The KDM6 contradiction resolved one step: GSK-J4 inhibits UTX too, so it should ANTAGONISE a menin inhibitor - and nobody has combined them
also called GSK-J4 selectivity · KDM6A vs KDM6B · kdm6b contradiction · ledger item O
This graph held two published claims pointing opposite ways: one says the cell's ability to strip a repressive chemical tag is what makes menin-inhibitor drugs work, while the other says a drug that BLOCKS that stripping kills leukaemia cells. Chasing the reagent rather than the biology moved it forward - the blocking drug turns out not to be selective for one of the two enzymes that do the stripping but to block both, including the one the first story depends on, which turns a vague disagreement into a sharp prediction that the two drugs should fight each other, and nobody has yet put them in the same dish.
The same thing, for a scientist
`gene-kdm6b` recorded a live contradiction and listed five untested reconciliations. The reagent arm is now bounded. GSK-J1/J4 is a **KDM6A/KDM6B** inhibitor at ~60 nM, not a KDM6B-selective one, and its selectivity was formally challenged in Nature by a Brief Communication Arising (PMID 25279926, Heinemann/Helin). Because GSK-J4 inhibits **UTX/KDM6A** - the enzyme `resistance-mll34-utx-tumour-suppressor-failure` makes the therapeutic effector of menin inhibition - the two literatures generate a **falsifiable prediction: GSK-J4 should antagonise a menin inhibitor.** A validated search (19 records, all reviews) finds **no primary study combining a KDM6 inhibitor with a menin inhibitor.**
What the contradiction was, and what moved
gene-kdm6b holds two published directions:
resistance-mll34-utx-tumour-suppressor-failure: menin inhibition works partly by letting MLL3/4-UTX activate a tumour-suppressive programme; cells that cannot do so resist. UTX activates by removing H3K27me3.- PMID 29594337: GSK-J4 raises global H3K27me3, abrogates HOX expression and kills AML cells. Blocking the removal is therapeutic.
The entry listed five candidate reconciliations - different paralogues, different loci, different disease contexts, KDM6A/KDM6B non-equivalence, off-target activity - and recorded that none had been tested. That was honest and it was where the entry stopped.
The reagent is the cheapest of the five to settle, and it settles partly.
GSK-J4 is not KDM6B-selective, and this is on the record twice
First, the compound's own target list. A 2023 open-access review tabulates GSK-J1/J4 against KDM6A/6B, H3K27me2/me3, IC50 60 nM - the two paralogues together, not KDM6B alone (PMID 36975603, table read from the PMC full text on 2026-09-20).
Second, the selectivity claim was formally disputed in the journal that published it. The original paper is PMID 22842901 (Kruidenier, Nature 2012), whose title asserts a selective H3K27 demethylase inhibitor. PMID 25279926 (Heinemann, Nielsen, ..., Helin, Nature 2014, DOI 10.1038/nature13688) is a Brief Communication Arising titled 'Inhibition of demethylases by GSK-J1/J4', indexed with publication types Letter and Comment against it.
I could not read the Brief Communication. It has no abstract (BCA format), no PMC record, isOpenAccess: N, and the DOI resolves to a Nature paywall. Five routes, unreadable. So this entry claims only what the PubMed record establishes: a formal published challenge to the selectivity of GSK-J1/J4 exists in Nature and is authored by a major chromatin laboratory. It does not claim to know which off-targets the Communication reports, and any entry in this graph that wants the KDM5 cross-reactivity specifically must obtain the text first.
That alone is enough for the argument here, because the KDM6A/KDM6B dual activity is independently tabulated and is the part that matters.
The sharpened claim: the two literatures predict that these drugs fight each other
This is the step the previous entry did not take. GSK-J4 inhibits UTX/KDM6A. The menin-resistance story makes UTX demethylase activity the effector of the menin inhibitor's benefit. Therefore:
If the menin-resistance model is right, GSK-J4 should blunt a menin inhibitor's effect, because it chemically removes the very activity that model says the drug depends on.
And yet GSK-J4 alone kills AML cells and abrogates HOX expression, which is the same direction as a menin inhibitor. Both cannot be straightforwardly true in the same cells, and that is a much more testable statement than 'the literatures disagree'.
Three ways it could resolve, all measurable:
- Locus separation. UTX acts at the tumour-suppressor loci menin inhibition de-represses, while KDM6B sustains HOX. One enzyme, two address spaces - the same shape as this project's central KMT2A finding, where the CXXC address and not the menin contact is what matters.
- The GSK-J4 phenotype is not KDM6 at all. A compound with a disputed selectivity record killing cells at micromolar concentrations in culture is a familiar way to get a real phenotype from the wrong protein. This project logged the identical problem for CK2 and CX-4945.
