KMT2A::USP8, t(11;15)(q23;q21) - three patients, and the second instance of a deubiquitinase partner
also called MLL-USP8 · KMT2A-USP8 · KMT2A::USP8 · t(11;15)(q23;q21)
A very rare leukaemia fusion in which KMT2A joins USP8, an enzyme that removes destruction tags from proteins. Only three patients are known, but they matter because they repeat the pattern seen with the much commoner USP2 fusion.
The same thing, for a scientist
KMT2A::USP8 is a balanced t(11;15)(q23;q21) that breaks in the KMT2A minor BCR and fuses the conserved UCH catalytic domain of USP8 to an extended 5'-KMT2A portion retaining the PHD1-3 cluster, bromodomain and ePHD4 [E1][E2]. It was described as a novel KMT2A partner in the 2019 minor-BCR discovery cohort (one patient, one of four balanced translocations in BCR2) [E1] and counts 3 patients in 3401 in the 2023 recombinome [E2]. A detailed independent case is a 16-month-old boy with B-ALL, trisomy 8, an atypical KMT2A FISH result and a focal 15q deletion with a breakpoint inside USP8, resolved only by mate-pair sequencing [E3]. Its significance is structural rather than epidemiological: with USP2 it forms the two-gene observation on which the UCH-rescue hypothesis rests, and because USP8's fusion geometry (balanced translocation) differs from USP2's (intrachromosomal inversion) while the retained domain is the same, the shared feature is the domain and not the mechanism of formation.
Why the geometry difference matters
KMT2A::USP2 is an intrachromosomal inversion over ~1 Mbp; KMT2A::USP8 is a balanced translocation between chromosomes 11 and 15 13. If both fusions were inversions, one could argue that whatever selects them is spatial - the two loci happen to be near each other and near each other in the nucleus. They are not. USP8 sits on a different chromosome entirely. What the two share is the UCH domain, not the way the break happened, which is exactly the shape of evidence that points at protein function rather than at genome geometry.
That is the strongest available argument that the deubiquitinase content of these fusions is selected for rather than incidental - and it rests on n = 30 plus n = 3, which is thin but not nothing.
Clinical picture, such as it is
Three patients across two reports and no series. The detailed case: 16-month-old boy, B-ALL, trisomy 8, t(11;15)(q23;q21), atypical KMT2A FISH, focal 15q deletion breaking inside USP8, resolved by mate-pair sequencing 3. The 2019 case is described only as one of four balanced minor-BCR translocations 1. The paper reporting the detailed case describes KMT2A-USP2 and KMT2A-USP8 together as high-risk with frequent CNS involvement, poor initial prednisone response and poor event-free survival 3 - but those descriptors were derived for USP2 and are restated for USP8, not measured in it. No outcome data specific to KMT2A::USP8 exist.
Nothing should be inferred about prognosis for this fusion. Three patients, no follow-up published, and the prognostic language in the literature is borrowed.
What has never been done
Everything. There is no cell line, no mouse, no xenograft, no protein-level characterisation, no localisation, no chromatin profile, no transcriptional signature, and no test of menin dependence for KMT2A::USP8. The per-patient junction sequence is unpublished, so even the exact composition of the protein is unknown. The entry exists to hold the position and to make the gap visible, not because there is content to summarise.
Analogy: the within-family pair with USP2, and the experiment the pair makes available
Closest analogue, and the axis. fusion-kmt2a-usp2, on the axis of the class of the 3-prime module - a USP/UCH deubiquitinase catalytic domain - plus a shared 5-prime extent, since both break in the KMT2A minor BCR and therefore both retain the PHD1-3 cluster, bromodomain and ePHD4. These are the only two deubiquitinase partners in the 3401-patient recombinome and both are minor-BCR-restricted, which is what hyp-uch-domain-rescues-phd-mediated-mll-degradation is built on.
