Premise (iii) was not falsified at this locus - Gothe 2019 measures MLL breakage and MLL-AF4 fusion in the same cells and they move TOGETHER under TDP2 loss; what TDP2 actually falsifies is a premise about break FORMATION that the triad never stated
also called the premise (iii) correction · TDP2 raises breaks and fusions together · formation is not persistence
This project believed that losing the repair enzyme TDP2 raises the number of cancer-causing rearrangements without raising the number of breaks, which would break the chain of reasoning it uses. In the one paper that counts both in the same cells at this gene, they go up together. The real problem is subtler: a break that is counted hours later is a break that failed to heal, not a break that formed.
The same thing, for a scientist
The Q105 triad's premise (iii) is *DSBs up implies rearrangement up*, and this beat's dispatch brief recorded it as **"already falsified at this locus"** by PMID 28794467 (TDP2 loss raises TOP2-induced translocations at *MLL* by changing rejoining, not breakage). **That is not what either paper says.** Gomez-Herreros 2017's own abstract states that **"TDP2 rejoins DSBs induced during transcription-dependent TOP2 activity in breast cancer cells and at the translocation 'hotspot', MLL"** [E1] - so TDP2 loss leaves *unrejoined* breaks, and a break-counting assay run after release will count more of them. And Gothe 2019 measures exactly that, in the same cells, at *MLL*, with both readouts: **"A similar increase in the percentage of cells with *MLL* breakage and *MLL-AF4* translocations was observed upon ETO-treatment in cells deficient for TDP2 or MRE11 nuclease activity"**, and **"TDP2 prevented the accumulation of ETO-induced DSBs within the *MLL* locus and the formation of *MLL* fusions"** [E2]. **Breakage and fusion co-move. Premise (iii) is supported at this locus in human cells, not falsified.** What TDP2 does refute is the *unstated* premise that the quantity an assay reports is break FORMATION: END-seq, sBLISS and C-Fusion 3D all count breaks that have not yet been repaired, so they read formation and persistence multiplied together [E1][E2].
The two quotes side by side
Gomez-Herreros et al. 2017 (PMID 28794467), abstract, verbatim 1:
"TOP2-induced DNA double-strand breaks are rejoined in part by tyrosyl-DNA phosphodiesterase 2 (TDP2)-dependent non-homologous end-joining (NHEJ), but whether this process suppresses or promotes TOP2-induced translocations is unclear. Here, we show that TDP2 rejoins DSBs induced during transcription-dependent TOP2 activity in breast cancer cells and at the translocation 'hotspot', MLL. Moreover, we find that TDP2 suppresses chromosome rearrangements induced by TOP2 and reduces TOP2-induced chromosome translocations that arise during gene transcription."
The mechanism named is rejoining. Nothing in it says breakage is unchanged; it says the step TDP2 performs is the ligation step. A cell that cannot rejoin accumulates breaks.
Gothe et al. 2019 (PMID 31202576), Results, verbatim 2:
"A similar increase in the percentage of cells with MLL breakage and MLL-AF4 translocations was observed upon ETO-treatment in cells deficient for TDP2 or MRE11 nuclease activity (Figure 7A,B). These findings indicate that TDP2 and the nuclease activity of MRE11 act as suppressors of MLL translocations by removing prone to breakage TOP2-DNA covalent complexes upon ETO-treatment..."
"...TDP2 prevented the accumulation of ETO-induced DSBs within the MLL locus and the formation of MLL fusions (Figure 7A,B), an effect that can be attributed to TDP2 functions in removing TOP2ccs accumulating predominantly during transcription (Figure 7C,D)."
Same cells (TK6, isogenic TDP2-/-), same locus (MLL), same assay family (4-color C-Fusion 3D), both quantities, and they rise together.
Gothe already performs the formation-versus-fate separation the graph proposes for CTCF
exp-ctcf-degron-breaks-and-rearrangements-same-cells-kmt2a proposes measuring breaks and rearrangements in the same cells under an occupancy perturbation. Gothe 2019 does exactly that design with a repair perturbation instead, and its isogenic TK6 panel is the machinery: TDP2-/-, MRE11-/-, MRE11-/H129N (nuclease-dead), MRE11 loxP, CtIP-/- / CtIP auxin-degron, LigIV-/-, RAD54-/-, LigIV-/- RAD54-/-, and the double TDP2-/-/MRE11-/H129N 2.
And the result decomposes cleanly along the two pathways:
| perturbation | MLL breakage | MLL-AF4 fusion | dependency |
|---|---|---|---|
TDP2-/- | up | up | transcription-dependent, replication-independent (DRB abolishes, aphidicolin does not) |
MRE11-/H129N | up | up | both transcription- and replication-dependent |
MRE11-/H129N TDP2-/- | up | up | both |
NHEJ-deficient (LigIV-/-) | - | up | "DSB repair by NHEJ suppresses the occurrence of MLL fusions by favoring intrachromosomal repair" |
HR-deficient (RAD54-/-, CtIP-/-) | - | no effect on repair kinetics | "TOP2-induced DSBs are predominantly repaired by NHEJ" |
The one genuine formation-versus-fate dissociation in the table is NHEJ, not TDP2: LigIV loss raises fusions by rerouting ends, which is a fate effect. TDP2 and MRE11 raise both, which is a persistence effect that propagates to fate. So the triad's premise (iii) is not the weak link the brief thought it was, and the right way to attack the triad is premise (i), which is unaddressed - see ana-hct116-is-the-wrong-cell-line-for-the-cohesin-arm-at-kmt2a.
