Oncofusion

start / graph / mechanism-top2-cleavage-at-kmt2a

Mechanismmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q1-fusion-genesis

Topoisomerase II cleavage and TOP2 poisons as a source of KMT2A breaks

also called TOP2B-mediated DSB · etoposide-induced MLL breaks · topoisomerase II poison mechanism

Topoisomerase II is an enzyme that deliberately cuts both strands of DNA to untangle it, then sews them back up. Drugs like etoposide jam it mid-cut, leaving a real double-strand break, and the KMT2A gene happens to sit at a favoured cutting site, which is why children treated with these drugs can later develop KMT2A-rearranged leukaemia.

The same thing, for a scientist

TOP2 creates transient covalent enzyme-DNA cleavage complexes; TOP2 poisons of the epipodophyllotoxin class (etoposide, teniposide) stabilise these complexes into persistent DSBs. Most etoposide-induced breaks at the KMT2A locus, and the bulk of etoposide genotoxicity overall, are TOP2B-dependent rather than TOP2A-dependent. However, TOP2A/TOP2B ChIP occupancy and measured etoposide-induced cleavage do not by themselves explain why breakpoints cluster where they do within the KMT2A BCR, so local TOP2 concentration is not the whole story; transcription and spatial co-localisation with the partner gene are required to complete the model. This is a well-supported mechanism for therapy-related KMT2A-r leukaemia and an unproven extrapolation for de novo infant KMT2A::AFF1. Two published results bound even the therapy-related claim: secondary-AML patients never exposed to epipodophyllotoxins break at the same TOP2-inducible hotspot [E1 counter], and the largest case-control test of dietary TOP2 inhibitors in infant leukaemia was essentially null [E1 counter].

Go deeper
The TOP2B result and why it matters1 deeper

Cowell et al. showed that most etoposide-induced chromosome breaks in the KMT2A locus, and the overall genotoxicity of etoposide, depend on TOP2B rather than TOP2A 1. That is mechanistically important because TOP2B is the isoform associated with transcription rather than replication, and it is expressed in non-dividing cells. It also means that, in principle, isoform-selective pharmacology exists as a concept, though no clinically usable TOP2B-sparing epipodophyllotoxin exists.

Local TOP2 concentration is not sufficient

The same study found that TOP2A and TOP2B occupancy and etoposide-induced DNA cleavage data suggest factors other than local TOP2 concentration determine the specific clustering of KMT2A translocation breakpoints in therapy-related AML 1. This is the honest caveat: TOP2 makes the break, but TOP2 occupancy does not predict the breakpoint. Something about transcription and nuclear geometry does the selecting.

Both TOP2 isoforms are required for the actual fusion

In a more recent genome-wide mapping study, TOP2-mediated DSBs were enriched at highly transcribed chromatin loop anchors, and the recurrent fusions that drive leukaemias arose at those hotspots in a manner requiring the activity of both TOP2 isoforms and requiring transcriptional elongation 2. So the isoform picture is: TOP2B dominates the break at KMT2A under etoposide, but formation of the productive fusion draws on both isoforms.

Reversibility: the cut is normally undone

TOP2 cleavage at the KMT2A-family breakpoint sites is reversible. Religation of drug-induced TOP2 cleavage sites at the AF9 breakpoint region was demonstrated directly 3. The relevance is that a TOP2 cut is not automatically a translocation; it becomes one only if the complex is trapped long enough and a second broken end is nearby. That is a kinetic window, and kinetic windows are in principle modifiable.

What this does NOT explain

De novo infant KMT2A::AFF1 arises in utero in children with no chemotherapy exposure. The classic mapping placed therapy-related breakpoints in the telomeric half of the BCR and de novo breakpoints in the centromeric half, implying distinct initiating mechanisms 4. That dichotomy should be treated as contested rather than settled. A review of the field states that a short telomeric sub-kilobase region of MLL is the most frequent site of rearrangement in infant acute leukaemia as well as in secondary leukaemia [E4 counter, Gole & Wiesmuller 2015], and within de novo t(4;11) ALL the breakpoint distribution splits by patient age rather than by aetiology [E4 counter, Reichel 2001]. Independently, five secondary-AML patients who never received etoposides carried near-identical MLL breakpoints at the TOP2-inducible hotspot, so hotspot position does not certify TOP2-poison causation [E1 counter, Langer 2003].

What survives is narrower and still useful: apoptosis-induced cleavage within the KMT2A BCR is not blocked by the catalytic TOP2 inhibitor merbarone, so at least one cleavage route into this locus is TOP2-independent 5. And the epidemiological form of the argument - that maternal dietary TOP2 inhibitors cause infant leukaemia - has now been tested directly and was essentially null [E1 counter, Spector 2005].

