The unnamed Cre is Mb1-Cre, which acts at the pro-B stage: the one resolution in this beat that lands on the right cell
also called GSE306634 provenance · Dot1l Mb1Cre · DOT1L B cell Cre driver
An entry on DOT1L in normal B-cell development notes that the paper never says which genetic switch was used or when it acts. The deposited data names it: Mb1-Cre, which turns on right at the pro-B stage - the same developmental point where infant KMT2A::AFF1 leukaemia arises.
The same thing, for a scientist
`mechanism-dot1l-normal-b-lymphopoiesis` carries the caveat *"Mouse conditional knockout - complete genetic loss in the B lineage, not pharmacological inhibition, and the Cre driver and its stage of action are not stated in the abstract"* [E1]. GEO series **GSE306634** names both. Twenty-one *Mus musculus* samples carry explicit genotype strings **`Dot1l^f/f Mb1^Cre/+`** (3) and **`Dot1l^+/+ Mb1^Cre/+`** (3) alongside 15 wild-type samples, with `chip antibody: H3K79me2 (Abcam, #ab3594)` on 12 samples and H3K27me3 and H3K4me3 on one each [E2]. The cell populations are annotated at a granularity the abstract never reaches: `Pro and pre B cells` from bone marrow (4), `Naive B cells` (4), `Follicular B cells` (6) and `Germinal center B cells` (7) from spleen, the last after `NP-KLH immunisation (+10% alum) 7 days prior` [E2]. The protocol states *"Follicular B-cells (CD19+CD21intCD23+) were sort-purified from Mb1Cre/+ and Dot1lf/fMb1Cre/+ mice"* [E2]. **Mb1-Cre (Cd79a-Cre) recombines from the pro-B stage onward**, at the earliest point of B-lineage commitment [E3]. That is the stage this project's disease occupies. So of every resolution in this beat, this is the only one whose cell type matches the index disease's cell of origin - and it is a **normal** B-lymphopoiesis study, which is exactly the comparator this graph needs and usually lacks.
Why the stage of action is the whole question
A conditional knockout's meaning is set by when the Cre fires. Cd19-Cre fires somewhat later and is notoriously incomplete; Mb1-Cre fires at pro-B and is efficient. The difference decides whether a phenotype reflects DOT1L's requirement during B-lineage commitment or only after it.
Because Mb1-Cre fires at pro-B, this paper reports what a developing B cell does when it has never had DOT1L - and it reports effects on repertoire establishment, marginal-zone development and germinal-centre function, i.e. across the whole subsequent lineage 12. For a project weighing DOT1L inhibition in infants, that is the toxicity model, and it is now specified rather than assumed.
The deposition also supplies the wild-type H3K79me2 landscape at the pro/pre-B stage (Developing B cells (B220+IgD-IgM-CD24+) isolated from WT mice (H3K79me2 ChIP)), which is the normal baseline against which a KMT2A::AFF1 blast's H3K79 distribution would have to be compared 2. No entry in this graph currently cites that.
The import that remains
Two axes still differ from the index disease and the entry should keep saying so: species (mouse) and state (normal development, not leukaemia). What no longer differs is lineage and stage, which is unusual enough in this beat to be worth stating positively. Of sixteen resolutions, thirteen are imports; this is one of three that are not imports on lineage, and the only one that is a B-lineage system at the pro-B stage.
| script | none - GEO accession-display text endpoint |
|---|---|
| status | run |
| result summary | Cre driver resolved to Mb1-Cre (Cd79a-Cre), acting from the pro-B stage. 21 murine samples: pro/pre-B from bone marrow, naive, follicular and germinal-centre B cells from spleen, H3K79me2 ChIP on 12. The only resolution in this beat whose cell type and developmental stage match the index disease's cell of origin. |
inputs
[ "PMID 40972012", "GSE306634", "PMID 16940357" ]
counts — the numbers behind the chart
{
"samples_total": 21,
"samples_mouse": 21,
"samples_conditional_ko": 3,
"samples_cre_control": 3,
"samples_wildtype": 15,
"h3k79me2_chip_samples": 12,
"b_cell_populations": 4
}entries affected
[ "mechanism-dot1l-normal-b-lymphopoiesis" ]
import axes
[ "species: mouse", "state: normal development, not leukaemia" ]
non import axes
[ "lineage: B-lymphoid, matching the index disease", "stage: pro-B onward, matching the index disease's cell of origin" ]
- DOT1L inhibition in an infant would act on a B-cell repertoire still being established, and this model says germline-onset loss disturbs repertoire, marginal-zone and germinal-centre biology - a humoral-immunity cost that is plausible, dose-dependent and not measurable in any leukaemia cell line.