- Context. PMID 29594337 is AML broadly; the menin-resistance entry is KMT2A-rearranged disease under a specific drug.
The decisive experiment is one plate. GSK-J4 plus revumenib in a KMT2A-rearranged line, reading viability and H3K27me3. Antagonism supports the UTX-effector model; additivity kills it. It also finally supplies the H3K27me3 measurement under menin inhibition that this project has been asking for since cycle 6 and that no study has reported H3K27me3 under a PHARMACOLOGICAL menin inhibitor - corrected 2026-09-20: Sparbier et al. 2023 (PMID 36635503) DID measure H3K27me3 by CUT&Tag, but in the MEN1-knockout arm, not the VTP50469 arm.
Correction 2026-09-20: the H3K27me3 absence is narrower than this entry said
This entry asserted that no menin-inhibitor study in any system reports an H3K27me3 measurement. Read as written that is false, and the counter-example is the paper this graph already cites for the polycomb phenocopy.
Sparbier et al. 2023 (PMID 36635503, Nature Cell Biology, full text PMC7614190) profiled H3K27me3 by CUT&Tag in MEN1-knockout versus control cells, reporting total H3K27me3, H3K4me3, H2AK119Ub, SUZ12 and EZH2 unchanged after MEN1 knockout, and at bivalent non-Menin-bound promoters increased H3K4me3 with reduced H3K27me3 and SUZ12.
The measurement is in the GENETIC arm, not the drug arm. The paper's VTP50469 experiments carry Menin ChIP-seq and RNA-seq and no H3K27me3.
So the defensible statement is the narrow one: no H3K27me3 measurement under a PHARMACOLOGICAL menin inhibitor has been reported (Sparbier et al. 2023, PMID 36635503, measured H3K27me3 by CUT&Tag in the MEN1-KNOCKOUT arm and found total H3K27me3 unchanged; its VTP50469 arm carries Menin ChIP-seq and RNA-seq and no H3K27me3). The experiment this graph keeps asking for - H3K27me3 under a clinical menin inhibitor, in the index fusion - is still unperformed, and knockout is not a substitute, because a drug that occupies a pocket and a deletion that removes the protein are different perturbations with different kinetics.
Correction 2026-09-20: the absence narrows a SECOND time
This entry's narrowed claim - no H3K27me3 measurement under a pharmacological menin inhibitor - is also false. PMID 38173411 treated human adipose-derived stem cell spheroids with 70 uM SNDX-5613 (revumenib) and blotted H3K27me3: "had no effect on H3K27me3".
The defensible statement is now: no locus-resolved H3K27me3 measurement under a pharmacological menin inhibitor has been reported in a KMT2A-rearranged leukaemia; one bulk Western blot exists, in a non-leukaemic system, and it was negative. Full reasoning and the bounded query in ana-the-h3k27me3-absence-narrows-a-second-time.
- GSK-J4 is a tool compound with a disputed selectivity record, not a candidate drug. Nothing here supports giving it to anyone, and a paper attributing a GSK-J4 phenotype specifically to KDM6B is naming rather than measurement.
- If the UTX-effector model of menin-inhibitor benefit is right, then any co-administered drug that inhibits KDM6A could blunt a menin inhibitor. That is an untested prediction and not a clinical warning - but it is the kind of interaction that would be found late and expensively if the combination reached patients without the one-plate experiment being done first.
- This analysis does not resolve which literature is correct. It narrows five untested reconciliations to a testable prediction and records that the test has not been run.
- This entry's central absence claim has now been narrowed TWICE, on 2026-09-20 both times. Treat the remaining claim as provisional and re-run the bounded query before quoting it.
- Does GSK-J4 antagonise revumenib in a KMT2A-rearranged line? One plate, and it settles the UTX-effector model.
- What does the Heinemann Brief Communication actually report? It is paywalled; an institutional copy would close the reagent question properly.
- Do UTX and KDM6B act at separable loci in KMT2A-rearranged cells - the locus-separation reconciliation, which is the same shape as this project's CXXC-address finding?
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
GSK-J1/J4 is tabulated as an inhibitor of KDM6A and KDM6B together, acting on H3K27me2/me3 with a half-maximal inhibitory concentration of 60 nM, rather than as an inhibitor selective for KDM6B.
Chemical Inhibitors Targeting the Histone Lysine Demethylase Families with Potential for Drug Discovery · Das ND, ..., Umehara T · Epigenomes 2023What this evidence cannot show
Read from the PMC full-text table of a 2023 review on 2026-09-20. **This is a secondary source**: the table aggregates values from primary references and the 60 nM figure is not re-derived here, nor is it stated which paralogue the number was measured against. Moderate for that reason. **What the table does establish firmly is the target assignment - KDM6A/6B, not KDM6B - which is the claim this analysis uses.** A primary enzymology paper should replace this citation.