Facts and figures
mechanism classenzymatic DNA double-strand break
isoform dependenceetoposide-induced KMT2A breaks predominantly TOP2B-dependent; productive fusion formation requires both isoforms
contestedBreakpoint position at the TOP2-inducible hotspot does not imply TOP2-poison exposure (Langer 2003, PMID 12619163); the field's own synthesis argues for multiple nucleases rather than TOP2 alone (Gole & Wiesmuller 2015, PMID 26161385).
effectors
[
  "TOP2A",
  "TOP2B"
]
pharmacological inducers
[
  "etoposide",
  "teniposide",
  "doxorubicin",
  "mitoxantrone"
]
aetiology fit
{
  "therapy_related_kmt2a_r": "strong",
  "de_novo_infant_kmt2a_aff1": "weak / extrapolated; the largest case-control test of dietary TOP2 inhibitors (Spector 2005, n=240 cases) was null for MLL-rearranged ALL"
}
How this could bite us back
  • TOP2A/TOP2B are essential for chromosome segregation and transcription; systemic TOP2 inhibition is itself leukaemogenic, so this mechanism is a diagnostic and epidemiological handle rather than an intervention target.
  • If TOP2B inhibition were used prophylactically it would block transcription-coupled decatenation in normal proliferating tissue, a non-starter in a fetus or infant.
Still open
  1. Is TOP2B occupancy at the KMT2A BCR higher in fetal liver haematopoietic progenitors than in adult bone marrow HSC? If yes, that would be the cleanest mechanistic explanation for the infant age peak.
  2. Can a TOP2B-sparing topoisomerase poison be designed, and would it reduce therapy-related KMT2A-r AML without losing anti-tumour efficacy?
  3. What is the in vivo half-life of a trapped TOP2 cleavage complex at the KMT2A BCR, and does it vary with transcription rate?
  4. Why do secondary-AML patients who never received epipodophyllotoxins break at the same MLL hotspot as those who did (Langer 2003, n=5)? Either the hotspot is generically fragile and TOP2 poisons only raise its usage, or those patients had an unrecorded TOP2-poison exposure. Nobody has repeated this observation at cohort scale, and it is the single cheapest test of how much of the therapy-related story is really about TOP2.
Every source behind this entry (5)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Most etoposide-induced chromosome breaks at the KMT2A locus and the overall genotoxicity of etoposide are TOP2B-dependent, but local TOP2 concentration does not explain breakpoint clustering.

    PMID 22615413DOI
    What this evidence cannot show

    Human cell lines, pharmacological etoposide, TOP2B depletion by genetic means; extrapolation to primary fetal progenitors is untested.

    Evidence that argues against this (3)
    • Causality, not just occupancy, is what is unproven. Langer et al. sequenced t(9;11) breakpoints in 11 children and found that all five patients with secondary AML who had NOT received etoposides during treatment of their primary disease nevertheless carried almost identical MLL breakpoints, very close to the breakage hot spot that TOP2 inhibitors (and apoptotic triggers) induce in vitro. A breakpoint sitting at the TOP2-inducible hotspot is therefore not evidence that a TOP2 poison made it: the same position arises without any epipodophyllotoxin exposure. This makes the TOP2 signal at the KMT2A BCR correlative with position rather than diagnostic of mechanism. Langer T et al. 2003
    • The field's own synthesis treats the TOP2-poison account as one proposal among several. Gole and Wiesmuller review clinical patient data and cell-based studies of a wider spectrum of stimuli and conclude that cellular stress signalling pathways - early apoptotic DNA fragmentation, transcription stalling and replication stalling - create secondary DNA structures, provide chromatin accessibility and activate nucleases OTHER than topoisomerase II at the MLL locus, and that these may act in concert. Their title states the position: multiple rather than a single mechanism. Gole B et al. 2015
    • The epidemiological extrapolation to de novo infant disease has been tested and is largely negative. In the largest case-control study of the hypothesis (Children's Oncology Group; 240 incident infant acute leukaemias 1996-2002, 255 random-digit-dialled controls, maternal diet by food frequency questionnaire), there was little evidence of an association between a dietary DNA-topoisomerase-II-inhibitor index and infant leukaemia overall or by subtype. The only suggestive signal was in MLL-rearranged AML (quartile 4 vs 1 OR 3.2, 95% CI 0.9-11.9, P for trend = 0.10), which is not significant and rests on small numbers, and there was no signal in MLL-rearranged ALL - the subtype this project cares about. Maternal fruit and vegetable intake was inversely associated with risk, the opposite direction to the flavonoid-as-TOP2-poison prediction. Spector LG et al. 2005
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    TOP2-mediated DSBs are enriched at highly transcribed chromatin loop anchors, and leukaemia-driving fusions arise at these hotspots dependent on both TOP2 isoforms and on transcriptional elongation.

    PMID 31202576DOI
    What this evidence cannot show

    Human cell lines with induced TOP2 poisoning; fusion formation was measured as a frequency in cultured cells, not in a fetal haematopoietic system.

