Oncofusion

start / graph / gene-jund

Geneestablished confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.Q-endocrine-liability

JUND - the AP-1 subunit menin holds down, and the one a menin inhibitor lets go

also called JunD · AP-1 subunit JunD · P17535

JUND makes a protein that switches genes on to drive cell growth, and menin normally sits on it and keeps it quiet. It is the only member of its protein family that menin binds.

The same thing, for a scientist

JUND encodes transcription factor JunD (UniProt P17535, 347 aa, nuclear), an AP-1 family subunit. It was identified as the first direct menin-interacting partner in a yeast two-hybrid screen, and menin bound **JunD alone** among Jun and Fos family members [E1]. Menin represses JunD-activated transcription both from a Gal4-fusion reporter and from an AP-1-responsive reporter, and several naturally occurring, clustered MEN1 missense mutations disrupt the menin-JunD interaction - which is the observation that ties this interaction to menin's tumour-suppressor function [E1]. Menin additionally uncouples JunD from upstream kinase signalling, inhibiting both ERK-dependent and JNK-mediated JunD phosphorylation downstream of MAPK activation [E2]. The consequence of releasing that hold is not neutral: JunD deprived of menin **switches from growth suppressor to growth promoter** [E3]. **Correction, 2026-09-19:** displacement of JunD by a menin inhibitor is no longer hypothetical - MI-503 and MI-463 disrupt menin-JunD binding by co-immunoprecipitation at 30-300 nM, and menin-MLL inhibitors disrupt menin-JUND in leukaemic cells by quantitative mass spectrometry, where the liberated JUND drives target genes and contributes to therapy resistance [counter-evidence to E3].

What it looks like
loading 3U86…
PDB 3U86 ↗
shape

This is real experimental data — atoms placed by crystallography or cryo-EM, deposited by the people who did the experiment.

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
Identity and how it was resolved

Queried UniProtKB with gene_exact:JUND AND organism_id:9606 AND reviewed:true and read back the primary gene name and length rather than trusting the search rank, per this project's standing rule after the TCF3/TCF7L1 incident:

FieldValue
AccessionP17535
Entry nameJUND_HUMAN
Primary gene nameJUND (no synonyms listed)
Protein nameTranscription factor JunD (Transcription factor AP-1 subunit JunD)
Length347 aa
Subcellular locationNucleus

The primary gene name matches the intended gene and there is no synonym collision of the kind that resolved TCF3 to TCF7L1 4.

The menin-binding motif

The deposited crystal structure of human menin in complex with JunD (PDB 3U86, X-ray, 2.84 A) contains a 15-residue JunD peptide, sequence SPGRLFPGAPPTAKK 5. Aligning that against the full-length JunD sequence places the native portion at approximately residues 27-38 (SPGRLFPGAPPT), with the trailing KK an engineered construct feature rather than contiguous native sequence; the core is LFPG at roughly residues 31-34.

The companion structure of menin with MLL1 (PDB 3U85, 3.0 A) contains the peptide SRWRFPARPGTTGGGGGGGRR, corresponding to the KMT2A MBM1 motif with RWRFP at residues 6-10 6.

Both peptides carry an F-P dipeptide - RWRFP in MLL1, RLFPG in JunD - and both occupy the same deep pocket. That is the structural reason the two cannot be separated pharmacologically by a pocket ligand, and it is the mechanistic core of the endocrine-liability question.

Correction (2026-09-19): displacement under drug is no longer an untested assumption, and the consequence is not primarily endocrine

This entry previously recorded that no co-immunoprecipitation of menin with JunD under a menin inhibitor had been published. That was wrong and is withdrawn. Two papers, neither cited here before, settle the pharmacology and move the argument:

  • Ma et al. 2021 (EMBO Mol Med, PMID 33821572) ran exactly the co-IP this beat kept asking for. MI-503 and MI-463 inhibited menin-JunD binding dose-dependently in HEK293T at 30, 100 and 300 nM. JunD recruits menin and HDAC3 to the Pbk promoter; a menin inhibitor disrupts that and de-represses Pbk.
  • Ozyerli-Goknar et al. 2026 (Blood, PMID 42371796) repeated the finding in the disease tissue and by an orthogonal method: quantitative mass spectrometry in MLL-rearranged leukaemic cells shows menin-MLL inhibitors disrupt menin-JUND interactions, JUND chromatin occupancy rises, its targets are upregulated, and this contributes to resistance to the drug.