- If DOT1L is required for normal pro-B development, then the therapeutic window in infant KMT2A::AFF1 ALL is bounded by the same lineage the disease occupies, which is the least favourable geometry a target can have.
- The knockout is complete and germline-onset; reading it as a drug-toxicity prediction overstates the risk, while ignoring it understates it, and the graph currently does neither explicitly.
- The deposition contains a wild-type H3K79me2 map of sorted murine pro and pre B cells. How does a KMT2A::AFF1 blast's H3K79me2 distribution compare against it? No entry in this graph cites that baseline.
- Does DOT1L loss at the pro-B stage change the cells that KMT2A::AFF1 transforms, or only their subsequent function? An inducible driver would separate these and this model cannot.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The DOT1L B-cell paper reports that DOT1L-mediated H3K79me2 directs B-cell repertoire establishment, marginal-zone development and germinal-centre function, and its abstract does not name the Cre driver or state its stage of action.
DOT1L-mediated H3K79me2 directs B-cell repertoire establishment, marginal zone development, and germinal center function. · Kealy LC, Russ BE, Turner SJ, Hobeika E, Good-Jacobson KL · J Immunol 2026What this evidence cannot show
Identifiers taken from the NCBI esummary record fetched on 2026-09-19. The full text could not be read: no PMCID appears in either the PubMed esummary or the Europe PMC core record (isOpenAccess N, inEPMC N) for this 2026 article, so this entry rests on the abstract and the deposition.
Evidence that argues against this (1)
- A complete germline-onset conditional knockout is a much more severe perturbation than pharmacological DOT1L inhibition in an already-formed repertoire, so the toxicity this model predicts is an upper bound on what a drug given to an infant would cause, not an estimate of it. Godfrey L et al. 2021
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.dataset
GEO series GSE306634 contains 21 Mus musculus samples with explicit genotypes 'Dot1l f/f Mb1Cre/+' and 'Dot1l +/+ Mb1Cre/+', spanning bone-marrow pro and pre B cells, splenic naive, follicular and germinal-centre B cells after NP-KLH immunisation with alum 7 days prior, with H3K79me2 ChIP (Abcam ab3594) on 12 samples.
DOT1L-mediated H3K79me2 directs B-cell repertoire establishment, marginal zone development, and germinal center function (GSE306634) · GEO submitters · NCBI Gene Expression Omnibus 2026What this evidence cannot show
Read on 2026-09-19 from the GEO accession-display text endpoint and enumerated sample by sample. Submitter-entered. Three samples carry each conditional genotype string, so the knockout comparison is n = 3 per arm in the deposited chromatin data; the repertoire-sequencing arm carries its own amplification-primer characteristics and was not separately counted here.
Evidence that argues against this (1)
- Fifteen of the 21 samples are annotated wild type, so most of the deposition is a normal H3K79me2 atlas rather than a knockout comparison; an entry citing the knockout phenotype is citing the smaller half of the series. Rogic S et al. 2026
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Mb1-Cre (Cd79a-Cre) mediates efficient recombination from the earliest committed pro-B stage of B-lymphocyte development.
Testing gene function early in the B cell lineage in mb1-cre mice. · Hobeika E, Thiemann S, Storch B, Jumaa H, Nielsen PJ, Pelanda R, Reth M · Proc Natl Acad Sci U S A 2006What this evidence cannot show
This is the founding characterisation of the mb1-cre line, fetched from PubMed on 2026-09-19; it establishes the driver's stage of action rather than anything about DOT1L. Recombination efficiency in any particular floxed allele depends on the locus and is not guaranteed by the driver, and the DOT1L deposition reports no deletion-efficiency measurement.
Evidence that argues against this (1)
- Even an efficient early driver leaves the question of whether a phenotype arises from loss during commitment or from cumulative loss thereafter, which requires an inducible driver to separate and which this deposition does not contain. Kealy LC et al. 2026
- related to →H3K79me2 builds up as B cells mature - so in B-ALL the drug's target is what the rescued compartment runs on
the driver is Mb1-Cre, acting from the pro-B stage
the deposition contains a wild-type H3K79me2 map of sorted murine pro and pre B cells
normal B-lineage H3K79me2 baseline across four B-cell populations
conditional Dot1l loss from the pro-B stage onward
resolved source in the unstated-system worklist