U-937 (CVCL_0007) - the single line carrying this graph's menin-independence argument, and it is adult, monocytic, historically contaminated, and exists as two variants nobody here distinguishes
also called U937 · U-937 · CVCL_0007 · CALM-AF10 comparator line
One cell line does almost all the work in this project's argument that a related leukaemia does not need the drug target. Looking that line up properly shows it comes from an adult with a different kind of leukaemia, some early stocks were contaminated with another line, and it exists in two versions that behave differently.
The same thing, for a scientist
**64 entries in this graph reference U-937 and none of them was a line entry until now**, although every other index line here has one. It is the `n=1` behind `ana-depmap-men1-dependency-calm-af10-u937`, the CALM-AF10 arm of the proposed menin-inhibitor panel, and by extension the main empirical support for `hyp-menin-axis-separable-from-hoxa-programme`. Verified live against Cellosaurus CVCL_0007 and PubMed: | property | value | |---|---| | disease | **Adult** acute monocytic leukaemia | | lineage | monocytic/myeloid — the index disease is **infant B-lineage** | | karyotype | has lost chromosome Y | | status | **flagged `Problematic cell line`** — partially contaminated; some ATCC stocks cross-contaminated by K-562 (PMID 7759961), later corrected | | variants | **Plus (permissive) and Minus (non-permissive)**, differing in IFN-γ-inducible TRIM22 and CIITA | **The variant split is the finding.** PMID 39205150 reports that Minus cells carry a **50 kDa truncated Cyclin T1** alongside the canonical 80 kDa protein, that the truncated form is **cytoplasmic** where the full-length protein is nuclear, and that CIITA interacts with Cyclin T1 while TRIM22 recruits CIITA into nuclear bodies with it. **Cyclin T1 is the regulatory subunit of P-TEFb, and every mechanistic account of KMT2A::AFF1 in this graph terminates at P-TEFb recruitment.** So the one line standing in for 'KMT2A-free HOXA leukaemia' has an aberrant form of the complex the whole mechanism runs through — in at least one of its two variants, and no entry here says which variant any result came from.
What this does and does not do to the comparison arm
It does not show CALM-AF10 is menin-dependent. Nothing here measures menin. What it does is add three unmodelled sources of difference to a comparison that already carried two.
The comparison is KMT2A-rearranged infant B-ALL against CALM-AF10 in U-937. It was already confounded by age (adult) and lineage (monocytic). It is now also confounded by:
- Variant ambiguity — Plus and Minus differ in CIITA and TRIM22 expression, and DepMap carries one U-937 record without saying which.
- An aberrant P-TEFb subunit in the Minus variant, in a graph where P-TEFb is the terminus of the fusion's mechanism.
- A historical contamination flag — some early ATCC stocks were cross-contaminated by K-562, since corrected, but provenance of any given published result is rarely stated.
The honest consequence is about weight, not direction. An n=1 whose single line is adult, wrong-lineage, variant-ambiguous and historically flagged should not carry a hypothesis. This project already bounded that MEN1 gene-effect value as underpowered on statistical grounds; these are the material grounds, and they point the same way.
The CIITA coincidence, which may not be one
Q62 established this cycle that CIITA is a nucleocytoplasmic shuttling protein whose export is leptomycin-B sensitive, i.e. XPO1-dependent, and that this graph's CIITA arguments are entirely transcriptional. mechanism-mhc-class-i-bivalency-under-menin-inhibition and the MHC class II thread both run through CIITA.
PMID 39205150 now places CIITA in a complex with Cyclin T1, recruited into nuclear bodies by TRIM22, and makes CIITA one of the two markers distinguishing U-937 Plus from Minus.
So the immune-evasion thread and the P-TEFb thread meet inside the comparator line, and neither thread knows it. That is offered as a lead rather than a result: the paper is an HIV-transcription study in a promonocytic line, nothing in it concerns leukaemia, and the connection may be nothing. But CIITA appearing as a P-TEFb-associated variable in the one line this graph uses as its menin-free control is worth an hour from someone.
| cellosaurus | CVCL_0007 |
|---|---|
| disease | Adult acute monocytic leukaemia |
| lineage | monocytic/myeloid |
| sex | male, lost chromosome Y |
| problematic flag | partially contaminated; some ATCC stocks cross-contaminated by K-562 |
| aberrant p tefb subunit | 50 kDa truncated Cyclin T1 in Minus, partly cytoplasmic |
| entries referencing u937 | 64 |
| role in this graph | the n=1 behind the MEN1 gene-effect argument and the CALM-AF10 arm of the proposed menin-inhibitor panel |
variants
[ "Plus (permissive)", "Minus (non-permissive)" ]
variant markers
[ "TRIM22", "CIITA" ]
confounds in the comparison
[ "age (adult vs infant)", "lineage (monocytic vs B)", "variant ambiguity", "aberrant P-TEFb subunit", "historical contamination flag" ]
- Do not read this as showing CALM-AF10 is menin-dependent. Nothing here measures menin; it adds unmodelled differences to a comparison that already carried age and lineage.
- Do not discount U-937 results because of the contamination flag. It is historical, was corrected, and modern panels STR-authenticate. It matters for provenance discipline and older literature.
