Oncofusion

start / graph / concept-junction-specific-antibody-to-a-leukaemia-fusion-tel-6f2

Explainerstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q115-absence-rebound

A junction-specific monoclonal antibody against a paediatric leukaemia fusion oncoprotein exists: clone TEL 6F2, raised against a 12-mer spanning the ETV6::RUNX1 fusion site and used for genome-wide ChIP

also called TEL 6F2 · TEL_6F2 · TEL-AML1 junction antibody · RRID:AB_3713372 · anti-ETV6-RUNX1 monoclonal

An antibody that recognises only the join between two fused proteins can see the cancer protein and nothing else, and one was made in 2013 for the commonest childhood leukaemia fusion by immunising with a twelve-letter peptide that straddles the join. It worked on blots, in flow cytometry and, most usefully, for mapping the fusion protein onto DNA across the genome.

The same thing, for a scientist

Clone **TEL 6F2** (rat IgG2a) was raised against the synthetic peptide **PIGRIAECILGM**, described by its authors as an immunising peptide *spanning the fusion site between the TEL and AML1 fusion partners* [E1]. It detects the fusion by western blot in an inducible BA/F3 pro-B system and by flow cytometry, and it was used for **ChIP-on-chip**, yielding 217 directly and 118 indirectly regulated ETV6::RUNX1 target genes; the majority of bound promoters were specific to the fusion and were not bound by native AML1 or native TEL [E1]. The reagent is catalogued in the Antibody Registry as **RRID:AB_3713372**, non-commercial, applications FACS / WB / ChIP, defining citation PMID 24121163 [E2]. This matters for the KMT2A programme because the standing argument for a junction antibody in this family rests on a **synovial sarcoma** exemplar (clone E9X9V, SS18-SSX) [E3]; the closer precedent is a **paediatric B-lineage leukaemia fusion**, and it demonstrates the two things the sarcoma case does not - that a short linear junction peptide is a sufficient immunogen for a chimeric transcription factor, and that the resulting antibody survives crosslinked chromatin. A second junction reagent exists for a leukaemia fusion at the protein level, **Bcr-Abl (b2a2 Junction Specific), clone L99H4, RRID:AB_490962**, but it is catalogued for western blot only [E4].

Go deeper
What exactly was immunised, and what that licenses

The immunogen is a 12-mer, PIGRIAECILGM, coupled to keyhole limpet haemocyanin and to ovalbumin, injected into Lou/C rats with CpG oligonucleotide and incomplete Freund's adjuvant; hybridomas were screened by ELISA against the immunising peptide and then by western blot 1. The design figure states the peptide spans the fusion site 1.

The consequence for this graph is arithmetical. absence-no-junction-specific-antibody-exists-for-any-kmt2a-fusion computes six distinct in-frame KMT2A::AFF1 protein junctions with a shared AFF1 acceptor octamer QTYSNEVH, and lists them as 16-mers. A 12-mer worked for ETV6::RUNX1. The immunogen length this family would need is not larger than the one that has already succeeded, which removes the only technical objection that is specific to peptide design rather than to breakpoint diversity.

What the ChIP result does and does not establish

ChIP-on-chip with TEL 6F2 was performed in an inducible murine BA/F3 pro-B line, not in primary human leukaemia, and the fusion was exogenous 1. So the demonstration is that a junction-raised monoclonal can be used to pull crosslinked chromatin at all - which is the step most likely to fail and the step the KMT2A field most needs - not that it performs in a patient sample.

An antibody that recognises a linear 12-mer in ELISA need not recognise the same sequence inside a folded, crosslinked chimera. The ChIP experiment is the only evidence in the paper that it does, and the specificity control is induced-versus-uninduced expression, not a junction point mutant.

Why a literature search could not have found this

The words junction, breakpoint and fusion-specific appear zero times in the full text of PMID 24121163 - counted on the retrieved fullTextXML, 50,231 characters 1. The abstract says only "a monoclonal antibody directed against TEL-AML1". No [tiab] query and no Europe PMC BODY: query on those three phrases can reach it.