And the transcription result at MLL, which is Gothe's headline and is directional
Figure 5 of Gothe 2019 is titled, verbatim, "The formation of MLL fusions is largely transcription-dependent" 2. The design: TK6 pre-treated with DRB 200 uM 3 h or aphidicolin 8 uM 3 h before etoposide 20 uM 4 h, then two days of release, then 4-color C-Fusion 3D for breakage and for each of MLL-AF4, MLL-AF6, MLL-AF9, MLL-ENL and the NPM1-ALK control. Result: "inhibition of transcription elongation by DRB, also reduced the frequency of MLL fusions with the various partners to almost background levels", and "Inhibition of transcriptional elongation, but not of replication, led to a great reduction in MLL breakage and translocations in TDP2 null cells" 2.
And the control that makes it mechanistic rather than an artefact of proximity: "In absence of DNA damage, inhibition of transcription by DRB, did not influence the percentage of cells with synapsed MLL and partner genes... suggesting that transcription does not contribute to the spatial proximity between MLL and the fusion partners." So transcription acts on breakage and on conversion, not on the 3D encounter 2.
This is the paper-level statement whose single-locus counterpart this beat measured in Gothe's own deposited sBLISS: DRB moves the KMT2A exon-11 element from the 98.016th to the 70.995th percentile of chr11 (see ana-kmt2a-exon-11-element-is-a-top-0-05-percent-etoposide-dsb-site). The paper's figure and the deposited track agree, at the element.
| status | run |
|---|---|
| result summary | The dispatch brief's premise that TDP2 falsifies triad premise (iii) at this locus is wrong. Gomez-Herreros 2017's mechanism is rejoining, which predicts more persistent breaks as well as more translocations; Gothe 2019 measures both at MLL in isogenic TDP2-/- TK6 and they rise together. The genuine formation-versus-fate dissociation in Gothe's panel is LigIV/NHEJ, not TDP2. What TDP2 does expose is that END-seq, sBLISS and C-Fusion 3D all report formation multiplied by persistence, so premise (iii) needs splitting into (iii-a) formation-to-rearrangement and (iii-b) persistence-to-rearrangement, of which only (iii-b) has evidence. |
| premise iii status | supported for the persistence reading; untested for the formation reading; the brief's 'already falsified' is withdrawn |
inputs
[ "PMID 28794467 abstract (efetch db=pubmed rettype=abstract)", "bioRxiv 10.1101/485763 v1 full text, Results and Figure 5 and 7 legends" ]
- Withdrawing the TDP2 falsification makes the triad harder to break, not easier, which is the opposite of what a project wants to hear. It also means the graph's current reading of mechanism-tdp2-repair-of-trapped-top2-complexes should be checked: if that entry says TDP2 changes fate without changing breakage, it carries the same error and needs the Gothe quote added.
- TDP2 inhibitors are in development as etoposide potentiators. This entry's reading implies they would raise MLL breakage and MLL fusion frequency together - a leukaemogenic risk in a paediatric context, and one that the transcription-dependence result says would be worst in cells where KMT2A is most highly expressed.
- Does PMID 28794467's full text report a breakage measurement at MLL under TDP2 loss, and in which direction? PMC5550487 is open access and unread; this is the one fetch that would settle whether the brief's paraphrase had any basis in the paper's body.
- Is there any perturbation anywhere that raises break FORMATION at a locus without raising rearrangement, or vice versa, measured in the same cells? Gothe's LigIV-/- arm is the closest candidate and its breakage panel was not extracted here.
- Every assay in this dispute counts unrepaired breaks. Is there a formation-rate readout at all - e.g. TOP2 cleavage-complex trapping by the ExoT-versus-ExoVII END-seq contrast, which Canela 2019 ran for every mouse condition and no human one?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
PMID 28794467's own abstract attributes TDP2's protective role to REJOINING, not to preventing break formation - 'TOP2-induced DNA double-strand breaks are rejoined in part by tyrosyl-DNA phosphodiesterase 2 (TDP2)-dependent non-homologous end-joining (NHEJ)' and 'we show that TDP2 rejoins DSBs induced during transcription-dependent TOP2 activity in breast cancer cells and at the translocation hotspot, MLL' - so TDP2 loss is expected to raise the count of unrejoined breaks as well as the translocation rate, and the paper additionally implicates TDP2-dependent NHEJ in FORMING a rare 4-bp-microhomology translocation subclass previously associated with therapy-related leukaemia.