- Do not treat the Plus/Minus split as established for leukaemia work. It is maintained in the HIV field, and no entry here has shown it touches HOXA regulation or menin dependence.
- Which U-937 variant is in DepMap, and which was used in the CALM-AF10 papers this graph cites? Neither is stated anywhere, and the MEN1 gene-effect argument rests on the DepMap record.
- Does the truncated cyclin T1 exist in the U-937 stocks used for leukaemia work, or only in the HIV field's sublines? One western blot would answer it.
- Is CIITA appearing as a P-TEFb-associated variable in this line a coincidence? The immune thread and the P-TEFb thread meet here and neither knows it.
- Should the CALM-AF10 arm use a second line? The comparison currently rests on one line that is adult, monocytic, variant-ambiguous and historically flagged.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
U-937 (Cellosaurus CVCL_0007) is an adult acute monocytic leukaemia line from a male donor that has lost chromosome Y and is flagged by Cellosaurus as a problematic cell line, partially contaminated because some ATCC stocks were cross-contaminated by K-562 before the problem was corrected.
Cellosaurus CVCL_0007 (U-937): adult acute monocytic leukaemia, male, lost chromosome Y, flagged Problematic cell line - partially contaminated, some ATCC stocks cross-contaminated by K-562 · Cellosaurus (SIB) 2026What this evidence cannot show
Read live from the Cellosaurus API on 2026-09-21. The contamination flag is historical and was corrected; it does not mean any particular published U-937 result is contaminated, only that stock provenance is a variable that publications rarely state. 'Adult' and 'monocytic' are curated record fields, not re-derived from the original 1976 report.
Evidence that argues against this (1)
- A CONTAMINATION FLAG ON A WIDELY USED LINE IS WEAK EVIDENCE AGAINST ANY GIVEN RESULT. U-937 is in the LL-100 blood cancer panel, the Human Protein Atlas Cell Atlas and DepMap, all of which authenticate their stocks by STR profiling, so modern results are far more likely to be on authentic material than the flag suggests. The flag matters for older literature and for provenance discipline, not as a reason to discount the line. Reid YA et al. 1995
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.journal-article
U-937 exists as permissive (Plus) and non-permissive (Minus) variants distinguished by IFN-gamma-inducible TRIM22 and CIITA, and Minus cells carry a 50 kDa truncated Cyclin T1 alongside the canonical 80 kDa protein which, unlike the nuclear full-length form, is also cytoplasmic; CIITA interacts with Cyclin T1 and TRIM22 recruits CIITA into nuclear bodies with it.
A Truncated Isoform of Cyclin T1 Could Contribute to the Non-Permissive HIV-1 Phenotype of U937 Promonocytic Cells. · Alberio T, Shallak M, Shaik AKB, Accolla RS, Forlani G · Viruses 2024What this evidence cannot show
**This is an HIV-transcription study in a promonocytic line and nothing in it concerns leukaemia.** Read at abstract level on 2026-09-21. The Plus/Minus distinction is specific to that laboratory's sublines and it is not established that the U-937 stock in DepMap, or in any paper this graph cites, corresponds to either. A truncated cyclin T1 detected by molecular weight is not necessarily a distinct gene product. The relevance asserted here - that P-TEFb is the terminus of this fusion's mechanism - is a fact about this graph's model rather than about the paper.
Evidence that argues against this (1)
- THE VARIANT SPLIT MAY BE IRRELEVANT TO EVERY RESULT THIS GRAPH USES. Most CALM-AF10 work in U-937 predates or ignores the Plus/Minus distinction, which is maintained in the HIV field rather than the leukaemia field, and a subline difference in interferon-inducible antiviral factors need not touch HOXA regulation or menin dependence at all. The argument for caring is that the difference lands on cyclin T1 specifically, not that it has been shown to matter here - and it has not. Cellosaurus (SIB) 2026
THE LINE BEHIND THAT n=1, described for the first time: adult, monocytic, variant-ambiguous and historically contamination-flagged. The statistical underpowering was already recorded; these are the material grounds and they point the same way.
THE COMPARISON'S MATERIAL BASIS. Its CALM-AF10 arm is one line that differs from the index disease in age, lineage, variant identity and P-TEFb composition.
WHICHEVER U-937 STOCK THIS PANEL USES SHOULD BE STATED AND STR-AUTHENTICATED, and a second CALM-AF10 line would remove the single-line dependency entirely.
THE ABERRANT SUBUNIT: a 50 kDa truncated cyclin T1, partly cytoplasmic, reported in the non-permissive U-937 variant - in the line this graph uses as its menin-free control, and in the complex every account of this fusion terminates at.
CIITA TURNS UP AS A P-TEFb-ASSOCIATED VARIABLE HERE, interacting with cyclin T1 and recruited to nuclear bodies by TRIM22, and it is one of the two markers separating the U-937 variants. The immune thread and the P-TEFb thread meet inside the comparator line.
the same failure the comparison arm found in U-937: a line's lineage and variant are taken from an annotation rather than measured in the stock used. MV4-11 is annotated monocytic and described by its founding paper as biphenotypic B-myelomonocytic