It is reachable in one request from the Antibody Registry, because a catalogue indexes the object rather than the prose about it 2. That is the census's instrument mismatch finding turned positive: the mismatched instrument was not merely weaker, it was structurally incapable, and the matching instrument returned the answer immediately.

Facts and figures
cloneTEL 6F2
rridRRID:AB_3713372
immunogen peptidePIGRIAECILGM
immunogen length AA12
immunogen designspans the fusion site between the TEL and AML1 partners
hostLou/C rat
isotyperat IgG2a
chip platformChIP-on-chip (promoter array)
chip systemmurine BA/F3 pro-B line with inducible human TEL-AML1
direct targets identified217
indirect targets identified118
commercialfalse
source labRegina Feederle / Elisabeth Kremmer, Helmholtz Center Munich; A Bockhardt, Heinrich-Heine University Duesseldorf
defining citation pmid24121163
full text chars read50231
relationship to existing entriesSupplies the missing precedent for `absence-no-junction-specific-antibody-exists-for-any-kmt2a-fusion`, which is correct that no such reagent exists for any KMT2A fusion but names synovial sarcoma as the only worked example. It does not duplicate that entry: that entry bounds the absence, this one documents the reagent that shows the approach works in a paediatric leukaemia fusion and was found in a different instrument.
word counts in pmid 24121163 full text
junction0.000junction: 0breakpoint0.000breakpoint: 0fusion-specific0.000fusion-specific: 0
applications recorded
[
  "FACS",
  "Western Blot",
  "ChIP"
]
other leukaemia fusion junction reagent
{
  "clone": "L99H4",
  "rrid": "RRID:AB_490962",
  "target": "BCR-ABL b2a2 junction",
  "vendor": "Cell Signaling Technology",
  "catalogue": "3908",
  "applications_recorded": [
    "Western Blot"
  ]
}
kmt2a fusion spellings returning zero in the registry
[
  "MLL-AF4",
  "MLL-AF9",
  "MLL-ENL",
  "KMT2A-MLLT3",
  "KMT2A-AFF1"
]
word counts in pmid 24121163 full text — the numbers behind the chart
{
  "junction": 0,
  "breakpoint": 0,
  "fusion-specific": 0
}
How this could bite us back
  • A junction antibody used for ChIP has a silent failure mode that a western blot does not: an epitope buried by a partner protein at exactly the chromatin sites that matter would produce a clean, reproducible, biologically plausible map of the sites where the fusion is NOT engaged. The ETV6::RUNX1 result carries this risk and was not controlled for it.
  • A reagent that sees only the fusion cannot see `protein-kmt2a-star-truncated-product-of-the-internal-promoter` or the wild-type allele, so a junction antibody narrows an attribution problem by discarding the comparator. Any experiment built on one needs a parallel N-terminal KMT2A antibody arm.
  • If a junction antibody became the diagnostic, patients whose breakpoint produces a seventh, rarer junction would score negative. The six-junction calculation is a coverage estimate from 1,262 cases, not a guarantee.
Still open
  1. Is clone TEL 6F2 still distributable? It is a non-commercial hybridoma from 2013 held at Helmholtz Munich, and the registry record carries no availability field; an email settles it and would decide whether this is a template or a usable reagent.
  2. Would the same immunisation design work against the six KMT2A::AFF1 junction 16-mers? The shared acceptor octamer QTYSNEVH means five of the six differ only on the KMT2A side, so a single failed campaign against one epitope would not condemn the rest.
  3. Does any KMT2A::AFF1 junction 16-mer contain a predicted rat or rabbit B-cell epitope of comparable quality to PIGRIAECILGM? That is a desk calculation and nobody has done it.
Every source behind this entry (4)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Monoclonal antibody clone TEL 6F2 was raised in Lou/C rats against the synthetic peptide PIGRIAECILGM spanning the TEL-AML1 fusion site, detects the fusion by western blot and flow cytometry in an inducible BA/F3 pro-B system, and was used for ChIP-on-chip identifying 217 direct ETV6::RUNX1 target genes whose promoters were largely not bound by native AML1 or native TEL.

