Losing the AF10 PZP reader is not a missing function - it is a removed brake, and putting it back cures the fusion
also called AF10 PZP loss · PZP restoration
The normal AF10 protein has a module that reads the histone tail and keeps DOT1L in check. Every leukaemia version of AF10 has lost it - and if you engineer it back into the fusion, the fusion stops causing leukaemia.
The same thing, for a scientist
The **PZP** module (PHD finger-Zn knuckle-PHD finger) of AF10 folds as a **single unit** that recognises histone H3 residues 22-27 and is abrogated by H3K27 methylation or acetylation; in cells, PZP recognition of H3 is required for H3K79 dimethylation, for expression of DOT1L target genes and for proliferation of DOT1L-addicted leukaemic cells [E1]. **It is consistently impaired or deleted in leukaemogenic AF10 translocations** [E2]. The decisive experiment is a cis restoration: **incorporating a functional AF10 PZP into CALM-AF10 prevents transformation of murine HSPCs in vitro and in vivo, abrogates leukaemogenesis, downregulates Hoxa genes and reverses established transformation** [E2]. So this is a *removed brake*, not merely a lost accessory - and it is a rare case in fusion biology where restoring a lost domain is therapeutic in a model.
The strongest counter-argument, which this graph already carries
The PZP-as-brake reading is not the only one, and gene-mllt10 already records the tension. Addition suffices: transformation by both MLL-AF10 and CALM-AF10 is Dot1l-dependent, which is an account in which the retained OM-LZ does the work and the lost PZP is incidental 3. And endogenous, PZP-containing AF10 is REQUIRED for transformation by MLL-AF9, MLL-AF6 and NUP98-NSD1 4 - so the same module that acts as a brake when restored in cis to CALM-AF10 is a requirement in trans for three other fusions. A model has to accommodate both, and the obvious reconciliation - that PZP controls where DOT1L acts rather than whether it acts, so its presence helps a fusion that has its own address and hinders one that does not - is a hypothesis, not a result.
The second and more practical caveat: the restoration construct is engineered cis restoration, not a patient allele. It shows PZP loss is permissive for transformation; it does not show PZP loss is the initiating event, and nothing in this beat's reading rules out that the PZP is simply on the wrong side of a breakpoint chosen for other reasons.
What it implies about the chromatin address
PZP reads unmodified H3K27, i.e. it is an anti-polycomb address. A fusion that has lost it can no longer be excluded from H3K27me3 domains by that mechanism. Whether CALM-AF10 therefore deposits H3K79me2 in territory endogenous AF10 would avoid is exactly the kind of locus-resolved question nobody has answered, and it is the same question this graph asks of the KMT2A CXXC module from the other direction.
| mechanism class | loss of a regulatory reader module |
|---|---|
| module | AF10 PZP (PHD-Zn knuckle-PHD), P55197-1 residues approximately 22-198 |
| ligand | unmodified H3K27, read via H3 residues 22-27; discriminates against H3K27me3 |
| fate in fusion | lost (PHD1 lost; the 2026 review records that CALM-AF10 retains the second PHD) |
| cis restoration effect | prevents and reverses transformation; downregulates Hoxa; promotes nuclear localisation and chromatin association |
| trans requirement elsewhere | endogenous PZP-containing AF10 is required for MLL-AF9, MLL-AF6 and NUP98-NSD1 transformation |
| unresolved | the same module is a brake in cis on CALM-AF10 and a requirement in trans for three other fusions; the reconciliation (PZP controls WHERE rather than WHETHER DOT1L acts) is a hypothesis |
| therapeutic shape | restoring a lost reader module is not a druggable modality, but a molecule that re-imposes H3K27-dependent restriction on the fusion complex would be the pharmacological equivalent, and none exists |
- No intervention follows directly: you cannot give a patient back a deleted domain. The therapeutic reading is indirect and depends on the unproven 'PZP controls where DOT1L acts' model.
- If the addition-suffices account is right, effort spent on PZP biology is effort not spent on the OM-LZ interface, which is the one with structures and chemical matter.
- Does CALM-AF10 deposit H3K79me2 in chromatin that endogenous AF10 would be excluded from? A PZP-restored versus PZP-less H3K79me2 map answers it and would convert the brake model from inference to measurement.
- Why is the same module a brake in cis and a requirement in trans? The 'addressing rather than activating' reconciliation is untested.
- Are there AF10 translocations that RETAIN an intact PZP, and are they non-leukaemogenic? The claim that PZP is 'consistently impaired or deleted' would be strongest if a counterexample were sought and not found.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
An AF10 region comprising PHD finger-Zn knuckle-PHD finger folds into a single module that recognises H3 residues 22-27, with binding abrogated by H3K27 modification; in cells, PZP recognition of H3 is required for H3K79 dimethylation, for DOT1L target-gene expression and for proliferation of DOT1L-addicted leukaemic cells.
The PZP Domain of AF10 Senses Unmodified H3K27 to Regulate DOT1L-Mediated Methylation of H3K79. · Chen S, et al. · Mol Cell 2015What this evidence cannot show
Structural plus cellular; the cellular arm is in DOT1L-addicted leukaemia lines, i.e. KMT2A-rearranged, not CALM-AF10.
