Oncofusion

start / graph / gene-setd2

Genemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q84-pan-fusion-core

SETD2 (Q9BYW2) - the H3K36 trimethyltransferase that binds all seven MLL fusions, and the only one of them with a published KMT2A::AFF1 dependency test

also called SETD2 · HYPB · KMT3A · HIF-1 · SET2 · KIAA1732

SETD2 is the enzyme that writes a chemical tag (H3K36me3) along the body of genes that are being read, and it sticks to every version of the MLL fusion protein that has been tested. Switching it off slows leukaemia cells carrying the exact fusion this project studies, yet other groups find leukaemias often break this same gene, so its role cuts both ways.

The same thing, for a scientist

SETD2 (Q9BYW2, 2,564 aa) is the sole mammalian H3K36 trimethyltransferase. In Skucha 2018 it co-purified with all seven Strep-HA-tagged MLL fusions expressed as single-copy doxycycline-inducible transgenes in isogenic Jurkat cells; the MLL-AF4 (= KMT2A::AFF1) column gives p = 0.0372 with 3|4 spectral counts across the two replicates, the weakest of its seven per-bait calls. It is the paper's headline finding, and uniquely among the twelve proteins in this beat it was carried into functional work: SETD2 knockdown was antiproliferative and pro-apoptotic in the two human t(4;11) KMT2A::AFF1 lines MV4-11 and SEM as well as in MLL-AF9 MOLM-13, while sparing MLL-wild-type K562 and HL-60, and it synergised with the DOT1L inhibitor EPZ5676 in MV4-11. The orthogonal co-immunoprecipitation mapping the contact to the MLL N-terminus and the SETD2 C-terminus was performed in transiently transfected HEK293, not in leukaemic cells. The dependency claim is contested: SETD2 loss-of-function mutations recur in MLL-rearranged acute leukaemia and behave as a haploinsufficient tumour suppressor, so the same gene is reported as both required and deleted in this disease.

Go deeper
What SETD2 does, independent of any fusion

SETD2 (originally HYPB) is the enzyme responsible for all histone H3 lysine-36 trimethylation in mammalian cells; H3K36me3 accumulates co-transcriptionally over gene bodies, peaking toward the 3' end, and is deposited on genes as they are induced 3. That places SETD2 on transcribed chromatin rather than at promoters - the opposite end of the gene from the H3K4me3 the KMT2A SET domain writes, and the KMT2A SET domain is the part the fusion loses.

The Skucha row, exactly as the table gives it

Supplementary Table 5 of Skucha 2018 records SETD2 (Q9BYW2) as interacting with 7 of 7 MLL fusions. Per-bait combined p-values and spectral counts: MLL-AF1p 0.00119 (12|6), MLL-AF4 0.0372 (3|4), MLL-AF9 2.74e-5 (28|23), MLL-CBP 9.84e-5 (12|15), MLL-EEN 9.48e-5 (30|25), MLL-ENL 1.41e-7 (36|36), MLL-GAS7 2.19e-4 (30|20); negative control 0|0 1. The MLL-AF4 call is the weakest of the seven by more than two orders of magnitude and rests on three and four spectra. That is the honest reading, and it is why the index-fusion knockdown data in the same paper matter more here than the table row does.

The part that is actually about KMT2A::AFF1

Skucha 2018 extended the shRNA screen beyond MLL-AF9 MOLM-13 to MV4-11 and SEM, both human t(4;11) KMT2A::AFF1 lines, and reports significant antiproliferative responses and induction of apoptosis on SETD2 knockdown in both, with no comparable effect in MLL-wild-type K562 and HL-60. In MV4-11, SETD2 knockdown combined with the clinical DOT1L inhibitor EPZ5676 potentiated growth inhibition, apoptosis and myeloid differentiation 2. This is the only entry in this beat whose interaction is followed by a functional experiment in the index fusion; every other protein here stops at the table row.

