LEDGF/p75 is required in the leukaemic cell and dispensable in ADULT MOUSE HAEMATOPOIESIS - not in the normal organism, which its germline knockout does not survive
also called LEDGF normal-versus-leukaemic dissociation · PSIP1 cell context
A protein called LEDGF is needed for MLL-rearranged leukaemia to form, and adult mouse blood cells without the fusion get along fine without it. But mice born without the gene at all mostly die, so the gap is between the leukaemia and one tissue, not between the leukaemia and the whole healthy animal.
The same thing, for a scientist
Conditional loss of Psip1/LEDGF in mice leaves haematopoiesis essentially intact while abolishing MLL-rearranged leukemogenesis [E1]. The DepMap measurement made in this beat is the same shape from the other direction: PSIP1 is depleted below -0.5 in only 1.1% of 1,258 screened cancer cell lines (mean gene effect +0.029) yet has a mean of -0.553 across eleven KMT2A-rearranged lines, a KMT2A-r-minus-all-lines difference of -0.582 that is the fourth largest in the panel [E2]. Both arms place PSIP1 in evidence class 2 (required) and neither places it in class 3: PSIP1 has never been shown to maintain a defined cell state. **SCOPE CORRECTED 2026-09-21 (Q68).** 'Not required in the normal cell' is true of the adult mouse haematopoietic compartment in a conditional model and false of the gene. Sutherland et al. (PMID 16980622, Mol Cell Biol 2006) disrupted Psip1/Ledgf by gene trap: the MAJORITY of homozygous mice died PERINATALLY, and survivors showed low fertility, absent epididymal fat pads, motor and behavioural defects, brachycephaly, small rib cages and **homeotic skeletal transformations** with incomplete penetrance - which the authors read as a role for Psip1 in the control of **HOX expression**, and link explicitly to PSIP1/LEDGF appearing as a NUP98 fusion partner in myeloid leukaemia. Hox control is the same function the leukaemia argument invokes, which makes the dissociation narrower than 'cleanest in the core set' implied: what is dissociated is the *tissue and developmental window*, not the *function*. The therapeutic-window argument survives - an adult patient is not an embryo - but it has to be made about adult haematopoiesis specifically, and this entry now says so.
Why this is the model case for the beat's distinction
The question this beat asks of every gene is which cell type does it matter in. For PSIP1 the honest answer is not a cell type at all — it is a cell condition: the presence of a KMT2A fusion. The normal pro-B cell and the normal HSC do not need it; the transformed cell does. 12
That is why writing 'LEDGF matters in blood' would be wrong while 'LEDGF is required in the leukaemic cell and not the normal one' is right, and it is the reason this graph keeps the cell_type.context field: whether a result is about the fusion context or about the wild-type protein in normal cells is not a caveat here, it is the entire content.
What is missing, stated as a bounded absence
Bounded query, NCBI esearch, 2026-09-20: (PSIP1 OR LEDGF OR "LEDGF/p75" OR "p75" OR "DFS70")[tiab] AND <18-term cell-state clause> AND <leukaemia clause> returns 5, of which 0 survive a title check — two are 1990s p75 tyrosine-phosphorylation papers matching on the alias p75, one is the unrelated NUP98::PSIP1 fusion in MDS, and two are reviews. Controls: the gene clause alone returns 7,404 and the state clause alone 96,743, so the instrument can return a one in both directions 3.
Zero in titles and abstracts; supporting figures not searched. The alias p75 inflates the bare count enormously and should not be used without title-checking every hit.
Correction, 2026-09-21 (Q68): 'the normal one' means adult mouse haematopoiesis
What was withdrawn. The unqualified 'not in the normal one' in the title, and the implication that PSIP1 is a dispensable gene.
Why. PMID 16980622: germline Psip1 disruption is mostly perinatally lethal, with homeotic skeletal transformations that the authors attribute to loss of Hox control.
Why this matters beyond the wording. The entry's bounded absence 3 records that no cell-state result exists for PSIP1. A homeotic transformation IS a positional-identity phenotype, i.e. a state claim in the developmental sense - it is simply written in a vocabulary that no leukaemia-scoped, eighteen-cell-state-term query can reach, in a 2006 paper with neither a leukaemia term nor a cell-state term in its title. The absence is real for the haematopoietic cell-state literature and is not an absence of state biology for this gene.
| gene | gene-psip1 |
|---|---|
| uniprot | O75475 |
| evidence class | 2 |
cell type
{
"required_in": "any cell carrying a KMT2A fusion; not lineage-restricted",
"dispensable_in": "normal murine haematopoiesis"
}cell state
{
"maintains": null,
"bounded_absence": true
}depmap — the numbers behind the chart
{
"pct_all_lines_depleted": 1.1,
"mean_all_lines": 0.029,
"mean_kmt2ar": -0.553,
"contrast_A": -0.582
}- A target that is dispensable in normal cells is also a target whose loss the leukaemia may tolerate by switching to HRP-2/HDGFL2, which is already reported to matter for leukaemic cell survival.
