You can lower LEDGF without a LEDGF ligand: its protein half-life is about 4 hours against a proteome average of 20, so blocking guanine nucleotide or rRNA synthesis takes it out first
also called IMPDH2 LEDGF · MMF CX-5461 LEDGF · indirect LEDGF lowering
LEDGF is an unusually short-lived protein, so anything that slows down protein-making removes LEDGF before it removes most other proteins. Two existing drugs do exactly that, and in mice they lower LEDGF, break up the leukaemia complex and make the menin drug work better.
The same thing, for a scientist
The LEDGF axis has no direct ligand, but it has an exploitable pharmacokinetic property. LEDGF protein turns over in **approximately 4 hours** against a proteome average of about 20, so a global suppression of translation depletes LEDGF disproportionately. Two mechanistically distinct agents do it: **mycophenolate mofetil** (IMPDH inhibitor, blocking guanine nucleotide biosynthesis) and **CX-5461** (RNA polymerase I inhibitor, blocking rRNA transcription). Both reduce LEDGF and menin protein dose-dependently in MLL-AF9 leukaemia stem cells without changing their half-lives - the effect is on synthesis, not degradation - and immunoprecipitation of an equal amount of menin recovers **almost no LEDGF** after either treatment. Menin occupancy at the Hoxa9 promoter falls, MLL target gene expression falls, myeloid differentiation follows, and the combination with the menin inhibitor VTP-50469 is synergistic in MOLM-13 and MV4-11. The bypass experiment is the cleanest part: a **PCE chimera** - LEDGF PWWP + MLL CXXC + ENL - which supplies the reader and the elongation contact in cis, rendered MOLM-13 resistant to MMF, CX-5461 **and** the menin inhibitor. This is currently the only intervention that lowers LEDGF in established KMT2A-rearranged disease in an animal.
Why LEDGF specifically
A global translation block should lower everything. It lowers LEDGF more, and the paper gives the reason: LEDGF's turnover rate of about 4 hours is faster than most of the proteome, whose average is about 20 hours 1. A short-lived protein sits at a steady state set almost entirely by its synthesis rate, so halving synthesis halves the protein within a few hours; a long-lived protein barely moves on the same timescale.
That is a general principle this graph should carry beyond LEDGF. Protein half-life is a selectivity axis for translation-directed and transcription-directed agents, and it is measurable. If LEDGF is unusually short-lived, the obvious follow-up question is which other members of the menin-KMT2A module are - menin also fell in these experiments - and whether a half-life census of the complex would predict which nodes a translation-blocking agent hits.
The caveat is the mirror image: every other short-lived protein in every other tissue is hit too, which is why MMF's dose-limiting problem in this setting is neutropenia, an unhappy toxicity in a leukaemia patient. The authors name it and offer CX-5461 as the alternative, noting it was tolerated in clinical trials for haematologic and solid cancers without evidence of neutropenia 1.
The bypass experiment, and what it proves
A minimal chimera of the LEDGF PWWP domain + MLL CXXC domain + ENL (PCE) transforms HSPCs like full-length MLL-ENL. Ectopically expressed in MOLM-13, PCE rendered the cells resistant to MMF, to CX-5461 and to the menin inhibitor VTP-50469 1. Separately, forced Hoxa9-Meis1 expression made THP-1 resistant to MMF-induced differentiation.
Two conclusions, and they point in opposite directions for this beat:
- For the mechanism: the drugs work by removing the LEDGF-menin bridge, because a construct that supplies the reader and the SEC contact in cis is immune to them. That is a strong genetic argument that the LEDGF/menin module is the operative target of a metabolic agent, which is not obvious a priori.
- Against the target: the same experiment is a blueprint for resistance. Anything that supplies the H3K36me2/3 reader in cis to the fusion - or that raises HOXA9/MEIS1 output by another route - escapes LEDGF-directed therapy entirely. The PCE construct is artificial, but HOXA9-MEIS1 amplification is not.
What it does and does not say about the LEDGF hole
This is the closest thing in the literature to removing LEDGF from established KMT2A-rearranged leukaemia in an animal, and it is close in a useful way: the treatment is given to leukaemia stem cells and to established models, the LEDGF reduction is measured, and the outcome is differentiation plus menin-inhibitor sensitisation.
It does not close the hole, for the reason that always applies to indirect interventions. MMF and CX-5461 lower LEDGF and menin and the general proteome, so the benefit cannot be attributed to LEDGF. The PCE bypass narrows this considerably - it says the effect runs through the LEDGF/menin module - but PCE supplies the PWWP and the ENL contact, so it also bypasses menin loss and SEC recruitment failure at the same time. A LEDGF-only rescue - shRNA-resistant full-length PSIP1 overexpressed under MMF - would separate them and is not reported.
The authors' own prediction for the resistance setting is worth recording verbatim in substance: they expect MMF and CX-5461 to be active against MEN1-mutant AML but possibly not against non-genetic resistant clones 1 - the same split this beat reached from the cooperativity argument, arrived at independently and pointing the same way.
