LEDGF PWWP chemistry exists after all - 68 fragment hits, double-digit micromolar leads - and the pocket has a structural ceiling the NSD PWWPs do not
also called LEDGF PWWP inhibitors · HRP family PWWP fragment screen
A 2024 fragment screen did find small molecules that sit in LEDGF's histone-reading pocket. They bind weakly, and the pocket is smaller than the related pockets where much stronger drugs were made, so there is a ceiling on how good they can get without a new idea.
The same thing, for a scientist
This graph's LEDGF entries state that there is essentially no chemical matter against PSIP1. For the IBD arm that remains true. For the **PWWP arm it is now false**: an X-ray crystallography-based fragment screen against the HRP family PWWP domain produced **68 diverse fragment hits** sampling the H3K36me2/3 pocket pharmacophore, and structure-guided expansion in three directions gave multiple compound series with **double-digit micromolar** affinity, verified by X-ray, NMR and DSF, with ligand efficiency above 0.5 kcal/mol per heavy atom. The screen used **HRP-2 PWWP crystals as a surrogate** because LEDGF PWWP crystals were unsuitable, and the resulting compounds bind both LEDGF and HRP-2 PWWP while showing no affinity for NSD2 PWWP1 or BRPF2 PWWP. The paper also states the structural ceiling explicitly: the LEDGF PWWP pocket is **considerably smaller** than the NSD2/NSD3 PWWP1 domains, which carry an additional subpocket that nanomolar compounds exploit, and its absence limits attainable affinity. The authors' own recommendation is that future compounds must also engage the DNA-binding residues, not only the aromatic cage - which is the same conclusion this graph reached from the avidity argument.
The numbers
| Item | Value |
|---|---|
| Fragment hits from crystallographic screening | 68 |
| Crystal system used | HRP-2 PWWP (LEDGF PWWP crystals unsuitable) |
| Best affinity after structure-guided expansion | double-digit micromolar |
| Ligand efficiency | > 0.5 kcal/mol per non-hydrogen atom |
| Orthogonal validation | X-ray, NMR, differential scanning fluorimetry |
| Selectivity | binds LEDGF and HRP-2 PWWP; no affinity for NSD2 PWWP1 or BRPF2 PWWP |
| Notable interaction | halogen sigma-hole contact with a conserved buried water |
A parallel fragment campaign against the HDGFRP2 (HDGFL2/HRP-2) PWWP domain was published the same year 2, so two independent groups were working the same subfamily pocket in 2024.
The ceiling, stated by the people who found it
The single most useful sentence in the paper for this graph is the structural comparison: the binding pocket in LEDGF PWWP is considerably smaller than the related PWWP1 domains of NSD2 and NSD3, which feature an additional subpocket and for which nanomolar compounds have been developed; the absence of that subpocket limits the attainable affinity 1.
This is the kind of statement that should change a portfolio decision. It says the gap between LEDGF PWWP chemistry and NSD2 PWWP chemistry is not a resourcing gap. It is a pocket-volume gap, and it does not close with more chemistry of the same kind.
The suggested escape is to extend ligands onto the DNA-binding surface, because nucleosome affinity is driven by the cage and by DNA contacts. That converts the problem from methyl-lysine-cage occupancy into interface disruption on a much larger surface, which is harder, and it is the same conclusion the existing interface-psip1-pwwp-h3k36me-nucleosome entry reached from the opposite direction - nucleosome binding being orders of magnitude tighter than peptide binding.
Selectivity: the good news and the bad news
Good: the compounds do not touch NSD2 PWWP1 or BRPF2 PWWP, so the subfamily structural differences translate into real selectivity and a LEDGF PWWP agent would not be a pan-PWWP reader inhibitor.
Bad: they bind HRP-2 as well as LEDGF, which is unavoidable given the screen was run on HRP-2 crystals. Whether that is a liability or a feature depends on an unanswered question. HRP-2 is dispensable for MLL-ENL transformation but important for leukaemic cell survival independently of MLL rearrangement 3, and HRP-2's IBD substitutes for LEDGF's in HIV integration. A dual LEDGF/HRP-2 PWWP ligand therefore removes the paralogue buffering that would defeat a LEDGF-selective agent - and simultaneously removes a protein with its own homologous-recombination function 4.
Where this leaves the two arms
| IBD arm | PWWP arm | |
|---|---|---|
| Genetic validation in KMT2A leukaemia | strong (IBM2 mutants abolish transformation) | none directly; the PWWP requirement for the fusion has never been separated |
| Structures with a leukaemia-relevant ligand | 2MTN, 2MSR, 6EMQ, 3U88 | six nucleosome cryo-EM structures, no leukaemic ligand |
| Small-molecule chemistry | none (ebselen aside) | 68 fragments, micromolar leads |
| Structural ceiling | shallow extended peptide surface | small cage, no subpocket, stated affinity limit |
| Selectivity liability | integrase, POGZ, JPO2, the whole IBM interactome | HRP-2 |
The arms fail in opposite directions: the validated arm has no chemistry, and the arm with chemistry has no validation. That is the honest shape of this target and it has not been stated this way in the graph before.
| target protein | gene-psip1 |
|---|---|
| domain | PWWP (H3K36me2/3 reader) |
| fragment hits | 68 |
| best affinity | double-digit micromolar |
| surrogate crystal system | HRP-2 PWWP |
| structural ceiling | LEDGF PWWP lacks the additional subpocket present in NSD2/NSD3 PWWP1, which the authors state limits attainable affinity |
| authors recommendation | future compounds should also target the DNA-binding region, not only the aromatic cage |
selectivity binds
[ "LEDGF PWWP", "HRP-2 PWWP" ]
selectivity spares
[ "NSD2 PWWP1", "BRPF2 PWWP" ]
- A stated structural ceiling on attainable affinity means more medicinal chemistry does not fix this; the honest options are a different surface, a different modality, or a different target.