Why the dependency is contested

A separate line of work finds SETD2 acting as a tumour suppressor in the same disease. Whole-genome sequencing of monozygotic twins discordant for KMT2A-rearranged leukaemia found biallelic SETD2 mutations, and loss-of-function SETD2 point mutations recurred in 6.2% of 241 acute leukaemia patients with global loss of H3K36me3 4. The same group later reported SETD2 loss of function in 22% of MLL-rearranged patients and showed that SETD2 inactivation accelerates MLLr leukaemia by further elevating H3K79me2 and deregulating a set of genes distinct from the canonical MLL-fusion targets 5. A 2026 study genotyped loss-of-function mutations to define SETD2 as a haploinsufficient tumour suppressor in this disease while simultaneously reporting that the SETD2 inhibitor EZM0414 is synergistic with revumenib, and states the double-sided role explicitly 6. A reader of Skucha alone would conclude SETD2 is required; a reader of Zhu and Bu alone would conclude it is deleted. Both are published and both are in this entry.

Localization, and what the database does and does not say

UniProt Q9BYW2 gives Nucleus and Chromosome - both carrying ECO:0000250 (by sequence similarity) only, with no experimental evidence code - which is unusual for a protein of this prominence and means the database is an annotation rather than a measurement here. The localization that is measured is indirect: the distribution of the mark it writes, H3K36me3, across gene bodies of induced genes in murine fibroblasts 3, plus Skucha's own subcellular fractionation showing all seven MLL fusion baits in the nuclear fraction 2. No published experiment places SETD2 in a named subnuclear compartment - nucleolus, speckle or PML body - in a KMT2A-rearranged cell.