Two revisions follow.

First, the structural inference is confirmed. A pocket ligand optimised against MBM1 does displace JunD. The selectivity the beat hoped for is not, on this evidence, achievable, and no inhibitor separating the two motifs has been reported.

Second, the direction of harm is not what this entry assumed. The published in-vivo endocrine phenotype of pharmacological menin inhibition is compensatory beta-cell proliferation with improved fasting glucose and glucose tolerance over 12 weeks at 70 mg/kg daily, and increased Ki67 in primary human islets at 10-100 nM [counter-evidence to E3]. The published leukaemia phenotype of the same displacement is acquired drug resistance. So the best-evidenced cost of JunD liberation today is loss of anti-leukaemic efficacy, not an endocrine tumour; the endocrine concern survives as a long-horizon question about exposures far longer than 12 weeks, in a tissue where menin loss takes years to produce disease.

Facts and figures
symbolJUND
uniprotP17535
length AA347
organismHomo sapiens
accession resolutiongene_exact:JUND AND organism_id:9606 AND reviewed:true; primary gene name read back as JUND and length confirmed at 347 aa on 2026-09-18
subcellular locationnucleus
familyAP-1 / Jun family
menin partner specificitythe only Jun or Fos family member menin binds directly
subnuclear localizationunpublished - no subnuclear compartment assignment for JUND was located; the published spatial facts are a nuclear/cytoplasmic partition and a chromatin occupancy change, not a compartment
shown for fusionYES for chromatin occupancy, which is the unusual case in this tranche. JUND chromatin binding was measured IN MLL-rearranged leukaemic cells and it MOVES under a menin-MLL inhibitor (PMID 42371796). NO for any subnuclear or imaging-level localisation, which has never been done in a KMT2A-rearranged cell.
cell context evidence classrequired_in (haemogenic endothelium, zebrafish development) + maintains_a_state (the menin-inhibited state, MLLr human and mouse). Two different cell contexts, two different evidence classes, and they must not be pooled - the developmental requirement says nothing about the drug-induced state and vice versa.
cell context shown for fusionSPLIT, and the split is the point. YES for the cell STATE: PMID 42371796 (Blood 2026) works in MLL-rearranged leukaemic cells with revumenib-class inhibitors, mass spectrometry, CRISPR and xenografts - the index fusion family, though AML and ALL are pooled and the partner is not resolved. NO for the cell TYPE: the haemogenic-endothelium requirement is zebrafish embryonic development, a cell type that does not exist after birth and has no KMT2A-rearranged counterpart.
cell context depmap reconciliationThe baseline screen and the drug-induced state agree once you stop expecting them to measure the same thing. JUND sits at the 75.4th percentile of 17,728 in this project's KMT2A-selectivity scan - the ANTI-selective half - at +0.002 in the 11 KMT2A-rearranged lines against -0.033 in 43 other leukaemia lines, i.e. not a fitness gene in KMT2A-rearranged cells at all. That is exactly what the Blood 2026 model predicts: JUND matters under menin inhibition, not at baseline. A CRISPR dropout screen run without the drug is the wrong instrument for a drug-induced dependency, and would have scored JUND as uninteresting.
menin binding motif
{
  "crystal_peptide": "SPGRLFPGAPPTAKK",
  "pdb": "3U86",
  "approximate_native_residues": "27-38",
  "core": "LFPG (approx. 31-34)",
  "note": "residue mapping is this entry's own alignment to P17535, not a depositor annotation"
}
PDB IDS
[
  "3U86"
]
comparator motif
{
  "partner": "KMT2A/MLL1 MBM1",
  "crystal_peptide": "SRWRFPARPGTTGGGGGGGRR",
  "pdb": "3U85",
  "core": "RWRFP (aa 6-10)"
}
localization
{
  "subcellular": "nucleus, chromatin-bound, as an AP-1 family transcription factor. The localisation result that matters in this graph is dynamic rather than static: menin-MLL inhibitor treatment of MLL-rearranged leukaemic cells INCREASES JUND chromatin occupancy, upregulating its target genes and contributing to resistance. Independently, MEN1 silencing in prostate cancer cells triggers nuclear translocation of JunD.",
  "subnuclear": "unpublished. No nuclear-body, speckle or other subnuclear compartment has been assigned to JUND in any source read for this beat; what is published is chromatin occupancy at specific loci (a genomic address) and a nuclear-versus-cytoplasmic partition, neither of which is a subnuclear compartment.",
  "how_shown": "chromatin binding assays plus RNA-seq in MLLr leukaemic cells under menin-MLL inhibitor, with quantitative mass spectrometry for the menin-JUND interaction (PMID 42371796); immunostaining and ChIP-qPCR for MEN1-knockdown-driven nuclear translocation (PMID 34446068)",
  "systems": "MLL-rearranged AML and ALL cells for the chromatin work - the index disease; AR-independent prostate cancer lines (DU145, PC3) for the translocation work",