    PMID 24121163DOI
    What this evidence cannot show

    Read in full through Europe PMC fullTextXML (PMC3816209, 50,231 characters) on 2026-09-22; the antibody-generation text and the peptide sequence were quoted from that retrieval, not from the abstract. The ChIP-on-chip is in a murine BA/F3 line carrying an inducible human fusion construct, not in primary human B-ALL, and ChIP-on-chip is a promoter-array method, not ChIP-seq. The authors do not report a titre, an epitope-mapping experiment, or cross-reactivity against a junction point mutant.

    Evidence that argues against this (1)
    • Specificity is shown as induced-versus-uninduced signal in one engineered line plus a V5-tag comparison. That design cannot distinguish a junction-specific antibody from one that recognises a TEL-side or RUNX1-side epitope that happens to be conformationally exposed only in the chimera; only a junction point mutant or a native-TEL/native-AML1 immunoblot at matched abundance would. The paper's own statement that most fusion-bound promoters are not bound by native AML1 or TEL is a statement about the promoters, not about the antibody. Blood Cancer Journal 2013
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    The Antibody Registry holds this reagent as a curated record: abId 3713372 (RRID:AB_3713372), name 'TEL-AML1 (ETV6-RUNX1) Antibody', target 'chimeric transcription factor TEL-AML1 (ETV6-RUNX1)', clone TEL_6F2, monoclonal, non-commercial (Helmholtz Center Munich and Heinrich-Heine University Duesseldorf), applications FACS / Western Blot / ChIP, defining citation PMID 24121163, zero recorded citations.

    What this evidence cannot show

    Fetched 2026-09-22 through the JSON API. The human-facing pages https://www.antibodyregistry.org/AB_3713372 and /antibody/AB_3713372 both return HTTP 200 with a 1,147-byte client-rendered shell containing neither 'TEL-AML1' nor '6F2', so the API URL is cited because it is the one that actually returns the record. The registry's `targetSpecies` field for this record is EMPTY and `numOfCitation` is 0; neither is evidence about the reagent, both are gaps in the catalogue.

    Evidence that argues against this (1)
    • A registry record is a catalogue entry, not a validation. RRID:AB_3713372 lists ChIP among its applications because the defining paper used it that way; nobody independent has repeated it, and the record carries zero citations. SciCrunch Antibody Registry 2026
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    The Antibody Registry returns exactly one record for SS18-SSX - the E9X9V XP rabbit monoclonal from Cell Signaling Technology, RRID:AB_3665243 - which is the exemplar this graph already cites, so the instrument demonstrably returns a known junction reagent when one exists.

    What this evidence cannot show

    This is the POSITIVE CONTROL for the search, and it travels through the same `/api/fts-antibodies?q=` URL template as every query reported here, not merely the same host. The negative control through that template, q=NOTAREALANTIGENXYZ, returns 0.

    Evidence that argues against this (1)
    • The registry's full-text search tokenises, so short alphanumeric tokens are unreliable operands: q=b2a2 returned 25 records dominated by '[Biotin]' and 'clone 2B4' names, none about BCR-ABL. A count from this endpoint is not a clean number and every returned name must be read. SciCrunch Antibody Registry 2026
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    A second junction-specific antibody against a leukaemia fusion protein is catalogued: Bcr-Abl (b2a2 Junction Specific), clone L99H4, Cell Signaling Technology catalogue 3908, RRID:AB_490962, mouse monoclonal, human target, applications 'W' (western blot) only.

    What this evidence cannot show

    Fetched 2026-09-22. The record carries no defining citation, no epitope field and zero recorded citations, and lists western blot as its only application - so unlike TEL 6F2 it is NOT evidence that a junction antibody can be used on chromatin. Curated 2014/2015.

    Evidence that argues against this (1)
    • BCR::ABL1 is a cytoplasmic kinase fusion and the b2a2 junction is one of a small number of recurrent transcript junctions, so this reagent speaks to feasibility of the immunogen and not to feasibility for a nuclear chromatin-bound chimera. TEL 6F2 is the relevant precedent, not this one. SciCrunch Antibody Registry 2026
Connected to