Facts and figures
uniprotQ9BYW2
length AA2564
accession resolutionResolved live 2026-09-21 via gene_exact:SETD2 AND organism_id:9606 AND reviewed:true, returning exactly one row whose gene_primary is SETD2: Q9BYW2, SETD2_HUMAN, 2,564 aa, synonyms HIF1 HYPB KIAA1732 KMT3A SET2. The accession printed in Skucha's Supplementary Table 5 is the same string, and this resolution is what verifies it.
localizationNucleus
subnuclear localizationGene bodies of transcribed genes, inferred from the distribution of the mark it writes (H3K36me3 over coding regions, 3'-skewed). No published experiment assigns SETD2 to a named subnuclear body in any KMT2A-rearranged cell.
localization evidenceUniProt Q9BYW2 gives Nucleus and Chromosome with ECO:0000250 (by similarity) ONLY - no ECO:0000269 experimental code on either. Fetched 2026-09-21. Treat as an annotation, not a measurement.
shown for fusionKMT2A::AFF1 (MLL-AF4) as one of seven AP-MS baits in Jurkat, AND functionally in the KMT2A::AFF1 lines MV4-11 and SEM; also MLL-AF1p, MLL-AF9, MLL-CBP, MLL-EEN, MLL-ENL and MLL-GAS7 by co-purification, and MLL-AF9 functionally in MOLM-13 and in mouse.
graph contextEighteen entries in this graph mentioned SETD2 before this one and none was a gene entry; they reach it through the menin-gammaH2A.X-SETD2-H3K36me3-LEDGF chain and through mark-h3k36me2-3. Measured 2026-09-21 by grep -rl SETD2 over data/graph/nodes (18 files) and by listing all 1,886 node ids (no gene-setd2).
skucha 2018
{
  "bait": "MLL-AF4 (= KMT2A::AFF1)",
  "p_value": 0.0372,
  "spectral_counts": "3|4",
  "n_fusions_of_seven": 7,
  "negative_control_counts": "0|0",
  "per_bait": {
    "MLL-AF1p": "0.00119 (12|6)",
    "MLL-AF4": "0.0372 (3|4)",
    "MLL-AF9": "2.74e-05 (28|23)",
    "MLL-CBP": "9.84e-05 (12|15)",
    "MLL-EEN": "9.48e-05 (30|25)",
    "MLL-ENL": "1.41e-07 (36|36)",
    "MLL-GAS7": "0.000219 (30|20)"
  },
  "source": "analyses/skucha-mll-fusion-interactome/results.json, re-derived from the cached MOESM4 xlsx on 2026-09-21"
}
cell types and states
[
  "Jurkat (human T-ALL) - AP-MS host, nuclear lysate",
  "MV4-11 (human t(4;11) KMT2A::AFF1) - shRNA knockdown: antiproliferative, apoptosis, differentiation, EPZ5676 synergy",
  "SEM (human t(4;11) KMT2A::AFF1 B-ALL) - shRNA knockdown: antiproliferative, apoptosis",
  "MOLM-13 (human MLL-AF9) - shRNA knockdown, antiproliferative",
  "K562, HL-60 (MLL-wild-type) - negative controls, spared",
  "mouse MLL-AF9/NrasG12D - differentiation and DNA damage",
  "HEK293, transiently transfected - the co-IP that maps the interaction domains"
]
bounded literature
{
  "query": "SETD2[tiab] AND (KMT2A[tiab] OR MLL[tiab] OR \"MLL-AF4\"[tiab] OR \"MLL-AF9\"[tiab])",
  "querytranslation": "\"SETD2\"[Title/Abstract] AND (\"KMT2A\"[Title/Abstract] OR \"MLL\"[Title/Abstract] OR \"MLL-AF4\"[Title/Abstract] OR \"MLL-AF9\"[Title/Abstract])",
  "count": 39,
  "date": "2026-09-21",
  "tool": "NCBI esearch db=pubmed",
  "positive_control": "the identical query shape with MEN1 in place of SETD2 returns 85",
  "reading": "Unlike the other eleven proteins in this beat, SETD2 has a substantial independent KMT2A-fusion literature - and it points in two directions at once."
}
functionally screened in the source paper
{
  "_added": "king, cycle 48, 2026-09-21, after Q86 flagged that cycle-47 entries understate the record. Verified by me directly from the supplementary file, not from the paper's prose.",
  "screened": true,
  "library_symbol": "SETD2",
  "shrnas_for_this_gene": 6,
  "design": "Supplementary Table 6 / MOESM5 is 768 data rows, 128 unique gene symbols, EXACTLY 6 shRNAs each -- counted with openpyxl on 2026-09-21. Negative-selection RNAi screen in MLL-AF9-expressing MOLM-13.",
  "what_this_DOES_establish": "Every one of the 128 conserved MLL-fusion interactors, this protein included, was functionally tested against an MLL fusion. An entry saying 'nothing in the paper validates this prey' is true of BINDING and understates the record.",
  "what_this_does_NOT_establish": "A per-gene result. Scores are presented as heatmaps (Fig. 2c/2d) and no per-gene depletion value is tabulated, so for any individual gene the outcome is measured-not-reported. It is also MLL-AF9 MOLM-13, a myeloid line, not the index fusion and not B-lineage.",
  "absence_class_for_the_per_gene_score": "measured-not-reported"
}
How this could bite us back
  • SETD2 is reported as both a dependency and a haploinsufficient tumour suppressor in MLL-rearranged leukaemia. A SETD2 inhibitor given to a patient whose blasts already carry a SETD2 loss-of-function allele may be pushing in the direction the tumour has already selected for.
  • SETD2 is the sole source of H3K36me3 genome-wide, so inhibiting it is not a fusion-selective action and carries the essential-function toxicity of removing a mark that guides mismatch repair and homologous recombination.
  • The proposed route pairs SETD2 inhibition with menin inhibition, and this graph already carries open questions about HR competence under menin inhibition. A combination that increases DNA damage on both arms needs a genotoxicity readout before it is called synergy.
  • The index-fusion evidence is knockdown, not inhibitor. shRNA removes the protein; EZM0414 removes the catalytic activity. If the fusion interaction is scaffolding rather than catalytic, the drug will not phenocopy the knockdown.
Still open
  1. Does SETD2 co-purify with the ENDOGENOUS KMT2A::AFF1 fusion in a t(4;11) B-ALL line such as SEM or RS4;11, rather than with a tagged transgene in Jurkat?
  2. Is the MLL-N-terminus to SETD2-C-terminus contact direct, and does it survive menin inhibition? The HEK293 co-IP can answer neither.
  3. How do the 22% SETD2 loss-of-function mutations in MLL-rearranged patients distribute across fusion partners, and is any of them in an infant t(4;11) case?
  4. Is the reported EZM0414 plus revumenib synergy reproducible in a KMT2A::AFF1 B-lineage model, given that SETD2 loss is reported to accelerate the same disease?
Every source behind this entry (6)
  1. E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In an affinity-purification survey of seven Strep-HA-tagged MLL fusions in isogenic Jurkat cells, SETD2 co-purified with all seven, including MLL-AF4 (= KMT2A::AFF1) at combined p = 0.0372 with 3|4 spectral counts across replicates.