  "node": "localization-chromatin-bound-versus-soluble-fraction-kmt2a"
}
cell type
{
  "required_in": "Haemogenic endothelium, in development. In zebrafish, deletion of the AP-1 family member jund impaired haematopoietic specification during the endothelial-to-haematopoietic transition, with the haemogenic endothelial cell retaining an ENHANCED endothelial identity instead of converting; mechanistically Jund and the haematopoietic transcription factor Hoxa9a jointly repress an endothelial dll4 enhancer through tight control of H3K27ac.",
  "reading": "This is a maintains-a-state result as much as a requirement: without jund the cell does not die, it stays the wrong kind of cell. It is also the only cell-type-level requirement anywhere in the JUND literature that is haematopoietic.",
  "how_shown": "cis-regulatory element and transcription-factor occupancy profiling across EHT, plus jund deletion, in zebrafish",
  "systems": "zebrafish embryonic haematopoiesis - a developmental cell type that does not exist postnatally",
  "nodes": [
    "cell-type-haemogenic-endothelium"
  ]
}
cell state
{
  "maintains_state": "The state under menin inhibition. Menin-MLL inhibitors disrupt the menin-JUND interaction as well as menin-MLL, and INCREASE JUND chromatin occupancy, upregulating JUND target genes and contributing to resistance; CRISPR inactivation of JUND or JNK inhibition synergises with menin-MLL inhibitors, reducing proliferation and causing cell-cycle arrest and apoptosis, and JUND-deficient leukaemia cells produce a lower tumour burden in engrafted immunocompromised mice.",
  "reading": "JUND does not define a baseline state in this disease. It defines a DRUG-INDUCED one - which is why a baseline fitness screen cannot see it.",
  "nodes": [
    "cell-state-under-menin-inhibition"
  ]
}
depmap kmt2a selectivity
{
  "chronos_mean_kmt2a_rearranged": 0.0018,
  "chronos_mean_other_leukaemia": -0.0331,
  "difference": 0.0348,
  "pan_essentiality_all_lines": -0.1571,
  "selectivity_rank": 13374,
  "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."
}
menin mbm1 kd conditions
{
  "_why": "Added by king sweep 2026-09-21 (Q82 corrections-owed #3). This entry quotes the menin-MBM1 53 nM figure. That number is NOT comparable to the KMT2A CXXC-DNA affinities without stating its buffer, and the omission is how a 1,300-fold 'discrepancy' was manufactured.",
  "value": "53 nM",
  "motif": "KMT2A MBM1, residues 4-15, FITC-labelled peptide at 15 nM",
  "method": "fluorescence polarisation",
  "buffer": "50 mM Tris, pH 7.5, 50 mM NaCl, 1 mM DTT",
  "nacl_mM": 50,
  "companion_itc": "72 nM by ITC, 50 mM phosphate pH 7.5, 50 mM NaCl, 1 mM beta-mercaptoethanol, 27 C",
  "source_pmid": "20961854",
  "source_quote": "\"in the FP buffer (50 mM Tris, pH 7.5, 50 mM NaCl, 1 mM DTT) used for binding experiments (Kd measurements)\" -- Grembecka 2010 JBC, Methods, fetched and read from PMC3003368 on 2026-09-21",
  "do_not_compare_naively": {
    "against": "KMT2A CXXC-DNA Kd 44 uM (PMID 20010842, Cierpicki 2010)",
    "because": "that measurement was made at 300 mM NaCl -- verbatim from its Methods, \"50 mM phosphate buffer (pH 7.1), 300 mM NaCl, 2 mM DTT at 25 C\" (PMC2908503, read 2026-09-21) -- SIX-FOLD more salt, on a salt-sensitive protein-DNA contact, against a hydrophobic peptide pocket that is not.",
    "best_matched_pair": "44 uM at 300 mM NaCl vs the 5.7 uM tagless CXXC measurement at 150 mM NaCl gives 108-fold against menin, not ~1,000-fold",
    "see": "ana-the-menin-affinity-comparison-was-made-at-one-sixth-the-salt"
  },
  "added": "2026-09-21"
}
cxxc dna kd conditions
{
  "_why": "Added by king sweep 2026-09-21 (cycle 46). This entry quotes the KMT2A CXXC-DNA 44 uM. That number is a property of five choices, not of the molecule, and quoting it bare is how this graph manufactured a 1,300-fold 'discrepancy' with the menin contact.",
  "value": "44 uM",
  "construct": "isolated CXXC domain, KMT2A 1147-1203, untagged, 15N-labelled",
  "ligand": "DNA pal oligonucleotide, one CpG element",
  "method": "NMR titration",
  "buffer": "50 mM phosphate, pH 7.1, 300 mM NaCl, 2 mM DTT, 25 C",
  "nacl_mM": 300,
  "source_pmid": "20010842",
  "source_quote": "\"Determination of the dissociation constants by NMR for the CXXC domain (MLL 1147-1203) and mutants was performed by titration of 15N-labeled protein with concentrated solutions of DNA pal oligonucleotide. Experiments were carried out in 50 mM phosphate buffer (pH 7.1), 300 mM NaCl, 2 mM DTT at 25 C.\" -- Cierpicki 2010 NSMB, Methods, fetched and read from PMC2908503 on 2026-09-21. The result sentence reads \"The wild type protein yielded a Kd of 44 uM\".",
  "this_is_the_high_salt_end": "300 mM NaCl is roughly twice physiological ionic strength and SIX TIMES the 50 mM the menin-MBM1 53 nM was measured in. The same domain measured tagless at 150 mM NaCl gives 5.7 uM -- 7.7-fold tighter from salt alone.",
  "do_not_compare_naively": {
    "against": "menin-MBM1 53 nM (PMID 20961854), measured at 50 mM NaCl by FP",
