Use centrifugal elutriation, not a chemical block: the one separation method demonstrated to work in KMT2A-rearranged cells, applied to both the cell-cycle question and the small-persister question
also called elutriation of KMT2A-rearranged cells · size-resolved persister enrichment · phase-resolved fusion occupancy
Standard drugs that freeze cells at one stage of division do not work on these leukaemia cells, but spinning them apart by size does. The same spin could separate the rare small quiet cells that survive treatment, which nobody has tried.
The same thing, for a scientist
Two unconnected observations in this graph point at the same instrument. (1) Five KMT2A-rearranged lines are refractory to mimosine, double thymidine and nocodazole, and the 2007 study that reported this recovered clean cell-cycle fractions by **centrifugal elutriation**, separating on size and sedimentation velocity [E1]. (2) The relapse-associated cells identifiable at diagnosis in infant MLL-rearranged ALL are **smaller**, alongside being quiescent and stem-like [E2]. Elutriation separates on exactly the property that distinguishes them, requires no antibody, no dye and no genetic modification, and processes primary material. This entry proposes one apparatus used for two questions: **(A)** phase-resolved fusion occupancy and composition, which has never been measured in this disease because everyone who tried to synchronise chemically failed; and **(B)** antibody-free enrichment of the small, quiescent compartment from primary marrow, which would make every downstream assay possible on cells currently too rare to study.
Arm A — phase-resolved chromatin and composition
Cells: SEM and RS4;11 (KMT2A::AFF1, lymphoid) plus MOLM-13 (KMT2A::MLLT3, myeloid), with a KMT2A-wild-type B-ALL line and a KMT2A-wild-type monocytic line as lineage controls. Including both lineages and both wild-type controls is what would settle whether the synchronisation refractoriness is a fusion property or a monocytic one — three of the five originally reported lines are monocytic.
Separation: counterflow centrifugal elutriation into G1, S and G2/M fractions, purity confirmed by propidium iodide DNA content on FACS, exactly as in the founding observation.
Readouts, per fraction:
- Fusion abundance by the published fusion-selective configuration: anti-AFF1 C-terminal IP, anti-KMT2A (aa 720–780) blot, size-discriminated. This reproduces the 2007 result and extends it to a second partner.
- Fusion occupancy by CUT&RUN. The honest limitation is that MLL-N antibodies do not distinguish fusion from wild-type, so the definitive version requires the endogenous C-terminal tag knock-in this graph already proposes.
- Menin occupancy in the same fractions, which would say whether the menin-KMT2A interface is engaged uniformly through the cycle.
- Chromatin accessibility and H3K79me2 per fraction.
Why it matters: the 2007 result is that fusion level is constant. Whether fusion occupancy is constant is a different and unasked question, and if occupancy varies with phase then a menin inhibitor's effect should vary with phase — which is the hypothesis hyp-menin-sensitivity-tracks-cell-cycle-state states and exp-cell-cycle-sorted-menin-sensitivity proposes to test.
Arm B — antibody-free enrichment of the small quiescent compartment
Input: primary diagnostic marrow or a KMT2A-rearranged patient-derived xenograft, which supplies enough cells.
Separation: elutriation fractions collected across the size range, with the smallest fractions retained rather than discarded. Conventional practice discards the small fraction as debris and dead cells, which is one reason this compartment may have been systematically thrown away.
Validation that the small fraction is the persister compartment:
- score the prednisone-dependent signature used to define relapse-associated cells, by single-cell or bulk RNA;
- Ki-67 and Hoechst/Pyronin Y for cycle status;
- limiting-dilution transplantation for leukaemia-initiating frequency;
- ex vivo prednisone and venetoclax sensitivity versus the large fraction.
If it works, the compartment becomes tractable for the assays that currently cannot be run on it: BH3 profiling, Seahorse metabolic flux, CUT&RUN, and drug screening. Several of this beat's open questions are gated on exactly that.
If it does not, the negative is still informative: it would mean size is a correlate of the state rather than a handle on it, and would redirect effort to dye-dilution or reporter-based isolation.
