The third job: LEDGF sits on transcribed gene bodies with splicing factors and changes exon choice in half the genes that have alternative isoforms
also called LEDGF splicing · PSIP1 SRSF1 adaptor · LEDGF exon choice
Besides holding onto chromatin and helping switch genes on, LEDGF helps decide which pieces of each gene end up in the final message. Removing it changes the messages cells make, in about half the genes that have a choice.
The same thing, for a scientist
LEDGF couples the H3K36-methylated transcribed gene body to the splicing machinery, and it does so through both isoforms by different routes. **p52** binds methylated H3K36 and splicing factors, including ASF/SF2 (SRSF1), and contributes to the regulation of alternative splicing - the founding observation, and the reason PSIP1's original name is 'PC4 and SF2-interacting protein 1'. **p75** independently associates with splicing factors and, in cells lacking LEDGF/p75 or lacking the p75 IBD specifically, contributes to splicing patterns in **half of the transcription units that have alternative isoforms**; the functional consequence in that study was that HIV-1 integration is directed to highly spliced genes, with intron density the strongest single predictor of integration density. A downstream example in stem cells: the PBX1a-to-PBX1b isoform switch that regulates pluripotency is linked to H3K36me3 through a PSIP1/SRSF1 adaptor. For this beat the splicing arm is a third bite-back surface, and a specific one: PBX1 is a HOX cofactor and PBX3 is a KMT2A-fusion target gene, so LEDGF-dependent splicing lands on the same transcription factor family the leukaemia runs on.
What each isoform contributes
p52 - binds H3K36me via PWWP, binds splicing factors including ASF/SF2, regulates alternative splicing 1. This is the classical function and the origin of the gene's name.
p75 - affinity purification shows association with a number of splicing factors; RNA-seq of cells lacking p75, and separately of cells lacking only the p75 IBD, shows changed splicing patterns in half of the transcription units with alternative isoforms 2. The IBD-deletion line is what makes this attributable to p75 rather than to total PSIP1, and it implies the splicing contribution runs at least partly through IBD-mediated partner recruitment - which is the module a KMT2A-directed IBD ligand would occupy.
That last point is the one that matters therapeutically and it is easy to miss: an IBD-directed agent would not be free of the splicing liability, because the splicing effect was demonstrated in an IBD-deletion background.
The PBX link
An analysis of human ESC differentiation concluded that the isoform switch from PBX1a to PBX1b links H3K36me3 to stem cell fate determination through the PSIP1/SRSF1 adaptor, producing exon skipping in PBX1 3.
PBX proteins are HOX cofactors; PBX3 is among the canonical KMT2A-fusion target genes and appears in the CUT&RUN panel of the menin-LSD1 preprint. Nobody has asked whether LEDGF-dependent splicing alters PBX isoform ratios in KMT2A-rearranged blasts, and it is answerable from existing RNA-seq: any dataset with PSIP1 knockdown in a leukaemic line would show it. The prediction, if the ESC mechanism generalises, is that LEDGF loss shifts PBX1 exon usage in a direction that changes HOX cofactor availability - a transcriptional consequence of a LEDGF drug that operates entirely outside the chromatin-tethering model.
Honest limits
The p75 splicing analysis is secondary to an HIV integration study in HEK293T cells, and the PBX1 link comes from a computational analysis of public ESC and differentiated-cell data with limited experimental validation. Neither is haematopoietic. The founding p52 work is a decade older and in different cells again. The three do not constitute a coherent, quantified account of LEDGF-dependent splicing in any disease-relevant cell, and the entry is graded accordingly.
| isoform resolved | true |
|---|---|
| ibd involved | yes - the p75 splicing effect was shown in an IBD-deletion background |
| leukaemia relevance | untested; PBX family are HOX cofactors and PBX3 is a KMT2A-fusion target |
functions
[ "p52: binds H3K36me and ASF/SF2, regulates alternative splicing", "p75: associates with splicing factors, changes exon choice in ~50% of alternatively spliced transcription units", "PSIP1/SRSF1 adaptor links H3K36me3 to PBX1 exon skipping" ]
systems
[ "HEK293T", "human ESC", "non-haematopoietic" ]
- An IBD-directed agent, usually assumed to be the clean modality, would carry the splicing liability, because the p75 splicing effect was demonstrated in cells lacking the IBD specifically.