Why USP8 is the more informative member despite being the rarer one. USP2 and USP8 arise by different mechanisms - an intrachromosomal inversion driven by USP2 lying ~1 Mbp from KMT2A in the right orientation, against a balanced t(11;15)(q23;q21) - so the shared feature cannot be local genome architecture. Two independent routes converging on the same domain class is the argument that the domain is selected for, and USP8 is the case that makes it, because a single inversion partner would be explicable by proximity alone.
Where the analogy breaks, and it is a break the UCH-rescue hypothesis does not currently account for: the two deubiquitinases have unrelated substrate biology.
| USP2 | USP8 | |
|---|---|---|
| canonical protein | O75604, 605 aa, USP domain 267-599 | P40818, 1118 aa, USP domain 777-1109 |
| best-characterised substrates | MDM2 and MdmX/MDM4, i.e. the p53 axis | EGFR and endosomal receptor trafficking |
| disease role outside leukaemia | overexpression stabilises MDM2 and promotes p53 degradation | somatic gain-of-function mutations in the 14-3-3-binding motif cause Cushing's disease by sustaining EGF signalling |
| fusion geometry | intrachromosomal inversion at 11q23.3 | balanced t(11;15)(q23;q21) |
And in both cases the fusion delivers the catalytic domain stripped of its N-terminal regulatory region, which is the one thing the two really do share at the protein level. For USP8 the Cushing hotspot - the 14-3-3-binding motif around residue 718 - sits in a disordered segment N-terminal to the USP domain, so the region whose mutation activates the enzyme in pituitary adenoma is most likely not in the fusion at all. For USP2 the region annotated as necessary for the MDM4 interaction spans residues 1-200, which a minor-BCR fusion carrying the conserved UCH domain does not retain. So the honest statement is that neither partner brings its known substrate biology into the fusion; both bring an unregulated catalytic module, and that is a cleaner statement of what the UCH-rescue hypothesis actually requires than 'a deubiquitinase is fused in'.
The discriminating experiment the pair makes available, which nobody has run. exp-usp2-uch-catalytic-dead-fusion-transformation is designed as a three-arm USP2 experiment. Adding a fourth arm - the matched KMT2A portion fused to the USP8 catalytic domain - converts it from a test of one fusion into a test of the hypothesis. If catalytic class is all that matters, the USP2 and USP8 arms should behave identically and both should be abolished by the active-site mutation. If they differ, the UCH-rescue model is incomplete and something partner-specific is doing work. Neither fusion has a cell line, a mouse or a xenograft, so this is buildable from constructs alone.
A caution on how much weight the pair can carry. n = 30 and n = 3 in the same recombinome, no protein-level evidence for either fusion, no published per-patient junction, and no western blot of either chimera. The pair is a good hypothesis-generator and a poor evidence base, and the existing counter-observation stands: two AFF1 cases and one MLLT3 case also break in the minor BCR and supply no deubiquitinase at all, so the PHD-retaining configuration is leukaemogenic without a UCH domain.
| gene5 | KMT2A |
|---|---|
| gene3 | USP8 |
| cytoband | t(11;15)(q23;q21) |
| mechanism class | balanced reciprocal translocation - explicitly NOT an inversion, unlike KMT2A::USP2 |
| uniprot5 | Q03164 |
| uniprot3 | P40818 |
| uniprot3 verified | P40818 = UBP8_HUMAN, PRIMARY gene name USP8, 1118 aa, Homo sapiens (UniProt REST, fetched 2026-09-19; primary gene name and length read back from the record). |
| prognosis | Not estimable. Three patients, no published follow-up, and the prognostic language in the literature is borrowed from the USP2 series. |
diseases
[ "paediatric B-ALL" ]
frequency
{
"dcal_recombinome_n_2023": 3,
"dcal_n_2019": 1,
"independent_cases": 1,
"pubmed_records_symbol_query": 1,
"pubmed_records_actually_relevant": 3,
"note": "The symbol query under-retrieves because the founding letter has no indexed abstract. See gene-usp8 E6."
