CTCF keeps CALM-AF10 leukaemia undifferentiated through TGM2 - an architectural dependency, not a menin one
also called CTCF requirement in CALM-AF10 · TGM2 in CALM-AF10
A protein that organises how DNA is folded, CTCF, is needed both to start and to keep CALM-AF10 leukaemia going in mice. It works by keeping one gene, TGM2, switched on, and blocking TGM2 makes the leukaemia cells mature into harmless macrophage-like cells.
The same thing, for a scientist
In mice, **CTCF deficiency significantly extends survival in CALM-AF10 AML**, and CTCF knockout reduces colony formation and drives differentiation into macrophage-like cells. RNA sequencing identifies **TGM2 (transglutaminase 2) as the most significantly downregulated gene**, and ChIP-seq shows that CTCF loss raises H3K27me3 and lowers H3K4me3 and H3K27ac at the Tgm2 transcription start site. Knocking down Tgm2, in murine CALM-AF10 AML cells **and in human U-937**, reproduces the phenotype, as does the TGM2 inhibitor GK921 [E1]. **This is the most explicit answer available to 'what holds the CALM-AF10 chromatin state open', and it is architectural** - a CTCF-dependent, polycomb-opposing configuration at a single downstream gene. It is not a menin answer, and menin is not examined.
Why it belongs in the menin argument
The menin-interface hypothesis says a fusion needs a tether that holds it at an address. CALM-AF10 has no menin tether, and the two candidate substitutes identified in this beat are both non-menin: CRM1 prebound at HOXA chromatin, and now CTCF-dependent architecture maintaining a permissive state at a key downstream target. Neither is a chromatin-reading module written into the fusion, which is what a KMT2A fusion has. If CALM-AF10 turns out menin-independent, this is the kind of mechanism that replaces menin in the account.
The honest limit: CTCF is required by a great many things, and a survival extension on CTCF deficiency is not by itself a fusion-specific dependency. What raises it above background is the TGM2 arm, which is specific, reproduced in a human line, and phenocopied by a small molecule.
| mechanism class | chromatin architecture / differentiation block |
|---|---|
| architectural factor | CTCF |
| effector gene | TGM2 (transglutaminase 2) |
| phenotype on loss | macrophage-like differentiation; reduced colony formation; extended survival in vivo |
| chemical matter | GK921 (TGM2 inhibitor) |
| human confirmation | TGM2 knockdown reproduced in U-937 |
| menin examined | false |
| relation to the menin question | One of two non-menin candidates identified in this beat for what supplies CALM-AF10 with a stable chromatin address; the other is chromatin-prebound CRM1. |
histone changes on ctcf loss
{
"H3K27me3": "increased at Tgm2 TSS",
"H3K4me3": "decreased",
"H3K27ac": "decreased"
}- CTCF is not druggable and is required nearly everywhere; the only actionable item here is TGM2, and GK921 is a tool compound with no clinical programme in leukaemia found in this reading.
- TGM2 has broad roles in extracellular matrix and apoptosis; a systemic TGM2 inhibitor's therapeutic window in a paediatric patient is entirely unbounded.
- Is the CTCF requirement specific to CALM-AF10 or general to leukaemia? The study reports no matched comparator fusion.
- Does TGM2 matter in PICALM::MLLT10 T-ALL, or only in the myeloid form? Both the model and the human confirmation line are myeloid.
- Does the CTCF-dependent state at Tgm2 have any counterpart at HOXA? The paper identifies TGM2 as the most downregulated gene, which is not the same as identifying it as the only relevant one.
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
In mice, CTCF deficiency significantly extends survival in CALM-AF10-induced AML; CTCF knockout reduces colony formation and induces macrophage-like differentiation; TGM2 is the most downregulated gene on RNA sequencing; CTCF loss raises H3K27me3 and lowers H3K4me3 and H3K27ac at the Tgm2 TSS; Tgm2 knockdown in murine CALM-AF10 AML cells and in human U-937 reduces colony formation and proliferation and promotes macrophage-like differentiation, as does the TGM2 inhibitor GK921.
CTCF Is Essential for the Development and Maintenance of CALM-AF10-Induced Leukemia. · Kuroki Y, et al. · Cancer Sci 2025What this evidence cannot show
Primarily a murine CALM-AF10 AML model, with the TGM2 arm confirmed in human U-937 - a myeloid line, in a disease that is 65% T-ALL. CTCF is a near-universal architectural factor and its requirement is not by itself fusion-specific; the fusion-specific content is the TGM2 axis. A single recent report, not yet independently replicated.
Evidence that argues against this (1)
- A CTCF requirement cannot be read as fusion-specific without a comparator: CTCF loss impairs many leukaemias and many normal tissues, and this study does not report a matched KMT2A-rearranged or non-leukaemic arm. Kuroki Y et al. 2025
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
CALM-AF10 activates HOXA genes at chromatin where CRM1 is already bound, and a CRM1-AF10 chimera reproduces leukaemia with HOXA activation.
Fusion of the CRM1 nuclear export receptor to AF10 causes leukemia and transcriptional activation of HOXA genes. · Aumann WK, et al. · Leukemia 2021What this evidence cannot show
Murine. Cited here as the other non-menin candidate for what supplies the address.
Evidence that argues against this (1)
- PUBLISHED ALTERNATIVE that makes this a three-way contest rather than a two-way one. CRM1-at-HOXA and CTCF-through-TGM2 are two candidate answers to 'what supplies the address', and a third has been mapped directly on this fusion: the MOZ/ENL complex recruits the DOT1L/AF10 fusion complex and maintains its chromatin retention through MOZ/MORF acetyltransferase activity, with ENL's YEATS-MOZ interaction critical for CALM-AF10 transformation. That is an acetylation-reader address on the fusion itself, not an architectural or a transport one, and it is the only one of the three demonstrated by a requirement experiment in CALM-AF10 cells. Naming CRM1 as 'the other non-menin candidate' omits the best-supported candidate. Komata Y et al. 2023
architectural dependency maintaining the differentiation block
CTCF loss releases the block into macrophage-like differentiation
- related to →CTCF/cohesin boundaries and the HOXA topological domain — the 3D scaffold the fusion program runs on
CTCF architecture at HOXA, approached from the CALM-AF10 side