In the only panel that tested both halves of four translocations, 3 of 4 reciprocal fusions transformed and only 1 of 4 direct ones did
also called 1 of 4 direct, 3 of 4 reciprocal · Sleeping Beauty MLL fusion panel · NEBL LASP1 MAML2 SMAP1
Somebody built both halves of four different chromosome swaps and tested each one for its ability to transform cells. Three of the four mirror-image halves worked. Only one of the four famous halves did.
The same thing, for a scientist
A Sleeping Beauty transposon system was built to express full-length MLL fusion cDNAs, which are too large for retroviral packaging, and used to test four partner pairs in both orientations: MLL-NEBL / NEBL-MLL, MLL-LASP1 / LASP1-MLL, MLL-MAML2 / MAML2-MLL and MLL-SMAP1 / SMAP1-MLL. Readouts were 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** [E1]. The authors describe this as pilot data motivating systematic analysis, and that framing should be kept: four partner pairs, rare partners, cell-line transformation surrogates, one laboratory, and no leukaemia. But it is the only symmetric test that exists - the only experiment in which direct and reciprocal products of the same translocations were assayed side by side under identical conditions - and it came out against the direct product. Set beside the t(4;11) result where AF4-MLL alone induced murine leukaemia and MLL-AF4 alone did not [E2], and the recombinome's observation that 10 of 102 KMT2A fusion genes are out of frame on the direct side [E3], the pattern is at least worth stating: the assumption that the direct fusion is the oncogenic one is an assumption, and where it has been tested symmetrically it has not held up well.
Why the symmetry of the design is what matters
Almost every functional experiment in this field tests a direct fusion. When a reciprocal product is tested at all it is usually in a separate paper with a different system, so a comparison across them confounds everything. Here the same vector, the same cells and the same four readouts were applied to eight constructs 1. Whatever the result's other limits, the direct-versus-reciprocal comparison inside it is internally controlled, and that is rare enough to be worth an entry on its own.
What would make this a finding rather than a curiosity
The four partners are NEBL, LASP1, MAML2 and SMAP1 - none is a common KMT2A partner and none is AFF1. Running the same symmetric panel on AFF1, MLLT3, MLLT1 and MLLT10 would settle whether the asymmetry generalises. The system to do it already exists and was built for exactly this purpose; the authors say so in their own conclusion.
| direct oncogenic | 1 of 4 |
|---|---|
| reciprocal oncogenic | 3 of 4 |
| system | Sleeping Beauty transposon, stable cell lines |
| partners tested are common | false |
| design property | the only symmetric direct-versus-reciprocal comparison in the KMT2A literature |
panel
[ "MLL-NEBL / NEBL-MLL", "MLL-LASP1 / LASP1-MLL", "MLL-MAML2 / MAML2-MLL", "MLL-SMAP1 / SMAP1-MLL" ]
readouts
[ "growth behaviour", "focus formation", "colony formation", "ectopic Hoxa transcription" ]
- If reciprocal products transform more readily than direct ones in a controlled comparison, then the entire target-selection logic of this field - aim at the direct fusion's menin interface - rests on an assumption that has been tested symmetrically once, in four rare partners, with the opposite result.
- The more deflationary reading is that focus formation measures something KMT2A direct fusions are simply bad at, in which case the assay and not the biology produced the asymmetry - and that reading is equally unexamined.
- Which of the four direct fusions was the one that scored positive? The abstract does not say and it matters for interpreting the asymmetry.
- Would the same symmetric panel run on AFF1, MLLT3, MLLT1 and MLLT10 reproduce the asymmetry? The system exists and the authors called for exactly this.
- Is focus formation in a cell line a meaningful surrogate for leukaemogenesis by a KMT2A fusion? For the direct fusions the answer is probably no, since MLL-AF4 is notoriously hard to transform with in any system - which may be what this panel is actually measuring.
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Using a Sleeping Beauty transposon vector system with no cargo size limit and no integration preference, MLL-NEBL, NEBL-MLL, MLL-LASP1, LASP1-MLL, MLL-MAML2, MAML2-MLL, MLL-SMAP1 and SMAP1-MLL were expressed in stable cell lines and assayed for growth behaviour, focus formation, colony formation and ectopic Hoxa transcription; only 1 of 4 tested direct MLL fusions but 3 of 4 tested 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 2014What this evidence cannot show
Explicitly pilot experiments in the authors' own words; four rare partners, none of them AFF1, MLLT3, MLLT1 or MLLT10; transformation surrogates rather than leukaemia; cell lines rather than haematopoietic progenitors; one laboratory, the same one that argues the reciprocal product matters, so the choice of design is not independent of the conclusion. Which single direct fusion scored positive is not stated in the abstract.
Evidence that argues against this (1)
- The best-powered test of the same proposition in the index fusion gives the opposite answer: removing the reciprocal transcript from human t(4;11) lines had no effect while removing the direct fusion inhibited growth and induced apoptosis. Kumar AR et al. 2011
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
AF4.MLL expressed in murine Lin-/Sca1+ progenitors induced pro-B ALL, B/T biphenotypic or mixed-lineage leukaemia, whereas MLL.AF4 alone caused no disease over 13 months.
The AF4.MLL fusion protein is capable of inducing ALL in mice without requirement of MLL.AF4. · Bursen A, Schwabe K, Rüster B, et al. · Blood 2010What this evidence cannot show
Murine, retroviral, partially penetrant; a different system from the Sleeping Beauty panel, so the two are concordant in direction rather than replicating each other.
Evidence that argues against this (1)
- The same construct in human cord-blood CD34+ cells produced enhanced engraftment but no serial self-renewal and no leukaemia within 36 weeks. Prieto C et al. 2017
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Of 102 KMT2A fusion genes characterised in the Frankfurt series, 10 were per se out of frame on the direct side, and 114 of 2,454 patients were diagnosed with the reciprocal allele alone and no KMT2A-X allele.
The reciprocal world of MLL fusions: A personal view. · Marschalek R · Biochim Biophys Acta Gene Regul Mech 2020What this evidence cannot show
Single-centre referral series framed as a personal view; out-of-frame status is assessed at the DNA level and does not exclude alternative splicing restoring a reading frame.
Evidence that argues against this (1)
- An out-of-frame direct fusion does not establish that the reciprocal one drives the disease; the leukaemia could be driven by KMT2A haploinsufficiency or by the partner gene's disruption. Kowarz E et al. 2007
The only symmetric direct-versus-reciprocal transformation comparison, and it favours reciprocal products.
Points the opposite way to the t(4;11) knockdown, in a symmetric design but in rare partners and a transformation surrogate.
- supports →About one KMT2A-rearranged patient in twenty has no direct fusion at all - only the reciprocal one
Supplies a functional rationale for why a reciprocal-only patient could have leukaemia at all.
A second, independent system in which reciprocal products show transforming activity.