    "best_matched_pair": "108-fold, not ~1,000-fold",
    "see": "ana-the-menin-affinity-comparison-was-made-at-one-sixth-the-salt"
  },
  "added": "2026-09-21"
}
How this could bite us back
  • JunD displacement by a menin inhibitor is now demonstrated (PMID 33821572 by co-IP at 30-300 nM; PMID 42371796 by mass spectrometry in leukaemic cells), so the assumption this entry rested on is confirmed - but the same result removes the escape hatch: there is no reported menin-pocket ligand that spares JunD, and selectivity for the MLL interface may simply not be achievable with this chemotype.
  • The best-documented cost of JunD liberation is anti-leukaemic, not endocrine: released JUND increases chromatin occupancy, upregulates targets and contributes to menin-inhibitor resistance (PMID 42371796). A drug whose on-target action arms a resistance program is a bite-back in the strictest sense, and it argues for combining menin inhibition with JNK inhibition or JUND depletion rather than for withholding it.
  • JunD release inverts to growth promotion in fibroblasts over stable overexpression, not over months of partial occupancy in a differentiated endocrine cell. In the one in-vivo endocrine test published, 12 weeks of menin inhibitor produced compensatory beta-cell proliferation with improved glucose tolerance and no reported tumour (PMID 33821572). The inference from E3 is about direction, not magnitude or timescale, and the timescale evidence currently runs against alarm.
  • The beat's cheapest kill shot is gone. A negative co-IP would have retired the endocrine concern in a week; the co-IP is positive, so the remaining question is the slow one - what months to years of partial pocket occupancy does to a parathyroid chief cell - and no cheap experiment answers it.
  • The localisation change IS the resistance mechanism. Menin-MLL inhibitor treatment does not merely fail to move JUND, it actively increases JUND chromatin occupancy and upregulates its targets (PMID 42371796). A resistance screen that looks for loss of drug binding or for MEN1 mutation will not see this, because the drug is working exactly as designed on its own target while redistributing a second protein that shares the pocket.
  • RESISTANCE — This is the resistance mechanism itself. A menin inhibitor working exactly as designed on its own target simultaneously relocates JUND onto chromatin and drives resistance, so the therapeutic gain and the resistance route come from the same molecular event. Combining a menin inhibitor with JNK inhibition or JUND depletion is the proposed answer, and it has been shown to synergise in cell lines and xenografts but not in any patient.
Still open
  1. What is the Kd of the menin-JunD peptide interaction, and how does it compare with menin-MBM1 (about 53 nM in this graph's interface entry)? If JunD binds an order of magnitude more weakly, a drug at a MBM1-blocking concentration would displace it fully and the concern grows; if more tightly, the drug may spare it.
  2. Is JunD displaced from menin in cells by revumenib, ziftomenib or bleximenib specifically, at clinical free concentrations? Displacement is established for the tool compounds MI-503 and MI-463 by co-IP (PMID 33821572) and for menin-MLL inhibitors as a class by quantitative mass spectrometry in leukaemic cells (PMID 42371796), so the remaining gap is a compound-by-compound potency ratio, not the existence of the effect.
  3. What does JunD do in the human parathyroid chief cell? The switch from suppressor to promoter was shown in fibroblasts; the MEN1 target tissues have never been tested.
  4. Do the MEN1 missense mutations that disrupt menin-JunD map to the same pocket residues that a menin inhibitor contacts? The structures exist and the comparison has not been published.
  5. JUND and HOXA9 cooperate to repress an endothelial enhancer in zebrafish EHT, and HOXA9 is the central output of KMT2A::AFF1. Do JUND and HOXA9 co-occupy sites in a KMT2A-rearranged blast, and does the increased JUND occupancy under menin inhibition land on HOXA9-cobound elements? The reagents for this exist in both papers.
  6. The resistance route described here would be invisible to every screen the field runs: no loss of drug binding, no MEN1 mutation, and no baseline fitness signal. How many other menin-pocket occupants relocate under inhibition, and has anyone done the unbiased chromatin-occupancy experiment across the whole pocket clientele rather than one factor at a time?
Every source behind this entry (10)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    A yeast two-hybrid screen using menin as bait identified JunD as a direct menin-interacting partner; menin did not interact directly with other Jun or Fos family members, menin repressed JunD-mediated transcriptional activation from both a Gal4-responsive and an AP-1-responsive reporter, and several naturally occurring clustered MEN1 missense mutations disrupted the menin-JunD interaction.