    MLL-fusion-driven leukemia requires SETD2 to safeguard genomic integrity. · Skucha A, Ebner J, Schmoellerl J, Roth M, Eder T, Cesar-Razquin A, Stukalov A, Vittori S, Muhar M, Lu B, Aichinger M, Jude J, Mueller AC, Gyorffy B, Vakoc CR, Valent P, Bennett KL, Zuber J, Superti-Furga G, Grebien F · Nature Communications 2018
    PMID 29777171DOI
    What this evidence cannot show

    SYSTEM: the AP-MS host is JURKAT, a T-ALL line, whereas KMT2A::AFF1 disease is B-lineage infant leukaemia, so a partner that needs a B-cell or infant proteome cannot appear here and a T-lineage partner can appear spuriously. The bait is a Strep-HA-tagged, SINGLE-COPY, doxycycline-INDUCED transgene, not the endogenous fusion. Prey was recovered from bulk nuclear lysate, so co-purification is not evidence of a direct contact. INCLUSION RULE: the paper kept the top 300 interactions per bait by combined p-value rather than applying a p threshold (Methods, verbatim: "The 300 most significant interactions per bait were retained"). Measured by this beat by reading the cached Supplementary Table 5 xlsx directly on 2026-09-21: 545 of the 2,049 per-bait p-values in the table (26.6%) exceed 0.05, to a maximum of 0.175. This graph's own king analysis ana-the-index-fusion-interactome-is-not-distinctive-and-two-earlier-answers-were-artifacts found the MLL-AF4 column reaches deepest into the weak tail of the seven, so an MLL-AF4 p-value is systematically weaker than the same protein's p-value under another bait and should not be read as a weaker interaction. The MLL-AF4 value is the weakest of SETD2's seven per-bait p-values by more than two orders of magnitude and rests on 3 and 4 spectra; read alone it would be a marginal call. Values taken from analyses/skucha-mll-fusion-interactome/results.json and re-read from the cached Supplementary Table 5 xlsx on 2026-09-21, not from memory.

    Evidence that argues against this (1)
    • IntAct has never curated this dataset (pubid:29777171 returns 0 rows against a working positive control), so no independent curator has re-examined these calls, and the raw data at ProteomeXchange PXD009338 have not been reanalysed here. Nothing in the paper validates this particular prey by an orthogonal assay: of the twelve proteins in this beat, only SETD2 is mentioned anywhere in the paper's body text (counted by regex over the PMC full text, 2026-09-21; SF3A3 appears once, as a named non-selective control). ProteomeXchange/PRIDE 2018
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In the same study, SETD2 knockdown was antiproliferative and pro-apoptotic in the human t(4;11) KMT2A::AFF1 cell lines MV4-11 and SEM and in MLL-AF9 MOLM-13, while sparing MLL-wild-type K562 and HL-60, and in MV4-11 it potentiated the effects of the DOT1L inhibitor EPZ5676.

    MLL-fusion-driven leukemia requires SETD2 to safeguard genomic integrity. · Skucha A, Ebner J, Schmoellerl J, Roth M, Eder T, Cesar-Razquin A, Stukalov A, Vittori S, Muhar M, Lu B, Aichinger M, Jude J, Mueller AC, Gyorffy B, Vakoc CR, Valent P, Bennett KL, Zuber J, Superti-Furga G, Grebien F · Nature Communications 2018
    PMID 29777171DOI
    What this evidence cannot show

    In human MV4-11 and SEM cells - both established t(4;11) KMT2A::AFF1 lines, neither an infant primary sample - plus MLL-AF9 MOLM-13 and mouse MLL-AF9/NrasG12D cells. shRNA knockdown, not genetic deletion, so residual SETD2 is present. The orthogonal co-immunoprecipitation that maps the contact to the MLL N-terminus and the SETD2 C-terminus was done in TRANSIENTLY TRANSFECTED HEK293 cells with MG-132 present (paper's Methods), so the physical interaction is supported in two heterologous systems, Jurkat and HEK293, and in neither at endogenous stoichiometry in a B-lineage blast. Full text read from PMC5959866 on 2026-09-21.