A framing disagreement worth flagging
The paper states that 'LEDGF plays a critical role in tethering the menin/MLL complex to chromatin' and cites the 2008 adaptor paper. That is the framing mechanism-ledgf-dispensable-for-fusion-chromatin-retention shows to be wrong for the fusion specifically, on direct occupancy measurement. It is not wrong for wild-type MLL, and this paper's own chromatin readout is menin occupancy, not fusion occupancy, so nothing in its data contradicts the correction. The framing sentence and the measured data are simply about different things - which is how the tethering model has stayed in circulation.
| ledgf half life h | 4 |
|---|---|
| proteome average half life h | 20 |
| mechanism | synthesis suppression, not increased degradation (cycloheximide chase unchanged) |
| combination | synergy with VTP-50469 in MOLM-13 and MV4-11 |
| bypass construct | PCE = LEDGF PWWP + MLL CXXC + ENL; confers resistance to MMF, CX-5461 and VTP-50469 |
| toxicity | MMF causes neutropenia; CX-5461 reported tolerated without neutropenia |
| authors resistance prediction | active against MEN1-mutant AML, possibly not against non-genetic resistant clones |
agents
[
{
"name": "mycophenolate mofetil",
"target": "IMPDH",
"effect": "lowers LEDGF and menin protein"
},
{
"name": "CX-5461",
"target": "RNA polymerase I",
"effect": "lowers LEDGF and menin protein"
}
]- MMF causes neutropenia, which is the worst available toxicity in a leukaemia patient, and the authors say so themselves.
- The PCE experiment is a blueprint for resistance: anything supplying the H3K36 reader in cis, or raising HOXA9-MEIS1 output by another route, escapes LEDGF-directed therapy entirely.
- Lowering LEDGF by suppressing global translation removes every short-lived protein in every tissue, so the selectivity is a kinetic accident rather than a designed property.
- Because the intervention lowers menin as well as LEDGF, a clinical benefit could not be attributed to the LEDGF axis and would not validate it as a target.
- Does shRNA-resistant PSIP1 overexpression rescue the MMF phenotype? That is the LEDGF-only rescue the PCE experiment does not provide, and it would establish whether the benefit runs through LEDGF specifically.
- What are the protein half-lives of the other members of this module - KMT2A, AFF1, the SEC subunits? A half-life census would predict which nodes any translation-directed agent hits, and this graph holds none of those numbers.
- Does the same mechanism operate in lymphoid KMT2A::AFF1 disease? Every model here is myeloid, and MMF is already a licensed immunosuppressant with known paediatric pharmacology, so the translational path is unusually short if it does.
- Is the short half-life a property of p75, p52 or both? The antibody used detects LEDGF and the isoform attribution is not stated.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In MLL-AF9 leukaemia stem cells, mycophenolate mofetil (IMPDH inhibitor) and CX-5461 (RNA polymerase I inhibitor) dose-dependently reduced LEDGF and menin protein without changing their half-lives by cycloheximide chase; LEDGF turnover is about 4 hours against a proteome average of about 20; menin immunoprecipitation recovered almost no LEDGF after either treatment; menin occupancy at the Hoxa9 promoter fell, as it did with VTP-50469; a PCE chimera of the LEDGF PWWP domain, MLL CXXC domain and ENL rendered MOLM-13 resistant to MMF, CX-5461 and VTP-50469; forced Hoxa9-Meis1 expression rendered THP-1 resistant to MMF-induced differentiation; and VTP-50469 combined with MMF or CX-5461 was synergistic in MOLM-13 and MV4-11.
Guanine nucleotide biosynthesis blockade impairs MLL complex formation and sensitizes leukemias to menin inhibition. · Shi X, Li M, et al. · Nat Commun 2025What this evidence cannot show
Murine MLL-AF9 leukaemia stem cells and human myeloid MLLr lines; no lymphoid model. The LEDGF reduction is a consequence of global synthesis suppression, so nothing here is LEDGF-selective. The half-life comparison uses a literature proteome average rather than a matched internal measurement. Read in full from Europe PMC fullTextXML (91,444 characters, body present; PMC11920272).
Evidence that argues against this (1)
- MMF's dose-limiting toxicity in this setting is neutropenia, which the authors name as a concern specifically for haematologic malignancies, so the agent that demonstrates the mechanism may not be the agent that can be used. Shi X et al. 2025
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The paper's framing that LEDGF tethers the menin/MLL complex to chromatin derives from the 2008 adaptor study; its own chromatin measurement is menin occupancy, not fusion occupancy.
Menin critically links MLL proteins with LEDGF on cancer-associated target genes. · Yokoyama A, Cleary ML · Cancer Cell 2008What this evidence cannot show
Cited to identify the source of the framing. The distinction between menin occupancy and fusion occupancy is this entry's observation, not the paper's.
Evidence that argues against this (1)
- Direct measurement shows that LEDGF knockdown increases rather than decreases MLL-fusion occupancy, so the tethering framing does not hold for the fusion even though it holds for wild-type MLL. Zhu L et al. 2016
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The PCE bypass is the same architecture as the published minimal MLL fusion: an H3K36me2/3 reader module plus a non-methylated-CpG binding module plus an AF4-family coactivator interaction domain is necessary and sufficient for MLL-fusion target recognition and can functionally substitute for an MLL fusion protein.
MLL fusion proteins link transcriptional coactivators to previously active CpG-rich promoters. · Okuda H, Kawaguchi M, et al. · Nucleic Acids Res 2014What this evidence cannot show
Establishes sufficiency of the module combination. Its use here is to explain why the PCE chimera escapes LEDGF-directed intervention.
Evidence that argues against this (1)
- A construct that supplies both the reader and the SEC contact in cis bypasses several steps at once, so its resistance does not isolate which lost step the drugs act through. Shi X et al. 2025
an indirect route to lowering this protein, exploiting its 4-hour half-life
menin protein falls alongside LEDGF under the same agents
the closest existing approach to removing LEDGF from established disease, and why it does not close the hole
independently predicts activity against MEN1-mutant but not non-genetic resistance
the PCE chimera bypass is the same module-sufficiency architecture