- Changing exon choice in half the alternatively spliced transcriptome is a large perturbation with no specific readout in any leukaemia study of this axis.
- PBX proteins are HOX cofactors, so a LEDGF agent could alter HOX cofactor isoform ratios in a direction nobody has predicted or measured.
- Does PSIP1 depletion change PBX1 or PBX3 isoform usage in a KMT2A-rearranged line? Answerable from any existing PSIP1-knockdown RNA-seq dataset in a leukaemic cell, if one exists.
- Is the splicing contribution of p75 IBD-dependent, as the IBD-deletion experiment implies? If so, an IBD-directed drug carries the splicing liability that an IBD-directed drug is usually assumed to avoid.
- Does the splicing function operate in haematopoietic cells at all? Every existing measurement is in HEK293T or ESC.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Psip1/Ledgf p52 binds methylated histone H3K36 and splicing factors and contributes to the regulation of alternative splicing.
Psip1/Ledgf p52 binds methylated histone H3K36 and splicing factors and contributes to the regulation of alternative splicing. · Pradeepa MM, Sutherland HG, et al. · PLoS Genet 2012What this evidence cannot show
The isoform characterised is p52, which lacks the IBD and is not the KMT2A-relevant isoform. Non-haematopoietic cells.
Evidence that argues against this (1)
- Attributing the splicing function to p52 alone is incomplete: p75 independently associates with splicing factors and contributes to exon choice. Singh PK et al. 2015
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
LEDGF/p75 is associated with a number of splicing factors by affinity purification, and RNA-seq of HEK293T cells lacking LEDGF/p75 or lacking the LEDGF/p75 integrase-binding domain shows that LEDGF/p75 contributes to splicing patterns in half of the transcription units that have alternative isoforms; intron density was the strongest predictor of HIV-1 integration density in transcription units.
LEDGF/p75 interacts with mRNA splicing factors and targets HIV-1 integration to highly spliced genes. · Singh PK, Plumb MR, et al. · Genes Dev 2015What this evidence cannot show
HEK293T cells, in an HIV integration-site study; the splicing analysis is secondary to that aim. Open access in Europe PMC (PMC4647561).
Evidence that argues against this (1)
- A change in splicing patterns after removing a chromatin protein is compatible with an indirect effect through altered transcription elongation rate rather than with a direct splicing role. Pradeepa MM et al. 2012
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
In human embryonic stem cell differentiation, the isoform switch from PBX1a to PBX1b links H3K36me3 to stem cell fate determination through the PSIP1/SRSF1 adaptor, producing exon skipping in PBX1.
Alternative splicing links histone modifications to stem cell fate decision. · Xu Y, Zhao W, et al. · Genome Biol 2018What this evidence cannot show
A computational analysis of transcriptomes and epigenomes across hESC and five differentiated cell types, with limited experimental validation of the PSIP1/SRSF1 step specifically; not haematopoietic and not leukaemic.
Evidence that argues against this (1)
- Correlation between a histone mark and an exon inclusion rate across cell types is weak evidence for a specific adaptor mechanism, and the PSIP1/SRSF1 step is inherited from earlier work rather than tested here. Pradeepa MM et al. 2012
- E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The chromatin context in which the splicing coupling occurs is the H3K36-methylated transcribed gene body that the PWWP domain reads multivalently.
Structure of H3K36-methylated nucleosome-PWWP complex reveals multivalent cross-gyre binding. · Wang H, Farnung L, et al. · Nat Struct Mol Biol 2020What this evidence cannot show
Structural; cited for the chromatin context rather than for splicing.
Evidence that argues against this (1)
- H3K36 methylation couples to splicing through several adaptors besides PSIP1, so the mark's presence does not make PSIP1 the operative link at any given exon. Xu Y et al. 2018
third documented function of this protein
both isoforms touch splicing by different routes
the p75 splicing effect was shown in an IBD-deletion background