CRISPR-engineered t(4;11) in human fetal liver — the only model that reproduces infant-specific biology
also called fetal liver MLL-AF4 CRISPR model
Researchers used gene editing to create the exact chromosome swap inside human fetal liver blood cells. Those cells reproduced the features of infant leukaemia that older children's leukaemia does not have.
The same thing, for a scientist
Using CRISPR-Cas9 in primary human **fetal liver** haematopoietic cells to produce a t(4;11)/MLL-AF4 translocation, Rice and colleagues generated a model that replicates the clinical features of infant-ALL and drives both infant-ALL-specific and fetal-specific gene-expression programmes. The framing result of the same paper is that fetal-specific gene expression is maintained in MLL-AF4 **infant**-ALL and not in MLL-AF4 **childhood**-ALL — which is the finding that makes donor age a biological variable rather than a demographic one, and therefore makes the whole model problem in this beat substantive. It is the only engineered system found here that engineers the endogenous translocation in the correct developmental compartment.
Why fetal liver and not cord blood
Infant t(4;11) arises in utero, so the correct compartment is fetal, not neonatal. This graph already holds concept-cord-blood-failure-is-method-not-cell, which corrects the shorthand that cord blood cannot transform — a CRISPR-engineered translocation does transform cord-blood HSPCs. The fetal-liver model is one developmental step earlier still, and it is the one that produces the infant-specific transcriptional programme rather than merely producing outgrowth 1.
Those are different bars. Indefinite growth in culture is not leukaemia; a fetal gene-expression programme is a phenotype the disease has and the older-child disease does not.
What it costs and what it cannot do
Primary human fetal liver tissue is ethically regulated, supply-limited, and in several jurisdictions unobtainable. That is not a scientific limitation but it is a decisive practical one, and it explains why nobody can simply substitute this model for a cell line. There is also no immortalised derivative of it: it is a primary-cell system, so it cannot be screened at genome scale, cannot be banked, and cannot be shared as a line.
So the model that best matches the disease is the one least able to answer a dependency question, and the models that answer dependency questions are the ones that match the disease least. That trade-off is the structure of this beat.
| model class | CRISPR-Cas9 engineered endogenous translocation in primary human tissue |
|---|---|
| compartment | human fetal liver haematopoietic cells |
| fusion | t(4;11) / MLL-AF4 (KMT2A::AFF1) |
| reproduces infant phenotype | true |
| reproduces fetal programme | true |
| immortalised derivative | false |
| screenable at genome scale | false |
| bankable | false |
| obtainable | no — primary fetal tissue, ethically regulated and supply-limited; no distributable line exists |
- A metabolic vulnerability that is fetal rather than fusion-specific would predict toxicity in any fetal or neonatal tissue, which is the population the drug would be given to.
- A model that cannot be banked cannot be replicated by another laboratory from the same material, so its results have the reproducibility profile of a primary-tissue experiment rather than of a cell-line experiment.
- If the fetal programme is required for the disease phenotype, then every dependency nominated in SEM, RS4;11 or MV4-11 was nominated in cells lacking it.
- Can this model be immortalised without losing the fetal programme? A bankable fetal-liver-derived KMT2A::AFF1 line would resolve the central trade-off of this beat in one step.
- Do the existing infant-derived cell lines (PER-490, ALL-PO, SCMC-L1, BLIN-3) retain the fetal programme? If they do, the fetal-tissue requirement applies to model *generation* and not to model *use*, and the lines become much more valuable than this entry implies.
- Is the fatty-acid requirement a fusion property or a fetal-cell property? Nothing in the graph separates them.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
CRISPR-Cas9 gene editing of primary human fetal liver haematopoietic cells to produce a t(4;11)/MLL-AF4 translocation generated a model that replicates the clinical features of infant-ALL and drives infant-ALL-specific and fetal-specific gene-expression programmes; in primary disease, fetal-specific gene-expression programmes are maintained in MLL-AF4 infant-ALL but not in MLL-AF4 childhood-ALL.
A human fetal liver-derived infant MLL-AF4 acute lymphoblastic leukemia model reveals a distinct fetal gene expression program · Rice S, Jackson T, Crump NT, et al. · Nature Communications 2021What this evidence cannot show
Abstract read; full text not read. 'Replicates the clinical features of infant-ALL' is the abstract's own phrasing and the specific features, the engraftment endpoint and the number of independent fetal donors are not established here. A model that reproduces a transcriptional programme has not thereby reproduced a disease.
Evidence that argues against this (2)
- A companion study from a different group defines the fetal origin of MLL-AF4 infant leukaemia in terms of specific fatty acid requirements, i.e. a metabolic rather than a transcriptional signature. Two fetal-origin models framed on different axes are not automatically the same model, and agreement on 'fetal' conceals disagreement about what is fetal about it. Symeonidou V et al. 2021
- The competing engineered route uses cord blood rather than fetal liver and reports monoclonal outgrowth with indefinite growth potential, which would make fetal tissue unnecessary. This graph's own correction entry holds that the cord-blood failure was a property of the construct, not the cell — so the argument that only fetal tissue will do is not settled. Secker KA et al. 2020
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The fetal origin of MLL-AF4 infant leukaemia has been independently defined with specific fatty acid requirements identified as a distinguishing feature.
Defining the fetal origin of MLL-AF4 infant leukemia highlights specific fatty acid requirements · Symeonidou V, Jakobczyk H, Bashanfer S, et al. · Cell Reports 2021What this evidence cannot show
Bibliographic record fetched from PubMed 2026-09-19; the paper was not read and the abstract was not read in full. The claim as stated is the title proposition; the identity of the fatty acids, the assay and the model system are not established here and should not be quoted from this entry.
Evidence that argues against this (1)
- A metabolic requirement measured in a fetal-derived system may be a property of fetal cells generally rather than of the fusion, and no entry in this graph currently holds the fetal-versus-adult metabolic comparison that would separate them. The same objection applies to every 'fetal-specific' claim in this beat. Rice S et al. 2021
- related to →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
endogenous t(4;11) engineered in primary human fetal liver cells
the compartment the model uses
one developmental step earlier than the cord-blood CRISPR route
- supports →Experiment: does MLL-AF4 transform a staged fetal progenitor when it cannot transform cord blood?
fetal programme maintained in infant but not childhood MLL-AF4 ALL
- ← related toThe pluripotent stem-cell route has failed from both ends, and the failures are informative
the fusion is obstructive at pluripotent stages and productive in fetal haematopoietic cells
fetal programme is maintained in infant and not childhood MLL-AF4 ALL, so a 5-year-old's line is not transcriptionally equivalent
the LIN28B perturbation run in this model
the KMT2A::AFF1 engineered human model that would carry the same ordered-accessibility design