    Menin interacts with the AP1 transcription factor JunD and represses JunD-activated transcription · Agarwal SK, Guru SC, Heppner C, Erdos MR, Collins RM, Park SY, Saggar S, Chandrasekharappa SC, Collins FS, Spiegel AM, Marx SJ, Burns AL · Cell 96:143-52 1999
    PMID 9989505DOI
    What this evidence cannot show

    Read as abstract. Overexpression and reporter-based assays in HeLa cells; the specificity claim (JunD alone among Jun/Fos members) rests on direct-binding assays and does not exclude indirect association with other AP-1 subunits in cells. The paper predates the crystal structures and does not localise the interaction to a pocket.

    Evidence that argues against this (1)
    • A yeast two-hybrid screen establishes a binary interaction in yeast and the specificity claim - no direct interaction with other Jun or Fos proteins - is a negative result in the same system, where false negatives are common. The interaction is now structurally confirmed, so the screen's value is historical rather than evidential. Cell 1999
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Menin inhibits ERK-dependent phosphorylation and activation of JunD and of Elk-1, represses inducible c-fos promoter activity, and inhibits JNK-mediated phosphorylation of JunD and c-Jun, acting downstream of MAPK activation rather than by blocking ERK2 or JNK1 activation itself; an N-terminal deletion mutant of menin that cannot block JNK-mediated JunD phosphorylation still represses ERK2-mediated phosphorylation, indicating two distinct inhibitory mechanisms.

