KMT2A*: a menin-blind, LEDGF-blind, SET-competent KMT2A protein made from out-of-frame reciprocal alleles

No three-dimensional model is available for this protein, so the map above is the whole picture. That is itself informative: AlphaFold does not model chains this long, and the parts that have been crystallised are held under the entries for the individual domains.

• KMT2A* is a KMT2A-derived, chromatin-competent, menin-blind protein that no assay detects and no eligibility criterion considers. If it contributes to disease, menin-inhibitor non-response would look idiopathic.
• It is also the species for which a WIN-site or SET inhibitor would be the only available handle - and those agents cannot distinguish it from the wild-type KMT2A allele.
• Because the internal promoter coincides with recombination hot spot II, transcription through that region is implicated in generating the translocation in the first place, so silencing it is not obviously safe even as a thought experiment.

1. Has KMT2A* ever been detected, as transcript or protein, in a patient sample? The recombinome says out-of-frame alleles 'may' express it and cites a promoter paper; nobody appears to have looked.
2. What is the initiator methionine of KMT2A*? Without it the protein's N-terminal boundary, and therefore whether PHD1 is intact, is unknown.
3. Does KMT2A* bind WRAD and methylate H3K4 in cells? It retains WIN and SET and has no menin handle, so it is the cleanest available test of whether a KMT2A protein needs menin to act on chromatin.
4. How many t(4;11) patients scored reciprocal-negative by RT-PCR carry an out-of-frame reciprocal allele capable of expressing KMT2A*? That number would change the denominator of every reciprocal-status analysis in this beat.

1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   A gene-internal promoter element in the mouse and human MLL gene colocalises with recombination hot spot II in the telomeric portion of the MLL breakpoint cluster region, and the mRNA it generates encodes an N-terminally truncated version of MLL that still exhibits many functional regions including the C-terminal SET domain; etoposide-induced double-strand breaks colocalise with the RNA polymerase II binding site and transcription initiation region rather than with a nearby topoisomerase II consensus.

   Transcription linked to recombination: a gene-internal promoter coincides with the recombination hot spot II of the human MLL gene. · Scharf S, Zech J, Bursen A, et al. · Oncogene 2007
   PMID 16983345DOI
   ▸What this evidence cannot show

   2007 promoter-mapping work in cell systems; establishes that the promoter exists and what its transcript encodes, not that it is used in leukaemic blasts carrying an out-of-frame reciprocal allele. The translocation-mechanism half of the paper is a separate claim from the protein half.

   ▸Evidence that argues against this (1)

   • Whether a promoter is used in a given cellular context is not established by its existence; no measurement of KMT2A* transcript or protein in a patient sample has been published. Meyer C et al. 2023
2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   Read in full text (PMC10169636): of 426 complex KMT2A rearrangements analysed, 40 reciprocal fusions were in frame while 386 were non-functional or out of frame at the genomic DNA level, but the majority of the out-of-frame reciprocal KMT2A fusions may still be transcribed and encode a 5'-truncated KMT2A protein termed KMT2A*, from a gene-internal promoter upstream of KMT2A exon 12; this shorter protein has no ability to bind Menin1, LEDGF or MYB but still carries all enzymatic functions and the domains needed to bind proteins performing H4K16 acetylation (MOF) or H3K4 methylation by the SET-domain complex, and retains the capacity to bind, read and modify chromatin.

   The KMT2A recombinome of acute leukemias in 2023. · Meyer C, Larghero P, Almeida Lopes B, et al. · Leukemia 2023
   PMID 37019990DOI
   ▸What this evidence cannot show

   Read in full text via Europe PMC fullTextXML (86,330 characters), not abstract only. The word 'may' is the authors' own: this is an inference from the promoter's existence rather than a measurement of KMT2A* in those 386 cases. The binding statements are architectural, since menin, LEDGF and MYB sites are all N-terminal, and no binding assay on KMT2A* is cited.

   ▸Evidence that argues against this (1)

   • The direct test of whether a menin-blind, SET-retaining KMT2A protein is consequential has only been done for a different species, NUP98::KMT2A, where the result was a leukaemia with reduced rather than elevated HOXA output - the opposite direction from the reciprocal-positive t(4;11) patients in whom HOXA is high. Fisher JN et al. 2020
3. E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

   In an inducible Sleeping Beauty vector system testing MLL-NEBL, NEBL-MLL, MLL-LASP1, LASP1-MLL, MLL-MAML2, MAML2-MLL, MLL-SMAP1 and SMAP1-MLL for growth behaviour, focus formation, colony formation and ectopic Hoxa transcription, only 1 of 4 direct MLL fusions but 3 of 4 reciprocal MLL fusions exhibited oncogenic functions.