- The murine dispensability does not license assuming a wide human therapeutic window: mouse and human haematopoiesis differ in exactly the stem-cell compartment where this would matter.
- Does the normal-cell dispensability hold in HUMAN haematopoiesis, or only in mouse? No human conditional system has been reported.
- Is PSIP1 required to ESTABLISH the leukaemia, to MAINTAIN it, or both? The mouse work is initiation-weighted.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In mice, LEDGF/p75 is dispensable for haematopoiesis but essential for MLL-rearranged leukemogenesis.
LEDGF/p75 is dispensable for hematopoiesis but essential for MLL-rearranged leukemogenesis. · El Ashkar S, Schwaller J, et al. · Blood 2018What this evidence cannot show
Murine system throughout (brief rule 73): mouse haematopoiesis and murine MLL-rearranged leukaemia, not human infant B-lineage KMT2A::AFF1. Abstract read in this beat; full text not opened.
Evidence that argues against this (1)
- HRP-2 (HDGFL2), LEDGF's paralogue, is dispensable for MLL-r leukemogenesis but important for leukaemic cell survival (PMID 33477970), so the two paralogues split the phenotype and a LEDGF-only account is incomplete. Cells 2021
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
In DepMap fetched 2026-09-20, PSIP1 is depleted below -0.5 in 1.1% of 1,258 screened lines with a mean gene effect of +0.029, while its mean across eleven KMT2A-rearranged lines is -0.553.
Open Targets Platform - DepMap CRISPR gene essentiality (depMapEssentiality), queried by Ensembl gene id 2026-09-20 · Open Targets Platform GraphQL API 2026What this evidence cannot show
Cancer cell lines in culture. Eleven KMT2A-rearranged lines, seven myeloid and four B-lineage. The myeloid half is -0.637 and the B-lineage half -0.406, so the dependency is present in both but not equal; with the fusion held constant at KMT2A::AFF1 the gap narrows to -0.201 on n = 1 against n = 3.
Evidence that argues against this (1)
- PUBLISHED CONFLICTING RESULT against 'not required in the normal cell', which is the half of this dissociation the DepMap number is cited to corroborate. Sutherland et al. disrupted Psip1/Ledgf by gene trap and the MAJORITY of homozygous mice DIED PERINATALLY; survivors had low fertility, absent epididymal fat pads, motor and behavioural defects, brachycephaly, small rib cages and homeotic skeletal transformations with incomplete penetrance. A gene whose germline loss is mostly lethal is not dispensable; what is dispensable is Psip1 in the conditional haematopoietic compartment of an adult mouse, and a 1.1% depletion rate across cancer lines in culture is a third, separate quantity. **And the graph already holds a finer version of this, in another directory:** `mechanism-psip1-in-normal-haematopoiesis-dispensable-is-not-silent` records that even within haematopoiesis 'dispensable' summarises halved white counts, perturbed HSC and progenitor numbers and failing serial replating. This entry's dissociation was therefore already qualified twice over elsewhere and had not been qualified here. The entry's title - 'the cleanest cell-context dissociation in the core set' - is defensible only if 'normal cell' is read as 'adult mouse haematopoiesis'. Sutherland HG et al. 2006
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database
A bounded PubMed query pairing PSIP1 and its aliases with an eighteen-term cell-state clause and a leukaemia clause returns five records, none of which survives a title check.
NCBI PubMed E-utilities esearch, db=pubmed, queried 2026-09-20 · NCBI Entrez Programming Utilities 2026What this evidence cannot show
Bounded by titles and abstracts only. Supporting figures were not searched and no full text was opened for PSIP1 in this beat, so this absence is weaker than the ones in this beat where a full text was opened.
Evidence that argues against this (1)
- PUBLISHED CONFLICTING RESULT about what a leukaemia-scoped, cell-state-term-scoped query can reach for this gene. A developmental-state phenotype for PSIP1 is published and is invisible to that query shape twice over: Sutherland et al. report HOMEOTIC SKELETAL TRANSFORMATIONS in Psip1-disrupted mice, read by the authors as a role in the control of HOX EXPRESSION, in a paper with no leukaemia term and no cell-state term in the title. Homeotic transformation is a positional-identity phenotype, which is a state claim in the developmental sense even though no word in the eighteen-term clause names it. The caveat correctly rates this absence as weaker than the beat's full-text-opened ones; the reason is the vocabulary, not the lack of an opened full text. Sutherland HG et al. 2006
the cell-context dossier for this gene
- related to →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
the requirement is conditional on the presence of a KMT2A fusion rather than on a cell type
the worked example of evidence class 2 with the normal-versus-fusion context separated
SCOPE CORRECTED: germline Psip1 disruption is mostly PERINATALLY LETHAL with homeotic skeletal transformations the authors attribute to loss of HOX control (PMID 16980622) - 'not required in the normal cell' is true of adult mouse haematopoiesis in a conditional model and false of the gene