    Menin uncouples Elk-1, JunD and c-Jun phosphorylation from MAP kinase activation · Gallo A, Cuozzo C, Esposito I, Maggiolini M, Bonofiglio D, Vivacqua A, Garramone M, Weiss C, Bohmann D, Musti AM · Oncogene 21:6434-45 2002
    PMID 12226747DOI
    What this evidence cannot show

    Read as abstract. Overexpression studies in HeLa cells; the separation into two mechanisms rests on a single deletion mutant. Does not address whether a small-molecule pocket ligand reproduces either effect.

    Evidence that argues against this (1)
    • Menin uncoupling Elk-1, JunD and c-Jun phosphorylation from MAPK activation is an overexpression result in cell lines, and the entry uses it to characterise endogenous menin function. It also complicates the JunD story this beat tells: if menin acts on several AP-1 components and on Elk-1, then displacing menin pharmacologically perturbs more than JunD alone. Oncogene 2002
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Stable overexpression of wild-type mouse JunD in JunD-null immortalised fibroblasts inhibited proliferation and reverted the transformed-like phenotype, whereas a JunD missense mutant with disabled menin binding had the opposite, growth-promoting effect; wild-type JunD overexpressed in Men1-null immortalised fibroblasts enhanced their transformed-like characteristics. JunD changes from growth suppressor to growth promoter when its binding to menin is prevented.

    Transcription factor JunD, deprived of menin, switches from growth suppressor to growth promoter · Agarwal SK, Novotny EA, Crabtree JS, Weitzman JB, Yaniv M, Burns AL, Chandrasekharappa SC, Collins FS, Spiegel AM, Marx SJ · Proc Natl Acad Sci U S A 100:10770-5 2003
    PMID 12960363DOI
    What this evidence cannot show

    Read as abstract. Immortalised mouse fibroblasts with overexpression, not endocrine cells at physiological expression, and the menin-null background is a complete genetic loss rather than pharmacological occupancy. The direction of the effect is nonetheless the one that matters: removing menin from JunD does not merely de-repress, it inverts the sign.

    Evidence that argues against this (2)
    • The pharmacological displacement question is no longer open, and the answer relocates the consequence away from endocrine tissue. Ozyerli-Goknar et al. show by quantitative mass spectrometry in MLL-rearranged leukaemic cells that menin-MLL inhibitors disrupt menin-JUND interactions; the released JUND gains chromatin occupancy, upregulates its target genes and CONTRIBUTES TO RESISTANCE to menin inhibition. CRISPR inactivation of JUND or JNK inhibition synergised with menin-MLL inhibitors, and JUND-deficient leukaemia cells gave reduced tumour burden in mice. So JunD liberation by a clinical-class menin inhibitor is documented - but the first-order consequence reported is a leukaemia-intrinsic escape route, not an endocrine one. This entry's framing, that JunD displacement matters mainly as an endocrine liability, is therefore incomplete: on current evidence the better-supported consequence is therapeutic failure in the marrow. Ozyerli-Goknar E et al. 2026
    • A direct co-immunoprecipitation of menin with JunD under a menin inhibitor was published in 2021 and this entry did not have it. Ma et al. show that MI-503 and MI-463 inhibit menin-JunD binding in a dose-dependent manner in HEK293T co-IP at 30, 100 and 300 nM, and that the JunD/menin/HDAC3 complex represses Pbk. Critically for the endocrine argument, the downstream effect in the relevant tissue was BENIGN AND BENEFICIAL, not neoplastic: menin inhibitor at 70 mg/kg daily by gavage increased beta-cell proliferation, lowered fasting glucose and improved glucose tolerance in high-fat-diet mice from week 6 through week 12, and MI-503 at 10-100 nM for 5 days raised Ki67 in primary HUMAN islets. Menin-inhibitor-driven JunD release in an endocrine cell has been observed and its published phenotype is compensatory proliferation with improved glycaemia over 12 weeks, with no islet tumour reported. Ma J et al. 2021
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProtKB entry P17535 (JUND_HUMAN) has primary gene name JUND, protein name transcription factor JunD, length 347 amino acids and nuclear subcellular location.

