a completely correct, time-resolved, genetically validated interface-disruption mechanism still needed a second chromatin drug to clear disease in vivo. A menin-interface programme should plan for a combination partner from the start rather than treat one as a fallback.
- If menin inhibition does recruit PRC1 at MYC enhancers, then PRC1.1 loss - already an established menin-inhibitor resistance route - acquires a second, locus-specific mechanism of escape, and the resistance is enhancer-local rather than global, which makes it harder to detect in a bulk assay.
- PRT4165, the tool used to inhibit RING1B activity in the source, is of limited selectivity; any therapeutic inference about blocking the swap is chemistry this field does not have.
- MYC enhancer usage in MLL-AF9 is documented to re-route under chromatin-drug pressure via WNT-recruiting alternative enhancers, so even a successful enhancer-directed intervention has a published escape route.
- Does menin inhibition recruit RING1B and evict BRG1 at BENC in KMT2A-rearranged cells? This is the analogy's central untested prediction and the experiment is a time-course CUT&RUN.
- Is there a released repressor in the menin axis analogous to RUNX1 in inv(16)? If not, the causal shapes differ and any swap seen after menin inhibition would have to be explained by PRC1.1 arrival at freed chromatin instead.
- Would the inv(16) result reproduce if H2AK119ub rather than H3K27me3 were measured, and would a PCGF1 or BCOR ChIP place PRC1.1 at those enhancers? That single addition would convert a PRC1 result into a PRC1.1 result and connect two literatures that currently do not touch.
- Does BRD4 remain at the enhancers after the BRG1-to-RING1B exchange? If it does, that is the mechanistic reason JQ1 was required, and it predicts the same requirement for any menin-based version.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Pharmacological disruption of the CBFbeta-SMMHC/RUNX1 interaction with AI-10-49 causes RUNX1 to bind three MYC distal enhancers where it mediates replacement of BRG1 with RING1B, with H3K27me3 gain at those enhancers; RUNX1 and RING1B occupancy at E3 rises from about 2.5 hours reaching 90% by 5 hours while BRG1 falls only between 4 and 6 hours; ChIP-re-ChIP shows drug-induced RUNX1 co-occupancy with RING1B but not BRG1 at E3; and MYC expression is partially rescued by RNF2 knockdown or by the RING1B-activity inhibitor PRT4165.
CBFbeta-SMMHC Inhibition Triggers Apoptosis by Disrupting MYC Chromatin Dynamics in Acute Myeloid Leukemia · Pulikkan JA, Hegde M, Ahmad HM, Belaghzal H, Illendula A, Yu J, O'Hagan K, Ou J, Muller-Tidow C, Wolfe SA, Zhu LJ, Dekker J, Bushweller JH, Castilla LH · Cell 2018What this evidence cannot show
Read in full from PMC6211564 on 2026-09-20. The chromatin experiments are overwhelmingly in one human line, ME-1, with a mouse model for the in vivo combination. The rescue by RNF2 knockdown is explicitly PARTIAL, so RING1B is a contributor and not the whole mechanism. PRT4165 is a tool compound of imperfect selectivity, described in the source as 'pharmacologic inhibition of RING1B activity'. Total RING1B levels were unchanged by the drug, so this is redistribution, not induction. The disease is inv(16) AML, not a KMT2A rearrangement.
Evidence that argues against this (1)
- The paper's own headline is that the complete, correctly targeted single-agent mechanism was therapeutically insufficient: eliminating inv(16) leukaemia required combining the interface inhibitor with JQ1. A correct chromatin mechanism is therefore not a sufficient basis for a single-agent programme, which is the direct warning to this project's menin-interface hypothesis. Rathert P et al. 2015
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Menin inhibition induces PRC1.1-dependent deposition of H2AK119ub that silences a subset of MLL-fusion-protein targets, identifying non-canonical PRC1.1 as a mediator of the cellular response to menin and DOT1L inhibitors.
DOT1L provides transcriptional memory through PRC1.1 antagonism. · Neville D, Ferguson DT, Heikamp EB, Lai Z, Magor GW, Lam C, et al. · Nat Cell Biol 2026What this evidence cannot show
Abstract-level only. The abstract does not name the silenced subset, does not mention MYC, and does not report any enhancer-level measurement - so it supports the existence of a menin-inhibitor-induced PRC1 event and says nothing about whether it occurs at MYC enhancers. That gap is precisely what makes the analogy a hypothesis rather than a finding.
Evidence that argues against this (1)
- Zhou et al. place the H2AK119ub at these non-canonical menin targets as a pre-existing bivalent feature rather than a drug-induced deposition, and describe the targets as being DOWNREGULATED after menin inhibition with PRC1.1 already present. Whether the PRC1.1 event is arrival or activation is not settled, and the distinction changes the predicted time course of any swap experiment. Zhou X et al. 2024
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
H2AK119ub1 dosage is largely disconnected from H3K27me3, and even complete loss of the PRC1 catalytic core and of H2AK119ub1 has only a limited effect on H3K27me3.