   Functional characterisation of different MLL fusion proteins by using inducible Sleeping Beauty vectors. · Wächter K, Kowarz E, Marschalek R · Cancer Lett 2014
   PMID 25016062DOI
   ▸What this evidence cannot show

   A pilot panel of four partner pairs in cell lines, explicitly described by its authors as pilot experiments motivating a systematic analysis; focus and colony formation are transformation surrogates, not leukaemia; the partners tested (NEBL, LASP1, MAML2, SMAP1) are rare ones, not AFF1, MLLT3 or MLLT1.

   ▸Evidence that argues against this (1)

   • The best-studied reciprocal product gives the opposite answer in the best-matched system: siRNA removal of AF4-MLL from human t(4;11) lines produced no growth, cell-cycle or apoptotic phenotype while removal of the direct fusion did. Kumar AR et al. 2011
4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

   UniProt Q03164 places PHD1-4 at 1431-1978, the bromodomain at 1635-1765, FYRN at 2018-2074, the TAD 9aaTAD core at 2847-2855, FYRC at 3666-3747, the WIN motif at 3762-3767 and the SET domain at 3829-3945, while the menin-binding motif (6-25) and both LEDGF integrase-binding motifs (123-134, 147-152) lie in the first 160 residues.

   UniProt Q03164 (KMT2A_HUMAN), feature table · UniProt Consortium · UniProtKB 2026
   ▸What this evidence cannot show

   Fetched 2026-09-19, primary gene name read back as KMT2A, length 3969. The mapping of these coordinates onto a protein initiating from the internal promoter assumes the initiation site lies upstream of PHD1 at 1431; the recombinome places the promoter upstream of exon 12, which is consistent, but the exact initiator methionine of KMT2A* was not resolved by this beat.

   ▸Evidence that argues against this (1)

   • A database annotation is opposable on scope: the region where KMT2A* would begin is the least structurally characterised part of the protein, so the N-terminal boundary of this species is not observed. PDBe 2026

• ← related toThe protein nobody can attribute by antibody is the one mass spectrometry finds most easily

  the assay resolves the in-frame reciprocal product and is blind to KMT2A*, which has no junction

• ← related toAn anti-MLL-N antibody sees the wild-type allele and the direct fusion, and nothing else - and cannot rank them

  KMT2A* has no N-terminus, so this reagent class cannot see it either

• ← depends onThe anti-MLL-C ambiguity is not a t(4;11) problem - it has three sources, and one KMT2A fusion family is measurably cleaner than the rest

  the third, unquantified source of MLL-C signal

• ← analogous toAnti-AF9 antibodies see three proteins, not two - and the third is an internal-promoter isoform discovered in 2025

  the same internal-promoter, junctionless, epitope-sharing architecture on the partner side of the fusion

• part of →KMT2A / MLL1 — the gene that breaks

  A 5'-truncated KMT2A protein initiating from a gene-internal promoter upstream of exon 12.

• confers resistance to →MEN1 / menin — the adaptor the fusion cannot work without

  Reported to have no ability to bind menin, because the binding motif is N-terminal and absent.

• confers resistance to →PSIP1 / LEDGF p75 — the tether with the best therapeutic index in the whole interactome

  Reported to have no ability to bind LEDGF.

• related to →KAT7 / HBO1 — an enzyme, a leukaemia-stem-cell dependency, and an inhibitor already in a crystal

  Retains the domains needed to bind the H4K16 acetyltransferase machinery, per the recombinome's description of MOF binding.

• analogous to →AFF1::KMT2A (AF4-MLL), the der(4) product: what it is made of, residue by residue

  Same KMT2A content without the partner N-terminus; the two are the in-frame and out-of-frame outcomes of the same allele.

• supports →Reciprocal status is a different fact depending on whether you assay DNA or RNA, and the whole prognostic literature is RNA

  Adds a fourth layer to the assay problem: a patient can be junction-negative and still make a menin-blind KMT2A protein.

• supports →Two classes of KMT2A rearrangement keep the SET domain, and the graph says otherwise

  A third SET-retaining, menin-blind KMT2A species beyond the in-frame reciprocal product and NUP98::KMT2A.