The bite-back surface: LEDGF tethers CtIP to active chromatin and is required for DNA-end resection and homologous recombination
also called LEDGF CtIP · LEDGF homologous recombination
Away from leukaemia, LEDGF helps cells repair broken DNA by bringing a cutting enzyme to the break. A drug that removes LEDGF would remove that too, in every cell in the body, in children who are also being given DNA-damaging chemotherapy.
The same thing, for a scientist
LEDGF/p75 promotes repair of DNA double-strand breaks by homologous recombination. Depletion impairs recruitment of **CtIP** to DSBs and the CtIP-dependent end resection that follows; LEDGF is constitutively chromatin-associated through its PWWP domain, binds CtIP in a **DNA-damage-dependent** manner, and thereby tethers CtIP to active chromatin and facilitates its access to breaks. A CRISPR knockout series later reproduced and extended this: LEDGF-deficient cells proliferate and migrate less, are **more sensitive to etoposide**, show reduced recruitment of RPA32 after etoposide, and - untreated - already carry increased DNA fragmentation with elevated gamma-H2AX and BRCA1; re-expression of **LEDGF/p75** rescued all knockout effects. The mark is the same one the leukaemic PWWP arm reads: H3K36me3 and PSIP1 associate with several DNA repair proteins at actively transcribing loci, which is why the repair function and the fusion-tethering function cannot be separated by targeting the PWWP cage. This is the single largest essential-function liability on the LEDGF axis, and it lands on a paediatric population that receives anthracyclines, etoposide and alkylators.
The measured chain
LEDGF sits on H3K36me2/3 chromatin constitutively -> DNA damage induces LEDGF-CtIP binding -> CtIP is tethered near the break within a transcribed region -> CtIP-dependent end resection proceeds -> RPA32 loads -> homologous recombination 12.
The knockout series adds two things the original paper did not have. First, an unstressed phenotype: LEDGF-null cells already show DNA fragmentation, gamma-H2AX and BRCA1 foci without any damaging agent, so the deficit is not only a response failure 2. Second, an isoform attribution: re-expressing LEDGF/p75 rescued, which places the function on the p75 isoform - the one that carries the IBD and the one relevant to KMT2A.
Why this is the dominant liability rather than one of several
The usual essential-function argument against an epigenetic-reader target is developmental or transcriptional. Here it is acute and it interacts directly with the standard of care.
- Infant KMT2A-rearranged ALL protocols use etoposide among other genotoxics, and etoposide sensitivity is the exact phenotype of LEDGF loss 2.
- LEDGF loss shifts repair pathway choice: 53BP1 foci increase and RAD51 foci decrease on PSIP1 depletion, and PSIP1-depleted cancer cells become more sensitive to PARP inhibitors and to DNA-damaging agents 3.
- The paralogue does not cover it: HRP-2/HDGFL2 also promotes homologous recombination 4, so an agent that hits both PWWP domains - which the only existing LEDGF chemistry does - removes two HR contributors at once.
The mirror of the liability is an opportunity, and it should be stated with the same force: if LEDGF loss sensitises to PARP inhibition and to etoposide, then a LEDGF-directed agent is a chemosensitiser and its natural development path is combination rather than monotherapy. That reframing also lowers the required potency, which matters given the micromolar ceiling on the PWWP chemistry.
What is not established
Nobody has measured any of this in a haematopoietic cell, let alone in a KMT2A-rearranged one. The CtIP work is in human cancer lines; the knockout series is in a single engineered line. The conditional Psip1 knockout mouse was not challenged with a genotoxic agent, so the in vivo question - does a LEDGF-deficient haematopoietic system tolerate chemotherapy - is open and is directly answerable in an existing mouse.
| pathway | homologous recombination / DNA end resection |
|---|---|
| key partner | CtIP (RBBP8) |
| isoform attributed | p75 (rescue by p75 re-expression) |
phenotypes of loss
[ "impaired CtIP recruitment", "impaired end resection", "reduced RPA32 loading", "etoposide hypersensitivity", "increased 53BP1, decreased RAD51 foci", "PARP-inhibitor sensitisation", "baseline DNA fragmentation" ]
untested in
[ "haematopoietic cells", "KMT2A-rearranged cells", "the conditional Psip1 knockout mouse under genotoxic challenge" ]
- A systemic LEDGF-directed agent removes a homologous-recombination contributor in every tissue, in children who are simultaneously receiving etoposide and anthracyclines.
- LEDGF loss already produces baseline DNA damage without any genotoxic insult, so the liability is not conditional on co-administration.
- Long-term HR impairment in a cured paediatric patient is a second-malignancy risk that would not show up in any preclinical model this field uses.
- The only existing LEDGF chemistry also binds HRP-2, which is a second HR contributor, so the current chemical series maximises rather than minimises this liability.
- Does a Psip1-deficient haematopoietic system tolerate etoposide or an anthracycline? The conditional knockout mouse exists and has never been challenged; this is a short experiment with a direct bearing on whether a LEDGF agent could be combined with standard infant-ALL therapy.