What would make this not worth doing
- If a modern re-test finds the five lines can be synchronised — reagent quality, protocols and cell-line substocks have all changed since 2007 — then arm A collapses to a routine experiment anyone can do with a thymidine block, and that is a good outcome that costs one week to discover. This should be the first step.
- Elutriation requires a specific rotor and a technician who has used it; it is not a plate assay. If the small-cell enrichment can be matched by FACS on forward scatter, which is universally available, then arm B should be done that way instead. A head-to-head against forward-scatter sorting is worth including.
- Elutriation's resolution by size is imperfect for cells of similar diameter, and a quiescent leukaemic blast and a contaminating normal lymphocyte are similar sizes, so arm B in primary marrow needs a leukaemia-specific marker for the readout even though the separation is marker-free.
| status | proposed |
|---|---|
| step zero | re-test whether SEM, RS4;11 and MOLM-13 can be synchronised by double thymidine today, with percent-in-phase reported; one week, and a positive result makes arm A unnecessary |
| why nobody has done it | elutriation fell out of routine use once FACS became universal, and the small fraction is conventionally discarded as debris |
tests hypotheses
[ "hyp-menin-sensitivity-tracks-cell-cycle-state" ]
arm a
{
"question": "does fusion OCCUPANCY vary through the cell cycle even though fusion ABUNDANCE does not?",
"cells": [
"SEM",
"RS4;11",
"MOLM-13",
"KMT2A-wild-type B-ALL control",
"KMT2A-wild-type monocytic control"
],
"separation": "counterflow centrifugal elutriation; PI/FACS purity confirmation",
"readouts": [
"fusion abundance: anti-AFF1 C-terminal IP then anti-KMT2A aa 720-780 blot, size-discriminated",
"fusion occupancy: CUT&RUN (limited by MLL-N antibody specificity; definitive version needs an endogenous C-terminal tag)",
"menin occupancy per fraction",
"chromatin accessibility and H3K79me2 per fraction"
]
}arm b
{
"question": "can the small, quiescent, relapse-associated compartment be enriched from primary material by size alone, without antibodies or dyes?",
"input": "primary diagnostic marrow or KMT2A-rearranged PDX",
"validation": [
"prednisone-dependent signature scoring on the small fraction",
"Ki-67 and Hoechst/Pyronin Y cycle status",
"limiting-dilution transplantation for LIC frequency",
"ex vivo prednisone and venetoclax sensitivity versus the large fraction"
],
"comparator": "FACS sorting on forward scatter, which is universally available; if it matches, use it instead"
}falsifiers of the premise
[ "the five lines synchronise normally with a modern double thymidine block", "the small elutriation fraction is not enriched for the prednisone-resistant signature", "forward-scatter sorting achieves the same enrichment, making elutriation unnecessary" ]
atlas refs
[
{
"label": "Elutriation as the method that worked where three chemical blocks failed",
"pmid": "17908926",
"title": "Bimodal degradation of MLL by SCFSkp2 and APCCdc20 assures cell cycle execution: a critical regulatory circuit lost in leukemogenic MLL fusions.",
"authors": "Liu H, Cheng EH, Hsieh JJ",
"venue": "Genes Dev",
"year": 2007,
"title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
},
{
"label": "Relapse-associated cells at diagnosis are smaller, quiescent and stem-like",
"pmid": "34304246",
"title": "Identification and characterization of relapse-initiating cells in MLL-rearranged infant ALL by single-cell transcriptomics.",
"authors": "Candelli T, Schneider P, Garrido Castro P, et al.",
"venue": "Leukemia",
"year": 2022,
"title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
}
]- The whole experiment rests on a nineteen-year-old narrative observation with no figure behind it, so the first outcome may be that the premise is wrong — which is worth one week to find out and would have saved several downstream designs.
- Enriching a therapy-resistant compartment in order to study it does not by itself suggest a drug, and the risk is spending effort characterising a state whose only actionable feature turns out to be that it is not dividing.
- Any phase-resolved occupancy measurement using MLL-N antibodies measures wild-type KMT2A and the fusion together, so a negative result could mean no phase variation in the fusion or phase variation masked by the wild-type signal.
