The cell types that need MLL1 are the uncommitted ones - dependence ends exactly where lineage commitment begins
also called Mll dependence is restricted to early multipotent haematopoiesis
The gene that breaks in infant leukaemia is needed by the earliest, most flexible blood cells, and stops being needed once a cell has decided what it is going to become. That boundary is measured, not assumed.
The same thing, for a scientist
Conditional Mll deletion defines the early multipotent stages of haematopoiesis as Mll-dependent: HSCs and myelo-erythroid progenitors both show phenotypes, and committed lymphoid and myeloid cells no longer require Mll [E1]. That is a cell-type statement with a boundary in it, which is rarer and more useful than 'expressed in blood'. It is not, however, an expression statement: human KMT2A RNA is low-tissue-specificity and detected in all tissues, with low immune-cell and low blood-lineage specificity [E2], so the compartment restriction is a requirement gradient, not an expression gradient. At the protein level the Human Protein Atlas calls KMT2A tissue-enriched in BONE MARROW on antibody staining with Enhanced reliability [E2], which is a third and weaker class of evidence and is recorded here as such rather than merged with the other two.
Three evidence classes, kept apart
This graph already separates expressed in, required in and maintains (concept-expressed-required-maintains-three-evidence-classes). KMT2A scores differently on each:
| class | KMT2A |
|---|---|
| expressed in | everywhere. HPA: RNA tissue specificity Low tissue specificity, distribution Detected in all; blood cell specificity Low immune cell specificity; blood lineage Low lineage specificity 2. |
| required in | HSCs and myelo-erythroid progenitors, in opposite cell-cycle directions; NOT in committed lymphoid or myeloid cells 1. |
| maintains | the Mll-dependent transcriptional network in HSCs, which is broader than the Hox cluster and contains both menin-dependent and menin-independent targets 3. |
The first row is why a transcriptomic argument cannot establish the compartment restriction and the second row is why the compartment restriction is real anyway.
The bone-marrow protein enrichment, and why it is held at arm's length
HPA's antibody-based data call KMT2A tissue enriched with bone marrow the top tissue (staining intensity 4,547,068.8 arbitrary units), and cell-type enriched with hepatocytes top (50,273,485.5), on three antibodies (CAB017794, CAB024270, HPA044910) with IH reliability Enhanced and IF reliability Supported 2.
Two of those statements point in different directions and the hepatocyte call has no support anywhere in the haematopoietic literature, which is the standard signature of an antibody-driven artefact rather than a biological finding. The bone-marrow call is recorded because it is consistent with the genetic requirement, not because IHC intensity is evidence of requirement. Neither number should be quoted as a measurement of abundance.
What this does and does not license about the infant disease
The brief that dispatched this entry asked for the haematopoietic compartment 'without inventing a hierarchy the sources do not support'. The sources support exactly one boundary - multipotent versus committed - measured in the mouse. They do not support a finer ordering among HSC, MPP, LMPP, CMP and GMP, and they do not speak to the human fetal compartments where the index fusion arises.
So the honest statement is: MLL1 is required in the uncommitted compartments and dispensable in committed ones, in mouse, for the wild-type protein. Everything finer than that in this graph comes from fusion experiments and belongs to the fusion entries, not here.
| evidence class key | concept-expressed-required-maintains-three-evidence-classes |
|---|
expressed in
{
"class": 1,
"summary": "ubiquitous. HPA RNA: Low tissue specificity / Detected in all; Low immune cell specificity; Low lineage specificity; Low single-cell type specificity [E2]. Expression carries NO compartment information for this gene."
}required in
{
"class": 2,
"compartments_required": [
"HSC",
"myelo-erythroid progenitor"
],
"compartments_not_required": [
"committed lymphoid",
"committed myeloid"
],
"direction_of_effect": "opposite in the two required compartments - quiescence maintained in HSC, proliferation promoted in progenitors",
"source": "PMID 18371366 (mouse conditional Mll)",
"granularity_limit": "multipotent versus committed is the ONLY boundary the sources establish. No finer ordering among HSC, MPP, LMPP, CMP and GMP is supported here."
}antibody based tissue call
{
"class": "weak - antibody staining, not requirement",
"top_tissue": "bone marrow",
"top_tissue_intensity_arbitrary_units": 4547068.8,
"top_cell_type": "hepatocytes",
"top_cell_type_intensity_arbitrary_units": 50273485.5,
"antibodies": [
"CAB017794",
"CAB024270",
"HPA044910"
],
"reliability_ih": "Enhanced",
"reliability_if": "Supported",
"why_held_at_arms_length": "the hepatocyte call has no support in the haematopoietic literature and points away from the bone-marrow call; two discordant calls from the same instrument are an antibody signature, not a biological one"
}graph links
[ "cell-type-hsc-bone-marrow", "cell-type-fetal-liver-hsc", "cell-type-mpp", "cell-type-cmp", "cell-type-gmp", "cell-type-clp" ]
- If MLL1 dependence really ends at commitment, then any therapy that degrades wild-type KMT2A spares committed cells and hits the stem compartment - exactly the wrong selectivity for a regenerating marrow after intensive chemotherapy.