    Evidence that argues against this (1)
    • SETD2 inactivation has been reported to ACCELERATE MLL-rearranged leukaemia rather than block it, which is the opposite sign to a dependency. Bu J et al. 2018
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    SETD2 (HYPB) is responsible for all histone H3 lysine-36 trimethylation in mammalian cells, and H3K36me3 is restricted to the coding regions of genes and increases across them transcription-dependently on induction.

    PMID 18157086DOI
    What this evidence cannot show

    Murine fibroblasts, c-fos and c-jun as the induced model genes; not a leukaemia system and not a fusion system. The 'all' claim is about H3K36 TRImethylation specifically - H3K36me1/me2 are deposited by NSD-family enzymes and are unaffected.

    Evidence that argues against this (1)
    • H3K36 methyltransferase activity is itself regulated by other marks, so 'SETD2 makes all H3K36me3' is a statement about the enzyme, not about the mark being unconditioned. Yuan G et al. 2013
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Loss-of-function SETD2 point mutations recur in human acute leukaemia - 6.2% of 241 patients, with global loss of H3K36me3 in mutant blasts - and biallelic SETD2 mutation co-occurred with a KMT2A rearrangement in monozygotic twins discordant for the disease.

    Identification of functional cooperative mutations of SETD2 in human acute leukemia. · Zhu X, He F, Zeng H, Ling S, Chen A, et al. · Nat Genet 2014
    PMID 24509477DOI
    What this evidence cannot show

    Patient sequencing in a mixed acute leukaemia cohort rather than infant t(4;11) specifically; the twin pair carried MLL-NRIP3, not KMT2A::AFF1. Read from the abstract, not the full text.

    Evidence that argues against this (1)
    • Skucha 2018 reports the opposite phenotype - SETD2 knockdown suppresses MLL-fusion leukaemia cell growth. Both results are in the record and this entry does not resolve them. Skucha A et al. 2018
  5. E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    SETD2 loss-of-function mutations were found in 22% of MLL-rearranged acute leukaemia patients, and SETD2 inactivation globally reduces H3K36me3, further elevates H3K79me2, and deregulates genes distinct from the canonical MLL-fusion targets.

    SETD2-mediated crosstalk between H3K36me3 and H3K79me2 in MLL-rearranged leukemia. · Bu J, Chen A, Yan X, He F, Dong Y, et al. · Leukemia 2018
    PMID 29249820DOI
    What this evidence cannot show

    MLL-rearranged as a class; the abstract does not break the 22% down by partner gene, so the fraction attributable to KMT2A::AFF1 specifically is not stated. Read from the abstract, not the full text. Note the direct collision with E2 - same mark, same disease class, opposite conclusions about whether SETD2 is required.

    Evidence that argues against this (1)
    • Skucha 2018's MV4-11 and SEM knockdowns are in KMT2A::AFF1 lines and show a growth requirement, which this paper's model does not predict. Skucha A et al. 2018
  6. E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    SETD2 behaves as a haploinsufficient tumour suppressor in MLL-fusion leukaemia by loss-of-function genotyping, and yet the SETD2 inhibitor EZM0414 is reported synergistic with the menin inhibitor revumenib against MLL-fusion and NPM1-mutant leukaemia, attenuating both H3K36me3 over gene bodies and H3K4me3 at promoters.

    Blocking SETD2 Enhances the Therapeutic Efficiency of Menin Inhibitor in MLL-Fusion Leukemia. · Li A, Zhang Y, Li Z, Liu A, Zhang L, Gao S, et al. · Cancer Sci 2026
    PMID 42170936DOI
    What this evidence cannot show

    Read from the abstract only. The cell lines, the fusion partners tested and whether any KMT2A::AFF1 model is included are NOT stated in the abstract and were not verified; the abstract does not name a single cell line. The paper's own framing - 'the role of SETD2 in MLL-fusion leukemia has remained controversial over the past decade' - is the correct status of E2 against E4 and E5. A synergy claim of this importance to this project must be re-read in full text before it is relied on.

    Evidence that argues against this (1)
    • A synergy reported in an abstract with no stated model system is the weakest form of this claim, and the direction of the combination - inhibit SETD2 plus inhibit menin - is hard to reconcile with SETD2 loss accelerating the same disease. Bu J et al. 2018
Connected to