Is MYC the node? Two papers, the same MLL-AF9 model, opposite answers - and the second one failed to rescue
also called Brg1-Myc contradiction · SWI/SNF MYC dependency dispute
One study found that SWI/SNF keeps MLL-rearranged leukaemia alive by keeping the MYC gene switched on, and proved it by showing extra MYC rescues the cells. A second study, using the same kind of leukaemia, found no change in MYC at all and could not reproduce the rescue. Nobody has resolved it.
The same thing, for a scientist
This is a direct, unresolved contradiction between two primary papers about the same mechanism in the same disease model, and it is the single most important caveat on every MYC-centred SWI/SNF claim in this graph. **Shi et al. 2013**: in MLL-AF9/NrasG12D RN2 cells, Myc is among the most down-regulated genes after Brg1 knockdown (p<0.01), falls within 24 h suggesting direct regulation, and **retroviral Myc renders cells resistant to the cell-cycle arrest and myeloid differentiation caused by Brg1 knockdown** - though Myc-transduced cells still die, indicating a Myc-independent survival function [E1]. **Cruickshank et al. 2015**: in human THP-1 and in mouse MLL-AF9 cells, SMARCA4/SMARCD2/DPF2 knockdown produces **no significant change in c-MYC transcript** by microarray or RT-qPCR, no loss of two independent MYC-defined gene modules, and **c-Myc over-expression failed to rescue** the Smarca4 knockdown proliferation defect - using a c-Myc cDNA that did complement RNAi against Jmjd1c as a positive control [E2]. The abstract states the disagreement explicitly: *"in contrast to previous results obtained in mouse cells, the three proteins are not required for the expression of c-MYC regulated genes."* Both were done in MLL-AF9. Both used shRNA. Both ran a rescue. **The rescue control in the second paper is the strongest part of it** - a Myc cDNA that works for a different gene in the same system does not work here [E2]. What Cruickshank et al. find instead is loss of a haematopoietic stem cell signature defined in human CD133+ cord blood cells [E2].
What the disagreement is, precisely
| Shi 2013 | Cruickshank 2015 | |
|---|---|---|
| system | mouse RN2 (MLL-AF9 + NrasG12D) | mouse MLL-AF9 spleen cells; human THP-1, MV4-11 |
| Myc transcript after knockdown | among most down-regulated, falls by 24 h | no significant change (microarray and RT-qPCR, mouse and human) |
| MYC gene sets | Myc signature globally suppressed | two MYC modules unchanged |
| Myc over-expression rescue | rescues arrest and differentiation (not death) | fails, with a validated positive control |
| what is lost instead | LSC signature, Hoxa9 (weakly) | HSC signature from CD133+ cord blood |
Possible reconciliations, none tested: (1) the NrasG12D in RN2 changes MYC dependence; (2) subunit versus ATPase - Cruickshank's Smarca4 result is the direct comparison and it still disagrees; (3) knockdown depth and timing; (4) mouse-versus-human, but Cruickshank report the negative in both species.
Option 2 is the one that would most change this graph's reading and it is already excluded by the data, because both papers knocked down the same ATPase. That leaves (1) and (3), which are experiment-design differences that a single side-by-side replication would settle.
What the graph should therefore assert
Assertable: SWI/SNF maintains a stem/progenitor transcriptional programme in KMT2A-rearranged leukaemia and its loss releases differentiation 12.
Not assertable without stating the dispute: that SWI/SNF works through MYC in this disease. A later paper reports FHD-286 repressing c-Myc along with PU.1 and CDK4/6 in MLL1-rearranged AML 3, which supports the Shi direction pharmacologically but does not resolve the genetic contradiction, because a dual ATPase inhibitor and a single-subunit shRNA are different perturbations.
The BRD9 literature adds a third voice on the same axis: AML cells require BRD9 to sustain MYC transcription 4. So of four relevant results, three point at MYC and one - the one with the failed rescue and the explicit statement of disagreement - does not.
| mechanism class | unresolved contradiction about the downstream node of a chromatin dependency |
|---|---|
| what both agree on | loss of a stem/progenitor transcriptional programme and release of myeloid differentiation |
dispute
{
"claim_a": "SWI/SNF maintains MLL-rearranged leukaemia by sustaining MYC transcription from distal enhancers (PMID 24285714)",
"claim_b": "SWI/SNF subunits are not required for MYC expression or MYC-regulated gene sets in MLL-rearranged cells, and MYC over-expression does not rescue (PMID 26571505)",
"shared_model": "MLL-AF9",
"decisive_control_in_b": "c-Myc cDNA that complements Jmjd1c RNAi in the same system fails to complement Smarca4 knockdown",
"status": "unresolved as of 2026-09-20; no published side-by-side replication found"
}third party evidence
[
{
"source": "PMID 38437498",
"direction": "supports MYC involvement",
"perturbation": "dual BRG1/BRM ATPase inhibitor, not knockdown"
},
{
"source": "PMID 27376689",
"direction": "supports MYC involvement",
"perturbation": "BRD9 (ncBAF) inhibition in AML, not KMT2A-selected"
}
]- A drug programme that nominates MYC suppression as its pharmacodynamic biomarker is building on a contested node; if the Cruickshank result is right, a trial could see a response with no MYC movement and conclude the drug missed.