    UniProtKB P17535 (JUND_HUMAN), Transcription factor JunD · UniProt Consortium · UniProtKB/Swiss-Prot 2026
    What this evidence cannot show

    Retrieved from the UniProt REST API on 2026-09-18 and the primary gene name read back explicitly to guard against the synonym-match failure mode that resolved TCF3 to TCF7L1 elsewhere in this project. No synonyms are listed for this entry, so there is no collision risk.

    Evidence that argues against this (1)
    • A UniProt record is a curated model. The subcellular annotation for JUND is inferred rather than experimentally determined for every isoform, and JunD localisation and activity are regulated, so a static nuclear annotation does not characterise its behaviour under menin displacement. Cell 1999
  5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    PDB entry 3U86 is the X-ray crystal structure at 2.84 A of human menin in complex with a 15-residue human JunD peptide of sequence SPGRLFPGAPPTAKK.

    PDB 3U86: Crystal structure of human menin in complex with JunD · Huang J, Gurung B, Wan B, Matkar S, Veniaminova NA, Wan K, Merchant JL, Hua X, Lei M · RCSB Protein Data Bank 2012
    DOI
    What this evidence cannot show

    Entity sequences and resolution retrieved from the RCSB data API on 2026-09-18. The mapping of the peptide onto JunD residues 27-38 is this entry's own alignment against UniProt P17535, not a statement from the depositors; the peptide's terminal KK does not match the contiguous native sequence and appears to be a construct feature. Interface residue lists were not extracted.

    Evidence that argues against this (1)
    • A 2.84 A peptide-bound crystal structure establishes the binding mode of a 15-residue peptide, not cellular competition. It does not give an affinity, a stoichiometry, or the fraction of menin engaged by JunD in a cell, all of which the displacement argument requires. PNAS 2003
  6. E6established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    PDB entry 3U85 is the X-ray crystal structure at 3.0 A of human menin in complex with a 21-residue MLL1/KMT2A peptide of sequence SRWRFPARPGTTGGGGGGGRR, corresponding to the MBM1 motif.

    PDB 3U85: Crystal structure of human menin in complex with MLL1 · Huang J, Gurung B, Wan B, Matkar S, Veniaminova NA, Wan K, Merchant JL, Hua X, Lei M · RCSB Protein Data Bank 2012
    DOI
    What this evidence cannot show

    Entity sequences and resolution retrieved from the RCSB data API on 2026-09-18. The construct lacks the native Cys5 present in full-length KMT2A, so the peptide is not a verbatim native fragment; RWRFP maps to KMT2A residues 6-10.

    Evidence that argues against this (1)
    • Same limitation for the MLL1 peptide structure at 3.0 A, and the pair together establish that the two peptides use one pocket - which is the premise of the displacement concern, not evidence that displacement occurs at therapeutic exposure. PNAS 2003
  7. E7strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In MLL-rearranged leukaemic cells, menin-MLL inhibitors disrupt the menin-JUND interaction as well as menin-MLL, and menin-MLL inhibitor treatment INCREASES JUND chromatin occupancy with upregulation of JUND target genes, contributing to resistance; CRISPR inactivation of JUND or JNK inhibition synergises with the menin inhibitor.

    Menin-MLL inhibitors enhance JUND activity in MLLr leukemic cells contributing to tumorigenesis and therapy resistance. · Ozyerli-Goknar E, Nizamuddin S, Fetsch V, Biniossek ML, Hartmann A, Perner F, McGeehan GM, Armstrong SA, Zeiser R, Timmers HTM · Blood 2026
    PMID 42371796DOI
    What this evidence cannot show

    The system is MLL-rearranged AML and ALL, which is the index disease, but the abstract does not separate which fusion partners were used. 'Chromatin binding assays' is not further specified in the abstract and the supporting figures were not opened for this beat, so per brief rule 80 the assay identity is bounded by the abstract. This is a 2026 paper and has not been independently replicated.

    Evidence that argues against this (1)
    • The prior model predicts the opposite. Menin was shown to repress JunD-mediated transcription by recruiting an mSin3A-histone deacetylase complex - a model in which JunD is ALREADY on DNA and menin supplies the repressor. On that account displacing menin should change JUND's activity without changing its occupancy, so the measured increase in chromatin binding is a genuine departure from the established mechanism and not a confirmation of it. Kim H et al. 2003
  8. E8moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    Loss of menin drives JunD into the nucleus: MEN1 knockdown in AR-independent prostate cancer cells triggers nuclear translocation of JunD together with beta-catenin, with augmented JunD binding at the MYC 5' enhancer by ChIP and increased enhancer-promoter looping by 3C.