- Are the CtIP-tethering model and the R-loop model of LEDGF's repair role the same mechanism seen twice, or two mechanisms? They make different predictions about which cells are most vulnerable.
- Is the PARP-inhibitor sensitisation large enough to be a development strategy? If so, the LEDGF axis is a chemosensitiser rather than a targeted agent, and the potency bar falls accordingly.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
LEDGF promotes repair of DNA double-strand breaks by homologous recombination: its depletion impairs recruitment of CtIP to breaks and the subsequent CtIP-dependent end resection, and LEDGF binds CtIP in a DNA-damage-dependent manner while remaining constitutively chromatin-associated through its PWWP domain.
LEDGF (p75) promotes DNA-end resection and homologous recombination. · Daugaard M, Baude A, et al. · Nat Struct Mol Biol 2012What this evidence cannot show
Abstract only: Europe PMC returns no PMCID for this record, so routes 2-4 of the unreadable test are unavailable and the cell systems and quantitation could not be checked. Human cancer cell lines, not haematopoietic cells.
Evidence that argues against this (1)
- A later CRISPR knockout study found that untreated LEDGF-null cells already carry increased DNA fragmentation with elevated gamma-H2AX and BRCA1, which is not what a pure resection-recruitment defect predicts and suggests the phenotype is broader than CtIP tethering. Liedtke V et al. 2021
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
CRISPR knockout of LEDGF reduced proliferation and migration, increased sensitivity to etoposide and reduced RPA32 recruitment after etoposide; untreated knockout cells showed increased DNA fragmentation with increased gamma-H2AX and BRCA1; UBC13 and PA28-gamma protein levels fell; and re-expression of LEDGF/p75 rescued all knockout effects.
LEDGF/p75 Is Required for an Efficient DNA Damage Response. · Liedtke V, Schröder C, et al. · Int J Mol Sci 2021What this evidence cannot show
A single engineered cell background with p75 and p52 re-expression lines; open access in Europe PMC (PMC8198318). The ubiquitin-pathway observations are correlative.
Evidence that argues against this (1)
- The rescue was by re-expression of LEDGF/p75 in the same system, so the result establishes sufficiency in that background rather than generality; the paper does not test a haematopoietic or KMT2A-rearranged context. Liedtke V et al. 2021
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
PSIP1 depletion increases 53BP1 foci and reduces RAD51 foci, indicating altered DNA repair pathway choice, and increases the sensitivity of cancer cells to PARP1 inhibitors and to DNA-damaging agents.
PSIP1/LEDGF reduces R-loops at transcription sites to maintain genome integrity. · Jayakumar S, Patel M, et al. · Nat Commun 2024What this evidence cannot show
The mechanism in that paper runs through R-loop accumulation rather than through CtIP tethering, so the two models of LEDGF's repair role are not yet unified. Open access (PMC10774266).
Evidence that argues against this (1)
- If the repair phenotype is secondary to R-loop-driven transcription-replication conflict rather than to a direct repair function, then the liability depends on transcriptional load and might be smaller in a quiescent cell than in a proliferating blast - or larger. Jayakumar S et al. 2024
- E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
H3K36me3 and PSIP1/LEDGF associate with several DNA repair proteins, suggesting a role in efficient DNA repair at actively transcribing loci - the same mark the leukaemic PWWP arm reads.
H3K36me3 and PSIP1/LEDGF associate with several DNA repair proteins, suggesting their role in efficient DNA repair at actively transcribing loci. · Sundarraj J, Taylor GCA, et al. · Wellcome Open Res 2017What this evidence cannot show
A proteomic association study published in Wellcome Open Research; associations rather than functional requirements. Open access in Europe PMC.
Evidence that argues against this (1)
- Co-association at an H3K36me3-marked locus does not establish that LEDGF recruits those repair proteins; both could be independently recruited by the mark. Li L et al. 2017
- E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The paralogue does not provide a safety margin here: HRP-2/HDGFL2 also promotes DNA repair by homologous recombination.
Hepatoma-derived growth factor-related protein 2 promotes DNA repair by homologous recombination. · Baude A, Aaes TL, et al. · Nucleic Acids Res 2016What this evidence cannot show
Cited for the HRP-2 repair function. The inference that a dual PWWP ligand compounds the liability is this entry's.
Evidence that argues against this (1)
- HRP-2 single knockout mice develop normally and are fertile, so HRP-2's repair contribution is evidently dispensable at the organismal level under unstressed conditions. Wang H et al. 2015
the non-leukaemic function that defines this target's toxicity surface
the essential-function liability the dossier's therapeutic-index argument omits
a dual LEDGF/HRP-2 PWWP ligand compounds this liability
two unreconciled accounts of the same genome-stability role
the homologous-recombination route to the same combination prediction
the menin/SETD2/LEDGF/CtIP arm licenses the step immediately upstream of EXO1/DNA2
the short-range half of the proposed single reaction
the effector that mechanism entry describes; the graph held the mechanism and not the protein
tests whether menin inhibition breaks the LEDGF-CtIP arm