Lost from the KMT2A half: both taspase1 cleavage sites (D2666 and D2718) - the fusion is never processed

Two things in the previous version of this entry were wrong and are corrected in the body, not only in the counter-evidence:

1. The loss is not functionally untested. The entry said the taspase1 axis was 'functionally untested in leukemia' and that a genetic tool existed which 'nobody has crossed into an MLL-AF9 or MLL-AF4 model'. The equivalent experiment was in fact done pharmacologically and published in 2018: CX-4945, a CKII inhibitor, blocks the phosphorylation that licenses taspase1 cleavage of MLL1, stabilises full-length wild-type MLL1, and in SEM cells - a KMT2A::AFF1 line - decreases genome-wide MLL-AFF1, AFF4 and ELL2 occupancy with decreased target-gene mRNA. In an MLL-AF9 syngeneic mouse it delayed progression and significantly extended survival by oral gavage and by intraperitoneal injection 6. The open question this entry posed has a partial answer, and the answer is positive.

2. 'Proteolytic maturation' is the wrong noun. The entry described cleavage as a maturation step the fusion skips. On the best-evidenced reading it is the opposite: cleavage primes MLL1 for degradation, and taspase1 loss leaves MLL1 uncleaved, longer-lived and more chromatin-bound, with localisation, bulk H3K4 methylation and COMPASS assembly all unchanged 6. The word 'maturation' carries the 1990s assumption that cleavage activates, which is exactly the assumption Zhao et al. set out to test. The entry now says 'processing' and states the direction explicitly.

What is genuinely contested. Takeda et al. reached the opposite conclusion from Taspase1-null mice - uncleaved precursor MLL has reduced HMT activity, H3K4me3 falls at Cyclin E1/E2, and non-cleavable MEFs proliferate poorly 7 - and a third study cited by Zhao found non-cleavable knock-in mice with no apparent defect at all. Three careful studies, three answers. This entry records the conflict and does not pick a winner, because the direction of the effect is load-bearing for everything downstream.

Taspase1 is a threonine aspartase that initiates a class of endopeptidases using an N-terminal threonine nucleophile and cleaving after aspartate; its proenzyme is intramolecularly processed into an active 28-kDa alpha / 22-kDa beta heterodimer. RNAi knockdown of taspase1 produces unprocessed MLL and loss of proper HOX gene expression. Taspase1 coevolved with MLL/trithorax as Arthropoda and Chordata emerged 1.

Cleavage generates N320 and C180, which heterodimerise. The FYRN domain of N320 directly interacts with the FYRC and SET domains of C180; disrupting that interaction markedly decreases N320 levels and redistributes C180 to a diffuse nuclear pattern. The model is that post-cleavage association confers stability on N320 and correct subnuclear localisation on the complex 2.

So the wild-type protein regulates itself by being taken apart and put back together. The fusion skips the entire process.

Follows directly (revised): the fusion protein's stability and subnuclear placement cannot be governed by the N320/C180 mechanism. Whatever governs them is partner-supplied or N-terminal. This is a testable, unexploited prediction about fusion half-life, and the graph already has a degron/half-life analysis (ana-fusion-halflife-degron-vs-menin) that could absorb it.

Does not follow: that taspase1 is irrelevant to KMT2A-rearranged leukemia. Taspase1 still cleaves the wild-type allele in the same cell, and non-cleavable MLL knock-in mice (MLL nc/nc) are protected from HER2/neu-driven breast tumorigenesis, establishing MLL as the primary taspase1 substrate whose cleavage is required in at least one cancer context 4. Whether the same is true in KMT2A-rearranged leukemia has not been reported, and that is a striking gap: a genetic tool exists (MLL nc/nc), the enzyme has published small-molecule inhibitors [E5], and nobody has crossed the tool into an MLL-AF9 or MLL-AF4 model. WITHDRAWN 2026-09-19 - this sentence was left standing when the cycle-31 correction was added below, so the entry asserted the gap and its closure in the same section. The perturbation has been made in KMT2A-rearranged material: blocking taspase1 cleavage of wild-type MLL1 displaces MLL-AFF1, AFF4 and ELL2 from chromatin in SEM cells and extends survival in an MLL-AF9 mouse 6. What remains genuinely untested is the narrow genetic version: the MLL nc/nc knock-in allele itself has never been crossed into an MLL-AF9 or MLL-AF4 mouse - as of 2026-09-19, taspase1[Title/Abstract] AND (MLL OR KMT2A)[Title/Abstract] returns 17 records and none reports that cross. Keep the narrow claim; the broad one is gone.

Also does not follow: that the fusion is 'unregulated'. It is unregulated by this mechanism. Absence of a known control is not absence of control.

Superseded, cycle 31. The paragraph above framed taspase1 as an untried target in KMT2A-rearranged leukemia. It has been tried, indirectly and successfully, and the route was not taspase1 itself: Zhao et al. report that their own virtual screen of 30 million compounds against the taspase1 structure produced 26 candidates and none of them worked, and that a 70% knockdown of taspase1 was not enough to change MLL1 cleavage at all 8. Taspase1 looks like a genuinely bad direct target. The tractable node is CKII, whose inhibitor CX-4945 (silmitasertib) is an orally available clinical-stage compound, and whose inhibition reaches the same endpoint through the substrate rather than the enzyme 6. That reverses the entry's previous recommendation, which pointed at 2009 covalent-warhead taspase1 chemistry.

One further negative result worth keeping, added 2026-09-19. Bier et al. asked whether catalytically dead taspase1 (T234V) or attenuated taspase1 (D233A) could act as a dominant negative in t(4;11) cells, and found that nine-fold overexpression inhibited neither cis- nor trans-cleavage, because inactive taspase1 does not form stable heterocomplexes with the wild-type enzyme 9. This is an explicit failed attempt to block the taspase1 axis in a KMT2A::AFF1 background, and it argues - as the authors say - for targeting catalytic activity rather than dimer assembly. It also reinforces why the successful route in 6 went around taspase1 via CKII.