- If MYC is not the node, the real node is a stem-cell programme that has no single measurable readout, which makes early-phase dose-finding harder rather than easier.
- Combination logic with BET inhibitors assumes a shared MYC axis; that assumption is exactly what is disputed here.
- Would a side-by-side experiment in one laboratory - RN2 with and without NrasG12D, plus THP-1, one shRNA set, one MYC rescue construct - resolve it? Nothing more complicated appears to be needed and nobody has done it.
- Does the NrasG12D allele in RN2 create the MYC dependence? RAS signalling stabilises MYC protein, which would explain the discrepancy and is directly testable.
- Is MYC protein, rather than transcript, changed by SWI/SNF loss in KMT2A-rearranged cells? Neither paper blotted MYC in the human lines after subunit knockdown.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In MLL-AF9/NrasG12D RN2 cells Myc is among the most down-regulated genes after Brg1 knockdown and falls within 24 h; retroviral Myc renders cells resistant to Brg1-knockdown-induced cell-cycle arrest and myeloid differentiation, though they still die, indicating a Myc-independent survival function.
Role of SWI/SNF in acute leukemia maintenance and enhancer-mediated Myc regulation. · Shi J, Whyte WA, Zepeda-Mendoza CJ, Milazzo JP, et al. · Genes Dev 2013What this evidence cannot show
Mouse line carrying NrasG12D in addition to MLL-AF9, so MYC dependence may be shaped by the RAS lesion; the rescue is a retroviral over-expression, which sets MYC at a non-physiological level.
Evidence that argues against this (1)
- The same paper reports Myc-transduced cells still undergo cell death after Brg1 knockdown, so MYC is at most part of the mechanism even on this paper's own reading. Xiu Y et al. 2018
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
SMARCA4, SMARCD2 or DPF2 knockdown produces no significant change in c-MYC transcript in THP-1 or mouse MLL-AF9 cells by microarray or RT-qPCR, no loss of two independent MYC-defined gene modules, and c-Myc over-expression fails to rescue the Smarca4-knockdown proliferation defect using a cDNA that does complement RNAi against Jmjd1c; SMARCA4 knockdown instead correlates with loss of a haematopoietic stem cell signature from human CD133+ cord blood.
SWI/SNF Subunits SMARCA4, SMARCD2 and DPF2 Collaborate in MLL-Rearranged Leukaemia Maintenance. · Cruickshank VA, Sroczynska P, Sankar A, Miyagi S, et al. · PLoS One 2015What this evidence cannot show
shRNA; the negative is a failure to detect, so knockdown depth and timing matter - but the Jmjd1c positive control makes the failed rescue a controlled negative rather than a null result. Read in full via PMC4646637.
Evidence that argues against this (1)
- Absence of a transcript change does not exclude a protein-level or translational MYC effect, which neither paper measured; and a failed over-expression rescue can also fail because the over-expressed level is wrong rather than because the pathway is wrong. Shi J et al. 2013
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
FHD-286 treatment of AML cells with MLL1 rearrangement or mutant NPM1 perturbs chromatin accessibility and represses c-Myc, PU.1 and CDK4/6.
BRG1/BRM inhibitor targets AML stem cells and exerts superior preclinical efficacy combined with BET or menin inhibitor. · Fiskus W, Piel J, Collins M, Hentemann M, et al. · Blood 2024What this evidence cannot show
Pharmacological dual ATPase inhibition, a different perturbation from single-subunit knockdown; repression of c-Myc is one of three reported and is not isolated as the causal node.
Evidence that argues against this (1)
- Drug-induced MYC repression is a common downstream consequence of arresting proliferation and is not by itself evidence of direct regulation - the Shi enhancer-looping data are what would make it direct, and they were not repeated here. Cruickshank VA et al. 2015
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
AML cells require the BRD9 subunit of SWI/SNF to sustain MYC transcription, rapid proliferation and a block in differentiation, with on-target proof from a bromodomain-swap resistance allele.
Sensitivity and engineered resistance of myeloid leukemia cells to BRD9 inhibition. · Hohmann AF, Martin LJ, Minder JL, Roe JS, et al. · Nat Chem Biol 2016What this evidence cannot show
Mouse and human AML lines, not selected for KMT2A rearrangement; the bromodomain-swap allele is an unusually rigorous on-target control and applies to the compound series, not to the MYC claim specifically.
Evidence that argues against this (1)
- BRD9 sits in ncBAF while the Shi and Cruickshank experiments concern the shared ATPase and cBAF subunits, so a MYC requirement through BRD9 is not the same claim and cannot arbitrate between them. Michel BC et al. 2018
this is the element at the centre of the Shi/Cruickshank disagreement, which the graph previously had no name for
arm 0 of the test measures the BRG1 occupancy neither Shi nor Cruickshank measured in a human line
the same unmeasured quantity - MYC protein - is the open discriminator in both
advances the recorded contradiction: the element is independently validated, the SWI/SNF route is not
the enhancer-to-MYC mechanism is contradicted by a failed MYC rescue in the same model
the contradiction is about the same ATPase in both papers
BRD9 is a third voice on the MYC axis, through a different complex
the arm-1 MYC rescue with a RAS-mutant and RAS-wild-type pair also resolves the MYC contradiction