- The compounds bind HRP-2 as well, and HRP-2 depletion impairs leukaemic colony formation regardless of MLL rearrangement, so the selectivity that justifies this target is lost at the chemical-matter stage.
- Both LEDGF and HRP-2 contribute to homologous recombination, so a dual PWWP agent carries a combined genome-instability liability in a paediatric population already receiving genotoxic chemotherapy.
- The screen was run on a surrogate crystal system; a compound optimised in HRP-2 crystals may be optimised for the wrong pocket geometry.
- Do any of these compounds displace LEDGF from chromatin in a cell, as opposed to binding the isolated domain? No PWWP protein has ever been shown to be displaced from chromatin by a cage ligand, so this is the field-level question and not only a LEDGF question.
- Is the PWWP arm required for the FUSION at all? The IBD arm has IBM2 mutants; the PWWP arm has no equivalent separation-of-function experiment in a KMT2A-rearranged system, which means the arm with chemistry is the arm without genetics.
- Is dual LEDGF/HRP-2 PWWP engagement a feature or a liability? HRP-2 depletion reduces colony formation regardless of MLL status, which suggests liability, but it also removes paralogue buffering.
- Would a LEDGF degrader sidestep the pocket-volume ceiling entirely? The avidity and subpocket arguments that cap occupancy inhibition do not apply to degradation.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
X-ray crystallography-based fragment screening against the HRP family PWWP domain, using HRP-2 PWWP crystals because LEDGF PWWP crystals were unsuitable, produced 68 fragment hits and, after structure-guided expansion, multiple compound series with double-digit micromolar affinity and ligand efficiency above 0.5 kcal/mol per heavy atom; the compounds bind both LEDGF and HRP-2 PWWP but not NSD2 PWWP1 or BRPF2 PWWP; and the LEDGF PWWP pocket is considerably smaller than NSD2/NSD3 PWWP1, lacking an additional subpocket, which the authors state limits the attainable affinity.
Rational fragment-based design of compounds targeting the PWWP domain of the HRP family. · Vantieghem T, Aslam NA, et al. · Eur J Med Chem 2024What this evidence cannot show
Abstract only: Europe PMC returns no PMCID for this record (inEPMC N, isOpenAccess N), so the four-route unreadable test stops at route 1 and the compound structures, exact Kd values and DSF conditions could not be checked. All affinities are biophysical against isolated domains; no cellular or nucleosome-displacement data are reported in the abstract.
Evidence that argues against this (1)
- Double-digit micromolar against an isolated domain is very far from displacing a PWWP protein from chromatin, where nucleosome binding is driven substantially by DNA contacts and is orders of magnitude tighter than peptide binding; no published example exists of a methyl-lysine cage ligand displacing any PWWP protein from chromatin. Koutná E et al. 2023
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
An independent 2024 fragment-based campaign discovered small molecule inhibitors of the HDGFRP2 PWWP domain, so two groups were working this PWWP subfamily pocket in the same year.
Fragment-based discovery of small molecule inhibitors of the HDGFRP2 PWWP domain. · Wei X, Li S, et al. · FEBS Lett 2024What this evidence cannot show
Cited from the record title and abstract only, as a statement that a second campaign exists. HDGFRP2 is HRP-2/HDGFL2, the paralogue, not LEDGF.
Evidence that argues against this (1)
- Chemistry against the paralogue is chemistry against the paralogue; cross-reactivity with LEDGF PWWP has to be demonstrated rather than assumed from subfamily membership. Vantieghem T et al. 2024
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
HRP-2 is dispensable for MLL-ENL transformation in HRP-2 knockout primary murine cells but its depletion reduces colony formation in leukaemic cell lines independently of MLL rearrangement, and HRP-2 interacts with MLL through a conserved interface that is less menin-dependent than the MLL-LEDGF interaction.
Unlike Its Paralog LEDGF/p75, HRP-2 Is Dispensable for MLL-R Leukemogenesis but Important for Leukemic Cell Survival. · Van Belle S, El Ashkar S, et al. · Cells 2021What this evidence cannot show
Systemic knockout mice showed only an increase in peripheral blood neutrophils, so the normal-tissue picture is thin. Open access in Europe PMC (PMC7835958).
Evidence that argues against this (1)
- The finding that HRP-2 depletion reduces colony formation regardless of MLL status means an agent hitting HRP-2 is not KMT2A-selective, which removes the therapeutic-index argument that makes the LEDGF axis attractive in the first place. Van Belle S et al. 2021
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
HRP-2/HDGFL2 promotes DNA repair by homologous recombination, so a dual LEDGF/HRP-2 agent would remove two homologous-recombination contributors at once.
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 combined-liability inference is this entry's, not the paper's.
Evidence that argues against this (1)
- Removing two homologous-recombination contributors could equally be an opportunity - a synthetic-lethal combination with a PARP inhibitor - rather than only a liability. Jayakumar S et al. 2024
supplies the chemical matter that entry recorded as absent, and its ceiling
68 fragment hits and micromolar leads against this protein's PWWP domain
corrects the dossier's zero-chemistry framing for the PWWP arm
a dual LEDGF/HRP-2 PWWP ligand compounds this liability
the existing PWWP chemistry binds both paralogues
the arm with chemistry is the arm without genetic validation
the PWWP arm is dispensable for gross chromatin retention, so PWWP engagement need not evict LEDGF