Combine an IKK inhibitor with a clinical menin inhibitor in a KMT2A::AFF1 B-ALL line, and read out fusion OCCUPANCY rather than viability
also called IKK plus revumenib occupancy experiment · two-route eviction test
Treat infant-type leukaemia cells with a menin drug, an IKK drug, or both, and then measure how much of the fusion protein is still sitting on its target genes — not just whether the cells died. Measuring the chromatin is the only way to tell whether the two drugs are doing the same job or two different ones.
The same thing, for a scientist
**Why occupancy and not viability.** A viability combination index cannot distinguish two independent requirements from one requirement hit twice, and both drugs kill these cells. The claim under test is mechanistic — that IKK/NF-kB and menin are **non-overlapping** requirements for fusion retention [E1] — and only a chromatin read-out addresses it. This design is the 2013 figure-8 panel, run in the index lineage, with a menin-inhibitor arm added. **Design.** Human KMT2A::AFF1 B-ALL lines **SEM** and **RS4;11**, plus a KMT2A-rearranged AML line (MV4;11 or MOLM-13) as a lineage comparator and a non-KMT2A line (K562 or REH) as a specificity comparator. Four arms: vehicle, menin inhibitor alone, IKK inhibitor alone, both — each at a dose giving roughly 30-50% of its own maximal transcriptional effect, so additivity is detectable. **Read-outs, in priority order.** (1) ChIP-qPCR at **HOXA9** and **MEIS1** for the fusion (anti-AFF1 C-terminus, which the wild-type KMT2A does not carry), for **wild-type KMT2A** (anti-KMT2A C-terminus, which the fusion does not carry), for RELA, and for **H3K79me2** and **H3K4me3**; plus Ey-globin-equivalent and H4 negative loci. (2) **Total fusion protein by western** in every arm — the stability control that makes 'eviction' meaningful [E1]. (3) HOXA9/MEIS1 transcript. (4) Nuclear RELA and an NF-kB reporter in the menin-inhibitor-alone arm, which is the direct test of the liberation prediction [E2]. **Cost tier: low.** No animals, no new chemistry, four arms and one ChIP panel.
Controls that are not optional
- Antibody epitope separation. The entire species claim in this literature rests on it: HA or anti-AFF1 for the fusion, anti-KMT2A C-terminus for wild-type. Run each antibody against a fusion-negative line to confirm it discriminates. This graph has already been burned on a fusion-versus-wild-type distinction (LEDGF retains wild-type MLL but not MLL fusions), and that is precisely the error this control prevents.
- Two structurally unrelated IKK inhibitors, because the founding result used three research-grade compounds with no target-engagement assay 1. If only one reproduces, the effect is off-target.
- A NEMO-directed arm if available (NBD peptide or a NEMO degrader), because it is canonical-selective and would show whether the eviction needs the non-canonical arm blocked too.
- Time course. Sample at 2, 6 and 24 hours. Occupancy loss before mark loss supports the paper's causal order; the reverse or simultaneous would not.
- A menin-inhibitor-resistant MEN1-mutant subline (for example M327I) as a specificity control: IKK inhibition should still evict the fusion in it, and if it does, that is the sharpest possible demonstration of route independence — and a directly clinical one, since these are the cells patients relapse with.
What each outcome means
| result | interpretation |
|---|---|
| IKK alone reduces fusion occupancy, protein unchanged | the 2013 mechanism holds in human B-lineage cells; the biggest gap in this folder closes |
| IKK alone does nothing to occupancy | mechanism is myeloid- or mouse-specific; most of this folder becomes background rather than target biology |
| combination beats both singles | two routes, as hypothesised; supports sub-maximal dosing of the toxic partner |
| combination equals the better single | one requirement measured twice; the co-IP negative underlying the hypothesis is wrong or irrelevant |
| menin inhibitor alone raises nuclear RELA | the liberation half of the hypothesis survives, and the combination is partly self-repair |
| menin inhibitor alone lowers NF-kB output | liberation half dies cheaply; combination reverts to a plain additivity claim |
| IKK evicts the fusion in a MEN1-mutant resistant subline | route independence demonstrated in the clinically relevant cell |
What this experiment deliberately does not attempt
It does not attempt a therapeutic claim. No dose used here is a tolerable human dose, no animal is treated, and nothing about the combination's safety follows from it — that is risk-ikk-inhibition-essential-immune-function. It also does not adjudicate the canonical/non-canonical question, because pan-IKK compounds cannot; the NEMO arm is the only part that touches it, and it is optional because no clean reagent may be available.
| design | Four-arm drug matrix (vehicle, menin inhibitor, IKK inhibitor, both) at sub-maximal doses in KMT2A::AFF1 B-ALL lines SEM and RS4;11, with a KMT2A-rearranged AML line as lineage comparator, a non-KMT2A line as specificity comparator, and a MEN1-mutant menin-inhibitor-resistant subline as a route-independence control. Two structurally unrelated IKK inhibitors. Optional canonical-selective arm via a NEMO-directed agent. Sampling at 2, 6 and 24 hours. |
|---|---|
| readout | Primary: ChIP-qPCR occupancy at HOXA9 and MEIS1 for the fusion (anti-3' partner), wild-type KMT2A (anti-C-terminus), RELA, H3K79me2 and H3K4me3, with negative loci. Secondary: total fusion protein by western in every arm; HOXA9/MEIS1 transcript; nuclear RELA and NF-kB reporter in the menin-inhibitor-alone arm. |
| controls | Antibody epitope discrimination verified against a fusion-negative line; two chemically unrelated IKK inhibitors; negative loci; IgG; total-protein stability control; MEN1-mutant resistant subline; non-KMT2A comparator line. |
| cost tier | low |
| timeline | 6-10 weeks with existing lines and reagents; longer if a tagged knock-in is needed for fusion ChIP |
- This experiment can succeed mechanistically and mean nothing therapeutically. Two drugs that are additive on chromatin in a dish may be additive on marrow suppression in a child, and nothing here measures that.
