The H2BK34ub HOXA9/MEIS1 result is HeLa: the full text contains the word leukaemia zero times
also called MSL2 H2BK34 system · PMC4119175 provenance · H2BK34ub non-leukaemic context
An entry on the histone mark H2BK34ub already suspected that its HOXA9/MEIS1 result came from a non-leukaemic setting. Reading the full text confirms it and names the cells: HeLa, a cervical cancer line. The word leukaemia does not appear in the paper at all.
The same thing, for a scientist
`mark-h2bk34ub` carries an unusually honest caveat - *"The HOXA9/MEIS1 result is in a non-leukaemic context - the abstract does not name a KMT2A-rearranged cell line and this beat does not assume one"* [E1]. The full text, recovered through NCBI `efetch db=pmc` for PMC4119175 after Europe PMC returned HTTP 500, confirms the caveat and supplies the system: the in vitro ubiquitylation assays use **HeLa nucleosomes** and the cellular work is in **HeLa cells** - *"we further tested their interactions in HeLa cells"* and *"We first measured HOXA9 and MEIS1 gene expression in HeLa cells after treatment of: 1) control shRNA; 2) MSL2 shRNA; 3) RNF20/40 overexpression; and 4) MSL2 shRNA followed by RNF20/40 overexpression"* [E2]. A regular-expression tally over the flattened full text returns **zero occurrences of the stem 'leukem'** and eighteen of 'HeLa' [E3]. No deposition exists: elink to GEO, SRA, BioSample and BioProject returns nothing for this PMID [E3]. This is the widest import in the beat. Every other resolved source is at least a haematopoietic cell. This one is a cervical adenocarcinoma epithelial line, and the HOXA9/MEIS1 readout - which in this graph means the leukaemic HOXA programme - is being measured where that programme does not exist as such.
Why an epithelial HOXA9 measurement is a different thing
HOXA9 and MEIS1 are expressed in HeLa and are measurable there, so the experiment is not invalid. But in this graph HOXA9/MEIS1 means the transcriptional programme a KMT2A fusion sustains in a haematopoietic progenitor to enforce a differentiation block. In HeLa it means two homeobox genes in an epithelial cancer line. A mark that supports their expression in one context has not been shown to support the leukaemic programme in the other, and the intervening biology - fusion occupancy, DOT1L recruitment, menin dependence, lineage priming - is entirely absent.
The entry was right to hedge. What it gains here is the ability to say how far the hedge extends: not a leukaemia line whose fusion we do not know but not a leukaemia at all.
The route that recovered it, because four papers in this beat failed it
Europe PMC's fullTextXML endpoint returned HTTP 500 for PMC4119175 on repeated attempts across two sessions. By the rule this project adopted in cycle 50 - a PMCID lookup and a fullTextXML fetch must both fail before you write unreadable - this paper was one fetch away from being filed as unreadable.
https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pmc&id=4119175&retmode=xml returns 55,724 characters of body text. Five of the nine papers in this beat that Europe PMC refused with HTTP 500 were recoverable this way (PMC4119175, PMC2888888, PMC4838499, PMC9827117, PMC7614282); four were not (PMC3593849, PMC6234363, PMC7566887, PMC3795288, which return front matter with no body element). The rule needs a third clause: a Europe PMC HTTP 500 is not a failure until NCBI efetch db=pmc has also returned no body.
| script | none - NCBI efetch db=pmc full text plus elink |
|---|---|
| status | run |
| result summary | System resolved to HeLa cells (human cervical adenocarcinoma epithelium). Zero occurrences of 'leukem' in 55,724 characters of full text; 18 of 'HeLa'. No deposition exists. This is the widest import in the beat - not a different fusion or species, but a different tissue and a non-malignant-haematopoietic context entirely. |
| method note | Europe PMC fullTextXML returned HTTP 500; NCBI efetch db=pmc recovered the body. 5 of 9 such papers in this beat were recoverable this way. |
inputs
[ "PMID 21726816", "PMC4119175" ]
counts — the numbers behind the chart
{
"fulltext_chars": 55724,
"hela_mentions": 18,
"leukemia_stem_mentions": 0,
"depositions": 0,
"leukaemia_cell_lines_named": 0
}entries affected
[ "mark-h2bk34ub" ]
import axes
[ "tissue: cervical epithelium, not haematopoietic", "disease: not leukaemia", "fusion: no fusion present" ]
- Any intervention aimed at H2BK34 ubiquitination in leukaemia would be aimed at a mark whose only functional link to the HOXA programme was measured in an epithelial cancer line, so the target rationale is one system removed from the disease.