Evidence that argues against this (2)
- The original report describes the compound as a selective H3K27 demethylase inhibitor, and selectivity against the KDM6 family as a whole is a different claim from selectivity between its two members - a compound can be genuinely selective for KDM6 over other JmjC demethylases while not discriminating KDM6A from KDM6B. The criticism here is of how the tool has been attributed in downstream papers, not necessarily of the original characterisation. Kruidenier L et al. 2012
- SUPERSEDED 2026-09-20 (second narrowing). A pharmacological menin inhibitor HAS now been assayed against H3K27me3: PMID 38173411 treated human adipose-derived stem cells with 70 uM SNDX-5613 (revumenib) and reported by Western blot that it 'had no effect on H3K27me3'. The defensible statement is now narrower again - no LOCUS-RESOLVED H3K27me3 measurement under a pharmacological menin inhibitor in a KMT2A-REARRANGED LEUKAEMIA - and the single existing datapoint is negative, in a non-leukaemic system, by bulk blot. See [[ana-the-h3k27me3-absence-narrows-a-second-time]]. Chang MM et al. 2024
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
A Brief Communication Arising titled 'Inhibition of demethylases by GSK-J1/J4', authored by Heinemann, Nielsen, Hudlebusch and colleagues with Kristian Helin, was published in Nature in 2014 and is indexed with publication types Letter and Comment against the 2012 report that introduced the compound as selective.
Inhibition of demethylases by GSK-J1/J4 · Heinemann B, Nielsen JM, Hudlebusch HR, ..., Helin K · Nature (Brief Communication Arising) 2014What this evidence cannot show
The PubMed record was fetched by efetch on 2026-09-20 and the title, authors, journal, year, DOI and publication types read back directly. **The content of the Communication was NOT read**: it carries no abstract, has no PMC record, is not open access, and the DOI resolves to a paywall - unreadable after five routes. **This evidence item therefore establishes the existence and target of a formal published challenge, and nothing about its findings.** Do not cite it in this graph for any specific off-target.
Evidence that argues against this (1)
- The existence of a Brief Communication Arising does not establish that its criticism was correct or accepted; Nature publishes an author reply alongside, which was also not read here. Treating an unread dispute as though it settles the matter would be the same error as treating the original title as though it settled selectivity. Kruidenier L et al. 2012
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database
No primary study has combined a KDM6 demethylase inhibitor with a menin inhibitor. A EuropePMC free-text search on 2026-09-20 for '"GSK-J4" AND ("menin inhibitor" OR revumenib OR VTP-50469 OR MI-3454 OR ziftomenib)' returned 19 records, and every one examined was a review naming the two agents separately rather than an experiment combining them.
EuropePMC REST search, GSK-J4 combined with menin inhibitors, with positive controls · Europe PMC 2026What this evidence cannot show
Run 2026-09-20 with positive controls through the identical code path, per brief rule 67: 'menin inhibitor' alone returned 538, revumenib 442, GSK-J4 1,255, so the instrument demonstrably returns results. **Only the first page of the 19 was examined by title**, so this is an absence of *findable primary reports*, not a proof that no such experiment exists - it could sit unnamed in a supplementary figure. Moderate for that reason. The search is title/abstract-weighted free-text and would miss a combination described only in methods.
Evidence that argues against this (1)
- Combination screens frequently include agents that never reach the title or abstract, and large drug-combination matrices in AML have included epigenetic tool compounds; absence from a free-text search is weaker evidence here than it would be for a headline biological claim. Li Y et al. 2018
CARRIES THE SAME SUPERSEDED CLAIM, now corrected in place.
- supports →KDM6B (JMJD3) - UTX's paralogue, which this graph has argued about for cycles without writing
ADVANCES THAT ENTRY'S RECORDED CONTRADICTION: its reagent caveat said GSK-J4's non-selectivity was 'naming and not measurement'. It is now sourced - GSK-J1/J4 is tabulated against KDM6A/6B at 60 nM, and the selectivity claim was formally challenged in Nature.
THE PARALOGUE THE CONTRADICTION TURNS ON: GSK-J4 inhibits UTX as well as KDM6B, so a compound used to argue that blocking demethylation is therapeutic also blocks the demethylase the menin-resistance model makes the therapeutic effector.
FALSIFIABLE PREDICTION: if MLL3/4-UTX demethylation is the effector of menin-inhibitor benefit, GSK-J4 should ANTAGONISE a menin inhibitor. Nobody has combined them.
THE MARK BOTH SIDES MOVE, IN OPPOSITE DIRECTIONS: the combination experiment would also supply the H3K27me3 measurement under menin inhibition that no study in any system has reported.
RULE 67 APPLIED IN ANGER: the counted absence here was recorded only after positive controls (538, 442, 1,255) showed the query returns results.