}breakpoints
[
{
"id": "bp-kmt2a-bcr2-usp8-uch",
"population": "KMT2A break in the minor BCR; USP8 contributes the conserved UCH domain. NO PER-PATIENT JUNCTION IS PUBLISHED, so the domain lists are positional inference from the BCR2 interval and the stated UCH retention.",
"domains_retained_5": [
"AT-hook 1",
"AT-hook 2",
"AT-hook 3",
"SNL1/2",
"MBM1",
"MBM2",
"CXXC",
"PHD1",
"PHD2",
"PHD3",
"Bromodomain",
"PHD4"
],
"domains_lost_5": [
"FYRN",
"TAD",
"FYRC",
"WIN",
"SET"
],
"domains_retained_3": [
"UCH (ubiquitin carboxyl-terminal hydrolase catalytic domain)"
],
"domains_lost_3": [
"USP8 dimerisation domain",
"Rhodanese-like domain"
],
"phenotype_notes": "The USP8 dimerisation domain is N-terminal (residues 8-115) and therefore probably lost, which removes forced oligomerisation as an explanation for the partner's contribution and leaves catalysis as the main candidate."
}
]never measured
[ "cell of origin, cell type, cell state", "subcellular and subnuclear localisation", "any epigenetic readout", "transcriptional signature", "menin dependence", "catalytic activity of the fusion", "any model system of any kind" ]
- USP8 gain-of-function mutations cause Cushing disease in humans, so this gene has a demonstrated systemic phenotype when its activity is perturbed. That is directly relevant to any proposal to drug the UCH domain class.
- Three patients is too few for any therapeutic inference, and the literature's habit of describing USP2 and USP8 together risks transferring USP2's clinical profile onto a fusion in which nothing has been measured.
- Unlike KMT2A::USP2, this fusion was visible to karyotype and microarray in the one detailed case, so it should not be assumed to hide in the same places.
- Is the WW domain (USP8 648-680) inside or outside the fusion? Without a junction sequence the protein's composition is unknown, and a WW domain would give the fusion a proline-rich-motif binding surface the UCH domain does not provide.
- Do all three recombinome USP8 patients break in the minor BCR, or only the one that was individually reported? The recombinome does not say, and this entry assumes it.
- Would a KMT2A::USP8 construct transform in the same assay as KMT2A::USP2? If the UCH domain is the operative shared feature, two different deubiquitinases fused to the same KMT2A portion should behave alike - which is a clean, cheap internal replication of the hypothesis.
- Does KMT2A::USP8 retain a catalytically competent USP domain, and is the 14-3-3-binding motif in or out? UniProt puts the catalytic domain at 777-1109 of 1118 residues and the Cushing-disease hotspot around 718, just outside it, so the fusion plausibly delivers an enzyme with its principal regulatory motif removed. With three patients in the world and one DNA-level breakpoint published, a junction-spanning RT-PCR and a single western blot would convert the whole UCH-rescue hypothesis from a sequence inference into a protein observation.
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
USP8 (15q21.2) was identified as a novel KMT2A fusion partner in the minor-BCR discovery cohort, where the KMT2A::USP8 case was one of four balanced translocations breaking in BCR2, and USP8 rearrangements fused the conserved UCH domain to an extended 5'-MLL portion.
Human MLL/KMT2A gene exhibits a second breakpoint cluster region for recurrent MLL-USP2 fusions. · Meyer C, Lopes BA, Caye-Eude A, Cavé H, Arfeuille C, et al., Marschalek R · Leukemia 2019What this evidence cannot show
One patient in this cohort, described in a single sentence plus a supplementary figure that was not opened in this session. The UCH-retention statement is made jointly for USP2 and USP8 and no junction sequence is given for either. Full main text read (PMC6756029).
Evidence that argues against this (1)
- The same cohort's other three balanced minor-BCR translocations were AFF1 and MLLT3, which contribute elongation modules and no deubiquitinase. The presence of a USP8 case among them does not establish that the UCH domain is what made that patient's fusion viable. Muntean AG et al. 2008
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The 2023 KMT2A recombinome records 3 USP8 patients among 3401, and a minor-BCR break places the complete PHD1-3 cluster, bromodomain and ePHD4 into the direct fusion protein.