- MSL2 sits in the MOF complex, which is required broadly for H4K16 acetylation and chromatin decompaction; inhibiting it carries a general essential-chromatin toxicity that a HeLa expression readout does not begin to bound.
- If the graph treats H2BK34ub as a KMT2A-relevant mark on the strength of this citation, it will propagate an epithelial result into a leukaemia mechanism, which is the exact failure mode this beat exists to find.
- Does MSL2-dependent H2BK34 ubiquitination support the HOXA9/MEIS1 programme in a KMT2A-rearranged haematopoietic cell? The mark's relevance to this project is entirely untested.
- Is the H2BK34ub-to-H3K79 crosstalk the same axis RNF20 uses at K120, or a parallel one? Both were measured in non-leukaemic or single-mouse-line settings, so the graph has no leukaemic comparison of the two marks.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The MSL2 paper reports that the RING finger protein MSL2 in the MOF complex is an E3 ubiquitin ligase for H2B K34 involved in crosstalk with H3 K4 and K79 methylation, and its abstract does not name a KMT2A-rearranged cell line.
The RING finger protein MSL2 in the MOF complex is an E3 ubiquitin ligase for H2B K34 and is involved in crosstalk with H3 K4 and K79 methylation. · Wu L, Zee BM, Wang Y, Garcia BA, Dou Y · Mol Cell 2011What this evidence cannot show
Identifiers taken from the NCBI esummary record fetched on 2026-09-19.
Evidence that argues against this (1)
- The crosstalk this paper establishes between H2B ubiquitination and H3K79 methylation is mechanistically the same axis the RNF20 leukaemia work rests on, so an epithelial demonstration is not irrelevant to leukaemia - it is untargeted rather than inapplicable. Wang E et al. 2013
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The paper's cellular experiments, including the HOXA9 and MEIS1 expression measurements under MSL2 knockdown and RNF20/40 overexpression, were performed in HeLa cells, and its in vitro E3 ligase assays used HeLa nucleosomes as substrate.
The RING finger protein MSL2 in the MOF complex is an E3 ubiquitin ligase for H2B K34 and is involved in crosstalk with H3 K4 and K79 methylation. · Wu L, Zee BM, Wang Y, Garcia BA, Dou Y · Mol Cell 2011What this evidence cannot show
Read from the NCBI efetch db=pmc full text for PMC4119175 on 2026-09-19, after Europe PMC fullTextXML returned HTTP 500 on repeated attempts. Quotations are from the results text. This entry does not claim the paper performed no other cellular work; it claims the HOXA9/MEIS1 experiment the graph cites is the HeLa one, which is what the text states.
Evidence that argues against this (1)
- HeLa is a reasonable substrate source for biochemistry and the in vitro nucleosome assays are context-independent by design, so the biochemical half of this paper transfers to leukaemia far better than the cellular half does - the import applies to the HOXA9/MEIS1 claim specifically. Shema E et al. 2008
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
A regular-expression tally over the paper's flattened full text returns zero occurrences of the stem 'leukem' and eighteen of 'HeLa', and NCBI elink from this PMID to GEO, SRA, BioSample and BioProject returns no records.
NCBI efetch db=pmc for PMC4119175 and elink pubmed to gds, sra, biosample, bioproject for PMID 21726816 · NCBI Entrez Programming Utilities 2026What this evidence cannot show
Run on 2026-09-19 over 55,724 characters of recovered body text. A stem tally over JATS body text excludes supplementary files and figure images, so a leukaemia experiment could exist in a supplement. The elink negative is weak for SRA, BioSample and BioProject, which returned zero for every paper this beat tested including papers with confirmed depositions; it is informative for db=gds, which did return series for other papers.
Evidence that argues against this (1)
- A zero count is exactly the clean, round negative this project has twice been wrong about, and the correct response is to note that the same full text was itself recovered only after a route that returned a clean negative had already failed once. Souren NY et al. 2022
resolves the non-leukaemic context to HeLa; the full text contains no occurrence of the stem leukem
recovered after Europe PMC returned HTTP 500, extending the unreadable rule with a third route
the two H2B ubiquitination citations in this graph rest on an epithelial line and a single mouse line respectively
resolved source in the unstated-system worklist