    Evidence that argues against this (1)
    • The genome-wide loop-anchor result and the locus-level break map disagree about what cuts KMT2A. Direct mapping of etoposide-induced double-strand breaks across the MLL breakpoint cluster region placed them at the RNA polymerase II binding site and the transcription initiation region of a gene-internal MLL promoter and explicitly NOT at a nearby topoisomerase II consensus sequence, leading those authors to attribute MLL's instability to transcriptional processes. Since TOP2 occupancy and Pol II occupancy are collinear at transcribed loop anchors, a population-level enrichment of TOP2-mediated breaks there cannot distinguish the two, and the one experiment that tried to separate them at this locus favoured transcription. Scharf S et al. 2007
  3. E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    Drug-induced TOP2 cleavage at KMT2A-family breakpoint cluster regions is reversible; religation of cleavage sites was demonstrated experimentally.

    DNA structural properties of AF9 are similar to MLL and could act as recombination hot spots resulting in MLL/AF9 translocations and leukemogenesis · Strissel PL, Strick R, Tomek RJ, Roe BA, Rowley JD, Zeleznik-Le NJ · Human Molecular Genetics 2000
    PMID 10861294DOI
    What this evidence cannot show

    Reversibility shown at the AF9 locus in cell lines; the quantitative half-life of a trapped complex in a primary human fetal progenitor is unknown.

    Evidence that argues against this (1)
    • Reversibility describes the majority outcome and says nothing about the rare tail that becomes a fusion, and the patient junctions show that tail was heavily processed rather than religated. In a therapy-related t(4;11) where both genomic breakpoint junctions were sequenced, the translocation breakpoints in MLL and AF-4 WERE functional topoisomerase II cleavage sites whose cleavage was enhanced in vitro by etoposide and its catechol and quinone metabolites but not by dactinomycin; recombination was near-precise, differing only by the gain of one templated nucleotide, and the authors conclude that exonucleolytic deletion and template-directed polymerisation must have processed the staggered topoisomerase II nicks before ligation. Their data are explicitly INCONSISTENT with a simple subunit-exchange/strand-transfer mechanism. So the events that matter are the ones where religation failed and the ends were enzymatically remodelled; a demonstration that cleavage complexes can reverse at AF9 does not bound that rate. Lovett BD et al. 2001
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Therapy-related KMT2A breakpoints concentrate in the telomeric, TOP2-consensus-rich half of the BCR while de novo breakpoints concentrate in the centromeric half, indicating distinct initiating mechanisms.

    PMID 8634439
    What this evidence cannot show

    Small sample (39 patients total) and pre-NGS mapping. The distinction has been reported by other groups but deserves re-testing with modern breakpoint sequencing across the Meyer/Marschalek cohort.

    Evidence that argues against this (2)
    • The de novo / therapy-related breakpoint dichotomy this evidence item rests on is contested. Gole and Wiesmuller state that rearrangements involving a short TELOMERIC region of under 1 kb of MLL are the most frequently observed molecular change in infant acute leukaemia as well as in secondary leukaemia - the opposite of the clean centromeric/telomeric split. If the same telomeric hotspot dominates both aetiologies, the mismatch argument in the section 'What this does NOT explain' loses much of its force and the distinction cannot carry an inference about initiating mechanism on its own. Gole B et al. 2015
    • Within de novo t(4;11) ALL alone, breakpoint position is stratified by AGE, not by aetiology: analysing 40 t(4;11) patients, Reichel et al. found significant differences in breakpoint distribution between patients below and above one year of age and defined two hotspot areas within the MLL BCR on that basis. Any dataset that compares de novo (infant-enriched) with therapy-related (older) cases therefore confounds aetiology with age, which is exactly the design of the 1996 study. Reichel M et al. 2001
  5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    Apoptosis-induced cleavage within the KMT2A BCR is not blocked by the catalytic TOP2 inhibitor merbarone, indicating the initial cleavage step in that pathway is TOP2-independent.

    PMID 16193084DOI
    What this evidence cannot show

    Single-group result in TK6 cells with episomal reporters; merbarone specificity is imperfect and residual TOP2 activity cannot be fully excluded.

    Evidence that argues against this (1)
    • A catalytic-inhibitor negative in one cell line does not generalise to patients, where sequenced breakpoints have been shown to sit at FUNCTIONAL topoisomerase II cleavage sites. In a therapy-related t(4;11), the der(11) and der(4) junctions both lay in MLL intron 6 and AF-4 intron 3, and DNA substrates containing the normal homologues of those breakpoints were cleaved by human topoisomerase IIalpha with cleavage enhanced by etoposide, etoposide catechol and etoposide quinone but not by dactinomycin — identifying the patient's breakpoints as bona fide TOP2 sites and the relevant exposure as etoposide or its metabolites. Merbarone-insensitive cleavage in TK6 episomes therefore establishes that AT LEAST ONE route to MLL breakage is TOP2-independent, not that the initial cleavage step is TOP2-independent in general. Lovett BD et al. 2001
Connected to