- Is the compartment boundary the same in human? Every requirement measurement here is murine, and the human fetal compartments in which KMT2A::AFF1 arises sit on the uncommitted side of a boundary that has only been drawn in adult mouse.
- Does the requirement boundary move when a fusion is present? A fusion-bearing cell has one wild-type allele and one neomorphic one, and no experiment has deleted the wild-type allele specifically in a fusion-bearing compartment.
- Why does HPA's antibody data call KMT2A hepatocyte-enriched? Either the antibodies cross-react in liver or there is an unexamined hepatic biology for a gene this project treats as haematopoietic. One two-antibody Western on primary hepatocytes would settle it.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Committed lymphoid and myeloid cells no longer require Mll, defining the early multipotent stages of haematopoiesis as Mll-dependent; within those stages Mll maintains quiescence in HSCs and promotes proliferation in myelo-erythroid progenitors.
Unique and independent roles for MLL in adult hematopoietic stem cells and progenitors. · Jude CD, Climer L, Xu D, Artinger E, Fisher JK, Ernst P · Cell Stem Cell 2007What this evidence cannot show
Mouse; abstract read, full text not opened (PMC2234224 returned an empty body on 2026-09-23). 'Committed lymphoid and myeloid cells no longer require Mll' is the authors' sentence and rests on lineage-restricted Cre drivers whose deletion efficiency is not quoted in the abstract; a failure to delete would produce the same negative result, so the negative half of this claim is weaker than the positive half.
Evidence that argues against this (1)
- A second conditional Mll allele reported grossly normal steady-state haematopoiesis in bone marrow, thymus and spleen after pan-haematopoietic deletion, with the defect appearing only on competitive transplantation. If that allele is the more faithful one, the requirement in uncommitted compartments is a stress-revealed requirement rather than a steady-state one, which changes what 'required in' means here. McMahon KA et al. 2007
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
For KMT2A (ENSG00000118058) the Human Protein Atlas records RNA tissue specificity 'Low tissue specificity' with distribution 'Detected in all', blood cell specificity 'Low immune cell specificity', blood lineage specificity 'Low lineage specificity', and single-cell 'Low cell type specificity'; the antibody-based data call it Tissue enriched with top tissue bone marrow and Cell type enriched with top cell type hepatocytes, on antibodies CAB017794, CAB024270 and HPA044910, IH reliability Enhanced, IF reliability Supported, subcellular main location Nucleoplasm with additional Cytosol.
Human Protein Atlas - KMT2A (ENSG00000118058), machine-readable gene record · Human Protein Atlas consortium · proteinatlas.org 2026What this evidence cannot show
Fetched 2026-09-23 from the per-gene JSON endpoint; HPA version not returned by that endpoint and is therefore NOT recorded, which means this row can drift silently and should be refetched before any number is requoted. The RNA calls are consensus tissue data and the protein calls are antibody staining scored by eye; they are different instruments and the hepatocyte call in particular has no independent support. Staining intensities are arbitrary units, not concentrations.
Evidence that argues against this (1)
- HPA's own RNA data contradict its protein data for this gene: the RNA call is 'Low tissue specificity, Detected in all' with low immune-cell and low lineage specificity, while the antibody data call the protein bone-marrow-enriched AND hepatocyte-enriched. Two of those three statements cannot all be describing the same abundance pattern. Human Protein Atlas consortium 2026
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The Mll-dependent transcriptional network in haematopoietic stem cells extends well beyond clustered Hox genes and contains both menin-dependent and menin-independent target genes.
An MLL-dependent network sustains hematopoiesis. · Artinger EL, Mishra BP, Zaffuto KM, Li BE, Chung EKY, Moore AW, Chen Y, Cheng C, Ernst P · Proc Natl Acad Sci U S A 2013What this evidence cannot show
Mouse; abstract read, full text not opened. The abstract does not name which targets are menin-dependent, so no gene-level assignment may be taken from this citation.
Evidence that argues against this (1)
- The compartment boundary and the network are not the same claim, and this network study does not test committed cells. If the Mll-dependent network is largely menin-independent, then the compartment restriction of MLL1 may not transfer to menin at all, and the therapeutic-index argument that borrows it would be borrowing across a boundary this evidence cannot cross. Maillard I et al. 2009
- ← regulatesKMT2A / MLL1 — the gene that breaks
Mll dependence is restricted to uncommitted haematopoietic compartments (PMID 18371366)
HSC: required, quiescence maintained
- related to →Common myeloid progenitor (CMP) — the fork at which KMT2A leukaemia almost always turns one way
myelo-erythroid progenitor: required, proliferation promoted
- related to →Granulocyte-monocyte progenitor (GMP) — the committed cell an MLL fusion can turn back into a stem cell
myelo-erythroid progenitor compartment in this graph's vocabulary
committed lymphoid: Mll no longer required (PMID 18371366); the negative half of the boundary
uses the graph's expressed/required/maintains separation