    MEN1 silencing aggravates tumorigenic potential of AR-independent prostate cancer cells through nuclear translocation and activation of JunD and β-catenin. · Luo Y, Vlaeminck-Guillem V, Baron S, Dallel S, Zhang CX, Le Romancer M · J Exp Clin Cancer Res 2021
    PMID 34446068DOI
    What this evidence cannot show

    Prostate cancer, not leukaemia (brief rule 73): the first clause of any downstream claim must say so. Genetic MEN1 knockdown is not pharmacological menin-MLL interface inhibition, and the two need not have the same effect on a partner that binds the same pocket. What makes it worth recording is that a second, independent system produces the same direction of travel as the leukaemia result - menin down, JUND further onto chromatin.

    Evidence that argues against this (1)
    • The leukaemia result and the prostate result agree in direction but were produced by different perturbations - a small-molecule interface inhibitor versus shRNA knockdown of the whole protein - and menin has interface-independent functions. Agreement between them is therefore weaker evidence than it looks; the prostate study cannot corroborate a claim about the menin-MLL POCKET. Ozyerli-Goknar E et al. 2026
  9. E9strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Menin-MLL inhibitors disrupt the menin-JUND interaction as well as menin-MLL in MLL-rearranged leukaemic cells and INCREASE JUND chromatin occupancy, upregulating JUND target genes and contributing to resistance; CRISPR inactivation of JUND or pharmacological JNK inhibition synergistically enhances the anti-leukaemic effect of menin-MLL inhibitors with reduced proliferation, cell-cycle arrest and apoptosis, and mice engrafted with JUND-deficient leukaemia cells show reduced tumour burden.

    Menin-MLL inhibitors enhance JUND activity in MLLr leukemic cells contributing to tumorigenesis and therapy resistance. · Ozyerli-Goknar E, Nizamuddin S, Fetsch V, Biniossek ML, Hartmann A, Perner F, McGeehan GM, Armstrong SA, Zeiser R, Timmers HTM · Blood 2026
    PMID 42371796DOI
    What this evidence cannot show

    MLLr leukaemic cell lines, quantitative mass spectrometry, RNA-seq, chromatin binding assays and immunocompromised-mouse engraftment. MLLr AML and ALL are discussed together and the abstract does not resolve which fusion partners the cell lines carry, so this cannot be attributed to KMT2A::AFF1 specifically, and there is no infant or primary-patient arm. Read: abstract only. Re-fetched from PubMed for this entry rather than copied from the sibling localisation evidence item.

    Evidence that argues against this (1)
  10. E10moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In zebrafish, the AP-1 family member jund is required for the endothelial-to-haematopoietic transition: jund deletion impairs haematopoietic specification and the haemogenic endothelial cell instead retains an enhanced endothelial identity, with Jund and the haematopoietic transcription factor Hoxa9a collectively repressing an endothelial dll4 enhancer through control of H3K27ac.

    PMID 40472028DOI
    What this evidence cannot show

    Zebrafish embryonic haematopoiesis. The cell type is developmental and transient and has no postnatal counterpart, and AP-1 family redundancy in zebrafish is not addressed by a single jund deletion. The Hoxa9a collaboration is intriguing in a graph about a HOXA-driven disease, but hoxa9a is the zebrafish orthologue in a developmental context, not HOXA9 in a blast. Read: abstract only. This is a cell-TYPE claim and carries no implication for the menin-inhibition cell state.

    Evidence that argues against this (1)
    • The developmental requirement and the leukaemic drug-induced requirement point in opposite directions for therapy: in the embryo, losing jund leaves a cell stuck in the wrong identity, while in MLL-rearranged leukaemia losing JUND is the desirable outcome and sensitises to menin inhibition. Nothing connects the two, and a JUND-directed agent would be acting on a factor whose only documented normal haematopoietic role is in a lineage-specification step. Ozyerli-Goknar E et al. 2026
Connected to