The KMT2A recombinome of acute leukemias in 2023. · Meyer C, Larghero P, Almeida Lopes B, Burmeister T, Gröger D, et al., Marschalek R · Leukemia 2023What this evidence cannot show
Three patients at one referral centre; a count of 3 cannot separate genuine rarity from assay failure. The domain-retention claim is positional, inferred from the BCR2 interval and the KMT2A domain map rather than from protein evidence in any patient. Counts read from the published article in this session.
Evidence that argues against this (1)
- It is not stated in the recombinome that all three USP8 patients broke in BCR2; the BCR2 assignment for USP8 comes from the single 2019 case. Treating all three as minor-BCR is an inference this entry makes and the source does not. Meyer C et al. 2019
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
An independent KMT2A-USP8 case in a 16-month-old boy with B-ALL showed trisomy 8, t(11;15)(q23;q21), an atypical KMT2A FISH result and a focal 15q deletion with a breakpoint inside USP8 on microarray, and was resolved by mate-pair sequencing; the authors describe KMT2A-USP2 and KMT2A-USP8 collectively as high-risk with frequent CNS involvement, poor prednisone response and poor event-free survival.
Cryptic and atypical KMT2A-USP2 and KMT2A-USP8 rearrangements identified by mate pair sequencing in infant and childhood leukemia. · Blackburn PR, Smadbeck JB, Znoyko I, Webley MR, Pitel BA, Vasmatzis G, Xu X, Greipp PT, Hoppman NL, Ketterling RP, Baughn LB, Lindsey KG, Schandl CA, Wolff DJ, Peterson JF · Genes, Chromosomes & Cancer 2020What this evidence cannot show
One patient for USP8, with a co-occurring trisomy 8 whose contribution cannot be separated. The prognostic descriptors are applied to the USP2/USP8 pair and derive from the larger USP2 series; no outcome data specific to KMT2A::USP8 exist anywhere. Abstract read in full.
Evidence that argues against this (1)
- Borrowing USP2's prognostic profile for USP8 assumes the two fusions behave alike, which is precisely what is unknown - they differ in geometry, in partner length and in the partner's normal biology, and USP8's case was not even cryptic to microarray. Meyer C et al. 2019
- E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
USP8's characterised disease biology is endosomal receptor trafficking, not chromatin: exome sequencing of corticotroph adenomas found somatic USP8 mutations in 4 of 10 tumours, clustered in the 14-3-3 protein-binding motif, which enhanced proteolytic cleavage and catalytic activity of USP8, increased deubiquitination of the EGF receptor, impaired its downregulation, sustained EGF signalling and raised proopiomelanocortin promoter activity - establishing that dominant USP8 mutations cause Cushing's disease through EGF receptor signalling.
Mutations in the deubiquitinase gene USP8 cause Cushing's disease. · Reincke M, Sbiera S, Hayakawa A, Theodoropoulou M, Osswald A, Beuschlein F, Meitinger T, Mizuno-Yamasaki E, Kawaguchi K, Saeki Y, Tanaka K, Wieland T, Graf E, Saeger W, Ronchi CL, Allolio B, Buchfelder M, Strom TM, Fassnacht M, Komada M · Nat Genet 2015What this evidence cannot show
Pituitary corticotroph adenoma, an endocrine epithelial tumour with no relationship to this fusion or to haematopoiesis; 10 tumours in the discovery exome set. Cited here only to establish what USP8's characterised biology IS, as the input to the disanalogy with USP2 in the analogy section, and not as evidence about KMT2A::USP8.