Evidence that argues against this (1)
- PUBLISHED CONFLICTING RESULT against the brake metaphor itself. A removed brake predicts that putting it back restores normal regulation. What is observed is a redirection: restoring a functional AF10 PZP domain to CALM-AF10 promotes NUCLEAR localisation of the fusion, is REQUIRED for its chromatin association, downregulates Hoxa genes and abolishes transformation and leukaemogenesis. A fusion with PZP restored is not a normal cell - it is a fusion that is more nuclear and more chromatin-bound and less oncogenic. That is a change of address, not a restored inhibitory constraint, and the brake framing predicts neither the increased chromatin association nor the loss of transformation. Klein BJ et al. 2021
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The PZP domain of AF10 is consistently impaired or deleted in leukaemogenic AF10 translocations, and incorporating a functional AF10 PZP into CALM-AF10 prevents transformation of bone-marrow HSPCs in vitro and in vivo, abrogates leukaemogenesis, downregulates Hoxa genes and reverses cellular transformation; AF10 PZP binds the nucleosome core particle through multivalent H3-tail and DNA contacts, colocalises with active methylation marks and discriminates against H3K27me3, and promotes nuclear localisation of CALM-AF10.
The role of the PZP domain of AF10 in acute leukemia driven by AF10 translocations. · Klein BJ, et al. · Nat Commun 2021What this evidence cannot show
Murine HSPCs with an engineered cis restoration. It establishes that PZP loss is permissive; it does not establish that PZP loss is the initiating or the decisive event.
Evidence that argues against this (1)
- An 'addition suffices' account explains the same data without needing a brake: transformation by both MLL-AF10 and CALM-AF10 is Dot1l-dependent, so the retained OM-LZ may do all the work. Chen L et al. 2013
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
MLL-AF10- and CALM-AF10-mediated transformation is Dot1l-dependent by both genetic knockout and pharmacological inhibition.
Abrogation of MLL-AF10 and CALM-AF10-mediated transformation through genetic inactivation or pharmacological inhibition of the H3K79 methyltransferase Dot1l. · Chen L, et al. · Leukemia 2013What this evidence cannot show
Murine. Cited here as the rival 'addition suffices' reading.
Evidence that argues against this (1)
- PUBLISHED CONFLICTING RESULT against the rival reading as well. 'Addition of the mark suffices' attributes the DOT1L dependence to the H3K79 that gets added at HOXA, and at achievable inhibitor exposures that is not the operative route in this leukaemia class: low-dose pinometostat is cytotoxic far below the concentrations that downregulate HOXA9 and MEIS1, the major pathway is transcriptional downregulation of FLT3, and constitutively active STAT5A largely rescues. A genetic Dot1l knockout cannot separate the routes because it removes all of them. So the cis-brake and addition-suffices readings are being compared using an instrument that collapses the very distinction at issue. Richter WF et al. 2021
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
AF10 inactivation reverses leukaemia-associated epigenetic profiles, precludes abnormal HOXA expression and impairs transformation by MLL-AF9, MLL-AF6 and NUP98-NSD1 - mechanistically distinct HOX-activating oncogenes - and AF10 regulates the H3K79me1-to-me2 conversion at HOX loci.
AF10 regulates progressive H3K79 methylation and HOX gene expression in diverse AML subtypes. · Deshpande AJ, et al. · Cancer Cell 2014What this evidence cannot show
Murine and human AML systems; CALM-AF10 is not among the fusions tested. This is the trans requirement that sits awkwardly beside the cis brake.
Evidence that argues against this (1)
- PUBLISHED ALTERNATIVE that reconciles the trans requirement with the cis brake, and which nobody has tested. AF10's requirement in trans across MLL-AF9, MLL-AF6 and NUP98-NSD1 need not be a requirement for AF10's own PZP reader function: the MOZ/ENL complex is reported as the recruiting factor for the AF10 fusion complex, maintaining its chromatin retention through MOZ/MORF acetyltransferase activity, with ENL's YEATS-domain interaction with MOZ critical for CALM-AF10 transformation. If AF10's trans role is to be part of a MOZ/ENL-anchored assembly, then losing PZP in cis and requiring AF10 in trans are statements about different modules of the same protein and there is no paradox to resolve - which also means this entry's tension is an artefact of treating AF10 as one functional unit. Komata Y et al. 2023
what the 3' partner loses at the break
PZP module of AF10, lost in the fusion
- related to →MLLT10 (AF10): the discarded PZP reader — the one partner where loss is proven to be the oncogenic event
the same subtraction, tested by cis restoration in CALM-AF10
- supports →Re-supply the AF10 PZP module in cis to KMT2A::MLLT10 and ask whether transformation is suppressed
the CALM-AF10 cis-restoration result that motivates the KMT2A::MLLT10 version
- related to →PICALM::MLLT10 (CALM-AF10), t(10;11)(p13;q14) - the DOT1L-dependent HOXA fusion that contains no KMT2A
restoring PZP makes the fusion MORE nuclear and MORE chromatin-bound and LESS transforming - a change of address, not a restored brake