TDP2 suppresses chromosomal translocations induced by DNA topoisomerase II during gene transcription · Gomez-Herreros F, Zagnoli-Vieira G, Ntai I, Martinez-Macias MI, Anderson RM, Herrero-Ruiz A, Caldecott KW · Nature Communications 8:233 2017What this evidence cannot show
This claim is made from the ABSTRACT, which was fetched verbatim; the full text at PMC5550487 was not read for this entry, so the paper's own breakage measurements at MLL under TDP2 loss - whether they went up, stayed flat, or were not made - are not established here. That is precisely the gap that made the dispatch brief's paraphrase possible, and the honest position is that the abstract names rejoining as the mechanism and says nothing either way about breakage. The system is breast cancer cells plus an MLL hotspot readout, not a haematopoietic cell.
Evidence that argues against this (1)
- The abstract also carries the finding that cuts the other way and that the brief's framing may have been reaching for: TDP2-dependent NHEJ is implicated in FORMING the 4-bp-perfect-homology translocation subclass associated with therapy-related leukaemia. So TDP2 has opposite signs for different junction classes, and a claim that TDP2 loss simply raises translocations is too coarse. That genuine subtlety is a statement about junction architecture, not about breakage versus rejoining, and it does not restore the brief's version. Gomez-Herreros F et al. 2017
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.preprint
Gothe 2019 measures MLL breakage and MLL-AF4 fusion frequency in the same isogenic TK6 cells by 4-color C-Fusion 3D and reports them rising together on TDP2 or MRE11-nuclease loss - 'A similar increase in the percentage of cells with MLL breakage and MLL-AF4 translocations was observed upon ETO-treatment in cells deficient for TDP2 or MRE11 nuclease activity' and 'TDP2 prevented the accumulation of ETO-induced DSBs within the MLL locus and the formation of MLL fusions' - with the TDP2 effect transcription-dependent and replication-independent and the MRE11 effect dependent on both, NHEJ loss (LigIV-/-) raising fusions, HR loss (RAD54-/-, CtIP-/-) leaving repair kinetics unchanged, and Figure 5 titled 'The formation of MLL fusions is largely transcription-dependent' on the finding that DRB 200 uM 3 h before etoposide 20 uM 4 h 'reduced the frequency of MLL fusions with the various partners to almost background levels' while not changing MLL-partner synapsis in undamaged cells.
Spatial chromosome folding and active transcription drive DNA fragility and formation of oncogenic MLL translocations - Results and Figures 5 and 7 with legends, complete preprint full text read 2026-09-22 · Gothe HJ, Bouwman BAM, Gusmao EG, Piccinno R, Sayols S, Drechsel O, Petrosino G, Minneker V, Josipovic N, Mizi A, Nielsen CF, Wagner EM, Takeda S, Sasanuma H, Hudson DF, Kindler T, Baranello L, Papantonis A, Crosetto N, Roukos V · bioRxiv (published as Mol Cell 75:267-283.e12, PMID 31202576) 2018What this evidence cannot show
Read from the December 2018 preprint v1, not the published July 2019 version; Figure 7 panel values were not extracted, only the legends and the Results sentences, so the effect SIZES for breakage versus fusion are not quoted and the claim is only that both rise. C-Fusion 3D scores breakage as separation of 5' and 3' MLL probes and fusion as co-localisation of MLL and partner probes by imaging - whole-gene resolution, not a sequenced junction - so 'breakage' here is probe separation persisting two days after release, i.e. unrepaired breakage, and 'fusion' is an imaging inference. Every arm is TK6, a human lymphoblastoid line, with etoposide at 10-20 uM; there is no infant, no B-precursor and no KMT2A-rearranged system. Cells analysed per sample were 1,500-16,500 across at least four independent experiments.
Evidence that argues against this (1)
- The strongest case for the brief's original reading is that C-Fusion 3D cannot distinguish 'more breaks formed' from 'the same breaks still visible two days later'. A persistence-only effect would present exactly as observed - both readouts up - because a fusion is scored at the same two-day timepoint as the breakage. So this entry establishes that breakage and fusion CO-MOVE and does NOT establish that formation rose; the claim that premise (iii) is 'supported' must be read as supported for the persistence reading of DSBs, which is the reading every assay in this dispute actually delivers. Gothe HJ et al. 2018
TDP2 loss raises MLL breakage AND MLL-AF4 fusions together in isogenic TK6; a rejoining mechanism predicts both, so it does not dissociate them
isogenic TDP2-null TK6: MLL breakage and MLL-AF4 fusion both rise, transcription-dependent and replication-independent
Gothe already runs this design with a repair perturbation instead of an occupancy one; the genuine formation-versus-fate dissociation in its panel is NHEJ, not TDP2
- related to →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
MLL-AF4 fusion formation measured directly by 4-color C-Fusion 3D in TK6, K562, HCT116, HT1080, Cal51 and primary CD34-positive cells at 6e-4 to 2e-3
every quote here comes from the full text that route audit retrieved
the TDP2 co-movement is the worked example that splits premise (iii)