Oncofusion

start / graph / ptm-kmt2a-k1364-sumoylation

Mechanismweak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.q165-the-retained-acetylation-site

KMT2A K1364 SUMO - the one modification site inside the breakpoint cluster, retained in some patients' fusion proteins and lost in others

One chemical-tag spot on MLL1 falls right inside the region where the chromosome usually breaks. Whether a patient's fusion protein has it depends on exactly where their break happened - so this single spot is different between patients with the same diagnosis.

The same thing, for a scientist

Q03164 K1364 carries a SUMOylation call (PRIDE PXD008003) and lies at residue 1364, inside the leukaemia breakpoint cluster (exons 8-14, ~1338-1607) and specifically in exon 9 territory, 2 residues past the annotated KMT2A-AF3P21/KNL1/CENPK breakpoint at 1362..1363 and 42 residues before the KMT2A-AFF4 breakpoint at 1406..1407. It is therefore VARIABLE: present in the fusion protein of patients whose break is 3' of it, absent in patients whose break is 5' of it. It is the only PTM site in this graph's census with that property.

Go deeper
Why a VARIABLE site is worth naming separately

Every analysis in this project that treats KMT2A-rearranged as one class implicitly assumes the retained portion is the same protein in every patient. For K1364 it is not. UniProt's own breakpoint annotations bracket it 2:

annotated breakpointfusionK1364 in the fusion protein?
1334..1335KMT2A-ZFYVE19❌ lost
1362..1363KMT2A-AF3P21, KMT2A-KNL1, KMT2A-CENPK❌ lost (by 1-2 residues)
1406..1407KMT2A-AFF4✅ retained
1444..1445KMT2A-GAS7, KMT2A-LPP✅ retained

The commonest clinical breakpoints — intron 9 and intron 11 for KMT2A::AFF1 — were not resolved to residue coordinates here, so this entry does NOT state whether K1364 is in the index fusion. That is the obvious next lookup and it is deliberately left open rather than guessed.

Facts and figures
uniprotQ03164
position1364
residueK
modificationSUMOylation
fusion statusVARIABLE
status in index fusionNOT DETERMINED - the commonest KMT2A::AFF1 intron 9 and intron 11 breakpoints were not resolved to residue coordinates here
shown for fusionNO
breakpoint cluster AA
[
  1338,
  1607
]
retained in
[
  "KMT2A-AFF4 (breakpoint 1406..1407)",
  "KMT2A-GAS7 / KMT2A-LPP (1444..1445)"
]
lost in
[
  "KMT2A-ZFYVE19 (1334..1335)",
  "KMT2A-AF3P21 / KMT2A-KNL1 / KMT2A-CENPK (1362..1363)"
]
pride deposits
[
  "PXD008003"
]
measurements
[
  {
    "quantity": "MS-evidenced PTM sites inside the KMT2A breakpoint cluster",
    "subject": "Q03164 aa 1338-1607",
    "value": 4,
    "units": "sites",
    "n": 174,
    "universe": "all 174 MS-evidenced site/modification calls on Q03164",
    "null": null,
    "note": "K1364 SUMO, K1517 Ub, K1529 Ac, K1574 Ub - the only four whose presence in the fusion protein is patient-specific"
  }
]
How this could bite us back
  • A patient-specific residue difference is attractive as an explanation for heterogeneity and is almost certainly too small to be one. Four sites, all single-deposit, none with a known function - this is a curiosity to be honest about, not a lead.
Still open
  1. Is K1364 present in the KMT2A::AFF1 fusion protein produced by an intron 9 break, and by an intron 11 break? This is a coordinate lookup against the exon map that this beat did not run, and it is the difference between a site the index fusion has and one it does not.
  2. Do any of the four breakpoint-cluster PTM sites differ systematically between breakpoint classes in a way that could contribute to the partner-independent/partner-dependent biology this graph tracks? Four sites is not many, but a per-breakpoint protein difference is a mechanism nobody has proposed.
Every source behind this entry (2)
  1. E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    Q03164 K1364 carries a SUMOylation call in PRIDE PXD008003 and is the only SUMO site on KMT2A that is not distal to the breakpoint cluster.

    EBI Proteins API - proteomics-ptm, Q03164 (KMT2A) - K1364 SUMOylation · EMBL-EBI Proteins API / PRIDE-reprocessed MS evidence 2026
    What this evidence cannot show

    ONE deposit. All four KMT2A SUMO calls are single- or low-deposit except K2528; this one has no corroboration and no functional data.

    Evidence that argues against this (1)
    • A single-deposit SUMO call with no corroboration is at the detection limit of what this endpoint should be trusted for. If it is a misassignment, this entry describes nothing.
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProt annotates KMT2A translocation breakpoints at 1334..1335, 1362..1363, 1362, 1406..1407 and 1444..1445, which bracket residue 1364.

    What this evidence cannot show

    Residue identity verified against the live UniProt Q03164 FASTA (3,969 aa) on 2026-09-24. Canonical isoform only. These are curated published fusion breakpoints, not a clinical breakpoint distribution. The commonest KMT2A::AFF1 breakpoints (introns 9 and 11) are not among UniProt's SITE annotations and were NOT resolved to residue coordinates in this beat.

    Evidence that argues against this (1)
    • UniProt's SITE breakpoint annotations are a curated list of published individual fusions, heavily weighted to unusual partners that merited a paper. They are NOT the clinical breakpoint distribution, and the commonest KMT2A::AFF1 breakpoints are absent from them - so the bracket around K1364 is drawn from an unrepresentative sample of breakpoints.
Connected to