USP7 regulates the ncPRC1 Polycomb axis to stimulate genomic H2AK119ub1 deposition uncoupled from H3K27me3. · Sijm A, Atlasi Y, van der Knaap JA, Wolf van der Meer J, Chalkley GE, Bezstarosti K, et al. · Sci Adv 2022What this evidence cannot show
Not leukaemia; a genome-scale dosage statement that does not exclude tight coupling at individual loci, which is exactly the scale at which the inv(16) enhancer result operates. Cited here to establish that H3K27me3 cannot substitute for H2AK119ub as a PRC1 readout, not to claim the two marks are unrelated.
Evidence that argues against this (1)
- At the specific enhancers in question, Pulikkan et al. observed RING1B recruitment and an H3K27me3 increase together, which is the coupled behaviour the classical hierarchical model predicts. A genome-wide dosage disconnection does not rule out local coupling, so the criticism of the readout stands as a limitation on inference rather than as a contradiction of the measurement. Pulikkan JA et al. 2018
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The MYC enhancer cluster 1.7 Mb downstream is occupied by SWI/SNF and by BRD4 in MLL-AF9 leukaemia cells, and Brg1 is required there to maintain transcription-factor occupancy and long-range looping to the Myc promoter.
Role of SWI/SNF in acute leukemia maintenance and enhancer-mediated Myc regulation. · Shi J, Whyte WA, Zepeda-Mendoza CJ, Milazzo JP, Shen C, Roe JS, et al. · Genes Dev 2013What this evidence cannot show
Mouse MLL-AF9 (RN2, which carries a cooperating NrasG12D allele). The MYC arm of this paper is directly contested in human cells, so the safe residue is the occupancy and looping observation rather than the dependency conclusion. Cited here to establish that the element this project would have to assay under a menin inhibitor is the same element and already has a documented BRG1/BRD4 occupancy baseline in a KMT2A-rearranged system.
Evidence that argues against this (1)
- Cruickshank et al. report that the same SWI/SNF subunits are not required for c-MYC-regulated gene expression in human MLL-rearranged cells, in explicit contrast to the mouse results. If SWI/SNF is not holding MYC in human KMT2A-rearranged cells, then there may be no BRG1 at the element to displace, and the proposed swap experiment could return a true negative for an uninteresting reason. Cruickshank VA et al. 2015
- E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Counted absence: a PubMed search on 2026-09-20 for (RING1B[tiab] OR RNF2[tiab]) AND MYC[tiab] AND enhancer[tiab] returns 2 records - the inv(16) paper itself and a mouse embryonic-stem-cell study of aryl hydrocarbon receptor repression - so no study has reported a RING1B or RNF2 measurement at MYC enhancers in KMT2A-rearranged cells.
Pluripotency factors and Polycomb Group proteins repress aryl hydrocarbon receptor expression in murine embryonic stem cells. · Ko CI, Wang Q, Fan Y, Xia Y, Puga A · Stem Cell Res 2014What this evidence cannot show
Bounded by the symbols RING1B and RNF2 and by [tiab]; a paper reporting the measurement under the phrase 'Polycomb' or 'PRC1' without naming the subunit would be missed, which brief rule 64 shows is a real failure mode for complex subunits. The complex form was run separately: (RING1B[tiab] OR RNF2[tiab]) AND (MLL[tiab] OR KMT2A[tiab]) returns 5 records, title-checked, none reporting MYC-enhancer occupancy. POSITIVE CONTROL: the identical query form returns the inv(16) paper, so the instrument can return a one. Authors of the co-hit verified from the fetched PubMed record on 2026-09-20.
Evidence that argues against this (1)
- The enhancer-naming problem recorded in concept-a-noncoding-element-has-no-symbol-and-three-names bounds this absence further than the symbol list does: a paper describing RING1B at 'BDME' or at 'the +1.7 Mb element' without using the word enhancer in its abstract would be invisible to this query. The absence is real for the symbols and phrasing searched and is not a claim about the whole literature. Pulikkan JA et al. 2018
- analogous to →CBFB::MYH11 (CBFbeta-SMMHC), inv(16)(p13;q22)
the source system: AI-10-49 interface disruption releases RUNX1 onto three MYC enhancers
the element at which the swap occurs (BDME element E3, plus ME1 and ME2)
the target of the analogy: does menin-interface disruption do the same at the same element?
RING1B is the replacing factor; RNF2 knockdown partially rescues MYC under drug
BRG1 is the displaced factor, leaving 4-6 h after RING1B arrives
the candidate KMT2A-side mechanism if the swap exists there: PRC1.1 arrival at freed fusion-target chromatin
the swap represses MYC and triggers apoptosis in inv(16)
generated by transplanting the inv(16) chromatin experiment onto the menin axis
the one published use of this chemistry at a MYC enhancer; the rescue was partial and the genetic arm carries the conclusion
the nearest sibling in form: another within-biology cross-fusion-family comparison in this graph, written to the same shape