- Research-grade IKK inhibitors at micromolar concentrations have substantial off-target liability. A positive result with one compound and not the other is the likeliest outcome and would consume the whole experiment.
- If the menin-inhibitor arm raises NF-kB output, the honest conclusion is that the approved drug partly works against itself in these cells - an uncomfortable finding that would need replication in primary blasts before it could be said aloud.
- Is there a ChIP-grade antibody against AFF1 that works in SEM or RS4;11? If not, the experiment needs an epitope-tagged knock-in and moves from low to medium cost tier.
- Which menin inhibitor? Revumenib is approved and available, but a tool compound with a cleaner off-target profile may be preferable for a mechanistic read-out.
- Should the IKK arm be an inhibitor or a degrader? The only agent with in vivo KMT2A evidence is a dual IKKb/NR4A1 degrader, which is a worse mechanistic tool and a better therapeutic one.
- Would CUT&RUN be preferable to ChIP at these loci, given the small cell numbers a resistant subline provides?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The design reproduces a published panel: IKK inhibition reduced promoter occupancy of HA-tagged MLL-AF10 and of wild-type MLL at Hoxa9 and Meis1, with reduced H3K79me2 and H3K4me3 and with stable total MLL-AF10 protein; RELA occupancy fell in parallel; and RELA/p50 did not co-immunoprecipitate with MLL fusion proteins in AML cells. Antibodies used were HA (ab9110), Rela (ab7970), Mll C-terminus (Millipore 05-765), H3K4me3 (ab8580), H3K79me2 (ab3594), IgG (ab46540), with qPCR primers at Hoxa9, Meis1, Ey-globin and H4.
Epigenetic roles of MLL oncoproteins are dependent on NF-κB. · Kuo HP, Wang Z, Lee DF, Iwasaki M, Duque-Afonso J, Wong SHK, Lin CH, Figueroa ME, Su J, Lemischka IR, Cleary ML · Cancer Cell 2013What this evidence cannot show
The original panel used an HA-tagged fusion in mouse cells; a human line requires an antibody against the 3' partner, which is a real technical risk - anti-AFF1 ChIP is not a routine reagent and a tagged knock-in line may be needed. Full text read this session (PMC3816582).
Evidence that argues against this (1)
- The experiment can give a positive occupancy result for the wrong reason. Pan-IKK inhibitors also block IKKa, the non-canonical kinase, and activating that arm independently suppresses MLL-AF9 leukaemia - so an anti-leukaemic effect in the combination arms may not be the eviction the design attributes it to. Only the chromatin read-out, not the viability read-out, distinguishes them, which is why occupancy is the primary endpoint. Xiu Y et al. 2018
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Menin binds p50, p52 and p65 directly and represses p65-mediated transactivation dose-dependently, with menin residues 305-381 essential - a region distinct from the MBM pocket occupied by clinical menin inhibitors.
The tumor suppressor protein menin interacts with NF-kappaB proteins and inhibits NF-kappaB-mediated transactivation. · Heppner C, Bilimoria KY, ... Burns AL · Oncogene 2001What this evidence cannot show
Abstract fetched and read this session. The statement that 305-381 is distinct from the MBM pocket is a structural inference from the graph's own menin entries, not a claim this paper makes; the paper predates every menin inhibitor. It is why read-out (4) is included as an explicit test rather than assumed.
Evidence that argues against this (1)
- This is a 2001 result on WILD-TYPE menin in the context of MEN1 endocrine tumour biology, using transactivation reporters rather than leukaemic chromatin. Menin's repression of p65 in that setting does not establish what menin does at NF-kB sites in a KMT2A-rearranged blast, where menin's dominant role in this graph is as the fusion's chromatin anchor - and the two functions could oppose one another. Kuo HP et al. 2013
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
Counted absence establishing that this experiment is unperformed: no published work combines a menin inhibitor with any IKK or NF-kB-directed agent in any system. (menin OR MEN1 OR revumenib OR ziftomenib) AND ("IKK"[tiab] OR "NF-kappaB inhibitor"[tiab] OR bortezomib OR "IKK inhibitor"[tiab]) returns 3 records and title-checking removes all three.
PubMed E-utilities esearch, db=pubmed, run 2026-09-20 · NCBI · PubMed 2026What this evidence cannot show
Bleximenib, enzomenib and preclinical MI-series compounds were not included as search terms, so a combination study naming only one of those would be missed.
Evidence that argues against this (1)
- A zero at the symbol level is not a zero at the pathway level, and on this beat that is demonstrable rather than hypothetical: PMID 29320732 studies the non-canonical arm directly in MLL-AF9 leukaemia and is unreachable from any of these subunit symbol queries, because it names the kinase and the pathway rather than the subunit. The count is honest about what was searched and should not be read as evidence that the biology is unstudied - only that it is not indexed under this symbol. Xiu Y et al. 2018
four-arm occupancy matrix in KMT2A::AFF1 B-ALL lines
runs the 2013 chromatin panel in the index lineage, which is the beat's biggest gap
SEM and RS4;11 KMT2A::AFF1 B-ALL lines as the primary system