Evidence that argues against this (1)
- The activating region this paper identifies is probably not in the fusion. UniProt places the USP catalytic domain of the 1118-residue protein at residues 777-1109, and the 14-3-3-binding motif that carries the Cushing hotspot lies around residue 718, inside a disordered segment (679-746) N-terminal to the catalytic domain. A KMT2A::USP8 chimera described as fusing the conserved UCH catalytic domain therefore most likely delivers an enzyme that has LOST its 14-3-3 regulatory motif rather than one carrying the pituitary gain-of-function lesion - which is the same conclusion the USP2 side reaches by a different route, and is why the analogy in this entry is about loss of regulation rather than about shared substrates. UniProt Consortium 2026
- E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
USP8 is UniProt P40818, 1118 aa, with an N-terminal MIT domain at residues 33-116, a rhodanese domain at 195-313 and the catalytic USP domain at 777-1109, so the conserved UCH catalytic module occupies the C-terminal third of the protein and every annotated N-terminal regulatory module lies outside it.
UniProtKB entry P40818 (UBP8_HUMAN), Ubiquitin carboxyl-terminal hydrolase 8, with domain boundaries · UniProt Consortium · UniProtKB 2026What this evidence cannot show
Database annotation for the wild-type canonical isoform, not for the fusion protein; accession resolved by reading the primary gene name (USP8) and length (1118 aa) back from the fetched record, per the graph's accession rule. Which USP8 residue actually begins the chimera is not published for any of the three recombinome patients, so the mapping of these boundaries onto the fusion is an inference from the founding report's statement that the conserved UCH domain is what is fused.
Evidence that argues against this (1)
- The DNA-level breakpoint has been resolved in one case and the protein-level junction still has not. The independent 16-month-old case was resolved only by mate-pair sequencing, which localised a focal 15q deletion with a breakpoint inside USP8 - so a genomic coordinate exists while the reading frame, the first USP8 residue in the chimera and the presence of an intact folded USP domain remain unverified for every published patient. No western blot of a KMT2A::USP8 protein exists. Blackburn PR et al. 2020
- part of →KMT2A / MLL1 — the gene that breaks
5' KMT2A through the minor BCR, retaining PHD1-3, bromodomain and ePHD4
conserved UCH catalytic domain (USP8 777-1106); dimerisation domain probably lost
one of four balanced translocations breaking in BCR2 in the discovery cohort
- supports →The partner's deubiquitinase domain is what lets a PHD-retaining KMT2A fusion survive and transform
second, independent deubiquitinase partner with a different fusion geometry - the shared feature is the UCH domain, not the way the break happened
the two deubiquitinase KMT2A fusions; 3 patients against 30, translocation against inversion, same retained domain class
partially: the one detailed case WAS visible to karyotype and microarray, so the cryptic profile does not transfer wholesale from USP2
AXIS: SHARED 3-PRIME MODULE CLASS (a USP/UCH deubiquitinase catalytic domain) PLUS SHARED 5-PRIME EXTENT (both break in the KMT2A minor BCR and retain PHD1-3, the bromodomain and ePHD4). The only two deubiquitinase partners in the 3401-patient recombinome, at 30 and 3 patients, both minor-BCR-restricted, and arriving by DIFFERENT geometries - intrachromosomal inversion against a balanced t(11;15) - so the shared feature is the domain and not local genome architecture. DISANALOGY: the two enzymes have unrelated substrate biology (USP2 on the MDM2/MDM4-p53 axis, USP8 on EGFR endosomal trafficking), and in both the fusion retains only the catalytic domain and leaves the N-terminal regulatory region behind - USP8's 14-3-3 Cushing motif at ~718 sits outside the 777-1109 catalytic domain, USP2's MDM4-interaction region at 1-200 outside 267-599. So both deliver an UNREGULATED catalytic module rather than their partner's known specificity. TEST THE PAIR MAKES AVAILABLE: add a USP8 arm to exp-usp2-uch-catalytic-dead-fusion-transformation; identical behaviour supports the UCH-rescue model, divergence refutes its sufficiency.
Localisation gap recorded for the fusion specifically: 1 record for USP8/UBPY/KIAA0055 against the nine-term KMT2A/MLL fusion clause and 0 with a localisation clause added; and the domain analysis says the wild-type endosomal address cannot be inherited because the MIT signal (UniProt 33-116) lies 5-prime of the retained UCH domain (777-1109) (Q95 2026-09-22)