- Is the 2007 synchronisation failure reproducible at all? Everything here is downstream of an unquantified observation, and the cheapest possible first step is to try a thymidine block and count.
- Does elutriation resolve leukaemic persisters from normal small lymphocytes in primary marrow, or does it require a leukaemia-specific readout to interpret the fraction?
- If fusion occupancy does vary with phase, is the variation large enough to produce a difference in menin-inhibitor sensitivity that a sensitivity assay could detect?
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Five human MLL leukaemia cell lines — SEM, RS(4;11), THP-1, NOMO-1 and MOLM13 — could not be synchronised with mimosine, double thymidine or nocodazole, and cell-cycle-phase populations were instead separated by centrifugal elutriation on the basis of physical properties with purity assessed by FACS analysis of DNA content; in the elutriated SEM fractions the endogenous MLL-AF4 protein, detected by anti-AF4 C-terminal immunoprecipitation followed by anti-MLL amino acid 720-780 immunoblot, did not fluctuate through the cell cycle.
Bimodal degradation of MLL by SCFSkp2 and APCCdc20 assures cell cycle execution: a critical regulatory circuit lost in leukemogenic MLL fusions · Liu H, Cheng EH, Hsieh JJ · Genes Dev 2007What this evidence cannot show
Read in full via the PMC HTML route (PMC1993870) after fullTextXML returned HTTP 500. The synchronisation failure is a narrative statement with no figure, no efficiency number and no supplementary data, so the premise of this experiment rests on an unquantified nineteen-year-old observation - which is why re-testing it is the proposed first step rather than an assumption. Three of the five lines are monocytic, so a lineage explanation is unexcluded. Cell lines only.
Evidence that argues against this (1)
- An isogenic test argues the fusion does not break checkpoints. MLL-AF4 or AF4-MLL knocked into the AAVS1 locus of HEK293 cells gave DNA damage repair kinetics identical to wild type by three assays, unchanged NHEJ, alt-NHEJ and HR protein levels and efficiencies, and etoposide induced S/G2/M arrest independently of fusion expression, which the authors describe as proper activation of the DNA damage checkpoints. If the synchronisation failure is not a fusion property, arm A's lineage controls become the main result rather than a control. Read: abstract only. Castano J et al. 2016
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Leukaemic cells associated with high relapse risk in diagnostic samples from infants with MLL-rearranged ALL, identified by single-cell RNA sequencing on two platforms scored against an independently derived prednisone-dependent signature, are characterised by basal activation of the glucocorticoid response, smaller size, and a quiescent gene expression programme with cell stemness properties.
Identification and characterization of relapse-initiating cells in MLL-rearranged infant ALL by single-cell transcriptomics · Candelli T, Schneider P, Garrido Castro P, Jones LA, Bodewes E, Rockx-Brouwer D, et al., Stam RW · Leukemia 2022What this evidence cannot show
Smaller size in a single-cell RNA experiment is typically inferred from transcript or gene counts rather than measured physically, so the link to a physical separation method is an inference this experiment would itself be testing. Diagnostic samples only; cohort size not stated in the abstract. The resistant population is defined by one external drug signature. Read: abstract only.
Evidence that argues against this (1)
- Size and quiescence are not independent: a cell that has not completed a growth cycle is smaller for that reason, so enriching by size may simply enrich for non-growing cells generally, including normal lymphocytes in a primary marrow, rather than for the leukaemic persister specifically. The clonal literature adds a second problem - quiescent and proliferating subclones in this disease trade places under changing pressure - so a size-based fraction taken at one moment captures current occupancy of the state and not a stable cell identity. Read: abstract only. Bardini M et al. 2015
step zero re-tests whether the five lines can be synchronised at all; arm A then measures phase-resolved fusion and menin occupancy by elutriation
arm B tests whether the small quiescent relapse-associated compartment can be enriched from primary marrow by size alone
supplies the separation method that makes the phase-stratified sensitivity question answerable in cells refractory to chemical blocks